Langmuir 2007, 23, 12473-12477
12473
Electrolyte-Induced Phase Separation and Charge Reversal of Cationic Zwitterionic Micelles Takeshi Aoki, Makoto Harada, and Tetsuo Okada* Department of Chemistry, Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8551, Japan ReceiVed June 15, 2007 The solution properties and anion recognition selectivity of the N-dodecyl-N,N,N′,N′,-tetramethyldiethylenediammoniopropane sulfonate bromide (DEPB) and N-dodecyl-N,N,N′,N′,-tetramethyl-1,3-propylenediammonio-propane sulfonate bromide (DPPB) micelles have been studied. These micelles have similar properties except for the phase behavior induced by coexistent anions. The addition of ClO4- or I- to aqueous DEPB causes phase separation, and further addition results in the dissolution of the separated phases. In contrast, no phase separation occurs for DPPB. Charge reversal of the micelles occurs during the addition of the anions. The difference in the phase behavior between DEPB and DPPB comes from the different sizes of the micelles; the addition of ClO4- allows the formation of large aggregates of DEP (∼50 nm in diameter).
The zeta potential of a micelle is a good measure for evaluating its interaction with ionic compounds, albeit whether the shear plane corresponds to the real surface or not is an essential question.1-8 This useful parameter is influenced by various physical and chemical factors, including the surface charge density, the dissociation degree of counterions, and the size and shape of a micelle, which depend not only on the nature of amphiphilic molecules but also on that of counteranions. In particular, the hydration nature of counterions plays an essential role in the determination of the electrostatic properties of micelles. The interaction of ions with electrically neutral micelles is largely affected by the hydration nature of ions rather than their electric charges. The zeta potential measurements have, for example, revealed that a negative potential is induced on a polyoxyethylated nonionic micelle because of the imbalance between anion and cation partitions.2 Ion-transfer voltammetry has verified that perchlorate is more preferably solvated in the micelle than in water by 3.5 kJ mol-1, whereas bromide does not undergo such favorable solvation.2 As a result, perchlorates induce a larger negative potential on the micelle than do halides. A similar trend has been reported for the ionic partition into a dodecyldimethylammoniopropane surfonate (DDAPS) micelle.1 DDAPS is an electrically neutral molecule but has an inner positive and an outer negative group. Although an electrostatic interaction should play some role in the ionic partition to the DDAPS micelle, it is known that the partition selectivity can be explained by the hydration energies of anions rather than by their charges. A similar rule may be applicable even for ionic micelles. However, most of the cationic surfactants form insoluble iodide or perchlorate salts, and thus the effect of counterions cannot be studied over a wide range of hydration energies.9,10 From this * Corresponding author. E-mail:
[email protected]. Phone and Fax: 81-3-5734-2612. (1) Iso, K.; Okada, T. Langmuir 2000, 16, 9199. (2) Ohki, T.; Harada, M.; Okada, T. J. Phys. Chem. B 2006, 110, 15486. (3) Sabate´, R.; Gallardo, M.; Estelrich, J. Electrophoresis 2000, 21, 481. (4) Imae, T.; Hayashi, N. Langmuir 1993, 9, 3385. (5) Cassidy, M. A.; Warr, G. G. J. Phys. Chem. 1996, 100, 3237. (6) Healy, T. W.; Drummond, C. J.; Grieser, F.; Murray, B. S. Langmuir 1990, 6, 506. (7) Imae, T.; Kohsaka, T. J. J. Phys. Chem. 1992, 96, 10030. (8) Johnsson, M.; Engbets, J. B. F. N. J. Phys. Org. Chem. 2004, 17, 934. (9) Udupa, M. R. Thermochim. Acta 1982, 52, 363. (10) Knock, M. M.; Bain, C. D. Langmuir 2000, 16, 2857.
perspective, novel cationic zwitterionic surfactants DEP and DPP have been synthesized.11
Their charge-stacked structures are expected to enhance the water solubility of their salts with poorly hydrated anions and to allow us to study the effects of counteranions on the properties of cationic micelle in more detail. In this letter, we report the properties of the DEP and DPP micelles; some are predictable from their structures, but some are unpredictable. Experimental Section Measurements. A 50 µm i.d. fused silica capillary (GL Science Inc.) was used for electrophoretic measurements of the mobility of micelles in various electrolytes. The total length of the capillary was 60 cm, and its effective length (i.e., the length between the sample injection end and the detection window) was 50 cm. Electrical voltage (10 kV) was applied between both ends of the capillary with a high-voltage electric power unit (Matsusada Precision Inc., HCZE30P no. 25). Solute migration was monitored with a UV-vis detector (Jasco CE-1570) set to 275 nm. These instruments were installed in an incubator to keep the temperature at 25 °C. A sample solution was injected by siphoning. An appropriate injection end was selected according to the direction of electroosmotic flow, which depended on the type of electrolyte and its concentration. Acetone and pyrene were added to sample solutions as the electroosmotic marker and the migration marker for the micelles, respectively. Dynamic light scattering measurements of micellar size were carried out with a Photal model Z. The concentrations of free Brin the micellar solutions were determined at 25 °C by potentiometry with a Br--selective electrode (8005-10C, Horiba). The potential measured against a double-junction Ag/AgCl reference electrode was converted to the free Br- concentration according to the literature.12 Similarly, the free ClO4- concentration was determined (11) Aoki, T.; Harada, M.; Okada, T. Langmuir 2007, 23, 8820. (12) Zana, R. J. Colloid Interface Sci. 1980, 78, 330.
10.1021/la7028565 CCC: $37.00 © 2007 American Chemical Society Published on Web 11/08/2007
12474 Langmuir, Vol. 23, No. 25, 2007
Letters
Figure 1. Phase diagrams obtained for DEPB/I- (upper) and DEPB/ClO4- (lower) at 25 °C. L1 and L′1 are homogeneous aqueous phases, and L2 is a surfactant-rich phase. Concentrations where phase separation occurs (O) and concentrations where the separated phases are remixed to the homogeneous L′1 phase (b). with a homemade ClO4--selective electrode.1 The X-ray absorption fine structure (XAFS) spectra were measured at the BL7C of the Photon Factory, High-Energy Accelerator Research Organization in Tsukuba, Japan. The data processing details are given in our previous paper.11 Phase Diagrams. A sodium salt of Br-, I-, or ClO4- was added to a DEPB or DPPB solution. The concentrations, where phase separation or dissolution of the separated phases occurred, were visually determined. This procedure was repeated with different DEP or DPP concentrations. All of the solutions were kept at 25 °C. Chemicals. The syntheses of DEPB and DPPB are briefly outlined here.11 DEPB was synthesized by the reaction of dodecylbromide with N,N,N′,N′-tetramethylethylenediamine and the reaction of the product of the first reaction with propanesultone. Crude DEPB was purified by treatment with Br--form anion-exchange resin (Amberlist A26) and repeated recrystallization from acetone-methanol. The same procedures were applied to the synthesis of DPPB except that N,N,N′,N′-tetramethyl-1,3-propylenediamine was used as an initial reactant. DPPB was finally recrystallized from acetonitrile. Ion chromatographic measurement confirmed that the Br- content in this purified DEPB and DPPB was higher than 98%. Sodium salts of inorganic anions were used as received. Aqueous solutions were prepared with Milli-Q water.
Results and Discussion The critical micelle concentration is 15 mM for both DEPB and DPPB. We have reported that the degree of dissociation (R) of the DEPB micelles is ca. 0.21.11 The same R value was confirmed for the DPPB as well. The average hydration number of the bromides partitioned into the DEP micelles (Non-mic) was reported to be 3.6-3.8 on the basis of the XAFS measurements at the Br-K edge; the corresponding Non-mic values for the DPPB solutions (0.03-0.15 M) have been determined to be 3.7-3.9. Thus, these physicochemical parameters strongly suggest that the DEPB and DPPB micelles have almost the same properties. In contrast, the aqueous solutions of DEPB and DPPB behaved in entirely different fashions when an electrolyte was added. Figure 1 shows the phase diagrams of aqueous DEPB obtained with the addition of I- or ClO4- at 25 °C. For DEPB, the addition of ClO4- or I- causes phase separation, which occurs at the lower boundaries in the diagram. Although the L2 phase is optically transparent, polarizing microscope observations have implied that it has a lamellar structure. The separated phases are dissolved into a single phase by the further addition of ClO4-
Letters
Langmuir, Vol. 23, No. 25, 2007 12475
Figure 2. Changes in the concentrations of free Br- (circles) and ClO4- (triangles) with the concentration of added ClO4-. Open symbols, DEP; solid symbols, DPP. The solid curve represents the change in the total concentration of free anions. The broken curve shows the change in the concentration of ClO4- bound to the micelles. The concentrations of DEPB and DPPB were 0.05 M.
Figure 3. Changes in micellar diameters with the concentration of added ClO4-. Error bars represent standard deviations. The concentrations of DEPB and DPPB were 0.3 M.
or I-. Interestingly, phase separation of the DPPB solutions was not caused by the addition of any electrolytes tested. The ion recognition selectivity of the DEPB and DPPB micelles was studied to obtain further insight into their different phase behavior. Figure 2 shows changes in the free ClO4- and Brconcentrations ([ClO4-]free and [Br-]free) with the concentration of ClO4- added ([ClO4-]add) to 0.05 M surfactant solutions. [Br-]free almost linearly increases with increasing ([ClO4-]add over the range of [ClO4-]add ) 0-0.03 M, whereas an increase in ([ClO4-]free is very small, implying that almost all of the added perchlorates are bound to the micelle. When the separated phases are mixed into a homogeneous solution, all of the bromides are excluded from the micelles. These changes in the free anion concentrations are entirely identical for both the DEP and DPP micellar systems. The ion-exchange selectivity coefficients of ClO4- over Br- in these micelles were 40-50, which depended on the concentration of the anions. Thus, ClO4-, which is partitioned into the DEPB and DPPB micelles much better than
Br- as predicted by their hydration energies, causes phase separation in DEPB solutions. It is important that the ion recognition selectivity is the same for the DEP and DPP micelles and cannot explain their different phase behaviors. Figure 3 shows changes in the diameters of the DEP and DPP micelles (0.3 M) with [ClO4-]add. The sizes of the micelles first increase with increasing [ClO4-]add and then decrease with further increases in [ClO4-]add; thus, the size changes are closely related to the phase separation for DEPB and the dissolution of separated phases. Because the micellar shapes are not necessarily spherical, the sizes determined by dynamic light scattering may not be very reliable. However, it should be noted that a more marked increase in a size is found for the DEP micelle; the diameter reaches ca. 50 nm just before phase separation occurs. This size increase should be caused by the binding of ClO4- and should be accompanied by the micellar shape transition from a sphere to a rod or disk; the coagulation of large aggregates causes phase separation. Because surfactants with smaller head groups are
12476 Langmuir, Vol. 23, No. 25, 2007
Letters
Figure 4. Changes in micellar mobility with added electrolyte. Micellar concentration, 50 mM for both DEPB and DPPB. Temperature, 25 °C. ClO4- (b), Br- (O), I- (2), and SO42- (0).
likely to form a nonspherical micelle, the shape transition may occur more easily for DEP, which has a smaller head group than DPP.13 The electrophoretic mobility of the DEP and DPP micelles in various electrolytes was measured to gain further insight into the ionic binding of these micelles as summarized in Figure 4. No clear differences in the electrophoretic mobility are found between the DEP and DPP micelles. We will discuss the mobility and charge of the DPP micelle because phase separation does not interfere with electrophoretic measurements. An increase in electrolyte concentration causes the compression of an electrical double layer and reduces the electrophoretic mobility even without specific interactions. Although the dicationic structures of the DPP molecule may lead to the expectation of a strong electrostatic interaction with multivalent anions, the effect of SO42- on the electrophoretic mobility of the micelle is smaller than that of Br-. The addition of ClO4- or I- reduces the micellar mobility to a greater extent. The electrophoretic mobility of the micelle continuously becomes less positive with increasing concentration of ClO4- or I-; zero mobility occurs at [ClO4-]add ) 0.045 M (13) Israelachvili, J. N. Intermolecular and Surface Forces; Academic Press: London, 1992.
or [I-]add ) 0.06 M. Further increases in the anion concentration result in the reversal of the electrophoretic direction because of the partitioning of excess anions into the DPP micelle. Thus, the partitioning of an anion into a micelle is principally governed by its hydration rather than its charge. The charge reversal of the micelles obviously comes from their dicationic structures, which allow the binding of excess anions to the micelle. The electrophoretic mobility (µ) was converted to the zeta potential according to Henry’s equation14
µ)
0ζ f(κa) η
where η is the viscosity of a medium, f(κa) is the Henry coefficient, κ is the Debye shielding parameter, and a is the radius of the micelle. Although the size of the micelle is varied with the concentration of ClO4- as illustrated in Figure 2, this effect on the calculated zeta potential is so small that a constant value (a ) 20 nm) was assumed in the above equation. The zeta potential of the DPPB micelles is 33 mV without added electrolytes, and decreases to -20 mV when ClO4- is excessively partitioned. (14) Ohshima, H. J. Colloid Interface Sci. 1996, 180, 299.
Letters
The former value corresponds to an excess of 10 positive charges per micelle, and the latter corresponds to 6 to 7 negative charges. If the aggregation number of the micelle is 50, then the degree of dissociation (R ) 0.21) agrees with its charge excess. Assuming the same aggregation number for the negatively charged DPP (or DEP) micelle, 10-15% of the surfactant molecules in the micelle bind two anions in 100 mM ClO4- solution. A number of examples of charge reversal are well known (e.g., for mineral particles interacting with polymers and layerby-layer materials).15-19 The charge reversal of a micelle by excessive partitioning of ions has been reported for the combination of salicylate with a cationic surfactant.7 Also, similar phenomena have been found for polyelectrolytes and multivalent ions.19 Thus, the charge reversal that we have confirmed for the (15) Becraft, K. A.; Moore, F. G.; Richmond, G. L. J. Phys. Chem. B 2003, 107, 3675. (16) Halaoui, L. I. Langmuir 2001, 17, 7130. (17) McNamee, C. E.; Matsumoto, M.; Hartley, P. G.; Mulvaney, P.; Tsujii, Yo.; Nakahara, M. Langmuir 2001, 17, 6220. (18) Wang, W.; Gu, B.; Liang, L.; Hamilton, W. A. J. Phys. Chem. B 2004, 108, 17477. (19) Lyklema, J. Colloids Surf., A 2006, 291, 3.
Langmuir, Vol. 23, No. 25, 2007 12477
DEP and DPP micelles is unique because it is caused by the partitioning of simple monovalent inorganic anions, such as Iand ClO4-. In conclusion, although the DEP and DPP molecules have similar micellar natures and ionic partitioning properties, the addition of poorly hydrated anions results in entirely different phase behavior, possibly because of different polar head sizes. In addition, the reversal of micellar charges is easily induced by adding I- and ClO4-. Molecular aggregates have been used as templates for the syntheses of nanomaterials, in which the sizes of the aggregates are an important factor. Surfactant adsorption on nanoparticles is also utilized to control their surface charges. From these perspectives, the present surfactants and their micelles are expected to be particularly useful in the development of novel nanomaterials. Acknowledgment. This work was supported in part by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science and by the Salt Science Research Foundation. LA7028565
