Proceedings of the International Biohydrometallurgy Symposium IBS 2003, Athens, Greece - Vol I

BIOHYDROMETALLURGY "A SUSTAINABLE TECHNOLOGY IN EVOLUTION" Proceedings of the International Biohydrometallurgy Symposi...

Author: Marios Tsezos | Artin Hatzikioseyian | Emmanouela Remoundaki (eds.)

11 downloads 921 Views 11MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

BIOHYDROMETALLURGY "A SUSTAINABLE TECHNOLOGY IN EVOLUTION"

Proceedings of the International Biohydrometallurgy Symposium, IBS 2003, held in Athens, Hellas, September 14-19, 2003

Part I Bioleaching Applications, Bioremediation Environmental Applications

Edited by Marios Tsezos Artin Hatzikioseyian Emmanouela Remoudaki

Associate Editors Pavlina Kousi Roza Vidali

NATIONAL TECHNICAL UNIVERSITY OF ATHENS School of Mining and Metallurgical Enginnering Laboratory of Environmental Science and Technology Heroon Polytechniou 9, 157 80 Zografou, Greece Tel: (+30) 2107722172, (+30) 2107722271, Fax: (+30) 2107722173 Contact: Professor Marios Tsezos, e-mail: [email protected]

© National Technical University of Athens. All rights reserved.

This work and the individual contributions contained in it are protected under the following terms and conditions: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Derivative works Permission of the publisher is required for all other derivative works, including compilations and translations. Electronic storage or usage Permission of the publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Contact the publisher at the address indicated. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without any prior written permission of the publisher. Address permission requests to: National Technical University of Athens, at the mail, fax and e-mail addresses noted above. Notice No responsibility is assumed of the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein.

First edition 2004

Electronic means of production – Cover design N. & J. ZOGRAPHAKIS Tel.: (+30) 2109731912 Fax: (+30) 2109707707

Printed in Greece. ISBN- (Set) : 960-88415-0-X ISBN- (Part I) : 960-88415-1-8

15th International Biohydrometallurgy Symposium (IBS 2003) September 14-19, Athens, Hellas "Biohydrometallurgy: a sustainable technology in evolution"

Preface The present edition includes the proceedings of the 15th International Biohydrometallurgy Symposium (IBS 2003) held in Athens, Greece on September 14th19th, 2003. Continuing the effort on the understanding of the interactions between metals and microbial cells and on developing and applying biohydrometallurgical processes, International Biohydrometallurgy Symposia offer the opportunity of exchanging international experience on a wide range of topics from metal extraction to environmental remediation. During IBS 2003, the issues of sustainability and environmental remediation, worldwide priorities, were addressed from different points of view. Biohydrometallurgy is a sustainable innovative technology, which in many cases, during the last decade, has successfully replaced classical metal extraction processes, from minerals and rocks. Combining a competitive technology with minimum environmental impact is the challenge for optimization of technologies applied today and/or to be applied in the future. Recent advances towards the quantitative description of the interactions between metals and microbial cells as well as the identification of key parameters controlling these interactions, play an important role in metal extraction processes optimization and in the development of treatment technologies for liquid and solid metallurgical discharges. The 15th International Biohydrometallurgy Symposium opened by the invited plenary lecture: "Biohydrometallurgy: a sustainable technology in evolution" given by Professor Giovanni Rossi from the University of Cagliari, Italy. Professor Rossi honored the Symposium with his presence, reviewed the state of the art and pointed out to the future trends in different areas of biohydrometallurgy. The Symposium was organized along five sessions: Bioleaching Applications and Technology Developments. Bioremediation – Environmental Applications. Biosorption Fundamentals and Technology Developments. Microbiology Fundamentals. Molecular Biology and Taxonomy. All papers included in the present edition were previously reviewed by a minimum of two experts selected among the International Scientific Committee Members as well as among prestigious researchers in the biohydrometallurgy science and technology fields. Among the 160 papers included in this edition, the Organising Committee aimed at providing the opportunity to the Symposium participants to attend as many oral presentations as possible, according to originality and scientific merit. Sixty-five oral i

Preface

presentations were made during IBS 2003. The rest of the communications were presented in the poster session. A Closing Session, chaired by a panel of experts and pioneers in the corresponding areas, was organized to conclude the main topics of the Conference and to point out future trends in scientific areas of Biohydrometallurgy. From this position, we wish to acknowledge all the members of the International Scientific Committee: Antonio Ballester, Barrie Johnson, Bohumil Volesky, Borje Lindstrom, Carlos Jerez, Corale Brierley, David Holmes, Dominique Morin, Douglas Rawlings, Edgardo Donati, Eric Guibal, Giovanni Rossi, Gregory Karavaiko, Henry Erhlich, James Brierley, John Duncan, K. A. Natarajan, Kishore Paknikar, Klaus Bosecker, Marios Tsezos, Olli Tuovinen, Paul Norris, Piet Bos, Ralph Hackl, Ricardo Amils, Stoyan Groudev, Tomas Vargas, Tsuyoshi Sugio, Virginia Ciminelli, Wolfgang Sand, for participating in the reviewing and selection of the manuscripts submitted to IBS 2003. We also wish to express our appreciation to prestigious researchers non members of the International Scientific Committee for assisting the reviewing procedure: Anthimos Xenidis, Frantz Glombitza, Georgios Anastassakis, Konstantinos Komnitsas, Ludo Diels, Lynne Macaskie, Nymphodora Papassiopi, Styliani Agatzini-Leonardou. We also thank our colleagues at IBS 2003 from the National Organizing Committee: Anthimos Xenidis, Emmanouil Zevgolis, Georgios Anastassakis, Konstatina Tsaimou, Konstantinos Komnitsas, Nymphodora Papassiopi, Simos Simopoulos, Styliani AgatziniLeonardou, for their valuable assistance and support. Acknowledgements are also due to the many others who participated in the organization of the Symposium, the authors and the many participants who represented many countries around the world. Special thanks also to Mrs Pavlina Kousi and Mrs Roza Vidali, Ph.D candidate students of our Laboratory, for editing the final manuscripts for the preparation of the hardcopies of the IBS 2003 proceedings. Finally, we wish to thank the National Technical University of Athens (NTUA), The Ministry of Development: General Secretariat of Science and Technology, The Hellenic Ministry of Culture, The Hellenic Technical Chamber, The National Institute of Geology and Mineral Exploration for supporting the Symposium. Professor Marios Tsezos Dr. Emmanouela Remoudaki Dr. Artin Hatzikioseyian National Technical University of Athens, School of Mining and Metallurgical Engineering, Laboratory of Environmental Science and Engineering

ii


Table of contents Preface .................................................................................................................................. i Table of contents ................................................................................................................ iii

PART I PLENARY LECTURE Biohydrometallurgy: a sustainable technology in evolution Giovanni Rossi ...................................................................................................................... 3

CHAPTER 1 BIOLEACHING APPLICATIONS A novel bio-leaching process to recover valuable metals from Indian Ocean nodules using a marine isolate Mukherjee A., Raichur A.M., Modak J.M., Natarajan K.A. ............................................... 25 A novel biotechnological process for germanium recovery from brown coal Xianwan Y., Yun Z., Yuxia G., Banghui G. ......................................................................... 35 Aerobic and anaerobic bacterial leaching of manganese Zafiratos J.G., Agatzini-Leonardou S. ................................................................................ 41 Anaerobic iron sulfides oxidation Schippers A. ........................................................................................................................ 55 Bacterial growth and propagation in chalcocite heap bioleach scenarios Petersen J., Dixon D.G. ...................................................................................................... 65 Bacterial leaching studies of a Portuguese flotation tailing Costa M.C., Carvalho N., Iglesias N., Palencia I. ............................................................. 75

iii

Table of contents

Bacterial tank leaching of zinc from flotation tailings Panin V.V., Adamov E.V., Krylova L.N., Pivovarova T.A., Voronin D.Yu., Karavaiko G.I. .................................................................................................................... 85 Behaviour of elemental sulfur in the biohydrometallurgical processing of refractory gold-sulfide concentrates of various mineral types Sedelnikova G.V., Savari E.E. ............................................................................................ 91 Beneficiation of phosphatic ores from Hirapur, India Agate A.D. ........................................................................................................................ 101 Biohydrometallurgy of antimony gold-bearing ores and concentrates Solozhenkin P.M., Nebera V.P. ........................................................................................ 107 Bioleaching of Argentinean sulfide ores using pure and mixed cultures Frizan V., Giaveno A., Chiacchiarini P., Donati E. ......................................................... 117 Bioleaching of complex gold-lead ores Ulberg Z., Podolska V., Yermolenko A., Yakubenko L., Pertsov N. ................................. 127 Bioleaching of electronic scrap material by Aspergillus niger Ten W.K., Ting Y.P. .......................................................................................................... 137 Bioleaching of metallic sulphide concentrate in continuous stirred reactors at industrial scale – Experience and lessons Morin D., d’Hugues P., Mugabi M. ................................................................................. 147 Bioleaching of natural zeolite – the processes of iron removal and chamfer of clinoptilolite grains Styriakova I., Kolousek D., Styriak I., Lengauer C., Tillmanns E. ................................... 157 Bioleaching of pyrite by defined mixed populations of moderately thermophilic acidophiles in pH-controlled bioreactors Okibe N., Johnson D.B. .................................................................................................... 165 Biolixiviation of Cu, Ni, Pb and Zn using organic acids produced by Aspergillus niger and Penicillium simplicissinum Galvez-Cloutier R., Mulligan C., Ouattara A. ................................................................. 175 Biooxidation of pyrite by Acidithiobacillus ferrooxidans in single- and multi-stage continuous reactors Canales C., Gentina J.C., Acevedo F. .............................................................................. 185 Chemical chalcopyrite leaching and biological ferric solvent production at pH below 1 Kinnunen P.H.-M., Salo V.L.A., Pehkonen S.O., Puhakka J.A. ....................................... 193

iv

Table of contents

Comparative study of the bioleaching of two concentrates of chalcopyrite using mesophilic microorganisms in the presence of Ag(I) Lopez-Juarez A., Rivera-Santillan R.E. ............................................................................ 203 Comparison of air-lift and stirred tank batch and semi continuous bioleaching of polymetallic bulk concentrate Tipre D.R., Vora S.B., Dave S.R. ...................................................................................... 211 Effect of pH and temperature on the biooxidation of a refractory gold concentrate by Sulfolobus metallicus Nancucheo I., Gentina J.C., Acevedo F. .......................................................................... 219 Effect of the pulp density and particle size on the biooxidation rate of a pyritic gold concentrate by Sulfolobus metallicus Valencia P., Gentina J.C., Acevedo F. ............................................................................. 227 Enhancement of chalcopyrite bioleaching capacity of an extremely thermophilic culture by addition of ferrous sulphate Rubio A., Garcia Frutos F.J. ............................................................................................ 235 Evaluation of microbial leaching of uranium from Sierra Pintada ore. Preliminary studies in laboratory scale Paulo P.S., Pivato D., Vigliocco A., Lopez J., Castillo A. ................................................ 243 Extraction of copper from mining residues and sediments by addition of rhamnolipids Mulligan C.N., Dahrazma B. ............................................................................................ 253 Improving of film coating bioleaching using biorotor process Shahverdi A.R., Oliazadeh M., Rohi R., Davodi M. ......................................................... 261 Isolation and evaluation of indigenous iron- and sulphur-oxidising bacteria for heavy metal removal from sewage sludge Matlakowska R., Sklodowska A. ....................................................................................... 265 Kinetics of ferrous iron oxidation with Sulfolobus metallicus at 70°C Meruane G., Carcamo C., Vargas T. ............................................................................... 277 Kinetics of sulphur oxidation: pH and temperature influence on bioleaching Patino E., Sandoval R., Frenay J. .................................................................................... 285 Leaching of iron from China clay with oxalic acid: effect of acid concentration, pH, temperature, solids concentration and shaking Mandal S.K., Banerjee P.C. ............................................................................................. 291 Mathematical modeling of the chemical and bacterial leaching of copper ores in stack Zeballos F., Filho O.B., de Carvalho R.J. ........................................................................ 301 v

Table of contents

Model for bacterial leaching of copper sulphides by forced aeration Sidborn M., Moreno L. ..................................................................................................... 311 Optimal oxygen and carbon dioxide concentrations for thermophilic bioleaching archaea de Kock S.H., Naldrett K., du Plessis C.A. ....................................................................... 319 Optimization study on bioleaching of municipal solid waste (MSW) incineration fly ash by Aspergillus niger Xu T.J., Ting Y.P. .............................................................................................................. 329 Production of an Acidithiobacillus ferrooxidans biomass using electrochemical regeneration of energetic substrate Morra C., Gondrexon N., Magnin J.-P., Deseure J., Ozil P. ........................................... 337 Removal of dibenzothiophene from fossil fuels with the action of iron(III)-ion generated by Thiobacillus ferrooxidans: Analytical aspects Beskoski V.P., Matic V., Spasic S., Vrvic M.M. ................................................................ 345 Solids loading in the bioleach slurry reactor: mechanisms through which particulate parameters influence slurry bioreactor performance Harrison S.T.L., Sissing A., Raja S., Pearce S.J.A., Lamaignere V., Nemati M. ............. 359 The development of a hybrid biological leaching-pressure oxidation process for auriferous arsenopyrite/pyrite feedstocks Dymov I., Ferron C.J., Phillips W. ................................................................................... 377 The development of the first commercial GEOCOAT® heap leach for refractory gold at the Agnes mine, Barberton South Africa Harvey T.J., Bath M. ........................................................................................................ 387 The electrochemistry of chalcopyrite bioleaching using bacteria modified powder micro-electrode Hongxu L., Dianzuo W., Yuehua H., Renman R. .............................................................. 399 The influence of crystal orientation on the bacterial dissolution of pyrite Ndlovu S., Monhemius A.J. ............................................................................................... 409 The influence of temperature and pH on the bioleaching of copper from a flotation concentrate of chalcopyrite Medrano-Roldan H., Salazar M.F.M., Pereyra-Alférez B., Solis-Soto A., Ramirez-Rodriguez D.G., Alvarez-Rosales E., Galan-Wong L.J. .................................... 419 The role of chemolitotrophic bacteria in the oxide copper ore heap leaching operation at Sarcheshmeh Copper Mine Seyed Baghery S.A., Shahverdi A.R., Oliazadeh M. ......................................................... 423

vi

Table of contents

Three-stage revolving drum biohydrometallurgical reactor for continuous operation Loi G., Trois P., Rossi G. ................................................................................................. 429 Use of biosurfactants for the mineral surfaces modification Sadowski Z., Maliszewska I., Polowczyk I. ....................................................................... 439

CHAPTER 2 BIOREMEDIATION ENVIRONMENTAL APPLICATIONS A new bench scale restoration method for a mercury-polluted soil with a mercury resistant Acidithiobacillus ferrooxidans strain SUG 2-2 Negishi A., Maeda T., Takeuchi F., Kamimura K., Sugio T. ............................................ 449 A novel type of microbial metal mobilization: cyanogenic bacteria and fungi solubilize metals as cyanide complexes Brandl H., Stagars M., Faramarzi M.A. ........................................................................... 457 An approach to cyanide degradation in wastewater of gold ore processing Podolska V., Ulberg Z., Pertsov N., Yakubenko L., Imanakunov B. ................................ 465 Available options for the bioremediation and restoration of abandoned pyritic dredge spoils causing the death of fringing mangroves in the Niger Delta Ohimain E. I. .................................................................................................................... 475 Bacterial reduction of TcO4- under the haloalkaline conditions Khijniak T., Medvedeva-Lyalikova N.N., Simonoff M. ..................................................... 483 Biodegradation of cyanides under saline conditions by a mixotrophic Pseudomonas putida Bipinraj N.K., Joshi N.R., Paknikar K.M. ........................................................................ 491 Bioleach of a fluvial tailings deposit material indicates long term potential for pollution Willscher S., Clark T.R., Cohen R.H., Ranville J.F., Smith K.S., Walton-Day K. ............ 497 Bioleaching of copper converter slag using A. ferrooxidans Seyed Baghery S.A., Oliazadeh M. ................................................................................... 507 Biooxidation of mine tailings using a mixed bacterial population Zahari M.A.K.M., Jaapar J., Bunyok M.A., Sohor S.H., Ahmad W.A. ............................. 513 Chromate reduction by immobilized cells of Desulfovibrio vulgaris using biologically produced hydrogen Humphries A.C., Penfold D.W., Forster C.F., Macaskie L.E. ......................................... 525 vii

Table of contents

Clean-up of mine waters from a uranium deposit by means of a constructed wetland Groudev S.N., Komnitsas K., Spasova I.I., Paspaliaris I. ................................................ 533 Degradation of tetracyanonickelate (II) by Cryptococcus humicolus in biofilm reactors Kwon H.K., Woo S.H., Sung J.Y., Park J.M. .................................................................... 541 Development of a bio-process using sulfate-reducing bacteria to remove metals from surface treatment effluents Battaglia-Brunet F., Foucher S., Denamur A., Chevard S., Morin D., Ignatiadis I. ....... 549 Effects of total-solids concentration on metal bioleaching from sewage sludge Villar L.D., Garcia O. Jr .................................................................................................. 559 Enhancement of electrodialytic soil remediation through biosorption Jensen P.E., Ottosen L.M., Ahring B.K. ........................................................................... 567 Fundamentals of the uranium separation in constructed wetlands Glombitza F., Karnatz F., Fischer H., Pinka J., Janneck E. ............................................ 575 Geomicrobiological risk assessment of abandoned mining sites Bosecker K., Mengel-Jung G., Schippers A. ..................................................................... 585 Immobilisation and growth of Acidithiobacillus ferrooxidans on refractory clay tiles Donati E., Martinez L., Curutchet G. ............................................................................... 595 Investigation of bioremediation techniques for cleaning-up arsenic contaminated soils Vaxevanidou K., Papassiopi N., Paspaliaris I. ................................................................ 603 Leaching characteristics of heavy metals from sewage sludge by Acidithiobacillus thiooxidans MET Cho K.S., Moon H.S., Yoo N.Y., Ryu H.W. ....................................................................... 613 Mercury removal by polymer-enhanced ultrafiltration using chitosan as the macroligand Kuncoro E.K., Lehtonen T., Roussy J., Guibal E. ............................................................ 621 Microbial recovery of copper from printed circuit boards of waste computer by Acidithiobacillus ferrooxidans Cho K.S., Choi M.S., Hong J.H., Kim D.S., Ryu H.W., Kim D.J., Sohn J.S., Park K.H. .. 631 Oxidation of iron, sulfur and arsenic in mine waters and mine wastes: an important role for novel Thiomonas spp Coupland K., Battaglia-Brunet F., Hallberg K.B., Dictor M.C., Garrido F., Johnson D.B. ................................................................................................................................... 639 viii

Table of contents

Oxidation of metallic copper by Acidothiobacillus Ferrooxidans Lilova K., Karamanev D. .................................................................................................. 647 Process monitoring of biodesulfurization of high sulfur coal in packed columns using molecular ecology methods Gómez F., Cara J., Carballo M.T., Moran A., Amils R., García Frutos F.J. .................. 653 Regeneration of hydrogen sulfide using sulfate reducing bacteria for photo catalytic hydrogen generation Takahashi Y., Suto K., Inoue C., Chida T. ........................................................................ 663 Remediation of sites contaminated by heavy metals: sustainable approach for unsaturated and saturated zones Diels L., Geets J., Vos J., Van Broekhoven K., Bastiaens L. ............................................ 671 Removal of chromium(VI) through a two-step process using sulphur-oxidising and sulphate-reducing bacteria Donati E., Viera M., Curutchet G. ................................................................................... 681 Removal of Mn(II) ions by manganese-oxidizing fungus at neutral pHs in the presence of carbon fiber Sasaki K., Endo M., Takano K., Konno H. ....................................................................... 689 Simultaneous removal of oil and heavy metals from waste waters by means of a permeable reactive barrier Groudeva V.I., Groudev S.N., Doycheva A.S. .................................................................. 697 The exploitation of sulphate-reducing bacteria for the reclamation of calcium sulphate sludges Luptakova A., Kusnierova M., Bezovska M., Fecko P. ..................................................... 703 The role of metal–organic complexes in the treatment of chromium containing effluents in biological reactors Remoudaki E., Hatzikioseyian A., Kaltsa F., Tsezos M. ................................................... 711 The selective precipitation of heavy metals by sulphate-reducing bacteria Luptakova A., Kusnierova M., Bezovska M., Fecko P. ..................................................... 719

APPENDIX Author index ................................................................................................................... A-3 Subject index ................................................................................................................. A-11

ix

Table of contents

PART II CHAPTER 3 BIOSORPTION A methodological approach to investigate the pH effect on biosorption process: experimental and modeling procedures Veglio F., Beolchini F., Pagnanelli F., Toro L. ............................................................... 731 A model for the copper biosorption in dried leaves de Carvalho R.P., De Sousa A-M.G., Freitas J.R., Rubinger C.P.L., Krambrock K. ...... 741 Agar-plate screening of effective metal biosorbents among year Podgorsky V.S., Lozovaya O.G., Kasatkina T.P., Fomina M.A. ...................................... 749 Bioremediation of chromium using Bacillus polymyxa Thyagarajan H., Subramanian S., Natarajan K.A. .......................................................... 759 Biosorption and bioaccumulation of heavy metals by bacteria isolated from contaminated sites of Karachi, Pakistan Nuzhat A., Uzma B., Fouad M. Qureshi, Fehmida F. ...................................................... 771 Biosorption equilibria with Spirogyra insignis Romera E., Fraguela P., Ballester A., Blazquez M.L., Munoz J.A., Gonzalez F. ............ 783 Biosorption of 226Ra and Ba by Sargassum sp. da Costa W.C., Garcia O. Jr., de Azevedo Gomes H. ...................................................... 793 Biosorption of arsenic and heavy metals on a ceramic-based biomass. Batch equilibrium experiments with Cu2+ model solutions Horak G., Willscher S., Werner P., Pompe W. ................................................................. 799 Biosorption of chromium (VI) by marine algal biomass Tan L.H., Chen J.P., Ting Y.P. ......................................................................................... 807 Biosorption of heavy metal ions from aqueous solutions by local seaweeds Sheng P.X., Chen J.P., Ting Y.P. ...................................................................................... 817 Biosorption of heavy metals onto an olive pomace: adsorbent characterisation and equilibrium modelling Pagnanelli F., Ubaldini S., Veglio F., Toro L. ................................................................. 825 Biosorption of Hg by vegetal biomasses Pimentel P.F., de Carvalho R.P., Santos M.H., Andrade M.C. ....................................... 835

x

Table of contents

Biosorption of lead in aquatic environment by Mucor rouxii biomass Som Majumdar S., Saha T., Bandhapadhyay T., Chatterjee S., Guha A.K. ..................... 843 Cadmium(II) biosorption by Aeromonas caviae: kinetic modeling Loukidou M.X., Karapantsios T.D., Zouboulis A.I., Matis K.A. ...................................... 849 Chromium uptake by pretreated cells of Aeromonas hydrophila isolated from textile effluents Zakaria Z.A., Ahmad W.A. ................................................................................................ 859 Copper ion adsorbed on chitosan beads: Physico-chemical characterization Chatterjee S., Som Majumdar S., Chatterjee B.P., Guha A.K. ......................................... 869 Development of a process for biosorptive removal of mercury from aqueous solutions Tupe S., Paknikar K. ......................................................................................................... 877 Effects of ionic strength, background electrolytes and heavy metals on the biosorption of hexavalent chromium by Ecklonia biomass Park D., Park J.M., Yun Y.-S. ........................................................................................... 883 Evaluation of silver recovery from photographic waste by Thiobacillus ferrooxidans and chitin Thiravetyan P., Nakbanpote W., Songkroah C. ................................................................ 891 Influence of the treatment of fungal biomass on sorption properties for lead and mercury uptake Spanelova M., Svecova L., Guibal E. ............................................................................... 899 Lanthanum and neodymium biosorption by different cellular systems Palmieri M., Garcia O. Jr. ............................................................................................... 911 Modeling of chromium biosorption by seaweed Sargassum sp. biomass in fixedbed column in series Cossich E.S., Silva E.A., Tavares C.R.G., Mesquita H.M., Eidan L.S. ............................ 919 Modelling and optimisation of copper ion uptake by Acidithiobacillus ferrooxidans Boyer A., Baillet F., Magnin J.-P., Ozil P. ....................................................................... 925 Platinum and palladium recovery from dilute acidic solutions using sulfate reducing bacteria and chitosan derivative materials Chassary P., de Vargas Parody I., Ruiz M., Macaskie L., Sastre A., Guibal E. .............. 935 Preliminary study of lead sorption by selected sorbents Ly Arrascue M., Bauer-Cuya J., Peirano Blondet F., Roussy J., Guibal E. .................... 947

xi

Table of contents

Regeneration of biomass after sorption of heavy metals Massacci P., Migliavacca E., Ferrini M. ......................................................................... 957 Structural modeling of arsenic biosorption using X-Ray spectroscopy (XAS) Teixeira M.C., Duarte G., Ciminelli V.S.T. ...................................................................... 965 Uranium and thorium removal by a Pseudomonas biomass: sorption equilibrium and mechanism of metal binding Sar P., Kazy S.K., D’Souza S. F. ...................................................................................... 975

CHAPTER 4 MICROBIOLOGY FUNDAMENTALS A model for iron uptake in Acidithiobacillus ferrooxidans based upon genome analysis Quatrini R., Veloso F., Jedlicki E., Holmes D.S. .............................................................. 989 Activity and occurrence of leaching bacteria in mine waste at Cartagena, Spain, in the years 1991 until 2000 Sand W., El Korchi-Buchert D., Rohwerder T. ................................................................ 997 An AFM-study on the adhesion of Acidithiobacillus ferrooxidans and Leptospirillum ferrooxidans to surfaces of pyrite Kinzler K., Sand W., Telegdi J., Kalman E. ................................................................... 1003 An X-ray photoelectron spectroscopy study of the mechanism of microbially assisted dissolution of chalcopyrite Parker A., Klauber C., Stott M., Watling H.R., Van Bronswijk W. ................................ 1011 Analysis of chalcopyrite (CuFeS2) electrodes utilizing galvanic current in the presence of Acidithiobacillus ferrooxidans Bevilaqua D., Benedetti A.V., Fugivara C.S., Garcia O. Jr. .......................................... 1023 Application of the bacterial weathering of silicate minerals in the improvement of quality of non-metallics Styriakova I., Styriak I. ................................................................................................... 1029 Assessment of acid production potential of sulphide minerals using Acidithiobacillus ferrooxidans and microbial sulphate reduction using Desulfotomaculum nigrificans Chockalingam E., Subramanian S., Natarajan K.A., Braun J.J. .................................... 1037 Comparative study on pit formation and interfacial chemistry induced by Leptospirillum and Acidothiobacillus ferrooxidans during FeS2 leaching Tributsch H., Rojas-Chapana J. ..................................................................................... 1047 xii

Table of contents

Composition of biofilm communities in acidic mine waters as revealed by combined cultivation and biomolecular approaches Kimura S., Coupland K., Hallberg K.B., Johnson D.B. ................................................. 1057 Computational fluid dynamics simulation of immobilized Acidothiobacillus ferrooxidans Metodiev B., Lilova L., Karamanev D. ........................................................................... 1067 Contribution to the quantification of the Acidithiobacillus ferrooxidans biomass concentration from the oxygen uptake rate Savic D.S., Veljkovic V.B., Lazic M.L. ............................................................................ 1077 Electrochemical and microbiological characterization of mercury in contact with mud Cruz F., Welzel A., Sampaio C., Englert G.E., Müller I.L. ............................................ 1085 Evaluating the growth of free and attached cells during the bioleaching of chalcopyrite with Sulfolobus metallicus Escobar B., Hevia M.J., Vargas T. ................................................................................. 1091 Experimental and modeling studies on inhibition effect of solution conditions on activity of Acidithiobacillus ferrooxidans during biooxidation of mixed sulphidic concentrates Chandraprabha M.N., Modak J.M., Natarajan K.A. ..................................................... 1099 Ferrous ion oxidation by an activated carbon cloth modified with Acidithiobacillus ferrooxidans de J. Cerino-Cordova F., Magnin J.P., Gondrexon N., Ozil P. ..................................... 1109 Heavy metal precipitation by off-gases from aerobic culture of Klebsiella pneumoniae M426 Essa A.M.M., Macaskie L.E., Brown N.L. ...................................................................... 1119 Influence of pH, Mg2+ and Mn2+ on the bioleaching of nickel laterite ore using the fungus Aspergillus niger O5 Coto O., Gutierrez D., Abin L., Marrero J., Bosecker K. .............................................. 1127 Mercury tolerance of thermophilic Bacillus sp. and Ureibacillus sp. Glendinning K.J., Brown N.L. ........................................................................................ 1137 Reduction of Pd(II) with Desulfovibrio fructosovorans, its [Fe]-only hydrogenase negative mutant and the activity of the obtained hybrid bioinorganic catalysts Mikheenko I.P., Baxter-Plant V.S., Rousset M., Dementin S., Adryanczyk-Perrier G., Macaskie L.E. ................................................................................................................. 1147 Removal of cobalt, strontium and caesium from aqueous solutions using native biofilm of Serratia sp. and biofilm pre-coated with hydrogen uranyl phosphate Paterson-Beedle M., Macaskie L.E. ............................................................................... 1155 xiii

Table of contents

Removal of soluble manganese from mine waters using a fixed bed column bioreactor Johnson D.B., Miller H., Ukermann S., Hallberg K.B. .................................................. 1163 Sulfane sulfur of persulfides is the actual substrate of the sulfur-oxidizing enzymes from Acidithiobacillus and Acidiphilium spp. Rohwerder T., Sand W. ................................................................................................... 1171 Sulfate reduction at low pH by mixed cultures of acidophilic bacteria Sen A.M., Kimura S., Hallberg K.B., Johnson D.B. ....................................................... 1179 Sulfur assimilation in Acidithiobacillus ferrooxidans Valdes J., Jedlicki E., Holmes D.S. ................................................................................ 1187 Survival of acidophilic bacteria under conditions of substrate depletion that occur during culture storage Johnson D.B., Bruhn D. F., Roberto F.F. ...................................................................... 1195 Synthesis of nanophase hydroxyapatite by Serratia sp. N14 Yong P., Sammons R.L., Marquis P.M., Lugg H., Macaskie L.E. .................................. 1205 The effect of maintenance on the ferrous-iron oxidation kinetics of Leptospirillum ferrooxidans Dempers C.J.N., Breed A.W., Hansford G.S. ................................................................. 1215 The kinetics of thermophilic ferrous-iron oxidation Searby G.E., Hansford G.S. ............................................................................................ 1227 The role of microorganisms in dispersion of thallium compounds in the environment Sklodowska A., Golan M., Matlakowska R. .................................................................... 1237

CHAPTER 5 MOLECULAR BIOLOGY AND TAXONOMY A promiscuous, broad-host range, IncQ-like plasmid isolated from an industrial strain of Acidithiobacillus caldus, its accessory DNA and potential to participate in the horizontal gene pool of biomining and other bacteria Goldschmidt G.K., Gardner M.N., van Zyl L.J., Deane S.M., Rawlings D.E. ............... 1249 Analysis of salt-induced outer membrane proteins in Acidithiobacillus ferrooxidans NASF-1 Kamimura K., Yamakado M., Shishikado T., Sugio T. ................................................... 1261

xiv

Table of contents

Bioinformatic analysis of biofilm formation in Acidithiobacillus ferrooxidans Barreto M., Rivas M., Holmes D.S., Jedlicki E. ............................................................. 1271 Diversity of Gram-negative bacteria at Malanjkhand copper mine, India Dave S.R., Tipre D.R. ..................................................................................................... 1279 Expression proteomics of Acidithiobacillus ferrooxidans grown in different metal sulfides: analysis of rhodanese-like proteins Ramirez P., Valenzuela L., Acosta M., Guiliani N., Jerez C.A. ..................................... 1287 Integration of metal-resistant determinants from the plasmid of an Acidocella strain into the chromosome of Escherichia coli DH5α Ghosh S., Mahapatra N.R., Nandi S., Banerjee P.C. ..................................................... 1297 Involvement of Fe2+-dependent mercury volatilization enzyme system in mercury resistance of Acidithiobacillus ferrooxidans strain MON-1 Sugio T., Fujii M., Takeuchi F., Negishi A., Maeda T., Kamimura K. ........................... 1305 Microbial diversity of various metal-sulphides bioleaching cultures grown under different operating conditions using 16S-rDNA analysis d’Hugues P., Battaglia-Brunet F., Clarens M., Morin D. .............................................. 1313 Molecular ecology of the Tinto River, an extreme acidic environment from the Iberian Prytic Belt González-Toril E., Llobet-Brossa E., Casamayor E.O., Amann R., Amils R. ................ 1325 Phenotypic characterization and copper induced stress resistance in the extremely acididophilic Archaeon Ferroplasma acidarmanus Baker-Austin C., Dopson M., Bowen A., Bond P. .......................................................... 1337 Pyrite oxidation by halotolerant acidophilic bacteria Norris P.R., Simmons S. ................................................................................................. 1347 Reversible loss of arsenopyrite oxidizing capabilities by Acidithiobacillus ferrooxidans is associated with swarming phenotype and presence of ISAfel Hurtado J.E. ................................................................................................................... 1353 Searching for physiological functions regulated by the quorum sensing autoinducer AI-1 promoted by afeI/afeR genes in Acidithiobacillus ferrooxidans Farah C., Banderas A., Jerez C.A., Guiliani N. ............................................................. 1361 Systematic analysis of our culture collection for "genospecies" of Acidithiobacillus ferrooxidans, Acidithiobacillus thiooxidans and Leptospirillum ferrooxidans Mitchell D., Harneit K., Meyer G., Sand W., Stackebrandt E. ....................................... 1369

xv

Table of contents

The strain genotypic heterogeneity of chemolithotrophic microorganisms Kondrateva T.F., Pivovarova T.A., Muntyan L.N., Ageeva S.N., Karavaiko G.I. .......... 1379

APPENDIX Author index ................................................................................................................... A-3 Subject index ................................................................................................................. A-11

xvi

Plenary Lecture


Biohydrometallurgy: a sustainable technology in evolution Giovanni Rossi Dipartimento di Geoingegneria e Tecnologie Ambientali – Università Piazza d’Armi, 19 – 09123 Cagliari, Italia Abstract In the mid to late 1990’s biohydrometallurgy used to be considered an innovative technology, but one can hardly continue to describe it as "innovative" today. Since the advent of biohydrometallurgical processing many major breakthroughs have been achieved and this economically profitable and sustainable technology now finds wide application in a variety of spheres, ranging from metal extraction to environmental remediation. Consequently, now that the pioneering days are over, what is required is a concerted effort by researchers to rationalize and optimize biohydrometallurgical processes. Three main areas of application can be identified: (i) environmental protection; (ii) metal extraction from minerals and rocks: (iii) pre-treatment of minerals to make them amenable to further processing. The fundamentals of biohydrometallurgy draw on a variety of disciplines, ranging from minerals engineering and mineralogy to microbiology, physical chemistry (with strong emphasis on surface science, colloid chemistry and electrochemistry) and solidstate physics. Researchers in all these fields have provided an equally important contribution to the development of this technology and this presentation endeavours to review their achievements and to briefly discuss those issues that remain open. Some of these issues continue to arouse controversy, stimulating the interest of scientists, which can also benefit other fields of science and technology. Based on his experience, the author would like to emphasize the need to establish a Biohydrometallurgical Society, to act as a point of reference to all those, from industry and academia, who are involved in implementing and further developing the technology. The Society should also provide a forum for information flow to decision makers in industry, about the potentials of biohydrometallurgy. Though it relies on the exploitation of the complex synergies between microoorganisms and minerals, this technology, when properly applied, is simple to implement, operationally stable and cost-effective. Finally, the author would like to invite academia to develop suitable curricula to ensure that new generations gain a specific working knowledge of biohydrometallurgy and industry to increase their funding for higher education and for private and academic research. 1.

INTRODUCTION In this review I will describe the evolution of a technology, biohydrometallurgy, which in my opinion offers promising prospects and can be highly rewarding for all those 3

Plenary Lecture

involved in providing humankind with the mineral resources indispensable for its progress with the minimum harm to the environment. Of course, I will focus most of my considerations on the engineering aspects of biohydrometallurgy and related physical-chemistry, solid state physics and earth sciences implications for two reasons with which I hope you will concur: first, in three recent excellent papers [1-3], all the most important aspects of microbiology applied to Biohydrometallurgy have been exhaustively reviewed in such a masterly way that there is really nothing significant to add at the present time; second, the degree of maturity attained by biohydrometallurgy as a new technology is such that its engineering developments and problems warrant attention. I would also like to point out that I have restricted the references to those necessary to justify some of my statements and that I have omitted many excellent contributions simply because otherwise this talk would have been more of a reference list than a presentation. In effect, on bioleaching kinetics alone I keep more than fifty papers in my files all of which deserve attention and mention in a paper concerning that topic. The Venn diagram shown in Figure 1 provides a visual representation of the interconnections among the various branches of science from which the fundamentals of biohydrometallurgy are derived. The engineering aspects are of paramount importance for commercial applications. In this regard three main groups of processes need to be distinguished: those concerned with metal extraction from rocks, those for mineral upgrading and environmental protection processes. For the sake of clarity these three groups will be examined separately.

Figure 1. Venn-Euler diagram showing biohydrometallurgy "parenthood"

4

Plenary Lecture

2.

ENVIRONMENTAL PROTECTION PROCESSES As is well known, the origins of biohydrometallurgy can be traced to the environmental problem created by the pollution of the Ohio River due to acid drainage of coalmines located in the river basin. Environmental protection continues to be one of the main areas of application of biohydrometallurgy and concerns either the development of remediation techniques aimed at low cost inhibition of dangerous effluents from old stopes or dumps or of systems for trapping toxic ions from effluents or for decontaminating polluted soils. Investigations into the properties of cell envelopes as adsorbents, of the so-called microbial derivatives [4,5], the proposed ingenious mathematical adsorption models, analysis of the factors influencing the processes, and the encouraging commercial applications [5] have opened new avenues and some promising research results have also been published recently [6,7] The interactions between microbes and solid surfaces play a very important role and three papers [8-10] on the fundamentals of this subject provide a useful background for those intending to advance in this field. The commercial applications rely on the knowledge of the mechanisms and the extent of adsorption of chemical elements or compounds by microorganisms. Providing the required information to the practitioners is the basic task of microbiology. 2.1 The biological fundamentals Organisms and microorganisms play the role of ion traps and, to some extent, can be considered the biological equivalents of inorganic exchange resins. This branch of biohydrometallurgy involves not only microorganisms but, more generally, all living things especially plants and algae. This distinctive feature already emerged at the time of what can be considered the First Symposium on Biohydrometallurgy, held in Braunschweig: on that occasion, the properties of the alga Hormidium fluitans (Gay) were described [11]. This is a broad field of research, as the ample literature published to date demonstrates. As pointed out by two recent reviews [5,12,13], the technology is promising but does not yet meet, for a number of reasons, the prerequisites for becoming a widespread cost-effective commercial application. For the time being the prospects of its application as a process in its own right seem to be limited. However, because of its characteristic feature of rapid intrinsic kinetics it is envisaged that biohydrometallurgy will be successfully integrated into water purification flowsheets consisting of hybrid technologies. For this type of technology the distinction between intra-biotechnological (IBT) or inter technological (IT) [12] depending on the type of associated processes can be helpful. The term IBT refers to biosorption, bioreduction or bioprecipitation, IT to biotechnology-based processes integrated with non-biotechnology based ones such as chemical precipitation, electrochemical processes, etc. 3.

BIOHYDROMETALLURGY AS A DEVELOPMENT OF EXTRACTIVE METALLURGY From its origins, in the 1940’s, up to the 1980’s, most biohydrometallurgical research work focussed on the ambitious target of developing an environmentally and costeffective process that could compete with pyrometallurgy-based processes for metal extraction from ores and concentrates.

5

Plenary Lecture

All of us old-timers certainly recall the enthusiasm with which we carried out our investigations and the encouraging results achieved by the first in-situ and dump leaching operations. Dare I say that almost all mining or metallurgical engineers involved in biohydrometallurgical research tested the process whenever an abandoned copper mine was available. The encouraging results obtained in several mining operations in the late 1970’s, that are still being confirmed today by the latest developments, like the Quebrada Blanca Mine [14] and the relatively low cost of the equipment required, certainly justify the impatience of several of us to start working on concentrate bioleaching. In effect, the results were quite frustrating. I still recall the disappointment after several months of very hard work in a well-equipped laboratory in Northern Italy. There, in 1977, in great secrecy, an Italian mining company entrusted an international team, of which I was part, with the task of developing a chalcopyrite flotation concentrate bioleaching flowsheet. We never succeeded in obtaining a 95% copper leaching in a single STR operated in batch. It was only two decades later that I began to understand the reasons for our failure, which depended simply on our ignorance of certain aspects of solid state physics typical of chalcopyrite. Subsequent research carried out in the light of the contributions of solid-state physicists and of new investigation methods like XPS, demonstrated the great potential of biohydrometallurgy. 3.1 Pre-treatment of run-of-mine ores or flotation concentrates and recovery of valuable metals The processes belonging to this group share the same type of problems and for this reason I will treat them together. The technology of this branch of biohydrometallurgy has already found successful application in cost-effective commercial operations. As summed up in the block diagram of Figure 2, their profitability depends on a series of interrelated factors whose investigation covers a wide range of basic sciences and of technological applications that had yet to be fully explored in the 1970’s. At this juncture I would like to stress the point that I purposely avoided the distinction between "pure" and "applied" sciences, as authoritatively stated more than a century ago by Pasteur [15]: ......No, a thousand times no, there does not exist a category of science to which one can give the name applied science. There are science and applications of science bound together as the fruit of the tree which bears it ..... The most encouraging commercial successes have been achieved in the pre-treatment of gold-bearing complex sulphide ores - notoriously refractory to conventional processing - for the subsequent cyanidation step. The excellent performance of numerous commercial plants is well documented [16,17]. Table 1 provides a summary of these achievements. It is no exaggeration to say that much of the recent research that has contributed to elucidating a great many problems posed by this technology has culminated in these successes. Research efforts directed to coal desulphurisation have produced some interesting results. A semi-commercial pilot plant, the first of its kind, jointly designed, built and operated in partnership between four European research groups in the framework of a project funded by the Commission of European Communities, demonstrated the practical

6

Table 1. Operating parameters of some commercial bioleaching operations Plant and location Fairview South Africa Sao Bento Brazil Olympia Greece

Ore minerals

Reactor type

% Solids concentration

Total useful bioreactor volume, m3

Daily throughput per bioreactor unit useful volume, tonn/m3.day

Residence time, hours

Reference

P, A

STR

20

90

0.444

96

[16]

A, P, Pr

STR

20

580

0.138

21

[16,17]

Complex Cu, Fe, As, Zn, Pb sulphides

STR

20

15,936 (3 moduli of 4 1,328 m3 each STR’s)

0.048

96

[16,18]

STR

20

23,376 (4 moduli of 6,974 m3 each STR’s)

0.047

96

[17]

P, A, Stb

STR

20

6x470 = 2,820

0.045

120

[17]

A, P, Pr, Mrc

STR

20

16,200 (3 moduli of 6 900 m3 each STR’s )

0.0444

96

[17,19]

Amantaytau Uzbechistan Wiluna Australia Ashanti Sansu Ghana

P = Pyrite; A = Arsenopyrite; Pr = Pyrrhotite; Mrc = Marcasite; Stb = Stibnite; C = complex Cu, Zn, Pb, As, Fe sulfides

Plenary Lecture

feasibility of coal biodepyritization. Removal of the so-called "organic sulphur" from coal, though investigated in depth by the same research groups, was not as successful, but did provide the guidelines for future research [20-22]. Solving this problem is of great environmental and economic significance, as billions of tons of fossil coal could be utilized were it possible to remove the organic sulphur therefrom [23].

Figure 2. Block diagram showing the factors affecting reactor bioleaching profitability 3.2 Base metal recovery from minerals As far as I am aware, no commercial bioleaching plant has ever been built for the extraction of base metals from mineral sulphides concentrates. One of the likely reasons for this is that the state of the art technology cannot yet compete with conventional pyrometallurgical processes. The problem here is obviously one of profitability. As Figure 2 shows, two main factors affect process profitability: bioleaching kinetics and reactor characteristics. Both factors are currently being researched, though bioleaching kinetics has received far more attention and will be dealt with first. The development of analytical expressions that mathematically relate the characteristics of: (i) microbial strains, (ii) solid substrate, (iii) process environment and (iv) reactor features, and their interactions with process kinetics has been and continues to be the main objective of bioleaching research. The composition of microbial populations and their synergies have been the almost exclusive hunting ground of microbiologists and can reasonably expected to be so in the future. A better understanding of the role played by the solid substrate requires the involvement of several fields of specialized nonmicrobiological knowledge, ranging from mineralogy associated to solid state physics, chemistry, electrochemistry and mineral engineering. Bioleaching is essentially the product of the the microbial population interacting with the solid substrate but the process 8

Plenary Lecture

is strongly conditioned by the physico-chemical environment, specifically the liquid phase pH and Eh, and consequently the related electrochemical phenomena, chemical composition, solids concentration, particle size distribution and evolution, and temperature. 3.2.1 Bioleaching microbiology Microorganisms can be considered, latu sensu, as the biocatalysts of mineral oxidation and solubilization processes. Up to now the following research lines have been pursued: (i) identification of the microorganisms involved in the process, (ii) how microorganisms interact among themselves and with the solid substrate, (iii) development of the most suitable microorganisms or microbial associations, (iv) enumeration of bioleaching microorganisms. Identification of the microorganisms involved in the bioleaching process that began with Temple and Hinkle’s discovery in 1948 still continues today. Major breakthroughs include the discovery of Leptospirillum ferrooxidans [24] and of the ability of Sulfolobus to bioleach metal sulphides [25]. These discoveries opened up a whole new world that continues to be investigated today. Incidentally, the discovery of L. ferrooxidans, may well be cited as yet another example of serendipity, because later investigations showed the organism to be quite atypical compared to those species that turned out to be so important in commercial operations [26]. The pioneering work on the physiology of Thiobacilli carried out by Kelly [citations in 27] the application of the theory of chemiosmotic mechanism to A. ferrooxidans proposed by Ingledew, Cox and Helling and by Ingledew [citations in 27] and the work by Don Kelly and Norris [citations in 27] on the moderate thermophiles and on the inhibition by ferrous and ferric ions have provided the tools for interpreting and the approaches for optimizing some laboratory and commercial performances. The results of investigations aimed at gaining a better understanding of the evolution of how bioleaching systems evolve in commercial operations carried out in CFSTR’s (work that appears to be continuing today) have provided a sound understanding of the roles of Thiobacilli and of Leptospirillum ferrooxidans [28-31] in metal sulphide bioleaching and of the importance of properly calibrating the physico-chemical environment on their oxidation activity. Undeniably, this information is invaluable to plant engineers. Identification methodologies involving nucleic acids represent the turning point for unequivocally distinguishing between strains of microorganisms as well as between species and genera and were exhaustively reviewed in a recent paper [32]. The confirmation of the wide diversity among Thiobacillus ferrooxidans strains [33], already observed by a number of researchers and reported two decades ago [34], as well as among the Leptospirillum ferrooxidans strains [35] are significant achievements of this methodology. Another group of methodologies are those based on immunological methods. This diversity is also of great practical significance. I still recall that about forty years ago, I was rather puzzled by the statement of a distinguished colleague as to the ubiquity of T. ferrooxidans. As matter of fact, I myself collected from numerous mines throughout Italy several strains of a microorganism that complied with the characteristics of T. ferrooxidans reported in the literature then available. However, the various strains responded quite differently under identical experimental conditions, a situation that seriously intrigued me, as at the time the differences in performance were inexplicable, and naturally I attributed it to some mistake in the laboratory procedure. The problem 9

Plenary Lecture

became dramatic when, a few years later, I was testing different strains for developing an industrial bioleaching process. As an engineer I would like to remark, however, that a major drawback of these methodologies is the need for fairly specialized equipment and skilled personnel [32]. I still recall the sophisticated van that housed a fine mobile laboratory where my Dutch colleagues very effectively carried out real time monitoring of the microorganisms used in the Porto Torres coal biodepyritization semi-commercial scale pilot plant [21]. It appears to me that this type of monitoring is very difficult to propose for a normal industrial operation. One potential area that warrants investigation is the development of kind of "diagnostic kit" that requires training accessible to technicians with a bachelor degree education background. The same considerations can be made for enumeration methodologies. In recent years microbial associations have been progressively identified and their importance recognized. Thus, it has been ascertained that several strains of heterotrophs contribute significantly to the activity of autotrophs f.i. acting as scavengers of their metabolites sometimes behaving as inhibitors when in sufficiently high concentrations [36-39]. Looking on the optimistic side, mutualism and synergism [37] should be considered important factors, for process optimisation also in continuous bioreactor systems, which so far have not received the attention they deserve. This is one way of remedying the other bad habit of many biohydrometallurgists – and I regret to say that I am one of them – of using enrichment cultures using substrates such as ferrous iron, sulphur and pyrite for isolating acidophiles. As vigorously pointed out recently [38], this unfortunately widespread practice may result in the selection of a relatively narrow range of acidophiles, depriving the cultures of the contribution of some important bacteria. The interactions between components of the microflora present in leaching operations were brought to light many years ago by Groudev [citations in 27]. In more recent times the existence and importance of this interaction seem to have been confirmed, in relation for instance to the role played by chemoorganotrophic microorganisms, such as Acidiphilium sp., [35] as a possible stimulating agent of Extracellular Polymeric Substance (EPS) excretion [40] by Leptospirillum ferrooxidans and as a kind of mineral surface scavenger/cleaner of the residual EPS after detachment of microorganisms [41] as well as the predatory activity of some protozoa on Thiobacilli. There is no doubt that the advances made, that started with some fundamental work by Costerton [42] and Characklis [43] on the microbial capsula and its importance in cell attachment to solid surfaces, fuelled the as yet unresolved controversy, that dates back more than thirty years [44,45] that has generated the fascinating and sometimes heated dispute between the supporters of direct [46] and those of indirect attack [47,40]. This issue has generated some very interesting and to a certain extent instructive critical reviews of the research work carried out on the topic. Probably the solution lies, as often happens, in a compromise and the papers mentioned above are the first premonitory signs. The mechanism by which microorganisms enhance metal sulphides oxidation and leaching are still surrounded by controversy. Consensus appears to have been momentarily reached on the current lack of evidence as to the existence of an enzyme that justifies the theory of the direct attack. The discovery of the extracellular polymeric substance (EPS) has given rise to some discrepancy in the interpretation of its function between the supporters of the purely electrochemical mechanism [47] and those who uphold a (surface-) chemical process where EPS is the localized environment cell/mineral surface for the action of an energy carrier (e.g. cysteine-based sulphur carrier) produced by some 10

Plenary Lecture

chemically-induced mechanism in the microorganism or for the local artificial increase in the concentration of an electron extracting agent (Fe3+) [46]. The peculiar properties of cysteine [48] appear, in this regard, extremely interesting and further research on this subject may be very rewarding. Consensus on the important role played by the EPS is of paramount importance to the process engineer because this provides vital information for the correct design of the bioreactor, as will be discussed later on. However, there still seems to be some disagreement as to the agents that generate the phenomena evolving within the EPS layer: Fe3+ concentration, hydrogen ion concentration, occurrence of thiol groups (e.g. cysteine). Experimental evidence based on redox potential measurements seems to favour the purely electrochemical mechanism [47]. However, further experiments are warranted in the light of recent contributions suggesting that redox potential at the surface of an electrode differs from the bulk of the solution in which it is immersed [49]. Similarly, further work is required to gain a better understanding of the ability of Thiobacillus ferrooxidans to excrete cysteine or other thiol group compounds. Hence, at least until the enzyme mediating the direct attack of the mineral by the microorganism is discovered, it seems quite appropriate to replace the term “direct attack” with the more cautious "direct contact" coined several decades ago in the very early days of Biohydrometallurgy [50]. Assessment of the significance of EPS is in my opinion a very important issue: the latest results obtained in research work being conducted by my group substantiate its crucial role in bioleaching. During the bioleaching of a gold-bearing pyrite/arsenopyrite concentrate in the new continuous bioreactor designed and constructed by our team, for which a patent is pending, the Biorotor [51] we came across a situation that we had never encountered when using batch bioreactors. The final leach liquor was observed to contain as much as 50 gram.dm-3 total iron, much higher than is consistent with its pH (0.9), with similar amounts of Fe3+, Fe2+. The intriguing point is that conventional analytical methods have proven unsatisfactory, owing to the fact that the iron seems to be encapsulated in something very similar to the recently reported glucuronic acid/iron ion complex containing 53 gram dm-3 [52,53], hence possibly the remnants of the EPS after pyrite corrosion and detachment of the microbial cell. This seems to prevent the precipitation of iron compounds that significantly impair bioleaching performance. Five of these substances’ properties appear to be quite remarkable, from the practical process engineering standpoint: (i) the fact that their chemical composition and surface activity are influenced by the substratum (ii) the fact that they form a particular, enlarged reaction space for the microbial cells, (iii) the ability of the cells to replenish their capsular material in a few hours when they loose it for any reason (for instance owing to mechanical action), (iv) EPS mediate the contact and (v) the microorganism looses its "catalyst" action if, for any reason, it is deprived of the EPS. This latter property appears to be significant [3] from the practical viewpoint: in the STR’s, at high suspension solids concentration, abrasion may seriously affect microbial action by tearing away the EPS from the cells. Two remedies are possible: either to modify the bioreactor design so as minimize abrasion and shear stresses or to design, employing genetic engineering methods, microorganisms resistant to abrasion and shear stresses. The Biorotor was designed in an attempt to pursue the first remedy. The opportunities offered by genetic engineering over the past twenty years, can be likened to medieval alchemists search for the philosophers’ stone with the obvious difference that genetic engineering is a very rigorous and well developed science. And like 11

Plenary Lecture

their search that contributed enormously to the development of modern chemistry, so too genetic engineering has provided a better understanding of the mechanisms underlying microbe-minerals interactions and enabled a far more accurate identification of microbial strains. Genetic engineering aims to develop microorganisms tailor-made for leaching, with fast kinetics and highly resistant to metal ions, the individual mineral. However, it needs to be said that the potential of "wild" or "indigenous" microorganisms has likely not yet been fully exploited through the development f.i. of suitable bioreactors. Genetic engineering may contribute, for instance, to suitably modifying the bacterial genomes of those microorganisms whose attachment to the mineral surface depends on the EPS, so that they are tailor engineered for the particular mineral. The observation, that as commercial biohydrometallurgical operations are not sterile, the risk of modified organisms being released into the environment may discourage or even prohibit the resort to such a technique is quite realistic [54]. A broad range of microbiological investigations has been conducted in the context of metal oxides bioleaching and the recovery of important elements from silicates as exhaustively discussed in a recent review paper by Ehrlich [2]. Attempts to use microorganisms for enhancing the solubilization kinetics of rocks, in particular of silicates, date back to the beginning of the last century, hence well before the discovery of Acidithiobacillus ferrooxidans. The solubilization of leucite, a potassium and aluminium silicate, to extract potassium and aluminium at low cost, was attempted in 1906 by De Grazia and Camiola [55] using molds. While the interesting properties of Acidithiobacillus ferrooxidans, were being investigated, "silicate" solubilizing bacteria were claimed to have been isolated from agricultural soils [56]. It seems that the silicate bacteria claimed by Aleksandrof and his school are in effect strains of Bacillus circulans, [57,58] but research in this area probably still has a long way to go before commercial applications can be contemplated. Some controversy still surrounds this discovery and compared to the advances made with Thiobacilli little progress has been reported in this field. Over the past fifty years no major advances have been achieved in research on the use of molds or, at least, they do not offer any promising commercial prospects. One of the major drawbacks of this technology is the dramatic volume of biomass involved and the practical problems posed by its handling. Some interesting proposals for applications of silicate solubilizing bacteria concerned the beneficiation of bauxites [59], but again no industrial applications have been reported. An interesting niche is represented by the investigations into the microbial recovery of manganese from manganese oxide [60-62]. Research carried out in this field is unravelling the mechanisms of microbial oxides reduction, and the results obtained so far seem quite encouraging. This area warrants further investigation as it offers interesting commercial prospects, for instance for recovering manganese from ocean nodules. 3.2.2 Microbe/solid matter interactions As far as solid matter is concerned, a distinction needs be made between rocks (minerals) and solid matter other than rocks. The latter is concerned more with environmental applications and has been dealt with above. Mineral bioleaching is a subject that embraces several topics. As sulphide minerals oxidation is a physico-chemical process, a better understanding of how the presence of microorganisms enhances electron transfer from the mineral to the end acceptor is crucial for optimising any bioleaching operation. 12

Plenary Lecture

The amenability of a given mineral to oxidation is related to the solid-state physics of both its bulk and its surface. A number of recent contributions have provided an insight into the importance of the semiconductor nature of the minerals involved and the modifications produced by the presence of foreign elements in the crystal lattices [22,6366]. Molecular orbital theory, valence bond theory and mineralogy may provide a decisive contribution to elucidating the solubility of metal sulphides in bioleaching systems. Thus, based on their solid state physics, Sand et al. [40] were able to classify metal sulphides into two main groups: the first, consisting of pyrite, molybdenite and tungstenite, and the second of sphalerite, galena, chalcopyrite, hauerite, orpiment and realgar, exhibiting strong differences in solubility in acid. One typical case is sphalerite: mining engineers and metallurgists are well aware of the fact that quite often sphalerite is not a stoichiometric compound of zinc and sulphur but contains dissolved iron in proportions that usually differ from one orebody to another. These varieties are called "marmatites" and have been widely investigated [67]. The importance of the proportion of iron dissolved in the crystal lattice of sphalerite had already been recognized in the mid 1950’s by flotation engineers. Now, in the light of considerations that have emerged in recent years [64], the importance of iron has been fully elucidated. The plot reported by Crundwell [68] demonstrating that the rate of reaction for the oxidative dissolution of marmatites is a linear function of the iron concentration in the zinc sulphide lattice, shows dramatically how iron affects the electrochemical, and hence the leaching behaviour, of marmatites. Similar considerations were made for pyrite [22]. Since the elements dissolved in the lattice are kinds of fingerprints of the mineral, it is clear that its origin warrants special attention and may well explain the differences in leaching behaviour. Mechanisms based on the valence bond theory have recently been proposed [46,64] for the oxidation of pyrite sulphur moiety by iron(III) ions and, though requiring further refinements, they are worthy of mention in that they are on the right track to solving the problem. Further research will likely provide the information necessary for process modulation. The surface structure and composition of minerals and their modifications during the bioleaching process as well as a kind of "activation" by some metals, like silver, have been recognized in some instances as important factors for the evolution of the process. Chalcopyrite is, in this respect, quite typical since the acidic ferric sulphate leaching or bioleaching evolve according to a well-documented pattern [27,69]. Copper dissolution proceeds in two phases: the initial phase is characterized by a relatively high dissolution rate, which can be expressed by a parabolic law. Over time, the rate decreases pronouncedly, the dissolution rate being represented by a gently sloping straight line. The dissolution rate of the first phase can only be restored by regrinding the leached residue, which usually contains no less than 50% of the copper in the initial feed. Among the explanations advanced for this behaviour, the one based on electroanalytical observations [70] seems to have been confirmed by XPS investigation of surface layers of bioleached chalcopyrite [71]. According to this model, during the first phase, a diffusion layer of copper-depleted chalcopyrite is formed on the mineral surface and it is this layer that governs the subsequent dissolution rate. The presence of silver in the reacting suspension has a catalytic effect [72] that strongly enhances copper solubilization and is explained by the formation of conductive

13

Plenary Lecture

compounds in the retarding barrier, which neutralize its passivating effect by acting as channels for electron flow. This is just one of many significant, still debated examples, of the interactions of a multiplicity of physico-chemical, electrochemical and solid-state physics factors. The intriguing micrographs showing the preferential adhesion of microbial cells to areas where emergence of dislocations is visible [40, 44, 73] seem to justify further and more sophisticated research on this subject. The resort to instruments that, like XPS, provide information on the chemical bonding between elements in the surface layers, may unveil the importance of increasing the surface area-to-bulk volume ratio and the extent to which ions in solution interact with the mineral surface and affect oxidation and solubilization. The importance of chemical composition is also significant, as sometimes the elements passing into solution are toxic to the microorganism, thus requiring continuous adaptation. The inhibitory action of these elements may affect the growth kinetics of the microbial populations and this again relates to the physiology of microorganisms. 3.2.3 Bioleaching kinetics In the light of the above considerations it is now possible to briefly discuss what may be considered one of the key issues in biohydrometallurgy: the bioleaching kinetics from a process engineering standpoint. As this audience is well aware, numerous researchers have explored this topic, most of whom are chemical engineers or hydrometallurgists, and at a rough estimate more than 50 papers have been published on the subject over the past sixty years. Further research is expected to provide useful data for modulation of the process. The dramatic increase in mineral sulphides oxidation and solubilization kinetics produced by the pseudocatalytic mediation of Thiobacillus sp. organisms has roused great interest in practitioners and researchers in extractive metallurgy since the very beginnings of biohydrometallurgy. In effect, it was just around this time that a greater awareness of environmental issues was developing, which was soon to place increasing constraints on new industrial projects. Up until then extractive metallurgy had been dominated by pyrometallurgical processes, which had a negative environmental impact. Clearly the development of mathematical models capable of predicting the economic performance of the new processes became a pressing need. As shown in Figure 2, the performance of a bioleaching process depends on a multiplicity of factors and, as pointed out in recent years [27,74], its kinetics are a combination of microbial growth kinetics, microbe-minerals and chemical compounds interaction kinetics and particulate solids oxidation and dissolution kinetics (a process that involves several branches of science and technology). The early models, which already endeavoured to develop a general theory of bioleaching, were relatively simple [69,75-77] but deserve mention in recognition of the originality and the initiative of those researchers. Developments in subsequent years are well documented in several excellent reviews [74,78,79] and numerous of proposed models (fifteen) are exhaustively summarized in a recent review paper [80]. The above-mentioned classification of metal sulphides proposed by Sand et al. [40] and the continuing speculation surrounding the importance of the glycocalyx in the biosolubilization process, would appear to definitively preclude the possibility of developing a general kinetic theory that embraces the entire field of bioleaching. At this point in time it seems expedient to revisit the existing models in the light of the advances in the fundamentals: a deeper insight into some of these may well lead to their unification. 14

Plenary Lecture

There seems to be general consensus that bioleaching kinetics is the result of the combination of several subprocesses, the most important being ferrous iron oxidation, bacterial growth (planktonic and sessile), and mineral sulphide dissolution reaction. The kinetic models The four recently published reviews mentioned above, propose a classification of the kinetic models. Three of them concern ferrous iron oxidation and mineral bioleaching, the fourth ferrous iron oxidation. Crundwell [81], proposes a classification criterion based on the relative importance of what are unanimously defined as "contact leaching" and "indirect leaching" [46] mechanisms and refers to two categories: (i) models that postulate a well defined mode of evolution of the leaching process (shrinking-core, propagating-pore) and derive bacterial growth; and (ii) models that postulate a well defined bacterial growth law (rate of growth related to attachment to- and detachment from the solid substrate, Monod law, the logistic equation) and derive the evolution of the leaching process. Haddadin et al. [80] based their classification (concerning both ferrous iron oxidation kinetics and mineral bioleaching) on the mass balance of each reactional system and on the underlying assumptions of each model and identified three categories: (i) well agitated reactors operating on a liquid phase; (ii) well agitated reactors operating on liquid and solid phases; (iii) the so-called "biofilm reactors". Hansford [74] considers three classes: (i) empirical models, based on the logistic equation; (ii) models based on attachment of microorganisms; (iii) the two subprocess mechanism (bacterial ferrous iron oxidation and chemical ferric leaching of the sulphide mineral, assumed to be an electrochemical process for which the Volmer-Butler equation holds) for pyrite bioleaching. The implications of this mechanism, as reported by Hansford [74] are rather interesting, insofar as it seems to be able to explain the findings that L. ferroooxidans is the dominant bacterial species in the bioleaching of arsenopyrite and pyrite [28,29] and that arsenopyrite is preferentially leached ahead of pyrite [82]. However, none of these bioleaching kinetics models take into account the population balance approach for describing in quantitative terms, by means of suitable distribution functions, the influence of material properties distribution on the overall behaviour of biohydrometallurgical systems, exhaustively described by extractive metallurgists in the late 1970’s [83,84]. The significance of the material properties appear to have been first understood by researchers investigating coal biodepyritization in relation to the size distribution of pyritic coal [85] but the population balance approach was not adopted. Crundwell [81] deserves recognition for his model based on the population balance for particulate leaching, for the bacterial cell number and material balance describing solution reactants and products. This model was found to provide an excellent fit to experimental data reported by other researchers for pyrite bioleaching, although he used the shrinking-core rather than the more realistic propagating-pore mechanism [74]. It should however be pointed out that the population balance approach can also have pitfalls, associated for instance with the practical difficulty, of which mineral processing practitioners are well aware, of developing a realistic function for the grain size distribution of a mineral powder. These difficulties can increase the complexity of such a model making its application impracticable. The kinetics of ferrous iron oxidation has long been recognized as playing a key role in bioleaching and several papers have been published on this subject over the past sixty 15

Plenary Lecture

years. I will not review every single paper here but will limit myself to a few general remarks, spotlighting the main features of this history, as they provide a good description of the evolution of biohydrometallurgy. Over the years the significance of the following factors in bioleaching processes listed in chronological order - has been recognized: Total initial iron concentration Temperature Type of culture (continuous or batch) Wall growth Total iron concentration Inhibition by ferric iron Threshold concentration of ferrous iron Product inhibition pH Bacterial concentration (mass or numeric) Bacterial decay Maintenance Carbon dioxide transfer Nemati et al. [86] reviewed the work carried out on the kinetics of ferrous iron oxidation by Thiobacillus ferrooxidans distinguishing two major classes: (i) freely suspended cells; (ii) immobilized cells. Whereas, for the immobilized cells processes, these authors compared the performances of different types of reactors (packed-bed, fluidised-bed, rotating biological contactors), in the paper by Pesic et al. [87] the experimental set-up consisted of a thermostated 300 cm3 plexiglass cell containing 250 cm3 of ferrous solution and Thiobacillus ferrooxidans inoculum and provided with a magnetic stirrer: no information was provided on the rpm of this stirrer. In another paper [88] ferrous conversion kinetics was experimentally determined using a set-up consisting of a small, thermostated, reaction cell (volume 20 cc) containing 10 cc of reaction liquid plus cell suspension, sparging air through the liquid contained in the cell at a flow rate of 8,3 cm3.s-1. This set-up can be regarded as an "ideal" reactor, where the bacterial cells are very likely only subjected to a minimum of mechanical stress. It is possible that experimental conditions in both cases ensured a fairly quiescent environment thus avoiding undue stress to the bacterial cells. These operating conditions differ significantly from those existing in the particulate mineral suspension of an STR fitted, for instance, with a Rushton-type turbine. Only recently, however, a mechanistically-based model adopting the initial rate method and relying on Michaelis-Menten kinetics [79,88] and a model based on Ingledew’s chemiosmotic theory, on the electrochemical theory and on the Monod and Michaelis- Menten models, take into account all of the above influencing factors and appear to provide a satisfactory fit to the bacterial ferrous iron oxidation data reported in the literature [74]. This is undoubtedly an important achievement, though the data fit refers to experiments carried out in a variety of bioreactors where, in terms of reactor dynamics, optimum operating conditions may well not have been ideal for bacterial cells. So, all the models for ferrous iron bioleaching kinetics or for mineral bioleaching kinetics appear to neglect the influence of reactor dynamics, which, as far as I am aware, was only mentioned [80] in relation to a paper [89] dealing with the geometry and operating characteristics of an air–lift bioreactor (Pachuca tank) designed by researchers of Bergbauforschung and still being used in my laboratory.

16

Plenary Lecture

Practically all the authors who have provided an overview of bioleaching kinetics models agree as to the difficulty of generalizing the results of the individual models, on account of the numerous influencing factors that characterize each substrate, such as mineralogy, solid-state physics, chemical composition, particle size, specific gravity, electrochemistry, and the microbial population (temperature, pH, strains, associated microbial strains…). Moreover, several authors have demonstrated that strong shear stresses are generated in conventional bioreactors (STR’s and ALR’s) during agitation of the suspension and that, beyond a certain threshold, abrasion produced by solid particles has to be taken into account. Such situations may become a limiting and determining factor seriously affecting both bacterial growth and bacterial adhesion and ultimately process kinetics. Quantification of these effects is an open and very likely promising research field, although I still doubt that general theories can be formulated. Probably, the most practicable and profitable way is to design suitable bioreactors where these effects are minimized by suitably selecting the proper operating conditions. There is general consensus that bioleaching process kinetics is directly dependent upon the number of active microbial cells present in the system. Already about twenty years ago this number had been estimated to be in the order of magnitude of 1012 cells.cm-3 [85,90,91] which, as far as I know, has not yet been achieved. It is therefore reasonable to attempt to maximize microbial growth. The question then arises as to what is the maximum number of microorganisms that can be achieved and if the bioreactors currently in use are the most suitable for attaining this population density in steady state regime and how such a reactor should be designed. The problems are very similar to those confronted by mineral beneficiation researchers and flotation plant designers and operators for about 100 years, from the very early days of flotation technology. Research has played a very important role in elucidating the problems, but no general theory could be developed. Basically, the flotation process is the same for any ore, but each ore requires equipment, reagents, and physico-chemical environment to be properly adjusted. Process performance can only be predicted by a wise combination of fundamentals (that, except for the biological agent, are practically the same as in biohydrometallurgy) and of bench- and laboratory scale pilot testing using devices that simulate commercial operation, that provide the experimental numeric parameters for the mathematical expressions for describing the specific process kinetics - supplemented by an up-to-date knowledge of microbiology. The future progress of Biohydrometallurgy much depends on the solutions of these problems: the fascinating aspect but also the intriguing feature of biohydrometallurgy is its multidisciplinarity. It is fascinating because it reveals how intimate the liaison between the various branches of science and technology can be; intriguing, because this technology requires a working knowledge of several disciplines, that are sometimes so different from one another and that cannot be reasonably mastered by a single specialist. Real advances will only be made possible by the cooperation of people skilled in the various facets of biohydrometallurgy. The timely exchange of information and a very complete documentation are the prerequisites for gaining an identity that this new technology seems to have not yet achieved. The importance of up-to-date documentation will prevent researchers from expensive and time-wasting repetition of investigations successfully carried out elsewhere. What I am going to say now may sound trivial, but it

17

Plenary Lecture

should be remembered that a well-documented failure is as useful as a successful achievement. 4.

A BIOHYDROMETALLURGY SOCIETY Based on his experience, the author would like to emphasize the need to establish a Biohydrometallurgical Society, to act as a point of reference to all those, from industry and academia, who are involved in implementing and further developing the technology. The Society should also provide a forum for information flow to decision makers in industry, about the potentials of biohydrometallurgy. Though it relies on the exploitation of the complex synergies between microoorganisms and minerals, this technology, when properly applied, is simple to implement, operationally stable and cost-effective. One task of the Society might be, for instance, the preparation and further updating of a Recommended Standard Terminology and Nomenclature for Biohydrometallurgy, like the one published several years ago by the Institution of Chemical Engineers Fluid Mixing Group [92]. This idea came to mind when I was reading a comprehensive review written by a distinguished microbiologist, who proposed an important terminology, justifying his intervention with the fact that the related topic had been studied by researchers from diverse cultural backgrounds other than microbiology. Last, but not least, the Society would be able to benefit from the experience and knowledge of all those involved in industry and academia to develop new university curricula aimed at training a new generation of specialists. REFERENCES 1. 2. 3. 4.

Ehrlich, Hydrometallurgy, 59 (2-3) (2001) 127. Ehrlich, Minerals and Metallurgical Processing, 19 (4) (2002) 220. Rawlings, D. E., Ann. Rev. Microbiol., 56 (2002) 65. Brierley, J.A., Brierley, C.L., and Goyak, G.M., Fundamental and Applied Biohydrometallurgy, Lawrence, R.W., Branion, R.M.R. and Ebner, H.G. (Eds.), Elsevier, Amsterdam (1986) 291. 5. Tsezos, M., Microbial Mineral Recovery, Ehrlich, H. L. and Brierley, C. L., (Eds.), McGraw-Hill Publishing Company, New York, (14) (1990) 325. 6. Pümpel, T., Ebner, C., Pernfuß, B., Schinner, F., Diels, L., Keszthelyi, Z., Stankovic, A., Finlay, J. A., Macaskie, L. E., Tsezos, M., and Wouters, H., Hydrometallurgy, 59 (2-3) (2001) 383. 7. Hatzikioseyian, A., Tsezos, M., and Mavituna, F., Hydrometallurgy, 59 (2-3) (2001) 395. 8. van Loosdrecht, M.C.M. and Zehnder, A.J.B., Experientia, 46 (1990) 817. 9. van Loosdrecht, M.C.M., Norde, W., Lyklema, J. and Zehnder, A.J.B., Aquatic Sciences, 52 (1) (1990) 103. 10. van Loosdrecht, M.C.M., Lyklema, J., Norde, W., Gosse, S. and Zehnder, A.J.B., Appl. Environm. Microbiology, 53 (8) (1987) 1893. 11. Madgwick, J.C., and Ralph, B.J., Conference Bioleaching, Schwartz, W. (Ed.) Weinheim (Germany) Verlag Chemie, (1977) 85. 12. Tsezos, M., Hydrometallurgy, 59 (2-3) (2001) 241. 13. Volesky, B., Hydrometallurgy, 59 (2-3) (2001) 203. 14. Schnell, H.A., Biomining: Theory, Microbes and Industrial Processe, Rawlings, D. E. (Ed.), Springer, Berlin, (11) (1997) 21. 15. Pasteur, L., La révue scientifique, 2e Sér. 1 (4) (1871) 73. 18

Plenary Lecture

16. van Answegen, P.C. and Marais, H.J., MINERAL BIOTECHNOLOGY, Kawatra, S.K. and Natarajan, K.A. (Eds.), Littleton, Colorado, USA Society for Mining, Metallurgy, and Exploration, Inc., (2001) 121. 17. Dew, D.W., Lawson, E.N., and Broadhurst, J.L., Biomining, Rawlings, D.E. (Ed.), Berlin, Springer, (1997) 45. 18. Taxiarchou, M., Adam, K., and Kontopoulos, A. Hydrometallurgy, 36 (2) 1994 169. 19. Nicholson, H.M., Smith, G.R., Stewart, R.J., Kock, F.W., and Marais, H.J., Biomine ’94, Australian Mineral Foundation: Glenside, South Australia, 1994 2-1. 20. Loi, G., Mura, A., Trois, P., and Rossi, G., Memorie dell'Associazione Mineraria Sarda – Iglesias, I (1) (1994a) 29. 21. Loi, G., Mura, A., Trois, P., and Rossi, G., Fuel Processing Technology, 40 (1994b) 26. 22. Rossi, G., Fuel, 72 (12) (1993) 1581. 23. International Scientific Committee of the Fourth Intrntl.Symp. on Biological Processing of Fossil Fuels, Rossi, G. (Guest Ed.), Coal Processing Technology, 40 (1994), 379. 24. Markosyan, G.E., Biol. Zh. Armenii, (1972) 25. 25. Brierley, C.L., and Murr, L.E., Science, 179 (1973) 488. 26. Rawlings, D.E., Biomining: Theory, Microbes and Industrial Processe, Rawlings, D. E. (Ed.), Springer Berlin, (1997) 229. 27. Rossi, G., Biohydrometallurgy, McGraw-Hill GmbH, Hamburg, (1990). 28. Rawlings, D. E., Tributsch, H., and Hansford, G. S., Microbiology, 145 (1999) 5. 29. Boon, M., Brasser, H. J., Hansford, G. S., and Heijnen, J. J., Hydrometallurgy, 53 (1) (1999) 57. 30. Breed, A. W., Dempers, C. J. N., Searby, G. E., Gardner, M. N., Rawlings, D. E., and Hansford, G. S., Biotechnology and Bioengineering, 65 (1) (1999) 44. 31. Battaglia-Brunet, F., d'Hugues, P., Cabral, T., Cezac, P., Garcia, J. L., and Morin, D. 1998, Minerals Engineering, 11 (2) (1998) 195. 32. Jerez, C. A. 1997, Biomining: Theory, Microbes and Industrial Processes. Rawlings, D. E. (Ed.), Springer, Berlin (1997) 281. 33. Kelly, D. P. and Wood, A. P., International Journal of Systematic and Evolutionary Microbiology 50 (2000) 511. 34. Harrison, A. P. J., Archives of Microbiology, 131 (1982) 68. 35. Hallmann, R., Friedrich, A., Koops, H., Pommering-Röser, A., Rohde, K., Zenneck, C., and Sand, W., Geomicrobiology Journal, 10 (1992) 193. 36. Johnson, B. and Roberto, F. F., Biomining: Theory, Microbes and Industrial Processe, Rawlings, D. E. (Ed.), Springer Berlin, (13) (1997) 260. 37. Johnson, D. B., FEMS Microbiol. Ecol., 35 (1998) 307. 38. Johnson, D.B., Hydrometallurgy, 59 (2-3) (2001) 147. 39. Johnson, B., Bacelar, Nicolau P., Okibe, N., Yahya, A., and Hallberg, K.B., Biohydrometallurgy - “Fundamentals, Technology and Sustainable Development”, Proceedings of the International Biohydrometallurgy Symposium IBS-2001, Part A, Ciminelli, V. S. T. and Garcia Jr., O. (Eds.), vol. A, Elsevier Amsterdam, (2001) 461. 40. Sand, W., Gehrke, T., Jozsa, P.-G. and Schippers, A., Hydrometallurgy, 59 (2-3) (2001) 159. 41. Sand W, Jozsa P-G, and Schippers A., Biohydrometallurgy and the Environment toward the Mining of the 21st Century, Amsterdam, Elsevier, Vol. A (1999) 27. 42. Costerton, J. W., Irvin, R. T., and Chen, K.-J., Ann Rev Microbiol, 35 (1981) 299. 43. Characklis, W. G., Biotechnology and Bioengineering, 23 (1981) 1923. 44. Berry, V. K. and Murr, L. E., Metallurgical Transactions, B, 6B, (1975) 488. 19

Plenary Lecture

45. Bennett, C. and Tributsch, H., J. Bacteriology, 134 (1) (1978) 310. 46. Tributsch, H., Hydrometallurgy, 59 (2-3) (2001) 177. 47. Crundwell, F.K., Biohydrometallurgy - “Fundamentals, Technology and Sustainable Development”, Proceedings of the International Biohydrometallurgy Symposium IBS2001, Part A, Ciminelli, V.S.T. and Garcia Jr., O. (Eds.), Elsevier, Amsterdam (2001) 149. 48. Rojas-Chapana, J. A., and Tributsch, H., 35 (2000) 815. 49. Nicol, M. J. and Lázaro, I., Hydrometallurgy, 63 (1) (2002) 15. 50. Silverman, M.P., and Ehrlich, H.L., Advan. Appl. Microbiol., 6 (1964) 153. 51. Loi, G., Trois, P., and Rossi, G., Biohydrometallurgical Processing, Vargas, T., Jedrez, C.A., Wiertz, J.V. and Toledo, H. (Eds.), University of Santiago, Chile, Vol. 1, (1995) 263. 52. Gehrke, T., Hallmann, R., Kinzler, K., and Sand, W., Appl. Environ. Microbiol., 65 (1997) 159. 53. Gehrke, T., Telegdi, J., Thierry, D., and Sand, W., Appl. Environ. Microbiol., 64, 7 (1998) 2743. 54. Rawlings, D.E., Hydrometallurgy, (2-3) 59 (2001) 187. 55. De Grazia, S. and Camiola, G., Le stazioni sperimentali agrarie italiane, 39 (9) (1906) 829. 56. Aleksandrov, V.G., Kiev Izv. Akad. Nauk. S. S. S. R., (1958) 62. 57. Tesic, Z.P. and Todorovic, M.S., Zemljiste i biljka, 1 (1) (1952) 3. 58. Tesic, Z.P. and Todorovic, M.S., Zemljiste i biljka, 8 (1-3) (1958) 233. 59. Groudeva, V.I. and Groudev, S.N., 5th International Congress of ICSOBA, Zagreb, Yugoslavia, Academie Yugoslave des Sciences et des Arts, 83 (18) 257. 60. Remezov, V. D., Agate, A. D., and Yurchenko, V. A., Biogeotechnology of Metals – Manual, Karavaiko, G. I., Rossi, G., Agate, A. D., Groudev, S. N., and Avakyan, Z. A. (Eds.), Centre for International Projects – GKNT Moscow U.S.S.R., (1988) 304. 61. Ehrlich, H. L. (and Rossi, G.), Microbial Mineral Recovery, Ehrlich, H. L., and Brierley, C. L. (Eds.), (7), MacGraw-Hill Publishing Company, New York, (1990) 149. 62. Ehrlich HL., Biohydrometallurgical Technologies, Torma, A. E., Apel, M. L., and Brierley, C. L. (Eds.), v. 2, The Minerals, Metals & Materials Society, Warrendale, Penna, USA, (1993) 415. 63. Pridmore, D.F., and Shuey, R.T., American Mineralogist, 61 (1976) 248. 64. Crundwell, F.K., AIChE Journal, 34(7) (1988) 1128. 65. Nesterovich, L. G., Biogeotechnology of Metals - Manual. Karavaiko, G. I., Rossi, G., Agate, A. D., Groudev, S. N., and Avakyan, Z. A. (Eds.), Centre for International Projects GKNT, Moscow, U.S.S.R., (1988) 101. 66. Thomas, B., Ellmer, K., Bohne, W., Rohrich, J., Kunst, M., Tributsch, H., Solid State Communications, 111 (1999) 235-240. 67. Kullerud, G., Norsk Geologisk Tidsskrift, 32 (1953) 61. 68. Crundwell, F.K., Hydrometallurgy, 21(2) (1988) 155. 69. McElroy, R.O. and Bruynesteyn, A., Metallurgical Applications of Bacterial Leaching and Related Microbiological Phenomena, Murr.L.A., Torma, A. E., and Brierley, J. A. (Eds.) Academic Press, New York (1978) 441. 70. Ammou-Chokroum, M., Cambazoglu, M. and Steinmetz, D., Bull. Soc. Fr. Miner. Cristallogr. 100 (1977) 161. 71. De Filippo, D., Rossi, A., Rossi, G., and Trois, P., International Biotechnology Symposium Proceedings, Durand, G., Bobichon, G., and Florent, J. (Eds.), J. Soc. Française de Microbiologie, Paris, France (1988) 1131. 20

Plenary Lecture

72. Miller, J.D., and Portillo, H.Q., Proceedings XIII Int. Min. Proc. Cong., Vol. 2 (A), Laskowski, J. (Ed.), Elsevier Scientific Publishing Comp., (1979) 851. 73. Rojas-Chapana, J.A., and Tributsch, H., Biohydrometallurgy and the Environment toward the Mining of the 21st Century, Amsterdam, Elsevier, Vol. 2 (1999) 597. 74. Hansford, G. S., Biomining. Rawlings, D. E., Springer: Berlin, (8) (1997) 153. 75. Torma, A. E., CANMET, Ottawa, Canada (1985) 5. 76. Torma, A. E. and Panneton, J. J., Ministère des Richesses Naturelles - Direction Générale des Mines-Centre de Recherches Minérales, Québec, Canada BLS-1 (1973). 77. Legault, G., and Torma, A.E., 421ème Congrès Annuel, Association Canadienne Française pour l'Avancement des Sciences, Province de Québec (1974) 1. 78. Barrett, J., Hughes, M. N., Karavaiko, G. I., and Spencer, P. A., Metal Extraction by Bacterial Oxidation of Minerals. Ellis Horwood Limited, NewYork. (1993) 103. 79. Nemati, M, and Webb, C., Biotechnology Letters, 20 (9) (1998) 873. 80. Haddadin, J., Dagot, C., and Fick, M., Enzime and Microbial Technology, 17 (1995) 290. 81. Crundwell, F.K., The Chemical Engineering Journal, 54 (1994) 207. 82. Miller, D. M. and Hansford, G. S., Minerals Engineering 5 (7) (1992) 737. 83. Sepulveda, J. E. and Herbst, J. A., AIchE Symposium Series, 74 (173) (1978) 41. 84. Herbst, J. A., Rate Processes of Extractive Metallurgy. Sohn, H. Y. and Wadsworth, M. E. (Eds.) Plenum Press New York, (2) (1979) 53. 85. Andrews G.F., Bioprocessing of Coal Workshop III, Idaho National Engineering Laboratory, Idaho Falls, ID, USA (1988) 234. 86. Nemati, M., Harrison, S.T.L., Hansford, G.S. and Webb, C., Biochemical Engineering Journal, 1 (1998) 171. 87. Pesic, B., Oliver, D.J., and Wichlacz, P., Biotechnology and Bioengineering, 33 (1989) 428. 88. Nemati, M. and Webb, C., Biotechnology and Bioengineering, 53, (5) (1997) 478. 89. Beyer, M., Ebner, H.G., and Klein, J., Appl. Microbiol. Biotechnol., 24 (1986) 342. 90. Andrews G.F. and Quintana J., First International Symposium on Biological Processing of coal, E.P.R.I. Palo Alto, California, U.S.A., (1990) 5-69. 91. Stevens, C. J., Noah, K. S., and Andrews, G. F., Fuel, 72 (12) (1993) 1601. 92. Anonymous, The Chemical Engineer, (1980) 557.

21

C HAPTER 1 Bioleaching Applications


A novel bio-leaching process to recover valuable metals from Indian Ocean nodules using a marine isolate Amitava Mukherjeea, Ashok M. Raichura, Jayant M. Modakb, K.A. Natarajana a

Department of Metallurgy Department of Chemical Engineering Indian Institute of Science, Bangalore – 560012, India b

Abstract A novel bio-leaching process to recover valuable nonferrous metals from Indian Ocean nodules at near neutral pH and ambient temperature is presented in this paper. Poly-metallic manganese nodules contain a lucrative resource for valuable strategic metals like Cu, Co and Ni. In view of rapid depletion of land -based resources of copper and nonavailability of significant resources of Ni and Co in India, these nodules have an enormous potential to effect future metal extraction trends in the country. The significant role of indigenous microorganisms in solubilization of valuable trace metals from the nodules was studied in detail. A marine organism was isolated from the nodules. The isolate was shown to be a Gram-positive heterotroph of Bacillus species. It was grown in artificial seawater medium. The growth studies of the isolate with respect to pH, temperature and salinity of the medium proved that, the isolate grows well at neutral pH, 30°C temperature and 0.25M NaCl concentrations. A growing culture of the isolate as well as cell-free spent liquor containing metabolites was employed to recover transition metals from the nodules. A few chemical leaching experiments in moderate acidic environment were also conducted to generate baseline data. The bioleaching of the nodules was carried out in two different ways (a) leaching during the growth of the isolate and (b) leaching by spent liquor after removal of fullygrown cells. The leaching efficiency was observed to be more or less the same in both the cases confirming that metabolites produced during growth of the microorganism, played a pivotal role in the leaching process. Around 50% cobalt and 30% of the Cu and Ni came out in the leach liquor at pH of 8.2 in a four-hour interval, indicating that the effect of metabolites was specific to cobalt recovery in comparison to copper and nickel. Kinetic studies revealed that the metal recovery stabilized after four hours. The results for cobalt were quite comparable to those achieved in highly acidic conditions. A strong reducing environment is required to break down the MnO2 and/or Fe-oxide matrix encapsulating the remaining part of the transition metals. So, the effect of adding a reductant like starch in spent liquor of the isolate was investigated. Starch enhanced the recovery of all the transition metals to about 80%-85%. The effect of pulp density and pH of the leach liquor on the bio-leaching process was investigated. The recovery was observed to be almost independent of size fraction of nodules over a wide range. 25

Bioleaching Applications

1.

INTRODUCTION In view of continuous depletion of land-based resources along with increasing consumption of valuable metals in India, development of environment friendly technologies for tapping alternate sources of metals has gained importance lately. One of them is recovery of strategic metals Cu, Ni and Co from polymetallic Indian Ocean nodules, by biological processing. Fuerstenau and Han (1983) extensively reviewed processing and extraction of valuable metals from manganese nodules discussing both hydro and pyrometallurgical routes. High porosity of the nodules resulting in high moisture content coupled with polluting effluent gases pose major hurdles in pyrometallurgical processing of the nodules. Therefore, hydrometallurgical techniques are potentially viable for extraction of metals from nodules. But often, slow kinetics and poor recovery with dilute acids and corrosiveness of the concentrated ones restrict application of hydrometallurgical extraction. Researchers have studied addition of reducing agents with either mineral acids or ammonia. The reducing environment, so created, enhances leaching, by breaking up the nodule matrix occluding the valuable metals (Niinae et al 1996; Kanungo et al 1988; Jana et al 1999; Trifoni et al 2001; Zhang et al 2001). These processes have varying degrees of success. However, most require high temperature pretreatment and/or costly, corrosive reagents to obtain a sizeable amount of metal recovery with favourable kinetics. As the nodules are low-grade ores of Cu, Co and Ni, use of costly chemical reagents as reducing agents may not be economically viable for large-scale runs. Ehrlich and his co-workers have studied the biogenesis and microbiology associated with Atlantic Ocean nodules extensively over the last four decades. Ehrlich (1963) isolated and characterized the Mn-reducing organism Bacillus 29 from the Atlantic Ocean nodules. Growing cultures of Bacillus 29 are able to reduce MnO2 aerobically and anaerobically using glucose as an electron donor. However, microbial ecology of the Indian Ocean nodules has not been studied in detail until now; utilizing microorganisms isolated from the nodules themselves to leach valuable metals from the nodules, still remains an unexplored route. In the recent past, researchers have been looking into bioprocessing as an alternative route of metal recovery from the nodules. Konishi et al (1997) showed nodule leaching by acidophilic sulphur oxidizing bacteria and thermophilic A. brierleyi. A. Kumari and Natarajan (2001) have been able to extract valuable metals from the nodules by electrobioleaching using acid producing chemolithotrophs like Acidithiobacillus ferrooxidans and Acidithiobacillus thiooxidans. But these processes employed a highly acidic environment supplemented with thermal or electrical energy for recovering a sizeable amount of metals. Major objectives of our present work were to isolate an organism from Indian Ocean nodules, characterize and grow the isolate, and ultimately use the isolate or its growth products for solubilisation of Cu, Co and Ni from the nodules. The recovery of metals through a biological route is compared with that of chemical leaching using dilute acids. The effect of reducing additives in the supernatant, pH of the leaching medium and pulp density were also investigated.

26


2.

MATERIALS AND METHODS

2.1 Materials Ocean nodule sample was collected from the bed of the Indian Ocean by the National Institute of Oceanography, Goa, India. The sample was ground in a mortar and pestle and sieved to obtain appropriate size fractions. The partial chemical composition of the different size fractions of the as-received sample is presented in Table 1. Phases revealed in the X-ray diffraction pattern of the nodule sample are depicted in Table 2. Table 1. Partial chemical analysis of the ocean nodule

Table 2. XRD analysis of the ocean nodules

Quartz

Chemical composition SiO2

1.15

Asbolan

NiMn2O3(OH)4

Ni

1.12

Co

0.16

Cobalt manganese oxyhydroxide

(Co, Mn) OOH

Fe

8

Element

Wt %

Mn

22

Cu

Phases present

2.2 Microorganisms 2.2.1 Isolation of the unidentified marine species The marine bacteria occurred inherently on the nodule sample as a spore-former. To remove the surface contaminants and germinate the spores, the nodule sample was boiled in water for 30 minutes (Ehrlich, 1963). After boiling, hot water was removed by decantation. The lump of nodule was transferred to an autoclaved porcelain mortar (outer diameter, 150mm) in an inoculating hood, which had been pre-sterilized for at least 30 min by UV rays. The lump was pulverized by pounding in the mortar with a sterilized pestle. One g of this nodule powder was added aseptically to 20ml of sterilized artificial seawater nutrient broth whose composition is given in Table 3. Henceforth the medium will be referred to as ASWNB medium. ASWNB with the nodule powder was incubated at 37°C for 24 hrs. From this enrichment media a loopful of the inoculum was streaked on to artificial seawater nutrient agar plates and incubated at 37°C for 24hrs. The next day round, creamy, smooth surface colonies were found growing on the same medium. A colony was picked from the plate and sub-cultured several times on the same medium to finally obtain the pure strain of the marine isolate. Periodic streaking was done to check for the purity of the isolated strain. Henceforth the bacterium will be referred to as "marine isolate". Table 3. Composition of Artificial Sea water nutrient broth (ASWNB) Chemical components

Wt. in g. in 100 ml of distilled water

Sodium chloride

28.13

Potassium chloride

0.77

Calcium chloride dihydrate

1.60

Magnesium chloride hexahydrate

4.80

Sodium bicarbonate

0.11 27

Bioleaching Applications Chemical components

Wt. in g. in 100 ml of distilled water

Magnesium sulphate heptahydrate

3.5

Peptone

5.0

Beef extract

3.0

Sodium chloride

28.13

Potassium chloride

0.77

Calcium chloride dihydrate

1.60

Magnesium chloride hexahydrate

4.80

Sodium bicarbonate

0.11

Magnesium sulphate heptahydrate

3.5

Peptone

5.0

Beef extract

3.0

2.2.2 Growth of the marine isolate The marine isolate was grown in ASWNB in 250-ml baffled Erlenmeyer flasks at 30°C on a rotary shaker (200 rpm). Ten percent of an active inoculum (from the late exponential phase) containing at least 109 cells/ml was added to the sterilized ASWNB medium. The growth of the microorganism was monitored, by measuring the cell count using a Petroff-Hauser counter employing phase-contrast microscopy. The sodium chloride concentration was kept at 0.25M, which was arrived at by testing at different salt concentrations. Growing cells as well as cell free growth supernatant, containing metabolites produced during growth, were used as bioleaching agents. 2.3 Methods 2.3.1 Chemical leaching experiments Chemical leaching experiments were carried out in 250-ml Erlenmeyer flasks on an incubated rotary shaker at 200 rpm at 30°C. In all the cases, 1 g of properly crushed nodule of 50-75 µm size fraction was used and the solid: liquid ratio was kept at 1:100(W/V). All of these experiments were conducted to generate baseline data to compare with bioleaching results. In order to optimize different parameters, the duration of leaching was fixed at four hours in some cases. Parameters such as choice of mineral acids, time of leaching, requirement of reducing agents, and the effects of organic and inorganic additives were investigated. HCl, HNO3 and H2SO4 solutions of 2.5M concentrations were used as leaching agents. H2SO4 solution (pH 2.0) alone or with reducing agents like sodium thiosulfate was employed for leaching also. After leaching, leach liquor was filtered using Whatman 42 filter papers and the collected residue was digested in 1:1 HCl at 60-70°C. The resultant solution, after proper dilutions were made, was analyzed for Cu, Co, Ni, Mn and Fe with an Inductively Coupled Plasma spectrophotometer (ICP). All chemicals used were of reagent grade. 2.3.2 Bioleaching experiments Leaching with growing culture: One g of pre-sterilized, pulverized ocean nodule was placed in 90 ml of sterilized ASWNB media in 250 ml conical flasks; 10% v/v actively growing culture (109cells/ml) of the marine isolate was added as inoculum. Growth flasks were removed from the rotary shaker after appropriate time intervals and the solution analyzed for leached metal content. 28


Leaching with cell-free growth supernatant: To obtain cell free growth supernatant, a fully-grown culture (after 10 hours of growth) was centrifuged at 10,000 rpm for 15 minutes followed by pressure filtration using Millipore ultra-filtration unit. The absence of any cells in the resultant supernatant was assured by observing under phase contrast microscope. One g of pulverized ocean nodule was added to100 ml of the growth supernatant and solid: liquid ratio was kept at 1:100. To optimize recovery of metals in leaching, pH of the growth supernatant was varied from an acidic to an alkaline range by adding 10N H2SO4or 0.1N NaOH. The duration of leaching was kept constant at four hours. The size fraction of the crushed nodules was in the range of 50 to 75 microns for all tests. Leaching with starch added to the growth supernatant: To observe the effect of starch addition to the growth supernatant, increasing proportions of starch were added to 100 ml of cell free growth supernatant and the solid: liquid ratio was kept constant at 1:100. The duration of leaching was maintained at four hours. The size fraction of the crushed nodule was in the range of 50 to 75 microns. Solution pH in all the cases was within 8-8.5 ranges In all the above tests, leach liquor collected after appropriate time intervals was filtered using Whatman 42 filter papers and the residue was digested in 1:1 HCl at 6070°C. The resultant solution, after proper dilutions were made, was analyzed for Cu, Co, Ni, Mn and Fe by an ICP spectrophotometer. 3.

RESULTS AND DISCUSSIONS

3.1 Characterization of the marine isolate A marine bacterium was isolated from an Indian Ocean nodule sample following the procedure discussed in section 2.2. A preliminary morphological and physiological examination of the marine isolate was carried out. Results are presented in Table 4. The isolated strain was rod shaped, motile and able to reduce Mn (IV) to Mn (II). Table 4. Characterization of the marine isolate Shape

Bacilli

Gram reaction

+

Motility

+

Manganese reduction

+

Catalase activity

+

Oxidase activity

+

Aerobic metabolism

+

3.2 Chemical leaching experiments Baseline leaching tests of the nodules in HCl, HNO3 and H2SO4 of 2.5M concentrations were carried out and the results are summarized in Table 5 below. We can observe from the table that, though Cu recovery is around 80% in all the cases and Ni recovery is about 60%, Co and Mn recoveries are very low. When leaching with HCl, only 30% Co leaches is leached. Co recovery is negligible for the other two acids. Low Co recovery in all the above tests may be attributed to low Mn recovery. A major part of Co in nodules is supposedly occluded in the MnO2 matrix, so disintegration of the matrix is an essential prerequisite for Co solubilization. Therefore presence of a reducing 29


agent, like glucose, under acidic condition with HCl remarkably enhances the recovery of Mn and Co. With glucose in a 2.5M HCl solution 50% Co is extracted while Mn recovery increased to 80%. Again the disparity in Co and Mn recovery proves that Mn and Co recovery do not follow a simple 1:1 correlation. It is possible, however, that all of the cobalt in the nodules might not be directly associated with Mn. Table 5. Leaching of ocean nodules by mineral acids with and without reducing agents Leaching reagent

% Recovery of metals Co

Cu

Ni

Mn

Fe

2.5 M HCl

30

80

55

30

60

2.5M HCl + 20% glucose

50

85

85

80

65

2.5M HNO3

2 FeOOH + Mn2+ + 2 H+

(3)

Fe(III) generated by the reaction of Fe(II) with MnO2 could react with FeS2 before it precipitates as FeOOH. D) The coupling of the redox pairs FeS2/Fe(III) and Fe(II)/MnO2 has been suggested for the dissolution of low grade ores or ocean bed nodules in acid media [20, 21]. E) Molecular-orbital theory considerations by Luther [29, 30] support Fe(II)/Fe(III) cycling. Fe(II) is a d6 (t2g6) electron configuration and Mn(IV) is a d3 (t2g3) electron configuration. The t2g orbitals are filled in case of Fe(II) and half-filled in case of Mn(IV) which imparts the stability for these metal ions. For both solids, FeS2 and MnO2, to react with each other, a ligand would have to dissociate as both reactants touched, but this does not occur. Soluble Fe(II), which has a t2g4 eg*2 electron configuration, is high spin and labile, thus it can adsorb to and react with MnO2 to form Fe(III). Soluble Fe(III) has d5 (t2g3 eg2) electron configuration and is therefore a labile cation that can undergo ligand exchange and is therefore able to react with the S22- ligand of FeS2. Summarizing the results, a model of FeS2 oxidation by MnO2 is shown in Fig. 2. It is postulated that electrons are transported via the Fe(II)/Fe(III)-shuttle if FeS2 and MnO2 are in a close contact. Fe(II) and Fe(III) should be adsorbed onto the surface of FeS2 because our experiments with amorphic Fe(III) oxide have shown that precipitated Fe(III) does not oxidize FeS2. However, while amorphic Fe(III) oxides precipitated to the FeS2 surface do not alone oxidize FeS2, they may serve as an electron conduit [31]. Electrons might flow from FeS2 via Fe(III) oxides to adsorbed Fe(III) or via Fe(III) oxides directly to MnO2. In our experiments with MnO2, only precipitated Fe(III) but not Fe(II) was detected by extraction with HCl, indicating that the reaction between Fe(II) and MnO2 is faster than the reaction between Fe(III) and FeS2. All reactions in this model were shown to be purely chemical, however, biological catalysis could be involved in the degradation of sulfur intermediates [9, 32, 33].

60


Figure 2. Model of anaerobic FeS2 oxidation by MnO2 via the Fe(II)/Fe(III)-shuttle [22]

In nature, Fe(III) might be stabilized in solution complexed to organic ligands, thus the Fe(II)/Fe(III)-shuttle might transport electrons even if FeS2 and MnO2 are not in close contact. Soluble complexed Fe(III) has been shown to exist in sediment pore waters [34, 35]. Furthermore, the FeS2 oxidation rate increases in presense of Fe(III)-chelating ligands [26, 30]. Besides MnO2, other manganese oxides might oxidize FeS2 as well because the standard redox potential of the couples MnO2/Mn2+, Mn3O4/Mn2+, and MnOOH/Mn2+ are all around 600 mV [36]. In contrast to FeS2 oxidation, FeS oxidation by MnO2 only produced elemental sulfur and some sulfate as oxidation products. According to the polysulfide mechanism [5, 6, 10], acid soluble metal sulfides are dissolved by Fe(III) and proton attack. The sulfide is oxidized via radicals and polysulfides mainly to elemental sulfur besides some sulfate, which is in agreement with our results. Therefore, the chemical FeS oxidation is described by the following equation: FeS + 1.5 MnO2 + 3 H+ ----> Fe(OH)3 + So + 1.5 Mn2+

(4)

Unlike MnO2, amorphous Fe(III) oxide was not an oxidant for FeS2 or FeS at pH 8 in the experiments of this study, even not in the presence of organic Fe-complexes or of the electron transporting compound AQDS. Luther et al., [30] showed a chemical FeS2 oxidation by 1 mM ferrihydrite and 10 mM salicylic acid in the pH range of 4 to 6.5. Ferrihydrite and salicylic acid form a Fe(III) salicylate complex which reacts with FeS2. Liu and Millero [37] showed that the solubility of Fe(III) in the presence of Fe(III) complexing humic acids is two orders of magnitude higher at pH 4-6 than at pH 8. Presumably, the concentration of complexed Fe(III) in the experiments of this study at pH 8 was too low to enable FeS2 dissolution. FeS2 and FeS were also not chemically oxidized by NO3- in the experiments of this study. Ottle y et al. [38] have shown, that a chemical oxidation of Fe(II) by NO3- can be catalyzed by metals such as copper. Thus, a chemical oxidation of FeS2 by NO3- might exist as well. A FeS2 oxidation by the reduction of NO3- has been suggested for aquifers based on geochemical data [39, 40] but clear experimental evidence is lacking. Bacteria may be involved in this process.

61


4.2 Anaerobic biological iron sulfides oxidation From anaerobic marine sediments bacteria could be enriched which anaerobically oxidize FeS with nitrate as electron acceptor. This finding is in agreement with results of Garcia-Gil and Golterman [41] who described a FeS-mediated denitrification for a marine sediment. FeS belongs to the acid soluble metal sulfides which are chemically oxidized via polysulfides to mainly elemental sulfur and some sulfate [5, 6, 10]. Due to its acid solubility, protons dissolve FeS according to:

FeS + H+ ----> Fe2+ + HS-

(5) -

Both products of this reaction may be oxidized by NO3 reducing bacteria. The Fe2+ can be oxidized according to Straub et al. [42]: 10 FeCO3 + 2 NO3- + 24 H2O ----> 10 Fe(OH)3 + N2 + 10 HCO3- + 8 H+

(6)

-

HS may be oxidized by e.g. Thiobacillus denitrificans or Thiomicrospira denitrificans [11]: 5 HS- + 8 NO3- + 3 H+ ----> 5 SO42- + 4 N2 + 4 H2O

(7)

In equation 6, protons are produced which continue to dissolve FeS. Bacteria might be attached to the FeS surface embedded in extracellular polymeric substances (EPS). Bacteria produce EPS to create a microenvironment which favours their metabolisms [5, 6]. In such a microenvironment, the pH might be much lower than 8 enabling FeS dissolution. Consequently, Fe2+ or HS- oxidizing and NO3- reducing bacteria can grow with FeS as a substrate, and I was able to enrich these bacteria from different marine sediments. With FeS2 as a substrate, bacteria did not grow since FeS2 cannot be dissolved by protons. Precipitation of Fe(III) hydroxide might explain the absense of 55FeS2 dissolution in a o S and Fe2+ oxidizing and NO3- reducing bacterial culture. The bacteria oxidize Fe(II) to Fe(III) which has to diffuse from the Fe-oxidizing enzyme of the bacteria to the FeS2 surface to serve as an oxidant for FeS2. Obviously, Fe(III) precipitates immediately and therefore cannot serve as an oxidant for FeS2. In aquifers where slightly acidic pH values were detected, a FeS2 oxidation by the reduction of NO3- has been suggested based on depth profiles of NO3- and SO42- [39, 40]. There, Fe2+ oxidizing and NO3- reducing bacteria and soluble organic Fe(III) complexes could probably catalyze an anoxic FeS2 oxidation with NO3- as electron acceptor. REFERENCES

1. H. Brandl, in H.-J. Rehm and G. Reed in cooperation with A. Pühler and P. Stadler (eds.), Biotechnology, Vol. 10, Wiley-VCH, Weinheim, Germany (1991) 191. 2. H.L. Ehrlich, Geomicrobiology, Marcel Dekker Inc., New York, 2002. 3. D.E. Rawlings, Ann. Rev. Microbiol., 56 (2002) 65. 4. T. Gehrke, J. Telegdi, D. Thierry and W. Sand, Appl. Environ. Microbiol., 64 (1998) 2743. 5. W. Sand, T. Gehrke, P.-G. Jozsa and A. Schippers, in R. Amils and A. Ballester (eds.), Biohydrometallurgy and the environment toward the mining of the 21st century, Part A, Elsevier, Amsterdam (1999) 27. 6. W. Sand, T. Gehrke, P.-G. Jozsa and A. Schippers, Hydrometallurgy 59 (2001) 159. 7. K.J. Edwards, B. Hu, R.J. Hamers and J.F. Banfield , FEMS Microbiol. Ecol., 34 (2001) 197. 8. Schippers, P.-G. Jozsa and W. Sand, Appl. Environ. Microbiol., 62 (1996) 3424. 62


9. A. Schippers, T. Rohwerder and W. Sand, Appl. Microbiol. Biotechnol., 52 (1999) 104. 10. A. Schippers and W. Sand, Appl. Environ. Microbiol., 65 (1999) 319. 11. J.G. Kuenen, L.A. Robertson and O.H. Tuovinen, in A. Balows, H.G. Trüper, M. Dworkin, W. Harder and K.-H. Schleifer (eds.), The Prokaryotes, Springer Verlag, Berlin (1992) 2638. 12. T. Brinkhoff, G. Muyzer, C.O. Wirsen and J. Kuever, Int. J. Syst. Bacteriol., 49 (1999) 385. 13. G.J.M.W. Arkesteyn, Plant and Soil, 54 (1980) 119. 14. A. Schippers, H. von Rège and W. Sand, Minerals Engineering, 9 (1996) 1069. 15. D. Emerson and C. Moyer, Appl. Environ. Microbiol., 63 (1997) 4784. 16. A. Das and A.K. Mishra, Appl. Microbiol. Biotechnol., 45 (1996) 377. 17. T.D. Brock and J. Gustafson, Appl. Environ. Microbiol., 32 (1976) 567. 18. J.T. Pronk and D.B. Johnson, Geomicrobiol. J., 10 (1992) 153. 19. K. Küsel, U. Roth, T. Trinkwalter and S. Peiffer, Environ. Experim. Bot., 46 (2001) 213. 20. S.B. Kanungo, Hydrometallurgy, 52 (1999) 313. 21. R.K. Paramguru and S.B. Kanungo, Can. Metal. Quart. 37 (1998) 389. 22. A. Schippers and B.B. Jørgensen, Geochim. Cosmochim. Acta, 65 (2001) 915. 23. A. Schippers and B.B. Jørgensen, Geochim. Cosmochim. Acta, 66 (2002) 85. 24. D.R. Lovley and E.J.P. Phillips, Appl. Environ. Microbiol., 51 (1986) 683. 25. D.R. Lovley and E.J.P. Phillips, Appl. Environ. Microbiol., 54 (1988) 1472. 26. S. Peiffer and I. Stubert, Geochim. Cosmochim. Acta, 63 (1999) 3171. 27. C.O. Moses and J.S. Herman, Geochim. Cosmochim. Acta, 55 (1991) 471. 28. D. Postma and C.A.J. Appelo, Geochim. Cosmochim. Acta, 64 (2000) 1237. 29. G.W. III Luther, Geochim. Cosmochim. Acta, 51 (1987) 3193. 30. G.W. III Luther, J.E. Kostka, T.M. Church, B. Sulzberger and W. Stumm, Mar. Chem., 40 (1992) 81. 31. C.M. Eggleston, J.-J. Ehrhardt and W. Stumm, Amer. Mineral. 81 (1996) 1036. 32. B. Thamdrup, K. Finster, J. Würgler-Hansen and F. Bak, Appl. Environ. Microbiol., 59 (1993) 101. 33. K. Finster, W. Liesack and B. Thamdrup, Appl. Environ. Microbiol., 64 (1998) 119. 34. M. Huettel, W. Ziebis, S. Forster and G.W. III Luther, Geochim. Cosmochim. Acta, 62 (1998) 613. 35. M. Taillefert, A.B. Bono and G.W. III Luther, Environ. Sci. Technol. 34 (2000) 2169. 36. B. Thamdrup, in B. Schink (ed.) Advances in Microbial Ecology, Kluwer Academic/Plenum Publishers, New York (2000) 41. 37. X. Liu and F.J. Millero, Geochim. Cosmochim. Acta, 63 (1999) 3487. 38. C.J. Ottley, W. Davison and W.M. Edmunds, Geochim. Cosmochim. Acta, 61 (1997) 1819. 39. D. Postma, C Boesen, H. Kristiansen and F. Larsen, Water Resources Research 27 (1991) 2027. 40. P. Engesgaard and K.L. Kipp, Water Resources Research, 28 (1992) 2829. 41. L.J. Garcia-Gil and H.L. Golterman, FEMS Microbiol. Ecol. 13 (1993) 85. 42. K.L. Straub, M. Benz, B. Schink and F. Widdel, Appl. Environ. Microbiol., 62 (1996) 1458.

63


Bacterial growth and propagation in chalcocite heap bioleach scenarios J. Petersena and D.G. Dixonb a

Department of Chemical Engineering, University of Cape Town, Rondebosch 7700, South Africa b Department of Metals and Materials Engineering, University of British Columbia, 6350 Stores Road, Vancouver, B.C. V6T 1Z4, Canada Abstract Heap bioleaching of chalcocite ores is widely practised as a relatively low cost process option, especially for marginal deposits. However, the rates of copper extraction achieved in actual operations are often slower than expected. A recent study by the authors has shown that chalcocite heap bioleaching is controlled by a two-stage mechanism. The first stage is rapid and ultimately controlled by the supply of acid to the reaction sites. The second stage is intrinsically much slower and controlled by mineral oxidation kinetics. In this work, the results of a column operated with a bacterial consortium grown under idealised laboratory conditions are compared to those of a column run with a native culture and high-TDS (total dissolved solids) raffinate solution from the mine site. The results clearly indicate that while the leach reactions follow the two-stage mechanism in both cases, the overall rate of copper extraction in the high-TDS column is significantly retarded. Bacterial growth in this column is slower and much more limited. It is postulated that in such high-TDS environments the rate of copper leaching is controlled by the rate of bacterial growth and oxidation. Bacterial growth and propagation trends observed in the experiments have been reproduced with a comprehensive heap simulation tool developed by the authors. The bacterial growth and oxidation model employed is briefly introduced. The simulations confirm that, if the bacterial growth rate is significantly retarded, it can become rate controlling over other factors. In full-scale heaps, however, the principal rate-limiting factor is still the diffusion of acid into large stagnant zones. Keywords: heap leaching, bio-oxidation, growth kinetics, yield, TDS, growth inhibition 1.

INTRODUCTION Heap bioleaching of chalcocite ores has become widely practised for the economic extraction of copper from low-grade deposits. However, in many operations the rate of copper extraction falls well behind what can be achieved in laboratory columns. Some concerns have been expressed that field conditions (high altitude, low temperatures, high TDS (total dissolved solids) of process waters) are adverse to bacterial growth, and may therefore be the cause for slow extraction rates. Some successes with improving heap performance by adapting cultures to such conditions have been reported [1], but there 65


appears to be only a limited understanding of the interactions between minerals and bacteria in chalcocite heap leach situations. The authors have conducted a comprehensive study into the dynamics of chalcocite heap bioleaching, and some results are presented in a previous paper [2]. Based on fundamental electrochemical investigations [3], extensive column studies and simulations with a comprehensive heap leach modelling tool, it has been found that the rate of chalcocite leaching by ferric ions, both in columns and in full scale unconfined heaps, is driven by a two-stage mechanism. In stage 1, approximately 40% of copper from chalcocite is released to form an interim pseudo-covellite compound: (1) Cu2S + 0.8 Fe2(SO4)3 → 0.8 CuSO4 + Cu1.2S + 1.6 FeSO4 The mineral undergoes significant structural changes during stage 1 leaching, such that the second stage, which releases the remaining copper, must be viewed as a separate reaction: (2) Cu1.2S + 1.2 Fe2(SO4)3 → 1.2 CuSO4 + S° + 2.4 FeSO4 In bioleaching the necessary ferric is in each case regenerated continuously through bio-oxidation: 2FeSO4 + 0.5 O2 + H2SO4 → Fe2(SO4)3 + H2O (3) The first stage of chalcocite oxidation is kinetically very rapid and is controlled primarily by the supply of ferric to the reaction site, and hence by the rate of ferrous to ferric oxidation facilitated by bacteria. As ferric consumption is virtually instantaneous, solution potentials are generally quite low (< 500 mV Ag/AgCl). In heaps the only acid available for reaction (3) during this stage is that coming in with the feed, since pyrite oxidation is typically limited and no elemental sulfur is being generated for subsequent bio-oxidation to sulfate. This restricts the stage 1 oxidation reaction to a narrow zone, which progresses downwards, commensurate with the rate of acid supply. In unconfined heaps this phenomenon is complicated further by transport effects in stagnant zones. These have been used to explain the large discrepancies in leach rates between full-scale heaps and column experiments [2]. Stage 2 oxidation proceeds in the wake of the stage 1 zone. This step is kinetically much slower at ambient temperatures. During bioleaching, the slow kinetics result in a build-up of ferric near the mineral surface and consequently the reaction proceeds at much higher solution potentials (> 650 mV Ag/AgCl). At the same time elemental sulfur is generated, which can be oxidised to generate sulfuric acid, and pyrite oxidation can also proceed more favourably. As the rate of reaction is relatively slow, it proceeds homogeneously over the entire height of the column (or heap), once stage 1 has passed through. Bacterial growth and oxidation kinetics in chalcocite heaps must be investigated against this background. It is conceivable that reaction (3) is severely retarded by bacteria growing under adverse conditions, and thus becomes overall rate limiting over other factors. In the present study, column leach experiments using a laboratory culture are compared with those using a native culture growing in a high-TDS raffinate. The different growth rates are quantified and the effect in large columns and heaps is investigated with the aid of a comprehensive modelling tool.

66


2.

EXPERIMENTAL

2.1 Column Studies The ore used in all experiments originated from a typical copper mining and heap leach operation. Top particle size was ¾ inch and the minus 100 micron fines content was 10%. The sample ore had an average Cu grade of 1.45%, approximately 30% of which was acid-soluble (mostly brochantite) and the rest was chalcocite, with some minor occurrences of chalcopyrite. The pyrite content was approximately 2.5%. The experimental work was conducted in mini-columns, 400 mm tall and 100 mm in diameter, immersed in a water-bath with the temperature controlled at 25°C. In each experiment a number of these columns were run in series, with solution from one column collected in a small, sealed interim container and pumped from there into the next column by means of a peristaltic pump. The standard experimental irrigation rate was 5 L/m2-hr and all columns were well aerated with air enriched with 1% v/v CO2. Before charging to the columns, the ore was acid agglomerated in accordance with operating practice at the mine site. All columns were operated on a once-through basis. Two column experiments, denoted Z5 and Z8, were run in the context of this study. Z5 had 8 columns in series with a total height of 2.5 m. The experiment was inoculated with a laboratory bacterial culture derived from a blend of pure cultures of Acidithiobacillus ferrooxidans, Acidithiobacillus thiooxidans and Leptospirillum ferrooxidans. This had been continuously maintained on a minus 40-mesh fines fraction of the original ore sample in standard K9 medium, in an incubator shaking at 150 rpm at 30°C, with weekly transfers. The feed solution consisted of 7.5 g/L H2SO4, 0.7 g/L Fe(III) as sulfate and 1.3 g/L Fe(II) as sulfate. This solution reflects the pH, potential and Fe concentrations reported from the mine. Column experiment Z8 consisted of four columns in series with a total height of 1.2 m. This experiment was inoculated with a native culture obtained from the mine site, which was well adapted to the high TDS raffinate solution, and the original raffinate sampled run-of-production at the mine site. The culture was maintained in the same way as the laboratory culture, but in the original raffinate augmented with K9 culture medium. The composition of the original raffinate is given in Table 1. Of note are the extremely high concentrations of Al and Mg, which exceed toxicity limits reported in the literature [1, 4, 5], and the elevated chloride levels. Table 1. Analysed composition of the mine site raffinate (only components >10 mg/L listed) Element Al Ca Co Cu Fe Mg Mn

Concentration [mg/L] 12,200 467 16.2 216 2,460 10,100 669

Element P K Na Zn Cl– F– NO2–

Concentration [mg/L] 221 29.0 1,670 376 1,300 80.1 28.1

NO3– o-PO4 SO4

Concentration [mg/L] 105.9 532 116,880

pH E (mV vs SHE)

1.24 640

Element

Figure 1 shows the copper extraction vs. time achieved in both tests. Although Z5 was halted to allow early assay of solids, the expected extraction trend (dashed line) - based on 67


observations from similar experiments - has been added for comparison. In both experiments the first 30% of copper extraction represents the rinsing of acid-soluble copper dissolved upon acid agglomeration. The next 30% of copper extraction corresponds roughly to stage 1 leaching, and it is in this phase where the two experiments differ significantly in terms of extraction rate: In Z5 this reaction is nearly complete within 20 days, whereas Z8 requires almost 40, although being only half as tall. Leaching beyond 60% extraction continues in Z8 at a rate similar to that expected for Z5 (although not measured in this particular experiment). 100%

Cu extraction

80%

60% 40% Exp. Z5

20%

Exp. Z8 0% 0

10

20

30

40

50

60

70

days on stream

Figure 1. Copper extraction in experiments Z5 and Z8. The dashed line indicates trends observed from similar experiments

Figures 2 shows the effluent pH and the progression of solution potentials along the height of the columns. With regard to Z5 these two figures illustrate that chalcocite stage 1 leaching progresses in a narrow zone down the column, consuming all available acid and ferric, and thus maintaining the solution potential at very low levels. 800

3

Pot ential (vs . Ag/AgCl)

Exp. Z5 Exp. Z8

pH

2.5

2

1.5

1

700

Z5 - 0.3 m Z5 - 0.9 m Z5 - 1.8 m

600

500

Z5 Z8 Z8 Z8

- 2.4 - 0.3 - 0.6 - 0.9

m m m m

400

Z8 - 1.2 m

300 0

10

20

30

40

days on s tream

50

60

70

0

10

20

30

40

50

60

70

days on s tream

Figure 2. pH measured in the effluent and solution potentials measured over the length of the column from experiments Z5 and Z8. The dashed lines indicate trends observed from similar experiments

In the wake of this zone (stage 2 leaching) the potential is rising rapidly to levels around 700 mV (vs. Ag/AgCl), and once the zone breaks through (not quite achieved in Z5 before shut-down, but observed in similar experiments (dashed line)), the solution pH gradually drops to feed levels. In Z8 the zone is still observed, but it progresses much more slowly, and the transition from stage 1 to stage 2 leaching is much more gradual. Only some of the available acid is consumed, but never depleted, and the pH remains consequently more or less stable with only a minor peak around day 50, when the low potential front breaks through. Thus it is clear that stage 1 leaching proceeds through Z8 68


within a broad band rather than a narrow zone, and it is not controlled by acid supply as opposed to Z5. The cause for these discrepancies becomes clear from Figure 3. Plotting the number of bacteria counted in solution as a function of time and depth in the heap shows counts lower by an order of magnitude in Z8 (native bacteria in high-TDS raffinate) as compared to Z5 (laboratory culture in artificial raffinate). Both sets of data show the same progression trends, however, with numbers moving down the columns in a "growing wave". The propagation rate of this wave corresponds to that of the high potential wave in the wake of the chalcocite stage 1 leach front. Therefore, it is postulated that the slower copper extraction rate in Z8 is linked to the slower propagation of bacteria through the column at much smaller numbers. The rate of copper leaching is hence controlled by bacterial growth kinetics rather than acid supply.

1.0E+08

1.E+0 8

1.E+0 7 Cell Count per mL

Cell Count per mL

1.0E+07

1.0E+06

1.0E+05

1.E+0 6

1.E+05 1.0E+04 m 0. 3

m 0. 6

m 0. 9

m 1. 2

Column Depth

m 1. 5

m 1. 8

m 2. 1

m 2. 4

we ek 3 we ek 2 we ek 1

1.E+04 0 .3

m

0 .6

m

0.9

m

1.2

m

we e we k 8 e we k 7 ek we 6 ek 5 we ek 4

Column Depth

Figure 3. Bacterial counts over the length of the column in experiments Z5 and Z8 2.2 Bacterial Growth Curves The growth characteristics of the cultures used in the column experiments were investigated in a series of shake-flask experiments. Two sets of seven flasks containing 75 mL of culture medium (standard K9 for the laboratory culture and original raffinate augmented with K9 growth medium for the native culture, each containing 2 g of ore fines) were prepared. At time zero, 25 mL of the respective mature culture (grown for 5 days in the same medium in a shaker incubator at 30°C, 150 rpm) was introduced into each medium and placed in the shaker. Both sets were prepared in duplicate, and a blind test was also prepared. Flasks were removed after 20 minutes (for the time-zero sample), 1, 2, 3, 5, 7 and 9 days, and left to settle for 5 minutes, after which a 5-mL sample was drawn from the supernatant, and from which the bacterial culture was enumerated. The resulting growth curves are shown in Figure 4. Clearly the laboratory culture grows much more rapidly and to larger numbers than the native culture. Interesting is the initial "overshoot" in the lab culture, before numbers settle to lower levels. This is thought to correspond to the stage-wise leaching of chalcocite, similar to the trends observed in the column experiments. The growth curves were analysed to extract initial growth rate and final cell yield, as reflected in Table 3. This suggests that the native culture grew about 6 times more slowly than the laboratory culture, and at one fifth of the yield. 69


Figure 4. Growth curves for the laboratory culture in the basic culture medium and the native culture in the original raffinate Table 2. Growth rate and final yield of growth experiments Culture

Growth rate (h–1)

Doubling time (h)

Final Yield (cells/mL)

Laboratory

0.109

6.36

1.10 × 108

Native

0.019

36.3

2.27 × 107

3.

MODELLING STUDY The authors have developed a comprehensive heap leach modelling tool (HeapSim) [2, 6], which combines an advection-diffusion model to account for transport of solutes through the heap (or column) to mineral sites in the ore, with a multi-reaction model at the mineral site. Gas adsorption, microbial growth and oxidation, and mineral leaching are represented through appropriate kinetic terms. A comprehensive description of the model is beyond the scope of this paper, but the equations describing bacterial growth and oxidation, as well as some key parameters, are discussed below. The bacterial growth model essentially follows Monod type kinetics [5]: dX = Xk g Π dt (4)

where X denotes the bacterial population per unit volume (i.e. [cells/mL]), kg is the growth rate (also commonly referred to as ⎧max, [h-1]), and Π denotes the product of a number of terms describing substrate and kinetic limitations (involving Fe2+, Fe3+, O2, acid, T, etc.). Some forms of the Π term have been reviewed by Nemati et al. [7]. The rate of ferrous oxidation is related to bacterial growth by a simple yield coefficient, Y [cells/mol Fe2+], thus: rFe2+ = −

1 dX Y dt

(5) The HeapSim code has been calibrated to model the column data generated in experiment Z5 and Z8 on the basis of as much independent bench and literature data as possible. With respect to bio-oxidation, the bacterial growth rate kg was obtained directly from the growth curve experiments described above (values in Table 2). The yield coefficient Y (which does not correspond to the maximum cell yield given in Table 2) was obtained for Z5 (and a number of other column studies using the laboratory culture, not 70


detailed here) by trial and error as 2 × 1012 cells/mol Fe2+. Based on the growth experiments, it was decided to take the yield coefficient for Z8 five times smaller than for Z5, i.e., as 0.4 × 1012 cells/mol Fe2+. The simulated copper extraction curves for both Z5 and Z8 and their closeness to the experimental data are shown in Figure 5. It should be stressed that the set of parameters for these simulations was identical except for the values of kg, Y, and column height. For Z5 the fit to the experimental data is very close, while the fit of Z8 is reasonable, considering that only the bacterial growth parameters were modified. These results thus lend some credence to the correctness of the modelling approach. 100%

Copper extraction

80% 60% 40%

Z5 Z5 Z8 Z8

20%

Data Modelled Data Modelled

0% 0

10

20

30

40

50

60

70

Time on stream [d]

Figure 5. Modelled extraction curves and experimental data for Z5 and Z8

Figure 6 shows the progression of the bacterial population in solution for both simulations. Both follow the rising wave pattern observed in the experiments, and the peak levels of bacterial counts correspond closely to those observed in the experiments. The simulated and measured peak levels between Z5 and Z8 are at a ratio of approximately 5, confirming that the selection of a yield coefficient for Z8 five times smaller than for Z5 was correct. Different from the measured data, however, the simulated curves do not progress as sharp fronts. In the model, bacterial transport is modelled as advectiondiffusion with concomitant Langmuir-type (physical) adsorption. The experimental data suggests, however, that bacteria do not migrate beyond the chalcocite stage 1 zone, but accumulate where there is good substrate (i.e. ferrous and acid) supply. Thus bacterial attachment appears to be part of the growth process rather than merely a physical phenomenon. This is not reflected in the model at present. 1.E+08

Bact erial count per mL

Bact erial count per mL

1.E+08

1.E+07

1.E+06

1.E+05

1.E+07

1.E+06

1.E+05

1.E+04

1.E+04 0.0

0.5

1.0

1.5

2.0

Col umn Dept h [m]

2.5

3.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Col umn Dept h [m]

Figure 6. Modelled trends for bacterial propagation in experiments Z5 (left) and Z8

The simulations support the hypothesis that the slower extraction rate in Z8 can be explained by the much lower bacterial growth rate and yield. However, as was stated in the introduction, chalcocite stage 1 leaching in columns is normally limited by acid supply. It is of interest, therefore, to investigate the conditions under which transition from 71


time for 98% stage 1 conversion [d]

acid-limited to bacterial growth-limited leaching occurs. A number of simulations of Z8 for various values of kg (0.015 to 0.17 h–1, corresponding to doubling times of 4 to 48 h) and Y (0.4 to 10×1012 cells/mol Fe2+) were run, and the predicted times required to achieve 98% chalcocite stage 1 conversion are plotted against growth rate constant in Figure 7. From this it becomes clear that for growth rates above about 0.07 h–1 little improvement in conversion time is achieved, indicating acid-limited conditions. This holds true even for growth rates considerably below 0.07 h–1, provided that the cell yield is sufficiently low. The conditions of the native culture in Z8, however, fall clearly outside acid-limited conditions. It must be stressed that this applies strictly only to a short column scenario as given in experiment Z8. 60 y = 0.4

50

y =2

40

y = 10

30

bacterial growth limited

20

acid limited

10 0 0

0.05

0.1

0.15

0.2

growth rate const. [1/hr]

Figure 7. Time taken for 98% chalcocite stage 1 conversion as a function of various growth rates and yield coefficients, modelled for a Z8-type experiment (1.2 m column). The dashed line indicates the transition from acid-limited to growth-limited conditions

It is now of interest to investigate the effect that growth-limited cultures like native culture used in the present study would have on the rate of extraction in a full-scale heap. Retaining the calibration achieved for Z5, the HeapSim code was run to simulate two 10m heap scenarios, one with the parameters of the laboratory culture, and one with those of the Zaldívar culture. In addition, two further simulations of 10-m columns, 10 cm in diameter, were also run. Changing from a column simulation to a heap simulation involves changing the model parameter, which describes the length of stagnant pores in the bed, by as much as an order of magnitude [2]. Copper extractions are plotted against time in Figure 8. While the retarded bacteria have quite a noticeable effect on the laboratory columns, the effect is marginal in the heap scenario. As discussed above, the native culture does indeed become overall rate limiting in a narrow-bore column scenario. In unconfined heaps, however, diffusion of acid from the flowing solution into large stagnant zones (through which no solution flow occurs) governs the rate of acid supply to the reaction sites. This process is so slow that even under growth-limiting conditions the rate of bacterial oxidation is still not limiting the overall process.

72

Bioleaching Applications 100%

Cu extraction

80%

60%

40% Column fast bacteria Column slow bacteria

20%

Heap fast bacteria Heap slow bacteria

0% 0

60

120

180

240

300

360

Time on stream [d]

Figure 8. Copper extraction vs. time simulated for 10 m columns and heaps with and without bacterial rate limiting parameters 4.

CONCLUSIONS Chalcocite bioleaching proceeds in two stages, the first of which requires rapid biooxidation of ferrous to ferric, which is typically limited by the availability of acid at the reaction site rather than bacterial growth. In the present study a native culture growing in a high-TDS raffinate was compared against a laboratory consortium. It was found that in laboratory columns the native culture displayed severely restricted growth behaviour, thus governing the overall rate of leaching. Simulations with the HeapSim modelling tool, calibrated on the basis of bench-scale data, emulated the measured extraction data fairly well and thus confirmed the observed trends in bacterial numbers and propagation. In fullscale heaps, however, acid diffusion through large stagnant zones would still most likely govern the overall rate of copper extraction, despite restricted bacterial growth.

ACKNOWLEDGEMENTS The authors wish to thank Placer Dome Technical Services Limited for their generous support. REFERENCES

1. C. Garcia G., J. Binvignat T., Jaime Roco R., and J. Campos B., Randol at Vancouver ’98 Copper Hydromet Roundtable, 1998, pp 249-254. 2. J. Petersen and D.G. Dixon in Hydrometallurgy 2003, Proceedings of the 5th International Symposium Honoring Professor Ian M. Ritchie, Volume 1, C.Young, A. Alfantazi, C. Anderson, A. James, D. Dreisinger, B. Harris (eds.), TMS Publishers, Warrendale, PA, 2003, pp 351-364. 3. S.A. Bolorunduro, “Kinetics of leaching of chalcocite in acid ferric sulfate media: chemical and bacterial leaching”, M.A.Sc. Thesis, University of British Columbia, 1999. 4. O.H. Touvinen, S.I. Niemelä and H.G. Gyllenberg, Antonie van Leeuwenhoek 37, 1971, pp 489-496. 5. G. Rossi, Biohydrometallurgy, McGraw-Hill, Hamburg, 1990. 6. J. Petersen and D.G. Dixon in P.R. Taylor (Ed.) EPD Congress 2002, TMS Publishers, Warrendale, PA, 2002, pp 757-771. 73


7. M. Nemati, S.T.L. Harrison, G.S. Hansford and C. Webb, Biochemical Engineering Journal 1, 1998, pp 171-190.

74


Bacterial leaching studies of a Portuguese flotation tailing M.C. Costaa, N. Carvalhoa, N. Iglesiasb, I. Palenciab a

Faculdade de Engenharia de Recursos Naturais, Universidade do Algarve, Campus de Gambelas, 8005-139 Faro, Portugal b Departamento de Ingeniería Química, Facultad de Química, Universidad de Sevilla, 41071 Sevilla, Spain

Abstract Two bioleaching processes - a direct bioleaching and an indirect bioleaching with chemical and biological separate stages - have been applied to the flotation tailings of a chalcopyrite ore. In the present paper the influence of experimental parameters is investigated and the obtained results are described and compared in order to test the efficiency of these processes for the recovery of copper and zinc. Keywords: bacterial leaching, copper, zinc, tailing 1.

INTRODUCTION Neves Corvo mine – one of the major mines of the Iberian pyrite Belt, located in Southern Portugal – produces high-grade copper (and tin) sulphide ores which are submitted to flotation processes in order to obtain copper and zinc concentrates for sale to international smelters. The content of copper and zinc in the flotation residue (0.85 and 0.64%, respectively) and the environmental problems caused by its accumulation justify the development of a biohydrometallurgical process, which could be at the same time an economical viable approach for its treatment and for the minimization of its environmental impact. Bacterial leaching can be a potential treatment for this type of residue. Bacterial leaching is an environmentally safe and flexible alternative. It is nowadays well established that the most widely used bacterium – Thiobacillus ferrooxidans – is able to oxidize Fe2+, S°, as well as other reduced inorganic sulphur compounds. Two mechanisms have been established: the direct mechanism that requires physical contact between bacteria and the particles of the metal sulphide (recently it has been suggested that this mechanism should be renamed to “contact” mechanism [1]) and the indirect mechanism, according to which the bacteria oxidize ferrous ion to the ferric state, thereby regenerating the ferric ion required for chemical oxidation of the sulphide mineral. According to the new integral model for bioleaching, metal sulphides are degraded by a chemical attack of iron (III) ions and/or protons on the crystal lattice [2]. The mechanism of degradation is determined by the mineral structure: pyrite via a thiosulphate mechanism with acid production, and chalcopyrite, sphalerite and galena via a polysulfide mechanism. 75


The use of bioleaching for copper recovery is usually limited because chalcopyrite is very refractory to oxidation in acid media and so presents problems to bioleaching [3]. Moreover, when bioleaching is carried out in a single reactor several phenomena take place that limit to a great extent the rate of the ferrous biooxidation. Firstly, the mineral particles exert an important abrasive effect on the bacteria with a partial breakdown of them, which has two negative effects on the process kinetics: the active bacterial population decreases and the resulting organic matter would reduce or inhibit the bacterial growth [4]. The slow kinetics results in residence times of several days, even weeks, which can limit its application. In addition, the values of pH, temperature and the use of catalysts are conditioned to those values compatible with bacterial growth. Several approaches, such as the use of thermoplilic microorganisms [3] and/or the addition of catalysts [3-6] have been attempted in order to overcome these problems. Copper concentrates can be effectively bioleached by performing chemical and biological oxidation in separate steps and using silver as a catalyst in the chemical oxidation. For copper-zinc concentrates both metals can be recovered to a great extent by conducting the ferric leaching in two stages, the first one without silver to recover zinc and the second one with silver as a catalyst to extract copper [5]. However, until now this approach has not been applied to flotation tailings in which the content of both copper and zinc are low and they are finely disseminated in the sulphide matrix. The main concern relative to the application of the indirect bioleaching with chemical and biological separate steps to the treatment of flotation tailings is that the kinetics and extent of the leaching reactions and the amount of catalyst required are economically competitive with the direct bioleaching of the residue. 2.

EXPERIMENTAL

2.1 Materials The tailings used in this work were produced by the copper/zinc selective flotation plant of the Neves Corvo mine in Southern Portugal. Its chemical composition is shown in Table 1. Table 1. Chemical composition of the tailings Element

Content

Element

Content

Cu (%)

0.85

As (ppm)

6438

Zn (%)

0.64

Sb (ppm)

654

Fe (%)

22.9

Bi (ppm)

114

S (%)

23.7

Hg (ppm)

21

Pb (%)

0.21

Sn (%)

0.27

The direct bioleaching experiments were performed using a mixed culture of Thiobacillus ferrooxidans and other related bacteria isolated from Aljustrel mines drainage waters (Portugal) and routinely maintained at 34ºC on a modified Silverman and Lundgren 9K nutrient medium at pH=2.0. Chemicals were of reagent grade and all solutions were made up with distilled water.

76


2.2 Procedure 2.2.1 Direct bioleaching experiments In general, bioleaching experiments were carried out in 250 cm3 conical flasks with 90 cm3 of 9K nutrient medium (without Fe(II)) at pH=2.0, 5 g of tailings (previously washed with ethanol to remove traces of flotation reagents that may inhibit bacterial growth) and 10 cm3 of inoculum. The incubated flasks were on a thermostatted bath shaker at 34ºC and at a constant speed of 280 min-1. The pH was maintained at a constant value by adding 3 M H2SO4. Suspended solids were allowed to settle and liquid samples were drawn daily for copper, zinc and total iron analysis. pH and redox potential were measured by a Pt electrode and an Ag/AgCl electrode as reference. Experiments with an ethanol solution containing 2% (v/v) of thymol instead of inoculum were used as controls. The effect of variables such as pulp density, air supply and particles size was investigated. A stirred tank bioleaching experiment was also carried out in a one dm3 glass jacketed vessel. One dm3 of the bioleaching solution was placed in the reactor vessel and heated to the working temperature (34ºC). The experiment was initiated by the addition of 50 g of dried solid. The agitation was obtained with a mechanical stirrer at 500 rpm. Air supply was provided (100 dm3.min-1). 2.2.2 Ferric sulfate leaching experiments Experiments were carried out in 250 cm3 conical flasks with 150 cm3 of ferric sulfate solution. The flasks were continuously agitated at 280 min-1 on an orbital shaker supplied with a forced air circulation thermostat. Ferric sulfate solutions were first heated to the desired temperature and the reaction was initiated by adding a dried mineral sample. In all tests, the water losses due to evaporation were taken into account during recovery calculations. At the end of the experiment, the slurry was filtered using 0.45 µm Millipore filters and the residue was washed with distilled water, dried and stored in a desiccator. The leach liquor was analyzed for copper, zinc, total iron and ferrous iron. In catalytic tests, the leaching medium consisted of ferric sulfate solutions with silver. An aliquot of a solution of silver sulfate in aqueous sulfuric acid at pH 1.40 containing 300 ppm of silver was added to the ferric sulfate solution. The amount of catalyst is expressed as milligrams of Ag+/gram of concentrate. Unless otherwise stated, the experimental conditions were: initial pH of solution 1.40, ferric iron concentration 12 g.dm-3 and duration of the test 8 hours [7]. Because all the experiments were carried out at low ferric iron concentration (12 g.dm-3) in batch systems, the studied pulp solids concentration have to be low. In a continuous operation the pulp solids concentration of leaching might be higher than the values considered in this study. The effect of the variables such as ferric iron concentration, temperature and amount of catalyst was investigated. Copper and zinc soluble in acid media were determined by performing a test in sulphuric medium at pH 1.40 without ferric iron. Copper, zinc and iron in leaching and bioleaching solutions were analysed by flame atomic absorption spectroscopy (AAS). Ferrous iron concentration was determined by standard potassium dichromate solution in an automatic titrator.

77


3.

RESULTS AND DISCUSSION Pyrite was identified by X-ray diffraction (XRD) as the predominant mineral in the solid material. Considering that the residue is coming from the flotation of chalcopyrite and sphalerite bearing ores, the presence of those minerals, although in much lower contents (and therefore not identified by XRD), should be taken into account. The presence of some oxides should be considered as well. 3.1 Direct bioleaching experiments Bioleaching tests were performed in order to investigate the possibility of using a biohydrometallurgical treatment for the recovery of copper and zinc from the flotation tailing. Therefore, preliminary studies were carried out taking into account the influence of some relevant experimental parameters on metals extraction. 3.1.1 Influence of pulp solids concentration Pulp solid concentration is an important parameter that must be considered in bioleaching. Thus, in order to check its influence, bioleaching experiments were carried out in 250-cm3 conical flasks using 5, 10, 15 and 20 g of solid material in 100-cm3 solution. The evolution of the redox potential in the bioleaching solution with 5% (w/v) pulp solids concentration as a function of time denotes a considerable increase which is consistent with the ferrous ion oxidation to ferric ion and thus, it confirms bacterial growth in the presence of the tailing. The amounts of copper and zinc recovered as a function of pulp solids concentrations are shown in Figure 1.

Figure 1. Effect of pulp solids concentration on copper and zinc extraction. Conditions: 34ºC, pH=2.0

This set of experiments shows that copper and zinc recoveries depend on the pulp solids concentration. A fast initial bioleaching rate is observed in all cases, after which it becomes very slow. The best copper and zinc recoveries, 36% and 77%, were achieved with the lower solids concentration. Although the production capacity increases with the use of high solids concentrations, when the solid concentration exceeds a value between 5 78


to 10%, the rate of metals leaching decreases probably due to important shearing stresses and lower dissolved oxygen concentration. In addition, when high solids concentration are used, the concentration of metals is high which may inhibit bacterial growth. In fact, experiments carried out with solids concentration higher than 5% showed a slight increase in the redox potential probably due to lower bacteria activity, which lead to lower metals recovery. Iron concentration also increased with time. At the end of the experiment performed with 5% (w/v) about 60% of iron was extracted probably from chalcopyrite and pyrite dissolution. The iron recovered in the rest of conditions varied considerably and was always lower than 30%. In the control experiments copper and zinc recoveries slightly increased with time and varied from 18% to 22% for copper and from 25% to 31% for zinc. These results show the important role of bacteria in the dissolution of those metals. In order to improve metals recovery other experiments with air supply and with previous grinding were performed. 3.1.2 Influence of air supply Figure 2 compares copper and zinc extraction in the absence and in the presence of air supply (50 dm3.min-1). Air supply does not have considerable influence on metals extraction: a slight increase was observed for copper and a small negative effect was detected for zinc.

Figure 2. Effect of air supply on copper and zinc extraction. Conditions: solids concentration 5% (w/v), 34ºC, pH=2.0 3.1.3 Influence of particle size The tailing was submitted to a granulometric separation after grinding. Five different fractions were separated: < 0.106 mm, 0.106-0.200 mm, 0.200-0.500 mm, 0.500-1.00 mm, 1.00-2.00 mm. The amount of copper, zinc and iron in each fraction was determined and the corresponding values were used for metals recovery calculations (Figure 3). As expected, grinding has a positive influence on zinc recovery. More than 95% of zinc was extracted in all experiments regardless of the particle size. On the other hand, copper recovery does not seem to be affected by grinding. The results also suggest that grinding has a positive effect on the rate of metals recovery, particularly for zinc since more than 95% of metal was dissolved in the first 250 hours from the fractions with particle size lower than 0.5 mm. 79


50

Zn extraction (%)

Cu extraction (%)

100

40 30 20 10

80

[1.0-2.0]mm [0.5-1.0]mm

60

[0.2-0.5]mm [0.1-0.2]mm

40

< 0.1 mm

20 0

0 0

250

500

750

0

1000

250

500

750

1000

Time (h)

Time (h)

Figure 3. Effect of particle size on copper and zinc extraction. Conditions: pulp solids concentration 5% (w/v), 34ºC, pH=2.0 3.1.4 Stirred tank bioleaching tests Considering the usual limitations of the shake flask technique (mainly because of continuously changing conditions which lead to a difficult control of experimental variables) stirred tank bioleaching experiments in similar conditions were carried out. Copper and zinc extractions are presented in Figure 4. A more efficient aeration and a complete mixing of suspended solids lead to a higher zinc recovery: 92% at 500 hours against only 77% at 623 hours in the shake flask experiment. An improvement of the copper leaching rate was also observed, but only a small increase in its maximum extraction (42% against 35%) was detected.

Extraction (%)

100 80 60

Cu

40

Zn

20 0 0

500

1000

Tim e (h)

Figure 4. Copper and zinc extraction (stirred tank bioleaching). Conditions: pulp solids concentration 5% (w/v), 34ºC, pH=2.0

In these tests the amount of iron recovered was 30% at 400 hours. After that time a decrease in iron concentration was observed probably due to jarosite precipitation which usually takes place in bioleaching of sulphide materials [8]. The high zinc recoveries obtained in the bioleaching experiments were probably due to the presence of chalcopyrite and pyrite. Sphalerite chemical and biological dissolution is reported to be improved in the presence of chalcopyrite and pyrite [9], due to galvanic interactions. Contrarily, the low results of copper dissolution obtained in all tests are probably due to incomplete oxidation of chalcopyrite by bacteria (bioleaching rates of Cuoxides are reported to be faster [10]). This phenomenon is well known and has been attributed to the formation of elemental sulphur on the mineral surface. 80


3.2 Ferric sulphate leaching experiments The leaching of the flotation tailing was initially studied at 70ºC at two different pH values, 1.25 and 1.40. Results from these tests are shown in Figure 5 in which copper and zinc soluble in both acid media (without ferric iron) are also shown. Both copper and zinc soluble in acid medium are very high, 26.5% to 28.0% of copper and 45.4% of zinc. The presence of ferric iron does not influence copper extraction, even it decreases slightly, and has a noticeable effect on zinc extraction, that markedly increases. Figure 6 shows the ferrous iron concentration and the final pH of the leaching liquor. In the absence of ferric iron there is a net consumption of acid, as the pH increases, and a small ferrous iron production (0.76 and 0.68 g/dm3 for initial pH values of 1.25 and 1.40 respectively). The presence of 12 g/dm3 of Fe (III) leads to a net production of acid, as the pH decreases, and to a high production of ferrous iron (6.11 and 6.03 g/dm3 for initial pH values of 1.25 and 1.40 respectively). These results, together with those shown Figure 5, indicate that an acid solution with 12 g/dm3 of ferric as ferric sulphate is not able to dissolve the non-acidsoluble copper of the flotation tailing (the majority being as chalcopyrite and other copper sulfides) and it is effective for the dissolution of the sphalerite present. The high ferrous iron production together with the net production of acid in ferric leaching suggests that some reductant component whose dissolution produces acid is being dissolved. This component could be pyrite. 100 90.9

90

82.8

80

Extraction (%)

70 %Cu, pHo=1,25

60

%Cu, pHo=1,40

50

45.4

45.4

%Zn, pHo=1,25 %Zn, pHo=1,40

40 30

28.0

25.9

26.5

24.5

20 10 0 [Fe(III)]=0g/L

[Fe(III)]=12g/L

Figure 5. Effect of the pH on the copper and zinc extraction with and without ferric iron. Conditions: pulp solids concentration 2% (w/v), 70°C, 8 hours 7 6

[Fe(II)] (g/L)

5 4

Fe(II) pHf

3 2

1,52

1,83 1,22

1,33

1 0

[Fe(III)]=0g/L pHo=1,25




Figure 6. Ferrous iron production and final pH in tests with and without ferric iron. Conditions: pulp solids concentration 2%(w/v), 70°C, initial pH 1.25 or 1.40, 8 hours 81


3.2.1 Effect of the ferric iron concentration The effect of the ferric iron concentration on the copper and zinc extractions was studied over the range 8-12 g/dm3 at 70ºC. The leaching results, shown in Figure 7, indicate that an increase of the ferric iron concentration in this range does not affect the copper extraction and has a positive effect on zinc extraction. 100 90 80 Extraction (%)

70 60

%Cu

50

%Zn

40 30 20 10 0 0

2

4

6

8

10

12

[Fe(III)] (g/L)

Figure 7. Effect of the initial ferric iron concentration on the copper and zinc extraction. Conditions: pulp solids concentration 2% (w/v), pH 1.40, 70°C, 8 hours 3.2.2 Effect of the pulp solids concentration The effect of the pulp solids concentration on the copper and zinc extractions was studied over the range 2% to 8% (w/v) at 70ºC. Figure 8 shows that as the pulp solids concentration increases both copper and zinc extractions decrease. This effect is not very important in the studied range of pulp density. The increase of the ferrous iron concentration as the pulp solids concentration increases suggests that the decrease in metal extraction observed at high pulp solids concentration could be due to the depletion of ferric iron in those conditions. 100

12

90 10

70

8

60 50

6

40 4

30 20

[Fe(II)] (g/L)

Extraction (%)

80

%Cu %Zn [Fe(II)]

2

10 0

0 0

1

2

3

4

5

6

7

8

9

Pulp solids concentration (w/v %)

Figure 8. Effect of the pulp solids concentration on the copper and zinc extraction and on the ferrous iron production. Conditions: pH 1.40, 70°C, 12 g/L Fe3+, 8 hours 82


3.2.3 Effect of the amount of catalyst The effect of the amount of silver on the copper and zinc extraction was studied over the range 0-1 mg silver/g of flotation residue at 70ºC and 2% (w/v) of pulp solids concentration. The leaching results, shown in Figure 9, indicate a marked effect of the presence of catalyst on the copper extraction. There is a noticeable increase in copper extraction as silver amount increases from 0.2 to 1 mg/g of flotation residue. Zinc extraction decreases as silver amount increase. These results are in agreement with previous results about the catalytic effect of silver on copper and zinc extraction from a mineral that contains both metals as sulphides [4]. 100 90

Extraction (%)

80 70 60

% Cu

50

% Zn

40 30 20 10 0 0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

Ag(I) (mg Ag/g residue)

Figure 9. Effect of the amount of catalyst on the copper and zinc extraction. Conditions: pulp solids concentration 2% (w/v), pH 1.40, 70°C, 12 g/L Fe3+, 8 hours 3.2.4 Effect of temperature Figure 10 shows copper and zinc extractions in the absence and in the presence of ferric iron from tests carried out at 34ºC and 70ºC. As it was observed at 70ºC, the presence of ferric iron does not have influence on the copper extraction at 34ºC and has a positive influence on the zinc extraction. Temperature of 34ºC was chosen since it is the temperature of the direct bioleaching experiments and with the aim of comparing the efficiency of chemical and bacterial leaching. The increase of temperature from 34ºC to 70ºC influence both copper and zinc extraction as it can be observed in Figure 10. 100

90.9

90 80

Extraction (%)

70 %Cu 70°C

60

%Cu 34°C

50

45.4

%Zn 70°C %Zn 34°C

40 30 20

26.5

26.0

25.9 19,7 14.0

14.5

10 0 [Fe(III)]=0g/L

[Fe(III)]=12g/L

Figure 10. Effect of the temperature on the copper and zinc extraction with and without ferric iron. Conditions: pulp solids concentration 2% (w/v), pH 1.40, 8 hours 83


4.

CONCLUSIONS More than 90% of zinc can be recovered by both chemical and biological leaching of the flotation tailing, while only 40% of copper can be recovered by direct bioleaching and 20 to 30% by ferric sulfate leaching. The whole experimental conditions needed to reach those recoveries either by chemical or biological leaching (temperature, pulp solids concentration, leaching time, reagents consumption) should be further evaluated in order to establish the most favourable process. In general, the obtained results are in agreement with previous studies showing that sphalerite is more easily dissolved than chalcopyrite. However, other bioleaching tests could be performed to extend this investigation to the effect of other relevant experimental parameters (i.e. amount of inoculum, initial Fe(II) concentration, pH), which can eventually have a positive influence on copper recovery. Other approaches, such as the use of catalysts, extreme thermophiles or regrinding of the leach residue, should be considered carefully, taking into account the low copper and zinc grade of this type of material. REFERENCES

1. H. Tributsch, In Biohydrometallurgy and the environment towards the mining of the 21st century. Elsevier (1999) 51. 2. W. Sand, T. Gehrke, P.G. Jozsa, A. Schippers. In Biohydrometallurgy and the environment towards the mining of the 21st century. Elsevier (1999) 27. 3. E. Gómez, A. Ballester, M.L. Blásquez, F. González, Hydrometallurgy, 51 (1999) 37. 4. F. Carranza, I. Palencia, R. Romero, Hydrometallurgy, 44 (1997) 29. 5. R. Romero, I. Palencia, F. Carranza, Hydrometallurgy, 49 (1998) 75. 6. F. Carranza, N. Iglesias, Min. Eng., 11 (4) (1998) 385. 7. Palencia, I., Romero, R., Carranza, F., Hydrometallurgy, 48 (1998) 101. 8. J. Frenay, X. Ciechanowski, Cours de Biometallurgie, Dep. de Métallurgie et Traitment des Minérais, Université de Liége, 1996. 9. A.P. Mehta, L.E. Murr, Biotech. Bioeng. 24 (1982) 919 10. H.T. Olli, In Microbial mineral recovery. Mc Graw Hill (1990) 67.

84


Bacterial tank leaching of zinc from flotation tailings V.V. Panina*, E.V. Adamova, L.N. Krylovaa, T.A. Pivovarovab, D.Yu. Voronina, G.I. Karavaikob a

Moscow State Institute of Steel and Alloys (Technological University), Leninsky Prospect, 4, Moscow, 119049, Russia b Institute of Microbiology, Russian Academy of Sciences, Prospect 60-let Octyabrya, 7/2, Moscow, 117312, Russia

Abstract The parameters of tank bacterial-chemical leaching of stale flotation tailings containing 5.6% Zn, 12.96% S and 11.6% Fe were investigated. The particle size of the tailings was 65% minus 44 µm. The main minerals were pyrite (25-30%), sphalerite (78%), pyrrhotite (7-10%), marcasite, chalcopyrite, galena (5-7%) etc. The gangue rock (55%) was predominantly present in the silicate form. Zinc was leached under continuous conditions at 28-30°C in agitated tanks with a pulp density of 16.7, 28.6 and 40.0% of solids. Acidithiobacillus ferrooxidans strain TFI was isolated and cultivated on zinc flotation tailings and adapted to zinc ions concentrations in the liquid phase of up to 40 g/L. Zinc extraction by leaching at a pulp density of 40% of solids with the return of 10% of the solution from the last tank reached 87.12% while the concentration of zinc ions in the leach solution was in the range of 31.4-32.4 g/L. The solids throughput was increased 8 fold, pulp flow rate 2.4 fold and overall leaching residence time was reduced 2.7 fold as compared to leaching at a pulp density of 16.7% of solids. Keywords: bioleaching, sphalerite, flotation tailings, acidithiobacillus, pulp density, process flowsheet 1.

INTRODUCTION Huge amounts of technogenic resources including flotation tailings are being accumulated worldwide in the areas adjacent to existent and closed mineral processing plants. The higher grades of ores processed earlier together with imperfection of the technology used are the main reasons why the stale flotation tailings are relatively rich with valuable components. In certain cases they are comparable by tenor with some of the ores mined nowadays and therefore can be reprocessed to extract metals. Generally the use of expensive pyrometallurgical and pressure leaching technologies to extract valuable components from low-grade technogenic resources is not economically viable. In such cases the technologies with lower operating costs come to the foreground. Primarily these are the cost-effective hydrometallurgical technologies and among them bacterial-chemical leaching should be considered [1-4]. *

Corresponding author: Tel./Fax: + 7-095-951-21-39, E-mail address: [email protected] (V.V. Panin)

85


The aim of this work was to investigate and develop the technological parameters of bacterial-chemical leaching of zinc from the flotation tailings. These parameters include solids content, specific pulp flow rate and flowsheet configuration. 2.

MATERIALS AND METHODS Quantitative chemical analysis showed that the stale zinc-containing flotation tailings with a particle size of 65% minus 44 µm contained 5.6% Zn, 12.96% S and 11.6% Fe. The gangue rock (55%) was predominantly present as silica – 32-38%, mica (biotite, muscovite) – 10-12%, graphite – 5-7% and Fe silicates. The main minerals were pyrite (25-30%), sphalerite (7-8%), pyrrhotite (7-10%), marcasite, chalcopyrite, galena (5-7%), sulphosalts, Cu sulfate, secondary Fe minerals, magnetite and haematite (partially substituted by Fe hydroxides). More than 70% of valuable minerals were intergrown with other valuable and gangue minerals. Zinc in flotation tailings was present both in the sulfide form (sphalerite) and in the oxide form (sulfates, oxides, silicates and ferrites). The major amount of zinc was found to occur in particle size range of 20-44 µm. Bacterial-chemical leaching of zinc flotation tailings was performed under continuous conditions at 28-30°C using three agitated tanks connected in a series. The volume of each tank was 1600ml with the mixer speed of 400 rpm. Aeration rate during the experiments was 1L/min per 1L of pulp volume. Acidithiobacillus ferrooxidans strain TFI was routinely isolated and cultivated on zinc flotation tailings as the energy substrate. Adaptation efforts resulted in obtaining a strain tolerant to 40 g/L of zinc ions in solution. The cell concentration was evaluated by both Lowry protein measurement [5] and end-point tenfold dilutions method. Strain identification in liquid phase of leaching pulp was performed by analyzing chromosomal DNA structure using pulsed field gel electrophoresis method [6]. Cells activity in leaching pulp and solutions was controlled by manometric method (O2 consumption) and by measuring the rate of ferrous iron oxidation to ferric iron. Total iron concentration (ferrous plus ferric iron) in solution, as well as zinc concentration in solution and in leach residue (after acid decomposition), were determined by atomic absorption spectrophotometry (Perkin-Elmer mod. 3100). Separately, ferric and ferrous iron concentrations in solution were determined by complexometric titration. Redox potentials were measured using a platinum electrode (combined with a silver/silver chloride reference electrode) and converted to Eh values (relative to a hydrogen reference electrode). 3.

RESULTS It is common practice to leach and oxidize sulfide concentrates in the presence of bacteria at a pulp density of 16-20% of solids. In the present work zinc was bacterially leached at pulp densities of 16.7, 28.6 and 40% of solids. Experimental results are reported in Table 1. The cells concentration and activity (measured as O2 consumption) together with ferric iron and zinc concentrations in leach solution increased with an increase in the pulp density. Experimental mass balance of bacterial-chemical leaching of zinc from the stale flotation tailings at different pulp densities is shown in Table 2. The dynamics of zinc leaching at different pulp densities are shown in Fig. 1. High zinc concentrations in leach

86


solutions were achieved by increasing the pulp density; zinc extraction was still at a high level (87.12%). Zinc content in solid leach residue was reduced from 5.6% to 0.75-0.87%. Table 1. Technological parameters of bacterial-chemical leaching of zinc from the flotation tailings at different solids content, % Factors Leach solution pH value Leach solution Eh value (mV) Ferric iron concentration (g/L) Zinc concentration (g/L) Cells respiratory activity (µL O2/mL⋅hour) Cells concentration (cells/mL) Flow rate (tank volume/hour) Pulp throughput (ml/day) Solids throughput (g/day) Single tank residence time (hour)

16.7% 1.5-2.0 650-750 5.5-6.0 8.7-10.0 20.0-44.0 108-109 0.013 500 100 96

Values 28.6% 1.4-2.0 700-750 10.0-14.0 13.2-19.0 36.0-40.0 109-1010 0.031 1200 480 40

40.0% 1.3-2.0 700-750 13.6-15.5 31.4-32.4 40.0-56.0 1010-1011 0.031 1200 804 35

Table 2. Mass balance of bacterial-chemical leaching of zinc from the flotation tailings at different solids content, % Tank

Leach residue yield (%)

Zinc concentration in solution (g/L)

Zinc content in leach residue (%)

Zinc extraction (%)

1.40 1.20 0.75 0.75

77.50 81.80 88.90 89.30

2.52 1.51 1.01

58.96 76.97 85.01

Pulp density 16.7% solids Feed preparation tank Leaching tank 1 Leaching tank 2 Leaching tank 3

90.0 85.0 83.0 80.0

8.68 9.16 9.96 10.00 Pulp density 28.6% solids

Feed preparation tank Leaching tank 1 Leaching tank 2 Leaching tank 3

95.0 90.0 86.0 83.0

13.21 17.40 19.04 Pulp density 40.0% solids

Feed preparation tank Leaching tank 1 Leaching tank 2 Leaching tank 3

95.0 89.8 85.7 82.8

31.36 32.39 32.37

1.02 0.89 0.87

83.64 86.38 87.12

Feed (flotation tailings)

100.0

⎯

5.6

100.0

High redox potentials and low pH values of the solution were observed during bacterial-chemical oxidation. At a pulp density of 16.7% of solids zinc extraction in the first tank was significantly higher than at a pulp density of 28.6% due to the extended residence time (96 hours vs. 40 hours). Increasing the pulp density to 28.6% with higher flow rate led to a decrease of zinc extraction in the first two tanks but in third tank extraction reached 85%. At a pulp density 87


of 40.0% of solids high zinc extraction and high flow rates in all tanks were achieved by returning 10% of the leach solution from the last tank to the first one. Returned liquid phase was characterized by high concentrations of active cells and ferric iron. It is possible at higher pulp densities to increase the flow rates and hence the overall leaching process throughput while keeping fast enough oxidation. Leaching at a pulp density of 40.0% of solids increased the solids throughput 8 fold, pulp flow rate 2.4 fold and decreased overall leaching residence time 2.7 fold as compared to the pulp density of 16.7% of solids. Mineralogical analysis of flotation tailings leach residues revealed significant phase transformations and the presence of coarse aggregates of up to 300 µm in size cemented by newly formed substance. Inside these aggregates relic mineral inclusions were found: sphalerite grains (10-15µm), pyrite, graphite and gangue. Only gangue minerals, undestructed pyrite and secondary Fe minerals were observed in "free" form. No "free" sphalerite grains were found in leach residue.

Figure 1. The dynamics of zinc leaching from flotation tailings at different pulp densities

The substance cementing the mineral aggregates was formed by incompletely destructed gangue rock fragments, whity-yellow and gray colored matter (elemental sulfur and bacterial metabolites) and unidentified porous material with sub-micron intermetallic inclusions. Mineralogical data on initial flotation tailings and on bacterial-chemical leach residues correlated the results of chemical analyses and gave evidence that "free" zinc sulfide grains and some other sulfide minerals were leached almost completely. Sphalerite bacterial-chemical oxidation mechanism can be described according to Fowler and Crundwell [7-9] by the following reaction:

ZnS + Fe2 ( SO4 ) 3 → ZnSO4 + 2 FeSO4 + S 0

(1) Iron oxidizing acidophilic bacteria regenerate ferric iron and oxidize elemental sulfur:

4 Fe 2+ + 4 H + + O2 → 4 Fe 3+ + 2 H 2 O

(2)

S 0 + H 2 O + 1.5O2 → H 2 SO4

(3)

88


As a result of the reaction according to Eqn. 3, the probability of sphalerite passivation by elemental sulfur layer is reduced greatly; the addition of sulfuric acid is not necessary because the latter is generated as a product of sulfur oxidation process. Surface-active properties of bacterial metabolites in leaching pulps have a technological significance since they increase the solids settling rate during thickening and dewatering. Since iron is present in bacterial-chemical leach solutions in the oxidized ferric form the lime consumption for Fe hydroxide precipitation prior to zinc extraction from the solution will be lower. 4.

CONCLUSIONS It has been shown in this work that unlike the common practice of leaching and oxidizing sulfide concentrates in the presence of bacteria at pulp densities of 16-20% of solids it is possible to bacterially leach zinc at a pulp density of 40% of solids with high technological results. Raising the pulp density in tank bacterial-chemical zinc leaching up to 40% of solids with the return of 10% solution from the last tank increased solids throughput 8 fold, pulp flow rate 2.4 fold and allows to reduce 2.7 fold the overall leaching residence time as compared to the leaching at a pulp density of 16.7% of solids. ACKNOWLEDGEMENTS This research was carried out with the support of Federal Scientific and Technical Program "Investigations and Developments on Priority Directions of Science and Technology" (years 2002-2006) on the subject "Biotechnology in mining and processing of mineral resources". REFERENCES

1. Sheveleva L.D., Abakumov V.V., Korkin B.I., Bishev L.Z. and Karavaiko G.I., 1995. Development of new technology for reprocessing of concentrating mill tailings. Tsvetnye Metally, 12, 23-26. 2. Panin V.V., Adamov E.V., Karavaiko G.I., Khamidullina F.G. and Voronin D.Yu., 1999. Use of the Bacterial Leaching Technology in Processing of Refractory CopperZinc Ores. Tsvetnye Metally, 5, 9-11. 3. Karavaiko G.I., Sedelnikova G.V., Aslanukov R.Ya., Savari E.E., Panin V.V. and Adamov E.V., 2000. Biohydrometallurgy of Gold and Silver. Tsvetnye Metally, 8, 2026. 4. Pol’kin S.I., Adamov E.V., Panin V.V. Technology of Bacterial Leaching of NonFerrous and Rare Metals. Nedra, Moscow, 1982, 288 p. 5. Lowry O.H., Rosenbrough N.J., Farr A.L. and Randell R.J., 1951. Protein Measurement with the Folin Phenol Reagent. Journal of Biological Chemistry, 193, 265-275 6. Kondratyeva T.F., Muntyan L.N. and Karavaiko G.I., 1995. Zinc- and ArsenicResistant Strains of Thiobacillus ferrooxidans have Increased Copy Numbers of Chromosomal Resistance Genes. Microbiology, 141 (5), 1157-1162. 7. Fowler T.A. and Crundwell F.K., 1998. Leaching of Zinc Sulfide by Thiobacillus ferrooxidans: Experiments with a Controlled Redox Potential Indicate No Direct Bacterial Mechanism. Applied and Environmental Microbiology, 64 (10), 3570-3575. 8. Fowler T.A. and Crundwell F.K., 1999. Leaching of Zinc Sulfide by Thiobacillus ferrooxidans: Bacterial Oxidation of the Sulfur Product Layer Increases the Rate of 89


Zinc Sulfide Dissolution at High Concentrations of Ferrous Ions. Applied and Environmental Microbiology, 65 (12), 5285-5292. 9. Driessens Y.P.M., Fowler T.A. and Crundwell F.K. A comparison of the bacterial and chemical leaching of sphalerite at the same solution conditions. In: Biohydrometallurgy and the Environment toward the Mining of the 21st century. R Amils and A. Ballester (eds.) Elsevier, 1999. Part A, pp.201-208.

90


Behaviour of elemental sulphur in the biohydrometallurgical processing of refractory gold-sulfide concentrates of various mineral types Sedelnikova G.V.*, Savari E.E. Central Research Institute of Geological Prospecting for Base and Precious Metals (TsNIGRI), Varshavskoe sh. 129 b, 117545 Moscow, Russia Abstract The kinetics of oxidation of gold-sulfide concentrates of various mineral types with the use of Acidithiobacillus ferrooxidans monoculture and mixed culture of mesophilic bacteria has been studied. It is shown that the oxidation of the main sulfide minerals arsenopyrite, pyrite and pyrrhotite is more efficient when applying mixed cultures. The oxidation of sulfide concentrates is accompanied by release of elemental sulfur. Most amount of elemental sulfur is derived in the course of bacterial oxidation of pyrrhotitepyrite-arsenopyrite concentrates. With actually complete oxidation of pyrrhotite, arsenopyrite and some pyrite, the process of elemental sulfur biooxidation is not terminated. Even additional bacterial leaching of pyrrhotite concentrate within 7 days does not completely oxidize elemental sulfur. As elemental sulfur has a negative effect on the process of gold recovery by cyanidation of biooxidation residues and leads to high consumption of sodium cyanide, some ways of additional oxidation of elemental sulfur by aeration in lime environment or electrolytic treatment were investigated. The research outcomes were taken into account in design and construction of the first commercial plant in Russia at the Olimpiada gold deposit.

Keywords: gold-sulfide concentrates, biohydrometallurgy, elemental sulfur 1.

INTRODUCTION Since 1986, the bacterial leaching of refractory gold ore and concentrates has been employed for gold recovery at a commercial scale. At present, more than 10 commercial plants are operating in the world: in Australia, Ghana, South Africa, Brazil, Peru, China etc. In Russia, the first plant was commissioned in 1997 at the Olimpiada gold deposit, the Krasnoyarsk territory, its daily capacity being 1 t concentrate with further increasing up to 400 t/day in 2001. The bacterial leaching is carried out around the year under severe climatic conditions, at winter temperatures of -25 to -45°C.

* Corresponding author: Fax/ Phone (095) 113-68-22, E-mail: [email protected]

91


The known BIOX® technology developed by Gencor, first tested at Fairview and later on applied at the other plants uses a mixed culture of mesophilic bacteria and biooxidation plants are operating within the range of 40° to 45°C. The other technology - Bac Tech process employs moderately thermophiliс bacteria at about 50º C at the Youanmi commercial plant in Australia [1]. Taking into account cold climate and refractory ore reserves in Russia which are to be developed, this paper considers the process of biooxidation of refractory gold-sulfide concentrates of various mineral types with the use of mesophilic bacteria at 28-32°C. While elaborating technology of treating the pyrrhotite - pyrite - arsenopyrite concentrates from the Olimpiada deposit, we faced some problems caused by release of a great amount of elemental sulfur as a result of biochemical oxidation of sulfides and first of all of pyrrhotite. It is known that S° [2] and S2, S2O32-, SO32- sulfur-bearing anions [3-6] produce (CSN)- and cause high cyanide consumption in the course of cyanidation of biooxidation residues. At the plants using biooxidation technology for the treatment of refractory gold concentrate the cyanide consumption attains 30 kg/t of concentrate [3]. Therefore, several methods of reducing cyanide consumption are applied, such as: increase of pulp density [4, 6], pre-leaching lime aeration [7, 8], pre-treatment in sodium hydroxide solution [9], electrolytic treatment [10]. This paper describes the results of study of the elemental sulfur behavior in the process of bacterial oxidation of various mineral types of concentrates and recommends pre-aeration in the lime environment and electrolytic treatment of pulp prior to cyanidation in order to reduce cyanide consumption and to increase gold recovery from the biooxidation residues. 2.


2.1 Sulfide concentrates Concentrate samples were produced by mineral processing of refractory gold ore of Russian deposits. Four samples of refractory gold - sulfide concentrates were studied under laboratory conditions, such as: gravity concentrate (1) from the Nezhdaninka deposit, Republic of Sakha -Yakutia, gravity - flotation concentrate (2) from the Albazin deposit, the Khabarovsk region, and two flotation concentrates (3) and (4), accordingly, from Nezhdaninka, Republic of Sakha – Yakutia, and Olimpiada, the Krasnoyarsk region. The main gold-bearing sulfide minerals in concentrates are arsenopyrite and pyrite. The concentrate (4) contains also pyrrhotite. The quantitative predominance of one or another mineral in decreasing order allows to distinguish three main mineral types of refractory gold-sulfide concentrates: arsenopyrite - pyrite concentrates (1) and (2); pyrite arsenopyrite concentrate (3); pyrrhotite - pyrite - arsenopyrite concentrate (4). Gold grade in concentrates is 21.6 - 150 g/t, silver 2.3 - 160 g/t. Total sulfur actually occurs in the sulfide form amounting to 5.68-28.8%; the content of elemental sulfur in pyrrhotite-free concentrates (1 -3) is insignificant (0.11-0.15%) while its grade in pyrrhotite-bearing concentrate (4) is higher by an order – 1.3%. Arsenic occurs, mainly, in the form of sulfide arsenic amounting to 4.5-18.9%. The concentrates also contain nonsulfide constituents, such as (%): 9.15 – 51.6 SiO2, 0 – 8.16 CO2, 0.54 – 4.5 C organic. The mineral composition of concentrates (Table 1) is represented by various contents of the main gold-bearing sulfide minerals - arsenopyrite, pyrite and pyrrhotite as well as sulfides grading 17 to 74.8%. Pyrrhotite is abundant (39%) only in the concentrate (4); 92


antimonite (1.7%) is also contained, mainly, in the concentrate (4). The grain size of concentrate samples is within the range of 80-90% of the class – 0.044 mm. Table 1. Mineral composition of concentrates Mineral type of concentrates, grade % Sulfides

FeS2 FeAsS FeS Sb2S3 Total sulfide content

Arsenopyrite-pyrite (1) 33.7 41.0 single grains 0.1 74.8

(2) 7.2 9.8 not found not found 17.0

Pyrite-arsenopyrite (3) 24.3 10.4 not found single grains 34.8

Pyrrhotite-pyritearsenopyrite (4) 12.0 11.8 39.0 1.7 64.5

2.2 Microorganisms The research on bacterial oxidation of the concentrate (1) employed mesophilic bacteria: the Acidithiobacillus (A.) ferrooxidans monoculture with concentration of 1,0 x 107cells/ml and mixed culture of A. ferrooxidans and A. thiooxidans with corresponding content of 2.5 x 107 and 2.5 x 104 cells/ml. (Table 2). Table 2. Caracterization of the applied culture* Mineral type of concentrates

Bacteria species, quantity of cells per milliliter of solution

Arsenopyrite-pyrite: (1)

Monoculture – A.ferrooxidans 1.0 x 1010

(1)

Mixed culture: A.ferrooxidans – 2.5 x 107, A.thiooxidans – 2.5 x 104

(2)

Mixed culture: A.ferrooxidans – 1.6 x 109, A.thiooxidans – 2.5 x 104

Pyrite-arsenopyrite (3)

Mixed culture: A.ferrooxidans – 3.5 x 109, A.thiooxidans – 5 x 103

Pyrrhotite-pyrite-arsenopyrite (4)

Mixed culture: A.ferrooxidans – 4.5 x 109, A.thiooxidans – 4.5 x 108 Leptospirillum ferrooxidans – 2.2 x 104, Sulfobacillus thermosulfudooxidans – 2.2 x 106

* The microorganism populations were studied by G.I.Karavaiko, T.F.Kondrateva and T.A.Pivovarova, specialists from the Institute of Microbiology, RAS.

The oxidation of concentrates (2, 3) was performed with the use of the mixed culture of A.ferrooxidans and A. thiooxidans bacteria with cell concentrations of 7.05x108 – 1.0x109 and 2.5x104 cells/ml correspondingly. The pyrrhotite-bearing concentrate (4) was tested with the use of the mixed culture of A. ferrooxidans – 4.4 x109, A. thiooxidans – 4.5 x108, Leptospirilllum ferrooxidans – 2.2 x104 and moderately thermophilic Sulfobacillus thermosulfidooxidans bacteria – 2.2 x106 cells/ml. Bacterial culture available in the TsNIGRI’s biotechnological laboratory which were previously adapted to pyrite-arsenopyrite substratum were used in researches. Adaptation and growth of the biomass was implemented for each particular concentrate (1-4) by transferring bacteria to the 9K environment with portioned addition of concentrates up to the ratio of S(solid):L(liquid) = 1:50-1:20. The biomass was further re-disseminated and 93


the content of solids increased up to S:L=1:5. The process of bacterial adaptation continued with checking pH= 1.5-2.2, Eh = 500-680 mV, concentration of iron oxide and protoxide, biomass activity measured by oxygen consumption on the Warburg device and the rate of iron protoxide oxidation. 2.3 Biooxidation tests The process of bacterial oxidation of concentrates was studied on the pilot laboratory unit under the conditions of continuous bacteria cultivating. The unit includes 5 reactors of capacity 2 litres each, furnished by devices for mechanical mixing, dispersion and additional 1000 cm³ /l per min air supply. Bacterial oxidation was performed at S:L = 1:5, pH=1.5-2.2 and temperature of 28 - 32º C. The pH value was maintained by addition of sulfuric acid or lime slurry depending on the material composition of the concentrate. The required temperature of the process was provided with the help of thermoelectric heaters placed directly in the reactors. The main parameters were constantly monitored, such as: pH, t°C, the biomass oxidizing activity; bacterial solutions were analyzed for Fe+3, Fe+2 and As. Once a day, 45 ml pulp samples were taken to make analyses for solid phase – sulfide As, sulfide and elemental S and sulfide Fe. 3.

TEST RESULTS

3.1 Comparison of the efficiency of FeAsS - FeS2 concentrate bioleaching with the use of A. ferrooxidans and a mixed culture of A. ferrooxidans and A. thiooxidans In order to select the most effective microorganisms for biooxidation of refractory concentrates containing arsenopyrite and pyrite at 28-32°C, the comparative tests were made on bacterial leaching of arsenopyrite - pyrite concentrate (1) with the use of A. ferrooxidans monoculture and a mixed culture of A. ferrooxidans and A. thiooxidans. Performance of biooxidation was estimated from the analyses of the contents of sulfide and elemental S and sulfide As in the residues of biooxidation of the concentrate (1) within 5 days (Fig. 1A), and also indexes of the completeness of arsenopyrite and pyrite oxidation (Fig. 1B). The kinetic curves show higher rate of sulfide components biooxidation in case of the mixed A. ferrooxidans and A. thiooxidans culture as contrasted to the single A. ferrooxidans (curves 1 and 1°). By the end of the 5-th day, residual content of sulfide S was, accordingly, 8.12 and 10.2%, sulfide As 0.1 and 1.0% and elemental S - 0.42 and 1.41%. When using the mixed bacterial culture, more complete oxidation of arsenopyrite – 99.4% and pyrite 71.4% is attained as contrasted to the monoculture (correspondingly, 94.7% and 62%). Joint use of A. ferrooxidans and A. thiooxidans was recommended as most effective tool for further investigation of bacterial oxidation of other mineral types of sulfide concentrates containing arsenopyrite and pyrite. 3.2 Kinetics of biooxidation Fig. 1 shows kinetic curves corresponding to biooxidation of sulfide components of three mineral types of concentrates at 28-32°C with the use of mixed bacteria, the performance of the latter being shown in Table 2. The kinetic curves of bacterial oxidation of the concentrates (1-3) containing only two main minerals - pyrite and arsenopyrite differ from kinetic curves of biooxidation of the concentrate (4) which additionally contains pyrrhotite. 94


95


1-4 residues of biooxidation of the concentrates (1 – 4) with mixed culture of bacteria. 1° residue of biooxidation of the concentrate (1) with A.ferrooxidans.

Figure 1. Biooxidation of sulfur, sulfide arsenic, elemental sulfur, pyrite and arsenopyrite during bioleaching of gold sulfide concentrates

The sulfide minerals are known to have different crystalline structures and different stability in sulphate solutions. With pH= 2.5 a row of sulfide stability is as follows: pyrrhotite (0.45B), arsenopyrite (0.50B), pyrite (0.55 – 0.6B). Pyrrhotite, as most electrochemically sensitive mineral, is oxidized, first of all, chemically and microbiologically according to the reactions: (1) FeS + Fe2(SO4)3 = 3 FeSO4 + S° (2) FeS + 0,5 O2 + H2SO4 = FeSO4 + S° + H2O bacteria (3) 2FeS + 4,5 O2 + H2SO4 = Fe 2(SO4)3 + 2 H3AsO4 In the course of biochemical oxidation of pyrrhotite Fe2+, Fe3+ sulfates and S° are created, which are further oxidized by bacteria: Fe2+up to Fe3+, S° up to SO4. In the first two days of concentrate biooxidation a considerable growth of S° grade in the residues of biooxidation of pyrrhotite-bearing concentrate (4) is observed - from 1.3 up to 6.85% while in the course of oxidation of the pyrrhotite-free concentrates (1-3) less amount of S° is produced – 0.1-1.3%. This testifies about predominant oxidation of pyrrhotite (90-95%) and lower oxidation rate of arsenopyrite (19-29%) and, especially, pyrite when leaching the FeS - FeS2 - FeAsS concentrate (4). In all other pyrrhotite-free concentrates, the arsenopyrite oxidation rate in the first two days is 2.5 - 3 times higher and it is oxidized for 78-91%. The lower rate of arsenopyrite oxidation in pyrrhotite-bearing concentrate as compared to pyrrhotite-free concentrates is, probably, explained by variations in electrochemical characteristics of the environment: increase of pH from 1.8 up to 2.2 and decrease of Eh value from 700 up to 500 mV due to Fe2+ accumulation and reduced 96


soluble sulfur compounds resulted from chemical oxidation of pyrrhotite and, partially, arsenopyrite that is evidenced by diminishing concentration of A.ferrooxidans bacteria from 2.5 x 107 up to 2.5 x 104 cells / ml in the first two days of oxidation of the pyrrhotitebearing concentrate (4). Then S° content decreases from 6.8 up to 3.9% due to microbiological oxidation by A. Thiooxidans bacteria the amount of which increases from 2.5 x 107 to 2.5 x 109 cells / ml. The Sulfacillus thermosulfidooxidans bacteria living inside the system also facilitate S° oxidation. At the same time, the rate of arsenopyrite oxidation increases that is evidenced by diminishing concentration of sulfide As from 4.4 up to 0.2-0.4%, accordingly, after 2 and 5 days of bioleaching, and by growing quantity of A.ferrooxidans bacteria from 2.5 x 104 up to 6 x 107 cells /ml. For 5 days of the Fe S - FeS2 - FeAsS concentrate (4) bioleaching, arsenopyrite is oxidized for 98%, pyrrhotite - 99%, pyrite 35%. The residual content of sulfide S accounts for 6.1% as compared to the initial 28.8% value. The completeness of sulfide oxidation in the FeAsS - FeS2 and FeS2 - FeAsS concentrates is governed by a total grade of sulfides and their quantitative ratio. All kinetic curves of oxidation of pyrite, arsenopyrite, sulfide S and As, and also elemental S in the concentrates (1-3) are ranked as 1, 3, 2. The content of sulfides in concentrates decreases in the same order, %: (1) – 74.8, (3) – 34.8, (2) – 17.0. Hence, the less is the content of sulfides, the more completely they are oxidized. This dependence is most pronounced for pyrite as an example. For 5 days, the highest degree of pyrite biooxidation (94.8%) is attained in the concentrate (2) which contains the least amount of sulfides - 17% (7.2 FeS2). The worst oxidation of pyrite (71.1%) occurs in the concentrate (1) that contains the greatest quantity of sulfides - 74.8% (33.7% FeS2). The duration of bacterial leaching of refractory concentrates is governed by their composition and, first of all, by sulfide content. The concentrate (2) containing the least quantity of sulfides (17%) is oxidized almost completely within 3-4 days. The oxidation of the concentrates (1,3,4) containing 34.8-74.8% sulfides requires 5-6 days of bioleaching with pH = 1.5-2.2, S:L = 1:5, t = 28-32ºC with the use of mixed culture of mesophilic bacteria. The analysis of kinetic curves of Sº biooxidation demonstrates that the process of Sº oxidation is not terminated after 5 days of concentrate leaching, except for the concentrate (2) with the lowest sulfide grade (17%). A great amount of Sº (3.9%) is retained in the pyrrhotite-bearing concentrate (4) while 0.42 and 0.24% Sº remains in the concentrates (1) and (3). For microbiologic oxidation of Sº the leaching was prolonged from 5 to 12 days (Fig. 1). This allowed to decrease the content of elemental sulfur from 3.9 only to 1.5% while the content of sulfide sulfur dropped from 6.1 up to 5%. 3.3 Oxidation of Sº by hydrometallurgy before cyanidation As a rule, the biooxidation residues have a complex composition and contain constituents which consume oxygen and cyanide, and deteriorate the process of gold cyanidation. It is shown in publication [8] that the residue of biooxidation of pyrite – arsenopyrite ore contain S2-, Fe (OH) compounds, residual FeAsS, jarosite and Sº that are the consumers of oxygen and sodium cyanide. In order to diminish their action, the pulp was treated, prior to cyanidation, with lime at рН=10.5 within 24-40 hours, then sodium cyanide was added to perform leaching within 24 hours. We studied an effect of several methods of treating the biooxidation residues before cyanidation on the decrease of the content of elemental sulfur, such as: 3-4 repeated water 97


washing, neutralization, aeration in limy environment, electrochemical treatment. The hydraulic washing of biooxidation residues does not affect the Sº content. The neutralization just insignificantly decreases the Sº content - from 3.9 up to 3.75%. The aeration of the pulp in lime environment causes a partial oxidation of Sº - from 3.75 up to 2.71%. Most effective is the electrolytic treatment of pulp within 1-2 hours at pH=11-12 (Fig. 2A).

Figure 2. Effect of treatment of biooxidation residues before cyanidation on content of sulfur (A), gold extraction and cyanide consumption (B)

In the course of electrolysis of aqueous solutions the gaseous oxygen is released; its solubility in solution increases up to 20-30 mg/l. The gas saturation of pulp with finely dispersed bubbles increases 100-1000 times as compared to aeration. The process of Sº oxidation is enhanced and its content decreases from 3.75 up to 2.17% for 1-2 hours of treating. A mechanism of pre-aeration in lime environment or electrolytic processing can be represented by two stages: 1. Oxidation and precipitation of oxygen and cyanide absorbers prior to cyanidation: –Fe2+→ Fe3+→ Fe (OH)3, S2-→ S° → S2O3 2- → SO42- CaO→ CaSO4 2.

Lime passivation of sorption-active surface of newly precipitated iron hydroxides and decrease of a sorption capacity of products of bacterial oxidation in relation to cyanide complexes of precious metals. The applying of aeration and electrolytic processing of pulp allows to stabilize the pulp ionic composition prior to cyanidation. As the duration of preparation increases, the concentration of oxygen and cyanide absorbants decreases, mg/l: from 52.5 up to 10.2 S2-; from 35 up to 0 S2O32-; from 6.4 up to 0 Fe2+. A chemical value of oxygen absorption decreases from 480 up to 320 mg/l and remains constant during the whole cyanidation process. The stabilizing of ion composition of pulp allows to reduce 3.3 times the CNSion concentration and 1.7 times the SO42- ions. The neutralization of harmful constituents allows to improve gold dissolution and to reduce cyanide consumption. The preliminary aeration within 24 hours allows to increase gold recovery from the residue of biooxidation of the concentrate (4) from 90 up to 96% 98


and to reduce the cyanide consumption from 16.0 up to 7.2 kg /t. The electrolytic processing of pulp before cyanidation increases gold recovery up to 97% and reduces the cyanide consumption from 16 up to 7 kg /t (Fig. 2B). Table 3 shows that biohydrometallurgical processing of refractory gold-sulfide concentrates is an effective technology of sulfide oxidation and recovery of finely disseminated precious metals. Gold recovery from the studied mineral types of concentrates with the use of biohydrometallurgical technology attains 92-98% as compared to low-effective parameters of gold recovery by cyanidation of source concentrates – 12.9-63.4%. Table 3. Effect of biooxidation on gold recovery from concentrates Mineral type of concentrate Processing method

Gold recovery by cyanidation, % Gold recovery by biohydrometallurgy, % Duration of biooxidation, days

(1) 63.4

(2) 12.9

(3) 37.5

Pyrrhotite-pyritearsenopyrite (4) 51.2

92.1 5

98.0 4

92.0 5

97.0 5

Arsenopyrite-pyrite

Pyrite-arsenipyrite

4.

CONCLUSION S° is formed during biooxidation of all mineral types of concentrates containing arsenopyrite, pyrite and pyrrhotite. The presence of pyrrhotite enables creating of the most amount of S°. The process of its biooxidation is slower than sulfides oxidation. S° is saved in biooxidation residues and has a negative effect on the process of cyanidation, leads to high consumption of cyanide because of (CNS) creating. Using of aeration in lime environment or electrochemical treatment enables to decrease cyanide consumption as much as 2.3 times and increase extraction of gold from 90 to 96-97%. REFERENCES

1. Bierly I.A., International Congress, Biotecnology, 2002 Moscou, Russia, (2002) 457. 2. Shrader, V.J. and S.X. Su, International Biohydrometallurgy Symposium '97 (1997) M3.3.1. 3. Lodeishikov V.V., Technology of gold and silver recovery from refractory ores. Irgiredmet JSC. Irkutsk, (1999). 4. Jones L. and Hacki R.P., International Biohydrometallurgy Symposium '99 (1999) 337. 5. Rossi, G., Biohydrometallurgy, McGraw-Hill, Hamburg, 1990. 6. Komnitsas C. and Pooley E.D., Minerals Engineering, 3 (1990) 295. 7. Hacki, R. P. and L.Jones, International Biohydrometallurgy Symposium '97 (1997) M14.2.1. 8. Livsesey, E., P. Norman and D.R.Livesey, International Biohydrometallurgy Symposium '83 (1983). 9. Xiang, L. and J. Ke, Transactions of Nfsos., 4(4) 1 (1994) 42. 10. Sedelnikova G.V. and A.V.Narseev, Proceedings of TsNIGRI, 233 (1989) 26.

99


Beneficiation of phosphatic ores from Hirapur, India A. D. Agate 7, Narmada Apts., United Western Society, Pune - 411 052, India Abstract The availability of phosphorous in tropical soils is generally low due to immediate precipitation and fixation of applied phosphorous to soil. In India, 98% of the soil contains insufficient amounts of available phosphorous to support maximum plant growth and hence, application of phosphorous is very essential. High grade rock phosphate is in short supply. Therefore, it becomes necessary to import di-ammonium phosphate from abroad. Hence, it was thought to beneficiate the low grade rock phosphate available at Hirapur in India. Unfortunately, it contains high amounts of iron and silica (3-5%). Hence, microbial beneficiation was attempted using Thiobacillus ferrooxidans to remove extra iron by ferrous iron oxidation and by removal of both iron and silica by sieving and cleaning the ore using various physical treatments. This resulted in a cleaner ore, with nearly 70% iron and 25% silica impurity removed. The factors for such a beneficiation process were standardized at laboratory level (optimum pH = 3, temperature = 30°C, 1% pyrite added as nutrient). These factors are used for scaling up the process at field site at one t.p.d. level. This is an important first step in production of indigenous phosphate source to eliminate import - dependence in agriculture.

Keywords: phosphatic ores, Thiobacillus ferrooxidans, beneficiate 1.

INTRODUCTION The availability of phosphorus in tropical soils is generally low due to immediate precipitation and fixation of the applied phosphorus. In India, 98% of the soils contain insufficient amounts of available phosphorus to support maximum plant growth. Application of phosphatic fertilizers, therefore, becomes very essential (1, 2). According to several workers (3, 4), about 85 to 90% of the added phosphorus in the fertilizer gets fixed in the soil within 24 to 48 hours of its application and becomes unavailable to the plants. Thus, efficiency of utilization of phosphatic fertilizers in soils ranges from 10 to 15% only. The problem of phosphorus fixation was found to be very acute in neutral to alkaline soils in India, where the phosphorus is precipitated as calcium phosphate upon application of superphosphate. This results in non-availability of applied phosphorus in soils, which is a considerable loss from the point of view of productivity and economy to the nation. Manufacturing of phosphatic fertilizers in India is faced with serious problems, as it requires the use of non-renewable resources, such as high grade rock-phosphate and sulphur, which are in short supply and are being depleted progressively; thereby, becoming costly. 101


The magnatic or igneous rocks are the principle sources of phosphorus on our planet and the igneous rocks rich in phosphorus are industrially exploited. In India, the Mussourie rock phosphate is currently exploited, as it has 8.6% phosphorus or 20% P2O5. At Hirapur, M.P. India, in addition to phosphorus it contains high amounts of iron along with silica. Therefore, it becomes difficult to remove iron and silica by conventional chemical and physical means. Hence, the aim of the present study was to screen and select microorgamisms, giving maximum phosphate solubilizing activity to remove the phosphorus part or removing the impurities of iron and silica from the phosphatic ore of Hirapur, India. 2.


2.1 Screening and selection of microorganisms 2.1.1 Phosphate solubilizing activity Our previous studies had indicated 10 species of bacteria, two of yeasts and four mycelial fungi to have a good phosphate dissolving activity. (5). These microorganisms were isolated from nine different ecosystems containing phosphates or having come into contact with phosphates. A chemical analysis of the rock phosphate from Hirapur is shown in Table 1. When these microorganisms were screened qualitatively by using Pikovskaya's agar medium or quantitatively in Pikovskaya's liquid medium, (6) it was found that the culture of Arthrobacter species gave maximum solubilization of 27.1% in 14 days (Table 2). This culture was used to find out the amount of phosphorus released from rock phosphate of Hirapur. 2.1.2 Removal of ferrous iron and silica from phosphate ore Such a beneficiation was attempted using Thiobacillus species on the phosphate ore crushed to -30 mesh size. From the different thiobacilli tried, Thiobacillus ferrooxidans strain B-101 from MACS Culture Collection MCM (7) was able to remove 86.5% ferrous iron from the ore along with silica. The phosphorus and silica content was analyzed by standard techniques (8) before and after the treatment and the analysis is shown in Table3. 2.1.3 Optimization of the beneficiation technique Since it was observed that instead of only solubilisation of phosphate, the beneficiation achieved by Thiobacillus species yielded good results, the optimisation was attempted only with Thiobacillus species, which removed the impurities of both silica and iron from the ore. The growth of the culture was attempted under various conditions of pH, temperature and nutrients to optimize the parameters and the results are reported in Table 4. 3.

RESULTS It can be seen from Table 1 that the ore contains, a high amount of iron and silica. Normal beneficiation operations, such as sieving and cleaning of ore using different techniques have very little effect in removal of these impurities.

102


Table 1. Chemical analysis of samples of phosphatic rocks from Hirapur, India Element CaO P2O5 SiO2 MgO A12O3 Fe2O3 and FeO

Average Occurrence 65 - 63% 25 - 20% 5 - 3% 2 - 0.5% 1 - 0.5% 5-3 %

It is also seen from Table 2 that even though the Arthrobacter species is able to solubilize phosphorus to some extent, the net effect is that most of the phosphorus in the ore remains insoluble, probably due to impurities associated with the ore. Therefore, the second aspect of the studies, i.e. beneficiating phosphatic rocks by removing silica and iron was considered promising and was attempted. Table 2. Phosphatic solubilizing activity of Azotobacter sp., Arthrobacter sp. and Candida sp. in Pikovskaya's broth Cultures used Bacteria: Azotobacter sp. Arthrobacter sp. Yeast: Candida sp.

after 7 days

% phosphatic solubilization after 14 days

after 21 days

5.75 13.82

10.50 27.10

7.04 17.82

6.74

14.05

8.71

When the Thiobacillus strain B-101 was tried, it was found that on an average 86.5% of iron and 30.4% silica impurities were removed at laboratory level (Table 3). Therefore, it was thought that this method could be used for beneficiating the phosphatic rocks from Hirapur after optimization of various parameters for heap leaching on a large scale. Table 3. Beneficiation of phosphatic ore from Hirapur, India using a culture of Thiobacillus feroxidans (MCM B-101) Ore analysis Before treatment After treatment

Fe 4.5 0.6

SiO2 2.3 1.6

Upon optimization, it was found that the Thiobacillus culture can optimally grow at pH 3 and at 35° C. It becomes necessary to add a source of sulfur, such as pyrite as part of the nutrient supplement to allow the cultures to grow optimally. When the experiments are scaled up from the laboratory level to the heap leaching level, it was found that the efficiency of the process decreased slightly, when it was observed that 70% of the iron and 25% silica impurities were removed at that level. However, the process takes place in a continuous manner and therefore is advisable to use. The process parameters for treatment of 300 kg heap of ore are listed in Table 5. Based on the encouraging results obtained, it is proposed that the beneficiation of rock phosphate first by using microbiological means followed by physical and chemical treatment would be an important first step for producing clean phosphatic rocks for use as fertilizer in India. 103


Table 4. Optimization of various parameters for growth of Thiobacillus ferrooxidans B-101 Parameter tested I. pH 2.0 3.0 4.0 5.0 II. Temperature 25°C 30°C 35°C 40°C III. Energy source Sulfur (1%) Ferrous sulfate (47%) Pyrite (1%)

Growth (cell density in 48 hrs) 4 x 107 1.3 x 108 2 x 107 0.5 x 107 6.6 x 107 1.3 x 107 4 x 108 1.6 x 108 3 x 107

Table 5. Beneficiation of phosphatic ore by heap leaching (Heap size = 300 kg) Optimum pH Optimum temperature Suitable Nutrient (Energy Source)

pH 3 30° C 1% Pyrite

Process Efficiency

Removal of Fe 70%

SiO2 25%

4.

DISCUSSION It was observed earlier in the microbial beneficiation of manganese ores, which was tried to remove phosphorus from the ore, that a combination of physical and chemical treatments, such as passing the crushed material after sieving through immersion magnetic separator (physical), followed by chemical precipitation in flotation cells, etc. had very little beneficial effect (9). Even though, some amount of iron and silica content was reduced, due to such treatments, it was not found effective in reducing the phosphorus content of the ore. However, a culture of Arthrobacter was observed to leach out 70% to 85% phosphorus from manganese ore (10). In the present study, it was observed that time 70% iron and 25% silica were removed from the phosphatic rocks in six days using a Thiobacillus culture. If physical and chemical beneficiation techniques are tried later on these 'cleaned' ores, it would provide an economically feasible operation to beneficiate phosphatic rocks, to be used as fertilizer. This is being tried at site on 1 t.p.d. plant now. REFERENCES

1. D.O. Norris, In: Tropical Pastures, W. Davis and C.L. Skidnore (eds.), Faber and Feber, London, (1968) 89-106. 104


2. K.R. Sonar and G.K. Zende, J. Maha Agric. Univ., 9 (1984), 74-81. 3. S. Banik and B.K. Dey, Plant and Soil, 69 (1982), 353-364. 4. J.M. Vincent, In: Soil Nitrogen, W.V. Bartholomew and F.E. Clark (eds.), American Soc. Agronomy 10, (1965) 384-385. 5. S. Beheray and A.D. Agate, In: Bio-fertilizer Technology Transfer, A.V. Gangavane (ed.) Associated Publishing Co., New Delhi (1992), 193-197. 6. R.I. Pikovskaya, Mikrobiologiya, 17 (1948), 362-370. 7. S.M. Hideaki, Juncai, M. Satoru, S. Junko and T. Youka (eds.) World Directory of Collection of Cultures of Microorgamisms - Bacteria, Fungi and Yeasts, WFCC World Data Center on Microorganisms, Saitama (Japan), (1993), 130-131. 8. F.J. Welcher, Standard Methods of Chemical analysis, Part A, Vol II, Industrial and Natural Products and Non-instrumental Methods. 6th ed. D. Van Norstrand Co., Princeton, N. J. (1979). 9. A.D. Agate, In: Biogeotechnology of Metals, G. I. Karavaiko and S. N. Groudev (eds.) Center for International Projects, Moscow (1985), 349-395. 10. A D. Agate, In: Proceedings of the Symposium "Biological approach to problems in medicine, industry and agriculture", N.K. Notani and K. Sundaram (eds.), Bhabha Atomic Research Center, Dept. of Atomic Energy, Bombay (1974), 161-176.

105


Biohydrometallurgy of antimony gold-bearing ores and concentrates P.M. Solozhenkin1 and V.P. Nebera2 a

Institute of Earth Resources Development RASc Kryukovsky Tupik, 4, Moscow, 111020, Russia; [email protected] b Moscow State Geological Prospecting University, Miklucho - Maklaya, 23, Moscow, 117997, Russia; [email protected] Abstract Experimental data on bacterial leaching of antimony-bearing ores of the Sarylakhsk and Sentochansk deposits (Yakutia) and ores of Tajikistan deposits. Bacterial accumulated culture Desulfovibrio desulfuricans have been employed as leaching and chemical agents. Encouraging results have been obtained in isolation antimony into the solution (recovery 96-98%). Thiobacillus ferrooxidans used for flotation selection of Hg minerals from antimonite and for transformation of antimonite into commercial product of antimony trioxide. 1.

INTRODUCTION Antimony gold-bearing ores (Sb2S3-Au) are wide spread in Russia, People's Republic of China, Republic of South Africa and Bolivia. In the Russian Federation, the main sources of antimony production are sarylakhsk flotation to produce antimony gold-bearing concentrates, obtained by enriching ores of the Sarylakhsk and Sentochansk deposits, Republic Sacha (Yakutiya) [1]. The current processing of Au-Sb concentrates is accomplished using pyrometallurgy. The gold is concentrated at the bottom of Au-containing antimony alloy ingots; the Au content ranges from 2 to 4% [2]. Only commercial trioxide of antimony is a salable product. When a hydrometallurgical alkaline leach method of production is applied, antimony is leached with the extraction of gold in cake. Soluble antimony is then subjected to cathodic electrodeposition to recover antimony [3, 4]. The existing methods for processing gold-antimony concentrates do not solve he problem of their complex use, since these processes do not make it possible to concentrate gold to products suitable for refining. Not all of Russia’s antimony needs are currently met. In this regard there are needs for geological prospecting of new deposits in Yakutiya, the Khabarovsk region, the Baikal region, in the Arctic and in Central Ural and new efficient technology developments for processing.

1 2

Prof. Dr. Sci., Academician Natur. Sci. (RAEN) and Tajik Acad Sci. Prof. Dr. Sci., Academician Natur. Sci. (RAEN) and Mining Sci. (RAGN)

107


A widely available cyanidation process is not applicable for gold extraction from Sb2S3-Au ores and concentrates, because of large cyanide consumption due to Sb reaction with NaCN and the locking of Au in the sulfide matrix. Use of thiourea allows to extract gold from gold-bearing stibnite, however, this development is still under laboratory investigation [5]. The problems of antimony leaching can be solved by applying an oxidizing acid leach. Other investigators have used ferric chloride to leach antimonite (stibnite) obtaining antimony trioxide [6-9]. Several microbial processes, including oxidizing and reducing type reactions, which can alter the physio-chemical, sorptive and flotation properties have been evaluated on various antimony minerals [10-12]. Some of these microbial processes are promising. When the surfaces of mercury and antimony sulfides are biomodified, selective separation is possible. It may also be possible to develop new microbial-produced solvents for Sb2S3 that would yield soluble antimony chlorides. Also, conventional bacterial leaching is promising. Kenzhalov et al. [17] from the Institute of Metallurgy and Mineral Processing, Republics Kazakhstan, have selected a heterothroph, Pseudomonos aureofaciens, from a Kazakhstan mineral deposit. They have studied in detail the influence of this heterothroph on antimonite, defining of the kinetics of the reaction, the order of reaction, the velocity constant, and the maximum velocity of product formation –(Mmax and the Michael’s constant, Km). Rapid dissolution of antimony from antimonite was obtained in the presence of the bacteria (0.11x10-4); without the bacteria the rate was 0.29x10-5. Bacterial affinity to antimony was quite high (Km = 0.07-0.37). Dissolution of sulphur surpassed that of antimony. Sulphate-reducing microorganisms (SRB) are wide-spread in nature. They utilize sulphate-sulphur as an electron acceptor. The most representative organism of the SRBs is Desulfovibrio desulfuricans. Intact cells of SRB rapidly reduce sulphate, while simple organic compounds or molecular hydrogen as follows: (1) Corg + SO42- = S2- + 2CO2 22(2) 4H2 + SO4 = S + 4H2O 2+ ∆F = -46410 kal (3) SO4 + H + 4H2 = HS + 4H2O A number of enzymes participate in the above reactions. Many studies have been conducted using SRBs in flotation of oxidized antimony and lead ores, as a desorbent and depressor in flotation of concentrates, for selection of lead from zinc minerals, and for the separation of molybdenite from chalcopyrite [10]. However researchers pay little attention to utilizing SRBs to produce H2S reagents for ore leaching and especially for leaching antimony-bearing materials. Lyubavina and coworkers (personal communication) have perfected a nutrient medium for the large-scale production of SRBs; this medium is based on the use of linter dust, a waste product of cotton-seed processing, as a carbon source for the organisms. Only alkaline leaching of antimony sulfides has achieved industrial application. This hydrometallurgical process results in high selectivity for the noble-metal groups, which remains in a leaching cake. Mel’nikov and coworkers found that dissolving antimony sulphide and antimony oxide in sodium sulphide and caustic soda results in sodium thioantimonite and thioantimonate formation according to the following reactions [2]: H2S + 2NaOH = Na2S + 2H2O (4) (5) Sb2Sb3 + 3H2S + 6NaOH = 2Na3SbS3 (6) Sb2S3 + 4NaOH = Na3SbS3+ NaSbO2+ 2H2O 108


Sb2O3+ 5Na2S + ÇÍ2Î = NaSb2S3+ 6NaOH (7) The objectives of this study were to: 1. Evaluate the use of sulphate-reducing bacteria (SRBs) as agents to produce H2S for the leaching of gold-bearing antimony sulfide minerals, and 2. Assess antimonite transformation into Sb2O3 using Thiobacillus ferrooxidans. 2.

EXPERIMENTAL

2.1 Materials and methods Antimonite (an antimony-sulphide mineral) ore and concentrates were investigated. A flotation sulphide concentrate, containing 39.55% antimony, was studied using 250-ml conical flasks at 28°C. An apparatus, specially designed for bacterial leaching, was also employed. Ores and concentrates for study were sterilized by boiling. In some experiments the solutions were filtered from insoluble residue. pH, the number of bacterial cells and antimony content were measured. Some antimony compounds were identified using spectral analyses with a quantometer VRA-2. For research purposes a SRBs were isolated from the Tyrny-Auz molybdenumwolfram deposit. Postgate medium [12] was used to cultivate the SRBs under laboratory conditions. Using a direct observation method a maximum of 220x106 cells.ml-1 of SRBs were found out to grow on the fourth day of culturing. During this period of time 400 mg.l-1 of soluble H2S was obtained; that was 10 times less than the maximum solubility of H2S in water. SRBs a ready-to-employ reagent solution of H2S for the leaching process after four days of growth. 2.2 The ores Ore from the Joint-Stock Co. Gold of Sacha (Yacutiya) was used to produce a flotation concentrate of antimony. This concentrate, produced by the Indigir-Gold plant, contained, %: 55 Sb; 22 S; 0.4 As; 30 g.t-1 Au and 3 g.t-1 Ag. An antimonite concentrate procured from the Adichanskiy operation at the Sentochanskaya Au-Sb deposit contained, %: 37 Sb; 14 S; 0.16 As; 60 g.t-1 Au, and 25 g.t-1 Ag. Thiobacillus ferrooxidans were cultured from the Bakyrchik and Olimpiadinsk deposits. The experimental testwork was performed in 250-ml Erlenmeyer flasks using 9K medium. 3.

RESULTS AND DISCUSSION

3.1 Sulfate-reducing bacteria as antimonite solvents Recovery of gold from refractory ores requires a pretreatment to liberate the gold particles from the host mineral. The antimony forms stable compound with NaCN during the cyanidation process. Pretreatment is usually an oxidation step. As an alternative, chemical or bioleaching can be applied to liberate the gold particles from the sulfur matrix. Emphases deserves an operation of sulfide-alkaline leaching as a way of selective separation of stibium from gold-bearing concentrates. Known developped by Irgiredmet (Irkutsk) technology of metallurgical conversion of rich gravity concentrates of Sarylachsk dressing plant (Au – 1050g.t-1; Sb - 65 %) [18]. The scheme includes sulfidealkaline leaching with the following electrolytic extraction of stibium from solutions. Three-phase recleaning of stibium leaching tailings from concentration tables, melting of secondary gravity concentrates (contents Au – 48 kg.t-1) on the dore metal and cyanidation 109


of tailig from final gravity concentration, received total extraction in corresponding commodity products Au - 98 %, Sb - 96,5%. The authors demonstrated the technical feasibility recovery of Sb by Na2S and NaOH leaching, the successive gold solubilisation by conventional cyanidation process and the recovery of Sb and Au from the respective leach solutions by electrowinnig [4]. Chemical basic leaching of pure stibnite by Na2S and NaOH under different experimental conditions at 40°C has been studied in order to optimise the reagents concentrations for the antimony dissolution process. Response surface methodology has been used to find the best experimental concentrations to maximize the Sb extraction yield. 98-100% of antimony recovery was obtained by using 1g Na2S and 1g NaOH per gram of pure stibnite. Treatment of a gold-bearing stibnite ore with cyanide yielded only 4% Au extraction; however, after seven days of bioleaching 85.5% Au recovery was attained (Table 1) [13]. Tests were conducted at laboratory scale utilising a refractory stibnite ore [14]. The gold content of the sample was 32 g.t-1. Bacterial cultures utilised in the biological test consisted predominantly of Thiobacillus. Table 1. Gold recovery by cyanidation, with and without biooxidation [14] Time, hours 1 12 24

Recovery Au, % 7 days bioleaching 52.2 63.5 85.5

No bioleaching 1.2 1.5 4.0

14 days bioleaching 53.2 64.1 86.0

At laboratory scale was investigated the best conditions for alkaline leaching of a refractory gold-bearing Sb2S3 (13.25 Sb2S3; 30 g.t-1 Au) coming from South America [19]. The solution was constituted by sodium sulfide and sodium hydroxide. Main parameters studied were: Na2S concentration, NaOH concentration, pulp density and temperature. It was reasonable to check a possibility of leaching stibium by bacteria. The selective flotation and separation of cinnabar from antimonite minerals using T. ferrooxidans [15] is illustrated in Figure 1. Bacterial conditioning of 5 h did not affect cinnabar flotation (recovery 89.6%), while the antimonite recovery by flotation decreased from 89 to 6.2%; this led to almost complete selection of the minerals. The results presented in Figure 1 indicate the superiority of biological separation compared to chemical separation. 100

Recovery, %

80 60 Sb

40

Hg

20 0 1

2

3

4

5

Time, h

Figure 1. Changes in flotation recovery (%) of Sb2S3 and HgS minerals in processing by T. ferrooxidans from time, h 110


As a result of bacterial oxidation, antimonite is converted to antimony trioxide, the mineral senarmontite. T. ferrooxidans oxidizes the surface of antimonite crystals while cinnabar remains intact. As a result, HgS is floated and extracted into the concentrate, while Sb2S3 is coated by a fine film of oxides (Sb2O3) and removed in the tailings. Partly regenerated culture can be recycled in this process. When the bacterially-oxidized products of the antimonite (stibnite) concentrate were analyzed by x-ray diffraction, no Sb2S3 was observed. However, XRD analysis of the original concentrate revealed intensive lines, belonging to antimonite [16]. Biooxidation also reduced the content of other elements in the stibnite concentrate. Antimony leaching was done by somewhat different technique. Antimony-containing portion was mixed up with SRB of different hydrogen sulphide concentration and with caustic soda solution. The pulp was heated up to 90°C and it was stirred with S:L = 1:16 ratio. Antimony solubility is most effective at 120 g.l-1 caustic soda concentration and maximum hydrogen sulphide concentration in SRB. Special experiments established that the time necessary for leaching of antimony is 1 or 1.5 h. To increase antimony transition into the solution, it is necessary to increase contact time with SRB during leaching. However, when the time of contact was increased, it was necessary to control antimony ions in the solution whose optimal concentration slowed down the leaching process. Therefore, antimony leaching was done in two stages with gradual addition of reagents. The first stage lasted 1 h, after that the solution was decanted, then once again necessary reagents were added and contacted for 0.5 h. Effective antimony leaching was observed under maintained optimal conditions. Isolation of antimony into the solution under those conditions was about 96.5-98.0%. Results obtained in antimony leaching from antimony-bearing raw materials are given in Table 2. These data show the extent of effective application of bacteria as compared to sodium sulphide in antimony leaching. Furthermore, the suggested technology of antimony leaching has a number of advantages. The presence of intact alkali in electrolyte leads to considerable increase of electric conductivity in the solution and promotes better results of the subsequent electrolysis. Table 2. Effect of sulphate-reducing bacteria (SRB) and Na2S on leaching Sb from antimony-bearing materials* Antimonite Stage I . -1

Stage II

SRB, mg l

8.7

17.5

43.7

52.5

8.7

7.5

43.7

52.5

Sb recovery, %

80.5

83.2

94.9

91.8

95.6

93.6

96.79

98.04

Antimony concentrate Stage I

Stage II

SRB, mg.l-1

25.4

70.9

86.7

139.1

25.4

70.9

87.7

139.1

Sb recovery, %

41.3

46.7

56.8

77.7

92.9

92.9

94.3

96.5

Antimonite . -1

Na2S, mg l

Sb recovery, %

60

20

180

240

300

360

51.4

78.8

92.2

98.0

98.9

98.9

. -1

*NaOH 120 g l , leaching time 1 h, temperature 90°C

111


Thermodynamic calculations show that hydrogene sulphide oxidation occurs at a lesser rate, than that of sodium sulphide. The concentration of S2- and SH- ions in the pulp containing SRB is weaker than that in the pulp, containing sodium sulphide. Traditional technology of antimony leaching requires higher concentrations of sodium sulphide, than in the case of SRB leaching, the latter reducing yield on the electric current. Given cake is processed by usual methods: - cyanidation, since contents of stibium does not render influences upon the leaching of gold; - presence in cake sulfur (more than 14%) must be sodium neutralized; - by gravity concentration methods in centrifugal devices of Knelson type with following separation of concentrate in ferro-magnetic liquid to receive rich Au-containing concentrate, ready for affinage. 3.2 Transformation of antimonite into antimony trioxide by Th. ferrooxidans Information on antimonite biooxidation is of certain interest [14]. According to the results of investigations carried out by Irgiredmet, Sb2S3 oxidation with Th. ferrooxidans was described in [20, 21]. Antimonite biooxidation realized in relatively pliant regime to improve the technological characteristics of cyanided material due to the release of gold associated with Sb2S3 and transformation of antimony to the less active chemical form [16]. It was established that bacterial oxidation of gold-arsenic concentrates by Th. ferrooxidans occurred within 100-120 h, the high degree of sulfide oxidation achieved (%) 96-98 of arsenopyrite, 97-98 of pyrrhotite, 92-95 of antimonite, and 65-84 of pyrite [17]. Under biochemical leaching was observed greater amount of the oxidized forms of stibium and also its oxides of high valences regardless of initial contents of stibium minerals. After bacterial influence an intensity of lines D 5.05 D 5.66 A, belonging to Sb2S3. Thionic bacteria oxidized animonite sulfur and formed antimony trioxide of cubic syngony of senarmontite type (the heat of formation of ∆H278 = 165.5 kJ) according to the following reactions: (8) 2Sb2S3 + 12O2 + 6H2O = Sb4O6 + 6H2SO4 (9) 3Sb2S3 + 3Fe2(SΟ4)3 + 6H2O = Sb4O6 + 12FeSΟ4 + 3S (10) S + H2Ο + 3/2Ο2 = H2SO4 (11) 2FeSΟ4 + l/2Ο2 + H2SO4 = Fe2(SΟ4)3 + H2Ο (12) Sb4O6+ H2SO4 = 2Sb2(SΟ4)2 + 6H2O The rhombic form of Sb2O3 is obtained in hydrolysis of antimony-chloride solutions. Table 3 presents the data of influence exerted by bacterial processing on the change in phase content of minerals under different conditions of the experiments. The action of diluted H2SO4 on antimony sulfate results in hydrolysis with formation of antimonyl sulfate: (13) Sb2(SΟ4)3 + 2H2Ο = (SbO)2SΟ4 + 2H2SO4 Senarmontite processed into antimony trioxide by the reactions: (14) Sb4O6 + 12HCl = 4SbCl3 (15) 2Sb2(SO4)2 + 6HC1 = 2SbC13 + H2SO4 (16) SbCl3 + H2O = SbOCl + 2HCl 112


SbOCl + 2NH4OH = Sb2O3 + 2NH4Cl + H2O

(17)

Table 3. Transformation minerals by Th. ferrooxidans Minerals Senarmontite Antimonite Pyrite Quartz

Initial content, % 0 71.7-81.5 13.3-6.8 15.0-11.7

No. 1 94.6 26.0 0.1 2.7

No 1' 93.9 1.5 — 46.0

Contents after treatment, % Samples numbers No 2 No 2' No.3 No 3' 87.9 88.3 72.0 69.3 7.5 37 8.7 10.8 1.1 — 6.2 6.3 35.0 80.0 12.9 13.6

No 4 40.9 79.0 50.0 46.1

No 4' 44.7 68.0 64.0 44.7

Figure 2. Recommended flow sheet of processing Au-Sb concentrates

Selected stibium trioxide, the contents not below 99.5% Sb2O3 and recovery of stibium in the product - 90.5 %. On the contents of allowing admixtures stibium trioxide corresponds to analytical grade (Tech.Cond. 6-09-3267-84) and exceeds requirements for the lavsan production (Tech.Cond. 6-09-2897-77). Offered technological conversion scheme for Sb-Au concentrates settles problems of their complex using and allows to get high-quality stibium trioxide and gold-containing product. One of the ways of oxidation of stibium minerals is a hydrogen peroxide oxidation, prodused by heterothrofic bacteria [17]. In the reactionary ambience, oppressing development of bacteria, occurs improvement of respiratory activity that brings about hypersynthesis of H2O2. Senarmontite disolves in the alkaline water-glycerin medium containing 250-300 g.l-1 glycerin, 50-60 g.l-1 NaOH, and 100-20 g.l-1 Sb with the complex formation: 113


Sb2O3 + 2NaOH + 2C3H8O3 = 2NaOSbO2C3H5OH + 3H2O (18) Leaching was carried out for 30-60 min without electrolyte heating and with mechanical air mixing. During electrolysis of water-glycerin solutions, antimony was separated from the complex anion on the cathode: (19) SbO3C3H5OH- + H2O + 3e = Sb + 2OH- +C3H5O2OH2Thus, the technological scheme is proposed for producing antimony trioxide or cathode antimony (Fig. 2). Realization of the technology will make it possible to begin development of the Sentachansk deposit (Yakutia). The bacterial transformation of Sb2S3 into Sb2O3 from gold-antimony concentrate favours production of the material suitable for cyanidation. In this case, several advantages can be expected: 1. An increase in market cost of antimony in concentrate due to its production in the form of trioxide (up to 83.53%), decrease in volumes of ore-mass transportation, and reduction in arsenic content in bacterial processing. 2. Reduction in time for gold and silver return to metallurgical conversion. 3. Removal of arsenic from concentrates, which will favour improvement of their quality and release of gold-bearing minerals. 4. Creation of small-scale production of antimony trioxide and diluted sulfuric acid suitable for ore processing and other purposes directly in the deposit. The bacterial transformation of antimony sulfide in trioxide from gold-antimony concentrates produced material applicable for the cyanidation. Herewith possible expect a number of essential advantages. 4.

CONCLUSIONS 1. The most advanced results in bioleaching of minerals were echieved with thiobacteria on sulphides of iron and some non-ferrous metals. These data showed the extent of effective application of bacterial strains as compared to sodium sulphide in antimony leaching. Furthermore, the suggested technology of antimony leaching has a number of advantages. Antimony recovery, as well as transformation of sulfates into Sb2O3, discussed here, are of great interest. 2. Increased stibium market value in the concentrate as Sb2O3 (up to 83.53%) in contrast with Sb2S3 (71.69%) and reduced volumes of transportation, reduction of arsenic contents in the process of biological conversion. 3. Increased return of gold and silver to metallurgical processing. 4. Removal of arsenic from concentrates promotes both: a raise of concentrate quality and opening gold-containing minerals.

REFERENCES

1. E.A. Kozlovskiy, Mineral and Raw Material Problems in Russia on the Eve of XXI Century (State and Prediction). Antimony. Mosk. Gos. Gorn. Universitet, Moscow, Russia (1999). 2. S.M. Mel'nikov (ed.), Antimony. Metallurgia, Moscow, Russia (1977). 3. Zhao Tian-cong, The Metallirgy of Antimony, Central South University of Technology Press, China (1988). 4. S .Ubaldini, F. Veglio, P. Fornari, C. Abbruzzese, Hydrometallurgy, 57 (2000) 187. 114


5. M.A. Meretukov and A.M. Orlov, Metallurgy of Noble Metalls, Foreign Experience, Metallurgia, Moscow, Russia (1991). 6. P.M. Solozhenkin, S.V. Usova, T.N. Aknazarova, and R. R. Fazylova, Tzvet. Metally, 1 (1994) 23-26. 7. P.M. Solozhenkin, Proceedings of the XIX IMPC, (1995) 223-226. 8. P.M. Solozhenkin, V.P. Nebera, and I.G. Abdulmanov, Proceedings of the XX IMPC, 4, Aachen (1997) 227-237. 9. P.M. Solozhenkin, V.P. Nebera, and I.G. Abdulmanov, Proceedings of the 7th International Symposium on Mineral Processing, Balkema, Rotterdam, Brookfield (1998) 495-500 10. P.M. Solozhenkin and V.P. Nebera, Proceeding of the IV International Conference on Clean Technologies for the Mining Industry, 1, University of Concepcion, Chile (1998) 399-407. 11. P.M. Solozhenkin, N.N. Lyalikova-Medvedeva, Journal of Mining Science, Vol. 37, No. 5 (2001) 534-541. 12. P.M. Solozhenkin, N.N. Lyalikova-Medvedeva, Physical technical problems of mining, No. 5 (2001) 95. 13. C. Abbruzzese, S Ubaldini, F. Veglio, Acta Metallurgica Slovaca, Special Issue, 4, 1 (1998) 19-24. 14. S. Ubaldini, F. Veglio, L. Toro, C. Abbruzzese, Minerals Engineering, Vol. 13, No. 14-15 (2000) 1641. 15. V.P. Nebera , P.M. Solozhenkin, N.N. Lyalikova-Medvedeva, Proceeding of the 7th International Conference on Mining, Petroleum and Metallurgical Engineering (MPM’7-Assiut) V. II Metallurgy & Mineral Processing, Assiut University, Egypt (2001), 295-303. 16. G.I. Karavaiko, G.V. Sedel'nikov, R.Ya. Aslanukov, U.U. Savari, V.V. Panin, E.V. Adamov, and E.F. Kondrat'eva, Tsvet. Metally, No. 8 (2000) 20-26. 17. B.K. Kendzhalov, G.V. Semenchenko, T.O. Ospanbekov, Ch.G. Absalyamov, M.M. Ignat’ev, Strain of bacteria Pseudomonas aureofaciens T 10 IMI, desolving gold from sulfide ores and concentrates, Kazakstan Patent PP¹ 11409 (2002). 18. F. Veglio, S. Ubaldini, EJMP&EP, Vol. 1, No. 2, (2001) 103-112. 19. A.F. Panchenko, V.V. Lodeyshchikov, V.Ya. Bivaltcev, Development of productive forces of Sybiria, Krasnoyarsk. V. 1, Pt. II (1985) 354-358. 20. V.V. Lodeishchikov, State of Investigations and Practical Developments in the Field of Bioliydrometallurgical Processing of Rebellious Gold-Bearing Ores and Concentrates: Review. Irgiredmet, Irkutsk, Russia (1993). 21. V.V. Lodeishchikov, A.F. Panchenko, 0.D. Khmel'nitskaya and L.P. Semenova, Ore beneficiation, Collected Works, Issue 5, Irkutsk, Russia (1994).

115


Bioleaching of Argentinean sulfide ores using pure and mixed cultures Frizán V.♦, Giaveno A.♦, Chiacchiarini P.♦, Donati E.♠ ♦

Facultad de Ingeniería, Universidad Nacional del Comahue, Buenos Aires 1400 (8300) Neuquén, Argentina* ♠ Centro de Investigación y Desarrollo de Fermentaciones Industriales (CINDEFI-CONICET), Facultad de Ciencias Exactas, Universidad Nacional de La Plata, 47 y 115 (1900) La Plata, Argentina Tel/Fax +54 221 4833794. E-mail: [email protected] Abstract The objective of this work was to evaluate the efficiency of the bioleaching of three different Argentinean sulfide ores using pure and mixed cultures of Acidithiobacillus ferrooxidans and Acidithiobacillus thiooxidans. The samples used were obtained from La Silvita, La Resbalosa and Mallín Quemado ores. Their main constituents are quartz galena, sphalerite, and chalcopyrite and their chemical composition includes 8.98% Zn, 0.046% Cu, 5.86% Pb and 0.55% Mn (La Silvita sample), 17% Zn, 0.064% Cu, 2.37% Pb and 0.56% Mn (La Resbalosa sample) and 7.95% Zn, 0.022% Cu, 64.7% Pb and 1.5% Mn (Mallín Quemado sample). Bioleaching experiments were carried out in glass columns with the percolation of the medium through the minerals. Solubilized metals (zinc, copper, manganese and total soluble iron) were determined using an atomic absorption spectrophotometer while iron(II) was measured by titration. Solid residues recovered by filtration were analyzed by means of X-ray diffraction. Metal recoveries from La Silvita and La Resbalosa were significantly enhanced by inoculation. The highest extraction was reached using Acidithiobacillus ferrooxidans.

Keywords: bioleaching, sulfide ores, zinc, acidithiobacillus 1.

INTRODUCTION Bioleaching is applied to ores which cannot be treated economically by conventional processes like flotation and roasting [1]. Acidithiobacillus ferrooxidans and Acidithiobacillus thiooxidans bacteria are capable of acting directly or indirectly on metallic sulfides, oxidizing the sulfides to sulfate and thus releasing the metals in those cases in which the respective sulfates are soluble. For this reason, these microorganisms were traditionally used in bioleaching processes of low-grade ore on a commercial scale [2].

* The authors wish to acknowledge the financial support provided by Agencia Nacional de Promocion Cientifica y Tecnologica (PICT99) and Universidad Nacional del Comahue, both of them from Argentina.

117


Although there are several minerals amenable to bioleaching or biooxidation on a commercial scale, only two metals, copper and gold, are currently recovered using this technology. Nowadays, important efforts are being done to adapt heap biooxidation technology to treat sphalerite concentrates [3]. On the other hand, in western Patagonia (Argentina) there are some reservoirs containing important amounts of base metal sulfides, which could be suitable to biohydrometallurgy commercial application. Therefore, it is interesting to study the application of this technique at these ores thoroughly. The objective of this work was to evaluate the efficiency of the bioleaching process of three different complex Zn-Mn-Fe-Pb sulfide ores from Neuquén (PatagoniaArgentina) using pure and mixed cultures of A. ferrooxidans and A. thiooxidans. 2.


2.1 Microorganisms and media Pure and mixed cultures of A. ferrooxidans (DSM 11477) and A. thiooxidans (DSM 11478) were used throughout this study. The first strain was cultivated routinely in 9K medium of initial pH 1.80 [4]. The cells were harvested when the culture had consumed 90% of the iron (II) available. The culture was filtered through blue ribbon filter paper to retain the jarosite deposits and then through a filter Millipore of 0.22 microns in order to retain the cells. Cells were washed at least twice with iron-free 9K medium and finally resuspended in the same medium pH 1.8. The second strain was cultivated routinely in iron-free 9K medium with sulfur as energy source. The cells were collected when the pH descended below 1.0. The procedure for the preparation of the inoculum was similar to the one applied for the other strain. These suspensions (with bacterial populations of approximately 2x108 cells/ml) were used as inoculum at the 10%v/v. 2.2 Mineral The samples used were obtained from La Silvita, La Resbalosa and Mallín Quemado ores (Province of Neuquén, Patagonia Argentina). Chemical Analysis and Mineral Composition of the samples are given in Table 1. Table 1. Chemical Analysis and Mineral Composition Mineral

Fe (%)

Zn (%)

Pb (%)

Mn (%)

Cu (%)

Mineral Composition

La Silvita

15.03

8.98

5.86

0.54

0.046

ZnS 5.5%, PbS 3%, FeS2 12%, CuFeS2 0.5%

La Resbalosa

9.34

17.01

2.37

0.56

0.064

ZnS 12%, PbS 2%, FeS2 8%, CuFeS2 1%

Mallín Quemado

2.94

0.795

64.75

1.55

0.022

PbS 96%

2.3 Detection of indigenous bacteria In order to detect sulfur or iron oxidizing bacteria two tests in 250 ml shake flasks on a rotatory shaker at 180 rpm and 30 ± 0.5 ºC were carried out. Each flask, containing the ore (pulp density of 10%), was respectively filled with 100 ml of the appropriate liquid medium (9K iron free pH=3.0 supplied with 1% w/v of sulfur powder or 9K medium pH=1.8).

118


2.4 Experiments in columns Tests were conducted in 50 mm inside diameter glass columns with 260 mm high at constant temperature of 30ºC ± 0.5ºC. Columns have a perforated plate in the bottom and a layer of glass wool to support the ore sample. Thirty grams of each ore sample with particle size ranging between 10 and 16 mesh were used in every column under flood conditions. Each column was charged with 300 milliliters of free-iron 9K medium at initial pH 1.8. Percolating solution was recirculated from the open top of the column to the bottom by continuous airflow rate at 1.2 VVM. In order to reach a pH condition compatible with the bacterial growth it was necessary to achieve an acid stabilization before the inoculation. Therefore, the pH value was initially adjusted at 1.8 by adding drops of sulfuric acid solution (4.8 N) and then it was not controlled again. The amount of sulfuric acid added was used to calculate the initial acid consumption. After the pH stabilization, the inoculation was done. Sterile control columns were prepared replacing the same volume of inoculum by a solution of 2% timol in methanol. Additionally, uninoculated control columns were prepared to check indigenous bacterial activity. Sterile distilled water was added to compensate evaporation. Samples from every column were taken at regular intervals. 2.5 Analytical methods Copper, total iron, manganese and zinc in solution were determined by atomic absorption spectrophotometry. Iron (II) concentration was determined by permanganimetry. Bacterial populations in solution were determined using a PetroffHausser camera in a microscope with a contrast phase attachment. This determination was not representative of the bacterial growth because the cells attached to the mineral were not determined. Both the redox potential and the pH were measured with specific electrodes Sulfuric acid production was analyzed by titration with sodium hydroxide solution. Solid residues were recovered by filtration and analyzed by X-ray diffraction (XRD) in Rigaku DII-Max equipment. 3.


3.1 Detection of indigenous bacteria After 25 days, ferrous iron was completely oxidized, but no free cells in media were observed at microscope. On the other hand, the pH values in the systems supplemented with sulfur did not decrease. Moreover the pH remained almost constant in the case of La Silvita (pH 3.5) whereas it increased in La Resbalosa and Mallín Quemado (pH 4.9). Since it was not possible to detect free bacterial cells, ferrous iron was oxidized more slowly than the usual by A. ferrooxidans and cells were not able to oxidize sulfur as energy source, it is possible to suggest that the indigenous bacterial activity was due to the presence of Leptospirillum ferrooxidans-like microorganisms. As it is known, these bacteria grow preferentially adhered to surfaces and their physiology is consistent with the result observed [5]. 3.2 Experiments in columns The experiments took one hundred days. After that period, the solid residues were removed from the columns and analyzed by XRD. The main mineralogical species originally present in the ores remained until the end of all leaching process indicating that 119


the oxidation of sulfides associated with the dissolution of metals was not complete. Additionally, after the leaching processes it was possible to identify sulfur and jarosite from La Silvita and La Resbalosa ores both in the biotic and abiotic systems. Galene present in the ore treated was oxidized to anglesite only in the biotic systems showing that bioleaching processes were important to improve this phase transformation. The initial sulfuric acid consumption before the inoculation was: 39.2 g H2SO4/kg ore to La Silvita, 40.5 g H2SO4/kg ore to La Resbalosa and 43.5 g H2SO4/kg ore to Mallín Quemado, which indicates the presence of slightly alkaline gangue content such as carbonate minerals. 3.3 La Silvita Figure 1 shows the percentages of zinc and manganese solubilized from La Silvita ore during the test. The highest zinc extraction (75%) was obtained in the column with A. ferrooxidans. In addition, zinc solubilization in the column with mixed culture (48%) was significantly higher than the one with A. thiooxidans (22%). The performance of the last column was similar to the uninoculated one (21%) throughout the experiment. The zinc extraction in the sterile column only reached 7%. These results suggest that A. ferrooxidans plays a key role in the bioleaching process for La Silvita ore contributing to release the zinc from the sphalerite. Meanwhile A. thiooxidans as pure or mixed cultures was not so efficient to improve the Zn solubilization. This was probably due to the mineralogical species present in this natural ore, since the extraction obtained in this work was lower than those reported using synthetic sulfide [6]. In contrast with the zinc solubilization, the amount of manganese released from La Silvita ore was higher when A. thiooxidans was present in the cultures. Figure 1 shows that the behavior displayed by this pure culture was very similar to the mixed one, reaching in both cases a manganese extraction close to 100%. Since the manganese solubilization was higher in presence of A. thiooxidans or mixed cultures and considering that these systems reached pH values lower than the other systems (Fig. 2), manganese could be present as a mineralogical species easily leachable by acid. This hypothesis could not be confirmed by XRD analysis probably due to the low amount of manganese present in this ore. In Figure 2, the pH evolution can be observed. During the first twenty-five days, all columns showed an increase of pH values indicating that the solubilization of some basic species present in the mineral continued beyond the inoculation. After that period, pH values decreased in all biotic system around 1.7 while the pH value increased to reach a value of 2.6 in the sterile system. These results could indicate that sulfur obtained from the redox dissolution of sulfides and identified by XRD analysis of leached residues, was partially oxidized by bacterial action and contributed to the acid production. Therefore, the whole bioleaching process had a positive acid balance. Additionally, in Figure 2, the redox potential evolution can be observed. Eh rapidly increased during the first days of the experience in columns inoculated with A. ferrooxidans reaching values over than 550 mV. Then, Eh values oscillated around this threshold until the end of the experiment. The uninoculated system took more than twenty days to reach the same final value. In the sterile column Eh remained constant close to the initial value.

120

Bioleaching Applications Af At Af/At Uninoculated Sterile

80

60

60 40

2+

40

2+

Zn Solubilization (%)

80

100

Mn Solubilization (%)

100

20 0

0

20

40

60

80 100

20 0

0

20

40

60

80

100

Time (days)

2.6 2.4

Eh (mV)

Figure 1. Comparison of zinc and manganese extraction from La Silvita ore using pure and mixed cultures of A. ferrooxidans (Af) and A. thiooxidans (At) with uninoculated and sterile control systems 600 500 400

0

20 40 60 80 100

Time (days)

Af At Af/At Uninoculated Sterile

pH

2.2 2.0 1.8 1.6 0

20

40

60

80

100

Time (days)

Figure 2. Evolution Eh and pH values from the La Silvita under different conditions: A. ferrooxidans (Af) cultures, uninoculated system and sterile control

Figure 3 shows the evolution of total soluble iron and ferrous iron concentrations in columns inoculated with A. ferrooxidans, uninoculated and sterile controls. The iron evolution was completely in agreement with Eh values shown in Figure 2. Ferrous iron was rapidly oxidized in A. ferrooxidans column and slightly slower in the uninoculated one. On the other hand, the ferrous iron remained reduced in the sterile controls reaching 2.5 g/l. Additionally a significant amount of iron was released from La Silvita ore during the leaching process mainly in A. ferrooxidans column (12 g/l). 121


Fe(II) (g/l)

12 10

FeTotal(g/l)

1 0 0

8

Af Uninoculated Sterile

2

20 40 60 80 100

Time (days)

6 4 2 0

0

20

40

60

80

100

Time (days)

Figure 3. Evolution of total iron and ferrous iron concentrations from La Silvita under different conditions: A. ferrooxidans (Af) cultures, uninoculated system and sterile control

These results show: i) A. ferrooxidans cultures remained active along the experience and the cells in suspension reached a value of 2.5x108cells/ml after one hundred days of operation. ii) This microorganism contributed to increase the solubilization of pyrite and other iron species. And iii) the bacterial activity detected in the uninoculated column was probably due to another ferrous iron oxidizing L. ferrooxidans-like bacteria. From Figures 2 and 3 the total iron dissolved with the pH evolution can be correlated either in the sterile or biotic systems. The amounts of iron dissolved were minimal when the values of pH were maximum. Correspondingly, an abundant amount of jarosite and other ferric oxyhydroxides covering solid residues was visually observed and then detected by X ray diffraction analysis. 3.4 La Resbalosa Figure 4 shows the percentages of zinc and manganese solubilized from La Resbalosa ore during the test. In the column with A. ferrooxidans the highest extraction of zinc was obtained, reaching 17.5%. In addition, zinc solubilization in the column with mixed culture (12.4%) was slightly higher than the one observed in the column inoculated with A. thiooxidans (9.7%) and in the uninoculated system (9.3%). The sterile control only removed 2.6% of zinc initially present in the ore. Figure 5 shows pH evolution and total iron solubilized throughout the experience. When the pH values increased, the iron in solution decreased. In A. ferrooxidans inoculated column the solubilization of different basic mineralogical species present in the ore caused an increase of pH values of over 3.3. After that, an important precipitation onto the mineral surface was observed. Moreover, the zinc extraction and the bacterial counts (1x108 bacteria/ml) were lower than those obtained from La Silvita. It was probably associated with diffusional barriers due to the large amount of brown amorphous ferric precipitated, which avoided further sphalerite dissolution. The inoculated systems with A. thiooxidans, as pure or mixed cultures reached the highest manganese solubilization (Fig. 4). Although La Resbalosa and La Silvita ores 122


showed a similar behavior, in the first ore the percentage of manganese extraction was only 70%. This was probably due to the fact that pH values reached in La Resbalosa systems were higher than those obtained in La Silvita. Af At Af/At Uninoculated Sterile

20

70 Mn Solubilization (%)

10

5

0

60 50

2+

2+


15

80

0

20

40

60

80 100

40 30

0

20

40

60

80

100

Time (days)

Figure 4. Comparison of zinc and manganese extraction from La Resbalosa using pure and mixed cultures of A. ferrooxidans (Af) and A. thiooxidans (At) with uninoculated and sterile control systems 4

pH

1200

Fe(total) Concentration (mg/l)

1000

Af At Af/At Uninoculated Sterile

3 2

800

0

20 40

600

60 80 100

Time (days)

400 200 0

0

20

40

60

80

100

Time (days)

Figure 5. Evolution of total iron and pH from La Resbalosa using pure and mixed cultures of A. ferrooxidans (Af) and A. thiooxidans (At) with uninoculated and sterile control systems 3.5 Mallín Quemado Figure 6 shows the evolution of zinc solubilization from Mallin Quemado ores. In contrast to the bioleaching experiences using La Silvita and La Resbalosa ores, Mallin Quemado reached higher Zn extraction when A. thiooxidans cultures were present. On the 123


other hand, the Zn solubilization was very similar in the A. ferrooxidans inoculated, uninoculated and sterile systems, and the metal extractions were not relevant in all these flasks. pH values at the end of the test were 2.0 in A. thiooxidans pure and mixed cultures and 2.3 in the other biotic systems while the sterile system reached a pH of 3.5. Eh values were increasing according to ferrous iron oxidation (data not shown). Af At Af/At Uninoculated Sterile

18

24

14

Mn Solubilization (%)

12 10 8

2+

2+


16

28

6 4

20 16 12 8

0

20

40

60

80 100

0

20

40

60

80

100

Time (days)

Figure 6. Comparison of zinc and manganese extraction from Mallín Quemado ore during leaching experiences using pure and mixed cultures of A. ferrooxidans (Af) and A. thiooxidans (At) with uninoculated and sterile control systems

The manganese extraction was higher in the systems inoculated with A thiooxidans cells. The different percentages of manganese removed could be attributed to the final pH in each system. The final percentages of Zn and Mn solubilized from Mallín Quemado ore were lower than those obtained using the other minerals. Bacterial counts were also very low (0.6 x 108 bacteria/ml) when this high-grade galena ore was tested. These results could indicate that the great lead percentage in the sample may inhibit the bacterial growth. Meanwhile the low solubility of lead (II) reduced the possibility of toxicity when it is present in lowgrade ores as La Silvita and La Resbalosa [7]. 4.

CONCLUSIONS The bioleaching treatment of La Silvita complex sulfide ore by bioleaching process appears to be technically feasible, since the zinc solubilization was increased ten fold when A. ferrooxidans was used as inoculum. Although La Resbalosa tests showed that the bioleaching processes were amenable, they could be improved if a rigorous pH control is done to avoid great oxides precipitation onto the mineral surface. Mallin Quemado tests did not show significant zinc extraction but it was possible to oxidize galene. Further tests should be carried out to analyze the extension of this transformation and the effects of lead toxicity on the bacterial grow. In summary, the results analyzed here are good enough to consider the metal biohydrometallurgical extraction as a promissory method to be applied for these regional ores to be treated using bacterial heap leaching technique. Moreover, future studies

124


involving isolation and identification of indigenous bacteria should be done in order to improve the bioleaching treatment of these Argentinean ores. REFERENCES

1. W. Sand, T. Gehrke, R. Hallmann, K Rohde, B. Sobotke and S. Wentzien. Biohydrometallurgical Technologies, E. Torma, J.E. Wey and V.L. Lakshmanan (eds). The Minerals, Metals & Materials Society (1993) 15. 2. D.E. Rawlings. Biomining: Theory, Microbes and Industrial Processes, SpringerVerlag, Berlin, 1997. 3. T.J. Harvey, W. Van Der Merwe K. and K. Afewu, The applicacion of the GeoBiotics GEOCOAT biooxidation technology for the tratment of sphalerite at Kumba resources´ Rosh Pinah mine. Minerals Engineering15 (2002) 823. 4. M.P. Silverman and D.G. Lundgren, J. Bacteriol., 77 (1959) 642. 5. W. Sand, K. Rohde, B. Sobotke and C. Zenneck. Evaluation of Leptospirillum ferrooxidans for leaching. Appl. Environ. Microbiol. 58 (1992) 85. 6. M. Pistorio, G. Curuchet, E. Donati and P. Tedesco. Direct zinc sulphide bioleaching by Thiobacillus ferrooxidans and Thiobacillus thiooxidans. Biotechnol. Lett. 16 (1994) 419. 7. J. Barrett, M.N. Hughes, G.I. Karavaiko and P.A. Spencer. Metal extraction by bacterial oxidation of minerals. Ellis Horwood Limited, 1993, 28.

125


Bioleaching of complex gold-lead ores Z. Ulberg, V. Podolska, A. Yermolenko, L. Yakubenko and N. Pertsov Ovcharenko Institute for Biocolloidal Chemistry, NAS of Ukraine, 42 Vernadsky blvd., 03142 Kyiv, Ukraine Abstract The present work is relevant to bioleaching of galena from gold-lead ore using Acidithiobacillus such as Thiobacillus ferrooxidans and Thiobacillus thiooxidans in the presence of magnetite as well as under action of DC electric field, which resulted in the acceleration of sulfide dissolution from the natural and synthetic galena. Various electrochemical mechanisms in the leaching process were considered. The electrokinetic properties of the thiobacteria under the conditions of galena microbial leaching were also studied.

Keywords: galena; bioleaching; magnetite; galvanic couple; zeta-potential 1.

INTRODUCTION In recent years, the use of microorganisms for metal solubilization from ores has increasingly attracted the attention of hydrometallurgists and biotechnologists. Microbial leaching is characterized by low cost, and its realization continues to create fewer problems compared to corresponding hydrometallurgical or pyrometallurgical processes. The bioleaching method is based on the ability of some microorganisms to oxidize Fe (II) ions or reduced sulfur compounds. As a result of the accumulation of sulfuric acid in the biosuspension, the decrease in pH and the metal solubilization from sulfides takes place. The cultures Thiobacillus ferrooxidans and Leptospirillum ferrooxidans are most-used [1]. Galena is the main industrially available source of lead. Hydrometallurgical processes for lead leaching have been studied by a number of researchers [2]. Quite a few studies have been dedicated to the use of microorganisms for lead leaching from lead-bearing ores [3, 4]. In gravitation and flotation separation of ores where the gold-bearing minerals are associated with the sulfide minerals of non-ferrous metals one obtains concentrates enriched with these metals, namely, lead. The Muzhiyev gold-mining deposit (Eastern Carpathians, Ukraine) belongs to such an ore deposits. A gravity concentrate with a galena upwards of 40% is produced at the operation. To avoid the formation of matte (melted sulfide), which is capable of dissolving a significant part of the gold during melting, the cleaner concentrate needs to be subjected to either an oxidizing roasting or some other process to remove the sulfur, arsenic, and antimony. Roasting, however, volatilizes sulfur and arsenic. The pyrometallurgical method for processing sulfide concentrates, therefore, has limits as to its application both economically and from the standpoint of environmental safety. Microbial oxidation of sulfide minerals (concentrates) is proposed as a viable alternative to the pyrometallurgical method. 127


Tomizuka and Yagisawa [4] examined the role of bacteria and proposed a base scheme for galena leaching in the presence of T. ferrooxidans; the leaching mechanism includes the following reactions: 2PbS + 2H2SO4 + O2 → 2PbSO4 + 2H2O + 2S

(1)

(2) 2S + 2H2O + 3O2 → 2H2SO4 According to these authors, at low pH values sulfur can serve as the sole energy source for the autotrophic microorganisms. The elemental sulfur is released as a result of the electrochemical reaction with oxygen (reaction 1), and further, sulfur serves as a substrate for thiobacteria which produce sulfuric acid (reaction 2). The aim of this investigation was to establish whether it is possible to improve the kinetics and the process efficiency of galena microbial leaching from the gold-bearing ore of the Muzhiyev deposit by the addition of natural iron-bearing raw material (for example, magnetite). Attention was also paid to the electrosurface properties of thiobacteria and galena in the microbial leaching process. 2.

MATERIALS AND METHODS The middlings of gravity separation (MGS) of the gold-bearing ore from the Muzhiyev deposit, which was produced in a Knelson centrifugal concentrator, was investigated. The phase composition of this sample was determined by four main components: galena PbS (∼42%), barite BaSO4 (∼20%), pyrite FeS2 (∼6%), and quartz. The sample also contained individual gold grains. MGS was very similar to the concentrate and differed from it only by low gold content. The sample was ground -0.063 mm. The mixed culture, containing mainly T. ferrooxidans as well as T. thiooxidans, was used for the microbial leaching. The culture was cultured from the samples of the same deposit and was adapted to a 10% (w/v) galena pulp density for 10 weeks. The accumulated culture was grown in basal 9K medium. The experiments were conducted in 0.5-L shaker flasks with 10 g of MGS, 2.5-15 g of magnetite, 50 ml of cell biosuspension (inoculum)) and 100 ml of 9K medium without iron at pH 2.1. The stirring speed was 150 rpm. The magnetite sample contained 64.5% of Fe3O4; the moisture content of the product was 10.5%. The efficiency of the microbial leaching (%) was estimated from the sulfidesulfur content (% w/w) of the dried mineral sample before and after treatment by bacteria; the iodine method of analysis was used. For this purpose the sample was dissolved in HCl; released hydrogen sulfide was absorbed by ammonia solution of zinc sulfate. The precipitate formed was dissolved in the mixture of HCl and titrated iodine solution; the sulfur quantity was determined on the iodine excess in the solution. The Fe3O4 dissolution was assessed on the Fe2+ and Fe-total concentration by colorimetric method with ortophenanthroline. The PbS electrode for rest potential investigations was prepared in the following way. The side surface of a circular graphite electrode with a cross-sectional area 0.3 cm2 was embedded in epoxy insulating glue. The working face surface was first polished and then rubbed with a finely dispersed powder of the synthetic PbS. Excess powder was removed with distilled water. Each electrode was placed in a vessel with 9K medium and bacterial inoculum. Stirring in the vessels with the biosuspension was accomplished with a magnetic-stirrer. The rest potential measurement was made with each electrode using an Ag-AgCl reference electrode. Zero time was considered to be the "control" without the microorganism action (Table 2). 128


The experiments on the influence of an electric field on the rate of leaching were carried out in the following way. Suspensions containing either MGS or synthetic PbS, the mixed thiobacteria culture, and 9K medium with 44.5 g/L iron (II) sulfate at pH 2.1 were prepared. All vessels containing the suspension were incubated on the shaker. Two vessels, one containing synthetic PbS and one containing MGS, served as controls; no electric field treatment was applied to the control vessels. The graphite electrodes were introduced into the other two vessels, and periodically an electric field with a voltage of 0.5 V/cm (anode potential was +0.22V C.S.E.) was applied. The duration of the electric field treatment was 5 min. followed by a pause of 10 min. The treatment by the discontinuous DC field was conducted in vessels with non-separated cathode and anode chambers. The electrophoretic mobility of T. ferrooxidans M1 was measured by the microelectrophoresis method in different buffers: in the citrate buffer of McIlven and in universal buffer mixture composed of phosphoric, acetic, and boric acids; these measurements were recalculated into the electrokinetic potential (ζ) using the Smolouchowsky formula. All the said experiments were carried out at least by duplicate. 3.


3.1 Galena oxidation in mixed biomineral suspension As seen from data given in Table 1 the MGS sample placed in the iron-free 9K medium was slowly oxidised. The galena oxidation reaction in water is described by this equation:

PbS + 4H2O → PbSO4 + 8H+ + e (3) The formation of insoluble anglesite (PbSO4) promotes the reaction shift to the right. However, the formation of an insoluble film on the surface of the mineral particles decreased the kinetics of the oxidation process. As a result, about 21% of the sulfides was oxidized from the original mineral sample in iron-free 9K medium in nine days. The second sample, which contained thiobacteria in the iron-free 9K medium with the MGS sample, was significantly oxidized. In three days about 73% of the sulfide was oxidized; in 9 days 82.5% of sulfide was oxidized. As can be seen, thiobacteria significantly improved both the oxidation kinetics and degradation of the mineral. Chemically produced sulfur (equation 1) served as the substrate for the T. thiooxidans, which metabolized it to sulfuric acid (equation 2). The data given in Table 1 for the sulfide oxidation (item 2) indicate that galena can serve as the energy source for the mixed culture resulting in degradation of the galena-containing ore. That is a reason why the degree of the sulfide destruction in the second sample considerably exceeded that of the first sample that did not contain bacteria. The data on the change in rest potential of lead sulfide with an increase in bioleaching time is testimony to the changes in the sulfide surface composition and the electrochemical properties of the surface. Table 2 gives the values for the electrode made from natural galena [5] and the electrode made from the synthesized PbS from day 0 through 14 days of leaching with the thiobacteria. The interaction between the mineral and thiobacteria was accompanied by the increase in rest potential with respect to that measured in fresh nutrient medium. The positive potential increases with the bioleaching time. Comparing the galena mineral and PbS electrodes, it is seen that pure product was oxidized more intensively because it had higher positive potential values. Note, however, 129


that the rest potentials for both the mineral and powder electrodes were not large. The reason for this phenomenon was discussed earlier; it is the formation of reaction products on the electrode surface diminishes oxidation. Table 1. The sulfide oxidation and Fe3O4 dissolution the mixed biomineral suspension No. and sample composition

S2S2- in the oxidised ore Sulfide oxidation (%) before (%) leaching In 3 days In 9 days In 3 days In 9 days (%)

Fe2+ /Fe3+

Fe3O4 dissol. (%)

1. MGS

5.6

5.2

4.4

7.2

21.5

-

-

2. MGS + bacteria

5.6

1.5

0.8

73.2

85.7

-

-

3. MGS + 2,5 g of Fe3O4

4.5

3.8

2.8

25.6

37.8

9.8

51

4. MGS + bacteria + 2,5 g of Fe3O4

4.5

0.6

0.5

86.7

89.0

4.0

65

5. MGS + bacteria + 5 g of Fe3O4

3.7

0.15

0.13

96.0

96.5

3.0

71


2.8

0.31

0.30

88.9

89.3

6.6

50.2


2.2

-

0.3

-

86.6

7.1

59

Table 2. Stationary potential of PbS-electrode in the 9K medium with microorganisms Bioleaching time, days 0 2 7 9 14

Rest potential of natural galena [5], mV (C.S.E.) 5 75 110

Rest potential of PbS-electrode, mV (C.S.E.) 27 68 97 140 159

Now we shall address the experiments in which magnetite was introduced simultaneously with thiobacteria. It is known that iron (II) as well as the reduced sulfur species can serve as an energy source for thiobacteria; this property is widely used for the microbial leaching of ores that contain sulfides with iron (e.g. pyrite, arsenopyrite, chalcopyrite, and pyrrotite) [6]. The majority of thiobacteria can use these substrates in their metabolism. The role of the added magnetite may iron (II) generation from the dissolution of Fe3O4. The autotrophic bacteria, in turn, form the oxidizing medium by generating the Fe (III) ions. Galvanic interactions among different minerals are known and sometimes used to promote the leaching of sulfide minerals [7-8]. The electrochemical concept of many galvanic leaching systems is the basis of these processes [9]. In our case the role of the added magnetite in promoting galena leaching from ore may be due to coupling the electrochemical reactions of magnetite reduction and galena oxidation. The cathodic reaction under acidic conditions could be: Fe3O4 + 6H+ + 2e → FeO + 2Fe2+ + 3H2O 130

(4)


The anodic reaction of galena oxidation to sulfur and then to sulfate could proceed according to polysulfide mechanism [10]: PbS + 4H2O → PbSO4 + 8H+ + 8e (5) However, the run of such galvanic pair can be retarded due to the formation of sulfur film on surface of galena particles and to low Fe3+ concentration in solution. Indeed, the sulfur leaching from galena in the magnetite presence was higher as compared in the absence of this addition, but in the whole remained still low (item 3, Table 1). The availability of the autotrophic bacteria and their metabolites promote galena oxidation by providing the Fe (III) generation and increasing the oxidizing conditions in a bioleaching system in the following way: Fe2+ bacteria Fe3+ + e, (6) as well as accordingly equation 2. Natural magnetite in 0.5 M sulfuric acid has a wide range, 0.5-1.4 V (S.H.E.), in which complete passivation occurs. Increasing the current during the cathodic scans at more negative potentials of 0.5 to –0.1 V may be necessary for the process described by equation 4. The mineral suspension, which contained sulfide ore, magnetite and thiobacteria culture had a positive Eh (0.2-0.4 V). Therefore, oxidative destruction of galena as well as microbial sulfuric acid production are possible. Superimposing several processes can have a number of consequences. For example, if the cell concentration is not high and these mechanisms have competitive character, then the appearance of the new substrate (Fe2+) for autotrophic bacteria can result in the decrease of the sulfuric acid production by microorganisms. On the other hand, the appearance of the new substrate can result in the increase of the cell concentration and then an acceleration of the sulfur oxidation rate. The appearance of the Fe3+ also must be accompanied by the acceleration of the solubilization process and greater dissolution of sulfide. The results of Table 1 (items 4-7) show that after magnetite introduction in the suspension, which included MGS, bacterial inoculum, and iron-free 9K medium, galena dissolution increased considerably. In three days sulfide oxidation reached ∼86-90%; this level of oxidation was not achieved until day 9 in the suspension without magnetite (sample 2). A number of experiments with different magnetite content, ranging from 2.5 to 15.0 g in the suspension, have been carried out. The best result was achieved in tests with 4 and 5 g of magnetite. In these tests, the sulfide oxidation level reached ∼96% after three days; little change in the oxidation level was observed with additional incubation. The sulfide content in the solid phase, the total concentration of the dissolved iron, and the Fe2+ concentration in incubation solution were all measured allowing assessment of magnetite degradation. As noted in sample 5 (Table 1), the magnetite was subjected to the highest solubilization level (∼71%) with a corresponding high degree of galena oxidation. The maximum concentration of oxidized iron Fe3+ was also achieved in this test. The results obtained lead to the conclusion that the addition of 2.5-5.0 g of magnetite for up to 10 g of MGS positively influenced galena leaching from ore. Addition of this mixed iron oxide promoted oxidizing conditions because of iron sulfate generation. The introduction of a larger quantity of magnetite also positively influenced the kinetics and efficiency of the sulfide leaching in relation to the control. However, with greater magnetite addition the Fe3O4 degradation declined as well as the oxidized iron concentration. Hence, it is possible to conclude that the increase in the magnetite 131


concentration in the suspension is not expedient; this would lead to a decrease in the redox-potential due to the cathodic process activation and will mean non-productive additive use. 3.2 Galena electro-bioleaching We also performed experiments on the influence of an electric field on the rate of the lead sulfide and galena-bearing ore bioleaching. Table 3 shows sulfide analysis results for the four experimental tests. Within two days those tests in which the electric field treatment was applied exhibited sulfide oxidation of 98.9% for the synthetic PbS and 90% oxidation for the MGS. After four days all tests, including the controls, were analyzed for residual sulfide. The difference was notable. In the controls the degree of synthetic galena oxidation was 87% and for MGS the sulfide oxidation was 78%; this is considerably less sulfide oxidation that was observed for the corresponding samples subjected to the field action (Table 3). Table 3. Microbial galena oxidation in field 0.5 V/cm Duration, days 2 4

Degree of the sulfide oxidation, % Without field In field MGS PbS MGS 90 78 87 99

PbS 98,9 99,5

We attempt to explain the results. Our previous experiments with iron alum (NH4Fe(SO4)2·12H2O) showed that during the DC electric field treatment in the electrochemical cell with non-separated electrodes, a decrease in Fe3+ ion concentration took place due to the cathodic processes. With an increase in the cathode potential the concentration of reduced iron rose. Thus, under the action of the DC electric field in the inoculated 9K medium, Fe3+ was continuously reduced, providing a substrate for microbial oxidation of the Fe2+ to Fe3+ (equation 6) [11]. The additional substrate stimulated bacterial growth. This, in turn, enhanced sulfide oxidation, resulting in increased dissolution of the lead sulfide and increased sulfuric acid formation. It appears that in the present experiments the system of coupling the electrochemical reactions is similar to that with magnetite, which was described earlier (section 3.1) in this paper. In the tests that employ non-separated electrodes, it is important not to apply too great of an applied potential in the negative direction; such an action could increase the Fe2+ concentration too much for the level of microbial activity. It is important to maintain a low voltage field for the biosuspension. The dynamics of the oxidation-reduction potential change indicates that at the beginning of the field action on the biosuspension (about 18 hours) the Fe3+ ions were regenerated in excess and the Eh potential increased to the extent expected by thiobacteria activity under the field action. It is likely that the acceleration of the galena oxidation in the biomineral suspension is related to the action of the discontinuous DC electric field. 3.3 Electrosurface properties Some investigations were undertaken to determine if the electrosurface properties of thiobacteria play any role in the interaction of the bacteria with galena and if these properties can be regulated in enhance the bioleaching efficiency [12]. The surface chemical and electrokinetic properties of galena were studied in detail by Pugh [13]; this study focused on establishing optimal conditions for the flotation separation of sulfide ores. Main attention was paid to the mineral properties under neutral and alkaline pH 132


values corresponding to the flotation conditions. Only a few studies have been dedicated to the electrical and hydrophilic/hydrophobic properties of the thiobacteria surface [1416]. The correlation between the oxidation-reduction potential and Zeta-potential of the bacteria during leaching was determined: the minimum ζ-potential values of bacteria corresponded to the maximum values of the Eh. One peculiarity connected with the presence of lead sulfide in the cultivation medium was noted. The ζ-potential value of the bacterial cells grown on 9K medium was higher by 1-2 mV compared to the cells grown in the same medium with MGS added. This was possibly due to the presence of highly dispersed particles of PbSO4 or PbCO3, which do not possess a high electrokinetic potential [17]. During the electrocoagulation of these minerals with the cells the decrease in the electrokinetic potential of the latter may have taken place. The interaction between a mineral and a cell depends on the pH value. Fig. 1 shows the Zeta-potential dependence with pH for galena as observed by Pugh [13] and our data for thiobacteria. In an acidic, oxidizing medium (NaNO3) the synthetic galena was positively charged (curve 4); under the same acidic, oxidizing conditions the natural Swedish galena was negatively charged (curve 5) and the value of the electrokinetic potential at pH 2 to 2.5 was very close to the ζ values of the cells T. ferrooxidans M1 (curve 1-3). The difference in the surface properties of various samples of galena is associated with its surface oxidation products including PbSO4 and PbCO3. As seen in Fig. 1, in the interval of pH 2.5-3.75 the Zeta-potential values closely coincide in both buffer solutions (curves 1-3). In the citrate buffer the maximum Zetapotential value was achieved at pH 3.75, however, the charge was not changed even at pH 1.75. It is likely that for the given culture the isoelectric point (IEP) does not exist. For two of six T. ferrooxidans strains studied by Skvarla and Kupka [13] the IEP was also not achieved. It is likely that the pK value observed at 2.60-2.70 represents the mixed value of two pKs related to the dissociated phosphate groups of phosphatidic acid and carboxyl groups of gluconic acid in the composition of the lipo-polysaccharide cell wall [18]. Consequently, the cellular envelope of T. ferrooxidans M1 bacteria, grown in the medium containing iron as an energy source, consists of acidic material determined by phosphate and carboxyl groups. The number of these groups changes depending on the growth phase, the bacterial oxidative activity, and the intensity of the exchange processes on the membrane. Electro-bioleaching experiments may be a proof of this. In the tests described in Section 3.2 the bacterial cells were separated from the solutions, two times washed off in 5 N H2SO4, resuspended in McIlven buffer, and after the mentioned preparation they were subjected to ζ-potentials measurement. As seen in Fig. 2, the cells separated from the suspension subjected to the DC field action had higher negative ζ-potential values compared to non-DC field-treated cells. The ζ-potential of the cells not subjected to the electric treatment during the initial incubation period decreased slightly at first and then increased. The cells grown under the Fe2+ ion electrochemical regeneration and electrostimulation conditions showed a stable increase of ζ-potential as the cells aged. On the 9th day of the incubation (approximately 216 hours), the electrokinetic potential increased nearly two times in both the DC field-treated and non-treated cultures. The electrokinetic potential of the thiobacteria changed with increases of electrolyte concentration at a constant pH. Changes in the electrokinetic potential could not be attributed to compression of the electric double layer with increase of the electrolyte concentration. As observed in Fig. 3 (curve 1), with an increase in the Fe (II) concentration from 10-6 to 10-2 M the negative potential increased by 5 mV, indicating the 133


higher affinity between T. ferrooxidans cells and iron. However, the cells proved practically indifferent to added calcium ions (curve 3); an insignificant decrease in the ζpotential value was observed when Ca(NO3)2 was added over a broad concentration range. Lead increased the negative charge of the cell surface at high Pb(NO3)2 concentrations (10-3 – 10-2 M). This same increase in the negative charge was not reached in pulp during galena leaching due to the extremely low solubility of the lead sulfate that was formed (curve 2). The aforementioned property of the T. ferrooxidans culture may indicate that the bonding locations of Fe (II) on the cell surface are not accessible to other cations and that its oxidation mechanism is protected from metals in the surrounding medium [19]. Figure 1. Zeta-potential versus pH plots of T. ferrooxidans M1 (1-3) and galena (4-6). The experimental conditions: 1 – growing in 9K medium with MGS (measurements in universal buffer mixture); 2 – growing in 9K medium, (measurements in McIlven buffer); 3 – growing in 9K medium with MGS (measurements in McIlven buffer); 4 – PbS in 2x10-3 M NaNO3 solution; 5 – natural Swedish galena in 2x10-3 M NaNO3 solution; 6 – natural Swedish galena in 2x10-3 M NaNO3 solution plus 1x10-5 M Pb(NO3)2. Curves 4-6 according to Pugh and Bergström [13]

Figure 2. Zeta-potential versus incubation time plot of T. ferrooxidans M1 bacteria in McIlven buffer. Previously incubated under the 0.5 V/cm DC field treatment (1); and without the DC field treatment (2)

134

Figure 3. Zeta-potential versus electrolyte concentration plot of T. ferrooxidans M1 bacteria dispersed in FeSO4 (1), Ca(NO3)2 (2), and Pb(NO3)2 (3). The ζ-potential value without electrolyte addition was –13.5 mV


5.

CONCLUSIONS 1. The addition of magnetite to lead sulfide in the ratio of 1-2:4 in the microbial leaching system enhanced the efficiency and kinetics of sulfide degradation. 2. The thiobacteria substantially changed the galena surface, surface composition and increased the mineral rest potential during the leaching study. 3. Application of a discontinuous DC electric field with low voltage application to the galena bioleaching system improved the kinetics of the sulfide degradation. 4. The thiobacteria cell and the natural galena surfaces possess the same electrokinetic potential signs but low in value, which promotes their heterocoagulation. The electrokinetic potential of the thiobacteria changes during galena leaching and under the action of the DC electric field.

REFERENCES

1. G.S. Hansford and T.Vargas, Hydrometallurgy, 59 (2001) 135 2. M.C. Fuerstenau, C.O. Neto, B.V. Elango, and K.N. Han, Met. Trans. B, 18 B (1987) 25. 3. A.E. Torma and K.N. Subramanian, Int. J. Miner. Process., 1 (1974) 125. 4. N. Tomizuka and M. Yagisava, in: L.E. Murr, A.E. Torma and J.A. Brierley (Eds.), Metallurgical Application of Bacterial Leaching and Related Microbial Phenomena, Academic Press, New York, 1978, pp. 321 - 344. 5. J.L. González-Chaves, F. González, A. Ballester, and M.L. Blazquez, Minerals and Metallurgical Processing, 17 (2000) 116. 6. H.L. Erlich, Hydrometallurgy, 59 (2001) 127. 7. R.K. Paramguru and B.B. Nayak, Metals and Materials Processes, 6 (1996) 23. 8. R.K. Paramguru and B.B. Nayak, Electrochem. Soc., 143 (1996) 3987. 9. N. Jyothi, K.N. Sudha and K.A. Natarajan, Int. J. Miner. Process., 127 (1989) 189. 10. A. Schippers and W. Sand, Appl. Environ. Mocrobiol., 65 (1999) 319. 12. V.I. Podolska, A.I. Ermolenko, L.N. Yakubenko, and Z.R. Ulberg, Colloid Journal, (in press). 13. R. Pugh and L.Bergström, Colloids Surfaces, 19 (1986) 1. 14. J. Skvarla and D. Kupka, Biotecnol. Techniques, 12 (1996) 911. 15. J. Skvarla, D. Kupka, A. Naveshakova and A. Skvarlova, Folia Microbiol., 47 (2002) 218. 16. M. Misra, K. Bukka and S. Chen, Minerals Engineering, 9 (1996) 157. 17. J. Bebie, M. Schoonen and D. Strongin, Geochim. Cosmochim., 62 (1998) 633. 18. W. Sand, T. Gehrke, P. Jozsa and A. Schippers, Hydrometallurgy, 59 (2001) 159. 19. O. Tuovinen and D. Kelly, Z. Allg. Mikrobiol., 12 (1972) 311.

135


Bioleaching of electronic scrap material by Aspergillus niger W.K.Ten and Y.P.Ting* Department of Chemical and Environmental Engineering, National University of Singapore, Kent Ridge Crescent, Singapore 119260 Abstract This work reports on the bioleaching of electronic scrap material (ESM) from a local waste recycling company. The most abundant elements present in the ESM dust were oxygen and silicon, followed by various base metals with concentration exceeding 10,000 mg/kg. Precious metals gold, silver and palladium were found in trace amount (6,000 mg/kg); most of these elements were found at a concentration higher than 10,000 mg/kg based on the result of digestion/ICP-AES and XRF analyses. Compared to the analyses of electronic scrap by Brandl et al. [9], the major elements found (Al, Cu, Pb, Sn and Zn) were similar with the ESM used in this work. Precious metals such as Ag, Au and Pd were found at lower concentration ( Hg > Cu > Cd >Cr >Ni >Pb >Co > Zn > Ca, Fe). In this sense, the presence of metals in either a soluble or exchangeable phase could significantly affect the biosynthesis of organic acids while using an in-situ bioaugmentation process. Thus, this toxicity needs to be controlled and 177


studied. On the other hand, strongly sorbed metals could increase operation costs since more organic acids would be needed to dissolve solid phases contributing to metal retention. Metal speciation in mining wastes before and after bioleaching were determined by using a selective sequential extraction (SSE) method [3]. Figure 2 shows the initial metal distribution of the 2 mining residues before bioleaching.

Figure 1. Partitioning of metals in soils under aerobic conditions (from (8)).

New Brunswick

Fraction

100%

Résiduels M.Organique

50%

Ox./Hyd.

0%

Carbonates Cu

Fe

Mn

Ni

Pb

Zn

Metals

Echangeable Solubles

New Caledonia

Fraction

100% 50% 0% Cu Fe Mn Ni Pb Zn Metal

Résiduels M.Organique Ox./Hyd. Carbonates Échangeables Solubles

Figure 2. Metal distribution under natural pH for the 2 residues (Résiduel = Residual, M. Organique = Organic Matter, Oxide/Hydroxides, Carbonates, Échangeable = Exchangeables and Solubles)

178


Bioleaching using fungi organisms is principally based on the following mechanisms: acidolysis (consisting of the solubilisation of the matrix by pH reduction), complexolysis (consisting of the complexation of metals by the excreted organic acids or amino acids and shown in Fig. 3), redoxolysis (the reduction of ferric iron which is mediated by the oxalic acid) and bioaccumulation of metals by the organism's mycelium. Metal bioleaching rates depend on acid concentration, complexation kinetics, contact time, pH, and metal geochemistry.

Figure 3. Complexation of heavy metals by organic acids 5.

BIOLEACHING PROCESS Bioleaching studies were funded by Natural Resources Canada and were recently performed at the environmental laboratory in Laval University. Organic acid production can be carried out within the mine residue or produced elsewhere in a reactor. Thus, in this study, organic acids and residues were mixed directly or indirectly using either beaker or column reactors as shown in Figure 4. Each option presents specific advantages and were tested in a preliminary activity. While in-situ bioaugmentation can accelerate the overall treatment process, ex-situ production of organic acids can facilitate the isolation of preferred compounds for more efficient application on the piles and can avoid difficulties related to keeping optimum culture conditions in the field. Indirectly, acids (supernatants) can be separated from fungi growth (biomass) and mixed with various concentrations of residues (1, 5, 7, 10 and 15% w/v). During preliminary experiments, fungus were grown in the presence of the same concentrations of residues up to 15 days. Various bioleaching results are presented in table 2 and figure 4. As it can be seen, indirect bioleaching give in some cases low extraction ratios than direct bioleaching, this is due to the purity and sterile condition in indirect bioleaching were organic acid production is optimized. Targeted metals for extraction were Cu and Pb from the New Brunswick mine and Cu, Ni and Zn from the New Caledonia mine.

179


a) Acid production phase for indirect leaching

P. simplicissimum

Aspergillus niger

b) Direct and indirect bioleaching in column Microorganisms culture

Mining Residues

Direct Bioleaching

Indirect Bioleaching

c) Direct bioleaching in beaker

New Caledonia residues

Figure 4. Bioleaching layouts 180

New Brunswick residues


Table 2. Results in % mass extraction from column bioleaching New Brunswick

Cu Fe Mn Ni Pb Zn

Aspergillus Niger Direct Indirect 8.9 12.3 1.8 1.5 78.7 83.7 0.0 0.1 19.4 12.5 6.0 3.5

Penicillium simplicissinum Direct Indirect 9.4 31.6 1.9 1.4 73.6 80.6 0.0 0.0 18.4 30.5 7.7 9.1

Metal Cu Fe Mn Ni Pb Zn

Aspergillus Niger 12.5 31.3 0.2 0.4 35.4 25.6 63.5 51.1 2.8 100 27.0 75.7

Penicillium simplicissinum 3.1 25 0.09 0.0 20.7 5.6 39.8 16.1 35.6 2.8 43.2 10.8

Metal

New Caledonia

Concentration (mg/kg de sol)

Biolixiviation PEN + CAL

Cu Fe Mn Ni

012 10 216 14 618 4 8

Pb

Jours

Zn

Concentration (mg/kg de sol)

Biolixiviation PEN + NOR Cu Fe Mn Ni Pb Zn

7000 5000 3000 1000 -1000

0

5

10 Jours

15

20

Figure 4. Bioleaching results for Penicillium S.

181


Results showed that organic acid production was not inhibited by the presence of mining residues, at least in the range of soil/liquid ratios under study. Preliminary tests were run to obtain optimal soil/liquid ratios for fungi growth. An analytical procedure combining HPLC and GC-MS was developped in order to dose the various organic acids produced. As shown in figure 5 Gluconic acid was produced at the highest concentration followed by citric acid. In a second part of the study, batch experiments evaluated continuous washing with the same initial organic acid load. After five washings, the metal efficiency extraction was: lead, 42%; cadmium, 96%; chromium, 85%; copper, 60% and zinc, 77%. Since pH influence the solubilization/ precipitation of some metal forms (exchangeable, oxides, carbonates), the study included a pH monitoring of both, acid and percolated metal containing effluent solutions. A slightly increase in metal extraction was observed with a decrease in the pH of the acid produced solution. It was also concluded that metal geochemistry played a significant role in metal bioleaching. Organic Acids AN

Concentration (%)

12 Acide gluconique

10

Acide Phytique Acide oxalique

8

Acide Tartarique

6

Acide Isocitrique Acide malique 1

4

Acide citrique Acide malique 2 Sucrose

2 0 37088

37090

37092

37094

37096

37098

37100

37102

Days Figure 5. Organic acids production as function of time for indirect bioleaching Indeed, the SSE results not only allowed to determine metal distribution within the soil matrix but also to conclude that carbonate and amorphous forms of metals were among the more difficult metal particulate species to destroy before extraction was possible. In revenge, exchangeable and oxides were more easily attacked by the acids. Also, a slight redistribution of metal particulate species after bioleaching was observed. We want to further explore the interrelations between metal species distribution, metal species re-adsorption and acid components (e.g. gluconic, citric) selectivity. By developing the application of the SSE method, the information gained will allow a better conception of reliable, and predictable methodologies for on-site metal bioleaching. 6.

USAGE OF ORGANIC WASTES AS SOURCE OF CARBON Parallel experiments were run under direct and indirect leaching conditions in order to evaluate different carbon sources. These experiments were run using a concentration of 10% residues and use only aspergillus niger’s produced acids. Various Carbon sources 182


included molasses, corncobs, and brewery wastes. These experiments lasted for about three weeks. The results are shown in Figure 6. It is clear that sucrose solution gave the best leaching results under both direct and indirect conditions. The pH was low (pH 3.6) as in the previous set of experiments. Molasses yielded lower leaching results than the sucrose and better than the corncobs. Corncobs yielded reasonably good results (up to 24% copper removal) under indirect leaching conditions. They contain simple sugars that can be used by A. niger. Brewery waste gave poor results. The grains would require pretreatment to release simple sugars for it to be a suitable substrate. (a)

(b) 40

60

Sucrose Molasses

40

Corn cob Brewery waste

20

% Cu leached

% Cu leached

80

Sucrose Molasses Corn cob

20

Brewery waste

0

0 0

10 Time (days)

20

0

10

20

Time (days)

Figure 6. Effect of C source on Cu extraction using indirect (a) and indirect (b) conditions 7.

DISCUSSION AND CONLUDING REMARKS Indirect leaching was clearly more favorable for metal removal. When the flasks were not sterilized, faster growing bacteria took over, causing problems for the fungus and decreasing acid yields. Another important parameter is the degree of selectivity of the organic acid mixture for precious metals rather than for non-precious metals (e.g. Ca, Mg, Na, K). Therefore, the bioleaching process must be conceived in such way of maximizing precious metal extraction, minimizing non-precious metal extraction, optimal organic acids ratios within the mixture, optimal pH to avoid early metal precipitation and optimal To. The process must also include a pre-screening geochemical characterization that will determine the real potential of metal extraction. In conclusion, metal recovery by fungi bioleaching is feasible, however, further applied research must be performed to provide greater understanding on the nature of complexes and bioleaching selectivity, on the influence of metal geochemistry on bioleaching as well as on the reuse of organic acids. REFERENCES 1. Bosecker, Klaus (1997). Bioleaching: metal solubilization by microorganisms. FEMS Microbiology Reviews, n. 20, pages 591-604. 2. Feasby, G. and Tremblay, G.A. (1995). New Technologies to reduce environmental liability from acid mine generating mine waste. Proc. Sudbury ’95 – Mining and the Environment, 6443 Ottawa. CANMET. 3. Galvez-Cloutier, R. (1995). Study of heavy metal accumulation mechanisms in the Lachine Canal sediments. Ph.D. Thesis. McGill University, Montreal.

183


4. Krebs, W. et al. (1997). Microbial Recovery of metals from solids. FEMS Microbiology Reviews, n. 20, pages 605-617. 5. Mulligan, C. et al. (1999). Biological leaching of copper mine residues by Aspergillus niger. Proc. of the Int. Biohydrometallurgy Symposium IBS’99. Madrid, Spain. Part A Bioleaching, Microbiology, pages 453-461. 6. Mind Environment Neutral Drainage Program (MEND)(1997). Annual Report. 202 pages. 7. Wheeland, K.G. and Feasby, G. (1991). Innovative decommission technologies via Canada’s MEND program. Proc. of the 12th Nat. Conference, Hazardous Materials. Control/Superfund ’91, Control Res. Inst., pages 23-38. 8. Galvez-Cloutier, R. Yong, R.N., Chan, J. and Bahout, E. (1995) Critical Analysis on Sediment Quality Criteria and Hazard-Risk Assessment Tools. Dredging, Remediation and Containment Contaminated Sediments ASTM STP No 1293, pp. 306-318.

184


Biooxidation of pyrite by Acidithiobacillus ferrooxidans in single- and multi- stage continuous reactors C. Canales, J.C. Gentina and F. Acevedo* School of Biochemical Engineering, Catholic University of Valparaiso, Av. Brasil 2147, Valparaíso, Chile Abstract The objective of this work was to study the extent of the bacterial attack on pyrite in the biooxidation of a refractory gold concentrate in continuous stirred tank reactors (CSTR). Two laboratory-scale biooxidation systems were installed. One of them consisted of a single stage 4-litre CSTR; the other one was a four-stage system with a total volume of 4 L. Acidithiobacillus ferrooxidans R2 was used; the concentrate, containing 66.7% pyrite, was suspended in 9K medium without ferrous sulfate. The single-stage CSTR was operated with residence times between 3.5 and 10 days, with Eh values of 600-650 mV. Up to 51% of iron solubilization was obtained, with negligible ferrous iron levels. This result suggests that the bacterial ferrous iron oxidation proceeded at a higher rate than the pyrite attack. The four-stage system operated with total residence times between 7 and 14 days, with maximum iron solubilization of 66% and Eh of 525 to 675 mV. Again, almost no ferrous iron was detected. Decreasing residence times had a large effect in diminishing the bacterial activity especially in the first stage because its low residence time. Oxygen demands measured in each stage revealed that decreasing total residence times caused a displacement of the main microbial activity towards the last stages. These results show that increasing residence times favor the multistage configuration, resulting in higher degrees of pyrite oxidation than in the single CSTR. This is because in the latter case the positive effect of residence time tends to saturation. Keywords: refractory gold concentrate, reactor configuration, CSTR, pyrite biooxidation 1.

INTRODUCTION Continuous biooxidation of refractory gold concentrates in tank reactors is currently a technologically and economically feasible technology that is being applied in several large-scale operations the world over [1, 2]. Reactors present significant advantages over heaps for bioleaching operations, in particular related to a more homogeneous reacting mass and the possibility of exerting close control on the main process variables. The biooxidation of gold concentrates, as well all other bioleaching processes, is as a whole an autocatalytic complex reaction. This autocatalytic character is given by the fact * [email protected]

185


that both cells and ferric iron act as reactants and products. This being the case, the optimal reactor configuration that minimizes the reaction volume for a given conversion is a continuous stirred tank reactor (CSTR) followed by a tubular reactor (TR) [3]. Conversely, given a defined volume, the multi-stage arrangement will render higher degrees of conversion. Because the need to maintain the solids in suspension and to supply oxygen and carbon dioxide rules out the operation of an actual TR, its kinetic behavior is simulated with a battery of CSTR’s connected in series [1, 4, 5]. This configuration is used in large-scale operations utilizing mesophilic bacteria [6] and has been tested at laboratory and pilot scale with extreme thermophiles [7, 8]. The objective of this work was to quantitatively characterize the behavior of a reaction system composed of four CSTR’s in series and compare it with the performance of a single CSTR of a volume equal to the total volume of the series arrangement. 2.


2.1 Microorganism, concentrate and culture medium Acidithiobacillus ferrooxidans R-2 was used. This strain was cultivated in the presence of the same gold concentrate for over one year before performing these experiments. The refractory gold concentrate contained 15 g gold/tonne, 67.6% pyrite and 9.0% chalcopyrite. It contained 36.1% Fe, 39.7% S and 3.3% Cu. A fraction of particle size less than 75 µm was used. Before each run the concentrate was washed with a 10% v/v aqueous solution of acetone, rinsed with diluted sulfuric acid pH 1.8 and dried overnight at 100ºC. 9K medium was used, replacing the ferrous sulfate with the concentrate at a pulp density of 6% w/v. 2.2 Analytical methods Ferrous ion iron was measured by the modified o-phenanthroline method [9] and total soluble iron was determined by reducing the ferric ion to ferrous ion with hydroxylamine and assaying with phenanthroline. Ferric iron was calculated as the difference between the two. Sulfate was assayed by turbidimetry [10]. Eh was monitored with an Ag/AgCl probe and dissolved oxygen was measured with a polarographic probe. Oxygen consumption rate was determined in each stationary state by the gassing-out method [11]. 2.3 Bioreactor systems Two continuous biooxidation systems were set up, each one of a total working volume of 4 litres. One was a single 4-L continuous stirred tank reactor (CSTR) and the other consisted of four CSTR’s connected in series. The first stage has a working volume of 1.75 L and the other three, which represent a TR, are 0.75 L each. The reactors were made of acrylic plastic and each one has four baffles, heating jacket, variable speed agitator with one pitched-blade turbine pumping down and Eh, pH and dissolved oxygen probes. The CSTR’s had constant geometrical ratios of HL/T = 1.0 and D/T = 0.30. Aeration was supplied by means of a perforated annular sparger. The tanks were fed and discharged by peristaltic pumps. Because of the very low flow rates involved, the pumps were turned on and off periodically by a programmable switch. Fresh pulp was fed from an agitated feed tank.

186


2.4 Operation conditions The reactors were operated at 33ºC, with agitation of 600 rpm and aeration of 0.5 volumes of air per volume of liquid per minute (vvm). pH was adjusted to 1.8 with sulfuric acid at the start of the batch phase operation. Pulp density was 6% w/v. These constant conditions were used in all runs of both biooxidation systems. The 4-L CSTR was operated at pulp residence times between 3.5 and 10 days. The four-stage system was operated at residence times in the range of 7 to 14 days. The reactors were operated continuously for at least three residence times before any measurements were done. The different steady states were always established by increasing the pulp flow rate. 3.


3.1 Single stage reactor Figure 1 depicts the results of the operation of the single stage reactor. Under the experimental conditions used in this work, 50% of the pyrite was solubilized as revealed by the total soluble iron measurements. At all times the concentration of ferrous ion was negligible, a fact compatible with the high Eh values attained. The absence of Fe2+ is an indication that the rate of bacterial oxidation of ferrous ion was higher than the rate of ferrous production in the ferric leaching of the concentrate. The low Eh value at a residence time of 3.5 days could be due to the high solids content of the reactor due to the low extraction at that operation condition. This effect has been repeatedly observed by the authors under similar circumstances.

Figure 1. Continuous solubilization of pyrite concentrate in 4-L CSTR at 6% pulp density, 0.5 vvm and 600 rpm Table 1 presents the production rates, oxygen consumption rates and ferric/sulfate ratios obtained at the different residence times. Maximum rates are obtained in the range of residence times of 5.5-7 days, while the consumption rate of oxygen decreases steadily with increasing residence times. The Fe3+/SO42- ratio varies between 0.60 and 0.69, higher than the stoichiometric value of 0.47 predicted by equation 1. It can also be noted that the oxygen consumption rate decreases with increasing ferric/sulfate ratios. These results 187


suggest that the pyritic sulfur is not completely oxidized to sulfate. This uncoupled behavior of the solubilization and oxygen consumption rates arises from significant changes in cell physiology mediated by the dilution rate. Table 1. Production rates, oxygen consumption rates and ferric/sulfate ratios in the single-stage reactor* Residence time, Fe production, SO42- production, O2 consumption rate, Fe3+/SO42(d) (g/L·d) (g/L·d) (g/L·d) w/w ratio 3.5 0.86 1.43 3.54 0.60 4.5 1.33 2.00 3.34 0.67 5.5 1.36 2.27 2.70 0.60 7 1.37 2.00 2.06 0.69 10 1.10 1.60 1.79 0.69 * Dissolved oxygen concentration was higher than 40% sturation at all operation conditions

4 FeS2 + 15 O 2 + 2 H 2 O → 2 Fe 2 (SO 4 )3 + 2 H 2SO 4

(1)

3.2 Multi stage system Steady state results are presented as a function of cumulative residence time in each reactor for the four total residence times considered. The dissolution of pyrite is presented in Figure 2. No ferrous ion was detected in the liquid. The first stage operated at residence times between 3 and 6 days. The increase of solubilization obtained in that range was over seven fold, significantly higher than the one obtained in the single CSTR under similar conditions (Figure 1). The solubilization that took place in the first CSTR increased steadily with residence time. Blank runs made with non-inoculated pulp showed that chemical leaching was almost nil, so the very low leaching that occurred at 3 days residence time must be due to the activity of a small bacterial population, as revealed by microscopic examination. By the other hand, the contribution of the three 0.75-L reactors was more important at low residence times than at high ones. As a whole, these data confirm the importance of the multi-stage design, pointing to an adequate operation residence time around 9 days. It is worth noting that in the intermediate range of residence times such as 8 or 9 days, highest solubilizations were obtained with two and three stages rather than with the four reactors. This result could be influenced by the fact that some solids accumulated in each stage because of difficulties in attaining a homogeneous pulp because of the small scale of the reaction system. Biooxidation expressed as percent iron extraction is shown in Figure 3. A maximum extraction of 66% is obtained at a cumulative residence time of 14 days. As a rule, the multi-stage system rendered higher extractions than the single vessel. This observation is consistent with the fact that the Eh values were also higher, as can be seen in Figure 4 as compared to Figure 1.The production of sulfate, presented in Figure 5, shows a similar pattern to the iron extraction, although the advantage of the four-stage system is apparent at lower residence times.

188


15

80

Fe extraction, %

Soluble iron, g/L

12 9 6 3

60

40

20

0

0 0

3

6

9

12

0

15

Cumulative pulp residence time, d

3

6

9

12

15


Figure 2. Continuous solubilization of pyrite concentrate in four-stage system at 6% pulp density. ♦: reactor 1; ■: reactor 2; ▲: reactor 3; x: reactor 4

Figure 3. Iron extraction in the four stage system as a function of cumulative residence time

30 Sulfate concentration, g/L

700

Eh, mV

600

500

400

25 20 15 10 5 0

0

3

6

9

12

15


Figure 4. Eh as a function of cumulative residence time in the four-stage system. ♦: reactor 1; ■: reactor 2; ▲: reactor 3; x: reactor 4

0

3

6

9

12

15


Figure 5. Sulfate production in the fourstage system. ♦: reactor 1; ■: reactor 2; ▲: reactor 3; x: reactor

The biological oxidation generated ferric/sulfate ratio values of 0.45 to 0.68, lower than the ones obtained in the single reactor unit, pointing to a more complete oxidation of sulfur in the complex arrangement. 189


The production rates of each stage are presented in Table 2. Except for stage 1, the rates increase with decreasing total residence times. The residence time in stage 1 at total residence time of 7 days across the system is 3 days; this is probably a too short time and did not allow the establishment of a stable cell population of a sufficient size. Data from Table 2 regarding sulfate production rate suggest that the bacterial activity is displaced successively from the first to the following stages as the residence time decreases. This effect can also be appreciated in Table 3 that shows the oxygen consumption rates. From the results of Tables 2 and 3 it can be concluded that the main variation in the behavior of the system is produced in the transition from 7 to 8.5 days of total residence time. The cumulative production rates of iron and sulfate in each stage are presented in Figures 6 and 7. Iron solubilization is less affected by residence time in stages 3 and 4 than in the first two stages, while sulfate production rate in each reactor increases up to a maximum and then decreases steadily with residence time. Cell activity is proportional to the slope of the curves of each stage. Similar behavior can be seen in the first three reactors, in which most of the pyrite solubilisation takes place. This saturation pattern could be due to the exhaustion of available active sites on the surface of the concentrate particles [12]. Table 2. Iron and sulfate production rates in the four-stage system (g/L·d)* Residence Stage 1 Stage 2 Stage 3 Stage 4 time** Fe SO42Fe SO42Fe SO42Fe SO42(d) 7 0.43 0.67 1.67 3.05 1.90 3.81 1.52 4.57 8.5 0.86 1.61 1.25 1.88 1.13 2.51 1.25 4.39 10 1.19 2.51 1.44 2.13 1.01 2.13 1.81 1.60 14 1.63 2.45 0.76 1.14 0.80 1.53 0.11 1.14 * Calculated as the ratio of the increase of production in each stage to the residence time. Dissolved oxygen concentration was higher than 40% saturation at all operation conditions ** Considering the total reaction volume of 4 litres

Table 3. Oxygen consumption rate in the four-stage system (g/L·d) Residence time* (d)

Stage 1

Stage 2

7 0.32 1.41 8.5 1.91 1.58 10 1.98 1.68 14 1.80 1.75 * Considering the total reaction volume of 4 litres

190

Stage 3

Stage 4

3.12 1.47 1.48 1.50

2.66 1.07 1.41 1.30


14 Sulfate production rate, g/L/d

Fe solubilization rate, g/L/d

8 7 6 5 4 3 2 1

12 10 8 6 4 2 0

0 0

3

6

9

12

15


Figure 6. Cumulative iron solubilization rates in the four-stage system. ♦: reactor 1; ■: reactor 2; ▲: reactor 3; x: reactor 4

0

3

6

9

12

15


Figure 7. Cumulative sulfate production rates in the four-stage system. ♦: reactor 1; ■: reactor 2; ▲: reactor 3; x: reactor 4

4.

CONCLUSIONS The biooxidation of a pyritic refractory gold concentrate has been compared using a single stage CSTR and a reactor arrangement consisting of four CSTR’s in series of equivalent volume. It is concluded that increasing residence times favor the multistage configuration, resulting in higher degrees of pyrite oxidation than in the single CSTR.

ACKNOWLEDGMENTS This work was supported by the National Commission of Science and Technology through the FONDECYT project 1000284. REFERENCES 1. F. Acevedo F, Electronic J. Biotechnol., 3 (2000) 184. Available from: www.ejb.org, ISSN 0717-3458. 1. J.A. Brierley and C.L.Brierley, Hydrometallurgy, 59 (2001) 233. 2. O. Levenspiel, Chemical Reaction Engineering, 3rd ed., J. Wiley & Sons, New York, 1998, chap. 6. 3. D.H. Dew. In: T. Vargas, C.A. Jerez, J.V. Wiertz and H. Toledo (eds.) Biohydrometallurgical Processing, Vol.1. Universidad de Chile, Santiago, Chile. p. 239, 1995. 4. R. González, J.C. Gentina and F. Acevedo. In: Proceedings XIII National Congress of Chemical Engineering, Antofagasta, Chile, 18–21 October, 1999. 5. D.E. Rawlings (ed.), Biomining. Theory, Microbes and Industrial Processes, Springer Verlag, New York, 1997. 6. M. Gericke, A. Pinches and J.V. van Rooyen, Int. Miner. Process., 62 (2001) 243. 191


7. P. d’Hugues, S. Foucher, P. Gallé-Cavalloni and D. Morin, Int. Miner. Process., 66 (2002) 107. 8. L. Herrera, P. Ruiz, J.C. Aguillon and A. Fehrmann, J. Chem. Technol. Biot., 44 (1989) 171. 9. Instituto de Hidrología de España, Análisis de aguas naturales continentales, 32. Centro de Estudios Hidrográficos, Madrid, 1980. 10. R. Jurecic, M. Bwerovic, W. Steiner and T. Koloini, Can. J. Chem. Eng., 62 (1984) 334. 11. G.F. Andrews, Biotechnol. Bioeng., 31 (1988) 378.

192


Chemical chalcopyrite leaching and biological ferric solvent production at pH below 1 P.H.-M. Kinnunena, V.L.A. Saloa,b, S.O. Pehkonenb and J.A. Puhakkaa a

Institute of Environmental Engineering and Biotechnology, Tampere University of Technology, P.O.Box 541, FIN-33101 Tampere, Finland b Department of Chemical and Environmental Engineering, National University of Singapore, 10 Kent Ridge Crescent, 119260 Singapore

Abstract Chalcopyrite ferric leaching experiments were conducted at pH 1.0 to 2.5 at 50°C and 65°C under air and nitrogen atmospheres. Copper leaching yields were higher (97%) at pH 1.0 than at pH 1.8-2.5 at both temperatures. Increase in the temperature increased the initial rate of leaching, but resulted in lower yields indicating high jarosite precipitation rates. The composition of the gas phase did neither affect redox potentials nor leaching rates. Jarosite and iron hydroxide precipitates were not formed and copper yields increased by using a ferric solution at pH 1.0. In tank reactor, similar copper leaching yields were obtained with different ferric supply regimes. Biological generation of ferric solution at low pH was studied in batch and continuous-flow fluidized-bed reactors (FBR). In batch assays at 35°C, biological iron oxidation rate was not affected by pH of 0.9-1.5 but started to decline at 0.7. The pH of the FBR was gradually decreased to 0.9 without changes in iron oxidation; the maximum iron oxidation rate at pH 0.9 was 10 g Fe2+ dm-3 h-1. The results indicate that biologically produced ferric solvent at pH 0.9 results in high chalcopyrite leaching yields. Keywords: chalcopyrite, ferric leaching, iron oxidation, passivation 1.

INTRODUCTION One of the major problems in hydrometallurgical applications is the formation of a hindering diffusion layer on the mineral surface or a contact hindrance between the leaching solution, microbes and mineral. A better understanding of the surface speciation under leaching conditions is a key factor in improving dissolution kinetics and yields in bioleaching. Passivation layer in sulphide mineral leaching with ferric iron has been proposed to consist of iron hydroxy precipitates or elemental sulphur formed as end products [1,2,3]. Polysulphides may act as transient intermediates of leaching [1]. Iron in the leaching solution precipitates as iron oxides or basic iron salts like jarosite [2,3,4]. Leaching conditions such as pH, temperature and ionic composition and concentration of the medium affect the formation of iron hydroxy precipitates [5]. Ferric iron precipitation diminishes the available ferric iron in the leach solution, forms kinetic barriers and tends to block pumps and valves. The iron precipitation highly 193


depends on the pH and temperature. [5] Thiobacillus ferrooxidans and Leptospirillum ferrooxidans are the most important mesoacidophilic microorganisms involved in ferrous iron oxidation. The optimum pH for bacterial iron oxidation is generally 2.0 to 2.5. [6] L. ferrooxidans tolerates lower pH values than T. ferrooxidans, which usually does not survive below a pH of 1.0 [5,6]. Even though microorganisms may adapt to physicochemical changes such as pH in their environment, there are limits to the extent to which this may occur [7]. In this study the influence of temperature, pH and atmospheric oxygen on indirect ferric leaching rates and surface precipitates were studied. Since the jarosite precipitation can be prevented at a very low pH, the potential of biologically producing ferric solvent at low pH was studied in batch and continuous flow reactors. The maximum iron oxidation rate at pH of 0.9 was determined. Further, different ferric leach solutions were used to maximize the copper leaching and minimize the iron precipitation. 2.


2.1 Influence of pH on the leaching rate The influence of pH on the leaching rate of chalcopyrite (Outokumpu Ltd, Pyhäsalmi, Finland, Cu 25.4%; Fe 27.9%) was studied at 50°C and 65°C in 150 cm3 erlenmeyer flasks using 3% (w/v) solids concentration. Leaching solution consisted of 0.4 g dm-3 (NH4)2SO4, 0.25 g dm-3 KH2PO4.2H2O, 0.25 g dm-3 MgSO4.7H2O, 0.02 g dm-3 yeast extract and a stoichiometric amount of ferric iron as Fe2(SO4)3 (Eq. 1). (Eq. 1) CuFeS2 + 4 Fe2+ Æ Cu2+ + 5 Fe2+ + 2 S° The pH was adjusted with H2SO4 or NaOH. Experiments under nitrogen atmosphere were conducted in order to study the influence of atmospheric oxygen on the precipitation. The leaching solution was sparkled with N2 before the addition of chalcopyrite (15 min) and after sampling (5 min). Samples (5 mL) were centrifuged (5000 rpm, 12 min) and the supernatants were used for analyses. Dissolved copper and iron concentrations were analyzed by inductively coupled plasma connected to mass spectrometry (ICP-MS) and the ferrous iron concentration spectrophotometrically by ferrozine method [8]. Ferric iron concentration was determined as the difference of total and ferrous iron concentration. The pH was measured with WTW Sentix 42 electrode and redox potential an ORION combination electrode 9678BN. Samples for SEM/EDX and XRD were centrifuged (4000 rpm, 3 min) and pellets were dried at 40°C. SEM/EDX-analyses were conducted with JEOL JSM-5600LV scanning electron microscope. In addition some of the residues were examined with x-ray diffraction (Cu anode, Kα radiation, λ = 1.54). 2.2 Effect of the ferric addition on chalcopyrite leaching The experiments were conducted in 2 L stirred tank reactors (Figure 1). The reactor conditions were as follows: liquid volume 1.5 L, solids concentration 10% (w/v), and temperature 50°C. The leach solution and a stoichiometric amount of Fe2(SO4)3 was added in either in 1, 2 or 5 shares. The pH of the leach liquor was adjusted to 1.5 with H2SO4 and NaOH at the beginning and at the sampling time. Redox potential was measured using Hamilton Pt-ORP electrode and pH using Orion model SA720 pH -meter after filtration (0.45 µm). The concentrations of dissolved copper and iron were measured using ICP-MS. The ferrous iron and SEM/EDX-analyses were performed as described above. The evaporation losses were accounted for in the results.

194


Figure 1. Schematic diagram of 2L agitated reactor for ferric leaching. 1) Hot water inlet from water bath, 2) hot water outlet, 3) mixing device, 4) mixing blade, 5) baffle, 6) sampling socket. 2.3 Influence of pH on iron oxidation The influence of pH on iron oxidation was studied in duplicate 120 mL serum bottles in 60 mL of total volume at 150 rpm in a rotary shaker at 35°C. The mineral medium was autoclaved (121°C, 0.1 MPa, 20 min) before the addition of FeSO4.7H2O after sterile filtration (0.2µm). The final concentration of the mineral medium was 0.35 g dm-3 (NH4)2HPO4, 0.05 g dm-3 K2CO3 and 0.05 g dm-3 MgSO4. The pH was adjusted with H2SO4 to 0.5, 0.7, 0.9, 1.1, 1.3 and 1.5 with 7 g dm-3 Fe2+ in the solution. The bottles were inoculated with 1 mL of an enrichment culture of iron-oxidizers from a fluidized-bed reactor [9]. In the chemical control bottles, sterile H2O was used instead of inoculum. After batch assay determinations of the pH tolerance limits, the pH in the fluidizedbed reactor with activated carbon as the biomass carrier [9] was gradually decreased to 0.9. The composition of the feed solution was 34.75 g dm-3 FeSO4.7H2O, 0.35 g dm-3 (NH4)2HPO4, 0.05 g dm-3 K2CO3 and 0.05 g dm-3 MgSO4 in tap water at pH of 0.9. The maximum iron oxidation rate was determined in the fluidized-bed reactor at the pH of 0.9 with 21 g dm-3 ferrous iron in the feed solution. The ferrous iron concentration was determined using the Shimadzu UV 1601 spectrophotometer by the colorimetric ortho-phenantroline method [10] modified as follows: 2 mL of 1,10-phenantroline (10 g/L) and 1 mL of ammonium acetate buffer were added to 3 mL of sample. Dissolved oxygen and temperature measurements were made using WTW OXI96 meter at the sampling time. 3.

RESULTS

3.1 The influence of pH, temperature and gas phase composition on leaching and precipitation Leaching experiments were conducted at different pH values under air and nitrogen atmospheres at 50°C and 65°C (Figure 2). Copper yields increased with decreasing pH; at 50°C they were 95%, 35% and less than 10% at pH 1.0, 2.0 and 2.5, respectively. At 195


65°C, the initial rate of leaching was higher, but resulted in lower yields than at 50°C indicating rapid precipitation. Iron precipitation was significant at pH of 1.8-2.5, which affected the leaching efficiency (Figure 3). At pH 1.0, the ferric iron did not precipitate, but was reduced to ferrous iron during leaching. Redox potentials decreased in the course of leaching, being highest at low pH values. The composition of the gas phase did not affect the redox potentials or leaching rates.

Figure 2. Copper leaching yields from chalcopyrite at 50°C (right) and at 65°C (left) In all experiments, SEM/XRD revealed jarosite precipitation layers on the surfaces of the CuFeS2. The layer consisted of large and small regular-shaped cubes with the average composition as presented in Table 1. The mineral surface was much less covered by the precipitation layer at pH 1.0 than at pH 1.5 and above (Figure 4). The surface of the ore (25% Cu, 16% Fe, 33% S) was partly covered by amorphous sulphur rich layer (60-97% S) typically formed under the cubic jarosite precipitates. Temperature and pH influenced the leaching yields, but not the composition of the residue-precipitates. Table 1. The range of elemental-% composition (EDX) of separate layers on CuFeS2 surfaces after 17 to 19 days of ferric leaching at pH 1.8-2.5

O, % S, % Fe, % Cu, % Na/K %

196

Surface composition after leaching

Mineral surface, unleached

Theoretical composition of jarosites

Larger precipitates

Smaller cubes

Typical mass precipitation

--25 11-22 14-24 ---

44-50 13 33-35 --~0

51-63 26-31 4-18 0-1 --

63-77 7.5-13 10-17 0 3-6

55-75 8-12 14-22 --3-4


40

A

g/L

30 20 10 0 0

5

10 time (d)

15

20

0

5

10 time (d)

15

20

0

5

10 time (d)

15

20

40

B

g/L

30 20 10 0

40

C

g/L

30 20 10 0

Figure 3. The fate of iron in chalcopyrite leaching at 50°C under the air atmosphere at A) pH 1.0, B) pH 1.8 and C) pH 2.0. (▲) calculated dissolved iron concentration (Fe from Fe3+-solvent plus Fe dissolved from CuFeS2), (x) dissolved iron (measured concentration), (♦) Fe2+ concentration, (■) Fe3+ concentration A

B

Figure 4. The mineral surface shown by scanning electron microscopy after 17-19 days of ferric leaching at 50°C A) at pH 1.0 and B) at pH 1.8 197


3.2 Effect of the ferric addition on chalcopyrite leaching Ferric iron was added to the 50°C stirred tank reactor at pH 1.6 in five, two or one doses and resulted in similar copper leaching yields (approximately 40%) with all supply regimes ferric iron becoming the limiting factor (Figure 5). The initial leaching rate was highest, and redox potential and pH decreased fastest when ferric solution was added in one dose in the beginning. The iron precipitation was highest with one dose and lowest with five doses. SEM/EDX revealed elemental sulphur layer on the surface of the mineral at the end of the experiment, but showed also mineral surfaces without passivating layer (Figure 6).

Figure 5. The yield, pH and redox potential with 1 (♦), 2 (▲) and 5 (■) ferric solution additions (shown by arrow). Dissolved iron (◊), Fe3+ (∆), Fe2+ (□) and (x) calculated dissolved iron (Fe from Fe3+-solvent plus Fe dissolved from CuFeS2) concentrations. Addition of ferric solution in five (A), two (B) or one (C) doses

198


CuFeS2

S0

Figure 6. Sulphur rich layer on the leached CuFeS2 surface at the end of the ferric addition experiment (day 15) 3.3 Effect of pH on iron oxidation The iron oxidation rate in the batch experiments remained unaffected at pH of 0.9-1.5 but started to decline at 0.7 (Figure 7). The lag phase of three weeks occurred at pH 0.5 and iron oxidation was strongly affected. The pH of the iron oxidizing fluidized-bed reactor was gradually decreased to 0.9 without changes in performance (Figure 8). The maximum iron oxidation rate at pH 0.9 was 10 g Fe2+ dm-3 h-1 (Table 2). The results show that the high-rate ferric production and regeneration in the FBR was possible at pH 0.9. 8 7

Fe2+ (g/L)

6 5 4 3 2 1 0 0

10

20

30

40

50

60

Time (days)

Figure 7. Influence of pH on ferrous iron oxidation by the fluidized-bed reactor enrichment culture in a batch assay. pH 0.5 (■), pH 0.7 (●), pH 0.9 (x), pH 1.1 (*), pH 1.3 (○), pH 1.5 (▲), pH 0.5 uninoculated control (□) and pH 1.5 uninoculated control (∆).

199


2,5

1,9 1,7 1,5

1,5

1,3

pH

Fe2+ (g L-1h -1)

2

1,1

1

0,9 0,5

0,7

0 0

20

40

60

80

100

0,5 120

Time (d)

Figure 8. The influence of pH decrease on iron oxidation rate in a fluidized-bed reactor. pH in FBR (■), pH in feed (□), load g Fe2+ dm-3 h-1 (∆) and oxidation rate g Fe2+ dm-3 h-1 (*). Table 2. Iron oxidation rate in the fluidized-bed reactor at 35°C. pH of the feed solution 0.9

Period 1 Period 2 Period 3

4.

Length of the period (days) 10 20 14

Number of data points

Fe oxidation max (%)

Fe oxid. min-max (g Fe2+ dm-3 h-1)

Fe oxid. mean (g Fe2+ dm-3 h-1)

7 4 3

94.6 99.2 99.5

7.3-9.7 9.7-10.4 9.2-10.4

8.1 10.0 9.8

DISCUSSION In this work, jarosite precipitates were typically found on top of the sulphur rich layer indicating the passivation of the mineral surface first by sulphur followed by jarosite precipitates. Increase in temperature from 50°C to 65°C increased the initial copper leaching rate, but resulted in reduced copper yields. This was likely due to faster precipitation at high temperature. Chalcopyrite leaching with ferric iron was initially fast regardless of pH, but slowed down or completely stopped due to intense precipitation. Decrease of the leaching pH resulted in less precipitates and copper yields close to 100%. Leaching proceeded similarly under air and nitrogen atmospheres. The results of this work demonstrated that stepwise adding of ferric iron did not improve the copper yields. For biological ferric regeneration, the pH of the iron oxidation reactor needs to be maintained low. In general the mesophilic iron oxidizers do not survive below pH 1.0 [5, 7], at which the ferric leaching experiments indicated copper yields close to 100%. The other mesophilic iron oxidation studies have been carried out at pH of 1.1-3.2 [11-23]. In this study, the iron oxidation rate remained unaffected at pH of 0.9-1.5. The maximum iron oxidation rate in a fluidized-bed reactor dominated by Leptospirillum like organisms (Kinnunen et al., paper in preparation) at pH 0.9 was 10 g Fe2+ dm-3 h-1, which was similar to the iron oxidation rate at pH 1.4 in the same reactor (8.2 g Fe2+ dm-3 h-1), when air was used for aeration [9]. The iron oxidation rate of this study compared favourably with the mesophilic iron oxidation rate (0.9 g Fe2+ dm-3 h-1) in the fluidized-bed reactor with

200


activated carbon as the carrier material at pH of 1.35-1.5 [23]. Ferric regeneration at 35°C was chosen to this study, because the moderately thermophilic iron oxidation rate at 60°C was considerably less than that of mesophilic [24]. In conclusions, the combination of high-rate biological ferric production at 35°C and pH 0.9 and below followed by chemical chalcopyrite leaching at 50°C to 65°C is promising for chalcopyrite leaching. REFERENCES 1. A. Parker, C. Klauber, H.R. Watling and W. van Bronswijk, Proc. Int. Biohydrometall. Symp. (2001), part B. ed. by V.S.T. Ciminelli and O. Garcia, Elsevier, 547-555 2. C. Klauber, A. Parker, W. Bronswijk and H.R. Watling, Int J Miner Process 62 (2001) 65-94 3. M.B. Stott, H.R. Watling, P.D. Franzmann and D. Sutton, Miner Eng 13 (2000) 11171127 4. S. Prasad and B.D. Pandey, Miner Eng 11 (1998) 763-781 5. M. Nemati, S.T.L. Harrison, G.S. Hansford and C. Webb, Biochem Eng J 1 (1998) 171-190 6. K. Bosecker, FEMS Microbiol Rev, 20 (1997) 591-604 7. D.B. Johnson, Hydrometallurgy 59 (2001) 147-157 8. L.L. Stookey, Anal Chem 42 (1970) 779-781 9. P.H.-M. Kauppi, H.J. Hautakangas and J.A. Puhakka, Proc. Int. Biohydrometall. Symp. (2001), part A. ed. by V.S.T. Ciminelli and O. Garcia, Elsevier, 385-392 10. Anonymous, Standard Methods for the Examination of Water and Wastewater. 18th ed. by A.E. Greenberg, L.S. Clesceri and A.D. Eaton, 1992, American Public Health Association 11. A.W. Breed and G.S. Hansford, Biochem Eng J 3 (1999) 193-201 12. M.J. Garcìa, I. Palencia and F. Carranza, Process Biochem (1989) 84-87 13. H.R. Diz and J.T. Novak, J Env Eng 125 (1999) 109-116 14. D.G. Karamanev and L.N. Nikolov, Biotechnol Bioeng 31 (1988) 295-299 15. M. Nemati and C. Webb, Appl Microbiol Biot 46 (1996) 250-255 16. H. Olem and R.F. Unz, Biotechnol Bioeng 19 (1977) 1475-1491 17. D.G. Karamanev, J Biotechnol 20 (1991) 51-64 18. M. Nemati and C. Webb, Biotechnol Bioeng 53 (1997) 478-486 19. S. Sandhya and R.A. Pandey, J Environ Sci Heal A27 (1992) 445-461 20. Mazuelos, I. Palencia, R. Romero, G. Rodríguez and F. Carranza, Miner Eng 14 (2001) 507-514 21. Mazuelos, R. Romero, I. Palencia, N. Iglesias and F. Carranza, Miner Eng 12 (1999) 559-564 22. Mazuelos, F. Carranza, I. Palencia and R. Romero, Hydrometallurgy 58 (2000) 269275 23. S.I. Grishin and O.H. Tuovinen, Appl Environ Microb 54 (1988) 3092-3100 24. P.H.-M. Kinnunen, W.J. Robertson, J.J. Plumb, J.A.E. Gibson, P.D. Nichols, P.D. Franzmann and J.A. Puhakka, Appl Microbiol Biot 60 (2003) 748-753

201


Comparative study of the bioleaching of two concentrates of chalcopyrite using mesophilic microorganisms in the presence of Ag(I) A. López Juárez and R. E. Rivera Santillán* Departamento de Ingeniería Metalúrgica, Facultad de Química, UNAM. Ciudad Universitaria, México D. F. 04510, México Abstract A bioleaching study of two copper concentrates from Sonora and Zacatecas, (Mexico), containing basically chalcopyrite, was performed. X-ray diffraction studies, XRD, showed certain variations in the composition of the concentrates. One of them, C1, showed a greater secondary copper sulfide presence and pyrite (23.2% Cu and 20.6% Fe). The concentrate C2, practically did not have secondary copper sulfides and it had very little pyrite amount (21.85% Cu and 31.14% Fe). Mesophilic microorganisms at 35ºC in presence and absence of Ag(I) as catalytic agent were used. The Ag(I) ion showed an important catalytic effect on concentrate C1 bioleaching, whereas on concentrate C2 the effect was not noticed, and an smaller copper extraction was observed. The copper extraction increased doping C2 concentrate with pyrite and chalcocite. Keywords: bioleaching, chalcopyrite, pyrite, chalcocite, catalytic effect, mesophilic 1.

INTRODUCTION The environmental requirements imposed on pyrometallurgical processes of sulfide mineral concentrates in the copper industry have forced the development of hydrometallurgical routes as alternatives for the conventional treatment of sulfide minerals concentrates in order to avoid the SO2 production. Those processes involve sulfide oxidation either to sulfur or sulphate using oxidating agents such as O2 or Fe(III) ions or by a direct anodic oxidation in an electrolyte. This oxidation can be considered as an electrochemical reaction (1), with the cathodic reduction of the oxidant and the anodic oxidation of sulfide (2). The first idea that the existing chemical interactions on the surface of minerals could be of electrochemical nature was proposed by Salamy and Nixon (3). Bacterial leaching is an economical and widely used method for metal extraction, but until now its application has been limited to low metallic content minerals (4). During the last two decades, this process has been successfully applied to refractory mineral of Au and Ag treatment, and recently has been used for cobalt recovery from pyritic material in Uganda (5). * Corresponding author. Main address: Depto de Ingeniería Metalúrgica. Circuito Institutos s/n., Facultad de Química UNAM. Ciudad Universitaria. México D. F. 04510. México. E-mail: [email protected] Phone +525556225241, fax: +525556225228.

203


Nevertheless, in many cases, the slow kinetic of the biooxidation processes has limited its commercial application. This slowness is attributed to different parameters such as biological, physicochemical, electrochemical and mineralogical factors (6). Among mineralogical factors affecting the bioleaching systems is the initial composition of the material that, will cause that the systems will respond different ways to the same treatment conditions. As an attempt to accelerate the kinetics of chalcopyrite dissolution, the use of catalytic agents has been proposed. The catalytic effect of silver, Ag(I), during the chalcopyrite leaching has been reported (7, 8, 9). The improvement in the dissolution rate is attributed to the formation of a film of Ag2S on the surface of chalcopyrite particles (8). According to the semi-conducting characteristics of sulfide minerals (6), the electrochemical interactions (galvanic pairs formation) originated among different sulfide minerals in a same bioleaching system could improve the selective dissolution of the most active minerals. Although these interactions have been known for some time, it is complicated either to explain or predict their effects on the bioleaching rate in a sulfide mineral mixture (10). In this work the results of bioleaching tests with mesophilic microorganisms, in absence and in the presence of Ag(I) of two concentrates of chalcopyrite, as well as the results of bioleaching of the concentrate C2 doped with pyrite and chalcocite, also in absence and in the presence of Ag(I) are presented. 2.


2.1 Minerals Two concentrates of chalcopyrite from Sonora and Zacatecas (Mexico), C1 and C2, respectively, were used. The samples of C2 were doped with 10% of pure minerals of pyrite (Py) or chalcocite (Ct), also from Zacatecas, Mexico. 2.2 Cultures The mesophilic microorganisms used were mixed cultures obtained from the own microflore of the concentrates, adapted by successive steps and developed in 100 mL of nutrient medium and 5g of concentrate. Three successive steps were carried out. 2.3 Bioleaching tests The bioleaching studies were made in an orbital incubator with temperature and stirring controlled to 35ºC and 150 rpm, respectively. The bioleaching tests were made with 90 mL of medium, 10 mL of inoculum and 5g of concentrate. All the tests were made in Erlenmeyer flasks of 250 mL. 2.4 Monitoring and control techniques Periodic measurements of pH and redox potentials, ORP, were made. The pH was fixed to the necessary value by addition of a diluted solution of H2SO4. The bacterial growth evolution was obtained determining the cellular concentration in solution samples counting the cells using an optical microscope with a Neubauer chamber. The analysis of metallic values in solution (Cu and Fetot) was conducted by atomic absorption spectrophotometry. The main mineral phases present in the concentrates were identified by means of xrays diffraction, XRD. 204


3.


3.1 Chemical analyses and characterization of the samples The chemical analyses of the resulting solutions of an acid attack to the concentrates allowed determining the following concentrate composition: Table 1. Chemical composition of the concentrates C1: 23.20% Cu, 20.60% Fe C2: 21.85% Cu, 31.14% Fe

Studies of XRD showed variations in the initial composition of the concentrates, exhibiting one of them, C1, a greater secondary copper sulfide and pyrite presence. The other concentrate, C2, did not have secondary copper sulfides, and contained very little pyrite and an important amount of arsenopyrite, figures 1A and 1B.

Figure 1A. Diffractogram of C1, and Figure 1B. Diffractogram of C2. (Cv: covellite, SiO2: silica, Cp: chalcopyrite, Py: pyrite y As: arsenopyrite) In figures 1A and 1B a clear difference between C1 and C2 is observed. With respect to the secondary copper sulfide, covellite, it was present in the C1 sample but not in the C2 sample. The pyrite presence in C1 was greater than in C2. The presence of arsenopyrite in C2 was not detectable in C1. Diffractograms of synthetic composites (results not shown) presented the signal corresponding to pyrite (sample C2Py) and chalcocite (sample C2Ct). 3.2 Copper and iron extraction Figure 2A shows an important difference between the reactors of C1, with and without Ag(I), results of reference (12), and the reactors of C2, with and without Ag(I). These last ones practically did not differ between each other. It is necessary to mention that the concentrate C1 contained secondary copper sulfides, which are less recalcitrant to the acid dissolution, present a faster dissolution wich was on a greater copper extraction: in 22 days 45% without Ag(I) and 62% with Ag(I). For C2 a copper extraction of only 16% in almost 80 days was reached in both cases. In figure 2B it can be observed that the iron extraction in systems with C1 and C2 was very similar with the exception of the C1 reactor with Ag(I), where a greater extraction in a smaller residence time than with C2 was reached: 25% of extraction in 30 days. The recalcitrance of C2 with respect to C1 was greater. In the reactors with C1, figure 2A, an important amount of copper came from secondary sulfides, because in the dissolution of the chalcopyrite equal amounts of both 205


copper and iron are practically dissolved. Except for the C1 with Ag(I), all of them had a very similar behavior for iron dissolution.

Figures 2A and 2B. Cu and Fe dissolution respectively. ▲C1 without Ag(I), • C1 with Ag(I), ■ C2 without Ag(I) and ♦ C2 with Ag(I) Under the testing conditions, a slower dissolution of C2 was obtained, with a residence time of 80 days. Because of the poor results obtained in the bioleaching tests of C2, it was decided to doped systems of this concentrate with pure minerals of pyrite and chalcocite in order to corroborate the influence of these sulfides on the copper extraction using Ag(I). Figures 3A and 3B show the copper and iron extraction in the bioleaching systems under different testing conditions: pyrite and chalcocite doped C2 systems in presence of Ag(I). In figure 3A it is observed that, at the beginning of the experiments, the reactors with greater copper dissolution were the chalcocite-C2 doped systems, due to the acid attack given by: CuS + 2H+ → Cu2+ + H2S

(1)

Figures 3A and 3B. Cu and Fe extraction respectively. ♦C2 + Ag(I), ■ Pyrite-C2 doped + Ag(I) and ▲Chalcocite-C2 doped + Ag(I) The massive dissolution of the chalcopyrite initiates after 70 days, obtaining a substantial increase in the copper extraction in the pyrite-C2 system reaching around 50% in 84 days. In all systems a greater copper dissolution is observed around day 70, agreeing this with the dissolution of iron, attributed, therefore, to the attack of the chalcopyrite. The higher copper recovery is obtained in the pyrite-C2 doped system, 50% in 84 days; surpassing by almost 20% that of the system also doped but with chalcocite for the same residence time. 206


The iron extraction curves, figure 3B, present a slightly different behavior from the curves of copper extraction, being at the end of the tests the reactors pyrite-C2 doped and C2 those reached the greater extraction. In the chalcocite-C2 doped systems and in the systems containing only C2, it can be seen a slight increase in the iron dissolution. A possible reason for this behavior is the acid attack only on chalcocite, and therefore, the attack to the crystalline structure of the chalcopyrite was almost negligible. Figures 4A and 4B show the copper and iron extraction in the bioleaching pyrite-C2 and chalcocite-C2 doped systems in absence of Ag(I).

Figures 4A and 4B. Cu and Fe extraction respectively. ♦C2, ■ Pyrite-C2 doped and ▲Chalcocite-C2 doped It can be seen, in figure 4A, the acid attack to chalcocite crystalline structure occurred based on the high copper extraction in that chalcocite doped system, almost the same that with the pyrite doped system, 31% in 84 days. This result verifies the fact that the Cu in solution in these systems came mainly from the secondary sulfides that were added, and in the case of the system containing only C2 it resulted from the slight attack to the chalcopyrite structure. Results of MEB (not shown) demonstrated a preferential attack on secondary copper sulfides greater than on the chalcopyrite particle surface. In the case of the pyrite doped system without Ag(I) a slight reduction in the time needed to initiate the massive iron dissolution, 5 days with respect to others systems was observed. The same behavior was observed in the copper extraction curve of C2, figure 4A and 4B. This agrees with observations made by other authors (6,10,11) in the sense that, in a bioleaching system where different sulfide minerals are present, their respective rest potentials (Erep) will create galvanic interactions that will influence the selective dissolution of the most active minerals. 3.3 pH and redox potential, ORP In figures 5A, 5B, 6A and 6B the results corresponding to representative curves of redox potential, ORP, and pH evolution for pyrite-C2 and chalcocite-C2 doped systems with and without Ag(I), respectively, are showed. Previous work have showed the effect of the presence of Ag(I) on the redox potential behavior on bioleaching systems (12). In figure 5A it is observed that the ORP evolution was affected by the presence of Ag(I) in the catalyzed systems.

207


The way in which Ag(I) affects the chalcopyrite dissolution is depicted by the following electrochemical reaction between Ag(I) and chalcopyrite: CuFeS2 + 4Ag+ → 2Ag2S↓ + Cu2+ + Fe2+ (2) Later, the silver sulfide generated in (2) reacts with the ferric ion, Fe(III), to form: (3) Ag2S(sup) + 2Fe3+ → 2Ag+ + 2Fe2+ + Sº This reaction consumes Fe(III) ion driving a decrease of the ORP value corresponding to Nernst equation on the Fe(III)/Fe(II) ratio. In the representative curves of catalyzed doped systems pH, figure 5B, a very similar behavior in all the cases was observed, except on chalcocite doped systems, where a slight increase of pH was recorded during the first 5 days due to the H+ consumption by the acid attack on chalcocite (reaction 1).

Figures 5A and 5B. Redox potential and pH respectively. ♦C2 + Ag(I), ■ Pyrite-C2 doped + Ag(I), ▲Chalcocite-C2 doped + Ag(I) From curves 6A it can be appreciated that the system that reached the highest value of ORP, and also with greater iron dissolution, figure 4B, was the pyrite-C2 doped system. In the other cases the values of redox potential are rather similar and only in those systems in which chalcocite was added the values of ORP were slightly low. In the final stage of the experiments, figure 6B, the reactors showed a slight increase in the pH value due to the ability of protons to attack the chalcopyrite (13) and due to a decrease in bacterial activity.

Figures 6A and 6B. Redox potential and pH of pyrite-C2 and chalcocite-C2 doped bioleaching systems in absence of Ag(I)

208


4.

CONCLUSIONS •

The initial composition of sulfide minerals affected the kinetics of bioleaching, which was influenced by the galvanic interactions generated when putting in contact minerals of different rest potentials.

•

The galvanic interactions follow the thermodynamic prediction of galvanic series of sulfides.

•

The chalcocite presence did not improve the copper extraction from chalcopyrite.

•

The presence of pyrite improved the chalcopyrite bioleaching.

•

The catalytic effect of the Ag(I) was affected by the presence (or absence) of pyrite in the bioleaching systems.

•

The highest copper extraction was reached in the pyrite-C2 doped system in the presence of Ag(I): 50% in 84 days.

ACKNOWLEDGEMENTS One of the authors, A. L. J., wishes to acknowledge the support of CONACyT and DGEP-UNAM for a Ph. D., scholarship. Both authors gratefully acknowledge DGAPAUNAM for the economical support through Project IN210000. REFERENCES 1. R. S. McMillan, D. J MacKinnon, and J. E. Dutrizac. J. of App. Electrochem 12 (1982) 743-757. 2. D. F. A. Koch. J’OM Bockris and B. E Conway, Modern Aspects in Electrochemistry. Plenum Press. 10 (1971) 211-237. 3. S. G. Salamy and J. C. Nixon, 2nd International Congress on Surface Activity 3 (1957) 369. 4. E. Gómez, A. Balleter, M. L. Blázquez, F. González, Hydrometallurgy 51 (1999) 3746. 5. T. A. Fowler, F. K. Crundwell. In Biohydrometallurgy and the Environmental Toward the Mining of the 21st Century, IBS’99 R. Amils and A. Ballester, (eds), Part A, (1999) 273. 6. K. A. Natajaran, Metal. Trans. 23B (1992) 5-11. 7. P. B. Munoz, J. D. Miller, M. E. Wadsworth. Metal. Trans. B. 10 (1979) 149-158. 8. J. D. Miller, H. Q. Portillo in 13th International Mineral Processing Congress J. Laskowwski (ed), Warsaw, (1979) 851-894. 9. M. L. Blázquez, A. Álvarez, A. Ballester, F. González and J. A. Muñoz. In Biohydrometallurgy and the Environmental Toward the Mining of the 21st Century, IBS’99 R. Amils and A. Ballester, (eds), Part A, (1999) 137. 10. Y. A. Attia and El-Zeky. Inter. J. of Min. Proc. 30 (1990) 99-111. 11. D. W. Dew, C. Van Buuren, K. McEwan and C. Bowker. In Biohydrometallurgy and the Environmental Toward the Mining of the 21st Century, IBS’99 R. Amils and A. Ballester, (eds), Part A, (1999) 229. 12. A. López-Juárez, R. E. Rivera-Santillán, A. Ballester Pérez, M. L. Blázquez Izquierdo and J. A. Muñoz Sánchez. In Biohydrometallurgy: Fundamentals, Technology and Sustainable Development. V. S. T. Ciminelli and O. Garcia Jr. (eds), Part B, (2001) 661. 13. A. Schippers and W. Sand. App. Environ. Microbiol. 65 (1999) 319-321. 209


Comparison of air-lift and stirred tank batch and semi continuous bioleaching of polymetallic bulk concentrate D. R. Tiprea, S. B. Vorab and S. R. Davea* a

Department of Microbiology, School of Sciences, Gujarat University, Ahmedabad 380 009, Gujarat, India b Gujarat Mineral Development Corporation, Khanij Bhavan, 132 ft. Ring Road, Ahmedabad 380 052, Gujarat, India

Abstract Metal bioextraction from GMDC polymetallic bulk concentrate was studied in stirred tank and air-lift reactors with a working volume of 5 dm3 and 20% (w/v) pulp density. The inoculum used in these studies was a consortium mainly consisting of Acidithiobacillus ferrooxidans, Leptospirillum ferrooxidans and Acidithiobacillus thiooxidans. Under the optimum conditions air-lift reactor gave 41, 75 and 65% of copper and zinc extraction and galena oxidation as compared to the STR giving 76, 70 and 80% respectively in 5 to 9 d of incubation time. During the process, 590 and 547 mV redox potential, 0.05 and 0.07 g/l of soluble ferrous and pH of 1.55 and 1.78 were recorded in air-lift reactor and STR respectively. When the total 20% (w/v) pulp density was added in four equal fractions in STR study, instead of one lot addition, the metal extraction rate increased by 1.22 and 1.1 fold for copper and zinc respectively. Use of semi-continuous fed-batch STR process further enhanced the copper and zinc extraction rate by 1.6 and 1.54 fold as compared to batch STR process. The highest metal extraction of 92 and 87% of copper and zinc respectively was achieved in semi continuous process. Acid consumption was 75% less in fed-batch semi continuous process as compared to batch process. Optimisation of the process and use of developed inoculum reduced the bioleaching time from 30 d to as low as 5 and 7 d for copper and zinc respectively with overall increased percent metal extraction. The reduction in contact time could make the polymetallic bulk concentrate bioleaching process economically viable. Keywords: polymetallic concentrate, semi continuous, air-lift, stirred tank 1.

INTRODUCTION Bioleaching has generated intense research activity in late nineties which, resulted in important findings in the field, such as the development of economics and engineering factors of biometal extraction [1]. Bioextraction processes are now applied on commercial scale for the extraction of copper, cobalt, gold and nickel from refractory ores and concentrates [2, 3]. Microbial leaching has been used at laboratory scale for the base metal sulphides of Co, Ga, Mo, Ni, Pb and Zn. Biohydrometallurgical processes are developed * Corresponding author: S.R. Dave, E-mail: [email protected]

211


for extraction of nickel and cobalt from their sulphidic concentrates using mesophilic iron and sulphur oxidising bacteria in stirred tank reactor [2]. Several factors such as oxygen requirement, nutrient availability, agitation speed and inoculum size plays an important role in metal extraction. However, the operating solid concentration constitutes one of the most critical parameter of a bioleaching process in terms of size of the equipment. At higher operating pulp concentration the above mentioned factors could limit the bioleaching efficiency [4-6]. The previous laboratory pilot scale studies were mainly carried out only up to 10% solids with about 50 to 80% metal extractions in 15 to 20 days [1, 7, 8]. The indirect two-stage bioleaching process from complex sulphidic Cu-Pb-Zn concentrate is developed in which, metal extraction is done at 70°C temperature [9]. But, the direct bioextraction process for such complex polymetallic bulk concentrate is poorly developed. India has well scattered small reserves of polymetallic ores and Ambamata Mine situated in Gujarat is one of them. Biomineral processing holds great potential for such reserves in India. In India, inspite of cheap labour, liberalised market, vast consumer base and strong foundation of hydrometallurgy, there exists a wide gap between the existing potential and the potentials to be exploited for economic metal growth [10, 11]. In biohydrometallurgical processes the stirred tank and air-lift reactors are widely used at laboratory scale but pulp density studied usually is upto 12.5% [12]. In this context, in the presented work air-lift and stirred tank batch and semi continuous bioleaching of polymetallic bulk concentrate was carried out after optimisation at shake flask leaching scale. Comparative bioleaching profiles of both these reactors at laboratory scale with 20% pulp density is discussed. 2.


2.1 Polymetallic concentrate Polymetallic concentrate was procured from Gujarat Mineral Development Corporation (GMDC), Ambamata Multimetal Project, Gujarat, India. Major constituents of the concentrate were sphalerite, chalcopyrite, galena and pyrite. The concentration of copper, lead, zinc, iron and sulphur in the concentrate were 2.5, 13, 30, 9 and 27% respectively. Detailed composition is given elsewhere [13]. The bulk concentrate received was of mixed particle size ranging between +150-325 # B.S.S. and was used as received. In all the experiments 20% (w/v) pulp density was used. 2.2 Inoculum The inoculum used for extraction of metals was developed from shake flask leaching experiment in the form of leachate. Leachate used in the experiment mainly consisted of a consortium of acidophilic chemolithotrophic auto- and heterotrophic iron and sulphur oxidisers [14]. All the experiments were carried out with 20% (v/v) leachate as inoculum and M-2 modified medium [15] prepared in tap water. 2.3 Reactors Bioextraction studies were compared at laboratory scale with two reactors. First, an Air Lift Glass Reactor (ALGR) which was fabricated in the laboratory consisting 5 dm3 working volume. Detailed design set up and configuration is given elsewhere [12]. The other bioreactor tested was a laboratory scale Stirred Tank Reactor (STR) with axial turbine type impellers which was designed in our laboratory and fabricated by Texfab 212


Manufacturers, India. The reactor consisted 5 vessels (3 in cascade and 2 in series) each 5 dm3 capacity with working volume of 3 dm3. The rate of agitation, aeration and CO2 supply was 300 rev/min, 0.5 l/min/v and 0.2% (v/v) of compressed air respectively. The detail configuration is quoted elsewhere [10, 11, 15]. Temperature during the investigation was 30 ± 5°C, i.e. ambient room temperature. 2.4 Bioleaching study In both the reactors typical batch leaching trial was operated until the logarithmic growth and steady leaching conditions were well established. In STR the pulp addition mode was tested with the four equal parts i.e. instead of addition of pulp 20% (w/v) at the 0th h, it was added in four fractions of 5% (w/v) each at 24 h of interval. The semi continuous bioleaching study was carried out in STR with 7 d of residence time. The detail design is given in the previous article [5]. 2.5 Analysis Supernatant of the leaching system was analysed periodically. pH and redox potential (Eh) was determined using Systronics Digital µ pH system 361 with platinum SCE couple electrode. Soluble ferrous iron was estimated by potassium dichromate titrimetric method. Copper, zinc and lead were estimated by spectrophotometric (Systronics UV–Vis spectrophotometer 119), polarographic (ELICO Polarograph Model CL-25 D) and titrimetric (tannic acid as external indicator) methods respectively from the leached solutions [16]. The analysis was confirmed by Atomic Absorption Spectrophotometer (Varian AA-175 model). 3.

RESULTS AND DISCUSSION After the amenability and bioleaching optimisation study of GMDC polymetallic bulk concentrate at shaken flask level [14, 17], the batch and semi continuous fed batch processes were performed with 20% (w/v) pulp density in a laboratory scale reactor. The performances of two reactors were compared for batch leaching with airlift glass reactor and stirred tank reactor, which were challenged with the consortium developed at shake flask as 20% (v/v) leachate. As can be seen from the data presented in Table 1, in ALGR highest copper extraction achieved was 40.1% while, lead and zinc were 65 and 75% respectively at 9 d of contact time in batch leaching. Throughout the study, the extraction of lead represent oxidation of galena to lead sulphate, which remain in insoluble form with the pulp. The system was stabilised after 6 d of residence time which can be noted from the pH, redox potential and soluble ferrous being 1.55, 590 mV and 0.05 g/l respectively at the end of 9 d. The total acid consumption recorded was 19.3 g sulphuric acid/Kg concentrate. This acid requirement was to bring pH back to 2.0 could be due to the acid consuming material present in the pulp and the chemical oxidation of metal sulphide [12].

Thereafter, the gradual decrease in the pH indicates the enhancement in biooxidation of sulphide content of the concentrate.

213


Table 1. Batch bioleaching profile of Air Lift Glass Reactor Bioleaching time (days) Leaching profile

Before inoc. 2.0 308 0.62

After inoc. 2.35 320 0.97

pH Redox potential (mV) Soluble Fe2+ (g/l) Acid consumption 2.9 (g/Kg concentrate) Copper (%) 2.2 4.8 Zinc (%) 12.91 15.92 Lead (%) 4.5 4.9 n.d.: not determined; inoc.: inoculation

2

4

6

8

9

2.7 330 1.06

2.6 370 0.09

1.6 520 0.06

1.55 580 0.05

1.55 590 0.05

6.4

10.0

-

-

-

11.22 24.38 n.d.

24.36 50.35 n.d.

41.0 64.98 n.d.

31.6 71.48 n.d.

40.1 75.0 65.0

When the leaching profile of STR and ALGR batch studies were compared (Table 2), STR performance was proved better than that of ALGR. The maximum copper, lead and zinc extraction recorded in STR were 76, 70 and 80% in 6, 9 and 9 d which comes out to be 2.84, 1.06 and 1.07 folds higher as compared to ALGR. The acid consumption was 10.7 g acid/Kg concentrate which was 1.8 times lower than the ALGR. This indicates highly stabilised system, which was due to the positive effect of aeration and agitation system adopted during the process [18]. The observed high chalcopyrite leaching indicates dominance of biological leaching over chemical leaching as chalcopyrite was very difficult to leach chemically. Table 2. Bioleaching profile of batch STR process Bioleaching time (days) Leaching profile pH Redox potential (mV) Soluble Fe2+ (g/l) Acid consumption (g/Kg concentrate) Copper (%) Zinc (%) Lead (%)

Before inoc. 2.0 326 0.66

After inoc. 2.52 344 1.22

2

4

6

8

9

2.5 354 1.53

2.29 382 0.99

1.84 429 0.73

1.78 528 0.09

1.78 547 0.07

-

5.73

4.97

-

-

-

-

3.8 8.4 5.1

8.4 20.68 5.5

34.36 39.36 n.d.

57.14 48.68 n.d.

76.0 62.93 n.d.

66.8 72.0 n.d.

65.96 80.0 70.0

Looking to the promising leaching result of STR, further experiment was performed with fractional pulp addition and results are shown in Table 3. As can be seen from the data 92.7% copper extraction was achieved in 140 h of contact time followed by 83.62 and 83% zinc and lead extraction respectively in 215 h of reactor run. The high metal extraction during this fractional addition of pulp could be due to the less shear effect generated and better gas exchange rate owing to low pulp density at any particular time throughout the process. The high biological activity throughout the process was responsible for constant neutralisation of the alkaline gangue present in the fraction pulp added at that time, which resulted in 1.96 times less acid consumption compare to the STR batch process. Even when 30% (w/v) pulp was added in 10+10+10% (w/v) fractions, extraction rates of 0.84 and 6.36 g/l/d for copper and zinc respectively (data not shown) were achieved. This suggests that with the developed inoculum it is possible to get the 214


considerable metal extraction even at such a high pulp density, which was almost equivalent to that achieved at 15% pulp density with fractional pulp addition [5]. Table 3. Bioleaching profiles during fractional addition of pulp in STR Bioleaching time (days) Leaching profile

Before inoc. 1.95 322 0.42

After inoc. 2.24 332 0.50

2

4

6

8

9

2.16 373 0.75

2.25 412 0.94

2.12 430 1.11

1.93 446 0.80

1.82 525 0.09

-

1.82

1.82

1.82

-

-

-

Copper (%)

6.2

18.4

56.80

64.57

92.7 (140h)

71.34

Zinc (%)

12.8

25.15

58.0

63.54

69.08

74.34

Lead (%)

5.3

6.0

n.d.

n.d.

n.d.

n.d.

pH Redox potential (mV) Soluble Fe2+ (g/l) Acid consumption (g/Kg concentrate)

83.62 (215h) 88.0 (215h) 83.0 (215h)

The better extraction results achieved during fractional addition of the pulp in STR study opened the way for semi continuous fed batch process. The semi continuous metal extraction was started with a stabilised leaching system which, reached to 75% metal extraction, thus even at initiation time the concentration of the extracted metals present in leachate was 33 ± 2.0%. At the time of calculation of the percent metal extraction this amount was substracted, and the results presented in Table 4 are of net average extraction of 15 cycles. The highest metal extractions of 92, 87 and 80% were achieved with copper, zinc and lead respectively. The metal extraction time was reduced by 35 h for copper while for lead and zinc it was reduced by 65 h as compared to the bioleaching with fractional pulp addition STR process. The bioleaching time is the major factor affecting the economy of the process. The other cost factor in the process was the external acid addition or acid consumption, which was 3.99 fold less or 75% reduction compare to STR batch leaching process. Table 4. Polymetallic bioleaching profile in semi continuous STR process based on average of 15 cycles Bioleaching time (days) Leaching profile

Before inoc.

After inoc.

2

4

6

7

pH Redox potential (mV) Soluble Fe2+ (g/l) Acid consumption (g/Kg concentrate)

2.06 321 0.79

2.23 338 1.06

2.15 397 1.23

2.06 448 0.94

1.96 502 0.08

1.80 521 0.07

-

1.73

0.95

-

-

-

Copper (%)

31.59

37.4

71.86

92.0 (105h)

73.54

65.98

Zinc (%)

32.63

45.68

57.07

63.08

73.81

Lead (%)

35.12

36.4

n.d.

n.d.

n.d.

87.0 (150h) 80.0 (150h) 215


The comparative metal extraction rates of all the four studied processes are depicted in Table 5. As can be seen from the data, a higher metal extraction rate was obtained in the STR semi continuous process as compared to any of the processes studied. This was 43.81 and 348.0 mg/l/h, which comes out to be 1.05 and 8.35 g/l/d copper and zinc solubilisation respectively. The total metal extracted from the polymetallic concentrate was 4.6 g/l copper, 52.5 g/l zinc and galena oxidised equivalent to 21.12 g/l or 105.60 g/Kg lead in leached residue in the form of lead sulphate. Table 5. Comparative soluble metals in leachate, galena oxidised and the overall metal extraction rate Metals Zinc

Copper

Process g/l

mg/l/h

g/l

mg/l/h

ALGR 2.05 14.64 45.0 209.30 STR (batch) 3.80 27.14 48.0 223.36 STR (fraction) 4.64 33.14 52.8 245.58 STR (semi 4.60 43.81 52.5 348.00 continuous) *: equivalent lead due to oxidation of galena to lead sulphate

g/l* 17.16 18.48 21.91 21.12

Lead g/Kg mg/l/h* concentrate 85.80 79.81 92.40 85.95 109.55 101.92 105.60 140.80

The selection and development of the efficient bioleaching consortium and modification of pulp addition resulted in reduction of leaching time from 30 d at shake flask level with wild type consortium [14] to as low as 5 to 6 d. The observed high metal extraction as well as high rate proved suitability of the semi continuous process over the other three processes studied. Moreover, the reduction in retention time as well as external acid consumption leads to economization of the process. On the basis of obtained data the bioextraction process was successfully scaled-up to 600 dm3 STR level in our laboratory and was efficiently operated for 17 cycles with more than 80% average metal extractions which is detailed elsewhere [13]. The noteworthy developments of this investigation are the room temperature operations, high pulp density as well as the higher zinc extraction rate obtained as compared to the reported values in literature [1, 7, 8, 19-21]. The zinc extraction data obtained in this investigation can be compared with those of Steemson et.al [21] who have reported 93.8% zinc extraction in the overall residence time of 3.7 days at 40-45°C temperature and 6-8% solids. The highest overall rate of zinc extraction achieved in this work is 348 mg/l/h. When the amount of zinc extraction is compared per day on one litre pulp volume basis, the extraction works out to be 8.35 g/l/d, which is comparable with the calculated value of 8.75 g/l/d from the Steemson et.al report [21]. What is interesting is, that the presented work was performed at ambient temperature compared to 40-45°C. Moreover, the pulp density is 3 times higher in this investigation compared to the above reference. When adopted on commercial basis the process may be more cost effective in terms of capital as well as operating costs. 4.

CONCLUSION To summarise, the achieved equivalent zinc extraction rate, higher than that reported by others, even at lower temperatures and higher pulp density using indigenously developed consortium, indicates comparatively higher cost effectiveness of the bioleaching process developed for GMDC polymetallic bulk concentrate. Interestingly, the 216


investigation was carried out with a Cu-Pb-Zn system, which allowed as much as 90% copper extraction and 80% oxidation of lead sulphide simultaneously. ACKNOWLEDGMENTS We are thankful to Gujarat Mineral Development Corporation for the project grant and research fellowship to D. R. Tipre. REFERENCES 1. Y. Rodriguez, A. Ballester, M. L. Blazquez, F. Gonzalez and J. A. Munoz, In: Biohydrometallurgy: Fundamentals, Technology and Sustainable Development, Part A, V. S. T. Ciminelli and O. Garcia Jr. (eds.), Proc. Intl. Biohydrometallurgy Symp., Elsevier, Amsterdam, (2001) 125. 2. J. A. Brierley and C. L. Brierley, In: Biohydrometallurgy and the Environment toward the Mining of the 21st Century, Part A, R. Amils and A. Ballester (eds.), Proc. Intl. Biohydrometallurgy Symp., Elsevier, Amsterdam, (1999) 81. 3. C. L. Brierley, In: Biomining: Theory, Microbes and Industrial Processes, D. E. Rawlings (ed.), Springer-Verlag, New York, (1997) 3. 4. A. D. Bailey and G. S. Hansford, Biotechnol. Bioengg., 42(10) (1993) 1164. 5. S. R. Dave, D. R. Tipre and S. B. Vora, In: Biohydrometallurgy: Fundamentals, Technology and Sustainable Development, Part A, V. S. T. Ciminelli and O. Garcia Jr. (eds.), Proc. Intl. Biohydrometallurgy Symp., Elsevier, Amsterdam, (2001) 561. 6. D. R. Tipre, S. B. Vora and S. R. Dave, In: Mineral Biotechnology, L. B. Sukla and V. N. Misra (eds.), Proc. Nat. Seminar, Allied Publishers Pvt. Ltd., New Delhi, (2002) 81. 7. R. E. Rivera, A. Ballester, M. L. Blazquez and F. Gonzalez, In: Biohydrometallurgy and the Environment toward the Mining of the 21st Century, Part A, R. Amils and A. Ballester (eds.), Proc. Intl. Biohydrometallurgy Symp., Elsevier, Amsterdam, (1999) 149. 8. A. Chin, In: Biotechnology comes of age, BIOMINE'97, Proc. Intl. Biohydrometallurgy Symp., Australian Mineral Foundation, Australia, (1997) M1.2.1. 9. R. Romero, I. Palencia and F. Carranza, Hydrometallurgy, 49 (1998) 75. 10. A. R. Tipre, Ph. D. Thesis, Gujarat University, Ahmedabad, Gujarat, India, 1999. 11. S. R. Dave, D. R. Tipre and S. B. Vora, In: Proc. Intl. Seminar on Mineral Processing Technology, S. Subramanian (ed.), Bangalore, (2002) (in press). 12. A. R. Tipre, S. B. Vora and S. R. Dave, Ind. J. Microbiol., 41 (2001) 173. 13. S. R. Dave, D. R. Tipre and S. B. Vora, In: Application of Chemical Engineering for Utilisation of Natural Resources, G. K. Roy, C. R. Mishra and K. Sarveswara Rao (eds.), New Age International Publishes, New Delhi, (2001) 169. 14. A. R. Tipre, S. B. Vora and S. R. Dave, J. Scientific Ind. Res., 57 (1998) 805. 15. D. R. Tipre, S. B. Vora and S. R. Dave, In: Biohydrometallurgy and the Environment toward the Mining of the 21st Century, Part A, R. Amils and A. Ballester (eds.), Proc. Intl. Biohydrometallurgy Symp., Elsevier, Amsterdam, (1999) 219. 16. A. I.Vogel, A Textbook of Quantitative Inorganic Analysis, 3rd ed., ELBS and Longman, London, 1962. 17. A. G. Menon and S. R. Dave, In: Biohydrometallurgical Technologies, Vol. 1, A. E. Torma, J. E. Wey and V. L. Lakshamanan (eds.), Proc. Intl. Biohydrometallurgy Symp., TMS, Wyoming, (1993) 137. 18. D. R. Tipre, S. B. Vora and S. R. Dave, In: Biotechnology of Microbes and Sustainable Utilisation, R. C. Rajak (ed.), Scientific Publishers (India), Jodhpur, (2002) 263. 217


19. T. A. Fowler and F. K. Crundwell, In: Biohydrometallurgy and the Environment toward the Mining of the 21st Century, Part A, R. Amils and A. Ballester (eds.), Proc. Intl. Biohydrometallurgy Symp., Elsevier, Amsterdam, (1999) 273. 20. P. Chawakitchareon, B. Buddharuksa and S. Pradit, In: Biotechnology comes of age, BIOMINE'97, Proc. Intl. Biohydrometallurgy Symp., Australian Mineral Foundation, Australia, (1997) M1.3.1. 21. M. Steemson, F. Wong, B. Goebel, In: Biotechnology comes of age, BIOMINE'97, Proc. Intl. Biohydrometallurgy Symp., Australian Mineral Foundation, Australia, (1997) M1.4.1.

218


Effect of pH and temperature on the biooxidation of a refractory gold concentrate by Sulfolobus metallicus I. Ñancucheo, J. C. Gentina* and F. Acevedo School of Biochemical Engineering, Catholic University of Valparaiso, Av. Brasil 2147, Valparaíso, Chile Abstract The biooxidation of refractory gold concentrates using thermophilic microorganisms is considered an interesting alternative to current processes. Many operation conditions influence the process, pH and temperature being two important parameters among others. The objective of this work was to evaluate the influence of these two variables on the biooxidation of an enargite-pyrite gold concentrate containing 40 g Au/ton, using the thermophilic archaeon Sulfolobus metallicus. The experiments were run in shake flasks with 1% pulp density and particle size under 200 mesh. The pH was kept constant at 1.0, 1.5 and 2.0 by daily addition of 0.5 M NaOH. One experiment was run at non-controlled condition with initial pH at 2.5. Every pH condition was tested at 60, 65, 70 and 75ºC. The extent of biooxidation was measured through the iron solubilisation. The best run at constant pH condition and at all different temperatures was 1.5, obtaining the highest percentage of iron extraction at 65ºC which amounted 75%. Iron extractions at pH 1.0 were between 10% and 24% of those obtained at pH 1.5, while at pH 2.0 they were in the range of 75% to 95%. All experiments run at 75ºC showed almost no iron solubilisation. However, the non-controlled pH experiment with initial pH at 2.5 was shown to be a more suitable condition for biooxidation. The highest iron extraction was 83% at 65ºC which is 10% higher than the one obtained at pH controlled at 1.5. It is concluded that, under the studied conditions, the best pH and temperature to perform the biooxidation are 65ºC and a non-controlled pH policy with initial pH of 2.5. Keywords: enargite, biooxidation, thermophiles, bioleaching, archaea 1.

INTRODUCTION Biooxidation has become an attractive alternative as a pretreatment of refractory gold concentrates in order to facilitate gold extraction by cyanidation [1-4]. Biooxidation competes against roasting and pressure oxidation presenting several advantages such as low capital demand, low energy input, mild operation conditions and low environmental impact. Nowadays there are operating no less than ten large scale biooxidation plants located in South Africa, Brazil, Australia, Ghana, Peru and USA [5, 6].

* [email protected]

219


Most of these commercial gold processing plants use stirred tank bioreactors for the biooxidation of refractory gold concentrates, although a few of them use heap leaching to pre-treat low-grade ores and tailings [7]. The advantage of using stirred tank bioreactors is the control that can be exerted on important environmental variables like pH and temperature [6, 8, 9]. Microorganisms that are involved in these operations occurs naturally at mineral sites and are mostly of the genus Acidithiobacillus preferentially Acidithiobacillus ferrooxidans, Acidithiobacillus thiooxidans and Acidithiobacillus caldus together with Leptospirillum ferrooxidans [11-13]. All of them are acidophiles and mesophiles, being used at operating pH between 1.2 to 2.2 and temperature between 30 and 40ºC [14]. Several attempts have been made in order to evaluate the performance of thermophilic bacteria and more recently hyperthermophilic archaea which grow at temperatures higher than 60ºC. Some examples of these archaea are Sulfolobus metallicus, Sulfolobus acidocaldarius, Sulfolobus solfataricus, Acidianus brierleyi and Acidianus infernus [15-17]. Although these archaea have natural environments different from mineral sites, they have adapted very well to oxidize different mineral species like pyrite, arsenopyrite, enargite, chalcopyrite, chalcocite and covellite [18-21] The bioleaching mechanisms used by archaea are less understood than those of Acidithiobacillus and much more basic information is needed. This work presents results that show the influence of pH and temperature on the solubilisation of a refractory gold concentrate with high content of enargite, a recalcitrant species present in several gold minerals in Chile using the archaeon Sulfolobus metallicus. 2.


2.1 Experimental conditions A strain of Sulfolobus metallicus, kindly supplied by Dr. Antonio Ballester (Faculty of Chemical Science, Universidad Complutense, Madrid), was used throughout this study. The cells were cultured in a medium containing 0.4 g/L of (NH4)2SO4, 0.5 g/L of MgSO4.7H2O, 0.2 g/L of KH2PO4, 0.1 g/L of KCl and 1% (w/v) of gold concentrate as energy source. The mineralogical composition of the gold concentrate (El Indio Mining Company, La Serena, Chile) was: 40.7% enargite, 42.8% pyrite, 3.9% chalcopyrite, 0.8% chalcosine and 0.3% covellite. Its elemental composition was: 42 g Au/ton, 21.1% Cu, 22.6% Fe, 37.8% S and 7.7% As. Particle size was lower than 200 mesh. Experiments were carried on in 500 mL erlenmeyer flasks with 100 mL of medium, incubated in a rotary shaker at 200 rpm. The different culture pH were kept constant at 1.0, 1.5 and 2.0 by daily addition of 0.5 M NaOH. One experiment was conducted under non-controlled pH policy with an initial pH of 2.5. Each pH condition was tested at 60, 65, 70 and 75ºC. The evaporated water was quantitatively replaced on a daily basis adding distilled water. The inoculum was obtained from a fully adapted culture grown on a medium of the same composition as described above. 2.2 Analytical procedures Total soluble iron, after reduction of ferric ion with hydroxylamine, was determined by the o-phenanthroline method [22]. Eh was measured with a Ag/AgCl probe.

220


3.

RESULTS AND DISCUSSION Figures 1, 2 and 3 depict the total soluble iron concentration profiles at all pH and temperature conditions studied, with the exception of the experiments run at 75ºC, a temperature at which after eight days iron solubilisation was not observed. This temperature was too high as to allow this archaeon to grow, a result that is consistent with the culture temperature reported by different authors that use Sulfolobus species in biooxidation studies [23, 24]. Under the operating conditions used in this work, the biooxidation of the mineral stopped after eight to ten days. Also, it was observed that under the different tested conditions biooxidation started after three to four days of lag, a situation that is related to the difficulties that the cells face when they are transferred as inoculum to a fresh medium. Experiments run at pH 1.0 showed almost no iron extraction independently of the cultivation temperature. It is thought that the initial iron solubilisation should be due mostly to a chemical action more than an initial biological activity. As pyrite is the most abundant mineral component of the refractory gold concentrate, the iron obtained in solution as a consequence of biooxidation must come from pyrite and therefore constitutes a measurement of the extent of pyrite oxidation. In this respect, from the results it is clear that the highest iron extractions were obtained at a temperature of 65ºC. In the case of the biooxidation experiments conducted at controlled pH, the highest extraction was obtained at a pH of 1.5, amounting to 75% of the iron present initially in the refractory gold concentrate. Table 1 summarizes the percentages of Fe extraction obtained under the different operating pH and temperature considered in this work. The extraction values obtained at pH 1.0 are between one fifth to one tenth of those obtained at pH 1.5, which reveals the importance to keep biooxidation pH above 1.0 in order to avoid a marked reduction on the process yield.

Figure 1. Iron solubilisation kinetics during the biooxidation of a refractory gold concentrate at 60ºC and different pH (nc: non-controlled)

221


Figure 2. Iron solubilisation kinetics during the biooxidation of a refractory gold concentrate at 65ºC and different pH (nc: non-controlled)

Figure 3. Iron solubilisation kinetics during the biooxidation of a refractory gold concentrate at 70ºC and different pH (nc: non-controlled) Table 1. Iron extraction by biooxidation of a 1% pulp density refractory gold concentrate at different pH conditions and temperature (%) Temperature (ºC) 60 65 70

Culture pH Non-controlled 64 83 69

1.0 14 7 14

1.5 58 75 67

2.0 44 55 64

In the case of experiments conducted under pH controlled condition and as a way to predict the optimal pH and temperature to carry out the biooxidation of the refractory gold concentrate using Sulfolobus metallicus, the experimental data was fitted to a second order polynomial function to generate a surface response. This equation represents the 222


solubilised iron concentration [Fe] as a function of biooxidation pH and temperature The following equation was drawn using a statistical software fed with the experimental results tabulated according to a central composite rotatable experimental design:

[ Fe] = −19.2326 + 7.4686 ⋅ pH + 0.4244 ⋅ T − 3.1629 ⋅ pH 2 + 0.046 ⋅ pH ⋅ T − 0.0036 ⋅ T 2

(1)

2

The values of the coefficient of determination (R ) and the standard deviation are 0.975891 and 0.138646 respectively. Calculating the first partial derivatives of equation (1) the optimum culture pH and temperature were quantified as 1.68 and 69ºC, respectively. These operating conditions predict a soluble iron concentration of 1.72 g/L, which represents a 76% of iron extraction The optimal pH and temperature fit in the ranges reported in the literature for Sulfolobus metallicus [25, 26]. However, as can be seen in Table 1, this figure still is lower than the iron extraction obtained in the experiment run at 65ºC and under non-controlled pH condition. In this case the iron extraction was 83%, showing that the biooxidation of a 1% pulp density refractory gold concentrate by Sulfolobus metallicus was more efficient when conducted under non-controlled pH condition. Although there is not a precise explanation for this phenomenon, it has been noted that sometimes microbial activity is higher under a policy of initially adjusting the pH and letting it change freely during the time course of the oxidation. The pH varied moderately during the biooxidation as can be seen in Figure 4, leveling off at a lowest value of 1.7, very close to the one predicted by equation (1). This policy is thought to be worthwhile for the biooxidation of solid suspension of low pulp density as in this work, since the pH downfall is restricted by the total amount of iron that can be extracted. As pulp density goes up, the total iron solubilisation increases and consequently the pH drop should become more important, reaching values that would interfere with microbial activity and extent of the biooxidation process. Also in Figure 4 is observed a typical Eh behavior during biooxidation, starting from low values and increasing as solubilisation of different elements goes up, especially iron.

Figure 4. pH and Eh evolution during the biooxidation of a refractory gold concentrate at 65ºC and under non-controlled policy with initial pH of 2.5 From the data contained in Figures 1, 2 and 3, it is possible to calculate the global volumetric productivity of iron solubilisation at each experimental condition. According to Table 2, again this parameter is maximum for biooxidation conducted at 65ºC and pH 223


controlled at 1.5. This time the global volumetric productivity is slightly higher at the pH controlled condition compared to non-controlled operation. This situation is clearly confirmed in Figure 2 where it is possible to infer that the rate of iron solubilisation at pH controlled at 1.5 is faster than at the non-controlled pH condition. Table 2. Global volumetric productivity of iron solubilisation of a batch biooxidation of a 1% pulp density refractory gold concentrate at different pH conditions and temperature (g/L·d) Temperature (ºC) 60 65 70

Culture pH Non-controlled 0.132 0.208 0.171

1.0 0.036 0.023 0.041

1.5 0.129 0.211 0.160

2.0 0.099 0.206 0.179

4.

CONCLUSIONS The results allowed the definition of the pH and temperature values that maximized the iron extraction from a refractory gold concentrate through biooxidation using Sulfolobus metallicus, a thermophilic archaeon. Under the experimental conditions studied in this work, the best temperature and pH to perform the biooxidation are 65ºC and a noncontrolled pH policy starting with initial pH of 2.5. On the other side, operating at a pH controlled condition, using surface response methodology it is shown that pH controlled at 1.68 and a temperature of 69ºC maximize the extent of iron solubilisation. On the other hand, the highest value of the global volumetric productivity of iron solubilisation is found in the pH controlled operation mode and correspond to a biooxidation conducted at pH 1.5 and 65ºC. ACKNOWLEDGEMENTS This work was supported by the National Commission of Science and Technology through the FONDECYT projects 1000284 and 1020768. REFERENCES 1. S. Hutchins, J. Brierley and C.L. Brierley, Mining Eng., April (1988). 2. C.L. Brierley, Australian Gold Conference, Kalgoorlie, Australia, 1992. 3. J.A. Brierley, R.Y. Wan, D.L. Hill and T.C. Logan, Biohydrometallurgical Processing (T. Vargas, C.A. Jerez, J.V. Wiertz and H. Toledo, eds.), Vol. I, p. 253, Universidad de Chile, Santiago, 1995. 4. A.W. Breed, C.J.N. Dempers and G.S. Hansford, J. South African Inst. Min. Metall. (2000) 389. 5. S. Gilbert, C. Bounds and R. Ice, CIM Bull., 81 (1988) 89. 6. F. Acevedo, Electronic J. Biotechnol., 3 (2000). Available from Internet: http://www.ejb.org/content/vol3/issue3/full/4/index.html. 7. C.L. Brierley, Biohydrometallurgy and the Environment toward the Mining of the 21st Century (R. Amils and A. Ballester, eds.), Part A, p. 91, Elsevier, Amsterdam, 1999. 8. A. Pinches, J.T. Chapman, W.A.M. te Riele and M. van Staden, Biohydrometallurgy (P.R. Norris and D.P. Kelly, eds.), p. 329, Science and Technology Letters, Surrey, 1988. 9. F. Acevedo and J.C. Gentina, Bioprocess Eng., 4 (1989) 223. 224


10. D.W. Dew, Biohydrometallurgical Processing (T. Vargas, C.A. Jerez, J.V. Wiertz and H. Toledo, eds.), Vol. I, p. 239, Universidad de Chile, Santiago, 1995. 11. D.W. Dew, E.N. Lawson and J.L. Broadhurst, Biomining: Theory, Microbes and Industrial Processes (D.E. Rawlings, ed.), p. 45, Springer Verlag, Berlin, 1997. 12. D.E. Rawlings, H. Tributsch and G.S. Hansford, Microbiol-UK, 145 (1999) 5. 13. F.K. Crundwell, Biotechnol. Bioeng., 71 (2001) 255. 14. R.W. Lawrence and P.B. Marchant, Biohydrometallurgy (P.R. Norris and D.P. Kelly, eds.), p. 359, Science and Technology Letters, Surrey, 1988. 15. P. Norris and J.P. Owen, FEMS Microbiol. Rev., 11 (1993) 51. 16. D.W. Dew, C. van Buuren, K. McEwan and C. Bowker, Biohydrometallurgy and the Environment toward the Mining of the 21st Century (R. Amils and A. Ballester, eds.), Part A, p. 229, Elsevier, Amsterdam, 1999. 17. Y. Konishi, S. Asai and M. Tobushige, Biotechnol. Progr., 15 (1999) 681. 18. F. Kargi and M. Robinson, Biotechnol. Bioeng., 24 (1982) 2115. F. Torres, M.L. Blázquez, F. González, A. Ballester and J.L. Mier, Metall. Mater. Trans. B, 26B (1995) 455. 20. Y. Konishi, S. Yoshida and S. Asai, Biotechnol. Bioeng., 48 (1995) 592. 21. P. d’Hugues, S. Foucher, P. Gallé-Cavalloni and D. Morin, Int. J. Miner. Process., 66 (2002) 107. 22. L. Herrera, P. Ruiz, J.C. Aguillon and A. Fehrmann, Chem. Tech. and Biotechnol., 44 (1989) 171. 23. Y. Konishi, M. Tokushige and S. Asai, Biohydrometallurgy and the Environment toward the Mining of the 21st Century (R. Amils and A. Ballester, eds.), Part A, p. 367, Elsevier, Amsterdam, 1999. 24. M. Nemati, J. Lowenadler and S.T.L. Harrison, Appl. Microbiol. Biotechnol., 53 (2000) 173. 25. J.L. Mier, A. Ballester, F. González, M.L. Blázquez and E. Gómez, J. Chem. Tech. Biotechnol., 65 (1996) 272. 26. M. Gericke, A. Pinches and J.V. van Rooyen, Int. J. Miner. Process., 62 (2001) 243.

225


Effect of the pulp density and particle size on the biooxidation rate of a pyritic gold concentrate by Sulfolobus metallicus P. Valencia, J.C. Gentina and F. Acevedo* School of Biochemical Engineering, Catholic University of Valparaiso, Av. Brasil 2147, Valparaíso, Chile Abstract It is recognized that increasing pulp densities and decreasing particle sizes have positive effects in the volumetric rate of biooxidation of refractory gold concentrates. It has also been noted that a variety of phenomena can limit this positive effect. The objective of this work was to determine the values of pulp density and particle size that maximize the volumetric rate of solubilisation of iron from a pyritic gold concentrate. The leaching was carried on in agitated flasks with the thermophilic archaeon Sulfolobus metallicus. The concentrate contained 66.7% pyrite, and the constant operation conditions were 220 rpm, 68ºC initial pH of 2.0. Pulp densities were 2.5, 5, 10 and 15% w/v and the size fractions were 150-106, 106-75, 75-38 and –38 µm. Total solubilised iron concentrations were in the range of 8 to 25 g/L. In the 2.5 and 5% pulp density runs, iron extractions were in the range of 80 to 100%. After 15 to 25 days of leaching the rate declined to almost zero in the runs with 2.5 and 5% pulp densities, while the same occurred in the other two runs after 20 to 45 days. A complete experimental design of 16 runs allowed the definition of response surfaces from which the optimal conditions that maximize the rate of iron solubilisation were determined. These optimal conditions are 7.8% pulp density and particle size of 35 µm. Keywords: optimal conditions, response surface, iron solubilisation, thermophilic archaeon 1.

INTRODUCTION The rate of biooxidation of refractory gold concentrates is influenced by several operational factors. It is recognized that increasing pulp densities and decreasing particle sizes have a positive effect in the volumetric rate of biooxidation, as both situations result in an increase in surface area. Nevertheless, it has also been noted that the interaction among these factors together with a variety of associated phenomena such as mechanical effects, metabolic stress and inhibitory concentrations of ferric ion, can limit this positive effect and even result in declining leaching rates [1].

* [email protected]

227


The negative effects of high pulp densities and small particle sizes were early reported in bioleaching with mesophilic bacteria [2-3]. The detrimental effect of high pulp densities is likely to be larger in operations with archaea because of their weaker cell wall [4-5] that make them susceptible to mechanical damage and metabolic stress caused by the intense agitation needed for maintaining a homogeneous suspension [6-9]. On the other hand, decreasing particle size can reduce the leaching rate probably because of difficulties in cell attachment when the diameters of the particles and cells become of similar magnitude. It is also likely that the rate of collision between particles increases as particle size diminishes [9-10]. The objective of this work was to determine the optimal values of pulp density and particle size that maximize the volumetric rate of solubilisation of iron from a pyritic gold concentrate when using the thermophilic archaeon Sulfolobus metallicus in shake flasks. 2.


2.1 Microorganism and culture conditions A strain of the thermophilic archaeon Sulfolobus metallicus, kindly supplied by Dr. Antonio Ballester from the Universidad Complutense, Madrid, was used throughout this work. The microorganism was maintained in Norris medium [11] (0.4 g/L (NH4)2SO4, 0.5 g/L MgSO4.7H2O, 0.2 g/L K2HPO4, 0.1 g/L KCl, H2SO4 to pH 2.0) in the presence of the gold concentrate. All experiments were run in 1-L shake flasks with 90 mL of the same culture medium with the concentrate at pulp densities of 2.5, 5, 10 and 15% w/v. The flasks were inoculated with 10 mL of a culture of adapted cells with total soluble iron concentration of 10 g/L, so in all runs initial concentration was 1 g Fe/L. Uninoculated flasks were run for each pulp density. Other culture conditions were 68ºC, initial pH of 2.0 and agitation in orbital incubator of 220 rpm. 2.2 Gold concentrate The concentrate contained 15 g gold/tonne and was supplied by Las Ventanas Copper Refinery, Las Ventanas, Chile. Its mineralogical and elemental composition is presented in Table 1. Table 1. Gold concentrate composition Component Mineralogical Pyrite (FeS2) Chalcopyrite (CuFeS2) Sphalerite (ZnS) Others Gangue Elemental S Cu Fe Zn Others 228

% w/w

67.55 8.98 0.56 0.71 22.20 39.47 3.32 34.43 0.38 22.40


Four fractions of the original concentrate were used in this work: 150-106, 106-75, 75-38 and –38 µm. 2.3 Analytical methods Ferrous ion was measured colorimetrically by the modified o-phenanthroline method [12]. Total soluble iron was determined by the same method after reduction of the ferric iron with hydroxylamine. Ferric ion was calculated as the difference between total and ferrous iron. Sulfate was determined volumetrically using the Cole-Parmer (Vernon Hills, IL) sulfate kit Nº 05542-23. 2.4 Experimental design A 24 factorial design was used. The factors were particle size and pulp density. The complete set of experiments is shown in Table 2. The coded variables were generated by defining the highest value of each variable as 1 and the lowest as –1. Table 2. Experimental design with the independent physical and coded variables Run

PS (µm)

PD (% w/v)

X1

X2

1

-38

2.5

-1

-1

2

75-38

2.5

-0.312

-1

3

106-75

2.5

0.132

-1

4

150

2.5

1

-1

5

-38

5

-1

-0.6

6

75-38

5

-0.312

-0.6

7

106-75

5

0.132

-0.6

8

150

5

1

-0.6

9

-38

10

-1

0.2

10

75-38

10

-0.312

0.2

11

106-75

10

0.132

0.2

12

150

10

1

0.2

13

-38

15

-1

1

14

75-38

15

-0.312

1

15

106-75

15

0.132

1

16

150

15

1

1

PS: particle size; PD: pulp density, X1: coded particle size; X2: coded pulp density

Results were modelled by multiple regression analysis [13] based in the empiric model: Q P = b 0 + b1 x1 + b 2 x 2 + b11 x12 + b 22 x 2 + b12 x1 x 2 (1) Equation (1) was used to generate response curves of the effect of particle size and pulp density on the maximum volumetric rate of production of soluble iron, QP. Maximum iron solubilisation rates were calculated as the slopes of the straight lines drawn from the iron concentration at time zero passing tangent to the solubilisation curve.

229


3.


3.1 Bioleaching kinetics The results of the 16 runs are presented in Figure 1. At all particle size fractions, pulp density of 15% proved to be deleterious, a similar result than that obtained by Nemati and Harrison [14] with a similar Sulfolobus strain. This effect can be due to a number of factors, namely mechanical, metabolic stress and gas transfer limitations [7, 10, 14-16]. Attrition of the cells by the concentrate particles can result in mechanical damage due to the delicate nature of the archeal cell envelopes, while this same situation can cause metabolic stress. It is believed that stress can be overcome by an extended adaptation period. High pulp densities can cause oxygen and carbon dioxide demands higher than the actual supplies, as has been suggested by Gerike el al. [15], d’Hugues et al. [16] and Boogerd et al. [17]. High percent iron solubilisations were obtained for pulp densities of 2.5 and 5%, as can be seen in Table 3. Iron solubilisation was affected by increasing pulp densities and by coarser particles. In fact, most of the soluble iron obtained with 15% pulp density and the coarser fraction (run 16) came from chemical leaching as evidenced by the results of the uninoculated flasks (not shown). The contribution of chemical leaching became less important with decreasing pulp densities and particle sizes, and was not significant in the 2.5 and 10% pulp density runs. Table 3. Maximum solubilisation rates and percent solubilisation at different particle sizes and pulp densities Run

PS* (µm)

PD* (% w/v)

Iron solubilisation (%)

Solubilisation rate (g/L·h)

1

-38

2.5

86.2

0.819

2

75-38

2.5

64.0

0.740

3

106-75

2.5

82.9

0.712

4

150

2.5

69.9

0.776

5

-38

5

81.4

0.948

6

75-38

5

73.7

1.042

7

106-75

5

68.1

0.985

8

150

5

59.0

0.646

9

-38

10

45.6

1.009

10

75-38

10

40.9

1.089

11

106-75

10

29.3

0.794

12

150

10

31.5

0.420

13

-38

15

32.9

0.618

14

75-38

15

21.5

0.582

15

106-75

15

29.3

0.318

16

150

15

6.4

0.084

* PS: particle size; PD: pulp density

230


Total iron (g/L)

25

25

150-106 µm

20

20

15

15

10

10

5

5

0

0

10

20

30

40

50

0

106-75 µm

0

10

Time (d)

Total iron (g/L)

25

25

75-38 µm

20

15

15

10

10

5

5

0

10

20

30

Time (d)

30

40

50

40

50

Time (d)

20

0

20

40

50

0

-38 µm

0

10

20

30

Time (d)

Figure 1. Iron solubilisation kinetics from a pyritic gold concentrate by Sulfolobus metallicus in shake flasks at 68ºC initial pH 2.0 and agitation of 220 rpm. Pulp densities: ■ 2.5%; • 5.0%; ▲ 10%; ▼15%

231


3.2 Iron solubilisation rates Maximum iron solubilisation rates are given in Table 3. Multiple regression analysis of these results allowed the evaluation of the coefficients of equation (1): Q P = 0.9591 − 0.2032x1 − 0.1862x 2 − 0.1304x12 − 0.3039x 22 − 0.1305x1x 2

(2) All the coefficients of equation (2) have a significant effect on QP as their F values are much higher than the tabulated F at the 95% confidence limit (F5,10 = 3.3, p < 0.05) [13]. F values also allow the conclusion that the interaction between particle size and pulp density had the weakest effect (F = 9.24) and that pulp density (F = 81.24) showed a stronger effect on QP than particle size (F = 14.15). The response surface generated by equation (2) is showed in Figure 2 and the corresponding contour plot is depicted in Figure 3. It can be seen that a maximum value of QP occurs at a pulp density of 7.8% and particle size of 35 µm, a value very near the fraction 75-38 µm. The -38 µm exhibits a negative effect on the biooxidation rate, but this effect is not as strong as that observed by Nemati et al. [10], who found a complete inhibition of microbial action at particle sizes under 25 µm.

1,0 QP (g Fe/L·d)

0,8 0,6 0,4 0,2

lp Pu

2.5

( ity ns de

5.0 10

% v) w/

15 -38

75 -3 8

10 675

15 010 6

(µm size e l c i t Par

)

Figure 2. Response surface of the effect of particle size and pulp density on the rate of iron solubilisation from pyrite by Sulfolobus metallicus in shake flasks at 68ºC initial pH 2.0 and agitation of 220 rpm

232


Particle size (µm)

150-106

Productividad (g Fe/L·d) 0.99 -- 1.10 0.88 -- 0.99 0.77 -- 0.88 0.66 -- 0.77 0.55 -- 0.66 0.44 -- 0.55 0.33 -- 0.44 0.22 -- 0.33 0.11 -- 0.22 0.00 -- 0.11

106-75

75-38

-38

2.5

5.0

10

15

Pulp density (%w/v)

Figure 3. Contour plot of the effect of particle size and pulp density on the rate of iron solubilisation from pyrite by Sulfolobus metallicus in shake flasks at 68ºC initial pH 2.0 and agitation of 220 rpm 4.

CONCLUSIONS It is concluded that the operation variables particle size and pulp density have an effect on the rate of iron solubilisation from pyrite by Sulfolobus metallicus in shake flasks. Under the experimental conditions used in this work, the set of these variables that produce the highest rate is 35 µm particle size and 7.8% pulp density. It is also concluded that, in the range studied, the interaction between the variables is weak and that pulp density has a much stronger effect than particle size. ACKNOWLEDGMENTS This work was supported by the National Commission of Science and Technology through the FONDECYT projects 1000284 and 1020768. REFERENCES 1. A.D. Bailey and G.S. Hansford, Biotechnol. Bioeng., 42 (1993) 1164. 2. A.E. Torma, C.C. Walden and R.M.R. Brannion, Biotechnol. Bioeng., 12 (1970) 501. 3. A.E. Torma, C.C. Walden, D.W. Duncan and R.M.R. Brannion, Biotechnol. Bioeng., 14 (1972) 777. 4. M.T. Madigan, J.M.Martinko and J. Parker, Brock: Biología de los Microorganismos, 8th ed., p. 741, Prentice Hall, Madrid, 1998. 5. Α. Balows, K.H. Schleifer, H.G Truper, M. Dworkin and W. Harder (eds.), Prokaryotes, Vol. 1, p. 684, Springer Verlag, New York. 6. M.K. Toma, M.P. Ruklisha, J.J. Vanags, M.O. Zeltina, M.P. Leite, N.I. Galinina, U.E. Viesturs and R.P. Tengerdy, Biotechnol. Bioeng., 38 (1991) 552.

233


7. P. d’Hugues, D. Morin and S. Foucher, Biohydrometallurgy: Fundamentals, Technology and Sustainable Development (V.S.T. Ciminelli and O Garcia Jr., eds.), Part A, p. 75, Elsevier, Amsterdam, 2001. 8. R.P. Hackl, F.R. Wrigh and L.S. Gormly, IBS ’89 International Symposium Proceedings, p. 533, Jackson Hole, WY, 1989. 9. D. Howard and F.K. Crundwell, Biohydrometallurgy and the Environment Toward the 21th Century (R. Amils and A. Ballester, eds.), Part A, p. 209, Elsevier, Amsterdam, 1999. 10. M. Nemati, J. Lowenadler and S.T.L. Harrison, Appl. Microbiol. Biotechnol., 53 (2000) 173. 11. P.R. Norris, IBS ’89 International Symposium Proceedings, p. 3, Jackson Hole, WY, 1989. 12. L. Herrera, P. Ruiz, J.C. Aguillon and A. Fehrmann, J. Chem. Technol. Biot., 44 (1989) 171. 13. J. Lawson, J. Madrigal and J. Erjavec, Estrategias Experimentales para el Mejoramiento de la Calidad en la Industria, p. 181, Grupo Editora Iberoamericana, México, 1992. 14. N. Nemati and S.T.L. Harrison, Biohydrometallurgy and the Environment Toward the 21th Century (R. Amils and A. Ballester, eds.), Part A, p. 473, Elsevier, Amsterdam, 1999. 15. M. Gericke, A. Pinches and J.V. van Rooyen, Int. Miner. Process., 62 (2001) 243. 16. P. d’Hugues, S. Foucher, P. Gallé-Cavalloni and D. Morin, Int. Miner. Process. 66 (2002) 107. 17. F.C. Boogerd, P. Bos, J.G. Kuenen, J.J. Heijnen and R.G.J.M. van der Lans., Biotechnol. Bioeng. 35 (1990) 1111.

234


Enhancement of chalcopyrite bioleaching capacity of an extremely thermophilic culture by addition of ferrous sulphate A. Rubioa and F.J. García Frutosb a

Instituto Geológico y Minero de España (IGME), La Calera nº1. 28760 Tres Cantos, Madrid, Spain b Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas (CIEMAT), Av. Complutense 22, Edificio 20, 28040 Madrid, Spain Abstract One of the mainly problems of the bioleaching processes applied to copper concentrates and refractory ores are the low kinetics of the reactions, with high residence time that does not permit it to be an economic process. For this reason, current researches into this field are focused on how to increase this bioleaching rate. Apart from improving engineering design of bioreactors, the possibilities to increase bioleaching rates depend on the use of catalyst and isolation and adaptation of new microorganisms with high capacity to leach these ores. Many researches have investigated the possibility of using thermophilic microorganisms to improve metal-leaching rates instead to mesophilic microorganisms. In this sense, it was obtained a mixed natural thermophilic culture from a typical chalcopyritic copper concentrate of the Spanish Pyritic Belt. This culture has selectivity with respect to leaching chalcopyrite, when this is present together with other sulphides in ores and concentrates, and presents high copper leaching rates at high pulp density [1]. In this paper the study of enhancement of the bioleaching capacity of this culture on a chalcopyritic copper concentrate of the Spanish Pyritic Belt when ferrous sulphate is added is presented. The results obtained show that an initial addition of 1.8 g/L of iron as ferrous sulphate increases the copper bioleaching rate in all pulp densities studied. These results are according with other investigations that suggest the enhancement in copper leaching of chalcopyrite obtained by addition of ferrous sulphate [2]. From these results authors suggest an indirect in situ mechanism of bioleaching of chalcopyrite for this culture in which ferrous ion have a significant role. Keywords: extremely thermophilic culture, ferrous iron catalysis, chalcopyrite, bioleaching 1.

INTRODUCTION The resources of high ores in the world are becoming more and more scarce being necessary the processing of more complex ores. Conventional mineral processing on complex sulphide ores, carried out by differential flotation, often produces concentrates 235


not enough clean and difficult to commercialise. Therefore, plenty effort has been put into of hydrometallurgical process development for treatment of these ores, but the greatest number of proposed methods are complex and expensive [3]. Biohydrometallurgical processes appear as an alternative. These processes were first applied industrially to copper and uranium extractions using bio-assisted heap, dump and in-situ technologies, are today successfully used in extraction of gold from refractory sulphide-bearing ores and concentrates [4]. However, for other metal concentrates this technology isn’t still viable alternative to conventional pyrometallurgical extractions. In case of chalcopyritic concentrates there is a more complicated situation due to natural refractivity of chalcopyrite. The leaching of chalcopyrite is slow and incomplete in relation to other sulphides, and this is thought to be as a result of formation of a passivating layer. The use of new microorganisms isolated and adapted and of catalyst could improve the bioleaching rates. Thermophilic microorganisms had been used in bioleaching process because can be to increase leaching rates due to high temperatures, tolerant capacity and metabolic characteristic [5, 6]. Moreover, utilisation of these microorganisms that naturally thrive on ore samples and in its aqueous environments could be good options, since that, those probably have chalcopyrite specificity and much higher capacity for adaptation. In this sense, the authors have obtained a mixed natural thermophilic culture from a typical chalcopyritic copper concentrate of the Spanish Pyritic Belt. This culture has selectivity with respect to leaching chalcopyrite, when this is present together with other sulphides in ores and concentrates [7]. The use of catalytic ions like silver can increase bioleaching copper rate. But other ions can be accelerating this process too. Although ferrous iron is a reagent involved in the process, its role in bioleaching process is not still overall understanding. In acidic solutions, chalcopyrite is oxidised by ferric ions and dissolved oxygen to release copper ions. Ferrous ions are rapidly oxidised to ferric ions in presence of iron oxidising bacteria such as Acidithiobacillus ferrooxidans. It appears that role of ferrous iron in leaching is only as a source of ferric ions. The role of iron in bioleaching of sulphur ores is complex and depends of the interactions between these in its different oxidation states, bacteria and minerals particulate. However, there are reports, which suggest that ferrous iron contributes to copper extraction from chalcopyrite. There is a critical potential at which the leaching rate is very great and the rate suddenly decreases above this potential. As the suspension potential increases with increasing ferric to ferrous ratio, these results indicate that the leaching rate is faster with an optimum concentration of ferrous ions that without ferrous ions [2]. Further, if addition of catalytic ions to bioleaching can effectively increase the oxidation rate, the combination of both thermophilic microorganisms and catalyses using ferrous iron could aid to increase the slow kinetic problem that exhibit this process. This paper is presented in context of copper bioleaching, where it is assumed that the ferric ions and protons produced by microbial action, act as the leach agents in copper dissolution. The objective of this study is to establish the rate of chalcopyrite dissolution as a function of to added ferrous iron, determining the role of this in copper extraction with this thermophilic culture.

236


2.

METHODS

2.1 Ore sample A chalcopyritic copper concentrate obtained from conventional differential flotation was used to obtain the mixed thermophilic culture. This concentrate came from the Spanish Pyritic Belt and its composition is given in Table 1. The mineralogical composition of this sample shows chalcopyrite and pyrite as main mineralogical species and galena and sphalerite as secondary mineralogical phases. The particle size distribution presents a passing d80 of 20µm. Table 1. Chemical composition of copper concentrate Chemical analysis (wt %) Cu

Zn

Pb

Fe

S

23.37

2.58

2.52

32.63

38.00

2.2 Microbial inoculum A mixed thermophilic culture (MTC) of native microorganisms, isolated on this chalcopyritic concentrate was used as inoculum. The optimal condition of this culture is a 65°C temperature, with a pH of 1.30. For all tests done, the inoculum was obtained by the following way. The final pulp of the bioleaching test was filtered, obtaining the leach liquor and the solid residue. The solid was intensively stirred during two hours with a solution at pH 4. This pulp was filtered again and the final liquid obtained, which contains most the bacteria, which were attached to the solid residue, was filtered through a 0.22 µm Millipore filter where bacteria were retained. Finally, the bacteria were re-suspended in 50 mL of the leach liquor obtained in the first filtration in order to get the inoculum volume (5% v/v) for the next bioleaching test. The cultures were successfully adapted to higher pulp density of copper concentrate until obtaining high copper extractions in short residence times. 2.3 Bioleaching experiments Bioleaching experiments were carried out in 1 L glass cylindrical reactors provided with a cap with four holes to allow mechanical stirring (at 130 rpm), aeration (10-15 L/h) and sampling. These reactors were placed in a thermostatic bath to keep the temperature constant at 65°C. During the experiments the pH was kept at 1.3 by the addition of 10N H2SO4 when were necessary. This was made to avoid the precipitation of iron in form of jarosites, which damage the bioleaching process. Redox potential and pH were measured daily, while the levels of copper, zinc, and iron in solution were analysed daily or every two days, depending on the test. Water was added to the reactors in order to compensate for evaporation losses. Once bioleaching tests were finished, solids were removed by filtration, and chemically characterised as well as the leachate.

237


Sterile bioleaching tests were carried out with ore sterilised by autoclaving at 121°C, 30 minutes and 1atm of pressure, and adding a solution of 10% ethanol to the leaching media. To study the effect of the addition of ferrous iron on the bioleaching process, the test were carried out using ferrous sulphate solutions of varying concentration that were added initially to the leaching media. The ferrous iron concentration varied from 400 mg/L to 9000 mg/L depending of the tests. 2.4 Analysis Soluble species of copper, zinc, lead, total iron and minor elements were analysed by ICP [8] using a spectrophotometer ICAP-61 Thermo Jarrel Ash. The ferrous iron was analysed by a volumetric method by titration with potassium dichromate [9]. Copper, zinc, lead and iron content samples and leaching solid residues were analysed by XRF [10] using a spectrophotometer Philips PW-1404, and minor elements by ICP. Total sulphur was gravimetrically determined and elemental sulphur was analysed by toluene extraction in a Soxhlet apparatus. The pH was measured with a 704 pH-meter Metrohm. The redox potential (Eh) was measured with a platinum electrode with an Ag/AgCl reference electrode. Mineralogical composition was determined by XRD using a diffractometer PW-1700 Philips. 3.

RESULTS AND DISCUSSION Initially the were carried out at 1% pulp density and using concentrations of ferrous iron of 450, 900, 1800, 3600 and 4500 mg/L. Figure 1 shows copper extraction during the tests carried out with more representative ferrous iron concentration. Initial Fe2+addition increase copper overall leaching, obtaining with 1800 mg/L more than 96% of extraction in only 69 hours. With 3600 mg/L of ferrous iron addition the catalyst effect is maintained, but with higher ferrous iron concentrations, 4500 mg/L, copper extractions was diminished. At early hours, this catalysed effect is not appreciable, having lower copper extractions when ferrous iron was added respecting to the control (no addition of ferrous sulphate). 100

Cu extraction (%)

90 80 70 60 50

Fe2+(mg/L) control 900 1800 3600 4500

40 30 20 10 0 0

15

30

45

60

75

90

105

120

135

150

165

Time (Hours)

Figure 1. Copper extraction evolution with the different ferrous iron concentration added. (1% pulp density (w/v), MTC culture at pH 1.30 and 65°C) 238


Relationship between both total iron and ferrous iron in solution are shows in Figure 2. In all tests during the first hours iron in solution was in ferrous form, without apparent oxidation of Fe2+ added. This form of iron was increased in solution with all ferrous iron concentration added, but after 45 hours of assay the concentration was practically constant in solution. Only is observed a reduction in the solution ferrous iron concentration, with 900 and 1800 mg/L of ferrous iron added, when copper had been leached. 7

FeT concentration (g/L) Control

0.9

1.8

7 3.6

4.5 6

5

5

4

4

3

3

2

2

1

1

0

2+

concentration (g/L)

Control

6

0

Fe

0 15 30 45 60 75 90 105 120 135 150 165 0 Time (Hours)

0.9

1.8

3.6

4.5

15 30 45 60 75 90 105 120 135 150 165 Time (Hours)

Figure 2. Relationship between iron forms present in solution. (Left) Total iron. (Right) Ferrous iron. (1% pulp density (w/v), MTC culture at pH 1.30 and 65°C) In this sense, it seems that both ferrous iron, ferrous iron added and ferrous iron produced by chalcopyrite leaching reaction, only are being oxidised when the chalcopyrite was not available. Ferric iron in solution was not higher than 30% of total iron in all tests. Figure 3 shows the zinc bioleaching of the copper concentrate during the tests. This present a typical kinetic observed in all tests, in which the zinc did not leached whilst copper was bioleaching. Thus, in 1800 mg/L of ferrous iron addition test, can be clearly observed zinc bioleaching after total copper bioleached (72 hours). 100

Zn extraction (%)

90

control

Fe2+ (mg/L) 900 1800

3600

4500

80 70 60 50 40 30 20 10 0 0

15

30

45

60

75

90

105

120

135

150

165

Time (Hours)

Figure 3. Zinc bioleaching evolution with the ferrous iron concentration added. (1% pulp density (w/v) (MTC culture at pH 1.30 and 65°C) 239


pH evolution was similar in all tests, increasing in the first hours but was enclose in 1.30 adding sulphuric acid (10N) to avoid iron precipitation. XRD analyses of final solids of all test is showed in Table 2. The copper leaching produced elemental sulphur as principal species, increasing the presence of jarosites when ferrous addition was increased. Table 2. XRD diffraction of final solids obtained in bioleaching tests Addition of Fe2+ (mg/L) Control 900 1800 3600 4500

Main species

Secondary species

Minor species

Anglesite, Pyrite Pyrite, Elemental sulphur Pyrite, Elemental sulphur Pyrite, jarosite, Elemental sulphur Pyrite, Elemental sulphur, jarosite

Anglesite -

Elemental sulphur Jarosite Jarosite, anglesite -

At the same way as done with 1% of pulp density, bioleaching tests adding ferrous iron were carried out at 5% and 10% of pulp density using ferrous iron addition of 900, 1800, 2400, 3600 and 9000 mg/L. In Figure 4 can be observed that with 1800 mg/L of ferrous iron addition, also better copper extractions was obtained, with a copper extraction of 80% in only 165 hours. 100

Cu extraction (%)

Fe2+(mg/L)

90

Control

80

900

70

1800

60

2700

50

3600

40

9000

30 20 10 0 0

15

30

45

60

75

90

105

120

135

150

165

Time (Hours)

Figure 4. Copper extraction evolution with ferrous iron concentration added. (5% of pulp density (w/v), MTC culture at pH 1.30 and 65°C) At 10% of pulp density, copper extraction of 90% in 270 hours was obtained when 1800, 2700 and 3600 mg/L of ferrous iron were added (Fig. 5). The evolution for the rest of parameters studied, such as ferrous and ferric iron in solution and zinc bleaching, were similar as obtained at 1% of pulp density. In all bioleaching tests carried out, addition of ferrous iron increase copper extraction with this thermophilic culture. Addition of 1800 mg/L of ferrous iron seems to have the optimum catalysed effect in all pulp densities studied (1, 5 and 10% (w/v)). This optimum extraction could be related wit lesser presence of jarosites in solid residues. Most of the soluble iron, was present in ferrous form and the ferric concentration was negligible. The ferrous iron was produced by chalcopyrite leaching, and only when chalcopyrite is almost oxidised, ferrous iron oxidation is observed. Ferrous oxidation by this thermophilic culture is very low in liquid medium (data confirmed in early tests in 9K medium). 240

Bioleaching Applications 100

Cu Extraction (%) Fe2+ addition (mg/L) control 1800 2700 3600

90 80 70 60 50 40 30 20 10 0 0

50

100

150

200

250

300

Time (Hours)

Figure 5. Copper extraction evolution with ferrous iron concentration added (10% of pulp density (w/v), MTC culture at pH 1.30 and 65°C)

Trying to explain leaching mechanisms with this thermophilic culture, authors think that, in acidic solution, chalcopyrite can be principally oxidised by dissolved oxygen according to the following reaction: CuFeS2 + O2 + 4H+ = Cu2+ + Fe2+ + 2So+ 2H2O (1) In addition to this, the ferric ion that can be produced by oxidation of ferrous iron in acidic medium through Eq. (2) was rapidly utilised to leach the chalcopyrite by Eq. (3). 4Fe2++ 4H+ + O2 = 4Fe3+ + 2H2O (2) 3+ 2+ 2+ o CuFeS2 + 4Fe = Cu + 5Fe + 2S (3) taking place the chalcopyrite oxidation. Considering the final products obtained in our case, an indirect leaching mechanism could be postulated. However, ferric iron in solution was minimum and MCT possess very low capacity to oxidise ferrous iron in solution. Besides, according with rest potentials, sphalerite should be the most easily mineralogical species to leach, but this is not true, in our case. Because of that, the authors think according to postulated in (2), ferrous oxidation to ferric ions is produce onto chalcopyrite surface, where ferrous ions added are adsorbed on chalcopyrite surface. This ferrous iron is oxidised on surface by culture not in leaching media. This process that occur also with no addition of ferrous iron, have a catalyse effect when initial ferrous ions were added. These, provide of ferric ions that rapidly oxidises chalcopyrite producing more ferrous iron that can be adsorbed in other points of the chalcopyrite surface acting in the same way. When the chalcopyrite leaching is almost exhausted, ferric ion concentration is high passing to the solution and could leach sphalerite by ferric leaching, as can be observed in Figs 2 and 3. 4.

CONCLUSIONS A natural mixed thermophilic culture was obtained from a chalcopyritic copper concentrate with an ability to preferentially leach chalcopyrite in concentrates and with high leaching rates at high pulp densities. In our case, the enhancement of chalcopyrite bioleaching capacity of the mixed thermophilic culture by addition of ferrous ions confirm the indirect in situ leaching mechanism, postulated in our previous work [1], for this culture respecting the chalcopyrite. 241


The results obtained in this study confirm the main role of iron forms in the bioleaching processes of chalcopyrite. From these laboratory results this culture can be considered as a promising advance in the biohydrometallurgical treatment of chalcopyritic concentrates and its potential use on an industrial scale. REFERENCES

1. A. Rubio and F.J. García Frutos (2002) Mineral Engineering, 15, 689-694. 2. Hiroyosi, N et al. Hydrometallurgy. 47 (1997), 37-45. 3. J.L. Alvarez, Simposio sulfuros polimetálicos de la Faja Pirítica Ibérica, Huelva, (1996). 4. M. A. Jordan, S. McGuiness and C.V Philips, Minerals Engineering, 9, No.2 (1996) 169. 5. C. L Brierley, International conference and workshop application of biotechnology to the mineral Industry, Australian Mineral Foundation (1993) 2.1. 6. D. A. Clark and P.R. Norris, Minerals Engineering, 9, No.11 (1996) 1119. 7. A. Rubio. PhD Thesis (1998). Dpto. Biología Molecular. Universidad Autónoma de Madrid. 8. S. Del Barrio, Boletín Geologico y Minero (Special issue) (1992). 9. I.M Kolthoff, Análisis químico cuantitativo, Ed. Niger SRL (1979). 10. J.A Martín Rubí, IV Simposio Internacional de sulfuretos polimetálicos de la Faixa Piritosa Ibérica, C10 (1998) 1.

242


Evaluation of microbial leaching of uranium from Sierra Pintada ore. Preliminary studies in laboratory scale Patricia Silva Paulo*1, Diego Pivato1, Ana Vigliocco1, Julian López2, Alberto Castillo3 1

Unidad Aplicaciones Tecnológicas y Agropecuarias, Centro Atómico Ezeiza Unidad de Operación de Instalaciones Nucleares, Centro Atómico Ezeiza 3 Unidad de Proyectos Especiales y Suministros Nucleares Comisión Nacional de Energía Atómica, Av. del Libertador 8250 (1429) Buenos Aires, Argentina 2

Abstract The feasibility of bacterial leaching was studied in our laboratory with mineral from Sierra Pintada mine (Province of Mendoza, Argentina). It is a low U grade ore with 6% of carbonates and low quantity of pyrite. The experiments were performed in shake flasks and scaled up to static flood tanks. Shake cultured leaching studies were carried out with 0.5 inch crushed ore and a pulp density of 10%. Acidithiobacillus ferrooxidans (ATCC 33020) was grown in 9K medium and used as inoculum. Ferrous sulphate (FeSO4) was added as energy source. Percent of U leached and acid demand to maintain acidity values of pH=2 were measured during the experiment. Three conditions were tested: ore, ferrous sulphate and inoculum; ore and FeSO4 and a control with ore but without inoculum or FeSO4. After 90 days, the acid added in the flask with bacteria and FeSO4 was 20 times less than the control, and the U leached yielded 90.5%, that is 83% more than the control. Scale up experiments were performed in static flood tank system. Sixteen kg of 0.5 inch crushed ore were placed in a static tank reactor and flooded with 32 liters of 9K medium without FeSO4. Ferrous sulphide (SFe) was added as energy source. Gentle fine bubble aeration was accomplished by passing air through diffusers, and liquid recirculation was forced by a pump. At. ferrooxidans was grown in 9K medium and then used as inoculum. Three different conditions were tested: ore, inoculum and SFe; ore and inoculum, and a control without inoculum (ore only). Daily pH was measured and sulphuric acid added to maintain pH=2.0 was recorded. Weekly U leached was measured. The assay was carried on for five months. An initial 7 weeks lag phase was observed for all three conditions. This was associated to pH adjustments required for neutralization of acid demand due to the high carbonate presence in the ore. After five months, 72.8% of U was extracted in the At. ferrooxidans and SFe tank, with a total of 43.0 g of sulphuric acid added per g of U leached. An uranium extraction of 32.3% and 73.2 g acid/g U leached was observed in the ore and At. ferrooxidans (no energy source added) tank, while the control showed 3.6% of U leached and 73.4g acid added /g U leached. * Corresponding author. E-mail: [email protected]

243


The slowness of the process may be due to gypsum and jarosite precipitation on the mineral, impairing bacterial action and liquid recirculation as well. Both, shake flasks and static flood systems results, showed that the addition of an energy source is needed to obtain a reasonable rate of uranium leaching. A second static flood system experiment was performed, with an acid cure treatment before bacteria addition and counter current media flow. Different energy sources were also tested. The four conditions were: ore, inoculum and carbon steel scrap as energy source; a blank without inoculum; ore, inoculum and exogenous pyrite as energy source, and its corresponding blank without bacteria. After a month, 90.7% of U was extracted from the inoculated and carbon steel scrap tank with an acid consumption of 60.9 g/g U leached. In the tank with pyrite and inoculum an 89.6% of U leached was observed with 54.9g acid /g U. The second static flood experiment was notoriously effective in diminishing the leaching time, and avoiding precipitation that cause blockages in tubing. Both scrap and pyrite have shown to be suitable and economic energy sources. The experiments describe here are a promising first step in the evaluation of a possible pilot scale application. 1.

INTRODUCTION In Argentina the uranium industry is controlled by the federal government through the Atomic Energy National Commission (CNEA). In the fifties, the geological group found several uranium deposits around the country. In the sixties, the mining exploitation began to supply the fuel to the nuclear power plants Atucha and Embalse de Río Tercero. Atucha (RWU type) works with low enriched uranium and Embalse de Río Tercero (Candu type) only with natural uranium, consuming nearly 120 tnU/year. From 1979 to 1995 CNEA produced, by heap leaching, 1,000 tU in yellow cake form in Sierra Pintada plant (Mendoza province), to supply the fuel to the two nuclear power plants. During the nineties, due to the lower price in the spot market and overvaluation of the currency, the Sierra Pintada mine and the yellow cake production plant were shut down and Argentina begun to buy yellow cake in the spot market. Nowadays both activities have been started again and, the engineering group started to work on some other methods to reduce the cost and improve leaching and waste treatment. The mineral of this mine is sandstone with high concentration of calcium carbonate. This situation demands huge amounts of sulfuric acid to extract the uranium from the rock. The sulfuric acid, the mining operation and the manpower, are the most significant costs in yellow cake production. The main tasks implemented were: new design in the open pit, changing heap leaching by flooded leaching, and studies on bioleaching. In 1999, combined work began between two groups from CNEA, the Engineering and the Microbiology groups, to investigate bioleaching, as reported in this paper. Microbial leaching is a simple and effective technology used for metal extraction from low-grade ore. Metal recovery is based on the activity of chemolithotrophic bacteria, mainly Acidithiobacillus ferrooxidans, At. thiooxidans and Leptospirillum ferrooxidans (1-3). 244


Most of the uranium ores occur as a mixture of mineral containing uranium in either the tetravalent or the hexavalent state. Uranium is soluble only in its most oxidized hexavalent state (4). Tetravalent uranium can be oxidized to the hexavalent state by ferric iron, but oxidation occurs much more rapidly in the presence of the iron-oxidizing At. ferrooxidans(5). Bacterial leaching of uranium occurs via an indirect mechanism in which the bacteria oxidize the pyrite in the ore, generating an acidic ferric sulfate solution which carries out the chemical oxidation of the tetravalent uranium to the soluble hexavalent state. The studies reported here were initiated in order to determine the feasibility of microbial leaching of Sierra Pintada ore and whether bioleaching could contribute in the costs reduction by increasing the uranium recovery rate, reducing the acid consumption and / or diminishing the leaching time. Bioleaching of uranium from Sierra Pintada ore was carried out using Acidithiobacillus ferrooxidans at bench scale; exogenous ferrous compound were added for laboratory tests due to the low content of pyrite in the ore. 2.


2.1 Bacterial culture preparation A pure strain of Acidithiobacillus ferrooxidans (ATCC 33020) was used in this study. The bacteria were cultured in 2L Erlenmeyer flasks with 1L of basal medium containing 0.1g KCl, 3.0g/l (NH4)2SO4, 0.5g/l MgSO4.7H2O, 0.5g/l K2HPO4.3H2O, 44.22 g/l FeSO4.7H2O, 0.01 g CaNO3 ( 9K medium) (6). The pH of the medium was adjusted to 1.80 with sulfuric acid. The culture was kept at 30°C in a rotary shaker at 100 r.p.m. The culture in exponential growth was used directly as inoculum in each experiment. 2.2 Analytical methods The majority of the uranium analyses were performed by LivestockGroup laboratory. The uranium content of the ore and pulp residues was determined by the Laboratory Section of Complejo Minero Fabril San Rafael, Province of Mendoza. Uranium in leach liquor and pulp residues was determined spectrophotometrically by using dibenzoylmethane method (8). Daily pH was measured and the volume of sulfuric acid added to maintain pH 2.0 was recorded. The bacterial viability in both shake flasks and static tanks was checked by subculturing 1 ml of supernatants in 9ml of 9K medium twice a month. 2.3 Uranium ore The Sierra Pintada ore consisted of moderately well-sorted grains of quartz, feldspar and rock fragments cemented by calcite with minor clay replacement. This mineral is a sandstone with high quantity of carbonates. It is a low-grade ore, with an average of U3O8 content of 0.15%, 3.8% of carbonates and a low quantity of pyrite. The radioactive mineralization occurs mainly as uraninite, brannerite and coffinite. Analysis of the ore gave average values of 0.26% magnesium, 0.14% phosphorous, 2.6% calcium and 1.72% total iron (9).

245


2.4 Pyrite Pyrite from an outcrop near Sierra Pintada mine was characterized and used in leaching experiments. Sulfur and iron contents were 30% and 38% respectively. 2.5 Shake culture leaching Four Erlenmeyer flasks with two liters of medium containing 0.5g/l K2HPO4, 0.5g/l (NH4)2SO4, 0.5g/l MgSO4.7H2O at pH: 2.0 (TyK iron-free medium) were inoculated with 100 ml of a 96 hour culture of Acidithiobacillus ferrooxidans (7). Two hundred grams of Sierra Pintada ore (>1.0, 1.0 1.0 1.0 10) within the bottom anaerobic zone, in the form of MnCO3 and Mn(OH)2, i.e. these precipitates were formed as a result of the microbial sulphate reduction and the microbial and chemical alkalization of the waters. A portion of the iron was also removed as a result of the prior oxidation (mainly biological) of the ferrous ion to the ferric state. The Fe3+ was then precipitated as Fe(OH)3. Different heterotrophic bacteria able to oxidize Fe2+ at slightly acidic, neutral and slightly alkaline pH (from 4.5 to 7.5) were found in the wetland, including the isolates related to Metallogenium, Leptothrix and Arthrobacter, which were able to oxidize also the bivalent manganese. It must be noted that the chemical oxidation of Fe2+ by O2 was possible at the above-mentioned values but proceeded at lower rates than the bacterial oxidation. Chemolithotrophic bacteria able to oxidize Fe2+ at low pH values (less than 4) were found in the wetland but in very low numbers (Table 1). They were related to the species Acidithiobacillus ferrooxidans and Leptospirillum ferrooxidans. The growth and activity of the indigenous microflora during the cold winter months (December – February) of the year, when the temperature of the waters in the wetland was usually less than 5°C and often about 0°C, were markedly inhibited. However, the removal of pollutants was efficient (Table 3) although the residence times were increased. The character of the sediments precipitated during this time of the year in most cases was similar to that of the freshly precipitated sediments during the warmer months. The pollutants were present mainly as the easily soluble exchangeable and carbonate fractions (Table 4). The contents of pollutants in the dead plant biomass were much higher then these in the living plants during the warmer months. The contents of pollutants in the clays present in the wetland steadily increased during the winter months. Some clay specimens contained more than 1.5 g uranium per kg dry clay and the content of manganese exceeded 538

Bioremediation Environmental Applications

12 g per kg (the initial content of manganese in this clay was less than 5 g/kg). These data revealed that the role played by some sorbents, mainly by dead plant biomass and clays, in the removal of pollutants was essential during the cold winter months. However, the decrease of the concentrations of sulphates and the relatively high number of viable sulphate-reducing bacteria in the soil layer at the bottom of the wetland were indications that the microbial sulphate reduction proceeded, although at lower rates, in this anaerobic zone where the temperature was higher than that of the upper water layer. The results obtained during this study showed that the treatment of waters polluted with radioactive elements and heavy metals can be efficiently carried out by constructed wetlands with a proper size and design, with suitable and well-developed plant and microbial communities. ACKNOWLEDGEMENTS Parts of this work were financially supported by the European Commission under COPERNICUS project No ICA2-CT-2000-10010. REFERENCES

1. Cambridge, M., (1995). Use of passive systems for the treatment of mine outflows and seepage, Minerals Industry International, May: 35-42. 2. Groudev, S.N., Georgiev, P.S., Komnitsas, K., Spasova, I.I. and Angelov, A.T., (1999). Treatment of waters contaminated with radioactive elements and toxic heavy metals by a natural wetland, Paper presented at the International Conference on Wetlands Remediation, Salt Lake City, Utah, USA, November 16-17, (1999). 3. Groudev, S.N., Georgiev, P.S., Spasova, I.I, Angelov, A.T. and Mitrov, T., (2001a). Biotechnological treatment of mine waters contaminated with radioactive elements, In: New Developments in Mineral Processing, G. Onal, S. Atak, A. Guney, M.S. Celik and A.E. Yűce, Eds., pp. 571-574, Beril Ofset, Istanbul. 4. Groudev, S.N., Komnitsas, K., Spasova, I.I., Georgiev, P.S. and Paspaliaris, I., (2001b). Contaminated sediments in a natural wetland in a uranium deposit, Paper presented at the Sediments Conference 2001, Battelle Geneva Research Centre, Venice, October 10-12, (2001). 5. Groudev, S.N., Nicolova, M.V., Spasova, I.I., Komnitsas, K. and Paspaliaris, I., (2001c). Treatment of acid mine drainage from an uranium deposit by means of a natural wetland, Paper presented at the ISEB 2001 Meeting on Phythoremediation, Leipzig, Germany, May 15-17, (2001). 6. Groudeva, V.I., Ivanova, I.A., Groudev, S.N. and Uzunov, G.C., (1993). Enhanced oil recovery by stimulating the activity of the indigenous microflora of oil reservoirs. In: Biohydrometallurgical Technologies, vol. II, A.E, Torma, M.L. Apel and C.L. Brierley, Eds., pp. 349-356, TMS, The Minerals, Metals & Materials Society, Warrendale, PA. 7. Gusek, J.J. (1995). Passive-treatment of acid rock drainage: What is the potential bottom line?, Mining Engineering, March: 250-253. 8. Karavaiko, G.I., Rossi, G., Agate, A.D., Groudev, S.N. and Avakyan, Z.A., (1988). Biogeotechnology of Metals. Manual, GKNT International Projects, Moscow. 9. Tessier, A., Campbell, P.G.C. and Bisson, M., (1979), Sequential extraction procedure for speciation of particulate trace metals, Analytical Chemistry, 51(7): 844-851. 10. Younger, P.L., Ed., (1997). Mine Water Treatment Using Wetlands, University of Newcastle, Newcastle, UK. 539


Degradation of tetracyanonickelate (II) by Cryptococcus humicolus in biofilm reactors Hyouk Kee Kwona, Seung Han Woob, Joon Yong Sunga and Jong Moon Parkb a

LG Institute of Environment Safety & Health, 134 Shinchon-Dong, Seodaemoon-Gu, Seoul 120-749, Korea b Department of Chemical Engineering, Division of Molecular and Life Sciences, Pohang University of Science and Technology (POSTECH), San 31, Hyoja-dong, Nam-gu, Pohang 790-784, Korea Abstract A new yeast capable of degrading high concentrations of a metallocyanide, TCN (tetracyanonickelate (II), K2[Ni(CN)4]), was isolated from coke-plant wastewater and identified as Cryptococcus humicolus. The strain was able to degrade cyanide compounds in both aerobic and anaerobic conditions. The highest degradation activity was achieved at pH 7.5 in aerobic condition and at pH 5.0 in anaerobic condition. The feasibility of biological detoxification of TCN-containing synthetic wastewater by the strain was investigated using two fixed-bed biofilm reactors operated in the aerobic and anaerobic conditions. At 24 h of hydraulic retention time, influent TCN of 50 mg CN dm-3 was completely degraded in an aerobic reactor and about 90% of the TCN was degraded in an anaerobic reactor. When the hydraulic retention time decreased to 9 h, the removal efficiency decreased to 83% and 61% for the aerobic and anaerobic reactors, respectively. Keywords: biodegradation, biofilm, bioreactor, cyanide, metallocyanide 1.

INTRODUCTION Cyanide compounds are mainly generated from various chemical industries, including metal extraction, electroplating, coal gasification, ore leaching, and production of synthetic fibers [1]. These compounds are of major environmental concern due to their toxicity to living organisms [2]. In metal-bearing wastewater, most of cyanides are present in the form of metal (such as Fe, Cu, Ni, and Zn) complexes rather than as free cyanide. Such metal-cyano complexes are highly stable and more resistant to biological attack compared with free cyanide. Biodegradation of cyanides has been studied with a number of microorganisms, including bacteria [3-5] and fungi [6-10]. Most of the microorganisms have been investigated in aerobic conditions to apply for conventional activated sludge processes. On the contrary, anaerobic treatment of cyanides has not been intensively studied except for the cases of some methanogenic mixed cultures [11, 12]. Anaerobic wastewater treatment could be more advantageous due to its low energy requirement compared with aerobic treatment. 541


In our previous report, for the first time, a new yeast strain (Cryptococcus humicolus) capable of degrading high concentrations of metallocyanide compounds was isolated from wastewater [13]. The strain was able to utilize free cyanide as a single nitrogen source by converting CN to ammonia and carbon dioxide in aerobic conditions. It was also elucidated that molecular oxygen from O2 was not directly incorporated during the conversion of the free cyanide. The ability of facultative degradation is greatly beneficial for practical applications since gradients of oxygen concentrations in microenvironment can seriously inhibit the biodegradability in the system requiring the strict aerobic or anaerobic condition. The aim of the present work was to investigate the treatability of metallocyanides in both aerobic and anaerobic conditions using the strain Cryptococcus humicolus. A ceramic "honey-comb" structure was used as support matrix for the biofilm in fixed-bed reactors. 2.


2.1 Microorganism and growth conditions Cryptococcus humicolus was grown in an enrichment medium containing 3 g glucose, 5 g KH2PO4, 4.2 g K2HPO4.3H2O, 0.5 g MgSO4, 0.04 g FeSO4.7H2O, 0.0015 g MnSO4.H2O per liter. The pH was adjusted to 7.5. The strain was maintained on mineral agar medium containing TCN of 260 mg CN dm-3 by subculturing at 7-day interval. Colonies on the agar medium were cultivated in a 500 ml Erlenmeyer flask containing 0.1 dm3 mineral medium and TCN of 260 mg CN dm-3 for 3 days in a rotary shaker at 250 rpm. Ten ml of culture was harvested by centrifugation (6,000 x g) and washed twice with mineral medium. This was resuspended in 0.01 dm3 of mineral medium and used as inoculum for biodegradation experiments. 2.2 Effect of pH The effect of initial pH ranging from 4.0 to 8.5 on TCN degradation was examined in flask cultures under aerobic and anaerobic conditions. Initial concentration of TCN was 260 mg CN dm-3. Anaerobic condition was accomplished by bubbling high purity N2 gas. During culture period, pH and TCN concentration were measured with time. 2.3 Fixed-bed biofilm reactor The schematic diagrams of honey-comb typed bioreactors and their dimensions are shown in Fig. 1 and Table 1, respectively. The influent was introduced into the bottom of the reactors by a peristaltic pump. For the aerobic bioreactor, air was sparged at the bottom. A ceramic structure (chemical composition: 2MgO.2Al2O3.5SiO2) was used as a support matrix for biofilm formation. One unit of structure contains 625 pixels and each pixel is a regular hexahedron void with a side length of 0.5 cm. 2.4 Reactor operation Inoculum was prepared by cultivation of C. humicolus in a fermenter using 4 dm3 enrichment medium. Each reactor was filled with the microbial inoculum prepared in the fermenter and synthetic wastewater. The synthetic wastewater contained 1 g of glucose and TCN (52 mg CN dm-3 tap water). The reactors were operated in a continuous mode at HRT (hydraulic retention time) of 24 h during 3 months before reaching steady state. For the aerobic reactor, the aeration rate was controlled at 3 dm3.min-1. The TCN biodegradation was measured at various HRT values and aeration rates. 542


Figure 1. Schematic diagrams of fixed-bed biofilm reactors Table 1. Summary of the dimensions of the fixed-bed biofilm reactor Items

Values

Inner diameter of the reactor Total volume of the reactor Working volume Height of the unit matrix Diameter of the unit matrix Void fraction of the matrix Hydrous water content Number of the matrices per reactor

15 cm 8.3 dm3 6.2 dm3 10 cm 15 cm 0.66 0.16 cm3 water / g-matrix 4

2.5 Liquid circulation Liquid circulation time (tc, min) was measured using hydrogen ion as a tracer at various aeration rates ranging from 3 to 8 dm3.min-1. Using the liquid circulation time and the liquid volume (VL, dm3), the liquid circulation rate (Qc, dm3.min-1) can be calculated as follows: V − (1 − ε )VB V (1) Qc = L = w tc tc

where ε is the void fraction of the bed, Vw (dm3) is the working volume, and VB (dm3) is the bed volume. Pressure drop between the top and the bottom of the bed was also measured by a manometer at various aeration rates. 2.6 Analytical methods The samples were filtered through a 0.2 µm pore size filter before analyses. TCN was analyzed by the silver nitrate titration method after distillation according to APHA standard methods [14]. Ammonia was determined by Nesslerization method [14]. Glucose concentration was determined using o-toluidine reagent kit (Sigma, USA). Dissolved 543


oxygen and pH value were measured by electrodes (Ingold, USA). Cell growth was determined by measuring the optical density of culture broth samples at 600 nm (OD600). Microbial cell dry weight was determined by the value obtained after centrifugation (6,000 x g for 10 min) and drying at 80°C for 24 h. 3.


3.1 Effect of pH on TCN degradation Figs. 2 and 3 show the effect of the initial pH on the degradation of TCN under both aerobic and anaerobic conditions. The optimal pH values for TCN degradation in aerobic and anaerobic conditions were quite different. While the highest degradation activity was achieved at pH 7.5 in the aerobic condition, pH 5.0 was the optimum in the case of anaerobic condition. In the aerobic condition, the TCN degradation rate did not show any significant difference in the range of 6.5 to 7.5. However, in the anaerobic condition, strong inhibition was observed above pH 7.0 as shown in Fig. 3. For all the tests, pH values did not change significantly with time (data not shown). As expected, the biodegradation rate was generally higher in aerobic condition than in anaerobic one. However, below pH 5.0, the aerobic growth of the strain C. humicolus was significantly inhibited (data not shown), while high degradation activity was observed in anaerobic conditions. It suggests that anaerobic treatment of cyanides using this strain could be beneficial for acidic wastewater compared to aerobic treatment. 300 300

200 150

TCN (mg dm-3)

-3

TCN (mg dm )

250

pH 5.5 pH 6.0 pH 6.5 pH 7.0 pH 7.5 pH 8.5

250

100

200 150 pH 4.0 pH 4.5 pH 5.0 pH 6.0 pH 6.5 pH 7.0

100 50

50

0

0 0

20

40

60

80

Time (h)

Figure 2. Effect of initial pH on TCN degradation under aerobic condition

0

20

40

60

80

Time (h)

Figure 3. Effect of initial pH on TCN degradation under anaerobic condition

3.2 Liquid circulation The aerobic reactor was operated as an air-lift with a riser part formed by aeration in the center of the reactor. This brought about repeated sinusoidal detection of pH in the trace test for measurement of liquid circulation time. The liquid circulation time in the aerobic reactor was measured at various aeration rates as shown in Fig. 4. The liquid circulation rate calculated by equation (1) was proportional to the aeration rate. Initially, no pressure drop was detectable. After 3-months operation, the pressure drop started to build up around 2.5 cm H2O due to the biofilm formation within the bed. The overgrown biomass was removed from the reactor when the pressure drop increased over 5.0 cm H2O. 544


Figure 4. Liquid circulation characteristics of the bioreactor at various aeration rates 3.3 Effect of HRT on TCN degradation Figs. 5 and 7 show the TCN degradation by C. humicolus in aerobic and anaerobic reactors at various HRTs. In the aerobic operation, TCN was completely degraded at 24 h of HRT. Decrease of HRT resulted in the gradual decrease of TCN removal efficiency. For example, at 9 h of HRT, the removal efficiency dropped to 83.2%. Ammonia generated from TCN degradation remained in the effluent after being used as a nitrogen source for microorganisms. The ammonia production rate was almost constant at 0.4 mg NH3-N dm-3 h-1 regardless of HRT. The pH increased from 6.7 to 7.4 with increasing HRT from 9 h to 24 h, respectively. This is possibly due to the higher production of ammonia by TCN degradation at higher HRT. The effect of aeration rate on TCN degradation in the aerobic operation was investigated at various HRTs (9, 11, 14, 18 and 24 h). The efficiency of TCN degradation increased proportionally with increasing aeration rate as expected (Fig. 6). However, above 7 dm3.min-1, the degradation rate was not further enhanced. Moreover, TCN was not completely degraded below 18 h of HRT despite high aeration rate.

Figure 5. Effect of HRT on TCN degradation in aerobic operation at an aeration rate of 7 dm3.min-1

545


Figure 6. Effect of aeration rate on effluent TCN at various HRTs in the aerobic reactor

Figure 7. Effect of HRT on TCN degradation in anaerobic operation

In the anaerobic operation, the removal efficiency of TCN was generally lower than in the aerobic operation. In addition, the decrease of removal efficiency by decreasing HRT in anaerobic condition was greater than under aerobic conditions. The removal efficiency dropped from 90% to 61.1% with decreasing HRT from 24 h to 9 h, respectively. In the anaerobic operation, ammonia was constantly generated at a rate of about 0.1 mg NH3-N dm-3.h-1 regardless of HRT, which was lower than that of the aerobic operation. Unlike the aerobic operation, the pH value decreased slightly from 5 to 4.4 with increasing HRT from 9 h to 24 h, respectively. This is thought to be due to the acid formation by fermentation of glucose in the anaerobic conditions. 3.4 TCN profile with bed depth Fig. 8 shows the TCN profile along the bed depth for the aerobic and the anaerobic operation. The dissolved oxygen values were zero at all the depths in the anaerobic reactor, and 7.6, 7.5, 7.4 and 7.2 at D1, D2, D3 and D4 in the aerobic reactor, respectively. In the anaerobic operations, TCN concentration gradually decreased as wastewater passed through the bed. However, in the aerobic operation, most of influent TCN was degraded within the first unit matrix and thereafter TCN was maintained almost constant. This is likely due to rapid liquid mixing in the aerobic reactor by the rigorous aeration. 546


Figure 8. Concentration profile of TCN along the depth of bed with HRT of 24 h in the anaerobic or aerobic reactor at air flow rate of 7 dm3.min-1 4.

CONCLUSIONS A new yeast strain, Cryptococcus humicolus, showed facultative property on TCN (tetracyanonickelate (II)) degradation, which is able to grow in both aerobic and anaerobic conditions. This facultative ability can be beneficial for practical applications since oxygen limitation in aerobic operation or the presence of oxygen in anaerobic operation is not detrimental to cell activity. Furthermore, in the acidic condition, TCN degradation rate in the anaerobic condition was comparable to that in the aerobic condition. Using fixedbed biofilm reactors, at 24 h of hydraulic retention time, influent TCN of 50 mg CN dm-3 was completely degraded in the aerobic operation and about 90% of the TCN was degraded in the anaerobic operation. The removal efficiency was not seriously decreased even when the hydraulic retention time decreased to 9 h in both aerobic and anaerobic operation. If the performance of anaerobic reactor is further optimized, it can be a competitive technology for the treatment of cyanide compounds in wastewater. REFERENCES

1. S.A. Raybuck, Biodegradation, 3 (1992) 3. 2. C.J. Knowles and A.W. Bunch, Adv. Microb. Physiol., 27 (1986) 73. 3. G. Rollinson, R. Jones, M.P. Meadows, R.E. Harris and C.J. Knowles, FEMS Microbiol. Lett., 40 (1987) 199. 4. J. Silva-Avalos, M.G. Richmond, O. Nagappan and D.A. Kunz, Appl. Environ. Microbiol., 56 (1990) 3664. 5. R. Harris and C.J. Knowles, J. Gen. Microbiol., 129 (1983)1005. 6. M.J. Cluness, P.D. Turner, E. Clements, D.T. Brown and C. O’Reilly, J. Gen. Microbiol., 139 (1993) 1807. 7. P. Wang, D.E. Matthews and H.D. VanEtten, Arch. Biochem. Biophys., 287 (1992) 569. 8. Dumestre, T. Chone, J. Portal, D. Gerard and J. Berthelin, Appl. Environ. Microbiol., 63 (1997) 2729. 9. M. Barclay, V. Tett and C.J. Knowles, Enzyme. Microb. Tech., 23 (1998) 321. 10. H. Yanase, A. Sakamoto, K. Okamoto, K. Kita and S. Sato, Appl. Microbiol. Biotechnol., 53 (2000) 328. 547


11. P.M. Fedorak and S.E. Hrudey, Wat. Sci. Tech., 21 (1987) 67. 12. R.D. Fallon, D.A. Cooper and M. Henson, Appl. Environ. Microbiol., 57 (1991) 1656. 13. H.K. Kwon, S.H. Woo and J.M. Park, FEMS Microbiol. Lett., 214 (2002) 211. 14. A.E. Greenberg, R.R. Trussell and L.S. Clesceri (eds.), Standard Methods for the Examination of Water and Wastewater, American Public Health Association, Washington, 1995.

548


Development of a bio-process using sulfate-reducing bacteria to remove metals from surface treatment effluents F. Battaglia-Bruneta, S. Fouchera, A. Denamura, S. Chevardb, D. Morina, I. Ignatiadisa a

BRGM, Environment and Process Division, Biotechnology Unit, Av. Claude Guillemin, 45060 Orléans Cedex 02, France b SETS S. A., ZI Les Vigneaux, B.P. 3, 36210 Chabris, France

Abstract The surface treatment activities are essential and associated to any industrial sector. However, they produce large quantities of effluents. The majority of these pollutants are currently removed through physical and chemical treatments, and metals are generally precipitated as hydroxides. The present project proposes to replace some steps of the classical treatment process by biological reactions. Sulfate-reducing bacteria (SRB) would perform the bio-reducing operations. The objective of this work was to reduce sludge volume and toxicity, and if possible to recover some metals selectively. Three processes involving sulfate-reducing bacteria were considered: (1) the sulfate produced by oxidation of sodium bisulfite, may be converted into H2S in a bioreactor, in which the acidic/alkaline metals-containing effluents would be directly injected; (2) the sulfate produced by oxidation of bisulfite may be injected in a SRB-bioreactor producing H2S that would be transferred in a separate precipitation reactor, in order to selectively recover metals; (3) the Cr(VI)-containing effluents may be directly treated in the bio-reactor, that would also produce some H2S to precipitate the other metals in a separate reactor. This last configuration was tested in a 20-L column bioreactor, fed in continuous mode with a real effluent, whose Cr(VI) concentration was adjusted at 30 mg/l. 1.

INTRODUCTION Surface treatment is a widespread industrial activity, necessary to many important economical sectors such as transports, building-trade. The expenses of environmental protection in surface treatment plants represent 25% of the investment and 10-15% of the operating costs [1]. In spite of significant improvement of the practices, the contamination of the environment due to surface treatment industries reaches 30% of the global industrial toxic discharge [1]. Metals-containing effluents are generally treated by physical and chemical processes generating voluminous amounts of metal-hydroxide sludge. The charge for sludge removal may reach 30% of the effluent treatment operating costs. Consequently, any improvement of the flowsheet that could reduce the volume and toxicity of the solid residues should be considered. The present study proposes to introduce a biological step in the process that could replace a fraction of the hydroxide sludge by metal sulfide precipitates, less voluminous and easier to dry. 549


2.

DATA ON THE INDUSTRIAL PLANT

2.1 Flowsheet of the effluent treatment plant at S.E.T.S (Fig. 1) The real industrial plant chosen for the present study is equipped with a classical and common effluent treatment plant. The three main types of effluents are cyanide rinse solution (4 m3.h-1), acid/alkaline rinse solution (10 m3.h-1) and chromate rinse solution (4 m3.h-1). Cyanides are oxidized with hypochloride (ClO-) in alkaline conditions. Chromate is reduced with sodium bisulfite (NaHSO3) in acidic conditions. The effluents from cyanide oxidation and chromate reduction are conducted to the neutralization tank, which also receives acid/alkaline bathes. The precipitation of metal hydroxides is performed in this reactor. cyanide rinse

acid/alkaline rinse

Cr(VI) rinse

lime HClO-

cyanide oxidation

neutralisation

Cr(VI) reduction

NaHSO3 H2SO4

NaOH flocculation decantation

solids

filter press

lime or NaOH

sand filter

treated effluent

Figure 1. Simplified flowsheet of S.E.T.S. effluent treatment plant 2.2 Composition of the effluents Samples of S.E.T.S. effluents were analyzed. Examples of effluent compositions are given in Table 1. The information given by effluent analyses was used to choose representative data. The further design of new flowsheets and comparison with the conventional process are based on these data (Table 2). 3.

MATERIAL AND METHODS

3.1 Pilot plant A glass column, 0.10 m inner diameter and 2.75 m total height was filled with pozzolana (8-10 mm) kindly supplied by Carrière de la Denise, le Puy, France. The temperature was kept at 32°C by water circulation in an outer double-jacket. The column was up-flow fed in continuous mode. Gas flow-rates were 20 l.h-1 for H2 and 1.2 to 1.6 l.h1 for CO2. The synthetic medium was made up of concentrated Industrial Urea Medium and concentrated Cr(VI) solution separately pumped into the column. The bioreactor was connected to a glass precipitation column, 0.04 m inner diameter and 1.75 m total height (Fig. 2). Metals-containing acid/alkaline effluent and alkaline effluent from the cyanide destruction tank fed the precipitation column.

550


Table 1. Composition of S.E.T.S. effluents (n.d: not determined) Compounds [mg.l-1]

Cr(VI) rinse effluent

Acid/alkaline rinse effluent

Cr(VI) reduction tank

Cyanides rinse effluent

Ca2+

50

49

56

6

9

8

9

6

32

153

52

251

15

60

8

5

1

3

0.4

8

Mg

2+

Na+ +

K

NH4

+

-

54

570

55

86

-

39

74

20

7

3-

0

20

1

0

Cl

NO3 PO4 -

F

30

3

4

0.4

2-

6

26

480

21

-

n.d.

n.d.

n.d.

42

3+

2

13.6

0.8

0.25

2+

0

0.06

0.08

1.2

Zn2+

SO4

CN Fe

Cu

4

34

21

5.5

Ni

2+

0.04

1.75

0.08

12.25

Al

3+

3

2.5

1.4

0

10

0

0.02

n.d.

Cr(VI)

Table 2. Data chosen for process design Effluents Cr(VI) rinse effluents Cr(VI) rinse effluents Cr(VI) reduction tank Acid-alkaline effluents Acid-alkaline effluents Acid-alkaline effluents Acid-alkaline effluents

Flow-rate [m3.h-1] 4 4 4 10 10 10 10

Compound Cr(VI) SO42SO42Zn2+ Fe3+ Ni2+ Cu2+

Concentration [mg.l-1] 25 85 925 60 50 2.5 0.1

3.2 Experimental programme The experimental programme was the following: (phase 1) the column was inoculated with a D. norvegicum-containing bacterial population (2 litres at 2x108 bact.ml-1), and the biofilm was allowed to develop by feeding the reactor with a 5 g.l-1 SO4-containing Industrial Urea Medium (IUM); (phase 2) the sulfate concentration in the feeding was decreased to 0.5 g.l-1; (phase 3) Cr(VI) concentration in the feed was increased from 0 to 30 mg.l-1; (phase 4) the residence time was reduced to 7 h by increasing the feed flow-rate; (phase 5) the column was fed with a real Cr(VI) containing electroplating effluent. Nutrients were added to the effluent, and a concentrated Cr(VI) solution was separately pumped in order to adjust the Cr(VI) concentration to 30 mg.l-1. The composition of the Industrial Medium with Urea was the same as in [4]. The liquid effluents from the bioreactor were conducted into the precipitation column, also fed with the outlet-gas from the biological step. A metals-containing acid/alkaline real effluent was injected into this 551


system. An alkaline effluent is used to increase the pH and favour sulfide precipitation (Fig. 2). Sulfide sludge settled down at the bottom of the column, and the gas phase was bubbled into a zinc sulfate solution (50 g.l-1) in order to remove H2S through ZnS precipitation. H2

CO2

gas

liquid

gas

effluent Cr(VI)

nutrients and sulfate

acid/alk. alkaline effluent

sludge

effluent

Figure 2. Schematic representation of the pilot plant 3.3 Analysis The bioreactor was equipped with pH and Eh (redox potential of Pt/Ag-AgCl) probes. Samples were daily taken in the outlet stream in order to analyse dissolved sulphide (potentiometric method), sulfate (kit MERCK spectroquant® 1.1458.0001) and Cr(VI) (kit MERCK spectroquant® 1.14758.0001). 4.

RESULTS

4.1 Design of potential process flowsheets including a biological step Three process flowsheets including a biological step are proposed. In these alternative flowsheets, metals (except chromium) are precipitated as sulphides. In the first option, the chemical neutralisation tank is replaced by a bioreactor (Fig. 3). The effluent from the Cr(VI)-reduction tank contains sulfate that will be converted into HS- by sulfate-reducing bacteria. Metals from other effluents will be precipitated into the bioreactor with biologically produced HS-. The second option proposes to perform the biological production of HS- and the precipitation of metal sulphides in separate reactors (Fig. 4). In this process, the chemical Cr(VI) reduction is maintained, and this chemical step still provides sulfate for the biological step. Some alkaline effluent from the cyanidesoxidation tank may be used to increase the pH in the bioreactor. In the third option (Fig. 5), the chemical Cr(VI)-reduction tank is replaced by a bioreactor in which SO42- and 552


Cr(VI) are biologically reduced. Some sulfate has to be added into the bioreactor, because chromate rinse effluents cannot provide enough SO42-. This last option was tested at the laboratory-pilot scale. cyanide rinse

acid/alkaline rinse

Cr(VI) rinse

lime cyanide oxidation

HClO-

Bio-reactor HS- production and precipitation of metals

NaHSO3

Cr(VI) reduction

H2SO4

NaOH Nutrients, H2

solids

flocculation decantation

filter press

lime or NaOH

sand filter

treated effluent

Figure 3. Alternative flowsheet, option 1 – neutralisation tank replaced by a bioreactor

Nutrients, H2

cyanide rinse

Bio-reactor

lime or NaOH

HS- production

Cr(VI) rinse

Precipitation of metals as sulfides

Cr(VI) reduction

lime HClONaOH

solids

cyanide oxidation acid/alkaline rinse

filter press

flocculation decantation

NaHSO3 H2SO4

lime or NaOH

sand filter

treated effluent

Figure 4. Alternative flowsheet, option 2 – neutralisation tank replaced by a bioreactor and a precipitation reactor

553


cyanide rinse

acid/alkaline rinse

Cr(VI) rinse

lime -

HClO

cyanide oxidation

Precipitation of metals as sulfides

Bioreactor Cr(VI) reduction HS- production

Nutrients, H2 Na2SO4

NaOH flocculation decantation

solids

filter press

lime or NaOH

sand filter

treated effluent

Figure 5. Alternative flowsheet, option 3 – Cr(VI) chemical treatment and neutralisation tanks replaced by a bioreactor and a precipitation reactor 4.2 Biological treatment of a Cr(VI)-containing effluent With 500 mg.l-1 SO42- in the feeding medium, Cr(VI) concentration was progressively increased to 30 mg.l-1 (Fig. 6, phase III), and the residence time was reduced to 7 h while maintaining the efficiency of the bioreactor: in these conditions, Cr(VI) concentration in the outlet was lower than 0.1 mg.l-1 (Fig. 6, phase IV), and sulfate was nearly entirely reduced. The accumulation of blue-green colored Cr(III) hydroxyde on the pozzolana was visible. When the synthetic feeding medium was replaced by the real Cr(VI) rinse effluent from S.E.T.S. (Fig. 6, phase V), the system was not disturbed as long as the residence time was 7 h or higher. This result was positive, and proved that the bacteria were able to resist to pollutants such as nitrate or cyanide, that were present in the effluent at low concentrations. When the residence time was decreased from 7 to 5.5 h, sulfate and Cr(VI) reduction processes were inhibited. The Cr(VI) concentration in the outlet reached 11 mg.l-1. A batch phase without chromate was necessary to restore the bacterial activity.

Figure 6. Evolution of Cr(VI) concentration in the feed and outlet of the bioreactor, and residence time vs. operating time 554


In the outlet stream, pH was in the range of 8 to 9, and redox potential fluctuated between –400 and –500 mV (ref. Ag/AgCl) (Fig. 7). The inhibition of the biofilm, when the flow-rate was too much high, was accompanied by a strong increase in redox potential, from –400 to –100 mV. This sensitive parameter will be useful as an indicator for the operating status of the bioreactor.

Figure 7. Evolution of pH and redox potential of the outlet liquid from the bioreactor vs. operating time

Elimination of metals from the acid/alkaline rinse was performed in the precipitation column. The liquid effluent from the bioreactor brought dissolved sulfide in the mixture. However, the bubbling bioreactor outlet gas, which contains H2S, greatly improved the precipitation efficiency. Only Zn2+ and Cu2+ were present in relatively high concentrations in the real effluent. These two metals were entirely precipitated by applying a 20-min residence time in the precipitation column. However, a residence time of 60 min was applied, because the effluent may contain metals with slower precipitation kinetics than Cu2+ and Zn2+, and it was thought to be more realistic for the cost estimate. 4.3 Preliminary technical and economical evaluation A preliminary evaluation of the process was performed based on the following configuration: Cr(VI) is reduced to Cr(III) hydroxide in a column fixed-film bioreactor filled with pozzolana, working at 30°C and pH in the range 7-8.5. The chromium hydroxides are recovered by settling at the bottom of the column. Gas and liquid from the bioreactor are mixed in a reactor with acid/alkaline effluents to precipitate metal sulfides, which are recovered by settling and filtration. The gas phase from the precipitation reactor is re-injected into the bioreactor. Consumption and production fluxes for this process are given in Table 3. For the evaluation of the classical treatment (bisulfite for Cr reduction and lime for metals precipitation), data from Rigaud and Girards [1] were used. The investment and operating costs of this process will largely depend on the way to obtain H2 (electron source for the anaerobic bacteria) and CO2 (for pH regulation). In-situ coal reforming would correspond to the highest investment and lowest operating cost. Buying the two gases separately as stock tanks would correspond to the lowest investment and highest operating cost. In this last configuration, the investment cost was estimated 555


closely to 281,000 €, and the operating cost reaches 114,000 €.year-1. As a comparison, the operating cost of the classical physico-chemical process is close to 75,000 €.year-1. The cost of waste solids elimination would be lower with the "biological" process (9,600 €.year-1), than with the classical process (17,300 €.year-1). Table 3. Process material balance Products urea (46% N) H2SO4 KOH MgSO4 Na acetate DAP pozzolana H2 CO2 Products liquids wet sulfides dry sulfides Wet Cr hydroxide Dry Cr hydroxide

5.

Consumption (tonnes.year-1) 1.5 10 12.6 3.4 17.5 1.6 37 1.6 28.3 Production 3. -1 m year tonnes.year-1 142,252 5.4 15.8 3.5 13.9 0.6 3.1 0.36 1.9

DISCUSSION AND CONCLUSION The study of a real effluent treatment plant resulted in proposing several process flowsheets including a biological step for the reduction and/or precipitation of metals using sulfate-reducing bacteria. These micro-organisms are able to reduce Cr(VI) into Cr(III) by two combined mechanisms: direct enzymatic reduction [2, 3, 4], and indirect chemical reduction by hydrogen sulfide [5]. The bacterial population was able to maintain its ability to reduce sulfate and chromate in the pilot plant fed with a real surface-treatment chromic effluent. The minimum residence time that could be applied was 7 h for a feed Cr(VI) concentration of 30 mg.l-1. In 2-litre column bioreactors, the sulfate-reducing biofilm was able to withstand a feeding Cr(VI) concentration of 100 mg.l-1 when the residence time was 19 h [5]. The hydrogen sulfide produced in the bioreactor was used to treat the other metals-containing effluents of the surface-treatment plant. A preliminary economic evaluation revealed that a new process including a biological step would be more expensive than the classical physical and chemical process, particularly in terms of operating costs. However, further optimisation of the biological step may lead to a cheaper configuration. The nutrients consumption can probably be reduced. Devices to produce H2 and CO2 at lower cost may be developed. Lowering the pH of the feed solutions could reduce carbon dioxide consumption. The biological step allows to reduce the amount of solid wastes. The cost of waste management, particularly for this type of solids containing toxic metals, will probably increase, and surface treatment plants will be encouraged to reduce their sludge production. In this context, the possibility to precipitate metals as sulfides rather than as hydroxides could become interesting. 556


AKNOWLEDGEMENTS This research was supported by ADEME (Contract No. 0002016), and by the Research Division of BRGM (Contribution N°01214). REFERENCES

1. J. Rigaud and L.M. Girard (eds.) Traitements de Surfaces – Epuration des eaux. Publication de l’Agence de l’Eau Rhône-Méditerranée-Corse et du Syndicat Général des Industries de Matériels et procédés pour les Traitements de Surface, (2002). 2. D.R. Lovley and E.J.P. Phillips, Appl. Environ. Microbiol., 60 (1994) 726. 3. C. Michel, M. Brugna, C. Aubert, A. Bernadac and M.Bruschi, Appl. Microbiol. Biotechnol., 55 (2001) 95. 4. F. Battaglia-Brunet, S. Foucher, A. Denamur, I. Ignatiadis, C. Michel and D. Morin, J. Ind. Microbiol. Biotechnol., 28 (2002) 154. 5. F. Battaglia-Brunet, S. Foucher, D. Morin, I. Ignatiadis Water, Air & Soil Pollution: Focus, in press

557


Effects of total-solids concentration on metal bioleaching from sewage sludge L.D. Villar* and O. Garcia Jr†. Department of Biochemistry and Chemical Technology, Institute of Chemistry, São Paulo State University, P.O. Box 355, Araraquara, SP 14.801-970, Brazil Abstract The possibility to remove metals from sewage sludge at high solids concentration by bacterial leaching may reduce the inherent cost of the process by lowering the sludge volume to be treated and increasing the metal concentration in the leachate. The effect of total-solids on the removal efficiency of chromium, copper, lead, nickel, and zinc from anaerobically digested sewage sludge was investigated. Indigenous sulphur-oxidizing acidithiobacilli were enriched from the sludge with 1% (w/v) S°. After 3 consecutive transfers, the enriched sludge was retained as the inoculum. The bioleaching assays were conducted in shake flasks containing fresh sludge at pH 7.0, which was inoculated with the enriched sludge (5% v/v), and supplemented with 0.5% (w/v) S°. Flasks were incubated at 30°C and 200 rpm. Total-solids concentration investigated varied from 10 to 40 g L-1 (10, 25, 32.5, and 40 g.L-1). The variation of pH and oxidation-reduction potential (ORP) was monitored during the bioleaching time course. Decrease in pH and increase in ORP values were faster at lower solids concentration. In spite of this, final solubilization efficiency was independent of solids variation for each metal investigated, resulting in the average final solubilization: Cr, 50.6%; Cu, 87%; Ni, 94.4%; Pb, 41.2%; and Zn, 99.5%. On the opposite, maximum solubilization rates, expressed as mg L-1.day-1, were greatly influenced by the solids concentration for each metal ion tested, resulting in either increase or decrease of solubilization rate depending on the metal considered. Thus, the effect of solids concentration on metal bioleaching was more pronounced on the kinetics of bioleaching than on the final removal achieved, thus encouraging the application of the process even at high solids concentration. Keywords: bioleaching, indigenous thiobacilli, metals, sewage sludge, total-solids

* Present address: Centro Técnico Aeroespacial, Instituto de Aeronáutica e Espaço, Divisão de Química 12228-904, São José dos Campos-SP, Brazil. ([email protected]) † Corresponding author ([email protected]) Thanks are due to FAPESP for supporting this research, and to the municipal sewage sludge works (SABESP) of Franca for providing the sludge samples. OGJ acknowledges CNPq for Researcher Fellowship.

559


1.

INTRODUCTION Disposal of sewage sludge generated during the treatment of municipal wastewater is becoming a growing problem. As an illustration of this issue, the metropolitan region of the city of São Paulo, Brazil, generates 295 ton.day-1 (dry basis) of sewage sludge, with an estimation of 750 ton.day-1 to be generated in year 2015 [1]. To accommodate this increase, sludge disposal in agriculture is being encouraged, since it appears to be the most attractive option for both economical and environmental reasons. Sewage sludge can be considered as a low-grade N-P fertilizer with a typical composition of 4% (w/w) nitrogen and 2% (w/w) phosphorous. Additionally, because of its organic composition, usually near to 40% (w/w), sludge can also be utilized as a soil ameliorant [2]. However, the accumulation of metals present in wastewaters into the solid phase of the sludge may pose some constraints to the sludge application into agricultural fields. Even when metal concentration in sludge is within the limits established through national regulations, there is still a risk they could reach toxic levels in soils [3]. Some recent studies [4, 5] have demonstrated that there is an accumulation of metals in soils after sludge application for long periods, although this accumulation was not showed to increase metal bioavailability in soil. In order to offer some alternative to the problems associated with the presence of metals in sewage sludge, chemical and biological processes have been investigated for the leaching of metals from sludge. It has been demonstrated that the chemical leaching is often more expensive than the bioleaching process due to the consumption of large amounts of inorganic acids [6]. Bioleaching of metals from sewage sludge employs sulphur- and iron-oxidizing bacteria from the Thiobacillus genus, which had some species recently reclassified into Acidithiobacillus, Halothiobacillus and Thermiothiobacillus [7]. Thiobacillus and the other new genera promote the acidification of sewage sludge through the oxidation of reduced sulphur compounds to sulphate by either direct or indirect mechanisms [8]. Most of the studies on bioleaching of metals from sludges have been performed in Canada. Some of these works have employed cultures of A. ferrooxidans and/or A. thiooxidans at the beginning of the studies on metal bioleaching from sludge. Afterwards, indigenous thiobacilli, enriched from sludge amended with sulphur, were successfully used, thus eliminating the requirement of chemical acidification of sludge to pH 4.0. Although several parameters can affect the bioleaching process, some of them are of greater importance when establishing the conditions for field application. Those parameters are initial pH, temperature and total-solids concentration. In previous works, we have addressed the effects of temperature on the kinetics and efficiency of metal bioleaching from sludge at two different initial pH values: pH 4.0 [9] and pH 7.0 (unpublished data). Higher efficiencies and solubilization rates were obtained at initial pH of 7.0 and temperature of 30°C for chromium, copper, and lead. Nickel and zinc showed higher solubilization rates for initial pH of 4.0 at the same temperature (unpublished data). In the operation of sewage treatment plants, it is less expensive to manage higher solids concentration sludges, thus implementation of bioleaching process would be more feasible if higher solids in sludge would not interfere in the efficiency of this process. In this context, we have investigated the effect of solids concentration, in the range of 10 to 40 g.L-1, on the kinetics and efficiency of metal solubilization from anaerobically treated sludge. 560


2.


2.1 Sampling and characterization of sludge A sample of anaerobically digested secondary sludge was obtained from the municipal wastewater treatment plant in the city of Franca, state of São Paulo, Brazil. Sludge was collected in sterilized polyethylene bottles, shipped cold, and kept at 4°C before utilization. Total solids concentration (TS) was determined as described in Standards Methods [10] in a number of 5 replicates. Total concentration for the metals chromium, copper, lead, nickel, and zinc in the sludge were determined after acid digestion with HNO3/H2O2 [11]. Dissolved metals in the sludge were determined by centrifugation of 20-mL samples at 3,300 g for 30 min, followed by membrane filtration. Determination of dissolved- and total-metal concentrations was carried out in triplicates, and analysis was conducted using plasma emission spectroscopy (ICP-AES). 2.2 Metal bioleaching assays Indigenous sulphur-oxidizing bacteria were enriched from the sludge with 1% (w/v) S°. After 3 consecutive transfers, the enriched sludge was retained as the inoculum. To obtain different total solids concentrations, the sludge (25 g total solids L-1) was either diluted with deionized water or concentrated by centrifugation. The solids concentrations obtained were 10, 32.5, and 40 g L-1. For the experiments, 250 mL of sludge (initial pH of 7.0) was inoculated with 5% (v/v) of the enriched sludge, and supplemented with 0.5% (w/v) S°. For each solids concentration investigated, duplicate flasks were used. Two controls were included for TS of 25 g L-1: (i) biological control: inoculated sludge without S° amendment, and (ii) chemical control: autoclaved sludge (at 121°C for 20 min) amended with S°, but not inoculated. All flasks were incubated in a gyratory shaker at 30°C and 200 rpm. 2.3 Chemical analysis Samples, 20 mL, were periodically withdrawn for pH (combined electrode, Orion 520A) and ORP (platinum electrode, Cole-Palmer) measurement and subsequent centrifugation at 3,300 g for 30 min, with the supernatant being analysed for solubilized metals by plasma emission spectroscopy (ICP-AES). Metal solubilization efficiency, defined as r, was calculated as the ratio between the solubilized metal due to bioleaching, and total metal present in the sludge (Equation 1). t [ Me sol ] − [ Me diss ] r(%) = × 100 (1) [ MeT ] × TS × 0.001

where, t [ Me sol ] , solubilized metal present in sludge at time t of the bioleaching experiment, in mg -1 L ; [ Mediss ] , dissolved metal in sludge before the bioleaching experiment, in mg L-1;

[ MeT ] , total-metal concentration in sludge, in mg kg-1 dry sludge; TS , total-solids concentration for sludge, in g dry sludge L-1.

561


3.

RESULTS AND DISCUSSION The enrichment of indigenous sulphur-oxidizing thiobacilli for inoculum production resulted in a reduction of the time required to decrease the sludge pH down to 1.3 from 8 to 6 days after three consecutive transfers (data not shown). Both controls performed for the initial acidification did not show any significant changes in pH (data not shown). The enriched sludge was kept at 4°C and used as inoculum. Total-solids concentration for the sludge samples collected was determined in five replicates and resulted in the concentration of 25 g L-1. The average total-metal concentration for this sludge is presented in Table 1 along with the U.S. EPA [12] recommended limits for land application. Table 1. Total-metal concentration (mg.kg-1 dry sludge) and dissolved-metal concentration (mg.L-1) for the anaerobically treated sludge U.S. EPA [12] recommended limits for total-metal concentration for land application. Values in parentheses are dissolved/total-metal concentration ratio (% w/w)

Total-Metal Concentration Anaerobic Sludge

Recommended Limits

Dissolved-Metal Concentration

Chromium

225

-*

0.08 (1)

Copper

253

1,500

0.04 (1)

Lead

129

300

0.10 (3)

Nickel

53

420

0.13 (9)

Zinc

929

2,800

0.39 (2)

Metal Ions

* No recommended limit.

Dissolved-metal concentrations were significant only for nickel (9% of total metal concentration present as dissolved metal), indicating that metal ions in sludge are mainly in the solid phase. Despite of the low values obtained for dissolved-metal concentration, they were taken into account when calculating the solubilization efficiency by using Equation 1. Although the metal contents for this sludge are below the EPA recommended limits, for all the metal ions analyzed, it is still interesting to look for processes for metal removal from sludge, since decontaminated sludge can be applied to soil at higher loads and for a longer period of time, with a lower risk of environmental contamination. Figure 1 presents the pH and ORP profiles obtained during the bioleaching assays. Both the decrease in pH and the increase in ORP were faster at lower solids concentrations due to changes in slope, as it can be verified from Figure 1. For solids-concentration values of 32.5 and 40 g L-1, a short lag phase of about 1 day was also observed for pH decrease (Figure 1). Some studies [13, 14], which had also determined the sulphate production during bioleaching, had attributed this observation to the increase in buffering capacity of the sludge with higher solids contents, thus preventing the pH to fast decrease. The controls showed in Figure 1 were run for the TS of 25 g L-1. Both controls did not show a significant variation in pH (decrease from 7.0 to 6.0), although some increase in ORP was observed, especially for the biological control. During bioleaching, the anaerobic sludge in the controls was submitted to aeration, which could explain the increase in ORP observed. 562


Figure 1. Variation in sludge pH (A) and oxidation-reduction potential (B) during bioleaching assays at various sludge solids concentration Symbols: (■) TS = 10g L-1; ( ) TS = 25 g L-1; (▲) TS = 32.5 g L-1; (○) TS = 40 g L-1; (+) biological control for TS = 25 g L-1; (x) chemical control for TS = 25g L-1

The solubilization efficiency (r) is presented in Figure 2 for the metals investigated.

Figure 2. Metal solubilization during bioleaching assays for the metals chromium, copper, lead, nickel, and zinc at various total-solids concentration Symbols: (■) TS = 10g L-1; (△) TS = 25 g L-1; (▲) TS = 32.5 g L-1; (○) TS = 40 g L-1; (+) biological control for TS = 25 g L-1; (x) chemical control for TS = 25g L-1

Fitting of the data obtained was conducted following the sigmoidal model described in Boltzmann equation (Equation 2). Both controls did not show any significant solubilization. A1 − A2 + A2 (2) y= 1 + e ( x − x0 ) / dx 563


where, xο is the centre of the linear range of the function, dx is the width of the linear range, A1 is the initial y value, i.e., y (-∞) and A2 is the final y value, i.e., y (+∞). The values of A2 obtained from Boltzmann’s fitting were compared by analysis of variance (ANOVA) for each metal. When the null hypothesis was rejected, Tukey’s test was applied to compare the means. The results obtained are presented in Table 2, along with the maximum solubilization rates (vmax) calculated from the slope of the linear part of the curves showed in Figure 2. The same statistical treatment mentioned above was applied to the maximum solubilization rates. Final solubilization values higher than 100% observed for nickel and zinc (Table 2) can be attributed to the metal income from the enriched sludge used as inoculum, which was not considered by Equation 1. Table 2. Final solubilization (in %) and maximum solubilization rate (in mg.L-1.day-1) for the bioleaching assays at various total-solids concentration (in mg.L-1) Metal Chromium

Copper

Nickel

Lead

Zinc

Total-Solids Concentration

Final Solubilization

Maximum Solubilization Rate (vmax)

10

53.3a*

1.8a

25

52.8a

1.2b

32.5

45.8a

1.2b

40

50.5a

1.2b

10

80.9a

1.5a

25

78.8a

1.7a

32.5

96.4a

4.1b

40

91.9a

4.1b

10

92.2a

0.23a

25

81.6a

0.58b

32.5

103a

0.82c

40

101a

0.89c

10

49.3a

0.34a

25

38.4a

0.74b

32.5

39.0a

0.72b

40

38.2a

0.53c

10

102a

3.6a

25

99.1a

10b

32.5

98.5a

24c

40

98.3a

18d

* For each metal, final solubilization and maximum solubilization rates were submitted to analysis of variance (ANOVA). No statistical differences were obtained for final solubilization at the solids concentrations investigated. For maximum solubilization rates, means followed by the same letter are not different by the Tukey’s test at 5%.

Although final solubilization for the metals investigated was not influenced by the solids variation, as showed by the results obtained by analysis of variance, at 5% of 564


significance level (Table 2), solids markedly modified maximum solubilization rates. For chromium, at the lower solids concentration it was obtained the higher solubilization rate, whereas for the other metals, there was a tendency of higher rates to be obtained at higher solids contents, especially at 25 and 32.5 g.L-1. In fact, higher solids concentration implies higher insoluble metal concentration in the medium, if expressed in mg metal.L-1. The increase in metal concentration accelerates the process of metal bioleaching through the chemical equilibrium. However, higher solids concentration also implies lower decrease in pH, and thus, lower hydrogen ions available for bioleaching, which decreases the rate of solubilization. For this reason, depending on the metal being leached, such equilibrium may be affected more by hydrogen ions availability than by insoluble metal concentration, or vice-versa. 4.

CONCLUSIONS Variations of the solids concentration in the range of 10 to 40 g.L-1 do not interfere in the final metal solubilization yield obtained during metal bioleaching from sewage sludge. However, the maximum solubilization rate is markedly influenced by solids variation, with higher rates being obtained at TS above 25 g.L-1 for copper, lead, nickel and zinc. An immediate advantage resulted from this study is the possibility to conduct metal bioleaching process in wastewater treatment plants at higher solids concentration, thus lowering the costs of this sludge treatment. REFERENCES

1. M.T. Tsutiya (ed.), Biossólidos na Agricultura (Biosolids in Agriculture), SABESP, São Paulo, 1997. (in Portuguese) 2. L. Korentajer, Water SA, 17 (1991) 189. 3. R. Renner, Environ. Sci. Technol., 34 (2000) 430A. 4. J.J. Kelly, M. Häggblom and R.L. Tate III, Soil Biol. Biochem., 31 (1999) 1467. 5. B.P. Knight, A.M. Chaudri, S.P. McGrath and K.E. Giller, Environ. Pollut., 99 (1998) 293. 6. D. Couillard and G. Mercier, Environ. Pollut., 66 (1990) 237. 7. D.P. Kelly and A.P. Wood, Int. J. System. Evolut. Microbiol., 50 (2000) 511. 8. A.M. Martin (ed.), Biological Degradation of Wastes, Elsevier, Amsterdam, 1991. 9. L.D. Villar and O. Garcia Jr., Int. Biohydrometallurgy Symp., Ouro Preto, Brazil (2001) 21B. 10. L.S. Clesceri, A.E. Greenberg and R.R. Trussell (eds.), Standard Methods for the Examination of Water and Wastewater, 17th ed., Am. Public Health Assoc./Am. Water Works Assoc./Water Pollut. Control Fed., Washington, DC, 1989. 11. J.J. Delfino and R. E. Enderson, Water & Sewage Works (1978) R32. 12. U.S. EPA., Standards for the Use and Disposal of Sewage Sludge (Code of Federal Regulations 40 CFR Part 503), Washington, DC, 1996. 13. T.R. Sreekrishnan, R.D. Tyagi, J.F. Blais and P.G.C.Campbell, Water Res., 27 (1993) 1641. 14. R.D. Tyagi, J.F.Blais, J.C. Auclair and N. Meunier, Water Environ. Res., 65 (1993) 196.

565


Enhancement of electrodialytic soil remediation through biosorption Pernille E. Jensena*, Lisbeth M. Ottosena and Birgitte K. Ahringb a

BYG.DTU, Building 204, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark b BiC-BioCentrum, Building 227, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark 1.

INTRODUCTION Toxic metals are being introduced to the environment through numerous different processes including different industrial production-processes, incineration emissions and waste disposal. The final sink for most metals in the environment is soil or sediment, where the metals adsorb strongly. Therefore metal-polluted soils are found in vast amounts in populated areas, where they possess a hazard to both environment and humans. Through the last decades a rising concern of metal pollution has developed, and different methods for decontamination of metal-polluted soils have been investigated. Since the metals cannot be degraded like organic pollutants, the only three ways to treat such metal-polluted land are i) to remove the polluted soil and deposit it in a place, where it does less harm or ii) stabilize the metals in the soil to make them less bioavailable and thereby less toxic or finally to iii) mobilize the metals in order to remove them from the soil followed by either deposition of the metals or possibly by reuse of these. At the moment deposition of polluted soil is the dominating procedure, when dealing with contamination problems. However the ideal solution seen in an environmental perspective would be the recovery of both metal and soil through separation of the two. This research aims at development of a method for such a separation. The separation process used is the electrodialytic remediation method described below, which has shown efficient in remediation of copper polluted soil [Ottosen et al., 1997]. The method has also been investigated for remediation of soils polluted with other metals, e.g. lead. It was shown that mobilization of lead only takes place at very low soil pH (< 3). At such low pH many other soil components dissolve, leaving the soil less usable after treatment [Ottosen et al., 2001]. In order to enhance the electrodialytic mobility of lead at higher pH values, addition of microbial biosorbents to the soil is suggested. The properties and possibilities of such biosorbents are further discussed below.

* Corresponding author: [email protected]

567


2.

ELECTRODIALYTIC SOIL REMEDIATION The electrodialytic soil remediation method was developed at the Technical University of Denmark for decontamination of soil polluted with heavy metals. The method takes advantage of electric current, which has the ability to desorb and move ions in soil [Hansen et al., 1997]. The electric current results in hydrolysis reactions at the electrodes, producing acid at the anode and base at the cathode. This results in development of an acidic front, which moves from the anode to the cathode and an alkaline front moving in the opposite direction. The development of an alkaline front is problematic to the remediation because lead and most other metals precipitate in alkaline environments. Therefore, in addition to electric current, the electrodialytic method involves a cation-exchange membrane, separating the catholyte from the soil to prevent hydroxide ions from entering the soil [Ottosen et al., 1997]. Similarly at the anode end an anion-exchange membrane is separating anolyte and soil. This makes it possible to control the chemistry of the system liquids. The anion exchange membrane does however not prevent development of an acidic front because water splitting is taking place at the membrane surface, followed by transport of the hydroxide-ion across the anion-exchange membrane [Ottosen et al. 2000]. In figure 1 a schematic illustration of a laboratory cell for investigation of the electrodialytic soil remediation method is given.

Figure 1. Schematic illustration of the electrodialytic remediation method. The soil is placed in compartment ΙΙ. Electrolytes are placed in compartment Ι and ΙΙΙ. Illustrated is the rejection of cations by the anion-exchange membrane, the rejection of anions by the cation-exchange membrane and the water splitting at the surface of the anion-exchange membrane. AN = Anion-exchange membrane. CAT =Cationexchangemembrane 3.

BIOSORPTION Biosorption has been extensively investigated for decontamination of heavy-metal polluted solutions [Volesky and Holan, 1995]. The process has many advantages which can also be of value for soil remediation: Cheap biomass can be obtained from different industrial waste streams [Volesky, 2001], a fast biosorption process shortens the remediation time [Gupta et al., 2000], possible regeneration and reuse of both biomass and heavy metals makes the method both economically and practically feasible [Volesky, 2001]. Also biosorption works well at low heavy metal concentrations (1-100 mg/L) [Gupta et al., 2000], which are the concentrations most often found in soil pore-water during electrodialytic remediation [Ribiero, 1998].

568


Biosorption of Pb from solution has been investigated for several different organisms including bacteria, algae, yeasts and filamentous fungi. Examples are the bacteria Phormidium laminosum [Blanco et el., 1999], Pseudomonas aeruginosa [Chang et al.,1998], several different marine algae [Holan and Volesky, 1994] and mixed species of green, reed and brown algae [Diniz et al., 2001], the filamentous fungi Aspergillus niger, Mucor rouxii and Rhizopus oryzae [Baik et al. 2002], Rhizopus oligosporus [Ariff et al., 1999] and Rhizopus arrhizus [Fourest et al., 1994; Brady and Tobin, 1995], and the yeasts Saccharomyces uvarum [Ashkenazy et al., 1997] and Saccharomyces cerevisiae [Engl and Kunz, 1995; Ashkenazy et al., 1999]. Also experiments have been performed with mixed aerobic wastewater culture [Ghosh and Bupp, 1992] as well as microbial products such as chitin and chitosan [Eiden et al., 1980] and cell walls [Baik et al., 2002]. Results are reported for both living [Ghosh and Bupp, 1992; Engl and Kunz, 1995] and dead biomass [Holan and Volesky, 1994; Diniz et al., 2001]. Also biomass that has been pretreated in different ways can be mentioned such as NaOH treated cell walls [Ashkenazy et al., 1997; Baik et al., 2002] or dried and powderized cells [Fourest et al., 1994; Ariff et al., 1999] and acetone washed cells [Ashkenazy, 1997]. Finally biomass which has been immobilized in variable matrixes has been tested for possible use in treatment of lead polluted wastewater. Examples are Blanco et al. (1999) who tested biomass immobilized in polysulfone and epoxy resin beads and Chang et al. (1998), who tested biomass entrapped in calcium alginate beads and polyacrylamide gel. It is shown that biomass can be as efficient an adsorbant as commercial ion exchangers. Holan and Volesky (1994) showed this for lead at solution equilibrium concentrations of 200 mg/g. However, when lowering the equilibrium concentration to 10 mg/g, the commercial ion exchangers showed a higher efficiency than the biomass. Maximum biomass-equilibrium capacities in the order of 300 mg/g lead are commonly seen [Aikin et al., 1979; Holan and Volesky, 1994; Diniz et al., 2001]. A maximum possible metal loading is often seen for solution-equilibrium concentrations above 200 mg/g lead [Ariff et al., 1999; Diniz et al., 2001]. For many biosorbents it is seen that lead is the metal sorbing most efficiently to biomass after uranium when looking at the mass adsorbed per gram dry biomass [Brady and Tobin, 1995]. Important parameters for biosorption are pH and growth state of the microorganisms [Ledin 2000]. At low pHs the sorption decreases because of protonation of the negatively charged groups on the biomass. At high pH sorption may also decrease. This is probably due to heavy metal precipitation or formation of negatively charged metal-complexes in alkaline environment. 4.

MICROBIAL ENHANCEMENT OF LEAD MOBILITY IN SOIL It is known, that microorganisms accumulate heavy metals in polluted soil systems [Ledin, 2000]. It has also been shown that heavy metals can be mobilized in aquifer materials by leaching with bacterial extracellular polymers [Chen et al., 1995], and in soil by stimulation of microbial growth [Chanmughathas and Bollag, 1987; Bender et al., 1989]. Further more it is shown that microorganisms can be transported in soil by addition of electric current [DeFlaun and Condee, 1997]. The purpose of this study is to add leadadsorbing microorganisms to lead polluted soil, making them work as “complexing agents” with mobilization of the metal as a consequence. The mobile and electrically charged microorganisms are then to be transported out of the soil by means of electrodialysis, carrying the lead with them.

569


5.

EXPERIMENTAL WORK An experimental setup for addition of microbial matter to soil with succeeding electrodialytic remediation is developed. The setup needs to be different than the one shown in figure 1, because the microorganisms are unable to pass an ion-exchange membrane. Generally the size of soil-microorganisms is around 1 µm. Consequently a barrier with pores of this size or larger must be used to separate soil and process-water.

Figure 2. Schematic illustration of the setup for experimental evaluation of the bioelectrodialytic soil remediation method. AN = Anion-exchange membrane. CAT = Cation-exchange membrane. NET = fiberglass armed insect-net with pore size of 1 mm. PM = Passive Membrane. Soil and biomass is mixed in compartment ΙΙΙ

A barrier consisting of fiberglass-armed insect-net with a pore-size of 1 mm is tested, since this represents a low-cost solution. With this setup it is possible to collect the heavymetal loaded biomass in the concentration-chamber ΙΙ in front of the anion-exchange membrane. Circulation of the process liquid at the soil side of the cation-exchange membrane is necessary in order to avoid fouling of the membrane and high electrical resistance. A passive membrane is placed between the insect-net and the cation-exchange membrane in order to avoid soil particles in the pump and tubes for this circulation. The setup is seen in figure 2. 5.1 Soil characteristics Laboratory experiments with two different lead-polluted soils are performed. Soils with medium carbonate content are chosen, because it was earlier shown that leadpolluted, carbonaceous soils are not easily decontaminated by electrodialysis [Pedersen, 1999]. Such soils are therefore candidates for combined bio-electrokinetic treatment. Also both soils contain lead at levels exceeding the value where Danish authorities prohibit human contact with the soil (400 mg/kg). Both soils are Danish. One soil is from a former army-site in Copenhagen: Holmen (soil 1). The other soil (soil 2) comes from the town Kalundborg situated in western Zeeland. The source of pollution at this site is not known. Parameters important for the remediation are shown in table 1. As is common for originally polluted soils, the lead is inhomogeneously distributed within the soil. This is illustrated by the high standard deviation on the analytical result. When experiments are made, the soil used is mixed as well as possible, and three samples are taken from this mixture before starting the experiments to give a more precise estimate of the exact amount of soil in the individual experiment. Apart from lead, soil 2 also contains copper at 764 +/- 424 mg/kg and Zn at 2028 +/- 544 mg/kg.

570


Table 1. Characteristics of experimental soils Pb [mg/kg] CaCO3 % (w/w) CEC meq/100g pH Clay % (w/w) Org. matter % (w/w)

Soil 1 637+/-242 7,7 8,2 7,4 0,7 7,0

Soil 2 765+/-760 9,1 4,9 7,6 2,8 3,6

5.2 Sequential extraction Sequential extraction of Lead has been performed in order to get an impression of how mobile the lead is in the soil before treatment. Results are shown in figure 3. The method used is the one recommended by the "Standards, Measurements and Testing Programme of the European Union" [Mester, 1998]. Here it is recognized that it is not possible to find a completely selective method for extraction of metals from certain soil fractions, but it is possible to use this method to get a qualitative impression of how the metals are bound and their mobility. Step 1 gives an estimate of how much metal is found as exchangeable ions and carbonates. Step 2 extracts metals that are reduceable. Step 3 extracts oxidizable metals and step 4 reveals the residual and least mobile fraction. The results of sequential extraction are shown in figure 3. It is seen that lead is binding tightly to the soil, as no lead is found as exchangeable ions and carbonates. Furthermore it is shown that the lead in soil 1 is less mobile than the lead in soil 2.

Figure 3. Results from sequential extraction of lead from the two soils 6.

CONCLUSIONS Lead polluted soil is not easily remediated, and improved remediation technology is a prerequisite for successful remediation. It is rendered that combination of biosorption and electrodialytic remediation can improve remediation efficiency in an economically feasible manner.

REFERENCES

1. Ariff, A.B., Mel, M., Hasan, M.A., Karim, M.I.A. (1999), The kinetics and mechanism of lead (II) Biosorption by powderized Rhizopus oligosporous, Worls Journal of Microbiology and Biotechnology, 15, 291-298. 571


2. Ashkenazy, R., Gottlieb, L., Yannai, S. (1997), Characterization of Acetone-Washed Yeast Biomass Functional Groups Involved in Lead Biosorption, Biotechnol. Bioeng., 55, 1-10. 3. Ashkenazy, R., Yannai, S., Rahman, R., Rabinovitz, E., Gottlieb, L., Fixation of spent Saccharomyces cerevisiae biomass for lead sorption, Appl. Microbiol. Biotechnol., 52, 608-611. 4. Baik, W.Y., Bae, J.H., Cho, K.M., Hartmeier, W. (2002), Biosorption of heavy metals using whole mold of mycelia and parts thereof, Bioresource Technology, 81, 167-170. 5. Bender, J.A., Archibold, E.R., Ibeanusi, V., Gould, J.P., (1989), Lead removal from contaminated water by a mixed microbial ecosystem, Wat. Sci. Tech., 21,1661-1664. 6. Blanco, A., Sanz, B., Llama, M.J., Serra, J.L. (1999), Biosorption of heavy metals to immobilized Phormidium laminosum biomass, Journal of Biotechnology, 69, 227-240. 7. Brady, J.M., Tobin, J.M. (1995), Binding of hard and soft metal ions to Rhizopus arrhizus biomass, Enzyme and Microbial Technology, 17, 791-796. 8. Chang, J.-S., Huang, J.-C., Chang, C.-C., Tarn, T.-J. (1998), Removal and recovery of lead fixed-bed biosorption with immobilized bacterial biomass, Wat. Sci. Tech., 38, 171-178. 9. Chanmugathas P., Bollag, J.-M. (1987), Microbial mobilization of cadmium in soil under aerobic and anaerobic conditions, J. Environ. Qual., 16, 161-167. 10. Chen, J.-H., Lion, L.W., Ghiorse, W.C., Shuler, M.L. (1995), Mobilization of adsorbed cadmium and lead in aquifer material by bacterial extracellular polymers, Wat. Res., 29(2), 421-430. 11. DeFlaun, M.F., Condee, C.W. (1997), Electrokinetic transport of bacteria, Journal of Hazardous Materials, 55, 263-277. 12. Diniz, V.G.S., Silva, V.L., Lima, E.S., Abreu, C.M.A. (2001), Lead Biosorption in “Arribadas” Algal Biomass, Biohydrometallurgy: Fundamentals, Technology and Sustainable Development, Part B 13. Eiden, C.A., Jewell, C.A., Wightman, J.P. (1980), Interaction of lead and chromium with chitin and chitosan, Journal of Applied Polymer Science, 25, 1587-1599. 14. Engl, A., Kunz, B. (1995), Biosorption of Heavy Metals by Saccharomyces cerevisiae: Effects of Nutrient Conditions, J. Chem. Tech. Biotechnol., 63, 257-261. 15. Fourest, E., Canal, C., Roux, J.-C. (1994), Improvement of heavy metal biosorption by mycelial dead biomasses (Rhizopus arrhizus, Mucor miehei and Penicillium chrysogenum): pH control and cationic activation, FEMS Microbiology Reviews, 14, 325-332. 16. Ghosh, S., Bupp, S. (1992), Stimulation of biological uptake of heavy metals, Wat. Sci. Tech., 26, 227-236. 17. Gupta, R., Ahuja, P., Khan, S., Saxena, R.K., Mohapatra, H. (2000), Microbial Biosorbents: Meeting challenges of heavy metal pollution in aqueous solutions, Current Science, 78(8), 967-973. 18. Hansen H.K., Ottosen L.M., Villumsen A. (1997), J. Chem. Tech. and Biotechnol. 70: 67-73. 19. Holan, Z.R., Volesky, B. (1994), Biosorption of lead and nickel by biomass of marine algae, Biotechnology and Bioengineering, 43, 1001-1009. 20. Ledin, M. (2000), Accumulation of metals by microorganisms – processes and importance for soil systems, Earth-Science Reviews, 51, 1-31. 21. Mester, Z., Cremisini, C., Ghiara, E., Morabito, R. (1998), Comparison of two sequential extraction procedures for metal fractionation in sediment samples, Analytica Chimica Acta, 359, 133-142.

572


22. Ottosen L. M., Hansen H. K., Laursen S., Villumsen, A. (1997), Electrodialytic Remediation of Soil Polluted with Copper from Wood Preservation Industry, Environ. Sci. Technol. 31(6): 1711-1715. 23. Ottosen, L.M., Hansen, H.K., Hansen, C.B. (2000), Water-splitting at ion-exchange membranes and potential differences in soil during electrodialytic soil remediation, Journal of Applied Electrochemistry, 30, 1199-1207. 24. Ottosen, L.M., Hansen, H.K., Ribeiro, A.B., Villumsen, A. (2001), Removal of Cu, Pb and Zn in an applied electric field in calcareous and non-calcareous soils, Journal of Hazardous Materials B85, 291-299. 25. Pedersen, A.J., Jensen, P.E.J. (1999), Electrodialytic remediation of Pb contaminated soil – effect of soil properties and Pb distribution, in: Proceedings of the 2.nd symposim, Heavy Metals in the Environment and Electromigration Applied to Soil Remediation, Technical University of Denmark, Lyngby, Denmark. 26. Ribeiro, A.B. (1998), Use of Electrodialytic remediation technique for removal of selected heavy metals and metals from soils, PhD.-thesis, Department of Geology and Geotechnical Engineering (now BYG-DTU), Technical University of Denmark, Denmark, 1998. 27. Volesky, B., Holan, Z.R. (1995), Biosorption of Heavy Metals, Biotechnol. Prog., 11, 235-250. 28. Volesky, B. (2001), Detoxification of metal-bearing effluents: biosorption for the next century, Hydrometallurgy 59, 203-216.

573


Fundamentals of the uranium separation in constructed wetlands F. Glombitza, F. Karnatz, H. Fischer, J. Pinka and E. Janneck Department of Biotechnology, G.E.O.S. Freiberg Ingenieurgesellschaft mbH, Gewerbepark Schwarze Kiefern, D 09633 Halsbrücke, Germany Abstract The microbial process of the uranium separation from mine flooding and drainage water was investigated and the results will be demonstrated. Uranium separation takes place in constructed wetlands under anaerobic conditions due to microbial reduction of the U6+ into insoluble U4+. The concentration of uranium can be reduced to values lower than 0.3 mg/l. Competitive reactions take also part and influence the degree of separation efficiency as well as the solubility and mobility of the precipitated uranium. The influences of the microbial community composition as well as the mine water composition on the separation were demonstrated. Sulphate-reducing microorganisms are able and responsible to reduce uranium in contrast with denitrifying bacteria. The formed HCO3- as final product from the transfer of the organic carbon source into biomass is the reason for the formation of uranylcarbonate complexes and the release of the stored uranium. Highly uranium concentrated fraction can therefore be detected sometimes in the case of missing SRB and of unfavourable HCO3- concentrations. Recommendations for the risk assessment and the evaluation of the stability of a constructed wetland will be derived after the treatment of a pilot plant. 1.

INTRODUCTION Uranium mine drainage and process waters contain different hazardous substances in most cases. Depending on the pH of the water, uranium and radium are the most important contaminants among different heavy metals. Iron, manganese, zinc, and some other heavy metals exist as cations, arsenic, sulphate and carbonate as anions. The classical water treatment technology passes three steps, a) separation of Uranium by precipitation or ion exchange, b) separation of Radium by BaCl2 precipitation and c) separation of Arsenic as well as some heavy metals by addition of FeCl3 and precipitation of Fe-As-compounds by addition of NaOH [1-2]. The sludge formed is separated, mixed with cement to form stable compounds as concrete and stored in special disposals [3]. The cost of this kind of water treatment is very high and lies in a range from 5-15 €/m³ water, depending on the treated water volume. Thus, cheaper methods for the treatment are necessary. Investigations for the separation of uranium by natural reactions in wetlands are a precondition for the development and use of cheaper and safer techniques in future. 575


2.

SUMMARY / COMPOSITION OF BASIC REACTIONS OF THE URANIUM SEPARATION The possibility to separate uranium in a wetland at anaerobic conditions is a known fact [4-6]. Knowledge of the different reactions, which take place in such a system, is a precondition to control the process and guarantee a long-time process stability. Summaries of some important possible reactions, which can take part in a wetland, are shown in table 1. Table 1. Some possible important uranium separation processes in an anaerobic wetland system

Microbiological reactions: SO42- + microorganisms + C org. NO3- + microorganisms + C org. U6+ + microorganisms (SRB) U6+ + microorganisms (DENI) Chemical reactions: U6+ + HS U6+ + S2Biosorption: U6+ + biomass

Æ Æ Æ Æ

HS- + biomass + HCO3- + OHN2 + biomass + HCO3- + OHU4+ (UO2) ? U4+ (UO2) ?

Æ Æ

U4+ + S° + H+ U4+ S2, U 6+ S3, UOS (?)

Æ

U - biomass

Microorganisms reduce sulphate and nitrate ions. Sulphide is one product of the reaction. It can be assumed, that the U6+-ion, which exists as uraniumsulphonyl or uraniumcarbonyl, is reduced and precipitated by means of the formed S2- or reduced and precipitated by the existing microorganisms [7-8]. Separation of Uranium by means of biosorption can also be taken into consideration as described by Tzesos [9]. 3.

INVESTIGATIONS OF THE REACTION

3.1 Uranium separation with the aid of sulphide and dithionite Uranium containing model solutions and mine waters were treated by the addition of increasing concentration of S2- as well as dithionite. Uranium precipitation could not be observed even after a long reaction time. Figures 1 and 2 show the behaviour of the uranium concentration by increasing concentration of sulphide and during a long reaction time. A decrease of the uranium concentration could not be observed in both cases.

576


Figure 1. Influence of the dithionite concentration on the uranium concentration

Figure 2. Demonstration of the influence of the reaction time on the uranium concentration

3.2 Separation by means of sulphate-reducing microorganisms The influence of the sulphate-reducing microorganisms on the uranium concentration was investigated by experiments with living SRB in uranium containing sulphate free mediums. The uranium containing solution was inoculated by living SRB and the uranium concentration was determined. The same experiment was carried out with dead "autoclaved" cells to determine and take into consideration the possible influence of the biosorption on the decrease of the uranium concentration. The results of these experiments are demonstrated in figure 3.

Figure 3. Separation of uranium by SRB

The results in figure 3 demonstrate the ability of the SRB to separate uranium. U6+ is reduced to U4+, which precipitates as black compound, and was specified by X–Ray as Uraninite. 4.

REACTION IN A CONTINUOUS FLOWED WETLAND

4.1 Continuous column tests The use of SRB for the reduction of uranium in a wetland process requires anaerobic conditions in the reaction basin. For this purpose, column tests were carried out for a period of some years. Conditions and results are summarised in Tables 2, 3 and 4. 577


Table 2. Process conditions of a long time column process Parameter Volume of the column Graves Volume of water in the column Water flow rate Residence time DOC concentration in the water (CH3OH) Uranium concentration in the drainage water Uranium concentration in the treated water

Number and Dimension 4,590 ml 5,277 g 1,720 ml 36 ml/h 48 h 80 – 100 mg/l 1 – 2 mg/l < 0. 3mg/l

Table 3. Consumption coefficients Component

Dimension

mg/l C - limited

mg/l NH4 - limited

Consumption of P

mg/l

0.65

0.13

Consumption of DOC

mg/l

104.4

89.6

-

mg/l

4.62

2.57

-

mg/l

143

216.87

+

mg/l

1.12

-

mg/l

1.99

0.61

mg/l

884

388.6

Formation of C

mg/l

173.9

76.45

Reduction of U

mg/l

0.84

1.24

Consumption of NO3 Consumption of SO4

Consumption of NH4

Consumption of total N Formation of HCO3

-

Table 4. Specific coefficients Kind of coefficients

Dimension

Amounts C – limited

Amounts without NH4 NH4 – limited

Related on DOC

g SO4/g DOC mg N/g DOC mg P/g DOC mg U/g DOC mg DOC/g SO4 mg N/g SO4 mg P/g SO4 mg U/g SO4 mg DOC/mg U mg N /mg U mg P/mg U mg SO4/mg U

1.37 19.06 6.22 8.04 730.07 13.92 4.55 5.87 124.29 2.37 0.77 170.24

2.93 6.9 1.49 13.93 340.99 2.33 0.51 4.75 71.77 0.49 0.107 210.48

Related on SO4

Related on U

Table 3 and 4 contain the obtained consumption and specific consumption coefficients for the separation and the reduction per unit sulphate and the consumed carbon. 340-730 mg of DOC are able to reduce 1 g sulphate. The formed biomass can separate between 4.8-5.8 mg of uranium simultaneously depending on the reaction 578


conditions. 71-124 mg of DOC are required for the separation of 1 mg uranium. The process conditions were changed for the simulation of possible disturbances and for checking different situations concerning the stability. The question turned up, what happens after a break or during the phase of restarting the process. The continuous process in a column was stopped and the column was flooded only with drinking water for some month. Drainage water was sprinkled again after this time and the concentration in the effluent as well as the composition of the microbial community and the CFU values were determined. Figure 4 shows the alteration of the U, DOC and HCO3- concentrations.

Figure 4. Representation of the start situation of a continuous process after a longtime break

Figure 5 shows the alteration of the sulphate concentration due to the reduction to sulphide. The results demonstrate a decrease of the DOC in the treated water due to its consumption and consequently an increase of the HCO3- concentration.

Figure 5. Alteration of the sulphate concentration

The uranium concentration increases in the treated water during the initial step. A release of uranium as [UO(CO3)3]2- uranylcarbonate by the formed HCO3 can be assumed and is possible [8]. A decrease of the Uranium concentration can be observed in the next steps although the HCO3- concentrations reach the same values. The analysis of the 579


microbial population and the behaviour of the CFU of the denitrifying and sulphatereducing bacteria are therefore investigated. Results are demonstrated in Figure 6. A rapidly increasing number of denitrifying bacteria can bee seen followed by an increase of the number of sulphate-reducing bacteria some days later.

Figure 6. Representation of the development of the denitrifying and sulphatereducing bacteria

Figure 7 shows the CFU-values and the alteration of the uranium concentration. This figure reveals a decrease of the uranium concentration in that moment where the sulphatereducing bacteria turned up in the system. That means: only sulphate-reducing bacteria are responsible for the uranium separation and the decrease of the concentration. Denitrifying bacteria do not reduce and separate uranium.

Figure 7. Connection between uranium separation and the alteration of the CFU values of denitrifying and sulphate-reducing bacteria

580


Figure 8. Separation of the uranium by means of the sulphate-reducing bacteria 4.2 Practical results and practical experiences from a real wetland These results are used for the controlled removing of uranium from drainage water in a constructed wetland. Main part of the wetland is an anaerobic reaction chamber with a total volume from 150 m³ and a volume of free water around 50-70 m³. The water residence times lie in a range from 50 to 100 h depending on the water flow rate. The plant was started in the summer of 2001. The separation process is based on the cultivation of sulphate-reducing bacteria for uranium reduction. Figure 9 shows the separation of uranium up to the point of the total consumption of the usable carbon sources. Uranium separation could not be observed and the process collapsed after this point. DOC values are lower than 10 mg/l. Figure 10 shows the situation after starting a carbon dosing followed by increasing number of SRB. Uranium concentration in the treated water was decreased to a value lower than 0.3 mg/l as a result of the cultivation of the sulphatereducing microorganisms.

Figure 9. Uranium concentration at different places in the reduction chamber and collapsing of the uranium separation due to a lacking of a carbon source (PB: Mine drainage water, S2o, S2m, S2u different sampling points)

581


Figure 10. Representation of the situation of uranium separation after dosing of carbon sources (PB mine drainage water, RK/S3 sampling point at the end of the reduction chamber) 5.

CONCLUSION Uranium can be separated in constructed wetlands at anaerobic conditions. The separation takes place by reduction of the U6+ by means of sulphate-reducing bacteria. Denitrifying bacteria do not separate uranium by the same mechanism. In the case of missing of SRB Uranium can be delivered by the microbial formed HCO3- and an increase of the concentration can be observed. Constant conditions concerning the growth of SRB are a precondition of stable process conditions and stable storage of the precipitated and stored uranium therefore. ACKNOWLEDGEMENTS The German Ministry for Education, Science, Research and Technology (BMBF FKZ 02 WB 0104), and the EU in the 5th frame programme (Project number: EVK1 - 1999 00168P), which are gratefully acknowledged, support this work. REFERENCES

1. G. Kießig, Ch. Kunze, Wasserbehandlung und Rückstandsentsorgung Geowissenschaften 14 (1996) 11 481-485. 2. M. Hüttl, H. Weinl, D. Laubrich Neue Wasserbehandlungsanlage für untertägige Sanierung des Grubenfeldes Ronneburg, WLB Wasser, Luft, Boden Zeitschrift für Umwelttechnik 11-12, 2002 pp. 35-37. 3. G. Kießig Erfahrungen aus realisierten Wasserbehandlungsvorhaben bei Wismut Tagungsband: 7. Wismut – Workshop of Water and sludge treatment, Konventionelle und innovative Lösungen 24-26.09.1997, Chemnitz. 4. H. Nisbet Wetland filtration research at ERA Ranger mine Wetland research in the wet dry tropics of Australia Workshop Jabiru NT 22-24 March 1995 p 165-72 Ed. by CM Finlayson.

582


5. St. Shinners An Overview of the application of constructed wetland Filtration at ERA Ranger mine, National engineering conference Darwin, 21-24 April 12996, Reprints of Papers, pp.367-378. 6. F. Glombitza, E. Janneck, F. Karnatz, G. Kießig, A. Küchler, M. Hüttl, Die Anwendung naturnaher Wasserbehandlung am Beispiel der Pilotanlage Pöhla. Natural Attenuation, neue Erkenntnisse, Konflikte, Anwendungen, Resümee und Beiträge zum 2. Symposium Natural Attenuation 07-08.12.2000 Ed. DECHEMA Gesellschaft für chem. Technik e.V. 2001 pp. 220-21, ISBN 3-89764-021-1. 7. W. Heymel Prinzipien und Methoden der technischen Uranfällung Ed. Wismut 1963. 8. Lovley D.R., Phillips E. J. P., Gorby Y.A., Landa E.R., (1991) Microbial reduction of uranium, Letters To Nature, Vol. 350, 4 April, 413-416. 9. M. Tsezos Engineering Aspects of Metal binding by biomasses in: Microbila minerla Recovery Eds.: H.L. Ehrlich, C. L. Brierly, Mc Graw Hill Inc., 1990, pp. 325-339, ISBN 0-07-007781-9.

583


Geomicrobiological risk assessment of abandoned mining sites K. Bosecker, G. Mengel-Jung and A. Schippers Federal Institute for Geosciences and Natural Resources (BGR), Section Geomicrobiology, Stilleweg 2, D 30655 Hannover, Germany e-mail: [email protected], A. [email protected] Abstract Some microorganisms colonize mineral surfaces, leading to acidification of the water draining from the mine or waste heap (AMD/ARD) and to contamination of the water with heavy metals. Depending on the mineral composition of the mine tailings and waste rock and on the environmental conditions (temperature, humidity, oxygen availability) the microbial activity can cause major environmental problems. Acid mine drainage from mine wastes may be predicted by investigating the site for the presence of metal-mobilizing bacteria (e.g. iron-oxidizing acidophiles, sulfur-oxidizing acidophiles, sulfur-oxidizing neutrophiles, acidophilic heterotrophs), quantifying the various indigenous populations, and measuring their leaching activity. We have studied mine waste dumps in a number of countries – with different metal contents and in different climate zones. Because bacterial leaching occurs mostly in sulfide ores, research has been concentrated on this kind of sulfide waste dump in Zimbabwe, Namibia, Bolivia, Peru, Brazil, Chile, Cuba and Kazakhstan. Metal-mobilizing bacteria were identified in all of the samples. The efficiency of the isolated bacteria and their significance for the mobilization of toxic substances from the waste dumps were determined. Possibilities for permanently minimizing environmental contamination of mining areas are shown. Keywords: acid mine drainage, mining waste, environmental risk potential, quantitative ecology, iron-oxidizing acidophiles, sulfur-oxidizing acidophiles, sulfur-oxidizing neutrophiles, acidophilic heterotrophs 1.

INTRODUCTION The mining and processing of metals creates spoil and waste heaps everywhere in the world. Some of the material in these heaps, which often have a volume of several million cubic meters, contains low concentrations of metallic ore. Sometimes the metal concentration may be remarkably high in the ore. Remediation measures are seldom carried out on these waste heaps to integrate them into the landscape. As a rule – at least in former times – no measures are taken when the mine is shut down and the waste heaps are left to be affected by weathering. Rainwater and residual water in the heaps dissolves the metals and sulfuric acid is produced. The metals and acid are transported into the surface water and groundwater, a process known as acid-mine drainage (AMD) or acid-rock drainage (ARD). Mobilization of the metals and generation of acid in the waste result 585


from chemical and microbiological oxidation. The conditions under which bacteria mobilize metals and other toxic substances are fairly well understood [1-4], but the extent to which this occurs at abandoned mining sites is not very well known. The bacterial activity depends on the mineral composition of the waste material, the sulfide, oxide or carbonate content and on the environmental conditions, such as pH, temperature, moisture content, and the availability of oxygen. In principle, the microbial leaching processes used at an industrial scale to recover metals from low-grade ores, refractory ores or concentrates, also occur in old mine dumps and at abandoned mining sites [5]. AMD is the most severe environmental problem the mining industry is faced with. It leads to contamination of the surface water and groundwater with toxic metals and degrades the groundwater quality downstream from the heaps and dumps. In the most extreme case reported, the Richmond Mine of the Iron Mountain, California, AMD/ARD contained metal concentrations as high as 200 g/L, sulfate concentrations as high as 760 g/L and had a negative pH as low as -3.6 [6]. For this reason, the Federal Ministry for Economic Cooperation and Development (BMZ) commissioned BGR to study bacterial mobilization of metals in mine waste heaps and to find ways to minimize groundwater contamination in order to improve the water supplies of the local population. During the past three years BGR has studied mine waste dumps in a number of countries – dumps with different metal contents and in different climate zones. Because bacterial leaching occurs mostly in sulfide ores, research has been concentrated on this kind of waste dump. Mining wastes in Bolivia, Brazil, Chile, Cuba, Kazakhstan, Namibia, Peru and Zimbabwe have been studied. The objective was to obtain enough information to predict the contamination load on the basis of the type of ore and the climatic conditions and to determine the conditions needed to permanently minimize contamination of surface water and groundwater. The solutions to the problems will be adapted to the economic and social conditions in the developing countries. The developing countries are to be given the basis for treatment of their waste heaps. 2.


2.1 Sampling The sampling sites, the number of samples and the respective climate conditions are given in Table 1. Table 1. Sampling sites, number of samples and climate conditions Country Bolivia Brazil Chile Cuba Kazakhstan

Mines 8 2 3 1 1 2 2

Samples 43 5 12 4 6 4 20

Namibia Peru Zimbabwe (Total)

1 7 4 (31)

16 31 24 (165)

586

tropical tropical tropical subtropical subtropical tropical temperate warm tropical tropical tropical

Climate cold warm warm continental warm warm continental

semi-humid semi-humid arid semi-humid semi-humid semi-arid semi-arid

warm cold warm

semi-arid humid semi-arid


After consultation with the mining geologists, samples were taken preferentially from sulfidic waste heaps more than five years old. Depending on the local conditions and the accessibility of the waste heaps, mostly the lower slopes of heaps were chosen for sampling. In some cases, additional sampling was carried out at flotation tailings ponds. After removing the upper 10-20 cm of the surface layer, approximately 250-300 g of waste material with a particle size less than 25 mm was collected in 125 mL sterile plastic jars. Before transport to the laboratory in Hannover, the jars were kept at ambient temperature, and from time to time the screw lids were carefully loosened for sufficient aeration. For transport by air, the jars were kept in the hand luggage and taken into the cabin. Special permits were needed to pass the security checks without the samples being x-rayed. In the extreme case, it was three weeks until processing was started in the laboratory. 2.2 Analytical measurements Five grams of each sample was suspended in 13.5 mL of 1 M KCl. The fluid was shaken for 5 min at 130 rpm/min at room temperature. The pH was measured in the supernatant after 2 hours and after 24 hours. Portions of the samples were dried to a constant weight using a Sartorius MA50 moisture analyzer. After grinding of the samples, major and trace elements were analyzed by x-ray fluorescence spectrometry (XRF). On-site, pH-value and concentrations of iron(III) and copper in seepage water from waste rock heaps were measured using analytical test strips (Merck). 2.3 Detection and quantification of microbial groups related to metal mobilisation To detach bacteria from the surface of the solid material, 10 g of fresh sample was placed in 100 mL of Leathen medium [7] without iron and shaken at 130 pm for 2h at room temperature. After settling of the solid phase, the supernatant was used for total- and viable-cell counting. The total number of cells was determined by acridine orange direct count using epifluorescence microscopy. The number of the viable cells of acidophilic sulfur-oxidizing, acidophilic iron-oxidizing and moderately acidophilic autotrophic bacteria was quantified by the "most probable number technique" (MPN), using serial 10fold dilutions in three tubes containing the suitable growth medium. The tubes were incubated on a shaking table at 120 rpm for 4-6 weeks in the dark at 30°C. Growth of acidophilic sulfur–oxidizing bacteria was indicated by high acid production in Starkeymedium [8]. Acidophilic iron-oxidizing bacteria were considered to be present if the Leathen-medium [7] became reddish-brown. Growth of moderately acidophilic autotrophic bacteria was indicated by a decrease in pH of more than one pH unit using the growth medium described by Matin and Rittenberg [9]. The presence of bacteria was demonstrated by microscopy. Acidophilic and neutrophilic heterotrophic microorganisms were counted by serial 10-fold dilution and spread-plating on agar as described by Harrison [10] and on R2Aagar [11], respectively. The agar plates were incubated in the dark at 30°C for 4-6 weeks, depending on when the number of colonies remained constant. 2.4 Calorimetric measurements The bioleaching activity of the microorganisms was determined by a microcalorimetric technique [12-17]. Heat is produced in mine waste heaps due to the oxidation of pyrite. Oxidation of pyrite to Fe(III) and sulfate has a reaction energy of – 1546 kJ/mol. The measured heat output correlates with the number and activity of the 587


leaching bacteria in the sample and the rate of pyrite oxidation can be calculated from these values. Heat output due to chemical oxidation was measured in control experiments. The calorimetric measurements were carried out in the laboratory of W. Sand, Department of Microbiology, University of Hamburg, Germany. 3.

RESULTS AND DISCUSSION The results for waste rock and tailings material from the different countries are presented separately in Tables 2-7. Large differences between waste rock and tailings material were observed with respect to the concentrations of the main metals. The metal content in waste rock material is generally higher than that in tailings material. Iron is the dominant metal (Tables 2 and 3). The waste rock from Bolivia and Peru contained very high concentrations of the listed main metals, and high contents of Cu and Zn were also detected in material from other countries (Table 2). Table 2. Main metals [g/kg] in waste rock Bolivia Brazil Chile Cuba Kazakhstan Namibia Peru Zimbabwe

As 6.2 4.1

Proceedings of the International Biohydrometallurgy Symposium IBS 2003, Athens, Greece - Vol I

Proceedings of the International Biohydrometallurgy Symposium IBS 2003, Athens, Greece - Vol II

Proceedings of the 24th international applied geochemistry symposium, vol 2

Proceedings of the 24th international applied geochemistry symposium, vol 1

Paideia: The Ideals of Greek Culture - Vol I: Archaic Greece - The Mind of Athens

Proceedings of Liverpool Singularities - Symposium I

Configurational Mechanics: Proceedings of the Symposium on Configurational Mechanics, Thessaloniki, Greece, 17-22 August 2003

Advances in Spatial and Temporal Databases: 8th International Symposium, SSTD 2003, Santorini Island, Greece, July 24 - 27, 2003. Proceedings

Frontiers of Fundamental Physics : Proceedings of the Sixth International Symposium

Frontiers of Applied Mathematics: Proceedings of the 2nd International Symposium

Frontiers of Applied Mathematics: Proceedings of the 2nd International Symposium

Computer and Information Sciences -- ISCIS 2003: 18th International Symposium, Antalya, Turkey, November 3-5, 2003, Proceedings

Ciba Vision Symposium: 16th Symposium, Osaka-tokyo, November 2003 Proceedings (Ophthalmologica Vol 218 Supplement 1)

High Performance Computing: 5th International Symposium, ISHPC 2003, Tokyo-Odaiba, Japan, October 20-22, 2003, Proceedings

Isocrates I (The Oratory of Classical Greece, vol. 4)

Quantum theory and symmetries. Proceedings of the 3rd international symposium

Proceedings of the Kommos Symposium, Scripta Mediterranean Vol. VI

Proceedings of the Ninth International Symposium on Cyclodextrins

Lepton and photon'01: proceedings of the XX International Symposium

New Panorama of Animal Evolution: Proceedings of the XVIII International Congress of Zoology, Athens, Greece, September, 2000

Artificial Neural Networks - ICANN 2006: 16th International Conference, Athens, Greece, September 10-14, 2006, Proceedings, Part I

Computer mathematics : proceedings of the Sixth Asian Symposium (ASCM 2003), Beijing, China, 17-19 April 2003

Foundations of Intelligent Systems: 14th International Symposium, ISMIS 2003, Maebashi City, Japan, October 28-31, 2003, Proceedings

Practical Aspects of Declarative Languages: 5th International Symposium, PADL 2003, New Orleans, LA, USA, January 13-14, 2003, Proceedings

Patras Logic Symposium: Proceedings

Distributed Computing: 12th International Symposium, DISC'98, Andros, Greece, September 24 -26, 1998, Proceedings

Zeolite Chemistry and Catalysis Proceedings of an International Symposium Prague

FME 2003: Formal Methods: International Symposium of Formal Methods Europe. Pisa Italy, September 8-14, 2003, Proceedings

FME 2003: Formal Methods: International Symposium of Formal Methods Europe. Pisa Italy, September 8-14, 2003, Proceedings

Proceedings of the Third Scandinavian Logic Symposium

Admixtures for Concrete - Improvement of Properties: Proceedings of the International RILEM Symposium (RILEM Proceedings 5)

Proceedings of the International Biohydrometallurgy Symposium IBS 2003, Athens, Greece - Vol I