trends in colloid and interface science v

PROGRESS IN COLLOID & POLYMER SCIENCE
Editors: H.-G. Kilian (Ulm) and G. Lagaly (Kiel)
Volume 84 (1991)
Trends in Colloid and Interface Science V Guest Editor: M. Corti (Pavia) and F. Mallamace (Messina)
0 Steinkopff Verlag • Darmstadt Springer-Verlag • N e w York 4
ISBN 3-7985-0885-2 (FRG) ISBN 0-387-91399-8 (USA) ISSN 0340-255-X
This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically these rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. Duplication of this publication or parts thereof is only permitted under the provisions of the German Copyright Law of September 9, 1965, in its version of June 24, 1985, and a copyright fee must always be paid. Violations fall under the prosecution act of the German Copyright Law.
© 1991 by Dr. Dietrich Steinkopff Verlag GmbH & Co. KG, Darmstadt. Chemistry editor: Dr. Maria Magdalene Nabbe; English editor: James Willis; Production: Holger Frey.
Printed in Germany.
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of specific state- ment, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.
Type-Setting: Graphische Texterfassung, Hans Vilhard, D-6126 Brombachtal Printing: betz-druck gmbh, D-6100 Darmstadt 12
Preface
The first meeting of the European Colloid and In- terface Society (ECIS) was held in Como, Italy, in 1987. Three years later, following meetings in Ar- cachon and Basel, the ECIS Conference was again held in Italy at Copanello di Catanzaro in September 1990. This gathering was attended by participants from 21 countries, including the USA and Russia. More than 150 papers were presented either orally or as posters. This volume includes most of these papers, which have been rather ar- bitrarily subdivided into six sections: Micelles, Microemulsions, Application of Colloids, Interac- tion and Ordering, Biological Macromolecules, and Layers and Interfaces. The interdisciplinary nature of these fields bordering between physics and chemistry is evident. Unfortunately, it was, of course impossible to reproduce in this volume the lively, friendly atmosphere of the meeting; discussions outside the conference room were wide-rang- ing and fruitful.
On behalf of the ECIS, we thank: all the participants for their contributions; the scientific c o r n -
mittee: M. Almgren, S. J. Candau, R. Klein, R. H. Ottewfll, R. Strey and M. Zulauf; from the Italian Ministry of University and Scientific Research, Prof. A. Ruberti, and the President of the Regional Government of Calabria, Dr. R. Olivo, both for their dedicated patronage; from the city of Catanzaro, Mayor Dr. M. Furriolo and Cultural Attach6 G. Guerriero; Prof. G. Stagno D~lcontres, Rector of Messina University; and the generous sponsors who made the Copanello meeting possible: The Italian Consiglio Nazionale delle Ricerche support- ing the publication of this issue, the Departiment of Physics of Messina University, the Assessorato Agricoltura della Regione Calabria, IBM-Italy, Spec- tra Physics, dB Electronics, and Chemifarm. Finally, particular thanks go to our hosts and the staff of Villaggio Guglielmo in Copanello for their hospitality.
Mario Corti Franco Mallamace
Micelles
Safran SA, MacKintosh FC, Pincus PA, Andelman DA: Spontaneous vesicle formation by mixed surfactants. 3 Thalberg K, Lindman B, Karlstr6m G: Electrolyte dependent phase separation in aqueous mixtures of a polyelec-
trolyte and an ionic surfactant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 van Stare J, Almgren M, Lindblad C: Sodium dodecylsulfate-poly(ethyleneoxide) interactions studied by time-
resolved fluorescence quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Cantu L, Corti M, Musolino M, Salina P: Spontaneous vesicle formation from a one-component solution of a
biological surfactant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Hoffmannn H, Hofmann S, Rauscher A, Kalus J: Shear-induced transitions in micellar solutions . . . . . . . . . . . 24 Lin T-L, Liu C-C, Roberts ME Chen S-H: Mixed short-chain lecithin/long-chain lecithin aggregates studied by
small-angle neutron scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Appell J, Porte G: Polymer-like giant micelles. An investigation by light scattering . . . . . . . . . . . . . . . . . . . . . . . . 41 Glatter O: Scattering studies on colloids of biological interest (Amphiphilic systems) . . . . . . . . . . . . . . . . . . . . . . 46 Baglioni P, Dei L, Ferroni E, Kevan L: Electron spin echo modulation and electron spin resonance studies of
sodium dodecylsulfate and dodecyltrimethylammonium bromide micellar solutons: Effect of urea addition 55 Hill A, Candau F, Selb J: Aqueous solution properties of hydrophobically associating copolymers . . . . . . . . . . 61 Despotovi4 R: On mixed surfactant systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Miinch C, Hoffmann H, Ibel K, Kalus J, Neubauer G, Schmelzer U, Selbach J: A shear-induced structure transi-
tion on a micellar solution measured by time-dependent small-angle neutron scattering . . . . . . . . . . . . . . . . . 69 J6hnannsson R, Almgren M: A fluorescence and phosphorescence study of AOT/H20/aikane systems in the L 2
reversed micellar phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 L6froth J-E, Johansson L, Norman A-C, Wettstr6m K: Interactions between surfactants and polymers. I: HPMC 73 L6froth J-E, Johansson L, Norman A-C, Wettstr6m K: Interactions between surfactants and polymers. Ih
Polyelectrolytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 Malliaris A, Binana-Limbele W: Solubilization of aprotic additives in aqueous micelles . . . . . . . . . . . . . . . . . . . . 83 Tsiourvas D, Paleos CM, Malliaris A: Aggregation of polyamphiphiles with the polar head on the main chain 86 Onori G, Ronca M, Santucci A: Properties of water solubilized in reversed AOT micelles from near-infrared
spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Onori G, Ronca M, Santucci A: Shape and solvation of water-containing reversed AOT micelles from viscosity
measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Ravey JC, Gherbi A, St6b6 MJ: Mixed systems of fluorinated and hydrogenated nonionic surfactants: The
air/water adsorbed film and micelles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Rauscher A, Rehage H, Hoffmann H: Stretched exponential relaxation processes in viscoelastic surfactant
solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Schubert K-V, Strey R, Kahlweit M: 3PHEX: A new surfactant purification technique . . . . . . . . . . . . . . . . . . . . . . 103 Tondre C, Derouiche A: Solubilization of electrolyte solutions in AOT reversed micelles. Conductivity percola-
tion and phase behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Treiner C, Bury R: Peculiar micellar solubilization of benzyl alcohol in binary benzyldimethyltetradecylam-
monium chloride and trimethyltetradecylammonium chloride solutions: A calorimetric investigation . . . . . . 108 Korolenko EC, Shokhirev NV: Spin-controlled reactions on the micellar surface . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Microemulsions
Teixeira J, Alba-Simionesco C, Angell CA: Glass transition in microemulsions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Senatra D, Lendinara L, Giri MG: W/O microemulsions as model systems for the study of water confined in
microenvironments: Low resolution 1H magnetic resonance relaxation analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 122
VIII Progress in Colloid & Polymer Science, Vol. 84 (1991)
Atkinson PJ, Clark DC, Howe AM, Heenan RK, Robinson BH: Characterization of microemulsion-based organogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Baglioni P, Gambi CMC, Goldfarb D: Pulse electron spin resonance and quasi-elastic light scattering of Winsor microemulsions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Rouch J, Safouane L, Cametti C, Codastefano P, Tartaglia P, Chen SH: A dynamic transition at the percolation threshold of a three-component microemulsion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Paillette M: Phase electric birefringence measurements in attractive-type W/O microemulsion systems . . . . . . . 144 Liano P, Duportail G: Fractal models for luminescence probing of organized assemblies. Studies with respect
to the nature of the assembly, the temperature, and the quencher concentration . . . . . . . . . . . . . . . . . . . . . . . . 151 Mallamace F, Magazu S, Micali N, Salvetti P: Microemulsion as model system for the study of the glass-like tran-
sition: Refractive index and calorimetric measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 Mallamace F, Micali N, Vasi C, D?~rragio G, Paparelli A: Hypersound velocity measurements in dense
microemulsions, evidence of a viscoelastic behavior connected with the percolation process . . . . . . . . . . . . . . 159
Interfaces
Woermann D: Critical phenomena in associative binary liquid mixtures with miscibility gap . . . . . . . . . . . . . . . 165 Kuzmin SV, Malomuzh NP: Surface-induced polarization properties of highly viscous liquids . . . . . . . . . . . . . . 171 Dgkrrigo G, MaUamace E Micali N, Paparelli A, Teixeira J, Vasi C: Aggregation phenomena in water-alcohol solu-
tions. Thermodynamic and dynamic studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Aveyard R, Binks BP, Fletcher PDI: Effects of subphase pH on the successive deposition of monolayers of
docosanoic acid onto mica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 Gambaro M, Gliozzi A, Robello M: Effect of surface charges on the electroporation process in lipid bilayers. 189 Meunier J, Henon S: Optical study of monolayers at liquid interfaces: Direct observation of first order phase tran-
sitions and measurement of the bending elastic constant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194 Nahrinbauer h The interaction between polymer and surfactant as revealed by interfacial tension . . . . . . . . . . 200 R6hl W, von Rybinski W, Schwuger MJ: Adsorption of surfactants on low-charged layer silicates. Part I: Adsorp-
tion of cationic surfactants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Bartolotta A, Di Marco G, Carini G, Tripodo G: Study of local and cooperative molecular movements in Po-
ly(ethylene oxide) -- Potassium thiocyanate complexes by mechanical measurements . . . . . . . . . . . . . . . . . . . . 215 Caminati G, Tomalia DA, Turro NJ: Photo-induced electron transfer at polyelectrolyte-water interface . . . . . . . 219 da Gra~a M. Miguel M, Burrows HD: Luminescence study of fluidity in the L a mesophase and liquid phase of
lead(H) decanoate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Bartolotta A, Di Marco G, Carini G, Tripodo G: Relaxation processes in polymeric electrolytes: Effect of the ca-
tion size and of the thermal history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 Gabrielli G, Puggelli M, Prelazzi G: Mono- and multi-layers containing ion carriers . . . . . . . . . . . . . . . . . . . . . . . 232 Gallegos C, Nieto M, Nieto C, Mufioz J: Influence of surfactant concentration on the time-dependent theological
behavior of the lamellar liquid crystal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236 G6bel S, Hiltrop K: Influence of organic counterions on the structure of lyotropic mesophases . . . . . . . . . . . . . 241 Miller CA, Gradzielski M, Hoffmann H, Kr/imer U, Thunig C: L 3 phases: Their structure and dynamic
properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Kaeder U, Hiltrop K: Alignment of lyotropic nematics by surface action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 Lachaise J, Sahnoun S, Dicharry C, Mendiboure B, Salager JL: Improved determination of the initial structure
of liquid foams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Papirer E, Perrin JM, Siffert B, Philipponneau G: Surface characteristics of colloidal aluminas and barium
titanates determined by inverse gas chromatography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Sch6n G, Peschel G, Stobbe H: Impedance-spectroscopic investigations of water structure near silica surfaces 262 Porte G, Appell J, Bassereau P, Marignan M, Skouri M, Billard I, Delsanti M, Candau SJ, Strey R, Jahn W, Snabre
P: Scaling laws for some physical properties of the L 3 phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 Paluch M: Effect of halogeno substituted ethyl alcohols on the surface potential and on the surface tension at
the water/air interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 Burger A, Rehage H: Two-dimensional model networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 Rolandi R, Dante S, Maga L, Robello M: Domains formation in polymerized monolayers revealed by
fluorescence microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 Schroder A, Candau SJ: Study of the swelling of latex particles by means of ultrasonic techniques . . . . . . . . . 275 Vandevyver M, Roulliay M, Bourgoin JP, Barraud A, Morand JP, Noel O: Structure-reactivity relationship in
Langmuir-Blodgett films of bisethylenedithio-tetrathiafulvalene (BEDT-TFF) derivatives . . . . . . . . . . . . . . . . . . . 279 Has M, Lfidemann H-D: p,T dependence of the hydrophobic interaction in a model solution . . . . . . . . . . . . . . 283 Shokhirev NV, Burshtein AI: The change in density and pressure tensor at the liquid-vapor interface . . . . . . . 285
Contents IX
Vituhknovsky AG, Sluch MI: Optical properties of Langmuir-Blodgett films: perylene excimer formation . . . . 288 Churaev N, Kotov A, Solometsev Y, Starov V: The influence of charged gel layers on the electrokinetic
phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 Kotov A, Solomentsev Y, Starov V: Direct approach of two particles covered with a porous layer . . . . . . . . . . . 293
Application of Colloids
Bongiovanni R, Ottewill RH, Rennie AR: Small-angle neutron scattering from dispersions of organophilic clays 299 Carpineti M, Giglio M, Paginini E, Perini U: Low-angle static light scattering by fast aggregation of polystyrene
latex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Siffert B, Badri F: Competition between micellization and adsorption of alkyl-PEO diblock copolymers on
titanium dioxide particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 Fucile E, Denti P, Saija R, Borghese F, Sindoni OI: Density dependence of the extinction coefficient of a disper-
sion of spherical metal particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 Carri6n Fit6 FJ: Electrokinetic behavior of polyester and solid impurity during washing process in the presence
of cellulose ethers and NTA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Herzog B: Micelle shape and capacity of solubilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 Callejas-Fern~indez J, de las Nieves FJ, Martinez-Garcfa R, Hidalgo-Alvarez R: Electrokinetic characterization and
colloid stability of calcium oxalate monohydrate dispersions in the presence of certain inhibitors . . . . . . . . . . 327 Jenta TR-J, Robinson BH, Batts G, Thomson AR: Enzyme kinetic studies using lipase immobilised in microemul-
sion-based organogels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334 Mendiboure B, Graciaa A, Lachaise J, Marion G, Bourrel M, Salager JL: Influence of the intensity of mixing on
the droplet size distribution of emulsions: Theory and experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338 Lisiecki I, Lixon P, Pileni MP: Synthesis in situ in reverse micelle of copper metallic clusters . . . . . . . . . . . . . . . 342 Tondre C, Claude-Montigny B, Ismael M, Scrimin P, Tecilla P: Metal-ion complexation by micelle-solubilized
long-chain complexing agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345 Te~ak D, Heimer S, Derek V, Strajnar F: Precipitation of aluminium with surfactant in sea-water . . . . . . . . . . . 348 Palberg T, Hartl W, Deggelmann M, Simnacher E, Weber R: Comparison of charge numbers of interacting latex
spheres from different experiments: Conductivity, electrophoresis, torsional resonance detection, and static light scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
Schulz SF, Maier EE, Hagenbfichle M, Graf Ch, Weber R: Structural properties of dilute aqueous solutions of charged rods studied by light-scattering techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
Di Biasio A, Bolle G, Cametti C, Codastefano P, Tartaglia P: Light scattering from aggregating colloids: Stretched exponential behavior of the time correlation function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359
Weyerich B, DAguanno B, Canessa E, Klein R: On the structure of suspensions of charged rodlike particles. 362
Interaction and Ordering
Candau SJ, Ilmain F, Moussa'id A, Schosseler F: Structure and properties of partially neutralized poly(acrylic acid) gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
Ostrowsky N, Gamier N: Brownian dynamics close to a wall, measured by quasi-elastic light scattering from an evanescent wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
Arauz JL, Ruiz-Estrada H, Medina-Noyola M, Nagele G, Klein R: Tracer-diffusion in binary mixtures of charged spherical macroparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
D~guanno B, M6ndez-Alcaraz J, Klein R: Structure and thermodynamics of mixtures of charged spherical colloidal particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
Granfeldt MK, J6nsson B, Woodward CE: The interaction between charged colloids with adsorbed polyelectrolytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
Palberg T, Simon R, Leiderer P: Forced Rayleigh scattering in mixtures of colloidal particles . . . . . . . . . . . . . . . . 397 Mimouni Z, Mathis C, Bossis G: Analysis of alignments of colloidal spheres by light scattering . . . . . . . . . . . . 402 Peschel G, van Brevern O: The contribution of hydration forces to particle-particle interaction in a silica hydrosol 405 Chang S-L, Chen S-H, Rill RL, Lin JS: Measurement and interpretation of counterion distribution around cyclin-
drical polyelectrolytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 Chabalgoity-Rodrfguez A, Marffn-Rodrfguez A, Galisteo-Gonz~lez F, Hidalgo-Alvarez R: Electrophoretic mobili-
ty, primary electroviscous effect and colloid stability of highly charged polystyrene latexes . . . . . . . . . . . . . . . 416 Lemaire E, Paparoditis C, Bossis G: Yield stress in magnetic suspensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425 Mallamace F, Micali N, Vasi C: Role of the ionic strength in the viscosity of charged colloids . . . . . . . . . . . . . . . 428
X Con~n~
Bio log ica l Macromole s
Margheri E, Bonosi F, GabrieUi G, Martini G: Spectroscopic investigation on the effect of the addition of ceramide into lipid vesicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
Giordono R, Grasso A, Teixeira J, Wanderlingh E Wanderlingh U: SANS in lysozyme solutions . . . . . . . . . . . . 439 Huruguen JP, Pileni MP: Changes in the percolation threshold by cytochrome c addition in AOT reverse micelles 442 Gallardo V, Bolivar M, Salcedo J, Delgado AV: A study of the effect of different amino acids on the electrical
properties of nitrofurantoin suspensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447 De Cuyper M, Joniau M: Effect of dimethylsulfoxide on the kinetics and thermodynamics of asymmetric
phospholipid fluxes between magnetic and non-magnetic vesicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456 Deriu A, Cavatorta E Cabrini D, Middendorf HD: Molecular structure and dynamics of biopoylmer gels by
neutron scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461 Cametti C, De Luca E D'nario A, Macri MA, Briganti G, Maraviglia B: The ripple phase in model membrane
systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465 Domazou AS, Mantaka-Marketou AE: Fluidity variation of DODAB vesicular membranes with estrogen hor-
mone using the lucigenin chemiluminescent reaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470 Edwards K, Almgren M: Solubilization of lecithin vesicles by C12E8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 472 Fisicaro E, Pelizzetti E, Lanfredi E, Savarino P: Osmotic coefficients of N-nonyl- and N-decyl-nicotinamide
chloride surfactant aqueous solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474 Aliotta E Fontanella ME, Magazu" S, Wandeflingth F: Hypersonic properties in macromolecular aqueous
solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483 Giordano R, Grasso A, Wanderlingh F, Wanderlingh U: Static and dynamic properties in thixotropic structures 487 G~ilvez-Ruiz MJ, Cabrerizo-Vflchez MA, Galisteo-Gonz~lez F, Hidalgo-Alvarez R: Study of temperature and pH
effects on phase transition liquid expanded/liquid condensed of cholesterol, lecithin and lithocholic acid mixed monolayers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494
Seras M, GrabieUe-Madelmont C, Paternostre M-T, Ollivon M, Handjani-Vila R-M: Study of non-ionic monoalkyl amphiphile-cholesterol vesicles solubilization by octylglucoside . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502
Staunton S, Quiquampoix H: The use of a trace amount of methylated bovine serum albumin as a probe of the state of bovine serum albumin adsorbed on montmorillonite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506
Xenakis A, Valis TP, Kolisis N" Microemulsions as a tool for enzymatic studies: The case of lipase . . . . . . . . . . 508
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 512
Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514
Progress in Colloid & Polymer Science Progr Colloid Polym Sci 84:3--7 (1991)
Spontaneous vesicle formation by mixed surfactants
S. A. Safranl'4), E C. MacKintosh1), P. A. Pincus2), and D. A. Andelman 3)
1) Exxon Research and Engineering, Annandale, New Jersey, USA 2) Materials Department, University of California, Santa Barbara, California, USA 3) Raymond and Beverly Sackler Faculty of Exact Sciences, School of Physics and Astronomy, Tel Aviv University,
Ramat Aviv, Israel 4) Polymer Dept., Weizmann Institute, Rehovot, Israel
Abstract: Although single surfactants rarely form vesicles spontaneously, mixtures of two surfactants can lead to spontaneous vesicle formation. By considering the curvature elasticity of the surfactant bilayer, we show theoretically how the energetic stabilization of mixed vesicles can occur. In- teractions between the two species (of the proper sign and magnitude) are crucial to stabilizing these vesicles. These interactions lead to composition asymmetries and effective spontaneous curvatures of the inner and outer layers that are of equal and opposite signs.
Key words: _Vesicle; surfactant; selfassociation
I. Introduction
Since vesicles rarely form as the equilibrium structure of simple surfactant-water systems, non- equilibrium methods, such as sonication of lameUar liquid crystalline phases, are usually necessary to obtain a metastable phase of vesicles, which may reequilibrate back into the multilamellar, liquid crystalline structure. Recently, however, Kaler et al. [1] have reported a general method for producing equilibrium phases of vesicles of a controlled size. The vesicles form spontaneously upon mixing simple surfactants with oppositely charged head groups. Most previous reports of spontaneous vesicle formation have also involved surfactant mixtures [2--5]. Using the charge as a control parameter has both chemical and physical advantages since a wide variety of head group, counterion, and salt chemistries can be prepared and studied.
In this paper, we use the concepts of curvature elastic theory [6] to explain the stability of vesicles formed in mixed surfactant systems. In systems composed of a single surfactant, the curvature energy of a bilayer dictates that the energy of a phase of spherical vesicles is never lower than that of a multilamellar, liquid crystalline phase [7, 8]. This is because the bilayer is composed of two am-
phiphilic monolayers which, in the single surfactant case, have the same spontaneous curvature [6]. Since the two layers have curvature of opposite sign (e.g., the inner one being concave with respect to the water, and the outer one convex), the system is frustrated. Small vesicles, where the vesicle radius is of the order of the surfactant size, can be of lower energy than fiat bilayer, as discussed in [9--12]. However, they may be of higher free energy than small micelles. In this work, we consider the case of large vesicles and discuss their stability with respect to lamellar phases; this feature can be compared with the experimental phase diagrams [13]. We find [7, 8] that the stabilization of the vesicles by surfactant mixtures only occurs when interactions of the surfactants is considered; ideal mixing of the two components does not yield vesicles as the ground state. These results can be used to see how the interactions can be exploited to control and stabilize the vesicle phase.
II. Mixed vesicles
In contrast to the situation for single amphiphiles, where large vesicles are usually not energetically stable in comparison with fiat bilayers, vesicles
4 Progress in Colloid & Polymer Science, Vol. 84 (1991)
composed of two amphiphiles can have lower curvature energies than fiat films. The curvature energy [6--8] per unit area of the vesicle is given by
fc = 2K[(c + c0) 2 + (c - -c i ) 2], (1)
where K is the bending elastic modulus [7, 8], q and c o are the spontaneous curvatures of the inner and outer monolayers, and c is the actual curvature of the inner layer. For the case of single surfactant systems, in the limit of small curvatures, c o = c~. In this case, the minimum of fc with respect to c im- plies that c = 0; fiat bilayers are the lowest bending energy state. For mixed surfactants, constitutive relations for effective spontaneous curvatures of the inner and outer layers, c i and c o are needed.
For simplicity, we consider a model where the spontaneous curvatures of films composed of each, single surfactant are equal, cl = c 2, and define ¢/as the volume fraction of surfactant type "2" in the system. In addition, we define ¢/~ and ¢/0 as the volume fraction of surfactant "2" in the inner and outer layers, respectively. The composition difference between these two layers is rp = 1/2(¢/0 -- ~i), with the constraint of fixed ~, --- 1/2(¢/0 + ~,~).
We now describe a simple statistical model for the surfactant head-head interactions which allows for a unified treatment of the free energy of the system including both the elastic, entropic, and interaction contributions. Our basic assumption is that the interaction between head groups alone determines the spacing between surfactants at the interfaces, while the resulting compression of the surfactant tails determines the spontaneous curvature of each monolayer. (In [11], we shall relax this assumption.) In this case, the spontaneous curvature depends directly on the mean spacing between surfactant head groups as a function of composition, ¢/.
We first consider a monolayer with a repulsive interaction +J between like head groups, and an attractive interaction, - J between opposite head groups. This suggests an Ising model description for the energy H of a two-component mixture:
H = ~ JS~Sj, ( 2 )
where the sum over (i]) includes only nearest neighbor pairs. The constituents are labeled by i, and S i = +1 (--1) denotes the presence of surfactant (2). Furthermore, the attractive or repulsive
interactions result in a local deformation of the bond distances compared to their values for the pure surfactants (which are assumed to have the same bond lengths). We describe this by a quantity Aij, which is the change in the bond length between surfactants at nearest-neighbor sites i and j. Finally, there is an elastic-restoring force, with spring constant k:
H = ~. S ~ S j - B(1 -- SiSj)A 0 + - ~ A . (3) (ij~
Here, B represents the strength of the coupling between the composition and elastic degrees of freedom. Equation (3) represents the compressible Ising model.
The mean-field value of (A,) is found by minimizing Eq. (3) with respect to (Ais):
(A~j) = B(1 -- (S~Sj))/k ; (4)
the resulting expression for the free energy per surfactant h is
B 2 h = l ( S , - (1 - ( s , sj ) 2 . (5)
In random mixing, the nearest-neighbor correlation function (SiSj) can be found by weighting the two possible values by the appropriate product of independent probabilities for finding surfactants 1 and 2 at each site:
( s i s j ) = (1 - + C - 2 ,(1 -
= ( 1 - - 2 ¢ / ) 2 . (6)
Simple models for the packing of surfactant molecules at a surface yield a spontaneous curvature which depends linearly on the mean spacing between polar head groups. Within the model of the previous section, the change in the spontaneous curvature depends on (Aij), and hence on I s i s j l :
/ / ( 1 - - ( S i S j ) ) = - - (7) - - c ( 0 ) = 7
The parameter ]/is of order a -1, where a is a microscopic length. The precise value of ] /can be obtained, although it is somewhat model specific [11].
Safran et al., Spontaneous vesicle formation by mixed surfactants 5
Considering now the propert ies of a bilayer, and using the definitions of the composi t ion asymmetries discussed above, we arrive at the following expressions for the effective spontaneous curvatures:
(8)
a = (c 1 - c2) - 8 ( 1 - 2 ~ , ) , (10b)
These formulae are writ ten for the general case where the individual spontaneous curvatures are unequal . For the case where q = c 2, the effective spontaneous curvature of the interacting system is reduced (for fl > 0) compared with q . This reduc- tion is just what is necessary to stabilize the vesicle so that the effective spontaneous curvatures of the inner and outer layers are equal and opposite, thus relieving the frustration present in the single surfactant case. For ideally mixed, or non-interacting surfactants (fl = 0), a vesicle composed of a single surfactant has an outer layer which satisfies the spontaneous curvature, but a frustrated inner layer. Interactions be tween the two surfactants, however, can result in a contribution to the spontaneous curvature which is opposi te in sign to both q and c 2. If more of these pairs are placed on the inner layer, one can stabilize the vesicle so that w h e n c = ci --- --c o , the system is at its lowest curvature energy state and the frustration is relieved. This is seen quantitatively from Eqs. (8) and (9) where the choice
¢ = + (e/p) v2 (11)
results in c i = --c 0. Note that this stabilization is only possible if the interaction terms are considered.
With this model , the curvature free energy of Eq. (1) then becomes
fc = 4K[(c - - a(o) 2 + (5(q]) -- fl~a2) 2] . (12)
Thus, the spontaneous curvature of the bilayer is c = a(a. This describes a flat bilayer, unless (p #0. We must now determine the value (o*, which minimizes
the free energy as a function of ¢. When c = a~o, the free energy per surfactant F c is
F c = 2 K a ( c ( u / ) - - fl¢2)2
= 2 K a [ c ( ¢ ) 2 _ 2flc0g)¢2 + fl2¢4], (13)
where a is the area per polar head group. The contribution of the interaction terms of Eq. (5) to the free energy per surfactant is
F ~ = / ( 1 - - 2 V ) 2 - - 8B z
k • ~ ( 1 - ¢ ) :
~p2
k
Similarly, for small values of ¢, the contribution due to the entropy of mixing is
U
F m = kT i ~ , l og ~/ + (1 - - ~) log (1 - - ~u) i_
1 ( 1 -)rp2 •
12 (1 - - ~,)8 •
The total free energy per surfactant can be wri t ten a s
F = F o --~,~p2 + Acp4, (16a)
where
8 c = 4 K a B c ( ¢ ) - - 4J + - - B2(1 - - 6~(1 - - ~))
k
and F 0 is independent of ~a. Equations (15) and (16) are valid in the high "temperature" limit. This corresponds to interaction terms J and B/k, which are small compared with kT. In this limit, e - ( T c - - T )
and B - T, where T c --- Kaf l c (g] ) . Then, a spontaneous vesicle phase characterized by ¢ ~= 0 will occur below a second order phase transition at T = To. This suggests that it will be fruitful to more ful- ly examine the case of low temperatures, or the case of strong interactions between the constituents [11].
III. D i scuss ion
For e < 0, the minimum free energy state is composed of flat bilayers where the two monolyers have identical compositions (~a = c = 0). When e > 0, the free energy is minimized by a non zero value of ~a and hence a non-zero curvature. However, the vesicle phase is limited to a finite region of the phase diagram as a function of the relative composition ¢/, as well as the absolute concentration of amphiphile ~s. This limitation arises from the imposition of packing constraints on the vesicles. This enables an estimate of the phase diagram at fixed values of temperature, ]/, cl, and c 2 as a function of concentration. Neglecting polydispersity, the volume fraction of the system occupied by vesicles is
4~ ¢~ = - - n R 3 , (17)
3
where R = 1/c* is the vesicle radius and n is the number density of vesicles. For large vesicles, the volume fraction of surfactant is
Cs = 8 n n r ~ R 2 . (18)
Eliminating n, we find that 6r3/R = CJcp. The vesicles cannot be overpacked (¢ must be less than one); we take the value of ¢ = 1 as the bound of stability of the vesicles with respect to the lamellar phase where steric constraints are much weaker. An approximation to the phase boundary as a function of (a s (the total volume fraction of surfactant) and g~ (the fraction of surfactant that is type "2") is then given by the locus of points which satisfy
r G = 6~c*(g]) , (19)
where
c* = a(g~)~a*, (20)
where ¢*(¢/) is the value of (a that minimize Eq. (16). A more detailed discussion of the phase diagram can be found in [8, 11].
In summary, we have shown how interactions between surfactants can stabilize a phase of spherical vesicles with respect to a fiat lamellar phase. These interactions require that the effective spontaneous curvature of the film have a term quadratic in the composition. The physical origin of this stabilization is the tendency of "1--2" surfactant pairs to have a different bond distance from the average of "1--1" and "2--2" pairs. It is then possible for the effective spontaneous curvature of a film composed mostly of "1-2" pairs to be quite different (even in sign) from the spontaneous curvature of the pure films. In the case where the curvature energy dominates, the vesicle is then stable; the outer layer, for example, may consist mostly of "1--1" pairs and the inner layer of the vesicle may be mostly "1--2". The concentration asymmetry of the two layers is such that the effective spontaneous curvatures of the inner and outer layer are equal and opposite; the frustration of one of the layers that destabilizes vesicles composed of a single surfactant is thus prevented.
Even within the context of this model, several outstanding issues remain. The first is to explore the interactions and mixing effects more generally for both the strong and weak interaction case [11]. In addition, the case of mixed amphiphiles of long and short chains should be studied. Finally, the microscopic interactions which determine the different head spacings in ionic systems should be ex- plored so that the interaction parameters fl can be related to charge and salinity.
Acknowledgements
The authors acknowledge useful discussions with J. Israelachvili, E. Kaler, D. Lichtenberg, Y. Talmon, and J. Zasadzinski. The support of the US-Israel Binational Science Foundation under grant no. 87-00338 is acknowledged. D. Andelman is grateful for the support of the Bat Sheva de Rothschild Foundation.
References
1. Kaler EW, Murthy AK, Rodriguez BE, Zasadzinski JAN (1989) Science 245:1371
Safran et al., Spontaneous vesicle formation by mixed surfactants
2. Carnie S, Israelachvili JN, Pailthorpe BA (1979) Biochirn et Biophys Acta 554:340
3. Gabriel NE, Roberts MF (1984) Biochemistry 23:4011; Hargreaves WR, Deamer DW (1978) Biochemistry 17:3759
4. Miller DD, Bellare JR, Kaneko T, Evans DF (1988) Langmuir 4:1363 and J Phys Chem, in press
5. Jain MK, de Haas GH (1981) Biochim et Biophys 642:203; Alrnog S, Kushnir T, Nir S, Lichtenberg D (1986) Biochemistry 25:6597
6. Helfrich W (1973) Z Naturforsch 28a:693 and in J de Phys (Paris) 47:321 (1986)
7. Safran SA, Pincus P, Andelman D (1990) Science 248:354
8. Safran SA, Pincus P, Andelman D, MacKintosh FC (1990) Phys Rev A 43:107 (1991)
9. Israelachvili JN, Mitchell DJ, Ninham BW (1972) Trans Far Soc II 72:1525
10. Israelachvili J, Mitchell DJ, Ninham BW (1977) Biochim et Biophys Acta 470:185
11. A unified theory which accounts for both the curvature energy and the in-plane interactions is given in E C. MacKintosh, S. A. Safran, P. Pincus, to be published
12. Wang ZG, to be published 13. Kaler EW, unpublished
Authors' address:
Dr. S. A. Safran Department of Polymer Research Weizmann Institute Rehovot 76100, Israel
Electrolyte dependent phase separation in aqueous mixtures of a polyelectrolyte and an ionic surfactant
K. Thalberg, B. Lindman*), and G. Karlstr6m 1)
Physical Chemistry 1 and 1) Theoretical Chemistry, Chemical Center, Lund University, Lund, Sweden
Abstract: A mixture of a polyelectrolyte and an oppositely charged ionic surfactant generally phase separates from an aqueous solution due to the strong attractive interaction between the two solutes. Under certain conditions the concentrated phase is a transparent gel. Redissolution can be achieved by electrolyte addition or by a high surfactant concentration. Over a wide range of electrolyte concentrations, there is no phase separation. However, at high electrolyte concentrations, separation into two isotropic phases occurs. While phase separation at low electrolyte contents results in one dilute solution and one phase concentrated in polymer and surfactant, phase separation at high electrolyte concentrations is of a different nature and results in one solution rich in sttrfactant and one rich in polymer. The phenomenon is related to, but different from that displayed by two polymers in a common solvent; called "polymer incompatibility", and can be referred to the elimination of electrostatic interactions. The phase diagrams can be modelled in Flory-Huggins type calculations with reasonable assumptions of the intermolecular interactions.
Key words: Cationic surfactant; polyanion; polyelectrolyte; phase separation; phase _behavior; coacervate
Polymer-surfactant interactions are important in biology and in several applications, such as emulsions, pharmaceuticals, cosmetics, detergents, paints, thickeners and foods. Polymer-surfactant systems have been extensively studied with respect to the binding of surfactants to polymers for relatively dilute systems [1]. A range of physico- chemical parameters has been used to map binding in terms of concentration of onset of binding (the ciritical aggregation concentration, CAC) and of saturation of binding. For many systems, binding isotherms have been obtained, notably by the use of surfactant-selective electrodes [2]. On the other hand, studies of more concentrated systems as well as of phase behavior are sparse, although it is well recognized that separation into two or more phases may easily occur and be of significant biological and technical interest. As regards structure of the systems, it can be inferred indirectly that surfactant molecules self-assemble to micelle-like clusters along the polymer chain, but there is also direct evidence for this type of structure, notably from
neutron scattering [3], but also from fluorescence quenching studies [4].
In systems containing a charged polymer (i.e., a polyelectrolyte) and an oppositely charged surfactant, the interactions are considerably reinforced, as is primarily seen in the low value of the CAC relative to the critical micelle concentration (CMC) of the surfactant. Also, for solutions of an ionic polymer and an oppositely charged ionic surfactant, phase separation is commonly observed over a wide range of mixing ratios. A conspicuous, but general feature is the redissolution and formation of a single isotropic solution phase at higher surfactant concentrations. Redissolution can also be ef- fected by electrolyte addition, which can be ac- counted for by the screening of the attractive polyion -- surfactant ion interaction.
We have studied the phase behavior for polyelectrolyte -- ionic surfactant systems over a wide range of electrolyte concentrations and report a further type of phase separation, which is distinct from the one which is found in the absence of electrolyte
Thalberg et al., Electrolyte dependence of polyelectrolyte - - surfactant systems 9
or at low electrolyte concentrations. The phenomenon is compared with other types of phase separation in colloidal systems and attempts have been made to model the behavior in theoretical calculations of phase diagrams applying the Flory- Huggins theory of polymer solutions for a three- component system of solvent, polymer, and cosolute.
Phase diagrams are presented here with hyaluronan, which appears to show a typical polyelectrolyte behavior, and cationic surfactants of the alkyltrimethylammonium type. Hyaluronan (or hyaluronic acid, here abbreviated Hy) is a linear anionic polysaccharide, built of alternating units of glucuronic acid and N-acetylglucosamine [5]. It plays an important role for the physico-chemical properties of the extracellular matrix [6], and is found in all mammals.
Samples containing polymer, surfactant, and water were thoroughly mixed and equilibrated. Under certain conditions, there is separation into one low-viscous supernatant phase, which is dilute in polymer and surfactant, and one concentrated, often gel-like bottom phase. The two phases were analyzed with respect to all the components and from the composition of the phases the phase diagram was traced. Systems of polyelectrolyte, ionic surfactant, and solvent must strictly, in thermodynamic considerations, be treated as four-component systems. However, an adequate representation for many purposes will, for an isothermal case, be in terms of a two-dimensional representation in a triangular diagram.
The phase diagram for the system hyaluronan- tetradecyltrimethyl ammonium bromide (C14TAB) - water is shown in Fig. 1. The two-phase region is located close to the water corner, and has a droplet- like shape. The supernatants are located at the left boundary of this region, and the concentrated phase to the right and upper sides of the region. The size of the two-phase region decreases when a surfactant analogue of shorter chain length is used, but its shape and location are largely retained. This phase behavior applies also to other systems of polyelectrolyte and oppositely charged surfactant, and seems to be of a wide generality.
We have been able to model the observed phase behavior in calculations based on the Flory-Hug- gins theory for polymer solutions [7]. Calculations for systems of polymer, solvent and cosolute show that it is impossible to obtain anything near the observed behavior without assuming the cosolute
Surfactant
7i
% Polyel.
Fig. 1. Phase diagram for the system hyaluronate- tetradecyltrimethylammonium bromide-water [8]. The compositions of some samples are indicated. Open circles refer to initial sample compositions, and filled circles connected by tie lines refer to the composition of the two phases in equilibrium. The dashed part of the phase boundary indicates larger uncertainty in this region
Polymer B /
/ -/z S.d
Water ~ - - - P o l y m e r A 10 20 30 40 % Polymer A
Fig. 2. Theoretically calculated phase diagram for a system of two polymers in a common solvent (water) [8]. Polymer A represents the polyelectrolyte and polymer B represents the surfactant. Index I refers to the solvent, index 2 to polymer A and index 3 to polymer B. The interaction parameters used are w12 = i200 J/mol, w23 = --5200 J/mol and the polymerization numbers are 300 for polymer A and 25 for polymer B
to have a high molecular weight, which is in agreement with our general view of a cooperative binding of surfactant to a polymer and the formation
of micelle-like clusters. The system is, therefore, treated as a system of solvent and two polymers, A and B, representing the polyelectrolyte and the surfactant respectively, and an effective interaction parameter wij is introduced between each pair of species in the system. (wij is related to the normal Flory interaction parameter by wij = R TXij). In addition to the interaction parameters, two other parameters have a major influence on the phase behavior, namely the polymerization numbers for the two polymers. Phase diagram calculations are performed by minimizing the total Helmholtz free energy of the system with respect to the composition in the different phases. (For a full description of the model and its limitations, see [8].)
By a proper choice of the five parameters, a theoretically calculated phase diagram is obtained (Fig. 2), which shows a reasonable agreement with the experimental one concerning the shape and the location of the two-phase region, as well as the slope of the tie-lines. The model is also able to account for the changes in phase behavior observed when the surfactant chain length is varied [9]. The driving force behind the phase separation is the favorable interaction between the polyelectrolyte and the surfactant molecules, relative to the interactions of these species with the solvent (water). A phase concentrated in both these components, which enables a higher degree of contact between polyelectrolyte and surfactant micelles, is then favored, in spite of the loss in entropy for redis- tributing the components of the system.
The phase separation seen in systems of a polyelectrolyte and an oppositely charged surfactant thus seems to be of the same origin as phase separation between two oppositely charged polymers or other colloidal species. Such systems were thoroughly investigated by Bungenberg de Jong in the 1930s, and he was able to qualitatively explain the observed phase behavior [10]. The resulting concentrated phase was called a "coacervate" and the phase separation was referred to as "complex coacervation'.
In Fig. 3, the phase diagram at addition of 75 mM of NaBr to the system is shown. The area of the two- phase region is reduced while its location in the phase diagram and the slopes of the tie-lines are rather unaffected. If we add more salt, the two- phase region shrinks and will finally vanish completely. Investigations show that this has occurred at 250 mM of added NaBr in the present system.
Very striking is that, at addition of a rather large salt concentration (i>500 mM of NaBr for this
Surfactant /
Poly- electrolyte
Fig. 3. Phase diagram for the system presented in Fig. 1 in the presence of 75 mM of NaBr. Symbols as in Fig. 1. The contour line of the two-phase region in the absence of salt is indicated
Water
Surfactant /
% Polyel. ,.
Fig. 4. Experimental phase diagram for the system presented in Fig. 1, in the presence of 1.0 M of NaBr. Symbols as in Fig. 1
system), two-phase separation reappears, but now the phase behavior is of an entirely different type. The two phases in equilibrium are again clear and isotropic, but now the supernatant phase is enriched in surfactant while the bottom phase is enriched in polyelectrolyte. A phase diagram is shown in Fig. 4, corresponding to a salt concentration of 1.0 M NaBr. The two-phase region is still located close to the water-surfactant axis of the phase diagram, but is not "anchored" at the water corner. Furthermore, the tie-lines now have a different
Thalberg et al., Electrolyte dependence of polyelectrolyte - - surfactant systems 11
direction. At a high salt concentration, a separation into one polyelectrolyte-rich and one surfactant-rich phase is the case, while the driving force for the phase behavior in the absence of salt (Fig. 1), is the attraction between these two species. It is, therefore, evident that we here deal with a totally different phase separation mechanism.
For systems of two nonionic polymers in a common solvent, a familiar phenomenon is the polymer incompatibility, leading to the separation of the polymers into two different solution phases (Fig. 5) [11]. Apparently, a behavior related to this can be expected for two oppositely charged colloids or polymers when the electrostatic attraction is eliminated. Although the phenomenon reported here to some extent can be considered as related to the common two-phase separation of polymer solutions, it can also from the phase diagram be seen to be distinct from that.
polymers A and B, representing the polyelectrolyte and the surfactant, respectively, remain unchang- ed, and, therefore, also the interaction parameter between these two species.
Addition of salt is known to facilitate the formation and growth of ionic micelles in surfactant- water systems [12]. In the model this corresponds to an increase in the polymerization number of polymer B. (Besides, the interaction parameter between water and polymer B has been slightly disfavored). Furthermore, the interaction between the polyelectrolyte and the salt-containing water will be more favored when the salt concentration is increased, due to the electrostatic interactions between the polyion and the salt ions. (This translates into a more favorable interaction between polymer A and the solvent in the model.) By appropriate changes in the interaction parameters according to the above reasoning, the phase behavior can be modelled (Fig. 6).
Water
Fig. 5. Phase diagram for the system PEG 6000 -- Dextran D17 -- water at 20°C. Experimental points (triangles) and tie lines (.. • ), and theoretical phase separation curve and tie line (full lines) calculated with the Flory-Huggins theory. See reference [11] for further information
Polymer B Poly (ethylene glycol) uu60 ~ /
j o o , o,o / \
/ / ~ ~ . ' . . " . ~ : ' - , "~.... \ Water t , , , ) - - - P o l y m e r A 10 20 30 40
lo is 2o 25 Dextran % Polymer A .
% Dextran
Fig. 6. Theoretically calculated phase diagram for a system of two polymers in a common solvent (water). In- dices as in Fig. 2. The interaction parameters used are W12 = --7200 J/mol, W13 = 1200 J/mol and W23 = --5200 J/mol and the polymerization numbers used are 300 for polymer A and 100 for polymer B
The influence of salt has also been modelled as described above. In order to circumvent the problem of the fourth component, we have chosen in the calculations to incorporate the added salt into the water component. In this way, the two
We can thus qualitatively rationalize the phase behavior of the system also at high salt concentrations. The dominating factor is that the polyelectrolyte prefers the salt-containing water to the surfactant. This, in combination with the poor inter-
action between water and the surfactant and the increase in size of the micellar aggregates, which allows the surfactant to separate out without too much loss in entropy, is the physical explanation to this new phase separation mechanism it is indicated in the model calculations that this phase behavior is very delicately balanced, with respect to both differences in the interactions between the three components and to the aggregation number of the micelles, as modelled by the polymerization number of polymer B. Especially, we note that for a large range of parameter values between those used in obtaining Figs. 2 and 6, phase diagrams without a two-phase region are calculated.
These results illustrate a rich new area of phase behavior in surfactant-polymer-solvent systems. Several features have been documented for systems other than that of hyaluronan and tetradecyltrimethylammonium bromide (mainly considered in this report), suggesting a considerable generality of the results. The fact that it can be successfully reproduced by a quite simple and general model (which does not depend on assumptions about the structure in the systems) supports this view, as well as provides a picture of the molecular interactions underlying the phase separation phenomena.
3. Cabane B, Duplessix R (1982) J Physique 43:1529; Colloids Surf (1985) 13:19
4. Turro NJ, Baretz BH, Kuo P-L (1984) Macromolecules 17:1321; Abuin EB, Scaiano JC (1984) J Am Chem Soc 106:6274; Chu D, Thomas JK (1986) J Am Chem Soc 108:6270
5. Hyaluronan was provided by Pharmacia AB, Upp- sala, Sweden in the form of sodium salt (i.e. sodium hyaluronate). It was of a highly purified quality, containing no appreciable amounts of protein or other impurities. The molecular weight of the Hy preparation used in this work is about 250000
6. Comper WD, Laurent TC (1978) Physiol Rev 58(1):255; Laurent TC (1987) A_cta Oto-Laryngol, Suppl 442:7
Z Flory PJ (1953) Principles of Polymer Chemistry; Cor- nell University Press: Ithaca, NY
8. Thalberg K, Lindman B, Karlstr6m G (1990) J Phys Chem 94:4289
9. Thalberg K, Lindman B, Karlstr6m G, J Phys Chem, in press
10. Bungenberg de Jong HG (1949) In: Colloid Science, vol II, Ed: Kruyt HR, Elsevier, Amsterdam; Chapter 10, p 259
11. Guastafsson .h,, Wennerstrtim H, Tjerneld F (1986) Polymer 27:1768
12. Lindman B, Wennerstr6m H (1980) Top Curr Chern 87:1
References
1. Goddard ED (1986) Colloids Surf 19:255/301; Hayakawa K, Kwak JCT (1991) In: Rubingh D, Holland PM (eds) Surf Sci Ser, Marcel Dekker, New York, chap 5
2. Hayakawa K, Kwak JCT (1982) J Phys Chem 86:3866; Hayakawa K, Santerre JP, Kwak JCT (1983) Macromolecules 16:1642, Malovikova A, Hayakawa K, Kwak JCT (1984) J Phys Chem 88:1930
Authors' address:
Prof. Dr. B. Lindman Physical Chemistry 1 Chemical Center University of Lund Box 124 S-22100 Lund, Sweden
Note added in proof:
We have recently noted an analytical error in the chemical analyses of the separating phases at high salt concentrations. Therefore, the two-phase region in Fig. 4
is underestimated and the phase behavior comes even closer to that of two nonionic polymers in a common solvent. The theoretical model accounts for the behavior with a lower w u than given in Fig. 6. For a full account of the phase behavior see ref. 9.
Sodium dodecylsulfate-poly(ethyleneoxide) Interactions studied by time-resolved fluorescence quenching
J. van Stam, M. Almgren, and C. Lindblad
Department of Physical Chemistry, Uppsala University, Uppsala, Sweden
Abstract: The interaction between sodium dodecylsulfate (SDS) and poly(ethyleneoxide) (PEO) has been studied by time-resolved fluorescence quenching at 20°C and 40°C in the dilute regime, i.e., 0.2% w/v, and in the semi-dilute regime, i.e., 2% w/v, with respect to PEO. Lifetime measurements show that PEO wraps around the micelle-like cluster formed by SDS upon interaction with PEO -- the polymer shields the probe, pyrene, from quenching by bulk water solubilized oxygen. The aggregation number determined at the SDS-concentration when interaction starts, CAC, is much lower than predicted by current theory. As the surfactant concentration is increased, the aggregation number is simultaneously increased in the dilute regime, but remains constant at low additions of SDS in the semi-dilute regime. This indicates a certain number of locations for the clusters on the polymer chain. At CAC clusters are formed at these locations. Added surfactant is consumed by forming new clusters until all locations are filled in competition with the growth of excisting clusters.
Key words: Fluorescence quenching; sodium dodecylsulfate; poly(ethyleneoxide); "_interactions; aggregation numbers
Introduction
Polymer-surfactant systems are of great interest, both from a fundamental point of view and for applications in a variety of industrial fields, e.g., enhanced oil recovery, paint, and medicine. Surfac- tants and polymers may interact with each other in a way where the surfactant forms micellar-like aggregates (in the following referred to as clusters to distinguish them from ordinary micelles) in contact with or in the vicinity of the polymer. In the case of ionic surfactants and a polyelectrolyte of opposite charge, electrostatic effects are of prime importance. Interactions between an ionic surfactant and an uncharged polymer, on the other hand, cannot be explained in this way, and theories have been suggested for this type of systems [1, 2]. For many reasons the systems of the anionic surfactant sodium dodecylsulfate (SDS) and the uncharged poly(ethyleneoxide) (PEO, in the literature some- times referred to as poly(ethyleneglycol)) have been used as model systems in the study of these interac-
tions. Many studies have appeared, both with classical methods [3, 4] and with new ones, such as small-angle neutron scattering, SANS, [5] time- resolved fluorescence quenching [6] and fluorescence quenching [7]. Some good reviews in this field have also been published, e.g., [8]. The over-all picture for interaction between an anionic surfactant and a neutral polymer is that the interaction starts at a critical aggregation concentration (CAC) which is lower than the ordinary critical micellization concentration (CMC) for pure surfactant solution. Moreover, it is also found that the aggregation numbers for the polymer-interacting dusters are lower than for ordinary micelles.
Even if results and suggested theories point in the same general direction, they do not agree in the detailed picture of the systems. For this reason it is of interest to continue studies of the SDS-PEO system, aiming at a detailed description and understanding of the interaction.
In this study, we present results for the natural fluorescence lifetime of an excited probe, pyrene,
the aggregation numbers for the clusters obtained from time-resolved fluorescence quenching measurements with dimethyl benzophenone as quencher, and the III/I vibronic peak ratio from pyrene steady-state fluorescence spectra, giving information of the micropolarity around the pyrene molecule for solutions without polymer, with 0.2% w/v PEO, i.e., a dilute solution, and 2% w/v, i.e., a semi-dilute solution.
Materials and methods
Poly(ethyleneoxide) (PEO) was purchased from Fluka (molecular weight 35000) and was used as supplied. Pyrene (Aldrich) and dimethylbenzophenone (DMBP) (Aldrich 99%) was twice recrystallized from ethanol. Sodium dodecylsulfate (SDS) was from BDH, specially pure. As conductivity measurements gave CMC in accordance with literature values (= 8 mM), the surfactant was used without further purification. All solutions were prepared with distilled water. For the deoxygenized experiments pure nitrogen was used to remove oxygen from the solutions prior to measurement. Dilute PEO solutions were 0.2% w/v and semi- dilute solutions were 2% w/v.
The preparation of samples for fluorescence measurements was described earlier [9]. To allow pyrene to dissolve completely in the micellar phase the solutions was stirred for at least 12 h. The pyrene concentration was kept low enough (< 10 -5 M, or less than one pyrene molecule per 50 clusters or miceUes) to prevent excimer formation. The DMBP concentrations were chosen to be less than one DMBP molecule per cluster or micelle.
Static fluorescence measurements were carried out on a SPEX Fluorolog 1680 combined with a SPEX Spectroscopy Laboratory Coordinator DMIB.
Time-resolved fluorescence decay data were collected with the single photon counting technique, as described earlier [10]. The set-up uses a mode- locked Nd-YAG laser (Spectra Physics, Model 3800) to synchronously pump a cavity-dumped dye laser (Spectra Physics Models 375, 344S) for the excitation, using DCM as dye, and a KDP crystal for fre- quency doubling. The excitation wavelength was 320 nm and the pyrene monomer emission was measured at 395 nm. The pulse width was less than 1.5 ns, which can be treated as a 0-pulse compared to the lifetime of pyrene in our experiments, about 150 ns. The excitation rate was low enough to prevent multi-photon excitation. The temperature was
held constant by thermostatting the cuvettes and the cuvette holder by the same standard water- bath. The measurements were performed at two temperatures, 20°C and 40°C.
All data were analyzed on a Digital Equipment VAXstation 2000 with the same method as described earlier [11].
Conductivity measurements were used to determine the CMC or CAC of each sample differing in either polymer concentration or temperature. For these measurements a standard platinum conductivity probe connected to a Philips PW 9505 conductivity meter was used. All solutions were ther- mostatted in a water bath and stirred to allow equilibrium conditions.
The method of time-resolved fluorescence quenching in microheterogeneous solutions is well described in the literature [12, 13]. Under the conditions that the excitation pulse is narrow compared to the fluorescence lifetime and that both probe and quencher molecules are stationary in their host micelles during the time window measured, the interpretation with the well-known Infelta model [14] is straight-forward.
In the Infelta equation,
F(t) = A l e x p [ - - A 2 t + A3(exp( - -A4t ) - - 1)] , (1)
and the parameters have the following meaning under the circumstances stated above:
A 1 is the fluorescence intensity at time t = 0, i.e., F(0). This has no physical meaning, but is only dependent on the time one allows each measurement to take.
A 2 is the decay rate at long time, i.e., when the decay shows an exponential tail.
A 3 is the average number of quencher molecules per micelle. If one knows the amount of bound surfactant molecules and the distribution of quenchers between micelles and the bulk phase, the aggregation number, (a), can be calculated from
A 3 x [surfactant]bo~ d (a) = ; (2 )
[quencherlbound
A 4 is the first-order quenching rate constant L. If it is assumed to be inversely proportional, roug~aly, to the hydrophobic volume of the micelle, the values can be used to check the reliability of the estimated aggregation numbers.
The natural lifetimes r 0 were determined in separate experiments without quencher, and the
1 a difference between 1/A 2 and r 0 is a measure of the condition that the probe and quencher are stationary during the time window of the quenching experiments. If this difference is close to zero, the quencher does not migrate between micelles, or micelles and the bulk phase. The probe, pyrene, is certainly stationary under our conditions, but the quencher, dimethylbenzophenone, have some solubility in the water bulk phase. It turned out, however, that also the quencher was stationary under the conditions of the investigated systems. The water-solubility of the quencher turned out to be significant only in solutions with a surfactant concentration close to CMC or its analogue in polymer-surfactant solutions, CAC. In all solutions the quencher concentrations were corrected for this solubility.
It should be noted here that the solutions in general were not deoxygenated, which means that the fluorescence decay was quenched by oxygen. However, this only affects the natural lifetime r 0 and does not influence the use of the Infelta model.
Results and discussion
The effect of PEO on the pyrene fluorescence quenching in SDS micelles is immediately seen in Fig. 1. Figure la shows the set of fluorescence quenching curves when PEO is absent, and Fig. lb when it is present. As all other parameters, i.e., SDS and DMBP concentrations, were the same, the decreased quenching in Fig. lb shows that the aggregation number is much lower when PEO is present.
For the quantitative discussion, the results can be divided into three parts: lifetime measurements, aggregation numbers, and III/I vibronic peak ratio from steady-state pyrene fluorescence spectra.
l b
4
2
0 1 O0 200 300 400 500 600 t ime / ns
4
0
I I I I I
0 100 200 300 400 500 600 t ime / ns
Fig. 1. Time-resolved fluorescence quenching measurements at 20°C. Both figures are 22.9 mM with respect to SDS and have the same DMBP concentrations. The upper curve is 0% and the lower curve 2% with respect to PEO, respectively. The DMBP concentration in mM is from above: 0, 0.18, 0.27, and 0.36
Lifetime measurements
In aerated solutions, the presence of PEO increases the lifetime of cluster-solubilized pyrene compared to pyrene in ordinary miceUes (Fig. 2 and Table 1). This increase is more enhanced in the semi-dilute solutions. Comparing the results from two temperatures, the same behavior is found, but is more pronounced at the lower temperature. In deoxygenated samples no difference in lifetime between the different systems is observed, as is also in
Fig. 2 and Table 1. Clearly, the polymer shields pyrene in the micellar aggregates from quenching by oxygen. Normal micellar solutions offer little protection in this respect. Assuming the oxygen concentration in air-saturated water at 20 °C to be 5.7 • 10 ~ M [15] the lifetimes in Table 1 for pyrene in SDS without PEO give a second-order quenching constant of 4.7 • 109 M -1 s -1, which is reduced to 2.9 • 109 M -1 s -1 in 2% PEO. The former value is as expected for a diffusion-controlled process in a homogeneous aqueous solution. The protective action can be understood if one imagines that the
th r =
[ S D S ] / raM
Fig. 2. Lifetime of pyrene. The symbols denote: circles = 0% PEO, squares = 0.2% PEO, triangles = 2% PEO, open symbols = 20°C, and filled symbols -- 40°C. The two upper curves refer to deoxygenated samples, squares with diagonal line = 0.2% PEO, 20°C; squares with cross mark = 0.2% PEO, 40°C. Inserted line shows lifetime in deoxygenated ordinary micelles
polymer wraps around the cluster in a rather com- pact layer, as has been suggested by others [5]. In this way the polymer will replace water molecules from the interface between the cluster and the bulk, and decrease the possibility for oxygen to quench the fluorescence.
It could be argued that oxygen contained in the clusters, and not molecules approaching from the bulk solution, were responsible for the quenching. There are two strong arguments against such a mechanism: i) the fluorescence decay is single-exponential, and ii) even if the solubility of oxygen in the micelles was 10 times that in water, very few of the aggregates would contain an oxygen molecule.
The fluorescence lifetime decreases with increasing surfactant/polymer concentration ratio. This is already seen at low concentrations of SDS in the dilute regime, but also at the highest concentrations of SDS in the semi-dilute regime. The protective action of the polymer starts to decrease before it is saturated and free micelles form.
Aggregation numbers
In the dilute polymer solutions, it is found that the aggregation number at CAC is about 20 (see Fig. 3a and Table 2) at both temperatures. The ag-
Table 1. Natural lifetimes, r 0, for the systems investigated. All lifetimes are given in ns-units
I. Air-saturated samples [SDS] 0% PEO 0.2% PEO 2% PEO mM
20°C 4 0 ° C 2 0 ° C 4 0 ° C 2 0 ° C 40°C
9.0 10.2 15.3 17.5 20.0 20.4 25.0 25.5 30.6 40.8 60.0
100.0 200.0 300.0 4O0.0
217.2 169.1 179.2 129.0 217.2 165.3 181.6 131.2 220.9 166.3
210.0 160.0 204.7 158.2
180.7 131.0 219.9 163.4 202.9 153.9
182.3 134.7 216.6 162.1 179.0 138.6 219.0 163.8 185.6 134.2 214.9 157.2
192.4 147.9 218.1 164.0 201.2 150.0 194.3 150.2 189.6 153.6
[SDS] mM
20°C 20°C 40°C
9.0 338.1 330.5 17.5 345.0 347.9 327.0 60.0 312.7 342.8
gregation numbers increase gradually upon increasing the SDS concentration, up to a limiting value. This value, approximately 60, is equal to that found in ordinary SDS micelles at this concentration.
The same behavior is seen in the semi-dilute solutions (see Fig. 3b and Table 2), but shifted to much higher surfactant concentrations; the value found in ordinary micelles at this concentration, approximately 100 [16, 17], is not reached until the SDS- concentration is as high as 300 raM. The only temperature studied in the semi-dilute solutions was 20°C, but reference measurements at 40°C show the same pattern.
A source of uncertainty in the interpretation in this case is the relatively high CMC of SDS. It is known from NMR measurements [18] that the amount of free surfactant decreases in aqueous solutions when the surfactant concentration is increased -- thus an uncertainty is introduced by set-
van Stare et al., Sodium dodecytsulfate-poly(ethyleneoxide) interactions 17
10
0 6O [SDS] / mM
0 400
Fig. 3. a. Aggregation numbers vs SDS concentration in 0.2% PEO. Open circles refer to 20°C and filled circles to 40°C, respectively. Dashed lines show results from model simulations. Inserted line shows aggregation number for a free micelle in this concentration range, b. Aggregation numbers vs SDS concentration in 2% PEO and at 20°C. Inserted line shows aggregation number for a free miceUe at the upper part of the concentration range
Table 2. Aggregation numbers, quenching rate constants and polydispersity indexes from fluorescence quenching measurements. For the dilute solutions, i.e., 0.2% PEO, the limiting duster aggregation number is 56 and 42 at 20°C and 40°C, respectively, from model simulations. Or- dinary SDS micelles have an aggregation number of about 60 in the concentration range in the dilute solution, and of about 100 in the upper part of the concentration range in semi-dilute solution. Ordinary STS micelles have an aggegation number of about 90 for the STS micelles in the concentration range investigated
I. SDS and 0.2% PEO [SDS] (a)w kq " 10 -7 S -1 G/~a)w mM
20°C 40°C 20°C 40°C 20°C 40°C
9.0 26.4 20.3 4.5 10.0 0.76 0.96 12.0 30.1 22.9 4.3 9.0 0.35 0.19 17.5 41.2 39.2 3.7 8.1 0.37 0.37 20.0 49.2 41.4 3.6 7.4 0.37 0.22 25.0 56.1 49.2 3.3 7.2 0.44 0.49 60.0 62.1 54.7 3.4 7.3 0.40 0.39
II. SDS and 2% PEO [SDS] la~w kq • 10 -7 s -1 a/la)w m M
20°C 40°C 20°C 20°C
15.3 24.5 3.8 0.93 22.9 20.6 5.1 1.32 51.0 26.0 3.9 0.32
100.0 58.3 55.4 3.2 0.51 200.0 83.4 77.5 2.8 0.43 300.0 99.0 92.4 2.6 0.39 400.0 103.8 97.5 2.5 0.37
III. (a)w for STS at 40°C [STS]/mM 0.2% PEO 2.0% PEO
9.0 54.8 28.7 25.0 89.4 33.9
ting the concentrat ion of free monomers equal to the CMC. This mus t play a great role w h e n the SDS concentrat ion is close to CMC, and there is no data on ho w the amoun t of free surfactant behaves w h e n PEO is present . To overcome this problem, some reference measurements were per fo rmed with the more hydrophobic surfactant sod ium tetradecyl- sulfate (STS), for which the two additional methylene groups have reduced the CMC to about 2 raM, and the CAC to about 1.5 mM at 40°C; at 20°C, STS is insoluble. The pat tern is just the same
as in the case of SDS - - the aggregation n u mb er increases already cont inuously at very low concentra- t ions in the dilute solut ion and was constant in the semi-dilute (Table 2).
At CAC, where only clusters and no micelles are present , the aggregation numbers were about 1/3 of those of the ordinary miceUes. This is lower than predicted by the theory of Nagarajan [1], whose model gives aggregation numbers in the range
45--55 for the SDS-PEO system, but in the same range as those found by Zana et al. [6] in a time- resolved study of pyrene excimer formation.
At higher concentration of surfactant, one must keep in mind that the method measures an average over the whole system. This means that if both small clusters interacting with the polymer and big- ger ordinary micelles are present, one gets an average aggregation number, approximately weighted by the hydrophobic volume of the two states. In this case, when the aggregation number for the clusters at CAC is about 20 and that for the ordinary micelles about 60, equal weights for the two states is reached when the number ratio between polymer-interacting clusters and ordinary bulk micelles are 3:1. On the other hand, it is possible to use this feature to understand the aggregation behavior. In the dilute solutions, a number of small
c 20.0 clusters are formed at CAC; this number is equal to .~ or just below the maximum number of clusters in ,- o the system. The main effect of increasing the surfac- ~ 15.0 tant concentration is not to form more clusters, but E to increase their aggregation number up to a limiting value, lower than the aggregation number o. 10.0 for ordinary micelles. When the clusters reach the limiting size, ordinary micelles are formed. In the semi-dilute solutions, also small clusters are formed ~ 5.0 at CAC, having the same size as in the dilute solu-
O} tions. But, added surfactants are first consumed by < forming more clusters; as the amount of polymer is 0.0 10 times higher, the maximum number of clusters is also 10 times higher, and the growth occurs much slower than in the dilute solution. This rough model is tested for the systems investigated by = 8.0 simulation and gives excellent concordance with "~
'- 7.0 the experimental points in the dilute solutions o (dashed lines in Fig. 3a) at both temperatures. In ~ 6.0 the semi-dilute solutions the model cannot E ->" 5.0 reproduce the experiments, as the model tested o Q.
4.0 does not take into account the possibility of forming more clusters, o. 3.0
There is also a temperature dependency in the ag- 2.0 gregation numbers. In the dilute solutions the P
model gave the limiting sizes for the cluster ag- m 1.0 gregation number to be 56 and 42 at 20 °C and 40 °C, < 0.0 respectively. This explains why the average aggregation number is always lower at the higher temperature at corresponding surfactant concentrations. The lower cluster size limit at higher temperature can be explained by the increased hydrophobicity of PEO at elevated temperatures. The polymer will then shrink and be less flexible,
avoiding water contact. Thereby, the interaction between the polymer and the ionic cluster is also restricted to lower cluster aggregation numbers.
The number of aggregates per polymer chain is calculated from A3-values in Eq. (1) and the quencher concentrations. According to the model, this value should be constant up to the point where they reach their maximum aggregation number. This is also indicated in the dilute solution; in Fig. 4a, this is most pronounced at 40 °C, but not seen at all in the semi-dilute solution (Fig. 4b). Evidently, growth of existing clusters and formation of new clusters occur simultaneously. This differs from the picture suggested by Cabane [5], as our interpretation does
0 60
4 a
[SDS] / mM
4 b
0 400 [SDS] / mM
Fig. 4. a. Number of aggregates per polymer chain at 0.2% PEO. Open circles refer to 20°C and filled circles to 40°C, respectively, b. Number of aggregates per polymer chain at 2% PEO and 20°C
van Stare et at., Sodium dodecylsutfate-poty(ethyleneoxide) interactions 19
not predict a stoichiometric composition of the cluster. Instead, our model is more like that suggested by Winnik and Winnik for the systems SDS- hydroxypropyl cellulose [7].
In a system consisting of two sets of aggregates of different sizes, the polydispersity would be of interest. From time-resolved fluorescence quenching measurements it is possible to calculate the polydispersity [13, 19] (Table 2). Due to the very small difference between the aggregation number for the free micelle and the limiting aggregation number for the cluster, no clear indication of a broader size distribution is seen in that concentration range. In- stead the data in Table 2 indicates a higher polydispersity at very low surfactant concentrations in the semi-dilute solution. The competition between cluster growth and cluster formation causes a very broad polydispersity in this region. When increasing the surfactant concentration, this polydispersity is first decreased and then increased again, which may be due to the fact that two different size sets are present in the system.
The values of the quenching rate constants, also collected in Table 2, would be expected to decrease with the increasing micellar size. This is also found, but not to the extent expected. Zana et al. [6] similarly found the excimer formation rate constant k¢ to be decreased much less than expected. Fur- thermore, in an investigation of the system CnTAB- hyaluronan [9], with n = 10, 12, a similar slow change of the quenching rate constant with the duster aggregation number was found. The interaction with polymer seems to slow down the quenching process in these systems.
Static fluorescence
The temperature also affects the III/I-values, the ratio between the third and the first vibronic peak in the pyrene steady-state emission spectrum. This value decreases when the surrounding environ- ment experienced by the pyrene molecule becomes more polar. Typical values in water are about 0.6, and in a SDS micelle, about 0.9--1.0 [20].
Comparing the aqueous solution, the dilute solution, and the semi-dilute solution at the two temperatures (Fig. 5), it can be concluded that the differences in III/I are greater at 20°C. For the aqueous SDS solution, the values are only slightly higher at 40 °C, but the difference is increased in the polymer solutions. In the dilute solution at high
1.0
0.9
z
0.8
0.7
40 [SDS] / mM
Fig. 5. III/I vibronic peak ratios from pyrene steady-state fluorescence spectra. The symbols denote: squares = 0% PEO, tirangles = 0.2% PEO, circles = 2% PEO, open symbols = 20°C, and filled symbols = 40°C
SDS concentration the ratio comes very close to the ratio in the aqueous solution at both temperatures, but in the semi-dilute solution this is true only at the higher temperature. This supports the idea of a limited number of interacting dusters on each polymer chain, as the III/I-values in aqueous micellar solution are approached in the dilute system at higher surfactant concentration, but not in the semi-dilute system. At higher temperature the approach to the values of the aqueous SDS solution occurs at lower surfactant concentration. This can be explained by a higher hydrophobicity of PEO at elevated temperatures, which would lead to a stronger interaction with the hydrophobic parts of the cluster interface, in accordance with the fin- dings from the lifetime measurements mentioned above -- the stronger interaction with the polymer leads to a less polar cluster with less difference between the lifetimes at higher temperature. The polymer itself does not interact with pyrene, as is seen from the III/I-values at zero surfactant concentration.
In the small clusters, with a higher curvature than ordinary micelles, pyrene would have been more exposed to water than in micelles, were it not for the shielding effect of the polymer. The increased shielding explains the small, but significant differences in III/I-values between the different PEO- concentrations. At higher temperature there will be a less polar pyrene microenvironment due to stronger interaction between the polymer and the
cluster, thus leading to a decreased difference in III/I-values at different PEO concentrations.
Conclusions
Each polymer chain can host a certain number of clusters that depends only on the chain length. At the critical aggregation concentration (CAC), surfactant aggregates or dusters start to form almost simultaneously at this number of locations.
At the CAC the aggregates, or clusters, have very low aggregation numbers -- about one-third that of the ordinary micelles. The polymer wraps around the cluster and replaces water at the cluster/bulk interface. Added surfactant is consumed by the growth of all clusters simultaneously, and not by the build-up of more clusters; almost all duster locations on the polymer chain become immediately occupied in the dilute solutions. When the clusters grow, the polymer will wrap around a decreasing part of the duster interface area, thus leading to a decreased shielding from oxygen quenching. In the semi-dilute solutions, however, at CAC, not all locations are occupied because of the 10 times higher polymer concentration. In this case, added surfactant will first mainly be consumed by occupying the remaining locations before the clusters start to grow, but a certain amount simultaneously takes part in the cluster growth.
From comparing model simulations and the ex- perimentally found aggregation numbers at the two temperatures, we conclude that the limiting size of the clusters is smaller than for ordinary micelles. This difference increases with increasing temperature.
Acknowledgement
This work was supported by the Swedish Natural Science Research Council and the Swedish Natio

trends in colloid and interface science v

Documents