CHEMICAL THERMODYNAMICS OF NICKEL
CHEMICAL THERMODYNAMICS Vol. 1. Chemical Thermodynamics of Uranium (H. Wanner and I. Forest, eds.) Vol. 2. Chemical Thermodynamics of Americium (R.J. Silva etal.) Vol. 3. Chemical Thermodynamics of Technetium (M.C.A. Sandino and E. Östhols, eds.) Vol. 4. Chemical Thermodynamics of Neptunium and Plutonium (OECD Nuclear Energy Agency, ed.) Vol. 5. Update on the Chemical Thermodynamics of Uranium, Neptunium, Plutonium, Americium and Technetium (OECD Nuclear Energy Agency, ed.) Vol. 6. Chemical Thermodynamics of Nickel (OECD Nuclear Energy Agency, ed.)
CHEMICAL THERMODYNAMICS 6
Chemical Thermodynamics of Nickel Heinz GAMSJÄGER (Chairman) Lehrstuhl für Physikalische Chemie Montanuniversität Leoben Leoben (Austria)
Jerzy BUGAJSKI Lehrstuhl für Physikalische Chemie Montanuniversität Leoben Leoben (Austria)
Tamás GAJDA Department of Inorganic and Analytical Chemistry University of Szeged Szeged (Hungary)
RobertJ. LEMIRE Deep River O N K0J 1P0
Wolfgang PREIS Lehrstuhl für Physikalische Chemie Montanuniversität Leoben Leoben (Austria)
(Canada)
Edited by FedericoJ. MOMPEAN (Series Editor and Project Co-ordinator) Myriam ILLEMASSÈNE (Volume Editor) and Jane PERRONE OECD Nuclear Energy Agency, Data Bank Issy-les-Moulineaux (France)
2005 ELSEVIER AMSTERDAM •BOSTON •HEIDELBERG •LONDON •NEW YORK •OXFORD PARIS •SAN DIEGO •SAN FRANCISCO •SINGAPORE •SYDNEY •TOKYO
Published by ELSEVIERB.V. Radarweg 29 P.O. Box 211, 1000 AE Amsterdam, The Netherlands
ISBN: 0-444-51802-9 ISBN: 0-444-51903-3 (CD-ROM) Copyright © Organisation for Economic Co-Operation and Development, 2005. All rights reserved. No part of this publication may be reproduced, stored in a retrieval, system or transmitted in any form or by any means, electronic, electrostatic, magnetic tape, mechanical, photocopying, recording or otherwise, without permission in writing from the copyright holders. Queries concerning permission or translation rights should be addressed to: Head of Publications Service, OECD, 2 rue André Pascal, 75775 Paris Cedex 16, France Special regulations for readers in the U.S.A. - This publication has been registered with the Copyright Clearance Center Inc. (CCC), Salem, Massachusetts. Information can be obtained from the CCC about conditions under which photocopies of parts of this publication may be made in the U.S.A. All other copyright questions, including photocopying outside of the U.S.A., should be referred to the OECD. No responsibility is assumed by the Publisher or by the OECD for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. The opinions expressed and arguments employed in this publication are the sole responsibility of the authors and do not necessarily reflect those of the OECD or of the governments of its member countries. Les idées exprimées et les arguments avancés dans cette publication sont ceux des auteurs et ne reflètent pas nécessairement ceux de l'OCDE ou des gouvernments de ses pays membres.
This volume has been reproduced by the publisher from the formatted pages prepared by the editors. @ The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Printed in The Netherlands
Preface This volume is the sixth of the series "Chemical Thermodynamics" edited by the OECD Nuclear Energy Agency (NEA). It is a critical review of the Thermodynamic Data Base of nickel and its compounds initiated by the Management Board of the NEA Thermochemical Database Project Phase II (NEA TDB II). The TDB Ni review team first met at the University of Leoben, Austria in January 1999. Three subsequent plenary meetings were held at NEA Headquarters at Issy-les-Moulineaux (France) in January 2001, October 2001 and in November 2002, two were held at Leoben in May 2002 and in November 2003. Smaller working subgroups met at Lulea, Sweden in October 1999 and in April 2001. The Executive Group of the Management Board provided scientific assistance in the implementation of the NEA TDB Project Guidelines. Hans Wanner participated in meetings of the Review Team as the designated member of the Executive Group. At the NEA Data Bank the responsibility for the overall co-ordination of the Project was placed with Eric Osthols (from its initiation in 1999 to February 2000), with Stina Lundberg (from March 2000 to September 2000) and with Federico Mompean (since September 2000). Federico Mompean was in charge of the preparation of the successive drafts, updating the NEA thermodynamic database and editing the book to its present final form, with assistance of Myriam Illemassene, Jane Perrone, Katy Ben Said and Cristina Domenech-Orti. Originally Willis Forsling, Lars Gunneriusson (Lulea University of Technology, Sweden), Jerzy Bugajski, Heinz Gamsjager, and Wolfgang Preis (University of Leoben, Austria) participated in the nickel project. In October 2001 Robert Lemire joined the Review Team. As member, chairman and peer reviewer of previous NEA TDB projects he contributed invaluable expertise to seeing the Ni review through to completion. In August 2002 time constraints and the pressure of other commitments forced Willis Forsling and Lars.Gunneriusson to resign from the Review Team and Tamas Gajda, University of Szeged, Hungary, joined as a new member. Although almost all of the members of the final review team contributed text and comments to several chapters, primary responsibility for the different chapters was divided as follows. Jerzy Bugajski prepared the sections on elemental nickel, silicon compounds and complexes, germanium compounds and complexes and, with the chairman, the section on boron compounds and complexes. Tamas Gajda drafted the sections for halide and pseudohalide complexes, the section for hydroxo complexes, and the sections on nitrato and thiocyanato complexes. Robert Lemire prepared Chapter VI, Discussion of auxiliary data selection, the sections on halates, sulphates, phosphorus compounds and complexes, arsenic compounds and complexes and, with the chairman, the section on solid halides (he also extensively reviewed several of the other sections, and revised the English throughout the text). Wolfgang Preis prepared the sections on the oxide, the sulphides and the hydrogensulphido complexes and, with the chairman, V
VI
PREFACE
the section on Ni(III,IV) hydroxides (for the Leoben group he also implemented a data base for ionic strength corrections and calculations ensuring internal consistency of the selected values). The chairman drafted the sections on aqua ions, hydroxides and carbonates. The initial contributions from Willis Forsling and Lars Gunneriusson on halide and pseudohalide complexes and (together with Wolfgang Preis) on hydroxo complexes are gratefully acknowledged as they constituted the starting point for subsequent discussions within the Review Team. While there is no need to repeat the general purpose of this review a side effect deserves mentioning. The selection of key values, e.g., for Ni2+(aq), revealed gaps in our knowledge which may stimulate rewarding projects on the experimental thermodynamics of nickel compounds.
Leoben, Austria, October 2004
Heinz Gamsjager, Chairman
Acknowledgements For the preparation of this book, the authors have received financial support from the NEA TDB Phase II Project. The following organisations take part in the Project:
ANSTO, Australia ONDRAF/NIRAS, Belgium RAWRA, Czech Republic POSIVA, Finland ANDRA, France IPSN (now IRSN), France FZK, Germany JNC, Japan ENRESA, Spain SKB, Sweden SKI, Sweden HSK, Switzerland NAGRA, Switzerland PSI, Switzerland BNFL, UK NIREX, UK DoE, USA Jerzy Bugajski, Heinz Gamsjager and Wolfgang Preis would like to express their gratitude to the Institut fur Physikalische Chemie of the Montanuniversitat Leoben for having provided the infrastructure necessary for their contributions to this project. Tamas Gajda gratefully acknowledges the technical support of the Department of Inorganic and Analytical Chemistry, University of Szeged. Robert Lemire wishes to thank Atomic Energy of Canada Limited (AECL) for allowing him to participate in this project, and for allowing him to use the library and other AECL facilities at the Chalk River Laboratories. The authors also thank Werner Sitte for his help with and comments on the section of silicon compounds and complexes. Heinz Gamsjager thanks the library staff at the University of Leoben and the Austrian Central Library for Physics for their literature searches and assistance in obtaining copies of many references, he also appreciates the help of Malcolm Rand, Alan Dinsdale, Heiko Kleykamp and Herbert Ipser for their
vn
Vlll
ACKNOWLEDGEMENTS
advice concerning papers difficult to come by. The unceasing efforts of Federico Mompean, coordinator for the TDB project during the time the main part of the draft of this book was assembled, edited for peer review and finally prepared for the press, is greatly appreciated. His work built on the earlier efforts of Erik Osthols and Stina Lundberg. The authors are indebted to Myriam Illemassene, Jane Perrone and Katy Ben Said who, with admirable alertness and scientific competence, transformed contributions of five authors, prepared in many different formats, into a consistent text with correctly numbered tables and figures. At the NEA Data Bank, Pierre Nagel and Eric Lacroix have provided excellent software and advice, which have eased the editorial and database work. Cynthia Picot, Solange Quarmeau and Amanda Costa from NEA Publications have provided considerable help in editing the present series. Their contributions and the support of many NEA staff members are highly appreciated. The entire manuscript of this book has undergone a peer review by an independent international group of reviewers, according to the procedures in the TDB-6 Guideline, available from the NEA. The peer reviewers have seen and approved the modifications made by the authors in response to their comments. The peer review comment records may be obtained on request from the OECD Nuclear Energy Agency. The peer reviewers were: Dr. Donald A. Palmer, Batelle, Oak Ridge National Laboratory, USA and Prof. Wolfgang Voigt, University of Freiberg, Germany. Their contributions are gratefully acknowledged.
Note from the Chairman of the NEA TDB Project Phase II The need to make available a comprehensive, internationally recognised and qualityassured chemical thermodynamic database that meets the modelling requirements for the safety assessment of radioactive waste disposal systems prompted the Radioactive Waste Management Committee (RWMC) of the OECD Nuclear Energy Agency (NEA) to launch in 1984 the Thermochemical Database Project (NEA TDB) and to foster its continuation as a semi-autonomous project known as NEA TDB Phase II in 1998. The RWMC assigned a high priority to the critical review of relevant chemical thermodynamic data of inorganic species and compounds of the actinides uranium, neptunium, plutonium and americium, as well as the fission product technetium. The first four books in this series on the chemical thermodynamics of uranium, americium, neptunium and plutonium, and technetium originated from this initiative. The organisation of Phase II of the TDB Project reflects the interest in many OECD/NEA member countries for a timely compilation of the thermochemical data that would meet the specific requirements of their developing national waste disposal programmes. The NEA TDB Phase II Review Teams, comprising internationally recognised experts in the field of chemical thermodynamics, exercise their scientific judgement in an independent way during the preparation of the review reports. The work of these Review Teams has also been subjected to further independent peer review. Phase II of the TDB Project consisted of: (i) updating the existing, COD ATAcompatible database for inorganic species and compounds of uranium, neptunium, plutonium, americium and technetium, (ii) extending it to include selected data on inorganic species and compounds of nickel, selenium and zirconium, (iii) and further adding data on organic complexes of citrate, oxalate, EDTA and iso-saccharinic acid (ISA) with uranium, neptunium, plutonium, americium, technetium, nickel, selenium, zirconium and some other competing cations. The NEA TDB Phase II objectives were formulated by the 17 participating organisations coming from the fields of radioactive waste management and nuclear regulation. The TDB Management Board is assisted for technical matters by an Executive Group of experts in chemical thermodynamics. In this second phase of the Project, the
IX
X
NOTE FROM THE CHAIRMAN OF THE NEA
TDB PROJECT
PiIASEII
NEA acts as coordinator, ensuring the application of the Project Guidelines and liaising with the Review Teams. The present volume is the second one published within the scope of NEA TDB Phase II and contains a database for inorganic species and compounds of nickel. We trust that the efforts of the reviewers, the peer reviewers and the NEA Data Bank staff merit the same high recognition from the broader scientific community as received for previous volumes of this series.
Mehdi Askarieh United Kingdom Nirex limited Chairman of TDB Project Phase II Management Board On behalf of the NEA TDB Project Phase II Participating Organisations: ANSTO, Australia ONDRAF/NIRAS, Belgium RAWRA, Czech Republic POSIVA, Finland ANDRA, France IPSN (now IRSN), France FZK, Germany JNC, Japan ENRESA, Spain SKB, Sweden SKI, Sweden HSK, Switzerland PSI, Switzerland BNFL, UK Nirex, UK DoE, USA
Editor's note
This is the sixth volume of a series of expert reviews of the chemical thermodynamics of key chemical elements in nuclear technology and waste management. This volume is devoted to the inorganic species and compounds of nickel. The tables contained in Chapters III and IV list the currently selected thermodynamic values within the NEA TDB Project. The database system developed at the NEA Data Bank, see Section II.6, assures consistency among all the selected and auxiliary data sets. The recommended thermodynamic data are the result of a critical assessment of published information.. The values in the auxiliary data set, see Tables IV-1 and IV-2 have been adopted from CODATA key values or have been critically reviewed in this or earlier volumes of the series.
XI
Xll
How to contact the NEA TDB Project Information on the NEA and the TDB Project, on-line access to selected data and computer programs, as well as many documents in electronic format are available at www.nea.fr. To contact the TDB project coordinator and the authors of the review reports, send comments on the TDB reviews, or to request further information, please send e-mail to
[email protected]. If this is not possible, write to: TDB project coordinator OECD Nuclear Energy Agency, Data Bank Le Seine-St. Germain 12, boulevard des lies F-92130 Issy-les-Moulineaux FRANCE The NEA Data Bank provides a number of services that may be useful to the reader of this book. •
The recommended data can be obtained via internet directly from the NEA Data Bank.
•
The NEA Data Bank maintains a library of computer programs in various areas. This includes geochemical codes such as PHREEQE, EQ3/6, MINEQL, MINTEQ and PHRQPITZ, in which chemical thermodynamic data like those presented in this book are required as the basic input data. These computer codes can be obtained on request from the NEA Data Bank.
Contents Preface
v
Acknowledgement
vii
Note from the chairman of the NEA TDB Project Phase II
ix
Editor's note
xi
Part I
1
Introductory material
I INTRODUCTION
3
1.1
Background
3
1.2
Focus of the review
5
1.3
Review procedure and results
6
II STANDARDS, CONVENTIONS, AND CONTENTS OF THE TABLES
11
11.1 Symbols, terminology and nomenclature II. 1.1 Abbreviations II. 1.2 Symbols and terminology II.1.3 Chemical formulae and nomenclature II. 1.4 Phase designators II. 1.5 Processes II. 1.6 Equilibrium constants II. 1.6.1 Protonation of a ligand II. 1.6.2 Formation of metal ion complexes II. 1.6.3 Solubility constants II. 1.6.4 Equilibria involving the addition of a gaseous ligand II. 1.6.5 Redox equilibria II. 1.7 pH 11.1.8 Order of formulae 11.1.9 Reference codes 11.2 Units and conversion factors
11 11 13 15 15 17 18 18 19 20 21 22 25 27 28 28
11.3
31
Standard and reference conditions
11.3.1 11.3.2 11.3.3
Standard state Standard state pressure Reference temperature
31 32 35 xiu
xiv
CONTENTS
11.4
Fundamental physical constants
35
11.5
Uncertainty estimates
36
H.6
The NEA-TDB system
36
II.7
Presentation of the selected data
38
Part II
Tables of selected data
41
III SELECTED NICKEL DATA
43
IV SELECTED AUXILIARY DATA
53
Part III Discussion of data selection
73
V
DISCUSSION OF DATA SELECTION FOR NICKEL
75
Elemental nickel
75
V.I.I Nickel gas V.I.2 Nickel crystal V.I.3 Nickel liquid V.2 Simple nickel aqua ions
75 76 78 79
V.2.1 Ni2+ V.2.1.1 Gibbs energy of formation of Ni2+ V.2.1.1.1 Temperature coefficient of the standard electrode potential Ni 2+ |Ni V.2.1.2 Enthalpy of formation of Ni2+ V.2.1.3 Partial molar entropy of Ni2+ V.2.1.3.1 NiSO4-7H2O(cr) cycle V.2.1.3.2 NiCl2-6H2O(cr) cycle V.2.1.3.3 Entropy of Ni2+from its AfG° and A f //° V.2.1.4 Heat capacity of Ni2+ V.2.2 Other oxidation states V.2.2.1 Ni3+ V.3 Oxygen and hydrogen compounds and complexes V.3.1 Aqueous nickel(II) hydroxo complexes V.3.1.1 Hydroxo complexes in acidic or neutral solutions V.3.1.2 Hydroxo complexes in alkaline solutions V.3.2 Solid nickel oxides and hydroxides V.3.2.1 Ni(II) oxide V.3.2.1.1 Solubility measurements V.3.2.2 Ni(II) hydroxides, Ni(OH)2(cr) V.3.2.2.1 Crystallography and mineralogy of nickel hydroxide
79 80
V.I
82 82 84 84 85 86 86 88 88 90 90 90 99 102 102 107 108 108
CONTENTS
V.3.2.2.1.1 p-Ni(OH)2 V.3.2.2.1.2 a-Ni(OH)2 V.3.2.2.2 Heat capacity and entropy of Ni(OH)2(cr) V.3.2.2.3 Thermodynamic analysis of solubility data of Ni(OH)2(s) V.3.2.3 Ni(III, IV) hydroxides V.3.2.3.1 P-NiOOH V.3.2.3.2 y-NiOOH V.4 Group 17 (halogen) compounds and complexes V.4.1 Nickel halide compounds V.4.1.1 Solid nickel fluoride, NiF2(cr) V.4.1.1.1 Enthalpy of formation of NiF2(cr) V.4.1.2 Solid nickel chloride, NiCl2(cr) V.4.1.2.1 Low-temperature heat capacity and entropy of NiCl2(cr) V.4.1.2.2 The high-temperature heat capacity of NiCl2(cr) V.4.1.2.3 Enthalpy of formation of NiCl2(cr) V.4.1.2.4 Gibbs energy of formation of NiCl2(cr) V.4.1.3 Hydrated NiCl2 solids V.4.1.3.1 Gibbs energy of formation of NiCl2-6H2O V.4.1.3.2 Enthalpy of formation of NiCl2-6H2O V.4.1.3.3 The entropy and heat capacity of NiCl2-6H2O(cr) V.4.1.3.4 Gibbs energy and enthalpy of formation of nickel chloride tetrahydrate V.4.1.3.5 The entropy and heat capacity of nickel chloride dihydrate V.4.1.3.6 Gibbs energy and enthalpy of formation of nickel chloride dihydrate V.4.1.4 Solid nickel bromide, NiBr2(cr) V.4.1.4.1 Low-temperature heat capacity and entropy of NiBr2(cr) V.4.1.4.2 The high-temperature heat capacity of NiBr2(cr) V.4.1.4.3 Enthalpy of formation of NiBr2(cr) V.4.1.5 Hydrated solid, NiBr 2 xH 2 O V.4.1.6 Solid nickel iodide, NiI2(cr) V.4.1.6.1 Entropy of NiI2(cr) V.4.1.6.2 Heat capacity of NiI2(cr) V.4.1.6.3 Enthalpy of formation of NiI2(cr) V.4.1.7 Solid nickel iodate compounds V.4.2 Aqueous nickel halide complexes V.4.2.1 Introduction V.4.2.2 Solution structural studies V.4.2.3 Aqueous Ni(II) - fluoro complexes V.4.2.4 Aqueous Ni(II) - chloro complexes V.4.2.5 Aqueous Ni(II) - bromo complexes V.4.2.5.1 Determination of the Ni2+ - Br~ ion interaction coefficient
XV
108 109 109 111 115 116 118 119 119 119 120 122 122 124 124 127 128 128 128 129 130 131 131 132 132 134 134 135 136 136 136 137 137 140 140 141 142 146 153 156
XVI
CONTENTS
V.4.2.6 Aqueous Ni(II) - iodine complexes V.4.2.6.1 Aqueous Ni(II) - iodo complexes V.4.3 Determination of the N i 2 + - C1O4 ion interaction coefficient V.5 Group 16 (chalcogen) compounds and complexes V.5.1 Sulphur compounds and complexes V.5.1.1 Nickel sulphides V.5.1.1.1 Solid nickel sulphides V.5.1.1.1.1 Crystallography and mineralogy of nickel sulphides V.5.1.1.1.1.1 Heazlewoodite, Ni3S2(cr) V.5.1.1.1.1.2 a-Ni7S6 V.5.1.1.1.1.3 Pentlandite, (Fe, Ni)9S8(cr) V.5.1.1.1.1.4 Godlevskite, Ni9S8(cr) V.5.1.1.1.1.5 Millerite, (3-NiS V.5.1.1.1.1.6 Polydymite, Ni3S4(cr) V.5.1.1.1.1.7 Vaesite, NiS2(cr) V.5.1.1.1.2 Ni - S phase diagram V.5.1.1.1.2.1 Ni-Smelt V.5.1.1.1.2.2 p,-Ni 3 S 2 andp 2 -Ni 4 S 3 V.5.1.1.1.2.3 Ni3S2(cr) (heazlewoodite) V.5.1.1.1.2.4 Ni7S6(cr) and Ni9S8(cr) V.5.1.1.1.2.5 NiS(a) and NiS(P) V.5.1.1.1.2.6 NiS2(cr), vaesite V.5.1.1.1.2.7 Ni3S4(cr), polydymite V.5.1.1.1.2.8 Ni4.5Fe4.5S8(cr), pentlandite V.5.1.1.1.3 Discussion of selected thermodynamic properties of nickel sulphides V.5.1.1.2 Solubility of NiS(s) and aqueous nickel hydrogen sulphido species V.5.1.2 Nickel sulphates V.5.1.2.1 Aqueous nickel (II) sulphato complexes V.5.1.2.1.1 Enthalpy of formation of NiSO4(aq) V.5.1.2.1.2 Heat capacity values V.5.1.2.2 Solid nickel sulphates V.5.1.2.2.1 NiSO4-7H2O(cr) V.5.1.2.2.1.1 a-NiSO4-6H2O V.5.1.2.2.1.2 Other hydrated nickel sulphate solids V.5.1.2.2.1.3 NiSO4(cr) V.5.2 Selenium compounds and complexes V.5.3 Tellurium compounds and complexes V.5.3.1 Nickel tellurides V.5.3.1.1 NiTeo.775(y,) V.5.3.1.2 NiTeo.85(Y2)
157 157 158 161 161 161 161 161 161 161 161 162 162 162 163 163 166 167 167 168 170 173 174 174 174 177 181 181 188 190 190 190 191 192 193 195 195 195 195 195
CONTENTS
V.6
Group 15 compounds and complexes
V.6.1 Nitrogen compounds and complexes V.6.1.1 Solid nickel nitrates V.6.1.1.1 Ni(NO3)2-9H2O(cr) V.6.1.1.2 Ni(NO3)2'6H2O(cr) V.6.1.1.3 Ni(NO3)24H2O(cr), Ni(NO3)2-3H2O(cr), Ni(NO3)2-2H2O(cr) V.6.1.1.4 Ni(NO3)2(anhydrous) V.6.1.2 Aqueous Ni(II)-nitrato complexes V.6.1.2.1 Determination of the N i 2 + - NO, ion interaction coefficient V.6.2 Phosphorus compounds and complexes V.6.2.1 Solid nickel phosphorus compounds V.6.2.1.1 Nickel phosphides V.6.2.1.2 Nickel phosphates V.6.2.2 Aqueous nickel phosphorus species V.6.2.2.1 Simple nickel phosphato complexes V.6.2.2.2 Aqueous nickel diphosphato complexes V.6.3 Arsenic compounds V.6.3.1 Nickel arsenides V.6.3.1.1 NiAs(cr) V.6.3.1.2 Ni n As 8 (cr) V.6.3.1.3 Ni5As2(cr) V.6.3.1.4 NiAs2(cr) V.6.3.2 Nickel arsenates V.6.3.2.1 Solid nickel arsenates V.6.3.2.2 Aqueous nickel arsenato complexes V.6.3.3 Nickel arsenites V.6.3.3.1 Solid nickel arsenites V.7 Group 14 compounds and complexes
XV11
196 196 196 196 196 198 200 200 203 204 204 204 204 205 205 207 210 210 210 211 212 212 212 212 213 214 214 214
V.7.1 Carbon compounds and complexes 214 V.7.1.1 Nickel carbonates 214 V.7.1.1.1 Solid nickel carbonates 214 V.7.1.1.1.1 NiCO3(cr) 214 V.7.1.1.1.1.1 Crystallography and mineralogy of nickel carbonate 214 V.7.1.1.1.1.2 Heat capacity and entropy of NiCO3(cr) 216 V.7.1.1.1.1.3 Thermodynamic analysis of solubility data on gaspeite ....217 V.7.1.1.1.2 NiCO3-5.5H2O(cr) 218 V.7.1.1.1.2.1 Crystallography and mineralogy of nickel carbonate hydrate 218 V.7.1.1.1.2.2 Heat capacity and entropy of NiCO3-5.5H2O(cr) 219 V.7.1.1.1.2.3 Thermodynamic analysis of solubility data onhellyerite ..219 V.7.1.1.2 Aqueous nickel carbonato complexes 223
xviii
CONTENTS
V.7.1.2 Nickel cyanides V.7.1.2.1 Aqueous Ni(II) cyano complexes V.7.1.2.1.1 Complexes in neutral and alkaline solution V.7.1.2.1.2 Protonated complexes V.7.1.3 Nickel thiocyanates V.7.1.3.1 Aqueous nickel thiocyanato complexes V.7.2 Silicon compounds and complexes V.7.2.1 Solid nickel silicates V.7.2.1.1 Nickel orthosilicate Ni2SiO4(cr) V.7.2.1.1.1 Crystal structure and phase transitions V.7.2.1.1.2 Crystallography and mineralogy of nickel orthosilicate V.7.2.1.1.3 Heat capacity and standard entropy V.7.2.1.1.4 Enthalpy of formation V.7.2.1.1.5 Gibbs energy of formation V.7.3 Germanium compounds and complexes V.7.3.1 Solid nickel germanates V.7.3.1.1 Nickel orthogermanate Ni2GeO4(cr) V.7.3.1.1.1 Crystal structure of nickel orthogermanate V.7.3.1.1.2 Enthalpy of formation from NiO(cr) and GeO2(cr) V.8 Group 13 compounds and complexes V.8.1 Boron compounds and complexes V.8.1.1 Solid nickel borates V.8.1.1.1 Crystal structure V.8.1.1.2 Thermodynamic data for NiO2B 2 0 3 (s) V.8.1.1.3 Solubility of Ni(BO2)2-4H2O(s) V.8.1.2 Aqueous nickel borato complexes
226 226 226 231 231 231 238 238 238 238 239 239 241 243 245 245 245 245 245 245 245 245 245 246 246 247
VI
DISCUSSION OF AUXILIARY DATA SELECTION
249
VI.l
Group 15 auxiliary species
249
VI. 1.1 Additional consistent auxiliary data for As compounds VI.2 Other auxiliary species VI.2.1
Auxiliary data for KCl(cr), KBr(cr) and KI(cr)
Part IV Appendices
249 249 249
251
A
Discussion of selected references
253
B
Ionic strength corrections
443
B.I The specific ion interaction equations B.I.I Background
445 445
CONTENTS
XIX
B.I.2 Ionic strength corrections at temperatures other than 298.15 K B.I.3 Estimation of ion interaction coefficients B.I.3.1 Estimation from mean activity coefficient data B. 1.3.2 Estimations based on experimental values of equilibrium constants at different ionic strengths B.1.4 On the magnitude of ion interaction coefficients B.2 Ion interaction coefficients versus equilibrium constants for ion pairs B.3 Tables of ion interaction coefficients
451 453 453
C
Assigned uncertainties
471
The general problem
471
C.2
Uncertainty estimates in the selected thermodynamic data
473
C.3
One source datum
474
C.4
Two or more independent source data
475
C.I
453 457 458 458
C.4.1 Discrepancies C.5 Several data at different ionic strengths
477 479
C.5.1 Discrepancies or insufficient number of data points C.6 Procedures for data handling C.6.1 Correction to zero ionic strength C.6.2 Propagation of errors C.6.3 Rounding C.6.4 Significant digits
481 483 483 485 486 487
Bibliography
489
List of authors
585
This Page Intentionally Left Blank
List of Figures
Figure II-1: Figure V-1: Figure V-2: Figure V-3: Figure V-4: Figure V-5:
Figure V-6: Figure V-7: Figure V-8: Figure V-9:
Figure V-10: Figure V-11: Figure V-12: Figure V-13: Figure V-14: Figure V-15: Figure V-16: Figure V-17: Figure V-18: Figure V-19: Figure V-20: Figure V-21:
Standard order of arrangement of the elements and compounds based on the periodic classification of the elements 27 The standard molar heat capacity of nickel as a function of temperature 77 Standard electrode potential of Ni | NiSO 4 | Hg 2 SO 4 | Hg at 2 5 ° C . . . . 8 1 Estimation of the Ni 3 + | Ni 2 + standard electrode potential 89 Extrapolation to Im=0 of the experimental data 96 for Reaction (V.23) in NaClO 4 media SIT analysis of the experimental data for Reaction (V.23) in chloride media, including log 1 0 *β1 ,1°((V.23), 298.15 K) 96 obtained in NaClO 4 Extrapolation to Im=0 of the experimental data 98 for Reaction (V.25) in perchlorate media 100 Solubility of Ni(OH)2(s) in alkaline solutions at 298.15 K Experimental results of heat capacity measurements on nickel oxide from 3.2 to 1809.7K 104 Comparison between the temperature dependence of the Gibbs energy of formation for NiO obtained from electrochemical as well as chemical reduction/oxidation equilibrium measurements and the prediction based on the present selection of thermodynamic properties for NiO 106 110 The heat capacity of Ni(OH)2(cr) as a function of temperature 114 Solubility constant of Ni(OH) 2 as a function of temperature β-Ni(OH)2 | β-NiOOH system. Variation of reversible potential 116 with the oxidation state of Ni at 298.15 K and I = 0 α-Ni(OH)2 | γ-NiOOH system. Variation of reversible potential 118 with the oxidation state of Ni at 298.15 K and I = 0 123 Molar heat capacity of anhydrous NiCl 2 between 2 and 30 K 123 Heat capacity of anhydrous NiCl2(cr) 125 A/?^ -function of Reaction (V.59) 126 Third-law analysis of data for Reaction (V.59) 127 Gibbs energy of formation of NiCl2(cr) versus temperature Values from the heat capacity studies of Stuve etal. [78STU/FER] and White and Staveley [82WHI/STA] 133 Extrapolation to I m = 0 of the experimental data 143 for Reaction (V.70) in NaClO 4 media Extrapolation to I m = 0 of the experimental enthalpies 145 for Reaction (V.70) in NaClO 4 media
xxi
XX11
Figure V-22: Figure V-23: Figure V-24: Figure V-25: Figure V-26: Figure V-27: Figure V-28: Figure V-29: Figure V-30: Figure V-31: Figure V-32: Figure V-33: Figure V-34: Figure V-35:
Figure V-36:
Figure V-37: Figure V-38: Figure V-39:
Figure V-40: Figure V-41:
LIST OF FIGURES
Extrapolation to I = 0 of the log 1 0 A values reported in [75LIB/TIA] for Reaction (V.71) in Ni(II) perchlorate solution . . . 149 Extrapolation to I = 0 of the accepted experimental data 150 for Reaction (V.71) in NaClO 4 media Extrapolation to I = 0 of the log 1 0 A values reported in [78LIB/KOW] for Reaction (V.73) in Ni(II) perchlorate media . . . 155 Plot of log 1 0 γ± versus molality of aqueous NiBr 2 solutions at 298.15 K 157 Plot of osmotic coefficient versus molality of aqueous NiBr 2 solutions at 298.15 K 158 Osmotic coefficient plotted as a function of molality 160 of aqueous Ni(ClO 4 ) 2 solutions at 298.15 K Plot of log 1 0 γ± versus molality of aqueous Ni(ClO 4 ) 2 solutions at 298.15 K 160 Phase diagram for the binary system Ni - S with T plotted versus mole fraction of sulphur 164 Section of the calculated phase diagram for the composition range 0.30 < x S < 0.55 165 168 Gibbs energy of reaction for (V.79) plotted versus temperature Partial pressure of sulphur for various phase equilibria in the binary system Ni - S plotted against temperature 170 Gibbs energy function for the stoichiometric composition 173 of NiAs-type Ni(II) sulphide, α-NiS, plotted versus temperature Plot of equilibrium partial pressure of sulphur versus temperature . . . 176 Distribution of nickel sulphide complexes as a function of total molality of nickel (II) in aqueous solutions saturated with H 2 S, T= 298.15 K, I=1.00 mol·kg 1 NaCl 179 Solubility of Ni(II) sulphides in aqueous solutions saturated with H 2 S as a function of initial concentration 181 of hydrochloric acid at 298.15 K The variation of molar conductances of aqueous nickel sulphate 186 solutions with concentration at 25°C Values of log 1 0 K\ for nickel sulphate association at different temperatures 189 Measured water vapour pressures over hydrated Ni(NO 3 ) 2 ·6H 2 O(cr) [37SAN], [72AUF/CAR] and Ni(NO 3 ) 2 ·4H 2 O(cr) [72AUF/CAR2] 199 Extrapolation to I= 0 of the constants for Reaction (V.96) in lithium perchlorate media based on the data in [73FED/SHM]....202 Plot of log 1 0 γ± versus molality of aqueous Ni(NO 3 ) 2 solutions at 298.15 K 203
LIST OF FIGURES
Figure V-42: Figure V-43: Figure V-44: Figure V-45: Figure V-46: Figure V-47: Figure V-48:
Figure V-49: Figure V-50: Figure V-51: Figure V-52: Figure V-53: Figure V-54: Figure A-1: Figure A-2: Figure A-3: Figure A-4: Figure A-5: Figure A-6: Figure A-7: Figure A-8: Figure A-9: Figure A-10: Figure A-11: Figure A-12:
Osmotic coefficient plotted as a function of molality of aqueous Ni(NO 3 ) 2 solutions at 298.15 K Temperature dependence of hellyerite solubility Ionic strength dependence of hellyerite solubility Thermodynamic analysis of hellyerite solubility Three-dimensional predominance diagram for the system Ni 2 + -CO 2 -H 2 O Calculated solubility of NiCO3(cr) in dilute sodium carbonate solutions under atmospheric conditions Extrapolation to I= 0 of experimental formation constants of Ni(CN)24 (V.1 13) determined in NaClO 4 medium (insert shows the corresponding plot of data reported for KClO 4 medium) Extrapolation to I = 0 of experimental equilibrium constants for Reaction (V.1 14) determined in NaClO 4 medium Extrapolation to Im=0 of the accepted formation constants + reported for the formation of NiSCN in perchlorate media Extrapolation to Im=0 of the accepted formation constants reported for the formation of Ni(SCN)2(aq) in perchlorate media Extrapolation to Im=0 of the accepted formation constants reported for the formation of Ni(SCN)–3 in perchlorate media The standard molar heat capacity of Ni2SiO4-olivine as a function of temperature Standard Gibbs energy of Reaction (V. 121) as a function of temperature Solubility of precipitated nickel carbonate High temperature enthalpy of anhydrous NiCl 2 Plot ofE(vs. SHE) versus log10([NiSO4]/m) Extrapolation to I m = 0 (using the SIT) of the formation constants for the species NiSCN+ as reported in [58YAT/KOR] Extrapolation of activity coefficients to saturated solution at 25 ° C . . . Extrapolation of osmotic coefficients to saturated solution at 25 °C .. Plot of log 1 0 β1 (A.17) reported in [61LIS/WIL] vs.1/T log10 β1 Im p l o t obtained from the potentiometric data (t=35 °C) 2+ SCN system in [61MOH/DAS] reported for the Ni Solubility product of Ni(BO 2 ) 2 ·4H 2 O(s) at 22°C from data reported in [61SHC] Nickel and cobalt borate complexes from data in [61SHC] Temperature dependence of log10 *ft j values derived from [63BOL/JAU] Temperature dependence of log 1 0 *β 4 , 4 values derived from [63BOL/JAU]
xxiii
204 220 221 222 223 225
229 229 235 236 236 240 244 268 274 276 290 295 295 298 299 300 301 306 307
XXIV
Figure A-13: Figure A-14: Figure A-15:
Figure A-16: Figure A-17: Figure A-18:
Figure A-19:
Figure A-20:
Figure A-21:
Figure A-22: Figure A-23: Figure A-24: Figure A-25:
Figure A-26:
Figure A-27: Figure A-28: Figure A-29: Figure A-30: Figure A-31:
LIST OF FIGURES
Extrapolation to I = 0 of the equilibrium constants log 1 0 *β1,1 ( A . 2 3 ) reported in [64PER] at 20°C, using the SIT Plot of ln *β°1,1 (A.23) recalculated from [64PER] vs.1/T Temperature dependence of the association constant – 3 for the reaction Ni(CN)42–+ CN ^ Ni(CN)5 – reported in [60MCC/JON] Molar enthalpy of solution of NiSO4·6H2O(cr) at 25°C + Temperature dependence of log 1 0 β1 values of NiSCN complex reported in [68MAL/TUR] Heats of reaction determined from the calorimetric data of [69IZA/EAT] by using the 298.15 K values of log 10 K1 from [73KAT] + Extrapolation to I = 0 of the formation constant of NiCl using Equation (A.40) and the experimental data published in [70HAL/VAN] Temperature dependence of log 1 0 β1 and log 1 0 β2 values for the Ni(II) - (SCN) system (I=0.2 M) as reported in [71TUR/MAL] Heats of dilution of hydrated nickel nitrate salts based on the experimental heats of solution and an assumed heat 1 of dilution to infinite dilution of 520 J·mol for a solution with a solvent: salt ratio of 8000:1 Experimental [73FED/SHM] and calculated solubility of nickel(II) in the aqueous phase with increasing nitrate concentrations Experimental [73FED/SHM] and calculated values of the solubility of Ni(IO 3 ) 2 ·3H 2 O in aqueous LiClO 4 solutions Variation of formation constants from Perlmutter-Hayman and Secco [73PER/SEC] with ionic strength Temperature dependence of log 1 0 K2 for the reaction Ni(CN)2–4+ H + ^ Ni(HCN)(CN)–3 at different ionic strengths using NaCl Extrapolation to I= 0 of experimental equilibrium constants + for the reaction Ni(HCN)(CN)3– + H ^ Ni(HCN)2 (CN) 2 (aq) at different temperatures rH°m(298 K) of reaction of B 2 O 3 (cr) with MO or M 2 O Heat capacity of NiS 2 plotted versus temperature SIT-plot of recalculated data from [80MIL/BUG], [65BUR/LIL2] and [71OHT/BIE], and the selected log 1 0 * / % ((A.70), 298.15 K) . . . Solubility of NiO in acidic solutions as a function of initial molality of HCl Solubility of NiO in basic solutions as a function of initial molality of NaOH
313 314
320 325 333
336
343
351
352 358 358 363
367
367 370 382 391 394 395
LIST OF FIGURES
Figure A-32: Figure A-33:
Figure A-34: Figure A-35:
Figure A-36: Figure A-37: Figure A-38: Figure A-39:
Figure A-40: Figure A-41: Figure B-1:
xxv
Values of Cp,m (NiBr2, cr) as a function of temperature as reported by White and Staveley [82WHI/STA] 400 Plot of logarithm of the oxygen partial pressure versus temperature for the phase equilibria α-NiS / β 2 -Ni 4 S 3 and β 2 -Ni 4 S 3 / NiO, respectively, a t p S O = 1 atm 407 Logarithm of the oxygen partial pressure plotted as a function of temperature for the equilibrium NiS(α) / NiO atp(SO 2 ) = 1 atm....408 Plot of logarithm of the oxygen partial pressure versus temperature for the phase equilibria NiS(α) / β 2 -Ni 4 S 3 and β 2 -Ni 4 S 3 / NiO, 419 respectively, a t p S O = 1 atm Heat capacity data for Ni 3 S 2 as a function of temperature 421 Heat capacity of "Ni 7 S 6 " consisting of a mixture of Ni 3 S 2 and Ni 9 S 8 as a function of temperature 427 Heat capacity of NiS as a function of temperature 429 Logarithm of stability constants of nickel bisulphide complexes in seawater (NaCl solutions) plus the Debye-Hückel term for ionic strength correction plotted as a function of ionic strength,T= 298.15 K 432 Solubility of theophrastite as determined by the pH variation method 440 Ionic strength dependence of theophrastite solubility 440 Plot of log 10 β1,m +4D versus Im for reaction (B.12), at 25°C and 1 bar 455
This Page Intentionally Left Blank
List of Tables
Table II-1: Table II-2: Table II-3: Table II-4: Table II-5: Table II-6: Table II-7: Table III-1: Table III-2: Table III-3: Table IV-1: Table IV-2: Table V-1: Table V-2: Table V-3: Table V-4: Table V-5: Table V-6: Table V-7: Table V-8: Table V-9: Table V-10: Table V-11: Table V-12: Table V-13: Table V-14: Table V-15:
Abbreviations for experimental methods 11 Symbols and terminology 13 Abbreviations used as subscripts of to denote the type of chemical process 17 Unit conversion factors 28 Factors g for the conversion of molarity, c B , to molality, mB, of a substance B, in various media at 298.15 K 30 Reference states for some elements at the reference temperature of 298.15 K and standard pressure of 0.1 MPa 31 Fundamental physical constants 36 Selected thermodynamic data for nickel compounds and complexes... 44 Selected thermodynamic data for reaction involving nickel compounds and complexes 48 Selected temperature coefficients for heat capacities of nickel 52 Selected thermodynamic data for auxiliary compounds and complexes 55 Selected thermodynamic data for reactions involving auxiliary compounds and complexes 68 The temperature coefficients of the heat capacity function for Ni(cr) 78 Enthalpy of Reaction (V.3) 83 Partial molar heat capacity values for Ni(II) salts in aqueous solution at 298.15 K 88 Experimental equilibrium data (logarithmic values) for the hydrolysis of Ni(II) in acidic or neutral solutions 93 Smoothed Cp°,m data and calculated values for the entropy of N i O . . . 103 Literature data for log 10 *K°s,0 (Ni(OH)2) 113 Thermodynamic properties of Ni(OH)2(cr) 115 Experimental stability constants (logarithmic values) of the species NiF+ 142 Experimental enthalpy values for the Reaction (V.70) 144 Experimental formation constants for the NiCl+ species 147 Experimental equilibrium constants for the Reaction (V.72) 152 Reaction enthalpies reported for Reaction (V.71) 153 Experimental formation constants of the NiBr+ complex 154 Reaction enthalpies reported for Reaction (V.73) 156 Gibbs energy functions for the binary system Ni - S relative to metallic nickel, Ni(s), and gaseous sulphur, S2(g) 165 xxvii
XXV111
LIST OF TABLES
Table V-16: Table V-17: Table Table Table Table Table
V-18: V-19: V-20: V-21: V-22:
Table V-23: Table V-24: Table V-25:
Table V-26:
Table V-27: Table V-28:
Table V-29:
Table V-30: Table V-31: Table V-32: Table V-33: Table VI-1: Table Table Table Table Table Table Table
A-1: A-2: A-3: A-4: A-5: A-6: A-7:
Table A-8:
Temperatures and enthalpies of transition for β-NiS ==^ α-NiS Temperatures for invariant three-phase equilibria in the binary system Ni - S Heat capacity functions of nickel sulphides Standard enthalpy of formation of nickel sulphides at 298.15 K Standard entropy of nickel sulphides at 298.15 K Heat capacity of nickel sulphides at 298.15 K Recalculated stability constants and SIT parameters for nickel sulphide complexes at T=298.15 K and I = 0 Solubility constants for Ni(II) sulphide at T=298.15 K and I = 0 Experimental values for the association constants and other thermodynamic quantities for nickel sulphate Results of reanalysis of NiSO4(aq) conductance data using a form of the Lee-Wheaton equation that is compatible with the SIT procedure Experimental values for the first association constant of nickel sulphate determined in the presence of supporting electrolytes Formation constants of the NiNO3+ complex (ion pair) Experimental values for the association constants and other thermodynamic quantities for nickel phosphate complexes Experimental values for the association constants and other thermodynamic quantities for nickel pyrophosphate complexes Experimental equilibrium data for the Ni(II) cyanide system Experimental enthalpy values for the Reaction (V.1 13) and (V.114) Experimental equilibrium data for the Ni(II) - thiocyanate system . . . Experimental reaction enthalpy values for the formation of Ni(II)-thiocyanate complexes Enthalpy of formation values for KCl(cr), KBr(cr), andKI(cr) at298.15K Equilibrium constants of Reaction (A.3) [21BER/CRU], [24CRU]... Equilibrium constants of NiCl2(cr) + Η2(g) = ^ Νi(cr) + 2 ΗCl(g) . . . Electrode potential of the cell: Ni|NiSO4(m)|Hg2SO4|Hg Standard electrode potential of Ni 2 + |Ni at 25 °C Equilibrium constants of Reaction (A.9) Equilibrium constants of Reaction (A.9) [53SAN] Enthalpy of formation values calculated from equilibrium gas compositions Equilibrium constants of Reaction (A.10) [54SHC/TOL]
172 175 176 177 177 177 178 180 183
185
187 201
206
209 228 230 233 237 250 258 260 262 276 278 279 282 282
LIST OF TABLES
Table Table Table Table
A-9: A-10: A-11: A-12:
Table A-13: Table A-14: Table A-15: Table A-16: Table A-17: Table A-18: Table A-19: Table Table Table Table Table Table Table
A-20: A-21: A-22: A-23: A-24: A-25: A-26:
Table A-27:
Table A-28: Table A-29: Table B-1:
Table B-2:
Table B-3:
Table B-4:
Enthalpy of solution of NiCl2(cr) in H2O(1) at 298.15 K [58MUL]... log 1 0 β1 versus Im Recalculated constants Comparison between experimental data [63LIN/LAF] and calculated values of log10[p(H2S/p(H2)] for the coexistence of N i 3 + x S 2 with metallic nickel Reported and recalculated log 1 0 *β°1,1 Dataset Experimental results for the enthalpy of formation for various nickel sulphides determined by calorimetric measurements Gibbs energy functions for various nickel sulphides Auxiliary data used in the recalculation of NaBr heat of formation values with the data from [78STU/FER] Corrected formation constants Stability constants for the complexes NiOH+, Ni(OH) 2 andNi(OH)–3 Stability constants log 1 0 *β3,°1forNi(OH)–3, I = 0 Enthalpies and entropies of reaction from Skeaff et al Enthalpies of formation for solid nickel sulphides at 298.15 K Enthalpy of solution of NiCl 2 in H 2 O at 298.15 K [90EFI/FUR] fHm° (NiCO 3 , 298.15 K) from decomposition data [91TAR/FAZ]... Enthalpy of solution of NiCl 2 in H 2 O at 298.15 K [95MAN/KOR]... Recommended formation constants (*/3° m) and reaction enthalpies for the hydroxo complexes formed in the acidic region, at 298.15 K, mNi 2 + + nH2O(1) ^ Nim(OH)2nm-n + nH + Recommended equilibrium constants ( * K ° n m ) and reaction enthalpies for the hydroxo complexes formed in the alkaline region, at 298.15 K, mNi(OH)2(cr) + nOH – Nim(OH)–2nm+n Recommended ion interaction coefficients ε(j, k) Standard thermodynamic functions of the aqueous species Water activities a H O at 2968.15 K for the most common ionic media at various concentrations applying Pitzer's ion interaction approach and the interaction parameters given in [91PIT] Debye-Hückel constants as a function of temperature at a pressure of 1 bar below 100°C and at the steam saturated pressure for t 100°C The preparation of the experimental equilibrium constants for the extrapolation to I = 0 with the specific ion interaction method at 25°C and 1 bar, according to reaction (B.12) Ion interaction coefficients ε(j, k) (kg·mol 1 ) for cations j with k = Cl–, ClO–4 and NO 3 , taken from Ciavatta [80CIA], [88CIA] unless indicated otherwise
XXIX
289 304 306
308 314 342 345 385 386 390 393 395 404 406 417 422 430
435
435 436 436
448
452
454
460
XXX
Table B-5:
Table B-6:
Table B-7: Table C-1:
LIST OF TABLES
Ion interaction coefficients ε(j, k) (kg·mol 1) for cations j with k=Li, Na and K, taken from Ciavatta [80CIA], [88CIA] unless indicated otherwise Ion interaction coefficients ε(1,j, k) and ε(2,j, k) for cations 7 with k = Cl–, ClO4– and NO–3 (first part), and for anions + + + 7 with k=Li , Na and K (second part), according to the relationship ε=ε1 + ε 2 log 1 0 I m . The data are taken from Ciavatta [80CIA], [88CIA] SIT interaction coefficient ε(j, k) kg·mol 1 for neutral species, j , with k, electroneutral combination of ions Details of the calculation of equilibrium constant corrected to I = 0, using (C.19)
467
470 470 483
Parti Introductory material
This Page Intentionally Left Blank
Chapter I
Introduction
I.I Background The modelling of the behaviour of hazardous materials under environmental conditions is among the most important applications of natural and technical sciences for the protection of the environment. In order to assess, for example, the safety of a waste deposit, it is essential to be able to predict the eventual dispersion of its hazardous components in the environment (geosphere, biosphere). For hazardous materials stored in the ground or in geological formations, the most probable transport medium is the aqueous phase. An important factor is therefore the quantitative prediction of the reactions that are likely to occur between hazardous waste dissolved or suspended in ground water, and the surrounding rock material, in order to estimate the quantities of waste that can be transported in the aqueous phase. It is thus essential to know the relative stabilities of the compounds and complexes that may form under the relevant conditions. This information is often provided by speciation calculations using chemical thermodynamic data. The local conditions, such as ground water and rock composition or temperature, may not be constant along the migration paths of hazardous materials, and fundamental thermodynamic data are the indispensable basis for dynamic modelling of the chemical behaviour of hazardous waste components. In the field of radioactive waste management, the hazardous material consists to a large extent of actinides and fission and activation products. Two isotopes of nickel (59Ni and 63Ni) pertain to the latter category. The scientific literature on thermodynamic data, mainly on equilibrium constants and redox potentials in aqueous solution, has been contradictory in a number of cases, especially in actinide chemistry. A critical and comprehensive review of the available literature is necessary in order to establish a reliable thermochemical database that fulfils the requirements for rigorous modelling of the behaviour of the actinides, fission and activation products in the environment.
3
4
1 Introduction
The International Atomic Energy Agency (IAEA) in Vienna published special issues with compilations of physicochemical properties of compounds and alloys of elements important in reactor technology: Pu, Nb, Ta, Be, Th, Zr, Mo, Hf and Ti between 1966 and 1983. In 1976, IAEA also started the publication of the series "The Chemical Thermodynamics of Actinide Elements and Compounds", oriented towards nuclear engineers and scientists. This international effort has resulted in the publication of several volumes, each concerning the thermodynamic properties of a given type of compounds for the entire actinide series. These reviews cover the literature approximately up to 1984. The latest volume in this series appeared in 1992, under Part 12: The Actinide Aqueous Inorganic Complexes [92FUG/KHO]. Unfortunately, data of importance for radioactive waste management (for example, Part 10: The Actinide Oxides, or data on fission and activation products) is lacking in the IAEA series. The Radioactive Waste Management Committee (RWMC) of the OECD Nuclear Energy Agency recognised the need for an internationally acknowledged, highquality thermochemical database for application in the safety assessment of radioactive waste disposal, and undertook the development of the NEA Thermochemical Data Base (TDB) project [85MUL], [88WAN], [91 WAN]. The RWMC assigned a high priority to the critical review of relevant chemical thermodynamic data of compounds and complexes for this area containing the actinides uranium, neptunium, plutonium and americium, as well as the fission product technetium. The first four books in this series on the chemical thermodynamics of uranium [92GRE/FUG], americium [95SIL/BID], technetium [99RAR/RAN] and neptunium and plutonium [2001LEM/FUG] originated from this initiative. Simultaneously with the NEA's TDB project, other reviews on the physical and chemical properties of actinides appeared, including the book by Cordfunke et al. [90COR/KON2], the series edited by Freeman et al. [84FRE/LAN], [85FRE/LAN], [85FRE/KEL], [86FRE/KEL], [87FRE/LAN], [91FRE/KEL], the two volumes edited by Katz et al. [86KAT/SEA], and Part 12 by Fuger et al. [92FUG/KHO] within the IAEA review series mentioned above. In 1998, Phase II of the TDB Project (TDB-II) was started to provide for the further needs of the radioactive waste management programs by updating the existing database and applying the TDB review methodology to other elements (occurring in waste as fission or activation products) and to simple organic complexes. In TDB-II the overall objectives are set by a Management Board, integrated by representatives of 17 organisations from the field of radioactive waste management. These participating organisations, together with the NEA, provide financial support for TDB-II. The TDB-II Management Board is assisted in technical matters by a group of experts in chemical thermodynamics (the Executive Group). The NEA acts in this phase as Project Coordinator ensuring the implementation of the Project Guidelines and liaising with the Review Teams. The present volume, the sixth in the series, is the second one to be published within this second phase of the TDB Project. A previous volume dealt with an update on the chemical thermodynamics of uranium, neptunium, plutonium, americium
1.2 Focus of the review
5
and technetium [2003GUI/FAN]. Three additional volumes are in preparation as this book goes to press dealing with the inorganic species and compounds of selenium, the inorganic species and compounds of zirconium and with compounds and complexes formed by the organic ligands oxalate, citrate, EDTA and iso-saccharinic acid with U, Np, Pu, Am, Tc, Zr, Ni and Se.
1.2 Focus of the review The first and most important step in the modelling of chemical reactions is to decide whether they are controlled by chemical thermodynamics or kinetics, or possibly by a combination of the two. This also applies to the modelling of more complex chemical systems and processes, such as waste repositories of various kinds, the processes describing transport of toxic materials in ground and surface water systems, the global geochemical cycles, etc. As outlined in the previous section, the focus of the critical review presented in this report is on the thermodynamic data of nickel relevant to the safety assessment of radioactive waste repositories in the geosphere. This includes the release of waste components from the repository into the geosphere (i.e., its interaction with the waste container and the other near-field materials) and their migration through the geological formations and the various compartments of the biosphere. As ground waters and pore waters are the transport media for the waste components, the knowledge of the thermodynamics of the corresponding elements in waters of various compositions is of fundamental importance. The present review therefore puts much weight on the assessment of the lowtemperature thermodynamics of nickel in aqueous solution and makes independent analyses of the available literature in this area. The standard method used for the analysis of ionic interactions between components dissolved in water (see Appendix B) allows the general and consistent use of the selected data for modelling purposes, regardless of the type and composition of the ground water, within the ionic strength limits given by the experimental data used for the data analyses in the present review. The interactions between solid compounds, such as the rock materials, and the aqueous solution and its components are as important as the interactions within the aqueous solution, because the solid materials in the geosphere control the chemistry of the ground water, and they also contribute to the overall solubilities of key elements. The present review therefore also considers the chemical behaviour of solid compounds that are likely to occur, or to be formed, under geological and environmental conditions. It is, however, difficult to assess the relative importance of the solid phases for performance assessment purposes, particularly since their interactions with the aqueous phase are in many cases known to be subject to quantitatively unknown kinetic constraints. Furthermore, in some circumstances sorption of aqueous ions at mineral-water interfaces may be a more important factor in determining migration of nickel than dissolu-
6
1 Introduction
tion and precipitation phenomena. This book contains a summary and a critical review of the thermodynamic data on compounds and complexes containing nickel (cf. Chapter V), as reported in the available chemical literature up to mid-2002, but a few more recent references are also included. A comparatively large number of primary references are discussed separately in Appendix A. Although the focus of this review is on nickel, it is necessary to use data on a number of other species during the evaluation process that lead to the recommended data. These so-called auxiliary data are taken both from the publication of CODATA Key Values [89COX/WAG] and from the evaluation of additional auxiliary data in the uranium and other volumes of this series [92GRE/FUG], [99RAR/RAN], [2003GUI/FAN], [2005OLI/NOL] and their use is recommended by this review. Care has been taken that all the selected thermodynamic data at standard conditions (cf. Section II.3) and 298.15 K are internally consistent. For this purpose, special software has been developed at the NEA Data Bank that is operational in conjunction with the NEATDB data base system, cf. Section II.6. In order to maintain consistency in the application of the values selected by this review, it is essential to use these auxiliary data when calculating equilibrium constants involving nickel compounds and complexes. This review does not include any compounds and complexes containing organic ligands or species in non-aqueous solvents. Nickel alloy systems have not been reviewed and treatment of gaseous species is very limited.
1.3 Review procedure and results The objective of the present review is the critical compilation of a database for the inorganic species of nickel. This aim is achieved by an assessment of all sources of thermodynamic data published until mid-2002. This assessment is performed in order to decide on the most reliable values that can be recommended. Experimental measurements published in the scientific literature are the main source for the selection of recommended data. Previous reviews are not neglected, but form a valuable source of critical information on the quality of primary publications. When necessary, experimental source data are re-evaluated by using chemical models that are either found to be more realistic than those used by the original author, or are consistent with side-reactions discussed in another section of the review (for example, data on carbonate and hydroxide solubilities might need to be re-interpreted to take into account crystal structure and particle size of the phases actually investigated). Re-evaluation of literature values might be also necessary to correct for known systematic errors (for example, if the junction potentials are neglected in the original publication) or to make extrapolations to standard state conditions (/ = 0) by using the specific ion interaction (SIT) equations (cf. Appendix B). For convenience, these SIT equations are referred to in some places in the text as "the SIT". In order to ensure that consistent
/. 3 Review procedure and results
1
procedures are used for the evaluation of primary data, a number of guidelines have been developed. They have been updated and improved since 1987, and their most recent versions are available at the NEA [99WAN], [99WAN/OST], [2000OST/WAN], [2000GRE/WAN], [2000WAN/OST]. Some of these procedures are also outlined in this volume, cf. Chapter II, Appendix B, and Appendix C. Parts of these sections, which were also published in earlier volumes [92GRE/FUG], [95SIL/BID], [99RAR/RAN], [2001LEM/FUG], [2003 GUI/FAN] have been revised in this review. Once the critical review process in the NEA TDB project is completed, the resulting manuscript is reviewed independently by qualified experts nominated by the NEA. The independent peer review is performed according to the procedures outlined in the TDB-6 guideline [99WAN]. The purpose of the additional peer review is to receive an independent view of the judgements and assessments made by the primary reviewers, to verify assumptions, results and conclusions, and to check whether the relevant literature has been exhaustively considered. The independent peer review is performed by persons having technical expertise in the subject matter to be reviewed, to a degree at least equivalent to that needed for the original review. The thermodynamic data selected in the present review (see Chapters III and IV) refer to the reference temperature of 298.15 K. and to standard conditions, cf. Section II.3. For the modelling of real systems it is, in general, necessary to recalculate the standard thermodynamic data to non-standard state conditions. For aqueous species a procedure for the calculation of the activity factors is thus required. This review uses the approximate specific ion interaction method (SIT) for the extrapolation of experimental data to the standard state in the data evaluation process, and in some cases this requires the re-evaluation of original experimental values (solubilities, emf data, etc.). For maximum consistency, this method, as described in Appendix B, should always be used in conjunction with the selected data presented in this review. The thermodynamic data selected in this review are provided with uncertainties representing the 95% confidence level. As discussed in Appendix C, there is no unique way to assign uncertainties, and the assignments made in this review are to a large extent based on the subjective choice by the reviewers, supported by their scientific and technical experience in the corresponding area. The quality of thermodynamic models cannot be better than the quality of the data on which they are based. The quality aspect includes both the numerical values of the thermodynamic data used in the model and the "completeness" of the chemical model used, e.g., the inclusion of all the relevant dissolved chemical species and solid phases. For the user it is important to consider that the selected data set presented in this review (Chapters III and IV) is certainly not "complete" with respect to all the conceivable systems and conditions; there are gaps in the information. The gaps are pointed out in the various sections of Part III, and this information may be used as a basis for the
8
1 Introduction
assignment of research priorities. The following example for a rewarding project in experimental thermodynamics must suffice: Precise measurements of the enthalpies of dissolution in aqueous media according to the reactions: NiX2 (cr) ^ Ni2+ + 2 X"
(where X = Cl, Br, I)
and NiCl2-6H2O (cr) ^ Ni2+ + 2 (X + 6 H2O(1) as well as low temperature heat capacity measurements on NiCl2-6H2O(cr) would lead to the important key values &fH°m (Ni2+, 298.15K) and 5° (Ni2+, 298.15K) by thermodynamic cycles. The results of experimental work done on formation of weak complexes are often ambiguous. In the TDB assessments, the intent is to provide an interpretation that is consistent with the experimental data. This is not necessarily a unique interpretation, but must be one that can be used within the SIT framework. It can be an interpretation with or without an association constant, but it must be clear which is to be used. We have attempted to state the assumptions at each stage so that a reader can: • reconstruct the arguments, • reconstruct the calculations and results, • with the final selected values (Tables III-1 and III-2) and interaction coefficients reported in Tables B-4 and B-5, be able to regenerate the original experimental data within the uncertainties of the selected values. In some cases, an association constant and related interaction coefficients have been used where it might have been possible to "make do" with a different set of interaction coefficients. We make no apologies for this so long as the calculation procedure is well defined and consistent. For example, in several hydrolysis studies, allowance is made for complexation of Ni2+ with chloride, using an association constant, the calculated activity of water. An interaction coefficient of NiOH+ with Cl" is then derived as part of the same calculation that defines a value for the first hydrolysis constant for Ni2+. Conversely, association between Ni2+ and Cl could have been neglected, an interaction coefficient for Ni2+ with Cl" then would have been specified, along with a different interaction coefficient for NiOH+ with Cl". In theory, there should be no effect on the value of the selected hydrolysis constant, K°, at / = 0.
/. 3 Review procedure and results
9
Our inherent assumptions about weak complexes can be stated as follows: • A weak complex forms between Ni2+ and some singly charged anion X~, • It is assumed that there is no complexation between Ni2+ and CIO4, • If a weak complex forms between Ni2+ and X", the interaction coefficient s(Ni2+, X~) is arbitrarily set to a "reasonable value", in some cases a value similar or identical to the interaction coefficient e(Ni2+, C1O4), • All other interaction between Ni2+ and X" is accounted for by the association constant (K°= a(NiX+) / a(Ni2+) a(X~)) and the interaction coefficient e(NiX+, X~) (and, if necessary, e(NiX+, CIO 4) also is specified). The e(Ni2+,X~) values are unlikely to be identical, and estimated uncertainties have been increased for values obtained from experiments in which a considerable part of the background electrolyte was replaced by the complex-forming anion. In some cases it has been necessary to make very arbitrary assumptions in generation of association constants for weak complexes. If used consistently, these should have little or no impact on geochemical modelling using the TDB database, or on regeneration of the experimental values. When the experimental data are sparse, or the experiments were not well designed, the estimated uncertainties in some association constant values are large. If we omitted these association constants, and only generated interaction coefficients, the propagated uncertainties would not be reduced. They would simply reappear in a different guise. In most cases this does not matter whatsoever, and a good argument can be made for eliminating the extraneous constants. However, modellers have found that when using a properly assembled database, there is only one thing worse than a species that has been included with a large uncertainty, and that is a species, for which there is qualitative or semi-quantitative evidence, but for which the database compiler has declined to propose a value. Therefore, there is a tendency for modellers to augment any database with extra values. In our view, it is better to supply association constants (with their inherent large uncertainties) than to allow the introduction of inconsistent "secondary" values.
This Page Intentionally Left Blank
Chapter II
Standards, Conventions, and Contents of the Tables This chapter outlines and lists the symbols, terminology and nomenclature, the units and conversion factors, the order of formulae, the standard conditions, and the fundamental physical constants used in this volume. They are derived from international standards and have been specially adjusted for the TDB publications.
II.l Symbols, terminology and nomenclature II.1.1 Abbreviations Abbreviations are mainly used in tables where space is limited. Abbreviations for methods of measurement are listed in Table II-1. Table II-1: Abbreviations for experimental methods
AES
Anion exchange Atomic Emission Spectroscopy
CAL
Calorimetry
CUR
Chromatography
C1X
Cation exchange
COL
Colorimetry
CON
Conductivity
A1X
COR
Corrected
COU
Coulometry
CRY
Cryoscopy
DIS
Distribution between two phases
DSC
Differential Scanning Calorimetry
DTA
Differential Thermal Analysis
EDS
Energy Dispersive Spectroscopy
EM
Electromigration
(Continued on next page)
11
12
II. Standards, Conventions, and Contents of the Tables Table II—1: (continued) EMF
Electromotive force, not specified
EPMA
Electron Probe Micro Analysis
EXAFS
Extended X-ray Absorption Fine Structure
FTIR
Fourier Transform Infra Red
IDMS
Isotope Dilution Mass-Spectroscopy
IR
Infrared
GL
Glass electrode
1SE-X
Ion selective electrode with ion X stated
ix
Ion exchange
KIN
Rate of reaction
UBD
Laser Induced Break Down
MVD
Molar Volume Determination
NMR
Nuclear Magnetic Resonance
PAS
Photo Acoustic Spectroscopy
POL
Polarography
POT
Potentiometry
PRX
Proton relaxation
QII
Quinhydrone electrode
RED
Emf with redox electrode
REV
Review
SEM
Scanning Electron Microscopy
SP
Spectrophotometry
SOL
Solubility
TC
Transient Conductivity
TGA
Thermo Gravimetric Analysis
TLS
Thermal Lensing Spectrophotometry
TRLFS
Time Resolved Laser Fluorescence Spectroscopy
uv
Ultraviolet
VLT
Voltammetry
XANES
X-ray Absorption Near Edge Structure
XRD
X-ray Diffraction
?
Method unknown to the reviewers
Other abbreviations may also be used in tables, such as SHE for the standard hydrogen electrode or SCE for the saturated calomel electrode. The abbreviation NHE has been widely used for the "normal hydrogen electrode", which is by definition identical to the SHE. It should nevertheless be noted that NHE customarily refers to a standard state pressure of 1 atm, whereas SHE always refers to a standard state pressure of 0.1 MPa (1 bar) in this review.
//. / Symbols, terminology and nomenclature
13
II. 1.2 Symbols and terminology The symbols for physical and chemical quantities used in the TDB review follow the recommendations of the International Union of Pure and Applied Chemistry, IUPAC [79WHI], [93MIL/CVI]. They are summarised in Table II-2. Table II-2: Symbols and terminology. Symbols and terminology length
/
height
h
radius
r
diameter
d
volume
V
mass
m
density (mass divided by volume)
P
time
t
frequency
V
wavelength internal transmittance (transmittance of the medium itself, disregarding boundary or container influence)
k Ti
internal transmission density, (decadic absorbance): Iogi0(l/Tj)
A
molar (decadic) absorption coefficient: A1 cj
£
relaxation time
T
Avogadro constant
NA
relative molecular mass of a substance1"'
M,
thermodynamic temperature, absolute temperature
T
Celsius temperature
t
(molar) gas constant
R
Boltzmann constant
k
Faraday constant
F
(molar) entropy
Sm
(molar) heat capacity at constant pressure
C,m
(molar) enthalpy
H
(molar) Gibbs energy
* Gm
chemical potential of substance B
/*
pressure
P
partial pressure of substance B: x&p
PB
fugacity of substance B
h (Continued next page)
14
//. Standards, Conventions, and Contents of the Tables Table II-2: (continued)
Symbols and terminology fugacity coefficient:/B/pB
/is
amount of substance"1'
n
mole fraction of substance B
XB
molarity or concentration of a solute substance B (amount of B divided by the volume of the solution)
cB, [B]
lc)
molality of a solute substance B (amount of B divided by the mass of the
m%
solvent) |d) factor for the conversion of molarity to molality of a solution: WB/CB ...
( V ++V )
(ci
mean ionic molality , m+
V+
Q
V
= m+ m_
m±
activity of substance B
a&
activity coefficient, molality basis: aB / mB
YB
activity coefficient, concentration basis: alt lcH
y&
• •
(el
(v^+v )
mean ionic activity , a+
v_ v
- oB = a+ a_
mean ionic activity coefficient|c), y+v*
%
= y"* y"
osmotic coefficient, molality basis ionic strength: / =— Z,"J2- o r / = — Z.c-z "' 2 2 ' SIT ion interaction coefficient between substance B| and substance B2 stoichiometric coefficient of substance B (negative for reactants, positive for
a± y± tj> / E(BJ, B2) vK
products) genera] equation for a chemical reaction equilibrium constant
(()
charge number of an ion B (positive for cations, negative for anions)
0 = XB VBB K ZB
charge number of a cell reaction
n
electromotive force
E
p H = - l o g > H , /(mol-kg-')]
pH
electrolytic conductivity
K
superscript for standard state18'
°
(a)
ratio of the average mass per formula unit of a substance to -^ of the mass of an atom of nuclide I2C.
(b)
cf. Sections 1.2 and 3.6 of the IUPAC manual [79WHI].
(c)
This quantity is called "amount-of-substance concentration" in the IUPAC manual [79WHI]. A solution with a concentration equal to 0.1 mol • dm ' is called a 0.1 molar solution or a 0.1 M solution.
(d) (e)
A solution having a molality equal to 0.1 mol • kg ' is called a 0.1 molal solution or a 0.1 m solution. For an electrolyte Nv Xv which dissociates into v + ( = v + + v _ ) ions, in an aqueous solution with molality m, the individual cationic molality and activity coefficient are m+(=vtm) and y v (= a+ I m+) . A similar definition is used for the anionic symbols. Electrical neutrality requires that v + z + = v_z_ .
(f)
Special notations for equilibrium constants are outlined in Section II. 1.6. In some cases, Kc is used to indicate a constant in molar units, and Km a constant in molal units.
(g)
See Section II.3.1.
II. 1 Symbols, terminology and nomenclature
15
11.1.3 Chemical formulae and nomenclature This review follows the recommendations made by IUPAC [71JEN], [77FER], [90LEI] on the nomenclature of inorganic compounds and complexes, except for the following items: •
The formulae of coordination compounds and complexes are not enclosed in square brackets [71 JEN] (Rule 7.21). Exceptions are made in cases where square brackets are required to distinguish between coordinated and uncoordinated ligands.
•
The prefixes "oxy-" and "hydroxy-" are retained if used in a general way, e.g., "gaseous uranium oxyfluorides". For specific formula names, however, the IUPAC recommended citation [71JEN] (Rule 6.42) is used, e.g., "uranium(IV) difluoride oxide" for UF2O(cr).
An IUPAC rule that is often not followed by many authors [71JEN] (Rules 2.163 and 7.21) is recalled here: the order of arranging ligands in coordination compounds and complexes is the following: central atom first, followed by ionic ligands and then by the neutral ligands. If there is more than one ionic or neutral ligand, the alphabetical order of the symbols of the ligating atoms determines the sequence of the ligands. For example, (UO2)2CO3(OH)3 is standard, (UO2)2(OH)3CO3 is nonstandard and is not used. Abbreviations of names for organic ligands appear sometimes in formulae. Following the recommendations by IUPAC, lower case letters are used, and if necessary, the ligand abbreviation is enclosed within parentheses. Hydrogen atoms that can be replaced by the metal atom are shown in the abbreviation with an upper case "H", for example: H3edta~, Am(Hedta)(s) (where edta stands for ethylenediaminetetraacetate).
11.1.4 Phase designators Chemical formulae may refer to different chemical species and are often required to be specified more clearly in order to avoid ambiguities. For example, UF4 occurs as a gas, a solid, and an aqueous complex. The distinction between the different phases is made by phase designators that immediately follow the chemical formula and appear in parentheses. The only formulae that are not provided with a phase designator are aqueous ions. They are the only charged species in this review since charged gases are not considered. The use of the phase designators is described below. •
The designator (1) is used for pure liquid substances, e.g., H2O(1).
•
The designator (aq) is used for undissociated, uncharged aqueous species, e.g., U(OH) 4 (aq), CO 2 (aq). Since ionic gases are not considered in this review, all ions may be assumed to be aqueous and are not designed with (aq). If
16
//. Standards, Conventions, and Contents of the Tables
a chemical reaction refers to a medium other than H 2 0 (e.g., D 2 0 , 90% ethanol/10% H 2 O), then (aq) is replaced by a more explicit designator, e.g., "(in D 2 O)" or "(sin)". In the case of (sin), the composition of the solution is described in the text. •
The designator (sin) is used for substances in solution without specifying the actual equilibrium composition of the substance in the solution. Note the difference in the designation of H2O in Eqs.(II.2) and (II.3). H2O(1) in Reaction (II.2) indicates that H2O is present as a pure liquid, i.e., no solutes are present, whereas Reaction (II.3) involves a HC1 solution, in which the thermodynamic properties of H2O(sln) may not be the same as those of the pure liquid H2O(1). In dilute solutions, however, this difference in the thermodynamic properties of H2O can be neglected, and H2O(sln) may be regarded as pure H2O(1). Example: UO2Cl2(cr) + 2HBr(sln) ^ UO2C12 • 3H2O(cr) ^ UO3(y) + 2HCl(sln) ^
UOBr2 (cr) + 2HCl(sln)
UO2C12 • H2O(cr) + 2 H2O(1) UO2Cl2(cr) + H2O(sln)
(II. 1) (II.2) (II.3)
•
The designators (cr), (am), (vit), and (s) are used for solid substances, (cr) is used when it is known that the compound is crystalline, (am) when it is known that it is amorphous, and (vit) for glassy substances. Otherwise, (s) is used.
•
In some cases, more than one crystalline form of the same chemical composition may exist. In such a case, the different forms are distinguished by separate designators that describe the forms more precisely. If the crystal has a mineral name, the designator (cr) is replaced by the first four characters of the mineral name in parentheses, e.g., SiO2(quar) for quartz and SiO2(chal) for chalcedony. If there is no mineral name, the designator (cr) is replaced by a Greek letter preceding the formula and indicating the structural phase, e.g., a-UF5,
P-UF5.
Phase designators are also used in conjunction with thermodynamic symbols to define the state of aggregation of a compound to which a thermodynamic quantity refers. The notation is in this case the same as outlined above. In an extended notation (cf. [82LAF]) the reference temperature is usually given in addition to the state of aggregation of the composition of a mixture.
//. 1 Symbols, terminology and nomenclature
17
Example: AfG° (Na + , 298.15 K)
standard molar Gibbs energy of formation of aqueous Na+ at 298.15 K
S°(UO 2 (SO 4 )-2.5H 2 O,cr, 298.15 K)
standard molar entropy of UO2(SO4)-2.5H2O(cr) at 298.15 K
C° m (UO 3 , a, 298.15 K)
standard molar heat capacity of a-UO 3 at 298.15 K
A f // m (HF, sin, HF-7.8H2O)
enthalpy of formation of HF diluted 1:7.8 with water.
II.1.5 Processes Chemical processes are denoted by the operator A, written before the symbol for a property, as recommended by IUPAC [82LAF]. An exception to this rule is the equilibrium constant, cf. Section II. 1.6. The nature of the process is denoted by annotation of the A, e.g., the Gibbs energy of formation, AfGm, the enthalpy of sublimation, A sub // m , etc. The abbreviations of chemical processes are summarised in Table II-3. Table II-3: Abbreviations used as subscripts of A to denote the type of chemical process. Subscript of A
Chemical process
at
separation of a substance into its constituent gaseous atoms (atomisation)
dehyd
elimination of water of hydration (dehydration)
dil
dilution of a solution
f
formation of a compound from its constituent elements
fus
melting (fusion) of a solid
hyd
addition of water of hydration to an unhydrated compound
mix
mixing of fluids
r
chemical reaction (general)
sol
process of dissolution
sub
sublimation (evaporation) of a solid
tr
transfer from one solution or liquid phase to another
trs
transition of one solid phase to another
vap
vaporisation (evaporation) of a liquid
The most frequently used symbols for processes are Af G and Af H , the Gibbs energy and the enthalpy of formation of a compound or complex from the elements in their reference states {cf Table II-6).
18
//. Standards, Conventions, and Contents of the Tables
11.1.6 Equilibrium constants The IUPAC has not explicitly defined the symbols and terminology for equilibrium constants of reactions in aqueous solution. The NEA has therefore adopted the conventions that have been used in the work Stability Constants of Metal ion Complexes by Sillen and Martell [64SIL/MAR], [71SIL/MAR]. An outline is given in the paragraphs below. Note that, for some simple reactions, there may be different correct ways to index an equilibrium constant. It may sometimes be preferable to indicate the number of the reaction to which the data refer, especially in cases where several ligands are discussed that might be confused. For example, for the equilibrium: (II.4)
mM + ? L - M , L ,
both Pq m and 8 (II.4) would be appropriate, and fiq m (II.4) is accepted, too. Note that, in general, K is used for the consecutive or stepwise formation constant, and /? is used for the cumulative or overall formation constant. In the following outline, charges are only given for actual chemical species, but are omitted for species containing general symbols (M, L). II.1.6.1 Protonation of a ligand H"1 + H ,L ^
HL
r H ' + L ^ H L
-- ,
K.r = — J ^ ^ — '•' [H+][H,._,L] p. =
[H X]
'
/v
(II.5)
(II.6)
[H+r[L] This notation has been proposed and used by Sillen and Martell [64SIL/MAR], but it has been simplified later by the same authors [71SIL/MAR] from Ki r to Kr. This review retains, for the sake of consistency, cf. Eqs.(II.7) and (II.8), the older formulation of*,.,For the addition of a ligand, the notation shown in Eq.(II.7) is used. HL , + L ^ H L
K =—
[HL ] "-1—
(II.7).
LHL,,-iJLLJ Eq.(II.8) refers to the overall formation constant of the species H,.Lg. XW T 1
rtf+^L^HL 8 = '' " (II.8). H Hq •" -' [H+]r[L]« In Eqs.(II.5), (II.6) and (II.8), the second subscript r can be omitted if r = 1, as shown in Eq.(II.7).
//. 1 Symbols, terminology and nomenclature
19
Example:
H+ + POJ- ^ HPOJ-
2 H+ + POJ- ^ H2PO4 4
2
[HP
fin=A=
4
[H
/?, 2 =
fil2
[H
+ 2
°%]
f°4j
] [PO34-]
II.1.6.2 Formation of metal ion complexes ML . + L ^
ML
M+ ? L f i M L
[MLq ] K =— — ' [ML,,][L]
P=~
*
(II.9)
[ML ] —
(H.10)
" [M][Lf
For the addition of a metal ion, i.e., the formation of polynuclear complexes, the following notation is used, analogous to Eq.(II.5):
M + M ,L ^ M L
K, = r \?mL*
'•"
n
(11.11).
[M][Mm.,L]
Eq.(II. 12) refers to the overall formation constant of a complex M m L ? .
mM + qL^ML
J3 =
"
"•m [Mf[Lf
(11.12)
The second index can be omitted if it is equal to 1, i.e., /3qm becomes J5q if m- 1. The formation constants of mixed ligand complexes are not indexed. In this case, it is necessary to list the chemical reactions considered and to refer the constants to the corresponding reaction numbers. It has sometimes been customary to use negative values for the indices of the protons to indicate complexation with hydroxide ions, OH". This practice is not adopted in this review. If OH" occurs as a reactant in the notation of the equilibrium, it is treated like a normal ligand L, but in general formulae the index variable n is used instead of q. If H2O occurs as a reactant to form hydroxide complexes, H2O is considered as a protonated ligand, HL, so that the reaction is treated as described below in Eqs.(II.13) to (11.15) using n as the index variable. For convenience, no general form is used for the stepwise constants for the formation of the complex MmL?Hr. In many ex-
20
//. Standards, Conventions, and Contents of the Tables
periments, the formation constants of metal ion complexes are determined by adding a ligand in its protonated form to a metal ion solution. The complex formation reactions thus involve a deprotonation reaction of the ligand. If this is the case, the equilibrium constant is supplied with an asterisk, as shown in Eqs.(II.13) and (11.14) for mononuclear and in Eq.(II.15) for polynuclear complexes. [ML YH + 1 ^[^[HL]
MVl+HL^ML?+H+
, ^
M + (?HL ^ M L , + 9 H +
mM + qUL
^
M L + qH' m
0
"
[ML YH*T [M][HLf
=
=
^
L
^ ^
TM L ¥ H + 7 "
JL
(11.15)
=*-
[M] m [HLf
m
- — £ v , In c,. (11.34) n¥ E'° is the potential E for a cell when the ratio of the concentrations (not the activities) on the right-hand side and the left-hand side of the cell reaction is equal to unity, and (11.35) E'a =E° - - ^ Z v , ]ng y, = - ^ nr nr where the yj are the molality activity coefficients and g is ( "Vj), the ratio of molality to molarity (cf Section II.2). The medium must be specified.
25
//. 1 Symbols, terminology and nomenclature
II.1.7 pH Because of the importance that potentiometric methods have in the determination of equilibrium constants in aqueous solutions, a short discussion on the definition of "pH" and a simplified description of the experimental techniques used to measure pH will be given here. For a comprehensive account see [2002BUC/RON]. The acidity of aqueous solutions is often expressed in a logarithmic scale of the hydrogen ion activity. The definition of pH as: pH= -log 10 a H . = - l o g l 0 ( w H . y r ) can only be strictly used in the limiting range of the Debye-Hiickel equation (that is, in extremely dilute solutions). In practice the use of pH values requires extra assumptions on the values for single ion activities. In this review values of pH are used to describe qualitatively ranges of acidity of experimental studies, and the assumptions described in Appendix B are used to calculate single ion activity coefficients. The determination of pH is often performed by emf measurements of galvanic cells involving liquid junctions [69ROS], [73BAT]. A common setup is a cell made up of a reference half cell (e.g. Ag(s)|AgCl(s) in a solution of constant chloride concentration), a salt bridge, the test solution, and a glass electrode (which encloses a solution of constant acidity and an internal reference half cell): Pt(s)
Ag(s)
AgCl(s)
: salt ; test KCl(aq) I I ; bridge ! solution
a where
KCl(aq)
AgCl(s)
Ag(s)
Pt(s)
(11.36)
b
stands for a glass membrane.
The emf of such a cell (assuming Nernstian behaviour of the glass electrode) is given by: £=£*+—lnaHf
+Ei
where E is a constant, and E-s is the liquid junction potential. The purpose of the salt bridge is to minimise the junction potential in junction "b", while keeping constant the junction potential for junction "a". Two methods are most often used to reduce and control the value of E1. An electrolyte solution of high concentration (the "salt bridge") is a requirement of both methods. In the first method, the salt bridge is a saturated (or nearly saturated) solution of potassium chloride. A problem with a bridge of high potassium concentration, is that potassium perchlorate might precipitate1 inside the liquid junction when the test solution contains a high concentration of perchlorate ions.
1
KClO4(cr) has a solubility of « 0.15 M in pure water at 25°C
26
//. Standards, Conventions, and Contents of the Tables
In the other method the salt bridge contains the same high concentration of the same inert electrolyte as the test solution (for example, 3 M NaClC^). However, if the concentration of the background electrolyte in the salt bridge and test solutions is reduced, the values of E-s are dramatically increased. For example, if both the bridge and the test solution have [C1O~] = 0.1 M as background electrolyte, the dependence of the liquid junction at "b" on acidity is E-}« - 440 X [H+] mV-dnr'-mor 1 at 25°C [69ROS] (p.l 10), which corresponds to a correction at pH = 2 of > 0.07 pH units. Because of the problems in eliminating the liquid junction potentials and in defining individual ionic activity coefficients, an "operational" definition of pH is given by IUPAC [93MIL/CVI]. This definition involves the measurement of pH differences between the test solution and standard solutions of known pH and similar ionic strength (in this way similar values of y^ and E-3 cancel each other when emf values are subtracted). In order to deduce the stoichiometry and equilibrium constants of complex formation reactions and other equilibria, it is necessary to vary the concentrations of reactants and products over fairly large concentration ranges under conditions where the activity coefficients of the species are either known, or constant. Only in this manner is it possible to use the mass balance equations for the various components together with the measurement of one or more free concentrations to obtain the information desired [61ROS/ROS], [90BEC/NAG], [97ALL/BAN], p. 326 - 327. For equilibria involving hydrogen ions, it is necessary to use concentration units, rather than hydrogen ion activity. For experiments in an ionic medium, where the concentration of an "inert" electrolyte is much larger than the concentration of reactants and products we can ensure that, as a first approximation, their trace activity coefficients remain constant even for moderate variations of the corresponding total concentrations. Under these conditions of fixed ionic strength the free proton concentration may be measured directly, thereby defining it in terms of- logi0[H+] rather than on the activity scale as pH, and the value of - logio[H+] and pH will differ by a constant term, i.e., log10^ 4 Equilibrium constants deduced from measurements in such ionic media are therefore conditional constants, because they refer to the given medium, not to the standard state. In order to compare the magnitude of equilibrium constants obtained in different ionic media it is necessary to have a method for estimating activity coefficients of ionic species in mixed electrolyte systems to a common standard state. Such procedures are discussed in Appendix B. Note that the precision of the measurement of-log 10 [H + ] and pH is virtually the same, in very good experiments, ± 0.001. However, the accuracy is generally considerably poorer, depending in the case of glass electrodes largely on the response of the electrode (linearity, age, pH range, etc.), and to a lesser extent on the calibration method employed.
Ill Symbols, terminology and nomenclature
27
II. 1.8 Order of formulae To be consistent with CODATA, the data tables are given in "Standard Order of Arrangement [82WAG/EVA]. This scheme is presented in Figure II-l below, and shows the sequence of the ranks of the elements in this convention. The order follows the ranks of the elements. For example, for uranium, this means that, after elemental uranium and its monoatomic ions (e.g.,U 4+ ), the uranium compounds and complexes with oxygen would be listed, then those with hydrogen, then those with oxygen and hydrogen, and so on, with decreasing rank of the element and combinations of the elements. Within a class, increasing coefficients of the lower rank elements go before increasing coefficients of the higher rank elements. For example, in the U-O—F class of compounds and complexes, a typical sequence would be UOF2(cr), UOF4(cr), UOF 4 (g), UO2F(aq), U O 2 F \ UO2F2(aq), UO2F2(cr), UO 2 F 2 (g), UO2F3", UO2F;~, U 2 O 3 F 6 (cr), etc. [92GRE/FUG]. Formulae with identical stoichiometry are in alphabetical order of their designators. An exception is made for ammonium salts, which are included in group 1 before the alkali metals. Figure II-l: Standard order of arrangement of the elements and compounds based on the periodic classification of the elements (from [82 WAG/EVA]).
28
//. Standards, Conventions, and Contents of the Tables
II.1.9 Reference codes The references cited in the review are ordered chronologically and alphabetically by the first two authors within each year, as described by COD ATA [87GAR/PAR]. A reference code is made up of the final two digits of the year of appearance (if the publication is not from the 20th century, the year will be put in full). The year is followed by the first three letters of the surnames of the first two authors, separated by a slash. If there are multiple reference codes, a "2" will be added to the second one, a "3" to the third one, and so forth. Reference codes are always enclosed in square brackets.
II.2 Units and conversion factors Thermodynamic data are given according to the Systeme International d'unites (SI units). The unit of energy is the joule. Some basic conversion factors, also for nonthermodynamic units, are given in Table II-4. Table II-4: Unit conversion factors To convert from
multiply by
to
(non-SI unit symbol)
(SI unit symbol)
angstrom (A)
metre (m)
1x10 '"(exactly)
standard atmosphere (atm)
pascal (Pa)
1.01325xl05 (exactly)
bar (bar)
pascal (Pa)
lxlO 5 (exactly)
thermochemical calorie (cal)
joule (J)
4.184 (exactly)
J K ' mol 1
entropy unit e.u. = cal • K ' • mol
4.184 (exactly)
Since a large part of the NEA-TDB project deals with the thermodynamics of aqueous solutions, the units describing the amount of dissolved substance are used very frequently. For convenience, this review uses "M" as an abbreviation of " mol • dm~3" for molarity, c, and, in Appendices B and C, "m" as an abbreviation of "mol-kg"' " for molality, m. It is often necessary to convert concentration data from molarity to molality and vice versa. This conversion is used for the correction and extrapolation of equilibrium data to zero ionic strength by the specific ion interaction theory, which works in molality units (cf. Appendix B). This conversion is made in the following way. Molality is defined as mB moles of substance B dissolved in 1 kilogram of pure water. Molarity is defined asc B moles of substance B dissolved in (p - cBM) kilogram of pure water, where pis the density of the solution in kgdm"3 and MB the molar weight of the solute in kg-moi"1. From this it follows that: mB =
p -
cBM
.
II.2 Units and conversion factors
29
When the ionic strength is kept high and constant by an inert electrolyte, I, the ratio mB /cB can be approximated by: 1
mB _ cB
p - clMl
where ct is the concentration of the inert electrolyte in mol-dnf3 and Mi its molar mass in kg-mol"'. Baes and Mesmer [76BAE/MES], (p.439) give a table with conversion factors (from molarity to molality) for nine electrolytes and various ionic strengths. Conversion factors at 298.15 K for twenty one electrolytes, calculated using the density equations reported by Sohnel and Novotny [85SOH/NOV], are reported in Table II-5. Example: 1.00MNaClO 4
=
l.OOMNaCl
= 1.02mNaCl
4.00MNaClO 4
= 4.95mNaClO 4
6.00MNaNO 3
= 7.55mNaNO 3
1.05mNaClO4
It should be noted that equilibrium constants need also to be converted if the concentration scale is changed from molarity to molality or vice versa. For a general equilibrium reaction, 0 = £ B V B ^ , the equilibrium constants can be expressed either in molarity or molality units, Kc or Km , respectively: log, 0 ^ r = Zv B log 10 c B B
O
l gl0 ^
m
=
XVBlog|0OTB
B
With (mB / c B ) = g, or (log10 mB - log l0 cB ) = Iogi0£>, the relationship between Kc and Km becomes very simple, as shown in Eq.(II.37). log10 Km = log 10 Kc + £ v B \ogw6 B
(11.37)
X B vB is the sum of the stoichiometric coefficients of the reaction, cf. Eq. (11.53) and the values of g are the factors for the conversion of molarity to molality as tabulated in Table II-5 for several electrolyte media at 298.15 K. In the case of very dilute solutions, these factors are approximately equal to the reciprocal of the density of the pure solvent. Then, if the solvent is water, molarity and molality may be used interchangeably, and Kc « Km. The differences between the values in Table II-5 and the values listed in the uranium NEA-TDB review [92GRE/FUG] (p.23) are found at the highest concentrations, and are no larger than ± 0.003 dm3-kg"1, reflecting the accuracy expected in this type of conversion. The uncertainty introduced by the use of Eq.(II.37) in the values of log10 Km will be no larger than ± 0.001 Z B VB •
30
//. Standards, Conventions and Contents of the Tables
Table II-5: Factors Q for the conversion of molarity, CB, to molality, ff?B, of a substance B, in various media at 298.15 K (calculated from densities in [85SOH/NOV]) )/ CB (dm3 of solution per kg of H2O) c(M)
HC1O4
NaClO4
LiClO4
NH4CIO4
Ba(C104)2
HC1
NaCl
LiCl
0.10
1.0077
1.0075
1.0074
1.0091
1.0108
1.0048
1.0046
1.0049
0.25
1.0147
1.0145
1.0141
1.0186
1.0231
1.0076
1.0072
1.0078
0.50
1.0266
1.0265
1.0256
1.0351
1.0450
1.0123
1.0118
1.0127
1.0523
1.0685
1.0172
1.0165
1.0177
0.75
1.0386
1.0388
1.0374
1.00
1.0508
1.0515
1.0496
1.0703
1.0936
1.0222
1.0215
1.0228
1.50
1.0759
1.0780
1.0750
1.1086
1.1491
1.0324
1.0319
1.0333
2.00
1.1019
1.1062
1.1019
1.2125
1.0430
1.0429
1.0441
1.3689
1.0654
1.0668
1.0666
1.0893
1.0930
1.0904
1.1218
1.1156
3.00
1.1571
1.1678
1.1605
4.00
1.2171
1.2374
1.2264
5.00
1.2826
1.3167
1.1147
6.00
1.3547
1.4077
1.1418
c(M)
K.C1
NH4C1
MgCl2
0.10
1.0057
1.0066
1.0049
0.25
1.0099
1.0123
1.0080
0.50
1.0172
1.0219
1.0135
0.75
1.0248
1.0318
1.00
1.0326
1.0420
1.50
1.0489
2.00
NaBr
HNO3
NaNO3
LiNO3
1.0044
1.0054
1.0056
1.0058
1.0059
1.0069
1.0090
1.0097
1.0102
1.0103
1.0119
1.0154
1.0169
1.0177
1.0178
1.0195
1.0176
1.0220
1.0242
1.0256
1.0256
1.0258
1.0239
1.0287
1.0319
1.0338
1.0335
1.0632
1.0393
1.0382
1.0428
1.0478
1.0510
1.0497
1.0662
1.0855
1.0540
1.0546
1.0576
1.0647
1.0692
1.0667
3.00
1.1037
1.1339
1.0867
1.0934
1.0893
1.1012
1.1090
1.1028
4.00
1.1453
1.1877 (
1.1241
1.1406
1.1240
1.1417
1.1534
1.1420
1.1974
1.1619
1.1865
1.2030
1.1846
1.2033
1.2361
1.2585
1.2309
5.00
1.2477
CaCl2
1.1423
6.00 c(M)
NHiNOs
H2SO4
Na2SO4
(NH4)2SO4
H3PO4
Na2CO3
K2CO3
NaSCN
0.10
1.0077
1.0064
1.0044
1.0082
1.0074
1.0027
1.0042
1.0069
0.25
1.0151
1.0116
1.0071
1.0166
1.0143
1.0030
1.0068
1.0130
0.50
1.0276
1.0209
1.0127
1.0319
1.0261
1.0043
1.0121
1.0234
0.75
1.0405
1.0305
1.0194
1.0486
1.0383
1.0065
1.0185
1.0342
1.00
1.0539
1.0406
1.0268
1.0665
1.0509
1.0094
1.0259
1.0453
1.50
1.0818
1.0619
1.0441
1.1062
1.0773
1.0170
1.0430
1.0686
2.00
1.1116
1.0848
1.1514
1.1055
1.0268
1.0632
1.0934
3.00
1.1769
1.1355
1.2610
1.1675
1.1130
1.1474
4.00
1.2512
1.1935
1.4037
1.2383
1.1764
1.2083
5.00
1.3365
1.2600
1.3194
1.2560
1.2773
6.00
1.4351
1.3365
1.4131
1.3557
31
II. 3 Standard and reference conditions
II.3 Standard and reference conditions II.3.1 Standard state A precise definition of the term "standard state" has been given by IUPAC [82LAF]. The fact that only changes in thermodynamic parameters, but not their absolute values, can be determined experimentally, makes it important to have a well-defined standard state that forms a base line to which the effect of variations can be referred. The IUPAC [82LAF] definition of the standard state has been adopted in the NEA-TDB project. The standard state pressure, p° = 0.1 MPa (1 bar), has therefore also been adopted, cf. Section II.3.2. The application of the standard state principle to pure substances and mixtures is summarised below. It should be noted that the standard state is always linked to a reference temperature, cf. Section II.3.3. • The standard state for a gaseous substance, whether pure or in a gaseous mixture, is the pure substance at the standard state pressure and in a (hypothetical) state in which it exhibits ideal gas behaviour. • The standard state for a pure liquid substance is (ordinarily) the pure liquid at the standard state pressure. • The standard state for a pure solid substance is (ordinarily) the pure solid at the standard state pressure. • The standard state for a solute B in a solution is a hypothetical liquid solution, at the standard state pressure, in which mB =m° = 1 mol-kg"', and in which the activity coefficient yB is unity. It should be emphasised that the use of superscript, °, e.g., in AtH^, implies that the compound in question is in the standard state and that the elements are in their reference states. The reference states of the elements at the reference temperature (cf. Section II.3.3) are listed in Table II-6. Table II-6: Reference states for some elements at the reference temperature of 298.15 K and standard pressure of 0.1 MPa [82WAG/EVA], [89COX/WAG], [91 DIN]. O2
gaseous
Al
crystalline, cubic
H2
gaseous
Zn
crystalline, hexagonal
He
gaseous
Cd
crystalline, hexagonal
Ne
gaseous
Hg
liquid
Ar
gaseous
Cu
crystalline, cubic
Kr
gaseous
Ag
crystalline, cubic
Xe
gaseous
Fe
crystalline, cubic, bcc
¥2
gaseous
Tc
crystalline, hexagonal
Cl2
gaseous
V
crystalline, cubic
(Continued on next page)
32
//. Standards, Conventions and Contents of the Tables
Table II-6: (continued) Br2
liquid
Ti
crystalline, hexagonal
I2 S
crystalline, orthorhombic
Am
crystalline, dhcp
crystalline, orthorhombic
Pu
crystalline, monoclinic
Se
crystalline, hexagonal ("black")
Np
crystalline, orthorhombic
Te
crystalline, hexagonal
U
crystalline, orthorhombic
N2
gaseous
Th
crystalline, cubic
P
crystalline, cubic ("white")
Be
crystalline, hexagonal
As
crystalline, rhombohedral ("grey")
Mg
crystalline, hexagonal
Sb
crystalline, rhombohedral
Ca
crystalline, cubic, fee
Bi
crystalline, rhombohedral
Sr
crystalline, cubic, fee
C
crystalline, hexagonal (graphite)
Ba
crystalline, cubic
Si
crystalline, cubic
Li
crystalline, cubic
Ge
crystalline, cubic
Na
crystalline, cubic
Sn
crystalline, tetragonal ("white")
K
crystalline, cubic
Pb
crystalline, cubic
Rb
crystalline, cubic
B
/}, crystalline, rhombohedral
Cs
crystalline, cubic
II.3.2 Standard state pressure The standard state pressure chosen "for all selected data is 0.1 MPa (1 bar) as recommended by the International Union of Pure and Applied Chemistry IUPAC [82LAF]. However, the majority of the thermodynamic data published in the scientific literature and used for the evaluations in this review, refer to the old standard state pressure of 1 "standard atmosphere" (= 0.101325 MPa). The difference between the thermodynamic data for the two standard state pressures is not large and lies in most cases within the uncertainty limits. It is nevertheless essential to make the corrections for the change in the standard state pressure in order to avoid inconsistencies and propagation of errors. In practice the parameters affected by the change between these two standard state pressures are the Gibbs energy and entropy changes of all processes that involve gaseous species. Consequently, changes occur also in the Gibbs energies of formation of species that consist of elements whose reference state is gaseous (H, O, F, Cl, N, and the noble gases). No other thermodynamic quantities are affected significantly. A large part of the following discussion has been taken from the NBS tables of chemical thermodynamic properties [82WAG/EVA], see also Freeman [84FRE]. The following expressions define the effect of pressure on the properties of all substances:
//. 3 Standard and reference conditions
33
ldT)P
{3p)r [dp)T
ldT)p
{T) = v
(II 41)
where
(11.42)
-
a = —(—)
For ideal gases, V = RT/p and a = R/pV = 1 IT. The conversion equations listed below (Eqs. (11.43) to (11.50)) apply to the small pressure change from 1 atm to 1 bar (0.1 MPa). The quantities that refer to the old standard state pressure of 1 atm are assigned the superscript =-5x0.03263 kJ-mor 1 at 298.15 K.
(H.48).
The changes in the equilibrium constants and cell potentials with the change in the standard state pressure follow from the expression for Gibbs energy changes, Eq.(II.47)
"'-bio
'•""••IH
= 5-0.005717
£ ±0.128
111.870 ±0.120
175.018 ±0.004
Br~ [92GRE/FUG]
-103.850 (a> ±0.167
-121.410 ±0.150
82.550 ±0.200
Br2(aq) [92GRE/FUG]
4.900 ±1.000
Br2(g) [92GRE/FUG]
3.105 ±0.142
30.910 ±0.110
245.468 ±0.005
Br2(l) [92GRE/FUG]
0.000 ±0.000
0.000 ±0.000
152.210 ±0.300
BrCT [92GRE/FUG]
-32.095 ±1.537 19.070 (a) +0.634
- 66.700 +0.500
161.500 ±1.300
(a)
-36.290 ±0.160
198.700 ±0.004
29.141 ±0.003
20.786 ±0.001
BrO3" [92GRE/FUG]
1
w
(J • K"
1
1
• mol" )
(J • K" 1 • m o l " 1 )
20.786 ±0.001
36.057 ±0.002
HBr(g) [92GRE/FUG]
- 53.361 ±0.166
HBrO(aq) [92GRE/FUG]
-81.356 (b) ±1.527
Kg) [92GRE/FUG]
70.172 (a) ±0.060
106.760 ±0.040
180.787 ±0.004
[92GRE/FUG]
-51.724 (a) ±0.112
- 56.780 ±0.050
106.450 ±0.300
I2(cr) [92GRE/FUG]
0.000 ±0.000
0.000 ±0.000
116.140 ±0.300
b(g) [92GRE/FUG]
19.323 (a) ±0.120
62.420 ±0.080
260.687 ±0.005
(a)
-219.700 +0.500
118.000 ±2.000
1.700 (a) ±0.110
26.500 ±0.100
206.590 ±0.004
29.157 ±0.003
0.000 ±0.000
32.054 ±0.050
22.750 ±0.050
277.170 ±0.150
167.829 ±0.006
23.674 ±0.001
r
IO3[92GRE/FUG] Hl(g) [92GRE/FUG] HIO3(aq) [92GRE/FUG] S(cr)(orthorhombic) [92GRE/FUG] S(g) [92GRE/FUG]
s2" [92GRE/FUG]
- 126.338 ±0.779
36.888 ±0.002
- 130.836 ±0.797 0.000 ±0.000 236.689 (a> ±0.151 120.695 ±11.610
(Continued on next page)
58
IV Selected auxiliary data
Table IV-1: (continued) Compound and NEA TDB Review
A f o: 1
(kJ • mol" )
Af//°
5°
(kJ • mol" 1 )
(J • K"' • mot 1 )
C (J • KT1 • m o r 1 )
S2(g) [92GRE/FUG]
79.686 ±0.301
(a)
SO2(g) [92GRE/FUG]
- 300.095 ±0.201
(a)
-296.810 ±0.200
248.223 ±0.050
SO?" [92GRE/FUG]
- 487.472 +4.020
S2Ol" [92GRE/FUG]
-519.291 ±11.345
SO4" [92GRE/FUG]
- 744.004 ±0.173
-110.530 ±0.170
197.660 ±0.004
29.141 ±0.002
CO2(aq) [92GRE/FUG]
- 385.970
CO2(g) [92GRE/FUG]
^ p.m
1
37.135 ±0.002
CN~ [2005OLI/NOL]
166.939 (b) +2.519
147.350 ±0.559
(kJ • mol"') 166.870 ±0.500
1
1
(J • IC • mor ) 65.740 ±0.800
p.m
(J • K-[ • m o r 1 )
HgO(montroydite, red) [92GRE/FUG] Hg2Cl2(cr) [92GRE/FUG]
- 58.523 ±0.154
(a)
- 90.790 ±0.120
70.250 ±0.300
-210.725 ±0.471
(a)
- 265.370 ±0.400
191.600 ±0.800
Hg2SO4(cr) [92GRE/FUG]
- 625.780 (a) ±0.411
- 743.090 ±0.400
200.700 ±0.200
0.000
33.150 ±0.080
24.440 ±0.050 20.786 ±0.001
Cu(cr) [92GRE/FUG]
0.000
Cu(g) [92GRE/FUG]
297.672 (a> ±1.200
337.400 ±1.200
166.398 ±0.004
(a)
64.900 ±1.000
- 98.000 ±4.000
Cu2+ [92GRE/FUG]
65.040 ±1.557
CuCl(g) [2003GUI/FAN] CuSO4(cr) [92GRE/FUG]
76.800 ±10.000 -662.185 ±0.801
147.100 ±0.800
148.648 ±0.003
Mg2* [92GRE/FUG]
- 455.375 (a> ±1.335
- 467.000 ±0.600
-137.000 ±4.000
MgO(cr) [92GRE/FUG]
- 569.312 ±0.305
(a)
-601.600 ±0.300
26.950 ±0.150
37.237 ±0.200
MgF2(cr) [92GRE/FUG]
-1071.051 ±1.210
(a)
-1124.200 ±1.200
57.200 ±0.500
61.512 ±0.300
Ca(cr) [92GRE/FUG]
0.000
0.000
41.590 ±0.400
25.929 ±0.300
Ca(g) [92GRE/FUG]
144.021 ±0.809
(a)
177.800 ±0.800
154.887 ±0.004
20.786 ±0.001
Ca2+ [92GRE/FUG]
- 552.806 ±1.050
(a)
- 543.000 ±1.000
- 56.200 ±1.000
CaO(cr) [92GRE/FUG]
- 603.296 ±0.116
- 252.140 ±0.080
101.200 ±0.200
(Continued on next page)
67
IV Selected auxiliary data
Table IV-1: (continued) Compound and NEA TDB Review
A
C° m
fGm
1
1
(kj • mol" )
(kJ • mol" )
KCl(cr) This Review
-436.461 ±0.129
KBr(cr) This Review
-393.330 +0.188
KI(cr) This Review
-329.150 +0.137
Rb(cr) [92GRE/FUG]
0.000
Rb(g) [92GRE/FUG]
53.078 ±0.805
Rb" [92GRE/FUG]
- 284.009 ±0.153
Cs(cr) [92GRE/FUG] Cs(g) [92GRE/FUG]
1
1
(J • K" • mol" )
(J • K"1 • mor 1 )
0.000
76.780 ±0.300
31.060 ±0.100
w
80.900 ±0.800
170.094 ±0.003
20.786 ±0.001
|a)
-251.120 ±0.100
121.750 ±0.250
0.000
0.000
85.230 ±0.400
32.210 ±0.200
49.556 (a> ±1.007
76.500 ±1.000
175.601 ±0.003
20.786 ±0.001
Cs" [92GRE/FUG]
-291.456 ±0.535
(a)
- 258.000 ±0.500
132.100 ±0.500
CsCl(cr) [2001LEM/FUG]
-413.807 ±0.208
(a)
-442.310 ±0.160
101.170 ±0.200
52.470
CsBr(cr) [2001LEM/FUG]
-391.171 ±0.305
-405.600 ±0.250
112.940 +0.400
52.930
(a) (b)
Value calculated internally using AfG° = Af H°m Value calculated internally from reaction data (see Table 1V-2).
68
IV Selected auxiliary data
Table IV—2: Selected thermodynamic data for reactions involving auxiliary compounds and complexes used in the evaluation of thermodynamic data for the NEA TDB Project data. All ionic species listed in this table are aqueous species. Unless noted otherwise, all data refer to 298.15 K and a pressure of 0.1 MPa and, for aqueous species, a standard state of infinite dilution (/ = 0). The uncertainties listed below each value represent total uncertainties and correspond in principle to the statistically defined 95% confidence interval. Systematically, all the values are presented with three digits after the decimal point, regardless of the significance of these digits. The reference listed for each entry in this table indicates the NEA TDB Review where the corresponding data have been adopted as NEA TDB Auxiliary data.The data presented in this table are available on computer media from the OECD Nuclear Energy Agency. Species and NEA TDB Review
Reaction log,u * ° (kj • mol" 1 )
HF(aq)
F" + H* ^
3.180 ±0.020
HF2" [92GRE/FUG]
F" + HF(aq) ^
CIO" [92GRE/FUG]
HClO(aq) ^
C1O2~ [92GRE/FUG]
HClO2(aq) ^
HClO(aq)
Cl2(g) + H2O(1) ^
HClO2(aq) [92GRE/FUG]
101.800 ±1.077
(a)
- 18.152 ±0.114
12.200 ±0.300
-2.511 ±0.685
3.000 ±2.000
18.486 (n) ±7.090
19.000 ±9.000
- 78.329 (a) ±30.289
HF2"
0.440 ±0.120
CIO -I-H*
- 7.420 ±0.130
42.354 ±0.742 C1O2" + FT
-1.960 +0.020
11.188 ±0.114 Cl" + FT + HClO(aq)
-4.537 ±0.105
25.900 ±0.600
H2O(1) + HClO(aq) ^± 2H+ + HClO2(aq) -t•2e~ - 55.400 ±0.700
(b)
HBrO(aq) ^
HBrO(aq)
Br2(aq) + H2O(1)
-8.630 ±0.030
- 8.240 ±0.200
316.230 ±3.996
BrO" fH+
BrO" [92GRE/FUG]
[92GRE/FUG]
(J • K" 1 • m o l " 1 )
HF(aq)
[92GRE/FUG]
[92GRE/FUG]
(Id • mol" 1 )
49.260 ±0.171
30.000 ±3.000
- 64.600 ±10.078
(a)
^ Br~ + H + + HBrO(aq) 47.034 ±1.142
(Continued on next page)
69
IV Selected auxiliary data
Table IV-2 (continued) Species and
Reaction ArG°
NEA TDB Review
(kJ • mol"1) HIO3(aq)
FT + \Oi ^
s2[92GRE/FUG] SO2," [92GRE/FUG]
s2or [92GRE/FUG] H2S(aq)
HS~ ^
-4.498 ±0.166
FT + S2"
- 19.000 ±2.000
108.450 ±11.416
- 31.400 (b) ±0.700
179.230 ±3.996
3H 2 O(I) + 2 S O r + 4 e - ^ 6 O F T + - 39.200 +1.400 H2S(aq) ^
(b)
223.760 ±7.991
H+ + HS"
-6.990 ±0.170
[92GRE/FUG]
FT + SO2," ^
HS2O3~ [92GRE/FUG]
H+ + S2O3" ^
H2SO3(aq)
FT + HSO3" ?=i H2SO3(aq)
HN3(aq)
H+ + N ; ^
16.000 ±5.000
88.891 ±16.840
w
-26.828 ±0.457
-15.000 ±10.000
39.671 ±33.575
(a)
52.725 ±0.126
52.090 ±0.210
-2.130 ±0.821
(a)
-10.503 ±0.457 HSOi - 11.302 ±0.285
HN3(aq)
4.700 ±0.080 NHJ ^
(a)
- 9.076 ±0.856
1.980 ±0.050
[92GRE/FUG]
359.590 ±100.630
HS2O3"
1.840 ±0.080 H+ + SO4" ^
66.000 +30.000
-41.212 ±0.457
1.590 ±0.150
USO4 [92GRE/FUG]
[92GRE/FUG]
HSO3"
7.220 ±0.080
[92GRE/FUG]
(a)
39.899 ±0.970
HSO3" [92GRE/FUG]
NH3(aq)
(J • KT1 • m o r 1 )
HIO3(aq)
0.788 ±0.029
[92GRE/FUG]
(kJ • mol"1)
H+ + NH3(aq) -9.237 ±0.022
(Continued on next page)
70
IV Selected auxiliary data
Table IV-2 (continued) Species and NEATDB Review
Reaction ArG°
iog,0K°
A r //° 1
(kJ • mor ) HNO2(aq) [92GRE/FUG]
[92GRE/FUG] P 2 O^ [92GRE/FUG]
H" + NO2~ ^ HNO2(aq) 3.210 -18.323 ±0.160 ±0.913
-12.350 +0.030 HP 2 O^ ^
70.494 ±0.171
- 9.400 ±0.150
H3PO4(aq)
H* + H2PO4" ^
-12.215 +0.171
H2P2O5" ^ H+ + HP2O^" - 6.650 37.958 ±0.100 +0.571
H2P2O5' [92GRE/FUG]
H3P2Of ^ H+ + H2P2O5" -2.250 12.843 ±0.150 +0.856
H3P2O7" [92GRE/FUG]
H4P2O7(aq) ^ H+ + H3P2Of -1.000 5.708 ±0.500 ±2.854
H4P2O7(aq)
2H3PO4(aq) ^
[92GRE/FUG] CO2(g) [92GRE/FUG]
14.600 +3.800
23.219 1 mM, became widely accepted. Consequently, as pointed out in [98PLY/ZHA], almost all previously reported values should be corrected. In such cases, re-evaluation was performed including oligonuclear species, especially Ni4 (OH) 4 + .
(iii)
In addition, although the experimental methods (e.g., pH-measurements) were already well established in the 1950s, the evaluation of experimental data has developed considerably since that time (in a great part of the cited works, graphical evaluation was used). Therefore, re-assessment of data was performed to check (and if necessary complete) the proposed equilibrium model.
The formation of two main species is generally recognised in this pH region: NiOH+, and Ni4 (OH) 4 + . Due to the presence of oligonuclear species, the onset of the hydrolysis reaction of Ni(II) is strongly concentration dependent. Under natural conditions above pH = 6.9, and at low Ni(II) concentrations (< 0.0001 M), only the dominant aqua complex and the species NiOH+ may be significant. In more concentrated solutions ([Ni2+]tot > 0.005 M), however, the tetramer Ni4(OH)4+ is the major hydroxo complex before precipitation occurs. The formation of a minor species Ni2OH3+ has been suggested by several authors [65BUR/LIL], [65BUR/LIL2], [66BUR/IVA], [71BUR/ZIN], [71OHT/BIE], [78BUR/KAM], because by adding this complex to the model a better fit of the data could be achieved. Some other species were also postulated by different authors (Ni 2 (OH)f [69KEN2], Ni 2 (OH)^ [71BHA/SUB], Ni3(OH)3+
92
V Discussion of data selection for nickel
[66BUR/IVA]). However, the formation of these species has never been independently confirmed. Actually the re-evaluation of data reported in [66BUR/IVA] indicated the presence of Ni4(OH)4+ rather than Ni3(OH)j+ (see discussion in Appendix A). The following equations describe the formation of the first mentioned three species: Ni2+ + H2O(1) ^ NiOH+ + H+
*/?u
(V-23)
2Ni2+ + H2O(1) ^ Ni2OH3+ + H+
*pu
(V.24)
4Ni2+ + 4 H 2 O ( l ) ^ Ni4(OH)J+ +4H +
*/?M
(V.25)
The reported and re-calculated hydrolysis constants are listed in Table V-4. As mentioned above, several early studies reported the formation of NiOH+ as the only hydrolytic species, but the experiments were conducted at relatively high concentrations of Ni(II), where a considerable amount of Ni4(OH)4+ species can be expected. To avoid its presence, the [Ni2+]tot should not exceed 0.001 M, and the pH should not exceed 8.6 to preclude precipitation. The studies of Perrin [64PER] and of Funahashi and Tanaka [69FUN/TAN] were the only ones in which conditions appropriate to the determination of a reliable formation constant for NiOH+ were used. The reevaluation of the experimental data reported in [63BOL/JAU], [65BUR/LIL], [65BUR/LIL2], [71BUR/ZIN] resulted in acceptable log,0 * $ , values. In a few cases, a simple correction was applied to take into account the formation of the NiCl+ complex [71OHT/BIE], [80MIL/BUG]. However, some of the reported values cannot be corrected, as insufficient experimental details or data have been communicated [52CHA/COU], [52GAY/WOO], [53SCH/POL], [56CUT/KSA], [59ACH], [63SHA/DES]. Tremaine and LeBlanc have reported a sharp increase of the first hydrolysis constant of Ni(II) between 150 and 300°C [80TRE/LEB2]. These data indicate that the formation of NiOH+ is more pronounced at higher temperatures. The mole fraction of NiOH+ never exceeds 15% at 25°C, but reaches 70% at 300°C. Although the reported data seem to be reliable, the temperature and pressure range used is well above of the scope of this review. The authors' extrapolation to the standard state resulted in considerably different values from those determined at 25°C.
93
V.3 Oxygen and hydrogen compounds and complexes
Table V-4: Experimental equilibrium data (logarithmic values) for the hydrolysis of Ni(II) in acidic or neutral solutions. Method
Ni2t + H2O(1) gl gl gl gl gl gl
gl
L
Medium
t/°C
X
Reference
retained"1*
^ NiOH+ + H 0.1
-
[52CHA/COU]
NiCl2 NiCl2, NiSO4
30 25
-9.4
->0
-(10.64 ±0.24)
-0.08 ->0
25 25
[52GAY/WOO] [53SCH/POL]
25 25
-
[56CUT/KSA]
->0
-9 -8.94 -(10.92 ±0.12)
-
Ni(C104)2 KC1O4, Ni(NO3)2 (Na)ClO4
(K)NO3
(Na)ClO4 (Na)Cl NaClO4 7
gl gl pH(sp)
(Na)Cl (Na)Cl (Na)ClO4
sol
HC1, NaOH
0.25 1.05 1.05 1.05
25 30 40
1.05
50
->0
15 20 25
(K)C1 NaClO4
-(9.76 ±0.05) - (9.65 ± 0.04) -(9.51 ±0.04)
-(10.05 ±0.04) -(9.86 ±0.03) - (9.75 ± 0.07)
- (10.07 ±0.20)' d) - (9.80 ± 0.20)""
25 25
0.10
25 7
-(9.5 ±0.1) -(11.0 ±0.2)
25
< - 10.5 -
0.82 ->0
60 25 150 200
0.51 1.57 0.01
25 25
-(8.93 ±0.1) - (8.05 ± 0.48) - (6.95 ± 0.22) - (6.03 ± 0.20) -(5.23 ±0.36) - (9.87 ± 0.40)
[59ACH] [63BOL/JAU]
-(9.71 ±0.30)' c) - (9.72 ± 0.30)(c) - (9.62 ± 0.30)'c) - (9.66 ± 0.30)(c)
3.50 3.20 ->0 3.20 3.20
- (9.84 ± 0.30)'c)
- (9.32 ± 0.06) -(10.22 ±0.04)
-(9.58 ±0.06) - (9.43 ± 0.03) -
250 300
sol
log,.
KC1
kin
gl
X
reported'"'
30 36 42 gl gl kin
log,.
- (10.22 ±0.20) (d)
[64PER]
- (9.74 ± 0.20)"" - (9.59 ± 0.20)(d) - (9.38 ± 0.20)(d) - (9.79 ± 0.12)(c) - (9.53 ± 0.20)(e) - (9.5 ± 0.2) -(10.23 ±0.20) (c) - (8.45 ± 0.30)(c)
[65BUR/LIL] [65BUR/LIL2] [69FUN/TAN] [71HOH/GEI] [71OHT/BIE] [71BUR/ZIN]
-
[75CLA/KEP] [80TRE/LEB2]
-(9.81 ±0.50) (c)
[80MIL/BUG]
-(10.06 ±0.40)
- (9.93 ± 0.50)(c)
-(8.35 ±0.10)
-
[97MAT/RAI]
(Continued on next page)
94
V Discussion of data selection for nickel
Table V-4 (continued) Method
Medium
Im
t/°C
l°g,o >„.„ reported"0
l°gio P,,m retained"5'
Reference
2Ni2f + H2O(1) ?=± Ni 2 OH 3 •t H * gl gl
(Na)ClO4
3.50
25
-(10.0 ±0.5)
- (9.8 ± 0.5)lc)
[65BUR/LIL]
(Na)Cl
25
(Na)NO3 (Na)Cl
- (9.3 ± 0.2) - (9.6 ± 0.2)
- (8.95 ± 0.40)(c)
gl gl
3.20 3.30
[65BUR/LIL2] [66BUR/IVA]
gl gl
(Na)Cl (Na)Br
-(10.5 ±0.5) - (8.5 ± 0.2)
^(10.0±0.6) l c ) - (8.58 ± 0.60)(c)
[71OHT/BIE] [71BUR/ZIN]
-(9.5 ±0.1)
- (8.82 ± 0.40)|c)
[78BUR/KAM]
-
- (27.34 ± 0.30)lc)
[63BOL/JAU]
4Ni gl
2+
3.20
25 25
3.20 3.30
60 25
+ 4H2O(1) ^=^ Ni4(OH)
[71OHT/BIE] [71BUR/ZIN]
- (27.25 + 0.16) - (26.99 ± 0.40)|c) -(27.15 ± 0.30)lc)
[73KAW/OTS] [75CLA/KEP] [78BUR/KAM] [80MIL/BUG]
-(27.0 ±0.3)
[91SPI/TRI]
(a) Reported values. (b) Accepted values corrected to molal scale. The accepted values reported in Appendix A are expressed on the molar or molal scales, depending on which units were used originally by the authors. (c) Re-evaluated value, see Appendix A. (d) Corrected to / = 0 by means of the SIT.
A single least squares SIT analysis was done using all the accepted (appropriately weighted) constants for 25°C in perchlorate and chloride media (Figure V-4 and Figure V-5). The available data for chloride media are somewhat less consistent than those for perchlorate media. This analysis results in the selected value of: log10 *P°X ((V.23), 298.15 K) = - (9.54 ± 0.14).
V.3 Oxygen and hydrogen compounds and complexes
95
In this analysis, the value recalculated from Perrin [64PER] for / = 0 and 25°C (weighted according to its estimated uncertainty) was also included (see Appendix A). From the slopes in Figure V-4 and Figure V-5, A E ( C 1 O ; ) = - (0.088 + 0.061) kg-moP1 and As(CF) = - (0.063 + 0.072) kg-mol"1 are calculated. Using the NEA-TDB values of 0.37, 0.17, 0.14 and 0.12 kg-moP1 for e(Ni2+,ClO^), e(Ni2+,Cr), E ( H , C 1 O 4 ) and e(H + ,Cr), respectively, leads to the selected values: e(NiOH+, ClO^) = (0.14 ± 0.07) kg-mol"1 and e(NiOH+, Cl") = - (0.01 ± 0.07) kg-moP1 . From the selected log10 *J3°, value, ArG^((V.23), 298.15 K) = (54.5 ±0.8) kJ-mol"1 can be derived. The Gibbs energy of formation of NiOH+ is calculated using the selected values for Ni2+ and H2O(1). AfG°(NiOH+, 298.15 K) = -(228.5 ±1.1) kJ-mol"1. When the discussion of Perrin [64PER] in Appendix A is also considered, the s(NiOH+, X") values (X = C l , NO,, ClO^) show a larger than expected spread - 0.01, - 0.27, 0.14 kg-mol"1, though the value for e(NiOH+, NO,) is not a selected value in the present review. The log10 */?°, ((V.23), 298.15 K) value reported above is somewhat higher than recommended by Baes and Mesmer (- (9.86 ± 0.03), [76BAE/MES]), but agrees well with the one obtained by [98PLY/ZHA] viz. - (9.50 ± 0.36). From the temperature dependence of the first hydrolysis constant determined by Perrin [64PER], A r //° = (51.7 ± 3.4) kJ-mol"1 can be obtained. The re-evaluation of the data reported in [68ARN] yielded A r // m = (54.3 ± 2.0) kJ-mol"' for 3.2 m NaCl medium. Combining the accepted hydrolysis constants of [65BUR/LIL2] and [71BUR/ZIN], A r // m = (59+10) kJ-mol"1 can be calculated for the formation of the species NiOH+ in 3.2 m NaCl medium. From the log]0 /?,, values reported for 150 300°C in [80TRE/LEB2], ArHm= (87 ± 15) kJ-mol"1 can be derived, but this value was not considered in this review, as it is probably valid only in the temperature range covered by the measurements. The temperature variation of log10 /?,, values accepted in [63BOL/JAU] is small compared to the uncertainties in the values. Consequently, no enthalpy value has been derived from these constants. As the relatively high uncertainties of the remaining three data cannot be used to extract a reliable ionic strength dependence, it was assumed to be negligible. Assigning statistical weights to the data , the value of: A,H°m ((V.23), 298.15 K) = (53.8 + 1.7) kJ-mol""1 is selected. The standard enthalpy of formation of NiOH+ is: A f //° (NiOH+, 298.15 K) = - (287.0 ± 1.9) kJ-mol"1.
96
V Discussion of data selection for nickel
Figure V-4: Extrapolation to /,„ = 0 of the experimental data for Reaction (V.23) in NaC104 media.
Figure V-5: SIT analysis of the experimental data for Reaction (V.23) in chloride media, including logm *#" ((V.23), 298.15 K) obtained in NaC104.
V.3 Oxygen and hydrogen compounds and complexes
97
The dinuclear Ni2OH3+ species is always a minor component, and its formation is considered to account for small deviations between the observed and calculated titration curves. Therefore, the rather wide scatter of the reported values is not surprising. As stated in [76BAE/MES], the reported constants for this minor species should be considered as the upper limits for its stability, if indeed such a species is present. In some cases, the accuracy of the reported experimental data was insufficient to take into account the formation of Ni2OH3+ during the re-evaluation [63BOL/JAU], [75CLA/KEP], thus all accepted constants refer to high ionic strength. Therefore, the extrapolation to lm= 0 was made assuming e(Ni2OH3+, ClO^) = (0.50 + 0.15) kg-mol"', based on the estimated value for e(Be2OH3+, C1O4), see Table B-4, (e(Ni2OH3+, C1O4) = (0.59 ± 0.16) kg-moF1 is reported in [98PLY/ZHA], based on the corrected data from [59ACH], [63BOL/JAU], [75CLA/KEP]). In this way, Ae((V.24), C1O~) is estimated to be - (0.1 ± 0.2) kg-mor', which leads to the selection: log10 V?,°2((V.24),298.15K) = -(10.6+ 1.0). From this value, ArG° ((V.24), 298.15 K) = (60.5 ± 5.7) kJ-moP1 and A f G°(Ni 2 0H 3+ , 298.15 K) = -(268.2 ± 5.9) kJ-mol"' can be derived. Two reliable values are available concerning the enthalpy of reaction to form Ni2OH3+. From the re-evaluation of the calorimetric data reported in [68ARN] A r // m = (45.9 ± 1.6) kJ-mol"1 can be calculated. Combining the accepted hydrolysis constants of [65BUR/LIL2] and [71BUR/ZIN], ArHm= (20 + 20) kJ-mol"1 can be obtained. Both values refer to 3.2 m NaCl medium. The more precise calorimetric value is accepted, but with an increased uncertainty, and negligible dependence on the ionic strength is assumed. A r #° ((V.24), 298.15 K) = (45.9 + 6.0) kJ-mol'1 is selected in this review, which leads to: A f #° (Ni2OH3+, 298.15 K) = - (350.0 ± 6.3) kJ-mol"1. At Ni(II) concentrations higher than 0.005 M, the Ni4(OH)4+ complex is dominant in the acidic pH region. Most of the accepted values concerning the formation of the tetrameric hydroxo species have been re-evaluated [63BOL/JAU], [65BUR/LIL], [65BUR/LIL2], [71BUR/ZIN], [75CLA/KEP], [78BUR/KAM] and corrected considering complexation with the background anion [65BUR/LIL2], [71BUR/ZIN], [71OHT/BIE], [78BUR/KAM]. The SIT analysis of the accepted constants for perchlorate medium (Figure V6) resulted in the selected value of: logl0 *J3°4 ((V.25), 298.15 K) = - (27.52 ± 0.15).
98
V Discussion of data selection for nickel
From the slope in Figure V-6, Ae((V.25), CIO;) = (0.16 ± 0.05) kg-mol"1 can be calculated. Using the selected values for e(Ni2+, C1O4) and e(H+, CIO;), Ae((V.25) ClO^) leads to a value of e( Ni 4 (OH) 4 + , CIO;) = (1.08 ± 0.08) kg-mol"1. In sodium chloride medium, only two values are accepted at 25°C [65BUR/LIL2], [71OHT/BIE]; both are valid for /,„ = 3.2 mol-kg"1. Combining them with the above selected log10 */?°4 ((V.25), 298.15 K), Ae((V.25), C\~) = (0.23+0.07) kg-mol"1 can be derived, which leads to a value of e(Ni4(OH)^+ ,C1~) = (0.43 ± 0.08) kg-mol"1. A similar treatment, using the value accepted for NaBr medium, results in 1 + AE((V.25), Br ) = (0.11 ± 0.07) kg-mol" . Since no value is currently available for e(H , + + Br"), only an estimate can be given for £(Ni4(OH)4 ,Br"). Assuming e(H ,Br") = e(H+,CF), e( Ni4(OH)4+ ,Br") = (0.71 ±0.12) kg-mol"1 can be estimated.
Figure V-6: Extrapolation to Im - 0 of the experimental data for Reaction (V.25) in perchlorate media.
V.3 Oxygen and hydrogen compounds and complexes
99
From the selected log,0 *#,°4 value, ArG° ((V.25), 298.15 K) = (157.1 ± 0.9) kJ-moF can be obtained. The Gibbs energy of formation of Ni4(OH)4+ is 1
A f G°( Ni 4 (OH) 4 + , 298.15 K) = - (974.6 ± 3.2) kJ-mol"1. From the temperature dependence of log10 /?4 4 values recalculated from the data in [63BOL/JAU], ATHm= (174+ 15) kJ-mor1 can be derived for 1.0 M NaClO4 medium (see Appendix A). The re-evaluation of the calorimetric data reported in [68ARN] yielded ArHm = (192.9 + 1.0) kJmor 1 for 3.2 m NaCl medium (see Appendix A). Combining the accepted hydrolysis constants of [65BUR/LIL2] and [71BUR/ZIN], A r // m = (186 + 6) kJ-mol ' can be obtained for the formation of Ni4(OH)4+ species in 3.2 m NaCl medium. Based on these three values in two different media, and the probability that the most accurate result is the value determined calorimetrically in 3.2 m NaCl(aq), the value: Ar//n°,((V.25), 298.15 K) = (190 ± 10) kJ-mor1 is selected in this review, which leads to A f //°(Ni 4 (OH) 4 + , 298.15 K) = -(1173.4+10.6) kJ'mor 1 . Electrostatic models such as the one presented in [94PLY/GRE] have been used to estimate the temperature dependence of formation constants and entropies of hydrolysis reactions of polynuclear hydroxo complexes, such as Ni4 (OH) 4 + . Such semi-theoretical treatments are beyond the scope of this review.
V.3.1.2
Hydroxo complexes in alkaline solutions
Three further complexes have been reported to form in Ni(II) solutions above ca. pH = 9: Ni(OH)2(aq), Ni(OH)3 and Ni(OH)4"", based on the increasing solubility of Ni(OH)2(cr) in alkaline solutions. According to a re-evaluation, carried out in this review, only the hydrolysis constant of the second species (V.27) can be determined with reasonable certainty: Ni2+ + 2H2O ^ Ni(OH)2(aq) + 2H+;
*/32]
(V.26)
Ni2+ + 3H2O ^
*J33,
(V.27)
*j34,.
(V.28)
Ni
2+
+ 4H 2 O ^
Ni(OH)~ + 3H+; +
Ni(OH) 4 " + 4H ;
The values of these hydrolysis constants selected by [76BAE/MES] and [98PLY/ZHA], (see Appendix A) are based on a few experimental points of a single paper [49GAY/GAR]. These data can, however, be equally well described when only Ni(OH)j is assumed to be present.
100
V Discussion of data selection for nickel
Three publications [49GAY/GAR], [80TRE/LEB2], [89ZIE/JON] report thermodynamic properties for the Ni(OH)2(aq) and Ni(OH)3 species. The hydrolysis constants for reactions (V.26) and (V.27) can be derived from solubility equilibria: Ni(OH) 2 (cr)^Ni(OH) 2 (aq)
K°sl,
(V.29)
Ni(OH)2(cr) + OH>± Ni(OH)~
K°xx
(V.30)
Ni(OH) 2 (cr) + 2H+ ^ Ni 2+ + 2H2O
*K°0
(V.31)
with
and assuming that hydrolysis of Ni + to NiOH+ was essentially complete in the experimental solutions used. Several hydrolysis models can be considered in interpreting nickel hydrolysis data obtained from experiments in basic solutions (see Figure V-7).
Figure V-7: Solubility of Ni(OH)2(s) in alkaline solutions at 298.15 K. • data from [49GAY/GAR], estimated error: + 0.5 log10 units, dash-dotted line: logi0[Ni(II)]tot = Iog 1 0 *; 3 1 + lo gl0 [OH-] mi , dotted line: log10 [Ni(II)]tot = log l 0 /C 3 , - Ae-[OHl ni + log l0 [OHl ni ; solid line: logI0[Ni(II)],ol = log10 (*,%,, + K°^ [OH^]in, 1 0 A E [ O H i-)
V. 3 Oxygen and hydrogen compounds and complexes Gayer and Garrett [49GAY/GAR] reported log10X°,o = ( 1 0 - 8 ± °-15)> log10 K°2, ~ - 7 and log10 K°M = - (4.2 ± 0.6). By combining these values, and the ionisation constant of water (pKw = 14.0), log10 */?2°i = - (17.8 + 1.0) and log10 7?3°i = - (29.0 ± 0.8) can be derived for reactions (V.26) and (V.27), respectively. Reevaluation of the solubility data reported by these authors in alkaline solutions, using the SIT approach, assuming formation of both Ni(OH)j and Ni(OH)2(aq) (but excluding the solubility values obtained in 8 M, 10 M and 15 M NaOH(aq)), results in log10 AT°2J ~ - (7.20 ± 0.22), log.oC.3,! = -(4.58 ±0.11), and Ae = (0.84±0.24) kg-mol"'. These equilibrium constants correspond to log10 P2X = —(18.00±0.30) and l o g . o X =-(29.38 ±0.30)'. A re-evaluation of the solubility data, using the SIT approach, and assuming Ni(OH)j (but not Ni(OH)2(aq)) is formed in the dissolution process, results in log.oA:^, = -(4.40 ±0.12), As = (0.66 + 0.17) kg-mol"1. The equilibrium constant corresponds to logl0 /?3°, = - (29.20 + 0.48). If activity coefficient effects are neglected, logI0 A"°3i1 = - (4.63 ± 0.22) and log10 *f5°x = - (29.43 + 0.48). Tremaine and LeBlanc obtained, logI0 *fl2l = - (20.26 + 1.77) and log10 *y8,°, = -(29.78 ± 1.97) at 298.15 K, by extrapolation, based on the solubility of NiO in NaOH solutions measured between 150 and 300°C [80TRE/LEB2]. A re-evaluation revealed that their experimental data can also be explained with Ni(OH)j as the only solubility controlling species in alkaline solutions (see Appendix A). The revised extrapolation to T = 298.15 K resulted in log10 *fi°A = - (30.9 ± 1.3), in reasonable agreement with [80TRE/LEB2], although Ni(OH)2, was neglected. Ziemniak et al. [89ZIE/JON] also measured the solubility of NiO between 18 and 289°C in alkaline solutions containing HPO4" ion. A re-evaluation of their data by [95LEM] resulted in log10 */?°, = - (19.77 ± 1.00) and log10 *ft°A = - (30.9 ± 1.3). As the numerical value of log10 Plx remains within its error limits whether Ni(OH)2 is taken into account or not, no thermodynamic quantities for the latter species are selected in this review. The solid line of Figure V-7 shows that a value of log10 K°2, = - (7.2 + 0.2) is also consistent with the data. No thermodynamic quantities for the latter species are selected in this review, though - 7 can be tentatively assigned as its upper limit. The value for logl0 /J,0, obtained by the re-evaluation of data reported in [49GAY/GAR] has been selected, but with an increased uncertainty to reflect the lower values obtained from use of the higher temperature data of [80TRE/LEB2] and [89ZIE/JON]: log10 X
1
((V.27), 298.15 K) = - (29.2 ± 1.7).
Gayer and Garrett [49GAY/GAR] did not report on the characterisation of their solid phase, which may, or may not, have been identical to p-Ni(OH)2, the solid discussed in Section V.3.2.2.1.1. Therefore, the value logl0 K°o = (10.8 ±0.15) was used for these recalculations, rather than the value for P-Ni(OH)2, log10 X°o = (11 02 ± 0.20), selected in the present review (also see the discussion of [49GAY/GAR] in Appendix A).
101
102
V Discussion of data selection for nickel
From this value, ArG° ((V.27), 298.15 K) = (166.7 + 9.7) kJ-mol"1 can be obtained, while the Gibbs energy of formation is AfG° (Ni(OH);, 298.15 K) = - (590.5 ± 9.7) kJ-mol"1. From the temperature dependence of log10 y92°, and log10 /?3°, measured between 150 and 300°C, ATHm(V.26) = (109 ± 15) kJ-mol"1 and A r # ra (V.27) = (119 ± 15) kJ-mor1 were obtained by [80TRE/LEB2]. From the re-evaluation of Ziemniak's data [89ZIE/JON] reported in [95LEM], A r //;(V.26) = (94 ± 10) kJ-mol"1 and A r //°(V.27) = (126 ± 15) kJ-mor1 can be estimated. The re-evaluation of the data of Tremaine and LeBlanc results in A r //° ((V.27), 298.15 K) = (121.2 ± 6.5) kJ-mor1 which agrees satisfactorily with the values given by [80TRE/LEB2] and [95LEM]. Although the quantities are not sufficiently accurate to allow selection of selected values for the standard reaction enthalpies in question (see above), they can be used as rough estimates. We have not found any convincing evidence for formation of Ni(OH)^~.
V.3.2 Solid nickel oxides and hydroxides V.3.2.1 Ni(II) oxide Bunsenite (NiO) is extremely rare, and a member of the periclase group of minerals. It had been discovered as early as 1868 in Johanngeorgenstadt, Erzgebirge, Saxony, Germany in a hydrothermal Ni-U vein. It is translucent to opaque, and occurs as wellformed fine sized crystals (up to 3 mm). Its hardness is 5.5, the colour pistachio-green, the lustre adamantine, and the streak brownish black. The lattice of NiO is face centred cubic, space group: Fm3m, with unit cell dimensions a0 = 4.177 A, Z = 4, />(calc.) = 6.806 gem"3 (according to JCPDS-ICDD card No.4-835, [53SWA7TAT]), p(obs.) = 6.4 - 6.8 g-cm~3. A very accurate technique for the determination of the standard entropy S^ (298.15 K) of a solid crystalline compound is the integration of low-temperature heat capacity data, C°pm, between 0 and 298.15 K: /•298.15 po
S°m (298.15 K ) =
-^-AT.
(V.32)
In the case of nickel oxide several publications dealing with heat capacity measurements in the temperature range 3.2 - 477.8 K are available [40SEL/DEW], [57KIN], [74WHI], [79DUB/NAU], [93WAT]. Seltz et al. [40SEL/DEW] investigated the heat capacity from 68 to 298 K, King [57KIN] performed low-temperature measurements between 54 and 296 K, and DuBose and Naugle [79DUB/NAU] determined the specific heat of NiO from 14 to 280 K. Whereas bulk-like NiO powder samples were employed in all these investigations, White [74WHI] used single crystals for the determination of C° m from 3.2 to 18.75 K. Furthermore, Watanabe [93WAT] measured
103
V.3 Oxygen and hydrogen compounds and complexes
the heat capacity of single crystals of NiO over the temperature range from 135.7 to 477.8 K. The experimental data of all studies mentioned above coincide remarkably, as can be seen in Figure V-8. The heat capacity values between 0 and 40 K are perfectly described by a Debye - T3 law. A fifth order polynomial fit to the experimental data in the range 40 - 477.8 K results in a C°m function which can be easily integrated. The smoothed C°m data as well as calculated entropy values are listed in Table V-5. The standard entropy of nickel oxide at 298.15 K, 5° (NiO), is equal to 38.4 JK^mol"'. This result is somewhat higher than the value (38.0 ± 0.2) J-KT'-mon1 obtained by King [57KIN] which was adopted by Kelley and King [61KEL/KIN], Kellogg [69KEL], the NBS tables [82WAG/EVA], Knacke et al. [91KNA/KUB] and Robie and Hemingway [95ROB/HEM]. As our value for 5° (NiO) is based on a simultaneous evaluation of five independent experimental studies of comparable accuracy, the selected standard entropy of NiO amounts to: S°m (NiO, cr, 298.15 K) = (38.4 ± 0.4) J-K^'mol"'. The uncertainty has been chosen, such that both values (the result of the present evaluation and the value of King [57KIN]) fall within the error limits.
Table V-5: Smoothed c r ) = (4110.64- 5.302412 (77K) + 3.52061xl0"3 (T/K)253039.297 (r/Kr 0 5 + 2.43067xl07(77K)-2)J-K"'-mor1
(V.35)
valid over the range from 298.15 to 519 K. Equations (V.34) and (V.35) describe the temperature dependence of the heat capacity above 298.15 K satisfactorily. Moreover, Equation (V.34) is consistent with the data of King and Christensen [58KIN/CHR] above 846 K, see the solid line in Figure V-8. Therefore, the Reviewers have selected the heat capacity function Equations (V.35) and (V.34), where C° m at 298.15 K amounts to: C; m (NiO, cr, 298.15 K) = (44.4 + 0.1) JKT'mor 1 . The uncertainty corresponds to the deviation between Equations (V.33) and (V.35) at 298.15 K. Boyle et ah [54BOY/KIN] determined the enthalpy of formation of nickel oxide directly by means of combustion calorimetry, A f //°(NiO) = —(239.7 ±0.4) kJ-mol"1. The application of a solid-state electrochemical cell to determine the Gibbs energy of formation of nickel oxide at high temperatures was originally introduced by Kiukkola and Wagner [57KIU/WAG]. By employing electrochemical cells of the type: Reference electrode | YSZ or CSZ | NiO, Ni, Pt
(V.36)
the temperature dependence of the Gibbs energy function of NiO was studied by a number of authors, where yttria- or calcia-stabilised zirconia served as the solid electrolyte and the reference electrode consisted of Pt, O2(air) or Fe, FeOx (wuestite) or Cu, Cu2O [63RAP], [65TRE/SCH], [65STE/ALC], [67RIZ/BID], [68CHA/FLE], [69MOR/SAT], [70HUE/SAT], [75MOS/FIT2], [76BER2], [78IWA/FUJ], [79KEM/KAT], [80TRA/BRU], [84COM/PRA], [84RAY/PET], [86HOL/NEI], [93NEI/POW], [94KAL/FRA]. Some of these references are discussed in Appendix A. Basically, the results of all these experimental studies agree within the limits of uncertainty. Especially, Charette and Flengas [68CHA/FLE], Berglund [76BER2], Comert and Pratt [84COM/PRA], Holmes et al. [86HOL/NEI] and O'Neill and Pownceby [93NEI/POW] performed highly reliable and precise electrochemical measurements of the Gibbs energy of formation as a function of temperature, as shown in Figure V-9. Equation (V.37) allows the calculation of the Gibbs energy of formation of NiO as a function of temperature:
106
V Discussion of data selection for nickel
AfGm(7) = A f //°(298.15K)-rA f ^(298.15K) + T J298.15
AfCp m
A{CpmdT-T( '
Pt
'
J298.15
dT
f
The auxiliary data for O2(g) at 298.15 K can be found in the NEA TDB auxiliary data set, heat capacities at higher temperatures were taken from CODATA [89COX/WAG] and from Robie and Hemingway [95ROB/HEM] in tabular and analytical form, respectively. When the selected data for Ni(cr), the calorimetric value for the standard enthalpy of formation of NiO, according to [54BOY/KIN], the heat capacity function and the standard entropy of NiO (selected above) are used, the predicted temperature dependence of the Gibbs energy of formation agrees remarkably well with experimental data obtained from various high-temperature electrochemical measurements, see Figure V-9. Thus, we select for the standard enthalpy of formation of nickel oxide at 298.15 K: A f //° (NiO, cr, 298.15 K) = - (239.7 ± 0.4) kJmol"' and the corresponding derived value for the Gibbs energy of formation at 298.15 K is: AfG° (NiO, cr, 298.15 K) = - (211.66 ± 0.42) kJ-mol""1.
Figure V-9: Comparison between the temperature dependence of the Gibbs energy of formation for NiO obtained from electrochemical as well as chemical reduction/oxidation equilibrium measurements and the prediction based on the present selection of thermodynamic properties for NiO.
V.3 Oxygen and hydrogen compounds and complexes
107
In addition, equilibrium constants for the reduction of NiO with H2 and CO are reported in [26PEA/COO], [73RAU/GUE] and [33WAT], [38BOG], [42FRI/WEI], [64ALC/BEL], [67ANT/WAR], respectively. The equilibrium constants for the reactions: NiO(cr) + H2(g) ^
Ni(cr) + H2O(g)
NiO(cr) + CO(g) ^
Ni(cr) + CO2(g)
can be transformed into the Gibbs energy of formation of NiO by using the pertinent auxiliary quantities for H2(g), H2O(g), CO(g) and CO2(g) listed in the CODATA tables [89COX/WAG] and Robie and Hemingway [95ROB/HEM]. The equilibrium data for the reduction/oxidation reactions given by [26PEA/COO], [33WAT], [42FRIAVEI], [64ALC/BEL], [67ANT/WAR], [73RAU/GUE] are in perfect agreement with calculated values for the Gibbs energy of formation using the thermodynamic data selected in the present compilation, as can be deduced from Figure V-9 (the results of [64ALC/BEL], [67ANT/WAR] are discussed in Appendix A). However, the data of Bogatzki [38BOG] deviate somewhat from the calculated line at lower temperatures, probably because thermodynamic equilibrium was not attained below 1000 K. The solid - gas equilibria: NiO(cr) + H2(g) ^ Ni(cr) + H2O(g) NiO(cr) + CO(g) ^ Ni(cr) + CO2(g) were likewise studied by [21WOH/BAL], [32SKA/DAB] and [36KAP/SIL], respectively, who obtained equilibrium constants deviating by approximately one order of magnitude from those predicted according to the thermodynamic model of the present assessment. The presumably erroneous results of these early works may be caused by (i) insufficient equilibration of the solid material with the gas phase, (ii) deposition of carbon by the Boudouard reaction, or (iii) inaccurate chemical analysis of the composition of the gas phase. Also, the relative mass change of NiO was monitored as a function of the oxygen partial pressure at 1473 K by means of thermogravimetry [89KIT]. Hence, the decomposition of NiO into O2(g) and Ni was found to occur at logi0(p(O2)/bar) = - 7.74 which concurs well with logio(p(02)/bar) = -7.84 calculated from the thermodynamic data of the present compilation. V.3.2.1.1
Solubility measurements
Due to the kinetically inert nature of nickel oxide with respect to its dissolution in aqueous media the solubility of NiO has been studied only at elevated temperatures so far [80TRE/LEB2], [89ZIE/JON]. These studies are not suitable for the calculation of any thermodynamic properties of NiO because of the high uncertainty of the measured solubilities compared to the high-temperature emf data and the low-temperature heat capac-
108
V Discussion of data selection for nickel
ity data discussed above. Moreover, the evaluation of the solubility experiments performed at hydrothermal conditions may cause an additional uncertainty for the solubility constant of NiO owing to the lack of heat capacity functions for the ionic species including the hydroxo complexes. Thus, the calculated value for the solubility constant of NiO at 298.15 K, according to the reaction: NiO(cr) + 2H+ ^ Ni2+ + H2O(1), derived from the thermodynamic data accepted in the present assessment, is : log10 *K°0 ( N i 0 > cr> 298.15 K) = (12.48 ± 0.15).
V.3.2.2
Ni(II) hydroxides, Ni(OH)2(cr)
Recently the thermodynamic property values of Ni(II) hydroxide have been reviewed by Archer [99ARC] and critically evaluated by Plyasunova et al. [98PLY/ZHA]. The data of Ni(OH)2(cr) reported are rather uncertain and, as this solid phase is of considerable importance for the deposition and remobilisation of nickel, they have been re-evaluated on the basis of recent solubility measurements [2002GAM/WAL]. V.3.2.2.1 V.3.2.2.1.1
Crystallography and mineralogy of nickel hydroxide p-Ni(OH)2
In a comprehensive review on bivalent metal hydroxides it was pointed out that the unit cell dimensions of synthetic P-Ni(OH)2 were determined by Lotmar and Feitknecht for the first time [36LOT/FEI], [77OSW/ASP]. Like most M(OH)2 hydroxides, the crystal structure is of the Cdl2- or brucite- type, trigonal, space group: P 3 m l , with unit cell dimensions a = 3.126 A, c = 4.605 A, Z= 1 (according to JCPDS-ICDD card No. 14-117). The natural occurrence of Ni(II) hydroxide in the Lord Brassey mine, Heazlewood, Tasmania, was first reported by Williams [60WIL2]. Almost pure Ni(OH)2 was found in the Vermion region, northern Greece, whereas Mg-bearing (Ni, Mg)(OH)2 occurs in Unst, Shetland, Scotland [81MAR/ECO], [82LIV/BIS]. The X-ray powder diffraction patterns of the former mineral match closely with synthetic p-Ni(OH)2. It was named theophrastite after Theophrastos, the first Greek mineralogist, 373/372-288/287 B.C. [81 MAR/ECO] (approved by IMA 1980). Theophrastite from the Vermion region is emerald green, translucent with vitreous luster and has a pale green streak. The Mohs hardness is 3.5. The specific gravity is /Xobs.) = 4.0 and p(calc.) = 3.95 g-cm~3. It is a gangue mineral in ore consisting of magnetite (Fe3O4), chromite (FeCr2O4) and Nisulphides as minor component. Theophrastite is formed from Ni-bearing solutions between 80 < ?(°C) < 115 in alkaline moderately oxidising media [83ECO/MAR].
V. 3 Oxygen and hydrogen compounds and complexes
V.3.2.2.1.2
109
a-Ni(OH)2
Other varieties of crystallised divalent nickel hydroxide, a-Ni(0H) 2 and a*-Ni(0H)2, differ from the thermodynamically stable P-form by the presence of a layer of water molecules in the van der Waals gap. Bode et al. [66BOD/DEH] proposed an a-3Ni(OH)2-2H2O unit cell (a = 3.08 A, c = 8.09A), whereas Braconnier et al. [84BRA/DEL]' synthesised a compound with a stoichiometric composition of a*-Ni(OH)2-0.75H2O and the following unit cell dimensions: a = 3.08 A, c = 23.41 A. In any case the intercalated crystal water of a-Ni(OH)2 causes an extension of the structure along the crystallographic c-axis, whereas the other dimension remains similar to that of theophrastite. The a-, a*- and P-forms of nickel hydroxide have the same trigonal structure. The P-form consists of well-ordered brucite type layers with a separation of 0.46 nm, whereas the hydrated a-form shows larger layer distances and turbostratic character as a result of intercalated water and distorted stacking, respectively. The a-form may also intercalate different anions so that the interlayer separation depends on the type as well as the extent of these ions and varies between 0.8 and 0.9 nm. By electron microscopy the very small crystallites appear as formless aggregates [84BRA/DEL]. Although it plays an important role in the charge/discharge cycle of nickel batteries no thermodynamic data can be definitively assigned to it (see Sections V.2.2 and V.3.2.3 about Ni(III, IV) ions and hydroxides). The natural occurrence of a—Ni(OH)2 has never been reported as it is probably too unstable to persist under ambient conditions. V.3.2.2.2
Heat capacity and entropy of Ni(OH)2(cr)
Sorai et al. [69SOR/KOS] measured the heat capacity of Ni(OH)2(cr) at low temperatures. These data have been used to determine the standard entropy, 5° (298.15 K), which has been accepted for the selected thermodynamic data of nickel compounds [76MAH/PAN]. Sorai et al. [69SOR/KOS] investigated three samples, (iii), (ii), and (i), of Ni(OH)2(cr) with different particle sizes, and reported C°m values up to 104K and 300 K for the coarsest (iii) and the finer crystals (ii, i), respectively. As shown in Figure V-10 the heat capacity and consequently the entropy turned out to depend on the particle size and thus on the specific surface area s. When the excess surface heat capacity estimated by Sorai et al. between 80 and 130 K was assumed to remain constant up to 300 K values of C°m(Ni(OH)2, cr, 298.15 K) = 81.7 J-K"'-mor' and S°m (Ni(OH)2, cr, 298.15 K) = 79.4 J-KT'-mor1 have been extrapolated for the coarsest material. When, however, the entropy increases linearly with s, extrapolation to s = 6 m2g~' (according to Sorai et al. the specific surface area of the coarsest sample) results in 5° (Ni(OH)2, cr, 298.15 K) = 80.0 JK^moP 1 , in
The latter phase has been called a* to emphasise that the composition and the intersheet distance are close to those of the turbostratic a-Ni(OH)2, whereas the texture and structure are different.
110
V Discussion of data selection for nickel
perfect agreement with the value selected by Mah and Pankratz [76MAH/PAN]. This value has been retained, but the uncertainty was increased from ± 0.4 to + 0.8: S°m (Ni(OH)2, p, 298.15 K) = (80.0 ± 0.8) JK~'mor'.
(V.38)
These increased error limits probably account also for the uncertainty introduced by Enoki and Tsujikawa's slightly differing results for bulky Ni(OH)2 between 4.2 to 35 K [78ENO/TSU], see Figure V-10. In addition the interpolated C° m value for sample (ii) at 298.15 K has been preferred to the extrapolated one of sample (iii). An uncertainty has been selected, so that both values fall within the error limits: C°(Ni(OH) 2 , p, 298.15 K) = (82.01 ± 0.30) .MC'-moP1.
Figure V-10: The heat capacity of Ni(OH)2(cr) as a function of temperature (A s(Ni(OH)2(iii)) = 6 m V [69SOR/KOS]; O s(Ni(OH)2(ii)) = 64 to 75 m2-g"' estimated by the reviewer; x s(Ni(OH)2(i)) = 320 m2-g"' [69SOR/KOS]; thick line: bulky Ni(OH)2 [78ENO/TSU]).
111
V.3 Oxygen and hydrogen compounds and complexes
V.3.2.2.3
Thermodynamic analysis of solubility data of Ni(OH)2(s)
For a methodological review of the experimental determination of solubilities of sparingly soluble ionic solids in aqueous media see [2003GAM/KON]. All auxiliary variables discussed in Section V.2.1 were used in the thermodynamic analysis of solubility data. The determination of the solubility product of Ni(OH)2(cr) is in principle a straightforward method to evaluate the standard Gibbs energy of formation, Af G° (Ni(OH)2, cr). It presumes, however, that the solubility product (defined by Equation (V.39)) or any appropriate solubility constant (as for example defined by Equation (V.40)) of the respective phase has indeed been determined. N i 2 + + 2OH"
Solubility product:
P-Ni(OH) 2 ^
Dissolution:
P-Ni(OH) 2 + 2 H
+
^ N i 2 + + 2H2O(1)
Ks0
(V.39)
*Ks0
(V.40)
Standard thermodynamic quantities from solubility measurements on hydroxides can be obtained using Equations (V.41) - (V.45). A
solGm
=
A
s0|/j^
- T
A
sol^
(V.41)
= -R
(V.42)
(51nX°o/57)
A
Af G°(Ni(OH) :.,P)=A f G°(r •Ji2+) 4-2AfG°JH2O,l) - A S O , G : A (H°m(Ni(OH) 2,P)=A f //:( Ni ) + 2A f //:(H 2 O, 2+
S !(Ni
2+
)=S° (Ni(OH)2,p)--2 5° (H 2 0,1) + AM|S,'n
/
Sol
(V.43) m
(V.44) (V.45)
As discussed in Section V.3.2.2.2, the entropy of nickel hydroxide was calorimetrically measured with sufficient accuracy [69SOR/KOS], [78ENO/TSU] that, in fact, the value of the entropy of dissolution contributes to the knowledge of the partial molar entropy 5° (Ni2+), see Equation (V.45). Actually most solubility data of Ni(OH)2(cr), reported so far and listed in Table V-6, suffer from an uncertainty in the physical state of the solid investigated [18ALM], [25BRI], [25WIJ], [38OKA], [43NAS], [49GAY/GAR], [50AKS/FIA], [53SCH/POL], [54DOB], [54FEI/HAR], [56CUT/KSA], [56JEN/PRA], [67BLA], [69MAK/SPI], [73KAW/OTS], [73NOV/COS], [80CHI/SAB], [96POU/DRE]. This is particularly surprising for Feitknecht and Hartmann [54FEI/HAR], because Lotmar and Feitknecht [36LOT/FEI] were the first to determine the structure of P-Ni(OH)2, and the concept that only well defined solid phases result in reliable solubility constants of metal oxides and hydroxides was developed by Schindler in the same laboratory [63SCH], [63FEI/SCH]. Moreover, the comprehensive review on Ni(II) hydroxide mentioned in Section V.3.2.2.1, is also based to a considerable extent on the work performed by Feitknecht and his co-workers [77OSW/ASP]. There it has been clearly pointed out that, apart from the well defined P-Ni(OH)2, a number of basic salts of changing com-
112
V Discussion of data selection for nickel
position exist. When nickel hydroxide is precipitated from aqueous NiCl2, Ni(NO3)2, or NiSO4 with NaOH or KOH solutions it is always contaminated with basic salts. The solubility of the latter varies depending on the anion and the molar ratio OH/Ni. This means that solubility studies on poorly defined nickel hydroxide or the respective basic salts are useless as an experimental basis to derive accurate thermodynamic functions of nickel hydroxide. They may, however, serve to find out relevant information concerning removal of Ni2+ from radioactive effluents. Sometimes solubility products have to be rejected, because they were calculated on the basis of quite inadequate experimental results, see Appendix A [18ALM], [53SCH/POL], [67BLA], [80CHI/SAB]. One of the most careful solubility studies reported so far was carried out by Mattigod et al. [97MAT/RAI], but according to their preparative method the respective results refer to a microcrystalline P-Ni(OH)2 that probably contained chloride. It is noteworthy that Poulson and Drever's [96POU/DRE] study of a commercially available nickel hydroxide (Johnson Matthey Chemical Co.) resulted in a solubility constant which after being extrapolated to the same temperature (log10 K°o = 12.08) agrees within 0.18 log,0-units with that of Mattigod et al [97MAT/RAI]. A re-evaluation of AfG° and A f //° of P-Ni(OH)2 has been based on the results of [2002WAL/GAT] and [2002GAM/WAL]. In Figure V-ll, the solubility constant log10 Ks0, is plotted as a function of temperature. It is quite obvious that after having rejected the most unreliable results [18ALM], [25WIJ], [67BLA], [80CHI/SAB], the maximum error limits according to Plyasunova et al [98PLY/ZHA] turned out to be too pessimistic—albeit the usual experimental error is exceeded by an order of magnitude. The results of several authors scatter closely around the curve adjusted to logI0 K°o of Mattigod et al. which seems to describe the solubility of finely divided P-Ni(OH)2, [25BRI], [54DOB], [56JEN/PRA], [96POU/DRE], [97MAT/RAI], whereas still higher solubilities presumably refer to amorphous Ni(OH)2 or even basic salts. Measurements carried out with carefully prepared coarse theophrastite led to lower solubilities, closer error limits, and a comparatively reliable prediction of thermodynamic quantities [2002GAM/WAL].
113
V.3 Oxygen and hydrogen compounds and complexes
Table V-6: Literature data(a) for log10 K°o (Ni(OH)2). Medium
/(molkg1)
NiCl2, NaCl 0.030-0.065 Ni(NO3)2,
0.3-0.4
T(K) 1Og.o X°c, (a) Method 291.15 298.15
12.20 13.36
NH4NO3
Remarks
Reference
H, titr.
Ni(OH)2-»Cl,(s)?
[25BRI]
[OH]->
incipient ppt.
[25WIJ]
[NH 3 ]/[ N H ; ]
Ni(NO3)2
dilute
298.15
13.50
pH, titr.
incipient ppt.
[38OKA]
NiCl2, KC1
0.002-2.05
298.15
12.92
H, titr.
Ni(OH)2-,Cl,(s)?
[43NAS]
NiCl2
0.038-0.15
298.15
10.78
pH, [Ni(II)]
saturated?
[49GAY/GAR]
NiCl2
0.038-0.15
308.15
10.78
PH, [Ni(II)]
sat. at 35°C ?
[49GAY/GAR]
NiSO4
0.006-5.37
291.15
13.40
H, QH(b), Sb(c) Ni(OH)2 pptd.
NiSO4
0.030-0.065 348.15
9.48
pH, titr.
Figs. 3 and 5.1
[54DOB]
NiCl2
0 corr
298.15
13.30
pH, ?
Ni(OH)2, fresh
[54FEI/HAR]
NiCl2
0 corr
298.15
10.80
pH,?
Ni(OH)2, aged
[54FEI/HAR]
[50AKS/FIA]
NaClO4
1.05
293.15
12.56
pH, [Ni(II)]
incipient ppt.
[56CUT/KSA]
NiCl2
0.028-0.28
298.15
12.12
PH, [Ni(II)]
Ni(OH)2 pptd.
[56JEN/PRA]
302.15
11.75
pH, rNi(II)]
Ni(OH)2 pptd.
[56JEN/PRA]
NiCl2
0.001 - 4 . 7 M 298.15
12.46
pH, [Ni(II)]
incipient ppt.
[69MAK/SPI]
incipient ppt.
[73KAW/OTS]
CH3COONa 0.10-0.50
LiClO4
3.48
298.15
12.99
n
NaCl
0.56
298.15
12.49
pH, [Ni(II)]
slow ppt.
[73NOV/COS]
NaCl
0.56
298.15
10.89
pH, [Ni(II)]
fast ppt.
[73NOV/COS]
NaC104
1.05
298.15
10.38
pH, [Ni(II)]
fast ppt.
[73NOV/COS]
NaNO3
0.02-0.2
295.15
12.24
pH, rNi(II)]
Johnson Matthey
[96POU/DRE]
NaClO4
0.01
298.15
11.90
PH, [Ni(II)]
X-ray,p- Ni(OH)2
[97MAT/RAI]
NaClO4
0.5 - 3.0, SIT 298.15
11.02
pH, [Ni(II)]
X-ray, P- Ni(OH)2
[2002GAM/WAL]
n
7
p H , Zmax
(d)
(a) Whenever possible the entries of column 4 have been recalculated from the experimental data. (b) quinhydrone electrode. (c) antimony electrode. (d) maximum number of H+ set free per Ni.
114
V Discussion of data selection for nickel
Figure V-ll: Solubility constant of Ni(OH)2 as a function of temperature. Thick solid curve: optimised from experimental data of Gamsjager et al. [2002GAM/WAL]; thin solid curves: corresponding error limits; thick dash-dotted curve: calculated according to [98PLY/ZHA]; thin dash-dotted curves: corresponding maximum error limits; thick dash curve: calculated with A f //° (Ni(OH)2, cr) adjusted to coincide with the result of Mattigod et al. [97MAT/RAI]; A [25BRI], ® [38OKA], A [43NAS], V [49GAY/GAR], • [50AKS/FIA], • [54DOB], x [54FEI/HAR] (aged), • [54FEI/HAR] (freshly prepared), 0 [56CUT/KSA], * [56JEN/PRA], 6 [56JEN/PRA], o [69MAK/SPI], T [73KAW/OTS], + [73NOV/COS] (high value: slow, lower values: fast precipitation), 8 [96POU/DRE], D [97MAT/RAI], 0 [2002GAM/WAL].
In Table V-7 the thermodynamic properties of Ni(OH)2 obtained in [2002GAM/WAL] and selected by Plyasunova et al. are compared. The latter authors overlooked a direct calorimetric determination of S° of Ni(OH)2(cr) [69SOR/KOS], [76MAH/PAN] which has been discussed in Section V.3.2.2.2 and Appendix A ([69SOR/KOS], [78ENO/TSU]). For the optimisation algorithm applied, the calorimetrically measured S°m (p-Ni(OH)2) = 80.0 J-K^mol"' (Equation (V.38)) and S°m (Ni2+) = - 131.8 J-KT'mor1 (Equation (V.I5)) were kept fixed. Based on solubility data, the
V.3 Oxygen and hydrogen compounds and complexes
115
precision and reliability of AfG° and A f //° (p-Ni(OH)2) can be improved considerably, leading to the selections: A f G;(Ni(OH) 2 , p, 298.15 K) = - (457.1 ± 1.4) kJ-mol"1 A f //° (Ni(OH)2, P, 298.15 K) =-(542.3 + 1.5) kJ-moP1.
Table V-7: Thermodynamic properties of Ni(OH)2(cr). !
Compound
°g|0 X°.O (V.40)
Ni(OH) 2 (cr) 52 K a transition of the antiferromagnetic to the paramagnetic state occurs [52BUS/GIA2]. V.4.1.2.1
Low-temperature heat capacity and entropy of NiCl2(cr)
Trapeznikova et al. measured the heat capacity of anhydrous Ni(II) chloride in the range of 14 to 130 K [36TRA/SCH]. Their results were not confirmed by Busey and Giauque who took great care to prepare high purity NiCl2 and measured its heat capacity between 15 and 300 K [52BUS/GIA2]. Kostryukova assessed the singularities of the magnetic energy spectrum of NiCl2 by heat capacity measurements at liquid helium temperatures [68KOS], [69KOS], [75KOS]. In Figure V-14 the smoothed values of Cpm/T are plotted versus T for temperatures up to 30 K. Figure V-14 clearly shows that the usual extrapolation to T= 0 with Debye's T 3 law is not applicable up to 15 K. Thus, in the present review the standard entropy of NiCl2 was re-evaluated by integration of the original and smoothed Cpm values listed in [52BUS/GIA2] and [75KOS] respectively. Two different linear spline functions, valid for the temperature range below and above the Neel temperature (r N = 52.32 K) were fitted to these data, shown in Figure V-15. The sum of these integrals resulted in the selected value: S° (NiCl2, cr, 298.15 K) = (98.14 + 0.30) J-mor'-K"1. This value is slightly higher than the one given in [52BUS/GIA2], which may be due to the more realistic extrapolation to T = 0 on the basis of Kostryukova's data [75K.OS] as well as to the influence of different fitting functions employed by [52BUS/GIA2] (unknown) and in the present review. The uncertainty selected is greater than the difference arising from using the more complete set of fitting functions. The standard heat capacity derived by interpolation of the fitted Cpmfunctions is essentially equal to the value given by Busey and Giauque (17.13 cal-moP'-KT1) who stated that near 300 K heat leaks may have introduced an error of several tenths of a percent into their Cp m -values. Thus the following value as well as its uncertainty are selected: C; m (NiCl 2 , cr, 298.15K) = (71.67 ± 0.30) J-moP'-K"1.
(V.55)
V.4 Group 17 (halogen) compounds and complexes
123
Figure V-14: Molar heat capacity of anhydrous NiCl2 between 2 and 30 K. —•— : smoothed values of [75KOS], — A — : smoothed values of [52BUS/GIA2] Dotted line: r ' - l a w up to 2 K, dash line T 3 -law up to 15 K, solid line: Cpm/T = 0.0011715-r 2 + 0.009121 I f up to 3.2 K [75KOS].
Figure V-15: Heat capacity of anhydrous NiCl2(cr). O: experimental data from [52BUS/GIA2]. X: smoothed data of [75KOS]. Dash line: linear spline T > TN, dotted line: linear spline T-(298.15)-2
(V.57)
data listed in From a least squares analysis of the (Hm(T)-H^(298A5)) [51COU] a C° m (T) -function was obtained which is numerically slightly different from the one given by Coughlin, but the calculated C° m -values differ only marginally. For the sake of consistency with the selected value of C°m(NiCl2, cr, 298.15 K) the following C° m (T) -function is selected: [c°P,Jim (NiCl2, cr) = 73.7949 + 1.22269* 1(T2 (T7 K) -5.12775xl0 5 -(77ICr 2 /J-mor l -K- 1 V.4.1.2.3
(V.58)
Enthalpy of formation of NiCl2(cr)
The enthalpy of formation of NiCl2 can be determined by a third-law analysis of the equilibrium studies on the reaction: NiCl2(cr) + H2(g) ^ Ni(cr) + 2 HCl(g)
(V.59)
carried out by Berger and Crut, [21BER/CRU], Crut, [24CRU], Jellinek and Uloth, [26JEL/ULO], Sano, [37SAN2], [53SAN], Busey and Giauque, [53BUS/GIA], as well as Shchukarev et al. [54SHC/TOL]. An inspection of Figure V-16 reveals that the data of [53BUS/GIA], [53SAN], and [54SHC/TOL] appear to be the most reliable and best suited for this analysis. This was confirmed by linear fits through the data pairs T, - RT-ln(Kp /bar) which resulted in lower standard errors for [53SAN], [53BUS/GIA] and [54SHC/TOL] than for [24CRU] or [26JEL/ULO]. An alternative method consists in subjecting high-purity nickel metal to chlorination in a calorimetric bomb according to reaction: Ni(cr) + Cl2(g) ^ NiCl2 (cr)
(V.60)
The advantage of the latter method is that the standard enthalpy of formation
V.4 Group 17 (halogen) compounds and complexes
125
A r //° (NiCb, cr, 298.15 K) can be determined directly at the reference temperature, [84LAV/TIM]. A third independent method has been developed by Efimov et al. and is described in Appendix A [88EFI/EVD].
Figure V-16: ArG° -function of Reaction (V.59). ArG° = -RTlnK°, linear fit of experimental T; experimental data from: • [53BUS/GIA], • [54SHC/TOL], • [37SAN2] [53SAN], V [26JEL/ULO], O [21BER/CRU] [24CRU]. ArG° data: solid line [53BUS/GIA], dotted line [53SAN], dashed line [54SHC/TOL]
The third-law analyses of the equilibrium constants obtained by Busey and Giauque [53BUS/GIA], Sano [53SAN] and Shchukarev et al. [54SHC/TOL] for reaction (V.59) are summarised in Figure V-17. A combination of the standard enthalpy of HCl(g) formation recommended by CODATA [89COX/WAG] with the reaction enthalpies of Figure V-17 resulted in A f //°(NiCl 2 , cr, 298.15 K)/kJmol"' = -(305.01 ±0.17) [53BUS/GIA]; -(305.5 + 2.4) [54SHC/TOL], -(307.07 ± 0.56) [53SAN]. Busey and Giauque's data show the least uncertainty. The results of Shchukarev et al. led to a similar value with a significantly enhanced uncertainty. Moreover, Busey and Giauque's as well as Sano's values scatter around the mean, whereas Shchukarev et al.'s results increase with increasing temperature.
126
V Discussion of data selection for nickel
In principle equilibrium constants of other reactions such as: NiO(cr) + 2 HCl(g) ^ NiCl2(cr) + H2O(g) can also be subjected to the third-law method, see Appendix A [77KAS/SAK]. Figure V-17: Third-law analysis of data for Reaction (V.59). Experimental data from • [53BUS/GIA], • [54SHC/TOL], and • [53SAN]. Solid line: A r / / ° / k J m O r ' = (120.41 + 0.10) [53BUS/GIA], dotted line: A r //° = (122.45 ± 0.56) kJ-mol"1, [53SAN], dash line: A r //° = (120.99 + 2.4) kJ-moP1 [54SHC/TOL], dash-dotted line: linear regression of [54SHC/TOL] data.
Lavut et al. [84LAV/TIM] determined the standard enthalpy of NiCl2 formation very accurately A f //° (NiCl2, cr, 298.15 K) = - (304.78 ±0.16) kJ-mol"1. This result overlaps with the value obtained from the third-law analysis of Busey and Giauque's data. Consequently the mean value of these investigations was selected in this review: A f //° (NiCl2, cr, 298.15 K) = - (304.90 ±0.11) Id-mot
(V.61)
V.4 Group 17 (halogen) compounds and complexes
V.4.1.2.4
127
Gibbs energy of formation of NiC^Ccr)
The Gibbs energy of formation of NiC^cr) can be calculated as a function of temperature using the selected values of A f i7°, S ° , C° m for NiCl2(cr) and for Ni(cr) as well as the respective CODATA values for Cl2(g) [89COX/WAG], adopted (Chapter VI) as NEA TDB auxiliary data (see Figure V-18). Gee and Shelton measured the emf of solid electrolyte cells and thus obtained, for example, ArG° of the displacement reaction (V.62) directly [76GEE/SHE]. Fe(cr) + NiCl2(cr) ^ FeCl2(cr) + Ni(cr)
(V.62)
The Gibbs energy of formation of FeCl2 appeared to be well established and thus the Fe, FeCl2 electrode was taken as the reference electrode. Figure V-18 shows that values measured in this manner and calculated Af G° (NiCl2, cr) values agree with each other satisfactorily. This is considered to be a strong argument in favour of the selected thermodynamic quantities for NiCl2(cr) and Ni(cr).
Figure V-18: Gibbs energy of formation of NiCl2(cr) versus temperature. Values (straight line) and uncertainty (dotted lines) derived from [76GEE/SHE]. Dashed line calculated from the selected values of A f // n ", S°m , C°m.
128
V.4.1.3
V Discussion of data selection for nickel
Hydrated NiCl2 solids
Thermodynamic data have been reported for the nickel chloride hexahydrate, tetrahydrate and dihydrate, and the existence of a heptahydrate has been reported at low temperatures (below - 30°C) [33BOY]. At 0.1 MPa, in contact with saturated aqueous solutions, the hexahydrate was reported [16DER/YNG] to be the stable form to approximately 36.25°C, and dehydration of the tetrahydrate to the dihydrate occurs above 60°C [33BOY]. The lower hydrates can also be prepared by equilibration with aqueous HC1 solutions at room temperature [23FOO]. V.4.1.3.1
Gibbs energy of formation of NiCl2-6H2O
Many solubility measurements for NiCl2-6H2O(cr) in water have been reported [33BOY], [37PEA/ECK], [79OIK], [79OJK/MAK], [79PET/SHE], [79PET/SHE2], [80PET/SHE], [83KIM/IMA], [85FIL/CHA], [86FIL/CHA], [87RAR3], [89LIL/TEP]. NiCl2-6H2O(cr) ^ Ni2+ + 2CP+ 6H2O(1)
(V.63)
Rard [92RAR2] has done an excellent assessment of the solubility of NiCl2-6H2O and the associated activity and osmotic coefficients of the saturated solution at 298.15 K. Based primarily on results from [37PEA/ECK], [87RAR3] and [92RAR2], Rard calculated ((4.9155 ± 0.0022) molkg 1 ), K° (V.63) = (1108 ±37) and ArG° (V.63) = - (17.38 + 0.08) kJ-mol"'. Rard's value is selected in the present review and, with the auxiliary data for CP (Chapter IV) and the selected value for AfG°(Ni2+, 298.15 K) (Table III.l), the value: AfG° (NiCI2-6H2O, cr) = - (1713.7 ± 0.8) kJ-mor1 is calculated. V.4.1.3.2
Enthalpy of formation of NiCl2-6H2O
Surprisingly, the only reported direct measurement of the heat of solution of the hexahydrate in water appears to be that of Thomsen [1883THO] (4.85 kJ-mol"1 at 18.5°C [20BUR]). Archer [99ARC] has reported that an attempt to correct this value to 25°C and infinite dilution (cf. Appendix A) resulted in the value 0.14 kJ-mol"1. Ashcroft [70ASH] reported the heat of solution of NiCl2-6H2O(cr) in 1 M HCl(aq) as (6.10 ±0.22) kJ-moI"1, and Gurrieri et al, [73GUR/CAL] reported (7.00 ± 0.13) kJ-mol"1. These two values do not agree within the authors' (2a) uncertainties, even though the final solution concentrations of the nickel salt were apparently quite similar in the experiments. Kakolowicz and Giera [75KAK/GIE] reported the heat of solution of NiCl2-6H2O(cr) in 4.36 M HCl(aq) as (21.32 ±0.13) kJ-mol"1.
129
V.4 Group 17 (halogen) compounds and complexes
The final nickel concentrations in Thomsen's experiments were quite high (approximately 0.14 m), thus applying a proper heat of dilution correction to determine A r //°(V.63) is problematic [99ARC]. The final nickel concentrations in the experiments of Ashcroft [70ASH] and Gurrieri et al. [73GUR/CAL] were about a factor of five lower, but to use these values a correction for transfer of the aqueous species from the dissolution medium to water is needed. The enthalpy of solution measurements of Manin and Korolev [95MAN/KOR] for dissolution of anhydrous nickel chloride in HCl(aq) indicate that if the final solutions have nickel chloride concentrations of less than 0.01 M, the enthalpy of solution of anhydrous nickel chloride is approximately 4 kJ-mor1 more positive than in water. This is in semi-quantitative agreement with results of the other measurements [1883THO], [70ASH], [73GUR/CAL]. Manin and Korolev [95MAN/KOR] also reported heats of solution of NiCl2(cr) in water as a function of the final solution concentration (for 0.0042 to 0.030 m). The variation in these values could be applied equally well to the heats of solution of the hydrated salt, but the experimental concentration range is not adequate to determine heats of dilution to / = 0 to better than ± 1 kJ-moP1. As discussed in Appendix A, using a reasonable value for the enthalpy of dilution, the heat of solution of NiCl2-6H2O(cr) to infinite dilution in water at 298.15 K can be estimated as (0.64 ± 1.80) kJ-mor1. From this, A f //° (NiCl2-6H2O, cr) = - (2104.41 + 2.01) kJmor 1 can be calculated. Another approach is to compare the heats of solution of the anhydrous salt and the hydrate in the same medium (with appropriate minor corrections for changes in the heat of formation of water in the medium). However, except for Thomsen [1883THO], none of these researchers reported enthalpy of solution values for the anhydrous solid, NiCl2(cr) and the hexahydrate in the same medium. As discussed in Appendix A, the heat of hydration of NiCl2(cr) to NiCl2-6H2O(cr) at 298.15 K can be estimated as - (85.06 + 1.80) kJ-mol 1 , and A f //° (NiCl2-6H2O, cr) = - (2104.9 ± 1.8) kJ-mol"1. In the present review, the weighted average of the two values based on the measurements of Thomsen [1883THO] is selected: A f //° (NiCl2-6H2O, cr, 298.15 K) = - (2104.7 + 1.8) kJ-moP1, but with an uncertainty somewhat larger than the statistical uncertainty because the values are based, in part, on the same measurement. V.4.1.3.3
The entropy and heat capacity of NiCl2-6H2O(cr)
Based on the selected values of AfG° (NiCl2-6H2O, A f //° (NiCl2-6H2O, cr, 298.15 K), the selected entropy is:
cr,
298.15 K)
and
5° (NiCl2-6H2O, cr, 298.15 K) = (341.0 ± 6.7) J-IC'-mor1. Robinson and Friedberg [60ROB/FRI] reported measurements of the heat capacity of NiCl2-6H2O(cr) between 1.6 K and 20.0 K, and found that the Neel tempera-
130
V Discussion of data selection for nickel
rare is at 5.34 K [73HAM/FRI]. Donaldson and Edmonds [65DON/EDM] extended the measurements to lower temperatures, with measurements between 0.7 K and 4 K. Similar measurements were reported by Okaji and Watanabe [76OKA/WAT]. Although Ko and Hepler [63KO/HEP] reported S°m (NiCl2-6H2O, cr, 298.15 K) = 344.3 kJ-mor1, based on unpublished Cp m data of Friedberg, the primary Cp m data apparently were never published. In the present review, no heat capacity value at 298.15 K is selected for NiCl2-6H2O(cr). V.4.1.3.4
Gibbs energy tetrahydrate
and
enthalpy
of formation
of
nickel
chloride
Measurements of water vapour pressures of mixtures of nickel chloride hydrates [16DER7YNG] and [40BEL], and over pure solids [93UVA/TIM] have been reported. Water vapour pressure values reported for the equilibrium: NiCl2-6H2O(cr) ^ NiCl2-4H2O(cr) + 2H2O(g)
(V.64)
by Derby and Yngve [16DER/YNG] and the equation of Bell [40BEL] between 20 and 36°C are in good agreement, and may even be based on the same experimental results (cf. Appendix A), whereas Uvaliev et ah [93UVA/TIM] reported slightly lower values. The latter (isopiestic) work was focussed on determining transition pressures at constant temperature (25 and 35°C) rather than on the temperature dependence of the water vapour pressure. The equilibrium water vapour pressure at 298.15 K is (1.30 ± 0.15) kPa, based on the average of the values reported by [16DER/YNG] (1.40 kPa) and Uvaliev et al. [93UVA/TIM] (1.19 kPa). The selected value is: ArG° (V.64) = (21.56 ± 0.60) kJmol"1. The uncertainty has been assigned to span the uncertainties in these inconsistent results. The enthalpy of reaction, determined only from the results of Derby and Yngve [16DER/YNG], is (102.4 ± 1.8) kJ-mot1; and (107.8 1 5.8) kJ-moF1 if the values from [93UVA/TIM] are included. If the water vapour pressure is fixed at the selected value at 298.15 K, the fitted slope of logl0 puo against (77K)~' leads to a calculated enthalpy of reaction of (111.2 1 5.6) kJ-mor'. In the present review a median value spanning all these values is selected: A r //° (V.64) = (107 + 9) kJ-mor1. Using the values selected above for the hexahydrate, and auxiliary data from Chapter IV, yields: AfG°(NiCl2-4H2O, cr, 298.15 K) = -(1234.91 1.0)kJ-mor' and A f //° (NiCl2-4H2O, cr, 298.15 K) = - (1514.0 1 9.2) kJ-mol"1.
131
V. 4 Group 17 (halogen) compounds and complexes From these and auxiliary values from Chapter IV, S°m (NiCl2-4H2O, cr, 298.15 K) = (250 ±31) J-IC'-mor1 is calculated. Bell [40BEL] reported on measurements of the vapour pressure of D2O over mixtures of the tetradeuterate and hexadeuterate, and it was found that the ratio of equilibrium values of pu o //J D , 0 was greater than 1.0 near 25°C. V.4.1.3.5
The entropy and heat capacity of nickel chloride dihydrate
Polgar et al. [72POL/HER] measured the specific heat of NiCl2-2H2O(cr) for temperatures between 1.2 and 24.5 K. Peaks were observed at 6.31 and 7.26 K, and at lower temperatures the heat capacity values are not a simple function of T3. The authors estimated that the contribution to the molar entropy between 0 and 1.2 K is 0.03 R. Using an adiabatic calorimeter, Juraitis et al. [90JUR/DOM] measured the heat capacity of the solid between 80 and 281 K, with emphasis on the effects of the structural transition near 220 K. It is not clear from these results (only reported graphically) whether the results from 250 to 281 K can be extrapolated smoothly to 298.15 K. A series of polynomials were fitted to the values of Cpm IT over the entire temperature range from 1.2 K to 281 K. The uncertainty in the extrapolation below 1.5 to 0 K is of the order of + 0.2 J-KT'-mor1. A single simple polynomial was fitted to the combined results from 20 to 25 K [72POL/HER] and 80 to 100 K [90JUR/DOM]. In the present review, it is assumed that there are no further anomalies in this temperature range, and the uncertainty contribution to 5° (NiCl2-2H2O, cr) over this range is estimated as ± 2.0 J-KT'-mor1. The uncertainties in the set of C°m values above 240 K (see Appendix A) independently introduce an uncertainty of ± 3.0 J-K^'-mol"' in the calculated value of 5 ° . In the present review, S°m (NiCl2-2H2O, cr, 298.15 K) = (186.2 ± 3.6) JKT'mor 1 is selected, where the uncertainty is an estimate. From the work of Juraitis et al. [90JUR/DOM] (see Appendix A), C; m (NiCl2-2H2O, cr, 298.15 K) = (230 + 15) J-KT'-mor1 is selected in the present review. Again, the uncertainty is estimated. V.4.1.3.6
Gibbs energy and enthalpy of formation of nickel chloride dihydrate
Derby and Yngve [16DER/YNG] and Uvaliev et al. [93UVA/TIM] both reported water vapour pressure measurements for the dehydration of nickel chloride tetrahydrate to the dihydrate. NiCl2-4H2O(cr) ^ NiCl2-2H2O(cr) + 2 H2O(1)
(V.65)
132
V Discussion of data selection for nickel
The values of logl0 pH 0 from the earlier study do vary monotonically with I IT, though they are in fair agreement with the results from Uvaliev et al. [93UVA/TIM] at 25 and 35°C. In the present review, the average of the discordant values from the two studies at 298.15 K is used to calculate a value for ArG° (V.65), but no value for A r //° (V.65) is selected, ArG° (V.65) = (23.4 + 0.7) kJ-moP1. Using the values selected above for the tetrahydrate, and auxiliary data from Chapter IV, AfG°(NiCl2-2H2O, cr, 298.15 K) = -(754.4 ± 1.3) kJ-mol"1. Based on this, auxiliary data from Chapter IV, and 5° (NiCl2-2H2O, cr, 298.15 K) from Section V.4.1.3.5, Afy/n°, (NiCl22H2O, cr, 298.15 K) = -(913.4 ± 1.7) kJ-mor1.
V.4.1.4
Solid nickel bromide, NiBr2(cr)
There are two different crystallographic forms of paramagnetic NiBr2(cr) [34KET]. Samples prepared by sublimation, or by melting and slow cooling, have a CdCl2 structure, rhombohedral space group R3m [33BIJ/NIE], [34KET], [80ADA/BIL]. Samples obtained by low-temperature dehydration of the hydrated salts or by recrystallisation of these solids from organic solvents have a "Wechselstruktur" [34KET]. Day el al. [76DAY/DIN] describe the structure as a disordered intergrowth of the CdCl2 and Cdl2 lattices [33BIJ/NIE]. Both forms have been used in experiments to determine the chemical thermodynamic properties of anhydrous nickel bromide, often without a recognition that two forms can exist. NiBr2(cr) orders to an antiferromagnetic form below 52 K, and to a helimagnetic structure below 23 K [83KAT/SUG]. V.4.1.4.1
Low-temperature heat capacity and entropy of NiBr2(cr)
There are two sets of low-temperature (8 to 300 K) heat capacity (adiabatic calorimetry) measurements [78STU/FER], [82WHI/STA] that are in good agreement between 30 and 250 K (as discussed in Appendix A). Both sets of results show a small excess heat capacity between 40 and 55 K, near the Neel temperature. White and Staveley [82WHI/STA] also found a small, sharp peak near 20 K, and apparently this is associated with the transition from an antiferromagnetic structure to an incommensurate phase with helimagnetic ordering [76DAY/DIN], [81ADA/BIL]. No corresponding peak was found in the earlier study [78STU/FER], and the reported heat capacities below 15 K decrease much more slowly with decreasing temperature than suggested by the results of [78STU/FER]. Unlike the results of White and Staveley, the values reported by Stuve et al. below 18 K do not fit well to a T 3 relationship (see Figure V-19). Above 250 K, the heat capacity values reported by White and Staveley increase rapidly with increasing temperature, and the authors suggested [82WHI/STA] that the sample may have been
V.4 Group 17 (halogen) compounds and complexes
133
contaminated by moisture (at 298.15 K, their value of C°pm (NiBr2, cr) is approximately 83 J-KT'-mor1, as opposed to 75.4 J-KT'-mor' from Stuve et al). Stuve et al. [78STU/FER] integrated their heat capacity values to obtain 5° (NiBr2, cr, 298.15 K) = (122.42 + 0.25) JKT'mor 1 . In the present review, the experimental Cpm(T) values reported in both papers were integrated. The results of White and Staveley [82WHI/STA] are accepted to 250 K (an entropy contribution of 108.23 J-KT'-mor1). The contribution to the entropy from 0 to 8 K (0.09 MC'-mol"1) was estimated by assuming C°m/T varies as T2 (Figure V-19), though there are no measurements to support this assumption. The results of Stuve et al. [78STU/FER] are accepted between 250 K and 298.15 K (an entropy contribution of 13.14 kJmol"'). The sum of these contributions provides the selected value: 5° (NiBr2,cr, 298.15 K) = (121.4 ± 1.0) HC'-mol' 1 . The uncertainty is an estimate based on the spread of the entropy contributions from the two studies in different temperature ranges, and considering that there may be an unknown contribution from the extrapolation from 8 to 0 K. The reported c;_m(NiBr2, cr, 298.15 K) value of Stuve et al. [78STU/FER] C; m (NiBr 2 , cr, 298.15 K) = (75.40 + 1.00) JK^'-mol"' is selected in the present review with a markedly increased uncertainty.
Figure V-19: Values from the heat capacity studies of Stuve et al. [78STU/FER] and White and Staveley [82WHI/STA]. The curves (from bottom to top) are based on the equation C° m (T) = bT2 for b = 0.000500, 0.000750, 0.001000, 0.001350 and 0.002000.
134
V Discussion of data selection for nickel
V.4.1.4.2
The high-temperature heat capacity of NiBr2(cr)
The fitted heat capacity equation reported by the same authors [78STU/FER] reproduces all their high-temperature experimental // m (7/)-//^ (298.15) values to better than 0.5% and is fixed to the 298.15 K value from the low-temperature measurements. The values from the equation deviate slightly from the experimental values below 298.15 K. If this heat capacity equation is used with the high-temperature equilibrium measurements of Shchukarev et al. [54SHC/TOL], the derived values of Af //° (NiBr2, cr, 298.15 K) vary with the measurement temperature (see the Appendix A discussion of [54SHC/TOL]). In the present review, this drift has been attributed to experimental difficulties in the work of Shchukarev et al, but it might also indicate problems with the drop calorimetry work of Stuve et al. It is also not clear whether the more stable (or the same) crystalline form of NiBr2 was used for all the measurements [34KET], [54SHC/TOL]. Nevertheless, in the present review, the high-temperature heat capacity equation of Stuve et al. is selected: [C°J™* V.4.1.4.3
K(NiBr2,
cryj-KT'-mor1 = 72.3163 + 0.01462777K- 113805 (77K)~2.
Enthalpy of formation of NiBr2(cr)
Stuve et al. [78STU/FER] measured the heats of heat of solution of samples of NiBr2(cr), (possibly the "Wechselstruktur" form), NiSO4(cr) and H2SO4-6H2O(1) in 4.36 m HCl(aq) and, from these measurements, as described in Appendix A, A f //° (NiBr2, cr, 298.15 K) = — (211.214 ± 1.752) kJ-mor1. Evdokimova and Efimov [89EVD/EFI] measured the heats of solution of Br(l), nickel and NiBr2(cr), (probably the "Wechselstruktur" form) in an aqueous KBr, Br2, HBr mixture. From these measurements, A f //° (NiBr2, cr) = - (211.93 ± 0.90) kJ-mol"1, 2c uncertainties (Appendix A). Several groups, Crut [24CRU], Jellinek and Uloth [26JEL/ULO2] and Shchukarev et al. [54SHC/TOL] have measured the equilibrium constant for the reaction: NiBr2(cr) + H2(g) ^ Ni(cr) + 2HBr(g). at temperatures between 623 K and 928 K. The most extensive, and least scattered set of values is that of [54SHC/TOL]. As discussed in Appendix A, these measurements can be used to calculate Af77° (NiBr2, cr) = - (215.02 + 0.91) kJmor'. Values based on the results of [24CRU] and [54SHC/TOL] are in rough agreement (-214 to - 2 1 7 JK~'mor'). A value of Af//",(NiBr2, cr) based on the sparse enthalpy of solution measurements of Paoletti [65PAO] would depend on use of the selected value of A f //° (Ni2+, 298.15 K) (also see the discussion of [90EFI/FUR] in Section V.2.1.2 and Appendix A). It is unclear whether the enthalpy of formation values based on the measurements of Stuve et al, Shchukarev et al. and Evdokimova and Efimov are for the identical solid. The solid of Evdokimova and Efimov was almost certainly the "Wechselstruktur" form, and that of Shchukarev et al. was possibly the rhombohedral form, at least for
V.4 Group 17 (halogen) compounds and complexes
135
some of their measurements. This seems to be reflected in the less negative value of AfH°y (NiBr2, cr) found by Evdokimova and Efimov. In the absence of other information, the average of the more precise values of Evdokimova and Efimov and of Shchukarev et al. is selected, and the uncertainty is selected to span the uncertainties of the these measurements: A f //° (NiBr2, cr, 298.15 K) = - (213.5 ± 2.4) k J m o l ' . This value results in the following derived Gibbs energy of formation: AfG° (NiBr2, cr, 298.15 K) = - (195.4 ± 2.4) kJ-mol"1. Based on this enthalpy of formation value and the selected value for A f //° (Ni2+, 298.15 K), the enthalpy of solution of NiBr2(cr) in water (at infinite dilution) is - (83.9 ± 2.6) kJ-mor1. This value is less negative than, but in marginal agreement with the - (86.18 ± 0.40) kJ-mol"1 reported by [90EFI/FUR], and is also in agreement, within the uncertainty limits, with the rather poorly documented result of Paoletti [65PAO](-82.1kJ-mor').
V.4.1.5
Hydrated solid, NiBiyxH2O
Although several hydrated compounds have been reported, information on wellcharacterised samples is sparse. The isolation of the hexahydrate (over a desiccant at room temperature) has been reported [66GME], [77BHA/CAR]. Conversion of the hexahydrate to the trihydrate may occur near 30°C. The low-temperature heat capacity of NiBr2-6H2O was reported by Spence et al. [59SPE/FOR] (between 4.0 and 13 K), and by Bhatia et al [77BHA/CAR] (between 1.5 and 30 K). There is a X-transition in the heat capacity measurements. Spence et al. indicate that the transition is at 6.5 K, whereas Bhatia et al. report that the transition is at (8.30 ± 0.02) K, the same temperature as the paramagnetic-antiferromagnetic transition found in their magnetic susceptibility measurements. The two groups report similar values for the entropy contribution of the transition (9.62 JKT'mor 1 [59SPE/FOR], (9.41 ±0.21) JKT'moi 1 [77BHA/CAR]). Because no values could be found for the heat capacity of NiBr2-6H2O for temperatures between 30 and 298.15 K, no value for S° (NiBr2-6H2O, cr) is selected. A value for AfH^ (NiBr2-3H2O) has been reported in several standard sets of tables [36BIC/ROS], [52ROS/WAG], [82WAG/EVA]. This value has been traced to enthalpy of solution measurements of Crut [24CRU], [24CRU2] (from - 79.1 kJ-mol"1 for the enthalpy of hydration of the anhydrous salt [24CRU]). However, the stoichiometry of the hydrated salt is not specified, and there are ambiguities in the reported values (see the discussion in Appendix A). In the present review, no chemical thermodynamic values are selected for the nickel bromide hydrates for 298.15 K.
136
V Discussion of data selection for nickel
V.4.1.6
Solid nickel iodide, NiI2(cr)
V.4.1.6.1
Entropy of NiI2(cr)
Worswick et al. [74WOR/COW] reported low-temperature (adiabatic calorimetry) heat capacity measurements (10 K to 300 K) for this solid, and calculated ^"(Nil^ cr) = 138.7 J-K '-mol '. A ^-transition at 58.9 K was reported. There are only limited confirmatory heat capacity measurements in the literature. Based on magnetic susceptibility measurements, Van Uitert et al. [65UIT/WIL] reported a ^-transition as occurring at approximately 75 K. Billerey et al. [77BIL/TER] found heat-capacity transitions at 59.5 K and near 76 K, and these were later identified by Kuindersma et al. [81KUI/SAN] as a structural transition and a magnetic phase transition, respectively. The raw data from Billerey et al. are not available, but the contributions to the entropy from the two transitions are estimated to be 0.40 and 0.37 J-KT'-mor1 (see Appendix A). Examination of the data from Worswick et al. indicates that their value for S° (Nil2, cr) included a contribution of 0.34 J-K^'-moP1 as an excess entropy in the region 50 K to 85 K. There are no reported heat capacity values at temperatures below 10 K, though the magnetic behaviour of the solid is very complex at low temperatures [81KUI/SAN], [90PAS/TAY], [92FRE/FAL]. The entropy, adjusted for the contribution of the transition at 76 K, is 139.1 J-IC'-mor1, and this value is accepted in the present review. A rather large uncertainty of ± 1.0 J-KT'-mor' is assigned to the selected value: S;(NiI 2 , cr, 298.15 K) = (139.1 ± 1.0) JKT'mor 1 . V.4.1.6.2
Heat capacity of NiI2(cr)
Worswick et al. [74WOR/COW] reported low-temperature (adiabatic calorimetry) heat capacity measurements (10 to 300 K) and differential scanning calorimetry (DSC) results from 300 to 550 K. The equation (V.66) (200 to 550 K) is not in the standard form generally used for solids {e.g., [93KUB/ALC], p. 166, Equation (116)), however, the original DSC data are unavailable. [C°J"Z. (Nil2, cr) / JKT'mor 1 = 65.5 + 0.0515 77K - 0.0000397 (77K)2
(V.66)
Equation (V.67), in a more standard form, was calculated to fit the adiabatic calorimetry results from 201.55 to 297.39 K and values from equation (V.66) at higher temperatures (as discussed in [74WOR/COW]). Equation (V.67) closely reproduces the values from the authors' equation (within 0.5 JKT'mor 1 between 300 and 550 K, and within the experimental uncertainties at lower temperatures). Equation (V.67) is accepted in the present review for the specified temperature range, but this equation should not be used at higher temperatures. NiI2(cr) is known to decompose at temperatures slightly above 700 K. [Cp.mlfooK ( NiI 2, cr) / J-K-'-mor1 = 73.5775 + 0.0164257- 1.057753xl05 (77K)"2 (V.67) The value of C°pm (Nil2, cr, 298.15 K) based on equation (V.67),
V.4 Group 17 (halogen) compounds and complexes
137
C; m (NiI 2 , cr, 298.15 K) = (77.28 ± 1.00) J-KT'-mor1 is selected in the present review with a markedly increased uncertainty (see Appendix A). V.4.1.6.3
Enthalpy of formation of NiI2(cr)
Values for the enthalpy of formation of Nil2(cr) have been reported (or can be calculated) based on enthalpy of solution measurements [65PAO/SAB], [90EFI/EVD] and from high temperature equilibrium measurements [80OPP/TOS]. The work of Oppermann and Toschew [80OPP/TOS] considers many of the difficulties in determining values from the rather complex Ni-I system at higher temperatures, and the problems in doing a proper extrapolation to determine thermodynamic values at 298.15 K. However, as discussed in Appendix A, this process can only be carried out within very large uncertainty bounds, by using the selected C°pm(T) equation for NiI2(cr), and drawing tentative conclusions related to the thermodynamic functions for NiI2(g) and Ni2l4(g). The earlier measurements of Jellinek and Uloth [26JEL/ULO2] suffer from many of the same problems, but are also too sparse to be properly re-interpreted. The value reported by Bartovska et al. [74BAR/CER] is based, in part, on estimated values. A value of A f //° (Nil2, cr) based on the enthalpy of solution measurements of Paoletti et al. [65PAO/SAB] or Efimov and Furkalyuk [90EFI/FUR] would depend on use of the selected value of A f //° (Ni2+, 298.15 K). However, experimental results reported by Efimov et al. [88EFI/EVD], Efimov and Evdokimova [90EFI/EVD] provide a value of A f //°(NiI 2 , cr, 298.15 K) that is independent of the value for A f //° (Ni2+, 298.15 K). From their work (see Appendix A), the value, A f //° (Nil2, cr, 298.15 K) = - (96.42 + 0.84) kJ-mof' is selected in the present review. This value is in agreement with - ( 9 1 ± 8) kJ-mol"1 based on Oppermann and Toschew [80OPP/TOS] (see Appendix A). This value results in the following derived Gibbs energy of formation: AfG° (Nil2, cr, 298.15 K) = - (94.4 ± 0.9) kJ-mol"1. Based on the selected values of A f //°(NiI 2 , cr, 298.15 K) and A f //°(Ni 2+ , 298.15 K), the enthalpy of solution (at infinite dilution) of NiI2(cr) in water is — (71.8 ± 1.2) kJ-mor1, in reasonable agreement with the rather poorly documented result of Paoletti [65PAO] (- 70.5 kJ-moF1) for slightly higher ionic strength. The value of - (73.38 ±0.12) kJmol"1 reported by Efimov and Furkalyuk [90EFI/FUR] is somewhat more negative.
V.4.1.7
Solid nickel iodate compounds
Pracht et al. [97PRA/LAN] have summarised and examined phase relationships in the Ni(IO3)2-H2O system. They concluded that, as reported in earlier papers [73COR], [73NAS/SHI], two tetrahydrates, a dihydrate and a higher hydrate, perhaps a decahy-
138
V Discussion of data selection for nickel
drate, may have regions of metastability or stability near 300 K. Meusser [01MEU] measured the solubilities of several different hydrated nickel iodates, and the most stable form near room temperature was the "P-dihydrate", which is the Ni(IO3)2-2H2O of later workers. The solubility results of Meusser [01MEU] and Cordfunke [73COR] for the dihydrate are quite similar, and extrapolation [73COR] and interpolation [01MEU] to 298.15 K provide values of- (5.17 ± 0.06) and - (5.17 + 0.04) for log10 K°o ((V.68), 298.15 K) (see Appendix A). This agreement is rather surprising as neither worker reported a measurement at 298.15 K, and there are doubts concerning whether equilibrium was attained, especially in the work of Cordfunke [73COR]. Ni(IO3)2-2H2O(cr) ^ Ni2+ + 2IO" + 2H2O(1)
(V.68)
Fedorov et al. [73FED/SHM] measured the solubility of a solid with the reported composition of a trihydrate, Ni(IO3)2-3H2O, in aqueous lithium perchlorate/nitrate mixtures at 298.15 K and ionic strengths from 0.5 to 4.0 M. Their solubility results near 25°C, on extrapolation to / =0, are similar to those from the other two studies. Based on the values obtained from experiments using solutions without nitrate, a value of the solubility product of Ni(IO3)2-3H2O, log10 K°o = - (5.09 ±0.16) is calculated (see Appendix A). It is clear from all the studies that the kinetics of the interconversion of the hydrates is slow at room temperature. The evidence for a true trihydrate is slim, yet might be the result of catalysed precipitation of a more stable form that is difficult to synthesize in other media. It might also reflect insufficient periods of equilibration, or that the solid was not properly handled prior to analysis. There does not seem to be any thermodynamic information (vapour pressure measurements) available on the interconversion of the hydrates. In the present review the solubilities of Fedorov et al. [73FED/SHM] are assumed to relate to the dihydrate rather than the "trihydrate". The three values for the dihydrate at 298.15 K. are then in marginal agreement within the statistical (2a) uncertainties. The average from the three studies is selected: logl0tf,°0(V.68) = -(5.14±0.10), where the uncertainty of 0.10 in logl0 K°o is assigned to reflect uncertainty in the crystalline form. From the solubility product, ArGn°, ((V.68), 298.15 K) = (29.34 ± 0.57) kJ-moP1. Using this selected value and the auxiliary data in Table IV-1, AfG° (Ni(IO3)2-2H2O, cr, 298.15 K) = - (802.1 ± 1.8)kJ-mor'. The effective enthalpy of the dissolution reaction can be calculated from the temperature dependence of the solubility products, provided that the heat capacity of
V.4 Group 17 (halogen) compounds and complexes
139
reaction is small over the experimental temperature range. For a dissolution reaction of an ionic salt, it is unlikely that the latter condition is met; nevertheless, as shown in Appendix A, the values based on the data of Meusser [01MEU] and Cordfunke [7 3 COR] do not appear to change significantly if the temperature range used for calculation of ATHm (V.68) is changed. On this basis, the average of the calculated temperatureaveraged values for the enthalpy of reaction (V.68), 21.4 kJ-mol"' (281 to 323 K) [01MEU] and 21.7 kJ-mol"1 (302 to 323 K) [73COR], is selected for 298.15 K, but with a slightly increased estimated uncertainty: A r //° ((V.68), 298.15 K) = (21.6 + 5.0) kJ-mol"1. Using this value and the auxiliary data in Table IV-1, A f //° (Ni(IO3)2-2H2O, cr, 298.15 K) = - (1087.7 + 5.2) kJ-mol 1 S° (Ni(IO3)2-2H2O, cr, 298.15 K) = (270.1 + 17.4) J'lC'mor 1 are calculated. The heat capacity of Ni(IO3)2-2H2O(cr) has been measured [64CHA/BOE], [72KLA/DOK] between 1 and 30 K, but no higher temperature measurements appear to have been reported. There are two forms of anhydrous nickel iodate, and there is some evidence [73NAS/SHI], [97PRA/LAN] that the yellow (hexagonal) (3 form is more stable than the green a form. The measurements of Meusser for the solubility of an anhydrous nickel iodate, probably the p-Ni(IO3)2 [73NAS/SHI], at four temperatures leads to logl0 K°o ((V.69), Ni(IO3)2, P, 298.15 K) = -(4.43 + 0.02), and an average value of A r // m (V.69) = - (7.3 ± 0.7) kJ-mor1 (303.15 K to 363.15 K). It is clear that at 298.15 K the dihydrate is less soluble than the anhydrous salt. In the absence of other data, the value of the solubility product is selected, but with an increased uncertainty: P-Ni(IO3)2 ^ Ni2+ + 2IO~
(V.69)
ArG^ ((V.69), 298.15 K) = (25.3 ± 0.1) kJ-mol"1. Using this value and the auxiliary data in Table IV-1, AfG° (Ni(IO3)2, P, 298.15 K) = - (323.7 + 1.7) kJ-mor'. The enthalpy of reaction value is somewhat suspect because it is an average value from measurements over a rather large temperature range (not including 298.15 K), and no attempt has been made to include heat capacity values in the data analysis. Nevertheless, the value is selected here with an increased estimated uncertainty. A r //° ((V.69), 298.15 K) = - (7.3 + 4.0) kJ-mor1. Using this value and the auxiliary data in Table IV-1,
140
V Discussion of data selection for nickel
AfH°m (Ni(IO3)2, p, 298.15 K) = - (487.1+4.2) kJ-mol"1 S° (Ni(IO3)2, p, 298.15 K) = (213.5 + 14.1) J-K"1 mol"1 are selected.
V.4.2 Aqueous nickel halide complexes V.4.2.1 Introduction Halide anions, with the exception of fluoride, form rather unstable complexes with Ni(II) in aqueous solution. This is mostly due to the strong hydration of Ni(II). Thus, water can efficiently compete with the essentially electrostatic Ni(II)-halide interaction. Consequently, high and varying excesses of ligand anions over Ni(II) have been used to assess the stability of the complexes formed. Under such conditions a large part of the medium ions are replaced by the complex-forming electrolyte, causing substantial change in the activity coefficients, and thus, the constant medium principle is violated in these measurements. According to Harned's rule, the logarithms of the activity coefficients in a mixture of two electrolytes vary linearly with their mole fractions. This variation, which depends mainly on the hydration of the ions involved, introduces an inherent and electrolyte dependent uncertainty into the determined stability constants of weak complexes [89BJE]. The hypothesis generally made in the earlier volumes of this series, was that in all studies of weak complexation, to a first approximation, the activity coefficients do not change significantly as the weak complex-forming anion substitutes for perchlorate, provided that the total ionic strength is maintained constant. In the present review the same procedure is accepted, emphasising the approximative validity of this hypothesis. Therefore, if considerable medium changes occurred during the experiments, we assign higher uncertainty to the accepted formation constants than was published in the original report, regardless of the quality of the experimental work. Furthermore, as it is almost impossible to distinguish between a medium effect and the formation of higher complexes, for lack of solid evidence, only NiX+ species are accepted in this review, if a substantial medium effect can be assumed. In some cases mixed or 'semi-thermodynamic' formation constants are reported [36JOB], [71PAA/HUM], [72RET/HUM], [88BJE], [90BJE]. In these sets of measurements, the formation of weak complexes was studied at low concentrations of the metal ion and up to the highest possible concentrations of the highly soluble salts (NX) of the complex-forming anions (X"). The 'semi-thermodynamic' formation constants can be written as: K
"
=
[NiXj [mxri_l]-[x]-r±
where y+ is the molar activity coefficient of the complex forming electrolyte (NX) in molar units [88BJE], [90BJE]. Others [71PAA/HUM], [72RET/HUM] used the so called complete equilibrium constant ( K n ) , in which the water is also taken into consideration in the equilibrium process,
141
V. 4 Group 17 (halogen) compounds and complexes [NiXn_,(H2O)7_n]3~" + X- ^
[NiX K (H 2 OVJ 2 -" + H2O(1)
[NiX.(H 2 0)^]-o. " [NiXn_,(H2O)7_J-[X]-r±
K.
Although these constants probably provide a better description of the given system in the case of a high and varying excess of complex-forming anion than do those determined using the constant medium principle, their chemical meaning is not clear [90BEC/NAG]. These 'semi-thermodynamic' formation constants are not equivalent to the value extrapolated to zero ionic strength using the SIT, except for the case when the ratio ^[NjX ,H20, ]/^ [ N i X (H 0) j= 1 over the whole range of the ionic strength studied. Besides the weakness of complex formation, the simultaneous presence of inner- and outer-sphere complexes, especially in the cases of chloride and bromide, further complicates the determination of the correct solution speciation. Using most of the experimental methods the measured formation constant is the sum of the formation constants of two types of complexes, [NiL(H2O)5]inner and [Ni(H2O)6L]0Uter: .exP = [NiL(H2O)s]inner + [Ni(H2O)6L]outcr [Ni(H2O)6] • [L]
=
inncr +
However, the d-d transitions of Ni(II) are certainly more influenced by innerthan outer-sphere complexation. In the case of weak complexes, inner-sphere interaction can be expected to dominate only at markedly reduced water activity, i.e., in case of concentrated solutions. Consequently, a visible spectrophotometric study, using high concentrations of the complex-forming anion, mostly provides information on innersphere complexes, and it is therefore not strictly comparable with e.g., potentiometric results obtained at considerably lower ligand concentrations.
V.4.2.2
Solution structural studies
A number of NMR [68LIN/APR], [79WEI/HER], [79WEI/MUL], [83GOW/DOD], EXAFS [80LAG/FON], [83LIC/PAS], [99HOF/DAR], solution X-ray [80CAM/LIC], [82WAK/ICH], [82MAG/PAS], [85MAG/MOR], [99WAI/KOU], neutron diffraction [83NEI/END] and neutron scattering [2002CHI/SIM] studies are available concerning the structure of concentrated NiCl2 and NiBr2 solutions. In presence of 2 M NiCl2 and 4 - 6 M LiCl/HCl, solution X-ray [82MAG/PAS] and EXAFS [83LIC/PAS] studies clearly demonstrated the formation of inner-sphere chloro complexes, although the average number of Ni 2+ -Cl~ contacts are relatively low (nC\- = 0.7-1.4). In stoichiometric solutions the situation is less clear. Some authors [80LAG/FON], [80CAM/LIC], [83NEI/END], [99HOF/DAR] have detected the formation of ion pairs only, but the majority of reports agree with the presence of inner-sphere complexes at higher concentrations. The first solid evidence in favour of the existence of inner-sphere [Ni(H2O)5Cl]+ complex in stoichiometric NiCl2 solutions was the detection of slow chloride exchange by 35C1 NMR at - 5 and + 1°C [83GOW/DOD]. This observation
142
V Discussion of data selection for nickel
demonstrates the presence of inner-sphere complexes, since outer-sphere interaction is unlikely to produce slow ligand-exchange. Solution X-ray diffraction studies of concentrated NiBr2 solutions [82WAK/ICH], [85MAG/MOR] confirmed the above observations, since the very large backscattering amplitude of a Br scatterer results in better identification of Ni 2+ -Br~ contacts than the Ni 2 + -CP ones. Although in [99HOF/DAR] the authors report the formation of ion pairs at 298 K, the temperature dependence of XAFS spectra clearly indicate the formation of tetrahedral [NiBr4]2~ complex above 698 K. The average number of Ni 2+ -X~ contacts reported in most of the publications mentioned above allowed the calculation of the stability constants of inner-sphere complexes. Although the applied ionic strengths are generally very high, some of these constants are reported in Sections V.4.2.4 and V.4.2.5.
V.4.2.3
Aqueous Ni(II) - fluoro complexes
The reported stability constants for Reaction (V.70) are listed in Table V-8. Ni2+ + F ^ NiF+
(V.70)
Table V-8: Experimental stability constants (logarithmic values) of the species NiF+. Method
Medium
K°c)
log,,,*:, reported
log..*,
w
Reference
accepted
ise-F
0.05 M (CH,)4NC1O4
25
(1.32 ±0.05)
(1.32 ±0.30)
[83SOL/BON]
ise-F
0.1 MNaC10 4
25
(1.02 ±0.12)
(1.02 ±0.20)
[81KUL/BLO]
kin
0.1 MNaClO 4
25
(1.1 ±0.1)
(1.10 ±0.20)
[69FUN/TAN]
ise-F
0.25 M NaClCX,
25
(0.91 ±0.10)
(0.91 ±0.30)
Mironov et a/.(b)
ise-F
0.5 M NaClO4
25
(0.76 ± 0.08)
(0.75 ± 0.20)
Mironov et al.(b)
ise-F
1.0MNaClO 4
25
(0.68 ± 0.09)
(0.66 ± 0.20)
Mironov et al.(b)
ise-F
1.0 M NaClO,
25
(0.34 ± 0.04)
(0.32 ± 0.30)
[72BON/HEF]
pol
1.0MNa(ClO 4 , F)
25
(0.48 ± 0.05)
(0.46 ± 0.30)
[71 BON]
qh
1.0MNaClO 4
20
(0.66 ± 0.05)
(0.64 ±0.10)
[56AHR/ROS]
(c>
25
(0.66 ± 0.10)
ise-F
2.0 M NaClO4
25
(0.65 ± 0.06)
(0.61 ±0.20)
[81KUL/BLO]
ise-F
3.0 M NaClO4
25
(0.76 ± 0.06)
(0.69 ± 0.20)
Mironov et al.(b)
(a) Corrected to molal scale. The accepted values reported in Appendix A are expressed on the molar or molal scales, depending on which units were used originally by the authors. (b) Average value from the studies reported by Mironov's [81KUL/BLO], [76KUL/BLO], (c) Corrected to 298.15 K using the selected value of A r //° .
group
[83AVR/BLO],
The majority of data in Table V-8 were obtained in NaClO4 solutions using a fluoride selective electrode, but some pH-metric [56AHR/ROS], kinetic [69FUN/TAN] and polarographic [71 BON] results are included, too. The SIT analysis of the experimental stability constants is depicted in Figure V-20.
V.4 Group 17 (halogen) compounds and complexes
143
Figure V-20: Extrapolation to /,„ = 0 of the experimental data for Reaction (V.70) in NaClO4 media. Experimental data from: • : [56AHR/ROS], A: [69FUN/TAN], O: [71BON], • : [72BON/HEF], • : [81KUL/BL0], • : [83SOL/BON], A: [76KUL/BLO], [81KUL/BLO], [83AVR/BLO].
The value given by Arhland and Rosengren [56AHR/ROS] determined at 293.15 K has been corrected to 298.15 K using the enthalpy of reaction selected below. The constant determined in 0.05 M tetraethylammonium perchlorate [83SOL/BON] was also considered (but with a somewhat higher uncertainty). Although there are notable differences between the values published by different authors, the data in Figure V-20 show acceptable consistency. We should note, however, that most of these data were reported by the Mironov's group [83AVR/BLO], [81KUL/BLO], [76KUL/BLO]. Only two of the five constants reported by other laboratories are consistent with the data of Mironov et al. The weighted linear regression using 11 data points yielded the selected value of: !og10 P° ((V.70), 298.15 K) = (1.43 ± 0.08). This value agrees well with that published in [81KUL/BLO] calculated by means of the Vasil'ev equation. The resulting Ae(V.70) value is - (0.049 ± 0.060) kg-mol"1. Using the selected values for e(Ni2+, C1O~) and E(Na+, F"), Ae(V.7O) leads to a value of e(NiF+, CIO") = (0.34 + 0.08) kg-mol"1.
144
V Discussion of data selection for nickel
From the above formation constant ArG°((V.7O), 298.15 K) = -(8.2 ±0.5) kJ-moP can be calculated. The Gibbs energy of formation of NiF+ is derived using the selected values for Ni2+ and F~. The selected value is: 1
AfG°(NiF+, 298.15 K) = -(335.5 ± l.l)kJ-mor'. The formation of higher complexes (NiFn2"", n > 1) is not reported in the literature, not even in presence of more than a thousand-fold excess of fluoride over Ni(II) [71 BON]. Reaction enthalpies for the formation of the NiF+ complex are collected in Table V-9. All data in Table V-9 except those of [74ARU] were determined in Mironov's laboratory. In order to determine A r //° ((V.70), 298.15 K), the SIT equation adapted to enthalpy format [97GRE/PLY2] has been used (cf. Figure V-21). The values at / = 0.1 and 0.25 M [81KUL/BLO] were not considered, as the amount of NiF+ formed is probably too small to yield reliable enthalpies. The value reported by Aruga was also neglected for the SIT, due to the different ionic medium used. The analysis of the remaining four points resulted in the selected standard reaction enthalpy: A r //° (V.70) = (9.5 ± 3.0) kJ-mol"1 and AsL(V.7O) = - (5 ± 35) X 10~5 kg-IC'-mol"1. The former value is fundamentally different from that reported in [81KUL/BLO] (A r //°((V.7O), 298.15 K) = -(2.9 + 2.5) kJ-mol"1; the method of extrapolation to / = 0 is not reported). It agrees well, however, with the general observation that the stabilities of most aqueous monofluoride complexes are governed predominantly by electrostatic interactions and thus are entropy controlled [74HEF], [90GRZ]. Table V-9: Experimental enthalpy values for Reaction (V.70). t/°C
Method
Medium
cal
0.1 MNaC10 4
25
0
cal
0.25 M NaClO4
25
(3.3 ±1.6)
cal
0.5 M NaC104
25
(6.7 ± 0.8)
1
A r // m (kj-mol ') reported
A r // m (kjmol""') Reference accepted [81KUL/BLO]
(b)
[81KUL/BLO] (6.7 ± 2.0) (c)
[81KUL/BLO]
cal
0.5 M self "
25
(8.1 ±0.2)
cal
1.0MNaClO 4
25
(7.9 + 1.2)
(7.9 + 2.0)
cal
2.0 M NaC104
25
(7.9 ±1.6)
(7.9 ± 2.0)
[81KUL/BLO]
cal
3.0MNaClO 4
25
(6.5 ± 2.0)
(6.5 ± 2.0)
Mironov et. al.(i)
(4.4 ± 2.0)
[74ARU] [81KUL/BLO]
(a) The solution of 0.17 M Ni(NO3)2 and 0.5 M NaF were mixed in different ratios. (b) Calculated from logl0/?, = 0.31, which was derived from the datum in [72BON/HEF], (c) Recalculated data using the selected value for log10 p° ((V.70), 298.15 K) and As (V.70). (d) Average value from the studies reported by Mironov's group [76KUL/BLO] and [81KUL/BLO],
V.4 Group 17 (halogen) compounds and complexes
145
From the above A r //° value, the following selected standard enthalpy of formation can be derived for NiF+: A f //° (NiF+, 298.15 K) = - (380.9 + 3.2) kJ-mol"1. The above selections result in: 5 ; (NiF+, 298.15 K) = - (86.4 ± 10.3) J-K"'-mor'.
Figure V-21: Extrapolation to Im = 0 of the experimental enthalpies for Reaction (V.70) in NaC104 media from [76KUL/BLO] and [81KUL/BLO] (squares) and [74ARU] (triangle). The data marked with open square were not considered in the SIT treatment, since the amount of NiF+ formed is too small. According to [97GRE/PLY2],
WM-
3A(z2)
ALJTm
where A(z2) is the stoichiometric sum of the charge numbers squared of the reactants (in the present case - 4), AL is the Debye-Hiickel parameter for the enthalpy, at 298.15 K and 1 b a r ^ L = 1.986 kJ-kg'/2-mor3/2.
146
V Discussion of data selection for nickel
V.4.2.4 Aqueous Ni(II) - chloro complexes The formation of Ni(II)-chloro complexes has been the subject of numerous investigations using several experimental methods (cf. Table V-10). Since in most of these studies the composition of the ionic medium was varied over a wide range, and no attention was paid to the medium effect, the literature reports on the complex formation are rather divergent concerning both the reported formation constants and stoichiometry of the formed complexes. Beside the formation of NiCl+ and NiC^Caq) complexes, the presence of other species is also proposed in aqueous solution at room temperature: e.g., Ni2Cl3+ [60LIS/ROS], NiCl3 [64GRI/LIB] or NiClf [68AND/KHA], [71KLY/GLE], [85SAP/CHI]. This reflects the fact that conventional methods are not always applicable to determine the stability of weak metal complexes if a high concentration of the ligand is needed to achieve a sufficient degree of complex formation. The majority of the experiments have shown that no anionic complexes are formed even at very high chloride concentrations. On the other hand, the formation of Ni2Cl3+ reported in [60LIS/ROS] was reconsidered and finally rejected in a subsequent paper of the authors [66KEN/LIS]. Therefore in this review only the formation constants for NiCl+ and NiCl2(aq) complexes are considered. The formation constants reported for Reaction (V.71), Ni2+ + C r ^ N i C l +
(V.71)
are listed in Table V-10. For reasons mentioned in Appendix A we do not consider the data reported in [57KIV/LUO], [58TRE], [62TRI/CAL], [65MOR/REE], [74GRA/WIL] and [76MUR/KUR]. Most of the accepted data are, however, also subject to substantial experimental errors, due to the medium effect, and in such cases we assigned significantly higher uncertainty to the selected constants than reported in the original publications. The data in [75LIB/TIA] are free of significant medium effects, and this is the only data set where the systematic errors can be assumed identical for each point. Therefore, these data were used to determine the ion interaction coefficient between NiCl+ and CIO4, in spite of the fact that the applied ionic strength (/,„ = 3 - 9 m) is well above of the recommended range for the SIT analysis. The plot (log10 /?, + 4£>) versus lm is depicted in Figure V-22. The weighted linear regression of data in Figure V-22 results in the values of log10/?° (V.71) = - (0.37 + 0.27) and Ae(V.71) = - (0.073 ± 0.040) kg-mol'. Using the selected values for s(Ni2+, CIO;), e(Ni2+,Cl~), (in the absence of data to allow calculation of e(Ni2+,Cl") while also considering association; see, for example, Section V.6.1.2), As(V.71) leads to a value of: e(NiCl+, C1O4) = (0.47 ± 0.06) kg-mor1. There is some uncertainty concerning this calculation. In [75LIB/TIA] Ni(ClO4)2 was used as a constant ionic medium (with a chloride concentration of- 0.01 m), therefore, Ae can be calculated as Ae = e(NiCl+, CIO;) - s(Ni2+,CO - s(Ni2+, CIO;).
147
V.4 Group 17 (halogen) compounds and complexes
In Appendix B.2 of previous TDB reviews it is mentioned that since the formation of chloride complexes is taken into account, the value of e(Ni2+, C\) as listed, for example, in Appendix B of [92GRE/FUG], [2001LEM/FUG] should be replaced by e(NiCl+, ClO^) in all calculations when chloride is part of the ionic medium. This listed value of e(Ni 2+ ,Cr) was obtained [80CIA] by neglecting association of nickel and chloride. Following this process, and using e(Ni2+, C1O4) = (0.37 ± 0.03) kg-mol"' (see Section V.4.3), from the above determined Ae= - (0.073 ± 0.04), e(NiCl + ,C10;) = (0.667 ± 0.06) kg-mor1 can be calculated. This value of e(NiCi+, C1O~) is too high, taking into account the relatively accurate value for £(NiF+, CIO4). In this review, e(NiCl+, C1O4) = (0.47 ± 0.06) kg moP1 is used to determine Ae in other media to calculate log10 /3° values from the rest of the data in Table V-10. According to Appendix B.2, the same treatment should be applied, e.g., for NO", i.e., s(M x+ ,NO~) should be replaced by E(M X+ , CIO4) if nitrate is part of the medium, but this recommendation is often disregarded. Using the remaining experimental values for NaClO4 media (from Table V-10) and s(Ni2+, CIO;) = 0.37 to determine e(NiCl+, CIO;), from the equation As = £(NiCl+, CIO;) - e(Na+, Cl") - s(Ni2+, CIO;), e(NiCl + ,C10;) = 0.51 kg-mol"1 can be obtained. This may support the value calculated using s(Ni2+, Cl~) from the data in [75LIB/TIA], however, this is not particularly convincing because most of the remaining data in the other papers have relatively low accuracy, due to substantial medium effects.
Table V-10: Experimental formation constants (logarithmic values) for the NiCl+ species. Method Ionic medium
t(°C)
log,,/?,
(a)
log.oA
lb)
iog,oA0(c)
Reference
pol
2 M Na(ClO4, Cl)
25
-(0.25 ±0.15)
[57KTV/LUO]
cix
1.5M(Ca,Ni)
?
-(0.66 + 0.01)
[58TRE]
co-
(NO3, Cl) 0.048 m KCIO4
-0.15
0.62
cry
0.258 m KCIO3
-0.8
0.23
ise-Cl 2 M (Na, Ni)ClO4
12 25 40 _2
-0.17 -0.15
cry
1.3MK(NO3/C1)
sp
1.5MNa(ClO 4 , Cl)
?
-(0.85 ±0.25)
sp
(0.38 ±0.08)
[59KEN] (0.23 + 0.30) - (0.25 ± 0.20)
(0.84 ± 0.30)
-(0.21 +0.20)"" -(0.18±0.20) ( d )
(0.88 + 0.30) (0.91 ±0.30)
(0.34 ± 0.30)ld)
(1.14 ±0.40)
[60LIS/ROS]
[62FAU/CRE] [62TRI/CAL]
5.7 M H(C1, CIO4)
25
- (0.5 ± 0.2)
cix
3 M H(C1O4, Cl)
25
- (0.64 ± 0.06) -(0.51 ±0.50) (d)
cix
0.69 H(C1O4, Cl)
20
(0.23 ± 0.03)
(0.79 ± 0.40) [59KEN]
(4) = 0.202 kg-mol" , estimated from e(Ni2+,C10^) = 0.375 kg-mol"1 and s(Na+, CV) using Equation (B.22) (see Appendix B). Although Iuliano's value is considerably lower, we recommend the value derived in the present review for e(NiCl+, CIO4), since it is based on experimental evidence. All accepted constants for NaClC>4 media are within the ionic strength range /,„ = 0 - 6 m. The SIT representation of these values is depicted in Figure V-23. Using the ion interaction coefficient e(Ni2+, C1O4) together with the above determined value of e(NiCl+, CIO;), Ae = (0.07 ±0.10) kg-mol"1 can be calculated for Reaction (V.71) in NaClO4 medium. From the plot of experimental data in Figure V-23 a very similar value, Ae = (0.11 + 0.06) kg-mol"1 can be obtained. However, this value must be treated with caution, due to the already mentioned medium change, which affects most of the data points in Figure V-23. Therefore, Ae((V.71), NaClO4) = (0.07 + 0.10) kgmor 1 was used to calculate the logl0 P° values for NaClO4 media given in Table V-10. The SIT extrapolation resulted in somewhat lower log,0 /3° values in LiClO4 (Ae((V.71),
150
V Discussion of data selection for nickel
LiClO4) = (0.0 + 0.1) kg-mol"1) and HC1O4 (Ae((V.71), HC1O4) = -(0.04 ±0.10) kg-mol"1) media.
Figure V-23: Extrapolation to / = 0 of the accepted experimental data for Reaction (V.71) in NaClO4 media (see text).
These extrapolated values are considerably higher than the value determined from the data in [75LIB/TIA]. This is probably a result of the disregarded medium changes. Obviously, the attribution of the global effect (complex formation and changes in activity coefficients) to complex formation alone results in higher constants. The well documented experimental results in [70HAL/VAN] offered the possibility to use the SIT to assess the medium effect caused by the replacement of the original background electrolyte (see Appendix A). In this work the background medium was changed gradually from 4 M NaClO4 to 2 M NaClO4/2 M NaCl. As explained in the discussion on [70HAL/VAN], the modified SIT treatment, in which the changes in ionic medium were taken into account, resulted in a value approximately half a unit lower in the logl0 J3° value (0.20 ± 0.40) than obtained by neglecting this effect. The medium changes, caused by the replacement of CIO" by CF ions, also were accounted for by SIT in [89IUL/POR]. Due to the different £(NiCl+, C1O4) value
V.4 Group 17 (halogen) compounds and complexes
151
used (see Appendix A), however, the authors reported a considerably lower log10 /3° than was recalculated in this review from their data. Bjerrum [88BJE] reported a 'semi-thermodynamic' formation constant (log10 ff° ), which is much lower than any other log10 (1° value in Table V-10. Due to the unknown ratio of yNj2t / / N j c r (see Appendix A), this constant is not compatible with the assumptions of the SIT model. Since Bjerrum applied a spectrophotometric method using the d-d transitions of Ni(II) and very high chloride concentrations, his constant probably refers to the formation of the inner-sphere NiCl(H2O)j complex (log10 fi°Mna ). Libus and Tialowska estimated a similar value for log,0/?,o'inner (-2.0, [75LIB/TIA]). Bjerrum's value is also within the uncertainty range of the 'semithermodynamic' formation constant derived from the data reported in [79WEI/HER] (log10 fi° = — (1.03 ± 1.00), see Appendix A). However, the NMR relaxation measurements in [79WEI/HER] also demonstrate the incompatibility of log10 ji° as defined by Bjerrum (see [88BJE] and [89BJE]) and the log,0/?° values extracted by means of the SIT analysis. Using the recommended ion interaction coefficients to calculate As(V.71) for KC1 medium, log I0 #° ((V.71), 298.15 K) = (0.16 ± 1.00) is derived from the experimental data in [79WEI/HER], which is considerably higher than Bjerrum's logi0 P°' value. Due to the medium effect, most of the constants listed in Table V-10 represent only the upper limits of the true values. Therefore, log10/?,° ((V.71), 298.15 K) is obtained as the average of the following two data: - (0.37 ± 0.27) (determined from the data in [75LIB/TIA]) and (0.52 + 0.38) (average of the remaining logI0/?° values reported in Table V-10 for 298 K in (H/Li/Na)ClO4 and KC1 media). This treatment resulted in the selected value: logio P° ((V.71), 298.15 K) = (0.08 ± 0.60). From this value, ArG° ((V.71), 298.15 K) = - (0.5 ± 3.4) kJ-mof1 can be derived. The Gibbs energy of formation of NiCl+ is calculated using the selected values for Ni2+ and Cl". AfG° (NiCf, 298.15 K) = - (177.4 ± 3.5) kJ-mol"1. Several authors reported equilibrium constants for Reaction (V.72), which are collected in Table V-ll: NiCf + Cl" ^ NiCl2(aq).
(V.72)
For reasons mentioned in Appendix A, and since in cases in which there are considerable changes in the ionic medium it is impossible to distinguish between a medium effect and the formation of higher complexes, we do not consider the data reported in [57KIV/LUO], [65MOR/REE] and [70HAL/VAN]. 'Semi-thermodynamic' constants are reported for Reaction (V.72) in [71PAA/HUM] and [88BJE] (see Appendix A), based on a spectrophotometric method using the d-d transitions of Ni(II) and very high
152
V Discussion of data selection for nickel
chloride concentrations ([Cl ] > 9 M). Therefore, these constants most probably refer to the formation of the inner-sphere complex NiCl2(H2O)4(aq). Iuliano et al. used solubility measurements to determine the equilibrium constant for Reaction (V.72). They reported a higher value for A^ than for K\, which is rather unlikely considering the low affinity of Ni(II) for chloride. Therefore, this review does not find it justified to select a recommended value for K2 (V.72).
Table V-ll: Experimental equilibrium constants (logarithmic values) for Reaction (V.72). log 10 K,' a)
t/°C
Method
Ionic medium
pol
2 M Na(C104, Cl)
25
(0.20 ±0.15)
cix
0.69 H(C1O4, Cl)
20
-(0.27 ±0.05)
[65MOR/REE]
sol(b)
4 M Na(ClO4, Cl)
25
-(0.89 + 0.01)
[70HAL/VAN]
sp
0 corr(c)
25
- ( 2 . 8 + 0.3)
[71PAA/HUM]
sp
0 corr(d)
25
- (2.3 ± 0.6)
[88BJE]
6 M Na(ClO4, Cl)
25
- (0.5 + 0.2)
[89IUL/POR]
sol
0 corr(c)
Reference [57KJV/LUO]
-0.3
(a) Reported values. (b) Corrected to / = 0 by means of the SIT, using the selected ion interaction coefficients. (c) 'Semi-thermodynamic' constant, in which the activity of chloride (using y± of the complex forming electrolyte) and water were taken into account. (d) 'Semi-thermodynamic' constant, in which the activity of chloride (using y± of the complex forming electrolyte) was taken into account. (e) Corrected to / = 0 by means of the SIT, using estimated ion interaction coefficients.
Higher chloro complexes NiClj; " with n = 3 - 4 readily form in several nonaqueous solutions [59GIL/NYH]. As was mentioned previously, such complexes are unlikely to be present in detectable concentrations in aqueous solution at room temperature. The log10/?4 in aqueous solution was estimated to be - 14 in [76BIA/CAR] (see Appendix A). Accepting this value, the calculated concentration of NiCl^ is negligible, even in highly concentrated chloride solutions. In aqueous solutions such complexes can be only expected at elevated temperature in melts [66ANG/GRU] or under hydrothermal conditions [96UCH/GOR]. The only exception is reported in [69GRI/SCA]. Using a concentrated aqueous solution of (CH3)4NC1 ( > 5 M) as background electrolyte, the formation of tetrahedral NiCl^ complex was detected by UVVIS spectrophotometry at as low a temperature as 50°C. This exceptional behaviour was attributed to the specific effect of tetramethylammonium cation, which reduces the hydration of chloride anion enhancing its coordination ability [69GRI/SCA].
153
V.4 Group 17 (halogen) compounds and complexes
Only a few studies have reported reaction enthalpies for the formation of NiCl+ species (Table V-12), and such data are not available for NiC^aq). Only the experimental results reported in [66KEN/LIS] for 298 K could be re-evaluated using log,o/?i°((V.71), 298.15 K) and As((V.71), NaC104), therefore no value is selected in the present review. Table V-12: Reaction enthalpies reported for Reaction (V.71). Method Medium cal
t (°C)
2 M (Na, Ni)ClO4
A r // m (kj-mol ') (a) A,.//m (kj-mol ')(b> Reference
25
(2.1±0.2) (c) (c)
cal
2 M (Na, Ni)ClO4
40
(1.8±0.2)
cal
3 M LKCICVCI)
25
(9.6 ± 0.4)
(8.2 + 2.5)(d)
[66KEN/LIS] [66KEN/LIS] [74BLO/RAZ]
(a) Reported value. (b) Accepted value. (c) Calculated using log, 0 # = - 0 . 1 7 (25°C) and log,,,/?, = - 0 . 1 5 (40°C) from [60LIS/ROS]. (d) Re-evaluated value, see Appendix A. (e) Calculated using log l 0 # = - 0 . 5 7 from [70M1R/MAK].
V.4.2.5 Aqueous Ni(II) - bronio complexes Only a few studies are available concerning the complex formation between Ni(II) and bromide ions. Most of the studies reported the formation of a monobromo complex (NiBr+), but at very high concentrations of the bromide ion the formation of bis[36JOB], [72RET/HUM], [90BJE] and tetrakis- bromo complexes [36JOB], [68AND/KHA2] are also reported. The graphical data presented in [36JOB] were later re-evaluated in [72RET/HUM], and were found to be consistent with the formation of NiBr+ and NiBr2(aq) complexes (see comments on [36JOB]). The qualitative work in [68AND/KHA2] using spectrophotometry may well suggest formation of 5 - 10% of the tetrakis - bromo complex, but only in extremely concentrated (18 M) HBr solutions (see Appendix A). Nevertheless, the assessment of values for formation of negatively charged complexes in aqueous solutions is not justified. The available quantitative information on the formation of NiBr+, according to Reaction (V.73), is collected in Table V-13. Ni2+ + Br" ^ NiBr+
(V.73)
There are large differences between the values listed in Table V-13. This is the consequence of the inherent difficulties that occur when formation constants of weak complexes are determined (see also Sections V.4.2.1 and V.4.2.4). The values reported in [63NET/DRO], [70MIR/MAK] and [73HUT/HIG] are of low precision, due to a considerable medium effect. The 'semi-thermodynamic' formation constant reported in [90BJE] probably gives an exact description of complex formation in 3.5 - 11 M LiBr,
V Discussion of data selection for nickel
154
but this constant is not compatible with the assumptions of the SIT model (see also Sections V.4.2.1 and V.4.2.4 and [89BJE]). The most reliable data set is given in [78LIB/KOW]. The SIT representation of the data reported in [78LIB/KOW] is depicted in Figure V-24. Although the ionic strengths in the experiments of [78LIB/KOW] are beyond the optimal range for SIT, the experiments are relatively free of interfering medium effects. Therefore, the estimation of e(NiBr+, C1O4) has been based on this data set. This probably leads to a more accurate value than reliance on the other experimental values in Table V-13, or on Equation (B.22). From the plot in Figure V-24, Ae(V.73) = - (0.046 + 0.09) kg-mol ' and logl0 fi° = - (0.5 ± 0.6) are obtained. Using the recommended values for s(Ni2+, B r ) (in the absence of data to allow calculation of s(Ni2+, Br~), while also considering association; see Section V.4.2.5.1) and e(Ni 2+ ,C10-), s(NiBr+, C1O4) = (0.59 ±0.10) kg-moF1 can be derived. Using this unexpectedly large value, it is possible to estimate the Ae values for all background electrolytes listed in Table V-13 (AE((V.73), NaC104) = (0.19 + 0.11) kg-mol"1, As((V.73), HC1O4) = (0.08 + 0.11) kgmol"', Ae((V.73), LiClO4) = (0.07 + 0.11) kg-mol"1, Ae((V.73), KC1) = (0.22 ± 0.11) kg-moP1). These coefficients were used to calculate the log10 fi° values given in Table V-13.
Table V-13: Experimental formation constants (logarithmic values) of the NiBr+ complex. Method Ionic medium
trc
2 M (Na, Ni)ClO4
25 «»
l°g, 0 A ( b > -(0.12 ±0.02) -(0.16 + 0.40)
log,.^(c) (1.20 + 0.50)
Reference
ise-Br sp
5.7 M H(Br, C1O4)
25
- (0.30 + 0.22) - (0.42 ± 0.50)
(1.28 ±0.60)
[63NET/DRO]
sp
3 M Li(ClO4, Br)
25
- (0.82 ± 0.03) - (0.89 ± 0.50)
(0.35 + 0.60)
[70MIR/MAK]
kin
1 M Na(ClO4, Br)
25
- (0.09 ± 0.04) -(0.11 +0.50)
(0.89 + 0.60)
[73HUT/HIG]
1.5mNi(ClO4)2
25
ise-Br
iog, 0 A
(a)
[61LIS/WIL]
-1.3
-(1.30 ±0.30) - (0.50 ± 0.60) [78LIB/KOW]
2 m Ni(C104)2
-1.3
-(1.30 ±0.30)
2.5 m Ni(C104)2
-1.3
-(1.30 + 0.30)
3 m Ni(ClO4)2
-1.16
-(1.16 + 0.30)
nmr
4mKBr
25
sp
0 corr'0'
25
-(1.10 ±0.60) - (2.37 + 0.08)
(0.80 + 0.60)
[79WEI/HER] [90BJE]
(a) Reported values. (b) Accepted values corrected to the molal scale. The accepted values reported in Appendix A are expressed on the molar or molal scales, depending on which units were used originally by the authors. (c) Corrected to / = 0 by means of the SIT, using the recommended ion interaction coefficients. (d) Data for several temperatures, see Appendix A. (e) 'Semi-thermodynamic' constant, in which the activity of bromide (using y± of the complex forming electrolyte) was taken into account.
V.4 Group 17 (halogen) compounds and complexes
155
As already mentioned in previous sections, experiments in which there was a considerable medium effect lead to formation constants that represent only upper limits for the true values. Therefore, the weighted average of the following two values is calculated to assess the most probable formation constant at infinite dilution: - 0.5 (log10 /3° calculated from the data in [78LIB/KOW], weight = 2) and 0.90 (average of the remaining log,0/?° values in Table V-13, weight = 1). This treatment yielded l°gio P\ (( v - 73 )> 298.15 K) = - (0.03 + 1.30). The uncertainty range was assigned to cover all the reported constants in Table V-13. Taking into account this large uncertainty, the above value can not be recommended, but can be used as the most probable value, until more precise data are published for Reaction (V.73).
Figure V-24: Extrapolation to / = 0 of the log,0 /?, values reported in [78LIB/KOW] for Reaction (V.73) in Ni(II) perchlorate media.
The formation of the dibromo complex NiBr2(aq) is dominant only at very high bromide concentrations (8 - 12 M LiBr or HBr). Therefore, only 'semi-thermodynamic' stability constants (log l0 K\ ) are available for this species. The reported values are as
156
V Discussion of data selection for nickel
follows: log,0A:2o1 = -(2.97 + 0.13)[90BJE], log10 K°' = - (4.25 + 0.06) [72RET/HUM] and logl0 K°' = - 3.7 (calculated from the data in [36JOB] by [72RET/HUM]). These data are not directly comparable since in the calculation of the last two values the activity of water was also taken into account. Thus these values refer to the following equilibrium: NiBr(H2O)5+ + Br" ^ NiBr2(H2O)4(aq) + H2O(1). The higher value reported in [90BJE] is reasonable when the very low water activity in concentrated LiBr or HBr solutions is taken into account. Two quantitative results are available for the enthalpy of Reaction (V.73) (Table V-14). The lack of experimental details and some unidentified experimental errors (see Appendix A), however, do not allow selection of accepted values for A r // m (V.73). Table V-14: Reaction enthalpies reported for Reaction (V.73). Method ise-Br cal
Medium 2 M (Na, Ni)ClO4 3 M Li(C104/Br)
/ /°C
A,.//m / kJ-mol ' l a ) (b>
25 25
(8 + 5)
(11.7 ± 0.4)
(c)
Reference [61LIS/WIL] [74BLO/RAZ]
(a) Reported value. (b) Calculated using the reported temperature dependence of log,,, /?, (V.73), see Appendix A. (c) Calculated using log l 0 # = - 0 . 8 2 from [70MIR/MAK].
V.4.2.5.1
Determination of the Ni2+ - Br~ ion interaction coefficient
Because the value of log,0 f}° ((V.73), 298.15 K) given above is low and rather uncertain, the description of the Ni(II)-bromo system in terms of ion-ion interactions may be in many cases more straightforward. Therefore, the ion interaction coefficient e(Ni2+,Br") has also been derived in this review from the osmotic and mean activity coefficients of NiBr2 solutions, using Equations (V.74) and (V.75), see Section V.4.3. The only relevant experimental data are reported in [78LIB/MAC]. Goldberg et al. reported recommended values for activity and osmotic coefficients of NiBr2 solutions [79GOL/NUT], based on the data in [78LIB/MAC]. In [79GOL/NUT] a corrected data set for the experimentally determined osmotic coefficients was given. Reference was made to a personal communication from Z. Libus that provided a corrected version of the erroneous tables in [78LIB/MAC]. Therefore, only the activity coefficients were taken from [78LIB/MAC].
V.4 Group 17 (halogen) compounds and complexes
157
Plots of Iog107± and ^ as a function of the molality, m, of NiBr2 (m = IJ2>) are depicted in Figure V-25 and Figure V-26, respectively. The experimental data can be reasonably described in the concentration range of 0 - 3 m (lm = 0 - 9 m) using e(Ni2+,Br ) = (0.268 ± 0.020) kgmol"1 (see the solid lines in Figure V-25 and Figure V26).
V.4.2.6 V.4.2.6.1
Aqueous Ni(II) - iodine complexes Aqueous Ni(II) - iodo complexes
No quantitative data are available for the formation of Nil^~' complexes. The formation of Nil+ and NiI2(aq) was postulated in [74KUT/GAL], to explain the polarographic prewaves of Ni(II) reduction in 5 M Ca(NO3)2 solutions containing 0.2 - 1 M KI. The spectrophotometric data reported in [68AND/KHA2] indicate formation of a considerable amount of Nil*" in 9 - 11 M HI solutions (see Appendix A).
Figure V-25: Plot of logl0 y± versus molality of aqueous NiBr2 solutions at 298.15 K. Solid line: SIT, e(Ni2+, Br~) = 0.268; D: Experimental data from [78LIB/MAC].
158
V Discussion of data selection for nickel
Figure V-26: Plot of osmotic coefficient versus molality of aqueous NiBr2 solutions at 298.15 K. Solid line: SIT, s(Ni2+, Br") = 0.268; D: Experimental data from [78LIB/MAC], [79GOL/NUT].
V.4.3 Determination of the Ni2+ - C1O4 ion interaction coefficient The ion interaction coefficients can be derived from the ionic strength dependence of the osmotic and mean activity coefficients of the corresponding electrolyte solutions. Two papers report relevant experimental data for Ni(ClO,4)2 solutions. Libus and Sadowska used an isopiestic method to determine the osmotic coefficients in 0.1 - 3.5 m Ni(C104)2 solutions [69LIB/SAD], and the mean activity coefficients were calculated from the Gibbs-Duhem equation. Malatesta et al. determined the activity coefficients in 0.0001 - 0.9 m Ni(C104)2 solutions from the emf of liquid-membrane cells [99MAL/CAR]. Goldberg et al. in [79GOL/NUT] have also reported recommended values for activity and osmotic coefficients of Ni(ClO4)2 solutions, but their data are based on [69LIB/SAD]. The ion interaction coefficient s(Ni2+, C1O4) was previously derived in [89IUL/POR] (e(Ni2+, C1O~) = 0.375 kg-mol"1) from the experimental data reported in [69LIB/SAD]. Since no uncertainty assignment is given in [89IUL/POR], and in the meantime, another relevant data set has been published [99MAL/CAR], the value of e(Ni2+, CIO4) has been re-determined in the present review.
159
V.4 Group 17 (halogen) compounds and complexes
According to the specific ion-interaction theory the mean activity coefficient y± of Ni(C104)2 can be expressed as (V.74): log10r± = - i ^ ^ - + ^ ( N i 2 + , C l O 4 ) / m , while the osmotic coefficient (breads:
1
^ -^[
1+L5 m
L
21n 1+1
+ (exp.) = 5.82 gem" 3 (according to JCPDS-ICDD card No.8-126). Fleet refined the crystal structure of artificial a-Ni3S2 using X-ray powder intensity data, and essentially confirmed Peacock's results a0 = 5.7465 A, c0 = 7.1349 A, Z= 1, p(calc) = 5.86 gem" 3 (according to JCPDS-ICDD card No.30-863) [77FLE]. Heazlewoodite is the natural form of the compound Ni3S2(cr) found intergrown with magnetite, Fe3O4, in narrow bands in the serpentine at Heazlewood, Tasmania. It occurs massive and granular, without crystal form or cleavage; it is characterised by the following properties: fracture even; brittle, hardness 4, lustre metallic, colour pale yellowish bronze, streak light bronze, non-magnetic, opaque. V.5.1.1.1.1.2
a-Ni7S6
This Ni-sulphide phase is orthorhombic with a^ = 3.274 A, bo = 16.157 A, Co = 11.359 A, space group: Bmmb. Based on /?(exp.) = 5.36 g-cm~3, the unit-cell content is 22.5 Ni and 19.3 S [72FLE]. There are three non-equivalent S sites and five nonequivalent Ni sites per unit cell. Four of the Ni sites are in square pyramidal coordination with S and one is in tetrahedral coordination. Each Ni site is related to at least one neighbouring Ni site by an interatomic distance similar to that of metallic Ni (2.492 A). It is apparent that metallic (Ni-Ni) bonding has been significant in stabilising this structure. V.5.1.1.1.1.3
Pentlandite, (Fe, Ni)9S8(cr)
Pentlandite (JPCRD card 8-90, [52ROB/BRO]) is an important ore of nickel. It is commonly associated with other sulphides such as pyrite (FeS2), chalcopyrite, (CuFeS2) and pyrrhotite (Fe^S) in basic igneous rock intrusions. The largest deposit of this important ore of nickel is at Sudbury, Ontario. Important deposits are mined in Manitoba, Russia, and Western Australia. However, the pure end-member, NJQSS, of pentlandite does not seem to occur in nature and has not been synthesised [87FLE], thus no thermodynamic data are proposed for this phase in the present review.
162
V.5.1.1.1.1.4
V Discussion of data selection for nickel
Godlevskite, Ni,Sg(cr)
Godlevskite forms light yellow microscopic crystals with metallic lustre and gray streak. The hardness is 4 - 5. Fleet showed that godlevskite has a crystal structure based on a distorted cubic close-packed array of 32 S atoms per unit cell, with 20 Ni atoms in tetrahedral coordination and 16 in square-pyramidal coordination [87FLE]. The ideal stoichiometry was established to be Ni9S8(cr). The empirical formula of the specimen was (Ni87Fe0.3)S8, the unit cell is orthorhombic, characterised by a0 = 9.3359 A, bo = 11.2185 A, c0 = 9.4300 A, Z = 4, space group: C222, p(calc) = 5.273 g-cirf3. Originally the X-ray single-crystal and powder diffraction patterns of godlevskite were reported to be consistent with the space groups C222, Cmm2, Amm2, and Cmmm, with a0 = 9.18 A, 60 = 11 -29 A, Co = 9.47 A, and the stoichiometric formula was assumed to agree with P-Ni7S6 (according to JCPDS-ICDD card No. 8-107) [69KUL/ERS]. V.5.1.1.1.1.5
Millerite, p-NiS
Millerite is a low temperature hydrothermal mineral found in cavities in carbonate rocks and as an alteration product of other nickel minerals. The high temperature form a-NiS has not been found in nature. The P-NiS phase is rhombohedral with hexagonal axes a0 = 9.620 A; c0 = 3.149 A, space group: R3m, Z = 9, p(calc) = 5.374 g-cm"3 (according to JCPDS-ICDD card No. 12-41). The crystal structure of P-NiS has been refined by Grice and Ferguson [74GRI/FER] as well as Rajamani and Prewitt [74RAJ/PRE]. The colour of the mineral is brassy yellow, its luster is metallic, crystals are opaque, fracture is uneven, hardness is 3 - 3.5, p(exp.) = 5.3 - 5.5 g-cm"3, the streak is dark green to almost black. Associated minerals are calcite, CaCO3, quartz, SiO2, dolomite, CaMg(CO3)2, bravoite (basically a nickel-rich pyrite), Feo.7Nio.2Coo.1S2, chalcopyrite, CuFeS2, grossular, Ca3Al2(SiO4)3, fluorite, CaF2, and pyrrhotite, Fe,_xS. Notable occurrences include south central Indiana, Keokuk, Iowa, Gap Mine, Pennsylvania and Sterling Mine, New York, USA; Freiberg, Germany; Glamorgan, Wales, England and Sherbrooke and Planet Mines, Quebec. V.5.1.1.1.1.6
Polydymite, Ni3S4(cr)
Gricaenko et al. [53GRI/SLU] investigated a sample from the Ural mountains, Russia, with respect to its X-ray powder patterns and synthesised polydymite, Ni3S4, from aqueous solution. The d values of the natural mineral and the artificial product agreed with each other. Polydymite has a spinel structure, the dimension of the cubic unit cell is a0 = 9.48 A, Z = 8, space group: Fd3m, p(calc) = 4.83 (4.75) g-cirf3 (the numerical value in parenthesis agrees with Ni3S4), p(exp.) = 4.81 g-cnf3 (according to JCPDSICDD card No. 8-107). The colour of the mineral is violet grey, copper red, light grey, or steel grey, lustre is metallic, its hardness is 4.5 - 5.5, its cleavage [001] is indistinct, the streak is black grey, the crystals are opaque. Polydymite occurs in Heazlewood, Tasmania, Aus-
V.5 Group 16 (chalcogen) compounds and complexes
163
tralia, Port Radium, Great Bear Lake, Canada, and in various sites in Germany and USA. Examples for the latter localities are the Grime Au Mine, Siegen District, Rhineland-Palatinate and the Gold Hill district, Boulder, Colorado. V.5.1.1.1.1.7
Vaesite, NiS2(cr)
The synthesis and X-ray diffraction pattern of NiS2(cr) has been reported by de Jong and Willems [27JON/WIL]. This nickel sulphide has a pyrite, FeS2, type structure. Material approaching NiS2(cr) in composition and having a pyrite like lattice has been found in the Kasompi mine of Zaire (formerly the Belgian Congo) by Johannes Vaes of the Union Miniere du Haut Katanga, whom it was named after. Vaes was an important contributor to the mineral development of Zaire. Two specimens close to the composition NiS2(cr), but containing minor amounts of Fe or Fe and Co, were examined microscopically, by an X-ray study, and analysed chemically [45KER]. The crystal structure is cubic, space group PA3, with lattice constant a0 = 5.670 A, Z = 4, p(calc) = 4.45 g-cirf3 (according to JCPDS-ICDD card No. 11-99). The density listed in this card has been calculated for the natural mineral. For pure NiS2(cr) a slightly different value, p(calc) = 4.476 g-cm~3, would be obtained with a0 = 5.670 A. The colour of the mineral is grey, its lustre is metallic, the crystals are opaque, its cleavage [001] is perfect. Vaesite occurs with pyrite randomly disseminated in dolomite. The type locality is the Kasompi mine, Katanga, Zaire. V.5.1.1.1.2
Ni — S phase diagram
The phase diagram of the binary system Ni — S, given by Elliott [65ELL], is based on a systematic investigation of the phase relationships in this system performed by Kullerud and Yund [62KUL/YUN] as well as Sokolova [56SOK]. Craig and Scott [76CRA/SCO] presented a phase diagram which is in accordance with experimental studies of Kullerud and Yund [62KUL/YUN] as well as Arnold and Malik [75ARN/MAL]. Apart from the elements the following mixture phases and stoichiometric compounds were reported: liquid phase (miscibility gap at high mole fractions of sulphur), Ni 3it S 2 (nonstoichiometric high-temperature modification), Nii^S(a) (hexagonal, nonstoichiometric high-temperature form), Ni7±xS6 (non-stoichiometric high-temperature modification), Ni7S6 (stoichiometric low-temperature modification), Ni3S2 (heazlewoodite), NiS(P) (millerite), Ni3S4 (polydymite), and NiS2 (vaesite). Whereas a monotectic reaction between NiS2 and the two liquid phases of the miscibility gap was proposed in [62KUL/YUN], Arnold and Malik [75ARN/MAL] re-investigated the phase relations for the sulphur-rich part of the phase diagram and found a syntectic equilibrium between NiS2 and two liquid phases, NiS2 ^ Li + L2. In 1980 Sharma and Chang [80SHA/CHA] optimised the Gibbs energy functions of all phases in the system Ni — S in order to calculate the phase diagram for the first time. Based on experimental work of Rau [76RAU] as well as Lin et al. [78LIN/HU] the high-temperature modification Ni3±xS2 was shown to consist of two non-stoichiometric phases, viz. Pi-Ni3S2 and p2-Ni4S3, where both phases exhibit a broad homogeneity range.
164
V Discussion of data selection for nickel
In contrast to earlier investigations [54ROS], [62KUL/YUN] the composition of the stoichiometric low-temperature phase, coexisting with Ni3S2 and NiS((3), has been found to be Ni9S8 [94STO/FJE], [96SEI/FJE]. Moreover, the thermodynamic properties of Ni7±xS6, Ni9S8, NiS (high- and low-temperature forms), and Ni3S2 were reinvestigated experimentally by St0len et al. [91STO/GRO], [94STO/FJE], [95GRO/STO]. In particular, in the case of NiS the enthalpy of transition estimated by Biltz et al. [36BIL/VOI] was revised and a new value was obtained by [95GRO/STO] based on careful heat capacity measurements employing an adiabatic-shield calorimeter. Thus, a new optimisation of the Gibbs energy functions of some phases in the system Ni - S has been necessary in order to recalculate the phase diagram and to recommend reliable thermodynamic data for the solid nickel sulphides. The phase diagram of the binary system Ni - S is depicted in Figure V-29 and Figure V-30. The calculation of the phase diagram is based on Gibbs energy equations listed in Table V-15. It is worth mentioning that the small homogeneity range of Ni7S6 has been neglected, i.e., this phase is treated as a stoichiometric compound.
Figure V-29: Phase diagram for the binary system Ni - S with T plotted versus mole fraction of sulphur, o [10BOR], • [54ROS], A [70NAG/ING], V [75MEY/WAR], x [75RAU2].
165
V.5 Group 16 (chalcogen) compounds and complexes
Figure V-30: Section of the calculated phase diagram for the composition range 0.30 < x s < 0.55. Dashed lines have been extrapolated, indicating regions where phase equilibria cannot be calculated by using the Gibbs energy functions of Table V-15.
Table V-15: Gibbs energy functions for the binary system Ni - S relative to metallic nickel, Ni(s), and gaseous sulphur, S2(g); AfG° = a + bT + cTln(T), Lt = a + bT +
cT\n(T). Range of validity
a (J-mol ')
b (J-K-'-mol"') c (JK ' m o r ' )
Liquid phase (three-suffix Margules model) Ni
900 10 3 5 mol-kg ') the uncommon complexes Ni2HS3+ and Ni3HS5+ become the most dominant species in aqueous solution. As this situation seems to be unrealistic and no studies on the structure of these complexes are reported in literature, we select thermodynamic data only for the aqueous species NiHS+: log, 0 #j((V.83), 298.15 K) = (5.18 ±0.20). This selection, with NEA-TDB selected auxiliary data, yields: AfG°,(NiHS+, 298.15 K) = - (63.1 + 2.5) kJmol 1 . Owing to the high uncertainty of the second dissociation constant of H2S the following dissolution reaction of nickel sulphide: NiS(cr) + 2 H + ^ Ni 2+ + H2S(aq)
*K°_0
(V.86)
K°o.
(V.87)
is thermodynamically better defined than NiS(cr) ^ Ni2+ + S2~
To the best of our knowledge no reliable studies on the solubility of nickel sulphides in aqueous media have been published yet. Some investigations refer to amorphous solid phases, resulting in too high values for the solubility constants [14THI/GES], [70CAR/LAI]. In many cases reaction times were far too short to attain solid-solute equilibrium, yielding values for the solubility constants that are too low [14THI/GES], [24MOS/BEH], [47DON]. Clearly thermodynamic quantities based essentially on experimental data of [14THI/GES] are unreliable [93MAK/VOE]. Recalculated values for both the solubility constant, log10 K°o, corresponding to the dissolution reaction (V.86) and the solubility product, log10 K°o, using the currently accepted value for the second dissociation constant of hydrogen sulphide (H2S(aq) ^ 2H+ + S2 logl0 K; = - 25.99 [92GRE/FUG]), are listed in Table V-23.
V.5 Group 16 (chalcogen) compounds and complexes
179
Figure V-35: Distribution of nickel sulphide complexes as a function of total molality of nickel (II) in aqueous solutions saturated with H2S, T= 298.15 K, / = 1.00 molkg ' NaCl. Solid line: Ni2+; dashed line: NiHS+; dotted line: Ni2HS3+; dash-dotted line: Ni3HS5+.
180
V Discussion of data selection for nickel
Table V-23: Solubility constants for Ni(II) sulphide at T= 298.15 K and / = 0. log10 A^°o
>°gio C o
Reference
-0.5
-27.5
[1883THO], [09BRU/ZAW]
-3.98
- 30.98
[24MOS/BEH], [31K0L]
-17.8
[70CAR/LA1]
9.20 -(1.53 ±0.30)
(a)
-(2.10±0.30) l b )
(a)
this review
-29.1(b»
this review
- 28.5
(a) NiS(a). (b) NiS(B), millerite.
The solubility data scatter remarkably and none of these is consistent with the solubility constants selected in this data assessment. The solubility constant given by Bruner and Zawadzki [09BRU/ZAW] is based on a calorimetric value for the enthalpy of reaction for the precipitation of NiS from aqueous solution: Ni2+ + S2 ^ N i S , reported by Thomson [1883THO]. Thiel and Gessner [14THI/GES] carried out a comprehensive study of the dissolution behaviour of NiS prepared by precipitation from aqueous solutions at various conditions. They found three fractions, viz. I-NiS, II-NiS and III-NiS, with different solubilities in hydrochloric acid. The variations of the solubility of these fractions, observed by the authors, can be explained in terms of different dissolution kinetics rather than in terms of different thermodynamic properties, since reaction times were certainly too short to attain equilibrium between the solid phase and the aqueous solution. Licht [88LIC2] has recalculated the results of [14THI/GES] and obtained Gibbs energies of formation for I-NiS, II-NiS and III-NiS, respectively. The solubilities of Ni(II) sulphides in solutions of hydrochloric acid according to [88LIC2], [14THI/GES] as well as the present assessment are depicted in Figure V-36. Whereas the solubility of I-NiS, which seems to correspond to an amorphous material, is too high, the solubilities of II-NiS and III-NiS are by far too low because of too short reaction times. The fractions II-NiS and III-NiS are apparently crystalline products. This assumption is likewise confirmed by Donges [47DON] who has pointed out that, besides amorphous products, the metastable high-temperature modification, NiS(a), is precipitated from aqueous solutions. Furthermore, it should be mentioned that millerite, NiS(P), is the only nickel sulphide identified so far in natural low-temperature anoxicsulphidic environments according to Thoenen [99THO].
V.5 Group 16 (chalcogen) compounds and complexes
181
Figure V-36: Solubility of Ni(II) sulphides in aqueous solutions saturated with H2S as a function of initial concentration of hydrochloric acid at 298.15 K. Dashed lines refer to I-NiS, II-NiS and III-NiS as experimentally investigated by Thiel and Gessner [14THI/GES] and thermodynamically interpreted by Licht [88LIC2]. Solid lines correspond to NiS(a) and NiS(p) using the thermodynamic data selected in the present compilation. Dotted lines are related to calculations taking into account the uncommon aqueous species Ni2HS3+ and Ni3HS5+, respectively.
V.5.1.2 V.5.1.2.1
Nickel sulphates Aqueous nickel (II) sulphato complexes
The complexation reactions of Ni2+ with SO^ have been the subject of a large number of investigations (Table V-24). These include studies by cryoscopy [55BRO/PRU], [56KEN], [58KEN], [71 ISO], conductivity [1895FRA], [56FIA/SHE], [73KAT], [74MAK/MAS], [76SHI/TSU], [79FIS/FOX], [93SHE], [2000TSI/MOL], [2001MOL/TSI], potentiometry and other electrochemical techniques [59NAI/NAN], [61TAN/OGI], [63TAN/SAI], [65POT], [68PRA], [70TUR/RUV], solubility [67GES/NEU], [70MIR/MAK], spectroscopy [64LAR], [70MIR/MAK], [77ASH/HAN], [78BEC/LIU], [80LIB/SAD] calorimetry [69IZA/EAT], [74BLO/RAZ], [85DRA/MAD] and isopiestic measurements [80LIB/SAD], [83HOL/MES]. There is a broad consensus, primarily based on the spectroscopic evidence, that, for temperatures below 100°C, association of these ions in dilute solutions
182
V Discussion of data selection for nickel
is predominantly an outer-sphere interaction, with sulphate gradually displacing water molecules in the inner sphere around Ni2+ at higher temperatures and concentrations [77ASH/HAN], [78BEC/LIU]. Indeed, for 25°C, it is not clearly defined what does and what does not constitute a nickel sulphate complex. The nature of a specific type of experiment and the interpretation of experimental results affect any proposed value for the formation constant of NiSO4(aq) according to reaction (V.88): Ni2+ + SO;;- ^
NiSO4(aq).
(V.88)
In particular, the reported value of the formation constant extrapolated to zero ionic strength is usually dependent on the particular method used to assign values to the ionic activity coefficients. In the present review, the objective is to assign a value that is consistent with the use of a particular formulation of the specific ion interaction equations (the SIT approach, cf. Appendix B). Among the assumptions inherent in the SIT treatment is that although the association constant depends on the total ionic strength of the solution, on the ionic interactions between Ni2+ and anions of the supporting electrolyte, and on the ionic interactions between SOj" and cations of the supporting electrolyte, all specific interactions between Ni + and SO^" are accounted for by the NiSC>4 formation constant. Nevertheless, Holmes and Mesmer [83HOL/MES] have shown that a Pitzer-type ion-interaction model can be used to provide a satisfactory interpretation of isopiestic data without assuming formation of a discrete NiSO4 complex, even at 140°C. Several studies on ionic strength effects have used nickel sulphate as a model 2:2 electrolyte [55BRO/PRU], [73KAT], [75TAM]. In such cases, for the sake of consistency, it has been necessary to abandon the more sophisticated treatments, and to reevaluate experimental data for complexation using the SIT. Perhaps because of the interest in nickel sulphate as a "typical" 2:2 electrolyte in aqueous solution, few of the association studies have been done as a function of concentration of a supporting electrolyte. Exceptions include the cryoscopic study of Kenttamaa [56KEN], [58KEN] and the polarographic studies of Tanaka and Ogino, [61TAN/0GI], and Tanaka, Saito and Ogino [63TAN/SAI], Some spectroscopic studies have also been done using supporting electrolytes, often at very high concentrations [70TUR/RUV], [77ASH/HAN]. In the temperature range (0 to 50°C) in which an outer-sphere complex predominates over inner sphere complexes, several studies [59NAI/NAN], [73KAT], [76SHI/TSU] have been used to determine the enthalpy of the complexation reaction (V.88). In addition, the enthalpy of reaction has been determined calorimetrically by Izatt et al. [69IZA/EAT], though a useful value can be obtained from their enthalpy titrations only if an independent method is used to determine the association constant [73POW] (cf. Appendix A). The comparison of derived values for the enthalpy of association also provides a tool for evaluating the association constant data.
183
V.5 Group 16 (chalcogen) compounds and complexes
Table V-24: Experimental values for the association constants and other thermodynamic quantities for nickel sulphate. Recalculated value(a)
Method
tl°C
Medium (aq)
Reported value
cond
25
0 (extrap.)
A", = 250
(220.7 ±13.4)
[32MON/DAV], recalc.of [1895FRA]
fpt fpt fpt
0 -0.2
0.005-0.09 self 0.0484 m KC1O4
K, =125-244 A", = 51
(161.3 ±20.0)
[55BRO/PRU] [56KEN]
-0.8
0.258 m KCIO3
AT, = 19
-2.8 0
1.15mKNO 3 < 0.05 m HC1
AT, = 4.90 A", = 121
10 15
< 0.05 m HC1 < 0.05 m HC1
AT, = 151 AT, = 174
25 35 45
< 0.05 m HC1 < 0.05 m HC1 < 0.05 m HC1
A", =211 Ki = 247 A", = 289
pot
A r / / ° = 13.85 kJmor 1 ArS° =90.8J-KTl-mor1 fpt
-3.06
1.15mKNO 3
[58KEN] (98.0 + 23.0) (126.8 ±16.0)
ir pot cal
25
0.2 M KNO3
25
1.0MNaClO 4
RT 35
1.0 M self 0 (extrap.) 0 (extrap.)
25
(236.9 + 30.0) see text
AT, = 11.6 AT, =3.7
[59ROS/ROS], recalc. of [58KEN] [61TAN/OG1] [63TAN/SAI]
dinner sphere 1P oucer sphere = 0.1
[64LAR]
A", = 1 1 1
[68PRA] [69IZA/EAT]
A", = 5.01
Ki = 646
spec/sol
25.0
3.0 M LiClO4
A r W°= 1.71 kJ-mor1 ArS° =59.0J-IC'-mor' AT, = 1.8
pol fpt
25.0 0
5.0 M NaClO4 0.005-0.03 self
A", = 15.6 A~,= 118-250
cond
0.0 5.0
0 0
A", = 156 A", = 162
(213.8 ±4.4)
10.0 15.0
0 0
Kt = 168 A", = 176
(223.9 ±4.9) (230.3 ± 4.7)
20.0 25.0 30.0 35.0
0
A", = 183 AT,= 187 A", = 191 A", = 201
(238.3 ± 5.3) (247.3 ± 5.8) (257.1 +7.1)
Ki = 209
(279.9 ± 8.8)
A", = 217
(295.6 + 9.4) (5.3 ± 2.0)
0 0
40.0
0 0
45.0
0
25.0
0
[59NAI/NAN]
(166.2 ±21.8) (172.8 ±31.9) (210.2 ±31.2)
A"2=33.6 pol pol
Reference
ArW^ = 5.31 kJ-mor'
see text [70MIR/MAK] [70TUR/RUV] (153.8 ±20.0)
[71 ISO] [73KAT]
(217.9 + 4.9)
(267.8 ± 7.8)
(Continued on next page)
184
V Discussion of data selection for nickel Table V-24: (continued) Recalculated value(a)
Method
t/°C
Medium (aq)
cal
25.0
0
A r //° =5.72-5.93 ld-mol"1
[73POW], recalc of [69IZA/EAT]
cal
25.0
A,«° = 2.59 kJ-mol"1
[74BLO/RAZ]
var
25.
1.5MUC1CV 0.75M Li2SO4 0.018 self+? 0.048 self+?
Reported value
Reference
Ar5° = 61.5J-K"'-mol"'
£, = 280
[74MAK/MAS]
0.069 self+?
£, = 170 £ , = 110
cond
0.0
0.
£, = 270
cond
15.0
0.
£,=457(0.1 MPa) £,=227 (160 MPa)
(187.7 + 2.1)
25.0
0.
£,=490(0.1 MPa) £,=269 (160 MPa)
(193.8+ 1.9)
40.0
0.
£,=575(0.1 MPa) £,=313(160 MPa)
(226.7 ±2.8)
25.0
0
ArW° = 6.90 kJ-mol"1
(6.0 + 2.5) kJ-mor1 (0.101 MPa)
25.0
0
Ar5^ = 109 J-Kr'-mor1 (0.101 MPa)
25.0
0
uv
25.0
0
15.0
0.
[77ASH/HAN]
£,=5.5 £,=5.4 A r #° = 5.31 kjmol"1 Ar
cond
[76SHI/TSU]
A,r = 8.6cm'mor'
25.0(b) 5 M Na2SO4/NaClO4 £,= 5.89 47 £,=6.8 60 70
cond
[75TAM], recalc. of [73KAT]
0 m
-
.
-1
[77KAT] recalc.of [73KAT]
1-1
£,=200(0.1 MPa)
(183.3 ±30.4)
[79FIS/FOX]
£,=105 (203 MPa)
cond
cond cond.
25.0
0.
25.0 25.0
0 0.
25.0
0
25.0 20.0
< 0.06 M 0
£,=196(0.1 MPa)
(199.4 + 28.5)
£,=93.5 (203 MPa) A1P°= 11.6 cm'-mor1 £,=169 A r //° = 5.24kJ-mol"' ArS° = 60.2J-K.~'-morl £,=230.1 £,=139
[81YOK/YAM] recalc. of [73KAT] [93 SHE]
[2000TSI/MOL], [2001MOL/TS1] (a) Uncertainties are the statistical uncertainties from the re-analysis of the data. The overall estimated uncertainties are discussed in the text. (b) This value was reported by the authors based on an extrapolation to 25°C.
185
V.5 Group 16 (chalcogen) compounds and complexes
Table V-25 provides a summary of the results from re-analysis of the conductance data for NiSO4(aq) using the series form [80PET/TAB]1 of the Lee-Wheaton equation [79LEE/WHE], modified so that the distance parameter is forced to the appropriate value for the SIT model used in the TDB. No value could be recalculated from the work of Shehata [93 SHE] because values of the concentration-dependent conductances were not reported. Results from the other conductance studies at 25°C are compared in Figure V-37. The molar conductances agree well at low concentrations at 25°C. At higher concentrations (> 10 3 M), values are more divergent, especially those from the early study of Frank [1895FRA]. The values of Fisher and Fox [79FIS/FOX] and Shimizu, Tsuchihashi and Furumi [76SHI/TSU] are in very good agreement. As can be seen from Table V-25, the agreement is slightly poorer for 15°C. Values of Ao from analysis of the molar conductances from these two studies are substantially smaller than the values calculated from the molar conductances reported by Katayama [73KAT]. Nevertheless, all four studies lead to values of log,0 K° ((V.88), 298.15 K) between 2.29 and 2.39. Table V-25: Results of reanalysis of NiSO^aq) conductance data using a form of the Lee-Wheaton equation that is compatible with the SIT procedure. t/°C
KA W
0
213.8
5
217.9
10
223.9
15
230.3 183.3 187.7
20
238.3 118.9
25
247.3 220.7 193.8 199.4
30
257.1
35
267.8
40
279.9 226.7
45
295.6
±oKA (statistical) ±4.4 ±4.9 ±4.9 ±4.7 ±30.4 ±2.1 + 5.3 + 7.2 + 5.8 + 13.4 ±1.9 ±28.5 ±7.1 ±7.8 ± 8.8 + 2.8 ±9.4
Assessed
logioATA
Reference
(2a) 68.82
[73KAT]
2.34
80.48
[73KAT]
2.35
92.96
[73KAT]
2.36
106.01
[73KAT]
2.26
105.28
[79F1S/FOX]
2.27
103.63
[76SH1/TSU],
2.38
119.82
[73KAT]
2.08
123.34
[2000TS1/MOL]
20
2.39
134.35
[73KAT]
60
2.34
136.57
[1895FRA]
30
2.29
132.99
[76SHI/TSU]
2.30
132.73
[79FIS/FOX]
2.41
149.50
[73KAT]
2.43
165.16
[73KAT]
2.45
181.47
[73KAT]
2.36
179.12
[76SHI/TSU]
2.47
198.58
[73KAT]
2.33
60
(a) The constant KA from the Lee-Wheaton treatment is assumed to be equivalent to K" (V.88).
1
In 1979, Dr. A.D. Pethybridge kindly provided one of the reviewers (R.J. L.) with a version of the computer program that has been used here (with modifications) for the analysis of conductance data.
186
V Discussion of data selection for nickel
As discussed in Appendix A, it appears that the reported molar conductance values from at least two of the studies [73KAT], [76SHI/TSU] are for "rounded" concentration values. If the values reported are "smoothed" molar conductances, the true experimental scatter may have been greater. The "statistical" uncertainties listed in Table V-25 are the values from the data re-analysis [80PET/TAB]. Fitting of the (unavailable) primary data from those studies probably would have led to larger calculated statistical uncertainties in KA (assumed equivalent to AT°(V.88)). Estimated (2a) uncertainties for all four values of AT°((V.88), 298.15 K) are listed in Table V-25. Based solely on the conductance data at 25°C, at / = 0 the value of A"°((V.88), 298.15 K) is (228.1 + 15.5), or log10 K° ((V.88), 298.15 K) = (2.36 + 0.03). Figure V-37: The variation of molar conductances of aqueous nickel sulphate solutions with concentration at 25°C. • : [1895FRA], • : [73KAT], A: [76SHI/TSU], A : [79FIS/FOX]
Earlier reviews [82WAG/EVA], [99ARC] gave substantial weight to the emf study of Nair and Nancollas [59NAI/NAN]. As shown in Appendix A, recalculation of values from that study to be consistent with the SIT procedure described in Appendix B, requires values for the dissociation constant of HSO^, and several interaction coefficients, e(Ni 2+ ,Cl), £(H+,C1), e(H + ,HSO 4 ) and e(Ni2+,HSO^). The values calculated for the formation constant of NiSO4(aq) are extremely sensitive to the activity coefficient model, to the values used for the interaction coefficients and to the values assumed
V.5 Group 16 (chalcogen) compounds and complexes
187
for the first protonation constant for the sulphate ion. As discussed in Appendix A, this work can be used to determine a value at / = 0 of K°((VM), 298.15 K) = (173 + 80), or logl0 #°((V.88), 298.15 K) = (2.24 ± 0.20). If the estimated uncertainty bounds are considered, the value from Nair and Nancollas overlaps with those from the conductance studies. Using the weighted average of the conductance and emf data ((228.1 ± 15.5) and (173 ± 80), respectively, weighted according to their uncertainties), the selected value of AT°((V.88), 298.15 K) is: AT°((V.88), 298.15 K) = (226.1 ±15.2) or
log10 K° ((V.88), 298.15 K) = (2.35 ± 0.03).
None of the studies at higher ionic strength at 25 °C is a systematic study in which the concentration of a single supporting electrolyte was varied. The results are summarised in Table V-26.
Table V-26: Experimental values for the first association constant of nickel sulphate determined in the presence of supporting electrolytes (see Appendix A). Values of log10 K° were determined using the SIT (Appendix B). Medium
//°C
lo
gl0 Kl la)
l°g,o*,°
Reference
0.0484 m KCIO4
-0.2
1.70
2.34
[56KEN]
0.258 m KC1O,
-0.8
(1.27 ±0.30)
(2.35 ± 0.30)
[58KEN]
1.15mKNO 3
-2.8
(0.69 + 0.20)
(2.16 + 0.20)
0.2 M KNO3
25
(1.06 ±0.20)
(2.13 ±0.20)
[61TAN/OGI]
1.0MNaClO 4
25
(0.57 ± 0.20)
(1.95 ±0.20)
[63TAN/SAI]
3.0 M LiClO4
25
0.26
1.07""
[70MIR/MAK]
5.0MNaClO 4
25
1.19
1.70(c)
[70TUR/RUV]
(a)
The added nickel sulphate has a substantial effect on the ionic strength and the derived value of logio^i.
(b)
The value is anomalously low compared to all other reported values.
(c)
The ionic strength is beyond the range for use of the SIT (Appendix B).
Except for the spectroscopic value of Mironov et al. [70MIR/MAK], the results are qualitatively in agreement with the results of the conductivity and emf studies at low ionic strengths. The association constant values from the polarographic studies [61TAN/OGI], [63TAN/SAI], [70TUR/RUV] do tend to be somewhat smaller than the values selected above. The estimated uncertainties in the values from [61TAN/OGI] and [63TAN/SAI] are large, and the raw data are unavailable for re-analysis. Therefore, those results are not incorporated in the weighted-average selected value. The cryoscopic results from experiments in different saturated solutions [56KEN], [58KEN] are in excellent agreement with the value (2.33 + 0.06) for 0°C from the conductance data of Katayama and the average value of (2.20 + 0.06) from the two cryoscopic studies in
188
V Discussion of data selection for nickel
water [71 ISO], [55BRO/PRU]. They are also in marginal agreement with the value (1.99 ±0.10) recalculated from the emf data of Nair and Nancollas [59NAI/NAN]. At high sulphate concentrations, there is some evidence for formation of Ni(SO4)2" in several studies {e.g., [58KEN], [59ROS/ROS], [63TAN/SAI]). However, the evidence is reasonably ambiguous, and may only reflect systematic errors in the experiments [58KEN]. No value is selected for K^ in the present review. There is considerable spectroscopic evidence that an inner-sphere sulphato complex exists at high sulphate concentrations and may become more important as the temperature is increased [77ASH/HAN], [78BEC/LIU]. Near 25°C, inner-sphere association is only about 10% of the outer-sphere association. No values are selected in this review for the formation constants of the inner sphere sulphato complexes [78BEC/LIU]. The polymeric species Ni 2 (SO 4 )^ proposed by Brintzinger and Osswald [34BRI/OSS] also is not considered credible [41KIS/ACS]. V.5.1.2.1.1
Enthalpy of formation of NiSO4(aq)
There have been several studies [59NAI/NAN], [73KAT], [76SHI/TSU] of the temperature dependence of the formation constant of the outer-sphere complex NiSO^aq), and, as discussed above, several other measurements at temperatures between — 2.8 and 45°C. The recalculated values of log10 K° are plotted in Figure V-38. The sets of values from most of the conductance and cryoscopic measurements are in good agreement. The temperature-dependencies of the values from these studies follow similar parallel curves. The higher temperature results of Nair and Nancollas [59NAI/NAN] follow the same type of curve for temperatures > 15°C (288.15 K), but the values for the two lowest temperatures are inconsistent with the cryoscopic and conductivity results. The recalculated values of log10 K° from Katayama [73KAT] are used to calculate (5.3 ± 2.0) kJmol ' for the average value of A r //° (V.88) between 0 and 45°C. Similarly, A r //° = (6.0 ± 2.5) kJ-mol' (average for 15 to 40°C) from the sparse data of Shimizu, Tsuchihashi and Furumi [76SHI/TSU] and A r #° = (12.6 + 3.0) kJ-mof' (average for 0 to 45°C, 2a uncertainty) from the work Nair and Nancollas [59NAI/NAN]. Izatt et al. [69IZA/EAT] determined the enthalpy of reaction at 25°C by a titration calorimetry method. The intent was to determine both the association constant and the enthalpy of association in a single experiment. Unfortunately, the enthalpy and association constant values, as derived iteratively [69IZA/EAT] or by the method of least squares [73POW], are highly correlated. Powell showed that useful enthalpy values could be obtained only if a value was assumed for K°. Starting from the 25°C value from Nair and Nancollas [59NAI/NAN], Powell used the data of Izatt et al. [69IZA/EAT] to obtain A r //°= (5.8 ± 0.2) kJmol '. The calculation requires values for the enthalpy of protonation of sulphate and the enthalpy of dilution of the tetramethylammonium sulphate titrant. Izatt et al. [69IZA/EAT] and Powell [73POW] also incor-
V.5 Group 16 (chalcogen) compounds and complexes
189
porated values for an association constant and enthalpy of association for the Me4NClC>4(aq) ion pair. Determination of the values for Me4NClO4(aq) required using an explicit value for the association constant between Na+ and S O ^ . As discussed in Appendix A, a reasonable value from this work is A r //° = (5.8 ± 2.0) kJ-mol ', based on Powell's recalculation and an increased estimated uncertainty. In the present review, we exclude the value from Nair and Nancollas [59NAI/NAN], and accept the weighted average of the results from the other three studies [69IZA/EAT], [73KAT], [76SHI/TSU]: A r //° ((V.88), 298.15 K) = (5.66 ± 0.81) kJmol '. From the selected values of log10 K° and A r / / ° , the following quantities are calculated: AfG°(NiSO4, aq, 298.15 K) =-(803.2 + 0.9) kJmor 1 A f //° (NiSO4, aq, 298.15 K) = - (958.7 + 1.3) kJ-mol' and
S° (NiSO4, aq, 298.15 K) = -(49.3 ± 3.1) J-K'-mol '.
Figure V-38: Values of log10 K° for nickel sulphate association at different temperatures. • : [1895FRA], *: [55BRO/PRU], O: [58KEN], • : [59NAI/NAN], +: [61TAN/OGI], V: [63TAN/SAI], X: [71ISO], n: [73KAT], A: [76SHI/TSU], A: [79FIS/FOX], T: [2000TSI/MOL].
190
V Discussion of data selection for nickel
V.5.1.2.1.2
Heat capacity values
Drakin, Madanat and Karlina [85DRA/MAD] reported measurements of the apparent molar heat capacities for nickel sulphate solutions at 50°C, 70°C and 90°C. The data are too sparse to be used to estimate the partial molar heat capacity of the ion pair. V.5.1.2.2
Solid nickel sulphates
Several hydrated nickel sulphate solids from NiSO4-7H2O to NiSO 4 H 2 O, including two forms of NiSO4-6H2O, a tetrahydrate, and a dihydrate have been reported [04STE/JOH], [39SIM/KNA], [39ROH], [64KOH/ZAS], though the domains of stability for and kinetics of interconversion between some of the intermediate hydrates remain contentious. In saturated solutions at 1 bar, the heptahydrate converts to the ot-hexahydrate just above 300 K. Therefore, samples of the heptahydrate often contain amounts of the hexahydrate, and samples of the a-hexahydrate can have the heptahydrate as a contaminant. This has complicated attempts to measure the physical properties for the individual solids. Dehydration of the monohydrate to anhydrous NiSO4(s) occurs in a stream of water-saturated N2 only above 630 K [78SIR/SEM]. Decomposition of NiSO4(s) to NiO(s) and sulphur oxides becomes important at temperatures above 900 K [83GAU/BAL]. V.5.1.2.2.1
NiSO4-7H2O(cr)'
The heptahydrate is the stable nickel sulphate hydrate at 298.15 K. Stout et al. [66STO/ARC] reported low-temperature heat capacity measurements (from 16.1 K to temperatures above 279 K) for samples that were slightly hypostoichiometric and hyperstoichiometric in water (NiSO4-6.867H2O(cr) and NiSO4-7.301H2O(cr)) with respect to the heptahydrate, and combined the results with those from an earlier study [41STO/GIA] to obtain 5° (NiSO4-7.00H2O, cr, 298.15 K). There do not appear to be any other comparable measurements for this solid, and the entropy and heat capacity values for NiSO4-7.00H2O(cr) at 298.15 K from Stout et al. are selected in the present review. Though some of the auxiliary data used by Stout et al. differ slightly from those used in the present review, such differences are much less than the rather small estimated experimental uncertainties (cf. Appendix A for [66STO/ARC]). The selected values are: S° (NiSO4-7.00H2O, cr, 298.15 K) = (378.95 ± 1.00) J-K '-mol ' C;m(NiSO4-7.00H2O, cr, 298.15 K) = (364.6 ±4.0) J-K ' m o l 1 . The value of the difference between the enthalpy of solution of a-NiSO4-6.00H2O(cr) and NiSO4-7.00H2O(cr) in water ((7.68 ±0.10) kJ-mol1) has been reported by Stout et al. [66STO/ARC]. In the present review, AS0]H°m (NiSO4-6.00H2O, a, 298.15 K) = (4.485 + 0.200) kJ-moF1 in water. Using this and the enthalpy of hydration, the value of Asol//° (NiSO4-7.00H2O, cr, 298.15 K) = 1
The dissolution of this compound is discussed in Section V.2.1.3.1.
V.5 Group 16 (chalcogen) compounds and complexes
191
(12.17 ± 0.23) kJ-mol 1 in water is obtained. The value 15.644 kJ-moF1 was reported by Przepiera et al. [93PRZ/WIS] as the heat of solution of the heptahydrate in water to produce a solution with a 5 x 10 4 mass ratio of salt to water. If the value is assumed to be for the enthalpy of solution to this stated dilution (approximately 0.002 m), the application of a heat of dilution correction to / = 0 gives a value (12.5 kJ-mor') in good agreement with that based on Goldberg et al. and Stout et al. Using the selected values of AfG° (Ni2+) = -(45.77 + 0.77) kJmol 1 (Section V.2.1.1), and Aso]G°m= (12.94 ± 0.11) kJ-mof1 (Section V.2.1.3) with standard auxiliary data, Af G° (NiSO4-7.00H2O, cr, 298.15 K) = - (2462.7 ± 0.9) kJ-moF1. With the selected entropy value, 5° (NiSO4-7.00H2O, cr, 298.15 K) = (378.9 ± 1.0) J - K ' m o l 1 is calculated and Af//n°, (NiSO4-7.00H2O, cr, 298.15 K) = - (2977.3 + 1.0) kJ-mol '. Bell [40BEL] reported vapour pressure measurements that showed that at 298.15 K removal of water from NiSO4-7H2O(cr) is less favoured than removal of D2O from NiSO4-7D2O (see Appendix A). In this review, no values are selected for the thermodynamic quantities of NiSO4-7D2O. V.5.1.2.2.1.1
cc-NiSO4-6H2O
In contact with saturated solutions, the a-hexahydrate becomes the stable nickel sulphate hydrate near 302 K, and is transformed to the p*-hexahydrate at approximately 327 K. Stout et al. [66STO/ARC] reported low-temperature heat capacity measurements (from 13.7 to 321.0 K) for samples that were slightly hyperstoichiometric (NiSO4-6.010H2O) in water with respect to the hexahydrate, and combined the results with those for 1.1 to 21.8 K from an earlier study [64STO/HAD] and with results for the heptahydrate to obtain 5° (NiSO4-6.00H2O, 298.15 K). There do not appear to be any other comparable measurements for this solid, and the entropy and heat capacity values for NiSO4-6.00H2O at 298.15 K from Stout et al. are selected in the present review. Though some of the auxiliary data used by Stout et al. differ slightly from those used in the present review, such differences are much less than the rather small estimated experimental uncertainties (cf. Appendix A for [66STO/ARC]). The Reviewers select: 5° (NiSO4-6.00H2O, a, 298.15 K) = (334.45 + 1.00) J-IC'mor 1 C;m(NiSO4-6.00H2O, a, 298.15 K) = (327.9 ± 1.0) J-K 'mol '. Values of the enthalpy of solution of a-NiSO4-6.00H2O in water have been reported by Stout et al. [66STO/ARC] and Goldberg et al. [66GOL/RID] and, allowing for the enthalpies of dilution, the values from the two sources are in good agreement.
192
V Discussion of data selection for nickel
The more extensive set of measurements by Goldberg et al. resulted in more dilute fmal solutions of nickel sulphate, and are preferred here. Goldberg et al. used enthalpy of dilution values of Lange [59LAN] and Lange and Miederer [56LAN/MIE] to extrapolate the enthalpy of solution to / = 0, and obtained Aso,//° = 4.81 kJ-mol' (as is normal for 2:2 electrolytes, the experimental heat of dilution for NiSO4 at low ionic strength changes much more strongly with concentration than would be predicted from simple Debye-Hiickel theory, cf. Appendix A for [66GOL/RID]). The largest uncertainty in this heat of solution value comes from the extrapolation to / = 0. In the present review the selected value is: Asol//° ((V.I6), NiSO4-6.00H2O, a, 298.15 K) = (4.485 + 0.200) kJ-moF1. This is one of the key quantities in the derivation of thermodynamic quantities ofNi (c/Section V.2.1.3). 2+
Based on A f //° (NiSO4-7.00H2O, 298.15 K) = -(2976.94 + 0.95) kJmol ', and the difference between the enthalpies of solution of a-NiSO4-6.00H2O and NiSO4-7.00H2O in water ((7.68 ± 0.10) kJmol 1 ) reported by Stout et al. [66STO/ARC], A f //° (NiSO4-6.00H2O, a, 298.15 K) = - (2683.8 ± 1.0) kJmol""'. The value of Af G; (NiSO4-6.00H2O, a, 298.15 K) is calculated from the selected values of A f //° (NiSO4-6.00H2O, 298.15 K) and S°m (NiSO4-6.00H2O, a, 298.15 K) and the TDB auxiliary data in Table IV-1: Af G° (NiSO4-6.00H2O, a, 298.15 K) = - (2225.5 ±1.1). V.5.1.2.2.1.2
Other hydrated nickel sulphate solids
Solubility data [34BEN/THI], [39ROH] indicate that the P-hexahydrate is the stable solid in contact with saturated solutions of nickel sulphate in water for temperatures between 325 and 358 K, where lower hydrates begin to predominate. As might be expected, addition of sulphuric acid results in the appearance of the lower hydrates at lower temperatures [39ROH]. Based on their differential scanning calorimetry results, Rabbeting et al. [75RAB/WAN] reported ATHm= (6.53 ± 0.17) kJmol 1 for the reaction: a-NiSO 4 -6.00H 2 O^ P-NiSO46.00H2O.
(V.89)
Even a difference in the heat capacities of the two solids of 3 J-K. 'mol ' would change this value by only 0.1 kJ-mol ' (approximately) between 298.15 K and the temperature of the experiment. Therefore, a value of the transition enthalpy Ar//n°,((V.89), a ^ P , 298.15 K) = (6.53 + 0.20) kJ-mor1 is assumed, and A f //° (NiSO4-6.00H2O, p, 298.15 K) = - (2677.3 ± 1.0) kJ-mol '
V.5 Group 16 (chalcogen) compounds and complexes
193
is retained for selection. If the transition temperature is 325 K, and the heat capacities of the two solids are assumed to be temperature independent and equal between 298.15 and 325 K, the selected value for molar entropy is: S°m (NiSO4-6.00H2O, p, 298.15 K) = (354.5 ± 5.0) J-K ' m o i 1 where the uncertainty has been increased to allow for the approximations in the calculation and the uncertainty in the transition temperature. These selections yield: AfG° (NiSO4-6.00H2O, p\ 298.15 K) = - (2224.9 ± 1.8) kJ-mol"1. Rabbering et al. [75RAB/WAN] and Siracusa et al. [78SIR/SEM] carried out thermal decomposition studies by a variety of techniques, and found evidence for decomposition of the P-hexahydrate to a tetrahydrate at temperatures near 400 K and to the monohydrate near 440 K. Nandi et al. [79NAN/DES] found that the dehydration pattern for thermogravimetric measurements carried out at 5 K-miiT1 were different for well-crystallised and powdered samples of the heptahydrate and hexahydrate. However, the intermediate hydrates were not well characterised. Several studies [62CAI/POB], [64KOH/ZAS], [75RAB/WAN], [78SIR/SEM] have shown that the monohydrate does not readily lose water below 500 K. Goldberg et al. [66GOL/RID] determined the heat of solution of a nickel sulphate hydrate having the composition NiSO4-5.059H2O, and a value for the enthalpy of hydration of NiSO4-H2O(s) to the hexahydrate was calculated by assuming that the mixed solid was a mixture of NiSO4-6.000H2O(s) and NiSO4H2O(s). It seems more likely that the solid was a mixture of NiSO4-6.000H2O(s) and NiSO4-4H2O(s) [65JAM/BRO]. Kohler and Za'ske [64KOH/ZAS] reported equations for water vapour pressures as a function of temperature for mixtures of the hexahydrate and tetrahydrate, the tetrahydrate and a dihydrate, and the dihydrate and a monohydrate. Simon and Knauer [39SIM/KNA] also identified the tetrahydrate as a probable intermediate product in the isobaric decomposition of the heptahydrate. The results from these studies are in rough agreement, but in the present review, no thermodynamic values are selected for these lower hydrates. V.5.1.2.2.1.3
NiSO4(cr)
Although Goldberg et al. [66GOL/RID] indicated that samples of NiSO4(cr) dissolved too slowly in water to allow determination of its heat of solution, Adami and King [65 AD A/KIN] reported the heat of solution in 4.360 m HCl(aq) at 303.15 K. The thermodynamic cycle involved comparison with the heat of dissolution of metallic nickel in a similar medium. The results of [65ADA/KIN] have been recalculated in the present review (Appendix A) to obtain the selected value: AfH°m (NiSO4, cr, 298.15 K) = - (873.28 + 1.57) kJmol '. Weller [65WEL] measured the heat capacity of anhydrous nickel sulphate between 52.0 and 296.3 K. Stuve et al. [78STU/FER] also measured the heat capacity by adiabatic calorimetry, but primarily at lower temperatures, between 9.7 and 69.6 K (a
194
V Discussion of data selection for nickel
sharp thermal transition was observed at 35.5 K), and used their results with those of Weller [65WEL] to determine: C; m (NiSO 4 , cr, 298.15 K) = (97.6 ±0.1) JIT'-mol 1 S°m (NiSO4, cr, 298.15 K) = (101.3 ± 0.2) J K '-mol"1. In the absence of other values, these values of 5°(NiSO 4 , 298.15 K), C° m (NiSO4, 298.15 K) are selected in the present review (with an estimated uncertainty of ± 0.2 J-KT'-mor1 for S°m). These selections yield: AfG° (NiSO4, cr, 298.15 K) = - (762.7 ± 1.6) kJ-mol"1. Stuve et al. [78STU/FER] also reported the results of a limited set of drop calorimetry experiments (402.9 to 1001.5 K). The authors fitted an equation to the experimental enthalpy differences such that the heat capacity values meshed smoothly with the value obtained from adiabatic calorimetry for 298.15 K. The authors indicated that their equation1 was applicable, within 0.6 per cent for temperatures between 298 and 1200 K. As discussed below, NiSO4 decomposes towards the upper end of this temperature range, and extrapolation of the heat capacities to 1200 K does not seem justified. Conversion of the equation from calorie to joule units leads to: [C;_m]^,K5K (NiSO4, cr) = (118.934 + 3.6761 x 10 2 (77K) - 2.8681 x 106 (T/Ky2) JKT'mol ' and this equation for C° m (7) from Stuve et al. [78STU/FER] is accepted in the present review. There have been several studies of the equilibrium conversion of NiSO4 to NiO from 880 K to 1200 K according to: NiSO4 ^ NiO + SO3 with possible concurrent decomposition of SO3: SO3 ^
SO2 + V2O2.
Both pressure measurements [25MAR], [78KAR/MAL] and electrochemical measurements [66ING], [64ALC/SUD], [73SKE/ESP], [77SAD/KAW], [83GAU/BAL] have been reported. The measurements lead to values of AfG° that agree within 5 - 6 kJmol ' near 900 K, and 7 - 8 kJmol 1 at 1150K. The values from Alcock et al. [64ALC/SUD] show better agreement with the vapour-pressure study of Karwan et al. [78KAR/MAL] than do the other electrochemical studies. In the present review, no attempt has been made to calculate heat-capacity values from the second derivatives of equations fitted to these measurements.
The rather similar equation for Hm (7) — //° (298 K), probably based on prepublication results from [78STU/FER], is incorrect as published in the review of Mah and Pankratz [76MAH/PAN].
V.5 Group 16 (chalcogen) compounds and complexes
195
V.5.2 Selenium compounds and complexes Nickel-selenium compounds have recently been reviewed by the NEA TDB review team on Se, (see [2005OLI/NOL] for the corresponding NEA TDB critical review).
V.5.3 Tellurium compounds and complexes V.5.3.1 Nickel tellurides The binary phase diagram nickel/tellurium as summarised by Lee and Nash [90LEE/NAS] shows a complex behaviour with six intermetallic solid phases. This diagram is essentially based on a seminal paper authored by Klepp and Komarek [72KLE/KOM]. The phases Pi, p 2 , P[ and NiTe2^(cr) have variable stoichiometry. In an early stage of this review it was decided that non-stoichiometric alloy systems will not be covered. Thus only the two y-phase line components, NiTeo.775(yi) and NiTeo.85(Y2) among the six solid phases, need to be discussed here. The thermodynamic data of the Ni - Te system have been critically assessed by Ball et at. [92BAL/DIC]. V.5.3.1.1
NiTe o . 775 ( Yl )
NiTeo.775(cr) is orthorhombic and has the following unit cell parameters: a — 3.916 A, b = 6.860 A, c = 12.32 A [72KLE/KOM]. The y,-phase appears to be stable up to 1015.5 K, when it decomposes peritectoidally into y2 and p 2 . There have not been any heat capacity measurements reported for y,. Ball et al. [92BAL/DIC] obtained estimates of C° m (NiTeo.775, cr, 298.15 K) = 45.8 (48.7) J K '-mol 1 and S°m(NiTeo.775, cr, 298.15 K) = 71.0 (67.6) J-K '-mol ' by taking a weighted average of heat capacities of the p and NiTe2 r(cr) phases. The values in brackets have been calculated with the cationic and anionic contributions to the heat capacity at 298 K and the revised 'Latimer' entropy contributions [93KUB/ALC]. If reasonable uncertainties of + 5.0 JK~'mol ' are chosen for AC° m and AS°^, both methods of estimation lead to results that are consistent within these limits. The enthalpy of formation of NiTeo.77s(cr) has been estimated from emf measurements in the y + NiTe 2x (cr) two-phase field [92BAL/DIC]. Neither of these standard entropy or standard enthalpy estimations (A f //° (NiTeo.775, cr, 298.15 K) = — (41.7 ± 4.4) kJ-mol ') is based directly on measurements of the NiTeo.77s(cr) phase, so no recommendation can be given in this review. This is not of any particular importance here as the yi phase is of peripheral interest only. V.5.3.1.2
NiTeo.85(y2)
Based on thermoanalytical studies Klepp and Komarek [72KLE/KOM] assigned somewhat arbitrarily a composition of NiTeo.85(cr) to the high temperature y2 phase. An attempt to characterise this phase by its X-ray powder diffraction pattern failed, because below 963 K it decomposes quickly into NiTeo.775(cr) + NiTe 2x (cr) and the phase equi-
196
V Discussion of data selection for nickel
librium could not be frozen in by quenching [72KLE/KOM]. Therefore, estimated thermodynamic quantities for the y2 phase are even less reliable than for the y, phase. According to Ball et al. [92BAL/DIC], the same values of the standard heat capacity and entropy were estimated for both y phases (based on x(Ni) + x(Te) - 1), resulting in C;m(NiTe0.85, cr, 298.15 K) = 47.7 (50.7) J-KT'-mor' and S°m (NiTeo.85, cr, 298.15 K) = 74.0 (70.7) J-Kr'-mor1. The values in brackets have been calculated with the cationic and anionic contributions to the heat capacity at 298 K and the revised 'Latimer' entropy contributions [93KUB/ALC]. The latter estimation method clearly leads also to identical heat capacity and entropy values for both y phases based on x(Ni) + x(Te) = 1. The enthalpy of formation of NiTeo.85(cr) has again been estimated from emf measurements in the y + NiTe 2x (cr) two-phase field [92BAL/DIC]. Without confirmation by experimental data the estimated standard entropy and standard enthalpy (A f //° (NiTeo.85, cr, 298.15 K) = -(50.1 + 5.6) kJ-mof1) of this phase cannot be recommended in this review.
V.6 Group 15 compounds and complexes V.6.1 Nitrogen compounds and complexes V.6.1.1 Solid nickel nitrates The hydrated nickel nitrate solids have been the subject of sporadic thermodynamic studies over the last 150 years, but the basic thermodynamic quantities for these materials are not well defined. Though careful synthesis yields solids with various Ni(NO3)2:H2O ratios, the regions of stability (if any) for the lower hydrates are still not well established [1899FUN], [34SIE/SCH], [98ELM/GAB], [2001MIK/MIG]. When hydrated nickel nitrate is heated at a pressure of approximately 0.1 MPa, the last water of hydration is lost between 450 and 500 K. Release of nitrogen oxides occurs at temperatures below 600 K, and continued heating leads to formation of basic nickel nitrates and then NiO [98ELM/GAB], [2001MIK/MIG]. V.6.1.1.1
Ni(NO3)2-9H2O(cr)
This solid has been reported as a stable solid in the Ni(NO3)2 - H2O system for temperatures below 270 K [34SIE/SCH]. The temperature for the equilibrium conversion of the nonahydrate to the hexahydrate in contact with saturated aqueous solution has not been well established. No values for Ni(NC>3)2-9H2O(cr) are selected in the present review. V.6.1.1.2
Ni(NO3)2-6H2O(cr)
The stable hydrate in equilibrium with a solution saturated in nickel nitrate at 298.15 K is Ni(NO3)2-6H2O(cr) [34SIE/SCH]. The water content of the commercially available solid tends to vary slightly. The solid can easily lose water on exposure to dry air [72AUF/CAR], [98ELM/GAB], but has also been reported to be slightly deliquescent in
197
V. 6 Group 15 compounds and complexes moist air [72AUF/CAR]. In a closed system, the hydrate begins to melt (or partially dissolve in its water of hydration) at 328 K [2001MIK/MIG]. Auffredic, Carel and Weigel [72AUF/CAR] measured the enthalpy of solution of the hexahydrate in water at 298.15 K. Calorimetric measurements were done for ratios of H2O:Ni(NO3)2-6H2O between 500 (0.11 m) and 10000 (0.0056 m), and a value of (31.23 + 1.00) kJmol ' for Asol//^ was reported. For the more dilute solutions, there is considerable scatter in the measurements as a function of molality, and the reported value of A sol //° was simply an extrapolated value from almost random values for concentrations between 0.01 m and 0.0056 m. As discussed below, and in the Appendix A discussion for [72AUF/CAR], the lack of concentration dependence over this concentration range is not unexpected. The average value of the enthalpies of solution for final concentrations below 0.01 kg-mol 1 is (31.28 + 0.87) kJmol 1 . A reasonable estimate of the integral heat of dilution of a nickel nitrate solution from a solution with a solvent:salt ratio of 8000:1 is -(0.52 ±0.10) k J m o l ' [59LAN] (see the Appendix A discussion of [72AUF/CAR]). In the present review, this estimated heat of dilution is applied to the average value of the enthalpies of solution for final concentrations below 0.01 kJ-mol ', and for Ni(NO3)2-6.00H2O(cr) ^ Ni2+ + 2 NO" + 6H2O(1)
(V.90)
A r //° ((V.90), 298.15 K) = (30.76 ± 1.00) kJmol ' is selected. Values based on measurements of the vapour pressure of water over the hexahydrate [72AUF/CAR] were not incorporated in the calculation of the assessed value because, as discussed below, the stoichiometry of the dehydrated salt in equilibrium with the hexahydrate and H2O(g) is not firmly established. This selection yields, using NEA-TDB auxiliary data: A f //° (Ni(NO3)2-6.00H2O, cr, 298.15 K) = - (2214.5 ± 1.6) kJmol"'. Goldberg et al. [79GOL/NUT], reviewed the available literature data, and provided tables of recommended values for the activity coefficients and osmotic coefficients for solutions of Ni(NO3)2 in water at 298.15 K. However, there were no studies that allowed the tables to be extended to saturated solutions of Ni(NO3)2-6.00H2O(cr) at 298.15 K (5.47 m [34SIE/SCH]), and no major studies seem to have been published since that review. No values of Asl),Gm at saturation are estimated here. Heat capacity measurements for Ni(NO3)26H2O(cr) were reported by Vasileff and Grayson-Smith [50VAS/GRA] for temperatures from 65 to 300 K. The authors estimated the entropy change between 65 and 273 K to be 364.4 J K '-mol 1 (87.1 cal-KT'-mol '). From the set of smoothed specific heats, the selected value of the molar heat capacity of the solid at 298.15 K is calculated to be C; m (Ni(NO3)2-6H2O, cr, 298.15 K) = (463.0 ± 2.0) J-K"'•mol"1,
198
V Discussion of data selection for nickel
where the uncertainty has been estimated in the present review. The temperature range of the measurements is not adequate for the calculation of a value for S° (Ni(NO3)2-6H2O, cr, 298.15 K). V.6.1.1.3
Ni(NO 3 ) 2 -4H 2 O(cr), Ni(NO 3 ) 2 -3H 2 O(cr), Ni(NO 3 ) 2 -2H 2 O(cr)
Dehydration of the hexahydrate leads to several lower hydrates, but the hydrate formed seems to depend markedly on the method used to carry out the dehydration. In few of the experiments were the dehydrated solids thoroughly characterised, nor was it established that the solids were stable over long periods. Chukurov et al. [73CHU/DRA] and Auffredic, Carel and Weigel [72AUF/CAR] have reported measurements of the heat of solution of a compound with the stoichiometry Ni(NO3)2-4H2O(cr) in water (8.49 k J m o l 1 and 8.66 k J m o l ' , respectively). The latter group prepared their solid by dehydration of the hexahydrate over calcium chloride at room temperature. They also reported [72AUF/CAR2] a value of the heat of solution (- 27.41 k J m o l ' ) of Ni(NO3)2-2H2O(cr) (prepared by dehydration of the tetrahydrate over calcium chloride at 50°C). Recalculation of the heats of dilution (cf. Appendix A for [72AUF/CAR]) gives heats of solution, corrected to "infinite dilution", for the dihydrate of -(27.64± 1.00) kJmol 1 and for the tetrahydrate (8.23 ± 1.00) kJmol ' (based on [72AUF/CAR]) and (8.49± 1.00) kJmol ' (based on [73CHU/DRA]). In contrast to Funk [1899FUN] who reported a stable trihydrate phase (55 to 95°C), Sieverts and Schreiner [34SIE/SCH] found that both the tetrahydrate (55 to 80°C) and dihydrate (85 to 120°C) were stable phases in the Ni(NO3)2 - H2O system, and could also be formed at equilibrium at 25 °C in the Ni(NO3)2 - H2O - HNO3 system at high nitric acid mole fractions. Studies of the dehydration of the hexahydrate using differential thermal analysis [74WEN], [98ELM/GAB], differential scanning calorimetry [99MIK/MIG] and thermogravimetric analysis [66KAL/PUR], [2001MIK/MIG] suggest that although a tetrahydrate can be obtained, it is easily converted to the trihydrate at slightly higher temperatures. For example, Kalinichenko and Purtov [66KAL/PUR] reported that dehydration of the tetrahydrate to the trihydrate occurred between 164°C and 170°C. Sano [37SAN] and Auffredic, Carel and Weigel [72AUF/CAR], reported measurements of the vapour pressure of water over Ni(NO3)2'6H2O(cr) as a function of temperature (Figure V-39). The measured vapour pressures differ systematically by a factor of 1.5 at similar temperatures. Sano interpreted his results in terms of the reaction: Ni(NO3)2-6.00H2O(cr) ^
Ni(NO3)2-3.00H2O(cr) + 3H2O(g)
(V.91)
whereas Auffredic, Carel and Weigel interpreted their results in terms of the reaction: Ni(NO3)2-6.00H2O(cr) ^
Ni(NO3)2-4.00H2O(cr) + 2H2O(g).
(V.92)
V.6 Group 15 compounds and complexes
199
As can be seen from the figure, the slope of the plot of- logi0 pHi0 against \IT is similar for both sets of results. This may indicate that whichever lower hydrate was involved in the equilibrium, it was the same solid for both sets of experiments. If the vaporisation corresponds to Reaction (V.92), the calculated enthalpy of vaporisation (110 kJ-mol') corresponds well with 116 kJ-mol"1, derived from the enthalpies of solution of the hexahydrate [72AUF/CAR] and the tetrahydrate [72AUF/CAR], [73CHU/DRA]. This suggests that the vaporisation follows Reaction (V.92), rather than (V.91). Similarly, the enthalpy for Reaction (V.93), Ni(NO3)2-4.00H2O(cr) ^
Ni(NO3)2-2.00H2O(cr) + 2H2O(g)
(V.93)
A r //°= 124 kJmol 1 , as derived from the reported calorimetric data [72AUF/CAR], [72AUF/CAR2], is similar to the value obtained from measurements of the vapour pressure of water over the tetrahydrate (122 kJ-mol"') [72AUF/CAR2], Nevertheless no value is selected in the present review for the Gibbs energy of Reaction (V.92) or (V.93) because there was no proper characterisation of the lower hydrate at equilibrium during any of these measurements.
Figure V-39: Measured water vapour pressures over hydrated Ni(NO3)2-6H2O(cr) [37SAN], [72AUF/CAR] and Ni(NO3)2-4H2O(cr) [72AUF/CAR2].
200
V Discussion of data selection for nickel
In the present review, the average calorimetry-based value is selected for the enthalpy of dissolution of the tetrahydrate: Ni(NO3)2-4.00H2O(cr) ^
Ni2+ + 2 NO3 + 4H2O(1)
(V.94)
1
A r //° ((V.94), 298.15 K) = (8.36 ± 0.80) kJmot . The selected uncertainty is slightly larger than the statistical uncertainty because the uncertainty in the heat of dilution is common to both results. The corresponding value for the dihydrate [72AUF/CAR2] has also been selected: Ni(NO3)2-2.00H2O(cr) ^
Ni2+ +2 NO3" + 2H2O(1)
(V.95)
A r / / ; ((V.95), 298.15 K) = - (27.64 + 1.00) kJmol '. In neither case should the selection of a value for the enthalpy of solution be taken to mean that it is certain that the solid is a stable phase in the Ni(NC>3)2-H2O system. These selections yield, using NEA-TDB auxiliary data: A f //° (Ni(NO3)2-4.00H2O, cr, 298.15 K) = - (1620.4 + 1.4) kJmol ' and A f #° (Ni(NO3)2-2.00H2O, cr, 298.15 K) = - (1012.7 ± 1.6) kJ-mol 1 . V.6.1.1.4
Ni(N03)2(anhydrous)
Except for salts containing singly charged cations, synthesis of anhydrous nitrate salts is very difficult, as the final steps of dehydration tend to cause loss of nitrogen oxides or nitric acid. Although some authors [82MU/PER], [2001MIK/MIG] claim anhydrous nickel nitrate forms as an intermediate decomposition product during heating of the hydrate, only Chukurov et al. [73CHU/DRA] claim to have carried out any chemical thermodynamic measurements (A f //° = - (402.1 ± 2.9) ld-mol"1, from determination of the heat of solution in dimethyl sulphoxide). The summary of their paper, all that was available to the reviewers, contains no details on the preparation, handling or analysis of the anhydrous salt. No values for the thermodynamic quantities for anhydrous Ni(NO3)2 are selected in the present review.
V.6.1.2
Aqueous Ni(II)-nitrato complexes
There are conflicting opinions concerning the interaction of Ni(II) with nitrate ion. Mironov et al. have failed to detect nitrato complexes of Ni(II) [70MIR/MAK]. On the other hand, neutron diffraction of 2 and 4.3 m Ni(NO3)2 solutions revealed ion pair formation between Ni2+ and nitrate ions [97HOW/NEI]. Quantitative information on nitrate complexation (ion pair formation) of Ni(II) has been published in [73FED/SHM] and [73HUT/HIG]. The Russian authors reported formation of mono-, bis- and triscomplexes [73FED/SHM], while only NiNO^ was reported in [73HUT/HIG]. In light of the results collected for similar weak complexes of Ni(II) (e.g., for halogeno complexes), the formation of higher complexes (Ni(NO 3 ) 2 ", n > 1) is unlikely, except in
201
V. 6 Group 15 compounds and complexes highly concentrated nitrate solutions. Therefore, the formation of such complexes is not considered in this review. The available association constant values for Reaction (V.96): Ni2+ + NO" ^
NiNO3+
(V.96)
are reported in Table V-27. The experimental data reported in [73FED/SHM] were re-evaluated taking into account only formation of the NiNO, species (see Appendix A). To minimise the medium effect, only half of the experimental data, for which [ NO, ] < [ ClO^ ], were taken into account. The recalculated constants are also listed in Table V-27. Table V-27: Formation constants (logarithmic values) of the NiNOj complex (ion pair). t(°C)
Method Medium kin sol
1 M Na(ClO4) NO3)
iog m A
a (b)
(a)
!og.0 H\
log,0^ NO3)
-(0.5510.09)
-(0.30 ± 0.40) 10) for temperatures from 290 to 560 K. As discussed in Appendix A and elsewhere [95LEM], it is difficult to interpret these solubility results quantitatively because of the limited range in hydroxide concentration and the wide variation in the ionic strength. The values for the formation constants of NiHPO4(aq) from the pH studies [67SIG/BEC], [96SAH/SAH] and for the species, Ni(OH)2HPO4~, as proposed by Ziemniak, Jones and Combs [89ZIE/JON], [95LEM] are not inconsistent for the pH values at which the experimental measurements were obtained.
Table V-28: Experimental values for the association constants and other thermodynamic quantities for nickel phosphate complexes. Method
/ (°C)
Medium (aq)
Reported value
Recalculated value (a)
Reference
log,,,*" ligancI:HPO; Ni2t + HPO; ^NiHPO 4 (aq) pot pot pot pol pot
25.0 15.0 25.0 25.0 25.0
0.1 MNaC10 4
log10 *, = 2.08
(2.97 ± 0.30)
[67SIG/BEC]
O.IMKNO3
logl0 *, = 2.00
(2.88 + 0.30)
[72FRE/STU]
0.1 MNaC10 4
log,,,*, =2.11
(3.00 ± 0.20)
[76TAY/DIE]
0.2 M NaC104
log 10 *, =3.26
0.1 MNaNO 3
log, 0 *,=2.20
P-Ni(OH)2 (cr)+ HPO; sol
25.0
^
0.1 MNaC10 4
[91SWI/PAW] (3.07 + 0.10)
[96SAH/SAH]
I<Ji(OH)2HPO^ logl0*=-(5.0+1.0)
[89ZIE/JON]
log,,,*, =0.5
[76TAY/DIE]
ligand: H,PO4 Ni 2+ + H21?O4 ^ pot
25.0
H,NiPO4
0.1 MLi/NaC104
(a) Uncertainties are the statistical uncertainties from the re-analysis of the data. The overall estimated uncertainties are discussed in the text.
V. 6 Group 15 compounds and complexes
V.6.2.2.2
207
Aqueous nickel diphosphato complexes
As in the NEA-TDB uranium review [92GRE/FUG], the only polyphosphate(V) species considered in the present review are the pyrophosphates (diphosphato complexes). Other polyphosphoric acid species have negligible equilibrium concentrations at total phosphate concentrations < 0.045 moldm 3 and at temperatures below 200°C [74MES/BAE]. There have been potentiometric (pH measurement) [56YAT/VAS], [64HAM/MOR], [73PER/SEC], calorimetric [56YAT/VAS2], spectrophotometric [58VAI/RAM] and amperometric [78KAR/HUB] studies of the formation of pyrophosphate complexes of Ni(II) (Table V-29). The complexes of the highly charged pyrophosphate ion with nickel are generally stronger than the phosphate complexes, but interpretation of the experiments is beset by the same difficulties as the interpretation of the phosphate studies with respect to unambiguous identification of the species. In the literature, results of studies of association of Ni2+ with pyrophosphate have been discussed in terms of the following equilibria: Ni2+ + P2O*~ ^ Ni2+ + HP2O'~ ^
NiP2O,NiHP2O;
Ni2+ + 2 P2O*- ;F± Ni(P2O7)2~
(V.98) (V.99) (V.I00)
The results of Vaid and Rama Char [58VAI/RAM] were obtained under illdefined conditions, and are in poor agreement with results from the other studies. The equilibrium constant of Yatsimirskii and Vasilev for Reaction (V.98) is consistent with other values, but was obtained at moderately high and variable ionic strength. Values from the measurements of Hammes and Morrell [64HAM/MOR] in 0.1 M Me4NCl(aq) for Reactions (V.98) and (V.99), after correction to / = 0, are logl0 K° (V.98) = (8.73 ± 0.25) and log,0 K° (V.99) = (5.14 + 0.25), at 298.15 K. Perlmutter-Hayman and Secco [73PER/SEC] did careful measurements of nickel pyrophosphate complexation at four different ionic strengths (KNO3(aq)), though for reasons discussed in Appendix A, only their values obtained in 0.1 M and 0.2 M KNO3(aq) are considered to be consistent. As discussed in Appendix A, the authors neglected association of pyrophosphate with K+, though the values they used for pyrophosphate protonation were determined in the corresponding aqueous KNO3 solutions. The apparent 298.15 K values (corrected to / = 0) of log10 K (V.98) ((8.73 + 0.25) [64HAM/MOR], (7.69 ± 0.25) and (7.78 ± 0.30) [73PER/SEC]) are in poor agreement; the value of Frey and Stuehr [72FRE/STU] for 288.15 K (7.94 ± 0.25) is in only marginal agreement with that of Perlmutter-Hayman and Secco (7.53 ± 0.25), and the value at 303.15 K of Karwelk and Huber [78KAR/HUB] is lower than would be expected from the results of Perlmutter-Hayman and Secco. The disagreement is primarily the result of neglect of complexation of the singly-charged cations, K+ and NH*, with pyrophosphate. As indicated in Appendix A, use of reasonable potassium
208
V Discussion of data selection for nickel
pyrophosphate complexation constant values brings the results of Perlmutter-Hayman and Secco into agreement with those of Hammes and Morrell. The raw data of Perlmutter-Hayman and Secco [73PER/SEC] are unavailable for re-analysis, and in this review, the selected value for log10 K° (V.98) is the value based on the work of Hammes and Morrell [64HAM/MOR] at 298.15 K: logl0 K° ((V.98), 298.15 K) = (8.73 + 0.25). Hence, ArG° ((V.98), 298.15 K) = - (49.83 ± 1.43) kJ-mol '. The apparent formation constants from Perlmutter-Hayman and Secco [73PER/SEC] and Frey and Stuehr [72FRE/STU] for Reaction (V.99) at 288.15 K agree well within the uncertainty bounds. After extrapolation to / = 0, the formation constants from Perlmutter-Hayman and Secco ((5.02 + 0.25) and (5.02 ± 0.30), based on their measurements in 0.1 and 0.2 M KNO3(aq), respectively) and from Hammes and Morrell (5.14 ± 0.25) also seem to be in good agreement. Unfortunately, the neglect of K+ association with pyrophosphate [73PER/SEC] (see Appendix A) and the unavailability of the raw data of Perlmutter-Hayman and Secco [73PER/SEC] preclude a proper reanalysis of the data from that study. The value of logl0 K (V.99) of Hammes and Morrell [64HAM/MOR] is selected in the present review for 298.15 K: logl0 JT ((V.99), 298.15 K) = (5.14 + 0.25). Hence, ArG° ((V.99), 298.15 K) = - (29.34 ± 1.43) kJ-mol '. There appears to be only one reported value [56YAT/VAS] for the cumulative second association constant (Reaction (V.I00)), J32 = (1.54 + 0.50) x 107. As discussed in Appendix A, this value was calculated from results obtained for solutions of ionic strength that ranged from 0.2 to 1.0 M. This value suggests that significant quantities of Ni(P2O7)j" should form only in solutions of moderate to high ionic strength where formation of NiP 2 O^ is less strongly favoured. Furthermore, the effect of association of Na+ with pyrophosphate [94STE/FOT] on the reported value has not been established. No value for the second association constant is selected in the present review. The values for the enthalpies of Reactions (V.98) and (V.99) from the formation constant measurements for 0.1 M KNO3 solutions [73PER/SEC] (A r //° ((V.98), (278 - 308) K) = (30.6 ± 7.4) kJ-mol""', (ArH°m ((V.99), (278 - 308) K) = (47.9 ± 10.0) kJ-mof1) are accepted to be the same as the values at / = 0 without correction for either ionic strength or association of K+ with pyrophosphate. The uncertainties are estimates, A r //° ((V.98), 298.15 K) = (30.6 + 10.0) kJ-mol ' Ar//n°((V.99), 298.15 K) = (47.9 + 15.0) kJ-moP1.
209
V.6 Group 15 compounds and complexes
The value for Reaction (V.98) is not inconsistent with the rough calorimetric value reported by Yatsimirskii and Vasilev (see discussion of [56YAT/VAS] in Appendix A). No value is selected for the enthalpy of Reaction (V.I 00). It must be emphasised that the values selected in this review for formation of NiHP2O, and NiP 2 O^ should not be used for solutions more than 0.01 M in alkali metal ions unless explicit values are introduced for the pyrophosphate-alkali metal ion association constants (the values of De Stefano et al. [94STE/FOT] may be useful, cf. Appendix A for that reference). The corresponding Gibbs energy of formation are: A f G°(NiP 2 O^,298.15K) = -(2031.1 ±4.8)kJ-mol ' A f G°(NiHP 2 O;, 298.15 K) = -(2064.3 ±4.8) kJmol '
Table V-29: Experimental values for the association constants and other thermodynamic quantities for nickel pyrophosphate complexes. Method
l/°C
Medium (aq)
Reported value
FLecalculated value "'•b) log,,, K"
Reference
ligand: P2O< (pj'rophosphate) Ni2( + n P,O? sol
25.0
var
Ni(P,O 7 ); 4 " log l0 AT, = 5.82
[56YAT/VAS]
'°g|0 g,o A"s0 (Ni2P2O7) = - 12.77 A,//, = (17.61 ±0.17)kJ-mor' lo
cal
25.0
var
[56YAT/VAS2]
gio K2= 1.60 A,H. , =-(9.25 +0.25) kJmol"1 lo
sp
25.0
var
pot
25.0
0.1 MMe 4 NCl
l°g,o K, = 3.62 •°g|0 AT, = 6.98
(8.73 ±0.25)
[64HAM/MOR]
pot
15.0
0.1 MKN0 3
togio K, = 6.22
(7.94 ± 0.25)
[72FRE/STU]
pot
5.0
0.1 MKNO3
lo
15.0
0.1 MKNO3
25.0
0.02 M KNO3
log,,, K, =5.81 !°g,o AT, =6.39
25.0
0.05 M KNO3
lo
25.0
0.1 MKNO3
'og,o K, = 5.94
(7.69 + 0.25)
25.0
0.2 M KNO3
logic K, =5.60
(7.78 ±0.30)
35.0
0.1 MKNO3
lo
0.1 MKNO3 amp
[58VAI/RAM]
30.0
0.1 MNH4NO3
g|0 AT, = 5.63
[73 PER/SEC] (7.53 ±0.25)
gio K, = 6.04
gio A:, =6.21
Ar Hm((278-308)K) >°gio AT, =5.35
= (30.6 ± 7.4) kJ-mor1 (7.11 ±0.30)
[78KAR/HUB]
(Continued on next page)
V Discussion of data selection for nickel
210
Table V-29 (continued) Method ligand:
,rc HP
Medium (aq)
Reported value
Recalculated value (a' b)
Reference
2°7 (hydrogen pyrophosphate) Ni2+ + HP2O] ^± HNiP,O7
pot
25.0
0.1 MMe 4 NCl
log l 0 *, =3.83
(5.14 + 0.25)
[64HAM/MOR]
pot
15.0
0.1 MKN0 3
log,,,*, =3.50
(4.79 + 0.25)
[72FRE/STU]
pot
5.0
0.1 MKNO3
log, 0 *, =3.15
15.0
0.1 MKNO3
log,,,*, =3.55
25.0
0.02 M KNO3
log, 0 *, =4.01
25.0
0.05 M KNO3
log 10 *, =3.78
25.0
0.1 MKNO3
log, 0 *,=3.71
(5.02 ± 0.25)
25.0
0.2 M KNO3
log, 0 *, =3.39
(5.02 ± 0.30)
0.1 MKNO3
log l 0 *,=4.07
0.1 MKNO3
A r // m ((278-308)K) = (47.9 ± 10.0)kJ-mor'
35.0
[73PER/SEC] (4.84 ±0.25)
(a) Uncertainties are the estimated uncertainties as discussed in the text. (b) Values from most studies were re-calculated neglecting explicit association of K or NH4 with pyrophosphate, and cannot be averaged directly with the values obtained in solutions of tetraalkylammonium salts.
V.6.3 Arsenic compounds V.6.3.1 Nickel arsenides The nickel-arsenic phase diagram [61YUN], [87SIN/NAS] is moderately complex, with the compounds Ni5As2(cr), Ni n As 8 (cr) (nickel-deficient Ni3As2(cr)), NiAs(cr) and NiAs2(cr) (with a (3 to a conversion near 870 K). V.6.3.1.1
NiAs(cr)
There are at least two sources available for the enthalpy of formation of NiAs(cr). Predel and Ruge [72PRE/RUG] reported - 36.0 kJg-at ' (- 72.0 kJ-mol ') for the formation enthalpy based on the heat of dissolution in liquid tin at an unstated temperature. As discussed in Appendix A, a third law analysis results in the value A f //° (NiAs, cr, 298.15 K) = -(73.60 ±8.00) kJ-mol '. Skeaff et al. [85SKE/MAI] used an electrochemical technique to determine the Gibbs energies of reaction of NiAs(cr) with oxygen to form NiO (bunsenite) and As4O6(g) at temperatures from 711 to 833 K. As discussed in Appendix A, a third law analysis with the heat capacity equation of Muldagalieva et al. [95MUL/CHU], and auxiliary data consistent with those used in the present review, leads to AfH^ (NiAs, cr, 298.15 K) = - (69.73 + 5.00) kJ-moP1. The values are consistent within the lower uncertainties. The selected value of A r //° is the weighted average of the two experimental values, and the selected value of 5°, is the recalculated value from Skeaff et al. [85SKE/MAI]:
V. 6 Group 15 compounds and complexes
211
A f //° (NiAs, cr, 298.15 K) = - (70.82 ± 4.24) kJ-moP1 S°m (NiAs, cr, 298.15 K) = (50.76 + 6.00) J-IC'mol '. The heat capacity equation of Muldagalieva et al. [95MUL/CHU] is selected in the present review, [C;nJ82°9°^5K(NiAs, cr) = (36.54 + 41.87 x 10"3 (T/K)+ 1.58xl05 (T/Ky2) J-K '-mol ' as is the value of C; m (NiAs, cr, 298.15 K) = (50.8 ± 4.0) J-K 'mol '. The heat capacity equation was claimed by the authors to be applicable over the temperature range 298.15 to 675 K, and in the present review it is accepted over a slightly expanded range to 800 K to accommodate the recalculation of the equilibrium data of Skeaff et al. [85SKE/MAI]. Nozue et al. [97NOZ/KOB] measured the specific heat of NiAs(cr) for temperatures between 1.7 K and 30 K, but data linking these values to those near 300 K are lacking. The above selections yield: AfG° (NiAs, cr, 298.15 K) = - (66.6 + 4.6) kJ-mol '. V.6.3.1.2
Ni,,As8(cr)
The heat capacity equation of Muldagalieva et al. [93MUL/ISA] is selected in the present review, [C; m ] 8 2 ^ 5K (Ni n As 8 , cr) = 380.12 + 367.27x10 3 (77K) + O.O18xlO5 (77K)2 J-K '-mol ' as is the value of: C; m (Ni u As 8 , cr, 298.15 K) = (490 + 40) J-K '-mol '. The heat capacity equation was claimed by the authors to be applicable over the temperature range 298.15 to 700 K, and in the present review it is accepted over a slightly expanded range to 800 K to accommodate the recalculation of the equilibrium data of Skeaff et al. [85SKE/MAI]. Skeaff et al. [85SKE/MAI] used an electrochemical technique to determine the Gibbs energies of reaction of Ni,|As8(cr) with oxygen to form NiO (bunsenite) and As4O6(g) at temperatures from 805 to 991 K. As discussed in Appendix A, a third law analysis with the heat capacity equation of Muldagalieva et al. [93MUL/ISA] and auxiliary data consistent with those used in the present review leads to the selected values: A f //° (Ni n As 8 , cr, 298.15 K) = - (743.0 ± 35.0) kJ-mol ' S^NinAsg, cr, 298.15 K) = (518 + 60) J-K '-mol '. The above selections yield: AfG^( Ni n As 8 , cr, 298.15 K) = - (715.8 + 39.3) kJ-mol '.
212
V Discussion of data selection for nickel
V.6.3.1.3
Ni5As2(cr)
Skeaff et al. [85SKE/MAI] also determined the Gibbs energies of reaction of Ni5As2(cr) with oxygen to form NiO (bunsenite) and As4C>6(g) at temperatures from 937 to 1139 K. No experimental heat capacity values appear to have been reported for Ni5As2(cr). Based on the approximation ATCpm = 0 used by the original authors, and auxiliary data from the present review, A f //° (Ni5As2, cr, 298.15 K) = - (244.6 ± 20.0) kJ-mol ', S° (Ni5As2, cr, 298.15 K) = (196.2 ± 30.0) J-K ' m o l 1 . Stolyrova [81STO] has reported a much less negative enthalpy of formation, - (204.0 ± 5.8) kJ-moP1, for the same solid, and Skeaff et al. [85SKE/MAI] noted that her reported value is inconsistent with the observed stability of Ni5As2(cr) solid with respect other phases [87SIN/NAS]. For that reason, the values of Skeaff et al. are selected in the present review: A f //° (Ni5As2, cr, 298.15 K) = - (244.6 ± 20.0) kJ-mol ' S°m (Ni5As2, cr, 298.15 K) = (196.2 ± 30.0) J-K 'mol '. The above selections yield: AfG° (Ni5As2, cr, 298.15 K) = - (237.6 ±21.9) kJ-mol"1. V.6.3.1.4
NiAs2(cr)
An experimental value for A f //° of- (90.06 ± 2.30) kJ-mol ' ( l a uncertainty) has been reported by Stolyrova [82STO] for this compound. Skeaff et al. [85SKE/MAI] were unable to establish a proper equilibrium with this compound in their apparatus, although some experiments suggested approximately the same value. In the present review, the value is selected, but with an increased uncertainty: A f //° (NiAs2, cr, 298.15 K) = - (90.1 ± 8.0) kJ-mol '.
V.6.3.2 V.6.3.2.1
Nickel arsenates Solid nickel arsenates
Several anhydrous [58TAY/HEY] and hydrated [66GME] nickel arsenates have been identified, but quantitative chemical thermodynamic data for these solids are sparse. Chukhlantsev [56CHU3] published values for the solubility of Ni3(AsO4)2-8H2O (annabergite) and its apparent solubility product [56CHU4] {K = (3.1 + 2.4)* 10 26, at 20°C for: Ni3(AsO4)2-8H2O(cr) ^
3Ni2+ + 2 AsO34" + 8 H2O(1).
(V.101)
Nishimura et al. [88NIS/ITO] measured the solubility at 25°C of what was probably the same solid, based on measurements over a wider range of pH (4 to 8), and reported log10 K (V.I02) = 9.9 (log10 K (V. 101) = - 26.76) for: Ni3(AsO4)2-8H2O(cr) + 4H+ ^
3Ni2+ + 2H2AsO~ + 8H2O(1).
(V.I02)
213
V. 6 Group 15 compounds and complexes
Langmuir et al. [99LAN/MAH] re-analysed these solubility data using a model that included formation of the complex NiHAsO4(aq), and reported log10^T (V.101) = - 28.38. The experiments of Nishimura et al. [88NIS/ITO] which involved equilibration for periods between one week and four months, indicate greater stability for the solid than was determined by Chukhlantsev [56CHU4]. The value of log10 AT°(V.1O1) selected in the present review: log 10 *°(V.101) = -(28.1±0.5) is based on the work of Nishimura et al. [88NIS/ITO] (see Appendix A), the auxiliary data for the present review, and the effect of formation of NiHAsO4(aq) as estimated by Langmuir et al. [99LAN/MAH]. From this, ArGm (V.101) can be calculated and AfG° (Ni3(AsO4)2-8H2O, cr, 298.15 K) = - (3491.6 ± 8.8) kJ-mol '. Omarova and Sharipov measured the heats of solution of NiSO4-7H2O, Na3AsO4, Na2SO4, H2O and Ni3(AsO4)2-8H2O in aqueous HC1, and reported A(H°m(Ni3(AsO4)2-8H2O, s, 298.15 K) = -(4179.0 + 9.2) kJmol ' [80OMA/SHA]. Artamonova and Kasenov [89ART/KAS] reported A f //° (Ni3(AsO4)2-8H2O, s, 298.15 K) = -(4065.4 ± 21.4) kJ-mol ', based on the heat of reaction of Na3As04(cr) with an aqueous nickel chloride solution to precipitate hydrated Ni3(As04)2. In the latter experiment, the less negative value found for the enthalpy of formation suggests that the initially precipitated compound might not have been well crystallised, and that the value of Omarova and Sharipov [80OMA/SHA] is to be preferred. In the present review, in the absence of information concerning the auxiliary data used by the authors, the value of Omarova and Sharipov [80OMA7SHA] is selected with an estimated uncertainty of 20kJmol ': A f //° (Ni3(AsO4)2-8H2O, 298.15 K) = - (4179.0 + 20.0) kJ-mol ', which leads to: 5° (Ni3(AsO4)2-8H2O, 298.15 K) = (540.8 ± 73.3) J K "'•mol"1. V.6.3.2.2
Aqueous nickel arsenato complexes
Langmuir et al. [99LAN/MAH] reported an estimated value of log10 K = 2.90 at 298.15 K for Ni 2+ + HAsO*~ ^
NiHAsO4(aq)
(V.I03)
based on work described in a thesis by Whiting [92WHI]. The value is similar to the value (3.05 + 0.09) for the corresponding phosphate complex (cf. Section V.6.2.2.1), and is an acceptable analogue value. In the present review, the value from Langmuir et al. [99LAN/MAH] is selected, but because of the
214
V Discussion of data selection for nickel
unavailability of experimental values for comparison, an uncertainty of ± 0.3 is assigned, log10 AT" ((V. 103), 298.15 K) = (2.9 + 0.3). This selection implies: Af G° (NiHAsO 4 , aq, 298.15 K) = - (776.9 ± 4.4) kJ-mol"1.
V.6.3.3
Nickel arsenites
V.6.3.3.1
Solid nickel arsenites
Several nickel arsenites [66GME] have been identified, but there are very limited chemical thermodynamic data for these solids. Chukhlantsev [57CHU] dissolved samples of nickel orthoarsenite in dilute nitric acid solutions at 20°C over 12 hours to form solutions with aqueous nickel concentrations of 8.7 x 10 3 mol-dm 3 at a "pH" value of 6.75 and 3.1x10 3 at a "pH" value of 7.10 mol-dm 3. The author did not report a solubility product based on their measurements but, as discussed in Appendix A, if the dissolution of the solid is assumed to correspond to the reaction: Ni3(AsO3)2-xH2O(cr, hydr.) + 6H+ ^
3Ni2+ + 2HAsO2(aq) + (2 + x)H2O(l) (V. 104)
the selected equilibrium constant can be calculated: log10 /C°((V.1O4), 298.15 K) = (28.7 ± 0.7). From this, ArGm (V.I 04) can be calculated and, if the water of hydration is not explicitly included in the chemical formula, AfG° (Ni3(As03)2, cr, hydr, 298.15 K) = - (1253.6 ± 9.3) kJ-mol '.
V.7 Group 14 compounds and complexes V.7.1 Carbon compounds and complexes V.7.1.1 Nickel carbonates V.7.1.1.1
Solid nickel carbonates
V.7.1.1.1.1
NiCO3(cr)
V.7.1.1.1.1.1 V.7.1.1.1.1.1.1
Crystallography and mineralogy of nickel carbonate NiCOi(cr), gaspeite
The crystal structure of NiCO3(cr) was definitively determined by Isaacs [63ISA]. It is of the calcite type, hexagonal, space group: R3 c, Z = 6, with unit cell dimensions aQ = 4.609 A, c0 = 14.737 A (according to JCPDS-ICDD card No. 12-771).
V. 7 Group 14 compounds and complexes
215
The natural occurrence of a new mineral (Ni0.49Mgo.43Feo.o8)C03 as a vein enclosed in siliceous dolomite in the Gaspe Peninsula, Quebec, was reported by Kohls and Rodda [66KOH/ROD]. The name gaspeite was selected for the nickel carbonate end member (approved by IMA 1966). The vein consists of essentially pure magnesian gaspeite with minor amounts of annabergite (Ni3(AsO4)2-8H2O), magnesite (MgCC^) and dolomite (CaMg(CO3)2). Enclosed in the magnesian gaspeite are a small amount of serpentine ((Mg, Fe)3Si2O5(OH)4) and a (Cr, Al) spinel. Well-formed crystals of millerite (NiS), niccolite (NiAs), annabergite, gersdorffite (NiAsS) and magnesite are found outside the vein in the buff siliceous dolomite. The X-ray powder diffraction patterns of magnesian gaspeite differ slightly from synthetic NiCO3(cr), see JCPDS-ICDD card No. 23-437. Gaspeite from the Gaspe Peninsula is light green, its luster is vitreous to dull and it has a yellow-green streak. The Mohs hardness is 4.5 to 5, the specific gravity is p(obs.) = (3.71 ±0.01) and /Xcalc.) = 3.748 g-cm 3. V. 7.1.1.1.1.1.2
Ni2(CO3)(OH)2, nullaginite
Nullaginite occurs as a secondary mineral in serpentinised peridotites in the Nullagine district (Western Australia). Mineral and name were approved by the IMA Commission in 1978. It is bright green with a luster varying from dull to silky. Composition and Xray powder diffraction pattern of nullaginite were investigated by Nickel and Berry [81NIC/BER]. Its crystal system is monoclinic, space group: P2)/m, cell dimension: a0 = 9.236 A, b0 = 12.001 A, c0 = 3.091 A, Z = 4, p= 90.48°, Vccn = 342.60 A3, p(obs.) = 3.56 and p(ca\c.) = 4.099 g e m 3 (according to JCPDS-ICDD card No 35-501). The large discrepancy between p(ca\c.) and p(obs.) is attributed to the presence of admixed water in the specimen. The ideal composition of Nullaginite is Ni2(CO3)(OH)2, but it occurs also together with Cr, Mg, Fe, and Cu and belongs to the rosasite group. This group is composed of minerals with either a monoclinic or triclinic symmetry and a general formula of M2(CO3)(OH)2. The M ion can be Co2+, Cu2+, Zn2+, Mg2+ and/or Ni2+. Minerals of this group composed of colour stimulating metals such as cobalt, copper and nickel are extremely colourful, usually green to blue. Most of these minerals range from rare to extremely rare. No synthesis of nullaginite has been reported and no thermodynamic data have been determined so far. V.7.1.1.1.1.1.3
Ni2(CO3)(OH)2-H2O, otwayite
The crystal structure of otwayite was investigated by Nickel et al. [77NIC/ROB]. Its fibre-rotation and X-ray powder diffraction patterns can be indexed on an orthorhombic unit cell with a = 10.18 A, b = 21A A, c = 3.22 A, Vce]l = 898.16 A3, Z = 8, p(obs.) = 3.41 and p(ca\c.) = 3.346 g-cm 3 (according to JCPDS-ICDD card No. 29-868). Otwayite occurs in Heazlewood (Tasmania), in the Nullagine region, but also together with Widgiemoolthalite in Widgiemooltha (both in Western Australia). Mineral and formula were approved by the IMA in 1977. Otwayite was named after Charles Albert Otway (1922 - ) , miner and prospector of Cosnells in Australia [77NIC/ROB].
216
V Discussion of data selection for n ickel
No synthesis of otwayite has been reported and no thermodynamic data have been determined so far. V.7.1.1.1.1.1.4
Ni3(CO3)(OH)4-4H2O, zaratite
Isaacs studied the structure of zaratite [63ISA]. The cell dimensions are: a = 6.16 A, Z = 1; f^ceii = 233.74 A3, p(calc.) = 2.31 g-cm 3, crystal system: isometric, space group: unknown (in part amorphous). Zaratite occurs in Cape Ortegal, Galicia, (Spain); in Broken Hill, New South Wales, (Australia); and in Pennsylvania, (USA). The mineral was named after Antonio Gil y Zarate (1793 - 1861). Data for physical and chemical properties vary significantly [63ISA]. The average chemical composition is close to Ni3(CO3)(OH)4-4H2O, however, analytical results scatter in the following range: NiO 56.9 - 61.2%, CO2 13.5 - 15.7%, H2O 23.2 - 27.1%. Zaratite is rare, forming emerald green secondary crusts on other nickel minerals. Its hardness is 3.5. From X-ray investigation of artificially synthesised as well as natural specimens it was concluded that zaratite is not a single mineral and should not be called one. Consequently no thermodynamic data can be ascribed to this poorly defined mixture of nickel hydroxide carbonates. V.7.1.1.1.1.1.5
Ni5(CO3)4(OH)f4~5H2O, widgiemoolthalite
The crystal structure of widgiemoolthalite was investigated by Nickel et al. [93NIC/ROB]. It has an ideal composition of Ni5(CO3)4(OH)2-4-5H2O. The monoclinic unit cell is similar to that of hydromagnesite, space group: P2i/c. The cell dimensions are: a = 10.06 A, b = 8.75 A, c = 8.32 A, Z = 2; jj= 114.3°, VccU = 667.48 A3, p(obs.) = 3.13 and p(calc.) = 2.97 g-cm 3 (according to JCPDS-ICDD card No. 46-1398). Widgiemoolthalite is found together with a number of other secondary nickel minerals in the weathered zone in the Widgie 132N mine in Widgiemooltha [93NIC/ROB]. It is bluish green and occurs mainly as a fibre. The mineral was discovered in 1992, named after its location and approved by IMA 1993. In reality the composition of widgiemoolthalite is close to (Ni, Mg)5(CO3)4.i5(OH)i.7-5.12 H2O. No thermodynamic data of this mineral are known. V.7.1.1.1.1.2
Heat capacity and entropy of NiCO3(cr)
A high temperature heat capacity function claimed to be valid at 298 < (77K) < 700 for NiCO3(cr) is given in thermodynamic data compilations: C°pm (NiCO3, cr) = (a + b (77K) + e (77K)2) J K ' -mol ' [C; n ,] 7 2 ^(NiCO 3 , cr) = (88.701 + 38.91x10
3
(77K) - 12.34xlO5(r/K)"2) J-K '-mol 1 (V.I 05) 3
Originally the following coefficients were given: a = 92.05, b = 38.91x10 and e = -12.34xl0 5 [77BAR/KNA]. In Knacke et al. [91KNA/KUB] and Kubaschewski et al. [93KUB/ALC] the coefficients b and e have been retained, but a
217
V. 7 Group 14 compounds and complexes has been slightly modified (a = 88.701). The only source of original C° m measurements up to 300 K found so far is [64KOS/KAL]. The latter authors estimated the maximum error of the standard entropy to be ± 0.4 J K '-mol 1 , however, they presented their results only graphically: S° (NiCO3, cr, 298.15 K) = (85.4 ± 2.0) J-K '-mol '. Robie and Hemingway assigned an uncertainty of ± 2.0 J-K~'-mor' to this quantity, which appears reasonable because, as mentioned above, the C°m function can only be evaluated graphically [95ROB/HEM]. Both the numerical value for this quantity and its error limits have been selected for this compilation. The so-called 'Latimer' contributions would result in S°(NiCO 3 , cr, 298.15 K) = (81.7 ±5.0) J-KT'mol 1 whereby the actually selected value falls within these error limits [93KUB/ALC]. The following value can be read off directly from Fig. 1 of [64KOS/KAL]: C°im(NiCO3, cr, 298.15 K) = (90.3 ± 4.1) J-K 'mol '. The uncertainty is assigned in such a way that the C° m values calculated according to Equation (V.I05) as well as estimated according to the cationic and anionic contributions, given by Kubaschewski et al. [93KUB/ALC], fall within the error limits. Although neither the experimental nor the semi-empirical basis for Equation (V.I05) could unambiguously be retraced, it has been retained in the current review. V.7.1.1.1.1.3
Thermodynamic analysis of solubility data on gaspeite
Solubility measurements are especially valuable for the determination of an internally consistent set of thermodynamic data for nickel carbonates. Other methods turned out to be less convenient. Thus the calorimetric determination of the dissolution enthalpy according to Reaction (V. 106) is bound to be inaccurate, because the result depends to a high extent on the state of carbon dioxide. It is well known that keeping CO2 quantitatively either in solution or in the gas phase is quite difficult. NiCO3(cr) + 2H+ ^ Ni2+ + CO2(g) + H2O(1) V.7.1.1.1.1.3.1
(V. 106)
Analysis of hydrothermaldecomposition
Another method to determine thermodynamic data of nickel carbonate consists in the investigation of the decomposition equilibrium (V.I07). Tareen et al. [91TAR/FAZ] determined the standard molar quantities of formation of gaspeite, NiCO3(cr) from hydrothermal T, p decomposition data of this reaction, but probably overestimated the precision achieved: AfG° = -(628.3611.24) kJ-moP1 (-(632.3 + 6.0) kJ-mol'), A f #° = - (703.38 + 1.24) kJmol 1 (- (709.2 ± 6.0) kJmol '). NiCO3(s) ^
NiO(s) + CO2(g)
(V.I07)
The numerical values in brackets have been found when the experimental results of [91TAR/FAZ] were re-evaluated according to the present state of the art, see Appendix A.
218
V Discussion of data selection for nickel
V. 7.1.1.1.1.3.2
Analysis of solubility studies
It was shown recently that solubility data obtained from NiCO3-5.5H2O(cr), hellyerite, were erroneously ascribed to NiCO3(cr), gaspeite [2001GAM/PRE]. Thus, in fact it was Reiterer [80REI] who determined the solubility constant of gaspeite defined by Reaction (V.I06) for the first time. He employed the pH variation method to investigate the temperature dependence of log10 AT 0 (V. 106) at constant ionic strength / = 1.0 mol-kg"1 (Na)C104. Kloger [82KLO] carried out additional experiments at / = 0.2 mol-kg ' (Na)C104 and 363.15K. The solubility data of synthetic NiCO3 were thermodynamically analysed to obtain the respective standard molar quantities of formation AfG° and A f / / ° . An analogous set of equations was used as described in Section V.3.2.2.3 (Equations (V.41) - (V.44)). Solubility constants were extrapolated to infinite dilution using the Specific Ion-interaction Theory [97GRE/PLY2]. All calculations were based on auxiliary quantities which were either recommended by COD ATA (see NEA TDB auxiliary data set) or recently critically evaluated (see Section V.2.1). The following set of reliable thermodynamic quantities has been derived [82GAM/REI], [98GAM/KON], [2002WAL/PRE]: log, 0 \, s , 0 ((V.106),298.15K) = (7.16±0.18) AfG° (NiCO3, cr, 298.15 K) = - (636.4 ± 1.3) kJ-mol 1 A f //° (NiCO3, cr, 298.15 K) = - (713.3 + 1.4) kJ-mol '. These values obtained from solubility measurements of pure synthetic NiCO3(cr) fall well within the error limits of those re-evaluated from decomposition studies, but are clearly more precise. V.7.1.1.1.2 V.7.1.1.1.2.1
NiCO3-5.5H2O(cr) Crystallography and mineralogy of nickel carbonate hydrate
The preparation and X-ray powder diffraction pattern of nickel carbonate hexahydrate, NiCO3-6H2O(cr), were first described by Rossetti-Francois [52ROS]. The crystal system is monoclinic-prismatic, space group: C2/c, cell dimensions: a$ = 10.770 A, bo= 7.299 A, c0 = 18.681 A, Z= 8; /?= 94.00°, Fcel, = 1464.94 A3, p(obs.) = 1.97 gem 3, p(calc.) = 1.975 g-cm 3 (according to JCPDS-ICDD cards 12-276 and 24-523, where the empirical formulae are NiCO3-6H2O(cr) in the former and NiCO3-5.5H2O(cr) in the latter). The natural occurrence of hellyerite, NiCO3-5.5H2O(cr), in the Lord Brassey nickel mine near Heazlewood (Tasmania), was first reported by Williams et al. [59WIL/THR]. The Mohs hardness is 2.5, its luster is vitreous (glassy) and its streak is white. The X-ray powder diffraction pattern was found to be similar to that of synthetic nickel carbonate hexahydrate, prepared by the method of Rossetti-Francois.
219
V. 7 Group 14 compounds and complexes The single crystal structure determination by Threadgold [63THR] of the natural product may be not perfect due to extensive twinning of the material. Hellyerite forms predominantly twin crystals to enhance the low crystal symmetry. Therefore the crystal structure needs to be refined to resolve its subtleties. Hellyerite has a prominent layered structure parallel to (001) and is made up of two distinct layers, a NiCO3(H2O)4 layer and an almost planar sheet of H2O molecules [63THR]. This layer structure with its unknown arrangement of H2O molecules is presumably responsible for chemical and thermogravimetric analyses of hellyerite resulting in a stoichiometric formula of NiCO3-5.5H2O(cr) i.e., NiCO3(H2O)41.5H2O(cr) instead of NiCO3-6H2O(cr) [2001GAM/PRE]. Crystallographic investigations to solve the hellyerite structure definitively are in progress. V.7.1.1.1.2.2
Heat capacity and entropy of NiCO3-5.5H2O(cr)
Apparently no experimental low-temperature heat capacity data of NiCO3(H2O)41.5H2O(cr) have been reported so far. A crude estimate was obtained by analogy to magnesium carbonate and its hydrates. When lansfordite (MgCO3-5H2O(cr)), nesquehonite (MgCO3-3H2O(cr)) and magnesite (MgCO3(cr)) are compared the heat capacity can roughly be approximated by Equation (V.I08) [99K.ON/KON], C;m(MgCO3-«H2O, cr, 298.15 K) = (76.09 + n * 58.0) J K 'mol '.
(V.108)
With Equation (V.I05) the following value can tentatively be used for the heat capacity of NiCO3-5.5H2O(cr): C;m(NiCO3-5.5H2O, cr, 298.15 K) = (405.4 ± 50.0) J K ' m o l 1 . The entropy of NiCO3-5.5H2O(cr) was determined by solubility measurements, 5°(NiCO3-5.5H2O, cr, 298.15 K) = (311.1 + 10.0) J-K '-mol ', see Section V.7.1.1.1.2.3, and is selected in this review. The standard entropy obtained from the thermodynamic analysis of solubility data agrees reasonably well with S°(NiCO3-5.5H2O, cr, 298.15 K) = 307.9 J-KT'-mor1 estimated according to Latimer [52LAT]. V.7.1.1.1.2.3
Thermodynamic analysis of solubility data on hellyerite
V. 7.1.1.1.2.3.1
Solubility measurements of hellyerite, NiCOy5.5H2O(cr)
As pointed out in Section V.7.1.1.1.1.3 solubility measurements of metal carbonates provide an important method to determine the thermodynamic quantities. For hellyerite, however, these data can be unambiguously assigned only when the stoichiometric composition is known. As formula and crystal structure are strongly interdependent, they should be determined simultaneously. So far neither has been established definitively and the thermodynamic quantities derived from solubilities must be used with caution.
220
V Discussion of data selection for nickel
With hellyerite prepared by a new method, solubility measurements were carried out at different temperatures and at / = 1.0 m (Na)C104 [2001GAM/PRE]. In addition, the solubility was also determined at 25°C and different ionic strengths [2002WAL/PRE]. In either case the pH variation method was used to study the dissolution reaction according to Reaction (V.I09): NiCO3-5.5H2O(cr) + 2H+ ^
Ni2+ + CO2(g) + 6.5H2O(1)
(V.I 09)
The thermodynamic calculations of this compilation are tentatively based on the formula NiCO3-5.5H2O(cr) which agrees with Threadgold's crystal structure analysis [63THR] and the composition of the synthetic product [2001GAM/PRE]. From the weak temperature dependence of the solubility constant, log10 Kps0, of hellyerite (see Figure V-43) A f //° (NiCO3-5.5H2O, cr) and ,S;(NiCO3(H2O)4l.5H2O, cr) were calculated using a non-linear least squares optimising routine.
Figure V-43: Temperature dependence of hellyerite solubility. / = 1.0 mol-kg ' NaClO4, • 278.15, x 288.15, • 298.15, • 308.15, * 313.15 K, solid line: specific ion-interaction formalism [97GRE/PLY2], dashed line: Pitzer model [91PIT].
V. 7 Group 14 compounds and complexes
221
Thermodynamic auxiliary data for H2O(1) and CO2(g) were taken from Table IV-1 (T= 298.15 K) and from CODATA [89COX/WAG] (7Y 298.15 K). Data for Ni2+ were used as selected in this review (see Section V.2.1). Nickel hydroxo and carbonato complexes are negligible [76BAE/MES], [2003BAE/BRA] in the pH ranges where the experiments of [2001GAM/PRE] and [2002WAL/PRE] were carried out. Figure V-44 shows the solubility constant of NiCO3-5.5H2O(cr) plotted as a function of ionic strength at 298.15 K. Both the Pitzer equations [91 PAP/PIT], [91 PIT], [99KON/KON] and the SIT, [97GRE/PLY2] were applied to model the ionic strength dependence of logl0 *Kps0. Whereas the Pitzer equations seem to predict the ionic strength dependence of the solubility constant more accurately than the SIT approach, the experimental data as a function of temperature coincide best with the latter, see Figure V-43. Figure V-44: Ionic strength dependence of hellyerite solubility, t — 25°C, Ionic medium: I = 0.5 to 3.0 mol-kg ' NaC104, A [2002WAL/PRE], solid line: specific ion-interaction formalism [97GRE/PLY2], dashed line: Pitzer model [91PIT].
A plot of log l0 (a 2t -a^Q -a\ ) versus logl0/?CO2 shown in Figure V-45 describes the thermodynamic analysis of the data presented in [11AGE/VAL], [30MUL/LUB] and [2002WAL/PRE] in comparison with the optimised line (slope = - 1.00). The activities of Ni2+ and H+ were calculated by using the SIT model. This result reveals a ratio of «[Ni]/«[CO2] close to unity in the solid phase, as is expected for NiCO3-5.5H2O(cr). A re-evaluation of the experimental data of Milller and Luber
222
V Discussion of data selection for nickel
[30MUL/LUB] as well as Ageno and Valla [11AGE/VAL] resulted in log10 *A^ 0 = (10.56 + 0.10), which compares favourably with logic X ° . . o ( ( V 1 0 9 ) ' 2 9 8 1 5
K
) = ( 1 0 6 3 ± °- 10 )
calculated from experimental data of [2001GAM/PRE], [2002WAL/PRE]. Thus it can be concluded that Ageno and Valla [11AGE/VAL] studied NiCO3-5.5H2O(cr) and not NiCO3 (log10 *K°^0 = (7.13 + 0.18)), see Appendix A. Obviously the solubility constants of hellyerite (Equation (V.109)) and gaspeite (Equation (V.106)) differ by 3.5 orders of magnitude (see Section V.7.1.1.1.1.3). This presents a further example that solubility data measured at different temperatures and ionic strengths provide a useful basis for the calculation of thermodynamic quantities when and only when the solid samples investigated are analytically and structurally unambiguously characterised. Moreover, a careful thermodynamic analysis showed that the experimental information of rather old solubility studies either falls in line with the optimised set of thermodynamic properties obtained [11AGE/VAL], [30MUL/LUB] or reveals that presumably a basic nickel carbonate was investigated [38SMU].
Figure V-45: Thermodynamic analysis of hellyerite solubility. Data of [11AGE/VAL]: •, datum of [30MUL/LUB]: A, data of [2002WAL/PRE]: 0 7 = 0.5, o / = 1.0, + / = 2.0, x 1= 3.0 molkg ' NaClO4. Solid straight line: calculated with log10 *K°j0 = 10.63 selected in this review.
V. 7 Group 14 compounds and complexes
223
Thus the selected standard Gibbs energy and standard enthalpy of formation have been derived from solubility measurements at various temperatures and ionic strengths. Af G°m (NiCO3'5.5H2O, cr, 298.15 K) = - (1920.9 ± 1.0) kJ-mof1 A f #° (NiCO3-5.5H2O, cr, 298.15 K) = - (2313.0 + 3.1) kJ-mol"1. A convenient overview of thermodynamic data on nickel carbonates and hydroxide is provided by a predominance diagram as depicted in Figure V-46. Figure V-46: Three-dimensional predominance diagram for the system Ni2+-CC>2-H2O. aNj2, = 10 4; black lines, solid phases: NiCO3, P-Ni(OH)2; gray lines, solid phases: NiCO3-5.5H2O(cr), p-Ni(OH)2; aqueous phase: Ni2+.
V.7.1.1.2
Aqueous nickel carbonato complexes
Recently the formation of carbonato complexes in the system Ni2+ - H2O - CO2 has been critically discussed and re-evaluated in a seminal review of Hummel and Curti [2003HUM/CUR]. So far only one paper has been published describing an attempt to determine the equilibrium constant, /?,, of Reaction (V.I 10) experimentally [87EMA/FAR]. Ni 2+ + HCO3 ^
NiHCO;
(V.I 10)
224
V Discussion of data selection for nickel
Emara et al. clearly misinterpreted their data and did not provide enough information to allow recalculation. Consequently, the stability constant of NiHCOj reported in [87EMA/FAR] cannot be included in this review. Values of equilibrium constants for Reaction (V.I 10) estimated by various procedures differ considerably 0.96 < log, 0 #°< 3.08 (T = 298.15 K) [76ZHO/BEZ], [77MAT/SPO], [79MAT/SPO], [84FOU/CRI]. The stability constant of the carbonato complex according to Reaction (V.I 11) has also been estimated leading to an even larger discrepancy: 2.56 < log10 K° < 6.87 (T= 298.15 K) [76ZHO/BEZ], [77MAT/SPO], [79MAT/SPO], [81TUR/WHI], [84FOU/CRI]. Ni2+ + CO*" ^ NiCO3(aq)
Kx
(V. 111)
One estimate exists for the equilibrium constant (log10 K^ = 3.24) of Reaction (V.112)[77MAT/SPO]. NiCO 3 (aq)+ C O ^ ^
Ni(CO 3 )^
K2
(V.I 12)
As the basis of the individual estimation procedures is rather dubious, variations of up to more than four log-units in these stability constants are to be expected [2003HUM/CUR]. Again neither of these values appeared suitable to be included in this review. Hummel and Curti proposed estimating K\ for Equation (V.I 11) using either the good correlation between the equilibrium constants of Ni(II) and Co(II) complexes and the poor data available for K{ of CoCO3(aq) or the rather poor correlation between Ni(II) and Zn(II) complexes and the excellent data for K\ for ZnCO3(aq) [2003HUM/CUR]. Both methods result in similar lower and upper bounds 4 8.5. At pH > 9 NiCC>3(aq) predominates.
Figure V-47: Calculated solubility of NiCO3(cr) in dilute sodium carbonate solutions under atmospheric conditions.
225
226
V.7.1.2
V Discussion of data selection for nickel
Nickel cyanides
V.7.1.2.1
Aqueous Ni(II) cyano complexes
The first quantitative data on the aqueous Ni(II) cyanide complexes were generated at the end of XIXth century, when Varet reported the heat of formation of the Ni(CN)4~ complex [1896VAR]. Varet's A r // m value is in good agreement with more recent determinations, but there has been an increase of nearly twenty orders of magnitude in the reported log,0/?4 value of Ni(CN)4~, since it was first measured by Masaki in 1932 [32MAS]. Besides the numerical value of log)0 /? 4 , conflicting results also have been obtained concerning the composition and protonation state of the species formed. Therefore, the discussion is divided into two parts. V.7.1.2.1.1
Complexes in neutral and alkaline solution
Most authors have agreed that the formation of NiCN+, Ni(CN)2(aq) and Ni(CN)3 can not be detected in the equilibrated solutions. Although, Shibata et ah claimed to detect Ni(CN)2(aq) and possibly Ni(CN)3(H2O)~ by electrophoresis of y-ray irradiated K2Ni(CN)4 solutions [66SHI/FUJ], the assignment of the measured peaks was rather speculative. The absence of the above-mentioned species indicates that the stepwise formation constants Ku K2 and K3 are much less than ljj34 , and thus K4 is probably greater than 1014 [68K.OL/MAR]. This very large value is likely related to the transformation of an octahedral to a square planar complex in the fourth stepwise complex formation process. The first reported value for logl0 /?4 [32MAS] was not accurate, since the electrode reactions involving Ni(II) are not reversible [50HUM/KOL]. Ni2+ + 4CN ^
Ni(CN)^
(V. 113)
Reliable values were reported from spectrophotometric [59FRE/SCH], kinetic [61MAR/BYD], [68KOL/MAR], [76PER], and pH-metric [63CHR/IZA], [71IZA/JOH], [74PER] measurements. Freund and Schneider used an incorrect pKHCN value, therefore the experimental data reported in [59FRE/SCH] were re-evaluated (see discussion in Appendix A). The experimentally determined log10 /?4 values are collected in Table V-30. Although only a limited number of data are available in NaC104 media, the precision of the constants is assumed to be good enough to perform an SIT analysis (Figure V-48). The weighted linear regression using five data points yielded the selected value of: logio A° ((V.I 13), 298.15 K) = (30.20 ± 0.12). From the slope in Figure V-48, As((V.113), NaC104) = -(0.465 ± 0.045) kg-mol"' can be calculated. Using the selected value for e(Ni2+, C1O4), and e(Na+, CN~) = (0.07 ± 0.03) kg-mol ' reported in [92BAN/BLI], e(Na+, Ni(CN)4~) = (0.185 + 0.081) kg-moP1 can be derived. In KC1O4 media, too narrow an ionic strength range is covered to perform the SIT analysis [59FRE/SCH], but after correction for the Debye-Huckel
V. 7 Group 14 compounds and complexes
227
contributions, the formation constants determined in these low ionic strength solutions are almost constant, and are nearly identical to the selected logl0 /?4° ((V.I 13), 298.15 K) value (see insert in Figure V-48). At ratios [CN~]/[Ni2+] > 4, the pale yellow solution of diamagnetic Ni(CN)4~ becomes successively orange, red and deep red upon addition of cyanide ion. The reason for this change is evidently the formation of higher complexes, but the composition and stability of these have been disputed for a long time. Several early reports [23JOB/SAM], [43SAM], [59KIS/CUP], [59BLA/GOL], [59BLA/GOL2] interpreted the colour change as evidence for the formation of hexacyanonickelate(II) ion, Ni(CN)^". Some of them reported surprisingly high equilibrium constant values for the reaction: Ni(CN)^ + 2CN~ ^
Ni(CN)j"
[59KIS/CUP], [59BLA/GOL2]. Later, the formation of a weak pentacyano complex and the non-existence of the earlier-suggested Ni(CN)*~ ion were definitely proven [60MCC/JON], [62BEC/BJE], [64GEE/HUM], [65COL/PET], [71 PIE/HUG]. The reported equilibrium constants for the reaction: Ni(CN)^ + CN ^
NiCCN);"
(V. 114)
are collected in Table V-30. Since the pentacyano complex is rather unstable, high cyanide concentrations were used to achieve its formation, which resulted in considerable replacement of the original background electrolyte by NaCN. Due to this medium effect, considerably higher uncertainties have been assigned to the equilibrium constants than originally reported. The SIT analysis of the constants determined in NaClO4 media (Figure V-49), resulted in logl0 AT5°((V.114), 298.15 K) = -(1.70 ±0.36), and Ae((V.l 14), NaClO4) = (0.00 ± 0.11) kg-mol"1. From the latter value, e(Na+, Ni(CN)^) = (0.25 + 0.14) kg-mol' can be derived. Using equilibrium constant from above, the overall formation constant of the Ni(CN)^ species is Ni2+ + 5 CN" ^
Ni(CN)f
log10 PI ((V.I 15), 298.15 K) = (28.5 + 0.5).
(V. 115)
228
V Discussion of data selection for nickel
Table V-30: Experimental equilibrium data for the Ni(II) cyanide system. Method
L
Medium
reported 2+
Ni + 4CN"
retained'"'
^ Ni(CN);
pot
NaCN
0.78 M
25
(11.83 + 0.17)
pol
KC1
0.10
9
>24
K(C104)
0.1
25
sp
Reference
log,. A
'(°C)
[32MAS] [50HUM/KOL] (29.37 + 0.20)
(b)
0.048
(29.45 ± 0.20)
(b)
0.014
(29.79 + 0.20)(b)
0.0028
(30.12 + 0.20)(b)
0 corr
[59FRE/SCH]
(31.0 + 0.08)
kin
NaCl
1.03
25
31.5
gl kin
(self)
0 corr
25
(30.1 ±0.2)
(30.1 ±0.2)
[63CHR/IZA]
0.10
25
(30.5 ±0.3)
(30.48 ± 0.6)
[68KOI7MAR]
gl
(self)
[71IZA/JOH]
gl kin
NaC104
10
(32.20 + 0.20)
(32.20 + 0.30)
25
(30.22 ± 0.05)
(30,22 ±0.15)
40
(27.43 ± 0.09)
(27.43 ± 0.20)
NaC104
3.50
25
(31.12 ±0.08)
(30.85 ±0.15)
[74PER]
NaC104
3.50
25
(31.08 ±0.05)
(30.81 ±0.15)
[76PER]
15.4
- (0.66 ± 0.02)
- (0.69 ± 0.30)
[60MCC/JON]
25.2
- (0.72 + 0.02)
- (0.75 + 0.30)
Ni(CN); +CN~ ^ sp
0 corr
[61MAR/BYD]
Na(C104)
Ni(CN)5 1.43
33.6
- (0.77 ± 0.02)
-(0.80 ±0.30)
ir
Na(ClO4)
4.95
25
- (0.55 ± 0.02)
- (0.65 ± 0.30)
sp
Na(C104)
2.84
23
- (0.70 ± 0.02)
sp
KF-KCN
4.00
25
(0.03 +0,0 l) (c)
sp
Na(ClO4)
2.21
20
-(0.77 ±0.01)
25
[63PEN/BAI] [64GEE/HUM]
- (0.77 + 0.30)
25 (a) Accepted values corrected to molal scale. (b) Re-evaluated value, see Appendix A. (c) In molal units. (d) Corrected to 25°C, using the selected enthalpy values.
[65COL/PET] [71 PIE/HUG] - (0.84 ± 0.30)(d)
V. 7 Group 14 compounds and complexes
229
Figure V-48: Extrapolation to / = 0 of experimental formation constants of Ni( (V. 113) determined in NaClO4 medium (insert shows the corresponding plot of data reported for KC1O4 medium).
Figure V-49: Extrapolation to / = 0 of experimental equilibrium constants for Reaction (V.I 14) determined in NaC104 medium.
230
V Discussion of data selection for nickel
The selected thermodynamic formation constants correspond to: ArG° ((V.I 13), 298.15 K) = - (172.4 ± 0.7) kJmol ', ArG^ ((V.I 15), 298.15 K) = - (162.7 ± 2.9) kJ-mol"' and thus the standard molar Gibbs energies of formation are: AfG° (Ni(CN)^ , 298.15 K) = (449.6 + 10.1) kJmol 1 , AfG° (Ni(CN)f, 298.15 K) = (626.6 ± 12.9) kJmol '. The reaction enthalpy of formation of the tetracyano complex has been studied calorimetrically by Varet [1896VAR] and by Izatt and his co-workers [63CHR/IZA], [71IZA/JOH] (see Table V-31). The value determined in [63CHR/IZA] is selected in this review: A r tf° ((V.I 13), 298.15 K) = - (180.7 + 4.0) kJmol '. Izatt et al. also determined the standard enthalpy of Reaction (V.I 13) at two other temperatures (10 and 40°C), which allowed calculation of the standard molar heat capacity of the reaction ArC°m((V.113), 298.15 K) = (150.6±42.0) J K ' m o l ' [71IZA/JOH]. Neglecting the ionic strength dependence, A r //° ((V.I 14), 298.15 K) = -(10.4 ±4.0) kJ-mol ' can be estimated from the temperature variation of log l 0 ^ 5 [60MCC/JON]. The combination of the above two enthalpy values yielded: Ar//n°,((V.I 15), 298.15 K)=-(191.1 ± 8.0) kJ-mol '. Table V-31: Experimental enthalpy values for Reactions (V. 113) and (V. 114). Medium Ar//m (kJ-mol ') Ar//m (kJmol ') Reference L '(°C)
Method
reported Ni2h + 4CN" ^ cal
retained
Ni(C:N)^ ?
14
-(180 + 4.2)
cal
(self)
0 corr
25
-(180.7 ±2.0)
-(180.7 ±4.0)
[63CHR/IZA]
cal
(self)
Ocorr
10
-(189.1 ± 1.2)
-(189.1 ±2.0)
[71IZA/JOH]
40
-(183.7 ±0.8)
-(183.7 ±2.0)
-12.5
-(10.4±4.0) (a)
Ni(CN)*~ + CN" sp
Na(ClO4)
[1896VAR]
^ Ni(CN),
1.43
15-34
[60MCC/JON]
(a) Calculated from the temperature dependence of the corresponding formation constant, see Appendix A.
From these data A f //° (Ni(CN)^, 298.15 K) = (353.7 ± 14.7) kJ-mor1, and
V. 7 Group 14 compounds and complexes
231
A f //°(Ni(CN)5", 298.15K) = (490.6+ 19.4) kJ-mol ' are selected in this review. These above selections yield: S° (Ni(CN)^ , 298.15 K) = (245.0 + 36.6) J-K '-mol ', S°(Ni(CN)f ,298.15 K) = (278.8 ±51.1) J-K'-mol 1 . V.7.1.2.1.2
Protonated complexes
The cyano complexes of several metal ions are reported to undergo protonation in the acidic pH-range [87BEC], but rather contradictory results are available for Ni(II). The protonation constants of Ni(CN)^ (log^AT, = (5.4 + 0.2), log10 K2 = (4.5 + 0.2), log10 K} = (2.6 ± 0.4) at t = 25°C and / = 0.1 M NaClO4) were first reported by Kolski and Margerum, based on a kinetic study of the acid dissociation of Ni(CN)J~ [68KOL/MAR]. Their equilibrium measurements between pH = 4 - 5 , using a spectrophotometric method, resulted in systematic changes in the calculated logl0 /?4 values. This was taken as further proof for the presence of protonated species, although the UV-VIS spectrum of Ni(CN)^ was found to be unaffected by the protonation. Two successive protonations of Ni(CN)J" have been reported using a pH-metric method in [74KAB], but the results are uncertain, and thus this work was not considered in the present review (see Appendix A). Recently, Monlien et al. reported a kinetic study on the cyanide exchange with the Ni(CN)^ complex [2002MON/HEL]. The kinetic data confirmed the presence of protonated species, although the protonation did not affect the 13C NMR spectrum of Ni(CN)^". On the other hand, spectrophotometric [59FRE/SCH] and careful potentiometric [63CHR/IZA], [71IZA/JOH], [74PER] studies performed between pH = 4 — 7 did not indicate the existence of protonated species in notable concentrations, although their kinetic role in the acid dissociation of Ni(CN)^~ was later corroborated [76PER]. To sum up, the kinetic role of the protonated species in some reactions of Ni(CN)4~ is well established, but the presence of such species in equilibrated solutions is at least questionable. Additional experimental data would be needed for a definitive conclusion. V.7.1.3 V.7.1.3.1
Nickel thiocyanates Aqueous nickel thiocyanato complexes
The thiocyanate ion is a widely studied ligand, mainly due to its ambidentate nature. Aside from some general reviews [75NEW2], [79GOL/KOH], a critical survey of stability constants of thiocyanate complexes has been published [97BAH/PAR]. The latter publication listed the majority of available thermodynamic data for the Ni(II) - thiocyanate system, but recommended values are given only for log,,,/?! (V.I 16) at / = 1.0 and 1.5 M.
232
V Discussion of data selection for nickel
Thiocyanic acid is a moderately strong acid. Only a few badly scattered data are available for estimation of its protonation constant [97BAH/PAR]. Among them, log10 K{ = - 0.87 at / = 2.0 M [82NEV/ANG] seems to be the most reliable. Thus, the protonation of the thiocyanate anion is negligible even at the most acidic conditions used [64TRI/CAL], [68MAL/TUR] for studying the Ni(II)-thiocyanate interaction. The thiocyanate anion is N-coordinated in most of its Ni(II) compounds [77CHE/BRO]. The aqueous Ni(II)-thiocyanate complexes have octahedral geometry, although the [Ni(SCN)4(H2O)2]2 species is probably in equilibrium with a minimal concentration of tetrahedral [Ni(SCN)4]2 [88BJE2]. Some early papers reported qualitative information on the Ni(II) — thiocyanate system [38CSO], [42MAJ] based on spectrophotometric studies. Later, several experimental techniques, namely ion exchange, spectrophotometric, polarographic, distribution, IR, potentiometric and kinetic methods, have been used to derive equilibrium data for the Ni(II)-thiocyanate complexes. Depending on the ligand-to-metal ratio, the formation of four species is generally recognised: Ni2+ + qSCN~ ^
Ni(SCN)^*
(V. 116)
with q = 1 - 4. The reported formation constants are listed in Table V-32. For reasons discussed in Appendix A, the data reported in [61DAV/SM1], [61MOH/DAS], [62FRO/LAR], [62TRI/CAL], [7IDAS/DAS], [71 LAN/CUM], [74NET/BAT] were not considered in the present review. In some cases [53FRO2], [64TRI/CAL], [74DIC/HOF], the formation constants reported for 20°C were corrected to 25°C, using the enthalpy values selected below. A majority of the accepted experimental data refers to the formation of the NiSCN+ species in perchlorate media. SIT analysis of these data shows acceptable consistency (cf. Figure V-50). The weighted linear regression using 12 data points yielded the selected value of: log10 P° ((V.I 16), q=\, 298.15 K) = (1.81 ± 0.04). Since no experimental data are available, in the above treatment e(H+, SCN ) and s(Li , SCN ) ion interaction parameters were assumed to be equal to s(Na+, SCN"). From the slope in Figure V-50, Ae((V. 116), q = 1, CIO;) = - (0.109 + 0.025) kg-moP1 can be calculated. Using the selected values for e(Ni2+, C1O4) and e(Na+, SCN ), Ae((V.l 16), q=\, C1O4) leads to a value of e(NiSCN+, CIO;) = (0.31 + 0.04) kg-moP1. +
Less data are available for the formation of the Ni(SCN)2(aq) and Ni(SCN)j species. The SIT treatment of the accepted data (cf. Figure V-51 and Figure V-52) resulted in the following selected thermodynamic formation constants: log10 PI ((V. 116), q = 2, 298.15 K) = (2.69 ± 0.07), logl0 Pi ((V.I 16), q = 3, 298.15 K) = (3.02 ± 0.18). From the slopes, Ae((V.116), q = 2, ClO^) = -(0.091 ±0.043) kg-mol"1 and Ae((V.116), q = 3, C1O4) = (0.14 + 0.12) kg-mol 1 can be derived. These parameters
233
V. 7 Group 14 compounds and complexes
lead to the values of e(Ni(SCN)2(aq), Na+ + CIO;;) = (0.38 + 0.06) kg-mol ' (also see [97ALL/BAN]) and s(Na+, NiCSCN)") = (0.66 ±0.13) kg-mol '. The use of an interaction coefficient like e(Ni(SCN)2(aq), Na+ + CIO 4) for a neutral species is not a standard procedure within the context of the SIT as described in Appendix B, but appears to be justified in the current case. Further experimental work could prove useful. Several experimental values are published for logl0 /?4 (or log10 K^) [64TRI/CAL], [71 LAN/CUM], (or [88BJE2]). In [64TRI/CAL] there was a rather small excess of ligand (cSCN_ < 0.5 M), and for both ionic strengths AT3 < K4 is reported. Considering that no geometrical change of Ni(II) occurs during the stepwise complex formation, the reported K4 values are probably overestimated. The method used in [71 LAN/CUM] and [90BJE2] is not compatible with the SIT. Although there is good evidence for the existence of Ni(SCN)4~ [88BJE2], the available data cannot be used to derive a selected value. The above listed selected logl0 j3q values correspond to: ArG° ((V.I 16), 9 = 1 , 298.15 K) = - (10.33 ± 0.22) kJ-mol ', ArG° ((V.I 16), q = 2, 298.15 K) = - (15.35 + 0.40) kJmol ', ArGn° ((V.I 16), q = 3, 298.15 K) = - (17.24 ± 1.03) kJ-mol ', and hence AfG° (NiSCN+, 298.15 K) = (36.60 + 4.08) kJ-mol 1 , AfG°(Ni(SCN)2, aq, 298.15 K) = (124.27 ± 8.04) kJ-mol ', A f G°(Ni(SCN)3, 298.15 K) = (215.09 ± 12.07) kJ-mol '. Table V-32: Experimental equilibrium data for the Ni(II) - thiocyanate system. Method
t(°C)
Medium
l o g | 0 /}r
reported >Ji2* + SCN" ^
NiSCN
Reference
_ (l.!2±0.06) ( b ) (1.69 + 0.15)
[53FRO2]
+
cix
(Na)ClO 4
1.05
sp sp pot sp ir pol sp
Ni(NO 3 ) 2
0 corrw
Ni(C10 4 ) 2 + KSCN
Ocorr
?
0.5 M
(NaClO 4 )
log,»A
retained*"'
20 25 23 25 35 1.4
3.5
(Na, H)C1O 4
1.05
?
(Na)ClO 4
0 corr
25
0.43
(1.18 ±0.02) (1.67 + 0.03) (1.50 ±0.04) (1.82 + 0.02) 1.38 1.18 (1.17 + 0.02) (1.762 + 0.005) -
(1.77 ± 0.I0)(d) (1.20 + 0.08)"1'
[58YAT/KOR] [61MOH/DAS] [61DAV/SMI] [62FRO/LAR] [62TRI/CAL] [62WIL2]
(Continued on next page)
234
V Discussion of data selection for nickel
Table V-32 (continued) Method
dis
Medium Ni2+ + SCN" ^± NiSCN+ (NaC104)
/m
/(°C)
l O g 10 y3v reported
log,. A accepted'3'
1.62
20 25 20 25 15 25 35 25 25 25 35 45 ? 5 15 25 35 25 45 20 25 25
(1.14 ±0.02) (1.19 ±0.02) (1.34 ±0.02) (1.24 ±0.02) (1.17 ± 0.03) (1.3 ±0.2) (1.34 ±0.01) (1.98 ±0.02) (1.85 ±0.02) (1.74 ±0.02) 2.22 1.55 1.46 1.37 1.29 (1.09 ±0.06) (1.00 ±0.03) 1.83
(1.08 ± 0.10)(b) (1.09 ± 0.10)(b) (1.33 + 0.10) (1.23 + 0.10) (1.16 + 0.10) (1.29 ±0.20) (1.27 ±0.10)
3.50 pol
(H)C1O4
0.67
dis sp pot
(Na)ClO4 (LiClO4) KSCN
0.26 3.48 Ocorr
aix pol
KSCN (KNO3)
kin
(NaC104)
0 corr 0.2 0.2 0.2(c) 0.2 1.05 Ocorr
kin cat ir dis cix dis
(NaClO4) (LiClO4) (NaC104)
1.05 3.48 1.05
Ni2! + 2SCN" ^Ni(SCN) 2 (aq) (Na)ClO4 1.05 (NaC104)
1.62 3.50
aix pol
KSCN (KNO3)
cal dis
(NaC104) (NaC104)
0 corr 0.2 0.2 0.2(c) 0.2 1.05 1.05
25
(1.14 ±0.02) 0.3-1.44 1.1
20 25 20 25 20 25 ? 5 15 25 35 25 25
(1.64 ±0.04) (1.76 ±0.06) (1.68 + 0.05) 2.43 2.23 2.04 1.86 1.69 (1.58 + 0.05) 1.6
7
Reference
[64TRI/CAL]
[68MAL/TUR]
[69SUB/COR] [70MIR/MAK] [7 IDAS/DAS]
[71 LAN/CUM] [71TUR/MAL]
(1.07 ±0.10) (0.98 ±0.10) (1.79 ± 0.10)(b) (1.12 ±0.02) (1.08 ±0.10) (1.54±0.06) (b) (1.63 ± 0.10/ b) (1.48 ± 0.10)(b)
[73HUT/HIG] [74DIC/HOF] [74KUL3] [74NET/BAT] [76MUR/KUR] [53FRO2] [64TRI/CAL]
[71 LAN/CUM] [71TUR/MAL]
(1.54 + 0.05) (1.56 ±0.20)
[74KUL3] [76MUR/KUR]
(Continued on next page)
235
V. 7 Group 14 compounds and complexes
Table V-32 (continued) Method
cix dis
Medium
/m
Ni 2+ + 3SCN-;F± Ni(SCN); (Na)ClO 4 1.05
0 corr 1.05
20 25 20 25 20 25 ? 25
Ni2+ + 4SCN" ;=i Ni(SCN)^ 1.62 (NaC10 4 ) 3.50 KSCN 0 corr 0 corr NaSCN
20 20 ? 25
(NaClO4)
1.62 3.50
aix cal dis aix sp
KSCN (NaC10 4 )
logl0 [5K reported
/(°C)
(1.81 ±0.07) (1.70 + 0.10) (1.3 ±0.3) 2.62 (1.6 + 0.2)
tog.. A
Reference
(1.66 ±0.10)"" (1.52 ±0.20)""
[53FRO2]
accepted'2'
[64TRI/CAL]
(1.01 ±0.30)"" (1.54 + 0.20)
(2.04 + 0.20) (1.54 ±0.35) 2.15 log,,, Kt = - (0.92 ±0.18)(e>
[71 LAN/CUM] [74KUL3] [64TRI/CAL] [71 LAN/CUM] [88BJE2]
(a) Accepted values corrected to the molal scale. The accepted values reported in Appendix A are expressed on the molar or molal scales, depending on which units were used originally by the authors. (b) Corrected to 25°C, using the accepted enthalpy values. (c) Data for several ionic strengths, see Appendix A. (d) Re-evaluated value, see Appendix A. (e) 'Semi-thermodynamic' constant, the activity of thiocyanate ion (using y± of NaSCN) was taken into account.
Figure V-50: Extrapolation to Im = 0 of the accepted formation constants reported for the formation of NiSCN+ in perchlorate media.
236
V Discussion of data selection for nickel
Figure V-51: Extrapolation to 7m = 0 of the accepted formation constants reported for the formation of Ni(SCN)2(aq) in perchlorate media.
Figure V-52: Extrapolation to Im = 0 of the accepted formation constants reported for the formation of Ni(SCN)3 in perchlorate media.
237
V. 7 Group 14 compounds and complexes
Five reliable reports are available for the reaction enthalpies of the formation of Ni(SCN)^~* (x = 1 - 3) complexes. The reported values are collected in Table V-33. Table V-33: Experimental reaction enthalpy values for the formation of Ni(II)thiocyanate complexes. Methc)d
/m
Medium
( (°C)
A,Hm (kJ-moP1) reported
Ni 2 ' + S O T ^
A,Hm (kJ-mor1)
IvliSCTT
cal
NiC104 + KSC1^i
pol
(H)C1O4
< 0.06
25
-(10.46 + 0.40)
0.67
15 - 3 5
- 14.43
- (9.25 ± 2.00)la)
[67NAN/TOR]
-(14.5 + 2.0)
[68MAL/TUR]
lb)
pol
(KNO3)
0.2
5 -35
-(14.3 + 2.0)
kin
(NaClO4)
1.05
25 - 4 5
- (8.3 ± 2.0)(b)
cal
(NaC104)
1.05
25
2t
Ni + 2SCN(KNO3)
0.2
cal
(NaClO4)
1.05
cal
(NaClO4)
-(12.02 + 0.15)
(12.0±0.5)
[71TUR/MAL] [73HUT/HIG] [74KUL3]
^ Ni(SCN)2(aq)
pol
Ni2+ + 3SC>T
Reference
r etained for selection
- (29.5 ± 2.0)(b)
[71TUR/MAL]
25
- (20.92 ± 1.15)
-(20.9 ±1.2)
[74K.UL3]
25
- (29.12 ±5.15)
-(29.1 ±5.2)
[74KUL3]
5 -35
^ Ni(SCN), 1.05
(a) Re-evaluated value, see Appendix A. (b) Calculated from the temperature dependence of the corresponding formation constant, see Appendix A.
These data do not allow correct evaluation of the ionic strength dependence, therefore it was assumed that reaction enthalpies are independent of the ionic strength. Assigning higher weight to the most reliable calorimetric data published in [74KUL3], and considering the above assumption, the weighted averages of the retained values are as follows: A r //°((V.116), g=l,298.15 K) = - (11.8 ±5.0) kJ-mol ', A r //° ((V.I 16), 9 = 2, 298.15 K) = - (21 ± 8) kJ-mol ', \H°m((V.I 16), q = 3, 298.15 K) = - (29 ± 10) kJ-mol '. From the above selected A r //° values, the following standard enthalpies of formation can be derived: A f tf° (NiSCN+, 298.15 K) = (9.59 + 6.46) kJ-mol ', A f //° (Ni(SCN)2, aq, 298.15 K) = (76.79 + 11.35) kJ-mol ', A f //° (Ni(SCN)", 298.15 K) = (145.19 ± 15.65) kJ-mol '. The above selections yield: 5° (NiSCN+, 298.15 K) = (7.5 + 25.4) J-K^-mol 1 , 5° (Ni(SCN)2, aq, 298.15 K) = (137.8 ± 46.5) J-K '-mol ',
238
V Discussion of data selection for nickel
S° (Ni(SCN)3, 298.15 K) = (261.5 ± 66.2) J-K"'-mor'.
V.7.2 Silicon compounds and complexes V.7.2.1 Solid nickel silicates V.7.2.1.1 V.7.2.1.1.1
Nickel orthosilicate Ni2SiO4(cr) Crystal structure and phase transitions
The only stable silicate in the system NiO - SiO2 is the orthosilicate 2NiOSiO 2 [54EIT], [66GME], [63PHI/HUT], [69LEV/ROB], [91KNA/KUB]. The existence of the pure metasilicate NiSiO3 has not been confirmed by any X-ray investigation [66GME]. However, an estimation of the thermodynamic data and the stability for the nickel metasilicate (pyroxene) was performed by Campbell and Roeder [68CAM/ROE] and by Navrotsky [71NAV]. In the present review, the thermodynamic data for the NiSiO3 are not evaluated because the pure nickel pyroxene is unstable relative to silica and the nickel olivine at a total pressure of one atmosphere [68CAM/ROE]. The 2NiOSi0 2 , orthorhombic crystals, have the olivine structure, but at higher temperature and pressure a spinel transformation can occur. This transformation was the subject of numerous studies [62RIN], [65AKI/FUJ], [68NAV/KLE], [73NAV], [73NAV2], [76NAV/KAS], [74YAG/MAR], [74MA] because of great geophysical interest. The approximate equilibrium pressure for the olivine - spinel transition in Ni2SiO4 at 650°C was found to be 18 kb [62RIN]. This value is about 40% lower than that determined by [74MA] who found a linear pressure - temperature dependence of the transformation, expressed by the equation: p/bar = 23000 + 1 1 . 8 / I°C The following thermodynamic data for the olivine (ol.) - spinel (sp.) transition in Ni2SiO4 were obtained by Navrotsky [73NAV] by measuring the heats of solution of both polymorphs in a molten oxide at 986 K: the enthalpy of the transition Ni2SiO4(ol.) —> Ni2SiO4(sp.), A trs //°(986 K) = (5.912.9) kJ-mol"', the heat content increments, ( / / m ( 9 8 6 K ) H°m (298 K)), olivine, (107.7 ± 1.8) kJ-mol ', spinel, (106.2 + 0.8) kJ-mol"1 and the entropy of the olivine - spinel transition AtrsS^= 12.6 to 14.6 .MC'-mol"1. Although the spinel polymorph of Ni2SiO4 also exists in metastable form at atmospheric pressure, the thermodynamic data for this high pressure compound are not comprehensively treated in the present review. Ma [74MA] found that Ni 2 Si0 4 olivine melts incongruently at high pressures to NiO plus melt and that it is a stable phase until melting occurs. Since such melting behaviour persists at a pressure as low as 4.9 kbar, he expected that similar melting behaviour also occurs at atmospheric pressure. The melting curve extrapolated to atmospheric pressure gave a melting temperature of 1848 K [74MA]. The phase diagram NiO-SiO2, drawn from data obtained by quenching and direct observational techniques shows, however, that the decomposition temperature of the nickel olivine is
239
V. 7 Group 14 compounds and complexes (1818 ± 5) K [63PHI/HUT], [69LEV/ROB]. The dissociation reaction occurring prior to melting is Ni2SiO4(cr) ^
2NiO(cr) + SiO2(cris.)
(V. 117)
where (cris.) means cristobalite. The uncertainty of the melting and decomposition temperature of Ni2SiO4 olivine was minimised by refining the thermodynamic data concerning this silicate [87NEI]. The decomposition temperature of Ni2SiO4 may therefore be given as (1820 + 5) K [87NEI]. This is in perfect agreement with Phillips et al. [63PHI/HUT]. V.7.2.1.1.2
Crystallography and mineralogy of nickel orthosilicate
Liebenbergite (Ni2Si04) is an end-member mineral of olivines being a group of minerals of the general formula X 2 Si0 4 , where X is a divalent metal cation (Mg, Fe, Mn, Ni, Ca, and Co) [82BRO]. The pure nickel olivine (liebenbergite) does not occur naturally [73WAA/CAL], [2001 HEN/RED]. The natural occurrence of liebenbergite (Ni, Mg)2Si04 of an approximate empirical formula Ni, 5Mg05SiO4 has been reported from Scotia Talc mine, Bon Accord, Barbeton distr., Transvaal, South Africa [73WAA/CAL]. Liebenbergite was named after W. R. Liebenberg, director of the National Institute of Metallurgy of South Africa. The crystal system of liebenbergite is orthorombic (space group Pbnm; Z = 4) [87BOS]. Cell data for the end-member Ni 2 Si0 4 are a = 4.727 k,b= 10.120 A, c = 5.911 A, V= 284.98 A3 [2001 HEN/RED]. The X-ray powder diffraction pattern is listed on JCPDS-ICDD card No. 15 -388. V.7.2.1.1.3
Heat capacity and standard entropy
Heat capacity measurements of the nickel silicate olivine were reported by [82WAT], [65EGO/SMI], and [84ROB/HEM]. However, the heat capacities reported by Watanabe in the temperature range 350 to 700 K [82WAT] are 4.6% greater than the heat capacities obtained by [84ROB/HEM] with the more accurate low-temperature calorimeter. Therefore, the results of Watanabe are not incorporated in this review. Egorov and Smirnova [65EGO/SMI] have determined the specific heat capacities of the nickel orthosilicate in the temperature range between 555 to 1570 K using a copper block calorimeter. The well-described, careful measurements allowed a determination of C°m to within + 0.5%. These high-temperature data were used in this review to extend the heat capacity equation up to 1570 K. Robie et al. [84ROB/HEM] have measured the heat capacities of Ni2SiO4-olivine between 5 and 387 K by cryogenic adiabatic-shield calorimetry and between 360 and 1000 K by differential scanning calorimetry. The standard molar heat capacity at 298.15 K, determined from these low-temperature measurements, was reported to be (123.20 ± 0.20) J-IC'-mor1 [84ROB/HEM]. This value, the only one experimentally determined by accurate calorimetric measurements, is selected by the present review: C° (Ni2Si04, olivine, 298.15 K) = (123.20 + 0.50) J-K '-mol '.
240
V Discussion of data selection for nickel
An assumption of a higher uncertainty for the accepted value of C° m (Ni2SiO4, olivine, 298.15 K) was necessary to achieve the coincidence of the standard heat capacity value at 298.15 K obtained from the low temperature measurements [84ROB/HEM] and from the high temperature thermal heat capacity function selected by this review (see below). According to [84ROB/HEM] between 300 and 1300 K the heat capacities can be represented by the equation: C° m (7) = (289.73 -0.024015 T+ 131045 T~2-2779.0
T
1/2
) J-K ' mol '
to within ± 0.5%. It should be mentioned that these authors carried out measurements only in the temperature range between 5 and 1000 K. Therefore, the extrapolation using the above equation up to 1300 K seems to be less reliable. In this review, we have used the calorimetric heat capacity measurements of [65EGO/SMI] and [84ROB/HEM] and fitted the thermal heat capacity function in the temperature range 270 < (77K) < 1570. The selected experimental data and the fitted curve are shown in Figure V-53. The coincidence of both data sets is satisfactory. Thus, we can select the following thermal function of the heat capacity for Ni2SiO4-olivine between 270 and 1570 K: Cp.mC7) = (147.61612 + 0.0486860 T- 1.50143-10 5 T2 - 3.34214-106 T 2) J-K '-mol ' with an uncertainty of ± 0.5%. Figure V-53: The standard molar heat capacity of Ni2SiO4-olivine as a function of temperature. The data are taken from [84ROB/HEM] (O) and [65EGO/SMI] (A). The fitted curve (solid line) corresponds to the thermal heat capacity function selected in this review.
241
V. 7 Group 14 compounds and complexes The standard molar entropy at 298.15 K for Ni2SiO4-olivine as determined by calorimetric measurements by Robie et al. [84ROB/HEM] is: S°m (Ni2SiO4, olivine, 298.15 K) = (128.10 ± 0.20) J K 'mol '. This value was also selected by the present review. The entropy value calculated from the temperature dependence of the Gibbs energy of the reaction of nickel with silica and oxygen e.g., as estimated by [68CAM/ROE], [76MAH/PAN], was not accepted by this review. V.7.2.1.1.4
Enthalpy of formation
The enthalpy of formation of Ni2SiC>4 olivine from the component oxides was measured by solution calorimetry in a molten oxide solvent at (965 1 2 ) K by Navrotsky [71NAV]. The value of ArHm ((V.I 18), 965 K) for the reaction: 2NiO(cr) + SiO2(q) ^ Ni2SiO4(ol.)
(V. 118)
was reported to be - (13.9 ± 1.9) kJmol '. Using the heat capacities of the components, the standard molar enthalpy of this reaction at 298.15 K was calculated: A r //° ((V.I 18), 298.15 K) = - (6.3 ± 6.7) kJmol"""' [78KOT/MUL]. The main interest, for this review, is the determination of the standard enthalpy of formation of Ni2SiO4-olivine from the elements in their standard state i.e., A f i/° for the reaction: 2Ni(cr) + Si(cr) + 2O2(g) ^ Ni2Si04(cr)
(V. 119)
The enthalpy of Reaction (V. 118), together with the enthalpies of the formation of SiO2(q), where (q) means quartz, and NiO(cr) and the thermal functions of heat capacity of all components can of course be used for the calculation of A f //° (298.15 K). Generally, the enthalpy of the reaction based solely on calorimetric data is the least reliable quantity, and contributes the major part of the uncertainty in the determination of the Gibbs energy of reaction. On the other hand, the standard entropy and the high-temperature heat capacities as well as the Gibbs energy of a reaction do provide more accurate data [87NEI]. Thus, using the third-law method, i.e., the equation: \H°m (298.15 K) = ATG°m (T) - £
|S ArC;,m
dT +
7-(Ar5° (298.15 K)+ £ 8i5 (A r C p o m IT)dT),
(V.120)
the standard molar enthalpy of a reaction at 298.15 K can be calculated. Based on his emf-measurements and the heat capacity values, among others those for the Ni 2 Si0 4 olivine determined by [84ROB/HEM], O'Neill [87NEI] used this method to calculate A r //° (298.15 K) for the reaction: 2Ni(cr) + SiO2(cr) + O2(g) ^
Ni2SiO4(cr)
(V.121)
242
V Discussion of data selection for nickel
to be -482.42 kJmol '. Using this value it was possible to calculate the Gibbs energy of Reaction (V. 118) and to achieve very good agreement between the calculated and experimentally observed decomposition temperature of Ni2SiO4-olivine, i.e., (1820±5)K[87NEI]. Robie and Hemingway [95ROB/HEM] used their accurate heat capacity measurements, combined with results from molten salt calorimetry, thermal decomposition of the Ni2SiO4-olivine into its constituent oxides, and equilibrium studies, both by CO reduction and solid state electrochemical cell measurement for Reaction (V.121) [87NE1], and calculated the following standard molar enthalpy of formation of Ni 2 Si0 4 olivine (liebenbergite): A f //° (298.15 K) = -(1396.5 ± 3.0) kJmol '. In the present review, the standard molar enthalpy of formation of Ni2SiO4-olivine at 298.15 K was also determined by the third-law method using the thermal heat capacity function of Ni2SiO4-olivine, the equation representing the Gibbs energy of Reaction (V.121) in the temperature range 970 - 1770 K, both derived by this review as described in Sections V.7.2.1.1.3 and V.7.2.1.1.5, as well as thermodynamic data of elemental nickel selected by this review. Elemental silicon, oxygen and silica (quartz) data are taken from the NEA-TDB auxiliary dataset. The value obtained and selected is: Af/7° (Ni2SiO4, olivine, 298.15 K) = - (1396.0 + 5.0) kJmol 1 and is in an excellent agreement with the value recommended by [95ROB/HEM]. A higher uncertainty than calculated was ascribed to A f //° (298.15 K) to overcome the problem of inconsistency arising during calculation of the decomposition temperature of Ni2Si04-olivine into NiO and SiO2-cristobalite. Using the selected value of A f //° it is possible to calculate the decomposition temperature to be (1820 + 5) K. It should be noted that our calculations of A f //° were carried out using the thermodynamic data for SiO2-quartz and not for SiO2-cristobalite. However, no thermodynamic data of SiO2(cristobalite) are available in the CODATA compilation [89COX/WAG] or in the TDB auxiliary data appendices used till now. The reason for it is probably the fact, that the thermodynamic data of SiO2(cristobalite) are less reliable. This issue was also discussed briefly by [87NEI]. Using the thermodynamic data of SiO2-cristobalite given by [95ROB/HEM] and the same procedure for the calculation of AfH°m (298.15 K) of Ni2SiO4-olivine as described above for SiO2-quartz, we have calculated: A f //° (298.15 K) = - (1401.15 + 3.40) kJ-moF1 and the proper decomposition temperature of Ni2SiO4-olivine (1820 + 5)°C. Different sets of values for low and high temperature SiO2(cristobalite) have been proposed by [98CHA]. The other value of the standard molar enthalpy of formation of Ni2SiO4-olivine proposed in the literature AfH°m (298.15 K) = - 1378.2 kJmol ' [61LEB/LEV] was not used by this review (see Appendix A).
V. 7 Group 14 compounds and complexes
V.7.2.1.1.5
243
Gibbs energy of formation
The Gibbs energy of formation of Ni2SiO4-olivine from metallic nickel, oxygen and silica according to Reaction (V.121) has been determined by equilibration of this silicate or metallic nickel and silica with a CO/CO2 atmosphere [61LEB/LEV], [63BUR/ABB], by determination of the oxygen fugacity over the mixture of nickel oxide-silica and the gas atmosphere of known CO2/H2-ratio [68CAM/ROE], or electrochemically by use of cells with a solid electrolyte [64TAY/SCH], [75LEV/GOL], [84ROG/BOR], [98ROG/KOZ], [2000ROG/KOZ], [87NEI] and [86JAC/KAL]. The results of the paper [86JAC/KAL] were not used for further evaluations, because they were presented only in graphical form. Thus, additional uncertainties in recovering the values from the original report were expected. A comparison of all results obtained by the determination of the Gibbs energy of formation of Ni2SiO4-olivine from metallic nickel, oxygen and silica, reported in the literature, is shown on Figure V-54. Although, different reactions were investigated, comparison was possible by recalculation, using the original data reported in the papers under consideration, to obtain the Gibbs energy of Reaction (V.121). Reaction (V.121) is the overall reaction in the solid electrolyte cells used for the determination of ArG° . Thus, it has been directly investigated in papers [64TAY/SCH], [75LEV/GOL], [84ROG/BOR], [98ROG/KOZ], [2000ROG/KOZ], [87NEI]. The study of Campbell and Roeder [68CAM/ROE] also directly refers to the equilibrium of Reaction (V.121). On the other hand, the studies of nickel orthosilicate reported by Lebedev and Levitskii [61LEB/LEV] and by Burdese et al. [63BUR/ABB] are concerned with the equilibrium of the reaction: 2Ni(cr) + SiO2(cr) + 2CO2(g) ^ Ni2SiO4(cr) + 2CO(g).
(V. 122)
The calculations of the Gibbs energy of Reaction (V.121) based on the Gibbs energy determinations of Reaction (V.I 22) requires the ArG° of Reaction (V.I 23) 2CO(g) + O2(g) ^ 2CO2(g).
(V. 123)
Using the ArG° values for Reactions (V.I22) and (V.I23) reported by [61LEB/LEV] and [63BUR/ABB], the Gibbs energies of Reaction (V.121) were calculated and are also shown on the diagram ArG° versus T, cf. Figure V-54, together with results of the solid electrolyte cell measurements. A comparison of the Gibbs energy of Reaction (V. 121) as obtained by different authors, Figure V-54, shows very good agreement between all solid electrolyte emfstudies and the oxygen fugacity study of Campbell and Roeder [68CAM/ROE]. The results of Lebedev and Levitskii [61LEB/LEV] and of Burdese et al. [63BUR/ABB] are significantly different. Therefore, they were disregarded by the present review and were not used for the non-linear curve fit. It should be mentioned that no significant difference between the value of the Gibbs energy of Reaction (V.121) was found when SiO2 cristobalite or quartz was used in solid electrolyte emf-studies reported by O'Neill
244
V Discussion of data selection fornickel
[87NEI]. The equation describing the fitted curve shown in Figure V-54 represents the Gibbs energy of Reaction (V. 121) in the temperature range 970 - 1770 K: A r G m (7) = (-(522.89819 ±8.94994) + (0.39674 ± 0.05714) T (0.02629 + 0.00700) T\nT) ld-mol"1. As already discussed in Section V.7.2.1.1.4, this equation was used in the present review for the calculation of the molar enthalpy of formation of Ni2SiO4-olivine at 298.15 K. The standard entropy change corresponding to Reaction (V.I 19) Ar5° (298.15 K) was calculated using S° (Ni2SiO4, olivine, 298.15 K) and the standard entropy of elemental nickel selected in this review. The other data i.e., the standard entropies of elemental silicon, oxygen and silica (quartz) at 298.15 K were taken from the NEA TDB auxiliary dataset. The Gibbs energy of formation of Ni2SiO4-olivine at 298.15 K was selected to be: AfG°m (Ni2SiO4, olivine, 298.15 K) = - (1288.44 ± 5.00) kJmol '. The selected value is in good agreement with AfG° = - (1289.0 ± 3.1) kJmol ' reported by [84ROB/HEM]. Figure V-54: Standard Gibbs energy of Reaction (V. 121) as a function of temperature. The data are taken from [68CAM/ROE] • , [64TAY/SCH] • , [75LEV/GOL] O, [84ROG/BOR] D, [98ROG/KOZ] O, [2000ROG/KOZ] A, [87NEI] +, [87NEI] SiO2(cr)X, [61LEB/LEV] A and [63BUR/ABB] T. The solid line refers to the nonlinear curve fit used by this review. The data of [61LEB/LEV] and [63 BUR/ABB] were disregarded.
V. 7 Group 14 compounds and complexes
245
V.7.3 Germanium compounds and complexes V.7.3.1 Solid nickel germanates V.7.3.1.1 V.7.3.1.1.1
Nickel orthogermanate Ni2GeO4(cr) Crystal structure of nickel orthogermanate
Ni - orthogermanate has a spinel structure and can easily be synthesised from the component oxides [66GME]. Germanates have long been regarded as suitable analogues for silicates, based on crystal chemistry systematics and, therefore, used for investigations to predict the strength of silicate spinel in the transition zone of the Earth [2001LAW/ZHA]. Even at high temperatures and pressures Ni2Ge04-spinel is a stable phase [66GME]. The cubic face-centred cell of the nickel orthogermanate, space group: Fd3m, has dimensions a = 8.221 A, Z = 8 (according to JCPDS-ICDD card No. 10 266). V.7.3.1.1.2
Enthalpy of formation from NiO(cr) and GeO2(cr)
Despite it being of geophysical interest, there are few thermodynamic data regarding nickel orthogermanate reported in the literature so far. Moreover, the reported data are incomplete. The direct determination of thermodynamic data for Ni2Ge04(cr) was performed by Navrotsky and Kleppa [68NAV/KLE] and by Navrotsky [71NAV]. These authors used oxide-melt solution calorimetry and determined the enthalpy of formation of Ni2Ge04-spinel from NiO(cr) and GeO2(quartz) at 965 K to be - ( 3 9 . 8 ± 1.7) kJ-mol"'. The enthalpy of formation values of Ni2GeC>4(cr) from the constituent oxides given by other authors [78KOT/MUL] and [72KUB] were taken from the original work of [68NAV/KLE] or [71NAV]. However, the value reported by [78KOT/MUL], - (18.1 ± 1.7) kJmol ', seems to be erroneously quoted. Lack of the thermal heat capacity function and the Gibbs energy of formation or entropy values for Ni2GeO4(cr) do not allow accurate calculation of thermodynamic data for this germanate at 298.15 K. Thus, the present review does not recommend any standard thermodynamic data for Ni2GeO4(cr).
V.8 Group 13 compounds and complexes V.8.1 Boron compounds and complexes Intermetallics are not treated in this review, therefore, thermodynamic data regarding borides are not included in this chapter.
V.8.1.1
Solid nickel borates
V.8.1.1.1
Crystal structure
Depending on the temperature, nickel borates can form crystals and glassy phases with different degrees of polymerisation, which are very often difficult to identify from the crystallographic point of view [66GME]. The numerous water-free and water-
246
V Discussion of data selection for nickel
containing nickel borate phases were characterised only by their stoichiometric composition and not by X-ray diffraction methods [66GME]. The X-ray diffraction pattern is known for Ni3B2O6 (JCPDS card No. 22-745, 26-1284 and 75-1809). V.8.1.1.2
Thermodynamic data for NiO-2B2O3(s)
So far no direct determination of thermodynamic data for water-free solid nickel borates has been reported in the literature. One of the reasons for this is probably the difficulty in obtaining well-defined crystalline phases. The only available data, published in a compilation of thermodynamic data for borate systems [74SLO/JON], are estimated values of the standard enthalpy of formation and the entropy for NiO2B 2 0 3 at 298.15 K. The value of A r //° (NiO2B 2 0 3 , s, 298.15 K) = - (62.8 ± 20.0) kJ-moF1 reported by [74SLO/JON] was obtained by a linear relationship between the enthalpy of reaction of the crystalline oxides and the ratio of cationic charge z to radius r. In view of the uncertainties with respect to the ionic radii selected, for (i) the z-rA versus A r //° correlation and for (ii) the Ni2+ ion, this value has not been accepted by this review. According to Latimer's method the standard entropies of predominantly ionic compounds can be obtained additively from values determined for the cationic and anionic constituents [51LAT]. Using the revised contribution of Ni2+ [93KUB/ALC] and the value for B 4 O^ derived by Richter and Vreuls [79RIC/VRE], the standard entropy was estimated to be S°m(NiO2B203, s, 298.15 K) = (129.9±6.0) J K 'moP 1 , which only slightly differs from the value given by [74SLO/JON] (137.6 ± 6.0). Generally this method is quite reliable, although the uncertainty would have to be increased to the 2o level (± 12.0). The problem, however, is that no well defined solid compound is known to which this entropy could be assigned. Thus it was not accepted by this review. V.8.1.1.3
Solubility of Ni(BO2)2-4H2O(s)
Shchigol [61SHC] studied the solubility of nickel orthoborate, Ni(BO2)2-4H2O(s). An average value of the solubility product of Ni(BO2)2-4H2O K°o= 2.16>2)2'4H2O(s) dissolves better in orthoboric acid solutions than in pure water. The formation of the nickel borate complex in solution was believed to occur according to the reaction: Ni(BO2)2-4H2O(s) + H3BO3(aq) ^ HNi(BO2)3(aq) + 5 H2O(1)
(V. 124)
The formula assigned to the proposed complex species was Ni(BO2)j . The dissociation constant of the aqueous nickel borate complex, according to Reaction (V.125), was calculated by Shchigol [61SHC], ^ 3 ,(V.125) = 2.82*10 9 mol3-dm 9. Ni(BO2 )~ ^ Ni2+ + 3 BO~
(V. 125)
An inspection of the experimental data revealed that they were misinterpreted, although Equation (V.I24) may be correct, see Appendix A. Consequently the reported value of the dissociation constant was not selected by the present reviewers.
This Page Intentionally Left Blank
Chapter VI
Discussion of auxiliary data selection VI.l Group 15 auxiliary species VI.1.1 Additional consistent auxiliary data for As compounds As was the case during preparation of an earlier TDB volume [92GRE/FUG], auxiliary data for arsenic compounds are not being critically assessed at this time, and several self consistent values from the U.S. National Bureau of Standards compilation have been accepted as an interim measure. The values of (// r a (7) - //° (298 K)) for As(cr) are from Herrick and Feber [68HER/FEB], although it is recognised that the value of S° (As, cr, 298.15 K) from the same paper is likely a better value than the TDB value, and differs from it by 0.6 J-K '-mol '. The high-temperature thermodynamic quantities for As4O6(g) are based on the work of Behrens and Rosenblatt, but are adjusted to account for the differences in the values of 5° (AS4O6, cubic, 298.15 K). The consistent values: A f #° = (As4O6, g, 298.15 K) = -(1196.25 ± 16.00) kJ-mol"1 and S° = (As4O6, g, 298.15 K) = (408.6 + 6.0) J-K ' mol', and the enthalpy and entropy differences of Behrens and Rosenblatt [72BEH/ROS] are used in the present review. The similar measurements of Murray et al. [74MUR/POT] on the equilibrium between gas-phase AS4O6 and the monoclinic solid are less well documented, and have not been taken into account here. The main contributions to uncertainties in the values for the values for As4O6(g) are from the uncertainties in the corresponding values for As4O6(cubic).
VI.2 Other auxiliary species VI.2.1 Auxiliary data for KCl(cr), KBr(cr) and KI(cr) The enthalpy of formation values for these solids in Glushko et al. [82GLU/GUR] are good, well-evaluated values that are very close to CODATA consistent. The value for A f //° (KC1, cr, 298.15 K) used in the first book in the CODATA series [87GAR/PAR] (but not included in the CODATA selected data [89COX/WAG]) is only 0.02 kJmol ' more positive than the value from [82GLU/GUR].
249
250
VI Discussion of auxiliary data selection
Detailed assessments of the aqueous enthalpy of solution values at 298.15 K for KCl(cr), KBr(cr) and KI(cr) were given in [82GLU/GUR]. In the present review, to obtain TDB-consistent values for the enthalpies of formation, these enthalpy of solution values are combined with the TDB enthalpy of formation values for K+ and CF, Br and I" for 298.15 K.
Table VI-1: Enthalpy of formation values for KCl(cr), KBr(cr), and KI(cr) at 298.15 K solid
A sln //° /kJ-mol-' [82GLU/GUR]
A r //° (KX, cr) / kJ-mor1 selected
KCl(cr)
(17.241 ±0.018)
-(436.461 ±0.129)
KBr(cr)
(19.780 ±0.080)
-(393.330 ±0.188)
KI(cr)
(20.230 + 0.100)
-(329.150 ±0.137)
The selected values of the enthalpies of formation differ by < 0.15 kJmol 1 from those tabulated by Glushko et al. [82GLU/GUR]. A TDB reassessment of the enthalpies of solution is not attempted, as it would involve analysis of a very large body of experimental data, and it is unlikely that the final result would be more accurate than the values from [82GLU/GUR]. It is possible that the enthalpy of solution values selected by Glushko et al. [82GLU/GUR] for these solids influenced the CODATA selections for the enthalpies of formation of the aqueous ionic species - the discussion in the CODATA volume [89COX/WAG] appears to lack the details that would be necessary to check this point. However, any extra uncertainties and inconsistencies from this source are expected to be small.
Part IV Appendices
This Page Intentionally Left Blank
Appendix A This appendix comprises discussions relating to a number of key publications, which contain experimental information cited in this review. These discussions are fundamental in explaining the accuracy of the data concerned and the interpretation of the experiments, but they are too lengthy or are related to too many different Sections to be included in the main text. The notation used in this appendix is consistent with that used in the present book, and not necessarily consistent with that used in the publication under discussion. [1883THO] Thomsen measured the heats of solution of NiCl2(cr) and NiCl2-6H2O(cr) in water. The final solution had a water to salt ratio of 400:1 (0.139 m). The reported heat of solution of the hexahydrate was 1.16 kcal-mol"1 (4.85 kJ-mor1) at 291.65 K. In the experiment, dilution was to 400 moles of H2O per mole of salt. Archer [99ARC] indicated that the NBS tables [82WAG/EVA] corrected the heat of solution value to / = 0 and 298.15 K, and obtained 0.138 kJ-mol"1 using - 188 J-K~'mor' for the heat capacity of reaction, and — 3.502 kJ-mor' for the integral heat of dilution. Based on the enthalpy of solution data of Manin and Korolev [95MAN/KOR] and the enthalpy of dilution (see Section V.4.1.3.2, Reference [59HAM] and the Appendix A discussion of [72AUF/CAR]), the enthalpy of dilution to / = 0 of the 1:400 solution of nickel chloride solution is here estimated as - (3.0 ± 1.0) kJ-mol"1. The total uncertainty from the heat of solution measurements themselves and the heat capacity correction is estimated as 1.5 kJ-moP1, and therefore the heat of solution of NiCl2-6H2O to infinite dilution in water at 298.15 K can be estimated as (0.64 ±1.80) kJ-mol"1. This leads to A f //° (NiCl2-6H2O, cr) = -(2103.33 + 2.01) kJ-mol"1. The reported heat of solution of the anhydrous salt was —19.17 kcal-mol"1 (- 80.21 kJmol"'), again at 291.65 K and dilution to 400 moles H2O per mole salt. If it is assumed that the heat of dilution is reasonably independent of the change in temperature, and that the heat capacity changes for the hydration reaction are not strong functions of temperature, the heat of hydration of NiCl2 to NiQ2-6H2O at 298.15 K can be estimated as -85.06 kJ-mor'. Thomsen and other early experimentalists found that NiCl2(cr) dissolved only with difficulty [66GME]. This might cause the reliability of the direct enthalpy of solution measurements of NiCl2(cr) to be cast in some in doubt. However, the heat of solution of the anhydrous solid is not inconsistent with the results of Manin and Korolev [95MAN/KOR].
253
254
A. Discussion of selected references
Based on the NEA TDB auxiliary value for A f //° (H2O, 1, 298.15 K), and the selected enthalpy of formation of NiCl2(cr) (-(304.90 ±0.11) kJ-mol"1), Thomsen's values would lead to an enthalpy of hydration o f - (2104.9+ 1.8) kJ-mor1, where the uncertainty (estimated here) is slightly lower than if the two heat of solution measurements were independent, because it is probable there were compensating errors. [1895FRA] In this early paper, results are reported for the conductance of nickel sulphate solutions for six concentrations between 9.8 * lO^M and 3.1 * 10~2M. The work appears to have been done carefully. Re-analysis is problematic because only two of the solutions were < 0.0025 M in nickel sulphate [79LEE/WHE]. Reanalysis using the solutions at the three lowest concentrations led to AT, = (221 ±13). The statistical deviation is undoubtedly smaller than the true uncertainty because of the limited data and the difficulties in properly calibrating the apparatus as discussed by Robinson and Stokes [59ROB/STO]. In the present review, we estimate the uncertainty in K{ at 25°C to be ±60. [1899FUN] Solubility measurements are reported for Ni(NO3)2-6H2O(cr) (-21°C to 41°C) and Ni(NO3)2-3H2O(cr) (58°C to 95°C) in water. The equilibrium mixtures below 0°C may also have contained the nonahydrate. The analyses on the solids indicated that good crystals of a tetrahydrate salt could not be obtained. [01MEU] Meusser measured the solubility of several hydrated forms of Ni(IO3)2, and of anhydrous Ni(IO3)2, between 0 and 90°C. Near 25°C, the least soluble compound was identified by the author as the P dihydrate. Later authors [73NAS/SHI], [97PRA/LAN] have concluded that the a-dihydrate of Meusser was actually an ill-defined amorphous solid, and Meusser's (3 form is what is now normally referred to as the dihydrate, Ni(IO3)2-2H2O. The solubility of this solid was measured at 8, 18, 50, 75 and 80°C. Equilibration times range from 24 h at the lowest temperature to 10 minutes at 80°C. The reported experimental (percent weight as Ni(IO3)2) values were converted to molalities, and the apparent solubility products were corrected to / = 0. These were plotted against (77K)~' to obtain values for logl0 K°o at 298.15 K and an average value of A r //° (A.I) over the temperature range of the measurements. Ni(IO3)2-2H2O ^ Ni2+ + 2 I O ' + 2H2O(1)
(A. 1)
If only the values to 50°C are included, log 10 £° 0 ((A.l), Ni(IO3)2-2H2O, 298.15 K) = - (5.17 ±0.11) and A r //°(A.l) = (21.4 ±6.1) kJ-mol"1, whereas, if the points at 75 and 80°C are included, log10 A:°0((A.l), Ni(IO3)2-2H2O, 298.15 K) = - (5.17 ± 0.08) and A r //° (A.I) = (22.5 ± 2.2) kJ-mol"1.
A. Discussion of selected references
255
A similar analysis based on the solubility measurements (30°C to 90°C) for a yellow anhydrous nickel iodate, probably the p—Ni(IO3)2 [73NAS/SHI], leads to log 1 0 £° 0 ((A.l), Ni(IO3)2-2H2O, 298.15 K) = -(4.427 ±0.015) and A r //°(A.l) = - (7.31 + 0.69) kJ-mol'1. This appears to be the origin of the values listed for Ni(IO3)2 in the US NBS tables [82WAG/EVA]. In the present review, greater uncertainties (+ 0.1 in log10 A^°o and 4.0 in A r // m ) are estimated because of the temperature extrapolation and the difficulties in establishing the purity of the solid at pseudo-equilibrium. The tetrahydrates are difficult to purify, and convert easily to the dihydrate in the presence of liquid water [73NAS/SHI]. This agrees with observation that the measured solubilities in water are greater than those of the dihydrate at all temperatures [01MEU]. However, the solubilities reported for the tetrahydrate are not used further in the present review. [06THO] See the entry for [1883THO]. [07WEI] The solubility of crystalline (millerite) as well as precipitated NiS in pure water at 291.15 K was determined by means of conductivity measurements. The solubility of millerite was found to be 1.629 x 10"5 mol-drrf3 whereas that of a precipitated product amounts to 3.987 x 10~5 mol-dnf3. No experimental details are given. [09BRU/ZAW] Based on the enthalpy of formation of NiS(s) given by Thomsen [1883THO] ( A f # ° (NiS, cr, 298.15 K) = -72.76 kJmol~') the solubility product was calculated, leading to log10 K°o = - 24.15 at T = 298.15 K. The entropy of formation for NiS(s) was assumed to be zero, which is likewise confirmed by the thermodynamic data of the present assessment. The authors used values for the second dissociation constant of H2S(aq), H2S(aq) ^ 2H+ + S2" \ogl0K°= -23.26 and the standard potential of Ni2+, E^.M2t = 0.219 V, which are no longer considered reliable. Employing the currently accepted values, logl0 K° - - 25.99 [92GRE/FUG] and E ^ . = 0.2372 V (this assessment), the recalculated solubility product amounts to log10 K°o = — 27.5. [10BOR] The phase relations in the system Ni - S in the composition range from 0 to 31 wt% sulphur were investigated by thermal (cooling curves) and microscopic techniques. The eutectic between metallic nickel and Ni3S2(cr) was found to lie at 918 K and 21.5
256
A. Discussion of selected references
wt% S. Moreover, it was reported that the high-temperature modification of Ni3S2(cr) melts incongruently at a temperature around 1083 K. [11AGE/VAL] Seemingly the solubility of NiCO3 was studied by Ageno and Valla for the first time, however, in their preliminary communication the authors described no attempt to characterise the chemical and physical state of the samples investigated. Regardless of this uncertainty the result was ascribed to anhydrous NiCO3 and accepted by thermodynamic compilations for calculation of AfG°(NiCO3, cr, 298.15 K) [35KEL/AND], [52LAT]. Ageno and Valla equilibrated their solid phase with water under a carbon dioxide containing atmosphere in a closed vessel at 25°C. The Ni(II) concentration and the partial pressure of CO2 were determined experimentally. Column 2 of their table on carbonate solubilities is labelled incorrectly. Under the conditions prevailing in their study an elementary stoichiometric consideration predicts: 2[Ni2+] = [HCO3"]. Consequently it has been assumed that the numerical values listed in the solubility table under NiCC>3/2 have to be ascribed to 2 [Ni(II)] instead. [14THI/GES] Nickel sulphide was prepared by adding Na2S or (NH4)2S to aqueous solutions of nickel chloride (typically 0.04 mol-dm~3) at ambient (around 298 K) and elevated temperatures, respectively. In addition, NiS was precipitated by reaction of H2S with NiCl2 solutions containing sodium acetate (typically 0.1 mol-dnT3) at ambient as well as elevated temperatures. In some cases the precipitated product was aged in the mother liquor for 10 weeks. The solubility of NiS at room temperature was investigated by treating both freshly prepared and aged solid phases with HC1 solutions (2 moldm"3) saturated with H2S. The amount of dissolved nickel(II) was determined by means of electrolysis. Three different fractions of the solid sulphide could be discerned with respect to their solubility behaviour. Fraction I dissolved rapidly in hydrochloric acid at room temperature. The solubility was estimated to be 0.1 mol'dnf3 at pH = 2. This high value indicates that fraction I corresponds to an amorphous product. Fraction II showed a solubility in H2S saturated 2 M HC1 of approximately 1 mmol-dirf3 at room temperature. As the reaction time was in the range from 0.5 to 42 hours, equilibrium was certainly not attained. Finally, the solubility of fraction III in H2S saturated 2 M HC1 was reported to be approximately 0.02 mmol-dnT3 at room temperature. At temperatures around the boiling point of the hydrochloric acid solution the solubility amounted to 0.22 mmoldm 3 with the reaction time ranging from 0.5 to 2 hours. Again, this time is not expected to be sufficient to attain equilibrium between the aqueous and the solid phase. According to
A. Discussion of selected references
257
Donges [47DON] the hexagonal NiAs-type modification of nickel sulphide is obtained when the precipitation is performed slowly by using H2S. Hence, it can be concluded that fractions II and III correspond to crystalline NiS of different particle sizes depending on the conditions of precipitation and ageing. The different nickel(II) concentrations found by dissolving these fractions may be caused by the variation of the dissolution rate with the particle size. However, thermodynamic solubility constants cannot be derived from these experimental studies. As none of these nickel sulphide phases was characterised by X-ray diffraction analysis the original notation I, II, and III-NiS has been retained in this review. [16DER/YNG] Using a static method, water vapour pressures were determined as a function of temperature above mixtures of the tetrahydrate and hexahydrate (1.8°C to 36.25°C) and the dihydrate and tetrahydrate (25.15°C to 79.06°C). The work established the existence of the tetrahydrate as a separate phase, and the temperature of the hexahydrate/tetrahydrate conversion at equilibrium with water vapour was reported as 36.25°C. The reported water vapour pressure above the tetrahydrate/hexahydrate mixtures at 25°C was (1.40 + 0.05) kPa, where the uncertainty is estimated in the present review. [18ALM] Almqvist suspended freshly precipitated nickel hydroxide in water for 24h, filtered off the residue, and determined the Ni(II) content of the clear solution. The value of the Ni(OH)2 solubility thus obtained ([Ni(II)] = 1.37 x 10"4 mol-dm"3) implies that, owing to the mass and charge balance of Equation (A.2), either the saturated solution is strongly alkaline or the uncharged hydroxo complex Ni(OH)2 is rather stable. As neither phenomenon has been observed, Almqvist's solubility is too high, as clearly documented in Figure 5 of [97MAT/RAI]. Therefore, it has not been taken into account in the calculation of a value of logl0 Ks 0 . Probably the high solubility observed by Almqvist was due to colloidal Ni(OH)2 particles which passed through the filter. [Ni(II)] = [Ni2+] + [NiOH+] + [Ni(OH)2] [OH] = 2 [Ni2+] + [NiOH+]
(A.2)
[21BER/CRU] Berger and Crut investigated the equilibrium of the reaction, NiCl2(cr)+H2(g) ^ Ni(cr) + 2 HCl(g)
(A.3)
using a static method. When the experimental data depicted in Figure V-16 and listed in Table A-l were compared with those of [53BUS/GIA], [53SAN], [54SHC/TOL] they did not seem to be of adequate quality for use in a third law analysis.
258
A. Discussion of selected references
Table A-l: Equilibrium constants of Reaction (A.3) [21BER/CRU], [24CRU]. 77 K
583 613 668 718
logioCK,, / atm)
logio(AT,, / bar)
-2.214 -1.627 - 0.784 -0.0556
-2.2083 -1.6213 -0.7783 - 0.0499
-Rrin(A:,,/bar)/kJmol ' 24.6475 19.0269 9.9532 0.6857
[23VIL/BEN] The solubility of nickel sulphate was determined by the floating equilibrium method. The following results were obtained at 25°C: 40.469 g NiSO4 per lOOg H2O and 40.720 g NiSO4 per lOOg H2O (floating equilibrium), 40.417 g NiSO4 per lOOg H2O (gravimetry). [24CRU] Crut described the method applied and the experiments carried out previously [21BER/CRU] in more detail. The reported constants are the same as those given in [21BER/CRU]. Only one chloride data set at 668 K is not identical. This was considered to be a misprint, as the other Kp values were exactly the same in [24CRU] as in [21BER/CRU]. The reported measurements were primarily for reduction of the chloride. Only very limited measurements for the bromide system were carried out (at 718 and 848 K). Also, see the discussion of [24CRU2]. [24CRU2] Crut [24CRU2] reported the heat of solution of anhydrous and hydrated NiBr2 as - 19.9 kcalmol 1 (- 83.26 kJ-mol"1) and 0 kcalmol"', respectively. The paper also reports a somewhat different derived value for the heat of hydration of the anhydrous salt as - 19 kcalmol"1 (- 79.50 kJ-mol"'). This value is consistent with the heats of solution of anhydrous and hydrated NiBr2, - 18.9 kcal-mol"1 (- 79.08 kJ-mor1) and 0 kcalmor', probably from the same experiments, reported in an earlier paper by the same author [24CRU]. The extent of hydration of the hydrated salt is not reported, nor is the salt purity, the final solution concentrations nor the measurement temperature. Because nickel bromide prepared at high temperatures dissolved slowly in water at room temperature, a sample of anhydrous NiBr2(cr) was prepared in a manner to enhance its rate of dissolution in water. The dissolution rate difference may reflect a structural difference that could have been affected the value determined in the heat of solution measurements. Bichowsky and Rossini [36BIC/ROS], in the forerunner to the US NBS Tables, estimated AfH°m (NiBr2-3H2O, cr) using the value - 18.8 kcal-moP1 (-78.62 kJ-mor') for the enthalpy of hydration of NiBr2(cr) to NiBr2-3H2O(cr) from Crut's 1923
A. Discussion of selected references
259
dissertation (unavailable to the current reviewers). This value corresponds approximately to the values reported in [24CRU]. The authors appear to have based their identification of the solid as NiBr2-3H2O on work reported in the thesis or elsewhere. There is a similar, but smaller discrepancy in the values for the cobalt bromides from the same two papers [24CRU], [24CRU2]. The analysis and value for A f //° (NiBr2-3H2O, cr) from Bichowsky and Rossini [36BIC/ROS] is the origin of the value for the same solid (at 298.15 K) in the 1952 version of the US NBS tables [52ROS/WAG] (and probably of the slightly different value in the 1982 tables [82 WAG/EVA]). [24MOS/BEH] The solubility of NiS at 293.15 K in H2S saturated 1 M H2SO4 was reported to be 1.04 x 10^* mol-dm"3. After an equilibration time of 10 hours the dissolved amount of nickel(II) was determined gravimetrically. The solid phase was precipitated at elevated temperatures (333 - 343 K) by adding H2S to an aqueous solution of NiSO4-7H2O. No characterisation of the solid product was performed. The reaction time of 10 hours seems to be too short for equilibrium between the solid phase and the aqueous solution to be attained. [24TAN] The solubilities of the crystalline nickel sulphate hydrates were determined to calculate the respective temperature coefficients. For 25°C 40.55 and 40.26 g NiSO4 per 100 g H2O were given. [25BRI] Britton titrated a nickel chloride solution with sodium hydroxide and employed a hydrogen electrode to follow the pH change. As only 1.66 moles of NaOH were needed to complete the reaction it is quite clear that a basic chloride had at least been coprecipitated with the pure nickel hydroxide. Thus, it is doubtful whether Britton's experimental data refer to Ni(0H) 2 at all. In addition Jena and Prasad have previously reported that Britton miscalculated the solubility product [56JEN/PRA]. Re-evaluation of the buffer region of Britton's Figure 2 has shown that logl0 ^ ° 0 = - (16.3 + 0.3) is consistent with the experimental data, and has been used instead of the value originally given (- 18.06) for the respective entry in Figure V-l 1 and Table V-6. [25JEL/ZAK] The partial pressure of sulphur in equilibrium with nickel sulphide was determined at 788.15 and 903 K. The ratio />(H2S)//?(H2) in the gas phase was measured by means of chemical analysis, resulting in sulphur partial pressures by taking into account the wellknown equilibrium: H 2 S ( g ) ^ H 2 ( g ) + '/2S2(g).
260
A. Discussion of selected references
The solid phase, NiS, was prepared by reaction of aqueous solutions of NiCl2 with ( N H ^ S . The product was characterised by chemical analysis, but no X-ray diffraction analysis was performed. As the results (logio[p(S2)/bar] = - 9.0 for the proposed equilibrium between NiS and Ni-iSf, at 788.15 K and logio[p(S2)/bar] = - 10.0 for the proposed equilibrium between Ni and Pi-Ni3S2 at 903 K) do not coincide with any other literature data, this investigation is not considered further in the present assessment. [25WIJ] In the course of a study of nickel ammine complexes, in two experiments turbidity occurred, which was interpreted as an indication of formation of Ni(OH)2. When the solubility product (log10 Ks 0 = - 13.8, 25°C) is extrapolated to / = 0 by means of the Davies equation the following constants were obtained ( l o g m ^ ° 0 = -14.64, log10 ^" 0 = 13.36) [62DAV]. These values are in line with others which refer to freshly precipitated probably amorphous Ni(0H) 2 , see Table V-6 [38OKA], [54FEI/HAR]. [26JEL/ULO] The reduction equilibrium of nickel(II) chloride was studied by a flow method. The results were probably flawed by vaporisation of NiCl2, thus they were not subjected to a third law analysis. The experimental data depicted in Figure V-16 are listed in Table A-2.
Table A-2: Equilibrium constants Of NlCl 2 (cr) + H 2 (g) ^ 77 K
\og^Kp 1 atm)
\o%w(Kp 1 bar)
573
- 2.574
-2.5683
613
-1.398
-1.3923
723
-0.136
-0.1303
Ni(cr) + 2 HCl(g). -Rrin(^/bar)/kJ-mol ' 28.1739 16.6060 1.8033
[26JEL/ULO2] The flow measurements for hydrogen reduction of NiBr2(cr) (683 K to 833 K) and NiI2(cr) (783 K to 983 K) suffer from problems similar to those discussed for the measurements on NiCl2(cr) [26JEL/ULO]. Furthermore, above 700 K, Nil2 has been reported to be subject to thermal decomposition [74WOR/COW]. The results of Jellinek and Uloth are not used in the present assessment. [27JON/WIL] Powdered NiS was heated with S at 170°C for 30 to 40 h. The empirical formula of the product was NiS2.ig. The X-ray powder pattern was quite similar to the pattern for FeS2.
A. Discussion of selected references
261
A cubic lattice constant a^ = 5.74 A was determined. [28JEL/RUD] The equilibrium of the reaction: NiF2(cr) + H2(g) ^ Ni(cr) + 2HF(g)
(A.4)
was studied from 573 K. to 773 K. The measured equilibrium constants varied with flow rate. Therefore, although the derived value of A f //° (NiF2, cr, 298.15 K) is consistent with those from other types of experiments [67RUD/DEV], the values of Jellinek and Rudat are not used in the assessment of A f //° (NiF2, cr, 298.15 K) in the present review. [28MUR] Murata reviewed older measurements with - 0.597 V < £°(NiSO4(aq) | Ni) < - 0.183 V and concluded that the state of the nickel electrode was responsible for this conspicuous disagreement [1894NEU], [04EUL], [04MUT/FRA], [04SIE], [09SCH], [09SCH2], [10SCH], [18SMI/LOB]. In his own work he used powdered nickel which was reduced with hydrogen directly in the following cell Ni | NiSO4 | KCl(sat.) | KCl(0.1 M), Hg2Cl21 Hg. Murata's data have been recalculated with the standard potential of the Hg2Cl21 Hg electrode ((0.2681 ± 0.0026) V, [89COX/WAG]) and the chloride activity of a 0.1 M KC1 solution (aKC1 = 0.077) [59ROB/STO]. The activity of Ni2+ in the solution was calculated using the SIT model. The corrected values are is°(Ni2+ | Ni) = - (0.246 ± 0.003) V at 25°C and -(0.245 + 0.005) V at 18°C. As the measurements were carried out under hydrogen, which according to other authors interferes, and no correction for the liquid junction was applied, Haring and Vanden Bosche's [29HAR/BOS] result has been preferred to Murata's. [29HAR/BOS] Haring and Vanden Bosche reviewed the literature on the standard potential of Ni2+ | Ni, discussed Murata's work and included the following papers into the discussion [00KUS], [01PFA], [05COF/FOR], [08THO/SAG], [26GLA]. They pointed out that in many of the numerous earlier studies, which led to results with £"°(Ni2+ | Ni) ranging from - 0.621 to - 0.178 V, appropriate corrections for activity coefficients and liquid junction potentials were lacking. Moreover, the absence of interfering gases such as oxygen or hydrogen and a constant temperature were not always ensured. These authors carried out most careful measurements using the cell: Ni | NiSO4(w) | Hg2SO41 Hg which involves no liquid junction. Finely divided, strain-free, electroplated nickel was used and precautions were taken to exclude oxygen and hydrogen completely. The
262
A. Discussion of selected references
measured potentials were constant within 1-5 mV for 18 days. The cell behaved completely reversibly. The reversibility of the electrode was demonstrated by the Nernst behaviour of the measured potentials with the activity of NiSC>4, see also Figure V-2. Moreover, after anodic and cathodic polarisation and short circuiting, which produced a temporary disturbance, the equilibrium value was quickly regained. The results are listed in Table A-3. It was necessary to recalculate the standard potential (£°(Ni2+| Ni, 298.15 K) = -(0.231 ± 0.002) V [29HAR/BOS]), because the SIT model leads to slightly different activity coefficients and the standard potential of the Hg|Hg2SO4 electrode had to be changed to the value consistent with the NEA-TDB auxiliary data. Using the function In y± = F (m (NiSO4)) given in the caption of Figure A-5 (page 295 under the discussion of [59ROB/STO]) for the recalculation of £°(Ni2+| Ni, 298.15 K) results in a value which agrees within 0.1 mV with the one derived using the SIT. Af G°m (Ni2+) selected in this review has been based on the latter value. An attempt by these authors to measure the standard potential of the nickel electrode with the similar cell: Ni | NiCl2(0.05 mol-kg"1) | Hg2Cl21 Hg was unsuccessful. Obviously nickel corroded in the chloride medium.
Table A-3: Electrode potential of the cell: Ni | NiSO4 (m) | Hg2SO4 | Hg [29HAR/BOS] m (NiSO4)
EIV
1
(mol-kg- )
£°/V
£°/V
£°/V
[29HAR/BOS]
SIT
[59ROB/STO]
0.0504
0.967
0.852
0.84848
0.84794
0.0504
0.967
0.851
0.84848
0.84794
0.0504
0.969
0.853
0.85048
0.84994
0.0506
0.968
0.852
0.84953
0.84900
0.0506
0.967
0.851
0.84853
0.84800
0.0507
0.971
0.855
0.85256
0.85204
0.0508
0.968
0.852
0.84959
0.84907
0.0509
0.967
0.851
0.84862
0.84811
0.0511
0.968
0.852
0.84968
0.84917
0.0511
0.967
0.851
0.84868
0.84817
0.0512
0.967
0.851
0.84871
0.84821
0.0515
0.970
0.854
0.85180
0.85131
(Continued on next page)
263
A. Discussion of selected references
Table A-3 (continued) m (NiSO4)
E/V
1
(mol-kg )
£°/V
E°/V
£°/V
[29HAR/BOS]
SIT
[59ROB/STO]
0.0515
0.968
0.852
0.84980
0.84931
0.0516
0.968
0.852
0.84982
0.84934
0.0525
0.971
0.854
0.85308
0.85263
0.1017
0.958
0.852
0.84967
0.85007
0.1018
0.958
0.852
0.84969
0.85008
0.1020
0.958
0.852
0.84971
0.85011
0.1024
0.958
0.852
0.84977
0.85017
0.1052
0.958
0.852
0.85016
0.85056
0.1503
0.953
0.852
0.85023
0.85046
0.1504
0.954
0.853
0.85124
0.85147
0.1519
0.953
0.852
0.85038
0.85060
0.1543
0.953
0.852
0.85060
0.85080
0.1543
0.952
0.851
0.84960
0.84980
0.1617
0.954
0.852
0.85226
0.85241
[30MUL/LUB] Miiller and Luber determined a single value of the solubility of nickel carbonate hydrate, which they believed to be NiCO3-6H2O, at p(CO2) — 50 atm, without giving the temperature of equilibration explicitly [30MUL/LUB]. When it is tentatively assumed that the experiment was carried out at 25°C, Muller and Luber's result agrees fairly well with results from [11AGE/VAL] and [2001GAM/PRE], [2002WAL/PRE] indicating that the same nickel carbonate phases were used in these studies. [31KOL] The solubility product of NiS was calculated by taking into account the solubility of nickel(II) sulphide in 1 M H2SO4 (T = 293.15 K), determined by Moser and Behr [24MOS/BEH], leading to log10 K°o = - 26.96. The author used a value for the second dissociation constant of H2S(aq), H2S(aq) ^
2H+ + S2~,
log)0 K°2 = — 21.97 which is no longer considered to be correct. Employing the currently accepted value, logl0 K° = - 25.99 [92GRE/FUG], the solubility product can be recalculated, resulting in log10 K°o = - 30.98.
264
A. Discussion of selected references
[32MAS] Potentiometric measurements of the nickel(II) - cyanide system are described. The addition of Ni(CN)2, NiSO4 or Ni(NO3)2 to sodium cyanide solution resulted in the formation of a single nickel(II) containing species, Ni(CN)J~. Its formation constant at 25°C and in 0.78 M NaCN solution was found to be logl0 /?4 = (11.83 + 0.17). The author used nickel metal as the working electrode. Since the electrode reactions involving nickel(II) are not reversible [50HUM/KOL], the reported formation constant is not reliable. [32MON/DAV] This paper provides a calculated value of 250 for K\ for nickel sulphate complexation based on the earlier measurements of Franke [1895FRA] (at the two lowest molarities for which results were reported, 9.768 x 104 and 19.53 * 104 M) at 25°C. The value of Ao was based on estimates of the molar ionic conductivities of sulphate (158 S-cm2-mor'), and for Ni2+ (112.2 Scm 2 mor'). There are no new experimental data. [34BRI/OSS] This paper reports the results of a dialysis study using aqueous solutions 1 M in ammonium sulphate and 0.05 M in various other metal sulphates. The sparse results for the nickel system are interpreted in terms of formation of Ni2(SO4)4~. There are insufficient details to derive useful complexation constants from the data. Also see [41KIS/ACS]. [34COL2] Colombier measured the potential of NiSO4(0.5 M) | Ni against a calomel reference electrode at 20°C. When the electrolyte was appropriately degassed and kept air-free the same equilibrium potentials were found for massive or powdered nickel, the latter having been electrolytically deposited or prepared by hydrogen reduction of NiO. As the author presented no details about the experimental data, the reference potential selected, how the liquid junction potential and the activity coefficients were accounted for, the final result, £°(Ni2+ | Ni, 293.15 K) = - (0.227 ± 0.002) V, cannot be recalculated and was not being considered any further. [34SIE/SCH] Sieverts and Schreiner [34SIE/SCH] carried out solubility measurements from - 27.8°C to 119.8°C. They found Ni(NO3)2-9H2O below - 11°C and Ni(NO3)2-6H2O to 50°C. They also reported that both the tetrahydrate (55 to 80°C) and dihydrate (85 to 120°C) were stable phases in the Ni(NO3)2 - H2O system, and could be formed at equilibrium at 25 °C in the Ni(NO3)2 - H2O - HNO3 system at high nitric acid mole fractions. The solution compositions at saturation do not differ markedly from those of Funk [1899FUN] except at the lowest temperatures.
A. Discussion of selected references
265
[35KEE/CLA] This paper reports the measurements of the specific heat capacity of nickel in the temperature range from 1.1 to 19.0 K. The hypothesis was advanced that the deviation from Debye's T3 law is connected with the energy of electron conduction. [36BIL/VOI] The decomposition pressure of NiS2(cr) was measured as a function of temperature by means of a tensimetric method. The solid phase, NiS2(cr), was prepared by solid state reaction between NiS(cr) and sulphur. The composition of the product was checked by chemical analysis (no X-ray diffraction analysis was performed). Furthermore, the phase transformation of millerite into its high-temperature modification was studied by an early DTA-like method. The transition temperature was reported to be 669 K and the heat of transition was estimated to be in the range between 2.1 and 2.9 kJ-mor'. Millerite was precipitated at room temperature by bubbling H2S gas through an ammonia-containing aqueous solution of NiCl2. The solid compound was characterised by chemical analysis. [36BRO/WIL] The mean heat capacities from - 80 to 120°C have been determined for nickel and a few other metals. The errors were estimated to amount to ± 0.1%. A discussion of the theoretical basis for the equations describing the temperature dependence of the heat capacity is given together with some calculations of heat capacities using data which involve no measurements of heat e.g., the expansion coefficient, compressibility etc. [36JOB] This is a spectrophotometric study on (among others) nickel(II) bromide complexes formed in aqueous HBr solution. The author identified two bromo complexes, NiBr2(aq) and NiBr42". The former species is reported to be dominant in 6 — 7 M, the latter above 10 M HBr. Since the ionic strength could not be maintained constant, the activity of bromide, instead of its concentration, was taken into account in the mass action expressions. These "semi-thermodynamic" constants include the ratios 7NiBr /yNiBr ? 2 (where n — 2 or 4), which are assumed to be nearly constant with increasing HBr concentrations. The reported constants are logl0 /?2° = - 3.24 and logl0 /3° = - 8.12. The graphical data presented in [36JOB] were later re-evaluated in [72RET/HUM], and were found consistent with the formation of NiBr+ and NiBr2(aq) complexes, if the activity of water is also taken into account in the mass action expressions. In [72RET/HUM] the recalculated value of logl0 /3° is also reported (log l0 /?° = -3.7). Since negatively charged halogeno complexes of nickel(II) have not been identified in aqueous solutions neither with fluoride nor with chloride, only the latter value is considered in this review.
266
A. Discussion of selected references
[36LOT/FEI] Lotmar and Feitknecht investigated variations in ionic distances of bivalent basic metal halogenides with layered structures of the Cdl2 type. In this context they needed the unit cell dimensions of nickel hydroxide as well as cobalt hydroxide. [36TRA/SCH] Trapeznikova et al. studied the heat capacity of NiCl2 at temperatures between 13 and 130 K. They used a glass calorimeter and a platinum thermometer heater. The discrepancy between their results and those of Busey and Giauque probably indicates that their glass calorimeter and their thermometer did not attain thermal equilibrium [52BUS/GIA]. [37BEL] Measurements are reported for dissociation pressures of nickel chloride hexadeuterate to tetradeuterate. The transition temperatures for dehydration of the hexahydrate and the hexadeuterate at 1 atm are reported as 36.25 and 35.9°C, respectively. It was noted that the solvent pressure ratio over the Ni-Cl solvates (p^o'.p^o) rises with increasing temperature from 25 to 35°C. There are no experimental numbers reported for the hexahydrate/tetrahydrate equilibrium. [37DOM] The equilibrium of the reaction: NiF2(cr) + H2O(g) ^ NiO(cr) + 2HF(g)
(A.5)
was studied from 773 K to 973 K. The measured equilibrium constants showed only a small variation with flow (at higher flow rates). In the present review, values of A r //°((A.5), 298.15 K) are calculated using the reported equilibrium constants, selected values of S°m and Cpm(T) from Tables III-l and III-3 for NiF2(cr) and NiO(cr), and the tabulated values for H2O(g) and HF(g) in the CODATA compilation [89COX/WAG]. Then, using A(H°m values from Tables III-l and IV-1 for NiO(cr), H2O(g), and HF(g), an average value, A f //° (NiF2, cr, 298.15 K) = -(666.13 +0.64) kJ-mor1, is calculated. The drift in the calculated value depends only slightly on the measurement temperature (< 1 kJ-mor1 for the 200 K range). The value calculated in the present review is 18 kJ-moF1 more negative than the value recalculated by Rudzitis et al. from the same data, though the values of A r // m (298 K) are in good agreement. Only part of the discrepancy can be explained as resulting from use of different auxiliary data (primarily the value for A f //° (HF, g, 298.15 K)). [37SAN] Sano reported the values for the vapour pressure of water over (initial) Ni(NO3)2-6H2O from 300 K to 326 K using a Jackson spring manometer. The product of the dehydration
A. Discussion of selected references
267
reaction was assumed to be Ni(NO3)2-3H2O, and the plot oi-\ogwpno against 1/7"is linear, but no estimate of the uncertainties is given. In this paper, the thermodynamic parameters for the dehydration reaction were calculated incorrectly from the data (the equilibrium constant depends on p^ 0 , not pH 0 as assumed by the author). No evidence was presented that the dehydration product was the trihydrate rather than the tetrahydrate, the dihydrate or a non-equilibrium mixture of hydrated salts. [37SAN2] See [53SAN]. [38OKA] Three titrations of pure as well as NaCl- or Na2SO4-containing nickel nitrate solutions with sodium hydroxide were carried out at 25°C. From the pH measured at each point taken on the titration curve (0.07 < «(NaOH) / «(Ni(NO3)2) < 1.62) the solubility product was calculated and extrapolated to zero ionic strength. A mean value of log10 K°o = - 14.50 was obtained and has been taken as a basis for the value listed in Figure V-l 1 and Table V-6, respectively. [38SMU] Smurov precipitated nickel carbonate by mixing aqueous sodium carbonate and nickel chloride solutions. The solid phase obtained was washed free of chloride and used for the solubility study, but no attempt was made to characterise its stoichiometry and/or its crystal structure. The precipitate was equilibrated with water in a temperature range between 278.15 and 353.15 K. For each temperature the partial pressure of carbon dioxide was varied from 0.0005 atm to pco + pu 0 = 1 atm. Smurov's data have been extrapolated to zero ionic strength under the assumption that the solid phase had been hellyerite. As shown in Figure A-l the solubility constants thus derived disagree not only with the values of hellyerite but also with those of gaspeite. The temperature dependence is more pronounced than is usually observed with neutral carbonates. This probably indicates that a basic nickel carbonate was investigated instead. However, without stoichiometric and structural evidence, no thermodynamic quantities can be assigned to it.
268
A. Discussion of selected references
Figure A-l: Solubility of precipitated nickel carbonate. • Experimental data [38SMU]; solid curve: experimental data fitted to log10 K s0 = A + BIT + ClnT; dotted curve: solubility constant of hellyerite, NiC(V5.5H2O; dash curve: solubility constant of gaspeite, NiCO3.
[38SYK/WIL] The specific heat capacity - temperature dependence for four different types of nickel have been determined and the effects of the method of manufacture, the heat treatment and the chemical composition have been ascertained. The results, together with those obtained by previous investigators were reviewed and an attempt was made to evaluate the most probable C°m versus T curve for nickel in the temperature range 100 to 600°C. [39ROH] After an equilibration period of four days at 25°C the solubility of NiSO4-7H2O was found to be 41.2 g NiSO4 per lOOg H2O. The 6 values taken from [23VIL/BEN], [24TAN], [39ROH], and cited in the respective Appendix A entries, were used to calculate mean and standard deviation (2a) of the NiSO4-7H2O solubility in water (see Section V.2.1.3.1).
A. Discussion of selected references
269
[39SCH/FOR] The high-temperature form of nickel(II) sulphide, NiS, was treated with hydrogen gas at three different temperatures, viz. 673, 773 and 873 K. The H2S(g) content in the gas phase was measured at constant temperature as a function of the sulphur content of the solid phases. Equilibrium compositions of the gas phase for univariant equilibria, involving NiS, Ni3S2 and Ni6S5 (according to St0len et al. [94STO/FJE] the composition of this phase is rather Ni7S6 (high-temperature modification) and Ni9Sg (lowtemperature modification), respectively), are given. No experimental details are reported. [39SIM/KNA] Results of study of the decomposition of nickel sulphate heptahydrate through nickel sulphate hexahydrate and nickel sulphate tetrahydrate at 0.015 bar (11 torr) are presented. X-ray diffraction patterns are presented for some of the different solids. Detailed measurement results were not tabulated, but the graphical presentation and the reported thermodynamic quantities for the dehydration reactions are in reasonable agreement with results reported in the later study of Kohler and Zaske [64KOH/ZAS]. [40BEL] This paper contains measurements of the water vapour pressures above samples of NiSO4-7H2O (at 20, 25 and 30°C) and NiSO4-7D2O (at 20 and 25°C). The values for the NiSO4-7H2O are very similar to those obtained by Schumb [23SCH], Bonnell and Burridge [35BON/BUR] and [66STO/ARC]. The vapour pressures above the deuterate were found to be about 80% of the vapour pressures above the ordinary hydrate. The enthalpy of removal of D2O from NiSO4-7D2O was found to be approximately 11 kJ-mol"1 less favourable than dehydration of NiSO4-7H2O. Equations for dissociation pressures of nickel chloride hexahydrate to tetrahydrate and the nickel chloride hexadeuterate to tetradeuterate are presented. The equation for the deuterate reaction is based on the experimental vapour pressures reported in [37BEL]. The equation for the hexahydrate/tetrahydrate equilibrium matches the experimental points from Derby and Yngve [16DER/YNG] so well that it raises the question of whether the reported equation is from a fit to the earlier data [16DER/YNG], rather than based on new experiments. The transition temperature reported in [37BEL] matches that in [16DER/YNG] exactly (within 0.01 K). [41KIS/ACS] This paper discusses problems with the dialysis method as used by Brintzinger and coworkers (cf. [34BRI/OSS]).
270
A. Discussion of selected references
[41STO/GIA] Results were reported for careful heat capacity and magnetic susceptibility measurements carried out between 1 and 15 K on a sample of "NiSO4-7H2O". The results indicated a maximum in the heat capacity for 1.8 K. In a later paper [66STO/ARC], the results were reinterpreted and combined with results for higher temperatures to determine the third-law entropy of the solid at 298.15 K. [42FRI/WEI] Mixtures of metallic nickel and nickel oxide were treated with CO/CO2 gas mixtures at temperatures between 1044 and 1289 K. The composition of the gas phase after equilibration with the solid phases was determined by chemical analysis, yielding equilibrium constants for the reaction: Ni(cr) + CO2(g) ^ NiO(cr) + CO(g) as a function of temperature. [43NAS] Nasanen titrated an excess of potassium hydroxide drop-wise with NiCl2 solutions and determined the equivalence point potentiometrically with a hydrogen electrode. The solubility product was calculated from the minimum of the buffer capacity. A recalculation taking Nasanen's experimental conditions into account has led to logl0 K°o = - 15.09 and log10 K°o= 12.92, respectively. The latter value has been listed in Figure V-ll and TableV-6. It agrees reasonably well with log[0K°0= -15.21 (log10 K°o = 12.79) given in the original paper. These values are at best valid for amorphous Ni(OH)2 immediately after its precipitation, but in the course of titration some Ni(OH)2_vClv may have formed as well and thus influenced the result. [45KER] Two sulphides have been identified as lying close to the Ni and Co end members of the Co-, Ni-, Fe-disulphide system. In the light of this study the pyrite group forms an isostructural series with pyrite, FeS2, vaesite NiS2 and cattierite CoS2 as end members, while bravoite (Co, Ni, Fe)S2 is intermediate. [47DON] Amorphous NiS was prepared by reaction of nickel chloride solutions with Na2S and (NH4)2S. If the nickel sulphide is precipitated by bubbling H2S(g) through a 1 M NiCl2 solution containing sodium acetate, crystalline NiS is obtained which shows the hexagonal NiAs-type structure according to X-ray diffraction analysis. After stirring the solid phase in 1 M HC1 for 10 minutes the nickel content of the aqueous phase was analysed spectrophotometrically. The solubility of NiS was reported to be in the range from
A. Discussion of selected references
271
0.006 to 0.142 mol-drrf3. As the reaction time of 10 minutes is by far too short to attain equilibrium, no further evaluation of these data has been performed. [47PEA] Peacock investigated two specimens of heazlewoodite from Heazlewood, Tasmania, by chemical and X-ray diffraction analyses. The empirical formula of one of the heazlewoodite samples was close to Ni3.04Fe0.02S2- The X-ray diffraction patterns of the natural and the artificial NJ3S2 phases agreed very well with each other and also with those reported by Westgren [38WES]. The cell dimensions given by the latter author, however, were incompatible with the respective d spacings. Consequently, the cell dimensions given by Peacock and not those of Westgren were accepted for JCPDS-ICDD card No.8-126. [49GAY/GAR] Nickel hydroxide was prepared by adding an equal volume of 0.02 M NaOH to a hot solution of 0.0034 M Ni(NC>3)2.The precipitate was washed free of alkali with water. Two sets of samples were prepared for each concentration of HC1 (0.0025 - 0 . 1 m) or NaOH (0.0016 - 1 5 m). One was equilibrated for 5 to 7 days at 35°C and then for 7 days at 25°C, the other was equilibrated for 5 to 7 days at 25°C only. In this way the authors intended to approach equilibrium from supersaturation as well as undersaturation. Inspection of Figure V-ll shows that in the acid range an aqueous suspension of (3-Ni(OH)2 saturated at higher temperatures is undersaturated at lower temperatures. As a rule of thumb a temperature decrease of 10 K slows down chemical reactions, such as the dissolution reaction according to Equation (V.40), by at least a factor of two. Long ago Schindler pointed out that the method applied by Gayer and Garrett may not be effective with slowly dissolving metal hydroxides [63SCH]. Thus, it is not quite sure that in their work the investigated solutions were indeed saturated. When it is assumed that Gayer and Garrett's result is valid at 308.15 K it agrees very well with the experimental data of Gamsjager et al. [2002GAM/WAL]. Jena and Prasad [56JEN/PRA] as well as Mattigod et al. [97MAT/RAI] presume that there were systematic errors in pH measurements by Gayer and Garrett, but this cannot be proven just by shifting the reported pH values until their own results are confirmed. Moreover, Gayer and Garrett determined the concentration of nickel(II) employing the dimethylglyoxime method; in addition the data in alkaline solutions were checked by radioactive tracer experiments. They calibrated the glass electrode for pH measurements in a then conventional way. Thus, errors between 0.5 and 0.8 pH units as suggested by Mattigod et al. are improbable. The solubility of (3-Ni(OH)2 certainly depends on the particle size, but whereas it is quite certain that Mattigod et al. equilibrated their solutions with the finest portion of their samples, no such information is available of Jena and Prasad's or Gayer and Garrett's experiments.
272
A. Discussion of selected references
The re-evaluation of Gayer and Garrett's data resulted in log10 Ks 0 = 10.78 (as compared to 10.8 given) only when the second column of their Table II was considered to contain the molality, myr , and not the activity, aH+ , as stated in the original paper. The ionic strength was not controlled, instead activity coefficients of Harned and Owen were used [58HAR/OWE]. The nickel(II) concentrations, over the pH range studied, were sufficiently low to neglect the formation of Ni4(OH)*+ . The authors reand logl0 Ksli = - (4.2 + 0.6), for the reactions: ported the values, logl0 Ksl2~-7 Ni(OH)2 (cr) ^
Ni(OH)2 (aq)
Ni(OH)2(cr) + OH" ^
Ni(OH)j
(A.6) (A.7)
The authors mentioned, that the dissolution of nickel hydroxide in alkaline solutions up to 15 M NaOH was so slight that only estimates of its solubility could be made. When the reported experimental data were re-evaluated it became evident that, at best, only one constant, Ks u , can be determined. The solubility of Ni(OH)2(cr) increases with increasing NaOH molality, but the solubility curve flattens out at [OH~] > 1 m. This may be due to ion-ion interactions at the high ionic strengths, but there is no evidence whatever for formation of Ni(OH)^'. Equation (A.7) can be fit reasonably to the data at lower NaOH molalities. When the SIT model is applied: log10 [Ni(II)]l0, = log10 KsU - A£-[OH-]ini + logl0[OH-]ini The logarithmic solubility curve, as obtained by regression analysis, turns sharply downwards at high hydroxide molalities. This demonstrates that the SIT approach is no longer valid, because the mass law requires an increasing solubility with an increasing NaOH molality. The regression analysis results in l o g ] 0 / C u = -(4.40 + 0.12) and As = (0.66 + 0.17) kg-mol"'. Where As = e(Na+, Ni(OH) 3 ) - e(Na+,OH") and s(Na+,OH") = 0.04 kg-mor'. The value of e(Na+, Ni(OH)~) = (0.70 ±0.12) kg-mot' is an unusually high value that leads to the strange looking solubility curve of Figure V-7. This indicates that the present analysis of the sparse data is probably oversimplified, especially for high base concentrations. [50AKS/FIA] Aksel'rud and Fialkov precipitated Ni(OH)2, which was not characterised any further, and reacted it with nickel sulphate solutions at 18°C, determined the equilibrium pH with hydrogen, quinhydrone and antimony electrodes, plotted y = log!0[Ni2+] + 21og10£w + 2pH versus x = [Ni2+], and extrapolated graphically to zero concentration. The authors used l o g ^ w valid at 25°C instead of 18°C. In Figure V-l 1 and Table V-6 the similarly extrapolated but appropriately corrected value is given. It should be emphasised that by equilibrating Ni(OH)2 with NiSO4 solutions only minute amounts of the solid phase dissolve and thus the solubility relates to its finest probably amorphous fraction.
A. Discussion of selected references
273
[50GLE/EIN] Glemser and Einerhand investigated (3-NiOOH, y-NiOOH and other nickel oxide hydroxides of higher oxidation state by X-ray powder diffractometry and electron micrography. [50HUM/KOL] Polarographic studies on the reduction of the Ni(CN)^~ ion have been performed. The results confirmed the irreversibility of the reduction, and indicated a minimum formation constant of about 1024. Polarograms of saturated solutions of Ni(CN)2 showed waves due to the aquonickel(II) ion and the Ni(CN)^ complex, but no free cyanide ions could be detected. This indicates that the solid is actually Ni[Ni(CN)4] (see also [51LON]. The solubility product of Ni[Ni(CN)4] in 0.1 M NaCl solution was reported to be l o g 1 0 ^ = - 8 . 7 7 . [50VAS/GRA] Heat capacity measurements were reported for temperatures from 65 to 300 K for several hydrated transition-metal salts, including Ni(NO3)2-6H2O. Results were not listed for each measurement, although the specific heat values are indicated as points on a graph. "Smoothed" values are reported at 5 to 20 K intervals over the temperature range. A rather broad second-order transition for Ni(NO3)2-6H2O was found between 140 K and 170 K, with a peak at 149 K. The authors estimated the entropy change between 65 K and 273 K to be 87.1 caMC'-mol"1 (364.4 JKT'mor 1 ). No magnetic transition was found in the temperature range of the transition. The authors stated that the random error "should not exceed 0.001 cal-K~'-g~'" (1.22 .MC'-mol"1). From the set of smoothed specific heats, the value of the molar heat capacity of the solid at 298.15 K is calculated to be (463.0 + 2.0) JK~'mol ', where the uncertainty has been estimated in the present review. The temperature range of the measurements is not adequate for the calculation of a value for S° (Ni(NO3)2-6H2O, 298.15 K). [51COU] Coughlin measured the enthalpy of nickel chloride from 298.15 to 1336 K, see Figure A-2. The NiCl2 samples for this investigation were provided by Busey and Giauque [52BUS/GIA]. The adopted melting point was 7}us(NiCl2, cr) = 1303 K.
274
A. Discussion of selected references
Figure A-2: High temperature enthalpy of anhydrous NiCl2. Solid curve: Hm(T)-H°m(298.15) =a • (7-298.15) + b -{T2 -298.15 2 ) + c-(T'1 -298.15 ' ) , with a = 7.37949-10"2, b = 6.11344-KT6, c = 512.775. O: Experimental data from [51COU].
[51HOO/WOY] The equilibrium of the reaction: NiCl2(cr) + 2 HF(g) ^ NiF2(cr) + 2 HCl(g)
(A.8)
was studied from 477 K to 830 K. The concentrations of HCl(g) and HF(g) were measured in a gas stream that had been brought to equilibrium over the solids. At the temperatures used, the measured equilibrium constants were independent of the flow rates of the nitrogen carrier gas containing HCl(g) or HF(g). In the present review, values of Ar.//^((A.8), 298.15 K) are calculated using the reported equilibrium constants, selected values of S°m and C°pm(T) from Tables III-l and III-3 for NiF2(cr) and NiCl2(cr), and the tabulated values for HCl(g) and HF(g) in the CODATA compilation [89COX/WAG]. Then, using A f //° values from Tables III-l and IV-1 for NiCl2(cr), HCl(g), and HF(g), an average value, A f //°(NiF 2 , cr, 298.15 K) = - (652.76 ± 0.93) kJ-moP1, is calculated. Calculated values of A f //° (NiF2, cr, 298.15 K) appear to be systematically more negative (by 1-2 kJ-mol"1) when based on the higher temperature measurements.
A. Discussion of selected references
275
[51LAT] In this paper the original "Latimer" entropy contributions of metals and negative ions in solid compounds are listed. [51LON] The exchange of cyanide ion between the Ni(CN)J" ion and the bulk solution was complete within 30 seconds, which was too fast to be followed properly by the method used. A study of the Ni2+ - Ni(CN)^" exchange was complicated by formation of the precipitate (Ni(CN)2) in the mixture of nickel(II) and tetracyanonickelate ions. The results indicated two non-equivalent metal ions in the solid, which confirmed that its structure is indeed Ni[Ni(CN)4]. The nickel(II) exchange between Ni(CN)4~ and four nickel(II) complexes indicated a direct bimolecular interaction. [52BUS/GIA] Busey and Giauque measured the heat capacity of nickel from 15 to 300 K. The entropy, heat content and free energy functions have been calculated. The authors used 99.98% nickel, in contrast to the numerous low temperature heat capacity studies quoted in this paper where rather low purity Ni-metal was investigated. Calculations of thermodynamic properties of nickel were extended to 800 K on the basis of available data. However, the results of such extrapolation seem to be less reliable. The standard molar entropy of Ni determined by Busey and Giauque was equal to 29.86 IlC'-mol" 1 . [52BUS/GIA2] The heat capacity of anhydrous nickel chloride was studied between 15 to 300 K. The anomalous region associated with the transition from the antiferromagnetic to the paramagnetic state was investigated in detail. A maximum of C°pm(T) was found near 52 K. This temperature corresponds with the temperature of the magnetic susceptibility maximum within the accuracy of the magnetic measurements. [52CAR/BON] Carr and Bonilla reported an approximate value of the standard electrode potential of nickel from the following cell: Ni | NiSO41 KCl(sat) | KCl(sat), Hg2Cl21 Hg. Their data have been recalculated with the standard potential of the Hg2Cl21 Hg electrode ((0.2681 ± 0.0026) V, [89COX/WAG]) and the chloride activity of a saturated KC1 solution a(KCl) = 2.829 [89LOB]. The activity of Ni2+ in the solution was calculated using the SIT model supported by ChemSage. The results are listed in Table A-4.
276
A. Discussion of selected references
Table A-4: Standard electrode potential of Ni2+ | Ni at 25°C. NiSOVM
log,0([NiSO4]/m)
lo
gio« N i ! -
E (vs. std. CE)
3
E (vs. SHE)
E° (vs. SHE) (mV)
(mV)
(mV)
0.0001
-4.0014
-4.10627
-590.4
- 349.0
-227.54
0.0010
- 2.999
-3.15545
-571.3
-329.9
-236.56
0.0100
-1.9987
-2.41766
-548.6
-307.2
-235.69
0.1000
-0.9989
-1.81732
-527.7
-286.3
-232.54
1.0000
0.0017
-1.26378
-499.2
-257.8
- 220.42
(moldm )
The authors rejected the results in rows 1, 4, and 5 altogether. As Figure A-3 shows only the data in row 5 deviate considerably from the curve predicted with a value of £°(Ni2+ | Ni, 298.15 K) = -(0.2331 ± 0.0082) V. This value agrees within the error limits with the value in the NBS tables [82WAG/EVA] and the result of [29HAR/BOS] as well as the compilation of [89BRA]. The reversibility of the cell was tested thoroughly and thus the value of [52CAR/BON] confirms the result of [29HAR/BOS] by an independent method. When only the results in row 2 and 3 are accepted, as recommended by the authors, [29HAR/BOS] and [52CAR/BON] agree within 1 mV. The uncertainty from the use of an uncorrected non-thermodynamic diffusion potential remains, and as a consequence the result of [29HAR/BOS] has been preferred.
Figure A-3: Plot of £ (vs. SHE) versus log10([NiSO4]/m)
A. Discussion of selected references
277
[52CHA/COU] Approximate hydrolysis constants for a number of cations, including Ni2+, were determined by potentiometric titrations at 30°C and 0.1 M (K)C1 ionic strength. The formation of NiOH+ as the sole hydrolytic species was assumed. However, the total metal concentration used (about 0.01 M) is too high to neglect formation of the tetranuclear hydroxo complex. The experimental data are not reported. Therefore, the results in this paper have not been considered further in the present review. [52GAY/WOO] The pH values of solutions prepared from purified NiCl2 ([Ni2+], = 0.0156 to 0.5 M) were measured to obtain data on the magnitude of the hydrolysis reactions. Only the formation constant, J3n , ofNiOH+: Ni2+ + H2O ^ NiOH+ + H+, was taken into account to explain the variation of the measured pH values. This paper was a short communication, and thus, several experimental details (such as the method of calculation and the source of the activity coefficients) were not reported. Furthermore, the [Ni2+], used was far too high to neglect formation of polynuclear hydroxo complexes. In addition, the effect of formation of chloro complexes was not addressed. The experimental data are inadequate for proper re-evaluation; therefore the reported constant was not considered in this review. [52ROS] An aqueous solution of nickel nitrate saturated with carbon dioxide was treated with ammonium carbonate at 0°C. After one week at 5°C bluish green crystals were formed. The analysis of this precipitate agreed within the error limits with the formula NiCO3-6H2O. The X-ray powder diffraction pattern was taken and listed. [52SUD] The reduction equilibria of synthetic nickel sulphides with hydrogen gas were measured with the help of a flow method. Ni7S6 was prepared by precipitation (reaction of H2S with an ammonia containing nickel sulphate solution). The solid product was annealed at 661 K in an evacuated silica tube for one hour. The composition was shown to be close to Ni7S6 by gravimetric analysis; no X-ray diffraction analysis was carried out. As the reduction experiments were mainly performed at temperatures below 675 K, the gas phase, characterised by a certain ratio /?(H2S)//?(H2), is in equilibrium with Ni3S2 and Ni9S8 in accordance with [94STO/FJE]. Thus, the ratios /?(H2S)/p(H2) given as a function of temperature should correspond to the equilibrium: >/2Ni9S8 + H2(g) ^
XN»3S2 + H2S(g).
In addition, the author prepared Ni3S2 by reduction of Ni7S6 with H2. Again, the solid product was characterised by chemical analysis, but no X-ray diffraction
278
A. Discussion of selected references
analysis was performed. Ni3S2 was partly reduced with hydrogen gas at temperatures ranging from 855 to 1000 K and the ratio p(H2S)/p(H2) in the gas phase was measured. The results were erroneously referred to the equilibrium between heazlewoodite (lowtemperature form of Ni3S2) and metallic nickel. However, at temperatures above 834 K the high-temperature form (pVNi3S2) is stable [80SHA/CHA], [91STO/GRO]. Moreover, the temperature for the eutectic between Ni, pVNi3S2 and the liquid phase is 905 K [80SHA/CHA], so that above this temperature some of the material equilibrated with the gas phase was molten. For these reasons the data of the experiments on Ni3S2 are discarded. [53BUS/GIA] Busey and Giauque investigated the equilibrium reaction: NiCl2(cr) + H2(g) ^ Ni(cr) + 2 HCl(g)
(A.9)
in order to provide evidence from the application of the third law of thermodynamics that a ferromagnetic substance approaches zero entropy at 0 K. This plausible implication was experimentally confirmed for the first time by this study. A third law analysis of the reported equilibrium constants was carried out employing the Gibbs energy functions of Ni(cr) and NiCl2(cr) selected in this review, whereas the Gibbs energy functions of H2(g) and HCl(g) were taken from [89COX/WAG]. The experimental data depicted in Figure V-18 are listed in Table A-5. Table A-5: Equilibrium constants of Reaction (A.9) 77 K.
Kp 1 atm
Kp 1 bar
-Rrin^/barVkJ-mol
661.67 676.65 693.00 645.33 676.18 707.11 738.02 707.02 646.19 692.14 692.12 707.42 722.73 737.89 630.18
0.2503 0.3949 0.6365 0.1488 0.3913 0.9568 2.135 0.954 0.1514 0.6223 0.6238 0.9565 1.423 2.095 0.0872
0.2536 0.4001 0.6449 0.1508 0.3965 0.9695 2.1633 0.9666 0.1534 0.6305 0.6321 0.9692 1.4419 2.1228 0.08836
7.5476 5.1532 2.5272 10.1516 5.2011 0.1822 -4.7349 0.1994 10.0720 2.6539 2.6400 0.1842 -2.1989 -4.6180 12.7133
'
A. Discussion of selected references
279
[53FRO2] The formation constants of Ni(SCN)^rA (x = 1 - 3) complexes have been determined at 20°C, at pH = 3, in 1 M (Na)ClO4 solution, using Amberlite IR-105 cation exchanger. The thiocyanate concentration ranged from 0 to 0.5 M, [Ni2+]tol varied between 1 and 5 mM. The equilibrium nickel(II) concentration was determined spectrophotometrically using dithiooxalate, and that of the thiocyanate ion by a Volhard titration. The results obtained were qualitatively confirmed by spectrophotometric measurements of the aqueous nickel(Il) - thiocyanate system. Considering the accepted enthalpy values for the formation of Ni(SCN)^"A (x = 1 - 3) species, the reported formation constants were corrected to 298.15 K: log,,,/?, = (1.14 + 0.06), log,0/?2 = (1.58 + 0.06), log,0/?3 = (1.72 + 0.10). [53GRI/SLU] NisS4 was synthesised by precipitation of NiS from aqueous solution with H2S. The suspended NiS was oxidised with oxygen to NiS2 which eventually disproportionated into Ni3S4 and H2SO4. [53SAN] This paper seems to be a slightly updated English version of [37SAN2]. Sano studied the reduction equilibria of cobalt(II) and nickel(II) chloride by hydrogen. He used the method of observing the total pressure and the partial pressure of hydrogen by means of a palladium membrane. Busey and Giauque tried to account for the difference of their results and Sano's results by assuming that thermal equilibrium was not attained between Sano's thermocouple and the reaction vessel. If Sano's temperatures were lower by about 14 K, the equilibrium constants would agree quite well with those of [53BUS/GIA]. In Figure V-18 the original experimental data have been depicted. They are listed in Table A-6. Table A-6: Equilibrium constants of Reaction (A.9) [53SAN] 77K 792 760 738 709 661
log10(A,/atm)
\ogm(Kp 1 bar)
0.7345 0.4141 0.1985 -0.1783 -0.8016
0.7402 0.4198 0.2042 -0.1726 -0.7959
-Rrin(^/bar)/kJTnol"' -11.2236 -6.1083 -2.8853 2.3426 10.0716
[53SCH/POL] Schwab et al. determined the solubility product of freshly precipitated Ni(OH)2 by measuring the pH when one half of the Ni2+ originally present in NiCl2 solutions had been precipitated by adding NaOH. The solid phase was not investigated any further.
280
A. Discussion of selected references
All experiments were carried out at "room temperature". The authors reported iogwKs0 = - 15.5. It is difficult to recalculate this value from the experimental data given; in any case, two different values of the pH of half precipitation can be found in [53SCH/POL]. If the first (pH = 7.75) is used, this results in log10 Ks0 = - 15.5. Use of the second value (pH = 8.4) would lead to log l 0 K s 0 = - 14.2. Thus, it was concluded that to extrapolate either of these results to / = 0 is not justified, and consequently neither was included in the final SIT calculations. In addition the hydrolysis of nickel (II) was studied by potentiometric titrations of NiSO4, Ni(NO3)2 and NiCl2 solutions. The initial concentration of nickel(II) was 2.44 mM for all titrations. The ionic strength was not controlled. The experimental data were interpreted assuming that NiOH+ was the only hydrolysis product formed. For all three solutions, irrespective of the anion present, logl0 /?,, = - 9 was reported. This value was not considered further in this review. [53SWA/TAT] Swanson and Tatge recorded the pattern of a NiO sample obtained from Johnson, Matthey & Co., Ltd. The estimated purity of 99.99% was corroborated by spectroscopic analysis, which showed only faint traces of Mg, Si, and Ca. After comparison with eight other published patterns the cell dimensions given by [53 SWA/TAT] were accepted for JCPDS-ICDD card No.4-835. [54DOB] Nickel sulphate solutions were titrated with sodium hydroxide and vice versa at 75°C. Each titration was carried out within 2 to 3 hours which appears to be a rather short period for equilibration even at 75°C. Ni(OH)2 and NiSO4'3Ni(OH)2 were precipitated in dilute and concentrated solutions, respectively. Neither solid phase was investigated with respect to its structure, e.g., by X-ray analyses. It is difficult to recalculate the results properly, because the ionic strengths are not given in Dobrokhotov's Table 1. The value for Ni(0H) 2 listed by the author (log10 Ks0 = - 16.2, log10 X , o = 9 - 2 5 a t 7 5 ° c ) agrees surprisingly well with our recent results. The recalculation of Dobrokhotov's digitalised data from Figures 3 and 5.1 with the SIT model resulted in log10 Ks0 = 9.48 at 75°C, whereby complex formation between Ni2+ and SO;;" had been taken into account (see Figure V-l 1 and Table V-6). [54FEI/HAR] This is an abstract of a conference paper containing no experimental details at all. The authors maintain that the solubility products of Ni(OH)2, freshly precipitated and aged at 110°C, differ by 2.5 orders of magnitude. As mentioned in Section V.3.2.2.3, the solid phases were not characterised with respect to their X-ray patterns. [54ROS] This important contribution deals with the investigation of phase relations in the system Ni—S by measurement of the partial pressure of sulphur in equilibrium with various
A. Discussion of selected references
281
solid phases. The nickel sulphides were prepared by solid state reaction from the elements. The solid sulphides were equilibrated with hydrogen gas at temperatures between 673 and 1273 K. The partial pressure of sulphur is given by the ratio /?(H2)/p(H2S) which is recorded in situ by density measurements employing a buoyancy balance. The compositions of the solid phases were determined by chemical analysis before and after equilibration with hydrogen. From these equilibrium studies the stability ranges of the different sulphides as well as their Gibbs energies of formation were derived. Gibbs energy functions are given for the following reactions: )/Ni+H 2 S ^
!/2Ni3S2+H2
[ A r G m S = (- 75479.4 + 32.217 (77K)) kJ-moP1, = (- 12761.2 - 7.7404 (77K)) kJ-moF1,
2Ni3S2+H2S ^ N i 6 S 5 + H 2
[\GJ™£
Ni6S5+H2S ^
6Ni,^S+H2
[ArGm]f733* = (-21840.5 + 14.477 (77K)) kJ-mor1,
)/Ni+H 2 S ^
!/2Ni3+;tS2+H2
[ A r G n i ] ^ = (- 49998.8 - 13.514 (77K)) kJ-mol"1,
Ni3+JS2+H2S ^ 3Ni,^S+H2
l A G J ^ = (- 69454.4 + 70.291 (77K)) kJ-mor1,
Ni,_xS+H2S ^ NiS2+H2
l A G J ^ = (- 19664.8 + 55.815 (77K)) kJ-mol"1.
It is worth mentioning that according to Stelen et al. [94STO/FJE], the composition of the phase Ni6S5 is actually Ni7S6. [54SHC/TOL] Shchukarev et al. measured the equilibrium concentrations of HBr(g) and H2(g) at equilibrium over solid samples of nickel and nickel bromide for temperatures from 623 K to 928 K. The primary reaction was: NiBr2(cr) +H2(g) ^ Ni(cr) + 2HBr(g). The values of the logarithm (base 10) of the reported equilibrium constant values (converted to the 1 bar standard state) are listed in Table A-7, where: log.o Kp = log10
L - -
lAi 2 (g) / b a r J
.
The authors indicated that equilibrium may not have been established successfully at the lowest temperatures (below 700 K), and that almost complete conversion of the NiBr2 to Ni at 928 K made the value for the highest temperature somewhat suspect. The reported equilibrium constant values were used here with the selected entropy (Section V.4.1.4.1) and equations and values for C°m(NiBr2, cr, 7), Section V.4.1.4.2, C; m (Ni, cr, 7), Section V.I.2, and C ^ ( H 2 , g) and C;,m(HBr, g) [89COX/WAG], to calculate values for A f //° (NiBr2, cr). The authors suggested that there was evidence for a transition for NiBr2(cr) just below 800 K (and an enthalpy of transition of 24 kJmol 1 ). Although two crystalline modifications of NiBr2 are known [34KET], the evidence from these sparse measurements [54SHC/TOL] seems to be insufficient to conclude that there is a transition with such a large transition enthalpy. The authors ap-
282
A. Discussion of selected references
parently did not confirm the crystalline form of the nickel bromide after measurements at each temperature. The calculated values of A f //° for 298.15 K become progressively more negative with increasing temperatures of the experiments on which the A f //° values are based (and the results for nickel chloride, as obtained by the same authors, also drift slightly with temperature). The average value of Af H^ (NiBr2, cr) from the measurements done between 700 and 900 K is - (215.02 ± 0.91) kJ-moF1. Table A-7: Enthalpy of formation values calculated from equilibrium gas compositions. 77K
logio*. -2.915
623 683 723
-1.930 -1.392 -1.070
748 773
-0.777 -0.295
823 873 928
0.130 0.497
\fH°m (NiBr2, cr) -210.86 -212.25. -214.37 -214.36 -214.48 -215.48 -216.42 -218.24
Shchukarev et ah investigated the decomposition reaction: NiCl2 ^ Ni + Cl2 by studying the equilibrium: NiCl2(cr) + H2(g) ^ Ni(cr) + 2 HCl(g)
(A. 10)
and combining the latter with the well known formation equilibrium of hydrogen chloride: H2 + Cl2 ^ 2 HC1. A third law analysis of the reported equilibrium constants was carried out using the same auxiliary Gibbs energy functions described under [53BUS/GIA]. The experimental data are depicted in Figure V-18 and listed in Table A-8. Table A-8: Equilibrium constants of Reaction (A. 10) [54SHC/TOL] 77 K
Kp 1 atm
Kp 1 bar
573 623
0.013 0.078
0.01317
693
0.59 1.23
0.5978 1.2463 2.0265
723 743 773
2.00 4.00
823
9.8
0.07903
-Rrin(X,/bar)/kJ-mol-' 20.6272 13.1460 2.9643
4.0530
-1.3236 -4.3633 - 8.9944
9.9299
-15.7079
A. Discussion of selected references
283
[55BRO/PRU] This paper reports cryoscopic measurements for NiSO4 aqueous solutions (0.00 > / /°C > - 0.21), and demonstrates the dependence of the calculated value of K\ on the assumptions in the complexation model. Based on the arguments of Brown and Prue, it is clear that use of the SIT as is done in the TDB project imposes constraints on the data analysis, and that these constraints should lead to a specific value for K\. However, recalculations show that the association constant is also strongly dependent on the concentration at which the data set is truncated. For the purposes of the present review, recalculations were done using the SIT, but only results for molalities < 0.03 were used in the selection of the value of log10 Kt . When data for higher concentrations were included, the calculated value for the association constant increased, and log10 K^ = 300 was obtained if the entire data set was used. [55CAT/STO] This adiabatic calorimetry study (12 K to 300 K) appears to be the only lowtemperature heat capacity study for NiF2(cr). From their large series of results the authors calculated 5° (NiF2, cr, 298.15 K) = (73.60 ± 0.17) J-K^-mol"1. A X transition was found at (73.22 + 0.05) K. The experimental values seem to have a non-systematic uncertainty of less 0.3 J-KT'-mor1. In a later paper, the authors indicated that the heat capacity values above 270 K may have a slightly greater uncertainty than those for lower temperatures [55STO/CAT]. Slightly lower values between 270 K and 300 K mesh better with the available high-temperature drop calorimetry results [70BIN/HEB]. [55KRA/WAR] Krauss and Warncke used two different calorimetric methods to determine the specific heat of high-purity nickel. The measurements in the temperature range between 180 and 600°C were carried out using a continuous calorimetric technique and between 500 and 1160°C by means of "reverse calorimetry". The source of heat in the last method was a Pt-rod of known temperature and heat capacity which was placed into the calorimeter containing a nickel sample. The separation of the determined heat capacity into a latticevibration term, a magnetic term and a residual term was briefly discussed in this paper. The reported data were used by this review for fitting of the thermal heat capacity function for nickel crystal. [56AHR/ROS] The stabilities of the fluoride complexes of nickel(II), copper(II) and zinc(II) have been determined by a pH-metric method using a quinhydrone electrode at / = 1 M Na(ClO4, F) and at 293.15 K. The pK value of HF2~ was reported in an earlier paper (pK = 2.93, [56AHR/LAR]). Under the conditions used ([F] t < 400 mM, [Ni2+]t < 100 mM) only the monocomplexes (MF+) were found to exist in appreciable concentration.
284
A. Discussion of selected references
[56CUT/KSA] Potentiometric titrations of NiC104 solutions with NaOH were performed at (20 + 0.1 )°C. To determine the solubility constant, the pH of incipient precipitation was measured for nickel concentrations of 0.0041 — 0.034 M at constant ionic strength, / = 1.0 M NaClO4. The value for log10 K°o, obtained by extrapolating these experimental data to / = 0 with the SIT model is listed in Table V-6. The hydrolysis constant was determined at nickel concentrations ranging from 0.0025 to 0.184 M. The ionic strength was not controlled, and the activity coefficients were calculated using an extended Debye-Hilckel expression. Such a general expression is normally assumed to be valid at ionic strengths below 0.1 M. In the course of this study, the ionic strengths varied from 0.007 to 0.55 M, with most data pairs at / > 0.1 M. The formation of the single complex NiOH+ was assumed, and a value of logl0 /?,° = - 8.94 was obtained from graphical extrapolation to / = 0. The experimental data used in the calculations referred to very low degrees of hydrolysis, thus, large unavoidable errors render the result dubious. For this reason, the reported hydrolysis constant was not included in the detailed evaluation. [56FIA/SHE] This was a scoping study. Conductance measurements were carried out for dilute solutions of nickel sulphate. However, the data were incorrectly analysed in terms of a 1:2 nickel:sulphate complex, and the data are not available for re-analysis. [56JEN/PRA] Nickel hydroxide was prepared by adding an excess of sodium hydroxide solution to a solution of nickel chloride. The precipitate was freed from alkali by washing with water and was then dried in an air-oven at 98 - 100°C. This solid phase was equilibrated either with HC1 (0.0189 - 0.1892M) at 25°C or buffer solutions of (1:1) acetic acid (0.04 - 0.20 M) and sodium acetate at 29°C. The authors preferred the result of the buffer solutions, however a recalculation shows that the mean values of either series agree within an experimental error of ±0.1 logio^ units and thus both have been recorded in Figure V-11 and Table V-6. [56KEN] This paper reports cryoscopic measurements for NiSC>4 aqueous solutions in saturated potassium perchlorate solutions (eutectic freezing point 272.99 K, freezing point depressions, AT, are between 0.0673 K and 0.181 K). Thus, the measurements were done with less variation of ionic strength than for measurements done in water. The results of this study are assessed in our discussion of a later paper [58KEN] by the same author. [56LAN/MIE] Lange and Miederer determined the relative apparent molar enthalpy, *L, of NiSC>4 and demonstrated that:
A. Discussion of selected references
limf^l
285
=28.9kJ-kg°-5-morL5
leads to the slope, which is over two times higher than the value of AL predicted by the Debye-Huckel limiting law. Lange and Miederer's experimental data were re-evaluated using the chord method described in [58HAR/OWE]. A still higher limiting slope has been obtained:
limf ^-jL I = (37.5 ± 2.6) kJkg° 5 -mor' 5 . The *I function thus derived was employed to extrapolate the enthalpies of NiSO4-6H2O(cr) solution determined by Goldberg et al. to zero ionic strength [66GOL/RID]. [56SOK] The application of thermal, metallographic, X-ray, conductivity and specific gravity measurements revealed the phase relations in the binary system Ni - S in the region from 30.0 to 50.0 at% S. The phase diagram given by Elliott [65ELL] is based on results of this study as well as the work of Kullerud and Yund [62KUL/YUN]. [56YAT/VAS] The solubility product at 25.0°C for hydrated Ni2P2O7, and 1:1 and 1:2 complexation constants between Ni2+ and P2O,~ were determined from a small set of solubility measurements in aqueous solutions of sodium pyrophosphate ranging in concentration from 0.032 to 0.123 M. The experiments seem to have been done carefully. Hydrolysis of pyrophosphate was considered to be small, and was neglected in the data analysis. This was probably not a bad approximation, as it can be calculated that < 5% of the unassociated pyrophosphate in the final solutions was protonated. The species NiHP2O, [64HAM/MOR], [73PER/SEC] was even less important under the experimental conditions. Because of the presence of many highly charged species, the ionic strength of the experimental solutions was high (0.5 to 1.2 M), and varied from solution to solution. Estimation of interaction coefficients would be highly speculative for some of the species, and it does not seem useful to attempt to use the value in this paper for the first association constant to determine thermodynamic values f o r / = 0. The value of 5.82 for log|0 Kx , is consistent with later studies in constant ionic strength media [64HAM/MOR], [73PER/SEC]. The value determined for the cumulative second association constant (Reaction (A.I 1)) was /? 2 = 1.54 x 107 Ni2+ + 2 ?2O47~ ^
Ni(P2O7)62"
(A.11)
(actually the cumulative dissociation constant, (6.5 + 2.0) x 10~8) was reported). Reexamination of the experimental data suggests that the 1:2 complex does form, but evidence for its formation is only marginal except in those solutions with higher initial
286
A. Discussion of selected references
pyrophosphate concentrations (> 0.08 M), and log10 K2 is probably < 1. The neglect of pyrophosphate complexation with sodium ions (an inherent assumption required within the context of the SIT) [94STE/FOT], introduces a further uncertainty. Therefore, no value for K2 is selected in the present review. In the absence of added sodium pyrophosphate, the solubility of the nickel pyrophosphate in 0.5 M to 1.0 M KNO3(aq) was reported as 3.4 x 10"* M, which is not inconsistent with the solubility product proposed by the authors (for higher ionic strengths) if the lower hydroxide concentration, and consequent formation of HNiPjO,, is considered. [56YAT/VAS2] Measurements of the enthalpy of solution of aliquots of a 1.92 M solution of nickel nitrate into solutions of sodium pyrophosphate (0.025 M to 0.184 M) were used to determine the enthalpy of the complexation reactions of Ni2+ with P2O*~. The final nickel: pyrophosphate ratios were also varied at the four different pyrophosphate concentrations. The values \Ht = (17.61 ± 0.17) kJmol ' and AtH2 = - (9.25 ± 0.25) kJ-mol"1 at 25°C were reported. A value of K~{ = 2.5 x 10~2 derived from the calorimetric results is also reported. It is not clear that the enthalpies of the two complexation reactions are as easily separable as suggested by the authors, especially at the higher pyrophosphate concentrations. The complexation constants cannot be accurately calculated for these solutions (see the discussion above for [56YAT/VAS]), and therefore there are large uncertainties in the enthalpies of the complexation reactions. Recalculations suggest that the value for A r //, is more positive than the reported value, that A r // 2 is more negative than the reported value, and that in none of the solutions does a single complex account for > 90% of the nickel in solution. For lower pyrophosphate concentrations, the measured enthalpies also must include a contribution for deprotonation of small amounts (< 5%) of monohydrogenpyrophosphate. [57CHU] The nickel orthoarsenite, not identified except by the Ni:As ratio, was found to dissolve in dilute nitric acid solutions at 20°C over 12 hours to form solutions with aqueous nickel concentrations of 8.7 x 10~3 mol-dm"3 at a "pH" value of 6.75 and 3.1 x io~3 mol-dm~3 at a "pH" value of 7.10. The Ni:As ratio of the solid was found to be unchanged after the experiment. The author did not report a solubility product based on their measurements. If the dissolution of the solid is assumed to correspond to the reaction: Ni3(AsO3)2-xH2O + 6H+ ^
3Ni2+ + 2HAsO2(aq) + (2 + x)H2O
(A. 12)
and the concentrations are assumed to be sufficiently low that the AE terms in the activity coefficient equations are negligible, the average value of logl0 K (A. 12) is (28.69 ± 0.16). In the near-neutral solutions, ionisation of HAsO2 can be assumed small,
A. Discussion of selected references
287
as can the extent of any complex formation. If the initial "pH" values are compared to those calculated based on the final concentrations of Ni2+, there are discrepancies of 19% and 31%. This suggests that the measured "pH" values were not particularly accurate. In the present review, the average value for 20°C is accepted as the value for 25°C, but with an increased uncertainty, i.e., log10 K ((A. 12), 298.15 K) is (28.7 ± 0.7). [57KIU/WAG] This pioneering work demonstrates the usefulness of emf measurements on galvanic cells, involving solid electrolytes for the determination of the standard molar Gibbs energy of formation of oxides, sulphides, selenides, and tellurides at elevated temperatures. These measurements have been carried out by means of the "open cell stacked pellet" technique. The reference electrode was a metal - metal oxide system of known Gibbs energy of formation. The emf of the cell Fe, FexO (wilstite) | (0.85 ZrO2 + 0.15 CaO) | Ni, NiO and the standard Gibbs energy change of the reaction: Ni(s) + '/2O2(g) ^
NiO(s)
were determined in the temperature range from 1023.15 to 1413.15 K. The reported reproducibility of emf was ± 2 mV or better, corresponding to an uncertainty in AfGm(T) of ±0.418 kJ-mol"1. The standard molar Gibbs energy of formation of nickel oxide was the following function of temperature: AfGm= (-235.171 + 0.08567 (77K)) kJ-mol"1. [57KIV/LUO] The formation constants of nickel(II) and zinc(II) chloro complexes were determined by a polarographic method at 25°C using lead(II) as the indicator ion. The ionic strength was held at 2 M using sodium perchlorate. The concentration of chloride ion was varied between 0.4 — 1.4 M. The authors reported the formation of two nickel(II) chloro species: NiCl+ and NiCl2(aq) and their formation constants fi\ = (0.56 ± 0.20) and /?2 = (0.9 + 0.5). From these data, it follows that K2 (= fh!P\ - 1.6) is three times higher than K] (= fi\). This is very unlikely taking into account the low complex forming ability of chloride ion with nickel(II), and is probably the result of some unidentified experimental error. Consequently, these data could not be used to formulate conclusions concerning the nature and stability of chloro complexes. [57MOR/ZEL] This report describes a carrier-gas method for measuring vapour pressure of liquid metals. Vapour pressures of nickel were determined at temperatures 1813 to 1893 K. The results can be represented by the following equation: logioPmn, = - 2 1 0 3 0 / r + 9.689, where/?„„, is vapour pressure (in mm of Hg). The heat of vaporisation of nickel at 0 K was determined to be 427.898 kJ-mol"1.
288
A. Discussion of selected references
[58KEN] This paper reports cryoscopic measurements for NiSC>4 aqueous solutions in saturated potassium chlorate solution (freezing point 272.55 K) and saturated potassium nitrate solution (freezing point 270.32 K). The eutectic freezing point depressions, AT, are between 0.0296 K and 0.178 K. Thus, together with the measurements reported in [56KEN], the variation in the complexation constant, K{, with ionic strength was examined. The experimental association constant values, K\, were extrapolated to the saturation ionic strength of the supporting electrolyte at the melting point. Therefore, the reported values at different limiting ionic strengths are also values for slightly different temperatures. For the cryoscopic measurements using potassium chlorate, the ionic strength contribution of the supporting electrolyte is generally less than that from the nickel sulphate. Only in the potassium nitrate solutions is the contribution of the nickel sulphate to the total ionic strength reasonably minor. Therefore, the extrapolation method used by the author is probably appropriate only for the potassium nitrate and (perhaps) the potassium chlorate systems. The author's extrapolations to determine K\ in the saturated solution of supporting electrolytes are not strictly compatible with the TDB use of the SIT, but the reported K\ values are probably adequate if the uncertainties are increased. The values logioAT, = (1.27 ± 0.30) (-0.8°C, / = 0.26 mol-kg"1 KClO3(aq)) and l o g , ^ , = (0.69 ± 0.20 (- 2.8°C, / = 1.15 mol-kg"1 KNO3(aq)) are accepted for use in the present assessment. In this review, the SIT is used to determine the values at / = 0 (with the effective interaction coefficient for Ni2+ with NOj (0.18 ±0.01) kg-mol1, neglecting nitrate association (Section V.6.1.2.1), and the estimated interaction coefficient for Ni2+ with CIO3 (0.28 ±0.10) kg-moF1). The latter value was estimated as the average of the interaction coefficients of Ni2+ with perchlorate and nitrate, with an increased uncertainty. The results are log]0 Kx = (2.35 ± 0.30) (- 0.8°C) and log10 K} = (2.16 ± 0.20) (-2.8°C). The curve fits to Kenttamaa's data could be used to determine values of the second cumulative association constant, but these values of /% are strongly dependent on activity coefficient assumptions and on the values used for the heats of solution. Kenttamaa did not propose /% values (and considered that the value corresponding to /?2 from the fitting process also incorporated systematic errors from the experiment and data analysis). Rossotti and Rossotti [59ROS/ROS] did report /?> values based on the measurements done by Kenttamaa using potassium nitrate mixtures. We concur with the opinion of the original author, and do not accept these values of approximately log10 K2 = 1 at / = 0.3 m and 1.2 m, although they are roughly comparable to a value reported by Tanaka, Saito and Ogino [63TAN/SAI] (log 10 A: 2 = 1.4 in 1.2 M NaClO4(aq)CH3COONa mixtures at 25°C).
289
A. Discussion of selected references
[58MUL] Muldrow synthesised anhydrous NiCl2 and determined Asol//m of reaction: NiCl 2 (cr)^Ni 2 + + 2Cr
(A. 13)
at 298.15 K. The mean value of A sol //° (A.13) has been obtained from column 4 of Table A-9 by Equation (A. 14) Asol//° (A.13) = AsolHm (A.13) - *L,
(A.14)
see Section V.2.1.3.2 (NiCl2-6H2O(cr) cycle). For the calculation of *L the last term of Equation (IX.44) in [97GRE/PLY2] had to be neglected, because the temperature derivative of e(Ni2+, Cl~) is unknown. Systematic errors may have arisen from the fact that the NiCl2 samples were not completely dry (w(H2O) / w(NiCl2) = 0.04 - 0.08).
Table A-9: Enthalpy of solution of NiCl2(cr) in H2O(1) at 298.15 K [58MUL]. b (NiCl2) (molkg-') 0.00545 0.00971 0.00192 0.00200 0.00373 0.00555 0.00138 0.00279 0.00685 0.01036
*L
A5ol//m(A.13) (kJmor1)
(kJ-mor1)
-82.508 -83.596 -82.341 -82.718 - 82.760 -83.011 -81.881 - 82.508 - 83.303 -83.387
0.667 0.854 0.417 0.424 0.564 0.672 0.357 0.494 0.736 0.878
Asol//°(A.13) (kJ-mor1) -83.175 -84.451 -82.758 -83.142 - 83.323 -83.683 -.82.238 - 83.003 - 84.040(a) -84.265
(a) Carried out in dilute acid.
[58TRE] The method of frontal analysis on an ion-exchange column was used to study the formation of weak chloro complexes of nickel(II), cobalt(II) and copper(II). Although, the reported dissociation constant for the complex NiCl+ (Kdiss = (4.6 ± 0.1), which can be converted to log10 Kt - - (0.66 ± 0.01)) is close to other reported values, some experimental data, such as the temperature of the measurements, were not published. Therefore, the reported constant is rejected in this review. [58VAI/RAM] The only part of this work that was related quantitatively to pyrophosphate complexation of nickel was a sparse set of spectrophotometric measurements from which an "in-
290
A. Discussion of selected references
stability constant" of 2.4 x 1CT4 was derived. The pH of the solutions is not indicated, though apparently the sodium salt was used. The association quotient is smaller by several orders of magnitude than values from other studies, and there are insufficient experimental details to determine if the value might refer to a protonated complex. The reported value is not used in the present review. [58YAT/KOR] The formation of MSCN+ complexes (M = Mn(II), Fe(II), Co(II) and Ni(II)) has been studied by a spectrophotometric method, by detecting the decrease of the absorbance of Fe(SCN)2+ complex at 465 and 495 nm with increasing concentrations of M(II). The concentration of Fe(III) and SCN~ was 5 and 0.5 mM, respectively, while the concentration of Ni(II) ranged from 0.02 to 0.5 M. Consequently, the ionic strength varied between 0.13 and 1.56 M. The Davies equation was used to calculate the activity coefficients. For the purposes of the present review, the reported data were re-evaluated using the SIT (cf. Figure A-4). From the plot of this figure, log10 fi° = (1.69 + 0.15) and Ae = - (0.075 ± 0.030) kg-mol"1 can be derived.
Figure A-4: Extrapolation to /,„ = 0 (using the SIT) of the formation constants for the species NiSCN+ as reported in [58YAT/KOR].
[59ACH] The pH of slightly alkaline 0.01 M K2SO4 and KC1O4 solutions was measured before and after addition of a Ni(NC>3)2 solution under N2 atmosphere. The concentration of
A. Discussion of selected references
291
nickel(II) in the titration vessel varied between 0.003 and 0.024 M. The ionic strength was varied during the measurements, and the Guntelberg equation was used to calculate the activity coefficients. Since the inertness of sulphate media is highly questionable, only the data in perchlorate solution will be discussed further. The solutions were equilibrated for 30 minutes. The experimental data were interpreted only in terms of the formation of NiOH+. Therefore, we re-evaluated the reported experimental data assuming the presence of both NiOH+ and Ni 4 (OH)J + . The following values were obtained: log,0 7?u = - ( 1 0 - 8 5 ± °- 10 ) a n d lo8io *A,4 = " ( 26 - 8 ± 1 0 ) - T h i s calculation revealed, however, that the degree of hydrolysis is very low. The highest concentration of NiOH+ and Ni4(OH)4+ was 5.4 x 10"7 and 4.4 x 10"9 M, respectively. Therefore, above data were not considered in the present review. [59BLA/GOL] The spin-lattice relaxation times of water protons have been measured to study complex formation between nickel(II) and cyanide ion in concentrated aqueous ammonia solution (p = 0.88 g-cirf3). The relaxation time increased up to the ratio [CN~]/[Ni2+] = 4/1, where a sharp breakpoint was observed, according to the transformation of the paramagnetic Ni(NH3 )l+ to the diamagnetic Ni(CN)4~. At higher cyanide concentration the relaxation time decreased with increasing cyanide concentration, and after a second breakpoint at the ratio [CN~]/[Ni2+] = 6/1 the relaxation time was unaffected by further cyanide addition. From these observations the authors were led to the conclusion that the addition of cyanide ion to Ni(CN)4~ results in quantitative formation of Ni(CN)^~. This finding was disputed later by many authors [60MCC/JON], [62BEC/BJE], [63PEN/BAI], [64GEE/HUM], [65COL/PET], [71PIE/HUG], and now the nonexistence of Ni(CN)^ is well established. [59BLA/GOL2] In this work an effort has been made to determine the formation constants of the Ni(CN)^" species, using relaxation times derived from *H NMR measurements. As already mentioned in the discussion of [59BLA/GOL], the existence of this species can be ruled out, even in the case of high excess of cyanide ion. [59FRE/SCH] Complex formation in the nickel(II) - cyanide system has been investigated by spectrophotometry (267.5 nm) at 24.92°C and at several ionic strengths (/ = 0.0028 to 0.1 M) using potassium perchlorate as the ionic medium, in the pH range from 5.3 to 7.7. The nickel(II) concentration was ~ 4 x 10~5 M. The [Ni2+]/[CN~] ratio was varied from 0.05 to 0.8 to obtain data so that the Job's method could be applied, and otherwise kept at around XA. The time required for the equilibration ranged from a few days for solutions of higher pH to several weeks for those of lower pH. Acetate and phosphate buffers were used to maintain the pH, but correction was made only to account for the formation of the acetato complex. Only formation of the complex Ni(CN)4~ was detected,
292
A. Discussion of selected references
with a formation constant of logl0 J3° = (31.0 + 0.1). The authors used pA:iICN = 9.398, which is the average value of some early determinations. Since the above value is not correct, and log10 /?4 depends on pKllCN to the fourth power, Christensen et al. [63CHR/IZA] recalculated the thermodynamic formation constant reported by Freund and Schneider, using the more reliable pA^°CN = 9.216, which resulted in logl0/?4° = (30.3+0.1). The method of this recalculation is unknown; therefore, a re-evaluation was carried out in the present review using the reported experimental data. Since the concentration of the acetate and phosphate buffers were not reported, corrections taking into account the acetato- and phosphato- complexes of nickel(II) was not possible. The values of pA^,ICN at different ionic strengths were calculated using the data reported in [92BAN/BLI]. The re-evaluation indicated that only Ni(CN)f formed in the equilibrated solutions, with log,0 fi4 = (29.39 ± 0.14) (/ = 0.1 M), (29.46 + 0.13) (/ = 0.048 M), (29.79 + 0.10) (/ = 0.014 M), (30.12 ± 0.12) (/ = 0.0028 M). For the purposes of this review, slightly larger uncertainties, + 0.20, were estimated for each of these constants. The ionic strength range is too narrow to perform an SIT analysis, but these values for logl0 PI determined at low ionic strengths are nearly identical to the value of logl0/?° selected in the present review. [59KEN] This paper reports cryoscopic measurements of NiCl2 solutions in saturated potassium chlorate (0.258 m, f.p. = 272.35 K) and potassium perchlorate (0.0485 m, f.p. = 272.99 K) solutions. The author determined the freezing point depression at different NiCl2 concentrations (in KC1O3 cNiCU = 0.0193 - 0.121 m and AT= 0.10 - 0.6FC, in KC1O4 c Nia, = 0.0234 - 0.1202 m and A f = 0.12 - 0.59°C), then extrapolated to cNicl, = 0 at the saturating ionic strength of the supporting electrode. The ionic strength contribution of NiCl2 is dominant in KC1O4 solutions, therefore the extrapolation used by the author can be accepted, with an increased uncertainty, only for the data measured in KCIO3 solutions. The value of log,0 /?, = (0.23 ± 0.30) (-0.8°C, / = 0.258 m KC1O3) is accepted in this review. From this value, logl0 P° = (0.79 ± 0.40) (- 0.8°C) can be calculated using the SIT, assuming the interaction coefficients s(Ni2+, CIO3) to be (0.28 + 0.10) kg-mol"1 (see also [58KEN]) and e(NiCl+, CIO") = 0.5 x {s(Ni2+, C1O3) + E(K + , CP)} = (0.135 ± 0.100) kg-mor'. [59KIS/CUP] Spectrophotometric measurements have been performed in the binary nickel(II)-cyanide and copper(I)-cyanide, and ternary nickel(II)-copper(I)-cyanide systems. It was concluded that Ni(CN)4~ and Ni(CN)g~ form in the binary nickel system. The equilibrium constant for the reaction: Ni(CN)'" + 2 CN~ ^
Ni(CN)*"
was determined in 5 M potassium acetate (log ]0 K = (3.33 ± 0.15)) and 10 M potassium nitrite (log l0 K - (3.42 + 0.15)) solutions (the temperature of the measurements was not
A. Discussion of selected references
293
reported). Today, the non-existence of Ni(CN)£~ is well established [60MCC/JON], [62BEC/BJE], [63PEN/BAI], [64GEE/HUM], [65COL/PET], [71 PIE/HUG]. According to Beck and Bjerrum [62BEC/BJE], the results reported in [59KIS/CUP] do not support the formation of a stable hexacyano complex, but rather of an unstable pentacyano species. [59LAF] The phase relationships between nickel monosulphide and Ni3S2 or Ni7S6 were studied by equilibrating NiS with hydrogen gas at temperatures ranging from 677.15 to 1006.15 K. The high-temperature form of NiS was in equilibrium with Ni7S6 from 677 to 823 K and with Ni3S2 (high-temperature modification) between 823 and 1006 K, respectively. The H2S(g) content of the gas, formed due to reduction of the monosulphide with hydrogen, was measured iodometrically and the composition of the solid phases was determined by X-ray diffraction analysis, magnetic susceptibility measurements, and chemical analysis. From the ratio of the partial pressures between H2S(g) and H2(g) the partial pressure of S2 can be easily calculated, leading to a temperature dependence of p(S2) which is in close agreement with the data reported by Rosenqvist [54ROS]. [59NAI/NAN] The paper reports on a series of potentiometric measurements (0 to 45 °C). The measurements were carried out in acidic media (HCl(aq)), and a Davies-type equation was used to calculate the activity coefficients (/ < 0.05 m). The reported formation constant (A. 15) for NiSO4(aq) at 25°C is (211 + 24), Ni2++ SO^ ^ NiSO4(aq).
(A. 15)
Sufficient data are provided to permit recalculation using the SIT and the TDB selected values [92GRE/FUG] for the first protonation constant (and enthalpy of protonation) for the sulphate ion. As in the original analysis, the initial concentrations, measured potential of the cell: H2(g), Pt| HCl(aq, m,), MSO4(aq, m2)|AgCl(s)|Ag(s), the selected value for the equilibrium constant (/ = 0) for: H+ + SO4- ^
HSO; ,
(A. 16)
and the selected activity coefficient equations (in this case the SIT equations) were used. A model assuming formation of a single 1:1 nickel sulphate complex was used, and calculations were carried out so that the ionic strength values in the activity coefficient equations were identical to those calculated from the final species concentrations. The values logl0 AT0 ((A. 16), 298.15 K) = (1.9810.05) and A r //° ((A. 16), 298.15 K) = (22.4 ±1.1) kJ'mor 1 were those selected in a previous TDB review [92GRE/FUG]. From these, the association constants for Reaction (A. 16) were calculated assuming constant values for C° m over the temperature range 0 to 45°C. As dis-
294
A. Discussion of selected references
cussed below, this is probably not adequate. The values e(Ni2+, Cl") = (0.17 ± 0.02) and E(H + , Cl") = (0.12 + 0.01) were from the same source (see also Section V.4.2.4). The value e(H+, HSO^) = (0.01 ± 0.02) is the value used by Grenthe et al. [92GRE/FUG] in their Appendix A discussion of [76PAT/RAM]. No specific value was adopted for s(Ni2+, HSO~), as this interaction was assumed to be incorporated in the value of the formation constant for the NiSO^aq) complex. Preliminary calculations indicated that the reported data for experiment 8 at 298.15 K [59NAI/NAN] could not be used in recalculations. The concentrations reported in Table 1 of the paper are not compatible with the reported complexation constant value or the reported final ionic strength. The calculations showed that the values calculated for the formation constant of NiSO4(aq) are extremely sensitive to the activity coefficient model and to the values assumed for the first protonation constant for the sulphate ion. For example, for 298.15 K, the formation constant was recalculated to be (173 ± 64). However, if the value for logl0 A"°(A.16) was changed from 1.98 to 1.96, well within the assessed uncertainty bound, the value became (140 ± 72). The uncertainties here are merely the 2a uncertainties in the set of seven values recalculated from [59NAI/NAN], and do not include uncertainties in the auxiliary data. Even though the solute concentrations are low (/ < 0.05), and the overall changes to the activity coefficients from the SIT interaction terms are small, the calculated value of log10 AT°(A.15) increases from (173 ± 64) to (201 + 35) if all the e values are set to zero (but the maximum change in ya_ from the interaction terms is 0.004). In the present review, the value logl0 K°((AA5), 298.15 K) = 173 is accepted, and an uncertainty of ± 80 is estimated. It is not clear whether the values of K(AA6) at temperatures other than 298.15 K are sufficiently well defined to generate usable values of ^T(A.15). The calculated values of K(A. 15) at the six different temperatures can be used to calculate A r // m =-(13.8 ± 1.0) kJ-mol"1 if the enthalpy of reaction is assumed to be constant from 273 to 318 K. However, Figure V-3 8 suggests that the values of K(AA5) at the two lowest temperatures diverge from those obtained from conductance and cryoscopic studies. [59ROB/STO] The data m, y+ and of NiSC>4 required in the plots of Figure A-5 and Figure A-6 were taken from Table 16, Appendix 8.10 of this textbook. The activity coefficient ^+(sat) has been calculated by a third order polynomial of In y± vs. 4m , as depicted in Figure A-5. For the activity of water, a w (sat), <j> was extrapolated to the saturated solution molality with a fourth order polynomial in m, as depicted in Figure A-6.
A. Discussion of selected references
295
Figure A-5: Extrapolation of activity coefficients to saturated solution at 25°C. Solid curve: calculated with a third order polynomial in Vm : In y± = A + B\-m '5 + B^m + Bym15 with A = -(0.90807 ± 0.02571), B{ = - (3.60463 + 0.09626), B2 = (1.42317 ± 0.10959) and S 3 = - (0.07339 ± 0.3843), with/? 2 = 0.9998, and a = 0.00699. • data from [59ROB/STO], • extrapolated value.
Figure A-6: Extrapolation of osmotic coefficients to saturated solution at 25°C. Solid curve: calculated with a fourth order polynomial in m: <j> = A + B\-m + B?m2 + Bym3 + B4-m4; with A = 0.61907, 5, = - 0.48613, B2 = 0.46844, fl3 = - 0.16846, B4 = 0.02584, and R2 = 0.99831 and a = 0.00326. • data from [59ROB/STO], • extrapolated value, with w(NiSO4) = 0.7488 ;^ = 0.7488.
296
A. Discussion of selected references
[59WIL/THR] Hellyerite forms a colourful association with bright green, amorphous zaratite, Ni3CO3(OH)4-4H2O, along shear planes in serpentine containing heazlewoodite (Ni3S2) and other nickel sulphides. The crystals are of a cobalt or pale blue. The empirical formula was found to be Ni(CO3)! ,8-5.67H2O. The mineral was named after Henry Hellyer, first Surveyor General of the Van Dieman's Land Company, who was responsible for exploring and surveying much of North Western Tasmania during 1826 - 1830 (approved by IMA 1959). Its rarity is probably due to the fact that hellyerite is relatively unstable and, if not kept in a cool, air-tight environment (Heazlewood is typically rather cold and damp), it decomposes to powdery zaratite-like phases, plus otwayite, Ni2(CO3)(OH)2H2O, and other minerals. [60LIS/ROS] See comments under [66KEN/LIS]. [60MCC/JON] See the discussion for [65COL/PET]. [60WIL2] Williams discovered a new mineral at Heazlewood, Tasmania, and identified it as nickel hydroxide, Ni(OH)2, by its X-ray powder pattern. [61CHU/ALY2] A value of logl0 Ks 0 for hydrated Ni3(PO4)2 at 20°C is reported as - 30.3. The value is based on sparse measurements, and no account is taken of complex formation between Ni2+ and HPO^ or other phosphate species. The results from this study were not used further in the present review. [61DAV/SMI] Spectrophotometric measurements were performed at 242.5 - 252.5 nm to determine the association constant for the NiSCN+ species. The concentration of the metal ion was in five- to twenty-fold excess over that of the thiocyanate ion so that the kinetic data could be interpreted. The ionic strength was 0.5 M, the background electrolyte is not specified. The association constants were measured between 1.4 and 45 °C, but only the value for 1.4°C is reported in the paper (although two interpolated values for 5.2 and 9.3°C are also given). Due to the lack of experimental details, the reported constants were not considered further in this review. [61LEB/LEV] Lebedev and Levitskii investigated the equilibrium of the reaction: Ni 2 Si0 4 + 2CO ^
2Ni + SiO2 + 2CO2
A. Discussion of selected references
297
in the range 800 - 1100°C using a circulation technique with an automatic and a volumetric gas analyser. The equilibrium of this reaction was approached both by reduction of nickel orthosilicate and by oxidation of nickel. Combining the experimentally determined Gibbs energy change for the above reaction with the Gibbs energy for the reaction: 2CO(g) + O2(g) ^ 2 C O 2 ( g ) , the authors [61LEB/LEV] calculated the Gibbs energy for the reaction: 2Ni + SiO2 + O2 ^ Ni 2 Si0 4 in the range 800 - 1100°C. These ArG° values are significantly different from results obtained by another direct equilibration technique [68CAM/ROE] and the emf-studies shown in Figure V-54. The results of Lebedev and Levitskii were, therefore, disregarded by this review. A possible explanation of the observed discrepancy seems to be difficulty in attaining equilibrium for the reduction of nickel orthosilicate by carbon monoxide in the temperature range chosen by [61LEB/LEV]. The values of the standard enthalpy of formation A f //° (Ni 2 Si0 4 , olivine, 298.15 K) = - 1378.21 kJ-mol ' and the Gibbs energy of formation A f G°(Ni 2 Si0 4 , olivine, 298.15 K) = - 1265.24 kJmol ' proposed by Lebedev and Levitskii were not used by this review and are significantly different from the recommended values. [61LIS/WIL] The emf of the cell Ag,AgBr|NaBr+M(ClO4)2+NaClO4|NaBr+NaClO4|AgBr,Ag (where M = Ni(II) or Co(II)) has been measured at / = 2 M (Na, M)C1O4 and between 273 323 K in order to determine the composition and stability of the nickel(ll) and cobalt(ll) complexes formed with bromide ion. The concentration of nickel(II) ranged from 0.08 to 0.31 M, while the chloride concentrations were nearly constant ([Cl ] ~ 0.008 M). The complex formation induced 1 to 6.5 mV differences between the electrodes. The average value of log10 /?, at 298.15 K is - (0.12 + 0.02). This value has been accepted with an increased uncertainty (± 0.40). From the temperature dependence of logl0 /?, , ArHm appears to fall from about 13 to 2.7 kJmol ' over the range 0 to 50°C (see Figure A-7). The authors could not explain this behaviour, which is probably due to some unidentified experimental error. Assuming constant A r // m between T = 273 - 323 K, and that the medium effect is additive in the logarithmic scale, from these data A r //° for the reaction, Ni2+ + Br > ^ N i B r + can be obtained as a rough estimation, A r //° (A. 17) = (8 ± 5) kJmol '.
(A. 17)
A. Discussion of selected references
298
Figure A-7: Plot of log,0 /?, (A.17) reported in [61LIS/WIL] vs. \IT.
[61MAR/BYD] The kinetics of the reaction: Ni(edta)2 +4CN ^
Ni(CN)f + edta4
(A. 18)
have been studied by spectrophotometry (285 nm) at 25°C in / = 0.11 and 1.01 M NaCl solutions around pH 11. From the kinetic data, logl0 /?4 (A. 18) = 31.52 was derived at / = 1.01 M, using pK[1CN= 9.36. Although, the latter pK is rather far from the correct value [92BAN/BLI], considering the pH used, the error in pKHCN has little effect on the reported log10 J34 (A. 18) value. [61MOH/DAS] The thermodynamic stability constant of the NiSCN+ complex was determined by spectrophotometric (t = (25 ± 1)°C) and potentiometric (t = 35°C) methods. The Davies equation was used to calculate the activity coefficients, and this is not compatible with the SIT. Therefore, the potentiometric data were re-evaluated, using the SIT approach. The resulting logl0 /?, - Im plot is rather scattered and the ionic strength range was relatively narrow (see Figure A-8). Therefore the reported data were not considered further in this review.
A. Discussion of selected references
299
Figure A-8: logl0 /?, - Im plot obtained from the potentiometric data (t = 35°C) reported for the Ni2+ - SCN system in [61M0H/DAS].
[61SHC] Shchigol synthesised nickel orthoborate by reacting nickel hydroxide or carbonate with orthoboric acid and determined its stoichiometric composition analytically. No attempt was made to characterise this substance structurally. First, the solubility of nickel orthoborate, Ni(BO2)2-4H2O(s), was studied at 22°C by equilibrating the solid phase with HC1 solutions. Second, solubility measurements in KOH solutions were interpreted by assuming simultaneous equilibrium between Ni(BO2)2-4H2O(s), Ni(OH)2(s) and the aqueous medium. A third method consisted of precipitating nickel nitrate with borax solutions. It should be mentioned that neither the solid nickel borate nor the precipitated nickel hydroxide were structurally characterised. An average value of the solubility product of Ni(BO2)2-4H2O(s) Ks0= 2.16x10 9 mol 3 dm 9 was obtained by these three methods. A recalculation of the solubility product was based on the experimental data obtained from undersaturation (method 1) and oversaturation (method 3). The acid dissociation constant of boric acid was taken to be log, 0 ^°= - 9.623 at 22°C [58HAR/OWE] and the Davies equation [62DAV] was used for extrapolation to / = 0. A value of logl0 K°o = - (8.80 + 0.19) was found in reasonable agreement with Shchigol's result (log l0 ^T°o = - 8.666), see Figure A-9. Shchigol found that the solubility of Ni(BO2)2-4H2O in aqueous media increases with increasing concentration of orthoboric acid. In electrolysis experiments the
300
A. Discussion of selected references
solution of the anode compartment was enriched in nickel indicating its presence in the complex borate anions. The formation of the nickel and cobalt borate complexes in solution was believed to occur according to the reaction: M(BO2)2(s) + «H3BO3(aq) ^
HBM(BO2)2+B(aq) + «H2O(1)
(A. 19)
The value n calculated from the solubility measurements on Ni(BO2)2-4H2O(s) at different concentrations of H3BO3 was calculated to be 0.7. Thus the co-ordination number for Ni2+ relative to the metaborate ion was assumed to be 3 and the proposed complex species was Ni(BO 2 )j. While Figure A-10 shows that for nickel n = 1 and for cobalt n = 2, Shchigol's conclusion that the dissolved Ni2+ essentially consists of Ni(BO2 )~ must be refuted. Where do the balancing cations come from ? When it was attempted to explain the data by the formation of a neutral species like NiH(BO2)3, the solubility product did not remain constant. Additional experiments (careful pH measurements, reagents free of protolytic impurities) are necessary to establish the formula and stability of nickel borate complexes. After standing 5 to 6 days, fine crystals of nickel hexaborate NiB6O|0 were precipitated from the complex nickel borate solution containing an excess of orthoboric acid possibly according to the reaction: HNi(BO2)3(aq) + 3H3BO3(aq) ^
NiB6O,0(s) + 5H2O(1)
(A.20)
Figure A-9: Solubility product of Ni(BO2)2-4H2O(s) at 22°C from data reported in [61SHC]. A oversaturation, — exp. slope: — 1.39, • undersaturation, — exp. slope both series: 1.06 ; — theor. slope: - 2.00
A. Discussion of selected references
301
Figure A-10: Nickel and cobalt borate complexes from data in [61SHC]. A M2+ : Co2+, —• Co(BO2)2-2H2O(s) + 2 H3BO3(aq) ^ CoH2(BO2)4(aq) + 4 H2O(1), • M2+ : Ni2+, — Ni(BO2)2-4H2O(s) + H3BO3(aq) ^ NiH(BO2)3(aq) + 6 H2O(1), O ([Ni 2+ ] tol -[Ni 2+ ])=/([H 3 BO 3 ].
[61TAN2] The standard potentials of cobalt and nickel in methanol were determined using the following cell without liquid junction: Ni | NiCl2, solvent | Hg2Cl2 | Hg. Whereas for the cobalt cell measurements were carried out successfully in pure water, methanol was the solvent for the nickel cell. Apparently the author experienced the same difficulties with nickel in the aqueous medium which were already described by [29HAR/BOS]. Thus, the results of this work provide no relevant information on thermodynamic data of Ni2+ in aqueous solution. [61TAN/OGI] Polarographic measurements were carried out in an acetate buffer (the assumed "pH" was 3.66), and concentrations of sulphate and acetate were varied in the presence of 1.00 x 10 3 M nickel nitrate and 1.09 x 10~3 M nitrilotriacetate at a constant ionic strength of 0.2 M (maintained with potassium nitrate). A dropping mercury electrode was used, and polyoxyethylene lauryl ether (1-2 x 10~6M) was used as a maximum suppressor. Although results are reported for 15, 25 and 35°C, the assumed "pH" value for the buffer at 15 and 35°C is not stated. The acetate molarities were varied from 0.025 to 0.100 M at 0.03 M sulphate, and the potassium sulphate molarities were varied from 0.01 to 0.05 M at 0.05 M acetate. The results indicated that for the experimental
302
A. Discussion of selected references
conditions only a monosulphato complex was formed for Ni(II), with no evidence for higher complexes or a mixed sulphato-acetato complex. Values of (14.3 + 0.4), (11.6 ± 0.9) and (14.6 ± 0.7) dm^mol 1 were reported for K{ at 15, 25 and 35°C, respectively. The reported statistical uncertainties are undoubtedly underestimates of the true uncertainties. For the purposes of the present assessment, the value log,0 Kt = (1.06 + 0.20) at 25°C is used; the measurements at 15 and 35°C are rejected because of the lack of experimental details. At this low ionic strength, assuming that the 8 value for the Ni2+-acetate interaction is the same as the value for the Ni 2+ -NO 3 interaction has little effect on the value calculated for/= 0 using the SIT, and log10 K° = (2.13 + 0.20) is obtained. [62BEC/BJE] The effect of different salts on the visible spectra of tetracyanonickelate(II) has been investigated, in order to elucidate the formation of further complexes, and to solve the discrepancy between two earlier papers, [59BLA/GOL] and [60MCC/JON]. All salts (KCN, KSCN, KI, KBr, KCl, KF, KNO3) added to Ni(CN)^ caused an increase in absorbance around 375 - 500 nm, but the effects were rather different. The data qualitatively confirmed the formation of four very weak complexes: Ni(CN)^, Ni(CN)4SCN3 Ni(CN)4 3 , and to a much lesser extent Ni(CN)4Br3 . Numerical estimates of their stability were not provided. The above-mentioned salts and some other compounds (NH3, pyridine, ethylenediamine, butylamine, ethanol, methanol, dioxane, dimethylformamide) were also added to solutions containing 0.04 Ni(CN)J~ and 0.98 M KCN. Relatively small concentrations of all these compounds result in a decrease of absorbance between 375 500 nm. Since ammonia has nearly the same effect as fluoride ion or dioxane, the authors rejected the possibility of further complex formation (such as Ni(CN)sF4), and explained the observation as a strong decrease of activity coefficients corresponding to the mass action expression: -'Ni(CN)5 •'CN"
•'Ni(CN)3
'
[62CON/GIL] Conway and Gileadi carried out electrochemical kinetic studies at the nickel oxide (Ni n -Ni in ) electrode. A polarisation decay method was employed in which the reversible potential was approached both from the anodic and the cathodic directions. [62FAU/CRE] Dissociation of some unstable chloro complexes has been studied by cryoscopic titration in saturated KNO3 (/ = 1.38 m, f.p. = 270.34 K). Using [Ni(NO3)2] = 0.02 M and [KCl] = 0.05 - 0.25 M; the freezing-point depressions were found to be between 0.065 and 0.393 K. The recalculation of the reported data using non-linear regression resulted in a value (log10 /?, = (0.36 + 0.06)) almost identical to that reported by the authors
A. Discussion of selected references
303
(log l0 # = (0.38 + 0.08)). The value of log10 ft = (0.36 + 0.30) is accepted for use in the present assessment. Assuming e(NiCl+, NO3) equal to 0.5((e(Ni2+, NO3) + s(K+, Cl)) = 0.09 kg-moP1 the extrapolation using the SIT resulted log,0 0° = (1.17 ± 0.40) (-2.0°C). [62FRO/LAR] The frequency and intensity of the C-N vibration have been measured for the SCN ion (at 2066 cm 1 ) and some MSCN+ complexes of 3d metal ions. In the presence of metal ions, new peaks at ~ 2100 cm ' appeared in all systems, and these peaks correspond to formation of the metal complexes. The intensities of the free and bound SCN ion were used to determine stability constants of the monothiocyanato complexes which formed. The NaSCN concentration was 0.36 M, [Ni2+]tot varied from 0.44 to 1.32 M, and (2.64 2[Ni2+]lot) M NaClO4 was used as a background electrolyte. Therefore, during the measurements, the ionic strength ranged from 3.44 to 4.32 M. Moreover, the temperature of the measurements was not reported in the paper. Although, the reported stability constant for the NiSCN+ complex (log,,,/?, = 1.18) is close to the value in other papers, taking into account the uncertain experimental conditions, the above value was not considered further in this review. Based on the comparison of different evaluation methods of the experimental data, the authors suggested the formation of 80% inner-, and 20% outer-sphere complexes. [62KUL/YUN] A systematic study of the phase relations in the binary system Ni - S was performed in the temperature range between 473 and 1303 K. The phase equilibria were investigated by means of quenching experiments, differential thermal analysis (DTA) and, to a lesser extent, high temperature X-ray powder diffraction. The phases present in the quenched samples were detected by application of X-ray diffraction analysis and reflecting microscopy. In the latter case the samples were polished and etched with concentrated nitric acid. The following stoichiometric phases were found by the authors: Ni3S2 (heazlewoodite), Ni7S6 (low temperature modification), NiS (millerite, rhombohedral structure), Ni3S4 (polydymite), and NiS2 (vaesite). The high temperature modifications of Ni3S2, NiySe, and NiS (hexagonal NiAs-type structure) were reported to show a nonstoichiometric homogeneity range. Whereas the homogeneity range of Ni7S6 was fairly narrow, the high temperature phases Nif _XS and Ni3+xS2 showed large stability fields in the respective phase diagram. In the sulphur-rich portion of the system a two-liquid phase field over a wide range of composition was revealed. The authors proposed a monotectic reaction between NiS2 and the two immiscible liquids at 1264 K. [62LOP] In this work a similar set-up to the one described previously by [52CAR/BON] was used for the measurement of the Ni2+ | Ni standard electrode potential. To avoid the formation of a Ni(OH)2 surface layer on the Ni electrode the NiSO4 solution was acidi-
304
A. Discussion of selected references
fied with H2SO4. According to other authors the nickel electrode corrodes in acid solution [52CAR/BON]. As a consequence the result, £°(Ni2+1 Ni, 295.15 K) = - (0.248 ± 0.004) V, (recalculated with the CODATA value for the saturated calomel electrode and corrected for an obvious misprint in the error limits) has been rejected. [62RIN] A direct search was made for the spinel polymorph of Ni2Si04(cr) using high-pressure techniques. An olivine - spinel transition was discovered at 650°C and (18000 ± 5000) bars. This result is in excellent agreement with the theoretical prediction. This enables the free energy of transition and the density of the closer packed polymorph to be found. From these data the transition pressure can be calculated. [62TRI/CAL] The decrease of the kinetic wave heights of the Ti(IV)-SCN complex at a Hg dropping electrode, in the presence of a second cation, allowed the stability constants of Ni(II), Co(II), Cd(II) and Mn(II) with thiocyanate ions to be determined at different ionic strengths (/ = 0.7 - 5 M (Na,H)C104). Due to the weak complex formation, this method could not be applied to the detection of formation of the NiCl+ complex. Therefore, the authors also determined /?, (NiSCN+) in a solution of 1.5 M (Na,H)C104 by means of spectrophotometry, using the UV-band at 275 nm, attributed to the NiSCN+ species. Then the decrease of the absorbance at 275 nm with increasing chloride concentration allowed calculation of /?, (NiCl + ). The following values were reported for the NiSCN+ complex. Table A-10 : l°SioPi versus /,„. 0 .73 •°g|0 P\
1.05
1.62
2.21
2 .84
6.58
(1 .21 ± 0.02) (1 .17 + 0.02) (1. 12 ±0.02) (1.10 + 0.04) (1 .11 + 0.02) (1 .27 ± 0.05)
Several experimental details are not given in the publication: (i) the temperature of the measurement, (ii) the chloride concentration used to determine /?, (NiCl+) (Chemical Abstracts reported [Cl~] < 1.3 M). The composition of the background electrolyte ((Na, H)C1O4) is also dubious. Therefore, the reported data were not taken into account in this review. [62VAG/UVA] Vagramian and Uvarov carried out emf measurements in the range of 18 < t < 250°C with the cell: Ni I NiSO4(0.501 mol-kg ') | Hg2SO41 Hg.
305
A. Discussion of selected references
Between 200 and 250°C the potential became a linear function of the temperature and was extrapolated to 25°C. The resultant £°(298.15 K) = 0.952 V combined with the standard potential of the Hg2SO4 | Hg electrode ((0.6127 + 0.0030) V, NEATDB auxiliary data) would lead to £°(Ni 2+ | Ni, 298.15 K) = -0.339 V. The value originally given (— 0.270 V) is based on an obsolete standard potential value for the Hg2SO41 Hg electrode (0.682 V). Finally, this result has been rejected due to the questionable calculation of the NiSO4 activity at / > 200°C and the rather long extrapolation to 25°C, although the slope (dE°/dT) of the solid straight line in Figure 1 of [62VAG/UVA] can be predicted accurately from the quantities selected by NBS and CODATA [82WAG/EVA] or using NEA-TDB auxiliary data. [62WIL2] The dissociation constants of CoSCN+ and NiSCN+ have been determined from UV spectrophotometric measurements at 25°C. The first set of data was measured at constant ionic strength (/ = 0.417 M (Na)C104) and was used to calculate the molar absorption coefficients of the MSCN+ complexes. A second set of data, measured at varying ionic strength, yielded the thermodynamic dissociation constants of thiocyanate complexes. The necessary activity coefficients were calculated from the Davies equation. The reported A"°ss for the NiSCN+ complex is (0.0173 ± 0.0002), which is equivalent to logl0 J3° = (1.762 ± 0.005). Since the activity model used by the author is not equivalent to the SIT treatment, the above value was not selected. Instead, the reported experimental data were re-evaluated for the purposes of this review. The recalculation of the first data set (measured at / = 0.417 M (Na)C104) resulted in the value of log10 J3, = (1.21 ± 0.08), and e(NiSCN+, 273 nm) = (149 ± 5). Using latter value for ^(NiSCN+, 273 nm), log10 fl° = (1.77 + 0.10) can be calculated from SIT analysis of the second data set. [63BOL/JAU] The hydrolysis of nickel(II) perchlorate ([Ni2+]t = 2 - 2 0 mM) in 0.25 M and 1.0 M NaC1O4 media has been investigated by potentiometric titration between - logio[H+] = 7 and 7.7 and at temperatures ranging from 25 - 50°C ( ± 0.05°C). Only the formation of NiOH+, Ni2+ + H2O ^ NiOH+ + H+
*/3i,,
was taken into account to explain the experimental data. The values of logl0 /?,, = -9.76, ArGm(298.15 K) = (13.25 + 0.10) kcal-mol"1 and A r // m (298.15 K) = (7.76 + 1.00) kcalmol ' are reported, probably for both ionic strengths used, since no "sensible" dependence of logl0 /?,, with ionic strength is reported. Under the conditions used, a substantial amount of Ni4(OH)4+ would be expected to form; therefore, the reported experimental data have been re-evaluated for the purpose of this review, taking into account the formation of the mononuclear and
306
A. Discussion of selected references
tetranuclear hydroxo species. Considering the formation constants reported by Baes and Mesmer [76BAE/MES], the presence of the dinuclear Ni2OH3+ species can be neglected under the conditions used. The recalculation resulted in the following constants (2a uncertainties).
Table A-l 1: Recalculated constants. /(°C)//(M) l°gio*A.i log,0X4
25/0.25
30/1.0
40/1.0
50/1.0
-(9.84 ±0.02)
-(9.74 ±0.02)
-(9.64 ±0.06)
-(9.68 + 0.20)
-(27.34 + 0.08) -(27.12 ±0.10) -(26.41 ±0.20) -(25.26 ±0.20)
Taking into account the limited experimental data available, the uncertainty + 0.30 is accepted for all the above values. From the temperature dependence of the formation constants (see Figure A-l 1 and Figure A-12) logl0 J3n = — (9.73 ± 0.30) and log,0 *#,,4 = - (27.68 ± 0.40) can be extrapolated for t = 25°C and / = 1.0 M. The temperature dependence of logl0 J3t, is too small, compared to the experimental error, to derive a value for the enthalpy of reaction, but A r // m (298.15 K, 1.0 M) = (174+ 15) kJ-mol ' can be obtained for the formation of Ni4(OH)*+.
Figure A-ll: Temperature [63BOL/JAU].
dependence
of
logl0
fiu
values
derived
from
A. Discussion of selected references
307
Figure A-12: Temperature dependence of log10 /? 44 values derived from [63BOL/JAU]
[63BUR/ABB] This study is similar to [61LEB/LEV]. The equilibrium of the reaction: Ni2SiO4 + 2CO ^ 2Ni + SiO2 + 2CO2 was investigated in the temperature range 700 - 1200°C. Results obtained by Burdese et al. do not coincide with results of other studies discussed in Section V.7.2.1.1.5 and shown in Figure V-54. Thus, they were not used for the evaluation of thermodynamic quantities selected by this review. [63CHR/IZA] The thermodynamic formation constant and heat of formation of Ni(CN)^ in aqueous solution have been determined at 25°C potentiometncally and calorimetrically, respectively. No background electrolyte was used, and the Debye-Huckel equation was applied to extrapolate the experimental thermodynamic data to zero ionic strength. Although, the activity model used is not compatible with the SIT, most of the experiments referred to / < 0.007 M; therefore, the reported value (A r //° = - 180.7 kJ-mol 1 at 25°C) is accepted as valid at / — 0, but the uncertainty is estimated in this review as ±4.0kJmol '.
308
A. Discussion of selected references
[63GEO/FER] The authors measured the heat capacity from 1120 to 1919°C and the heat of nickel fusion by a calorimetric method. The reported heat of fusion is A ^ / / " = (17.47 + 0.23) kJmor'. [63ISA] The X-ray diffraction patterns of hydrothermally synthesised anhydrous nickel carbonate were taken, indexed and the constants of the unit cell ascertained. The T, p phase diagram of NiCO3(cr) was determined semi-quantitatively. Solid solution formation between NiCO3 and MgCO3 was established. In addition hellyerite, NiCO3-5.5H2O, and zaratite were investigated. The latter turned out not to be a single mineral, but a composite of amorphous and fibrous components. [63KO/HEP] Ko and Hepler claimed that S. Friedberg (Department of Physics, Carnegie Inst. Technol., 1961) gave them S°m (NiCl2-6H2O, cr) = 344.3 J-K '•mol ' "from unpublished heat capacity data". No subsequent article containing these heat capacity data was found. Hence, no documentation of this value can be provided for. [63LIN/LAF] The partial pressure ratio /?(H2S)/p(H2) of the gas phase in equilibrium with the hightemperature modification of Ni3S2 was determined as a function of composition at 853.15 and 933.15 K, respectively. The same experimental procedure was performed as described by Laffitte [59LAF]. The ratios /?(H2S)//?(H2) for the coexistence of Ni3+xS2 with metallic nickel at 853.15 and 933.15 K are listed in Table A-12. The experimental values coincide well with those predicted by the present compilation of thermodynamic properties of nickel sulphides. Table A-12: Comparison between experimental data [63LIN/LAF] and calculated values of logio[p(H2S)/p(H2)] for the coexistence of Ni3+xS2 with metallic nickel. 77K
lo gl0 b(H 2 S)/p(H 2 )], exp.
log,0[p(H2S)/p(H2)], calc.
853.15
-3.04
-3.03
933.15
-2.90
-2.79
[63NET/DRO] The stability constants of NiCl+ and NiBr+ complexes have been determined by a spectrophotometric method at an ionic strength of 5.7 M (H(X, C1O4), where X = Cl or Br~). The concentration of the complex forming anion was varied from 0 to 5.44 M. The calculation of the formation constants was based on the slight changes of extinction
A. Discussion of selected references
309
coefficients at 395 and 400 nm. The values of the formation constants accepted in this review at / = 5.7 M for the NiCl+ and NiBr+ complexes are logl0 /?, = - (0.5 ± 0.4) and - (0.3 ± 0.5), respectively. [63PEN/BAI] See discussion for [65COL/PET]. [63PHI/HUT] Phillips et al. investigated the phase equilibria in the system NiO - A12O3 - SiO2 by quenching and direct observational techniques. The crystalline phases were identified by petrographic microscopy and X-ray diffraction methods. The equilibrium diagram of the system NiO - SiO2 drawn from the experimental data shows that the only intermediate compound is Ni 2 Si0 4 , which has the olivine structure. The authors found, that unlike other olivines which melt congruently, nickel olivine has an upper temperature of stability 1545°C and at temperatures 1545 and 1650°C, NiO and SiO2(cristobalite) coexist in equilibrium. [63SHA/DES] Solutions of unspecified nickel(II) salts ([Ni]tot = 0.029 to 0.0855 M) were titrated with NaOH solution at (28.0 ± 0.5)°C at an ionic strength of 1.0 M using NaClO4 as the ionic medium. The pH values of the solutions were determined by unspecified "external electrodes", which were calibrated using a single buffer solution (pH = 7). The added volume of NaOH solution was measured in "drops", assuming 20.6 drops per cm3. The uniform drop-size over the range of 0 - 25 cm3 is questionable. The results were analysed assuming the formation of NiOH+ as the sole hydrolytic species (log10 /?,, = - (10.03 ± 0.08)), although under the conditions used, a considerable amount of Ni4(OH)4+ would be expected. Therefore we re-evaluated the reported data, assuming (i) only the presence of Ni4(OH)4+ (log l0 *fl44 = - (27.85 ± 0.3)) and (ii) the formation of both NiOH+ and Ni4(OH)*+ (log10 *^,, = -(10.03 ± 0.04) and l o g l 0 X 4 = - (30.2 ± 0.7)). Both calculations revealed substantial errors in the measured pH, probably due to the relatively low precision of the experimental work. Therefore, these constants were not considered further in this review. The correction applied by Plyasunova et al. in [98PLY/ZHA] for the data in [63SHA/DES] resulted in log]0 */?,,= - (10.39 ± 0.17), logl0 */3l2=~ (10.4 ± 0.7) and log10 *foA=- (28.2 ± 1.0). Even these constants do not provide a better description of the experimental data. [63TAN/SAI] This work is a continuation of an earlier study [61TAN/OGI]. Polarographic measurements were carried out at 25°C in an acetate buffer (the assumed "pH" was 3.72), and concentrations of sulphate and acetate were varied. The sulphate complexation constant was then obtained by accounting for acetate complex formation. At this higher ionic strength, the reported nickel sulphate association constant was K\ = 3.7 dm3-mor'. This
310
A. Discussion of selected references
was determined using an extrapolation procedure that should have minimised any effect of acetate. The SIT is used with e(Ni2+, CIO;) equal to (0.37 + 0.03) kg-mol ' and log10 K° = 1.95 is calculated, with an estimated uncertainty of + 0.20. At the higher acetate concentrations (to 0.5 M) and sulphate concentrations (to 0.20 M) there was evidence for formation of Ni(SO4)2~ (A> 26 dm6-moi2), Ni(OAc)(SO4)~ (K= 1 dm6-mol2) and Ni(OAc)(SO4)32" (K= 3 dm'-mol'3). There is insufficient information to re-analyse the experimental data, and the values of the higher complexation constants are not accepted in the present review. [63THR] In the course of structure refinement Threadgold points out that the water content of natural hellyerite corresponds to the composition NiCO3(H2O)4-1.5H2O, whereas the nickel content indicates NiCO3(H2O)4-2H2O [59WIL/THR]. From the packing model of the structure it was concluded that it is not possible for hellyerite to contain two water molecules per formula unit in the interlayer water sheet. Consequently, the agreement of the diffraction pattern with Rossetti-Francois' NiCO3(H2O)4-2H2O [52ROS] was considered a problem. It should be emphasised, however, that a new synthesis also resulted in crystals with the empirical formula NiCO3(H2O)41.5H2O [2001GAM/PRE], [2002WAL/PRE]. [64ALC/BEL] This work deals with the experimental determination of the oxygen activity in lead melts by performing emf measurements over the temperature range from 820 to 1120 K. Galvanic cells of the type Pb(l) | YSZ | NiO, Ni were applied, where the Ni / NiO oxygen buffer served as the reference electrode. The Gibbs energy function for the reference electrode was obtained from measurements of the solid-gas equilibrium: NiO(cr) + CO(g) ^ Ni(cr) + CO2(g) provided by Tomlinson and Young (quoted as a private communication). Tomlinson and Young determined the CO/CO2 ratios in equilibrium with Ni/NiO by chemical analysis as a function of temperature. They obtained the following Gibbs energy function for the reaction mentioned above: \Gm (7) = (- (47.15 ± 0.71) + (0.293 ± 0.711) x 10 3 (77K)) kJ-mol '. Based on this equation, Alcock and Belford derived for the Gibbs energy function of formation for NiO: AfGm (7) = (- 235.27 + 0.08682 (77K)) kJ-mol ', 820 < 77K < 1120, with a standard deviation of ± 1.1 kJ-mol"1. [64GEE/HUM] The authors repeated (and extended) the measurements reported by Blackie and Gold in [59BLA/GOL]. It was found that the solutions, for which the ratio of cyanide to nickel(II) was between 4 and 8, have the same relaxation time as the solvent (both in 17 M ammonia solution and in water), in sharp contrast with the findings of Blackie and
A. Discussion of selected references
311
Gold [59BLA/GOL]. At a much higher excess of cyanide, a marked change in the UVVIS spectra, between 300 - 520 nm, was observed. This spectral change was used to determine the stability constant for the reaction: Ni(CN)2" + CN~" ^
Ni(CN)j"
at 23°C and / = 2.5 M Na(C104). The concentration of the tetracyanonickelate(II) ion ranged between 0.4 and 70 mM, while the concentration of cyanide ion varied from 0.05 to 2.5 M. The reported stability constant (log l0 Ks = - (0.69 ± 0.02)) was not corrected to 25°C, as the difference is assumed to be negligible. [64GRI/LIB] Equilibria established in a solution containing a cation (Cu2+, Co2+, Ni2+ or Cd2+) and chloride have been studied by means of chromatography on ion-exchange papers. Complex formation constants were determined for Ni2+ at 25°C and / = 3 M H(C1O4, Cl). The concentration of nickel(II) was 0.4 M, while the chloride concentration varied between 0 - 3.0 M. The reported constants are jix = (0.23 + 0.03) M '; /% = (0.06 + 0.03) M 2; /?3 = (0.001 ± 0.006) M 3 . The authors stated that the results are less accurate then ones obtained by means other standard methods. Due to the considerable change in the ionic medium during the measurements, only the formation of the first complex is considered in this review. After recalculation of the experimental data using non-linear regression, a value of logl0 /?, = - (0.45 ± 0.50) is accepted. [64HAM/MOR] Complexation concentration quotients for 25CC and a medium of 0.1 M (CH3)4NCl(aq) were reported for the formation of NiP 2 O^ (log10 K (A.21) = 6.98) and NiHP2O; ( l o g l 0 ^ (A.22) = 3.83). Ni 2+ + P2O* ^ Ni2+ + HP2O, ^
NiP 2 O^ NiHP2O;
(A.21) (A.22)
The ligand solutions were prepared by titration of the free acid with tetramethylammonium hydroxide (to a pH value of approximately 10). Aliquots containing the ligand, metal (as Ni(NO3)2), and tetramethylammonium chloride were titrated with 0.0685 M HCl(aq), with the primary measurements being carried out at pH values between 4.5 and 6.5. The electrodes were probably calibrated against pH buffers rather than concentration standards, and the hydrogen ion activity coefficient, y , was assigned the value of 0.796; the SIT procedure described in Appendix B would have led to essentially the same value, yw = 0.80. For Reaction (A.21) in 0.1 M (CH3)4NC1, log10 K (A.21) = 6.98 and logl0 K° (A.21) = (8.73 + 0.1 As). For Reaction (A.22) in 0.1 M (CH3)4NC1, logl0 K (A.22) = 3.83 and logl0 K° (A.22) = (5.14 + 0.1 As).
312
A. Discussion of selected references
As the ionic strength is low, the effect of the Ae term on the final calculated values of log10 K° is limited. Rather than use estimated 8 values for interactions involving the pyrophosphate moieties (some of which are highly charged), it is simply assumed that Ae is zero. As a consequence, the uncertainty in the log10 K° values is increased to + 0.25 for the values from these measurements in 0.1 M Me4NCl solutions. [64KOH/ZAS] Thermal decomposition of hydrated nickel sulphates was carried out as a function of temperature at 0.026 bar ("20 torr"). Samples were equilibrated at constant temperature (± 0.1 K) and pressure for at least 24 hours. Equations were reported for water vapour pressures as a function of temperature for mixtures of hydrates (NiSO4-7H2O/NiSO4-6H2O; NiSO4-6H2O/NiSO44H2O; NiSO4'4H2O/NiSO4-2H2O; NiSO4-2H2O/NiSO4-lH2O), and for a saturated solution (18 to 59°C). Rehydration of the monohydrate at 60°C with water vapour pressure of 0.078 bar was reported to generate a mixture of the mono-, di- and tetrahydrates. Decomposition of the monohydrate led to an amorphous solid, and vapour pressure measurements over mixtures of the anhydrous solid and the monohydrate were not reproducible. The authors indicated that they did not feel it was useful to use their empirical equations to calculate dissociation enthalpies and entropies, and values calculated from the equations are not used in the present review. [64KOS/KAL] The temperature functions of the heat capacities of hydrothermally synthesised calcite type carbonates MnCO3, FeCO3, CoCO3, and NiCO3 were investigated between 70 and 300 K. The respective standard entropies were calculated by the integration of Cp m IT versus T curves. [64LAR] The author used infrared spectroscopic results to conclude that only about 10% of the nickel sulphate ion pairs are inner-sphere complexes in 0.3 M and 1 M NiSO4. The derived association constants are useful for comparisons for the different di- and trivalent systems studied, but are not suitable for calculating a selected value in the present review. [64LEE/ROS] The sulphur pressures (sulphur activities) for the equilibrium between NiS2 and Ni l x S were studied at temperatures ranging from 673 to 873 K. Pure NiS2 was prepared by solid state reaction between sulphur and NiS. The solid product was checked by X-ray diffraction analysis before and after the decomposition experiments. The equilibrium sulphur pressure for the decomposition of NiS2 into Ni ! x S and S2(g) was measured by means of a Rodebush manometer [30ROD/HEN].
A. Discussion of selected references
313
[64PER] The first hydrolysis constant (AT, = /?,,) of nickel(II) was determined at 20°C by potentiometric titrations using various ionic strength and nickel(II) nitrate concentrations: 0.0016 M < / < 1.503 M, 0.00050 M < [Ni(II)]tot < 0.01000 M. In another set of titrations, the temperature was varied from 15 to 42°C, applying three different ionic strengths: 0.0016 ([Ni(II)]tot = 0.0005 M), 0.0130 and 0.0430 M ([Ni(H)]tot = 0.001 M). In all titrations, KNO3 was used to adjust the ionic strength. The author assumed that the only important hydrolysed species was NiOH+. For the experiments made at [Ni(II)]tol < 0.001 M, the formation of Ni4(OH)4+ (and Ni2(OH)34) can, indeed, be ignored, in the pH range of the experiments. The formation of NiNOj was not considered by the author, although at 20°C, where a wide range of ionic strength (/ = 0.0016 - 1.5) was used, its concentration can not be neglected. Therefore, for the purposes of this review, the reported constants were corrected for the formation of NiNOj, and the SIT approach was used to derive log10 /?°, (see Figure A-ll, page 306). From the plot in Figure A-13, for the reaction: Ni2+ + H2O ^ NiOH+ + H+
(A.23)
log10 *fi°, = -(10.07 ± 0.07) and Ae = -(0.38 + 0.10) kgmol ' are obtained. From the latter value and £(Ni2+,NO") = (0.182 ± 0.010) kgmol ', £(NiOH+,NC>3) = - (0.27 ±0.12) kg-mol ' can be derived. Figure A-13: Extrapolation to / = 0 of the equilibrium constants log10 ySj, (A.23) reported in [64PER] at 20°C, using the SIT.
314
A. Discussion of selected references
At the other temperatures, only a few percent of NiNOj can be expected (/max = 0.046 M), thus no correction was applied, but the SIT was used to derive log10 J3°t. The reported log10 /?,", values and those recalculated using the SIT are collected in Table A-13.
Table A-13: Reported and recalculated log1( ?(°C) log,//?" (reported)
15
20
25
30
36
42
-10.22
-10.05
-9.86
-9.75
-9.58
-9.43
l°g.o *A°, (recalculated) -10.22
-10.07
-9.80
-9.74
-9.59
-9.38
From the plot of In *#° vs. 1/7 (see Figure A-14) A r //° (A.23) = (51.7 ± 3.4) kJ-mor1 can be derived.
Figure A-14: Plot of In *#°, (A.23) recalculated from [64PER] vs. \IT.
A. Discussion of selected references
315
[64TAY/SCH] This is one of the pioneering studies of thermodynamic data by emf-measurements with a solid electrolyte galvanic cell. Zirconia, doped with CaO and MgO, was used as an electrolyte to determine the free energy of the reaction: Ni + 0.5 SiO2 + 0.5O2 ^
0.5Ni2SiO4)
and the free energy of formation of numerous silicates and titanates at temperatures from 750 to 1200°C. The data of Taylor and Schmalzried coincide well with the newest emf-study of the above reaction and were used by this study to determine the Gibbs energy of formation of Ni 2 Si0 4 . [64TRI/CAL] The formation constants of thiocyanate complexes of Ni(II), Cd(II) and Ni(II) were determined by an extraction method at 20°C. The aqueous media were 1.5 and 3.0 M (Na)ClO4 solutions. The authors used rather acidic conditions (pH = 0.5 to 1.0), as only the protonated form of the ligand (HSCN) can be extracted into the organic phase (Ni(SCN)2(aq) is reported to be weakly extractable). The method used to determine the concentration of the extracted HSCN was not reported. The concentration of the thiocyanate ion ranged from 0.01 to 0.55 M, while [Ni2+]tol was - 0 . 1 M. The experimental data were described by the formation of four Ni(SCN)^~A (x = 1 — 4) complexes. For both ionic strengths used, K3 < K4 is reported. Considering that no geometrical change of nickel(II) occurs during the stepwise complex formation [90BJE2], the reported K4 values are probably overestimated. Using the accepted enthalpy values for the formation of Ni(SCN)^ A (x = 1 - 3) species, the reported formation constants were corrected to 298.15 K and the uncertainties were increased: log10 y3, = (1.10 ±0.10), log,0/?2 = (1.69 ±0.10), log, 0 /? 3 = (1.62 ±0.20) (/ = 1.5 M), and log lo j0,= (1.15 + 0.10), logl0 p2 = (1.61 ± 0.10), log]0 fi3 = (1.21 ± 0.30) (/= 3.0 M). [64WEL/KEL] Low-temperature heat capacities were determined for NiS (millerite) and Ni3S2 (heazlewoodite) between 50 and 298.15 K. The solid compounds were obtained by reaction of certain amounts of nickel oxide with sulphur. All products were characterised by X-ray diffraction analysis. Standard entropy values at 298.15 K were derived by integration of the respective heat capacity functions. In the case of millerite, the standard entropy is equal to 5° (NiS, P, 298.15 K) = (52.97 ± 0.33) J-KT'mor1 and the heat capacity at 298.15 K is C;_m(NiS, p\ 298.15 K) = 47.11 JKT'-mor1, which agrees well with the result published by Grenvold and Stolen [95GRO/STO], C° m (NiS, (3, 298.15 K) = 47.08 J-K'-mol '. The standard entropy of Ni3S2 was found to be 5° (Ni3S2, cr, 298.15 K) = (133.89 ± 0.84) J-K ' mol ' which is in close agreement with the value given by St0len et al. [91STO/GRO] (5^(Ni 3 S 2 , cr, 298.15 K) = 133.2 J-K-'-mol"1). Weller and Kelley [64WEL/KEL] obtained C° m (Ni 3 S 2 , cr, 298.15 K) =
316
A. Discussion of selected references
117.65 JK~'mol ', which is in excellent agreement with the values recommended by Ferrante and Gokcen [82FER/GOK], C; m (Ni 3 S 2 , cr, 298.15 K) = 117.7 J-K '-mol 1 , and Stelen etal. [91STO/GRO], C°p m (Ni3S2, cr, 298.15 K) =118.2 J-K'-mor 1 . [65AD A/KIN] Adami and King reported heats of solution in 4.360 m HCl(aq) at 303.15 K. The thermodynamic cycle used heats for the following reactions: H2SO4-7.068H2O(l) ^ 2H+(sln) + SO^ (sin) + 7.068H2O(sln) Ni(cr) + 2H+(sln, H2(sat)) ^ Ni2+(sln, H2(sat)) + H2(g) 7.068H2O(l) ^
7.068H2O(sln) 2+
NiSO4(cr) ^ Ni (sln) + SO'~ (sin) to obtain a value of (3.724 + 0.460) kJ-mol ' for the heat of the reaction: Ni(cr) + H2SO4-7.068H2O(l) ^ NiSO4(cr) + 7.068H2O(l) +H2(g)
(A.24)
at 303.15 K. The authors corrected the value for the heat of Reaction (A.24) to (3.598 ± 0.460) kJ-mol ', at 298.15 K. The reported experimental uncertainties are ICT uncertainties. Also, the authors estimated the uncertainly in the reaction involving dissolution of nickel with evolution of hydrogen as + 0.3 kJ-mol ', after applying a correction to account for heat absorbed in vaporising the water and hydrogen chloride that accompanied the evolution of hydrogen. Solution calorimetry involving reactions resulting in gas evolution is notoriously difficult, and in the present assessment the 1 a uncertainty has been doubled. Based on this, a value of (3.724 + 1.400) kJ-mol ' is calculated for the heat of the Reaction (A.24) at 303.15 K (2a uncertainty). The difference in the values of ArC°m for 298.15 K and of ArC°m for the average reaction temperature, 300.65 K, are small enough that they can be ignored relative to the uncertainties in the values of C° m . Using the values at 298.15 K, C°m (NiSO4, cr, 298.15 K) = (97.63+0.10) J-K^-mol ' (see SectionV.5.1.2.2.1.3, and [78STU/FER]), C; m (Ni, cr, 298.15 K) = (26.07 ±0.10) J-K '-mol ' (see Section V. 1.2), C°pm(U2, g, 298.15 K) = (28.836 + 0.002) J-K~'-mol ' (cf. Table IV-1), C; m (H 2 O, 1, 298.15 K) = (75.351 ± 0.080) J-K ' m o l ' (cf. Table IV-1) and C; m (H 2 SO 4 -7.068H 2 0,1, 298.15 K) = (614+10) J-K '-mor 1 [56HOR/BRA], [82WAG/EVA], we obtain ArC°m = (20 ± 10) J-K'mor 1 , A(A r //°) = (0.10±0.10) kJ-mol 1 , and A r //°= (3.59±1.40) kJ-mor1. Then, using A f //° (H 2 0,1, 298.15 K) = - 285.83 kJ-mol ' and Af//n° (H2SO4-7.068H2O, 1, 298.15 K) = - 2897.05 kJ-moP1 [82WAG/EVA], A f //° (NiSO4, cr, 298.15 K) can be calculated. The value for A f //° (H2SO4-7.068H2O, 1, 298.15 K) is adjusted to - (2897.12 + 0.64) kJ-moP1 to allow for the differences in the values for Af H°m ( S O ^ , 298.15 K) from Table IV-1 and from the US NBS table [82WAG/EVA]; the uncertainty
A. Discussion of selected references
317
is estimated to incorporate the uncertainty for A f H° for the sulphate ion and the "hydration" to H2SO4-7.068H2O. From these values, A f //°(NiSO 4 , cr, 298.15 K) is calculated to be — (873.28 ± 1.57) kJ-mol 1 . The difference between this number and the value originally reported by Adami and King is related primarily to the different values used for A(H°m (H2SO4, 1, 298.15 K). Adami and King used values from the 1952 US NBS Circular 500 [52ROS/WAG]. [65AKI/FUJ] In this paper the olivine - spinel transition in Fe 2 Si0 4 and Ni 2 Si0 4 have been investigated over the temperature range 700 to 1500°C in the pressure range 20 to 40 kilobars for Ni2SiO4 using a tetrahedral-anvil type of high-pressure apparatus. For both silicates the transition has proved to be reversible. The equation for the transition curve for Ni2SiO4 was determined to be/? = 19 + 0.012 t, wherep is in kilobars and / is in degrees centigrade. On the basis of this equation, the standard Gibbs energy of formation of spinel from the olivine was calculated to be (1200 + 0.98 7) calmol ' for Ni 2 Si0 4 , where T is absolute temperature. [65BUR/LIL] The hydrolysis of nickel(II) was studied at (25.0 + 0.1 )°C in 3 M (Na)ClO4 medium by potentiometric titrations using glass electrodes. The concentration of nickel(II) ranged from 0.1 to 0.8 M. Since [C1O4] was kept at 3 M during the measurements, the ionic strength ranged from 3.1 to 3.8 M. This is a common problem of investigations using rather concentrated nickel(II) solutions. Nevertheless, this is one of the highest quality papers concerning the hydrolysis of nickel(II). Both graphical and computer evaluation of the experimental data indicated that the major species is Ni 4 (OH) 4 + , under the experimental conditions used. The remaining small deviations between the observed and calculated Z values (number of OH bound per nickel(II)) were explained by small analytical errors and the presence of Ni2OH3+ as a minor complex. The optimum fit resulted log l0 */? 44 = -(27.37 ±0.02 (3a)) and log10 */? u = -(10.0 + 0.5). The authors listed some additional minor species, remarking that the experimental data give no support for their existence. Due to the relatively important change of ionic strength (and thus activity coefficients) between the measurements at different nickel(II) concentrations, we reevaluated the reported experimental data referring to both [Ni2+]tot = 0.1 to 0.4 M and [Ni2+]tot = 0.1 to 0.8 M. Using the whole dataset, and the model proposed by the authors, logl0 7?4,4 = " ( 2 7 - 3 7 ± °- 04 ) ( 3CT ) a n d log10 X 2 = - (9.42 ±0.10) were obtained. The agreement between the recalculated and original values is excellent for the main hydrolytic species, and reasonable for the minor complex. The difference is very likely due to the more sophisticated data analysis software now available (for all calculations in this chapter the PSEQUAD program was used [91ZEK/NAG]). Introducing the species
318
A. Discussion of selected references
NiOH+ into the equilibrium model, a substantial (40%) decrease of the fitting parameter was obtained. From the data referring to [Ni2+]lot = 0.1 - 0.4 M, logl0 /? 44 = - (27.43 ± 0.04) (3cr), log10 *#,, = - (9.86 ±0.12) and logl0 */3l2 = - (9.87 ± 0.50), using the whole dataset logl0 */?4,4 = - (27.42 ± 0.03), log10 */?,,= -(9.89 ±0.06) and log10 /?, 2 = — (9.78 ±0.16) were calculated. It is important to note that the authors also tried to include NiOH+ in the equilibrium model, but their calculations gave no real support for its existence. Taking into account the above values, log10 J3A 4 = - (27.43 ± 0.05) (3a), logl0 *J3U = - (9.86 ± 0.12) and logl0 */? u = " ( 9 - 8 ± °-5) a r e accepted for use in this review. [65BUR/LIL2] The hydrolysis of nickel(II) was studied at (25.0 ± 0.1)°C in 3 M (Na)Cl medium by a method identical to the one described in [65BUR/LIL]. The total nickel(II) concentration was varied between 0.2 and 1.4 M, which resulted in a change of ionic strength from 3.2 to 4.4 M. The analysis of experimental data indicated the presence of Ni4(OH)4+ as the dominant species (log l0 */?4,4= - (28.42 ± 0.05)), while the deviations at lower pH values suggested the presence of Ni2OH3+ (log ]0 /?, 2 = - (9.3 ± 0.2)). For log10 /?!, a value of- 10.0 was obtained from the calculations, but was rejected as too uncertain. During these measurements, a large part of the Na+ ion concentration was replaced by Ni2+, consequently the constancy of activity coefficients at different nickel(II) concentrations is questionable. Furthermore, the formation of chloro complexes (both binary and ternary) was not considered. For the purpose of this review, an attempt has been made to minimise the influence of the medium effect (using only the data reported for [Ni2+]tol = 0.2 and 0.4 M) and to take into account the presence of NiCl+ complex. For the reaction: Ni2+ + Cl ^ N i C l +
(A.25)
log,0 /?, = —0.59 can be calculated at /,„ = 3.2 m in NaCl medium, using the recommended value for log10 # ° (A.25) and e(NiCl+, Cl) = '/2{e(Ni2+, Cl) + e(Na+, CO} = (0.1 ± 0.03) kg-moF1. Considering the whole data set and the same model as proposed, logl0 */?44 = - (28.32 ± 0.09) (3a) and log10 *0l2= -(9.41 ±0.13) can be calculated, which are in reasonable agreement with the published values (see also the discussion of [65BUR/LIL]). Using only the data sets reported for [Ni2+]tot = 0.2 and 0.4 M, and taking into account the formation of NiCl+, log10 /? 44 = -(27.52 ±0.10 (3a)) and log10 /?, 2 = - (8.55 ± 0.06) can be derived. Introducing the species NiOH+ into the equilibrium model, a 50% decrease in the fitting parameter was found, and log10 /?4 4 = - (27.55 ± 0.06) (3a), logl0 *j3t, = - (9.56 ± 0.10) and log10 *# 2 = - (8.95 ± 0.20) were obtained. These values are accepted for use in this review, with increased uncertainties: l o g I 0 X 4 = -(27.55 ±0.20), l o g l 0 X , = -(9.56 ±0.20) and log10 % = - (8.95 ± 0.40).
A. Discussion of selected references
319
[65COL/PET] Penneman and his co-workers published three consecutive papers concerning formation of the Ni(CN)^ ion [60MCC/JON], [63PEN/BAI], [65COL/PET]. In the first paper [60MCC/JON], the complexes formed in aqueous solutions of Ni(CN)^ and cyanide ions were studied by IR and spectrophotometric methods. The results indicated the formation of only one new species (Ni(CN)j~), even at a very high excess of cyanide concentrations. The magnetic measurements indicated diamagnetic behaviour of the pentacyano species, which suggests square pyramid geometry around the metal ion. From the spectrophotometric data, measured at 430, 450 and 470 nm, logl0 K5 = - (0.72 ± 0.02) was determined at 25.2°C and / = 1.34 M (NaClO4). From the IR measurements, log]0 Ks = - (0.52 ± 0.20) was obtained. Identical spectrophotometric studies at 15.4 and 33.6°C have been performed. From the temperature dependence of logl0 K5 (see Figure A-15) A r // m = -(10.4±0.2) kJ-mol ' can be obtained. The uncertainty in this value does not include systematic uncertainties, nor the uncertainty in determining an enthalpy of reaction from equilibrium data over a small temperature range. In the present review, we assign an uncertainty of + 4.0 kJ-mol ' to this value. While [60MCC/JON] was in press, two papers by Blackie and Gold [59BLA/GOL], [59BLA/GOL2] appeared, with substantially different conclusions, therefore Penneman et al. reinvestigated the Ni(CN)^- CN system [63PEN/BAI], using IR spectroscopy at 25°C and / = 4 M NaClO4. The authors suggested the formation of two new species Ni(CN)^ and Ni(CN)£~. The latter complex was taken into account to explain some deviations at very high cyanide concentrations ([CN ] > 2.5 M). The values of log10 K5 = - (0.55 ± 0.02) and log10 K6 = - (1.0 + 1.0) have been reported. Due to the substantial medium effect, only formation of the pentacyano complex is considered in this review, and its formation constant is accepted with an increased uncertainty (± 0.30). In a further experiment, the authors found that the addition of NaCl to the solution of 2 M NaCN and 0.075 M Ni(CN)^ (the main species is Ni(CN)f) resulted in a significant decrease in the peak height attributed to Ni(CN)^. This observation was explained as formation of Ni(CN)5Cl4~ (log l0 K = - 0.66). Two years later, Coleman et al. in [65COL/PET] tried to solve the discrepancy between their two earlier papers [60MCC/JON] and [63PEN/BAI]. In this third paper the authors again used IR and spectrophotometric methods to study the Ni(CN)4~ - CN system in a 4 m KF - KCN mixture at 25°C (30°C for IR). They showed that, under the conditions used, the formation of a single species, Ni(CN)^, is sufficient to account for the experimental data. Based on the spectrophotometric data, a value of logl0 K5 = (0.03 + 0.01) (in molal units) has been reported. From the IR measurements at 30°C, log,0 K5 = - (0.08 ± 0.06) was derived. In light of these findings, the authors explained their earlier results reported in [63PEN/BAI] (the formation of Ni(CN)4^ and Ni(CN) 5 Cl 4 ) by the neglected medium effect caused by the substantial change of NaCN/NaClO4 (NaCl/NaClO4) ratio during the measurements.
320
A. Discussion of selected references
Figure A-15: Temperature dependence of the association constant for the reaction Ni(CN)^ + CW ^± Ni(CN)f reported in [60MCC/JON].
[65EGO/SMI] This is a very careful study of the specific heat capacity of Ni 2 Si0 4 and Zn 2 Si0 4 , the only one for the nickel orthosilicate for temperatures as high as 1570 K. Egorov and Smirnova used a copper block calorimeter and 99 NiCr / Constantan thermocouples for the measurement of the temperature change of calorimeter. The average, least square, precision of measurements performed in the range between 555 - 293 K and 1570 — 293 K. was ± 0.4. This high temperature data were used in this review to fit the heat capacity equation up to 1570 K. [65LIN/SEI] According to Linke and SeidelPs critical evaluation of the nickel sulphate solubility in water the stable phase from the eutectic temperature to 31°C is NiSO4-7H2O. The reported solubilities differ by 1 - 3%. [65MOR/REE] The formation of nickel(II) chloro complexes in aqueous solution has been studied using a cation-exchange method at 20°C, in a solution of 0.691 M H(C1O4,C1). The distri-
A. Discussion of selected references
321
bution of nickel at equilibrium between the resin (Amberlite IR-120) and the solution has been determined spectrophotometrically. The reported formation constants are J3\ = (1.7 ± 0.1) M '; fc = (0.92 + 0.10) MT2. Due to the almost complete change of the ionic medium during the measurements ([Cl ] = 0 to 0.659 M) the real formation of the biscomplex is doubtful. As the presented data do not allow an independent recalculation, they were not considered in this review. [65PAO] The heat of solution of NiBr2 in water was reported as - (82.09 + 0.17) kJmol 1 based on limited measurements and analysis. The solid was obtained by dehydration over sulphuric acid in a vacuum desiccator, and the Ni and Br analyses were within 0.2% of theoretical. The number of samples measured was not reported, but some values in the same paper were based on only two measurements. Neither the dilution of the final solution nor the sample size was given. The enthalpy value probably was not corrected for dilution, as there is no description of a correction procedure. The uncertainty reported was probably la. The resulting value is approximately 4 kJ-mol' less exothermic than that of Efimov and Furkalyuk [90EFI/FUR] (although the final solutions probably were somewhat different). [65PAO/SAB] A value of - (70.46 + 0.33) kJ-mol"1 was determined for the enthalpy of solution of NiI2(cr) in water containing («-Pr)4NI (0.0005 m) and a "small quantity" of HI. The final solution concentrations were approximately 0.00025 m in nickel. The solid was obtained by dehydration over sulphuric acid in a vacuum desiccator, and the Ni and I analyses were within 0.2% of theoretical. The reported value was an average of "at least two determinations". The authors concluded from similar measurements on the dissolution of ZnI2(cr) that the effects of association were small; yet, their own measurements suggest that the uncertainties from this source could be greater than 1 kJ-mol '. The resulting value is less exothermic than that of Efimov and Furkalyuk [90EFI/FUR], though the final solutions also are not identical. [65PAW/STA] The specific heat capacity of high-purity nickel and several nickel-rich alloys from room temperature to 600°C is reported. The data, obtained with an adiabatic calorimeter, were reproducible to within + 0.25 per cent and have an estimated absolute error of less than + 0.5 per cent. This high quality data were credited by this review and used for the determination of the heat capacity function. [65POT] This reports a study of the complex dielectric constants of 1 M NiSC>4 in aqueous solution as a function of temperature.
322
A. Discussion of selected references
[65WEL] Heat capacity measurements, 52.0 to 296.3 K, were reported for anhydrous NiSO4(cr). A maximum entropy for 298.15 K was estimated by extrapolation to 0 K and incorporation of an estimated magnetic contribution. These heat capacity results were incorporated with results at lower temperatures in the later analysis of Stuve et al. [78STU/FER]. [66BOD/DEH] Bode et al. investigated the reactions and reaction products on the nickel oxide electrode of the alkaline storage batteries. The following aspect of Bode et al.'s model is well established by now: the charged (NiOOH) as well as the discharged (Ni(OH)2) material of nickel oxide hydroxide electrodes can exist in two forms. In this paper the synthesis and structure of a-Ni(OH)2 has been described and its formula 3Ni(OH)2-2H2O was determined. This nickel hydroxide phase consists of brucite type layers of Ni(OH)2 with larger layer distances than those of P-Ni(OH)2. This is a result of interlayers of water. The definitive characterisation of its structure was not possible, because no single crystals could be grown. [66BUR/IVA] The hydrolysis of nickel(II) has been studied at (25.0 ± 0.1 )°C in 3 M (Na)NO3 medium by a method identical to the procedure described in [65BUR/LIL]. The nickel(II) concentration ranged from 0.2 to 1.0 M, which resulted in a change of ionic strength from 3.2 M to 4.0 M. The analysis of experimental data indicated the presence of Ni3(OH)3+ ( l o g , 0 X 3 = -(21.5810.03)) as major and Ni2OH3+ (log ] 0 7? l 2 = - ( 9 . 6 + 0.2)) as minor complexes. Since no spectroscopic evidence was found up to 4 M nitrate concentration, for the formation of the species NiNO,, the formation of binary or ternary nitrato complexes was not considered by the authors. In fact, a substantial amount of nitrato complex is present at 3 M nitrate concentration, therefore the reported constants are underestimated. In any other medium (NaClO4, NaCl, KC1, NaBr), the tetramer is reported to dominate, and the formation of a trimer has never been confirmed by others; therefore, we re-evaluated the reported experimental data. Considering the suggested hydroxo complexes and neglecting the formation of NiNO 3 , logl0 /? 33 = -(21.61 ±0.03) and log10 */312= -(9.32 ±0.11) can be calculated, in good agreement with the reported values. However, an equally good fit to the experimental data can be achieved by assuming the presence of N i ^ O H ) ^ , NiOH+ and Ni2OH3+ (log l0 /? 44 = - (28.00 ± 0.04), log10 */?i,i= -(9.70 ±0.11) and l o g 1 0 X 2 = - (9.28 ± 0.2)). An attempt was made to consider the formation of NiNO3 species. However, the tentatively accepted equilibrium constant for the reaction, Ni2+ + NO; ^
NiNO;,
(A.26)
logl0 P° ((A.26), 298.15 K) = (0.5 ± 1.0)) represents only the higher limit of the 'true' constant due to the neglected medium effect. Using the SIT approach, logl0 /?, (A.26),
A. Discussion of selected references
323
298.15 K) = - (0.3 ± 1.0) can be calculated for 3.31 m NaNO3 medium. Accepting this value, l o g , 0 X = -(26.52 + 0.05), log10 */?,,,= -(9.14 + 0.12) and logI0 > l 2 = - (8.77 + 0.25) were obtained. These constants are much higher than the values of [65BUR/LIL] and [65BUR/LIL2], which is very likely the consequence of the overestimated formation constants of NiNOj. Fixing the value of logl0 /? 44 at -27.4, log]0 * # . = - (9.49 ± 0.15), log10 % = - (9.02 ± 0.15) and logl0 fit ((A.26), 298.15 K) = - (0.84 + 0.04) can be calculated. Considering the great uncertainty of log10 fi° ((A.26), 298.15 K), none of the above values are accepted for use in this review. [66FLO] The formation and stability of the NiCl(H2O)J complex have been studied by a spectrophotometric method, in 10 M Li(NO3, Cl) medium at 303 K, to explain the polarographic behaviour of nickel(II) in concentrated chloride media. The concentration of LiCl was varied between zero and 10.0 M. This fundamental change in the ionic medium results in a large uncertainty in the reported formation constant (/?, = (0.094 ± 0.009)), although the author noted that the absorption spectra of 0.03 M nickel(II) in 10 M LiCl, 10 M HC1, 5 M CaCl2 and 5 M MgCl2 are almost identical. This indicates a relatively minor impact of the medium effect. Therefore, logl0 /?, = - (1.0 + 0.5) is accepted in this review. The ionic strength applied is too high to extrapolate the reported constant to infinite dilution using the SIT. In addition, LiNC>3 can not be regarded as an 'innocent' background electrolyte, since nickel(II) forms an ion pair with the nitrate ion. An attempt to determine the stability constants of the nickel bromo complexes was complicated by the slow formation of bromine from oxidation of bromide by the nitrate ion. However, the results indicated that the bromo and chloro complexes of nickel(II) have similar stability. [66GOL/RID] In this paper, heats of solution in water at 298.15 K were reported for a-NiSO4-6.000H2O(s) (final solution compositions 0.00009 m to 0.051 m) and for a mixed hydrate with a nominal composition of NiSO4-5.059H2O (final solution compositions 0.02 m). The mixed solid was assumed to be a mixture of NiSO4-6.000H2O(s) and NiSCVH2O(s), and the enthalpy of hydration of NiSO4-H2O(s) to the hexahydrate was calculated. The standard enthalpies of solution for the solids were calculated using heat of dilution values reported by Lange [59LAN], and Lange and Miederer [56LAN/MIE]. The enthalpy of solution of NiSO4-6.000H2O(s) in water was reported as A sol //° ((A.27), 298.15 K) = 1.15 kcalmol ' (4.81 kJmol"'). NiSCy6H2O(cr) ^
Ni2+ + SOf + 6H2O(1)
(A.27)
324
A. Discussion of selected references
The largest uncertainty in this value comes from the extrapolation to / = 0. Measurements by Lange and co-workers [56LAN/MIE] on CdSO4, CuSO4, ZnSO4, and NiSO4 lead to values for the integral heats of dilution that approach 7 = 0 with a slope almost twice the expected Debye-Hiickel limiting slope, AL. If an extended DebyeHilckel treatment (such as the SIT [97ALL/BAN]) is used to describe the behaviour of a strong 2:2 electrolyte, the electrolyte will be found to behave as if it is associated (as we have noted for NiSO4). However, because higher order electrostatic terms have been omitted, the extended Debye-Hiickel theory is inadequate for extrapolation of enthalpies of dilution and heat capacity values to / = 0 [97MAL/ZAM], [98ARC/RAR]. Thus, the large slope found by Lange and Miederer [56LAN/MIE] is not unexpected. In Figure A-16 the original data are plotted versus the square root of the ionic strength. The experimental enthalpies of dilution for NiSO4 reported by Lange and Miederer [56LAN/MIE] were subtracted from Asol#ra values of [66GOL/RID], resulting in values of Asol//° . The mean value obtained from Equation (A.28) A sol //: (A.27) = Asol//m (A.27) - *L
(A.28)
is A sol //° (A.27) = (4.485 ± 0.200) k J m o l ' , and this value is accepted in the present review. For comparison ^L was calculated with Equation (IX.44) of [97ALL/BAN], where the last term had to be neglected because the temperature derivative of e(Ni2+, SO^) is unknown. Figure A-16 shows that relying on the Debye-Hiickel limiting law leads to an error of- 1.7 kJmol ' in the enthalpy of solution of NiSO4-6H2O(cr) at infinite dilution. It seems likely that the NiSO4-5.059H2O was a mixture of NiSO4-6H2O(s) and NiSO4-4H2O(s) rather than a mixture of the hexahydrate and monohydrate. Jamieson et al. [65JAM/BRO], using samples dehydrated at 100°C, noted changes in dissolution kinetics and enthalpies for hydrated nickel sulphates with less than 29% water (the water content of the tetrahydrate would be 31%). They also showed that vacuum dehydration at 40°C gave products with a different heat of hydration per mole of water of hydration. Goldberg et al. [66GOL/RID] reported similar difficulties in dissolving more highly dehydrated samples, and presented no proof that dehydration of the hexahydrate over P2O5 led to formation of the monohydrate.
A. Discussion of selected references
325
Figure A-16: Molar enthalpy of solution of NiSO4-6H2O(cr) at 25°C. Experimental data from Goldberg et al. [66GOL/RID] o; dotted curve: calculated according to Equation (IX.44) in [97ALL/BAN]; dash dotted curve: calculated according to the chord method, see Equations (8 - 2 - 19) and (8 - 2 - 20) of Harned and Owen [58HAR/OWE]; solid curve: calculated with an adjustable Debye-Huckel parameter, AL, and a term linear in /, Asol//° ((A.27), 298.15 K) derived from dilution enthalpies [56LAN/MIE] x ; mean value of Asoltf° ((A.27), 298.15 K) *, dash line.
[66KEN/LIS] This paper reports calorimetric results for the heats of formation of (among others) NiCl+ and NiBr+ complexes at an ionic strength of 2.0 at 298 and 313 K. In a typical experiment 350 cm3 of the nickel(II) perchlorate solution (0.67 M) was mixed with 50 cm3 2.0 M sodium halide solution. In the first step, values of the stability constants of the complexes formed were taken from two earlier papers from Lister's laboratory [60LIS/ROS] and [61LIS/WIL]. In [60LIS/ROS] the authors measured (among others) the emf of the cell Ag,AgCl | NaCl+M(ClO4)2+NaClO41 NaCl+NaClO41 AgCl,Ag (where M = Ni(II) or Co(II), [M] = 0.12 - 0.31 M and [Cl ] = 0.008 M) at / = 2 M (Na,M)C104 and at 285, 298 and 313 K, to determine the composition and stability of the nickel(II) complexes formed with chloride ion. The complex formation induced 1-6 mV differences between the electrodes. The experimental data were interpreted by tak-
326
A. Discussion of selected references
ing into account the formation of two complexes, NiCl+ (K[) and Ni2Cl3+(Ay. The latter complex was considered to explain a small drift in the K\ value. The experimental data reported in this earlier paper were reinterpreted in this paper by Kennedy and Lister [66KEN/LIS], as the temperature dependence of K\ and K2 determined earlier was not consistent with their calorimetric results. Taking into consideration only the formation of NiCl+, a coherent model can be given. Therefore, the authors recalculated the earlier reported Kx values for 298 and 313 K neglecting the formation of Ni2Cl3+. For the purpose of this review we re-evaluated the experimental data reported in [60LIS/ROS] for all the three temperatures. This calculation resulted in only slightly different values for 298 and 313 K than were tabulated in [66KEN/LIS]. The following formation constants are accepted in this review: logl0 fit = - (0.21 ± 0.20) (285 K), - (0.16 ± 0.20) (298 K) and - (0.14 ± 0.20) (313 K). It is worth mentioning that the impact of the medium effect on these constants, induced by the relatively high concentrations of Ni(ClC>4)2 (0.12 -0.31 M) used, is probably considerably lower than for most of the studies reported in Table V-12. From the equilibrium constants at 285 and 298 K, a considerably higher reaction enthalpy (A r // m = (5.6 + 1.4) kJmol 1 ) can be calculated using the integrated van't Hoff equation than from the data of 298 and 313 K (A r // m = (3.1 + 1.2) kJ-mol '). The latter value is closer to the calorimetrically determined reaction enthalpies (A r // m = (2.1 + 0.3) kJmol ' and (1.8 ± 0.2) kJmol ' at 298 and 313 K, respectively). In the present review we re-evaluated the calorimetric data at 298 K of the equilibrium: Ni 2 + +Cl ^ N i C l +
(A.29)
using the recommended value for log10 fi° and Ae((A.29), NaClO4). At / = 2 M, /?, = 0.1 can be calculated. Since the authors used self-media for the calorimetric measurements, by mixing 0.632 M Ni(C104)2 and 2 M NaCl solutions, the /?, value calculated above is only an approximation for the conditions used. Nevertheless, using this constant A r // m ((A.29), 298 K) = (8.1 + 2.5) kJ-mol ' can be derived. [66KOH/ROD] The nickel mineral discovered at Lemieux Township, Gaspe Peninsula, Canada was investigated by X-ray, wet chemical, spectrographic, infrared, differential thermal, and petrographic analyses. It was found to be a previously undescribed nickel, magnesium, iron carbonate. It was named after the locality of its discovery. [66LOF/MCI] Lofgren and Mclver carried out measurements for the cells Mg,MgF2|CaF2|NiF2|Ni and Mg,MgF2|CaF2,YF3,(3% by weight)|NiF2|Ni. The unweighted average of the values from the three sets of measurements judged by the authors to be unaffected by moisture is ArGm (A.30)= - (451.90 ± 0.70) kJ-mol '. NiF2(cr) + Mg(cr) ^ MgF2(cr) + Ni(cr)
(A.30)
A. Discussion of selected references
327
A third law analysis of the results using auxiliary values from the present review, and with heat capacity functions for Mg(cr) and MgF2(cr) consistent with those in the CODATA review [89COX/WAG], leads to A f //;(NiF 2 , cr, 298.15 K) = -(657.0+1.4) kJ-mor1. The authors also determined potentials for the cells: Mg,MgF2|CaF2,YF3,(3% by weight)|UF4,UF3|Pt and Pt,UF3,UF4|CaF2|NiF2|Ni and Pt,UF3,UF4|CaF2,YF3(3% by weight)|NiF2|Ni. The combined measurements from these two cells were consistent with, indeed almost identical to, those from the simpler Mg,MgF2|CaF2|NiF2|Ni cell. Results reported for the cell Fe,FeF2|CaF2|NiF2|Ni cannot be used without a proper re-evaluation of the thermodynamic quantities for FeF2(cr), and such an evaluation is outside the scope of the present review. [66SHI/FUJ] A solution of K2Ni(CN)4 was irradiated with y-rays for 15 minutes, then the chemical species in the irradiated sample (labelled with 57Ni by the reaction 58Ni(y,n)57Ni)) was separated by paper electrophoresis. The authors detected four main peaks. Beside the Ni2+ and Ni(CN)5" ions, the presence of Ni(CN)2 and Ni(CN)3(H2O) species was also suggested, though the assignment of the peaks is rather speculative. [66STO/ARC] Results were reported for low-temperature calorimetric heat capacity measurements of samples of hydrated nickel sulphate with compositions NiSO4-6.010H2O, NiSO4-6.867H2O and NiSO4-7.301H2O. These were combined with results from earlier heat-capacity measurements [41STO/GIA], [64STO/HAD] to determine smoothed sets of entropy and heat capacity values from 1 to 300 K for a-NiSO4-6H2O and NiSO4-7H2O. The authors corrected the measurements for NiSO4-7.301H2O for the heat of fusion of ice (5.879 kJmol ' at 269.75 K) on the basis of solubility data at the temperature of the NiSO4-7H2O-ice eutectic [39ROH]. The heat of dehydration, NiSO4-7H2O ^
NiSO4-6H2O + H2O(NiSO4(aq, sat)) 1
was reported as (7.98 ± 0.21) kJ-mol" . The authors also measured the enthalpies of solution of samples of NiSO4-6.000H2O and NiSO4-6.940H2O in water such that the final solutions had essentially the same compositions (0.1 M in NiSO4). Thus, in that medium, the enthalpy of the dehydration reaction was (7.68 + 0.10) kJ-mol '. The authors assumed that this enthalpy difference was unaffected by the final medium, and that it was equal to the difference at infinite dilution in water. This value is accepted in the present review, and used in the calculation of the standard enthalpy of solution of the heptahydrate.
328
A. Discussion of selected references
The authors also carried out vapour pressure measurements to determine the Gibbs energy of hydration of a-NiSO4-6H2O at 293.15, 298.15 and 303.15 K. Their results were within the experimental uncertainties of values determined by others [23SCH], [35BON/BUR]. From the enthalpy of solution measurements and the vapour pressure measurements the difference between 5° (NiSO4-6H2O, a) and 5° (NiSO4-7H2O) was calculated, and compared to the difference in the S° values calculated from the low-temperature heat capacity measurements. The differences were of the order of 0.7 J-Kr'-mor1. This is small considering the many experimental quantities used in the calculation. Though they are dependent on fewer assumptions, and are therefore the basis of the selected values in the present review, the third-law entropy values for the two hydrated solids may incorporate compensating errors. Therefore, an uncertainty of 1.0 J-K '-mol ' is estimated in the present review for the values of 5° for the two solids at 298.15 K. For NiSO4-6H2O, the uncertainties in the heat capacity values near 298.15 K are estimated in this review to be ± 1.0 J-Kr'-mor1, and the same uncertainty could be assigned to values for C°m(NiSO4-7H2O) for temperatures below, but close to the NiSO4-7H2O-ice eutectic. However, because of the experimental problems encountered near room temperature, an uncertainty of ± 4.0 J-K~'-mor' is assigned to C° m (NiSO4-7H2O, 298.15 K). [66VOL/KOH] Vollmer et al. determined the enthalpy of fusion Afus//° = (17.15 ± 0.42) kJ-moP1 and the heat capacity of nickel in the liquid region up to 1820 K. Based on the earlier work of one of the authors, R. Kohlhaas, the heat capacity of nickel in the solid state in the temperature range 300 to 1725 K is also reported in this paper. The reported values were used to determine the thermal heat capacity function recommended by this review. The heat capacity of nickel in the liquid region, C° m (Ni, 1) = 39.0 J-K '-mol', was found to be independent of the temperature. [67ANT/WAR] A mixture of metallic nickel and nickel oxide was treated with a CO2/CO gas mixture at temperatures ranging from 853 to 1289 K. The partial pressure of CO in the equilibrated gas mixture was measured by means of a 14C tracer method after the carbon dioxide had been removed by a liquid nitrogen trap. The authors found for the temperature dependence of the partial pressure ratio: log,0(/>co I pCQ ) = - 2381 (KIT) - 0.025. From this relationship, the Gibbs energy of formation of NiO was derived: AfGm (7) = (-236.81 + 0.087446 (77K)) kJ-mol1, 853 < 77K< 1289. [67BLA] Blaszkiewicz determined the solubility of Ni(OH)2 by a 63Ni tracer method and found [Ni(II)] = 1.49 x 10~5 mol-drn'3. Although this value is an order of magnitude lower than Almkvist's, the mass and charge balance with Ni2+ and NiOH+ still leads to a much too
A. Discussion of selected references
329
high value of Ks 0 or else Ni(OH)2(aq) is unrealistically stable. Consequently this result was also rejected. At room temperature directly measured aqueous solubilities of nickel hydroxide are an insufficient basis for the determination of the respective solubility constant [18ALM], [67BLA]. As the system is not adequately buffered, freshly precipitated Ni(OH)2 forms colloidal particles, and the concentration of dissolved Ni(II) at the solubility minimum is at or below the instrumental detection limit [97MAT/RAI]. [67GES/NEU] The solubility of NiSC>4 hydrate is reported for temperatures from 160 to 240°C, and in ammonium nitrate solutions to 0.768 m to 300°C. Activity coefficients over the concentration and temperature range were calculated. [67NAN/TOR] Differential calorimetry has been used to measure the enthalpies of formation of the MSCN+ complexes (M = Mn(II), Co(II), Ni(II), Cu(II), Zn(II), Cd(II) and Pb(II)) in aqueous solutions at 25°C. No background salt was used to attempt to maintain constant ionic strength. The concentration of nickel(II) perchlorate varied between 5.04 and 8.9 mM, while [SCN ] lol ranged from 2.44 to 4.31 mM. Thus only the mono-thiocyanato complex was formed in the calorimeter: Ni 2+ +SCN ^ N i S C N + .
(A.31)
The concentrations of NiSCN+ formed were calculated by means of the thermodynamic stability constant reported in [62WIL2] and the Davies equation. For the purposes of the present review, the experimental data given in [67NAN/TOR] have been re-evaluated using the recommended values for logl0 /?,° (A.31) and AE((A.31), C1O4). The recalculated reaction enthalpy (-(9.25 + 0.40) kJ-mol') is somewhat higher than the original value (- (10.46 + 0.40) kJ-mol"1). Although, these values refer to NaClO4 medium, the ionic strength in [67NAN/TOR] was always less than 0.063 M. Therefore, the recalculated reaction enthalpy is accepted with an increased uncertainty (- (9.25 ± 2.00) kJ-mol 1 ). [67RUD/DEV] Rudzitis et al. used fluorine bomb calorimetry to determine the enthalpy of formation of anhydrous nickel fluoride as -657.7 kJ-mol ' at 298.15 K. During the reaction, only approximately 50% of the nickel underwent reaction, and the fraction of unreacted material was determined by mechanical and magnetic separation. Despite these difficulties, the authors estimated that the overall 2a uncertainty in their value of A f //° (NiF2, cr, 298.15 K) was 1.7 kJ-mol 1 (0.4 kcal-mol"1), slightly greater than the statistical uncertainties of the eight heat of combustion measurements. Because of the need to separate the nickel fluoride from the residual nickel at the end of the experiments, a bias in the final enthalpy of formation value cannot be ruled out. The paper provides an assessment of earlier fluoride enthalpy of formation determinations.
330
A. Discussion of selected references
[67SIG/BEC] The determination of the complexation constant between Ni2+ and HPO2, was almost incidental to the main set of experiments, which were intended to compare the effects of addition of bipyridyl and organophosphate ligands on nickel complexation. Nevertheless, this is one of the studies most often cited when an association constant is required for nickel-phosphate complexation (and the values were cited and used by the same group in later publications, e.g., [81BAN/KAD]). Potentiometric titrations were carried out at 25°C using solutions 0.01 or 0.02 M in nickel and 2 x 10 4 M in HPO 2 ", with 0.02 M NaOH(aq) as titrant, in a 0.1 M perchlorate medium (NaClO4(aq)). To determine the complexation constant, the titration curves for solutions containing the metal and ligand were compared with those with the metal or ligand omitted. The primary data were not reported except in a figure showing the smoothed data on a coarse scale, and, therefore, the data analysis cannot be properly evaluated. Also, the stoichiometry of the complex was not clearly established. The value logl0 K (A.32) = 2.08 is given for the equilibrium constant for Ni2+ + HPO2" ^ NiHPO4(aq)
(A.32)
but the estimated uncertainty is not reported. In the same paper, the association constant value for the reaction: H + + HPO^ ^
H2PO4
(A.33)
is reported as (6.70 ± 0.02) in 0.1 M NaClO4(aq), whereas the value calculated using the SIT (cf. Appendix B) is (6.781 + 0.015). Correction to zero ionic strength (Appendix B) gives log10 AT°(A.32) = (2.97 ± 0.30), where the uncertainty has been estimated in the present review. The small As correction has been based on the value for e(Na+, HPO 4 ") from Table B-5 (rather than using the equation in Table B-6), and e(Ni2+, ClO^) = 0.37 kg-mol'. [68AND/KHA2] This is qualitative work on the complex formation of nickel(II) in concentrated aqueous HBr (0.1 - 18 M) and HI (0.1 - 11 M) solutions, done by comparing the electronic absorption spectra of nickel(II) at varying HX concentrations with those of crystalline Ni(II) complexes. The authors suggested the formation of NiBrA2"A (x = 1 - 6) and Nil2"1 (x = 1 - 4) complexes. Considering the molar absorption coefficient of NiBr42~ complex in nitromethane (£rnax(698 nm) = 173 [59GIL/NYH]), only the formation of approximately 5-10 % NiBr42" complex can be estimated in 18 M HBr solution from the published electronic spectra, while the higher bromo complexes can not be justified. On the other hand, the spectra presented suggest, indeed, the formation of a Nil2" complex in considerable amounts in 9 - 11 M HI solutions. The absorption coefficients reported for 11 M HI solution (£mm(~ 800 nm) = 400, £;mx(~510 nm) = 2600) suggest the formation of a tetrahedral complex.
A. Discussion of selected references
331
[68ARN] Calorimetric studies by acid titration of hydrolysed nickel(II) solutions have been performed at 25°C in 3 M (Na)Cl ionic medium in order to obtain the enthalpy of the hydrolysis reactions. The experimental data have been interpreted using the hydrolysis mechanisms proposed by Burkov and Lilich [65BUR/LIL2] and Biederman and Ohtaki (cited as a personal communication to the author [68ARN]). The latter values (log10X4= -(28.57 + 0.03), log10 */?,.= -(10.5 ±0.1) and lo gl0 */? 12 = - (10.3 ± 0.2)) are slightly different from those reported in [71OHT/BIE]. The concentration of nickel(II) ranged from 0.055 to 0.750 M (/ = 3.055 to 3.750 M). A value of (180 ± 2) kJ mol ' was obtained for the heat of formation of the main hydrolysis product Ni 4 (OH)4 + , irrespective of which assumptions were made about the minor species. On the other hand, the rather uncertain enthalpy values for the formation of the possible minor species (NiOH+, Ni2OH3+) strongly depended on the model (formation constants) used. The best fit to the data was obtained using the hydrolysis model proposed by Biederman and Ohtaki: \Hm= (64.4 ± 6.0) and (16.0 ± 8.4) kJ-mol' for the formation of NiOH+ and Ni2OH3+, respectively. The formation constants used in the calculations were obtained without considering chloride complexation. Therefore, for the purpose of this review we re-evaluated the reported experimental data, using the selected formation constants for 3 M chloride medium and assuming a value of (8.2 ± 2.5) kJmol ' for the formation of NiCl+ under the conditions used (the latter value was calculated from the data measured at / = 2 M [66KEN/LIS]). The recalculation resulted in the following enthalpy of reaction values, valid for 298 K and / = 3M, A r // m = (54.3 + 2.0) kJmol ', (45.9± 1.6) kJmol ' and (192.9 ± 1.0) kJmol ' for the formation of NiOH+, Ni2OH3+ and Ni 4 (OH) 4 + , respectively. [68CAM/ROE] This paper describes a study of the stability of olivine and pyroxene in the Ni - Mg - Si - O system at temperatures of 1300 - 1500°C at controlled oxygen fugacities using gas mixtures of known mixing ratio. The Gibbs energy of formation of nickel oxide and the Gibbs energy change for the reaction: 2Ni + SiO2 + O2 ^
Ni2SiO4
determined in this temperature range coincide very well with results of all emf-studies discussed by this review. Moreover, a close approach to ideal behaviour of the magnesium - nickel olivine solid solutions was reported. [68CHA/FLE] A "closed system" solid electrolyte electrochemical cell has been designed to investigate the thermodynamic properties of metal oxides. This technique gave a rapid cell response and highly stable potentials through the elimination of mixed potentials. The
332
A. Discussion of selected references
standard Gibbs energy of formation of NiO was found to be the following function of the temperature: A f G°(7) = (-233.651 + 0.085 (77K)) kJ-moP1. The accuracy calculated from the maximum deviation from the computed least-squares line is equal to ± 0.209 kJ-mol"1. This corresponds to the maximum deviation in cell potential (emf) of+l.OmV. [68KOL/MAR] The rate of formation and dissociation of Ni(CN)2." has been studied at 25°C and / = 0.1 M NaClO4. The rate of formation between pH 5.5 to 7.5 showed a second-order dependence on both [CN ] and [HCN], at higher pH the reaction order in [CN~] was greater than 2, but the total cyanide dependence remained fourth order. The fact that the formation rate depended on [HCN]2 suggested that the reaction product might be Ni(HCN)2(CN)2(aq). Equilibrium measurements, using a spectrophotometric method between pH 4 - 5, resulted in systematic change in the calculated log10 fiA values. This was taken as further proof for the presence of protonated species, although the UV-VIS spectrum of Ni(CN)4~ was found to be unaffected by the protonation. The evaluation of combined kinetic and spectrophotometric data resulted in logl0 /?4 = (30.5 + 0.3) and three protonation constants of the tetracyanonickelate(II) species: log10 KH 0 = (5.4 ± 0.2), log10 Km = (4.5 ± 0.2), log10 KH2 = (2.6 + 0.2), for the reactions: Ni(HCN)v(CN)4I2 +H + ^
Ni(HCN), +I (CN)£:i
(where x = 0, 1,2). On the other hand, spectrophotometric [59FRE/SCH] and potentiometric [63CHR/IZA], [71IZA/JOH], [74PER] studies performed in the same pH-range did not indicate the existence of protonated species in notable concentration, although their kinetic role in the acid dissociation of Ni(CN)4~ was later corroborated [76PER]. The value for logl0 /?4 is accepted, but with an increased uncertainty (± 0.6). [68KOS] Kostryuka and Zarubina measured the heat capacity of NiCl2 between 1.8 and 16 K to ascertain the singularities of the energy spectrum of layered antiferromagnets. The data have been presented graphically only. [68LAR/CER] Larson et al. obtained for the reaction: NiSO4-7H2O(cr) ^
Ni2+ + SO2" + 7H2O(1)
(A.34)
AsolG° (A.34) = 12.929 kJ-mol ' (3.09 kcalmol 1 ) which agrees very well with the value recalculated in Section V.2.1.3.1. A misprint seems to have occured in this paper, as the standard enthalpy of reaction (A.34) is given as Asol//° (A.34) = 2890 instead of 2986 cal-mol"1. The latter value follows from the data of [66STO/ARC] and the original value given by [66GOL/RID].
A. Discussion of selected references
333
[68MAL/TUR] The instability constants of NiSCN+ were determined at 15, 25 and 35°C by a polarographic method in aqueous solutions at / = (0.65 ± 0.03) M using (H)C1O4 as background electrolyte. The thiocyanate concentration ranged from 0.5 mM to 7 mM, with [Ni2+]tot = 55.7 mM, thus the formation of higher complexes (Ni(SCN)« x > 1) was neglected. The determination is based on the measurement of the decrease in catalytic polarographic current of the Ti(IV) - SCN complexes in the absence and in the presence of nickel(H). Although, the authors applied rather acidic conditions, and thus the thiocyanate ion was partly protonated, the reported values are accepted with increased uncertainties. From the temperature dependence of log,,,/?, (see Figure A-17) ATHm = - (14.5 + 1.2) kJmol ' can be derived.
Figure A-17: Temperature dependence of log,,,/?, values of the NiSCN+ complex reported in [68MAL/TUR].
[68NAV/KLE] The enthalpies of formation from the oxides of thirty-two 2-3 and 2-4 spinels, among others Ni2Ge04, have been determined by solution calorimetry in molten oxide solvents at 970 K. Regularities in the thermodynamics of spinel formation and the olivine spinel transition for orthosilicates of Mg, Fe, Mn, Co, and Ni have been discussed. [68PRA] This is a description of a general method used by Prasad and co-workers to carry out potentiometric measurements to determine complexation constants, and describes how the measurements were extrapolated to / = 0. For NiSC>4 only a single value is reported (K\ l = 9.0 x 10 3 at 35°C). This suggests weaker complexation than was found in other
334
A. Discussion of selected references
work (e.g., [59NAI/NAN]). For experimental details and results, reference was made to unpublished work by S.C. Sircar, P.K. Jena and B. Prasad. However, no related paper on nickel sulphate complexation by these authors seems to have been reported in Chemical Abstracts, and the work may never have been published. The unavailability of the primary data precludes recalculation of the results using the standard SIT procedures, and the reported value is not used in the present assessment. The data reported for the nickel(II) - thiocyanate system in this paper, were later published in detail in [7 IDAS/DAS]. [68VOG] This paper presented evidence for the existence, in potassium hydroxide melts (0.85 w KOH) between 498 and 523 K, of reversible homogeneous nickel redox couples in which nickel exhibits valencies greater than two. The data can be interpreted by the following two simultaneous redox reactions: Ni2+ ^ Ni3+ + e~ Ni3+ + 4OH ^ NiO2(s) + 2H2O + e". The nature of the black NiO2 crystals (+4-valent nickel or peroxidic oxygen) was left undecided. [69FUN/TAN] Kinetic and mechanistic studies of the reaction between 4-(2-pyridylazo)resorcinol (PAR) and (among others) NiOH+ and NiF+ were performed at an ionic strength of 0.10 M (NaClO4) and at (298.15 ± 0.20) K. Nickel(II) perchlorate solutions were mixed with ligand solution, then borate buffer solution (8 x 10 3 M) was added to adjust the pH (7 9.5). The concentration ranges were as follows: [Ni2+] ~ 5 - 50 * 10 7 M, [PAR] ~ 10 5 M, [F ] ~ 0.05 M. The influence of the buffer was examined and it was concluded that it had no effect on the rate constants. The kinetic results also provided data for the equilibrium constant for the reaction: Ni2+ + OH" ^ NiOH+
(A.3 5)
log10 K (A.35) = (4.3 ±0.1)) and for Ni2+ + F ^ N i F + ,
(A.36)
log,0A-(A.36) = (l.l±0.1)). Using log10 ATW = - 13.78, log10 K (A.35) can be can be converted to the corresponding first hydrolysis constant of nickel(II): Ni2+ + H2O ^ NiOH+ + H+, log10 Kf — - (9.5 ±0.1). In this review, these values are accepted, but with the uncertainties increased to ± 0.2.
A. Discussion of selected references
335
[69IZA/EAT] Titration calorimetry was used to determine values of log10 K, Ar77^ and Af S°,, reported for the formation of NiSO4(aq) for 25°C and 7 = 0 . The data analysis is suspect because of a possible correlation between the value for the calculated equilibrium constant and the enthalpy of association. A good reassessment was reported by Powell [73POW], who pointed out that if the data of Izatt et al. were used with the association constant at 298.15 K set equal to the value from Nair and Nancollas [59NAI/NAN], then an enthalpy of ion-pair formation of 5.7 to 5.9 kJmol ' was obtained, not the reported 1.7 kJmol ' [69IZA/EAT]. In the present review, several approaches were tried to determine the enthalpy of association within the constraints of the SIT formulation in Appendix B. As in earlier reviews [92GRE/FUG], [2001LEM/FUG], for low ionic strengths the assumption was made that A r // m is independent of ionic strength. Powell [73POW] suggested that the effect of this approximation for sulphate complexation was < 1 J-Kr'-mol1. Izatt et al. [69IZA/EAT] and Powell [73POW] incorporated values for an association constant and enthalpy of association for the Me4NClO4(aq) ion pair in their calculations. A full evaluation of these values has not been attempted here. Determination of the values for Me4NClO4(aq) required using an explicit value for the association constant between Na+ and SOj", whereas the SIT, as described in Appendix B, would treat these interactions using an interaction coefficient rather than a specific association constant. Initially, the effect of the enthalpy of protonation of sulphate has been ignored. Using 7 = 0 values for the formation constant of NiSO4(aq), as recalculated from the data of Nair and Nancolas [59NAI/NAN] and of Katayama [73KAT], the molar concentration of NiSO4(aq) formed at each step of the titration was calculated iteratively. The average enthalpy of ion pair formation is 5.8 kJ-mol"1 and 4.7 kJ-mol 1 , respectively. However, it is apparent that the calculated enthalpy of association is strongly dependent on the ionic strength of the solution. If the values at the measured ionic strengths are extrapolated to 7 = 0, the limiting value (Figure A-18) is 1.2 kJ-mol' or 0.4 kJmol ' depending on whether the association constant data of Nair and Nancolas [59NAI/NAN] or from Katayama [73KAT] are used. This cyclic inconsistency may be a result of the neglect of the heats of the secondary association reactions and heats of dilution. It may also represent a failure of the SIT to adequately represent the system. At present, the enthalpy of association as calculated by Powell is accepted, but with an increased uncertainty Ar77° = (5.8 + 2.0) kJmol '.
336
A. Discussion of selected references
Figure A-18: Heats of reaction (•) determined from the calorimetric data of [69IZA/EAT] by using the 298.15 K values of log10 AT, from [73KAT]. The line represents a least-squares fit to these enthalpies as a function of ionic strength.
[69KOL/KIL] The reaction rate and mechanism of the acid catalysed decomposition of Ni4(OH)4+ have been studied by stopped-fiow measurements. Potentiometric titrations of 0.46 M nickel(II) perchlorate solutions at 25°C and 1.5 M ionic strength, maintained by NaC1O4, were also performed to study the hydrolysis under the conditions used for the kinetic measurements. The authors suggested the presence of the Ni4 (OH)^ species, only, and logl0 /? 44 = -(27.03 ±0.06) was obtained. This value compares favourably with the corresponding value reported in [65BUR/LIL]. The reported data indicate a slight tendency for a decrease in the calculated log]0 J34 4 values with decreasing pH, which may be due to the presence a minor hydroxo complex. Therefore, the reported experimental data were re-evaluated for the purpose of this review, but no further species can be identified with any confidence. [69KOS] The measurements carried out in [68KOS] were discussed from the point of view of the magnetic spectrum of layered antiferromagnets. The molar heat capacities of NiC^ from 1.8 to 16 K have been listed.
A. Discussion of selected references
337
[69KUL/ERS] On a godlevskite specimen from the Cu - Ni sulphide deposit Noril'sk, Siberia, Russia, Kulagov et al. carried out chemical and mineralogical investigations as well as a crystal structure analysis. Whereas the mean values of the chemical analyses agreed with (Ni, Fe, Co)S the authors maintained that the upper limits of the metal and the lower limits of the sulphur contents are consistent with Ni7S6. [69LIB/SAD] Osmotic coefficients have been determined by an isopiestic method for (among others) the aqueous solution of Ni(C104)2 in the concentration range m = 0.1 - 3.5 mol-kg ' at 25°C. The mean activity coefficients were calculated from the Gibbs-Duhem equation. For the purposes of this review, these data, together with those listed in [99MAL/CAR], have been used to derive the ion interaction coefficient £(Ni2+, C1O4), see also the discussion on [99MAL/CAR]. [69MAK/SPI] Makovskaya and Spivakovskii added sodium hydroxide to 0.001 - 4.7 M NiCl2 solutions at 25°C until a slight precipitate formed. The pH and the chloride activity were measured with glass and silver chloride electrodes, respectively. The concentration of Ni(II) in the filtered solutions was determined by complexometric, photometric, and polarographic methods. Ni(OH)2, Ni(OH)) 5CI05, and Ni(OH), 75CI0.25 were detected by thermodynamic analysis, but the latter phase disappeared during aging. None of these phases were characterized by X-ray analyses. The solubility product of the freshly precipitated nickel hydroxide turned out to be approximately an order of magnitude higher than after an undefined aging period. So, only the value of the aged phase (log l0 A^°o = - 15.54) has been accepted for Table V-6 and Figure V-l 1. [69MOR/SAT] The standard Gibbs energy of the reaction: 2Ni + O2 ^
2NiO
was investigated as a function of temperature in the temperature range from 700 to about 1100°C. The Gibbs energy function reads: AfG° (7) = -476.976 + 0.175770 (77K)) kJmol '. The uncertainty of emf measurements, caused by different kinds of solid electrolytes, was around + 3 mV. [69RAN/BRE] The potential of nickel sulphide electrodes was measured against the saturated calomel electrode at 298.15 K. The sulphide electrodes were prepared by covering platinum with nickel sulphide which was obtained by precipitation in aqueous solutions of low acidity (1 M Na2S solution was dropped into 0.5 M NiSO4 solution). The composition of the
338
A. Discussion of selected references
sulphide was altered by anodic as well as cathodic polarisation. Unfortunately, X-ray diffraction analysis revealed that the nickel sulphides were amorphous. Hence, no reliable thermodynamic data can be evaluated from this study. [69SOR/KOS] For the calculation of S°(Ni(OH)2, cr, 298.15 K) according to Equation (A.37) the original C° m data have been plotted versus inT:
Sm = J C ^ dlnT.
(A.37)
Figure V-10 shows the experimental Cpm values of three Ni(OH)2 samples (iii, ii, i) up to the temperature where the data for each have been measured. At T Cp m (ii) > Cpm (i), but at temperatures higher than at the minimum of Cp m = f (InT), this order is reversed. Using the CurveExpert routine the functions C pm (ii) = f(ln7) and Cpm(i) = f(ln7) were fitted to linear splines and integrated according to Equation (A.37) [97HYA]. Finally, the Debye T3 extrapolation down to T= 0 was applied. As no data above T = 104 K were measured for sample (iii), 0.4 and 1.2 J-K '-mol' were subtracted from C pm (ii) and C pm (i) above 104 K, respectively, and the values thus obtained were added to series (iii) [69SOR/KOS]. Again the CurveExpert routine was used to integrate the true as well as the dummy part of this Cpm (iii) function. [69SUB/COR] The rate of extraction of nickel(II)- and zinc(II)-dithizonates into CHC13 has been studied in presence of several auxiliary ligands (L), e.g., thiocyanate, at 25°C and at ionic strength of 0.25 M using NaC104. The concentrations used were not reported, but were probably sufficiently low compared to the total ionic strength, since the concentration of the Ni(ClO4)2 stock solution was 1 mM. The variation of the apparent rate constant of the extraction as a function of the concentration of the auxiliary ligand, allowed calculation the stability constants of the ML, complexes. The authors reported log10 /?, = (1.3 + 0.2) for the stability of the species NiSCN+. [70ASH] The heat of solution of nickel chloride hexahydrate into 1.0M HCl(aq) was measured, and (6.10 ± 0.22) kJ-mol ' was reported (2a uncertainties). The final solution was 0.02 M in Ni. A double correction to 0.0 M HCl(aq) and to / = 0 [95MAN/KOR] would be required to use the reported value to calculate the enthalpy of solution of the salt at infinite dilution in water. The corrections themselves have large uncertainties, and this heat of solution value is not used further in the present review.
A. Discussion of selected references
339
[70BIN/HEB] The authors reported drop calorimetry results obtained from samples initially at temperatures between 381.9 K and 1461.7 K. The scatter in the enthalpy difference results, based on measurements done at similar temperatures, seems to be of the order of 200 kJ-moF1. Direct fitting of the drop calorimetry results without incorporating some low temperature heat capacity constraint leads to a poor extrapolation for temperatures below 400 K. The authors fitted an equation to their values of ( # m ( r ) - / / ° ( 2 9 8 . 1 5 K ) ) , "weighting the point at 298.15 K for the best fit of all results". This seems to have entailed the use of some of the heat capacity results of Catalano and Stout [55CAT/STO]. The reported value of C;ro(298.15 K) (62.93 J K ' m o l ') differs by 1.1 J-IC'mol ' from the value of [55CAT/STO], though the S° value from that source is used. Also, the expression reported by Binford and Hebert [// m (7')-//°(298.15K)]/J-mor 1 = 4.184-(15.677(77K) + 1.7344 x 10 3 (77K)2 + 1.4872 x 105(77K) ' -5161.4) is non-zero (693 J-mol ') for 298.15 K. In the present review, the drop calorimetry results are used together with the heat capacity results of Catalano and Stout [55CAT/STO] to derive a properly constrained expression for (// n ,(r)-// n °,(298.15K)) and C°pm(T). Above BOOK, the experimental values of ( / y m ( r ) - / / ; ( 2 9 8 . 1 5 K ) ) appear to be slightly larger (by 0.5 kJmol ' to 1.3 J K " ' m o r ' ) than might be expected by extrapolation of values from lower temperatures. Because there are only two measurements in that temperature range, these values have not been treated differently from the other measurements in the fitting process. [70CAR/LAI] The solubility product of NiS was determined by linear voltage sweep voltammetry using a mercuric sulphide coated electrode (hanging mercury drop). The peak potential for the exchange reaction between the mercuric sulphide coated mercury electrode and Ni2+ ions of a nickel perchlorate solution, HgS + 2e + Ni2+ ^
Hg + NiS
(A.38)
yields the ratio between the solubility constants of HgS and NiS. At 298.15 K the authors obtained log10 A"°o= - 17.8 for the solubility product of NiS. Unfortunately, neither any further experimental details nor the second dissociation constant for H2S(aq) are indicated in this contribution. Thus, the data cannot be recalculated and compared with results of different studies. [70EFI/KUD] The enthalpies of solution in water of nickel chloride, cobalt chloride and the coordination compounds RbNiCl3, RbCoCl3, Rb2CoCl4 and Rb3CoCl5 were determined in a calorimeter with an "isothermal jacket". The authors reported for the reaction,
340
A. Discussion of selected references NiCl2(cr) ^ Ni2+ + 2C1
(A.39)
an enthalpy of solution of- (83.26 + 0.42) kJ-mor1 for 10000 ^ n(H2O) / «(NiCl2, cr) ^ 40000); this leads to a value of Asol//° (A.39), between - 83.62 kJmol ' and - 83.93 kJ-moF1, which is approximately 1 kJ-mol ' less negative than the value reported in [90EFI/FUR]. The latter (-(84.89 + 0.27) kJmol 1 ) has been better documented and thus is accepted in this review. [70HAL/VAN] The formation of nickel(II) chloro complexes has been studied at 298.15 K in 4 M Na(GO4,Cl). The solution containing a known total concentration of nickel (0.1, 0.2 or 0.5 M) and chloride (0 - 2.0 M) was saturated with AgCl labelled with radioactive "°Ag. The measured radioactivity was correlated with the free chloride concentration taking into account the solubility of AgCl and the different chloro complexes of silver^). The experimental data have been interpreted in terms of the unique formation of the bis-complex NiCl2 (i.e., /?, = (0.000 + 0.015) M '; /32 = (0.129 + 0.002) M~2). This interpretation solely in terms of the formation of the NiCl2 complex is questionable, and probably was adopted because of effects from the marked change of the ionic medium during the measurement. The medium effects, however, are less than in most of the other work on chloro complexation of Ni(II). Therefore, we re-evaluated the published experimental data. Taking into account only the bis-complex, a better fit can indeed be achieved (log10 Pi — - (0.89 + 0.02)), compared to the result obtained using the assumption of the unique formation of NiCl+ (log, 0 /?, = - (0.62 ± 0.04)). Nevertheless, in this review we prefer the latter value with a higher uncertainty (log, 0 fii = - (0.62 ± 0.30)). This choice is also confirmed by several equilibrium studies performed in non-aqueous solvents, which have higher precision since the complexes are more stable than in aqueous solution. For example, in methanol, ethanol and propanol, Kt is much higher than K2 [95KAH/CRO]. Using the recommended values for £(Na+, Cl"), e(Ni2+, C1O~) and the e(NiCl+, CIO4) derived from the data published in [75LIB/TIA], the extrapolation to / = 0 resulted in logI0 P° = (0.68 ± 0.40). The experimental work is well documented in this paper, which offers the possibility of using the SIT to assess the medium effect caused by the replacement of the original background electrolyte. The activity of a cation M of charge zM in a mixture of two electrolytes (NX and NY, with a molality mx and mY) at constant ionic strength can be expressed using the SIT as follows: logio 7M = - z2MD + e(M, X) mx + e(M, Y) mY where D is the Debye-Hilckel term. The same equation for an anion Y is: logio TY = ~ A D + E(N, Y) (mx + mY). Substituting the log10 ^l values in Equations (B.3) (Appendix B) with these expressions, applying it to the equilibrium:
A. Discussion of selected references
Ni2+ + Cl ^
341
NiCf
and rearranging leads to: log10 # +4D- {s(Na+, Cl) + e(Ni2+, C104 )}/,„ - e(Ni2+, C1O4) mY - s(Ni2+, Cl") mY = log,0/?,° ~E(NiCl+, C10 4 )/ m + e(NiCl+, CIO,") mY - s(NiCl+, Cl) mY provided that N = Na+, M = Ni2+, X = CIO;, Y = Cl and lm = mx + mY. Since all quantities on the left-hand side of this equation are known, it can be written: Q = logl0 P° - s(NiCl+, C1O4 )/,„ + e(NiCl+, ClO^) mY - s(NiCl+, Cl") mY The three unknown quantities on the right-hand side are strongly correlated with each other; thus, to solve this equation we should accept the e(NiCl+, ClO^) value derived from the data published in [75LIB/TIA]. Then the equation is transformed into ff = log]0 ft - E(NiCl+, CD mY
(A.40)
Plotting the Q' values vs. ma gives a straight line, with the intercept and slope equal to log l0 /?° and — e(NiCl+, Cl), respectively. Before applying this equation to the data published in [70HAL/VAN], two further simplifications should be used. With increasing chloride concentration the ionic strength of the solution expressed in molarity is constant, but in the molal scale it changes. Since no density data are available for such an electrolyte mixture, /,„ was calculated using the equation: _ M a .x fl + (4-Ma.)xfr where M a is the concentration of chloride in molar units, and a and b are the molality of 4 M NaCl and NaClC>4 solution, respectively. The density of solutions with different compositions was calculated assuming that 4 M NaCl and 4 M NaClO4 solutions mix ideally. With these assumptions, the following dataset can be calculated. The plot of Q' versus mC] is depicted in Figure A-19. From this plot log10 fi° = (0.20 + 0.03) and e(NiCl+, Cl) =-(0.16 + 0.02) are calculated. This value of logl0 P° is 0.5 log units lower than that calculated above from the accepted logl0 /?, value at / = 4 M Na(ClO4, Cl) using the conventional SIT extrapolation. This difference is, in great part, due to the medium effect. But it is also the consequence of the sensitivity of function Q' to the values of different interaction coefficients used (especially on whether the value of e(NiCl+, CIO;,) is correct), and on the validity of the assumptions made concerning lm and mC\-. In this review we accept the value obtained from the plot in Figure A-19, but with a considerably increased uncertainty: log I 0 ^° = (0.20 ±0.50).
342
A. Discussion of selected references
Table A-14: Dataset. [ci- ]lM
[crw
[Ni2+]IM
P\
IIog10/J, (molal)
0.25 0.30 0.38 0.40 0.45 0.50 0.65 0.70 0.75 0.90 1.00 1.15 1.20 1.25 1.40 1.45 1.50 1.65 1.75 1.90 1.95 2.00 2.15 0.20 0.95 1.70 0.40
0.2375 0.2835 0.3525 0.3900 0.4440 0.4651 0.6299 0.6880 0.7000 0.8640 0.9049 1.0960 1.1700 1.1241 1.3292 1.4099 1.3262 1.5570 1.5325 1.7840 1.8863 1.7362 1.9989 0.2000 0.8890 1.6106 0.3940
0.50 0.50 0.50 0.20 0.10 0.50 0.20 0.10 0.50 0.20 0.50 0.20 0.10 0.50 0.20 0.10 0.50 0.20 0.50 0.20 0.10 0.50 0.20 0.10 0.10 0.10 0.20
0.1078 0.1201 0.1335 0.1345 0.1433 0.1616 0.1770 0.1979 0.1587 0.2543 0.2596 0.3377 0.3656 0.2994 0.4120 0.4741 0.4019 0.5578 0.5024 0.7737 0.9322 0.6433 1.5441 0.0007 1.7597 5.2113 0.0785
-1.0600 -1.0129 - 0.9672 - 0.9639 - 0.9363 -0.8841 - 0.8447 - 0.7962 -0.8919 - 0.6872 -0.6783 - 0.5640 - 0.5295 -0.6162 - 0.4777 -0.4167 - 0.4885 -0.3461 -0.3915 - 0.2040 -0.1230 - 0.2842 0.0961 -3.2532 0.1529 0.6244 -1.1980
m
cr 0.3058 0.3665 0.4572 0.4873 0.5474 0.6074 0.7863 0.8456 0.9047 1.0809 1.1975 1.3710 1.4285 1.4858 1.6567 1.7133 1.7697 1.9378 2.0489 2.2143 2.2690 2.3236 2.4861 0.2450 1.1393 1.9934 0.4873
lm
D
Q'
4.9139 4.9067 4.8958 4.8922 4.8850 4.8778 4.8561 4.8489 4.8416 4.8200 4.8055 4.7838 4.7766 4.7694 4.7477 4.7405 4.7333 4.7116 4.6971 4.6755 4.6682 4.6610 4.6393 4.9211 4.8127 4.7044 4.8922
0.2610 0.2609 0.2609 0.2608 0.2608 0.2608 0.2606 0.2606 0.2605 0.2604 0.2603 0.2602 0.2601 0.2601 0.2599 0.2599 0.2598 0.2597 0.2596 0.2595 0.2594 0.2594 0.2592 0.2610 0.2603 0.2597 0.2608
0.2453 0.2754 0.2955 0.2904 0.3011 0.3364 0.3254 0.3572 0.2449 0.4000 0.3760 0.4413 0.4596 0.3567 0.4471 0.4921 0.4044 0.4993 0.4225 0.5634 0.6288 0.4523 0.7866 -1.9308 1.2236 1.4541 0.0563
A. Discussion of selected references
343
Figure A-19: Extrapolation to / = 0 of the formation constant of NiCl+ using Equation (A.40) and the experimental data published in [70HAL/VAN]. The outlying points (open diamond) were not taken into account in the calculation.
[70MAC] Potential sweep experiments were performed on film nickel hydroxide electrodes in an effort to determine the redox mechanism. The over-all reaction for a-nickel hydroxide was found to be: 2[3Ni(OH)2-2H2O] + 6OH" + | K O H ^
2[3NiOOH
^H2O-|KOH]
+ 3H2O + 6e .
The rate-controlling step was assumed to be proton diffusion. [70MIR/MAK] This paper reports association constants for the inner- and outer-sphere complexes (fix and W\, respectively) of Co(II), Ni(II), and Cu(II) hexa-aqua ions with chloride, bromide, thiocyanate, sulphate, and nitrate ions, based on spectrophotometric and solubility measurements in 3 M Li(ClO4, X) media at 298.15 K. The authors used the UV-region of the electronic spectra to determine the sum of/?) and a\, while the value of 0\ was determined from the VIS-region. This is obviously only a rough approximation. Only the thiocyanate ion is proposed to form an inner-sphere complex (approximately 10% in the case of nickel(II)). The sulphate value reported is an average of values determined by UV spectroscopy and from the effects of nickel ions on the solubility of calcium sulphate. Apparently the two techniques gave reasonable agreement, as the average value of logl0 p° was reported as (1.8 + 0.2). This value is smaller than most, and there are insufficient details in the paper to allow the reviewer to properly assess or recalculate the results. In attempting to assess the values for the present review, the SIT was used to
344
A. Discussion of selected references
calculate logl0 K° = 1.07, again indicating that the value obtained by Mironov et al. is anomalously low. The authors provided insufficient experimental details to assess the reliability of their data, although the sum of the J3\ and a\ values are in agreement with other reports. [70NAG/ING] The partial pressure of sulphur over Ni - S melts was derived from the equilibrium weight of the melt in gas streams of known H2S/H2 compositions. The mass of the melt was monitored continuously, while the composition of the gas stream was maintained constant by appropriate flow meters. Experimental results for melts saturated with nickel as well as for homogeneous melts, with compositions ranging from 33 to 42 at% S, were obtained at temperatures between 973 and 1373 K. The sulphur partial pressure for a given gas composition was calculated from the well-known equilibrium constants for the reaction: H 2 S(g)^H 2 (g) + 1/2S2(g). [70ROT] Experimental levels of the configuration 3d94p, 3d84s4p and 3d95p of (Nil) were compared with corresponding calculated values. The electrostatic interactions between the configurations were considered explicitly. [70TUR/RUV] This paper describes polarographic measurements done at (25.0 ± 0.1)°C, / = 5 . 0 M (NaC104), 1 x 10 3 M Ni(NO3)2, 0.0 - 0.2M Na2SO4 and with enough HC1O4 to adjust the pH to 3. A value of Kx = 15.6 is reported for nickel sulphate association. The ionic strength is too high (6.6 m) to properly apply the SIT. If the SIT is used with e(Ni2+, CIO;) equal to (0.37 + 0.03) kg-mol"1, log10 K° = 1.70 is calculated. [71ARI/MOR] The enthalpies of formation were determined for several nickel sulphides, including Ni3S2, NiS (various compositions of the high-temperature form) and NiS2. The results are summarised in Table A-15. No experimental details are given. The mean value for the standard enthalpy of formation of the NiAs-type NiS is calculated to be A f //° = — (85.8 ± 2.4) kJmol"1 which agrees within the limits of uncertainty with the value recommended in the present compilation ( A f / / ° = - ( 8 8 . 1 ± 1.0) kJ-mol"'').
A. Discussion of selected references
345
Table A-15: Experimental results for the enthalpy of formation for various nickel sulphides determined by calorimetric measurements. Composition
Phases
V / J (kJ-mor1)
NiSo.67
Ni3S2
-(202.1 ±3.9)
NiSo.,8
Ni3S2 + NiS(a)
-(85.0 ±2.6)
NiS,.oo
NiS(a)
- (84.9 ± 2.6)
NiSi.oi
NiS(a)
- (87.4 ± 2.5)
NiSi.02
NiS(a)
-(84.9 ±2.5)
NiSi.04
NiS(a)
- (86.5 ± 2.4)
NiSi.053
NiS(a)
-(85.4 ±2.4)
NiSi.oe
NiS(a) + NiS2
-(85.7 ±2.4)
[71BHA/SUB] The hydrolysis of nickel(II) was studied in 0.5 M NaClO4 medium at 30°C. The concentration of the metal ion ranged from 0.01 to 0.04 M, while the pH varied between 6.8 and 7.1. The graphical analysis of the data, assuming only one hydroxo species, indicated the presence of Ni2(OH)j:~ with a logl0 *fi62 = - (42.94 ± 0.04). The authors used too narrow a range of metal ion concentrations (and too narrow a range of pH for each [Ni2+]toi) to allow a satisfactory evaluation of the composition of the oligonuclear species. The reliability of the results is questionable; therefore, the results from this publication were not used further in this review. [71BON] The author earlier suggested a simple calculation method for interpreting metal-ion complexation behaviour from polarographic studies of irreversible electrode processes [70BON]. The present paper was devoted to a test of the validity of the proposed method. One of the major conditions is that the heterogeneous charge-transfer rate constant should not change markedly with addition of the ligand to the aquo complex, i.e., the removal of coordinated water also should remain the rate determining step of the reduction in the presence of ligand. This condition was found to be fulfilled in case of simple 1:1 complexes of nickel(II), due to the slow water exchange in both the aquo and [Ni(H2O)5X]+ complexes. Therefore, the proposed simple method, combined with the De Ford-Hume equation, was used to determine the stability of the monofluoro, monochloro and mononitrato complexes of nickel(II). During the measurements, to achieve a measurable effect of complex formation, the background electrolyte (NaClO4) was entirely replaced by the sodium salt of the corresponding anion. The author mentioned that, taking into account the limits of the method, the strong medium effect, and the weak complexation (especially in the cases of chloride {fl\ = 0.6) and nitrate (J3\ = 0.4)), the reported values can be considered only as order of magnitude estimates.
346
A. Discussion of selected references
[71BUR/ZIN] The hydrolysis of nickel(II) in 3 M (Na)Cl was studied at 60°C by potentiometric titrations. The nickel(II) concentration was [NiCytot = 0-2 to 1.5 M and pH range was from 4.9 to 6.3. The reported constants are log10 /? 44 = - (25.33 ±0.02) and logl0 /?, 2 = - (8.5 ± 0.2). The medium effect and thus the change of ionic strength was very important in these measurements (/ varied from 3.2 to 4.5 M, i.e., pure NiCl2 solution was used for the highest nickel(II) concentration). Therefore, the reported experimental data were re-evaluated for the purposes of this review. For the reaction: Ni2+ + Cl ^ NiCl+
(A.41)
at /,„ = 3.2 m, and T = 333.15 K, log10 # (A.41) = - 0.44 can be estimated for the NiCl+ complex in NaCl medium, accepting the values of log10 /?, ((A.41), 298.15 K.) = - 0.59 (see discussion on [65BUR/LIL2]) and A r // m = (8.2 + 2.5) kJmol ' [66KEN/LIS]. Considering the whole dataset and the same model proposed by the authors, logl0 /? 44 = - (25.29 + 0.05) (3a) and log10 */3{1 = - (8.50 ± 0.09) can be calculated, in good agreement with the published values. If only the data sets reported for [Ni2+]lol = 0.2, 0.4, 0.6 and 0.8 M are used, taking into account the formation of NiCl+, logl0 /? 44 = - (24.04 ± 0.06) (3c) and log I0 */?, 2 = -(7.92 + 0.10) can be derived. If the species NiOH+ is added to the equilibrium model, a substantial (30%) decrease of the fitting parameter was achieved, and the values log10 /?44 = - (24.09 ± 0.03) (3a), logl0 /?,, = - (8.48 + 0.10) and log]0 *fil2 = - (8.58 ± 0.40) were obtained. The latter values, with increased uncertainties, are accepted for use in this review: logl0 /? 44 = - (24.09 + 0.30), logl0 X i = " ( 8 - 4 8 ± °- 3 °) a n d loSio *A.2 = " ( 8 - 5 8 ± °- 60 )- Combining these constants with the values from [65BUR/LIL2], A r // m = (59 ± 10) kJmol 1 , (20 ±20) kJ-mol 1 and (186 ±6) kJ-moF1 are calculated for the formation of NiOH+, Ni2OH3+and Ni4(OH)4+ species, respectively (in 3.2 m aqueous NaCl). [71 CON/LOO] Connelly et al. measured the specific heat of pure single-crystalline nickel near the Curie temperature using a calorimetric technique which permits direct observation of C°pm{T) as a continuous function of T, with a temperature resolution of about 10 2 K. Special attention has been devoted to the determination of the analytical form of the magnetic contribution to C°m (7). [71DAS/DAS] Potentiometric measurements have been performed at 25, 35 and 45°C to determine thermodynamic parameters of the mono-thiocyanato complexes of five 3d metal ions, among them nickel(II). The activity coefficients of the different species were calculated by the Davies equation. Since the Davies equation is not compatible with the SIT, and no experimental data are provided, the reported thermodynamic functions were not considered in this review.
A. Discussion of selected references
347
[71HOH/GEI] A stopped-flow method was used to study the hydrolysis of nickel(II). The ionic strength of the solutions was not controlled ([Ni2+]tot = 5x10 4 to 2.5x10 3 M, [OH]lot ~ 3X10^* M); the measurement temperature was not specified in the paper. The results indicated, that using different time scales, the deprotonation of Ni(H2O)j;+ ion and the polynuclear hydroxo species can be separated. The deprotonation reaction for the mononuclear species is estimated to occur within 10 7 s, while the polynuclear complex formation needs several seconds (20 min. is reported to reach the equilibrium in [65BUR/LIL] and [69KOL/KIL]). The pH of the solutions, measured between 2 - 1 0 0 0 ms by spectrophotometry using phenolphthalein as the indicator (the value used for the pK of the indicator was not reported), was extrapolated to / = 0 s. From these data log10 /?,, = (11.0 ± 0.2) has been derived. This value has not been considered further in this review, because of the lack of experimental details. [71 ISO] Freezing point measurements were carried out in a manner similar to that of Brown and Prue [55BRO/PRU], and similar results were obtained. Analysis of the saturated solutions was done by comparison of the conductivity with the conductivity of standard solutions. An extended Debye-Huckel treatment was used to analyse the data, and values ofK] =118 to 250 dm 3 mol ' were obtained for values of the Debye-Huckel "a" parameter of 0.4 to 1.4 run. The author reported that the calculated association constant also depends on the choice of the maximum solute concentration for which the data analysis can be considered reliable. In the present review the data were reanalysed using the SIT, and only data for total nickel sulphate molalities < 0.03 were used. [71IZA/JOH] The thermodynamic parameters for the formation of several metal-cyanide complexes, among others those of Ni(CN)^", have been determined using pH-metric and calorimetric methods at 10, 25 and 40°C. In case of nickel(II), the thermodynamic data were determined by titration of Ni(C104)2 solutions with NaCN solutions. The ionic strength of the solutions were / < 0.02 M in all cases. The Debye-Huckel equation, related to the SIT model, was used to correct the formation constants to thermodynamic constants valid at / = 0. Since previous experiments indicated that the dependence of A r //° in the ionic strength in dilute aqueous solutions is small compared to the experimental error, the measured heats of reaction (A r // m = - 189.1 kJ-mol 1 at 10°C; A r // m = -183.7 kJ-mol"1 at 40°C) were taken to be valid at / = 0, but the uncertainties were estimated in this review as ± 2.0 kJ-mol"1. From the values of A r // m as a function of temperature, average ArC° values were calculated.
348
A. Discussion of selected references
[71LAN/CUM] The distribution of 63Ni(II) between anion exchange resin and MSCN (M = Na+, K+, NH^ and Li+) solutions was determined using liquid scintillation techniques. The temperature of the measurements was not given in the paper. The thiocyanate concentration ranged between 0.05 and 5.0 M, while [63Ni(II)],ot was varied from 4.26x10 7 to 2.92x10^ M. No background electrolyte was used, thus the ionic strength varied over a wide range. The authors reported "semi-thermodynamic" formation constants (see [89BJE]) for the four thiocyanato complexes detected (Ni(SCN)^"v with x = 1 - 4), using literature data for the main activity coefficients of KSCN and NaSCN solutions, and estimated values for NH4SCN. Since the temperature of the measurements was not specified, and the "semi-thermodynamic" formation constants are not compatible with values extrapolated to zero ionic strength using SIT, the reported data were not considered further in this review. [71LEB/SAV] This paper describes the measurements of the heat of fusion of Ni and other refractory metals by an electrical explosion method i.e., the rapid heating by a current of high density. The metal wires 0.05 - 0.1 mm in diameter and about 1 cm in length were heated by single square current pulses of density 1.5 — 4.0 x 10 10 A-crri2. Under these conditions, one can neglect the self-induction, the skin effect, and the energy losses, so the wire resistance and energy supplied can be calculated if the voltage and the time of the pulse are known. The maximum possible systematic error in energy determination was 8%. The authors assume that in the case of Ni this error is 1.5%. Unfortunately, A ta / / m( N i ) determined by the electrical explosion method Afus//° = (18.43 + 0.27) kJ-mor1 is significantly different from the values obtained by calorimetric studies of Vollmer et al. [66VOL/KOH] A t a //° = (16.90 + 0.25) kJ-mol 1 , and Geoffray et al. [63GEO/FER] Afus//° = (17.47 + 0.23) kJmol 1 upon which our selected value is based. One possible reason for the observed differences could be the low-purity of nickel, 99.5%, used in the study of Lebedev and Sawatimskii. [71NAV] Navrotsky has measured the enthalpies of formation of Mg2SiO4, Co2SiO4, Ni2SiO4, Zn2SiO4, MgGeO3, Co2GeO4, CoGeC>3, Ni2GeO4 and Zn 2 Ge0 4 from the component oxides by solution calorimetry in a molten oxide solvent at (965 + 2) K. Regularities in the thermodynamics of formation of the mentioned silicates and germanates were discussed. In the case of Ni2GeO4, 9Pb03Cd04B 2 03 was used as a solvent and the enthalpy of formation from oxides was determined to be -39.75 kJ-mol"1. This value is identical with that previously given by Navrotsky and Kleppa [68NAV/KLE]. Using the calorimetrically determined enthalpy of formation from oxides for Ni 2 Si0 4 , an attempt was made to estimate the temperature of decomposition of Ni 2 Si0 4 into NiO and SiO2. The obtained value, 2065 K, is significantly higher then the value accepted by this review (1820 ± 5) K.
349
A. Discussion of selected references
[71OHT/BIE] The hydrolysis of nickel(II) ion was studied by potentiometric titrations in 3 M (Na)Cl medium at 25°C. The concentration of nickel(II) ranged between 0.0145 and 1.0 M, thus the ionic strength varied from 3.0145 to 4 M. The evaluation of emf data, performed with both graphical and numerical methods, indicated the dominant formation of Ni4(OH)4+ , together with some minor species NiOH+ and Ni2OH3+. The best fit to data was obtained by the following formation constants: log10 /? 44 = - (28.55 + 0.02), log10 */?i,i = - (10.5 ± 0.1) and logl0 */?i 2 = - (10.5 ± 0.5). The authors attempted to consider the impact of formation of the NiCl+ complex, but they considerably underestimated this effect. Accepting for the reaction, Ni2+ + Cl ^ NiCl+
(A.42)
logl0 /?, (A.42) = - 0.59 in 3.2 m NaCl medium (see the discussion on [65BUR/LIL2]), ca. 38 - 42 % of nickel(II) is present as the chloro complex. Considering this in the mass action equations, the corrected formation constants are as follows: logl0 /? 44 = - (27.60 ± 0.02), log10 */?i,i = - ( 1 0 - 2 6 ±0.10) and log10 *ft2 = - (10.0 ± 0.5). Taking into account the medium effect caused by replacement of NaCl by NiCl2, the changing ionic strength and the uncertainties of the latter correction
log10 X 4 = " ( 27 - 6 ± °-3)> loSio % = ~ O°- 2 6 ± °-2°)
are accepted in this review.
a n d lo
gi« V..2 = -(10.0 + 0.6)
[71PAA/HUM] This is a spectrophotometric study on nickel(II) chloride complexes formed in aqueous NaOCVLiCl solution. Complex formation was studied first at / = 6 M. The lithium chloride concentration was varied between 0 and 6 M; the ionic strength was maintained by sodium perchlorate. Under these conditions the results indicated the unique formation of the NiCl+ complex and its conditional stability constant is reported to be /3\ = (0.32 ± 0.05). During this study, both the anion and the cation of the original background electrolyte were entirely replaced, which induced additional uncertainty in the value determined for the formation constant. The impact of the C1O4 -> Cl change is difficult to quantify, but the discussion on [70HAL/VAN] may give an approximation. The impact of the Na+ -> Li+ change is easier to assess. As discussed in [97GRE/PLY2] the following expression can be applied for log10 /?, using the SIT at constant ionic strength and total perchlorate concentration: Alog 1 0 # = log10 A,Nacio4 -
lo
gio A(Na,Li)cio4 = [e(Na+, CO - s(Li+, Cl)] mL])
At / = 6 M, and in case of a complete Na+ —» Li+ exchange, A logl0 /?, = — 0.48 can be calculated. It means that the replacement of both the cation and the anion of the background electrolyte each induces (as a rough approximation) an increase of 0.5 in the determined value of log10 /?, compared with the "true" value in NaClO4 (it is very
350
A. Discussion of selected references
difficult to say if these effects are additive or not). Therefore, the reported constant can not be accepted. At LiCl concentrations above 6 M, the formation of a further complex (NiCl2(aq)) was detected. Since the ionic strength between [LiCl] = 6-13.68 M could not be maintained constant, the activity of chloride ion and water was taken into account, while the ratios y , >>. h , >. and y , ,, /y , ^ were assumed /Ni(II,O);*
'.NiCI(H,O)r
'NiCl(H2O);
' NiCl,(H2O)4(aq)
to be nearly constant with increasing LiCl concentrations and were included in the reported "semi-thermodynamic" formation constants (/?" = [NiCl v (H 2 O)^]xa w / [NiCl^_,(H 2 O)7^]xa cl , where x = 1, 2). The reported values are /9I°=(4.9 + 0.8)xl0 2, and /?2° = (7.8 ± 4.1)xlO 5. Although the use of such constants generally provide higher correctness in case of weak complexes (see the discussion on [89BJE]), /7,° was obtained from the rather uncertain value of /3] valid for / = 6 M, thus only K°' = fi° I P\ can be accepted in this review (log l0 K^ = - (2.8 ± 0.5)). [71PIE/HUG] The equilibrium constant for the reaction, Ni(CN)f +CN ^
Ni(CN)^
Ks
was determined by spectrophotometry (400 - 500 nm) at 20°C, / = 2 M (NaClO4(aq) containing 0.01 M NaOH (pH ~ 12)). The formation of the hexacyano complex was not observed even at a 100-fold excess of CN over Ni(CN)5". The reported stability constant (log10 K5 = - 0.77) was corrected to 25°C (log10 K5 = - 0.80), using the selected reaction enthalpy. [71TUR/MAL] The effects of temperature (t = 5 to 35°C) and ionic strength (/ = 0.075 to 0.2 M, KNO3 + KSCN) on the stability and kinetics of formation of nickel(II)-thiocyanate complexes were investigated at pH = 6.0 - 6.2, by detecting the catalytic polarographic currents in the nickel(II) - thiocyanate system. The metal ion concentration varied from 0.2 to 9.5 mM, while the thiocyanate concentration ranged between 4 and 40 mM. Under the conditions used, the formation of Ni(SCN)+ and Ni(SCN)2(aq) was detected. The reported formation constants are accepted, but uncertainties of +0.1 and ±0.2 are assigned to log10 /?, and logl0 /? 2 , respectively. From the temperature dependence of the formation constants (see Figure A-20) A r // m = - (14.3 ± 0.5) kJ-moP1 and ArHm = - (29.5 ± 0.5) kJ-moP1 can be derived for the formation of Ni(SCN)+ and Ni(SCN)2(aq), respectively. These values are selected in this review with both uncertainty values increased to ± 2.0 kJmol '.
A. Discussion of selected references
351
Figure A-20: Temperature dependence of log10 /?, (filled squares) and logl0 fl2 (open squares) values for the Ni(II) - (SCN)~ system (/ = 0.2 M) as reported in [71TUR/MAL].
[72AUF/CAR] The heats of solution in water of the hexahydrate (actually Ni(NO3)2-6.055H2O) and the tetrahydrate were measured as a function of final solution concentration. Extrapolated "infinite dilution" values of 31.23 kJ-moP1 and 8.66 ld-mol"1 were reported. The authors did separate extrapolations for the two sets of measurements (and also for the measurements using the dihydrate [72AUF/CAR2]). The enthalpy of dilution correction to obtain a value for the enthalpy of solution at / = 0 should be identical (within the experimental uncertainties) for the hexahydrate, tetrahydrate and dihydrate [72AUF/CAR2], and the correction should be consistent with whatever correction is applied to the heat of solution results of Chukurov et al. [73CHU/DRA] (for the tetrahydrate). The experimental scatter is fairly large (as much as 0.4 kJ-mor1) for final nickel concentrations < 0.015 m. It would appear that the enthalpy of dilution curve is similar to those for potassium sulphate and barium nitrate [59LAN]. For each of the three salts [72AUF/CAR], [72AUF/CAR2], the values for the five lowest concentrations were averaged to estimate a value at the third-lowest concentration with a solventsalt ratio of 8000:1. The difference between each experimental heat of solution measurement and the measurement (for the same hydrate) resulting in a solution with a solventsalt ratio of 8000:1 (0.00694 m) was calculated. Only values from the measurements with final solution concentrations below 0.056 m were used. Different values from 650 J-mol ' to 200 J m o l 1 were assumed for the heat of dilution to zero molality of the solution with a solventsalt ratio of 8000:1 (0.00694 m). Chukurov et al. [73CHU/DRA] provided an equation for heat of solution of the tetrahydrate in the concentration range 0.007 m to 0.03 m, and difference values based on that equation
352
A. Discussion of selected references
and the same assumed heat of dilution were also calculated. The derived heats of dilution were plotted as a function of mA (Figure A-21), as was a line with the limiting slope for 1:2 and 2:1 electrolytes [59LAN]. If the enthalpy of dilution from a solution with a solvent:salt ratio of 8000:1 to infinite dilution is assumed to be of the order of (520 + 100) J-mol ', all the experimental results enthalpy of solution are in reasonable agreement with each other. However, the values including those of Chukurov et al. [73CHU/DRA] at the lower concentrations) are compatible with the limiting-law slopes found for other 1:2 and 2:1 electrolytes [59LAN]. Figure A-21: Heats of dilution of hydrated nickel nitrate salts based on the experimental heats of solution and an assumed heat of dilution to infinite dilution of 520 J-mol ' for a solution with a solventsalt ratio of 8000:1. Heat of solution values for [73CHU/DRA] were based on the equation of the concentration dependence as reported in that paper (see text and Appendix A entry for [73CHU/DRA]).
Using the estimated heat of dilution and average values of the heats of solution at concentrations at or below 0.01 m, the heats of solution corrected to "infinite dilution" for the hexahydrate, tetrahydrate and dihydrate are (30.76+1.00) kJ-mol1, (8.23 + 1.00) kJ-mol 1 and -(27.64 ± 1.00) kJ-mol '. The uncertainties have been estimated in the present review based on the values shown in the authors' Figure 1 and systematic uncertainties generally associated with similar measurements.
A. Discussion of selected references
353
The equilibrium water vapour pressure at which the hexahydrate begins to lose water to form a lower hydrate was determined from 15°C to 42°C. The vapour pressure was found to follow the equation: l°gio( Ai 2 o / a t m ) = ((7.7226 - 3039.2-r 1 ) ± 0.078) or p,,, 0 /bar= 1.01325 x _
HP™-™'-"")*"-™)
i Q ( ( 7 . 7 2 8 3 - 3 0 3 9 . 2 T " ') ± 0.078)
and at 298.15 K, if the reaction is assumed to be: Ni(NO3)2-6.00H2O(cr) ^ Ni(NO3)2-4.00H2O(cr) + 2H2O(g)
(A.43)
ArG°((A.43), 298.15 K) = (28.14 ±0.89) kJmol 1 and A r //° ((A.43), 298.15 K) = (116.4+ 10.9) kJ-mol '. If the heat of vaporisation is calculated from the enthalpy of solution measurements, and the accepted values for the enthalpies of formation of H2O(1, g), A r //° ((A.43), 298.15 K) = (110.4 ± 1.3) kJmol '. The enthalpy of reaction value from the vapour pressure measurements agrees with the value from calorimetry within the combined uncertainties, though the uncertainty is much lower for the value from calorimetry. Unfortunately, the method used to measure the water vapour pressure does not allow unambiguous identification of the lower hydrate. [72AUF/CAR2] The heat of solution in water of the dihydrate (Ni(NC>3)2-2.00H2O) was measured as a function of final solution concentration. An extrapolated "infinite dilution" value of - (27.41 ± 1.00) kJ-mol ' was reported, and a recalculation has been done in the present review (see the Appendix A for [72AUF/CAR]). The water vapour pressure over mixtures of the tetrahydrate and dihydrate were also measured from 32°C to 73°C. The vapour pressure was found to follow the equation l°gio(p H ,o /atm )= ((7.5527 - 3179.4 T ' ) ± 0.094) or
A,, 0 /bar= 1.01325 x 1o«™«-"™-'"1>±°-°*> _ i/\((7.5584-3l79.4-r ') +0.094)
and at 298.15 K, if the reaction is assumed to be: Ni(NO3)24.00H2O(cr) ^
Ni(NO3)2-2.00H2O(cr) + 2H2O(g)
(A.44)
ArG°((A.44), 298.15 K) = (35.45 ±1.09) kJmol ' and A r //° ((A.44), 298.15 K) = (121.7 + 9.3) kJmol '. If the heat of vaporisation is calculated from the enthalpy of solution measurements, and the accepted values for the enthalpies of formation of H2O(1, g), A r //°((A.44), 298.15 K) = (124.0 ± 1.3) kJ-mol '. The enthalpy of reaction value from the vapour pressure measurements agrees with the value from calorimetry
354
A. Discussion of selected references
within the combined uncertainties, though the uncertainty is much lower for the value from calorimetry. Unfortunately, the method used to measure the water vapour pressure does not allow unambiguous identification of the lower hydrate, and no value for ArG° ((A.44), 298.15 K) can be accepted based on this work. [72BON/HEF] The fluoride complexes of the divalent ions from Mn(II) to Zn(II) have been studied by potentiometric titration, using a fluoride selective electrode at 298.15 K and / = 1.0 M NaOC>4. Under the experimental conditions employed (25 cm3 0.05 M metal(II) perchlorate solution was titrated by 10 cm3 0.004 M NaF solution), only monofluoride complexes are formed but in rather low concentrations (at the end of titrations [NiF+] ~ 2.2x10 6 M can be calculated). This yielded only a moderate effect ( 1 - 2 mV) in the measured electrode potential, therefore we estimate higher uncertainty in K\ ((2.2 ± 1.0) for NiF4) in the present review, than suggested by the authors. The authors reported an anomalous stability sequence in the Mn(II)-Zn(II) series which does not fit an IrvingWilliams type trend. This behaviour was later confirmed by several authors (see e.g., [83SOL/BON]). [72FLE] The high temperature phase a-Ni7S6 was synthesised by heating Ni sponge and S crystals in an evacuated silica glass tube at 504°C for 9 days; the Ni sponge was reduced by hydrogen at 900°C before use. [72FRE/STU] Complexation concentration quotients for 15°C and a medium of 0.1 M KNO3(aq) were reported for the formation of NiHPO4 (K(AA7) = 100), NiP 2 O^ (logl0K(AA5) = 6.22) and NiHP2O~ (log l0 K (A.46)= 3.50). Ni2+ + P2O,~ ^ Ni2+ + HP2O37" ^
NiP2O," NiHP2O,
Ni2+ + HPOl ^ NiHPO4(aq)
(A.45) (A.46) (A.47)
The values for the pyrophosphate species were cited and used in a later paper from this group [78FRE/STU]. The potentiometric measurements were carried out by titration of the free acid with KOH(aq) solution in the presence and absence of Ni(NO3)2 in the solutions. The metal:ligand concentrations were varied from 1:2 to 2:1. Ligand concentrations were 2x10 3 to 5xl0~3 M. The authors estimated 5% to 10% uncertainties in the experimentally determined stability constant values. The electrodes were calibrated against pH buffers rather than concentration standards, and the hydrogen ion activity coefficient, yH, was assigned the value of 0.83; the SIT procedure described in Appendix B would have led to yH, = 0.79. If the concentration quotients are adjusted
A. Discussion of selected references
for this, the value for formation of NiHPO4 is K = 105) and for log10 K (A.45) = 6.24 (the value for log10 K (A.46) is unchanged).
355
NiPjO,",
Correction to zero ionic strength (Appendix B) gives log10 K° (A.47) = (2.88 ± 0.30), where the uncertainty has been assigned in the present review. The small Ae correction has been based on the value for s(Na+, HPO^") from Table B-5 (rather than using the equation in Table B-6), and e(Ni2+, NOj) = 0.18 kg-moP1. For Reaction (A.45) in 0.1 M KNO3, logl0 K (A.45) at 15°C = (7.94 + 0.1 As). For Reaction (A.46) in 0.1 M KNO3, logl0 K (A.46) at 15°C = (4.79 + 0.1 Ae). As the ionic strength is low, the effect of the Ae term on the final calculated values of logl0 K° is limited. Rather than use estimated e values for interactions involving the pyrophosphate moieties (some of which are highly charged), it is simply assumed that AE is zero. As a consequence, the uncertainty in the logl0 K° values is increased to ± 0.25 for the values from measurements in 0.1 M KNO3 solutions. As discussed in Appendix A for [73PER/SEC], the fact that the calculations were done without directly considering the effect of association of K+ with pyrophosphate means that these are "apparent" values (useful for comparisons, but not adequate as selected values). The raw data are unavailable for re-analysis. [72KLE/KOM] Based on thermal and X-ray measurements the complete Ni-Te phase diagram was constructed. [72POL/HER] Polgar et al. carried out calorimetric measurements of the specific heat of NiCl2-2H2O(cr) for temperatures between 1.2 and 25.5 K. Peaks were observed at 6.31 and 7.26 K, and at lower temperatures the heat capacity values are not a simple function of T3. The reasons for these magnetic transitions are not clear. The authors estimated that the contribution to the molar entropy between 0 and 1.2 K is 0.03 R. In the present review, the uncertainty in the extrapolation below 1.5 K is estimated to be of the order of±0.2J-K 'mol '. [72PRE/RUG] Predel and Ruge reported a value of - 8600 calg-atom ' for the heat of formation of NiAs(cr) based on an unreported value for the heat of solution of NiAs in liquid tin over an unspecified temperature range. Based on another paper from the same group [72PRE/RUG2], it would appear that the reaction was carried out between 600 and 900 K. The value is not claimed to be a value corrected to 298.15 K (though Mah and Pankratz [76MAH/PAN], Barin et al. [77BAR/KNA], and Skeaff et al. [85SKE/MAI] appear to have accepted it as the value for that temperature). If the value is assumed to be an average value obtained at 800 K, and the extrapolated heat capacity function of
356
A. Discussion of selected references
Muldagalieva et al. [95MUL/CHU] is applied (with values for // m (800 K) //° (298 K) for Ni(cr) from the present review and for As(cr) from Herrick and Feber [68HER/FEB]), the value: A f //° (NiAs, cr, 298.15 K) = - 71.965 - 30.21 + 15.42 + 13.15 k J m o l ' , i.e., A f //°(NiAs, cr, 298.15 K) = -73.60 kJ-mol 1 can be calculated. The authors claimed that their results were accurate within 5%. In the present review, this is considered to be a 1 a uncertainty, and an overall uncertainty of ± 8.00 kJmol 1 is accepted. [72RET/HUM] The "semi-thermodynamic" equilibrium constant (K°') for the reaction: NiBr(H2O); + Br ^
NiBr2(H2O)4(aq) + H2O(1)
was estimated from spectrophotometric measurements in 8.5 - 11.7 M LiBr solutions. In the evaluation of K% , several simplifying assumptions were made: (i) the activity coefficient of bromide ion can be approximated by the mean activity coefficients of LiBr, (ii) the ratio of the activity coefficients of the two Ni bromide complexes is constant with increasing LiBr concentration, (iii) in 7 - 8 M LiBr solution the only nickel(II)-containing species is the NiBr(H2O)5 complex. [73CHU/DRA] The heat of solution of Ni(NO3)2'4H2O in water was found to follow the equation: KiK
= ( 8 - 4 9 4 + 7 - 4 4 8 m°'5) kJ-mol '
for final solution concentrations of nickel from 0.007 m to 0.03 m. Although the value as extrapolated to / = 0 (8.49 kJ-mol ') is very similar to the one obtained by Auffredic et al. [72AUF/CAR] (8.66 kJ-mol '), the concentration dependence (and hence, it is assumed, the actual values for the measurements at finite concentrations) is less similar—with differences of 0.2 kJ-mol ' to 0.4 kJmol ' in the concentration range of the measurements (see the Appendix A discussion for [72AUF/CAR] for further details). Using the estimated heat of dilution from 0.007 m, and the value of the heat of solution as calculated using the authors' equation for that concentration, the heat of solution for the tetrahydrate, corrected to "infinite dilution", is (8.59± 1.00) kJmol '. The uncertainty is estimated in the present review. The authors also reported the heats of dissolution of "anhydrous" Ni(NO3)2 (-(117.2 ±1.67) kJ-mor1), water (-(5.31 ± 0.08) kJmol ') and Ni(NO3)2-4H2O (- (63.18 + 0.84) kJ-moF1) in dimethyl sulphoxide. This would indicate an enthalpy of (75.23 + 1.90) kJ-moP1 for the dehydration reaction: Ni(NO3)2-4H2O ^
Ni(NO3)2 + 4H2O.
Except for salts containing singly charged cations, synthesis of anhydrous nitrate salts is very difficult, as the final steps of dehydration tend to cause loss of nitrogen
A. Discussion of selected references
357
oxides or nitric acid. Only the summary of the paper was available to the reviewers, not the full deposited document, and, in the summary, there were no details on the preparation of the anhydrous salt. The experimental approach used to determine the enthalpy of formation is reasonable, and may lead to a useful value if the deposited paper contains proper analysis results for the solid and solutions. No value for anhydrous Ni(NO3)2 is selected from these data. [73COR] Cordfunke measured the solubility of Ni(IO3)2-2H2O between 29°C and 61°C. It is not clear from the text whether a pure sample of the dihydrate was used for the measurements or whether the samples were contaminated with the tetrahydrate. Also, the equilibration times are not reported. The results were shown graphically, and it was necessary to retrieve values by digitisation of the points plotted in the author's Figure 2. These values (percent weight as Ni(IO3)2) were converted to molalities, and the apparent solubility products were corrected to / = 0. These were plotted against (77K) ' to obtain values for logl0 K°o at 298.15 K and an average value of A r //° (A.48) over the temperature range of the measurements. Ni(IO3)2-2H2O ^ Ni2+ + 2 IO3 + 3H2O(1)
(A.48)
If only the values to 50°C are included, logl0 K% ((A.48), Ni(IO3)2-2H2O, 298.15 K) = -(5.18 + 0.04) and A r //°(A.48) = (21.7 ±4.7) kJmol ', whereas, if the points for temperatures above 50°C also are included, log10 K°o ((A.48), Ni(IO3)2-2H2O, 298.15 K) = -(5.17 ±0.04) and A r //° (A.48) = (20.5 ± 2.4) kJ-mol '. [73FED/SHM] The solubility of Ni(IO3)2 in nitrate containing LiClO4 solutions ( / = 0.5, 1.0, 2.0, 3.0, 4.0 M) has been measured to determine the stability and composition of the nitrato complexes formed with nickel(II). The authors reported the formation of Ni(NO3)J A complexes (x = 1 - 3). Since, during the measurements, the original background electrolyte (LiClC^) was entirely replaced by LiNO3, the formation of bis- and tris-complexes can not be verified. Therefore, the experimental data published in [73FED/SHM] were re-evaluated taking into account solely the formation of NiNOj species. To minimise the medium effect, only the experimental data [LiNO3]/[LiClO4] < 1 (see Figure A-22) were taken into account. The non-linear regression of these data resulted in the following values: logl0 /?, = - (0.20 + 0.40) (/= 1.0 M), logl0 /?, = - (0.21 ± 0.40) (/= 2.0 M), log10 /?, = - (0.30 ± 0.40) (/= 3.0 M), log10 /?, = - (0.01 ± 0.40) (/= 4.0 M). The uncertainties were assigned by the reviewer, taking into account the still substantial medium effect (also see Section V.6.1.2).
358
A. Discussion of selected references
Figure A-22: Experimental [73FED/SHM] and calculated solubility of nickel(II) in the aqueous phase with increasing nitrate concentrations.
The values for the solubility of Ni(IO3)23H2O in aqueous 0.5 M to 4.0M LiClO4 solutions (i.e., those solutions with no added nitrate) were used to calculate the value of the solubility product of Ni(IO3)2'3H2O (Figure A-23) from a standard SIT plot. This gives the value logl0 K°o = - (5.09 + 0.16).
Figure A-23: Experimental [73FED/SHM] and calculated values of the solubility of Ni(IO3)2'3H2O in aqueous LiClO4 solutions.
A. Discussion of selected references
359
[73FLE] The data concerning NiO are essentially the same as in [68CHA/FLE]. [73GUR/CAL] The heat of solution of nickel chloride hexahydrate into 1 .OM HCl(aq) was measured, and (6.996 ± 0.133) kJmol ' was reported (2a uncertainties). The final solution was 0.02 to 0.03 M in Ni. A double correction to 0.0 M HCl(aq) and to 7=0 [95MAN/KOR] would be required to use the reported value to calculate the enthalpy of solution of the salt at infinite dilution in water. The corrections themselves have large uncertainties, and this heat of solution value is not used further in the present review. [73HUT/HIG] A kinetic method, based on the metal ion catalysed reaction of: [Co(EDTA)Cl]2 ^
[Co(EDTA)] + Cl ,
has been used to determine the stability constants of several [MA] complexes, among others those of the chloro-, bromo-, nitrato- and thiocyanato complexes of nickel(II) (pH = 4.5 - 5.5, r = 298 K, / = 1 M Na(C104, X)). The concentration of nickel(II) was varied up to 0.051 M. The highest concentrations of the complex-forming anions (X ) for nickel(II) in a given series were [Br ] = 0.49 M, [Cl ] = 0.278, 0.466 and 0.844 M, [NO~] = 0.422 M and [SCN""] = 0.15, 0.318 M. The formation constant determined at [Cl ] = 0.844 M was not considered in this review, in order to minimise the need to assess the medium effect. The authors acknowledged several disadvantages of the chosen reaction. The most important is that the monocationic complex [NiX]+, not just Ni2+, may catalyse the reaction. This effect was taken into account by a simple correction factor. Therefore, we assigned higher uncertainties to the reported values. The formation constants of chloro-, nitrato- and thiocyanato- complexes also were determined at 318.15 K. The stability of the nitrato complex does not change at higher temperatures, but those of the chloro and thiocyanato complexes decrease moderately; this results in negative A r // m values. This observation is the opposite to what was observed in [60LIS/ROS] and [66KEN/LIS] for the NiCl+ complexes, but is in reasonable agreement with other reports [67NAN/TOR], [74KUL3] in the case of NiSCN+. Based on the entropies of association, the formation of mostly outer-sphere complexes (ion pairs) was suggested in the case of nickel(II), except for thiocyanate. From the temperature dependence of the formation constant for the NiSCN+ species ArHm = — (8.3 + 1.0) kJ-mol"1 can be derived. This value is accepted with an increased uncertainty (A r // m = - (8.3 ± 2.0) kJ-mor1). [73KAT] This is a report on a careful conductometric study. Sets of data were obtained for 5°C intervals for temperatures between 0 and 45°C. The sets of equivalent conductivities are reported for specific concentrations from 0.0002 to 0.003 M, and have the appearance
360
A. Discussion of selected references
of being smoothed values from unreported primary data. In the data analysis, the ionsize parameter in the Debye-Hiickel equation was optimised for each temperature. It is not clear that this is consistent with the use of the Fuoss and Accascina conductance equation [59FUO/ACC]. The result is a slightly greater temperature dependence of the association constant than if a constant value were used for the ion-size parameter. Nevertheless, the temperature dependence is calculated to be less than found in the potentiometric study of Nair and Nancollas [59NAI/NAN]. For the purposes of the present review, the data have been re-analysed using a version of the Lee-Wheaton equation modified so that the activity-coefficient equation matches the SIT equation usually used in the TDB project (Appendix B). At comparable temperatures, the calculated nickel sulphate association constants are slightly larger than those recalculated from the results of other conductivity studies [1895FRA], [76SHI/TSU], [79FIS/FOX]. In the present review, we estimate the uncertainty in K^ at 25°C to be + 20. The same uncertainty is estimated for 20 and 30°C, but for temperatures above 30°C and below 20°C, an uncertainty of + 30 is estimated. The average enthalpy of reaction for 0 to 45CC was recalculated, and found to be 5.3 kJ-mor1. This value is accepted for 25°C, and an uncertainty of ± 2.0 kJmol ' is assigned, in part by considering separately the average enthalpies of reaction calculated from the association constants for the four highest temperatures (7.5 kJmol ') and for the four lowest temperatures (3.3 kJ-mol '). [73KAW/OTS] The hydrolytic reactions of nickel(II) ion were studied at 25°C in water and dioxanewater mixtures containing 3 M (Li)GO 4 as an ionic medium. The total nickel(II) ion concentration, B, varied between 0.025 - 0.8 M (/ = 3.025 - 3.8 M) in water and the average number of hydrogen ions set free per nickel atom, Z, was measured as a function of pH. In the aqueous solutions a range of 5.7 < pH < 7.3 was studied. Approximate values of the solubility constants of Ni(OH)2(s) (log10 *Ks 0 = (13.5 + 0.5)) and Ni(OH)ClO4(s) (log ]0 *ATV 0 = (6.9 ± 0.5)) were estimated from maximum values of Z. The former constant, converted into the molality scale with density data taken from Lobo and Quaresma [89LOB/QUA] (Part B), and extrapolated to zero ionic strength, has been given in Table V-6 and Figure V-l 1. At lower Z values, the function Z = Z(pH) can be interpreted by the formation of Ni4(OH)4+ as the predominant hydrolysis product. None of the minor species NiOH+ and Ni2OH3+ could be established from the data in this study. In aqueous solution log10 /?4 4 = - (27.32 + 0.08) has been obtained, which is in good agreement with the corresponding value reported in [65BUR/LIL]. This value has been accepted for use in the present review, but with the uncertainty increased to ± 0.16.
A. Discussion of selected references
361
[73NAV] The enthalpy of the olivine - spinel transition of Ni 2 Si0 4 was obtained by measuring the heat of solution of both polymorphs in a molten oxide solvent, 2PbOB 2 O 3 , at 713°C. The following values were found: enthalpy of the transition Ni2SiO4(oliv) —> Ni2SiO4(spin), A trs // m (986 K) = (5.9 + 2.9) kJ-mol ', the heat content increments, // m (986 K) - //°(298 K), olivine, (107.7 ±1.8) kJ-mol"1, spinel, (106.2 ± 0.8) kJ-mol"' and the entropy of olivine - spinel transition A^S^ 12.6 to 14.6 J-IC'mol '. [73NAV2] Navrotsky reported an experimental approach that was used to obtain values of Gibbs energy differences between olivine, spinel, and phenacite forms of silicates and germanates from the thermodynamics of terminal solid solutions in ternary systems. This is applied to the systems N i O - M g O - G e O 2 and C o O - M g O - G e O 2 at 1200°C in air and to the system NiO - MgO - GeO2 at 800°C and 0.57 kbar water pressure. The Gibbs energy of transformation from the olivine to spinel structure for Ni2GeO4 at 1200°C was estimated to be - 34.31 kJ-mol"1. [73NOV/COS] The authors point out that Ni(OH)2 phases of different solubilities were obtained depending on the speed of precipitation. The solubility products given at 25°C for / = 0.55 M NaCl and / = 1.0M NaClO4 seem to have been calculated from Equation (A.49): logl0 K,fi = logl0([Ni2+]-[H+] 2) + 2 log10 Kw .
(A.49)
The authors used Ni2+ and presumably H+ concentrations in the first term on the right hand side of Equation (A.49) but the ionic activity product of H2O in the second term. Thus, adequately corrected numerical values are given in Table V-6 and Figure V-ll. [73PER/SEC] The complexation of Ni2+ with partially protonated pyrophosphate ion was studied as a function of ionic strength (KNO3, 0.02 M to 0.20 M), temperature (5 to 35°C) for 0.1 M KNO3 in solutions with pH values from 4.5 to 7.0. Values are reported for the formation constants of NiP 2 O^ and NiHP2O~ . Ni2+ + P2O,~ ^ 2+
Ni + HP2O7 ^
NiP2O7~ NiHP2O7
(A.50) (A.51)
Values for the first and second protonation constants for P2O*~ also were determined for the same ionic media and temperatures. The ionic strengths are sufficiently low that it is unlikely that meaningful As values can be established for any of these reactions based on the data in this paper alone. Furthermore, the pH electrode appears to have been calibrated against standard buffer solutions, with no allowance being made
362
A. Discussion of selected references
for different junction potentials in the solutions of different ionic strength. The values of the reported 25°C constants after correction for the extended Debye-Hiickel terms (nD, cf. Appendix B), but without consideration of complex formation between K+ and pyrophosphate ions, are plotted in Figure A-24. The corrected complexation constants of pyrophosphate with nickel show a marked ionic strength dependence that would correspond to Ae values of (2.5 ± 0.7) kg-mor' and (1.7 ± 0.7) kg-mol"1. The former, at least, is inconsistent with any reasonable selection of e values (though the same is not true for the Ae values for the first protonation constant for pyrophosphate, which also involves highly charged ions). Based on these calculations, only the results from the higher ionic strength solutions (7=0.1 M and 0.2 M) are used in the present review. For Reaction (A.50) in 0.1 M KNO3: log10 K (A.50) = 5.94 and logl0 AT°(A.50) = (7.69 + 0.1 Ae) at 25°C, logI0 K (A.50) = 5.81 and logl0 K° (A.50) = (7.53 + 0.1 AE) at 15°C, in 0.2 M KNO3, log10 A (A.50) = 5.60 and log10 K° (A.50) = (7.78 + 0.2Ae) at 25°C. For Reaction (A.51) in 0.1 M KNO3: log10A:(A.51) = 3.71 and log10 AT" (A.51) = (5.02 + 0.1 AE) at 25°C, log10 K (A.51) = 3.55 and log10 A"0 (A.51) = (4.84 + 0.1 AE) at 15°C, in 0.2 M KNO3, log10 K (A.51) = 3.39 and logl0 K° (A.51) = (5.02 + 0.2Ae) at 25°C. As the ionic strength is low, the effect of the Ae term on the final calculated values of logl0 K° is limited. Rather than use estimated e values for interactions involving the pyrophosphate moieties (some of which are highly charged), it is simply assumed that Ae is zero. As a consequence, the uncertainty in the log10 K° values is increased to ± 0.25 for the values from measurements in 0.1 M KNO3 solutions, and to ± 0.3 for the values from measurements in 0.2 M KNO3 solutions. There is a further complication. Within the context of a standard (or extended) Debye-Hiickel treatment of activity coefficients, the interaction of the highly-charged pyrophosphate ion with alkali metal ions can be considered most easily by assuming weak complex formation [49MON], [57LAM/WAT]. The same is true if the Specific Ion Interaction approach described in Appendix B is used. De Stefano et al. [94STE/FOT] measured the protonation constants for pyrophosphate, as a function of temperature between 5 and 45°C, in aqueous solutions of (CH^NCl, NaCl and KC1 (0.00 to 0.75 M). The values for a 0.1 M KCl(aq) medium agreed well (within 0.05 in log]0 K) with the values reported for a 0.1 M KNO3(aq) medium in the studies of Perlmutter-Hayman and Secco [73PER/SEC]. From their apparent protonation constants, De Stefano et al. [94STE/FOT] calculated values of the formation constants of KP2O37~ and KHP 2 O^ (and of the formation constants for the corresponding sodiumion species).
A. Discussion of selected references
363
Selection of values for the association constants for pyrophosphate ion with alkali metal ions is beyond the scope of the present review. If the formation constants values for KPjO," and KHP 2 O^ from [94STE/FOT] are used with the values of the pyrophosphate protonation constants from the same source, and if the nickel pyrophosphate complexation constants from Hammes and Morrell [64HAM/MOR] are used, the calculated concentrations of Ni2+ and NiP 2 O^ are within 15% of those calculated by Perlmutter-Hayman and Secco (by neglecting the association of K+). This provides strong support for the value of K(A.5O) of Hammes and Morrell [64HAM/MOR]. The species NiHP2C>7 is calculated to be less important (generally < 20% of the total nickel in solution), and its calculated concentration is approximately 50% higher when association is neglected. This would suggest that the formation constant for NiHP2O, reported by Perlmutter-Hayman and Secco is greater than would be the case if association of pyrophosphate with K+ were considered explicitly. The raw data are unavailable for re-analysis. From the temperature dependencies of the equilibrium constants, the enthalpies of reaction in 0.1 M KNO3(aq) can be calculated as, Ar//m(A.5O) = (47.93 ±10.16) kJ-mol ',and A r // m (A.51) = (30.56 ± 7.42) kJmol '.
Figure A-24: Variation of formation constants from Perlmutter-Hayman and Secco [73PER/SEC] with ionic strength. For the first and second protonation constants of pyrophosphate, n = - 8 and - 6; for Reactions (A.50) and (A.51), n = - 16 and - 12.
364
A. Discussion of selected references
[73POW] This paper presents a reassessment of the data of Izatt et al. [69IZA/EAT]. The author used the K{ value for the nickel sulphate association constant that was determined by Nair and Nancollas [59NAI/NAN], and reinterpreted the results from the later titration calorimetry experiments. The variation of the enthalpy of ion-pair formation with ionic strength was also reported. The recalculated value agrees well with those from some conductance studies, but is still low with respect to the value from [59NAI/NAN]. [73RAU/GUE] A mixture of metallic nickel and nickel oxide (particle size around 1 um) was treated with mixtures of water vapour and hydrogen gas at temperatures ranging from 711 to 860 K. The ratio of equilibrium partial pressures pHOlpH was determined by a tensimetric method using a palladium-membrane apparatus. The authors applied a third law analysis to their data and arrived at A f //° = - (239.5 ± 1.0) kJmol 1 for the standard enthalpy of formation of NiO at 298.15 K. [73VEN/GEI] The activity of sulphur in dilute solutions of sulphur in pure nickel melts was investigated at three different temperatures, viz. 1773, 1823 and 1848 K, for concentrations up to 0.7 wt% S. A gas mixture of H2(g) and H2S(g) was equilibrated with liquid nickel at a constant gas stream through the furnace. The compositions of both the gas mixture and the equilibrated liquid phase were determined by chemical analysis. When S2(g) is the reference state for sulphur, the activities of sulphur in the nickel melt can be calculated from the well-known gas phase equilibrium: H 2 S ( g ) ^ H2(g)+'/2S2(g). Since this contribution is confined to highly diluted nickel melts only, the experimental results are not re-evaluated in the present assessment. [73WAA/CAL] Liebenbergite, a natural nickel mineral is described mineralogically in this work. [74ARU] The heat of association, in aqueous solution, of fluoride ion with metal ions of the Mn(II)-Zn(II) series has been measured by a direct calorimetric method at / = 0.5 M and 298.15 K. The measurements were performed in a self medium (to 83 cm3 of a 0.1665 M Ni(NO3)2 solution were added three amounts of 5 cm3 0.5 M NaF solution). The molar enthalpies of the reaction were calculated using the stability constants reported in [72BON/HEF]. For the purpose of the present review, the reported experimental data for the reaction, Ni2+ + F ^ N i F +
(A.52)
A. Discussion of selected references
365
have been recalculated using the recommended values for logl0 j5° ((A.52), 298.15 K) and As (A.52). At / = 0.5 m log10 /3X ((A.52), 298.15 K) = (0.76 ± 0.08) can be calculated. Taking into account the different ionic media used by Aruga, a greater uncertainty in log10 fix ( ± 0.25) was assigned. The recalculated formation enthalpy of the NiF+ species is (4.4 ± 2.0) kJ-mol 1 . [74BIX/LAR] Chloride selective electrode measurements were used to determine the stability constants of nickel(II) chloride complexes at 25°C in 1.00 M NaClO4 and 0.10 M HC1O4. The total chloride and metal ion concentrations ranged from 0.0003 - 0.01 M and 0.02 0.04 M, respectively. The authors noted that the response of the chloride selective electrode was not linear below a chloride concentration of 0.2 mM. The effect of complex formation on the measured potential was rather modest (0.4 - 1 . 1 mV), therefore we assigned a higher uncertainty to the formation constants than was reported by the authors. [74BLO/RAZ] The enthalpy of formation of M(H2O)6X(2 y)+ complexes (where M = Ni(II), Co(II), or Cu(II) and X y = C l , Br or SO\ ) was determined, using the stability constants reported in [70MIR/MAK] and experimental results on heat of mixing of 0.05 M M(C1O4)2 solution (/ = 3.0 M maintained with LiClO4) with the solution of the corresponding lithium salt (/ = 3 M). The authors did not reporte sufficient experimental details to allow the reliability of their data to be assessed. [74DIC/HOF] The kinetics of formation and dissociation of Ni(SCN)2(aq) have been studied in water and in several organic solvents, using the pressure-jump and shock wave relaxation technique at 20°C. The concentration of Ni(SCN)2(aq) ranged between 0.001 and 0.1 M. In water, only the formation of the monothiocyanato complex was observed. No background electrolyte was used, and the activity coefficients were calculated by an extended Debye-Hiickel expression. Although, this activity model is not compatible with the SIT, the ionic strengths were low. Therefore, the reported result was corrected to 25°C and the resulting value was accepted with an increased uncertainty (log l0 J3° = (1.79 + 0.10)). [74GRA/WIL] The ternary and quaternary complexes of nickel(II) and palladium(II) with asparaginate (asn), chloride and hydroxide ions were studied using glass and silver-silver chloride ion selective electrodes (/ = 3 M NaClO4, T = 298.15 K). The formation of Ni(asn)^(OH)J,2"J~y) complexes was not reported, only the ternary (quaternary) complexes with chloride ions, Ni(asn)ACl(OH)^A'~y) , in addition to the binary NiCl+ complex, were found to exist. Taking into account the low affinity of chloride for nickel(II)
366
A. Discussion of selected references
this is unlikely. Therefore, the reported logl0 A:(NiCl+)= (0.687 + 0.040), which is much higher than other reported values, was not taken into account in this review. [74GRI/FER] The structure of a millerite crystal from Marbridge Mine, Malartic, Quebec, with the empirical formula Nio.98iFeo.oi6Coo.004S has been refined. The hexagonal axes were found to be a0 = 9.607(1) A and c0 = 3.143(1) A. Within the lattice each Ni is coordinated by five S atoms and two Ni atoms. The observed Ni-S bond lengths are comparable to the expected value for a covalent bond. Molecular orbital theory was invoked to show that the millerite structure with five-fold coordination around Ni is more stable than the nickeline structure (oc-NiS) with six-fold coordination about each Ni. Thereby it is rationalised that the low temperature phase (3-NiS occurs in nature and the high temperature phase a-NiS does not. [74HEF] This is a review paper on fluoride complexes of some 50 metal ions. The author found that the stability constants of these complexes correlate very well with simple electrostatic parameters such as ion size and charge. Only a few metal ions, all found in the lower right hand side of the Periodic Table, seem to be exceptions from this general rule. The enthalpy of formation is positive even for the strongest complexes. Only Be2+ and Ag+ appear to have negative A r // m values. Consequently, the formation of the great majority of fluoro complexes is entropy controlled. [74KAB] The protonation of tetracyanonickelate(II) ion was investigated by a pH-metric method at temperatures from 25 - 50°C and at ionic strengths of 0.01, 0.02, 0.05, 0.1, 0.2 and 0.5 M, adjusted using NaCl. The author reported two successive protonations of the Ni(CN)J~ ion(log l0 A^ and \ogwK2 , respectively). Titrations of aqueous Na2Ni(CN)4 solutions by 0.1 M HC1 were performed while nitrogen gas was bubbled through the solution; therefore HCN volatilisation could have been occurring. Several authors have reported slow equilibration in acidic solutions of Na2Ni(CN)4, sometimes lasting several days [63CHR/IZA], [68KOL/MAR], [74PER]. During the titration performed in [74KAB], equilibrium was probably not attained. The dependence of the determined protonation constants on the temperature and ionic strength shows notable deviation from what would be expected (see e.g., Figure A-25 and Figure A-26). Therefore, the reported data were not considered further in this review.
A. Discussion of selected references
Figure A-25: Temperature dependence of logl0 K2 for the reaction Ni(CN)^ + H+ Ni(HCN)(CN)j at different ionic strengths using NaCl.
367 ^
Figure A-26: Extrapolation to / = 0 of experimental equilibrium constants for the reaction Ni(HCN)(CN)3" + H + ^ Ni(HCN)2(CN)2(aq) (log10 K2) at different temperatures.
368
A. Discussion of selected references
[74KUL3] The stability constants and the enthalpy changes for the formation of nickel(II) (and copper(II)) thiocyanate complexes were determined by calorimetric titrations at 25°C in 1 M (NaC104) aqueous solution. The concentrations of nickel(II) and thiocyanate ranged from ~ 0.01 - 0.1 M and 0 - 0.8 M, respectively. The formation of NiSCN4, Ni(SCN)2(aq) and Ni(SCN)~ was assumed according to [53FRO2]. Both graphical and computer evaluations of the experimental data were performed. The formation constants obtained by the computer programme "Kalori" are as follows: log10 /?, = (1.12 ± 0.04) (3a), logl0 /32 = (1.57 ± 0.14), log10 /?3 = (1.28 ± 1.00) (two years later the author reported [76KAR/KUL] somewhat different constants: log I0 /?,= (1.12 ± 0.02) (3a), log,oA= (1.57 + 0.04), log,0/?3 = (1.25 ±0.20)). The reported "best" values are slightly different from both above mentioned sets, since Fronaeus' data [53FRO2] were also taken into account: log10/ff,= (1.14 ±0.02) (3a), log, 0 /? 2 = (1.58 + 0.05), log10 J33 = (1.6 ± 0.2). The latter constants have been used to calculate the corresponding stepwise enthalpy values: A,Hm= - (12.02 ± 0.15), - (8.9 ± 1.0) and - (8.2 ± 4.0) kJ-mol"1 for the three successively formed complexes, respectively. The enthalpy of reaction values corresponding to the three cumulative formation constants are A r // m = - (12.0 ± 0.5), - (20.9 ± 1.2) and - (29.1 + 5.2) kJmol ', respectively, where the uncertainties have been estimated. [74MA] In this work the olivine - spinel transformation in Ni2Si04 at high pressures was investigated over a. p - t range of 20 - 40 kbar and 650 - 1200°C, using a piston-cylinder apparatus. The transformation curve is essentially a straight line and can be expressed by an equation: p(bar) = 23300 + 11.8-f (°C). The melting behaviour of Ni2SiO4 olivine was investigated map - t range of 5 - 13 kbar and 1600 - 1700°C. The results show that Ni2Si04, olivine melts incongruently at high pressures to NiO plus melt. The reported dissociation of Ni2SiO4, olivine, to NiO plus cristobalite prior to melting [63PHI/HUT] was not observed. The melting curve has ap — t slope of approximately 105 bar-K '. It can be extrapolated to atmospheric pressure to give a melting temperature of approximately 1575°C. This value is 75°C lower than that given by [63PHI/HUT]. [74MAK/MAS] Values for the association of nickel with sulphate for temperatures from 25 to 95°C and ionic strengths to 2.6 M were measured using a variety of pH titration, conductivity and transport number measurement techniques. Although the reported association constants are in rough agreement with those from other studies, there are insufficient experimental details (or results) to use this work in deriving a recommended value for K\.
A. Discussion of selected references
369
[74NET/BAT] Complex formation of Ni(II), Pr(III) and Ho(III) with thiocyanate was studied in aqueous solutions (/ = 3 M (LiClO4)) by IR spectroscopy. The concentration of thiocyanate was 0.36 M, the concentration of metal ions ranged from 0.36 to 0.86 M. The formation constants of the mono-complexes (MSCN+) were calculated by three different methods, which resulted in considerably different values. The temperature of the measurements is not given in the paper. Therefore, the reported data were not considered further in this review. [74PER] The formation of complexes between nickel(II) and cyanide ion was studied by spectrophotometric and potentiometric measurements with a glass electrode at 25°C, at an ionic strength of 3 M using aqueous sodium perchlorate as the ionic medium. The potentiometric data are best described by the formation of two complexes NiCl^ and Ni(CN)^ (log10 # = (7.03 ± 0.2) and log10 /?4 = (31.06 + 0.03)). On the other hand, the spectrophotometric measurements gave no positive evidence for the formation of NiCN+; therefore the logl0 /?, value given above should be regarded as a maximum value. If only formation of Ni(CN)^" was postulated, log 10 /? 4 = (31.12 + 0.08) was obtained, with a somewhat higher uncertainty in the fitted parameter. This value is accepted in the present review, but the uncertainty has been estimated here as ± 0.15. The potentiometric investigation was performed at 4 < pH < 5.5, but protonated tetracyano complexes, as reported in [68KOL/MAR], were not detected. [74RAJ/PRE] The structure of a single crystal of millerite from Quebec, Canada, with hexagonal axes «o = 9.6190(5) A, Co = 3.1499(3) A and composition Ni103Feo.oi6Coo.004S has been refined. The Ni and S atoms are in five-fold coordination (tetragonal pyramidal) with each other. The average Ni-S distance, 2.310 A, is consistent with a divalent Ni in five-fold coordination. Rajmani and Prewitt suggested that the metal-metal bonding and the formation of the trinuclear cluster stabilise the millerite structure. [74RUT/HAU] This paper describes vapour pressure measurements over Ni in the range 1277 - 1658 K. as obtained using the Langmuir technique. The results yield the enthalpy of sublimation (447.7 ± 24.3) kJ-mol"1 and (431.8 + 16.7) kJ-mol""1 according to 2nd and 3 rd law analysis, respectively. In addition Rutner and Haury [74RUT/HAU] determined the "best values" of the parameters in the vapour - pressure equations of solid and liquid nickel, and the heat of vaporisation and sublimation of nickel by a statistical treatment of data available in the literature and their own results. The results of Rutner and Haury [74RUT/HAU] were not used by [98CHA] for the determination of the enthalpy of formation for Ni(g) because of an inconsistency with the melting data.
370
A. Discussion of selected references
[74SLO/JON] In their Figure 2 Slough and Jones plotted z-r~x versus A r //° of reactions, MO(cr) + 2B2O3(cr) ^
MB4O7(s)
(A.53)
and M2O(cr) + 2B2O3(cr) ^
M2B4O7(s)
(A.54)
where r are presumably the Pauling radii [40PAU] and A r //° were determined experimentally by reactions of the crystalline oxides, see Figure A-27. The value for NiB4O7, given by [74SLO/JON], can only be estimated when the Goldschmidt radius of Ni2+ (r = 0.078 nm) [26GOL] and not the Pauling radius (r = 0.070 nm) is employed. The effective ionic radii of Shannon (for coordination number 6 [76SHA]) lead to an inferior correlation and a completely different A r //° for the respective nickel reaction, even on the plot based on the Pauling radii. This example demonstrates that methods of thermodynamic data estimation also must be reviewed critically.
Figure A-27: A r //° (298 K) of reaction of B2O3(cr) with MO or M2O.
[74WOR/COW] Heat capacities from 10 K to 300 K were determined by adiabatic calorimetry, and by differential scanning calorimetry (DSC) from 300 K to 550 K. In the paper, only rounded values are supplied for the adiabatic calorimetry results for NiI2(cr) except for values near the (structural phase [81KUI/SAN]) transition at approximately 59 K. The authors also supply an equation (A.55)
A. Discussion of selected references [C;J5250°0K
(Nil2, cr) / J-K-'-mor1 = 65.5 + 0.0515 77K - 0.0000397 (77K)2
371
(A.55)
that provides a fit to the combined heat capacity results from adiabatic calorimetry and differential scanning calorimetry between 200 K and 550 K, but do not separately provide measured values above 300 K. This equation is not in a form normally used for the temperature dependence of heat capacities of solids near room temperature [93KUB/ALC]. The adiabatic calorimetry heat capacity data were available as supplementary material from the British Library Lending Division (SUP 21075), and were used in the preparation of the current review. The authors integrated the low-temperature heat capacity measurements and reported 138.7 J K r ' m o l 1 for 5°(NiI 2 , cr). No contribution was added for the magnetic phase transition at 75 K to 76 K reported by Billerey et al. [77BIL/TER] and Kuindersma et al. [81KUI/SAN], and this may have led to an underestimation of between 0.3 JK. '-mol ' and 0.5 J K ' m o l 1 in the value of 5° above 80 K. Indeed, the two most relevant values from the authors "Run I" (for 77.69 K and 79.36 K) show no evidence whatsoever of an anomaly. However, the authors' "Run I" values also do not provide a well-defined peak for the anomaly near 59 K. The "Run II" experiments covered only the temperature range of the 59 K anomaly, and appear to have been done with much more care and with mush smaller temperature increments. The values of entropies of reaction as derived from adiabatic calorimetry and from measured dissociation pressures and heats of solution of Nil2, Ni(NH3)2l2 and Ni(NH3)6I2 were found to be in good agreement. This indicates that it is likely that assuming a smooth extrapolation of Cpm between 10 K and 0 K does not introduce any substantial error in the calculated S1" value. The adiabatic calorimetry results between 200 K and 300 K showed a random scatter of approximately + 0.4 J K ' m o l 1 , unusually large for this type of measurement. The experimental scatter for the differential scanning calorimetry values was reported to be within 4%, but it is not reported how well the authors' equation corresponded to the measured values, or how well measurements obtained from the two techniques agree near 300 K. Unfortunately, the DSC data were not part of the supplementary material. To obtain an equation for Cpm(T) in a more standard form, values of the heat capacity were calculated at 25 K intervals between 300 K and 550 K using the authors' equation (A.55). An equation of the form (A + B(77K) + C(7VK)"2) was fitted to these values and to the experimental low-temperature adiabatic calorimetry values from 200 K to 300 K. Values from the adiabatic calorimetry and from the equation based on the DSC measurements were weighted in a ratio of 0.4:2.0. Because the number of DSC measurements used in the original fitting by Worswick et al. is unknown, these weightings are arbitrary by definition. The obviously erroneous datum at 229.92 K was not used. The resulting equation (A.56), [C° ]™l ( NiI 2, cr) / J K 'mol ' = 73.5775 + 0.0164257- 1.057753*105 (77K)2 (A.56)
372
A. Discussion of selected references
fit the low temperature data as well as the authors' equation (and well within the experimental scatter). As would be expected, agreement between values from the two equations was poorer for 300 K to 550 K, but the differences were always less than 0.5 JKr'-mor 1 —again much less than the experimental uncertainties. The values from equations (A.55) and (A.56) for 298.15 K agree well with the reported smoothed value from the adiabatic calorimetry results (78.4 J K 'mol ') and with each other (within 0.05 J-K^'-mol"1). Because the actual DSC results are unavailable, and there is considerable scatter in the adiabatic calorimetry results above 200 K, an uncertainty of ± 1.0 J K '-mol ' is assigned to the heat capacity value for nickel iodide at 298.15 K. [74YAG/MAR] Yagi et al. have investigated structures of the spinel polymorphs of Fe2SiC>4 and Ni2SiO4 in detail using single crystals synthesised at high pressures and temperatures. Ni 2 Si0 4 spinel with a = 8.044(1) A had a normal spinel structure. [75ARN/MAL] The sulphur-rich part of the system Ni - S above 1253 K was re-investigated by metallographic methods. Vaesite (NiS2) was prepared from NiS and sulphur by solid state reaction. Mixtures of NiS2 and S were heated in evacuated silica glass tubes. After quenching, polished samples were examined microscopically. In contrast to Kullerud and Yund [62KUL/YUN], who reported a congruent melting point of vaesite at 1280 K and a monotectic reaction at 1264 K, the heating experiments indicated that NiS2 melts syntectically, forming two immiscible liquids at (1295 ± 3) K, i.e., NiS2 ^ L, + L2. The liquid L2 was estimated to contain 0.5 wt% Ni, while the composition of the liquid L| amounted to (52.0 ± 1.8) wt% Ni. [75ARU] Enthalpies of association of nitrate and chlorate with Mn(II), Co(II), Ni(II), Cu(II) and Zn(II) have been determined by direct calorimetry in an aqueous medium at 298 K and an ionic strength / = 1 M. In a typical experiment, 2.5 and 5.0 cm3 of 1 M NaNO3 solution were added to 93 cm3 of 0.34 M Ni(ClO4)2 solution, and this was repeated twice under identical conditions. The association between Na+ and nitrate ion was considered during the calculations (Kass = 0.25 M '). The molar enthalpies of the reaction: Ni2+ + NO; ^
NiNOj
(A.57)
were calculated using the stability constants reported in [73HUT/HIG]. For the purpose of the present review, the reported experimental data have been recalculated using the tentative values for log10 fi° ((A.57), 298.15 K) and As((A.57), (NiClO4)2) = -(0.11 ±0.15) kgmol '. At / = 1.05 m Iog10/J, ((A.57), 298.15 K) = -(0.21 ± 1.00) can be calculated. The recalculated formation enthalpy of the NiNO, species is (6 ± 4)
A. Discussion of selected references
373
kJ-mol"1. The relatively high uncertainty is assigned due to the large error in log 10 ^ o ((A.57), 298.15 K). [75BAR/MAS] Barvinok et al. measured the enthalpy of solution of NiCl2(cr) in 2.0 M HC1 at 298.15 K. A value for the molar enthalpy of solution, Asol//m = - (74.43 ±0.17) kJ-mol~', was reported. As other experimental details are lacking, this single value was disregarded in this review. [75CLA/KEP] The kinetics of hydrolysis of nickel(II) were studied by mixing nickel(II) sulphate or nickel(II) perchlorate solutions with NaOH solutions in a stopped-flow apparatus at 25 and 41.7°C. The ionic strength was maintained at 0.8 M by additions of sodium sulphate or perchlorate. The hydrolysis occurred in two stages, as already mentioned in [71HOH/GEI]: (i) the first stage (probably the formation of NiOH+) was complete within the mixing time (2 ms), (ii) the second stage resulted in the formation of Ni4(OH)4+ within a few seconds (after a longer time interval, slight precipitation of nickel hydroxide was also noted). In perchlorate medium some runs were made to follow the changes of pH, after addition of base, using Neutral Red as an indicator (the value used for the pK of the indicator was not reported) at [Ni2+]lol = 0.203 M. The pH was measured just after mixing and at equilibrium. From the pH values measured at "t = 0 s", log10 y?|, = - (8.93 + 0.10) was obtained for the reaction : Ni2+ + H2O ^
NiOH+ + H+.
(A.58)
Since this value is considerably lower than other literature values, the authors concluded that some oligomerisation occurred even during mixing. From the pH values at equilibrium, the authors calculated the formation constants of Ni4 (OH)4+ assuming it as the only hydrolysis species that was formed. Taking into account the relatively low pH values at equilibrium (pH = 6.50 — 6.75), the formation of NiOH+ should also be considered. Since only 6 experimental data points were reported, the value of logl0 /?,, was fixed at - 9.87 (calculated from the selected logl0 */?, i ((A.58), 298.15 K) for /,„ = 0.82), and only the logl0 /? 44 value was refined during our re-assessment. With this constraint, logl0 fiA 4 = - (26.98 ± 0.20) was obtained, a value close to the reported value. Taking into account the limited number of experimental data, log10 /? 44 = - (27.0 + 0.4) is accepted for use in this review. [75KAK/GIE] The heat of solution of nickel chloride hexahydrate into 4.36 M HCl(aq) was measured ((21.32 ±0.13) kJ-mor1). Few experimental details are provided, and a double correction (to 0.0 M HCl(aq) and to / = 0) with large uncertainties would be required to use the reported value to calculate the enthalpy of solution of the salt at infinite dilution in water. This heat of solution value is not used further in the present review.
374
A. Discussion of selected references
[75KOS] The molar heat capacities of NiCl2 from 2 to 30 K have been investigated. Smoothed values in this region were listed. The results were compared with the predictions of the spin-wave theory. [75LEV/GOL] The authors investigated the equilibrium potentials of the following galvanic cell with a solid electrolyte: Pt | (Ni + SiO2) Ni 2 Si0 4 | O2 | Ni, NiO | Pt. Based on the Gibbs energy of the cell reaction determined in the temperature range of 1228 to 1355 K, the standard Gibbs energy of formation and the standard molar enthalpy of formation of nickel orthosilicate at 298.15 K were calculated. We used the raw data of this work to represent the Gibbs energy of the reaction: 2Ni + SiO2 + O2 ^
Ni2SiO4
as a function of temperature. [75LIB/TIA] Emf and spectrophotometric methods have been used to determine the stability constants of MC1+ complexes in M(C1O4)2 medium (M = Mn2+, Co2+, Ni2+ and Zn2+) at different concentrations. The following cell was used for the potentiometic determination of formation constants: Ag | AgCl.HCKw^MgCClC^Mm) || M(ClO4)2(m),HCl(w,), AgCl | Ag, where rrt\ was fixed at approximately 0.01 m, while the molalities of the metal perchlorate (m) were 1.0, 1.5, 2, 2.5 and 3 m. The emf of the cells varied between 1.0 and 26.4 mV. The calculation was based on the following plausible assumptions: (i) the activity coefficient of the chloride is identical in the equimolar solution of Mg(C104)2 and M(C1O4)2, (ii) the liquid-junction potential of the above cell is negligible, (iii) Mg(II) does not form complexes or ion pairs with chloride. In the spectrophotometric study, the strong absorption of the CuCl+ complex at 272 run was used to assess the formation constants of the MC1+ complexes. At this wavelength the molar absorptivities of both the M(H2O)2+ and the MC1+ complexes were assumed to be negligible. The free chloride concentration and, thus, the formation constants of the MC1+ complexes were calculated using the formation constant and the £x=272nm value of the CuG + complex determined earlier [73LIB]. The two methods resulted in nearly coincident values. The main advantage of the above experimental procedure is that in Ni(C104)2 media the medium effect can be eliminated. In the case of nickel, the results are interpreted in terms of the existence of [Ni(H2O)5]Cl+ and [NiCl(H2O)5]+ complexes. The overall stability constant of the NiCl+ species (fix = 0.37 at / = 0) is due mainly to the stability of the outer-sphere ion pair. The fi\ value for the inner-sphere complex was reported to be around 0.01 M"1.
A. Discussion of selected references
375
[75MEY/WAR] The equilibrium partial pressure of sulphur over Ni — S melts was determined by bubbling H2/H2S gas mixtures through the melts at temperatures between 1373 and 1873 K. The compositions of the melt varied from nickel saturation to 27 wt% sulphur. The composition of the gas phase was monitored with the help of an optical interferometer, and the composition of the melt was determined by chemical analysis. The sulphur partial pressure for a given gas composition was calculated from the well-known equilibrium constants for the reaction: H 2 S(g)^H 2 (g)+'/ 2 S 2 (g). In addition, the liquidus line of the binary phase diagram Ni - S was obtained in the region from 0 to 20 wt% sulphur. [75MOS/FIT] The data concerning NiO are essentially the same as in [75MOS/FIT2]. [75MOS/FIT2] The least-squares line representing the temperature dependence of AfG°(NiO) was determined to be AfGm(7) = (- 244.019 + 0.092 (77K)) kJ-mol '. However, no detailed information concerning uncertainties of the measured values was available. [75RAU2] The partial pressure of sulphur in equilibrium with the non-stoichiometric hightemperature modification of Nii^S was measured as a function of composition in the temperature range from 781 to 1250 K. The sulphur pressure was determined directly by a tensimetric method. In addition, the solid sample was reduced with hydrogen gas. The amount of H2S(g) formed during the reduction as well as the ratio between H2S(g) and H2(g) in the gas phase were analysed by tensimetric measurements. From these data both the composition of the solid phase and the temperature dependence of the sulphur activity in Ni,^S, with S2(g) being the reference state for sulphur, could be calculated from the well-known equilibrium: H2S(g) ^
H2(g) + >/2S2(g).
Moreover, phase boundaries of the homogeneity range of Nil ^S are likewise obtained from this equilibrium study. [75TAM] This is a reanalysis of some of the data of Katayama [73KAT] using different activity coefficient assumptions.
376
A. Discussion of selected references
[75WEE/KOE] Weenk et al. determined the enthalpies of solution of a series of alkalitrichloronickelates ANiCl3 (A = Cs, Rb, K, Tl, NH4) and their constituting chlorides (NiCl2 and AC1) in a medium of 0.1 M HC1 at 298.15 K with «(H2O) / «(NiCl2) = 1200. A solution calorimeter of the isoperibol type was used. The molar enthalpy of solution of NiCl2 was found to be A sol //° = - (76.82 + 0.42) kJ-mol"1 for the reaction: NiCl2^Ni2++2Cl. This value refers to an ionic strength / = 0.24 mol-kg ', which is too high for the usual extrapolation to / = 0 to be applicable. Consequently this single value was disregarded in this review. [76BAE/MES] This is a well-known standard work concerning metal ion hydrolysis in water. Selected experimental values for 298.15 K were converted to zero ionic strength using the equations from Pitzer and Brewer [61 LEW/RAN]. For the water soluble nickel(II) hydroxo complexes, N i ^ O H ^ the following log10 j3°x values are reported: log10 fi°} = - (9.86 ± 0.03), logl0 */?2°, = - (19 + 1) log,0 %= - (30.0 ± 0.5), logl0 > ° , < - 44, logl0 */?°2= -(10.7 + 0.5) and log10 *J3°4= -(27.74 + 0.02). No thermodynamic data have been taken directly from this reference. [76BER2] A high quality data set was obtained by accurate measurements of the Gibbs energy of formation of NiO by using a solid state emf method with a gaseous hydrogen/water buffer atmosphere. The result of the weighed linear regression for the temperature range from 825 to 1675 K was given by the equation: AfG° (NiO) =-(235.071 + 0.141) +(0.08620+ 0.00010)-(77K)kJ-mol '. [76BIA/CAR] A method using variation of solvent composition to study the formation equilibria of complexes and its application to the nickel(II)-chloride system is reported. In pure acetone the formation of the complex NiCl^" was detected, and this species readily decomposed with increasing water content. The extrapolation of experimental results, determined between 0 and 0.05 mole fraction of water, to pure water, resulted in a value of log10 /?(NiCl^~) — 14 as a rough approximation. [76GEE/SHE] The emf measured for solid electrolyte cells of the type: Pt | M', M'C121 electrolyte | M, MC121 Pt provided Gibbs energies of formation for CrCl2, MnCl2, CoCl2 and NiCl2 at
A. Discussion of selected references
377
470 < 7 7 K < 910. Depending on the temperature range either PbCl2 or BaCl2 were used as the solid electrolyte. The Fe, FeCl2 electrode served as reference electrode. From results two-parameter Gibbs energy of formation equations have been derived, which obviously contain the uncertainties of the thermodynamic parameters ascribed to the substances of the reference half-cell. The Gibbs energy equation related to NiCl2 has been plotted in Figure V-18. [76KAR/KUL] This is a description of the "Kalori" computer programme (see the discussion of [74KUL3]). As one of the applications, the nickel(II) - thiocyanate system was mentioned. Using the identical dataset, the authors reported stability constants and enthalpy values that differ slightly from those reported for the same complexes in the original paper [74KUL3]. [76KUL/BLO] See comments under [81KUL/BLO]. [76MUR/KUR] A solvent extraction method was used to determine, the stability of NiCl+ and NiSCN+ complexes, among others. The aqueous phase was 1 M NaClC>4 solution, while the organic phase was CCI4 containing 0.05 M thenoyltrifluoroacetone and 0.03 M trioctylphosphine. The concentration of nickel(II) was ~ 1.0 x 10 4 M, while that of the complex forming anion varied up to 0.4 M for thiocyanate and ca 0.85 M in case of chloride. As the amount of metal extracted was very low in the case of chloride, even at high chloride concentrations, only the formation constant reported for the thiocyanate complex is considered in this review. [76NAV/KAS] The enthalpy of formation from the oxides of Mg2SnO4 and Co2SnO4 were found by oxide solution calorimetry. Using thermochemical data for the formation of olivine, for olivine - spinel transitions and for the transformation of quartz to stishovite, pressures for the disproportionation of silicate spinels were calculated to be in the range 150 - 200 kbar. [76PAN/PAT] Pant and Pathak measured the solubilities of Ni3(PO4)2-7H2O and NiHPO4 in water from 30 to 60°C, and reported the metal concentration in solution (0.012 M to 0.027 M), the metal to phosphate ratio in solution (near 2.0 in all cases) and the "pH" (between 7.1 and 7.7, without allowance for the change in Kw with temperature). In the absence of a good set of values for the complexation constants for Ni2+ with phosphate ligands as a function of temperature, and good analyses of the final solids, no useful thermodynamic quantities can be extracted from the information in this paper.
378
A. Discussion of selected references
[76PER] The kinetics of formation of Ni(CN)4~ have been studied spectrophotometrically at 367.5 nm in the pH-range 4.7 - 7.0, at / = 3 M NaClO4(aq) and / = 25°C. The author found that, at the equilibrium, only Ni2+ and Ni(CN)^ exist in substantial concentrations, although the complexes NiCN+, Ni(CN)2(aq) and Ni(CN)j are kinetically significant. In contrast to the acidic dissociation of Ni(CN)4" [76PER/EK.S], its association from nickel(II) can be described, within the experimental errors, without the protonated complexes. This is probably due to the higher pH range used in this study. From the kinetic data presented in this paper, the author determined the formation constant of Ni(CN)^ (log l0 PA = (31.08 ± 0.05), at / = 3 M NaC104 and t = 25°C, using pKHCN = 9.484 [71 PER]). This value of logl0 /?4 is retained in the present review, but with the uncertainty increased to ± 0.15. [76PER/EKS] The acidic dissociation of Ni(CN);;~ was studied spectrophotometrically in the pH range 0 < pH < 3, at / = 3 M NaClO4 and t = 25°C, in order to elucidate the presence of protonated H A Ni(CN)^ 2 species. Persson [74PER] reported Ni(CN)^ as the only complex existing in the nickel(II) - cyanide system. On the other hand, Kolski and Margerum, in [68KOL/MAR], reported extensive formation of protonated complexes. Persson and Ekstrom [76PER/EKS] acknowledged the formation of protonated species and their role in the acidic decomposition, but their data indicated that the protonated species have steady state concentrations during the decomposition. Therefore, according to the authors, the pK values of the protonated complexes should be much lower than suggested in [68KOL/MAR]. [76RAU] The activity of sulphur in the high temperature modification of Ni3S2 was determined as a function of composition at temperatures between 815 and 1044 K. The solid sample was reduced with hydrogen gas in a stepwise manner. The amount of H2S(g) formed during the reduction as well as the ratio between H2S(g) and H2(g) in the gas phase were obtained from tensimetric measurements. From these data both the composition of the solid phase and the temperature dependence of the sulphur activity in Ni3S2, with S2(g) being the reference state for sulphur, could be calculated from the well-known equilibrium H2S(g) ^=i H2(g) + '/2S2(g). Neither the initial compound nor the solid products formed owing to reduction with hydrogen gas were checked by X-ray diffraction analysis. [76SHI/TSU] This was a conductance study done at pressures from 0.1 to 160 MPa and at 15, 25 and 40°C. The sets of equivalent conductivities are reported for a very limited range of concentrations from 0.0001 to 0.0003 M, and have the appearance of being smoothed val-
A. Discussion of selected references
379
ues from unreported primary data. The data analysis used the simple Onsager equation to extrapolate the NiSO4(aq) dissociation constant to / = 0, and values for A r // m and AV are reported. The reported association constants are greater than those found in many other studies. Re-analysis of the "1 atm." results (assumed to be identical at 0.1 MPa) using a version of the Lee-Wheaton equation that is consistent with the SIT equation used in the present review (Appendix B) leads to results that are consistent with the other studies. In the present review, we estimate the uncertainty in K\ at 25°C (0.1 MPa) to be ± 30. The high-pressure results are consistent with values found in a later study of Fisher and Fox [79FIS/FOX]. The average enthalpy of reaction, calculated from the equally weighted values for the temperature range 15 to 40°C, is 6kJmol ', in excellent agreement with the value obtained from the data of Katayama [73KAT]. An uncertainty of + 2.5 kJ-mol ' is estimated in the present review. [76SMI/MAR] This is a well-known compilation of stability constants including a number of metal ions and a variety of inorganic ligands. For the water soluble nickel(II) hydroxo complexes the following logl0 P x for the reaction: 2+
XNi
+yOH
^
Ni v (OH)' AV
are reported: log]0 fl° = 4.1, log10 £°, = 8, log,0 &» = 11, logl0 / ? u = 4.2 (/ = 3 M) and log10 *P°AA = 28.3 (/ = 0), (29.37 + 0.03) ( 7 = 3 M). No thermodynamic data have been taken directly from this reference. [76TAY/DIE] This paper reports a pH study of complexation between HPO^ and Ni2+ at 25°C in 0.1MNaClO 4 : Ni2+ + HPO' ^ NiHPO4(aq)
(A.59)
Nickel(II) concentrations were between 7 x 10 3 and 3.3 x 10~2 M, and the initial ligand concentration was approximately 2 x 10 2 M. The association constant between Ni2+ and HPO^" was reported as (130 ±6) dm3-mol ', and the association constant between Ni2+ and H2PO4 to form NiF^PO* as (3.5 ± 1) dm 3 -mol'. There is insufficient information to recalculate the results. For the protonation constants for HPO 2 ", the reported log10 K values are within + 0.2 of values from other studies, and in the present review a similar uncertainty is assigned to the reported NiHPO4(aq) formation constant (log 1 0 ^ = (2.11 ±0.20)). Correction to zero ionic strength (Appendix B) gives log,0 A^° (A.59) = (3.00 + 0.20), where the uncertainty has been estimated in the present review. The small Ae correction has been based on the value for s(Na+, H P O j ) from Table B-5 (rather than using the equation in Table B-6), and e(Ni2+, C1O4) = 0.37 kgmol '. The evidence for H2NiPO4 is not considered to be sufficient to assign a formation constant for this species.
380
A. Discussion of selected references
[77ASH/HAN] UV spectroscopic measurements (/ = 5 mol-dnT3, NaC104/Na2S04) were compared for Na2SO4 concentrations of 0.1 mol-dnf3 and 1.0 mol-dnT3. Measurements for 47, 60 and 70°C were reported. The authors extrapolated their measurements to 25°C, assuming the effects of temperature were small (A r // m was assumed to be 0 kJmol '), and reported logl0 Kx (298.15 K, / = 5 M NaC104) = 0.77. The assumption concerning A r // m is reasonable in a general sense, but probably introduces an uncertainty of the order of ± 0.2 in the value of log10 Ki (298.15 K). The ionic strength (6.6 m) is too high to properly apply the SIT. If the SIT is used with £(Ni2+, ClO^)equal to (0.37 ± 0.03), log10 K° = 1.3 is calculated. [77BAR/KNA] This volume contains the Planck function, the enthalpy and the Gibbs energy tabulated at 100 K intervals for pure substances. It continued the work started by [73BAR/KNA]. [77BIL/TER] In this paper, the heat capacity and magnetisation of Nil2 were reported from measurements done between 20 K and 150 K. The heat capacity measurements showed peaks at 59.5 K and 76 K (later ascribed by Kuindersma et al. [81KUI/SAN] to a structural phase transition and a magnetic phase transition, respectively), and a shoulder and a peak were found at the corresponding temperatures in the magnetic measurements. The measurements are reported only in graphical form. In the present review, the values of Cp m (7) from the authors' Figure 2(a) were digitized and used to calculate values of Cpm/T. Equations were fit to the values over short ranges of temperature and to a set of values including points for temperatures above and below the anomalies. If the contributions from the two transitions are neglected, the contribution to the entropy between 50 K and 85 K is calculated to be 27.2 J-K '-mol '. The additional contributions to the entropy from the transitions near 59.5 K. and 76 K are estimated to be 0.40 J K - ' m o l 1 and 0.37 J-K ' m o l 1 , respectively. [77CHE/BRO] The I4N nuclear quadripole resonance spectra were measured for several Pd(II) and Ni(II) thiocyanate complexes. The thiocyanate is N-bonded in all nickel(II) compounds. [77FLE] Fleet investigated synthetic a-Ni3S2 and demonstrated unambiguously that there are four similar Ni—Ni 'bonds' per Ni atom. The short Ni-Ni distances in heazlewoodite are only slightly greater than those for metallic Ni thus indicating metallic bonding.
A. Discussion of selected references
381
[77KAS/SAK] The chloride-formation reaction of nickel oxide and other metal oxides with HC1 gas were studied at temperatures from 25 to 500°C and at 1.0 atm. The experiments were carried out by using gas analysis, a "thermobalance" and X-ray analysis to determine compositions for the reaction: NiO (cr) + 2 HC1 (g) ^ NiCl2 (cr) + H2O (g). A third-law analysis of the reported equilibrium constants at 298.15 and 773.15 K resulted in AfH^ (NiCl2, cr, 298.15 K) = - (316.2 ± 6.4) kJ-mol '. This value has a comparatively large uncertainty, does not overlap with the values obtained by the definitive studies of [53BUS/GIA] and [84LAV/TIM], and is therefore rejected. [77KAT] This paper (in Japanese) uses the same nickel sulphate conductiometric data reported previously by the same author [73KAT], and compares the results with those for other electrolytes. [77NIC/ROB] The otwayite was investigated with respect to its chemical composition, its optical and physical properties and powder diffraction patterns. No synthesis of otwayite has been attempted so far. [77OGA] The heat capacities of FeS2, CoS2, NiS2 and binary solid solutions of these sulphides were measured over the temperature range from 15 to 350 K in order to investigate the magnetic contribution to the specific heat at low temperatures. The solid samples were not characterised by X-ray diffraction analysis. The temperature dependence of the heat capacity data for NiS2 is depicted in Figure A-28. Between 0 and 20.5 K, a Debye - T3 function is fitted to the data, and at higher temperatures a spline fit of polynomials of the general form (a + bT+ el2 + dT2 + eT~'A) is applied, leading to the solid line in Figure A-28. The integration j.298.i5K^ /T^dT r e s u i t s i n s° = (79.6+ 2.4) J-K '-mol 1 . The integration of the Cpm function of FeS2 yields 5° = 54.4 J-K" '-mol"1 which deviates from the value listed in the JANAF tables [98CHA], 5°= 52.9 J-K ' m o l 1 , by 3%. Thus, the same relative error is assumed for the standard entropy of NiS2, leading to the uncertainty of +2.4 J-K~'-mor'.
382
A. Discussion of selected references
Figure A-28: Heat capacity of NiS2 plotted versus temperature, o [77OGA], solid line: polynomial fit to the experimental data.
[77OSW/ASP] Oswald and Asper reviewed the preparation and properties of bivalent metal hydroxides. [78BEC/LIU] In this study, the effect of temperature (0 to 180°C) on formation of inner-sphere and outer-sphere nickel sulphate complexes was investigated by 17O nmr spectroscopy. The solutions used were 0.097 m NiSC>4 in 1.07 m and 2.34 m U^SO^aq). Results indicated that inner-sphere complexation increased with increasing temperature, and that at least two different inner-sphere complexes were formed, involving displacement of one and two inner-sphere water molecules, respectively. It was not clearly established whether the second complex was formed by bidentate complexation of a single inner-sphere sulphate ion, or by unidentate complexation of two sulphate ions. Near 25°C, the estimated value of the formation constant for the first inner-sphere complex was approximately an order of magnitude less than the value of the formation constant as measured by methods [59NAI/NAN], [73KAT] that did not distinguish between inner- and outersphere complexation. The complex formed by displacement of two inner-sphere water molecules became predominant at temperatures above 160°C.
A. Discussion of selected references
383
[78BUR/KAM] A potentiometric study of the hydrolysis of nickel(II), similar to the studies of [65BUR/LIL], [65BUR/LIL2], [66BUR/IVA] and [71BUR/ZIN], at 25°C, and in 3 M (Na)Br medium is described. The metal ion concentration ranged from 0.2 to 1.4 M (/ = 3.2 - 4.4 M). The best fit to the experimental data was obtained assuming the presence of Ni4(OH)4+ and Ni2OH3+ species, with formation constants log10 */?44 = - (28.18 ± 0.05) and logl0 */3]2 = -(9.5 ±0.1). During these measurements, a large part of the Na+ ion is replaced by Ni2+; consequently the constancy of activity coefficients at different nickel(II) concentrations is questionable. Furthermore, the formation of bromo complexes (both binary and ternary) was not considered. Therefore, the reported experimental data were re-evaluated for the purposes of this review. At / = 3.0 M, equilibrium constant for the reaction: Ni2+ + Br ^ N i B r +
(A.60)
logl0 /?, (A.60) = - 0.49 can be calculated for the NiBr+ complex in NaBr medium, using the recommended value log,0 /?" (A.60) and s(NiBr+, Br ) = !/2{£(Ni2+, Br) + s(Na+, Br~)} = (0.16 + 0.03) kg-moP1. Considering the whole dataset and the same model as proposed, logl0 */34A= - (28.22 ± 0.04) (3a) and logl0 */?i,2= - (9.34 + 0.08) can be calculated, in good agreement with the published values. Using only the data sets reported for [Ni2+]tot = 0.2, 0.4 and 0.8 M (to reduce the influence of medium effects), and taking into account the formation of NiBr+, logl0 /? 44 = - (27.19 + 0.06) (3a) and log10 /?, 2 = - (8.82 ±0.10) can be derived. If the formation of NiOH+ is also considered, the fit to the data does not improve. Thus, the values for the other two constants are selected in this review, with increased uncertainties: logl0 j34A= - (27.19 ± 0.30) and log10 Vi,2 = - (8-82 ± 0.40). [78ENO/TSU] Enoki and Tsujikawa measured the heat capacity of hydrothermally aged, bulky Ni(OH)2 crystals from 4.2 to 35 K. Whereas the magnetic entropy was found to be higher by 0.37 J K '-mol' than that of Sorai et al. [69SOR/KOS], Cpm values above the transition temperature were lower. However, as the results were presented graphically in the rather narrow temperature range mentioned above, they were taken into account only by increasing the error limits. [78FOR] The equilibria in the nickel(II)-imidazole-chloride ternary system have been investigated by means of pH-metric and spectrophotometric titrations in 3 M Na(ClO4, Cl) solution. The latter method was used to determine the stability constant of the binary NiCl+ complex in the nickel(II)-chloride system. The concentration range studied was [Ni2+] = 0.096 to 0.300 M and [Cl ] = 1.25 to 3.00 M. The log10/?, value obtained (- (0.47 + 0.10)) is in good agreement with that determined from potentiometric data in
384
A. Discussion of selected references
an indirect way from the ternary system (- (0.53 ± 0.10)). In this review, the mean value of these constants is accepted with an increased uncertainty (log l0 /?, = - (0.50 ± 0.40)). [78KAR/HUB] The stability constant of a complex of pyrophosphate with Ni2+ (presumably NiP2Oy~ though this is not specified by the authors) was measured by an amperometric titration procedure. Small amounts of the nickel nitrate salt were titrated with the pyrophosphate in an ammonium nitrate buffer solution (0.1 M), adjusted to a pH value of 8 with aqueous ammonia. A PbO2 indicating electrode was used with a platinum foil counter electrode and a saturated calomel reference electrode. The procedure was found to be very sensitive to pyrophosphate even in the presence of phosphate. At 30°C, the measured value of logl0 K was 5.35, and the authors used estimated activity coefficients to obtain log10 K = 6.35 at / = 0. The original data are not reported and cannot be re-analysed. For reaction: Ni2+ + P2O'" ^
NiPjO'-
in 0.1 M KNO3, logl0 K at 30°C = (7.11 + 0.1 Ae) As the ionic strength is low, the effect of the AE term on the final calculated value of log)0 K° is limited. Rather than use estimated s values for interactions involving the pyrophosphate moieties (some of which are highly charged), it is simply assumed that AE is zero. As a consequence, the uncertainty in the logl0 K° value is increased to ± 0.30 for the values from measurements in 0.1 M NH4NO3 solutions. [78LIB/KOW] The experimental procedure is very similar to that used in [75LIB/TIA]. In the 1978 work, a potentiometric method was used to determine the stability constants of MBr+ complexes in M(C1O4)2 medium (M = Mn2+, Co2+, Ni2+ and Zn2+) at different concentrations. The following cell was used for the potentiometric determination of formation constants: Ag|AgCl,HBr(w1),Mg(C104)2(w) || M(ClO4)2(m),HBr(7M,),AgCl|Ag, where rri\ was fixed at around 0.01 m, while the molalities of the metal perchlorate (m) were 1.5, 2, 2.5 and 3 m. The emf of the cells varied between 1.6 and 12 mV. Among the metal ions studied, nickel(II) forms the least stable bromo complex. The variation of the stability constants within the series is dictated by the formation of inner-sphere complexes to different degrees. [78LIN/HU] The activity of sulphur in the high-temperature modification of Ni3S2 was determined as a function of composition of the solid solution at various temperatures ranging from 823 to 1023 K. The sulphide was equilibrated with gas mixtures of H2(g) and H2S(g) and the
385
A. Discussion of selected references
composition of the gas phase was obtained from the flow rates of the gas streams. When S2(g) is the standard state for sulphur the activity of sulphur can be calculated from the well-known thermodynamic data for the gas phase equilibrium H2S(g) ^ H2(g) + 1/2 S2(g). The composition dependence of the sulphur activity at constant temperature revealed that the high-temperature modification of Ni3S2 is composed of two nonstoichiometric solid solutions, viz. p r Ni 3 S 2 and (32-Ni4S3. The homogeneity range as well as the phase relationships of these phases were described by means of a sub-regular model. Based on the thermodynamic properties of these mixture phases (Pi-Ni3S2 and (32-Ni4S3) a new phase diagram for the binary system Ni - S was calculated, opposing the nickel-rich portion of the phase diagram proposed by Kullerud and Yund [62KUL/YUN] as well as Rau [76RAU]. [78SCH/MIL] The phase relations in the binary system Ni - S were investigated in the composition range from xNi = 0.33 to xNj = 0.83 at temperatures between 509 and 1287 K by employing isothermal sublimation in Knudsen effusion cells. The authors derived Gibbs energies of formation for Ni3S2, Ni7S6, NiS, Ni3S4, and NiS2 as a function of temperature as well as enthalpies of formation for T— 298.15 K. No original experimental data are reported. The results given by the authors (see Table A-16) are not consistent with any other thermodynamic study on nickel sulphides, published so far. Thus, the data given in this paper are discarded. Table A-16: Gibbs energy functions for various nickel sulphides. Reaction 3Ni + S 2 ^
A r //° (298.15 K)
(K.)
(kJmol"')
152.297-0.065689 T
925-1070
-65.27
167.36-0.09037 7"
640 - 840
- 90.79
2NiS
184.93-0.11004 T
6 7 0 - 1200
-72.38
T
212.97-0.08703 T
600 - 625
- 79.08
180.33-0.12761 T
509 - 730
-87.86
N13S2
j / Ni + S2 ^ 2Ni +S2 ^ y2 Ni + S2 ^ Ni + S2 ^
Temperature range (kJ-mol ')
'/3Ni7S6
/2Ni3S4
NiS2
[78STU/FER] For anhydrous NiBr2, the authors report low temperature Cpm measurements between 7.9 K and 303 K, drop calorimetry measurements to 1150 K, and heat of solution measurements at 298 K. Samples of NiBr2(cr), NiSO4(cr) and H2SO4-6H2O(1) were dissolved in 4.36 m HCl(aq). Prior to use, the NiBr2 was heated for several weeks at 200°C, and the X-ray diffraction pattern was claimed to be in good agreement with the ASTM pattern. Nevertheless, it is not certain which crystalline form of the solid was used [34KET]. The values were combined with the results of earlier measurements to determine A r // m (A.61) = (12.320 ± 0.319) kJ-mol"1.
386
A. Discussion of selected references
2[HC112.731H2O](1) +2NaBr(cr) +NiSO4(cr) ^ 19.462H2O(1) +2NaCl(cr) +H2SO4-6H2O(1) +NiBr2(cr)
(A.61)
The auxiliary data used by the authors were updated using the values in Table A-17. Table A-17: Auxiliary data used in the recalculation of NaBr heat of formation values with the data from [78STU/FER]. Reaction #
Equation
A r //°
Reference
(A.62)
0.5H2(g) + 0.5Cl 2 (g)+12.731H 2 O(1)^HC1-12.731H 2 O
-(162.344 + 0.167)
a
(A.63)
Na(cr) + 0.5Br,(l) ^
-(361.160 ± 0.200)
b
(A.64)
Ni(cr) + S(rh) + 2O2(g) ^ NiSO4(cr)
- (873.280 ± 1.570)
c
(A.65)
Na(cr) + O.5Cl2(g) ^
- (411.260 ± 0.120)
d
(A.66)
H2(g) + 2O2(g) + 6H2O + S(rh) ^
-(874.295 + 0.418)
a
NaBr(cr)
NaCl(cr) H2SCV6H2O(1)
(a) [82WAG/EVA] adjusted to be consistent with values in Table IV. 1 (b) [82GLU/GUR] (selected auxiliary datum) (c) This review. Section V.5.1.2.2.1.3. (d) Table IV. 1 ([82GLU/GUR], selected in [2001LEM/FUG])
Thus, for Ni(cr) +Br2(l) ^ NiBr2(cr)
(A.67)
ArH°m (A.67) = A r #° (A.61) + 2 A r tf ° (A.62) + 2 ArH°m (A.63) + A r //° (A.64) 2A r i/°(A.65)-A r tf°(A.66), and
Af//n°,(NiBr2, cr) = -(211.214 + 1.752) kJmol '.
The method used is a reasonable way to determine A f //° (NiBr2, cr), but any consequent determination of the value of the enthalpy of formation of Ni2+ with this value for AfH^ (NiBr2, cr) is linked to the value from the sulphate cycle. The heat of solution of Ni(c) in the sulphuric acid solution is included in part of the cycle to determine the required A f //° (NiSO4, cr) auxiliary value. The authors' integration of their low-temperature heat capacity measurements (adiabatic calorimetry) led to 5° (NiBr2, cr, 298.15 K) = (122.42 ± 0.25) J-K 'mol '. The reported value of C°m(NiBr2, cr, 298.15 K) was (75.40 + 0.02) J-K '-mol"'. The heat capacity results showed no sharp transition above 7.8 K, though there was a small excess heat capacity between 40 and 55 K, near the Neel temperature. There was no indication of the transition near 20 K [81ADA/BIL], [82WHI/STA], and the reported heat capacity values for temperatures between 7.8 K and 15 K are markedly higher than the corresponding values measured by White and Staveley [82WHI/STA]. Also, as shown in Figure V-19, for the purposes of extrapolation to 0 K, an equation in T3 does not provide a satisfactory representation of the experimental Cpm values between 7.89 K and 18 K. It appears that the values reported for temperatures below 20 K are
A. Discussion of selected references
387
unreliable. In the present review, the contribution to the entropy from integration of CVJTbetween 22 K and 45.4 K was calculated to be only 0.4 J-KT'-mol1 greater than the contribution as calculated from the results of White and Staveley; and the contribution to the entropy between 46.5 K and 250.0 K was calculated to be 94.17 J-KT'-mor1, in good agreement with 94.31 J-K"1 moP1 from the later work [82WHI/STA]. The authors' fitted heat capacity equation reproduces all the high-temperature experimental Hm (T)-H°m (298.15) values to better than 0.5% and is fixed to the 298.15 K value from the low-temperature measurements. The values from the equation deviate slightly from the experimental values below 298.15 K. In the same report, the authors reported low-temperature calorimetry heat capacity measurements (9 to 70 K) for anhydrous NiSO^cr). Limited sets of dropcalorimetry measurements were carried out for the same solid (403 to 1001.5 K), and these were used to generate equations for thermodynamic functions for NiSO4 (cr) from 298 to 1200 K. [79FIS/FOX] This paper presents a thorough discussion of the pressure dependence of ion association parameters for several 2:2 electrolytes, including NiSO4(aq). New conductance data are reported for 15 and 25°C to 2000 atm (analysed using the Fernandez-Prini modification of the Fuoss-Hsai equation). The trends in the association constants for Ni2+ with sulphate are in agreement with earlier work [76SHI/TSU]. Reanalysis of the 1 atm results (assumed to be identical at 0.1 MPa) using a version of the Lee-Wheaton equation that is consistent with the SIT equation used in the present review (Appendix B) leads to results that are in good agreement with the other studies. The calculated association constants have a rather large (10%) statistical uncertainty, but it does not appear that the data are unusually badly scattered. In the present review, we estimate the uncertainty in Kx at 25°C (0.1 MPa) to be ± 60. The high-pressure results are consistent with values found in the study of Shimizu et al. [76SHI/TSU]. [79GOL/NUT] Goldberg et al. reviewed the available literature data, and provided recommended values for the activity coefficients and osmotic coefficients for solutions of Ni(NC>3)2 in water at 298.15 K. The authors commented on the "unusually large amount of scatter" between (and within) the osmotic coefficient measurements. The tables extend only to solutions 4.623 m in Ni(NO3)2 and, based on the study of Sieverts and Schreiner [34SIE/SCH], the saturation solubility is 5.47 m at 298.15 K. Therefore, in the present review, no thermodynamic quantities for Ni(NO3)2 solids are based on this careful assessment.
388
A. Discussion of selected references
[79KEM/KAT] Emf measurements on an electrochemical cell involving the Ni - NiO system were carried out in the temperature range from 1191 to 1823 K. No values of AfG°(NiO) have been reported hitherto at temperatures above 1726 K (the melting point of nickel). For this review, however, the results in the lower temperature range are of primary interest. The results were expressed by the authors as A f G°(Ni0) = ((-(232.463 ±0.251) + (0.083596 + 0.000167) (77K)) kJmol 1 . The standard deviation of the AfG° values amounted to ± 0.209 kJmol '. [79RIC/VRE] The standard molar entropy of solid NiB4O7 was estimated by the linear dependence of entropies on the radius of the compound's cation constituent. The extrapolation to r(cation) —> 0 led to the borate contribution of the standard molar entropy depending on the cationic charge. [79WEI/HER] Proton and deuteron relaxation rates in aqueous solutions of NiX2 (X = Cl, Br) in the concentration range 0.2 - 4 m are reported. The deuteron relaxation rates were found to depend substantially linearly upon the NiX2 concentration. Comparison of the concentration dependences of the deuteron relaxation rate in these solutions with those in diamagnetic MgX2 solutions yielded information about the average number («x) of halide ions bound in the first coordination sphere of Ni2+. For NiCl2 solutions na = (0.12 + 0.02)mNiCi2, for NiBr2 nBr = (0.085 ± 0.02)wNiBr2 was found. Some additional measurements with solutions containing 0.02 m NiX2 and 1 - 4 m KX were also performed, and «Ci = (0.045 ± 0.04)/wKCi, and «Br = (0.06 ± 0.04)mKBr were found. From the latter data the equilibrium constant for the reaction: Ni2+ + Cl ^ N i C l +
(A.68)
logl0 /?, ((A.68), 298.15 K) = - (1.26 ± 1.00) and for: Ni2+ + Br ^ N i B r + ,
(A.69)
log l 0 # ((A.69), 298.15 K) = -(1.11 ±0.60) can be derived in 4.0 m KX solutions. Using the recommended ion interaction coefficients to calculate Ae (A.68) and Ae (A.69) for KX background electrolytes, log10 fi° ((A.68), 298.15 K) = (0.18 ± 1.00) and logl0 P° ((A.69), 298.15 K) = (0.8 ± 0.6) can be derived. For the sake of comparison, we also calculated the "semi-thermodynamic" formation constants (log l0 (5°) as defined in [88BJE], using y± (4 m KC1) = 0.577 and y± (4 m KBr) = 0.608 [49ROB/STO]: log 10 #°'((A.68), 298.15 K) = -(1.03 ±1.00) and logl0/?,o'((A.69), 298.15 K) = - (0.89 + 0.60).
A. Discussion of selected references
389
[80BAR/RAN] Reversible potentials of partially charged a- and (3-Ni(OH)2 electrodes have been measured up to an oxidation state of « 2.5 over a range of KOH concentrations from 0.01 10.0 M. Couples derived from the parent a- and |3-Ni(OH)2 systems can be distinguished by the relative change in KOH level on oxidation and reduction as well as the difference in the formal potentials with respect to Hg | HgO | KOH. [80CHI/SAB] The solubility of Ni(OH)2 in pure water was measured from 298.15 to 313.15K. Only the solubility products calculated by the assumption 2[Ni(II)] = [OH ] have been given (at 25°C log l0 K° o = - 11.82). Thus, the conclusion has been confirmed that by direct reaction of Ni(OH)2(cr) with water no reliable solubility data can be obtained [18ALM], [67BLA]. Thus, these results were not included in either Table V-6 or Figure V-l 1. [80DUB/CLA] The phase transition between millerite, NiS((3), and the NiAs-type high-temperature modification, NiS(a), was studied by calorimetric and emf measurements. NiS(a) was prepared by solid state reaction from the elements. Millerite was obtained by annealing NiS(a) at 627 K for 3 - 4 days. Both solid compounds were checked by X-ray diffraction analysis. The calorimetric measurements of enthalpy increments during heating or cooling the sample yielded the temperature and the enthalpy of transition (p -> a), respectively, TUs = (667 ± 2) K, A tIS //°= (5.86±0.36) kJmol""'. In addition, the measurement of the emf of the electrochemical cell C (graphite) | Ni | NiCl21 NiS | C (graphite) as a function of temperature allowed the determination of an independent value for the entropy of transition, Atrs5° = (9.62 ± 0.84) J-K'-mol '. [80LIB/SAD] Isopiestic results (relative to KCl(aq)) are reported for NiSO4 at 25°C for nickel sulphate molalities between 0.1271 and 1.8526 mol-kg"1, and changes in the UV and visible spectra as a function of the concentration of nickel sulphate are discussed. The authors only present semi-quantitative estimates for the degree of association, and conclude that except for concentrations above 1 M, complexation between Ni2+ and SOj" is primarily outer-sphere complexation. [80MIL/BUG] The hydrolysis of nickel(II) in (K)C1 medium was studied by emf titrations at 25°C. In the first series of titrations, the chloride concentration was held constant at 2.0 M, and [Ni2+]t was varied from 0.05 - 0.50 M (/ = 2.05 - 2.5 M). In the second series, the [Ni2+]t was held constant at 0.10 M while the chloride concentration was varied from 0.5 to 3.0 M (/ = 0.6 - 3.1 M). Taking into account the smaller medium effect, the results of
390
A. Discussion of selected references
the second series are more reliable. The experimental data were interpreted by assuming that Ni4(OH)^+ was the main hydrolysis product formed. At / = 0.5 and 1.5 M the value of log10 /?,, was also estimated, although the authors noted that the formation of NiOH+ is not completely certain. The formation of the NiCl+ complex was not considered in the calculation; therefore a correction was applied for the purposes of this review, using the recommended logl0 fi° (NiCl+) and Ae(298.15 K, KC1) = - (0.085 + 0.05) kg-mol"1 (see the discussion of [71OHT/BIE]). The corrected formation constants are as follows:
Table A-18: Corrected formation constants. /(M)
0.5
1.0
2.0
3.0
log,0*/J,, reported
-(9.87 ±0.40)
•ogio'A.4 reported
-(28.04 ±0.10) - (28.14 ± 0.06) - (28.28 ± 0.06) - (28.24 ± 0.04) - (28.58 ± 0.06)
log,,, & (NiCl1)
-0.58
-
1.5
-0.65
log.o'X, corrected -(9.82 ±0.50)
-
-(10.06 ±0.40)
-0.67
-0.67
-0.63
-(9.95 ±0.50)
l o g , 0 X 4 corrected - (27.83 ± 0.30) - (27.80 ± 0.30) -(27.80 + 0.30) -(27.63 + 0.30) -(27.66 + 0.30)
The SIT-plot of the recalculated data, together with the accepted values for [65BUR/LIL2] and [71OHT/BIE], as well as the selected log)0 *,0°4 for the reaction: 4Ni2+ + 4H2O ^
Ni4(OH)4+ + 4H+,
(A.70)
is presented in Figure A-29. As can be seen, the analysis of the data from [80MIL/BUG] results in the determination of a considerably different logl0 /?4°4 value than the value selected in this review. The different medium cation can not be the reason of this deviation, since both the aquo ion and hydroxo complex of nickel(II) are positively charged. Therefore, the above data were not considered further in this review.
A. Discussion of selected references
391
Figure A-29: SIT-plot of recalculated data from [80MIL/BUG] (open diamond), [65BUR/LIL2] and [71OHT/BIE] (open squares), and the selected log10 */?°4 ((A.70), 298.15 K) (filled square).
[80OPP/TOS] Oppermann and Toschew studied the thermal decomposition and the sublimation of crystallised nickel(II) iodide using a membrane manometer and transport ampoules. The temperature functions of the total pressure (membrane manometer), as well as of the sublimation pressure (transport ampoules), in a range from 800 to 1000 K, were thermodynamically analysed, and the following quantities were obtained: A f //° (Nil2, cr, 298.15 K) = - (83.7 + 8.4) kJ-mol 1 , A f //°(NiI 2 , g, 298.15 K) = (130.5 + 20.9) kJmol" 1 , 5°(NiI 2 , cr, 298.15 K) = (146 + 8) J-K'-mol '-, 5n°,(NiI2, g, 298.15 K) = (335 + 8) J-K '-mol \ and 5^(Ni2I4, g, 298.15 K) = (536+13) JKT'mor 1 -. In this review the uncertainties given by the authors were accepted. The recalculation of these quantities suffered from the total pressure function having been presented only graphically. For the sake of comparison with independent measurements, an attempt was made to recalculate A f //° (Nil2, cr, 298.15 K) and S^ (Nil2, cr, 298.15 K), regardless of the uncertainty implicit in reading the relevant coordinates from a Figure. Based on the auxiliary data used throughout this review the second-law and approximate third-law methods employed by Oppermann and Toschew resulted in A f //° (Nil2, cr, 298.15 K) = -(91.45 +4.18) kJ-moP' and A f //° (Nil2, cr, 298.15 K) = -(91.00 + 8.37) kJmol ', respectively. Within the uncertainty limits, both of these values overlap with those reported by the original authors. The recalculated enthalpies of formation also agree with the value selected in this review (- (96.42 ± 0.84) kJ-mol'), but their uncertainty is 5 to 10 times greater.
392
A. Discussion of selected references
As a reliable value for S°m (Nil2, cr, 298.15 K) is a prerequisite for the third-law analysis, this quantity only can be determined by the second-law method. The standard entropy of nickel(II) iodide is recalculated to be S°m (Nil2, cr, 298.15 K) = (138.0 ± 8.4) J K '-mol ', which, within the large uncertainty limits, agrees very well with the value selected in this review (5° (Nil2, cr, 298.15 K) = (139.1 + 1.0) J K 'mol 1 ). [80REI] In this work neutral anhydrous transition metal carbonates MCO3 (M = Mn, Fe, Co, Ni, Zn) were synthesised by a hydrothermal autoclave method. The solubility of the well crystallised pure carbonates (size of rhombohedral crystals 10 - 100 \xm) in HOOVNaClC^ solutions of constant ionic strength (1.0 mol'kg ') and at various partial pressures of CO2 was studied by means of the pH variation method. At T = 323 K equilibrium was usually attained within a few days. Only NiCO3 was so inert that it had to be equilibrated at 363 K > T> 348 K. When the Gibbs energy of formation was calculated from the solubility constant obtained it was noted that it is remarkably more negative than the then-accepted value [77BAR/KNA]. [80SHA/CHA] The thermodynamic properties of all phases in the binary system Ni - S were optimised for the first time, allowing the calculation of the phase diagram for this system. The liquid phase was modelled by applying the three suffix Margules equation, the hightemperature phases p r Ni 3 S 2 and (52-Ni4S3 were described using sub-regular models, and the non-stoichiometric hexagonal modification of nickel monosulphide was modelled by application of a statistical model [69LIG/LIB] adopted previously for pyrrhotite, Fei XS, [77LIN/IPS], [79SHA/CHA]. The phase Ni7S6, which shows a narrow homogeneity range [94STO/FJE], was assumed to be a stoichiometric compound, neglecting its transformation into Ni9S8 at 675 K [94STO/FJE]. The phases Ni3S2, NiS (low-temperature form), Ni3S4, NiS2 were likewise assumed to be stoichiometric compounds. The calculated phase diagram enables excellent predictions of available experimental phase diagram data [62KUL/YUN], [75ARN/MAL]. However, the phase relations of Ni7S6 and Ni9Ss, respectively, were not considered properly. Furthermore, it should be mentioned that the thermodynamic data for NiS2, the low- and high-temperature forms of NiS are not in perfect coincidence with decomposition pressure measurements [54ROS], [64LEE/ROS] as well as heat capacity measurements published recently [95GRO/STO]. [80TRE/LEB2] The solubility of NiO, characterised by X-ray diffraction analysis, was measured in a flow apparatus from 423 to 573 K. The solid phase was treated with acidic solutions, containing HC1 with molalities ranging from 10 5 to 5 x 10 4 mol-kg"1, or basic solutions, containing NaOH with molalities ranging from 10 6 to 4 x 10 2 mol-kg"1. An ex-
393
A. Discussion of selected references
tended Debye-Hiickel expression was used by the authors in order to calculate the activity coefficients. The pH of the leach solutions was measured with a glass electrode, and varied between 3 - 1 2 . The nickel content of solutions equilibrated with NiO was determined continuously by means of an atomic-absorption spectrometer. As dilute solutions were investigated, only the formation of mononuclear hydroxo complexes (NiOH+, Ni(OH)2 and Ni(OH)j) was considered. The values for the hydrolysis constants at 25°C were taken from [64PER] and [49GAY/GAR]. From the temperature dependence of the solubility data the authors derived the first, second and third hydrolysis constants, see Table A-19. Table A-19: Stability constants for the complexes NiOH+, Ni(OH)2 and Ni(OH)~ trc
log10X,(A.71)
log,0 X , (A.72)
log,0 X , (A.73)
4, by means of a fluoride-selective electrode and a differential double calorimeter. The experiments were performed in the presence of an excess of metal ions (0.1 M Me(ClO4)2) to assure the formation of only monofluoro complexes. In the second paper [81KUL/BLO] the authors reported the ionic strength dependence of thermodynamic quantities using the same experimental techniques. The initial concentration of the metal ions was varied with varying ionic strength maintained with NaClO4 ([M] = 0.02 M for / = 0.1 M, [M] = 0.05 M for / = 0.25 M and [M] = 0.1 M for / = 0.5, 1.0, 2.0 and 3.0 M). The data revealed that the stability of MF+ complexes varies in the following order: Mn > Co ~ Ni < Cu > Zn, which is not consistent with the Irving-Williams series. The authors also reported thermodynamic properties extrapolated to / = 0, but a reassessment of these data using the SIT was done in this review. The authors classified NiF+ as an outer-sphere complex, since the addition of fluoride ions to a solution of nickel perchlorate is not accompanied by any changes in the electronic spectrum, and the thermodynamic quantities are very similar to those of the complex [Co(NH3)5NO2]F+, which is clearly an outer-sphere complex. In the third paper [83AVR/BLO], the stability constant values for NiF+ have been reported for 0 - 6 0 vol. % H2O/EtOH mixture (/ = 0.25, 0.5 and 1.0 M NaC104), using a fluoride-selective electrode. In contrast with the earlier conclusions, the authors suggest inner-sphere association for NiF+. This assumption was based on the determination of the so called complete equilibrium constant, in which the water is also taken into consideration in the equilibrium process: Ni(H2O)^+ + F ^
NiF(H2O)5+ + H2O(1).
A. Discussion of selected references
397
The reported thermodynamic quantities seem to have been re-determined in each study, since the three papers report somewhat different values for the parameters referring to identical conditions. To avoid the over-representation of the data from this laboratory during our SIT analysis, we used an average value for each ionic strength studied. [81MAR/ECO] Marcopoulos and Economou found almost pure nickel hydroxide in the Vermion region of northern Greece, described it comprehensively, named it theophrastite [81 MAR/ECO], and studied its genesis [83ECO/MAR]. The new mineral and its name were approved of by the Commission on New Minerals and Mineral Names, the International Mineralogy Association, IMA. [81NIC/BER] Crystallography and composition of nullaginite, found as nodular grains and cross-fibre vein lets in a nickel-rich assemblage from the Otway deposit, Nullagine district, Western Australia were investigated. The close relationship between nullaginite, Ni2(OH)2CO3, and rosasite, (Cu, Zn)2(OH)2CO3, as well as glaukosphaerite, (Cu, Ni)2(OH)2CO3, was pointed out. [81SCH] The partial pressure of oxygen for the equilibrium between (32-Ni4S3 and NiO at /?SOi = 1 arm was measured as a function of temperature between 968.6 and 1053.7 K. The emf of the electrochemical cell Pt, (32-Ni4S3, NiO, SO2 (1 atm) | ZrO21 O2 (0.0122 atm), Pt can be easily transformed into the equilibrium partial pressure of oxygen. The solid phases were checked by X-ray diffraction analysis. During equilibration sulphur was incorporated in excess in the non-stoichiometric high-temperature phase p2-Ni4S3 due to reaction with SO2 of the gas phase. The sulphur content of quenched samples was determined by standard analytical procedures. A composition of Ni2.515S2 was calculated from these chemical analyses. However, it is not expected that the sulphur content of p2~Ni4S3 remains constant when the temperature is varied. As the variation of the composition of p2-Ni4S3 in equilibrium with NiO and SO2 was not investigated as a function of temperature, the results of this work are not suitable for derivation of the Gibbs energy function of this phase. [81YOK/YAM] The data of Katayama [73KAT] were re-interpreted and compared to a calculated theoretical value. No new data are presented. [82CHI/SAB] Chickerur et al. [82CHI/SAB] reported solubility product values for Ni3(PO4)2-8H2O of 4.07 x 10 25 and 1.95 x 10 21 at 30 and 35°C, respectively. The paper provides few de-
398
A. Discussion of selected references
tails, though there is a suggestion that the measurements were done in solutions with "pH" values < 5.75. At higher temperatures, the authors indicated that Ni3(PO4)28H2O converted to "NiHPCV^O". No allowance was made for complexation, no raw data were presented, and no values can be recalculated from the sparse information in this paper. [82FER/GOK] Enthalpy increments from 298.15 to 1197 K were measured on Ni3S2 by means of drop calorimetry, leading to heat capacity values in this temperature region. The samples were prepared by solid state reaction from the elements and characterised properly (X-ray diffraction). The transition temperature for the phase transformation between the high- and low-temperature form was found to be 834 K. The heat capacity at 298.15 K amounts to 117.7 J-K '-mol ', which is in fair agreement with that obtained by Stolen et al. [91STO/GRO], C; m (298.15 K) = 118.2 J-K '-mol '. [82GAM/REI] The solubility constants measured at 363 K > T > 348 K were extrapolated to T = 298.15 K implying a linear dependence of logl0 Kp s 0 for the reaction: NiCO3(s) + 2H+ ^
Ni2+ + CO2(g) + H2O(1)
on \IT [80REI]. Data obtained at / = 1.0 and 0.2 molkg ' were extrapolated to / = 0 with the Davies approximation [62DAV]. On this basis AfG°(NiCO3, cr, 298.15 K) was calculated. [82KLO] The solubility of NiCO3(cr) synthesised by the hydrothermal autoclave method was determined at T= 363.15 K and / = 0.2 mol-kg ' NaC104. [82LIV/BIS] Livingstone and Bish reported X-ray powder diffraction, infrared, thermal, chemical, optical, and physical data for a Mg-bearing theophrastite from Hagdale Quarry, Unst, held in the Scottish Mineral Collection. These authors endeavoured to establish a mineral name for the compound (Ni, Mg)(OH)2 by a submission to the IMA in late 1979. Two months later IMA received another submission for pure Ni(OH)2 [81 MAR/ECO]. In the vote the Unst material and name were narrowly defeated in favour of the Greek Ni(OH)2. The values of c determined by Livingstone and Bish are intermediate between those for pure Ni(OH)2 and the isomorphous Mg(OH)2, brucite, indicating that a solidsolution series exists between these end-members.
A. Discussion of selected references
399
[82WAT] Watanabe carried out measurements of the molar heat capacities of high pressure compounds relevant to the earth's mantle by differential scanning calorimetry between 350 and 700 K. The investigated compounds were, among others, a-Ni2SiO4 and y-Ni2Si04. In general, the heat capacity for the high pressure phase was found to be smaller than that for the isochemical low-pressure phase i.e., Cpm decreases in the phase sequence olivine (a), modified spinel ((3), and spinel (y). The calculated standard entropies at 298 K for high-pressure phases are presented and entropy changes for various transformations of silicate compounds were discussed. The heat capacities of Ni2SiO4, olivine, reported by Watanabe are 4.6% larger than the heat capacities obtained by [84ROB/HEM] with the more accurate low-temperature calorimeter. Therefore, the results of Watanabe are not used by this review. [82WHI/STA] The authors report low-temperature heat capacity measurements (adiabatic calorimetry) from 8.03 K to 299.83 K. A sample of NiBr2 that had been resublimed at 600°C was used. When plotted against temperature, the results show a shoulder at 46 K, and a spike at 19.5 K (Figure A-32). The values between 20 K. and 250 K are in reasonable agreement with the earlier work of Stuve et al. [78STU/FER] (cf. discussion in Appendix A). At lower temperatures, the values are lower, and the value of Cpm(NiBr2, cr, 8.0 K) is only 30% of the value reported by Stuve et al. [78STU/FER]. As shown in Figure V-19, for the purposes of extrapolation to 0 K, an equation in T3 provides a satisfactory representation of the experimental Cpm values between 8.03 K and 18 K. In the present review, polynomials were fitted to Cp m IT, and integration resulted in Sm (NiBr2, cr, 250 K) = 108.23 kJ-mor1. The authors reported difficulties with their measurements between 250 K and 300 K, and the values of Cpm increase more rapidly than expected. Integration of CpJT between 250 K and 300 K gives a contribution to Sm (NiBr2,cr) that is 0.61 J K 'mol ' greater than the contribution obtained by integration of values from Stuve et al. over the same temperature range. The results from White and Staveley for the temperature range 250 K to 300 K are not accepted in the present review. [83AVR/BLO] See comments under [81KUL/BLO]. [83ECO/MAR] Marcopoulos and Economou found almost pure nickel hydroxide in the Vermion region of northern Greece, described it comprehensively, named it theophrastite [81 MAR/ECO], and studied its genesis [83ECO/MAR]. The new mineral and its name were approved of by the Commission on New Minerals and Mineral Names, the International Mineralogy Association, IMA.
400
A. Discussion of selected references
Figure A-32: Values of C pm (NiBr 2 , cr) as a function of temperature as reported by White and Staveley [82WHI/STA].
[83GOW/DOD] A 35C1 NMR study of NiCl2 solutions has shown that at — 5 and + 1°C the system is in the slow-exchange region. This allowed the concentration of the bound chloride ion to be calculated, and thus the stability constant for the inner-sphere Ni(H2O)5Cl+ complex at different concentrations of the salt. The average number of inner-sphere-bound chlorides varied between na = 0.07 and 0.27, depending on the concentration of NiCl2. The determined stability constants do not show notable temperature or ionic strength dependence, and logl0 /?, = (0.034 ± 0.020) was determined from the whole dataset. The applied ionic strength is out of the validity range of the SIT. Nevertheless, if it is used, logl0 p° for the reaction: Ni2
Cl ^ NiCl+
can be calculated to be — (0.95 ± 0.50). The method applied has substantial inherent inaccuracy, but since outer-sphere interaction is unlikely to produce slow ligandexchange, the results verify the existence of an inner-sphere complex.
A. Discussion of selected references
401
[83HOL/MES] The authors carried out isopiestic measurements on aqueous solutions of nickel sulphate at 383.14 K and 413.22 K for nickel sulphate molalities between 1.52 and 5.66. The results were interpreted using a modified Pitzer equation without invoking formation of NiSO4(aq) as discrete species. The results at these high ionic strengths were not used in the present assessment. [83SOL/BON] The thermodynamics {K, ATHm, ArSm ) for the formation of monofluoride complexes of the Mn(II) - Zn(II) series have been studied in methanol at / = 0.05 M (CH3)4NC1O4 and 298.15 K with use of fluoride selective electrode potentiometry. Additionally, the K values have been determined in aqueous solution (/ = 0.05 M (CH3)4NC1O4 and 298.15 K). In a typical measurement, a solution containing 2 — 40x10 5 M fluoride was titrated with the solution of the given metal ion, to reach a final metal ion concentration of 2 - 80x10"4 M. This work was devoted to gaining further insight into the anomalous stability order of the monohalide complexes of 3d transition metal ions first reported in [72BON/HEF]. The results indicated that the monofluoride complexes are a clear and distinct exception to the Irving-Williams sequence; the stability of MF+ complexes was found to be Mn ~ Fe > Co > Ni < Cu > Zn in both water and methanol. This stability order does not follow even the trend predicted by ionic radii. Therefore the authors suggested that the monofluoride complexes have some characteristics of ion pairs rather than those of inner-sphere complexes. The thermodynamic data indicated that the complex (ion pair) formation reactions are all distinctly entropy controlled. [84BRA/DEL] Sodium nickelate, NaNiO2, was prepared by direct reaction of sodium oxide and nickel oxide at 600 — 800°C under a stream of oxygen. This compound was hydrolysed and reduced by so-called soft chemical reactions in aqueous media at room temperature. [84COM/PRA] The standard molar Gibbs energy of formation of NiO was determined at temperatures as low as 760 K up to 1275 K with an excellent accuracy of ± 0.039 kJmol ' (which is ± 0.2 mV for the standard deviation of the emf). The obtained results were represented by the equation: Af G° (7) = - 232.450 + 0.083435 (77K) kJ-mol"1. [84LAV/TIM] High-purity nickel (wimpurities < 6 xlO 5) was subjected to chlorination in a calorimetric bomb with a microfurnace for sample heating. The microfurnace developed a temperature of about 1070 K over approximately 10 min at a power input of 75 W. Under these conditions nickel chlorination proceeded practically to completion.
402
A. Discussion of selected references
[84ROB/HEM] Robie et al. have carried out excellent measurements of the heat capacity of Ni2SiO4 olivine between 5 and 387 K by cryogenic adiabatic-shield calorimetry and between 360 and 1000 K by differential scanning calorimetry. At 298.15 K the molar heat capacity and entropy of Ni 2 Si0 4 olivine were determined. This molar heat capacity and entropy were accepted by this review. The thermal heat capacity function proposed by Robie et al. for the temperature range between 300 and 1300 K was, however, not adopted by this review, because it is based on measurements which were made only at temperatures up to 1000 K. Combining the heat capacity measurements with results of molten salt calorimetry, thermal decomposition of Ni 2 Si0 4 olivine into its constituent oxides, and equilibrium studies, both by CO reduction and solid state electrochemical cell measurements for the reaction: 2Ni + SiO2 + O2 ^ Ni2SiO4, Robie et al. calculated for Ni 2 Si0 4 olivine AfG° (298.15 K) = - (1289.0 + 3.1) kJ-mol ' and A f //° (298.15 K) = - (1396.5 ± 3.0) kJ-mol '. [84ROG/BOR] The authors carried out emf-measurements using a galvanic cell with a Ni2+ exchanged P-alumina solid electrolyte to determine the Gibbs energy, enthalpy and entropy of formation of nickel orthosilicate from the constituent oxides. The Gibbs energy of the cell reaction as a function of temperature, given by Rog and Borchard for the temperature range 970 to 1370 K, coincides very well with results of other emf-studies of Ni 2 Si0 4 , as shown in this review. The estimated temperature of decomposition of Ni2SiO4 to NiO and SiO2, 1790 K, is significantly lower than the value recommended by this review (1820 ± 5) K. [84VAS/VAS] Vasil'ev et al. determined A r // m of Reaction (A.74) calorimetrically at 298.15 K. Metallic nickel was dissolved in 2.0 to 6.0 M HC1 or HC1O4 solutions containing 1.0 to 1.5%H2O2. Ni(cr) + 2H+ + H2O2(aq) ^ Ni2+ + 2H2O(1)
(A.74)
When the authors plotted the measured enthalpies minus the Debye-Hiickel term versus the ionic strength (which appears to be dominated by the acid concentration) the data fell on straight lines with slopes depending on the H2O2 content, but the intercepts coincided within the error limits. Thus, the results were extrapolated semiempirically to / = 0, resulting in Ar/7° (A.74) = - (432.59 + 0.43) kJ-mol '. As the SIT model cannot predict the H2O2 dependence of the slopes this evaluation has been accepted as is. With the standard enthalpies of A f //°(H 2 O, 1) = -(285.830 + 0.040) kJ-mol ' and A f //°(H 2 O 2 , aq) = - (191.170 ± 0.100) kJ-moP1 taken from CODATA
403
A. Discussion of selected references
and the NBS Tables, respectively, A f //°(Ni 2+ ) = -(52.10 ±0.45) kJ-mol ' has been obtained [82WAG/EVA], [89COX/WAG]. [85DRA/MAD] Apparent molar heat capacities were determined for nickel sulphate solutions at 50, 70 and 90°C. The lowest molalities used were 0.3 molkg ', and the extrapolation to / = 0 is problematic, especially if a model that invokes association between nickel and sulphate ions is to be used. The values from this study are not used in the present assessment. [85MEH/TAR] The partial pressure of oxygen for the solid phase equilibrium p2-Ni4S3 / NiO at pso = 1 atm was determined in the temperature range 973 — 1173 K by performing emf measurements on the electrochemical cell Pt, O2 (air) | YSZ | NiO, P2-Ni4S3, SO2 (1 atm), Pt or Au. Yttria-stabilised zirconia served as the solid electrolyte. The solid phases were prepared by solid state reactions from the elements. Phase identification was carried out by means of X-ray diffraction analysis and the composition of the solid nickel sulphide was checked by chemical analysis. The authors found a composition of Ni2.666S2- The variation of this composition with temperature was reported to be not appreciable, but was not checked carefully by the authors. Hence, the results of this work are apparently not suitable for the derivation of the Gibbs energy function for this phase. [85SKE/MAI] Skeaff et al. [85SKE/MAI] used an electrochemical technique to determine the Gibbs energies of reaction of NiAs(cr) with oxygen to form NiO (bunsenite) and As4O6(g) from 711 K to 833 K. Similar measurements were carried out with Ni| jAsg(cr) (805 K to 991 K) and Ni5As2(cr) (937 K to 1139K). Experiments with NiO-NiAs2 and NiAs-NiAs2 mixtures were unsuccessful in that there was no indication of any change in potential with changes in partial pressure of As4O6(g). Based on the approximation that ATCpm = 0 for xNiO(cr) + ^ As4O6(g) ^ NixAsy(cr) + 4
(2x + 3y)
4
O2(g)
the authors' estimated heat capacity equations (in J K '-mol 1 ) were: C°m (NiAs, cr)
= 63.7532 + 6.4616 *10 3 (77K) - 8.6016 * 105 (77K)2
C° m (NinAsg, cr) = 605.483 + 70.7846 xlO 3 (77K) - 66.3005 *105 (77K)"2 C; m (Ni5As2, cr) = 222.965 + 32.0067 x l O 3 (77K) - 14.6911 xlO5 (T/Ky2 and the following values were reported: Af/7° (NiAs, cr, 298.15 K) = - 73.29 kJmol ', S° (NiAs, cr, 298.15 K) = 45.4 .MC'-mol"1, A f //° (Ni,,As8, cr, 298.15 K) = -777.04 kJ-mol"1, 5° (NinAsg, cr, 298.15 K) = 468.9 Jlt'-mol 1 , A f //° (Ni5As2, cr, 298.15 K) = - 251.10 kJ-mol"1, and S° (Ni5As2, cr, 298.15 K) = 190.6 J K r ' m o l 1 .
404
A. Discussion of selected references
The authors used values for the heat capacity and entropy of As4C>6(g) from the analysis of Behrens and Rosenblatt [72BEH/ROS], and tied the value for Af//no, (As4O6) to the vapour pressure measurements of [72BEH/ROS] and Murray et al. [74MUR/POT]. Behrens and Rosenblatt, in turn, relied on values calculated from the thorough sets of low-temperature heat capacity measurements of Chang and Bestul [71CHA/BES] for the As4O6 solids. As was discussed by Grenthe et al. [92GRE/FUG], auxiliary data for arsenic compounds have not been critically reassessed in the TDB Project, and at 298.15 K several of the self-consistent values from the U.S. National Bureau of Standards compilation [82WAG/EVA] are accepted as an interim measure. The values A f //°(As 4 0 6 , g, 298.15 K) = -(1196.2 ± 16.0) kJ-mol ' and S°(As 4 O 6 , g, 298.15 K) = (408.6 + 6.0) J-K. '-mop' were obtained by adjustment of the corresponding values reported by Behrens and Rosenblatt [72BEH/ROS], and have been used in the present review. The reported reaction enthalpies and entropies for Reactions (A.75), (A.76), (A.77) are listed in Table A-21 NiO(cr) + i As4O6(g) ^ NiAs(cr) + | O2(g)
(A.75)
11 NiO(cr) + 2 As4O6(g) ^ Ni,,As8(cr) + 2102(g)
(A.76)
5 NiO(cr) + I As4O6(g) ^ Ni5As2(cr) + 4 O2(g)
(A.77)
Table A-21: Enthalpies and entropies of reaction from Skeaff et al. [85SKE/MAI]. arsenide
c
T,* (kJ-mol"')
(J-K ' mor')
(K.)
NiAs(cr)
465.458
159.057
772
Nii]As8(cr)
4254.395
1564.75
800
Ni5As2(cr)
1544.845
604.452
1038
approximate mid-temperature for the measurements
Equations for C° m (NiAs) and C (NinAsg) from work of Muldagalieva et al. [93MUL/ISA], [95MUL/CHU] are used with auxiliary data for As4O6(g) (see Chapter IV), O2(g) [98CHA], and values for NiO(cr) from the present review to obtain A f //°(NiAs, cr, 298.15 K) = -(69.73 + 5.00) kJ-mol ', 5° (NiAs, cr, 298.15 K) = (50.76 ±6.00) J-K •"'•mor1, A f //° (Ni n As 8 , cr, 298.15 K) = - (743.0 ± 35.0) kJ-mol 1 and 5° (Ni n As 8 , cr, 298.15 K) = (518 ± 60) J-K '-mol '. In the absence of the original data except as points on a graph, but with the reported uncertainties in the values for ArGnl from the equations, the uncertainties here are estimated by the reviewer. For both solids, the heat capacity equations are used slightly beyond the range suggested by the original authors, but the Cpm{T) curves appear to be monotonic, and there are no better
A. Discussion of selected references
405
data. For NisAs2(cr), no heat capacity equation was available, and based on the approximation ATCp m = 0 used by the original authors, and auxiliary data from the present review, A f ^ ( N i 5 A s 2 , cr, 298.15 K) = - (244.6 + 20.0) kJmol 1 , and S°(Ni5As2, cr, 298.15 K) = (196.2 + 30.0) J K ' - m o l 1 is obtained. The uncertainties have been increased to account for the lack of proper heat capacity data. [86BHA/MUK] Bhattacharya et al. report the electrochemical generation of the first discrete NiinO6 species, Ni(bpyO 2 )j + . The cation is low spin and rhombic. This paper is interesting in the present context, because the redox potential of Ni(bpyO2 )\+ | Ni(bpyO2),+ can be measured and correlated to the redox potential of Ni3+ | Ni2+. [86CEM/KLE] The standard enthalpies of formation of various phases in the system Ni - S were determined by means by drop calorimetry. Synthetic NiS was prepared by solid state reaction from the elements. X-ray diffraction and microprobe analysis revealed that the high-temperature modification of NiS was obtained. The minerals vaesite (NiS2) and heazlewoodite (Ni3S2) were synthesised by reaction of NiS with sulphur and metallic nickel, respectively. All samples were proved to be well-crystallised material. The enthalpy of formation for the stoichiometric composition of the hexagonal NiAs-type nickel monosulphide, NiS(a), at 298.15 K was determined by dropping evacuated silica ampoules containing a stoichiometric mixture of powdered metallic nickel and sulphur from room temperature (298.15 K) into the hot calorimeter (1021 K) where the formation of NiS occurred. In a second experiment the silica ampoules containing NiS formed during the first experiment were dropped from room temperature into the hot calorimeter. Obviously, the difference of the enthalpy effects corresponding to the dropping experiments directly yields the enthalpy of formation of the high-temperature form of NiS at 298.15 K, A f #° = -(88.1 ± 1.0) kJmol'.The enthalpy of formation of the lowtemperature modification, NiS((3), cannot be obtained directly because in the high temperature calorimeter NiS(a) always is formed, and this solid does not transform into NiS(|3) upon quenching. The enthalpy of formation of heazlewoodite was obtained by dropping evacuated silica capsules containing a mixture of powdered metallic nickel and NiS from room temperature into the calorimeter at 873 K. Again, in a second experiment the silica capsules containing the product of the first reaction (N3S2) were dropped into the calorimeter. Taking into account the enthalpy of formation of NiS determined earlier, the difference of the enthalpy changes associated with these experiments leads to the enthalpy of formation of heazlewoodite at 298.15 K, Af H°m = - (217.24 ± 1.60) kJmol '. The enthalpy of formation of vaesite (NiS2) at 298.15 K was derived from the reaction:
406
A. Discussion of selected references
NiS2 + Ni ^
2NiS(a)
at 873 K in the calorimeter. In a second experiment the heat content of the silica ampoule and NiS formed during the reaction was measured. The difference between the enthalpy changes associated with these experiments results in the enthalpy of formation ofvaesite at 298.15 K, A f //° = -(124.9 + 1.0)kJmol '. The enthalpy of formation for Ni7S6 was likewise determined by reaction of NiS with Ni at 833 K in the calorimeter. At these conditions the non-stoichiometric high-temperature modification of Ni7S6 is formed. However, when this product is quenched, metastable modifications are obtained depending on the cooling rate [96SEI/FJE]. Therefore, the enthalpy of formation reported by the authors instead should be assigned to formation of metastable phases of Ni7S6 or to a mixture of Ni3S2 and Ni9S8. All experimental results of this interesting paper are listed in Table A-22. Table A-22: Enthalpies of formation for solid nickel sulphides at 298.15 K. Solid phase
A r //° (kJ-mol1)
NiS(a), high-temperature form
— (88.1 ± 1.0)
Ni,S2
-(217.2+1.6)
NiS2
-(124.9 ±1.0) 2
Ni7S6, maybe !/jNi3S2 + /3Ni9Ss
- (582.8 + 5.7)
[86HOL/NEI] The standard Gibbs energy of formation of NiO has been experimentally determined over the temperature range from 900 to 1400 K using a galvanic cell with the solid electrolyte made of 15% calcia-stabilised zirconia. The measured value of AfG°(NiO) at 1300 K was - 123.555 kJmol ' with a precision of+0.057 kJ-mol ' and an estimated accuracy not worse than 0.200 kJ-mor1. This precision is equivalent to an error of only 0.2 — 0.3 mV in the cell potential (emf). In comparison, most previous studies have reported a precision of 1 - 2 mV. Using the third law analysis, the authors obtained for the enthalpy of formation of NiO, A f //° (NiO, 298.15 K) = - 240.110 kJmol '. [86HSI/CHA] Equilibria between two solid phases in the ternary system Ni - S - O at p&Q = 1 atm were studied by using the electrochemical cell Pt, O2(air) | YSZ or CSZ | mixture of phases, SO2 (1 atm), Au, where yttria- or calcia-stabilised zirconia served as the solid electrolyte. The following phase equilibria were studied: NiS(ot) / p2-Ni4S3; p2-Ni4S3 / NiO; NiS(a) / NiO; and NiO / NiSO4. While the solid sulphides and NiO were prepared by solid state reaction from the elements, NiSO4 was synthesised by dehydrating NiSO4-6H2O. All solid phases were identified by X-ray diffraction analysis. The emf of
A. Discussion of selected references
407
the electrochemical cell, recorded as a function of temperature, corresponds directly to the partial pressure of oxygen at p = 1 atm for the phase equilibria mentioned above. Figure A-33 shows the temperature dependence of the oxygen partial pressure for NiS(cc) / p2-Ni4S3 as well as p2-Ni4S3 / NiO, where the experimental results of [81SCH], [85MEH/TAR] are likewise included. It can be deduced from Figure A-33 that the thermodynamic models of both [80SHA/CHA] and the present data assessment are in close agreement with the experimental results for the phase equilibrium NiS(a) / P2-Ni4S3. The experimental data from various literature sources [81SCH], [85MEH/TAR], [86HSI/CHA] for the equilibrium p2-Ni4S3 / NiO coincide remarkably well with each other despite the fact that Mehrotra et al. [85MEH/TAR] and Schaefer [81SCH] gave constant but different values for the composition of the nonstoichiometric phase p2-Ni4S3. However, it is to be expected that the composition of P2Ni4S3, which is fixed by the phase equilibrium with NiO at psQ = 1 atm, should change with temperature. Moreover, it is worth mentioning that the temperature dependence of the sulphur content in p2-Ni4S3 in equilibrium with NiO has not yet been investigated experimentally. In addition, the variation of the oxygen partial pressure with temperature for the equilibrium NiS(a) / NiO at pso = 1 atm is depicted in Figure A-34. The experimental data agree reasonably with equilibrium calculations using the present thermodynamic model. Figure A-33: Plot of logarithm of the oxygen partial pressure versus temperature for the phase equilibria a-NiS / p2-Ni4S3 and p2-Ni4S3 / NiO, respectively, at p&0 = 1 atm. •, • [86HSI/CHA]; A [81SCH]; dotted line: [85MEH/TAR]. Dashed line: thermodynamic model according to [80SHA/CHA]; solid line: thermodynamic model of the present assessment.
408
A. Discussion of selected references
Figure A-34: Logarithm of the oxygen partial pressure plotted as a function of temperature for the equilibrium a-NiS / NiO at /?(SO2) = 1 ami. • [86HS1/CHA]. Solid line: thermodynamic model of the present assessment.
[86JAC/KAL] The emf-study of Ni2Si04 done by Jacob et al. was not used for the evaluation of thermodynamic data by this review, because the results were presented only in graphical form, making accurate recovery of experimental values difficult. [86VAS/DMI] Vasil'ev et al. reported another value of Af H°m (Ni2+) = -(51.83 + 0.88) kJmol ' determined calorimetrically by i) dissolving Ni(cr) in solutions containing 0.3 M Br2, 0.05 M HC1O4, 0.95 to 3.95 M NaBr, ii) accounting for the enthalpy of mixing NiBr2 as well as iii) Br2 to make up the final solutions, and iv) extrapolating to / = 0. The re-calculation with the CODATA value for A f //°(Br) = -(121.41 ±0.15) kJ-mol 1 resulted in a minute correction only, but the error limits appear too optimistic in view of four different contributions to the final result, and thus were set to 2a = ±2.0 kJmol 1 . Thus, A f //° (Ni2+) = - (51.85 ± 2.00) kJ-moP1 has been assigned to this experimental information.
A. Discussion of selected references
409
[87BEC] The formation constants of the proton and metal ion complexes of cyanide ion have been critically surveyed. The recommended value for pK°1CN is correct, but the tentative values for higher ionic strengths seem to be erroneous [92BAN/BLI]. Several literature values for the formation constants of Ni(CN)4~ and Ni(CN)4~ as well as reaction enthalpies for the formation of Ni(CN)^~ were reported, but no recommended values were given. In addition, the formation constants of several NiX(CN)x complexes (X = multidentate organic ligands, x = 1 or 2) were listed. [87BJE] These "semi-thermodynamic" formation constants are not equivalent to the value determined by extrapolation to zero ionic strength using the SIT, except for the case when the ratio: Y[NiX,,(H2O)6_n] ^[NiX,,.|(H 2 O) 7 _J~ *
in the whole range of the ionic strength studied. [87CEM/KLE] The standard enthalpy of formation of synthetic pentlandite, Fe4 5Ni4 5 S 8 , and natural violarite, (Feo.294iNio.7059)384, were measured by solution drop calorimetry. Synthetic pentlandite was prepared from the elements and checked by X-ray diffraction analysis. First a stoichiometric mixture of FeS, NiS, Ni, and Fe was dropped from 298.15 K into a melt of Ni0.6S0.4 at 1100 K. In a second experiment synthetic pentlandite was dropped into the Nio.6So.4 melt. The enthalpy of formation of pentlandite from its elementary and binary compounds is obtained from the difference of the enthalpy effects associated with these two experiments, &rH°m = - (78.41 ± 14.45) kJ-mol ' for: 4 NiS + 4 FeS + 0.5 Ni + 0.5 Fe ^ Ni4.5Fe4.5Sg. Taking A f //° = - (101.67 ± 0.20) kJ-moP1 for the standard enthalpy of formation of FeS from the JANAF tables [98CHA] and A f //°= - (88.1 ± 1.0) kJmol ' for the high-temperature modification of NiS [86CEM/KLE], the standard enthalpy of formation of pentlandite at 298.15 K is calculated to be A f //° = - (837 + 15) kJmol '. By using a similar procedure, the authors arrive at A f //°= — (378 + 12) kJmol 1 in the case of violarite with the composition (Feo.294iNio.7O59)3S4. [87EMA/FAR] Emara et al. gave an example of how they made up their solutions. For a solution intended to contain 2.4xlO'4 M HCO3" they started with 6.18> P) = (54.008 - 0.01437 (77K) + 3.2282 x 10 5 (77K)2 -490363 (f/ICrV-tC'-mor 1 . (A.102)
Figure A-38 shows the temperature dependence of the heat capacity of NiS, where the dashed line corresponds to the metastable NiAs-type phase below 660 K. The steep increase of Cp m at temperatures above 900 K can be attributed to the formation of small traces of Ni3+xS2.
A. Discussion of selected references
429
Figure A-38: Heat capacity of NiS as a function of temperature, o, solid lines: stable NiS phases. A, dashed line: metastable NiAs-type modification below 660 K.
[95MAN/KOR] Manin and Korolev reported having measured the enthalpy of solution of NiC^Ccr) into water and into aqueous solutions of other electrolytes. The measurements were performed with an isoperibol calorimeter with a claimed accuracy of ± 0.6%. Their results for the enthalpy of solution into H2O are listed in Table A-25. The mean value of A sol //° for the reaction: N i C l 2 ( c r ) ••
- Ni2+ + 2 Cl
(A. 103)
agrees nicely with the one derived from Muldrow's data. According to the Cl content (54.0%) determined by argentometric titration, the sample of NiCl2 used was not completely anhydrous («(H2O) / «(NiCl2) ~ 0.09). This may lead to systematic errors similar to those discussed in the Appendix A entry for [58MUL].
430
A. Discussion of selected references
Table A-25: Enthalpy of solution of NiCl2 in H2O at 298.15 K [95MAN/KOR]. m (NiCl2)
Asol//m(A.103) 1
*I
ABl//°(A.103)
(kgmol )
(kJ-mor )
(kJ-mol )
(kJmol •')
0.004230
- 82.67
0.5961
-83.2661
0.004570
- 82.46
0.6169
- 83.0769
0.006320
- 82.92
0.7111
-83.6311
0.009250
- 83.08
0.8371
-83.9171
0.009260
- 82.83
0.8375
- 83.6675
0.009630
- 82.87
0.8515
-83.7215
0.012410
-82.10
0.9465
- 83.0465
0.014380
- 82.63
1.0054
-83.6354
0.027340
-82.10
1.2957
-83.3957
0.030460
-82.14
1.3499
- 83.4899
1
1
The variation of the enthalpy of solution of the anhydrous solid should parallel the enthalpy of solution of the hydrated solid as a function of salt molality. The experimental values for dissolution in water drift to slightly more negative values at lower NiCl2(aq) concentrations, but the measurements themselves have fairly large uncertainties of approximately ± 0.5 kJ-mol '. The solution concentrations are not sufficiently low to provide a good extrapolation to / = 0, but values based on a theoretical extrapolation [97GRE/PLY2] are provided in Table A-25. The authors' Figure 1 compares the values for the heats of solution in water to those taken from two compilations. However, there is no evidence as to how the values from the compilations were based on experimental heats of dilution. (There is also a typographical error in the key to the points in the authors' Figure 1. The points designated " 3 " appear to be the data from [95MAN/KOR]). The heats of dissolution of anhydrous nickel chloride in aqueous solutions containing several different concentrations of HCl(aq) also were determined. For measurements with final solutions having nickel chloride molalities of ca. 0.007 m, the enthalpy of solution of anhydrous nickel chloride for each mol% of HC1 is about 3 kJmol ' more positive than in water. [95MUL/CHU] This paper is similar to the earlier paper on Ni n As 8 (cr) [93MUL/ISA], but contains more details. The heat capacity equation, from dynamic calorimetric measurements, is reported as: [CU%H (NiAs, cr) = (36.54 + 41.87x10 3(77K) + 1.58xlO5 (77K) 2) J K 'mol '.
A. Discussion of selected references
431
[96BRU/PIA] The authors reported an extensive set of torsion balance and Knudsen cell measurements for the determination of the enthalpy of sublimation of NiF2 between 950 and 1250 K. Third law and second law values are in much better agreement if the values of (G m (7')-//°(298.15 K))/T for the sublimation reaction for CoF2(cr) are substituted for those in the literature for NiF2(cr). The work appears to have been carefully carried out, and if the interpretation were correct, this would indicate that (a) 5° (Ni, cr, 298.15 K) is 8 - 11 J K '-mof1 less than based on [55CAT/STO], or (b) Cpm(T) from [70BIN/HEB] is incorrect (although the values used for C0F2 are from work by the same group [67BIN/STR]) or (c) the value of A f //°(NiF 2 , cr, 298.15 K) based on Rudzitis et al. [67RUD/DEV] is too negative by about 8 kJ-mol"1. [96KEH] The table presented gives second virial coefficients of about 110 inorganic and organic gases as a function of temperature. Selected data from the literature have been fitted by least squares to the equation:
B (cm3-mor') = |>(i)[(r o IT)-\fx where To = 298.15 K. For the present purpose, use has been made of the a(i) values of CO2. [96LUT/RIC] The stability constants for the aqueous species NiHS+, Ni2HS3+ and Ni3HS5+ at 298.15 K in seawater as well as sodium chloride solutions (artificial seawater) have been determined by application of square wave voltammetry. Solutions of sodium sulphide ( 1 - 1 0 umol-kg"1) in diluted seawater of constant ionic strength ranging from 0.07 to 0.7 mol-kg"1 NaCl were titrated with nickel (II) solutions, where care was taken to remove traces of oxygen from all solutions by purging with high purity argon. The molality of free sulphide was determined from both the peak potential and the peak current at constant pH for the electrochemical reaction: HS + Hg ^ HgS + H+ + 2e occurring at the hanging mercury drop electrode. The titration curves were evaluated by means of a standard method [51DEF/HUM], [77HEA/HEF], leading to stability constants for NiHS+, Ni2HS3+ and Ni3HS5+, respectively. Plots of the peak potential versus pH indicate the existence of bisulfide (HS ) complexes at pH values above 7. The ionic strength dependence of the nickel bisulphide stability constants is depicted in Figure A-39. By applying the SIT concept [97GRE/PLY2] the respective stability constants are extrapolated to zero ionic strength. As well as the stability constants, the optimisation procedure also yields the SIT interaction parameters Ae in chloride media, T= 298.15 K:
432
A. Discussion of selected references
Ni2+ + HS ^ NiHS+ log10 ]3° = (5.18 + 0.20); AE(A. 104) = - (0.97 ± 0.39) kgmol 1 .
(A. 104)
2Ni2+ + HS ^ Ni2HS3+ log10 P° = (9.92 + 0.10); Ae(A.105) = - (0.05 ± 0.22) kg-mol1.
(A. 105)
3Ni 2+ + HS ^ Ni3HS5+ (A. 106) logl0 j3° = (14.008 + 0.099); Ae(A.106) = (0.59 ± 0.22) kg-mol '.
Figure A-39: Logarithm of stability constants of nickel bisulphide complexes in seawater (NaCl solutions) plus the Debye-Hiickel term for ionic strength correction plotted as a function of ionic strength, T= 298.15 K. The solid lines correspond to least squares fits of the SIT model to the experimental data, • [96LUT/RIC], • [94ZHA/MIL]. (a) log10 J3° = (5.18 + 0.20);
Ae(A.104) = - (0.97 ± 0.39) kg-moP1.
(b) logl0 J3° = (9.92 ± 0.10);
Ae(A.105) = - (0.05 ± 0.22) kg-mol '.
(c) log10 J5l = (14.008 ± 0.099); Ae(A.106) = (0.59 ± 0.22) kg-mol.
A. Discussion of selected references
433
[96POU/DRE] Nickel hydroxide purchased from Johnson Matthey Chemical Co. was equilibrated with 0.002 to 0.200 mol-kg"1 NaNO3 over a pH range of approximately 5.5 - 6.5 at 22°C. The experimental data were evaluated using the Davies equation in MINTEQA2 at these ionic strengths [93ALL/BRO]. The solubilities obtained a t / = 0.2 molkg ' were apparently lower than those at / = 0.002 - 0.02 mol-kg"1. The authors preferred the higher values, because their model might become incorrect at / > 0.02 molkg '. When the activity data were recalculated assuming that Equation (A. 107) is valid a mean value of log10 K°o = - (15.96 ±0.10) was obtained at 295.15 K. log10 *K°_0= 12.24 recorded in Table V-6 and Figure V-l 1 was based on this evaluation. log]0 am2t = log10 K°,fi - 2 log10 am_
(A. 107)
[96SAH/SAH] This paper reports log10 Kx = (2.20 ± 0.02) for the association constant of Ni2+ with HPO4" at 25°C in 0.1 M NaNO3. The study appears to have been done carefully, though several of the minor difficulties found for an earlier paper [67SIG/BEC] from the same group still occur (see Appendix A for [67SIG/BEC]). Correction to zero ionic strength (Appendix B) gives \ogmK° = (3.07 + 0.10), where the uncertainty has been assigned in the present review. The small As correction has been based on the value for e(Na+, HPO^") from Table B-5 (rather than using the equation in Table B-6), and e(Ni2+, NO 3 ) = 0.18kg-mol '. [96SEI/FJE] The relative stability and structural descriptions of four metastable phases with composition close to Ni7S6 were obtained from X-ray diffraction, transmission electron microscopy, calorimetry, and thermal analysis. The structures of these phases are closely related to that of the disordered high-temperature phase Ni7S6 which is formed eutectoidally from Ni3S2 and Ni9S8 at 675 K [94STO/FJE]. When Ni7S6 is cooled to room temperature, this phase decomposes into the stable compounds Ni3S2 and NigSg or a transformation of Ni7S6 into metastable modifications of similar composition occurs. [97BAH/PAR] This is a recent compilation of thermodynamic data of metal thiocyanate complexes. The paper listed the majority of available thermodynamic data for the nickel(II) - thiocyanate system, but recommended values are given only for log10 /?, at / = 1.0 (log I0 fi = (1.14 ± 0.09)) and 1.5 M (log10 # = (1.14 ± 0.02)). [97BAL/DIV] Revised E—pH and E-m(KOH) diagrams for nickel at 25°C are calculated and graphically represented using the thermodynamic data derived for hypothetical NiO2-xH2O
434
A. Discussion of selected references
[93BAL/DIV], [93BAL/DIV2]. This E-pH diagram is meant to supersede previous versions [63DEL/ZOU], [81SIL]. [97MAT/RAI] Nickel hydroxide was precipitated by adding NaOH to a 0.2 M NiCl2 stock solution. The final pH of the Ni(OH)2 suspension was approximately seven. The precipitate was washed with water until it was free of chloride. The X-ray diffraction data of the freeze dried precipitate matched with JCPDS data file 14-117 for the Ni(OH)2 mineral, theophrastite. Approximately 300 mg of Ni(0H) 2 were suspended in 30 cm3 of 0.01 M NaClO4. The pH was adjusted between 7 and 12 by HC1O4 and NaOH, respectively. Thus, only a minute Ni(OH)2 portion, doubtlessly having consisted of the finest particles, was actually dissolved. When equilibrium was approached from oversaturation again finely dispersed Ni(OH)2 will have been precipitated. As solubilities of metal oxides and hydroxides depend on the particle size [65SCH/ALT], the numerical value of the solubility constant will probably be overestimated by this method, see Figure V-l 1. [97PAN/CAM] This is probably the most comprehensive set of heat capacity results available for any nickel salt in aqueous solution. Apparent molar heat capacities of aqueous Ni(C104)2 were measured calorimetrically from 25 to 85°C over a molality range of 0.02 to 0.80 molkg '. Standard molar heat capacities of Ni2+ for the same temperature range were obtained by using the additivity rule and data for HClO4(aq), given in literature. The results for C° m (Ni2+) can be fitted with a conventional heat capacity model valid from 298.15 to 358J5 Kusing C° m (HC10 4 , aq, T) from Lemire etal. [96LEM/CAM]: [C;m]3254;!5K(Ni2+) = (842.427- 1.8166 (77K)-3.06653 x 107/(77K)2) J-K 'mol '. The values change only slowly with temperature in this range, with an apparent shallow maximum value near 50°C. The authors extrapolated their results to 300°C using the Helgeson-Kirkham-Flowers equation [74HEL/KIR2], [81HEL/KIR], [97PAN/CAM]. The mean value for the standard partial molar heat capacity of Ni2+ at 25°C in the authors' Table III is - 44.0 kJ-mol ' whereas the result in the authors' Table IV is C; m (Ni 2+ , 298.15 K) = - 49.6 J-K 'mol '. The difference reflects the sensitivity of the C° m value to the use of two different, but reasonable, auxiliary values for the partial molar heat capacity for HClO4(aq). As noted in Section V.2.1.4, results for other nickel salt solutions lead to similar uncertainties in the value of C° m (Ni 2+ , 298.15 K). [97RIC] This monograph describes the synthesis, structure and reactivity of aqua ions. Only a short paragraph deals with Ni3+, because this ion occurs only as a 'transient' species.
435
A. Discussion of selected references
[98GAM/KON] A plot of log 10 {[Ni 2+ ]xp co } versus pH showed that the corresponding solubility constant is almost independent of temperature at least in the range of 348.15 to 363.15 K. Apparently other neutral transition metal carbonates behave similarly. [98JAI/ELM] The redox reaction of nickel hydroxide and nickel oxide hydroxide, the electrochemically active compounds at the positive electrode of a nickel battery, was investigated. The thermodynamics of non-ideal solid solutions were applied to the reversible potential as a function of the state-of-discharge. In a temperature range 5 to 55°C two parameter activity coefficient models perform significantly better than one parameter models. [98PLY/ZHA] Plyasunova et al. critically evaluated the standard thermodynamic quantities of Ni2+, its hydrolysis reactions and hydroxo-complex formation on the basis of published experimental studies and the specific interaction theory (SIT) for activity coefficient modelling [97GRE/PLY2]. Recommended thermodynamic functions and interaction coefficients relevant for the present review and its compounds were presented, see Table A26. Table A-26: Recommended formation constants ( J3°m ) and reaction enthalpies for the hydroxo complexes formed in the acidic region, at 298.15 K, mW\2+ + «H2O(1) ^ Ni m (OH)f'" +«H + Species
log,, 'PL
NiOH"
\Hl (kJmol"')
-(9.50 ±0.36)
(50 ±21)
Ni2OH3t
-(9.8 ±1.2)
(43.3 ± 3.0)
Ni4(OH):+
-(27.9± 1.0)
(180 ±2)
Table A-27: Recommended ecluilibrium constants ( t'", m ) and reaction enthalpies for the hydroxo complexes forme;d in the alkaline region, at 298.15 K, wNi(OH)2(cr) + «OH ^ Nim(OH)":+n log,. X;,»,
Ar//° (kj-mol"')
- (7.52 ± 0.80)
-
Ni(OH);
-(3.7± 1.8)
(7.1 +3.0)
Ni(OH);-
(6.43 + 0.23)
-(14 + 8)
Species Ni(OH)2(aq)
436
A. Discussion of selected references
Table A-28: Recommended ion interaction coefficients e(/, k).
cr
C1O4
Na+
> 0.31
(0.30 + 0.18)
-
(0.41 ±0.56)
(0.59 + 0.16)
-
(0.68 ±0.38)
(0.90 + 0.32)
-
Ni(OH) 2(aq)
-
-
- (0.07 ± 0.03)
Ni(OH)3
-
Ni(OH) 4
-
k
i
NiOH + Ni2OH
3+
Ni4(OH)4*
(0.18 ±0.05) (0.27 ± 0.06)
-
Table A-29: Standard thermodynamic functions of the aqueous species. AfG° (kJ-mol-')
A f //; (kJmol ')
^(J-K'-mol1)
Ni2+
- ( 4 5 . 5 ±3.4)
-(54.1 +2.5)
- ( 1 3 0 + 3)
NiOH +
-(228.4 ±5.7)
- (290 ± 24)
- ( 7 4 ±81)
- ( 2 7 2 + 14)
- (359 + 20)
- (260 ± 86)
Ni 4 (OH);
-(971 ±19)
-(1190 ±27)
- (203 ± 90)
Ni(OH), (aq)
- ( 4 1 7 + 11)
-(540+13)
- (48 ± 32)
Ni(OH) 3
- ( 5 8 7 + 15)
-(791 + 18)
- (85 + 47)
Ni(OH) I
- (734.0 + 8.6)
-
-
Species
Ni 2 OH
3f 4> 4
The analysis was useful in considering some of the references, but the values from [98PLY/ZHA] are not used in the present review. [98ROG/KOZ] Exchange of sodium ions in P"-alumina, performed in molten salts, made possible a preparation of a series of divalent P' '-aluminas. In this way the solid electrolytes conducting zinc, cobalt, nickel, calcium, manganese and copper ions were prepared. Their application to the solid-state galvanic cells have been presented. The Gibbs energy of formation of nickel orthosilicate, from oxides, determined by [98ROG/KOZ] in the range 973 - 1323 K was used by this review to fit best experimental values in the temperature range 970 - 1770 K. [99 ARC] Archer detailed the sources of previous thermodynamic property values for nickel and some of its compounds being of importance to environmental processes. This review is particularly relevant to the origin of values contained in the NBS Thermodynamic Tables [82 WAG/EVA].
A. Discussion of selected references
437
[99KON/KON] Konigsberger et al. presented a low-temperature thermodynamic model for the Na2CO3-MgCC>3—CaCO3-H2O system. The model was based on calorimetrically determined A f tf°(298 K) values, S° (298 K) values and C°pm(T) functions taken from the literature as well as on yi° (298 K) values of solids derived from solubility measurements. [99MAL/CAR] The activity coefficients of (among others) aqueous Ni(C104)2 solutions (mmaOth = 7.3x10 5 - 0.89 mol-kg ') have been determined from the emf of liquid-membrane cells. For the purposes of this review, these data, together with the activity and osmotic coefficients listed in [69LIB/SAD], have been used to derive the ion interaction coefficient e(Ni2+, C1O4). The selected value is s(Ni2+, ClO^) = (0.37 ± 0.03) kg-mol ' (see Section V.4.3). Ciavatta [80CIA] proposed to use ionic strength dependent ion interaction coefficients if the uncertainty is ± 0.03 or greater (Appendix B, Section B.3). The above uncertainty is at the limit. Accepting Ciavatta's proposal e(Ni2+, CIO4) = Ei + E2 logio/m, where e, = (0.287 ± 0.02) kg-mol ' , e2 = (0.121 ± 0.03) kg-mol1 can be calculated. However, these values provide a better fit to the experimental data only for m > 2.5 molkg ' (/ > 7.5 mol-kg '), and lead to negative ion interaction coefficients for m < 0.0015 mol-kg '. Therefore, the use of an ionic strength dependent e(Ni2+, CIO4) was rejected in this review. [99MCB] McBreen comprehensively reviewed nickel hydroxide battery electrodes, the solid state chemistry of nickel hydroxides, and the electrochemical reactions of the Ni(OH)21 NiOOH couple. Any critical discussion of the thermodynamic data of nickel oxide hydroxides with higher oxidation states has to refer to this splendidly written account of nickel solid state electrochemistry. [2000ROG/KOZ] This paper of Rog et al. is a description of emf-measurements of Ni2SiO4 and Co2SiC>4 with a solid electrolyte. A calcium fluoride-based composite containing SiO2 was applied as a solid electrolyte. The results obtained in the temperature range 823 to 1273 K coincide perfectly with results of other emf-studies of Ni 2 Si0 4 carried out by Rog et al. [84ROG/BOR], [98ROG/KOZ] and also by other authors quoted in this review. [2000TSI/MOL] The main object of this conductance study was to obtain information on changes in association at a function of solvent composition. However, the value for the nickel sulphate association constant in water (calculated from the data using the Lee-Wheaton equation) at 20°C is reported. For the purposes of the present review, the data were re-
438
A. Discussion of selected references
analysed with the distance parameter fixed to the value consistent with the SIT procedure described in Appendix B. The recalculated value of K\ is (118.6 ± 7.2) compared to (139 + 1) reported in the original paper. This value is markedly lower than the results from other studies [73KAT], [76SHI/TSU], [79FIS/FOX]. [2001GAM/PRE] In order to synthesise hellyerite, Rossetti-Francois' method was applied and an aqueous solution of nickel chloride was reacted with ammonium carbonate. Fragmented crystals (d< 2um) and amorphous particles were precipitated [52ROS]. Therefore a new method was developed. The initially precipitated amorphous particles were dissolved and larger crystals (d = 0.05 mm) obtained when Ni2+ was kept in excess and the solution was slowly acidified by CO2. The X-ray data of hellyerite prepared by this new method agreed satisfactorily with the data given in the JCPDS-ICDD card 12- 276. The aqueous solubility of hellyerite, then believed to have the stoichiometric formula NiCGy6H2O, was studied at varying temperatures and constant ionic strength (/ = 1.0 mol-kg '). The crystals used were X-rayed before and after the dissolution experiment, but each time only the characteristic peaks of hellyerite were found. Data of log 10 {[Ni 2+ ]xp co } plotted versus pH fell on straight lines with slopes of approximately - 2.0 for all temperatures. From the experimental data obtained a preliminary set of the thermodynamic quantities Af G°, Af //° and S° for hellyerite was derived using the ChemSage optimiser routine [95KON/ERI]. [2001 HEN/RED] This is a study of the non-convergent ordering of Mg and Ni in synthetic olivines by means of the neutron powder diffraction and X-ray absorption spectroscopy (EXAFS). The implications of the intracristalline ordering to olivine crystal chemistry and their application to the consideration of thermodynamic relations, geothermometry, geospeedometry, and minor or trace element distribution control is discussed in detail. [2001MOL/TSI] The same data for nickel sulphate are reported as in [2000TSI/MOL]. However, values for the specific conductance are also reported. [2002GAM/WAL] The re-evaluation of AfG° and A f #° of P-Ni(OH)2 discussed in Section V.3.2.2.3. has been based on the results of this paper. Theophrastite, P-Ni(OH)2, was synthesised by an improved method based on the hydrolysis of Na2[Ni(OH)4]. Pure single crystals were obtained with sizes up to 0.25 mm [2002WAL/GAT]. The solubility of P-Ni(OH)2 was measured at different
A. Discussion of selected references
439
temperatures and ionic strengths, and the data observed were thermodynamically analysed to obtain the respective standard molar quantities of formation A,G,° and A{H^ . For parameter optimisation non-linear least squares routines were used. The tendency of precipitated Ni(OH)2 to form larger, well-shaped crystallites is extraordinarily small (see Figure 13 of [77OSW/ASP]). Consequently it seemed doubtful that solubility equilibria according to the following equations p-Ni(OH)2(cr) ^ Ni2+ + 2 OH P-Ni(OH)2(cr) + 2 H V N i 2 + + 2 H2O(1) can be approached both from over- and undersaturation. Thus the method of pHvariation (i.e., the initial acidity was varied from 0.005 to 0.100 molkg ' HC1O4) was employed [63SCH], but met in the case of P-Ni(OH)2 with two difficulties: •
The equilibrium pH in the neutral range is poorly buffered by solute species.
•
At 25°C Ni(OH)2 is notoriously inert, thus the solid-solute buffer system doesn't work either.
However, when the experiments were carried out in a temperature range between 35 and 80°C constant pH values indicating equilibrium were attained within periods from 3 weeks to as little as 48 hours. In a first series of solubility experiments on theophrastite the temperatures were varied, and the ionic strength was kept constant at 1.0 mol-kg ' NaClCv Figure A40 shows results typical for the pH-variation method. Data of logi0[Ni2+],ol plotted versus pH fall on straight lines with the theoretical slope of- 2.0. At 50°C a second series of solubility measurements was carried out at different ionic strengths varying from 0.5 to 3.0 mol-kg ' NaCIO,). Solubility constants were extrapolated to infinite dilution using the SIT [97GRE/PLY2]. Figure A-41 shows that the experimental data of P-Ni(OH)2 coincide reasonably well with the optimised curve obtained when the SIT equation is employed.
440
A. Discussion of selected references
Figure A-40: Solubility of theophrastite as determined by the pH variation method; t(°C); 80 A, 70 n, 60 T, 50 o, 35 •; solid straight lines; calculated with mean values of log10 *Ks0 and theoretical slope = - 2.0; / = 1.0 mol-kg"1 (Na)C104.
Figure A-41: Ionic strength dependence of theophrastite solubility. Inert electrolyte: NaC104. / (mol-kg '): 0.5 •, 1.0 •, 2.0 •, 3.0 A, + mean value and error bar (A logl0 Ks = ± 0.2). Solid curve: calculated according to the SIT model [97GRE/PLY2] with the interaction parameter s(Ni2+, CIO;) = 0.37 kgmol ' selected in this review.
A. Discussion of selected references
441
[2002MON/HEL] A kinetic study of cyanide exchange in [M(CN)4]2~ square-planar complexes (M = Ni, Pd and Pt) was performed as a function of pH using I3C NMR. The kinetic data confirmed the presence of the pentacyano complex at a higher excess of cyanide, and the protonated species below pH 7. Surprisingly, the protonation does not affect the chemical shift of the 13C NMR complexed-cyanide signal (Kolski and Margerum detected identical UV-VIS spectra for Ni(CN)^ and its protonated derivatives [68KOL/MAR]). The observed kinetic behaviour was described using log10 /34 = 31.0 for the formation constant of Ni(CN)^ [59FRE/SCH], log10 K5 = - 0.77 for the reaction: Ni(CN)4~ + CN ^
Ni(CN)5"
[71 PIE/HUG], and log10 K] H = 5.4 and logl0 K2 H = 4.5 for the two successive protonations of Ni(CN)^ [68KOL/MAR]. Under the conditions used (25°C, /,„ = 0.6 m), log10 /?4 = (29.75 ± 0.20) seems to be more reliable, but the impact of the erroneous /?4 on the conclusions (e.g., the presence of protonated species) is rather difficult to judge. [2002WAL/GAT] A new method to synthesize P-Ni(OH)2 has been described. It is based on the hydrolysis of Na2[Ni(OH)4] and leads to pure single crystalline nickel hydroxide with crystal sizes up to 0.25 mm. The influence of temperature and concentration on the crystal size was studied. [2002WAL/PRE] Solubility measurements of hellyerite were carried out at different ionic strengths of NaClO4 (/= 0.5, 1.0, 2.0 and 3.0 molkg '). These and all other solubility data on gaspeite [80REI], [82KLO] and hellyerite [11AGE/VAL], [30MUL/LUB], [2001GAM/PRE] available were thermodynamically analysed to obtain the respective standard thermodynamic quantities AfG° and A t / / ° . For hellyerite, 5°, was also determined. Note that in this paper all calculations on hellyerite were based on the stoichiometric formula NiCO3-6H2O. [2003BAE/BRA] The stability constants of nickel carbonato complexes were investigated by measuring the solid/liquid distribution ratio of Ni in the absence and presence of varying carbonate concentrations. For this purpose a 63Ni tracer method was employed. A considerable part of the uncertainty (log10 K°, for the reaction Ni2+ + CO^ ^ NiCO3(aq), is calculated to be (4.2 ± 0.4)) inherent in this method arises from the poorly known formation constants of neutral and anionic hydroxo complexes of nickel.
442
A. Discussion of selected references
[2003HUM/CUR] Hummel and Curti comprehensively reviewed complex formation between nickel and carbonato and hydrogen carbonato ligands. Not only was the basis for experimental measurements discussed, but also methods used for estimation of the formation constants [76ZHO/BEZ], [77MAT/SPO], [79MAT/SPO], [81TUR/WHI], [84FOU/CRI], [87EM A/FAR]. Therefore a short summary of their arguments suffices. As suggested by Garrels and Christ linear relationships between - log10 A^° or - log10 (i° and the electronegativity of the metals were used by Zhorov et al. to determine the respective constants for Ni(II) [65GAR/CHR], [76ZHO/BEZ]. The former case was calibrated with CuCO3(aq), MgCO3(aq) and CaCO3(aq), the latter with MgHCOj and CaHCOj only, although Garrels and Christ warned against an overestimation of the then available data on alkaline earth hydrogen carbonato complexes. Mattigod and Sposito correlated the complex formation constants with the sum of the radii of the metal ion and the ligand forming the ion-pair. Their estimation formula for the nickel carbonate system was calibrated to a single experimental data point (CuCO3(aq)) [77MAT/SPO]. Two years later, Mattigod and Sposito correlated these stability constants with a multiple parameter involving ionic charge, electronegativity, integral weighting factors, and a contribution calculated from atomic shielding constants [79MAT/SPO]. The numerical values obtained by the two estimation methods differ considerably, but their empirical bases remained equally doubtful. Turner et al. used the following correlation between the stability constants of carbonato and oxalato complexes: logic A,co3 = - 0.011 + 1.042 log10 y9MC2o4
(A. 108)
Hummel and Curti succeeded in revealing the rationale of Equation (A. 108) and pointed out that it leads to log10 K° (NiCO3, aq) = (5.4 + 1.5) rather than log10 K° = 5.36 [81TUR/WHI]. Fouillac and Criaud [84FOU/CRI] applied different versions of electrostatic models, but finally resorted to the numbers of Zhorov et al. [76ZHO/BEZ] which they mistook for experimental data and therefore corrected to / = 0. This vicious circle was probably one of the reasons which led to the subtitle "A thermodynamic elegy" for Hummel and Curti's review [2003HUM/CUR].
Appendix B Ionic strength corrections1 Thermodynamic data always refer to a selected standard state. The definition given by IUPAC [82LAF] is adopted in this review as outlined in Section II.3.1. According to this definition, the standard state for a solute B in a solution is a hypothetical solution, at the standard state pressure, in which mB = m = 1 molkg"1, and in which the activity coefficient yB is unity. However, for many reactions, measurements cannot be made accurately (or at all) in dilute solutions from which the necessary extrapolation to the standard state would be simple. This is invariably the case for reactions involving ions of high charge. Precise thermodynamic information for these systems can only be obtained in the presence of an inert electrolyte of sufficiently high concentration that ensures activity factors are reasonably constant throughout the measurements. This appendix describes and illustrates the method used in this review for the extrapolation of experimental equilibrium data to zero ionic strength. The activity factors of all the species participating in reactions in high ionic strength media must be estimated in order to reduce the thermodynamic data obtained from the experiment to the state 7 = 0 . Two alternative methods can be used to describe the ionic medium dependence of equilibrium constants: •
One method takes into account the individual characteristics of the ionic media by using a medium dependent expression for the activity coefficients of the species involved in the equilibrium reactions. The medium dependence is described by virial or ion interaction coefficients as used in the Pitzer equations [73PIT] and in the specific ion interaction theory.
This Appendix essentially contains the text of the TDB-2 Guideline written by Grenthe and Wanner [2000GRE/WAN], earlier versions of which have been printed in the previous NEATDB reviews [92GRE/FUG], [95SH7BID], [99RAR/RAN], [2001LEM/FUG] and [2003GUI/FAN]. The equations presented here are an essential part of the review procedure and are required to use the selected thennodynamic values. The contents of Tables B.4 and B.5 have been revised. 443
444
B Ionic strength corrections
•
The other method uses an extended Debye-Hiickel expression in which the activity coefficients of reactants and products depend only on the ionic charge and the ionic strength, but it accounts for the medium specific properties by introducing ion pairing between the medium ions and the species involved in the equilibrium reactions. Earlier, this approach has been used extensively in marine chemistry, cf. Refs. [79JOH/PYT], [79MIL], [79PYT], [79WHI2].
The activity factor estimates are thus based on the use of Debye-Hiickel type equations. The "extended" Debye-Hiickel equations are either in the form of specific ion interaction methods or the Davies equation [62DAV]. However, the Davies equation should in general not be used at ionic strengths larger than 0.1 mol • kg"1. The method preferred in the NEA Thermochemical Data Base review is a mediumdependent expression for the activity coefficients, which is the specific ion interaction theory in the form of the Brensted-Guggenheim-Scatchard approach. Other forms of specific ion interaction methods (the Pitzer and Brewer "B-method" [61 LEW/RAN] and the Pitzer virial coefficient method [79PIT]) are described in the NEA Guidelines for the extrapolation to zero ionic strength [2000GRE/WAN]. The specific ion interaction methods are reliable for intercomparison of experimental data in a given concentration range. In many cases this includes data at rather low ionic strengths, / = 0.01 to 0.1 M, cf. Figure B-l, while in other cases, notably for cations of high charge (> + 4 and < - 4), the lowest available ionic strength is often 0.2 M or higher, see for example Figures V.I2 and V.I3 in [92GRE/FUG]. It is reasonable to assume that the extrapolated equilibrium constants at / = 0 are more precise in the former than in the latter cases. The extrapolation error is composed of two parts, one due to experimental errors, and the other due to model errors. The model errors seem to be rather small for many systems, less than 0.1 units in logl0 K°. For reactions involving ions of high charge, which may be extensively hydrolysed, one cannot perform experiments at low ionic strengths. Hence, it is impossible to estimate the extrapolation error. This is true for all methods used to estimate activity corrections. Systematic model errors of this type are not included in the uncertainties assigned to the selected data in this review. It should be emphasised that the specific ion interaction model is approximate. Modifying it, for example by introducing the equations suggested by Ciavatta ([90CIA], Eqs. (8-10), cf. Section B.1.4), would result in slightly different ion interaction coefficients and equilibrium constants. Both methods provide an internally consistent set of values. However, their absolute values may differ somewhat. Grenthe et al. [92GRE/FUG] estimate that these differences in general are less than 0.2 units in log10 K°, i.e., approximately 1 kJ-mol"1 in derived AfG° values.
B. I The specific ion interaction equations
445
B.I The specific ion interaction equations B.I.I Background In the following discussion of the SIT model strict quantity calculus was sacrified in order to improve readability {cf. [91STO]). The Debye-Huckel term, which is the dominant term in the expression for the activity coefficients in dilute solution, accounts for electrostatic, non-specific long-range interactions. At higher concentrations, short range, non-electrostatic interactions have to be taken into account. This is usually done by adding ionic strength dependent terms to the Debye-Huckel expression. This method was first outlined by Brensted [22BRO], [22BRO2] and elaborated by Scatchard [36SCA] and Guggenheim [66GUG]. Biedermann [75BIE] highlighted its practical value, especially for the estimation of ionic medium effects on equilibrium constants. The two basic assumptions in the specific ion interaction theory are described below. • Assumption 1: The activity coefficient Yj of an iony of charge z, in the solution of ionic strength /,„ may be described by Eq. (B.I): logl0 Yj = ~z) D+ Ze(j,k,ljmk
(B.I)
D is the Debye-Huckel term: D=
A
^
r
_
(B.2)
\ + Baj4Fm where /,„ is the molal ionic strength:
A and B are constants which are temperature and pressure dependent, and a, is an ion size parameter ("distance of closest approach") for the hydrated ion j . The DebyeHuckel limiting slope, A, has a value of (0.509 + 0.001) kg^ -mol^ at 25°C and 1 bar, {cf Section B.1.2). The term So, in the denominator of the Debye-Huckel term has been assigned a value of Baj = 1 . 5 kg^ m o l ^ at 25°C and 1 bar, as proposed by Scatchard [76SCA] and accepted by Ciavatta [80CIA]. This value has been found to minimise, for several species, the ionic strength dependence of e{j,k,Im) between /,„ = 0.5 m and /,„ =3.5 m. It should be mentioned that some authors have proposed different values for Ba, ranging from Ba} = 1.0 [35GUG] to Baj = 1.6 [62VAS]. However, the parameter Baj is empirical and as such is correlated to the value of z(j,k,Im). Hence, this variety of values for Baj does not represent an uncertainty range, but rather indicates that several different sets of So, and e{j, k, Im) may describe equally well the experimental mean activity coefficients of a given electrolyte. The ion interaction coefficients at 25°C listed in Table B-4, Table B-5 and Table B-6 have thus to be used with Baj= 1.5 kg^-mol"^.
446
B Ionic strength corrections
The summation in Eq. (B.I) extends over all ions k present in solution. Their molality is denoted by wi)+\zJ\g(T,p)
[90OEL/HEL], where g(T,p) is a
temperature and pressure function which is tabulated in [88TAN/HEL], [92SHO/OEL], and is approximately zero at temperatures below 175°C. The values of e(j,k,lm), obtained with the methods described in Section B.I.3 at temperatures other than 25°C, will depend on the value adopted for Baj. As long as a consistent approach is followed, values of z{j,k,Im) absorb the choice of Baj, and for moderate temperature intervals (between 0 and 200°C) the choice Ba, = 1.5 kg^ • mol'"• is the simplest one and is recommended by this review. The variation of c(j,k,Im) with temperature is discussed by Lewis et al. [61 LEW/RAN], Millero [79MIL], Helgeson et al. [81HEL/KIR], [90OEL/HEL], Giffaut et al. [93GIF/VIT2] and Grenthe and Plyasunov [97GRE/PLY]. The absolute values for the reported ion interaction parameters differ in these studies due to the fact that the Debye-Hiickel term used by these authors is not exactly the same. Nevertheless, common to all these studies is the fact that values of (de/BT)p are usually + e(PuOr,ClO-)
m^
+
l o g 1 0 r r = - £ > + e(H ,ClO4-) w a04 Hence, log l 0 /f = l o g 1 0 ^ - 8 D + (e(Pu4+, ClOi) -eCPuOf, CIO;) - 4 e ( H + , C1O4) ) « a 0 4 +21og loflH20
(B15)
The relationship between the equilibrium constant and the redox potential is: nF \nK = ~E0' RT
(B.16)
lnK° = — E°. (B.17) RT E°' is the redox potential in a medium of ionic strength /, E° is the corresponding standard potential at / = 0, and n is the number of transferred electrons in the reaction considered. Combining Eqs. (B.I5), (B.16) and (B.17) and rearranging them leadstoEq.(B.18):
£ -- ffD -2i*.^SZm)-£'-*^«2Jf«tt)
„,,„
For n = 2 in the present example and T= 298.15 K, Eq.(B.18) becomes: £°'[mV]-236.6Z) + 59.161og 10 a Hi o =£°[mV]-29.58Ae maowhere As = £(Pu4+, CIO;) -e(PuOf, CIO;) - 4 s ( H + , CIO;). The value of aH0 can be taken from experimental data or calculated from equations (B. 10) and (B. 11). In general, formal potentials are reported with reference to the standard hydrogen electrode, cf. Section II. 1.6.5, as exemplified in Tables V.2 and V.3 of the uranium
B.I The specific ion interaction equations
457
NEA review [92GRE/FUG]. In that case, the H+ appearing in the reduction reaction is already at standard conditions. For example, experimental data are available on the formal potentials for reactions: PuC>2++ 4H + + 2e" ^
Pu 4+ + 2H2O(1)
(B.I9)
and PuOf+ e
^
PuO;.
(B.20)
While reaction (B.19) corresponds to (B.14), reaction (B.20) is equivalent to: PuOf + | H 2 ( g ) ^
PuO;+ H+
(B.21)
where the designator "(r)" has been omitted, since in these equations only the H+ in the reference compartment is relevant. The cations in reaction (B.14) represent aqueous species in the ionic media used during the experiments. In reaction (B.21) H+ represents the cation in the standard hydrogen electrode, and therefore it is already in standard conditions, and its activity coefficient must not be included in any extrapolation to / = 0 of experimental values for Reaction (B.20). Reactions (B.20) and (B.21) are equivalent, as are reactions (B.14) and (B.19), as can be seen if any of these equations are combined with reaction (11.26). Hence Eq. (B. 18) can be obtained more simply by using Eq. 11.33 for reaction (B. 19).
B.I.4 On the magnitude of ion interaction coefficients Ciavatta [80CIA] made a compilation of ion interaction coefficients for a large number of electrolytes. Similar data for complexations of various kinds were reported by Spahiu [83SPA] and Ferri et al. [83FER/GRE]. These and some other data for 25°C and 1 bar have been collected and are listed in Section B.3. It is obvious from the data in these tables that the charge of an ion is of great importance for determining the magnitude of the ion interaction coefficient. Ions of the same charge type have similar ion interaction coefficients with a given counter-ion. Based on the tabulated data, Grenthe et al. [92GRE/FUG] proposed that it is possible to estimate, with an error of at most ± 0.1 kg • mol"1 in e, ion interaction coefficients for cases where there are insufficient experimental data for an extrapolation to / = 0. The error that is made by this approximation is estimated to + 0.1 kg • mol"1 in As in most cases, based on comparison with As values of various reactions of the same charge type. Ciavatta [90CIA] has proposed an alternative method to estimate values of s for a first or second complex, ML or ML2, in an ionic media NX, according to the following relationships: s(ML,NorX)« (e(M,X) +e(L,N) )/2
(B.22)
458
B Ionic strength corrections
e(ML 2 ,NorX) « (e(M,X) +2e(L,N) )/3
(B.23) 1-1
Ciavatta obtained [90CIA] an average deviation of ± 0.05 kg • mol between e estimates according to Eqs. (B.22) and (B.23), and the e values at 25°C obtained from ionic strength dependency of equilibrium constants.
B.2 Ion interaction coefficients versus equilibrium constants for ion pairs It can be shown that the virial type of activity coefficient equations and the ionic pairing model are equivalent provided that the ionic pairing is weak. In these cases the distinction between complex formation and activity coefficient variations is difficult or even arbitrary unless independent experimental evidence for complex formation is available, e.g., from spectroscopic data, as is the case for the weak uranium(VI) chloride complexes. It should be noted that the ion interaction coefficients evaluated and tabulated by Ciavatta [80CIA] were obtained from experimental mean activity coefficient data without taking into account complex formation. However, it is known that many of the metal ions listed by Ciavatta form weak complexes with chloride and nitrate ion. This fact is reflected by ion interaction coefficients that are smaller than those for the noncomplexing perchlorate ion, cf. Table B-4. This review takes chloride and nitrate complex formation into account when these ions are part of the ionic medium and uses the value of the ion interaction coefficient, e(M"+,C10~), as a substitute for e(M" + ,Cr) and e(M" + ,NO3). In this way, the medium dependence of the activity coefficients is described with a combination of a specific ion interaction model and an ion pairing model. It is evident that the use of NEA recommended data with ionic strength correction models that differ from those used in the evaluation procedure can lead to inconsistencies in the results of the speciation calculations. It should be mentioned that complex formation may also occur between negatively charged complexes and the cation of the ionic medium. An example is the stabilisation of the complex ion, UO2(CO3)3~ , at high ionic strength, see for example Section V.7.1.2.1.d (p. 322) in the uranium review [92GRE/FUG].
B.3 Tables of ion interaction coefficients Table B-4, Table B-5, Table B-6 and Table B-7 contain the selected specific ion interaction coefficients used in this review, according to the specific ion interaction theory described. Table B-4 contains cation interaction coefficients with Cl", C1O~ and NO 3, Table B-5 anion interaction coefficients with Li+, Na+ (or NH4) and K+, and Table B-7 neutral species—electroneutral combination of ions. The coefficients have the units of kgrnol"1 and are valid for 298.15 K and 1 bar. The species are ordered by charge and appear, within each charge class, in standard order of arrangement, cf. Section II. 1.8.
B.3 Tables of ion interaction coefficients
459
In some cases, the ionic interaction can be better described by assuming ion interaction coefficients as functions of the ionic strength rather than as constants. Ciavatta [80CIA] proposed the use of Eq. (B.24) for cases where the uncertainties in Table B-4 and Table B-5 are ± 0.03 kg-mol"1 or greater. 6 = 8, + e2 log10 lm
(B.24)
For these cases, and when the uncertainty can be improved with respect to the use of a constant value of e, the values e, and z2 given in Table B-6 should be used. It should be noted that ion interaction coefficients tabulated in Table B-4, Table B-5 and Table B-6 may also involve ion pairing effects, as described in Section B.3. In direct comparisons of ion interaction coefficients, or when estimates are made by analogy, this aspect must be taken into account.
460
B Ionic strength corrections
Table B-4: Ion interaction coefficients s(j,k)(kgmol ') for cations j with k = Cl , C1O~ and NO", taken from Ciavatta [80CIA], [88CIA] unless indicated otherwise. The uncertainties represent the 95% confidence level. The ion interaction coefficients marked with | can be described more accurately with an ionic strength dependent function, listed in Table B-6. The coefficients £(M" + ,Cr) ande(M" + ,NO3) reported by Ciavatta [80CIA] were evaluated without taking chloride and nitrate complexation into account, as discussed in Section B. 2.
cr
cio4
(0.12 ±0.01)
(0.14 ±0.02)
I H+ NH4
-(0.01 ±0.01)
H 2 giy
+
- (0.08 ± 0.04)
(0.07 ±0.01) f
- (0.06 ± 0 . 0 3 /
- (0.06 ± 0.02)
Tl +
-(0.21 +0.06) t
ZnHCOj CdCl Cdl
NO;
0.2
(a)
+
(0.25 ± 0.02)
+
(0.27 ± 0.02)
CdSCN^ HgCl
(0.31 ±0.02)
+
(0.19 ±0.02)
Cu +
(0.11 ±0.01) (0.00 ±0.01)
Ag* NiOH NiF
+
- (0.01 ± 0.07)
(K)
+
(0.14±0.07) (L) (0.34 ± 0.08)(M)
NiCl +
(0.47 + 0.06)(N)
N1NO3
(0.44 ± 0.14)(0)
NiBr+
(0.59 ± 0.10)(P)
NiHS
-(0.12±0.05) f
+
- (0.85 ± 0.39)(Q)
NiSCN+
(0.31 ±0.04)"*' (0.17±0.04)
UO2NO3
(0.33 + 0.04)'°'
UO2SCN+
(0.22 ± 0.04)'°'
2+
(0.15 ±0.02)
A1OH 2 + A12CO3(OH)2
0.09 (s) +
0.26
-(0.20±0.12) f
0.3l' s) (0.33 ± 0.03)
(0.16 ±0.02)
(0.35 ± 0.05)'a>
Cd 2+ Hg
(0.41 ±0.22)'"'
(s)
Zn
ZnCof
(0.51 ±1.4)'"'
(0.33 + 0.04)'"'
UOJCIOJ
Pb
NO;
(0.09 ± 0.02)
2+
(0.34 ±0.03)
-(0.1 ±0.1) f
(0.09 ± 0.02)
-(0.2 + 0 . 1 ^
< Cu 2+
(0.08 ±0.01)
(0.32 + 0.02)
(0.11+0.01)
Ni 2+
(0.17 ±0.02)
(0.370 ± O.O32)(S)
(0.182 ± 0.010) m
Co 2+
(0.16 ±0.02)
(0.34 + 0.03)
(0.14 + 0.01)
(Continued on next page)
462
B Ionic strength corrections
Table B-4 (continued) j
cr
k^
i
cio;
FeOH2+ FeSCN
0.38 ( b )
2+
0.45 ( b >
Mn2+
(0.13 ± 0 . 0 1 )
YHCof
(0.39 ± 0.04) ( b )
AmOH2+
- (0.04 + 0.07) ( q )
(0.39 ± 0.04) ( c ) (0.39 ± 0.04) ( c )
AmF AmCl
NO;
2+
(0.39 ± 0.04) (based on E(Am !i ,Cl") = (0.23 + 0.02) kg • mol"1 and £(Na*, CO,") = - (0.08 ± 0.03) kg • mor 1 .
(s)
Evaluated in Appendix A, discussion of [98ALM/NOV] from AE for the reactions KNpO,COj(s) + 2CO3~ ^ NpO 2 (CO 3 ); +K + (in K2CO3-KC1 solution) and K3NpO2(CO3)2(s) + CO," ^ NpO, (CO3 )\~ + 3 IC (in K2CO3 solution) (based on E(K* , CO3~) = (0.02 ± 0.01) kgmol"1).
(t)
Evaluated
in
Ni2+ + 3SCN~ ^ (u)
Evaluated
in
Ni2+ + 4CN~ ^ (v)
Evaluated
in
Section
V.7.1.3.1
from
Aen
Section
perchlorate
media
for
the
reaction
V.7.1.2.1.1
from
Aen
in
perchlorate
media
for
the
reaction
from
AE,, in
perchlorate
media
for
the
reaction
Ni(CN)*~ . Section
V.7.1.2.1.1
2+
Ni + 5 C N ~ ^ N i ( C N ) ' ~ . (w) Evaluated in [2003GUI/FAN], Section 9.4.2.2.1.1. (x)
in
Ni(SCN); .
As reported in [92BAN/BLI].
470
B Ionic strength corrections
Table B-6: Ion interaction coefficients, s(l J,k) and e(2J,k) kg • mol ', for cations y with k = Cl~, ClO^ and NO3 (first part), and for anionsy with k - Li+, Na+ and K+ (second part), according to the relationship z = Z\ + e2 logialm. The data are taken from Ciavatta [80CIA], [88CIA], The uncertainties represent the 95% confidence level. C
I
r
c:ic>4
£1
£1
NO3 E2
-(0.088+0.002) (0.095+0.012) Tl +
-(0.18±0.02)
61
62
-(0.075+0.001)
(0.057±0.004)
(0.09±0.02)
+
-(0.1432+0.0002) (0.0971+0.0009
2+
-(0.329+0.007)
(0.288+0.018)
Hg 2 +
-(0.145±0.001)
(0.194±0.002)
Ag Pb
-(0.230010.0004) (0.194+0.002) 1Ma+
L;i +
I
El
OH"
-(0.039±0.002) (0.02+0.01)
NO2
£|
S2
K+ £2
£1
£2
(0.072+0.006) (0.11+0.01)
NO3
-(0.049+0.001) (0.044+0.002)
-(0.131+0.002)
H 2 PO 4
-(0.109+0.001) (0.095±0.003)
-(0.1473+0.0008) (0.121+0.004)
B(OH) 4
-(0.092+0.002) (O.1O3±O.OO5)
SO 2 ~
-(0.125+0.008) (0.106±0.009)
SO4"
-(0.068+0.003)
(0.082±0.006)
(0.093+0.007) -(0.184+0.002) (0.139±0.006)
S 2 O|"
-(0.125+0.008) (0.106+0.009)
HPO4"
-(0.19±0.01)
(0.11±0.03)
-(0.152+0.007)
(0.123+0.016)
CrO^"
-(0.090±0.005) (0.07±0.01)
-(0.123+0.003)
(0.106+0.007)
PO3A~
-(0.29+0.02)
(0.10+0.01)
Table B-7: SIT interaction coefficient e(j,k) kg • mol for neutral species,/ with k, electroneutral combination of ions. j k ->
Na + + CIO4
I Ni(SCN)2(aq)
(0.38 + 0.06)
see discussion in Section V.7.1.3.1
Appendix C Assigned uncertainties1 This Appendix describes the origin of the uncertainty estimates that are given in the TDB tables of selected data. The original text in [92GRE/FUG] has been retained in [95SIL/BID], [99RAR/RAN], [2001LEM/FUG] and [2003GUI/FAN] except for some minor changes. Because of the importance of the uncertainty estimates, the present review offers a more comprehensive description of the procedures used.
C.I The general problem The focus of this section is on the uncertainty estimates of equilibria in solution, where the key problem is analytical, i.e., the determination of the stoichiometric composition and equilibrium constants of complexes that are in rapid equilibrium with one another. We can formulate analyses of the experimental data in the following way: From N measurements, yh of the variable y we would like to determine a set of n equilibrium constants knr — 1, 2,..., n, assuming that we know the functional relationship: y=j[kl,k2,
...kr...kn;a\,a2,....)
(C.I)
where a\, a2, etc. are quantities that can be varied but whose values (au; a-u\ etc.) are assumed to be known accurately in each experiment from the data sets (y,, au, a2i,...), i - 1,2, ...N. The functional relationship (C.I) is obtained from the chemical model proposed and in general several different models have to be tested before the "best" one is selected. Details of the procedures are given in Rossotti and Rossotti [61ROS/ROS]. When selecting the functional relationship (C.I) and determining the set of equilibrium constants that best describes the experiments one often uses a least-squares method. Within this method, the "best" description is the one that will minimise the residual sum of squares, U:
U= Z w -b/-/(*.-*^ a w.« 2 ,--)] 2
(C2)
where w, is the weight of each experimental measurement yh
This Appendix essentially contains the text of the TDB-3 Guideline, [99WAN/OST], earlier versions of which have been printed in the previous NEA TDB reviews [92GRE/FUG], [95SIL/BID], [99RAR/RAN], [2001LEM/FUG] and [2003GUI/FAN], Because of its importance in the selection of data and to guide the users of the values in Chapters III and IV, the text is reproduced here with minor revisions.
471
472
C. Assigned uncertainties
The minimum of the function (C.2) is obtained by solving a set of normal equations: ! ^ - = 0,r = l, dkr A "true" minimum is only obtained if:
n
(C.3)
•
the functional relationship (C.I) is correct, i.e., if the chemical model is correct.
•
all errors are random errors in the variable y, in particular there are no systematic errors.
•
the random errors in y follow a Gaussian (normal) distribution.
•
the weight w,{yh au, a2i,....) of an experimental determination is an exact measure of its inherent accuracy.
To ascertain that the first condition is fulfilled requires chemical insight, such as information of the coordination geometry, relative affinity between metal ions and various donor atoms, etc. It is particularly important to test if the chemical equilibrium constants of complexes that occur in small amounts are chemically reasonable. Too many experimentalists seem to look upon the least-squares refinement of experimental data more as an exercise in applied mathematics than as a chemical venture. One of the tasks in the review of the literature is to check this point. An erroneous chemical model is one of the more serious types of systematic errors. The experimentalist usually selects the variable that he/she finds most appropriate to fulfill the second condition. If the estimated errors in au, a2;... are smaller than the error in yt, the second condition is reasonably well fulfilled. The choice of the errorcarrying variable is a matter of choice based on experience, but one must be aware that it has implications, especially in the estimated uncertainty. The presence of systematic errors is, potentially, the most important source of uncertainty. There is no possibility to handle systematic errors using statistics; statistical methods may indicate their presence, no more. Systematic errors in the chemical model have been mentioned. In addition there may be systematic errors in the methods used. By comparing experimental data obtained with different experimental methods one can obtain an indication of the presence and magnitude of such errors. The systematic errors of this type are accounted for both in the review of the literature and when taking the average of data obtained with different experimental methods. This type of systematic error does not seem to affect the selected data very much, as judged by the usually very good agreement between the equilibrium data obtained using spectroscopic, potentiometric and solubility methods.
C.2 Uncertainty estimates in the selected thermodynamic data.
473
The electrode calibration, especially the conversion between measured pH and - logio[H+] is an important source of systematic error. The reviewers have when possible corrected this error, as seen in many instances in Appendix A. The assumption of a normal distribution of the random errors is a choice made in the absence of better alternatives. Finally, a comment on the weights used in least-squares refinements; this is important because it influences the uncertainty estimate of the equilibrium constants. The weights of individual experimental points can be obtained by repeating the experiment several times and then calculating the average and standard deviation of these data. This procedure is rarely used, instead most experimentalists seem to use unit weight when making a least-squares analysis of their data. However, also in this case there is a weighting of the data by the number of experimental determinations in the parameter range where the different complexes are formed. In order to have comparable uncertainty estimates for the different complexes, one should try to have the same number of experimental data points in the concentration ranges where each of these complexes is predominant; a procedure very rarely used. As indicated above, the assignment of uncertainties to equilibrium constants is not a straightforward procedure and it is complicated further when there is lack of primary experimental data. The uncertainty estimates given for the individual equilibrium constants reported by the authors and for some cases re-estimated by this review are given in the tables of this and previous reviews. The procedure used to obtain these estimates is given in the original publications and in the Appendix A discussions. However, this uncertainty is still a subjective estimate and to a large extent based on "expert judgment".
C.2 Uncertainty estimates in the selected thermodynamic data. The uncertainty estimate in the selected thermodynamic data is based on the uncertainty of the individual equilibrium constants or other thermodynamic data, calculated as described in the following sections. A weighted average of the individual log\oK values is calculated using the estimated uncertainty of the individual experimental values to assign its weight. The uncertainty in this average is then calculated using the formulae given in the following text. This uncertainty depends on the number of experimental data points — for N data points with the same estimated uncertainty, a, the uncertainty in the average is o/yjN . The average and the associated uncertainty reported in the tables of selected data are reported with many more digits than justified only in order to allow the users to back-track the calculations. The reported uncertainty is much smaller than the estimated experimental uncertainty and the users of the tables should look at the discussion of the selected constants in order to get a better estimate of the uncertainty in an experimental determination using a specific method.
474
C. Assigned uncertainties
One of the objectives of the NEA Thermochemical Data Base (TDB) project is to provide an idea of the uncertainties associated with the data selected in this review. As a rule, the uncertainties define the range within which the corresponding data can be reproduced with a probability of 95% at any place and by any appropriate method. In many cases, the statistical treatment is limited or impossible due to the availability of only one or few data points. A particular problem has to be solved when significant discrepancies occur between different source data. This appendix outlines the statistical procedures, which were used for fundamentally different problems, and explains the philosophy used in this review when statistics were inapplicable. These rules are followed consistently throughout the series of reviews within the TDB Project. Four fundamentally different cases are considered: 1. One source datum available 2. Two or more independent source data available 3. Several data available at different ionic strengths 4. Data at non-standard conditions: Procedures for data correction and recalculation.
C.3 One source datum The assignment of an uncertainty to a selected value that is based on only one experimental source is a highly subjective procedure. In some cases, the number of data points, on which the selected value is based, allows the use of the "root mean square" [82TAY] deviation of the data points, Xh to describe the standard deviation, sx, associated with the average, X :
The standard deviation, sx, is thus calculated from the dispersion of the equally weighted data points, Xh around the average X, and the probability is 95% that an^Y, is within X ±1.96 sx, see Taylor [82TAY] (pp. 244-245). The standard deviation, sx, is a measure of the precision of the experiment and does not include any systematic errors. Many authors report standard deviations, sx, calculated with Eq. (C.4) (but often not multiplied by 1.96), but these do not represent the quality of the reported values in absolute terms. Therefore, it is thus important not to confuse the standard deviation, sx, with the uncertainty, a. The latter reflects the reliability and reproducibility of an experimental value and also includes all kinds of systematic errors, Sj, that may be involved. The uncertainty, a, can be calculated with Eq. (C.5), assuming that the systematic errors are independent.
°x = J4+Us2j)
(C5)
C.4 Two or more independent source data
475
The estimation of the systematic errors Sj (which, of course, have to relate to X and be expressed in the same units), can only be made by a person who is familiar with the experimental method. The uncertainty, o, has to correspond to the 95% confidence level preferred in this review. It should be noted that for all the corrections and recalculations made (e.g., temperature or ionic strength corrections) the rules of the propagation of errors have to be followed, as outlined in Section C.6.2. More often, the determination of sx is impossible because either only one or two data points are available, or the authors did not report the individual values. The uncertainty a in the resulting value can still be estimated using Eq. (C.5) assuming that s\ is much smaller than Ufa/) > which is usually the case anyway.
C.4 Two or more independent source data Frequently, two or more experimental data sources are available, reporting experimental determinations of the desired thermodynamic data. In general, the quality of these determinations varies widely, and the data have to be weighted accordingly for the calculation of the mean. Instead of assigning weight factors, the individual source data, Xh are provided with an uncertainty, O;, that also includes all systematic errors and represents the 95% confidence level, as described in Section C.3 The weighted mean X and its uncertainty, a^ , are then calculated according to Eqs. (C.6) and (C.7).
CT
^=
/
x
pi?)
(Q7)
Eqs. (C.6) and (C.7) may only be used if all the X-, belong to the same parent distribution. If there are serious discrepancies among the Xt, one proceeds as described below under Section C.4.1. It can be seen from Eq. (C.7) that a^ is directly dependent on the absolute magnitude of the a, values, and not on the dispersion of the data points around the mean. This is reasonable because there are no discrepancies among the Xh and because the a, values already represent the 95% confidence level. The selected uncertainty, a^ , will therefore also represent the 95% confidence level. In cases where all the uncertainties are equal, a, =a, Eqs. (C.6) and (C.7) reduce to Eqs. (C.8) and (C.9).
X=^-ix,
(C.8)
476
C. Assigned uncertainties
(C9)
.12) 2 + (3.5x0.12) 2 = 0.44.
The selected, rounded value is: log I 0 >(C.20) = -(4.9±0.4).
C.6.2 Propagation of errors Whenever data are converted or recalculated, or other algebraic manipulations are performed that involve uncertainties, the propagation of these uncertainties has to be taken into account in a correct way. A clear outline of the propagation of errors is given by Bevington [69BEV]. A simplified form of the general formula for error propagation is given by Eq.(C21), supposing thatXis a function of Y\, Y2,...,YN.
Eq. (C.21) can be used only if the variables, Y\, Y2,...,YN, are independent or if their uncertainties are small, i.e., the covariances can be disregarded. One of these two assumptions can almost always be made in chemical thermodynamics, and Eq. (C.21) can thus almost universally be used in this review. Eqs. (C.22) through (C.26) present explicit formulas for a number of frequently encountered algebraic expressions, where c, c\, c2 are constants. a2x = (c^ f + ( c ^ f
X = ciYl±c2Y2:
™*-*-f: ^HfHfj x=cr-
2 L=c ^ x
••
X = cxe±c{i : X =Cl\n(±c2Y)
:
(C.22)
(C23)
(C24)
— = c2ar
(C.25)
ax = c , ^
(C.26)
Example C.7: A few simple calculations illustrate how these formulas are used. The values have not been rounded. Eq. (C.22):
ArGm = 2-[ - (277.4 ± 4.9)] kJ-mol"1 - [ - (467.3 ± 6.2)] kJ-mol"1 = -(87.5 ± 11.6)kJ-mor'.
Eq.(C23):
K=
( °'038±0'°Q2) (0.004710.0005)
= ( 8.09±0.96)
486
C. Assigned uncertainties
Eq. (C.24):
AT= 4-(3.75 ± 0.12)3 = (210.9 ± 20.3) -A r G°
Eq. (C.25):
K° = e
Rr
;
ArG° = - (2.7 ± 0.3) kJ-mol"1 R = 8.3145 HC'-moP 1 T= 298.15 K
A"0 =(2.97 ±0.36). Note that powers of 10 have to be reduced to powers of e, i.e., the variable has to be multiplied by ln(10), e.g., \ogwK = (2.4510.10); K = 10log'°* =e(l" (3 du monosulfure de nickel, C. R. Seances Acad. Set, Ser. C, 290, (1980), 89-92, in French. Cited on pages: 172,389.
[80LAG/FON]
Lagarde, P., Fontaine, A., Raoux, D., Sadoc, A., Migliardo, P., EXAFS studies of strong electrolytic solutions, J. Chem. Phys., 72, (1980), 3061-3068. Cited on page: 141.
[80LIB/SAD]
Libus, W., Sadowska, T., Libus, Z., Correlation between thermodynamic properties and coordination states of aqueous bivalent transition metal sulfates, J. Solution Chem., 9, (1980), 341354. Cited on pages: 181,389.
[80MIL/BUG]
Milic, N. B., Bugarcic, Z. D., Vasic, M. V., Hydrolysis of the nickel(II) ion in potassium chloride medium, Bull. Soc. Chim. Beograd, 45, (1980), 349-357. Cited on pages: xxv, 92, 93, 94, 389, 390,391.
[80OMA/SHA]
Omarova, F. M , Sharipov, M. S., Determination of the standard heat of formation of nickel orthoarsenate octahydrate, VINITI, Report 5859-80, (1980). Cited on page: 213.
[80OPP/TOS]
Oppermann, H., Toschew, A., Zur thermischen Zersetzung und Sublimation des Nil2, Z. Anorg. Allg. Chem., 436, (1980), 43-55, in German. Cited on pages: 137,391.
[80PET/SHE]
Petrov, G. V., Shevchuk, V. G, Solubility polytherm of the KC1NiCl2-H2O system, Russ. J. Inorg. Chem., 25, (1980), 1715-1716. Cited on page: 128.
[80PET/TAB]
Pethybridge, A. D., Taba, S. S., Precise conductimetric studies on aqueous solutions of 2:2 electrolytes. Part 2. Analysis of data for MgSC>4 in term of new equations from Fuoss and from Lee and Wheaton, J. Chem. Soc. Faraday Trans., 76, (1980), 368-376. Cited on pages: 185, 186.
[80REI]
Reiterer, F., Loslichkeitskonstanten und Freie Bildungsenthalpien neutraler Ubergangsmetallcarbonate, Ph. D. Thesis, Montanuniversitat, (1980), Leoben, Austria, in German. Cited on pages: 218,392,398,441.
550
BIBLIOGRAPHY
[80SHA/CHA]
Sharma, R. C , Chang, Y. A., Thermodynamics and phase relationships of transition metal-sulfur systems: IV. Thermodynamic properties of the Ni-S liquid phase and the calculation of the Ni-S phase diagram, Metall. Trans. B, 11, (1980), 139-146. Cited on pages: 163, 166, 167, 170, 172, 175, 176, 278, 392, 407, 419.
[80TRA/BRU]
Tran, T., Brungs, M. P., Applications of oxygen electrodes in glass melts. Part 3: An oxygen concentration cell for thermodynamic studies of CO-CO2 and Ni-NiO systems, Phys. Chem. Glasses, 21, (1980), 184-188. Cited on page: 105.
[80TRE/LEB2]
Tremaine, P. R., LeBlanc, J. C , The solubility of nickel oxide and hydrolysis of Ni2+ in water to 573 K, J. Chem. Thermodyn., 12, (1980), 521-538. Cited on pages: 92, 93, 95, 100, 101, 102, 107, 392, 394, 395.
[81ADA/BIL]
Adam, A., Billerey, D., Terrier, C , Bartholin, H., Regnault, L. P., Rossat-Mignod, J., Hydrostatic pressure effect on the commensurate-incommensurate phase transition of NiBr2, Phys. Lett., 84A, (1981), 24-27. Cited on pages: 132, 386.
[81BAN/KAD]
Banerjea, D., Kaden, T. A., Sigel, H., Enhanced stability of ternary complexes in solution through the participation of heteroaromatic N bases. Comparison of the coordination tendency of pyridine, imidazole, ammonia, acetate, and hydrogen phosphate toward metal ion nitrilotriacetate complexes, Irtorg. Chem., 20, (1981), 25862590. Cited on page: 330.
[81 BAR/RAN]
Barnard, R., Randell, C. F., Tye, F. L., Studies concerning charged nickel hydroxide electrodes: part III. Reversible potentials at low states of charge, J. Electroanal. Chem., 119, (1981), 17-24. Cited on pages: 116, 117, 118,396.
[81COR/SUG]
Corlis, C , Sugar, J., Energy levels of Ni I through Ni XXVIII, J. Phys. Chem. Ref. Data, 10, (1981), 197-290. Cited on page: 75.
[81HEL/KIR]
Helgeson, H. C , Kirkham, D. H., Flowers, G. C , Theoretical prediction of the thermodynamic behavior of aqueous electrolytes at high pressures and temperatures: IV. Calculation of activity coefficients, osmotic coefficients, and apparent molal and standard and relative partial molal properties to 600°C and 5 kb, Am. J. Sci., 281,(1981), 1249-1516. Cited on pages: 434,451.
BIBLIOGRAPHY
551
[81KUI/SAN]
Kuindersma, S. R., Sanchez, J. P., Haas, C , Magnetic and structural investigations on Nil2 and Col2, Physica, 11 IB, (1981), 231-248. Cited on pages: 136, 370, 371, 380.
[81KUL/BLO]
Kul'vinova, L. A., Blokhin, V. V., Makashev, Y. A., Mironov, V. E., Thermodynamics of the formation of fluoride complexes of transition metal in water-salt solution, Koord. Khim., 7, (1981), 201206, in Russian, English translation in [81KUL/BLO2]. Cited on pages: 142, 143, 144, 145, 377, 396, 399.
[81KUL/BLO2]
Kul'vinova, L. A., Blokhin, V. V., Makashev, Y. A., Mironov, V. E., Thermodynamics of formation of fluoride complexes of transition metal in water-salt solution, Sov. J. Coord. Chem., 7, (1981), 104108. From a citation in this Bibliography
[81 MAR/ECO]
Marcopoulos, T., Economou, M., Theophrastite, Ni(OH)2, a new mineral from northern Greece, Am. Mineral, 66, (1981), 1020-1021. Cited on pages: 108,397,398,399.
[81NIC/BER]
Nickel, E. H., Berry, L. G., The new mineral nullaginite and additional data on the related minerals rosasite and glaukosphaerite, Can. Mineral, 19, (1981), 315-324. Cited on pages: 215, 397.
[81PER/ROU]
Perron, G., Roux, A., Desnoyers, J. E., Heat capacities and volumes of NaCl, MgCl2, CaCl2 and NiCl2 up to 6 molal in water, Can. J. Chem., 59, (1981), 3049-3054. Cited on pages: 86, 87, 88.
[81SCH]
Schaefer, S. C , Electrochemical determination of Gibbs energies of formation of cobalt and nickel sulfides, United States Department of the Interior, Bureau of Mines, Report RI 8588, (1981). Cited on pages: 397,407.
[81SIL]
Silverman, D. C , Revised EMF-pH diagram for nickel, Corrosion NACE, vol. 37, pp.546-548, (1981). Cited on page: 434.
[81STO]
Stolyrova, T. A., Thermochemistry of arsenides of iron, cobalt, and nickel, Vses. Soveshch. Eksp. Tekh. Mineral. Petrogr., 10th, (1981), 107-113, in Russian. Cited on page: 212.
552
BIBLIOGRAPHY
[81TUR/WHI]
Turner, D. R., Whitfield, M., Dickson, A. G., The equilibrium speciation of dissolved components in freshwater and seawater at 25°C and 1 atm pressure, Geochim. Cosmochim. Acta, 45, (1981), 855-881. Cited on pages: 224, 442.
[81Y0K/YAM]
Yokoyama, H., Yamatera, H., Ion association of some 2:2 electrolytes in water at 25°C. III. A new interpretation of experimental result and the determination of formation constants of inner-sphere ion-pairs, Bull. Chem. Soc. Jpn., 54, (1981), 22862289. Cited on pages: 184, 397.
[82BRO]
Brown, Jr., G. E., Reviews in mineralogy, Orthosilicates, 2nd ed., Mineralogical Society of America, Ribbe, P. H., Ed., vol. 5, pp.275381, Bookcrafters Inc., Chelsea, Michigan, (1982). Cited on page: 239.
[82CAM]
Caminiti, R., Nickel and cadmium phosphates in aqueous solution. Cation-anion complex formation and phosphate - H2O interactions, J. Chem. Phys., 11, (1982), 5682-5686. Cited on page: 205.
[82CHI/SAB]
Chickerur, N. S., Sabat, B. B., Mahapatra, P. P., Solubility of trinickel phosphate octahydrate in aqueous media, Indian J. Chem., 21A, (1982), 924-925. Cited on pages: 205, 397.
[82DRE]
Drever, J. I., The geochemistry of natural waters, Prentice-Hall, Englewood Cliffs, N. J., (1982), 388 p. Cited on page: 24.
[82FER/GOK]
Ferrante, M. J., Gokcen, N. A., Relative enthalpies of Ni3S2, United States Department of Interior, Bureau of Mines, Report RI 8745, (1982). Cited on pages: 167, 316, 398, 420,421.
[82GAM/REI]
Gamsjager, H., Reiterer, F., Heindl, R., Solubility constants and free enthalpies of metal sulphides and carbonates, Ber. Bunsen-Ges., 86, (1982), 1046-1049. Cited on pages: 218, 398.
[82GLU/GUR]
Glushko, V. P., Gurvich, L. V., Bergman, G. A., Veits, I. V., Medvedev, V. A., Khachkuruzov, G. A., Yungman, V. S., Thermodynamic properties of individual substances, Glushko, V. P., Ed., Vol. IV, Nauka, Moscow, USSR, (1982), 623 pp., in Russian. Cited on pages: 249, 250, 386.
BIBLIOGRAPHY
553
[82HAM]
Hamann, S. D., The influence of pressure on ionization equilibria in aqueous solutions, J. Solution Chem., 11, (1982), 63-68. Cited on page: 38.
[82KLO]
Kloger, W., Beitrag zur Thermodynamik des Ni(II)-carbonates, Ph. D. Thesis, University of Leoben, (1982), Austria, in German. Cited on pages: 218,398,441.
[82LAF]
Laffitte, M , A report of IUPAC commission 1.2 on thermodynamics: Notation for states and processes, significance of the word "standard" in chemical thermodynamics, and remarks on commonly tabulated forms of thermodynamic functions, J. Chem. Thermodyn., 14, (1982), 805-815. Cited on pages: 16, 17, 31, 32, 35, 443.
[82LIV/BIS]
Livingstone, A., Bish, D. L., On the new mineral theophrastite, a nickel hydroxide, from Unst, Shetland, Scotland, Mineral. Mag., 46, (1982), 1-5. Cited on pages: 108, 398.
[82MAG/PAS]
Magini, M , Paschina, G., Piccaluga, G., Ni-Cl bonding in concentrated Ni(II)aqueous solutions at high Cl /Ni2+ ratios. An xray diffraction investigation, J. Chem. Phys., 76, (1982), 1116-1121. Cited on page: 141.
[82MU/PER]
Mu, J., Perlmutter, D. D., Thermal decomposition of metal nitrates and their hydrates, Thermochim. Ada, 56, (1982), 253-260. Cited on page: 200.
[82NEV/ANG]
Neves, E. A., Angnes, L., Chierice, G. O., Mazo, L. H., Potentiometric study of the association constant of isothiocyanic acid, Talanta, 29, (1982), 335-337. Cited on page: 232.
[82SAD/LIB]
Sadowska, T., Libus, W., Thermodynamic properties and solution equilibria of aqueous bivalent transition metal nitrates and magnesium nitrate, J. Solution Chem., 11, (1982), 457-468. Cited on pages: 203,204.
[82SAR/COV]
Sarbar, M , Covington, A. K., Nuttall, R. L., Goldberg, R. N., Activity and osmotic coefficients of aqueous nickel(II) nitrate solutions, J. Chem. Thermodyn., 14, (1982), 537-545. Cited on pages: 203,204.
554
BIBLIOGRAPHY
[82SMI/WOO]
Smith-Magowan, D., Wood, R. H., Tillett, D. M., Heat capacities of aqueous solutions of NiCl2 and NiCl2.2NaCl from 0.12 to 30 mol.kg"1 and 321 to 572 K at a pressure of 17.7 MPa, J. Chem. Eng. Data, 27, (1982), 335-342. Cited on pages: 86, 87.
[82STO]
Stolyrova, T. A., Enthalpy of Formation of Co2As and Ni2As, Geokhimiya, 8, (1982), 1211-1213. Cited on page: 212.
[82TAY]
Taylor, J. R., An introduction to error analysis: The study of uncertainties in physical measurements, University Science Books, Mill Valley, CA, USA, (1982), 270 p. Cited on page: 474.
[82WAG/EVA]
Wagman, D. D., Evans, W. H., Parker, V. B., Schumm, R. H., Halow, I., Bailey, S. M , Churney, K. L., Nuttall, R. L., The NBS tables of chemical thermodynamic properties: Selected values for inorganic and C, and C2 organic substances in (SI) units, J. Phys. Chem. Ref. Data, 11, Suppl. 2, (1982), 1-392. Cited on pages: 27, 31, 32, 80, 81, 82, 103, 135, 186, 253, 255, 259, 276, 305, 316, 386, 403, 404, 436.
[82WAK/ICH]
Wakita, H., Ichihashi, M., Mibuchi, T., Masuda, I., The structure of nickel(II) bromide in highly concentrated aqueous solution by x-ray diffraction analysis, Bull. Chem. Soc. Jpn., 55, (1982), 817-821. Cited on pages: 141,142.
[82WAT]
Watanabe, H., Thermochemical properties of synthetic highpressure compounds relevant to the earth's mantle, High-pressure research in geophysics, Akimoto, S. and Manghnani, M. H., Eds., pp.441-446, Reidel: Dordrecht, (1982). Cited on pages: 239, 399.
[82WHI/STA]
White, M. A., Staveley, L. A. K., A thermodynamic study of the magnetic phase transitions of NiBr2, J. Phys. C: Solid State Phys., 15, (1982), L169-L173. Cited on pages: xxi, xxv, 132, 133, 386, 387, 399, 400.
[83AVR/BLO]
Avramenko, N. I., Blokhin, V. V., Kul'vinova, L. A., Mironov, L. A., Study of the formation of nickel(II) monofluoride complexes in water-ethanol solutions, Koord. Khim., 9, (1983), 658-659, in Russian. Cited on pages: 142, 143, 396, 399.
BIBLIOGRAPHY
555
[83ECO/MAR]
Economou, M., Marcopoulos, T., Genesis of the new mineral theophrastite, Ni(OH)2) Chem. Erde, 42, (1983), 53-56. Cited on pages: 108,397,399.
[83FER/GRE]
Ferri, D., Grenthe, I., Salvatore, F., Studies on metal carbonate equilibria. 7. Reduction of the tris(carbonato)dioxouranate(VI) ion, UO2(CO3)3~, in carbonate solutions, lnorg. Chem., 22, (1983), 31623165. Cited on page: 457.
[83GAU/BAL]
Gauthier, M., Bale, C. W., Oxide-sulfate equilibria in the Mg, Ni and Mn systems measured by K2SO4 concentration cell, Metail. Trans., 14B, (1983), 117-124. Cited on pages: 190, 194.
[83GOW/DOD]
Gower, S. A., Dodgen, H. W., Hunt, J. P., Inner-sphere coordination of Cl" by Ni2+ as determined by 35C1 NMR, lnorg. Chem., 22, (1983), 1952-1953. Cited on pages: 141,400.
[83HOL/MES]
Holmes, H. F., Mesmer, R. E., Isopiestic studies of aqueous solutions at elevated temperatures. VII. MgSO4 and NiSO4, J. Chem. Thermodyn., 15, (1983), 709-719. Cited on pages: 181, 182,401.
[83KAT/SUG]
Katsumata, K., Sugiyama, K., Date, M., Experimental observation of the antiferro-fan transition in NiBr2, J. Phys. Soc. Jpn., 52, (1983), 3312-3314. Cited on page: 132.
[83KIM/IMA]
Kim, T. P., Imanakunov, B. I., Kazybaev, S. A., Interaction of dimethylformamide with some nickel(II) and zinc(II) halides in aqueous media at 25°C, Russ. J. lnorg. Chem., 28, (1983), 122-124. Cited on pages: 128,423.
[83LIC/PAS]
Licheri, G., Paschina, G., Piccaluga, G., Pinna, G., EXAFS study of Ni-Cl bonding in Ni(II) aqueous solutions at increasing Cl /Ni2+ ratios, J. Chem. Phys., 79, (1983), 2168-2171. Cited on page: 141.
[83NEI/END]
Neilson, G. W., Enderby, J. E., The structure of an aqueous solution of nickel chloride, Proc. R. Soc. London, A390, (1983), 353-371. Cited on page: 141.
[83RAM/FAN]
Ramette, R. W., Fan, G., Copper(II) chloride complex equilibrium constants, lnorg. Chem., 22, (1983), 3323-3326. Cited on page: 413.
556
BIBLIOGRAPHY
[83SOL/BON]
Solomon, L. R., Bond, A. M., Bixler, J. W., Hallenbeck, D. R., Logsdon, K. M., Stability of monofluoride complexes of the IrvingWilliams series acceptors in methanol, Inorg. Chem., 22, (1983), 1644-1648. Cited on pages: 142, 143, 354, 401.
[83SPA]
Spahiu, K.., Carbonate complex formation in lanthanoid and actinoid systems, Ph. D. Thesis, The Royal Institute of Technology, (1983), Stockholm, Sweden. Cited on pages: 457, 464, 465.
[84ANA/ATK]
Ananthaswamy, J., Atkinson, G., Thermodynamics of concentrated electrolyte mixtures. 4. Pitzer-Debye-Hiickel limiting slopes for water from 0 to 100°C and from 1 atm to 1 kbar, J. Chem. Eng. Data, 29, (1984), 81-87. Cited on page: 451.
[84BRA/DEL]
Braconnier, J. J., Delmas, C , Fouassier, C , Figlarz, M., Beaudouin, B., Hagenmuller, P., A novel nickel (+11) hydroxide obtained by soft chemistry, Rev. Chim. Miner., 21, (1984), 496-508. Cited on pages: 109,401.
[84COM/PRA]
Comert, H., Pratt, J. N., The standard molar Gibbs free energy of formation of NiO from high-temperature e.m.f. measurements, J. Chem. Thermodyn., 16, (1984), 1145-1148. Cited on pages: 105, 401.
[84FOU/CRI]
Fouillac, C , Criaud, A., Carbonate and bicarbonate trace metal complexes: critical reevaluation of stability constants, Geochem. J., 18, (1984), 297-303. Cited on pages: 224, 442.
[84FRE/LAN]
Handbook on the physics and chemistry of the actinides, Freeman, A. J. and Lander, G. H., Eds., vol. 1, North-Holland, Amsterdam, (1984), 515 p. Cited on page: 4.
[84FRE]
Freeman, R. D., Conversion of standard (1 atm) thermodynamic data to the new standard state pressure, 1 bar (105 Pa), J. Chem. Eng. Data, 29, (1984), 105-111. Cited on page: 32.
[84HOS]
Hostettler, J. D., Electrode electrons, aqueous electrons, and redox potentials in natural waters, Am. J. Sci., 284, (1984), 734-759. Cited on page: 24.
BIBLIOGRAPHY
557
[84LAV/TIM]
Lavut, E. G., Timofeyev, B. I., Yuldasheva, V. M , Galchenko, G. L., Enthalpy of formation of nickel chloride, J. Chem. Thermodyn., 16, (1984), 519-522. Cited on pages: 125, 126, 381, 401, 411.
[84MAR/MES]
Marshall, W. L., Mesmer, R. E., Pressure-density relationships and ionization equilibria in aqueous solutions, J. Solution Chem., 13, (1984), 383-391. Cited on page: 38.
[84PAN/STU]
Pankratz, L. B., Stuve, J. M., Gokcen, N. A., Thermodynamic data for mineral technology, US Bureau of Mines, Dpt. of the Interior, Washington D.C., Report 677, (1984), 355 p. Cited on page: 121.
[84PAN]
Pankratz, L. B., Thermodynamic properties of halides, US Bureau of Mines, Dpt. of the Interior, Washington D.C., Report 671, (1984), 841 p. Cited on page: 121.
[84RAY/PET]
Ray, D. K., Petot-Ervas, G., Farhi, R., Petot, C , Influence of the temperature on the heat of formation of nickel monoxide at high temperature, Solid State Commun., 52, (1984), 207-211. Cited on page: 105.
[84ROB/HEM]
Robie, R. A., Hemingway, B. S., Ito, J., Krupka, K. M., Heat capacity and entropy of Ni2SiCvolivine from 5 to 1000 K and heat capacity of Co2SiO4 from 360 to 1000 K, Am. Mineral, 69, (1984), 1096-1101. Cited on pages: 239, 240, 241, 244, 399, 402, 410.
[84ROG/BOR]
R6g, G., Borchardt G., Thermodynamics of nickel orthosilicate, J. Chem. Thermodyn., 16, (1984), 1103-1105. Cited on pages: 243, 244, 402, 437.
[84VAS/VAS]
Vasil'ev, V. P., Vasil'eva, V. N., Dmitrieva, N. G., Kokurin, N. I., The standard enthalpy of formation of Ni2+ in aqueous solution at 298.15 K, Zh. Neorg. Khim., 29, (1984), 1123-1127, in Russian. Cited on pages: 83,402.
[85BAR/PAR]
Bard, A. J., Parsons, R., Jordan, J., Standard Potentials in Aqueous Solution, Bard, A. J. and Parsons, R. and Jordan, J., Eds., International Union of Pure and Applied Chemistry. Marcel Dekker, Inc., New York, (1985), 834 p. Cited on page: 24.
558
BIBLIOGRAPHY
[85DRA/MAD]
Drakin, S. I., Madanat, A. T., Karlina, O., Heat capacity of nickel sulphate (NiSC^) and lanthanum trinitrate aqueous solutions at 50, 70 and 90°C, Tr. Inst.-Mosk. Khim.-Tekhnol. lnst. im. D. I. Mendeleeva, 136, (1985), 47-51, in Russian. Cited on pages: 87, 181, 190,403.
[85DYR]
Dyrssen, D., Metal complex formation in sulphidic seawater, Mar. Chem., 15, (1985), 285-293. Cited on page: 177.
[85FER/GRE]
Ferri, D., Grenthe, I., Hietanen, S., Naer-Neumann, E., Salvatore, F., Studies on metal carbonate equilibria: 12. Zinc(II) carbonate complexes in acid solution, Acta Chem. Scand., A39, (1985), 347353. Cited on page: 464.
[85FIL/CHA]
Filippov, V. K., Charykov, N. A., Solechnik, N. D., Thermodynamics of the systems NiCl, SO4-H2O and CoCl, SO4H2O at 25°C, Zh. Prikl. Khim. (S.-Peterburg), 58, (1985), 18111814. Cited on pages: 128,423.
[85FRE/KEL]
Handbook on the physics and chemistry of the actinides, Freeman, A. J. and Keller, C , Eds., vol. 3, North-Holland, Amsterdam, (1985), 520 p. Cited on page: 4.
[85FRE/LAN]
Handbook on the physics and chemistry of the actinides, Freeman, A. J. and Lander, G. H., Eds., vol. 2, North-Holland, Amsterdam, (1985), 503 p. Cited on page: 4.
[85MAG/MOR]
Magini, M , De Moras, M., Licheri, G., Piccaluga, G., Composition of the first coordination sphere of Ni2+ in concentrated aqueous NiBr2 solutions by x-ray diffraction, J. Chem. Phys., 83, (1985), 5797-5801. Cited on pages: 141,142.
[85MEH/TAR]
Mehrotra, G. M., Tare, V. B., Wagner, Jr., J. B., The standard Gibbs energy of formation of Ni3+.XS2, J. Electrochem. Soc, 132, (1985), 247-250. Cited on pages: 403, 407.
[85MUL]
Muller, A. B., (1985), Private communication, OECD Nuclear Energy Agency, Paris. Cited on page: 4.
BIBLIOGRAPHY
559
[85SAP/CHI]
Saprykova, Z. A., Chichirova, N. D., PMR study of the formation of chloride complexes of copper(II), cobalt(II) and nickel(II) in hydrochloric acid, water-acetonitrile, water-acetone and waterdioxane media, Zh. Neorg. Khim., 30, (1985), 939-943, in Russian. Cited on page: 146.
[85SKE/MAI]
Skeaff, J. M., Mainwaring, P. R., Speelman, J. L., Thermochemical arsenide-oxide equilibria in the system Ni-As-O determined electrochemically between 438 and 866°C, Can. Metall. Q., 24, (1985), 349-382. Cited on pages: 210, 211, 212, 355, 403,404,425.
[85SOH/NOV]
Sohnel, O., Novotny, P., Densities of aqueous solutions of inorganic substances, Elsevier, Amsterdam, (1985), 335 p. Cited on pages: 29, 30.
[86BHA/MUK]
Bhattacharya, S., Mukherjee, R., Chakravorty, A., A nickel(III) complex with a NiO6 coordination sphere, Inorg. Chem., 25, (1986), 3448-3452. Cited on pages: 89, 405.
[86BJE/SKI]
Bjerrum, J., Skibsted, L. H., Weak chloro complex formation by copper(II) in aqueous chloride solution, Inorg. Chem., 25, (1986), 2479-2481. Cited on page: 413.
[86BRU]
Bruno, J., Stoichiometric and structural studies on the Be2+-H2OCO2(g) system, Ph. D. Thesis, The Royal Institute of Technology, (1986), Stockholm, Sweden. Cited on page: 465.
[86CEM/KLE]
Cemic, L., Kleppa, O. J., High temperature calorimetry of sulfide systems. I. Thermochemistry of liquid and solid phases of Ni + S, Geochim. Cosmochim. Ada, 50, (1986), 1633-1641. Cited on pages: 167, 169, 171, 172, 173, 405, 409, 423.
[86COD]
CODATA, The 1986 adjustment of the fundamental physical constants. A report of the CODATA Task Group on Fundamental Constants, prepared by Cohen, E. R. and Taylor, B. N., Pergamon Journals, Oxford, CODATA Bulletin, Report 63, (1986), 36 p. Cited on pages: 35, 36.
[86FIL/CHA]
Filippov, V. K., Charykov, N. A., Fedorov, Y. A., The NaCl-NiCl2H2O and NaCl-CuCl2-H2O systems at 25°C, Russ. J. Inorg. Chem., 31, (1986), 1071-1074. Cited on pages: 128, 423.
560
BIBLIOGRAPHY
[86FRE/KEL]
Handbook on the physics and chemistry of the actinides, Freeman, A. J. and Keller, C , Eds., vol. 4, North-Holland, Amsterdam, (1986), 567 p. Cited on page: 4.
[86HOL/NEI]
Holmes, R. D., O'Neill, H. S. C , Arculus, R. J., Standard Gibbs free energy of formation for Cu2O, NiO, CoO, and FexO: High resolution electrochemical measurements using zirconia solid electrolytes from 900-1400K, Geochim. Cosmochim. Acta, 50, (1986), 2439-2452. Cited on pages: 105,406.
[86HSI/CHA]
Hsieh, K., Chang, Y. A., A solid-state Emf study of ternary Ni-S-O, Fe-S-O, and quaternary Fe-Ni-S-O, Metall. Trans. B, 17, (1986), 133-146. Cited on pages: 406, 407, 408.
[86JAC/KAL]
Jacob, K. T., Kale, G. M , Ramachandran, A., Shukla, A. K., Gibbs' energy of formation of nickel orthosilicate (Ni2SiO4), High Temp. Mater. Processes (London, U.K.), 7, (1986), 141-148. Cited on pages: 243,408.
[86KAT/SEA]
The chemistry of the actinide elements, 2nd. Edition, Katz, J. J. and Seaborg, G. T. and Morss, L. R., Eds., Chapman and Hall, London, (1986), 1674 p. Cited on page: 4.
[86NOR/MUN]
Nordstrom, D. K., Munoz, J. L., Geochemical Thermodynamics, Blackwell Sci. Publ., Palo Alto, (1986). Cited on page: 24.
[86RAM]
Ramette, R. W., Copper(II) complexes with chloride ion, Inorg. Chem., 25, (1986), 2481-2482. Cited on page: 413.
[86TOD/HAS]
Toda, Y., Hashimoto, K., Hashimoto, K., Arai, Y., Hydrothermal synthesis and particle properties of spherical crystals of nickel(II) phosphate hydrate, Shikizai Kyokaishi, 59, (1986), 725-731. Cited on page: 204.
[86VAS/DMI]
Vasil'ev, V. P., Dmitrieva, N. G., Vasil'eva, V. N., Yashkova, V. I., Belokurova, N. A., The standard enthalpy of formation of Ni(II) ion in aqueous solution at 298.15 K, Zh. Neorg. Khim., 31, (1986), 3044-3047, in Russian. Cited on pages: 83, 408.
[87BEC]
Beck, M. T., Critical survey of stability constants of cyano complexes, Pure Appl. Chem., 59, (1987), 1703-1720. Cited on pages: 231,409.
BIBLIOGRAPHY
561
[87BJE]
Bjerrum, I , Determination of small stability constants. A spectrophotometric study of copper(II) chloride complexes in hydrochloric acid, Acta Chem. Scand, A41, (1987), 328-334. Cited on pages: 409,413.
[87BOS]
Bostrom, D., Single-crystal X-ray diffraction studies of synthetic Ni-Mg olivine solid solutions, Am. Mineral, 72, (1987), 965-972. Cited on page: 239.
[87CEM/KLE]
Cemic, L., Kleppa, O. J., High temperature calorimetry of sulfide systems. II. Standard enthalpies of formation of pentlandite and violarite, Phys. Chem. Miner., 14, (1987), 52-57. Cited on pages: 174,409.
[87CIA/IUL]
Ciavatta, L., Iuliano, M., Porto, R., The hydrolysis of the La(III) ion in aqueous perchlorate solution at 60°C, Polyhedron, 6, (1987), 1283-1290. Cited on page: 40.
[87CUD/LEC]
Cudennec, Y., Leserf, A., Riou, A., Gerault, Y., Syntheses et etude d'hydrogeno-phosphates de formule MHPO4.3H2O et (MHPO4)2.3H2O (M=Co, Ni, Zn), Rev. Chim. Miner., 24, (1987), 234-247. Cited on page: 204.
[87EMA/FAR]
Emara, M. M., Farid, N. A., Shehata, H. A., Ionic association of transition metal ions with bicarbonate using spectrophotometric method. II. Nickel and cobalt bicarbonates in aqueous and aqueous alcohol mixtures, J. Indian Chem. Soc, 64, (1987), 119-122. Cited on pages: 223, 224, 409, 442.
[87FLE]
Fleet, M. E., Structure of godlevskite, Ni9Sg, Acta Crystallogr., Sect. B: Struct. Set, C43, (1987), 2255-2257. Cited on pages: 161, 162, 410.
[87FRE/LAN]
Handbook on the physics and chemistry of the actinides, Freeman, A. J. and Lander, G. H., Eds., vol. 5, North-Holland, Amsterdam, (1987), 375 p. Cited on page: 4.
[87GAR/PAR]
Garvin, D., Parker, V. B., White, Jr., H. J., CODATA thermodynamic tables: Selection of some compounds of calcium and related mixtures: A prototype set of tables, Springer-Verlag, Berlin, (1987), 356 p. Cited on pages: 28, 249, 410.
562
BIBLIOGRAPHY
[87NEI]
O'Neill, H. S. C , Free energies of formation of NiO, CoO, Ni 2 Si0 4 , and Co2SiO4, Am. Mineral, 11, (1987), 280-291. Cited on pages: 239,241,242,243,244,410.
[87RAR3]
Rard, J. A., Isopiestic investigation of the osmotic and activity coefficients of aqueous NiCl2, Pr(NO3)3, and Lu(NO3)3 and solubility of NiCl2 at 25°C, J. Chem. Eng. Data, 32, (1987), 334341. Cited on pages: 128,423.
[87RIG/VIT]
Riglet, C , Vitorge, P., Grenthe, I., Standard potentials of the (MC>2+ /MOj) systems for uranium and other actinides, Inorg. Chim. Acta, 133, (1987), 323-329. Cited on page: 464.
[87SIN/NAS]
Singleton, M., Nash, P., The As-Ni (arsenic-nickel) system, Bull. Alloy Phase Diagrams, 8, (1997), 419-422. Cited on pages: 210, 212.
[88BJE]
Bjerrum, J., Estimation of small stability constants in aqueous solution. The nickel(II)-chloride system, Acta Chem. Scand., A42, (1988), 714-716. Cited on pages: 140, 148, 151, 152,388,410,412, 416.
[88BJE2]
Bjerrum, J., Estimation of small stability constants in aqueous solution. The fourth consecutive stability constant in the nickel(II)thiocyanate system, Acta Chem. Scand., A42, (1988), 721-723. Cited on pages: 232,233,235,411.
[88CIA]
Ciavatta, L., (1988), Universita di Napoli, Naples, Italy, Private Communication, (from the citation in [92GRE/FUG]). Cited on pages: xxx, 460, 467,470.
[88DYR]
Dyrssen, D., Sulfide complexation in surface seawater, Mar. Chem., 24, (1988), 143-153. Cited on page: 177.
[88EFI/EVD]
Efimov, M. E., Evdokimova, V. I., Medvedev, V. A., Zirel'nikov, V. I., The standard enthalpy of formation of crystalline nickel chloride, Zh. Fiz. Khim., 62, (1988), 1961-1963, in Russian. Cited on pages: 83, 125, 137,411,416,417.
BIBLIOGRAPHY
563
[88HED]
Hedlund, T., Studies of complexation and precipitation equilibria in some aqueous aluminium(III) systems, Ph. D. Thesis, University of Umea , (1988), Sweden. Cited on page: 464.
[88LIC2]
Licht, S., Aqueous solubilities, solubility products and standard oxidation-reduction potentials of the metal sulfides, J. Electrochem. Soc, 135, (1988), 2971-2975. Cited on pages: 180,181,411.
[88NIS/ITO]
Nishimura, T., Itoh, C. T., Tozawa, K., Stabilities and solubilities of metal arsenites and arsenates in water and the effect of sulfate and carbonate ions on their solubilities, Arsenic metallurgy fundamental and applications, pp.77-92, (1988). Cited on pages: 212,213,412.
[88SHO/HEL]
Shock, E. L., Helgeson, H. C , Calculation of the thermodynamic and transport properties of aqueous species at high pressures and temperatures: Correlation algorithms for ionic species and equation of state predictions to 5 kb and 1000°C, Geochim. Cosmochim. Ada, 52, (1988), 2009-2036. Cited on pages: 38,451.
[88TAN/HEL]
Tanger, IV, J. C , Helgeson, H. C , Calculation of the thermodynamic and transport properties of aqueous species at high pressures and temperatures: Revised equations of state for the standard partial molal properties of ions and electrolytes, Am. J. Sci., 288,(1988), 19-98. Cited on pages: 38,451.
[88WAN]
Wanner, H., The NEA Thermochemical Data Base Project, Radiochim. Ada, 44/45, (1988), 325-329. Cited on page: 4.
[89ART/KAS]
Artamonova, E. A., Kasenov, B. K., Thermochemistry of nickel orthoarsenate octahydrate, Izv. Vyss. Uchebn. Zaved., Tsvetn. Metall.,2, (1989), 112-113. Cited on page: 213.
[89BAL]
Balej, J., Solubility in the system NiCl2-ZnCl2-H3BOrH2O at 15, 20 and 25°C, J. Chem. Soc. Faraday Tram. 1, 85, (1989), 3327-3334. Cited on page: 246.
[89BJE]
Bjerrum, J., Problems concerning the determination of small stability constants of aniono complexes in aqueous solution, Coord. Chem. Rev., 94, (1989), 1-15. Cited on pages: 140, 151, 154, 348, 350,411,412.
564
BIBLIOGRAPHY
[89BRA]
Bratsch, S. G., Standard electrode potentials and temperature coefficients in water at 298.15 K, J. Phys. Chem. Ref. Data, 18, (1989), 1-21. Cited on pages: 80, 82, 89, 276, 413.
[89COX/WAG]
Cox, J. D., Wagman, D. D., Medvedev, V. A., CODATA Key Values for Thermodynamics, Hemisphere Publ. Corp., New York, (1989), 271 p. Cited on pages: 6, 31, 37, 39, 53, 54, 55, 80, 81, 83, 106, 107, 117, 125, 127, 221, 242, 249, 250, 261, 266, 274, 275, 278, 281,327,403.
[89EVD/EFI]
Evdokimova, V. P., Efimov, M. E., Standard enthalpy of formation of crystalline nickel bromide, Russ. J. Phys. Chem., 63, (1989), 436437. Cited on pages: 83, 134, 413.
[89HOV/HEP]
Hovey, J. K., Hepler, L. G., Apparent and partial molar heat capacities and volumes of aqueous HCIO4 and HNO3 from 10 to 55°C, Can. J. Chem., 67, (1989), 1489-1495. Cited on page: 87.
[89IUL/POR]
Iuliano, M , Porto, R., Vasca, E., Chloride complexes of copper(II) and nickel(II) ions in 6 m NaClO4, Ann. Chim. (Rome), 79, (1989), 439-454. Cited on pages: 148, 149, 150, 152, 158, 414.
[89K.IT]
Kitayama, K., Thermogravimetric study of the M-Ni-O system. I. M = Ta and Nb, J. Solid State Chem., 83, (1989), 37-44. Cited on page: 107.
[89LIL/TEP]
Lileev, A. S., Tepavitcharova, S., Lyashchenko, A. K., Balarev, K., Dielectric properties of saturated solutions and the solubility diagram of the CsCl-NiCl2-H2O system, Russ. J. lnorg. Chem., 34, (1989), 281-284. Cited on page: 128.
[89LOB/QUA]
Lobo, V. M. M , Quaresma, J. L., Handbook of electrolyte solutions, part B, Elsevier, Amsterdam, (1989). Cited on page: 360.
[89LOB]
Lobo, V. M. M., Handbook of electrolyte solutions, Part A, Elsevier Science Publishing Company Inc., New York, (1989). Cited on page: 275.
[89RIG/ROB]
Riglet, C , Robouch, P., Vitorge, P., Standard potentials of the (MO^/MOj) and (M4+/M3+) redox systems for neptunium and plutonium, Radiochim. Ada, 46, (1989), 85-94. Cited on pages: 464, 465.
BIBLIOGRAPHY
565
[89ROB]
Robouch, P., Contribution a la prevision du comportement de l'americium, du plutonium et du neptunium dans la geosphere; donnees chimiques, Commissariat a l'Energie Atomique, Report CEA-R-5473, (1989). Cited on pages: 464, 465.
[89SHO/HEL]
Shock, E. L., Helgeson, H. C , Corrections to Shock and Helgeson Geochimica et Cosmochimica Acta, 52, (1988) 2009-2036, Geochim. Cosmochim. Acta, 53, (1989), 215. Cited on pages: 38, 451.
[89SHO/HEL2]
Shock, E. L., Helgeson, H. C , Sverjensky, D. A., Calculation of the thermodynamic and transport properties of aqueous species at high pressures and temperatures: Standard partial molal properties of inorganic neutral species, Geochim. Cosmochim. Acta, 53, (1989), 2157-2183. Cited on pages: 38,451.
[89ZIE/JON]
Ziemniak, S. E., Jones, M. E., Combs, K. E. S., Solubility and phase behavior of nickel oxide in aqueous sodium phosphate solutions at elevated temperatures, J. Solution Chem., 18, (1989), 1133-1152. Cited on pages: 100, 101, 102, 107, 115, 205, 206, 414.
[90ARC/WAN]
Archer, D. G., Wang, P., The dielectric constant of water and Debye-Huckel limiting law slopes, J. Phys. Chem. Ref. Data, 19, (1990), 371-411. Cited on page: 451.
[90BEC/NAG]
Beck, M. T., Nagypal, I., Chemistry of complex equilibria, 2nd. Edition, Horwood Limited Publishers, New York, (1990), 402 p. Cited on pages: 26, 141.
[90BJE]
Bjerrum, J., Estimation of small stability constants in aqueous solution. The nickel(II)-bromide system, Acta Chem. Scand., 44, (1990), 401-403. Cited on pages: 140, 153, 154, 156, 412, 416.
[90BJE2]
Bjerrum, J., The cobalt(II) and nickel(II) thiocyanate systems: a spectrophoto-metric study, Coord. Chem. Rev., 10, (1990), 105-118. Cited on pages: 233,315,416.
[90CIA]
Ciavatta, L., The specific interaction theory in equilibrium analysis: Some empirical rules for estimating interaction coefficients of metal ion complexes, Ann. Chim. (Rome), 80, (1990), 255-263. Cited on pages: 444,457,458.
566
BIBLIOGRAPHY
[90COR/KON2]
Cordfunke, E. H. P., Konings, R. J. M., Ouweltjes, W., The standard enthalpies of formation of MO(s), MCl2(s), and M2+(aq, oo), (M = Ba, Sr), J. Chem. Thermodyn., 22, (1990), 991-996. Cited on page: 4.
[90DYR/KRE]
Dyrssen, D., Kremling, K., Increasing hydrogen sulfide concentration and trace metal behavior in the anoxic Baltic waters, Mar. Chem., 30, (1990), 193-204. Cited on page: 177.
[90EFI/EVD]
Efimov, M. E., Evdokimova, V. I., The standard enthalpies of formation of crystalline iron and nickel iodides, Zh. Fiz. Khim., 64, (1990), 242-245, in Russian. Cited on pages: 83, 137,416,417.
[90EFI/FUR]
Efimov, M. E., Furkalyuk, M. Y., Standard enthalpies of formation of Co2+ and Ni2+ in aqueous solution, Zh. Fiz. Khim., 64, (1990), 2560-2562, in Russian. Cited on pages: xxix, 82, 83, 134, 135, 137, 321,340,417.
[90GRZ]
Grzybkowski, W., Variation in stability of monohalide complexes and some properties of the solvated cations within the Mn2+-Zn2+ series, Polyhedron, 9, (1990), 2257-2261. Cited on pages: 144, 418.
[90HEM]
Hemingway, B. S., Thermodynamic properties for bunsenite, NiO, magnetite, Fe3C>4, and hematite, Fe2O3, with comments on selected oxygen buffer reactions, Am. Mineral, 75, (1990), 781-790. Cited on page: 104.
[90JUR/DOM]
Juraitis, K. R., Domiciano, J. B., Sano, W., Structural transition study of NiCl2-2H2O by specific heat, Phys. Rev. B: Condens. Matter Mater. Phys., 42, (1990), 2586. Cited on pages: 131,418.
[90LEE/NAS]
Lee, S. Y., Nash, P., Ni-Te, Binary alloy phase diagrams, 2nd edition, Massalski, T. B. and Okamoto, H. and Subramanian, P. R. and Kacprak, L., Eds., vol. 3, pp.2869-2872, ASM International, U.S.A., (1990). Cited on pages: 195,418.
[90LEI]
Leigh, G. J., Nomenclature of inorganic chemistry, recommendations 1990, issued by the Commission on Nomenclature of Inorganic Chemistry, IUPAC, Ed., Blackwell Sci. Publ., Oxford, (1990), 289 p. Cited on page: 15.
BIBLIOGRAPHY
567
[90MON]
Monnin, C , The influence of pressure on the activity coefficients of the solutes and on the solubility of minerals in the system Na-Ca-ClSO4-H2O to 200°C and 1 kbar, and to high NaCl concentration, Geochim. Cosmochim. Acta, 54, (1990), 3265-3282. Cited on page: 38.
[90OEL/HEL]
Oelkers, E. H., Helgeson, H. C , Triple-ion anions and polynuclear complexing in supercritical electrolyte solutions, Geochim. Cosmochim. Acta, 54, (1990), 727-738. Cited on page: 451.
[90OSA/ROS]
Osadchii, E., Rosen, E., Saitton, B., Equilibrium studies of the system Ni-S-O using the solid electrolyte galvanic cell technique, Acta Chem. Scand, 44, (1990), 476-480. Cited on pages: 175, 176, 418.
[90PAS/TAY]
Pasternak, M. P., Taylor, R. D., Chen, A., Meade, C , Falicov, L. M., Giesekus, A., Jeanloz, R., Pressure-induced metallization and the collapse of the magnetic state in the antiferrromagnetic insulator Nil2, Phys. Rev. Lett., 65, (1990), 790-793. Cited on page: 136.
[90RIG]
Riglet, C , Chimie du neptunium et autres actinides en milieu carbonate, Commissariat a 1'Energie Atomique, Report CEA-R5535, (1990), 267 p. Cited on pages: 464, 465.
[90TRO]
Trojan, M., A study of the reactions during formation of c-Ni2P4O|2, Thermochim. Acta, 160, (1990), 361-366. Cited on page: 205.
[91AND/CAS]
Anderson, G. M , Castet, S., Schott, J., Mesmer, R. E., The density model for estimation of thermodynamic parameters of reactions at high temperatures and pressures, Geochim. Cosmochim. Acta, 55, (1991), 1769-1779. Cited on page: 38.
[91 DIN]
Dinsdale, A. T., SGTE data for pure elements, CALPHAD: Comput. Coupling Phase Diagrams Thermochem., 15, (1991), 317-425. Cited on page: 31.
[91FRE/KEL]
Handbook on the physics and chemistry of the actinides, Freeman, A. J. and Keller, C , Eds., vol. 6, North-Holland, Amsterdam, (1991), 742 p. Cited on page: 4.
568
BIBLIOGRAPHY
[91KNA/KUB]
Thermochemical properties of inorganic substances, 2nd. Edition, Knacke, 0. and Kubaschewski, O. and Hesselmann, K., Eds., I and II, Springer Verlag, Berlin, (1991), 2412 p. Cited on pages: 103, 216,238,419.
[91PAP/PIT]
Pabalan, R. T., Pitzer, K. S., Activity coefficients in electrolyte solutions, Mineral solubilities in electrolyte solutions, Pitzer, K. S., Ed., pp.435-490, CRC Press, Boca Raton, Florida, (1991). Cited on page: 221.
[91 PIT]
Pitzer, K. S., Ion interaction approach: theory and data correlation, Activity coefficients in electrolyte solutions, Pitzer, K. S., Ed., pp.75-153, CRC Press, Boca Raton, Florida, (1991). Cited on pages: xxx, 220, 221,448.
[91SPI/TRI]
Spivakovskii, V. B., Trizna, L. G., Homogeneous hydrolyses of nickel ions in chloride solutions, Zh. Neorg. Khim., 36, (1991), 1281-1285, in Russian. Cited on pages: 94,420.
[91STO/GRO]
Stelen, S., Grenvold, F., Westrum, Jr., E. F., Kolonin, G. R., Heat capacity and thermodynamic properties of synthetic heazlewoodite, Ni3S2, and of the high-temperature phase Ni3+.XS2, J. Chem. Thermodyn., 23, (1991), 77-93. Cited on pages: 164, 167, 168, 278, 315,316,398,420,421,427.
[91STO]
Stokes, R. H., Thermodynamics of solutions, Activity coefficients in electrolyte solutions, Pitzer, K. S., Ed., p. 10, CRC Press, Boca Raton, Florida, (1991). Cited on page: 445.
[91SWI/PAW]
Swiatek, J., Pawlowski, T., Polarographic study of Ni2+ and Zn2+ ion interaction with tRNAs of wheat germ. Lupine phenylalanine tRNA and related subunits, J. Inorg. Biochem., 44, (1991), 163-171. Cited on pages: 205,206,421.
[91TAR/FAZ]
Tareen, J. A. K., Fazeli, A. R., Basavalingu, Bhandage, G. T., Hydrothermal decomposition curves and thermodynamic data for spherocobaltite (C0CO3) and gaspeite (NiCC^), Eur. J. Mineral., 3, (1991), 501-505. Cited on pages: xxix, 217, 421, 422.
BIBLIOGRAPHY
569
[91 WAN]
Wanner, H., On the problem of consistency of chemical thermodynamic data bases, Scientific basis for nuclear waste management XIV, vol. 212, pp.815-822, Materials Research Society, (1991). Cited on page: 4.
[91ZEK/NAG]
Zekany, L., Nagypal, I., PSEQUAD: a comprehensive program for the evaluation of potentiometric and/or spectrophotometric equilibrium data using analytical derivatives, Computational methods for the determination of formation constants, Leggett, D. J., Ed., Plenum Press, New York, (1991). Cited on page: 317.
[92BAL/DIC]
Ball, R. G. J., Dickinson, S., Cordfunke, E. H. P., Konings, R. J. M., Drowart, J., Smoes, S., Thermochemical data acquisition - Part II, Commission of the European Communities, Report EUR 14844 EN, (1992). Cited on pages: 195, 196, 423.
[92BAN/BLI]
Banyai, I., Blixt, J., Glaser, J., Toth, I., On the dissociation of hydrogen cyanide in aqueous solutions containing different ionic media. A combined potentiometric and carbon-13 NMR study, Ada Chem. Scand, 46, (1992), 138-141. Cited on pages: 226, 292, 298, 409, 469.
[92CAP]
Capdevila, H., Donnees thermodynamiques sur l'oxydoreduction du plutonium en milieux acide et carbonate. Stabilite de Pu(V), Ph. D. Thesis, Universite de Paris-Sud, Orsay, France, (1992), in French. Also published as CEA-R-5643 Commissariat a l'Energie Atomique, France, (1993). Cited on pages: 464, 465.
[92FRE/FAL]
Freericks, J. K., Falicov, L. M., Thermodynamic model of the insulator-metal transition in nickel iodide, Phys. Rev. B, 45, (1992), 1896-1899. Cited on page: 136.
[92FUG/KHO]
Fuger, J , Khodakovsky, I. L., Sergeyeva, E. I., Medvedev, V. A., Navratil, J. D., The chemical thermodynamics of actinide elements and compounds, Part 12. The actinide aqueous inorganic complexes, International Atomic Energy Agency, Vienna, (1992), 224 p. Cited on page: 4.
570
BIBLIOGRAPHY
[92GRE/FUG]
Grenthe, I., Fuger, J., Konings, R. J. M., Lemire, R. J., Muller, A. B., Nguyen-Trung, C , Wanner, H., Chemical Thermodynamics of Uranium, Wanner, H. and Forest, I., Nuclear Energy Agency, Organisation for Economic Co-operation and Development, Eds., vol. 1, Chemical Thermodynamics, North Holland Elsevier Science Publishers B. V., Amsterdam, The Netherlands, (1992), 715 p. Cited on pages: 4, 6, 7, 27, 29, 39, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 147, 178, 207, 249, 255, 263, 293, 294, 335, 404, 426, 443, 444, 457, 458, 464, 465, 469, 471,482,483.
[92RAR2]
Rard, J. A., Isopiestic investigation of water activities of aqueous NiCl2 and CuCl2 solutions and the thermodynamic solubility product of NiCl2-6H2O at 298.15 K, J. Chem. Eng. Data, 37, (1992), 433442. Cited on pages: 85, 128, 423.
[92SHO/OEL]
Shock, E. L., Oelkers, E. H., Johnson, J. W., Sverjensky, D. A., Helgeson, H. C , Calculation of the thermodynamic properties of aqueous species at high pressures and temperatures. Effective electrostatic radii, dissociation constants and standard partial molal properties to 1000°C and 5 kbar, J. Chem. Soc. Faraday Trans., 88, (1992), 803-826. Cited on page: 451.
[92VID/KOR]
Vidavskii, L. M., Korotkevich, I. I., The standard molar enthalpy of formation of heazlewoodite Ni3S2 determined by its synthesis in a calorimeter, J. Chem. Thermodyn., 24, (1992), 577-584. Cited on pages: 167,423.
[92WHI]
Whiting, K. S., The thermodynamics and geochemistry of arsenic with application to subsurface waters at the Sharon Steel Superfund site at Midvale, Utah, Ph. D. Thesis, Colorado School of Mines, (1992), Dept. of Chemistry and Geochemistry. Cited on page: 213.
[93ALL/BRO]
Allison, J. D., Brown, D. S., Novo-Gradac, K. H., MINTEQA2/PRODEFA2, A Geochemical Assessment Model for Environmental Systems: Version 3.0 User's Manual, U.S. EPA Office of Research and Development, Environmental Research Laboratory, Report EPA/600/3-91/021, (1993). Cited on page: 433.
[93BAL/DIV]
Balej, J., Divisek, J., On the composition of higher solid nickel oxocompounds, Ber. Bunsen-Ges., 97, (1993), 929-933. Cited on pages: 118,423,434.
BIBLIOGRAPHY
571
[93BAL/DIV2]
Balej, J., Divisek, J., On the thennodynamics of higher solid nickel oxocompounds, Ber. Bunsen-Ges., 97, (1993), 933-939. Cited on pages: 118,424,434.
[93GIF/VIT2]
Giffaut, E., Vitorge, P., Capdevila, H., Corrections de temperature sur les coefficients d'activite calcules selon la TIS, Commissariat a l'Energie Atomique, Report CEA-N-2737, (1993), 29 p. Cited on page: 451.
[93KUB/ALC]
Kubaschewski, O., Alcock, C. B., Spencer, P. J., Materials thermochemistry, 6th. Edition, Pergamon Press Ltd., Oxford, (1993), 363 p. Cited on pages: 117,136, 195, 196,216,217,246,371,424.
[93MAK/VOE]
Makarov, G. V., Voevoda, N. Y., Batrakov, V. V., Kudasheva, T. V., Thermodynamics of the solubility of symmetric nickel and cobalt chalcogenides, Russ. J. Phys. Chem., 67, (1993), 1200-1203. Cited on pages: 178,424.
[93MIL/CVI]
Mills, I., Cvitas, T., Homann, K., Kallay, N., Kuchitsu, K., Quantities, units and symbols in physical chemistry. (The green book), IUPAC, Blackwell Scientific Publications, Oxford, (1993), 176 p. Cited on pages: 13,22,26.
[93MUL/ISA]
Muldagalieva, S. A., Isabaev, A. S., Zhambekov, M. I., Isabaev, S. M., Thermodynamic properties of nickel arsenide, High Temp., 31, (1993), 794. Cited on pages: 211, 404, 424, 430.
[93NEI/POW]
O'Neill, H. S. C , Pownceby, M. I., Thermodynamic data from redox reactions at high temperatures. I. An experimental and theoretical assessment of the electrochemical method using stabilized zirconia electrolytes, with revised values for the Fe-FeO, Co-CoO, Ni-NiO and Cu-Cu2O oxygen buffers, and new data for the W-W0 2 buffer, Contrib. Mineral. Petrol, 114, (1993), 296-314. Cited on pages: 105,425.
[93NIC/ROB]
Nickel, E. H., Robinson, B. W., Mumme, W. G., Widgiemoolthalite: the new Ni analogue of hydromagnesite from western Australia, Am. Mineral, 78, (1993), 819-821. Cited on pages: 216, 425.
572
BIBLIOGRAPHY
[93OBE/KAS]
Obendrauf, W., Kaschnitz, E., Pottlacher, G., Jager, H., Measurements of thermophysical properties of nickel with a new highly sensitive pyrometer, Int. J. Thermophys., 14, (1993), 417426. Cited on pages: 79,425.
[93PRZ/WIS]
Przepiera, A., Wisniewski, M., Dabrowski, W., Jablonski, M., Enthalpy of solution and effect of temperature on activity coefficients in the MSO4-H2SO4-H2O system, J. Therm. Anal., 50, (1993), 1139-1143. Cited on pages: 191, 425, 426.
[93SHE]
Shehata, H. A., Studies of formation of some metal(II) sulphate complex using conductometry technique. I- Bivalent ions of the first transition series and alkaline earth metals in aqueous media at 25°C, Ultra Sci. Phys. Sci, 5, (1993), 108-113. Cited on pages: 181, 184, 185,426.
[93UVA/TIM]
Uvaliev, Y. K., Timofeeva, T. G., Boldamdaeva, O. K., Transformation equilibria of nickel chloride hydrates, Izv. Nats. Akad. Nauk resp. Kaz. Ser. Khim., 1, (1993), 23-26. Cited on pages: 130, 131, 132,426.
[93 WAT]
Watanabe, H., Thermal constants for Ni, NiO, MgO, MnO, and CoO at low temperatures, Thermochim. Ada, 218, (1993), 365-372. Cited on pages: 102, 104.
[94CAP/VIT]
Capdevila, H., Vitorge, P., Potentiels redox des couples PuO^/PuOj et Pu4+/Pu3+ a force ionique et temperature variables. Entropie et capacite calorifique, Commissariat a l'Energie Atomique, Report CEA-N-2762, (1994), 73 p. Cited on pages: 464,465.
[94GIF]
Giffaut, E., Influence des ions chlorure sur la chimie des actinides, Ph. D. Thesis, Universite de Paris-Sud, Orsay, France, (1994), 259 pp., in French. Cited on page: 464.
[94HAS/TOD]
Hashimoto, K., Toda, Y., Miura, K., Hashimoto, K., Synthetic process and thermal behaviour of nickel hydrogenphosphate hydrate prepared by wet chemical method, Shikizai Kyokaishi, 67, (1994), 615-623. Cited on page: 205.
[94KAL/FRA]
Kale, G. M., Fray, D. J., Oxygen potentials in Ni + NiO and Ni + Cr2O3 + NiCr2O4 systems, Metall. Mater. Trans. B, 25, (1994), 373378. Cited on page: 105.
BIBLIOGRAPHY
573
[94PLY/GRE]
Plyasunov, A. V., Grenthe, I., The temperature dependence of stability constants for the formation of polynuclear cationic complexes, Geochim. Cosmochim. Ada, 58, (1994), 3561-3582. Cited on page: 99.
[94STE/FOT]
De Stefano, C , Foti, C , Gianguzza, A., Ionic strength dependence of formation constants. Part 19. The effect of tetramethylammonium, sodium and potassium chlorides on the protonation constants of pyrophosphate and triphosphate at different temperatures, J. Chem. Res. (M), (1994), 2639-2661. Cited on pages: 208, 209, 286, 362, 426.
[94STO/FJE]
Stolen, S., Fjellvag, H., Granvold, F., Seim, H., Westrum, Jr., E. F., Phase stability and structural properties of Ni7+.8S6 and Ni9S8. Heat capacity and thermodynamic properties of Ni7S6 at temperatures from 5 K to 970 K and of Ni9S8 from 5 K to 673 K, J. Chem. Thermodyn., 26, (1994), 987-1000. Cited on pages: 164, 168, 169, 174, 175, 269, 277, 281, 392, 426, 433.
[94ZHA/MIL]
Zhang, J.-Z., Millero, F. J., Investigation of metal sulfide complexes in sea water using cathodic stripping square wave voltammetry, Anal. Chim. Ada, 284, (1994), 497-504. Cited on pages: 177, 178, 428, 432.
[95CAP/VIT]
Capdevila, H., Vitorge, P., Redox potentials of PuO^ / PuOj and Pu / Pu at different ionic strengths and temperatures. Entropy and heat capacity, Radiochim. Acta, 68, (1995), 51-62. Cited on pages: 464, 465.
[95GRE/PUI]
Grenthe, I., Puigdomenech, I., Rand, M. H., Sandino, M. C. A., Corrections to the Uranium NEA-TDB review, Appendix D, in: Chemical Thermodynamics of Americium (authored by Silva, R. J., Bidoglio, G., Rand, M. H. and Robouch, P. B., Wanner, H., Puigdomenech, I.), Nuclear Energy Agency, Organisation for Economic Co-operation and Development, Ed., pp.347-374, North Holland Elsevier Science Publishers B. V., Amsterdam, The Netherlands, (1995). Cited on page: 469.
574
BIBLIOGRAPHY
[95GRO/STO]
Granvold, F., Stelen, S., Heat capacity and thermodynamic properties of millerite from 298.15 to 660 K and NiAs-type nickel(II) sulfide from 260 to 1000 K. Thermodynamics of the NiAs-type to millerite transition, Thermochim. Ada, 266, (1995), 213-229. Cited on pages: 164, 171, 172,315,392,428.
[95KAH/CRO]
Kahn, M. A., Cronier, D., Bouet, G., Vierling, F., Transition Met. Chem (London), 20, (1995). Cited on page: 340.
[95KON/ERI]
Konigsberger, E., Eriksson, G., A new optimization routine for chemsage, CALPHAD: Comput. Coupling Phase Diagrams Thermochem., 19, (1995), 207-214. Cited on page: 438.
[95LEM]
Lemire, R. J., The solubility of nickel and copper(II) oxides in water and aqueous phosphate solutions at high temperatures: a reanalysis of the data of Ziemniak, Jones and Combs, Atomic Energy of Canada Limited Research Company, Report RC-1330, (1995). Cited on pages: 101, 102, 206, 415.
[95MAN/KOR]
Manin, N. G, Korolev, V. P., Thermochemistry of NiCl2 dissolution in aqueous electrolyte solution, Russ. J. Inorg. Chem., 40, (1995), 1007-1011. Cited on pages: xxix, 82, 83, 129, 253, 338, 359, 417, 429, 430.
[95MUL/CHU]
Muldagalieva, R. A., Chunaeva, V. D., Isabaev, S. M., Heat capacity and thermodynamic properties of nickel monoarsenide, High Temp., 33, (1995), 642-643. Cited on pages: 210, 211, 356, 404, 430.
[95ROB/HEM]
Robie, R. A., Hemingway, B. S., Thermodynamic properties of minerals and related substances at 298.15 K and 1 bar (105 Pascals) pressure and at higher temperatures, US Geol. Surv. Bull., 2131, (1995). Cited on pages: 78, 103, 106, 107, 175, 177,217,242.
[95SIL/BID]
Silva, R. J., Bidoglio, G., Rand, M. H., Robouch, P., Wanner, H., Puigdomenech, I., Chemical Thermodynamics of Americium, Nuclear Energy Agency, Organisation for Economic Co-operation and Development, Ed., vol. 2, Chemical Thermodynamics, North Holland Elsevier Science Publishers B. V., Amsterdam, The Netherlands, (1995), 374 p. Cited on pages: 4, 7, 443, 464, 465, 469,471.
BIBLIOGRAPHY
575
[96BRU/PIA]
Brunetti, B., Piacente, V., Torsion and knudsen measurements of cobalt and nickel fluorides and their standard sublimation enthalpies, J. Alloys Compd., 236, (1996), 63-69. Cited on pages: 121,431.
[96CRI/MIL]
Criss, C. M., Millero, F. J., Modeling the heat capacities of aqueous 1-1 electrolyte solutions with Pitzer's equations, J. Phys. Chem., 100,(1996), 1288-1294. Cited on page: 88.
[96DOJ/NOV]
Dojcilovic, J., Novakovic, L., Napijalo, M. M , Napijalo, M. L., Structural phase transitions in Co(II) and Ni(II) phosphate octahydrates in the temperature range 100-300 K, Mater. Chem. Phys., 43, (1996), 275-277. Cited on page: 205.
[96HEP/HOV]
Hepler, L. G., Hovey, J. K., Standard state heat capacities of aqueous electrolytes and some related undissociated species, Can. J. Chem., 74, (1996), 639-646. Cited on page: 88.
[96KEH]
Kehiaian, H. V., CRC Handbook of Chemistry and Physics, 76th. Edition, Lide, D. R. and Frederikse, H. P.R., Eds., CRC Press, Boca Raton, (1996). Cited on pages: 422, 431.
[96LEM/CAM]
Lemire, R. J., Campbell, A. B., Apparent molar heat capacities and volumes for NpO2ClO4(aq) to 100°C: Calculation of the temperature dependence of the MO2n+ (VI/V) couple, Radiochim. Acta, 73, (1996), 131-137. Cited on pages: 87,434.
[96LUT/RIC]
Luther III, G. W., Rickard, D. T., Theberge, S., Olroyd, A., Determination of metal (bi)sulfide stability constants of Mn2+, Fe2+, Co2+, Ni2+, Cu2+, and Zn2+ by voltammetric methods, Environ. Sci. Technol, 30, (1996), 671-679. Cited on pages: 177, 178, 431, 432.
[96POU/DRE]
Poulson, S. R., Drever, J. I., Aqueous complexing of nickel and zinc with 3-(N-morpholino)propanesulfonic acid and the solubility products of nickel and zinc hydroxides, Talanta, 43, (1996), 19751981. Cited on pages: 111, 112, 113, 114,433.
[96SAH/SAH]
Saha, A., Saha, N , Ji, L.-N., Zhao, J., Gregan, F , Sajadi, S. A. A., Song, B., Sigel, H., Stability of metal ion complexes formed with methyl phosphate and hydrogen phosphate, J. Biol. Inorg. Chem., 1, (1996), 231-238. Cited on pages: 205, 206, 433.
576
BIBLIOGRAPHY
[96SEI/FJE]
Seim, H., Fjellvag, H., Gr0nvold, F., Stolen, S., Metastable nickel sulfides with composition close to Ni7S6 - stability and structural properties, J. Solid State Chem., 121, (1996), 400-407. Cited on pages: 164, 169, 406, 426, 433.
[96STU/MOR2]
Stumm, W., Morgan, J. J., Aquatic Chemistry. Chemical equilibria and rates in natural waters, 3rd. Edition, John Wiley and Sons, New York, (1981), 1022 p. Cited on page: 24.
[96UCH/GOR]
Uchida, E., Goryozono, Y., Naito, ML, Aqueous speciation of magnesium, strontium, nickel and cobalt chlorides in hydrothermal solutions at 600°C and 1 kbar, Geochem. J., 30, (1996), 99-109. Cited on page: 152.
[97ALL/BAN]
Allard, B., Banwart, S. A., Bruno, J., Ephraim, J. H., Grauer, R., Grenthe, I., Hadermann, J., Hummel, W., Jakob, A., Karapiperis, T., Plyasunov, A. V., Puigdomenech, I., Rard, J. A., Saxena, S., Spahiu, K., Modelling in Aquatic Chemistry, Grenthe, I. and Puigdomenech, I., Eds., Nuclear Energy Agency, Organisation for Economic Cooperation and Development, Paris, France, (1997), 724 p. Cited on pages: 26,233,324,325.
[97BAH/PAR]
Bahta, A., Parker, G. A., Tuck, D. G., Critical survey of stability constants of complexes of thiocyanate ion, Pure Appl. Chem., 69, (1997), 1489-1548. Cited on pages: 231, 232, 433.
[97BAL/DIV]
Balej, J., Divisek, J., Er-pH and Er-mKon diagrams for nickel at 25°C, Collect. Czech. Chem. Commun., 62, (1997), 696-708. Cited on pages: 118, 119, 433.
[97GRE/PLY]
Grenthe, I., Plyasunov, A. V., On the use of semiempirical electrolyte theories for the modeling of solution chemical data, Pure Appl. Chem., 69, (1997), 951-958. Cited on page: 451.
[97GRE/PLY2]
Grenthe, I., Plyasunov, A. V., Spahiu, K., Estimations of medium effects on thermodynamic data, Modelling in Aquatic Chemistry, Grenthe, I. and Puigdomenech, I., Eds., pp.325-426, Nuclear Energy Agency, Organisation for Economic Co-operation and Development, Paris, France, (1997). Cited on pages: 80, 116, 144, 145, 177, 218, 220, 221, 224, 289, 349, 430, 431, 435, 439, 440.
BIBLIOGRAPHY
577
[97HOW/NEI]
Howell, I., Neilson, G. W., Ni2+ coordination in concentrated aqueous solution, J. Mol. Liq., 73-74, (1997), 337-348. Cited on page: 200.
[97HYA]
Hyams, D., CurveExpert 1.3: A comprehensive curve fitting system for windows., 112B Crossgate Street, Starkville, MS 39759, USA, (1997). Cited on page: 338.
[97KON/FAN]
Konnecke, T., Fanghanel, T., Kim, J. I., Thermodynamics of trivalent actinides in concentrated electrolyte solutions. Modelling the chloride complexation of Cm(III), Radiochim. Acta, 76, (1997), 131-135. Cited on page: 465.
[97MAL/ZAM]
Malatesta, F., Zamboni, R., Activity and osmotic coefficients from the EMF of liquid membrane cells. VI-ZnSO4, MgSO4, CaSO4 and SrSO4 in water at 25°C, J. Solution Chem., 26, (1997), 791-815. Cited on page: 324.
[97MAT/RAI]
Mattigod, S. V., Rai, D., Felmy, A. R., Rao, L., Solubility and solubility product of crystalline Ni(OH)2, J. Solution Chem., 26, (1997), 391-403. Cited on pages: 93, 112, 113, 114, 257, 271, 329, 434.
[97NOZ/KOB]
Nozue, T., Kobayashi, H., Sato, M., Uesawa, A., Suzuki T., Kamimura, T., Specific heat and de Haas-van Alphen effect in NiAs, Physica B (Amsterdam), (1997), 174-176. Cited on page: 211.
[97PAN/CAM]
Pan, P., Campbell, A. B., Apparent and standard molar volumes and heat capacities of aqueous Ni(ClO4)2 from 25 to 85°C, J. Solution Chem., 26, (1997), 461-478. Cited on pages: 86, 87, 88, 434.
[97PRA/LAN]
Pracht, G., Lange, N., Lutz, H. D., High-temperature Raman spectroscopic studies on nickel iodates, Thermochim. Acta, 293, (1997), 13-24. Cited on pages: 137, 139, 254.
[97PUI/RAR]
Puigdomenech, I., Rard, J. A., Plyasunov, A. V., Grenthe, I., Temperature corrections to thermodynamic data and enthalpy calculations, Modelling in Aquatic Chemistry, Grenthe, I. and Puigdomenech, I., Eds., pp.427-493, Nuclear Energy Agency, Organisation for Economic Co-operation and Development, Paris, France, (1997). Cited on pages: 39, 40.
578
BIBLIOGRAPHY
[97RIC]
Richens, D. T., The chemistry of aqua ions, John Wiley and Sons, Chichester, (1997). Cited on pages: 88,434.
[98ALM/NOV]
Al Mahamid, I., Novak, C. F., Becraft, K. A., Carpenter, S. A., Hakem, N., Solubility of Np(V) in K-C1-CO3 and Na-K-Cl-CO3 solutions to high concentrations: measurements and thermodynamic model predictions, Radiochim. Acta, 81, (1998), 93-101. Cited on page: 469.
[98ARC/RAR]
Archer, D. G., Rard, J. A., Isopiestic investigation of the osmotic and activity coefficients of aqueous MgSC>4 and the solubility of MgSO4.7H2O(cr) at 298.15 K: Thermodynamic properties of the MgSO4 + H2O system to 440 K, J. Chem. Eng. Data, 43, (1998), 791-806. Cited on page: 324.
[98CHA]
Chase, Jr., M. W., Monograph No. 9. NIST-JANAF Thermochemical Tables, J. Phys. Chem. Ref. Data, (1998). Cited on pages: 75, 76, 77, 78, 79, 173, 175, 177, 242, 369, 381, 404, 409.
[98ELM/GAB]
Elmasry, M. A. A., Gaber, A., Khater, E. M. H., Thermal decomposition of Ni(II) and Fe(III) nitrates and their mixture, J. Therm. Anal. Calorim., 52, (1998), 489-495. Cited on pages: 196, 198.
[98GAM/KON]
Gamsjager, H., Konigsberger, E , Preis, W., Solubilities of metal carbonates, Pure Appl. Chem., 70, (1998), 1913-1920. Cited on pages: 218,435.
[98JAI/ELM]
Jain, M., Elmore, A. L., Matthews, M. A., Weidner, J. W., Thermodynamic considerations of the reversible potential for the nickel electrode, Electrochim. Acta, 43, (1998), 2649-2660. Cited on pages: 117,435.
[98PLY/ZHA]
Plyasunova, N. V., Zhang, Y , Muhammed, M., Critical evaluation of thermodynamics of complex formation of metal ions in solutions. IV. Hydrolysis and hydroxo-complexes of Ni2+ at 298.15 K, Hydrometallurgy, 48, (1998), 43-63. Cited on pages: 80, 82, 83, 85, 86,90,91,95,97,99, 108, 112, 114,115,309,426,435,436.
BIBLIOGRAPHY
579
[98RAI/FEL]
Rai, D., Felmy, A. R., Hess, N. J., Moore, D. A., Yui, M., A thermodynamic model for the solubility of UO2(am) in the aqueous K+-Na+-HCOrCO^~-OH'-H2O system, Radiochim. Ada, 82, (1998), 17-25. Cited on page: 469.
[98ROG/KOZ]
Rog, G., Kozlowska-R6g, A., Divalent ion-p"-alumina electrolytes for the study of thermodynamic properties of oxide systems, Z. Phys. Chem. (Munich), 207, (1998), 83-92. Cited on pages: 243, 244, 436, 437.
[99ARC]
Archer, D. G., Thermodynamic properties of import to environmental processes and remediation. II. Previous thermodynamic property values for nickel and some of its compounds, J. Phys. Chem. Ref. Data, 28, (1999), 1485-1507. Cited on pages: 80, 82, 83, 108, 128, 129, 186, 253, 425, 436.
[99HOF/DAR]
Hoffmann, M. M., Darab, J. G., Palmer, B. J., Fulton, J. L., A transition in the Ni2+ complex structure from six- to four- coordinate upon formation of ion pair species in supercritical water, an x-ray absorption fine structure, near-infrared, and molecular dynamics study, J. Phys. Chem. A, 103, (1999), 8471-8482. Cited on pages: 141, 142.
[99KON/KON]
Konigsberger, E., Konigsberger, L.-C, Gamsjager, H., Low temperature thermodynamic model for the system Na2CO3-MgCO3CaCO3-H2O, Geochim. Cosmochim. Ada, 63, (1999), 3105-3119. Cited on pages: 219,221,437.
[99LAN/MAH]
Langmuir, D., Mahoney, J., MacDonald, A., Rowson, J., Predicting arsenic concentrations in the porewaters of buried uranium mill tailing, Geochim. Cosmochim. Ada, 63, (1999), 3379-3399. Cited on pages: 213,412.
[99MAL/CAR]
Malatesta, F., Carbonaro, N., Fanelli, N., Ferrini, S., Giacomelli, A., Activity and osmotic coefficients from the Emf of liquid-membrane cells. VII: Co(ClO4)2, Ni(C104)2, K2SO4, CdSO4, CoSO4, NiSO4, J. Solution Chem., 28, (1999), 593-619. Cited on pages: 158, 159, 337,437.
[99MCB]
McBreen, J., Chapter 3, Nickel hydroxides, Handbook of battery materials, Besenhard, J. O., Ed., pp. 135-151, Wiley-VCH, Weinheim, (1999). Cited on page: 115.
580
BIBLIOGRAPHY
[99MIK/MIG]
Mikuli, E., Migdai-Mikuli, A., Wrobel, S., Grad, B., DSC investigations of the phase transitions of [M(H2O)g](NO3)2, where M = Mn, Co, Ni, Cu and Zn, Z. Naturforsch., 54a, (1999), 595-598. Cited on page: 198.
[99RAR/RAN]
Rard, J. A., Rand, M. H., Anderegg, G., Wanner, H., Chemical Thermodynamics ofTechnetium, Sandino, M. C. A. and Osthols, E., Nuclear Energy Agency Data Bank, Organisation for Economic Cooperation and Development, Eds., vol. 3, Chemical Thermodynamics, North Holland Elsevier Science Publishers B. V., Amsterdam, The Netherlands, (1999). Cited on pages: 4, 6, 7, 39, 53,54,64,443.
[99THO]
Thoenen, T., Pitfalls in the use of solubility limits for radioactive waste disposal: the case of nickel in sulfidic groundwaters, Nucl. Technol, 126, (1999), 75-87. Cited on page: 180.
[99WAN/OST]
Wanner, H., Osthols, E., TDB-3: Guidelines for the assignment of uncertainties, Nuclear Energy Agency Data Bank, Organisation for Economic Co-operation and Development, Report , (1999), 17 p. Cited on page: 7.
[2000GRE/WAN] Grenthe, I., Wanner, H., Osthols, E , TDB-2: Guidelines for the extrapolation to zero ionic strength, OECD Nuclear Energy Agency, Data Bank, Report, (2000), 34 p. Cited on pages: 7,443,444,450. [2000OST/WAN] Osthols, E., Wanner, H., TDB-0: The NEA Thermochemical Data Base Project, Nuclear Energy Agency Data Bank, Organisation for Economic Co-operation and Development, Report , (2000), 21 p. Cited on page: 7. [2000ROG/KOZ] Rog, G., Kozlowska-R6g, A., Bucko, M., Glowacz, E., Determination of the standard molar Gibbs free energies of formation of the silicates of cobalt and nickel by solid-state galvanic cells involving the CaF2-based composite electrolyte, J. Chem. Thermodyn., 32, (2000), 931-935. Cited on pages: 243, 244, 437.
BIBLIOGRAPHY
[2000TSI/MOL]
581
Tsierkezos, N. G., Molinou, I. E., Limiting molar conductances and thermodynamic association constants for nickel(II), cadmiun(II), magnesium(II), and copper(II) sulfates in mixtures of methanol and water at 298.15 K, J. Chem. Eng. Data, 45, (2000), 819-822. Cited on pages: 181, 184, 185, 189,437,438.
[2000WAN/OST] Wanner, H., Osthols, E., TDB-1: Guidelines for the review procedure and data selection, Nuclear Energy Agency Data Bank, Organisation for Economic Co-operation and Development, Report, (2000), 5 p. Cited on page: 7. [2001GAM/PRE] Gamsjager, H., Preis, W., Wallner, H., Solid-solute phase equilibria in aqueous solutions XIV. Thermodynamic analysis of the solubility of hellyerite in water, Monatsh. Chem., 132, (2001), 411-415. Cited on pages: 218, 219, 220, 221, 222, 263, 310,438, 441. [2001 HEN/RED] Henderson, C. M. B., Redfern, S. A. T., Smith, R. I., Knight, K. S., Charnock, J. M., Composition and temperature dependence of cation ordering in Ni-Mg olivine solid solutions: a time-of-flight neutron powder diffraction and EXAFS study, Am. Mineral, 86, (2001), 1170-1187. Cited on pages: 239,438. [2001LAW/ZHA] Lawlis, J. D., Zhao, Y.-H., Karato, S., High-temperature creep in a Ni 2 Ge0 4 : a contribution to creep systematics in spinel, Phys. Chem. Miner., 28, (2001), 557-571. Cited on page: 245. [2001LEM/FUG] Lemire, R. J., Fuger, J., Nitsche, H., Potter, P. E., Rand, M. H., Rydberg, J., Spahiu, K., Sullivan, J. C , Ullman, W. J , Vitorge, P., Wanner, H., Chemical Thermodynamics of Neptunium and Plutonium, Nuclear Energy Agency Data Bank, Organisation for Economic Co-operation and Development, Ed., vol. 4, Chemical Thermodynamics, North Holland Elsevier Science Publishers B. V., Amsterdam, The Netherlands, (2001). Cited on pages: 4, 7, 39, 58, 61, 67, 68, 147, 335, 386, 443, 464, 465, 469,471. [2001MIK/MIG]
Mikuli, E., Migdat-Mikuli, A., Chyzy, R., Grad, B., Dziembaj, R., Melting and thermal decomposition of [Ni(H2O)6](NO3)2, Thermochim. Ada, 370, (2001), 65-71. Cited on pages: 196, 197, 198, 200.
582
BIBLIOGRAPHY
[2001MOL/TSI]
Molinou, I. E., Tsierkezos, N. G., Conductance studies on manganese(II), cobalt(II), nickel(II), and cadmiun(II) sulfates in water + N,N-dimethylformamide mixtures at 298.15 K, J. Chem. Eng. Data, 46, (2001), 1399-1403. Cited on pages: 181, 184, 438.
[2001OAK/RAI]
Oakes, C. S., Rai, D., Thermodynamic properties of aqueous HC1O4 to 373 K and at low pressure: a comprehensive Pitzer ion-interaction model, J. Chem. Eng. Data, 46, (2001), 875-884. Cited on page: 87.
[2001VOG]
Vogt, N., Equilibrium bond lengths, force constants and vibrational frequencies of MnF2, FrF2, C0F2, NiF2, and ZnF2 from least-square analysis of gas-phase electron diffraction data, J. Mol. Struct., 571, (2001), 189-195. Cited on page: 121.
[2002BUC/RON] Buck, R. P., Rondini, S., Covington, A. K., Baucke, F. G. K., Brett, C. M. A., Camoes, M. F., Milton, M. J. T., Mussini, T., Naumann, R., Pratt, K. W., Spitzer, P., Wilson, G. S., Measurement of pH. DEfinition, standards, and procedures (IUPAC recommendations 2002), Pure Appl. Chem., 74, (2002), 2169-2200. Cited on page: 25. [2002CHI/SIM]
Chialvo, A. A., Simonson, J. M., Simulations and neutrons scattering methodology, Mol. Phys., 100, (2002), 2307-2315. Cited on page: 141.
[2002GAM/WAL] Gamsjager, H., Wallner, H., Preis, W., Solid-solute phase equilibria in aqueous solutions XVII. Solubility and thermodynamic data of nickel(II) hydroxide, Monatsh. Chem., 133, (2002), 225-229. Cited on pages: 108, 112, 113, 114, 115,271,438. [2002MON/HEL] Monlien, F. J., Helm, L., Abou-Hamdan, A., Merbach, A. E., Mechanistic diversity covering 15 orders of magnitude in rates: cyanide exchange on [M(CN)4]2" (M = Ni, Pd and Pt), lnorg. Chem., 41, (2002), 1717-1727. Cited on pages: 231, 441. [2002PAT/WOO] Patterson, B. A., Woolley, E. M., Thermodynamics of ionization of water at temperatures 278.15 K < T/K < 393.15 K and at the pressure p = 0.35 MPa: apparent molar volumes and apparent molar heat capacities of aqueous of potassium and sodium nitrates and nitric acid, J. Chem. Thermodyn., 34, (2002), 535-556. Cited on pages: 87,88.
BIBLIOGRAPHY [2002SCH]
583
Schlesinger, M. E., The thermodynamic properties of phosphorus and solid binary phosphides, Chem. Rev., 102, (2002), 4267-4301. Cited on page: 204.
[2002WAL/GAT] Wallner, H., Gatterer, K., Growth of pure Ni(OH)2 single crystals from solution-control of the crystal size, Z. Anorg. Allg. Chem., 628, (2002), 2818-2820. Cited on pages: 112, 438, 441. [2002WAL/PRE] Wallner, H., Preis, W., Gamsjager, H., Solid-solute phase equilibria in aqueous solutions XV. Thermodynamic analysis of the solubility of nickel carbonates, Thermochim. Ada, 382, (2002), 289-296. Cited on pages: 218, 220, 221, 222, 263, 310, 441. [2003BAE/BRA] Baeyens, B., Bradbury, M. H., Hummel, W., Determination of aqueous nickel-carbonate and nickel-oxalate complexation constants, J. Solution Chem., 32, (2003), 319-339. Cited on pages: 221,224,225,441. [2003GAM/KON] Gamsjager, H., Konigsberger, E., In Hefter, G. T., 4.2 Solubility of sparingly soluble solids in liquids, The experimental determination of solubilities, Tomkins, R. P. T., Ed., pp.315-358, John Wiley & Sons, Ltd, (2003). Cited on page: 111. [2003GUI/FAN]
Guillaumont, R., Fanghanel, T., Fuger, J., Grenthe, I., Neck, V., Palmer, D. A., Rand, M. H., Update on the chemical thermodynamics of Uranium, Neptunium, Plutonium, Americium and Technetium, Nuclear Energy Agency Data Bank, Organisation for Economic Co-operation and Development, Ed., vol. 5, Chemical Thermodynamics, North Holland Elsevier Science Publishers B. V., Amsterdam, The Netherlands, (2003). Cited on pages: 5, 6, 7, 53, 54, 59, 65, 66, 67,443, 464,469, 471.
[2003HUM/CUR] Hummel, W., Curti, E., Nickel aqueous speciation and solubility at ambient conditions: A thermodynamic elegy, Monatsh. Chem., 134, (2003), 941-976. Cited on pages: 223, 224, 225, 442. [2004BRO/MER] Brown, B. R., Merkley, E. D., McRae, B. R., Origlia-Luster, M. L., Woolley, E. M , Apparent molar volumes and apparent molar heat capacities of aqueous nickel(II) nitrate, copper(II) nitrate, and zinc(II) nitrate at temperatures (from 378.15 to 393.15) K at the pressure 0.35 MPa, J. Chem. Thermodyn., 36, (2004), 437-446. Cited on page: 88.
584
[2005OLI/NOL]
BIBLIOGRAPHY
Olin, A., Nolang, B., Ohman, L-O., Osadchii, E. G, Rosen, E., Chemical thermodynamics of Selenium, Nuclear Energy Agency Data Bank, Organisation for Economic Co-operation and Development, Ed., vol. 6, Chemical Thermodynamics, North Holland Elsevier Science Publishers B. V., Amsterdam, The Netherlands, (2005). Cited on pages: 6, 53, 62, 72, 195.
List of cited authors This chapter contains an alphabetical list of the authors of the references cited in this book. The reference codes given with each name corresponds to the publications of which the person is the author or a co-author. Note that inconsistencies may occur due to variations in spelling between different publications. No attempt has been made to correct for such inconsistencies in this volume. Author Abbattista, F. Abou-Hamdan, A. Accascina, F. Achenza, F. Acs, V. Adam, A. Adami, L. H. Aditya, S. Ageno,F. Ahrland, S. Ajdinjan, N. H. Akimoto, S.-I. Aksel'rud, N. V. Al Mahamid, I. Alcock, C. B. Allard, B. Allison, J. D. Almqvist, G. Althaus, H. Alyamovskaya, K. V. Ananthaswamy, J. Anderegg, G. Anderson, C. T. Anderson, G. M. Andreev, S. N. Andreeva, M. N. Angell, C. A. Angnes, L. Antill, J. E.
Reference [63BUR/ABB] [2002MON/HEL] [59FUO/ACC] [59ACH] [41KIS/ACS] [80ADA/BIL], [81ADA/BIL] [65ADA/KIN] [61MOH/DAS] [11AGE/VAL] [56AHR/LAR], [56AHR/ROS] [53GRI/SLU] [65AKI/FUJ], [74YAG/MAR] [50AKS/FIA] [98ALM/NOV] [64ALC/BEL], [64ALC/SUD], [65STE/ALC], [93KUB/ALC] [97ALL/BAN] [93ALL/BRO] [18ALM] [65SCH/ALT] [61CHU/ALY2] [84ANA/ATK] [99RAR/RAN] [35KEL/AND] [91 AND/CAS] [68AND/KHA], [68AND/KHA2] [68AND/KHA2] [66ANG/GRU] [82NEV/ANG] [67ANT/WAR1 (Continued on next page)
585
586 Author Aprile, F. Arai, Y. Archer, D. G. Archibald, R. C. Arculus, R. J. Ariya, S. M. Arnek, R. Arnold, R. G. Artamonova, E. A. Aruga, R. Arvia, A. J. Asao, H. Ashcroft, S. J. Ashurst, K. G. Asper, R. Atkinson, G. Auffredic, J.-P. Avramenko, N. I. Baes, Jr., C. F. Baeyens, B. Bahta, A. Bailey, S. M. Bain, R. Balarev, K. Bale, C. W. Balej, J. Ball, R. G. J. Balz, 0 . Banerjea, D. Banwart, S. A. Banyai, I. Bard, A. J. Barin, I. Barnard, R. Bartholin, H. Bartholomew, C. H. Bartovska, A. Bartovsky, A. Barvinok, M. S. Basavalingu Bates, R. G. Bath, G. A. Batrakov, V. V. Batyaev, I. M.
LIST OF CITED A UTHORS
Reference [68LIN/APR] [86TOD/HAS] [90ARC/WAN], [98ARC/RAR], [99ARC] [66STO/ARC] [86HOL/NEI] [71ARI/MOR] [68ARN] [75ARN/MAL] [89ART/KAS] [74ARU], [75ARU] [63BOL/JAU] [69MOR/SAT] [70ASH] [77ASH/HAN] [77OSW/ASP] [84ANA/ATK] [72AUF/CAR], [72AUF/CAR2] [83AVR/BLO] [74MES/BAE], [76BAE/MES] [2003BAE/BRA] [97BAH/PAR] [82WAG/EVA] [63PEN/BAI] [89LIL/TEP] [83GAU/BAL] [89BAL], [93BAL/DIV], [93BAL/DIV2], [97BAL/DIV] [92BAL/DIC] [21WOH/BAL] [81BAN/KAD] [97ALL/BAN] [92BAN/BLI] [85BAR/PAR] [73BAR/KNA], [77BAR/KNA] [80BAR/RAN], [81 BAR/RAN] [81ADA/BIL] [69IZA/EAT] [74BAR/CER] [74BAR/CER] [75BAR/MAS] [91TAR/FAZ] [73BAT] [71BHA/SUB] [93MAK/VOE] [74NET/BAT] (Continued on next page)
587
LIST OF CITED A UTHORS
Author Baucke, F. G. K. Beaudouin, B. Bechtold, D. B. Beck, M. T. Becker, K. Becraft, K. A. Behr, M. Behrens, R. G. Belford, T. N. Bell, J. Belokurova, N. A. Bender, J. A. Benrath, A. Berger, E. Berglund, S. Bergman, G. A. Berry, L. G. Bestul, A. B. Bevington, P. R. Bezborodov, A. A. Bhandage, G. T. Bhatia, S. N. Bhattacharya, S. Biader Ceipidor, U. Bichowsky, F. R. Bidoglio, G. Bidwell, L. R. Biedermann, G. Bigliardi, G. Bijvoet, J. M. Bilinski, H. Billerey, D. Biltz, W. Binford Jr., J. S. Bish, D. L. Bishop, J. J. Bixler, J. W. Bjerrum, J. Blackie, M. S. Blaszkiewicz, G. Blixt, J. Blokhin, V. V. Bode, H.
Reference [2002BUC/RON] [84BRA/DEL] [78BEC/LIU] [62BEC/BJE], [87BEC], [90BEC/NAG] [67SIG/BEC] [98ALM/NOV] [24MOS/BEH] [72BEH/ROS] [64ALC/BEL] [37BEL], [40BEL] [86VAS/DMI] [23VIL/BEN] [34BEN/THI] [21BER/CRU] [76BER2] [82GLU/GUR] [81NIC/BER] [71CHA/BES] [69BEV] [76ZHO/BEZ] [91TAR/FAZ] [77BHA/CAR] [86BHA/MUK] [76BIA/CAR] [36BIC/ROS] [95SIL/BID] [67RIZ/BID] [71OHT/BIE], [75BIE] [63FER/BRA] [33BIJ/NIE] [73NOV/COS] [77BIL/TER], [80ADA/BIL], [81ADA/BIL] [36BIL/VOI] [67BIN/STR], [70BIN/HEB] [82LIV/BIS] [61MAR/BYD] [74BIX/LAR], [83 SOL/BON] [62BEC/BJE], [86BJE/SKI], [87BJE], [88BJE], [88BJE2], [89BJE], [90BJE], [90BJE2] [59BLA/GOL], [59BLA/GOL2] [67BLA] [92BAN/BLI] [74BLO/RAZ], [76KUL/BLO], [81KUL/BLO], [83AVR/BLO] [66BOD/DEH1 (Continued on next page)
588
Author Boerstoel, B. M. Bogatzki, D. P. Boldamdaeva, 0 . K. Bolzan, J. A. Bond, A. M. Bonilla, C. F. Bonnell, D. G. R. Borchardt G. Bornemann, K. Bostrom, D. Bouet, G. Boye, E. Boyle, B. J. Brackett, T. E. Braconnier, J. J. Bradbury, M. H. Bradley, D. J. Braibanti, A. Branica, M. Bratsch, S. G. Braun, M. Breckpot, R. Brett, C. M. A. Britton, H. T. S. Brodale, G. E. Broers, G. H. J. Bronson, H. L. Brensted, J. N. Brooker, E. J. Brown, B. R. Brown, D. S. Brown, G. R. Brown, Jr., G. E. Brown, P. G. M. Brown, T. L. Bruner, L. Brunetti, B. Brungs, M. P. Bruno, J. Buck, R. P. Bucko, M. Bugarcic, Z. D. Burdese, A. Burke, K. A.
LIST OF CITED A UTHORS
Reference [64CHA/BOE] [38BOG] [93UVA/TIM] [63BOL/JAU] [70BON], [71 BON], [72BON/HEF], [83SOL/BON] [52CAR/BON] [35BON/BUR] [84ROG/BOR] [08BOR], [ 1OBOR] [87BOS] [95KAH/CRO] [33BOY] [54BOY/KIN] [56HOR/BRA] [84BRA/DEL] [2003BAE/BRA] [79BRA/PIT] [63FER/BRA] [73NOV/COS] [89BRA] [66VOL/KOH] [69RAN/BRE] [2002BUC/RON] [25BRI] [66STO/ARC] [75WEE/KOE] [36BRO/WIL] [22BRO], [22BRO2] [52ROB/BRO] [2004BRO/MER] [93ALL/BRO] [65JAM/BRO] [82BRO] [55BRO/PRU] [77CHE/BRO] [09BRU/ZAW] [96BRU/PIA] [80TRA/BRU] [86BRU], [97ALL/BAN] [2002BUC/RON] [2000ROG/KOZ] [80MIL/BUG] [63BUR/ABB] [20BUR1 (Continued on next page)
LIST OF CITED A UTHORS
Author Burkov, K. A. Burridge, L. W. Busey, R. H. Bydalek, T. J. Caillere, S. Caldero, J. M. Cali, R. Calk, L. C. Caminiti, R. Camoes, M. F. Campbell, A. B. Campbell, F. E. Cantor, S. Capdevila, H. Carbonaro, N. Carel, C. Carlin, R. L. Carney, J. H. Carpenter, S. A. Carr, D. S. Carunchio, V. Castet, S. Catalano, E. Cemic, L. Cerny, C. Cerutti, P. Chaberek, Jr., S. Chakravorty, A. Chandrasekharaiah, M. S. Chang, S. S. Chang, Y. A. Charette, G. G. Charnock, J. M. Charykov, N. A. Chase, Jr., M. W. Chattopadhyay, G. Chen, A. Cheng, C. P. Chialvo, A. A. Chichirova, N. D. Chickerur, N. S. Chierice, G. O. Childs, C. W.
589
Reference [65BUR/LIL], [65BUR/LIL2], [66BUR/IVA], [71BUR/ZIN], [78BUR/KAM] [35BON/BUR] [52BUS/GIA], [52BUS/GIA2], [53BUS/GIA] [61MAR/BYD] [62CAI/POB] [62TRI/CAL], [64TRI/CAL] [73GUR/CAL] [73WAA/CAL] [80CAM/LIC], [82CAM] [2002BUC/RON] [96LEM/CAM], [97PAN/CAM] [68CAM/ROE] [65CAN] [92CAP], [93GIF/VIT2], [94CAP/VIT], [95CAP/VIT] [99MAL/CAR] [72AUF/CAR], [72AUF/CAR2] [77BHA/CAR] [70CAR/LAI] [98ALM/NOV] [52CAR/BON] [76BIA/CAR] [91 AND/CAS] [54STO/CAT], [55CAT/STO], [55STO/CAT] [86CEM/KLE], [87CEM/KLE] [74BAR/CER] [68LAR/CER] [52CHA/COU] [86BHA/MUK] [75CHA/KAR] [71CHA/BES] [77LIN/IPS], [78LIN/HU], [79SHA/CHA], [80SHA/CHA], [86HSI/CHA] [68CHA/FLE] [2001 HEN/RED] [85FIL/CHA], [86FIL/CHA] [98CHA] [75CHA/KAR] [90PAS/TAY] [77CHE/BRO] [2002CHI/SIM] [85SAP/CHI] [80CHI/SAB], [82CHI/SAB] [82NEV/ANG] [70CHI] (Continued on next page)
590
LIST OF CITED A UTHORS
Author Christ, C. L. Christensen, J. J. Christensen, Jr., A. U. Chukhlantsev, V. G. Chukurov, P. M. Chunaeva, V. D. Churney, K. L. Chyzy, R. Ciavatta, L. Ciret, N. Claire, Y. Clare, B. W. Clark, C. W. Coffetti, G. Coleman, J. S. Colombier, L. Combs, K. E. S. Comert, H. Conard, B. B. Connelly, D. L. Conway, B. E. Conway, K. C. Cook, R. S. Cordes, S. M. Cordfunke, E. H. P. Corlis, C. Cosovic, B. Coughlin, J. P. Courtney, R. C. Covington, A. K. Cowell, J. C. Cox, J. D. Craig, J. R. Crego, A. Criaud, A. Criss, C. M. Cronier, D. Crut, G. Csokan, P. Cucinotta, V. Cudennec, Y. Cummiskey, C. J. Cuprova Curti, E. Cuta, F.
Reference [65GAR/CHR] [63CHR/IZA], [69IZA/EAT], [71IZA/JOH] [58KIN/CHR] [56CHU3], [56CHU4], [57CHU], [61CHU/ALY2] [73CHU/DRA] [95MUL/CHU] [82WAG/EVA] [2001MIK/MIG] [80C1A], [87CIA/IUL], [88CIA], [90CIA] [77BIL/TER] [80DUB/CLA] [75CLA/KEP] [35KEE/CLA] [05COF/FOR] [65COL/PET] [34COL2] [89ZIE/JON] [84COM/PRA] [77CON/SRI] [7 ICON/LOO] [62CON/GIL] [54BOY/KIN] [26PEA/COO] [69SUB/COR] [73COR], [90COR/KON2], [92BAL/DIC] [81COR/SUG] [73NOV/COS] [51COU] [52CHA/COU] [82SAR/COV], [2002BUC/RON] [74WOR/COW] [89COXAVAG] [76CRA/SCO] [62FAU/CRE] [84FOU/CRI] [96CRI/MIL] [95KAH/CRO] [21BER/CRU], [24CRU], [24CRU2] [38CSO] [78SIR/SEM] [87CUD/LEC] [71 LAN/CUM] [59KIS/CUP] [2003HUM/CUR] [56CUT/KSA] (Continued on next page)
LIST OF CITED A UTHORS
Author Cvitas, T. Dabrowski, J. Dabrowski, W. Damiani, R. D'Aprano, A. Darab, J. G. Das, R. C. Dash, A. C. Dash, U. N. Date, M. Davies, A. G. Davies, C. W. Davis, C. E. S. Day, P. De Jong, W. F. De Jonge, W. J. M. De Moras, M. De Nobel, J. De Ranter, C. De Souza, B. C. De Stefano, C. De Waal, S. A. De Wijs, H. J. De Zoubov, N. DeFord, D. D. Dehmelt, K. Delmas, C. Deltombe, E. Derby, I. H. Desai, P. D. Deshpande, D. A. Desnoyers, J. E. DeWitt, B. J. Dickinson, S. Dickson, A. G. Diebler, H. Dinsdale, A. T. Divisek, J. Dmitrieva, N. G. Dobrokhtov, G. N. Dodgen, H. W. Dojcilovic, J. Dokoupil, Z. Domange, L. Domash, L.
591
Reference [93MIL/CVI] [32SKA/DAB] [93PRZ/WIS] [63BUR/ABB] [71 APR] [99HOF/DAR] [61M0H/DAS], [71DAS/DAS] [7 IDAS/DAS] [7 IDAS/DAS] [83KAT/SUG] [61DAV/SMI] [32MON/DAV], [62DAV] [77NIC/ROB] [76DAY/DIN] [27JON/WIL] [72POL/HER] [85MAG/MOR] [64CHA/BOE] [69RAN/BRE] [63SHA/DES] [94STE/FOT] [73WAA/CAL] [25WIJ] [63DEL/ZOU] [51DEF/HUM] [66BOD/DEH] [84BRA/DEL] [63DEL/ZOU] [16DER/YNG] [73HUL/DES] [79NAN/DES] [81PER/ROU] [40SEL/DEW] [92BAL/DIC] [81TUR/WHI] [76TAY/DIE] [76DAY/DIN], [91 DIN] [93BAL/DIV], [93BAL/DIV2], [97BAL/DIV] [84VAS/VAS], [86VAS/DMI] [54DOB] [68LIN/APR], [78BEC/LIU], [83GOW/DOD] [96DOJ/NOV] [72KLA/DOK] [37DOM] [55TOM/DOM] (Continued on next page)
592
LIST OF CITED A UTHORS
Author Domiciano, J. B. Donaldson, R. H. Donges, E. Drakin, S.I. Drever, J. I. Droll, H. A. Drowart, J. Du Chatenier, F. J. DuBose, J. B. Dubusc, M. Dunn, L. Dyrssen, D. Dziembaj, R. Eatough, D. J. Eckstrom, H. C. Economou, M. Edmonds, D. T. Edwards, O. W. Efimov, A. I. Efimov, M. E.
Reference [90JUR/DOM] [65DON/EDM] [47DON] [73CHU/DRA], [85DRA/MAD] [82DRE], [96POU/DRE] [63NET/DRO] [92BAL/DIC] [64CHA/BOE] [79DUB/NAU] [80DUB/CLA] [73EDW/FAR] [85DYR], [88DYR], [90DYR/KRE] [2001MIK/MIG] [69IZA/EAT], [71IZA/JOH] [37PEA/ECK] [81MAR/EC0], [83ECO/MAR] [65DON/EDM] [73EDW/FAR] [70EFI/KUD] [88EFI/EVD], [89EVD/EFI], [90EFI/EVD], [90EFI/FUR] [65EGO/SMI] [64EHL/KEN] [36BIL/VOI] [50GLE/EIN] [54EIT] [76PER/EKS] [65ELL] [98ELM/GAB] [98JAI/ELM] [87EMA/FAR] [83NEI/END] [78ENO/TSU] [97ALL/BAN] [95KON/ERI] [69KUL/ERS] [73SKE/ESP] [04EUL] [52ROS/WAG], [82WAG/EVA] [88EFI/EVD], [90EFI/EVD] [89EVD/EFI] [90PAS/TAY], [92FRE/FAL] [83RAM/FAN] [99MAL/CAR] [97KON/FAN], [2003GUI/FAN] (Continued on next page)
Egorov, A. M. Ehler, T. C. Ehrlich, P. Einerhand, J. Eitel, W. Ekstrom, C. G. Elliott, R. P. Elmasry, M. A. A. Elmore, A. L. Emara, M. M. Enderby, J. E. Enoki, T. Ephraim, J. H. Eriksson, G. Erstigneeva, T. L. Espelund, A. W Euler, H. Evans, W. H. Evdokimova, V. I. Evdokimova, V. P. Falicov, L. M. Fan, G. Fanelli, N. Fanghanel, T.
LIST OF CITED A UTHORS
Author Farber, M. Farhi, R. Farid, N. A. Fair, T. D. Faucherre, J. Fazeli, A. R. Feber, R. C. Fedorov, V. A. Fedorov, Y. A. Fedotov, N. V. Feitknecht, W. Felmy, A. R. Ferguson, R. B. Ferier, A. Fernelius, W. C. Ferrante, M. J. Ferrari, A. Ferri, D. Ferrini, S. Fialkov, Y. A. Figlarz, M. Filho, A. P. Filippov, V. K. Filippova, E. P. Fisher, F. H. Fitzner, K. Fjellvag, H. Fleet, M. E. Flengas, S. N. Florence, T. M. Flowers, G. C. Fontaine, A. Foote, H. W. Forsling, W. Forstat, H. Forster, F. Foti, C. Fouassier, C. Fouillac, C. Fox, A. P. Frank, D. F. Franke, E. Fraunberger, F. Fray, D. J. Freeman, R. D.
593
Reference [58FAR/MEY] [84RAY/PET] [87EMA/FAR] [73EDW/FAR] [62FAU/CRE] [91TAR/FAZ] [68HER/FEB] [73FED/SHM] [86FIL/CHA] [74MAK/MAS] [36LOT/FEI], [54FEI/HAR], [63FEI/SCH] [97MAT/RAI], [98RAI/FEL] [74GRI/FER] [63GEO/FER] [77FER] [78STU/FER], [82FER/GOK] [63FER/BRA] [83FER/GRE], [85FER/GRE] [99MAL/CAR] [50AKS/FIA], [56FIA/SHE] [84BRA/DEL] [77BHA/CAR] [85FIL/CHA], [86FIL/CHA] [74MAK/MAS] [79FIS/FOX] [75MOS/FIT], [75MOS/FIT2] [94STO/FJE], [96SEI/FJE] [72FLE], [77FLE], [87FLE] [68CHA/FLE], [73FLE] [66FLO] [81HEL/KIR] [80LAG/FON] [23FOO] [78FOR] [59SPE/FOR] [05COF/FOR] [94STE/FOT] [84BRA/DEL] [84FOU/CRI] [79FIS/FOX] [67RIZ/BID] [1895FRA] [04MUT/FRA] [94KAL/FRA] [84FRE] (Continued on next page)
594 Author Freericks, J. K. Freiser, H. Freund, H. Frey, C. M. Fricke, R. Friedberg, S. A. Fronaeus, S. Fuger, J. Fujimura, K. Fujinaga, T. Fujisawa, H. Fulton, J. L. Funahashi, S. Funk, R. Fuoss, R. M. Furkalyuk, M. Y. Furumi, Y. Gaber, A. Galchenko, G. L. Galus, Z. Gamsjager, H. Garber, H. K. Garrels, R. M. Garrett, A. B. Garvin, D. Gatterer, K. Gauthier, M. Gayer, K. H. Gee, R. Geier, G. Geiger, G. H. Geoffray, H. Gerault, Y. Gessner, H. Giacomelli, A. Gianguzza, A. Giauque, W. F. Giera, E. Giesekus, A. Giffaut, E. Gil, N. S.
LIST OF CITED A UTHORS
Reference [92FRE/FAL] [69SUB/COR] [59FRE/SCH] [72FRE/STU], [78FRE/STU] [42FRI/WEI] [60ROB/FRI], [73HAM/FRI] [53FRO2], [62FRO/LAR] [92FUG/KHO], [92GRE/FUG], [2001LEM/FUG], [2003GUI/FAN] [78IWA/FUJ] [66SHI/FUJ] [65AKI/FUJ] [99HOF/DAR] [69FUN/TAN] [1899FUN] [59FUO/ACC] [90EFI/FUR] [76SHI/TSU] [98ELM/GAB] [84LAV/TIM] [74KUT/GAL] [82GAM/REI], [98GAM/KON], [99KON/KON], [2001GAM/PRE], [2002GAM/WAL], [2002WAL/PRE], [2003GAM/KON] [68LAR/CER] [65GAR/CHR] [49GAY/GAR] [87GAR/PAR] [2002WAL/GAT] [83GAU/BAL] [49GAY/GAR], [52GAY/WOO] [76GEE/SHE] [71HOH/GEI] [73VEN/GEI] [63GEO/FER] [87CUD/LEC] [14THI/GES] [99MAL/CAR] [94STE/FOT] [41STO/GIA], [52BUS/GIA], [52BUS/GIA2], [53BUS/GIA], [56HOR/BRA], [66STO/ARC] [75KAK/GIE] [90PAS/TAY] [93GIF/VIT2], [94GIF] [59GIL/NYH] (Continued on next page)
LIST OF CITED A UTHORS
Author Gilbert, G. Gileadi, E. Glaser, J. Glasstone, S. Glebov, V. A. Gleiser, M. Glemser, O. Glowacz, E. Glushko, V. P. Gokcen, N. A. Gold, V. Goldberg, R. N. Goldschmidt, V. M. Golovanova, J. G. Golub, A. M. Goncharov, V. I. Goryozono, Y. Goto, S. Gower, S. A. Grad, B. Graham, R. D. Grauer, R. Grayson-Smith, H. Gregan, F. Grenthe, I.
Gricaenko, G. S. Grice, J. D. Griffiths, T. R. Grimaldi, M. Granvold, F. Gruen, D. M. Gruener, D. W. Grzybkowski, W. Guedes de Carvalho, J. R. F. Guggenheim, E. A. Guillaumont, R. Gurrieri, S. Gurvich, L. V. Haas, C. Hadermann, J. Hadley, W. B.
595
Reference [63PEN/BAI] [62CON/GIL] [92BAN/BLI] [26GLA] [71KLY/GLE] [73HUL/DES] [50GLE/EIN] [2000ROG/KOZ] [82GLU/GUR] [82FER/GOK], [84PAN/STU] [59BLA/GOL], [59BLA/GOL2] [66GOL/RID], [79GOL/NUT], [82SAR/COV] [26GOL] [75LEV/GOL] [79GOL/KOH] [73CHU/DRA] [96UCH/GOR] [77SAD/KAW] [83GOW/DOD] [99MIK/MIG], [2001MIK/MIG] [74GRA/WIL] [97 ALL/BAN] [50VAS/GRA] [96SAH/SAH] [83FER/GRE], [85FER/GRE], [87RIG/VIT], [92GRE/FUG], [94PLY/GRE], [95GRE/PUI], [97ALL/BAN], [97GRE/PLY], [97GRE/PLY2], [97PUI/RAR], [2000GRE/WAN], [2003GUI/FAN] [53GRI/SLU] [74GRI/FER] [69GRI/SCA] [64GRI/LIB] [91STO/GRO], [94STO/FJE], [95GRO/STO], [96SEI/FJE] [66ANG/GRU] [65JAM/BRO] [90GRZ] [73RAU/GUE] [35GUG], [66GUG] [2003GUI/FAN] [73GUR/CAL] [82GLU/GUR] [81KUI/SAN] [97 ALL/BAN] [64STO/HAD] (Continued on next page)
596
LIST OF CITED A UTHORS
Author Hagenmuller, P. Hakem, N. Hale, J. D. Hallenbeck, D. R. Halloff, E. Halow, I. Hamann, S. D. Hamburger, A. I. Hamer, W. J. Hammes, G. G. Hancock, R. D. Hansen, J. W. Haring, M. M. Harned, H. S. Hartmann, L. Hashimoto, K.
Reference [84BRA/DEL] [98ALM/NOV] [63CHR/IZA] [83 SOL/BON] [70HAL/VAN] [82WAG/EVA] [82HAM] [73HAM/FRI] [59HAM] [64HAM/MOR] [77ASH/HAN] [71IZA/JOH] [29HAR/BOS] [58HAR/OWE] [54FEI/HAR] [86TOD/HAS], [86TOD/HAS], [94HAS/TOD], [94HAS/TOD] [73EDW/FAR] [74RUT/HAU] [73HUL/DES] [55TOM/DOM] [77HEA/HEF] [70BIN/HEB] [88HED] [72BON/HEF], [74HEF], [77HEA/HEF] [82GAM/REI] [56CUT/KSA] [74HEL/KIR], [74HEL/KIR2], [81HEL/KIR], [88SHO/HEL], [88TAN/HEL], [89SHO/HEL], [89SHO/HEL2], [90OEL/HEL], [92SHO/OEL] [2002MON/HEL] [84ROB/HEM], [90HEM], [95ROB/HEM] [2001 HEN/RED] [30ROD/HEN] [63KO/HEP], [66GOL/RID], [68LAR/CER], [78SPI/SIN], [78SPI/SIN2], [79SPI/OLO], [89HOV/HEP], [96HEP/HOV] [67BIN/STR] [68HER/FEB] [79WEI/HER], [79WEI/MUL] [72POL/HER] [98RAI/FEL] [58TAY/HEY] [85FER/GRE] [73HUT/HIG] (Continued on next page)
Hatfield, J. D. Haury, G. L. Hawkins, D. T. Hay, R. G. Heath, G. A. Hebert, T. H. Hedlund, T. Hefter, G. Heindl, R. Hejtmanek, M. Helgeson, H. C. Helm, L. Hemingway, B. S. Henderson, C. M. B. Henry, W. F. Hepler, L. G. Herbert, T. H. Herrick, C. C. Hertz, H. G. Herweijer, A. Hess, N. J. Hey ding, R. D. Hietanen, S. Higginson, W. C. E.
LIST OF CITED A UTHORS
Author Hofer, F. Hoffmann, H. Hoffmann, M. M. Hohl, H. Holmes, H. F. Holmes, R. D. Homann, K. Hood, G. C. Hopkins, H. P. Homung, E. W. Hostettler, J. D. Hounslow, A. W. Hovey, J. K. Howell, I. Hsieh, K. Hu, D. C. Hubbard, W. N. Huber, C. O. Huebner, J. S. Hugel, R. Hultgren, R. Hume, D. N. Hummel, W. Hummelstedt, L. Hunt, J. P. Hutchinson, M. H. Hutta, J. J. Hyams, D. Ichihashi, M. Imanakunov, B. I. In Hefter, G. T. Ingraham, T. R. Ipser, H. Isaacs, T. Isabaev, A. S. Isabaev, S. M. Isono, T. Ito, J. Itoh, C. T. Iuliano, M. Ivanova, L. V. Iwase, M. Izatt, R. M. Jablonski, M.
597
Reference [65SCH/ALT] [74DIC/HOF] [99HOF/DAR] [71HOH/GEI] [83HOL/MES] [86HOL/NEI] [93MIL/CVI] [51HOO/WOY] [66GOL/RID] [56HOR/BRA] [84HOS] [59WIL/THR] [89HOV/HEP], [96HEP/HOV] [97HOW/NEI] [86HSI/CHA] [78LIN/HU] [67RUD/DEV] [78KAR/HUB] [70HUE/SAT] [71 PIE/HUG] [73HUL/DES] [50HUM/KOL], [51DEF/HUM], [64GEE/HUM], [72RET/HUM] [97ALL/BAN], [2003BAE/BRA], [2003HUM/CUR] [71PAA/HUM] [68LIN/APR], [78BEC/LIU], [83GOW/DOD] [73HUT/HIG] [63 PHI/HUT] [97HYA] [82WAK/ICH] [83KIM/IMA] [2003GAM/KON] [66ING], [70NAG/ING] [77LIN/IPS] [63ISA] [93MUL/ISA] [93MUL/ISA], [95MUL/CHU] [71 ISO] [84ROB/HEM] [88NIS/ITO] [87CIA/IUL], [89IUL/POR] [66BUR/IVA] [78IWA/FUJ] [63CHR/IZA], [69IZA/EAT], [71IZA/JOH] [93PRZ/WIS] (Continued on next page)
598
LIST OF CITED A UTHORS
Author Jacob, K. T. Jaffe, I. Jager, H. Jain, M. Jakob, A. Jamieson, J. W. S. Janjic, T. Jauregui, E. A. Jeanloz, R. Jellinek, K.
Reference [86JAC/KAL] [52ROS/WAG] [93OBE/KAS] [98JAI/ELM] [97ALL/BAN] [65JAM/BRO] [74DIC/HOF] [63BOL/JAU] [90PAS/TAY] [25JEL/ZAK], [26JEL/ULO], [26JEL/ULO2], [28JEL/RUD] [56JEN/PRA] [71 JEN] [96SAH/SAH] [23JOB/SAM] [36JOB] [04STE/JOH] [92SHO/OEL] [79JOH/PYT] [71IZA/JOH] [74SLO/JON] [60MCC/JON], [63PEN/BAI] [89ZIE/JON] [54ORI/JON] [85BAR/PAR] [90JUR/DOM] [74KAB] [81BAN/KAD] [95KAH/CRO] [75KAK/GIE] [86JAC/KAL], [94KAL/FRA] [66KAL/PUR] [64KOS/KAL] [93MIL/CVI] [78BUR/KAM] [97NOZ/KOB] [66SHI/FUJ] [36KAP/SIL] [73CHU/DRA] [97 ALL/BAN] [2001LAW/ZHA] [75CHA/KAR] [85DRA/MAD] [76KAR/KUL] [68AND/KHA] (Continued on next page)
Jena, P. K. Jensen, K. A. Ji, L.-N. Job, A. Job, P. Johnson, F. M. G. Johnson, J. W. Johnson, K. S. Johnston, H. D. Jones, G. P. Jones, L. H. Jones, M. E. Jones, T. S. Jordan, J. Juraitis, K. R. Kabir-Ud-Din Kaden, T. A. Kahn, M. A. Kakokowicz, W. Kale, G. M. Kalinichenko, 1.1. Kalinkina, I. N. Kallay, N. Kamenetskaya, N. V. Kamimura, T. Kanchiku, Y. Kapustinsky, A. F. Karapet'yants, M. K. Karapiperis, T. Karato, S. Karkhanavala, M. D. Karlina, O. Karlsson, R. Karpinskaya, N. M.
LIST OF CITED A UTHORS
Author Karwan, T. Karweik, D. H. Kasaoka, S. Kaschnitz, E. Kasenov, B. K. Kasper, R. B. Katayama, I. Katayama, S. Katsumata, K. Kawai, T. Kawakami, M. Kazybaev, S. A. Keesom, W. H. Kehiaian, H. V. Kelley, K. K. Kellogg, H. H. Kemori, N. Kennedy, B. Kennedy, M. B. Kent, R. A. Kenttamaa, J. Kepert, D. L. Kerr, P. F. Ketclaar, J. A. A. Khachkuruzov, G. A. Khaldin, V. G. Khan, G. A. Khater, E. M. H. Kher, V. G. Khodakovsky, I. L. Kildahl, N. K. Kim, J. I. Kim, J. J. Kim, T. P. King, E. G. Kipp, R. L. Kirkham, D. H. Kisova, L. Kitayama, K. Kiukkola, K. Kivalo, P. Klaaijsen, F. W. Kleinclauss, J.
599
Reference [78KAR/MAL] [78KAR/HUB] [77KAS/SAK] [93OBE/KAS] [89ART/KAS] [76NAV/KAS] [79KEM/KAT] [73KAT], [77KAT] [83KAT/SUG] [73KAW/OTS] [77SAD/KAW] [83KIM/IMA] [35KEE/CLA] [96KEH] [35KEL/AND], [60KEL], [61KEL/KIN], [64WEL/KEL], [73HUL/DES] [69KEL], [75MEY/WAR] [79KEM/KAT] [69KEN2] [66KEN/LIS] [64EHL/KEN] [56KEN], [58KEN], [59KEN] [75CLA/KEP] [45KER] [34KET] [82GLU/GUR] [68AND/KHA], [68AND/KHA2] [59SPE/FOR] [98ELM/GAB] [79NAN/DES] [92FUG/KHO] [69KOL/KIL] [97KON/FAN] [74PIT/KIM] [83KIM/IMA] [54BOY/KIN], [57KIN], [58KIN/CHR], [61KEL/KIN], [65ADA/KIN] [57MOR/ZEL] [74HEL/KIR], [74HEL/KIR2], [81HEL/KIR] [59KIS/CUP] [89KIT] [57KIU/WAG] [57KIV/LUO] [72KLA/DOK] [77BIL/TER] (Continued on next page)
600
LIST OF CITED A UTHORS
Author Klepp, K. O. Kleppa, O. J. Kloger, W. Klygin, A. E. Knacke, O. Knauer, H. Knight, K. S. Ko, H. C. Kobayashi, H. Koekoek, F. R. J. Kohler, H. Kohler, K. Kohlhaas, R. Kohls, D. W. Kokurin, N. I. Kolonin, G. R. Kolski, G. B. Kolthoff, I. M. Kolyada, N. S. Komarek, K. L. Konigsberger, E.
Reference [72KLE/KOM] [68NAV/KLE], [86CEM/KLE], [87CEM/KLE] [82KLO] [71KLY/GLE] [73BAR/KNA], [77BAR/KNA] [39SIM/KNA] [2001 HEN/RED] [63KO/HEP], [78STU/FER] [97NOZ/KOB] [75WEE/KOE] [79GOL/KOH] [64KOH/ZAS] [66VOL/KOH] [66KOH/ROD] [84VAS/VAS] [91STO/GRO] [68KOL/MAR], [69KOL/KIL] [31KOL], [50HUM/KOL] [71KLY/GLE] [72KLE/KOM] [95KON/ERI], [98GAM/KON], [99KON/KON], [2003GAM/KON] [99KON/KON] [90COR/KON2], [92BAL/DIC], [92GRE/FUG] [97KON/FAN] [58YAT/KOR] [95MAN/KOR] [92VID/KOR] [69SOR/KOS] [64KOS/KAL] [68KOS], [69KOS], [75KOS] [78KOT/MUL] [78LIB/KOW], [78LIB/MAC] [98ROG/KOZ], [2000ROG/KOZ] [69MOR/SAT], [79KEM/KAT] [55KRA/WAR] [76DAY/DIN] [90DYR/KRE] [84ROB/HEM] [70MIR/MAK] [56CUT/KSA] [70TYS/KSE] [72KUB], [77BAR/KNA], [93KUB/ALC] [93MIL/CVI] [93MAK/VOE] (Continued on next page)
Konigsberger, L.-C. Konings, R. J. M. Konnecke, T. Korableva, V. D. Korolev, V. P. Korotkevich, 1.1. Kosaki, A. Kostryukov, V. N. Kostryukova, M. O. Kother, W. Kowalewska, G. Kozlowska-R6g, A. Kozuka, Z. Krauss, F. Krausz E. R. Kremling, K. Krupka, K. M. Kryzhanovskii, M. M. Ksandr, Z. Ksenzhek, O. S. Kubaschewski, O. Kuchitsu, K. Kudasheva, T. V.
LIST OF CITED A UTHORS
Author Kudryashova, Z. P. Kuindersma, S. R. Kukhtina, V. A. Kulagov, E. A. Kullberg, L. Kullerud, G. Kul'vinova, L. A. Kurakane, K. Kutner, W. Laffitte, M. Lagarde, P. Laitinen, H. A. Lambert, S. M. LaMontagne, R. A. Landers, L. Lange, E. Lange, N. Langmuir, D. Larson, J. W. Larson, T. M. Larsson, R. Latimer, W. M. Lavut, E. G. Lawlis, J. D. Lebedev, B. G. Lebedev, S. V. LeBlanc, J. C. Lee, S. Y. Lee, W. H. Leegard, T. Leigh, G. J. Lekae, V. A. Lemire, R. J. Leserf, A. Levin, E. M. Levine, S. Levitskii, V. A. Lewis, G. N. Liberti, A. Libowitz, G. G. Libus, W. Libus, Z. Licheri, G.
601
Reference [70EFI/KUD] [81KUI/SAN] [73FED/SHM] [69KUL/ERS] [74KUL3], [76KAR/KUL] [62KUL/YUN] [76KUL/BLO], [81KUL/BLO], [83AVR/BLO] [76MUR/KUR] [74KUT/GAL] [59LAF], [63LIN/LAF], [82LAF] [80LAG/FON] [70CAR/LAI] [57LAM/WAT] [65JAM/BRO] [71 LAN/CUM] [56LAN/MIE], [59LAN] [97PRA/LAN] [99LAN/MAH] [68LAR/CER] [74BIX/LAR] [56AHR/LAR], [62FRO/LAR], [64LAR] [51LAT], [52LAT] [84LAV/TIM] [2001LAW/ZHA] [61LEB/LEV] [71LEB/SAV] [80TRE/LEB2] [90LEE/NAS] [79LEE/WHE] [64LEE/ROS] [90LEI] [71KLY/GLE] [92GRE/FUG], [95LEM], [96LEM/CAM], [2001LEM/FUG] [87CUD/LEC] [69LEV/ROB] [52ROS/WAG] [61LEB/LEV], [75LEV/GOL] [61 LEW/RAN] [64GRI/LIB] [69LIG/LIB] [80LIB/SAD], [82SAD/LIB] [69LIB/SAD], [73LIB], [75LIB/TIA], [78LIB/KOW], [78LIB/MAC], [80LIB/SAD] [80CAM/LIC], [83LIC/PAS], [85MAG/MOR] (Continued on next page)
602
LIST OF CITED A UTHORS
Author Licht, S. Lightstone, J. B. Lileev, A. S. Lilic, L. S. Lilich, L. S. Lin, R. Y. Lincoln, S. F. Line, G. Linke, W. F. Lister, M. W. Liu, G. Livingstone, A. Lobo, V. M. M. Lobry de Bruyn, C. A. Lofgren, N. L. Logsdon, K. M. Long, F. A. Loomis, J. S. Lopez-Lopez, A. Lotmar, W. Luber, A. Luoto, R. Luther III, G. W. Lutz, H. D. Lyashchenko, A. K. Ma, C.-B. Mac Arthur, D. M. MacDonald, A. MacDonald, R. D. Maciejewski, W. Madanat, A. T. Magini, M. Mah, A. D. Mahapatra, P. P. Mahoney, J. Mainard, R. Mainwaring, P. R. Major, E. Makarov, G. V. Makarow, L. L. Makashev, Y. A. Makovskaya, G. V. Maksimova, I. N. Malatesta, F. Malik, O. P.
Reference [88LIC2] [69LIG/LIB] [89LIL/TEP] [65BUR/LIL], [65BUR/LIL2] [71BUR/ZIN] [77LIN/IPS], [78LIN/HU] [68LIN/APR] [63LIN/LAF] [65LIN/SEI] [60LIS/ROS], [61LIS/WIL], [66KEN/LIS] [78BEC/LIU] [82LIV/BIS] [89LOB/QUA], [89LOB] [18SMI/LOB] [66LOF/MCI] [83 SOL/BON] [51LON] [7 ICON/LOO] [62LOP] [36LOT/FEI] [30MUL/LUB] [57KIV/LUO] [96LUT/RIC] [97PRA/LAN] [89LIL/TEP] [74MA] [70MAC] [99LAN/MAH] [77NIC/ROB] [78LIB/MAC] [85DRA/MAD] [82MAG/PAS], [85MAG/MOR] [76MAH/PAN] [80CHI/SAB], [82CHI/SAB] [99LAN/MAH] [80ADA/BIL] [85SKE/MAI] [42MAJ] [93MAK7VOE] [79OJK/MAK] [70MIR/MAK], [74BLO/RAZ], [81KUL/BLO] [69MAK/SPI] [74MAK/MAS] [97MAL/ZAM], [99MAL/CAR] [75ARN/MAL] (Continued on next page)
LIST OF CITED A UTHORS
Author Malinowski, C. Mal'kov, I. V. Malyavinskaya, O. N. Manin, N. G. Mapother, D. E. Marchal, G. Marcopoulos, T. Margerum, D. W. Margrave, J. L. Marshall, W. L. Martell, A. E. Marumo, F. Masaki, K. Mashkov, L. V. Mashovets, V. P. Masuda, I. Matarrese, L. M. Matthews, M. A. Mattigod, S. V. Mavrina, I. Y. Mayorga, G. Mazo, L. H. McBreen, J. McCormick, D. B. McCullough, R. L. McCurdy, K. G. McDonald, H. J. Mclver, E. J. McMurdie, H. F. McRae, B. R. Meade, C. Medvedev, V. A. Mehrotra, G. M. Meisel, K. Merbach, A. E. Meriel, P. Merkley, E. D. Merrill, P. W. Mesmer, R. E. Meusser, A. Meyer, G. A. Meyer, R. T.
603
Reference [78KAR/MAL] [79VOR/VAS] [68MAL/TUR], [71TUR/MAL] [95MAN/KOR] [7 ICON/LOO] [25MAR] [81MAR/ECO], [83ECO/MAR] [61MAR/BYD], [68KOL/MAR], [69KOL/KIL] [58FAR/MEY], [64EHL/KEN] [84MAR/MES] [52CHA/COU], [64SIL/MAR], [71SIL/MAR], [76SMI/MAR] [74YAG/MAR] [32MAS] [75BAR/MAS] [74MAK/MAS] [82WAK/ICH] [54MAT/STO] [98JAI/ELM] [77MAT/SPO], [79MAT/SPO], [97MAT/RAI] [70MIR/MAK] [73PIT/MAY], [74PIT/MAY] [82NEV/ANG] [99MCB] [67SIG/BEC] [60MCC/JON] [78SPI/SIN] [40SEL/DEW] [66LOF/MCI] [69LEV/ROB] [2004BRO/MER] [90PAS/TAY] [82GLU/GUR], [88EFI/EVD], [89COX/WAG], [92FUG/KHO] [85MEH/TAR] [36BIL/VOI] [2002MON/HEL] [80ADA/BIL] [2004BRO/MER] [68MOO/MER] [74MES/BAE], [76BAE/MES], [83HOL/MES], [84MAR/MES], [91 AND/CAS] [01MEU] [75MEYAVAR] [58FAR/MEY] (Continued on next page)
604
LIST OF CITED A UTHORS
Author Mibuchi, T. Miederer, W. MigdaJ-Mikuli, A. Migliardo, P. Mikhailin, B. V. Mikuli, E. Milic,N. B. Miljutin G. Miller, A. R. Miller, M. L. Millero, F. J. Mills, I. Mills, K. C. Milton, M. J. T. Minder, W. Mironov, L. A. Mironov, V. E.
Reference [82WAK/ICH] [56LAN/MIE] [99MIK/MIG], [2001MIK/MIG] [80LAG/FON] [79VOR/VAS] [99MIK/MIG], [2001MIK/MIG] [80MIL/BUG] [36TRA/SCH] [78SCH/MIL] [56MIL/SHE] [79MIL], [94ZHA/MIL], [96CRI/MIL] [93MIL/CVI] [74MIL] [2002BUC/RON] [65SCH/ALT] [83AVR/BLO] [70MIR/MAK], [73FED/SHM], [74BLO/RAZ], [76KUL/BLO], [81KUL/BLO] [94HAS/TOD] [61MOH/DAS] [2000TSI/MOL], [2001MOL/TSI] [32MON/DAV] [49MON] [2002MON/HEL] [90MON] [55TOM/DOM] [68MOO/MER] [98RAI/FEL] [96STU/MOR2] [78IWA/FUJ] [69MOR/SAT] [71ARI/MOR] [64HAM/MOR] [65MOR/REE] [57MOR/ZEL] [24MOS/BEH] [75MOS/FIT], [75MOS/FIT2] [82MU/PER] [98PLY/ZHA] [86BHA/MUK] [95MUL/CHU] [93MUL/ISA] [58MUL] [85MUL], [92GRE/FUG] [79WEI/MUL] (Continued on next page)
Miura, K. Mohapatra, P. P. Molinou, I. E. Money, R. W. Monk, C. B. Monlien, F. J. Monnin, C. Montgomery, C. W. Moore, C. E. Moore, D. A. Morgan, J. J. Mori, T. Moriyama, J. Morozova, M. P. Morrell, M. L. Morris, D. F. C. Morris, J. P. Moser, L. Moser, Z. Mu, J. Muhammed, M. Mukherjee, R. Muldagalieva, R. A. Muldagalieva, S. A. Muldrow, C. N. Muller, A. B. Muller, C.
LIST OF CITED A UTHORS
Author Muller, E. Muller, F. Mumme, W. G. Munoz, J. L. Murai, R. Murata, K. Murray, J. J. Mussini, T. Muthmann, W. Naer-Neumann, E. Nagamori, M. Nagypal, I. Nair, V. S. K. Naito, M. Nancollas, G. H. Nandi, P. N. Napijalo, M. L. Napijalo, M. M. Nasanen, R. Nash, P. Nassau, K. Naugle, D. G. Naumann, R. Navratil, J. D. Navrotsky, A. Neck, V. Neilson, G. W. Netsvetaeva, V. S. Netzel, D. A. Neumann, B. Neuschiitz, D. Neves, E. A. Newmann, A. A. Nguyen-Trung, C. Nickel, E. H. Nieuwenkamp, W. Nikol'skaya, N. A. Nishimura, T. Nitsche, H. Nolang, B. Nordstrom, D. K. Novak, C. F. Novak-Adamic, D. M. Novakovic, L.
605
Reference [30MUL/LUB] [78KOT/MUL] [93NIC/ROB] [86NOR/MUN] [76MUR/KUR] [28MUR] [74MUR/POT] [2002BUC/RON] [04MUT/FRA] [85FER/GRE] [70NAG/ING] [90BEC/NAG], [91ZEK/NAG] [59NAI/NAN] [96UCH/GOR] [59NAI/NAN], [67NAN/TOR] [79NAN/DES] [96DOJ/NOV] [96DOJ/NOV] [43NAS] [87SIN/NAS], [90LEE/NAS] [73NAS/SHI] [79DUB/NAU] [2002BUC/RON] [92FUG/KHO] [68NAV/KLE], [71NAV], [73NAV], [73NAV2], [76NAV/KAS] [2003GUI/FAN] [83NEI/END], [97HOW/NEI] [74NET/BAT] [63NET/DRO] [1894NEU] [67GES/NEU] [82NEV/ANG] [75NEW2] [92GRE/FUG] [77NIC/ROB], [81NIC/BER], [93NIC/ROB] [33BIJ/NIE] [71KLY/GLE] [88NIS/ITO] [2001LEM/FUG] [2005OLI/NOL] [86NOR/MUN] [98ALM/NOV] [73NOV/COS] [96DOJ/NOV] (Continued on next page)
606
LIST OF CITED A UTHORS
Author Novo-Gradac, K. H. Novotny, P. Nozue, T. Nuttall, R. L. Nyholm, R. S. Oakes, C. S. Obendrauf, W. Obozova, L. A. Oelkers, E. H. Ogawa, S. Ogino, H. Ohman, L-O. Ohtaki, H. Oikova, T. Ojkova, T. Oka, Y. Okaji, M. Olette, M. Olin, A. Olofsson, I. V. Olroyd, A. Omarova, F. M. O'Neill, H. S. C. Oppermann, H. Oranskaya, M. A. Oriani, R. A. Origlia-Luster, M. L. Osadchii, E. Osadchii, E. G. Osswald, H. Osthols, E.
Reference [93ALL/BRO] [85SOH/NOV] [97NOZ/KOB] [79GOL/NUT], [82SAR/COV], [82WAG/EVA] [59GIL/NYH], [63PEN/BAI] [2001OAK/RAI] [93OBE/KAS] [75BAR/MAS] [90OEL/HEL], [92SHO/OEL] [77OGA] [61TAN/OGI], [63TAN/SAI] [2005OLI/NOL] [71OHT/BIE], [73KAW/OTS] [79OIK] [79OJK/MAK] [38OKA] [76OKA/WAT] [63GEO/FER] [2005OLI/NOL] [78SPI/SIN2], [79SPI/OLO] [96LUT/RIC] [80OMA/SHA] [86HOL/NEI], [87NEI], [93NEI/POW] [80OPP/TOS] [54SHC/TOL] [54ORI/JON] [2004BRO/MER] [90OSA/ROS] [2005OLI/NOL] [34BRI/OSS] [99WAN/OST], [2000GRE/WAN], [2000OSTAVAN], [2000WAN/OST] [77OSW/ASP] [73KAW/OTS] [90COR/KON2] [58HAR/OWE] [71PAA/HUM] [91 PAP/PIT] [63CHR/IZA] [99HOF/DAR] [2003GUI/FAN] [97PAN/CAM] [76MAH/PAN], [84PAN/STU], [84PAN] [76PAN/PAT] [65PAO/SAB], [65PAO] (Continued on next page)
Oswald, H. R. Otsuka, H. Ouweltjes, W. Owen, B. B. Paatero, J. Pabalan, R. T. Pack, R. T. Palmer, B. J. Palmer, D. A. Pan, P. Pankratz, L. B. Pant, G. N. Paoletti, O.
LIST OF CITED A UTHORS
Author Parker, G. A. Parker, V. B. Parsons, R. Paschina, G. Pascucci, G. Pasternak, M. P. Pathak, D. N. Patil, S. K. Patterson, B. A. Pauling, L. Pavlinova, L. A. Pawel, R. E. Pawlowski, T. Payne, S. L. Peacock, M. A. Pearce, J. N. Pease, R. N. Peiluck, R. V. Penneman, R. A. Perlmutter, D. D. Perlmutter-Hayman, B. Perrin, D. D. Perron, G. Persson, H. Petersen, Jr. H. Peterson, J. R. Pethybridge, A. D. Petot, C. Petot-Ervas, G. Petrov, G. V. Pfanhauser, W. Phillips, B. Piacente, V. Piccaluga, G. Pierrard, J-C. Pinna, G. Pitzer, K. S. Plyasunov, A. V. Plyasunova, N. V. Pobegun, T. Polgar, L. G.
607
Reference [97BAH/PAR] [82WAG/EVA], [87GAR/PAR] [74PAR], [85BAR/PAR] [80CAM/LIC], [82MAG/PAS], [83LIC/PAS] [76BIA/CAR] [90PAS/TAY] [76P AN/PAT] [76PAT/RAM] [2002PAT/WOO] [40PAU] [71ARI/MOR] [65PAW/STA] [91SWI/PAW] [57MOR/ZEL] [47PEA] [37PEA/ECK] [26PEA/COO] [65JAM/BRO] [60MCC/JON], [63PEN/BAI], [65COL/PET] [82MU/PER] [73PER/SEC] [64PER] [81PER/ROU] [71 PER], [74PER], [76PER/EKS], [76PER] [65COL/PET] [78PIT/PET] [80PET/TAB] [84RAY/PET] [84RAY/PET] [79PET/SHE], [79PET/SHE2], [80PET/SHE] [01PFA] [63PHI/HUT] [96BRU/PIA] [80CAM/LIC], [82MAG/PAS], [83LIC/PAS], [85MAG/MOR] [71 PIE/HUG] [80CAM/LIC], [83LIC/PAS] [73PIT/MAY], [73PIT], [74PIT/KIM], [74PIT/MAY], [75PIT], [76PIT/SILJ, [78PIT/PET], [79BRA/PIT], [79PIT], [91PAP/PIT], [91 PIT] [94PLY/GRE], [97ALL/BAN], [97GRE/PLY], [97GRE/PLY2], [97PUI/RAR] [98PLY/ZHA] [62CAI/POB] [72POL/HER] (Continued on next page)
608
LIST OF CITED A UTHORS
Author Polydoropoulos, K. Ponomareva, V. L. Popov, C. G. Popov, N. I. Porto, R. Pottel, R. Potter, P. E. Pottie, R. F. Pottlacher, G. Poulson, S. R. Pourbaix, M. Powell, H. K. J. Pownceby, M. I. Pracht, G. Prasad, B. Pratt, J. N. Pratt, K. W. Predel, B. Preis, W.
Reference [53SCH/POL] [71ARI/MOR] [75LEV/GOL] [76ZHO/BEZ] [87CIA/IUL], [89IUL/POR] [65POT] [2001LEM/FUG] [74MUR/POT] [93OBE/KAS] [96POU/DRE] [63DEL/ZOU] [73POW] [93NEI/POW] [97PRA/LAN] [56JEN/PRA], [68PRA] [84COM/PRA] [2002BUC/RON] [72PRE/RUG], [72PRE/RUG2] [98GAM/KON], [2001GAM/PRE], [2002GAM/WAL], [2002WAL/PRE] [73NAS/SHI] [74RAJ/PRE] [55BRO/PRU] [93PRZ/WIS] [78KAR/MAL] [95GRE/PUI], [95SIL/BID], [97ALL/BAN], [97PUI/RAR] [74MUR/POT] [66KAL/PUR] [79JOH/PYT], [79PYT] [89LOB/QUA] [75RAB/WAN] [97MAT/RAI], [98RAI/FEL], [2001OAK/RAI] [74RAJ/PRE] [58VAI/RAM] [86JAC/KAL] [76PAT/RAM] [83RAM/FAN], [86RAM] [95GRE/PUI], [95SIL/BID], [99RAR/RAN], [2001LEM/FUG], [2003GUI/FAN] [61 LEW/RAN] [80BAR/RAN], [81 BAR/RAN] [97MAT/RAI] [75MEY/WAR] [80LAG/FON] (Continued on next page)
Prescott, B. E. Prewitt, C. T. Prue, J. E. Przepiera, A. Ptak, W. Puigdomenech, I. Pupp, C. Purtov, A. I. Pytkowicz, R. M. Quaresma, J. L. Rabbering, G. Rai, D. Rajamani, V. Rama Char, T. L. Ramachandran, A. Ramakrishna, V. V. Ramette, R. W. Rand, M. H. Randall, M. Randell, C. F. Rao, L. Rao, Y. K. Raoux, D.
LIST OF CITED A UTHORS
Author Rapp, R. A. Rard, J. A. Rau, H. Ray, D. K. Razmyslova, L. I. Reddy, G. K. N. Redfern, S. A. T. Reed, G. L. Regnault, L. P. Regnault, V. Reiterer, F. Retajczyk, T. F. Rey, J. Richens, D. T. Richter, J. Rickard, D. T. Riddell, R. G. Riglet, C. Ringwood, A. E. Riou, A. Rizzo, F. E. Robbins, C. R. Robbins, D. J. Robie, R. A. Robinson, B. W. Robinson, R. A. Robinson, S. C. Robinson, W. K. Robouch, P. Robov, A. M. Rodda, J. L. Rodebush, W. H. Roeder, P. Rog, G. Rohmer, R. Rondini, S. Rosen, E. Rosenblatt, G. M. Rosenblum, P. Rosengren, K. Rosenqvist, T. Rossat-Mignod, J. Rossetti-Francois, J. Rossini, F. D.
609
Reference [63RAP] [87RAR3], [92RAR2], [97ALL/BAN], [97PUI/RAR], [98ARC/RAR], [99RAR/RAN] [73RAU/GUE], [75RAU2], [76RAU] [84RAY/PET] [74BLO/RAZ] [63PEN/BAI] [2001 HEN/RED] [65MOR/REE] [80ADA/BIL], [81ADA/BIL] [1841 REG] [80REI], [82GAM/REI] [72RET/HUM] [80DUB/CLA] [97RIC] [79RJC/VRE] [96LUT/RIC] [66GOL/RID] [87RIG/VIT], [89RIG/ROB], [90RIG] [62RIN] [87CUD/LEC] [67RIZ/BID] [69LEV/ROB] [76DAY/DIN] [84ROB/HEM], [95ROB/HEM] [77NIC/ROB], [93NIC/ROB] [49ROB/STO], [59ROB/STO] [52ROB/BRO] [60ROB/FRI] [89RIG/ROB], [89ROB], [95SIL/BID] [73FED/SHM] [66KOH/ROD] [30ROD/HEN] [68CAM/ROE] [84ROG/BOR], [98ROG/KOZ], [2000ROG/KOZ] [39ROH] [2002BUC/RON] [90OSA/ROS], [2005OLI/NOL] [72BEH/ROS] [60LIS/ROS] [56AHR/LAR], [56AHR/ROS] [54ROS], [64LEE/ROS] [80ADA/BIL], [81ADA/BIL] [52ROS] [36BIC/ROS], [52ROS/WAG] (Continued on next page)
610
LIST OF CITED A UTHORS
Author Rossotti, F. J. C. Rossotti, H. Roth, C. Roux, A. Rowson, J. Rubin, J. J. Rudat, A. Rudzitis, E. Ruge, H. Rutner, E. Ruvinskii, O. E. Rydberg, J. Sabat, B. B. Sabatini, A. Sadakane, K. Sadoc, A. Sadowska, T. Sage, M. W. Saha, A. Saha, N. Saito, Y. Saitton, B. Sajadi, S. A. A. Sakata, Y. Salvatore, F. Samuel, A. Sanchez, J. P. Sandino, M. C. A. Sano,K. Sano, W. Saprykova, Z. A. Sarbar, M. Sato, M. Sato, N. Satyanarayan, D. Sawatimskii, A. I. Saxena, S. Scarrow, R. K. Scatchard, G. Schaefer, S. C. Schenck, R. Schiffman, R. A. Schildbach, R. Schindler, P. Schlesinger, M. E.
Reference [59ROS/ROS], [61ROS/ROS] [59ROS/ROS], [61ROS/ROS], [69ROS] [70ROT] [81PER/ROU] [99LAN/MAH] [65UIT/WIL] [28JEL/RUD] [67RUD/DEV] [72PRE/RUG], [72PRE/RUG2] [74RUT/HAU] [70TUR/RUV] [2001LEM/FUG] [80CHI/SAB], [82CHI/SAB] [65PAO/SAB] [77SAD/KAW] [80LAG/FON] [69LIB/SAD], [80LIB/SAD], [82SAD/LIB] [08THO/SAG] [96SAH/SAH] [96SAH/SAH] [63TAN/SAI] [90OSA/ROS] [96SAH/SAH] [77KAS/SAK] [83FER7GRE], [85FER/GRE] [23JOB/SAM], [43SAM] [81KUI/SAN] [95GRE/PUI] [37SAN], [37SAN2], [53SAN] [90JUR/DOM] [85SAP/CHI] [82SAR/COV] [70HUE/SAT], [97NOZ/KOB] [69MOR/SAT] [7 IDAS/DAS] [71LEB/SAV] [97ALL/BAN] [69GRI/SCA] [36SCA], [76SCA] [81SCH] [39SCH/FOR] [78SCH/MIL] [10SCH] [63FEI/SCH], [63SCH], [65SCH/ALT] (Continued on next page)
LIST OF CITED A UTHORS
Author Schmalzried, H. Schneider, C. R. Schoch, E. P. Schott, J. Schreiner, L. Schubnikow, L. Schuijff, A. Schumb, W. C. Schumm, R. H. Schwab, G.-M. Schweitzer, A. Scott, S. D. Secco, F. Seidell, A. Seim, H. Seki, S. Sekine, T. Seltz, H. Seminara, A. Sergeyeva, E. I. Shankar, J. Shannon, R. D. Sharipov, M. S. Sharma, R. C. Shchigol, M. B. Shchukarev, S. A. Shehata, H. A. Sheka, Z. A. Shelton, R. A. J. Sheridan, X. Sherwood, R. C. Shevchuk, V. G. Shibata, N. Shiever, J. W. Shimizu, K. Shirata, M. Shmyd'ko, 1.1. Shock, E. L. Short, E. L. Shukla, A. K. Siemens, A. Sigel, H. Silberman, A. Sillen, L. G.
611
Reference [64TAY/SCH], [65TRE/SCH] [59FRE/SCH] [09SCH] [91 AND/CAS] [34SIE/SCH] [36TRA7SCH] [75RAB/WAN] [23SCH] [82WAG/EVA] [53SCH/POL] [09SCH2] [76CRA/SCO] [73PER/SEC] [65LIN/SEI] [94STO/FJE], [96SEI/FJE] [69SOR/KOS] [76MUR/KUR] [40SEL/DEW] [78SIR/SEM] [92FUG/KHO] [63SHA/DES] [76SHA] [80OMA/SHA] [79SHA/CHA], [80SHA/CHA] [61SHC] [54SHC/TOL] [87EMA/FAR], [93SHE] [56FIA/SHE] [76GEE/SHE] [56MIL/SHE] [65UIT/WIL] [79PET/SHE], [79PET/SHE2], [80PET/SHE] [66SHI/FUJ] [73NAS/SHI] [76SHI/TSU] [77KAS/SAK] [73FED/SHM] [88SHO/HEL], [89SHO/HEL], [89SHO/HEL2], [92SHO/OEL] [65MOR/REE] [86JAC/KAL] [04SIE] [67SIG/BEC], [81BAN/KAD], [96SAH/SAH] [36KAP/SIL] [64SIL/MAR], [65BUR/LIL], [71SIL/MAR] (Continued on next page)
612
LIST OF CITED A UTHORS
Author Silva, R. J. Silverman, D. C. Silvester, L. F. Simaeva, L. S. Simon, A. Simonson, J. M. Singh, P. P. Singleton, M. Siracusa, G. Skapsi, A. Skeaff, J. M. Skibsted, L. H. Slater, D. N. Slough, W. Sludskaja, N. P. Smirnov, Y. B. Smirnova, I. D. Smirnova, M. F. Smirnova, M. N. Smith, MacF. W. Smith, R. I. Smith, R. M. Smith-Magowan, D. Smits, A. Smoes, S. Smurov, A. A. Sohnel, O. Sokolova, M. A. Solechnik, N. D. Solomon, L. R. Song, B. Sorai, M. Spahiu, K.
Reference [95SIL/BID] [81SIL] [76PIT/SIL], [78PIT/PET] [73FED/SHM] [39SIM/KNA] [2002CHI/SIM] [78SPI/SIN], [78SPI/SIN2], [79SPI/OLO] [87SIN/NAS] [73GUR/CAL], [78SIR/SEM] [32SKA/DAB] [73SKE/ESP], [85SKE/MAI] [86BJE/SK.I] [65MOR/REE] [74SLO/JON] [53GRI/SLU] [71LEB/SAV] [71KLY/GLE] [68AND/KHA], [68AND/KHA2] [65EGO/SMI] [61DAV/SMI] [2001 HEN/RED] [76SMI/MAR] [82SMI/WOO] [18SMI/LOB] [92BAL/DIC] [38SMU] [85SOH/NOV] [56SOK] [85FIL/CHA] [83SOL/BON] [96SAH/SAH] [69SOR/KOS] [83SPA], [97ALL/BAN], [97GRE/PLY2], [2001LEM/FUG] [85SKE/MAI] [59SPE/FOR] [93KUB/ALC] [78SPI/SIN], [78SPI/SIN2], [79SPI/OLO] [2002BUC/RON] [69MAK/SPI], [91SPI/TRI] [77MAT/SPO], [79MAT/SPO] [77CON/SRI] [65PAW/STA] [79GOL/NUT] [82WHI/STA] (Continued on next page)
Speelman, J. L. Spence, R. D. Spencer, P. J. Spitzer, J. J. Spitzer, P. Spivakovskii, V. B. Sposito, G. Sridhar, R. Stansbury, E. E. Staples, B. R. Staveley, L. A. K.
LIST OF CITED A UTHORS
Author Staveley, L. A. W. Steele, B. C. H. Steele, B. D. Stokes, R. H. St0len, S. Stolyrova, T. A. Stout, J. W. Strohmenger, J. M. Stuehr, J. E. Stumm, W. Stuve, J. M. Subbaraman, P. R. Subrahmanya, R. S. Sudo, K. Suga, H. Sugar, J. Sugiyama, K. Sukiennik, M. Sullivan, J. C. Suzuki T. Sverjensky, D. A. Swanson, H. E. Swiatek, J. Sykes, C. Taba, S. S. Tamamushi, R. Tanaka, M. Tanaka, N. Tanger, IV, J. C. Taniewska-Osinska, S. Tantzov, N. V. Tare, V. B. Tareen, J. A. K. Tatge, E. Taylor, G. Taylor, J. B. Taylor, J. R. Taylor, R. D. Taylor, R. S. Taylor, R. W. Tepavitcharova, S. Terrier, C.
613
Reference [74WOR/COW] [65STE/ALC] [04STE/JOH] [49ROB/STO], [59ROB/STO], [91STO] [91STO/GRO], [94STO/FJE], [95GRO/STO], [96SEI/FJE] [81STO], [82STO] [41STO/GIA], [54MAT/STO], [54STO/CAT], [55CAT/STO], [55STO/CAT], [64STO/HAD], [66STO/ARC] [67BIN/STR] [72FRE/STU], [78FRE/STU] [96STU/MOR2] [78STU/FER], [84PAN/STU] [69SUB/COR] [71BHA/SUB] [52SUD], [64ALC/SUD] [69SOR/KOS] [81COR/SUG] [83KAT/SUG] [78KAR/MAL] [2001LEM/FUG] [97NOZ/KOB] [89SHO/HEL2], [92SHO/OEL] [53SWA/TAT] [91SWI/PAW] [38SYK/WIL] [80PET/TAB] [75TAM] [69FUN/TAN] [61TAN/0GI], [63TAN/SAI] [88TAN/HEL] [61TAN2] [24TAN] [85MEH/TAR] [91TAR/FAZ] [53 SWA/TAT] [59SPE/FOR] [5 8T AY/HEY] [82TAY] [90PAS/TAY] [76TAY/DIE] [64TAY/SCH] [89LIL/TEP] [77BIL/TER], [80ADA/BIL], [81ADA/BIL] (Continued on next page)
614
LIST OF CITED A UTHORS
Author Theberge, S. Thiel, A. Thiemann, W. Thoenen, T. Thompson, D. M. Thomsen, J. Threadgold, I. M. Tialowska, H. Tilden, W. A. Tillett, D. M. Timofeeva, T. G. Timofeyev, B. I. Toda, Y. Tolmacheva, T. A. Tomlinson, J. R. Torrance, K. Toschew, A. Toth, I. Tozawa, K. Tran, T. Trapeznikowa, O. Tremaine, P. R. Tremillon, B. Tretjakow, J. D. Tribalat, S. Trizna, L. G. Trojan, M. Tsertsov, V. N. Tsierkezos, N. G. Tsuchihashi, N. Tsujikawa, I. Tuck, D. G. Turner, D. R. Tur'yan, Y. I. Tye, F. L. Tysyachnyi, V. P. Uchida, E. Uesawa, A. Ullman, W. J. Uloth, R. Uvaliev, Y. K. Uvarov, L. Vacca, A. Vagramyan, A. Vaid, J.
Reference [96LUT/RIC] [14THI/GES] [34BEN/THI] [99THO] [08THO/SAG] [1883THO], [06THO] [59WIL/THR], [63THR] [75LIB/TIA] [02TIL], [02TIL2] [82SMI/WOO] [93UV A/TIM] [84LAV/TIM] [86TOD/HAS], [94HAS/TOD] [54SHC/TOL] [55TOM/DOM] [67N AN/TOR] [80OPP/TOS] [92BAN/BLI] [88NIS/ITO] [80TRA/BRU] [36TRA/SCH] [80TRE/LEB2] [58TRE] [65TRE/SCH] [62TRI/CAL], [64TRI/CAL] [91SPI/TRI] [90TRO] [75LEV/GOL] [2000TSI/MOL], [2001MOL/TSI] [76SHI/TSU] [78ENO/TSU] [97BAH/PAR] [81TUR/WHI] [68MAL/TUR], [70TUR/RUV], [71TUR/MAL] [80BAR/RAN], [81 BAR/RAN] [70TYS/KSE] [96UCH/GOR] [97NOZ/KOB] [2001LEM/FUG] [26JEL/ULO], [26JEL/ULO2] [93UVA/TIM] [62VAG/TJVA] [65PAO/SAB] [62VAG/UVA] [58VAI/RAM] (Continued on next page)
LIST OF CITED A UTHORS
Author Valla, E. Van Deventer, E. H. Van Geet, A. L. Van Uitert, L. G. Vanden Bosche, E. G. Vannerberg, N.-G. Varet, R. Vasca, E. Vasic, M. V. Vasileff, H. D. Vasilev, V. A. Vasil'ev, V. P. Vasil'eva, V. N. Vasilyev, V. P. Veits, I. V. Venal, W. V. Vidavskii, L. M. Vierling, F. Vilbrandt, F. C. Vitorge, P. Voevoda, N. Y. Vogel, W. M. Vogt, N. Voigt, A. Vollmer, O. Von Brintzinger, H. Von der Forst, P. Von Dickert, F. Von Gesell, H. Von Kiss, A. V. Von Sieverts, A. Vorob'ev, A. F. Vreuls, W. Wagman, D. D. Wagner, C. Wagner, Jr., J. B. Wakita, H. Wallner, H. Wang, P.
615
Reference [11AGE/VAL] [67RUD/DEV] [64GEE/HUM] [65UIT/WIL] [29HAR/BOS] [70HAL/VAN] [1896VAR] [89IUL/POR] [80MIL/BUG] [50VAS/GRA] [79VOR/VAS] [56YAT/VAS2], [62VAS], [84VAS/VAS], [86VAS/DMI] [84VAS/VAS], [86VAS/DMI] [56YAT/VAS] [82GLU/GUR] [73VEN/GEI] [92VID/KOR] [95KAH/CRO] [23VIL/BEN] [87RIG/VIT], [89RIG/ROB], [93GIF/VIT2], [94CAP/VIT], [95CAP/VIT], [2001LEM/FUG] [93MAK7VOE] [68VOG] [2001VOG] [36BIL/VOI] [66VOL/KOH] [34BRI/OSS] [39SCH/FOR] [74DIC/HOF] [67GES/NEU] [41KIS/ACS] [34SIE/SCH] [79VOR/VAS] [79RIC/VRE] [52ROS/WAG], [73HUL/DES], [82WAG/EVA], [89COX/WAG] [57KIU/WAG] [85MEH/TAR] [82WAK/ICH] [2001GAM/PRE], [2002GAM/WAL], [2002WAL/GAT], [2002WAL/PRE] [90ARC/WAN] (Continued on next page)
616
LIST OF CITED A UTHORS
Author Wanner, H.
Reference [88WAN], [91WAN], [92GRE/FUG], [95SIL/BID], [99RAR/RAN], [99WAN/OST], [2000GRE/WAN], [2000OST/WAN], [2000WAN/OST], [2001LEM/FUG] [75RAB/WAN] [67ANT/WAR] [55KRA/WAR] [75MEY/WAR], [77CON/SRI] [63PHI/HUT] [82WAT], [93WAT] [33WAT] [76OKA/WAT] [65MOR/REE] [63CHR/IZA] [57LAM/WAT] [75WEE/KOE] [36BIL/VOI] [98JAI/ELM] [72AUF/CAR], [72AUF/CAR2] [07WEI] [79WEI/HER], [79WEI/MUL] [42FRI/WEI] [64WEL/KEL], [65WEL] [74WEN] [38WES] [91STO/GRO], [94STO/FJE] [79LEE/WHE] [79WHI2] [74WHI] [87GAR/PAR] [82WHI/STA] [79WHI], [81TUR/WHI] [92WHI] [38SYK/WIL] [27JON/WIL] [74GRA7WIL] [65UIT/WIL] [59WIL/THR], [60WIL2] [62WIL2] [36BRO/WIL] [61LIS/WIL] [2002BUC/RON] [66GOL/RID] [93PRZ/WIS] [66BOD/DEH] [21WOH/BAL] (Continued on next page)
Wanrooy, J. Warburton, J. B. Warncke, H. Warner, J. S. Warshaw, I. Watanabe, H. Watanabe, M. Watanabe, T. Waters, D. N. Watt, G. D. Watters, J. I. Weenk, J. W. Weibke, F. Weidner, J. W. Weigel, D. Weigel, O. Weingartner, H. Weitbrecht, G. Weller, W. W. Wendlandt, W. W. Westgren, A. Westrum, Jr., E. F. Wheaton, R. J. Whiffen, D. H. White, H. W. White, Jr., H. J. White, M. A. Whitfield, M. Whiting, K. S. Wilkinson, H. Willems, H. W. V. Williams, D. R. Williams, H. J. Williams, K. L. Williams, T. Wilson, A. J. C. Wilson, D. W. Wilson, G. S. Wingard, M. R. Wisniewski, M. Witte, J. Wohler, L.
LIST OF CITED A UTHORS
Author Wood, R. H. Woolley, E. M. Woontner, L. Worswick, R. D. Woyski, M. M. Wrobel, S. Wulff, C. A. Yagi, T. Yamatera, H. Yashkova, V. I. Yashko-Zakharova, O. E. Yatsimirskii, K. B. Yngve, V. Yokoyama, H. Yoshihara, K. Yui, M. Yuldasheva, V. M. Yund, R. A. Yungman, V. S. Zador, S. Zakowski, J. Zakulski, W. Zamboni, R. Zaske, P. Zavrazhnova, D. M. Zawadzki, J. Zekany, L. Zellars, G. R. Zhambekov, M. I. Zhang, J.-Z. Zhang, Y. Zhao, J. Zhao, Y.-H. Zhorov, V. A. Ziemniak, S. E. Zinevich, N. I. Zirel'nikov, V. I.
Reference [82SMI/WOO] [2002PAT/WOO], [2004BRO/MER] [52GAY/WOO] [74WOR/COW] [51H00/W0Y] [99MIK/MIG] [66GOL/RID] [74YAG/MAR] [81YOK/YAM] [86VAS/DMI] [69KUL/ERS] [56YAT/VAS], [56YAT/VAS2], [58YAT/KOR] [16DER/YNG] [81Y0K/YAM] [66SHI/FUJ] [98RAI/FEL] [84LAV/TIM] [61YUN], [62KUL/YUN] [82GLU/GUR] [64ALC/SUD] [25JEL/ZAK] [75MOS/FIT] [97MAL/ZAM] [64KOH/ZAS] [71KLY/GLE] [09BRU/ZAW] [91ZEK/NAG] [57MOR/ZEL] [93MUL/ISA] [94ZHA/MIL] [98PLY/ZHA] [96SAH/SAH] [2001LAW/ZHA] [76ZHO/BEZ] [89ZIE/JON] [71BUR/ZIN] [88EFI/EVD]
617
This Page Intentionally Left Blank
