Unusual findings on studying surfactant

Colloids and Surfaces A: Physicochemical and Engineering Aspects 205 (2002) 215– 229 www.elsevier.com/locate/colsurfa Unusual findings on studying surfactant solutions: displacing solvatochromic pyridinium N-phenolate towards outlying areas of rod-like micelles? Nikolay O Mchedlov-Petrossyan a,*, Natalya A Vodolazkaya a, Christian Reichardt b a Chemical Faculty, Kharko6 V.N Karazin National Uni6ersity, S6oboda Sq 4, Kharko6 61077, Ukraine b Department of Chemistry, Philipps Uni6ersity, D-35032 Marburg, Germany Received September 2001; accepted 24 January 2002 Abstract During studies of colloidal surfactants solutions by means of acid/base and solvatochromic indicators an unusual finding was observed The properties of micelles of cetyl trimethylammonium bromide (CTAB) and N-cetyl pyridinium bromide (CPB), modified by introducing various organic anions (altogether 24 anions), were examined by means of 4-(2,4,6-triphenylpyridinium-1-yl)-2,6-diphenylphenolate, used to establish the empirical ET(30) solvent polarity scale, at working concentrations of ca × 10 − M Surprisingly, on addition of electrolytes (0.01 M) containing some organic anions (tosylate, alkyl sulfonates, etc.) to aqueous solutions of a cationic surfactant (0.001 M), in the presence of the betaine dye, instead of the expected red shift, a blue shift (up to 75 nm) of its long-wavelength solvatochromic intramolecular charge-transfer (CT) Vis absorption band was observed This gives evidence for displacing the zwitter-ionic dye from a less polar to a more polar (i.e more hydrated) microenvironment The band shift is more or less expressed for various surfactants and for anions of different structure, and in some cases it was not observed at all This blue shift can be explained by a micellar ‘sphere “ rod’ transitions Such transformations, caused by the addition of hydrophobic counter-ions, are known to be unfavourable for normal solubilizations and manifest themselves in moving of the large-sized dye towards outlying areas of the micellar palisade An increase in both, surfactant and anion concentration, results normally in a red shift of the CT band, as compared with its spectrum in CTAB, respectively, CPB micelles In order to clarify further the behaviour of the dye, its acid/base properties were studied, its concentration was changed, ethanol was added, and both more hydrophilic and more hydrophobic betaine dyes were analogously examined © 2002 Elsevier Science B.V All rights reserved Keywords: Solvatochromic probes; Pyridinium N-phenolates; Absorption spectra; Blue shift; Surfactants; CTAB; CPB; Organic anions; Micellar ‘sphere “rod’ transitions * Corresponding author Tel.: + 380-57-2457445 E-mail addresses: nikolay.o.mchedlov@univer.kharkov.ua (N.O Mchedlov-Petrossyan), reichard@ps1515.chemie.uni-marburg.de (C Reichardt) 0927-7757/02/$ - see front matter © 2002 Elsevier Science B.V All rights reserved PII: S - 7 ( ) 0 2 - 216 N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 Introduction For the study of lyophilic ultramicroheterogeneous systems solvatochromic probe dyes come more and more into use [1 – 19] Suitable solvatochromic dyes, i.e dyes of which the position of their UV/Vis absorption bands is strongly solventdependent, have been applied above all for the empirical determination of the polarity of solvents [20 – 24] Their sometimes extreme UV/Vis spectroscopic sensitivity against small changes in their molecular microenvironment make them also useful for the study of the interior of all kinds of microheterogeneous systems such as microemulsions, micelles, phospholipid bilayers, etc [1– 19] This approach implies that suitably selected solvatochromic dyes characterize the region of their location in the microheterogeneous system, which can differ from one indicator dye to the other [25] In studying the properties of both, micelles of colloidal surfactants [26] and of phospholipid liposomes [27], the problem of having often various sites for the location of the probe dye is one of the important aspects in these investigations According to some results referring to this [28,29], an equilibrium is possible between species dissolved in the hydrocarbon core (‘oil droplet’) and bound by micellar interfaces (i.e absorption ? adsorption) By the way, a certain analogy seems to exist between the location of various probe solutes in micelles and the preferential solvation of solutes in binary solvent mixtures [30– 33] The present communication illustrates some unexpected difficulties which can arise in the study of common micellar systems with a widely used solvatochromic betaine dye During studies of the influence of inorganic and organic counter-ions upon the protolytic properties of acid/base indicators bound to micelles of colloidal cationic surfactants, we have found a sharp decrease in the values of ‘apparent’ ionization constants, K aa, of such indicators in the presence of an excess of organic anions According to the electrostatic model, the pK aa value of acid/base indicators completely bound to the micelles is connected with the thermodynamic ‘aqueous’ value, pK w a , and the electrical potential of the Stern layer, c, according to Eq (1) [3– 5,34–36] pK aa = pK w a + log = pK ia − kR fm cF R + log m − kHR f HR 2.3RT cF 2.3RT (1) Here, kR and kHR are the activity coefficients of transfer from water to the pseudophase, f m the concentration activity coefficients of the species, F is the Faraday constant, R the gas constant, T the absolute temperature, and K ia the so-called ‘intrinsic’ equilibrium constant According to the pseudophase ion-exchange model, the pK aa value of the acid/base indicator completely bound to the micellar surface depends on the concentration of the supporting electrolyte due to shielding of the micellar surface charge, as described by Eq (2), bi pK aa = const+ log % Si [X− i,w] , (2) j where Si is the selectivity parameter of the given counter-ion relative to a standard ion, for which Si is taken to be unity, and [X− i,w] is the equilibrium concentration of the given counter-ion in the bulk phase; the term ‘const’ includes pK ia [37,38] The term containing Si appears as a result of CMC dependence on [X− i,w] The meaning of the Si parameter is close to that of the constant of ion exchange In the particular case of a single counter-ion, Eq (2) can be rewritten as Eq (3) pK aa = Bi + bi log[X− i,w] (3) The slope bi in Eq (3) can be regarded as the degree of neutralization of surfactant ions by the counter-ions in the Stern layer According to the results obtained by us earlier [37,38], substitution of the Br− ions in CTAB micelles by more hydrophobic organic anions leads to a distinct increase in the values of Si, Bi, and bi (e.g bi : 1) In the latter case, the colloidal particles in the surfactant solution are to be considered uncharged (c “ 0) within the framework of the electrostatic model On the other hand, for uncharged surfaces the pK aa value of the acid/base indicator should not be influenced at all by the electrolyte concentration in the bulk phase From the viewpoint of the N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 pseudophase ion exchange model, the salt effect can be regarded as the result of an exchange of HO− anion in the Stern region for X− [34] Information on the possibility of an excessive adsorption of organic anions in the micellar system [H29C14 N(CH3)3]+Br− +salicylate−, obtained from electrophoretic measurements and with pK aa probes, is somewhat contradictory [39,40] Our investigations led us to the conclusion, that the explanation of the observed effects only by the ion-exchange is insufficient In particular, it is well known, that addition of electrolytes containing organic anions to micellar solutions of cationic surfactants (e.g CTAB and CPB) causes a rebuilding of the micelles, which is testified by viscosimetry [41– 43], electron microscopy [42], and NMR spectroscopy [44] In aqueous CTAB and CPB solutions, an increase in the surfactant concentration and the introduction of an excess of Br− ions lead undoubtedly to micellar ‘sphere“ rod’ transitions [43,45] In the presence of relatively hydrophobic organic counter-ions, the latter display an even decisive impact upon such micellar transitions, which is demonstrated by the pK aa values of the acid/base indicators applied in these systems [37– 39], by the marked decrease in the CMC values [46], and by the increase in viscosity of micellar solutions of cationic surfactants [37,41– 43] Alterations in the micellar shape can possibly reflect changes in the character of the hydration of micellar surfaces If the latter really takes place, it should be registered by solvatochromic indicators with varying hydrophobicity Suitable probe dyes for these investigations are the strongly solvatochromic pyridinium N-phenolate dyes – shown in Scheme 1, the hydrophobicity of which increases in the sequence B B3, according to their molecular structure [47–49] The long-wavelength intramolecular charge-transfer (CT) Vis absorption band of dye is hypsochromically shifted by Du =357 nm in going from nonpolar diphenyl ether (umax =810 nm) to polar water (umax = 453 nm) as solvent, which corresponds to an increase in the solvent-induced electronic Transition Energy, ET(2) [=ET(30); see Scheme 1], of ca 28 kcal mol − (1 kcal = 4.184 kJ) Therefore, 217 betaine dye (dye number 30 in Ref [47]) has been used as standard dye for a UV/Vis-spectroscopically derived solvent polarity scale, called the ET(30) scale [21,47,49] Sometimes, betaine dye has been called the ‘ET(30) dye’ or simply ‘ET(30)’ in the literature, which is somewhat unfortunate, because ET(30) is defined as the molar electronic transition energy (in kcal mol − 1) of dye number 30, i.e the formula number given to this dye in the first publication on this topic [47] The extreme sensitivity of the Vis absorption band of the zwitter-ionic, highly dipolar betaine dyes 1–3 to small changes in their microenvironment make them ideal indicator molecules not only for the empirical determination of solvent polarities, but also for the study of the polarity of the interior of micelles However, in the lastnamed systems the character of indicator molecules location is of especial importance, with respect to the complexity of micellar structure As a rule, both the increase in concentrations of cationic surfactants and addition of inorganic Scheme 218 N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 Fig Vis absorption spectrum of betaine dye 2, measured in water (line 1), after addition of 0.001 M CTAB (line 2), and of 0.001 M CTAB +0.01 M Tos− (line 3) salts are known to lead to transitions of spherical micelles into rod-shaped ones, the latter being less hydrated And indeed, the increase in umax values of dye was registered under such conditions [1,3] Earlier [37], we have already carried out such a study with betaine dye embedded in the micellar system CTAB/H7C7 SO− (tosylate ion) with a remarkable abnormal result: at low CTAB concentration (C =0.001 M), an increase in the tosylate concentration (up to 0.01 M) leads, contrary to our anticipation, instead of a bathochromic band shift, to a distinct hypsochromic dislocation of the solvatochromic CT absorption band of dye to umax = 465 nm, as compared with the analogous band measured in the micellar system CTAB/Br− within the same concentration range (umax = 540 nm; in the presence of M NaBr, the umax value grew to 545 nm) As this Vis spectrum of dye is similar to that measured in pure water (cf Fig 1), it could be supposed that the micellar transformations of cetyl trimethylammonium tosylate lead to a ‘pushing-out’ of the rather large betaine dye molecules (with altogether five peripheral phenyl groups) into the direction of the aqueous phase Variations of the CTAB and tosylate concentration, particularly their increase, result eventually in a disappearance of the unexpected band at umax =465 nm and in a restoration of the former absorption band at umax : 540 nm Naturally, it is of importance to clarify these unusual findings in order to see whether it is of general character or not Therefore, the present study was performed with typical cationic surfactants such as CTAB and CPB dissolved in water, with addition of electrolytes containing organic − anions such as H7C7 SO− (tosylate, Tos ); salicy− − late (HSal ); H5C6 CO2 (benzoate, Benz−); ophthalate (o-Phth2 − ); o-sulfobenzoate (o-SO3 Benz2 − ); m-nitrobenzoate (m-NO2Benz−); ochlorobenzoate (o-ClBenz−); p-nitrophenolate (pNO2Phen−); the carboxylates H3C CO− , H7C3 − − − CO− , Cl C CO , F C CO , and H C 3 13 CO2 ; − the alkyl sulfonates H11C5 SO3 , H13C6 SO− , − − H15C7 SO− , H C SO , H C SO , and 17 19 − H21C10 SO− ; 5,5-diethylbarbiturate (Ver ); p− H2N C6H4 SO3 (sulfanilic acid anion); 2,4- and 2,6-dinitrophenolate; 2,4,6-trinitrophenolate In this investigation, we examined the solvating properties of various aqueous micellar systems, hydrophobizated by the organic anions added, by means of the solvatochromic betaine dyes 1–3 in the concentration region of C= 0.001–0.022 M for CTAB and CPB and, as a rule, of C= 0.001– 0.025 M for the organic anions mentioned above Materials and methods 2.1 Chemicals Cetyl trimethylammonium bromide (CTAB; Sigma –Aldrich, purity 99%) and N-cetyl pyridinium bromide (CPB; Minkhimprom, USSR) were used as commercially obtained p-toluenesulfonic acid was purified with trichloromethane according to a procedure recommended by Perron [50] Sodium alkyl sulfonates (chromatographically pure); ammonium o-carboxybenzosulfonate, 2,4- and 2,6-dinitrophenol; picric, o-chlorobenzoic, 5,5-diethylbarbituric, sulfanilic, acetic, trifluoroacetic, trichloroacetic, butanoic, and heptanoic acids (analytical grade) were used without further purification Salicylic acid, sodium salicylate, benzoic acid, sodium benzoate, potas- N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 219 sium o-phthalate, m-nitrobenzoic acid, and p-nitrophenol were purified by recrystallization Potassium and sodium bromides, potassium chloride, aqueous hydrobromic, hydrochloric and phosphoric acids, borax, and tris(hydroxymethyl)aminomethane (TRIS) were of analytical grade The betaine dyes 4-(2,4,6-triphenylpyridinium-1yl)phenolate (1× ca – 12H2O; dye number in Ref [47]), 4-(2,4,6-triphenylpyridinium-1-yl)-2,6diphenylphenolate (2 × ca 2H2O; dye number 30 in Ref [47]), and 4-[2,4,6-tri(4-t-butylphenyl)pyridinium-1-yl]-2,6-di(4-t-butylphenyl)phenolate (3×ca 3H2O; dye number 1d in Ref [49]) were synthesized and purified as described earlier [47– 49,51,52] In spite of drying these somewhat hygroscopic dyes with P4O10 in vacuo, they still contain some water of re-crystallization The standard betaine dye 2, commonly used for the determination of ET(30) solvent polarity values [47,49], is also commercially available (Aldrich, order no 27,244-2) Bromophenol blue (BPB, Minkhimprom, USSR) was used as received [36– 38,53]; its UV/Vis spectrum and its pK w a value is in agreement with the values given in the literature Ethanol (96 mass%) of high quality was purified by distillation; the absence of aldehydes was confirmed UV/Vis spectroscopically Methanol was purified using standard procedures The aqueous NaOH stock solution, prepared by using CO2 – free water, was kept protected from carbon dioxide tion band disappears) On addition of the anions (0.01 M in the working solution), H15C7 SO− (0.01 M), and the picrate anion H17C8 SO− (0.001 M) to aqueous CTAB solutions (0.001 M), a strong turbidity was observed; at alkyl sulfonate concentrations of 0.005 M the solutions were transparent The addition of 2,4- and 2,6-dinitrophenol causes roughly the same increase in the pK aa value of BPB as that of H15C7 SO− However, the yellow colour of these aromatic anions prevents their use in working solutions of dye at the required concentrations The UV/Vis absorption spectra were measured with the spectrophotometer SP-46 The umax values of dyes 1–3 were measured as described in Refs [47–49,54] pH measurements were performed at 2590.1 °C with a standard deviation of (0.01–0.02) with a potentiometer P 37-1 and a pH-meter pH-121, equipped with an ESL-63-07 glass electrode and an Ag/AgCl reference electrode in a cell with liquid junction (1 M KCl) Standard buffers (pH=1.68, 4.01, 6.86, and 9.18) and dilute aqueous HCl solutions were used for the cell calibration Suitable pH values of the working solutions used for pK aa determinations were provided with borate, phosphate, and TRIS buffers in case of dye (ionic strength of the buffer solutions 0.0125, 0.01, and 0.01 M, respectively) and with hydrobromic acid in case of BPB The pK aa values were obtained using the dependence of the absorbance, A, on the pH value at constant indicator concentrations, according to Eq (4), 2.2 Procedure pK aa = pH+ log[(AR − A)/(A − AHR)], Stock solutions of dyes – were prepared with ethanol (96 mass%) as solvent Some experiments were also carried out with stock solutions of dye in methanol as well as in micellar CTAB solutions Aliquots of the dye stock solutions were of such volume that the alcohol content of the aqueous working solutions did not exceed 2% On evaluating the umax of dyes – their Vis absorption spectra were measured as a rule in solutions with pH ca 12 (adjusted with NaOH) to ensure their presence in the coloured zwitter-ionic form (in acidic media, the betaine dyes are reversibly protonated and the solvatochromic Vis absorp- (4) in which pH corresponds to the bulk (aqueous) phase, AR and AHR are the respective absorbances observed under conditions in which only R or HR exist in solution, and A the absorbance at current pH All the solutions were prepared and measured at 25 °C Results and discussion 3.1 Influence of organic anions on acid/base equilibria in surfactant micelles The pK aa values of BPB (HR− = R2 − + H+) N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 220 determined in surfactant micelles in the presence of various organic anions are presented in Table 11 The error in the pK aa values does not exceed 90.1 and is on the average 0.03 The essential increase in the pK aa values of BPB, as compared with the value found for the bromide system, probably depends not so much on shielding of the micellar surface charge by the organic anions but on possible structural transitions of the micelles caused by substitution of hydrophobic counter ions for Br− Such transitions, in turn, can result in additional changes in pK ia And really, if the effects are to be explained exclusively in terms of simple ion-exchange, with constant pK ia value, then the Si values Eq (2) evaluated from the pK aa (Table 1) are higher than the Si calculated from the CMC values, available in literature For instance, the Si values for H13C6 SO− H7C7 SO− and HSal− in , , + CTA system are 1492, 2395 [38], and ca 60, respectively (S− Br 1), while the CMC values of cetyl trimethylammonium salicylate and of several substituted benzoates are only five times lower than that of CTAB [46] Table Vis absorption maxima of betaine dye 2, measured in aqueous solution in the ‘critical region’ of concentrations (0.001 M CTAB or CPB, 0.01 M of the organic anion) Organic anion H21C10 SO− H19C9 SO− H7C7 SO−a H17C8 SO−c H13C6 SO− H15C7 SO−c H11C5 SO− H13C6 CO− −a HSal o-sulfobenzoate2− Benzoate−a o-chlorobenzoate−d o-phthalate2−a F3C CO− C13C CO− H3C CO− − H7C3 CO2 Ver− NH2 C6H4 SO− p-NO2phenolate− m-NO2 benzoate− umax/nm CTA+ CPB+ 462 462 465b 470 471 473 497 498 510 510 510 510d 525 528 530 532 535 535 535 540 558 – – 485 – 487 – – 514 530 – – – – – – – – – – – 555 a Table The pK aa values of bromophenol blue (BPB), determined in aqueous micellar solutions of cationic surfactants (0.001 M) in the presence of various organic anions (0.01 M) Anion Br− NO− HSal− H7C7 SO− H13C6 SO− o-SO3Benz2− H11C5 SO− m-NO2Benz− a b The stock solution of the dye was prepared with methanol b From Ref [37] c 0.005 M d If the anion concentration is 0.001 M, max is 528 nm 3.2 Vis spectra of the sol6atochromic indicator dye pK aa of BPB CTAB CPB 2.35a 2.56a 3.82a 3.35b 3.36 3.02 2.93 3.40 2.10 2.34 3.18 3.12 3.13 2.59 2.68 3.34 Obtained by A.V Timiy From Ref [37] A much wider set of data are to appear in a future publication Examination of the Vis spectra of embedded in the systems ‘CTAB (or CPB) –organic anion’ within the concentration ranges C= 0.001– 0.022 and 0.001–0.025 M, respectively, reveals that the ‘dye displacing effect’, manifesting itself in the shift of long-wavelength solvatochromic Vis absorption band of 2, is clearly observed at C(surfactant)= 0.001 M and C(organic anion)= 0.01 M (called ‘critical region’ or ‘critical zone’ of concentrations) Typical umax values and Vis absorption spectra of are presented in Table and in Figs and (if the cation accompanying organic anions is not shown, it is Na+) N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 Fig Vis absorption spectrum of betaine dye 2, measured in aqueous solutions containing different molar concentrations of CTAB and H13C6 SO3Na (see insert) Within the concentration range under discussion the surfactant: dye ratio is ca 20 Evidently, one CTAB micelle contains several dye molecules However, the umax values of the dye as well as its pK aa values (see below) are rather close in 0.001 and 0.005 M CTAB solutions The ‘abnormal’ effects appear only if some organic counterions are introduced On the other hand, Fig demonstrates a distinct red shift against the spectrum in CTAB solutions, if concentrations of both surfactant and H13C6 SO− are higher Besides, under Fig Vis absorption spectrum of betaine dye 2, measured in aqueous solutions containing different molar concentrations of CTAB and o-SO3Benz2 − (see insert) 221 such conditions (curve 4) the solutions become very viscous In 0.001 M CTAB+ 0.01 M H13C6 SO− mixtures, with build-up NaBr concentrations (0.01; 0.5; M), the ‘unusual’ band with umax ca 470 nm finally disappears, and absorption, typical for the CTAB –NaBr system, is registered As the interaction of the dye with Br− or Na+ ions, as well as + that of H13C6 SO− with Na , was never registered in diluted aqueous solutions, the competition between X− and the inorganic ions seems to be improbable On the other hand, at 400-fold excess of bromide anions the CTA+ H13C6 SO− micelles certainly convert into CTA+Br− ones And the umax value of the solvatochromic dye (ca 545 nm) indicates, that its microenvironment becomes ‘traditional’ (as in CTAB micelles, even rod-shaped due to high NaBr concentrations) According to their ability to induce a hypsochromic band shift, as compared with the band position of observed in pure CTAB micellar solutions (umax = 540 nm), the organic anions can be classified into three groups Group I (with a blue shift of Du : 70 nm): − H13C6 SO− H19C9 SO− H7C7 SO− (Tos ), , , − − H21C10 SO3 , H15C7 SO3 (0.005 M), and H17C8 SO− (0.005 M) Group II (with a smaller blue shift of Du:10– − − 15 nm): H11C5 SO− , H13C6 CO2 , Cl3C CO2 , − − − 2− HSal , Benz , o-ClBenz , o-Phth , oSO3benz2 − , and F3C CO− Group III (with practically no blue shift, Du : 0): H3C CO− Ver−, p-H2N C6H4 SO− p2 , , − NO2 C6H4 O It is quite probable that with anions of this group changes in the micellar structure still occur, but of such kind that no shifting of the dipolar betaine dye to the more hydrated area takes place In the case of 0.01 M mNO2Benz− Table 2, the Vis absorption band of undergoes a significant red shift of Du = 18 nm, evidently due to the relatively high hydrophobicity of the cetyl trimethylammonium m-nitrobenzoate micellar surface It should be noted, that in case of using CPB as cationic surfactant, the Vis absorption band of is shifted towards the blue at addition of anions of Groups I and II, and the umax values are 15–20 nm higher, as compared with the cetyl trimethy- 222 N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 lammonium system The reason could be the intermolecular interaction between the y-electron systems of CP+ and that of the dye chromophore For the organic anions of Group I, the ‘displacing’ effect of dye can be interpreted as follows: under conditions with deficiency of surfactant (at C = 0.001 M), the enlargement of the micelles results in such a sharp decrease in their number, that the ratio dye:micelle is ] 10, which hinders a complete binding of dye On the other hand, the conversion of spherical micelles into anisometrical ones (e.g rod- or even worm-like micelles) makes them less suitable for binding such large-sized molecules as that of dye For organic anions of Group II, the effect of ‘displacing’ seems to be intermediate between ‘normally bound’ and ‘strongly displaced’ states In any cases the effects observed cannot be connected with the (partial) transformation of CTAB micelles into CTA+HO−, because at lower pH the umax values remain unaltered, even if the intensity decreases Such observations were made during determination of pK aa (see below) The different molecular structure and properties of the organic counter-ions determine the peculiarities of micellar transformations and, in particular, the non-identity of the respective Stern regions in the various micellar systems For example, detailed information on the micellar structure of decylammonium carboxylates (H2n + 1Cn CO− , up to n = 5) and alkyltrimethylammonium 5ethylsalicylates is available in literature [55,56] We have also carried out analogous measurements in aqueous 0.001 M CTAB solutions with the addition of tosylate or n-amyl sulfonate ions at concentrations of dye being fivefold smaller than in the series described above In the tosylate system, the Vis band position of turned out to be practically unchanged, whereas with n-amyl sulfonate a red shift of Du :13 nm was observed This can indicate, that under these experimental conditions the numbers of micelles and that of dye molecules are commensurable, and a complete binding of dye by the micelles is to take place However, the value of umax =510 nm, obtained in the system CTAB (0.001 M)+H11C5 SO− (0.01 M), deviates considerably from that of umax =540 nm, observed for complete binding of (see above) The absorption observed in the region around 510 nm can be the result of a superposition of two bands with umax : 470 and 540 nm In addition, for organic anions of Group III (e.g for the 5,5-diethylbarbiturate ion), a fivefold increase in the concentration of dye does not lead to spectral changes at all The problem of the position of pyridinium Nphenolate betaines within the ‘normal’ micelles of colloidal surfactants has been touched on by many authors [1–3,19], who have mostly proceeded from the NMR data Though the proximity of the phenolate moiety of these dyes to the quaternary nitrogen atom of the cationic surfactant is generally recognized, the viewpoints on the orientation of the rest portion of the dipole molecule are somewhat contradictory Taking into account the versatility and complexity of the systems studied in the present work, we yet not propose a distinct picture of dye location within the modified micelles, the more so in the ‘displaced state’ 3.3 Betaine dye as pKa -probe The modification of micellar pseudophases as solvating medium was also studied by using betaine dye as an acid/base indicator (HR+ = R+ H+) with the colourless N-(hydroxyphenyl)pyridinium ion as acid and the coloured zwitter-ionic betaine dye as base, according to the following acid/base equilibrium (Scheme 2): The pK aa values of dye determined in aqueous CTAB and CPB micellar solutions in the presence of various counter ions are compiled in Table Under conditions of ‘normal binding’ of to cationic micelles (0.005 M CTAB) and in the presence of tosylate ions (0.01 M Tos−; umax = 540 nm), the pK aa value obtained (i.e 8.08 and Scheme N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 223 Table The pK aa values of betaine dye 2, determined in aqueous micellar solutions of the cationic surfactants CTAB and CPB Surfactant umax/nm of Csurf, M Anion Canion, M pK aa CTAB CTAB CTAB CTAB CTAB CTAB 540 540 465 497 540 540 0.005 0.005 0.001 0.001 0.001 0.005 H7C7 SO− H7C7 SO− H7C7 SO− H11C5 SO− Br− Br− 0.01 0.01 0.01 0.01 0.01 0.01 8.08a 8.25b 8.61b 8.34b 7.58c 7.36c,d CPB CPB None 540 540 453 0.001 0.005 Br− Br− None 0.01 0.01 7.55c 7.18c 8.64e a TRIS buffer Borate buffer c Phosphate buffer d Obtained by A.V Timiy e From Ref [3,51] b 8.25; see Table 3) depends somewhat on the buffer composition As to the location of the colourless species HR+ in the cationic micelles, they must be considered as completely bound, because the salt HR+Br−, being practically insoluble in water, does not precipitate in the presence of CTAB micelles The pK aa value of dye in the ‘displaced state’ equals to 8.61 (Table 3) This value coincides with the pK w a values determined for in water by Drummond et al [3] and by Kessler and Wolfbeis [51] Nevertheless, we suppose that the betaine dye does not exist in the aqueous phase as singular hydrated molecules, but is bound in some ‘pre-micellar’ aggregates with the surfactant and/ or with the organic anions, or stay even bound but on the peripheral, extremely outlying and hence strongly hydrated area of the transformed micelles And indeed, in aqueous solutions of dye not only the umax value is somewhat lower, but the solutions are rather unstable because of the low dye solubility, whereas in the case under discussion the solutions are rather stable Blank experiments were performed without surfactant The spectra of the dye with and without − Tos−, H13C6 SO− coincide pre3 , or other X cisely, while in the presence of CTAB they differ, even if the maxima positions are close Moreover, precipitates appear during pK aa determination after a period of time in aqueous dye solution as well as in dye+ X− systems, the case being not such a one if CTAB is present The pK aa value of dye in the state which corresponds to a umax value of 497 nm (0.001 M CTAB + 0.01 M H11C5 SO− ) turns out to be the average between the values obtained for the ‘normally bound’ and the ‘displaced’ state, namely pK aa = 8.34 On the other hand, the umax value of 497 nm coincides with that observed in aqueous sodium dodecyl sulfate (SDS) micelles Though, our electrophoretic studies with Sudan III as a tracer show that the micelles in the system mentioned before (0.001 M CTAB+0.01 M H11C5 SO− ) retain their positive charge, and the pK aa values of in SDS solutions are with ca 11 essentially higher [3,57] In aqueous 0.01 M SDS solutions with 0.01 M n-Bu4N+, the pK aa value is with ca 9.9 also markedly higher than in the systems discussed before By the way, it is of interest to compare the pK aa values of the dye in micellar solutions of various type [3,36,53,57] (see also Table 3) with the newly reported data obtained in sol–gel entrapped micelles [58] 224 N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 3.4 On the possibility of aggregation of betaine dye 3.5 Studies with the hydrophobic penta-t-butyl-substituted betaine dye It would have been possible to explain the observed abnormal spectral effects by aggregation (e.g dimerization) of the dipolar molecules of betaine dye In particular, the formation of so-called H-aggregates of some polymethine dyes in aqueous solutions, consisting in a parallel ‘stacking’ of planar cationic cyanine chromophores, leads to an appearance of a new blue-shifted absorption band in comparison with the monomer absorption band [59] However, the validity of Lambert– Beer’s law was checked for betaine dye in 1,4-dioxane, pyridine, and methanol as solvents (unfortunately, not in water because of the low solubility of the dye 2) within the concentration range of C(2) =10 − –10 − M [47,48] This law was found to be exactly fulfilled, which means, that in these solvents and in this concentration range exists no association of the zwitter-ionic dye molecules [47,48] Such an association seems also to be unlikely for sterical reasons According to the X-ray structure analysis of a bromo-substituted derivative of dye 2, the betaine dye is far from being planar: the five peripheral phenyl groups are heavily twisted, and the interplanar angle between pyridinium and phenolate ring amounts up to 65° [60], which is in good agreement with a recently calculated value of 68° for the electronic ground state of dye in the gas phase [61] Taking into account the validity of Lambert–Beer’s law and assuming that the conformations of dye in solution should not differ too much from that in the crystal lattice and in the gas phase, the explanation of the discussed unusual spectral effects through dimerization of dye seems to be less probable than through displacing of toward the more hydrated area of micelles It should be mentioned, however, that recent calorimetric measurements of the enthalpies of dilution of solutions of dye in some aliphatic alcohols at concentrations ] 10 − M showed some evidence of dye aggregation at higher concentrations [62] In contrast to dye 2, the penta-t-butyl-substituted betaine dye is practically insoluble in water, but better soluble in nonpolar solvents such as hydrocarbons Introduction of five t-butyl groups into the p-position of the peripheral phenyl groups of dye leads to a small bathochromic shift of the long-wavelength solvatochromic Vis absorption band by Du :5 –20 nm, depending on the solvent used For example, in ethanol dye absorbs at umax = 551 nm and dye at umax = 562 nm (Du = 11 nm) [49] Solubilized into an aqueous 0.022 M CTAB solution, dye absorbs at umax = 561 nm, which corresponds to a red shift of its Vis absorption band by ca 20 nm, as compared with the analogous band of dye measured in the same system With high probability, this can be explained in terms of a deeper penetration of the more hydrophobic betaine dye into the surfactant micelles In case of some microemulsions [63], the deviation reaches even 90 nm In addition, our experiments with liposomes of some phospholipids have shown that there is a remarkable red shift (ca 50 nm) of the band against that of as well In micellar solutions of a non-ionic surfactant n-nonylphenol-12, the umax values of dyes and are 540 and 589 nm, respectively (to be published) This result can also be attributed to different depths of dye incorporation into the oxyethylene mantle or even into the hydrocarbon core In this case, a certain analogy between the aforementioned effect and the preferential solvation of solutes in solvent mixtures can be traced Similar Vis absorption maxima were measured for the dyes and in aqueous 0.022 M CTAB solution in the presence of 0.025 M tosylate: umax = 548 [37] and 589 nm This can be analogously explained in terms of different incorporation depths However, another explanation is also possible Let us consider the behaviour of the betaine dye in binary water/ethanol mixtures In 96 mass% ethanol, dye absorbs at umax = 553 nm, in agreement with the value found for dye [64] In 43 mass% aqueous ethanol, dye absorbs at umax = N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 526 nm and dye at umax =572 nm This unusual absorption value of 3, even higher than that measured in ethanol (96 mass%), cannot be explained in terms of preferential solvation We suppose that a dispersion of the aggregated dye (e.g a sol or a suspension) may appear under this experimental conditions, unfavourable for the formation of a normal solution Introduction of an ethanolic stock solution of dye into water leads to its precipitation as a blue deposit, even with extremely diluted solutions Therefore, the pronounced red shift observed with dye in 0.001 M CTAB both, with 0.01 M tosylate (umax =630 nm) and without tosylate (umax =593 nm), can be explained by formation of a dispersed solid dye phase, stabilized by the surfactant It should be noted that another bulky and hydrophobic dye such as tetratolylporphine aggregates in aqueous media even in the presence of micellized SDS (C=3 × 10 − M) and converts into a solubilized molecular form on increasing the surfactant concentration and addition of sodium chloride [65] In any case, the experiments with dye confirm significant changes of the CTAB micelles on addition of tosylate ions 225 explained by a deficiency of CTAB (or CPB) micelles and sufficient amounts of ethanol The hypothesis of CTAB micelles dissolution in water/ ethanol mixtures seems to be improbable for several reasons For example, the umax value of the BPB dianion, registered by us in the system given, coincides with the umax value of 603 nm, observed in aqueous CTAB micellar solutions, whereas in a water/ethanol mixture of the same composition without surfactant, the umax value is 592 nm, only nm higher than that in water At a CTAB (or CPB) concentration of 0.001 M and mass% ethanol, and a CTAB (or CPB) concentration of 0.01 M and 16 mass% ethanol, the umax values of dye are 520 and 540 nm, respectively (Fig 5b) Besides, it is known [66] that such concentrations of ethanol display relatively small influence on the CMC values of surfactants 3.7 The beha6iour of the less hydrophobic sol6atochromic betaine dye 3.6 The influence of ethanol addition to solutions of betaine dye To make this investigation of the solvating properties of cationic micelles, modified with organic anions, more versatile, we have also used the less hydrophobic betaine dye 1, without the two phenyl groups in the phenolate moiety of dye [47] The Vis absorption band of dye in aqueous micellar CTAB solutions is red-shifted Addition of 16 mass% of ethanol into the systems ‘dye 2+cationic surfactant+Group II anions’ in the ‘critical region’ of concentrations leads to a hypsochromic shift of the Vis band of (Figs and 5) In the case of o-SO3Benz2 − , umax =469 nm, which corresponds to a blue shift of Du =40 nm as compared with the results given in Fig In aqueous ethanol (16 mass% of EtOH), dye absorbs at umax =475 nm (Fig 5a) Introduction of the same amount of ethanol into the systems ‘dye +cationic surfactant+Group III anions’ results in smaller hypsochromic band shifts (Du :10–15 nm) Note that even in the system ‘CTAB (or CPB) +EtOH’ without organic anions a transfer of dye into the bulk phase can be registered However, this phenomenon is observed only at concentrations of 0.001 M CTAB (or CPB) and 16 mass% ethanol, which can be Fig Vis absorption spectrum of betaine dye 2, measured in aqueous solutions containing ethanol (16 mass%) as well as different molar concentrations of CTAB and o-SO3Benz2 − (see insert) 226 N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 Fig Vis absorption spectrum of betaine dye 2, measured in (a) water (line 1) and water/ethanol (16 mass% of EtOH; line 2) as well as in (b) water/ethanol solutions containing various concentrations of EtOH and CTAB (lines – 3) with increasing surfactant concentration, in accord with the data obtained by Plieninger and Baumgaă rtel [2] In particular, at CTAB concentrations of 0.001 and 0.022 M, dye absorbs at umax =417 and 465 nm, respectively This result testifies an incomplete binding of the probe dye by CTAB micelles in dilute surfactant solutions All the micellar systems investigated by means of dye were examined with dye as well.2 Direct comparison of the results obtained with the two different indicators is impossible because of the incomplete binding of dye by the micelles, although some analogies can be revealed The absorption maxima of dye in various micellar systems are given in Table In systems with Group I organic anions, the ‘displacement’ of the dye can be registered even at CTAB and anion concentrations of 0.001 and 0.005 M, respectively, expressed by the small blue shift observed (Du =5 nm) These experiments were carried out only with CTAB; in case of CPB, the umax values of the dye are lower even at high surfactant concentration (see Table 4), probably due to interaction of the y-electron systems of the phenolate moiety of the dye and pyridinium moiety of CP+ In the ‘critical region’ of concentrations for Group II anions, in particular for H11C5 SO− and HSal−, the umax value of dye corresponds to the ‘aqueous’ one Probably, due to its greater hydrophilicity, this dye can be easier dislocated to the outlying area of micelles or is located in the bulk aqueous phase The most probable explanation of the observed effects is a shift of the equilibrium between the bound and unbound dye 1, influenced by the modification of the micelles In the system ‘dye 1+ 0.022 M CTAB+0.025 M tosylate’, the umax value of indicates that this Table Vis absorption maxima of betaine dye 1, measured in water and various micellar systems umax/nm System Water 0.001 M 0.022 M 0.001 M 0.001 M 0.022 M 0.001 M 0.001 M CTAB CTAB CTAB–0.005 M H7C7 SO− CTAB–0.01 M H7C7 SO− CTAB–0.025 M H7C7 SO− CTAB–0.01 M H11C5 SO− CTAB–0.01 M HSal− 412 417 465a 412 412 417 412 412 a In CPB solution of equal concentration umax equals 444 nm N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 Table Vis absorption maxima of betaine dye 1, measured in aqueous ethanol with and without the surfactant CTAB System umax/nm Water 16 mass% EtOH 0.001 M CTAB+8 mass% EtOH 0.001 M CTAB+16 mass% EtOH 0.010 M CTAB+16 mass% EtOH 412 422 417 422 432 dye is absent from micelles, contrary to dye applied under the same experimental conditions Thus, the changes in micellar size and shape and/or the hydrophobization of the micellar surface prevent dye from binding Investigations of aqueous micellar CTAB solutions with ethanol addition by means of dye gave the following results (cf Table 5): at small additions of aqueous ethanol (8 mass% EtOH), the equilibruim stays unchanged With increasing ethanol content (16 mass% EtOH), the dye is transferred into the bulk phase, which is actually a water/ethanol mixture At equal ethanol contents (16 mass% EtOH), but at higher CTAB concentration, the umax value of is intermediate The data collected in Tables and demonstrate the clear response of the Vis absorption spectra of dye to the changes occurring in the micellar pseudophase 227 The spectral changes are sometimes observable even by the naked eye (Du up to 75 nm) The influence of the organic anions varies due to the peculiarities of their molecular structure and the character of their incorporation into the micellar pseudophase The addition of aqueous ethanol (16 mass% EtOH) to micellar solutions of cationic colloidal surfactants at concentrations near the CMC display a significant impact, i.e betaine dye is now easier transferable into the bulk phase from the micelles All solvatochromic probe dyes 1–3 indicate the possibility of micellar transformations, leading to unfavourable conditions for dye binding However, dye is in general poorer bound by CTAB micelles, whereas the penta-tbutyl-substituted dye tends to move from the CTAB micelles into the aqueous phase in form of aggregates, and not of monomers The results obtained in this work encourage to further experimental investigations with micellar solutions of various cationic surfactants Acknowledgements The authors are grateful to Anna V Timiy (Kharkov National University) for participation in some preliminary experiments Conclusions The solvating properties of micellar solutions of cationic surfactants modified by addition of various organic anions and of ethanol were studied by means of the solvatochromic pyridinium N-phenolate betaine dyes – (having in this sequence an increasing hydrophobicity), leading to the following results: In the systems ‘cationic surfactant+ organic anion’, the anions display a substantial influence on the hydration of micellar surfaces and on micellar transitions, which results in displacing the probe dye from its ordinary position in the micelles, probably accompanied by the formation of some pre-micellar species References [1] K Zachariasse, N Van Phuc, B Kozankiewicz, J Phys Chem 85 (1981) 2576 [2] P Plieninger, H Baumgaă rtel, Ber Bunsenges Phys Chem 86 (1982) 161; Justus Liebigs Ann Chem (1983) 860 [3] C.J Drummond, F Grieser, T.W Healy, Faraday Discuss Chem Soc (1986) 95; Chem Abstr 106 (1987) 126 332k [4] C.J Drummond, F Grieser, T.W Healy, J Phys Chem 92 (1988) 2604 [5] F Grieser, C.J Drummond, J Phys Chem 92 (1988) 5580; and refs cited therein [6] M.A Kessler, O.S Wolfbeis, Chem Phys Lipids 50 (1989) 51; Chem Abstr 111 (1989) 59 537u [7] U Ueda, Z.A Schelly, Langmuir (1989) 1005 228 N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 [8] I.A Koppel, J.B Koppel, Org React (Tartu) 26 (1989) 78; Chem Abstr 113 (1990) 230 702m [9] D.-M Shin, K.S Schanze, D.G Whitten, J Am Chem Soc 111 (1989) 8494 [10] T Handa, M Nakagaki, K Miyajima, J Coll Interface Sci 137 (1990) 253; Chem Abstr 113 (1990) 46 955v [11] R Varadaraj, J Bock, N Brons, S Pace, J Phys Chem 97 (1993) 12 991 [12] J.M Perera, G.W Stevens, F Grieser, Colloids Surf., A 95 (1995) 185 [13] J Kriwanek, R Miller, Colloids Surf., A 105 (1995) 233 [14] D Banerjee, P.K Das, S Mondal, S Ghosh, S Bagchi, J Photochem Photobiol., A: Chemistry 98 (1996) 183 [15] M.F Vitha, P.W Carr, J Phys Chem B 102 (1998) 1888 [16] R Helburn, Y Dijiba, G Mansour, J Maxka Langmuir 14 (1998) 7147 [17] A Mishra, P.K Behera, R.K Behera, B.K Mishra, G.B Behera, J Photochem Photobiol., A: Chemistry 116 (1998) 79 [18] Z Shervani, Y Ikushima, Colloid Polym Sci 277 (1999) 595 [19] L.P Novaki, O.A El Seoud, Phys Chem Chem Phys (1999) 1957; Langmuir 16 (2000) 35; and refs cited therein [20] C Reichardt, Solvents and Solvent Effects in Organic Chemistry, 2nd edition, VCH, Weinheim, 1988, pp 339 – 405 Chapter [21] C Reichardt, Chem Rev 94 (1994) 2319 [22] C Reichardt, Kharkov University Bulletin (1999) No.437, Chemistry, Issue 3, 9; Chem Abstr 132, (2000) 180 091f [23] J.-L.M Abboud, R Notario, Pure Appl Chem 71 (1999) 645 [24] J Catala´ n, Handbook of Solvents In: G Wypych (Ed.), ChemTec Publishing, Toronto, and William Andrew Publishing, Norwich/N.Y., 2001, Chapter 10.3, pp 583 –616 [25] T Handa, C Ichihashi, I Yamamoto, M Nakagaki, Bull Chem Soc Jpn 56 (1983) 2548 [26] K.K Karukstis, A.A Frazier, C.T Loftus, A.S Tuan, J Phys Chem B 102 (1998) 8163 [27] R.J Clarke, Biochim Biophys Acta 1327 (1997) 269 [28] P Mukerjee, J Pharm Sci 60 (1971) 1528 [29] A.D James, B.H Robinson, N.C White, J Colloid Interface Sci 59 (1977) 328 [30] H Langhals, Angew Chem 94 (1982) 739; Angew Chem., Int Ed Engl 21 (1982) 724ff [31] R.D Skwierczynski, K.A Connors, J Chem Soc., Perkin Trans (1994) 467 [32] M Rose´ s, C Ra´ fols, J Ortega, E Bosch, J Chem Soc., Perkin Trans (1995) 1607 [33] E.B Tada, L.P Novaki, O.A El Seoud, J Phys Org Chem 13 (2000) 679 [34] N Funasaki, J Phys.Chem 83 (1979) 1998 [35] M.S Fernandez, P Fromherz, J Phys Chem 81 (1977) 1755 [36] G.P Gorbenko, N.O Mchedlov-Petrossyan, T.A Chernaya, J Chem Soc., Faraday Trans 94 (1998) 2117 [37] A.V Timiy, N.O Mchedlov-Petrossyan, E.N Glaskova, N.A Pinchukova, O.E Zhyvotchenko, Kharkov University Bulletin (1998), No 420, Chemistry, Issue 2, 235; Chem Abstr 132 (2000) 98 516g [38] N.O Mchedlov-Petrossyan, A.V Timiy, N.A Vodolazkaya, N.A Pinchukova, Kharkov University Bull (1999) No 454, Chemistry, Issue (27) 203 [39] M.A Cassidy, G Warr, J Phys Chem 100 (1996) 3237 [40] T Imae, T Kohsaka, J Phys Chem 96 (1992) 10 030 [41] V Hartmann, R Cressely, Colloids Surf., A 121 (1997) 151 [42] T Wolff, C.-S Emming, G von Bunau, K Zierold, Colloid Polym Sci 270 (1992) 822 [43] A Heidl, J Strnad, H.-H Kohler, J Phys Chem 97 (1993) 742 [44] S.J Bachofer, U Simonis, T.A Nowicki, J Phys Chem 95 (1991) 480 [45] J.N Ness, D.K Moth, J Colloid Interface Sci 123 (1988) 546 [46] A.L Underwood, E.W Anacker, J Phys Chem 88 (1984) 2390 [47] K Dimroth, C Reichardt, Th Siepmann, F Bohlmann, Justus Liebigs Ann Chem (1963) [48] C Reichardt, Ph D Thesis, Univ Marburg/Germany, 1962 (for a proof of the validity of Lambert – Beer’s law for solutions of betaine dye see p 23) [49] C Reichardt, E Harbusch-Gцrnert, Justus Liebigs Ann Chem (1983) 721 [50] R Perron, Bull Soc Chim Fr 11 – 12 (1952) 966 [51] M.A Kessler, O.S Wolfbeis, Synthesis (1988) 635 [52] B.R Osterby, R.D McKelvey, J Chem Educ 73 (1996) 260, 737 [53] N.O Mchedlov-Petrossyan, A.V Plichko, A.S Shumakher, Chem Phys Reports 15 (1996) 1661 [54] H Langhals, in: R.I Zalewski, T.M Krygovski, J Shorter (Eds.), Similarity Models in Organic Chemistry, Biochemistry and Related Fields, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1991, p 317 [55] M Jansson, B Jonsson, J Phys Chem 93 (1989) 1451 [56] R.T Buwalda, M.C.A Stuart, J.B.F.N Engberts, Langmuir 16 (2000) 6780 [57] N.O Mchedlov-Petrossyan, G.P Gorbenko, N.A Vodolazkaya, V.I Alekseeva, L.P Savvina, Funct Mater (2000) 138 [58] C Rottmann, D Avnir, J Am Chem Soc 123 (2001) 5730 [59] N.O Mchedlov-Petrossyan, S.A Shapovalov, V.L Koval, T.A Shakhverdov, Yu.A Bochkaryov, Dyes and Pigments 19 (1992) 33 [60] R Allmann, Z Kristallogr 128 (1969) 115 [61] S Hogiu, J Dreyer, M Pfeiffer, K.-W Brzezinka, W Werncke, J Raman Spectrosc 31 (2000) 797 [62] C Machado, M da Graca Nascimento, M.C Rezende, A.E Beezer, Thermochim Acta 328 (1999) 155 N.O Mchedlo6-Petrossyan et al / Colloids and Surfaces A: Physicochem Eng Aspects 205 (2002) 215–229 [63] N.O Mchedlov-Petrossyan, Yu.V Isaenko, O.N Tychina, Zh Obsh Khim (Russ J Gen Chem.) 70 (2000) 1963 [64] T.M Krygowski, P.K Wrona, U Zielkowska, C Reichardt, Tetrahedron 41 (1985) 4519 229 [65] K Kano, Y Ueno, S Hashimoto, J Phys Chem 89 (1985) 3161 [66] K Shinoda, T Nakagawa, B.-I Tamamushi, T Isemura, Colloidal Surfactants (Russian translation) Moscow, Mir, 1966