Rate-Dependent Lowering of Surface Tension during Transformations

Bolaform to Monomeric Structures. Lana I. Jong and Nicholas L. Abbott*. Department of Chemical Engineering and Materials Science, University of Califo...
0 downloads 0 Views 64KB Size
Langmuir 1998, 14, 2235-2237

2235

Rate-Dependent Lowering of Surface Tension during Transformations of Water-Soluble Surfactants from Bolaform to Monomeric Structures Lana I. Jong and Nicholas L. Abbott* Department of Chemical Engineering and Materials Science, University of California, Davis, California 95616 Received February 26, 1998 We report the synthesis of water-soluble, bolaform surfactants containing disulfide bonds (Br-N+(CH3)3(CH2)8SS(CH2)8N+(CH3)3Br-, 1) and measurements of the surface tensions of aqueous solutions of these surfactants during their reduction to monomeric, thiol-containing fragments (Br-N+(CH3)3(CH2)8SH, 2). Surface tensions measured during the redox-induced transformation of 1 to 2 were found to be lower, by up to 10 mN/m, as compared to surface tensions of solutions of 1 or 2. The extent of lowering of the surface tension was dependent on the rate of transformation of 1 to 2. These results and others suggest that, during the transformation of 1 to 2 at the surface of the solution, slow desorption of 2 leads to a transient accumulation of 2 at the surface of the solution and thus to low surface tensions. We believe that redox-induced transformations of surfactants may form the basis of a general principle to create low (transient) surface tensions.

We report measurements of the time-dependent surface tensions of aqueous solutions during redox-induced transformations of water-soluble surfactants from bolaform (PRRP, where P is a polar headgroup and R is a hydrophobic chain) to monomeric (RP) structures. Whereas past studies have demonstrated that the surface tensions of aqueous solutions measured before and after a redoxinduced change in the structure of a surfactant can differ by as much as 23 mN/m,1 the time-dependent evolution of the surface tension during the transformation was not investigated. The results reported herein demonstrate that the surface tensions measured during redox-induced transformations of bolaform surfactants to monomeric ones can be substantially lower (∼10 mN/m) than the initial or final (equilibrium2) surface tensions of the system and that the extent of lowering of the surface tension is dependent upon the rate of the redox-induced transformation of PRRP to RP. These observations suggest that redox-induced transformations of surfactants can be exploited to create low (transient) surface tensions.3 The study reported in this paper is based on measurements of surface tension during the reduction of a disulfidecontaining, bolaform surfactant (1) to less surface-active, monomeric, thiol-containing fragments (2).4 Watersoluble surfactants containing disulfide bonds are a * To whom correspondence should be addressed. E-mail: [email protected]. (1) Electrochemical oxidation and reduction of ferrocene-based surfactants in aqueous solution can lead to large and reversible changes in surface tension: (a) Gallardo, B. S.; Hwa, M. J.; Abbott, N. L. Langmuir 1995, 11, 4209-4212. (b) Bennett, D. E.; Gallardo, B. S.; Abbott, N. L. J. Am. Chem. Soc. 1996, 118, 6499-6505. (c) Gallardo, B. S.; Metcalfe, K. L.; Abbott, N. L. Langmuir 1996, 12, 4116-4124. (2) We use the term “equilibrium surface tension” to mean the surface tension of the system in the absence of an ongoing chemical reaction. See also ref 14 below. (3) Contact of a crude oil containing natural acids and an alkaline aqueous phase can give rise to transient minima in oil/water interfacial tensions. See, for example: Rubin, E.; Radke, C. J. Chem. Eng. Sci. 1980, 35, 1129-1138. Rudin, J.; Bernard, C.; Wasan, D. T. Ind. Eng. Chem. Res. 1994, 33, 1150-1158, and references therein. (4) Water-soluble, disulfide-containing, cationic surfactants have been reported in the past. Measurements of the surface tensions of aqueous solutions of these surfactants during their reduction to thiol fragments have not been described. Pinazo, A.; Diz, M.; Solans, C.; Pes, M. A.; Erra, P.; Infante, M. R. J. Am. Oil Chem. Soc. 1993, 70, 37-42.

potentially useful class of amphiphiles because disulfide bonds can be reduced chemically,5 electrochemically,6 and photochemically,7 can participate in thiol-disulfide exchange reactions,8 and can be processed enzymatically by thiol-disulfide oxidoreductases.9 Because previously investigated lipids10 and nonionic amphiphiles11 containing disulfide bonds have been sparingly soluble in water, these amphiphiles are not useful for studies of surfactancy in aqueous systems. Surfactant 1 is, in contrast, watersoluble and easy to prepare, and is used herein to demonstrate that a redox-induced change of 1 to 2 can create transient surface tensions that are substantially lower than any equilibrium surface tensions of the same system. Surfactants 1 and 2 (Chart 1) were synthesized according to procedures described in the Supporting Information. Figure 1 shows surface tensions of aqueous solutions (50 mM phosphate buffer, pH 6.9, deoxygenated)12 of 1 and 2 measured by using a maximum bubble pressure tensiometer (SensaDyne, Mesa, AZ) bubbled at 0.1-0.3 Hz and 25 °C.13,14 The threshold concentration of surfactant causing a measurable reduction in surface tension was found to be an order of magnitude lower for 1 (∼0.1 mM) as compared to 2 (∼1 mM). Reduction of 1 (5) Lamoureux, G. V.; Whitesides, G. M. J. Org. Chem. 1993, 58, 633-641. (6) Zagal, J.; Herrera, P. Electrochim. Acta 1985, 30, 449-454. (7) Barltrop, J. A.; Hayes, P. M.; Calvin, M. J. Am. Chem. Soc. 1954, 76, 4348-4367. (8) Samuel, N. K. P.; Singh, M.; Yamaguchi, K.; Regen, S. L. J. Am. Chem. Soc. 1985, 107, 42-47. (9) Gravina, S. A.; Mieyal, J. J. Biochemistry 1993, 32, 3368-3376. (10) Ringsdorf, H.; Schlarb, B.; Venzmer, J. Angew. Chem., Int. Ed. Engl. 1988, 27, 113-158. (11) Cuomo, J.; Merrifield, J. H.; Keana, J. F. W. J. Org. Chem. 1980, 45, 4216-4219. (12) The surface tension of a solution of 2 prepared with deoxygenated buffer did not measurably change over a period of 8 h. (13) The upward curvature of the plot of surface tension of 1 shown in Figure 1 (for concentrations greater than ∼1 mM) may be caused by premicellar aggregation: Menger, F. M.; Littau, C. A. J. Am. Chem. Soc. 1993, 115, 10083-10090. (14) Surface tensions measured at bubble frequencies less than 0.1 Hz were similar to those measured at 0.1 Hz. The surface tensions of the reacting system reported in this paper correspond, therefore, to a local steady state at the surface of the solution. In the absence of reaction, these surface tensions are equilibrium ones.

S0743-7463(98)00235-2 CCC: $15.00 © 1998 American Chemical Society Published on Web 04/04/1998

2236 Langmuir, Vol. 14, No. 9, 1998

Figure 1. Surface tensions of aqueous solutions (50 mM phosphate buffer, 25 °C, pH 6.9) of 1 (9) and 2 (b). Chart 1. Structures of Bis(8-trimethylammonium bromide) Dioctyl Disulfide (1), (8-Mercaptooctyl)trimethylammonium Bromide (2), and Dithiothreitol

to 2 leads to changes in surface tension of up to 17 mN/m. Because the average area occupied by 1 at the surface of a 1 mM solution of 1 is estimated to be 117 Å2/molecule whereas the average area occupied by 2 at the surface of a 2 mM solution of 2 is estimated to be >500 Å2/molecule,15 we conclude that the overall increase in surface tension upon reduction of 1 to 2 is caused by desorption of surfactant from the surface of the solution (see below).16 Rates and equilibria of redox reactions involving thiols and disulfides have been measured in past studies of the autoxidation of cysteine-containing enzymes, protein folding, and the pharmacological and therapeutic actions of a number of drugs.17 Thiol-disulfide exchange is a base-catalyzed, SN2 displacement reaction in which the thiolate anion RS- is the nucleophile.18 In this paper, we report measurements of the rate of change of surface tension caused by thiol-disulfide exchange between 3 and 1; the exchange process results in formation of 2 and oxidized 3. We used 3 in our experiments for four reasons: (i) 3 is a strong reducing agent19 and thus the equilibrium lies toward 2;20 (ii) the formation of mixed (15) The molecular areas were estimated by using the Gibbs adsorption equation in the presence of swamping concentrations of electrolyte and by assuming ideal solution behavior in the bulk of the solution. For a discussion of this calculation, see ref 1c. (16) Dimeric, trimeric, and polymeric surfactants that possess disulfide bonds can, we believe, provide the capability to effect large changes in surface tension over wide ranges of concentration. See, for example: Gallardo, B. S.; Abbott, N. L. Langmuir 1997, 13, 203-208. (17) (a) Szajewski, R. P.; Whitesides, G. M. J. Am. Chem. Soc. 1980, 102, 2011-2026. (b) Gilbert, H. F. Adv. Enzymol. 1990, 63, 69-172. (c) Kim, P. S.; Baldwin, R. L. Annu. Rev. Biochem. 1990, 59, 631-660. (d) Woghiren, C.; Sharma, B.; Stein, S. Bioconjugate Chem. 1993, 4, 314318. (18) Singh, R.; Whitesides, G. M. In The Chemistry of the Thiol Group, Supplement; Patai, S., Rappoport, Z., Eds.; Wiley: New York, 1993. (19) (a) Cleland, W. W. Biochemistry 1964, 3, 480-482. (b) Jocelyn, P. C. Methods Enzymol. 1987, 143, 246-256. (20) Whitesides, G. M.; Lilburn, J. E.; Szajewski, R. P. J. Org. Chem. 1977, 42, 332-338.

Letters

Figure 2. Surface tensions of aqueous solutions (50 mM phosphate buffer at 25 °C) of 1 mM of 1 following the addition of 3: 20 mM of 3 at pH 6.9 (b); 20 mM of 3 with step change in pH from 6.9 to 4.0 (phosphoric acid) to 7.0 (sodium hydroxide) (O). Also shown is the surface tension of a 1 mM solution of CTAB following the addition of 10 mM of 3 (2).

disulfide products (HS-R3-S-S-R1) is small because oxidized 3 is a stable, cyclic disulfide;20 (iii) 3 and oxidized 3 are not surface-active;21 (iv) 3 has been reported to reduce water-insoluble, disulfide-containing lipids leading to the rupture of lipid membranes.10 Figure 2 (filled circles) shows measurements of the surface tension of aqueous solutions of 1 (1 mM, pH 6.9) following the addition of 3 (20 mM). The surface tension of a 1 mM solution of 1 was initially 57 mN/m. Approximately 10 min after the addition of 3, however, the surface tension decreased by ∼10 mN/m even though the product of the exchange reaction was confirmed to be species 2, a species that is less surface-active than 1.22 As the exchange reaction progressed, the surface tension passed through a minimum. At the completion of the reaction, the surface tension of the solution was 72 mN/m (i.e., surface tension of a solution of 2). Both the overall rate of reduction of 1 to 2 and the magnitude of the transient reduction in surface tension were observed to change with the concentration of 3.23 The addition of acid, which is known to quench thiol-disulfide exchange reactions by protonation of the thiolate anion,24 was observed to cause an abrupt (and reversible) increase in surface tension (open circles, Figure 2). No measurable change in surface tension followed the addition of 3 to a solution of cetyltrimethylammonium bromide (CTAB). Two additional observations lead us to conclude that the minimum in surface tension measured during the reduction of 1 to 2 is not an equilibrium property of a mixture containing 1 and 2 and is not a result of the surface activity of 3, oxidized 3, or intermediate chemical species (e.g., HS-R3-S-S-R1). First, the surface tensions of all solutions prepared by directly mixing 1 and 2 (not reacting) were greater than those for 1 alone (Figure 3). That is, the so-called “synergism” of surfactant mixtures25 does not explain our results. Second, the minimum in surface tension was observed when 1 was reduced to 2 by using either Na2SO3 or 3. We do not believe, therefore, that the transient lowering of surface tension is caused by the (21) The addition of 10 mM DTT and 1 mM oxidized DTT to water did not result in a change in surface tension. (22) The product of reduction was identified as 2 by using thin-layer chromatography (aluminum oxide plates, methanol as eluent). (23) Jong, L. I.; Abbott, N. L. Unpublished results. (24) Krisovitch, S. M.; Regen, S. L. J. Am. Chem. Soc. 1992, 114, 9828-9835. (25) Rosen, M. J. Surfactants and Interfacial Phenomena, 2nd ed.; Wiley: New York, 1989.

Letters

Figure 3. Surface tensions of preformed (nonreacting) mixtures of 1 and 2 of various compositions (50 mM phosphate buffer at 25 °C, pH 6.9). The solutions contained X1 mM of 1 and 2(1-X1) mM of 2, where X1 is the mole fraction of 1 in solution.

presence of intermediate chemical species such as unsymmetrical disulfides formed during the reduction of 1 to 2 using 3. The results presented in Figures 2 and 3 demonstrate that the surface tensions measured during the reduction of 1 to 2 are substantially lower than surface tensions of nonreacting mixtures corresponding to the same composition of the bulk solution. We conclude, therefore, that the low surface tensions measured during the reduction of 1 to 2 arise from nonequilibrium surface states created by the transformation of a bolaform surfactant to a monomeric structure. We hypothesize that the low, transient surface tensions measured during the reduction of 1 to 2 are caused by a transient surface excess concentration of surfactant (1 and 2) that is greater than the equilibrium value corresponding to the instantaneous bulk composition of the solution. This situation can arise if the rate of reduction of 1 to 2 within the surface monolayer is greater than the rate at which 2 desorbs from the surface of the solution. We propose that a barrier to desorption of 2 from the surface of the solution leads to the accumulation of 2 at the surface of the solution. This hypothesis is

Langmuir, Vol. 14, No. 9, 1998 2237

supported by preliminary results of a diffusion-kinetic model3,23 and by the observation (based on Figure 1) that a 1 mM solution of 1 has a surface excess of 0.0085 molecule/Å2 and a surface tension of 57 mN/m whereas a solution of 2 with a surface excess of 2 × 0.0085 molecule/ Å2 can have a surface tension that is less than 57 mN/m.26 In summary, the results reported herein demonstrate the importance of understanding the nonequilibrium properties of surfactant solutions when developing strategies for active control of these solutions:1 slow kinetics of desorption can cause the surface tensions of solutions of redox-surfactants during transformations of state to deviate substantially from the initial and final surface tensions of the solutions. Our results suggest that nonequilibrium states created during redox-induced transformations of surfactants may be exploited to obtain systems with transient (over minutes), low interfacial tensions. Finally, the results reported in this paper also demonstrate that water-soluble, disulfide-based surfactants, when used in thiol-disulfide exchange reactions, provide a simple and general experimental system with which to investigate the equilibrium and nonequilibrium properties of aqueous solutions of water-soluble surfactants. Acknowledgment. This work was supported in part by the Camille and Henry Dreyfus Foundation (New Faculty Award), the David and Lucile Packard Foundation, the donors of the Petroleum Research Fund, and the National Science Foundation (CTS-9410147 and CTS9502263). The authors thank George M. Whitesides for helpful comments. Supporting Information Available: Synthesis and characterization of compounds 1 and 2 (4 pages). Ordering information is given on any current masthead page. LA980235C (26) This observation is consistent with a past study that investigated oxidation of water-insoluble, 1-octadecanethiol to dioctadecyl disulfide at the surface of a solution. At constant surface pressure, oxidation was accompanied by a decrease in surface area: Ahmad, J. Langmuir 1996, 12, 963-965.