30
Langmuir 2004, 20, 30-32
Ion Exchange at a Surface Monolayer Makoto Harada and Tetsuo Okada* Department of Chemistry, Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8551, Japan Received August 4, 2003. In Final Form: November 11, 2003 Total reflection total conversion electron yield X-ray absorption fine structure is applied to the evaluation of ion exchange occurring at the surface monolayer of two-tailed ammonium ions. X-ray absorption measurements at a Br K-edge allow us to detect ion-exchange equilibria between Br- and Cl- added in a subphase. The ion-exchange selectivity of Br- over Cl- basically increases as the monolayer is compressed, indicating that Cl- is selectively squeezed out by compression because of its larger hydrated ionic radius.
Introduction Ion-exchange experiments are usually carried out with bulk ion exchangers, such as resins and membranes. However, since ion exchange occurs at the interface between an ion exchanger and electrolyte solution, use of a bulk ion exchanger is not essential. Bulk ion exchangers are required for an experimental reason, that it is extremely difficult to evaluate ionic distribution when the interface having ion-exchange functionality comprises a marginal sector of the entire working volume; various approaches have been attempted to selectively probe such interfaces.1 In bulk ion exchangers, polymer matrixes often influence ion-exchange selectivity and make the interpretation of experimental results complicated; such effects as “polymer effect” and “matrix effect” have been wellknown.2 Monolayer ion exchangers without matrixes should be ideal materials for the investigation of the interaction between charged groups. Although there are various ways to prepare monolayer ion exchangers, a direct and selective access to them requires special techniques. Langmuir-Blodgett membranes have been used for monolayer ion-exchange experiments, where the transference of the monolayer onto a substrate obviously makes the application of spectroscopic and electrochemical approaches easier.3 However, no one doubts that in situ experiments are essential to understand phenomena occurring at a monolayer/solution interface. Although, to evaluate monolayer properties, several methods have been developed such as interfacial tension measurements,1a flotation,4 Brewster angle microscopy,5 X-ray reflectivity,6 vibration spectroscopy,7 etc., most of these methods focus on a surface monolayer (or a surfactant molecule) itself and not on counterions. We have recently developed an efficient approach to monolayers at the air/solution interface on the basis of * To whom correspondence may be addressed. Phone and Fax: +81-3-5734-2612. E-mail:
[email protected]. (1) (a) Bunton, C. A.; Nome, F.; Quina, F. H.; Romsted, L. S. Acc. Chem. Res. 1991, 24, 357. (b) Cuccovia, I. M.; da Silva, I. N.; Chaimovich, H.; Romsted, L. S. Langmuir 1997, 13, 647. (c) Souza, S. M. B.; Chaimovich, H.; Politi, M. J. Langmuir 1995, 11, 1715. (2) Ciardelli, F.; Carlini. C.; Pertici, P.; Valentini, G. J. Macromol. Sci., Chem. 1989, A26, 327. Green, B. R.; Hancock, R. D. Hydrometallurgy 1981, 6, 353. (3) Fujii, M.; Li, B.; Fukada, K.; Kato, T.; Seimiya, T. Langmuir 2001, 17, 1138. Du, X.; Liang, Y. J. Phys. Chem. B 2001, 105, 6092. Peng, J. B.; Barnes, G. T.; Gentle, I. R. Adv. Colloid Interface Sci. 2001, 91, 163. (4) Morgan, J. D.; Napper, D. H.; Warr, G. G.; Nicol, S. K. Langmuir 1994, 10, 797. Morgan, J. D.; Napper, D. H.; Warr, G. G. J. Phys. Chem. 1995, 99, 9458. (5) Marshall, G.; Dennin, M.; Knobler, C. M. Rev. Sci. Instrum. 1998, 69, 3699. (6) Cuvillier, N.; Rondelez. F. Langmuir 1999, 15, 5547. (7) Li, C.; Zhao, B, Lu. Y.; Liang, Y. J. Colloid Interface Sci. 2001, 235, 59. Calvez, E.Le; Blaudez, D.; Bauffeteau, T.; Desbat, B. Langmuir 2001, 17, 670.
X-ray absorption fine structure (XAFS).8 This method allows us to probe both the amount and local structure of ions attracted by the surface monolayer. The incidence X-ray is introduced onto the solution surface at a grazing angle, which satisfies the total reflection condition. X-ray absorption occurs within ca. 10 nm from the surface (the detection depth is determined by the incidence angle and energy of X-ray) and thus involves surface-selective information. Total conversion electron yield detection allows us to detect very weak X-ray absorption signals with a satisfactory sensitivity. This method is, hereinafter, named total reflection total conversion electron yield XAFS (TRTCY-XAFS). The surface charge density is an important factor governing an electrostatic interaction and in turn ion-exchange equilibrium. As already reported, the surface molecular density can be successively varied by moving two PTFE barriers in the present TRTCY-XAFS cell.8b Thus, the present approach is suitable for studying ion exchange taking place just below surface monolayers. Experimental Section The TRTCY-XAFS instrument is basically the same as that previously reported except for minor modifications.8b The solution temperature was maintained at 25 °C by circulating thermostated water. The size of the trough was 250 mm × 210 mm × 2 mm. The effective surface area, which was varied by moving PTFE barriers, ranged from 63 to 399 cm2. The incidence X-ray was introduced onto the solution surface between the barriers at a grazing angle, which was set to 0.8 mrad smaller than the critical angle at the Br K-edge energy (2.1 mrad). The beam size of the incidence X-ray was typically 50 µm (height) × 5 mm (width), and thus the observation area on the solution surface was calculated as 4.2 cm × 5 mm. The space above the solution was filled with He, and ca. 150 V bias voltage was applied between the bottom of the trough and a carbon electrode (5 × 10 cm) placed ca. 1 cm above the solution. He atoms were successively ionized by the photoelectrons and Auger electrons generated by the absorption of evanescent wave just below the solution surface, and the resulting He+ was collected by the carbon cathode. The incidence X-ray intensity was monitored by a 4 cm long ion chamber filled with N2. All XAFS measurements were performed at BL-7C of the Photon Factory, High Energy Accelerator Research Organization in Tsukuba, Japan. Didodecyldimethylammonium bromide (DMAB) and dihexadecyldimethylammonium bromide (HMAB) were purchased from Tokyo Kasei and used as received. An appropriate aliquot (3040 µL) of 1 mM DMAB or HMAB benzene solution was dropped on the surface of an aqueous solution to allow the surface monolayer formation. Subphases were prepared with MilliQ water. Other reagents were of the highest grade available. (8) (a) Harada, M.; Okada, T.; Watanabe, I. J. Phys. Chem. B 2003, 107, 2275. (b) Harada, M.; Okada, T.; Watanabe, I. Anal. Sci. 2002, 18, 1167.
10.1021/la0354180 CCC: $27.50 © 2004 American Chemical Society Published on Web 11/27/2003
Letters
Langmuir, Vol. 20, No. 1, 2004 31
Figure 1. π-A curves for DTAB and HTAB surface monolayers: (a) subphase 5 mM KBr; (b) subphase (1) 5 mM KBr, (2) 3 mM KBr + 2 mM KCl, and (3) 1 mM KBr and 4 mM KCl; monolayer HTA.
Results and Discussion Figure 1 shows surface pressure (π)-surface molecular area (A) curves obtained with DMAB and HMAB surface monolayers on (a) NaBr and (b) NaBr-NaCl mixtures as subphases. The surface monolayers were compressed at 0.175 cm2/s. It is known that insoluble monolayers can be formed on a solution surface though the solubility of DMAB is rather high (ca 10-5 M).9,10 No anomalies originating from the dissolution of the surface monolayers are found for the π-A curves obtained with a NaBr subphase even for DMAB. These simple curves indicate the gradual transition from an expanded to condensed liquid phase upon compression. The collapse of the monolayer occurs at A ) ca. 60 Å2. Molecular modeling indicates that the cross-sectional areas parallel to the surface (Acs) of DMAB and HMAB are 73 and 95 Å2, respectively, assuming that the molecules are situated on the surface with two hydrophobic chains similarly extended into the air. When A becomes smaller than Acs by compression, the molecules are aligned on the surface with the two-dimensionally closest-packed structures. When Cl- replaces Br- in the subphase, the surface pressure tends to decrease as shown in Figure 1b. This result disagrees with that reported by Souza et al.,1c who showed that replacing Br- by Clresulted in the higher surface tension due to the larger hydrated ionic radius of the latter. When a glass plate, which was used for surface tension measurements in the present study, was immersed in the surface monolayer for a long time period, surfactant molecules were strongly adsorbed on its surface and thereby reproducibility became worse. For this reason, we confine our discussions based (9) Dynarowicz, P.; Romeu, N. V.; Trillo, J. M. Colloids Surf., A 1998, 131, 249. (10) Haas, S.; Hoffmann, H. Prog. Colloid Polym. Sci. 1996, 101, 131.
Figure 2. (a) Changes in signal intensities obtained at the Br K edge with the surface molecular area (A) of DTA and HTA: subphase 5 mM KBr. (b) Dependence of signal intensity-A curves on subphase compositions: (1) 5 mM KBr, (2) 3 mM KBr + 2 mM KCl, and (3) 1 mM KBr and 4 mM KCl with the HTA monolayer.
on the surface tensions to verification of the homogeneous monolayer formation and its smooth transition upon compression. The charge density is an important factor governing electrostatic interactions and in turn affecting ionexchange separation. It is one of the advantages of the present setup that TRTCY-XAFS measurements are feasible with the surface molecular density successively changed by moving PTFE barriers. The Debye length in 5 mM aqueous solution of a 1:1 electrolyte is ca. 4.3 nm, which almost corresponds to the detection depth of the present method, indicating that we can selectively probe the interior of the electrical double layer and all of counterions are accumulated in the detection volume. In this solution, the total mole of the electrolyte in the detection volume (2.1 nL) is ca. 10 pmol, which is negligible in comparison with the amount of counterions of the surface monolayer (0.35 nmol at A ) 100 Å2). Thus, TRTCY-XAFS signals totally come from the ions influenced by the surface monolayer. Our previous work indicated that ca. 30% of the bromide ions attracted by the zwitterionic surface monolayer are dissociated from the ammonium groups of surfactant molecules, indicating that counterions accumulated in the electrical double layer significantly contribute to the XAFS signals.8a It should also be noted that the usual transmission XAFS method cannot selectively access the interface nor detect ions at nanomole to sub-nanomole levels. Figure 2a shows changes in the signal intensities measured at Br K-edge with A on a 5 mM NaBr subphase. The surface monolayers were compressed at the same rate as used for π-A curve measurements. When the monolayer is compressed, the signal intensities should increase in a
32
Langmuir, Vol. 20, No. 1, 2004
hyperbolic fashion as drawn by broken curves. The TRTCY-XAFS signals follow the broken curves when A is relatively large. However, the departures of experimental values from the curves become obvious when the surface monolayers are further compressed, and finally the signal intensities are leveled off. Although results are not shown, lower compression rates resulted in more marked departure from the hyperbolic curve; constant signals were observed despite substantial compression in some cases. One possible origin of this observation is the dissolution of the surface monolayers as pointed out for similar ammonium ions.9 However, the surface pressure increases with increasing A, suggesting that other effects are possibly responsible for this effect. We would like to point out another possibility that the compression of the monolayer allows the formation of a dense layer of surfactant hydrophobic chains just above the solution and increases the refractive index, which causes a change in the total reflection plane. We are studying this aspect in detail. Figure 2b shows similar results obtained with the subphases containing both Br- and Cl-. The edge jump at the Br K edge decreases with increasing concentration of Cl- in the subphase principally due to the ion-exchange between Cl- and Br- at the surface monolayer. The ion-exchange selectivity coefficient of Br- over Cl(KIE) at a given A was calculated from XAFS signal intensities based on the following relation
KIE )
EJBr[Cl-] (EJBr - EJBr-Cl)[Br-]
where EJBr and EJBr-Cl are the edge-jump magnitudes measured for the subphase containing Br- as a counteranion and that containing both Cl- and Br-, respectively. KIE is varied with A as depicted in Figure 3. KIE increases with decreasing A for both DMAB and HMAB surface monolayers, indicating that the compression of the monolayer enhances the selectivity toward Br-. Abrupt changes in KIE occur when A is decreased up to ca. 100 Å2, which is almost equal to Acs of DMAB and HMAB. This must be due to the phase transition or changes in the configurations of the surface monolayers. Although the present experiments are so delicate that the reproducibility of the values is not very high, repeated experiments have confirmed the trends shown in Figure 3. The edge jumps may involve 5% errors, and thus resulting KIE values contain >10% errors. Changes in KIE are significant if this error is taken into account. There are two possibilities to account for the Brselectivity enhanced by compressing the monolayer; these
Letters
Figure 3. Changes in ion-exchange selectivity of Br- against Cl- with A.
should be related to the leveling-off of the XAFS signals at the Br K-edge illustrated in Figure 2. One is the dissolution of the monolayer, which is likely to be accompanied by Cl- because of high water solubility of chloride salts. The other is a change in the total reflection plane due to a change in the refractive index just above the solution surface. The total reflection plane should shift up by the compression of the monolayer. If the shift is responsible for the enhanced Br- selectivity, it is suggested that Br- be accumulated nearer to the surface than Cl-. In either case, Cl- is selectively squeezed out by compressing the monolyaer because it has a larger hydrated radius than Br-. Hence, the present approach has proved efficient for studying ion exchange occurring at surface monolayers. Although light elements are not studied directly by this method (Cl- is the case), ion exchange involving such elements can be probed by selecting an appropriate ionexchange pair (Br- in the present case). An important feature of this method is that ions attracted from the subphase are selectively detected. Its extensive application to various systems and comparison of the data with existing ones obtained with bulk ion exchangers facilitate the further understanding of ion-exchange phenomena and interpretation of ion-exchange data. Acknowledgment. The authors are grateful for financial support from the Salt Science Research Foundation and the Mukai Science and Technology Foundation. This work was performed under the approval of Photon Factory Advisory Committee (Proposal Nos. 2001G113 and 2003G082). LA0354180
