摘要:

Contents 4259 4269 4277 4287 4295 431 1 4321 4335 Protonation Constant of Caffeine in Aqueous Solution M. Spiro, D. M. Grandoso and W. E. Price Ionic Equilibria in Acetonitrile Solutions of 2-, 3- and 4-Picoline N-oxide Perchlorates, studied by Potentiometry and Conductometry L. Chmurzynski, A. Wawrzyn6w and Z. Pawlak Liquid-phase Adsorption of Binary Ethanol-Water Mixtures on NaZSM-5 Zeolites with Different Silicon/Aluminium Ratios W-D. Einicke, M. Heuchel, M. v.Szombathely, P. Brauer, R. Schollner and 0. Rademacher Influence of Oxidation/Reduction Pretreatment on Hydrogen Adsorption on Rh/TiO, Catalysts. An lH Nuclear Magnetic Resonance Study J. P. Belzunegui, J. M. Rojo and J. Sanz Volumetric Properties of Mixtures of Simple Molecular Fluids A. C. Colin, E. G. Lezcano, A.Compostizo, R. G. Rubio and M. D. Peiia Study of Ultramicroporous Carbons by High-pressure Sorption. Part 4.-Iso- thems and Kinetic Transport in Activated Carbons J. E. Koresh, T. H. Kim, D. R. B. Walker and W. J. Koros Kinetic and Equilibrium Studies associated with the Solubilisation of n- Pentanol in Micellar Surfactants G. Kelly, N. Takisawa, D. M. Bloor, D. G. Hall and E. Wyn-Jones The effect of Carboxylic Acids on the Dissolution of Calcite in Aqueous Solution. Part 1 .-Maleic and Fumaric Acids R. G. Compton, K. L. Pritchard, P. R. Unwin, G. Grigg, P. Silvester, M. Lees and W. A. House 130-2Contents 4259 4269 4277 4287 4295 431 1 4321 4335 Protonation Constant of Caffeine in Aqueous Solution M. Spiro, D. M. Grandoso and W. E. Price Ionic Equilibria in Acetonitrile Solutions of 2-, 3- and 4-Picoline N-oxide Perchlorates, studied by Potentiometry and Conductometry L.Chmurzynski, A. Wawrzyn6w and Z. Pawlak Liquid-phase Adsorption of Binary Ethanol-Water Mixtures on NaZSM-5 Zeolites with Different Silicon/Aluminium Ratios W-D. Einicke, M. Heuchel, M. v.Szombathely, P. Brauer, R. Schollner and 0. Rademacher Influence of Oxidation/Reduction Pretreatment on Hydrogen Adsorption on Rh/TiO, Catalysts. An lH Nuclear Magnetic Resonance Study J. P. Belzunegui, J. M. Rojo and J. Sanz Volumetric Properties of Mixtures of Simple Molecular Fluids A. C. Colin, E. G. Lezcano, A. Compostizo, R. G. Rubio and M. D. Peiia Study of Ultramicroporous Carbons by High-pressure Sorption. Part 4.-Iso- thems and Kinetic Transport in Activated Carbons J. E. Koresh, T. H. Kim, D. R. B. Walker and W. J. Koros Kinetic and Equilibrium Studies associated with the Solubilisation of n- Pentanol in Micellar Surfactants G. Kelly, N. Takisawa, D. M. Bloor, D. G. Hall and E. Wyn-Jones The effect of Carboxylic Acids on the Dissolution of Calcite in Aqueous Solution. Part 1 .-Maleic and Fumaric Acids R. G. Compton, K. L. Pritchard, P. R. Unwin, G. Grigg, P. Silvester, M. Lees and W. A. House 130-2

ISSN:0300-9599    

DOI:10.1039/F198985FX009

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY ASSOCIAZIONE ITALIANA DI CHIMICA FlSlCA DEUTSCHE BUNSEN-GESELLSCHAFT FUR PHYSIKALISCHE CHEMIE KONINKLIJKE NEDERLANDS CHEMISCHE VERElNlGlNG SOCIETE FRANGAISE DE CHIMIE, DIVISION DE CHlMlE PHYSIQUE FARADAY DIVISION GENERAL DISCUSSION No. 90 Colloidal Dispersions University of Bristol, 10-12 September 1990 Orga nising Com mitte e Professor R. H. Ottewill (Chairman) Professor P. Botherol Professor E. Ferroni Or J. W. Goodwin Professor H. Hoff mann Professor A.L. Smith Professor P. Stenius Dr Th. F. Tadros Professor A. Vrij Dr D. A. Young The joint meeting of the Societies will be directed towards examining current understanding of the behaviour of colloidal dispersions. In particular, stability and instability, short range interactions, dynamic effects, non-equilibrium interaction, concentrated dispersions and order-disorder phenomena will form topics for discussion.The preliminary programme is now availablemay be obtained from: Mrs Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W1V OBN. THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY SYMPOSIUM No. 26 Molecular Transport in Confined Regions and Membranes Oxford, 17-18 December 1990 Experimental, theoretical and simulation studies which address fundamental aspects of molecular transport will be discussed in the following main areas: a) Transport of atoms and molecules in pores, zeolite networks and other cavities; time-dependent statistical mechanics of small systems in confined geometries b) Molecular transport through synthetic membranes, biological membranes, smectic liquid crystalline phases and Langmuir Blodgett films; the dynamics of the molecules forming the membrane c) Diffusion, reorientation, conformational dynamics, viscosity and conductivity of polymer melts, to include papers dealing with bulk systems since the segments of the polymer will move in the anisotropic environment of the complete chain d) Applications of Brownian dynamics to the study of diffusion in porous media and across membranes including the transport of flexible aggregates such as microemulsions e ) The growth of crystals, colloidal aggregates and droplets on irregular surfaces and in pores Contributions for consideration by the Organising Committee are invited and abstracts of about 300 words should be sent by 31 December 1989 to: Dr D.J. Tildesley, Department of Chemistry, The University, Southampton SO9 SNH. Full papers for publication in the Symposium Volume will be required by August 1990.THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY ASSOCIAZIONE ITALIANA DI CHIMICA FlSlCA DEUTSCHE BUNSEN-GESELLSCHAFT FUR PHYSIKALISCHE CHEMIE KONINKLIJKE NEDERLANDS CHEMISCHE VERElNlGlNG SOCIETE FRANGAISE DE CHIMIE, DIVISION DE CHlMlE PHYSIQUE FARADAY DIVISION GENERAL DISCUSSION No. 90 Colloidal Dispersions University of Bristol, 10-12 September 1990 Orga nising Com mitte e Professor R. H. Ottewill (Chairman) Professor P. Botherol Professor E. Ferroni Or J. W. Goodwin Professor H. Hoff mann Professor A.L. Smith Professor P. Stenius Dr Th.F. Tadros Professor A. Vrij Dr D. A. Young The joint meeting of the Societies will be directed towards examining current understanding of the behaviour of colloidal dispersions. In particular, stability and instability, short range interactions, dynamic effects, non-equilibrium interaction, concentrated dispersions and order-disorder phenomena will form topics for discussion. The preliminary programme is now availablemay be obtained from: Mrs Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W1V OBN. THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY SYMPOSIUM No. 26 Molecular Transport in Confined Regions and Membranes Oxford, 17-18 December 1990 Experimental, theoretical and simulation studies which address fundamental aspects of molecular transport will be discussed in the following main areas: a) Transport of atoms and molecules in pores, zeolite networks and other cavities; time-dependent statistical mechanics of small systems in confined geometries b) Molecular transport through synthetic membranes, biological membranes, smectic liquid crystalline phases and Langmuir Blodgett films; the dynamics of the molecules forming the membrane c) Diffusion, reorientation, conformational dynamics, viscosity and conductivity of polymer melts, to include papers dealing with bulk systems since the segments of the polymer will move in the anisotropic environment of the complete chain d) Applications of Brownian dynamics to the study of diffusion in porous media and across membranes including the transport of flexible aggregates such as microemulsions e ) The growth of crystals, colloidal aggregates and droplets on irregular surfaces and in pores Contributions for consideration by the Organising Committee are invited and abstracts of about 300 words should be sent by 31 December 1989 to: Dr D.J. Tildesley, Department of Chemistry, The University, Southampton SO9 SNH. Full papers for publication in the Symposium Volume will be required by August 1990.

ISSN:0300-9599    

DOI:10.1039/F198985BX011

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ISSN 0300-9599 JCFTAR 85(3) 479-781 (1989) 479 485 493 503 52 1 53.7 55 1 561 579 585 60 1 609 62 1 633 645 655 663 677 69 1 699 JOURNAL OF THE CHEMICAL SOCIETY Faraday Transactions I Physical Chemistry in Condensed Phases Heat of Solution and Electric Conductivity of Electrolytes in Water- Tetrahydrofuran Mixtures s. Taniewska-Osinska, Z. Kozlowski, B. Nowicka, A. Bald and A. Szejgis A Kirkwood-Buff Theoretical Approach to Debye-Huckel Theory. Inter- pretation of Electrolyte Activity Coefficients in Both Dilute and Concentrated Solutions K. E. Newman X-Ray Diffraction and Raman Spectral Analysis of Molten CdCl, Y. Takagi, N. Itoh and T. Nakamura Photosensitised Oxidation of Water by CdS-based Suspensions A. Mills and G. Williams Acid-Base Equilibria in Aqueous Micellar Solutions.Part 1 .-' Simple ' Weak Acids and Bases Acid-Base Equilibria in Aqueous Micellar Solutions, Part 2.-Sulphone- phthalein Indicators Acid-Base Equilibria in Aqueous Micellar Solutions. Part 3.-Azine Derivatives Acid-Base Equilibria in Aqueous Micellar Solutions. Part 4.-Azo Indicators C. J. Drummond, F. Grieser and T. W. Healy Surface Oxygen Species involved in the Oxidation of Carbon Monoxide over Chromium(Ir1) Oxide M. Kobayashi and T. Kanno Stepwise Adsorption at the Same Site. A Thermodynamic Treatment E. Garrone and P. Ugliengo Electron Spin Resonance Investigation of the Copper(I1)-P-Glucosidase Interaction in Aqueous Solution Photocatalysed Isomerization of Butenes on MgO Powders with Coordinatively Unsaturated Surface Ions M. Anpo, Y. Yamada, S.Coluccia, A. Zecchina and M. Che Stability and Structure of Formamide and Urea Dimers in Aqueous Solution. A Theoretical Study P. Cristinziano, F. Lelj, P. Amodeo, G. Barone and V. Barone Reactivation of Zeolite and Oxide Catalysts using Nitrous Oxide G. J. Hutchings, H. Comninos, R. G. Copperthwaite, L. Jansen van Rensburg, R. Hunter and T. Themistocleous A Study of the Exchange Sites per Unit Area of External Surface of Zeolite AI-ZSM-5 Reversible Volume Change of Microparticles in an Electric Field R. Kishi and Y. Osada Solutions of Organic Solutes. Part 4.-Structure and the Effect of Solute Concentration on the Volume Reactions of the Hydroxyl Free Radical with Coppe(I1)-Amino-acid Complexes in Aqueous Solution G. R. A. Johnson, N. B. Nazhat and R. A. Saadalla- Nazhat Fourier Transform Infrared Study of the Adsorption and of Reactions of Acetaldehyde on Dispersed Silica W.Hill, H. Miessner and G. Ohlmann Studies of Electrical Transport Processes in Polyelectrolyte Solutions H. Vink C. J. Drummond, F. Grieser and T. W. Healy C. J. Drummond, F. Grieser and T. W. Healy C. J. Drummond, F. Grieser and T. W. Healy F. Laschi and C. Rossi G. P. Handreck and T. D. Smith J. V. Leyendekkers 17 F A R IContents 711 Formation of 'CF, and 'CCl, Radicals by Unimolecular Decomposition of (CF,COR)'+ and (CCl,COR)'+ Radical Cations C. J. Rhodes, L. Portwood and M. C. R. Symons Infrared Study of the Adsorption of Ethyl Ethanoate on Barium Sulphate W. Neagle and C. H. Rochester A Computer Simulation of the Solvation of a Solvatochromic Pyridinium Betaine Transfer Chemical Potentials, Solubilities and Reactivities in Binary Aqueous Mixtures of two Complex Iron(I1) Cations : Tris[(3-methoxyphenyl)imino- p hen y 1- 2-p yrid ylme t hane] iron( 11) and 1 ,8 - Bis[ (2-quinol ylme t h y 1ene)aminol- 3,6-diazaoctane Iron(r1) M. J. Blandamer, J. Burgess, P. Guardado and C. D. Hubbard Alkaline Hydrolyses of Ester-linked Detergents in the Solution Phase and at the Air-Water Interface as studied by Surface-tension Measurements T. Okubo and Y. Ohyama Rotating-disc Electcodes and the Theory of CE Processes. Arbitrary Rate Constants and Diffusion Coefficients R. G. Compton and R. G. Harland An Alternating Current Impedance Study of Polypyrrole/Poly(vinyl chloride) Composites A. M. Waller, A. N. S. Hampton and R. G. Compton 719 727 735 M. A. Beckett and J. G. Dawber 749 761 773

ISSN:0300-9599    

DOI:10.1039/F198985FP031

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

The following papers were accepted for publication in Faraa2zy Transactions I during December 1988. 8D3273D 8P3332C 8/03351 J 8/03623C 8~03627F 8D3646B 8D3726D 8/03727B 8/03797C 8P3798A 81039421 8/04131H 8/04239J 8/02034E 8/020405 8D2553C 8/02795A The Growth of Phase IV Ammonium Nitrate Crystals and their Transformation to the Phase I11 Structure Davey, R. J., Guy, P. D., Mitchell, B., Ruddick, A. J. and Black, S. N. Kinetics of Reaction between Bromophenol Blue and Hydroxide Ions in Aqueous Salt Solutions at 298-15 K. Applications of Piker’s Equations for Ionic Activity Coefficients to Kinetic Salt Effects Blandamer, M. J., Burgess, J., Cottrell, M. R., Hakin, A. W., Horn, I. M. and Sanchez, F. Rotating Disc Electrode Voltammetry: The Digital Simulation of the Current-Voltage Behaviour of Electrode Processes involving Reversible Electron Transfer and Coupled Homogeneous Kinetics Compton, R.G. and Unwin, P.R. Modification of Mordenite Zeolites by Chemisorption of Disilane and its Influence on the Adsorption Properties. Part 1.- A Modification Parameter Study Yan, Y.-A., Verbiest, J., de Hulsters, P. and Vansant, E. F. Modification of Mordenite Zeolites by Chemisorption of Disilane and its Influence on the Adsorption Properties. Part 3.- An Adsorption Study Yan, Y.-A., Verbiest, J., de Hulsters, P. and Vansant, E. F. Stereochemistry of Solutions. Part 21.- Inner- and Outer-sphere Complexes of Lithium with Thiocyanate in Acetonitrile Solutions Gans, P., Gill, J. B. and Longdon, P. J. Isomeric Equilibria of Monosaccharides in Solution. Influence of Solvent and Temperature Franks, F., Lillford, P, J.and Robinson, G, Silicon-29 Shielding Tensors in Solid Organosilicon Compounds, Studied by Slow-spinning N.M.R. Harris, R. K., Pritchard, T. N. and Smith, E. G. Electrostatic Bond Dipole Moments from Dimethylether, Methanol, Methane and Water Huiszoon, C. Reactions of Unsaturated Hydrocarbons on Rutile and Anatase Brookes, B.I., Bird, R., Kemball, C. and Leach, H. F. On the Second Emission of the Uranyl Ion in Aqueous Solution Marcantonatos, M. D. and Pawlowska, M. M. I.R. Study of the Thermal Behaviour of Heteropolyacids Bielanski, A., Malecka, A. and Kubelkowa, L. Perhydrotriphenylene Radical: An Initiator for Inclusion Polymerization as Characterized by E.S.R. Spectroscopy Sozzani, P., Scotti, R.and Morazzoni, F. Effect of Temperature on the Salt-induced Sphere-Rod Transition of Dodecyltrimethylammonium Bromide Micelles in Aqueous NaBr Solutions Zielinski, R., Ikeda, S., Nomura, H. and Kato, S. A B .E.T.-like Three-solvation-stage Sorption Isotherm Timmermann, E. 0. Role of Solvent Dynamics on Kinetics of Electro-oxidation of Ferrocene at a Platinum Electrode Khan, S. U. M. Heat Capacities for the Formation of 1-1 Complexes of Lasalocid with Alkali-metal and Alkaline-earth-metal Cations in Methanol Woznicka, J., Lhermet, C., Morel-Desrosiers, N., Morel, J.-P. and Juillard, J. Mechanistic Studies of Capillary Processes in Porous Media. Part 1.- Probabilistic Description of Porous Media Cruz, M., Mayagoitia, V. and Rojas, F. Mechanistic Studies of Capillary Processes in Porous Media.Part 2.- Construction of Porous Networks by Monte-Carlo Methods Cruz, M., Mayagoitia, V. and Rojas, F. (98/03023E 8/03025A 8/03057J 8/03059F 8/03202E 8103207F 8/03333A 8/03334J 8/033741 8/03629B 8/03631D 8/03636E 8/03822H 8/03842B 8/0384411 8/0348921 8/03945C 8/04179B 8/04360D 8/04465A Crystal Structures of Different Dealuminated Y-Type Zeolites. Determination of Framework Vacancies and Non-framework Species Jeanjean, J. Dereigne, A., Aouali, L. and Delaf'osse, D. An E.S.R. Study on the Interaction of Oxygen with Ag/SiO2 Yeh, C.-T., Wang, Y.-P. and Chien, S.-H. Photovoltaic and Photocatalytic Behaviour of Ferroelectric Semiconductor, Lead Strontium Zirconate Titanate, with Polarization Axis Perpendicular to the Surface Inoue, Y., Sato, K.and Sato, K. Complex Formation and Self-association in Ternary Mixtures . Apparent Heat Capacities for Alcohol-Methyl Acetate-Hydrocarbon and Acetone-chloroform- Cyclohexane Costas, M., Yao, 2.-T. and Patterson, D. Study of Ultamicroporous Carbons by High-pressure Sorption. Part 3.- Complex Transport Phenomena as sensed by COZ and N2 Kinetics Koresh, J. E., Kim, T. H., Walker, D. R. B. and Koros, W. J. Solution of Hydrogen in Thin Films of Pd-Ag Alloy Kishimoto, S., Yoshida, N., Tanabe, T. and Flanagan, T. B. Diffusion of Ethane in Silicate- 1 by Frequency Response, Sorption Uptake and N.M.R. Techniques Van-Den-Begin, N., Rees, L. V. C. , Caro, J., Bulow, M., Hunger, M. and Karger, J. Structure of Germanium CVD Zeolite by EXAFS and XPS Hibino, T., Sano, M., Miki, N.and Murakami, Y. Chromia/Silica-Titania Cogel Catalysts for Ethene Polymerization: Polymer Characteristics Conway, S. J., Falconer, J. W. Rochester, C. H. and Downs, G. W. Photolysis of Cyclotetrasilanes. Remarkable Dependence of Molecular Structures Shizuka, H., Murata, K., Arai, Y., Tonokura, K-i., Tanaka, H., Matsumoto, H., Nagai, Y., Gillette, G. R. and West, R. Peroxy Radical Reactions in the Photo-oxidation of CH3CHO Moortgat, G. K., Cox, R. A., Schuster G., Burrows, J. P. and Qndall, G. S. Reaction of Preadsorbed Methane with Oxygen over Magnesium Oxide at Low Temperatures Ito, T., Watanabe, T., Tashiro, T, and Toi, K. Flat-Vertical Transitions of Fluoranthene Molecules Adsorbed on a Mercury Electrode Compton, R. G., Harland, R. G. andNorthing, R.J. The Effect of Pressure on the Electrical Conductivity of Liquid Iodine, Iodine Chloride, Iodine Bromide and Bromine Trifluoride Cleaver, B. and Condlyffe, D. H. Fluorescence of Cyclotetrasilanes in Rigid Glass at 77 K. A Remarkably Large Stokes Effect Shizuka, H., Murata, K., Arai, Y., Tonokura, K-i., Hiratsuka, H., Matsumoto, H. and Nagai, Y. Electrochemical Characterization of BaSn~-~Sbfls Perovskites De Silva Pereira, M. I., Melo, M. J. B. V., Da Costa Fernanda, M. A., Nunes, M. R. and Peter, L. M. High-resolution 'H N.M.R. Study of Hydrogen Sulphide Adsorption on Heterogeneous Catalysts Mastikhin, V. M., Mudrakovski, I. L.,Nosov, A. V. and Mashkina, A. V. Photochemical Properties of Iron Oxide incorporated in Clay Interlayers Miyoshi, H. and Yoneyama, H, Computer Simulations of Fluids in Zeolites X and Y Woods, G.B. and Rowlinson, J. S . Solubilities, Solubility Products and Solution Chemistry of Lanthanon Trifluoride-Water Systems Menon, M. P. and James, J. (ii)Cumulative Author Index 1989 Aguilella, V. M., 223 Akitt, J. W., 121 Albuquerque, L. M. P. C., 207 Allen, G. C., 55 Amodeo, P., 621 Anpo, M., 609 Apelblat, A., 373 Asakura, K., 441 Bald, A., 479 Barone, G., 621 Barone, V., 621 Beckett, M. A., 727 Bengtsson, L., 305, 317 Berry, F. J., 467 Bertoldi, M., 237 Blandamer, M. J., 735 Bond, G. C., 168 Borowko, M., 343 Boss, R. D., 11 Bowker, M., 165 Brimblecome, P., 157 Burgess, J., 735 Busca, G., 137, 237 Chadwick, A. V., 166 Che, M., 609 Chen, L-f., 33 Clegg, S. L., 157 Coluccia, S., 609 Comninos, H., 633 Compton, R.G., 761, 773 Conway, S. J., 71, 79 Copperthwaite, R. G., 633 Cox, B. G., 187 Cristinziano, P., 621 Datka, J., 47 Dawber, J. G., 727 De Giglio, A., 23 Dell’Atti, A., 23 Donini, J. C., 91 Drummond, C. J., 521, 537, 551, el Torki, F. M., 349 Falconer, J. W., 71, 79 Finch, J. A., 91 Fletcher, P. D. I., 147 Foo, C . H., 65 Frey, H. M., 167 Fubini, B., 237 Gabriel, C. J., 11 Gabrys, B., 168 Garrone, E., 585 Geus, J. W., 269, 279, 293 Giamello, E., 237 Gilbert, P. J., 147 Gottschalk, F., 363 Grieser, F., 521, 537, 551, 561 Guardado, P.. 735 56 I Hampton, S., 773 Handreck, G. P., 645 Harland, R. G., 761 Hasted, J. B., 99 Hatano, M., 199 Healy, T. W.. 521, 537, 551, 561 Hesselink, W. H., 389 Hester, R. E., 171 Higgins, J. S., 170 Higuchi, A., 127 Hill, W., 691 Holmberg, B., 305, 317 Hong, C.T., 65 Howarth, 0. W., 121 Hubbard, C. D., 735 Hunter, R., 363, 633 Hutchings, G. J., 363. 633 Ichikawa, K., 175 Ikeda, R., 111 Ishida, H., I 1 1 Itoh, N., 493 Iwasawa, Y., 441 Jin, T., 175 Johnson, G. R. A., 677 Jonkers, G., 389 Jutson, J. A., 55 Kanno, T., 579 Katoh, T., 127 Keeler, J. H., 163 Kelebek, $., 91 Kishi, R., 655 Knijff, L. M., 269, 293 Kobayashi, M., 579 Kozlowski, Z., 479 Laschi, F., 601 Lawrence, K. G., 23 Lelj, F., 621 Levy, O., 373 Leyendekkers, J. V., 663 Lorenzelli, V., 137 Loudon, R., 169 Lund, A., 421 Mafe, S., 223 Manzurola, E., 373 Marcus, Y., 381 Markovits, G., 373 Masiakowski, J. T., 421 Matsuhashi, N., 1 1 1 Matsumoto, T., 175 Meima, G. R., 269, 279, 293 Miessner, H., 691 Mills, A., 503 Mosier-Boss, P.A., 1 1 Nakagawa, T., 127 Nakamura, D., 1 I 1 Nakamura, T.. 493 Nazhat, N. B.. 677 (iii) Neagle, W., 429, 719 Newman, K. E., 485 Nicholas, A., 773 Nowak, R. J., 11 Nowicka, B., 479 Ohlmann, G., 691 Ohyama, Y., 749 Okubo, T., 455, 749 Orchard, S. W., 363 Osada, Y., 655 Otsuka, K., 199 Pandey, J. D., 331 Pellicer, J., 223 Piwowarska, Z., 47 Portwood, L., 71 1 Price, W. E., 415 Rai, R. D.. 331 Ramis, G., 137 Rao, K. J., 251 Reed, W. F., 349 Rees, L. V. C., 33 Reis, J. C. R., 207 Rhodes, C. J., 711 Rochester, C. H., 71, 79, 429, Rosen, D., 99 Rossi, C., 601 Rowlinson, J. S., 171, 172 Saadalla-Nazhat, R. A., 677 Sacco, A,, 23 Said, M., 99 Schmehl, R. H., 349 Schneider, H., 187 Schneider, I., 187 Selvaraj, U., 251 Shido, T.. 441 Shukla, R. K., 331 Smith, G.W., 91 Smith, J. J., 1 1 Smith, M. R.. 467 Smith, T. D., 645 Stroka, J., 187 Sundar, H. G. K., 251 Symons. M. C. R., 71 1 Szejgis, A., 479 Szpak, S., 11 Takagi, Y., 493 Taniewska-Osinska, S., 479 Thamm, H., 1 Themistocleous, T., 633 Ugliengo, P., 585 Vaccari, A.. 237 van Buren, F. R., 269. 279, 293 van Dillen, A. J., 269, 279, 293 van Leur, M. G. J . . 279 van Rensburg, L. J., 633 van Veen. J. A. R., 389 719Vink, H., 699 Vis, R. J., 269, 279 Wacker, T., 33 Waller, A. M., 773 AUTHOR INDEX Waugh, K. C., 163 Weale, K. E., 165 Williams, G., 503 Yamada, Y., 609 Yeh, C-t., 65 Young, D. A., 173 Zecchina, A., 609THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION No. 87 Catalysis by Well Characterised Materials University of Liverpool, 11 -1 3 April 1989 Organking Committee: Professor R.W. Joyner (Chairman) Professor A. K. Cheetham Professor F. S. Stone Dr K. C. Waugh Professor P. B. Wells The understanding of heterogeneous catalysis is an important academic activity, which complements industry’s continuing search for novel and more efficient catalytic processes. The emergence of rele- vant, in particular in sifu techniques and new developments of well established experimental approaches to catalyst characterisation are making a very significant impact on our knowledge of catalyst composition, structure, morphology and their inter-relationships. Well characterised cata- lysts, which will be the subject of the Faraday Discussion, include singlecrystal surfaces, whether of metals, oxides or sulphides; crystalline microporous solids, such as zeolites and clays, and ap- propriate industrial catalysts.The elucidation of structure/function relationships and catalytic mechanism will be important aspects of the scientific programme. Contributions describing novel methods for synthesising well characterised catalysts and also reporting important advances in char- acterisation techniques will also be included. The final programme and application form may be obtained from: Mrs Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W1V OBN. THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION No. 88 Charge Transfer in Polymeric Systems University of Oxford, 11-13 September 1989 This Discussion aims to bring together physicists and chemists interested in the mechanism of elec- tron and ion transport in polymeric systems. The systems include conducting polymers, redox polymers, ion exchange membranes and modified electrodes.Discussion topics will cover ex- perimental evidence from spectroscopy, electrochemistry and new techniques such as the quartz microbalance. Theoretical models ranging from band theory through polarons to localised chemical structures will be critically evaluated and compared with experiment. The following have agreed to participate in the Discussion: R. Murray W. J. Albery M. B. Armand D. Bloor P. G. Bruce R. Friend A. J. Heeger A. R. Hillman A. G. MacDiarmid M. Ratner S. Roth W. Salaneck G. Tourilton C. Vincent G. Wegner The preliminary programme may be obtained from: Mrs Y.A. Fish, The Royal Society of Chemistry, Burlington House, London W1V OBN.THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION No. 89 Structure of Surfaces and Interfaces as Studied using Synchroton Radiation University of Manchester, 4-6 April 1990 Organising Committee: Professor J. N. Shetwood (Chairman) Professor D. A. King Dr G. King The Discussion will focus on the wealth of novel information which can be obtained on the nature and structure of surfaces using the full spectral range of synchroton radiation. Emphasis will be placed on the scientific results of recent investigations rather than on technical aspects of experimen- tation. Papers will be welcome which shed new light on the structure of the complete range of interfaces: solidlsolid, solid/gas, solidniquid, gasniquid and "clean" surfaces including both static and dynamic in sifu examinations.It is hoped that the discussion will define the utility of synchroton radi- ation examinations in surface science studies at a time of expansion of the availability of such sources and the inauguration of new and more powerful sources. Contributions for consideration by the Organising Committee are invited and abstracts of about 300 words should be sent by 31 May 1989 to: Professor J. N. Sherwood, Department of Pure and Applied chemistry, University of Strath- Clyde, Thomas Graham Building, 295 Cathedral Street, Glasgow G1 1XL Full papers for publication in the Discussion Volume will be required by December 1989. Dr C. Norris Dr R.Oldman Dr G. Thornton THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY SYMPOSIUM No. 25 Large Gas Phase Clusters University of Warwick, 12-14 December 1989 Organising Committee: Professor K. R. Jennings (Chairman) Professor P. J. Derrick Professor D. Phillips The Symposium will focus on recent developments in the rapidly expanding field of large gas phase clusters, including the preparation, structure and reaction of both neutral and ionic clusters. It is hoped that the meeting will bring together scientists working on many different types of cluster, e.g. rare gas atoms, metals, inorganic and organic species, and biomolecules, to discuss the chemistry and physics of clusters from different viewpoints. Contributions for consideration by the Organising Committee are invited.Titles and abstracts of about 300 words should be submitted by 15 February 1989 to: Professor K. R. Jennings, Department of Chemistry, University of Warwick, Coventry CV4 7AL, England. Full papers for publication in the Symposium volume will be required by 14 August 1989. Dr N. Quirk Dr R. P. H. Rettschnick Dr A. J. StaceDEUTSC H E BU NSEN-GESELLSCHAFT FUR PHYSIKALISCH E CH E MI E ASSOCIAZIONE ITALIANA DI CHIMICA FlSlCA FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY SOCIETE FRANGAISE DE CHIMIE, DIVISION DE CHlMlE PHYSIQUE JOINT DISCUSSION MEETING 1989 Transport Processes in Fluids and in Mobile Phases Aachen, 25-27 September 1989 Organised by: H.Versmold (F. R.G.) Al. Weiss (F.R.G.) M. Zeidler (F.R.G.) G. R. Luckhurst (U.K.) P. Turq (France) The purpose of the meeting is to bring together scientists working on transport and related phenomena in simple and complex fluids, colloidal and micellar systems, and surface phases.Experimental techniques considered indude classical methods, optical spectroscopy, light scattering, nuclear magnetic resonance, and neutron scattering. The following persons have accepted invitations to present talks: D. Evans, Canberra; B. U. Felderhof, Aachen; D. Frenkel, Amsterdam; A. Geiger, Dortmund; W. GIBser, Grenoble; H. G. Hertz, Karlsruhe; S. Hess, Berlin; J. Jonas, Urbana; R. Klein, Konstanz; K. Lucas, Duisburg; H.-D. Ludermann, Regensburg; H. Posch, Wen; P. Pusey, Malvern; J. P. Ryckaert, Brussels; W. A. Steele, Penn State; D. J. Tildesley, Southampton; H. Weingartner, Karlsruhe.Further details may be obtained from: Professor H. Versmold, lnstitut fur Physikalische Chemie, RWTH Aachen, Templergraben 59, D-5100 Aachen. Federal ReDubllc of Germanv. (vii)FARADAY DIVISION INFORMAL AND GROUP MEETINGS Theoretical Chemistry Group Graduate Students’ Meeting To be held at University College, London on 1 March 1989 Further information from Dr P. Fowler, Depament of Chemistry, University of Exeter, Exeter EX4 4QD ~~~ Neutron Scattering Group Neutron and X-ray Scattering: Complementary Techniques To be held at the University of Kent at Canterbury on 29-30 March 1989 Further information from Dr R. J. Newport, Physics Laboratory, University of Kent, Canterbury CT2 7NR Polar Solids Group Atomic Mechanisms of Mass Transport in Solids To be held at Mansfiekl College, Oxford, on 29-31 Mar& 1989 Further information from Professor R.A. Catlow, Department of Chemistry, University of Keele, Keele, Staffordshire Division jointly with the Colloid and lnterface Science Group Annual Congress: Surfactant Interactions in Colloidal Systems To be held at the University of Hull on 4-7 April 1989 Further information from Dr J. F. Gibson, The Royal Society of Chemistry, Burlington House, London W1 V OBN Molecular Beams Group Surfaces, Ions and Clusters To be held at the University of Liverpod on $1 1 Apnll989 Further information from Dr J. M. Hutson, Department of Chemistry, University of Durham, South Road, Durham DH13LE Electrochemistry Group Spring Informal Meeting To be held at the University of Warwick on 1 @12 April 1989 Further information from Dr S.P. Tyfield, CEGB, Berkeley Nudear Laboratories, Berkeley, Gloucestershire GL13 9PB Electrochemistry Group with the Electroanalytical Group Electroanalysis To be held at Loughborough University of Technology on 12-14 Apnll989 Further information from Dr S. P. Tyfield, CEGB, Berkeley Nudear Laboratories, Berkeley, Gloucestershire GL13 9PB Gas Kinetics Group Developments in Gas Kinetics: New Techniques, Results and their Interpretation To be held at the University of York on 3-4 July 1989 Further information from Professor R. J. Donovan, Department of Chemistry, Uniwrsity of Edinburgh, West Mains Road, Edinburgh EH9 3JJ lndustrial Physical Chemistry Group with the Thin Films and Surfaces Group of the IOP Materials for Non-linear and Electrosptics To be held at Girton College, Cambridge on 4-7 July 1989 Further information from The Meetings Officer, Institute of Physics, 47 Belgrave Square, London SWlX 8QX Polymer Physics Group Biologically Engineered Polymers 89 To be held at Churchill College, Cambndge, on 31 July to 2 August 1989 Further information from Dr M.J. Richardson, Division of Maten’als, National Physical Laboratory, Queens Road, Tddngton, Middlesex lW11 OLW (viii)Polymer Physics Group Biennial Meeting To be held at the University of Reading on 13-1 5 September 1989 Further information from Dr M. J. Richardson, Division of Materials, National Physical Laboratory, Queens Road, Teddington, Middlesex lW11 OLW Colloid and Interface Science Group Inorganic Particulates To be held at Chester College on 1921 September 1989 Further information from Dr R.Buscall, ICI plc, Corporate Colloid Science Group, Po Box 11, The Heath, Runcorn, Cheshire WA7 4QE Division with the Institute of Physics Sensors and their Applications To be held at the University of Kent at Canterbury on 1922 September 1989 Further information from The Meetings Officer, Institute of Physics, 47 Belgrave Square, London SW1X 8QX Division with the Deutsche Bunsen Gesellschafi, Division de Chimie Physique of the Societe Franpise de Chimie and Assouazione ltaliana di Chimica Fisica Transport Processes in Fluids and Mobile Phases To be held at the Physikalische InstiMt, Aachen, West Germany on 2528 September 1989 Further information from Professor G.Ludhurst, Department of Chemistry, University of Southampton, Southampton SO9 5NH Division Autumn Meeting: Chemistry at Interfaces To be held at Loughborough University of Thndogy on 26-28 September 1989 Further information from Professor F. Wilkinson, Department of Chemistry, toughborough University of Technology, Loughborough LE11 3TUJOURNAL OF CHEMICAL RESEARCH Papers dealing with physical chemistry or chemical physics which appear currently in J. Chem. Research, The Royal Society of Chemistry's synopsis + microform journal, include the following: Absolute Rate Data for the Reaction of Atomic Germanium, Ge(43P~), with Halogenated Olefins and Aromatic Compounds by Tme-resolved Atomic Resonance Absorption Spectroscopy Subhash C. Basu and David Husain (1988, Issue 10) Benzo-bisdithiazole) Gotthelf Wolmershausser, Gerhard Wortmann and Martin Schnau ber (1988, Issue 11) Frank Hibbert and Rowena J.Sellens (1 988, Issue 11) Structural and Magnetic Properties of the Radical-cation SaR BBDTA"FeCl4- (BBDTA = Electrolyte Effects on the Reactions of Hydroxide Ion in 70% (vh) Dimethyl Sulphoxide-Water Studies on Sulphate Complexes. Literature Data Analysis of the Stability of HS04- and NaS04- Species in Aqueous Perchlorate Solution at Different Temperatures and Ionic Strengths Concetta De Stefano, Carmelo Rigano, Sihrio Sammartano and Rosario Scarcella (1988, Issue 11) Jaroslav Pecka (1 988, Issue 12) Contribution of an Intramolecular Hydrogen Bond to the Dipole Moment Otto Exner andNOMENCLATURE AND SYMBOLISM For many years the Society has actively encouraged the use of standard IUPAC nomenclature and symbolism in its publications as an aid to the accurate and unambiguous communication of chemical information between authors and readers.Nomenclature. The following publications provide the IUPAC nomenclature rules and guidance on their use: Nomenclature of Organic Chemistry, Sections A, 8, C, D, El F, and H (Pergamon, Oxford, 1979 edn.) Nomenclature of lnorganic Chemistry (Butterworths, London, 1971, now published by Pergamon). Biochemical Nomenclature and Related Documents (The Biochemical Society, London, 1978). Where there are no IUPAC rules for the naming of particular compounds or authors find difficulty in applying the existing rules, they should seek the advice of the Society’s editorial staff.Units and Symbols. A detailed treatment of units and symbols with specific application to chemistry, based on the Systhme lnternationale d’Unites (SI), is given in Quantities, Units and Symbols in Physical Chemistry, published for IUPAC by Blackwell Scientific Publications, Oxford (1 988 edn.). A comprehensive list of IUPAC publications on nomenclature and symbolism appears in the January issue of J. Chem. Soc., Faraday Transactions.NOMENCLATURE AND SYMBOLISM For many years the Society has actively encouraged the use of standard IUPAC nomenclature and symbolism in its publications as an aid to the accurate and unambiguous communication of chemical information between authors and readers. Nomenclature. The following publications provide the IUPAC nomenclature rules and guidance on their use: Nomenclature of Organic Chemistry, Sections A, 8, C, D, El F, and H (Pergamon, Oxford, 1979 edn.) Nomenclature of lnorganic Chemistry (Butterworths, London, 1971, now published by Pergamon). Biochemical Nomenclature and Related Documents (The Biochemical Society, London, 1978). Where there are no IUPAC rules for the naming of particular compounds or authors find difficulty in applying the existing rules, they should seek the advice of the Society’s editorial staff. Units and Symbols. A detailed treatment of units and symbols with specific application to chemistry, based on the Systhme lnternationale d’Unites (SI), is given in Quantities, Units and Symbols in Physical Chemistry, published for IUPAC by Blackwell Scientific Publications, Oxford (1 988 edn.). A comprehensive list of IUPAC publications on nomenclature and symbolism appears in the January issue of J. Chem. Soc., Faraday Transactions.

ISSN:0300-9599    

DOI:10.1039/F198985BP033

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I , 1989, 85(3), 479483 Heat of Solution and Electric Conductivity of Electrolytes in Water-Tetrahydrofuran Mixtures Stefania Taniewska-Osinska,* Zygmunt Kozlowski, Boiena Nowicka, Adam Bald and Adam Szejgis Department of Physical Chemistry and Department of Chemial Didactics, University of t d d i , ul. Nowotki 18, 91-416 t o d i , Poland Integral heats of solution of sodium perchlorate (NaCIO,) in water- tetrahydrofuran (THF) mixtures from 0 to 100molY0 THF have been measured. The electric conductivity of NaCIO, solutions in binary solvents containing from 0 to 80 mol % THF have also been measured, as have the conductivities of solutions of sodium chloride (NaCI) and sodium iodide (Nal) in mixtures containing from 0 to 35 mol% THF. Molar electro- conductivities, Walden products and association equilibria constants have been determined.Taking into account ion-pair association, standard enthalpies of solution have been recalculated. Our earlier thermochemical investigations of solutions of electrolytes and of urea in water-tetrahydrofuran (THF) mixtures were limited to the water-rich region.'. The variations of the standard enthalpies of solution us. mixed solvent composition appeared to be similar to analogous behaviour for the solute-water-t-butyl alcohol system. In this paper we present the results of measurements of AH: (NaClO,) over the whole composition range of water-THF mixtures. Taking into' account the low electric permittivity of solution with high tetrahydrofuran concentrations we considered it necessary to examine the effect of ion-pair formation on the standard enthalpies of electrolyte in solution.For this reason the electric conductance was measured for NaClO,, NaCl and NaI solutions in the mixed solvent under investigation. Experimental The purification of the salts NaClO,, NaCl and NaI and of solvent THF has been described earlier.' The enthalpies of solution of NaClO, in water-THF mixtures were measured in a calorimeter also described in ref. (1). The conductivity measurements were performed using a bridge of the type E 3 15A (Mera-Tronik, Poland). A cell similar to that described by Dagett3 was used. The temperature of the thermostat, 298.15 K, was kept constant to within k0.005 K. There have been no investigations of electric conductivity carried out in mixtures with THF contents > 35 mol%, nor have any measurements been made of heats of solution for NaCl and NaI, because of the extremely low solubility of these salts.'~2 Since the ion association is large, it was possible to perform conductometric measurements for NaC10, solutions with satisfactory precision only up to 80 mol % THF.Results and Discussion In order to determine the enthalpy of solution at infinite dilution, i.e. the standard enthalpy of solution, of NaC10, in water-THF mixtures, as well as of NaCl and NaI examined earlier1T2 in the same mixed solvent, the method presented by Barthel et al., 479 17-2480 Solution Heats and Conductivities of Electrolytes in THF-H,O Table 1. Electric conductivty of electrolytes in water-tetrahydrofuran mixtures at 298.15 K THF A0 dAo" K* dK," 6," (mol %) /cm2 mo1-I R-' /cm2 mol-' R-' /dm3 mol-' /dm3 mol-I /cm2 mol-I l2-l nb 4' 1 .o 2.5 5.0 10.0 15.0 20.0 25.0 30.0 35.0 2.5 5.0 15.0 25.0 30.0 35.0 2.5 4.2 10.0 15.0 20.0 25.0 40.0 60.0 80.0 114.46 100.87 84.90 68.18 59.79 54.52 50.01 47.12 44.85 98.85 80.79 58.41 54.96 56.35 57.41 92.78 80.66 58.70 53.86 52.57 53.53 58.87 71.20 93.33 f 0.02 f 0.02 f 0.02 f 0.02 f 0.02 f 0.02 f 0.02 f 0.04 & 0.04 - + 0.02 - + 0.02 & 0.02 f 0.02 - + 0.04 f 0.06 f 0.02 f 0.02 kO.01 f 0.02 f 0.02 fO.O1 f 0.02 fO.01 f 0.02 NaCl - - - 2.6 5.7 10.3 22.2 42.2 73.5 NaI - - - 7.9 16.7 33.8 NaClO, - - - - - I .o 31.3 352 6773 - - - fO.1 f0.1 f0.2 f 0.2 f 0.6 f 0.7 - - - kO.1 - + 0.5 f 1.0 - - - - - f0.I kO.1 + I k 6 f 0.05 f 0.03 f 0.03 +_ 0.04 f 0.04 k 0.04 f 0.04 f 0.08 f 0.08 f 0.05 f 0.05 f 0.03 f 0.04 - + 0.08 fO.10 f 0.05 f 0.05 k 0.03 f 0.05 & 0.04 f 0.0 1 f 0.04 f 0.02 k 0.04 12 8 14 14 14 16 15 15 13 14 12 13 13 12 12 11 13 14 14 10 14 10 16 19 3.69 3.88 4.23 5.02 5.95 7.02 8.21 9.5 1 10.88 3.88 4.23 5.95 8.2 1 9.5 1 10.88 3.88 4.1 1 5.02 5.95 7.02 8.2 1 12.30 18.68 26.8 1 a dAo, d K , and 6, are the standard deviations of Ao, K,, and A, respectively."Number of experimental points. Bjerrum distance parameter. was employed. In this method the relative apparent molal heat content of a solution containing 'free' ions and ion pairs is described by the following expression: where QL(FI) is the relative apparent molal heat content of a solution with 'free' ions, AH: is the enthalpy of association and a is the degree of dissociation.Taking into (2) account one obtains an expression determining the enthalpy of solution at infinite dilution : AH," = AH,"-a@,,(FI)-(l - a ) A H i (3) mL = -AH: = AHF-AHZ where AH," is the integral enthalpy of solution at a molality m. The @ , , ( F I ) values were calculated using equations proposed by Barthel et aL4 and by Wachter et al.' The necessary values of the degree of dissociation were obtained by us from conductometric measurements. Molar conductivities were analysed by means of the Fuoss-Justice equation :6 A = a[A,-~~aora+E(col)ln(col)+ J ( c ~ ) + J $ U ~ ] . (4) Analytical forms of the remaining quantities have been presented elsewhere."-"'S. Taniewska-Osinska et al. 48 1 C 10 0 6 L g I 2 -10 2 2 -2 0 -3c -\ THF (mol %) 1 1 I 1 I I I I I 1 THF (molOr0) 20 40 60 80 100 Fig.1. Standard enthalpies of solution, AH,", of electrolytes in water-tetrahydrofuran mixtures at 298.15 K. (-) A% obtained by use of eqn (3) for a = 1 ; (---) A% taking into account the effect of ionic association; 0 and ., NaI; 0 and 0 , NaCIO,; A and A, NaCl. The degree of dissociation and the association constant KA were calculated from the relationship I -a -- - KA a2CY: where y+ - is the activity coefficient and A cfat Iny, = - 1 + BqchB where A and B are coefficients of the Debye-Hiickel equation. According to Justice's' suggestions, in our work we adopted the Debye-Huckel closest-approach parameter a, equal to the Bjerrum distance q (q = e2/2&kBT). In this way eqn (4) becomes a diparametric equation, and was resolved by a least-squares rnethod.'*-l2 The values obtained of the parameters A.and KA and their standard errors dAo and 6KA are included in table 1. As can be seen from these data, ionic association was not observed, in the water-rich solutions, up to 10 mol % THF for NaCl and 20 mol O h482 Solution Heats and Conductivities of Electrolytes in THF-H,O 20 40 60 80 Q4 THF (mol%) Fig. 2. Walden product A, 17 of electrolytes in water-tetrahydrofuran mixtures at 293.15 K, A, NaCl ; 0, NaI;O, NaCIO,; A, for solutions of electrolytes in water from ref. (14); 17, our values, ref. (1 5) and (16). THF for NaI and NaC10,. In the systems with higher THF contents ionic association becomes significant, with the KA values increasing with an increase in THF content.In order to obtain true values AH," we used eqn (3). To solve this equation it is sufficient to know the experimental values of AH? [this work and ref. (1) and (2)] and the ion-pair formation equilibrium constants for NaClO,, NaCl and NaI in water-tetrahydrofuran mixtures. Eqn (3) was solved by a least-squares method which simultaneously permits one to calculate the values AH," and AH:. In fig. 1 standard enthalpies of solution AH," of electrolytes obtained by the above mentioned method are plotted along with those directly extrapolated from calorimetric data. The plots of AH," us. mol% THF both taking into account ionic association and neglecting it are of similar character. In all cases ionic association diminishes the enthalpies of solution.The plot of the standard enthalpy of solution of NaC10, passes through a minimum in tetrahydrofuran-rich solutions, suggesting a change in the interactions or structure of the mixture. This function resembles an analogous dependence in electrolyte- water-butanol systems.13 The Walden product (fig. 2) for solutions of the three investigated electrolytes in water-tetrahydrofuran mixtures was calculated. The only conclusion we were able to draw from it is that there is difference in behaviour for NaCl solutions with regard to NaI or NaClO, solutions. The same conclusion results from calorimetric data (fig. 1). Financial support for this work from the CPBP-01 .15 programme is acknowledged. References 1 S. Taniewska-Osinska, B. Piestrzyriska and R.togwinienko, Can. J. Chem., 1980, 58, 1584. 2 S. Taniewska-Osinska and B. Nowicka, Thermochim. Acta, 1987, 115, 129. 3 H. M. Dagett, E. J. Bair and C. A. Kraus, J . Am. Chem. SOC., 1951, 73, 799. 4 J. Barthel, H. J. Gores, G. Schmeer and R. Wachter, Non-Aqueous Electrolyte Solutions in Chemistry 5 R. Wachter and K. Riederer, Pure Appl. Chem., 1981, 53, 1301. and Modern Technology, Topics in Current Chemistry (Springer, Berlin, 1983), vol. I l l , p. 49.S. Taniewska-Osinska et al. 483 6 J . C. Justice, Electrochim. Acta, 1971, 16, 701. 7 R. M. Fuoss and K. L. Hsia, Proc. Natl Acad. Sci. USA, 1967, 57, 1550. 8 R. Fernandez-Prini, Trans. Faraday SOC., 1968, 64, 2146. 9 R. M. Fuoss and F. Accascina, Electrolytic Conductance, (Interscience, New York, 1959), p. 195. 10 J. Barthel, Angew. Chem., 1968, 80, 253, (Angew. Chem. Int. Ed. Engl., 1968, 7 , 260). 11 J. C. Justice, R. Bury and C. Treiner, J. Chem. Phys., 1968, 65, 1708. 12 J. Barthel, J. C. Justice and R. Wachter, 2. Phys. Chem., N.F., 1973, 84, 100. 13 S. Taniewska-Osinska and H. Piekarski, J . Solutn Chem., 1978, 7 , 12. 14 J. L. Hawes and R. L. Kay, J. Phys. Chem., 1965,69, 2420. 15 B. Nowicka, A. Kacperska, J. Barczynska, A. Bald and S. Taniewska-Osinska, J . Chem. Soc., Furaduy 16 S. Taniewska-Osinska and B. Nowicka, unpublished data. Trans. I, in press. Paper 71001 501 ; Received 5th October, 1987

ISSN:0300-9599    

DOI:10.1039/F19898500479

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1 , 1989, 85(3), 485492 A Kirkwood-Buff Theoretical Approach to Debye-Huckel Theory Interpretation of Electrolyte Activity Coefficients in Both Dilute and Concentrated Solutions Kenneth E. Newman? De'partement de Chimie, Faculte' des Sciences, Universite' de Sherbrooke, Sherhrooke, Que'bec, Canada J I K 2RI A set of equations is developed for the thermodynamic properties of 1 : I electrolytes using as its starting point the non-linearized version of the Poisson-Boltzmann radial distribution function and Kirkwood-Buff theory to forge the connection with the salt chemical potential. It is shown that the Debye-Huckel (DH) limiting law arises not from the linear term in the exponential expansion of the distribution but from the square term. The extended DH equation arises as a direct consequence of the treatment, although the interpretation of the distance of closest approach term is slightly different. By including only two parameters, one representing the distance of closest approach and a constant to represent ion-solvent interactions, excluded-volume effects and/or higher-order ion-ion inter- actions, the theory is able to model the activity coefficients of aqueous KCl accurately from infinite dilution to 4 mol kg-'.The theory is compared with other approaches. The thermodynamic properties (in particular the activity coefficient) of electrolyte solutions have been a major preoccupation of physical chemists through the years. The pioneering work of Debye and Huckel (DH)l provided a quantitative theory of their behaviour in the limit of very low concentration, but attempts to extend the theory have met with varied success.The statistical-mechanical shortcomings2 of Debye-Huckel theory are known not to affect the correctness of the limiting law, but they cause major problems for the higher-order terms. However, very major advances have been made through the years, especially in the work of M a ~ e r , ~ Kirkwood and Poirier,* Fried- man5 and Pitzer.' It is the purpose of this paper to explore a slightly different direction, namely the use of Kirkwood-Buff (KB) theory,' to forge the link between the ion atmosphere (ion-ion distribution) and the thermodynamic properties. This procedure was suggested in the original KB work but has not, to the author's knowledge, been explored in any depth.In this paper we will show that the Debye-Hiickel limiting law can be readily obtained from the ion distribution function without the need to invoke sometimes problematical charging cycles. However, the limiting law arises not from the first term in the linearization of the exponential of the distribution function but from the square term. In addition, an expression similar to that of the extended Debye-Huckel treatment arises as a direct consequence of the treatment, although the detailed form of the expressions are somewhat different. The present treatment also allows a great insight into the extension to concentrated solution in a logical and yet very simple way, and yields an expression capable of fitting the activity coefficients of KCl accurately up to saturation with only two parameters (apart from the DH slope).t Present address : Chemistry Department, The King's College, 107697th Street, Edmonton, Alberta, Canada T5H 2MI. 48 5486 Kirkwood-Buf Approach to Debye-Huckel Theory The Debye-Huckel Limiting Law The development of Debye-Huckel theory is so well known that it will not be repeated here. Following Pitzer,* we write for the radial distribution function gJr) of ionj around ion i for a 1 : 1 electrolyte MX ( i , j = M, X) (after conversion to rationized S1 units) sij(r) = exp (- 4ij) (1) with & 5 e2 exp (Ica) exp ( - K T ) 4.. = l3 4 n ~ , ~ , k , T(l +rca)r and K 2 = 2e2n/&, E, k , T (3) with n the salt number density. Eqn (2) is valid for all values of r > a, the distance of closest approach. Expanding out the exponential we obtain g&) = 1 - qii + &/2 ! - &/3 ! + .. . . (4) The Kirkwood-Buff treatment' involves integrals of the form G,, = lom 4nr2 [gkl(r) - 11 dr where the subscripts kl include all species in solution. For the terms Gii and qj it is convenient to perform the above integral in two parts, one between a and infinity, where we can use the expansion (4), the other between zero and a, over which gii(r) and gij(r) are zero. Including only the linear term in qij ( 6 ) Cj = - 0.54 q / n - where equation is in accord with the electroneutrality conditions :9 = 4na3/3, herein referred to as the excluded volume. It is clear that this GMM = Gxx = 1 /n + GMx (7 a) providing the distance of closest approach is the same for cation-cation, cation-anion and anion-anion interactions.For completeness we include the other electroneutrality condition :' Gfw = Gxw (7 b) with subscript W implying solvent. S in a solvent W is given by The Kirkwood-Buff expression for the chemical potential of a non-electrolyte solute It is not immediately obvious how to extend eqn (8) to an electrolyte solution. Initial attempts including treating the system as a ternary mixture (two solutes M and X in solvent W) and constraining M and X to have equal concentrations did not yield sensible results. We note that the final expression for the differential of the salt chemical potential (divided by k , T ) must have as its leading term 2/12 and must also yield the Debye-Huckel limiting law for the activity coefficient.Returning now to the integral ( 5 ) of eqn ( 4 ) it is clear that (apart from the first term) only the even powers of qij should be included, in order to preserve electroneutrality. For convenience we shall write GMM = -0.5/n + GLM - K. (9) The G/MM term includes only the integral of the even terms. If we assume eqn ( 8 ) is a correct representation of the chemical potential of an electrolyte solute and we simplyK . E. Newman 487 identify terms in SW and SS with those for MW and MM we do in fact obtain an equation of the correct form.? Thus if we substitute eqn (9) into (8), we obtain after some simplification We should note at this stage that in the conventional Debye-Huckel ion-charging treatments only the linear term in eqn (4) is included so as to obtain the limiting law. By contrast, in this treatment, which is formally correct from a statistical-mechanical viewpoint, the inclusion of only the linear term implies zero for GLM, i.e.ideal behaviour [eqn (lo)] ! As we shall now show the DH limiting law arises from the term in &. Thus we write S iraigh tforward evaluation of this integral gives At this point it is convenient to change from a number density (n) to a molar concentration scale (c), and thus we obtain for the mean ionic activity coefficient of the (13) salt, Y , (!*) = 2(GMW - GLM + K) T , p +2c(GbM- K-GMW) with the dimensions of qj now being equal to volume per amount of substance. GhM is now given by GIMM = A / [ ~ c ’ . ~ ( 1 + B~cO.~)~] (14) where A and B are the two Debye-Huckel parameters [ A = 1.76 (mol dm-3)-0.5; B = 0.329 x 1O1O m-’ (mol dm-3)-0.5 for water at 298.15 K and 101 325 Pa].Substitution of eqn (14) into eqn (13) readily yields the DH limiting law. The Kirkwood-Buff approach is thus able to forge the link between the ion distribution derived in the Debye-Hiickel treatment and the solute chemical potential without recourse to the inherently problematical charging cycles. The surprise in the approach is that the linearization of the distribution function [eqn (l)] does not yield the limiting law; the term in qii arises simply as a thermodynamic requirement related to electroneutrality constraints. The non-ideality arises from the square term, and we note in passing that Card and Valleau have reported that this distribution is closely similar to that calculated by Monte Carlo methods for the restricted primitive modello even up to relatively high concentrations (0.425 mol dm-3).Non-limiting Behaviour It is clear from eqn (10) that the thermodynamic properties of the solute are determined by both solute-solute and solute-solvent terms. It is shown in the Appendix that in the limit of infinite dilution the term GW is simply given by K k B T - pM,/2, where K is the solvent isothermal compressibility and pMx is the solute partial molar volume at infinite dilution. We may recall that, with the exception of the salts of certain large organic ions, electrolyte partial molar volumes rarely exceed the range 0.04 dm3 mol-l. In addition, for such ions we do not expect the term to exceed 0.08 dm3 mol-l, and thus we may t Such an approach does not of course imply that only cation-interactions are important.The electroneutrality conditions (7 a) and (7 b) require that cation and anion interactions are inexorably related.488 Kirkwood-Buf Approach to Debye-Huckel Theory conclude up to 0.1 mol dm-3 the ion-solvent terms in eqn (1 3) are not important. Thus as our next level of approximation we shall neglect both GMw and in eqn (13), which may noy be integratTd directly after substitution of eqn (14) with the approximation (1 + BUC~)' = 1 + ~ B u c ~ : lny, = - A ln[l+(2Bu+A/2)~'.~]. - 2Ba+A/2 If we expand out the logarithmic term on the right-hand side of eqn (1 5 ) including the first two terms, we obtain {noting that [ 1 - (Ba + A / ~ ) c ' .~ ] = 1 /[ 1 + (Ba + A / ~ c ' . ~ ] in the limit of small c> In y, - = - Ac'.~/[ I + (Bu + A/4) c ' . ~ ] . This equation should be compared with the conventional extended Debye-Hiickel equation : In y, - = - Ac'.~/( l + Buc'.~). (17) Eqn (1 6) and (1 7) clearly h$ve the same functional form, although the derived values of a from eqn (16) are 0.89 A smaller than from eqn (1 7). We note that these smaller values are more closely similar to crystal radii than the original values; however, both sets of values are anomalous in that they are not ion-additive. Eqn (15) is clearly less restrictive than eqn (16), although it is in a less convenient form. However, it is clear that even (16) is only formally valid up to 0.1 mol kg-l or less owing to the approximate expansion of (I + I C U ) ~ .Extension to Higher Concentrations Extension of the treatment to higher concentrations follows immediately from the above ideas if we accept that we are no longer able to obtain an analytical expression for the integral of eqn (13). As our next level of approximation we shall assume that the term (GMw - GLM + K) in eqn (13) is given by a constant C minus the expression given in eqn (14). This could imply that GdM is simply given by eqn (14) with the constant C equal to (GW+ K), or it could imply that some part of the constant term should be identified with the GLM term. We do not currently know which is the correct explanation, although evaluation of a and C for a whole variety of salts may enable us to resolve the matter. At this stage we simply treat C as an adjustable parameter.Thus the full expression for the activity coefficient becomes with GLM given by eqn (1 4). The numerical integration is very straightforward ; the ready availability of microcomputing facilities should help minimize this complication. Comparison with Experiment and with the Other Approaches It is not the goal of this paper to perform a detailed analysis of all available activity coefficients for 1: 1 electrolytes. Rather, we wish to demonstrate the method by comparison of observed values for KC1 with those obtained using some of the differeat equations developed in this work. Table 1 summarizes the results. The value of 2.95 A was chosen by optimizing the dilute solution data (where the value of C is not critical) using a numerical integration of eqn (18).The value of C was then optimized so as to give the best visual fit to all the data. Such a crude analysis does not of course permit a reliable estimate of the uncertainties in the parameters. However, it is clear that the equation does represent the overall shape of the activity coefficient data remarkably well. The equation slightly overestimates the depth of the minimum in the data, which implies that C may not be truly a constant but may diminish slowly with concentration. NeitherK. E. Newman 489 Table 1. Comparison of observed and calculated activity coefficients for KCl in water at 298.15 K rn/mol kg-' c/mol dm-3 y*(obsd)a eqn (18)b eqn (18)' eqn (15)d eqn (16)d 0.00 1 0.002 0.01 0.02 0.05 0.1 0.2 0.5 1 .o 2.0 3.0 4.0 0.000 997 0.001 994 0.009 968 0.019 93 0.049 79 0.099 43 0.198 3 0.491 4 0.968 9 1.883 2.742 3.55 1 0.964 8 0.95 1 0.90 1 0.868 0.817 0.771 0.724 0.660 0.620 0.606 0.620 0.647 0.965 0 0.952 0.902 0.870 0.8 17 0.770 0.721 0.659 0.624 0.61 I 0.623 0.648 0.964 9 0.951 0.901 0.868 0.814 0.765 0.71 1 0.637 0.583 0.535 0.510 0.494 0.964 9 0.95 1 0.901 0.868 0.812 0.761 0.704 0.622 0.559 0.498 0.465 0.443 0.964 9 0.952 0.901 0.869 0.8 15 0.767 0.715 0.646 0.597 0.554 0.533 0.519 0 a Calculate$ from data in ref.(1 1). a = 2.95 A; C = 43 cm3 mol-'. a = 2.95 A ; C = 0. a = 2.95 A. the simplified eqn (1 5 ) or ( I 6) which neglect C nor the full eqn ( I 8) with the value of C set to zero are able to reproduce the data accurately much beyond 0.05 mol kg-l.It would at this stage be foolhardy to suggest that a two-parameter equation would suffice to characterize the activity coefficient data of electrolytes in general. However, the variations of the term d In y , /dc with concentration for a whole variety of I : I , 2 : 1 and 2: 2 electrolytes12 all show-a remarkably similar shape simply scaled by the different values of the Debye-Huckel slope consistent with the simple picture represented by eqn (18). We are thus confident that the approach discussed in this paper will be able to characterize electrolyte activity coefficients in a manner much simpler than hitherto available. To date most of the distribution function theories used in electrolyte solution work have been 'integral equation' theories and involve the use of the Ornstein-Zernike equation to define a direct correlation function followed by the use of different closure approximations to solve this equation and thus obtain the variation of ion-ion radial distribution functions with distance.The approximations used are generally the hypernetted-chain (HNC), Percus-Yevick or mean spherical (MSA). Such studies have been well reviewed elsewhere,2.13 but it is probably useful to make a brief comparison of these approaches with the work presented here. All of the approximations require the use of the intermolecular potential. In addition, use is made McMillan-Mayer (MM) theory such that the derived thermodynamic parameters are functions only of the effective ion-ion pair potential; this is achieved by considering the solution in osmotic equilibrium with the pure solvent, and is often referred to as an MM standard state, as opposed to the normal or Lewis-Randall (LR) standard state.Conversion of experimental data from the LR to the MM standard state can be made providing density and compressibility data are available. Once the radial distribution has been calculated then the thermodynamic parameters are calculated by means of the compressibility, the volume or the energy equations. The latter two equations require again the intermolecular potential. The relation between MM standard states and Kirkwood-Buff theory is readily apparent if we consider the KB formulation developed by Hall" who showed that (19) RTd In c, = (1 + c, G5s) dps + c, G,, dpw.490 Kirkwood-Bufl Approach to Debye-Hiickel Theory If dpW is eliminated by use of the Gibb-Duhem equation, one readily obtains eqn (8), which is in the LR standard rate.If we note that under a MM standard state ,uw is constant, then the resultant equation for chemical potential is identical to eqn (8) but with the term Gw set to zero. This MM equation is in fact identical with the compressibility equation referred to above. In summary, the integral equation approaches allow calculation, from the intermolecular potential, of the radial distribution as a function of distance and hence allow evaluation of the thermodynamic properties for a MM standard state. The approach is ideally suited for testing simple ion-ion intermolecular potentials to see how well they are able to produce the thermodynamic parameters of electrolyte solutions and as such it has had a profound impact on our understanding of such solutions.The Kirkwood-Buff approach outlined above does not require the intermolecular potential, and by allowing explicit treatment of solute-solvent interactions can be used with either MM or LR standard states. Much of the discussion about the interpretation of electrolyte activity coefficients through the years has centred on the role of ion-solvent interactions. It would seem preferable that discussion, within the framework of distribution function theories, of the effect of ion hydration be formulated in a LR rather than MM standard state, and as such Kirkwood-Buff theory is ideally suited. A limitation of the theory is that only the radial distribution function integrals Gij are accessible, The major utility of the approach probably lies in its ability to commence with experimental thermodynamic data for real electrolyte solutions and to derive structural information. As such the integral equation and the Kirkwood-Buff approach would appear complementary.There is, however, clearly much fruitful work that could be performed to forge a stronger connection between the two. It is perhaps useful also to make a short comparison of the present treatment with the currently very popular approach of Pitzer.’ This latter approach, like the present work, starts with the radial distribution function as derived through DH theory. Pitzer then obtains a connection with the thermodynamic properties via the ‘pressure’ equation of statistical mechanics.Such a procedure allows him to show that the virial term [the term linear in concentration which is generally added to equation of the form of eqn (1 5)] is in fact a function of ionic strength. He then writes out plausible expressions for the excess free energy as the sum of long-range electrostatic terms, pair and triplet interactions. After including a reasonable expression for the variation of the pair interaction with ionic strength and assuming that the triplet interaction is constant, final equations are adopted on the basis of ‘empirical effectiveness’.6 This treatment which is formulated in the molality concentration scale has been extensively tested with both single and mixed electrolytes of various charge types and fits the available data remarkably well.In the data analysis, the distance of closest approach was fixed [the value of Ba in an equation similar to eqn (15) was set at 1.2 (mol kg-1)-0.5]. The three parameters so optimized were the pair interaction at low ionic strength, the pair interaction at high ionic strength and the triplet interaction. (The changeover from low ionic strength to high ionic strength pair interaction also involves certain additional parameterization.) In the present treatment, Kirkwood-Buff theory (which is formally correct) was used to make the connection between the DH radial distribution function and the thermodynamic properties. Straightforward evaluation of the equations yields the final expressions. The treatment is formulated in the molar concentration scale, and the final results occur in differential form, which needs to be integrated numerically.The differential nature of the equations, although perhaps at times inconvenient, does hold certain advantages. Much of the available literature on electrolytes is in the form of osmotic coefficients, i.e. solvent vapour pressures, and the differential nature of the theory allows ready conversion between solute and solvent activities uia the Gibbs-Duhem equation. Should analytical expression be required they can always beK. E. Newman 49 1 obtained by suitable Taylor series zxpansiccs of eqn (1 6 ) [after substitution of (2)J so as to obtain suitably integrable power series. The advantage of such an approach is that one is always aware of the limits of applicability of the expansion.In conclusion, we suggest that the approach developed above holds considerable promise for the analysis of the thermodynamic properties of electrolyte solutions. The work fully supports the idea expressed by Card and ValleaulO and by Pitzer6P8 that the non-linearized Debye-Hiickel equation for the ion-ion radial distribution function is a good function, our approach simply demonstrates an accurate way of obtaining the thermodynamic properties from the distribution. Questions still in need of resolution include the ability of the theory to fit data for all 1 : 1 electrolytes, the extension- to both higher charge types and to mixed salts, extension of the theory to temperature derivatives, e.g. enthalpies and specific heats, as well as interpretation of the value of the constant C and related questions of ion additivity. Work is continuing on several of these areas.Appendix According to Kirkwood-Buff theory, the expression for the molar volume of a solute S in solvent W is (after conversion from number density to molar concentration) As before, we shall assume that this equation is also correct for a 1 : 1 electrolyte solution if we identify V, with the salt partial molar volume VMx and we identify the SW and SS terms with those for MW and MM. Following Hall13 we note that in very dilute solution the distribution of W around M is due to its direct environment plus the solvent distribution around the ions which make up its ion atmosphere. Thus we may write for example G,, = G& + c q , G:w + c q i GT, where G:w is the integral of the radial distribution function of W around i (i,j = M or X) with the upper limit so chosen to exclude all solvent associated with the ion atmosphere.According to eqn (9) and (12), in the limit of infinite dilution cG,, = -0.5 and c q j = 0.5, and we thus obtain from eqn (A 1) and (A 2) (A 3) PMx = lim (Y,,) = 2 / ~ , + 2Gww - (GLW + Giw). c-0 Substituting the Kirkwood-Buff expression for the compressibility of the pure solvent we obtain14 (A 4) PMx = 2 ~ k , T - (GLw + G?&) = 2 ~ k , T - 2GMw which has the required ion-additivity properties for an electrolyte partial molar volume. References 1 P. Debye and E. Huckel, Phys. Z . , 1923, 24, 185; 334; 1925, 25, 97. 2 H. C. Anderson, in Modern Aspects of Electrochemistry, ed. J. O’M. Bockris and B. E. Conway (Plenum, New York, 1975), vol. 1 1 , p. 1 . 3 J. E. Mayer, J . Chem. Phys., 1950, 18, 1426. 4 J. G. Poirier and J. G. Kirkwood, J. Phys. Chem., 1954, 58, 591. 5 H. L. Friedman, Ionic Solution Theory (Wiley Interscience, New York, 1962). 6 K. S. Pitzer, J . Phys. Chem., 1973, 77, 268; K. S. Pitzer and G. Mayorga, J . Phys. Chem., 1973, 77, 7 J. G. Kirkwood and F. P. Buff, J. Chem. Phys., 1951, 19, 774. 8 K. S. Pitzer, Acc. Chem. Res., 1977, 10, 371. 2300; K. S. Pitzer and G. Mayorga, J. Soln Chem., 1974, 3, 539.Kirkwood-Buf Approach to Debye-Huckel Theory 9 J. L. Beeby, J. Phys. C, 1973, 6, 2262. 10 D. N. Card and J. P. Valleau, J . Chem. Phys., 1970, 52, 6232. 1 1 C. B. Monk, Electrolytic Dissociation (Academic Press, London, 1961), p. 33. 12 Y. Cheng and K. E. Newman, unpublished work. 13 J. Enderby and G. W. Neilson, Rep. Progr. Phys., 1981, 44, 38. 14 D. G. Hall, Trans. Faraday SOC., 1971, 67, 2516. 15 D. G. Hall, J. Chem. SOC., Faraday Trans. 2, 1972, 68, 25. 16 K. E. Newman, J. Chem. Soc., Faraday Trans. I , 1988, 84, 1387. Paper 7/001811; Receiued 9th November, 1987

ISSN:0300-9599    

DOI:10.1039/F19898500485

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

J. Chern. SOC., Faraday Trans. I, 1989, 85(3), 493-501 X-Ray Diffraction and Raman Spectral Analysis of Molten CdCl, Yoshiki Takagi,* Noriko Itoh? and Tetsur6 Nakamura Research Laboratory of Engineering Materials, Tokyo Institute of Technology, Nagatsuta-cho 4259, Midori-ku, Yokohama 227, Japan The structure of molten CdCl, has been investigated by X-ray diffraction and Raman spectral analysis. The tetrahedral configuration (CdCI,) was confirmed from the first and second peaks of the radial distribution function. Crystalline MCl,(M = Mg, Mn, Fe, Co, Ni and Cd) has a CdC1,-type layer structure.' In which C1 atoms are close-packed f.c.c., a layer of M atoms alternates with every two Cl layers and every M atom has six nearest-neighbour CI atoms in octahedral coordination. For molten MCI,-ACl systems where A is an alkali metal, many workers have suggested that tetrahedral complex MCIi- ions exist predominantly in ACI-rich In molten MgCI,, MnCl,, FeCI,6 and COCI,~, MCI, in a tetrahedral configuration has been suggested to occur, as shown by the enthalpy of mixing.In this paper the radial distribution function (r.d.f.) of molten CdCI, is obtained, and a structural model of the molten state is proposed by considering the geometrical orientation of the ions. The coordination states of molten CdCI, have also been studied by Raman spectroscopy. To date, the Cd-Cl stretching frequencies in the crystalline, molten and solution states have been reported. However, as shown in table 1, various assignments have been made. For instance, Bues' made an assignment on the basis of a 4-coordinated Cd2+ ion, while Clarke et al.' based their assignment and 6-coordination state by reference to previous data.'" Both assignments are reasonable from the viewpoint of selection rules.Ex per imen tal The experimental procedures and analysis of the observed intensities are identical to those described previously. 11, l2 Cadmium chloride (99.9 O/O, Rare Metallic Co.) was dried in vacuo at 473 K for a few days. The anhydrous cadmium chloride was melted at 923 K in a silica glass tube, purified by bubbling dry HCI over it for 1 h and then sealed in a silica glass tube (fig. 1) following evacuation. The X-ray measurements were made using parafocusing reflection geometry. Mo K2 radiation was used and the beam was monochromatized by use of a curved graphite single crystal mounted in the diffracted beam.The correction for background, polarization, absorption and Compton scattering were applied to the observed data and were scaled to the independent scattering factor for the stoichiometric unit using the method of Krogh-Moe13 and N ~ r m a n . ' ~ The radial distribution function D(r), average correlation function G(r) and reduced intensity function Si(S) were calculated using methods reported previously. l l . l2 Parameters used in the calculation are given in table 2. Japan. t Present address: Tokyo Metropolitan Institute of Technology, 6-6 Asahigaoka, Hino, Tokyo 191, 493Table 1. Raman shifts of CdCl, solid molten salt Raman shift/cm-' T/"C assignment Raman shift / cm-I T/ "C assignment ref. no.2 ref. no. $ 242 hexagonal 2 % 1 $ 213 20 A , of tetragonal CdCIt- 1 212+5 580-650 A, of tetragonal CdCli- in layer lattice 3 - - 3 205 & 3 233 & 3 - 23 5 25 } A , , of triply shared CdC1;- 4 215 > 600 essentially CdC1;- 4 2 z 3 229" 580" associated with low net coordinated Cd2+ in disordered ionic melt B 239 room - this work 229 580, - this work K. temp. 760 & 5 Tanemo tob 210-217 580, 605 a Not observed but extrapolated to 580 "C. A member of staff in our laboratory.Y. Takagi, N . Itoh and T. Nakamura h diffracted beam Fig. 1. X-Ray diffraction cell for volatile sample. Table 2. Parameters used in the calculation of the r.d.f. temperature/K 923 DJstoichiornetric unit A-3 0.011 067 effective electron number : K C d 48.580 0 16.810 0 12.00 K, I Srnax/A-l 495 incident beam The Raman spectra were recorded on a JEOL spectrometer with 514.5 nm excitation '(ca.300 mW) by an argon-ion laser (Spectra Physics). Results and Discussion X-Ray Diffraction The correlation function G(r) and the radial distribution function D(r) are shown in fig. 2. The function D(r)/r is also shown in fig. 2. The first peak in D(r) in fig. 2 evidently corresponds to, the nearest-neighbour interaction of Cd-C1 pairs. The nearest Cd-Cl distance, 2.42 A, in the molten state is slightly shorter than that in the crystalline state. The number of chlorine atoms around each cadmium species wa; calculated from the first peak assuming a Gaussian distribution centred at 2.42 A. The observed number of chlorine atoms which are nearest to cadmium is 3.9, and decreases markedly on melting in comparison with the number for the crystalline state, which is 6.0.The second and third peaks of the correlation function G(r) are due to the contribution of the nearest Cl-Cl and Cd-Cd ion pairs, respectively. The numbers and distances of these ion pairs obtained by analysis of the radial distribution function are listed in table 3. Raman Spectra We observed a single polarized band at 239 cm-l for the solid CdCl, (fig. 3). This Raman shift is in good agreement with the previous values of 242,15 and 235 cm-'.' When the CdCI, was heated to 853 and 1033 K, a single polarized band was observed at 229 cm-l (fig. 3). This Raman shift is apparently different from the previous values~9*1s of ca.212 cm-', rather it is same as the Raman shifts of solid CdCl,.g.16 A different Raman spectrum was obtained in our laboratory, although the experimental procedure was the496 Structure of Molten CdC1, Fig. 2. Radial 9 8 7 5 6 4 * 5 2 4 3 2 1 0 2 n A 1 u 0 distribution 1 2 3 4 6 6 7 8 9 1 0 l I A I I I I I I I I 4 3 4 5 6 7 8 9 1 0 r/A function D(r), correlation function G(r) and function CdCl, at 923 K. Table 3. Average distance,, ,,/A, root mean square displacement, ( Arij)H, and coordination number, n,,, for i-j pairs i-j pair T i , <Ari,>: nij Cd-Cl 2.42 0.055 3.90 Cd-Cd 4.88 1.32 3.6 Cl-Cl 3.80 0.294 10.88 10.9 Cd-Cl 5.56 - 1 I I I I I I I l I I I (d n 229 (4 239 P 400 300 200 100 400 300 200 100 400 300 200 100 wavenumber/cm-' of molten Fig. 3. The normal (n) and polarized (p) Raman spectra of CdCl,: (a) solid, (b) 853 K and (c) 1038 K.Y.Takagi, N . Itoh and T. Nakamura 497 Table 4. The observed frequencies, v, of the local vibration of tetrahedral CdC1,2- molten salt solution selection ~ mode rule vlcrn-l state ref. vlcm-' state ref. a, R," pol. 256f3 CdCI,..uCsCI 4 (x = 2,3,4.3) 253+3 CdCl,.xKCl 4 (-u = 2,3,4.3) 250 & 2 CdCI, . xLiC1 4 (x = 2,3,4.3) 259 CdCI,..yKCI 6 (x = 1,2) f, R, depol. - - - 264 (Et,N),CdCI, in 13 HCI 26 1 (Et,N),CdCI, in 13 CH,NO, 260 (Et,N),CdCI, in 13 CH,,CN 258 (Et,N),CdCI,, 13 tetra-n- butyl phosphate 28 I (Et,N),CdCl,, in 13 CH,CN (Et,N),CdCI,, 13 tetra-n-butyl _______- ~~ - .- - ' I Raman. Table 5. GF matrix for tetrahedral CdCIi- 1 I G(a,) = -(1 +3cos 109.4")+- m(.d MCI I I G(f,) = - (1 - cos 109.4") + - mcd %I '(a,) = If, + ? f r , l m, is the mass of atom X.same. The Raman shift observed at 2 10-2 17 cm-I by Tanemoto agreed with the reported l6 This suggests that the coordination state is not so simple nor so regular at long range. The Raman shifts due to CdC1;- local structure are 260 cm-' (a,) and 281 cm-' (f2),9917*18 as shown in table 4. The existence of the CdC1;- ion is supported also by measurements of electrical conductivity," viscosity2" and surface tension.'' Table 4 shows that the a, frequencies are the same, although the circumstances are changed from the molten state to the various organic solutions. This means that the force constants obtained from the CdCIi- unit have universal values for various CdCl:-,-" unit structures, if the bond length is same.Thus the force constants of CdCIt- were determined by the GF method22 using the generalized force field as given by V is the potential energy, f, and f,, are the force constants and Ar is the change in the Cd-CI bond distance from the equilibrium position. The block matrices of G and F are498 Structure of Molten CdCI, Table 6. The stretching frequencies of CdC12,-" (n = 2, 3 and 6) ~ observed calculated unit point state of ref. structure group mode the sample no. vlcrn-' vlcm-' CdCl, D,, v,(Ci) TBP"/KCI s o h 13 280 248 CdCI, D,, v,(ai) TBP"/KCl s o h 13 265 254 v,(e') TBP"/KCI s o h 13 287 - ;E; ( i f bond angle - c,, Vlb,) - v3(e) - 281 = 110" - - CdC1;- 0, v,(A,,) solid 2,4, this work 237 + 2 265 a Extract from KCI aqueous solution to the tri-n-butyl phosphate soh tion.Fig. 4. (a) and (b) Corner-shared and (c) edge-shared models of bitetrahedra. shown in table 5. Thus f, = 1.225 x 10, N m-' and f,, = 0.0624 x lo2 N m-' were obtained. The value of f, + 3f,, ( = 1.412 x 10, N m-') corresponds to f = 47t2cm,, v: (1.41 x 10, N m-') calculated by Davies and Long.ls The values off, andf,, of CdCIi- were used to calculate the stretching frequencies of possible unit structures such as CdCl,, CdCl; and CdCli-. The v1 and v3 frequencies calculated for planar or bent CdCI, agreed with the observed values, as seen in table 6. However, v1 frequencies calculated for the respective CdCl, and CdCli- species did not agree with the observed values. This is because the Cd-CI bond distance is unchanged in the case of CdCl,, but becomes short in the case of CdCl,, and becomes long in the case of CdCIt-, compared with CdCli-.If we assume that molten CdC1, contains all or some species of CdCI,, CdCl;, CdC1,2- and CdC1,4- unit structures, the peak frequency of the Raman band synthesized by superposing the Raman bands of the unit structures is at least 237 cm-l. The peak frequency of 213 or 229 cm-' in table 3 cannot be reproduced by such a superposition.Y. Takagi, N . Itoh and T. Nakamura 12 0 140 160 180 a/" Fig. 5.fb values plotted against the bent angle a used in the present calculations. Table 7. Description of the normal coordinates (normalizing factors omitted) 499 - Q , Q2 Q3 Q4 Q5 QG Q, - Qa - 1 1 1 1 1 1 1 1 - 1 - 1 - 1 3 - 1 - 1 - 1 3 1 1 1 1 - 1 - 1 - 1 - 1 - 1 - 1 - 1 3 1 1 1 - 3 - 1 - 1 2 0 - 1 - 1 2 0 1 - 1 0 0 1 - 1 0 0 - 1 - 1 2 0 1 1 - 2 0 - 1 - 1 0 0 - 1 1 0 0 The isolated CdC12,-" unit structure exists only in systems containing two or more components. In the glassy state the calculation should be carried out for the bridged structures. It seems difficult for the octahedral unit structure to have various bridge angles.Rather, the tetrahedral unit structure is more suited to have flexible bridge angles. Therefore the stretching frequencies were calculated using a CdC1,-Cl-CdCl, model. A generalized configuration is illustrated in fig. 4. The CdCli- at one side retains symmetry and the bridge angle between bonds rs and rs, a, has values ranging from 109.4 to 180". If a = 180°, CdCl, units at both sides are related by an eclipsed or staggered configuration.The potential energy, 5, is given by (2) & = v, + fb Ar8 Ars. We have already determinedf, = 1.225 x 10, andf,, = 0.0624 x 10, N m-', but not the value of fb. Thus an assumption was introduced that fb =A, if a = 109.4", fb = 0 if a = 180°, and fb varies with a as shown in fig. 5. The normal coordinates are given in table 7. The calculated frequencies are shown in table 8. The same values were obtained from the staggered and eclipsed configurations. As the Q, and the Q, modes are totally symmetric, the Raman-scattering intensity is most strong among modes Q,-Q,. The value of v, is independent of a, but v2 shifts to lower frequencies as the bridge angle becomes wider. If the v, and Y , bands overlap and give rise to a single band, the peak frequencies appear between 270 and 132 cm-'.For example, the Raman shifts observedv1 0 0 Table 8. Symmetry of the modes and the calculated frequencies, vralrd(crn-') of Cd,CI;- ~~ Cd-C1-Cd bridge conformations Q, Q 2 Q3 Q4 Q5 QG Q, Qfl linear D3d (staggered) (eclipsed) Dm CS a = 109.4" 'calrd 120" 130" 140" 150" 160" 170" A', A, R 267 R, i.r. 270 268 268 267 267 267 267 4 A, R 132 R, i.r. 239 213 196 I69 159 142 130 E" E" 1.r. i.r. A," p2" i.r. 1.r. I?. A," A," E' E' E" E" 1.r. i.r. R, i.r. R, i.r. R R 5 % % B Z B, A2 inactive R, i.r. R 4 B, A, cp 266 358 28 I 28 1 28 1 28 1 R, i.r. R, i.r. R, i.r. 264 302 28 1 28 1 all vibrations are R and i.r. active $ 265 325 28 1 28 1 B 265 332 28 1 28 1 p 3 265 357 28 1 28 1 265 350 28 1 28 1 266 363 28 1 28 1 266 373 28 1 28 1Y.Takagi, N . Itoh and T. Nakamura 50 1 at 213 and 229 cm-' coincided with the calculated frequencies: if the oscillator strengths of modes Q, and Q, are equal, ( v , + v,)/2 = 21 3 cm-' for a = 150" and (v, + vJ/2 = 232 cm-' for a = 130", respectively. On the other hand, the depolarized Raman shift observed at 281 cm-I for CdC1,2- coincided with the calculated frequency v 5 , v ~ , vi or v8. These were independent of the bridge angle and of the unit structure. The calculation of normal frequencies should also be made for the edge-shared model, CdCl,. Assuming that ( a ) the molecular symmetry is DZh, (h) the CdCl, CI,Cd unit remains a regular tetrahedron and (c) the Cd-CI bond distance is equal to that of the isolated CdC1;- unit, the frequencies of the totally symmetric modes (Alg) are 275 and 230 cm-l.As the Cd-Cl bond used for bridging is longer than the terminal Cd-CI bond, the A,, frequencies will become lower in wavenumber that 275 or 230 cm-'. Also, a simple expression for the potential energy such as eqn (2) seems insufficient to treat such a complex structure as the edge-shared model. The number of chlorine atoms around each cadmium atom was 3.9, and the ratio of the nearest Cl-C1 and nearest Cd-CI distances was close to the value 1.63 [i.e. (8/3)2]. Thus the existence of [CdCl,] tetrahedral units in the melt was confirmed by analysis of the radial distribution function. The result is consistent with the Raman spectral study reported in the present work./cl \ \c1/ The computations were carried out on an M-180 and VAX8600 computer at the Nagatsuta Branch of the Computer Centre of the Tokyo Institute of Technology. References 1 R. W. G. Wyckoff, Crystal Structures (Interscience, New York, 1960), vol. 1, p. 272. 2 K. Tanemoto and T. Nakamura, Chem. Lett., 1975, 4, 351. 3 A. S. Kucharski and S. N. Flengas, J. Elecrrochem. Soc., 1974, 121, 1298. 4 G. N. Papatheodorou and 0. J. Kleppa, J. Inorg. Nucl. Chem., 1971, 33, 1298. 5 B. R. Sundheim and M. Kukk, Discuss. Faraduy SOC., 1961, 32, 49. 6 G. N. Papatheodorou and 0. J. Kleppa, J. Inorg. Chem., 1971, 33, 1249. 7 W. Trzebiatowski and A. Kisza, Bull. Acad. Pol. Sci., Ser. Sci. Chim., 1961, 9. 605. 8 W. Bues, Z . Anorg. Allg. Chem., 1955, 279, 104. 9 J. H. R. Clarke, P. J. Hartley and Y. Kuroda, J . Phys. Chem., 1972, 76, 1831. 10 A. F. Wells, Structural Inorganic Chemisrry (Oxford University Press, London, 1962). 11 H. Ohno, K. Furukawa, K. Tanemoto, Y. Takagi and T. Nakamura, J . Chem. Sac., Furaduy Truns. I , 12 Y. Takagi, T. Nakamura, T. Sata, H. Ohno and K. Furukawa, J . Chem. Soc., Faraduy Trans. I, 1979, 13 J. Krogh-Moe, Acta Crystallogr., 1956, 9, 951. 14 N. Norman, Acta Crystallogr., 1957, 10, 370. 15 C . S. Vankateswaren, Proc. Indian Acad. Sci., 1935, A l , 850. 16 M. Tanaka, K. Balasubramanyan and J. O'M. Bockris, Electrochim. Acta, 1963, 8, 621. 17 U. A. Maroni and E. J. Hathway, Electrochim. Acfa, 1970, 15, 1837. 18 J. E. D. Davies and D. A. Long, J. Chem. Soc, A , 1968, 2054. 19 H. Bloom and E. Heymann, Proc. R. Soc. London, Ser. A , 1947, 188, 392. 20 B. S. Harrap and E. Heymann, Trans. Faraday Soc., 1955, 51, 268. 21 N. K. Boardman, A. R. Palmer and E. Heymann, Trans. Furuduy Soc., 1955, 51, 277. 22 S. Mizushima and T. Shimanouchi, Infrared Absorption and Raman Eflect (Kyoritsu, Tokyo, 1989), 1978, 74, 804. 75, 1161. p, 113. Paper 7/00078B; Received 15th December, 1987

ISSN:0300-9599    

DOI:10.1039/F19898500493

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1989, 85(3), 503-519 Photosensitised Oxidation of Water by CdS-based Suspensions Andrew Mills* and Geraint Williams Department of Chemistry, University College of Swansea, Singleton Park, Swansea SA2 8PP Dispersions of CdS powder with or without metal or metal oxide deposits, such as Pt, Au, Ag, Rh, Rh,03 and RuO,, have been used to sensitise the oxidation of water by PtCl:-. The most active of the CdS-based photosensitisers was found to be one in which Pt was deposited by precipitation of a Pt colloid onto the surface of a CdS sample which had been annealed in air for 3 h prior to platinisation. A study was made to determine the optimum conditions for 0, evolution. The initial rate of 0, evolution was found to depend upon a number of factors including: pH, [CdS/Pt], [PtClE-1, [O,], Pt loading and CdS annealing temperature and environment.Although a number of different electron acceptors were tried, including PtCIi-, PtCII-, Pt(OH)i-, %AuO;, S,Oi-, Co(NHJ,C12+ and Fe(CN)i-, 0, photogeneration was observed with only PtCIi- and Fe(CN)i-. A number of other materials were tested as photocatalysts for the oxidation of water by PtCli-, including TiO,, TiOJPt, Ti0,/Rh,03, CdO, CdO/Pt, HgS/Pt, Si0,/Rh,03 and Al,O,/Pt; however, only the Ti0,-based materials showed any activity. The results of this work are discussed with respect to the current controversy over the mechanism for 0, evolution. 1. Introduction Cadmium sulphide (CdS) is an n-type semiconductor which has attracted a great deal of attention from workers in the field of solar-energy conversion.' One reason for this attention is its low band gap (2.4 eV), which allows it to absorb an appreciable fraction of the solar spectrum (i.e.1 < 526 nm).2 Another reason is the strongly reducing nature of its conductance band electrons, e-, (- 0.9 V vs. standard calomel electrode, SCE) and the strongly oxidising nature of its valence-band holes, h+ (1.5 V us. SCE), both of which are generated simultaneously3 on absorption of a photon of light, 1 < 526 nm, by CdS. This combination of properties has encouraged many to attempt the photocleavage of water with visible light, using CdS as the photosensitiser, i.e. CdS hv 2 2.4 eV 2&0 - 2 4 + 02t. Although some groups have claimed succesP in their attempts, others have failed.'. The problem does not appear to be associated with the reduction side of the reaction, i.e.2e-+2H+ -+&T (2) since many group^^-^^ have used CdS, coupled to a H, catalyst, to efficiently sensitise this reaction in the presence of a sacrificial electron donor, such as EDTA,'-" TEOA,12 cysteine," sulphide13 or formic acid.'* The role of the sacrificial electron donor (D) is to react irreversibly with the photogenerated holes to prevent electron-hole recorn bina t ion, i.e. h+ + e- -+ heat (3) 503504 Photosensitised Oxidation of KO by CdS and encourage the reduction of water by the remaining photogenerated electrons, i.e. reaction (2). Essential to the overall efficiency of the photosystem is the inclusion of an active, stable H, catalyst, such as Pt or Cd metal or RuO,, to mediate reaction (2).'-14 In contrast to the photoreduction of water, the photo-oxidation of water, i.e.4h' + 2 h O --* 4H' + 0,t (4) when sensitised by CdS, is much more difficult and is generally considered to be the major obstacle preventing the efficient photocleavage of water sensitised by CdS.4-8 At the heart of the problem is that the photogenerated holes tend to react with the semiconductor, i.e. 2h+ + CdS + Cd2+ + S (5) or, in air-saturated water, hv > 2.4 eV CdS + 20, Cd2+ + SO:- rather than with water, via reaction (4). In order to prevent this it is necessary to have present an 0, catalyst to promote reaction (4) over reactions (5) and (6). Much of the current controversy which surrounds the research into CdS as a photocatalyst for the oxidation or dissociation of water centres on the existence of such an 0, ~ata1yst.l~ and photochemical systems''.l7 that sacrificial electron acceptors (A), such as hexachloroplatinate (PtC1,2-) or persulphate, allow systems capable of photosensitising the oxidation of water to 0, to be studied and optimised. Ideally, the role of the sacrificial electron acceptor is to react irreversibly with the photogenerated electrons to prevent the electron-hole back reaction, reaction (3), and encourage the oxidation of water by the remaining photogenerated holes, reaction (4). The use of such ' sacrificial ' electron acceptors in this way has become a popular method for testing the effectiveness of both new sensitisers and catalysts for the oxidation of water.In this paper we describe the results of a detailed study of the oxidation of water by PtC1:- and other sacrificial electron acceptors, sensitised by CdS with and without a number of different metal or metal oxide deposits which have been proved capable of mediating the oxidation of water. It has been demonstrated for both photoelectrochemicall~ 2. Experimental 2.1. Materials The main source of cadmium sulphide (purity > 99.99%) used throughout this work was Koch-Light Laboratories. However, samples of CdS from other sources were also tested, including those from Fluka (99.99 YO), Johnson Matthey (99.99 YO) and B.D.H. (98 O h ) . Chloroplatinic acid (KPtCg), rhodium trichloride (RhCl,), chloroauric acid (HAuCl,) and ruthenium trichloride (RuCl, - x&O) were obtained from Johnson Matthey. Chloroauric acid undergoes rapid hydrolysis to &AuO, under the highly alkaline conditions (pH 13) used throughout this work; thus we refer to the Au'" used in this work as &AuO,.The cobalt(II1) pentammine dichloride was synthesised as described elsewhere;" all other materials were obtained from B.D.H. in their purest available forms. The water used to make up solutions was always doubly distilled and deionised. 2.1. I. Photocatalyst Preparation Unless stated otherwise, platinisation of the CdS was carried out using the following procedure. Initially a very stable Pt/citrate colloid, black in colour, was prepared byA . Mills and G. Williams 505 refluxing for 4 h a solution containing 30 mg of chloroplatinic acid, 30 cm3 of a 1 O h sodium citrate solution and 120 cm3 of water.Transmission electron micrographs of the Pt colloid showed that the Pt particles formed clusters, with an average diameter of 29+4 nm. The average diameter of the Pt particles which made up these clusters appeared to be ca. 4 nm. Two thirds of the resultant Pt colloid (100 cm3) were stirred with 1 g of the CdS as 11.6 g of NaCl were added. Although destabilisation of the Pt sol followed by Pt precipitation appeared to be complete within seconds, the solution was left stirring for ca. 1 h before being filtered and washed thoroughly with distilled water to remove the NaCl and citrate from the CdS/Pt. A variation in the amount of Pt deposited on the 1 g of CdS was achieved by using different volumes of the Pt/citrate colloid.The method of depositing Pt onto CdS described above can also be used for depositing other materials provided they can be prepared in colloidal form,19 Thus it proved possible to prepare both Au and Ag deposited CdS samples by using a Au/citrate and a Ag/citrate colloid, respectively. In addition, using the method of platinisation described above, it proved possible to platinise supports other than CdS, including TiO,, A1,03, CdO and HgS. Samples of CdS/Rh,03,17 CdS/Ru0,17 and photoplatinised CdS (CdS/Pt*)9 were prepared as described in the literature. 2.2. Methods A dispersion (5 mg ~ m - ~ ) of the photocatalyst under test (30 cm3) was placed in a thermostatted (25 "C) quartz cell with an 0,-membrane polarographic detector (0,- MPD) in its base and stirred continuously.The details of the 0,-MPD have been given in a previous paper.20 Unless stated otherwise, prior to illumination the dispersion was saturated with nitrogen (B.O.C., white spot) for at least 15 min. Illumination of the dispersion in the cell was carried out using a 250 W xenon lamp (Applied Photophysics) whose U.V. output was removed by means of a 400 nm cut-off filter. In some of the prolonged irradiations, i.e. t > 1 h, the photogeneration of 0, was monitored via the headscape (volume = 17 cm3) using gas chromatography. The gas chromatograph used (Perkin-Elmer, model F33) was fitted with two 2 m stainless-steel columns (3 mm 0.d. and packed with 5A molecular sieve) and a thermistor-type thermal- conductivity detector.The carrier gas used was argon (B.O.C.). In some of the experiments atomic absorption spectroscopy (a.a.s.) was carried out, using a Perkin-Elmer Alpha 2 instrument, to monitor the levels of Cd2+ ions in solution before and after irradiation of a CdS/Pt dispersion. In this work an irradiated dispersion was acidified to pH 5 using lo-, mol dmP3 acetic acid and stirred for 2 h to dissolve any Cd(OH), generated. The particles were removed using a 0.2pm membrane filter (Schleicher and Schuell) incorporated into the syringe used to sample the dispersion. The level of Cd2+ ions in the filtrate was then determined using a.a.s. A blank experiment was carried out in which Cd(N0,) (5.1 x mol dm-3) was added to a solution (20 cm3) containing CdS/Pt (5 mg crnp3) and 0.1 mol dm-3 NaOH.This solution was stirred in the dark for 2 h and then analysed for Cd2+ ions as described above. After subtracting for the blank (i.e. [Cd2+] when no Cd(N03), was added) the concentration of Cd2+ was determined as 4.1 x lop5 mol dm-3; thus ca. 82 % of the Cd2+ added was detected by a.a.s. In addition, a.a.s. was used to monitor the amount of Pt deposited onto the surface of a CdS/Pt dispersion before and after irradiation. This was achieved by first filtering off, washing and drying in air the CdS/Pt sample under test. A part of this sample (typically 0.1 1 g) was then dissolved in 10 cm3 of a boiling aqua regia solution and the solution made up to 50cm3. The amount of Pt dissolved in solution, and therefore contained in the original CdS/Pt sample, was then determined by a.a.s.Electron microscopy coupled with energy-dispersive analysis (e.d.a.) was carried outPhotosensitised Oxidation of KO by CdS Fig. 1. Typical dissolved 0, concentration us. irradiation time observed for dispersions of the following CdS-based photosensitisers : (a) CdS/Pt (0.8 YO Pt), (b) CdS/Pt (photoplatinised sample, 0.6% Pt), ( c ) CdS and ( d ) no photosensitiser. The irradiations were carried out using light of R. > 400 nm with a solution (30 containing 150 mg of the photosensitiser under test, PtCIi- ( mol dm-3) and NaOH (0.1 mol dm-3). Table 1. Initial rates of 0, evolution for a variety of CdS-based photocatalysts" metal/metal oxide initial rate of O2 deposited evolution photocatalyst (Yo w/w) /pmol dmP3 CdS/RuO, CdS/Rh,03 CdS (Fluka)/Pt CdS (BDH, GPR grade)/ Pt CdS (Johnson Matt hey)/ Pt CdS/Pt CdS/Pt CdS/Pt CdS/Rh CdS/Au CdS/Ag CdS/Pt*b CdS none 1.2 1 .o 0.6 0.8 0.4 0.8 0.6 0.4 0.3 0.5 0.4 0.6 0 0 0.2 2.2 3.3 0.5 0.2 2.3 1.9 1.7 0.5 0.4 0.1 0.3 0.1 0 a Unless stated otherwise the CdS used was supplied by Koch-Light (puriss, > 99.999%). In this case the platinum was photodeposited onto the CdS.A .Mills and G. Williams 507 Fig. 2. Initial rate of 0, evolution us. wavelength of irradiation (Ak20 nm); all other reaction conditions were as described for fig. 1. The broken line is the absorption spectrum of a PtCIi- solution ( lo-, mol dmP3) recorded in a 1 cm cell. using a 120C TEM-SCAN instrument (JEOL Ltd) and was used to examine the CdS/Pt samples before and after irradiation.The electron micrographs recorded at 100 000 magnification showed that the CdS/Pt powder comprised CdS microcrystals (diameter ca. 1.2 pm) with a sparse but uniform covering of Pt particle clusters (average diameter ca. 29 +4 nm). Calculations show that the Pt clusters cover typically only 1.8 % of the surface area of any one CdS particle. Energy-dispersive analysis for Pt carried out on the CdS/Pt particles confirmed these observations. CdS/Pt particles from samples taken both before and after prolonged irradiation showed no evidence in their transmission micrographs of corrosion. X-Ray powder diffraction patterns recorded for the CdS before and after heat treatment (up to 400 "C) showed no other lines except those due to CdS. 3. Results 3.1. Different CdS-based Photocatalysts In a study of the photocatalytic activities of a number of different CdS-based powders, 150 mg of the photocatalyst under test were dispersed in 30 cm3 of a solution containing PtC1;- mol dm-3) and NaOH (0.1 mol dm-3).The solution was then purged with nitrogen and subsequently irradiated. Fig. 1 illustrates the variation in 0, concentration vs. time observed for several of the samples, and table 1 lists the measured initial rates of O2 generation. 3.2. Evolution sensitised by CdS/Pt Using the most active of the CdS-based photocatalysts, i.e. CdS/Pt; 0.8% w/w, the initial rate of O2 evolution was determined as a function of irradiation wavelength (Ak20 nm). The irradiations were performed using a 250 W xenon arc lamp coupled to a high-radiance monochromator, and the results are illustrated in fig.2. In another set of experiments the concentrations of Cd2+ ions were determined before and after a typical irradiation. The dissolved oxygen generated during this irradiation was determined as 2 x lo-' mol dm-3 and the concentration of Cd2+ ions was found to508 Photosensitised Oxidation of KO by CdS [CdS/Pt]/mg cM3 0 0.4 0.0 1.2 1.6 2 [PtCg -3 / 1 O-2mol dm-3 Fig. 3(a-c). For legend see facing page.A . Mills and G. Williams 509 100- 0 n 0- I I I 0.2 0.4 0.6 0.8 Pt (%) f I I 1 I 200 300 400 0' loo CdS baking temperature/"C Fig. 3. Relative rate of 0, evolution as a function of the following different reaction parameters :, (a) solution pH (b) [CdS/Pt] (in mg~m-~), (c) PtCli-] (in mol dmP3), ( d ) Pt loading on the CdS/Pt photocatalyst (in YO) and (e) temperature used to bake the CdS powder, in air for 3 h, prior to platinisation.All other reaction conditions were as described for fig. 1. be 7.8 x mol dm-3 before and after irradiation. The high background [Cd2+] (7.8 x mol dm-3) was most likely due to the CdS/Pt particles with diameters < 0.2 pm which were able to pass through the membrane filter employed to remove the CdS/Pt particles when sampling the liquid phase of the dispersion (see section 2.2). In addition to the [Cd2+] measurements, the Pt loadings on the CdS/Pt photocatalyst were determined before and after an irradiation in which 69 mm3 of 0, were generated in the head-space. The Pt loading before irradiation was determined as 0.81 %, which compares well with the value of 0.84% predicted from the known concentration and volume of the original Pt colloid used in the platinisation procedure.After irradiation the Pt loading on the CdS/Pt sensitiser had increased to 1.2 YO. Thus, from these results it appears that 3 . 0 ~ 10-6mol Pt were deposited onto the surface of the CdS/Pt photocatalyst during the generation of at least 2.8 x lop6 mol 0,. 18 F A R I510 Photosensitised Oxidation of 40 by CdS irradiation time/min Fig. 4. Volume of O2 photogenerated and detected by gas chromatography in the head-space (1 7 cm3) of the irradiation cell us. irradiation time. The illumination of a CdS/Pt dispersion was carried out under optimum reaction conditions, i.e. pH 13, [CdS/Pt] = 5 mg CIII-~, [PtCli-] = I OP2 mol dmP3, Pt = 0.8 O h and a CdS powder which was annealed in air for 3 h at 100 "C prior to platinisation.3.2.1. Determination of the Optimum Reaction Conditions The most active of the CdS-based photocatalysts tested was the sample with a high platinum content (0.8 % w/w) (see section 3.1). Using this sample as the photocatalyst, a series of experiments was carried out in which the reaction conditions were systematically varied and the effect on the initial rate of 0, evolution [R(O,)] monitored. The observed variation in R(0,) as a function of pH, [CdS/Pt] and [PtCli-] is illustrated in fig. 3(a), (b) and (c), respectively. R(0,) was also found to vary with the amount of Pt deposited onto the CdS, as illustrated in fig. 3(d). In addition, R ( 0 , ) was found to vary in a surprising manner when the CdS starting material was annealed in air at various temperatures above ambient, prior to platinisation ; the observed variation is illustrated in fig.3(e). In contrast to these latter findings, the photocatalytic activity of a CdS/Pt sample was reduced appreciably if it was annealed in air at T > 100 "C, after the CdS has been platinised. 3.2.2. Prolonged and Repeated Irradiations From the results of the work described in the above section the optimum conditions for 0, evolution were taken as [PtCli-] = lo-, mol dm-3, pH 13 and [CdS/Pt] = 5 mg ~ m - ~ . These conditions were used in a series of experiments involving prolonged irradiation of the CdS/Pt sensitiser in the presence of PtCl:-. The CdS/Pt sensitiser itself was prepared by platinisation of a CdS sample pre-baked in air for 3 h.The 0, generated during the long-term irradiation of the optimised photosystem was monitored via the head-space using gas chromatography. The observed variation in the volume of 0, photogenerated vs. irradiation time is illustrated in fig. 4. Prolonged irradiation of the photosystem did not, however, alter significantly [PtCli-1, as determined by u.v.-visible absorption spectroscopy. From the gas chromatograms there was clear evidence that a small amount of H, was generated along with the 0, after prolonged irradiation (> 60 min) of the CdS/Pt photosystem, although work with an h-MPD indicated that no 4 was photogenerated by the system in the first 30 min of irradiation. In another set of experiments the optimised CdS/Pt system was irradiated for 20 min, purged with nitrogen and then re-irradiated for another 20 min.This cycle was repeatedA . Mills and G. Williams 51 1 Fig. 5. Typical [O,] us. irradiation time profiles for the same dispersion of CdS/Pt irradiated for four 20 min periods with only N2 purging in between each irradiation, resulting in curves (a)-(d), respectively. The initial reaction conditions were as described for fig. 4. another two times. The [O,] us. irradiation time profiles for all of these irradiations are illustrated in fig. 5. In this work [PtC1,2-] did not vary appreciably from the first to the last of the irradiations. In addition, when the used CdS/Pt photocatalyst was filtered, washed thoroughly with water and then placed in a fresh PtCli- solution, no recovery in photocatalytic activity was observed.In attempt to regenerate the CdS/Pt photocatalyst, other experiments were carried out in which, after washing with water, the used CdS/Pt powder was stirred for 1 h in the presence of acetate, citrate or EDTA (typically lo-, mol dm-3) and then washed again with water before being placed in a fresh PtC1:- solution. Upon irradiation all of these dispersions showed a partial recovery in photocatalytic activity, typically 50 %. 3.2.3. Sensitised Oxygen Reduction vs. Generation Experiments carried out using the optimised photosystem described in section 3.2.1, in the absence of PtC1:- but in the presence of oxygen (air-saturated solution), showed that the rate of 0, photoreduction sensitised by CdS/Pt was six times that for CdS.However, in the presence of PtC1,2- mol dm-3) the rate of oxygen photoreduction in air- saturated solution by either CdS/Pt or CdS was greatly diminished (see fig. 6). A more rigorous study was carried out on the effect of the initial PtCli- concentration on the rate of 0, photoreduction and photogeneration using a fixed concentration of CdS/Pt (5 mg ~ m - ~ ) and 0, (12 % air-saturated). The results of this work are illustrated in fig. 7. 3.2.4. Alternative ‘ Sacrlficial’ Electron Acceptors A number of sacrificial electron acceptors other than PtC1;- were tested, using the optimised reaction conditions described in section 3.2.1 ., and the observed initial rates of 0, evolution for PtC1:- and these other sacrificial electron acceptors are given in table 18-2512 Photosensitised Oxidation. of &O by CdS I 1 I " b 5 lo 15 20 Fig.6. [O,] us. irradiation time profiles observed for dispersions of CdS/Pt [(a) and (c)] and CdS (6) in air-saturated solution. In this work, for curves (b) and (c) PtC1;- was absent. Unless stated otherwise all other reaction conditions were as described for fig. 4. Fig. 7. [O,] us. irradiation time profiles recorded for a dispersion of CdS/Pt in ca. 12 % air- saturated solution containing different concentrations of PtCli-. The PtC1;- concentrations used were: (a) 5 x (b) 2 x (c) ( d ) 5 x (e) and cf) 0 mol dm-3. All other reaction conditions were as described for fig. 4. 2, along with some relevant redox infdrmation.21* 22 Interestingly, when KAuO; and, to some extent, PtC1;- were used as electron acceptors, the CdS/Pt photocatalyst darkened considerably within a short time of irradiation (20 min), even though no 0, evolution was observed.A .Mills and G. Williams 513 Table 2. Initial rate of 0, generation observed using different sacrificial electron acceptors acceptor initial rate of O2 generation / 1 0-6 mol dm-3 min-' PtCli- PtC1;- P t( OH):- &AuOS Co(NH&,C12+ Fe(CN)i- w- 2.7 0 0 0 0 0 0.3 relevant redox 22 E" = 0.68 V US. NHE PtC1,2- + 4e + Pt + 4C1- E" = 0.72 V US. NHE PtC1;- + 2e -, Pt + 4C1- E" = 0.76 V US. NHE PtCli- + 2e -, PtC1;- + 2C1- 0) (ii) (iii) &AuO; + H' + 3e 4 Au + 30H- (iv) ( 9 (vi) E" = (1.8-0.079 x PH) V US. NHE &Oi- + 2e 4 2SO;- Fe(CN)t- + le 4 Fe(CN):- E" = 2.01 V US. NHE E" = 0.36 V US.NHE Table 3. Initial rates of 0, evolution for a variety of different photocatalysts metal/metal oxide initial rate of deposited 0, evolution photocatalyst (% w/w) / p o l dm-3 TiO, Ti0,/Rh,03 Ti02/Pt CdO CdO/Pt HgS/Pt Si02/Rh,03 A1203 0 1 .o 0.8 0 0.8 0.4 5.0 0.8 3.9" 1.3" 0 0 0 0 0 3!\7" " Irradiations were performed using the full output of the 250 W Xe lamp. In the absence of the photocatalyst no 0, evolution was observed upon illumination of the PtCIi- solution. 3.3. Other Photocatalysts In addition to CdS and CdS/Pt other materials, such as TiO,, TiO,/Pt, CdO, CdO/Pt, HgS/Pt, SiO,/Pt and Al,O,/Pt, were tested for photocatalytic activity. The reaction conditions were similar to those used in the CdS/Pt optimised system, i.e.[PtCli-] = lo-, mol drn-,, [photocatalyst] = 5 mg cm-3 and solution pH 13. The initial rates of 0, evolution recorded for the other photocatalysts are given in table 3.514 Photosensitised Oxidation of -0 by CdS 4. Discussion From the work described in section 3.1. it appears that a number of different CdS based materials are able to photosensitise the oxidation of water by chloroplatinic acid (PtCIi-). It is usually belie~edl’.~~ that PtCli- is reduced to Pt metal by the photogenerated electrons, and, in the presence of an ideal 0, catalyst, the photogenerated holes oxidise water. Thus the overall reaction may be expressed as PtCIi- + 2€-40 -+ Pt + 6C1- + 4H’ + 0, (7) AG = -4F[(0.72- 1.23) - 0.0591 x pH]. In support of this we found that the amount of Pt deposited onto the CdS/Pt sensitiser was approximately equal to that of the 0, photogenerated (see section 3.2).In addition, it was found that the concentration of Cd2+ ions before and after a typical irradiation remained unchanged, indicating that the CdS sensitiser does not undergo extensive photocorrosion [reactions (5) and (6)] during the irradiation. The initial rate of 0, generation was found to vary with the wavelength of irradiation under conditions of constant light flux ( I ) (see fig. 2). The observed variation of R(0,) with 13. ( f 2 0 nm) is similar to the expected absorption spectrum of a direct band-gap semiconductor such as CdS.’ As the band-gap of CdS is 2.4eV, then only light of R < 516 nm should be effective in driving reaction (7) forward if CdS is acting as the sensitiser. This prediction is confirmed by the results illustrated in fig.2. In the absence of CdS/Pt or in the presence of a large-band-gap semiconductor, such as A1,03/Pt or Si0,/Rh,03, no 0, evolution was observed (see tables 1 and 3). These results indicate that CdS/Pt is responsible for photosensitising the oxidation of water by PtCli- when irradiated with ultraband-gap light (i.e. 3, < 516 nm). The work of Rajeshwar and K a n e k ~ , ~ and other^^^,^^ has demonstrated that the interface between n-CdS and a metal or metal oxide deposited onto its surface is difficult to predict and can vary from an ohmic to a Schottky barrier. For example, Gissler et a1,26 have reported that RuO, forms a Schottky barrier of 0.5 V with CdS, whereas Rajeshwar and K a n e k ~ , ~ have found that it forms an ohmic barrier.Work carried out by Aspnes and Heller25 has shown that Pt forms a high Schottky barrier (> 1 V) with CdS; however, they also suggest that if the surface of the CdS is damaged then an ohmic contact might be formed. As a consequence it may be possible to have two or more types of contact for only one material deposited onto the CdS surface. The results of table 1 show that deposits of RuO,, often used as an 0, catalyst,27, 28 do not enhance significantly the photocatalytic activity of CdS. This would not be surprising if the majority of the contacts formed between the two materials were ohmic, since the RuO, would then reflect the potential of the conductance-band electrons and thus be more likely to mediate the transfer of the photogenerated electrons to the PtCli- than to mediate raction (4).The low rate of 0, evolution observed for photoplatinised CdS may indicate that Pt deposited in this manner occurs mainly on damaged surfaces of the semiconductor and, like RuO,, forms mainly ohmic contacts with the CdS particles. However, enhanced rates of 0, generation were observed for CdS samples platinised by the standard method employed in this work (see section 2.1.1 .); this may be due to the formation of a larger number Schottky barriers on the CdS particles. A Schottky barrier between CdS and a Pt site would channel photogenerated holes to the Pt site owing to the directing influence of the electric field associated with the barrier.Once at the Pt site the holes could oxidise water via reaction (4). Since it is generally considered unlikely that the oxidation of water could be made to occur on the bare surface of CdS,’ it is surprising that CdS alone was able to photosensitise reaction (7) (see table 1). From the [O,] us. irradiation time profile [fig, l(c)] for CdS alone, it appears that 0, generation is prompt, indicating that 0,A . Mills and G. Williams 51 5 generation is possible even if there is initially little or no 0, catalyst (such as photodeposited Pt) on the surface of the CdS. In the mechanism for the photocatalysis of reaction (7) discussed above, 0, evolution is most likely to occur at a Pt site which forms a Schottky junction with the CdS. Thus when CdS alone is used to sensitise reaction (7), it is necessary to propose that the oxidation of water occurs on some of the freshly deposited Pt sites produced by the initial photoreduction of PtCl:-.However, the low rate of 0, generation for both CdS, after prolonged irradiation (see fig. l), and photoplatinised CdS/Pt (see table 1) indicates that the photodeposition of Pt does little to enhance the ability of CdS to sensitise reaction (7), possibly due to the predominant formation of ohmic, rather than Schottky, CdS-Pt contacts. From the variation of the initial rate of 0, generation [R(O,)] as a function of pH [see fig. 3(a)] it would appear that reaction (7) cannot be made to proceed at a measurable rate if the solution pH is < 12 although, at this low pH, AG for reaction (7) is still very negative (-77 kJ mol-l).At the optimum pH (13) the driving force for reaction (7) is increased (AG = - 100 kJ mol-’), and therefore it is not surprising that the rate of 0, evolution is also increased. The decrease in R(0,) at pH > 13, despite a further increase in AG, may be due to an ionic-strength effect. The observed variation of R(0,) with [CdS/Pt] illustrated in fig. 3(b) is typical of semiconductor dispersion^.^. 29 Initially, by increasing [CdS/Pt] the amount of light absorbed increases, and therefore so does R(0,). Eventually a semiconductor dispersion concentration is reached at which the amount of incident light which is absorbed reaches a maximum. A further increase in the concentration of the dispersion serves only to reduce the penetration depth of the incident light and thereby increase the likelihood of losing to the surroundings any scattered light which could be absorbed.At high [CdS/ Pt] the penetration depth is reduced to such a level that the light lost to the surroundings due to scattering is significant, but constant. The observed variation of R(0,) with [PtCli-] is illustrated in fig. 3(c) and is the product of at least two effects. As [PtCl:-] is increased it is expected that the probability of reaction between PtC1:- and any photogenerated reducing species, such as conduction- band electrons or S- radicals, will increase. In addition, as illustrated by the broken line in fig. 2, an increase in the [PtCli-] will increase the amount of light absorbed by the electron acceptor, and therefore decrease the amount of light absorbed by the CdS/Pt.However, this latter effect is not expected to be large, since the [CdS/Pt] employed in this work was high ( 5 mg CM-~) and the PtC1;- does not absorb very strongly light of 2 > 400 nm (see fig. 2). Under the experimental conditions employed in this work most of the irradiation light (1 > 400 nm) will be absorbed by the CdS/Pt particles close (1-2 mm) to the front cell wall and not by the PtCl:-, even at concentrations of lop2 mol dm-3. The rise in R(0,) with increasing Pt deposited onto the surface of the CdS is illustrated in fig. 3(d) and may be taken as a reflection of the increasing probability of electron transfer to the PtCl;-, rather than electron-hole recombination [(reaction (3)], with increasing Pt present.This trend in R(0,) with Pt indicates that the role of some of the Pt sites may be that of a ‘sink’ for electrons and a catalyst for electron transfer to the PtC1:- from the reducing species (e- and/or S- radicals). In support of this, several studiesg* 3o on the photoreduction of water sensitised by other semiconductors have demonstrated that the rate of H, evolution varies with the Pt deposited in a manner similar to that illustrated in fig. 3(d). In this latter work the majority of the Pt sites functioned as & catalysts. The effect on R(0,) of baking the CdS in air, prior to platinisation is illustrated in fig. 3(e). We refer to this process as ‘thermal activation’, since it leads to an increase in photocatalytic activity of the CdS/Pt.Separate experiments showed that the CdS could not be ‘thermally activated’ in the absence of 0,. This latter finding suggests that the process of ‘ thermal activation’ is most likely due to the oxidation of at least some of the516 Photosensitised Oxidation of 5 0 by CdS surface of the CdS powder, probably to CdO. Since increasing the baking temperature above 100 "C leads to a gradual decline in activity, it would appear that increasing the thickness of the oxidised CdS surface layer above an optimum level is detrimental to the activity of the CdS/Pt sensitiser. How the CdO film enhances the activity of the CdS is not clear. It may be that the CdO film produces an improvement in the electrical contact between the CdS and the Pt deposits and so increases the efficiency of electron transfer from the CdS to the PtC1;- via the Pt sites.A second possibility is that the CdO film removes recombination centres, such as surface SH- groups, and so increases the lifetime of the photogenerated reducing and oxidising species, such as e- and h+. Supporting evidence for the latter suggestion comes from the work of Henglein and co-worker~.~~ This group has demonstrated recently that a precipitate of cadmium hydroxide onto the surface of colloidal CdS particles not only increases their luminescence lifetimes and intensities but also stabilises them against anodic corrosion. The most likely surface product formed on placing CdO in solution at pH 13 is Cd(OH),. However, in contrast to the observations made by Henglein et aZ.,31 in our work 'thermal activation', which presumably produces a surface layer of CdO, does not produce any detectable enhancement in the very weak, if any, luminesence of the CdS powders and increased the activity of the original CdS.However, it was found that a precipitate of Cd(OH), [0.73% w/w Cd(OH),/CdS/Pt] onto the surface of a CdS/Pt powder did lower (by ca. 2 fold) its photocatalytic activity. The results illustrated in fig. 4 and 5 indicate that both prolonged and repeated irradiation of the CdS/Pt leads to a deterioration in R(O,), even under optimum reaction conditions and with no significant decrease in [PtCl;-]. Washing the used CdS/Pt with water does not regenerate the photocatalytic activity of th CdS/Pt. However, stirring with acetate, citrate or EDTA does produce a partial recovery in the photocatalytic activity of the CdS/Pt, and it is relevant to note that Cd(OH), is known to dissolve in these washing solutions.These results suggest that during the photocatalysis of reaction (7) by CdS/Pt some anodic corrosion of the CdS occurs, thereby generating Cd(OH), on its surface which, with time, accumulates and eventually destroys the photocatalytic action of the CdS/Pt, possibly in a manner similar to that described by Hengelein et al.31 and outlined above. Although our a.a.s. studies of the CdS/Pt systems indicated no significant generation of Cd2+ ions, the irradiation time employed was short (20 min) and the background concentration of Cd2+ ions was high (7.8 x mol dm-3), see section 3.2. It seems quite possible, therefore, that some photocorrosion of the CdS/Pt does occur during illumination but not to an extent sufficient for detection by our method of analysis.Experiments carried out in the presence of O,, but in the absence of PtC1;- (see section 3.2.3 and fig. 6 ) demonstrate that CdS/Pt is ca. six times more active than CdS at sensitising the photoreduction of 0, ; similar findings have been reported by others." Work carried out by Memming and his co-workers and others on CdS single crystals7* 159 32 and ~ o l l o i d s ~ ~ - ~ ~ has shown that 0, is reduced by the photogenerated electrons (e-) to OH-, i.e. (8) 0, + 2H+ + 4e- + 20H-. In addition, these workers found that 0, can also be reduced via a photoelectrochemical oxidation process, i.e. (9) S2- + 2KO + 0, + 4h+ --+ SO:- + 4H+ in which the initial step is believed3, to be S2-+ h+ --* S- (10) followed by s- + 0, 40,.(1 1)A . Mills and G. Williams 517 In an air-saturated solution, with [PtCli-] = lop2 mol dm-3, no overall 0, evolution or reduction was observed (see fig. 6 ) . Thus it seems likely that under these conditions the rates of 0, evolution and reduction were similar. At a lower [O,] (12 YO air-saturated) and a high [PtCli-] (1 x mol dm-3) it was possible to observe 0, evolution (see fig. 7) and at the same rate as observed in the absence of 0,. Thus from the results of the latter set of experiments it would appear that PtCli-, at concentrations > lo-, mol dmP3, is more efficient at scavenging the photogenerated electrons and S- radicals than O,, at concentrations < 12 % air-saturated.Since most of the work reported in this paper was carried out under N,-purged conditions and in the presence of a high [PtClt-1, it was assumed that under such conditions the rate of 0, evolution was not affected by the generation of 0, until the concentration of dissolved 0, reached ca. 9 12% air saturation. As a result, at very low concentrations of [O,], as found in the initial part of any irradiation, the likely major processes are e- + PtCli- -+ PtCli- (12) and possibly s- + PtCli- -+ PtC1;- + s. (13) The likelihood of reaction (13) taking place will depend upon whether reaction (10) predominates over reaction (4) even in the presence of a suitable 0, catalyst, such as some, or all, of the Pt sites on the CdS. In a recent paper Memming and co-~orkers'~ have suggested that 0, evolution photosensitised by CdS does not occur via reaction (4) but rather that an unstable Pt"' species (PtCli- or some other Pt"' species) which is generated via steps (12) and/or (13), is responsible, i.e.(14) 4PtCl:- + 6&0 -+ 4Pt + 12H' + 24C1- + 30,. It has yet to be proved whether reaction (14) occurs, if at all, in bulk solution and/or on the surface of a CdS/Pt powder particle. However, if the latter were the case it would be more likely to occur at a Pt site than on the easily oxidised CdS surface. Some evidence for and against the Pt"' intermediate mechanism comes from our work with sacrificial electron acceptors other than PtCli- (section 3.2.4). As can be seen from table 2, the use of sacrificial agents, such as PtClt-, Pt(OH)i-, Co(NH,),C12+, &AuO, and $Oi-, did not lead to the photogeneration of 0,.However, most of the sacrificial electron acceptors [PtCl;-, Pt(0H)i- and KAuO,] did reduce to a large extent the ability of CdS/Pt to photoreduce 0, under air-saturated conditions. This indicates that these electron acceptors are efficient at scavenging the photogenerated electrons and S- radicals. The observed rapid darkening of the CdS/Pt upon illumination in the presence of KAuO; or PtCl:-, presumably owing to their reduction to the metal, provided further evidence of the effective scavenging action of these two electron acceptors. Given that several of the electron acceptors, other than PtC1,2-, appear to act as efficient scavengers of any photogenerated reducing species, it is difficult to explain using the conventional mechanism, proposed by Gratzel and co-workers and summarised by reaction (7), why the photogenerated holes do not bring about the oxidation of water via reaction (4).Such difficulties are not encountered using the Pt"' mechanism, since the intermediate species PtCli- will only be generated when PtC1,2- is used as the sacrificial electron acceptor. Interestingly, the Pt 'I1 mechanism provides an alternative, and slightly simpler explanation for the observed photogeneration of 0, by CdS alone, i.e. some photoreduction of the PtC1:- to Pt"' occurs and water oxidation then occurs via reaction (14). The enhancement in R(0,) observed for CdS/Pt over CdS may be interpreted in terms of the Pt"' mechanism as evidence that some Pt sites serve to mediate the reduction of PtC1;- via reactions (12) and (1 3), and that other Pt sites possibly catalyse reaction (14).518 Photosensitised Oxidation of KO by CdS The Pt"' mechanism does not offer a ready explanation as to why the CdS/RuO, sample was little better than CdS in sensitising the oxidation of water, despite the likely formation of ohmic or low-Schottky-barrier CdS-RuO, junctions and the established27.28 0, catalytic activity of RuO,. In addition, the Pt"' mechanism does not provide an adequate explanation for the observed photogeneration of 0, when Fe(CN)i- was used as a sacrificial electron acceptor (see table 2). In the absence of CdS/Pt, and in the presence or absence of A1,03/Pt, illumination of the Fe(CN),3- does not lead to 0, generation.These findings suggest that the oxidation of water photosensitised by CdS/ Pt is possible in the absence of PtCli- and this, in turn, lends support to the idea that at least some oxidation of water can occur, via reaction (4), on the surface of a CdS/Pt particle. It may be, however, that the reduction of Fe(CN):- at pH 13 can produce an intermediate species capable of oxidising water; this would represent a similar route for 0, evolution, as suggested by the Pt"' mechanism. The variations in R(0,) observed using different photocatalysts are summarised in table 3. Although TiO, was able to photocatalyse reaction (7) r e a d i l ~ , , ~ . ~ ~ the presence of Pt or Rh,03 on the TiO, reduced, if anything, its photocatalytic activity.Similar observationsg have been made for TiO, when used to sensitise the photo-oxidation of water to 0, by Fe3+ ions at a concentration of lop3 mol dm-3 in 5 x mol dm-3 &SO,. Although this effect may be simply due to the deposited material 'screening' the TiO, from the incident light,37 it may also be that the deposited material enhances electron- hole recombination. 38 Like CdS, both Cd03' and HgS40 are n-type semiconductors which, upon absorption of ultra-band-gap irradiation, generate conductance-band electrons capable of reducing PtCli- and holes capable of oxidising water. However, neither CdO or CdO/Pt photosensitised the photogeneration of 0, in the presence of PtCli-. In addition, these two materials failed to photoreduce 0, in the absence of PtCli-.The lack of any photoactivity for either of the two materials is disappointing, given the reported ability of CdO to sensitise the photo-oxidation of water by Fe(CN)i- under the same conditions of PH.~' The lack of photocatalytic activity by Si0,/Rh,03 and Al,03/Pt was not unexpected, since both these materials possess band gaps too large to absorb the incident irradiation (A > 400 nm). 5. Conclusions Cadmium sulphide, with or without metal or metal oxide deposits such as Pt, Au, Ag, Rh, RuO, and Rh,03, is able to photocatalyse the oxidation of water by PtCli- at pH 13. A CdS-based photocatalyst with a high activity is produced by depositing Pt onto the CdS (0.8% w/w) using a method involving precipitation of a Pt colloid onto CdS which had been annealed at 100 "C for 3 h prior to platinisation (see section 2.1.1 .).Using this photocatalyst the optimum conditions for the photogeneration of 0, are as follows: pH 13, [CdS/Pt] = 5 mg cmP3, Pt content 0.8 YO and [PtCli-] = lo-, moi dm-3. The majority of the experimental evidence described in this paper supports the original findings of Gratzel and co-w~rkers~~ and favours their suggested mechanism in which the CdS/Pt truely acts as a photocatalyst for reaction (7), i.e. corrosion of the CdS is not a major reaction. The lack of 0, evolution observed when electron acceptors other than PtC1;- were used, such as PtCli-, Pt(OH)i-, KAuO;, Co(NH3),C12+ and &Oi-, is difficult to explain using this mechanism. Results such as these have led other workers15 to propose an alternative mechanism in which the water is oxidised by a Pt'" species, generated by the reduction of PtCli- by a photogenerated reducing species, such as e- or S - .However, the Pt"' mechanism fails to explain many of the results reported here, including the photo-oxidation of water by Fe(CN)i-, sensitised by CdS/Pt. In order to resolve these differences, further work is now in progress to elucidate the mechanism for 0, evolution when CdS/Pt is used as the photosensitiser.A . Mills and G. Williams 519 We thank the S.E.R.C. for supporting this work. In addition we thank Professor Memming and his group for sending us preprints describing their latest work on the photoelectrochemistry of CdS. References 1 Energ?, Resources through Photochemistry and Cutalysis, ed.M. Gratzel (Academic Press, New York, 1983). 2 K. Rajeshwar, P. Singh and J. DuBow, Electrochim. Acta. 1978, 23, 11 17. 3 T. Watanabe, A. Fujishima and K. Honda, Chem. Left., 1974, 897. 4 K. Kalyanasundaram, E. Borgarello and M. Gratzel, Helv. Chim. Acta, 1981, 64, 362. 5 A. J. Frank and K. Honda, J . Phys. Chem., 1982, 86, 1933. 6 M. M. T. Khan, R. C. Bhardwaj and C. M. Iadhev, J. Chem. Soc., Chem. Commun., 1985, 1690. 7 D. Meissner, R. Memming, B. Kastening and D. Bahnemann, Chem. Phys. Letf., 1986, 127,419. 8 A. Mills and G. Williams, J . Chem. SOC., Chem. Commun., 1987, 606. 9 A. Mills and G. Porter, J . Chem. SOC., Faraday Trans. I , 1982, 78, 3659. 10 J. R. Darwent, J. Chem. Soc., Faraday Trans. 2, 1981, 77, 1703. 1 1 J. R. Harbour, R.Wolkow and M. L. Hair, J . Phys. Chem., 1981, 85, 4026. 12 B. Hubesch and B. Mahieu, hog. Chim. Acta., 1982, 65, L65. 13 D. H. M. W. Thewissen, K. Timmer, E. A. Zouwen-Assink, A. H. A. Tinnemans and A. Mackor, 14 I. Wilner and Z . Goren, J. Chem. SOC., Chem. Commun., 1986, 172. 15 I. Lauermann, D. Meissner and R. Memming, J . Electroanal. Chem., 1987, 228, 45 and references 16 J. P. Collin, J. M. Lehn and R. Ziessel, Nouv. J . Chim., 1982, 6, 405. 17 N. M. Dimitrijevoic, S. Li and M. Gratzel, J. Am. Chem. SOC., 1984, 106, 6565. 18 G. Pass and H. Sutcliffe, Practical Inorganic Chemistry (Wiley, New York, 1974), p. 108. 19 A. Mills, J . Chem. SOC., Chem. Commun., 1982, 367. 20 A. Mills, A. Harriman and G. Porter, Anal. Chem., 1981, 53, 1254. 21 G. Milazzo and S. Caroli, Tables of Standard Electrode Potentials (Wiley, Chichester, 1978). 22 M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions (Pergamon Press, Oxford, 23 J. S. Curran, J. Dominech, N. Jaffrezic-Renault and R. Philippe, J . Phys. Chem., 1985, 89, 957. 24 K. Rajeshwar and M. Kaneko, J . Phys. Chem., 1985, 89, 3587. 25 D. E. Aspnes and A. Heller, J . Phys. Chem., 1983, 87, 4919. 26 W. Gissler, A. J. McEvoy and M. Gratzel, J . Electrochem. Soc., 1982, 129, 1733. 27 A. Mills, S. Giddings, and I. Patel, J . Chem. Soc., Faraday Trans. I , 1987, 83, 2317. 28 A. Mills, S. Giddings, I. Patel and C. Lawrence, J . Chem. SOC., Faraday Trans. 1, 1987, 83, 2331. 29 M. V. Rao, K. Rajeshwar, V. R. Pal Verneker and J. DuBow, J . Phys. Chem., 1980, 84, 1987. 30 T. Sakata, T. Kawai and K. Washimoto, Chem. Phys. Lett., 1982, 88, 50. 31 L. Spanhel, M. Haase, H. Weller and A. Henglein, J . Am. Chem. SOC., 1987, 109, 5649. 32 D. Meissner, C. Benndorf and R. Memming, Appl. Surf Sci., 1987, 27, 423. 33 D. Meissner, R. Memming S. Li, S. Yesodharan and M. Gratzel, Ber. Bunsenges. Phys. Chem., 1985, 34 A. Henglein, Ber. Bunsenges. Phys. Chem., 1982, 86, 301. 35 R. Hayes, P. A. Freeman, P. Mulvaney, F. Grieser and T. Healy, Ber. Bunsenges. Phys. Chem., 1987, 36 B. Kraetler and A. J. Bard, J. Am. Chem. Soc., 1978, 100, 4317. 37 P. Pichat, J. M., Herrmann, J. Disdier, H. Courbon and M. N. Mozzanega, Nouv. J . Chim., 1981, 5, 38 T. Kawai and T. Sakata, J . Chem. Soc., Chem. Commun., 1980, 694. 39 A. Harriman, J . Chem. SOC. Faraday Trans. I , 1983, 79, 2875. 40 R. S. Davidson and C. J. Willsher, J . Chem. SOC., Faraday Trans. I , 1980, 76, 2587. J . Chem. Soc., Chrm. Commun., 1985, 1485. therein. 1966). 89, 121. 91, 231. 627. Paper 7/00090A ; Receired 21s1 December, 1987

ISSN:0300-9599    

DOI:10.1039/F19898500503

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1989, 85(3), 521-535 Acid-Base Equilibria in Aqueous Micellar Solutions Part 1 .-' Simple' Weak Acids and Bases Calum J. Drummond,*T Franz Grieser and Thomas W. Healy Department of Physical Chemistry, The University of Melbourne, Parkville, Victoria, 3052, Australia The acid-base equilibria of a number of phenols, amines and carboxylic acids in aqueous micellar solutions and organic solvent-water mixtures have been examined. For the majority of the molecules investigated, the differences between the plr$ values in pure water and the apparent plr$ values when the molecules reside within micellar interfacial microenvironments can primarily be ascribed to the differences between the mean intrinsic solvent properties of the interfacial and bulk phases, with an additional contribution from the electrostatic micellar surface potential in the case of the charged aqueous micellar solutions.The apparent p& of a weak acid or weak base residing at or in the vicinity of a charged interface, pKtbs, is generally considered to be composed of an electrostatic component due to the surface potential and an inherent interfacial non-electrostatic component. This relationship is often expressed as1-26 where pK," is the apparent p& of the molecule in the absence of any potential, Y is the mean field potential at the average interfacial site of residence for the prototropic moiety, F i s the Faraday constant, R is the universal gas constant and Tis the absolute temperature. In spite of the fact that eqn (1) is increasingly being used to rationalize acid-base behaviour at charged interfaces, very little quantitative work9* 1 9 7 23* 26 h as been done to isolate the principal factors contributing to the magnitude of the pK," component.Therefore, the present series of investigations were designed with the following major objectives : (a) to ascertain the magnitude of the non-electrostatic component to the values [i.e. the pK," value in eqn (l)] for a wide range of interfacially located 'simple' weak acids and bases and commonly used acid-base indicators; and (b) to determine the reasons for the difference, if any, between the pK," values of these interfacially located ' simple ' weak acids and bases and commonly used indicators and their p& values in bulk aqueous solution.In this first paper in the series, the acid-base equilibria of a number of 'simple' weak acids and weak bases in aqueous micellar solutions are examined and a quantitative assessment of their pK," values is given. There is an enormous pool of data in the literature concerned with the acid-base equilibria of 'simple' weak acids and bases in organic solvent-water mixtures. Some of these data are used to employ a procedure first developed by Fernandez and Fromherzg to compare the acid-base equilibria in organic solvent-water mixtures with those in aqueous non-ionic and charged micellar solutions. t Present address : CSIRO, Division of Chemicals and Polymers, G.P.O. Box 433 1, Melbourne, Victoria, 3001, Australia. 52 1522 ' Simple' Acid-Base Equilibria in Micelles pH-Titration results obtained in the present work for p-nitrophenol, p-(t-buty1)phenol and p-toluidine in both non-ionic and charged micellar systems are analysed. In addition, the results of other researchers for myristic ~tearylamine~~ and 4-octadecyloxy- 1 -naphthoic acid" in micellar systems are also examined.Since a number of biologically important molecules exhibit either weak acid or weak base behaviour there is an added incentive to investigate the acid-base equilibria of 'simple' weak acids and bases at lipid-water interfaces. For example, the apparent p& values of fatty bile acids,2s uncouplers of oxidative phosphorylation,' ph~sphatidylserine,'~~~~ phosphatidylethanolamine20 and some local anaesthetics15 located in rudimentary model biological membranes are all shifted relative to their p& values in pure water.Experiment a1 p-Nitrophenol, p-t-butylphenol and p-toluidine were purchased from Tokyo Kasei, Kogyo Co. The purity of each of these reagents was confirmed by established tests.29 Dodecyltrimethylammonium bromide (DTAB) was supplied by Tokyo Kasei and specially purified sodium dodecyl sulphate (SDS) by B.D.H. Chemicals Pty. Ltd. Both surfactants were recrystallized prior to use. Brij-3 5 and n-dodecyloctaoxyethylene glycol monoether (C12E,) were obtained from B.D.H. and Nikko Chemical Co., respectively. These surfactants were used as received. The inorganic reagents, tetraethylammonium chloride (TEAC), NaOH, HCl, KC1 and NaBr, were analytical grade and were employed without further purification.Buffer concentrates were from Merck. The aqueous solutions were prepared with Millipore-filtered water which had been distilled at least once before being filtered (conductivity < 1 x R-' cm-l and air-water surface tension equal to 72.0 mN m-' at 25 "C). All experiments were conducted at 25 "C. Both the conjugate acid and conjugate base forms of p-nitrophenol, p-(t-buty1)phenol and p-toluidine are relatively water soluble, hence high concentrations of surfactant were employed in an attempt to solubilize most of the species in the micellar phase. For the same reason micellar systems that had the same surface charge as the charged forms of the probe molecules were not examined. In order to ensure the removal of any undissolved weak acid or weak base, each solution was filtered before the pH-titration was performed.The solution under investigation was kept in a water-thermostatted reservoir fitted with a glass electrode and a double-junction reference calomel electrode. This dual electrode system was utilized so that the excessive drift in the pH-meter reading which is sometimes observed when a single combined electrode is used in concentrated micellar solution could be avoided.30 The upper compartment of the reference electrode contained a saturated KCl solution whilst the lower compartment contained a lop3 mol dm-3 solution of TEAC. The electrodes were standardized with a two-buffer adjustment procedure and the pH of the standardizing buffer solutions were always chosen to bracket the apparent p& of the weak acid or weak base in the system being examined. A Radiometer PHM84 research pH-meter was used to monitor the pH.The bulk pH of the solution in the reservoir was adjusted with small aliquots of concentrated solutions of NaOH or HCl in order to minimize dilution effects. After the solution had equilibrated, and a stable pH-meter reading had been reached, a sample was withdrawn from the reservoir and the u.v.-visible absorption spectrum was measured on a Varian Cary Model 210 spectrophotometer. This procedure was repeated until a full pH-titration profile was obtained. All the acid-base equilibria investigated were treated as being of the formC. J . Drummond, F. Grieser and T. W. Healy 523 where HX is the conjugate acid form of the molecule with charge z, H' is the proton and X is the conjugate base form of the molecule.The thermodynamic acid-base equilibrium constant for this reaction is given by The ratio of the concentrations of the conjugate acid-base forms could be determined from the spectra as a function of pH. In pure water the activity coefficient, y , of any singly charged species was approximated by the mean activity coefficient, y + , for HCl in water at the same total ionic strength.31 The influence of electrolyte on-the activity coefficient of a neutral species was assumed to be negligible, therefore the activity coefficient for a neutral species was set equal to unity. Since there is no available information on the mean activity coefficients of species residing within the interfacial microenvironment of micelles, the last term in eqn ( 3 ) had to be neglected for the apparent p& determinations in micellar solutions.The p& values in pure water and the apparent p& values in the micellar solutions were determined from the u.v.-visible absorption spectra as a function of the bulk aqueous pH using eqn (4) (4) Yx log - YH X A-AH, Ax - A m with a = where A is the absorbance at the long-wavelength band maximum of the conjugate base form, A,,,, at the particular pH being examined. AHx and A, are the absorbances at R,,, when the pH values are set such that only acid and base species are present, respectively. In order to ascertain whether or not the apparent p& of the molecules were well-defined, at least seven intermediate a values were looked at for each micellar solution.Examples of the change in the spectrum of p-nitrophenol, p-(t-buty1)phenol and p-toluidine with bulk aqueous pH can be found in ref. ( 3 2 ) . General Considerations The apparent pK, of the prototropic moiety of a weak acid (or weak base) residing within the aqueous interfacial microenvironment of a micelle can be split into a number of quantitative components. This section outlines the relationships and accompanying assumptions which allow one to derive these components. The initial part of this section closely follows reasoning presented el~ewhere.~~ 26 Nevertheless, it is restated herein because a knowledge of this reasoning is necessary for the understanding of the analysis and discussion of the results contained in this work and subsequent parts in this series.The apparent acid-base equilibrium for a weak acid (or weak base) located within an aqueous micellar interfacial phase can be represented as where subscripts i and w denote the interfacial phase and the bulk aqueous phase, respectively. The ' two-phase ' thermodynamic acid-base equilibrium constant for this reaction is defined by524 ' Simple' Acid-Base Equilibria in Micelles where a;, al,, and a;+ denote the activities of the various species in the phases. The experimentally determined apparent pKd has the form . . (8) Y'x Y L X and therefore = p c + log - . The standard Gibbs free energy of the 'two-phase' reaction given by eqn (9, AGO, and the ' two-phase ' thermodynamic acid-base equilibrium constant are related and with AGO expressed in terms of the standard chemical potentials for the species involved in the equilibrium where v/ is set equal convenient to define Also note that (9) to zero when a non-ionic interface is considered.In addition, it is 1 (Jpwf+p;-jp Fry 2.303 RT HX) - 2.303RT pKZ = Fv/ 2.303 RT (Jp;+ + p: - p;x). pPa = pK,"+ 1 2.303 RT - and the pK," component of eqn (1) and p E are related by PK," = p&+lOg-. Yfc Y h X The acid-base equilibrium for a weak acid (or weak base) in an organic solvent-water mixture (m) can be represented as HXg)$H;+Xg-l). (13) The ' single-phase ' thermodynamic acid-base equilibrium constant for this reaction is defined by pK," = -logTaR+ a: 'HX 1 If one assumes (a) that there are no specific molecular interactions which interfere with the interfacial acid-base equilibrium, and (b) that the solvent properties of an interfacial phase can be approximated by an organic solvent-water mixture with equivalent solvent properties, then for the particular equivalent organic solvent-water mixture and pK: and pK," differ solely by the work required to transfer the proton from the bulk aqueous phase to the interfacial phase, i.e.C. J .Drummond, F. Grieser and T. W. Healy The change of free energy for the transfer process is related to the medium effect on the proton, myH+, where H+ (std. state, w) -+ H+ (std. state, m) 525 by the formula33 /A;? -pLW+ = 2.303 RT log (18) PK; = pK," -log (19) and as a result The medium effect has also been referred to as the 'primary medium effect', the 'degenerate activity coefficient' and the 'distribution coefficient '.34 35 as to the nature of the medium effect on the proton.The problem is that a direct measurement of the medium effect on a single ionic species is not possible and it can only be estimated by making various non- thermodynamic assumptions. In this series of investigations the procedure of Fernandez and Fromherzg has been followed and it has been assumed that for a particular organic solvent-water mixture the log ,y,+ value can be approximated by the log ,y, value for HCl in the same organic solvent-water mixture. The justification for taking this approach will be given in the forthcoming discussion. The values of logmy, for HC1 in the various organic solvent-water mixtures were obtained from the change in the standard potential, E O , for the cell Pt/H,(g), HCl in solvent s, AgCl/Ag as an organic solvent is added to pure water.The relationship is given by3' At present, there is considerable - (,E" - ,E") F 2RTln 10 logmy, = where ,E" and ,E" are the standard potentials for the cell when pure water and an organic solvent-water mixture are the solvents, respectively. The value used for ,E" was 0.22234 V at 25 0C.37 The values for ,E" were either obtained from the compilation of Feakins and French38 or the one of Bates.34 The one exception to this was the ,E" value for the 82 weight % 1,4-dioxane-water mixture which was taken from the work of Danyluk et af.39 It should be mentioned that all the standard potentials used were based on the molar scale and as a result the calculated my, values relate to molarity.Fig. 1 shows logmy+ for HCl as a function of the dielectric constant of 1,4-dioxane-water, ethanol-water and methanol-water mixtures. The values for the dielectric constants were obtained by interpolation of the data of AkerloP' and Critchfield et aL4' For a particular weak acid (or weak base) the p& values obtained in micellar solutions and in organic solvent-water mixtures are, in general, referred to the pEi, value determined in pure water. Therefore it is convenient to define ApK," may be separated into an electrostatic and a non-electrostatic component :42 A P C = (APK,")~, +(APK,")none,* (24) The electrostatic component is due to the different amounts of work required to charge526 2.0 i-* E 4 8 1.0 ' Simple ' Acid-Base Equilibria in Micelles 20 40 60 dielecnic constant Fig.1. logmy, Values as a function of the dielectric constant of methanol-water (e), ethanol-water (0) and 1,4-dioxane-water (0) mixtures. the ions in two media of different dielectric constant. It can be estimated from the Born equation :42* 43 where N is Avogadro's number, e is the elementary charge, E is the dielectric constant and rj is the ionic radius of species j . The non-electrostatic component incorporates all the additional influences on A p e , such as specific solute-solvent interactions. Using the relationships given in eqn (19) and (24), eqn (22) can be rewritten as: (26) AP& = (ApK,">el + (APK,")nont?l - log rnYHf. Hence, if the quantity (ApC),,,,, -log myH+ is either small in magnitude compared with (ApK,"),, or equivalent for different solvent mixtures with the same dielectric constant, one would expect to observe for a particular molecule the same ApFi behaviour as a function of solvent dielectric constant irrespective of the particular kind of solvent mixture.In this and subsequent parts in this series, unless specifically indicated to the contrary, it is taken for granted that in charged micelles the prototropic moieties of the guest molecules reside, on average, in the plane of the surface charge. As will become apparent the results justify this assumption. Independent n.m.r. also indicate that the vast majority of non-lipoidal micelle solubilized aromatic molecules reside on average within the interfacial headgroup region of charged micelles.Results and Discussion The logmy, data in fig. 1 were utilized to construct reference ApK: curves for the molecules investigated from their pK," values in organic solvent-water mixtures of1.0 .- m 3 0.5 C. J . Drummond, F. Grieser and T. W. Healy I I 1 527 2 0 4 0 6 0 dielectric constant Fig. 2. p-Nitrophenol ApKj reference values as a function of the dielectric constant of methan~l--water~~~~~-~~ (a) and ethano1-waterg9 (0) mixtures. known dielectric constant. Fig. 2-7 show the ApKi values as a function of the solvent dielectric constant. All the pK," values employed in this study were obtained from the literature. Some of the pK," values were based on the molality scale so these were converted into molarity units before subtracting log The various sources of the pK," values43.45-61 are indicated in the relevant figure captions. The ApKA curve for p-(t-butyl)phenol, fig. 3, was actually constructed from ApK," values for phenol. This was done because literature pK," data for phenol are far more abundant. From the work of Parsons and Roche~ter'~ with substituted phenols, it is clear that the ApK," behaviour of phenol and p-(t-buty1)phenol are equivalent. For myristic acid and stearylamine, fig. 5 and 6, the ApK," behaviour with dielectric constant was assumed to be the same as for their water soluble homologues propionic acid and hexylamine. Similarly, 4-octadecyloxy- 1 -naphthoic acid, fig. 7, was assumed to have the same ApK," behaviour as benzoic acid. As shown in fig. 2, there is a small difference between the methanol-water and ethanol-water ApKi values of p-nitrophenol as a function of solvent dielectric constant.Fig. 3 indicates that the ApKi behaviour of phenol with solvent dielectric constant is equivalent for dioxane-water and ethanol-water mixtures. Interestingly, however, there is a significant difference between this ApKA behaviour and that found for meth- anol-water mixtures. Evidently, there is some kind of aberrant specific solute-solvent interaction in this particular methanol-water system. The reference ApK: values for p - toluidine as a function of dielectric constant, fig. 4, also show some solvent specificity as indicated by the different dioxane-water and ethanol-water results. In the organic solvent-water mixtures examined for propionic acid, hexylamine and benzoic acid, ApKL is a unique function of solvent dielectric constant.This can be seen in fig. 5-7. In this study, it has been assumed that the log (yk/yLx) term in eqn (1 2) is negligibly small. It has also been assumed that there is no contribution to pK," values due to specific528 ' Simple ' Acid-Base Equilibria in Micelles 2.0 .- * 3 1 .o 20 40 60 dielectric constant Fig. 3. p-(t-Buty1)phenol ApKl reference values, based on phenol, as a function of the dielectric constant of rnethan~l--water~~*~~ (a), e t h a n o l - ~ a t e r ~ ~ ~ ~ ~ (0) and 1 ,4-dioxane-water51 (0) mixtures. 2 0 40 60 dielectric constant Fig. 4. p-Toluidine ApK: reference values as a function of the dielectric constant of ethanol- ~ a t e r ~ ~ , ~ ~ , ~ ~ (0) and 1,4-dio~ane-water~~ (0) mixtures.C.J . Drummond, F. Grieser and T. W. Healy 529 20 40 60 dielectric constant Fig. 5. Myristic acid ApKi reference values, based on propionic acid, as a function of the dielectric constant of methan~l-water~~.~~ (O), e t h a n ~ l - w a t e r ~ ~ * ~ ~ . ~ ’ (0) and 1,4-dioxane-~ater~~-~~ (0) mixtures. -1.0 -2.0 2 0 40 6 0 dielectric constant Fig. 6. Stearylamine ApKi reference values, based on hexylamine, as a function of the dielectric constant of methanol-wateP (0) and ethan~l-water~~ (0) mixtures.530 ' Simple' Acid-Base Equilibria in Micelles fl 20 40 60 dielectric constant Fig. 7. 4-Octadecyloxy-1-naphthoic acid ApK: reference values, based on benzoic acid, as a mixtures. function of the dielectric constant of rnethan~l-water~~ (a) and ethan~l-water~~,~~.~' (0) molecular interactions. Thus by comparing a ApK," value with the plot of reference ApP; values as a function of the solvent dielectric constant, one can estimate the effective dielectric constant of the interfacial microenvironment of micelles.As the ApKA values of myristic acid, stearylamine and 4-octadecyloxy- 1 -naphthoic acid apparently respond uniquely to changes in the solvent dielectric constant, these molecules can provide unambiguous estimates of the interfacial Eeff at their average site of residence. For p- nitrophenol, p-(t-buty1)phenol and p-toluidine there will obviously be some ambiguity associated with the Eeff estimates. Tables 1, 2 and 3 contain the pH-titration results obtained for p-nitrophenol, p-(t- buty1)phenol and p-toluidine, respectively.Table 4 contains the pH-titration results of Ptak et aL2' for myristic acid and stearylamine and those of Lovelock et aI.l7 for 4- octadecyloxy-1-naphthoic acid. Also included in all the tables are the ApK," and Eeff estimates. Where the reference ApKA behaviour was not a unique function of solvent dielectric constant, eeff values referring to each calibrating organic solvent-water mixture have been given. For the charged micellar systems the pK," values were determined from the known micellar surface potential^,^^^^^ the P K , " ~ ~ values and eqn (1). The electrostatic surface potential of a DTAB micelle in a 2, 5 and 10 weight % DTAB solution is considered to be + 114, +98 and +91 mV, respectively, whilst in the presence of 4 mol dm-3 NaBr it is considered to be + 18mV.19924 The surface potential of a CTAB micelle in a 0.05 mol dm-3 CTAB solution is considered to be + 141mV.19 With water soluble molecules the possibility exists that the PK,"~~ value is a composite value, comprising contributions from species within the interfacial phase and the bulk aqueous phase.In this study we attempted to avoid this occurrence by using highC. J. Drummond, F. Grieser and T. W. Healy 53 1 Table 1. pH-Titration results for p-nitrophenol in pure water and aqueous micellar solutions with the corresponding ApK," and ceff estimates 'max/nm solution WtYo" ROH (RO-) pKibS APK," Eef? water 0 3 14 (397) 7.15 f 0.05 0 - Brij-35 5 312 (396) 7.86k0.12 0.71 3 8 f 3 (M) 3 7 f 4 (E) Brij-35 10 3 10 (396) 8.08 & 0.17 0.93 35 f 2 (M) 31 f 4 (E) DTAB 2 3 16 (399) 6.27 k 0.06 1 .05c 34 f 1 (M) < 31 (E) DTAB 5 316 (398) 6.45f0.06 0.96' 35f 1 (M) < 31 (E) DTAB 10 315 (397) 6.62k0.03 1.01" 35f 1 (M) < 31(E) DTAB-NaBr (4 mol dm-3) 5 314 (393) 8.01 k0.03 1.16' < 34 (M) <31 (E) a Weight YO surfactant in solution.M or E denotes whether methanol-water or ethanol-water mixtures were used as the reference, respectively. ' Calculated with eqn (1) and the 'yo value estimated by utilizing the ET(30) molecule. l 9 s Z 4 Table 2. pH-Titration results for p-(t-buty1)phenol in pure water and aqueous micellar solutions with the corresponding ApK,O and cerf estimates water 0 274 (291) 10.31 f 0.02 0 - CI2El-4 5 277 (293) 11.96 f 0.05 1.65 33f 1 C12El3 10 276 (293) 12.04 & 0.03 1.73 3 1 f l DTAB 2 276 (296) 9.83 0.02 1 .45c 37f 1 DTAB 5 277 (297) 10.07 f 0.02 1.42' 37+ 1 DTAB 10 275 (296) 10.30 f 0.03 1.53' 35f 1 DTAB-NaBr (4 mol dm-3) 5 277 (297) d - - a Weight YO surfactant in solution.1,4-dioxane-water and ethanol-water reference ApK: curve used. ceff values based on methanol-water all < 34. ' Calculated with eqn (1) and the ry, value estimated by using the ET(30) molecule.1g~24 pK:hs was not well defined varying from 11.45 at a = 0.092 to 10.75 at a = 0.925. concentrations of surfactants. Nevertheless, the PK,"~" results for p-(t-buty1)phenol in 5 wt % DTAB solution with 4 mol dm-3 NaBr and those for p-toluidine in both C,,E, and SDS solution suggest that a high percentage of the species in these systems may not have partitioned into the interfacial phase.Alternatively, the finding that the pK,"bs value of p-toluidine in micellar SDS solution is not well-defined, table 3, can be accounted for if there is specific interaction between the positively charged p-toluidinium ion and a negatively charged sulphate headgroup. Indeed there is some other evidence6,. 63 which suggests that this may be the case. Henceforth the p-toluidine results will be ignored in the discussion. It is clear from the results of tables 1, 2 and 4 that the ApKZ values of p-nitrophenol, p-(t-butyl)phenol, myristic acid, stearylamine and 4-octadecyloxy- 1 -naphthoic acid in532 ' Simple' Acid-Base Equilibria in Micelles Table 3. pH-Titration results for p-toluidine in pure water and aqueous micellar solutions with the corresponding ApK," and eeff estimates 4n,x/nm solution wt Yo" RNH, pK;" APK," &elfb - water 0 285 5.05 f 0.03 0 C12E8 5 288 4.62 + 0.05 - 0.43 59 f 2 (D) 65 k 2 (E) C12E8 10 289 4.39 + 0.02 -0.66 51 f 1 (D) 59f 1 (E) SDS 2 286 6.60 -, 6.89" - - SDS 5 286 6.63 +6.96" - - SDS 10 286 6.28-6.56' - - a Weight YO surfactant in solution.D or E denotes whether 1,4-dioxane-water or ethanol-water mixtures were used as the reference, respectively. ' pKibS not well defined; low a value -, high a value. Table 4. Literature apparent p q results for myristic acid,27 stearylamineZ7 and 4-octadecyloxy- 1 - naphthoic acid17 in aqueous micellar solution with the corresponding ApK: and eePf estimates from this study molecule PK," a pKibS APK: 'eff myristic acid 5.0 6.7b + 1.7 31 f 4 stearylamine 10.7 8.95b - 1.75 40k4 4-octadecyloxy- 1 -naphthoic acid 4.2 6.60" + 2.4 35+ 1 4-octadecyloxy- 1 -naphthoic acid 4.2 4.20d + 2.38" 35f 1 ~~ a Estimated p& value in water based on water soluble homologues.Micellar Triton X-100 solution. 0.005 mol dm-3 Brij-35 or 0.05 mol dm-3 CI2E8 solution. 0.05 mol dm-3 CTAB solution. " Calculated using eqn (1) and the yo value estimated by utilising the &(30 rn01ecule.'~ non-ionic micelles comprising surfactant molecules with poly(ethy1ene oxide) head- groups simply reflect the interfacial solvent properties. These interfacial solvent properties are well characterized by an effective dielectric constant of 35 + 5. Similarly, the calculated ApK: values for p-nitrophenol, p-(t-buty1)phenol and 4- octadecyloxy- 1 -naphthoic acid in cationic micelles comprising surfactant molecules with quaternary ammonium headgroups, are indicative of an interfacial microenvironment which possesses an effective dielectric constant of ca.35. This finding also validates the practice of splitting PK,"~" values into an electrostatic surface potential component and a non-electrostatic component as is done in eqn (1). Moreover it clearly establishes the independence of the two components. Using an identical procedure to the one employed in this work, Fernandez and Fromherzg obtained an eeff value of 32+ 1 for the interfacial microenvironment of non- ionic Triton X-100 micelles. This value was determined by comparing the ApK," values of 7-hydroxy-4-undecylcoumarin and 7-amino-4-heptadecylcoumarin in Triton X- 100 micelles with the Ap& behaviour of their water soluble homologues in 1,4- dioxane-water mixtures.The eeff values gauged in the present investigation are also in agreement with a number of other estimates gained by utilizing totally different techniques. From the solvent-sensitive u.v.-absorption spectrum of solubilized dode- cylpyridinium iodide, Mukerjee et aL6* estimated the Eeff value for the interfacial regionC. J. Drummond, F. Grieser and T. W. Healy 533 of Brij-35 micelles to be 36. This estimate was based on alcohols and alcohol-water mixtures as the reference solvents. Law65 employed the solvent-sensitive fluorescence emission maximum of 2-[6-(2,2-dicyanovinyl)-3,4-dihydro-2,2,4-trimethyl- 1 (2H)-qui- noyllethylbenzoate, with a range of neat organic solvents and alcohol-water mixtures as reference solvents, and estimated that the eeff values for the interfacial micro- environments of Triton X-100 and CTAB micelles are 28 & 8 and 36 & 8, respectively.Utilizing the solvatochromic visible absorption-band maximum of ET(30) and n-alcohols, ethanol-water mixtures and dioxane-water mixtures as reference solvents, both Zachariasse et al." and Drummond et al." have obtained Eeff estimates of 29-36, 28-33 and 27-30 for DTAB, CTAB and C,,E, micelles, respectively. The significance of the E~~~ value characterizing the interfacial region of non-ionic micelles comprising surfactant molecules with poly(ethy1ene oxide) headgroups may be ascertained by comparing the interfacial solvent properties with those of poly(ethy1ene glycol)-water systems.Static dielectric constant measurements of poly(ethy1ene glycoljwater mixtures indicate that bulk solution dielectric constants of 30-40 are attained at an average of 0.5-1 .O water molecules per poly(ethy1ene oxide) subunit." Thus the Eeff values are consistent with the interfacial region of the non-ionic micelles possessing low water activity. It seems likely that the low Eeif values found for the interfacial microenvironment of CTAB and DTAB micelles may also be attributed to a low interfacial water activity.ss~s8 Nevertheless it should be mentioned that the low interfacial E~~~ values may also be a result of the hydrogen-bond donor properties of the water in the interfacial regions being different from that of bulk water, and/or the presence of electrostatic image interactions caused by the proximity of the low dielectric hydrocarbon core.69 Conclusions The acid-base equilibria of a wide range of ' simple ' weak acids and bases residing within the interfacial microenvironment of both non-ionic and charged micelles have been investigated. Concordant estimates have been obtained for the effective dielectric constant of the interfacial region of both non-ionic micelles comprised of surfactant molecules with poly(ethy1ene oxide) headgroups and charged micelles composed of surfactants with quaternary ammonium headgroups. The concordance of the Eeff estimates determined for the same interfacial microenvironment from the acid-base behaviour of the different weak acids and bases indicates that (a) for the molecules and systems examined, with the probable exception of p-toluidine in micellar SDS solution, the differences between the p c and pK," values can primarily be ascribed to the difference between the mean solvent properties of the interfacial phase and the bulk aqueous phase ; (b) interfacial solvent effects on the acid-base equilibria investigated can be likened to those of an organic solvent-water mixture; (c) for the molecules and systems studied, with the probable exception of the p-toluidinium ion in SDS micelles, there is no notable contribution to pK," from any kind of specific molecular interaction within the interfacial region; ( d ) the medium effect on the proton, myH+, can reasonably be approximated by the mean ionic medium effect on HCl, J + ; and (e) for the weak acids and weak bases investigated in this work the logy!JyL, term in eqn (12) is negligibly small.This work was supported by the Australian Research Grants Scheme. References 1 G. S. Hartley and J. W. Roe, Trans. Faraday SOC., 1940, 36, 101. 2 P. Mukerjee and K. Banerjee, J. Phys. Chem., 1964, 68, 3567. 3 S. H. Yalkowsky and G. Zografi, J . Colloid Interface Sci., 1970, 34, 525.534 ' Simple' Acid-Base Equilibria in Micelles 4 F. Tokiwa, A h . Colloid Interface Sci., 1972, 3, 389. 5 M. Montal and C. Gitler, J. Bioenerg., 1973, 4, 363. 6 P. Fromherz and B. Masters, Biochim. Biophys. Acta, 1974, 356, 270. 7 E. P. Bakker, J. C. Arents, J.P. M. Hoebe and H. Terada, Biochim. Biophys. Acta, 1975, 387, 491. 8 N. Funasaki, J. Colloid Interface Sci., 1977, 60, 54. 9 M. S. Fernandez and P. Fromherz, J. Phys. Chem., 1977, 81, 1755. 10 W. L. C. Vaz, A. Nicksch and F. Jahnig, Eur. J. Biochem., 1978, 83, 299. 11 T. Mashimo, I. Ueda, D. D. Shieh, H. Kamaya and H. Eyring, Proc. Natl Acad. Sci. USA, 1979, 76, 12 J. Frahm, S. Diekmann and A. Haase, Ber, Bunsenges. Phys. Chem., 1980, 84, 566. 13 G. Cevc, A. Watts and D. Marsh, Biochemistry, 1981, 20, 4955. 14 S. Lukac, J. Phys. Chem., 1983, 87, 5045. 15 J. Garcia-Soto and M. S. Fernandez, Biochim. Biophys. Acta, 1983, 731, 275. 16 P. Fromherz and R. Kotulla, Ber. Bunsenges. Phys. Chem., 1984, 88, 1106. 17 B. Lovelock, F. Grieser and T. W. Healy, J. Phys. Chem., 1985, 89, 501.18 B. Lovelock, F. Grieser and T. W. Healy, Langmuir, 1986, 2, 443. 19 C. J. Drummond, F. Grieser and T. W. Healy, Faraday Discuss. Chem. Soc., 1986, 81, 95. 20 F. C. Tsui, D. M. Ojcius and W. L. Hubbell, Biophys. J., 1986, 49, 459. 21 G. V. Hartland, F. Grieser and L. R. White, J. Chem. SOC., Faraday Trans. I , 1987, 83, 591. 22 C. J. Drummond and F. Grieser, Photochem. Photobiol., 1987, 45, 19. 23 C. J. Drummond and F. Grieser, Langmuir, 1987, 3, 855. 24 C. J. Drummond, F. Grieser and T. W. Healy, Chem. Phys. Lett., 1987, 140, 493. 25 J. Kibblewhite, C. J. Drummond, F. Grieser and T. W. Healy, J. Phys. Chem., 1987, 91, 4658. 26 C. J. Drummond, F. Grieser and T. W. Healy, J. Phys. Chem., 1988, 92, 2604. 27 M. Ptak, M. Egret-Charlier, A. Sanson and 0.Bouloussa, Biochim. Biophys. Acta, 1980, 600, 387. 28 A. F. Hofmann and A. Roda, J. Lipid Res., 1984, 25, 1477. 29 G. Ackermann, L. Sommer and W. I. Stephen, Pure Appl. Chem., 1985, 57, 845. 30 A. L. Underwood, Anal. Chim. Acta, 1977, 93, 267. 3 1 H. S. Harned and B. B. Owen, The Physical Chemistry of Electrolyte Solutions (Reinhold, New York, 32 C. J. Drummond, Doctoral Dissertation (The University of Melbourne, 1987). 33 R. G. Bates, Determination ofpH: Theory and Practice (Wiley, New York, 2nd edn, 1973), p. 213. 34 R. G. Bates, in Hydrogen-bonded Solvent Systems, ed. A. K. Covington and P. Jones (Taylor and 35 0. Popovych, Crit. Rev. Anal. Chem., 1970, 1, 73. 36 R. G. Bates, Determination of pH: Theory and Practice (Wiley, New York, 2nd edn, 1973), p. 270.37 R. G. Bates, Determination of pH: Theory and Practice (Wiley, New York, 2nd edn, 1973), p. 334. 38 D. Feakins and C. M. French, J. Chem. SOC., 1957, 2581. 39 S. S. Danyluk, H. Taniguchi and G. J. Janz, J. Phys. Chem., 1957, 61, 1679. 40 G. Akerlof, J. Am. Chem. SOC., 1932, 54,4125. 41 F. W. Critchfield, J. A. Gibson and J. L. Hall, J. Am. Chem. SOC., 1953,75, 1991. 42 R. G. Bates and R. A. Robinson, in Chemical Physics of Ionic Solutions, ed. B. E. Conway and R. G. 43 G. H. Parsons and C. H. Rochester, J. Chem. SOC., Faraday Trans. 1, 1974, 70, 1058. 44 K. N. Ganesh, P. Mitra and D. Balasubramanian, J. Phys. Chem., 1982, 86, 4291, and references 45 C. L. De Ligny, H. Loriaux and A. Ruiter, Recueil, 1960, 80, 725. 46 R. A. Robinson and R. G. Bates, J. Res. Nut. Bur. Stand., Sect. A , 1966, 70, 553. 47 R. Gaboriaud, Ann. Chim. (Paris), 1967, 2, 201. 48 J. Juillard, C.R. Acad. Sci., Ser. C, 1969, 268, 2251. 49 R. Thuaire, J. Chim. Phys., 1972, 69, 23. 50 C. C. Panichajakul and E. M. Woolley, Anal. Chem., 1975, 47, 1860. 51 J. M. Sanchez-Ruiz, J. Llor and M. Cortijo, J. Chem. SOC., Perkin Trans. 2, 1984, 2047. 52 B. Gutbezahl and E. Grunwald, J. Am. Chem. SOC., 1953, 75, 559. 53 W. J. Gelsema, C. L. De Ligny and G. F. Visserman, Recueil, 1965, 84, 1129. 54 H. P. Marshall and E. Grunwald, J. Am. Chem. SOC., 1954, 76, 2000. 55 A. L. Bacarella, E. Grunwald, H. P. Marshall and E. L. Purlee, J. Org. Chem., 1955, 20, 747. 56 G. Papanastasiou, G. Stalidis and D. Jannakoudakis, Bull. SOC. Chim. Fr., 1984, 9-10, 255. 57 E. Grunwald and B. J. Berkowitz, J. Am. Chem. SOC., 1951, 73, 4939. 58 H. S. Harned and T. R. Dedell, J. Am. Chem. SOC., 1941, 63, 3308. 59 L. G. Van Uitert and G. G. Haas, J. Am. Chem. SOC., 1953, 75,451. 60 0. Dusart, C. Piekarski and S. Piekarski, J. Chim. Phys., 1975, 72, 97. 61 C. L. De Ligny, Recueil, 1960, 79, 731. 62 T. Yamashita, H. Yano, S. Harada and T. Yasunaga, J. Phys. Chem., 1984, 88, 2671. 51 14. 3rd edn, 1958), p. 716. Francis, London, 1968), p. 49. Barradas (Wiley, New York, 1964), p. 211. therein.C. J . Drummond, F. Grieser and T. W. Healy 535 63 T. Yamashita, M. Sumino, H. Yano, S. Harada and T. Yasunaga, Bull. Chem. SOC. Jpn, 1984, 57, 2352. 64 P. .Mukerjee, J. R. Cardinal and N. R. Desai, in Micellization, Solubilization and Microemulsions, ed. 65 K. Y. Law, Photochem. Photobiol., 1981, 33, 799. 66 K. A. Zachariasse, N. Van Phuc and B. Kozankiewicz, J. Phys. Chem., 1981, 85, 2676. 67 D. Dunstan, Dept. Physical Chemistry, The University of Melbourne, personal communication. 68 M. Gutman, Methods Biochem. Anal., 1984, 30, 1. 69 C. Ramachandran, R. A. Pyter and P. Mukerjee, J . Phys. Chem., 1982, 86, 3198. K. L. Mittal (Plenum Press, New York, 1977), vol. 1, p. 241. Paper 7/00 103G ; Receiiied 2 1st December, 1987

ISSN:0300-9599    

DOI:10.1039/F19898500521

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I , 1989, 85(3), 537-550 Acid-Base Equilibria in Aqueous Micellar Solutions Part 2.-Sulphonephthalein Indicators Calum J. Drummond,t* Franz Grieser and Thomas W. Healy Department of Physical Chemistry, The Uniuersity of Melbourne, Parkville, Victoria, 3052, Australia The acid-base equilibria of three sulphonephthalein indictors, uiz. bro- mocresol green, bromothymol blue and phenol red, in 1,4-dioxane-water mixtures and aqueous solutions of both non-ionic and cationic micelles have been studied. The difference between the acid-base equilibrium constant of a sulphonephthalein indicator in pure water, pK,", and the apparent acid-base equilibrium constant of a sulphonephthalein located entirely within the interfacial microenvironment of a non-ionic micelle has been explained quantitatively in terms of the different intrinsic solvent properties of the two systems.A number of factors are considered to be primarily responsible for the difference between the apparent acid-base equilibrium constant of a sulphonephthalein residing within the interfacial micro- environment of a cationic alkyl quaternary ammonium micelle and its pK," value. These factors are the electrostatic micellar surface potential, the low effective interfacial dielectric constant and either an interfacial ' salt-effect ' or the formation of an ion pair between the negatively charged sulphonate group of a sulphonephthalein and a cationic quaternary ammonium surfactan t headgroup. Sulphonephthalein acid-base indicators have been employed to probe both the electrostatic and solvent properties of the aqueous interfaces of micelles,1-8 pr0teins,~9 polyelectrolytes,1° vesicles1' and microemu1sions.12-'4 Nevertheless, to date very little progress has been made towards clearly and quantitatively establishing all the factors which contribute to the interfacial acid-base equilibrium of a sulphonephthalein indicator.The aqueous acid-base equilibrium of a sulphonephthalein indicator is far more complex than that of a simple weak acid. As shown in fig. 1, several species have been postulated to be involved in the acid-base equilibrium of a sulphonephthalein. Machida et al.15 interpreted the resonance Raman spectra of the sulphonephthalein indicators in aqueous solution at various pH values as indicating the existence of equilibrium (A).Kolthoff and GUSS" interpreted the acid-base behaviour of the sulphonephthaleins in various buffer systems on the basis of equilibrium (B). In Banyai's treatise" on acid-base indicators, the combined species of equilibrium (A) and (B) have been given as contributing to the overall acid-base equilibrium. In the present study, the acid-base equilibria of bromocresol green, bromothymol blue and phenol red in 1,4-dioxane-water mixtures and aqueous solutions of non-ionic micelles comprising surfactant molecules with poly(ethy1ene oxide) headgroups, n-octyl P-D-glucoside (OG) micelles and n-dodecyltrimethylammonium bromide (DTAB) micelles were investigated. The low-pH structures of the three sulphonephthalein indicators are depicted in fig.2. The primary objective of the present study was to obtain t Present address: CSIRO, Division of Chemicals and Polymers, G.P.O. box 4331, Melbourne, Victoria, 3001, Australia. 537538 Sulphonephthalein Acid-Base Equilibria in Micelles + H+ &SOi b s o ; +H+ Fig. 1. Species postulated to be involved in the acid-base equilibrium of a sulphonephthalein indicator. The example given is phenol red. Fig. 2. Structures of the sulphonephthalein indicators investigated. Phenol red: R = R' = R" = R"' = H; Bromocresol green: R = R" = Br, R = CH,, R = H; Bromothymol blue: R = Br, R' = CH,, R" = CH(CH,),, R"' = H. a quantitative assessment of the factors which are responsible for the difference between the apparent acid-base equilibrium constant of an interfacially located sulphonephthalein indicator and its bulk aqueous pK, value (pK,").Experimental Bromocresol green was a guaranteed reagent supplied by Tokyo Kasei Kogyo Co. Bromothymol blue and phenol red were purchased from Ajax Chemicals and Hopkins and Williams, respectively. IUPAC recommended tests were conducted to establish the identity and purity of each of these acid-base indicators.'* UV spectroscopic grade 1,4-dioxane from Fluka was passed through an aluminium oxide column (active neutral Brockmann grade 1 from BDH Chemicals) prior to use. This was done to remove any residual water or peroxides in the 1,4-dioxane. n-Octyl/h-glucopyranoside (OG) was purchased from Sigma Chemical Co. and was used as received. The source and quality of the water and the other chemical reagents employed in this study have been given in Part 1 of this series1' The experimental set-up for performing the pH-titrations was identical to the one that was described in Part 1 of this series.19 All experiments were carried out at 25 "C.C.J . Drummond, F. Grieser and T. W. Healy 539 1.5 0.5 20 40 60 80 1,4 - dioxane/vol % Fig. 3. The pH-meter correction factor, log uDH, as a function of volume% 1,4-dioxane in 1,4-dioxane-water mixtures. In pure water and the aqueous micellar solutions, the negative logarithm of the hydrogen-ion activity was taken as being equal to the pH-meter reading. However, for organic solvent-water mixtures the pH-meter reading is not a direct measure of the negative logarithm of the hydrogen-ion activity. Van Uitert and Haas20 have shown that an empirical calibration can be made so that the pH-meter reading can be converted into the stoichiometric hydrogen-ion concentration.The equation they derived for 1,4- dioxane-water mixtures was (1) where B is the pH-meter reading and log PH is a correction factor which is independent of ionic strength and is attributable to two effects :'O (i) the liquid junction potential being a function of the solvent composition, and (ii) the medium effect on the activity coefficient of the hydrogen ion varying with solvent composition. y y in eqn (1) is the mean activity coefficient for hydrochloric acid, referred to the p a r h l a r 1,4-dioxane-water mixture at infinite dilution. The values for y y can be obtained by interpolation of the values given by Harned and Owen.21 The method employed in this work to obtain the correction factor, log VH, was based on the dilution method which has been described by Sanchez-Ruiz et The experimental procedure involved taking a 100 % aqueous solution of known volume and hydrogen-ion concentration and successively diluting this solution with known volumes of 1,4-dioxane. After each dilution and a sufficient time delay to allow for equilibration, the pH-meter reading, B, was taken.The values of log VH for each of the various 1,4- dioxane-water mixtures were then calculated on the basis of eqn (l), and the tabulated y: values. For the 100% aqueous solution -log[H+] was assumed to be equal to the qiantity [PH - log( 1 /y",]. For each successive 1,4-dioxane-water mixture the stoi- chiometric hydrogen-ion concentration was calculated by taking the dilution and the density of the solution into account. The data of Harned and 0wen2l were used to interpolate the densities of the 1,4-dioxane-water mixtures.As no extra electrolyte was added to the solutions the ionic strength was considered to be equal to the stoichiometric hydrogen-ion concentration and the 77 values were calculated accordingly. In fig. 3 the values obtained for log -VH are plotted as a function of the volume % of -log [H+] = B + log VH + log y y -540 Sulphonephthalein Acid-Base Equilibria in Micelles Table 1. Values of the pH-meter correction factor, log uDH, in 1,4- dioxane-water mixtures at 25 "C vol Yo this ref. ref. ref. ref. 1,4-dioxane work (23) (24) (25) (22) ~ 10 20 30 40 50 60 70 80 0.02 0.01 0.01 0.01 0.02 0.06 0.04 0.02 0.02 0.03 0.09 0.06 0.04 0.04 0.04 0.14 0.11 0.08 0.08 0.08 0.23 0.22 0.20 0.20 0.16 0.41 0.44 0.38 0.40 0.35 0.80 0.85 0.78 0.80 0.71 1.66 - - - - 1,4-dioxane in the 1,4-dioxane-water mixtures.Each log VH value is the average of three separate experiments where the initial stoichiometric hydrogen-ion concentrations were 1.41 x lop3, 3.10 x mol dm-3, respectively. Table 1 shows how the correction factors calculated in this work compare with the values obtained by other researcher~.~~-~~ It should be noted that the correction factors determined by the other researchers were all acquired with the use of a combined electrode whereas in this work a dual electrode system was employed. When determining the various pK, values it was assumed that the acid-base equilibria of the sulphonephthalein indicators could be described by and 6.87 x tcB HIN e H+ + IN where HIN, IN and H+ denote the protonated (acid) and deprotonated (base) forms of the sulphonephthaleins and the proton, respectively.For the sulphonephthalein indicators in aqueous micellar solution, the apparent pK, values were obtained from the change in the visible absorption spectrum of each indicator with bulk aqueous pH by means of the expression with and [IN] - a [HIN] 1-a (4) where A represents the absorbance at the long-wavelength band maximum of the deprotonated form of the sulphonephthalein, A,,,, at a given pH, AHIN the absorbance at A,,, when all the indicator molecules are protonated and 4, the absorbance at A,,, when all the indicator molecules are deprotonated. Representative visible absorption spectra for bromocresol green, bromothymol blue and phenol red in aqueous micellar solutions as a function of bulk aqueous pH can be found in ref.(26). Table 2 gives the positions of the long wavelength absorption band maximum of the sulphonephthaleins in the media investigated.C. J . Drummond, F. Grieser and T. W. Healy 54 1 Table 2. The position of the long wavelength absorption band maximum of the conjugate acid-base forms of the sulphonephthaleins bromocresol bromothymol green blue phenol red medium' Eb IN (HIN) IN (HIN) IN (HIN) Water 78.48 20% D 61.86 40% D 44.54 60% D 27.21 80% D 11.86 5% Brij-35 - 5% OG - 2% DTAB - 5% DTAB - 10 Yo DTAB - - 10% BriJ-35 - 5 YO DTAB-4 mol dmP3 NaBr 613 (438) 618 (432) 621 (428) 622 (422) 623 (416) 624 (422) 623 (420) 622 (428) 623 (420) 623 (418) 623 (418) 625 (41 8) 61 1 (430) 619 (425) 622 (420) 623 (41 5) 624 (409) 621 (414) 622 (41 3) 621 (415) 622 (41 5) 621 (414) 622 (4 12) - 557 (426) 560 (424) 562 (422) 565 (416) 568 (410) 557 (420) 559 (421) 568 (41 7) 568 (416) 568 (416) 566 (414) - a YO refers to weight YO; D denotes 1'4-dioxane.Dielectric constants from ref. (27). The thermodynamic acid-base equilibrium constant for the reaction described by eqn (2) in 1,4-dioxane-water mixtures is given by where ypN and yEIN denote the activity coefficients of the base and acid forms of the sulphonephthalein indicators, respectively, referred to the particular 1,4-dioxane-water mixture at infinite dilution.It is not completely clear how one should approximate the activity coefficients of large complex organic ions such as the sulphonephthaleins.28 Consequently, the pK," values given in this work neglect the activity coefficient term. pH-Titrations in 174-dioxane-water mixtures were carried out with low indicator concentrations, (2.4 f 0.1) x 1 W5 mol dm-3, and with no added electrolyte other than the HCl and NaOH required to perform the titrations. These HCl and NaOH additions were always kept to a minimum. The high pH and low pH spectra were always the final spectra to be taken in an experiment to determine a pK,". By taking into account the changing ion product of water with added dioxane22 it was estimated that, in the pH range where the [IN]/[HIN] ratios were determined, the total added electrolyte never exceeded 1 O-* mol dm-3.At this background ionic strength, the interpolated magnitudes of the mean activity coefficients of HCl in dioxane-water mixtures2' suggest that for the 0, 20, 40 and 60 weight% dioxane-water mixtures, the activity coefficient term in eqn (5) is probably negligibly small. Hence the pK," values of this work in these mixtures are believed to approximate well the thermodynamic acid-base equilibrium constants. Results In Part 1 of this series,l9 it was demonstrated that, when there are no specific molecular interactions or interfacial ' salt-effects ' involved, the influence of a micellar interfacial microenvironment on acid-base equilibria can be likened to that of an organic 19 F A R 1542 4.0 $ 2.0 Sulphonephthalein Acid-Base Equilibria in Micelles 20 40 60 dielectric constant Fig.4. Bromocresol green ApK," (0) and ApKk (0) values as a function of the dielectric constant 4.0 $ 2.0 of 1,4-dioxane-water mixtures. I I I 2 0 40 60 $electric constant Fig. 5. Bromothymol blue ApK," (0) and ApK; (a) values as a function of the dielectric constant of 1,4-dioxane-water mixtures.C. J . Drummond, F. Grieser and T. W. Healy 543 20 40 60 dielectric constant Fig. 6. Phenol red ApK," (0) and ApKa (@) values as a function of the dielectric constant of 1,4-dioxane-water mixtures. solvent-water mixture. Moreover, it was shown how a pKr value can be transformed into a pK: value by accounting for the medium effect on the proton, myH+, i.e.pKL = pK? -log JH+ (6) Providing the interfacial acid-base equilibria of the sulphonephthaleins are unperturbed by any form of specific solute-solvent interaction pKa is related to pK," by Y : N pK; = pK," -log - Y h J (7) where pK," is the pK:bS value measured in non-ionic micellar systems and is defined by eqn (8) in charged micellar systems, i.e." where pK," is the apparent acid-base equilibrium constant of the indicator in the micellar interfacial region if the surface potential, yo, is zero and F, R and Tare the Faraday constant, the universal gas constant and the absolute temperature, respectively. Fig. 4, 5 and 6 display the ApK," (i.e. pK,"-pK,") and ApK: (i.e. pKg-pK,") values of bromocresol green, bromothymol blue and phenol red as a function of the dielectric constant of l,4-dioxane-water mixtures.The data points refer to 0, 20, 40, 60, and 80 weight % dioxane-water mixtures. The pK," values were converted into pK: values by utilizing eqn (6) and the procedure detailed in Part 1 of this series." 19-2544 Sulphonephthalein Acid-Base Equilibria in Micelles Table 3. pH-titration results for bromocresol green in pure water and aqueous micellar solutions mediuma P q h S PK," APK," %ffb ref. water 5% Brij-35 10% BriJ-35 40 mg Brij-58 cmP3 10 mmol dm-3 Triton X-100 10 mmol dm-3 Tween 80 Brij-35 OG 5 % OG 2 % DTAB 5% DTAB 10% DTAB 5 70 DTAB-4 mmol dm-3 NaBr 4.87 f 0.03' 6.22 f 0.05 6.21 f0.09 6.50 f 0.05 6.2 6.0 6.00 5.76 f 0.03 d 3.78 f 0.07 3.93 f 0.05 4.13 f 0.07 5.61 fO.10 6.22 6.2 1 6.50 6.2 6.0 6.00 5.76 d 5.7 1 5.59 5.67 5.91 1.35 1.34 I .63 1.33 1.13 1.13 0.89 0.87d 0.84 0.72 0.80 1.04 38f 1 3 8 f 2 32+ 1 38 42 42 49+ 1 49 50 f 2d 54 f 2d 51 f 2 d 45 f 3d this work this work this work 29 30 30 8 this work 8 this work this work this work this work ~ ~~ YO refers to weight YO surfactant.See text for implicit assumptions. pK,". See discussion in text. Table 4. pH-titration results for bromothymol blue in pure water and aqueous micellar solutions ref. mediuma pKZt)' pK," ApK," E , ~ ( . ~ water 5% Brij-35 10% Brij-35 BriJ-35 OG 2 % DTAB 5 % DTAB 10 Yo DTAB 5 % DTAB-4 mmol dm-3 NaBr 7.41 f0.13' 9.23f0.19 9.20 f 0.07 8.96 d 6.8 1 f 0.08 7.02 f 0.08 7.19 k0.05 8.58 f 0.02 9.23 9.20 8.96 d 8.73 8.68 8.73 8.88 1.82 1.79 1.55 1 .32d 1.32 1.27 1.32 1.47 - this work 4 0 f 4 this work 40 f 2 this work 44 7, 8 49 7, 8 49f2d this work 50 f 2d this work 49f2d this work 46f I d this work _ _ _ _ _ _ _ _ ~ a YO refers to weight % surfactant.text. See text for implicit assumptions. pK r. See discussion in Tables 3-5 summarize the pH-titration results for bromocresol green, bromothymol blue and phenol red, respectively. In addition tables 3 and 4 include some literature pH- titration results for bromocresol green and bromothymol blue in aqueous non-ionic micellar solution^.^^ '* 29* 30 The pK," values found for the sulphonephthalein indicators in pure water are in reasonable agreement with literature values. pK, values of 4.90,173 31 5.033,31 4.98531 and 4.98731 have been reported for bromocresol green in pure water, while for bromothymol blue values of 7.30,17*31 7.199;' 7.341,31 7.5,'O and 7.446,3' and for phenol red values of 7.97,17* 31 8.03531 and 8.08.31 The pK," values of the sulphonephthaleins in the micellar DTAB solutions were determined with the aid of eqn (8) and the known micellar surface potential^.^'-^* All theC.J . Drummond, F. Grieser and T. W. Healy 545 Table 5. pH-titration results for phenol red in pure water and aqueous micellar solutions mediumo pK,ObS PK,O APK &errb water 8.01 0.05" - - - 5% Brij-35 10% Brij-35 8.83f0.16 8.83 0.82 5 3 f 3 d 9.02 & 0.21 9.02 1.01 49*5d 2% DTAB 7.29 & 0.17 9.22 1.21 45f4d 5 % DTAB 7.47 f 0.06 9.13 1.12 47f2' 10% DTAB 7.66 & 0.04 9.20 1.19 45+ld 5 Yo DTAB4 mmol dm-3 NaBr 8.77 * 0.09 9.07 1.06 48f2d a YO refers to weight YO surfactant.text. See text for implicit assumptions. pK;. See discussion in Eeff values were obtained by comparing the ApC values of the sulphonephthaleins with their ApKi values in 1,4-dioxane-water mixtures, fig. 4-6. Note that these E , ~ ~ values are based on three major assumptions. It is assumed that:19 ( i ) both the protonated and deprotonated forms of the sulphonephthalein indicators have quantitatively partitioned into the interfacial phase; (ii) the activity coefficient term in eqn (7) is negligibly small so that the ApK," values are directly comparable with the ApK: behaviour in different solvent dielectric-constant bulk media ; and (iii) there is no contribution to the interfacial acid-base equilibrium from specific molecular interactions.All the literature pH-titrations were performed in the presence of extraneous inert electrolyte. The acid-base equilibrium of an indicator located within the interfacial microenvironment of a non-ionic micelle comprising surfactant molecules with poly(ethy1ene oxide) headgroups appears to be unaffected by the presence of inert electrolyte in the bulk aqueous ~ o l u t i o n . ~ ~ . ~ ~ Therefore it was felt that the literature pKibs data obtained in Brij-58, Brij-35, Triton X-100 and Tween 80 (a polyoxyethylene sorbitan ester) micellar solutions could be analysed. The acid-base equilibrium of an indicator residing within the interfacial microenvironment of non-ionic micelles comprising surfactant molecules with saccharide headgroups is definitely influenced by the presence of electrolyte in the bulk aqueous However, in this system the 'salt- effect' on the interfacial acid-base equilibrium appears to mimic the ' salt-effect ' on the bulk aqueous acid-base eq~ilibrium.~~ Thus for the literature results in OG solution it was considered appropriate to compare the ApKa values with the ApKZ values derived from the pK," and pKibs values obtained in identical electrolyte solutions, rather than using the pK," value measured in pure water.Discussion Non-ionic Micelles The pKzbs values of bromocresol green and bromothymol blue situated within the interfacial microenvironment of non-ionic micelles comprising surfactant molecules with poly(ethy1ene oxide) headgroups can be quantitatively accounted for by assuming that the interfacial microenvironment of these micelles possesses an effective dielectric constant of 38f6.This Eeff value, which characterizes the influence of the interfacial solvent properties on the acid-base equilibria of bromocresol green and bromothymol blue, is in close agreement with the Eeff estimate gained from the acid-base behaviour of 'simple' weak acids and weak bases located within this type of interfacial regi011.l~ Thus it can be inferred that the log (y:N/&IN) term does not contribute significantly to the apparent acid-base equilibrium constants of bromocresol green and bromothymol blue546 Sulphonephthalein Acid-Base Equilibria in Micelles residing within the interfacial region of non-ionic micelles comprising surfactant molecules with poly(ethy1ene oxide) headgroups.The agreement also indicates that there is no marked contribution to the pKzbs values of bromothymol blue and bromocresol green in these micellar systems arising from specific molecular interactions. The seemingly anomalous phenol red Eeff values are most likely due to the highly water-soluble deprotonated form of phenol red not partitioning quantitatively into the non-ionic interfacial phase. This contention is supported by the observation that the A, value of the deprotonated form of phenol red in the 5 wt YO Brij-35 solution is the same as the A,,, value in pure water, and in the 10 wt YO Brij-35 solution there is still very little difference between the Amax value and the Amax value in pure water, table 2. In contrast with this observation, the A,,, values of the deprotonated forms of bromocresol green and bromothymol blue in the Brij-35 solutions are markedly different to their A, values in pure water, table 2.If it is assumed that the pK," value of phenol red, with both forms close to being fully solubilized within the interfacial microenvironment of Brij-35 micelles, is ca. 0.5 units greater than the pK," value in DTAB solution, as is the case for bromocresol green and bromothymol blue, a ApK," value of ca. 1.7 and a corresponding eeff estimate of ca. 35 is obtained for phenol red in micellar Brij-35 solution. Intriguingly, Gutman et aL2' found it necessary to propose a reaction scheme involving two independent steps in order to explain their kinetic data for the rates of protonation and deprotonation of bromocresol green molecules located within the interfacial region of Brij-58 micelles. The following reaction sequence was proposed : kr H; + IN, f HINi kd kin HIN, e HIN, (9) k0"t where subscript c denotes a location within a more hydrophobic region.The relationship between the observed acid-base equilibrium constant and the various rate constants was given by (1 1) KL = - kd kr and As embodied in reactions (9) and (lo), the proposed reactions were a diffusion- controlled reaction between a proton and an interfacially located bromocresol green molecule followed by translocation of the prototropic part of the molecule to a more hydrophobic region, c, in the micelle. For example, with bromocresol green, in an aqueous 40 mg Brij-58 per ml of 100 mmol 1-1 KC1 solution, Gutman et al.29 measured a pKPbs value of 6.50+0.05 and the kinetic parameters given to fit eqn (12) were pKi = 5.5 and kin/kOut = 9.This two-step reaction scheme is at odds with the present findings. The concordance between the ceif estimate derived from the acid-base behaviour of bromocresol green in Brij-58 micellar solution, which is given in table 3, and the estimates gauged from the acid-base behaviour of many other molecules in similar non-ionic micellar 3 6 9 37 indicates that if there is a two-step reaction sequence governing the interfacial acid-base equilibrium of bromocresol green it must also occur in 1,4- dioxane-water mixtures. Therefore any form of translocation reaction is precluded. In addition, a translocation reaction such as (10) seems highly unlikely because molecules with a similar hydrophobicity and size to the sulphonephthaleins have been deduced38* 39 to rotate, relatively unhindered, within the interfacial region of micelles.C.J . Drummond, F. Grieser and T. W. Healy 547 Interestingly, the pKibS values measured in micellar OG solutions suggest that the interfacial microenvironments of these micelles comprising surfactant molecules with monosaccharide headgroups is characterized by an E~~~ value of 49f 1. Thus it appears that the interfacial region of OG micelles is more 'aqueous-like' in nature than the interfacial region of non-ionic micelles comprising surfactant molecules with poly(ethy1ene oxide) headgroups. A similar interfacial Eeff estimate for OG micelles has been ascertained from the acid-base behaviour of both neutral red and methyl orange.36* 37 The Eeff estimate derived from the acid-base behaviour of bromocresol green and bromothymol blue residing within the interfacial microenvironment of OG micelles can also be compared with estimates obtained by employing other investigative techniques.Mukerjee and c o - w o r k e r ~ ~ ~ ~ ~ have reported Eeff values for the interfacial region of OG micelles, based on the absorption spectrum of micelle-solubilized dodecylpyridinium iodide and 2,2,6-tetramethylpiperidin- 1 -yloxy with n-alcohols and alcohol-water mixtures as the reference media, of 45 1 and 40, respectively. The slight differences between the Eeff estimates may be due to slightly different solute-solvent interactions in each case or alternatively slightly different average residential sites for each type of probe within the interfacial region.Charged Micelles As shown in tables 3 and 4, the pK," value for a quantitatively micelle-solubilized sulphonephthalein indicator in Brij-35 solution is not the same as the pK," value in DTAB solution. In this section, a number of possible explanations for the disparity between the two types of pK," values are assessed. In view of the overwhelming amount of evidence in support of a reasonably similar interfacial ceff value for DTAB and Brij-35 type m i c e l l e ~ , ~ ~ ' ~ ~ the explanation that the DTAB Eeff values are as given in tables 3-5 was considered to be improbable. For the remaining alternative explanations it is assumed that the Eeff for the Brij-35 micellar interface is similar to the Eeff for the DTAB micellar interface.One can postulate two types of specific molecular interactions between a cationic micellized surfactant headgroup and an overall negatively charged sulphonephthalein indicator to account for the disparity between the Brij-35 pK," values and the DTAB pK: values. Specifically, these are the formation of an ion pair between a cationic quaternary ammonium surfactant headgroup and either the ' phenoxide-like' proto- tropic part of a sulphonephthalein indicator or the sulphonate group. The formation of an ion pair between a cationic quaternary ammonium surfactant headgroup and the ' phenoxide-like' prototropic part of a sulphonephthalein, could further stabilize the deprotonated forms of the sulphonephthaleins within the interfacial region of the DTAB micelles.However, this type of interaction, if present, does not appear to influence significantly the interfacial acid-base equilibria of any other molecules possessing a similar kind of prototropic group.lg* 32, 41* 43 Therefore, specific molecular interaction of this type is not considered to be responsible for the observed pK," difference. The negative charge of the sulphonate group in a sulphonephthalein indicator has an effect on the acid-base equilibrium of the indicator. The quantitative formulation of the electrostatic effect of a neighbouring charged group on the pK, of the prototropic moiety of a 'simple' weak acid may be given as41,44-46 (13) z , e2 2.303 Ddk, T pK, = pK," - where pKi is the intrinsic pK, of the prototropic moiety, z , is the charge of the neighbouring group, D is the effective dielectric constant separating the charged group and the prototropic moiety, d is the distance between the charged group and the548 Sulphonephthalein Acid-Base Equilibria in Micelles prototropic moiety and the other parameters have their usual definition.The resonance hybrid nature of the sulphonephthalein indicators, fig. 1, means that eqn (1 3) cannot be employed to determine quantitatively the effect of the negative sulphonate group on the pK, of a sulphonephthalein indicator. However, eqn (13) can be used to discuss qualitatively what effect the formation of an ion pair between a cationic quaternary ammonium headgroup and the sulphonate group of a sulphonephthalein indicator would have on the apparent pK, of the indicator.The ion-pairing of the sulphonate group of a sulphonephthalein indicator with a quaternary ammonium headgroup would result in the charge on the sulphonate group being electrically neutralized. From eqn (13) it is clear that the intrinsic interfacial pK,, pK,", value of an indicator ion-paired in this manner would be less than the pK," value of the un-complexed indicator. Therefore the formation of an ion pair between the negatively charged sulphonate group of a sulphonephthalein indicator and a cationic quaternary ammonium headgroup is totally consistent with the observed difference between the Brij-35 pK," value and the DTAB pK," value for a particular sul- phonephthalein indicator.This, however, is not the only explanation which is consistent with the results. By analogy with the effect of salt on the pKa values of the sulphonephthalein indicators in aqueous solution, one expects the magnitude and sign of the log (yiN/yLIN) term for a sulphonephthalein residing within the interfacial phase of a DTAB micelle, where there is a high effective counter-ion and headgroup concentration, to be relatively large and negati~e.~',~' Indeed the present study's finding, that the pK," values of bromocresol green and bromothymol blue in Brij-35 micellar solutions are greater than their pK," values in DTAB micellar solutions, is in complete agreement with this expectation. Therefore the difference between the pK," values obtained for a sulphonephthalein in Brij-35 and DTAB micellar solutions can also be rationalized in terms of an interfacial 'salt-effect' on the activity coefficients of the protonated and deprotonated forms of a sulphonephthalein indicator when it is located within the interfacial region of a DTAB micelle.Note that the perturbation of the interfacial acid-base equilibria by either an interfacial 'salt-effect' or the formation of an ion pair means that the pK," value of a sulphonephthalein residing within the interfacial region of a cationic micelle comprising quaternary ammonium surfactant molecules cannot be compared validly with the ApKa behaviour as a function of solvent dielectric constant to determine an interfacial E~~~ value. It also means that the value measured for a sulphonephthalein residing within the interfacial microenvironment of a non-ionic micelle comprising surfactant molecules with poly(ethy1ene oxide) headgroups cannot serve as a reasonable substitute for the pK," value of the same sulphonephthalein in a cationic m i ~ e l l e .~ , ~ Conclusions In spite of the complexity of the postulated aqueous acid-base behaviour of the sulphonephthalein indicators, which is depicted in fig. 1, the interfacial acid-base equilibria of bromocresol green, bromothymol blue and phenol red in a number of aqueous micellar solutions has proved amenable to analysis. The difference between the pK," of a sulphonephthalein indicator and the apparent acid-base equilibrium constant, PK:~', of a sulphonephthalein residing within the interfacial microenvironment of a non-ionic micelle has been attributed to the different intrinsic solvent properties of the two systems.Estimates of the E , ~ ~ of the non-ionic interfacial regions have been made by comparing modified pKa values for the sulphonephthaleins in 1,4-dioxane-water mixtures of known dielectric constant with the pKibS values. The interfacial microenvironment of non-ionic micelles comprising surfactant molecules with poly(ethy1ene oxide) headgroups is characterized by an E,,,., ofC. J . Drurnrnond, F. Grieser and T. W. Healy 549 38f6. This ceff estimate is in reasonable agreement with the estimates determined from the acid-base behaviour of other acid-base indicators in this type of interfacial system.’’* 36* 37 The interfacial microenvironment of OG micelles appears to be more ‘aqueous-like’ in nature and is characterized by an Eeff of 49f 1.Several factors have been accredited with being primarily responsible for the difference between a sulphonephthalein’s pK,” value and its pKibs value in micellar DTAB solution. These factors are the electrostatic surface potential, the low interfacial ceff and either the formation of an ion pair between the negatively charged sulphonate group of a sulphonephthalein indicator and a cationic quaternary ammonium surfactant headgroup or an interfacial ‘salt-effect ’. This work was supported by the Australian Research Grants Scheme. References 1 P. Mukerjee and K. Banerjee, J. Phys. Chem., 1964, 68, 3567. 2 C. E. Williamson and A. H. Corwin, J .Colloid Interface Sci., 1972, 38, 567. 3 M. Montal and C. Gitler, J. Bioenergetics, 1973, 4, 363. 4 J. V. Moller and U. Kragh-Hansen, Biochemistry, 1975. 14, 2317. 5 N. Funasaki, Nippon Kagaku Kaishi, 1976, 5, 722. 6 N. Funasaki, J . Colloid Interface Sci., 1977, 60, 54. 7 P. Mukerjee, J. R. Cardinal and N. R. Desai, in Micellization, Solubilization and Microemulsions, ed. 8 N. R. Desai, Doctoral Dissertation (University of Wisconsin, Madison, 1981). 9 C. E. Williamson and A. H. Corwin, J. Colloid Interface Sci., 1972, 38, 577. K. L. Mittal (Plenum Press, New York, 1977), vol. 1, p. 241. 10 E. Baumgartner, R. Fernandez-Prini and D. Turyn, J . Chem. SOC., Faraday Trans. I , 1974, 70, 11 T. Mashimo, I. Ueda, D. D. Shieh, H. Kamaya and H. Eyring, Proc. Natl Acad. Sci.USA, 1979, 76, 12 R. A. Mackay, K. Jacobsen and J. Tourian, J. Colloid Interface Sci., 1980, 76, 515. 13 H. Fujii, T. Kawai and H. Nishikawa, Bull. Chem. SOC. Jpn, 1979, 52, 2051. 14 H. Fujii, T. Kawai, H. Nishikawa, and G. Ebert, Colloid Polym. Sci., 1982, 260, 697. 15 K. Machida, H. Lee and T. Uno, J. Raman Spectrosc., 1979, 8, 172. 16 I. M. Kolthoff and L. S . Guss, J. Am. Chem. Soc., 1938, 60, 2516. 17 E. Banyai, in Indicators, ed. E. Bishop (Pergamon Press, Oxford, 1972), chap. 3. 18 G. Ackermann, L. Sommer and W. 1. Stephen, Pure Appl. Chem., 1985, 57, 845. 19 C. J. Drummond, F. Grieser and T. W. Healy, J. Chem. SOC., Faraday Trans. I , 1989, 85, 521. 20 L. G. Van Uitert and C . G. Haas, J. Am. Chem. Soc., 1953, 75, 541. 21 H. S. Harned and B. B. Owen, The Physical Chemistry of Electrolyte Solutions (Reinhold, New York, 22 J.M. Sanchez-Ruiz, J. Llor and M. Cortijo, J. Chem. SOC., Perkin Trans. 2, 1984, 2047. 23 H. M. N. H. Irving and U. S . Mahnot, J. Inorg. Nucl. Chem., 1968, 30, 1215. 24 J. P. Shukla and S. G . Tandon, J . Electroanal. Chem., 1972, 35, 423. 25 Y. K. Agrawal, Talanta, 1973, 20, 1354. 26 C. J. Drummond, Doctoral Dissertation (University of Melbourne, 1987). 27 F. W. Critchfield, J. A. Gibson and J. L. Hall, J. Am. Chem. Soc., 1953, 75, 1991. 28 R. G. Bates, Determination of p H : Theory and Practice (Wiley, New York, 2nd edn, 1973), p. 145. 29 M. Gutman, E. Nachliel, E. Gershon and R. Giniger, Eur. J. Biochem., 1983, 134, 63. 30 M. Gutman, D. Huppert, E. Pines and E. Nachliel, Biochim. Biophys. Acta, 1981, 642, 15. 31 M. Meloun and S. Kotrly, Collect. Czech. Chem. Commun., 1977, 42, 21 15. 32 C. J. Drummond, F. Grieser and T. W. Healy, Faraday Discuss. Chem. SOC., 1986, 81, 95. 33 C. J. Drummond, F. Grieser and T. W. Healy, Chem. Phys. Lett., 1987, 140, 493. 34 J. Kibblewhite, C . J. Drummond, F. Grieser and T. W. Healy, J. Phys. Chern., 1987, 91, 4658. 35 G. G. Warr, C. J. Drummond, F. Grieser, B. W. Ninham and D. F. Evans, J . Phys. Chem., 1986, 90, 36 C. J. Drummond, F. Grieser and T. W. Healy, J. Chem. SOC., Faraday Trans. I , 1989, 85, 551. 37 C. J. Drummond, F. Grieser and T. W. Healy, J. Chem. SOC., Faraday Trans. I , 1989, 85, 561. 38 W. Reed, M. J. Politi and J. H. Fendler, J. Am. Chem. SOC., 1981, 103, 4591. 39 M. J. Politi and J. H. Fendler, J. Am. Chem. SOC., 1984, 106, 265. 40 C. Ramachandran, R. A. Pyter and P. Mukerjee, J . Phys. Chem., 1982, 86, 3198. 41 C. J. Drummond, F. Grieser and T. W. Healy, J. Phys. Chem., 1988, 92, 2604. 1518. 51 14. 3rd edn, 1958). 458 1.550 Sulphonephthalein Acid-Base Equilibria in Micelles 42 F. Grieser and C. J. Drummond, J . Phys. Chem., 1988, 92, 5580. 43 C. J. Drummond and F. Grieser, Photochem. Photobiol., 1987, 45, 19. 44 J. T. Edsall and J. Wymann, Biophysical Chemistry (Academic Press, New York, 1958), vol. 1 , chap. 8. 45 S. H. Yalkowsky and G. Zografi, J. Pharm. Sci., 1970, 59, 798. 46 E. E. Mehler and G. Eichele, Biochemistry, 1984, 23, 3887. Paper 7/00109F; Received 21st December, 1987

ISSN:0300-9599    

DOI:10.1039/F19898500537

出版商:RSC    

年代:1989

数据来源: RSC