J . Chem. Soc., Faraday Trans. I, 1984, 80, 73-86 Raman Studies in Micellar Sodium Octyl Sulphate Solutions BY MURRAY H. BROOKER* Chemistry Department, Memorial University of Newfoundland, St John's, Newfoundland, Canada A1B 3x7 AND DAVID J. JOBE AND VINCENT C . REINSBOROUGH Chemistry Department, Mount Allison University, Sackville, New Brunswick, Canada EOA 3CO Received 13th April, 1983 Raman spectra of micelle-forming sodium octyl sulphate have been measured as a function of concentration over the critical micelle concentration. Accurately measured frequency shifts of the C-H stretching region and relative intensities at 1080 and 3125 cm-l characteristic of gauche isomeric forms can be used as indicators of micelle formation. Quantitative Raman measurements can be used to measure solubilization of organic additives to micellar solutions.Studies of the effects of cations on the halfwidth of the 0-SO, symmetric stretching vibration indicate a strong interaction with Ca2+, perhaps even complex formation. This blocking effect may account for the decreased catalytic ability of micelles in the presence of Ca2+. Investigations in micellar systems have utilized almost every physicochemical technique available to throw light upon the properties and structures of micelles and the process by which micelles effect the solubilization of a wide variety of substances. One method that has received scant attention in this regard is Raman spectroscopy: yet in its latest refinements this technique is proving to be a sensitive and versatile indicator of changes in the microenvironments of solutes.Accurate measurements of peak maxima, peak halfwidths and band intensities can provide information about the average local field, the vibrational relaxation time (as affected by the dynamic local field) and species concentrations, respective1y.l Current techniques are largely limited to concentrations above 0.05 mol dm-3 but improved instrumentation and computer- based signal-averaging methods could lower this limit by two orders of magnitude into the range of concentrations most useful for micelle studies. The present study was initiated with three goals in mind. (i) Micelle formation: studies were performed to determine which, if any, parameters of the Raman spectrum could serve as a sensitive probe of micelle formation. Previous studies had indicated that relative intensity changes between the solid and aqueous states or as a function of temperature could be employed to follow conformational changes of the surfactant alkyl (ii) Solubilization of additives: it is well known that the solubility of organic solutes in water is significantly increased by the presence of micelles, and previous Raman studies7t8 were able to detect solute peaks.In the present study the effect of sodium octyl sulphate on the solubility of benzene in water was measured quantitatively. (iii) Effect of cations : the presence of anionic micelles has been shown to have a catalytic effect on certain ligand exchange reactions which are believed to be associated with the high charge density at the anion head group. The effects of several cations on the Raman spectra were measured to investigate this phenomenon.7374 RAMAN STUDIES OF MICELLAR SDS EXPERIMENTAL Sodium n-octyl sulphate (SOS) and sodium n-lauryl sulphate (SLS) were obtained from Eastman Kodak and were reported to be 99% pure with respect to other alkyl isomers. Reagent-grade CaC1, - 4H,O, NiC1, * 6H,O, NaCl and MgC1,. 6H,O and spectroscopic-grade benzene, n-hexane and butanol were used without further purification. Solutions were prepared by weighing the appropriate amount of solid into volumetric flasks. Solutions were treated with small amounts of activated charcoal to remove fluorescent impurities and filtered through fine frits to remove particular matter which would cause a bothersome stray-light background. The benzene-saturated water and benzene-saturated 0.50 mol dm-3 SOS solution were prepared by vigorously shaking a mixture which contained an excess of benzene and allowing the two phases several hours to separate before drawing off the appropriate layer with a syringe and recording the spectrum immediately.Solids were measured in thin-walled capillary tubes at 77 and 298 K while solutions were measured in 1 cm diameter Pyrex test-tubes at 298 K. Raman spectra were obtained with a Coderg PHO spectrophotometer equipped with photon counting and digital output. Sample excitation was achieved with either the 488.0 or 514.5 nm line from a Control model 553A argon-ion laser operating at ca. 800 mW. Narrow-bandpass interference filters were employed to remove unwanted plasma lines but these same plasma lines were used as required for frequency calibration. The standard 90° angle between the incident and scattered light was employed throughout.Two polarization techniques were used. Method one, which involved changing the polarization of the incident light with a half-wave plate, was used for rough depolarization measurements, X ( Z 5 ) Y and X( Yg) Y orientations. Method two involved placement of Polaroid analysers to accept scattered light polarized parallel or perpendicular to the incident light, X(Zz> Y and X(ZX) Y orientations. A quarter-wave plate before the slit compensated for the different response of the gratings for the two polarizations of light. Method two was employed for accurate depolarization studies.Most spectra were recorded in the normal recorder mode, but selected spectra in the 900-1 160 and 2800-3000 cm-l ranges were collected digitally and stored on a PDP 11/70 computer. Spectra obtained in this manner could be signal-averaged through repetitive scans, smoothed with a 7 point Savitsky-Golay smoothing program, baseline-corrected and curve-resolved as required. RESULTS AND DISCUSSION Raman frequency, polarization data and tentative assignments are collected in table 1 for the SOS and SLS salts in both the solid and aqueous phases. Data are presented for 0.5 mol dmP3 micellar solutions; the effect of dilution will be discussed below. Peak positions are in reasonable agreement with previously published although a number of discrepancies were noted.Surprisingly, previous workers did not report depolarization measurements. Both the polarized and depolarized spectra of SOS are shown in fig. 1. Most of the peaks of the alkyl chain are strongly polarized and must be due to symmetric vibrations. The peak at 1128 cm-l has previously been assigned to an antisymmetric C-C stretch but the depolarization ratio of ca. 0.4 puts this assignment in doubt. Peaks at ca. 1080 and ca. 1300 cm-l and the 1400-1470 cm-l envelope are clearly depolarized. The depolarized peaks at 420-460 and 580-630 cm-l and the strongly polarized peak at 1063 cm-' are due to the sulphate head groups and are similar to observations of HSO;.l. The polarized peak at ca. 827 cm-l appears to be due to the C-0-SO; stretch. Except for the symmetric stretching mode of -0-SO; the peak frequencies do not differ significantly from the solid to aqueous phase.The symmetric stretching mode of -0-SO; occurs at ca. 1063 cm-l in the aqueous phase but at 1084 cm-l in the solid salts. Attempts to measure the effects of micelle formation on the C-C stretching vibrations in the 1020-1 150 cm-l region were complicated by the presence of the intense -0-SO; peak and special efforts were required to identify and measure peaks in this region.75 M. H. BROOKER, D. J. JOBE AND V. C. REINSBOROUGH Raman studies of n-hexane indicate that the -CH,-CH, chain has peaks at ca. 1010 p, 1040 p, 1066 dp, 1080 dp and 1140 p (fig. 2). In solid SOS and SLS the -O-SO, peak tends to mask the C-C stretching peak at ca. 1080 cm-l, although peaks at ca.1076 (sh), SOS, and 1081 cm-l, SLS, could be resolved in the solid at 77K along with a peak at 1063 cm-l due to another C-C stretching vibration (fig. 2). For the aqueous micellar solution, the 1081 cm-l C-C stretching peak appears as a shoulder on the 1063 cm-l 0-SO, stretching peak in the polarized spectrum but as a distinct peak in the depolarized spectrum (fig. 1 and 3). Intensity from a C-C vibration probably contributes to the 1063 cm-l peak intensity and may provide most of the intensity in the depolarized spectrum (fig. 3), since the depolarization ratio of the 0-SO, symmetric stretching vibration is expected to be much smaller than the measured value of 0.10. The presence of the 1045 cm-l peak for the aqueous micellar solution can only be inferred from attempts to curve-fit the 1000-1 150 cm-l envelope (table 1).The C-H stretching regions of aqueous SOS and SLS were essentially totally polarized and must be due to symmetric stretching vibrations (fig. 4). Previouslys the peak at 2930cm-l was assigned to an antisymmetric stretching vibration, but this cannot be correct. In the depolarized spectrum the most intense peak (2897 cm-l) does not even coincide with a peak in the polarized spectrum. Several sharp peaks on the C-H profile of SLS reported by Kalyanasundaram and Thomas3 were not observed in this work (fig. 5) and would appear to be due to plasma lines superimposed on their spectra. Studies of the effect of SOS and SLS on the water spectrum revealed almost no change from the spectrum of pure water.The hydrogen-bond peak at ca. 180 cm-l is reasonably sensitive to structure-breaking effects;' therefore the present results indicate that the micelle does not disrupt the water structure significantly. MICELLE FORMATION The critical micelle concentration (c.m.c.) for SOS is 0.1 15 mol drn-,,1° so the concentration range chosen for this study was 0.05-0.50 mol dm-3. The 0.05 mol dm-3 concentration represented the practical lower limit for reasonably resolved spectra (4 cm-l resolution) recorded with a single scan. Studies of micelles with lower c.m.c. would require signal averaged spectra. The spectra of SOS above and below the c.m.c. showed only subtle differences. The most significant change in the spectrum occurred in the C-H stretching region from 2850 to 2960 cm-l, where all the peak maxima shifted by 5 cm-l to lower values from the dilute solution of monomers to the micellar solution (fig.4 and 6). The plot of the peak maximum against concentration for several of the peaks indicated that the shift occurred fairly abruptly over a 0.1 mol dm-, concentration range (fig. 7). Similar results were obtained for the shift which occurred for the C-H stretching region of butanol between pure butanol and a saturated butanol +water solution. These results suggest that in general one should expect a ca. 5 cm-l shift to lower frequency for the C-H stretch when an aqueous environment is replaced by a hydrocarbon environment. Small shifts of 1 or 2 cm-l in the same direction may also occur in the 1 128 and 108 1 cm-l peaks over the same concentration range (fig.3). Previous workers have not measured the peak frequencies with significant precision to detect the above shifts. It would appear that even these small changes could be used to follow micelle formation. In the context of the present controversy regarding the degree of hydration of surfactant chains within micellesll* l2 it would thus appear that in the picosecond observation range of the Raman technique the carbons of the octyl sulphate chain are predominantly not wetted in the micelle. It must be stressed that sodium octyl sulphate micelles are relatively unstable andTable 1. Raman peak positions (cm-l) for sodium octyl sulphate and sodium lauryl sulphatea sodium octyl sulphate sodium lauryl sulphate 0.5 mol dm-, (aq) solid, 298 K 0.5 mol dm-3 (as) solid, 298 K assignment ca.180 br, dp 350 w, p 422 m, dp ca. 460sh,dp 510, 526 w, p 582 m, dp 626 w, dp 735 w, p 767 w, p 810 m, p 826 m, p 846 m, p 877 m, p 895 m, p 930 w, p 968 w, p 1016 m, p ( 1045)b w, p (1060y w, dp 1063 s, p 1081 w, dp - ca. 360w 422 m ca. 430 - 579 m 595 w, 629 w 760 br 812 m 846 m 892 m 928 m 964 w 1014 w 1040 w 1063 m 1084 s ca. 1076 sh - - - ca. 180 br, dp 360 w, p 420 m, dp - - 584 m, dp ca. 620w,dp 735 w, p 765 w, p 827 m, p 867 m, p 890 m, p - - - - ca. 1015w,p - (1060)c w, dp 1063 s, p ca. 1080 sh, dp - ca. 360w 406 w, 420 m 503 w ca. 575 m, 598 m ca. 630 w 770 w - - - - 837 m 867 890 m - - 1020 w 1063 m 1085 s 1081 - H,O hydrogen bond CCO deformation -SO; rock -SO,- deform C-O-S03- str CH, rock CH, rock CH,-0 CH, twist C-C sym str 0-SO,- sym str C-C sym str1128 m, p (partly) 1152 w, p 1190 w, p 1218 w, dp 1250 w, dp 1305 m, dp 1370 w, p 1392 w, p 1441 m, dp 1455 m, dp 1470 sh, dp 1637 br, p 2859 s, p 2875 vs, p 2904 s, p 2934 s, p 2963 m, p ca.3250vs,p CQ. 3410vs,p - - 1128 m 1150 sh, w 1196 w 1218 w 1262 w 1300 m - ca. 1400 w 1440 m 1456 m 1470 sh 2850 s 2862 s 2876 s 2885 s 2902 s 2914 s 2934 s 2962 m - 1128 m, p 1145 sh, p 1129 m ca. 1150 w C-C str - 1294 m, dp 1370 w, p 1395 w, p 1437 m, dp 1453 m, dp 1468 sh, dp 1637 br, p 2854 vs, p 2875 s, p 2901 s, p 2931 s, p 2962 m, p ca. 3250vs, p ca. 3410 vs, p - - - - 1296 m 1373 w 1436 m 1455 m - - 2862 m 2882 vs 2897 s 2934 m 2960 m - - - 1 0-SO; antisym str CH, twist CH2 scissor v 2 H2O CH2 C-H sym str and C-H sym str of CH, P ? a !z ?i U 5 0 a p, polarized peak; dp, depolarized peak; w, weak; m, medium; s, strong; br, broad; sh, shoulder.Inferred from curve-resolution attempt. Inferred from studies of other phases.78 RAMAN STUDIES OF MICELLAR SDS loosely packed, so that in micelles where the chains are longer there should be even less hydration within the micelle. Previous studies of alkyl carboxylates and alcohols have shown that the relative intensity changes between solid and aqueous solution and as a function of temperature for the 1040-1 140 cm-l region and that the 2800-3000 cm-l region can be used to 1305 1081 I I I I r I I I I I I I I 1 I I 1600 1400 1200 1000 800 600 LOO 200 wavenumber/crn-' Fig. 1. Raman spectra of 0.50 mol dm-3 SOS obtained with the 514.5 nm line at 4.0 cm-' slit width for the parallel (a) and perpendicular (b) polarizations. estimate the relative concentrations of trans and gauche isomer~.~-~ For instance a decrease in the relative intensities of peaks at ca.2925 and 1080cm-l has been associated with an increase in the trans isomer and a decrease in the gauche isomer. Comparison of the dilute SOS monomer solution (0.070 mol dm-3) to that of the SOS micellar solution (0.50 mol dm-3) indicated a decrease in relative intensities at 2925 and 1081 cm-l for the micellar solution (fig. 6). These results indicate that the more randomly ordered gauche form of t5e alkyl chain is more stable in a dilute aqueous environment and that formation of micelle aggregates favours the more ordered trans isomers.For both the studies it was necessary to record the spectra digitally and to apply a baseline correction to the data so that the spectra could be directly super- imposed for comparison. The computer-collected spectra for the C-H region also show the ca. 5 cm-1 offset required to overlay the maxima (fig. 6). Unfortunately the symmetric stretching frequency of 0-SO; at 1063 cm-l dominates the polarized spectrum in the C-C region and, although the 1081 cm-l peak can be seen as a shoulder on the 1063 cm-l peak (fig. 8), it was difficult to measure relative intensity changes with concentration. Fortunately the 0-SO; band is completely polarized while the 1081 cm-l band is depolarized, and better resolution is possible in the depolarized spectrum (fig.1 and 3). Because the intensities of the peaks were weak the measurements were made for 10 scans smoothed and averaged for both the dilute and concentrated SOS solutions. The results (fig. 3) clearly indicated that the 108 1 cm-l peak decreased in relative intensity for the 0.50 mol dm-3 micellar solution.M. H. BROOKER, D. J. JOBE AND V. C. REINSBOROUGH 19 1200 1100 1000 w avenum berlcrn-' Fig. 2. Raman spectra of (a) n-hexane liquid at room temperature, (b) solid SOS at 77 K and (c) solid SLS at 77 K. The 514.5 nm laser line and 2.0 cm-l slit widths were employed. SOLUBILIZATION OF BENZENE Organic molecules can have greatly enhanced solubility in micellar solutions com- pared with pure water, It is desirable to establish techniques to measure the solubility directly and to establish the micellar site of solubilization.To obtain information on this subject studies were performed on saturated solutions of aqueous benzene and aqueous benzene in 0.50 mol dm-3 SOS. At 25 OC the solubility of benzene in water is 0.01 1 mol dm-3 and the two major peaks of benzene, the 991 cm-l ring-breathing80 RAMAN STUDIES OF MICELLAR SDS and the 3070 cm-l C-H stretch, were easily observed in the Raman spectrum (fig. 9). The 991 cm-I peak occurs at the same frequency for benzene in water as for pure benzene but the C-H stretching peak at 3072 cm-l is ca. 8 cm-l higher for benzene in water than for pure benzene (3062 cm-l). Again the C-H stretching frequency is sensitive to whether the environment is aqueous or hydrocarbon.The presence of wavenumber/cm-' Fig. 3. Depolarized Raman spectra, X(ZX) Y orientation, for (---) 0.07 mol dm-3 and (-) 0.50 mol dm-3 SOS to illustrate the intensity decrease of the 1081 cm-' peak and slight frequency shifts which accompany micelle formation. Spectra are smoothed averages of 10 scans which have been baseline corrected. Spectra were obtained with the 488.0 nm line and 2.0 cm-l slits. 0.5 mol dm-3 SOS greatly enhanced the solubility of benzene (fig. 9). Measurement of the relative intensity of the benzene peaks to the water peaks at 3400 and 1637 cm-l were used to measure a 7.0 times increase in the solubility of benzene in 0.50 mol dm-3 SOS over that of pure water. Jobe et all3 found a roughly ten-fold increase in benzene solubility at the same SOS concentration from ultraviolet studies but only after several days of equilibrium.It usually requires at least a day before maximum uptake of solubilizate is achieved in micellar solutions. Since we were not particularly interested in benzene solubilities, this aspect of the work was not pursued further. However, it is clear that measurements of relative intensities of Raman lines could easily be utilized to monitor solubilizate concentrations. More precise quantitative measure- ments could be achieved if an internal standard anion such as NO; or ClO, were employed since these ions have peaks which are more easily measured than those of water. Evidence that the benzene dissolves in the micelle and is removed from bulk water comes from the fact that the C-H stretching frequency of benzene in 0.50 mol dmP3 SOS at 3062 cm-l corresponds to the value in pure benzene and not to the value for benzene in pure water (3072 cm-I).Other techniques which throw light on solubiliza- tion sites in micelles confirm that benzene in concentrations > 0.01 mole fraction is principally located in the micellar ~ o r e . ' ~ - ' ~ The Raman results are in agreement. Since similar results were obtained for both the 488.0 and 514.5 nm laser lines it wouldM. H. BROOKER, D. J. JOBE AND V. C. REINSBOROUGH 81 6 I - - cm-' 2897 x 2 3 000 2900 wavenumber/cm-' 2 000 Fig. 4. Raman spectra for the C-H stretching region for SOS solutions of (a) 0.07 and (6) 0.50 mol dm-3 concentration to illustrate frequency shifts and small depolarization ratio. Spectra were obtained with the 488.0 nm line and 2.0 cm-l slits.appear that the Raman signal due to benzene was not resonance enhanced by interactions with the hydrocarbon environment as has been reported for anthracene by Beck and Brus.ls EFFECTS OF IONS ON sos It has been shown that micelles serve as catalysts in ligand exchange reactions and it is generally believed that the high charge density at the polar micelle surface enhances the ligand exchange of reacting ~ati0ns.l~ In order to investigate this phenomenon we have studied the effect of various cations on the spectrum of SOS. Particular emphasis was placed on the 0-SO; symmetric stretch because studies of aqueous HSO, and SO:- have demonstrated that the frequency shifts, halfwidth82 500 - 400 - 300- 200 - 100- I 0 RAMAN STUDIES OF MICELLAR SDS I I 2882 2897 9 I , l , , , , l , , , , ~ ( , L 2950 2850 2700 wavenum berlcm-' Fig.5. Raman spectra of the C-H stretching region for (a) solid SLS and (b) solid SOS, at room temperature. Spectra were obtained with the 514.5 nm line and 2.0 cm-l slits. 600 4 IM. H. BROOKER, D . J. JOBE AND V. C. REINSBOROUGH -2.0 IB I!, -L.O -6.0 83 + * 0 A + - 1 5 + - 0 0 - + A + Q 0 0 Q * I - I 0.1 0.2 0.3 0.4 concentrationlmol dm-3 Fig. 7. The frequency difference (v-v,,) between the peak maxima of C-H vibrations from the dilute solution value, vD, and the value observed at a specific concentration, V, plotted against concentration of SOS: 0, 2864; +, 2879; *, 2941; A, 2069 cm-l; 0, all points. I ~ I I I I ~ I I I I ~ I I I I ~ I I I 1150 1100 1050 1000 wavenum ber/cm -l Fig.8. Raman spectra of the 1000-1150 cm-l region of (a) 0.070 mol dm-3 SOS, (b) 0.50 mol dm-3 SOS and (c) 0.50 mol dm-3 SOS + 2.5 mol dm-3 CaC1,. Spectra were obtained with the 488.0 nm line and 2.0 cm-l slit. 4 FAR 184 RAMAN STUDIES OF MICELLAR SDS 3062 'L A I t , , , I , , , , l , , , , I , , , , I , , 31 00 3050 11 00 1000 wavenumber/cm-' Fig. 9. Portions of the Raman spectra for (a) benzene-saturated and 0.50 rnol dm-3 SOS and (b) benzene-saturated water. Spectra were obtained with the 488.0 nm line at 2.0 cm-l slits. Table 2. Peak positions and halfwidths for the symmetric stretching frequency of aqueous sodium octyl sulphate with different added salts (A) SOS, 0.070 mol dmP3 1063.0 13.5 B+2.5 mol dmP3 HC1 1063.0 14.5 NaCl 1 064.5 14.0 MgCl2 1064.0 12.5 NiC1, 1064.0 13.2 CaCl, 1065 .O 17.0 (B) SOS, 0.50 rnol dm-3 1063 .O 12.0 changes and even doubling of peaks due to complex formation can be used to gain information about the ionic interactions.' In the present study the halfwidth and peak frequencies were measured for solutions of 0.070 mol dm-3 SOS, 0.5 mol dm-3 SOS and 0.5 mol dm-3 SOS plus 2.5 mol dm-3 HCl, NaCl, MgCl,, NiCl, and CaCl,.The results are presented in table 2. Only the symmetric stretching 0-SO; peak at 1063 cm-l exhibited a measurable change due to the addition of salt. The presence of added salt caused a small increase in peak position and halfwidth. In studies of other systems it has been found that frequencies of concentrated solutions tend toward the values of the solid salts, i.e.1084 cm-l for pure SOS solid. Halfwidth changes ofM. H. BROOKER, D. J. JOBE AND V. C. REINSBOROUGH 85 the isotropic polarized symmetric stretching frequency will be primarily due to the vibrational dephasing effects of elastic collisions due to changes in the environment about the 0-SO; group. Note that the concentration increase from 0.070 to 0.50 mol dm-3 was accompanied by a 1.5 cm-l halfwidth decrease from 13.5 to 12.0 cm-l. This could be interpreted to indicate a less dynamic environment about the 0-SO; micellar head groups. The magnitude of the effect of cation on the 0-SO; head group as measured by the peak frequencies and halfwidth changes follows the order: Mg2+ < Ni2+ < H+ < Na+ < Ca2+.Calcium exhibits the greatest influence on both the halfwidth and peak maximum and attests to the strong interaction between Ca2+ and the 0-SO; head group. The high-frequency asymmetry of the peak (fig. 8) may be an indication that the Ca2+ resides on the micelle head group in an inner-sphere-type interaction long enough to create a second type of 0-SO; environment in equilibrium with the uncomplexed site. Hicks and Reinsboro~ghl~ have found that the presence of small amounts of Ca2+ significantly reduced the catalytic effect of micelles for the ligand exchange reaction of Ni(H,O);+ with pyridine-2-azo-p-dimethylaniline (PADA), whereas Mg2+ and Na+ additions of the same scale had little effect. Calcium ion apparently competes much more effectively than nickel ion for adsorption into the Stern layer of the micelles.The present Raman results corroborate this view. The very small effect of Mg(H20)2+ and Ni(H,O);+ is not unexpected since the ions are strongly hydrated and will not readily lose the hydrated water, which would be a necessity for the Mg2+ or Ni2+ to form an inner-sphere complex with the head group. Peaks due to the M-0 symmetric stretching vibration of the M(H,O);+ complex were observed in the present study at 355 and 395 cm-l for the Mg(H,O);+ and Ni(H,O);+ complexes. No similar bands have been reported for aquated Ca2+ and Na+ ions, a fact which attests to the weaker hydration of these i0ns.l These results are similar to complex formation studies of NO; and NO; where Ca2+ accepts these weak ligands much more readily than does Mg(H,O);+ or Ni(H,O);+.l The effect of HCl(aq) was very small but one could predict a greater effect if the H+ concentration were increased to > 10 mol dm-3 since at this concentration the proton would be forced onto the 0-SO; group.The effect of Ca2+ on the SOS solutions was also studied as a function of concentration of SOS to concentrations below the c.m.c. of the pure SOS (0.1 15 mol dm-3). The peak positions and halfwidths of the SOS for a 2.5 rnol dm-3 CaCl, concentration were unaffected by decreasing SOS concentrations as low as 0.08 mol dm-3. These results indicated that the Ca2+ - - - OSO; interaction was not affected by the SOS concentration but it was not possible to determine whether the Ca2+ - * 0-SO; interaction was the same for monomer and micellar forms of SOS because as indicated by the C-H stretching frequencies the presence of 2.5 mol dm-3 CaCl, had lowered the c.m.c.below 0.08 mol dm-3 and prevented the study of Ca2+ with monomer, Generous support of the National Science and Engineering Research Council of Canada for separate grants to M. H. B. and V. C. R. is gratefully acknowledged. 4-286 RAMAN STUDIES OF MICELLAR SDS D. E. Irish and M. H. Brooker, in Advances in Infrared and Raman Spectroscopy, ed. R. J. H. Clarke and R. E. Hester (Heyden, New York, 1976), vol. 2, chap. 6, p. 212. J. L. Lippert and W. L. Peticolas, Proc. Nut1 Acad. Sci. USA, 1971, 68, 1572. K. Kalyanasundaram and J. K. Thomas, J. Phys. Chem., 1976,80, 1462. H. Okabayashi, M. Okuyama, T. Kitagawa and T. Miyazawa, Bull. Chem. SOC. Jpn, 1974,47, 1075. H. Okabayashi, M. Okuyama and T. Kitagawa, Bull. Chem. SOC. Jpn, 1975,48, 2264. K. Larsson, Chem. Phys. Lipids, 1972,9, 181. J. B. Rosenholm, K. Larsson and N. Dinh-Nguyen, Colloid Polym. Sci., 1977, 255, 1098. T. Takenaka, K. Harada and T. Nakanaga, Bull. Inst. Chem. Res. Kyoto Univ., 1975, 53, 173. H. Chen and D. E. Irish, J. Phys. Chem., 1971, 75, 2672; D. E. Irish and H. Chen, J. Phys. Chem., 1970, 74, 3796. lo P. D. I. Fletcher and V. C. Reinsborough, Can. J. Chem., 1981, 59, 1361. l1 H. Wennerstrom and B. Lindman, J. Phys. Chem., 1979,83, 2931. l* F. M. Menger and B. J. Boyer, J. Am. Chem. SOC., 1980, 102, 5936. l3 D. J. Jobe, V. C. Reinsborough and P. J. White, Can. J. Chem., 1982, 60, 279. l4 J. R. Hicks and V. C. Reinsborough, Surfactants in Solution, ed. K. L. Mittal and B. Lindman l5 S. J. Rehfeld, J. Phys. Chem., 1970, 74, 117. l6 S. M. Beck and L. E. Brus, J. Chem. Phys., 1981, 75, 1031. l7 J. R. Hicks and V. C. Reinsborough, unpublished results. (Plenum Press, New York, in press). (PAPER 3/590)