摘要:

THE ROYAL SOCIETY OF CHEMISTRY Journal of Materials Chemistry Scientific Editor Staff Editor Professor Anthony R. West Mrs. Janet M. Leader Department of Chemistry The Royal Society of Chemistry University of Aberdeen Thomas Graham House Meston Walk Science Park Aberdeen AB9 2UE, UK Cambridge CB4 4WF, UK Assistant Editor: Mrs. F. J. O’Carroll Editorial Secretary: Miss J. E. Chapman Materials Chemistry Editorial Board Anthony R. West (Aberdeen) (Chairman) C. Richard A. Catlow (London) David A. Rice (Reading) David A. Dunmur (Sheffield) Rodney P. Townsend (Bebington) H. Monty Frey (Reading) Allan E. Underhill (Bangor) John W. Goodby (Hull) Graham Williams (Swansea) John D. Wright (Canterbury) International Advisory Editorial Board M. A.Alario- Franco (Madrid) D. Kohl (Aachen) K. Bechgaard (Copenhagen) M. Lahav (Rehovot) J. D. Birchall (Runcorn) A. J. Leadbetter (Daresbury) D. Bloor (Durham) P. M. Maitlis (Sheffield) A. K. Cheetham (Oxford) J. S. Miller (Wilmington) E. Chiellini (Pisa) P. S. Nicholson (Hamilton) M. G. Clark (Wembley) M. Nygren (Stockholm) P. Day (Grenoble) V. Percec (Cleveland) D. Demus (Halle) C. N. R. Rao (Bangalore) B. Dunn (Los Angeles) M. Ratner (Evanston) W. J. Feast (Durham) J. Rouxel (Nantes) A. Fukuda (Tokyo) R. Roy (University Park, PA) D. Gatteschi (Florence) J. L. Serrano (Zaragoza) A. M. Glass (Murray Hill) J. N. Sherwood (Glasgow) J. B. Goodenough (Austin) J. Simon (Paris) G. W. Gray (Poole) J. F. Stoddart (Sheffield) A. C. Griffin (Cambridge) S.Takahashi (Osaka) S-i. Hirano (Nagoya) G. J. T. Tiddy (Bebington and Salford) P. Hodge (Manchester) 6. J. Tighe (Birmingham) H. lnokuchi (Okazaki) Yu. D. Tretyakov (Moscow) W. Jeitschko (Munster) R. J. P. Williams (Oxford) 0. Kahn (Orsay) R. Xu (Changchun) Journal of Materials Chemistry (ISSN 0959-9428) is published six times a year by The Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge CB4 4WF, UK. All orders accompanied with payment should be sent directly to The Royal Society of Chemistry, Turpin Transactions Ltd., Blackhorse Road, Letchworth, Herts SG6 1 HN, UK. NB Turpin Transactions Ltd., distributors, is wholly owned by The Royal Society of Chemistry. 1991 Annual subscription rate EC (inc.UK) f175.00, USA $395.00, Rest of World €1 95.00. Customers should make payments by cheque in stirling payable on a UK clearing bank or in US dollars payable on a US clearing bank. Air freight and mailing in the USA by Publications Expediting Inc., 200 Meacham Avenue, Elmont, NY 11 003. USA Postmaster: send address changes ta J’uurnal of Materials Chemistry, Publications Expediting Inc., 200 Meacham Avenue, Elmont, NY 11 003. Second Class postage paid at Jamaica, NY 11 431. All other dispatches outside the UK by Bulk Airmail within Europe, Accelerated Surface Post outside Europe. PRINTED IN THE UK. @ The Royal Society of Chemistry, 1991. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form, or by any means, electronic, mechanical, photographic, recording, or otherwise, without the prior permission of the publishers.Professor A. R. West, Scientific Editor Mrs. J. M. Leader, Staff Editor Tel.: Aberdeen (0224) 27291 8 Tel.: Cambridge (0223) 420066 Fax: (0224) 272938 E-Mail (JANET): Telex: 73458 UNIABN G RSCl @UK.AC.RL.GB Fax: (0223) 420247 or 423623 Telex: 818293 ROYAL G INFORMATION FOR AUTHORS The Royal Society of Chemistry welcomes submission of manuscripts intended for pub- lication in two forms, Articles and Materials Chemistry Communications. These should describe original work of high quality dealing with the synthesis, structures, properties and applications of materials, particularly those associated with advanced technology.Artides Full papers contain original scientific work that has not been published previously. How- ever, work that has appeared in print in a short form such as a Materials Chemistry Com- munication is normally acceptable. Four copies of Articles including a top copy with figures etc. should be sent to The Editor, Journal of Materials Chemistry, The Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge CB44WF, UK. Materials Chemistry Communications Materials Chemistry Communications contain novel scientific work in short form and of such importance that rapid publication is war-ranted. The total length is rigorously restric- ted to two pages of the double-column A4 format.The manuscript will be returned for reduction if this length is exceeded. For a Communication consisting entirely of text and ten references, with no figures, equations or tables, this corresponds to approximately 1600 words plus an abstract of up to 40 words. Submission of a Materials Chemistry Com- munication can be made either to The Editor, Journal of Materials Chemistry, The Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge CB4 4WF, UK, or viaa member of the Interna- tional Advisory Editorial Board. In the latter case, the top copy of the manuscript includ- ing any figures etc., together with the name of the person to whom the Communication is being submitted, should be sent simultan- eously to the Editor at the Cambridge address. Authors may wish to contact the Board mem- ber to ensure that he is available to arrange review of the manuscript within reasonable time. In order to avoid delay in publication, proofs of Communications are not sent to authors unless this is specifically requested. Full details of the form of manuscripts for Articles and Materials Chemistry Communi- cations, conditions for acceptance etc. are given in issue number one of Journal of Materials Chemistry published in January of each year, or may be obtained from the Staff Editor. There is no page charge for papers published in Journal of Materials Chemistry. Fifty reprints are supplied free of charge. Any author who is publishing in Journal of Materials Chemistry for the first time is entitled to a free copy of the issue in which the paper appears.

ISSN:0959-9428    

DOI:10.1039/JM99101FX017

出版商:RSC    

年代:1991

数据来源: RSC

ISSN:0959-9428    

DOI:10.1039/JM99101BX019

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ISSN 0959-9428 JMACEP(5) 715-901 (1991) Journal of Materials Chemistry Synthesis, structures, properties and applications of materials, particularly those associated with advanced technology CONTENTS 715 FEATURE ARTICLE. Charge fluctuations and an ionic-covalent transition in La, -,Sr,CuO, J. B. Goodenough, JS. Zhou and K. Allan 725 Behaviour of the adsorbed C1' intermediate in anodic C1, evolution at thin-film RuO, surfaces B. E. Conway, G. Ping, A. De Battisti, A. Barbieri and G. Battaglin 735 Chemical reduction of FeC1,-graphite intercalation compounds with potassium-napthalene complex in tetrahydrofuran A. Messaoudi, M. Inagaki and F. Beguin 739 Porous chromia-pillared a-zirconium phosphate materials prepared via colloid methods P. Maireles-Torres, P.Olivera-Pastor, E. Rodriguez-Castellon, A. Jimenez-Lopez and A. A. G. Tomlinson 747 New perovskite phases in the systems Li,O-(Nb,O,, Ta,05)-Zr0, M. E. ViUafuerte-Castrejon, C. Kuhliger, R.Ovando, R. I. Smith and A. R. West 751 Flash calcination of kaolinite: Mechanistic information from thermogravimetry R. C. T. Slade, T. W. Davies and H. Atakiil 757 Synthesis and thermal behaviour of mesomorphic Cu" Schiff's base complexes K. P. Reddy and T. L. Brown 765 Columnar mesophases of cyclic trimers of disubstituted acetylenes C. Pugh and V. Percec 775 Preparation and second-harmonic generation properties of tris(pyrocatecholato)stannate(Iv)compounds C. Lamberth, J. C. Macheil, D. M. P. Mingos and T. L. Stolberg 781 Low-temperature synthesis of perovskite solids LaMO, (M =Ni, Co, Mn) via binuclear complexes of compartmental ligand N,N'-bis(3- carboxysalicy1idene)thylenediamine S.P. Skaribas, P. J. Pomonis and A. T. Sdoukos 785 Surface segregation of Sr in doped MgO. Comparison between X-ray photoelectron spectroscopy and atomistic ionic model simulations L. L. Cao, R. G. Egdell, W. R. Flavell, K. F. Mok and W. C. Mackradt 789 Preparation of submicrometre spherical oxide powders and fibres by thermal spray decomposition using an ultrasonic mist atomiser T. Ogihara, T. Ookura, T. Yanagawa, N. Ogata and K. Yoshida 795 Electrodeposition and study of multilayered Co/Cu structures V. M. Fedosyuk, 0.I. Kasyutich and N. N. Kozich 799 Preparation and thermal properties of new liquid crystals with a 2-benzoyloxy-2,4,6-cycloheptatrien-1-one core A.Mori, N. Kato, H. Takeshita, M. Uchida, H. Taya and R.Nimura 805 X-Ray photoelectron and Mossbauer spectroscopies of a binary iron phosphate glass L. Armelao, M. Bettinelli, G. A. Rizzi and U. Rum 809 Dopant effects on the response of gas-sensitive resistors utilising semiconducting oxides D. E. Williams and P. T. Moseley 815 Structural phase transition of VO,(B) to VO,(A) Y. Oka, T. Yao and N. Yamamoto 819 Docosanoyl itaconate/ 1-docosylamine alternate-layer Langmuir-Blodgett films: Polymerisation, pyroelectric properties and infrared spectroscopic studies J. Tsibouklis, M. Petty, Y-P. Song, R. Richardson, J. Yarwood, M. C. Petty and W. J. Feast 827 Crystal structure and pressure dependence of electrical conductivities of [(CH,),N][Pt(dmit),], A.Kobayashi, A. Miyamoto, H. Kobayashi, A. Clark and A. E. Underhill 831 Thermochromism and solvatochromism of bis[1,2-bis(3,4-di-n-alkoxyphenyl)ethanedione dioximato]nickel(~~) complexes K. Ohta, H. Hasebe, M. Moriya, T. Fujimoto and I. Yamamoto 835 Alkyl-substituted oligothiophenes: Crystallographic and spectroscopic studies of neutral and doped forms S. Hotta and K. Waragai 843 Synthesis and phase behaviour of mesomorphic transition-metal complexes of alkoxydithiobenzoates. Crystal and molecular structure of three metal alkoxydithiobenzoates H. Adams, A. C. AlbeNz, N.A. Bailey, D. W. Bruce, A. S. Cherodian, R. Dhillon, D. A. Dunmur, P. Espinet, J. L. Feijoo, E.Lalinde, P. M. Maitlis, R. M. Richardson and G.Ungar 857 Linear dichroism of mesomorphic transition-metal complexes of alkoxydithiobenzoates D. W. Bruce, D. A. Dunmur, S. E. Hunt, P. M. Maitlis and R. Orr 863 Improved method for the determination of the oxygen content in YB~,CU,O,-~ H. Vlaeminck, H. H. Goossens, R. Mouton, S. Hoste and G. Van der Kelen 867 Magnetic frustration, spirals and short-range order in Cr,-Fe, -,V04-I solid solutions J. P. Attfield, A. K. Cheetham, D. C. Johnson and T. Novet 875 27Al Nuclear magnetic resonance spectroscopy investigation of thermal transformation sequences of alumina hydrates. Part 2.-Boehmite, y-Al00H R. C. T. Slade, J. C. Southern and I. M. Thompson 881 Electrochromism in titanyl and vanadyl phthalocyanine thin films J. Silver, P.Lukes, P. Hey and M. T. Ahmet ~ ~~~~~~ MATERIALS CHEMISTRY COMMUNICATIONS 889 Oriented nucleation of CaCO, from metastable solutions under Langmuir monolayers J. B. A. Walker, B. R. Heywood an d S. Mann 891 Effect of nickel oxide on the sintering characteristics and phase stability of zirconia-12 mol% ceria ceramics S. C. Br adwell, A. Kapusta and S. J. Milne 893 Diol-based sol-gel system for the production of thin films of PbTiO, N. J. Phillips and S. J. Milne 895 Mesomorphic behaviour of some cross-linked derivatives of all-conjugated main-chain polyesters F. Navarro G6mez and J . L. Serrano Ostariz 897 Arylaldehyde and arylketone hydrazones as a new class of amorphous molecular materials K. Nishimura , T. Kobata, H .Inada and Y. Shirota 899 Book Reviews: I. C. Callagban; L. Shields, P. A. Holmes; J. M. G. Cowie; P. D. Lickiss Note: Where an asterisk appears against the name of one or more authors, it is included with the authors’ approval to indicate that correspondence may be addressed to this person. Spelling of Sulfur The new (1990) edition of IUPAC‘s ‘Nomenclature of Inorganic Chemistry’ contains a table of IUPAC-approved names ‘for use in the English language’. These include ‘caesium’, ‘aluminium’, and ‘sulfur’ (spellings as given here). There is increasing use of the ‘f’ rather than the ‘ph’ spelling for sulfur in English publications, in particular the English language versions of IS0 and European standards, and those British Standards which implement IS0 standards verbatim. Furthermore, there is no good etymological basis for preferring the ‘ph’ spelling. In view of these considerations, the Royal Society of Chemistry’s Nomenclature Committee has recently recommended that RSC change to using the ‘f’ spelling in all its publications. This recommendation will be implemented for RSC‘s primary journals in 1992. Alan McNaught Manager, RSC Journals

ISSN:0959-9428    

DOI:10.1039/JM99101FP055

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

Cumulative Author Index 1991 Abe M., 483 Adams D. M., 487 Adams H., 843 Cook M. J., 121, 703 Copley R. C. B., 583 Cowie J. M. G., 899 Hague J. D., 507 Hall G. P., 685 Hardin S., 423 Kojima M., 559 Kordas G., 97, 175, 181 Korgul P., 239 Nakanishi Y., 357 Nakano C., 37 Nakayama A., 707 Adams J. M., 43 Cox P. A., 51,451 Harrison K. J., 121, 703 Kozich N. N., 795 Nalini V., 201 Adams P. N., 141 Ahmet M. T., 881 Alagna L., 319 Albeniz A. C., 843 Alcock N. W., 569 Allan K., 715 Allen Sir Geoffrey, 1 Cracknell S. J., 703 Crayston J. A., 381 Crennell S. J., 113, 297 Currie D. B., 295 Dalas E., 473 Daniel M. F., 121 Daolio S., 191 Harrison W. T. A,, 153 Hasebe H., 831 Hawley M., 525 Hayter J. B., 181 Hey P., 881 Heywood B. R., 889 Higashi N., 365 Krok F., 705 Krongauz V., 331 Kuhliger C., 747 Kuhn H.H., 525 Kurmoo M., 51 Kusabayashi S., 169 Kusumoto T., 707 Narayan K. S., 479 Nardella A., 129 Navarro Gomez F., 895 Nebesny K. W., 213 Nimura R., 799 Nishihata Y., 169 Nishimura K., 897 Allen G. C., Alonso P. J., Annen M. J., 69, 73 197 79 Davey A. P., Davies T. W., Davis M. E., 245 361, 751 79 Hirao M., 293 Hirst P. R., 281, 429 Hitterman R. L., 175 Kuzuya M., 387 Lacroix P., 475 Lalinde E., 843 Niwa M., 365 Noguchi A., 387 Novet T., 867 Apblett A. W., 143 Aragon-Santamaria P., Arhancet J. P., 79 409 Day P., 597 De Battisti A,, Demus D., 347 191, 51 1, 725 Hiyama T., 707 Hoang M., 423 Hodes G., 339 Lamberth C., 583, 775 Lee C. K., 149, 595 Lee G.R., 381 Nowinski J., 705 OBrien P., 139 Ocaiia M., 87 Armelao L., Armes S.P., 805 525 Dennington N. R., Dent A. J., 103 663 Hoffman D., 87 Hoke J. B., 551, 701 Lee M., 611 Leech D., 629 OConnor C. J., 555 OConnor P., 103 Atakul H., 751 Attfield J. P., 867 Audebert P., 699 Desiraju G. R., 201 Dhillon R., 843 Dickens P. G., 105, 137, 415 Holmes P. A,, 899 Homer J., 327 Hori K., 667 Legge C. H., 303 Le Lagadec R., 251 Lewis J., 485 Ogata N., 789 Ogihara T., 789 OHare D., 5I, 205, 69 1 Audiere J. P., 475 Dodd S. M., 11 Horner P. J., 271 Li H-X., 79 Ohashi Y., 667 Babu T. G. N., 677 Doi T., 169 Hoste S., 863 Lickiss P. D., 899 Ohta K., 831 Bailey N. A,, 843 Drache M., 649 Hotta S., 835 Liddell K., 239 Okay., 815 Ball R. G. J., Barbieri A,, Barboux P., 191, 725 681 105, 415 Dragone R., 531 Dray G.R., 485 Dunmur D. A., 251, 255, Howe S.D., 29 Howlin B., 29 Hudson M. J., 375 Livage J., 621, 681 Lo Jacono M., 129, 531 Lomas L., 475 Okamoto N., 591 Olivera-Pastor P., Olsson P-O., 239 3 19, 739 Barley S. H., Barrer R. M., 481 305 843, 857 Dunn B., 265 Hunt S. E., 251, 857 Hunter J. A., 709 Lukes P., 29, 881 Maceira-Vidan A., 409 Ookura T., 789 Orr R., 255, 857 Barron A. R., Bartlett P. N.. 143 569 Dyer A., 43 Eastwick-Field V. M., 569 Hursthouse M. B., Imaeda K., 37 139 Machell J. C., 583, 775 Mackrodt W.C., 785 Osawa M., 707 O’Sullivan T. P., 393 Battaglin G., 191, 725 Beery J. G., 525 Beguin F., 735 Benedetti A., 51 1 Benhamza H., 681 Bettinelli M., 437, 805 Bicelli L. P., 259 Binks J. H., 289 Birdwhistell T.L. T., 555 Eddy M. M., 223 Edge S., 103 Edwards D. J., 223 Egdell R. G., 63, 451, 489, Epstein A. J., 479 Espinet P., 843 Esteruelas M. A., 251 Faber J., 175 785 Inada H., 897 Inagaki M., 735 Ingletto G., 437 Inokuchi H., 37 Inomiya M., 483 Inoue M., 213 Inoue M. B., 213 Inversi M., 531 Irvine J. T. S., 147, 289 Madsen L., 503 Maffi S., 259 Maireles-Torres P., 3 19, 739 Maitlis P. M., 251, 255, 843, Maldotti A., 51 1 Male S. E., 69 Malitesta C., 259 Mandal K. C., 301 857 Ovando R., 747 Owen J. J., 113 Percec V., 217, 61 1, 765 Perry M. C., 327 Persson C., 577 Petty M., 819 Petty M.C., 819 Phillips N. J., 893 Pike G. A., 569 Blau W. J., 245 Block H., 709 Bond S. P., 327 Borrill C. J., 655 Feast W. J., 819 Fedosyuk V. M., 795 Feijoo J. L., 843 Fernando Q., 213 Ishikawa M., Ishikawa Y., Ito H., 387 Iwasawa N., 387 483 37 Mann S., 889 Manterfield M.M., Marcos M., 197 Marsden J. R., 251 255 Ping G., 725 Poliui S., 51 1 Polo-Diez L. M., 409 Pomonis P. J., 781 Boscolo Boscoletto A,, 191 Bouhaouss A,, 681 Bradwell S. C.. 891 Brambley D. R., 401 Brock T., 151 Brookes J., 691 Brown I. T., 69 Brown N. M. D., 469, 517 Brown T. L., 757 Bruce D. W., Bruce L. A,, 423 Bruce P. G., 705 Byrne H. J., 245 Cahen D., 339 Callaghan 1. C., 899 Campillos E., 197 Cao L. L., 785 Cardin D. J., 245 Carneiro K., 503 Catlow C. R. A., 233 Centonze D., 259 Chambers R. D., 59 Chatakondu K., 205 Cheatham L. K., 143 Cheetham A. K., 113, 223, 297, 867 Chen C-Y., 79 Chen X., 643 Cherodian A. S., 843 Cho C. G., 217 Christy A. G., 487 Chvatal Z., 59 Clark A., 827 Clarke J.H. R.. 487 Clement R., 475 Cogle T. J., 289 Conflant P., 649 Conway B. E., 725 251, 255, 843, 857 Fierro G., 531 Fish D. J., 677 Fitch A. N., 461 FitzGerald E. T., 51 Flavell W. R., 63, 451, 489, Flint C. D., 437 Follet-Houttemane C., 649 Formstone C. A., 51, 205 Forster R. J., 517, 629 Friend R. H., 485 Fujimoto T., 831 Fujita T., 559 Fujiyasu H., 357 Garzon F., 525 Gelder A., 327 Geus J. W., 539 Giatti A., 191 Gibb T. C., 23 Gibson V. C., 705 Gier T. E., 153 Gilbert A., 303, 481 Glasser F. P., 305 Golden M. S., 63, 489 Golden S. J., 63 Gonsalves K. E., 643 Goodby J. W., 5, 307 Goodenough J. B., 7 15 Goossens H. H., 863 Gottesfeld S., 525 Gourier D., 265 Greaves C., 17, 677 Green M. L. H., 205 Grey C. P., 113 Griesmar P., 699 Grimes R.W., 461 Grins J., 239 Groenendijk P. E., 539 Grossel M. C., 223 785 Jacob K. T., 477, 545 Jarman R. H., 113,297 Jeffries T., 555 Jiang M. R. M., 11 Jimenez-Lopez A., 319, 739 Jin J. Y., 457 Johnson B. F. G., 485 Johnson D. C., 867 Johnson O., 223 Johnstone R. A. W., 457 Jones A. C., 139 Jones M.N., 447 Jones R. G., 401, 673 Jorgensen J. D., 175 Josien F-A., 681 Judeinstein P., 621 Jutson J. A., 73 Kaduk J. A., 113, 297 Kaharu T., 145 Kakkar A. K., 485 Kamiya K., 387 Kane J., 447 Kapusta A., 891 Kasyutich 0.I., 795 Kathirgamanathan P., 103, Kato N., 799 Kawaguchi T., 387 Kawahashi N., 577 Kellar E. J. C., 331 Kemp J. P., 451 Khan M. S., 485 Kimura K., 293 Klein C. L., 555 Klinowski J., 709 Kobata T., 897 Kobayashi A., 827 Kobayashi H., 827 Kg11 P-O., 239 141 Maruyama Y., 37 Mateus C.A. S., 289 Mathews T., 545 Matijevic E., 87, 577 Matsubara H., 145 Matsuda T., 559 Matsushima R., 591 McKeown N. B., 121 McLean R. S., 479 McWhinnie W. R., 327 Messaoudi A., 735 Milburn G. H. W., 155 Miller J. S., 479 Miller Tate P. C., 401, 673 Milne S. J., 891, 893 Mingos D. M. P., 583, 775 Mitchell G. R., 303, 481 Miyamoto A., 827 Mohr K., 347 Mok K.F., 785 Mombourquette C., 525 Moon B. M., 97 Moore G. A., 175, 181 Moretti G., 129, 531 Mori A., 799 Mori T., 37 Moriya M., 831 Morris M., 43 Moseley P. T., 809 Motevalli M., 139 Mouton R., 863 Muramatsu H., 357 Murphy D. M., 583 Murray H. H., 551 Musicanti M., 129 Mutlu M., 447 Mutlu S., 447 Nakamura K., 707 Nakamura T., 357 Porta P., 129, 531 Postle S. R., 223 Powell A.V., 137 Powell H., 583 Prassides K., 597 Pressman H. A,, 429, 685 Pringle P. G., 569 Pugh C., 217, 765 Quill K., 141 Ramasesha S. K., 477 Rao C. N. R., 299 Rappaport M., 339 Rastomjee C. S., 451 Rayment T., 299 Reddy K. P., 757 Richardson R., 8 19 Richardson R. M., 121,843 Riello P., 511 Rizzi G. A., 805 Rodriguez-Castellon E., Rosenberg M. F., 447 Roser S. J., 121 Rosseinsky D. R., 487 Rosseinsky M. J., 597 Rotella F., 175 Russo U., 805 Sabbatini L., 259 Saito G., 37 Sakurai Y., 169 Salamon M. B., 181 Salmon L., 265 Sanchez C., 699 Sankar G., 299 Santos-Delgado M. J., 409 Sat0 K-i., 707 Savadogo O., 301 Schafer W., 347 Schwartz M.. 339 319, 739 1 Scolnik Y., 339 Sdoukos A. T., 781 Serrano J. L., 197, 895 Stacey J. M., 251 Stern E. W., 551, 701 Stevens E.D., 555 Thomson A. J., 121 Tilley R.J. D., 155 Tomlinson A. A. G., 3 19, Vlaeminck H., 863 Vochten R., 637 Volin K. J., 175 Wittmann F., 485 WongT.K., 643 Workman A. D., 375 Sherrington D. C., 151, 371 Shields L., 899 Shirota Y., 897 Silver J., 29, 881 Simmons J. M., 121 Sinclair D. C., 147 Singh N., 441 Skaribas S. P., 781 Slade R. C. T., 281, 361, Slaney A. J., 5 Slater P. R., 17 Smith R.I., 91, 747 Smyth M. R., 629 Sobry R., 637 Sola E., 251 Song Y-P., 819 Sonoda K., 483 Southern J. C., 563, 875 429,441, 563,685, 751, 875 Stobbe D. E., 539 Stolberg T. L., 775 Stucky G. D., 153 Takahashi S., 145 Takehara S., 707 Takematsu M., 365 Takenaka S., 169 Takeshita H., 591, 799 Takeuchi Y., 357 Tamatani A., 169 Tavakkoli K., 705 Taya H., 799 Taylor S.E., 393 Templeton-Knight R., 59 Terauchi H., 169 Thompson D. P., 239 Thompson I. M., 563, 685, Thompson W. C., 305 875 Tsibouklis J., 819 Turney T. W., 423 Twyman J. M., 205 Uchida M., 483, 799 Underhill A. E., 103, 141, Ungar G., 843 Vadgama P., 447 van Buren F. R., 539 Van der Kelen G., 863 van der Sluijs M., 503 van Dillen A. J., 539 Van Haverbeke L., 637 Vazquez C., 479 Verweij P. D., 371 Villafuerte-Castrejon M. E., 747 Vivien D., 265 739 503, 827 Vos J. G., 517, 629 Walker J. B. A., 889 Walsh J. R., 139 Waragai K., 835 Watanabe H., 483 Watkin D. J., 691 Waugh K. C., 709 Wedler W., 347 Weissflog W., 347 Weller M. T., 11, 295 West A. R., 91, 147, 149, 157, 163, 595, 747 West B. C., 281 Whitcombe M. J., 303 Whiteley R. H., 271, 655 Williams D. E., 809 Williams G., 331 Williams J.O., 663 Wiseman P. J., 205 Wright A. C., 663 Wright P. V., 507 Yamamoto I., 831 Yamamoto N., 8 15 Yanagawa T., 789 Yao T., 815 Yarwood J., 819 Yee G.T., 479 Yitzchaik S., 331 Yokoyama M., 293 Yoshida K., 789 You H. X., 469, 517 Zaba B. N., 503 Zambonin P. G., 259 Zaschke H., 347 Zhang X., 233 Zheng C., 163 Zheng Q., 611 Zhou J-S., 715 11 September 1-6 September 2-4 September 2-6 September 2-6 September 4-6 September 4-6 September 8-13 September9-11 September 9-12 September9-12 September 9-14 September 17-19 September 18-20 September 22-25 September 23-27 September 24 September 26-27 Conference Diary 1991 XVI International Symposium on Macrocyclic Chemistry (ISMC 1991) Sheffield, UK Dr Norma A.Stoddart, Departmemt of Chemistry, The University, Sheffield S3 7HF, UK. Tel.: (0742) 768555 Ex. 4522. FAX: (0742) 739826 Crystalline Structures and Defa in Ceramics Ljubljana, Yugoslavia M. Dmfenik, Institute Jozef Stefan, Yu-61001 Ljubljana, PO Box 100, Yugoslavia International Conference on Magnetism Edinburgh,UK ICM '91 Secretariat, 47 Belgrave Square, London SWlX 8QX International Conference on Pdymer-Sdid Interfaces Namur, Belgium Relations Publiques, conferenceICPSI 91, Rue de Bruxeues 53, B-5000 Namur, Belgium IWASES-I1(International Workshop on Auger Spectroscopy and Nectronic Structure) Lund, Sweden Dr C-0. Almbladh, Depattmemt ofTheoretical Physics, Sijlvegatan 14 A, S-22362 Lund, Sweden E-mail: COA@SELDC52.BlTNET. FAX: 46-(0)46-104710 Polymer Stabilization: Mechanisms and Applications Birmingham, UK Profe.ssor N.S.Allen, Department of Chemistry, Manchester Polytechnic, John Dalton Building, Chester Street, Manchester M15GD International Symposium: Zeolite Chemistry and Catalysis Prague, Czechoslovakia Dr B. Wichterlov4 (Zeolite Chemistry and Catalysis), The J. Heyrovsky Institute of Physical Chemistry and Electrochemistry, Dolejskova 3,182 23 Prague 8, Czechoslovakia Tel.: (422) 858 412Q (422) 815 35%. Telex: 122018 ATOM C. FAX: (422) 858 4569. 6th Biennial Polymer Conference Leeds,UK Meetings office, The Institute of Physics, 47 Belgrave Square, London SWlX SQX, UK ECOSS 12: 12th European Conference on Surface Science Stockholm,Sweden CONGREX, Att: ECOSS-12, Box 5619, S-114 86 Stockholm,Sweden 18th International Symposium on Gallium Arsenide and Related Compounds Seattle,Washington,USA VIPMeetings and Conventions, Awl: GaAs & Related Compounds Symposium, 1515 Palisades Dr., Suite I, Pacific Palisades, California 90272, USA.Tel.: (213) 459-4691. FAX: (213) 459-0605. Physics of Polymer Networks Ballenstedt, Germany Organizing Committee, Physics of Polymer Networks, Technical University of Merseburg, Dept of Physics, Merseburg, DDR-4200, Germany Interfacial Phenomena in Composite Materials '91 Leuven, Belgium Caroline Sumner, Conference Manager, Interfacial Phenomena in Composite Materials '9 1, Butterworth-Heinemann Ltd, Linacre House, Jordan Hill, Oxford OX2 8DP, UK Recyding of Polymers -lla,spainW. Heitz, Philip-Universitiit Marburg FB 14 Physikalische Chemie, Polymere, Hans-Meerweinstrasse, D-3550 Marburg, Germany 5th Sensors and their Applications Conference Edinburgh,UK Meetings Office, The Institute of Physics, 47 Belgrave Square, London SWlX SQX, UK 7th International Symposium on Electrets Potsdam, Germany ProfessorMeV Geiss, Academy of Sciences, Institute of Polymer Chemistry 'Erich Correns', Kantstrasse 55, 1530 TeltowSeehof, Germany Firth Symposium: Clusters, Partides and Catalysts Sheffield, UK Professor P.M. Maitlis FRS,Department of Chemistry,University of Sheffield, Sheffield S3 7HF Vth Colloquium on Biomaterials Aachen,Gemany Dipl.€hem. Dirk Andeheiden, Klinikum der RWTH-Aachen, Institut fiir Pathologie, Pauwelsstrasse 30,5100 Aachen, Germany.Tel.: 0241/8088392; FAX:49-241-8089284 ... 111 September30-October 2 September 30- October2 Septepber 30-October 5 October 2-4 october6-11 October7-8 October7-1 1 October 13-17 October 14-18 October 14-18 October 20-26 October 23 -25 October 28- November 2 November 4 November 4-8 November 1 1 -15 November 17-22 EumnsorsV Rome, Italy Professor A. D'Amico, Universita' di Rana "TOR VERGATA, Dipartimento di Ingegneria Elettrmica, Via Orazio Raimondo,140173 Rma, Italy Speciality Polymers '91 Mainz, Germany Conference Organizer, SP'91, Butterworth Scientific Ltd,PO Box 63, Westbury House, Bury Street, Guildford, Surrey GU2 SBH, UK Mammdecule-Metal Complexes IV Siena, Italy Professor Roland0 Barbucci, Dipartimento di Chimica, Universita di Siena, Piano dei Mantellini 44,53 100 Siena, Italy The American Assodation of Crystal Growth Emstern Region Fourth Conference on Crystal Growth Atlantic City, NJ, USA Ed Porbansky.AIlidSignal, Inc,PO Box 1021, CTC-1, Momstown. NJ 07%2, USA. Tel.:(201) 455-6879 Sixth International Workshop on Glasses and Ceramics from Gels Seville, SpainDr Luis Esquivias Fedriani (Workshop Chairman), Deparuunento de Estructura y Pqnedades de 10s Materiales, Apartado, 40-1 1510 Puerto Real (cbdiz),Spain Superconductivity: Materials, Physics and Applications: ICMAS-91 Paris,France IITI-International,40Pmenade Marx Domy, F-93460 Goumay-sur-Mame, France.Tel.: 33-1 45921771; FAX: 33-1 45929215: Telex: 250303 (Att.iitt) IV International TopicalMeeting on Optics of Liquid Crystals (Optical Properties and Applications of Liquid Crystals and Organic Materials) Cocoa Beach, Florida,USA Donna Wilson,OW '91, Meeting Coordinator, CREOL,12424 Research Parkway, Suite 400, Orlando, FL 32826, USA Tel.: (407)658-6834. FAX: (407)658-6880. Unconventional Photoactive Solids (UPS) Symposium on Molecular Systems Okazaki, Japan Tadadri Mi&, Institute for Molecular Science, Myodaiji, Okazaki 444,Japan ECASIA 91 (4th European Conference on Applications of Surface and Interface Analysis) Budapest, Hungary ECASIA 91, MTA ATOMIKI, Pf. 51, H-4001 Debrecen, Hungary. Tel.: (36)-52-16181. Telex: 72210 (atam h).FAX: (36)-52-16181 Olefin and Vinyl Polymerization and F'unctionalization Hangzhou, Qlina Professor Kun-Yuan Qiu, Secretary-General of Symposium 91 Hangzhou, Chemistry Department, Peking University, Beijing 100871, China Solid State Ionics 8th International Conference Lake Louise, Canada DrP. S. Nicholson, McMaster University, JHE-249, Hamilton,Ontario, Canada L8S 4L7. Tel.: (416) 529-7575. FAX: (416) 529-5994 Pdymers in a Marine Environment III London, UK Rhian Bufton,Conference Organiser, The Institute of Marine Engineers, The Memorial Building, 76 Mark Lane, London EC3R 7JN, UK Europhysics Conference on Macromolecular Physics: Polymer Networks Wemigerode, Germany S. Wattewig, 'Carl Schorlemmer' Technical University Leuna-Merseburg, Department of Physics, DDR-4200 Merseburg, Germany Copenhagen Opt(r)ode Workshop Roskilde, Denmadc Mrs Hanne Jensen, MODECS. Copenhagen Opt(r)ode Workshop.RIS0 National Laboratorys MILKEM, DK-4000Roskilde, Denmark.Tel.:int. + 42 37 12 12 Ext. 5333. FAX: int.+ 42 37 0264 EMRS 1991Fall Meeting: 15th International Symposium on the Scientific Basis for Nuclear Waste Management Strasbourg, France P. Sif€ert,EMRS 1991 Fall Meeting, C. T. T., B. P. 20,67037 Strasbourg, Cedex, France 5th International Kyoto Conference on New Aspects of Organic Chemistry (IKCOC-5) (Session 2: Organic Synthesis for Materials Science) Kyoto, Japan Professor Yoshiki Ohshiro, Chairman of IKCOC-5, Dept of Applied Chemisw, Osaka University, Yamadaoka 2-1, Suita, Osaka 565, Japan ICFRM 5: F'ifth International Conference on Fusion Reactor Materials Qearwater, Florida,USA Dr E.E. Bloom, General Chairman, ICFRM-5, Oak Ridge National Laboratory, PO Box 2008, Building 450023, Oak Ridge, TN 3783 161 17, USA iv November 30- December 1 December 1-6 December2-6 December 16-17 December 16-18 January 6-8 February 3-5 March 24-26 March 30-3 1 April 5-10 April 5-17 April 6-10 April 24-28 May 17-21 July 5-12 July 13-18 July 14-16 August 12-18 International Symposium on New Polymers Kyoto, Japan ProfessorToshinobu Higashiura, Department ofPolymer Chemistry, Kyoto University, Yoshida, Sakyoku, Kyoto 606, Japan Redrid, Optical, and Magnetic Properties of Organic Sdid State Materials (MRS Symposium) Boston,USA D.3. sandman, GTE Laboratories Incorporated,40 Sylvan Road, Waltham, MA 02254, USA. Tel.: (617) 466-4216 Cellulose'91 New Orleans, USA DrNoelie R. Bemniere, USDA ARSSouthern Regional Research Center, 1100 Robert E. Lee Boulevard, PO Box 1%87, New Orleans,LA 70179-0687, USA Prdein Engineering: Structure Prediction and Ligand Interactions London,UK DrNeera Brokakoti, c/o Roche Products Ltd,Broadwater Road, Welwyn Garden City, Hens AL7 3AY. UK Faraday Division Symposium on The Conformation of Flexible Molecules in Fluid Phases Southampton,UK ProfessorG. R. Ludchurst, Depamnent of Chemistry, University ofSouthampton, Southampton SO9 5NH, UK ~ ~~ Conference Diary 1992 Fluoropdymers '92 Manchester, UK DrE. Smith or Dr S.Dunn,Department of Chemistry, UMIST, Manchester M60 1QD. UK 19th Australian Pdymer Symposium Fremantle, Australia DrGraeme George, SecretaryRACI Polymer Division, Chemistry Department, University of Queensland, St. Lucia QLD 4067, Australia Fifth International Conference on Fibre Reinforced Composites Newcastle, UK conference Secretariat, FRC '92, The Plastics and Rubber Institute,11 Hobart Place, London SWlW OHL, UK Kautschukelastische Pdymersysteme Bad Nauheim, Germany GDCh-Geschaftsstelle, Abteilung Tagungen, Postfacb 90 0440, D-6000Frankfurt am Main, Germany ACS Spring Meeting San Francisco, USA ACS, International Activities Office, 155 16th Street NW,Washington DC20036, USA Nab AS1 on Multifunctional Mesoporous Inorganic Solids SiIltIq Portugal Professor&ar A.C. Sequeira,Instituto Superior T&nico, Av. Rovisco Pais, 1096 Lisboa Codex,Portugal ICMCIR International Conference on Metallurgical Coatings and Thin Films San Diego, USA S.V.Krishnaswamy, Westinghouse Science and Technology Center, 1310 Beulah Road, Pittsburgh, PA 15235, USA Fourth World Biomaterials Congress Berlin,Germany ProfessorDr Ulrich M. Gross, Instituteof Pathology, Steglitz Clinic, FU Berlin, Hindenburgdamm 30, D-1000 Berlin 45, Germany. TeL: 030/ 798 2296; FAX: 030/ 798 41 41. 7thInternational Conference on Relation Between Homogeneous and Heterogeneous Catalysis Tokyo, Japan Professor Yasuhiro Iwasawa, Chairman of WSHHC, Department ofChemistry, The University of Tokyo, Hongo, Tokyo 113, Japan EICS IV (European Gordon Conference) (Organometallic Compounds in Catalysis and in the Synthesis of Advanced Materials) Sheffield, UK ProfessorP. M.Maitlis FRS,Department ofChemistry,The University, Sheffield S3 7HF 34thIUPAC International Symposium on Maavrmdecules Prague, Czechoslovakia IUPAC Macro '92 Secretariat, Instituteof Macromolecular Chemistry, Czechoslovak Academy of Sciences, Heyrovkeho nam.1888/2,162 06 Prague 6,Czechoslovakia Speciality Polymers '92 London,UK conference Organizer, Speciality Polymers '92, Butterworrh Scientific Limited, PO Box 63, Westbury House, Bury Street, Guildford, Surrey GU2 SBH, UK ICSM '92:International Conference on Sdence and Technology ofSynthetic Metals Gteborg, Sweden S.StafsWCSM '92, Department of Physics, IFM,Linkaping University, S-581 83 Linktiping, Sweden. Tel.: + 46-13 281352; FAX: +46-13 137568; E-Mail: IFM::SST@SELIUCSl (BITNET) SST@IFM.LIU.SE (INTERNET) V August 23-28 ACS Autumn Meeting Washington, USA ACS, Intematicmal Activities Office, 155 16th Street NW,Washington DC 20036, USA August 24-28 European Conference on Molecular Electronics PadmItalyProfessorR Bozio, ECME 92, Department of Physical Chemistry, 2 Via Loredan, 1-35131 Padua, Italy August 31-Macromolecules '92 September 4 Canterbury, UK Dr A Amass, Department of Chemical Engineeringand Applied Chemistry, University of Aston, Aston Triangle, Birmingham B4 7ET, UK September 7-11 2nd International Workshop on the Crystal Growth of Organic Materials (CGOM2) Glasgow, UK Drs.D. Pugh and K. J. Roberts, Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow G1 1XL Tel.: 041-552-4400, FAX: 041-552 5664 Septembex 17-20 Implant Retrieval Symposium St Charles, Illinois,USA Society for Biomatenals, Post office Box 717, Algonquin, IL 60102 USA. Tel.: 708/658-2909, FAX: 708/658-2921. October 25-30 International Symposium on Chemistry and Physics of Molecular Based Magnetic Materials Shuzenji,Japan Professor Hiizu Iwamura, Department of chemistry, Faculty of Science,The University of Tokyo, Bunkyo-ku, Tokyo 113, Japan vi

ISSN:0959-9428    

DOI:10.1039/JM99101BP057

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

J. MATER. CHEM., 1991, 1(5), 715-724 FEATURE ARTICLE Charge Fluctuations and an Ionic-Covalent Transition in La,-,SrxCuO, John B. Goodenough, J-S. Zhou and K. Allan Center for Materials Science and Engineering, ETC 5.160, University of Texas at Austin, Austin, TX 78712- 1084, USA The search for the mechanism responsible for high-T, superconductivity in the copper oxides begins with a consideration of their distinctive structures, normal-state properties, and superconductive properties. All have intergrowth structures, and some consequences of this situation are enumerated. The normal-state properties are discussed with reference to the La,-,Sr,CuO, system in particular. Transport data are presented that provide evidence for a transition from small polarons to itinerant electrons with increasing x and, in the range of compositions 0.0355~50.125where this transition occurs, the onset of dynamic charge fluctuations in the normal state below a Tp> T,; Tp exhibits a maximum value near 150 K in a narrow interval about ~~0.075.These charge fluctuations appear to separate domains containing antiferromagnetic spin fluctuations from domains containing weakly correlated electrons within which there could be a formation of disordered large bipolarons. X-Ray photoelectron spectroscopy (XPS) data are cited and interpreted in terms of a transition from more ionic to more covalent Cu-0 bonding on passing from the antiferromagnetic to the superconductive compositions. A critical temperature T, proportional to the hole concentration pzx in the oxidized CuO, sheets is interpreted to be a consequence of a small superconductive coherence length and a two-dimensional conduction band.A change in the character of the condensation to the superconductive state in the interval 0.12<x<0.15 takes place; in the range xI0.12, condensation occurs from a state containing dynamic charge fluctuation within which large bipolarons may be formed, whereas for x20.15 condensation of Cooper pairs occurs directly from the normal state. It is noted that these observations fulfil two important conditions for our ‘correlation bag’ model of superconductivity to be viable: (1) stabilization of charge fluctuations by co-operative, dynamic atomic displacements and (2)an on-site electron-electron Coulomb repulsion that decreases sensitively with increased screening via an increase in the covalent character of the Cu-0 bond with increasing hole concentration and decreasing mean Cu-0 bond length.Keywords: Feature article; Oxide superconductor; Superconductivity Sorting out the distinctive structure-property features com- mon to all the copper oxide superconductors is a necessary first step in the identification of the superconductive mechan- ism operative in these materials. To this end, we call attention to the following structural, normal-state and, superconductive properties. (i) All (but one) of the copper-oxide superconduc- tors have intergrowth structures, and we consider some conse- quences of this crystal architecture.We consider, for example, the role of internal stresses and dimensionality in the stabiliz- ation of unusual electronic states. (ii) All these superconductors are mixed-valent systems, and their normal-state properties exhibit a confluence of three important cross-overs with changes in the electron/copper ratio: from antiferromagnetic insulator to normal metal, from small-polaron to itinerant- electron conductor, and from more ionic to more covalent Cu-0 bonding. (iii) In all these oxides, the superconductive pairs have a small coherence length and their condensation below T, reduces the density of states at the Fermi energy. There is mounting evidence that electron-phonon interactions play a key role in the condensation of the superconductive pairs; we argue for a condensation of Cooper pairs than a Bose condensation of bipolarons. In this paper, particular emphasis is placed on the La,-,Sr,CuO, system as it has the simplest structure and is chemically relatively stable. We present normal-state transport data that provide indirect evidence for a dynamic phase segregation (in the form of charge fluctuations) that separates electron-rich domains con- taining antiferromagnetic spin fluctuation from electron-poor regions containing itinerant electrons that, at lower tempera- tures, support superconductivity.We also cite XPS evidence for a sharp change from more ionic bonding in the antiferro- magnetic phase to more covalent bonding in the superconduc- tive phase and show how this change supports the appearance of electron states near the Fermi energy on doping, strong electron-phonon interactions, and a screening of the Hubbard U required for a ‘correlation-bag’ model of superconductivity.Intergrowth Structures Except for the recently discovered infinite-layer superconduc- tors,l all the known high-T, copper oxide superconductors have intergrowth structures consisting of superconductive layers alternating with non-superconductive layers. The super- conductive layers contain one or more CuOz sheets and have a fixed oxygen content. The non-superconductive layers may have a variable oxygen content; they also provide an internal stress on the superconductive layers that determines how the CuOz sheets may be doped.Superconductive layers with more than one CuOz sheet have these sheets separated by divalent or trivalent cations more stable in eight-fold than in twelve- fold oxygen co-ordination when located between CuOz sheets. The infinite-layer compounds consist of an infinite sequence of CuOz planes separated by cation planes as in a supercon- ductive layer. Without a non-superconductive layer to stabilize stressed Cu-0 bonds, high pressure is required to accomplish doping of the CuOz planes. The CuOz sheets contain a simple square of Cu atoms with 0 atoms forming a ca. 180" Cu-0-Cu bridge across each edge of the square. Within these structures, the Cu of a CuO, sheet may be co-ordinated by four, five, or six oxygen atoms depending upon the occupation of the apical (c axis) co- ordination sites.On the other hand, Cu in the non-supercon- ductive layers has been found co-ordinated to two, three, four, five or six oxygen atoms. Such flexibility of stable oxygen co- ordination is clearly one of the distinctive features of copper chemistry. Such a structural architecture has five immediate conse- quences, as described in the following sections., (1) The electronic potential at the Cu atoms varies sensitively with 0 co-ordination and mean Cu-0 bond length. Since an intergrowth structure holds the Cu-Cu distances essentially fixed within a CuO, sheet, the electrostatic Madelung energy increases with the number of apical oxygen and with decreas- ing apical Cu-0 bond length.This fact has four important consequences: (a) since superconductivity requires itinerant conduction electrons in a periodic potential and the Cu-0-Cu interactions only allow the formation of a narrow conduction band, superconductivity is restricted to those copper oxides in which all the Cu of a CuO, sheet have the same oxygen co-ordination; (b) correlated, dynamic dis- placements of the oxygen atoms can alter locally the electronic potential in a CuO, sheet so as to create charge fluctuations; (c) where Cu atoms in the non-superconductive layers have a variable oxygen co-ordination as in the YBa2CU306+, system, oxidation of the CuO, sheets depends not only on the oxygen concentration, but also on the oxygen co-ordi- nation in a CuO, sheet relative to that in the non-supercon- ductive layers; (d) whether a CuO, sheet can be oxidized or reduced may depend on whether the Cu-0 bonds are under compression or tension as well as the oxygen co-ordination.(2) Oxidation/reduction of CuO, sheets can be accomplished without perturbing too seriously its periodic potential. For example, aliovalent substitutions for the cations between CuO, sheets of a superconductive layer or for cations in the non-superconductive layer need not change the oxygen co- ordination at the Cu atoms of a CuO, sheet. Moreover, anion insertion or extraction into or from the non-superconductive layer within planes that are removed from the CuO, sheets does not change the oxygen co-ordination at the Cu atoms of a CuO, sheet.These are the types of doping strategies that have been used to convert an antiferromagnetic (CuO,), -sheet into a superconductor. (3) The electronic and elastic properties are strongly anisotropic. This strong anisotropy creates special problems for the fabri- cation of practical wires and films. It may also remove the orbital degeneracy of the conduction band. It is not yet clear whether removal of the degeneracy of the antibonding u* bands is critical for high-T, superconductivity. However, a two-dimensional electronic system tends to stabilize short- range fluctuations over long-range order, and we argue below for the presence of charge fluctuations that provide a dynamic phase segregation. (4) Internal electronic Jields normal to the intergrowth layer may be present to shift the energy levels in the superconductive layers relative to those in the non-superconductive layer^.^ For example, the La,CuO, structure of Fig. l(a) may be rep- resented as LaOlCuO, lLaO La01Cu0, [La0 2-2+ 2-J.MATER. CHEM., 1991, VOL. 1 where the vertical lines separate the superconductive CuO, layers from the non-superconductive LaOLaO rocksalt layers. The internal electric field introduced by the alternating formal charges 2- and 2+ for the CuO, and the LaOLaO layers of a tetragonal unit cell ensure that the introduction of an interstitial oxygen atom Oi between the two La0 planes of a rocksalt layer oxidizes the CuO, sheets by a capture of two electrons at an 02-i~n.~,~Oxidation of the CuO, sheets makes L~,CUO,+~ (6x 0.08) a superconductor.In the Y -,Ca,Ba2 -,La,Cu,O, +, system, it is possible to make x>l in the CuO, planes of the non-superconductive layer so as to create Cu(1) atoms in this plane with five-fold oxygen co-ordination that can compete for the holes associ- ated with the five-fold-co-ordinated Cu(2) atoms of the oxid- ized CuO, sheets. The substitution of Ca2+ for Y3+ ions modulates the c axis electric field so as to shift the value of x for which this competition is equal.6 (5) Bond-length mismatch across an intergrowth interface intro- duces internal elastic strains. The stabilization of an intergrowth structure requires epitaxial growth of successive layers.Since the thermal expansions of the different layers are generally not matched, intergrowth structures experience alter- nating and temperature-dependent tensile and compressive stresses within the successive intergrowth 1a~er.s.~ In the case of the La,Cu04 structure of Fig. l(a), for example, a measure of the bond-length matching is the tolerance factor t =(A-O)/J2(B-O) (2) that was first proposed by Goldschmidt for the cubic ABO, perovskites. The A-0 and B-0 bond lengths may be calculated at room temperature from the empirically tabulated ionic radii for the several ions; epitaxial matching occurs for t=l. In La2Cu04, where A=La3+ and B=Cu2+, a t<l at room temperature signals a tensile stress within the LaOLaO layers and a compressive stress within the CuO, planes of a tetragonal structure.Moreover, as has been discussed else- where,, nature finds three successive ways of relieving these internal stresses, which increase with decreasing temperature because the La-0 bond is softer than the Cu-0 bond. First, the antibonding 3d electron at a Cu" ion of the CuO, plane is removed from the 03-~2 band of 3d,2-,2 parentage, where the z axis is taken normal to the plane; the octahedral co-ordination at a Cu atom is strongly distorted to tetragonal (c/a>1) symmetry. Secondly, interstitial oxygen is introduced as 02-ions within the LaOLaO layers as the temperature is reduced; but at lower temperatures a reduced oxygen mobility inhibits further oxygen insertion even though t <1 continues to decrease with decreasing temperature.Thirdly, the CuOd tetragonal (c/a>1) octahedra tilt co-operatively, as indicated by the arrows in Fig. l(a), to cause a buckling of the Cu-0-Cu bond angle from 180"; the room-temperature symmetry is orthorhombic, and we refer to this structural situation as a T/O composition to include both tetragonal and orthorhombic polymorphs. With six-fold oxygen co-ordination at the copper atoms, the CuO, sheets of this phase may be oxidized, but not reduced to any significant extent. Substitution of a smaller Ln3+ ion for La3+ to obtain Ln2Cu04, where Ln =Pr.. .Gd, yields the T' tetragonal struc- ture of Fig. l(b). The smaller Ln3+ ion results in so small a tolerance factor that the apical, c axis oxygen atoms of the T tetragonal La2Cu04 phase are displaced to the interstitial sites to give a fluorite-type Ln-02-Ln layer intergrowing with CuO, sheets having Cu" in square-coplanar co-ordi- nation with no apical oxygen at the Cu.For Ln=Pr and Nd, the Cu-0-Cu bond lengths are stretched and the CuO, sheets are readily reduced by the substitution of Ce4+ or J. MATER. CHEM., 1991, VOL. 1 Fig. 1 (a) The tetragonal T structure of La,CuO,; arrows show direction of co-operative tilt of CuO, octahedra in low-temperature orthorhombic phase. (h)The tetragonal T structure of Nd,CuO, Th4+ for some of the Ln3+, but they cannot be oxidized to any significant extent. Use of a low-temperature preparative route has permitted Manthiram and Goodenough' to obtain the phase diagram of Fig.2, which illustrates well the tempera- ture dependence of the tolerance factor t and its control of the T' and T/O phase fields. Of particular interest is the fact that extrapolation of the phase boundary to lower tempera- tures indicates that La2Cu04 may be stabilized in the T' structure if prepared by a suitable low-temperature route. The T" phase in Fig. 2 appears to be a T' phase with a 3: 1 ordering of the La3+ and Nd3+ ions within the fluorite layers. Unlike the Ln, -,M,Cu04 systems, where the non-super- conductive layer controls the stress on the Cu02 sheets and t 0.855 0.860 0.865 0.870 I I I I T' T' 1000 .0 E $ 800 Q,E + T' 0,C .-.-L c 600 ..'..' I. ' 400 2.0 1.6 1.2 0.8 0.4 0.0 hence whether the parent compound can be doped p-type, the infinite-layer structure is formed at atmospheric pressure only as the parent compound Cao,84Sr0.16C~02 in which the Ca/Sr ratio is constrained to that which leaves the Cu-0 bond length in the CuO, planes near their equilibrium value. All attempts at atmospheric pressure to dope this material p-type or n-type so as to render it superconductive have failed. However, the different compressibilities of the Sr-0 and Cu-0 bonds allowed preparation of SrCuO, under a pressure of 25 kbar, the larger Sr2+ ion places the Cu-0 bonds under tension on removal of the pressure. This strategy allows relief of this tension by n-type doping; Sr, -,Ln,CuO, n-type superconductors can be prepared at 25 kbar.' Normal-state Properties There are three distinctive features of the normal-state proper- ties to which we call attention; we illustrate them for the system La, -,Sr,Cu04: (1) the evolution with increasing x from antiferromagnetic insulator to superconductor to nor- mal, Pauli paramagnetic metal; (2) a change from primarily ionic to primarily covalent Cu-0 bonding on passing from the antiferromagnetic to the superconductive phase; (3) the evolution with increasing x from small-polaron conduction to a hopping without activation energy to itinerant-electron transport behaviour with indirect evidence for the formation of charge fluctuations at temperatures T < Tpin the intermedi- ate regime.In order to introduce the discussion, it is necessary first to consider the electronic energy levels of the primarily ionic model, which appears to be applicable to the parent compound La,CuO,. La2Cu04 Fig. 3 represents schematically the steps taken in the construc- tion of a band diagram for CuO on starting from an ionic model. The free 0,-ion has a negative electron affinity, which places the 0-I2-redox potential above the vacuum level E,,,. Let EI be the energy required to take the second electron from a Cu atom, i.e. to remove an electron from the free-ion Cu2+/+ redox level, and put it on an 0-ion at infinite separation. The ionization energy, EI,lost is compen- sated by the electrostatic Madelung energy EM associated with an assembly of 2' and 2-point charges at the Cu and 0 lattice sites of the CuO structure.An E,>E, causes the Cu2+/+ 3d" redox energy to cross the 0-/2-redox energy, which is a necessary condition for an ionic model to be valid. Covalent mixing returns electronic charge from the O2-ions to the Cu2+ ions to reduce EM,but it also introduces a ' *'Cu 3d"Ei ,'.'*; \.-I I cu3+/2+ 3dg free ion point charge crystal field Fig. 3 Derivation of energy bands of CuO on an ionic model quantum-mechanical repulsion between the antibonding and bonding states that compensates for the loss in Madelung energy. Therefore, the mean energies of the bonding bands are not changed appreciably, and a point-charge model gener- ally gives an excellent estimate of the binding energy of an 'ionic' solid. However, the antibonding s and p bands tend to be shifted to higher energies relative to the d bands because of their stronger covalent mixing.The Cu-0 interactions are introduced into the crystal-field wavefunctions; the Cu-0-Cu and the 0-0 interactions are then treated separately to determine the widths of the antibonding and bonding bands, respectively. In the case of La,CuO,, an anomalously large tetragonal (c/a>l) distortion of the CuO, octahedra [Fig. l(a)] signals that the empty Cu2+/+ 3d" level at the Cu atoms would represent a hole in the x2-y2 crystal-field orbital, which is transformed into a narrow ~ band of width 2 W via the3~~ Cu-0-Cu interactions. In an oxide, the electrostatic inter- action between electrons of the crystal-field d-state manifold on a CU" ion remains large.Consequently, the Cu3+l2+ 3d" and Cu2+/+ 3d9 energy levels are split by an on-site 'corre- lation' energy U> W,which is why La,CuO, is an antiferro- magnetic insulator, see Fig. 4. In fact, what distinguishes silver and copper oxides are the relative magnitudes of the ratios W/U.W>U in an Ag" oxide can result in a disproportion into Ag' and Ag"'; W<U in the Cu" oxides stabilizes antiferro- magnetic ordering rather than a 'negative- U' charge-density wave, i.e. a disproportion into Cut and Cu"'. A Uz5-6 eV at a Cu" in La2Cu04 places the lower a3-,2 band below the top of the highest occupied band, which would appear to be a nzy band of primarily oxygen character because the t2g electron energies are not raised above the 0 2p, energies by the Madelung energy, EM.The measured energy gap' E, z2.0 eV for La,CuO, is therefore said to be a charge-transfer gap, A, rather than a correlation gap, u.The x2-y2 crystal-field wavefunctions have the form $x2 -y2 = NU(fX2 -y2 -MS -Ms) (3) where #s and 6, are the symmetrized 0 2s and 0 2p, orbitals of the nearest-neighbour oxygen in a CuO, plane that are summed so as to give the same symmetry as the atomic fxz -,,2 orbital; the covalent-mixing parameters are 2 = bCa/AE (4) for an ionic model where the covalence can be treated in second-order perturbation theory. The anion-cation reson-1 j!-La 5d CP2 2 I WE)-+ Fig.4 Schematic energy us. density of states for La,CuO, J. MATER. CHEM., 1991, VOL. 1 ance integral is b'", and AE is the energy required to transfer an electron from an oxide ion in the crystal to a neighbouring Cu2+ ion. A measure of the strength of the interatomic Cu-0-Cu interaction is the electron-transfer energy (resonance) intergral b= ($i,H'@j)X&u($i,@j) (5) where H' is the perturbation of the electron potential at site Rj by the presence of an atom at Ri,E, is a one-electron energy, and the crystal-field wavefunctions (3) are used in the overlap integral to yield b, -&,(A,' + 2,") (6) For localized electrons, the superexchange perturbation theory gives a Neel temperature TN, b:/U (7) and for itinerant electrons, the tight-binding approximation leads to a bandwidth WX2zb, = 8b, (8) where z=4 is the number of the nearest neighbours to a Cu atom in a CuO, sheet. From the pressure dependence" of the Neel temperature in antiferromagnetic La2Cu04, it would appear that stoichio- metric La2Cu04 has a b, close to the transition from localized- electron to itinerant-electron antiferromagnetism; in itinerant- electron antiferromagnets the superexchange perturbation expansion fails to converge, and TN decreases with increasing b,.l1 La, -3rxCu0, Stoichiometric La2Cu04 is a single-valent compound; all the copper atoms are present as Cu". High- T, superconductivity is a mixed-valent phenomenon; it occurs in a narrow range of compositions within which the CuO, planes are partially reduced or oxidized from the single-valent (CuO,), -state of La,Cu04.We consider here the p-type superconductors La, -,Sr,CuO, in which Sr2 + substitution for La3 + causes an oxidation of the CuO, sheets. The Phase Diagram Fig. 5 is a preliminary phase diagram for the system La, -,Sr,CuO,. The orthorhombic-tetragonal transition tem- perature Tt is seen to decrease with increasing x; the larger Sr2+ ion relieves the tensile stress in the rocksalt layers owing to a t<l, and removal of antibonding electrons from the CuO, sheets relieves their compressive stress. This relief of the internal stress also suppresses the tendency to incorporate excess oxygen. As prepared, L~,CUO,+~ generally has a maximum 6 x 0.02 excess oxygen in the non-superconductive phase; in the range 0.05 < x < 0.25 a 6 x0 is normally attained for La, -xSr,CuO, +a; in the range x 20.25 it is necessary to apply an oxygen partial pressure p(O,)>l atmt to oxidise the system fully, i.e.to prevent a 6<0. For our analysis, we assume 6 = 0 over the range 0.035 < x< 0.25 in our samples. The critical temperature, T,, below which superconductivity occurs appears to be restricted to the mixed-valent compo- sitional range 0.05 Ix < 0.27; T, increases nearly linearly with x in the range 0.05 IxI0.12 and saturates by x x0.15. The long-range antiferromagnetic ordering temperature TN drops precipitously with x, essentially vanishing by x x 0.02.A spin-glass ordering at low temperature is found in the interval 0.02 < x< 0.05. l2 On the other hand, short-range spin t 1 atm= 101 325 Pa. J. MATER. CHEM., 1991, VOL. 1 0.1 0.2 0.3 x in La,-xSr,CuO,,, Fig. 5 Preliminary phase diagram for the system La, -,Sr,Cu04 fluctuations persist through the compositional range where superconductivity occurs. T,,, in Fig. 5 represents the tem-perature at which the curve of paramagnetic susceptibility uersus temperature reaches its maximum value; a broad maxi-mum at a T,,, below which short-range spin fluctuations are stabilized is a common feature of two-dimensional spin sys-tems. With increasing x, T,,, decreases to zero to leave only an enhanced Pauli paramagnetism for compositions x >0.27 that exhibit normal metallic beha~iour.'~This feature of the phase diagram provides clear evidence that the system changes from being strongly correlated (W< U)for x <0.035 to weakly correlated (W>U)for x>0.27.A change from p-to n-type cond~ction'~near xzO.27 also testifies to a collapse of the correlation splitting U with increasing x. Nevertheless, the on-site correlation energy Uremains to stabilize antiferromag-netic spin fluctuations throughout the superconductorcompo-sitional range. These spin fluctuations have been observed by neutron scattering" and nuclear magnetic resonancel6 measurements. Their coherence length decreases with increas-ing x. Antiferromagnetic spin fluctuations can be expected to compete with superconductivity; each phenomenon depletes the density of one-electron states at the Fermi energy.There-fore, the observation of a coexistence of antiferromagnetic spin fluctuations and superconductivity in the same phase raises fundamental questions. Whereas some have proposed radically new theories of the electronic state, we17 have suggested a lattice instability that expresses itself as dynamic charge fluctuations that segregate electron-rich domains sup-porting antiferromagnetic spin fluctuations from electron-poor domains in which superconductive pairs are formed. In this connection, it appears particularly relevant that in the oxygen-rich system La2Cu04+, a static segregation into two phases occurs below room temperature, one antiferromag-netic and the other superconductive as illustrated in Fig.6.18 Segregation into two separate phases can occur at such low temperatures because the interstitial oxygen atoms Oi remain mobile to below room temperature. Such a phase separation between magnetic and non-magnetic phases commonly occurs where the interatomic interactions are approximately the same as the intraatomic correlation energy, i.e. where WzU.For example, lattice instabilities leading to a segregation into T 50( Yc-25( I ; Onr,Iq+orI I I I 1 I IT,I I\L... ---.--.--... I I I 0 0.04 0.08 0.12 6 Fig. 6 Preliminary phase diagram for La,Cu04 +d magnetic and metallic phases has been documented for the system LiVl -,Cr,02 where a transition from itinerant-elec-tron to localized-electron behaviour occurs in a single-valent system.Ig In the system La2Cu04+,, the two low-temperature ortho-rhombic structures are distinguishable; the antiferromagnetic phase with 6 50.02 exhibits a co-operative tilting of the Cu06 octahedra as illustrated in Fig.l(a),the superconductive phase has a distortion that appears to minimize the elastic inter-actions between the local distortions about the interstitial Oi atoms. This structural distinction presumably allows the inherent electronic instability associated with a change from localized to itinerant electrons to be expressed at a high enough temperature for a static phase segregation. In the system La2-,Sr,Cu04, on the other hand, the Sr2+ ions are not mobile at lower temperatures; any electronic instability can only express itself in this case by the formation of either a static charge-density wave or dynamic charge fluctuations, both of which would be stabilized by co-operative atomic displacements.In a mixed-valent system, any static charge-density wave tends to be incommensurate and rela-tively unstable, so the stabilization of dynamic charge fluctu-ations at low temperatures becomes highly probable. Fig. 7 gives a schematic representation of a charge fluctuation with a coherence length determined by a short-wavelength acoustic-mode vibration. Indirect evidence for dynamic charge fluctuations is begin-ning to appear from several types of experiment.For example, Egami et al.2o have used pair-distribution-function analysis of neutron-scattering data to identify the presence of local-scale dynamic inhomogeneities at temperatures TL T, in a superconductorsample of nominal T12Ba2CaCU208and Haga et al.21 have used temperature-dependent Rutherford back-C +-hole rich +electron rich -4 Fig. 7 Schematic representation of how acoustic-mode atomic dis-placements can create dynamic charge fluctuations scattering studies from [OOl] channels in YBa2Cu,06 +x samples to identify the onset of complicated lattice dynamics in the vicinity of T,. We22 have explored the evolution of the temperature dependence of normal-state resistance, R, and the Seebeck coefficient, a, of polycrystalline pellets of La2-,Sr,Cu04 pressed at ca.4kbar, fired at 1050 "C, and furnace-cooled, all in air. Fig. 8 shows the temperature variations of R and a2 for the superconductor composition x =0.055. Two critical tempera- tures can be identified in the Seebeck data; a TFx240K above which the Seebeck coefficient is temperature indepen- dent and a Tpbelow which a2 drops linearly with decreasing temperature. There is no anomaly in R us. T at TF;but below Tpthe resistance R deviates from a linear temperature depen- dence, goes through a minimum value, and begins to rise with decreasing temperature before reaching the superconductor critical temperature T,. In Fig. 5, Tp is obtained from the break in da2/dT. The linear temperature dependence of R above Tp is consistent with a diffusive hopping motion of the mobile holes of concentration p, but without any activation energy in their mobility pp: R-p=(pep,)-l-T (9) given a mobility pp =eDo/kB (10) and a temperature-independent mobile-hole concentration p.This behaviour differs from a Fermi liquid where R-T2 holds. In eqn. (lo), Do is the pre-exponential factor for the diffusion coefficient of the mobile holes; the exponential factor goes to unity for a motional enthalpy AH,,,=O. Such an interpretation anticipates a temperature-independent Seebeck coefficient, a, in which the configurational-entropy term is dominant where the motional enthalpy is small. $! z(kB/e)1n[2(1 -P)/Pl (1 1) Indeed, for T > TF a temperature-independent a is found (Fig.8), which suggests that in the temperature interval Tp< T< TF a change in the configurational entropy occurs 2.1I 4-0 0 100 200 300 T1K Fig. 8 Temperature variations of the square of the Seebeck coefficient, ct2, and of the resistance, R J. MATER. CHEM., 1991, VOL. 1 without any change in p or ,up.We suggest that below TF the period of a short-range spin fluctuation is longer than the time q, for a hole to jump from one cu atom to a neighbour; in this case, transfer of a hole to a nearest-neighbour Cu atom costs an exchange energy, so the hopping involves a double jump to a next-nearest-neighbour Cu atom. The resulting reduction of the concentration of available sites changes the statistics and hence the logarithmic term in eqn.(1 1) so as to reduce the magnitude of a; but it does so without changing p or pp.NMR data23 support such an interpretation; the spin- lattice relaxation term, l/Tl, is clearly defined and varies linearly with temperature in the inteval Tp< T< TF,dropping off sharply with decreasing T for T< Tp and being less well defined for T> TF. As the compositional parameter x increases to ~20.1,the Seebeck coefficient above TF becomes temperature dependent; it increases with decreasing temperature at temperatures T> TF, reaching a maximum value at TF. Such a variation is characteristic of a mass-enhanced narrow-band cond~ction,~~ so we conclude that a smooth transition from hopping conduc- tion with AH,,,=0 to itinerant-electron conduction occurs within the range 0.075<x <0.1.The hole mobility pp appears to be activated for x<O.O2. In the itinerant-electron compo- sitional range ~~0.075,TF decreases with increasing x, as does TmaX. The compositional dependence of Tp in Fig. 5 exhibits a maximum at xw0.075 and has a shape reminiscent of a spinodal decomposition. However, at these temperatures no atomic diffusion can occur; only atomic displacements can separate domains of strongly correlated electrons from domains of itinerant electrons. In the absence of a static charge-density wave, such displacements would introduce dynamic charge fluctuations. Associated with the charge fluc- tuations is a narrow band of states at EF within the charge- transfer energy gap.The fact that the maximum value of Tp occurs at x=0.075, the composition at which there is a transition from diffusive to itinerant-electron conduction, is also suggestive of the onset of charge fluctuations below Tp. It has been assumed by many that the change from a metallic to a semiconductor temperature dependence of the resistance R below Tpis due to the trapping out of charge carriers and the onset of variable-range hopping. But such an interpre- tation is inconsistent with the observed decrease in a with decreasing temperature for T< Tp (Fig. 8). On the other hand, the onset of charge fluctuations that create regions rich in holes would provide domains of high hole concentration that percolate through the crystal and have a reduced a; at the same time they would reduce the hole mobility.In fact, Hundley et al.25have shown that the static phase separation below room temperature in the normal state of La2Cu04+d, 0.02<6 <0.08, produces a sharp decrease in a and an increased resistivity with decreasing temperature in a manner similar to what is observed below Tpin La2-,Sr,Cu04. Therefore we conclude that the onset of charge fluctuations within the normal state occurs for T,< T< Tpin the compositional range 0.035 IX I 0.12, which is the range within which T, appears to increase linearly with the concentration of mobile holes in the Cu02 sheets. For x 20.15, superconductive-pair fluctu- ations appear just above T,;condensation to the superconduc- tive state occurs directly from a normal state without any charge fluctuations other than pair fluctuations just above TC.Our data are not able to distinguish between two limiting interpretations of the charge fluctuations. In one, the charge fluctuations represent the direct formation of large bipolarons, and Bose condensation occurs below T, for x I0.12. In the other, the charge fluctuations create fluctuating itinerant- electron domains within which large bipolarons may form J. MATER. CHEM., 1991, VOL. 1 (the on-site energy U is assumed to remain too large to allow the formation of small bipolarons), and condensation into superconductive Cooper pairs below T,occurs where the hole concentration is large enough for an overlap of the supercon- ductive pairs and the itinerant-electron domains have sufficient volume to percolate through the lattice.In this second case we can envisage the condensation of Cooper pairs below T, for all superconductive compositions. In either case, strong electron-phonon interactions are implicated; these results do not support those proposed mechanistic models that are based on electron-electron interactions alone. Nevertheless, it is necessary to inquire into the origin of the strong electron- phonon interactions and whether this origin implies a change in the electronic 'screening' of the coulombic repulsions between the electrons (holes) of a superconductive pair that would provide the additional stabilization needed for conden- sation into the superconductive ~tate.'~ Energy Levels Oxidation or reduction of a single-valent compound having correlation-split (U > w)energy bands is commonly pictured as a simple lowering or raising of the Fermi energy EF.In such a picture, the critical parameter is the electron-hopping time q, relative to the period of an optical-mode lattice vibration oR-'.A q,>uR-' leads to a trapping of the mobile charge carrier by a local lattice relaxation to form a small polaron. In such a picture, the total number of states associ- ated with the partially occupied d" redox energy stays fixed, but there is a change in the weights of the d"-' and d"+l levels with changing occupancy of the d" configurations. The small-polaron energies are shifted from the edge of occupied (p-type) or unoccupied (n-type) redox energies by 50.3 eV.Such a simple picture fails to take account of the shift in the redox potential energies that must take place as the Madelung electric field at the cation array, for example, shifts with oxidation or reduction. At issue is whether the Fermi energy moves across a correlation-energy gap (U -W) between fixed band edges with oxidation or reduction as it does in a broad-band semiconductor, where Koopmanns' theorem applies, or whether midgap states are created in which the Fermi energy becomes pinned. Optical spectroscopy9 has provided the first suggestion that midgap states are created in La, -,Sr,Cu04 as illustrated in Fig.9(a),and photoemission studies26 have shown that these midgap states have 0, not n, character. Fujim01-i~~ has used photoemission spectroscopy to show that the formation of midgap states may be a more f A N(a I ME)+--+ 72 1 general phenomenon, and Uchida et have used optical spectra to observe the existence of midgap states. In the case of La, -,Sr,CuO,, small-polaron behaviour is found only at small carrier concentration; at x=O.O75 there appears to be a transition from diffusive to itinerant electronic behaviour in the normal state. Nevertheless, the onset of dynamic charge fluctuations in the normal state below Tpin the interval 0.02 <x <0.15 indicates that strongly correlated electrons remain associated with the electron-rich domains where antiferromagnetic spin fluctuations are found, whereas the midgap states are associated only with the electron-poor 3 ~2domains.This duality of character of the 0~ electrons is represented in Fig. 9(a) by the presence of both midgap and correlation-split 03-y2 electron energies. Association of the 'midgap' states with 0,*2-~2 electrons follows from the observation that the mobile holes have 0, not n,character and lie in the Cu02 planes. Given a charge- transfer gap, A, in La,CuO, (see Fig. 4), this association can follow from the more covalent Cu-0 bonding due to a sharply reduced energy AE for transfer of an electron from oxygen to a Cu"' vs. a Cu" ion. In fact, the A. EbCa/AEof the ionic model for Cur'-0 bonding becomes too large for the perturbation expansion used in that model to converge.Consequently, a rather abrupt change from more ionic to more covalent bonding can be anticipated to occur with increasing x. In a covalent Cu-0 bond, the holes become stabilized in the most strongly antibonding states, which 03~ 2would be the ~orbitals associated with the shortest 3~~2Cu-0 bonds. The midgap 0 orbitals would contain a nearly equal (probably between 40-60 and 60-40) admixture of Cu 3d and 02p states. Finally, we emphasize that a change from more ionic to more covalent bonding necessarily increases the screening of the coulombic repulsion between pairs of mobile electrons. A measure of the consequent reduction in the electron corre- lations is the conversion of the antiferromagnetic semicon- ductor La2Cu04 to the normal metal Lal.7Sro.3Cu04 of Fig.9(b) and the associated vanishing of T,,, with increasing x in Fig. 5. Therefore, we" have previously taken the change U-0 in the screened on-site correlation energy on going from an electron-rich domain associated with spin fluctuations to an electron-poor domain associated with midgap states to be the added electronic stabilization needed to create super- conductive pairs of small coherence length via strong electron- phonon interactions. XPS Data X-Ray photoelectron spectroscopy (XPS) on the high-T, copper oxides has been plagued by the fact that this technique only probes CQ. 20A below the surface and the surfaces of many of these oxides are highly reactive with the water and C02 in the atmosphere.In order to obtain dry surfaces free of carbonate species, workers have felt forced to look only at surfaces obtained by cleavage in vacuum, preferably cleaned single crystals. However, scraping or cleaning an oxide in vacuum to remove contaminants may result in a reduced and/ or reconstructed surface. Without a subsequent anneal within the chamber under oxygen pressure, the 0 1s and Cu 2p3,, binding energies (Eb)have appeared to remain in the same positions as for the parent compound La,CuO,. This obser- vation has led to the speculation that oxidation of the CuO, sheets introduces holes into 0 2p bands. According to Fig. 3, too small a Madelung energy EM to cause the Cu3+I2+ 3d9 redox energy to cross the 0-l2-level would make the top of the nzy band primarily 0 2p in character, which could be Fig.9 Schematic energy us. density of states for (a) La1,85Sro,15Cu04the only basis for such a conclusion. and (b) Lal,,Sr,,,Cu04 However, these observations seemed surprising to us, so we2' undertook an XPS study of the metallic perovskite LaCuO,. This perovskite contains Cu"' with 180" Cu-0-Cu interactions, not 90" Cu-0-Cu interactions as in the CuO, chains of NaCuO,. Relative to the peak positions found in La2Cu04, we observed a definite shift (ca. 1 eV, but not the 2 eV associated with surface OH-ions) of the 0 1s peak to higher Eb and also a larger shift of the main Cu 2p3/2 peak (22 eV) to higher Eb.On the other hand, the main Cu 2p3/, peak of Cu"' in NaCuOz was found3' to be shifted by only ca. 1.3 eV. This finding has led us to consider whether the enhanced covalent bonding responsible for reduc- ing the 'screened' on-site correlation energy U on going from Cu" to Cu"' should not manifest itself in the energies of the '3d" configurations' associated with the main Cu 2p3,, peak and in the positions of the satellite peaks as well. If this were the case, then the greater screening associated with six-fold oxygen co-ordination and delocalized o* electrons in metallic LaCuO, relative to four-fold oxygen co-ordination and a localized di2 configuration in NaCuO, could be responsible for the shift to higher E, of the Cu2p3,, main peak, corre- sponding to 12p'3d''L2 > in conventional notation, found in LaCu0, relative to NaCuO,.The more Cu 3d is the character in the ligand-state holes L, the greater is the covalent screening of the on-site electrostatic energy U and hence the greater the E, of the 12p13d''L2> excited state. Furthermore, the Cu 2p3,, main peak is-broadened in LaCu03 relative to NaCuO,. We believe this difference reflects a broader ~,*2-~2 band in LaCuO, where 180" Cu-0-Cu interactions allow strong overlap; in NaCu02 on the 90" Cu-0-Cu interac-03 ~2tions make the ~ orbitals on neighbouring Cu atoms [Fig. l(b)] orthogonal to one another. This line of reasoning leads to the following idea. Since a superconductor is expected to have itinerant electrons in the normal state as a result of enhanced Cu-0 covalent mixing, the transition from a localized-electron configuration in anti- ferromagnetic La2Cu04 to an itinerant-electron configuration for at least the 0,*2-~2 electrons in the superconductor La1~85Sro~15Cu04should result in a clear shift to higher E, of the main Cu 2p3,, peak in the superconductor due to the greater screening of the energy parameter U.Moreover, in the n-type superconductors with the tetragonal T' structure of Fig. I(b), a shift to higher Eb (albeit a smaller shift than in La1~85Sro~15Cu04)might occur in spite of a reduction of the formal valence state on the copper. The upper limit of the possible peak positions would have to be that of LuCuO, for the p-type superconductors and that of NaCuO, for the n-type superconductors. In order to test these conclusions, we set out to re-examine the XP spectra of the superconductor compositions La 1.85srO.1 Scu04 (p-type) and Nd 1.8SCe0. 1ScU04 (n-type)*Without access to either single crystals or the means to create fresh surfaces in situ within the spectrometer, we were forced to employ an ex situ sample-preparation procedure. Although we were able to obtain reproducible results consistent with our hypothesis for spectra obtained at normal collection angle, our best surfaces remained contaminated with graphite (not CO or CO,) and surface OH- groups, but not La(OH),. The results are consistent with cleaner spectra obtained after an oxygen anneal in the spectrometer of p-type single crystals cleaved in situ that are now in the literature.Although these literature spectra were not taken on the La2-,Sr,Cu04 system, we refer to these spectra as they must be more convincing than ours. Flavell and Egdell,,l for example, have compared the Cu 2p core-level photoemission spectra for YBa2Cu306 +, in both metallic (x >0.5) and non-metallic (x <0.5) phases. The differ- ence spectrum between these two phases shows a pronounced shift of nearly 2eV toward higher E, of the main CU~P,,, J. MATER. CHEM., 1991, VOL. 1 peak on going from the non-metallic, antiferromagnetic to the metallic, superconducting phase. We presume the peak at higher energy is due to oxidation of the CuO, sheets. More recently, Parmigiani et reported XPS data taken on a cleaved crystal of Bi,Sr,CaCu208 annealed (temperature unspecified) in situ under 12 atm 02.They found a shift of the 529eV 0 1s peak to a higher Eb of ca.530eV, which is what we found in our spectra, and a broadening of the main Cu 2p3,, peak due to the appearance of a shoulder on the higher Eb side; this broadening is again indicative of a new peak position at ca. 2eV higher E,. Our 'good' samples of La1~85Sro.15Cu04showed a similar shift of the main Cu 2p3/, peak to higher E,. In the case of the n-type Nd,-,Ce,CuO, system, scraping of the surface was found to degrade enormously the XP spectrum; the unscraped, cleaved surface gave a main Cu 2p3,, peak that was greatly broadened toward higher E, despite a reduction of the CuO, sheets.,, In our samples, we found a shift of ca.1 eV, for normal collection angle, to higher E,, but no shift for collection at a grazing angle. Fig. 10 compares the positions of the higher E, of the main Cu 2p3,, peak we obtained for p-type Lal~85Sro~lsCu04 and n-type Ndl .85Ceo. 15Cu04 superconductors relative to the peak postions found for La2Cu04, NaCuO, and LaCuO,. The shift to higher E, of the Cu 2p3,, peaks in the reduced CuO, plane of n-type Nd,,85Ceo,15Cu04 is in accordance with the qualitative predictions we made in the last section. The shifts to higher E, found for Cu co-ordinated to five oxygen appear to be intermediate to those for Cu in six-fold co-ordination in La1~85Sro~15Cu04 and for Cu in four-fold co-ordination in Nd .85Ce0, sC~04.Moreover, the evidence points to the presence of two types of Cu atoms in a Cu0, sheet, one with an E, close to that of Cu" in CuO or La2Cu04 and the other at a higher E,, the magnitude of the shift corresponding to the oxygen co-ordination at the Cu atoms. Such a situation can be understood if a transition from more ionic bonding at Cu" ions is transformed to more covalent Cu-0 bonding at some, at least, of the Cu of a CuO, sheet and the more covalent Cu-0 bonding results in a screening of the on-site o* electrons that sharply reduces the on-site, screened covalent energy U. Superconductive Properties Attention is called to five superconductive properties: (1) As in conventional superconductors, a gap A is opened up at the 3 2.3.0 2.0 1.o 0 AEJeV Fig. 10 Schematic diagram of the increase in binding energy AEb relative to La,CuO, and CuO of the CU~P,,, peak for several copper-oxide compounds: (a) La,CuO, or CuO; (b) Nd,,,,Ce,.,, CuO,; (c) N~CUO,;,~(d)La1,,,Sr,,,,Cu0,; (e) LaCuO, J. MATER. CHEM., 1991, VOL. 1 Fermi energy below T,; however, states remain in the gap, and the gap is highly anis~tropic.~~ (2) Muon spin rotation has shown that T, of a p-type superconductor is proportional to p/m* as the hole concentration p approaches the value psat where T, reaches a saturation value.35 (3) The superconduc- tive-pair coherence length r is anisotropic and small (tLx3 A and tllx10 A) in the copper oxides relative to its value in conventional superconductors, e.g.r,,,, z1000A.36(4) In the system La, _,Sr,CuO4, the oxygen isotope effect increases monotonically with hole concentration to x x0.12, i.e. to p zpsat,beyond which it drops precipitiously. (5) The Meissner effect shows that only a fraction of the sample becomes superconductive and that this fraction increases rapidly with x in the range 0.05 <x<O.12. The first of these properties together with a critical hole concentration for superconductivity to appear abruptly with a finite Tc37supports Cooper-pair formation rather than Bose condensation of bipolarons, particularly as a small Meissner fraction shows that the superconductivity is confined to only a fraction of the total volume that increases with x in the range 0.05<x I0.1 2.However, the small coherence length indicates that large bipolarons may be formed within the electron-poor charge fluctuations forming in the temperature range T, <T <Tp. We have argued el~ewhere’~ that, from the Uncertainty Principle, a small requires a larger range of momenta in reciprocal space, and hence a larger energy range hw,,, of states about EF that are perturbed by the formation of superconductive pairs below T,. The critical temperature T, is given by an expression of the form where the pairing potential E,_has been derived for strong electron-lattice interaction^,^^ V, is the coulombic repulsion between electrons of a pair, and N(EF) is the density of one- electron states at EF for T> T,.Since all the holes contribute to superconductive pairing so long as the energy difference between the top of the conduction band and EF is EFIhumax, it follows that for a two-dimensional conductor (the CuO, sheets) and that E~z~w,,, at p=ps,,x0.12; a broadening of the narrow midgap band also occurs with increasing x zp. It is not necessary to have a Bose condensation of two-dimensional bipolarons to achieve a T, -p. The observation that, in the system La,-,Sr,CuO,, not only does Tpapproach T,, but also a precipitous drop occurs in the oxygen isotope effect at pzpPsatwould seem to be significant, If our interpretation of Tpis correct, the system is segregated below Tp by dynamic charge fluctuations into domains of strongly correlated electrons and domains of more weakly correlated electrons.The former give rise to spin fluctuations, the latter restrict the volume within which super- conductivity can occur. For p <0.075, the domains of strongly correlated electrons occupy the major volume; for p >0.075 the domains of weakly correlated electrons dominate. The condensation of Cooper pairs below T, within the weakly correlated domains requires a sufficient hole concentration for any itinerant-electron bags (or large bipolarons) formed to be strongly interactive. Moreover, although superconduc- tivity is found in the range 0.05<xI 0.75 where hopping conduction occurs, charge fluctuations would create a p >0.075 within the itinerant-electron domains where Cooper pairs are formed.From this perspective, the model provides a simple explanation of the conspicuous increase in Meissner 723 fraction over the compositional range 0.05 Ix I0.12 where we find Tp>T,. To this point in our discussion we have avoided a commit- ment to any particular model of the specific mechanism responsible for the formation of superconductive Cooper pairs. We have argued for charge fluctuations associated with the instability inherent in a transition from small-polaron to itinerant-electron behaviour within a narrow midgap band of the mixed-valent system. Now we once more call attention to the associated collapse in the correlation energy U. From the variation of T,,, and TF with x in the phase diagram of La2-xSrxC~04(Fig. 5) we conclude that within the more weakly correlated domains produced by charge fluctuations there may be an additional instability associated with a condensation of Cooper pairs within bags of smaller corre- lation energy via a different charge fluctuation.This model leads to17 where U-0 is the change in the on-site coulombic repulsion between holes of a superconductive pair on passing from without to within a single ‘bag’ of charge fluctuation. This model requires a reasonable magnitude for the change U -U; but we have argued from the XPS data (as well as the evidence for midgap states asEociated with more covalent 0-0 bonding) for a U-U-1 eV resulting from the increased screening of the on-site energy U due to the greater covalent mixing within a ‘bag’.Finally, a change from one type of charge fluctuation to another at T, for XI 0.12 would be in contrast to the initiation of charge fluctuations of the second type at T, for x>O.12, which may account for the anomalous change in the oxygen isotope effect in the range 0.12 <x <0.15. REFERENCES 1 M. G. Smith, A. Manthiram, J. Zhou, J. B. Goodenough and J. T. Markert, Nature (London), 1991 351, 549. 2 J. B. Goodenough, Supercond. Sci. Technol., 1990,3,26;Supercon-ductivity and its Applications, American Institute of Physics Con$ Proceedings 219, ed. Y-H. Kao, P. Coppens and H-F. Kwak, Buffalo, 1990, p. 26. 3 J. Fontcubert, X. Obradors and J.B. Goodenough, J. Phys. C, 1987, 20, 40. 4 J. Zhou, S. Sinha and J. B. Goodenough, Phys. Rev. B, 1989, 39, 12331. 5 J. D. Jorgensen, B. Dabrowski, S. Pei, D. R. Richards and D. G. Hinks, Phys. Rev. B, 1989, 40, 2187; J. D. Jorgensen, P. Lightfoot and S. Pei, Supercond. Sci. Technol., 1991, 4, S11. 6 A. Manthiram and J. B. Goodenough, Physica C, 1989,159,760; 162-164, 69. 7 J. B. Goodenough, J. Less-Common Met., 1986, 116, 83. 8 A. Manthiram and J. B. Goodenough, J. Solid State Chem., 1991, 92, 231. 9 M. Suzuki, Phys. Rev. B, 1989, 39, 2312. 10 M. C. Aronson, S-W. Cheong, F. H. Garzon, J. D. Thompson and Z. Fisk, Phys. Rev. B, 1989,39, 11445. I1 J. B. Goodenough, Prog. Solid State Chem., 1971, 5, 145. 12 S. Uchida, in Proc.IX Winter Meeting on Low Temperature Physics: Progress in High- Temperature Superconductivity. ed. J. Heiras, R. A. Barrio, T. Akathi and J. Taguena, World Scientijc, Singapore, 1988, vol. 5, p. 13. 13 J. B. Torrance, A. Bezinge, A. I. Nazzal, T. C. Huang, S. S. P. Parkin, D. J. Keane, S. J. La Placa, P. M. Horn and G. A. Heald, Phys. Rev. B, 1989, 40, 8872. 14 H. Tokagi, T. Ido, M. Ishibashi, M. Vota, S. Uchida and Y. Tokura, Phys. Rev. B, 1989,40, 2254. 15 R. J. Birgeneau, D. R. Gabbe, H. P. Jenssen, M. A. Kastner, P. J. Picone, T. R. Thurston, G. Shirane, Y. Endoh, M. Sato, K. Yamada, Y. Hidaka, M. Oda, Y. Enomoto, M. Suzuki and T. Murakami, Phys. Rev. B, 1988, 38, 6614; G. Shirane, R. J. Birgeneau, Y. Endoh, P. Gehring, M. A. Kastner, K.Kitazawa, 724 J. MATER. CHEM., 1991, VOL. 1 16 H. Kojima, I. Tanaka, T. R. Thurston and K. Yamada, Phys. Rev. Lett., 1989, 63, 330. Y. Nokamura and K. Kumagai, Physica B, 1990, 165 & 166, 1303; Y. Kitaoka, K. Ishida, K. Kondo and K. Asaymo, Physica B, 1990, 165 & 166, 1309. 27 28 29 A. Fujimori, personal communication, 199 1. S. Uchida, T. Ido, H. Takogi, T. Arima, Y. Tokura and S. Tajima, Phys. Rev. B, 1991,43, 7942. K. Allan, A. Campion, J-S. Zhou and J. B. Goodenough, Phys. Rev. B, 1990,41, 11 572. 17 18 19 20 J. B. Goodenough and J-S. Zhou, Phys. Rev. B, 1990, 42,4276. J. D. Jorgensen, P. Lightfoot and Shiyou Pei, Supercond. Sci. Technol., 199 1, 4, S11. J. B. Goodenough, A. Manthiram and G. Dutta, Phys. Rev. B, 1991, in the press. T. Egami, B. H. Toby, S. J. L. Billings, Chr. Janot, J. D. Jorgensen, D. G. Hinks, M. A. Subramanian, M. K. Crawford, W. E. 30 31 32 33 P. Steiner, V. Kinsinger, I. Sander, B. Siegwart, S. Huffner, C. Politis, R. Hoppe and H. P. Muller, Z. Phys. B, 67, 497. W. R. Flavell and R. G. Egdell, Phys. Rev. B, 1989, 39, 231. F. Parmigiani, Z. X. Shen, D. B. Mitzi, I. Lindau, W. E. Spicer and A. Kapitulnik, Phys. Rev. B, 1991, 43, 3085. T. Suzuki, M. Nagoshi, Y. Fukunda, K. Oh-ishi, Y. Syono and M. Tachiki, Phys. Rev. B, 1990, 42, 4263. 21 Farneth and E. M. McCarron, in Proc. Univ. Miami Workshop on Electronic Structure and Mechanisms for High- Temperature Superconductivity, ed. G. Vezzoli, J. Ashkenazi, S. Barnes, F. Zuo and B. M. Klein, Plenum, New York, 1991, in the press. T. Haga, K. Yamaya and Y. Abe, Phys. Rev. B, 1990,41, 826. 34 35 36 0.G. Olson et al, Science, 1989,245,73 1; R. Manzke, T. Buslaps, R. Claessen and J. Fink, Europhys. Lett., 1989, 9, 477. Y. J. Uemura et al, Phys. Rev. Lett., 1989, 62, 2317. W. C. Lee, R. A. Klemm and D. C. Johnston, Phys. Rev. Lett., 1989, 62, 1012. 22 23 J-S. Zhou and J. B. Goodenough, unpublished. T. Imai, Y. Kazuyoshi, T. Uemura, H. Yasuoka and K. Kosuge, 37 A. Rigamoni, F. Borsa, M. Corti, T. Rega, J. Ziolo and F. Waldner, Earlier and Recent Aspects of Superconductivity, J. G. 24 25 J. Phys. SOC. Jpn., 1990, 59, 3864. R. E. Pronge and L. P. Kadanoff, Phys. Rev. A, 1964, 134, 566; M. A. Howson and B. L. Gallagher, Phys. Rep., 1988, 170,265. M. F. Hundley, J. D. Thompson, S-W. Cheong and Z. Fisk, Phys. Rev. B, 1990,41, 4062. 38 Bednorz and K. A. Muller, 1990, Springer-Verlag, Berlin-Heidel- berg, p. 441. A. S. Aleksandrov and V. F. Elesin, Fiz.Tverd. Tela (Leningrad), 1983, 25, 456 (Sou. Phys.-Solid State, 1983, 25, 257). 26 E. E. Alp, J. C. Campuzano, G. Jennings, J. Guo, D. E. Ellis, L. Beaulaigue, S. Mini, M. Faiz, Y. Zhou, V. W. Veal and J. Z. Liu, Phys. Rev. B, 1989,40, 9385. Paper 1/02389F; Received 23rd May, 1991

ISSN:0959-9428    

DOI:10.1039/JM9910100715

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

J. MATER. CHEM., 1991, 1(5), 725-734 Behaviour of the Adsorbed CI' Intermediate in Anodic CI, Evolution at Thin-film RuO, Surfaces Brian E. Conway," Gu Ping," Achille De Battisti,b Andrea Barbierib and G. Battaglin"" Chemistry Department, University of Ottawa, 140 Louis Pasteur Street, Ottawa, Ont. KIN 6N5, Canada Dipartmento de Chimica, Universita ' di Ferrara, Via Borsari 46, Ferrara 44 100, Italy Dipartimento di Fisica, University of Padova, Padova, Italy Studies of the adsorption of the CI' intermediate in anodic CI, evolution at a series of four thin-film RuO, electrodes formed on a Ti substrate have been made by means of potential-relaxation measurements, following interruption of previous steady-state currents, coupled with determination of the current vs.overpotential kinetic relationships. Experiments at rotated RuO, electrodes show only a small effect, indicating that neither diffusion-controlled supply of CI- nor effects of CI, supersaturation in the boundary region at the electrode materially effect the kinetic behaviour. This behaviour is manifested as continuously curved Tafel relations, which approach limiting currents at overvoltages of 0.2-0.3 V. Therefore, on these thin-film RuO, materials, kinetic control of CI, evolution seems to be by the CI' recombination pathway; this is supported by linearity of Conway-Novak test plots for the recombination mechanism. Analysis of the potential-relaxation transients enables the pseudocapacitance, C,, for adsorbed CI to be determined.C, shows ascent to large values below 50-100 mV of overpotential, depending on temperature. Some problems arise, however, in reconciling this adsorption behaviour with expectations associated with recombination control. The near-surface composition profiles of the RuO, films have been determined by means of Rutherford back- scattering and the average compositions, deeper into the films, by energy-dispersive X-ray emission analyses. Keywords: Electrocatalyst; Ruthenium dioxide; Thin film; Electrode The development of RuO, electrodes (the so-called DSA or materials), for use in commercial cells for anodic C1, pro-duction, is well known, e.g. ref. 1. It represents one of the 2M0,C1 (3) k3 -2M,, +C1,Tmost successful applications of electrochemical technology. In parallel with these advances, considerable fundamental Electrode-kinetic diagnostic criteria indicate that the research has been carried out on the nature2p3 of RuO, films recombination pathway (l), (3) seems to apply to a number that can be thermochemically formed on Ti substrates and of cases of the Cl, evolution reaction at various anode on the electrochemical formation4 of RuO,; also, the electro- materials.'*'0-'2 It h as generally been indicated,"-12 in chemical redox behaviour of thermally and electrochemically deducing mechanisms for anodic Cl, evolution, that the generated forms of RuO, has been studied.' In complementary product of C1- discharge is a chemisorbed C1' atom that is work, the kinetics and mechanisms of anodic Cl, evolution then recombinatively desorbed as Cl, by either step (2) or have been examined at RuO, surface^^-^ and, comparatively, step (3).Only by the mechanism (l), (3) involving C1. can the at IrO, films of various thicknesses.' Generally, the properties observation of Tafel relations with low slopes of ca. 0.03 V at of the RuO, electrocatalyst material depend significantly on low overpotentials, which increase continuously to a limiting the Ru loading at the Ti substrate and especially the tempera- current condition at high overpotential, be accounted for. In ture of formationg of the RuO, from painted films of RuC13/ fact, the Conway-Novak test plot procedure, lo applicableTiC1, or other additives. over the whole range of overpotentials, gives the most In previous works, the kinetic behaviour of RuO, prep- unambiguous confirmation of the recombination mechanism arations for anodic C12 evolution has been examined almost involving adsorbed C1 .Also, in Krishtalik's mechani~m,~ entirely by means of log(current, i) vs. overpotential, q, curves involving desorptive recombination between the C1+ inter- (Tafel relations) with mechanistic interpretations being based mediate and C1- (in the double-layer), the Cl+ entity is on values of the Tafel slopes (b=dq/d log i). Various mechan- derived from adsorbed C1 by electrochemical oxidation. isms have been propo~ed,~-~,'~-'~ as for Cl, evolution at Hence, the involvement of C1' is well established, albeit Pt."," Generally, these follow the types of successive steps somewhat indirectly.A significantly different desorption step envisaged and experimentally indicated for cathodic H2 evol-from those shown in the reactions (1)-(3), but still involving ution, to which the anodic Cl, reaction processes are formally Cl', was proposed by Krishtalik and co-workers7 for the C1, similar, uiz. discharge of the C1- ion evolution process on RuO,: M,,Cl * +M,,Cl + +e -(4) followed by a presumably rapid step followed by either of the desorptive steps M,,Cl+ +Cl-+Cl,f (5) with the possibility that Cl', the chloronium ion, could exist M,,C1' +C1-A M,,+Cl,f+e-(2) in the solution boundary layer, near the anode surface or be adsorbed at it. However, the oxidation of a very electronega- tive element to an onium-type ion is improbable; so an equivalent alternative is the OC1- intermediate, generated from Cl'(C1' +H,O-+OCl- +2H+ +e-), from which C1, can be formed.In the above mechanisms, it is to be emphasized that the elementary steps involved are envisaged as proceeding on an anodically formed4*I3 oxide film'3'0314 of the anode metal M (indicated above as Mox)or on a thermochemically formed oxide film, as for the cases or RuO, or IrO,. In the case of oxide films generated in Cl -solution at Pt, co-adsorbed C1- exists together with electrodeposited OH and 0 in the oxide fiim.15 Alternatively, the kinetics of C1, evolution can be studied on oxide films, separately pre-formed on PtI4 and charac- terized by linear sweep-voltammetry in aq.H2S04, and then transferred to a Cl--containing solution from which Cl, can be anodically generated. In both fundamental and R and D work,l,16 no attempt has been made to examine the adsorption behaviour of the C1' intermediate'.l0 in steps (l), (2) or (3) above. In general, as was emphasized in our recent paper^,'^-^' catalysis in an electrode process must be characterized not only by log io and b values, together with surface-structure, band-structure and surface-composition evaluations, but also through deter- mination of the adsorption behaviour of the chemisorbed intermediate@) involved. In the present paper, we address this question by examin- ation of the potential-relaxation behaviour of C1,-evolving Ru0, anodes.This enables the interfacial capacitance behav- iour to be evaluated using previously described2lV2' pro- cedures that have been applied to the H2 l8 and 0,l9 evolution reactions. Experimental General Procedure The required information on the kinetics of anodic C1, evolution at RuO, surfaces and the involvement of the adsorbed C1 intermediate was obtained by complementary measurements of the steady-state log(current-density, i) us. potential relations and of the potential-relaxation behaviour, following interruption of prior steady-state currents at a series of controlled potentials, E, as decribed previously. The7322 results are treated19v21 as follows, defining the potential- relaxation rate, -dq/dt, in relation to the interfacial capaci- tance, C, and the kinetics of the electrode process in terms of the exchange current-density, io, the transfer coefficient a, and the overpotential q -C(dq/dt)= -(Cdl+ C,)dq/dt =io exp(aqF/RT) (6) Here, C is represented as being a combination of the double- layer capacitance, Cdl, and an adsorption pseudo-capacitance, C,, arising from the potential dependence of the coverage, 8, by the Cl intermediate at appreciable overpotentials.C, is defined as qldO/d~ where q1 is the charge percm2 for com- pletion of a monolayer of Cl'. Assumptions concerning, and the validity of, eqn. (6) have been treated in previous pap- er~~~*~~where the operational pseudo-capacitance C, is related to other capacitance quantities of theoretical interest." Potential-relaxation Transients and Tafel Relations Following interruption of prior anodic steady currents by means of a vacuum Hg relay, potential-relaxation transients were recorded over a time span of ca.2 ps to 10 s on two Nicolet digital oscilloscopes to cover the desired time range, J. MATER. CHEM., 1991, VOL. 1 as described in previous paper^.'^,'^.^^ The data were pro- cessed with a Hewlett-Packard computer. Tafel relations were determined by means of a computer-controlled potentiostat, as in ref. 23. Electrode Preparation In the present work, a special series of electrodes with thin films of RuO,, thermochemically deposited on Ti, were stud- ied. The morphology of such electrodes is substantially differ- ent from that of the regular 'mud-cracked' type of surface found with thicker-film RuO, DSA electrodes as seen in SEM photographs (Fig.1) where the surfaces of thin- and thick- film Ru02/Ti02 electrodes are compared. The purpose of using thin-film Ru0, electrodes was (a) to examine how electrodes with relatively low loadings of RuO, behaved as anode materials for Cl, evolution and (b) to have surface regions that were materially less porous than those at thick- film electrodes so that porosity and macroheterogeneity effects in potential-relaxation and ax. impedance meas~rernents~~ could be minimised. The thin-film RuO, electrodes were prepared in the follow- ing way, which is similar to that employed in the usual preparation of thick-film electrodes. Solutions of RuCl, -3H20 with TiCl, in isopropyl alcohol were painted on freshly cleaned Ti metal tabs, and each coating was given a prelimi- nary drying at 100 "C followed by pyrolysis at 400 "C for 15 min in a dry 0, atmosphere.The procedure was repeated up to 10 times, giving an Ru0, coating of 1 pm per 25-27 coatings. Final annealing of the accumulated coating was carried out at 400 "C for 2 h. Electrodes having five levels of relative RuO, loading were employed; their compositions are listed in Table 1. Thicknesses varied from ca. 100 to ca. 400 nm. The RuO, loadings in the thin-film oxide surfaces were ca. 0.05-0.25 mg cm-, of original apparent (smooth) Ti surface. Chemical Compositions of the Oxide Films The compositions of the near-surface regions of the films were determined by means of X-ray energy dispersive spectrometry (Table 1) expressed as the wt.% of Ru and Ti.The relative compositions of the films were found to be different from those of the respective coating solutions, especially for the thin films (1-4), an effect that may arise from penetration of the Ru atoms into the TiO, substrate or of Ti into the oxide film. For the thick film, 5, this difference was not, however, significant. Preparation of a Rotating RuOz Electrode In some experiments on possible effects of C12 supersaturation (cf:ref. 25),a thin layer of RuO, was thermochemically formed at the end of a rod of Ti embedded, after thermal preparation of the RuO, film, in a rotatable Teflon holder. This was used as a 'rotating-disc' RuO, electrode (employing a Pine Instru- Table 1 Compositions of RuOJTiO, films Ru ("0 in electrode RU (Wt.Yo) Ti (wt.%) coating solution) I 9 91 30 2 29 71 60 3 38 62 80 4 50 50 I00 5 22 78 25 J.MATER. CHEM., 1991, VOL. 1 Fig. 1 SEM photos of (a), (b) thin-film RuO, coatings on a Ti substrate and (c) thick-film, DSA-type coating. Lenth of bar: (a)and (b)4 Clm, (4 2 Clm ment rotator), although the surface was, of course, not micro- scopically smooth. Composition Depth Profiling Composition depth profiling of the RuO, films was carried out by means of Rutherford back-scattering measurements 127 conducted at the University of Padova. Although this tech- nique is ideally applicable only at smooth surfaces, it gave meaningful results for the relatively smooth thin RuO, films.The composition profiles for the four thin RuO, films are shown in Fig. 2. Cell and Solutions Aqueous NaC1-HC1 solutions were made up in doubly dis- tilled water and transferred to a three-compartment all-glass cell of the usual design.".14 Glass-sleeved stopcocks were employed so that the cell could be partially immersed in a thermostat fluid without contamination. Reference Electrodes All potential measurements were referred to the potential of a Cl2/Cl--platinized Pt reference electrode in a separate compartment but in a solution of the same composition as that in the working electrode compartment; the measured potentials were therefore overpotentials (q)for C1, evolution, since C1, was also bubbed in the working-electrode com-partment.Instrumentation Instrumentation required for digital recording of log i us. q and of potential-relaxation measurements over 5 decades of time, was as described in previous papers"*22 and referred to above. Cyclic voltammetry on the RuO, films was carried out using conventional equipment comprising a PAR potentiostat and function generator. Real Surface Areas In previous work,,' using thicker thermochemically prepared RuO, films, the rea1:apparent surface area ratios, R, were determined by the Krypton desorption B.E.T. method as ca. 450. However, with the present thin-film electrodes, R values were too small to be reliably measured, i.e.<ca. 50. R determinations at Ru02 by the double-layer capacitance method are inappropriate since Ru02 electrodes give a large 1.0 -4-t 0)e,c 0 3 LT aJ .-2 0.5-z Y s2 1....1.3 I 1,0 *I 0 50 100 depth/1016 molecules cm-' Fig. 2 Rutherford back-scattering depth profiles for relative Ru con- tent (normalized) N/N, in the thin-film RuO, electrodes. RuC13:TiCl,: 0, 80 :20; 0,60 :40; x ,40 :60 and A, 20 :80 in the electrode preparations. 30 x 10l6 molecules cmF2=ca. 100 nm thick- ness of film 728 redox pseudo-~apacitance~~~ in impedance or cyclic-voltam- metry measurements, and not a true double-layer capacitance, from which R values could be derived.Therefore, apparent current densities, I, are the quantities used for the kinetic data plots of Fig. 5 (q.u. later) and for derived capacitance values, based on the geometrical surface areas. IR-drop Corrections Even thin-film RuO, electrodes are quite active for anodic C1, evolution; therefore IR-drop corrections were required in the polarization measurements in the range q =0.1-0.4 V and were directly determined from the initial sharp fall of potential in the q(t) open-circuit decay transients in the usual way; they were correctly linear in I. Results and Discussion Characterization of the RuOz Electrodes by Cyclic Voltammetry For comparative purposes, cyclic voltammograms for each of the RuO, electrode preparations (cf: ref.5) were recorded at a sweep-rate of l00mVs-’ (Fig. 3). The qualitative and quantitative reproducibility of the cyclic voltammograms taken at the Ru02/Ti02 electrodes was excellent so that these results served to give ‘electrochemical fingerprints’ of the five oxide preparations and some relative indication (Table 2) of their electrochemically active surface areas. The currents gen- erated in the cyclic voltammetry, I = Cd V/dt, provide a meas- ure of the electrochemical ~apacitance,~.~ C, of the oxide 1.50-1.00-h E hI 0.50-Y i I Y 0.00-NI5 -0.50-a5-1.00-C 2 -1.50-2 vi --2.00 -2.50-t.’I I I 1 I 0 0.4 0.8 1.2 Fig. 3 Cyclic voltammograms of the five RuO,-TiO, electrodes in 0.1 mol dm-3 H,SO, solution.For (a) 1; (b)2; (c) 3; (d) 4 and (e)5, the thick-film electrode [(c) in Fig. 13 Table 2 Anodically injected charges at the five RuO,/TiO, electrodes between 0.00 and 1.4 V us. RHE electrode surface redox charge/ mC cm-, 0.44 0.87 1.42 1.84 4.68 J. MATER. CHEM., 1991, VOL. 1 systems, i.e. for RuO, a double-layer capacitance and the pseudo-capacitance associated with Ru redox processes in the film.5 The RuO,/TiO, film materials are usefully characterized not only by their compositions (Table 1) but also by the integrated values of the charge injected between two potentials (here 0.00-1.4 V us. RHE) (Table 2 and Fig. 4). Electrode 5 was coated with a thick film [see SEM photo, Fig. l(c)]. Primary Results and Derived Quantities The primary results are presented in the form of log(apparent current density) us.overpotential, q, relations coupled with the potential-relaxation transients plotted as log(time). Derived results are shown in the form of q us. log(-dq/dt) plots and as C us. q plots calculated from eqn. (l), using dq/ dt derived by differentiating the digital q(t) data of the transients and coupling the resulting data with the Tafel polarization results in the form i(q)= i, exp(aqF/RT) [from eqn. (l)]. The significance of log(-dqldt) us. q plots follows from eqn. (6) in relation to the Tafel equation in the form i(q)=io exp(aqF/RT) and any potential dependence of C, as discussed for other reactions in previous paper^'^.'^,^^,^^ and theoretically in ref.21, 19, 27. Thus, if C is virtually indepen- dent of q, e.g. when C,<<Cdl,log(-dqldt) us. q plots will have the same slopes as those of the respective polarization relations, as follows from eqn. (1). The aim of the analyses of results is the evaluation of C, and especially its C, component [eqn. (6)], as a function of potential in relation to the kinetic behaviour of the C1, evolution reaction at Ru02. In particular, if there is appreci- able and potential-dependent coverage by the C1 intermedi-ate, it will be expected (cf: ref. 21, 27, 28) that C, will have values appreciably larger than Cdl and possibly exhibit a maximum (cf: ref. 17, 23) as shown ‘theoretically’ in a simu- lation treatment.Ig Plots of Primary and Derived Data log( Current Density) vs.Potential Relations First we show the Tafel relations and respective potential- relaxation transients taken from several overpotentials corre- sponding to various initial I values at 298 K (Fig. 5, curves 1-4 for the four thin-film electrodes having different RuO, loadings). In general, the higher the concentration of Ru in RuO,/TiO, mixed films, the better is the electrocatalytic behaviour, but the structure of the oxide film surface can also play a significant role that may account for the behaviour of I I I5.00I /I1N 4 001 /*‘ I //*5 3.00 E */. g 2.00 /,I_ /* i I I0.00 -I 0.4 0.8 1.2 1.6 potential/\/ vs. RHE Fig. 4 Integrated values of the anodically injected charge from Fig.3 for the five RuO,,TiO, electrodes. For: +, 1; x, 2; 0,3; W, 4 and *, 5 J. MATER. CHEM., 1991, VOL. 1 0.301 10.251 .-c -2 0.15 e g 0.10 --0.05 Fig. 5 ZR-compensated Tafel plots for the C1, evolution reaction at the four thin-film RuO,,TiO, electrodes (1-4) in 0.1 mol dmP3 HCl- 0.9 mol dm-3 NaCl solution. 1-4 represent the kinetic behaviour at electrodes 1-4, respectively electrode 4, which is less active than electrodes 2 and 3 (Fig. 5). Thus, the SEM photo (Fig. 1) shows some extended 'black' areas which appear to be lower than the main surface area, corresponding to lack of uniformity in the Ru02/Ti02 film structure. The IR-corrected Tafel relations are all non-linear, with ap- proaches to limiting currents in the -1 to 0 region of log I.An experiment in which an RuO, electrode was rotated up to 6400 rpm (Fig. 6) was performed to establish whether the approach to limiting currents was due to diffusion limitation. Since similar behaviour is found even up to 4.1 mol dmP3 NaCl (see Fig. 8, later) it was unlikely that mass-transfer limitation could involve the C1- ion itself. However, (cf: ref. 17,25) supersaturation involving anodically generated C12 could give rise to such behaviour, and can be 'spun away' by rotation. However, there is only a rather small effect (Fig. 6) of electrode rotation (up to 6400rpm) on the shapes of the q us. log i relations, so the curvature must be attributed to an 0.4 + +* 0.' .-> .-m A activation-controlled kinetic effect lo rather than mass-transfer effects.On account of the curvature in the IR-corrected plots of Fig. 5 and 6, a definite value of the Tafel slope, as observe$ e.g. by Erenburg et aL7 (ca. 42 mV at other types of RuO, surfaces at low I) is not found. This difference probably depends very much on the surface state of the thermochem- ically or anodically formed RuO, deposit, its composition, its microscopic structure and the real :apparent area ratio or porosity. Such properties depend' on the baking temperature and the preparation history of RuOz materials,'.' whereas anodically and thermochemically prepared Ru02 films' can behave in substantially different ways. Hence, the parameters of electrocatalysis specifically depend on the state and prep- aration history of the RuO, surface at which the C1,-evolution kinetics are characterized, i.e.for 'RuO,' surfaces, there is probably no unique polarization behaviour that can be defined. Also anodically5 and thermochemically formed RuO, films can behave in substantially different ways. Since there are evidently continuously curved, activation- controlled relationships between q and log i (Fig. 5 and 6), tests were made by means of Conway-Novak plots" to examine whether the kinetic behaviour (apparent approaches to limiting currents) was associated with C1' recombination control [eqn. (3)]. Such tests involve" plotting i-1/2 exp(qF/ RT) us. exp(qF/RT); a resulting linear relation indicates recombination control and, from the slopes and intercepts of such plots, the recombination rate constant k3 and the quasi- equilibrium constant K1 for C1- *C1' +e- [eqn.(l)] can be determined." Such plots, shown in Fig. 7 for the four thin- film electrode surfaces at 298 K, evidently indicate the prob- ability of recombination control under the present conditions. This conclusion is not necessarily in conflict with that of Erenburg et aL7 since their preparations of the RuO, films were different from those in our experiments. Recombination control in Cl, evolution is also found, but at Pt, in other works on the C1, evolution reaction."-12 That the ascents of currents at RuO, towards limiting values with increasing q are not due to progressive increase of oxide-film formation, as is possible at Pt,13914 follows from: (a) the fact that a relatively thick RuO, film is already thermochemically formed to a definite extent; (b)the log I us.q relation is the same for descending as for ascending changes of q. Behaviour at Higher C1-Concentration In Fig. 8 are compared the log I us. q relations for [Cl-] = 0.1, 0.5, 1.0 and 4.1 mol dm-3 at pH 1. It is clear that the I I I I ow 0.1 0.2 03 04 log (//A crn-') lo3 exp (VF/RT) Fig.6 Tafel plots for the C1, evolution reaction at a rotated RuO, Fig. 7 Conway-Novak plots for testing recombination-control in the electrode for various rotation rates in 0.1 mol dmP3 HC1-0.9 mol mechanism of anodic C1, evolution based on the data of Fig.5 for dm-3 NaCl solution: +, 400; *, 900; A, 2500; 0,3600; x ,6400 rpm electrodes 1-4. A, 1; 0,2; V, 3; 0,4 0.501 I I 1 I 1 0.40I 4 0.00-4O.lOL 0 log (//A cm-*) Fig. 8 Tafel plots for the C1, evolution reaction at electrode 4 at four C1- concentrations at constant ionic strength: *, 0.1; V, 0.5; x, 1.0 and 0, 4.1 mol dm-3 C1-. T=298 K non-diffusion-controlled limiting currents that are approached with increasing q, increase with [Cl-), but this increase itself appears to be approaching a limit as [Cl-] is raised from 1.0 to 4.1 mol dm-3. For a recombination-controlled process, proceeding at its limiting rate, the reaction-order should be zero in [Cl-1, according to conventional representations of the kinetics, since &--.being 1 is no longer influenced by kl (and kk1). Therefore, the continuous increase of Ilimwith [Cl-] can be J. MATER. CHEM., 1991, VOL. 1 interpreted only as due to a dependence of k3 itself on C1- concentration and this must be attributed to an effect of coverage by adsorbed C1- ion since no direct effect of the concentration of C1- in solution could arise. We suggest this effect of [Cl-] on Ilimcould arise on account of the interaction between adsorbed C1- and Cl', weakening the adsorption of the latter, leading to a more facile recombination desorption. Similar effects were observed in the recombination-controlled kinetics of Cl, evolution at Pt.29 It follows that if C1- adsorption can affect the rate constants of steps in the C1, evolution mechanism (I), (2), and in particular (3), reaction-order determination as a means of evaluation of the reaction mechanism can be unreliable, or, rather, the conventional approach3' in interpreting (dln I/ dlnc), becomes inapplicable.Further information on electrocatalysis in anodic C1, evol- ution at these RuO, surfaces is afforded by the potential- relaxation experiments and derived data, to be discussed below. Potential-Relaxation Transients, q(t), and Derived Information Potential-relaxation data, plotted as log(time), derived from digital recording of the q(t) transients following interruption of currents at five different potentials for prior steady-state C1, evolution are shown in Fig. 9(a) and (b)for two of the four RuO, surfaces.Apart from the initial potential fall in the first 20 ps, all transient plots show two distinct regions with an arrest developing at qx30 mV. Rotation of the electrode to 3600 rpm produced [Fig. 9(c) and (d)] only a small effect on the course (but not the shape) of the q(t) us. log t plots and on the plots of q(t) us. log(-dqldt). L 0) 0 0.301 I0.30r-----l >.- I R1.- c c) Q)c 0Cl L Q) 0 0.10 -6 -5 -4 -3 -2 -1 n' 1I 0 1 0.05 0.00 -6 -5 -4 -3 -2 -1 0 1 Fig. 9 Overpotential us. log(time) plots for potential relaxation at two of the four thin-film RuO, electrodes corresponding to the polarization relations of Fig. 5: (a) 1; (b)2; (c) 5 at zero rotation rate and (d) 5 at 3600 rpm.Numbers indicate initial polarization potentials corresponding to the respective Tafel relations shown in Fig. 5: (1) 0.34; (2) 0.20; (3) 0.17; (4) 0.16; (5) 0.15 V. Annotations I-IV indicate distinguishable sections of the curves J. MATER. CHEM., 1991, VOL. 1 73 1 0 . 3 0 ~ 0.251 (a (b 1 .> O.l5t .2l 1 0.10- , .: ,_.::.' 1 2-' 3 . 4, 0.10 0.05-/ 10.05u0.00-5 -4 -3 -2 -1 0 1 2 3 0.00-4 log (-dV/dt) log (-dV/dt) (c >'2-0.20-.-c. c a,c E g 0.10-.A*0.00 '-Fig. lO(a) and (b)show the q(t)us. log(-dq/dt) plots resulting from the application of eqn. (6)in log form to the results of Fig. 9(a) and (b). These plots will evidently be similar [cj eqn. (6)] to the corresponding Tafel plots only if C is indepen- dent of q.I7 It is clear that this is not the case and the evaluation of C from the dq/dt and the q us.log i information then gives the potential dependence of C as shown in Fig. 1l(a) and (b). These plots all show a rapid rise of C with decreasing q, below q z 50 mV, corresponding to the flattening out or arrest in the q(t)us. log t plots of Fig. 9(a)and (b).Results of similar experiments on the behaviour of the adsorbed intermediates in the H2, O2and C1, evolution reactions at Pt17,18,23 show a different behaviour within 50-10 mV of the respective reversible potentials, where the maxima in C 40.00 5 t are observed. It is possible that a maximum in C for C1' adsorption does exist at lower q values but is not observable over the range of potentials covered. However, in some experiments, with potential relaxation measured from higher overpotentials, a maximum in C for RuO, electrode 2 as an example is actually realisable, as shown in Fig.ll(b) and the peak is well resolved from the continuing rise of C at lower potentials. It may be thought that the rapid ascent of values of C which arise when q x 50 mV is an artefact due to the appear- ance of the back-reaction current. However, (a) C is deter- mined by dq/dt and the net observed current in eqn.(6), i.e. including any back-reaction component, and (b) a similar situation arises in the case of the HER at active Pt or at Ni, yet clear maxima in C are actually 0b~erved.l~ Therefore, the U .E 30.001 ?: 1aJ ctu c.-I0 n 9 *. ... . . . 1 . I . I I 0 0.05 0.1 0.15 0 0.025 0.05 0.075 0.10 overpotential/V overpotential/V Fig. 11 Capacitance us. q plots derived from the dq/dt and log i as Art) at 298 K; results for electrodes (a)2; (b) 2 from an initial q of 0.30 V. Numbers refer to initial polarization potentials as indicated respectively on Fig. 9. Capacitance values are calculated per apparent cm2 732 J. MATER. CHEM., 1991, VOL. 1 rapid rise of C with decreasing q for all four electrodes is a I1 true characteristic feature of the reaction at these Ru02 surfaces. For some other surfaces examined, the rise in C sets in at somewhat higher q values, 0.07-0.1 V, than for (a) and (4.At q ~0.04V, the apparent values of C rise to ca.40 mF per apparent cm2. While it was not possible to obtain an estimate of the real area per apparent cm2, previous estimates of the real :apparent area ratio, R, for thermochemically formed RuO, films give a figure R =300-400; then, the above value of C would be ca. 115 pF ern-,. Since the present electrodes appear smoother in SEM photos [Fig. l(a), (b)] than DSA-type surfaces [Fig. l(c)], a smaller R value may reasonably apply, giving C, possibly of the order of 500-1000 pF cm-, in the rising limit, i.e. in the range e~pected'~~~~ for appreciable coverage by Cl'. Comparative experiments conducted at a lower tempera- ture, 278 K, gave qualitatively similar results to those at 298 K but the following differences are significant: (i) values at a given q are less, as expected (activation energy effect); (ii) the potential-relaxation transients, taken from different initial potentials (exemplified in Fig.12) do not tend to become superimposable at longer times (say log t> -3) in the e~pectedl~.~'way, as actually observed for the HER at Nil7 or Pd.31 This is confirmed by the clearly non-coincident series of piots of q(t) us. log(-dqldt) [for which the integration constant z (ref. 32) of eqn. (6) has no influence on the relative shapes of such plots] shown in Fig. 12(b) for 278 K, corre-sponding to the respective data in Fig. 12(a). The derived C us. q plots [Fig. 13(a) and (b)] for 278 K also depend, correspondingly, much more on the initial potential from which potential-relaxation was initiated, than for 298 K, which suggests that the state of the Ru02 surface determining the 1 V.I" 0.10- 0.05- 0.00 -6 -5 -4 -3 -2 -1 0 1 2 log (t/s) 0.25-(b) .l -0.201 log (-dV/dt) Fig. 12 (a)q us. log t plots for electrode 1 taken at 278 K from the following initial overpotentials: (1) 0.5; (2) 0.45; (3) 0.40; (4) 0.35 and (5) 0.30 V. (b)q us. log(-dq/dt) plots, corresponding to the respective curves of Fig. 12(a) iI (a) 0 0.05 0.1 0.15 0.2 0.25 overpotential/\/ 0 0.05 0.1 0.15 0.2 0.25 overpotential/V Fig. 13 Capacitance us. q plots for electrodes (a)1 and (b)4 from the data of Fig.12(a) and (b) for 278 K, and corresponding q us. log i data. Designation of initial potentials as in Fig. 12(a); T=278 K. Capacitance values are calculated per apparent cm2 kinetics of the C1, evolution reaction, and particularly the adsorption of Cl', depends appreciably on the initial anode potential. Such an effect could arise from a potential-or temperature-dependent change of oxidation state of the RuO, surface; for example, depending on the potential, higher oxidation states of Ru in the surface, up to Ru", can arise, as indicated by ESCA33 and indirectly from the charge injection data shown in Fig. 4. Mechanisms of Potential Relaxation In the absence of significant coverage by an adsorbed inter- mediate, the potential relaxation is determined simply by the Tafel characteristic i =io exp(aVF/RT) (expressed as a Fara- daic reaction resistance) in combination with the double-layer capacitance, as shown by Butler and Arm~trong.~~ When step (2) is rate-controlling, the intermediate (C1 *) must be desorbed on open-circuit, predominantly by the coupled anodic and cathodic processes:21 M,,Cl' +C1-+Mox +C1, +e-(2) e-+Mo,C1' +Max +C1-(-1) since the charge that can be dispensed from the double-layer capacitance is normally only some 10% of that associated with electrochemical deposition or desorption of a monolayer of an intermediate, such as Cl'.In the case of the recombination process (3), the situation must be different since no passage of charge is directly associated with the desorption step, as it is in (2).Hence, the open-circuit desorption of C1 must proceed by continuing recombination in (3), with the potential being determined by J. MATER. CHEM., 1991, VOL. I electrosorption associated with quasi-equilibrium in step (1): e ___ exp(go)=Klccl-exp( qF/RT) (7)1-8 where g is a lateral interaction parameter.27 From dO/dq, the resultant pseudo-capacitance is given by C,=--9w1 cc1-exp(?F/R TI/[ 1+K 1cc1-exp( V/R 771RT (8) or, in terms of 8 from eqn. (7), where q1 is the charge required for monolayer coverage (6-1) by the intermediate. Then, eqn. (6)for the potential relaxation rate becomes -Cdq/dt = -qlFK1 ccl-exp(qF/RT)~ RT x [l +Klccl-exp(qF/RT)I2dq/dt =2Fk3t12=2Fk3 1+Klccl-exp(qF/RT) i.e.2RT 41 -dq/dt =-k3K ccl -exp( qF/R T) (11) Integrating eqn. (1 1) gives -"'..pi 41 -")=RT k3K,ccl-(t+z) (12) where z is an integration In log form, we find ln(t+z) (13) i.e. the slope of the logarithmic potential decay curve would be expected to be dq(t)/d ln(t+z)= -RT/F (14) The experimental potential relaxation curves (Fig. 9) are more complex than is indicated by eqn. (14) and show three or four distinguishable regions marked I-IV on Fig. 9. Region I is simply where z is comparable to or greater than t giving the usualI7 run-in effect when q is plotted as Alog t). This region is also influenced by the dielectric capacitance of the underlying oxide film on the Ti as shown in ref. 20. The well defined intermediate regions in Fig.9 labelled I1 do have slopes dq/d log t approaching -2.3 RT/F, corre-sponding to eqn. (14); however, these regions are followed by those labelled I11 [Fig. 9(a) and (b)] and IV having substan- tially smaller slopes. These regions correspond to the rapidly rising sections of the C us. q plots of Fig. 1 l(a) and (b)or to the maximum in Fig. ll(c). As in the other mechanism, a small net flow of charge from the double-layer capacitance must always" pass through the reaction pathway as the potential declines from its initial value, at the prior steady-state current, down to lower values towards the reversible potential, but this charge will normally be substantially smaller than that corresponding to the change of 8 of Cl'.An 'arrest' in the potential-relaxation transient arises" when the declining potential becomes 'buffered' around 8zO.5, according to the form of eqn. (7) which is similar to that for a redox titration curve. 733 capacitance/pF cm-2 0.00 20.00 60.00 80.00 0.40 >2- I0.00I 1 --' I _c-> 1 -4 -3 -2 -1 0 1 23 log (//A crnp2) Fig. 14 Numerical simulation of the q us. log i (solid lines) and C(q) us. q behaviour (dashed lines) taking the rate constants k2= 10-l2 and k,=4.87 x10-* and the quasi-equilibrium constant for step 1 as 0.5 with q=7 pC cm-2. Curves (a), (b) and (c) are calculated for interaction parameter g values of 0, 2 and 5, respectively Relation of C1 Adsorption to Recombination-controlled Kinetics The approach to what appear to be recombination-controlled limiting currents (cJ Fig.7 and ref. 10) in Fig. 5 and 6 suggests that levels of full coverage by C1' are approached. According to the basis of the Conway-Novak plots" of Fig.7, the approach to full coverage should follow a Langmuir (g=O) or Frumkin type (g#O) isotherm [eqn. (7)]. The corresponding pseudo-capacitance relations for non-equilibrium conditions were worked out by Gileadi and C~nway~~ for various values of g. The sharpness of the ascent of C values when q 550 mV (at 298 K), albeit in the absence of an observable maximum, suggests35 that g actually takes negative values (see below), so that 8 can change from a small value to near unity over a relatively small potential range.[For Langmuir conditions (g= 0), this range for 0.5 <8<0.95 is ca. 200 mV.] In connection with C1' recombination control in the C1, evolution mechanism, the course of calculated log(current) us. potential relations can be plotted for a selected set of rate constants, in relation to the corresponding profiles of C, us. overpotential, on the same scale of potentials, as in Fig. 14. Such simulation plots are shown for three g values in the Frumkin-type isotherm, eqn. (7), from which the C, us. q [eqn. (10) or using (1 1) with (9)] curves of Fig. 14 have been calculated. The C, us. q relations, which relate the C1' coverage to q differentially, are also shown in Fig. 14 as the dashed curves. It is seen (a) that the approaches to the expected limiting currents do not arise until the maxima in C, us.q have been passed and (b) that the range of potentials over which curved Tafel relations arise, become much extended with increasing positive g values. Correspondingly, the range would become narrower35 if g <0. These effects are as expected on the basis of lateral intera~tion.~~ The positions of the maxima on the potential axis depend on the value chosen for K, and on the g values indicated in the caption of Fig. 14. Grateful acknowledgment is made to the Natural Sciences and Engineering Research Council of Canada for support of the main body of this work with Mr. Gu Ping and by B.E.C. to Professor A. de Battisti for providing facilities for a short leave of absence at the University of Ferrara where this work was commenced.734 J. MATER. CHEM., 1991, VOL. 1 References 17 B. E. Conway and L. Bai, J. Chem. SOC., Faraday Trans. 1, 1985, 1 2 3 4 D. M. Novak, B. V. Tilak and B. E. Conway, Modern Aspects of Electrochemistry, ed. B. E. Conway, J. O’M. Bockris and R. White, Plenum, New York, 1980, vol. 14, ch. 4. S. Trasatti and G. Lodi in Electrodes of Conductive Metal Oxides, Part B, ed. S. Trasatti, Elsevier, Amsterdam, 1981. A. E. Newkirk and D. W. McKee, J. Catal., 1968, 11, 370; see also G. W. Jang and K. Rajeswar, J. Electrochem. SOC., 1987, 134, 1830. D. Galizzoli, F. Tantardini and S. Trasatti, J. Appl. Electrochem., 18 19 20 21 22 81, 1841. B. E. Conway and T. C. Liu, Langmuir, 1990, 6, 268; J.Chem. SOC., Faraday Trans. I, 1987,83, 1063. D. A. Harrington and B. E. Conway, J. Electroanal. Chem., 1987, 221, 1. B. E. Conway and T. C. Liu, Ber. Bunsenges. Phys. Chem., 1987, 91, 461. B. V. Tilak and B. E. Conway, Electrochim. Acta, 1976, 21, 745. B. E. Conway, L. Bai and D. F. Tessier, J. Electroanal. Chem., 1984, 161, 39. 5 6 7 8 9 10 11 1974, 4, 57. S. Hadzi-Jordanov, M. Vukovic, H. A. Kozlowska and B. E. Conway, J. Electrochem., SOC., 1978, 125, 1471. S. Trasatti and G. Buzzanca, J. Electroanal. Chem., April 1971, 29, App. 1, following p. 448. R. G. Erenburg, L. I. Krishtalik and V. I. Bystrov, Elektrokhimiya, 1972,8, 1240; R. G. Erenburg, L. I. Krishtalik and I. P. Yatoshev- skaya, Electrokhimiya, 1975, 11, 1068. B. E. Conway and J.Mozota, Electrochim. Acta, 1983,28, I and 9. S. Trasatti and G. Buzzanca, J. Electroanal. Chem., April 1971, 29. D. M. Novak and B. E. Conway, J. Electroanal. Chem., 1977, 99, 133. G. Faita and G. Fiori, J. Appl. Electrochem., 1972,2,31; G. Faita, 23 24 25 26 27 28 29 30 31 B. E. Conway and Gu Ping, J. Chem. SOC., Faraday Trans., 1990, 86, 923; Gu Ping, Ph.D. Thesis, University of Ottawa, 1990. R. de Levie, J. Electroanal. Chem., 1965, 9, 117; Electrochim. Acta, 1965, 10, 113. I. V. Kadija, J. Electrochem. SOC., 1964, 131, 601. A. G. C. Kobussen, H. Willems and G. H. Broers, J. Electroanal. Chem., 1982, 142, 67 and 85. E. Gileadi and B. E. Conway, J. Chem. Phys., 1963,39, 3420. A. Matsuda and T. Ohmori, J. Res. Inst. Catal., Hokkaido University, 1962, 10, 203 and 215. D. M. Novak and B. E. Conway, J. Chem. SOC., Faraday Trans. I, 1981, 77, 2341. K. J. Vetter, Z. Phys. Chem., 1950, 194, 284. M. Elam and B. E. Conway, J. Electrochem. SOC., 1987, 135, 1678. 12 G. Fiori and N. Nidola, J. Electrochem. SOC., 1970, 117, 1333. W. E. Triaca, C. Solomons and J. O’M. Bockris, Electrochim. 32 H. B. Morley and F. E. W. Wetmore, Can. J. Chem., 1956, 34, 359. 13 14 15 16 Acta, 1968, 13, 1949. E. L. Littauer and L. L. Shrier, Electrochim. Acta, 1966, 11, 527. S. G. Roscoe and B. E. Conway, J. Electroanal. Chem., 1987, 224, 183. D. M. Novak and B. E. Conway, J. Chem. SOC., Faraday Trans. I, 1979, 75, 244. D. W. F. Hardie, Electrolytic Manufacture of Chemicals from Salt, The Chlorine Institute, Atlanta, 1975. 33 34 35 J. Augustynski, L. Balsenc and J. Hinden, Proc. Chem. SOC. Symp. Novel Electrode Materials, Brighton, U.K., 1975; J. Electrochem. SOC., 1978, 125, 1093. J. A. V. Butler and J. F. Armstrong, Trans. Faraday SOC., 1933, 29, 11261. H. Angerstein-Kozlowska, J. Klinger and B. E. Conway, J. Elec-troanal. Chem., 1977, 75, 45. Paper 0/05279E; Received 23rd November, 1990

ISSN:0959-9428    

DOI:10.1039/JM9910100725

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

J. MATER. CHEM., 1991, 1(5), 735-738 Chemical Reduction of FeCI,-Graphite Intercalation Compounds with Potassium-Naphthalene Complex in Tetrahydrofuran Ali Messaoudi,a Michio Inagaki" and Franqois Beguinb a Faculty of Engineering, Hokkaido University, Kita-ku, Sapporo, 060 Japan Centre de Recherche sur la Matiere Divisee, CNRS la, rue de la Ferollerie, 45071 Orleans Cedex 02, France The reduction reaction between FeCI, intercalated into graphite and potassium has been studied. The reduction of FeCI, to metallic Fe and KCI occurred in the graphite gallery and, as a consequence, the reduction products were supposed to be confined within the graphite matrix. By coupling energy-dispersive X-ray spectroscopy (EDX) and X-ray diffraction (XRD) analyses of the products obtained during the reaction, it was concluded that the reaction between tetrahydrofuran(THF)-solvated K and FeCI, was the rate-determining step, and not the diffusion of THF-solvated K ions into the graphite gallery. Experimental results indicated that a strong constraint for the reaction was the limited space in the graphite gallery.When heated at different temperatures, the flakes of the product showed remarkable exfoliaton with subsequent appearance of metallic Fe and KCI as reaction products on the surface. Keywords: Graphite; Intercalation; Reduction; lron(II1) chloride; Exfoliation It has been pointed out by various authors that the catalytic activity of transition metals depends not only on their disper- sion state but also on the nature of their support.So far, several studies have been carried out on the preparation of catalysts of finely dispersed transition-metal particles on graphite using graphite intercalation compounds (GICs) as starting materials. Principally, two reactions have been used to disperse fine metallic particles on graphite: reduction of metal chlorides by a binary K-GIC, KC8, in an organic solvent' -3 and reduction of transition-metal chloride-GICs by potassium vapo~r.~.~ In the former, the metallic particles are formed on the surface of graphite, but are easily oxidized during handling in The catalysts thus prepared, there- fore, have two disadvantages: they cannot be handled in air and their support has a small specific surface area.The latter reaction results in very stable products in which the fine metallic particles formed are mostly included in the graphite matrix; however, they do not show high activity as catalysts. In the present work, the reducing reaction of iron(II1) chloride in the graphite gallery with a potassium-naphthalene complex in THF at room temperature was followed kinetically to prepare finely dispersed metal particles. The metal particles formed were mostly included between the graphite layers and, as a conse- quence, were inert to air. However, these metal particles could be activated by exfoliation of the graphite support. Experimental The pure FeCl,-GICs used as starting materials were synthe- sized by heating mixtures of anhydrous iron(@ chloride and natural graphite powder with an average particle size of 400 pm in a sealed tube; the stage (1) compound was prepared at 300 "C with a graphite: FeCl, ratio of 4 and the stage (2) at 500 "C with a graphite: FeCl, ratio of 10.The GICs were then separated from the unreacted FeCl, by washing with water. A large particle size for the host graphite flakes was chosen to make it easier to follow the kinetics of the GIC reducing reaction. The ideal chemical formulae for stage (1) and (2) FeC1,-GICs have been shown to be C6FeCl, and CI2FeCl,, respectively. However, the starting GICs used in the present work have a lower content of FeCl,, as shown by the presence of free graphite in the XRD patterns (Fig. 1 after 7hI 12 after 13h I 1 I 1 I I 10 20 30 40 50 60 261' Fig.1 Evolution of X-ray powder pattern from stage (1) FeC1,-GIC with reaction time in the THF solution of K-naphthalene radicals and 2); the free graphite is formed by decomposition during washing. After washing, the GICs were very stable, with no change in the XRD patterns over long periods of time, as reported previously. The aromatic anion radical, potassium-naphthalene, was prepared under an argon atmosphere in anhydrous THF, the FeC13-GIC, a. stage 2 o 0. hl al0 cl c h after 1 h I I I 1 I I 10 20 30 40 50 60 201' Fig. 2 Evolution of X-ray powder pattern from stage (2) FeC1,-GIC with reaction time in the THF solution of K-naphthalene radicals concentration of K being roughly 0.5 rnoldm-,.THF was purified in advance by refluxing under an argon atmosphere with a sodium wire for 5 h and then distilled. The formation of the aromatic anion radical was instantaneous and so the solution turned green immediately after adding potassium to the THF solution of naphthalene. The FeC1,-GIC was then added to the thus prepared K-napthalene-THF solution. A large excess of K in THF solution was used to reduce intercalated FeCl, completely. The suspension of GIC par- ticles was stirred at room temperature under an argon atmos- phere. During the reaction, a small amount of the solution was taken out by pipetting; the solid products were filtered off and then washed several times with acetone and finally with a water-ethanol mixture in order to remove adsorbed potassium and naphthalene.The reaction products thus sampled were analysed by XRD and EDX. They were heated up to different temperatures in the range 300-1000 "C under a flow of pure nitrogen gas and examined by scanning electron microscopy (SEM). In order to avoid any species being adsorbed on the surface of the graphite particles, the flakes of the products were sometimes cleaved by using an adhesive tape just before EDX analysis. Results Reduction of FeC1, in the Graphite Gallery The XRD diagram in Fig. 1 and 2 correspond to the reaction products obtained from the stage (1) and (2) FeC1,-GICs, respectively, after different reaction times. In both cases we can observe progressive decreases in diffraction intensities of 001 lines of the starting GICs.The reaction seems to proceed faster from the stage (2) GIC than from the stage (1). Note that the decrease in diffraction intensities for the starting intercalation compounds is accompanied by the appearance and growth of new diffraction lines which can be J. MATER. CHEM., 1991, VOL. 1 easily attributed to the reaction products, Fe metal and KC1. The diffraction lines due to Fe metal were weak because of the extremely small size of the particles, as will be discussed later. In any case, no lines attributable to Fe-GICs, as stated by Volpin et aL9 were identified. After complete destruction of the starting intercalation structure [13 h for the stage (I) and 6 h for the stage (2)], the XRD patterns showed only trace amounts of graphite, suggesting extensive destruction of the graphite structure during the reducing reaction.These changes in the XRD patterns, also imply that the starting- stage structure persists until complete decomposition, and any observable stage transformation is not apparent. Both Fe metal and KCl thus formed were very stable; no oxidation of Fe and no decrease in the relative amount of KCl were observed during washing. This suggests that neither could be leached away by water; they are probably included in the graphite matrix despite its extensive destruction as mentioned above. It was impossible to remove KCl even by washing with concentrated HCl solution (6 mol dm-3).The particle sizes of KCl and Fe in the final reduction products, calculated from the line broadening of XRD, were ca. 16 and 9 nm, respectively. Fig. 3 and 4 show representative results for the distribution of K, Fe and C1 in the GIC flakes, measured by EDX. After a short reaction time (e.g. 10 min), the distribution profiles of Fe and C1 are almost the same, consistent with XRD results, but, surprisingly K is detected even at the centre of the flakes of the stage (1) GIC as well as those of the stage (2)[Figs 3(a) and 4(a), respectively]. For the final products obtained after a long reaction time of 20 h, the EDX analysis Fig. 3 The distribution of (b) K, (c) C1 and (d) Fe along the line indicated by the arrows on the flakes (a) obtained from stage (1) FeC1,-GIC: Left-hand side, after reaction for 10 min; right-hand side, after reaction for 20 h J.MATER. CHEM., 1991, VOL. 1 Fig. 4 The distribution of (b) K, (c) C1 and (d) Fe along the line indicated by the arrows on the flakes (a) obtained from stage (2) FeC1,-GIC: Left-hand side, after reaction for 10 min; right-hand side, after reaction for 20 h indicates that the flakes still contain Fe, C1 and K, the amount of K increasing [Fig. 3(b) and qb)]. The chemical compo- sitions of the final product obtained from the stage (1) GIC was measured as C19.3FeK2.6C12.5, which is consistent with the XRD pattern showing the presence of KCl and Fe. The distribution profiles of K in the flakes seem to be very similar to those of C1 and Fe, which suggests the existence of three elements as KCl and Fe.For intermediate products, the coexistence of three elements with similar distribution profiles was also observed in the same flakes. These EDX observations reveal very rapid diffusion of THF-solvated potassium towards the centre of the flakes. However, XRD shows that it did not immediately destroy the structure of the starting GIC. The diffraction lines of the starting GIC became broader and weaker with reaction time (Fig. 1 and 2), which indicates a decrease in the amount of GIC and also the increase of stacking disorder in GIC structure. This experimental result suggests that the reaction of FeCl, with THF-solvated potassium in the graphite gallery is not as fast as that of free FeCl, with K in THF.From the SE micrographs presented in Fig. 5(a)and (b),the reaction in THF is found to provoke a pronounced decrease in the size of the flakes. This seems to be consistent with the assumption that the intercalated FeCl, is reduced with the THF-solvated K in the graphite gallery, which results in the destruction of the layer stacking in the graphite structure. Morphology Change of the Reaction Product When the final products were heated up to high temperatures under nitrogen, appreciable exfoliation of the flakes occurred Fig. 5 SE micrographs of the flakes: (a) the starting stage (1) FeC1,-GIC; (b) after complete reduction for 20 h (the final product); (c) after heating the final product at 300 "C for 1 h; (6)after heating at 400 "C for 1 h; (e) after heating at 500 "C for 1 h; and (f)-(h) after heating at 1000 "C for 1 min [Fig.S(c)-(f)]; this was more remarkable the higher the tempera- ture. At temperatures higher than 300 "C, minute particles appear on the surface of the exfoliated flakes. When the products were heated at 1000 "C for 1 min, one could see clearly cubic particles of different size and minute particles with undefined shapes [Fig. 5(f)-(h)].The cubic particles were characterized by EDX to be KCl. The reduced product obtained from the stage (2) FeC1,-GIC showed the same exfoliation behaviour. In addition to the exfoliation of the graphite matrix and the appearance of the particles, the restoration of the graphite structure by heating was observed on XRD patterns.The potassium chloride in the product exfoliated at 500 "C could be removed easily by washing with a mixture of water and ethanol, but at the same time Fe metal was quickly oxidized to a-Fe203, which were clearly detected by XRD. Discussion For short reaction times (10 min), XRD patterns of the products show that the starting structure is retained, but EDX analysis indicates the presence of potassium even at the centre of the flakes. Potassium must, therefore, be in the same gallery as FeCl,. The reaction between intercalated FeC1, and potass- ium is reasonably supposed not to occur at this stage, because the structure is retained and it is known that KC1 is not intercalated into the graphite gallery." At this reaction stage, potassium is expected to be solvated by THF molecules in the gallery." Therefore, the present work reveals that the diffusion of THF-solvated potassium ions, which have the relatively large thickness of 0.55 nm, is very fast even when the graphite galleries are occupied by FeCl, in advance.It can be assumed that the diffusion of THF-solvated potassium ions is much easier for an unoccupied gallery than for an occupied. If so, a kind of bi-intercalation compound must be obtained from the stage (2) FeC1,-GICs. However, this was not the case under our experimental conditions because the stage (2)structure was retained even though large amounts of potassium were detected by EDX. So, the THF- solvated potassium ions seem to diffuse preferentially into the occupied gallery.After complete reduction of FeC1,-GIC, the reaction prod- ucts, Fe and KCl, were detected, but only a trace of graphite. This is most likely due to the disturbance of the parallel stacking of the graphite layers by the inclusion of the reaction products Fe and KC1. After exfoliation at high temperatures, the 002 line of graphite reappeared, owing to the partial recovery of the graphite structure by expulsion of the included Fe and KC1 particles. On the way to complete reduction of FeCl, intercalated into graphite, the starting stage structure was preserved, though the size decreased. The reaction between FeC1, and THF-solvated potassium in the gallery seems to be the rate- determining step in the whole reducing reaction because the diffusion of the solvated K ions into the graphite gallery is extremely fast as discussed above.This is unexpected because the reducing reaction between free FeC1, and K in solution is instantaneous. In the present case, therefore, the limited space within the graphite gallery must be a strong constraint for the reaction. The following experimental results also suggest the strong effect of limited space: (1) there was not much change in the distribution of Fe and C1, the latter bonding with Fe in the beginning of the reaction and with K at the end of the reaction; (2)the fragmentation of the graphite layers occurred during the reaction; (3) a certain number of THF molecules were probably trapped in the graphite gallery, which caused exfoliation of the graphite matrix on heating.A similar exfoliation phenomenon has been observed on ternary GICs with K and THF.12 J. MATER. CHEM., 1991, VOL. 1 It is supposed that the reduction product of metallic Fe particles is confined within the graphite gallery together with KC1 and possibly THF molecules. It was reasonable to expect the metal particles to be inactive because they were enclosed within the graphite matrix; indeed the reaction products could be handled in the air. These metallic particles could be activated by heating to exfoliate the graphite matrix. After exfoliation at 300 "C the size of Fe metal particles was so small that it was difficult to identify them under SEM, and they were so active that they were immediately oxidized by washing with water.The present products prepared from the reduction of metal chloride intercalated into graphite by THF- solvated potassium seem to give us the following advantages over other similar catalysts: (i) they can be handled in air and (ii) they can be converted to finely dispersed metal particles supported on graphite with high surface area by simple heating just before their use. High activity can be expected because of the small size of the metallic particles and the high surface area of the graphite support after exfoliation. This work was partly supported by a grant for the Inter- national Joint Research Project from the NEDO, Japan.References 1 P. Braga, A. Ripamonti, D. Savoia, C. Trombini and A. Umani- Ronchi, J. Chem. SOC., Chem. Commun., 1981,40. 2 M. Inagaki, Y. Shiwachi and Y. Maeda, J. Chim. Phys., 1984, 81, 847. 3 A Messaoudi, R. Erre and F. Beguin, Carbon, 1991, 29, 515. 4 G. Bewer, W. Wichmann and H. P. Boehm, Mater. Sci. Eng., 1977, 31, 73. 5 R. Erre, A. Messaoudi and F. Beguin, Synth. Met., 1988, 23,493. 6 P. Kaiser, A. Messaoudi, D. Bonnin, R. Erre and F. Beguin, J. Chim. Phys., 1989,86, 1787. 7 A. Furstner and H. Weidmenn, J. Chem. SOC., Dalton Trans., 1988,2023. 8 Z. D. Wang and M. Inagaki, Synth. Met., 1988, 26, 181. 9 M. E. Volpin, N. Novikov, N. D. Lapkina, V. I. Kasatochkin, Y. T. Struchov, M. E. Kazakov, R. A. Stukan, V. A. Povitskij, S. Karimov and A. V. Zvarikina, J. Am. Chem. SOC., 1975, 97, 3366. 10 M. Inagaki and Z. D. Wang, Synth. Met., 1987, 29, 1. 11 M. Nomine and L. Bonnetain, C. R. Acad. Sci., 1967, 264. 12 M. Inagaki, K. Muramatsu and Y. Maeda, Synth. Met., 1983,8, 335. Paper 1/00333J; Received 23rd January, 1991

ISSN:0959-9428    

DOI:10.1039/JM9910100735

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

J. MATER. CHEM., 1991, 1(5), 739-746 Porous Chromia-pillared a-Zirconium Phosphate Materials prepared via Colloid Methods Pedro Maireles-Torres," Pascual Olivera-Pastor,*" Enrique Rodriguez-Castellon," Antonio Jimenez- Lopez" and Anthony A. G. Tomlinson*b a Departamento de Quimica lnorganica, Cristalogra fia y Mineralogia, Universidad de Malaga, Apartado 59,29071 Malaga, Spain 1. T.S.E., Area delta Ricerca di Roma, C. N.R., Monterotondo Staz., 00016 Rome, Italy The reaction of Cr(CH3C0,), [Cr(OAc),] with colloidal n-propylammonium a-zirconium phosphate and subsequent calcination of the products have been investigated. Depending on [Cr(OAc),] :[initial phosphate] ratios and Cr3+ concentrations, a series of polyhydroxy acetato-Cr3+ intercalated precursor materials can be obtained, in which topotactic interface reactions have occurred to give materials with interlayer distances (doo2)ranging from 13.0 to 39.0A.These precursors show higher layer expansions than the analogous pillared clays (PILCS; doo,= 16.8-27.6A).A model invoking ordered in situ polymerisation of the Cr(OAc), on the phosphate surfaces is put forward. Calcination of these precursors under N, (400 "C) leads to a series of chromia-pillared materials in which the interlayers do not collapse to a single (much lower) value, as found previously for most PILCS, but instead provide a wide range of interlayer distances (10-27A).These correspond to free heights of 3.5-20.5k the widest ranging and highest yet found for such materials. These nanoscale oxide-pillared materials have N, surface areas (B.E.T., 77 K) of 250-330 m2 g-', with pore radii (cylindrical pore method) ranging from 8.5 to 13.8A, and very narrow pore-size distributions.Calcination conditions are crucial for obtaining porous solids. If calcination is carried out in air at 400 "C, although pillared powders and films are again obtained, surface areas are only ca. 40 m2 g-' (B.E.T., N,, 77 K). Furthermore, higher calcination temperatures (500 "C, under N,) give rise to X-ray amorphous materials, again having high surface areas and narrow pore-size distributions. All the materials can be processed in thin- film form without loss of textural characteristics. Keywords: a-Zirconium phosphate; Pillaring; Porosity Insertion of chemical moieties between the layers of swellable layered materials to give a fixed separation between the layers, and ultimately 'tuning' of the pores after thermal processing, is an attractive alternative method for inducing holes into materials' to the hydrothermal chemistry underlying zeolite synthesk2 It may prove of practical importance, given the limitations of (small-pore) zeolites3 for many types of applications (including, but not confined to: catalytic cracking').This strategy has to date been restricted to smectite clays, and especially use of the Keggin ion [A1,304(0H)12(0H2)24]7+(a very naive representation of the species present in the starting precursor solutions6) as a precursor for producing alumina pillar^.^ We have recently found that methods previously successful for preparing chrom- ia- and iron-oxide-pillared montmorillonite* (prior formation of polyhydroxy-Cr"' precursors or use of polynuclear com- plexes) instead give amorphous materials when applied to colloidal suspensions of the base material 'NMe,-SnP' (a-Sn[NMe4]o,9-l. 'H1, 1-o.9(P04)204H,0).~ Instead, when Cr(OAc), (OAc- =CH3C0,) reacts with NMe,-SnP under varying conditions (temperature, differing precursor :base material ratios, Cr3 + concentration) it is possible to obtain intercalate precursors having well defined polyhydroxyaceta- to-Cr3+ species inserted and highly expanded interlayers.On calcination in N,, chromia-pillared materials are obtained having a narrow pore-size distribution (>80% of pores with rp <20 A).These colloid manipulation methods have now been extended to a-zirconium phosphate, a base material exten- sively investigated in the past for ion-exchange purposes,'O membrane technology, and materials design strategies.' The objective is again to find new general methods for obtaining thermally stable, large-pore materials. Experimental Materials a-ZrP was prepared and stored by well established methods.12 n-Propylamine (n-PrA), Cr3 + acetate [Cr(OAc),] and all other chemicals used were best available analytical grade reagents. In initial experiments, NMez -exchanged a-ZrP phases were used. These proved unsuccessful, because considerable amounts of Me4N+ remained intercalated and could not be completely removed during the reaction.The use of n-PrA colloids instead provided pure phases of chromium oligomers on a-ZrP. The reasons for these differences lie in details of surface reactivity, which are being investigated. l3 Preparation of n-PrAH-a-ZrP Suspensions a-ZrP (5 g) was dispersed in water and a solution of 200 cm3 of 0.1 mol dm -n-PrA [60% cation-exchange capacity (c.e.c.) of a-ZrP] was slowly added under vigorous stirring for 2 h. The suspension was centrifuged and the solid resuspended in 1 dm3 of water. Gravimetric determination of the concen- tration of n-PrA-a-ZrP in this suspension gave a value of 4.84 g dm-3. The X-ray diffraction pattern (XRD) of a film prepared from this suspension had a 001 progression (rel.height in parentheses): 17.00A (100); 14.69 A (66); 10.64 A (21). There was no evidence for the presence of hkO reflections and the material appears to be the same as that reported by Alberti et all4 Adsorption Isotherm of Cr(OAc)3 on 60% n-PrAH-ZrP The above suspension (100 cm3) was contacted with increasing quantities of Cr(OAc), dissolved in 100 cm3 of H20: from 8.8 to 351 mequiv. Cr3+ added per gram of ZrP207, i.e. [Cr3+] varying from 0.005 to 0.194 mol dm-3. The effect of Cr3+ concentration was then studied using two-fold diluted solutions. In all cases, each suspension (green in colour) was refluxed for 4 days, cooled, centrifuged, and the dirty-green solids separated, washed well with water (first washing green, second washing colourless), air-dried, and analysed (C, H, N, TG, Cr analysis).Chemical and Physical Measurements XRD, on both powders and cast films (which usually gave narrower 001 reflexions), were recorded on a Siemens D501 diffractometer (graphite monochromator, Cu-Ka radiation). TG and DTA were measured on a Rigaku Thermoflex instru- ment (calcined Al,03 reference, 10 "C min-' heating rate). +Cr3 was analysed colorimetrically using the chromate method (A=372 nm) on alkaline solutions, after treatment with NaOH-H,02. Electronic spectra were registered on a Shimadzu MPC 3 100 spectrophotometer (BaSO, reference) and IR spectra on a Perkin-Elmer 883 spectrometer as KBr disks. Adsorption-desorption of N2 was measured on a conventional volumetric apparatus (77 K, degassing at 200 "C and lo-, mbar overnight).Results and Discussion Inspection of the uptake curve for Cr(OAc), on n-PrAH-ZrP (Fig. 1) immediately shows that uptake does not occur in a simple fashion, but involves complex Cr3 + species distributed between solution and solution/solid interface. Within a narrow range of [Cr3+] :[phosphate] ratios, the material takes up large amounts of Cr3+. At higher dilution (curve B) this range is extended, but in both cases addition of very high amounts of Cr(OAc), leads to materials with much lower Cr3+ loadings. As for the a-tin phosphate analogue, we ascribe this to the setting up of solution equilibria between small and large oligomers, which at equilibrium eventually give rise to higher oligomers no longer capable of inserting into the matrix through those already attached." The analysis of the materials separated at the points of J.MATER. CHEM., 1991, VOL. 1 50 100 150 200 250 300 mequiv. Cr3+ added/g ZrP,O, Fig. 1 Uptake of Cr(OAc), by colloidal cr-nPrAH-Zr(HPO,), *H20; each batch refluxed for 4 days. Conditions as detailed in Table 1 Table 1 Experimental conditions for the uptake of polyhydroxyaceta-to-Cr3+ by nPrA-a-ZrP (as in Fig. 1) sample mequiv. Cr3+ added/g ZrP207 mequiv. Cr3+ taken up/g ZrP207 [Cr3 +Ira 8.78 8.20 0.005 26.34 24.60 0.015 43.89 38.22 0.024 87.78 36.66 0.049 181.27 35.71 0.100 351.13 20.86 0.194 43.89 27.80 0.012 87.78 39.47 0.024 181.27 4 1.43 0.050 351.13 2 1.09 0.097 " Total molar concentration. curve A is shown in Table2.(Note that very small amounts of starting n-PrA were still present in several materials pre- pared from more concentrated solutions; this does not signifi- cantly change the discussion below.) The empirical formulations are also given in Table 2. They are based on the assumption that six-co-ordinate geometry is present for the Table 2 Analyses of the materials of Table 1 material empirical formulation doo2lA c ("/.I N (Yo) H (Yo) 13.2 1.17 0.3 1 2.39 19.1 1.72 0.10 2.28 32.0 39.3 27.3 20.0 3.48 4.25 4.4 1 4.38 0.10 0.15 0.19 0.20 2.45 2.47 2.50 2.5 1 32.9 4.38 0.02 2.37 33.7 36.6 20.6 2.15 3.24 3.82 0.02 0.02 0.02 2.28 2.42 2.63 wt.loss at 200 "C(Toy total H20 (YO) Cr3+:OAc- ratio exothermic effect/ "C" 9.7 10.4 11.9 19.6 10.7 13.4 13.1 28.1 28.8 15.1 14.3 13.4 11.5 19.9 19.5 18.4 17.2 27.9 27.8 18.0 -3.0 2.0 2.5 3.0 1.5 3.0 2.0 2.0 1.5 388 318 280 283 290 335 275 275 273 305 (I From TG/DTA. J. MATER. CHEM., 1991, VOL. 1 Cr3+ throughout and that the oligomers are mixed polyhyd- roxy-acetato Cr3+ moieties (apart from material 1, there was no evidence from ion-exchange experiments that OAc- was ever present as free anion or as an insalation compound). Local probes of structure (optical spectra, EPR, EXAFS, etc.) can give only indirect evidence to support the presence of trimeric, tetrameric, and pentameric Cr3 + species. Neverthe- less, we note that the 4T2g+4A2g(vl) and 4Tl,+4A2,(v2) transitions in chromium hydroxides move to higher energy with polymerisation degree,16 as found here. Secondly, much work on hydroxy-Cr3 oligomers in solution has suggested + the criterion E 1/~21.17- 1.20 (monomer, dimer), 1.6 (trimer), 1.95 (tetramer) and 1.5-1.6 (pentamer and hexamer), i.e.this intensity ratio increases as the connectivity of the chromium cluster increase^.'^ Applying this criterion, the data of Table 3 and Fig. 2 point to: (i) the presence of single species (bands are not broadened), and (ii) formation of extended structures, rather than the compact closed structures believed to exist in solution (but see ref. 18). The IR spectra are relatively well defined for this type of material (see Fig.3) and all show a difference v,,~~~(CO;)-~,~,,(CO;) in the range 91-101 cm-', which is characteristic of bidentate OAc- groups." Furthermore, the v3(P04)region is shifted by ca. 50cm-' with respect to the starting a-ZrP, is unusually distinct, and split, with two bands (at 1137 and 1009 cm-') rather than a single very broad one. This is evidence that the oligomers are attached to the pendant P-OH groups. Unfortunately, there is ambi- guity between a symmetry-lowering of the v3 vibration and the presence of Cr-0-Cr bridges (Cr-0 is expected to have a strong band at 1100-1200 cm-' with h(Cr-O-Cr) vibrations in a wide range, 450-950 cm-1).20 The XRD also favour the presence of well defined oligomers within the layers; all are characteristic of highly crystalline materials, although only a single 001 reflexion is observed, i.e.there is no interstratification or phase segregation (see Fig. 4). This contrasts with the much less defined XRD found in polyhydroxy precursors with smectite clays.I6 These highly expanded precursors re-acquire water after dehydration, as expected for zeolite-like materials. More important is the fact that, when all the zeolitic water is Table 3 Electronic spectra (diffuse reflectance) material v 1/nm v2b other/nm AIIA2 precursors -599 429 1.24 594 420 259 1.02 594 426 263 1.10 59 1 425 -1.10 587 41 1 262 0.96 583 417 259 1.08 597 428 254 1.27 590 425 -1.09 590 426 259 1.10 585 415 256 1.06 calcined at 400 "C N, 61 lsh 454sh 357sh, 323sh, 273 6 17sh 454sh 349, 272 623sh 453sh 351sh, 272 620sh 456sh 366, 320sh, 270 619 450sh 346sh, 273 61 lsh 456sh 332,277 calcined at 400 "C air 61 Ish, 577 454 366, 322sh, 273 598 460 320, 274 617sh, 580 45 Ish 309,272 60 1 453 330sh, 274 741 A '\ B IlIIII1 1IIIIlIII IIIIIIIIIIIIII 400 600 800 Ilnrn Fig.2Optical spectra (as diffuse reflectance) of selected materials. A, Precursors; B, materials calcined at 400 "C under N2; C, materials calcined at 400°C in air (Cr,03 shown for comparison). Numbers refer to materials of Fig. 1 J. MATER. CHEM., 1991, VOL. 1 hi 1'37\ IV 1009 #. .*I,?. 700 1400 1100 , 800 500 2 0 v/cm -' Fig. 3 IR spectra: (a) material 3, as prepared; (b) material 3, 400 "C 24 h N,; (c) Cr(OAc), removed (at 200 "C) they do not collapse to give materials with a restricted range of interlayer distances, but instead provide a range spreading from 9.7 to 35 A (see Fig.5). This behaviour is different from that in the a-tin phosphate ana- logues, which even at this low calcination temperature do collapse considerably' (as do all the PILCS reported to date).16 The electronic and IR spectra do not undergo any marked changes at this stage. The TG/DTA of the materials are very similar to those of the tin phosphate analogues (see Fig. 6) and a similar decomposition scheme operates,' apart from the presence of a plateau at 350-420 "C in TG, which is much more defined than in the tin phosphate analogues.Nevertheless, all water molecules are removed only at much higher temperatures (>600 "C). A further point of interest is that the temperature at which the acetate ion is decomposed decreases more or less linearly as the interlayer distance increases. The difference is large: 113 "C between material 1 and material 3 (Table 2), and it directly reflects the ease with which decomposition products leave the solids. Note also that there is no clear evidence for the presence of a-ZrP,O, phase (as confirmed by XRD) expected at 450-600 "C, constant weight being achieved at 950 "C. As before,' we take this to be indirect evidence that cross-layer condensation between -P-OH groups is not possible because of the pillaring (condensation between adjacent OHs in the layers may also contribute to the TG plateau at 420 "C).A possible mechanism can now be put forward, assuming that as for hexaaquo-Cr3+ ion, Cr3 + acetate forms hydroxo- bridged polymers in solution and polymerisation degree 39.3 Fig. 4 XRD powder patterns of material 4 (thermal treatment under N2) 40 30 * (N*) 5 8 20 t? 10 I I 100 200 300 Tl"C Fig. 5 Changes in interlayer distances on calcination in air. The numbers refer to the materials separated along curve A of Fig. 1. Ir (N,): material 4 calcined under N, increases with Cr3+ concentration. This is shown as (i) in Scheme 1, and addition of more Cr3+ (region 3-4 of Fig. 1) increases oligomer nuclearity and charge, with simultaneous buffering action of AOc-.At low amounts of added Cr3+, competition between oligomers in solution is pushed towards low-nuclearity clusters by electrostatic interaction with the J. MATER. CHEM., 1991, VOL. 1 + high -polymersH20 \ 461 + n -PrAH OAc higher polymer -2 H20 intercalates (iii) Scheme 1 Mechanism (schematic) of the forced polymerisation of Cr(OAc), on colloidal surfaces of a-nPrA-a-ZrP -283 "C0 10 EX0 h s 4 vr 20 DTA u) c -0 v 3c V ENDO 85 4c r.t. 250 500 750 T/"C Fig. 6 TG/DTA of material 4 TC phosphate layer. The higher negative charge on the layer now accommodates the higher positive charge of the oligomer (ii). After attachment of the first, trinuclear, oligomers, olation in the usual manner for hydroxy-bridged Cr3 + complexes then occurs (iii).OAc -must be involved in this oligomerisation stage, and in such a way as to preclude formation of high- connectivity cluster^,'^ as shown. As even more Cr3+ is added, both Cr3+ and negative charge on the phosphate layer increase. An effective way for such a highly charged species to neutralise a high-charge layer is by tilting. This well known 'covering' effect becomes more important as polymerisation proceeds, leaving progressively less space for other species. This nicely rationalises both the uptake curve, the decrease in the total amount of oligomer and the drop in dOo2between materials 4 and 5. The evidence thus points to different orderings of the precursors in a-tin phosphate and a-zirconium phosphate, the former probably containing monolayers and considerable amounts of water on re-forming from the colloidal solution (as suggested previously'), and the latter a bilayer (Scheme 2).Furthermore, we suggest that oligomer chain-endings are different in the two cases, with specific binding possible in the a-zirconium phosphate precursors. Calcination and Formation of Oxidic Pillars Calcination of the precursors with complete removal of organics and formation of chromia pillars supports the above suggestions. Calcining in air at 400 "C in all cases gave pillared products with only low B.E.T. surface areas: ca. 40m2 g-' (after thorough washing). (This is as found for chromia- pillared smectites,8 but different from the a-tin phosphate analogue^.^) Calcination under N2 at 400 "C instead leads to a series of materials (with still recognisable basal spacings in the XRD ranging between 9.5 and 26.9 A) and surface areas in the range 250-330 m2 g-'.The basal spacings of all mater- ials did not change after treatment with 0.2 mol dm-3 HCl or CuCl, (in other words, there are no extractable Cr3+ species remaining between the layers). The most highly pillared are those derived from precursor materials 3 and 4 of the uptake isotherm, i.e. those with the highest Cr3+ content (in these cases, broad dOo2peaks were observed at low angle). The reflectance spectra of the final porous materials show clear differences from that of Cr203 itself, the 4T2, band lying J.MATER. CHEM., 1991, VOL. 1 Table 4 Textural parameters for chromia-pillared a-zirconium phosphates" ' material"/calcination temp SB.E.T.lm2 g- CB.E.T./m2 g- Sb/m2 g-' V,"/cm g - 31400 "C 272 75 31 1 0.224 33 41400 "C 308 87 344 0.199 26 41500 "C 33 1 77 369 0.254 31 5(b)/400"C 321 102 379 0.378 47 'All samples calcined under nitrogen. Parameters as defined and calculated in Cranston and Inkley." m Scheme 2 Illustrating straight (i) and tilted (ii) orderings of oligomers in precursors 1 , , , . , , , , . 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8 PIP0 at higher wavelengths (Table 3) by 20-30 nm. This is in agreement with an oxy-hydroxy formulation for the 'chromia' pillar, rather than simply an oxide one.Conversely, the non- porous materials obtained via calcination in air clearly involve mixed-oxidation species, including Cr2OS, and, possibly, Cr5 + (very-low-energy broad band in the near-IR region (Table 3). The pore radii of the porous series from the plateau of the uptake curve [i.e. materials 3-5, and 4(b) and 5(b)] were calculated from the adsorption branches of the adsorption- desorption curves of Fig. 7 using the cylindrical pore model.21 They are characteristic of orderly pillared materials, and the distributions are narrow (Fig. 8). The best derive from mater- ials 3 and 4, in which ca. 75% of the pores have fp<15 A) that from material 5(b)has only ca. 55% of pores <15 A.We underline the fact that these materials are predominantly mesoporous with a microporous contribution (i.e. the iso- therms are Type IV isotherms in the BDDT classification),22 and that the cylindrical pore method has limitations (in particular, it is unable to provide a reliable estimate of srnall- pore contributions less than the Kelvin limit of 8 It is therefore not obvious just what pore is being accessed by N, (vertical or horizontal to the layers?) nor how this is correlated with the free height deduced from the XRD. Despite this uncertainty, the pore-size distributions represent some of the most homogeneous yet obtained for a pillared material.". 24 This supports the presence of a true pillaring model rather than one in which porosity derives from 'card-house' or 'book- house' models based on face-to-edge associations, both of 20 3.5 15 7 10 0) r I 45 0E z X (c1 h 1.3Gn20--.rd, 15 10 5 \ \ ,ALI-20 40 60 80 20 40 60 80 FpIAFig. 7 Adsorption-desorption isotherms for materials calcined under N, (77 K, N,) on: (a) material 3 400 "C; (b) material 4 400 "C; (c) Fig.8 Pore-size distributions of the 'chromia'-pillared materials of material 4 500 "C; (d) material 5(b)400 "C Fig. 7; same legend J. MATER. CHEM., 1991, VOL. 1 Fig. 9 SEM micrographs of material 5, (a) before and (b) after calcination in N, at 400 “C which are expected to give rise to a scattered pore distri- b~tion).~~Previously, we invoked the ‘book-house’ model to rationalise the wide distribution of pores found in alumina- pillared tin phosphates prepared by ex situ methods.26 As regards gross morphology of materials, SEM micrographs show that after calcination straight and pleated layers are still visible (Fig.9). On calcining at 500 “C under N2, amorphous chromia- pillared zirconium phosphates are obtained, with virtually the same pore size and pore-size distribution as the crystalline form (see the example of material 4 in Fig. 8). This poses the problem of whether XRD alone is a good measure of porosity induced by pillaring. SEM and SAXS investigations have shown that smectite-water, kaolinite-water and sepiolite- water systems give different particle arrangements depending Inon particle fle~ibility.~~ a similar vein, the amorphous nature of the porous ‘chromia’ zirconium phosphates may be rationalised by assuming that it arises from end-to-end sin- tered particles which are so small (<50 A) as not to preclude long-range order yet give only ill defined (amorphous) XRD patterns (Scheme 3).We suggest that these most expanded phases are close to a limit of layer expansion, which renders them more likely to be amorphous. Or, put crudely, the calcination step probably causes gross rearrangements within the pillar, which break up the layers. Although superficially similar, the materials reported here Scheme 3 ‘Break-up’ of layer structure on calcination >400 “C, giving amorphous, but still porous, materials are in reality very different from those obtained in the reflux reactions of Cr(OAc)3 with CX-Z~N~H(PO~)~ 5H20, which are non-porous with stuffed layers, and possess no clear precursor phases.28 Although also believed to involve in situ partially hydrolysed chromium oligomers, the hydrolysis occurs in the restricted space (interlayer distance 11.8 A) of NaH-ZrP 5H20particles.Ordered hydrolysis of the chromium acetate with diffusion of NaOAc out of the matrix is not possible and interlayer fragments are formed. Conclusions The forced hydrolysis of metal complexes on the surfaces of colloidally dispersed layered phosphates can be used success- fully to obtain porous solids, the final products obtained depending critically on the details of surface structure and pH.Further kinetic and structural details are being investi- gated and SAXS experiments planned on these novel, tuned- pore solids. We thank the EEC (European Project 1-0027) and CICYT (Spain, Project No. MAT90-0298) for financial support. References 1 R. M. Barrer, Pure Appl. Chem., 1989, 61, 1903; T. J. Pinnavaia, M. S. Tzou, S. D. Landau and R. H. Raythatha, J. Mol. Catal., 1984, 27, 195. 2 A. Dyer, Introduction to Zeolite Molecular Sieves, Wiley, Chichester, 1988; R. M. Barrer, The Hydrothermal Synthesis of Zeolites, Academic Press, London, 1986. 3 J. M. Newsam, Science, 1986,231, 1093; J. V. Smith, Chem. Rev., 1988, 88, 149; M. Szostak, Molecular Sieves, Academic Press, London, 1988. 4 G.A. Ozin, A. Kugerman and A. Stein, Angew. Chem. Int. Ed. Engl., 1989, 28, 359; P. A. Diddams, J. M. Thomas, W. Jones, J. A. Ballantine and J. H. Purnell, J. Chem. SOC.,Chem. Commun., 1984, 1340. 5 D. E. W. Vaughan, P. K. Maher and E. W. Albers, U.S.Pat., 3 775 345, 1973; M. Occelli, Ind. Eng. Chem. Prod. Res. Dev., 1983, 22, 553. 6 J. W. Akitt, J. M. Elders, X.L. R. Fontaine and A. K. Kindu, J. Chem. Soc., Dalton Trans., 1989, 1897, and refs. therein: A. Clearfield, in Surface Organometallic Chemistry: Molecular Approaches to Surface Catalysis, ed. J. M. Bassett, Kluwer, Amsterdam, 1988, p. 271. 7 D. E. W. Vaughan and R. J. Lussier, Proc. 5th Int. Con$ Zeolites, ed. L. V. C. Rees, Heyden, London, 1980, p. 94; G. W. Brindley and R. E. Sempels, Clay Minerals, 1977, 12, 229.8 M. S. Tzou and T. J. Pinnavaia, in Catalysis Today, ed. R. Burch, Elsevier, London, 1988, p. 243; D. H. Doff, N. H. J. Gangas, J. E. M. Alan and J. M. D. Coey, Clay Minerals, 1988, 23, 367. 9 P. Maireles-Torres, P. Olivera-Pastor, E. Rodriguez-Castellon, A. Jimenez-Lopez, A. A. G. Tomlinson, in Pillared Layered Struc- tures. Current Trends and Applications, ed. I. V. Mitchell, Elsevier Applied Science, London, 1990, p. 137; J. Solid State Chem., 1991, in the press. 10 Inorganic Exchange Materials, ed. A. Clearfield, CRC Press, Boca Raton, 1982, ch. 1; G. Alberti, in Recent Developments in Ion Exchange, ed. P. A. Williams and M. J. Hudson, Elsevier Applied Science, London, 1987, p. 233. 11 A. Clearfield, in Design of New Materials, ed.D. L. Cocke and A. Clearfield, Plenum Press, New York, 1987, p. 121; G. Alberti, U. Costantino, F. Marmottini, R. Vivani and P. Zappelli, in Pil-lared Layered Structures. Current Trends and Applications, ed. I. V. Mitchell, Elsevier Applied Science, London, 1990, p. 119. 12 G. Alberti and E. Torracca, J. Znorg. Nucl. Chem., 1968, 30,317. 13 P. Maireles-Torres, P. Olivera-Pastor, E. Rodriguez-Castellon, A. Jimenez-Lopez and A. A. G. Tomlinson, in preparation. 14 G. Alberti, A. Casciola and U. Costantino, J. Colloid Interface Sci., 1985, 107, 256. 15 C.H. Giles, T.H. McEwan, S.N. Nakhwa and D.Smith, J. Chem. SOC., 1960, 3973. 16 T. J. Pinnavaia, M. S. Tzou and S.O. Landau, J. Am. Chem. Soc., 1985, 107, 4783.746 J. MATER. CHEM., 1991, VOL. 1 17 H. Stunzi, F. P. Rotzinger and W. Marty, Inorg. Chem., 1984, 23, 2160. 25 T. J. Pinnavaia, in Chemical Physics of Intercalation, NATO, AS1 Ser. B, vol. 172, p. 233; H. van Damme, P. Levitz, J. J. Fripiat, 18 L. Monsted, 0. Monsted and J. Springborg, Inorg. Chem., 1985, J. F. Alcover, L. Gatineau, and F. Bergaya, in Physics of Finely 24, 3496. 19 K. Nakamoto, Infrared and Raman Spectra, Wiley, New York, 1986, p. 233. 26 Divided Matter, ed. N. Boccara and M. Daoud, Springer, 1985, p. 24. P. Maireles-Torres, P. Olivera-Pastor, E. Rodriguez-Castellon, 20 J. R. Ferraro, R. Driver, W. R. Walker and W. Wozniak, Inorg. A. Jimenez-Lopez, L. Alagna and A. A. G. Tomlinson, J. Chem. Chem., 1967,6, 1586. SOC., Chem. Commun., 1989, 741; J. Muter. Chem., 1991, 319. 21 R. W. Cranston and F. A. Inkley, Adv. Catal., 1957, 9, 143. 27 D. Tessier, G. Pedro and L. Camara, C.R. Acad. Sci. Paris, 1980, 22 S. J. Gregg and K. S. W. Sing, Adsorption, Surface Area and 2WD, 1169; C. H. Pons, F. Rousseaux and A. Tchoubar, Clay Porosity, Academic Press, Orlando, 2nd edn., 1982, p. 4. Minerals, 1981, 16, 23. 23 F. Rodriguez-Reinoso and A. Linares-Solano, in The Chemistry 28 D. J. MacLachlan and D. M. Bibby, J. Chem. SOC., Dalton Trans., and Physics of Carbon, Marcel Dekker, New York, Vol. 21; 1989, 895. K. S. W. Sing, Pure Appl. Chem., 1982, 54, 201. 24 J. Sterte, Clays Clay Minerals, 1986, 34,658. Paper 1 /00500F; Received 4th February, 1991

ISSN:0959-9428    

DOI:10.1039/JM9910100739

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

J. MATER. CHEM., 1991, 1(5), 747-749 New Perovskite Phases in the Systems Li,O-(Nb,O,,Ta,O,)-ZrO, Maria Elena Villafuerte-Castrejon,' Carlos Kuhliger," Rosalba Ovando,' Ronald 1. Smithb and Anthony R. Westb a Instituto de Investigaciones en Materiales, UNAM, Apdo Postal 70-360, Mexico, D.E 04510, Mexico University of Aberdeen, Department of Chemistry, Meston Walk, Aberdeen AB9 2UE, Scotland -&,Two new orthorhombic perovskite phases of formula, Lio~704~xZro,074+4xTa,where 0.08<x< 0.15 and Li,.,,, -xZro.057 where 0 <x< 0.15, have been synthesized. Their stoichiometries were determined as +4xNb,-3x03, part of a phase-diagram study of the systems Li20-Nb205-Zr0, and Li,O-Ta,O,-ZrO,. In addition to these new phases, extensive solid solutions based on Zr-doped LiNbO,, LiNb308, LiTaO, and LiTa308 have been prepared. Keywords: Li,O-( Nb205, Ta,O,)-ZrO, sysfem; Perovskife; Sfoichiomefry LiTaO, and LiNb03 are important electro-optic materials, the properties of which can be modified by doping them with various cations.For waveguide applications, the surface properties of LiNb03 crystals can be modified by in-diffusion of Ti, so as to cause a variation of refractive index with depth of penetration.' Similar results may be obtained by partial exchange of H+ for Li+ at the crystal surfaces.2 Various studies have been reported in which the Curie temperature of LiTaO, is modified by cation substitution^.^-^ In the present work, a study of the effects of Zr substitution has been made, by means of a phase-diagram determination of the systems Li20-Nb205-Zr02 and Li20-Ta205-Zr02. Solid solutions of Zr02 in LiTaO, and LiNbO, have been prepared and of particular interest, two new orthorhombic perovskite phases have been discovered.Experimental Reagents used were Li2C03, Nb205, Ta205 and ZrO,, of reagent grade or better. They were used direct from the bottle as drying was found to be unnecessary. Mixtures of 3-4g were ground into a paste using acetone with an agate mortar and pestle, dried and reacted in Pt crucibles, initially at 700 "C for a few hours to expel CO,, followed by temperatures in the range 900-1350 "C, depending on composition. General- purpose muffle furnaces, controlled and measured to & 30 "C were suitable since the phase equilibria and phase stabilities were not markedly temperature dependent.Samples were heated in air and cooled in 5-10 min by removing them from the furnace and allowing them to cool naturally. Reaction times varied between 1 and 12 days, depending on the time required to reach equilibrium. For the longer reaction times, samples were removed occasionally from the furnace, crushed to a fine powder and then returned to the furnace. The conditions were chosen so as to avoid significant evaporation of lithium oxide, especially for compositions rich in Li20. Generally, reactions in the Ta205 system required tempera- tures 50-100 "C higher than those in the Nb205 system. Products were analysed by X-ray powder diffraction using a Siemens D500 diffractometer, Cu-Ka radiation.For accurate d-spacings, KCl was added as an internal standard. Results and Discussion Heating experiments were carried out on 91 mixtures in the system Li,0-Ta205-Zr02 and on 87 compositions in the system Li,0-Nb205-Zr02. The results are summarised in Supplementary Data Tables A and Bt and were used to construct the phase diagrams shown in Fig. 1 and 2, respect- ively. Both diagrams contain three features of particular interest. (i) Each system contains a new, non-stoichiometric, ortho- rhombic perovskite of composition fairly close to LiNb0, and LiTa03. In both cases, the perovskite appears to have a limited stoichiometry range. This is shown on an expanded scale in Fig. 3 and 4, respectively.In the Nb system, one compositional limit of the perovskite solid solutions lies on the line of hypothetical stoichiometry: Lil -4yZryNb03,corre-sponding to y=0.057, point A, Fig. 4. The solid solutions extend from point A in the direction given by the substitution mechanism, Li +3Nb 4Zr, giving a general formula for the solid solutions of Li -4y -xZry+4xNb1-3x03, for which y = 0.057 and 0 <x <0.15. In the Ta system, Fig. 3, a similar solid solution forms, but with y=0.074 and 0.08<x<0.15. In this case, the solid solutions do not appear to extend as far as the 'ideal' composition, with x =0, point A. Indexed X-ray powder data, based on primitive orthorhom- bic unit cells, are given in Tables 1 and 2. The data are very similar to those of CaZrO,, an orthorhombic perovskite, space group Pnrna.From the experimentally determined com- positions of the perovskites, it appears that, in the Nb system f' Supplementary data available (SUP56841, 13 pages); details from Editorial Office Li20 Fig. 1 Phase diagram of the system Li20-Zr02-Ta20, at subsolidus temperatures, 950-1250 "C. Compositions studied are shown as closed circles. The area of single-phase LiTaO, solid solutions is shaded J. MATER. CHEM., 1991, VOL. 1 A Fig. 2 Phase diagram of the system Li,0-ZrOz-Nb,05 at subsolidus temperatures, 950-1 150 "C. Compositions studied are shown as closed circles. The area of single-phase LiNbO, solid solutions is shaded A %Ta,O, \ J= 0.074 \ \-Li1dyZry Ta03 \\ 10 20 %ZrO, Fig.3 Expanded region of the phase diagram Li,0-Zr0,-Ta,05 showing the occurrence of the LiTaO, solid solutions and the perovskite phase at least, the ideal structure is based on a complete three- dimensional array of corner-sharing Nb06 octahedra. Substi- tution of Zr onto the Nb sites (i.e.B sites) occurs in the solid solutions, with increasing x. However, since the Zr content is greater than is the deficiency in Nb content by an amount (y+ x), some Zr must presumably occupy the larger, A cavities, which also contain Li ions. Preliminary powder neutron + diffraction results have confirmed the essentially perovskite- like structure of these new phases. Further structural studies are required to explain why these perovskites exhibit an overall cation deficiency, relative to the ideal AB03 formula and why the phase limits are different in the Nb and Ta systems.(ii) LiTa03 and LiNb03 both form limited ranges of solid solutions containing <lo% ZrO,. These occur over an area of compositions in the phase diagrams and therefore do not form by a single substitution mechanism. In similar solid solutions containing Ti4 instead of Zr4 +,the solid solutions + were found to be most extensive in the direction of the hypothetical composition Li2Ti409.7 Consequently, the main solid solution mechanism was deduced to be Li+3(Nb, 'y A%, ,Li+3Nb+4Zr, \ I/ v V 10 20 %ZrO, Fig. 4 Expanded region of the phase diagram Li,O-Zr0,-Nb,O, showing the occurrence of the LiNbO, solid solutions and the perovskite phase Ta)e4Ti, which corresponded to a mechanism of constant overall cation content.In the present materials, the solid solutions are less extensive, presumably because the Zr4+ ion is somewhat larger than Li', Nb5+ and Ta5+. It is less clear, therefore, whether the mechanism of constant overall cation content, which would extend the solid solutions in the direc- tion of hypothetical Li2Zr409 (Fig. 1 and 2) is particularly significant or dominant. (iii) LiNb308 and LiTa308 both form extensive solid solu- tion series in the direction of ZrO,. In these, the solid solution mechanism is clearly the one that retains a constant overall cation content, i.e. Li +3(Nb,Ta) 4Zr, giving rise to the general formula, Lil -x(Nb,Ta)3 -3xZr4x08, where 0 <x <0.1 1 for Ta and 0 <x <0.14 for Nb. The remaining areas of the two phase diagrams are divided into a number of two- and three-phase regions, Fig.1, 2. In constructing these diagrams, it was assumed that the literature data on the various binary phases and phase diagrams was essentially correct, viz. Li20-Nb205,8~9 Li20-Ta205,8,10-12 Li20-Zr02,13 Ta205-Zr0214-16 and Nb205-Zr02.17-19 Table 1 X-Ray powder diffraction data for Li0.704 -x~ro.o74+4xTa1- 3AP 0203.739 lOl} 100 2.720 2.729 200 17 2.678 2.679 121 41 2.593 2.593 002 15 2.139 2.145 7 1.880 { ;:2 202 19 1.760 1.752 23 1 3 1.716 301 15{ K1: 1321 1.683 { ::% 1.649 1.648 1.562 240{ !:Z 1.528 1.537 042 1.511 1.513 123 1.363 1.364 400 "~=5.457(6)A; b=7.637(7) A; ~=5.185(2)A J.MATER. CHEM., 1991, VOL. 1 749 Table 2 X-Ray powder diffraction data for 2 C. E. Rice and J. L. Jackel, J. Solid State Chem., 1982, 41, 308. 3 J. L. Fourquet, M. F. Renou, R. de Pap, H. Theveneau, P. P.Li0.774 -xzr0.057 +4xNbl -3x03 Man, 0. Lucas and J. Pannetier, Solid State Ionics, 1983, 9-10, hkl 1 1011. 4 R. R. Neurgaonkar, T. C. Limy E. J. Staples and L. E. Cross, 020 Ferroelectrics, 1980, 27, 63.3.755 lOl} 100 5 G. T. Joo, J. Ravez and P. Hagenmuller, Rev. Chim. Miner., 3.368 3.365 111 3 1985, 22, 18. 2.743 2.743 200 19 6 B. Elouadi, M. Zriouil, J. Ravez and P. Hagenmuller, Ferroelec-2.660 2.66 1 121 30 trics, 1984, 56, 21.002 7 M. E. Villafuerte-Castrejon, A. Aragon-Piiia, R. Valenzuela and 2.58 1 { ;:E3 210) 25 A. R. West, J. Solid State Chem., 1987, 71, 103. 2.131 2.131 022 14 8 R. S. Roth, H. S. Parker, W. S. Brower, and J. L. Waring in Fast Ion Transport in Solids, Solid State Batteries and Devices ed. W. 1.881 { i:::; 24 van Gool, North Holland, Amsterdam, 1973 p. 217. 1.724 1.724 30 1 3 9 L. 0.Svaasand, M. Eriksrud, A. P. Grande and F. Mo, J. Cryst. 1.703 1.711 132 5 Growth, 1973, 18, 179. 10 R. L. Barnes and J. R. Carruthers, J. Appl. Crystallogr., 1970,3,1.683 f:::; 14 395. 1.644 { 1.644 103 12 1 1 A. Reisman, J. Phys. Chem., 1962,66, 15. 12 C. D. Whiston and A. J. Smith, Acta Crystallogr., 1965, 19, 169.;::I 042 1.562 10 13 J. L.Enriquez, P. Quintana and A. R. West, Trans. Br. Ceram. 1.522 1.522 3 SOC.,1982, 81, 17. 1.506 1 SO6 123 5 14 B. W. King, 3. Schultz, E. A. Durbin and W. H. Duckworth in 1.371 I .372 400 2 Phase Diagrams for Ceramists, American Ceramic Society, Col- 1.330 1.331 242 3 umbus, 1964, No. 374. ~ ~~ 15 R. S. Roth and J. L. Waring, in Phase Diagramsfor Ceramists, a=5.486(3) A; b=7.531(8) A; ~=5.168(2)A American Ceramic Society, Columbus, 1975, No. 4458. 16 C. Zheng and A. R. West, J. Mater. Chem., 1991, 1, 163. We thank the British Council for supporting the Aberdeen- 17 R. S. Roth and L. W. Coughanour, J. Res. Nut. Bur. Std., 1955, 55, 209.Mexico collaboration programme, CONACyT, Mexico for 18 R. S. Roth, J. L. Waring, W. S. Brower and H. S. Parker, NBSfinancial support, L. M. Banos for assistance with the X-ray Spec. Publ., 1972, 364,183. diffraction and SERC for the provision of neutron scattering 19 C. Zheng and A. R West, Br. Ceram. Trans. J., 1990, 89, 138. facilities at ISIS, Rutherford Appleton Laboratory. Paper 1/00563D; Received 6th February, I99 IReferences 1 R. V. Schmidt and I. P. Kaminov, Appl. Phys. Lett., 1974, 25, 458.

ISSN:0959-9428    

DOI:10.1039/JM9910100747

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

J. MATER. CHEM., 1991, 1(5), 751-756 751 Flash Calcination of Kaolinite: Mechanistic Information from Thermogravimetry Robert C. T. Slade,*" Thomas W. Daviesb and Husnu AtakulaIb Departments of "Chemistry and bChemical Engineering, University of Exeter, Exeter EX4 4QQ UK Kinetically frozen samples of flash-calcined kaolinite (rapidly heated to high temperature, maintained at that temperature for a variable residence time and then rapidly cooled) have been produced in a laboratory calciner. The degrees of dehydroxylation (fl) of calcines relative to the parent kaolinite have been determined using thermogravimetry (TG) and found to increase with calciner temperature, heating rate and residence time in the calciner. Variations in the forms of the TG curves for the subsequent dehydroxylation of calcines as a function of provide information concerning the sequential loss of hydroxyls from kaolinite during flash calcination.This is postulated to occur in two waves: interlamellar hydroxyls (75%) are preferentially removed, with a secondary loss of intralamellar hydroxyls (25%) becoming dominant as loss of interlamellar hydroxyls approaches com p let io n. Keywords: Kaolinite; Flash calcine; Thermogravimetry; Dehydroxylation Kaolinite [china clay, A12Si205(0H), sometimes written as Al2O3-2SiO2.2H20] is a raw material of considerable indus- trial significance on an international scale.' The ideal kaolinite structure consists of stacks of lamellae, each composed of a pair of silica and alumina sheets.The silica sheets contain vertex-shared Si04 tetrahedra, while the alumina sheets con- tain edge-shared A106 octahedra (some 0s being those of hydroxyl groups). Kaolinite dehydroxylation to create a thermally stable product is an important step in the manufac- ture of clay products. Three of the four hydroxyl groups associated with the alumina sheets lie adjacent to the interla- mellar space between successive sheet pairs, while the fourth is intralamellar (between the silica and alumina sheets). Dehy- droxylation is a rate process and calcines can be kinetically frozen at various stages of structural reorganisation. In the absence of structural collapse, an idealised kaolinite would lose 13.95% of its mass on complete dehydroxylation and its density would decrease from 2.64 to 2.27 g cm-3.2 Two industrial calcination techniques, which lead to entirely different materials with differing physical characteristics, can be contrasted.Soak calcination is total dehydroxylation under slow heating to a fixed soak temperature (held for >30 min). This has been the subject of numerous investigations (e.g. ref. 3-6), but is still incompletely understood. The form of the dehydroxylation curve (mass uersus temperature) for kaolinite under such conditions has been well known for some time (e.g.ref. 7), but the exact behaviour depends on variables such as particle size, crystallinity, sample volume and chemical composition of the materia1.2.6.8-" It will be evident from the years of publication of the references cited that studies of soak calcination are by no means a new phenomenon. More recent literature, providing additional insight into the process and its structural consequences includes ref.12-16 and refer- ences therein. Flash calcination involves passing cold powdered clay through a gas or oil flame and then quenching by injection of cold air. The laboratory simulation of this very rapid process involves plunging a stream of clay particles into a co- flowing stream of inert gas (which is in downward laminar flow) in a vertical electrically heated reaction tube. Flash calcines are partially dehydroxylated (as opposed to fully dehydroxylated soak calcines) and have different properties from those of soak calcine^,'^*'^ e.g.lower densities than kaolinite (soak calcines are denser than kaolinite) and internal voidsI8 (absent from soak calcines) with diameters similar to the wavelengths for visible light (causing light scattering and so imparting opacity to the material). During flash calcination kaolinite particles are heated at such a speed that the steam generated within them is generated faster than it can escape by diffusion, and structural disruption is a consequence. Studies aimed at the in-depth characterisation of flash calcines and the effects of process variables have been reported only very recently, this being a consequence of the necessary sophistication of the laboratory calciner. A systematic investigation of the flash calcination process is in hand at Exeter using a purpose-built furnace (described elsewhere") allowing kaolinite particles to be subjected to thermal histories comparable to those in industrial flash calciners.We have recently reported examination of physical2' and spectroscopic20*2' properties of flash calcines of kaolinite produced using different tailored combinations of process variables, such as calciner temperature, heating rate from cold to the calciner temperature, residence time in the calciner and calciner atmosphere [flowing dry He(g) or N2(g)]. X-Ray powder diffraction pat terns of calcines2' typically have an 'amorphous' background arising from transformed material/ regions, with an additional low-intensity contribution from untransformed kaolinite-like regions (decreasing progressively with increasingly severe calcination conditions).There was no evidence in the X-ray pattern of any flash calcine for the formation of any crystalline mullite or cristobalite (generally considered undesirable owing to the increased abrasiveness imparted to the calcine). This contrasts with the case of soak calcines produced at ca. 1000 "C(see e.g. ref. 22). Time reso- lution of some of the structural changes occurring during flash calcination has been achieved by following MAS NMR spectra of flash calcines as a function of residence time in the calciner." 29Si Spectra can be deconvoluted into a minimum of four Gaussian components with shifts ranging from mullite- like (-90ppm) to Q4 (-110ppm) Si environments.27Al spectra show peaks for four-co-ordinate and six-co-ordinate Al, but no peak assignable to five-co-ordinate A1 (seen for soak calcines) is observed. The derived composite picture of flash calcination is progressive transformation of kaolinite to a single product which undergoes little further chemical reaction during its short time in the calciner. We now report examination of the further dehydroxylation of tailored flash calcines using thermogravimetry (TG) tech-niques, and the insight these studies give on the residual hydroxyls ('water') in the flash calcines themselves. 752 Experimental The feedstock was commercial grade SPS clay (English China Clays International) from the Lee Moor (Devon) deposit.X-Ray fluorescence (XRF) analysis of the material gave Si02 46.2, A1203 38.7, Fe203 0.56, Ti02 0.09, CaO 0.20, MgO 0.20, K20 1.01, Na20 0.07%. The potassium content arises from the presence of some mica (hydromuscovite); the Lee Moor deposit gives materials which are particularly suited to paper coating and are 293% kaolinite. The X-ray powder diffraction pattern and infrared spectrum of the material have been given elsewhere, along with those for the calcines2' Scanning electron microscopy (SEM) revealed the particles to be flat platelets. The particle size distribution obtained by scattering of laser light (Malvern MasterSizer) is shown in Fig. 1. The calculated specific surface area was 2.0 m2 g- '. The particle size distribution was also examined using sedi- mentation techniques (Micromeritics Sedigraph), giving 78% by mass with particle size <2 pm.The latter technique tends to exaggerate the incidence of smaller particles, because of the 'parachute' effect of clay platelets and the increased significance of Brownian motion. Flash calcines were produced employing a range of process variables in a laminar flow furnace described e1~ewhere.l~ A stream of clay particles in flowing He(g) or N2(g) was heated from room temperature to a set calciner temperature T by a computer-controlled heating system. Calcined material col- lected in a water-cooled collector probe was quenched by massive dilution with room temperature N2(g). Residence time z for the particles in the isothermal reaction zone (which exists between the powder inlet and collection points) at a given T was set by controlling gas flow rate and position of the collector probe.Heating rate A was calculated from gas flow rate and temperature difference that solid particles experienced in the heating zone between the furnace entrance and the isothermal reaction zone.16 The design and conditions of the furnace are such that laminar flow (and hence uniform resi- dence time) is always established in the furnace. Analysis of the heat transfer and flow processes within the reaction tube shows that the particles effectively achieve the set calciner temperature T in a few milliseconds and follow the transport gas temperature and velocity." Thermogravimetry was carried out in a Stanton TG-750 instrument.Samples (ca. 10 mg) were heated from ambient temperature to 960 "C at 20 "C min-' in flowing dry N2(g) and soaked at 960 "C to constant mass. Variation of heating rate made little difference to the dehydroxylation curves. X-Ray powder diffraction studies of the final products found 100 I20f-Il0 c i //" n 0 6.1 1 10 100 particle size/pm Fig. 1 Particle size distribution for kaolinite feedstock as determined by scattering of laser light (Malvern MasterSizer). Both the frequency distribution (bar chart) and the cumulative frequency curve (continu- ous line) are shown J. MATER. CHEM., 1991, VOL. 1 metakaolinites with traces of mullite, tending towards the products of soak calcination.The term degree of dehydroxylation is used in two senses (at a point on a TG run and as a characteristic of an as- prepared calcine) with respective definitions as follows. (i) At the point on the dehydroxylation (TG) curve at which rn g of the initial M g of 'water' present in the sample under analysis (either kaolinite or a calcine) has been lost, the degree of dehydroxylation a is defined by a=(m/M)x100% (1) When degree of dehydroxylation during a TG run is relative to a parent calcine (already partially dehydroxylated; p >0, see below), we use the modified symbol a'. a and a' therefore vary along the dehydroxylation curves of kaolinites and calcines, respectively. (ii) For a calcine prepared from kaolinite containing MI g of 'water' which loses mlg during calcination (as subsequently determined by difference after a TG run), the degree of dehydroxylation p of the calcine is defined by p=(rn'/M') xlOO% (2) /3 is therefore a constant characteristic of a given material.The dehydroxylation curve for the parent material (p=0% by definition) and storage conditions used in this study (oven drying at 105 "C) is shown in Fig. 2: a= 100% corresponds to a mass loss of 12.2%, arising from loss of the 'water' originally present. The general form of the dehydroxylation curve is in accord with the literature and has been well documented (e.g. as long ago as 19577) and discussed else- where (e.g. ref. 2,6,8-11). The initial low-temperature mass decrease corresponds to loss of non-constitutional (sorbed) water.Dehydroxylation of kaolinites takes place at T>400 0C,2 the onset temperature being affected by particle size, crystallinity, atmosphere and mass of sample.2*6*8-'1 Kaolinites with the lowest particle sizes and poorest crystal- linities have the lowest onset temperature. As discussed above, the material in this study contains a little mica, which itself will undergo dehydroxylation in the calciner and in TG runs. The dehydroxylation curve is typical of a 'kaolinite' (very nearly all of the material is kaolinite phase) and further discussion neglects the mica content, this being justified by the uncertainties in the different hydroxyl populations deduced (see below). It should be noted, however, that the properties of calcines produced from different feedstocks may well have significant slight differences.60 h s Y 8 40 0 200 ' 400 I 600 ' I 800 ' I L 1000 1200 T/"C Fig. 2 The dehydroxylation curve for the kaolinite used in this work. a is the degree of dehydroxylation (relative to kaolinite) during the TG run J. MATER. CHEM., 1991, VOL. 1 100 ' I ' I ' I ' I ' I ' I ' I ' (b) 0 00loo! 80 0 00 0 0 60 1 0 00"1 ...@ Fig. 3 Degrees of dehydroxylation B (relative to kaolinite) of flash calcines produced at a calciner temperature T= 1000"C as a function of residence time (z) and heating rate (A): (a) in He(g), (b) in N2(g). Values of A/K s-':O, 4700; 0,8000 100 ' I ' I ' I ' I ' I ' I ' ((a1 A -1 0 0 0 0 0 00 0 11 t OQ' '"I 0A""'l'"l'l'' 600 700 800 900 1000 1100 1200 1300 600 700 800 900 1000 1100 1200 1300 T/"C TI" C Fig.4 Degrees of dehydroxylation fi (relative to kaolinite) of flash calcines produced at fixed residence time z =0.5 s as a function of calciner temperature (T) and heating rate (A): (a)in He(g), (b) in N,(g). Values of A/K s-': 0,4700; 0,8000; 0,10 000; W, 12 000; A,15 000 Results The laboratory flash calciner used in this work enables exploration of the influence of changes in process variables (T, A, z, atmosphere; see above) both on the degrees of dehydroxylation (8) of product calcines and on the form of the dehydroxylation curves.Degrees of dehydroxylation of product calcines, /I Fig. 3 shows the variations in 8 for individual (single-batch) flash calcines produced at a fixed calciner temperature T= 1000 "C as a function of heating rate (A =4700, 8000 K s-I) and residence time (z =0-1.5 s) in He(g) and N2(g). /Ivalues shown in Fig. 3 and 4 are averages from repeated dehydroxyl- ation curves, with standard deviations uniformly <1%. Scatter (deviations from smooth variations of 8 with z) in the data shown in Fig. 3 (and in variations of p with T in Fig. 4) thus reflects the level of controllability of the process variables provided by the calciner. As might be anticipated, /Iis higher for calcines corresponding to longer residence times in the calciner, as dehydroxylation then proceeds further towards completion.Higher heating rates also lead to more complete dehydroxylation (higher fi calcines), this being clearer in the data referring to an N2(g) atmosphere. This is likely to be related to greater structural disruption (see above) arising with a higher heating rate; the results of density and X-ray studies20 are compatible with such an interpretation. In flash dehydroxylation, layers near particle surfaces are dehy-droxylated much faster than others, generating a crustal barrier to diffusion of H20 from the bulk.I6 This situation is likely to be exacerbated at the higher heating rate, leading to greater disruption. Fig.4 shows variations in fl for calcines produced with a fixed residence time z =0.5 s as a function of heating rate (A = 4700-15 000 K s -I) and calciner temperature (T=700-1200 "C) in He(g) and N2(g).Other than for calcines produced in He(g) at the lowest heating rates, dehydroxylation levels 8270% are achieved at T2900 "C. This could correspond to removal of all interlamellar hydroxyls (see introduction), with intralamellar hydroxyls being more difficult to remove. A slowing in loss of water noted in studies of soak calcination has been similarly a~signed.~~.~~ Fig. 3 and 4 combine to show that the most highly dehy- droxylated materials are obtained at T>900 "C with high heating rates and longer residence times. Such materials have the lowest densities,20 this arising from the highest degree of structural disruption.Such disruption provides more facile routes for the exit of 'water' from the structure. 754 Dehydroxylation Curves of Flash Calcines These curves provide information on the nature of residual 'water' in flash calcines. If the calcines were simply admixtures of kaolinite and anhydrous product, the form of these curves would closely resemble that of kaolinite itself (Fig. 2), which has been fully discussed in the literature (see above). However, this is not the case (see below). Fig. 5 shows the dehydroxylation curves obtained for cal- cines produced in N2(g) at a fixed calciner temperature T= 1000 "C and with heating rates A=4700 and 8000 K s-' as a function of residence time z. Characteristics of the parent calcines are given in Fig.3. Only for the least dehydroxylated calcines (low heating rate, short residence time) is the curve similar to that for kaolinite, deviations occurring for more dehydroxylated parent calcines. The increased significance of the low temperature mass decrease (loss of water sorbed by the calcine) is a simple consequence of the decreased hydroxyl content (relative to kaolinite) of the calcine. NMR studies reveal structural information consistent with a decreasing content of intraparticle kaolinite-like regions as z increases;21 the present TG results corroborate that picture. Deviations from the curve for kaolinite become marked when pr60% for the parent calcine (combining the information in Fig. 3 and 5). fiz6O% corresponds to calcines in which not all the interlamellar hydroxyls (75YOof hydroxyls originally present, see introduction) have been removed from all particles; dehy- droxylation of particles to different extents is a likely conse- quence of the particle size distribution discussed above.The deviations observed indicate that in the most highly dehy- droxylated particles/calcines a sizeable proportion of the remaining hydroxyls (originally intralamellar) are more easily removed than any present in kaolinite or during its slow dehydroxylation (during TG or soak calcination) and that this proportion increases at longer residence times. These are the calcines with the lowest densities;20 this arises from greater structural disruption, which enables facile removal of some of the 'water' present.Fig. 6 shows the dehydroxylation curves obtained for cal- cines produced in N2(g) with fixed calciner residence time z = 0.5 s and with heating rates A=4700 and 12 000 K s-' as a function of calciner temperature T. Characteristics of the parent calcines are given in Fig. 4. Combination of the data in Fig. 4 and 6 reveals a picture very similar to that described in the previous paragraph uiz. at fi >60% a proportion of the residual hydroxyls are lost relatively easily. This proportion increases with increasing T. J. MATER. CHEM., 1991, VOL. 1 Dehydroxylation curves for calcines produced in He(g) were very similar in form to those discussed above [calcines pro- duced in N2(g)], and can be interpreted similarly.Fig. 7 exemplifies this point and shows the dehydroxylation curves obtained for calcines produced in He(g) with fixed calciner residence time z=0.5 s and with heating rate A = 12 000 K s-' as a function of production temperature T. Characteristics of the parent calcines are given in Fig. 4. Discussion During flash calcination the kaolinite particles are raised from ambient to high temperature very rapidly (heating rate A zlo5 K s -'). The dehydroxylation almost all occurs at high temperature, this being in marked contrast to dehydroxylation during slow heating (soak calcination and TG). It follows that the course of dehydroxylation during flash calcination is not necessarily the same as that characteristic of slower heating; one clear difference is the production of voids in the calcine particles during flash calcination only.Both initial heating rate and calciner temperature have profound effects on the product, increases in either variable leading to increasingly dehydroxylated (higher p) calcines (Fig. 3 and 4). The lowest density calcines are also those which are the most dehy-droxylated.20 An attempt to explain these gross effects of flash calcination in terms of the heat- and mass-transfer phenomena which occur during the transient heating of a reactive particle has been made;25 the resulting mathematical model was used to demonstrate the sensitivity of the dehydroxylation process to surface heat transfer events. In this study, control over process variables was used to produce calcines kinetically frozen at different degrees of dehydroxylation (p).Subsequent (slow) TG dehydroxylation of these materials provides information concerning the nature of the residual hydroxyls present in the calcines as a function of p. The results permit the following model of the sequence of events occurring within the particles during flash calcination to be postulated. (i) The first wave of dehydroxy- lation may remove progressively the 75% of hydroxyls in the parent kaolinite that are interlamellar. For calcines kinetically frozen during this stage the subsequently determined (slow) dehydroxylation curve is similar in form to that of kaolinite. NMR studies of these calcines have shown the existence of kaolinite-like regions which decrease in extent as /? is increased.21 (ii) A second wave of dehydroxylation (involving removal of the hydroxyls originally intralamellar in kaolinite) 0 200 400 600 800 1000 1200 V"C V°C Fig.5 Dehydroxylation curves for flash calcines produced at fixed calciner temperature T=1000"C in @N,(g)as a function of residence time (T) and calciner heating rate (A): (a)A =4700, (b)A =8000 K s-'. a' is the degree of dehydroxylation (relative to the appropriate calcine) during the TG run. z/s: X, 0.1; m, 0.2; 0,0.3; 0,0.4; @, 0.5; +, 0.6; 0, 1.0; A,1.50.7; +, 0.8; A,0.9; ., J. MATER. CHEM., 1991, VOL. 1 100 ' I ' II(a) v 8 b40 40 20 20 n 0 v-0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200 T/"C T/"C Fig.6 Dehydroxylation curves for flash calcines produced with a fixed residence time 7=0.5 s in N,(g) as a function of calciner temperature (T) and calciner heating rate (A): (a) A=4700, (b) A= 12 000 K s-', a' is the degree of dehydroxylation (relative to the appropriate calcine) during the TG run. T/ "C: X, 700 N, 800; 0,900; 0,950; 0,1000; U, 1050; 0, 1100; A, 1150; A, 1200 100 I 80 h 60 s v -8 40 20 0 200 400 600 800 1000 1200 V"C Fig. 7 Dehydroxylation curves for flash calcines produced with a fixed residence time z =0.5 s in He(g) and calciner heating rate A = 12 000 K s-' as a function of calciner temperature T (symbols as for fig. 6). LY'is the degree of dehydroxylation (relative to the appropriate calcine) during the TG run may have commenced in calcines kinetically frozen at p 2 60%.The structural disruption accompanying loss of an intralamel- lar hydroxyl could cause nearby remaining hydroxyls to be less strongly held. This may be evident in the onset and form of observed deviations of the dehydroxylation curves of cal- cines from the form for kaolinite itself. For these calcines initial dehydroxylation occurs more readily than for parent kaolinite (for which p =0% by definition). Otero-Arean et al. heated kaolinite in a laboratory furnace to 693 K (slow heating analogous to soak calcination) for different periods of time to prepare calcines with increasing degrees of dehydr~xylation.~~ Variations in broad-line 'H NMR absorption spectra permitted construction of an incom- plete picture of the locations of residual hydroxyls in those calcines.From variations in the second moments (M,) of the absorption spectra they concluded that the initial stages of dehydroxylation during soak calcination involve layer-by- layer loss of interlamellar hydroxyls. For soak calcines with fl> 70% some intralamellar hydroxyls must also have been removed and information about remaining 'H environments followed from deconvolution of a minimum of two contribu- tory Gaussian spectra. Environments for remaining hydroxyls are discussed in terms of two classes: the first is said to be 'patches' of groups of hydroxyls and the second to be isolated individual hydroxyls.As /3 is increased the patches appear to shrink and there is a higher proportion of isolated hydroxyls. As discussed above, the course of dehydroxylation during flash calcination in this study (involving rapid dehydroxylation at much higher temperatures) need not necessarily be the same as that during slow heating (though it appears attractive to interrelate the model postulated in this study and that of Otero-Arean et al.). A broad-line 'H NMR is necessary to examine the detailed distribution of hydroxyls in flash calcines. Preliminary "Si and 27Al NMR studies of flash calcines have revealed massive structural disorder.21 Atomic-level structural aspects of flash calcines are the subject of continuing NMR studies. Further thermal investigations are in hand, using the techniques of isothermal dehydroxylation and differential scanning calorimetry.We thank SERC for supporting this study under grant GR/ E 81999. H.A. thanks Istanbul Technical University for study leave. References 1 W. D. Keller, Geology Today, 1985, 109. 2 R. E. Grim, Clay Mineralogy, McGraw-Hill, New York, 2nd edn., 1968. 3 R. Roy, D. L. Roy and E. E. Francis, J. Am. Ceram. Soc., 1955, 38, 198. 4 G. W. Brindley and M. Nakahara, J. Am. Ceram. SOC., 1957, 40, 346. 5 F. Toussaint, J. J. Fripiat and M. C. Gastuche, J. Phys. Chem., 1963, 67, 26. 6 G. W. Brindley, J. H. Sharp, J. H. Patterson and B. N. N. Achar, Am. Mineral., 1967, 52, 201. 7 R. C. Mackenzie, The Diflerential Thermal Investigation of Clays, Mineralogical Society, London, 1957.8 P. Murray and J. White, Proc. Br. Ceram. SOC.,1949,45, 187. 9 J. N. Weber and R. Roy, Am. Mineral., 1965, 50, 1038. 10 J. B. Holt, I. B. Cutler and M. E. Wadsworth, J. Am. Ceram. SOC., 1962, 45, 133. 11 V. J. Hurst and A. C. Kunkle, Clays Clay Minerals, 1985, 33, 1. 12 A. C. D. Newman, Chemistry of Clays and Clay Minerals, Mineral Society, London, 1987. 13 S. Mazumdar and B. Mukherjee, J. Am. Ceram. SOC., 1983,66,610. 14 K. J. D. Mackenzie, I. W. M. Brown, R. H. Meinhold and M. E. Bowden, J. Am. Ceram. SOC., 1985, 68, 293. 756 J. MATER. CHEM., 1991, VOL. 1 15 I. W. M. Brown, K. J. D. Mackenzie, M. E. Bowden and 23 C. Otero-Arean, M. Letellier, B. C. Gerstein and J. J. 16 17 18 19 20 21 R. H. Meinhold, J. Am. Ceram. Soc., 1985, 68, 298. T. W. Davies, Chem. Eng. Res. Dev., 1985, 63, 82. T. W. Davies, J. Muter. Sci. Lett., 1986, 5, 186. D. Bridson, T. W. Davies and D. P. Harrison, Clays Clay Minerals, 1985, 33, 258. T. W. Davies, High Temperature Technology, 1984, 2, 141. R. C. T. Slade, T. W. Davies, H. Atakiil, R. M. Hooper and D. J. Jones, J. Muter. Sci., in the press. R. C. T. Slade and T. W. Davies, J. Muter. Chem., 1991, 24 25 Fripiat, in Proceedings of the International Clay Confer- ence 1981 -Bologna, Van Olphen and Veniale, Elsevier, Amsterdam, 1982, p. 73. J. M. Criado, A. Ortega, C. Real and E. Torres de Torres, Clay Minerals, 1984, 19, 653. T. W. Davies and A. D. Diaper, in Proceedings of the 2nd UK Conference on Heat Transfer -Glasgow, Insti-tution of Mechanical Engineers, London, 1988, vol. 2, p, 1201. 22 1, 361. R. H. Meinhold, K. J. D. Mackenzie and I. W. M. Brown, Paper 1 /00612F; Received 1 1 th February, 1991 J. Muter. Sci. Lett., 1985, 4, 163.

ISSN:0959-9428    

DOI:10.1039/JM9910100751

出版商:RSC    

年代:1991

数据来源: RSC