Langmuir 2004, 20, 33-36
33
Surfactant Aggregation within Room-Temperature Ionic Liquid 1-Ethyl-3-methylimidazolium Bis(trifluoromethylsulfonyl)imide Kristin A. Fletcher and Siddharth Pandey* Department of Chemistry, New Mexico Institute of Mining and Technology, Socorro, New Mexico 87801 Received August 27, 2003. In Final Form: November 6, 2003 On the basis of the response of solvatochromic probes [Reichardt’s betaine dye, pyrene, and 1,3-bis(1-pyrenyl)propane], we have investigated the aggregation behavior of common anionic, cationic, and nonionic surfactants when solubilized within a low-viscosity room-temperature ionic liquid 1-ethyl-3methylimidazolium bis(trifluoromethylsulfonyl)imide (emimTf2N). We observed possible aggregate formation by all nonionic surfactants included in the study (Brij-35, Brij-700, Tween-20, and Triton X-100), while no aggregation was observed for the cationic surfactant cetyltrimethylammonium bromide. The anionic surfactant sodium dodecyl sulfate does not appear to solubilize within emimTf2N at ambient conditions.
Introduction With growing concern over the environmental fate, impact, and health hazards of traditional organic solvents, researchers continue to search for greener alternatives. Room-temperature ionic liquids (RTILs), a relatively new class of solvents, have been shown to be viable substitutes for organic solvents in many reactions of industrial importance. Several properties make RTILs highly desirable. For instance, they have a negligible vapor pressure, are stable up to very high temperatures (300 °C or more), are not flammable, and can easily be recycled for continued use. A number of solutes are not soluble in RTILs; however, organized media within RTILs could significantly increase the solubility of sparingly soluble solutes and, thus, enhance RTIL applications. Surfactants are amphiphilic molecules with a hydrophilic polar headgroup (nonionic, anionic, cationic, or zwitterionic in nature) and a hydrophobic hydrocarbon chain (the tail). In this letter, we report the behavior of several surfactants (anionic, cationic, and nonionic) when dissolved in 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide (emimTf2N, viscosity ∼28 cP,1 Figure 1). Merrigan et al. have demonstrated that imidazolium cations with attached long fluorous tails act as surfactants and appear to self-aggregate within imidazolium-based RTILs;2 however, there has been no systematic study of the behavior of traditional surfactants within RTILs so far. It is important to mention here that, during the submission of this letter, a communication presenting the evidence of micelle formation in the RTILs 1-butyl-3-methylimidazolium hexafluorophosphate (bmimPF6) and -chloride (bmimCl) upon addition of Brij700, Brij-35, dioctyl sulfosuccinate, sodium dodecyl sulfate (SDS), and caprylyl sulfobetaine by the Armstrong group appeared in the literature.3 This group reports the decrease in surface tension of the RTIL solutions upon addition of surfactants as the first possible indication of the presence * Corresponding author. Phone: (505) 835-6032. Fax: (505) 8355364. E-mail:
[email protected]. (1) McEwen, A. B.; Ngo, H. L.; LeCompte, K.; Goldman, J. L. J. Electrochem. Soc. 1999, 146, 1687. (2) Merrigan, T. L.; Bates, E. D.; Dorman, S. C.; Davis, J. H., Jr. Chem. Commun. 2000, 2051. (3) Anderson, J. L.; Pino, V.; Hagberg, E. C.; Sheares, V. V.; Armstrong, D. W. Chem. Commun. 2003, 2444.
Figure 1. Molecular structure of RTIL emimTf2N.
of micellar aggregates. They further investigate the solvation behavior of these micellar aggregates in RTILs using a linear free energy relationship in concert with inverse gas chromatography. The ability of surfactants to self-aggregate depends on one or more of the following: the structure of the surfactant, its concentration, the solubilizing media, and the method with which the self-assemblies are prepared. The critical aggregate concentration (CAC) represents the minimum surfactant concentration required for aggregation to occur. Traditionally, the CAC can be determined by observing sharp changes in a number of physical properties such as surface tension, turbidity, electrical conductivity, and solute solubility, among others.4 In addition, the response of solvatochromic probes may be affected by the formation of organized assemblies.4 This observation provides a simplistic and convenient method to establish early evidence of aggregate formation. It is the latter property that is reported here. Solvatochromic probes respond to one or more solventsolute interaction(s) in solution.5 Therefore, they provide a convenient method to investigate solute-solvent or solvent-solvent interactions. As the process of surfactant aggregation becomes favorable, solution properties such as viscosity, hydrogen-bond donating ability, static dielectric constant, polarizability, and so forth could change. Variations in these properties can be successfully reported by the judicious selection of solvatochromic probes. We report the behavior of three such probes [absorbance probe (4) (a) Evans, D. F.; Wennerstrom, H. The Colloidal Domain; WileyVCH: New York, 1999; Chapter 4. (b) Pramauro, E.; Pelezetti, E. Surfactants in Analytical Chemistry; Elsevier: New York, 1996; Vol. XXXI. (5) (a) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Kluwer Academic/Plenum Press: New York, 1999. (b) Marcus, Y. Chem. Soc. Rev. 1993, 409. (c) Acree, W. E., Jr. In Encyclopedia of Analytical Chemistry; Meyers, R. A., Ed.; John Wiley & Sons, Ltd.: Chichester, 2000; pp 10280-10305.
10.1021/la035596t CCC: $27.50 © 2004 American Chemical Society Published on Web 11/19/2003
34
Langmuir, Vol. 20, No. 1, 2004
Letters
I corresponds to a S1(v ) 0) f S0(v ) 0) transition and band III is a S1(v ) 0) f S0(v ) 1) transition. As the solvent dipolarity increases, the II/IIII ratio increases.7 A change in the microviscosity of the cybotactic region is effectively manifested through the steady-state fluorescence spectrum of BPP. As the microviscosity of the cybotactic region decreases, the two pyrene moieties more easily fold together to form an intramolecular excimer, and the ratio of monomer intensity to excimer intensity, IM/IE, decreases.8 Experimental Section
Figure 2. Molecular structures of the probes used in this study.
Figure 3. Molecular structures of the surfactants used in this study.
Reichardt’s betaine dye, fluorescence probes pyrene, and 1,3-bis(1-pyrenyl)propane (BPP), see Figure 2] as incremental amounts of surfactant are added to emimTf2N. The concentration-dependent behavior of several different surfactants such as Brij-35, Brij-700, Tween-20, and Triton X-100 (nonionic) as well as cetyltrimethylammonium bromide (CTAB, cationic) and SDS (anionic) are investigated within neat emimTf2N (molecular structures of these surfactants are presented in Figure 3). The ET(30), one of the most widely used empirical scales of solvent dipolarity, is calculated from the wavelength maximum of the lowest-energy, intramolecular chargetransfer π-π* absorption band of the zwitterionic Reichardt’s dye, λmax, using ET(30) (kcal mol-1) ) 28 951/ λmax (nm) or ET(30) (kJ mol-1) ) 1.2 × 105/λmax (nm).6 Pyrene is a small neutral fluorescence probe whose polarity scale is defined as the II/IIII emission intensity ratio where band (6) Reichardt, C. Chem. Rev. 1994, 94, 2319 and references therein.
Highest purity Reichardt’s betaine dye, pyrene, and BPP were purchased from Aldrich, AccuStandard, and Molecular Probes, respectively. All the probes were used as received. Dilute stock solutions were prepared by dissolving the probe in ethanol or methylene chloride in precleaned amber-glass vials and stored in a refrigerator at 4 °C. Very high purity electrochemical-grade emimTf2N (g99.9+%) was obtained from Covalent Associates, Inc. Brij-35, SDS, and CTAB (99+%) were purchased from Acros. Tween-20 (enzyme grade) and Triton X-100 (electrophoresis grade) were obtained from Fisher. Brij-700 was purchased from Aldrich. Samples for spectroscopic studies were prepared as described elsewhere.9 More specifically, samples for spectroscopic studies were prepared in the following manner: appropriate aliquots of probe stock solutions were transferred into 1-cm2 quartz cuvettes and evaporated under ultrahigh-purity nitrogen. Sufficient emimTf2N was added so the resulting solution was 10 µM in probe, mixed thoroughly, and allowed to equilibrate (typically 1-2 h). Liquid surfactant (Tween-20 and Triton X-100) was added directly to the sample solution in increments using micropipets. Incremental amounts of solid surfactant (Brij-35, Brij-700, CTAB, SDS) were weighed using a Mettler-Toledo AB54 analytical balance and added to the RTIL solution. After each addition of surfactant, the sample was mixed thoroughly and allowed to equilibrate for 40-60 min, heating to ∼60 °C to aid the dissolution of the solid surfactants. Steady-state fluorescence experiments were performed with a PTI QuantaMaster model C-60/2000 L-format scanning spectrofluorimeter with a 75-W xenon arc lamp as the excitation source and single-grating monochromators as wavelength selection devices. All fluorescence spectra were corrected for emission monochromator response and were background subtracted using appropriate blanks. Absorption spectra were recorded on an Agilent Hewlett-Packard 8453 photodiode array spectrophotometer in the usual manner. All the fluorescence and absorbance data were collected in a 1-cm2 quartz cuvette at 25 °C.
Results and Discussion Initial attempts to dissolve solid SDS within emimTf2N at a concentration of ∼8 × 10-3 M proved to be difficult. We speculate that, in the absence of water, the lack of hydration surrounding the SDS headgroup prevented the solid surfactant from dissolving in emimTf2N. Additional studies are currently underway to better understand the relationship between the SDS concentration, water concentration, and temperature. Interestingly, Armstrong et. al. also do not report on the behavior of SDS in bmimPF6; however, they do present the results of SDS dissolved in bmimCl.3 At this point, we cannot rule out the possibility that the insolubility of SDS in emimTf2N may arise because of the nature of the salts formed (7) (a) Dong, D. C.; Winnik, M. A. Can. J. Chem. 1984, 62, 2560. (b) Karpovich, D. S.; Blanchard, G. J. J. Phys. Chem. 1995, 99, 3951. (8) Zachariasse, K. A. Chem. Phys. Lett. 1978, 57, 429. (9) (a) Fletcher, K. A.; Storey, I. A.; Hendricks, A. E.; Pandey, S.; Pandey, S. Green Chem. 2001, 3, 210. (b) Fletcher, K. A.; Pandey, S. Appl. Spectrosc. 2002, 56, 266. (c) Fletcher, K. A.; Pandey, S. Appl. Spectrosc. 2002, 56, 1498. (d) Fletcher, K. A.; Pandey, S. J. Phys. Chem. B. 2003, in press. (e) Fletcher, K. A.; Baker, S. N.; Baker, G. A.; Pandey, S. New J. Chem. 2003, in press.
Letters
(NaTf2N and emim dodecyl sulfate) by the statistical mixing rather than because of the absence of hydration water. Our solvatochromic probe studies provided some interesting results. For all emimTf2N-surfactant systems, the response of Reichardt’s dye [i.e., the λmax or ET(30) value] remained constant (within our experimental error), indicating that the probe may reside primarily in the emimTf2N-rich region and is not affected to a measurable extent by any surfactant aggregation regardless of the surfactant structure. We observed no change in the solvatochromic probe behavior of pyrene when the cationic surfactant CTAB was added to emimTf2N for CTAB concentrations ranging from 5 × 10-3 to 0.10 M; however, according to the response of BPP we did see a gradual increase in the microviscosity of the cybotactic region over this concentration range, which may be attributed to a slight increase in the bulk viscosity of the solution upon surfactant addition (see Supporting Information). The limit of solubility of CTAB in emimTf2N was surpassed at 0.25 M. We suspect structural similarities between the surfactant quaternary ammonium headgroup, and the RTIL imidazolium cation may, among others, prevent CTAB from aggregating to any significant extent. However, we cannot rule out the possibility that we may have reached the solubility limit of CTAB in emimTf2N before the critical aggregation concentration was reached. The Armstrong group has reported critical micelle concentrations (cmcs) of several surfactants within bmimPF6 and bmimCl that are above this concentration.3 Interestingly, all the nonionic surfactants investigated appear to self-aggregate when solubilized within emimTf2N; the aggregation behavior seems to be surfactant-concentration dependent. A significant decrease in pyrene II/IIII was observed within a very small surfactant concentration range, indicating a more nonpolar microenvironment encountered by excited-state pyrene as the surfactant concentration is increased. This may be the result of aggregation by surfactant molecules giving rise to the ionic liquid solvatophobic cybotactic region for this probe. In addition, we observed a significant and rather drastic increase in microviscosity reported by the BPP probe as the concentration of surfactant is increased. This further implies the possibility of the existence of aggregates of these nonionic surfactants within emimTf2N solutions. Reichardt’s dye probe behavior seems insensitive to these aggregates, or perhaps it is not solubilized by or partitioned into these aggregates to a significant extent. Being zwitterionic in nature, it is not inconceivable that this probe may prefer to reside in the emimTf2N-rich region. The concentration of Brij-35 ranged from 2 × 10-3 to ∼0.50 M. On the basis of the response of pyrene and BPP, we observed aggregation behavior for Brij-35 at a concentration of ∼5 × 10-2 M (Figure 4). Within experimental error, the response of Reichardt’s dye was unaffected as the surfactant concentration increased (vide supra; see Supporting Information). As expected, Brij-700, a nonionic surfactant containing an ethoxy chain considerably longer than that in Brij-35, also displayed self-aggregation behavior over a range of surfactant concentrations. The response of pyrene and BPP indicate that aggregation began at ∼1 × 10-2 M (lower than that in Brij-35, as expected; see Supporting Information). It is important to mention that the Armstrong group reported cmc values of 115 and 20 mM for Brij-35 and Brij-700 in bmimPF6, respectively.3 Our approximate cmc values of 50 and 10 mM for the two nonionic
Langmuir, Vol. 20, No. 1, 2004 35
Figure 4. Experimental response of pyrene (II/IIII) and BPP (IM/IE) when solubilized within emimTf2N as the concentration of Brij-35 is increased up to 0.50 M. The concentration of each probe is 10 µM. The excitation wavelength is 337 nm, and the excitation and emission slit widths are 2 and 1 nm, respectively.
Figure 5. Experimental response of pyrene (II/IIII) and BPP (IM/IE) when solubilized within emimTf2N as the concentration of Triton X-100 is increased up to 0.50 M. The concentration of each probe is 10 µM. The excitation wavelength is 337 nm, and the excitation and emission slit widths are 2 and 1 nm, respectively.
surfactants, respectively, are in good agreement considering emimTf2N is ∼10 times less viscous than bmimPF6.1 The concentration of Tween-20 was varied from 1 × 10-3 to 0.25 M. The behavior of pyrene and BPP indicate that aggregation occurs for surfactant concentrations greater than ∼5 × 10-2 M (see Supporting Information). The aggregation of Triton X-100 is known to depend on the structure of the solvent, where self-aggregation is promoted in solvents with two or more potential hydrogenbonding sites and not supported in polar solvents with the capability to form one hydrogen bond.10 It has been reported that, in similar RTILs, hydrogen bonding may exist between the cation and the anion.11 Furthermore, several research groups, including our own, have determined that the hydrogen-bond donating ability of several RTILs is very close to that of short-chain alcohols.9,12 The response of pyrene and BPP indicate that self-aggregation of Triton X-100 occurs within emimTf2N for surfactant concentrations greater than 0.10 M (Figure 5). It is important to mention here that, for each of the four nonionic surfactants investigated, a decrease in the optical density of Reichardt’s dye is observed as the surfactant concentration is increased (data not shown). We tentatively attribute this observation to the interaction between the cation moiety (-N+) on Reichardt’s dye and the lone pairs on many oxygen atoms present on these surfactants or (10) Reverse Micelles; Luisi, P. L., Straub, B. E., Eds.; Plenum Press: New York, 1984. (11) (a) Huang, J.-F.; Chen, P.-Y.; Sun, I.-W.; Wang, S. P. Inorg. Chim. Acta 2001, 320, 7. (b) Headley, A. D.; Jackson, N. M. J. Phys. Org. Chem. 2002, 15, 52. (12) (a) Aki, S. N. V. K.; Brennecke, J. F.; Samanta, A. Chem. Commun. 2001, 413. (b) Muldoon, M. J.; Gordon, C. M.; Dunkin, I. R. J. Chem. Soc., Perkin Trans. 2 2001, 433. (c) Baker, S. N.; Baker, G. A.; Bright, F. V. Green Chem. 2002, 4, 165. (d) Anderson, J. L.; Ding, J.; Welton, T.; Armstrong, D. W. J. Am. Chem. Soc. 2002, 124, 14247.
36
Langmuir, Vol. 20, No. 1, 2004
the hydrogen bonding between the anionic site (-O-) of the dye and the surfactant headgroup. The lack of the presence of any such functionalities on the other two probes (i.e., pyrene and BPP) suggest the extent of any interaction between these probes and the nonionic surfactant monomers to be minimum or completely absent. Therefore, at this point, we believe these two probes to be solubilized by the surfactant aggregates within emimTf2N. We have presented preliminary investigations showing evidence that several common nonionic surfactants may display aggregation behavior within the low-viscosity RTIL, emimTf2N, on the basis of the changes observed in the solvatochromic probe response. This method may provide an initial indication that some sort of aggregation may be occurring within the ionic liquid solution; however, it has some limitations. Even at fairly low concentrations,
Letters
the probe may distort/interact with such aggregates, distorting/modifying them in the process. Alternatively, probe self-association or its interaction with the surfactant monomer may also produce changes in the solvatochromic probe response. Currently, we are in the process of performing surface tension and conductivity measurements as well as utilizing electron microscopy along with light scattering techniques to explore the structures, if any, of these aggregates as well as their size(s) and shape(s). Supporting Information Available: Probe response versus surfactant concentration for all (emimTf2N + surfactant) systems investigated. This material is available free of charge via the Internet at http://pubs.acs.org. LA035596T