摘要:

physicochemical topics, thereby encouraging scientists of different disciplines to contribute their varied viewpoints to a coiiinion theme. A recent Discussion is :- The Royal Soci- of Chemistry- No.75 lntraamolecwlar Kinetics No. 75 in the series, this publication is the result of a general discussion held at the University of Warwick in April 1983. Contents: The Spiers Meniorlal Lecture; Vibrational Redistribution within Excited Electronic States of Polyatomic Molecules Inrraniolecular R e h u t i o n o f 1. vcited States lsomerization of Intcrnal~ncrgy-selected Ions Kinetics of Ion-Molecule Collision Coinple\es in the Gas Phase, E\periinent and Theory lntrainolccular Decay 01' Soinc Open-shell Pulya t o niic Ca lions On tlic Theory u i Iiitrdniolccul~r I n e r g y Transfer Pulsed Laser Preparation and Ouaniuin Superposition Statc Evolution in ReguLtr and Irregular Systems A Ouantuiii-iiicclianical Internal-collision Model for State-sclcctcd Uniinolccular Decoiiiposilio n The Correspondence Principle and Intramolecular Dynamics lntrainoleculdr Dcphasiiig.t'icusecond Evolution of Wavepacket States in a Molecule with Int erinediate-casc level Struct urc Energy Conversion in van der Waals C'u~~iplc\c\ ol s-Tetrarine and Argon Tim-dependent Processes in Polyatuinic Molecules During and After Intense Intrarcd Irradiation Energy Distributions in tlic (.N(X'L+) bragnient froiii tlie Infrared Multiplepholun Dissociation ol' CI. ICN. A Coinparison between 1:xperiiiiental Results and the Predictions ot Statistical Theories of ChFO + Product Energy Partitioning in the Decoiii- position of State-selectively Excited HOON and IIOOD Low-power Inl-rarcd Laser I'hoiolysis o f Tetramethy ldioxetan Uniinolecular Reactions lnduccd by Vibrational Overtone Excitation Uniiiiolecular Decomposition of t-Butylhydro- peroxide by Direct Excitation of the 6-0 0-11 Stretching Overtone I'icosecond-jet Spectroscopy and Photoclieinistry.Energy Redistribution and its Iiiipact'on Coherence, Isoincrization, Ihssociatiun and Solvalioii knergy Redistribution in Large Molecules. Duect St ud y o f In1 rainolucular Rehxa lion in the Gas Phase with Picosecond Gating Rotation-dependent Intrainolecuhr I'r~)cessc.sofSO:(A'A.) in a Superwnic Jct Role of Rotation-Vibration Interaction in Vibrational Keh\ation. Energy Kcdistribution in k,xcitcd Singlet I~'ornialdc1iyde Sub-lhppler.Spectroscopy of Benrcnc in the "('liaiinel-lliree" Region Intraiiiulccular 1:lectronic Kclau~tion and I'liotois~)iiieruati[)n Processes in tlie lsuhted Azabenrene Molecules Pyridinc, Pyrazinc and I'yriiiiidinc Softcover 434pp 0 85186 658 1 Price f25.00 ($48.00) Rest of the World f26.00 RSC Members f 16.25 Faraday Discussions of the Chemical Society 7< lnrruniolei u h r Kincrit I Faraday Symposia are usually held annually and are confined to more specialiscd topics than Discussions, with particular reference to recent rapidly developing lines of rescuch. A recent Symposium is :- No.l?The Hydrophobic Interadion No. 17 in the series, this publication is the result of a symposium on The Hydrophobic Interaction held at the Uiiiversity of Reading in December 1982.Contents: Hydrophobic Interdctionr a llistaric.11 Per spect ivr llydrupliobic Ilydration Geometric Kelaution in Water. Its Role in Precise Vapour-pressure Measureiiients of the SolubilkdtiorI of Benzene by Aqueous Sodiuiii Octylsulphate Solutions Nuclear Magnetic Resonance R e b u t i o n Investigation of Tetrahydrofuran and Methyl Iodide Clathrdtes Infrared and Nuckar Magnetic Kcwnance Studies Pertaining to the (age Model t o r Solutions oS Acetone in Water Irothernial Transport Properties in Solutions o f S y mmet r ica I Tet ra-alk y hmnioniuiii Bromides Thermodynamics of Cavity I'oriiiaiion in Water. A Molecular Dynamics Study Molecular Librations and Solvent Oricnt- ational Correlations in Hydrophobic Phenomena Monte Carlo Computer Siniulation Study of the Hydrophobic Effect.Potential ot Mean Force for ECfir)gaq at 25 and SOv C Hydroplicibic Moments and Protein Structure Application 01' the Kirkwood-Buff Theory to the t'roblcin 01 Hydrophobic Interactions Ihentangleinent of Ilydrophubic and IFlcctrosi~tic Contributions t o the I.ilni Pressures O i Ionic Surfactants llydrophobir. Intcracliuns in Dilute Su lut io ns u t 1'0 1 y (vin y I a Ico lio I) ('onioriii;tiionaI and 1:unc.i ional I'ropertics of tiaeiiwglobin in Water+Alcohol Mixtures. Dependence o f Bull. Electrostatic and tlydrupliohic I n t c r x t i o n s upon ptl and KCI concentrations Softcover 24Opp 0 85186 668 9 Price f36.50 ($70.00) Rest of the World f38.50 RSC Members f 23.75 ORDERING RSC Members should send their orders to: The Royal Society of Chemistry.The Membership Officer. 30 Russell Square, Non-RSC Members The Royal Society of Chemistry, Distribution Centre, Blackhorse Road, L London WC1 B 5DT. Letchworth, Herts SO6 IHN, England. Faradaj Symposia of the Chemical Society hGi 17 I hc HI drophohr' Inrcrm rron 1 9 X ? (viii)physicochemical topics, thereby encouraging scientists of different disciplines to contribute their varied viewpoints to a coiiinion theme. A recent Discussion is :- The Royal Soci- of Chemistry- No.75 lntraamolecwlar Kinetics No. 75 in the series, this publication is the result of a general discussion held at the University of Warwick in April 1983. Contents: The Spiers Meniorlal Lecture; Vibrational Redistribution within Excited Electronic States of Polyatomic Molecules Inrraniolecular R e h u t i o n o f 1.vcited States lsomerization of Intcrnal~ncrgy-selected Ions Kinetics of Ion-Molecule Collision Coinple\es in the Gas Phase, E\periinent and Theory lntrainolccular Decay 01' Soinc Open-shell Pulya t o niic Ca lions On tlic Theory u i Iiitrdniolccul~r I n e r g y Transfer Pulsed Laser Preparation and Ouaniuin Superposition Statc Evolution in ReguLtr and Irregular Systems A Ouantuiii-iiicclianical Internal-collision Model for State-sclcctcd Uniinolccular Decoiiiposilio n The Correspondence Principle and Intramolecular Dynamics lntrainoleculdr Dcphasiiig. t'icusecond Evolution of Wavepacket States in a Molecule with Int erinediate-casc level Struct urc Energy Conversion in van der Waals C'u~~iplc\c\ ol s-Tetrarine and Argon Tim-dependent Processes in Polyatuinic Molecules During and After Intense Intrarcd Irradiation Energy Distributions in tlic (.N(X'L+) bragnient froiii tlie Infrared Multiplepholun Dissociation ol' CI.ICN. A Coinparison between 1:xperiiiiental Results and the Predictions ot Statistical Theories of ChFO + Product Energy Partitioning in the Decoiii- position of State-selectively Excited HOON and IIOOD Low-power Inl-rarcd Laser I'hoiolysis o f Tetramethy ldioxetan Uniinolecular Reactions lnduccd by Vibrational Overtone Excitation Uniiiiolecular Decomposition of t-Butylhydro- peroxide by Direct Excitation of the 6-0 0-11 Stretching Overtone I'icosecond-jet Spectroscopy and Photoclieinistry. Energy Redistribution and its Iiiipact'on Coherence, Isoincrization, Ihssociatiun and Solvalioii knergy Redistribution in Large Molecules.Duect St ud y o f In1 rainolucular Rehxa lion in the Gas Phase with Picosecond Gating Rotation-dependent Intrainolecuhr I'r~)cessc.sofSO:(A'A.) in a Superwnic Jct Role of Rotation-Vibration Interaction in Vibrational Keh\ation. Energy Kcdistribution in k,xcitcd Singlet I~'ornialdc1iyde Sub-lhppler. Spectroscopy of Benrcnc in the "('liaiinel-lliree" Region Intraiiiulccular 1:lectronic Kclau~tion and I'liotois~)iiieruati[)n Processes in tlie lsuhted Azabenrene Molecules Pyridinc, Pyrazinc and I'yriiiiidinc Softcover 434pp 0 85186 658 1 Price f25.00 ($48.00) Rest of the World f26.00 RSC Members f 16.25 Faraday Discussions of the Chemical Society 7< lnrruniolei u h r Kincrit I Faraday Symposia are usually held annually and are confined to more specialiscd topics than Discussions, with particular reference to recent rapidly developing lines of rescuch.A recent Symposium is :- No.l?The Hydrophobic Interadion No. 17 in the series, this publication is the result of a symposium on The Hydrophobic Interaction held at the Uiiiversity of Reading in December 1982. Contents: Hydrophobic Interdctionr a llistaric.11 Per spect ivr llydrupliobic Ilydration Geometric Kelaution in Water. Its Role in Precise Vapour-pressure Measureiiients of the SolubilkdtiorI of Benzene by Aqueous Sodiuiii Octylsulphate Solutions Nuclear Magnetic Resonance R e b u t i o n Investigation of Tetrahydrofuran and Methyl Iodide Clathrdtes Infrared and Nuckar Magnetic Kcwnance Studies Pertaining to the (age Model t o r Solutions oS Acetone in Water Irothernial Transport Properties in Solutions o f S y mmet r ica I Tet ra-alk y hmnioniuiii Bromides Thermodynamics of Cavity I'oriiiaiion in Water.A Molecular Dynamics Study Molecular Librations and Solvent Oricnt- ational Correlations in Hydrophobic Phenomena Monte Carlo Computer Siniulation Study of the Hydrophobic Effect. Potential ot Mean Force for ECfir)gaq at 25 and SOv C Hydroplicibic Moments and Protein Structure Application 01' the Kirkwood-Buff Theory to the t'roblcin 01 Hydrophobic Interactions Ihentangleinent of Ilydrophubic and IFlcctrosi~tic Contributions t o the I.ilni Pressures O i Ionic Surfactants llydrophobir. Intcracliuns in Dilute Su lut io ns u t 1'0 1 y (vin y I a Ico lio I) ('onioriii;tiionaI and 1:unc.i ional I'ropertics of tiaeiiwglobin in Water+Alcohol Mixtures. Dependence o f Bull. Electrostatic and tlydrupliohic I n t c r x t i o n s upon ptl and KCI concentrations Softcover 24Opp 0 85186 668 9 Price f36.50 ($70.00) Rest of the World f38.50 RSC Members f 23.75 ORDERING RSC Members should send their orders to: The Royal Society of Chemistry. The Membership Officer. 30 Russell Square, Non-RSC Members The Royal Society of Chemistry, Distribution Centre, Blackhorse Road, L London WC1 B 5DT. Letchworth, Herts SO6 IHN, England. Faradaj Symposia of the Chemical Society hGi 17 I hc HI drophohr' Inrcrm rron 1 9 X ? (viii)

ISSN:0300-9599    

DOI:10.1039/F198480FX029

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

AUTHOR INDEX xxxv Sabbatini, L., 1029 Sacco, A., 2669 Sanders, J. V., 571 Sangster, D. F., 291 1 Sarka, K., 521 Sasahira, A., 473 Sasse, W. H. F., 571 Satchell, P. W., 2395 Sato, K., 841 Sato, Y., 341 Savino, V., 759 Sayers, C. M., 1217 Schiller, R. L., 1257 Schmidt, J., 1 Schmidt, P. P., 2017 Schneider, H., 3275, 3285 Schulz, R. A., 489, 1323 Scott, J. M. W., 739, 1651, 2287, Scott, S. K., 3409 Segall, R. L., 2609 Sehested, K., 2929, 2969 Seidl, V., 1367 Sem, P., 297 Serratosa, J. M., 2225 Seyama, H., 237 Seyedmonir, S. R., 87, 2269 Shanahan, M. E. R., 37 Sheppard, A., 2999 Sherwood, P. M. A., 135, 2099, Shindo, Y., 879, 2199 Shiotani, H., 2145 Shizuka, H., 383, 341 Siekhaus, W. J., 61 Sircar, S., 1101, 2489 Smart, R. St C., 2957, 2609 Smith, I. M., 3021 Smith, R., 3233 Snow, R.L., 3463 Solar, S., 2929 Solar, W., 2929 Solymosi, F., 1841 Soma, M., 237 Soupart, J-B., 3209 Sourisseau, C., 3257 Spink, J. A., 3469 Spoto, G., 1875, 1891 Spotswood, T. M., 3147 Staricco, E. H., 2631 Stassinopoulou, K., 3095 Stedman, D. H., 285 Stout, D. R., 3481 Strohbusch, F., 1757 Strumolo, D., 1479 Struve, P., 813, 2167 Styring, M. G., 3051 Subramanian, R., 2405 2881, 3359 2549, 2867 Sundar, H. G. K., 3491 Sutcliffe, L. H., 669, 3021 Sutton, H. C., 2301 Sutton, L. E., 635 Suzuki, H., 803 Suzuki, T., 1925, 3157 Symons, M. C. R., 423, 1005, Szamosi, J., 1645 Szczepaniak, W., 2935 Takagi, Y., 1925 Takahashi, K., 803 Takahashi, N., 629 Takanaka, J., 941 Takao, S., 993 Takasaki, S., 803 Takegami, H., 1221 Tam, S-C., 2255 Tamamushi, R., 2751 Tamaru, K., 29, 1567, 1595 Tamilarasan, R., 2405 Tanabe, S., 803 Tanaka, K., 2563,2981 Tanaka, T., 119 Taniewska-Osinska, S., 1409 Tascon, J.M. D., 1089 Teo, H. H., 981, 1787 Tetenyi, P., 3037 Thomas, J. K., 1163 Thompson, L., 1673 Thomson, M., 1867 Thomson, S. J., 1689 Tiddy, G. J. T., 789, 3339 Tittarelli, P., 2209 Tominaga, T., 941 Tomkinson, J., 225 Tonelli, C., 1605 Toprakcioglu, C., 13,413 Tran, T., 1867 Trasatti, S., 913 Tripathi, A. D., 1517 Tronc, E., 2619 Troncoso, G., 2127 Truscott, T. G., 2293 Tsurusaki, T., 879 Tuck, J. J., 309 Turner, P. S., 2609 Tusk, M., 1757 Tvarbikova, Z., 2639 Tyrrell, H. J. V., 1279 Ueki, Y.. 341 Ueno, A., 803 Unno, H., 1059 Valencia, E., 2127 van de Ven, T. G. M., 2677 van Ommen, J. G., 2479 van Truong, N., 3275, 3285 Vargas, I., 1947 2767, 2803, 21 1, 1999 Vedrine, J.C., 1017 Veith, J., 2313 Velasco, J. R., 3429 Vesala, A., 2439 Vickerman, J. C., 1903 Vincent, B., 2599 Vinek, H., 1239 Vink, H., 507, 1297 Waghorne, W. E., 1267 Wagley, D. P., 47 Walker, R. W., 435, 3187, 3195, Wallington, T. J., 2737 Wang, G-W., 1039 Watkins, P. E., 2323 Watkiss, P. J., 1279 Watt, R. A. C., 489 Webb, G., 1689 Webster, B. C., 255, 267 Weiner, E. R., 1491 Wells, C. F., 2155. 2445 Wells, J. D., 1233 Whang, B. C. Y., 291 1 Whittle, E., 2323 Wichterlova, B., 2639 Wiesner, S., 3021 Wilhelmy, D. M., 563 Williams E. H., 3147 Williams, P. A., 403 Williams, R. J. P., 2255 Wokaun, A., 1305 Wolff, T., 2969 Wood, S. W., 3419 Woolf, L. A., 549, 1287 Wright, C. J., 1217 Wu, D. C., 1795 Wiirflinger, A., 3221 Wyn-Jones, E., 1915 Yamabe, M., 1059 Yamamoto, S., 941 Yamashita, H., 1435 Yamauchi, H., 2033 Yamazaki, A., 3245 Yariv, S., 1705 Yasumori, I., 841 Yeates, S.G., 1787 Yide, X., 969, 3103 Ylikoski, J., 2439 Yokokawa, T., 473 Yoneda, N., 879 Yonezawa, T., 1435 Yoshida, S., 119, 1435 Zambonin, P. G., 1029 Zanderighi, L., 1605 Zecchina. A., 2209, 2723, 1875, Zipelli, C., 1777 Zundel, G., 553 348 1, 2827 1891AUTHOR INDEX xxxv Sabbatini, L., 1029 Sacco, A., 2669 Sanders, J. V., 571 Sangster, D. F., 291 1 Sarka, K., 521 Sasahira, A., 473 Sasse, W. H. F., 571 Satchell, P. W., 2395 Sato, K., 841 Sato, Y., 341 Savino, V., 759 Sayers, C. M., 1217 Schiller, R. L., 1257 Schmidt, J., 1 Schmidt, P. P., 2017 Schneider, H., 3275, 3285 Schulz, R. A., 489, 1323 Scott, J. M. W., 739, 1651, 2287, Scott, S.K., 3409 Segall, R. L., 2609 Sehested, K., 2929, 2969 Seidl, V., 1367 Sem, P., 297 Serratosa, J. M., 2225 Seyama, H., 237 Seyedmonir, S. R., 87, 2269 Shanahan, M. E. R., 37 Sheppard, A., 2999 Sherwood, P. M. A., 135, 2099, Shindo, Y., 879, 2199 Shiotani, H., 2145 Shizuka, H., 383, 341 Siekhaus, W. J., 61 Sircar, S., 1101, 2489 Smart, R. St C., 2957, 2609 Smith, I. M., 3021 Smith, R., 3233 Snow, R. L., 3463 Solar, S., 2929 Solar, W., 2929 Solymosi, F., 1841 Soma, M., 237 Soupart, J-B., 3209 Sourisseau, C., 3257 Spink, J. A., 3469 Spoto, G., 1875, 1891 Spotswood, T. M., 3147 Staricco, E. H., 2631 Stassinopoulou, K., 3095 Stedman, D. H., 285 Stout, D. R., 3481 Strohbusch, F., 1757 Strumolo, D., 1479 Struve, P., 813, 2167 Styring, M. G., 3051 Subramanian, R., 2405 2881, 3359 2549, 2867 Sundar, H.G. K., 3491 Sutcliffe, L. H., 669, 3021 Sutton, H. C., 2301 Sutton, L. E., 635 Suzuki, H., 803 Suzuki, T., 1925, 3157 Symons, M. C. R., 423, 1005, Szamosi, J., 1645 Szczepaniak, W., 2935 Takagi, Y., 1925 Takahashi, K., 803 Takahashi, N., 629 Takanaka, J., 941 Takao, S., 993 Takasaki, S., 803 Takegami, H., 1221 Tam, S-C., 2255 Tamamushi, R., 2751 Tamaru, K., 29, 1567, 1595 Tamilarasan, R., 2405 Tanabe, S., 803 Tanaka, K., 2563,2981 Tanaka, T., 119 Taniewska-Osinska, S., 1409 Tascon, J. M. D., 1089 Teo, H. H., 981, 1787 Tetenyi, P., 3037 Thomas, J. K., 1163 Thompson, L., 1673 Thomson, M., 1867 Thomson, S. J., 1689 Tiddy, G. J. T., 789, 3339 Tittarelli, P., 2209 Tominaga, T., 941 Tomkinson, J., 225 Tonelli, C., 1605 Toprakcioglu, C., 13,413 Tran, T., 1867 Trasatti, S., 913 Tripathi, A.D., 1517 Tronc, E., 2619 Troncoso, G., 2127 Truscott, T. G., 2293 Tsurusaki, T., 879 Tuck, J. J., 309 Turner, P. S., 2609 Tusk, M., 1757 Tvarbikova, Z., 2639 Tyrrell, H. J. V., 1279 Ueki, Y.. 341 Ueno, A., 803 Unno, H., 1059 Valencia, E., 2127 van de Ven, T. G. M., 2677 van Ommen, J. G., 2479 van Truong, N., 3275, 3285 Vargas, I., 1947 2767, 2803, 21 1, 1999 Vedrine, J. C., 1017 Veith, J., 2313 Velasco, J. R., 3429 Vesala, A., 2439 Vickerman, J. C., 1903 Vincent, B., 2599 Vinek, H., 1239 Vink, H., 507, 1297 Waghorne, W. E., 1267 Wagley, D. P., 47 Walker, R. W., 435, 3187, 3195, Wallington, T. J., 2737 Wang, G-W., 1039 Watkins, P. E., 2323 Watkiss, P. J., 1279 Watt, R. A. C., 489 Webb, G., 1689 Webster, B. C., 255, 267 Weiner, E. R., 1491 Wells, C. F., 2155. 2445 Wells, J. D., 1233 Whang, B. C. Y., 291 1 Whittle, E., 2323 Wichterlova, B., 2639 Wiesner, S., 3021 Wilhelmy, D. M., 563 Williams E. H., 3147 Williams, P. A., 403 Williams, R. J. P., 2255 Wokaun, A., 1305 Wolff, T., 2969 Wood, S. W., 3419 Woolf, L. A., 549, 1287 Wright, C. J., 1217 Wu, D. C., 1795 Wiirflinger, A., 3221 Wyn-Jones, E., 1915 Yamabe, M., 1059 Yamamoto, S., 941 Yamashita, H., 1435 Yamauchi, H., 2033 Yamazaki, A., 3245 Yariv, S., 1705 Yasumori, I., 841 Yeates, S. G., 1787 Yide, X., 969, 3103 Ylikoski, J., 2439 Yokokawa, T., 473 Yoneda, N., 879 Yonezawa, T., 1435 Yoshida, S., 119, 1435 Zambonin, P. G., 1029 Zanderighi, L., 1605 Zecchina. A., 2209, 2723, 1875, Zipelli, C., 1777 Zundel, G., 553 348 1, 2827 1891

ISSN:0300-9599    

DOI:10.1039/F198480BX031

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

JOURNAL OF THE CHEMICAL SOCIETY F A R A D A Y TRANSACTIONS, PARTS I A N D I 1 The Journalof the Chemical Society is published in six sections, ofwhich five are termed Transactions; these are distinguished by their subject matter, as follows: Dalton Transactions (Inorganic Chemistry). All aspects of the chemistry of inorganic and organometallic compounds; including bioinorganic chemistry and solid-state inorganic chemistry; of their structures, properties, and reactions, including kinetics and mechanisms; new or improved experimental techniques and syntheses. Faraday Transactions I (Physical Chemistry). Radiation chemistry, gas-phase kinetics, electrochemistry (other than preparative), surface and interfacial chemistry, heterogeneous catalysis, physical properties of polymers and their solutions, and kinetics of polymerization, etc.Faraday Transactions 11 (Chemical Physics). Theoretical chemistry, especially valence and quantum theory, statistical mechanics, intermolecular forces, relaxation phenomena, spectroscopic studies (including i.r., e.s.r., n.m.r., and kinetic spec- troscopy, etc.) leading to assignments of quantum states, and fundamental theory. Studies of impurities in solid systems. Perkin Transactions I (Organic Chemistry). All aspects of synthetic and natural product organic, organometallic and bio-organic chemistry, including aliphatic, alicyclic, and aromatic systems (carbocyclic and heterocyclic). Perkin Transactions II (Physical Organic Chemistry). Kinetic and mechanistic studies of organic, organometallic and bio-organic reactions.The description and application of physicochemical, spectroscopic, and theoretical procedures to organic chemistry, including structure-activity relationships. Physical aspects of bio-organic chemistry and of organic compounds, including polymers and biopolymers. Authors are requested to indicate, at the time they submit a typescript, the journal for which it is intended. Should this seem unsuitable, the Editor will inform the author. The sixth section of the Journal of the Chemical Society is Chemical Communications, which is intended as a forum for preliminary accounts of original and significant work, in any area of chemistry that is likely to prove of wide general appeal or exceptional specialist interest. Such preliminary reports should be followed up eventually by full papers in other journals (e.g.the five Transactions) providing detailed accounts of the work. NOTES I t has always been the policy of the Faraday Transactions that brevity should not be a factor influencing acceptability for publication. In addition however to full papers both sections carry at the end of each issue a section headed ‘Notes’, which are short self-contained accounts of experimental observations, results. or theory that will not require enlargement into ‘full’ papers. The Notes section is not used for preliminary communications. The layout of a Note is the same as that of a paper. Short summaries are required. The procedure for submission, administration, refereeing, editing and publication of Notes is the same as for full papers.However, Notes are published more quickly than papers since their brevity facilitates processing at all stages. The Editors endeavour to meet authors’ wishesas to whether an article is a full paper or a Note, but since there is no sharp dividing line between the one and the other, either in terms of length or character of content, the right is retained to transfer overlong Notes to the full papers section. As a guide a Note should not exceed I500 words or word-equivalents. 0)NOMENCLATURE AND SYMBOLISM Units and Symbols. The Symbols Committee of The Royal Society, of which The Royal Society of Chemistry is a participating member, has produced a set of recommendations in a pamphlet ‘Quantities, Units, and Symbols’ (1975) (copies of this pamphlet and further details can be obtained from the Manager, Journals, The Royal Society of Chemistry, Burlington House, London W 1V OBN).These recommendations are applied by The Royal Society of Chemistry in all its publications. Their basis is the ‘Systeme International d’Unites’ (SI). A more detailed treatment of units and symbols with specific application to chemistry is given in the IUPAC Manual of Symbols and Terminology for Physicochemical Quantities and Units (Pergamon, Oxford, 1979). Nomenclature. 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. In order to encourage authors to use IUPAC nomenclature rules when drafting papers, attention is drawn to the following publications in which both the rules themselves and guidance on their use are given: Nomenclature of Organic Chemistry, Sections A, B, C , D, E, F, and H (Pergamon, Oxford, 1979 edn).Nomenclature of Inorganic Chemistry (Butterworths, London, 197 1, now published by Pergamon). Biochemical Nomenclature and Related Documents (The Biochemical Society, London, 1978). A complete listing of all IUPAC nomenclature publications appears in the January issues of J. Chem. Soc., Faraday Transactions. It is recommended that 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.FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY 1- Marlow Medal and Prize Applications are invited for the award of the Marlow Medal for 1985 and prize of f 100. The award will be open to any member of the Faraday Division of the Royal Society of Chemistry who, by the age of 32, had made in the judgement of the Council of the Faraday Division, the most meritorious contribution to physical chemistry or chemical physics. The award will be made on the basis of publications (not necessarily in the Transactions) on any subject normall published in J. Chem. SOC., Faraday Transactions I and 11, that carry a date of receipt for publication not later than the candidate‘s 32nd birthday. Candidates should be members and under 34 on 1 st January 1985, the closing date for applications, which may be made either by the candidate himself or on his behalf by another member of the Society.Copies of the rules of the award and application forms may be obtained from: Mrs Y. A. Fish, The Royal Society of Chemistry. Burlington House, London W1V OBN (ii)THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION NO. 78 ~ Radicals in Condensed Phases University of Leicester, 4-6 September 1984 ~ Organrung Committee I Professor M C R Symons (Chairman) 1 Dr T A Claxton Dr R L. Willson Dr K A McLauchlan Dr G B Buxton Professor Lord Tedder ' 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 discussion will be primarily concerned with the structure and reactions of radicals in liquids and solids.It is designed to bring together theoretical work on structure, environmental effects and reactivity with spectroscopic and mechanistic studies directly concerned with radicals. Fundamental aspects will be stressed, and particular attention will be given to new developments including measurement at short time intervals, special solvent effects, and the effects of external fields. A special area for inclusion will be electron gain and loss processes including trapped and solvated electrons, electrochemical reactions, and specific electron capture and electron loss in low-temperature systems. Photochemical charge-transfer processes will also be included. THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY DEUTSCHE BUNSENGESELLSCHAFT FUR PHYSIKALISCHE CHEMIE SOClI!TI! DE C H l M l E PHYSIQUE ASSOCIAZIONE ITALIANA D I CHIMICA FlSlCA Joint Discussion Meeting on : Laser Studies in Reaction Kinetics Evangelische Akademie, Tutzing, West Germany, 2&27 September 1984 Organising Committee R.Ben Aim (Gif sur Yvette) G. Giacometti (Padova) P. Rigny (Gif sur Yvette) E. W. Schlag (Munchen) I. W. M. Smith (Cambridge) J. Troe (Gottingen) K. Welge (Bielefeld) The aim of this meeting is the discussion of the latest experiments and related theories in the field of laser studies of elementary chemical reactions in molecular beams, in the gas phase, and in the condensed phase. The discussion will include oral contributions and poster presentations.Further information may be obtained from: Professor Dr J. Troe, lnstitut fur Physikalische Chemie, Universitat Gottingen. Tammannstrasse 6, 03400 Gottingen, West Germany. Authors of accepted contributions will be required t o provide a manuscript for publication in a special issue of the Berichte der Bunsengeseflschaft fur Ph ysikalische Chemie.THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY SYMPOSIUM NO. 19 Applications University of Cambridge, 12-1 3 December 1984 I N.B. Please note change of date Molecular Electronic Structure Calculations- Methods and Molecular electronic structure calculations have now developed into a powerful predictive tool and are necessary in several different fields to aid the understanding and interpretation of experimental observations.The meeting will review the current state of this rapidly developing discipline and will bring together experts on some of the most advanced methods and their applications. The meeting will provide an opportunity for discussion and comparison of the various techniques currently in use. It will therefore not only be a valuable forum for discussion among research workers in the field, but should also show the non-specialist what theoretical calculations can be expected to achieve now and in the near future. 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. 79 (in conjunction with the Polymer Physics Group) Polymer Liquid Crystals University of Cambridge, 1-3 April 1985 The object of the meeting will be to discuss all aspects of the developing subject of polymeric liquid crystals.The hope is to bring together scientists from the fields of conventional polymer science and monomeric liquid crystals who are active in this field. The discussion is aimed at understanding the following facets: (a) The chemical characteristics that give rise to polymer liquid crystalline behaviour. (b) The nature of the high local anisotropy of these systems and their structural organisation at the molecular, micron and macroscopic levels. (c) The physical properties and their industrial exploitation, with particular reference to the influence of external force fields such as flow, electric and magnetic fields.(d) The inter-relations of polymer liquid crystals with small-molecule mesophases, conventional flexible polymers and biopolymers which exhibit liquid-crystalline behaviour. The preliminary programme may be obtained from: 1 Mrs Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W1 V OBNTHE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION NO. 80 Layers l University of Oxford, 17-18 December 1985 Physical Interactions and Energy Exchange a t the Gas-Solid Interface McMaster University, Hamilton, Ontario, Canada, 23-25 July 1985 Organising Committee : Professor J. A. Morrison (Chairman) Dr M. L. Klein Professor G. Scoles Professor W. A. Steele Professor F. S. Stone Dr R. K. Thomas 1 Full papers for publication in the symposium volume will be required by August 1985.The discussion will be concerned with certain aspects of current research on the gas-solid interface: elastic, inelastic and dissipative scattering of atoms and molecules from crystal surfaces, and the structure and dynamics of physisorbed species, including overlayers. Emphasis will be placed on the themes of physical interactions and energy exchange rather than on molecular-beam technology or the phenomenology of phase transitions on overlayers. The interplay between theory and experiment will be stressed as they relate to the nature of atom and molecule surface interaction potentials, including many-body effects. Further information may be obtained from: Professor J. A. Morrison, Institute for Materials Research, McMaster University, Hamilton, Ontario, Canada U S 4M1 THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY SYMPOSIUM NO.2 0 Phase Transitions in Adsorbed Organising Committee : Professor J. S. Rowlinson (Chairman) Dr E. Dickinson Dr R. Evans Mrs Y. A. Fish Dr N. Parsonage Dr D. A. Young The aim of the meeting is t o discuss phase transitions at gas/liquid, liquid/liquid and solid/fluid interfaces, and in other systems of constrained geometry or dimensionality less than three. Emphasis will be placed on molecularly simple systems, whereby liquid crystal interfaces and chemisorption phenomena are excluded. Contributions for consideration by the Organising Committee are invited and abstracts of about 300 words should be sent by 12 October 1984 to: Professor J.S. Rowlinson, Physical Chemistry Laboratory, University of Oxford, South Parks Road, Oxford OX1 3QZ.FARADAY DIVISION INFORMAL AND GROUP MEETINGS Industrial Pbysica I Chemistry Group Bacterial Adhesion To be held at Girton College, Cambridge on 11-1 3 September 1984 Further information from Dr I. D. Robb, Unilever Research, Port Sunlight, WirraI, Merseyside L63 3JW Electrochemistry Group with the SCI Electrochemical Technology Group Etectrolytic Bubbles To be held at Imperial College, London on 13-1 4 September 1 984 Further information from Dr G. H. Kelsafl, Imperial College, London SW7 Polar Solids Group with British Ceramic Socie fy, Institute of Physics and Mineralogical Society d Gre8t Britain Kinetics and Mass Transport of Silicate and Oxide Systems -To be held at The Geological Society, London on 13-1 4 September 1984 Further information from Or R.Freer, Department of Electrical and Electronic Engineering, North Staffordshire Polytechnic, Beaconside, Stafford $Tl8 OAD Molecular Beams Group New Directions in Molecular Beams To be held at the University of Oxford on 13-1 4 September ‘I 984 Further information from Dr J. C. Whitehead, Department of Chemistry. University of Manchester, Manchester M13 9PL Division with the Gas Kinetics Group Autumn Meeting: Combustion Chemistry in the Gas Phase To be held at the University of Hufl on 18-20 September 1984 Further information from Dr R. W. Walker, Department of Chemistry, The University, Huit HU6 7RX Colloid and Interface Science Group with the SCI Dynamic Surface Tension Effects in Foaming and Aeration To be held at Imperial College, London on 20-21 September 1984 Further information from Professor A.Bailey, Imperial College, London SW7 Electrochemistry Group witb Statistical Mechanics and Thermodynamics Group The Eiectrical Double Layer To be hetd at the University of Southampton on 25-26 September 1984 Further information from Dr E. Dickinson, Procter Department of Food Science, University of Leeds, Leeds LS2 9JT Carbon Group Refactory Applications of Carbon To be held at the University of Sheffield on 25-27 September 1984 Further information from Dr 8. Rand, Department of Ceramics, Glasses and Polymers, University of Sheffield, Sheffield S10 2TZ Electrochemistry Group witb ICI New Science Group Organic Electrosynthesis To be held at ICI, Runcorn on 27 September 1984 Further information from Dr A.J. B. Cutler, CEGB, Kelvin Avenue, Leatherhead, Surrey KY22 7SENeutron Scattering Group Neutron Scattering from Aqueous Solutions To be held at the University of Bristol on 27-28 September 1984 Further information from Dr G. W. Neilson, H. H. Wills, Physics Laboratory, University of Bristol, Tyndall Avenue, Bristol BS8 1TL Polymer Physics Group with the Institute of Marine Engineers Polymers in a Marine Environment To be held at the Institute of Marine Engineers, London on 30 October and 1 November 1984 Further information from Dr G. J. Lake, MRPRA, Brickendonbury, Hertford SG13 8ML Neutron Scattering Group Neutrons in Magnetism To be held at the University of Southampton in December 1984 Further information from Dr R.B. Rainford, Department of Chemistry, University of Southampton, Southampton SO9 5NH High Resolution Spectroscopy Group CARS, Diode Laser and Microwave Spectroscopy To be held at the University of Reading on 17-18 December 1984 Further information from Dr J. M. Hollas, Department of Chemistry, University of Reading, Reading RG6 2AD Colloid and Interface Science Group with the Colloid and Surface Chemistry Group of the SCI Colloidal Aspects of Cohesive Sediments (provisional) To be held at the SCI Headquarters, London on 18 December 1984 Further information from Dr R. Aveyard, Department of Chemistry, The University, Hull HU6 7RX Division Annual Congress: Solid-state Chemistry To be held at the University of St Andrews on 26-28 March 1985 Further information from Professor P.A. H. Wyatt, Department of Chemistry, University of St Andrews, The Purdie Building, St Andrews KY16 9ST Neutron Scattering Group Small Angle Neutron Scattering from Organised Systems To be held at Imperial College, London on 17-18 April 1985 Further information from Dr R. W. Richards, Department of Pure and Applied Chemistry, University of Strathclyde, Glasgow G1 1 XL (vii)physicochemical topics, thereby encouraging scientists of different disciplines t v contribute their varied viewpoints to a common theme. The Royal Society of Chemistry- A recent Discussion is :- NO.?~ Intramolecular Kinetics No. 75 in the series, this publication is the result of a general discussion held at the University of Warwick in April 1983.Contents : The Spiers Memorial Lecture; Vibrational Redistribution within Elcited Electronic States of Polyatornic Molcculcs Intramolecular Relautiuii of ILscited Stdtes lsumeruation ~i Internal-encrgy-selected Iuns Kinetics 01. Ion-Molecule Collision C o m p l e w in the Gas Phase, E\perment and Theory Intramolecular Decay of Some Open-shell Polyatomic Cations On the Theory of I n t r a i i i ~ l e c i r l ~ r 1:nerp) Transfer Pulsed Laser Preparation and Quantum Superposition State L'volution in Reeuhr and Irregular Systenis A Vudntuni-mechanicdl Inlerndl-colhion Model lor State-selected Unimolecular Decomposition The Correspondence Principle and Intramolecular Dynamics lntrainolecular Dephasing. l'icoaecoiid Evolution of Wavepacket States in a Molecule u i t h Intermediate-case level Structure Energy Conversion in van der Wads Complr\es of a-TetraLine and Argon Tune-dependent Processes in Polydtomic Molecules During and After Intense Infrared Lr rad iat io n Energy Distributions in the C N ( X ' 1 t ) Fragment from the Infrared Multiplephoton Ihssociation of CFICN.A Comparison between Cxperimcntal Results and the Predictions of Statistical Theories Of C 6 F h + Prvduct L-nerpy Partitioning in the Deconi- position ot Statc-scleLtively l:\cited HOON and t IOOD Low-power Infrarcd I.JSCI I'liotiilysia vt l e t r a n i e l Iivldivuetan Uniniolccular Reactions tnduced by Vibrational Overtone Ekcitatioti Uniiiiolecular Decoinposition of t-Butylhydro- pertihide by Ilirect E\citdtion 01.the 6 - 0 0-tl Stretching Overtunc Pliotocheniiatry. Lnergy Redistribution And i t s Iiiipdct'iin ('olierence. Isoincrimtion, 1)issociation and Solvation 1.ncrgy Redistribution in Large Molccules. Duect Study of- Intrninolccular Relaxation in the Gas Ph~se with I'icosccond Gating Rotdtion-dependent Intramolecular Proccsscc~rt S O . ( A ' A . ) i n a S u p r r \ ~ i n i c Jet Role of Roldtion Vibrdtion Interaction in Vihrdtvinal R e b u t i o n . Liiergy Redistribution in Excited Singlet I~ornialdcliydc Sub-Doppler. Spectroscopy of Benzene in the "Channel-t hree" Region Intraniolecular 13ectronic Rela xdt ion and Pliotoisoineru.at ion Processes in the Isolated Azabenxne Molccules Pyridine, Pyrazine and Pyrimidine Picosecond-jrt Spectroscopy and Softcover 434pp 0 85186 658 1 Price f25.00 ($48.00) Rest of the World f26.00 RSC Members f 16.25 Farada! Discussions of the Chemical Society 7 $ Irirronioletuliir hinctiii Faraday symposia are usually held annually and are confined to more specialised topics than Discussions, with particular reference to recent rapidly developing hnes of research.No. 17 in the series, this publication is the result of a symposium on The Hydrophobic Interaction held at the University of Reading in December 1982. Contents: Hydrophobic Interactions-a Historical Perspective Precise Vapour-pressure Measureinenta of the Solubilization of Benzene by Aqueous Sodium Octylsulphate Solutions Nuclear Magnetic Rewnance Relaxation Investigation of Tetrahydrofuran and Methyl Iodide Clathrates Infrared and Nuclear Magnetic Resonance Studies Pertaining t o the Cage Model for Solutions of Acetone in Water Isothermal Transport Properties in Solutions of Symmetrical Tetra-alkylamnioniurn Bromides Thermodynamics of Cavity Formation in Water. A Molecular Dynamics Study Molecular Librations and Solvent Orient- ational Correlations in Hydrophobic Phenomena Monte Carlo Computer Simulation Study 01 the Hydrophobic Effect. Potential of Mean Force for C(CH4)laq at 25 and 50" C Geometric Relaxation in Water. Its Role in Hydrophobic Hydration Hydrophobic Moments and Protein Structure Application of the Kirkwood-Buff Theory to the Problem of Hydrophobic Interactions I)isentanglenient of Hydrophobic and Flectrostatic Contributions lo the I.ilni Pressures of Ionic Surfactants Hydrophobic Interaction\ in Dilute Solutions 01. Poly(viny1 alcohol) Conformational and functional l'roperties ot Harnioglobin in Water+Alcohol Mixtures. Dependence of Hulk I:lectrostatic and liydrophobrc Interactions upon pH and KCI concentrations Softcover 240pp 0 85186 668 9 Price f36.50 ($70.00) Rest of the World f38.50 RSC Members f23.75 ORDERING RSC Members should send their orders to: The Royal Society of Chemistry. The Membership Officer, 30 Russell Square, Non-RSC Members The Royal Society of Chemistry, Distribution Centre, Blackhorse Road, L London WClB 5DT. Letchworth, Herts SG6 1HN. England. Faraday S! mposia of the Chemical Societ) \<, 17 (viii)

ISSN:0300-9599    

DOI:10.1039/F198480FP061

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. 1, 1984,80, 2017-2026 Band-like States for Ionic Conduction BY PARBURY P. SCHMIDT* AND CAROL KORZENIEWSKI~ Department of Chemistry, Oakland University, Rochester, Michigan 48063, U.S.A. Received 26th May, 1983 Facile, ultrafast migration of ions is observed in several systems. Our concern in this paper is primarily with cation transport across phospholipid membranes via cylindrical channels. By means of an analogy with the Mott-Hubbard metal/non-metal transition, we show that band-like ionic conduction through cylindrical channels is possible. In particular, if the height of the intrinsic barrier between potential-energy minima is approximately the same as the width of the ‘conduction band’ which can be defined for the system, then band-like transfer is possible.The existence of band-like states does not guarantee high conductivity. However, the underlying band-like structure may interact non-linearly with the phonons to yield solitary waves and even solition states. Such solitary waves migrate with minimal degradation and thus ensure high conductivity . The transport of small cations (primarily of Group I, viz. sodium and potassium) across lipid membranes via channels for the conduction appears to be an accepted interpretation of experimental observations. Nevertheless, the mechanism of trans- fer is still largely unresolved. It has been suggested that the channel for conduction consists of protein macro- molecules which form channels with diameters < 6 at the surface of the membrane.l Within the channel, the diameter narrows; this narrowing has been suggested to act as a selective filter.2.Furthermore, along the channel, regions of varying potential energy present a sequence of energy barriers through which the ion must tunnel. In this limit, ionic migration is viewed as a ‘hopping’ mechani~m.~ However, even in an adiabatic limit, a hopping mechanism for ionic conductivity seems to us to embody an intrinsic slowness for the conduction process which is at odds with the apparently fast trans-membrane ionic migration, especially across the axonal membrane. We therefore suggest, and provide theoretical support for, a mechanism of conductivity which involves band-like states. Band states alone cannot assure high conductivities. However, non-linear interactions between the conduction bands and the phonons can yield solitary waves ; solitary waves migrate without substantial degradation over considerable distances.A simple model shows that band-like states are possible. Within a chain of harmonic-oscillator wells a particle, an ion for example, interacts with each well. As the spacing between the wells decreases, the height of the intrinsic barrier between the wells approaches the width of the ‘conduction band’ for the ion. Under this condition (drawing an analogy with the Mott-Hubbard metal/non-metal tran- sition59 6, we assert that band-like states for ionic conduction can exist. The analysis of the model system shows that local vibrational states of the ion in the conducting channel combine to yield band states.However, in order to consider the transport process completely, it is necessary to include the influence of the t Present address: Department of Chemistry, University of Utah, Salt Lake City, Utah, U.S.A. 20172018 BAND-LIKE STATES FOR IONIC CONDUCTION environment. Environmental factors can be isolated and studied with the use of a Born-Oppenheimer type of separation. For this system the analysis resembles the Holstein analysis of the small p01aron.~ In order to carry out an analysis to prove the possible existence of band states, however, it is sufficient merely to use adiabatic degrees of freedom of the conducting subsystem. The channel for conduction contains structures within which local vibrational degrees of freedom for the mobile species can be defined.The use of these local degrees of freedom establishes the limit for the onset of band-like states. For the activated, hopping mode of transfer, environmental factors (uiz. the polarizability of the environment) contribute importantly to the value of the energy of activation and must be considered. For band-like ionic conductivity, as is also the case for electronic conduction in metals, perturbing phonons scatter the charge carriers. The result is a resistance to conduction. Yet, as noted above, in systems of this type solitary-wave states can arise; therefore such states may account for ion transfer which is fast and relatively insensitive to scattering processes. In the following section the basic model is justified with the use of a simple simu- lation of the conduction channel.Next, the model is parameterized as a sequence of oscillator wells. An analysis of that parameterization suggests the limits for which band states for ionic conduction can be argued to exist. THE MODEL The channel-forming species alamethicin is an appropriate model for study. The structure of the channel for conduction, as formed from twisted, six-stranded, helical aggregates of protein [cf. ref. (l)], is important. Such an aggregate can project a sequence of potential minima onto the transfer axis. Within each energy minimum it is possible for the cation to execute harmonic oscillations near to the ground vibrational state. Because of the microscopic character of the direct interactions between the mobile ion and the elements of the conduction channel, the ion interacts individually with atoms (or at worst with small groups of atoms) of the monomers in the chain.It is clear, therefore, that as the ion migrates through the protein channel it experiences the influence of a tight sequence of potential-energy minima along the trajectory it follows. Under the appropriate conditions, the projections of potential minima onto the trajectory for ion transfer can merge into a barrierless continuum. We now present a calculation to support this hypothesis. Assume that the migrating ion interacts with a microscopic, probably atomic, portion of the protein which makes up the channel according to the potential-energy function : (1) V(r) = -Zion e2a/2r4 -B/rs + 6/r12. The three contributions to this potential-energy function consist of an ion-induced- dipole interaction, a Londonlvan der Waals attraction and a repulsion, which for computational simplicity is taken to be an inverse- 12 law.In eqn (1) a is polarizability, /3 is a parameter which is associated with the van der Waals attraction and 6 is a parameter associated with the repulsion. As the potential-energy function corresponds to an atomic model, permanent ion-dipole interactions are not considered. Ion-dipole contributions are realistic for interactions which involve molecules beyond the immediate boundary of the conduction channel; such interactions, however, are screened by the intervening material and therefore contribute less than the immediate interactions.P. P. SCHMIDT AND C. KORZENIWESKI 2019 Table 1.Values of the parameters used in the model potential-energy function parameter value ct (polarizability) /3 (van der Waals constant) 6 (repulsion) ro (ion-monomer distance) D (ion-monomer dissociation energy) (monomer-monomer distance) 5.0 x 10-23 A 3 281 2A 0.67 kJ A6 mol-l 53.10 kJ A12 mol-I 193 kJ mol-l Table 1 summarizes the values of the parameters used in eqn (1). The parameters were chosen arbitrarily to yield an ion-monomer dissociation energy of 2eV (193 kJ mo1-l). (From now on, ‘monomer’ refers to an atomic or small-molecular component of the protein chain which makes up the channel.) An equilibrium ion-monomer separation of 2 A was also assumed. A potential-energy function of the type of eqn (1) reflects an average of the kinds of interaction which can be expected for the variety of atoms in the protein chains and the distances between atoms which are involved.Several calculations were carried out, and the results are illustrated in the following figures. An important feature of the model, which emerges clearly in the analysis of the harmonic wells to follow in the next section, is the decrease of the intrinsic energy barrier between harmonic wells as the distance between the wells decreases. The individual chains of protein which make up the channel contribute potential minima to the sequence which is projected onto the trajectory for ion transfer. Considering typical structures for such molecules, it is evident that the polar atoms, which alone account for the major art of any local interaction within the polymeric together such that the polar atoms on the chains all occur in the same (repeat) planes, then a sequence of deep wells forms.This situation is illustrated in fig. 1. It is difficult to see how ionic migration can occur easily in such a system. The activation energy for conduction would be high and tunnelling would be vanishingly small owing to the large distances between the wells. On the other hand, if each of the six strands is slipped slightly with respect to the others, then it is easy to see that the projections of the potential minima onto the transfer axis can merge. Such a situation is shown in fig. 2. In fig. 2(a) the spacing between monomeric units (i.e. polar atoms) is taken to be 2 A. The slip distance between the six strands is 0.2 A.This slip distance allows a small periodic void to arise. As a result, potential minima are seen; the barriers are now considerably lower in energy than in the case illustrated in fig. 1. Moreover, these barriers arise only between the voids. They are not seen within the sequence of slipped strands. In fig. 2(6) the spacing between monomers has been extended to 3 A and the slip distance is 0.3 A. The height of the barriers is seen to decrease. This decrease in height of the barrier is due simply to the increase in the distance between the test charge on the axis and the atoms along the chain. In fig. 2(c) the spacing between monomers is again 2 A, but now the slip distance is 0.33 A (uiz. 6 of the monomer spacing). No clear wells are seen to form in this case.Thus it seems reasonable to assume that the slippage of strands of protein within an aggregate which makes up a channel for conduction can account for the presence of a sequence of closely spaced potential-energy wells. This is exactly what is required to argue the existence of energy-band states for conducting ionic systems. protein strand, are separated 2-3 K from each other. If six such strands are wound2020 BAND-LIKE STATES FOR IONIC CONDUCTION -27 5 8 11 14 distance through membrane Fig. 1. Segment of the potential-energy profile within the channel for the case in which the sources (which are part of the channel) all lie in repeat planes. The alternation of voids and sources results in deep minima and high barriers. In addition to plotting the potential profile along the transfer axis, we determined the value of the force constant for a harmonic oscillation of the transfer species at one of the energy minima within the body of the channel; we took care to examine a potential-energy minimum which lies well removed from the surface.A general expression for the force constant for an arbitrary potential-energy function is given where the summation is over the individual atomic or molecular contributions. The application of this formula to the potential-energy function, eqn (l), yields Ze2a k = x [ -~ Rf (2f5 “P 2( C O S ~ ~ ) ) -8 :i( 10+24P,(~0~Oi) i-1 where P2(cos Oi) = i ( 3 cos2 Bi - 1) and Oi is the angle between the z axis and the line to the monomer i. For a general spacing of 2 A, with the values of the other parameters as listed in table 1, we find a fundamental frequency for sodium, for example, of ca.200 cm-l; the approximation arises as a result of the uncertainty in the value of the reduced mass for the cation in the chain. This value for the basic oscillation of the cation in one potential well is consistent with values for the vibration of sodium in cages of solvent. The analysis which follows is an extension of the classical treatment of the tunnelling between two harmonic-oscillator potential wells which was considered by Wall and Gl~ckler.~ The difference between our model and the smaller system of Wall and Glockler is simply the fact that we must consider many more wells and the fact that no limit goes to infinity.P. P. SCHMIDT AND C . KORZENIWSKI 202 1 7 4 - 7 - 17 - 2 7 .-7 - ( 0 ) -17 - 2 7 % - 27 0 I/-- 10 20 30 -27 0 10 2 0 30 . - ( c ) THE ANALYSIS OF THE BAND STATES Given the uniform sequence of local harmonic-oscillator wells, as illustrated in fig. 3, the Hamiltonian operator is given by P2 H = - - + V 2m (4) where the potential-energy operator V is given by V = $k[zf2(m+ l)q2 ( 5 ) where m = 0, 1,2,. . . and k is the uniform force constant, which is the same for each of the wells. The width of a well is 2. The basis function for the ground vibrational state of an ion which is located in any of the wells is &z,) = (a/.): exp ( - iazf) (6) with z , = z+ml. (7)2022 BAND-LIKE STATES FOR IONIC CONDUCTION distance Fig. 3. Chain of harmonic-oscillator wells. The width of each well at the points where the well intersects the parabolas of its two neighbouring wells is 2.The labelling of the wells runs from m = 0 at the (chosen) centre to - t [ ( N - 1)/2]. Here rn has the same meaning as in eqn (5). The state function for the system is With the use of this form of the state function, the energy levels are easily found to be and the limit of the summation, lim, is given by lim = [$(N- l ) ] where the bracketing operation (. . .] indicates the taking of the integer. The single-centre matrix element a in eqn (8) is found to be a = (OIrrlO) = gi4o+(O(VlO) (1 1) with The error integral is given by erf (q), q = d a 1. The two-centre matrix element is found to be P m = <OIHlm> = ghSom -&co(m + 1 ) 2 q2S,, + (01 r/lm>P. P. SCHMIDT AND C.KORZENIWKI with 2023 + [ 1 +2(2n-m+ 2)2 q2] (erf [(2n -m + 3) q] -erf [(2n -m + 1) q]} -- 2q (6n + 3m + 7) exp [ - (2n+m + 3)2 q2] - (6n + 3m + 5 ) d n x exp [ - (2n + m + 1)2 q2] + (6n - 3m + 7) exp [ - (2n - m + 3)2 q2] - (6n - 3m + 5) exp [ - (2n - m + q2] . 11 The overlap integral between 0 and m is Som = exp ( -mq2). THE IONIC CONDUCTOR TRANSITION By analogy with the metal-non-metal transitions we assume that whenever the bandwidth A (which arises owing to the broadening of the ground-state vibrational levels through interaction with the sequence of wells): A = Emax-Emin (16) Ebarrier = khq2 (17) is approximately the same value as the height of the barrier: the conduction of the ion occurs as a band-type migration as opposed to a hopping migration. The condition is A Ebarrier* (18) The intersection of the barrier height with the bandwidth for the sodium or potassium cation as a function of q is illustrated in fig.4. If, for the sodium cation, we assume that the fundamental frequency o is ca. 200 cm-l, the value of q at the intersection of the bandwidth and the barrier height is ca. 2.1. This value of q, together with the mass of sodium, gives a value of 1 equal to 0.18 A; this value is half of the spacing between the wells. For potassium we used 150 cm-l for the fundamental frequency. The value of q at the intersection is higher than it is for sodium, but the value of 1 is less, 0.16 A. Thus the appearance of band-like conductivity for potassium requires, by our criterion, a closer spacing of the monomers. In this paper we have treated only the case of one ion within a channel.It is possible that several ions may occupy positions in the channel at once. A result of this, multiple occupancy of the channel can be complicated. However, even in the absence of additional calculations on models, we can suggest the following. The channel is not rigid. Although our modelling has assumed rigidity of the channel (in order to demonstrate the possible existence of band states in the systems),2024 BAND-LIKE STATES FOR IONIC CONDUCTION 3 -0 2.0 3 rx G 1 .o 0 2.00 2.05 2.10 2.15 2.20 4 Fig. 4. Plot of the bandwidth against q for sodium and potassium. The energy is plotted in reduced (scaled) units in order to show the relative intersections. The height of the barrier is plotted as E,, = q 2 / 2 .The point of intersection of the two curves marks the boundary in q and the spacing I for which one can expect to see band-like conduction. within the neighbourhood of an ion the attraction (or intrinsic repulsion) between the elements of the channel and the ion can result in a local constriction (or swelling if a repulsion operates). The self-energy of the ion within local region will differ from that which we found above for the ion in a rigid system. More to the point will be the fact that the effective interaction between two ions of like or opposite charge within the channel will differ considerably from the form of the direct interaction between the ion. There can arise, through mediation of the environment, a diminution of repulsion or attraction.If the distortion of the channel by an ion is sufficiently great, this distortion can have a strong influence on the other ions in the vicinity. It is possible for pairs of ions to be bound together by such distortions and to migrate as a correlated pair. There is an apparent analogy to be drawn with the case of p-alumina. Such ion-pairing has been demonstrated to be (physically) reasonable, and it has been shown that the pairing can bring about significant lowerings of the barriers to ionic migration.1° The implication of the barrier-lowering through ionic associations in biological channel systems is straightforward; it may be possible for groups of ions to occupy the space in the channel for conduction and thereby, as with the aluminas, result in the lowering of the effective barrier.It is reasonable to expect that, through leakage, the channel for conduction across the lipid film should contain a number of cations. The rapid migration of such species for an excitable tissue in its resting state would not be expected. However, the rate-governing step for the leakage current in the resting state is clearly the slow transfer of ions across the membrane. As facile transfer of ions is necessary in the excited state (which manifests the action potential), it is clear that facile transfer inside the channel must be assured for all states of the system; this situation must apply for all states in an excitable system if the ‘gating’ of the ionP. P. SCHMIDT AND C. KORZENIWESKI 2025 transfer occurs, as supposed, at the surface of the membrane.If concentrations of ions are required to occupy the channel for conduction, then excitability would not occur unless the channels contained the requisite minimum number of cations (such cations, as indicated, being required to ensure a small intrinsic barrier to conduction). The condition, eqn (18), which predicts the beginning of band-like states for extended ion-transfer systems does not depend upon excess. ambient cations which, in the resting state, reside within the channel for conduction. If, as the previous paragraphs suggest might occur, cations are required to account for the lowering of the intrinsic barrier, then the model shall have to be extended. It may be necessary, for example, to invoke one form or another of the solitary-wave phenomenon to explain the conductivity.In order to decide if modificaton of the model is necessary, much m x e information is needed about the structure of the channels, particularly in the region of the path of the ion transfer. At this point, nevertheless, we are still free to explore some additional consequences of the condition (1 8) simply in terms of the fundamental strength of interaction which operates between the transfer ion and the atoms and molecules of the monomers of the protein chain. Wide spacing between the sources for the ion-monomer interaction in the conduction channel assures the trapping of ions at local sites; the bandwidth will be narrow, dependent upon vibrational quantum numbers of the phonons, and therefore a maximum value (as with the small polaron7) will occur only at absolute zero.Such a condition favours a hopping model for the conductivity. On the other hand, the close spacing between wells, as we have shown, broadens the band of oscillator states and at the same time lowers the heights of the intrinsic barriers between the wells. Thus a limit is reached for a tubular conduction channel where, in effect, the migrating ion ceases to ‘see’ the molecular structure. It is then free to migrate without activation along the length of the channel. The driving force is the electrochemical gradient (which is almost discontinuous) which exists between the ends of the cylinder for conduction. As noted in the Introduction, Hille4 has suggested that conduction channels may have constrictions which act as selective filters for ions.If such constrictions indeed exist, then the ions must tunnel through these barriers for any current to be observed; the channels for conduction can consist of sequences of bands and traps. Also, although in the excited state (which is manifested by the action potential in the nerve) gating can occur at the surface of the lipid membrane, interior traps may limit the number of channels which can accommodate any particular ion. Alamethicin, as a model for an ionic-conduction channel, shows effectively no ionic exclusivity; i.e. many small cations migrate relatively freely through .the channels.’ This fact does not rule out the possibility of a differential mobility for ions. For example, if the spacing of the monomers in the conduction channel were set to accommodate and yield a band-like conduction for sodium, but not for potassium, then a particular channel would be ion selective.As our calculations suggest, the difference in spacing between the wells which is required for band-like conduction for sodium as opposed to potassium is smali. Although a channel may satisfy the criterion for band-like conduction for one ion and not another, the hopping migration for the other ion (for example potassium) may be sufficiently facile as to ensure the easy transport of that ion through the same channel. Finally, we note with respect to ionic conductivity in the protein channels that this conduction probably takes place for a desolvated (deaquated) ion. Ions at the surface of the membrane, partially or fully solvated, pass from their sheaths of solvent into the channel.One may regard the channel for conduction as essentially a cylindrical ‘solvent’. The ion which is contained within the channel occupies an environment2026 BAND-LIKE STATES FOR IONIC CONDUCTION which is not significantly different from its original aqueous environment. Indeed, conduction is aided through the ' hydrophobic' regions of the lipid system by allowing an ion to migrate in an environment which mimics many of the interactions of its solvent, water, in free solution. SUMMARY The analysis of the model for a conducting channel which' we have carried out shows that it is possible to define band-like states which may be important in the transport of the ion across lipid membranes. The onset of the band-like phenomenon, we assert, occurs when the width of the band of vibrational states equals or exceeds the height of the intrinsic barrier between the individual potential-energy wells which can be defined within the channel. This condition, we argue, is the analogue for ionic conductivity of the Mott-Hubbard condition for the metal-non-metal transition. We have suggested that the main consequence of this band-like conductivity is facile ion transfer. The plastic, deformable environment can influence this transport in one of two important ways: it can reduce the facility of ion transfer by phonon scattering and, on the other hand, it can ensure an almost unattenuated transfer through non-linear interactions with the phonons; the consequence of this is the appearance of solitary-wave phenomena. This work was supported in part by the U.S. Office of Naval Research, Arlington, Virginia. R. Nagaraj and P. Balaram, Acc. Chem. Res., 1981, 14, 356. C. M. Armstrong, Q. Rev. Biophys., 1975, 7 , 179. B. Hille, Biosystems, 1977, 8, 195. B. Hille, Biophys. J., 1978, 22, 283. N. F. Mott, Philos. Mag., 1961, 6, 287. J. Hubbard, Proc. R. SOC. London, Ser. A , 1963, 276, 238; 1964, 281, 49. J. M. McKinley and P. P. Schmidt, J. Chem. SOC., Faraday Trans. 2, 1982, 78, 876. F. T. Wall and G. Glockler, J. Chem. Phys., 1937, 5, 314. ' T. Holstein, Ann. Phys. (NY), 1959, 8, 325; 343. lo J. C. Wang, M. Gaffare and S . 4 Choi, J. Chem. Phys., 1975, 63, 722. (PAPER 3/854)

ISSN:0300-9599    

DOI:10.1039/F19848002017

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. 1, 1984, 80, 2027-2032 Estimation of Partial Molar Volumes of a-Amino Acids in Water BY MANDAPATI V. R. RAO,* MAHALAKSHMI ATREYI AND MOGANTY R. RAJFSWARI Department of Chemistry, University of Delhi, Delhi, India Received 15th June, 1983 A method of estimating partial molar volumes of a-amino acids in water, starting from the partial molar volume of glycine, is described. For eleven zwitterionic amino acids the estimated partial molar volumes are found to be in very good agreement with the experimental values given in the literature. Amino acids having ionisable side chains can exist in aqueous solution as different ionic species; the partial molar volumes of these obtained from our densitimetric studies have been compared with the estimated values.The method gave good results for the ionic species of lysine and arginine, where the charge centre in the side chain is well separated from a-NH, and a-COO-, but not for all species of aspartic and glutamic acids. However, for the two dicarboxylic amino acids a different procedure, using data for a,w-amino acids, gave good estimates of the partial molar volumes. Recently there has been renewed interest in the volumes,1-6 heat capacities'? 7-9 and compre~sibilities~~ of a-amino acids, mainly with a view to elucidating the contribution of the amino-acid side chains to the thermodynamic properties of biopolymeric systems; some of these studies were concerned with aspects of estimating partial molar volumes 3 9 Shahidi and Farrel12 modified the spherical-cavity method of Edward et al.l0 to take account of electrostriction and hydrogen bonding in estimating of zwitterionic amino acids and succeeded in obtaining good agreement with experimental data for alanine, valine, serine and methionine. Millero and coworkers3 used data of functional groups to calculate the volume contribution of various side chains of a-amino acids, but could get fair agreement with experimental values only for five amino acids with non-ionisable side chains.Cabani et al.,5 using the group-volume method, estimated the electrostriction of zwitterions of alanine, valine, leucine and norleucine to be - 14.7 0.8 cm3 mol-l, which is higher than the accepted value for glycine (- 13.5 cm3 mo1-l).l6 The spherical-cavity method2 and group-volume method3? were found to be inadequate for estimating P of amino acids with ionisable side chains; it has so far not been possible to estimate P of glycine, the simplest of the a-amino acids.The present paper deals with the estimation of partial molar volumes of protein a-amino acids with ionisable as well as non-ionisable side chains. RESULTS AND DISCUSSION PARTIAL MOLAR VOLUMES OF a-AMINO ACIDS WITH NON-IONISABLE SIDE CHAINS Assuming the volume-additivity principle,ll P for any zwitterionic a-amino acid other than glycine can be written as + + Vo{H3NCH(R)COO-} = Vo(H3NCH2COO-) + P(CH3R) - Vo(CH4) 2027 where R denotes the side chain.2028 PARTIAL MOLAR VOLUMES OF OI-AMINO ACIDS Table 1. Partial molar volumes, FO, of a-amino acids, H,kCH(R)COO-, with non-ionisable side chains in water at 25 "C V0/cm3 mo1-l abbrevia- A P/cm3 mo1-l a-amino acid tion side chain experimental estimateda [ V'(est) - v"(exptl)] 1.alanine 2. valine 3. leucine 4. asparagine 5. glutamine 6. serine 7. threonine 8. cysteine 9. methionine 10. phenylalanine 11. tyrosine Ala Val Leu Asn Gln Ser Thr CYS Met Phe TYr CH3 CH(CH3)2 CH2CH(CH3)2 CH2CONH2 CH20H CHOHCH, CH,SH (CH,),CONH, (CH2)2SCH3 CH2C6H5 CH,C6H,0H 60.3b 91.3c 107.5" 77.6d 93.9b 60.8c 76.8e 73.g 1O5.lb 121.9 1 23.6b 61.3 91.3 107.8 77.1 93.4 61.3 78.2 73.4 105.7 121.5 124.5 1 .o 0.0 0.3 -0.5 - 0.5 0.5 1.4 0.0 0.6 0.0 0.9 a Literature references for P' values used in the estimations are: glycine, ref. (3); n-alkanes and benzene, ref. (12); ethanol and ethyl sulphide, ref. (14); propan-2-01, ref.(15); n-alkyl amides, ref. (18); hydrogen, ref. (19); phenol ref. (13); ref. (2); ref. (16); ref. (17); ref. (4); f ref. (3). In order to calculate P of amino acids it is necessary that the Vo data of glycine, methane and CH,R species in aqueous media are used, so that interactions of the solvent with the solutes are contained in the P data. Values of P of eleven a-amino acids with non-ionisable side chains were estimated using the above expression (method 1). The relevant data are incorporated in table 1. For eight systems (Val, Leu, Phe, Asn, Gln, Ser, Cys and Met) there is excellent agreement with experimental P values (table l), while good agreement was obtained for the remaining three amino acids (Ala, Tyr and Thr). values of zwitterions of non-acidic/basic a-amino acids is found to be successful irrespective of whether the side chain is made up of hydrophobic alkyl or aromatic groups, polar alcoholic or amide groups or sulphhydryl groups.The partial molar volume of glycine, with its connotations of zwitterionic nature, partial overlap of positive and negative charges and associated electrostriction, is obviously a basic and common component in the of all the eleven zwitterionic a-amino acids. The fact that the volume contributions of the side chains are related to those of appropriate CH,R groups would imply that the charges on the a-amino and a-carboxylate groups do not in any significant way affect either the volumes of the side-chain functional groups, whether they are in the 8- or y-position,2 or tEe hydrophobic hydration of the CH, groups.2o The present method of deducing PARTIAL MOLAR VOLUMES OF a-AMINO ACIDS WITH IONISABLE SIDE CHAINS BASIC AMINO ACIDS, LYSINE (Lys) AND ARGININE (Arg) In the case of amino acids with an additional charge centre located on the side chain, their aqueous solutions contain not only zwitterionic species but also singly, doublyMI V.R. RAO, M. ATREYI AND M. R. RAJESWARI 2029 and triply charged species, depending on the pH of the solution. For example, various species of lysine that can exist in solution are: + + + + H,NCH[(CH,),NH,]COOH H,NCH[(CH,),NH,]COO- cation (i) cation (ii) + H,NCH[(CH,),NH,]COO- H,NCH[(CH,),NH,]COO-. zwitterion anion So, any ro values determined experimentally for such systems would be a weighted average of the P of all the solute species present in solution. In view of the above, To values of individual ionic species of lysine were estimated by the method used for a-amino acids with non-ionisable side chains. In the case of cation (i) of lysine, for example, of the glycine cation6 instead of that of the zwitterion needs to be taken into account; similarly, the values of appropriately charged (CH,R) species must be used.In the case of the other basic amino acid, arginine, the method of estimation was slightly modified because appropriate data on the n-butyl guanidine cation, [CH,(CH,),g&], required for the estimation were not available in the literature. v" of cation (i) of arginine was therefore calculated as indicated below: + + ~(H,NCH[(CH,),NHC(NH,),]COOH) + + = V"(H,NCH,COOH)+ P[C(NH,),] + P(C,H,) -2P(H,).To values of cation (ii) and the zwitterion of arginine were also estimated using the above procedure. However, Vo of the arginine anion could not be evaluated as v" data on guanidine (unchanged) are not available in the literature. The Vo data for the two basic amino acids are collated in table 2 along with the experimental values determined in our laboratory using precision densitimetry.6 The agreement between experimental and estimated values is good for the zwitterion and cation (ii) of both lysine and arginine, while in the case of cation (i) the agreement is only fair. No experimental data on the lysine anion were available for comparison. Estimation of partial molar volumes of various species of the other basic amino acid, histidine, could not be made as there are no data available for imidazole and the imidazole ion.ASPARTIC ACID (Asp) AND GLUTAMIC ACID (Glu) In the case of the acidic amino acids Asp and Glu, one needs to consider one cationic, one zwitterionic and two anionic species. The various species of aspartic acid are : H,NCH(CH,COOH)COOH H,NGH(CH,COOH)COO- + + cation zwitterion + H3NCH(CH,C00-)C00- H2NCH(CH,C00-)C00-. anion (i) anion (ii)2030 PARTIAL MOLAR VOLUMES OF a-AMINO ACIDS Table 2. Partial molar volumes, p, of ionic species of lysine and arginine in water at 20 "C species experimental* estimateda lysine cation (i), [H,&CH(k)COOH] cation (ii), [H,fiCH(k)COO-] zwitterion, [H,NCH(k)COO-] anion, [H,NCH(R)COO-] arginine cation (i), [H,&CH(k)COOH] cation (ii), [H,fiCH(k)COO-] zwi t terion, [H ,NCH (k)COO-] anion, [H,NCH(R)COO-] 108.71 102.60 109.01 - 126.27 116.11 126.26 - 105.73 104.07 1 10.40 120.10 120.93 116.66 125.60 - a Literature references for values used in the estimations are: glycine, ref.(6); n-pentyl- amine, and its cation, ref. (21) and (22); guanidinium, ref. (23); methane and propane, ref. (12); hydrogen, ref. (1 9). The p values for the Asp and Glu systems are estimated as explained earlier and are given in column 3 of table 3; experimental data for the various ionic species are also included in the same table. A comparison of the estimated To and our experimental valuess shows that there is good agreement only in the case of two species, the cation and anion (i). However, p of the zwitterionic species is overestimated by ca.4 cm3 mol-l, and on the other hand p of the doubly negatively charged anion (ii) is underestimated by ca. 6-10 ~ m ~ m o l - ~ . The group volume of the COOH group has a higher value (34.3 cm3 mol-l) in monocarboxylic acids than in dicarboxylic acids (25.8 cm3 m0l-l);~~9 25 this is also true for the volume occupied by the carboxylate ion in mono- and di-carboxylate salts (26.3 and 17.3 cm3 m01-~).~~9 25 Since Asp and Glu are dicarboxylic acids, the procedure for estimating values was modified by starting with the volume of the dicarboxylic acid instead of that of glycine. For example, for the aspartic acid cation the modified procedure would be as follows (method 2): F'[H,&CH(CH,COOH)COOH] = P[(CH,COOH),] + p(H3C&H3) - P(CH,).The results of estimating ro by method 2 are included in column 4 of table 3. Again there is agreement only for two of the four ionic species. It is significant, however, that there is now good agreement with experiment for the cation and zwitterion in which the side-chain /l-COOH or y-COOH is un-ionised. However, when the side-chain carboxy group is ionised, as in both anions (i) and (ii), the are underestimated. This clearly suggests that the volumes of /l- and y-carboxylate groups in dicarboxylic amino acids are not the same as those in dicarboxylate salts. A possible reason for this is the following: the side-chain carboxy groups can come fairly close to the aminoM. V. R. RAO, M. ATREYI AND M. R. RAJESWARI 203 1 Table 3. Partial molar volumes, P, of ionic species of aspartic acid and glutamic acid in water at 20 "C estimateda by method : species experimentals 1 2 3 aspartic acid cation, [H,&CH(R)COOH] zwitterion, [H,kCH(R)COO-] anion (i), [H,&CH(R-)COO-] anion (ii), [H,NCH(R-)COO-] 77.48 77.63 78.34 76.72 70.89 62.72 74.00 70.62 69.30 76.19 73.69 76.16 68.14 79.82 68.29 74.43 glutamic acid cation, [H,kCH(R)COOH] zwi tterion, [H,&CH( R)COO-] anion (i), [H,kCH(R-)COO-] anion (ii), [H,NCH(R-)COO-] 94.43 94.33 94.54 93.46 88.12 92.86 86.75 87.36 78.60 89.88 84.30 91.70 84.42 102.83 83.83 90.43 a Literature references for P values used in the estimations are : glycine, /I-alanine and y-amino butyric acid (cations, zwitterions and anions), ref.(6), (2) and (28); n-mono- and di-carboxylic acids and their anions, ref.(21), (24) and (25); methyl amine and its cation, ref. (21) and (22); methane, ref. (12). group; in fact ring closure between the a-amino and y-carboxy groups in glutamic acid is known to lead to the formation of pyroglutamic acid.2s*27 Thus there could be an overlap of side-chain negative charge and the positive charge at the a-position, leading to reduced electrostriction and hence a higher observed Vo. Further, Shahidi and Farrel12 showed that in a,o-amino acids reduced electrostriction caused by the overlap of charges is effective until the charges are separated by six or more CH, groups. In view of the above, it would be of interest to consider the use of V O data for systems wherein the interaction of hH3 with COO- is built into the data, as for example in a,o-amino acids.This procedure is delineated below for the aspartic acid anion (i) (method 3): P[H,&CH(CH,COO-)COO-] = P(H3&CH2CH,COO-) + p(CH3COO-) - P(CH,). Essentially, this method is similar to the methods described earlier, except that the a-COO- species is now considered to be part of a monocarboxylate anion. Partial molar volumes obtained by this method are given in column 5 in table 3. This method is clearly superior to the other two methods in that the estimation of To for the cation, zwitterion and anion (i) is in good agreement with experiment. However, T O of anion2032 PARTIAL MOLAR VOLUMES OF a-AMINO ACIDS (ii), particularly for Glu, is considerably underestimated; this was also the case with the other two methods.The present procedure of estimating v" values of a-amino acids by method 1 is found to be applicable not only to zwitterionic systems but also to species with charged side chains, as in the case of amino acids with E - ~ H , and 8-guanidinium groups; however, the method was not successful with dicarboxylic amino acids. This should be seen in the following context: the side-chain functional group in qasic amino acids (Lys and Arg) is separated by as many as 5 or 4 CH, groups from the a-NH, and a-COO- species. Shahidi and Farrel14 reported that the volumes of proton ionisation of amino acids with nitrogen-containing side chains (iys, Arg and His) are independent of electrostatic effects caused by the terminal charged groups in the side chains. However, in Asp and Glu COO- is in the p- or y-position relative to a-NH, and the effect of overlapping of charges cannot be overlooked.M.R.R. thanks the Indian Council of Scientific and Industrial Research for financial assistance. J. C. Ahluwalia, C. Ostiguy, G. Perron and J. E. Desnoyers, Can. J. Chem., 1977, 55, 3364. * F. Shahidi and P. G. Farrell, J. Chem. SOC., Faraday Trans. I , 1978, 74, 858. F. J. Millero, A. LuSordo and C. Shin, J. Phys. Chem., 1978, 82, 784. F. Shahidi and P. G. Farrell, J. Chem. SOC., Faraday Trans. I , 1981, 77, 963. S. Cabani, G. Conti, E. Matteoli and M. R. Tine, J. Chem. SOC., Faraday Trans. I , 1981, 77, 2377; 2385. M. V. R. Rao, M. Atreyi and M. R. Rajeswari, J. Phys. Chem., 1984, in press. C. H. Spink and I. Wadso, J. Chem. Thermodyn., 1975,7, 561.K. P. Prasad and J. C. Ahluwalia, J. Solution Chem., 1976, 5, 491. S. Cabani, G. Conti, E. Matteoli and A. Tani, J. Chem. SOC., Faraday Trans. I , 1977, 73, 476. lo J. T. Edward, P. G. Farrell and F. Shahidi, J. Chem. SOC., Faraday Trans. I , 1977, 73, 705. l 1 J. Traube, Sammlung Chemischer und Chemisch-Technisch Vurtrage, 1899, 4, 255. l 2 W. L. Masterton, J. Chem. Phys., 1954, 22, 1830. l 3 S. D. Hammann and S. C. Lim, Austr. J. Chem., 1954, 7, 329. l4 S. Teresawa, H. Itsuki and S. Arakawa, J. Phys. Chem., 1975, 79, 2345. l5 C. Jolicoeur and G. Lacroix, Can. J. Chem., 1976, 54, 624. l6 E. J. Cohn, T. L. McMeekin, J. T. Edsall and M. H. Blanchard, J. Am. Chem. SOC., 1934, 56, 784. I s F. Shahidi, P. G. Farrell and J. T. Edward, J. Chem. SOC., Faraday Trans. I , 1977, 73, 715. l9 D. D. Eley, Trans. Faraday SOC., 1939, 1421. *O S. Cabani, V. Mollica, L. Lepori and S. T. Lobo, J. Phys. Chem., 1977, 81, 987. 21 M. Sakurai, T. Komatsu and T. Nakagawa, Bull. Chem. SOC. Jpn, 1975,48, 3491. 22 S. Cabani, G. Conti and L. Lepori, J. Phys. Chem., 1974, 78, 1030. 23 K. Miyajima, H. Yoshida and M. Nakagaki, Bull. Chem. SOC. Jpn, 1978, 51, 2508. 24 H. Heriland, Acta Chem. Scand., Part A, 1974, 28, 699. 25 H. Heriland, J. Chem. SOC., Faraday Trans. I , 1975, 71, 797. 26 A, Menozzi and G. Appiani, Gazz. Chim. Ital., 1892, 22, 105; 1894, 24, 370. 27 E. Abderharden and K. Kautzsch, Z. Physiol. Chem., 1909, 64,447; 1910, 68, 487. 28 F. Shahidi, J. Chem. SOC., Faraday Trans. I , 1980, 76, 101. J. P. Greenstein and J. Wymann, J. Am. Chem. SOC., 1936, 58, 463. (PAPER 3/ 10 1 5)

ISSN:0300-9599    

DOI:10.1039/F19848002027

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. I, 1984, $0. 2033-2038 Physical and Chemical Properties of Inactive Surface Hydrogen-bonded Hydroxyl Groups BY SEIICHI KONDO,* HIROFUMI YAMAUCHI, YASUO KAJIYAMA AND TATSUO ISHIKAWA Laboratory of Physical Chemistry, Osaka University of Education, 4-88 Minamikawahori-cho, Tennoji-ku, Osaka 543, Japan Receitled 25th July, 1983 The physical and chemical properties of surface hydrogen-bonded OH groups, formed by the hydrothermal reaction of silica hydrogels, xerogels, precipitated silicas (made with sodium silicate in the liquid phase at pH 11 and > 373 K) and aerosils, have been studied. In contrast to free OH groups, surface hydrogen-bonded OH groups, having an absorption band at 3500 cm-l, have low enthalpies of adsorption of water, methanol and methyl iodide, can be reversibly dehydroxylated and rehydroxylated by the adsorption of water vapour, and can be easily changed to OD groups by the adsorption of heavy water.These OH groups cannot be methoxylated with chlorotrimethylsilane or methanol. Cations such as Co2+ do not seem to be ion exchanged with the protons of these OH groups. Surface OH groups play an important role in controlling the properties of colloidal silicas. According to a series of studies made in this laboratory, silica gels made under acidic or neutral conditions have three kinds of OH groups: (a) free OH groups, which are physically and chemically active and have a sharp i.r. band at ca. 3750 cm-l, (b) interparticle hydrogen-bonded OH groups, which are at the area of contact between primary silica particles and whose i.r.band is broad with maximum at ca. 3600 cm-l, and (c) inner OH groups, which lie inside the closed pores and/or in the substrate and have a broad and unsymmetrical i.r. band at 3670 cm-l. The population of these OH groups depends upon the conditions of thermal and hydrothermal pretreatments. 1-4 In contrast to these OH groups, silica gels made by hydrothermal treatment of silica hydrogels at pH > 11, adjusted with either sodium hydroxide or .ammonia, and at 373 K for 2 h (denoted ATHG) were found to possess a new type of OH group with an i.r. band at ca. 3500 cm-l. This band is accompanied by a small band from free OH groups and a large band from inner OH groups. The OH groups with a band at 3500 cm-l can be changed to OD groups as easily as free OH groups by the ad- sorption of heavy water.Isosteric enthalpies of adsorption of water and methanol onto this surface at coverages below monolayer are lower than those of condensation, in contrast to the case of ordinary silica gels which have free OH g r o ~ p s . ~ Since this type of inactive surface hydrogen-bonded OH groups is new, it is interesting to see if it is possible to produce this type of surface OH group from different materials and to characterize these OH groups in more detail. EXPERIMENTAL Silica hydrosols made by hydrolysis of redistilled tetraethoxysilane at pH 2 and 343 K for ca. 100 min were gelled to thin flat plates -= 0.1 mm thick, washed with water to remove ethanol 20332034 SURFACE OH GROUPS ON SILICAS Table 1.Surface properties of the samples investigated materials As/m2 g-l C 5/cm3 g-' r/nm No,/nm-2 a NH20/nm-2 UTXG 74 1 73 0.65 1.88 5.5 5 19 9 ATHG 260 0.70 - 10.4 25.5 8 ATXG 172 267 - UTA 191 109 5 ATA 169 126 - PS 376 62 - 3 2.3 - - - - - - - - ~~ a Number of OH groups per nm2. Number of monolayer water molecules per nm2. and the pH was adjusted with hydrochloric acid. These hydrogels were hydrothermally treated with water at pH 5.7 for 6 h and dried to xerogels (denoted UTXG). Xerogels mentioned above were treated at pH 1 1.9, the pH being adjusted with sodium hydroxide or ammonia, and 373 K for 2 h, washed with dilute hydrochloric acid and water and then dried (denoted ATXG). Precipitated silicas were made from sodium silicate as follows.6 A 13 wt % solution of sulphuric acid was mixed with a 0.5 wt% solution of sodium silicate of molar ratio 3.25 at 363 K, keeping the pH at ca.10, and then a 3 wt% solution of sulphuric acid and a 10.7% solution of sodium silicate were added simultaneously at 363 K keeping the pH at 11.4. The precipitates were washed thoroughly with water using a centrifuge, then with dilute hydrochloric acid at pH 3.0 and finally with water before being dried (denoted PS). Alkali-treated aerosils (denoted ATA) were made by treating aerosils 200 (denoted UTA) hydrothermally at pH 11.8 and 373 K for 12 h, and then by following the same washing procedure as for PS. Some of the surface properties of these materials are listed in table 1. Isosteric enthalpies of adsorption of water for these samples were computed from the Clausius-Clapeyron equation using adsorption isotherms obtained at 288, 298 and 308 K by means of a gravimetric method.' The number of monolayer water molecules per nm2 (NFz0) was computed using the B.E.T.equation. 1.r. spectra from 4000 to 2000 cm-l were obtained using thin plates of UTXG and ATXG and thin discs of PS, UTA and ATA (pressures as low as 800 kg were used in order to minimize the effect of particle contact) and also using samples physically and chemically treated after synthesis. An ordinary dispersion-type spectrometer and a vacuum and high-temperature cell reported elsewhere' were used. OH concentrations at a given temperature were calculated from the weight difference between that at 443 K and that at the given temperature.Methoxylation of some of these samples was carried out with methanol in an autoclave at 473 K. Trimethylsilylation was done by immersing the samples in a benzene solution of trimethylsilyl chloride at room temperature. The ion-exchange ability was examined by immersing the samples in a 1 mol dm-3 solution of cobalt(1r) chloride at pH 5 and 323 K. RESULTS AND DISCUSSION Nitrogen adsorption isotherms for UTXG, ATXG and ATA are illustrated in fig. 1. The isotherm for UTXG is typical of mesoporous gels, while ATXG is macro- porous with a small fraction of micropores, according to the shape of this isotherm,* similar to ATHG.5 The formation of a very small amount of micropores was also observed for ATA and PS. The reason for the formation of micropores in highly alkali treatment is not clear.1.r. spectra of the OH stretching region are shown in fig. 2 for UTXG, ATXG and ATXG after H-D isotope exchange, the spectra being taken in vacuo at room temperature after heating the samples at 443 K for 2 h in vacuo. UTXG has a prominent surface-free OH band and a small hydrogen-bonded OH band assigned to interparticle hydrogen-bonded OH groups, as reported el~ewhere.~ The free OHS. KONDO, H. YAMAUCHI, Y. KAJIYAMA AND T. lSHIKAWA 2035 400 - I ," 300 E --- z" B e 2 200 2 100 0.2 0.4 0.6 0.8 1.0 PIP, Fig. 1. Nitrogen adsorption isotherms for UTXG (- - -), ATXG (-) and ATA ( * * - a ) . 3800 3400 3000 2600 Fig. 2.1.r. absorption spectra for UTXG (-), ATXG ( - * * -) and ATXG after H-D isotope exchange by adsorption of heavy water (- - -).wavenumber/cm-'2036 SURFACE OH GROUPS ON SILICAS UTXG 3 I I I I 1 I adsorbed water/prnol g-' 2.0 4.0 6.0 Fig. 3. Isosteric enthalpies of adsorption of water to UTXG (O), ATXG (A), UTA (0) and ATA (0). Arrows show monolayer adsorption. band of ATXG is absent, an inner OH band is seen at 3670 cm-I and a new band appears at 3500 cm-'. This new band is similar to that of ATHG and can be changed to two OD bands by adsorption of heavy water, with maxima at ca. 2750 and 2600cm-'. The isotope ratio of the free OH band and the sharper OD band (3750/2750 = 1.36) is equal to that of the hydrogen-bonded 3500 cm-I band and the remaining OD band (3500/2600 = 1.36), so the 2600 cm-' band seems to correspond to the 3500 cm-l band. This and other experimental evidence mentioned below suggest that the OH groups responsible for the 3500 cm-l band may exist on open pores, being easily accessible to adsorptives of large molecular size.This band disappears on heat treatment at 673 K in vacuo and is rehydroxylated almost reversibly by in situ adsorption of water vapour at room temperature. The behaviour of the i.r. spectra is the same as that of ATHG. UTA has a strong and fairly sharp surface free OH band with a small interparticle hydrogen-bonded OH band. However, this spectrum changes markedly to that of ATA, which has a much weaker free OH band, an inner OH band and a 3500 cm-I band. Ca. 12 h were necessary in order to obtain the same intensity of the 3500 cm-l band as that of ATXG. PS also has an OH band of similar shape to that of ATXG.The B.E.T. N, specific surface area of ATXG after heating at 673 K is equal to that before heating, within experimental error, in spite of the dehydroxylation of these 3500 cm-l OH groups. This suggests that the dehydroxylation reaction does not accompany the change in substrate structure, unlike the case of dehydrqxylation of interparticle hydrogen-bonded OH groups. These experimental results show that all these alkali-treated silicas have surface hydrogen-bonded OH groups which are very similar to each other.S. KONDO, H. YAMAUCHI, Y. KAJIYAMA AND T. ISHIKAWA 2037 wavenumberlcm-' Fig. 4. 1.r. spectra of ATA (-) and ATA after treatment with liquid methanol at 472 K for 4 h ( * * - a). There was no change in the powder X-ray diffraction, using a 100 mA X-ray source, before and after alkali treatment, so there seems to be no observable change in the long-range bulk structure.The total number of OH groups (NOH) for sample ATXG in table 1 is much larger than that for UTXG, because ATXG has a large number of inner OH groups besides the surface hydrogen-bonded OH groups and free OH groups. This is compared with UTXG, which is mainly covered with free OH groups with a surface density of 3 nm-2, with interparticle hydrogen-bonded OH groups and inner OH groups constituting only a small minority. The amount of surface hydrogen-bonded OH groups can be obtained from the difference in weight at 423 and 673 K, which is the temperature of dehydroxylation of these OH groups and at which free and inner OH groups do not dehydroxylate.This value is ca. 5.5 nm-2. It might be of interest to postulate the structure of this surface as well as the free OH surface. The short-range bulk structure of these amorphous silicas is assumed to be similar to that of a-cri~tobalite,~ since the densities of both materials are nearly the sameand diffuse X-ray scattering gives approximately the same diffraction angles for strong diffractions. The surface covered with surface hydrogen-bonded OH groups appears to be similar to the (1 10) plane of this crystal, which can form a zigzag-type hydrogen-bond chain of OH groups on this surface plane, while the (111) plane can have isolated free OH groups. According to this model, surface hydrogen-bonded OH groups can be reversibly de- and re-hydroxylated and there should be little change in surface area.The solubility of the (1 10) plane would be lower than that of the (1 1 1) plane, which has more reactive free OH groups, so the area of the (1 10) plane can grow relatively more than that2038 SURFACE OH GROUPS ON SILICAS of the (1 11) plane under the severe hydrothermal treatment where the solubility of silica is very high. The isosteric enthalpies of adsorption (Qiso) of water onto UTXG, ATXG, UTA and ATA are plotted in fig. 3. For coverages less than monolayer, Qiso of UTXG and UTA are higher than the enthalpy of condensation of water. The number of water molecules in a monolayer for these samples is 5 and 2.3 nmW2, respectively, which suggests a one to one interaction between water molecules and free OH groups. On the other hand, Qiso of ATXG and ATA are high in the beginning, but become very low afterwards and then approach the enthalpy of condensation near monolayer coverage.This is because water molecules are adsorbed first by free OH groups remaining on the surface and then by the surface hydrogen-bonded OH groups, which may have a very low energy of interaction with water molecules. The large value of NHzo for ATXG, 8 nm-2, suggests also that water molecules are adsorbed non- specifically on this surface with weak interactions. NHzo for ATA is larger than for UTA but smaller than for ATXG, because the surface of ATA is covered with more free OH groups than the surface of UTXG, as is seen from the i.r. spectra of both materials in fig. 2. Qiso of methanol for ATXG below monolayer coverage is less than the enthalpy of condensation and Qiso of methyl iodide is roughly equal to the enthalpy of condensation, as was reported elsewhere. lo This suggests that surface hydrogen- bonded OH groups lie in pores of size greater than that of these molecules.All these results show that a similar inactive surface hydrogen-bonded OH group can be produced from different amorphous silicas by means of hydrothermal reaction at high pH and temperature. Surface free OH groups are reactive and can be alkoxylated easily by heating silica in liquid alcohol in an autoclave. l1 However, surface hydrogen-bonded OH groups cannot be methoxylated by thislmethod. Fig. 4 shows the i.r. spectra of ATA before and after treatment with methanol at 473 K for 4 h.The free OH band was displaced by C-H bands after the treatment, but there was almost no change in the surface hydrogen-bonded and inner OH bands. Methoxylation did not occur when methanol was adsorbed onto ATXG after dehydroxylation at 673 K in vacuo. By attacking ATXG and ATA with a reactive agent, trimethoxymonochlorosilane in benzene solution, at room temperature, free OH groups were replaced by trimethylsilyl groups, but there was no change in the 3500 cm-l band. The protons of the free OH groups can be exchanged by cations such as Co2+ in an aqueous environment and at suitable pH. However, the intensity of the 3500 cm-l band shows no observable change on immersing ATXG in cobalt(I1) chloride solution under the same conditions as mentioned elsewhere,12 although the free OH band disappeared. These results show that surface hydrogen-bonded OH groups are more stable to chemical treatment than free OH groups. S. Kondo, M. Muroya and K. Fujii, Bull. Chem. Soc. Jpn, 1974, 47, 553. S. Kondo, H. Fujiwara and M. Muroya, J. Colloid Interface Sci., 1976, 55, 421. S. Kondo, K. Tomoi and C. Pak, Bull. Chem. SOC. Jpn, 1979,52, 2046. S. Kondo, Y . Saito, T. Ishikawa and H. Yamauchi, Stud. Surf. Sci. Catal., 1982, 10, 241. R. K. Iler, The Chemistry of Silica (Wiley-Interscience, New York, 1979). K. Naoki, E. Amano and S. Kondo, Mem. Osaka Univ. Educ. III, 1977, 26, 1. * K. S. W. Sing and S. J. Gregg, Adsorption, Surface Area and Porosity (Academic Press, London, 1982). R. K. Iler, The Colloid Chemistry of Silica and Silicates (Cornell University Press, New York, 1955). lo T. Ishikawa, E. Amano, M. Muroya and S. Kondo, Chem. Lett., 1983, 415. l1 S. Kondo, H. Fujiwara, E. Okazaki and T. Ichii, J. Colloid Interface Sci., 1980, 75, 328; H. Utsuki, l2 S. Kondo, T. Ishikawa, H. Yamauchi, H. Yamaoka, E. Amano, A. Fukushima and M. Ono, Nippon .I S. Kondo, H. Fujiwara, T. Ichii and I. Tsuboi, J. Chem. Soc., Faraday Trans. 1, 1979, 75, 646. Hyomen, 1978, 16, 525. Kagaku Zasshi, 1983, 61 1. (PAPER 3/ 1280)

ISSN:0300-9599    

DOI:10.1039/F19848002033

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1, 1984, 80, 2039-2052 Thermodynamic and Kinetic Study of the Binding of Azo-dyes to a-Cyclodextrin BY ANNE HERSEY~ AND BRIAN H. ROBINSON* University Chemical Laboratory, University of Kent at Canterbury, Canterbury, Kent CT2 7NH Received 5th August, 1983 A thermodynamic and kinetic study concerning the binding of azo-dyes to a-cyclodextrin is described. Joule-heating temperature-jump and stopped-flow techniques with optical detection have been used to elucidate the mechanism of inclusion-complex formation. Azo-dyes form 1 : 1 inclusion complexes with a-cyclodextrin in which the dye is located inside the cyclodextrin cavity. The mechanism of insertion involves a fast pre-equilibrium step to form an intermediate. In a subsequent, slower, rate-determining step the intermediate is converted into the final stable complex. Comparing rates of inclusion for a number of azo-dyes containing two substituted phenyl groups separated by an azo-linkage, it is possible to determine which aromatic ring of the dye preferentially enters the cyclodextrin cavity during the binding process.The forward and reverse rate constants for the inclusion process are very dependent on the nature of the substituent groups on the phenyl rings. Charge and steric effects are found to be particularly important in influencing the kinetics. The overall equilibrium constant for complex formation does not vary much as the dye substituents are altered. An attempt is made to interpret the proposed mechanism in terms of the molecular rearrangements which take place as the dye binds to the cyclodextrin.Cyclodextrins are cyclic compounds containing glucose units fused in a ring. The ring-like structure of the cyclodextrin is formed with a central void. a-, p- and y-cyclodextrins are made up of 6, 7 and 8 glucose units and have internal diameters of 0.50,l 0.672 and 0.85 nm,3 respectively. Cyclodextrins provide interesting models for certain aspects of enzyme behaviour. They form inclusion complexes in aqueous solution with a wide variety of substrates including dye^,^-^ drugs7 and small ions.* These substrates normally form inclusion complexes with a 1 : 1 stoichiometry, although other stoichiometries have been observed. Much of the work undertaken on cyclodextrins has been concerned with the thermodynamics of inclusion-complex formation.To a lesser extent the kinetics have also been studied. The insertion process is representative of the first step in enzyme-substrate reactions. In common with enzymes, cyclodextrins also catalyse a number of chemical reactions after complexation has taken place.10-12 It is generally accepted that substrate/cyclodextrin complexes are indeed inclusion complexes, evidence being provided from a variety of sources, e.g. ~pectrophotometry,~~ fluorescence6 and n.m.r. There has been much discussion as to which forces are responsible for the stability of inclusion complexes. The binding interaction may involve van der Waals forces,15 t Present address : Biochemistry Department, Wellcome Research Laboratories, Beckenham, Kent. 20392040 BINDING OF AZO-DYES TO a-CYCLODEXTRIN hydrogen bonding,16 release of incompletely hydrogen-bonded water (so called ' high-energy' water) from the cyclodextrin cavity1' and release of strain energy in the cyclodextrin ring.lx Apparently all these forces are involved in the inclusion process, but their relative magnitudes depend on the nature of the substrate.To date relatively few kinetic studies have been undertaken of the rates, and hence mechanism, of formation of inclusion complexes. Dyes have been largely favoured as species suitable for kinetic studies because they frequently exhibit a visible spectral change on binding to cyclodextrins which is easily monitored. Cramer et a1.6 studied the kinetics of inclusion-complex formation using temperature- jump relaxation spectrometry. They studied dyes of the type - 0 3 3 / Y N'h-+x where the groups X, Y and Z were varied.Dyes of this type were chosen because the naphthyl ring was too large to be incorporated into the cavity. The rate constants for inclusion are therefore measured unambiguously for the smaller phenyl ring. Their kinetic results were consistent with a simple reaction scheme of the type k, D + aCD + D/aCD k-i where D is the dye, aCD is a-cyclodextrin and D/aCD represents a 1 : 1 dyela- cyclodextrin complex. As the group Z is varied from -H to -CH, to -C,H, (X = Y = H) the rate constants for association and dissociation both decreased dramatically. The equilibrium constants K( = k l / k - , ) were largely unchanged. It was concluded that rate variations can only be explained in terms of increased steric hindrance as the substituent Z is increased in size. More recent studies by Rohrbachlg and Fujimoto et ul.,20~21 also using the temperature-jump technique, were also consistent with a 1 : 1 interaction.However, there may be some indication from Rohrbach's experimental data that k1/k-, < K , where K is the value determined spectrophotometrically. Rohrbach attributed this to experimental error, but it may be that the kinetics of these systems are more complex than the above A + B $ C model suggests. The nature of the driving force for inclusion and the mechanism of the inclusion process are still not resolved. To this end, in this paper the kinetics of interaction of simple azo-dyes with a-cyclodextrin have been studied.Azo-dyes based on two phenyl rings are used, in preference to the larger structures of Cramer el because of their similarity to certain drug molecules, However, both ends of the dye are then potentially able to penetrate into the cavity. This is, of course, also a problem encountered when studying drug/receptor binding. The kinetics of the inclusion process are investigated over a temperature range in order to obtain values for the activation parameters AH+ and A,!?, which should further help to elucidate the mechanism of the inclusion process.A. HERSEY AND B. H. ROBINSON 204 1 Table 1. Azo-salicylate dyes used in this work -CH3 -H -H -H -H -H -H -CH, -H -H -H -H -CH, -H -H -H -CH3 -H -so, -so, -NH, -H -c1 -OC,H, -CH3 -H -CH,CH,CH, -H -H -H -CH3 -H -H -H -CH3 -H EXPERIMENTAL MATERIALS The dyes used in this work were azo-salicylates of the type R 4 \ where substituents R, to R, are described in table 1.Their abbreviated forms are also indicated by a letter. Most dyes were prepared by the diazo-coupling reaction between appropriately substituted diazonium salts and the substituted salicylic acid. The diazonium salt was first prepared by the reaction of nitrous acid on the aniline derivative. Standard procedures were used for purification and this was checked by elemental analysis. a-Cyclodextrin was obtained from Koch-Light and used without further purification. Biebricht Scarlet (I) Two other dyes were used: -0 -Q- "\\N q - SO3- ykg / \ - \ / and pyridine-2-azo-p-dimethyl aniline (J) METHODS All spectra were recorded on a Cary 219 spectrophotometer using matched 1 cm silica cells.The main apparatus used for the kinetic experiments was a joule-heating double-beam temperature-jump instrument wi th spectrophotometricdetection from Messanlagen (Gottingen). 67 FAR 12042 BINDING OF AZO-DYES TO a-CYCLODEXTRIN Sodiumchloride (at 0.1 mol dm-3) wasusedas theconductingelectrolyte in the temperature-jump experiments as C1- binds very weakly to a-cyc10dextrin.l~ A small-volume stopped-flow instrument with spectrophotometric detection was also used. The temperature control on all equipment was to & 0.1 "C and the temperature was monitored using a Cr/Al thermocouple connected to a Comark 1604 electronic thermometer. All reactions were studied under conditions such that the kinetics yielded single-exponential traces, within experimental error.These traces were analysed using a trace-matching procedure based on that of Crooks et a1.22 The absorption spectra of the dyes were recorded over the pH range 2-13 to ensure that at the pH of a given experiment (usually ca. pH 7) there was only one absorbing species of the dye present. The solution pH was adjusted manually using sodium hydroxide or hydrochloric acid as appropriate. The use of buffers was avoided as many inorganic salts are known to bind to cyclodextrins.*. l9 RESULTS In order to study the rates and mechanism of incorporation of azo-dyes into the cyclodextrin cavity it is important to ascertain which end of the dye molecule is being inserted and hence for which substituted phenyl ring the rate of inclusion is being measured.The azo-dye (A) is particularly interesting. By comparison with the work of Cramer et aL6 the methylsalicylate group of (A) is too large to be incorporated into the a-cyclodextrin cavity. Therefore only the phenyl ring containing the sulphonate group is able to insert into the cavity. Subsequent studies on the binding of other azo-dyes containing a sulphonate group at one end should enable the determination of which substituted phenyl group is being incorporated in each case. Values of k , similar to (A) would indicate incorporation by a sulphonate-substituted phenyl ring whereas a faster rate would indicate inclusion by the other substituted phenyl group. It is assumed that the phenyl ring not being inserted does not significantly influence the insertion rate.However, there may well be an effect on the exit rate. The absorbance changes which occur when (A) binds to a-cyclodextrin are shown in fig. 1 . The equilibrium constant ( K ) for binding of (A) to a-cyclodextrin was determined from the spectra using a modified Benesi-Hildebrand method. The equilibrium constant measured over a 20-fold range of [aCD], (where the subscript T denotes weighed-in concentrations) was apparently consistent with a 1 : 1 interaction of the type D+aCDeD/aCD. Kinetic measurements were made using the stopped-flow method. Pseudo-first-order conditions were employed with [aCD], 9 [DIT and single-exponential traces were observed. Observed rate constants (kobs) were independent of wavelength and dye concentration.For a reaction of the type K k, k-1 D + aCD + + D/aCD (where K = k , / k - , ) under the conditions of the experiment where [aCD], is the total concentration of a-cyclodextrin added. Hence for the above scheme a linear plot of kobs against [aCD], with slope k , and intercept k-, should be obtained. Plots of kobs against [aCD], at different temperatures are shown in fig. 2. The plotsA. HERSEY AND B. H. ROBINSON 2043 Fig. [AIT 300 350 400 hlnm 1. Absorption spectra for (A) binding to a-cyclodextrin. T = 25 “C; = 3 x mol dm-3; [aCD], = 0 (a), (b), 3 x (c), 5 x (d), 1 x (e) and 2 x low3 (j) mol dm-3. are clearly not linear and hence it has to be concluded that the results are not consistent with the simple 1 : 1 interaction described above. The plots can, however, be most simply interpreted in terms of a mechanism of the type K1 kz D + aCD D/aCD e D / a C D ’ fast k-2 slow where the dye reacts with the cyclodextrin in a fast-equilibrium step to give an intermediate (D/aCD) which then reacts more slowly to form the final complex (DlaCD’).that the rate expression (for [aCD], % [D])T reduces to By assuming a mechanism of this type it can be + k2. (2) KI k2WDlT (1 + &[aCDI,) kobs = 67-22044 BINDING OF AZO-DYES TO a-CYCLODEXTRIN 2 .o - I v) > a0 1.0 0 0 1 2 3 4 [ C X C D ] ~ / I O - ~ mol dm-3 Fig. 2. Plots of kobs against [aCD], for (A) binding to a-cyclodextrin. [AIT = 3.0 x lop5 mol dmp3; T = 15.0 ( x), 25.0 (0) and 35.0 (0) "C. 12.0 5 8.0 I h N * 0 .y, 4.0 0 0 1 2 3 4 [a CDI +' / 1 O3 dm3 mol-' Fig.3. Plots of (kobs - k-J1 against [ a c D ] ~ ~ for (A) binding to a-cyclodextrin. [AIT = 3.0 x lop5 mol dm-3; T = 15.0 ( x), 25.0 (0) and 35.0 (0) "C. At high [aCD],, KJaCD], 9 1 and kobs = k,+k-,. At low [aCD],, K,[aCD], Q 1 and kobs = Kl k,[aCD], + k-,. This is consistent with the shape of the plots in fig. 2. Rearrangement of eqn (2) gives (kobs - k-,)-l = (Kl kz[aCD],)-l + k;l. (3) A good estimate of k , can be made from the intercept of the plot of kobs against [aCD], and consequently if (kobs - k-?)-' is plotted against [aCD]+l a straight line of slope ( K , k2)-l and intercept (k2)-l will result. The ' best-fit ' of the data was actually achievedA. HERSEY AND B. H. ROBINSON 2045 Table 2. Rate and equilibrium data for the binding of (A) to a-cyclodextrin K , + K , K , = KT/dm3 mol-' spectro- kinetic photometric T/"C k,/s-' k-z/s-2 K1/dm3 molt' K, determination determination ~ 15.0 0.78k0.08 0.1 1 640+80 7.1 f0.7 5200f810 - 25.0 1.6f0.1 0.28 430 f 40 5.7 f0.4 2900 f 260 2800 f 300 35.0 3.0f0.2 0.61 350+40 4.9k0.3 2100k220 __ Table 3.Comparison of equilibrium constants obtained by spectrophotometry (Kspec) and stopped-flow kinetics (Kkin) Kspe,/dm3 mo1-I range of [aCD], Kkin/dm3 mol-l range of [aCD], dye (T= 25°C) /mol dmP3 (T = 25 "C) /mol dm-3 3700 f 400 1 x 10-4-5 x 1 800 f 400 9900 f 1000 5 x 10-5-1 x 10-3 1200 * 200 3300 Ifr 200 1 x 10-4-5 x 1300 f 300 5 x 10-5-2 x lop3 800 & 200 4000 f 200 4100 & 300 5 x 10-5-2 x 10-3 1100f200 5 x 10-4-2 x 10-3 5 x 10-5-1 x 10-3 I x 10-4-2 x 103 5 x 10-5-2 x I 0-3 3 x 10-5-2 x 10-4 (B) (C) (D) (E) (F) 3 x 10-5-1.5 x lo3 (G) 1300 k 100 1 x 10-4-3 x 400 2 80 using an iterative computer procedure.Plots of (kobs -kJ1 against [aCDl~l are shown in fig. 3 and are linear as predicted. that a mechanism of the type described above is also consistent with the appearance of an isosbestic point in the spectra, as is observed (see fig. 1). The concentration-independent equilibrium constant previously calculated using a modified Benesi-Hildebrand method is also consistent with this scheme. The equili- brium constant (KT) is, however, a composite one and equal to ( K l + K l K J . A summary of the rate and equilibrium data at several temperatures for (A) binding to a-cyclodextrin is shown in table 2. Good agreement is obtained between kinetic and equilibrium determinations of KT.The binding of a number of other azo-dyes to a-cyclodextrin was also studied. The temperature-jump technique was used for all the dyes except (B) at ca. pH 13 and (I) when the stopped-flow method was used. When the temperature-jump results were analysed according to the simple scheme involving a single product, it was found that K, determined spectrophotometrically, was greater than k , / k - , , determined kinetically (table 3). Again it is implied that a more complex mechanism than the simple 1 : 1 interaction is in operation. If a mechanism of the type used to interpret the results of dye (A) is employed, the discrepancy between Kkin and Kspec can adequately be explained. On first inspection plots of z-l (= /cobs) against [aCD], appear to be reasonably linear, but the experimental error is larger than that for the stopped-flow measurements as AHo (which determines the amplitude of the temperature-jump traces) is small.Closer examination of the plots reveals a slight but systematic curvature. The temperature-jump method is a relaxation technique, and therefore in order to perturb It can be2046 BINDING OF AZO-DYES TO Q-CYCLODEXTRIN 8 - I m 0 - ---- 3 4 0 4 8 12 [a!CDIT/1O4 mol dm-3 Fig. 4. Plot of kobs against [aCD], for (C) binding to a-cyclodextrin. [A], = 4 X lop5 mOl dm-3; T = 25 "c. t 0 4 8 12 ([aCDl + [D])-'/103 dm3 mol-' Fig. 5. Plot of (kobs - kLJ1 against ([aCD] + [D])-' for (C) binding to a-cyclodextrin. [AIT = 4 x mol dm-3; T = 25 "C.the equilibrium significantly the concentration range studied must correspond to ca. 10-90% conversion of reactants into products. It was only at relatively high concentrations of aCD in the stopped-flow experiments with dye (A) that the curvature in the rate constant against concentration plots became pronounced, and this concentration range is inaccessible to study using the temperature-jump method. Relaxation theory predicts that for the two-step mechanism two relaxations are, in principle, observable. T'nese correspond to the fast and slow steps of the reaction. In practice the faster step, which has a relaxation time comparable with the heating time, gave very-low-amplitude traces presumably because either AHo was small and/or E,, x cDlaCD, where E is the extinction coefficient.Kinetic information couldA. HERSEY AND B. H. ROBINSON 2047 therefore only be obtained for the slow relaxation. Values of kobs were then obtained from the observed relaxation time (kobs = 7-l). Plots of kobs against [acD], and (kobs-k-2)-1 against ([D]+[aCD])-l are shown in fig. 4 and 5 for dye (C). For dye (I), where the kinetics were investigated using the stopped-flow method, pronounced curvature in the kobs against [aCDIT plot was observed. In addition to the work performed on the dye species which exists at ca. pH 7, charge effects on substrate binding were investigated by variation of the pH of the solution, hence changing the charge on the dye. This procedure was carried out for dyes (3) and (C) at ca. pH 13. At this pH the species of dye present in solution was In table 4 they are indicated as (B’) and (C’).From the effect on the inclusion rate, information can be obtained on which phenyl group of the dye is being incorporated into the cavity. In practice the results for dyes (B’) and (C’) at ca. pH 13 could adequately be explained in terms of a 1 : 1 interaction. If K,, as defined in eqn (2), is small then K1[aCD], 6 1 and eqn (2) approximates to kobs = Kl k,[aCDIT+k-,. A plot of kobs against [aCD], will then be linear with slope Klk, and intercept k-,. Similarly the equilibrium constant KT( = Kl + K , K,) becomes equal to K, K , when K2 $ K l . Under these conditions Kspec = Kkin, as is observed for these dyes binding to a-cyclodextrin. Hence in this situation no information concerning Kl (other than that it is small compared with [aCD],l) can be obtained. A summary of the rate and equilibrium constants for dyes binding to a-cyclodextrin at 25 “C is shown in table 4.Table 5 shows a comparison of the kinetic and thermodynamic parameters determined for dyes (A), (I) and (J). DISCUSSION In order to interpret properly the thermodynamic and kinetic data, it is necessary to identify which end of the dye substrate preferentially penetrates into the cyclodextrin cavity. After this has been established, an attempt can be made to consider the proposed reaction scheme in terms of (i) the nature of the intermediate (DIaCD) and the kinetically stable product (DlaCD’) and (ii) structural changes, e.g. water loss from the cavity, which occur on dye binding. As previously observed,6 our studies confirm that association equilibrium constants for a variety of azo-dyes of similar skeletal structure are of comparable magnitude, but rate constants for insertion and release vary over many orders of magnitude.These effects are not related to steric factors alone, and clearly both steric and charge effects are important. To establish which end of the dye is preferentially incorporated into the cavity, the following criteria have been used. (a) A single-exponential transient was always observed on the time scale for insertion, which suggests that for each dye investigated the kinetics and thermodynamics were dominated by the insertion and release of a preferred end. (b) For dyes with the same inserting group, the rates at which these groups enter the cavity are comparable, regardless of the structure and chemical nature of the rest of the dye molecule.(c) The release rate of the dye (as measured by k-,) is more dependent on the structure of the whole dye molecule. Both (b) and (c) imply that the transition state for insertion is reached when the dye is close to its final equilibrium position in the cavity. It was found that (H) did not bind to a-cyclodextrin. Therefore, on steric2048 BINDING OF AZO-DYES TO a-CYCLODEXTRIN Table 4. Summary of rate and equilibrium constants for dye binding to a-cyclodextrin KT/dm3 dye structure mol-l Kl/dm3 kP2/s-' k,/s-l mol-l K2 -025 I -02 c )=7 -02: CH3 -W, - 0 2 5 2800 3 700 3 50 9900 420 3300 4000 4100 1300 1.6 0.28 2.9 x 103 6.0 x 102 3 .0 ~ lo3 8.0 (Kl k2) 1 . 1 x 104 2 . 7 ~ 103 > lo7 ca. 3 x lo4 (K, k2) 4.4 x 103 7.0 x 102 5.7 x lo4 2.2 x lo4 1.9 x lo4 9.8 x 10' 3.1 x lo4 3 . 0 ~ lo4 430 620 - 2000 - 480 1100 1500 740 5.7 4.7 - 4.3 - 6.3 2.6 1.9 1 .oA. HERSEY AND B. H. ROBINSON 2049 Table 4. (conrinued) KT/dm3 K, /dm3 structure mol-' k-,/s-' k,/s-' mol-I K2 CH3 no inclusion 0.6 0.5 1.8 x 104 4.8 x 103 1100 1.2 670 3.7 6100 5 . 4 ~ lo3 2 . 6 ~ lo3 2000 2.1 1.2 x 104 3.2 x 102 6.3 x lo4 1 . 4 ~ lo2 (Kl k2) 160 38 considerations both the 3,Sdimethylbenzene species and the 3-methylsalicylate species are too large to be incorporated into the cavity. It is therefore clear that the benzenesulphonate group of (A) and the salicylate group of (D) are the relevant groups which are incorporated for these dye substrates. The rates of insertion and release of the benzenesulphonate group of (A) are similar to those for (I).On steric grounds it is clear that the benzenesulphonate group in (I) is the preferred site for incorporation so the consistency of kinetic behaviour for (A) and (I) supports the general approach. For (D) it is certain that the inclusion rate of the salicylate ion is being measured. Since (B) is incorporated at the same rate, it is clear that, in this case also, the salicylate group is the preferred end for insertion rather than the benzenesulphonate group, which is already known to interact much more slowly. The similarity of exit rates for (B) and (D) suggests that the nature of the other end of the molecule [whether charged as in (B) or providing a steric contribution as in (D)] does not dramatically affect the2050 BINDING OF AZO-DYES TO a-CYCLODEXTRIN Table 5.Thermodynamic and kinetic parameters for dyes (A), (I) and (J) parameter (A) (I) (J) AfG 47 51 50 A€E2 61 52 46 A% -83 -78 + 4 Asp2 - 5 2 -76 -19 A f C -22 -19 -33 A% -14 - 1 +4 A g -23 -5 -57 A$ -33 -2 +23 Units: AH, kJ mol-l; AS, J K-l mol l , exit rates for these particular dyes. Similarly the exit rates for (A) and (I) are not much different. For (C), (E), (F), (G) and (K) the rate of incorporation and/or release was faster than that measured for (D), so it would appear that the non-salicylate end of these dye molecules is preferentially involved in the insertion process. With (K) an unambiguous assignment is not possible, so it is likely that both ends can be incorporated.In the case of MO, the rate constant k , was very much larger than that measured for the benzenesulphonate groups of (A) and (I); hence the N-dimethylaniline group must be the preferred group for insertion. In the case of (J) the assignment is more difficult, but the available evidence favours preferential inclusion of the 2-pyridyl group. Confirmation of these assignments is provided from data for deprotonated dye species. The rate of incorporation and release of (B’) decreases quite dramatically compared with (B), as expected if the salicylate group is incorporated. However, from the data obtained for (B’) it is difficult to determine which of the two groups in (B’) is being incorporated, but the rather more tentative evidence, based on exit rates, favours the salicylate group.For (C) and (C’) the effect of deprotonation is actually to increase the exit rate constant, while the forward rate is not much affected. Hence in both cases the anilino group is being incorporated. In table 4 the group which inserts into the cyclodextrin cavity is indicated by an asterisk. In contrast to previous studies on substrate-cyclodextrin interactions, our kinetic studies suggest that a reaction scheme of the type K , ki D + aCD D/aCD 0 DIaCD’ fast k-2 slow is required in order to explain the experimental observations. In the inclusion mechanism the following considerations are likely to be of importance: (a) Approach of dye and cyclodextrin. The dye is likely to insert at the more open side of the conical cyclodextrin.(b) A loose interaction (manifested by K l ) will occur between the dye and the cyclodextrin before extensive penetration has occurred. (c) Solvating water molecules are removed from the phenyl group ofA. HERSEY AND B. H. ROBINSON 205 1 the inserting dye, and water (so-called ' high-energy ' water) is released from the cavity. (d) The dye inserts further into the cavity. This process may be associated with structural changes in the cyclodextrin and dye, and it is at this stage that the transition state is reached. (e) The final stable location of the dye is achieved, which probably corresponds to the location of the cyclodextrin collar over the azo group of the dye. (f) There may be resolvation of exposed parts of the dye protruding from the cavity following insertion.The values of K , are quite large, indicating that a reasonably strong complex is formed in the first stage of the reaction. The association rate constant is likely to be close to diffusion-controlled, so the 'lifetime' of the D/aCD complex is likely to be of the order of microseconds. The process is associated with a negative enthalpy change (AH," = - 20 to - 30 kJ mol-l), which is largely responsible for the observed stability. A detailed interpretation of these thermodynamic parameters is not justified, but it is likely that changes in hydrogen bonding, van der Waals interactions and solvation are involved. The similarity of K , values for all dyes studied suggests that an interaction between the phenyl group and the hydroxy groups of aCD may be involved.The subsequent slow step is associated with a large positive enthalpy of activation (ca. 50 kJ mol-1) for both neutral and charged dye-inserting groups. The dyes containing a charged group (e.g. SO;) are slowed down by an additional unfavourable contribution from the entropy of activation, which implies a mechanistic difference between the two types of inserting group. This may be caused by the charged group interacting with water in the cavity preventing its removal and hindering access. Neutral groups are all inserted with a relatively large rate constant ( k , ) which is largely independent of the inserting group. The salicylate group behaves differently to the --SO; group probably because an intramolecular hydrogen bond is formed between the -OH and COY groups.As a result this group is less able to interact with water molecules in the cavity. Removal of the intramolecular hydrogen bond by deprotonation results in a much slower rate of insertion, consistent with the proposed hypothesis. Protonation on the azo-nitrogen (in MO+) also has a dramatic effect on the rate of insertion. If K , is taken to be the same as for MO, the rate of insertion is slowed down by a factor of 50. Rather curiously, the exit rate is not much affected. A detailed analysis of the other thermodynamic (AH", ASo) and kinetic (AH+, A S ) parameters is not justified in view of the large number of weak interactions which are involved. We thank the S.E.R.C. and the Wellcome Foundation for a CASE Award (to A.H.).We are especially grateful to Dr R. M. Hyde and Dr R. A. Paterson of the Biochemistry Department, Wellcome Research, Beckenham, for many useful discussions. P. C. Manor and W. Saenger, J . Am. Chem. SOC., 1974 96, 3630. K. Lindner and W. Saenger, Angew. Chem., Int. Ed. Engl., 1978, 17, 694. J. M. MacLennan and J. J. Stezavski, Biochem. Biophys. Res. Commun., 1980, 92, 926. W. Lautsch, W. Broser, W. Biedermann and H. Grichtel, J. Polym. Sci., 1955, XVII, 479. M. Suzuki, Y. Sasaki and M. Sugiura, Chem. Pharm. Bull., 1979, 27, 1797. F. Cramer, W. Saenger and H-Ch. Spatz, J. Am. Chem. SOC., 1967,89, 14. ' G. E. Hardee, M. Otargri and J. H. Perrin, Acta Pharm. Suec., 1978, 15, 188. R. P. Rohrbach, L. J. Rodriguez, E. M. Eyring and J. F. Wojcik, J . Phys. Chem., 1977, 81, 944. K. A. Connors, S. Lin and A. B. Wang, J . Pharm. Sci., 1982, 71, 217. lo R. L. Van Etten, G. A. Clowes, J. F. Sebastian and M. L. Bender, J . Am. Chem. Soc., 1967,89,3253. D. E. Jutt and M. A. Schwartz, J . Am. Chem. SOC., 1971,93, 767. l 2 N. Hennrich and F. Cramer, J . Am. Chem. SOC., 1965,87, 1121. l 3 R. L. Van Etten, J. F. Sebastian, G. A. Clowes and M. L. Bender, J . Am. Chem. SOC., 1967,89,3242. l4 P. V. Demarco and A. L. Thakhar, Chem. Commun., 1970, 2.2052 BINDING OF AZO-DYES TO M-CYCLODEXTRIN l 5 R. J. Bergeron, M. A. Channing, G. J. Gibeily and D. M. Pillor, J . Am. Chem. SOC., 1977, 99, 5146. l 6 E. S. Hall and H. J. Ache, J. Phys. Chem., 1979, 83, 1805. D. W. Griffiths and M. L. Bender, Ado. C a u l . , 1973, 83, 1805. K. Lindner and W. Saenger, Agnew. Chem., Int. Ed. Engl., 1978, 17, 694. l R R. P. Rohrbach, Ph.D. Thesis (Villanova University, 1975). 2o N. Yoshida and M. Fujimoto, Chem. Lett., 1980, 23 1 . *I N. Yoshida and M. Fujimoto, Chem. Lett., 1980, 1377. ** J. E. Crooks, M. S . Zetter and P. A. Tregloan, J . Phys., 1970, 3, 73. 23 A. Riches, Ph.D. Thesis (University of Kent at Canterbury, 1982). (PAPER 3/ 1380)

ISSN:0300-9599    

DOI:10.1039/F19848002039

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1984, 80, 2053-2071 Raman Spectroscopic Studies of a Thionine-modified Electrode BY KENNETHUTCHINSON AND RONALD E. HESTER* Department of Chemistry, University of York, Heslington, York YO1 5DD AND W. JOHN ALBERY* AND A. ROBERT HILLMAN Department of Chemistry, Imperial College of Science and Technology, South Kensington, London SW7 2AY Received 6th September, 1983 Raman spectroscopy has been used to determine the structure of the surface coat of a thionine-modified gold electrode. Spectra have been observed from the thionine form of the coat; no spectra could be obtained from the leucothionine form. Evidence is presented to show that the surface coat exists as a polymeric film with the individual thionine units linked via an amino nitrogen in a secondary amine structure.There is no evidence of chemical bonding between the film and the electrode. Raman spectroscopy has also been used to study the movement of charge through the film as it is oxidised and reduced. Observed spectra are shown to be due to a combination of resonance- and surface-enhanced Raman scattering. No spectra could be obtained from either the thionine or the leucothionine forms of a thionine-modified platinum electrode. The deliberate modification of electrode surfaces by the immobilization of one or more layers of electrochemically reactive species has received considerable attention recently.lt2 The range of modification procedures now available and the variety of immobilized species employed lead to many applications in electro~atalysis,~ electroanalysi~,~ corrosion pr~tection,~ semiconductors,s electrochromism7 and solar- energy conversion.An understanding of these electrode systems requires not only kinetic information from electrochemical experiments but also the use of spectroscopic methods to obtain complementary structural information. For instance, X-ray photoelectron spectroscopy has been successfully applied both qualitatively, to show the presence of redox centre^,^ and quantitatively, to distinguish signals from both substrate and overlayer;lo in this last study peak intensities were correlated with electrochemically measured surface concentrations of redox centres. Secondary ion mass spectrometry has been used for qualitative measurements on electropolymerized films at electrode surfaces.11 Both these techniques require ultra-high-vacuum conditions but better correlation with kinetic data can be obtained from in situ spectroscopic studies. For instance, u.v.-vis.absorption spectroscopic measurements on a variety of highly coloured species coated on transparent electrodes have been made, but these electronic spectra normally yield little structural information.12 Vibrational spectroscopy there- fore offers the greatest promise of obtaining structural information from species at the electrode/electrolyte interface. Infrared spectroscopy using a double-pass technique has been used and in situ studies in aqueous media are yielding interesting res~1ts.l~ Many of the difficulties intrinsic in these infrared methods are obviated by using Raman spectroscopy since the wavelength of the incident light can be selected to suit 20532054 RAMAN STUDY OF THIONINE-MODIFIED ELECTRODE the solvent and the species modifying the electrode surface.Following the pioneering work of Fleischmann et al.,14 recent Raman spectroscopic studies utilizing both resonance and surface enhancement effects have provided much valuable information concerning species adsorbed at electrode s ~ r f a c e s . ~ ~ - l ~ However, to date there have been few studies of modified electrodes.l*. l9 H thionine leucothionine Scheme I In this paper we apply Raman spectroscopy to the study of the thionine-coated electrode. The thionine redox couple in aqueous acid solution is shown in scheme I. At an oxidizing potential thionine can be readily coated onto electrode substrates such as platinum or gold to make a thionine-modified electrode. Such electrodes are valuable for use in photogalvanic cells where it is essential for the illuminated electrode to discriminate between the products of the photoredox 21 The results from the Raman study of these electrodes give information about the structure of the coating, the mechanism of charge transfer through the modified layer and the Raman enhancement processes involved.EXPERIMENTAL Thionine (Fluka) was purified by column chromatography on alumina, washed with chloro- form and acetone and dried in uacuu at room temperature. [2H,]Thionine was prepared by the coupling22 of [2H,]aniline (99 + % , Aldrich) with 2,3,5,6-[2H,]phenylenediamine which itself was prepared by standard procedures23 (see scheme 11).[2H,]Thionine was purified by separation on alumina with methanol as solvent2, and was estimated (using n.m.r. spectroscopy) to be ca. 95% ring deuterated. Iron(II1) chloride, ethanoic acid, [2H2]sulphuric acid (99 o/, ), sodium chloroaurate(II1) and ethanoic anhydride (98 % ) were laboratory reagent grade from Aldrich. Sulphuric acid, hydrochloric acid, nitric acid, ethanol, methanol, sodium ethanoate and sodium sulphate all were AnalaR grade. Hydrazine monohydrate (98%) was supplied by Rose Chemicals. All aqueous solutions were prepared using distilled water. The design of the Raman spectroelectrochemical cell has been reported p r e v i o ~ s l y . ~ ~ Experiments were carried out under potentiostatic control using a Chemical Electronics type TR40-3A potentiostat and WG-Ol/ 1 waveform generator.A Hewlett Packard model 701 5B X-Y-t recorder was used for recording voltammograms and current against time traces. All potentials were measured and are reported relative to a saturated calomel electrode (SCE). The gold and platinum working electrodes consisted of ca. 3 mm diameter cylindrical discs embedded within Kel-F rods, with exposed electrode areas of 0.071 and 0.070 cm2. Prior to use they were polished to a mirror finish with a slurry of 1 ,um alumina polishing powder (Banner Scientific) and washed with distilled water. The surface of the working electrode when inserted into the Raman spectroelectrochemical cell was ca. 2 mm from the optical window. The Raman spectrum of solid thionine was obtained from a mixture of sodium sulphate and thionine compressed into a disc and rotated at ca. 2000 r.p.m.in the laser beam. Raman spectra were recorded on a Spex model 1403 spectrometer fitted with a cooled photomultiplier tube, RCA type C31034-AO2, and linked to a Spex DPC-2 digital photometer. A Compudrive and Scamp minicomputer served to control the instrument. Exciting radiation was provided by a Spectra Physics model 170 krypton laser or a dye laser, Coherent RadiationK. HUTCHINSON, R. E. HESTER, W. J. ALBERY AND A. R. HILLMAN 2055 D D$: D NH.COC H3 NH.COCH3 i' D 70% O2 SO4/ D 0 reflux 2 / y 2 ?2 N2H2 /5/0 Pd-C D ' q D EtOH/5O0C Scheme I1 model 590, containing a solution of Rhodamine 6G and pumped by a Coherent Radiation model 52A argon laser.The dye laser made available exciting wavelengths in the region 580-640 nm. The laser power at the sample was ca. 50 mW for coated electrodes and 150 mW for solids. Spectra were recorded with a spectral bandpass of 5 cm-', a dwell time of 1 s and a step interval of 1 cm-I. Where appropriate, spectra were digitally smoothed and a linear sloping back- ground subtracted. The scattered light was collected at 90" to the incident beam, which was itself polarized perpendicular to the scattering plane. The working electrode was parallel to the cell window and was set at an angle of ca. 60" with respect to the laser beam. In some experiments, Raman intensity was monitored during cyclic voltammetry of the layer between the thionine and leucothionine forms.ELECTROPLATING PROCEDURE It has been claimedz5 that electroplating increases the Raman signal on gold electrodes. We therefore prepared a number of 'electroplated' gold electrodes as opposed to the 'polished' gold electrodes described above. Prior to electroplating, the gold electrode was cleaned with 600 grit Sic paper, wetted with distilled water. The gold electrode was then electroplated from an unstirred, deoxygenated, 5 x mol dm-3 aqueous solution of sodium chloroaurate(rI1) in the absence of supporting electrolyte. The electroplating was carried out at a potential of -0.05 V for 200 s,25 with the current density being ca. 0.4 mA cm-2. COATING PROCEDURE After electroplating and washing with distilled water, the electrode was coated from a ca. 30 pmol dm-3 aqueous solution of thionine with 0.05 mol dm-3 sulphuric acid as supporting electrolyte at a potential of + 1.2 V for 900 s.21 After coating and washing with distilled water the electrode was inserted into the Raman spectroelectrochemical cell containing 0.05 mol dm-3 sulphuric acid and brought under potential control.Replacement of the aqueous H,SO, by D,SO, in D,O and scanning the potential several times2056 RAMAN STUDY OF THIONINE-MODIFIED ELECTRODE 6 - I h/nm Fig. 1. Absorption spectrum of thionine in solution (- - -) and of thionine on an n-SnO, electrode (-) together with excitation profiles of some prominent Raman bands ( 0 , 4 0 5 ; 0 , 4 7 9 ; A, 804 cm-’) from a pressed disc of solid thionine. Band intensities are measured as peak heights relative to the sulphate 992 cm-’ band as an internal standard.between the thionine and leucothionine forms (ie. from 0.6 to - 0.1 V) was used to exchange hydrogen on the amino groups for deuterium. In the coating of the platinum electrode no electroplating stage was performed and the thionine was coated at a potential of + 1.15 V for 900 s. The amount of thionine coated on an electrode was determined from the areas under cyclic voltammetric peaks for the thionine/leucothionine couple. Scan rates of 0.1 V s-’ or less were used to ensure complete oxidation/reduction of the coat. The coverage was then calculated. Assuming a ‘flat’ orientation, leading to a cross-sectional area of 0.05 nm2,26 we may convert this into a number of idealized layers, using the factor 0.3 nmol cm+ (layer)-’.RESULTS AND DISCUSSION Fig. 1 shows the absorption spectra for an aqueous thionine solution and thionine coated on a transparent n-SnO, electr~de.~’ For both, the absorption maximum is at ca. 600 nm, although for the coated form the spectrum is broader and less intense. Also shown in fig. 1 are excitation profiles of solid thionine measured for some of theK. HUTCHINSON, R. E. HESTER, W. J . ALBERY AND A. R. HILLMAN 2057 s s r I I I I 1 1 1 200 600 1000 1400 1700 wavenumber/cm -' Fig. 2. Raman spectra from solid thionine recorded between 200 and 1700 cm-l using (A) 676, (B) 647, (C) 630, (D) 614, (E) 578 and (F) 514nm exciting wavelengths. Sulphate band marked S. more intense Raman bands in fig. 2, which shows Raman spectra from solid thionine and illustrates the dependence of the band intensities on changes in the exciting wavelength; at wavelengths < 615 nm the spectrum becomes very weak.Band intensities were measured as peak heights relative to that of the 992 cm-l sulphate band and have been corrected for v4 scattering dependence and instrumental sensitivity variations. Fig. 1 clearly shows that the excitation profiles for all bands measured maximize at ca. 645 nm. The broadening of absorption bands from thin films of solid dyes prepared by vacuum sublimation has been reported previously.28 A thin film of thionine (ca. 0.1 pm thick), deposited on a quartz plate and held in vacuo, has been shown28 to give a broad absorption structure between 490 and 700 nm which consists of at least 2 bands.The strongest of these occurs at 580 nm with a second almost as intense band at 640 nm.2058 RAMAN STUDY OF THIONINE-MODIFIED ELECTRODE lei thionine I I I I I -0.2 0.0 0.2 0.4 0.6 E/V (VS SCE) Fig. 3. Cyclic voltammogram of a thionine-coated gold electrode in 0.05 mol dmP3 H,SO,. Sweep rate 0.05 V s-l. When exposed to water vapour changes occur in this spectrum, most noticeably a decrease in intensity of the 640 nm band and a shifting of the 580 nm band to 530 nm. It has been suggested that the broad absorption band observed in the spectrum from the thionine film in vacuo is due to interactions of thionine molecules with each other and with the substrate. The wavelength and intensity changes observed in the absorption spectra on exposure to water vapour are explained in terms of a weakening of these interactions, the formation of aggregates and changes in the surface crystalline However, it has also been suggested that the two bands from the thionine film in vacuo are due to a splitting of the excited state of the dye molecule into two levels by the interactions of neighbouring m01ecules.~~ Results from the thionine-coated electrode appear to be qualitatively similar to those observed from a thionine film in vacuo; the spectrum, however, is closer to that observed in solution, which is not surprising since the thionine film is hydrated.Note that the excitation profiles of all Raman bands from the pressed disc of solid thionine maximize at 645 nm, which corresponds almost exactly to the maximum of the long-wavelength absorption band observed from the thionine films in vacuo.Thus, although this absorption band has been reported to decrease in intensity on hydration of the film,29 the resonance Raman data testify to the importance of the associated transition in the present work. Since this transition corresponds to the longest- wavelength component of the absorption band, it may be attributed to the 0-0 vibronic transition of the thionine; the changes in wavelength and intensity as compared with the free solution species arise from distortions of the molecular geometry in the surface adsorbed species. Fig. 3 shows a typical cyclic voltammogram for a polished gold electrode coated with thionine with 0.05 mol dm-3 sulphuric acid as supporting electrolyte. The amount of thionine coated on an electroplated gold electrode was greater than that normally coated on a polished gold electrode, indicating a greater surface area available for coating.Similar surface coverages to those observed for the polished gold electrode could be obtained by assuming a ratio of 4 for the real: geometric area. Maximum thionine coverages on gold and platinum electrodes corresponded to ca. 12 and 15 monolayers, respectively (see Experimental section).K. HUTCHINSON, R. E. HESTER, W. J. ALBERY AND A. R. HILLMAN 2059 J C I A 1 I I 8 I 1 200 60 0 1000 1400 ,1700 Fig. 4. Raman spectra from (A) solid thionine, (B) thionine on gold in 0.05 mol dmP3 H,SO,/H,O (Th-NH,) and (C) thionine on gold in 0.05 mol dm-3 D,SO,/D,O (Th-ND,). Lexc = 647.1 nm, 5 cm-l bandpass.wavenumber/cm -' In general, good-quality Raman spectra were obtained from the thionine form coated on gold electrodes with and without the electroplating stage. Only marginally higher signal-to-noise ratios were achieved when the electroplating stage was included and certainly no dramatic enhancement was observed. No Raman spectra from the leucothionine form of the coated gold electrode could be obtained using either 647.1 or 514.5 nm exciting radiation even with surface coverages of 10-15 layers. Unlike thionine, the leucothionine form has no absorption in the visible region of the spectrum. The 647.1 nm exciting radiation is coincident with an absorption band of the coated thionine, enabling a resonance-enhancement process to play a major role in the generation of Raman signals.Despite surface coverages of > 15 layers Raman spectra could not be obtained from the thionine or leucothionine forms coated on platinum. The inability to observe Raman spectra from the thionine form on platinum, as opposed to gold, suggests that surface-enhancement processes also are involved in the generation of the Raman signals; species at platinum electrodes do not give surface-enhanced Raman scattering. Multiplication of surface and resonance en- hancement has also been observed for some coloured dyes adsorbed at a silver e l e ~ t r o d e . ~ ~ We conclude that both effects are necessary for the observation of Raman spectra from the thionine-modified electrode.Table 1. Comparison of Raman band wavenumbers (cm-l) for thionine and [2H,]thionine, solid, coated on a gold electrode, in G.05 rnol dm-'< h, 0 m ~~~ ~ 0 H,SO,/H,O and in 0.05 mol dmP3 D,SO,/D,O (%, = 647.1 nm) _ _ ~ .. thionine coated thionine coated [,He] thionine [,HH,] thionine electrode in electrode in electrode in electrode in 0.05 rnol dm-3 0.05 rnol dmP3 0.05 mol dm-3 0.05 mol dm-" thionine solid solid H,SO,/H,O D,SO,/D,O H,SO,/H,O D,SO,/D,O on a gold on a gold coated on a gold coated on a gold [ H ,] t hi o ni n e (F) - (E) ~~ F (A) (B) (B) -(A) (C) (D) (D) - (C) (El (F) 1618 m 1503 w-m 1475 vw 1387 m 1320 vw 1283 w-m 1224 w 1150m 1033 vw 961 vw 906 w-m 887 sh 804 m-s 684 vw 605 m 479 vvs 405 s 312 w - - - - - - - 1595 m-s 1498 m - - - - 1375 m 1324 vw 969 w 1232 vw 817 w-m 762 w-m 946 vw - - - 870 w-m 780 w 589 m 472 vvs 400 s 310 w __ - - - 23 -5 - - - - - 13 + 4 -314 +8 - 333 -271 - 13 - - - - - - 24 - - 16 -7 -5 -2 - 1618 m-s 1587 vw 1500 m 1480 vw 1428 m 1388 m-s 1315 w 1289 w 1224 vw 1150 m 1127 w-m 1036 m 963 vw 953 w-m 909 w-m 879 vw 806 w-m 682 w-m 609 m 480 vvs 435 sh 399 w-m 312 w - - 1621 m-s 1587 vw 1012 sh 1493 sh 1414 w 1396 m-s 1316 w 1291 w 1148 m 1127 w-m 1034 w-m 941 vw 966 w 910 vw 897 w 612 sh 682 w-m 605 m 474 vvs 435 sh 400 w - - __ - + 3 0 - 488 - - + 14 -8 -!- 1 + 2 -2 0 -2 - 22 + 13 + I + 18 - 194 0 -4 -6 0 + I - - - 1590 m-s 1488 w-m - - - 1418 m 1377 m-s 1304 w 968 vw 1248 vw 825 w-m 1171 m 763 w-m 935 w 975 w-m 862 w-m 817 sh 492 sh 589 m 472 vvs 435 sh 390 w 312 vw - - 1597 m-s 1568 vw 1010 w 1484 sh 1416 m-s 1377 m-s 1299 w 963 w 1206 w 832 w-m 1142 w 758 w 934 sh 957 w-m 861 w-m 592 sh 489 sh 586 m 468 vvs 430 sh 400 w ~- _.- - +7 F 2 - 2: v1 - 478 - U .e __ +2 0 4 0 -5 2: -42 ' ? + 7 z 0 - 5 5 O g - 5 5 -27 - 1 8 rn r rn cl - 1 el Y 0 - - -225 - 3 -3 -4 - 5 + 10 ~- Key: v, very; s, strong; m, medium; w, weak, sh, shoulder.K. HUTCHINSON, R. E. HESTER, W. J. ALBERY AND A. R. HILLMAN 206 1 I I I I r I f I 1 200 600 1000 1400 1700 Fig. 5. Raman spectra from (A) solid [2H,]thionine, (B) [2H,]thionine on gold in 0.05 mol dm-3 H,SO,/H,O (Th-d,-NH,) and (C) [,H,]thionine on gold in 0.05 mol dm-3 D,SO,/D,O (Th-d,-ND,). A,,, = 647.1 nm, 5 cm-* bandpass. wavenumber/cm -' Fig. 4(A) shows the Raman spectra from solid thionine and fig. 4(B) the thionine form of a coated gold electrode with 0.05 mol dm-3 H,SO, as supporting electrolyte.Comparison of the spectra shows several striking changes in relative band intensities although few wavenumber shifts (> 5 cm-l) are observed. Shown in fig. 4(C) is the spectrum of thionine obtained from a coated electrode in 0.05 mol dm-3 D,SO,/D,O where the amino groups will be deuterated. Comparison of this spectrum with that in H,SO, (B) shows several significant band shifts which have been listed in table 1. Fig. 5(A) shows the Raman spectra from solid pH,]thiOnine, fig. 5(B) that from [2H,]thionine coated on a gold elstrode and fig. 5(C) the same with deuteration of the amino groups. Comparisons of these spectra together with band shifts caused by deuteration of carbon and amino groupt are again given in table 1 . Raman band assignments are given in table 2.Assignments of vibrational modes for molecules of this complexity can only be approximate. In making assignments use has been made of previous Raman work on phenothiazine and its radical cation,32* 33 1,4-diaminobenzene and its radical cation,34 diphenyl ~ u l p h i d e , ~ ~ aniline3, and other2062 RAMAN STUDY OF THIONINE-MODIFIED ELECTRODE Table 2. Assignments of major Raman bands (cm-I) for solid thionine and thionine coated on a gold electrode in 0.05 mol dmP3 H2S0,/H,0 (Aexc = 647.1 nm) thionine coated on a gold electrode in 0.05 mol dm-3 thionine solid H2SO,/H,O assignmentsa 1618 1503 1475 1387 1320 1283 1224 1150 1033 96 1 906 887 804 684 605 479 405 312 - - - - 1618 1587 1500 1480 1428 1388 1315 1289 1224 1150 1127 1036 963 953 909 879 806 682 609 480 435 399 312 v(C-C)ring P(NH2) Y(CW S(CCC) i.p. skel.def. G(CNC) skel. def. G(CNC) skel. def. S(CSC) skel. def. W H , ) a Approximate description, only major contributions listed. v, Stretching (asym, asymmetric ; sym, symmetric); p, in-plane bending; i.p., in-plane; 6, bending; p, rocking; skel. def., skeletal deformation ; z, torsion; y, out-of-plane bending. simple sulphur and amino compounds. 37 The phenothiazine data are particularly relevant in that these may be seen as relating specifically to leucothionine while those of its radical cation relate to semithionine. Use was also made of Raman spectra from isotopically substituted thionine forms. Bands of high intensity were assigned to modes involving sulphur, consistent with its large polarizability, and modes involving the ring nitrogen through resonance Raman effects.All the spectra in fig. 2, 4 and 5 are dominated by the band at ca. 480 cm-l. For solid thionine [fig. 4(A) and 5(A)] the second most prominent band is that at 405 cm-l. Deuteration does not affect the position of either of these bands and they are therefore, assigned to skeletal deformation modes: S(CSC) at 405 cm-l and G(CNC) at 479 cm-l. Comparison with the equivalent bands in the spectrum of the phenothiazine radical cation shows a shift of + 181 cm-l for the S(CSC) mode and + 8 cm-l for the G(CNC) mode. The large increase in the S(CSC) wavenumber suggests a larger contribution to the thionine structure from a canonical form containing a C=S-C group.TheK. HUTCHINSON, R. E. HESTER, W. J. ALBERY AND A. R. HILLMAN 2063 Table 3. Values of PKA and u(CN) compound PKA u(CN) /cm-l 10.7a 1080’ \ TC-NH, H2N NH2 6.2c 1289 a;a < - I d 1387 H*N NH2 + >C=NH, - 1 680e a Typical value from ref. (40). Typical value from ref. (37). From ref. (40). From ref. Typical value from (41), assuming that the first proton goes on the central nitrogen atom. ref. (42). Scheme III smaller wavenumber shift of the G(CNC) mode suggests the presence of the C=N-C group since this is also present in the most prominent canonical form attributed to the phenothiazine radical The principal canonical forms for thionine may be written as in scheme 111. There have been no reported crystallographic studies of thionine, although there have been two studies on a similar thiazine dye, methylene blue.These have produced contradictory conclusions in that the positive charge has been stated to be localized on the amino nitrogen [structure (I)]38 and on the sulphur atom [structure (II)].3g Both have confirmed a planar structure, with only the substituents on the amino nitrogens out of plane. In our work on solid thionine, bands assigned to modes involving the amino groups include the .r(NH,) torsional mode at 312 cm-l, the p(NH,) rocking2064 RAMAN STUDY OF THIONINE-MODIFIED ELECTRODE mode at 804 cm-l and the v(CN) stretching mode at 1387 cm-l. Equivalent modes in 1,4-diaminobenzene were associated with bands at 356, 881 and 1289 cm-l, respectively. 34 In table 3 we collect data for the v(CN) stretching mode together with typical pK, values for typical primary aliphatic amines, diaminobenzene, thionine and typical iminium compounds.These data show that the C-N bond in thionine is intermediate between a single and double bond and suggest that structures (I) and (11) contribute about equally to the overall structure of thionine. Comparison of the Raman spectra from solid thionine and [2H,]thionine [fig. 4(A) and 5 (A)] show several band shifts, listed in table 1, which have been used in making assignments. For example, the 1283 cm-l band associated with a @(CH) vibration shifts to 969 cm-l in the deuterated form. The ratio of these band wavenumbers is 0.76, which is close to the ratio of 0.73 predicted on the basis of reduced masses [ d ( p / p ’ ) , where p and p’ are the reduced masses of CH and CD, respectively]. Similar ratios are obtained for the shifting of the 1 150 cm-l band [@(CH) mode] to 8 17 cm-l (ratio 0.71) and the 1033 cm-l band [B(CH) mode] to 762 cm-l (ratio 0.74).Several band shifts are also observed in the Raman spectra of thionine coated on a gold electrode when amino hydrogen atoms are replaced by deuterium atoms [fig. 4(B) and (C)]. These shifts are listed in table 1. Most noticeable in this regard is the 1500 cm-l band [6(NH2) mode] which shifts to 1012 cm-l on amino deuteration. The ratio 1012/1500 of 0.67 is a little lower than the ratio of 0.73 predicted on the basis of reduced masses. This difference probably reflects some accompanying mode decoupling occurring in the deuterated form. The shift of the 806 cm-l band b(NH2) mode] to 612 cm-1 gives a ratio of 0.75, in close agreement with predicted values.Next we compare the spectra of solid thionine and coated thionine in fig. 4 and 5 . From the absence of band shifts in the Raman spectra of solid and coated thionine it may at first sight appear that there is little change in the structure of thionine on coating. However, several new bands and relative band intensity changes are observed and from these we do infer important structural modifications. One clearly observed change is the marked increase in intensity of the complex structure at ca. 1400 cm-l. The emergence of a new band at 1428 cm-l is evident and is one of four new bands observed in the Raman spectra of all coated forms. Two others occur at 1127 and 953 cm-l and we infer a third band at ca.435 cm-l which causes broadening and a shoulder on the strong 480 cm-l band. These band positions vary only slightly with ring and/or amino deuteration and are therefore assigned to skeletal modes. Characterization of these new bands provides information on the bonding between thionine molecules in the film. Examination of the canonical forms of thionine suggests three probable sites for this bonding: via the central nitrogen atom, via the central sulphur atom or via the amino nitrogen atoms. Bonding via the central nitrogen atom does not account adequately for the new bands observed and the strong 479 cm-l band, assigned to a G(CNC) skeletal deformation mode, would be expected to shift considerably if the coated form were bonded via this central nitrogen atom.Bonding via the sulphur atom also is ruled out since this also cannot adequately account for the new bands observed. In particular, the band at 1428 cm-l lies at a higher wavenumber than is reasonable for a vibrational mode involving sulphur motion. Furthermore, Bauldreay and have shown that phenoxazines can be coated on to electrodes and therefore a central sulphur atom is not necessary for the bonding between thiazine molecules. There also are steric problems inherent in bonding to either of the central heteroatoms.K. HUTCHINSON, R. E. HESTER, W. J. ALBERY AND A. R. HILLMAN 2065 that bonding is via an amino nitrogen atom to a carbon atom on another thionine molecule. This type of bonding accounts for three of the new bands observed.On the basis of a secondary amine structure with an amino nitrogen atom bridging two thionine molecules, the 1428 cm-l band and the 953 cm-l band may be assigned to v,,,,(CNC) and vsym(CNC) stretching vibrations, respectively. The higher-than-expected wave- number shift of the vasym(CNC) band possibly reflects some increase in the bond order or mode mixing. The low-wavenumber band at ca. 435 cm-l is assigned to a G(CNC) skeletal deformation mode. In secondary amines bands associated with thisdeformation mode occur at 427 14 cm-l, depending upon the nature of the sub~tituents.~’ The three structures possible with this type of bonding are shown in scheme IV. We agree, therefore, with the conclusion of Bauldreay and Scheme IV All three structures ensure that the coated thionine remains electroactive and the formation of the leucothionine form does not cause any breaking of bonds holding the coat together.This bonding arrangement agrees with that deduced by Bauldreay and Archer from their cyclic-voltammetric studies of amino-substituted phenothiazines, phenoxazines and phena~ines.~~ In these studies cyclic voltammograms of an aqueous thionine solution with 1 mol dm-3 H,SO, as supporting electrolyte at an n-SnO, electrode show an oxidation wave at ca. + 1.4 V (us NHE), i.e. ca. + 1.15 V (us SCE). This is reported as being due to a one-electron oxidation of the thionine monomer forming a radical cation. A mechanism is proposed in which the polymer film is built up by the coupling of these radical cations, linked via their amino nitrogens.Such a free-radical mechanism would eliminate structure (c) given in scheme IV as at no time is there a canonical form in which an electron resides at the carbon C, position. Our ESCA studies of the thionine-coated electrode4* have shown that there are two sulphur signals. The main signal, corresponding to two-thirds of the material, is2066 RAMAN STUDY OF THIONINE-MODIFIED ELECTRODE I 1 t 200 9 0 800 1X)o 1400 MOO wavenumber/cm-' Fig. 6. Raman spectra from thionine coated on a gold electrode in 0.05 moldm-3 H,SO,/H,O : (A) thionine coverage 6 layers, (B) thionine coverage 0.8 monolayer. Aexc = 647.1 nm, 5 cm-l bandpass. similar to that from solid thionine. In addition, a smaller sulphur peak, corresponding to approximately one-third of the material, is found at a higher binding energy and therefore originates in a more oxidized form of sulphur.It is known that N-substituted phenothiazines can be oxidized to sulphoxides at ca. 0.8 V (us SCE) in 1 mol dm-3 H,S04.459 46 It is plausible to infer that the more delocalized thionine becomes partially oxidized during the coating procedure at 1 . 1 V. This hypothesis is supported by the presence of a band in the Raman spectrum at 1127 cm-l; this band is only found in the coated material and does not shift significantly with deuteration. We assign this band to a v(S=O) stretching vibration. Strong Raman spectra could be obtained from sub-monolayer coverages of thionine on the coated gold electrode. These spectra are compared with those from multilayered thionine-coated electrodes in fig.6 . There is a marked decrease in the intensity of the 1428 cm-l band previously assigned to a vasym(CN) mode bridging two thionineK. HUTCHINSON, R. E. HESTER, W. J. ALBERY AND A. R. HILLMAN 2067 50 pA cm-' I 480 cm-' band - Raman background I -Ol? OlO 0.1 0.2 0.3 0.4 0.5 0.6 E/V (us SCE) I I I I I Fig. 7. Variation of intensity of 480 cm-l Raman band and background at 300 cm-l together with the simultaneously obtained cyclic voltammogram. A,,, = 647.1 nm, 10 cm-l bandpass, C.V. sweep rate 0.05 V s-l, surface coverage 18 nmol cm-2 geometric area (ca. 15 layers). molecules; this is to be expected since the thionine in the sub-monolayer coat will not be so polymerized. Other bands showing intensity reductions are one at 405 cm-l and two at 803 and 1503 cm-l, both assigned to modes involving the amino groups and which are implicated in the polymerization.Charge transport in polymeric films can occur by electron hopping or by molecular diffusion of the redox centre^.^' In the thionine-coated electrode the redox centres are fixed in position by covalent bonds, so charge transport must occur by electron hopping. The reduction of thionine to leucothionine at pH < 3 requires 2e- and 3H+ : 2e- 3H+ Th+-L2+. Hence, for overall charge neutrality, an influx of counter ions (A-) is required. The electrons diffuse out from the electrode/film interface and the counter ions diffuse in from the film/electrolyte interface. The reverse process occurs during oxidation.Fig. 7 shows a plot of the variation of the intensity of the strong 480 cm-l Raman band (a bandpass of 10 cm-l was used to compensate for any small band shifts with potential) together with the simultaneously obtained cyclic voltammogram. To compensate for changes in background intensity, similar experiments were carried out to monitor background levels during the cyclic voltammogram. As thionine forms, Raman intensity rises and finally reaches a plateau. On the return sweep, correspond- ing to the reduction of thionine, hysteresis is observed and there is no decrease in Raman intensity until a substantial amount of thionine has been reduced. Similar2068 RAMAN STUDY OF THIONINE-MODIFIED ELECTRODE 0.0 0.1 0.2 0.3 q/mC cm-2 Fig. 8. Plots of background signal normalized to its maximum value at 0.6 V, IBG/Igtx, from fig.7 against charge (4) from the integrated cyclic voltammogram: 0, anodic scan; a, cathodic scan. results were obtained for the 1387 cm-l band. The hysteresis effect was less pro- nounced for thinner films. The change with electrode potential of the background signal is caused by fluorescence from the coated thionine. In fig. 8 the background signal is plotted against the amount of charge present in the coat obtained by integration of the cyclic voltammogram in fig. 7. A common plot is found for both the oxidation and the reduction sweeps. The hysteresis in the background signal is therefore caused by the asymmetry in the cyclic voltammogram; there is a good correlation between the charge in the coat and the intensity of the thionine fluorescence.The latter can therefore be used to measure the composition of the coat. In fig. 9 we plot the Raman intensity with the background signal subtracted against the background signal itself. Now there is marked hysteresis between the oxidation and reduction sweeps. We consider that this arises because the surface-enhancement effect results in a greater contribution to the Raman signal from thionine centres close to the electrode surface.48 This conclusion is confirmed by. the intensity of Raman signals observed for films ofK. HUTCHINSON, R. E. HESTER, W. J. ALBERY AND A. R. HILLMAN 2069 n A 0.5 1.0 o.oo 0 I*GIIEX Fig. 9. Plots of Raman signal corrected for background and normalized, ( I , - IBG)/(ZR - IBG)maX, against normalized background signal, IBG/Igtx : 0, anodic scan; A, cathodic scan.different thickness; as the film thickness increases beyond five layers there is only a small increase in intensity. Furthermore no hysteresis is observed for thin films because all the centres are too close to the electrode for there to be significant differential sensitivity . This differential sensitivity allows us to distinguish as to whether the conversion of the film redox centres proceeds from the electrode/film interface outwards or from the film/electrolyte interface inwards. The results in fig. 9 show that in both directions conversion proceeds from the film/electrolyte interface inwards; the Raman signal lags behind th.:: fluorescence signal and the innermost layers are the last to change.The reason for this surprising behaviour is that the rate-limiting process for conversion of the thionine layer is diffusion of the counter ions49 to match the change in charge on the thionine, as shown in scheme V. The anions are supplied or released through the electrolyte/film interface and their concentration gradient at the electrode/film interface is zero. When the anions adjust their concentration the reduction or oxidation of the thionine couple can take place; th13 process starts at the electrolyte interface and spreads inwards. Thus rate-limiting counter-ion diffusion is an outside to inside process. This may be contrasted with the alternative possibility where the rate of conversion of the film redox centres is limited by electron transfer. This change will start at the electrode and spread towards the electrolyte; it is an inside to outside process.The differential sensitivity of the Raman technique allows us to distinguish2070 RAMAN STUDY OF THIONINE-MODIFIED ELECTRODE reduction of coated thionine -, leucothionine Th+ +3H+. L2+ + 2e- N Th+Th+Th+ & h 7lEli: -1lzlIl - A- A- A- A- A- A-s A- ' reduction oxidation of leucothionine + thionine 7 $+ $+ L2+ N N cv N 0[10 z A- A- A- A- A- A- - 7 'u h Tu - A- A- A- A--9 A- A- + oxidation Scheme V directly between the two mechanisms. The use of the technique for this purpose does not require vibrational-mode analysis and can be applied to all modified electrodes for which characteristic Raman bands can be measured. The present results confirm our earlier conclusions that for the particular case of thionine layers, counter-ion diffusion is rate limiting in 0.05 mol dmP3 H,SO,.We are grateful for financial support from the S.E.R.C., B.P. Research, I.C.I. and the Wolfson Foundation. We thank Mr R. B. Girling for technical assistance and Dr A. J. McQuillan for helpful discussions. R. W. Murray, Acc. Chem. Res., 1980, 13, 135. W. J. Albery and A. R. Hillman, Annual Report C (The Royal Society of Chemistry, London, 1981), p. 377. R. W. Murray, Philos. Trans. R. Soc. London, Ser. A , 1981, 302, 253. N. Yamamoto, Y. Nagasawa, S. Shuto, M. Sawai, T. Sudo and H. Tsubomara, Chem. Lett., 1978, 245. M. S. Wrighton, Ace. Chem. Res., 1979, 12, 303. Fu-R. Fan and L. R. Faulkner, J. Am. Chem. SOC., 1979, 101, 4779.K.HUTCHINSON, R. E. HESTER, W. J. ALBERY AND A. R. HILLMAN 207 1 D. Ellis, M. Eckhoff and V. D. Neff, J. Phys. Chem., 1981, 85, 1225. W. J. Albery, Ace. Chem. Res., 1982, 15, 142. P. R. Moses and R. W. Murray, J. Electroanal. Chem., 1977, 77, 343. 111. lo W. J. Albery, A. R. Hillman, R. G. Egdell and H. Nutton, J. Chem. Soc., Faraday Trans. I , 1984,80, l1 G. Tourillon, J-E. Dubois and P-C. Lacaze, J. Chim. Phys. Phys.-Chim. Biol., 1979, 76, 369. l2 A. Akahoshi, S. Thoshima and K. Itaya, J. Phys. Chem., 1981, 85, 818. l 3 J-E. Dubois, P-C. Lacaze and M-C. Pham, J. Electroanaf. Chem., 1981, 117, 233; A. Bewick, l4 M. Fleischmann, P. J. Hendra and A. J. McQuillan, Chem. Phys. Lett., 1974, 26, 163. l 5 Surface Enhanced Raman Scattering, ed. R. K. Chang and R. E. Furtak (Plenum Press, New York, l6 R.P. Van Duyne, in Chemical and Biologicaf Applications of Lasers, ed. C. B. Moore (Academic Press, l7 R. P. Cooney, M. R. Mahoney and A. J. McQuillan, in Advances in Infrared and Raman Spectroscopy, lo C. A. Melendres and F. A. Cafasso, J. Electrochem. Soc., 1981, 128, 755. l9 V. S. Srinivasen and W. J. Lamb, Anal. Chem., 1977, 49, 1639. 2o W. J. Albery and M. D. Archer, Nature (London), 1977, 270, 399. *l W. J. Albery, A. W. Foulds, K. J. Hall and A. R. Hillman, J. Electrochem. Soc., 1980, 127, 654. 2 2 V. Vesely and D. Terijska, Chem. Prum., 1955, 5, 388. 23 A. I. Vogel, Elementary Practical Organic Chemistry (Longmans, New York, 1957), part 1. 24 F. Souto-Bachiller, unpublished work. 25 C. C. Busby and J. A. Creighton, J. Electroanal.Chem., in press. 26 M. D. Archer, I. Ferreira, W. J. Albery and A. R. Hillman, J. Electroanal. Chem., 1980, 111, 295. 27 W. J. Albery, W. R. Bowen, F. S. Fisher, A. W. Foulds, K. J. Hall, A. R. Hillman, R. G. Egdell and K. Kunimatsu, J. Robinson and J. Russell, J. Electroanal. Chem., 1981, 119, 175. 1982). New York, 1979), vol. 4. ed. R. J. H. Clark and R. E. Hester (Heyden, London, 1982), vol. 9. A. F. Orchard, J. Electroanal. Chem., 1980, 107, 37. L. A. Lyzina and A. T. Vartanyan, Opt. Spectrosc., 1959, 6, 110. 29 L. A. Lyzina and A. T. Vartanyan, Opt. Spectrosc., 1959, 6, 307. 30 J. Weigl, J. Chem. Phys., 1956, 24, 364. 31 T. Watanabe and B. Pettinger, Chem. Phys. Lett., 1982, 89, 501. 32 B. Kure and M. D. Morns, Talanta, 1976, 23, 398. 33 R. E. Hester and K. P. J. Williams, J. Chem. Soc., Perkin Trans. 2, 1981, 852. 34 E. E. Ernstbrunner, R. B. Girling, E. Mayer, K. P. J. Williams and R. E. Hester, J. Raman Spectrosc., 35 J. H. S. Green, J. Chem. SOC., 1956, 1350. 36 J. C. Evans, Spectrochim. Acta, 1960, 16, 428. 37 F. R. Dollish, W. G. Fately and F. F. Bentley, Characteristic Raman Frequencies of Organic Com- pounds (Wiley-Interscience, New York, 1974). ‘3* H. E. Marr and J. M. Stewart, Chem. Commun., 1971, 131. G. S. Zhadanov, Z. V. Zvankova and L. G. Vorontsova, Sou. Phys.-Crystaflogr. (Engl. Transl.), 1956, 1, 44. ‘lo D. D. Perrin, Dissociation Constants of Organic Bases in Aqueous Solution (Butterworths, London, 1965). 41 M. I. C. Ferreira, Ph.D. Thesis (University of London, 1978). 42 D. Dolphin and A. E. Wick, Tabulation of Infrared Spectral Data (Wiley, New York, 1977). 4 3 J. M. Bauldreay and M. D. Archer, Electrochim. Acta, 1983, 28, 1515. 44 W. J. Albery, A. R. Hillman, R. G. Egdell and H. Nutton, unpublished work. 4 5 A. N. Merkle and C. A. Discher, Anal. Chem., 1964, 36, 1639. 36 P. Kabasakalian and J. McGlotten, Anal. Chem., 1959, 31, 431. 4 7 D. A. Buttry and F. C. Anson, J. Electroanal. Chem., 1981, 130, 333. 4 R A. Otto, I. Pockrand, J. Billmann and C. Pettenkofer, in Surface EnhancedScattering, ed. R. K. Chang j9 W. J. Albery, M. G. Boutelle, P. J. Colby and A. R. Hillman, J. Electroanal. Chem., 1982, 133, 135. 1981, 10, 161. and T. E. Furtak (Plenum Press, New York, 1982), p. 147. (PAPER 3/ 1554)

ISSN:0300-9599    

DOI:10.1039/F19848002053

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J. Chem. Soc., Faraday Trans. 1, 1984,80, 2073-2086 The Pep tide-Urea Interaction Excess Enthalpies of Ternary Aqueous Solution of Sarcosylsarcosine Diketopiperazine and Alkylureas at 298.15 K BY GUIDO BARONE,* PASQUALE CACACE AND VITTORIO ELIA Department of Chemistry, University of Naples, Via Mezzocannone 4, 801 34 Napoli, Italy AND ATTILIO CESARO Institute of Chemistry, University of Trieste, Italy Received 4th October, 1983 The excess enthalpies of ternary aqueous solutions of sarcosylsarcosine diketopiperazine and alkylureas have been determined and compared with the properties of other diketopiperazine- urea systems at 298.15 K. A group-contribution analysis shows that cyclic dipeptides are characterized by more favourable peptide-peptide and peptide-urea interactions than the linear oligopeptides.The other contributions, and in particular the cross-interactions between peptides and alkyl chains, seem to assume values equal to those of non-cyclic solutes. The particular properties of sarcosylsarcosine diketopiperazine aqueous solutions are discussed and an unexpected analogy with monosaccharide aqueous solutions is found. The peptide-urea interaction is of great interest in biophysical chemistry, since it concerns the conformational stability of proteins and naturally occurring oligo- and poly-peptides. It is generally accepted that the peptide-urea interaction provides an important driving force in the denaturation processes in aqueous solutions containing urea or urea-like compounds and derivatives. Indeed, favourable interactions between urea and model compounds of the peptide repeating units have been suggested on the basis of spectroscopic1.and thermodynamic However, the nature and strength of the interactions between two simple, non-ionic molecules in water are still the subject of controversy. For instance, the concept of hydrophobic interaction is periodically re~ised,~ and the importance of mixed interactions between the polar and apolar parts of the solute l1 and of the role of water12 has recently been stressed. Controversial opinions are currently reported as to whether the inter- action between molecules such as peptides and ureas is due to direct hydrogen-bond formation or to solvent-mediated effects. Evidence in favour of the second interpretation has been provided by spectroscopic studies13-16 and by the strict similarity in the thermodynamic excess properties of urea and thiourea,l79 l8 since it is unlikely that the latter can, in water, establish favourable hydrogen bonds with other molecules of the same species.Urea solutions seem rather to be characterized by an extended perturbation of the structure of the solvent, which appears much more disordered than in pure water.14-16 An interesting class of model solutes is that of the diketopiperazines, whose main framework is characterized by a dipeptide cyclic structure. Within the ring, two -CO-NR- groups (R being a hydrogen atom or an alkyl chain) are present in the unusual cis conformation. Some solution properties, including free energies and ent halpies of dilution and of mixing diketopiperazines with urea, have previously been 68 2073 F A R 12074 THE PEPTIDE-UREA INTERACTION reported.lg9 2o The present paper reports the results of a calorimetric investigation of ternary aqueous solutions of the monosubstituted alkylureas and sarcosylsarcosine diketopiperazine (Sar,dkp), formula LCON (CH,) CH, CON (CH,) CH,’ at 298.15 K.One aim of the work was to explain the differences in the behaviour of the cis- and trans-peptide conformers. This attempt was prompted by the recent publication of new data on linear ~ligopeptides.~l-~~ A second aim was to study the influence of the nature and conformation of neighbouring polar group (peptide, amide, urea residue) on the hydrophobic interactions. For this purpose, alkylureas and Sar,dkp were chosen: the latter was preferred to the isomeric alanylalanine diketo- piperazine because of its greater solubility.THERMODYNAMICS Excess thermodynamic properties are an important source of data for the formulation and testing of models for intermolecular interactions. For operational purposes, the excess enthalpy is defined as6? lo* 18* 20-25 n H E = H - H ~ - C m,Hi 2-1 where HE and H refer to an amount of solution containing 1 kg of water and m, . . . m, moles of each solute species, H g is the standard enthalpy of 1 kg of water and ET: is the standard partial molal enthalpy of each solute species. Therefore HE represents the deviation from the behaviour of an athermal or ideal solution and is determined by the nature and intensity of the molecular interactions.However, HE is only one of the contributions to the excess Gibbs free energy of the solution, GE, which also includes an entropic term. The excess enthalpies of binary or ternary solutions can be calculated from experimental data for heats of dilution, AHdi1. Those of a ternary solution can also be obtained by combining the data of the heats of mixing, A P i X , with the heats of dilution of two binary solutions. By adopting the Lewis-Randall- Friedman notation24 it is possible to express H E as a power series in the molalities. For a ternary solution one can write H E = h,, m i + 2hx, m, my + h,, m; + h,,, m i + ~ ~ , , , ~ ~ ~ , + 3 h , , , m x m ~ + h , , , m ~ + . . . . (2) For each of the h coefficients, relationships with the excess free-energy and entropy coefficients can be written as for the overall properties and so on.According to the McMillan-Mayer theory, adapted to non-ionic solution~,~@-~~ the g coefficients are a measure of the interactions among pairs, triplets and higher multiplets of solute molecules of the same or different species. EXPERIMENTAL MATERIALS Sarcosylsarcosine diketopiperazine (Sar,dkp) (Fluka AG product) was crystallized from ethylacetate; monomethylurea (MMU), monoethylurea (MEU), monopropylurea (MPU) andG . BARONE, P. CACACE, v. ELIA AND A. CESARO 2075 Table 1. Heats of mixing for aqueous solutions of sarcosylsarcosine diketopiperazine (Sar,dkp) and monomethylurea (MMU) at 298.15 K 0.7493 0.7493 0.7493 0.4989 0.3740 0.4989 0.2489 0.3740 0.3740 0.2489 0.2489 0.4989 0.2489 0.2489 0.2489 0.3075 0.4349 0.1606 0.2252 0.1589 0.1853 0.1463 0.1519 0.1508 0.1221 0.1401 0.0983 0.0541 0.3047 0.2182 0.7505 0.7505 0.7505 0.49 12 0.7505 0.49 I2 0.49 12 0.2488 0.1970 0.2488 0.2488 0.49 12 0.2488 0.2488 0.2488 0.4425 0.3149 0.5897 0.2695 0.43 17 0.3087 0.2025 0.1477 0.1 176 0.1267 0.1087 0.3944 0.1947 0.2183 0.3073 160.6 159.7 11 1.4 76.5 55.7 72.5 24.8 37.1 34.8 20.5 19.7 24.4 13.2 8.6 8.3 90.3 89.3 62.2 44.9 44.2 42.8 20.2 18.8 16.0 12.8 12.1 9.2 7.9 5.2 4.9 a 0 = 2.5.monobutylurea (MBU) (Fluka or K & K products) were crystallized twice from ethanol + water mixtures. All the products were dried in vacuo at room temperature. The solutions were freshly prepared with deionized, twice-distilled and degassed water. CALORIMETRY Heats of mixing, A T i x , of pairs of binary solutions and heats of dilution, AHdi1, of ternary solutions weredetermined, at 298.15 K, with two standard LKB 10700-1 flow microcalorimeters.Two kinds of experiments were set up : for Sar,dkpMMU, Sar,dkpMEU and Sar,dkpMBU, binary aqueous solutions were mixed over a range of solute concentration; in the case of Sar,dkpMPU, ternary solutions of the two components were diluted with water. The two methods turned out to be equivalent,18 because of the weakness of the intermolecular interactions in these solutions. (In the case of a strong association the dilution of a ternary solution would lead only to the release or absorption of the amount of heat due to the shift in the equilibrium.) Repeated checks on the reproducibility of the data obtained with the two calorimeters were carried out by determining the heats of dilution of MMU28 and MEU.,@ Other experimental details are reported in preceding papers.l89 20, 28-33 The cross-coefficients of eqn (2) can be obtained by a least-squares treatment using an analytical form of the auxiliary functions defined asla' 20, 2 5 7 309 31 and The fitting was tried with polynomials of increasing degree, choosing that of highest degree for which all the coefficients are significant with respect to their own 95% confidence limit.RESULTS The experimental data are reported in tables 1-4. Tables 1, 2 and 4 give the heats of mixing and AH* values at 298.15 K for Sar,dkp solutions with aqueous solutions of MMU, MEU and MBU, respectively, as functions of the initial and final molalities.Table 3 gives heats of dilution and AH** values at 298.15 K for ternary aqueous 68-22076 THE PEPTIDE-UREA INTERACTION Table 2. Heats of mixing for aqueous solutions of sarcosylsarcosine diketopiperazine (Sar,dkp) and monoethylurea (MEU) at 298.15 K - AHrniX -AH* a mkar2dkp mSarzdkp mMEu ~ M E U /J kg-l /J kg-l 0.7493 0.7493 0.7493 0.4989 0.7493 0.4989 0.3740 0.3740 0.3740 0.2489 0.2489 0.2489 0.2489 0.2489 0.2489 0.5545 0.4340 0.3069 0.2257 0.1603 0.1861 0.1556 0.1508 0.1503 0.1464 0.1224 0.1399 0.054 1 0.0304 0.2176 0.7607 0.7607 0.7607 0.4958 0.7607 0.4958 0.7607 0.2058 0.2506 0.4958 0.2506 0.2506 0.2506 0.2506 0.2506 0.1978 0.3201 0.4492 0.271 5 0.5980 0.3108 0.4441 0.1228 0.1499 0.2042 0.1274 0.1098 0.1962 0.2199 0.03 15 105.4 126.2 125.2 59.6 87.4 56.7 57.1 28.2 28.9 24.7 16.5 15.6 10.3 6.8 6.7 21.3 19.0 18.1 12.6 12.4 12.2 8.7 7.0 6.9 5.0 4.8 4.1 2.4 1.8 1.6 Q = 2.7.Table 3. Heats of dilution of ternary aqueous solutions of sarcosylsarcosine diketopiperazine (Sar,dkp) and mono-n-propylurea (MPU) at 298.15 K m$ar2dkp mSarsdkp mMPv mMPU 1.0058 0.9024 0.8599 0.7627 0.7530 0.5613 0.4168 0.351 1 0.3659 0.3 150 0.2727 0.2355 0.2561 0.1902 0.4419 0.3977 0.3798 0.3393 0.3363 0.2719 0.2040 0.1731 0.1763 0.1523 0.1350 0.1 170 0.1270 0.0935 0.31 16 0.2796 0.2664 0.2363 0.2333 0.1739 0.1292 0.1089 0.1 134 0.0976 0.0845 0.0730 0.0793 0.0589 0.1369 0.1232 0.1 177 0.1051 0.1042 0.0843 0.0633 0.0537 0.0546 0.0472 0.041 8 0.0362 0.0393 0.0288 210.3 172.1 157.1 121.5 117.0 66.9 37.3 26.8 27.1 20.1 16.1 12.3 14.2 7.5 58.6 49.9 46.1 34.2 31.8 19.2 11.0 8.2 6.9 5.1 4.8 3.9 4.3 2.2 u ~ = 1.3.solutions of Sar,dkp and MPU. The values of the coefficients h,., h,, etc., relative to the binary aqueous solutions, are given in table 5. Only the coefficients h,, of eqn (2) were found to be significant in the range of concentrations studied. The values of these cross-coefficients, together with their 95% confidence limits, are in table 5. The values of h,, for Sar,dkpMBU and Sar,dkpMPU pairs are positive, as are the respective values of h,. and h,,. In the case of Sar,dkpMEU h,, < 0, while bothG. BARONE, P. CACACE, v. ELIA AND A. CESARO 2077 Table 4. Heats of mixing for aqueous solutions of sarcosylsarcosine diketopiperazine (Sar,dkp) and mono-n-butylurea (MBU) at 298.15 K 0.7493 0.7493 0.7493 0.7493 0.4989 0.4989 0.4989 0.2489 0.2489 0.2489 0.2489 0.2489 0.3 103 0.4379 0.5540 0.1622 0.2268 0.1871 0.1004 0.1229 0.141 1 0.0542 0.0306 0.2178 0.7422 0.7422 0.7422 0.7422 0.5003 0.5003 0.5003 0.2559 0.2559 0.2559 0.2559 0.2559 0.4348 0.3084 0.1935 0.581 5 0.2729 0.3 127 0.3996 0.1296 0.1109 0.2001 0.2244 0.0320 148.1 145.2 118.3 107.9 60.9 58.4 41.6 15.0 14.6 10.4 6.5 6.2 135.5 131.5 96.2 96.1 59.3 56.3 38.5 13.6 13.4 9.4 6.1 6.0 a 0 = 3.2.Table 5. Enthalpic pair interaction coefficients for binary and ternary aqueous solutions of diketopiperazines and ureas at 298.15 K solute Sar,dkp Sar,dkp Sar,dkp Sar,dkp Sar2dkp GVDkp GAdkp A2dkP G2dkP MBU MPU MEU MMU U U U U - 486( 1 2)c 388( 10) -77(10) - 339( 1 O)i - 758(33)i - 674( 1 5)i - 783(35)i - 9 1 7(25)i - 577d 577d 577d 577d 577d 121d -213d - 509d - 1 138d (- 1075z) 420e 92f 3 79 21h 21,i 22" 21: 22" 21: 22" 21: 22" 21: 22" a Units: J mol-l (mol kg-l)-l.Units: J mol-l (mol kg-l)-,. are the 95% confidence limits. 76, 3065. The numbers in parentheses Ref. (28). Datum from S. J. Gill and L. Noll, J. Phys. Chem., 1972, Ref. (19). Ref. (33). f Ref. (32). g Ref. (29). Ref. (20). j Ref. (51). Ref. (10). h,, and h,, are positive. This is very unusual behaviour for aqueous solutions of simple non-ele~trolytes.~~ Finally, for the pair Sar,dkpMMU h,, < hMMU-MMU < 0. The results of the present study are compared in table 5 with those obtained from similar experiments with solutions containing urea and other diketopiperazines :199 2o glycylglycine (G,dkp), L-alanyl-L-alanine (A,dkp), glycyl-L-alanine (GAdkp) and glycyl-L-valine (GVdkp). The values of h,, change gradually from the positive value for Sar,dkpMBU and Sar,dkpMPU to the negative ones, characterizing the solutions of urea and cyclic dipeptides.Moreover, only for the pairs Sar,dkpMPU2078 THE PEPTIDE-UREA INTERACTION and G,dkpU is h,, < h,, < h,,, while for all other pairs h,, < h,, < hYy. This behaviour is also unusual and, among the known pairs of simple non-electrolytes in water, it is exhibited exclusively by some sugars with urea.10*35 DISCUSSION The macroscopic thermodynamic properties are related, albeit in a complex way, to the molecular parameters of the components, which may be used for the statistical-mechanical evaluation of the physico-chemical properties of the solution.To compare the experimental data for the excess enthalpy and Gibbs free energy with the results of theoretical calculations, based on molecular models, other properties such as isothermal compressibility, partial molar volumes and their derivatives should be available. When such data are known, it will be possible, in principle, to transpose the constant pressure properties into the relations describing the system at constant number of 36-38 Both the lack of experimental data and the difficulties in solving the problem of mutual orientational effects among the molecules of polar solutes and water have discouraged attempts to treat ternary solutions. In the preceding paperslap 2o strong specific associations were shown to be the origin of values of the second virial coefficients, g,, and h,,, up to two orders of magnitude greater than the corresponding ones for aqueous systems containing urea or diketopiperazines.Moreover, in these cases third virial coefficients are present and are greater in absolute terms than the second virial coefficient^.^^ For the systems studied in the present paper, the measured effects must therefore be ascribed to weak non-bonding interactions. In the previous papersl9* 2o favourable enthalpic contri- butions were shown to characterize the interaction between urea and cyclic dipeptides, or between two diketopiperazine molecules, even in the presence of methyl substituents on the ring. However, the data show that, upon increasing the number and size of the alkyl chains on both the two interacting molecules, the hydrophobic effects gradually prevail, promoting large changes in the values of the thermodynamic parameters.This behaviour is quite similar to that observed in aqueous solutions of linear peptides, amides and ureas.l09 20-23? 28, 29 It is worthwhile to show that changes in the thermodynamic properties to positive values are due not only to the hydrophobic interactions but also to alkyl-peptide, water-mediated, interactions. The analysis can be carried on by the group-additivity approach, developed by Wood and coworkers,loT l1, 4 0 v 41 which allows a formal factorization of the contributions arising from the interactions between each pair of functional groups to the overall pairwise interactions.Before applying the group-additivity approach, the properties of Sar,dkp aqueous solutions require some comments, since the second virial coefficients of the excess free energy, enthalpy and entropy are all p ~ s i t i v e : ~ ~ ~ ~ ~ g,, = 27, h,, = 577 and Ts,, = 550 J mol-l (mol kg-l)-l, respectively. This distinguishes, at 298.15 K, Sar,dkp not only from the urea-like set of non-electrolytic solutes18 but also from more similar solutes. This is shown by comparison with the properties of N,N'-dimethylformamide and N,N-dimethylacetamide, other examples of molecules not able to establish hydrogen bonds with one another. In aqueous solutions of these two compounds the *l g,, = - 106 and - 1200 J mol-1 (mol kg-l)-I, respectively, are found.Moreover, Ts,, > h,, > 0 for both the systems. These values indicate that the two alkylamides are predominantly hydrophobic molecules interacting with other molecules of the same species because of the existence of a favourable non-bonding hydrophobic effect. To some extent the behaviour of Sar,dkp binary solutions resembles that of theG. BARONE, P. CACACE, V. ELIA AND A. C E S h o 2079 ~ u g a r s ; ~ ~ - ~ ~ even more, because of structural similarities, it resembles that of he~amethylentetramine.~~-~~ The similarity between the values of the parameters? of the last two compounds can probably be attributed to the crowding of alternating donor polar groups and methyl or methylene groups. All the properties of hexa- methylenetetramine solutions suggest that this molecule is extensively hydrated, the water of the hydrophilic and hydrophobic cospheres being in some way compatible.Although related to solid crystalline packings (energetically different), the high solubility of this solute and of Sar,dkp is also remarkable, especially when compared with that of other diketopiperazines, which are much less soluble. Favourable solute- solvent interactions probably prevent solute-solute interactions, as hypothesized for sugar^.^^-^^ Because of the large differences in molecular structure, the similarity with this family of solutes must be regarded as a functional analogy, rather than as a structural one. THE GROUP-CONTRIBUTION APPROACH The additivity principle is of general applicability in chemical thermodynamics for obtaining the contributions of groups of atoms to molecular properties and has been recently used also for excess solution properties.Within the framework of the McMillan-Mayer theory, Woodlo has obtained an expression for the second coefficient of the excess enthalpy: h,, = nf ny ( H i j ) . ( 5 ) i j The enthalpic coefficient is therefore given by the sum of all the contributions (Hi,) obtained by coupling each group i of the solute molecule x with each group j of the solute molecule y (of the same or different species). Although of practical interest, eqn ( 5 ) contains some approximations, whose origins are in part implicit in the cluster-expansion treatment.ll9 2 3 9 51 For example, the additivity of group interactions is invalidated by any cooperative or concerted effect of several groups leading to the specific mutual recognition of two Eqn ( 5 ) seems to work well when the coefficients are averaged over a large number of structurally similar solutes, characterized by weak and scarcely specific interactions.Therefore the approach can provide a useful basis to ascertain whether unexpected effects are present in a given system. Woodlo has made some assumptions to reduce to a minimum the number of group interactions to be considered [since it increases according to the function n(n+ 1)/2, n being the number of chosen groups]. Accordingly, we selected -CONH-, U and -CH,-groups, assuming CH = OSCH, and CH, = 1 .5CH2 and that all thecontribu- tions involving CONH and U have the same value, independent of their degree of N-substitution. In a preceding paper20 it has been found that the (Hu, CONH) and (HCONH, CONHI contributions for aqueous solutions of diketopiperazines differ from those obtained for amide so1utions.10~40 This was ascribed to the different stereochemistry of the -CONH- group.At least in the solid state, stereochemical studies on the -CONH- group show that stable structures are to be expected when the angle of internal rotation about the C(0)-N bond is close to that of the trans conformation. In the cyclic dipeptides, on the other hand, the two -CONH- groups are forced t Quadrifoglio er aL4' and later Garrod and Herringt~n~~ found h,, = 1142 and 1272 J mol-I (mol kg-')-l in the ranges 0-3 and M . 3 mol kg-l, respectively. More recently Tasker and WoodS0 found in the range 0-1 mol kg-' h,, = 872.4f24.3 J mol-I (mol kg-l)-l, h,,, = 262148.3 J mol-l (mol kg-')-* and h,,,, = - 90.3 f 29.0 J mol-' (mol kg-1)-3.The difference in the second virial coefficient can be partially due to the large third coefficient found by Tasker and Wood, but also to the different purification procedures.2080 THE PEPTIDE-UREA INTERACTION Table 6. Values of the coefficients of the excess enthalpies used to derive the (Hi'} for amides, peptides and ureas in water at 298.15 K NMF NMA NMP NBA NMF NMF NMF NMA NMA NMP FA AA PA DMF DMA FA NAGA NAAA NAVA NALA NAGA NAGA NAGA NAAA NAAA NAVA NAG,A NAG,A NAGA NAGA NAA,A NAAA NAAGA NAA,A U U U U U MMU MEU MPU 1,3DMU 1,lDMU 1,3DEU 1,lDEU U NMF NMA NMP NBA NMA NMP NBA NMP NBA NBA FA AA PA DMF DMA DMF NAGA NAAA NAVA NALA NAAA NAVA NALA NAVA NALA NALA NAG,A NAG,A NAG,A NAG,A NAA,A NAA,A NAAGA NAA,A U NMF NMA NMP NBA MMU MEU MPU 1,3DMU 1,lDMU 1,3DEU 1,lDEU MMU MMU 1,3DMU 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 4 2 2 3 2 3 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 4 3 4 3 3 3 4 0 1 1 1 1 0 0 0 0 0 0 0 0 0 2 3 4 6 2 2 2 3 3 4 0.5 1.5 2.5 3.5 4.5 0.5 2.5 3.5 5.5 6.5 2.5 2.5 2.5 3.5 3.5 5.5 3.5 4.5 2.5 2.5 5.5 3.5 4.5 7.5 0 0 0 0 0 1.5 2.5 3.5 3 3 5 5 0 1.5 2 3 4 6 3 4 6 4 6 6 0.5 1.5 2.5 3.5 4.5 3.5 2.5 3.5 5.5 6.5 3.5 5.5 6.5 5.5 6.5 6.5 3.5 4.5 3.5 4.5 5.5 5.5 4.5 7.5 0 2 3 4 6 1.5 2.5 3.5 3 3 5 5 1.5 3 272" 236" 636c 1477" 368" 54OC 883" 393" 62gC 1 079" -115d 1 2e 249e 737d 962"*f 1 55d - 2209 2689 12598 17148 689 3858 5478 5919 8998 14869 - 646h - 1499h -21 l h - 544h 939h 641h 284h 4880hvi - 350" - 109" 0" 180" 264" - 85' 1 60k 292' 3 9 38' 101 1" 791' - 151m 87" 126 393 693 1394 25 1 377 627 535 819 1010 -21 1 5 256 539 856 89 259 1294 1861 17 468 693 743 985 1569 - 208 - 635 - 1277 - 368 - 528 766 449 49 1525 - 47 28 102 25 1 - 89 127 376 247 247 813 813 60 - - 238G.BARONE, P. CACACE, v. ELIA AND A. CES~RO 208 1 Table 6. (cont.) MMU MEU 0 0 1.5 2.5 76" 11 MMU 1,3DEU 0 0 1.5 5 410" 260 1,3DMU MEU 0 0 3 2.5 131" 185 1,3DMU 1,3DEU 0 0 3 5 388" 497 MEU 1,3DEU 0 0 2.5 5 508" 418 MBU MBU 0 0 4.5 4.5 1 039i7* 659 Abbreviations used : NMF : N-methylformamide; NMA: N-methylacetamide; NMP: N- methylpropionamide; NBA: N-butylacetamide; FA: formamide; AA: acetamide; PA; pro- pionamide; DMF : N,N'-dimethylformamide; DMA : N,N-dimethylacetamide; NAGA : N-acetylglycinamide; NAAA : N-acetyl-L-alaninamide ; NAVA : N-acetyl-L-valinamide ; NALA : N-acetyl-L-leucinamide ; NAG,A : N-acetylglycylglycinamide; NAG,A : N-acetyl- glycylgl ycylglycinamide ; NA A,A : N-acet yl-L-alanyl-L-alaninamide ; NAAGA : N-acetyl-L- alanylglycinamide ; NAA,A : N-acetyl-L-alanyl-L-alanyl-L-alaninamide ; U : urea ; MMU : monomethylurea ; MEU : monoethylurea ; MPU : mono-n-propylurea ; MBU : mono-n-butyl- urea; 1,3DMU : 1,3(N,N)-sym-dirnethylurea; 1,l DMU : 1, l(N,N)-asym-dimethylurea; 1,3DEU : 1,3(N,N')-sym-diethylurea; 1,l DEU : 1,l (N,N)-asym-diethylurea. a Number of [--CONH-],,,,,, [--CON--],,,,, or --CONH, groups and number of -CH2- groups, or fractions, respectively, on the solute molecule x or y .Note that the number of urea residues is n t = 1 for urea and urea derivatives and n2, = 0 for all other solutes considered. ' Units: J mo1-I (mol kg-l)-I: Ref. (10). Ref. (53). Ref. (40). f Evaluated from G. C. Kresheck, J. Phys. Chem., 1969, 73, 2441. Ref. (23) Not included in the fit. j Ref. (28). Ref. (29). Ref. (32). Ref. (18). Ref. (31). O Ref. (33). Ref. (21). into the domain of the cis conformations, to minimize the intramolecular constraints. The most stable angle of internal rotation about the C(0)-N bond depends on the substituents, which determine the overall conformation of each diketopiperazine ring (quasi-planar, chair or boat).52 The new data given in the present paper justify an attempt at improving the group-contribution analysis.The following procedure was adopted. First, the set of six (H,,J contributions arising from the interactions of the three selected groups CONH, CH, and U was evaluated from a set of 52 experimental figures, concerning the excess enthalpies of solutions containing linear peptides, amides and ureas. The experimental data, which are listed in table 6 together with number of groups present on each of the two solute molecules, were used for fitting eqn (8) to obtain the mean {Hii} contributions, given in the third column of table 7. The results are in very good agreement with more recent ones reported by Lilley and on the basis of data obtained exclusively for the N-acetyl-di- or tri-peptidoamides (fourth column of table 7). With respect to earlier fittingslO3 2 1 g * O , 53 there is a clear improvement in the estimated confidence limits, as the basic set of data is enlarged to all the peptide, amide and urea derivatives.This is proof that the overall group of data reported in table 6 is homogeneous. The recent data of Lilley and represent, however, the decisive step, because the number of CONH-CONH pairwise interactions available for the fitting is at least comparable to the number of other pairs. Instead, owing to scarcity of experimental data concerning urea-peptide systems, a relatively high value for the 95% confidence limit for {H,,CONH} is still obtained. Finally, because the refined values of {Hu, u> converge to the experimental value (or to a value very near2082 THE PEPTIDE-UREA INTERACTION to it), we finally assumed {flu, u) = hu, = - 350 J mol-1 (mol kg-I)-l.lo- 2 5 9 51 If a fitting is carried out without assuming (flu, u> as a fixed parameter, a value differing by < 4% from the experimental hu, is obtained.However, this value would be affected by a 95% confidence limit of k 78 J mol-1 (mol kg-l)-l, which is very large and attri- butable in part to the relatively small number of U-U interactions involved in the systems studied and, in part, to small specific effects between the alkylureas. In this case the other ( H i j ) contributions involving urea change by only 3 % , while the others do not change at all. The standard deviation of the fitting would decrease by < 1 % . These considerations are sufficient in our opinion for assuming {Hu, u> = hu, and for confirming its invariance with the degree of substitution.It is possible to see, from the last two columns of table 6, that the h,, values calculated by the preceding procedure differ remarkably from the experimental ones in a few cases only. There are no reasons, at present, for excluding them from the fitting, nor is other information available to explain these discrepancies. However, for binary solutions of the tripeptide N-acetyl-L-alanyl-L-alanyl-L-alanineamide and of mono-n-butylurea, not included in the fitting, the discrepancy is very large. In both cases it can be argued that cooperative hydrophobic interactions lead to a non-monotonic increase in the h,, values with respect to the lower homologous compounds, and hence to the failure of the additivity.One of the more interesting conclusions of the analysis carried out by Lilley and on the sign and values of the ( H i i ) contributions, as well as of free-energy and entropy data, is that the peptide-peptide interaction has been underestimated in the past and that an unfavourable peptide-alkyl or urea-alkyl interaction can considerably attenuate the effects of intramolecular hydrophobic forces or of peptide-peptide interactions in proteins. Using the {Hij) values obtained by the above-discussed procedure (table 7, third column) an attempt was made to calculate the h coefficients for the diketopiperazine systems. All the coefficients resulting from these calculations are greatly overestimated, far beyond the standard deviation of the preceding fitting (see the last column of table 8).This means that the set of data for the cyclic dipeptides is not homogeneous with that chosen as a reference. Therefore, the h,, data reported in table 8 have been fitted separately from those of table 6, with the implicit assumption that the [-CONH-I,,, conformer has different behaviour from that of the [-CONH-],,,,, conformer. The value of {Hu, u> was assumed to be equal to the experimental hU, value, as before, and the (HcH,, CH2} and {Hu, CHz} contributions were assumed from the third column of table 7. The last assumption is justified by the fact that equivalent results are obtained if these contributions were considered adjustable parameters. The result of the fitting is given in the last column of table 7 and shows that the (H,,,,, C H p ) contribution has the same value as that found for linear peptides, whilst this does not occur for {HCONH, CONH) and {HCONH, u} contributions.Therefore the main results of this work are that the enthalpic parts of the pairwise interactions are more favourable for peptide-urea and peptide-peptide interactions, when dealing with the diketopiperazine systems. The extra contribution, calculated from the differences between the values of the {Hii) involving trans and cis conformers, as reported in table 7, is ca. -640 J mol-l (mol kg-l)-l for the diketopiperazine- diketopiperazine interaction and - 660 J mol-l (mol kg-I)-l for the diketopiperazine- urea interaction. It is not possible at present to ascertain whether the behaviour of the diketopiperazine systems is due to the intrinsic nature of the [-CONH-],i, conformer (and of its hydration cosphere) or whether it originates partially or totally from a concerted action of the entire diketopiperazine molecule. The shape, size and geometry of the six-membered ring of the cyclic dipeptides could, in fact, make the interaction with another diketopiperazine, urea or a urea derivative easier.The second hypothesis, although stimulating, needs other data in order to be supported.G . BARONE, P. CACACE, v. ELIA AND A. CESARO 2083 Table 7. Group contributions {Ifij} to the second virial coefficients of the excess enthalpies for amides, ureas, linear peptides and diketopiperazines functional amides and linear peptides groups cyclic ref.(lo), dipeptides, 1 .i this work ref. (23) ref. (53) (21), (40) this work CONH CONH -307(22)" -31 l(57) -252(105) -251(103) -470(38)' - - - 525(27)" CONH U - 196(75)" - U U - 350b - - 280 - 350b CONH CH, 92( 1 1)" 95(29) 66(37) 42( 3 3) 9 1 (9)c - 29 74b 74( 1 0)" - U CH2 CH2 CH2 17(5)" 14(13) 26( 13) 42@) 1 7b - a Values obtained by the fitting of the data of table 6, CONH being the -CONH- or Values assumed as fixed Value obtained by the fitting of the data of table 8, CONH being, in this case, -CON- groups in trans conformation or the -CONH, group. parameters. the -CONH- or -CON- groups in the cis conformation. Units: J mol-1 (mol kg-l)-l. Table 8. Values of the coefficients of the excess enthalpies used to derive the {Hij} for diketopiperazines in water at 298.15 K G2dkP A2dkP G2dkP A2dkP GAdkp GVdkp Sar,dkp GAdkp Sar,dkp Sar,dkp Sar,dkp Sar,dkp Sar,dkp G2dkP A2dkP GAdkp GVdkp Sar,d kp U U U U MMU MEU MPU MBU 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 0 0 0 0 0 0 0 2 3 4 5 5 2 3 4 5 5 5 5 5 2 3 4 5 5 0 0 0 0 1.5 2.5 3.5 4.5 - 1 138" - 509" -213" 121" 577" -917f - 783f - 674 - 758f - 339 - 77 388 486 - 1085 - 637 - 156 358 358 - 900 - 826 -751 - 677 - 278 - 13 253 518 - 429 22 505 1022 1022 - 242 - 168 - 93 - 19 382 648 915 1182 a Number of groups on the solute molecule x or y as in table 6.Units: J mol-l (mol kg-l)-l. Calculated by using the {Hij} From ref. (19). f From ref. (20). ' Calculated by using the {Hij) of the last column of table 7. of the third column of table 7 (amides and linear peptides). CONCLUSIONS The main conclusion of this work is that urea-peptide and peptide-peptide interactions are more (enthalpically) favoured for cyclic dipeptides than for linear peptides and amides.From the analysis of pairwise contributions this is to be ascribed to (IfCONH, v> and (HcoNH, CONH} terms. Whatever the molecular mechanism ac- counting for the difference, in this case the group-contribution approach, based on2084 THE PEPTIDE-UREA INTERACTION a large set of data, can discriminate stereochemical differences. On the contrary no such a result was found for alcohols,lo p o l y 0 1 s ~ ~ - ~ ~ and sugars.42 The unique value of {H,,,,, CH,}, on the contrary, proves that the orientational effects, very important for interactions between polar groups, play only a marginal role when the -CHz- group is involved, regardless of its partner (cis- or trans-peptide group).The positive signs of the {HCONH, C H p } and { H I , , CH,} contributions, as well as that of (HCH,, cIT2}, suggest that the hydration cosphere of the hydrophobic groups releases ~ a t e r , ~ ~ - ~ l as is assumed to occur in hydrophobic interaction^.^^? 25y 6 2 v 63 If the mechanism of interactions implies the two cospheres and involves the partial disruption of the more labile hydrophobic one, then the conformation of the peptide group is expected to play a minor role. Differences are found between the value of {WCHp,CH2} reported in this work and those used in earlier studies10. 21, 40 or obtained for the aqueous solutions of hydroxylated ~ o m p o u n d s ~ ~ ~ 55-57 [3543 J mol-l (mol kg-l)-l, depending upon the basic set of data assumed].Moreover, large differences are found for the {Hu,CH,} contribution, when compared with the value reported for the solutions containing urea and an alcohol or a polyol [ 136 J mol-1 (mol kg-1)-1],60~ 61 and for {HCONH, CH2}, when compared with the value given for the interaction of amides with alcohols, polyols, sugars and cyclic ethers [3 1 J mol-l (mol kg-1)-1].53, 64 These discrepancies probably originate from the fact that the {Hii) values are affected, for each set of solutes, by different intramolecular contributions. These arise from neighbouring groups (which promote inductive effects etc). and from the perturbation exerted by neighbouring groups and their cospheres on the cospheres of the interacting molecules.A real limitation of the method is that the fitting procedure does not explicitly consider these intramolecular perturbations and distributes them among the intermolecular contributions { Hii}. Therefore the other conclusion of this work is that the { H i j } values reported in table 7 can be used for predicting exclusively the enthalpies of aqueous solutions of simple unknown solutes, characterized by alternating short alkyl chains and polar groups similar to those considered here. The accumulation of near polar groups, as shown for instance by bi~ret,~O invalidates the predictions of the method. Analogously, the prediction of the properties of compounds with alkyl chains longer than C , is hazardous. The example of the tripeptides shows that quantitative extensions to the biopolymers are only speculative at present.However, it can qualitatively be accepted that the overall conformations of the proteins are determined mainly by the intramolecular hydrophobic forces.? The polar-apolar cross interactions seem to be, vice versa, destabilizing factors. Finally it must be emphasized that the action of urea in the denaturation processes of the protein is multiple: urea interacts favourably with peptide groups, competing with the intramolecular interactions, and probably has a statistically perturbing action on the extended, ordered structures (chaotropism12). However, the action on the alkyl groups is controversial because of the large uncertainties that still affect the free-energy t As 2 5 * 29 also by model solute molecules, predominantly hydrophobic, the favourable Gibbs free energy for the hydrophobic interaction is determined by the entropic term, which overwhelms the unfavourable enthalpic one at least at 298.15 K.It seems that, by increasing the molecular weight of the peptide and by increasing the length of the alkyl side chains, the hydrophobic interaction becomes more cooperative. As a consequence, globular well defined tertiary structures are stabilized in solution, occasionally showing short segments o'f a-helix, 8-turns elc. If the hydrophobic forces are weak, then the possibility of arranging a geometrically favourable network of intermolecular hydrogen predominates, giving almost insoluble products.G.. BARONE, P. CACACE, v.ELIA AND A. CESARO 2085 group contributions.ll For the alkylureas (as well as for alcohols and alkylamides) there is competition between the destabilizing effect of the intermolecular alkyl-alkyl interactions and the stabilizing intramolecular ones, leading to an increase in the denaturing effectiveness of the alkyl derivatives. The cross interactions between the peptide groups of the protein backbone and the alkyl groups of the cosolute seem to lead, vice versa, to a salting-out effect that attenuates the denaturing action. This work was carried out with financial support from the Italian C.N.R. (Rome) and from the Ministry of Public Education, M.P.I. C. A. Swenson and L. Koob, J. Phys. Chem., 1970,74, 3376. H. Schonert and L. Stroth, Biopolymers, 1981, 20, 817. D.R. Robinson and W. P. Jencks, J. Am. Chem. SOC., 1965,87, 2462. H. Uedaira, Bull. Chem. SOC. Jpn, 1972, 45, 3068. M. Roseman and W. P. Jencks, J. Am. Chem. SOC., 1975, 97, 631. T. H. Lilley and R. P. Scott, J. Chem. SOC., Faraday Trans. I , 1976, 72, 184. K. P. Prasad and C. Alhuwalia, Biopolymers, 1980, 19, 273. L. Stroth and H. Schonert, J. Chem. Thermodyn., 1980, 12, 653. See for instance Faraday Symp. Chem. Soc., 1982, 17. lo J. J. Savage and R. H. Wood, J. Solution Chem., 1976, 5, 733. l 1 B. Y. Okamoto, R. H. Wood and P. T. Thompson, J . Chem. SOC., Faraday Trans. I , 1978,74, 1990. l2 F. Franks and D. Eagland, CRC Crit. Rev. Biochem., 1975, 3, 165. l 3 I. M. Klotz and J. S . Franzen, J. Am. Chem. SOC., 1962, 84, 3461. l4 G. E. Walrafen, J. Chem.Phys., 1966, 44, 3726. l5 G. Barone, E. Rizzo and V. Vitagliano, J. Phys. Chem., 1970,74,2230. l7 G. Barone, P. Cacace, G. Castronuovo and V. Elia, Gazz. Chim. Itai., 1980, 110, 215. l9 V. Crescenzi, A. Cesaro and E. Russo, Int. J. Pept. Protein Res., 1973, 5, 427. *O A. Cesaro, E. Russo and G. Barone, Int. J. Pept. Protein Res., 1982, 20, 8. 21 G. M. Blackburn, T. H. Lilley and E. Walmsley, J. Chem. SOC., Faraday Trans. I , 1980,76, 915. 22 G. M. Blackburn, T. H. Lilley and E. Wa!msley, J, Chem. Soc., Chem. Commun., 1980, 1091. 23 G. M. Blackburn, T. H. Lilley and E. Walmsley, J. Chern. SOC., Faraday Trans. I , 1982,78, 1641. 24 H. L. Friedman and C. V. Krishnan, J. Solution Chem., 1973, 2, 119. 25 F. Franks, M. D. Pedley and D. S. Reid, J. Chem. SOC., Faraday Trans.I, 1976, 72, 359. 26 W. G. McMillan and J. E. Mayer, J. Chem. Phys., 1945, 13, 276. 27 J. J. Kozak, W. S. Knight and W. Kaumann, J. Chem. Phys., 1968, 48, 675. 28 G. Barone, G. Castronuovo, V. Elia and A. Menna, J . Solution Chem., 1979, 8, 157. 29 G. Barone, G. Castronuovo, V. Crescenzi, V. Elia and E. Rizzo, J. Solution Chem., 1978, 7, 179. 30 G. Barone, P. Cacace, G. Castronuovo and V. Elia, Can. J. Chem., 1981,59, 1257. 31 G. Barone, G. Castronuovo, A. Cesaro and V. Elia, J. Solution Chem., 1980, 9, 867. 32 V. Abate, G. Barone, G. Castronuovo, V. Elia and P. Masturzo, Gazz. Chim. Itai., 1981, 111, 85. 33 V. Abate, G. Barone, P. Cacace, G. Castronuovo and V. Elia, J. Mol. Liquids (Adv. Mol. Relaxation 34 G. Barone, G. Castronuovo and V. Elia, J. Solution Chem., 1980, 9, 607.35 V. Abate, G. Barone, G. Castronuovo, V. Elia and V. Savino, J . Chem. Soc., Faraday Trans. 1, 36 H. L. Friedman, J. Solution Chem., 1972, 1, 387. 37 H. L. Friedman, J. Solution Chem., 1972, 1, 413. 3R A. M. Clark, F. Franks, M. D. Pedley and D. S. Reid, J. Chem. SOC., Faraday Trans. I, 1977,73,290. 39 A. Cesaro, E. Russo and D. Tessarotto, J. Solution Chem., n80, 9, 221. 40 R. H. Wood and L. H. Hiltzik, J. Solution Chem., 1980, 9, 45. 12 G. Barone, G. Castronuovo, D. Doucas, V. Elia and C. A. Mattia, J. Phys. Chem., 1983,87, 1931. 43 G. Barone, P. Cacace, G. Castronuovo and V. Elia, Carbohydr. Res., 1981, 91, 101. ‘*4 G. Barone, P. Cacace, G. Castronuovo, V. Elia and F. Iappelli, Carbohydr. Res., 1981, 93, 11. ‘” G. Barone, P. Cacace, G. Castronuovo, V. Elia and U. Lepore, Carbohydr. Res., 1983, 115, 15. 46 V. Crescenzi, F. Quadrifoglio and V. Vitagliano, J. Phys. Chem., 1967, 71, 2313. 4 7 F. Quadrifoglio, V. Crescenzi, A. Cesaro and F. Delben, J. Phys. Chem., 1971, 75, 3633. E. G. Finer, F. Franks and M. J. Tait, J. Am. Chem. SOC., 1972, 94, 4424. G. Barone, P. Cacace, G. Castronuovo and V. Elia, J. Chem. SOC., Faraday Trans. 1, 1981,77, 1569. Interaction Processes), 1983, 27, 59. in press. T. H. Lilley and R. H. Wood, J. Chem. Soc., Faraday Trans. 1, 1980, 76, 901.2086 THE PEPTIDE-UREA INTERACTION 48 G. Barone, G. Castronuovo, C. Della Volpe, V. Elia and L. Grassi, J. Phys. Chem., 1979,83, 2703. 4g J. E. Garrod and T. M. Herrington, J. Chem. SOC., Faraday Trans. I , 1981,77,2559. 50 I. R. Tasker and R. H. Wood, J. Solution Chem., 1982, 11, 729. 51 F. Franks and M. D. Pedley, J. Chem. SOC., Faraday Trans. I , 1981,77, 1341. 52 E. Benedetti, in Peptides, ed. M. Goodman and J. Meienhofer (J. Wiley, New York, 1977). 53 I. R. Tasker and R. H. Wood, J. Solution Chem., 1982, 11, 295. 54 G. Barone, B. Bove, G. Castronuovo and V. Elia, J. Solution Chem., 1981, 10, 803. 55 I. R. Tasker and R. H. Wood, J. Solution Chem., 1982, 11, 469. 56 I. R. Tasker and R. H. Wood, J. Phys. Chem., 1982, 86,4040. 57 G. Barone, P. Cacace, G. Castronuovo and V. Elia, Carbohydr. Res., 1983, 119, 1. 58 F. Franks and M. D. Pedley, J. Chem. Soc., Faraday Trans. 1, 1983, 79, 2249. 5g J. E. Desnoyers, G. Perron, L. Avedikian and J-P. Morel, J. Solution Chem., 1976, 5, 631. 6o G. Barone, G. Castronuovo and V. Elia, Adv. Mol. Relaxation Interaction Processes, 1982, 23, 279. 61 G. Barone, V. Elia and E. Rizzo, J. Solution Chem., 1982, 11, 687. ** Water: A Comprehensive Treatise, ed. F. Franks (Plenum Press, New York, 1975), vol. 4. F. Franks, Philos. Trans. R. Soc. London, Ser. B, 1977, 278, 33. 64 1. R. Tasker and R. H. Wood, J. Solution Chem., 1982, 11, 481. (PAPER 3/ 1749)

ISSN:0300-9599    

DOI:10.1039/F19848002073

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I , 1984, 80, 2087-2098 Kinetics of Polyelectrolyte Complexation Studied by the Conductance S t opped-flow Technique The Systems Polyacrylate-Polybrene and Pol y( E t hylene Sulp honate)-pol ye thylenimine BY TSUNEO OKUBO,* KENICHI HONGYO AND AKIRA ENOKIDA Department of Polymer Chemistry, Kyoto University, Kyoto, Japan Received 4th October, 1983 Polyelectrolyte complexation between weakly acidic (or basis) macroions and strongly basic (or acidic) ones, i.e. sodium polyacrylate (NaPAA) + polybrene(PB) and polyethylenimine (PEI) + sodium poly(ethy1ene sulphonate) (NaPES), has been analysed. The equilibrium association constants (KJ of PAA- + PB+ and PEIH+ + PES- and those (K2) of PAA + PB+ and PEIf PES- are estimated from pH measurements. Kl (lo5-lo7 dm3 mol-l) increases with degree of neutralization (a) of weak macroions and also with decreasing polymer concentration.K2 is insensitive to o! and to the concentration. Two relaxation times (z, and z2) are obtained in the traces of the relaxations of the polyelectrolyte complexations from conductance stopped-flow measurements and are assigned to a fast association process in complexation and a slow reorientation of the complex formed, respectively. The association (k,) and dissociation (kb) rates are between 1 x lo6 and 1.3 x lo7 dm3 rno1-l s-l and between 0.1 and 10 s-l. k, first increases with a but then decreases at large a, whereas k, is either insensitive or decreases with a. Both the enthalpy and entropy of activation of the complexation increase drastically with increasing a, which suggests significant hydration of the macroions and the importance of the dehydration of water molecules in the course of complexation. Polyelectrolyte complexes are a group of the complexation products of two highly but oppositely charged polymers in aqueous media.Intensive studies of their preparation, stability and other basic properties have been reported.l-* In particular, electrostatic interactions between macroions are helpful in understanding biological sy~tems.~-l~ Our interests have also been concerned with the application of these complexes to the fields of medicinel57 l6 and membrane ~cience.l~-~O However, almost no attention seems to have been paid to the kinetic properties of complexation, e.g. the reaction rate constant of complexation.This scarcity of kinetic measurements is partly due to limitations of the experimental technique. Most polyelectrolyte complex systems do not show clear changes in optical properties in the course of complexation and so the usual techniques for the analyses of fast reactions, such as temperature-jump and stopped-flow, are not applied. However, we expect that complexation is accom- panied by a significant change in conductance. Recently we constructed a conductance stopped-flow (c.s.f.) apparatus and applied it to various ionic reaction systems, i.e. Ni2+-murexide complexations catalysed by macroions,21 micellar equilibria of ionic surfactants,22 conforrnational changes of rnacr~ions,~~ polymer complexation between polyacrylic acid and poly~inylpyrrolidone,~~ the cationic polymerization of p-methoxystyrene in non- aqueous media24 and metal-macroion c~mplexation.~~ In the present work, we extend the application of this technique to polyelectrolyte complexes.20872088 KINETICS OF POLYELECTROLYTE COMPLEXATIONS EXPERIMENTAL MATERIALS Poly(acry1ic acid) (HPAA) was donated by the Nippon Junyaku Co., Tokyo (type P-1 lM, degree of polymerization 1500). The polymer was purified by dialysis for 7 days against pure water using Visking tube followed by ion exchange with columns of cation- and anion-exchange resins (Amberlite IR- 120B and IRA-400). Sodium poly(ethy1ene sulphonate) (NaPES) was purchased from Polyscience and further purified by repeated precipitation, dialysis and ion-exchange methods.Polybrene{a 3-6-type ionic polymer, poly[(dimethylimino)propan- 1,3 -diyl(dimethylimino)hexan- 1,6-diyl dibromide]) from Aldrich Co. was used after purificatior. by precipitation (ethylacetate as a precipitant). Polyethylenimine (PEI) from Polyscience was purified by ion-exchange methods. PH, CONDUCTANCE AND CONDUCTANCE STOPPED-FLOW MEASUREMENTS Potentiometric titration was carried out on a pH titrator (Radiometer Co., Copenhagen, model RTS 622). Electric conductance was measured on a Wayne-Kerr autobalance precision bridge (model B33 1 , mark 11) with Jones-Ballinger cell. The details of the c.s.f. apparatus have been described in a previous paper.22 RESULTS AND DISCUSSION STATIC PARAMETERS OF POLYELECTROLYTE COMPLEXATION The polyelectrolyte complexes covered in this study are those between weakly acidic (or basic) macroions and strongly basic (or acidic) macroions, i.e. PAA- and PB+ or PEI+ and PES-.Conductometric and potentiometric studies of the complexation of PAA- and PB+ have been carried out by Tsuchida et ~ l . , ~ although quantitative parameters such as stability constants were not obtained. The three equilibria kC PB+ + PAA- +complex kb Kl = kf/kb k; PB+ + HPAA +complex + H+ k l K, = k;/kL (2) HPAA $ H+ + PAA- K, = K 2 / K , . (3) should be taken into account in the kinetic analysis of the PAA--PB+. K, and K2 are the association constants of PB+ with polyacrylate anions (PAA-) and with poly(acry1ic acid) (HPAA), respectively, and K, is the apparent dissociation constant of HPAA. We obtain a further three equations from the principle of electroneutrality : [HPAA] = a( 1 -a) - [H+] (4) a = [HPAA] + [PAA-] + [complex] ( 5 ) b = [PB+]+[complex] (6) where a is the degree of neutralization of PAA, and a and b are the total concentrations of PAA and PB.From the above six equations, we derive Ax) = (1 -&) X3+[K2a(l -a)+h+(l -K,)(aa-b+K,/K,)] X 2 +K2a(l -a)[aa-b+(2K2- l)/Kl] X - ( e / K l ) a 2 (I -a)2 = o as a function of X (= [H+]). (7)T. OKUBO, K. HONGYO AND A. ENOKIDA 2089 0 0.5 1 added PB/cm3 Fig. 1. Observed (solid) and calculated (broken) curves of pH titration for PB-PAA complexation at 25 "C. [PB], = 0.006 mol dm-3, [PAA], = 3.16 x mol dm-3, a = (1) 0, (2) 0.1, (3) 0.2, (4) 0.4 and (5) 0.6. Table 1. Values of 8, K,, K, and K, for the complexation of PAA with PB at 25 "Ca 0.0005 0 0.1 0.2 0.4 0.6 0.8 0.0004 0.2 0.0003 0.2 0.0002 0.2 0.000 1 0.2 - - - - - 0.33 0.34 0.4 1 0.5 1 0.65 0.8 1 0.43 0.46 0.51 0.63 6.0 x 1 0 4 1.5 x 1 0 5 4.5 x 105 1.2 x 106 1.5 x lo6 2.0 x 106 7.0 x 105 7.0 x 105 9.0 x 105 1 .7 ~ lo6 0.25 0.20 0.25 0.25 0.25 0.25 0.30 0.40 0.70 2.0 4.2 x 1.3 x 5.6 x 10-7 2.1 x 10-7 1.7 x 10-7 1.3 x 10-7 4.3 x 10-7 7.8 x 10-7 1.4 x 1.2 x 10-6 a Experimental errors in 8, K,, K, and Ka are estimated as 10, 20, +20 and k 20%, respectively . Eqn (7) was solved by the Newton-Raphson method for arbitrary values of Kl and K,, and the pH curves thus obtained were compared with the experimental values obtained from pH measurements. From this comparison the most appropriate values of Kl and K2 were deduced. Fig. 1 gives a comparison of the pH curves observed and calculated for various values of a, Kl and K,.The agreement seems to be reasonable, especially at low degrees of neutralization. The deviation of the calculated values from the observed values may be due to the fact that the dissociation of H+ from PAA- macroions is strongly retarded at large a by the electrostatic attraction between the two kinds of ions. The most reasonable parameters thus estimated are compiled in table 1. 8 denotes the ratio of2090 KINETICS OF POLYELECTROLYTE COMPLEXATIONS Table 2. Values of 8, K,, K, and K,, for the complexation of PES with PEI at 25 "C" [PES], = [PEI], K , Kb /mol dm-3 a 6 /dm3 mol-l K2 /mol dmP3 0.0005 0.2 0.0004 0 0.1 0.2 0.4 0.6 0.8 0.0003 0.2 0.0002 0.2 - - - - - 0.29 0.17 0.28 0.33 0.48 0.64 0.8 1 0.26 0.29 3 x 106 1.2 x 106 2 x 106 5 x 106 I x 107 2 x 107 3 x 107 7 x los 9 x 106 0.05 0.04 0.1 0.1 0.15 0.2 0.3 0.03 0.05 2 x 10-8 3 x 10-8 5 x 10-8 2 x 10-8 1.5 x loP8 I x 10-8 1 x 10-8 6 x 4~ 10-9 a Experimental errors in 8, K,, K,, and Kb are estimated as k 10, +20, +20 and & 20%, respectively.the complexed macroions to the total concentration of the macroions, i.e. 6 = [complex]/[PAA],. The equality [PAA], = [PB], holds in our experiments. The increase in Kl with a may be interpreted by an increase in the electrostatic attraction between the dissociated macroanions and polybrene macrocations. K , was insensitive to a, because the complexation reactions proceed between neutral polymer (HPAA) and the macrocations without significant electrostatic interactions.At degrees of neut- ralization close to zero, 0 > a, whereas 6 -+ a at larger a. Similar results for the relation between 8 and a have been reported for other polyelectrolyte complexes.s Kl increased with decreasing polymer concentration. This is reasonable, because the macrocation-macroanion interactions decrease sharply with polymer concentration owing to the effect of electrostatic shielding. The equilibrium parameters of the PEI-NaPES complexes are estimated in table 2, fig. 2 shows a comparison of the observed pH curves with those calculated and the relevant equations are : kf PES-+ HPEI+ +complex kb Kl = kf/kb k f PES- + PEI + H20 complex + OH- K , = k;/kL (9) K h PEI + H,O + HPEI+ +OH- Kb = K,/Kl (10) f( Y) = K , a Y + K , K,(a + b - ba) + (1 - K,) ( K , + K , ba) Y + K,b(l -a) (Kl ba + K, - 1 + K , - K,a) Y - Kl( 1 - a),b2 =0, Y=[OH-] (1 1) a = [PES-] + [complex] b = [HPEI+] + [PEI] + [complex].(12) (1 3) The dependences on a of 8, K,, K2 and Kb seem to be very similar to those of the system PAA--PB+. However, the agreement between the observed and calculatedT. OKUBO, K. HONGYO AND A. ENOKIDA 209 1 I I 0 0.5 added PES/cm3 Fig. 2. Observed (solid) and calculated (broken) curves of pH titration for PES-PEI complexa- tion at 25 "C. [PES], = 0.01 mol dm-3, [PEI], = 5.26 x mol dm-3, a = (1) 0, (2) 0.1, (3) 0.2, (4) 0.4, (5) 0.6 and (6) 0.8. values is not so good, especially at large a, because our assumption that Kb is constant irrespective of a is too simple. DYNAMIC PARAMETERS OF POLYELECTROLYTE COMPLEXATION The changes in solution conductance during complexation are illustrated in fig.3. Clearly, the conductance is increased on complexation. Typical traces of the relaxations of PB+-PAA- complexation obtained from c.s.f. measurements are displayed in fig. 4. When aqueous solutions of PB and PAA were mixed rapidly two relaxations appeared, i.e. a rapid increase in conductance (7,) and a slow decrease in conductance (z2). In other words, the amplitude of the relaxation was much larger for z, than z,. The former was obtained from clear traces of the relaxation curves and were highly reliable, whereas the latter contained rather larger experimental errors (i- 15 %). The values of 7;' increased with increasing polymer concentration and decreased with increased ionic strength (see fig.5 later). The latter result may be explained by an electrostatic shielding effect of the foreign salt on the macrocation-macroanion attraction. From the above results we assigned z, and z, to the association process of the macroions and the reorientation of the complex formed, respectively. The two relaxations are given in scheme 1. In our c.s.f. measurements the concentration-jump occurs on mixing and is followed by relaxation to a new equilibrium concentration. Thus the chemical relaxation treatment is applicable. The kinetic equation for the rates of the polyelectrolyte complexation at a time t is written as d[PB+] dt --- - kf[PB+] [PAA-] - k,[complex] + k;[PB+] [HPAA] - kL[complex] [H+].2092 KINETICS OF POLYELECTROLYTE COMPLEXATIONS 0 0.5 1 added polybrenelcm3 Fig.3. Solution conductances in the titration of PAA with PB at 25 "C. [PAA], = 5 x rnol dm-3, [PB], = 5 x lop3 mol dmP3. Fig. 4. Typical traces of relaxation of the PAA-PB complexation reaction, a = 0.2. ( 1 ) full scale = 1 s, [PAA], = [PB], = 5 x mol dm-3 (20 "C); (2) full scale = 2 s, [PAA], = [PB], = 3 x mol dmp3 (25 "C); (3) full scale = 10 s, [PAA], = [PB], = I x lop4 mol dmp3 (15 "C). The small deviation of [PB+] from the new equilibrium concentration at time t , i.e. 6[PB+], is therefore given by dd[PB+] dt --= kf([PB+] G[PAA-] + [PAA-] d[PB+]) - k, G[complex] + k;([PB+] 6[HPAA] + [HPAA] 6[PB+]) - kb([complex] 6[H+] + [H+] S[complex]).T. OKUBO, K. HONGYO AND A. ENOKIDA 40 I--+- 2093 I I I 5x10-" 1.5X10'3 5x10 O L J 0 [ NaBr] /mol dm-3 Fig.5. Influence of the addition of NaBr on z;' for the complexation of PAA with PB at 25 "C. a = 0.4, [PAA], = [PB], = 5 x mol dmP3. PAA PB complex Scheme 1. Reaction mechanism for the complexation of PAA with PB. The left-hand side of eqn (1 5) corresponds to the [PB+]/z,, by relaxation theory. Thus z;l of the system PB+-PAA- is given by eqn (16) zll = Ak, + Bk; (16) K,[PB+] + [PB+] [PAA-] 1 A = +[PAA-]+- K,[PAA-] + X Kl (18) X[PB+I X [complex] X B = K+ [PAA-] + X +[HPAA1+%+K,(K,+[PAA-]+X) in terms of k, and k; with the help of the relationsfor the mass balance of the reaction components, i.e. G[PB+] + G[complex] = 0 G[PAA-] + G[complex] + G[HPAA] = 0 and d[H+]+d[HPAA] = 0 K,G[HPAA] - [PAA-] G[H+] - [H+] G[PAA-] = 0.2094 KINETICS OF POLYELECTROLYTE COMPLEXATIONS 1 I I 1 7 1 I Fig.6. (z,B)-l A IB mol dm-3, a = 0.2. plotted against A / B for PB-PAA complexation at 15 "C. (1- 5 ) x [PAA], = The association and/or dissociation of protons with polyacrylate anions given by eqn (3) is safely assumed to be very fast compared with the complexations given by eqn (1) and (2). Therefore, the fast relaxation time z1 results from eqn (1) and (2). The slopes of linear part of plots of (z,B)-l against A / B give k,. Values of k; were not estimated from the intercepts because of their large experimental errors. Typical plots are displayed in fig. 6. The values of k, and kb( = k,/Kl) thus estimated are given in fig. 7 and 8, respectively. k, increased sharply with increasing degree of neutralization.This is reasonable, because the electrostatic attraction between PAA- and PB+ would be strengthened on increasing a. The tendency of k, to decrease at large a may be due to the hydrolysis of the weakly acidic polymer PAA and resultant electrostatic shielding of the dissociated NaCI. On the other hand, kb was rather insensitive to a or slightly decreased with increasing a. From the temperature dependence of kf, the activation parameters [i.e. the free energy (Act), enthalpy ( A m ) and entropy ( A S ) of activation] were evaluated and are given in table 3. AGt was not sensitive to a. However, both AH1 and ASS increased strikingly with increasing a. The increase of AH$ reflects the enhanced hydration of PAA- ions and the raised energy barrier to the dehydration of the previously hydrated macroions in the course of complexation.A large positive value of AS$ at large a supports the idea that the dehydration of water molecules, which were bound originally to the macroions by electrostatic forces, does occur in the course of macrocation-macroanion binding. An overall picture of dehydration is shown in scheme (2). The gegenions are not shown in this scheme. This type of dehydration becomes very significant compared with the dehydration of simple electrolytes, as was discussed p r e v i o ~ s l y . ~ ~ - ~ ~ A negative value of AS3 at a = 0 probably means that the decrease in the entropy of the macroions themselves on complexation is more important than the dehydration effect. A Free water \ Hydrated water + (-7 <- F Scheme 2.Dehydration in the course of polyelectrolyte complexation.T. OKUBO, K. HONGYO AND A. ENOKIDA 0 2095 0.2 0.4 0.6 01 Fig. 7. k, plotted against a for the complexation of PAA with PB. 0, 10; x , 15; A, 20 and 0, 25 "C. 10 - 8 5 0 & 0 I I I I 0 01 Fig. 8. k , plotted against a for the complexation of PAA with PB. 0, 10; x , 15; A, 20 and 0 , 2 5 "C. Table 3. Activation parameters for the complexation of PAA with PB at 17.5 O C a degree of neutralization (a) 0 0.1 0.2 0.4 0.6 AGl/kJ mol-l 38.5 37.6 34.7 33.9 34.3 A#/kJ mol-l 4.2 46.8 48.5 75.7 86.1 A$/J K-l mol-l -117 33 46 146 180 a The experimental uncertainty is believed to be kO.6 for AGt, 1.0 for AH$ and f 2 0 for AS:.2096 KINETICS OF POLYELECTROLYTE COMPLEXATIONS 0 \ 0 - 0.2 0.4 0.6 0.8 Fig.9. k, plotted against a for the complexation of PEI with PES at 25 "C. 4 3 - I z 2 A? 1 0 I I I I 0 0 0 I I 110 0.2 0.4 0.6 0.8 a Fig. 10. k , plotted against a for the complexation of PEI with PES at 25 "C. Kinetic analyses of the PEI-PES systems were also made from eqn (1 6), (19) and (20) : qPES-] Y ncom plex] B = + [PEI]+-+ Kb + [HPEI+] + Y K , &(Kb + [HPEI+] 4- r)' The magnitudes and a dependences of k, and k, are very similar to those of the system PB-PAA, as seen in fig. 9 and 10, respectively.T. OKUBO, K. HONGYO AND A. ENOKIDA 2097 0 2 4 6 [PAA] and [PB]/10-4 mol dm-3 Fig. 11. Macroion concentration dependences of ~ ; l for the complexation of PAA with PB at 20 "C. a = 0.4 (O), 0.6 (0) and 0.8 (A). Slow relaxations (t2) observed in the c.s.f.measurements are displayed in fig. 11. Two points are seen: (1) as a increases, z;' decreases; (2) z;l is insensitive to polymer concentration except at small a (a = 0.4 in the figure). Result (1) is very similar to the dependence of the reciprocal of the relaxation times for conformational changes of sodium p~lyacrylate.~~ This would probably mean that the reorientation velocity of the complex formed is larger for the complex having a more compact 'coil' in its conformation. Result (2) may indicate that the reorientation process is a mono- molecular and essentially intra-complex reaction [see scheme (I)]. However, the change in z, with polymer concentration, especially that observed for the complex at a =: 0.4, suggests that inter-complex interactions are not neglected.We thank Prof. N. Ise, Kyoto University, for encouragement and discussion. This work was supported by a grant-in-aid for special project research on biomedical materials administered by the Japanese Ministry of Education, Science and Culture. R. M. Fuoss and H. Sadek, Science, 1949, 110, 552. A. S. Michaels and R. G. Miekka, J. Phys. Chem., 1961, 65, 1765. A. Veis, Biological Polyelectrolytes (Marcel Dekker, New York, 1970). A. Nakajima and H. Sato, Biopolymers, 1972, 11, 1345. B. A. Kabanov and N. M. Papisov, Vysokomol. Soed., Ser. A , 1979, 21, 243. A. Domard and M. Rinaudo, Macromolecules, 1980, 13, 898. H. Morawetz and W. L. Hughes Jr, J. Phys. Chem., 1952, 56, 64. lo G. Bhat, A. C. Roth and R. A. Day, Biopolymers, 1977, 16, 1713. l 1 T.M. Lohman, P. L. DeHaseth and M. T. Records Jr, Biophys. Chem., 1978, 8, 281. l 2 S. L. Kielland, L. A. Slotin and R. E. Williams, Can. J. Chem., 1978, 56, 2650. l 3 R. B. Cundall, J. B. Lawton and D. Murray, Makromol. Chem., 1979, 180, 2913. l 4 L. C. Yip and M. E. Balis, Biochemistry, 1980, 19, 1849. l5 M. K. Vogel, R. A. Cross, H. J. Bixler and R. J. Guzman, J. Macromol. Sci., Chem., 1970, 4, 675. l6 H. P. Gregor, Biomedical Application of Polymers (Plenum Press, New York, 1975). l 7 E. Renkin, J. Gen. Physiol., 1954, 38, 225. l8 A. S. Michaels, Znd. Eng. Chem., 1965, 57, 32. l9 M. F. Pefojo, J. Appl. Polym. Sci., 1965, 9, 3417. ti E. Tsuchida and T. Osada, Makromol. Chem., 1974, 175, 583. ' J. Diiszeln, V. Martin and J. Klein, Makromol. Chenr., 1979, 180, 255.2098 KINETICS OF POLYELECTROLYTE COMPLEXATIONS *O R. I. Kalyuzhnaya, A. L. Volynskii, A. R. Rudman, N. A. Vengerova, Ye. F. Razvodovskii, B. S. 21 T. Okubo and N. Ise, Polym. Bull., 1978, 1, 109. 22 T. Okubo, H. Kitano, T. Ishiwatari and N. Ise, Proc. R. SOC. London, Ser. A , 1979, 366, 81. 23 T. Okubo, Biophys. Chem., 1980, 11, 425. 24 M. Sawamoto, T. Higashimura, A. Enokida and T. Okubo, Polym. Bull., 1980, 2, 309. 25 T. Okubo and A. Enokida, J. Chem. Soc., Faraday Trans. I , 1983, 79, 1648. 26 K. J. Laidler and P. S. Bunting, The Chemical Kinetics and Enzyme Action (Clarendon Press, Oxford, 27 N. Ise, T. Maruno and T. Okubo, Proc. R. SOC. London, Ser. A , 1980, 370, 485. ** T. Okubo, T. Maruno and N. Ise, Proc. R. SOC. London, Ser. A , 1980,370, 501. 29 T. Okubo and N. J. Turro, J. Phys. Chem., 1981,85,4034. El’tsefon and A. B. Zezin, Vysokomol. Soed., 1976, 18, 71. 2nd edn, 1973). (PAPER 3/ 1750)

ISSN:0300-9599    

DOI:10.1039/F19848002087

出版商:RSC    

年代:1984

数据来源: RSC