POLYMER/SURFACTANT INTERACTION 31 1.O 0.5 I i -4.5 -3.5 LOG Mi• Figure 7. Binding isotherms of polyacrylate (5 X 10 -4 M): comparison of alkyl-pyridinium and -tri- methylammonium ions. (/•) C•2PyCI (O) C•4PyBr (&) C•2TABr (O) C•4 TABr (18). group is replaced by a CF 3 group) showed a characteristic NMR shift (8) when this surfactant underwent micellization, i.e., when the CF 3 groups experienced a change from an aqueous environment to that of the micelies. In the presence of PEO (constant level) and at low F3SDS concentration, the chemical shift was the same as that of submicellar concentrations of F3SDS (21). At a certain concentration ("T•"), lower than the CMC of F3SDS, a slope change in the 8 vs. reciprocal concentration plot was ob- served. At a second concentration ("T2"), above the CMC, the slope changed again to a value close to that of polymer-free miceliar F3SDS solutions. T• was found to be independent of polymer concentration, while T2 increased with it. In other words, the results obtained are in good agreement with the behavior observed by other methods, e.g., surface tension and dialysis. The calculated binding ratio was 0.25 mol F3SDS/ base mol PEO, also in good agreement with the value obtained by direct binding studies. Results for PEO 7000 and 20,000 were essentially the same, but interaction was less pronounced for PEO 1500. The similar values of 8 for the polymer-bound surfactant and micellized surfactant confirmed again the concept of the surfactant mole- cules existing as aggregates in the former state. In an extensive study of the PEO/SDS system, Cabane investigated the •3C NMR shifts of the different carbons in SDS on adding increasing amount of PEO to miceliar solu- tions (22) of SDS. Substantial chemical shifts were noted for the three carbons (C•, C=, C 3) closest to the sulfate headgroup but were virtually absent for carbons C 4 through C•=. Furthermore, the shifts for C•, C=, and C 3 were linear up to a concentration of added PEO corresponding to the T= condition, i.e., where the polymer is just saturated with surfactant. When the experiment was done in reverse, i.e., SDS was added in increasing amounts to a fixed concentration solution of PEO, a linear shift in the •3C line of PEO was observed up to a concentration close to the T 2 value for the system.

32 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS 10 • 1.O ß • O.1 C C SPSPSP P //SP SP SPCC C C SPSPSP P/ P SP C C C C SP SP P// P T C C CC SPSP P/ P T CC - / - / - / - / - / / / I I I I I llll 0,• I I C- CLEAR T-TURBID P= PRECIPITATE SP- SLIGHT PRECIPITATE I '1 IIIII I • • I i II! 1 .O 10 TEALS, WT % Figure 8. Solubility diagram of polyquaternium 10/triethanolamine lauryl sulfate (TEALS) system (12). One can draw a number of conclusions from Cabane's work. First, the environment experienced by carbons C 4 through C•2 of the surfactant in the polymer/surfactant ag- gregate is indistinguishable from that in regular SDS micelies, suggesting that the aggregates themselves are modified micelies. On the other hand, the NMR data show that carbons C• through C 3 of SDS in the polymer/surfactant aggregate are in a different environment, and Cabane suggests that EO groups of the polymer replace water mole- cules in the outer region of the micelies. These ideas will be discussed later in the article. GEL FILTRATION A direct and convenient way to obtain information on the formation and properties of polymer/surfactant complexes involves the use of gel permeation. Sasaki et alo (23) in-