Layered Double Hydroxides (Structure and Bonding) (Structure and Bonding)

119 Structure and Bonding Series Editor: D. M. P. Mingos

Editorial Board: P. Day · T. J. Meyer · H. W. Roesky · J.-P. Sauvage

Structure and Bonding Series Editor: D. M. P. Mingos Recently Published and Forthcoming Volumes

Layered Double Hydroxides Volume Editors: Duan, X., Evans, D. G. Vol. 119, 2005 Semiconductor Nanocrystals and Silicate Nanoparticles Volume Editors: Peng, X., Mingos, D. M. P. Vol. 118, 2005 Magnetic Functions Beyond the Spin-Hamiltonian Volume Editor: Mingos, D. M. P. Vol. 117, 2005 Intermolecular Forces and Clusters II Volume Editor: Wales, D. J. Vol. 116, 2005 Intermolecular Forces and Clusters I Volume Editor: Wales, D. J. Vol. 115, 2005 Superconductivity in Complex Systems Volume Editor: Müller, K. A. Vol. 114, 2005 Principles and Applications of Density Functional Theory in Inorganic Chemistry II Volume Editors: Kaltsoyannis, N., McGrady, J. E. Vol. 113, 2004 Principles and Applications of Density Functional Theory in Inorganic Chemistry I Volume Editors: Kaltsoyannis, N., McGrady, J. E. Vol. 112, 2004

Supramolecular Assembly via Hydrogen Bonds II Volume Editor: Mingos, D. M. P. Vol. 111, 2004 Applications of Evolutionary Computation in Chemistry Volume Editors: Johnston, R. L. Vol. 110, 2004 Fullerene-Based Materials Structures and Properties Volume Editor: Prassides, K. Vol. 109, 2004 Supramolecular Assembly via Hydrogen Bonds I Volume Editor: Mingos, D. M. P. Vol. 108, 2004 Optical Spectra and Chemical Bonding in Transition Metal Complexes Special Volume II dedicated to Professor Jørgensen Volume Editor: Schönherr, T. Vol. 107, 2004 Optical Spectra and Chemical Bonding in Inorganic Compounds Special Volume I dedicated to Professor Jørgensen Volume Editor: Schönherr, T. Vol. 106, 2004 Group 13 Chemistry III Industrial Applications Volume Editors:Roesky, H. W., Atwood, D. A. Vol. 105, 2003

Layered Double Hydroxides Volume Editors: X. Duan, D. G. Evans

With contributions by X. Duan · D. G. Evans · J. He · Y. Kang · A. I. Khan F. Leroux · B. Li · F. Li · D. O’Hare · R. C. T. Slade C. Taviot-Gueho · M. Wei · G. R. Williams

123

The series Structure and Bonding publishes critical reviews on topics of research concerned with chemical structure and bonding. The scope of the series spans the entire Periodic Table. It focuses attention on new and developing areas of modern structural and theoretical chemistry such as nanostructures, molecular electronics, designed molecular solids, surfaces, metal clusters and supramolecular structures. Physical and spectroscopic techniques used to determine, examine and model structures fall within the purview of Structure and Bonding to the extent that the focus is on the scientific results obtained and not on specialist information concerning the techniques themselves. Issues associated with the development of bonding models and generalizations that illuminate the reactivity pathways and rates of chemical processes are also relevant. As a rule, contributions are specially commissioned. The editors and publishers will, however, always be pleased to receive suggestions and supplementary information. Papers are accepted for Structure and Bonding in English. In references Structure and Bonding is abbeviated Struct Bond and is cited as a journal. Springer WWW home page: http://www.springer.com Visit the SAB content at http://www.springerlink.com/

Library of Congress Control Number: 2005930165

ISSN 0081-5993 ISBN-10 3-540-28279-3 Springer Berlin Heidelberg New York ISBN-13 978-3-540-28279-2 Springer Berlin Heidelberg New York DOI 10.1007/b100426

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable for prosecution under the German Copyright Law. Springer is a part of Springer Science+Business Media springer.com c Springer-Verlag Berlin Heidelberg 2006
Printed in Germany The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Cover design: Design & Production GmbH, Heidelberg Typesetting and Production: LE-TEX Jelonek, Schmidt & Vöckler GbR, Leipzig Printed on acid-free paper 02/3141 YL – 5 4 3 2 1 0

Series Editor Prof. D. Michael P. Mingos Principal St. Edmund Hall Oxford OX1 4AR, UK [email protected]

Volume Editors Prof. David G. Evans

Prof. Xue Duan

Ministry of Education Key Laboratory of Science and Technology of Controllable Chemical Reactions P.O. Box 98 Beijing University of Chemical Technology Beijing 100029, P.R. China [email protected]

Ministry of Education Key Laboratory of Science and Technology of Controllable Chemical Reaction P.O. Box 98 Beijing University of Chemical Technology Beijing 100029, P.R. China [email protected]

Editorial Board Prof. Peter Day

Prof. Herbert W. Roesky

Director and Fullerian Professor of Chemistry The Royal Institution of Great Britain 21 Albermarle Street London W1X 4BS, UK [email protected]

Institut for Anorganic Chemistry University of Göttingen Tammannstr. 4 37077 Göttingen, Germany [email protected]

Prof. Jean-Pierre Sauvage Prof. Thomas J. Meyer Department of Chemistry Campus Box 3290 Venable and Kenan Laboratories The University of North Carolina and Chapel Hill Chapel Hill, NC 27599-3290, USA [email protected]

Faculté de Chimie Laboratoires de Chimie Organo-Minérale Université Louis Pasteur 4, rue Blaise Pascal 67070 Strasbourg Cedex, France [email protected]

Structure and Bonding Also Available Electronically

For all customers who have a standing order to Structure and Bonding, we offer the electronic version via SpringerLink free of charge. Please contact your librarian who can receive a password or free access to the full articles by registering at: springerlink.com If you do not have a subscription, you can still view the tables of contents of the volumes and the abstract of each article by going to the SpringerLink Homepage, clicking on “Browse by Online Libraries”, then “Chemical Sciences”, and finally choose Structure and Bonding. You will find information about the – – – –

Editorial Board Aims and Scope Instructions for Authors Sample Contribution

at springeronline.com using the search function.

Preface

Intercalation in layered solids is a long-established phenomenon. It has been suggested [1] that the first example, dating from over two thousand years ago, involved intercalation in kaolinite (an aluminosilicate clay) and explains the secret behind the production of fine Chinese porcelain. In modern times, many thousands of papers have been devoted to intercalation chemistry in clays, graphite and other materials. In this volume, various aspects of the chemistry of layered double hydroxides (LDHs) are reviewed. LDHs are a class of synthetic anionic clays whose structure is based on brucite (Mg(OH)2 )-like layers in which some of the divalent cations have been replaced by trivalent ions giving positively-charged sheets. This charge is balanced by intercalation of anions in the hydrated interlayer regions. The identities and ratios of the di- and trivalent cations and the interlayer anion may be varied over a wide range, giving rise to a large class of isostructural materials. The parent material of this class is the naturally occurring mineral hydrotalcite and LDHs are consequently also known as hydrotalcite-like materials. Although the basic features of the structure are well understood, detailed structural aspects have been the subject of some controversy in the literature. In the first chapter of this volume, Evans and Slade review the wide range of experimental and theoretical studies in this area, highlighting areas of consensus and currently unresolved issues. Although simple to prepare in the laboratory in principle, LDHs are not always easy to synthesize as pure phases. In the second chapter of this volume He et al. review methods of synthesis of LDHs, with an emphasis on the way in which the physicochemical properties of the materials vary with the synthesis method. In the following chapter, Taviot-Guého and Leroux review the synthesis and structural features of nanocomposite materials prepared by the assembly of LDHs and polymers or polymer precursors and discuss how the properties of the resulting hybrid materials are related to their structure. Although a great many intercalation compounds of LDHs have been prepared and characterized, there have been few studies of the kinetics and mechanism of the intercalation process. In their chapter, O’Hare and colleagues describe how detailed information about these aspects may be obtained by means of time-resolved in situ X-ray powder diffraction experiments.

X

Preface

The wide range of possible compositions of LDHs means that materials with a great variety of different properties can be produced. A detailed understanding of the structural chemistry of LDHs as well as the kinetics and mechanism of the formation process should facilitate the design of materials with properties precisely tailored for specific applications. In the final chapter of this volume, Li and Duan review current and potential applications of LDHs in this light. In recent years, there has been an explosive growth in publications concerning LDHs in both academic journals and the patent literature. These reviews have attempted to cover work published up to early 2005, but many examples of excellent science have no doubt been excluded, either inadvertently or because of space limitations. Beijing, October 2005

X. Duan, D. G. Evans

Reference 1. Weiss A (1963) Angew Chem 75:755; Angew Chem Int Ed Engl 2:697

Contents

Structural Aspects of Layered Double Hydroxides D. G. Evans · R. C. T. Slade . . . . . . . . . . . . . . . . . . . . . . . . .

1

Preparation of Layered Double Hydroxides J. He · M. Wei · B. Li · Y. Kang · D. G Evans · X. Duan . . . . . . . . . . .

89

In Situ Polymerization and Intercalation of Polymers in Layered Double Hydroxides C. Taviot-Gueho · F. Leroux . . . . . . . . . . . . . . . . . . . . . . . . 121 Mechanistic and Kinetic Studies of Guest Ion Intercalation into Layered Double Hydroxides Using Time-Resolved, In-situ X-ray Powder Diffraction G. R. Williams · A. I. Khan · D. O’Hare . . . . . . . . . . . . . . . . . . . 161 Applications of Layered Double Hydroxides F. Li · X. Duan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 Author Index Volumes 101–119 . . . . . . . . . . . . . . . . . . . . . . 225 Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233

Contents of Volume 118 Semiconductor Nanocrystals and Silicate Nanoparticles ISBN: 3-540-27805-2

Electrochemistry and Electrogenerated Chemiluminescence of Semiconductor Nanocrystals in Solutions and in Films A. J. Bard · Z. Ding · N. Myung Intraband Spectroscopy and Semiconductor Nanocrystals P. Guyot-Sionnest Controlled Synthesis of High Quality Semiconductor Nanocrystals X. Peng · J. Thessing The Zintl–Klemm Concept Applied to Cations in Oxides. II. The Structures of Silicates D. Santamaría-Pérez · A. Vegas · F. Liebau

Struct Bond (2006) 119: 1–87 DOI 10.1007/430_005 © Springer-Verlag Berlin Heidelberg 2005 Published online: 2 December 2005

Structural Aspects of Layered Double Hydroxides David G. Evans1 (u) · Robert C. T. Slade2 (u) 1 Ministry

of Education Key Laboratory of Science and Technology of Controllable Chemical Reactions, Beijing University of Chemical Technology, 100029 Beijing, P.R. China [email protected] 2 Department of Chemistry, University of Surrey, Guildford GU2 7XH, UK [email protected] 1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Basic Structural Features . . . . . . . . . . . . . . . . . Brucite Layers . . . . . . . . . . . . . . . . . . . . . . . Cation Substitution in Brucite-like Layers . . . . . . . . Possible Identity of MII /MIII Ions and Range of Values of the MII /MIII Ratio . . . . . . . . . . . . . . . . . . . 2.2.2 LDHs with MIV Ions? . . . . . . . . . . . . . . . . . . . 2.3 LDH Interlayers . . . . . . . . . . . . . . . . . . . . . . 2.4 Layer Rigidity . . . . . . . . . . . . . . . . . . . . . . . 2 2.1 2.2 2.2.1

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3 3 3 4

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

4 11 11 12

3 3.1 3.2 3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.2.6 3.2.7 3.3 3.3.1 3.3.2 3.4 3.5

Detailed Structural Characterization of LDHs . . . . . . Stacking Sequences in LDH Polytypes . . . . . . . . . . . Arrangement of Anions and Water in Interlayer Galleries Evidence from X-ray Diffraction . . . . . . . . . . . . . . Rietveld Refinement of Interlayer Structure . . . . . . . . Spectroscopic Measurements . . . . . . . . . . . . . . . . Molecular Modeling . . . . . . . . . . . . . . . . . . . . . Guest-guest Interactions . . . . . . . . . . . . . . . . . . Molecular Dynamics Simulations . . . . . . . . . . . . . Interlayer Galleries and Hydration . . . . . . . . . . . . . Long-range Cation Order-disorder . . . . . . . . . . . . . Experimental Studies . . . . . . . . . . . . . . . . . . . . Theoretical Studies . . . . . . . . . . . . . . . . . . . . . Short-range Cation Order . . . . . . . . . . . . . . . . . . Anion Ordering . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

12 12 24 24 27 30 37 46 48 53 58 58 64 66 70

4 4.1 4.2 4.3 4.4

Unusual Structural Features of Certain LDHs Non-octahedral Coordination of Layer Cations Staging of Interlayer Anions . . . . . . . . . . Double Layers of Anions or Cations/Anions . Absence of MIII Ions . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

76 76 76 78 78

5

Summary and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

79

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

80

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

2

D.G. Evans · R.C.T. Slade

Abstract Layered double hydroxides (LDHs) have been known for a considerable time and have been widely studied. The basic features of their structure, involving positively charged brucite-like layers together with charge-balancing anions and water in interlayer galleries are well understood, but some detailed aspects of their structure have been the subject of controversy in the literature. In this article we review the wide range of experimental and theoretical studies of the structure of LDHs, highlighting areas of consensus and currently unresolved issues. We focus on the range of composition for which LDHs may be formed, possible layer stacking polytypes, arrangement of guest species in the interlayer galleries and the extent of order-disorder phenomena, both long-range and short-range, in the layers and interlayer galleries. Keywords Layered double hydroxide · Hydrotalcite · Structure · Diffraction · Polytype · Order-disorder Abbreviations AEC Anion exchange capacity AFM Atomic force microscopy AQ15 Anthraquinone-1,5-disulfonate AQ2 Anthraquinone-2-sulfonate AQ26 Anthraquinone-2,6-disulfonate BPC p-Benzoylbenzoate CSC 4
-chloro-4-stilbenecarboxylic acid DSC Differential scanning calorimetry EDAX Energy dispersive analysis of X-rays EXAFS Extended X-ray absorption fine structure HH Head-to-head HT Head-to-tail ICP Inductively coupled plasma emission spectrophotometry IP Image plate IR Infrared JCPDS Joint committee on powder diffraction standards LB Langmuir-Blodgett LDH Layered double hydroxide MBSA 5-Benzoyl-4-hydroxy-2-methoxybenzenesulfonic acid NEXAFS Near edge X-ray absorption spectroscopy (synonym for XANES) NHA 4-Nitrohippuric acid NMR Nuclear magnetic resonance SAED Selected area electron diffraction STEM Scanning transmission electron microscopy STM Scanning tunneling microscopy UV Ultraviolet XAS X-ray absorption spectroscopy XANES X-ray absorption near-edge spectroscopy XRD X-ray diffraction

Structural Aspects of Layered Double Hydroxides

3

1 Introduction Layered double hydroxides (LDHs) [1–3] have been known for over 150 years since the discovery of the mineral hydrotalcite, and a large class of minerals with closely related structures are usually known to mineralogists as the sjögrenite-hydrotalcite group [4]. Although the stoichiometry of hydrotalcite, [Mg6 Al2 (OH)16 ]CO3 ·4H2 O, was first correctly determined by Manasse in 1915, it was not until pioneering single crystal X-ray diffraction (XRD) studies on mineral samples were carried out by Allmann [5] and Taylor [6, 7] in the 1960s that the main structural features of LDHs were understood. Nearly 40 years later, however, many of the fine details of the structure such as the range of possible compositions and stoichiometry, the extent of ordering of metal cations within the layers, the stacking arrangement of the layers and the arrangement of anions and water molecules in the interlayer galleries are not fully understood and have been the subject of some controversy in the literature. In this article we review the wide range of experimental and theoretical studies of the structure of LDHs, highlighting areas of consensus and currently unresolved issues. In addition to due reference to key papers from the earlier literature, recent work published up to early 2005 has been covered.

2 Basic Structural Features 2.1 Brucite Layers The basic layer structure of LDHs is based on that of brucite [Mg(OH)2 ] which is of the CdI2 type, typically associated with small polarizing cations and polarizable anions. It consists of magnesium ions surrounded approximately octahedrally by hydroxide ions. These octahedral units form infinite layers by edge-sharing, with the hydroxide ions sitting perpendicular to the plane of the layers as shown in Fig. 1 [8]. The layers then stack on top of one another to form the three-dimensional structure. From the point of view of close-packing, the structure can be said to be composed of close-packed planes of hydroxyl anions that lie on a triangular lattice. The metal cations occupy the octahedral holes between alternate pairs of OH planes and thus occupy a triangular lattice identical to that occupied by the OH ions. In actual fact, both the local geometry around the metal and the close-packing of the hydroxyl anions are strongly distorted away from the idealized arrangements. The octahedra are compressed along the stacking axis, so that the local geometry at the metal is D3d , rather than Oh . This has the effect of increasing the O· · · ·O and Mg· · · ·Mg distances parallel to the plane from 0.2973 nm (ideal

4

D.G. Evans · R.C.T. Slade

Fig. 1 Stereographic projection of brucite. Reprinted with permission from L. Eriksson, U. Palmqvist, Studsvik Neutron Research Laboratory Report 482. Copyright University of Stockholm

Oh geometry) to 0.3142 nm (experimental distance) and decreasing the thickness of the layers from 0.2427 to 0.2112 nm, with the O – M – O bond angles becoming 96.7◦ and 83.3◦ rather than a regular 90◦ [9]. This distortion has been discussed from a molecular orbital viewpoint [10] and an almost constant distortion has been suggested for all brucite-like hydroxides on the basis of the correlation between unit cell parameters and M – O distances [11]. The distortion of the brucite layers does not change the hexagonal symmetry (ao = bo = 0.3142 nm, co = 0.4766 nm, γ = 120◦ ) and the space group is P3m1. Note that the value of ao is equal to the nearest neighbor Mg· · · ·Mg distances parallel to the plane. The O – H bonds are directed along the threefold axes towards the vacant tetrahedral site in the adjacent layers. The nature of the weak forces between the layers has been the subject of considerable controversy in the literature with different importance being attached to contributions from dispersion forces and hydrogen bonding [8, 12–14]. 2.2 Cation Substitution in Brucite-like Layers 2.2.1 Possible Identity of MII /MIII Ions and Range of Values of the MII /MIII Ratio The basic structure of an LDH may be derived by substitution of a fraction of the divalent cations in a brucite lattice by trivalent cations such that the layers acquire a positive charge, which is balanced by intercalation of anions (and, usually, water) between the layers. It is the possibility of varying the identity and relative proportions of the di- and trivalent cations as well as the identity of the interlayer ions that gives rise to the large variety of materials having the general formula [MII 1–x MIII x (OH)2 ]x+ [An– ]x/n ·yH2 O, which

Structural Aspects of Layered Double Hydroxides

5

belong to the LDH family. Many ternary LDHs involving mixtures of different MII and/or MIII cations can also be prepared and even quaternary LDHs have been reported [15]. In fact the range of materials is even larger than suggested by this formula because materials containing monovalent lithium ions of the type [LiAl2 (OH)6 ]+ [An– ]1/n ·yH2 O are also known [16] and have similar structures. A number of empirical studies have attempted to define the boundaries that indicate which metal ions can form LDHs. For example, the roles of the absolute values of ionic radii and the difference in ionic radii of MII and MIII [3, 17–21] as well as the values of the solubility products of MII (OH)2 and MII CO3 [22] have been speculated on in the literature. In earlier publications [1, 23], it was often stated that ions such as Cu2+ that are subject to a Jahn-Teller distortion do not form LDHs unless diluted by other cations [24], but more recently examples of CuII -containing LDHs such as [Cu0.69 Cr0.31 (OH)2 ]Cl0.31 · 0.61H2 O [25, 26] have been well characterized (see Sect. 3.2.2), although the material does show a corrugation of the sheets associated with the Jahn-Teller distortion around individual CuII centers as discussed in Sect. 3.4. More recently a number of studies of the thermodynamic factors affecting the formation of LDHs have been reported. The enthalpies of formation from the elements of a series of Mg/Al – CO3 LDHs with different Mg/Al ratios have been measured by high-temperature oxide-melt solution calorimetry [27]. Compared with a mixture of Mg(OH)2 , Al(OH)3 , MgCO3 and water and it was found that the LDHs are more stable by 10–20 kJmol–1 , but there is no correlation with Mg/Al ratio. Similar results were obtained [28] from calorimetric measurements of the heats of formation of LDHs of the type [Co0.8 Al0.2 (OH)2 ](NO3 )0.2y (CO3 )0.1(1–y) ·(0.7 – 0.3y)H2 O and [(Niy Co1–y )0.8 Al0.2 (OH)2 ](CO3 )0.1 · 0.7H2 O. Based on measurements of solubility products of [M4 Al2 (OH)12 ]CO3 ·nH2 O, it has been concluded [29, 30] that the thermodynamic stability of the LDHs is greater than that of the corresponding divalent hydroxides below pH ≈ 10 (M = Zn), 9 (M = Co) and 8 (M = Ni); for M = Mg, the LDH is more stable up at least pH 12. It has been proposed [31, 32] that simple thermodynamic models can be used to determine the standard molar Gibbs free energy of formation of LDHs by, for example, using the tabulated values for the corresponding divalent and trivalent metal hydroxides and the salt of the divalent cation according to the following stoichiometric equation: (1 – 3x/2)MII (OH)2 + xMIII (OH)3 + (x/2)MII (An– )2/n = [MII 1–x MIII x (OH)2 ](An– )x/n ,

(1)

although water was apparently not included in these calculations. Standard Gibbs free energies of formation of several so-called green-rust type LDHs of the type [FeII 1–x FeIII x (OH)2 ]x+ (An– )x/n ·nH2 O have been tabulated [33]. It

6

D.G. Evans · R.C.T. Slade

has been suggested that equilibria between these species and related ternary species such as [FeII 1–x–y Mgy FeIII x (OH)2 ](OH)x ·nH2 O are responsible for controlling the distribution of iron in hydromorphic soils and relevant solubility products have been determined [34–36]. Braterman and coworkers [37, 38] have analyzed the stabilities of LDHs of the type [MII 1–x MIII x (OH)2 ]Clx · nH2 O relative to the corresponding metal hydroxides/hydrous oxides. The LDH stabilities were found to vary in the order Mg < Mn < Co ≈ Ni < Zn for MII and Al < Fe for MIII . On the basis of pH titration curves for the reaction of NaOH with the mixed chloride salt solutions, it was suggested that the hydroxide or hydrous oxide is generally formed first at lower pH and subsequently reacts with the aqueous MII ions at higher pH to form the LDH. The LDHs were shown to be generally thermodynamically unfavorable with respect to mixtures of metal hydroxides, but mass action effects associated with the high concentration of the anion in solution drive the system in the direction of the LDH. In contrast, for the case of MIII = Cr, the formation of [MII 1–x CrIII x (OH)2 ]Clx ·nH2 O occurs in a single step without formation of Cr(OH)3 . The solubility products of the LDHs vary in the order Zn < Ni ≈ Co. For MII = Mg, the corresponding LDH chloride could not be prepared although the sulfate analogue could be isolated via a two-step process. It is often said [1, 3, 19, 21] that pure LDH phases can only be formed for stoichiometries in the range 0.20 < x < 0.33, i.e. MII /MIII ratios in the range 2–4. It has been argued that for x > 0.33, the presence of MIII – O – MIII linkages would be required and these are unfavorable from the point of view of charge repulsion (the so-called cation avoidance rule) [36, 39, 40]. It has been demonstrated, however, that such MIII – O – MIII linkages can occur in LDHs, at least in the case of M = Cr. Reaction of a perchlorate solution of the dimer [(H2 O)4 Cr(µ – OH)2 Cr(OH2 )4 ]4+ with ZnO and subsequent anion exchange affords [41] grayish-blue [Zn7 Cr4 (OH)22 ]CO3 ·5H2 O, which has a ligand field splitting less than that of blue-green [Zn4 Cr2 (OH)12 ]CO3 · 3H2 O. The existence of the dimeric unit in the Zn7 Cr4 LDH was confirmed by UV-visible spectroscopy and ion-exchange chromatography after dissolution of the LDH in perchloric acid. In the dioctahedral gibbsite [γ -Al(OH)3 ] lattice only two-thirds of the octahedral sites are occupied by cations. A novel variant of LDHs having the formula [MII Al4 (OH)12 ](NO3 )2 · nH2 O (MII = Co, Ni, Cu, Zn) have been prepared [42] by reaction of activated gibbsite with the corresponding MII (NO3 )2 solutions. Half of the cation vacancies have been filled with MII cations, so that although the MII /Al ratio is 1 : 4, there are still cation vacancies unlike in the trioctahedral brucite-like layers of conventional LDHs where all the octahedral sites are occupied. In the case of monovalent Li+ cations, the vacancies in gibbsite can be completely occupied giving a series of LDHs containing [LiAl2 (OH)6 ]+ layers that have been widely studied and fully structurally characterized as noted above [16].

Structural Aspects of Layered Double Hydroxides

7

Theoretical calculations [43] based on first principles molecular dynamics discussed in Sect. 3.2.6 have suggested that Mgn Al LDHs are most stable for n = 3 (i.e. x = 0.25) and indeed many minerals, including hydrotalcite itself, have this stoichiometry [4]. It has been reported that the synthesis of LDHs (with benzoate or terephthalate anions in the interlayers) from solutions containing Mg/Al = 2, leads to LDHs having the same composition when the synthesis is carried out at moderate temperatures but LDHs with Mg/Al = 3 (plus AlOOH) when the reaction is carried out under hydrothermal conditions [44]. It was proposed that the latter ratio represents the thermodynamically most favorable product. A similar observation has been reported [45] for solutions with NiII /FeIII = 2, where hydrothermal preparation led to segregation of an LDH with NiII /FeIII = 3 and NiII FeIII 2 O4 . An attempt to synthesize a CoII 5 Al LDH resulted in partial oxidation of the CoII and formation of a CoII 0.7 CoIII 0.3 LDH with complete migration of Al3+ from the layers to generate interlayer aluminum oxy-species [46]. There have been many claims that LDHs can be formed with stoichiometries outside the range 0.20 < x < 0.33 [17, 21], with reported values of x down to as low as 0.07 for Mg/Ga – CO3 LDHs [47] and as high as 0.41–0.48 for MII /MIII – CO3 LDHs (MII = Mg, Ni, Co, Cu; MIII = Al, V) [48–52] and 0.5 for FeII /FeIII LDHs [53]. It has been suggested [54] that ion-exchange reactions at relatively low pH favor leaching of MII ions from the layers and thus gives rise to high values of x. There are, however, many difficulties in determining the exact value of x in LDHs. Elemental analysis of the metal content of a solid phase will give erroneous values if the LDH is mixed with MII (OH)2 , MII (OH)3 /MIII OOH or other phases that might be expected to segregate [36] when the synthesis mixture contains either very high or very low MII /MIII ratios. Whilst such phases can sometimes be observed by XRD [44, 55–58], more often than not they are likely to be amorphous and remain undetected by XRD [59]. It has been noted that preferred orientation of LDH samples tends to make it difficult to observe weak reflections from small amounts of such impurities [60]. Other methods sometimes provide evidence of impurity phases. Small quantities of Co3 O4 admixed with Mg/CoII /CoIII LDHs have been identified by FTIR [61]. The presence of significant quantities of amorphous iron(III) oxides such as ferrihydrite Fe5 HO8 · 4H2 O mixed with Mg/FeIII LDHs has been detected [62, 63] by Mössbauer spectroscopy at 4.2 K, since ferrihydrite and related species are magnetically ordered at this temperature and give a sextet in the Mössbauer spectrum whereas the FeIII ions in the LDH are not and give a doublet. It has also been suggested, however, that the presence of FeIII – O – FeIII moieties in the LDH itself can be inferred from Mössbauer spectroscopy and it has been speculated [64] that the acidic nature of the MgFe2OH units favors deprotonation so that the anion exchange capacity of the materials remains constant when x > 0.33, i.e. the structure can be written as [Mg1–x Fex (OH)2–y Oy ](CO3 )(x–y)/2 ·nH2 O.

8

D.G. Evans · R.C.T. Slade

It has been proposed that, if no change in Mg/Al ratio is observed after washing an Mg/Al – CO3 LDH with hot NaOH, this can be taken to indicate the absence of amorphous aluminum oxy-species [51]. It has also been suggested [65] that consistent values of the metal content of the bulk phase (e.g. as determined by ICP) and that of the surface (as determined by EDAX) is good evidence for the formation of a single LDH phase. It has been proposed that use of a high resolution scanning transmission electron microscope (STEM) coupled with EDAX allows the composition of LDH platelets to be determined in the presence of non-LDH impurities, which have particles with clearly different morphology [49]. Since carbonate-containing LDHs are characterized by the formula [MII 1–x MIII x (OH)2 ](CO3 )x/2 ·yH2 O, it is sometimes suggested that analysis of the carbon content gives a more reliable guide to the value of x than metal analysis [57]. Other studies have, however, indicated [66, 67] that LDHs can adsorb significant amounts of CO2 or carbonate anions on external non-gallery surfaces, affecting the carbon content and so suggesting that this may not be a reliable indication of the value of x. On balance, it seems, therefore, that the range of stoichiometries for which LDH phases can be prepared is relatively narrow and that many of the materials reported in the literature with unusually high or low values of x are probably not pure single phases. Structural studies of LDHs discussed in Sects. 3.3 and 3.4 indicate that the metal-oxygen octahedra in LDHs are compressed along the c axis exactly as for brucite with the O – M – O bond angles being distorted considerably (typically by 7–8◦ ) from an ideal octahedral arrangement. It has been suggested that the observed broadening of the d–d transitions of CrIII ions in LDHs is associated with this distortion from Oh to D3d symmetry [41, 68]. As noted in Sect. 2.1, the value of the unit cell parameter ao of brucite (0.3142 nm) is equivalent to the mean distance between adjacent cation centers in the closepacked sheets and the value of the corresponding parameter for LDHs can be correlated with the average radii of the metal cations in the layers. To a first approximation, the value of ao may be calculated [56, 69] assuming Vegard’s √ law and an ideal atomic arrangement for the cations; in that case ao = 2d(M – O), where the metal-oxygen bond length d(M – O) is related to the ionic radii by the equation d(M-O) = (1 – x)r(MII ) + xr(MIII ) . From this it follows that √ δao /δx = – 2[r(MII ) – r(MIII )] . The values of ao for Mg1–x Alx LDHs (≈ 0.302–0.307 nm) are smaller than the value for brucite and decrease with x, since the Shannon crystal radius of Al3+ is smaller than that of Mg2+ (0.0675 vs. 0.0860 nm) [70]. Some typical data are shown in Fig. 2 for Mg/Al – CO3 and Zn/Al – CO3 LDHs [19] (the Shannon crystal radius of Zn2+ is 0.088 nm) and there are

Structural Aspects of Layered Double Hydroxides

9

many other examples in the literature for a wide variety of LDHs [20, 39, 46, 55, 71–77]. Ab initio molecular orbital calculations on dimers of the type [(H2 O)2 (OH)2 Mg(µ-OH)2 Mg(OH)2 (OH2 )]2– and [(H2 O)2 (OH)2 Mg(µOH)2 Al(OH)2 (OH2 )]– have reproduced the experimental values of the M· · · ·M distances in brucite and LDHs [78]. Where there is no correlation between the value of ao and the apparent composition of the LDH, this may indicate that other non-LDH phases are also present [58]. It has often been observed in the case of Mg/Al LDHs with interlayer carbonate [51] or nitrate [55] anions that the value of ao remains essentially constant (≈ 0.304 nm) for values of x ≥ 0.33. In some cases the presence of an impurity such as gibbsite has been observed by XRD [55], sug-

Fig. 2 Variation of the hexagonal unit cell parameters of Mg/Al – CO3 and Zn/Al – CO3 LDHs with the MII /MIII ratio used in their preparation. Reprinted with permission from [19]. Copyright Cambridge University Press

10

D.G. Evans · R.C.T. Slade

gesting that the LDH phase has x ≈ 0.33 with the excess Al3+ incorporated in the impurity. It has been argued, however, [51] that a value of (≈ 0.304 nm) represents the closest possible approach of Mg2+ and Al3+ cations and pure LDH phases with x > 0.33 also have ao ≈ 0.304 nm. With [Fe(CN)6 ]3– as the guest anion in Mg/Al LDHs, a linear variation in the value of ao against x in the range 0.10–0.48 has been reported, however, [76] with a value of x = 0.48 being associated with ao = 0.302 nm. A correlation of the value of ao with values of x up to 0.44 has also been reported for Zn/Al – CO3 LDHs [49]. Although in principle, the value of ao provides a useful way of estimating the value of x (at least up to x ≈ 0.33), as discussed in Sect. 3.1, the experimental value of ao is determined from the position of the (110) reflection in the powder XRD pattern. This reflection is often weak and may be broadened and overlap with the adjacent (113) reflection so that its position may be difficult to determine sufficiently accurately to give a meaningful measure of x. In some cases, deconvolution of the overlapping (110) and (113) reflections has been reported [55, 79] in an attempt to overcome this problem. The basal spacing of an LDH (the distance from the center of one layer to that in the adjacent layer) is obviously much greater than that in brucite (co = 0.4766 nm) because of the absence of any interlayer anions and water in the latter. The LDH basal spacing shows some correlation with composition of the layers, although clearly the size of the anion and the extent of hydration will also have a major influence [1, 17, 19]. For Mgn Al – CO3 LDHs it has been reported that the basal spacing decreases from 0.7928 nm for n = 5 to 0.7591 nm for n = 24. [N.B. As discussed below, the crystallinity of LDH materials reported in the literature varies over a very wide range. Structural data are therefore given to varying numbers of significant figures in the literature, (hopefully!) taking into account the degree of crystallinity of the sample; the quoted values are reproduced in this review without any attempt at harmonization]. There is a small contraction in the layer thickness from 0.2001 nm (n = 5) to 0.1959 nm (n = 2) (as compared with 0.2101 nm in brucite [8, 13]), which parallels the decrease in ao . Almost all of the change in basal spacing, however, can be attributed to the decrease in the thickness of the interlayer galleries [39]. The charge density of the layers increases with decreasing n leading to stronger electrostatic bonding with interlayer anions and stronger hydrogen bonding with interlayer water molecules. In the case of [Ni6–x Mgx Fe2 (OH)16 ]CO3 ·4H2 O, the basal spacing was reported to show a monotonic increase from 0.778 nm (x = 0) to 0.791 (x = 6) although the overall cationic charge on the layers does not change [80]. This was interpreted as indicating that OH groups bonded to Ni2+ are more strongly polarized than those bonded to Mg2+ and hence form stronger hydrogen bonds with interlayer carbonate anions. A similar increase in basal spacing in LDHs of the type [Co2 Fey Al1–y (OH)2 ]Cl·nH2 O from 0.7637 nm (y = 0) to 0.7893 nm (y = 1) was ascribed to a variation in the thickness of the layers, however [72].

Structural Aspects of Layered Double Hydroxides

11

2.2.2 LDHs with MIV Ions? Although it has long been established that many combinations and permutations of divalent and trivalent cations can form LDHs and that one monovalent cation (Li+ ) is also able to form LDHs based on [LiAl2 (OH)6 ]+ layers, the question of possible incorporation of tetravalent ions into the LDH layers has been more controversial. A number of recent papers have reported the possibility of synthesizing LDHs containing MIV ions including Mg/Al/ZrIV – CO3 [81–84], Mg/Al/SnIV – CO3 [85], MII /Al/SnIV – CO3 (MII = Co, Ni) [86], MII /Al/SnIV – CO3 (MII = Mg, Ni, Zn) [87], Mg/Al/TiIV – CO3 [88], Mg/Al/SiIV – CO3 [89] and even a Zn/TiIV – CO3 LDH containing no trivalent cations [90]. Recently, however, considerable doubt has been cast on these conclusions. In the case of the putative Mg/Al/SnIV – CO3 , Mg/Al/ZrIV – CO3 and Co/Al/SnIV – CO3 materials it has been unambiguously shown by X-ray absorption spectroscopy (XAS) and Mössbauer spectroscopy that the tetravalent cations are segregated from the LDH structure and form amorphous MIV oxide-like particles [69]. It was further demonstrated that the increased values of ao previously attributed to the introduction of the large MIV cations could equally well be explained by the changes in Mg/Al ratio in an LDH phase with no incorporation of MIV . A similar analysis of a material described as a Co/Al/TiIV – CO3 LDH has also clearly shown that the TiIV cations are not incorporated in the LDH sheets [91]. These results highlight the dangers of postulating the composition of an LDH based solely on analytical data and the value of unit cell parameters, as discussed in Sect. 2.2.1 for MII /MIII LDHs. 2.3 LDH Interlayers The interlayer galleries of LDHs contain both interlayer anions and water molecules and there is a complex network of hydrogen bonds between layer hydroxyl groups, anions and water molecules. The interlayers are substantially disordered and hydrogen bonds are in a continuous state of flux so that the precise nature of the interlayer is extremely complex [3]. The bonding between octahedral layers and interlayers involves a combination of electrostatic effects and hydrogen bonding. Hydroxyl groups, particularly those bonded to trivalent cations are strongly polarized and interact with the interlayer anions. Every anion has to satisfy excess positive charges on both of the sandwiching octahedral layers, which are electrically balanced by two neighboring interlayers; it has been suggested that charge compensation in LDHs has many of the characteristics of resonance effects [92]. The structure of the interlayer galleries is discussed in detail in Sect. 3.2.

12

D.G. Evans · R.C.T. Slade

2.4 Layer Rigidity The rigidity of the layers in a two-dimensional solid has a significant effect on its properties [93]. An interlayer rigidity parameter p ≈ 5 was determined for LDHs by fitting the basal spacing of [Zn0.61 Al0.39 (OH)2 ](CO3 )0.195(1–y) Cl0.39y ·n H2 O as a function of composition y to the standard version of the discrete finite layer rigidity model [94, 95]. This value indicates LDH layers are much more rigid than those in graphite (p ≈ 2) and much less rigid than those of 2 : 1 phyllosilicate clays such as vermiculite (p ≈ 7) and are more comparable to, although somewhat more rigid than, metal dichalcogenides (p ≈ 3.5) [93]. These values are consistent with the number of planes of atoms from which the layers are composed (one for graphite, three for LDHs and metal dichalcogenides and seven for 2 : 1 clays). As a result of the relatively rigid layers, strong forces are exterted on the interlayer guests in LDHs. From the magnitude of the red shift in the photoluminescence observed on intercalation of [SmW10 O36 ]9– in LDHs, it has been estimated that the layers exert a uniaxial stress of about 14 GPa on the guest anion [96, 97].

3 Detailed Structural Characterization of LDHs 3.1 Stacking Sequences in LDH Polytypes The brucite-like layers in LDHs may be stacked in different ways, which gives rise to a variety of possible polytype structures. All sites in the (110) plane of the close packed hydroxide layers may be represented as A, B or C related by lattice translations of (1/3, 2/3, 0) or (2/3, 1/3, 0), and the location of octahedral holes occupied by metal cations can be described analogously as a, b or c. Thus a single brucite layer can be represented as AbC (since, if close packed hydroxyls occupy A and C sites, the cations must, of necessity, occupy b sites the abbreviation is sometimes simplified to AC, but we retain the full description here). AbC layers may be stacked in various ways giving rise to a large number of possible polytypes. At the most basic level, these polytypes may be classified in terms of the number of sheets stacked along the c axis of the unit cell, but such an approach overlooks some interesting detailed structural features associated with the stacking of the sheets. If the opposing OH groups of adjacent layers lie vertically above one another (say both in C sites), a trigonal prismatic arrangement (denoted by =) results; if the hydroxyls are offset (say one layer in C sites and those of an adjacent layer in either A or B sites) then the six OH groups form an octahedral arrangement denoted by ∼. Thus

Structural Aspects of Layered Double Hydroxides

13

brucite itself can be denoted as ...AbC∼AbC... or 1H, where the “1” denotes a one-layer polytype and the “H” denotes a stacking sequence with hexagonal symmetry. Bookin and Drits [98–100] have systematically derived all of the possible polytypes for other stacking sequences. There are three possible twolayer polytypes, each of which has hexagonal stacking of the layers, which can be denoted as follows: ...AbC=CbA=AbC... ...AbC∼AcB∼AbC... ...AbC∼BcA=AbC...

2H1 2H2 2H3

Note that in the case of the 2H1 polytype, the cations all occupy b positions, i.e. are aligned on a normal to the layers (this is also the case for the 1H polytype), whereas for the other two polytypes the cations alternate between b and c sites. Note also that the interlayers in the 2H1 polytype are all prismatic and those in the 2H2 polytype are all octahedral, whilst in the 2H3 polytype both types of interlayers are present. There are nine possible three-layer polytypes. Two of these have rhombohedral symmetry (3R): ...AbC=CaB=BcA=AbC... ...AbC∼BcA∼CaB∼AbC...

3R1 3R2

whilst the remaining seven have hexagonal symmetry: ...AbC∼AcB∼AcB∼AbC... ...AbC∼AcB∼CaB∼AbC... ...AbC∼AcB=BcA∼AbC... ...AbC∼AbC=CbA=AbC... ...AbC∼AcB=BaC∼AbC... ...AbC∼AcB∼CbA=AbC... ...AbC∼AbA∼BcA=AbC...

3H1 3H2 3H3 3H4 3H5 3H6 3H7

Note that the cations are homogenously distributed over a, b and c sites in the case of the 3R1 , 3R2 and 3H2 polytypes. For the 3R1 polytype, the interlayers are all prismatic and in the case of 3R2 , 3H1 and 3H2 they are all octahedral; other polytypes involve both types of interlayers. Bookin and Drits [98–100] have also described the large number of possible six-layer polytypes, some of which have rhombohedral symmetry (6R) and the remainder hexagonal symmetry (6H). The identity of the polytype present in a given LDH sample may, in principle at least, be determined from the powder XRD pattern, although as we shall see for many LDHs this is not possible, as the amount of useful information therein is limited. By convention, the indexing of powder patterns for rhombohedral polytypes is based on a triple hexagonal unit cell (see Fig. 3).

14

D.G. Evans · R.C.T. Slade

Fig. 3 Relationship between the triple hexagonal cell with axes OE, OF, OG to the primitive rhombohedral cell with axes OA, OB, OC

Although such a cell is not primitive, it is easier to visualize and facilitates direct comparison with hexagonal structures. The reflections in the XRD pattern of an LDH fall into three groups: (1) A series of strong basal (00l) reflections at low angles allow direct determination of the basal spacing normal to the (00l) plane (co ), which equals the thickness of one brucite-like layer plus one interlayer. For an n-layer polytype, the unit cell parameter c = nco , and the lowest angle reflection is indexed as (00n) although its spacing always corresponds to co . Higher order (00l) reflections generally have spacings corresponding to co /2, co /3, ... However, this relationship will break down if there is imperfect stacking along the c axis. This can arise if there is interstratification (interlayers have varying anions and/or water composition) [101, 102] or varying extents of disorder in the interlayers [100, 103]. In addition, if the LDH is composed of very thin platelets, the small coherent scattering domains in the c direction lead to the basal reflection being shifted to a slightly lower angle, giving an apparent spacing that is greater than the true value of co calculated from the higher order basal reflections [100]. Where basal and non-basal reflections overlap, useful information can sometimes be obtained by comparing the intensities of XRD patterns of preferentially ordered samples with those of randomly oriented samples. The former may be prepared by drying suspensions of finely ground LDHs in water on glass slides [104] or by thorough pressing and sliding of the finely ground sample [105]. The latter may be obtained by means of the McMurdie [106] or Petrov [105] methods. The intensities of reflections from planes oriented approximately parallel to the sample holder (i.e. (00l) and other planes closely inclined to the basal plane) are enhanced relative to those of other planes for preferentially ordered samples. If the two XRD patterns are subtracted from one another (after preliminary normalization of the inten-

Structural Aspects of Layered Double Hydroxides

15

sity of a reflection from a plane inclined at approximately 45◦ to the basal plane), overlapping basal and non-basal reflections can be resolved [105]. For example, using this method Stanimirova [105] has concluded that a hydrotalcite mineral sample from Snarum in Norway contains a mixture of three polytypes. (2) The position of the (110) reflection at high angle (near 60◦ 2θ for Cu Kα radiation) allows the value of the lattice parameter ao to be determined, since ao = 2d(110). The value of ao corresponds to the distance between two metal cations and its value should therefore reflect the radii of the cations, as discussed in Sect. 2.2.1. (3) Finally, the positions of the (01l) and/or (10l) reflections at intermediate angles can be used to determine the stacking pattern of the layers. Using the values of ao and co determined as above, the number of layers stacked along the c axis of the unit cell can be determined by calculating the values of d(hkl) for each possible value of c (co , 2co , 3co ...) using the relationship 1/d2 (hkl) = 4(h2 + hk + k2 )/3a2o + l2 /c2 . The true c value gives coincident calculated and experimental values for the positions of the (01l) and/or (10l) reflections within experimental error. Furthermore, rhombohedral and hexagonal structures can be distinguished by systematic absences: for rhombohedral structures, reflections are systematically absent unless – h + k + l = 3n, where n is an integer and h, k and l are the (hexagonal) Miller indices, whilst the presence of strong reflections with – h + k + l  = 3n indicates hexagonal symmetry [44]. The diffraction pattern in Fig. 4a is that of a synthetic hydrotalcite [Mg6 Al2 (OH)16 ](CO3 )·4H2 O [107]. The observed reflections can be indexed in a three-layer 3R polytype with rhombohedral symmetry (space group R3m) with unit cell values of ao = 0.306 nm and c = 2.34 nm. In this case, the basal repeat distance, co , is c/3. The pattern in Fig. 4b [107] is that of the mineral manasseite (JCPDS 14–525) which has the same formula as hydrotalcite and the reflections can be indexed in a two layer 2H polytype with hexagonal symmetry (space group P63 /mmc) with unit cell values of ao = 0.306 nm and c = 1.56 nm. The basal spacing co corresponds to c/2 and is numerically identical to that in hydrotalcite. In order to distinguish between different polytypes with closely related stacking sequences it is essential to analyze the intensities of the non-basal reflections. Bookin and Drits [98–100] have simulated the diffraction patterns of many different polytypes. Although only qualitative results were obtained because the calculations neglect scattering of anions and water molecules in the interlayers, a set of criteria were derived that can be used to distinguish between different polytypes. For example, for the 3R1 polytype, intensities of (01(l+1)) reflections are stronger than (10l) (l = 3n + 1, n = integer) whereas the reverse is true for the 3R2 polytype. The presence of (012), (015) and (018) reflections in Fig. 4a indicates that hydrotalcite has the 3R1 structure

16

D.G. Evans · R.C.T. Slade

Fig. 4 Diffraction patterns of a a synthetic hydrotalcite [Mg6 Al2 (OH)16 ](CO3 )·4H2 O which has a 3R polytype and b the mineral manasseite which has the same formula as hydrotalcite but a 2H polytype. Reprinted with permission from [107]. Copyright Kluwer Academic Publishers

and it is generally believed that most carbonate-containing LDHs have this structure [7, 100] since, the prismatic arrangement of hydroxyl groups allows those in both upper and lower layers to form hydrogen bonds with the oxygen atoms of the carbonate anions (see Sect. 3.2). Jones et al. [108] have reported the synthesis of the 3R2 polytype of an LDH by hydrothermal treatment of a mixture of a flash calcined gibbsite and magnesium oxide, characterized by strong reflections that can be indexed as (101), (104) and (107) (Fig. 5). It was suggested that the majority of the interlayer anions are hydroxide, rather than carbonate, consistent with the observation that carbonate ions generally occupy prismatic interlayers. In this case there would be a cubic close packing sequence formed by hydroxyl groups of neighboring layers and the interlayer hydroxide ions. Based on an analysis of the intensities of the (01l) reflections, Moggridge et al. [44] have shown that Mg/Al LDHs containing interlayer benzoate or terephthalate anions both have the 3R1 structure. Distinguishing between different polytypes with hexagonal symmetry is more difficult, since the intensities of reflections are sensitive to the identity of the interlayer anion. Simulated XRD patterns suggest that manas-

Structural Aspects of Layered Double Hydroxides

17

Fig. 5 Diffraction pattern of a 3R2 polytype of an Mg/Al – CO3 LDH (peaks ∗ 1 and ∗ 2 are probably due to brucite). Reprinted with permission from [108]. Copyright Royal Society of Chemistry

seite and sjögrenite (one polytype of [Mg6 Fe2 (OH)16 ]CO3 ·4.5H2 O) have the 2H1 structure, which was confirmed by single crystal X-ray diffraction on the latter mineral [7]. Whereas carbonate favors prismatic interlayers (as found in 3R1 and 2H1 polytypes), hydrotalcite-like minerals containing interlayer sulfate ions have been found [99] to exist in polytypes with both prismatic (3R1 ) and octahedral (3R2 , 3H2 ) interlayers with the latter being more common. In mineral samples, different polytypes are often found intergrown. For example, hydrotalcite and manasseite often occur together with the latter on the inside and the former on the outside of the grains [71]. This has been interpreted as indicating that the latter is favored by high temperature since the inside of the grains cools more slowly than the outside. Pons et al. [71] have suggested that the 3R1 polytype can be constructed from a stack of identical brucite AbC layers by translations of (2a/3 + b/3) between consecutive layers whilst formation of the 2H1 polytype requires a translation of (2a/3 + b/3) followed by a 60◦ rotation. Thus formation of the latter requires higher temperatures. An alternative explanation was suggested by Hofmeister and von Platen [92], who noted that although the interlayers are prismatic in each case so that the layer-interlayer interactions are the same, they differ in that the cations in adjacent layers are directly superimposed on lattice rows parallel to the c axis in the case of 2H1 but are shifted by (2a/3, b/3) between adjacent layers in the case of the 3R1 polytype. This indicates that long-range interactions between cations in adjacent layers may influence the stacking of the layers. Thus hexagonal stacking of the layers to give the 2H1 polytype with eclipsed cations requires higher energy than rhombohedral stacking to give the 3R1 polytype where the cations are stag-

18

D.G. Evans · R.C.T. Slade

gered. Among the large number of synthetic LDH carbonates, no examples of the 2H1 polytype have been directly prepared [100]. Interestingly, however, it has also recently been claimed [109] that under pressure a synthetic hydrotalcite transforms irreversibly to a manasseite-like phase at about 1.5 GPa. Further compression leads to formation of an amorphous phase at pressures of 4.0–4.5 GPa. Thus the occurrence of manasseite inside hydrotalcite in mineral grains may reflect the pressure exerted on the core of the grains as the surface cools. In some cases, the stacking sequence in synthetic samples changes as a function of temperature. For example [95], the XRD pattern of a Zn/Al – Cl LDH shows that it exists as the 3R1 polytype at ambient temperature, whilst heating at 150 ◦ C results in partial dehydration accompanied by a transformation to the 2H1 polytype. A NiII /CoIII – CO3 LDH having the 3R1 structure was converted to a 1H polytype with a reduced basal spacing on heating at 200 ◦ C; the 3R1 polytype was regenerated by rehydration [110]. It was suggested that the carbonate anions are monografted to the slabs in the 1H polytype. Pausch et al. [51] reported the synthesis of LDHs with interlayer hydroxide anions. At lower reaction temperatures, the 3R1 polytype was formed. At higher temperatures (e.g. 150 ◦ C, 100 MPa for Mg3 Al) a second polytype was formed, for which a disordered layer stacking was proposed. Bookin and Drits [100] subsequently suggested, however, that the XRD pattern of the latter phase could be indexed as the 3R2 polytype with a cubic close packing of layer hydroxyl groups and interlayer hydroxide ions. As noted above, the same polytype was observed for an Mg/Al LDH with interlayers containing predominantly hydroxyl anions [108]. In the case of the Friedel’s salt halide series [Ca2 Al(OH)6 ]+ [X]– ·2H2 O, the materials exist as rhombohedral polytypes [6R (X = Cl), both 6R and 3R (X = Br) and 3R (X = I)] at high temperature and undergo a phase transition to a monoclinic polytype on cooling [111, 112]. A sample of the mineral stichtite with the formula [(Mg5.94 Ca0.01 )5.95 (Cr1.29 Al0.51 FeIII 0.25 )2.05 (OH)15.1 ](CO3 )1.47 ·3.7H2 O exists as pink grains that have the 3R1 structure with a basal spacing co of 0.776 nm [113]. On irradiation with visible light or X-rays, a green species is formed which has the 1H polytype stacking with a reduced basal spacing of 0.737 nm. In the case of Li/Al LDHs, it has been shown that [LiAl2 (OH)6 ]Cl·nH2 O can exist in both rhombohedral (when n = 2) [114] and hexagonal (when n = 1) [16] polytypes, although in this case the materials are prepared from different Al(OH)3 precursors rather than being interconverted by heating. Interestingly, the two polymorphs show some differences in their intercalation behavior [114]. The related compound [LiAl2 (OH)6 ]OH·nH2 O has a random stacking of layers that can be modeled with a 54 layer structure and has pseudohexagonal symmetry [115]. Unfortunately, for many of the synthetic LDHs that have been reported in the literature the XRD patterns are less informative than those in Figs. 4

Structural Aspects of Layered Double Hydroxides

19

and 5. Methods of synthesis that give highly crystalline samples have been reviewed by He et al. in Chapter 2 of this volume, and in these cases materials often do have diffraction patterns that allow the polytype to be determined unambiguously. Experimental and simulated diffraction patterns for a CoII 2.19 FeIII – CO3 LDH produced [116] by aerial oxidation of mixed CoII /FeII solutions are shown in Fig. 6 and indicate a 3R1 polytype with ao = 0.312 nm and c = 2.278 nm. When an Mg2.01 MnIII – CO3 LDH is formed [117] by aerial oxidation of Mg(NO3 )2 /MnCO3 mixtures at constant pH a highly crystalline material with a 3R1 polytype is formed as shown by the diffraction pattern in Fig. 7a, but when an analogous material is prepared by traditional coprecipitation methods a much less crystalline material is obtained (Fig. 7b). Many synthetic LDHs give diffraction patterns even less well defined than that in Fig. 7b (typical examples [118] are shown in Fig. 8) and in such cases it may only be possible to determine the basal spacing from the positions of the (00l) reflections. The lines can then be indexed as a one-layer polytype since there is no need or justification for using a larger unit cell [118]. In fact, however, in the literature many similar patterns are still indexed using the 3R stacking sequence. If the (110) reflection is resolved, then the unit cell parameter ao can be calculated, but if this reflection is broadened and overlaps with other reflections at higher angle, even this may not be possible with any degree of certainty as discussed in Sect. 2.2.1.

Fig. 6 Experimental A and simulated B diffraction patterns for a CoII /FeIII – CO3 LDH produced by aerial oxidation of mixed CoII /FeII solutions which indicate a 3R1 polytype. Reprinted with permission from [116]. Copyright Academic Press

20

D.G. Evans · R.C.T. Slade

Fig. 7 XRD patterns of a an Mg/MnIII – CO3 LDH formed by aerial oxidation of Mg(NO3 )2 /MnCO3 mixtures at constant pH and b an analogous material prepared by traditional coprecipitation methods. Reprinted with permission from [117]. Copyright The Mineralogical Society

There are a number of reasons why the XRD peaks of LDH samples are often rather broad. The relatively small domain size, particularly in the (00l) direction, leads to line broadening. The Scherrer equation may be used to estimate the domain size in the a and c directions from the width of the (110) and (00l) reflections, respectively [119, 120], although the inherent approximations in this method should always be borne in mind [121, 122]. If the broadening of peaks is non-uniform and characteristic of certain families of planes, or is so excessive as to give unrealistically small crystallite sizes, factors other than crystallite size must also be considered [123]. Various kinds of disorder may be incorporated during the crystallization process

Structural Aspects of Layered Double Hydroxides

21

Fig. 8 Examples of diffraction patterns of poorly crystalline LDHs that can best be indexed as one layer polytypes. Reprinted with permission from [118]. Copyright American Chemical Society

leading to a loss of “crystallinity” (see [124] for recommendations regarding use of the term “crystallinity”). Several authors have investigated the different types of disorder present in divalent metal hydroxides having the brucite structure [125–128]. The XRD patterns may be simulated with the DIFFaX program [39], which utilizes a statistical recursive approach in order to model extended planar faults. The various types of slabs present in the material are constructed from the ideal atomic positions and the unit-cell parameters. These slabs are then packed along the c direction using the stacking vectors (1/3, 2/3, 1) and (2/3, 1/3, 1) along with (0, 0, 1). The use of the former two stacking vectors at low (< 0.15) probabilities simulates the effect of stacking faults by introduction of cubic close packed ABC motifs in the normal hexagonal close packed ABAB packing and leads to selective broadening of the (h0l) (101 and 102) reflections [103, 126, 128]. The use of all three stacking vectors with equal (0.3333) or almost equal probabilities affords a more random packing of layers, corresponding to turbostratic disorder. In these cases, in addition to an increased broadening of the (h0l) reflections, the (hk0) (100 and 110) reflections become essentially two-dimensional hk in nature and acquire a “saw tooth” or “shark’s fin” line shape, rising sharply and displaying an asymmetry on the high angle side [20, 128]. Similar simulations can be carried out for LDHs and also show that if the positions of cations in adjacent layers are not correlated along the c axis,

22

D.G. Evans · R.C.T. Slade

i.e. there are stacking faults, then significant diminution in intensities of certain reflections are expected. For example, for the case of the 3R1 polytype, if a small number of such faults are present, the position, intensities and narrow linewidths of the (11l) reflections are unchanged whereas the intensity of the set of (10l) reflections is reduced and the lines are broadened, and they are shifted to lower angle than that expected on the basis of the cell defined by the (11l) reflections [100, 103]. If the number of faults is so large that the cations are completely uncorrelated along the c direction, the (10l) reflections have the “shark’s fin” peak shape associated with turbostratic structures [20, 100, 129, 130]. In synthetic samples, stacking faults often arise from an intergrowth of rhombohedral and hexagonal polytypes that gives rise to broad asymmetric (01l)/(10l) reflections in the XRD patterns as shown in Fig. 7b. Rebours et al. [39] have proposed that the occurrence of faults can be quantified in terms of a “fault probability” (FP), where FP = 1.0 describes a pure rhombohedral polytype 3R1 and FP = 0.0 a pure hexagonal polytype 2H1 , whilst a completely random stacking has FP = 0.5. They estimated that FP∼ = 0.6 for a number of Mg/Ga and Ni/Al LDHs whilst an Mg1.8 Al LDH had an essentially pure rhombohedral structure, as shown by the sharpness and symmetry of the (01l) reflections. This was rationalized by noting that, as described above, the arrangement of cations is different for the two polytypes and that although the local symmetry of the interlayer is the same for both polymorphs, charge compensating anions will not interact with the layers in the same way [92]. In the case of the 2H1 polytype, the compensating forces are perpendicular to the layers, but in the case of the 3R1 polytype they are directed more diagonally through the interlayer to the next octahedral layer. When the cations have a high average charge density (e.g. Mg2 Al), leading to extensive polarization of the hydroxyl groups, the staggered arrangement of cations in the 3R1 polytype may be favored over the eclipsed arrangement in the 2H1 polytype. When the cations have lower average charge density (e.g. Mg3 Al or Mg2 Ga), the difference between the two polytypes may be less marked, leading to a greater prevalence of stacking faults. Kamath et al. [103], however, have suggested that the XRD pattern of another Mg/Al LDH with very similar composition can best be simulated using the same approach by assuming a random intergrowth of 3R1 (60%) and 3R2 (40%) polytypes (see the simulated and calculated patterns in Fig. 9). They chose the 3R2 polytype on the grounds that a synthetic example has been reported [108] as discussed above, but it should be noted that this is believed to contain mostly hydroxide, rather than carbonate, anions in the interlayers. This indicates that the nature and degree of stacking faults observed may vary according to method of preparation as well as sample composition. It should also be noted that, as in any empirical fit, there may not be a unique solution and several solutions involving different stacking sequences may exist.

Structural Aspects of Layered Double Hydroxides

23

Fig. 9 a Experimental diffraction pattern of an Mg/Al – CO3 LDH and b the simulated pattern assuming a random intergrowth of 3R1 (60%) and 3R2 (40%) polytypes. Reprinted with permission from [103]. Copyright The Clay Minerals Society

Stacking faults and turbostratic distortion have been associated [101] with a mismatch in the geometries of layers and interlayer anions, which prevents an ideal packing of the slabs of the type that is often observed with interlayer carbonate anions. Delmas et al. [131, 132] have suggested, however, that an alternative explanation, local distortions within slabs due to microstrains arising from the distribution of electrostatic charges, is more likely. They based this conclusion on the fact that a γ -oxyhydride with the formula H0.20 Na0.12 K0.21 (H2 O)0.47 Ni0.70 Co0.30 O2 can be converted to a vanadate-intercalated LDH by reduction with hydrogen peroxide in the presence of ammonium metavanadate (a so-called chimie douce method of synthesis) and then subsequently regenerated by oxidation [131]. The XRD pattern of the γ -oxyhydride precursor exhibits sharp (01l) (strong) and (10l) (weak) reflections characteristic of a regular rhombohedral stacking (space group R3m, 3R1 polytype in the notation used above, P3 type in the notation used for layered metal oxides), and thus extensive rotation of the slabs would be required if the reduction-oxidation processes were to be accompanied by a switch between rhombohedral and turbostratic stacking. It was suggested that this is unlikely under the mild conditions employed.

24

D.G. Evans · R.C.T. Slade

3.2 Arrangement of Anions and Water in Interlayer Galleries 3.2.1 Evidence from X-ray Diffraction A considerable amount of information can be obtained about the location and bonding of the interlayer anions by XRD, although the detailed structure of the highly disordered hydrogen bonding network in the interlayer galleries is not amenable to study by this method. When the intercalated anions do not contain atoms with large scattering power, intensities of the basal (00l) reflections are mainly governed by the X-ray intensity scattered by metal cations in the host layers as discussed in Sect. 3.1 and the intensities generally decrease as l increases, although some anomalous intensities can be observed even with organic guests containing only light atoms [20, 44, 133]. Care should be taken where the positions of the (006) reflections of the intercalated species are close to those of the (003) reflections of possible impurities such as LDH-Cl/NO3 precursors or LDHCO3 derived by uptake of atmospheric carbon dioxide [104]. The positions of the basal reflections of NiII /FeIII and NiII /CoIII LDHs containing interlayer carbonate anions show no change on heating up to 160 ◦ C, but the intensity of the (006) reflection (I(006)) decreases relative to I(003). Simple simulations suggest a reduction in interlamellar electron density associated with loss of water is responsible [110]. If the interlayer contains a metal complex anion as the guest, the second basal reflection is generally although not always [134] more intense than the first, which has been attributed to an increased electron density at the midpoint of the interlayers, where the metal is presumed to be located [54, 135–138]. A similar inversion of intensity has been observed for sulfonated 9,10-anthraquinones [139, 140]. Many examples of polyoxometallate anions intercalated in LDHs have been reported and inversion of intensities of basal reflections is sometimes, but not always, observed [60, 141, 142]. (In these cases, an additional broad peak is often observed between the first and second basal reflections, which is usually attributed to a partially hydrolyzed salt of the polyoxometallate [141]). More detailed studies can indicate the orientation of the guest anion in the interlayer galleries since the resultant distribution of electron density as a function of fractional coordinate along the c axis is reflected in the intensities of the basal reflections. Intercalation of paramolybdate ([Mo7 O24 ]6– ) in Mg/Al LDHs gave a material showing no (003) reflection [143]. The intensities of the (00l) reflections were calculated from structure factors and Lorentz polarization factors for two model structures (having the same basal spacing) with different orientations of the paramolybdate anion (Fig. 10). The calculated intensities for model A [0.1 (l = 3), 100.0 (l = 6), 45.8 (l = 9), 0.2 (l = 12) and 12.4 (l = 15)] were very close to the observed intensities.

Structural Aspects of Layered Double Hydroxides

25

Simulation with the DIFFaX program of the XRD patterns of simple model structures involving MO2 (M = Mo, W) slabs inserted to varying extents between slabs of NiO2 were shown [101] to be remarkably effective in reproducing the XRD patterns of materials composed of [M2 O7 ]2– anions intercalated in NiII /CoIII LDHs, where I(003) < I(006) and the ratio of the intensities varies with the amount of interlayer guest. Intercalation of [Cr2 O7 ]2– anions in NiII /Al LDHs [144] gives a material with similar interlayer spacing but with I(003) > I(006), presumably associated with the presence of first row transition metals in both layers and interlayers. Intercalation of the same anion in Mg/Al LDHs [145] gives a material with I(003) < I(006). A significant broadening of the (006) reflection was observed in the XRD patterns of [M2 O7 ]4– (M = V, P) intercalated in the latter host, which was attributed to a disorder in the periodicity of the (00l) planes, possibly related to the formation of polymeric chains in the interlayer galleries [146]. In the case of 9,10-anthaquinone-2,6-disulfonate, two intercalated phases with basal spacings of 1.9 and 1.2 nm could be prepared [140]. It was proposed that these involve almost vertical and tilted guest anions, respectively. Interestingly the (00l) peak intensities show very different patterns, which presumably reflect the different locations of the sulfur atoms along the c axis. Ferrocenecarboxylate and 1,1
-ferrocenedicarboxylate anions have been intercalated in Zn/Al LDHs giving materials [120] with basal spacings of 2.00 and 1.55 nm. These were interpreted in terms of a bilayer and monolayer of guest species, respectively, as discussed in Sect. 3.2.4. The (00l) peak intensities also show very different patterns, again presumably reflecting the different locations of the one or two iron atoms along the c axis. When the intensities of several (5 or more) (00l) reflections are well-defined, a one-dimensional electron density distribution along the c axis may be calculated. In this method, which

Fig. 10 Two possible orientations of paramolybdate ([Mo7 O24 ]6– ) in the interlayer of LDHs. Reprinted with permission from [143]. Copyright American Chemical Society

26

D.G. Evans · R.C.T. Slade

has been widely used for other layered solids [147–149], peak intensities are obtained from integrated peak areas and, after correction for Lorentz polarization effects, used to generate structure factors and then finally electron density maps can be generated for z = 0–1. To date, there have been only a few reports of such calculations for LDHs. In the case of a tetrasulfonated perylene dye intercalated in an Mg/Al LDH, the electron density distribution plot obtained is shown in Fig. 11. The high electron densities at the edges (z = 0 and z = co ) are associated with the close packed metal hydroxide layers, whereas the two maxima in electron density in the interlayer gallery are attributed to the sulfur atoms, since these have the highest scattering power. The two minima in electron density can be associated with the C – N single bonds, which are the site of minimum electron density in the molecule. On this basis, the structural model shown in Fig. 11 was proposed [150].

Fig. 11 One-dimensional electron density projection along the c-axis for a perylene dye intercalated in an Mg/Al LDH and a structural model based on this data. Reprinted with permission from [150]. Copyright Wiley

Structural Aspects of Layered Double Hydroxides

27

In the case of stearate ions intercalated in Mg/Al LDHs, the electron density shows a pronounced minimum in the center of the interlayer region, which is consistent with the presence of a bilayer of guest molecules [151, 152]. The electron density distribution for an anionic azobenzene derivative intercalated in an Mg/Al LDH has also been reported [153]. 3.2.2 Rietveld Refinement of Interlayer Structure When sufficiently large, defect-free single crystals of a material are available, its structure can be precisely determined by X-ray diffraction. Unfortunately, most LDHs are only available in the form of fine crystallites, and this is generally not possible. When suitable crystals are not available, the Rietveld technique can be used to refine structural details (atomic coordinates and site occupancies) for LDHs, if their idealized structure is known. If there is disorder present, such as stacking faults or interstratification, which as discussed in Sect. 3.1 are common for layered species such as LDHs, then a Rietveld refinement will not be possible [39]. Care must be taken to avoid preferential orientation of the platelets [105, 106, 123, 154]. A Rietveld analysis for an LDH with the formula [Zn4 Al2 (OH)12 ]Cl2 ·nH2 O has been reported [106]. The layers stack in the 3R1 polytype (R3m space group). Several possible locations for interlayer chloride ions and water molecules were considered: the 3(b) sites (site symmetry 3m) at the midpoint of the O· · · ·O vector between oxygen atoms of adjacent layers, and adjoining sets of sites distributed in groups of six around the three-fold O· · · ·O axis – 18(g) positions (site symmetry .2) (Fig. 12) or 18(h) positions (site symmetry .m).

Fig. 12 Possible location of interlayer species in [Zn4 Al2 (OH)12 ]Cl2 ·nH2 O. Reprinted with permission from [106]. Copyright Academic Press

28

D.G. Evans · R.C.T. Slade

Geometrical considerations suggest that for chloride ions, OH· · · ·Cl· · · ·HO hydrogen bonding will be maximized if the anions occupy 3(b) sites, whereas OH· · · ·O· · · ·HO hydrogen bonding is optimized if the water molecules occupy 18(g) or 18(h) sites. For the fully hydrated ambient temperature material (n = 4), the results indicate that the water molecules do indeed occupy the 18(g) or 18(h) sites and the chloride ions exhibit strong oscillations around the equilibrium position 3(b) associated with their hydrogen bonding interaction with water molecules. Experimental, calculated and difference powder XRD patterns are shown in Fig. 13 [106]. At higher temperatures as water is progressively removed, the chloride ions become increasingly localized in 3(b) positions. This is reflected in an increase of the intensity of (110) reflection relative to that of the (113) reflection observed in both experimental and simulated XRD patterns. Rietveld refinement of analogous [M4 Cr2 (OH)12 ]Cl2 ·4H2 O (M = Zn, Cu) materials suggest that the chloride ions also occupy 18(g) sites with the water molecules disordered over 18(g) (M = Zn) or 36(i) (M = Cu) sites [25]. It was suggested that the layer oxygen-chloride distances (O· · · ·Cl = 0.2827 nm for M = Zn and 0.2916 nm for M = Cu) are consistent with the presence of OH· · · ·Cl hydrogen bonding since they are similar to the O· · · ·O distances between hydroxyl groups and water molecules (0.2827 nm for M = Zn and 0.3151 nm for M = Cu), for which hydrogen bonding can be demon-

Fig. 13 Experimental, calculated and difference powder XRD patterns for [Zn4 Al2 (OH)12 ] Cl2 ·nH2 O at ambient temperature. Reprinted with permission from [106]. Copyright Academic Press

Structural Aspects of Layered Double Hydroxides

29

strated by IR spectroscopy. Refinement of the structure of the related green rust material with approximate stoichiometry [FeII 4 FeIII 2 (OH)12 ]Cl2 ·4H2 O at ambient temperature suggested that chloride and water both occupy sites off the three-fold O· · · O axes, with the water slightly closer to the axis than the chloride (0.032 nm vs. 0.080 nm) [155]. This was ascribed to the larger radius of Cl– . A single crystal study of the mineral iowaite, [(Mg5.9 FeII 0.1 )FeIII 2 (OH)16 ]Cl1.4 (OH)0.48 (CO3 )0.06 ·4H2 O (space group R3m) suggested that disordered interlayer Cl– /OH– /H2 O species were located off the three-fold O· · · O axes [156]. A Rietveld refinement of a second green rust with approximate stoichiometry [FeII 4 FeIII 2 (OH)12 ]SO4 ·ca. 8H2 O revealed a different packing sequence and interlayer structure [157]. The material has the single layer 1H polytype and the interlayer consists of a double layer of sulfate anions and water molecules (Fig. 14), as discussed in more detail in Sects. 3.5 and 4.3. The layer hydroxyl groups and interlayer oxygen atoms have an almost ideal hexagonal close packed arrangement and the stacking sequence may be described as AbC(∼A)(∼C)AbC, (where (∼A) represents an interlayer species located in, or close to, an A site). On geometrical grounds, it can be seen that sulfate is unable to form the 3R1 polytype with a single layer of anions as described above for either planar (e.g. carbonate) or spherical (e.g. chloride) anions, where the stacking sequence can be denoted AbC(∼C)CaB(∼B) BcA(∼A)AbC. Rietveld refinements of the structures of Mgn Al – CO3 LDHs (n = 2.0, 5.0) have been reported [39]. The oxygen atoms of both carbonate ions and water molecules were found to occupy a single set of 18(h) sites distributed around the three-fold O· · · O axis with a 3R1 polytype stacking (space group R3m).

Fig. 14 View of the ordered interlayer structure of the green rust with stoichiometry [FeII 4 FeIII 2 (OH)12 ]SO4 · ca. 8H2 O. Reprinted with permission from [157]. Copyright Elsevier SAS

30

D.G. Evans · R.C.T. Slade

3.2.3 Spectroscopic Measurements The coordination environment around the metal in complex anions intercalated in LDHs has been studied by XAS. Extended X-ray absorption fine structure (EXAFS) studies of noble metal chloro-complexes such as [IrCl6 ]2– [137] and [PtCl6 ]2– [138] indicate that intercalation in LDHs is accompanied by partial hydrolysis of the M – Cl bonds. Molybdenum K-edge EXAFS measurements suggested that reaction of a Zn/Al LDH containing intercalated 2,2
-bipyridine-5,5
-dicarboxylate anions with MoO2Cl2 (THF)2 afforded a material containing a hydrolyzed species based on [O2 Mo – O – MoO2 ] units rather than the anticipated bipyridyl complex, since there was no evidence of Mo – Cl or Mo – N coordination shells [50]. In the case of the oxalato-complexes [MII (C2 O4 )2 ]2– (MII = Cu, Co) and [MIII (C2 O4 )3 ]3– (MIII = Ga, Mn), intercalated in Mg2 Al and Zn2 Al LDHs, EXAFS studies indicated only minor changes in coordination environment compared with that in salts of the complexes, with the perturbations induced by Mg2 Al layers being slightly more significant [135]. Comparison of the third coordination shell of [M(C2 O4 )3 ]3– (M = Ga, Mn) indicated that the ions were oriented differently in the interlayer galleries, with the former having two short and four long Ga· · · O distances and the latter having three short and three long Mn· · · O distances, corresponding to the arrangements shown in Fig. 15. EXAFS studies of [M2 O7 ]2– (M = Mo, W) intercalated in NiII /CoIII LDHs indicated that the geometry of the intercalated anions was very similar to that in their salts [101] with one long M – O bond (involving a bridging M – O – M unit) and three shorter M – O bonds making up a distorted tetrahedral coordination shell around the metal. A second coordination shell containing two oxygen atoms was ascribed to water molecules pointing towards triangular faces of MO4 tetrahedra. A similar conclusion was drawn in the case of [CrO4 ]2– and [Cr2 O7 ]2– intercalated in Ni/Al LDHs [144]. Since the degree of oligomerization in polyoxometallates is very sensitive to pH, changes in the structure of the anion may be observed on intercalation in LDHs. Comparison of EXAFS data for intercalated species with those of reference compounds often provide a useful insight into the structure of the guest [158]. For example, X-ray absorption near-edge spectroscopy (XANES) and EXAFS data at the V K-edge of a material produced by ion-exchange of a Zn/Al – Cl LDH with ammonium metavanadate are essentially identical to that of a decavanadate salt, suggesting that [V10 O28 ]6– anions are present in the interlayer galleries of the material [159]. When the intercalation reactions were carried out at higher pH, the EXAFS data suggest that the interlayers contain polymeric anions based on VO4 tetrahedra. EXAFS cannot be used to distinguish between [Fe(CN)6 ]4– and [Fe(CN)6 ]3– because the difference in Fe – C and Fe· · · N distances is within experimental error. Simulation of the Fe K-edge XANES absorption edge indicates, how-

Structural Aspects of Layered Double Hydroxides

31

Fig. 15 Different arrangements of [MIII (C2 O4 )3 ]3– in LDH intergalleries: M = Mn (top) and M = Ga (bottom). Reprinted with permission from [135]. Copyright American Chemical Society

ever, that intercalation of either [Fe(CN)6 ]4– or [Fe(CN)6 ]3– in LDHs leads to a partial redox reaction and a mixture of both species being present in the interlayers [160]. Vibrational spectroscopy has been widely applied in the study of LDHs [161, 162] but a somewhat confusing variety of spectral data and interpretations have appeared in the literature. In this section, we focus on the information that can be obtained regarding the structure of the interlayer anions. The unperturbed carbonate ion has point symmetry D3h . Group theoretical analysis predicts four normal modes: the ν1 symmetric stretch of A
1 symmetry at 1063 cm–1 , the ν2 out of plane bend of A

2 symmetry at 880 cm–1 , the ν3 asymmetric stretch of E
symmetry at 1415 cm–1 , and the ν4 in plane bend of E
symmetry at 680 cm–1 [22]. The ν2 mode is IR active only, the ν1 mode is Raman active only, whilst the two E
modes are both IR and Ra-

32

D.G. Evans · R.C.T. Slade

man active. Although many publications merely report a broad band in the C-O stretching region of the IR spectrum, band component analysis suggests that the ν3 band is split into two bands separated by some 30–60 cm–1 [163]. This can be interpreted in terms of a lowering of the symmetry to C2v or Cs . In such cases the ν1 stretch should become allowed and indeed a weak band can often be seen [161] at about 1060 cm–1 . It has been suggested [161] that lowering of symmetry is associated with hydrogen bonding of the carbonate anion to layer hydroxyl groups and/or interlayer water molecules. A number of studies by Frost, Kloprogge and co-workers [163–166] on LDHs containing a variety of cations have indicated that the position of the two peaks arising from the ν3 band depend on the nature of the divalent cation, suggesting that the variation in polarizing power of the cation affects the hydrogen bonding between the hydroxyl groups bound to it and the carbonate anion. A similar suggestion has been advanced for a series of LDHs with varying trivalent cations [73]. Some authors have suggested that the ν3 band of the nitrate ion in LDHs shows a similar splitting, although others have reported a single band indicative of no loss of symmetry, with additional bands being assigned to carbonate impurities [22, 161]. The arrangement of nitrate ions in the interlayer galleries of LDHs has been the subject of some controversy in the literature. Xu and Zeng [55, 79] have reported that there is an abrupt increase in basal spacing in LDHs of the type [Mg1–x Alx (OH)2 ](NO3 )x ·nH2 O when x exceeds 0.26, although Marcelin et al. [167] reported that the difference was much smaller if the preparation was carried out under hydrothermal conditions (This may indicate that the actual Mg/Al ratio in the layers tends towards 3 under hydrothermal conditions as discussed in Sect. 2.2.1). It was suggested [55, 79] that when x < 0.26, the nitrate anions lie in the center of the interlayer galleries with their planes parallel to the layers analogous to the position adopted by carbonate anions, with their oxygen atoms located either between hydroxyl groups on adjacent layers where there are prismatic layers or hydrogen bonded more directly to just one layer if there are octahedral interlayers [168]. When x > 0.26, it was proposed that alternate nitrate ions are shifted up and down the c axis with their planes remaining parallel to the layers so that they are strongly hydrogen bonded to one layer and do not interact significantly with the other (the so-called “stick-lying” model) [55, 79]. It was suggested that this structure is not observed for carbonate-containing LDHs since for a given value of x the number of divalent carbonate anions incorporated is half that of the monovalent nitrate ions, and there is still space for the former to be accommodated in the center of the interlayer galleries even at high layer charge densities. It was proposed that the “stick-lying” model is more plausible than the alternative model involving tilting of the nitrate anions in the interlayer galleries on several grounds: the nitrate ion shows no evidence of distortion from D3h symmetry in the IR spectrum whilst the νOH stretching band shows two components sepa-

Structural Aspects of Layered Double Hydroxides

33

rated by ≈ 100 cm–1 , which was taken to indicate that some layer hydroxyl groups are hydrogen bonded to nitrate whilst others are not; 15 N NMR measurements [169] between – 100 ◦ C and + 80 ◦ C on LDHs with Mg/Al ≈ 3 are consistent with either the nitrate being rigidly held or undergoing a rapid rotation about the three-fold axis, but it should be noted that no data have been reported for LDHs with the higher interlayer spacing associated with Mg/Al ≈ 2. The “stick-lying” model been supported by Gonçalves et al. [50] who noted that on replacing nitrate by 2,2
-bipyridine-5,5
-dicarboxylate anions, which pack almost vertically with their longest dimension nearly parallel to the layers, two components were no longer observed in the νOH stretching band (although the band is very broad), suggesting that there are no longer two types of layer hydroxyl groups. On heating LDHs of the type [Mg1–x Alx (OH)2 ](NO3 )x ·nH2 O at 400 ◦ C it was suggested [170] on the basis of IR spectroscopy and XRD, that most nitrate ions are retained in the interlayer space, some with D3h symmetry and the remainder arranged vertically, directly attached to one partially hydroxylated sheet (corresponding to C2v symmetry). Raman studies on a single crystal of a related compound [Ca4 Al2 (OH)12 ](NO3 )2 ·nH2 O (n = 4, 2, 1) indicated [171] that as water is progressively removed, the symmetric stretching mode ν1 with A
1 symmetry shifts from 1059 (n = 4) to 1070 (n = 2) to 1055 cm–1 (n = 1) whilst the ν4 in plane bend with E
symmetry, which is split slightly when n = 4, is split more clearly into two components for n = 2 and gives a single peak when n = 1. These results were interpreted in terms of a mondentate nitrate ion bonded to one sheet, a bidentate nitrate bridging two sheets and a nitrate ion parallel to the layers, respectively. These suggestions were confirmed [171] by single crystal XRD for n = 4 and n = 2. In these materials, as described in Sect. 4.1, the calcium ion is seven coordinate with the six hydroxyl groups being augmented by oxygen atoms of either water or nitrate ions perpendicular to the layers (50% probability of each) when n = 4 or nitrate ions when n = 2 [171, 172]. The nitrate ions in the former compound are highly disordered, showing essentially free rotation about an axis perpendicular to the layers. Partial dehydration and change in coordination mode of the nitrate ion is accompanied by a contraction in basal spacing from 0.862 to 0.805 nm. A combined heating stage Raman microscopy and IR emission spectroscopy study of an Mg/Al LDH containing a mixture of carbonate and nitrate in the interlayers suggested [173] that nitrate ions are present in the centers of interlayers at room temperature but become grafted to the layers in a unidentate fashion at 150–170 ◦ C. Grafting is commonly observed if LDHs are heated, as discussed in Sect. 4.1. A variety of anionic transition metal cyano-complexes have been intercalated in LDHs and their structure investigated by vibrational spectroscopy [161]. There is general agreement that octahedral complexes such as [Fe(CN)6 ]4– and [Fe(CN)6 ]3– intercalated in LDHs align themselves with their three-fold axis perpendicular to the sheets, but the spectra are gen-

34

D.G. Evans · R.C.T. Slade

erally complicated by the presence of a number of peaks apart from those associated with the anion that was initially intercalated. These have been variously ascribed to the results of redox processes, hydrolysis, or the presence of salts [134, 174–176]. Ignoring these complications, for the case of intercalated [Fe(CN)6 ]4– for example, the expected one IR active band (T1u in Oh symmetry) is split into two bands, which can be attributed to a decrease in symmetry to D3d (in which case the bands can be assigned as A2u and Eu ) [174, 175, 177]. The Eu band is (x, y) polarized whilst the A2u band is z-polarized. Braterman et al. [134, 174, 177] have compared IR spectra of conventional (non-oriented) materials with those of an oriented sample prepared by allowing an aqueous suspension to evaporate on a barium fluoride disk. In the case of the oriented sample, the molecular three-fold z-axis, the crystallographic c axis and the direction of propagation of the light all coincide [177]. In this case, the material is probed with light polarized in the (x, y) direction for which the A2u mode is forbidden. Thus whilst the conventional spectrum shows two bands, in the spectrum of the oriented sample the lower frequency E2u component is suppressed [174, 177]. This method could potentially be applied to investigate the orientation of other guest anions, although care must be taken to avoid artifacts caused by uneven coverage [177]. Solid state 13 C MAS NMR studies of some organic guests have been reported [178, 179]. This method also allows the presence of co-intercalated carbonate and/or bicarbonate anions to be detected [179]. It was found that the UV absorbing agent 5-benzoyl-4-hydroxy-2-methoxybenzenesulfonic acid (MBSA), having both a strongly acidic sulfonic acid group and a weakly acidic phenolic group, could be intercalated as either a monovalent or divalent anion depending on the pH [180]. The sulfonate groups were proposed to be hydrogen bonded to triads of hydroxyl groups, since the IR spectral features in the 1000–1300 cm–1 region were nearly identical to those observed for the guanidinium salt of MBSA1– , which has been shown by single crystal XRD to have the sulfonate group in a site of C3v symmetry, hydrogen bonded to guanidinium protons [181]. A 2 H solid state NMR investigation of terephthalate dynamics and orientation in LDHs of the type [Mg1–x Alx (OH)2 ]{(p-C6 D4 (COO)2 )y (CO3 )1–y }x/2 · nH2 O has been reported [182]. When x = 0.37 and y = 0.9, the observed interlayer spacing was 1.42 nm, corresponding to a vertical arrangement of terephthalate anions [183], and the variable temperature 2 H spectra in the range 245 to 355 K were simulated using a model involving free rotation about the C – COO axis, rather than discrete 180◦ flips. It was suggested that the mobility of the terephthalate anions is inhibited as the carbonate content of the interlayer gallery is increased. When x = 0.29 and y = 0.51, the interlayer spacing was 0.76 nm, corresponding to a horizontal orientation of interlayer terephthalate anions [183]. A fraction of the terephthalate anions were found to be undergoing a similar rotation, but since this seems unlikely in the constrained environment of the contracted interlayers, it was

Structural Aspects of Layered Double Hydroxides

35

suggested this might be due to rotation of surface adsorbed terephthalate anions. A 35 Cl NMR study of the structure and dynamical behavior of ClO4 – intercalated into Mg/Al and Li/Al LDHs has shown that the anion is rigidly held at low relative humidities and temperatures, but undergoes rapid isotropic reorientation at high relative humidities and temperatures [184]. In contrast, 77 Se NMR measurements by the same authors on LDHs containing interlayer SeO4 2– indicate that the anion does not undergo isotropic reorientation at any relative humidity. This was ascribed to the greater ionic charge of the selenate ion compared with the perchlorate [185]. A 35 Cl NMR study of chloride ions intercalated in an Mg3 Al – Cl LDH and the related LDH-like hydrocalumite [Ca4 Al2 (OH)12 ]Cl2 ·4H2 O has been reported [186]. In the case of the Mg3 Al – Cl LDH, the 35 Cl NMR data show a poorly resolved signal indicating a range of Cl– environments in a disordered interlayer. There is a change from triaxial to uniaxial or nearly uniaxial symmetry over a broad temperature interval below – 40 ◦ C, suggesting a phase transition, which can also be observed by differential scanning calorimetry (DSC). In contrast, in the Ca/Al system where water molecules are coordinated to calcium ions in a seventh coordination site as discussed in Sect. 4.1, the interlayer water and chloride ions are well ordered. This may be associated with the known ordered distribution of Ca2+ and Al3+ cations in the layers (Sect. 3.3). NMR and DSC data suggest that there is a well-defined, essentially first order phase transition near 6 ◦ C, at which the symmetry of the Cl– site changes from triaxial in the low temperature phase to uniaxial or nearly so in the high temperature phase. The latter is a result of dynamic averaging of the hydrogen bonding interactions associated with highly mobile water and chloride ions [186]. 1 H NMR studies on [LiAl2 (OH)6 ]Cl ·xH2 O have also demonstrated the presence of dynamically disordered water molecules [187]. An early 1 H NMR study [167] claimed that the C2 axis of the water molecules in an Mg/Al LDH containing interlayer nitrate ions was perpendicular to the crystallographic c axis, but it was later argued this was incorrect [188]. This latter study indicated that both the C3 axis of carbonate ions and the C2 axis of water molecules are parallel to the c axis and that the water possesses rotational freedom about its C2 axis. Mössbauer spectroscopy has been used to investigate the oxidation state of iron in complexes such as [Fe(CN)6 ]3–/4– intercalated in LDHs [176]. By means of Mössbauer spectroscopy, it has also been demonstrated that ferrocene sulfonates intercalated in LDHs decompose on attempted ionexchange with sodium carbonate solutions and that the liberated Fe3+ ions become incorporated in the layers [189]. XANES (also known as near edge X-ray absorption spectroscopy, NEXAFS) at the carbon edge provides a means of determining the orientation of organic molecules relative to a planar surface and has traditionally been used to investigate organic adsorbates on single crystal surfaces. Moggridge and

36

D.G. Evans · R.C.T. Slade

co-workers proposed [104, 190] that XANES might be a potential way of determining the orientation of organic guests in the interlayer galleries of LDHs, using benzoate as an example. At room temperature, the plane of the benzoate ion in an Mg2 Al LDH was found to be tilted 35 ± 10◦ to the layers, whereas after heating at 50–100 ◦ C, the benzoate lies parallel to the layers. It was later shown [44] that the first figure is an underestimate because partial dehydration and reorientation of the benzoate anions occurs under the vacuum used during the XANES experiment, and the observed result must therefore be the average of the figure for flat and almost vertical benzoate anions. This correlates well with the observed [44] decrease in basal spacing from 1.54 to 0.90 nm. In the case of Mg3 Al LDHs, the lower benzoate loading required results in the anions adopting an essentially flat orientation even at room temperature and no evidence for an expanded structure was obtained, although it has been reported [191] in non-dried samples. The area occupied by a flat lying benzoate ion, excluding van der Waals radii, is ≈ 0.21 nm2 , whereas the area available per unit charge in Mg2 Al and Mg3 Al LDHs is ≈ 0.24 and ≈ 0.36 nm2 . Thus a flat orientation of benzoate anions results in considerable steric crowding in the former, but not in the latter, case [44]. The photophysical and photochemical properties of intercalated chromophores is of interest because of the potential application of these materials in photophysical devices and in nonlinear optics; furthermore the photophysical and photochemical response of the chromophore can provide information about the interlayer organization. In the case of fluorescein, the absorption and emission spectra of intercalated and surface adsorbed dye are markedly different, with the former being similar to the crystalline disodium salt and the latter being similar to that of the dye in solution. It was concluded that the intercalated anions are arranged in an ordered form resembling that in the crystal with π–π interactions between the chromophores [192]. The fluorescence of methyl orange-intercalated LDHs is similar that of the microcrystalline dye but is shifted to slightly higher energy in the case of the hydrated species, whilst dehydration leads to a shift to lower energy. This was interpreted in terms of the packing of the anions in the dehydrated form being closer to that in the dye itself [193]. The absorption spectrum of 9-anthracenecarboxylate intercalated in LDHs shows an unexpected band around 490 nm that was attributed to an aggregate of the guest anions. Excitation around 500 nm generates an emission attributed to the same type of aggregate [133]. UV spectroscopic evidence for aggregates of salicylate anions intercalated in LDHs has also been reported [194]. It has been suggested that conventional and space-resolved fluorescence spectra obtained by confocal fluorescence spectroscopy can give valuable information about the distribution of fluorescent dyes in the interlayer galleries of LDHs [195]. Reaction of 4-nitrohippuric acid (O2 NC6 H4 CONHCH2 COOH) (NHA) with [LiAl2 (OH)6 ]Cl leads to intercalation of the neutral organic molecule with-

Structural Aspects of Layered Double Hydroxides

37

out loss of chloride ion. The resulting material exhibits frequency-doubling characteristics, indicating an ordered array of guest molecules within the interlayer galleries [196]. Interestingly, crystals of pure NHA do not exhibit any nonlinear optical properties as a result of the centrosymmetric packing in the crystal. The alignment of 4
-chloro-4-stilbenecarboxylic acid (CSC) intercalated in Mg/Al LDHs has been investigated [197] by electric linear dichroism, which indicated that the guest anions were intercalated as a double layer, with the tilt of the molecular plane being ≈ 40◦ . 3.2.4 Molecular Modeling It is generally accepted that the interlayer arrangement in LDHs depends strongly on the area available to each anion. Since the distance between adjacent metal ions in the layers is equal to the unit cell parameter ao , the area occupied by one M(OH)2 unit (shown by the rhomb highlighted in Fig. 16) is a2o sin 60◦ [193]. For an LDH with layers of the type [MII 1–x MIII x (OH)2 ]x+ , the area per unit charge is therefore (1/x)a2o sin 60◦ . When a monovalent anion balances the positive charge on one side of the sheet, the same charge is not “available” for another anion approaching the sheet on the other side. This means a monolayer of intercalated monovalent anions can be formed if their cross-sectional area is ≤ (1/x)a2o sin 60◦ , regardless of whether the anions “see” the positive charges of the layer on only one side or on both sides of the sheet. The gallery height in LDHs is normally estimated [1] by subtracting the thickness of the brucite-like layers (assumed [198] to be 0.48 nm) from the

Fig. 16 Top view of brucite-like layer, where the area occupied by one M(OH)2 unit is shown by the highlighted rhomb. Reprinted with permission from [193]. Copyright American Chemical Society

38

D.G. Evans · R.C.T. Slade

basal spacing determined by XRD, i.e. dbasal = 0.48 + lanion (nm), where lanion includes the van der Waals radii of appropriate external atoms of the anion. Alternatively it has been suggested [130, 189, 199] that the basal spacing in intercalated carboxylate or sulfonate species can be estimated by the equation dbasal = llayer + 2lO–H–O + l
anion , where the layer thickness (llayer ) is 0.21 nm (the intralayer O· · · O distance perpendicular to the layers in brucite [13]), lO–H–O corresponds to the length of a strongly hydrogen bonded O – H··O unit (0.27 nm) and l
anion is the length of the anion. Thus dbasal = 0.75 + l
anion (nm). A theoretical model for estimating the textural properties of LDHs by combining geometrical models of the layers and the intercalated anions has recently been proposed [200]. The model allows the estimation of interpillar distances, interlamellar and external areas, the interlamellar free volume, fraction of external anions and the apparent and true density of the LDH. For well-crystallized LDH samples, good agreement between calculated and experimental results was found, whereas for poorly crystalline samples, a correlation between the degree of crystallite agglomeration and experimental values was proposed. For example, the very high surface area of an Mg3.3 Al – [Fe(CN)6 ]3– LDH was satisfactorily rationalized [200] and the model has also been used to aid in the discussion of the orientation of [H4 Co2 Mo10 O38 ]6– anions in Mg/Al LDHs [201]. Calculations based on simple molecular models and the charge density of the layers suggest that sulfopropylated-β-cyclodextrin and carboxyethylatedβ-cyclodextrin are arranged in the interlayer galleries with their conical axis parallel to the layers with a packing structure which is similar to that in crystalline cyclodextrin complexes, where the molecules are arranged in a brickwork pattern [202]. The molecular dimensions of the (4-phenylazophenyl)acetate anion have been calculated [153] based on the molecular structure determined by the MM2 semiempirical molecular dynamics method together with appropriate van der Waals radii (Fig. 17). The area per unit negative charge on the layers can be calculated from the crystallographic structure. At low anion loadings, it was found that the guest anions lie parallel to the layers for LDHs with Mg/Al ratios of either 2 or 3. At higher loadings, a vertical orientation is observed. In the case of Mg/Al = 3, the cross-sectional area of the guest is smaller than the area per unit charge on the layers so that the molecules are arranged in an antiparallel fashion with little interaction between them. For Mg/Al = 2, the cross-sectional area of the guest is larger than the area per unit charge, so the molecules are slightly staggered in order to reduce the lateral interactions, leading to an increase in interlayer distance (Fig. 18). The most probable arrangement of fluorescein anions in LDHs has been investigated [192] using Hyperchem, a molecular visualization and simulation program and the predicted arrangement was consistent with both the

Structural Aspects of Layered Double Hydroxides

39

Fig. 17 Calculated molecular dimensions of the (4-phenylazophenyl)acetate anion. Reprinted with permission from [153]. Copyright American Chemical Society

Fig. 18 Staggered arrangement of the (4-phenylazophenyl)acetate anion in LDH galleries. Reprinted with permission from [153]. Copyright American Chemical Society

observed basal spacing and the fluorescence behavior of the intercalate discussed in Sect. 3.2.3. The approximate van der Waals dimensions of the methyl orange molecule have been calculated using the Hyperchem program and are shown in Fig. 19 [193]. It was calculated that the cross-sectional area is ≈ 0.22 nm2 at the – SO3 – group (although other workers [203] have estimated the area to be as large as 0.27 nm2 ) and ≈ 0.29 nm2 at the – N(CH3 )2 group. When intercalated in a Zn2 Al host, for which the unit layer charge area is ≈ 0.25 nm2 , the fact that the cross-sectional area of the – N(CH3 )2 moiety is larger than the area available suggests that the molecules will pack in an antiparallel fashion, as shown by the computer model in Fig. 20a. It was suggested that in the hydrated state, water molecules may occupy the space between – N(CH3 )2 groups and the layers, whereas when the structure is dehydrated the methyl orange anions can further interpenetrate as shown in Fig. 20b. Interestingly, the difference in basal spacing between hydrated and dehydrated forms, 0.27 nm, is very close to the diameter of a water molecule [193].

40

D.G. Evans · R.C.T. Slade

Fig. 19 Approximate van der Waals dimensions of the methyl orange molecule calculated using the Hyperchem program. Reprinted with permission from [193]. Copyright American Chemical Society

The intercalation of dodecyl sulfate, for which the calculated crosssectional area is ≈ 0.28 nm2 , in Znn CrIII LDHs with n = 2, 3 and 6 has been reported [204]. The observed basal spacing was similar in each case and characteristic of a monolayer of surfactant. The estimated area per unit charge in the LDHs is 23, 30 and 54 nm2 , respectively. It was suggested in the case of Zn3 Cr, the similarity between the cross-sectional area of the surfactant and the available area per unit charge leads to the formation of a closely packed monolayer of surfactant anions. In the case of Zn6 Cr, intercalation of just sufficient anions to balance the layer charge would lead to them being too widely spaced to allow significant hydrophobic interactions between the alkyl chains. Therefore additional surfactant anions along with sodium cations are incorporated to give a closely packed monolayer of surfactant species. The presence of significant amounts of sodium was confirmed by elemental analysis. In contrast, for Zn2 Cr, there is insufficient space in a monolayer to accommodate the necessary amount of surfactant anions to balance the high layer charge and as a result, cointercalation of nitrate ions, confirmed by elemental analysis, is observed. In the case of Mgn Al LDHs (n = 2–5) intercalated with dodecyl sulfate, it was shown that the affinity for chlorinated organic solvents reached a maximum for n = 3, which was interpreted in terms of the arrangement of surfactant being optimized for inclusion of hydrophobic guest molecules [205].

Structural Aspects of Layered Double Hydroxides

41

Fig. 20 Models showing arrangement of methyl orange anions in a hydrated and b dehydrated LDHs. Reprinted with permission from [193]. Copyright American Chemical Society

The molecular dimensions of the non-steroidal anti-inflammatory drugs salicylate and naproxen have been determined using the CS Chem 3D Pro program. Comparison of the dimensions of the anions (taking into account the van der Waals radii of the external atoms) with the observed basal spacings suggested that both anions form a tilted bilayer of anti-parallel molecules with carboxylate groups pointing towards the layers and the hydrophobic parts of the molecule oriented towards the center of the interlayer galleries [179]. An analogous anti-parallel packing with interpenetrating chains was proposed for dodecylbenzene sulfonate intercalated in an Mg2 Al LDH [203]. It was suggested that there are strong hydrophobic interactions between the alkyl chains because the calculated accessible area of a methylene group in the galleries (0.22 nm2 ) is similar to that in a close-packed monolayer (0.20 nm2 ),

42

D.G. Evans · R.C.T. Slade

and that this contributes to the high affinity of the surfactant for the host layers (as shown by the observation that it partly displaces carbonate anions from the interlayer galleries). In the case of reactions of ethanolic solutions of myristic acid, CH3 (CH2 )12 COOH, with an Mg3 Al – Cl LDH, it was found that the acid was intercalated as neutral molecules in a bilayer and chloride anions were retained as charge-balancing anions in the interlayers [206]. A model for the arrangement of a bis(2-mercapto-2,2-diphenylethanoate) dioxomolybdate(VI) complex, {MoO2 [OOCC(S)(C6 H5 )2 ]2 }2– , in an LDH has been proposed on the basis of the crystal structure of the ammonium salt of the anion precursor [207]. Models of the arrangement of ferrocenecarboxylate and 1,1
-ferrocenedicarboxylate in LDHs have been suggested, using the conformation and dimensions of the guest molecules obtained from single crystal XRD studies on their heterobimetallic complexes, together with the van der Waals radius of oxygen [120]. The area occupied by a 1,1
ferrocenedicarboxylate dianion having its longest dimension perpendicular to the layers was estimated to be ≈ 0.40 nm2 , which is smaller than the calculated maximum area available and a monolayer arrangement of the dianions was proposed; for the ferrocenecarboxylate, a monolayer is not possible on packing grounds and a bilayer of monoanions was proposed. This is consistent with the observed interlayer spacings of 1.55 and 2.00 nm respectively. On dehydration, the dianion reorientates so that its longest dimension is almost parallel to the layers and a reduced layer spacing of 1.23 nm is observed. Napthalene-2,6-disulfonate was reported to intercalate into a Ca2 Al layered double hydroxide-like host material as a tilted monolayer, whereas naphthalene-2-sulfonate was intercalated as a perpendicular bilayer [208]. Similarly when intercalated in Zn2 Al LDHs, anthraquinone-1,5-disulfonate (AQ15) and anthraquinone-2,6-disulfonate (AQ26) are arranged in a monolayer, whereas anthraquinone-2-sulfonate (AQ2) is arranged in a bilayer with interdigitated antiparallel guest anions [209]. Interestingly, competitive intercalation experiments showed that the affinity of the LDH for the three ions varies in the order AQ2 > AQ26 AQ15, which was taken to indicate that the hydrophobic interactions between the anions in the bilayer structure contribute significantly to the stability of the intercalated monovalent species. This type of molecular recognition capability has also been demonstrated for other LDHs by means of the selective uptake of fumarate from fumarate/maleate mixtures, of one isomer from a mixture of naphthalene disulfonates, and of terephthalate from mixtures of all three benzenedicarboxylate isomers by Li/Al LDHs [210, 211]. The extent of separation was found to be both temperature and solvent dependent. The latter suggests that differences in solvation architecture around the anion in the bulk solvent and in the interlayer galleries play an important role in the affinity of LDHs for a particular anion [19]. The relative disposition of the negative charges in the dianions also presumably has a significant effect [19].

Structural Aspects of Layered Double Hydroxides

43

Experimental measurements and computer simulations have demonstrated that the orientation of terephthalate anions in the interlayer galleries of LDHs is strongly dependent on both the charge density of the layers and the interlayer water content [183]. LDHs with high layer charge density (e.g. Mg2 Al) and high water content give materials with a basal spacing of 1.40 nm, suggesting a vertical orientation of the anion with respect to the layers is favored. For lower layer charge density (e.g. Mg3 Al) and low water content, a basal spacing of 0.84 nm is observed, consistent with a horizontal orientation of the anion. The two phases may be interconverted by cycles of dehydration-rehydration. It has been suggested [212] that under true coprecipitation conditions near pH ≈ 10 (at lower pH, precipitation of the trivalent metal hydroxide occurs first, followed by reaction with M2+ cations in solution), that Mg/Al and Mg/Ga LDHs with interlayer terephthalate ions only form when the Mg2+ : Al3+ (Ga3+ ) ratio is 2 : 1 and that chloride intercalates are formed at higher ratios. At lower pH, it was suggested that mixtures of an LDH having Mg2+ : Al3+ (Ga3+ ) = 2 : 1 with α-FeOOH or GaOOH were formed when the Mg2+ : Al3+ (Ga3+ ) ratio in the reaction mixture exceeded 2 : 1. This was interpreted in terms of a high layer charge density being necessary for the incorporation of hydrophobic anions such as terephthalate in LDH interlayers in order to create a continuous hydrophobic layer between the sheets. Many workers have shown [3, 19] that the gallery height in LDHs containing long chain aliphatic carboxylate, dicarboxylate, sulfonate or sulfate guests increases as the chain length increases. For the case of α, ω-dicarboxylate anions – OOC(CH2 )n COO– , the basal spacing shows a mean increase of ≈ 0.127 nm/CH2 from n = 3 to 12 as shown in Fig. 21 [213] for Mg/Al LDH

Fig. 21 Variation in basal spacing of α, ω-dicarboxylate anions – OOC(CH2 )n COO– , intercalated in Mg/Al LDH nanocrystals supported on silicon substrates. Reprinted with permission from [213]. Copyright American Chemical Society

44

D.G. Evans · R.C.T. Slade

nanocrystals supported on silicon substrates. Since this is approximately equal to the projection of the C – C bond length on the chain axis [214], it implies a monolayer of anions oriented vertically in the interlayer galleries. A bilayer of anions would imply an increase of twice this value if the molecules are vertical, or less than twice if the molecules are oriented at an angle to the layer [1, 215]. Most examples of alkyl carboxylates (as well as other long-chain monofunctional surfactant anions such as alkyl sulfates and alkyl sulfonates) intercalated in LDHs have their hydrocarbon chains closely packed in a monolayer, as an interdigitated structure or in a bilayer tilted at an angle of about 55◦ with respect to the layers [19, 204]. This is similar to the tilting habit of saturated fatty acids in their crystals and allows both carboxylate oxygen atoms to form hydrogen bonds equally to the hydroxide layers [216]. It has been reported that formation of monolayer/interdigitated structures is favored by high temperature [152]. It was shown that reaction of Mgn Al LDHs (n = 2, 3 or 4) with sodium stearate led to the uptake of C18 guest species of up to 230% of the anion exchange capacity (AEC), giving interlayer arrangements very similar to the Langmuir-Blodgett (LB) bilayer structure [151]. Two-dimensional XRD measurements on cast films of the intercalates were carried out using an image plate (IP) detector. The IP image (Fig. 22) shows a series of basal reflections in the vertical direction and additional peaks in the lateral direction (Fig. 23), which is consistent with a regularly aligned disposition of the C18 guests in a distorted hexagonal structure in the LDH interlayers (Fig. 24). This arrangement is also similar to that adopted by the headgroups of long chain aliphatic acids in LB films. Intercalation at over 100% AEC was shown to be a result of co-intercalation of sodium cations or neutral stearic acid molecules. In the case of trans-CH3(CH2 )7 CH = CH(CH2 )7 COOH (elaidic acid), two different intercalated LDHs with a monolayer of elaidate anions

Fig. 22 Image plate study of a stearate/LDH cast film. Reprinted with permission from [151]. Copyright American Chemical Society

Structural Aspects of Layered Double Hydroxides

45

Fig. 23 XRD diffraction patterns in the lateral and vertical directions in the image plate study of a stearate/LDH cast film. Reprinted with permission from [151]. Copyright American Chemical Society

Fig. 24 Structural model of a stearate/LDH composite showing a tilted C18 bilayer and b regular packing of C18 in the interlayer galleries. Reprinted with permission from [151]. Copyright American Chemical Society

(basal spacing of 3.08 nm) and a mixed bilayer of elaidate anions and neutral elaidic acid molecules (basal spacing of 4.88 nm) have been prepared (Fig. 25). In the case of the corresponding cis-isomer (oleic acid), however, the kink in the chain imposed by the double bond allows the chains to overlap only in the region below the double bond, sticking the chains together like Velcro (Fig. 25), resulting in a basal spacing of 3.56 nm. The geometry of the oleate anions is similar to that adopted by oleic acid itself [216].

46

D.G. Evans · R.C.T. Slade

Fig. 25 Proposed packing arrangements in LDHs intercalated with a monolayer of elaidate anions (on the left), a mixed bilayer of elaidate anions and neutral elaidic acid molecules (center) and oleate anions (on the right). Reprinted with permission from [216]. Copyright American Chemical Society

3.2.5 Guest-guest Interactions The photo-induced [2 + 2] dimerization of crystalline trans-cinnamic acid has been shown to require both close contacts of approximately 0.4 nm and a specific orientation of the unsaturated bonds, and can therefore be used to explore the arrangement of cinnamate ions in the interlayer galleries of LDHs [19, 217]. It was suggested that the packing of o-chlorocinnamate and p-chlorocinnamate as bilayers between LDH layers is qualitatively similar to the packing of the same ions between the cations in their hydrated sodium salts [218]. Head-to-head (HH) cyclodimers were formed selectively on irradiation of LDHs containing interlayer cinnamate ions, indicating that the arrangement of anions in the interlayer galleries is such as to preclude the formation of the head to tail (HT) dimer. The ratio of anti-HH to syn-HH increases markedly as the Mg/Al ratio in the layers increases; it was argued that longer guest-guest distances give an appropriate geometry for formation of the anti-HH isomer [219]. This system has also been studied [217] using molecular mechanics computer simulations of the type discussed in Sect. 3.2.6. In agreement with experiment, it was found that for Mg/Al ratios of 2 and 3, the arrangement of monomer pairs was unfavorable for HT dimerization and the selectivity to anti-HH dimer increased with increasing Mg/Al ratio. A model with Mg/Al of 6 and 20 wt % water content was observed to

Structural Aspects of Layered Double Hydroxides

47

have a very disordered interlayer arrangement with a preference for syn-HT dimer formation, whilst increasing the water content to 30 wt % led to cistrans isomerization rather than dimerization being favored since the guest species are relatively far apart. This highlights the role of both layer charge density and water content on reactivity of guest species. In all cases, the water molecules were found to have a tendency to migrate towards the faces of the hydroxyl sheets. The packing model discussed earlier for 4
-chloro4-stilbenecarboxylic acid (CSC) intercalated in Mg/Al LDHs was found to be consistent with the observed stereochemistry of the photodimerization of CSC in the interlayer galleries [197]. Intercalation of (Z, Z)-2,4-hexadienedioic acid (muconic acid) in an LiAl2 LDH has been reported. On photoirradiation of an aqueous suspension of the material, the guest molecules underwent polymerization. It was proposed that a vertical arrangement of guest molecules in the interlayer galleries results in the 1,3-diene moieties being arranged in a face-to-face manner, with a suitable orientation and distance for the propagation of radicals induced by photoirradiation [220]. Aromatic ketocarboxylate anions p-CH3 C6 H4 CO(CH2 )n CO2 – with varying chain lengths (n = 4–10) have been intercalated in Mg/Al LDHs [221]. The observed gallery heights show a much smaller increase with n (2.2 nm for n = 4 to 2.5 nm for n = 10) than that in the anion itself (1.36 nm for n = 4 and 2.12 nm for n = 10), suggesting that the chains become increasingly interdigitated as their length increases. Norrish type II photochemical reactions of the intercalates have been studied [221], since these reactions have been widely used as a probe for investigating ordered anisotropic systems. Either elimination (E) products (p-methylacetophenone and alkenecarboxylates) or cyclization (C) products (cyclobutanols) can be obtained. In solution, the ratio of C/E products was found to be independent of n for the above anions, whereas the intercalated anions with longer chain lengths (n = 6–10) gave negligible amounts of C products. Formation of C products proceeds via a cisoid diradical, whereas E products are derived from a transoid diradical and it was proposed that formation of the more bulky cisoid diradical was inhibited in interlayer galleries containing the anions with longer more interpenetrating chains [221]. Photochemical hydrogen abstraction reactions from aliphatic carboxylate anions by p-benzoylbenzoate (BPC) co-intercalated in LDHs have been reported. Compared with reactions in solution, the cointercalated species showed a high selectivity for hydrogen abstraction from methylene groups close to the methyl terminus of the alkyl chain [18]. This was interpreted in terms of an antiparallel packing of BPC and carboxylate anions. The enthalpy changes associated with exchange of chloride ions in a Zn2.2 Al LDH by a variety of dicarboxylate anions (oxalate, succinate, adipate and tartrate) have been measured by microcalorimetry [222]. With the exception of tartrate, the exchange processes are endothermic and the enthalpies

48

D.G. Evans · R.C.T. Slade

of exchange vary linearly with the number of carbon atoms, consistent with the intercalated dicarboxylate anions being perpendicular to the layers. The energetic parameters are consistent with the presence of hydrophobic interactions between the CH2 groups of succinate and adipate anions and hydrogen bonding between the OH groups of the tartrate anions. Since the calculated distance between uniformly spaced tartrate anions is too high to allow hydrogen bonding of this type, it was proposed that pairs or domains of closely spaced tartrate anions were present. 3.2.6 Molecular Dynamics Simulations Far IR spectroscopy is a useful technique to study the structural environment and dynamics of water molecules and ions in the interlayer galleries of LDHs, since it directly probes the intermolecular and hydrogen bonding interactions [223]. Spectra, which can be recorded by pressing small amounts of powder onto one side of a piece of Scotch tape, are often difficult to interpret because of the complex network of intermolecular interactions present and molecular dynamics simulations have been employed in order to assist in the interpretation [223]. The principal limitation of the molecular dynamics technique is the accuracy of the forcefield employed, and several methods have been proposed for calculations involving LDHs [223, 224]. It was found that the forcefield used in early work was unsatisfactory since it predicted an unreasonable contraction in the ab plane and a modified forcefield was subsequently employed [224, 225]. Calculations using the latter forcefield have suggested that carbonate anions lie midway between, and generally coplanar with, the hydroxide layers [224]. Calculations using the same forcefield suggested that in the case of (S)-phenylalanine intercalated in Mg/Al LDH, the guest molecules adopted an interdigitated bilayer arrangement with their long axis approximately perpendicular to the layers [225]. The phenyl rings form a hydrophobic region in the middle of the interlayer with an edge to face orientation being preferred, possibly indicative of weak but directional C – H· · · · π bonding (Fig. 26). The oxygen atoms of water molecules together with those of the carboxylate groups form an oxygen monolayer on the surface of the hydroxide layers, giving hydrogen bonding interactions with an H· · ··O distance of ≈ 0.22 nm and variable O-H· · ··O angle (Fig. 27). In LDHs, the layer hydroxyl groups can act as hydrogen bond donors to both the interlayer anions and the oxygen atoms of interlayer water molecules. The interlayer water molecules can also form hydrogen bonds between themselves and can act as hydrogen bond donors to the interlayer anions. A molecular dynamics study of an Mg3 Al – Cl LDH has been reported by Wang et al. [223], and indicated that the interlayer species are distributed between two sublayers such that none of them is able to form direct hydrogen

Structural Aspects of Layered Double Hydroxides

49

Fig. 26 Molecular dynamics simulation cell of (S)-phenylalanine intercalated in an Mg/Al LDH. Reprinted with permission from [225]. Copyright American Chemical Society

bonds to both neighboring octahedral sheets simultaneously (Fig. 28). Each Cl– ion accepts ≈ 2.5 hydrogen bonds from layer hydroxyl groups but ≈ 4 hydrogen bonds from neighboring water molecules. The average structural environment of the Cl– anions was said to be strikingly similar to that of the ion in bulk aqueous solution. The average local structural environment of the interlayer water molecules involves ≈ 3.8 hydrogen bonds to layer hydroxyl groups, Cl– and other water molecules, which is quite different from that in bulk liquid water and more similar to that in ice (which have ≈ 3.2 and 4 hydrogen bonds per molecule, respectively). A good agreement was observed between simulated far IR spectra using the above structural model and the experimental spectra [223]. For [LiAl2 (OH)6 ]Cl·H2 O, molecular dynamics simulations suggest that the both interlayer water molecules and chloride anions are located in the middle

50

D.G. Evans · R.C.T. Slade

Fig. 27 Possible hydrogen bonding interactions for (S)-phenylalanine intercalated in an Mg/Al LDH. Reprinted with permission from [225]. Copyright American Chemical Society

Fig. 28 Results of a molecular dynamics study of an Mg3 Al – Cl LDH: the structure viewed along the (010) direction is shown on the left and the arrangement of interlayer Cl– and H2 O molecules viewed along the (001) direction is shown on the right. Dashed lines represent hydrogen bonding. Reprinted with permission from [223]. Copyright Mineralogical Society of America

of the interlayer [226]. The oxygen atoms of water molecules act as hydrogen bond acceptors with one hydroxyl group from the layer above and one from the layer below. The plane of the water molecules is parallel to the layers, such that its hydrogen atoms can form hydrogen bonds to separate chloride anions, giving a distorted tetrahedral environment about the oxygen atom. A minority of the chloride ions are in one of two types of trigonal prismatic sites coordinated to three hydroxyl groups from the layer above and three from the layer below (one type is directly above lithium cations, the other between vacant tetrahedral sites in the layers), whilst the majority are in distorted octahedral sites, forming six hydrogen bonds to two hydroxyl groups from the upper layer, two from

Structural Aspects of Layered Double Hydroxides

51

the lower layer and two hydrogen atoms of different water molecules. Separate signals can be observed in the 35 Cl NMR spectrum at room temperature, which become nearly fully averaged at 70 ◦ C. In the anhydrous material, the chloride anions all occupy prismatic sites along three-fold Cl· · ··(OH)3 · · · ·Li· · ··(OH)3 axes, consistent with the observed crystal structure [16]. Molecular dynamics simulations of the structure of hydrocalumite, [Ca2 Al(OH)6 ]Cl·2H2 O, have been reported [227, 228]. In this material, unlike other LDHs, the calcium ions do not occupy sites in the center of the layers but half are shifted up along the c direction and half down. The calcium ions are seven-coordinate, being bonded to a water molecule in the interlayer towards which they are shifted in addition to the usual six hydroxyl groups. Thus the water molecules form two sublayers on either side of the center of the interlayer along the c axis. The simulations suggest the presence of a 10 coordinate chloride ion, with the water molecules undergoing rapid librations (hindered hopping rotations) about the Ca – O bond perpendicular to the layers (note that in this case, as a result of coordination to the calcium ion, the C2 axis of the water molecules is neither parallel to, or perpendicular to the layers but at an angle of ≈ 50◦ ). Each water molecule is hydrogen bonded to three pairs of chloride ions for 2/3 of the time, so at any instant is bonded to four chloride ions. With respect to the chloride ion, at any instant there are four hydrogen bonds to neighboring water molecules, but averaged over time there are six hydrogen bonds with 2/3 occupancy. Structural studies involving single crystal XRD [111] and synchrotron [112] and conventional [229] powder XRD are essentially in agreement with this model. Above ≈ 308 K, the structure adopts a rhombohedral 6R polytype (space group R3c) where the chloride is surrounded by a distorted trigonal prism of hydroxyl groups from two adjacent layers and six additional water molecules giving a slightly distorted icosahedral coordination shell [111]. The water molecules have orientational disorder with partial occupancy of 2/3, giving a net total of four hydrogen bonds to water molecules in addition to the six to layer hydroxyl groups [112]. On cooling, the compound transforms to a monoclinic two layer polytype (space group C2/c) with a shift of chloride anions along (010)h and water molecules along (210)h of the hexagonal cell as indicated by the arrows in Fig. 29. This results in an ordering of the hydrogen bonding, with each chloride forming four strong hydrogen bonds to water molecules rather than six weaker bonds as in the high temperature polytype. The phase transition has also been studied by variable temperature 27 Al MAS NMR spectroscopy [230]. It was suggested that the changes in 27 Al quadrupole coupling parameters confirm the key role played by hydrogen bonding in the structural changes which occur during the phase transition. A similar phase transition is observed in the case of the corresponding bromide and iodide analogues [111]. The transition temperature between high and low temperature polytypes decreases from 309 K (X = Cl) to 123 K (X = I) from which it was concluded that

52

D.G. Evans · R.C.T. Slade

Fig. 29 Interlayer structure of [Ca2 Al(OH)6 ]Cl·2H2 O: the high-temperature rhombohedral form (represented by a projection along the ch axis is shown on the left and the low-temperature monoclinic form (represented by a projection on the (am , bm ) plane is shown on the right. Reprinted with permission from [112]. Copyright Elsevier Science Ltd

the smaller chloride ion favored the low temperature polytype (four strong hydrogen bonds) over the high temperature polytype (six weaker hydrogen bonds) because of its stronger hydrogen bonding capability [111]. Similar phenomena have also been reported in the literature for related materials. For example, two different modifications of [Ca4 Al2 (OH)12 ]CO3 ·5H2 O have been prepared at different temperatures and both have been characterized by single crystal XRD [231, 232]. In the high temperature form [231], which crystallizes in the non-centrosymmetric space group P1, 3/4 of the calcium ions in each layer are coordinated to water molecules and the remaining 1/4 to carbonate anions (in addition to six layer hydroxyl groups). In the low temperature form [232], which crystallizes with pseudohexagonal symmetry (centrosymmetric space group P1), the calcium ions in alternate sheets are all coordinated to water molecules, whereas those in the remaining sheets are coordinated to carbonate or water molecules with 50% probability. This system has not yet been studied by molecular mechanics, however. Molecular dynamics simulations have been carried out on [Mg2 Al(OH)]Cl ·n H2 O [233, 234]. The results suggest that the material has a similar structure to hydrocalumite with each magnesium being displaced along the c axis from the center of the layers and coordinated to a water molecule or chloride anion, in addition to six hydroxyl groups. The studies indicate that a minimum in the hydration energy is observed for n = 2, in which a well-developed, although disordered, hydrogen bonding network similar to that in hydrocalumite is formed with each available site in the O-type interlayer being occupied by either an anion or water molecule. In contrast, structural studies of magnesium-containing LDHs have always suggested that the Mg2+ ion is six-coordinate, however.

Structural Aspects of Layered Double Hydroxides

53

The structure of anhydrous intercalates of the type [LiAl2 (OH)6 ]X (X = Cl, Br, NO3 ) has been investigated using energy minimization methods [235]. In the case of the halide ions, the optimized structures were found to adopt a hexagonal 2H1 polytype with eclipsed (P type) metal hydroxide layers with the halide anions forming Li· · ··X· · ··Li· · ··X chains through the lattice perpendicular to the layers. In the case of X = NO3 , the nitrogen atom was found to occupy the same site as the halide ions with the oxygen atoms disordered over six sites, leading to two nitrate ions related by a 180◦ rotation in the ab plane existing in a disordered structure. The calculations are in good agreement with the experimental structures determined by synchrotron X-ray and neutron powder diffraction [16]. First principles molecular dynamics calculations have also been used to investigate the structure of Mg/Al LDHs. It was concluded that the 2H1 polytype with P type interlayers is more stable than the 1H polytype with O type interlayers and that the aluminum and chloride ions are aligned vertically along the c axis [43]. In this case, computing limitations precluded calculations on a threelayer polytype (e.g. 3R1 with P type interlayers and a staggered arrangement of cations along the c axis) or the introduction of any interlayer water molecules. The conclusions from these calculations are not consistent with the results from Rietveld analyses of Zn/Al or FeII 2 FeIII LDHs with interlayer chloride anions [25, 106, 155] in which the chloride and water molecules occupy P type interlayers and are located in sets of sites distributed in groups of six around the three-fold O· · ··O axis or exhibit strong oscillations around an equilibrium position along the O· · ··O axis (see Sect. 3.2.2). It should be noted, however, that as discussed above, the structure of the interlayer in [LiAl2 (OH)6 ]Cl·nH2 O is different for the anhydrous material (n = 0) and the hydrated material (n = 1), so any simulations which exclude water may lead to significant differences between experimental and optimized calculated structures. There has been an attempt [236] to optimize the arrangement of IO3 – ions intercalated in an Mg3 Al LDHs using the CASTEP (Cambridge serial total energy package) code, which is allows density functional theory calculations for crystal structures. The calculations suggested that the oxygen atoms of the IO3 – anion were preferentially hydrogen bonded to Mg2 Al(OH) groups rather than Mg3 (OH) groups. The calculations were carried out on anhydrous systems, however, whereas the experimental systems are hydrated [237] and only a very limited section of the lattice was included in the calculations. 3.2.7 Interlayer Galleries and Hydration As we have seen, the interlayer galleries in LDHs can be considered to be made up of hexagonal close-packed sites parallel to the close-packed layers of hydroxyl groups and metal cations, with one crystallographic site per metal [17, 238]. The maximum amount of water which can be accommo-

54

D.G. Evans · R.C.T. Slade

dated in the interlayer galleries of an LDH with [MII 1–x MIII x (OH)2 ]x+ layers is therefore given by (1 – Nx/n) where N is the number of sites occupied by an anion of charge n [143]. The three oxygen atoms of a carbonate group occupy three sites and the remainder can be occupied by water or left vacant. The general formula for an LDH-carbonate can be expressed as [MII 1–x MIII x (OH)2 ](CO3 )x/2 · (H2 O)1–3x/2–∆ where ∆ represents the number of vacant interlayer sites [7]. Thus in the mineral hydrotalcite, there is one vacant layer site per [Mg6 Al2 (OH)16 ]CO3 ·4H2 O formula unit. In synthetic LDHs, the experimentally determined amount of water can exceed (1 – Nx/n) in which case it is assumed that there is both interlayer or intrinsic water [95] and extrinsic water adsorbed on external non-gallery surfaces [239]. The maximum amount of intrinsic water that can be co-intercalated with other, larger, anions can be estimated in an analogous manner. As discussed in Sect. 3.2.1, the intensities of the (00l) reflections of a paramolybdate-intercalated LDH can be simulated using a model [143] involving three close-packed layers of interlayer oxygen atoms from the polyoxometallate and water molecules. A comprehensive study by Hou et al. [240] has shown that LDHs exhibit a diverse range of swelling and water sorption behavior, and can be divided into three categories. Type 1 have significantly expandable basal spacings (0.15–0.30 nm); Type 2 are slightly expandable (< 0.05 nm) and have significant interlayer water exchange; Type 3 are essentially non-expandable with little interlayer water exchange. For Type 1 (e.g. Mg3 Al – SO4 ), the fully expanded phases have a two-water layer structure, and the phase transition from one layer to two layers as determined by XRD correlates with both a significant step in the water sorption isotherm and changes in the interlayer structure and dynamics as observed by NMR. Type 2 (e.g. Mg3 Al – ClO4 ) only form one-water layer structures and the interlayer anion may undergo dynamical disordering at high relative humidities. For Type 3 phases (e.g. Mg3 Al – Cl) there is little interlayer water sorption because their interlayers are essentially closed, due to the small size or planar shape of the anions. The arrangement of anions in green rust materials of the type [FeII 1–x FeIII x (OH)2 ]Clx ·yH2 O has been explored in detail [241]. According to a Rietveld refinement, the layers adopt the 3R1 polytype and the stacking sequence can be written in the form AbC(C)CaB(B)BcA(A)AbC, where the letters in italics denote the interlayer species. Within a single layer, the O· · ··O distance in the hexagonal pavement is equal to the unit cell parameter ao (0.32 nm), whereas the diameter of a chloride anion is 0.36 nm. Thus in the hexagonal pavement of the interlayer, the six nearest-neighbor sites to a chloride ion cannot be occupied by another chloride anion (although they can be filled at random by water molecules). If this exclusion criterion is met, the maximum √ value of x is 1/3, with each chloride having six nearest neighbors at ao 3. For x ≤ 1/3 some of these next nearest neighbor

Structural Aspects of Layered Double Hydroxides

55

sites are vacant or may be filled by water molecules. The amount of water is y ≤ 1 – x. A random array of chloride anions that respects the exclusion rule is shown in Fig. 30a. There exists a correlation between the distribution of anions from one interlayer to the next (see Fig. 30b and c). Each anion shares its charge with one FeIII of the layer underneath and one of the layer above. As shown by the stacking sequence AbC(C)CaB(B)BcA, for the chloride in position B, the adjacent FeIII ions in the layer above occupy one of three a positions whilst the interlayer chloride ions in the interlayer above occupy one of six C positions. Three of these involve cis-arrangements of O – H· · ··Cl about the metal and other three have trans-O – H· · ··Cl with the former being more favorable [241]. Mössbauer spectroscopy confirms that only one type of FeIII site is present. There are two types of FeII site, however, as shown in Fig. 30d: one type denoted FeII (Cl) is coordinated to the two O – H· · ··Cl units in an analogous manner to FeIII , whilst the other, denoted FeII (H2 O) is coordinated to an O – H· · ··OH2 or O – H· · ··[vacancy] moiety in one interlayer. The FeIII : FeII (Cl) : FeII (H2 O) ratio is x : x : 1 – 2x, consistent with the results from Mössbauer spectroscopy [241]. A similar an-

Fig. 30 (001) projections of the pavements of anions in green rust materials of the type [FeII 1–x FeIII x (OH)2 ]Clx ·yH2 O: a a first interlayer of Cl– ions; b the adjacent Fe cation layer; c the next interlayer of Cl– ions; d the superimposition of (a), (b) and (c). Reprinted with permission from [241]. Copyright Elsevier SAS

56

D.G. Evans · R.C.T. Slade

alysis of the arrangement of anions and cations in FeII /FeIII green rust type LDHs containing interlayer carbonate and sulfate ions has been carried out and the results used in aiding the interpretation of their Mössbauer spectra [241]. There have been many studies of the nature of the interlayer water in LDHs using a variety of methods. The structure of bulk liquid water is dominated by distorted local tetrahedral arrangements similar to those in ice Ih , although the long range order in the latter is lost in liquid water and on average there are fewer than four nearest neighbor water molecules. Molecular dynamics calculations [242] have shown that water molecules confined in brucite galleries show significant deviations from bulk water. In comparison with brucite, the presence of interlayer anions in the galleries of LDHs introduces an additional level of complexity [3]. Jones et al. [243] have shown that in Mg/Al – CO3 LDHs, interlayer water is more strongly bound than physisorbed water by some 12 kJmol–1 . Quasi-elastic neutron scattering measurements [244, 245] on Mg/Al LDHs containing either interlayer carbonate anions or terephthalate anions have demonstrated that the water molecules are not fixed in one position, but rather exhibit translational diffusion as well as reorientational motions. There is rapid breaking and forming of hydrogen bonds as water molecules hop from one site to another. The diffusion coefficients of the interlayer water are lower than that for bulk water, with the value being higher for the terephthalate intercalate than for the carbonate analogue [244]. Thermodynamic measurements on Mg/Al – CO3 LDHs and mixtures of Mg(OH)2 , Al(OH)3 , MgCO3 and water or ice have indicated [27] that the entropy of formation of the LDH is negative when water is used but positive in the case of ice, suggesting that interlayer water in LDHs has a higher entropy than ice, but lower than that in bulk water; this is consistent with the above spectroscopic data. Molecular dynamics simulations [183] have suggested that the diffusion coefficient increases with increasing MII /MIII ratio in the layers, which can be interpreted in terms of decreasing anion content and charge density on the layers. Comparison of the dielectric properties of mica-type silicates and LDHs has indicated [246] that the latter show anomalous low frequency dispersion, which is associated with charge carriers of low mobility. This suggests that the water molecules in silicate clays have a greater variety of available reorientational motions than those in LDHs. Impedance measurements have also been used to calculate the spatial disorder of the host layers, expressed by the fractal dimension ds . This decreases from 2.7 to ≈ 2 on replacing nitrate ions with polyoxometallate anions, with the latter value indicating a molecularly smooth surface, suggesting that the stacking disorder decreases with increasing charge on the interlayer anions [246]. Far IR spectroscopy and molecular dynamics calculations have demonstrated the structural and dynamic similarity of chloride anions in the interlayer galleries of LDHs to those in bulk aqueous solution [223], as discussed

Structural Aspects of Layered Double Hydroxides

57

in Sect. 3.2.6. Molecular mechanics calculations have also suggested that the hydration energy of LDHs involves two components: one is controlled primarily by the formation of a hydrogen bonding network in the interlayer and the other is related to the decreasing electrostatic attraction between layers and interlayers as the structure expands along the c axis [233]. As noted in Sect. 3.2.6, for [Mg2 Al(OH)6 ]Cl·yH2 O, the most stable interlayer arrangement is for y = 2, in which all the interlayer sites are occupied giving a well-developed, although disordered, hydrogen bonding network. The key role of hydrogen bonding in the interlayer of an LDH containing tert-butoxide anions has been investigated by means of ab initio plane-wave density functional theory calculations [247]. It has been suggested that the active catalyst is in fact a hydroxide charge-balanced LDH with co-intercalation of neutral tert-butyl alcohol molecules giving a hydrophilic region adjacent to the layers and a hydrophobic region in the center of the interlayer galleries. This results in organic substrates with polar functional groups being intercalated in a particular orientation, facilitating reactions and promoting catalytic behavior. The hydration dynamics of a series of LDHs of the type [Zn0.61 Al0.39 (OH)2 ] (CO3 )0.195(1–y) Cl0.39y ·nH2 O [0 ≤ y ≤ 1, 0 ≤ n ≤ (0.4 + 0.2y)] have been studied by thermogravimetric analysis and XRD [95, 248]. For the fully hydrated samples, the basal spacing difference (0.019 nm) between end member compounds (y = 0, 1) is less than the height difference between Cl– and CO3 2– (0.053 nm). Furthermore, in the dehydrated materials the basal spacing of the LDH-chloride (y = 1) is smaller than that of the LDH-carbonate, i.e. dehydration of the former species causes a collapse in basal spacing although the amount of the (larger) ion remains constant. It was proposed that the chloride ion is able to nest in the trigonal pockets formed by the hydroxyl groups of the layers, but that this is not possible for carbonate. It was also suggested that in the fully hydrated LDH-chloride, the chloride anion is surrounded hexagonally by six water molecules, which must tilt about their C2 axes in order to be accommodated and leading to an increased interlayer spacing. Removal of a small amount of water gives a five member hydration shell around the chloride which can be accommodated without any tilting of the water molecules and is consistent with the almost immediate decrease in basal spacing from 0.774 to 0.760 nm when a sample of the fully hydrated material is purged with dry nitrogen. Extensive drying leads to a further decrease in basal spacing to 0.729 nm, associated with formation of the material containing nested chloride ions [95]. The vibrational modes of layered double hydroxides have been examined using a combination of IR and Raman spectroscopy and inelastic neutron scattering [249]. All three techniques highlight the existence of an extensive hydrogen bonding network between the lattice water and the layer hydroxyl groups, but show that the lattice hydroxyl modes are not significantly affected by altering the interlayer anion (carbonate, nitrate or hydroxide).

58

D.G. Evans · R.C.T. Slade

The dynamic disorder in the interlayers of LDHs is also confirmed by single crystal XRD, since the interlayer anions and water molecules generally have occupancy factors less than unity and elevated temperature/displacement factors [172]. 3.3 Long-range Cation Order-disorder The extent of long-range or short-range ordering of the cations within the layers of LDHs has been widely discussed in the literature and contradictory conclusions have been drawn. Although interesting in its own right as a theoretical problem, the arrangement of cations is also of practical significance since where LDHs are used as catalysts [250], or as precursors to catalysts [251, 252], ferrimagnetic spinels [253–255] or other materials it should be advantageous to have a homogeneous distribution of cations without segregation of “lakes” of separate cations. 3.3.1 Experimental Studies Magnetic susceptibility measurements [256] on CuII /CoII /Zn/Al LDHs indicate that the Curie–Weiss law is obeyed with a Weiss temperature close to zero, indicating the lack of magnetic interactions between paramagnetic ions and thus a random distribution in the lattice. It has been proposed, however, that the catalytic activity of Mgn Al LDHs (n = 2, 3) containing interlayer carbonate anions in the oxyethylation of 1-dodecanol with ethylene oxide can be correlated with the distance between Al3+ cations in ordered lattices [250]. In terms of local order about the cations, it is generally argued that according to Pauling’s rules [257], MIII cations should not occupy adjacent sites (unless accompanied by vacant cation sites, i.e. a gibbsite “unit” within a brucite layer [56]): the corollaries of this are that the minimum possible MII /MIII ratio is 2 and that when MII /MIII is exactly equal to 2 there is also a long range order with a superlattice involving an ordered array with each MIII cation having six MII nearest neighbors and each MII cation having three MII and three MIII nearest neighbors [40].√As shown in Fig. 31, the superlattice is characterized by a cell parameter a 3 and is rotated by 30◦ compared with the lattice √ √of non-ordered brucite-like layers with parameter a and is denoted ( 3 × 3)R30◦ . Some workers, notably Hofmeister and von Platen [92], have argued that most or all LDHs, including those with MII /MIII ratio greater than 2, also actually have a completely ordered distribution of cations but that this is not easily demonstrated experimentally because of high pseudosymmetry, micro-crystallinity and stacking faults. They argued that the prevalence of MII /MIII ratios close to integral values of 2 and 3 in natural and synthetic

Structural Aspects of Layered Double Hydroxides

59

√ √ Fig. 31 a Unit cell and lattice for non-ordered brucite-like layers and b ( 3 × 3)R30◦ suII III perlattice of ordered layers with M /M ratio of 2. Reprinted with permission from [92]. Copyright Gordon and Breach Science Publishers

LDHs indicate a preference for ordered layers and that if the layers, as well as the interlayers, were disordered, long range ordered stacking of the layers in two- or three-layer polytypes would be unlikely to occur. Two possible ordered arrangements of cations with MII /MIII ratios of 3 are shown in Fig. 32 [92, 206, 258]: one is based on a hexagonal supercell √ with a = 2ao and the other is based on an orthorhombic supercell with a = ao 3 and b = 2ao . Hofmeister and von Platen [92] simulated the XRD powder patterns for a variety of LDHs with different metal ions, stoichiometries and stacking

Fig. 32 Two possible ordered arrangements of cations with MII /MIII ratios √ of 3: a a hexagonal supercell with a = 2ao and b an orthorhombic supercell with a = ao 3 and b = 2ao . Reprinted with permission from [206]. Copyright American Chemical Society

60

D.G. Evans · R.C.T. Slade

sequences in both ordered and disordered forms and found that the calculated intensity of superlattice reflections was always very much weaker than those from the subcell with dimension ao , particularly when the scattering power of the MII and MIII cations were similar (e.g. Mg and Al). Even where the scattering powers are quite different, simulations of XRD patterns show that a slight degree of either defective cation sequences or non-stoichiometry lead to a marked loss of intensity of the superlattice reflections [100]. For example in the case of Mg2 Fe layers, the cation sequence along (210), (120) and (110) diagonal of the supercell of ordered cations is · · · Fe – Mg – Mg – Fe· · · (Fig. 33). Introduction of a small number of defective · · · Fe – Mg – Mg – Mg – Fe· · · and · · · Fe – Mg – Fe· · · sequences does not change the stoichiometry, but the superlattice reflections show a significant decrease in intensity. Note that even if the number of these defective sequences is large, there is still not complete disorder because no · · · Fe – Fe· · · or · · · Fe – Mg – Mg – Mg – Mg – Fe· · · sequences are present. Relatively small deviations in stoichiometry from Mg/Fe = 2 also lead to a dramatic decrease in the intensities of superlattice reflections. Rebours et al. have investigated the structure of Mg/Al and Mg/Ga LDHs with interlayer carbonate anions [39]. The former did not reveal any evi√ dence of cation ordering even for Mg2 Al, but a supercell of dimension a 3

Fig. 33 Mg2 Fe LDH in the 3R1 polytype with ordered cation sequence · · · Fe – Mg – Mg – Fe· · · within the layers but no correlation of cation position between layers. Light large circles represent Mg2+ , dark large circles represent Fe3+ ; small circles represent OH groups below (open circles) and above (filled circles) the metal ion plane. Reprinted with permission from [100]. Copyright Nova Science Publishers

Structural Aspects of Layered Double Hydroxides

61

was detected for the Mg2 Ga LDH as shown by the presence of a reasonably strong (100) reflection at 0.463 nm. The authors rationalized this by noting [39] that the brucite layers in LDHs and related structures are always compressed along the c axis and expanded along the plane as discussed in Sect. 2.1. The compression of the layer is limited by the close approach of oxygen atoms along shared octahedral edges – in brucite the O· · · O distance is 0.2785 nm whilst in gibbsite the O· · · O distance between shared edges of occupied and unoccupied octahedra is 0.2635 nm. In Mg/Al LDHs, the O· · · O distance was found to be close to the latter value and it was suggested that the structure cannot sustain the further shortening which would be associated with an ordering of cations. In contrast, since the ionic radii of Mg2+ and Ga3+ are similar, the ordering can take place without any significant distortions. An alternative explanation is that both Mg2 Al and Mg2 Ga systems are ordered, but the (100) reflection of the superlattice is not observed with the fomer because of the similar scattering powers of the isoelectronic Mg2+ and Al3+ ions. Refinement of the single crystal structure of the mineral shigaite discussed below has shown [259] that ordered Mn2 Al layers are accompanied by the presence of shared octahedral edges (0.2586–0.2610 nm) considerably shorter than those in gibbsite, suggesting that layer flattening is not a significant factor in determining whether or not cation ordering occurs. The presence of weak broad reflection (d = 0.457 nm) in the XRD pattern of an Mg2 Al LDH prepared by placing initial flocculated precipitates in cellulose dialysis bags immersed in water (changed daily) at 60 ◦ C has √ been attributed [260] to the (100) reflection of a superlattice with a = ao 3 = 0.528 nm. Although other values have also been claimed [38, 261], it has generally been asserted [25] that Znn Cr – Cl LDHs can only be prepared for the case of n = 2, strongly suggesting the existence of ordered layers. This has been rationalized [262] by proposing that the synthesis proceeds via direct condensation of two hexaaquozinc(II) complexes with a deprotonated monomeric chromium species to give a trimeric species which subsequently condenses to give the [Zn2 Cr(OH)6 ]+ layers. The scattering powers of zinc and chromium are similar and no superlattice reflections are observed in the XRD: it has been calculated, however, that any such reflections would have an intensity of only ≈ 0.1% of that of the most intense (003) reflection [25], so their absence cannot be taken as evidence of a lack of long-range order. Nickelaluminum LDHs have been studied using a high intensity (rotating anode) X-ray generator [263]; long range ordering was indicated by the presence of a weak broad reflection could be assigned to a superposition of three √ which √ reflections from a ( 3 × 3)R30◦ superlattice in the case of x = 0.33, but is lost at other compositions. In the case of a synthetic Mg2 Al – CO3 LDH, selected area electron diffraction (SAED) [264] patterns of a few √crystals with zone axis (00l) showed evidence of a superlattice with a = ao 3 = 0.532 nm [71], although this was

62

D.G. Evans · R.C.T. Slade

not the case for the majority of the sample. No superlattice reflections were observed in the XRD pattern and a Rietveld type analysis gave a unit cell with a = 0.30502(4) and c = 2.2925(3) nm. Partially ordered regions in pyroaurite and sjögrenite crystals have also been detected by SAED [92, 265]. Rietveld refinement of the XRD patterns of an Mg2 Al LDH with interlayer carbonate [266] and Zn2 Al LDHs containing interlayer chloride [106] or carbonate [267] anions also gave no indication of cation ordering. It has been suggested that powder XRD patterns of some mineral samples of LDH minerals show evidence of superlattice reflections [7, 113] but there is no clear consensus [100]. It should also be borne is mind that, as discussed in Sect. 3.5, superlattice reflections may be due to anion, rather than cation, ordering although the latter may be an indication of the former. In contrast to the situation for most LDHs, there is ample evidence from powder diffraction [115, 258] and synchroton X-ray and neutron powder diffraction [16] that those based on [LiAl2 (OH)6 ]+ units always have an√ordered array of cations with a hexagonal supercell of dimensions a = ao 3 = 0.532 nm in which vacancies in the dioctahedral gibbsite lattice are filled by lithium cations, which naturally gives rise to ordering [268]. 27 Al MAS NMR spectra of [LiAl2 (OH)6 ]Cl·nH2 O have one narrow symmetric peak indicating that the aluminum cations occupy highly ordered and symmetric positions [269], consistent with the refined X-ray and neutron diffraction structures. In contrast, gibbsite has two 27 Al resonances with different quadrupolar products, which arise from the two crystallographically distinct sites in gibbsite. It was proposed that introduction of Li+ into site vacancies in gibbsite leads to a more uniform distribution of positive charge within the octahedral sheet, thereby reducing the differences between individual Al – O bond lengths and O – Al – O angles [269]. In view of the difficulties in detecting any long-range cation ordering by powder XRD, it would be useful to have single crystal diffraction data. Unfortunately the microcrystalline nature of most synthetic LDHs means that suitable single crystals are not generally available. Even when they are available, there is always a possibility that they may not be representative of the bulk sample [232]. Single crystals of a variety of [Ca2 Al(OH)6 ]+ LDHs containing anions such as carbonate [231, 232], nitrate [171, 172] and chloride [229, 270] have been prepared starting from mixtures of powdered Ca(OH)2 , Al(OH)3 and the corresponding calcium salt of the anion and water in sealed silver capsules [231]. These materials show complete ordering of cations with each aluminum cation surrounded by six nearest neighbor calcium cations and no edge-sharing between AlO6 octahedra. In these cases, as discussed in Sect. 4.1, the structure differs from that of most LDHs in that the calcium ions are seven coordinate, being coordinated to an anion and/or oxygen atom of an interlayer water molecule in addition to six layer hydroxyl groups. It has generally been assumed in the literature that the fact that Ca/Al LDH-like materials are only formed with Ca/Al = 2 and have an ordered array of cations

Structural Aspects of Layered Double Hydroxides

63

is associated with the large difference in ionic radii between Ca2+ and Al3+ . However, it has recently been shown that replacement of Al3+ by larger Ga3+ , Fe3+ or Sc3+ ions does not lead to a loss of cation ordering. This has been interpreted as indicating that the cation ordering is associated with large size and the pronounced anisotropy of the coordination sphere of the Ca2+ ion, rather than the difference in cation radii [271]. There have been some studies of single crystals of minerals [4] that have indicated ordering of cations [7, 92, 100]. In the case of mineral samples, however, there are often considerable problems with intergrowth and twinning of crystals [92, 272]. The single crystal structure of shigaite, [MII 6 Al3 (OH)18 ][(SO4 )2 {Na(OH 6 }(H2 O)6 ] (M = Mn), shows an ordered √ 2 )√ array of layer cations in a ( 3 × 3)R30◦ array with a three layer rhombohedral polytype (space group R3) [259]. The interlayer is also ordered and contains an array of sulfate anions and cationic hexaaquosodium cations with a periodicity of 3ao ; there are two distinct Al sites with mean Al – O bond lengths of 0.1914 and 0.1912 nm together with a single MnII site (mean Mn – O = 0.2192 nm) (Fig. 34). The mineral nikischerite (M = Ni) has recently been shown [273] to be isostructural with shigaite, and motukoreaite, [Mg5.6 Al3.4 (OH)18 ](SO4 )1.3 (CO3 )Na0.6 (H2 O)12 also has essentially the same structure [274]. A single crystal refinement of the mineral zaccagnite, [ZnII 4 Al2 (OH)12 ] CO3 ·3H2 O, has been reported [275]. The material has a hexagonal two-layer

Fig. 34 Interlayer species in shigaite, [Mn6 Al3 (OH)18 ][(SO4 )2 {Na(OH2 )6 }(H2 O)6 ], projected down (001): {Na(OH2 )6 } octahedra are cross-shaded, SO4 tetrahedra are randomdot shaded, oxygen atoms of water are shown as black circles, hydrogen atoms of water molecules as small open circles and dotted lines indicate hydrogen bonds. Reprinted with permission from [259]. Copyright Mineralogical Association of Canada

64

D.G. Evans · R.C.T. Slade

2H1 polytype stacking (space group P63 /mmc) and in addition to sharp reflections corresponding to cell parameters ao = 0.307 and co = 1.511 nm, √ diffuse streaks parallel to c∗ at positions defining a supercell with a = ao 3 are observed. These can be indexed as (100) (0.462 nm), (200) (0.231 nm) and (210) (0.175 nm) (where the indices are referred to the supercell). The diffuseness of these reflections suggests that there is no correlation between the ordering in subsequent layers. Cation ordering has also been suggested [276] by the presence of superlattice reflections in single crystal X-ray photographs of other minerals of the type [MII 4 Al2 (OH)12 ]CO3 ·3H2 O known as charmarite (M = Mn), quintinite (M = Mg) and caresite (M = Fe). For charmarite and quintinite, both hexagonal two-layer and trigonal three-layer polytypes have been observed, whereas caresite has only been observed in the√trigonal √ form. The reflections were indexed in a superlattice of the type (2 3 × 2 3)R30◦ . It was subsequently shown [277] by means of a single crystal X-ray study on a mineral from the same source with the composition [Mg4 Al2 (OH)12 ]CO3 ·3H2 O that it had a two-layer hexagonal polytype √ (space group P62m) with an ordered cation distribution giving rise to a ( 3 × √ √ 3)R30◦ superlattice with a = ao 3 = 0.5283 nm and c = 1.5150 nm. In contrast to these results for LDHs with MII /MIII = 2, determination of the single crystal structure of LDH minerals of the type [MII 6 MIII 2 (OH)16 ] CO3 ·nH2 O such as pyroaurite [MII = Mg, MIII = Fe, n = 4.5, 3R1 polytype], sjögrenite [MII = Mg, MIII = Fe, n = 4.5, 2H1 polytype] and desautelite [MII = Mg, MIII = Mn, n = 4.0, 3R1 polytype] show no evidence of cation ordering [100]. 3.3.2 Theoretical Studies There have been a small number of theoretical studies of cation ordering in LDHs. First principles molecular dynamics calculations [43] on [Mg3 Al(OH)8 ]Cl LDHs discussed in Sect. 3.2.6 suggested that structures with adjacent aluminum cations were energetically less favorable than one without, although the chosen arrangement for the latter lacked either hexagonal or rhombohedral supercell. Calculations on a two-dimensional discrete Coulomb alloy A1–x Bx , in which A and B have different charges and are free to move on the sites of a triangular lattice in order to achieve the lowest energy configuration have been reported [278]. It was suggested that this is a valid model of the Coulombic interactions between cations within the sheets of LDHs, since the screening of interactions between layers is sufficiently large. At certain values of the composition x it is possible to form regular superstructures. The most important class of these is when one set of ions, say B, lies on a triangular √ network where the separation between the nearest neighbor B ions is ao k, where k = (n2 + nm + m2 ) and n, m are non-negative integers (i.e. k = 3, 4, 7, 9, 12, etc.). This gives rise to a set of compositions with x = 1/k, of which an ex-

Structural Aspects of Layered Double Hydroxides

65

Fig. 35 Computer simulation of a regular triangular superlattice of a Coulomb alloy A1–x Bx with x = 1/3 as described in the text. Reprinted with permission from [278]. Copyright American Physical Society

ample with k = 3 is shown in Fig. 35. For some values of k, it was found that other superlattice structures may have an energy very close to the minimum energy of triangular lattice superstructures. For example a piece of triangular superlattice with k = 4 (the kagomé structure [279]) is shown in Fig. 36 together with a piece of rectangular lattice with the domain boundaries marked (note that these two lattices have the unit cells shown in Fig. 32). For other values of x < 1/3, the B atoms form a triangular

Fig. 36 Computer simulation of a regular triangular superlattice of a Coulomb alloy A1–x Bx with x = 1/4. A piece of triangular superlattice is shown together with a piece of rectangular lattice where the domain boundaries are marked. Reprinted with permission from [278]. Copyright American Physical Society

66

D.G. Evans · R.C.T. Slade

lattice at the commensurate points and otherwise a pseudotriangular lattice. For 1/3 < x < 1/2, it was suggested that the lowest energy structures consist of chains of finite length arranged in a rectangular array. Commensurate configurations (those having rational ratios of subcell and supercell periodicities in the ab plane [263]) of this type are only possible for x = n/2(n + 1) where n is any integer. For x = 1/2 it was found that the lowest energy structure has infinite straight chains of nearest neighbor ions. These results are consistent with those √ given in Sect. 3.3.1, where ordering of cations in a superlattice with a = ao 3 is often observed for LDHs with MII /MIII = 2 (i.e. x = 1/3). The presence of mixed domains of triangular and rectangular structure is a possible explanation for the lack of superlattice reflections generally observed in LDHs with MII /MIII = 3 (i.e. x = 1/4). 3.4 Short-range Cation Order Although the evidence for and against long-range cation ordering in LDHs is somewhat inconclusive, there is much stronger evidence for local ordering of cations. EXAFS is a powerful tool for studying local cation ordering. Analysis of the fine structure beyond the edge in XAS depends on the interference of the backscattered electron waves with the ejected photoelectron and as a consequence of the short mean free path of electrons in a solid, information can only be obtained to a distance of several tenths of nm from the absorbing atom. In the case of Mgn FeIII – CO3 LDHs with nominal values of n = 2, 3 and 4, it has been shown that FeIII cations never (within the detection limits of EXAFS) occupy neighboring sites [40]. Powder XRD studies of the same samples showed no evidence of long-range cation ordering. It was noted that for Mg/FeIII = 2 these two results can be consistent only if there are reasonably frequent defects to the ideal cation pattern, probably roughly every fifth to tenth cation in any direction – an example of one type of defect has been discussed in Sect. 3.3.1. This would result in coherent superlattice domains never exceeding a few nm and remaining X-ray silent, while only a few percent of sites would be affected and the average environment around FeIII cations would remain effectively unchanged [40]. Other defects may involve cation vacancies or Mg2+ cations occupying sites that would be ideally occupied by FeIII : indeed in the sample with a nominal Mg/FeIII ratio of 2, the actual value is 2.17 indicating an excess of magnesium ions. Ordering in Zn2 Cr systems with interlayer terephthalate anions has been reported on the basis of EXAFS measurements [280] with the Cr having 6 O at 0.198 nm and 6 Zn at 0.311 nm and the Zn having 6 O at 0.206–0.207 nm, 3 Zn at 0.309 nm and 3 Cr at 0.311 nm and 6 Zn at 0.536–0.537 nm. The local description of the arrangement around the cations in Zn2 Cr and Cu2 Cr LDHs with interlayer chloride anions has been studied by EXAFS at the cation K edges [25, 281]. The EXAFS Fourier transform in each case can be modeled

Structural Aspects of Layered Double Hydroxides

67

with a six shell model in which the first peak P1 arises from the oxygen atoms coordinated to the peaks P2 to P6 are mainly caused by metallic √ metal whilst √ neighbors at a, a 3, 2a, a 7 and 3a as shown in Fig. 37 [281]. The intensities of peaks P4 and P6 are characteristically enhanced by the so-called focusing and superfocusing effects, associated with multiple scattering phenomena from the large number of three and four atoms in a collinear arrangement with paths twice or three times that of the first metal-metal shell respectively [282]. Although the difference in scattering power is relatively small for Cu/Cr and Zn/Cr pairs, the marked difference in CuO6 and CrO6 coordination polyhedra associated with the Jahn-Teller distortion in the former allows it to be demonstrated unequivocally that the cations in the Cu2 Cr LDH are in an ordered distribution and that as a result the sheets show a pronounced corrugation, with two Cr – Cu distances observed for P2 at the Cr edge and one Cu – Cu distance for P2 at the Cu edge [281]. Strong evidence for local cation ordering in the Zn2 Cr LDH was also obtained [25, 262], but in this case the sheets preserve a relative flatness. It was noted that in this case, as in others [283], the interlamellar species do not make any contribution to the Fourier transform and it was suggested that this is possibly due to their thermal motion. EXAFS studies at the Zn K edge of Zn2 Al – Cl LDHs have also demonstrated the ordered nature of the cation distribution [77]. The ordering is demonstrated by the presence of P3 and P4 peaks in the pseudo-radial distribution function, arising from multiple scattering by metal cation shells at √ a 3 and 2a. Replacement of interlayer chloride ions by dodecylsulfate did not affect the intralayer cation arrangement. In contrast, however, delamination of the material followed reformation of the layered structure by evaporation of the solvent suggested that although the first (6 O) and second (3 Zn + 3 Al) coordination shells around Zn were retained, the intensity of the P3 and P4 bands was greatly reduced which suggests a loss of longer range order. Consistent with this interpretation, IR bands assigned to metal-oxygen lattice vibrations were broadened after the delamination-evaporation process. EXAFS at the Co and Fe K edges has demonstrated [72] that the cations in LDHs of the type [Co2 (Fey Al1–y )(OH)6 ]Cl·nH2 O (0 ≤ y ≤ 1) are ordered. In particular, in an ordered CoII 2 MIII lattice of the type shown in Fig. 37 each coordination shells with 3 CoII and 3MIII ions at a, CoII cation √should have II II 6Co at a 3 and 3Co and 3MIII at 2a.√ The data indicate that the contribution at the Co K edge from the shell at a 3 does not vary when aluminum is replaced by iron, but the contribution from the shell at 2a does vary with y over the range from 0 to 1, which is consistent with the ordered arrangement of cations shown in Fig. 37. In related species of the type [(Co1–y Cuy )2 Al(OH)6 ]Cl·nH2 O (0 ≤ y ≤ 1) it was shown [284] by EXAFS that although the local order of MII and Al cations was maintained as y increased, the layers consisted of mixtures of domains of {Co2 Al} and {Cu2 Al}. The distortion in CuII coordination polyhedra associ-

68

D.G. Evans · R.C.T. Slade

Fig. Cu2 Cr layer showing shells P2 to P6 containing metallic neighbors at a, √ 37 Ordered √ a 3, 2a, a 7 and 3a. Reprinted with permission from [281]. Copyright Elsevier BV

ated with the Jahn-Teller effect results in a non-linear arrangement of metal ions as shown by the disappearance of the focusing effect peaks in the case of the {Cu2 Al} domains. A Zn /Al LDH with Zn/Al = 1.45 has been analyzed by Zn K edge EXAFS at 40 K [50]. The data were fitted to three Zn· · · O and four Zn· · · M (M = Zn, Al) shells and were also found to be consistent with an ordered distribution of cations. In the case of an LDH with Zn/Al ≈ 1.5, the cation avoidance rule can no longer be sustained and it was assumed that the trivalent cations were distributed as far apart as possible. EXAFS data at the Co and Al K edges for Con Al – Cl LDHs were consistent with an ordered array of cations for n = 2, 3 and 4. In each case the Al has a second coordination shell of ≈ 6 CoII ions and the CoII has a second shell of ≈ 6/n Al and (6–6/n) CoII next nearest neighbors [46]. The peaks arising from the focusing effect are still observed in the Co K edge EXAFS of the material showing that the cobalt cations in the layers remain aligned. The presence of local cation ordering in Mg2 Ga and Mg5 Ga – CO3 LDHs noted in Sect. 3.3.1 has been confirmed by means of both EXAFS and by calculation of the electron radial distribution function from the Fourier transform of the diffracted X-ray intensity. In each case the gallium was found to have six magnesium ions and no gallium ions as next-nearest neighbors [39].

Structural Aspects of Layered Double Hydroxides

69

IR and Raman spectroscopy can give some insight into local cation order [77] and there have been a few other attempts to probe this aspect of the structure of LDHs using these techniques [285]. Each hydroxyl group is coordinated to three cations (and may also be hydrogen bonded to interlayer anions or water) and the number of distinct types of M3 (OH) units will depend on the stoichometry of the LDH and the arrangement of cations in the lattice. For example, given hexagonally ordered lattices, for 2 : 1 LDHs MII 2 MIII (OH) units should be present, whereas for 3 : 1 LDHs both MII 2 MIII (OH) and MII 3 (OH) units should be present with no MII MIII 2 (OH) MIII 3 (OH) units in either case. It has been suggested [286] that the presence of five sharp lattice vibrations in the region 800–250 cm–1 of the IR spectra of Mg2 Al, Ni2 Al and Mg2 Fe LDHs containing interlayer carbonate anions is consistent with an ordered array of MII 2 Al(OH) units, whereas the presence of five broader bands in the spectra of the analogous MII 3 Al LDHs is indicative of a disordered array of cations. Deuteration studies later suggested the presence of six sharp lattice bands in the IR spectrum of Ni2.5 Al analogue [287]. It should be noted that even for simpler species such as β-Ni(OH)2 , the interpretation of the vibrational spectrum has been the subject of controversy in the literature [288]. It has been suggested, however, [173] that in the Raman spectrum of an Mg2.3 Al LDH with a mixture of interlayer carbonate and nitrate ions, separate hydroxyl stretching bands can be assigned to Mg3 (OH), Mg2 Al(OH) and MgAl2 (OH) units; separate hydroxyl stretching bands have also been assigned to Mg3 (OH), Mg2 Cr(OH) and Cr3 (OH) units in the Raman spectrum of stichtite ([Mg6 Cr2 (OH)16 ]CO3 ·4H2 O) [289] and to Mg3 OH, Zn3 OH and Al3 OH moieties in the hydroxyl stretching region of the Raman spectrum of [Mgx Zn6–x Al2 (OH)16 ](CO3 )· 4H2 O (x = 0–6) [290]. On the basis of the cation avoidance rule, however, some of these “units” would not be expected. Ab initio quantum mechanical modeling of the variation in vibrational frequencies of the OH group with the charge density and mass of the cations in dimers of the type [(H2 O)2 (OH)2 M(µ-OH)2 M
(OH)2 (OH2 )]n– (M, M
= Mg, Al, FeII , FeIII ) have been used in an attempt to understand the vibrational spectra of phyllosilicate clays [78, 291, 292] and these studies may also be useful in analyzing IR data for LDHs. Raman studies on [Cux Zn6–x Al2 (OH)16 ]CO3 ·4H2 O have suggested that for x < 2, the cations are arranged in a regular array, but for larger values of x, separate OH stretching bands associated with “CuOH”, “ZnOH” and “AlOH” were identified suggesting that the cations separate into “lakes” or “islands” of like cations [165]. As noted above, segregation of {Co2 Al} and {Cu2 Al} domains in other LDHs has been observed by EXAFS and attributed to the Jahn-Teller distortion associated with CuII , but segregation of Cu-only domains is inconsistent with the EXAFS results in this case [284]. Mössbauer spectroscopy is a potentially very useful way to study the cation ordering in LDHs containing iron [293–295], although care must be taken in order to avoid misinterpreting the results as has often happened

70

D.G. Evans · R.C.T. Slade

in publications in clay and soil science [296]. In pyroaurite type materials [Mg1–x FeIII x (OH)2 ](CO3 )x/2 ·nH2 O, a single FeIII site is generally found to be present when x ≈ 0.25 [238, 283]. (The quadrupole doublet is often asymmetric but this is usually ascribed to the effects of preferred orientation rather than the presence of a second doublet). At higher values of x, two quadrupole doublets can be observed, one corresponding to FeIII with 6 Mg as nextnearest neighbors and another involving one or more FeIII nearest neighbors (cluster-type arrangements). Results for so-called green rusts containing layers of the type [FeII 1–x FeIII x (OH)2 ]x+ indicate that they contain one type of FeIII but generally (except in the case of sulfate [157, 297]) contain two types of FeII site as indicated by the presence of two quadrupole doublets in a ratio which depend on the anion [238, 241, 283, 297], as discussed in Sect. 3.2.7. It has been suggested that the observed range of stoichiometries (x) and extent of cation ordering in these materials is connected with the ordering of anions [241]. Related materials of the type [FeII 1–x–y Mgy FeIII x (OH)2 ]x+ have been shown to have two FeII quadrupole doublets as for the green rusts, and also two FeIII quadrupole doublets [283]. The latter have been assigned to FeIII ions with all Fe nearest neighbors and with Fe/Mg nearest neighbors and their intensity varies with magnesium content consistent with this interpretation. The spectral linewidths are relatively small and similar to what is observed at the same temperature for green rusts without magnesium cations. This is consistent with an ordered arrangement of cations in which each FeIII cation is surrounded by six divalent cations. Mössbauer spectra of [FeII 4 FeIII 2–y Aly (OH)12 ](SO4 )·nH2 O show that there is only one FeIII and one FeII doublet in the unsubstituted material (y = 0) but that for y > 0, it is necessary to have two FeII doublets in order to fit the data [298]. This was interpreted in terms of an ordered array of cations in which Al progressively replaces FeIII in the next nearest neighbor shell at distance ao around FeII . UV-visible diffuse reflectance spectra of Zn/Cr – Cl LDHs have been reported [38, 261]. On the basis of the relatively narrow bands and the greater size of ligand field splitting compared with that in Cr(OH)3 , it was concluded that the CrIII O6 sites are surrounded by six Zn2+ ions as next nearest neighbors for Zn2 Cr and Zn3 Cr. The existence of Znx Cr – Cl LDHs with x < 2 was ascribed to the formation of dimeric Cr – O – Cr links. As noted in Sect. 2.2.1, such links have been demonstrated [41] in [Zn7 Cr4 (OH)22 ]CO3 ·5H2 O, which has a ligand field splitting less than that of Zn2 Cr – Cl. It has been suggested [299] that the presence of similar FeIII – O – FeIII dimers in coalingite [Mg10 Fe2 (OH)24 ]CO3 ·5H2 O can be demonstrated by UV-visible spectroscopy. 3.5 Anion Ordering In addition to long-range ordering of the intralayer cations, there also exists the possibility of anion ordering within the interlayer galleries. This may be

Structural Aspects of Layered Double Hydroxides

71

a manifestation of an ordered arrangement of cations and/or a reflection of the hydrogen bonding network in the interlayers. It was suggested [44] that a peak at 0.454 nm in the powder XRD of an Mg2 Al-benzoate√LDH resulted from the (100) reflection of a superlattice with dimensions ao 3. In view of the similar scattering powers of Mg2+ and Al3+ cations, the reflection was attributed to an ordered arrangement of benzoate anions associated with an ordering of Al3+ cations, rather than directly from an ordered superlattice within the layers. A similar conclusion was drawn regarding a reflection at 0.46 nm observed [300] in the XRD pattern of an oxalate-intercalated LDH with Mg/Al = 2.2. On the basis √ of reflections observed at 0.465 and 0.369 nm, a superlattice (with a = ao 3 = 0.5305 nm) associated with the ordering of silicate anions in the interlayers of a Zn3 Al LDH has been proposed [301]. This is possibly associated with the hexagonal symmetry of the condensed silicate anions, which matches that of the layers. In the case of a silicateintercalated Mg3 Al LDH, no superlattice reflections were observed in the XRD pattern but selected area electron diffraction demonstrated the presence of a superlattice with a ≈ 0.52 nm [302]. It should be noted, however, that there is some evidence that LDHs decompose in the intense electron beam in SAED experiments [300] and this should be borne in mind when interpreting such data. As discussed in Sect. 3.2.2 sulfate occurs in different types of interlayers in LDH minerals [99, 303]. When the basal spacing is ≈ 0.88 nm as in [Mg4 Al2 (OH)12 ]SO4 ·3H2 O, there is no anion ordering and a = ao . When the layer is expanded to ≈ 1.1 nm by inclusion of additional water and/or hydrated cations, the sulfate √ ions are generally ordered. The periodicity of the lattice may be ao 3 (as expected for the cations in a 2 : 1 lattice) as in [Mg3.96 (Al1.98 FeIII 0.06 )(OH)12 ][(SO4 )1.30 Na0.56 (H2 O)7.3 ] [99, 303] or 3ao as discussed in Sect. 3.3.1 for shigaite [259], [MnII 6 Al3 (OH)18 ][(SO4 )2 {Na(OH2 )6 }(H2 O)6 ]. The same phenomenon occurs in synthetic LDHs [304, 305]. For example, the powder XRD patterns shown in Fig. 38 for [Zn0.67 Cr0.33 (OH)2 ][(SO4 )0.165 (H2 O)0.73 ] and [Zn0.67 Cr0.33 (OH)2 ][(SO4 )0.22 Na0.11 (H2 O)1.25 ] may be indexed in two-layer√ polytypes according to cells with (ao = 0.3120, c = 1.782 nm) and (a = ao 3 = 0.5411, c = 2.2116 nm) respectively [305]. Note that in the case of the ordered cation distribution, the (01l) and (11l) indices determined for the disordered cation distribution are replaced by (11l) and (03l) respectively. The structure of a green rust LDH-type material with the formula [FeII 4 FeIII 2 (OH)12 ]SO4 ·ca. 8H2 O has been determined by Rietveld analysis [157]. The material exists as a one-layer polytype with the interlayers containing two planes of sulfate and water molecules giving a basal spacing of 1.1011 nm. The sulfate ions are oriented with their C3 axes perpendicular to the layers and alternate anions point up and down (as √ shown earlier in Fig. 14) and form a superlattice with parameter a = ao 3 = 0.5524 nm (Fig. 39).

72

D.G. Evans · R.C.T. Slade

Fig. 38 XRD patterns of [Zn0.67 Cr0.33 (OH)2 ][(SO4 )0.165 (H2 O)0.73 ] (top) and [Zn0.67 Cr0.33 (OH)2 ][(SO4 )0.22 Na0.11 (H2 O)1.25 ] (bottom) which may be indexed in √two-layer polytypes according to cells with (ao = 0.31204, c = 1.782 nm) and (a = ao 3 = 0.5411, c = 2.2116 nm) respectively. Reprinted with permission from [305]. Copyright Academic Press

Fig. 39 Projection on the (001) plane of the ordered interlayer structure of the green rust with stoichiometry [FeII 4 FeIII 2 (OH)12 ]SO4 ·ca.8H2 O. Reprinted with permission from [157]. Copyright Elsevier SAS

It was assumed that the sulfate oxygen atoms pointing towards the layers are oriented towards FeIII cations. This does not necessarily imply long range ordering of the cations, however, because only half of the FeIII sites are interacting with sulfate anions in this way and the other half of the sites could

Structural Aspects of Layered Double Hydroxides

73

be occupied by FeII cations (although this would also lead to the presence of adjacent FeIII sites which would violate the cation avoidance rule). In the Mössbauer spectrum, there are two quadrupole doublets, one characteristic of FeII and one of FeIII . It was suggested this is consistent with the ordered nature of the interlayers, since green rusts with disordered layers are characterized by two FeII signals [157]. LDHs of the type [Ni1–x Alx (OH)2 ](CO3 )x/2 ·yH2 O have been studied using a high intensity X-ray generator [263]. In addition to the ordering of cations noted in Sect. 3.3.1 for the case of x√= 0.33,√it was found that for x = 0.30, the carbonate anions form an ordered ( 13 × 13)R13.90◦ superlattice as shown by the presence of four weak reflections which can be indexed as (10), (11), (21) and (30) in the superlattice (the reflections only require (hk) labels) as shown in Fig. 40 [263]. A possible arrangement of the carbonate anions is shown in the inset to the Figure. The anions are each centered over a triad of hydroxyl groups and oriented so that the hydroxyl oxygen atoms are on lines joining carbonate centers. For x = 0.30 and ao = 0.3032 nm, the charge density of the layers can be calculated (as described in Sect. 3.2.4) to be 3.768 nm–2 , corresponding to a carbonate density of 1.88 ions nm–2 . This corresponds closely to the value calculated for the carbonate anion density in the supercell (1.93 nm–2 ). The superlattice reflections have an asymmetric Warren lineshape and simulations of the diffraction patterns indicated that this is a result of uncorrelated stacking of the anions in the c direction, even though they are well-ordered in the ab plane. This was attributed to the high (26-fold) degeneracy of possible carbonate sites [306]. The effect of ordering of layer cations and/or ordering of interlayer carbonate anions on the crystal chemistry of LDHs has been analyzed [277]. The authors reported the single crystal structure of a mineral with the

Fig. 40 Measured (top) and simulated (bottom) XRD patterns of [Ni0.7 Al0.3 (OH)2 ] √ √ (CO3 )0.15 ·yH2 O showing ordered ( 13 × 13)R13.90◦ superlattice of carbonate anions. [The data are given as a q plot, where q = 4π sin θ/λ]. Rhombohedral primitive cells of hydroxyl and carbonate ions are shown as solid and dashed lines respectively in the inset figure. Reprinted with permission from [263]. Copyright Elsevier Science Ltd

74

D.G. Evans · R.C.T. Slade

formula [Mg4 Al2 (OH)12 ]CO3 ·3H2 O which has an ordered distribution of layer cations. The stacking along the c axis is unprecedented in that alternate interlayers are composed of (i) carbonate anions with the three oxygen atoms displaced towards the carbon atom located on one of the three-fold axes and (ii) a net of water molecules whose oxygen atoms are arranged in a close-packed motif, i.e. the layer stacking is of the type – 1 – 2 – 1 – 3 – 1 – where layer 1 is the metal hydroxide, layer 2 the carbonate ions and layer 3 the √ water molecules. This gives a structure with space group P62m with a = ao 3 = 0.5283 nm and c = 1.5150 nm. The usual disordered interlayer structure in carbonate-containing LDHs can be thought of as a superposition of nets 2 and 3 giving a statistical distribution of carbonate ions and water molecules projected onto a unit cell with parameter ao . A highly crystalline Zn1.56 Al – Cl LDH has been prepared by displacing the carbonate anions in a precursor LDH synthesized using the urea method (see Chapter 2 by He et al. in this volume). The good crystallinity of the material combined with the use of a high intensity X-ray generator the √ allowed √ observation of three reflections (see Fig. 41) arising from a ( 3 × 3)R30◦ superlattice of chloride anions in the dehydrated form [95]. The superlattice is not present in the hydrated form, however. As noted in Sect. 3.1, dehydration is accompanied by a change in stacking sequence from 3R1 to 2H1 and it was concluded that in the dehydrated form the chloride ions must occupy trigonal pockets of layer hydroxyl groups, if they are to form a commensurate superlattice arrangement. The adsorption of anions on the surface of LDHs has been studied by scanning tunneling microscopy (STM) and atomic force microscopy (AFM) in the belief that the arrangement of anions adsorbed on the external (001) surface of the platelets may mimic the arrangement of the same anions in the interlayer galleries. The AFM image of the crystal surface of an LDH with

Fig. 41 Measured √ simulated XRD patterns of a Zn1.56 Al LDH consistent with the √ and presence of a ( 3 × 3)R30◦ superlattice of chloride anions. Reprinted with permission from [95]. Copyright American Physical Society

Structural Aspects of Layered Double Hydroxides

75

the formula [Mg6 Al2 (OH)16 ](CO3 )0.5 Cl·2H2 O in contact with an aqueous solution of Na2 SO4 shows two-dimensional periodicity with a unit lattice of a = 0.31 ± 0.02 nm, b = 0.31 ± 0.02 nm and α = 58 ± 3◦ , which was interpreted as being due to the hydroxyl anions of the basal plane, any adsorbed anions having been removed by the AFM tip [307]. (It should be noted, however, that in some cases anions preferentially adsorb on the lateral (101) faces rather than the basal surface e.g. HPO4 2– on [FeII 4 FeIII 2 (OH)12 ]CO3 ·nH2 O [308] and adsorption of catalytically active species on these faces of LDHs has also been proposed [309, 310]; the sulfate may therefore be adsorbed on these faces). STM of the same sample in air showed a two-dimensional lattice with a = 0.75 ± 0.04 nm, b = 1.10 ± 0.03 nm and α = 70 ± 3◦ . One bright spot, every 1.1 [Mg6 Al2 (OH)16 ]2+ unit was attributed to chloride anions adsorbed on the surface. STM of the same crystal at less negative bias voltage exhibited a two-dimensional lattice with a = 0.62 ± 0.03 nm, b = 0.62 ± 0.03 nm and α = 63 ± 3◦ , which was attributed to an ordered array of Al3+ cations with periodicity 2ao . The STM images of the crystal in contact with aqueous solutions of [Fe(CN)6 ]n– (n = 3, 4) showed two-dimensional lattices with a = 1.43 ± 0.06 nm, b = 1.86 ± 0.06 nm and α = 90 ± 4◦ (n = 3) and a = 2.20 ± 0.06 nm, b = 2.20 ± 0.06 nm and α = 75 ± 3◦ (n = 4) respectively, which were attributed to ordered arrays of the respective anions on the surface of the crystal. The calculated surface density of the anions is significantly less than that required in order to balance the positive charge on the layers, indicating that anion-anion repulsions are significant. The same workers reported similar results for [Fe(CN)6 ]n– anions adsorbed on an LDH with the formula [Mg6 Al2 (OH)16 ](CO3 )0.75 (Cl)0.5 ·2H2 O, although the lattice parameters of the anion arrays were significantly smaller in this case [311]. AFM studies suggested [180] that mono- and dianions of 5-benzoyl-4hydroxy-2-methoxybenzenesulfonic acid (MBSA) form ordered arrays on the surface of an LDH and that the arrangement adopted is governed by an interplay between the Coulombic interactions between the MBSA anions and the positively charged surface, steric interactions between neighboring anions as well as the hydrogen bonding interactions between the sulfonate groups and hydroxyl triads groups discussed in Sect. 3.2.3. A floating stearate monolayer has been used as a template for the formation of an LDH by means of spreading a chloroform solution of stearic acid on an aqueous Mg2+ /Al3+ (3 : 1) solution and followed by deposition onto mica as a Z-type film [312]. From the AFM images of the resulting films, it was concluded that the largest crystals were obtained when the stearate monolayer was compressed to give an area of 0.36 nm2 per molecule, which is close to the area occupied by one negative charge for an Mg3 Al LDH (0.34 nm2 ). These studies all suggest that ordering of some organic anions is relatively common, in contrast to the generally disordered nature of interlayers containing inorganic anions.

76

D.G. Evans · R.C.T. Slade

4 Unusual Structural Features of Certain LDHs 4.1 Non-octahedral Coordination of Layer Cations Although the structure of LDHs generally involves six coordinate MII and MIII cations, as discussed previously there are exceptions in the case of MII = Ca with a number of examples confirmed by single crystal XRD [7, 111, 112, 171, 172, 231, 270, 313] and reasons for this have been analyzed [271]. Molecular dynamics calculations on [Mg2 Al(OH)6 ]Cl·2H2 O have suggested that the Mg2+ ion is seven coordinate, being coordinated to water or chloride in addition to the six layer hydroxyl groups and it has been speculated that this coordination geometry may be more common in LDHs than previously thought [233, 314], although to date, there is no direct experimental evidence to support this. Intriguingly, however, whilst the basal spacing in Mg/Al – CO3 LDHs (≈ 0.76 nm) is generally said [7] to be incompatible with carbonate anions tilted at an angle to the layers, in two polymorphs of [Ca4 Al2 (OH)12 ]CO3 ·5H2 O where the carbonate ions are tilted at an angle of 21.8◦ [231] and 20.5◦ [232] to the layers and bonded to the calcium ions, the basal spacings are only 0.755 nm in each case. The seven coordination of the calcium ion, however, results in a higher water content than in the magnesium analogue [Mg4 Al2 (OH)12 ]CO3 ·3H2 O. It has also been argued that the high affinity of sulfonate compounds for LDHs (to the extent that, unlike most anions, they are not displaced by carbonate) indicates a coordinate bond between sulfonate oxygen and aluminum ions. It has been claimed that FT-IR and 13 C CP/NMR data support this hypothesis [139]. With the exception of calcium-containing LDHs, there is scant experimental evidence in support of anion coordination in addition to that by six hydroxyl groups but there is ample evidence that on heating LDHs, the interlayer anions can become grafted to the layers. In these cases, the anion replaces one of the hydroxyl groups, however, so that the cation retains octahedral six-coordination. Grafting has been widely observed for organic anions containing carboxylate, sulfonate, and phosphonate groups [199], oxometallate [146] and polyoxometallate anions [118, 131, 144] as well as simple oxoanions such as carbonate [110]. 4.2 Staging of Interlayer Anions The majority of LDH structures have all interlayers identical, but an increasing number of materials with more than one type of interlayer, giving rise to so-called staged structures [315] are being identified. Some min-

Structural Aspects of Layered Double Hydroxides

77

erals having this type of structure are known, as has been described in Sect. 3.5 for the material with composition [Mg4 Al2 (OH)12 ]CO3 ·3H2 O having alternate interlayers of carbonate anions and water molecules [277]. The mineral coalingite [Mg10 Fe2 (OH)24 ]CO3 ·3H2 O has interlayers containing carbonate and water which alternate with vacant interlayers (as found in brucite) giving a basal spacing of 1.25 nm [316]. There is a tendency to random interstratification, however, and the mineral contains intergrown regions of brucite and sjögrenite or pyroaurite. The mineral motukoreaite [Mg5.6 Al3.4 (OH)18 ](SO4 )1.3 (CO3 )Na0.6 ·12H2 O which has a basal spacing of 1.11 nm, readily changes to a dehydrated phase with reduced basal spacing of 0.87 nm. An intermediate staged phase with basal spacing of 1.98 nm (corresponding to 1.11 + 0.87 nm) with alternate hydrated and dehydrated interlayers can also be obtained [317]. A mineral with the stoichiometry [Mg4 Al2 (OH)12 ](SO4 )0.5 (CO3 )0.5 ·nH2 O has cell parameters a = 0.304 and c = 5.562 nm [303] and has been shown to have alternate interlayers of carbonate (0.756 nm) and sulfate (1.098 nm) anions, so that the basal spacing co is the sum of these two values (1.854 nm) and c = 3co . It has been proposed that the material has the 6R4 polytype in which alternate Oand P-type interlayers are occupied by sulfate and carbonate ions respectively [303]. In the case of synthetic materials, staging has been observed in the case of Mg/Al LDHs with intercalated terephthalate anions. At low temperatures, the terephthalate anions are oriented vertically in the interlayer galleries, whilst on heating dehydration is accompanied by a reorientation of the anions to become parallel to the layers as noted in Sect. 3.2.4 [44]. At intermediate temperatures an interstratified intermediate containing alternate layers of vertical and horizontal anions can be isolated which has a six-layer rhombohedral repeat unit AbC=CaB-CaB=BcA-BcA=AbC-AbC. Staging has also been reported during reaction of these materials with inorganic anions giving partial exchange of terephthalate anions [318]. Staging in LDHs containing alternate interlayers of inorganic anions and organic anions such as (4-phenylazophenyl)acetate [153] or succinate or tartrate [281, 319] anions has also been reported. A series of experiments by O’Hare et al. [320, 321] described in Chapter 4 of this volume, demonstrate that staged intermediates can be observed in a variety of exchange reactions of LDHs. To date, all examples of staging observed with LDH have involved second stage materials (i.e. the interlayers are filled according to an ABAB pattern and no examples of interconversion of third stage (AABAAB)) and second stage materials have been observed. This suggests that the staging phenomena can be explained by the Rüdorff model [322], rather than the Daumas and Hérold model involving buckling of the layers usually invoked for graphite [315]. From the observed change in the magnitude of the red shift in the photoluminescence of [SmW10 O36 ]9– on its intercalation in LDHs, it has been suggested that LDHs do give rise to staging phe-

78

D.G. Evans · R.C.T. Slade

nomena involving buckled layers at low temperature, however [96, 97]. As discussed in Sect. 2.4 the layers in LDHs are more rigid than graphite but much less rigid than silicate clays, so buckling of the layers would seem to be plausible. 4.3 Double Layers of Anions or Cations/Anions Although the interlayer galleries of most LDHs contain a single layer of anions and water molecules, more complex interlayer structures are also possible. In the case of sulfate anions, expansion of the interlayer to give a double layer of anions and water molecules is often observed as has been described in Sect. 3.5 for the green rust [FeII 4 FeIII 2 (OH)12 ]SO4 ·ca.8H2 O [157]. In the case of Mg/Al – SO4 LDHs with interlayer sulfate anions, a material with a basal spacing of 1.115 nm is observed at high relative humidity corresponding to a similar double layer, which collapses on drying to a material with a basal spacing of 0.865 nm, corresponding to a single layer of anions and remaining water molecules [323]. There are also examples of expanded phases containing sulfate anions and hydrated alkali or alkaline earth metal cations. Some of these have been discussed in Sect. 3.5, including examples of synthetic materials such as Zn/CrIII LDHs [305] and natural minerals [303] such as shigaite [259] and motukoreaite [317]. The mineral mountkeithite [324] contains interlayer Mg2+ cations and sulfate anions. Although it is generally assumed that anions such as chloride or nitrate are replaced by organic anions, this is not always the case and neutral organic molecules such as 4-nitrohippuric acid [196] or sodium salts of surfactants [204] may be incorporated in the interlayer galleries of LDH materials. 4.4 Absence of MIII Ions A growing number of materials that have structures related to those of LDHs but contain only divalent, rather than a mixture of divalent and trivalent, cations in the layers have been reported. These include α-hydroxides of nickel and cobalt [129, 325], and layered hydroxy double salts such as Zn5 (OH)8 (NO3 )2 ·2H2 O [326] and Ni1–x Zn2x (OH)2 (CH3 CO2 )2x ·nH2 O [327, 328]. In this latter class of materials, there are vacancies in the brucite-like layers with cations coordinated in tetrahedral sites above and below the vacancies. Some minerals also have this type of structure, e.g. lawsonbauerite [(Mn, Mg)9 Zn4 (OH)22 ](SO4 )2 ·8H2 O which has been characterized by single crystal XRD, has vacancies in two-ninths of the cation sites in a brucitelike lattice with oxygen atoms of the vacant sites coordinated to tetrahedral zinc ions [329]. Although the anions are generally coordinated to the tetrahedral metal anions, there are also examples of this type of material where

Structural Aspects of Layered Double Hydroxides

79

the anion is intercalated in the interlayer galleries as for LDHs. One such example, which has been recently characterized [330] by single crystal XRD is [Co7 (OH)12 ](O3 SC2 H4 SO3 )·2H2 O.

5 Summary and Outlook This review has covered experimental and theoretical studies of the structure of LDHs and highlighted areas where our undertstanding of the detailed structure is still incomplete. Although LDHs are easy to prepare, they are not necessarily easy to prepare as a pure single phase and there is still no clear consensus of opinion on a number of aspects of their structural chemistry, such as (i) the range of possible cations and the relative amounts of each which may be incorporated in the layers of a pure LDH phase, (ii) the different stacking arrangements which may be adopted and (iii) the extent of long-range of order of cations within the layers. Attention needs to be focused on the types of synthetic procedures (see Chapter 2 by He et al. in this volume) which give highly crystalline materials having powder patterns with a high information contents [116, 117, 131, 193, 266, 267] or even single crystals [231] in order to obtain the most accurate structural information. Technical advances such the ready availability of high intensity X-ray generators [95, 263], area detectors [331] and image plate technology [151] as well as improved methods of modeling static and dynamic disorder [332] may also facilitate detailed studies of the structure of LDHs. Possible avenues of approach include using the intensities of several (00l) reflections to calculate one-dimensional electron projections along the c axis and hence derive a model for the orientation of a guest anion in the interlayer galleries [150–153], comparing accurately measured intensities of reflections in order to determine polytype structure [98–100, 108] and observing weak superlattice reflections associated with long-range cation and/or anion ordering [95, 263]. A few examples of these have been discussed above, but others will surely follow. Some theoretical studies on LDHs and related simple models have been discussed above [43, 78, 247, 291, 292], and improvements in both hardware and software should also allow such studies to make future contributions to our understanding of the structure of LDHs. Further spectroscopic, thermodynamic and molecular dynamics studies as well as investigation of in situ reactions aimed at probing the way in which the arrangement of the guest anions in LDHs is determined by the interplay between layer-guest and guestguest (including hydration) interactions as discussed in Sect. 3.2 would also be welcome.

80

D.G. Evans · R.C.T. Slade

References 1. Cavani F, Trifirò F, Vaccari A (1991) Catal Today 11:173 2. Rives V (ed) (2001) Layered Double Hydroxides: Present and Future. Nova Science Publishers, New York 3. Khan AI, O’Hare D (2002) J Mater Chem 12:3191 4. Gaines RV, Skinner HCW, Foord EE, Mason B, Rosenzweig A (1997) Dana’s New Mineralogy, 8th edn. Wiley, New York 5. Allmann R (1968) Acta Cryst B 24:972 6. Taylor HFW (1969) Miner Mag 37:338 7. Taylor HFW (1973) Miner Mag 39:377 8. Catti M, Ferraris G, Hull S, Pavese A (1995) Phys Chem Miner 22:200 9. Greaves C, Thomas MA (1986) Acta Cryst B 42:51 10. Peterson RC, Hill RJ, Gibbs GV (1979) Canad Mineral 17:703 11. Brindley GW, Kao CC (1984) Phys Chem Miner 10:187 12. Parise JB, Leinenweber K, Weidner DJ, Tan K, von Dreele RB (1994) Amer Mineral 79:193 13. Nagai T, Hattori T, Yamanaka T (2000) Amer Mineral 85:760 14. Shinoda K, Aikawa N (1998) Phys Chem Miner 25:197 15. Marchi AJ, Apesteguía CR (1998) Appl Clay Sci 13:35 16. Besserguenev AV, Fogg AM, Francis RJ, Price SJ, O’Hare D, Isupov VP, Tolochko BP (1997) Chem Mater 9:241 17. Alberti G, Costantino U (1996) In: Alberti G, Bein T (eds) Solid State Supramolecular Chemistry: Two- and Three-Dimensional Inorganic Networks. Comprehensive Supramolecular Chemistry, vol 7, Chapter 1. Pergamon and Elsevier Science, Oxford UK 18. Takagi K, Saito N, Shichi T, Sawaki Y (1999) Chem Lett 275 19. Newman SP, Jones W (2001) In: Jones W, Rao CNR (eds) Supramolecular Organization and Materials Design. Cambridge University Press, Cambridge 20. Prakash AS, Kamath PV, Hegde MS (2000) Mater Res Bull 35:2189 21. de Roy A, Forano C, Besse JP (2001) Chapter 1 in [2] 22. Kloprogge JT, Wharton D, Hickey L, Frost RL (2002) Amer Mineral 87:623 23. Alejandre A, Medina F, Salagre P, Correig X, Sueiras JE (1999) Chem Mater 11:939 24. Iglesias AH, Ferreira OP, Gouveia DX, Souza Filho AG, de Paiva JAC, Filho JM, Alves OL (2005) J Solid State Chem 178:142 25. Roussel H, Briois V, Elkaïm E, de Roy A, Besse JP (2000) J Phys Chem B 104:5915 26. Bigey L, Depège C, de Roy A, Besse JP (1997) J Physique IV 7:C2-949 27. Allada RK, Navrotsky A, Boerio-Goates J (2005) Amer Mineral 90:329 28. Allada RK, Navrotsky A, Berbeco HT, Casey WH (2002) Science 296:721 29. Johnson CA, Glasser FP (2003) Clays Clay Miner 51:1 30. Johnson CA, Glasser FP (2003) Clays Clay Miner 51:357 31. Bravo-Suárez JJ, Páez-Mozo EA, Oyama ST (2004) Quim Nova 27:574 32. Bravo-Suárez JJ, Páez-Mozo EA, Oyama ST (2004) Quim Nova 27:601 33. Hansen HCB (2001) Chapter 13 in [2] 34. Bourrié G, Trolard F, Génin JMR, Jaffrezic A, Maître V, Abdelmoula M (1999) Geochim Cosmochim Acta 63:3417 35. Génin JMR, Bourrié G, Trolard F, Abdelmoula M, Jaffrezic A, Refait P, Maître V, Humbert B, Herbillon A (1998) Environ Sci Technol 32:1058 36. Bourrié G, Trolard F, Refait P, Feder F (2004) Clays Clay Miner 52:382 37. Boclair JW, Braterman PS (1999) Chem Mater 11:298

Structural Aspects of Layered Double Hydroxides

81

38. Boclair JW, Braterman PS, Jiang J, Lou S, Yarberry F (1999) Chem Mater 11:303 39. Bellotto M, Rebours B, Clause O, Lynch J, Bazin D, Elkaïm E (1996) J Phys Chem 100:8527 40. Vucelic M, Jones W, Moggridge GD (1997) Clays Clay Miner 45:803 41. Gutmann N, Müller B (1996) J Solid State Chem 122:214 42. Fogg AM, Williams GR, Chester R, O’Hare D (2004) J Mater Chem 14:2369 43. Trave A, Selloni A, Goursot A, Tichit D, Weber J (2002) J Phys Chem B, 106:12291 44. Vucelic M, Moggridge GD, Jones W (1995) J Phys Chem 99:8328 45. del Arco M, Malet P, Trujillano R, Rives V (1999) Chem Mater 11:624 46. Leroux F, Moujahid EM, Taviot-Guého C, Besse JP (2001) Solid State Sci 3:81 47. López-Salinas E, García-Sánchez M, Montoya JA, Acosta DR, Abasolo JA, Schifter I (1997) Langmuir 13:4748 48. Tsuji M, Mao G, Yoshida T, Tamaura Y (1993) J Mater Res 8:1137 49. Thevenot F, Szymanski R, Chaumette P (1989) Clays Clay Miner 37:396 50. Gago S, Pillinger M, Valente AA, Santos TM, Rocha J, Gonçalves IS (2004) Inorg Chem 43:5422 51. Pausch I, Lohse HH, Schürmann K, Allmann R (1986) Clays Clay Miner 34:507 52. Labajos FM, Sánchez-Montero, Holgado MJ, Rives V (2001) Thermochim Acta 370:99 53. Legrand L, Abdelmoula M, Géhin A, Chausé A, Génin JMR (2001) Electrochim Acta 46:1815 54. Inacio J, Taviot-Guého C, Morlat-Thérias S, de Roy ME, Besse JP (2001) J Mater Chem 11:640 55. Xu ZP, Zeng HC (2001) J Phys Chem B 105:1743 56. Brindley GW, Kikkawa S (1979) Amer Mineral 64:836 57. Turco M, Bagnasco G, Costantino U, Marmottini F, Montanari T, Ramis G, Busca G (2004) J Catal 228:43 58. Kaneyoshi M, Jones W (1999) J Mater Chem 9:905 59. Kooli F, Chisem IC, Vucelic M, Jones W (1996) Chem Mater 8:1969 60. Gardner EA, Yun SK, Kwon TH, Pinnavaia TJ (1998) Appl Clay Sci 13:479 61. Zeng HC, Xu ZP, Qian M (1998) Chem Mater 10:2277 62. Sanchez-Valente J, Millet JMM, Figueras F, Fournes L (2000) Hyperfine Interact 131:43 63. Kumbhar PS, Sanchez-Valente J, Millet JMM, Figueras F (2000) J Catal 191:467 64. Kanezaki E (1995) J Mater Sci 30:4926 65. Basile F, Fornasari G, Gazzano M, Kiennemann A, Vaccari A (2003) J Catal 217:245 66. Hibino T, Yamashita Y, Kosuge K, Tsunashima A (1995) Clays Clay Miner 43:427 67. Rives V (1999) Inorg Chem 38:406 68. Gutmann NH, Spiccia L, Turney TW (2000) J Mater Chem 10:1219 69. Intissar M, Jumas JC, Besse JP, Leroux F (2003) Chem Mater 15:4625 70. Shannon RD (1976) Acta Cryst A 32:751 71. de la Calle C, Pons CH, Roux J, Rives V (2003) Clays Clay Miner 51:121 72. Intissar M, Segni R, Payen C, Besse JP, Leroux F (2002) J Solid State Chem 167:508 73. Pérez-Ramírez J, Mul G, Kapreijn F, Moulijn JA (2001) J Mater Chem 11:2529 74. Ribet S, Tichit D, Coq B, Ducourant B, Morato F (1999) J Solid State Chem 142:382 75. Kukkadapu RK, Witkowski MS, Amonette JE (1997) Chem Mater 9:417 76. Mao G, Tsuji M, Tamaura Y (1993) Clays Clay Miner 41:731 77. Leroux F, Adachi-Pagano M, Intissar M, Chauvière S, Forano C, Besse JP (2001) J Mater Chem 11:105 78. Sainz-Diaz CI, Timon V, Botella V, Hernandez-Laguna A (2000) Amer Mineral 85:1038

82

D.G. Evans · R.C.T. Slade

79. Xu ZP, Zeng HC (2001) Chem Mater 13:4555 80. Frost RL, Erickson KL (2004) J Thermal Anal Calorim 76:217 81. Velu S, Ramaswamy V, Ramani A, Chanda BM, Sivasanker S (1997) Chem Commun 2107 82. Velu S, Sabde DP, Shah N, Sivasanker S (1998) Chem Mater 10:3451 83. Velu S, Suzuki K, Osaki T, Ohashi F, Tomura S (1999) Mater Res Bull 34:1707 84. Das NN, Konar J, Mohanta MK, Srivastava SC (2004) J Colloid Interface Sci 270:1 85. Velu S, Suzuki K, Okazaki M, Osaki T, Tomura S, Ohashi F (1999) Chem Mater 11:2163 86. Velu S, Suzuki K, Kapoor MP, Tomura S, Ohashi F, Osaki T (2000) Chem Mater 12:719 87. Tichit D, Das N, Coq B, Durand R (2002) Chem Mater 14:1530 88. Das N, Samal A (2004) Micropor Mesopor Mater 72:219 89. Ren QL, Luo Q, Liu DZ, Chen ST (2003) Mater Sci Forum (Funct Graded Mater VII) 423-425:157 90. Saber O, Tagaya H (2003) J Incl Phenom Macrocyclic Chem 45:109 91. Intissar M, Holler S, Malherbe F, Besse JP, Leroux F (2004) J Phys Chem Solids 65:453 92. Hofmeister W, von Platen H (1992) Cryst Rev 3:3 93. Solin SA, Hines D, Yun SK, Pinnavaia TJ, Thorpe MF (1995) J Non-Cryst Solids 182:212 94. Hines DR, Solin SA, Costantino U, Nocchetti M (2000) Mol Cryst Liq Cryst 341:1181 95. Hines DR, Solin SA, Costantino U, Nocchetti M (2000) Phys Rev B 61:11348 96. Park TR, Park TY, Kim H, Min P (2002) J Phys: Condens Matter 14:11687 97. Park TR (2004) J Korean Phys Soc 44:398 98. Bookin AS, Drits VA (1993) Clays Clay Miner 41:551 99. Bookin AS, Cherkashin VI, Drits VA (1993) Clays Clay Miner 41:558 100. Drits VA, Bookin AS (2001) Chapter 2 in [2] 101. Vaysse C, Guerlou-Demourgues L, Demourgues A, Lazartigues F, Fertier D, Delmas C (2002) J Mater Chem 12:1035 102. Guerlou-Demourgues L, Denage C, Delmas C (1994) J Power Sources 52:269 103. Thomas GS, Rajamathi M, Kamath PV (2004) Clays Clay Miner 52:693 104. Moggridge GD, Parent P, Tourillon G (1994) Clays Clay Miner 42:462 105. Stanimirova T (2001) Ann Univ Sofia Geol 94:73 106. Ennadi A, Legrouri A, de Roy A, Besse JP (2000) J Solid State Chem 152:568 107. Stanimirova TS, Kirov G, Dinolova E (2001) Mat Sci Lett 20:453 108. Newman SP, Jones W, O’Connor P, Stamires DN (2002) J Mater Chem 12:153 109. Parthasarathy G, Kantam ML, Choudary BM, Reddy CV (2002) Micropor Mesopor Mater 56:147 110. Vaysse C, Guerlou-Demourgues L, Delmas C (2002) Inorg Chem 41:6905 111. Renaudin G, Rapin JP, Elkaïm E, François M (2004) Cement Concrete Res 34:1845 112. Rapin JP, Renaudin G, Elkaïm E, François M (2002) Cement Concrete Res 32:513 113. Bookin AS, Cherkashin VI, Drits VA (1993) Clays Clay Miner 41:631 114. Fogg AM, Freij AJ, Parkinson GM (2002) Chem Mater 14:232 115. Thiel JP, Chiang CK, Poeppelmeier KR (1993) Chem Mater 5:297 116. Hansen HCB, Koch CB, Taylor RM (1994) J Solid State Chem 113:46 117. Hansen HCB, Taylor RM (1991) Clay Miner 26:507 118. Ménétrier M, Han KS, Guerlo-Demourgues L, Delmas C (1997) Chem Mater 36:2441 119. Zhao Y, Li F, Zhang R, Evans DG, Duan X (2002) Chem Mater 14:4286 120. Gago S, Pillinger M, Santos TM, Rocha J, Gonçalves IS (2004) Eur J Inorg Chem 1389 121. Drits V, ´ Srodo´ n J, Eberl DD (1997) Clays Clay Miner 45:461

Structural Aspects of Layered Double Hydroxides

83

122. Jaboyedoff M, Kübler B, Thélin P (1999) Clay Miner 34:601 123. Ennadi A, Legrouri A, de Roy A, Besse JP (2000) J Mater Chem 10:2337 124. Guggenheim S, Bain DC, Bergaya F, Brigatti MF, Drits VA, Eberl DD, Formoso MLL, Galan E, Merriman RJ, Peacor DR, Stanjek H, Watanabe T (2002) Clays Clay Miner 50:406 125. Radha AV, Kamath PV, Subbanna GN (2003) Mater Res Bull 38:731 126. Delmas C, Tessier C (1997) J Mater Chem 7:1439 127. Rajamathi M, Kamath PV, Seshadri R (2000) J Mater Chem 10:503 128. Ramesh TN, Jayashree RS, Kamath PV (2003) Clays Clay Miner 51:570 129. Xu ZP, Zeng HC (2000) Int J Inorg Mater 2:187 130. Prévot V, Forano C, Besse JP (1998) Inorg Chem 37:4293 131. Han KS, Guerlou-Demourgues L, Delmas C (1996) Solid State Ionics 84:227 132. Rey S, Mérida-Robles J, Han KS, Guerlou-Demourgues L, Delmas C, Duguet E (1999) Polymer Int 48:277 133. Aloisi GG, Costantino U, Elisei F, Latterini L, Natali C, Nocchetti M (2002) J Mater Chem 12:3316 134. Boclair JW, Braterman PS, Brister BD, Yarberry F (1999) Chem Mater 11:2199 135. Beaudot P, de Roy ME, Besse JP (2004) Chem Mater 16:935 136. Kanezaki E, Maeda K (2002) Nippon Kagaku Kaishi 393 137. Beaudot P, de Roy ME, Besse JP (2004) J Solid State Chem 177:2691 138. Beaudot P, de Roy ME, Besse JP (2001) J Solid State Chem 161:332 139. Kanezaki E (2003) J Incl Phenom Macrocyclic Chem 46:89 140. Kanezaki E, Sugiyama S, Ishikawa Y (1995) J Mater Chem 5:1969 141. Weir MR, Kydd RA (1998) Inorg Chem 37:5619 142. Kooli F, Rives V, Ulibarri MA (1995) Inorg Chem 34:5122 143. Hibino T, Tsunashima A (1997) Chem Mater 9:2082 144. Malherbe F, Bigey L, Forano C, de Roy A, Besse JP (1999) J Chem Soc Dalton Trans 3831 145. Malherbe F, Besse JP (2000) J Solid State Chem 155:332 146. Rives V, Ulibarri MA (1999) Coord Chem Rev 181:61 147. Liu YJ, Schindler JL, DeGroot DC, Kannewurf CR, Hirpo W, Kanatzidis MG (1996) Chem Mater 8:525 148. Zhang X, Lerner MM (1999) Chem Mater 11:1100 149. de La Calle C, Tejedor MI, Pons CH (1996) Clays Clay Miner 44:68 150. Bauer J, Behrens P, Speckbacher M, Langhals H (2003) Adv Funct Mater 13:241 151. Itoh T, Ohta N, Shichi T, Yui T, Tagaki K (2003) Langmuir 19:9120 152. Kanoh T, Shichi T, Takagi K (1999) Chem Lett 117 153. Iyi N, Kurashima K, Fujita T (2002) Chem Mater 14:583 154. McMurdie MF, Morris MC, Evans EH, Paretzkin B, Wong-Ng W, Hubbard CR (1986) Powder Diffraction 1:40 155. Refait PH, Abdelmoula M, Génin JMR (1998) Corrosion Sci 40:1547 156. Braithwaite RSW, Dunn PJ, Pritchard RG, Paar WH (1994) Miner Mag 58:79 157. Simon L, François M, Refait P, Renaudin G, Lelaurin M, Génin JMR (2003) Solid State Sci 5:327 158. Rives R (2002) Mater Chem Phys 75:19 159. Barriga C, Jones W, Malet P, Rives V, Ulibarri MA (1998) Inorg Chem 37:1812 160. Holgado MJ, Rives V, Sanroman MS, Malet P (1996) Solid State Ionics 92:273 161. Kloprogge JT, Frost RL (2001) Chapter 5 in [2] 162. Kloprogge JT (ed)(2005) The Application of Vibrational Spectroscopy to Clay Minerals and Layered Double Hydroxides. Clay Minerals Society, Aurora 163. Kloprogge JT, Frost RL (1999) J Solid State Chem 146:506

84

D.G. Evans · R.C.T. Slade

164. Frost RL, Martens W, Ding Z, Kloprogge JT, Johnson TE (2003) Spectrochim Acta A 59:291 165. Frost RL, Ding Z, Martens W, Johnson TE, Kloprogge JT (2003) Spectrochim Acta A 59:321 166. Kloprogge JT, Hickey L, Frost RL (2004) J Raman Spec 35:967 167. Marcelin G, Stockhausen NJ, Post JFM, Schutz A (1989) J Phys Chem 93:4646 168. Xu R, Zeng HC (2003) Chem Mater 15:2040 169. Hou XQ, Kirkpatrick RJ, Yu P, Moore D, Kim Y (2000) Amer Mineral 85:173 170. Xu ZP, Zeng HC (2001) Chem Mater 13:4564 171. Renaudin G, Rapin JP, Humbert B, François M (2000) Cement Concrete Res 30:307 172. Renaudin G, François M (1999) Acta Cryst C 55:835 173. Kloprogge JT, Hickey L, Frost RL (2001) Appl Clay Sci 18:37 174. Boclair JW, Braterman PS, Brister BD, Wang Z, Yarberry F (2001) J Solid State Chem 161:249 175. Kloprogge JT, Weier M, Crespo I, Ulibarri MA, Barriga C, Rives V, Martens WN, Frost RL (2004) J Solid State Chem 177:1382 176. Hansen HCB, Koch CB (1994) Clays Clay Miner 42:170 177. Braterman PS, Tan CQ, Zhao JX (1994) Mater Res Bull 29:1217 178. Raki L, Rancourt DG, Detellier C (1995) Chem Mater 7:221 179. del Arco M, Gutierrez S, Martin C, Rives V, Rocha J (2004) J Solid State Chem 177:3954 180. Franklin KR, Lee E, Nunn CC (1995) J Mater Chem 5:565 181. Cai H, Hillier AC, Franklin KR, Nunn CC, Ward MD (1994) Science 266:1551 182. Maxwell RS, Kukkadapu RK, Amonette JE, Cho H (1999) J Phys Chem B 103:5197 183. Newman SP, Williams SJ, Coveney PV, Jones W (1998) J Phys Chem B 102:6710 184. Hou XQ, Kirkpatrick RJ (2002) Chem Mater 14:1195 185. Hou XQ, Kirkpatrick RJ (2000) Chem Mater 12:1890 186. Kirkpatrick RJ, Yu P, Hou XQ, Kim Y (1999) Amer Mineral 84:1186 187. Kozlova SG, Gabuda SP, Isupov VP, Chupakhina LE (2003) J Struc Chem 44:198 188. van der Pol A, Mojet BL, van de Ven E, de Boer E (1994) J Phys Chem 98:4050 189. Morlat-Thérias S, Mousty C, Palvadeau P, Molinié P, Léone P, Rouxel J, Taviot-Guého C, Ennaqui A, de Roy A, Besse JP (1999) J Solid State Chem 144:143 190. Moggridge GD, Parent P, Tourillon G (1995) Physica B 209:269 191. Meyn M, Beneke K, Lagaly G (1990) Inorg Chem 29:5201 192. Costantino U, Coletti N, Nocchetti M, Aloisi GG, Elisei F, Latterini L (2000) Langmuir 16:10351 193. Costantino U, Coletti N, Nocchetti, Aloisi GG, Elisei F (1999) Langmuir 15:4454 194. Xing Y, Li D, Ren L, Evans DG, Duan X (2003) Acta Chim Sinica 61:267 195. Latterini L, Elisei F, Aloisi GG, Costantino U, Nocchetti M (2002) Phys Chem Chem Phys 4:2792 196. Cooper S, Dutta PK (1990) J Phys Chem 94:114 197. Sasai R, Shin’ya N, Shichi T, Takagi K, Gekko K (1999) Langmuir 15:413 198. Carlino S, Hudson MJ (1995) J Mater Chem 5:1433 199. Prévot V, Forano C, Besse JP (2001) Appl Clay Sci 18:3 200. Bravo-Suárez JJ, Páez-Mozo EA, Oyama ST (2004) Micropor Mesopor Mater 67:1 201. Bravo-Suárez JJ, Páez-Mozo EA, Oyama ST (2004) Chem Mater 16:1214 202. Choy JH, Jung EY, Son YH, Park M (2004) J Phys Chem Solids 65:509 203. Xu ZP, Braterman PS (2003) J Mater Chem 13:268 204. Crepaldi EL, Pavan PC, Tronto J, Valim JB (2002) J Colloid Interface Sci 248:429 205. Zhao H, Nagy KL (2004) J Colloid Interface Sci 274:613

Structural Aspects of Layered Double Hydroxides 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228. 229. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249.

85

Borja M, Dutta PK (1992) J Phys Chem 96:5434 Corma A, Fornés V, Rey F, Cervilla A, Llopis E, Ribera A (1995) J Catal 152:237 Raki L, Beaudoin JJ, Mitchell L (2004) Cement Concrete Res 34:1717 Kuk WK, Huh YD (1997) J Mater Chem 7:1933 Fogg AM, Green VM, Harvey HG, O’Hare D (1999) Adv Mater 11:1466 Fogg AM, Dunn JS, Shyu SG, Cary DR, O’Hare D (1998) Chem Mater 10:351 Weir MR, Morre J, Kydd RA (1997) Chem Mater 9:1686 Lee JH, Rhee SW, Jung DY (2004) Chem Mater 16:3774 Ocko BM, Wu XZ, Sirota EB, Sinha SK, Gang O, Deutsch M (1997) Phys Rev E 55:3164 Newman SP, Jones W (1998) New J Chem 105 Xu ZP, Braterman PS, Yu K, Xu H, Wang Y, Brinker CJ (2004) Chem Mater 16:2750 Greenwell HC, Jones W, Newman SP, Coveney PV (2003) J Mol Struct 647:75 Valim J, Kariuki BM, King J, Jones W (1992) Mol Cryst Liq Cryst 211:271 Shichi T, Takagi K, Sawaki Y (1996) Chem Commun 2027 Rhee SW, Jung DY (2002) Bull Korean Chem Soc 23:35 Takagi K, Harata E, Shichi T, Kanoh T, Sawaki Y (1997) J Photochem Photobiol A: Chem 105:47 Morel-Desrosiers N, Pisson J, Israëli Y, Taviot-Guého C, Besse JP, Morel JP (2003) J Mater Chem 132:2582 Wang JW, Kalinichev AG, Amonette JE, Kirkpatrick RJ (2003) Amer Mineral 88:398 Newman SP, Greenwell HC, Coveney PV, Jones W (2001) Chapter 3 in [2] Newman SP, Di Cristina T, Coveney PV, Jones W (2002) Langmuir 18:2933 Hou XQ, Kalinichev AG, Kirkpatrick RJ (2002) Chem Mater 14:2078 Kalinichev AG, Kirkpatrick RJ, Cygan RT (2000) Amer Mineral 85:1046 Kalinichev AG, Kirkpatrick RJ (2002) Chem Mater 14:3539 Renaudin G, Kubel F, Rivera JP, Francois M (1999) Cement Concrete Research 29:1937 Andersen MD, Jakobsen HJ, Skibsted J (2002) J Phys Chem A 106:6676 François M, Renaudin G, Evrard O (1998) Acta Cryst C54:1214 Renaudin G, François M, Evrard O (1999) Cement Concrete Research 29:63 Wang JW, Kalinichev AG, Kirkpatrick RJ, Hou XQ (2001) Chem Mater 13:145 Kalinichev AG, Wang J, Kirkpatrick RJ, Cygan RT (2001) AIChE Symp Ser 325:251 Fogg AM, Rohl AL, Parkinson GM, O’Hare D (1999) Chem Mater 11:1194 Toraishi T, Nagasaki S, Tanaka S (2002) Radiochim Acta 90:671 Toraishi T, Nagasaki S, Tanaka S (2002) Appl Clay Sci 22:17 Drissi, SH, Refait P, Abdelmoula M, Génin JMR (1995) Corrosion Sci 37:2025 Yun SK, Pinnavaia TJ (1995) Chem Mater 7:348 Hou XQ, Bish DL, Wang SL, Johnston CT, Kirkpatrick RJ (2003) Amer Mineral 88:167 Génin JMR, Ruby C (2004) Solid State Sci 6:705 Wang J, Kalinichev AG, Kirkpatrick RJ (2004) Geochim Cosmochim Acta 68:3351 Pesic L, Salipurovic S, Markovic V, Vucelic D, Kagunya W, Jones W (1992) J Mater Chem 2:1069 Kagunya W, Dutta PK, Lei Z (1997) Physica B 234:910 Mitra S, Pramanik A, Chakrabaty D, Mukhopadhyay R (2004) Pramana – J Phys 63:437 Mehrotra V, Giannelis EP (1992) J Appl Phys 72:1039 Greenwell HC, Stackhouse S, Coveney PV, Jones W (2003) J Phys Chem B 107:3476 Hines DR, Solin SA, Costantino U, Nocchetti M (1998) Solid State Comm 12:971 Kagunya W, Baddour-Hadjean R, Kooli F, Jones W (1998) Chem Phys 236:225

86

D.G. Evans · R.C.T. Slade

250. Kim D, Huang C, Lee H, Han I, Kang S, Kwon S, Lee J, Han Y, Kim H (2003) Appl Catal A: Gen 249:229 251. Basile F, Vaccari A (2001) Chapter 10 in [2] 252. Monzón A, Romeo E, Marchi AJ (2001) Chapter 11 in [2] 253. Chellam U, Xu ZP, Zeng HC (2000) Chem Mater 12:650 254. Liu JJ, Li F, Evans DG, Duan X (2003) Chem Commun 542 255. Li F, Liu JJ, Evans DG, Duan X (2004) Chem Mater 16:1597 256. Porta P, Morpurgo S, Pettiti I (1996) J Solid State Chem 121:372 257. Pauling L (1929) J Am Chem Soc 51:1010 258. Serna CJ, Rendon JL, Iglesias JE (1982) Clays Clay Miner 30:180 259. Cooper MA, Hawthorne FC (1996) Canad Mineral 34:91 260. Gastuche MC, Brown G, Mortland M (1967) Clay Miner 7:177 261. Boclair JW, Braterman PS (1998) Chem Mater 10:2050 262. Roussel H, Briois V, Elkaïm E, de Roy A, Besse JP, Jolivet JP (2001) Chem Mater 13:329 263. Solin SA, Hines DR, Seidler GT, Treacey MMJ (1996) J Phys Chem Solids 57:1043 264. Steeds JW, Morniroli JP (1992) Rev Miner Geochem 27:37 265. Ingram L, Taylor HFW (1967) Miner Mag 36:465 266. Costantino U, Marmottini F, Nocchetti M, Vivani R (1998) Eur J Inorg Chem 1439 267. Costantino U, Casciola M, Massinelli L, Nocchetti M, Vivani R (1997) Solid State Ionics 97:203 268. Sissoko I, Iyagba ET, Sahai R, Biloen P (1985) J Solid State Chem 60:283 269. Hou X, Kirkpatrick RJ (2001) Inorg Chem 40:6397 270. Terzis A, Filippakis S, Kuzel HJ, Burzlaff H (1987) Z Kristallogr 181:29 271. Rousselot I, Taviot-Guého C, Leroux F, Léone P, Palvadeau P, Besse JP (2002) J Solid State Chem 167:137 272. Dunn PJ, Peacor DR, Palmer TD (1979) Amer Mineral 64:127 273. Huminicki DMC, Hawthorne FC (2003) Canad Mineral 41:79 274. Rius J, Plana F (1986) Neues Jahrb Miner Monatsh 263 275. Merlino S, Orlandi P (2001) Amer Mineral 86:1293 276. Chao GY, Gault RA (1997) Canad Mineral 35:1541 277. Arakcheeeva AV, Pushcharovskii DY, Rastsvetaeva RK, Atencio D, Lubman GU (1996) Krystallografiya 41:1024; Crystallog Rep 41:972 278. Xiao YQ, Thorpe MF, Parkinson JB (1999) Phys Rev 59:277 279. Chen J, Thorpe MF, Davis LC (1995) J Appl Phys 77:4349 280. Evans J, Pillinger M, Zhang J (1996) J Chem Soc Dalton Trans 2963 281. Taviot-Guého C, Leroux F, Payen C, Besse JP (2005) Appl Clay Sci 28:111 282. Scheinost AC, Sparks DL (2000) J Colloid Interface Sci 223:167 283. Refait P, Abdelmoula M, Trolard F, Génin JMR, Ehrhardt JJ, Bourrié G (2001) Amer Mineral 86:731 284. Leroux F, Moujahid EM, Roussel H, Flank AM, Briois V, Besse JP (2002) Clays Clay Miner 50:254 285. Frost RL, Weier ML, Clissold ME, Williams PA (2003) Spectrochim Acta A – Molec Biomolec Spec 59:3313 286. Hernandez-Moreno MJ, Ulibarri MA, Rendon JL, Serna CJ (1985) Phys Chem Miner 12:34 287. d’Espinose de la Caillerie JB, Kermarec M, Clause O (1995) J Am Chem Soc 117:11471 288. Mockenhaupt C, Zeiske T, Lutz HD (1998) J Mol Struct 443:191 289. Frost RL, Erickson KL (2004) Spectrochim Acta A 60:3001

Structural Aspects of Layered Double Hydroxides 290. 291. 292. 293. 294. 295. 296. 297. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 308. 309. 310. 311. 312. 313. 314. 315. 316. 317. 318. 319. 320. 321. 322. 323. 324. 325. 326. 327. 328. 329. 330. 331. 332.

87

Johnson TE, Martens W, Frost RL, Ding Z, Kloprogge JT (2002) J Raman Spec 33:604 Martínez-Alonso S, Rustad JR, Goetz AFH (2002) Amer Mineral 87:1224 Martínez-Alonso S, Rustad JR, Goetz AFH (2002) Amer Mineral 87:1215 Koch CB (2001) Chapter 7 in [2] Hansen HCB, Guldberg S, Erbs M, Koch CB (2001) Appl Clay Sci 18:81 Génin JMR (2004) Hyperfine Interact 156/7:471 Rancourt DG (1998) Hyperfine Interact 117:3 Refait P, Génin JMR (1997) Corrosion Sci 39:539 Aissa R, Ruby C, Gehin A, Abdelmoula M, Génin JMR (2004) Hyperfine Interact 156/7:445 Mattson SM, Rossman GR (1984) Phys Chem Miner 11:225 Roelofs JCAA, van Bokhoven JA, van Dillen AJ, Geus JW, de Jong KP (2002) Chem Eur J 8:5571 Depège C, El Metoui FZ, Forano C, de Roy A, Dupuis J, Besse JP (1996) Chem Mater 8:952 Schutz A, Biloen P (1987) J Solid State Chem 68:360 Drits VA, Sokolova TN, Sokolova GV, Cherkashin VI (1987) Clays Clay Miner 35:401 Bish DL (1980) Bull Minéral 103:170 Khaldi M, de Roy A, Chaouch M, Besse JP (1997) J Solid State Chem 130:66 Hines DR, Seidler GT, Treacy MMJ, Solin SJ (1997) Solid State Sci 101:835 Yao K, Taniguchi M, Nakata M, Takahashi M, Yamagishi A (1998) Langmuir 14:2410 Bocher F, Géhin A, Ruby C, Ghanbaja J, Abdelmoula M, Génin JMR (2004) Solid State Sci 6:117 Sels B, de Vos D, Buntinx M, Pierard F, Kirsch-de Mesmaeker A, Jacobs P (1999) Nature 400:855 Choudary BM, Madhi S, Chowdari NS, Kantam ML, Sreedhar B (2002) J Amer Chem Soc 124:14127 Yao K, Taniguchi M, Nakata M, Yamagishi A (1998) J Electroanal Chem 458:249 He JX, Yamashita S, Jones W, Yamagishi A (2002) Langmuir 18:1580 Allmann R (1977) Neues Jahrb Miner Monatsh 136 Cygan RT, Liang JJ, Kalinichev AG (2004) J Phys Chem B 108:1255 Fogg AM, Dunn JS, O’Hare D (1998) Chem Mater 10:356 Pastor-Rodriguez J, Taylor HFW (1971) Miner Mag 38:286 Brindley GW (1979) Miner Mag 43:337 Kaneyoshi M, Jones W (1998) Chem Phys Lett 296:183 Pisson J, Taviot-Guého C, Israëli Y, Leroux F, Munsch P, Itié JP, Briois V, MorelDesrosiers N, Besse JP (2003) J Phys Chem B 107:9243 Williams GR, Norquist AJ, O’Hare D (2004) Chem Mater 16:975 Williams GR, Norquist AJ, O’Hare D (2003) Chem Commun 1816 Timp G, Dresselhaus MS (1984) J Phys C: Solid State Phys 17:2641 Brindley GW, Kikkawa S (1980) Clays Clay Miner 28:87 Hudson DR, Bussell M (1981) Miner Mag 44:345 Kamath PV, Therese GHA, Gopalakrishnan J (1997) J Solid State Chem 128:38 Stählin W, Oswald HR (1970) Acta Cryst B26:860 Choy JH, Kwon YM, Han KS, Song SW, Chang SH (1998) Mater Lett 34:356 Choy JH, Kwon YM, Song SW, Chang SH (1997) Bull Korean Chem Soc 18:450 Treiman AH, Peacor DR (1982) Amer Mineral 67:1029 Forster PM, Tafoya MM, Cheetham AK (2004) J Phys Chem Solids 65:11 Bhuvanesh NSP, Reibenspies JH (2003) J Appl Cryst 36:1480 Izumi F (2004) Solid State Ionics 172:1

Struct Bond (2006) 119: 89–119 DOI 10.1007/430_006 © Springer-Verlag Berlin Heidelberg 2005 Published online: 28 September 2005

Preparation of Layered Double Hydroxides Jing He · Min Wei · Bo Li · Yu Kang · David G Evans · Xue Duan (u) Ministry of Education Key Laboratory of Science and Technology of Controllable Chemical Reactions, Beijing University of Chemical Technology, Box 98, 100029 Beijing, People’s Republic of China [email protected] 1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 2.1 2.2 2.3 2.4

Coprecipitation . . . . . . . . . . . . . . . . . . . . . . . Precipitation at Low Supersaturation . . . . . . . . . . . Precipitation at High Supersaturation . . . . . . . . . . . A Method Involving Separate Nucleation and Aging Steps Urea Hydrolysis Method . . . . . . . . . . . . . . . . . .

. . . . .

92 93 95 97 100

3

Ion-exchange Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

103

4

Rehydration Using Structural Memory Effect . . . . . . . . . . . . . . . .

106

5

Hydrothermal Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

108

6

Secondary Intercalation (Pre-pillaring Method) . . . . . . . . . . . . . . .

108

7

An Intercalation Method Involving Dissolution and Re-coprecipitation Procedures . . . . . . . . . . . . . . . . . . . . . .

109

8 8.1 8.2 8.3 8.4 8.5 8.6

Other Methods . . . . . . . . . . Salt-oxide (or Hydroxide) Method Non-equilibrium Aging Method . Non-conventional Aging Methods Surface Synthesis . . . . . . . . . Templated Synthesis . . . . . . . . Miscellaneous Methods . . . . . .

. . . . . . .

110 110 110 111 111 111 112

9

Summary and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

115

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

115

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . .

. . . . . . .

. . . . .

. . . . . . .

. . . . .

. . . . . . .

. . . . .

. . . . . . .

. . . . .

. . . . . . .

. . . . .

. . . . . . .

. . . . .

. . . . . . .

. . . . .

. . . . . . .

. . . . .

. . . . . . .

90

Abstract Layered double hydroxides (LDHs) comprise an extensive class of materials that are very easy to synthesize in the laboratory, albeit not always as pure phases. In this chapter, we review the wide variety of methods that are available for the synthesis of LDHs and focus on the way in which the physicochemical properties of the materials (such as phase purity, crystallinity and surface area) vary with synthesis method. The flexibility of the different methods is also discussed: some methods can be used to synthesize LDHs containing a wide range of constituent cations and anions, whilst others are more limited in scope. In some cases, the potential for scale-up of a method to produce larger quantities of material is also noted.

90

J. He et al.

Keywords Layered double hydroxide · Hydrotalcite · Preparation · LDH Abbreviations CSC 4
-chloro-4-stilbenecarboxylic acid DNA Deoxyribonucleic acid edta Ethylenediamine tetraacetate EG Ethylene glycol LDH Layered double hydroxide HTLc Hydrotalcite-like compounds ICTA Indole-2-carboxylate MAS Magic angle spinning NMR Nuclear magnetic resonance Phe Phenylalanine PXRD Powder X-ray diffraction SEM Scanning electron microscope TEM Transmission electron microscope THF Tetrahydrofuran

1 Introduction The terms layered double hydroxides (LDHs) or hydrotalcite-like compounds (HTLc) [1] are used to designate synthetic or natural lamellar hydroxides with two or more kinds of metallic cations in the main layers and hydrated interlayer domains containing anionic species. This large family of compounds is also commonly referred to as hydrotalcites or anionic clays, the latter term indicating a complementarity with the more usual cationic clays whose interlamellar domains contain cationic species. The mineral hydrotalcite, which was discovered in Sweden around 1842, is a hydroxycarbonate of magnesium and aluminum and can be considered a prototype of this class of materials. As noted in Chapter 1, the correct formula for hydrotalcite, [Mg6 Al2 (OH)16 CO3 · 4H2 O], was first suggested by Manasse, who was the first to recognize that carbonate ions were an essential feature of the mineral [2]. LDHs may be described by the general formula [M(II)1–x M(III)x (OH)2 ]x+ n– (A )x/n · mH2 O, where M(II) and M(III) are di- and trivalent metals respectively, and An– is an anion [3]. As discussed in detail in Chapter 1, their structure is based on brucite-like layers, in which a divalent metal cation is located in the center of oxygen octahedra and two-dimensional infinite layers are formed by edge-sharing of the octahedra. Partial isomorphous substitution of trivalent cations for divalent ones results in a positive charge on the layers. Organic or inorganic anions are intercalated between the layers in order to maintain charge balance, and water of crystallization is also generally found in the interlayer galleries [4, 5]. Fig. 1 shows the idealized structure of a layered double hydroxide with interlayer carbonate anions [6].

Preparation of Layered Double Hydroxides

91

LDHs may be synthesized with a wide range of compositions and a large number of materials with a wide variety of M(II)/M(III) cation combinations as well as M(I)/M(III) cation pairs (e.g. Li/Al) with different anions in the interlayer can be obtained. It is often said that only M(II) and M(III) ions having an ionic radius not too different from that of Mg2+ may be accommodated in the octahedral sites of the close-packed hydroxide ions in the brucite-like layers to form LDH compounds [7] and that cations which are too small, such as Be2+ , or too large, such as Cd2+ , give rise to other types of compounds, and also that the most reliable composition range corresponds approximately to 0.2 ≤ x ≤ 0.4. As discussed in Chapter 1, many LDHs with composition and stoichiometry well outside these limits have been reported in the literature, but they may not be pure phases. The interlayer domains contain anions, water molecules and sometimes other neutral or charged moieties. Relatively weak bonding occurs between the interlayer ions or molecules and the host sheets. A key feature of these materials is therefore their anionic exchange capacity, which makes them unique as far as inorganic materials are concerned. A great variety of anionic species can be introduced into the interlayer regions through one-pot syntheses such as coprecipitation, or post-synthesis modifications, such as ion exchange. Incorporated anions can be simple ones, such as carbonate, nitrate or chloride, or larger organic anions, such as carboxylates or sulfonates or inorganic polyoxometalates such as Keggin, Finke, or Dawson-type anions [8, 9]. Layered solids composed of alternating inorganic and organic sheets have received considerable attention because of their many practical applications, including as catalysts [10–12],

Fig. 1 Idealized structure of a layered double hydroxide with interlayer carbonate anions. Several parameters are defined. Reprinted with permission from [6]. Copyright Elsevier

92

J. He et al.

functional materials [13–15] and nanocomposite materials [16] and in separation science [17, 18]. The attractive feature of such materials is that they can serve as a template for the formation of supramolecular structures [16]. The host layers can impose a restricted geometry on the interlayer guests, leading to enhanced control of stereochemistry, rates of reaction, and product distributions. Many species can be assembled by reaction of guest species in the LDH matrices, and the study of novel functional materials based on LDHs is therefore a rapidly growing field. A number of synthetic techniques have been successfully employed in the preparation of LDHs. The most commonly used are simple coprecipitation methods; the second method is based on the classical ion exchange process; the third, reconstruction, is based on the so-called “memory effect” in Sect. 4. Furthermore, some other methods, such as sol-gel synthesis using ethanol and acetone solutions [19], and a fast nucleation process followed by a separate aging step at elevated temperatures [20], have also been reported. The synthesis of LDHs has been reviewed from both experimental [3, 7] and thermodynamic viewpoints [21]. Using a thermodynamic approach, the synthesis of LDHs using hydrothermal reconstruction and coprecipitation methods was discussed in terms of the standard molar Gibbs free energy change of reaction. Good qualitative agreement was found between theoretical and experimental data using a simple model for the estimation of the thermodynamic parameters associated with the formation of LDHs [21]. This review is focused on the methods of preparation of LDHs, emphasizing those reported during the last decade.

2 Coprecipitation The method of coprecipitation is the most commonly used preparative technique for LDHs. It has been used extensively for the one-pot direct synthesis of LDHs containing a variety of layer cations and interlayer anions [22]. It also amenable to scaling up in order to produce large quantities of material. In the coprecipitation method, aqueous solutions of M2+ (or mixtures of M2+ species) and M3+ (or mixtures) containing the anion that is to be incorporated into the LDH are used as precursors. Of the various (M(II)/M(III)) LDH systems, some display a variable composition in terms of M(II)/M(III) ratio, while certain phases can be obtained only within a narrow range of ratios. For example, Zn/Al – CO3 LDHs have been reported [7] with a variable Zn/Al ratio. This flexibility in the layer charge density allows the packing density of the interlamellar anion to be varied. Another interesting feature of this preparative method is that a wide variety of anionic species can be directly intercalated between the hydroxylated sheets. This synthetic route is often the method of choice for the preparation of organic anion-containing

Preparation of Layered Double Hydroxides

93

Table 1 pH values of precipitation of some M(III) and M(II) hydroxides [3] Cation

pH at 10–2 M

pH at 10–4 M

pH at which hydroxide re-dissolves

Al3+ Cr3+ Cu2+ Zn2+ Ni2+ Fe2+ Co2+ Mn2+

3.9 5.0 5.0 6.5 7.0 7.5 7.5 8.5

8.0 9.5 6.5 8.0 8.5 9.0 9.0 10.0

9.0–12.0 12.5 14

LDHs which are difficult to obtain in other ways. In order to ensure simultaneous precipitation of two or more cations, it is necessary to carry out the synthesis under conditions of supersaturation. Generally, supersaturation conditions are reached by controlling the pH of the solution. In particular, it is necessary to precipitate at a pH higher than or equal to the one at which the most soluble hydroxide is precipitated [3]. Table 1 lists the appropriate pH for precipitation of the hydroxides of the most common metals forming LDHs. Following coprecipitation, a thermal treatment process is often performed in order to increase yields and/or the crystallinity of amorphous or badly crystallized materials. Thermal treatment in order to enhance crystallinity can be classified according to the temperature employed. A conventional aging process consists of heating a reactor containing an aqueous suspension of the LDH product at a temperature between 273–373 K over a few hours or several days. The second method involves hydrothermal treatment, which involves heating the sample in a gold or silver capsule or in stainless steel autoclave under high pressure ranging from 10 to 150 MPa [23]. The mechanism of coprecipitation relies upon the condensation of hexaaquo complexes in solution in order to build brucite-like layers having a uniform distribution of both metallic cations and solvated interlamellar anions [7]. Two methods of coprecipitation have been commonly used: precipitation at low supersaturation and precipitation at high supersaturation. 2.1 Precipitation at Low Supersaturation In general, precipitation at low supersaturation is performed by slow addition of mixed solutions of divalent and trivalent metal salts in the chosen ratio into a reactor containing an aqueous solution of the desired interlayer anion. A second solution of an alkali is added into the reactor simultaneously at such a rate as to maintain the pH at a selected value leading to the coprecipitation of the two metallic salts. The rate of addition can be controlled

94

J. He et al.

by manual monitoring of the pH, but the best results are achieved by use of an automatic titration device [7]. The anion that is to be introduced should have a high affinity for the LDH layers and be present in excess, otherwise the counter-anions of the metal salts may be incorporated by competing reactions. Consequently, metal nitrate and chloride salts are commonly used because of the low selectivity of LDHs toward these anions. Furthermore, LDHs have a high affinity for carbonate anions and hence, unless this is the target anion, reactions are generally carried out under nitrogen in order to avoid absorption of atmospheric carbon dioxide which would generate carbonate ions in situ. One advantage of this method is that in many cases it allows careful control of the charge density (M(II)/M(III) ratio) of the hydroxide layers of the resulting LDH by means of precise control of the solution pH. The second advantage is that low supersaturation conditions usually give rise to precipitates with higher crystallinity than those obtained under high supersaturation conditions, because in the former situation the rate of crystal growth is higher than the rate of nucleation. Many kinds of anion-intercalated LDHs have been prepared by precipitation at low supersaturation, including CO3 2– [19, 24–27], NO3 – [28], naphthalene-2,6-disulfonate [29], tetraphenylporphyrins [30], [Ni(edta)]2– [31], an anionic azobenzene derivative (AzAA) [32], and alizarin red S [33]. Prinetto et al. [27] reported the preparation of Mg/Al and Ni/Al LDHs by coprecipitation at a constant pH 9. The samples were found by TEM to exist as aggregates of fibrous particles with lengths up to 300 nm. Bonnet et al. [30] reported the intercalation of meso-tetrakis (o-carboxyphenyl)porphyrin (o-TCPP) into a Zn/Al LDH using the coprecipitation method. In recent years, considerable attention has been paid to the intercalation of biomolecules or pharmaceutical agents into LDHs, such as amino acids [34–36], nucleoside monophosphates and deoxyribonucleic acid (DNA) [37], and vitamins (A, C, E) [38]. LDHs are biocompatible and have found pharmaceutical applications as an antacid, or as an ingredient in sustained-release pharmaceuticals containing nifedipine, for stabilizing pharmaceutical compositions and for preparing magnesium-aluminium salts of antipyretic, analgesic and anti-inflammatory drugs [39]. Aisawa et al. [34] reported the direct intercalation of phenylalanine (Phe) into various LDHs (Mg/Al, Mn/Al, Ni/Al, Zn/Al, Zn/Cr) by coprecipitation at low supersaturation. It was found that the extent of coprecipitated Phe was strongly influenced by the solution pH and reached a maximum in the pH 8–10 region. Some functionalized materials have also been prepared by coprecipitation at low supersaturation. A perylene chromophore, for example, has been intercalated into LDH in an attempt to prepare stabilized pigments [40]. Catalytically active species have also been introduced into the interlayers of LDHs by direct synthesis, e.g. the intercalation of (PW12 O40 )3– or (SiW12 O40 )4– gives catalysts or catalyst precursors containing interlayer polyoxometalate anions [41]. Velu et al. reported the synthesis of Zr-containing LDH-like

Preparation of Layered Double Hydroxides

95

compounds that possess excellent catalytic performance for the selective hydroxylation of phenol to catechol in the liquid phase [42]. The Zr4+ cations were said to be well-dispersed in both uncalcined and calcined samples (at 723 K) [43]. Incorporation of Sn4+ was reported to influence the reducibility of Ni/Al LDHs and Co/Al LDHs, as well as their thermal decomposition products [44]. The reducibility of both Ni- and Sn-containing phases was reported to be enhanced in the materials derived from Ni-containing LDHs and this was attributed to the existence of an interaction between the Nicontaining phase and the SnO2 phase formed in situ in these samples. In contrast, the reducibility of Co and Sn decreased in the materials derived from Co-containing analogues, which was attributed to an enhancement in the polarization of the Co – O bond due to the incorporation of Sn4+ in the Co-containing non-stoichiometric spinel lattice [44]. As discussed in Chapter 1, however, recent work has suggested that the M4+ ions may not actually be incorporated in the LDH layers in any of these materials and that they are in fact mixed phases [45, 46]. LDHs containing variable amounts (0.04–5% by atomic ratio) of different noble metal ions Rh3+ , Ir3+ , Ru3+ , Pd2+ and Pt2+ have been successfully prepared by coprecipitation at pH 10.0 in aqueous solution [47]. Problems arise only for the synthesis of the Pt-containing sample, as a consequence of the strong preference of these cations for square planar coordination, which does not favor their insertion in the octahedral sites of the brucite-type sheets. The presence of noble metal ions inside an inert matrix leads to materials characterized by good thermal stability and relatively high surface areas, which facilitates good dispersion of the noble metal particles after reduction, allowing the resulting materials to be employed as catalysts used under severe reaction conditions, such as those in CO2 -reforming or in partial oxidation of methane. 2.2 Precipitation at High Supersaturation This method requires the addition of a mixed M(II)/M(III) salt solution to an alkaline solution containing the desired interlayer anion. Preparations under conditions of high supersaturation generally give rise to less crystalline materials, because of the high number of crystallization nuclei. Because this method leads to a continuous change in solution pH, the formation of impurity M(OH)2 and/or M(OH)3 phases, and consequently an LDH product with an undesired M(II)/M(III) ratio, often results. Thermal treatment performed following coprecipitation may help increase the crystallinity of amorphous or badly crystallized materials. The first published patent on LDHs as catalyst precursors used high supersaturation conditions. Mg/Al – CO3 , Ni/Al – CO3 , Co/Mn/Al – CO3 , Co/Mn/Mg/Al – CO3 , and Ni/Cr/Al – CO3 LDHs, as well as other bimetallic and multimetallic LDHs were synthesized by adding a solution containing the

96

J. He et al.

salts to an NaHCO3 solution [48]. Many subsequent publications also involved the synthesis of catalysts and catalyst precursors [49, 50]. Constantino and Pinnavaia [51] prepared a series of Mg/Al LDH intercalation compounds containing interlayer OH– , Cl– or SO4 2– anions by precipitation at high supersaturation. In the synthesis of LDHs using the system [(Cu, Zn)/Al – CO3 /SO4 ], impurity phases were observed [52]. Alkoxide-intercalated derivatives of LDHs have been directly synthesized using coprecipitation at high supersaturation from solutions of magnesium and aluminum salts in the presence of sodium hydroxide as a base and an alcohol, preferably methanol or ethanol, as the solvent [53]. The XRD line widths of Mg3 Al LDH products prepared in alcohols were found to decrease with increasing chain length of the alcohol, suggesting that the size of the fundamental particle increases with decreasing acidity of the alcohol. The scattering domain size along the layer stacking direction was estimated from the Scherrer equation to be 8–10 nm for the products prepared with methanol and ethanol, corresponding to crystallites containing stacks of 9–12 LDH layers. The products obtained from methanol and ethanol were formulated as mixed anion intercalates of the type [M(II)1–x M(III)x (OH)2 ]An– (x/n – y) (OR)y · mH2 O, where the alkoxide anion (OR) substitutes for some of the gallery anions. Dispersing the methoxide LDH in water at ambient temperature overnight led to complete hydrolysis of the methoxide ion and the formation of a nearly transparent colloidal LDH suspension. The resulting colloidal suspensions were used as precursors to transparent thin films. Recently, Duan et al. reported the synthesis of M/Fe(II)/Fe(III) LDHs (M = Mg, Co, and Ni) [54, 55], Mg/Fe(III) LDH [56], and [Fe(CN)6 ]3– intercalated Mg/Fe(III) LDHs [57] by a coprecipitation method. These iron-containing LDHs are potential precursors for MFe2 O4 spinel ferrites possessing excellent magnetic properties. In addition to inorganic anions, some organic anions have also been incorporated into the interlayers of LDHs by coprecipitation at high supersaturation, including caprate [58], indole-2-carboxylate (ICTA) [59], 4
-chloro-4stilbenecarboxylate [60], terephthalate [61, 62], aspartate and glutamate [63], polyamino acids [63], dicarboxylates [64], anti-inflammatory drugs [65, 66], and L-aspartic acid [67, 68]. To prepare ICTA-intercalated LDH, for example, a sufficiently high ratio of ICTA to the nitrate anion present in the mother liquor is required [59]. Initial molar ratios of Zn/Al/ITCA of 2/1/6 and 3/1/6 at pH 7.5 were found to be suitable for the formation of ITCAintercalates. Electric linear dichroism (ELD) measurements in combination with XRD data for the LDH complex involving 4
-chloro-4-stilbenecarboxylic acid (CSC) indicated that the CSC species were intercalated as a double layer, with the tilt angle of the molecular plane being ca. 40◦ and the interlayer spacing ca. 3.94 nm [60]. Preferential intercalation of certain isomers of anthraquinone sulfonate ions in Zn/Al LDHs was observed using coprecipitation [69]. In contrast to other processes described in this section,

Preparation of Layered Double Hydroxides

97

the intercalation compounds of anthraquinone sulfonates were synthesized by adding dropwise an aqueous NaOH into a mixed solutions of ZnCl2 , AlCl3 and anthraquinone-1,5-sulfonate (AQ15) and/or anthraquinone-2,6sulfonate (AQ26) and/or anthraquinone-2-sulfonate (AQ2). The ability of intercalation into Zn/Al LDH decreases in the order AQ2 > AQ26  AQ15, probably resulting from the difference in the intermolecular forces between the host and the guest. 2.3 A Method Involving Separate Nucleation and Aging Steps Formation of crystallites involves two stages – nucleation and aging. The processes occurring while a crystal is undergoing aging in its mother liquor are very complex and can involve crystal growth, agglomeration, breakage, and other processes such as Ostwald ripening. Since the mixing process of salt and alkali solutions takes a considerable time in the case of conventional coprecipitation at either high or low supersaturation, nuclei formed at the start of the addition process have a much longer period of time to undergo aging than those formed at the end of the addition process. In other words, nucleation and aging take place simultaneously during the prolonged addition process. The inevitable consequence is that a wide dispersion of crystallite sizes is obtained after aging. It is therefore difficult to control the particle size and distribution of LDHs using the traditional coprecipitation methods. Valim and coworkers [22] have confirmed that, in the case of magnesium-aluminum and zinc-chromium LDH materials containing interlayer terephthalate or dodecyl sulfate anions, a wide distribution of crystallite sizes was obtained using coprecipitation at either variable or constant pH. A new method involving separate nucleation and aging steps was reported by Zhao et al. [20]. The key features of this method are a very rapid mixing and nucleation process in a colloid mill followed by a separate aging process. The design of the colloid mill is schematically illustrated in Fig. 2. The structural and morphological properties of LDH carbonates with different compositions produced using the new method have been compared with those produced using the constant pH low supersaturation method employed by Yun and Pinnavaia [26] and others [3, 9]. The LDH materials produced using the new method have similar chemical compositions and structural parameters as well as thermal behavior to those of LDHs synthesized using the conventional coprecipitation process at constant pH. The method involving separate nucleation and aging steps results in a slightly higher crystallinity of the LDH materials than the conventional process. Most notably, the new method has a significant advantage in that it affords smaller crystallites with a higher aspect ratio, having a very narrow distribution of crystallite size, as can be seen from Figs. 3 and 4 [20]. In contrast to the conventional coprecipitation procedure, in the colloid mill process the mixing

98

J. He et al.

Fig. 2 Schematic illustration of a colloid mill

Fig. 3 TEM micrographs of Mg/Al – CO3 LDHs with different Mg2+ /Al3+ ratios prepared using the new method using rapid mixing and nucleation in a colloid mill followed by a separate aging step (a–c) and conventional coprecipitation at constant pH (d–f). The new method affords smaller crystallites with a much narrower range of length. Reprinted with permission from [20]. Copyright ACS Journal Archives

Preparation of Layered Double Hydroxides

99

Fig. 4 Profiles of particle diameter distribution for Mg/Al – CO3 LDHs with different Mg2+ /Al3+ ratios prepared using (a) the new method using rapid mixing and nucleation in a colloid mill followed by a separate aging step and (b) conventional coprecipitation at constant pH. The new method affords materials with a much narrower range of diameter. Reprinted with permission from [20], Copyright ACS Journal Archives

100

J. He et al.

and nucleation are complete in a very short time and are followed by a separate aging process. It was suggested that the extreme forces to which the nucleation mixture is subjected in the colloid mill prevent aggregation of the nuclei and result in the nuclei having a uniform small size. When the resulting mixture is aged in a separate process, well-formed crystallites with a similarly narrow range of diameters result. This method has also been successfully applied in the synthesis of Cu(II)containing LDHs, although well crystallized materials are difficult to prepare as a consequence of the Jahn-Teller distortion found in the coordination shell of Cu(II) [70, 71]. Incorporation of Ni(II) in the layers was found to improve the crystallinity and structural stability of such LDHs. For the synthesis of Cu/Ni/Al – CO3 and Cu/Ni/Mg/Al – CO3 LDHs [70] by the method with separate nucleation and aging steps, LDHs with both smaller particle size and narrower distribution of particle size were obtained compared with those prepared using a conventional coprecipitation method, similar to the case for Mg/Al – CO3 LDHs [20]. Well crystallized Cu/Ni/Cr – CO3 LDHs [71] were obtained when the Cu/Ni/Cr atom ratio ranged from 1 : 2 : 1 to 1 : 3 : 1 in the reaction mixture with hydrothermal aging conditions at 180 ◦ C for 4 h. This method has proved to be very readily scaled-up [72, 73] and is now being used to produce LDHs commercially in China. 2.4 Urea Hydrolysis Method Urea has a number of properties that makes its use as an agent for precipitation from “homogeneous” solution very attractive, and it has long been used in gravimetric analysis to precipitate several metal ions as hydroxides or as insoluble salts when in the presence of a suitable anion [74]. Urea is a very weak Brønsted base (pKb = 13.8), highly soluble in water, and its hydrolysis rate may be easily controlled by controlling the temperature of the mixture. Hydrolysis of urea proceeds in two steps, the formation of ammonium cyanate (NH4 CNO) being the rate determining step, with subsequent fast hydrolysis of the cyanate to ammonium carbonate: CO(NH2 )2 → NH4 CNO NH4 CNO + 2H2 O → 2NH4 + + CO3 2– The hydrolysis reactions of ammonium ions to give ammonia and carbonate to give hydrogen carbonate result in a pH of about 9, depending on the temperature. This pH is suitable for precipitating a large number of metal hydroxides. Costantino et al. [75] prepared M(II)/Al-carbonate LDHs (M(II) = Mg, Zn and Ni) by this method. The effects of varying the temperature, total metal cation concentration, molar fraction Al/Al+M(II) and molar fraction

Preparation of Layered Double Hydroxides

101

urea/M(II)+Al in solution on the composition and the crystallinity of the samples were investigated. The optimum conditions to prepare LDHs with a good crystallinity in a relatively short time using a simple procedure were suggested to involve dissolving solid urea in a 0.5 M solution of the chosen metal chlorides [molar ratio Al/Al+M(II) equal to 0.33] to give a urea/metal ion molar ratio of 3.3. The clear solution was maintained at 100 ◦ C for 36 h. The urea method is not suitable for the preparation of Mg/Al LDHs with low charge density but allows the preparation of compounds with high charge density not easily obtainable using other procedures [21]. If the M(II) hydroxide is much more insoluble than Mg(OH)2 , the LDH obtained has a composition near to that predicted. This is the case, for example, for the Zn/Al and Ni/Al systems, in which the Zn/Al or Ni/Al molar ratio in the solid was found to be very near to that in the original solutions. It was found [76] for Mg/Al-carbonate LDHs that better crystallinity was observed as the aging time was prolonged and the total metal concentration was decreased, as can be seen from Fig. 5. Since the decomposition rate of urea in aqueous solution depends on the temperature, the particle size distribution was controlled by altering reaction temperature. Larger particles were formed at lower temperatures due to the lower nucleation rate [77]. Larger particle size was obtained using the urea method than by other methods because the hydrolysis of urea proceeds very slowly, which leads to a low degree of supersaturation during precipitation. Thus the resulting LDH materials generally give PXRD patterns having intense and narrow peaks. Adachi-Pagano et al. [78] prepared mono-disperse sub-micron sized Mg/Al LDHs by the urea hydrolysis method, and the results were compared to the constant pH coprecipitation method. The compounds prepared using the urea method displayed platelet-like primary particles with a hexagonal shape. Changing the synthesis conditions such as the precipitation rate or the solvent medium led to a variation in both the particle size and its distribution. Using ethylene glycol (EG) as medium together with water allows control of the range of particle sizes, which depends on the water/EG ratio. In particular, sub-micron primary particles are obtained for water/EG = 1/4. A narrow particle size distribution is obtained for high reaction temperatures, low Mg/Al ratio and high urea/metal ion ratio. When the urea method is carried out in water/EG = 1, formation of Mg/Al LDHs with a higher Mg/Al ratio is obtained but the particle size is much smaller. Studies have shown that the urea method is not suitable for preparing LDHs containing Cu(II) or Cr(III) [75] and it remains to be seen how widely applicable it will be. Although LDHs with a homogeneous size and a welldefined hexagonal shape can be prepared by the urea method, which is very interesting from the viewpoint of nanotechnology because the LDHs offer nano-size two-dimensional spaces for creation of functional materials, the LDHs prepared by the urea method usually contain carbonate ions which cannot be generally deintercalated because of their high affinity for LDHs.

102

J. He et al.

Fig. 5 Scanning electron microscope images for LDHs synthesized by the urea method with a aging times 6, 30, 45, and 69 h, and b the concentration of metal ions, 0.87, 0.65, 0.44, and 0.06 M (scale bar = 2.5 µm). Reprinted with permission from [76], Copyright Elsevier

Costantino et al. [75] have shown, however, that treatment with gaseous or aqueous HCl can lead to replacement of carbonate with chloride, which itself can be replaced by large organic anions such as methyl orange [79] or fluorescein [80]. It has recently been suggested that use of HCl/NaCl mixtures also leads to more effective displacement of carbonate by chloride [81]. Another reagent, hexamethylenetetramine (HMT) has also been used in place of urea for the homogeneous preparation of LDHs. HMT hydrolyzes at high temperature in aqueous solution with the release of ammonia, which makes the solution alkaline, and formaldehyde, which would not be expected to be incorporated into the LDH. Using HMT, well-crystallized 1–5 µm – sized particles of chloride-intercalated LDHs were prepared in a pressure

Preparation of Layered Double Hydroxides

103

vessel after a 1-day treatment at 140 ◦ C [82]. The resulting LDHs contained some carbonate anion impurity, which was successfully deintercalated using an NaCl/HCl mixed solution without any morphological change.

3 Ion-exchange Method The ion-exchange method is especially useful when the coprecipitation method is inapplicable such as when, for example, the divalent or trivalent metal cations or the anions involved are unstable in alkaline solution, or the direct reaction between metal ions and guest anions is more favorable. In this method, the guests are exchanged with the anions present in the interlayer regions of preformed LDHs, as shown in Fig. 6, to produce specific anion pillared LDHs. In thermodynamic terms, ion exchange in LDHs depends mainly on the electrostatic interactions between the positively-charged host sheets and the exchanging anions and, to a lesser extent, on the free energy involved in the changes of hydration [83, 84].

Fig. 6 Schematic course of an ion exchange reaction

Ordinarily, the ion-exchange method is carried out in one of two processes shown schematically as follows: LDH · Am– + Xn– → LDH · (Xn– )m/n + Am– or LDH · Am– + Xn– + mH+ → LDH · (Xn– )m/n + Hm A In the first process, the precursor contains univalent anions, such as chloride, nitrate or perchlorate, which have a weak electrostatic interaction with the layers. In the second process, the precursor contains anions susceptible to acid attack, such as carbonate or carboxylates, e.g. terephthalate. This method was first proposed by Bish [85], who demonstrated the anion exchange of carbonate by chloride, nitrate, bromide and sulfate. There are several factors which determine the extent of ion-exchange in any given case:

104

J. He et al.

1) Affinity for incoming anion Generally, the exchange ability of incoming anions increases with increasing charge and decreasing ionic radius. The order for simple inorganic anions decreases in the order CO3 2– > HPO4 2– > SO4 2– for divalent anions and OH– > F– > Cl– > Br– > NO3 – > I– for monovalent anions. The co-intercalation of a second anion was found to have no effect on the order of ion exchange preference [86]. Because nitrate is exchanged most easily, nitrate pillared LDHs are usually used as the precursors for ion exchange [87]. 2) Exchange medium The interlayer space of LDHs can be expanded to some extent in a suitable solvent medium, which favors the ion exchange process. An aqueous medium, for example, favors the exchange by inorganic anions, whilst an organic solvent favors exchange by organic anions [14]. 3) pH value For anions such as terephthalate or benzoate which are conjugate bases of weak acids, the lower the pH of the reaction solution, the weaker the interaction between the layers and interlayer anions [88]. A low pH value therefore favors liberation of the original anion as the conjugate acid and incorporation of a less basic anion from solution. The pH value should not be lower than ∼ 4.0, however, because the basic LDH sheets themselves would begin to dissolve. 4) Chemical composition of the layers The chemical composition of the LDH sheets influences the charge density of the sheets and the hydration state, thereby affecting the ion exchange process. Some other factors such as temperature also have an impact on the ion exchange process. It is usually accepted that higher temperatures favor ion exchange [89]. It should be noted, however, that too high a temperature might have an adverse effect on the structural integrity of the LDHs. The ion exchange method is especially useful for the preparation of noncarbonate LDHs. A large number of organic and inorganic anions have been incorporated in LDHs using the ion-exchange process. The organic anions intercalated include carboxylates [83], surfactant ions [90], phosphonates [91], β-cyclodextrin detratives [92], poly(ethylene oxide) derivatives [93, 94], polystyrene sulfonate [95], pharmaceutically active species [96, 97], biomolecules [98], amino acids [99], glyphosate anions [100], and dye anions [79, 80]. Intercalated inorganic anions include metal oxo species [101, 102], polyoxometalates [8, 9], phosphate ions [103, 104], and metal complex ions [105–114]. It was found [9] that the synthesis of pure crystalline LDHs intercalated with (H4 Co2 Mo10 O38 )6– required a compromise between reaction conditions where the hydrolytic stability of the anion is maximized and those where the layers are most stable to hydrolysis. The optimum conditions for intercala-

Preparation of Layered Double Hydroxides

105

tion were achieved using meixnerite-type precursors, which have interlayer hydroxide anions, with Al/(Mg + Al) molar ratios larger than 0.33, a reaction pH of about 4.7, and a nonaqueous medium (ethanol/water 40 vol%) at 353 K, with an exchange time of 1 h [9]. The interlayer arrangement of organic anions depends strongly on the area available to each interlayer anion [28]. A small available area was found to force alkyl chain compounds (fatty acid anions, dicarboxylates and alkyl sulfates) to point away from the interlayer surface and to form monolayer films of high regularity. Secondary alkane-sulfonates aggregated in the interlayer spaces forming bilayer films of constant thickness, and well ordered organoLDHs resulted. In some cases, shape-selective intercalation of isomers [115–117] has been observed in the ion-exchange process. 1,4-benzenedicarboxylate and fumarate anions were selectively intercalated intoLDHs from equimolar mixtures of 1,2-, 1,3-, and 1,4-benzene-dicarboxylates, and fumarate and maleate mixtures, respectively. Time-resolved in situ energy-dispersive X-ray diffraction studies described in detail in Chapter 4 have shown that these are the thermodynamic products resulting from the initial insertion of all the isomers followed by collapse to the most stable intercalate phase by expulsion of the disfavored anions. Second staging structures as intermediate phases during the intercalation of organic guests by ion-exchange [32, 88, 118, 119] have been observed, as discussed in Chapter 4. Phase segregation during the intercalation of hexacyanoferrate(III) in the LDH-chloride system [114] has also been observed in the ion exchange process. The reactivity toward grafting of LDHs intercalated by organic aromatic anions using ion-exchange or coprecipitation methods has been investigated under thermal treatment [120]. Strong contractions of the basal spacing were observed for hybrid phases arising from the reaction of hydroxyl groups of the host and either the anionic functionality or a hydroxyl group substituted on the aromatic molecules. An alternative method of anion exchange in LDHs based on the formation of a salt between an anionic and a cationic surfactant, as shown in Fig. 7, was reported by Crepaldi et al. [121, 122]. This method makes use of an LDH precursor intercalated with an anionic surfactant, for example, dodecylsulfate (DS) or dodecylbenzenesulfonate (DBS). The anion exchange can be promoted by the addition of a cationic surfactant, for example, N-cetylN,N,N-trimethylammonium bromide, (CTAB), to a mixture containing the precursor suspended in an aqueous solution of the anion of interest in contact with an immiscible organic phase (e.g. chloroform). The exchange is achieved by the formation of a water insoluble salt between the cationic surfactant and the anionic surfactant of the precursor. The surfactant salt is rapidly removed from the aqueous suspension to the organic phase. Such an exchange mechanism makes this method easy and fast to carry out. It was shown that the exchange of dodecylsulfate by a set of different anions (chloride, carbon-

106

J. He et al.

Fig. 7 Scheme illustrating anion exchange by means of the formation of a salt between an anionic and a cationic surfactant in LDHs. Reprinted with permission from [122]. Copyright RSC

ate, terephthalate, cholate and a sulfonated Cu-phthalocyanine) was achieved within 30 minutes with efficiency higher than 98.5%. The materials obtained showed phase purity and relatively high crystallinity. The precursors are also easy to synthesize and no further treatment prior to use in the exchange reaction are required.

4 Rehydration Using Structural Memory Effect Calcination of LDHs removes the interlayer water, interlayer anions and the hydroxyl groups, resulting in a mixed metal oxide that cannot be achieved by mechanical means. It is especially interesting that the calcined LDH is able to regenerate the layered structure when it is exposed to water and an-

Preparation of Layered Double Hydroxides

107

ions [123–126]. Water is absorbed to reform the hydroxyl layers, and anions and water are incorporated into the interlayer galleries. The anions incorporated do not necessarily need to be the anion that was in the original LDH material, and thus this is an important method to obtain various inorganic and organic anion forms of LDHs as well as pillared structures [11, 127– 141]. Organic chromophores [129], surfactants [130], ε-caprolactam [131], a sulfonated spiropyran dye [134], the herbicide glyphosate [135], naphthalenedisulphonates [138], metal complex anions [137, 139], amino acids and peptides [140], hexose [141] and even non-ionized guest pentoses [133] have been incorporated through such rehydration processes. The conversion of the mixed metal oxides into LDHs has been variously referred to as regeneration, reconstruction, restoration, rehydration or the “calcination-rehydration process”, “structural memory effect” or simply “memory effect”. This method is usually employed when large guests are intercalated. It also avoids the competitive intercalation of inorganic anions arising from the metal salts. The procedure is more complicated than coprecipitation or ion-exchange methods, however, and amorphous phases are often produced simultaneously. It should be noted that both the calcination temperature and chemical composition of the LDH sheets have significant influence on the reconstruction process. The “memory effect” is reduced on increasing the calcination temperature of the parent LDH, because increased calcination temperature causes the solid-state diffusion of divalent cations into tetrahedral positions, which results in the progressive formation of stable spinels. After calcination of a Cu/Co/Zn/Al LDH at 873 K, for instance, all the divalent cations occupy tetrahedral sites and the “memory effect” is lost [125]. For Mg/Al LDHs [126], PXRD and 27 Al NMR revealed that calcination at increasing temperatures leads to migration of Al3+ ions to tetrahedral sites while the crystalline structure is progressively destroyed. Upon rehydration with water vapour, the layered LDH structure is recovered to varying extents, depending on the previous calcination temperature, as well as on the rehydration time at 25 ◦ C. Reconstruction is complete after rehydration for 24 h when the sample has been calcined at or below 550 ◦ C, while equilibration for 3 days is required to reconstruct samples previously calcined at 750 ◦ C. Only partial reconstruction is observed after calcination at 1000 ◦ C. However, solid state 27 Al MAS NMR indicates that tetrahedral Al and spinel phases still exist in the solids for which PXRD shows a complete recovery of the LDH structure. An inert nitrogen atmosphere is required during the reconstruction process when a non-carbonate anion is incorporated by rehydration because the carbonate LDH generally forms preferentially in the presence of atmospheric CO2 . The reconstruction method fails to give LDH materials if an anion such as tartrate, which has a strong tendency to complex metal cations, is used [142].

108

J. He et al.

5 Hydrothermal Methods When organic guest species with low affinity for LDHs are required to be intercalated into the interlayers, anion-exchange reactions using presynthesized LDHs as precursors or coprecipitation using soluble metal salts such as chlorides and nitrates are not applicable. Hydrothermal synthesis has been shown to be effective in such cases because insoluble hydroxides, for example magnesium and aluminium hydroxides can be used as the inorganic sources ensuring that the desired anions occupy the interlayer space since no other competing anions (apart from hydroxide, which has a very low affinity) are present [143]. Hydrothermal synthesis has been used to control the particle size and its distribution when soluble magnesium and aluminium salts were used together with alkali solution to prepare Mg/Al-carbonate LDHs [76, 144], but is particularly useful when LDH-like materials are prepared using powders as starting materials. Single crystals of a layered compound with the formula 3CaO · Al2 O3 · CaCO3 · 11H2 O were synthesized by a hydrothermal process using Ca(OH)2 , Al(OH)3 and CaCO3 as starting materials, with a molar ratio of 3.5 : 2 : 0.5 [145]. It was found that the hydrothermal temperature influenced the crystal structure of the resulting material. An ordered structure was obtained at 120 ◦ C, and a disordered one at 100 ◦ C. Both disordered and ordered structures have identical main layers [Ca4 Al2 (OH)12 ]2+ , but they have different arrangements of water molecules and carbonate groups forming the interlayer region of composition (CO3 · 5H2 O)2– . This leads to major differences in respect of the stacking of the slabs as discussed in Chapter 1.

6 Secondary Intercalation (Pre-pillaring Method) This method is a development of the ion-exchage method. From a kinetic point of view, the rate-determining step of the ion-exchange reaction is the diffusion of the incoming anions within the interlayer, provided that “infinite solution conditions” are respected [7]. The diffusion of large anions into the interlayer can be inhibited if the basal spacing of the precursor is too small. It is therefore difficult to intercalate large guest anions directly into the interlayer regions of LDHs. Species with low charge density are another class of guests difficult to intercalate as a result of the reduced driving force associated with the weak interaction between the anions and host layers. In such cases, pre-intercalation by smaller guests using coprecipitation or ionexchange methods is an effective way to enlarge the interlayer space, so that

Preparation of Layered Double Hydroxides

109

it is possible to introduce large or low charge target guests into the interlayer regions by ion-exchange, encapsulation or chemical bonding. Using an LDH with dodecyl sulfate counteranions as precursor, C60 molecules [146] and poly(ethylene oxide) [147] have been introduced into LDH interlayers. Neutral C60 molecules were intercalated by dissolving the molecules into the interlayer hydrophobic phase of the surfactant-pillared LDH. After heating the resultant compound under vacuum to decompose the dodecyl sulfate, C60 molecules were left sandwiched between the double hydroxide layers. The simultaneous intercalation of dodecyl sulfate and poly(ethylene oxide) resulted in a new nanocomposite. A three-step process was used for the immobilization of penicillin G acylase (PGA) in the interlayer galleries of a layered double hydroxide (LDH). Glutamate-pillared LDH was first prepared. Subsequent reaction of the amino groups of the immobilized glutamate ions with a glutaraldehyde linker followed by addition of PGA affords the immobilized enzyme (IME) [148]. Inorganic polyoxometalatepillared LDH derivatives have also been prepared using organic anionintercalated LDHs as precursors [61, 149]. Finally, a dioxomolybdenum(VI) complex MoO2 Cl2 (THF)2 [150] was intercalated into the interlayer galleries of Zn/Al LDH using 2,2-bipyridine-5,5-dicarboxylate anion-pillared LDH as a precursor.

7 An Intercalation Method Involving Dissolution and Re-coprecipitation Procedures This method allows the preparation of carboxylate-intercalated LDHs using carbonate-containing LDHs as precursors. It is a simple procedure involving dissolution of the precursor by adding an aqueous solution of the appropriate carboxylic acid, followed by re-precipitation on mixing with a basic solution. This method avoids the problem of competitive intercalation by carbonate or other anions arising from atmospheric carbon dioxide or metal salts respectively, which is often observed with conventional methods such as coprecipitation, ion-exchange and rehydration. In contrast to other synthetic procedures for intercalated LDHs, it does not need to be carried out under nitrogen or other inert atmosphere. Glutamate [148], citrate, oxalate, tartrate and malate anion-pillared [151] LDHs have been prepared by this procedure. Typically, carbonate-containing LDHs are first dispersed in distilled water at the chosen temperature, typically 50 ◦ C. Addition of the carboxylic acid results in the suspension dissolving slowly with effervescence giving a clear solution. This solution was added dropwise to an alkaline solution maintaining the pH above 9, followed by refluxing of the mixture. The absence of any co-intercalated carbonate ions was confirmed by elemental analysis, IR spectroscopy and the lack of evolution of carbon dioxide on reaction with dilute acid.

110

J. He et al.

8 Other Methods 8.1 Salt-oxide (or Hydroxide) Method Preparation of LDHs using salts and oxides as metal sources was first proposed by Boehn et al. [152] for the preparation of Zn/Cr-Cl LDHs. Cu/Cr – Cl LDHs, which are difficult to obtain by coprecipitation, were subsequently synthesized using a similar process from Cu(II) oxide and chromium(III) chloride [153]. An Mg/Al LDH was recently prepared from magnesium oxide [154] or magnesium hydroxide [155] and sodium aluminate. The sodium aluminate can be a commercial product [154] or prepared from aluminum hydroxide [155] dissolved in an aqueous solution of sodium hydroxide. Isupov et al. [156] synthesized Mg/Al LDHs via mechanical activation of a mixture of magnesium hydroxide and an aluminium salt using a high-energy planetary-type AGO-2 activator. Aging of trivalent metal hydroxide/oxide gels in divalent metal salt solutions also led to the formation of LDHs. It was found that aging of freshly precipitated Al(OH)3 gels in solutions of Mg2+ and Ni2+ salts leads to LDH formation at high pH (> 12), whilst aging of “Fe(OH)3 ” leads to LDH formation in Mg2+ salt solutions but not in Ni2+ salt solutions [157]. 8.2 Non-equilibrium Aging Method Non-equilibrium crystallization was developed [158] to operate in tandem with the method involving separate nucleation and aging steps discussed in Sect. 2.3. In a general coprecipitation process, prolonged aging time scarcely influences the crystal growth after a period of crystallization at a certain temperature, due to thermodynamic restrictions. In the non-equilibrium aging or crystallization method, the salt and alkali solutions are rapidly mixed and nucleated in a colloid mill, followed by an aging step. After a period of aging, another portion of salt and alkali solutions is supplied simultaneously to the aging mixture, to ensure that the metal ions in the solution are always supersaturated. As is well known [159], increasing concentration and/or decreasing temperature favor nucleation over crystal growth. Thus a high temperature and slow rate of introduction favor the adsorption of the added metal ions on the preformed crystal particles, rather than the formation of new nuclei. It was shown that Mg/Al – CO3 LDHs with a stable composition and increased crystallinity can be prepared by this method [158]. The particle size and its distribution was controlled to some extent by changing the quantities of either salt and alkali introduced.

Preparation of Layered Double Hydroxides

111

8.3 Non-conventional Aging Methods Non-conventional aging techniques such as microwaves and ultrasound have been employed in chemical reactions because of their unique features which make them different from conventional energy sources. The effect of microwave irradiation on chemical reactions is a combination of thermal effects which result from absorption of energy and transmission by diectric loss, as well as non-thermal effects, the origin of which is still somewhat controversial [160]. The preparation of LDHs using microwave sources has been found to be rapid and convenient [161, 162]. Similar crystallinities and chemical compositions for Mg/Al and Mg/Ga LDH samples were obtained by microwave irradiation and conventional aging of synthesis gels. The microwave aging technique allows the use of highly concentrated solutions and reduces the synthesis time considerably [162]. Similarly, ultrasound has also been used in the synthesis of LDHs. The LDHs synthesized under ultrasonic conditions showed a larger crystallite size and a larger adsorption capacity of humic substances than those synthesized without ultrasonic treatment [163]. 8.4 Surface Synthesis In situ synthesis of a compound or functional material on the surface of a support is usually used to prepare hybrid or composite materials. Not only are the properties of the adsorbate including mechanical performance, thermal stability, and degree of dispersion significantly enhanced by being supported or loaded on a surface by physical or chemical means, but the resulting hybrid or composite materials can also be expected to have the merits of both the compound itself and the support. The synthesis of LDHs on the surface of α-Al2 O3 was reported as early as 1983 [164]. Ni/Al LDHs were synthesized in situ using the urea hydrolysis procedure on the surface of α-Al2 O3 . Co, Ni, Zn-containing LDHs were prepared on the surface of γ -Al2 O3 by impregnating the support with Co2+ , Ni2+ , or Zn2+ solution, followed by ventilation with ammonia gas to control the pH between 7–8.2 [165–167]. Mao et al. [168] have prepared MII /Al LDHs on the surface of γ -Al2 O3 using aqueous ammonia and urea as precipitatants as well as for pH control. Nickel and magnesium nitrates were used as divalent metal sources and the “activated” surface Al sites were utilized as the source of the trivalent cations. 8.5 Templated Synthesis Template-directed synthesis has been the object of much recent attention in the field of material science [169]. Using self-assembled aggregrates as a tem-

112

J. He et al.

plate, inorganic minerals or materials can be directed to an ordered structure with specific morphology and size by replication of the structure of the selfassembled aggregrates. Typically, reactions of inorganic precursors take place at the interface between the organic self-assembled aggregrates and the solution, forming inorganic-organic composites. Removal of the organic template from the composites produces inorganic materials with some structure and morphology. The size and size distribution of the inorganic particles can also be controlled or tailored by template synthesis. The templates commonly employed for the morphology and size control of inorganic materials include peptides, polysaccharides, amphiphilic block polymers, Langmuir-Blodgett films, as well as microemulsions, vesicles, and micelles consisting of surfactants. There have been only a few reports of the synthesis of LDHs using templates to date because control and tailoring of the morphology and particle size of LDH is much more difficult than for materials such as CaCO3 and BaSO4 . The Langmuir-Blodgett method has been used to prepare hybrid films of an anionic Ru(II) cyanide polypyridyl complex with LDHs [170]. An LDH film was formed on mica owing to the interaction between LDHs particles and the Ru(II) cyanide polypyridyl complex that was pre-dispersed on the surface of mica. Water-in-oil emulsions composed of octane, water and sodium dodecyl sulfate (SDS) have been used to synthesize Mg/Al LDHs with carbonate as the interlayer anion [171] by constant pH or variable pH methods. A floccule or fiber-like LDH material that possesses similar chemical composition and properties to that synthesized using a conventional variable pH method was obtained. The resulting LDH shows high surface area and a narrow distribution of mesopores. 8.6 Miscellaneous Methods In addition to those mentioned above, a variety of other methods have also been employed to prepare LDHs. Sol-gel method LDHs with Ni/Al molar ratio of 2.5 have been synthesized by a sol-gel method using nickel acetylacetonate and aluminium isopropylate as precursors. The sol-gel synthesized samples exhibited lower crystallite dimensions and greater BET surface areas compared with coprecipitated samples prepared from an aqueous solution of nickel and aluminium nitrates [19, 172]. Electrosynthesis LDHs containing nickel with trivalent cations such as aluminum, chromium, manganese and iron, have been synthesized electrochemically [173]. The electrosynthesized LDHs have similar structural features to both the chemically synthesized LDHs and α-nickel hydroxide. A new Fe(II)/Fe(III)-

Preparation of Layered Double Hydroxides

113

carbonate LDH, of the type generally known as a green rust (GR), has also been electrochemically synthesized using iron as the starting material [174]. It was found that GRs prepared by chemical oxidation of Fe(OH)2 and by the electrochemical method differ in their values of unit cell parameter a and c and Fe(II)/Fe(III) ratio. It was proposed that the electrosynthesized GR results from a coprecipitation of Fe(II) and Fe(III) species generated by oxidation of the metallic iron that was used as starting material. NH3 · H2 O as a base LDH materials with nitrate anions in the interlayers could serve as a potential candidates for the development of slow-release nitrate fertilizers. However, as noted above, the preparation of such materials with nitrate has been found to be quite difficult because of the very high affinity of LDH layers for CO3 2– anions which are ubiquitous due to the CO2 in the atmosphere. The synthesis of the nitrate form of LDHs therefore generally has to be carried out using degassed solutions along with an inert atmosphere. A simplified synthesis of the NO3 – forms of Mg/Al LDHs with Mg/Al ratios of 2 : 1, 2.5 : 1 and 3 : 1 using ammonium hydroxide as a base has been developed by Olanrewaju et al. [175]. No incorporation of CO3 2– was observed using this approach even though an inert atmosphere was not employed. In situ oxidation of MII An LDH with the approximate stoichiometry Mg0.3 Co(II)0.6 Co(III)0.2 (OH)2 (NO3 )0.2 · H2 O has been synthesized by oxidation of Co(II) using an ammoniacal solution and hydrothermal treatments under various O2 : N2 atmospheres [176]. The ammoniacal solution plays a number of roles in the synthesis. Firstly, it provides a basic medium. Secondly, it acts as ligand by forming a complex [Co(NH3 )6 ]2+ , which facilitates oxidation of Co2+ to [Co(NH3 )6 ]3+ because of the low standard reduction potential (E◦ ): [Co(NH3 )6 ]3+ + e = [Co(NH3 )6 ]2+

E◦ = 0.11 V

On the other hand, Co(OH)2 can be oxidized to Co(OH)3 [or CoO(OH)] within the solid phase in basic media: Co(OH)3 (s) + e = Co(OH)2 (s) + OH–

E◦ = 0.17 V

or CoO(OH)(s) + H2 O + e = Co(OH)2 (s) + OH–

E◦ = 0.17 V

It was found that 23% of Co2+ was oxidized to Co3+ . The conversion of a brucite-like material to an LDH structure was observed in response to an increase in O2 partial pressure. Anionic species, such as NO3 – , were intercalated between the brucite-type layers of LDHs and the extent of anion intercalation increased as a function of O2 partial pressure.

114

J. He et al.

Mg/FeIII LDHs have been synthesized by Hansen et al. [177, 178] by oxidation of Mg/FeII solutions with air and the oxidation of freshly precipitated Mn(OH)2 in an Mg2+ solution has been used by the same author to prepare Mg/MnIII LDHs [179]. The key in these reactions is careful control of pH during the oxidation since the FeIII /FeII and MnIII /MnII redox couples are very sensitive to pH. By controlling the rate at which the trivalent cation is formed, highly crystalline LDH materials without stacking faults can be obtained as discussed in Chapter 1. Although limited to LDHs where the trivalent cations can be prepared by oxidation, a wide variety of anions can be incorporated using this method in for example LDHs of the GR type [Fe(II)4 Fe(III)2 (OH)12 ]2+ (An– )2/n · mH2 O (An– = Cl– , SO4 2– , CO3 2– oxalate etc.) [180]. “Chimie douce” method Using coprecipitation methods with a suitable mixture of solutions described above, the resulting LDH materials are often poorly crystallized and exhibit compositional fluctuations due mainly to the difference in the values of the pH at which the precipitation of M(II)(OH)2 and M(III)(OH)3 hydroxides occurs. Consequently, the chemical formula of the final material may not reflect the composition of the solution prior to the precipitation as noted in Chapter 1. Controlling the amount of anion incorporated under such conditions is very difficult. A “chimie douce” method has been proposed by Delmas et al. in an effort to overcome this problem [181, 182]. The process is illustrated schematically in Fig. 8. Since the synthesis starts from a highly crystalline layered γ -oxyhydroxide precursor, it was suggested that this favored the formation of very crystalline LDHs with controllable M(II)/M(III)

Fig. 8 Schematic illustration of the successive reaction steps in the preparation of LDH by “chimie douce”. Reprinted with permission from [182]. Copyright Elsevier

Preparation of Layered Double Hydroxides

115

ratios. Using this method, LDHs with simple inorganic anions of the type [Ni1–x Cox (OH)2 ]Xx/n · mH2 O (Xn– = CO3 2– , SO4 2– , NO3 – , OH– ) [181, 182] and as well as with a large anionic metavanadate as the guest anion have been prepared. Although a wide variety of anions can be incorporated, this method is only likely to be applicable to a small range of layer compositions.

9 Summary and Outlook This review has demonstrated the wide variety of methods that are available for the synthesis of LDHs. Clearly the method of choice will depend on the purpose for which the LDH is to be used. If accurate structural information is required, then pure phases with high crystallinity are necessary. Possible methods involve hydrothermal synthesis starting from soluble or insoluble [143, 145] precursors, gradual increase of pH to produce a carbonateintercalated LDH which can be subsequently exchanged by first Cl– and then the anion of interest [79–81], slow formation of MIII cations by controlled oxidation of MII precursors [177–179] or “chimie douce” methods [181, 182]. In contrast, if the LDHs are to be employed as catalysts or catalyst precursors, low surface areas and absence of structural or compositional defects associated with highly crystalline materials are probably disadvantageous and coprecipitation or other methods are generally more appropriate. Finally in view of the increasing interest in actual and potential applications of LDHs discussed in Chapter 5, it is useful to consider the ease of scaling up any new synthetic procedure if larger quantities of material should be required.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Martin KJ, Pinnavaia TJ (1986) J Am Chem Soc 108:541 Manasse E (1915) Atti Soc Toscana Sc Nat Proc Verb 24:92 Cavani F, Trifirò F, Vaccari A (1991) Catal Today 11:173 Allmann R (1968) Acta Crystallogr B 24:972 Taylor HFW (1973) Mineral Mag 39:377 Rives V, Ulibarri MA (1999) Coordin Chem Rev 181:61 de Roy A, Forano C, Besse JP (2001) In: Rives V (ed) Layered Double Hydroxides: Present and Future. Nova Science Publishers Inc, New York p 8 Gardner EA, Yun SK, Kwon T, Pinnavaia TJ (1998) Appl Clay Sci 13:479 Bravo-Suárez JJ, Páez-Mozo EA, Oyama ST (2004) Chem Mater 16:1214 Sels B, de Vos D, Buntinx M, Pierard F, Kirsch-De Mesmaeker A, Jacobs P (1999) Nature 400:855 Vaccari A (1999) Appl Clay Sci 14:161 Rousselot I, Taviot-Guého C, Besse JP (1999) Int J Inorg Mater 1:165 Ogawa M, Kuroda K (1995) Chem Rev 95:399 Newman SP, Jones W (1998) New J Chem 105

116

J. He et al.

15. Tagaya H, Sato S, Kuwahara T, Kadokawa J, Masa K, Chiba K (2002) J Mater Chem 4:1907 16. Newman SP, Jones W (2001) In: Rao CNR, Jones W (eds) Supramolecular Organization and Materials Design. Cambridge University Press, Cambridge UK p 295 17. Fogg AM, Green VM, Harvey HG, O’Hare D (1999) Adv Mater 11:1466 18. Millange F, Walton RI, Lei LX, O’Hare D (2000) Chem Mater 12:1990 19. Aramendia MA, Borau V, Jimenez C, Marinas JM, Ruiz JR, Urbano FJ (2002) J Solid State Chem 168:156 20. Zhao Y, Li F, Zhang R, Evans DG, Duan X (2002) Chem Mater 14:4286 21. Bravo-Suárez JJ, Páez-Mozo EA, Oyama ST (2004) Quim Nova 27:601 22. Crepaldi EL, Pavan PC, Valim JB (2000) J Braz Chem Soc 11:64 23. Rabenau A (1985) Angew Chem Int Ed 24:1026 24. Corma A, Fornes V, Martin-Aranda RM, Rey F (1992) J Catal 134:58 25. Hibino T, Yamashita Y, Kosuge K, Tsunashima A (1995) Clays Clay Miner 43:427 26. Yun SK, Pinnavaia TJ (1995) Chem Mater 7:348 27. Prinetto F, Ghiotti G, Graffin P, Tichit D (2000) Micropor Mesopor Mater 39:229 28. Meyn M, Beneke K, Lagaly G (1990) Inorg Chem 29:5201 29. Kanezaki E (1999) Mater Res Bull 34:1435 30. Bonnet S, Forano C, de Roy A, Besse JP (1996) Chem Mater 8:1962 31. Tsyganok AI, Suzuki K, Hamakawa S, Takehira K, Hayakawa T (2001) Chem Lett 24 32. Iyi N, Kurashima K, Fujita T (2002) Chem Mater 14:583 33. Kanezaki E (2003) J Incl Phen Macro Chem 46:89 34. Aisawa S, Takahashi S, Ogasawara W, Umetsu Y, Narita E (2001) J Solid State Chem 162:52 35. Fudala A, Palinko I, Hrivnak B, Kiricsi I (1999) J Therm Anal Calorim 56:317 36. Aisawa S, Takahashi S, Ogasawara W, Umetsu Y, Narita E (2000) Clay Sci 11:317 37. Kwak SY, Jeong YJ, Park JS, Choy JH (2002) Solid State Ionics 151:229 38. Hwang SH, Han YS, Choy JH (2001) Bull Korean Chem Soc 22:1019 39. Ambrogi V, Fardella G, Grandolini G, Perioli L (2001) Int J Pharm 220:23 40. Bauer J, Behrens P, Speckbacher M, Langhals H (2003) Adv Funct Mater 13:241 41. Kagunya W, Hassan Z, Jones W (1996) Inorg Chem 35:5970 42. Velu S, Veda R, Ramani A, Chanda BM, Sivasanker S (1997) Chem Commun 2107 43. Velu S, Sabde DP, Shah N, Sivasanker S (1998) Chem Mater 10:345 44. Velu S, Suzuki K, Kapoor MP, Tomura S, Ohashi F, Osaki T (2000) Chem Mater 12:719 45. Intissar M, Jumas JC, Besse JP, Leroux F (2003) Chem Mater 15:4625 46. Intissar M, Holler S, Malherbe F, Besse JP, Leroux F (2004) J Phys Chem Solids 65:453 47. Basile F, Fornasari G, Gazzano M, Vaccari A (2000) Appl Clay Sci 16:185 48. French Patent 2,091,785 (1971) to BASF AG 49. Reichle WT, Kang SY, Everhardt DS (1986) J Catal 101:352 50. Fornasari G, Gazzano M, Matteuzzi D, Trifirò F, Vaccari A (1995) Appl Clay Sci 10:69 51. Constantino VRL, Pinnavaia TJ (1995) Inorg Chem 34:883 52. Correa JAM (2004) In: Pecchio M, Andrade FRD, D’Agostino LZ, Kahn H, Sant’Agostino LM, Tassinari MMML (eds) Applied Mineralogy. Proceedings ICAM-BR, Sao Paulo, Brazil, 1:399 53. Gardner E, Huntoon KM, Pinnavaia TJ (2001) Adv Mater 13:1263 54. Liu J, Li F, Evans DG, Duan X (2003) Chem Commun 542 55. Li F, Liu J, Evans DG, Duan X (2004) Chem Mater 16:1597 56. Meng W, Li F, Evans DG, Duan X (2004) Mater Res Bull 39:1185 57. Meng W, Li F, Evans DG, Duan X (2004) Mater Chem Phys 86:1 58. Carlino S, Hudson MJ (1995) J Mater Chem 5:1433

Preparation of Layered Double Hydroxides 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102.

117

Hussein MZB, Long CW (2004) Mater Chem Phys 85:427 Sasai R, Shin’ya N, Shichi T, Takagi K, Gekko K (1999) Langmuir 15:413 Drezdzon MA (1988) Inorg Chem 27:4628 Xu ZP, Zeng HC (2000) J Phys Chem 104:10206 Whilton NT, Vickers PJ, Mann S (1997) J Mater Chem 7:1623 Raki L, Rancourt DG, Detellier C (1995) Chem Mater 7:221 Li BX, He J, Evans DG, Duan X (2004) Appl Clay Sci 27:199 Li BX, He J, Evans DG, Duan X (2004) Int J Pharm 287:89 Yuan Q, Wei M, Wang ZQ, Wang G, Duan X (2004) Clay Clay Miner 52:40 Yuan Q, Wei M, Evans DG, Duan X (2004) J Phys Chem B 108:12382 Kuk WK, Huh YD (1997) J Mater Chem 7:1933 Feng YJ, Li DQ, Li CX, Wang ZH, Evans DG, Duan X (2003) Clay Clay Miner 51:566 Jiang XR, Li F, Evans DG, Duan X (2004) Acta Chimica Sinica 62:16 Duan X, Jiao Q (1999) Chinese Patent 99119385.7 to Beijing University of Chemical Technology Duan X, Jiao Q (2000) Chinese Patent 00132145.5 to Beijing University of Chemical Technology Vogel AI (1989) Quantitative Inorganic Chemical Analysis. 5th edn. Longman, Harlow, Essex, UK p 424 Costantino U, Marmottini F, Nocchetti M, Vivani R (1998) Eur J Inorg Chem 1439 Oh JM, Hwang SH, Choy JH (2002) Solid State Ionics 151:285 Ogawa M, Kaiho H (2002) Langmuir 18:4240 Adachi-Pagano M, Forano C, Besse JP (2003) J Mater Chem 13:1988 Costantino U, Coletti N, Nocchetti M, Aloisi GG, Elisei F (1999) Langmuir 15:4454 Costantino U, Coletti N, Nocchetti M, Aloisi GG, Elisei F, Latterini L (2000) Langmuir 16:10351 Iyi N, Matsumoto T, Kaneko Y, Kitamura K (2004) Chem Mater 16:2926 Iyi N, Matsumoto T, Kaneko Y, Kitamura K (2004) Chem Lett 1122 Morel-Desrosiers N, Pisson J, Israëli Y, Taviot-Guého C, Besse JP, Morel JP (2003) J Mater Chem 13:2582 Israëli Y, Taviot-Guého C, Besse JP, Morel JP, Morel-Desrosiers N (2000) J Chem Soc, Dalton Trans 791 Bish DL (1980) Bull Minéral 103:175 Bontchev RP, Liu S, Krumhansl JL, Voigt J, Nenoff TM (2003) Chem Mater 15:3669 Newman SP, Jones W (1999) J Solid State Chem 148:26 Ravi KK, Marc SW, James EA (1997) Chem Mater 9:417 Xing Y, Li DQ, Ren LL, Evans DG, Duan X (2003) Acta Chimica Sinica 61:267 Xu ZP, Braterman PS (2003) J Mater Chem 13:268 Williams GR, Norquist AJ, O’Hare D (2004) Chem Mater 16:975 Wang J, Wei M, Rao R, Evans DG, Duan X (2004) J Solid State Chem 177:366 Yang QZ, Sun DJ, Zhang CG, Wang XJ, Zhao WA (2003) Langmuir 19:5570 Leroux F, Aranda P, Besse JP, Ruiz-Hitzky E (2003) Eur J Inorg Chem 1242 Moujahid EM, Besse JP, Leroux F (2002) J Mater Chem 12:3324 Khan AI, Lei L, Norquist AJ, O’Hare D (2001) Chem Commun 2342 Wei M, Shi S, Wang J, Li Y, Duan X (2004) J Solid State Chem 177:2534 Choy JH, Kwak SY, Park JS, Jeong YJ, Portier J (1999) J Am Chem Soc 121:1399 Fudala A, Palinko I, Kiricsi I (1999) Inorg Chem 38:4653 Li F, Zhang LH, Evans DG, Forano C, Duan X (2004) Thermochim Acta 424:15 Villegas JC, Giraldo OH, Laubernds K, Suib SL (2003) Inorg Chem 42:5621 Malherbe F, Besse JP (2000) J Solid State Chem 155:332

118

J. He et al.

103. Badreddine M, Legrouri A, Barroug A, de Roy A, Besse JP (1998) Collect Czech Chem Commun 63:741 104. Badreddine M, Legrouri A, Barroug A, de Roy A, Besse JP (1999) Mater Lett 38:391 105. Arco MD, Gutierrez S, Martin C, Rives V (2003) Inorg Chem 42:4232 106. Prévot V, Forano C, Besse JP (2000) J Solid State Chem 153:301 107. Beaudot P, de Roy ME, Besse JP (2001) J Solid State Chem 161:332 108. Carpani I, Berrettoni M, Ballarin B, Giorgetti M, Scavetta E, Tonelli D (2004) Solid State Ionics 168:167 109. Bhattacharjee S, Dines TJ, Anderson JA (2004) J Catal 225:398 110. Isupov VP, Chupakhina LE, Mitrofanova RP, Tarasov KA, Rogachev AY, Boldyrev VV (1997) Solid State Ionics 101–103:265 111. Tsyganok AI, Tsunoda T, Hamakawa S, Suzuki K, Takehira K, Hayakawa T (2003) J Catal 213:191 112. Beaudot P, de Roy ME, Besse JP (2004) Chem Mater 16:935 113. Li C, Wang G, Evans DG, Duan X (2004) J Solid State Chem 177:4569 114. Jobbágy M, Regazzoni AE (2005) J Phys Chem B 109:389 115. Fogg AM, Dunn JS, Shyu SG, Cary DR, O’Hare D (1998) Chem Mater 10:351 116. Khan AI, O’Hare D (2002) J Chem Mater 12:3191 117. Rhee SW, Lee JH, Jung DY (2002) J Colloid Interface Sci 245:349 118. Williams GR, Norquist AJ, O’Hare D (2003) Chem Commun 1816 119. Pisson J, Taviot-Guého C, Israëli Y, Leroux F, Munsch P, Itié JP, Briois V, MorelDesrosiers N, Besse JP (2003) J Phys Chem B 107:9243 120. Prévot V, Forano C, Besse JP (2001) Appl Clay Sci 18:3 121. Crepaldi EL, Pavan PC, Valim JB (1999) Chem Commun 155 122. Crepaldi EL, Pavan PC, Valim JB (2000) J Mater Chem 10:1337 123. Erickson KL, Bostrom TE, Frost RL (2005) Mater Lett 59:226 124. Stanimirova TS, Kirov G (2001) J Mater Sci Lett 20:453 125. Marchi AJ, Apesteguía CR (1998) Appl Clay Sci 13:35 126. Rocha J, del Arco M, Rives V, Ulibarri MA (1999) J Mater Chem 9:2499 127. Kooli F, Rives V, Ulibarri MA (1995) Inorg Chem 34:5114 128. Koolf F, Depege C, Ennaqadi A, de Roy A, Besse JP (1997) Clays Clay Miner 45:92 129. Aloisi GG, Costantino U, Elisei F, Latterini L, Natali C, Nocchetti M (2002) J Mater Chem 12:3316 130. You YW, Zhao HT, Vance GF (2002) J Mater Chem 12:907 131. Katahira SI, Yasue K (1999) J Mater Chem 14:1178 132. Tagaya H, Ogata A, Kuwahara T, Ogata S, Karasu M, Kadokawa JI, Chiba K (1996) Micropor Mater 7:151 133. Aisawa S, Hirahara H, Ishiyama K, Ogasawara W, Umetsu Y, Natira E (2003) J Solid State Chem 174:342 134. Kuwahara T, Tagaya H, Chiba K (1995) Micropor Mater 4:247 135. Martín MJS, Villa MV, Sanchez-Camazano M (1999) Clays Clay Miner 47:777 136. Morioka H, Karasu TM, Kadokawa J, Chiba K (1995) J Solid State Chem 117:337 137. Corma A, Fornés V, Rey F, Cervilla A, Llopis E, Ribera A (1995) J Catal 152:237 138. Kanezaki E (1995) J Mater Sci 30:4926 139. Barbosa CAS, Ferreira AMDC, Constantino VRL, Coehlo ACV (2002) J Incl Phen Macro Chem 42:15 140. Nakayama H, Wada N, Tsuhako M (2004) Int J Pharm 269:469 141. Aisawa S, Hirahara H, Takahashi S, Umetsu Y, Narita E (2004) Chem Lett 306 142. Prévot V, Forano C, Besse JP, Abraham F (1998) Inorg Chem 37:4293 143. Ogawa M, Asai S (2000) Chem Mater 12:3253

Preparation of Layered Double Hydroxides

119

144. Xie H, Jiao Q, Duan X (2001) Chin J Appl Chem 18:70 145. Renaudin G, Francois M, Evrard O (1999) Cement Concrete Res 29:63 146. Tseng WY, Lin JT, Mou CY, Cheng S, Liu SB, Chu PP, Liu HW (1996) J Am Chem Soc 118:4411 147. Bubniak GA, Schreiner WH, Mattoso N, Wypych F (2002) Langmuir 18:5968 148. Ren LL, He J, Zhang SC, Evans DG, Duan X (2002) J Mol Catal B: Enzymatic 18:3 149. Dimotakis ED, Pinnnavaia TJ (1990) Inorg Chem 29:2393 150. Gago S, Pillinger M, Valente AA, Santos TM, Rocha J, Gonçalves IS (2004) Inorg Chem 43:5422 151. Zhang J, Zhang F, Ren LL, Evans DG, Duan X (2004) Mater Chem Phys 85:207 152. Boehn HP, Steinle J, Vieweger C (1977) Angew Chem Int Ed Engl 16:265 153. El Malki K, de Roy A, Besse JP (1989) Euro J Solid State Inorg Chem 26:339 154. Zhang J, Li DQ, Ren LL, Evans DG, Duan X (2004) Chin J Inorg Chem 20:1208 155. Li L, Luo QS, Duan X (2002) J Mater Sci Lett 21:439 156. Isupov VP, Chupakhina LE, Mitrofanova RP (2000) J Mater Syn Pro 8:251 157. Radha AV, Kamath PV (2003) Bull Mater Sci 26:661 158. Zhao Y, Jiao QZ, Li F, Evans DG, Duan X (2001) Chin J Inorg Chem 17:830 159. Shedam MR, Rao AV (1998) Mater Chem Phy 52:263 160. de la Hoz A, Díaz-Ortiz Á, Moreno A (2005) Chem Soc Rev 34:164 161. Tichit D, Rolland A, Prinetto F, Fetter G, Martinez-Ortiz MJ, Valenzuela MA, Bosch P (2002) J Mater Chem 12:3832 162. Fetter G, Botello A, Lara VH, Bosch P (2001) J Porous Mater 8:227 163. Seida Y, Nakano Y, Nakamura Y (2002) Clays Clay Miner 50:525 164. Mok KB, Ross JRH, Sambrook RM (1983) In: Poncelet G, Grange P, Jacobs PA (eds) Prep Catal III. Elsevier, Amsterdam, p 291 165. Paulhiac JL, Clause O (1993) J Am Chem Soc 115:11602 166. d’Espinose de la Caillerie JB, Kermarec M, Clause O (1995) J Am Chem Soc 117:11471 167. Merlen E, Gueroult P, d’Espinose de la Caillerie JB, Rebours B, Bobin C, Clause O (1995) Appl Clay Sci 10:45 168. Mao S, Li DQ, Zhang FZ, Evans DG, Duan X (2004) Chin J Inorg Chem 20:596 169. Estroff LA, Hamilton AD (2001) Chem Mater 13:3227 170. He JX, Kobayashi K, Takahashi M, Villemure G, Yamagishi A (2001) Thin Solid Films 397:255 171. He J, Li B, Evans DG, Duan X (2004) Colloid Surf A: Physicochem Eng Aspects 251:191 172. Jitianu M, Bãlãsoiu M, Zaharescu M, Jitianu A, Ivanov A (2000) J Sol-Gel Sci Tech 19:453 173. Indira L, Dixit M, Kamath PV (1994) J Power Sources 52:93 174. Legrand L, Abdelmoula M, Géhin A, Chaussé A, Génin JMR (2001) Electrochim Acta 46:1815 175. Olanrewaju J, Newalkar BL, Mancino C, Komarneni S (2000) Mater Lett 45:307 176. Zeng HC, Xu ZP, Qian M (1998) Chem Mater 10:2277 177. Hansen HCB, Koch CB (1995) Appl Clay Sci 10:5 178. Hansen HCB, Koch CB, Taylor RM (1994) J Solid State Chem 113:46 179. Hansen HB, Taylor RM (1991) Clay Miner 26:507 180. Hansen HCB (2001) In: Rives V (ed) Layered Double Hydroxides: Present and Future. Nova Science Publishers Inc, New York 181. Delmas C, Borthomieu Y (1993) J Solid State Chem 104:345 182. Han KS, Guerlou-Demourgues L, Delmas C (1996) Solid State Ionics 84:227

Struct Bond (2006) 119: 121–159 DOI 10.1007/430_001 © Springer-Verlag Berlin Heidelberg 2005 Published online: 3 December 2005

In Situ Polymerization and Intercalation of Polymers in Layered Double Hydroxides Christine Taviot-Guého (u) · Fabrice Leroux Laboratoire des Matériaux Inorganiques, Université Blaise Pascal, UMR 6002-CNRS, 24 av. des Landais, 63177 Aubière, France [email protected], [email protected] This chapter is dedicated to Dr. Jean-Pierre Besse on the occasion of his retirement. 1

Introduction: State-of-the-Art for Inorganic/Polymer Nanocomposites . .

LDH/Polymer Nanocomposites . . . . . . . . . . . . . . . . . . . Improved Strategies to Incorporate Polymers Into LDH Materials Insights Into the In Situ Polymerization Process . . . . . . . . . . LDH Host Features: Local Structure and Layer Charge Density . . Characterization of LDH/Monomer Assembly: Matching Between the Two “Sub-Lattices” . . . . . . . . . . . . . 2.2.3 In Situ Polymerization Evidenced by 13 C Solid-State NMR and ESR Spectroscopies . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Thermal Behavior of LDH/Polymer Nanocomposites . . . . . . . 2.4 LDH/Biopolymers . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

128 128 135 135

. . . . .

137

. . . . . . . . . . . . . . .

143 148 151

Future Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

152

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

156

2 2.1 2.2 2.2.1 2.2.2

3

. . . .

. . . .

. . . .

. . . .

122

Abstract This chapter is intended to provide a state-of-the-art review of nanocomposite materials prepared by the assembly of layered double hydroxides (LDH) and polymers, including their synthesis and characterization, and point out their potential applications. Owing to the highly tunable LDH intralayer composition coupled with the wide possible choice of organic moiety, a large variety of LDH/polymer systems may be tailored. The incorporation of a polymer in the interlayer galleries may proceed via different pathways such as coprecipitation, exchange, surfactant-mediated incorporation, hydrothermal treatment, reconstruction, or restacking. Alternatively, various monomers can be intercalated and polymerized in situ within the interlamellar space of LDH. The spatial confinement is believed to increase the degree of polymerization, and, in addition, this type of in situ radical polymerization process makes it possible to tune the tacticity and the molecular weight of the resulting polymer by varying the layer-charge density and the particle size of the host structure, respectively. From several studies, it has been observed that the multi-component LDH/polymer systems are thermally more stable than the pristine inorganic compounds, leading to potential applications such as flame-retardant composites. Bio-related polymers and large bio-macromolecules such as poly(aspartate), alginate, and deoxyribonucleic acid (DNA), have also been incorporated within the galleries of LDH materials. One obvious interest for these bio-organoceramics is the drug release aspect, but acquiring a fundamental understanding of biomechanisms and other biomimetic phenomena is also important. The incorporation of polymers into

122

C. Taviot-Guého · F. Leroux

hydrocalumite-like LDHs, known collectively as AFm phases in the literature of cement science, is of current concern in relation to the macro-defect-free (MDF) cement concept. Another incentive aspect is the use of LDH materials as nanolayers for fillers in polymeric matrices. Largely studied for the case of smectite-type materials, some recent results show similar trends for LDH nanofillers, i.e., an enhancement of the mechanical properties and increase in the polymer glass transition temperature. Keywords Layered double hydroxide (LDH) · Organic inorganic hybrids · Nanocomposite polymers · In situ polymerization · Nanoscopically confined polymers

1 Introduction: State-of-the-Art for Inorganic/Polymer Nanocomposites To put the current research in LDH/polymer nanocomposites into context, we will first give a brief review of 2D-hybrid materials composed of a guest polymer and an inorganic host structure. In materials science, considerable attention has been devoted to how to prepare and shape new multifunctional materials. For the past two decades, 2D-hybrid materials composed of a guest polymer and an inorganic host structure have been extensively studied [1–10]. The lamellar host structure supplies a constrained environment in which the polymer is forced to locate and both parts may also act synergistically. For example, from the organic point of view, polymers such as poly(ethylene oxide) (PEO) and polyaniline (PANI), among others, have been widely studied due to their interesting conductive properties. PEO displays relatively high ionic conductivity due to the motion of the intracrystalline cations confined in its chains, and PANI exhibits good electronic conductivity, but this is associated with rather poor processability. Their incorporation within 2D inorganic host structures may be of a great interest for applications as electrode materials, solid-polymer electrolytes for fuel cells, or for others requiring corrosion protection by gas barriers. Alternatively, from the viewpoint of the host structure, the presence of the polymer is found to increase the diffusion of mobile species. The formation of a sandwich assembly in which a polymer and an inorganic host structure alternate may proceed by several pathways (i) direct intercalation of polymer, (ii) polymer inclusion, and (iii) monomer incorporation followed by subsequent polymerization. (i) Direct intercalation of polymers can be achieved via dispersion of a clay mineral or transition metal oxide in dissolved polymers, such as for PEO/montmorillonite [11] and PEO/MoO3 [12]. To facilitate the diffusion of the polymer into the galleries, the interlayer space may be either increased or chemically modified. For instance, for epoxy-clay systems, the key was found to be to load the clay gallery with hydrophobic surfactants, thus overcoming the incompatibility between the polymer and the host. Lyophilization

In Situ Polymerization and LDH/Polymer Nanocomposites

123

of the clays by incorporation of alkyl ammonium cations acts to reduce the surface polarity of the silicate layers, and therefore provides a good affinity between the silicate and the polymer matrix. Using the same idea, the socalled “refined guest displacement” and “surfactant mediated pathway” have been employed for the PVP–kaolinite system via a kaolinite-ammonium acetate precursor [13] and for the incorporation of poly(p-phenylene) (PPP) into molybdenum bronzes [14], respectively. The mechanism is displayed in Fig. 1. The surfactant-mediated approach was found to be appropriate for the transport of insoluble, hydrophobic polymers such as poly(p-phenylene) within an inorganic interlayer space. ii) A neat procedure consists of polymer inclusion. This was reported for different systems such as PANI/MoS2 [15, 16], and PEG, PVP, PEO/NbSe2 [17]. The formation of the sandwich assembly proceeds by restacking the layers in an appropriate solvent. Intercalation of polymers into the interlayer space of smectites has been extensively studied, however, it was found to be difficult to avoid reaggregation of the layers when the polymer interacted strongly with the fully delaminated smectite colloid [18]. iii) Monomer incorporation and its subsequent in situ polymerization has been exemplified in several cases. Among others, one may report the polymerization of aniline or pyrrole in HNbMoO6 [19], FeOCl [20], α-RuCl3 [21], MoO3 [22], and graphitic oxide [23–26]. For such non-oxidizing structures, an external agent such as FeCl3 , (NH4 )2 S2 O8 or a soft thermal treatment may be used to induce the radical polymerization process. Solventless oxidative polymerization was recently reported for a hectorite clay exchanged with Cu2+ and Fe3+ cations [27]. As in the case of layered double hydroxide host structures in which the external agent may be partially exchanged as discussed below, thermal postsynthesis treatment has been employed to induce the monomer linkage, as exemplified by the insertion of the precursor p-xylene-α-dimethylsulfonium chloride into a hydrated MoO3 bronze [28] or by the incorporation of ε-aminocaproic acid into α-zirconium phosphate (α-ZrP) leading to a nylon nanocomposite after treatment at 200 ◦ C under nitrogen [29]. Conversely, when the host structure has a strong oxidizing capacity, such as xerogel V2 O5 , the in situ polymerization is concomitant with the in-

Fig. 1 Schematic showing surfactant-mediated incorporation of PPP in MoO3 (Reprinted from [14] with permission from ACS)

124

C. Taviot-Guého · F. Leroux

corporation of the monomer. The so-called “reductive intercalative polymerization” (RIP) process developed by Kanatzidis et al. has been mostly exemplified by the incorporation of conductive polymers (polythiophene, polypyrrole, polyaniline) [30]. The polymerization occurs via the redox reaction induced by V5+ cations, although Leroux et al. have demonstrated that a fully reoxidized V2 O5 framework may be obtained after mild oxygen treatment [31]. The oxidative technique has been extended to other hybrid phases obtained by the RIP process. When dealing with assemblies involving a polymer/inorganic framework, it is interesting to consider the matching between the host and the guest, i.e., the organization of the in-plane inorganic sheets compared to the size of the monomer unit. The best example is supplied by the PANI–FeOCl system [32]. The presence of PANI chains between FeOCl slabs was found to double the periodicity along the a and c axes relative to the original FeOCl unit cell (Fig. 2). According to the authors, the origin of this superlattice was due to the substantial long-range order of PANI in FeOCl. The polymer orientation inside the host structure was explained by the matching between the repeat unit of PANI and the Cl – Cl distance. By orientating parallel to the (101) direction, the polymer can spatially match each NH unit with a Cl “partner” (Fig. 3). The presence of H – Cl hydrogen bonding, surmised by the authors but not directly demonstrated, may act as an additional stabilizing force in the hybrid assembly. Such ordering of one material in the interior of another was defined as an endotactic reaction. The matching between the guest molecules and the host structure was also considered to explain the in situ polymerization of ε-aminocaproic acid in zirconium phosphate [29]. The schematic representation of the arrangement

Fig. 2 Crystal structure of FeOCl viewed perpendicularly to the b axis (left) and proposed arrangement of PANI chains in the galleries of FeOCl viewed from two different directions (middle and right) (Reprinted from [32] with permission from ACS)

In Situ Polymerization and LDH/Polymer Nanocomposites

125

Fig. 3 Crystal structure of FeOCl viewed down the b axis (left) (large open circles represent chlorine, small open circles represent oxygen atoms, and crossed circles represent iron atoms) and, for the hybrid phase PANI/FeOCl, projection of the relative orientation of PANI chains with respect to an FeOCl layer viewed down the stacking b axis (right) (Reprinted from [32] with permission from ACS)

of ε-aminocaproic acid within the zirconium phosphate interlayer gap, displayed in Fig. 4, was said to be consistent with an orientation of the monomer molecules parallel to the phosphate layers and with the stoichiometry of the hybrid phase. The amino acid molecules are favorably disposed for the condensation reaction, and the condensation and formation of interleaved polyamide linkages to form nylon was observed at an elevated temperature. On the basis of a comparison of 13 C NMR data for the inserted nylon and nylon prepared ex situ, the authors concluded that there was a strong interaction of the carbonyl groups of the polymer with the interlayer P – OH groups.

Fig. 4 Schematic representation of the arrangement of ε-aminocaproic acid in zirconium phosphate in the layer-parallel phase (Reprinted from [29] with permission from ACS)

126

C. Taviot-Guého · F. Leroux

To further understand the arrangement of the guest molecules within a host structure, one-dimensional Patterson function and electron density calculations may be carried out. From the observed intensities of the diffraction lines characteristic of the stacking sequence, a one-dimensional (1D) Patterson function projected along the stacking axis and then a 1D electron density plot are helpful to locate the arrangement of the guest molecules. This is illustrated here by the system PEO/V2 O5 [33]. As theoretically expected and observed experimentally, if PEO confined in a V2 O5 framework has a straight-chain planar zigzag conformation, the 1D electron density map can show several different patterns which depend on the packing arrangement of the polymer chains (Fig. 5a). The presence of interleaved PEO was manifested by four peaks, allowing the authors to conclude that the polymer

Fig. 5 a Two schematic arrangements of planar-zigzag PEO chains in the V2 O5 framework and their expected electron density projections along the c axis: (top) the plane containing the polymer chain is perpendicular to the V2 O5 sheet and (below) the plane is parallel to the V2 O5 sheet and the PEO bilayer arrangement is in a zigzag-like fashion. b Projection of the electron density of (PEO)1.0 V2 O5 · nH2 O and illustrations of the deduced arrangement of PEO in the interlayer space (Reprinted from [33] with permission from ACS)

In Situ Polymerization and LDH/Polymer Nanocomposites

127

was not in a helical coil structure but rather in a fully extended arrangement, with molecules parallel to each other and arranged in a bilayer fashion (Fig. 5b). This method will be fully detailed in the following section taking the example of the hybrid phase styrene sulfonate/hydrocalumite (see Sect. 2.2). The benefit of a hybrid phase for the intercalation-deintercalation of mobile species such as Li+ cations is well illustrated by the study of conductive polymers such as polyaniline or polypyrrole intercalated into a V2 O5 framework as potential electrode materials in lithium batteries [34]. For PANI/V2 O5 , an oxidative post-treatment performed under an oxygen atmosphere allowed the authors to compare the conductivity attributed to the polymer, as in absence of reduced V4+ cations, there was no electronic hopping between V5+ /V4+ ions, and the conductive state was due only to the presence of PANI in its emeraldine salt form. V2 O5 xerogel has a conductivity of 8 × 10–6 S cm–1 whereas PANI/V2 O5 after the oxidative treatment possesses a conductivity of 2.5 × 10–3 S cm–1 . Moreover, the lithium insertion was completely reversible, and relaxation studies demonstrated that the polymer, in addition to its capacitive response slightly increasing the electrochemical performance, was also facilitating the insertion-deinsertion of the lithium ions between the layers (Fig. 6). Lastly, the authors observed that the electrochemical response of the hybrid phases was greater than the sum of the two components (organic and inorganic) when the polymer was truly

Fig. 6 Galvanostatic intermittent titration curves for a V2 O5 and b PANI0.44 V2 O5 (Reprinted from [34] with permission from the Electrochemical Society)

128

C. Taviot-Guého · F. Leroux

incorporated between V2 O5 slabs and not only adsorbed on the surface, underlining the synergy that occurs when they are intimately assembled. The next section is devoted to layered double hydroxide (LDH)/polymer nanocomposites, which may be referred to as organoceramics. A large variety of anionic polymers have been introduced into the interlayer space of LDH. First, the different synthesis routes reported for LDH/polymer nanocomposites are described with an exposition of the factors that can influence assembly characteristics. The in situ polymerization process is discussed in more detail in Sect. 2.2. The matching between the monomeric repeat unit and the layer-charge density is illustrated by a comparative study of the in situ polymerization of vinyl benzene sulfonate in disordered ZnR Al (R = 2, 3, 4) and ordered Ca2 Al LDH hosts. This section will also demonstrate the utility of NMR and ESR spectroscopies in probing the polymerization reaction. The textural properties and temperature behavior will be presented in Sect. 2.3. The following section (2.4) deals in some details with LDH/biopolymer nanocomposites. The last part is devoted to future developments and trends in the field of LDH/polymer nanocomposites.

2 LDH/Polymer Nanocomposites 2.1 Improved Strategies to Incorporate Polymers Into LDH Materials In this section, we consider the different synthesis routes that have been employed to incorporate polymers into LDH with emphasis on the in situ polymerization process, an approach which gives rise to well-defined nanocomposites. LDH are built up by the stacking of two kinds of layers, weakly held together: metal hydroxide layers with positive charge maintained by covalent forces and an interlamellar gallery composed of easily exchangeable anionic species surrounded by water molecules. The LDH structure can accommodate numerous cations within the hydroxide layers as well as various anions in the interlayer space. This wide range of composition is a result of several special features. First, the LDH hydroxide layers are constituted of one layer of edge-sharing octahedra and are more flexible than other two-dimensional frameworks such as 2/1 layered silicates. In addition, the anionic exchange capacity (AEC) can be varied over a wide range depending on the divalent to trivalent metal ratio R = M2+ /M3+ within the hydroxide layers: the values range between 450 and 200 meq/100 mg for LDH while cationic clays present limited AEC near 100 meq/100 mg. Finally, the synthesis procedures offer numerous parameters that make possible a fine tuning of the physico-chemical properties. A large variety of LDH/polymer systems can thus be tailored by

In Situ Polymerization and LDH/Polymer Nanocomposites

129

Table 1 Polymer/LDH nanocomposites Polymer/LDH

Synthetic pathway

Refs.

Cu2 Al/PANI Cu2 Cr/PANI LiAl2 /PANI Ca2 Al/PVA Mg3 Al/PSS Zn2 Al/PSS MAl/PA, PVS (M = Mg, Co, Zn) CaAl/PA, PVS, PSS NiFe/PA M2 Ni/PA (M = Mn, Fe, Co) Zn2 Al/PSS MCr/PEG-(DC and AS) (M = Cu, Zn)

(d) (a) (a) (b) (b) (a, b) (b) (b) (a) (a) (a, b, c) (b, d)

[54] [55] [122] [94] [37] [37, 55] [38, 111] [38] [50, 51, 53] [52] [43] [123]

Polymer: (PANI) poly(aniline); (PVA) poly(vinyl)alcohol; (PSS) poly(styrene sulfonate); (PVS) poly(vinyl sulfonate); (PA) poly(acrylic acid); (PEG-DC) poly(ethylene glycol) dicarboxylate; (PEG-AS) poly(ethylene glycol) alkyl (3-sulfopropyldiether). Pathway: (a) in situ polymerization, (b) polymer direct incorporation, (c) restacking or reconstruction, and (d) guest displacement method.

considering the intralayer composition and the choice of the organic moiety. The nanocomposites mentioned hereafter are listed in Table 1. In the following, the pristine material and organic derivatives are noted as M2+ R M3+ /X where X represents the interlayer anion. To overcome the problem of diffusion of cumbersome molecules (such as polymers) into the constrained interlamellar space of LDH, several methods have been employed, which are presented in Fig. 7. According to the general classification established by Schöllhorn et al. [35], there are three principal options to obtain a polymer intercalated inorganic host: (a) Direct intercalation of extended polymer chains in the host lattice. (b) Transformation of the host material into a colloidal system and precipitation in the presence of the polymer. (c) Intercalation of the monomer molecules and subsequent in situ polymerization. An additional route consists of using the reconstruction ability of some LDH materials after a moderate thermal treatment. This is a peculiarity of LDH systems called the “memory effect”: the LDH lamellar framework is reconstructed in the presence of the polymer with concomitant intercalation. Finally, a post-synthesis hydrothermal treatment can be applied in all cases, which may improve the inter- and intralamellar organization of the LDH/polymer nanocomposites.

130

C. Taviot-Guého · F. Leroux

Fig. 7 Scheme of the preparation of LDH/polymer nanocomposites: (a) in situ polymerization, (b) direct incorporation of polymer, (c) restacking or reconstruction, and (d) guest displacement or solvent-assisted method

The first method involving the incorporation of the polymer as a whole, by coprecipitation or exchange reaction, was applied in the intercalation of poly(acrylic acid), poly(vinyl sulfonate), poly(styrene sulfonate) and poly(ethylene glycol) alkenyl sulfonic acid into LDH [36, 37]. Generally, the incorporation of polymer chains of small molecular weight is obtained via exchange reactions while the coprecipitation route allows the incorporation of larger molecules. This latter method can be considered as a self-assembly process involving the in situ formation of LDH layers around the polymer intercalate. Concerning the exchange route, a pre-intercalation may prepare the interlayer space for subsequent incorporation as evidenced by the early work of Drezdzon with the pre-intercalation of terephthalate anions [38] or the swelling with glycerol reported by Dimotakis and Pinnavaia [39] prior to incorporation of polyoxometallates (POM). This method, referred to as guest displacement or the solvent-assisted method, compatibilizes the host interlayer space to the guest molecules. The presence of hydrophobic alkyl chains

In Situ Polymerization and LDH/Polymer Nanocomposites

131

also makes it possible to entrap non-functionalized polymers as reported for the incorporation of poly(ethylene oxide) into LDH using a dodecyl sulfate (DDS) intercalate as precursor [40]: the organo-modified LDH has A, which suggests a highly interdigitated a basal-plane repeat distance of 26.2 ˚ arrangement of the alkyl chains; when PEO is incorporated, the d-spacing is A. increased to 38.2 ˚ The delamination-restacking process should be a promising way for the preparation of LDH nanocomposites since the conditions of small particle size and low layer-charge density, required for the formation of colloidal or exfoliated LDH suspensions, are appropriate for the subsequent entrapment of large polymers. LDH nanolayer exfoliation also enables the dispersion of LDH as particles reinforcing the polymeric matrix. As a result of their high layer-charge density, however, LDH do not have a natural tendency to exfoliate, contrary to cationic clays and only a few attempts have been successful. The modification of LDH with organic anions as described above was found to be effective in bringing about subsequent delamination. In this way, Zn2 Al LDH modified with DDS was delaminated after refluxing in butanol [41, 42] and the subsequent restacking of the layers with PSS was observed [43]. Mg2 Al LDH intercalated with DDS also underwent delamination in a solution of 2-hydroxyethyl methacrylate (HEMA) under high shear [44], which was then polymerized to poly(acrylate) with the inorganic component still in the delaminated form. A DDS intercalate was also reported to be well dispersed in a polyethylene matrix after a refluxing procedure in xylene [40]. More recently, it has been demonstrated that amino acid-containing LDH in formamide are readily delaminated, since the reaction occurs spontaneously and does not need any heat or refluxing [45]. Using this method, the entrapment of poly(vinyl alcohol) between the layers of Mg3 Al LDH was achieved [46]. Immobilization of a guest molecule can be achieved through the “memory effect” of LDH. The partial dehydroxylation of LDH hydroxide layers under a moderate thermal treatment leads to the formation of amorphous oxides usually denoted as layered double oxides (LDO). For some LDH compositions these oxides can reconstruct the parent LDH either on cooling in air (uptake of carbonate anions) or on soaking in an aqueous solution [47]. This technique was first used to incorporate large anions such as polyoxometallates [39]. By contacting an amorphous LDO with a solution containing PSS, it was also possible to obtain a nanocomposite, which presents similarities in terms of crystallinity and morphology to the phase prepared by a direct exchange (Fig. 8) [48]. Similarly, the oxide salt method developed by Boehm et al. [49] and commonly used for the preparation of M2 Cr (M = Cu and Zn) LDH was reported to immobilize polymers through the formation of Zn2 Cr layers, although no further characterization was provided [35]. All these pathways give rise to poorly defined nanocomposites with only a few harmonics visible in the X-ray diffraction patterns. Generally, the polymer is found to influence strongly not only the textural properties but also the

132

C. Taviot-Guého · F. Leroux

Fig. 8 Powder X-ray diffraction patterns (CuKα ) of Zn2 Al/PSS nanocomposites obtained via: a in situ polymerization (1). The patterns of the monomer-exchanged phase (2), and of the pristine material at room temperature (3) and at 150 ◦ C (4) are also reported for comparison. b polymer direct exchange (1), reconstruction (2) or restacking of the layers over the polymer (3) or the monomer (4). Patterns are offset for clarity

intralayer composition of the inorganic structure. The sheets are more or less crumpled depending of the pathway. The influence of the preparation method for Zn2 Al/PSS is evidenced by SEM micrographs (Fig. 9) [43]. The sand-rose morphology typical of the pristine materials is modified considerably after in situ polymerization of 4-styrene sulfonate; large chunks are obtained by direct incorporation through the coprecipitation method. In the latter case, the crystallinity may be enhanced by a post-synthesis hydrothermal treatment, which both increases the coherence length along the stacking direction and improves the organization of the hydroxide layers (Fig. 10) [40]. Ultrasound was also found to ameliorate slightly the crystallinity of the nanocomposite Mg3 Al/poly (vinyl alcohol) obtained by a delamination-restacking process [39]. In situ polymerization is generally a highly suitable method for the obtention of LDH/polymer nanocomposites. Various monomers can be intercalated and polymerized within the interlamellar space of LDH and this spatial confinement is believed to increase the degree of polymerization. Yet, the process is limited by two factors [43]: • The distance between the monomers, which depends on the charge density of the hydroxide layer. • The conditions required for polymerization (pH, oxidizing agent, temperature), which must leave the layered structure intact. Tanaka et al. were the first to report the in situ polymerization in an LDH system for Mg2 Al/acrylate LDH obtained via an exchange reaction with Cl– or NO3 – -containing LDH [50, 51]. Polymerization was observed after a thermal

In Situ Polymerization and LDH/Polymer Nanocomposites

133

Fig. 9 Scanning electron micrographs of a Zn2 Al/Cl, and its PSS derivatives obtained b via monomer intercalation c followed by in situ polymerization, or d by direct polymer incorporation using the “memory effect” with e subsequent hydrothermal treatment and f after a delamination-restacking process. The bar represents 5 µm. Reprinted from [43] and [48] with permission from ACS and RSC, respectively

treatment at 80 ◦ C leading to a slight decrease of the interlamellar distance A to 13.4 ˚ A; it was also evidenced by the disappearance of the from 13.8 ˚ C = C vibration band in the IR spectrum. Acrylic acid was intercalated in the lamellar structure of an iron-substituted nickel LDH material [52, 53]. In this study, potassium persulfate was used to initiate the polymerization reaction. The resulting nanocomposite shows a net decrease of the interlamellar A to 12.6 ˚ A, which was interpreted in terms of the disdistance from 13.6 ˚ appearance of the electrostatic repulsions between the C = C double bonds. Noteworthy is the formation of a sulfate-containing LDH as a result of persulfate decomposition, together with the acrylate anion monomer-intercalated phase. More recently, Vaysse et al. have reported the in situ polymerization of acrylate in iron-, cobalt-, or manganese-substituted nickel hydroxides [52]

134

C. Taviot-Guého · F. Leroux

Fig. 10 X-ray diffraction (left part) and absorption (right part) of the nanocomposites Zn2 Al/PSS before (a) and after (b) hydrothermal treatment. The distances are not corrected from phase shifts. Reprinted from [48] with permission from RSC

using potassium persulfate. In this work, the preparation method of the LDH precursor, either via the oxide salt method or coprecipitation depending on the nature of the trivalent cations, was found to strongly influence the intralamellar arrangement of the macromolecule. In addition, for Co and Mn-containing LDH, the intercalation and polymerization processes appear to proceed concomitantly. Styrene sulfonate was also polymerized between Zn2 Al LDH sheets [48], giving rise to well-defined nanocomposites while Zn3 Al and Zn4 Al hosts, with lower charge density, exhibit incomplete polymerization; this will be further discussed in Sect. 2.2.2. As already noted in Sect. 1, in the case of conjugated polymers, when the host structure presents a strong oxidizing capacity such as V2 O5 , the in situ polymerization is concomitant with the monomer intercalation. For nonoxidative matrices such as LDH, this so-called RIP process is not possible and use of an external oxidizing agent like FeCl3 or (NH4 )2 S2 O8 is required as reported for the polymerization of aniline or pyrrole into other nonoxidative matrices like HNbMoO6 [19], FeOCl [20], α-RuCl3 [21], MoO3 [22], and graphitic oxide [23–26]. Insertion of conjugated polymers into an LDH framework was first described by Challier and Slade [54]. Terephthalate and hexacyanoferrate exchanged Cu2 Cr LDH phases were used as host matrices for the oxidative polymerization of aniline. The reaction performed under reflux conditions led to a rather poorly defined material with a basal spacing of ∼ 13.5 ˚ A. An alternative method consists of incorporating a soluble anionic monomer such as aniline-2-sulfonate or metanilic acid (3-amino benzenesulfonic acid – H2 NC6 H4 SO3 H) that can undergo polymerization under less drastic conditions thus preserving the host structure [55]. According to

In Situ Polymerization and LDH/Polymer Nanocomposites

135

the authors, the electrophilic function decreases the potential of polymerization [56, 57] and the sulfonic acid ring-substituted polyaniline (PANIS) is capable of self-doping [58, 60]. The absence of external oxidizing agent also avoids the competitive exchange with chloride or disulfate anions. Atmospheric oxygen may initiate in situ polymerization [55] as well as light as exemplified by the photo-induced isomerization/polymerization of (Z,Z)-muconate anions in the gallery space of Li2 Al LDH [61] and the photoisomerization of indolinespirobenzopyran [62]. 2.2 Insights Into the In Situ Polymerization Process 2.2.1 LDH Host Features: Local Structure and Layer Charge Density As can easily be understood, the in situ polymerization process is favored when distances between monomers in the interlamellar space are close to the monomeric repeat distance displayed by the polymer. Such a compatibility of distances was first demonstrated for the polymerization of aniline in V2 O5 and FeOCl hosts with an exact matching between NH functions for PANI and the (1–30) crystallographic direction for V2 O5 or the (201) direction for FeOCl [63]. In the case of LDH hosts, since only the average structure is known, it is important to consider the local structure by means of X-ray absorption spectroscopy: extended X-ray absorption fine structure (EXAFS). According to an ideal model based on edge-sharing octahedra [64], it is possible to define the local environment around each type of cation in relation to the layer-charge density. Figure 11 shows the local cation environment for M2+ R M3+ LDH with R = 2 and 3. The correlation between cations M – M is as follows: the first √ correlation is at a distance of a noted as P2, then P3, P4, P5 and P6 at a 3, 2a, √ a 7 and 3a, respectively. P1 represents the oxygen shell (M – O). Regardless of the ratio R = M2+ /M3+ , the local environment P2 around M3+ is composed of 6 M2+ whereas, for M2+ , it depends on this ratio, as illustrated in Fig. 11. LDH host structures such as Mg2 Fe [65], Co2 Fe1–y Aly [66], and M2 Cr (M = Cu, Zn) [67] generally exhibit a local order. Figure 12 shows the moduli of the Fourier transform for CoR Al (R = 2 and 3) at the Co K-edge; the increase in the intensity of the peaks P2 and P4 has to be understood as indicating a greater number of heavier backscattering atoms Co compared to Al atoms in the corresponding shells. The presence of such a local order is helpful in order to better understand the matching between the LDH host structure and a guest molecule. The anionic exchange capacity for some LDH compositions is reported in Fig. 13. The values range from 450 to 200 meq/100 g; lower values are difficult to reach since the ratio R = M2+ /M3+ would be too high to maintain

136

C. Taviot-Guého · F. Leroux

Fig. 11 Ideal local order for a cation composition M2+ /M3+ of a 2 : 1, and b 3 : 1. The circles represent the correlations between the cations P2 → P6 defined in the text

the LDH structure. In terms of the formulation M2+ 1–x M3+ x (OH)2 · X–m x/m · nH2 O, LDH materials are usually obtained in the domain 0.20 ≤ x ≤ 0.33 A2 /e. For comparison, leading to a surface per unit charge between 25 and 40 ˚ cationic clays display lower exchange capacity near 100 meq/100 g, associA2 /e. Owing to their ated with a higher surface area per unit charge of ∼ 70 ˚ high layer-charge density, LDH hydroxide layers are tightly stacked via attractive forces with interlayer anions and the latter are densely packed in order to balance the layer charge, leaving no free space available. Such a situation is thus unfavorable for both ion-exchange and exfoliation processes. Indeed,

Fig. 12 Moduli of the Fourier transform for Cox Al at Co K-edge (x = 2 (black circle) and x = 3 (empty square)). The distances are not corrected from phase shifts

In Situ Polymerization and LDH/Polymer Nanocomposites

137

Fig. 13 Variation of anion exchange capacity (meq/100g) as a function of the amount of trivalent cation x in M2+ 1–x M3+ x (OH)2 · X–m x/m · nH2 O

LDH materials are not readily exfoliated in contrast to other layered materials such as smectite clays or MS2 type-chalcogenides (MoS2 , NbSe2 , etc...) [15– 17]; the delamination of LDH sheets requires elaborate syntheses as discussed above in Sect. 2.1. 2.2.2 Characterization of LDH/Monomer Assembly: Matching Between the Two “Sub-Lattices” To illustrate the matching between the LDH host and the monomeric repetitions in the polymer, one can compare the in situ polymerization of vinyl benzene sulfonate in two LDH hosts ZnR Al [49, 68], and Ca2 Al [69, 70]. Most synthetic LDH phases are hydrotalcite-like materials characterized by a disordered cation distribution within the hydroxide layers and a variable layercharge density. In contrast, the Ca2 Al host belongs to the hydrocalumite family; the unusual coordination environment of the Ca2+ ions induces an ordered distribution of the cations within the hydroxide layers and these are present for a unique Ca2+ /M3+ molar ratio equal to 2. In this respect, hydrocalumite can serve as a model for less ordered LDH. The incorporation of polymers into a hydrocalumite host is of current concern for cement applications [71–75]. Indeed, like the calcium silicates C – S – H, these calcium aluminium hydroxide salts called Friedel salts are formed during the hydration process and may play an important role in determining the mechanical properties of the cement.

138

C. Taviot-Guého · F. Leroux

Vinyl benzene sulfonate (styrene sulfonate – VBS) was incorporated into these two host structures using the coprecipitation method. The layer-charge density of ZnR Al LDH was varied with R = 2, 3 or 4. The incorporation of VBS was characterized by X-ray diffraction, and the in situ polymerization process was evidenced by 13 C solid-state CP-MAS NMR spectroscopy. VBS incorporation between the layers of LDH was first confirmed by a large increase in the basal spacing, the value of which represents the sum of the layer thickness (approximately 4.8 ˚ A) and the gallery height (Fig. 8 (Sect. 2.1) and Fig. 14). The interlamellar distances for M2 Al/Cl (M = Zn and Ca) phases, and their VBS and PSS intercalates obtained by direct incorporation via coprecipitation are reported in Table 2. For hydrotalcite-like materials, the reflections are generally indexed in a hexagonal lattice with an R-3m rhombohedral symA and metry; refined cell parameters for Zn2 Al/Cl are the following: a = 3.07 ˚

Fig. 14 XRD patterns of a Ca2 Al/Cl and hybrid derivatives: Ca2 Al/VBS before b and after in situ polymerization c and d Ca2 Al/PSS obtained via coprecipitation Table 2 Chemical analysis of M2 Al/Cl (M = Ca, Zn) and the organic derivatives obtained by direct incorporation using the coprecipitation method. The basal spacing is displayed (Reprinted from [70] by permission of Elsevier) Sample

Chemical formulae

d-spacing (˚ A)

Ca2 Al/Cl Zn2 Al/Cl Ca2 Al/VBS Zn2 Al/VBS

Ca0.67 Al0.33 (OH)2 Cl0.33 · nH2 O Zn0.67 Al0.33 (OH)2 Cl0.33 · nH2 O Ca0.67 Al0.33 (OH)2 (C8 H7 SO3 )0.32 · nH2 O Zn0.67 Al0.33 (OH)2 (C8 H7 SO3 )0.32 · nH2

7.9 7.8 17.7 18.2

Ca2 Al/PSS Zn2 Al/PSS

Ca0.67 Al0.33 (OH)2 (C8 H7 SO3 )0.32 · nH2 O Zn0.67 Al0.33 (OH)2 (C8 H7 SO3 )0.33 · nH2 O

19.2 19.3

In Situ Polymerization and LDH/Polymer Nanocomposites

139

A (= 3 × d003 ). For Ca2 Al/Cl, the structure is often described in c = 23.37 ˚ the rhombohedral space group R-3 [76–80], with the refined cell parameters A and c = 23.37 ˚ A (3 × d003 ). The gallery heights obtained for VBS a = 5.74 ˚ A for Ca2 Al/VBS and 13.4 ˚ A for all ZnR Al/VBS samples. derivatives are 12.9 ˚ A), the gallery dimensions indicate In view of the size of VBS anions (∼ 6.8 ˚ a bilayer intercalate arrangement in all cases. The structural model places the three oxygen atoms in each anionic – SO3 head group in a plane parallel to the hydroxide layers, with the hydrophobic aromatic part thus orientated toward A, VBS the center of the gallery. Yet, with a gallery dimension smaller than 14 ˚ molecules cannot fit between the layers in a perpendicular orientation. Either the bilayers nestle into one another or VBS molecules are inclined with regard to the c axis or both. Compared to the chloride precursors, a broadening of the (00l) and (110) reflections is observed for VBS derivatives, which is likely to be due to a significant disorder along the stacking direction (turbostratic effects) and a partial disorder occurring within the hydroxide layers, respectively. For the Ca2 Al host, a change from the space group R-3 to C2/c is observed with the following reA, b = 5.74 ˚ A, c = 36.10 ˚ A and β = 104.91 ◦ . The fined cell parameters a = 9.96 ˚ C2/c space group involves the presence of a two-fold axis located midway in the interlamellar space consistent with a bilayer arrangement. 13 C CP-MAS spectra of VBS derivatives are presented Fig. 15 and for comparison, the spectrum of Na-VBS is displayed in the same figure. The resonance peaks were assigned according to the literature [81]: C1 113.9 ppm, C2 137.4 ppm, C3 141.5 ppm, C4 125 ppm, and C5 143.8 ppm for C1 H2 = C2 H – C3 (C4 4 H4 )C5 – SO3 (Na). The incorporation of the organic molecule induces an up-field shift of C2 , C3 and C5 carbon atoms, corres-

Fig. 15

13 C

CP-MAS NMR spectra of a VBS, b Ca2 Al/VBS and c Zn2 Al/VBS

140

C. Taviot-Guého · F. Leroux

ponding to a shielding effect and consistent with an electrostatic interaction between the sulfonate function and the hydroxide layers. The interaction weakens the electrophilic character of the monomer through the carbon atom C5 . The shielding propagates through the benzene backbone down to C2 . It is interesting to note that the resonance peak of C1 = CH2 is then deshielded, indicating that π electrons are preferentially located on the terminal carbon C1 . This may explain why the in situ polymerization is achieved in the absence of chemical initiator and requires a soft thermal treatment only. Furthermore, it is noteworthy that whatever the layer-charge density, similar shifts are observed indicating that the forces of attraction are equivalent. This means that VBS molecules always interact with the same local charge on the hydroxide layers and only the number of VBS molecules is changing when the layer charge is varied. No difference is apparent between Zn2 Al and Ca2 Al hosts, probably due to the small difference in the layer charge densities, 24.7 or 28.6 ˚ A2 /e, respectively. The interaction may be studied by FTIR. Shifts in the symmetric and asymmetric stretching modes of (SO3 – ) functional groups constitute strong evidence of a geometric disturbance. The observed reduction in frequency corresponds to a weakening of the S = O bond strength. This suggests the presence of an electrostatic binding with the hydroxide layer through hydrogen bonding as follows S = O...H – O – Me (Me = Zn or Al) [48]. Owing to the relatively good crystallinity and the large number of 00l reflections observed for the hydrocalumite derivative, the electron density distribution along the c axis can be estimated using a series of 00l reflections, in accordance with previous literature [82, 83]. One-dimensional electron density calculations based on X-ray diffraction are often carried out to probe the structure of the intercalated species in two-dimensional inorganic hosts [33, 84–86]. This yields specific information about the orientation and structure of the intercalated species or at least eliminates certain conformational possibilities, which are incompatible with the diffraction data. In LDH systems, however, such calculations are usually impossible because the X-ray diffraction patterns of hybrid materials are often very ill defined [87, 88]. The electron density maps were obtained from Eq. 1:   ∞
2πlz ρ(z) = (1) F00l cos c l=0

The structure factors of the 00l reflections F00l were derived from their intensities corrected for Lorentz-polarization effects. The signs of the structure factors were directly obtained from the scattering contributions of Ca2 Al hydroxide layers assuming that the contribution of intercalated VBS is relatively small. c is the unit cell parameter. Seven reflections were used for Ca2 Al/Cl out A and eight reflections Ca2 Al/VBS out to d0016 = 2.13 ˚ A. The to d0021 = 1.12 ˚ 1D electron-density map of Ca2 Al/Cl was calculated from the structural data

In Situ Polymerization and LDH/Polymer Nanocomposites

141

(Fig. 16) [79]. As expected, the projection of the Ca2 Al/Cl structure along the c A due to Ca, axis gives rise to two strong symmetrical peaks separated by ∼ 1.4 ˚ the most electron-rich atoms in the hydroxide layers, shifted alternately up and down along the c axis. To enable visualization of the hydroxide layer structure, a Gaussian deconvolution into five components was applied. The peak positions and their relative areas are in good agreement with the structure: Al3+ ions are located midway between the pair of Ca layers and OH groups on either side. Three additional peaks are observed between Ca2 Al hydroxide layers, due to water molecules arranged on either side of the central Cl anions. The 1D electron-density map of Ca2 Al/VBS (Fig. 17) shows a broadening A against ∼ 3.8 ˚ A for Ca2 Al/Cl) due to the water of the hydroxide layers (∼ 8 ˚ molecules and the sulfonate groups flanked at the outer parts of the interlayer space. The aromatic ring and the vinyl bond cause a broad region of medium electron density in the middle of the gallery. Gaussian deconvolution indicates four components. From the peak positions and their relative area it is inferred that VBS molecules are more likely to be perpendicularly oriented towards the hydroxide layers and that the bilayers are interpenetrating A at the gallery center. This arrangement brings the vinyl groups into by 1–2 ˚ an appropriate conformation for the polymerization reaction. The distances between anionic sites in the ordered hydrocalumite [76–80], A and 5.74 ˚ A along the a and b axes, respectively, are suitable for i.e., 9.96 ˚ a syndiotactic polymerization process with monomers alternating above and below the central plane, running along the a axis as illustrated in Fig. 18. However, in order to connect with each other, the VBS monomers must incline, which is consistent with the lamellar contraction observed upon heating and associated with the polymerization process (discussed in Sect. 2.2.3).

Fig. 16 a One-dimensional electron density distribution of Ca2 Al/Cl along the c axis b schematic structure of Ca2 Al/Cl c Gaussian deconvolution of the hydroxide layer

142

C. Taviot-Guého · F. Leroux

Fig. 17 a One-dimensional electron density distribution of Ca2 Al/VBS along the c axis b Gaussian deconvolution of the hydroxide layer c Gaussian deconvolution of the interlayer space and the resulting conformation of VBS molecules; in C2/c space group, VBS molecules are related by an helico¨ıdal axis 21 located midway in the interlamellar space

Fig. 18 Schematic representation of the in situ syndiotactic polymerization of VBS in Ca2 Al host running along the a axis. The signs “+” and “–” denote the location of VBS molecules above or below the central plane, respectively

For the less ordered hydrotalcite type sample, a syndiotactic polymer is also probable. The amount of intercalated species and likely location within the interlamellar space is determined by the amount of trivalent cations within the

In Situ Polymerization and LDH/Polymer Nanocomposites

143

Fig. 19 13 C CP-MAS spectra of samples after treatment at 200 ◦ C for 6 h in air a VBS molecule, b Zn3 Al/VBS and c Zn2 Al/VBS

√ hydroxide layers. The shortest M3+ – M3+ distance being a 3, a syndiotactic PSS can only be considered. Similarly, the in situ polymerization requires good matching between the host structure and the guest molecules, since it was established that the driving force is the mutual attraction between the positive charge of the layers and the sulfonate group [48]. Experimentally, it was shown that for lower layer-charge density i.e., Zn3 Al and Zn4 Al, the amount of interleaved monomers VBS was not sufficient to give rise to complete polymerization. Vinyl carbons were still observed after thermal treatment at 200 ◦ C whereas for the Zn2 Al host structure, which has the optimum charge density, the complete disappearance of any C = C contribution is observed (Fig. 19). The distance between interleaved monomers in the galleries of Zn3 Al is then too large for them to connect with each other. Furthermore, the polymerization reaction for free VBS was found to be incomplete under the same conditions, showing that the constraint supplied by the LDH structure is necessary. Lamellar confinement is forcing the monomers to accommodate initially as a bilayer and then upon thermal activation to complete the polymerization when the requirements stated above are met. 2.2.3 In Situ Polymerization Evidenced by 13 C Solid-State NMR and ESR Spectroscopies For the above M2 Al/VBS materials, the in situ polymerization phenomenon was clearly evidenced by 13 C solid-state NMR (Fig. 20). The signal of the car-

144

C. Taviot-Guého · F. Leroux

Fig. 20 13 C CP-MAS spectra of a Ca2 Al/VBS, and b Zn2 Al/VBS after heat treatment at different temperatures in air. (Reprinted from [70] with permission from Elsevier)

bon atom 1 C associated with the vinyl bond disappears progressively with the increase of the temperature and large humps located at ≈ 40–50 ppm are characteristic of CH and CH2 responses. Eventually, the resonance peaks of the materials resemble that of PSS. For Zn2 Al/VBS, the polymerization happens at a lower temperature than for Ca2 Al/VBS: the inception of the process occurs at 120 ◦ C, whereas it is initiated at 180 ◦ C for Ca2 Al/VBS. In addition, the conversion is spread over a large temperature range for Zn2 Al/VBS: a few monomers are still observed at 200 ◦ C, and thermal treatment at higher temperature is needed to observe the complete disappearance of any vinyl bond signal. In contrast, complete conversion is achieved at 200 ◦ C for Ca2 Al/VBS. The in situ polymerization process is accompanied by a lamellar contraction. Figure 21 gives the variation of the interlamellar distance as a function of the temperature in air for ZnR Al/VBS and Ca2 Al/VBS in the temperature range 150–200 ◦ C. As can be seen, the lamellar structure is greatly conA for d003 of Zn2 Al/VBS and 13.9 ˚ A for tracted in both cases, down to 14.5 ˚ Ca2 Al/VBS. Consistent with the NMR data, the variation of the interlamellar distance is spread over a wide range of temperature for Zn2 Al/VBS while for Ca2 Al/VBS the contraction occurs at a definite temperature as a consequence of the structural disorder and order, respectively, that differentiates these two LDH hosts. Above 280 ◦ C, the (110) reflection for Zn2 Al disappears, although the lamellar organization is maintained. On the contrary, for Ca2 Al, the intralamellar organization is still visible indicating a stronger structural cohesion for the latter LDH host. By comparison with the chloride phases M2 Al/Cl, the nanocomposites sustain a lamellar structure at higher tem-

In Situ Polymerization and LDH/Polymer Nanocomposites

145

Fig. 21 Variation of the interlamellar distance as a function of the temperature for ZnR Al/VBS and Ca2 Al/VBS (Reprinted from [70] with permission from Elsevier)

perature. The presence of polymer thus reinforces the inorganic framework. Similar contractions have been reported elsewhere under thermal treatment and attributed to in situ polymerization processes: for α, β-aspartate (from A) [89] or for acrylate (from 13.6 to 12.6 ˚ A) [50, 51, 53] in LDH and 11.0 to 9.0 ˚ for conversion of ε-aminocaproic acid to nylon in a α-ZrP matrix (from 16.3 A) [29]. to 12.2 ˚ Surfactant molecules are usually used as a spacer, facilitating the incorporation of insoluble guest molecules (see the surfactant-mediated pathway in Sect. 2.1) or helping the delamination process to occur [40, 55, 56]. In addition to such properties, the possibility of polymerization may be of great interest. Based on this idea, sulfopropylmethacrylate (denoted as SPMA), a surfaceactive monomer, was recently incorporated into an LDH structure using the coprecipitation method [90]. By means of FTIR and 13 C CP-MAS solid-state NMR results, the authors provide clear evidence of the in situ polymerization of SPMA in Zn2 Al LDH galleries by thermal treatment (Fig. 22). When the surface-active monomer was confined, the reaction was complete without the use of any external chemical agent. Nevertheless, the thermal behavior of the free SPMA molecules and Zn2 Al/SPMA suggested that oxygen molecules may act as an external initiator for the polymerization and that the reaction was kinetically limited. Dispersed particles of small size were obtained after the polymerization process; this, added to an organophilic nature, makes them attractive for incorporation as nanofillers in polymers (see Sect. 3). Recent work reports the incorporation of aniline sulfonic acid derivatives, o- and m-aminobenzenesulfonic denoted as I and II, respectively, 3-amino4-methoxybenzenesulfonic (III), 3-aniline-1-propane sulfonic acid (IV), and

146

C. Taviot-Guého · F. Leroux

Fig. 22 13 C CP-MAS NMR spectra of SPMA (left part) and Zn2 Al/SPMA (right): a before and b after thermal treatment at 200 ◦ C in air for 2 hours. The labelling is according        to H2 1 C = 2 C(3 CH3 )4 CO2 5 CH2 6 CH2 7 CH2 SO3 and H2 1 C – 2 C(3 CH3 )4 CO2 5 CH2 6 CH2 7 CH2 SO3

4-aniline-1-butane sulfonic acids (V) in Cu2 Cr LDH via direct coprecipitation [55]. After a subsequent thermal treatment at 200 ◦ C in air, the degree of connectivity between guest monomers, i.e., dimerization and/or polymerization, was evaluated by ESR spectroscopy. ESR spectra of the hybrid materials exhibit a single Gaussian line with a peak-to-peak linewidth, noted as ∆Hpp , close to 1100 G ± 20 G and a Lande (g) factor of 2.0621 ± 0.0005, characteristics attributed to the combination of paramagnetic ions Cu2+ (3d9 ) and Cr3+ (3d3 ) (Fig. 23). By increasing the temperature, some additional narrow signals were observed, which may be classified according to their Lande factor. For the hybrid phases H(I, IV and V), the g value of the narrow signal (g = 2.0034 ± 0.0004) was typical of organic radicals and/or conduction electrons. For H(II and III), a broad asymmetric line exhibiting a partially resolved hyperfine structure (hfs) was assigned to isolated Cu2+ ions in an axially distorted octahedral environment (g// ≈ 2.34 and g⊥ ≈ 2.06). Hfs is due to the interaction of the unpaired electron with the nuclear spin I = 3/2 of Cu2+ ion, even when present in a small amount [91, 92]. The authors concluded that the presence of an organic structure impedes the formation of CuO-type clusters and pointed out the stabilizing role of the organics on the lamellar structure when connected to each other [93]. In contrast to the narrow signal of H(I), which exhibits a single Lorentzian profile, the feature observed for H(IV) and H(V) was more complex (Fig. 23 1b) with splitting into five lines indicating a superhyperfine structure (line denoted A in Fig. 23 2). This phenomenon is due to the hyperfine interaction between localized spin and two neighboring nitrogen nuclei (nuclear spin

In Situ Polymerization and LDH/Polymer Nanocomposites

147

Fig. 23 ESR spectra of Cu2 Cr/aniline sulfonic acid derivatives after a thermal treatment at 200 ◦ C for 4 h in air recorded at 105 K with a sweep width of a 6500 G b 150 G. H(I): o–aminobenzenesulfonic acid; H(I): m–aminobenzenesulfonic acid; H(III): 3-amino-4-methoxybenzenesulfonic acid; H(IV): 3-aniline-1-propane sulfonic acid; H(V): 4-aniline-1-butane sulfonic acid (left). Simulation of the ESR spectrum recorded at 105 K of the phase H(IV) after treatment at 473 K. The assignment of the lines is reported in the text (right)

I = 1) yielding (2nI + 1) = 5 lines. The simulation of the signal leads to the hyperfine parameter A = 12 ± 1 G, associated with a linewidth ∆Hpp = 12.0 G ± 0.5 G and a g-factor of 2.0033 ± 0.001. Line B results from the non-resolved hyperfine structure of the unpaired electrons interacting with the neighboring nitrogen and/or from another type of spin carrier with different localization within the organic structure. The presence of such a hyperfine interaction implies that a part of the unpaired electrons in the organic species are trapped in an environment composed of two nitrogen nuclei without delocalization. For the hybrids H(IV) and H(V), the spin carriers behave as isolated spins; this was evidenced by the evolution of the ESR parameters with temperature. The evolution of ∆Hpp was independent of the temperature, and the intensity of the signal increases with decreasing temperature. This behavior is characteristic of Curie-type magnetism, whereas for H(I), the positive temperature dependence of the ESR line width is typical of conduction electrons and the intensity of the signal is independent of the temperature which were interpreted by a Pauli-type mechanism. The authors concluded that the poly-

148

C. Taviot-Guého · F. Leroux

merization process was not favorable either when the amino group is located in the o-position to the sulfonate or in presence of the additional methoxy group, which both hinder the polymerization. The monomer connectivity was also not improved by using alkyl chain spacers (molecules IV and V) which gives rise to the formation of dimers. 2.3 Thermal Behavior of LDH/Polymer Nanocomposites We have seen that the presence of a polymer not only affects the crystallinity but also the dimension and morphology of the host material. Furthermore, it results in several other effects on the assembly. First by holding together the layers, it generally enhances the thermal stability. The collapse of the lamellar structure on heating is delayed. Ca2 Al/PVA was found to be stable up to a temperature of 400 ◦ C [94]. According to the authors, the nature of the interface between the organic and inorganic counterparts is responsible for such improved thermal stability. At high temperatures, the organoceramic transforms into inorganic solids with different compositions from those resulting from heat treatment of the pristine host material. As can be seen in the SEM micrographs in Fig. 24, the organic residue is encompassing the inorganic crystallites, thus preventing the crystallization of CaO. Onset of formation of ZnO crystallites on heating was found to be significantly delayed for Zn2 Al/PSS (Fig. 24c and d) [68]. Secondly, the polymer may act as a protective shell, since it not only delays the crystallization of by-products but also induces the formation of unusual solids after thermal treatment under an inert atmosphere. The nature of these solids clearly depends on the cations initially present in the hydroxide layers

Fig. 24 SEM micrographs of Ca2 Al/PVA treated at a 500 ◦ C and b 1000 ◦ C under air, and of c Zn2 Al/Cl and d Zn2 Al/PSS after thermal treatment at 600 ◦ C under N2 atmosphere. The bar represents 2 µm. (a and b reprinted from [94] with permission from ACS, c and d reprinted from [68] with permission from RSC)

In Situ Polymerization and LDH/Polymer Nanocomposites

149

but also on the organic functionality of the polymer. Chalcogenides such as ZnS or CaS [68, 70] or nitrides such as AlN [95] were thus obtained after thermal decomposition under an N2 atmosphere. When the calcined residues are amorphous, the XAS technique is useful to identify the decomposition products. For instance, XANES measurements at the sulfur K-edge for Zn2 Al/PSS were carried out. The spectra are dominated by a single white-line feature reflecting the 1s to np transitions and corresponding to localized, unfilled atomic or molecular states (Fig. 25). At 600 ◦ C, the white line for the PSS/LDH phase initially located precisely at the same energy as that of the PSS macromolecule, i.e., 2481.6 eV, is shifted either to higher energy after treatment in air, or to lower energy when the sample is heated under N2 atmosphere, indicating the formation of Na2 SO4 or ZnS, respectively. This corresponds in the first case to the oxidation of the sulfonate to sulfate, in the second to the reduction of sulfur to sulfide. EDX analyses show a phase segregation (Fig. 25 left part), with small particles arranged in stars composed mostly of Zn and S. EXAFS refinements indicate marked similarities with 2H-p wurtzite phase

Fig. 25 Combined analysis of the products obtained after thermal treatment of Zn2 Al/PSS at 600 ◦ C under a N2 and b air atmospheres. In the left part, the EDX analysis is presented, and in the right part, XANES (above) and EXAFS (below) spectra are displayed in comparison to ZnS c

150

C. Taviot-Guého · F. Leroux

of ZnS, as evidenced by the comparison of the two kχ(k) signals (Fig. 25, right part (bottom)). The thermal behavior of an Mg3 Al/acrylonitrile nanocomposite was investigated under an N2 atmosphere [95]. The acrylonitrile was intercalated into a surfactant-modified Mg3 Al LDH by means of displacement of Me(CH2 )11 OSO3 – anions, and subsequent polymerization in the presence of an initiator (benzoyl peroxide). Below 1600 ◦ C, the by-products were AlN, MgS and MgO, whereas above this temperature, submicron AlN grains were obtained as a single phase. During the carbothermal reduction, by acting as a shell the stratified polymer (PAN) prevents the collapse of the layers and the formation of mixed (Mg, Al) oxide, while the crystal growth is occurring in the interior of the particles, thus giving rise to submicron particles after decarbonization. The by-products were identified by X-ray diffraction.

Fig. 26 27 Al MAS spectra of M2 Al/Cl samples after thermal treatment at a 25 ◦ C, b 150 ◦ C, c 200 ◦ C, d 250 ◦ C, e 275 ◦ C, and f 300 ◦ C. (Black line for M = Ca and grey line for M = Zn). The chemical shifts are not corrected from the secondary effect of the quadrupolar interaction. (Reprinted from [70] with permission from Elsevier)

In Situ Polymerization and LDH/Polymer Nanocomposites

151

For Al-based LDH, the structural changes occurring during the thermal treatment are generally accompanied by a conversion of intralayer Al(OH)6 octahedra to interlayer AlO4 tetrahedra (Fig. 26). This conversion can be evaluated by single-pulse 27 Al solid-state or triple-quantum (3Q) 27 Al NMR under magic angle conditions in both cases [96, 97]. As much as 10% of the initially Oh -coordinated Al nuclei can be thus converted to Td without the collapse of the structure. Generally, the presence of polymer delays the temperature of conversion [48, 70]. 2.4 LDH/Biopolymers Only a few biopolymers have been incorporated between LDH lamellae. For instance, the incorporation of poly(α,β aspartate) was reported [89] to proceed by the condensation of aminosuccinic acid via a polysuccinimide intermediate, which rearranges to give polyaspartate at 220 ◦ C. It was found that the basal spacing decreases during the condensation process from 11.1 A, giving an available space of only 4.2 ˚ A for the accommodation of to 9.0 ˚

Fig. 27 SEM micrograph of a Zn2 Al/alginate with the schematic representation of the nanocomposite (right part) and Mg2 Ga/DNA obtained via direct coprecipitation carried out at two different temperatures: b room temperature and c 60 ◦ C. (Reprinted from [98] with permission from Elsevier)

152

C. Taviot-Guého · F. Leroux

the polymer. As mentioned in Sect. 2.1, the polymer may influence the intralayer composition: starting initially from a ratio Mg(II) to Al(III) of 2, the final product after incorporation of poly(α,β aspartate) has a ratio of 1.2 indicating an uptake of Al3+ cations. The presence of biopolymer also influences strongly the textural properties of the hybrid systems. This is exemplified in Fig. 27 by two bio-nanocomposites: Zn2 Al/alginate [98] and Mg2 M/DNA (M = Al, Ga) [99–107]. Alginic acid produced by brown seaweed is a heteropolysaccharide having a nonregular structure and was incorporated directly. LDH/DNA nanocomposites were prepared by substituting NO3 – anions in presynthesized Mg2 Al LDH or by direct incorporation for Mg2 Ga/DNA. These assemblies of biopolymers and LDH hosts give rise to unusual submicron morphologies: highly 2D-oriented materials are observed with a tubular shape completely different from the sand-rose morphology of the pristine materials. The bio-polymer is apparently acting as a glue to consolidate nanosized LDH particles into larger scale aggregates.

3 Future Developments Taking into account organo-modified and polymer LDH assemblies, some future developments may be envisioned. First, considering the assembly on its own, they may be of interest as electrode materials and sensors. The role of the LDH here is to increase the processability, since most of the intercalated polymers have excellent physical properties such as conductivity (PANI, PPY) but are difficult to process due to their lack of mechanical strength. It was shown that conjugated polymers interleaved into a 2D-host structure generally promote ion diffusion, as for PANI or PPY/V2 O5 nanocomposites (see Sect. 1) [31]. This is illustrated by the polymer-clay system, PPY/montmorillonite, which shows promising properties for both sensor and electrolysis applications [108]; polymer/LDH systems may also meet this challenge. Another important role of LDH materials is to encapsulate and effectively neutralize the negative charge of guest molecules. Nevertheless, surface adsorption and border effects overcompensate for the initial charge of LDH layers, leaving a residual negative charge. That has been evidenced by electrophoretic measurements on PSS- or DNA/LDH systems [107, 109]. This is of great interest when either the neutralization of the guest molecule is required or a weak affinity toward cationic species is desired. This was recently illustrated by the incorporation of bio-macromolecules such as AMP, CMP, GMP, ATP [103] which are nucleoside monophosphates, and DNA in LDH materials, which act as a barrier to prevent extra-cellular degradation but also as a vector by cancelling the electrostatic repulsive interaction between the biomolecule and the negatively charged cell; once incorporated into the lyosome, the LDH moiety dissolves allowing a progressive release of the

In Situ Polymerization and LDH/Polymer Nanocomposites

153

biomolecule [101]. LDH-based bio-hybrids may provide new opportunities as reservoirs and delivery carriers of functional biomolecules such as DNA, and therefore may find applications in gene therapy and drug delivery systems [100]. Choy et al. have shown that a biomolecule such as ATP can be exogenously introduced into eucaryotic cells. An overview of the chemistry and further delivery of these systems is provided in Fig. 28. Alginate is another biopolymer extensively studied for its gel formation property with applications in the food packaging and pharmaceutical industries as well as for membranes and as biosensors. Such LDH hybrid assemblies were recently used for the detection of cations such as Ca2+ [110]. In addition to the protective role, the organophilic properties of the resulting hybrid materials can be used to great benefit in a number of applications. This illustrates perfectly the enhanced properties of hybrid materials compared with those of the two separate components. On the other hand, lamellar structures present characteristic features that make them interesting in polymer reinforcement. In 1979, Kato et al. reported

Fig. 28 Schematic illustration of the hybridization and expected transfer mechanism of the bio-LDH hybrid into a cell. (Reprinted from [101] with permission from Wiley)

154

C. Taviot-Guého · F. Leroux

for the first time the properties of a nanocomposite formed by nylon-6 obtained from aminocaproic acid and a montmorillonite [111]. A few years later, researchers at Toyota achieved a great enhancement of the mechanical properties by using the same assembly but with the inorganic material dispersed in the polymer matrix [112–114]. Similarly by using organomodified LDH dispersed in either polyimide or epoxy resin, tensile strengths at break were found to increase. LDH/poly(imide) (PI) was prepared from an organo-modified LDH/aminobenzoic acid (AB) in the presence of poly(amic acid) [115]. The (AB) organo-modified LDH was mixed with pyromellitic anhydride (PMDA) and 4 -oxidianiline (ODA) in N,N-dimethylacetamide, giving rise to poly(amic acid) (PAA) precursors which transform into PI after an imidization reaction. By reacting an amine group with an anhydride, one can prepare a multi-branched polymer grafted into LDH galleries: one anhydride functional group of PMDA reacts with the amino group of AB and the other with ODA; ODA molecules which present two amino groups are acting as the linkage agent. At high molecular weight, extensive polymer penetration of PAA induces delamination of LDH-AB nanolayers. In the case of the reinforced organo-modified LDH/epoxy resin, it was prepared by mixing the epoxy diglycidyl ether of bisphenol (DGEBA), with a diamine curing agent polyoxypropylene diamine Jeffamine D400. The inorganic Mg2 Al hydroxide layers were reacting with aminolauric acid (AL) during the coprecipitation and the resulting material displays a paraffinic structure. The layers of Mg2 Al/AL were dispersed in the epoxy resin with different loadings as observed in Fig. 29. As for the LDH/AB-PI system, the glass transition temperature of Mg2 Al/AL-epoxy was found to increase with the inorganic loading as a result of the restriction of polymer chain mobility in presence of LDH nanolayers. The rigid Mg2 Al nanolayers enhance the stiffness of the nanocomposite and the thermal resistance, while the thermal expansion coefficient decreases. The formation process of the LDH/epoxy nanocomposites is represented in Fig. 29. Similarly, a poly(methyl methacrylate)/LDH nanocomposite was prepared using Mg3 Al/glycine organo-modified LDH which was first delaminated in formamide and then mixed with a solution of PMMA in acetone [116]. The LDH hydroxide layers were found to retain an exfoliated state when dispersed in the polymer. In terms of reinforcement, hydrocalumite/polymer nanocomposites may also be considered as promising in the field of cement-related materials. The so-called macro-defect-free (MDF) cements are based on the filling by polymer of the macroscopic voids responsible for the breakdown of cement. The hydrocalumite phase is formed during the hydration of cement and is known to readily incorporate polymers between its layers [71–75]. Clay-polymer nanocomposites have proven to be interesting candidates as gas barrier materials preventing permeation of volatile gases by creating a long path for diffusion and as flame-retardant materials. Previous work mainly involves the utilization of cationic clays, although LDH materials

In Situ Polymerization and LDH/Polymer Nanocomposites

155

Fig. 29 TEM micrographs of LDH/epoxy nanocomposites with various LDH contents: a 7 wt %, b 5 wt %, and c 3 wt %. The bar length is 50 nm (top) and the process of formation of LDH/epoxy nanocomposites is also shown (bottom). (Reprinted from [115] with permission from Elsevier)

may also be promising for this type of application by virtue of their large number of hydroxyl groups. Poly(ethylene-grafter-maleic acid) (PE-g-MA) presents a slower thermo-oxidative behavior with temperature than pure PEg-MA when organo-modified LDH was dispersed in the material [117]. In the same vein, flame-retardant properties of polyamide 6 (PA-6)/polypropylene (PP) blends containing nano-LDH and NH4 -polyphosphate were found to be improved by the synergistic effect of the mineral addition, and the LDH promotes cross linking and char formation during the thermal degradation of the blends [118]. Finally, in addition to the different possible functions (protection, transport, compatibilizer), the notion of constraint cannot be discarded. Indeed, when constrained, organic polymers give rise to carbonaceous materials presenting a high surface area associated with a rather good control of the porosity, the effect of the constraint being to inhibit cross-linking reactions between carbon species and thus increase the microporosity. After charring, carbons exhibiting high surface area, 1020 m2 /g, associated with a microporous volume of 0.31 cm3 /g [119] were obtained from poly(styrene 4-sulfonate)/LDH, more than twice the values obtained in the absence of con-

156

C. Taviot-Guého · F. Leroux

straint [120]. Recently, it was shown that even better textural properties may be obtained when using the in situ polymerization of styrene sulfonate [121]. Taking advantage of the large microporous volume, the carbons were studied as electrochemical supercapacitors, and capacitance of 100 F/g was obtained for carbons obtained from PSS/LDH in an acidic medium. Acknowledgements The authors would like to thank Dr. Marc Dubois, El Mostafa Moujahid, and Laetitia Vieille for their contributions.

References 1. Hawthorne DG, Hodgkin JH, Loft BC, Solomon DH (1974) J Macromol Sci Chem A 8:649 2. Wang S, Ahmad Z, Mark JE (1993) Polymer Bull 31:323 3. Messersmith PB, Giannelis EP (1994) Chem Mater 6:1719 4. Bujdak J, Hackett E, Giannelis EP (2000) Chem Mater 12:2168 5. Krishnamoorti R, Vaia RA, Giannelis EP (1996) Chem Mater 8:1728 6. Landry CJT, Coltrain BK, Landry MR, Fitzgerald JJ, Long VK (1993) Macromolecules 26:3702 7. Choi MH, Chung IJ, Lee JD (2000) Chem Mater 12:2977 8. Wang Z, Lan T, Pinnavaia TJ (1996) Chem Mater 8:2200 9. Wang Z, Pinnavaia TJ (1998) Chem Mater 10:1820 10. Shi H, Lan T, Pinnavaia TJ (1996) Chem Mater 8:1584 11. Aranda P, Ruiz-Hitzky E (1992) Chem Mater 4:1395 12. Nazar LF, Wu H, Power WP (1995) J Mater Chem 5:1985 13. Komori Y, Sugahara Y, Kuroda K (1999) Chem Mater 11:3 14. Kerr TA, Leroux F, Nazar LF (1998) Chem Mater 10:2588 15. Kanatzidis MG, Bissessur R, DeGroot DC, Schindler JL, Kannewurf CR (1993) Chem Mater 5:595 16. Wang L, Schindler J, Kanewurf CR, Kanatzidis MG (1997) J Mater Chem 7:1277 17. Tsai HL, Schindler JL, Kannewurf CR, Kanatzidis MG (1997) Chem Mater 9:875 18. Lagaly G (1999) Appl Clay Sci 15:1 19. Nam HJ, Kim H, Chang SH, Kang, SG, Byeon SH (1999) Solid State Ionics 120:189 20. Kanatzidis MG, Tonge LM, Marks TJ (1987) J Am Chem Soc 109:3797 21. Wang L, Brazis P, Rocci M, Kannewurf CR, Kanatzidis MG (1998) Chem Mater 10:3298 22. Kerr TA, Wu H, Nazar LF (1996) Chem Mater 8:2005 23. Liu P, Gong K (1999) Carbon 37:701 24. Liu P, Gong K, Xiao P, Xiao M (2000) J Mater Chem 10:933 25. Matsuo Y, Hatase K, Sugie Y (1998) Chem Mater 10:2266 26. Matsuo Y, Tahara K, Sugie Y (1997) Carbon 35:113 27. Eastman MP, Bain E, Porter TL, Manygoats K, Whitehorse R, Parnell RA, Hagerman ME (1999) Appl Clay Sci 15:173 28. Nazar LF, Zhang Z, Zinkweg D (1992) J Am Chem Soc 114:6239 29. Ding Y, Jones DJ, Maireless-Torres P, Rozière J (1995) Chem Mater 7:562 30. Wu CG, DeGroot DC, Marcy HO, Schindler JL, Kannewurf CR, Bakas T, Papaefthymiou V, Hirpo W, McCarthy WJ, Kannewurf CR, Marks TJ (1999) Solid State Ionics 32/33:594

In Situ Polymerization and LDH/Polymer Nanocomposites

157

31. Leroux F, Koene B, Nazar LF (1996) J Electrochem Soc 143:L181 32. Wu CG, DeGroot DC, Marcy HO, Schindler JL, Kannewurf CR, Bakas T, Papaefthymiou V, Hirpo W, Yesinowski YP, Liu YJ, Kanatzidis MG (1995) J Am Chem Soc 117:9229 33. Liu YJ, Schindler JL, DeGroot DC, Kannewurf CR, Hirpo W, Kanatzidis MG (1996) Chem Mater 8:525 34. Leroux F, Goward G, Power WP, Nazar LF (1997) J Electrochem Soc 144:3886 35. Schöllhorn R (1996) Chem Mater 8:1747 36. Oriakhi CO, Faar IV, Lerner MM (1996) J Mater Chem 6:103 37. Oriakhi CO, Faar IV, Lerner MM (1997) Clays Clay Miner 45:194 38. Drezdzon MA (1988) Inorg Chem 27:4628 39. Dimotakis ED, Pinnavaia TJ (1990) Inorg Chem 29:2393 40. Bubniak GA, Schreiner WH, Mattoso N, Wypych F (2002) Langmuir 18:5967 41. Adachi-Pagano M, Forano F, Besse JP (2000) Chem Comm 91 42. Leroux F, Adachi-Pagano M, Intissar M, Chauvière S, Forano C, Besse JP (2001) J Mater Chem 11:105 43. Leroux F, Besse JP (2001) Chem Mater 13:3507 44. O’Leary S, O’Hare D, Seeley G (2002) Chem Comm 1506 45. Hibino T, Jones W (2001) J Mater Chem 11:1321 46. Li B, Hu Y, Zhang R, Chen Z, Fan W (2003) Mater Res Bull 38:1567 47. Rajamathi M, Nataraja GD, Ananthamurthy S, Kamath PV (2000) J Mater Chem 10:2754 48. Moujahid El M, Leroux F, Besse JP (2002) J Mater Chem 12:3324 49. Boehm HP, Steinle J, Vieweger C (1977) Angew Chem Int Ed Engl 16:265 50. Tanaka M, Park IY, Kuroda K, Kato C (1989) Bull Chem Soc Jpn 62:3442 51. Rey S, Mérida-Robles J, Han KS, Guerlou-Demourgues L, Delmas C, Duguet E (1999) Polym Int 48:277 52. Vaysse C, Guerlou-Demourgues L, Duguet E, Delmas C (2003) Inorg Chem 42:4559 53. Rey S, Mérida-Robles J, Han KS, Guerlou-Demourgues L, Delmas C, Duguet E (1999) Polym Int 48:277 54. Challier T, Slade RCT (1994) J Mater Chem 4:367 55. Moujahid El M, Dubois M, Besse JP, Leroux F (2002) Chem Mater 14:3799 56. Wie X, Epstein AJ (1995) Synth Met 74:123 57. Kuramoto N, Geniès EM (1995) Synth Met 68:191 58. Chen SA, Hwang GW (1997) Polymer 38:3333 59. Tang H, Kitani A, Yamashita T, Ito S (1998) Synth Met 96:43 60. Shimizu S, Saitoh T, Uzawa M, Yuasa M, Kano K, Maruyama T, Watanabe K (1997) Synth Met 85:1337 61. Rhee SW, Jung DY (2002) Bull Kor Chem Soc 23:35 62. Tagaya H, Sato S, Kuwahara T, Kadokawa J, Masa K, Chiba K (1994) J Mater Chem 4:1907 63. Kanatzidis MC, Wu CG (1989) J Am Chem Soc 111:4139 64. Hofmeister W, von Platen H (1992) Cryst Rev 3:3 65. Vucelic M, Jones W, Moggridge GD (1997) Clays Clay Miner 45:803 66. Intissar M, Segni R, Payen C, Besse JP, Leroux F (2002) J Solid State Chem 167:508 67. Roussel H, Briois V, Elkaim E, de Roy A, Besse JP (2000) J Phys Chem B 104:5915 68. Moujahid El M, Besse JP, Leroux F (2003) J Mater Chem 13:258 69. Vieille L, Taviot-Guého C, Besse JP, Leroux F (2003) Chem Mater 15:4369 70. Vieille L, Moujahid El M, Taviot-Guého C, Besse JP, Leroux F(2004) J Phys Chem Solids 65:385 71. Bonapasta AA, Buda F, Colombet P, Guerrini G (2002) Chem Mater 14:1016

158

C. Taviot-Guého · F. Leroux

72. Popoola OO, Krive WM, Young JF (1991) J Am Ceram Soc 74:1928 73. Birchall JD, Howard AJ, Kendall K (1981) Nature 289:388 74. Merlin F, Lombois H, Joly S, Lequeux N, Halary JL, Van Damme H (2002) J Mater Chem 12:3308 75. Drabik M, Slade RCT (2004) Interface Sci 12:375 76. Allmann R (1977) N Jb Miner Mh, p. 136 77. Terzis A, Philippakis S, Kuzel HJ, Burzlaff H (1987) Z Kristallogr 181:29 78. Sacerdoti M, Passaglia E (1988) N Jb Miner Mh, p. 462 79. Renaudin G, Kubel F, Rivera JP, François M (1999) Cem Concr Res 29:1937 80. Rousselot I, Taviot-Guého C, Leroux F, Leone P, Palvadeau P, Besse JP (2002) J Solid State Chem 167:137 81. Geismar G, Lewandowski J, de Boer E (1991) Chemiker Zeitung 115:335 82. Zeng X, Ungar G (1998) Polymer 19:4523 83. Adach T, Takahashi H, Ohki K, Hatta I (1995) Biophys J 68:1850 84. Liu YJ, Schindler JL, DeGroot DC, Kannewurf CR, Hirpo W, Kanatzidis MG (1996) Chem Mater 8:525 85. De La Calle C, Tejedor MI, Pons CH (1996) Clays Clay Miner 44:68 86. Zhang X, Lerne MM (1999) Chem Mater 11:1100 87. Bauer J, Behrens P, Speckbacher M, Langhals H (2003) Adv Funct Mater 13:241 88. Itho T, Ohta N, Shichi T, Yui T, Tagaki K (2003) Langmuir 19:9120 89. Whilton NT, Vickers PJ, Mann S (1997) J Mater Chem 7:1623 90. Roland C, Besse JP, Leroux F (2004) Chem Mater 16:5512 91. Ardelean I, Peteanu M, Burzo E, Ciorcas F (1996) Solid State Comm 98:351 92. Ardelean I, Cozar O, Filip S, Pop V, Cenan I (1996) Solid State Comm 100:609 93. Moujahid El M, Dubois M, Besse JP, Leroux F (2005) Chem Mater 17:373 94. Messersmith PB, Stupp SI (1995) Chem Mater 7:454 95. Sugahara Y, Yokoyama N, Kuroda K, Kato C (1988) Ceram Int 14:163 96. Aramendia MA, Aviles Y, Borau V, Luque JM, Marinas JM, Ruiz JR, Urbano FJ (1999) J Mater Chem 9:1603 97. Rocha J, del Arco M, Rives V, Ulibarri MA (1999) J Mater Chem 9:2499 98. Leroux F, Gachon J, Besse JP (2004) J Solid State Chem 117:245 99. Choy JH, Kwak SY, Park JS, Jeong YJ, Portier J (1999) J Am Chem Soc 121:1399 100. Choy JH, Park JS, Kwak SY, Jeong YJ, Han YS (2000) Mol Cryst Liq Cryst 341:425 101. Choy JH, Kwak SY, Jeong YJ, Portier J, Park JS (2000) Angew Chem Int Ed 39:4042 102. Hwang SH, Han YS, Choy JH (2001) Bull Kor Chem Soc 22:1019 103. Choy JH, Kwak SY, Park JS, Jeong YJ (2001) J Mater Chem 11:1671 104. Kwak SY, Jeong YJ, Park JS, Choy JH (2002) Solid State Ionics 151:229 105. Kwak SY, Kriven WM, Wallig MA, Choy JH (2004) Biomater 25:3059 106. Choy JH (2004) J Phys Chem Solids 60:373 107. Leroux F, Ben Belkacem M, Guyot G, Taviot-Guého C, Léone P, Cario L, Desigaux L, Pitard B (2004) In: Chujo Y, Laine RM, Sanchez C, Schubert U (eds) (2005) Organic Inorganic Hybrid Materials. Mater Res Soc Fall Symp 847:223–228 108. Faguy PW, Ma W, Lowe JA, Pan WP, Brown T (1994) J Mater Chem 4:71 109. Wilson OC, Olorunyolemi T, Jaworski A, Borum L, Young D, Siriwat A, Dickens E, Oriakhi C, Lerner M (1999) Appl Clay Sci 15:265 110. Darder M, Lopez-Blanco M, Aranda P, Leroux F, Ruiz-Hitzky E (2005) Chem Mater 17:1969 111. Kato C, Kuroda K, Misawa M (1979) Clays Clay Miner 27:129 112. Usuki A, Kojima Y, Kawasumi M, Okada A, Fukushima Y, Kuruachi T, Kamigaito O (1993) J Mater Res 8:1179

In Situ Polymerization and LDH/Polymer Nanocomposites

159

113. Kamigaito O, Kawasumi M, Hohzaki M, Kojima Y, Okada A (1993) US Patent 4810743 114. Yano K, Usuki A, Okada A, Kurauchi T, Kamigaito O (1993) J Polym Sci A 31:2493 115. Hsueh HB, Chen CY (2003) Polymer 44:1151 116. Li B, Hu Y, Liu J, Chen Z, Fan W (2003) Colloid Polym Sci 281:998 117. Chen W, Qu B (2003) Chem Mater 15:3208 118. Xue J, Hao J, Zhao Y, Duan X (2002) Xiandai Huagong 22:34 (in Chinese) 119. Putyera K, Bandosz TJ, Jagieo J, Schwarz JA (1995) Appl Clay Sci 10:177 120. Putyera K, Jagiello J, Bandosz TJ, Schwarz JA (1995) Carbon 33:1047 121. Leroux F, Moujahid El M, Roland C, Vieille L, Taviot-Guého C, Besse JP, RaymondoPinero E, Béguin F (2004) In: Chujo Y, Laine RM, Sanchez C, Schubert U (eds) (2005) Organic Inorganic Hybrid Materials. Mater Res Soc Fall Symp 847:195–202 122. Isupov VP, Chupakhina LE, Ozerova MA, Kostrovsky VG, Poluboyarov VA (2001) Solid State Ionics 141:231 123. Leroux F, Aranda P, Besse JP, Ruiz-Hitzky E (2003) Eur J Inorg Chem 6:1242

Struct Bond (2006) 119: 161–192 DOI 10.1007/430_002 © Springer-Verlag Berlin Heidelberg 2005 Published online: 8 November 2005

Mechanistic and Kinetic Studies of Guest Ion Intercalation into Layered Double Hydroxides Using Time-Resolved, In-situ X-ray Powder Diffraction Gareth R. Williams · Aamir I. Khan · Dermot O’Hare (u) Chemistry Research Laboratory, Department of Chemistry, University of Oxford, Mansfield Road, Oxford OX1 3TA, UK [email protected] 1 1.1 1.2 1.3 1.4 1.5

Introduction . . . . . . . . . . . . . . . . . . . Background . . . . . . . . . . . . . . . . . . . Solid State Kinetics . . . . . . . . . . . . . . . Energy Dispersive X-ray Diffraction (EDXRD) In-situ Diffraction Apparatus . . . . . . . . . . Data Analysis . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

162 162 163 166 167 169

2 2.1 2.1.1 2.1.2 2.2 2.3 2.4 2.5

Survey of Recent Results . . . . . . . . . . . . . . . . . . . Intercalation and Deintercalation of Li salts into γ -Al(OH)3 Intercalation . . . . . . . . . . . . . . . . . . . . . . . . . . Deintercalation . . . . . . . . . . . . . . . . . . . . . . . . Staging Intermediates During Ion-exchange Reactions . . . Selective Intercalation Reactions . . . . . . . . . . . . . . . Intercalation of Biomolecules . . . . . . . . . . . . . . . . . Intercalation of Agrochemicals . . . . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

169 169 170 174 175 180 184 185

3

Conclusions and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . .

189

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

190

Abstract In recent years, a tremendous increase in our understanding of intercalation reactions has been attained through the use of in-situ techniques. In such studies, a noninvasive probe is employed, allowing the collection of large amounts of data on a reaction process. A non-invasive probe is preferred, rather than arresting the reaction and analysing the product thereby obtained (quenching). This is because the quenching process often affects the reaction product, and hence what is isolated can never be guaranteed to be typical of the reaction matrix. In the study of crystalline solids, one of the most useful non-invasive probes is X-ray diffraction, since this technique can provide evidence on the nature of the species present in the reaction and their concentrations. Through the use of synchrotron radiation, X-ray diffraction patterns may be collected on timescales as short as seconds, allowing accurate quantitative information about an intercalation process to be acquired. In this review, the application of this technique to a variety of intercalation processes relating to layered double hydroxides is surveyed. Keywords In-situ · Intercalation · Layered Double Hydroxide

162

G.R. Williams et al.

1 Introduction 1.1 Background The reaction of a guest molecule with a host lattice is a heterogenous process. This means that the kinetics as well as the thermodynamics of the process must be considered [1]. When intercalation into a layered double hydroxide (LDH) occurs, interactions between the positively charged layers and the initial interlayer anions are broken, and new bonding interactions form between the host and the new guest anions (Fig. 1). Additionally, since these reactions tend to take place in aqueous or alcoholic media, solvation factors are also relevant. The energy barrier to reaction will be determined by a balance of the relative strengths of the initial host-guest interactions, the post-reaction host-guest interactions and the solvation energies of both the initial and final anions. One way in which the overall activation energy barrier may be altered is by staging. This is a phenomenon in which some interlayer regions are completely vacant while other regions are fully or partially occupied. The number of layers between successively filled or partially filled layers defines the order of staging. The traditional models describing staging involve the layers having some degree of flexibility [2]. However, LDHs have relatively rigid layers [3], which means that these models are inapplicable here. Staging is illustrated by the schematic in Fig. 2. Determining the kinetics and mechanisms of intercalation reactions is not trivial. Quenching studies (in which an aliquot of the reaction suspension is removed and the solid product recovered through filtration) have frequently proved to be unreliable. The material isolated is often atypical of the reaction matrix as a whole, having been affected by the quenching process. Therefore, it is desirable to use a non-invasive probe to observe the reaction in situ , in real time. This allows the extraction of both qualitative and quantitative information on the kinetics of a process and the exact steps lying on the reaction pathway. A variety of techniques have previously been applied to monitor re-

Fig. 1 Schematic showing the process of intercalation into a layered double hydroxide

Mechanistic and Kinetic Studies of LDH intercalation processes

163

Fig. 2 Staging in a layered host material

actions in situ, including transmission electron microscopy [4], neutron and X-ray diffraction, optical microscopy, light scattering, NMR, IR/Raman spectroscopy and EXAFS [5–9]. In-situ techniques have previously been used to investigate a variety of intercalation and decomposition processes of LDHs. For example, Pérez-Ramirez et al.. have used in-situ IR and Raman studies to investigate the differences in the decomposition processes of cobalt, copper and aluminum containing LDHs [10]. The work described in this review involves the use of in-situ X-ray powder diffraction to investigate a variety of intercalation reactions involving LDHs. 1.2 Solid State Kinetics The fundamental processes involved in a solid state reaction are twofold. First, there is the reaction itself – the breaking and forming of bonds. Second, there is the transport of matter to the reaction zone. A number of models aiming to describe solid state reactions exist. They are generally based on sigmoidal kinetic curves. The general form of the kinetic equation is as follows: f (α) = kt

(1)

where αhkl (t) =

Ihkl (t) max Ihkl

(2)

Ihkl (t) represents the integrated intensity of the (hkl) reflection at time t, and max is the maximum intensity of that peak. Ihkl There exists a functional relationship between the extent of reaction, α, and the reaction time, t, correlated by the rate constant, k. The form of the function f is significant in terms of the reaction mechanism. It has, however, been suggested that the full mechanism of a reaction may not be understood solely from kinetic data [11, 12].

164

G.R. Williams et al.

The most commonly employed approach is that of Avrami and Erofe’ev. This rate law has been derived in a number of ways, indicating its general applicability and validity. This equation takes the form [– ln(1 – α)]1/n = k(t – t0 )

(3)

where t0 is the induction time for the reaction. This expression has been successfully applied to a number of solid state processes, including phase transformations [13], decompositions [14], crystallisation and intercalation reactions [15, 16]. The equation is found to fit most closely with the experimental data within the range 0.15 < α < 0.85. The exponent n is linked to the number of steps in the formation of a nucleus (this is a zone in the solid matrix at which the reaction occurs), β, and the number of dimensions in which the nuclei grow, λ. It can be difficult to distinguish β and λ without independent evidence, and β can fall to zero following the consumption of external nuclei sites. Hulbert has analysed the possible values of the exponent, n, for a variety of conditions of instantaneous (β = 0), constant (β = 1) and deceleratory (0 < β < 1) nucleation and for growth in one, two and three dimensions (λ = 1 – 3) [17]. He also considered the effects of a diffusion contribution to the reaction rate. This reduces the importance of the acceleratory process and reduces the value of n. For diffusion controlled processes, n = β + λ/2, whereas for a phase boundary controlled process n = β + λ. Possible values of n are summarised in Table 1. Interpretation of these values can be difficult, and a given value does not unequivocally allow the determination of the reaction mechanism. Nuclei growth may occur along any or all of the three cartesian axes. 3D growth corresponds to spheres, 2D growth to plates and 1D growth to rods. A nucleus can be defined as a point in space where two reactant phases come into contact and can react. For an intercalation reaction into a layered material, the potential nucleation sites (active sites) are simply the layer edges, where the guest molecules can gain access to the interlayer space. For a layered material, it is expected that the growth will be in two dimensions, corresponding to the movement of the guests between the layers of the host. If a perfect host lattice (without defects) is considered, then there are a given number of nucleation sites, s, at the crystalline faces. These are all identical, so the possibility of nucleation occurring is the same, p. Therefore, the nucleation rate is p × s. As the reaction proceeds, some of the layers become filled. The edges of these layers are no longer potential nucleation sites, and so as the reaction progresses, s decreases and hence the nucleation rate decreases. Thus, one might expect intercalation reactions to be 2D processes with deceleratory nucleation. However, there exist some other possibilities. The first of these is that nucleation may be instantaneous. This means that all the nucleation sites are instantly saturated as soon as the reaction begins. This will occur when the

Mechanistic and Kinetic Studies of LDH intercalation processes

165

Table 1 Nuclei growth models for solid state reactions. Values possible for intercalation into a layered host are highlighted in bold Dimension of growth (λ)

Nucleation rate (β)

Exponent value Phase boundary controlled (n)

Diffusion controlled (m)

1

Zero (instantaneous) Deceleratory Constant

1

0.5

1–2 2

0.5 – 1.5 1.5

Zero (instantaneous) Deceleratory Constant

2

1

2–3 3

1–2 2

Zero (instantaneous) Deceleratory Constant

3

1.5

3–4 4

1.5 – 2.5 2.5

2

3

transport of guests to the host in the reaction matrix is very rapid. Deceleratory nucleation will be seen when this transport rate is slower. In both the above cases, we have 2D processes. Following nucleation, the reaction may be either phase boundary controlled (i.e. the rate is limited by the rate at which the interlayer space expands to accommodate the guest) or diffusion controlled (i.e. the reaction rate is controlled by the rate at which the guests diffuse between the layers – the interlayer spacing expands instantly as the guests move). In addition to the above, there are further possibilities. When the rate of guest diffusion between individual layers is very large compared with the rate of nucleation at the edge of the crystal, there exists a situation in which the individual layers appear to fill instantly. In this case, when Avrami kinetics are applied to the system, the diffusion process being observed is not the diffusion of guest species between the layers, but the diffusion of filled layers parallel to the c-axis. Such 1D processes will consist of nucleation followed by diffusion control in the vast majority of cases, although phase boundary control is also possible if the rate of advancement of the phase boundary is also very rapid with respect to nucleation. In this case, instantaneous nucleation is not a possibility [18]. Finally, there is the possibility of a reaction mechanism that is entirely diffusion controlled, where the rate of nucleation does not play a role in determining the rate of reaction. In this case, Avrami-Erofe’ev kinetics do not

166

G.R. Williams et al.

Table 2 Possible equations describing diffusion controlled reactions Growth model

Equation

1D diffusion 2D diffusion 3D diffusion

α2 = kt (1 – α) ln(1 – α) + α = kt [1 – (1 – α)1/3 ]2 = kt

apply; different models exist to describe such entirely diffusion controlled processes with diffusion in various dimensions [19]. These are summarised in Table 2. n is found to be approximately 0.5 for these processes. The values of the exponent n that can be obtained for an intercalation reaction into a layered host are therefore limited. Possible values are highlighted in bold in Table 1. The most reliable way to find n is to use the method of Sharp and Hancock. This consists of taking logarithms of the Avrami-Erofe’ev equation, to give ln(– ln(1 – α)) = n ln k + n ln t .

(4)

The gradient of this graph therefore permits the determination of n, and the intercept allows k to be calculated. The advantage of using a Sharp-Hancock plot rather than a least squares fitting process with the Avrami equation is that if Avrami kinetics are not applicable, this can be seen in the former plot, and hence other kinetic models may be investigated. Purely diffusion controlled processes can be identified using a Sharp-Hancock plot: n is found to be 0.5 in such cases. 1.3 Energy Dispersive X-ray Diffraction (EDXRD) The desirability of using a non-invasive in-situ probe has already been discussed. There is, however, a problem, in that standard characterisation techniques are unable to penetrate bulky reaction vessels. As a result of this, little is known about the reaction dynamics or kinetics of intercalation reactions. A non-invasive probe which can interrogate a typical intercalation process is required. It is also necessary to employ short data collection times in order that kinetic information may be obtained. X-ray powder diffraction is a highly appropriate tool. It is non-invasive, and is a powerful characterisation technique when used in combination with ex-situ analyses. Angular-dispersive X-ray diffraction is used as a standard characterisation technique in the majority of solid-state laboratories. In this method, a constant-wavelength X-ray source is used. A detector sweeps a range of angles, and therefore Bragg reflections are separated by a spatial coordi-

Mechanistic and Kinetic Studies of LDH intercalation processes

167

nate. For in-situ studies, the alternative technique of energy dispersive X-ray diffraction (EDXRD) is preferred. This allows the simultaneous observation of a wide range of d-spacings. Each species involved in an intercalation reaction tends to have a distinct set of Bragg reflections. Hence, the simultaneous observation of the host material, any intermediate phases, and the product is possible using EDXRD. The change in intensity of the Bragg reflections can be monitored as a function of time, since data collection times can be as little as 10 s. This is well within the average reaction time for the intercalation reactions of LDHs. From these data, both qualitative and quantitative information regarding the mechanism and kinetics of a reaction can be obtained. A wide variety of reactions have been monitored using EDXRD [15, 20–25], in addition to the intercalation reactions of LDHs. In order to perform in-situ EDXRD experiments, a high-intensity polychromatic beam of X-rays is needed. Therefore, synchrotron radiation is required for these reactions. Much of the recent work in this area has been performed by O’Hare and co-workers on Station 16.4 of the UK Synchrotron Radiation Source. This station has been designed for kinetic studies of hightemperature, high-pressure reactions using energy dispersive diffraction and high energy X-rays. The polychromatic nature of the radiation allows the diffraction patterns of all the species present in a reaction matrix to be simultaneously recorded. For the energy dispersive method, combination of Bragg’s law with the Planck relationship E= gives

hc λ

(5)

hc (6) E where E is the energy of a photon in keV diffracted by a crystallographic plane A. At Station 16.4, a three-element detector system is with a d-spacing of d˚ used. Each detector is separated by ca. 2◦ in 2θ, and covers a different range of d-spacings. There is overlap between the range covered by each detector to ensure that no reflections are missed. 2d sin θ =

1.4 In-situ Diffraction Apparatus An experimental apparatus designed for the study of intercalation reactions by EDXRD has been developed by O’Hare and co-workers. This is shown in Fig. 3 [26, 27]. Typically, a Pyrex or quartz reaction vessel (normally fitted with a greaseless Teflon stopcock) is inserted into an aluminum heating block. Slits are cut into the side of the block in order to allow the incident X-ray beam in and

168

G.R. Williams et al.

Fig. 3 The experimental apparatus used to monitor intercalation reactions in situ. Reproduced with permission from Chem Mater (2005) 17:2632–2640

Fig. 4 Schematic representation of the detector arrangement on Station 16.4 of the SRS, Daresbury Laboratory, UK

Mechanistic and Kinetic Studies of LDH intercalation processes

169

the diffracted beam out. The block is mounted on a stand, allowing adjustment of the height and lateral position of the cell with respect to the incident beam. The reaction mixture is stirred by adding a magnetic flea to the reaction mixture. This is required in order that a constant sample concentration is maintained in the beam, and so that the solid does not settle. The apparatus illustrated is used for performing reactions at room temperature and higher temperatures. An alternative apparatus exists for low temperature reactions. This is essentially the same as the equipment in Fig. 3, except that a glycol cooled chamber is used. The experimental rig is mounted on an optical bench, along which it can be moved. This ensures that the diffracted intensity is maximised by locating the ideal alignment of the diffraction lozenge and the sample (Fig. 4). The position of the solid-state detector is controlled by a series of motors which drive it remotely to the required angle. 1.5 Data Analysis Gaussian curves are fitted to the data [28], and a peak area calculation program employed to integrate the area under each peak. This method is used to determine the evolution of peak area with time. The value of the extent of reaction, α, at a time t for any given Bragg reflection (hkl) is calculated using Eq. 2. EDXRD is a very powerful technique, although limitations include the requirement for synchrotron radiation. This limits the number of experiments that can be performed, due to the high cost and low availability of synchrotron beam time. Because of the large volume of the reaction vessel and the geometry of the instrument, the peak resolution of the energy dispersive data is also rather poor (∆E/E). This means that although it is possible to accurately monitor the course of a reaction, using the data for ab initio structure solution or structure refinement is precluded.

2 Survey of Recent Results 2.1 Intercalation and Deintercalation of Li salts into γ -Al(OH)3 The reactions of gibbsite (γ -Al(OH)3 ) with LiX salts (X = e.g. Cl, Br, I, NO3 , 1 1 2 CO3 , 2 SO4 ) in aqueous media to yield the LDHs [LiAl2 (OH)6 ]X · mH2 O (LiAl – X) are rare examples of reactions in which both cations and anions are similarly intercalated into the host lattice. Although other routes are available to synthesise this family of LDHs, including hydrolysis of aluminum

170

G.R. Williams et al.

tri(sec-butoxide) in the presence of lithium carbonate, and a hydrothermal preparation from hydrated alumina gel and LiOH, direct reaction of gibbsite or bayerite with the LiX salt is a very attractive method of preparing highly crystalline samples of these LDHs [29–32]. The work described in this section involves the use of EDXRD to study the intercalation of LiX into gibbsite. These are seen to be reversible processes – if LiAl – X is stirred in water, LiX leaches from the LDH matrix to give gibbsite and LiX. Under hydrothermal conditions at T > 150 ◦ C, boehmite (AlOOH) forms in addition to the Li salt [33]. 2.1.1 Intercalation The intercalation of a variety of LiX salts into gibbsite using EDXRD has been extensively studied by Fogg et al. [34]. In all cases, the reaction proceeded directly from the host gibbsite matrix to the reaction product – no other crystalline phases were seen. The (001) and (110) reflections of gibbsite are observed to decrease smoothly in intensity. The (002) reflection of the LiAlCl LDH is the major peak seen to grow into the diffraction pattern at 49.1 keV A). A smaller peak corresponding to the (004) reflection of (equivalent to 7.65 ˚ the product is also observed. No changes in either the peak position or its fullwidth half-maximum could be resolved. The inherently poor resolution of the energy dispersive detector means that the resolution function of the detector is poorer than the sample broadening term. Nonetheless, highly accurate kinetic parameters for the reactions were extracted by integrating the (002) reflection of the product phase as described in Sect. 1.5. The time evolution of the reaction between a 7.5 M solution of LiCl and gibbsite at 120 ◦ C is displayed in Fig. 5. Reactions were performed over a range of temperatures. Sharp-Hancock plots were used to determine the values of n and k for each temperature. The plots are given in Fig. 6; and the parameters extracted in Table 3. Plots of α vs. t/t0.5 (t0.5 is the half-life for the reaction) were found to be superimposable within experimental error, confirming the reaction mechanism to be consistent across the temperature range studied. The exponent n is found to be approximately 1 – indeed, a value of n = 1 was found to give a satisfactory fit to the majority of the experimental data. This corresponds to a 2D diffusion controlled reaction following instantaneous nucleation. Further evidence for this model is provided from an analysis of the intersection of the alpha vs. time curves. These intersect at ca. 0.5, which suggests that the loss of coherent diffraction from the host is matched by the gain in coherence of the product. In cases where nucleation is random, the rate of loss of coherent diffraction from the host lattice exceeds the rate of growth of the product phase, giving intersection at α < 0.5.

Mechanistic and Kinetic Studies of LDH intercalation processes

171

Fig. 5 Time-resolved in-situ EDXRD data showing the intercalation of LiCl into gibbsite at 120 ◦ C. a 3D stacked plot. b Plot of extent of reaction of the (001) reflection of gibbsite (◦) and the (002) reflection of [LiAl2 (OH)6 ]Cl· H2 O () as a function of time. Reproduced with permission from Chem Mater (1999) 11:1771–1775

The activation energy for the intercalation process was calculated using the Arrhenius relationship. k = Ae–Ea/RT .

(7)

A plot of ln k vs. 1/T (Fig. 7) gives an activation energy of 27 kJ mol–1 . A second set of experiments was performed to discover the effect of the LiCl concentration. The reaction rate was found to increase markedly as [LiCl] increased. When the concentration falls below 5 M the reaction failed

172

G.R. Williams et al.

Fig. 6 Sharp-Hancock plots for the temperature dependence of the intercalation of 10 M LiCl into gibbsite at 140 ◦ C (
) 120 ◦ C (), 100 ◦ C (◦), 80 ◦ C (∇) and 60 ◦ C (♦). Reproduced with permission from Chem Mater (1999) 11:1771–1775 Table 3 Kinetic parameters obtained from Sharp-Hancock analysis of the intercalation of LiCl into gibbsite at varying temperatures T/◦ C

n

k/10–3 s–1

t1/2 /s

140 120 110 100 90 80 60

1.06 1.00 1.07 1.06 1.05 0.80 0.82

1.25 0.740 0.582 0.476 0.480 0.277 0.181

720 960 1140 1320 1830 2280 4080

to reach completion in 3 h. A value of n = 1 gave a satisfactory fit to the experimental data in the majority of cases. A two-dimensional diffusion controlled mechanism following instantaneous nucleation operates in all cases. The order of reaction was determined from a plot of ln k vs. ln [LiCl] (Fig. 8). This plot has a gradient of 0.52, suggesting that the reaction is half-order with respect to LiCl. The rate equation is therefore rate ∝ [LiCl]0.5

(8)

The influence of the nature of the anion on the intercalation process was also studied. The intercalation of 5 M solutions of LiX (with X = Br, NO3 , OH and 1 ◦ 2 SO4 ) were followed at 120 C. The extent of reaction plots vary greatly between the different salts (Fig. 9). The plots shown in Fig. 8 are reduced time plots, in which the time is divided by the half-life of the reaction.

Mechanistic and Kinetic Studies of LDH intercalation processes

173

Fig. 7 Arrhenius plot for the intercalation of 10 M LiCl into gibbsite. Reproduced with permission from Chem Mater (1999) 11:1771-1775

Fig. 8 Plot of ln k vs. ln [LiCl], used to determine the order of reaction with respect to the concentration of LiCl. Reproduced with permission from Chem Mater (1999) 11:1771– 1775

Fig. 9 Reduced time (t/t0.5 ) plot for the intercalation of LiX into gibbsite. X = NO3 (◦), Br (), Cl (♦), OH (
), SO4 (∇). Reproduced with permission from Chem Mater (1999) 11:1771–1775

174

G.R. Williams et al.

The reaction half-life varies over a remarkable range, from 450 s for sulfate to 4800 s for nitrate. The very rapid reaction of sulfate is assumed to be a result of the high charge on this anion. The reaction rates correlate roughly with the ionic radii of the anions being intercalated, with the exception of nitrate. Inspection of the curves in Fig. 9 suggests that the reaction mechanism is significantly different for different anions. The t/t0.5 plots for LiCl, LiBr and Li2 SO4 are all almost superimposable, suggesting that the same mechanism is operational for all these reactions. For the intercalation of LiOH, there is a long induction period, but once the reaction begins it reaches completion within 10 min. In contrast, the nitrate reaction begins immediately but the reaction remains incomplete after 6 h. The values of n for LiBr and Li2 SO4 lie between 1 and 2, implying a twodimensional diffusion-controlled mechanism with deceleratory nucleation. However, the LiNO3 process has n = 0.5, indicating that this process is completely nucleation controlled. LiOH has n = 2.2, consistent with a phase boundary controlled process in two dimensions, with deceleratory nucleation again. In order to gain more insight into these reactions, further studies are ongoing in the group, and we hope to submit a paper imminently. 2.1.2 Deintercalation It was observed on many occasions that leaching of Li+ ions from LiAl–X occurs readily, yielding Al(OH)3 . These deintercalation reactions of the LiAl–X systems were studied by suspending the host in water and heating [35]. The Bragg reflections corresponding to the Al(OH)3 product were observed to be less intense and broader than those from conventionally prepared gibbsite. This is presumed to be a result of smaller crystallite domain size, due to stacking disorder in the layers. No Al3+ cations were observed in the supernatant solution, confirming that there is no dissolution of the hydroxide matrix during the reaction. The reaction is therefore perfectly topochemical. As with the forward reaction, deintercalation is a one-step process proceeding directly from the host to the product. Plots showing the progress of the reaction with LiAl – Cl are given in Fig. 10. The temperature dependence of the reaction was studied, and the activation energy of the reaction was calculated to be approximately 100 kJ mol–1. The exponent n was found to lie in the range 1 – 2, which is consistent with a 2D diffusion controlled reaction mechanism with deceleratory nucleation. The rate of reaction increases markedly with the amount of water added to the LDH: with very small amounts of water added, the deintercalation process does not go to completion. This effect is a result of the LiCl being leached into solution. An equilibrium exists between the LDH and gibbsite/LiCl in solution. The greater [LiCl], the further to the LDH side this lies.

Mechanistic and Kinetic Studies of LDH intercalation processes

175

Fig. 10 Plot of extent of reaction of the (002) Bragg reflection of [LiAl2 (OH)6 ]Cl · H2 O (
) and the 001 reflection of gibbsite (◦) at 80 ◦ C. Reproduced with permission from J Mater Chem (2004) 14:1443–1447

Experiments were performed with various LiAl – X LDHs, with X = Br, NO3 and 12 SO4 . As with the intercalation process, the nature of the anion exerts a powerful influence on the reaction. In the case of sulfate, the deintercalation reaction does not go to completion – only 40% of the available lithium sulfate was released. The deintercalation reaction initially proceeds very quickly, but the process is then halted. The rate of deintercalation is NO3 – > Cl– > Br– . This series does not correspond with data on the anion selectivity for intercalation into Al(OH)3 , which is SO4 2– > Cl– > Br– > NO3 – . Neither is there a correlation of the release data with the heats of hydration of the anions. The series observed arises because the intercalation and deintercalation processes are a balance of a number of factors, including interactions between the guest ions and the host matrix. 2.2 Staging Intermediates During Ion-exchange Reactions Staging is a method by which the energy barrier to an intercalation reaction may be reduced. However, the traditional models of staging [36] require the layers of the host material to bend, and therefore to be flexible. In contrast, LDHs have rigid layers, and are hence not expected to undergo staging processes. Fogg et al. performed a series of time-resolved experiments following the intercalation of a variety of carboxylic acids into the hexagonal form of the LiAl – Cl LDH using the EDRXD technique [37]. Surprisingly, they discovered

176

G.R. Williams et al.

that the intercalation does not proceed directly, in a one-step process, but instead proceeds via a then unprecedented second stage intermediate. Similar results have been observed for the intercalation of phosphonate anions into this LDH [38, 39]. Data for the intercalation of methylphosphonate (MPA) at pH 8 are shown in Fig. 11. A schematic showing the staging process is given in Fig. 12. Experiments have also been performed to investigate the effect of the layer stacking sequence and the initial interlayer anion on the intercalation process. Whereas intercalation of LiX salts into gibbsite gives the hexagonal form of the LiAl – X LDH (with an ABA stacking sequence), performing similar reactions with bayerite or nordstrandite produce rhombohedral LDHs with an ABCA stacking sequence.

Fig. 11 Time-resolved in-situ EDXRD data showing the intercalation of methylphosphonate into hexagonal [LiAl2 (OH)6 ]Cl · H2 O. a 3D stacked plot. b Plot of extent of reaction vs time for the (002) reflections of the host and product and the (004) reflection of the intermediate. Reproduced with permission from Chem Commun (2003) 15:1816–1817

Fig. 12 Schematic illustrating the intercalation of methylphosphonate into hexagonal [LiAl2 (OH)6 ]Cl · H2 O. a Host material, with all interlayer spaces occupied by Cl. b Second stage intermediate in which alternate interlayer spaces are occupied by Cl and methylphosphonate. c First stage product with all interlayer spaces occupied by methylphosphonate. Reproduced with permission from Chem Commun (2003) 15:11816–1817

Mechanistic and Kinetic Studies of LDH intercalation processes

177

Ex-situ experiments by Fogg and co-workers suggested that the rhombohedral LiAl – Cl LDH did not form staging intermediates when reacted with dicarboxylate salts [40]. In-situ measurements have confirmed that these reactions are indeed direct one-step processes. Similarly, staging is not seen for the intercalation of phosfonate salts (Fig. 13a). The alteration of the layer stacking sequence therefore has a profound effect on the reaction pathway [41]. The initial interlayer anion also plays a strong role in determining the pathway of the reaction. Neither the hexagonal nor rhombohedral forms of LiAl – NO3 exhibit staging; the alpha vs. time curves cross at α ≈ 0.5, strongly suggesting a direct transformation from host to product. This is illustrated in Fig. 13b,c. When Br is the initial interlayer anion, the situation is more interesting. For hexagonal LiAl – Br, no staging was seen for phthalate, nor for BPA and PPA. The alpha vs. time curves of the host and product are observed

Fig. 13 3D stacked plots of the in-situ EDRXD data for the intercalation of methylphosphonate into a Rhombohedral LiAl – Cl. b Hexagonal LiAl – NO3 . c Rhombohedral LiAl – NO3

178

G.R. Williams et al.

to cross at α ≈ 0.5. In contrast, a poorly crystalline intermediate is seen for MPA intercalation at pH 8 (Fig. 14). For EPA, succinate, fumarate, maleate and terephthalate with hexagonal LiAl – Br second stage intermediate phases are also observed. Again, the crystallinity of these tends to be rather poor, which makes detection difficult, although once detected these phases can be assigned to be second stage intercalates with a high degree of certainty. In the case of the rhombohedral form of LiAl – Br, the results are also interesting. The intercalation of maleate, phthalate, terephthalate, EPA, BPA and PPA is observed to proceed in a one-step process, directly from the host to the product, again with the alpha vs. time curves intersecting at ca. 0.5. The intercalation of MPA at pH 8, proceeds differently however (Fig. 15). From the 2D plot, it is clear that the host and product (003) alpha vs. time curves cross at ca. α ≈ 0. However, it is not possible to resolve an intermediate reflection at any time. Similarly, in the 3D plot, the host reflection can be seen declining in intensity, but there is a period in which no Bragg reflections are observed before the product phase finally grows in. If this reaction is quenched halfway between the disappearance of the host reflection and the emergence of the product reflection then a material with the X-ray diffration pattern given in Fig. 16 is isolated. The assignments of the Bragg reflections depicted in Fig. 16 are presented in Table 4. From Fig. 16 and Table 4, it is possible to see that the isolated material contains three phases: the host, the product and an intermediate phase. The intermediate is identified as being A. The failure to observe the ina second stage intermediate, with d003 = 20.8 ˚ termediate phase in-situ can be attributed to a combination of two factors: the poor crystallinity of the intermediate phase, and also the relatively poor resolution of the EDXRD detector on Station 16.4. Similarly, poorly crystalline intermediates are observed on the reaction pathway for the intercalation of succinate and fumarate.

Fig. 14 Time-resolved data for the intercalation of methylphosphonate into hexagonal LiAl – Br. a 3D stacked plot. b Extent of reaction vs. time plots for host, intermediate and product

Mechanistic and Kinetic Studies of LDH intercalation processes

179

Fig. 15 Time-resolved data for the intercalation of methylphosphonate into rhombohedral LiAl – Br. a 3D stacked plot. b Extent of reaction vs. time plots for host, intermediate and product

Fig. 16 Ex-situ X-ray powder diffraction pattern for the material isolated by quenching the reaction between rhombohedral LiAl – Br and MPA at the point where the concentration of intermediate is expected to be greatest Table 4 Assignments of the Bragg reflections shown in Fig. 16 No.

d-spacing

index

1 2 3 4 5 6

20.8 12.6 10.4 7.7 6.8 6.3

2nd stage 003 1st stage 003 2nd stage 006 host 003 2nd stage 009 1st stage 006

180

G.R. Williams et al.

Therefore, both the layer stacking sequence and the initial interlayer anion play a major role in determining the reaction mechanism. This is expected to be a result of these factors affecting the strength of the initial host-guest interactions, and the strength of the interactions between the layers and the final guest species. More information is given in [41]. Additionally, staging has recently been reported for the LDHs [Zn2 Al(OH)6 ]Cl · mH2 O and [Zn2 Cr(OH)6 ]Cl · mH2 O [42, 43]. 2.3 Selective Intercalation Reactions The selective intercalation of guests into solid hosts offers the potential for application in catalysis and separation science. An excellent case in point is zeolites, which exhibit shape and size selective inclusion properties and are used for an enormous variety of processes [44, 45]. Additionally, a number of layered materials have been reported to possess selective intercalation properties, including layered metal phosphonates [46, 47], montmorillonite [48], magnesium aluminum oxide [49], and layered double hydroxides [50–59]. Studying such reactions in-situ could yield vast amounts of useful information on the factors causing the selectivity and the mechanisms by which one isomer is favored over others. Unfortunately, in many cases EDXRD is not a suitable technique. The interlayer separations for the intercalates of two isomeric competing anions are often very similar, and in many cases cannot be resolved using the EDXRD detector. However, in some favorable cases EDXRD has been used to study selective intercalation reactions. [LiAl2 (OH)6 ]Cl · H2 O has been shown to exhibit shape selectivity for isomeric dicarboxylates and sulfonates [50, 51]. In-situ EDXRD has been used to study the competitive intercalation of 1,2-, 1,3- and 1,4-benzenedicarboxylate (BDA). The reaction is found to be over 95% selective for 1,4-BDA. A syringe pump was used to add a 1:1:1 mixture of the isomers to a suspension of the host material in water. The early stages of the reaction proceed with the decay of the host (002) reflection, accompanied by the growth of reflections corresponding to second stage intermediates (see Sect. 2.2). Following this, the first product peak observed has a d-spacing of A. This corresponds to the interlayer spacing of either the 1,2-BDA or 15.1 ˚ 1,3-BDA intercalate. The decay of this peak then ensues, accompanied by the A (Fig. 17). This corresponds to the terephthalate growth of a peak at 14.2 ˚ intercalate. Similar experiments have been performed with fumaric and maleic acids (trans and cis 1,4-but-2-enedicarboxylic acid). Identical observations were made – at the beginning of the reaction, a broad Bragg reflection centered A is seen to grow in to the patterns. This corresponds to the second at 10.0 ˚ stage intercalates of maleate and fumarate. Upon continued addition of the diA appears. This is the maleate intercalate basal carboxylates, a phase at 12.9 ˚

Mechanistic and Kinetic Studies of LDH intercalation processes

181

Fig. 17 Integrated intensity of the Bragg reflections at 15.1 ˚ A (
) and 14.2 ˚ A () as a function of time during the addition of all three benzecarboxylic acids to hexagonal LiAl – Cl. Reproduced with permission from Chem Mater (1998) 10:351–356

spacing. This then undergoes rapid exchange with fumarate anions, giving A. a shift in the peak position to 12.2 ˚ If the guests are intercalated separately, then the two product Bragg reflections should grow in smoothly at constant energy values, indicating that the shift in peak position is due to exchange rather than guest reorientation in between the layers. However, it is likely that the initial phase is not the pure maleate intercalate, but that both maleate and fumarate are intercalated between the same layers, leading to the observation of the higher d-spacing. The

Fig. 18 Schematic diagram showing the kinetic and thermodynamic intercalate phases of hexagonal LiAl – Cl. a Host material. b Kinetic intercalate phase containing a mixture of 1,4- and 1,2-BDA. c Thermodynamic intercalate phase containing soley 1,4-BDA, following the extrusion of 1,2-BDA

182

G.R. Williams et al.

kinetic phase then rapidly collapses, with extrusion of the thermodynamically less favoured anion to give the final product. An analogous process occurs for the BDA system. A schematic illustrating this process is given in Fig. 18. Similar studies have been performed using [Ca2 Al(OH)6 ](NO3 ) · 2H2 O (CaAl – NO3 ) [52]. Stirring the LDH with an equimolar mixture of the disodium salts of 1,2-BDA and 1,4-BDA resulted in intercalation of the 1,4 isomer with > 95% selectivity. The reactions are found to be very rapid when studied in-situ at 80 ◦ C, and so a dropwise addition approach was again employed. The reaction cannot be studied at lower temperatures because of the poor crystallinity of the product at low temperature. The reaction of CaAl – NO3 with 1,2-BDA is depicted in Fig. 19.

Fig. 19 Extent of reaction vs time for the reaction between CaAl – NO3 and 1,2-BDA. The reflection at 8.7 ˚ A corresponds to the host material, that at 14.8 ˚ A to the 1,2-BDA intercalate, and the 11.2 ˚ A reflection to Ca(1,2-BDA). Reproduced with permission from Chem Mater (2000) 12:1990–1994

Mechanistic and Kinetic Studies of LDH intercalation processes

183

Initially, the starting material, which exhibits a strong Bragg reflection at A, is observed. As addition of the guest solution continues, a peak at 8.7 ˚ A evolves. This is the first stage intercalate of 1,2-BDA, with all interlayer 14.8 ˚ spaces occupied by the dicarboxylate. No second stage intermediates are observed. Continued addition of 1,2-BDA causes a reduction in the intensity of A reflection, and a reflection with a d-spacing of at 11.2 ˚ A to evolve. the 14.8 ˚

Fig. 20 Intensity vs. time plot for the competition reaction between CaAl – NO3 and an equimolar mixture of 1,2- and 1,4-BDA. The 1,2-BDA intercalate has a d-spacing of 14.8 ˚ A (•) and the 1,4-BDA intercalate a d-spacing of 13.4 ˚ A (
). Reproduced with permission from Chem Mater (2000) 12:1990–1994

184

G.R. Williams et al.

This latter peak is believed to be due to the formation of calcium phthalate. Identical behavior is observed with 1,4-BDA. When an equimolar solution of the two salts is used, the reaction is shown to be extremely rapid – the host peak had decayed completely before data A collection commenced. Bragg reflections with d-spacings of 14.8 and 13.4 ˚ appear in the diffraction pattern, corresponding to the 1,2-BDA and 1,4-BDA intercalates respectively (Fig. 20). The phase with a d-spacing of 14.8 ˚ A is present in the reaction mixture for less than 15 min, while the 1,4-BDA reflection increases steadily in intensity, and at the end of the process is the sole phase present. This suggests that the same mechanism as for the LiAl – Cl selective intercalation reactions is operating here. That is, both anions are initially intercalated, followed by extrusion of the less favored isomer to give the thermodynamically favored product. 2.4 Intercalation of Biomolecules In recent years there has been significant interest in the intercalation of biomolecules, specifically vitamins, nucleic acids and drugs [60–71]. The latter is of particular importance, since the encapsulation of drug molecules into a host matrix is a useful way of achieving the controlled release of the active agent at the target site. This method has a number of advantages over traditional methods of drug delivery; most importantly, with a controlled release formulation it is possible to maintain an effective and non-toxic concentration of active agent in the body over an extended period of time. In contrast, with a traditional formulation the dosage is only in the effective range for a relatively short period of time, either side of which the drug concentration can be dangerously high or ineffectively low. Recent work by O’Hare and co-workers, among others, has concentrated on using LDHs as possible matrices for the storage and controlled release of drugs. Drugs such as those depicted in Fig. 21 have been successfully intercalated into LiAl – Cl, CaAl – NO3 and MgAl – NO3 ([Mg2 Al(OH)6 ]NO3 · xH2 O) LDHs. The intercalation of these species has been studied using time-resolved EDXRD. For intercalation into the LiAl – Cl system, a kinetic analysis of the data for naproxen (Nx), diclofenac (Df) and 4-biphenylacetic acid (4-Bpaa) suggests that the reactions are 2D diffusion controlled processes following instantaneous nucleation. In a number of cases, the importance of nucleation decreases at higher temperatures (T > 60 ◦ C), with a corresponding reduction in the value of n from 1 to 0.5. This latter value corresponds to a situation where nucleation plays no part in controlling the reaction rate. The data in Fig. 22 relate to the intercalation of Nx. The activation energies for these reactions were found to lie in the range 25 to 55 kJ mol–1 , consistent with a nucleation controlled process (diffusion con-

Mechanistic and Kinetic Studies of LDH intercalation processes

185

Fig. 21 Some of the drug molecules which have been successfully intercalated

trolled processes in aqueous media typically have activation energies of ca. 15 kJ mol–1). In contrast, the intercalation of gemfibrozil (Gz), 2-propylpentanoic acid (2-Pp) and ibuprofen (Ib) were found to be extremely rapid at room temperature, reaching completion in less than 2 min. Therefore, a dropwise addition process was employed to monitor these reactions. In the cases of Gz and 2-Pp, the reaction proceeded directly from the host to the product, with no intermediate phases being observed. However, the intercalation of Ib proceeded A decays in intensity, and via an intermediate phase. The host reflection at 7.7 ˚ A. This latter reflection then a new reflection grows in at a d-spacing of 10.4 ˚ A (Fig. 23). It is believed that the reflecshifts position to a d-spacing of 11.4 ˚ A is the (006) reflection of a second stage intermediate, and the tion at 10.4 ˚ A reflection is the (004) reflection of the first stage product phase. Ex-situ 11.4 ˚ studies offer additional evidence to support this possibility. The intercalation of the drugs into MgAl – NO3 was found to be too rapid to study using EDXRD. The intercalation of the drugs into CaAl – NO3 was seen to differ somewhat from the LiAl – Cl results. For the intercalation of Nx, the value of n is ca. 1.5, suggesting a 2D diffusion controlled reaction with deceleratory nucleation. In contrast, for Ib and 4-Bpaa, a value of 1 for n suggests 2D diffusion control again, but in these cases with instantaneous nucleation. The values of the activation energies for all three of these processes are consistent with a nucleation controlled process, ranging between 45 and 85 kJ mol–1. 2.5 Intercalation of Agrochemicals The possibility of using LDHs for the storage and controlled release of agrochemicals is another which has excited significant interest in the scientific

186

G.R. Williams et al.

Fig. 22 In-situ EDXRD data for the intercalation of naproxen into hexagonal LiAl – Cl. a 3D stacked plot for the reaction at 31 ◦ C. b Plot of extent of reaction vs time for the (004) reflection of the intercalate over a range of temperatures

Mechanistic and Kinetic Studies of LDH intercalation processes

187

Fig. 23 3D stacked plot illustrating the intercalation of ibuprofen into hexagonal LiAl – Cl as monitored using EDXRD. The presence of an intermediate phase is clearly visible

Fig. 24 A selection of agrochemicals which have been successfully intercalated

community in recent years. Agrochemicals are bioactive agents employed to improve the quality and production yields of commercial food crops. Their usage is increasing year-on-year, but their persistence in the soil, air and sur-

188

G.R. Williams et al.

Fig. 25 Time-resolved EDXRD data for the intercalation of MCPA into hexagonal LiAl – Cl at 50 ◦ C. a 3D stacked plot. b Extent of reaction vs time plot showing the decline in intensity of the (002) reflection of the host (
) and increase in intensity of the product (002) (•)

face and ground water presents significant enviromental issues. A number of studies have been undertaken to assess the suitability of clay type minerals (including LDHs) as immobilising matrices for agrochemical species [72–85]. Recent work by O’Hare and coworkers has been directed towards the intercalation of a variety of agrochemicals, such as those in Fig. 24, into the LiAl – Cl, CaAl – NO3 and MgAl – NO3 LDH systems.

Mechanistic and Kinetic Studies of LDH intercalation processes

189

All the agrochemical moieties were successfully intercalated, and fully characterised. Several phases were observed for the intercalation of glyphosate, depending on the pH at which the reaction was carried out, owing to the fact that glyphosate has multiple ionisable protons. Kinetic analyses of the herbicide intercalation reactions were performed using time-resolved in-situ EDXRD. For the intercalation of 2,4-D, MCPA and picloram, n was found to be approximately 1, indicating a 2D diffusion controlled reaction following instantaneous nucleation. For the mecoprop reaction, n was found to be ca. 1 at lower temperatures, but was reduced to 0.5 at higher temperatures, suggesting a change from a nucleation control mechanism to diffusion control. The reaction with glyphosate was seen to be very rapid, and hence diffusion control operates here. Adding the guest solution to the host suspension dropwise showed that the reaction is a one-stage process, as in the case of the other reactions, proceeding directly from the host to the first stage products, with no intermediate phases lying on the reaction coordinate. EDRXD data for the intercalation of MCPA at 50 ◦ C are given in Fig. 25.

3 Conclusions and Outlook The above has been a brief discussion of the use of in-situ, time-resolved EDXRD techniques to study the intercalation reactions of LDHs. A variety of other hosts such as metal dichalcogenides have also been investigated, but these fall outside the scope of this review. In general, it is possible to obtain good kinetic and mechanistic information for the LiAl – Cl and CaAl – NO3 host materials. With MgAl – NO3 , the intercalation reactions tend to be too rapid to observe. Other LDH hosts such as [Cu2 Cr(OH)6 ]Cl · H2 O have also been investigated, but problems regarding the crystallinity of the host matrix have been encountered. In this article, a variety of systems have been presented in the hope of showing the wide range of intercalation processes to which the EDXRD technique may be applied. It has been shown that in some cases EDXRD may be used to extract quantitative kinetic information for a reaction. This allows us to gain significant insights into the probable reaction mechanisms, even if in some cases it is not possible to unambiguously determine the reaction mechanism. Activation energies may be calculated, which give additional credence to the mechanisms deduced from use of the Avrami-Erofe’ev model. In other cases, the reactions are observed to be complete in a few minutes, even at very low temperatures. In these cases, it is necessary to add the guest solution dropwise to a suspension of the host. This precludes a full kinetic treatment of the data but the mechanism of the reaction under these conditions can still be determined. In some cases, this has allowed the observation of staging intermediates, which are both very rare in rigid host materials such

190

G.R. Williams et al.

as LDHs, and very difficult to observe ex-situ. Additional information on the reaction pathway is necessary to isolate such intermediates. This work has produced significant advances in our understanding of intercalation reactions. A wide range of mechanisms is observed, which can be rationalised on the basis of the vast array of factors governing an intercalation reaction. Work in this area is continuing, and recent studies have included the intercalation of bicyclic and tricyclic carboxylates. The construction of a new third-generation synchrotron facility in the UK (Diamond) should allow us to make further advances in this area. Acknowledgements The authors would like to acknowledge the tremendous help of Dr. Alex Norquist, who drew many of the figures in this work, and also of Dr. Dave Taylor and Mr. Alfie Neild at the SRS. Additionally, the financial help of the EPSRC (Engineering and Physical Sciences Research Council) is gratefully acknowledged.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Bruce DW, O’Hare D (1996) Inorganic Materials. Wiley Herold A (1979) Intercalated Layered Materials. Dordrecht Solin SA (1997) Annu Rev Mater Sci 27:89 Cherns D, Ngo GP (1983) J Solid State Chem 50:7 Shi JM, Anderson MW, Carr SW (1996) Chem Mater 8:369 Twu J, Dutta PK, Kresge CT (1991) Zeolites 11:672 Shannon IJ, Maschmeyer T, Sankar G, Thomas JM, Oldroyd RD, Sheehy M, Madill D, Waller AM, Townsend RP (1997) Catal Lett 44:23 Schoemann BJ (1997) Zeolites 18:97 Walton RI, Francis RJ, Halasyamani PS, O’Hare D, Smith RI, Done R, Humphreys R (1999) Rev Sci Instrum 70:3391 Pérez-Ramirez J, Mul G, Mouljin JA (2001) Vib Spectrosc 27:75 Carter RE (1961) J Chem Phys 34:2010 Sharp JH, Brindley GW, Achar BNN (1966) J Am Ceram Soc 49:379 Sheridan AK, Anwar J (1996) Chem Mater 8:1042 Hancock JD, Sharp JH (1972) J Am Ceram Soc 55:74 Price SJ, Evans JSO, Francis RJ, O’Hare D (1996) Adv Mater 8:582 Wilkinson AP, Speck JS, Cheetham AJ, Natarajan S, Thomas JM (1994) Chem Mater 6:750 Hulbert SF (1969) J Br Ceram Soc 6:11 Evans JSO, Price SJ, Wong HV, O’Hare D (1998) J Am Chem Soc 120:10837 Jander W (1927) Z Anorg Chem 163:1927 Rey F, Sankar G, Thomas JM, Barrett PA, Lewis DW, Catlow CRA, Clark SM, Greaves GN (1995) Chem Mater 7:1435 Rey F, Sankar G, Thomas JM, Barrett PA, Lewis DW, Catlow CRA, Greaves GN (1996) Chem Mater 8:590 Davies AT, Sankar G, Catlow CRA, Clark SM (1997) J Phys Chem B 101:10115 Walton RI, Loiseau T, O’Hare D, Ferey G (1999) Chem Mater 11:3201 Norquist AJ, O’Hare D (2004) J Am Chem Soc 126:6673

Mechanistic and Kinetic Studies of LDH intercalation processes

191

25. Walton RI, Norquist AJ, Neeraj S, Natarajan S, Rao CNR, O’Hare D (2001) Chem Commun 19:1990 26. Clark SM, Nield A, Rathbone T, Flaherty J, Tang CC, Evans JSO, Francis RJ, O’Hare D (1995) Nucl Ins Meth 97:98 27. Evans JSO, Francis RJ, O’Hare D, Price SJ, Clarke SM, Flaherty J, Gordon J, Nield A, Tang CC (1995) Rev Sci Instrum 66:2442 28. Clark SM (1995) J Appl Crystallogr 28:646 29. Serna CJ, White JL, Hem SL (1977) Clays Clay Miner 25:384 30. Poppelmeier KR, Hwu S-J (1987) Inorg Chem 26:3297 31. Nemurdy AP, Isupov VP, Kotsupalo NP, Boldyrev VV (1986) Russ J Inorg Chem 31:651 32. Nayak M, Kutty TRN, Jayaraman V, Periaswamy G (1997) J Mater Chem 7:2131 33. Thomas GS, Kamath PV (2002) Mater Res Bull 37:705 34. Fogg AM, O’Hare D (1999) Chem Mater 11:1771 35. Tarasov KA, Isupov VP, Chupakhina LE, O’Hare D (2004) J Mater Chem 14:1443 36. Daumas N, Herold A (1969) Comptes Rendues Acad Sci Paris C 268:373 37. Fogg AM, Dunn JS, O’Hare D (1998) Chem Mater 10:356 38. Williams GR, Norquist AJ, O’Hare D (2003) Chem Commun:1816 39. Williams GR, Norquist AJ, O’Hare D (2004) Chem Mater 16:975 40. Fogg AM, Freij AJ, Parkinson GM (2002) Chem Mater 14:232 41. Williams GR, O’Hare D (2005) Chem Mater 17:2632 42. Pisson J, Taviot-Gueho C, Israeli Y, Leroux F, Munsch P, Itie JP, Briois V, MorelDesrosiers N, Besse JP (2003) J Phys Chem B 107:9243 43. Taviot-Gueho C, Leroux F, Payen C, Besse JP (2005) Appl Clay Sci 28:111 44. Dai PSE (1995) Catal Today 26:3 45. Venuto PB (1994) Microporous Mater 2:297 46. Cao G, Mallouk TE (1991) Inorg Chem 30:1434 47. Brousseau LC, Aoki K, Garcia ME, Cao G, Mallouk TE (1993) NATO ASI Ser, Ser C 400:225 48. Ogawa M, Hirata M, Kuroda K, Kato C (1992) Chem Lett 21:365 49. Sato T, Kato K, Yoshioka Y, Okuwaki A (1992) J Chem Technol Biotechnol 55:385 50. Fogg A, Dunn JS, Shyu S-G, Cary DR, O’Hare D (1998) Chem Mater 10:351 51. Fogg AM, Green VM, Harvey HG, O’Hare D (1999) Adv Mater 11:1466 52. Millange F, Walton RI, Lei LX, O’Hare D (2000) Chem Mater 12:1990 53. Rhee SW, Lee JH, Jung D-Y (2002) J Colloid Interface Sci 245:349 54. Lei LX, Millange F, Walton RI, O’Hare D (2000) J Mater Chem 10:1881 55. Lei L, Vijayan RP, O’Hare D (2001) J Mater Chem 11:3276 56. Cardoso LP, Valim JB (2004) J Phys Chem Solids 65:481 57. Kuk W-K, Huh Y-D (1997) J Mater Chem 7:1933 58. Kuk W-K, Huh Y-D (1998) Bull Korean Chem Soc 19:1032 59. Yahaya AH, bin Hussain MZ, Seh WS (2003) J Phys Chem Solids 64:1113 60. Choy JH, Kwak SY, Park JS, Jeong YJ (2001) J Mater Chem 11:1671 61. Choy JH, Park JS, Kwak SY, Jeong YJ, Han YS (2000) Mol Cryst Liq Cryst Sci Technol, Sect A 341:425 62. Inoue T, Yasunaga T (1991) Kinki Daigaku Kogakubu Kenkyu Hokoku 25:51 63. Lotsch B, Millange F, Walton RI, O’Hare D (2001) Solid State Sci 3:883 64. Hwang S-H, Han YS, Choy JH (2001) Bull Korean Chem Soc 22:1019 65. Tronto J, dos Reis MJ, Silverio F, Balbo VR, Marchetti JM, Valim JB (2004) J Phys Chem Solids 65:475 66. Ambrogi V, Fardella G, Grandolini G, Perioli L (2001) Int J Pharm 220:23

192

G.R. Williams et al.

67. Fardella G, Ambrogi V, Perioli L, Grandolini G (1998) Proc Int Symp Controlled Release Bioact Mater 25:774 68. Fardella G, Grandolini G, Ambrogi V, Chiappini I (1997) Acta Technol Legis Med 7:125 69. Fardella G, Perioli L, Costantino U, Nocchetti M, Ambrogi V, Grandolini G (1997) Proc Int Symp Controlled Release Bioact Mater 24:1033 70. Tyner KM, Schiffman SR, Giannelis EP (2004) J Control Release 95:2419 71. Nakaeyama H, Takeshita K, Tsuhako M (2003) J Pharm Sci 92:2419 72. Lagaly G (2001) Appl Clay Sci 18:205 73. Weber JB (1970) Residue Rev 32:93 74. Weber JB (1972) Adv Chem Ser 111:55 75. Nennenmann A, Kulbach S, Lagaly G (2000) Appl Miner 2:867 76. El-Nahhal Y, Nir S, Polubesova T, Margulies L, Rubin B (1999) Pest Sci 55:857 77. Aguer JP, Hermosin MC, Calderon MJ, Cornejo J (2000) J Environ Sci Health, Part B 35:279 78. Carrizosa MJ, Calderon MJ, Hermosin MC, Cornejo J (2000) Sci Total Environ 247:285 79. Carrizosa MJ, Koskinen WC, Hermosin MC, Cornejo J (2001) Appl Clay Sci 18:223 80. Celis R, Cornejo J, Hermosin MC, Koskinen WC (1998) Soil Sci Soc Am J 62:165 81. Cox L, Koskinen WC, Celis R, Yen PY, Hermosin MC, Cornejo J (1998) Soil Sci Soc Am J 62:911 82. Hermosin MC, Cornejo J (1993) J Environ Qual 22:873 83. Casal B, Merino J, Serratosa JM, Ruiz-Hitzky E (2001) Appl Clay Sci 18:245 84. Lakraimi M, Legrouri A, Barroug A, De Roy A, Besse J-P (2000) J Mater Chem 10:1007 85. Inacio J, Taviot-Gueho C, Forano C, Besse J-P (2001) Appl Clay Sci 18:255

Struct Bond (2006) 119: 193–223 DOI 10.1007/430_007 © Springer-Verlag Berlin Heidelberg 2005 Published online: 23 September 2005

Applications of Layered Double Hydroxides Feng Li · Xue Duan (u) Ministry of Education Key Laboratory of Science and Technology of Controllable Chemical Reactions, Beijing University of Chemical Technology, 100029 Beijing, P.R. China [email protected] 1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

194

2 2.1 2.2 2.3 2.4 2.5 2.6

Applications in Catalysis . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . As Catalyst Supports . . . . . . . . . . . . . As Catalysts in Important Organic Reactions Environmental Catalysis . . . . . . . . . . . Catalysts in Natural Gas Conversion . . . . . Pillared LDHs as Catalysts . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

195 195 195 197 199 200 201

3 3.1 3.2 3.3 3.4

Applications in Ion-exchange and Adsorption . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adsorption of Inorganic Anions on Uncalcined and Calcined LDHs Adsorption of Organic Species on Uncalcined and Calcined LDHs . Adsorption of Organic and Inorganic Species on Organo-LDHs . .

. . . . .

. . . . .

. . . . .

. . . . .

203 203 204 205 207

4 4.1 4.2

Applications in Pharmaceutics . . . . . . . . . . . . . . . . . . . . . . . . . Applications in Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications in Biochemistry . . . . . . . . . . . . . . . . . . . . . . . . . .

208 208 210

5

Applications in Photochemistry . . . . . . . . . . . . . . . . . . . . . . . .

211

6

Applications in Electrochemistry . . . . . . . . . . . . . . . . . . . . . . .

213

7

Applications as Additives in Functional Polymer Materials . . . . . . . . .

214

8

Other Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

215

9

Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

217

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

217

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

Abstract Layered double hydroxides (LDHs), also known as anionic clays or hydrotalcite (HT)-like materials, are a family of materials that have attracted increasing attention in recent years. LDHs have anionic exchange capacity, and the ability to capture organic and inorganic anions makes them almost unique as inorganic materials. This paper deals with current and potential applications of these materials, including in catalysis, ionexchange/adsorption, pharmaceutics, photochemistry and electrochemistry; these are related to the wide range of possible compositions and versatile preparation methods for both uncalcined and calcined LDHs, as well as pillared LDHs.

194

F. Li ·X. Duan

Keywords Layered double hydroxides · Application · Intercalation · Calcination Abbreviations AEC Anion-exchange capacity CPO Catalytic partial oxidation DDS Dodecyl sulfate FCCU Fluidized catalytic cracking units HCH Hexachlorocyclohexane HT Hydrotalcite-like IBU Ibuprofen LDHs Layered double hydroxides MPV Meerwein-Ponndorf-Verley POM Polyoxometalate PEO Poly(ethylene oxide) SCR Selective catalytic reduction SRM Steam reforming of methanol UV Ultraviolet

1 Introduction Layered double hydroxides (LDHs), also known as hydrotalcite (HT)-like materials, are a class of synthetic two-dimensional nanostructured anionic clays whose structure can be described as containing brucite-like layers, where a fraction of the divalent cations coordinated octahedrally by hydroxyl groups have been replaced isomorphously by trivalent cations, giving positively charged layers with charge-balancing anions between them; some hydrogen bonded water molecules may occupy any remaining free space in the interlayer region [1, 2]. LDHs may be represented by the general formula [M2+ 1–x M3+ x (OH)2 ]x+ (An– )x/n · mH2 O, where M2+ (M = e.g. Mg, Fe, Co, Cu, Ni, or Zn) and M3+ (M = e.g. Al, Cr, Ga, Mn or Fe) are di- and trivalent cations respectively; the value of x is equal to the molar ratio of M2+ /(M2+ + M3+ ) and is generally in the range 0.2–0.33; An– is an anion. As a result, a large class of isostructural materials, which can be considered complementary to aluminosilicate clays, with widely varied physicochemical properties can be obtained by changing the nature of the metal cations, the molar ratios of M2+ /M3+ as well as the type of interlayer anions. LDHs, both as directly prepared or after thermal treatment, are promising materials for a large number of practical applications in catalysis, adsorption, pharmaceutics, photochemistry, electrochemistry and other areas [1–6]. This is due to their high versatility, easily tailored properties and low cost, which make it possible to produce materials designed to fulfil specific requirements.

Applications of Layered Double Hydroxides

195

LDH materials have relatively weak interlayer bonding and as a consequence exhibit excellent expanding properties. Therefore, over the past few years, increasing interest has been devoted to the use of these layered inorganic solids as host materials in order to create inorganic-organic host-guest hybrid (supramolecular) structures with desirable physical and chemical properties, in which the brucite-like layers may impose a restricted geometry on the interlayer guests leading to enhanced control of stereochemistry, rates of reaction, and product distributions. The possibility of the formation of three-dimensional pillared layered structures by appropriate intercalation processes opens up new perspectives for functional materials with novel properties [3, 4, 7].

2 Applications in Catalysis 2.1 Introduction The flexibility in composition of LDHs has led to an increase in interest in these materials. As a result of their relative ease of synthesis, LDHs represent an inexpensive, versatile and potentially recyclable source of a variety of catalyst supports, catalyst precursors or actual catalysts. In particular, mixed metal oxides obtained by controlled thermal decomposition of LDHs have large specific surface areas (100–300 m2 /g), basic properties, a homogeneous and thermally stable dispersion of the metal ion components, synergetic effects between the elements, and the possibility of structure reconstruction under mild conditions. In this section, attention is focused on recently reported catalytic applications in some fields of high industrial and scientific relevance (including organic chemistry, environmental catalysis and natural gas conversion). 2.2 As Catalyst Supports In 1971, LDHs containing different metal cations (such as Mg, Zn, Ni, Cr, Co, Mn and Al) with carbonate as interlayer anions, calcined at 473–723 K and partially or completely chlorinated, were reported to be effective as supports for Ziegler catalysts in the polymerization of olefins [8], with the maximum catalytic activity of polyethylene production observed for Mg/Mn/Al – CO3 LDH calcined at 473 K. Even earlier, calcined Mg/Al LDHs were used to support CeO2 for SOx removal from the emissions from fluidized catalytic cracking units (FCCU) [9, 10]. Some transition metal oxides have also been

196

F. Li ·X. Duan

supported on calcined Mg/Al LDHs materials for use in the selective catalytic reduction (SCR) of NO by NH3 [11], the oxidative dehydrogenation of n-butane [12, 13] and the vapor phase synthesis of isobutyraldehyde from methanol and n-propanol [14]. Recently, LDHs and their organic-derived products have been reported as supports for immobilization of enzymes [15–17]. For example, Ren et al. investigated the immobilization of penicillin G acylase in the interlayer galleries of an Mg/Al LDH pillared by glutamate ions [15]. Immobilized PGA was assembled in the interlayer galleries by a three-step process (see Fig. 1). It was found that that after 10 recycles, carried out in succession in a batch reactor, the immobilized enzyme displayed 90 % activity retention. The same authors also reported the immobilization of penicillin G acylase by physical adsorption on calcined Mg/Al – CO3 LDHs [16]. Immobilization on the support increases the acid resistance of the enzyme, and the activity of the immobilized enzyme was found to increase with decreasing Mg/Al molar ratio. The activity of the immobilized enzyme reached a maximum when the calcination temperature was between 723 and 823 K. Calcined Mg/Al – CO3 LDHs had a higher affinity for the enzyme than calcined Zn/Al – CO3 LDHs, but a lower percentage expressed activity. Both uncalcined and calcined LDHs have also been shown to be effective supports for noble metal catalysts [18–25]. For example, palladium supported on Cu/Mg/Al LDHs has been used in the liquid phase oxidation of limonene [24], and on calcined Mg/Al LDHs for the one-pot synthesis of 4methyl-2-pentanone (methyl isobutyl ketone) from acetone and hydrogen at atmospheric pressure [25]. In the latter case, the performance depends on the interplay between the acid-base and hydrogenation properties. More recently,

Fig. 1 Schematic illustration of the immobilization of penicillin G acylase on glutamatepillared LDHs. Reaction of the primary amino groups of the intercalated glutamate ions with glutaraldehyde through Schiff ’s base formation is followed by covalent coupling of the residual aldehyde groups with the free amino groups of the enzyme

Applications of Layered Double Hydroxides

197

a bimetallic Pd/Pt catalyst supported on calcined Mg/Al LDHs prepared by incipient wetness impregnation has also been used effectively for the hydrogenation and hydrogenolysis of naphthalene [26], and exhibits an intrinsic sulfur resistance due to the high hydrodesulfurization activity of the Pd/Pt pair, regardless of a possible contribution from the acid sites of the support. In addition, Choudary et al. studied nanopalladium(0) catalysts supported on Mg/Al LDHs prepared by ion-exchange with PdCl4 2– followed by reduction [27]. They found that, when used in ionic liquids, the catalysts not only exhibited higher activity and selectivity than the homogeneous PdCl2 system in the Heck olefination of electron-poor and electron-rich chloroarenes, but also displayed superior activity in the C – C coupling reactions of chloroarenes compared with other heterogeneous catalysts involving nanopalladium(0) on supports such as silica, alumina or Merrifield resin. A multifunctional ruthenium-grafted Mg/Al LDH catalyst has also been used in direct α-alkylation of nitriles with primary alcohols; a cooperative effect between the Ru species and the surface base sites was demonstrated [28]. It was shown that use of the same catalyst system could be extended to the onepot synthesis of α,α-dialkylated phenylacetonitriles via the base-catalyzed Michael addition reaction of α-alkylated phenylacetonitrile with activated olefins. 2.3 As Catalysts in Important Organic Reactions Many workers have reported the use of non-calcined LDHs in a large number of catalytic reactions, including epoxidation reactions of styrene using Mg/Al LDHs [29], Knoevenagel condensation using Ni/Al LDHs [30] or LDHs containing fluoride [31], hydroxylation of phenol over Co/Ni/Al LDHs [32], and liquid-phase carbonylation of methanol to methyl acetate catalyzed over tin-promoted Ni/Al LDHs [33]. In addition, Choudary et al. have reported a catalyst derived from calcined/rehydrated Mg/Al LDH (Mg/Al= 2.5 : 1), which was used for selective Michael addition reactions on methyl vinyl ketone, methyl acrylate, and simple and substituted chalcones by donors such as nitroalkanes, malononitrile, diethyl malonate, cyanoacetamide and thiols, with quantitative yields obtained under mild reaction conditions [34]. Recently, Corma et al. studied a similar catalyst for the synthesis of chalcones of pharmaceutical interest through the Claisen-Schmidt condensation between benzaldehyde and acetophenone [35]. They found that an Mg/Al (3 : 1) mixed oxide with a water content of 35 wt.% was the optimized solidbase catalyst, and could be used in the synthesis of several chalcones with anti-inflammatory, antineoplasic, and diuretic activities; in all cases excellent activity and selectivity were observed. Many reports have focused on the use of LDHs as precursors to mixed oxide catalysts formed by thermal decomposition. Such metal oxides are

198

F. Li ·X. Duan

known to promote a wide variety of industrially important base-catalyzed reactions [36]. For example, high activity and selectivity with calcined Mg/Al LDH catalysts have been reported in a wide variety of reactions, such as aldol [37–42], Claisen-Schmidt [43], and Knoevenagel condensations [44, 45], Meerwein-Ponndorf-Verley (MPV) reduction [46–48], Henry reaction of aldehydes [49], isomerizations [50, 51] and the polymerization of propylene oxide [52]. Velu et al. reported that Mg/Al mixed oxides derived by calcination of LDHs catalyzed the diasteroselective synthesis of nitroalkanols from aldehydes and nitroalkanes; these facile hydrogenation reactions, which occur with retention of configuration, allow the formation of pharmacologically important derivatives [49]. Di Cosimo et al. investigated the structural requirements and reaction pathways in condensation reactions of alcohols, using Mg/Al mixed oxides obtained by decomposition in N2 at 673 K for 4 h of LDH precursors with a wide range of composition [53], and found that the mechanistic pathway of the condensation reactions is affected not only by the catalyst acid-base properties but also by the chemical nature of the alcohols as well as steric factors. Kumbhar et al. found that Mg/Al-CO3 LDHs heated in N2 up to 823 K are highly active and selective catalysts for the liquid phase MPV reduction of carbonyl compounds using 2-propanol as the hydrogen donor [46]; the catalysts can also readily be regenerated. The same LDH precursor after thermal decarbonation and rehydration has been used in the cyanoethylation of alcohols [54], showing high activity and air stability (in contrast to other solid base catalysts). LDHs may also be suitable precursors for transition metal-containing heterogeneous catalysts [55–60]. For example, an Mg/Fe mixed oxide obtained from an LDH precursor was reported to be effective in the reduction of aromatic nitro-compounds using hydrazine hydrate under mild reaction conditions. The catalyst was readily regenerated and the high activity and selectivity were unaffected [59]. On calcination and reduction, these and similar materials usually give rise to well-dispersed and stable supported metal particles, possessing both basic and acidic sites. Velu et al. studied the vapor phase alkylation of phenol or m-cresol to give gasoline additives or chemical intermediates using calcined Mg/Al, Mg/Cr or Mg/Fe LDHs, and found a direct correlation of composition and acid-base properties of the catalyst with the observed activity and selectivity towards O- and C-alkylated products [61]. Oxides obtained from Ni/Mg/Al LDHs have been recently studied as catalysts for reactions such as the hydrogenation of acetonitrile and the partial oxidation of methane [62–64], as well as the oxidative dehydrogenation of n-butane and propene [65]. Furthermore, optimization of the catalytic properties for mixed oxides in the hydrogenation of nitriles involves a compromise between metal reducibility and the acid-base nature of the catalyst [66]. Choudary et al. have reported the selective reduction of aldehydes

Applications of Layered Double Hydroxides

199

to alcohols by calcined Ni/Al LDHs [67]. Calcined Cu/Mg/Al LDHs have been studied as potential catalysts for hydrogenation of cinnamaldehyde [68]. It was found that Cu2+ ions embedded in the Mg/Al oxide matrix are the key redox site for this reaction. Recently, Gao et al. reported that Mg/Zn/Al/Fe mixed oxides obtained from an LDH precursor gave ethylbenzene conversion as high as 53.8 % and a styrene selectivity of 96.7 % at 773 K in the dehydrogenation of ethylbenzene [69]. The high catalytic activity and stability of the Mg/Zn/Al/Fe catalyst were attributed to the presence of a large number of strong acid sites and a moderate number of base sites on the catalyst. A higher content of iron oxide species favors the redox cycle and also enhances the dehydrogenation activity. Li/Fe/Al oxide systems obtained from Li/Fe/Al LDH precursors have been proposed as effective catalysts for ethylbenzene dehydrogenation using CO2 as the oxidant [70]. The best catalytic performance was found for catalysts containing 15–45 wt.% of Fe. Rives et al. reported the use of Mg/V mixed oxides obtained from V(III)substituted LDH precursors as catalysts for the oxidative dehydrogenation of propane and n-butane [71]. Their results indicated that the relative amounts of Mg3 VO4 and MgO, which depend on the V(III) content of the starting LDHs, determine the performance of the catalysts. Recently, the efficacy of LDHs as catalyst precursors for the synthesis of carbon nanotubes via catalytic chemical vapor deposition of acetylene has been reported by Duan et al. [72]. Nanometer-sized cobalt particles were prepared by calcination and subsequent reduction of a single LDH precursor containing cobalt(II) and aluminum ions homogeneously dispersed at the atomic level. Multi-walled carbon nanotubes with uniform diameters were obtained. In addition, Mg/Al and Co/Al LDHs containing Ru3+ in the layers have also been used as precursors to mixed oxides that are highly efficient catalysts for the oxidation of allylic and benzylic alcohols [73, 74]. 2.4 Environmental Catalysis Calcined LDHs have been investigated as potential materials for the reduction of SOx and NOx emissions from FCCU in oil refineries [75, 76]. Corma et al. found that mixed oxides obtained from a Cu/Mg/Al LDH precursor were the most effective at catalyzing both the oxidation of SO2 to SO4 2– in the FCC regenerator and the reduction of sulfates to H2 S, which may be recovered, in the reducing atmosphere of the cracking zone [76]. Calcined Cu/Mg/Al LDHs [77] and Co/Mg/Al LDHs [78], subsequently activated by heating under H2 , can simultaneously remove SOx and NOx . In both cases, reduced transition metal species have been proposed as the active sites.

200

F. Li ·X. Duan

Ni/Al, Mg/Co, Co/Al and Cu/Al LDHs, in some cases containing Rh, La or Pd, have also been studied as precursors to similar de-NOx and de-SOx additives [79–82]. Catalysts obtained by calcination of Cu/Mg/Al LDH precursors exhibit enhanced catalytic performance in the SCR of NO compared with the traditional Cu2+ -impregnated Mg/Al mixed oxide [79], because the SCR reaction occurs between gas-phase NO and NH3 strongly adsorbed on Cu-containing phases and/or highly dispersed CuO clusters. The use of LDH precursors allows the concentration of each metal component in the mixed oxide to be controlled and leads to well-defined, high surface area catalysts. Calcination of LDHs at 773 K results in an intermediate non-stoichiometric mixed oxide structure, which undergoes rehydration in aqueous media, and regenerates the original layered structure with different anions in the interlayers; this reconstruction process is often called the “memory effect” and has been discussed in Chapter 2. Recently, Corma et al. made use of this effect as an elegant way of improving the catalytic reduction of nitrates in water [83]. The use of calcined Cu/Mg/Al LDHs as a support was found to reduce the problems associated with mass transfer limitations observed with Pd-Cu/Al2 O3 catalysts, and the new concept of “active support” was introduced. In this case the nitrate ions are forced to be located between the positively charged layers of the regenerated LDH and therefore are reduced to nitrite ions, nitrogen or, to a much lower extent, ammonia. These neutral species are released into the solution, reducing the problems related to diffusion limitations that strongly affect the selectivity of the reaction with traditional catalysts. Hydroxide anions, formed during the reduction process, are intercalated into the interlayer galleries and can be exchanged in the presence of more nitrate anions, leading to the start of a new reaction cycle. 2.5 Catalysts in Natural Gas Conversion In the future, hydrogen is expected to become an important fuel in automobiles and electric power plants. Therefore, applications of Ni/Al, Ni/Mg/Al or Ni/Ca/Al catalysts obtained from LDH precursors in the preparation of syngas by catalytic partial oxidation (CPO) of methane have been widely investigated, with promising results [84–89]. More recently Vaccari et al. reported a series of active and stable catalysts for the CPO reaction, based on Rh and Ru nanoparticles interacting strongly with MgO or spinel matrices, which were prepared by calcination and reduction of LDH precursors [90, 91]. They found that high Mg/Al ratios depressed the formation of the spinel and led to a narrow distribution of particle size of the support. The sample with the highest percentage of MgO phase showed the highest activity. Tsyganok et al. studied a combined partial oxidation and dry reforming of methane (PO-DRM) to syngas at 1123 K over noble metals supported on Mg/Al mixed oxides formed from corresponding LDH precursors [92]. The results indi-

Applications of Layered Double Hydroxides

201

cated that Ru showed the most attractive catalytic performance. The amount of noble metal could be lowered from 2.0 to 0.1 % of the weight of support without any decrease in catalytic activity or in selectivity to syngas. Ni/Al, Co/Al, Cu/Al, Rh/Mg/Al and Rh/Ni/Mg/Al catalysts obtained from LDH precursors have been investigated in CO2 -reforming of CH4 [93, 94]. In addition, Tsyganok et al. also found that Mg/Al LDHs intercalated with an [Ni(EDTA)]2– chelate complex could be used as a precursor to an efficient catalyst for the above reaction [95]. Copper-containing catalysts, especially Cu/Zn or Cu/Zn/Al mixed oxides, have been the most studied catalysts for steam reforming of methanol (SRM) due to their high selectivity and activity [96–98]; the active sites in Cu-containing catalysts are metallic Cu species [97]. A few reports have shown that Cu-based catalysts, such as Cu/Zn/Al mixed metal oxides derived from LDHs [99] and as-prepared Cu/Al LDHs that have not been preactivated [100], are active catalysts in SRM. The high activity and selectivity of the non-calcined Cu/Al LDH for producing H2 is related to the ease of reduction of Cu2+ species to a highly dispersed metallic Cu phase. In addition, Velu et al. have recently demonstrated that an oxidative steam reforming of methanol reaction over a series of Cu/Zn/Al(Zr)-oxide catalysts derived from Cu/Zn/Al(Zr)-hydroxycarbonates containing LDH/aurichalcite phases, gives hydrogen selectively with a methanol conversion of around 90 % at around 503 K [101, 102]. Since ethanol is a renewable raw material that can be cheaply produced by the fermentation of biomass, ethanol reforming as a source of H2 has environmental benefits. Recently, Velu et al. reported for the first time the steam reforming of ethanol in the presence of added O2 , which was performed over a series of Cu/Ni/Zn/Al mixed oxide catalysts derived from LDH precursors [103]. They observed an ethanol conversion close to 100 % at 573 K over all the catalysts. Cu-rich catalysts favor the dehydrogenation of ethanol to acetaldehyde. The addition of Ni was found to favor C – C bond rupture, producing CO, CO2 and CH4 . Segal et al. have also studied the catalytic decomposition of alcohols, including ethanol, for in situ H2 generation in a fuel stream using a non-calcined Cu/Al LDH-derived catalyst [104]. 2.6 Pillared LDHs as Catalysts The intercalation of anionic species into LDHs is an interesting alternative for the immobilization of catalytic complexes. Special attention is being paid to LDHs containing bulky and stable anions, e.g. polyoxometalates (POMs), since they can give rise to a wide range of microporous materials [4]. Many years ago, Pinnavaia et al. reported the intercalation of POM anions into LDHs, and found that the products showed significant photocatalytic activity in the oxidation of isopropanol to acetone in the presence

202

F. Li ·X. Duan

of oxygen [105, 106]. In addition, the use of LDHs as layered host systems for intergallery immobilization of anionic photocatalysts has also been proposed by Pinnavaia et al. It was shown that anionic molecular species such as [Ru(4,7-diphenyl-1,10-phenanthrolinedisulfonate)]4– ions could be intercalated in LDHs without affecting their photochemical excited state lifetimes [107]. Guo et al. reported that trace amounts of aqueous organochlorine pesticides, such as hexachlorocyclohexane (HCH), could be totally degraded and mineralized into CO2 and HCl by near-UV irradiation of a suspension of Mg/Al LDH intercalated with paratungstate anions [108]. They demonstrated that photocatalytic degradation of the pesticide occurs in the interlayer galleries. It was also found that Zn/Al/W(Mn) mixed oxides, formed by calcination of POM-intercalated Zn/Al LDHs [109], exhibited higher photocatalytic activity in the degradation of HCH compared with the POM-LDH precursors. The immobilization of active components that possess biomimetic catalytic properties in LDHs, with the aim of increasing the lifetime of the catalyst system, has attracted the interest of many researchers. For example, Sels et al. described tungstate-exchanged LDHs as biomimetic catalysts for mild oxidative bromination and bromide-assisted selective epoxidation reactions [110, 111]. Furthermore, the activity of LDH – WO4 2– could be affected by varying the elemental composition of the layers of the LDH [112]. The long lifetime, stability toward leaching and high H2 O2 concentrations, and low cost of this heterogeneous catalysis system raise the prospect of being able to develop a clean and efficient industrial route to important organic chemicals. Recently, Choudary et al. reported the first example of catalytic Noxidation of tertiary amines by tungstate-exchanged Mg/Al LDHs in water [113], and the halodecarboxylation of α,β-unsaturated aromatic carboxylic acids to β-bromostyrenes has also been achieved for the first time, using a molybdate-exchanged Mg/Al LDH catalyst [114]; this latter catalyst was active for selective oxidation [115, 116]. Earlier, Barloy et al. [117] reported the application of Mn-porphyrins intercalated in LDHs in oxidation catalysis. They found that the catalyst was suitable for the expoxidation of cyclooctene, but that it was a poor catalyst for the hydroxylation of heptane. This was related to the accessibility of the metal center in the catalyst. Metallophthalocyanines possess biomimetic catalytic properties for the autoxidation of organic molecules in aqueous solution, but their longevity is very limited, however, and there has been increasing interest [118–124] in supporting these materials on LDHs. This type of catalytic performance makes LDHs an exceptionally promising class of supports for immobilization of metallomacrocycles for various applications. For example, Pinnavaia et al. [121] have reported the use of LDHimmobilized Co(II)phthalo-cyaninetetrasulfonate ([CoPcTs]4– ) for the autoxidation of a thiolate to a disulfide. Incorporation of the phthalocyanine

Applications of Layered Double Hydroxides

203

complex into an Mg/Al LDH was found to significantly improve the catalyst reactivity and longevity, compared with the homogeneous catalyst. A new stereoselective epoxidation catalyst based on a novel chiral sulfonato-salen manganese(III) complex intercalated in Zn/Al LDH was used successfully by Bhattacharjee et al. [125]. The catalyst gave high conversion, selectivity, and enantiomeric excess in the oxidation of (R)-limonene using elevated pressures of molecular oxygen. Details of the catalytic activities with other alkenes using both molecular oxygen and other oxidants have also been reported [126].

3 Applications in Ion-exchange and Adsorption 3.1 Introduction There has been considerable interest in the use of LDHs to remove negatively charged species by both surface adsorption and anion-exchange. Their high uptake levels of anionic species can be accounted for by their large surface area and high anion-exchange capacities (AEC) [127] and flexible interlayer space, which is accessible to polar molecular species as well as anions, and can accommodate very diverse materials such as contaminants from soils, sediments and water. The anion-exchange capacity of LDHs is affected by the nature of the interlayer anions initially present and the layer charge density (i.e. the M(II)/M(III) molar ratio in the brucite-like sheets). When the layer charge density is very high the exchange reaction may become difficult. LDHs have greater affinities for multivalent anions compared with monovalent anions [128, 129]. In particularly, the favorable lattice stabilization enthalpy associated with CO3 2– results in these anions being difficult to displace in anion-exchange reactions. LDHs can take up anion species from solution by three different mechanisms: surface adsorption, interlayer anion-exchange and reconstruction of a calcined LDH precursor by the “memory effect”. The “memory effect” [130] of LDHs, discussed in 2.3 above, is one of their most attractive features as adsorbents for anionic species. Calcination allows the recycling and reuse of the adsorbents with elimination of organic contaminants as CO2 and water [131]. The main advantages of LDHs over traditional anionic exchange resins are their higher AEC values and the fact that LDHs are resistant to high temperature treatments. In summary, contaminants that can be adsorbed by LDHs are those of anionic character, inorganic as well as organic. Although some polar organic

204

F. Li ·X. Duan

molecules can also be incorporated in the interlayer galleries, this type of reaction has only been rarely described [127, 132]. 3.2 Adsorption of Inorganic Anions on Uncalcined and Calcined LDHs To date, literature references concerning inorganic anions adsorbed by uncalcined and calcined LDHs cover anionic species such as oxoanions (e.g. NO3 – , SO4 2– , PO4 3– , AsO4 3– , CrO4 2– , Cr2 O7 2– ) [133–141] and monoatomic anions (e.g. F– , Cl– , Br– , I– ). Parker et al. investigated the anion sorption capacity of calcined Mg/Al LDHs in a cycle involving heat treatment, anion sorption, and carbonate exchange to remove the sorbed anions [135]. The total amounts of sorbed anions using samples of freshly calcined LDHs (673 K) and carbonate-free solutions decreased in the order SO4 2– > F– > HPO4 2– > B(OH)4 – > NO3 – . The most strongly bound anions are those with the smallest highest charge density. Moreover, the calcined LDHs are more effective sorbents than LDHs because interlayer carbonate anions in the latter are difficult to displace with the above anions. Goswamee et al. investigated the adsorption of Cr(VI) by various uncalcined and calcined Mg/Al, Ni/Al and Zn/Cr LDHs [136]. Adsorption of Cr2 O7 2– through ion exchange with uncalcined LDHs occurs more slowly and gives lower loadings compared with calcined LDHs, where the adsorption process occurs via rehydration. The nature and relative amounts of di- and trivalent metal ions in the LDH influence the adsorption process. Calcined Mg/Al LDHs with higher Al3+ content shows higher adsorption capacity. Furthermore, adsorption is higher for the Mg/Al LDH than for Ni/Al and Zn/Cr LDH. Dousova et al. [142] found that calcined Mg/Al LDHs were effective in removal of As (V) compounds from aqueous solutions at 293 K and neutral pH utilizing the “memory effect”. More than 70 % of As (V) compounds were removed from aqueous solution at low sorbent-solution ratios. Parida et al. also studied the affinity of Mg/Fe LDHs toward the removal of inorganic selenite (SeO3 2– ) from aqueous media [143]. The results indicated that the efficacy of SeO3 2– removal increases with a decrease in either pH or temperature. The adsorption behavior of Cr2 O7 2– and SeO3 2– on uncalcined and calcined Zr4+ -substituted Zn/Al/Mg/Al LDHs has also been described [144]. Samples calcined at 723 K exhibit high adsorption capacities for Cr2 O7 2– (1.6–2.7 meq/g) and SeO3 2– (1.1–1.5 meq/g). Incorporation of Zr4+ increases the adsorption capacity by up to 20 %, thus providing an alternative way of enhancing the adsorption capacity of these type of materials. As discussed in Chapter 1, however, recent work has suggested that the Zr4+ ions may not actually be incorporated in the LDH layers in these materials and that they are in fact mixed phases [145, 146].

Applications of Layered Double Hydroxides

205

Recently, Zhang et al. [147] investigated the removal of B(OH)4 – , SeO4 2– , CrO4 2– and MoO4 2– from wastewater by incorporation into hydrocalumite ([Ca4 Al2 (OH)12 ](OH)2·6H2 O) and ettringite ([Ca6 Al2 (OH)12 ](SO4 )3·26H2 O). They found that anion uptake by hydrocalumite was larger than that by ettringite, and that the former was capable of reducing anion concentrations to below the standards set for drinking water. Ettringite showed an anion preference in the order of B(OH)4 – > SeO4 2– > CrO4 2– > MoO4 2– . In contrast, borate showed the lowest affinity for hydrocalumite. In addition to anionic species, some metal cations can also be removed via adsorption processes with LDH materials. Recently, Lazaridis reported an interesting removal of two anions (PO4 3– , SCN– ) and three cations (Cd2+ , Pb2+ , Ni2+ ) from aqueous solutions in single batch systems using uncalcined and calcined (773 K) Mg/Al LDH carbonate materials [148]. It was found that the calcined material showed higher sorption capacities than the uncalcined material for all the ions. Since the sorption capacities are relatively high, the author suggested that LDHs could be considered as a potential materials for sorption of both anions and cations in wastewater treatment systems. Seida et al. have also reported the rapid removal of dilute Pb2+ from dilute aqueous solutions by a column packed with a pyroaurite-like Mg/Fe – CO3 LDH compound over a wide range of space velocity (Sv = 150–800 min–1 ) [149]. Both uncalcined and calcined LDHs have also been used as sorbents for decontamination of radioactive wastewater [150, 151]. For example, Toraishi et al. [151] reported the adsorption behavior of IO3 – in radioactive wastewater by LDHs with interlayer CO3 2– or NO3 – anions. It was found that the adsorption equilibrium was reached more quickly for LDH – CO3 than for LDH – NO3 . Moreover, the adsorption isotherm for LDH – CO3 consisted of two Langmuir type adsorption processes, while the desorption behavior was quite different for LDH – CO3 and LDH – NO3 , suggesting that IO3 – is adsorbed on the external surface of LDH – CO3 whereas IO3 – is exchanged for interlayer NO3 – . 3.3 Adsorption of Organic Species on Uncalcined and Calcined LDHs LDHs are also promising materials as sorbents for anionic organic contaminants via both ion-exchange and reconstruction reactions. There have been a large number of reports of the use of LDHs for removal of species such as aromatic carboxylic acids, phenols, pesticides, and humic or fulvic acids. Recently, Cardoso et al. [152] found that the sorption process of terephthalate anions from aqueous solutions by calcined Mg/Al – CO3 LDHs takes place by reconstruction of the LDHs and involves the intercalation and adsorption of terephthalate anions. Calcined Mg/Al – CO3 LDHs were found to be capable of removing 40 to 85 % of the benzoate from solutions in the concentration

206

F. Li ·X. Duan

range usually found in industrial wastewater [153]. It was suggested that the sorption process could be attributed to the fast regeneration of the LDH with intercalated hydroxyl anions from the calcined species, followed by a slow process of anion exchange by benzoate. Phenols, particularly the highly chloro-or nitro-substituted variety, are an important group of organic contaminants which, at typical ambient pH, can be present in groundwater predominantly as phenolate anions. Ulibarri et al. [154] studied the adsorption capacity of 2,4,6-trinitrophenol (TNP) on Mg/Al LDHs and their calcined products. The adsorption of TNP on LDHs by anionic exchange is dramatically affected by the identity of the interlayer anion and LDH chlorides have an adsorption capacity of more than 4 times that of LDH carbonates. However, calcined LDH carbonates are more effective adsorbents than those derived by calcination of LDH chloride samples. This possibly reflects the higher surface area of the former species. Anionic surfactants are present in surface water, resulting in serious environmental pollution. Therefore, adsorption of surfactants, such as sodium dodecylsulfate [155, 156], on Mg/Al LDHs has received considerable attention. Ulibarri et al. also published the results of sorption of an anionic surfactant (sodium dodecylbenzenesulfonate) from water by LDHs and calcined samples (773 K), focusing both on their potential application as a sorbent and on the possibility of their recycling [154, 157]. They found that anionic exchange was complete when the interlayer anion in the LDH precursor was Cl– , reaching 100 % of AEC, and calcined LDH-carbonates were better adsorbents than those derived from LDH-chloride samples, however. It was also claimed that an increase in the crystallinity of the LDH samples probably leads to better ordered calcined mixed oxides, facilitating reconstruction of the layers and enlarging the absorption capacity. The presence of color in many industrial effluent streams is highly undesirable. LDHs have been found to be particularly effective at removing various synthetic dyes (Table 1) [158]. For example, Acid Blue 29 could be adsorbed on the surface or enter the interlayer region of the LDH by anion exchange; an equilibrium time of 1 h with 99 % dye removal was obtained. Furthermore,

Table 1 Adsorption capability of LDHs for various synthetic dyes Dye solution

Removal (%)

Dye charge

Formula weight

Acid Blue 29 Eosin B Reactive Blue 48 Disperse Red 1 Basic Blue 66 Basic Blue 9

96 96 8 86 54 44

Anionic Anionic Anionic Nonionic Cationic Cationic

616.50 624.08 840.12 314.35 530.16 373.90

Applications of Layered Double Hydroxides

207

the spent sorbents can be readily regenerated by heating at 723 K, which results in removal of all adsorbed organics. As expected, the reused sorbents displayed greater adsorption capabilities than the newly prepared LDH as a consequence of the “memory effect”. Although anionic dyes were generally removed most effectively, as expected, nonionic and even cationic dyes can also be removed to reasonable extents. Pesticides, one of the main groups of organic contaminants in water, are found in increasing amounts because of their widespread use in agriculture. Adsorption of pesticides, such as MCPA (4-chloro-2-methylphenoxyacetic acid) [159], imazamox (2-[4,5-dihydro-4-methyl-(1-methylethyl)-5-oxo-1Himidazol-2-yl]-5-(methoxymethyl)-3-pyridinecarboxylic acid) [160], glyphosate (N-phosphonomethyl glycine) [161, 162] and dicamba (3,6-dichloro-2methoxy benzoic acid) [163], on LDHs and calcined LDHs has been widely studied. In addition, iron-containing LDHs and their calcined samples have shown a high adsorption capacity for humic substances dissolved in water, as reported by Seida et al. [164]. The authors demonstrated that the removal of humic substances occurs both by intercalation into the interlayer galleries and adsorption by hydroxyl groups of the materials. Moreover, the buffering pH function of the LDHs enhances the removal by producing hydroxide ions through their slight dissolution. These hydroxide ions coagulate the humic substances with the sorbents, resulting in a sludge with low water content. 3.4 Adsorption of Organic and Inorganic Species on Organo-LDHs The anionic exchange capability of LDHs can be utilized to good advantage by intercalating large organic anions, such as surfactants, in the interlayers. The change of LDH surface properties from hydrophilic to hydrophobic and accessibility of the interlayer region of the resulting so-called organo-LDHs result in enhanced sorption capacity for a diverse range of organic pollutants and even non-ionic organic materials. A few reports [160, 161, 165–175] have been published on the use of organic-modified LDHs for removing organic molecules from water and the results indicate the considerable potential of these materials for that purpose. For example, Kopka et al. [173] found that Zn/Cr LDHs containing surfactant anions have the ability to take up a large variety of small organic molecules such as ethylene glycol, propanediol and glycerol into their interlayers. Dutta et al. [174] also found that sorption of large guest molecules, e.g. pyrene, by an organic acid-exchanged Li/Al LDH could be regulated by varying the chain length of the carboxylic acids or type of anion in the LDH. In addition, the potential application of one such species, a myristate-intercalated Li/Al LDH, as the stationary phase in gas chromatography has been demonstrated, based on the partitioning effect exhibited by this material [175]. The selectivity of a hydrophobic LDH toward organic liquid mixtures has also been

208

F. Li ·X. Duan

reported [167], and this material was suggested for use as a stationary phase in gas chromatography or as self-assembled films [170, 175] in membrane and separation design. You et al. [171] recently investigated dodecylbenzenesulfonate-exchanged LDHs as sorbent for trichloroethylene and tetrachloroethylene; no significant difference in sorption capacity was found between products derived from Mg/Al LDHs with different charge density. In addition, Zhao et al. have also studied sorption of trichloroethylene and tetrachloroethylene by dodecyl sulfate (DDS)-intercalated Mg/Al LDHs [172]. The results showed that DDS-intercalated Mg/Al LDH (Mg/Al= 3) had the highest affinity for both trichloroethylene and tetrachloroethylene in water while DDS-intercalated Mg/Al LDH (Mg/Al= 2) had the lowest sorption affinity, although containing the highest amount of DDS. This is related to the fact that Mg/Al LDH (Mg/Al= 3) has a charge equivalent area of 32.2 ˚ A2 /charge, which allows the formation of the optimal DDS configuration for uptake of additional guest species within its interlayers, thus resulting in the highest affinity for the chlorinated compounds. In the case of the Mg/Al LDH (Mg/Al= 2), the higher packing density of the surfactant molecules restricts uptake of additional guest species. In addition, the adsorption properties of organo-LDHs may also be utilized in the preparation of new materials. The intercalation of C60 molecules into the hydrophobic interlayer of an Mg/Al LDH with intercalated dodecyl sulfate anions has recently been reported [176]. It was found that heating the resulting compound under vacuum in order to decompose the dodecyl sulfate left C60 molecules sandwiched between the hydroxide layers, although the crystallinity of the sample was poor.

4 Applications in Pharmaceutics 4.1 Applications in Medicine The earliest medical applications of LDHs were mainly as antacid and antipepsin agents [177, 178], and an increased demand in this area is expected in the future. Furthermore, they have also been suggested for the removal of phosphate anions from the gastrointestinal fluid with the aim of preventing hyperphosphatemia [179]. More recently, LDHs have found other significant applications in medicine, especially in pharmaceutical formulations. Recent studies have focused on the intercalation and controlled release of pharmaceutically active compounds from LDH materials, taking advantage of their biocompatibility, variable chemical composition, ability to intercalate anionic drugs, and their alkaline character.

Applications of Layered Double Hydroxides

209

Current trends in pharmaceutical technology require formulations to be able to maintain pharmacologically active drug levels for long periods, avoiding repeated administration, and/or to localize the drug release at its pharmaceutical target [180]. The interlayer region of LDHs may be considered a microvessel in which the drug is stored in an ordered way, while maintaining its integrity, and protected from the action of light and oxygen. After administration of the intercalation compound, the drug may be released via a deintercalation process, occurring because of ion exchange or displacement reactions. Such a drug delivery system could reduce the side effects of drugs associated with high plasma levels and prolong active drug life at the same time. The rate of drug diffusion out of the intercalation compound is controlled by the strength of the host-guest interaction, the rigidity of the layers and the diffusion path length. Mg/Al LDHs have already been used, but only as an excipient, in sustained-release formulations containing nifedipine, an antihypertensive drug [181]. A variety of anionic drug molecules, such as indomethacin [182], the anti-inflammatory drug fenbufen [183] and 1-hydroxyethylidene-1,1diphosphonic acid (HEDP) [184], have also been intercalated into Mg/Al LDHs. The aim of these studies is to determine the feasibility of using these intercalation compounds as materials for the storage, transport and, ultimately, modified and controlled release of the drug. For example, O’Hare et al. intercalated a series of pharmaceutically active compounds including diclofenac, gemfibrozil, ibuprofen, naproxen, 2-propylpentanoic acid, 4-biphenylacetic acid and tolfenamic acid into LDHs [185]. The results show that the intercalation of pharmaceutically active compounds that form stable anions is a feasible approach for the storage and subsequent controlled release of bioactive agents (Table 2) as discussed in Chapter 4. Ambrogi et al. [186] intercalated the drug ibuprofen (IBU) in the interlayer region of an Mg/Al – Cl LDH by an anion-exchange method in order

Table 2 Release profile data

Guest molecules

Release times/min a pH 4 pH 7 b b t50 t90 t50 b

t90 b

Diclofenac (λmax = 279 nm) Naproxen (λmax = 317.330 nm) Gemfibrozil (λmax = 275 nm) Tolfenamic acid (λmax = 290 nm) 4-Biphenylacetic acid (λmax = 290 nm)

1 1 1 1 1

28 17 5 23 35

a b

4 9 5 21 2

Release (%) = (I0 /Imax ) × 100, at λmax t50 and t90 are the times for 50 and 90 % release, respectively

1 1 1 1 1

210

F. Li ·X. Duan

to prepare a modified release formulation. The result of dissolution tests for LDH-IBU at pH 7.5 were compared with those obtained using a physical mixture of LDH – Cl and IBU as well as the commercial formulation NeoMindol (containing ibuprofen sodium salt), in simulated intestinal fluid. It was found that the latter two samples gave an immediate release of total drug content, whilst LDH-IBU released 60 % of the drug after 20 min and 100 % after 100 min. They suggested that the mechanism of modified drug release was based on the ion-exchange process of the ibuprofen anion intercalated in the lamellar host and phosphates contained in the buffer solution used as model intestinal fluid. Recently, Choy et al. also reported that LDHs are an efficient drug reservoir for folate derivatives [187]. Folic acid derivatives, folinic acid and methotrexate (MTX), have been successfully hybridized with Mg/Al LDHs by ionexchange reactions. Cellular uptake tests with the MTX-LDH hybrids were carried out in the fibroblast (human tendon) and osteosarcoma (SaOS-2) cell lines by in vitro assay. They found that the LDH not only plays a role as a biocompatible delivery matrix for drugs but also facilitates a significant increase in the delivery efficiency. However, because of their basic character, unmodified LDHs are unsuitable as an oral drug delivery system because they will be destroyed in the stomach where the pH is 1.2. Therefore, Duan et al. [188] prepared a core-shell material as a drug delivery system. An LDH intercalated with a non-steroidal anti-inflammatory drug, fenbufen, as the core was coated with enteric polymers, Eudragit® S 100 or Eudragit® L100, as a shell, giving a composite material which shows effective controlled release of the drug under in vitro conditions which model the passage of a material through the gastrointestinal tract. 4.2 Applications in Biochemistry Incorporation of biologically important molecules into LDHs has become of interest in recent years and materials such as DNA [189], ATP [190], amino acids or enzymes [191, 192] and vitamins [193] can be stabilized in the interlayer space of LDHs. If enzymes and proteins, for example, can be immobilized in the interlayer galleries of LDHs, new types of selective catalysts (see Sect. 2.2) as well as new delivery systems and carrier materials can be expected. Choy et al. [193] reported that vitamin A (retinoic acid), vitamin C (ascorbic acid) and vitamin E (tocopherol) could be intercalated into Zn/Al LDHs by the coprecipitation method. In solutions, these vitamins are normally all sensitive to light, heat and oxygen, and it was proposed that incorporating the molecules into a layered inorganic lattice may lead to their stabilization, resulting in a wider range of potential applications.

Applications of Layered Double Hydroxides

211

Choy et al. have also intercalated biological macromolecules such as DNA, ATP and nucleosides into Mg/Al – NO3 LDHs [189, 190, 194, 195], where the host lattice may protect relatively delicate biomolecules from degradation and also aid their transport to specific targets within the body, and hence the intercalation reactions lead to the formation of novel bioinorganic nanohybrids with potential practical significance, such as new DNA reservoirs or carriers for the delivery of genetic material to cells [189]. Nano-sized LDHs have also been shown to be effective as delivery carriers for drugs and genes by hybridizing with DNA and c-antisense oligonucleotide (As-myc) [196]. A strong suppression of cell growth (65 %) is observed when HL-60 leukemia cells are incubated with 20 µM As-myc-LDH hybrid. The LDH itself was found to be non-cytotoxic against HL-60 cells. Consequently, LDHs can act as a new type of inorganic carrier that is completely different from existing nonviral vectors in terms of its chemical bonding and structure. Recently, Choy et al. reported that bioinorganic DNA-LDH nanohybrids could be used as a genetic molecular coding system [197]. Such a molecular-level coding method utilizing DNA base pairs as code units has been systemized. The system consists of four steps: encoding, encrypting, decrypting and decoding. The nanohybrids were found to give solutions to the inherent problems that hamper DNA molecular coding systems. In addition, LDHs are able to concentrate selectively and organize organic molecules [198–200] For example, Pitsch et al. reported that glycolaldehyde phosphate intercalated in LDHs from highly dilute aqueous solution can condense to racemic aldotetrose-2,4-diphosphates and aldolhexose-2,4,6triphosphates [200], which has led to speculation about the possible role of LDHs in chemical evolution and the origins of life.

5 Applications in Photochemistry The interlayer region of LDHs can provide a novel environment for photochemical reactions of guest molecules [201–204]. For example, Takagi et al. reported that the controlled photodimerization of a variety of unsaturated carboxylate species intercalated in the interlayer galleries of Mg/Al LDH could occur between the layers [203]. Syn head-to-head cyclodimers were selectively formed in the irradiation of intercalated cinnamate ions, whereas two isomers of syn head-to-head and syn head-to-tail cyclodimers were formed for the case of phenylethenylbenzoates. The product selectivity was shown to be controlled by the Mg/Al ratio of the host LDH, and hence the packing density of the anions in the interlayer region [204]. Tagaya et al. [205, 206] studied the photoisomerization of photochromic molecules sulfonated indolinespirobenzopyran (SP – SO3 – ) to merocyanine (MC) in the interlayer region of an Mg/Al LDHs. They found that the photo/-

212

F. Li ·X. Duan

Fig. 2 Possible mechanism for the reversible photoisomerization between (SP – SO3 – ) and merocyanine (MC) in the interlayer of an Mg/Al LDH

isomerization was irreversible for an Mg/Al LDH containing SP – SO3 – due to the stability of MC in the interlayer polar environment. However, in the presence of p-toluenesulfonate (PTS), reversible photoisomerization was observed (Fig. 2). Organic dyes are extensively used in the textile industry and other industrial applications, but suffer from limited UV and/or oxygen and thermal stability. Therefore, ways of improving their properties attract a great deal of interest. There has been considerable work on intercalation of cationic dyes in zeolites [207], aluminosilicate clays [201] and metal(IV) phosphonates [208]. This should have the effect of stabilizing the dye against UV, oxidative or thermal degradation and give a superior type of pigment. Such materials also have potential applications as nonlinear optical materials provided that they have sufficient thermal and optical stability under relatively strong laser irradiation [209]. Several examples of the intercalation of organic dyes in LDHs have been reported in the literature [210–214]. The dyes may be slowly leached from the host, however [210]. For example, the incorporation of two anionic dyes, indigo carmine and new coccine, into Mg/Al LDHs may yield materials with potential applications as pigments [213]. Miyata [128] determined the Cl– /Naphthol Yellow S exchange isotherms for an Mg/Al LDH and found that the dye was exchanged with high selectivity. Recently, a large anionic pigment has been intercalated into an LDH host by ion-exchange of an Mg/Al LDH nitrate precursor with a solution of C.I. Pigment Red 48 : 2 (the calcium salt of 4-((5-chloro-4-methyl2-sulfophenyl)azo)-3-hydroxy-2-naphthalene-carboxylic acid), in ethane-1,2diol [215]. The UV-visible diffuse reflectance spectra of C.I. Pigment Red

Applications of Layered Double Hydroxides

213

48 : 2 itself show marked changes after heating at 200 ◦ C and above, whereas there are no significant changes in the spectra of the intercalated pigment after heating at temperatures up to 300 ◦ C, indicating that the thermostability is markedly enhanced by intercalation in the LDH host. Furthermore, the pigment-intercalated LDHs exhibits much higher photostability to UV light than the pristine pigment, in the case of both the pure solids and their composites with polypropylene. Cooper and Dutta [216] found that Li/Al LDHs intercalated with 4-nitrohippuric acid could exhibit second harmonic generation, which is a frequencydoubling nonlinear optical process. This is due to a perpendicular monolayer packing of the guest in the interlayer, resulting in an ordered arrangement of dipoles and hence bulk dipole moment in the solid. LDH materials may be used as infrared absorbing materials in agricultural plastic films, affording higher night-time temperatures in greenhouses. Xu et al. prepared polyethylene (PE) films with Mg/Al LDHs [217] which have significantly greater absorption of blackbody infrared radiation than a talccontaining PE film; other properties of the film are unaffected by replacing talc with LDH. El-Toni et al. reported that a material prepared by coating a Zn/Al LDH intercalated with 4,4 -diaminostilbene-2,2 -disulfonic acid nanocomposite with silica has effective UV barrier properties and is suitable for use in sunscreen formulations [218]. The problem of deintercalation of organic molecules from LDHs by the anion-exchange reaction with carbonate ion was greatly reduced by forming a protective film of silica on the surface. Duan et al. prepared a salicylate-pillared LDH by ion-exchange of a Zn/Al LDH precursor in an organic solvent [219]. They found that the salicylate-pillared LDH had good thermal stability and could block UV radiation over a wider range than that afforded by a mixture of the two components.

6 Applications in Electrochemistry Organic polymers are widely used in modified electrodes [220], but inorganic materials such as zeolites, clays or microporous solids are attractive as replacements since they have much better stability, tolerance to high temperatures and oxidizing conditions, and chemical inertness. Due to the capability of clays to exchange intercalated ions, clay modified electrodes have been extensively studied. Layered nickel hydroxide can be used as an electrode for alkaline secondary cells. To improve its properties, modification has been carried out by incorporation of other metal elements to form Ni/M LDHs, including Co [221], Zn [222], Al [223], Cr or Mn [224] and Fe [225]. For example, Chen et al. reported the electrochemical performance of Al-substituted layered α-

214

F. Li ·X. Duan

Ni(OH)2 with appropriate additives in Ni-metal hydride batteries [223]. They found that the addition of Mg2+ increases the unit cell parameters and the charge and discharge potentials, and decreases the electrochemical polarization. A higher charge efficiency and cycle life is obtained by the addition of Co2+ . The authors suggested that the addition of both Mg2+ and Co2+ could improve the stability in alkaline solution at high temperature. In addition, the potential application of organo-LDHs as new modified electrodes has been investigated. For example, Mousty et al. [226, 227] prepared new electroactive materials derived from LDHs containing electroactive organic molecules, such as anthraquinonesulfonates, m-nitrobenzenesulfonate, and 2,2 -azinobis-3-ethylbenzothiazoline-6-sulfonate. Recently, Liao and Ye [228] have synthesized poly(ethylene oxide)/LDH (PEO/LDH) nanocomposite polymer electrolytes, which were formed by addition of an oligo(ethylene oxide) modified LDH (OLDH) as a nanoscale nucleating agent to PEO nanocomposite electrolytes. The authors found that exfoliated nanocomposites were formed due to the compatibility of the surface modified LDH layers with the PEO matrix, and that the nanoscale dispersed OLDH layers inhibit the growth of PEO crystallites and result in an abundance of intercrystalline grain boundaries within the PEO/LDH nanocomposites. The ionic conductivities of the nanocomposite electrolytes are increased by 3 orders of magnitude compared with the pure PEO polymer electrolytes at ambient temperature. Such novel nanocomposite electrolyte systems with high conductivities should find application in the fabrication of thin-film type Li-polymer secondary batteries.

7 Applications as Additives in Functional Polymer Materials Chlorine-containing polymers such as poly(vinyl chloride) PVC undergo an autocatalytic dehydrochlorination reaction under the influence of elevated temperature and UV radiation. Since the HCl originating from the dehydrochlorination of the PVC chains is believed to sustain this autocatalytic process, stabilizers that irreversibly bond HCl can thus inhibit the degradation. Heavy metal compounds such as cadmium stearate or lead stearate are currently used for this purpose. However, alternatives are required due to environmental problems associated with the use of heavy metals. Indeed, the largest current application of LDH materials is in the polymer industry, mainly to stabilize PVC [3, 229–232]. Kyowa Chemical Industries (Japan) were the first to demonstrate that adding Mg/Al LDHs to PVC in combination with other additives such as zinc stearate and tin maleate leads to an enhancement in thermal stability of the resin [231]. The role of the LDH in absorbing HCl was confirmed experimentally by Van der Ven et al. [232], who measured the capacity of LDHs

Applications of Layered Double Hydroxides

215

having the same MII /MIII ratio and different counter-ions to react with HCl gas and found a linear correlation between increasing HCl capacity and thermal stability of the LDH in the order SO4 2– < Cl– < OH– ∼ NO3 – < CO3 2– < C17 H35 COO– . They suggested that the reaction between the LDHs and the HCl occurs in a two-step process: firstly (if possible), the interlayer anions react with the HCl gas, and, secondly, the LDH layers themselves react with the HCl giving complete destruction of the LDH structure and formation of metal chlorides. Recently, Duan et al. [233] found that Mg/Al LDHs with Mg/Al = 2 has the optimum stabilizing effect on PVC because of its higher layer charge density and consequent stronger driving force for uptake of Cl– into the interlayer galleries. However, this LDH contains the highest quantity of interlayer CO3 2– ions but the lowest overall HCl absorption capacity, suggesting that it is the reaction of HCl and CO3 2– ions that is most important in stabilizing the PVC against thermal degradation. Flame retardant materials may be formulated so as to be more resistant towards ignition or to have slower rates of flame spread in a major fire. LDHs have also found useful application as flame retardants in PVC and other polymers as well leading to reduced quantities of smoke during combustion [234]. Braglia et al. compared the flame retardant effect of LDH in ethylene vinyl acetate (EVA) copolymer with that of inorganic hydroxides, such as aluminum hydroxide and magnesium hydroxide [235]. The most significant flame retardant effects, observed using a mass loss calorimeter, indicated that the EVA polymer filled with 50 wt.% of LDHs has the slowest heat release rate and the lowest evolved gas temperature. Duan et al. have recently shown [236] that reaction of an Mg/Al – CO3 LDH with boric acid affords a polyborate-pillared LDH. The flame retardant properties of this material and the carbonate precursor in composites with ethylene vinyl acetate copolymer were compared. Introduction of the borate anion leads to a significant enhancement in smoke suppression during combustion without compromising the flammability of the material. This is related to the synergistic effect between the host layers of the LDH and the borate anions uniformly distributed in the interlayer region.

8 Other Applications Today, magnetic information carriers are widespread, and no other technology can compete in storage density and rate of access to stored data. Further improvement in performance requires the development of new magnetic materials with high saturation magnetic flux density, appropriate coercivity and suitable magnetic characteristics.

216

F. Li ·X. Duan

It is well known that spinels can be obtained after calcination of LDHs. In general pure spinels cannot be prepared, however, because whilst the MII /FeIII ratio in LDHs is typically in the range 2–4 (see Chapter 1), the ratio in spinels is 0.5. Li et al. synthesized pure MFe2 O4 (M = Mg, Co, and Ni) spinel ferrites from tailored LDH precursors [237, 238]. MII -FeII -FeIII layered double hydroxides were first prepared and single phase ferrites were obtained after calcination (1173 K) since the FeII is oxidized to FeIII under these conditions. The saturation magnetization of the as-prepared ferrites is significantly higher than that of ferrites made by conventional methods. This is because the metal elements in the layered precursor are uniformly dispersed at the atomic level. An alternative way of producing LDH precursors with MII /FeIII ratios appropriate for formation of MFe2 O4 spinel ferrites involves intercalation of iron-containing guests such as [Fe(CN)6 ]3– has been reported by Duan et al. [239, 240]. On calcination, the iron from the interlayer anions is incorporated in the spinel phase along with the iron originally in the layers and is not converted to a separate iron oxide phase. Wong and Buchheit [241] utilized the structural memory effect of Li/Al LDHs in order to embed it in a water-permeable epoxy matrix for sensing water uptake in organic coatings, leading to a novel approach for remotely and non-destructively detecting water uptake in optically opaque organic coatings. This work suggests that addition of calcined LDHs to organic coatings may lead to methods for sensing early-stage coating degradation due to water uptake and may give advance warning of substrate corrosion. Mousty et al. [242] studied a novel inexpensive and simple amperometric biosensor, based on the immobilization of horseradish peroxidase (HRP) in redox active Zn/Cr-ABTS LDH [ABTS = 2,2 -azinobis(3-ethylbenzothiazoline6-sulfonate)], which was applied in the determination of cyanide. The electrochemical transduction step corresponds to the reduction at 0.0 V of ABTS (+) enzymatically formed in the presence of H2 O2 . The biosensor has a fast response to H2 O2 (8 s) with a linear range of 1.7 × 10–9 to 2.1 × 10–6 M and a sensitivity of 875 mA M–1 cm–2 . The detection of cyanide is performed via its non-competitive inhibiting action on the HRP/[ZnCr-ABTS LDH] electrode. Sustainable development of concrete infrastructure continues to be of importance to the construction industry. The use of supplementary cementitious materials in concrete is an integral component of these strategies. Raki et al. [243] demonstrated the promise of LDH-like materials as suitable hosts for intercalation of organic admixtures with the long-term view of controlling their release rate in concrete by blending the inorganic-organic nanocomposites (in small amounts) with the cement. In their study, nitrobenzoic acid (NBA), naphthalene-2,6-disulfonic acid (26NS), and naphthalene-2 sulfonic acid (2NS) salts, commonly used as admixtures in the making of concrete [244], were intercalated by anion-exchange of nitrate in the host material, [Ca2 Al(OH)6 ]NO3 · nH2 O. It was suggested that potential future

Applications of Layered Double Hydroxides

217

applications of these composite materials could be to control the effect of admixtures on the kinetics of cement hydration by programming their temporal release.

9 Concluding Remarks LDHs represent one of the most technologically promising materials as a consequence of their low cost, relative ease of preparation, and the large number of composition/preparation variables that may be adopted. At present, even though a great deal of work of academic and commercial interest on LDH materials has been carried out, still more remains to be done in order to exploit completely their potential applications. In the future, we believe work on applications of these layered compounds will continue to expand rapidly.

References 1. Cavani F, Trifirò F, Vaccari A (1991) Catal Today 11:173 2. De Roy A, Forano C, Malki KE (1992) In: Occelli ML, Robson HE (eds) Expanded Clays and Other Microporous Solids. Chapter 7, Synthesis of Macroporous Microporous Solids, vol II. Van Nostrand Reinhold, New York p 108 3. Trifirò F, Vaccari A (1996) In: Atwood JL, Davies JED, MacNicol DD, Vögtle F (eds) Comprehensive Supramolecular Chemistry, vol 7. Pergamon, Oxford p 251 4. Rives V, Ulibarri MA (1999) Coord Chem Rev 181:61 5. Kagunya W, Hassan Z, Jones W (1996) Inorg Chem 35:5970 6. Carlino S (1997) Chem Brit 33:59 7. Bröcker FJ, Marosi L (1970) German Pat. 2,024,282 to BASF AG 8. Müller-Tamm H, Frielingsdorf H, Schweier G, Reuter L (1971) German Pat. 2,163,851 to BASF AG 9. Bhattacharyya AA, Woltermann GM, Yoo JS, Karch A, Cormier WE (1988) Ind Eng Chem Res 27:1356 10. Yoo JS, Bhattacharyya AA, Radlowski CA (1991) Ind Eng Chem Res 30:1444 11. Pasel J, Käßner P, Montanari B, Gazzano M, Vaccari A, Makowski W, Lojewski T, Dziembaj R, Papp H (1998) Appl Catal B 18:199 12. Blasco T, López Nieto JM, Dejoz A, Váquea MI (1995) J Catal 157:271 13. López Nieto JM, Dejoz A, Váquea MI (1995) Appl Catal A 132:41 14. Dinka P, Prandová K, Hronec M (1998) Appl Clay Sci 13:467 15. Ren L, He J, Zhang S, Evans DG, Duan X (2002) J Mol Catal B: Enzym 18:11 16. Ren L, He J, Evans DG, Duan X, Ma R (2001) J Mol Catal B: Enzym 16:65 17. Rahman MBA, Basri M, Hussein MZ, Rahman RNZA, Zainol DH, Salleh AB (2004) Appl Biochem Biotech 118:313 18. Chen YZ, Hwang CM, Liaw CW (1998) Appl Catal A 169:207 19. Davis RJ, Derouane EG (1991) Nature 349:313 20. Narayanan S, Krishna K (1998) Appl Catal A 174:221 21. Narayanan S, Krishna K (1999) Catal Today 49:57

218

F. Li ·X. Duan

22. Bennur TH, Ramani A, Bal R, Chanda BM, Sivasanker S (2002) Catal Comm 3:493 23. Choudary BM, Kantam ML, Reddy NM, Gupta NM (2002) Catal Lett 82:79 24. Bussi J, Lopez A, Pena F, Timbal P, Paz D, Lorenzo D, Dellacasa E (2003) Appl Catal A 253:177 25. Das N, Tichit D, Durand R, Graffin P, Coq B (2001) Catal Lett 71:181 26. Albertazzi S, Busca G, Finocchio E, Glöckler R, Vaccari A (2004) J Catal 223:372 27. Choudary BM, Madhi S, Chowdari NS, Kantam ML, Sreedhar B (2002) J Am Chem Soc 124:14127 28. Motokura K, Nishimura D, Mori K, Mizugaki T, Ebitani K, Kaneda K (2004) J Am Chem Soc 126:5662 29. Kirm I, Medina F, Rodriguez X, Cesteros Y, Salagre P, Sueiras J (2004) Appl Catal A 272:175 30. Costantino U, Curini M, Montanari F, Nocchetti M, Rosati O (2003) J Mol Catal A: Chem 195:245 31. Choudary BM, Kantam ML, Neeraja V, Rao KK, Figueras F, Delmotte L (2001) Green Chem 3:257 32. Rives V, Prieto O, Dubey A, Kannan S (2003) J Catal 220:161 33. Kapoor MP, Matsumura Y (2004) Catal Today 93-95:287 34. Choudary BM, Kantam ML, Reddy CRV, Rao KK, Figueras F (1999) J Mol Catal A: Chem 146:279 35. Climent MJ, Corma A, Iborra S, Velty A (2004) J Catal 221:474 36. Tanabe K, Holderich WF (1999) Appl Catal A 181:399 37. Reichle WT (1985) J Catal 94:547 38. Dumitriu E, Hulea V, Chelaru C, Catrinescu C, Tichit D, Durand R (1999) Appl Catal A 178:145 39. Reichle WT (1980) J Catal 63:295 40. Kelkar CP, Schuta AA (1998) Appl Clay Sci 13:417 41. Di Cosimo JL, Diez VK, Apesteguia CR (1998) Appl Clay Sci 13:433 42. Perez CN, Perez CA, Henriques CA, Monteiro JLF (2004) Appl Catal A 272:229 43. Climent MJ, Corma A, Iborra S, Primo J (1995) J Catal 151:60 44. Climent MJ, Corma A, Guil-Lopez R, Iborra S, Primo J (1999) Catal Lett 59:33 45. Corma A, Fornes V, Martin-Aranda RM, Rey F (1992) J Catal 134:58 46. Kumbhar PS, Sanchez-Valente J, Lopez J, Figueras F (1998) Chem Comm 535 47. Aramendia MA, Borau V, Jimenez C, Marinas JM, Ruiz JR, Urbano FJ (2002) J Chem Soc Perkin Trans II 1122 48. Aramendia MA, Borau V, Jimenez C, Marinas JM, Ruiz JR, Urbano FJ (2003) Appl Catal A 249:1 49. Bulbule VJ, Feshpande VH, Velu S, Sudalai A, Sivasankar S, Sathe VT (1999) Tetrahedron 55:9325 50. Fishel CT, Davis RJ (1994) Langumir 10:159 51. Fishel CT, Davis RJ (1994) Catal Lett 25:87 52. Laycock DE, Collacott RJ, Skelton DA, Tchir MF (1991) J Catal 130:354 53. Di Cosimo JL, Apesteguía CR, Ginés MJL, Iglesia E (2000) J Catal 190:261 54. Kumbhar PS, Sanchez-Valente J, Figueras F (1998) Chem Comm 1091 55. Li F, Zhang L, Evans DG, Duan X (2004) Colloids Surf A 244:169 56. Carja G, Nakamura R, Niiyama H (2002) Appl Catal A 236:912 57. Chen AM, Xu HL, Yue YH, Hua WM, Wei S, Zi G (2004) Appl Catal A 274:101 58. Zhang L, Li F, Evans DG, Duan X (2004) Mater Chem Phys 87:402 59. Kumbhar PS, Sanchez-Valente J, Figueras F (1998) Tetrahedron Lett 39:2573 60. Velu S, Swamy CS (1997) Appl Catal A 162:81

Applications of Layered Double Hydroxides

219

61. Velu S, Swamy CS (1994) Appl Catal A 119:241 62. Basile F, Basini L, Damore M, Fornasari G, Guarinoni A, Matteuzzi D, Del Piero G, Trifirò F, Vaccari A (1998) J Catal 173:247 63. Dung NT, Tichit D, Chiche BH, Coq B (1998) Appl Catal A 169:179 64. Cabello FM, Tichit D, Coq B, Vaccari A, Dung NT (1997) J Catal 167:142 65. Holgado MJ, Rives V, San Roman MS (2001) Appl Catal A 214:219 66. Tichit D, Coq B, Ribet S, Durand R, Medina F (2000) In: Corma A, Melo FV, Mendioroz S, Fierro JLG (eds) Proc. 12th Int. Cong. Catal. Stud Surf Sci Catal 130:503 67. Choudary BM, Kantam ML, Rahman A, Reddy CRV (2003) J Mol Catal A: Chem 206:145 68. Barrault J, Derouault A, Courtois G, Maissant JM, Dupin JC, Guimon C, Martinez H, Dumitriu E (2004) Appl Catal A 262:43 69. Ye XN, Ma N, Hua WM, Yue YH, Miao CX, Xie ZK, Gao Z (2004) J Mol Catal A: Chem 217:103 70. Dziembaj DR, Kustrowski P, Chmielarz L (2003) Appl Catal A 255:35 71. Holgado MJ, Labajos FM, Sontero MMJ, Rives V (2003) Mater Res Bull 38:1879 72. Li F, Tan Q, Evans DG, Duan X (2005) Catal Lett 99:151 73. Kaneda K, Yamashita T, Matsushita T, Ebitani K (1998) J Org Chem 63:1750 74. Matsushita T, Ebitani K, Kaneda K (1999) Chem Comm 265 75. Yoo JS, Bhattacharyya AA, Rodlowski CA, Karch JA (1991) Ind Eng Chem Res 30:1444 76. Corma A, Palomares AE, Rey F (1994) Appl Catal B 4:29 77. Corma A, Palomares AE, Rey F, Marquez F (1997) J Catal 170:140 78. Palomares AE, Lopez-Nieto JM, Lazaro FJ, Lopez A, Corma A (1999) Appl Catal B20:257 79. Lojewski T, Dziembaj R, Chyzy R, Vaccari A, Montanari B (1997) Pol J Environ Stud 6:71 80. Kannan S, Swamy CS (1999) Catal Today 53:725 81. Kannan S (1998) Appl Clay Sci 13:347 82. Perez-Ramirez J, Kapteijn F, Moulijn JA (1999) Catal Lett 60:133 83. Palomares AE, Prato JG, Rey F, Corma A (2004) J Catal 221:62 84. Bhattacharyya A, Chang WD, Kleefisch MS, Udovich CA (1997) US Pat. 5,653,774 to AMOCO Co 85. Bhattacharyya A, Chang WD, Kleefisch MS, Udovich CA (1998) US Pat. 5,767,040 to AMOCO Co 86. Bhattacharyya A, Chang WD, Kleefisch MS, Udovich CA (1997) US Pat. 5,591,238 to AMOCO Co 87. Takehira K, Shishido T, Wang P, Kosaka T, Takaki K (2004) J Catal 221:43 88. Lee KM, Lee WY (2002) Catal Lett 83:65 89. Morioka H, Shimizu Y, Sukenobu M, Ito K, Tanabe E, Shishido T, Takehira K (2001) Appl Catal A 215:11 90. Basile F, Fornasari G, Gazzano M, Kiennemann A, Vaccari A (2003) J Catal 217:245 91. Basile F, Fornasari G, Rosetti V, Trifirò F, Vaccari A (2004) Catal Today 91–92:293 92. Tsyganok AI, Inaba M, Tsunoda T, Suzuki K, Kakehira T, Hayakawa T (2004) Appl Catal A 275:149 93. Bhattacharyya A, Chang WD, Schumacher DJ (1998) Appl Clay Sci 13:317 94. Basile F, Fornasari G, Poluzzi E, Vaccari A (1998) Appl Clay Sci 13:329 95. Tsyganok AI, Suzuki K, Hamakawa S, Takehira K, Hayakawa T (2001) Catal Lett 77:75 96. Gunter MM, Ressler T, Jentoft RE, Bems B (2001) J Catal 203:133 97. Shen GC, Fujita S, Matsumoto S, Takezawa N (1997) J Mol Catal A: Chem 124:123

220

F. Li ·X. Duan

98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108.

Peppley BA, Amphlett JC, Kearns LM, Mann RF (1999) Appl Catal A 179:21 Velu S, Suzuki K, Osaki T (2000) Catal Lett 69:43 Segal SR, Anderson KB, Carrado KA, Marshall CL (2002) Appl Catal A 231:215 Velu S, Suzuki K, Osaki T (1999) Chem Comm 2341 Velu S, Suzuki K, Kapoor MP, Ohashi F, Osaki T (2001) Appl Catal A 213:47 Velu S, Satoh N, Gopinath CS, Suzuki K (2002) Catal Lett 82:145 Segal SR, Carrado KA, Marshall CL, Anderson KB (2003) Appl Catal A 248:33 Kwon T, Pinnavaia TJ (1992) J Mol Catal A: Chem 74:23 Kwon T, Tsigdinos GA, Pinnavaia TJ (1988) J Am Chem Soc 110:3653 Giannelis EP, Nocera DG, Pinnavaia TJ (1987) Inorg Chem 26:203 Guo YH, Li DF, Hu CW, Wang YH, Wang EB, Zhou YC, Feng SH (2001) Appl Catal B 30:337 Guo YH, Li DF, Hu CW, Wang EB, Zhou YC, Ding H, Feng SH (2002) Micropor Mesopor Mater 56:153 Sels BF, De Vos D, Buntinx M, Pierard F, Kirsch-De Mesmaeker A, Jacobs PA (1999) Nature 400:855 Sels BF, De Vos D, Jacobs PA (2001) J Am Chem Soc 123:8350 Sels BF, De Vos D, Buntinx M, Jacobs PA (2003) J Catal 216:288 Choudary BM, Bharathi B, Reddy CV, Kantam ML, Raghavan KV (2001) Chem Comm 1736 Choudary BM, Someshwar T, Kantam ML, Reddy CV (2004) Catal Comm 5:215 Choudhary VR, Dumbre DK, Narkhede VS, Jana SK (2003) Catal Lett 86:229 Choudhary VR, Indurkar JR, Narkhede VS, Jhan R (2004) J Catal 227:257 Barloy L, Laliier JP, Battioni P, Mansuy D, Piffard Y, Tournoux M, Valim JB, Jones W (1992) New J Chem 16:71 Ukrainczyk L, Chibwe M, Pinnavaia TJ, Boyd SA (1994) J Phys Chem 98:2668 Carrado KA, Forman JE, Botto RE, Winans RE (1993) Chem Mater 5:472 Chibwe M, Pinnavaia TJ (1993) Chem Comm 278 Perez Bernal ME, Ruano Casero R, Pinnavaia TJ (1991) Catal Lett 11:55 Ukrainczyk L, Chibwe M, Pinnavaia TJ, Boyd SA (1995) Environ Sci Technol 29:439 Shannon IJ, Maschmeyer T, Sankar G, Thomas JM, Oldroyd RD, Sheehy M, Madill D, Waller AM, Townsend RP (1997) Catal Lett 44:23 Park IY, Kuroda K, Kato C (1989) Chem Lett 2057 Bhattacharjee S, Dines TJ, Anderson JA (2004) J Catal 225:398 Bhattacharjee S, Anderson JA (2004) Chem Comm 554 Bish DL (1980) Bull Mineral 103:170 Miyata S (1983) Clays Clay Miner 31:305 Dutta PK, Puri M (1989) J Phys Chem 93:376 Chibwe K, Jones W (1989) Chem Comm 926 Ulibarri MA, Pavlovic I, Hermosin MC, Cornejo J (1995) Appl Clay Sci 10:131 Dimotakis ED, Pinnavaia TJ (1990) Inorg Chem 29:2393 Châtelet L, Bottero JV, Yvon J, Bouchelaghem A (1996) Colloids Surf A 111:167 Shin HS, Kim MJ, Nam SY, Moon HC (1996) Water Sci Technol 34:161 Parker LM, Milestone NB, Newman RH (1995) Ind Eng Chem Res 34:1196 Goswamee RL, Sengupta P, Bhattacharyya KG, Dutta DK (1998) Appl Clay Sci 13:21 Kozai N, Ohnuki T, Komarneni S (2002) J Mater Res 17:2993 Seida Y, Nakano Y (2002) Water Res 36:1306 You YW, Zhao HT, Vance GF (2001) Environ Tech 22:1447 Rhee SW, Kang MJ, Kim H, Moon CH (1997) Environ Tech 18:231 Das DP, Das J, Parida K (2003) J Colloid Interface Sci 261:213

109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141.

Applications of Layered Double Hydroxides

221

142. Dousova B, Machovi V, Kolousek D, Kovanda F, Dornicak V (2003) Water Air Soil Pollut 149:251 143. Das J, Das D, Dash GP, Parida KM (2002) J Colloid Interface Sci 251:26 144. Das NN, Konar J, Mohanta MK, Srivastava SC (2004) J Colloid Interface Sci 270:1 145. Intissar M, Jumas JC, Besse JP, Leroux F (2003) Chem Mater 15:4625 146. Intissar M, Holler S, Malherbe F, Besse JP, Leroux F (2004) J Phys Chem Solids 65:453 147. Zhang M, Reardon EJ (2003) Environ Sci Tech 37:2947 148. Lazaridis NK (2003) Water Air Soil Pollut 146:127 149. Seida Y, Akano NY, Nakamura Y (2001) Water Res 35:2341 150. Kang MJ, Rhee SW, Moon H, Neck V, Fanghanel T (1996) Radiochim Acta 75:169 151. Toraishi T, Nagasaki S, Tanaka S (2002) Appl Clay Sci 22:17 152. Crepaldi EL, Tronto J, Cardoso LP, Valim JB (2002) Colloids Surf A 211:103 153. Cardoso LP, Tronto J, Crepaldi EL, Valim JB (2003) Mol Cryst Liq Cryst 390:49 154. Ulibarri MA, Pavlovic I, Barriga C, Hermosín MC, Cornejo J (2001) Appl Clay Sci 18:17 155. Pavan PC, Crepaldi EL, Gomes GD, Valim JB (1999) Coll Surf A 154:399 156. Pavan PC, Gomes GA, Valim JB (1998) Micropor Mesopor Mater 21:659 157. Pavlovic I, Ulibarri MA, Hermosín MC, Cornejo J (1997) Fresenius Envir Bull 6:226 158. Orthman J, Zhu HY, Lu GQ (2003) Separ Purif Technol 31:53 159. Inacio J, Taviot-Gueho C, Forano C, Besse JP (2001) Appl Clay Sci 18:255 160. Celis R, Koskinen WC, Cecchi AM, Bresnahaw GA, Carrizosa MJ, Ulibarri MA, Pavlovic I, Hermosin MC (1999) J Environ Sci Health B34:929 161. Villa MV, Sanchez-Martin MJ, Sanchez-Camazano M (1999) J Environ Sci Health B34:509 162. Sanchez-Martin MJ, Villa MV, Sanchez-Camazano M (1999) Clays Clay Miner 47:777 163. You YW, Zhao HT, Vance GF (2002) Appl Clay Sci 21:217 164. Seida Y, Nakano Y (2000) Water Res 34:1487 165. Zhao HT, Vance GF (1998) J Incl Phenom Mol Recogn Chem 31:305 166. Esumi K, Yamamoto S (1998) Colloids Surf A 137:385 167. Dékány I, Berger F, Imrik K, Lagaly G (1997) Colloid Polym Sci 275:681 168. Zhao HT, Vance GF (1998) Clays Clay Miner 46:712 169. Celis R, Koskinen WC, Hermosín MC, Ulibarri MA, Cornejo J (2000) J Soil Sci Soc Am 64:36 170. Dékány I, Haraszti T (1997) Colloids Surf A 123:391 171. You YW, Zhao HT, Vance GF (2002) J Mater Chem 12:907 172. Zhao H, Nagy KL (2004) J Colloid Interface Sci 274:613 173. Kopka H, Beneke K, Lagaly G (1988) J Colloid Interface Sci 123:427 174. Dutta PK, Robins DS (1994) Langmuir 10:1851 175. Jakupca M, Dutta PK (1995) Chem Mater 7:989 176. Tseng WY, Lin JT, Mou CY, Cheng SF, Liu SB, Chu PP, Liu HW (1996) J Am Chem Soc 118:4411 177. Miyata S (1985) US Patent 5,514,389 178. Jettka W, Gajdos B, Benedikt MD (1996) Eur. Patent EP 715846, to Rhone-Poulenc Rorer Gmbh, Germany 179. Ookubo A, Ooi K, Hayashi H (1992) J Pharm Sci 81:1139 180. Gardner CR (1985) Biomaterials 6:153 181. Doi N, Nitta S, Kusakari M, Takahashi N (1984) Jpn. Kokai Tokkyo Koho JP 60,225,719 [85,225,719] 182. Li B, He J, Evans DG, Duan X (2004) Appl Clay Sci 27:199

222

F. Li ·X. Duan

183. Del Arco M, Cebadera E, Gutierrez S, Martin C, Montero MJ, Rives V, Rocha J, Sevilla MA (2004) J Pharm Sci 93:1649 184. Nakayama H, Takeshita K, Tsuhako M (2003) J Pharm Sci 92:2419 185. Khan AI, Lei L, Norquist AJ, O’Hare D (2001) Chem Comm 2342 186. Ambrogi V, Fardella G, Grandolini G, Perioli L (2001) Int J Pharm 220:23 187. Choy JH, Jung JS, Oh JM, Park M, Jeong J, Kang YK, Han OJ (2004) Biomaterials 25:3059 188. Li B, He J, Evans DG, Duan X (2004) Int J Pharm 287:89 189. Choy JH, Kwak SY, Park JS, Jeong YJ, Portier J (1999) J Am Chem Soc 121:1399 190. Choy JH, Kwak SY, Park JS, Jeong YJ (2001) J Mater Chem 11:1671 191. Aisawa S, Takahashi S, Ogasawara W, Umetsu Y, Narita E (2001) J Solid State Chem 162:52 192. Nakayama H, Wada N, Tsuhako M (2004) Int J Pharm 269:469 193. Hwang SH, Han YS, Choy JH (2001) Bull Korean Chem Soc 22:1019 194. Choy JH, Kwak SY, Park JS, Jeong YJ, Park JS (2000) Angew Chem Int Ed 39:4042 195. Portier J, Campet G, Treuil N, Poquet A, Kim YI, Kwon SJ, Kwak SY, Choy JH (1998) J Korean Chem Soc 42:487 196. Kwak SY, Jeong YJ, Park JS, Choy JH (2002) Solid State Ionics 151:229 197. Choy JH, Oh JM, Park M, Sohn KM, Kim JW (2004) Adv Mater 16:1181 198. Krishnamurthy R, Pitsch S, Arrhenius G (1999) Origins Life Evol Biosphere 29:139 199. DeGraaf RM, Visscher J, Xu Y, Arrhenius G, Schwartz AW (1998) J Mol Evol 47:501 200. Pitsch S, Eschenmoser A, Gedulin B, Hui S, Arrhenius G (1995) Origins Life Evol Biosphere 25:297 201. Ogawa M, Kuroda K (1995) Chem Rev 95:399 202. Valim JB, Kariuki BM, King J, Jones W (1992) Mol Cryst Liq Cryst 211:271 203. Takagi K, Shichi T, Usami H, Sawaki Y (1993) J Am Chem Soc 115:4339 204. Shichi T, Takagi K, Sawaki Y (1996) Chem Comm 2027 205. Tagaya H, Kuwahara T, Sato S, Kadokawa J, Karasu M, Chiba K (1993) J Mater Chem 3:317 206. Tagaya H, Sato S, Kuwahara T, Kadokawa J, Masa K, Chiba K (1994) J Mater Chem 4:1907 207. Schulz-Ekloff G, Wöhrle D, Van Duffel D, Schoonheydt RA (2002) Micropor Mesopor Mater 51:91 208. Hoppe R, Alberti G, Costantino U, Dionigi C, Schulz-Ekloff G, Vivani R (1997) Langmuir 13:7252 209. Zlatanova K, Markovsky P, Spassova I, Danev G (1996) Opt Mater 5:279 210. Bin Hussein MZ, Zainal Z, Yahaya AH, Aziz ABA (2002) Mater Sci Eng B 88:98 211. Jung W, Huh Y (1996) Bull Korean Chem Soc 17:547 212. Costantino U, Coletti N, Nocchetti M, Aloisi GG, Elisei F (1999) Langmuir 15:4454 213. Park LY, Kuroda K, Kato C (1990) J Chem Soc Dalton Trans, p 3071 214. Bauer J, Behrens P, Speckbacher M, Langhals H (2003) Adv Funct Mater 13:241 215. Guo S, Li D, Zhang W, Pu M, Evans DG, Duan X (2004) J Solid State Chem 177:4597 216. Cooper S, Dutta PK (1990) J Phys Chem 94:114 217. Xu G, Guo C, Duan X (1999) Chin J Appl Chem 16:45 218. El-Toni AM, Yin S, Sato T (2004) J Solid State Chem 177:3197 219. Jiao QZ, Zhao Y, Xie H, Evans DG, Duan X (2002) Chin J Appl Chem 19:1011 220. Merz A (1990) Chemically Modified Electrodes. In: Topics in Current Chemistry, vol. 152. Springer, Berlin 221. Jayashree RS, Kamath PV (2002) J Electrochem Soc 149:761

Applications of Layered Double Hydroxides

223

222. Chen H, Wang JM, Pan T, Xiao HM, Zhang JQ, Cao CN (2002) Int J Hydrogen Energy 27:489 223. Chen H, Wang JM, Pan T, Zhao YL, Zhang JQ, Cao CN (2003) J Electrochem Soc 150:1399 224. Jayashree RS, Kamath PV (2002) J Power Sources 107:120 225. Sakaebe H, Uchino H, Azuma M, Shikano M, Higuchi S (1998) Solid State Ionics 15:35 226. Mousty C, Therias S, Forano C, Besse JP (1994) J Electroanal Chem 63:374 227. Therias S, Mousty C (1995) Appl Clay Sci 10:10 228. Liao CS, Ye WB (2004) Electrochim Acta 49:4993 229. Vaccari A (1998) Catal Today 41:53 230. Vaccari A (1999) Appl Clay Sci 14:161 231. Miyata S, Kuroda M (1981) US Patent 4,299,759 232. Van Der Ven L, Van Gemert MLM, Batenburg LF, Keen JJ, Gielgens LH, Koster TPM, Fischer HR (2000) Appl Clay Sci 17:25 233. Lin Y, Li D, Evans DG, Duan X (2005) Polym Degrad Stabil 88:286 234. Wang XD, Zhang Q (2004) Polym Int 53:698 235. Camino G, Maffezzoli A, Braglia M, De Lazzaro M, Zammarano M (2001) Polym Degrad Stabil 74:457 236. Shi L, Li D, Wang J, Li S, Evans DG, Duan X (2005) Clays Clay Miner 53:293 237. Li F, Liu J, Evans DG, Duan X (2004) Chem Mater 16:1597 238. Liu J, Li F, Evans DG, Duan X (2003) Chem Comm 542 239. Meng W, Li F, Evans DG, Duan X (2004) J Mater Sci 39:4655 240. Meng W, Li F, Evans DG, Duan X (2004) Mater Phys Chem 86:1 241. Wong F, Buchheit RG (2004) Prog Org Coatings 51:91 242. Shan D, Cosnier S, Mousty C (2004) Biosens Bioelectron 20:390 243. Raki L, Beaudoin JJ, Mitchell L (2004) Cem Concr Res 34:1717 244. Renaudin G, François M, Evrard O (1999) Cem Concr Res 29:63

Author Index Volumes 101–119 Author Index Vols. 1–100 see Vol. 100

The volume numbers are printed in italics Alajarin M, see Turner DR (2004) 108: 97–168 Aldinger F, see Seifert HJ (2002) 101: 1–58 Alfredsson M, see Corà F (2004) 113: 171–232 Aliev AE, Harris KDM (2004) Probing Hydrogen Bonding in Solids Using State NMR Spectroscopy 108: 1–54 Alloul H, see Brouet V (2004) 109: 165–199 Amstutz N, see Hauser A (2003) 106: 81–96 Anitha S, Rao KSJ (2003) The Complexity of Aluminium-DNA Interactions: Relevance to Alzheimer’s and Other Neurological Diseases 104: 79–98 Anthon C, Bendix J, Schäffer CE (2004) Elucidation of Ligand-Field Theory. Reformulation and Revival by Density Functional Theory 107: 207–302 Aramburu JA, see Moreno M (2003) 106: 127–152 Arˇcon D, Blinc R (2004) The Jahn-Teller Effect and Fullerene Ferromagnets 109: 231–276 Atanasov M, Daul CA, Rauzy C (2003) A DFT Based Ligand Field Theory 106: 97–125 Atanasov M, see Reinen D (2004) 107: 159–178 Atwood DA, see Conley B (2003) 104: 181–193 Atwood DA, Hutchison AR, Zhang Y (2003) Compounds Containing Five-Coordinate Group 13 Elements 105: 167–201 Autschbach J (2004) The Calculation of NMR Parameters in Transition Metal Complexes 112: 1–48 Baerends EJ, see Rosa A (2004) 112: 49–116 Bard AJ, Ding Z, Myung N (2005) Electrochemistry and Electrogenerated Chemiluminescence of Semiconductor Nanocrystals in Solutions and in Films. 118: 1–57 Barriuso MT, see Moreno M (2003) 106: 127–152 Beaulac R, see Nolet MC (2004) 107: 145–158 Bellandi F, see Contreras RR (2003) 106: 71–79 Bendix J, see Anthon C (2004) 107: 207–302 Berend K, van der Voet GB, de Wolff FA (2003) Acute Aluminium Intoxication 104: 1–58 Bianconi A, Saini NL (2005) Nanoscale Lattice Fluctuations in Cuprates and Manganites 114: 287–330 Blinc R, see Arcˇcon D (2004) 109: 231–276 Boˇca R (2005) Magnetic Parameters and Magnetic Functions in Mononuclear Complexes Beyond the Spin-Hamiltonian Formalism 117: 1–268 Bohrer D, see Schetinger MRC (2003) 104: 99–138 Boulanger AM, see Nolet MC (2004) 107: 145–158

226

Author Index Volumes 101–119

Boulon G (2004) Optical Transitions of Trivalent Neodymium and Chromium Centres in LiNbO3 Crystal Host Material 107: 1–25 Bowlby BE, Di Bartolo B (2003) Spectroscopy of Trivalent Praseodymium in Barium Yttrium Fluoride 106: 193–208 Braga D, Maini L, Polito M, Grepioni F (2004) Hydrogen Bonding Interactions Between Ions: A Powerful Tool in Molecular Crystal Engineering 111: 1–32 Brouet V, Alloul H, Gàràj S, Forrò L (2004) NMR Studies of Insulating, Metallic, and Superconducting Fullerides: Importance of Correlations and Jahn-Teller Distortions 109: 165– 199 Buddhudu S, see Morita M (2004) 107: 115–144 Budzelaar PHM, Talarico G (2003) Insertion and β-Hydrogen Transfer at Aluminium 105: 141–165 Burrows AD (2004) Crystal Engineering Using Multiple Hydrogen Bonds 108: 55–96 Bussmann-Holder A, Keller H, Müller KA (2005) Evidences for Polaron Formation in Cuprates 114: 367–386 Bussmann-Holder A, see Micnas R (2005) 114: 13–69 Canadell E, see Sánchez-Portal D (2004) 113: 103–170 Cancines P, see Contreras RR (2003) 106: 71–79 Cartwright HM (2004) An Introduction to Evolutionary Computation and Evolutionary Algorithms 110: 1–32 Christie RA, Jordan KD (2005) n-Body Decomposition Approach to the Calculation of Interaction Energies of Water Clusters 116: 27–41 Clot E, Eisenstein O (2004) Agostic Interactions from a Computational Perspective: One Name, Many Interpretations 113: 1–36 Conley B, Atwood DA (2003) Fluoroaluminate Chemistry 104: 181–193 Contreras RR, Suárez T, Reyes M, Bellandi F, Cancines P, Moreno J, Shahgholi M, Di Bilio AJ, Gray HB, Fontal B (2003) Electronic Structures and Reduction Potentials of Cu(II) Complexes of [N,N
-Alkyl-bis(ethyl-2-amino-1-cyclopentenecarbothioate)] (Alkyl = Ethyl, Propyl, and Butyl) 106: 71–79 Corà F, Alfredsson M, Mallia G, Middlemiss DS, Mackrodt WC, Dovesi R, Orlando R (2004) The Performance of Hybrid Density Functionals in Solid State Chemistry 113: 171–232 Crespi VH, see Gunnarson O (2005) 114: 71–101 Daul CA, see Atanasov M (2003) 106: 97–125 Day P (2003) Whereof Man Cannot Speak: Some Scientific Vocabulary of Michael Faraday and Klixbüll Jørgensen 106: 7–18 Deeth RJ (2004) Computational Bioinorganic Chemistry 113: 37–69 Delahaye S, see Hauser A (2003) 106: 81–96 Deng S, Simon A, Köhler J (2005) Pairing Mechanisms Viewed from Physics and Chemistry 114: 103–141 Di Bartolo B, see Bowlby BE (2003) 106: 191–208 Di Bilio AJ, see Contreras RR (2003) 106: 71–79 Ding Z, see Bard AJ (2005) 118: 1–57 Dovesi R, see Corà F (2004) 113: 171–232 Duan X, see He J (2005) 119: 89–119 Duan X, see Li F (2005) 119: 193–223 Egami T (2005) Electron-Phonon Coupling in High-Tc Superconductors 114: 267–286 Eisenstein O, see Clot E (2004) 113: 1–36

Author Index Volumes 101–119

227

Evans DG, see He J (2005) 119: 89–119 Evans DG, Slade RCT (2005) Structural Aspects of Layered Double Hydroxides. 119: 1–87 Ewing GE (2005) H2 O on NaCl: From Single Molecule, to Clusters, to Monolayer, to Thin Film, to Deliquescence 116: 1–25 Fontal B, see Contreras RR (2003) 106: 71–79 Forrò L, see Brouet V (2004) 109: 165–199 Fowler PW, see Soncini A (2005) 115: 57–79 Frenking G, see Lein M (2003) 106: 181–191 Frühauf S, see Roewer G (2002) 101: 59–136 Frunzke J, see Lein M (2003) 106: 181–191 Furrer A (2005) Neutron Scattering Investigations of Charge Inhomogeneities and the Pseudogap State in High-Temperature Superconductors 114: 171–204 Gàràj S, see Brouet V (2004) 109: 165–199 Gillet VJ (2004) Applications of Evolutionary Computation in Drug Design 110: 133–152 Golden MS, Pichler T, Rudolf P (2004) Charge Transfer and Bonding in Endohedral Fullerenes from High-Energy Spectroscopy 109: 201–229 Gorelesky SI, Lever ABP (2004) 107: 77–114 Gray HB, see Contreras RR (2003) 106: 71–79 Grepioni F, see Braga D (2004) 111: 1–32 Gritsenko O, see Rosa A (2004) 112: 49–116 Güdel HU, see Wenger OS (2003) 106: 59–70 Gütlich P, van Koningsbruggen PJ, Renz F (2004) Recent Advances in Spin Crossover Research 107: 27–76 Gunnarsson O, Han JE, Koch E, Crespi VH (2005) Superconductivity in Alkali-Doped Fullerides 114: 71–101 Guyot-Sionnest P (2005) Intraband Spectroscopy and Semiconductor Nanocrystals. 118: 59–77 Habershon S, see Harris KDM (2004) 110: 55–94 Han JE, see Gunnarson O (2005) 114: 71–101 Hardie MJ (2004) Hydrogen Bonded Network Structures Constructed from Molecular Hosts 111: 139–174 Harris KDM, see Aliev (2004) 108: 1–54 Harris KDM, Johnston RL, Habershon S (2004) Application of Evolutionary Computation in Structure Determination from Diffraction Data 110: 55–94 Hartke B (2004) Application of Evolutionary Algorithms to Global Cluster Geometry Optimization 110: 33–53 Harvey JN (2004) DFT Computation of Relative Spin-State Energetics of Transition Metal Compounds 112: 151–183 Haubner R, Wilhelm M, Weissenbacher R, Lux B (2002) Boron Nitrides – Properties, Synthesis and Applications 102: 1–46 Hauser A, Amstutz N, Delahaye S, Sadki A, Schenker S, Sieber R, Zerara M (2003) Fine Tuning the Electronic Properties of [M(bpy)3 ]2+ Complexes by Chemical Pressure (M = Fe2+ , Ru2+ , Co2+ , bpy = 2,2
-Bipyridine) 106: 81–96 He J, Wei M, Li B, Kang Y, G Evans D, Duan X (2005) Preparation of Layered Double Hydroxides. 119: 89–119 Herrmann M, see Petzow G (2002) 102: 47–166 Herzog U, see Roewer G (2002) 101: 59–136

228

Author Index Volumes 101–119

Hoggard PE (2003) Angular Overlap Model Parameters 106: 37–57 Höpfl H (2002) Structure and Bonding in Boron Containing Macrocycles and Cages. 103: 1–56 Hubberstey P, Suksangpanya U (2004) Hydrogen-Bonded Supramolecular Chain and Sheet Formation by Coordinated Guranidine Derivatives 111: 33–83 Hutchison AR, see Atwood DA (2003) 105: 167–201 Iwasa Y, see Margadonna S (2004) 109: 127–164 Jansen M, Jäschke B, Jäschke T (2002) Amorphous Multinary Ceramics in the Si-B-N-C System 101: 137–192 Jäschke B, see Jansen M (2002) 101: 137–192 Jäschke T, see Jansen M (2002) 101: 137–192 Jaworska M, Macyk W, Stasicka Z (2003) Structure, Spectroscopy and Photochemistry of the [M(η5 -C5 H5 )(CO)2 ]2 Complexes (M = Fe, Ru) 106: 153–172 Jenneskens LW, see Soncini A (2005) 115: 57–79 Jeziorski B, see Szalewicz K (2005) 116: 43–117 Johnston RL, see Harris KDM (2004) 110: 55–94 Jordan KD, see Christie RA (2005) 116: 27–41 Kabanov VV, see Mihailovic D (2005) 114: 331–365 Kang Y, see He J (2005) 119: 89–119 Keller H (2005) Unconventional Isotope Effects in Cuprate Superconductors 114: 143–169 Keller H, see Bussmann-Holder A (2005) 114: 367–386 Khan AI, see Williams GR (2005) 119: 161–192 Koch E, see Gunnarson O (2005) 114: 71–101 Kochelaev BI, Teitel’baum GB (2005) Nanoscale Properties of Superconducting Cuprates Probed by the Electron Paramagnetic Resonance 114: 205–266 Köhler J, see Deng (2005) 114: 103–141 van Koningsbruggen, see Gütlich P (2004) 107: 27–76 Lein M, Frunzke J, Frenking G (2003) Christian Klixbüll Jørgensen and the Nature of the Chemical Bond in HArF 106: 181–191 Leroux F, see Taviot-Gueho C (2005) 119: 121–159 Lever ABP, Gorelesky SI (2004) Ruthenium Complexes of Non-Innocent Ligands; Aspects of Charge Transfer Spectroscopy 107: 77–114 Li B, see He J (2005) 119: 89–119 Li F, Duan X (2005) Applications of Layered Double Hydroxides. 119: 193–223 Liebau F, see Santamaría-Pérez D (2005) 118: 79–135 Linton DJ, Wheatley AEH (2003) The Synthesis and Structural Properties of Aluminium Oxide, Hydroxide and Organooxide Compounds 105: 67–139 Lux B, see Haubner R (2002) 102: 1–46 Mackrodt WC, see Corà F (2004) 113: 171–232 Macyk W, see Jaworska M (2003) 106: 153–172 Mahalakshmi L, Stalke D (2002) The R2M+ Group 13 Organometallic Fragment Chelated by P-centered Ligands 103: 85–116 Maini L, see Braga D (2004) 111: 1–32 Mallia G, see Corà F (2004) 113: 171–232

Author Index Volumes 101–119

229

Margadonna S, Iwasa Y, Takenobu T, Prassides K (2004) Structural and Electronic Properties of Selected Fulleride Salts 109: 127–164 Maseras F, see Ujaque G (2004) 112: 117–149 Micnas R, Robaszkiewicz S, Bussmann-Holder A (2005) Two-Component Scenarios for Non-Conventional (Exotic) Superconstructors 114: 13–69 Middlemiss DS, see Corà F (2004) 113: 171–232 Mihailovic D, Kabanov VV (2005) Dynamic Inhomogeneity, Pairing and Superconductivity in Cuprates 114: 331–365 Millot C (2005) Molecular Dynamics Simulations and Intermolecular Forces 115: 125–148 Miyake T, see Saito (2004) 109: 41–57 Moreno J, see Contreras RR (2003) 106: 71–79 Moreno M, Aramburu JA, Barriuso MT (2003) Electronic Properties and Bonding in Transition Metal Complexes: Influence of Pressure 106: 127–152 Morita M, Buddhudu S, Rau D, Murakami S (2004) Photoluminescence and Excitation Energy Transfer of Rare Earth Ions in Nanoporous Xerogel and Sol-Gel SiO2 Glasses 107: 115–143 Morsch VM, see Schetinger MRC (2003) 104: 99–138 Mossin S, Weihe H (2003) Average One-Center Two-Electron Exchange Integrals and Exchange Interactions 106: 173–180 Murakami S, see Morita M (2004) 107: 115–144 Müller E, see Roewer G (2002) 101: 59–136 Müller KA (2005) Essential Heterogeneities in Hole-Doped Cuprate Superconductors 114: 1–11 Müller KA, see Bussmann-Holder A (2005) 114: 367–386 Myung N, see Bard AJ (2005) 118: 1–57 Nishibori E, see Takata M (2004) 109: 59–84 Nolet MC, Beaulac R, Boulanger AM, Reber C (2004) Allowed and Forbidden d-d Bands in Octa-hedral Coordination Compounds: Intensity Borrowing and Interference Dips in Absorption Spectra 107: 145–158 O’Hare D, see Williams GR (2005) 119: 161–192 Ordejón P, see Sánchez-Portal D (2004) 113: 103–170 Orlando R, see Corà F (2004) 113: 171–232 Oshiro S (2003) A New Effect of Aluminium on Iron Metabolism in Mammalian Cells 104: 59–78 Pastor A, see Turner DR (2004) 108: 97–168 Patkowski K, see Szalewicz K (2005) 116: 43–117 Patoˇcka J, see Strunecká A (2003) 104: 139–180 Peng X, Thessing J (2005) Controlled Synthesis of High Quality Semiconductor Nanocrystals. 118: 137–177 Petzow G, Hermann M (2002) Silicon Nitride Ceramics 102: 47–166 Pichler T, see Golden MS (2004) 109: 201–229 Polito M, see Braga D (2004) 111: 1–32 Popelier PLA (2005) Quantum Chemical Topology: on Bonds and Potentials 115: 1–56 Power P (2002) Multiple Bonding Between Heavier Group 13 Elements. 103: 57–84 Prassides K, see Margadonna S (2004) 109: 127–164 Prato M, see Tagmatarchis N (2004) 109: 1–39 Price LS, see Price SSL (2005) 115: 81–123

230

Author Index Volumes 101–119

Price SSL, Price LS (2005) Modelling Intermolecular Forces for Organic Crystal Structure Prediction 115: 81–123 Rao KSJ, see Anitha S (2003) 104: 79–98 Rau D, see Morita M (2004) 107: 115–144 Rauzy C, see Atanasov (2003) 106: 97–125 Reber C, see Nolet MC (2004) 107: 145–158 Reinen D, Atanasov M (2004) The Angular Overlap Model and Vibronic Coupling in Treating s-p and d-s Mixing – a DFT Study 107: 159–178 Reisfeld R (2003) Rare Earth Ions: Their Spectroscopy of Cryptates and Related Complexes in Glasses 106: 209–237 Renz F, see Gütlich P (2004) 107: 27–76 Reyes M, see Contreras RR (2003) 106: 71–79 Ricciardi G, see Rosa A (2004) 112: 49–116 Riesen H (2004) Progress in Hole-Burning Spectroscopy of Coordination Compounds 107: 179–205 Robaszkiewicz S, see Micnas R (2005) 114: 13–69 Roewer G, Herzog U, Trommer K, Müller E, Frühauf S (2002) Silicon Carbide – A Survey of Synthetic Approaches, Properties and Applications 101: 59–136 Rosa A, Ricciardi G, Gritsenko O, Baerends EJ (2004) Excitation Energies of Metal Complexes with Time-dependent Density Functional Theory 112: 49–116 Rudolf P, see Golden MS (2004) 109: 201–229 Ruiz E (2004) Theoretical Study of the Exchange Coupling in Large Polynuclear Transition Metal Complexes Using DFT Methods 113: 71–102 Sadki A, see Hauser A (2003) 106: 81–96 Saini NL, see Bianconi A (2005) 114: 287–330 Saito S, Umemoto K, Miyake T (2004) Electronic Structure and Energetics of Fullerites, Fullerides, and Fullerene Polymers 109: 41–57 Sakata M, see Takata M (2004) 109: 59–84 Sánchez-Portal D, Ordejón P, Canadell E (2004) Computing the Properties of Materials from First Principles with SIESTA 113: 103–170 Santamaría-Pérez D, Vegas A, Liebau F (2005) The Zintl–Klemm Concept Applied to Cations in Oxides II. The Structures of Silicates. 118: 79–135 Schäffer CE (2003) Axel Christian Klixbüll Jørgensen (1931–2001) 106: 1–5 Schäffer CE, see Anthon C (2004) 107: 207–301 Schenker S, see Hauser A (2003) 106: 81–96 Schetinger MRC, Morsch VM, Bohrer D (2003) Aluminium: Interaction with Nucleotides and Nucleotidases and Analytical Aspects of Determination 104: 99–138 Schmidtke HH (2003) The Variation of Slater-Condon Parameters Fk and Racah Parameters B and C with Chemical Bonding in Transition Group Complexes 106: 19–35 Schubert DM (2003) Borates in Industrial Use 105: 1–40 Schulz S (2002) Synthesis, Structure and Reactivity of Group 13/15 Compounds Containing the Heavier Elements of Group 15, Sb and Bi 103: 117–166 Seifert HJ, Aldinger F (2002) Phase Equilibria in the Si-B-C-N System 101: 1–58 Shahgholi M, see Contreras RR (2003) 106: 71–79 Shinohara H, see Takata M (2004) 109: 59–84 Sieber R, see Hauser A (2003) 106: 81–96 Simon A, see Deng (2005) 114: 103–141 Slade RCT, see Evans DG (2005) 119: 1–87

Author Index Volumes 101–119

231

Soncini A, Fowler PW, Jenneskens LW (2005) Angular Momentum and Spectral Decomposition of Ring Currents: Aromaticity and the Annulene Model 115: 57–79 Stalke D, see Mahalakshmi L (2002) 103: 85–116 Stasicka Z, see Jaworska M (2003) 106: 153–172 Steed JW, see Turner DR (2004) 108: 97–168 Strunecká A, Patoˇcka J (2003) Aluminofluoride Complexes in the Etiology of Alzheimer’s Disease 104: 139–180 Suárez T, see Contreras RR (2003) 106: 71–79 Suksangpanya U, see Hubberstey (2004) 111: 33–83 Sundqvist B (2004) Polymeric Fullerene Phases Formed Under Pressure 109: 85–126 Szalewicz K, Patkowski K, Jeziorski B (2005) Intermolecular Interactions via Perturbation Theory: From Diatoms to Biomolecules 116: 43–117 Tagmatarchis N, Prato M (2004) Organofullerene Materials 109: 1–39 Takata M, Nishibori E, Sakata M, Shinohara M (2004) Charge Density Level Structures of Endohedral Metallofullerenes by MEM/Rietveld Method 109: 59–84 Takenobu T, see Margadonna S (2004) 109: 127–164 Talarico G, see Budzelaar PHM (2003) 105: 141–165 Taviot-Gueho C, Leroux F (2005) In situ Polymerization and Intercalation of Polymers in Layered Double Hydroxides. 119: 121–159 Teitel’baum GB, see Kochelaev BI (2005) 114: 205–266 Thessing J, see Peng X (2005) 118: 137–177 Trommer K, see Roewer G (2002) 101: 59–136 Tsuzuki S (2005) Interactions with Aromatic Rings 115: 149–193 Turner DR, Pastor A, Alajarin M, Steed JW (2004) Molecular Containers: Design Approaches and Applications 108: 97–168 Uhl W (2003) Aluminium and Gallium Hydrazides 105: 41–66 Ujaque G, Maseras F (2004) Applications of Hybrid DFT/Molecular Mechanics to Homogeneous Catalysis 112: 117–149 Umemoto K, see Saito S (2004) 109: 41–57 Unger R (2004) The Genetic Algorithm Approach to Protein Structure Prediction 110: 153–175 van der Voet GB, see Berend K (2003) 104: 1–58 Vegas A, see Santamaría-Pérez D (2005) 118: 79–135 Vilar R (2004) Hydrogen-Bonding Templated Assemblies 111: 85–137 Wei M, see He J (2005) 119: 89–119 Weihe H, see Mossin S (2003) 106: 173–180 Weissenbacher R, see Haubner R (2002) 102: 1–46 Wenger OS, Güdel HU (2003) Influence of Crystal Field Parameters on Near-Infrared to Visible Photon Upconversion in Ti2+ and Ni2+ Doped Halide Lattices 106: 59–70 Wheatley AEH, see Linton DJ (2003) 105: 67–139 Wilhelm M, see Haubner R (2002) 102: 1–46 Williams GR, Khan AI, O’Hare D (2005) Mechanistic and Kinetic Studies of Guest Ion Intercalation into Layered Double Hydroxides Using Time-resolved, In-situ X-ray Powder Diffraction. 119: 161–192 de Wolff FA, see Berend K (2003) 104: 1–58

232

Author Index Volumes 101–119

Woodley SM (2004) Prediction of Crystal Structures Using Evolutionary Algorithms and Related Techniques 110: 95–132 Xantheas SS (2005) Interaction Potentials for Water from Accurate Cluster Calculations 116: 119–148 Zerara M, see Hauser A (2003) 106: 81–96 Zhang Y, see Atwood DA (2003) 105: 167–201

Subject Index

Acid Blue 29, 206 Acrylonitrile 150 Aging method 110 Agrochemicals, intercalation 186 Alginate 121 Aminosuccinic acid 151 Anion-exchange capacities 203 Antipepsin agents 208 Benzenedicarboxylate 180 Biocompatibility 208 Bio-LDH hybrid 152 Biomolecules, intercalation 184 Biphenylacetic acid 184 Bragg reflection 169, 179 Brucite 3 C60 109 Calcination 194 Calcination-rehydration effect 107 Carboxylic acids 175 Catalysis, environmental 199 Catalyst supports 195 Catalysts, Cu-based 201 Catalytic partial oxidation 200 Chlorocinnamate, LDH layers 46 Cinnamate, LDH layers 46 Coprecipitation 92 Diclofenac 184 DNA 121, 210 DNA-LDH hybrid 152, 211 Drug carriers 211 Drug delivery system 209 Elaidic acid, LDH 45 Electrochemistry 213 Electrosynthesis 112 Enzymes, immobilized 109

Ethanol 201 Ethylene vinyl acetate 215 Ettringite 205 Fluidized catalytic cracking units 195 Galleries, interlayer 53 Glutamate-pillared LDHs 109, 196 HCH 202 Hexamethylenetetramine 102 Hydrocalumite 51, 141, 205 Hydrogen 200 Hydrolysis method 100 Hydrotalcite 1, 89, 194 Hydrothermal methods 108 Hydroxide method 110 Hydroxides, double, layered 1, 89 Indolinespirobenzopyran, sulfonated 211 Indomethacin 209 Intercalation 161 – secondary 108 Ion-exchange method 103 Layered double oxide 131 LDH/biopolymers 151 Li salts 169 Limonene 203 Macro-defect-free cement 122 Magnetic materials 215 Memory effect 106, 129, 204, 207 Merocyanine 211 Metallomacrocycles 202 Mn-porphyrins 202 Montmorillonite 122

234

Subject Index

Naproxen 184 Nickel hydroxide, layered 213 Nucleation 164 Nuclei growth 165 Oxoanions

204

Paramolybdate, LDH 25 Penicillin G acylase 109, 196 PEO 122 Pesticides 207 Phenylethenylbenzoates 211 Photochemistry 211 Phthalocyanine 202 Polyaniline (PANI) 122 Polymer/LDH nanocomposites 129 Polyoxometalates 129, 201 Poly(acrylic acid) 129 Poly(aspartate) 121, 151 Poly(ethylene oxide) 109, 214 Poly(vinyl chloride) 214 Poly(vinyl sulfonate) 129 PPP/MoO3 123 Pre-pillaring method 108 PVP-kaolinite 123

Quenching

161

Reductive intercalative polymerization 123, 134 Rietveld refinement, LDH, interlayer structure 27 Salt-oxide method 110 Selective catalytic reduction 196 Shigaite 63 Sj¨ ogrenite-hydrotalcite 3 Sol-gel method 112 Stearate/LDH 45 Structural memory effect 106 Surface synthesis 111 Synchrotron radiation 161 Vinyl benzene sulfonate 138 Vitamin C/E 210 X-ray diffraction 161 – energy dispersive 166 Zaccagnite 63