In Situ Crystallization of Ionic Liquid [Emim][PF6] from Methanol

The solubility of 1-ethyl-3-methylimidazolium hexafluorophosphate ([Emim][PF6]) in methanol under high pressure is newly measured quantitatively accor...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/JPCB

In Situ Crystallization of Ionic Liquid [Emim][PF6] from Methanol Solution under High Pressure Haining Li,† Lei Su,*,† Xiang Zhu,† Xuerui Cheng,† Kun Yang,† and Guoqiang Yang*,‡ †

The High Pressure Research Center of Science and Technology, Zhengzhou University of Light Industry, Zhengzhou, 450002, China ‡ Key Laboratory of Photochemistry, Institute of Chemistry, Chinese Academy of Sciences, Beijing, 100080, China ABSTRACT: The solubility of 1-ethyl-3-methylimidazolium hexafluorophosphate ([Emim][PF6]) in methanol under high pressure is newly measured quantitatively according to the correlation between the ratios of Raman intensity and the concentrations. In situ crystallization and cation conformation of [Emim][PF6] from methanol solution under high pressure have been investigated by using Raman spectroscopy in detail. Remarkably, crystal polymorphism was observed and two crystalline phases (phases I and II) coexisted under high pressure up to ∼1.4 GPa. However, only phase II was obtained by recrystallization at ∼2 GPa. Our findings may facilitate the development of an effective way for crystallization and purification of ionic liquids under high pressure.

1. INTRODUCTION Recently, crystallization of ionic liquids (ILs) is receiving great attention. Some scientists tried to give insight into the phase transition of ILs from the viewpoint of a solid-state chemist.1 The value of crystallization for purification of ILs has been widely realized. However, homogeneous crystallization of ILs from the melt is often inhibited. Some ILs solidify in the form of glasses, plastic crystals, or liquid crystals, and several experimental techniques are developed to crystallize ILs for purification, such as zone melting in the presence of seed crystals2,3 and layer crystallization.4 Besides the methods above, ILs could also be crystallized from solution. In preparation of ILs, the sample was often dissolved in the volatile solvent, which could be evaporated upon heating to purify ILs. For example, [Emim][PF6] (Emim = 1-ethyl-3-methylimidazolium) was crystallized from methanol5 and [Emim]Br, [Emim]I,6 and [Bmim]Cl7 (Bmim = 1-butyl-3-methylimidazolium) were crystallized from acetonitrile. Their success highly depended on the melting point and chemical nature of the IL as well as on the physicochemical properties of the chosen solvent. Crystallization from solution is currently by far the most prominent technique used for crystal growth of IL material. Temperature and concentration are important parameters when growing crystals from solution. In fact, there is another parameter that can be tuned, that is, pressure. It will be very intriguing to explore crystallization from solution under high pressure, which may provide an effective way for purification of ILs that do not tend to crystallize from the melt upon cooling. Moreover, the solubility of ILs in organic solvent has also led to growing interest. Urszula Domańska et al. systematically studied the solubility of ILs in organic solvent by a dynamic © 2014 American Chemical Society

method from 290 K to the melting point of IL or to the boiling point of the solvent at ambient pressure, such as [Emim][PF6] in alcohols8 and 1-butyl-, decyl-, or dodecyl-3-methylimidazolium chloride [C4, C10, or C12mim][Cl] in alcohols.9−12 They also measured the solid−liquid phase equilibrium (SLE) in binary mixtures that contained an IL and an organic solvent: {1-ethyl-3-methylimidazolium tosylate, [Emim][TOS] + cyclohexane, or benzene} and {1,3-dimethylimidazolium methylsulfate, [Mmim][CH3SO4] + hexan-1-ol} under high pressures up to about 900 MPa in the temperature range from 328 to 363 K.13 A piston−cylinder device was combined with a water thermostat, and liquid−solid phase transitions were determined by discontinuities of volume−pressure curves. The experiment procedures were complicated, and the “overpressure” effect might affect the accuracy of the discontinuity. In this paper, it was attempted to obtain the solubility of [Emim][PF6] in methanol under high pressure by using Raman spectroscopy and a diamond anvil cell (DAC). A procedure to quantify the concentration of [Emim][PF6] in methanol was developed, which could provide a novel and simple method of quantitative measurement of solubility under high pressure. Meanwhile, in situ crystallization of [Emim][PF6] from methanol solution was investigated in detail. The crystalline phase of [Emim][PF6] was determined under different pressures and different compression processes. Received: January 24, 2014 Revised: June 11, 2014 Published: June 26, 2014 8684

dx.doi.org/10.1021/jp500866z | J. Phys. Chem. B 2014, 118, 8684−8690

The Journal of Physical Chemistry B

Article

Figure 1. (a) Raman spectra of pure methanol, pure [Emim][PF6], and the solution of 10 wt % [Emim][PF6] in methanol. (b) Relationship between the ratios of peak areas and relative concentrations (mol) of [Emim][PF6] and methanol.

where σ, η, and F are the Raman scattering cross section, instrumental efficiency, and Raman quantification factor, respectively. This approach has been successfully applied to a CH4−H2O system in aqueous solution.17,18 The Raman spectra of pure methanol, pure [Emim][PF6], and the solution of 10 wt % [Emim][PF6] in methanol under ambient temperature (297 K) are shown in Figure 1a. It can be seen that there is a single observable Raman band at 738 cm−1, which is assigned to the frequencies of P−F symmetric stretching of [Emim][PF6],19 and the band at 2834 cm−1, which is assigned to C−H symmetric stretching of the methanol.20 The relationship between Raman peak areas and relative concentrations (mol) of [Emim][PF6] and methanol is shown in Figure 1b. By linear fitting, the correlation between the ratios of Raman intensity and the concentrations could be described by

2. EXPERIMENTAL SECTION [Emim][PF6] was supplied by Henan Lihua Pharmaceutical Co., China, whose purity was more than 99.5 wt %. Before all of the measurements, the sample was kept under a vacuum at a moderate temperature (353 K) for at least 3 days to reduce the moisture content and volatile compounds to negligible values. Its molecular weight is 256.13 g/mol, and the melting point was reported to be 331−333 K5 or 335 K.14 Methanol was obtained from Sinopharm Chemical Reagent Co., Ltd., China, whose purity was more than 99.5 wt %. A DAC, with a diamond culet size of 500 μm, was used for generating pressures up to about 2 GPa. The sample was contained in a 200 μm diameter hole in a T301 gasket which was preindented to a thickness of about 100 μm and clamped between the two diamond anvils. Pressures were calculated from the shift of the ruby R1 fluorescence line.15 The Raman experiments are carried out using a Renishaw inVia Raman microscope (Renishaw, U.K.) with 532 nm wavelength excitation. Raman spectra were collected in a backscattering geometry with a 2400 gr/mm holographic grating, and the slit width was selected as 65 μm corresponding to a resolution of ca. 0.5 cm−1. The sample image can be collected through an achromatic lens and then focused onto a CCD detector for visual monitoring during experiments. The samples were held under each pressure and temperature for enough time until the equilibrium was established. The obtained Raman spectra were fitted with a Gaussian− Lorentzian mixing function using the WIRE 3.3 software (Renishaw, U.K.) for analyzing the spectral data.

S[Emim][PF6]/SMethanol = 0.71152C[Emim][PF6]/C Methanol

(2)

To investigate the influence of pressure on correlation between the ratios of Raman intensity and the concentrations, a certain concentration unsaturated solution of [Emim][PF6] in methanol was sealed in DAC under ambient temperature (297 K). The ratios of the Raman intensity of [Emim][PF6] to that of methanol under high pressure are shown in Figure 2. It can be seen that S[Emin][PF6]/SMethanol remains constant with

3. RESULTS AND DISCUSSION The intensity of the Raman vibrational band is proportional to the amount of molecules or ions, and it may theoretically serve as a plausible parameter for quantitatively determining their concentrations. However, the effect of any fluctuation in the laser position and intensity, the time of scanning, and scattering cross sections of the sample would cause unpredictable variations of Raman intensity, which would render it unreliable and therefore unsuitable as a quantitative descriptor of concentration. For two Raman active species a and b in a fluid phase, relative concentrations C (e.g., molar or mol %) are related to their Raman peak areas S which could be expressed by16 Sa /S b = (Ca /C b)(σa /σb)(ηa /ηb) = (Ca /C b)(Fa /Fb)

Figure 2. Ratios of the Raman intensity of [Emim][PF6] to that of methanol under high pressure.

(1) 8685

dx.doi.org/10.1021/jp500866z | J. Phys. Chem. B 2014, 118, 8684−8690

The Journal of Physical Chemistry B

Article

increasing pressure. Therefore, the influence of pressure on eq 2 can be neglected under pressures up to about 1.6 GPa, and relative concentrations can still be calculated by eq 2 under the given experimental conditions. A saturated solution of [Emim][PF6] in methanol was sealed in a DAC under ambient temperature (297 K). It was proved to be saturated solution by the crystals in the cell, which is shown in Figure 3a. With increasing pressure, Raman spectra of the

Figure 4. Solubility of [Emim][PF6] in methanol under high pressure: (a) compression process and (b) recrystallization process at room temperature (297 K).

observation could also provide clear evidence on the change of solubility of [Emim][PF6] in methanol under high pressure, which is shown in Figure 3a−d. It could be seen that [Emim][PF6] crystallized from saturated solution with increasing pressure. The growth of the crystals was rapid upon compression initially; however, the size of [Emim][PF6] crystals was almost the same under pressures above ∼0.9 GPa, which was in accordance with the change of solubility of [Emim][PF6] in methanol under high pressure in Figure 4a. The high density crystals with high stability under high pressures formed, so the solubility decreased sharply below ∼0.9 GPa. The reason why the solubility remained constant above ∼0.9 GPa is that there might be a dissolution equilibrium of [Emim][PF6] in methanol under high pressure. Compared with the previous study about the influence of high pressure on the solubility of ILs using a piston−cylinder device,13 it could be seen that the method by using DAC and Raman spectra to quantitatively measure solubility under high pressure was superior. First, the higher pressure could be obtained by using DAC, which means that the solubility under higher pressure could be obtained by this method. Second, only one liquid−solid phase transition point could be obtained after one assembling in the previous method, while the solubility under different pressures could be obtained by this method after one assembling of the DAC. That was to say, this method was simpler and more convenient. Third, the transparency of diamonds of the DAC allowed in situ observation for the

Figure 3. Photographs of the DAC under high pressures. (a−d) crystallization from solution upon compression at room temperature (297 K); (e) melting completely after heating; (f−i) recrystallization upon compression at room temperature (297 K) after being cooled naturally.

saturated solution were measured by focusing the laser on more than four different positions of the solution until the ratio of Raman intensity was constant. That was to say, the spectra were taken until the equilibrium was established and the ratio was averaged to stand for the equilibrium in the whole cell. According to the correlation between the ratios of Raman intensity and the concentrations, the solubility of [Emim][PF6] in methanol in the form of the relative concentrations (mol) of [Emim][PF6] and methanol (C[Emin][PF6]/CMethanol) under high pressure could be obtained by the Raman spectra of the solution, which was plotted in Figure 4a. The solubility decreased sharply at the initial stage but remained almost constant in the range from 0.9 to 1.4 GPa. Microscopic 8686

dx.doi.org/10.1021/jp500866z | J. Phys. Chem. B 2014, 118, 8684−8690

The Journal of Physical Chemistry B

Article

Figure 5. Raman spectra of the original crystals of [Emim][PF6] in methanol that already exist initially under high pressure: (a) the CH2(N) bending band reflecting the conformation of [Emim][PF6]; (b) the C−H stretching vibrational region.

phenomenon might also be a kind of “overpressure” effect, which is similar to that in the literature.13 With increasing pressure up to 2.0 GPa, [Emim][PF6] recrystallized from the solution abruptly (Figure 3i). The understanding of these processes was of fundamental importance, although the first step of formation of nucleation was still under intense scrutiny. The solubility of [Emim][PF6] in methanol under high pressure during the recrystallization is shown in Figure 4b. It could be seen that the solubility remained constant up to ∼1.7 GPa and decreased sharply at ∼2.0 GPa, which was in agreement with the photographs in the DAC (Figure 3f−i). Raman spectra of the crystals of [Emim][PF6] in methanol were measured by focusing the laser on different parts of the crystal. Figures 5 and 6 presented the spectra of the original crystal which already existed initially under ambient pressure and the newly grown crystal, which were located on positions A and B, respectively, in Figure 3b. The crystal of [Emim][PF6] was surrounded by the methanol solution, so the spectra were a combination of the crystal of [Emim][PF6] and methanol solution. For comparison, the Raman spectra and assignment of pure [Emim][PF6] and methanol were shown in the same figure. According to previous studies,21 [Emim]+ has two stable conformers for the CH2(N) bending, the trans (planar) form and the gauche (nonplanar) form, due to the rotational isomerism of the ethyl group. Raman bands at 241, 297, 387, 430, and 448 cm−1 of [Emim][PF6] are assigned to the gauche, gauche, gauche, gauche, and trans conformers, respectively.

sample under an optical microscope, which could provide direct evidence of the solid−liquid phase equilibrium. Finally, on a microscopic level, crystallization occurred stepwise and could be divided into nucleation and crystal growth. In a previous method, discontinuities of volume−pressure curves were used to determine the liquid−solid phase transitions. The initiation of freezing was noticed under a higher pressure than the equilibrium value of the phase transition, which was observed as an “overpressure” effect.13 Actually, the “overpressure” effect was a kind of supersaturation, which was also observed in our study. If the saturated solution was contained in the DAC without a small amount of crystals, [Emim][PF6] did not separate out as soon as the pressure was increased but crystallized from the solution abruptly at a higher pressure. In other words, small crystals already assembled in the DAC in this study served as crystal nuclei. The crystallization under high pressure was just a crystal growth process, which meant that the static solubility was measured in this method. Therefore, a novel and simple method of quantitative measurement of solubility under high pressure was provided in this study. In order to investigate the recrystallization of [Emim][PF6] in methanol solution, the [Emim][PF6]/methanol system was heated until the crystals of [Emim][PF6] dissolved in the methanol solution gradually (Figure 3e). Then, the sample was cooled naturally to room temperature (297 K) by turning off the power. Surprisingly, [Emim][PF6] did not crystallize from the solution (Figure 3f), and it could be speculated that this 8687

dx.doi.org/10.1021/jp500866z | J. Phys. Chem. B 2014, 118, 8684−8690

The Journal of Physical Chemistry B

Article

Figure 6. Raman spectra of the newly grown crystal of [Emim][PF6] in methanol under high pressure: (a) the CH2(N) bending band reflecting the conformation of [Emim][PF6]; (b) the C−H stretching vibrational region. The asterisks represent new peaks.

Figure 7. Raman spectra of different parts of the [Emim][PF6] crystal after recrystallization at ∼2.0 GPa: (a) the CH2(N) bending band reflecting the conformation of [Emim][PF6]; (b) the C−H stretching vibrational region. The asterisks represent new peaks.

From Figure 5a, it can be seen that there is no Raman band at 448 cm−1 during the whole experimental process, and it means that the conformation of [EMIM]+ in the original crystal is gauche, not trans under high pressure, which is the same as that in pure [Emim][PF6].22 The Raman spectra shape of C−H stretching vibration in Figure 5b is almost the same under high pressure. Therefore, it could be concluded that the original

crystal of [Emim][PF6] in methanol remained a crystalline phase (hereafter designated as phase I) up to 1.4 GPa, which showed a gauche conformer. As shown in Figure 6a, the newly grown crystal of [Emim][PF6] in methanol also presented a gauche conformer. Intriguingly, there was a split of the band at 469 cm−1 which was assigned to P−F symmetric bending.19 In addition, as for 8688

dx.doi.org/10.1021/jp500866z | J. Phys. Chem. B 2014, 118, 8684−8690

The Journal of Physical Chemistry B

Article

solution, but the gauche conformation might be more beneficial for crystallization and more stable than the trans one under high pressure. In summary, the solubility of [Emim][PF6] in methanol under high pressure was quantitatively measured by Raman spectra according to the correlation between the ratios of Raman intensity and the concentrations. The relative concentrations (mol) of [Emim][PF6] and methanol decreased with compression and remained constant under the pressure above 0.9 GPa. A novel and simple method of quantitative measurement of solubility under high pressure was provided. The investigation of SLE under high pressure was of important significance for crystallization and purification under high pressure regardless of providing a good tool for examining the thermodynamic nature of many systems. Meanwhile, in situ crystallization of [Emim][PF6] from methanol solution under high pressure was investigated in detail. Remarkably, crystal polymorphism was observed under the pressure. The original crystal and the newly grown crystal in the solution showed different phases (phase I and phase II) under high pressures up to 1.4 GPa; however, only phase II was obtained by recrystallization, which might be due to kinetic effects. Meanwhile, phase II with the gauche conformer is more stable under high pressure. Our findings might facilitate the development of an effective way for crystallization and purification of ionic liquids under high pressure.

the C−H stretching vibrational region in Figure 6b, the full width at half-maximum (fwhm) of the band at 2984 cm−1 was obviously larger than that of the crystal under atmospheric pressure, and it was divided into two peaks under higher pressure. According to previous studies,20 methanol crystallized at ∼3.5 GPa at room temperature from sharp split peaks of Raman spectra. Actually, methanol was difficult to crystallize but usually formed a superpressed liquid, and the Raman bands remained broad and no band splits with pressure up to 9.2 GPa. Therefore, the nonsplit Raman band of methanol occurred in this study with pressure below ∼2.0 GPa and the new peak that appeared at 2984 cm−1 belonged to the newly grown crystal of [Emim][PF6]. From the results above, it implied that the newly grown crystal of [Emim][PF6] in methanol was another crystal phase of [Emim][PF6] (designated as phase II). In other words, the original crystal and the newly grown crystal in the solution showed different phases under high pressure up to 1.4 GPa; however, both phases presented only a gauche conformer. Raman spectra of the recrystallization phase were also investigated under pressures up to 2.0 GPa, as shown in Figure 7. Raman spectra of different parts of the crystal were similar in the characteristic of the new peaks for the P−F symmetric bending (evolving into a sharp peak and a shoulder under high pressure, Figure 7a) and the C−H stretching vibrational modes (Figure 7b). That is, the recrystallization phase under a pressure of 2.0 GPa is similar to phase II of the newly grown crystal under high pressure. In addition, there are no Raman bands at 448 cm−1 in Figure 7, and the [EMIM]+ conformation of the crystal after recrystallization is also a gauche former. In this study, phase I and phase II coexisted under high pressure up to 1.4 GPa, and only phase II was obtained by recrystallization under pressures up to 2.0 GPa. For high pressure crystal phases, Mollard et al. discussed decompressioninduced crystallization in hydrated silica-rich melts, and the authors thought that the crystallization was markedly dependent on effective undercooling and melt dehydration as the pressure decreases.23 Fanetti et al. found that pyridine could either crystallize in phase II or give rise to a glass, and phase I was never obtained by compression of the fluid but only decompression of the sample brought to the formation of phase I.24 Fanetti et al. thought that the missed crystallization of phase I upon compression was more likely related to the kinetic effect. In this study, the newly grown crystals upon compression and the crystals obtained through the thermodynamic process and recrystallization were phase II, which might be more stable under high pressure. However, phase I originated from assembling crystal nuclei of phase I. From the results above, it could be speculated that this phenomenon might also be induced by a kinetic effect, which is similar to Fanetti’s result.24 Furthermore, phase I and phase II coexisted in a certain pressure range, which suggested that there are similar energies for both phases and high energy barriers among them making the transformations kinetically driven. It should be pointed out that, as for the conformation of [Emim][PF6] crystals in methanol under high pressure, only a gauche conformer was observed, just as the expectation that only certain conformations are crystallized from the solution.25 Fabbiani et al. reported that high-pressure crystallization of an aqueous solution of the GABA analogue gabapentin resulted in the formation of a previously unknown structure.26 They thought the crystal structure can be seen as a “snapshot”one of manyof what happens in solution. In this study, the trans conformation and the gauche conformation coexisted in the



AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work is supported by the National Science Foundation of China (No. 21273206 and No. 31201377), Natural Science Foundation of Henan Province (No. 2010GGJS-110), and the Basic Research Plan on Natural Science of the Henan Province education department of China (No. 2009A140009).



REFERENCES

(1) Mudring, A. V. Solidification of ionic liquids: theory and techniques. Aust. J. Chem. 2010, 63, 544−564. (2) Choudhury, A. R.; Winterton, N.; Steiner, A.; Cooper, A. I.; Johnson, K. A. In situ crystallization of low-melting ionic liquids. J. Am. Chem. Soc. 2005, 127, 16792−16793. (3) Choudhury, A. R.; Winterton, N.; Steiner, A.; Cooper, A. I.; Johnson, K. A. In situ crystallization of ionic liquids with melting points below −25°C. CrystEngComm 2006, 8, 742−745. (4) König, A.; Stepanski, M.; Kuszlik, A.; Keil, P.; Weller, C. Ultrapurification of ionic liquids by melt crystallization. Chem. Eng. Res. Des. 2008, 86, 775−780. (5) Fuller, J.; Carlin, R. T.; De Long, H. C.; Haworth, D. Structure of 1-ethyl-3-methylimidazolium hexafluorophosphate: model for room temperature molten salts. J. Chem. Soc., Chem. Commun. 1994, 299. (6) Elaiwi, A.; Hitchcock, P. B.; Seddon, K. R.; Srinivasan, N.; Tan, Y. M.; Welton, T.; Zora, J. A. Hydrogen bonding in imidazolium salts and its implications for ambient-temperature halogenoaluminate (III) ionic liquids. J. Chem. Soc., Dalton Trans. 1995, 3467−3472. (7) Nockemann, P.; Thijs, B.; Driesen, K.; Janssen, C. R.; Van Hecke, K.; Van Meervelt, L.; Kossmann, S.; Kirchner, B.; Binnemans, K. Choline saccharinate and choline acesulfamate: Ionic liquids with low toxicities. J. Phys. Chem. B 2007, 111, 5254−5263.

8689

dx.doi.org/10.1021/jp500866z | J. Phys. Chem. B 2014, 118, 8684−8690

The Journal of Physical Chemistry B

Article

(8) Domańska, U.; Marciniak, A. Solubility of ionic liquid [emim][PF6] in alcohols. J. Phys. Chem. B 2004, 108, 2376−2382. (9) Domańska, U.; Bogel-Łukasik, E.; Bogel-Łukasik, R. Solubility of 1-dodecyl-3-methylimidazolium chloride in alcohols (C2-C12). J. Phys. Chem. B 2003, 107, 1858−1863. (10) Domańska, U.; Bogel-Łukasik, E.; Bogel-Łukasik, R. 1-Octanol/ Water Partition Coefficients of 1Alkyl-3-methylimidazolium Chloride. Chem.Eur. J. 2003, 9, 3033−3041. (11) Domańska, U.; Bogel-Łukasik, E. Measurements and correlation of the (solid+liquid) equilibria of [1-decyl-3-methylimidazolium chloride+alcohols (C2-C12)]. Ind. Eng. Chem. Res. 2003, 42, 6986− 6992. (12) Domańska, U.; Bogel-Łukasik, E. Solid−liquid equilibria for systems containing 1-butyl-3-methylimidazolium chloride. Fluid Phase Equilib. 2004, 218, 123−129. (13) Domańska, U.; Morawski, P. Influence of high pressure on solubility of ionic liquids: experimental data and correlation. Green Chem. 2007, 9, 361−368. (14) Ngo, H. L.; LeCompte, K.; Hargens, L.; McEwen, A. B. Thermal properties of imidazolium ionic liquids. Thermochim. Acta 2000, 357, 97−102. (15) Mao, H. K.; Bell, P. M.; Shaner, J. T.; Steinberg, D. J. Specific volume measurements of Cu, Mo, Pd, and Ag and calibration of the ruby R1 fluorescence pressure gauge from 0.06 to 1 Mbar. J. Appl. Phys. 1978, 49, 3276−3283. (16) Wopenka, B.; Pasteris, J. D. Limitations to quantitative analysis of fluid inclusions in geological samples by laser Raman microprobe spectroscopy. Appl. Spectrosc. 1986, 40, 144−151. (17) Lu, W. J.; Chou, I. M.; Burruss, R. C.; Yang, M. Z. In Situ Study of Mass Transfer in Aqueous Solutions under High Pressures via Raman Spectroscopy: A New Method for the Determination of Diffusion Coefficients of Methane in Water Near Hydrate Formation Conditions. Appl. Spectrosc. 2006, 60, 122−129. (18) Lu, W.; Chou, I. M.; Burruss, R. C. Determination of methane concentrations in water in equilibrium with sI methane hydrate in the absence of a vapor phase by in situ Raman spectroscopy. Geochim. Cosmochim. Acta 2008, 72, 412−422. (19) Berg, R. W.; Deetlefs, M.; Seddon, K. R.; Shim, I.; Thompson, J. M. Raman and ab initio studies of simple and binary 1-alkyl-3methylimidazolium ionic liquids. J. Phys. Chem. B 2005, 109, 19018− 19025. (20) Mammone, J. F.; Sharma, S. K.; Nicol, M. Raman spectra of methanol and ethanol at pressures up to 100 kbar. J. Phys. Chem. 1980, 84, 3130−3134. (21) Umebayashi, Y.; Fujimori, T.; Sukizaki, T.; Asada, M.; Fujii, K.; Kanzaki, R.; Ishiguro, S. I. Evidence of conformational equilibrium of 1-ethyl-3-methyl imidazolium in its ionic liquid salts: Raman spectroscopic study and quantum chemical calculations. J. Phys. Chem. A 2005, 109, 8976−8982. (22) Endo, T.; Masu, H.; Fujii, K.; Morita, T.; Seki, H.; Sen, S.; Nishikawa, K. Determination of Missing Crystal Structures in the 1Alkyl-3-methylimidazolium Hexafluorophosphate Series: Implications on Structure−Property Relationships. Cryst. Growth Des. 2013, 13, 5383−5390. (23) Mollard, E.; Martel, C.; Bourdier, J. L. Decompression-induced crystallization in Hydrated silica-rich melts: empirical models of experimental plagioclase nucleation and growth kinetics. J. Petrol. 2012, 53, 1743−1766. (24) Fanetti, S.; Citroni, M.; Bini, R. Structure and reactivity of pyridine crystal under pressure. J. Chem. Phys. 2011, 134, 204504. (25) Glusker, J. P.; Lewis, M.; Rossi, M. Crystal Structure Analysis for Chemists and Biologists; Wiley-VCH Publishing: New York, 1994; pp 455−456. (26) Fabbiani, F. P.; Levendis, D. C.; Buth, G.; Kuhs, W. F.; Shankland, N.; Sowa, H. Searching for novel crystal forms by in situ high-pressure crystallisation: the example of gabapentin heptahydrate. CrystEngComm 2010, 12, 2354−2360.

8690

dx.doi.org/10.1021/jp500866z | J. Phys. Chem. B 2014, 118, 8684−8690