A simple colorimetric method for the quality control of 1-alkyl-3-methylimidazolium ionic liquid precursors

Abstract

A simple colorimetric method to monitor the production of ionic liquid precursors is developed, which is based on the determination of 1-methylimidazole with copper(II) chloride. The synthesis of 1-ethyl-3-methylimidazolium chloride, an industrially important ionic liquid precursor, can be followed and the purity of the final product can be readily assessed in a quick and convenient manner.

---

Green Context

Ionic liquids are attracting a lot of attention as novel, VOC-free solvent systems. One of the barriers to acceptance industrially is the assurance of quality and reproducibility. A contribution to this is the development of a simple and rapid technique to monitor levels of the parent imidazole in the final product. The method is rapid, sensitive and requires only standard lab equipment.

DJM

Introduction

Room-temperature ionic liquids^1–5 have been widely studied as solvents for electrochemical technologies and more recently as clean, environmentally benign solvents for industrial chemical processes.^6–8 Room-temperature ionic liquids typically comprise an organic cation and weakly coordinating anion;³ systems of greatest current interest contain 1-alkyl-3-methylimidazolium cations (Fig. 1). Most ionic liquid systems are derived from chloride salts (of the 1-alkyl-3-methylimidazolium cation) either by mixing [i.e. tetrachloroaluminate(III) ionic liquids⁹] or via metathesis;¹⁰ direct alkylation routes have also been applied¹¹ The chloride salts may be prepared by alkylation of 1-methylimidazole with the corresponding chloroalkane.^9,12

Fig. 1 The structure of the 1-alkyl-3-methylimidazolium cation (n = 2 is conventionally designated as [emim]⁺).

However, despite the current level of interest in ionic liquids, from both academic groups and industry,^13–15 these materials are not readily available on a commercial basis. The key to the synthesis of ionic liquids is the relatively simple preparation of 1-alkyl-3-methylimidazolium chloride salts by alkylation of 1-methylimidazole with the corresponding 1-chloroalkane, in particular, the synthesis of 1-ethyl-3-methylimidazolium chloride ([emim]Cl). As part of a development program in the scale-up of ionic liquid production, it was necessary to monitor the synthesis of [emim]Cl and determine the purity of the final product using a procedure that could be applied at both a laboratory scale and under industrial production conditions as a quality control method. In particular, the presence of unreacted 1-methylimidazole starting material in the ionic liquids from incomplete alkylation poses potential downstream problems from catalyst poisoning and introduction of protic impurities. 1-Methylimidazole is a coordinating base, and has proved difficult to remove from ionic liquids at a later stage.

We present here a method to monitor the end-point of the alkylation reaction and determine the levels of unreacted 1-methylimidazole contamination to <0.2 mol%, based on the complexation of 1-methylimidazole with copper(II) chloride in ethanol to form the [Cu(mim)₄]²⁺ ion¹⁶ (mim = 1-methylimidazole), which has an intense blue colour. In contrast, the interaction of copper(II) chloride with [emim]Cl in ethanolic solution gives a yellow solution in the absence of 1-methylimidazole and provides the necessary spectral shift to allow a colorimetric determination based on the position of λ_max of the absorption band in the electronic absorption spectrum.

Experimental

1-Methylimidazole (ex Aldrich) was distilled from CaH₂ prior to use; ethanol (spectroscopic grade) and copper(II) chloride dihydrate (Analar) were used as received. Electronic absorption spectra of samples in 1 mm or 1 cm quartz cells were obtained using a Perkin Elmer Lambda-9 spectrometer over the range 500–1500 nm. A 3 l autoclave, equipped with overhead, magnetically coupled stirrer was manufactured by Strata Technology Ltd, Sunbury-on-Thames, UK.

Preparation of 1-ethyl-3-methylimidazolium chloride

1-Ethyl-3-methylimidazolium chloride ([emim]Cl) was prepared on a bulk scale from 1-methylimidazole and chloroethane. Samples were then taken and recrystallised twice from ethanenitrile–ethyl ethanoate and then stored under dinitrogen in a dry-box. Purity was confirmed by ¹H, and ¹³C NMR spectroscopy, CHN microanalysis, and determination of Cl⁻ content.

An autoclave (3 l) was charged with 1-methylimidazole (1030 g, 12.54 mol, 1000 cm³; freshly distilled from CaH₂) and cooled to 0 °C under dinitrogen. Chloroethane (1000 g, 19.8 mol) was condensed into a partially evacuated 2 l Schlenk tube and transferred to the autoclave, which was then sealed and pressurised to 5 bar with dinitrogen. The autoclave was then heated and stirred at 75 °C for 48 h. The reactor contents were transferred to a round-bottomed flask under dinitrogen and the excess of chloroethane was removed with heating at 70 °C under reduced pressure and finally in vacuo to give a colourless, viscous oil. The molten product was then transferred to a dry-box and poured into a shallow tray, where it crystallised on cooling as a white solid (yield 1818 g, 99%; C₆H₁₁ClN₂: found (calc.): C, 48.96 (49.15); H, 7.49 (7.56); N, 18.89 (19.10%); Mp 87 °C (lit. 84 °C⁹).

Job plot analysis and calibration curves

For the Job plot analysis, a standard solution (1000 cm³) of [emim]Cl (77.70 g, 0.530 mol) in ethanol was prepared ([emim]Cl concentration, 0.530 mol l⁻¹). Two stock solutions (250 cm³) were then prepared from copper(II) chloride CuCl₂·2H₂O (2.26 g, 13.3 mmol) with the standard ethanolic [emim]Cl solution, (53.2 mmol l⁻¹ in [Cu(II)]) and from 1-methylimidazole (1.09 g, 13.3 mol) with the standard ethanolic [emim]Cl solution (1-methylimidazole concentration 53.2 mmol l⁻¹).

Samples of 10 cm³ volume, in each case with constant overall [Cu(II)] + [mim] concentration, were prepared from the two stock solutions (1–10, Table 1) using the quantities indicated in Table 1; overall concentrations were 0.530 mol l⁻¹ in [emim]Cl and 53.2 mmol l⁻¹ in combined copper(II) chloride and 1-methylimidazole. Electronic absorption spectra were obtained over the range 500–1500 nm using 1 mm path length quartz cuvettes.

Table 1 Solutions prepared for Job plot composition analysis, indicating the volume of the standard solutions of Cu(II)Cl₂ and 1-methylimidazole (in ethanolic [emim]Cl) used, concentration of each component in the final solution and position of λ_max in the electronic absorption spectrum. Total [Cu(II)] + [mim] concentration was 53.2 mmol l⁻¹ and 0.530 mol⁻¹ in [emim]Cl for each solution

Cu(II)Cl2		mim
Solution	Volume/ cm^3	Conc./ mmol l^−1	Volume/ cm^3	Conc./ mmol l^−1	λmax/nm
1	10.0	53.2	0	0.0	1090
2	9.0	47.7	1.0	5.3	1046
3	8.0	42.4	2.0	10.6	996
4	7.0	37.1	3.0	15.9	939
5	6.0	31.8	4.0	21.2	898
6	5.0	26.5	5.0	26.5	856
7	4.0	21.2	6.0	31.8	856
8	3.0	15.9	7.0	37.1	773
9	2.0	10.6	8.0	42.4	735
10	1.0	5.3	9.0	47.7	648

---

For the calibration curve, a standard ethanolic solution (100 cm³) of copper(II) chloride solution (CuCl₂·2H₂O, 1.00 g, 5.87 mmol) and a stock solution (1000 cm³) of pure [emim]Cl (51.210 g, 0.3493 mol) in ethanol were prepared. Solutions (11–19, Table 2) of known [emim]Cl:[mim] concentration ratios were prepared from 1-methylimidazole (accurately measured in the range 0–0.2500 g, ca. 0–3 mmol) and the stock solution of [emim]Cl (100 cm³). Finally a 10 cm³ aliquot of the [emim]Cl/[mim] solutions in ethanol was taken and copper(II) chloride solution (1.0 cm³) was added, to give an immediate colour change. The overall copper concentration was 6.73 mmol l⁻¹, which was calculated to be sufficient to completely complex 1-methylimidazole at concentrations up to 27 mmol l⁻¹. Under the conditions chosen, this corresponds to ca. 8 mol% 1-methylimidazole in a sample of [emim]Cl. Electronic absorption spectra were obtained of the samples over the range 500–1500 nm using 1 cm path length quartz cuvettes.

Table 2 The maxima in the electronic absorbance spectra (λ_max) with concentration of 1-methylimidazole in the calibration solutions, containing fixed concentration of copper(II) (5.336 mmol l⁻¹) and an excess of [emim]Cl (317.8 mmol l⁻¹).

Solution	[mim]/mmol l^−1	λmax/nm
11	0.000	1093
12	0.000	1096
13	0.886	1065
14	2.768	963
15	3.875	945
16	6.864	873
17	11.072	796
18	14.725	777
19	25.686	754

---

Results and discussion

The change in the electronic absorbtion spectrum of the composite mixture [Cu(II)]–[emim]Cl–mim (Table 1) was investigated as a function of [Cu(II)]:[mim] ratio using a Job plot analysis¹⁷ to characterise the copper–imidazole species as a function of composition (Fig. 2). Electronic absorption spectra of 1–10 with constant overall concentration and varying [Cu(II)]:[mim] ratios were obtained. The [Cu(II)]:[mim] molar ratio was varied in the range 10–100 mol% Cu(II) while maintaining constant overall ionic concentration with a ten-fold excess of [emim]Cl and the electronic spectra taken over the range 500–1500 nm.

Fig. 2 Variation in the absorption maxima (λ_max) of the electronic spectra of solutions 1–10 with copper(II)–1-methylimidazole molar composition in ethanol. The solutions (Table 1) have a constant combined [Cu(II)] + [mim] concentration (53.2 mmol l⁻¹) and constant [emim]Cl concentration (0.530 mol l⁻¹).

The maximum values (λ_max) in the absorption spectra vary continuously and linearly with copper:[mim] concentration and are independent of the excess of [emim]Cl. This indicates a continuous gradual transition from a solvated copper(II) chromophore to a mononuclear [Cu(mim)₄]²⁺ chromophore on increasing [mim]:[Cu] ratio. This is in agreement with electron paramagnetic resonance (EPR) measurements on mixtures of copper(II) nitrate and 1-methylimidazole.¹⁸ The spectrum obtained at 4∶1 [mim]:[Cu(II)] (20% copper content) is identical to the spectrum of [Cu(mim)₄]Cl₂ dissolved in ethanol containing an excess of [emim]Cl.

The Job plot analysis shows that there is a continuous, gradual change in the absorption spectrum as a function of composition. The positions of the absorption maxima depend only on the relative Cu(II):mim ratio, and are independent of the [emim]Cl:mim ratio. Thus, by monitoring the change in λ_max with 1-methylimidazole concentration while maintaining constant copper concentration, a calibration curve can be constructed which will enable the 1-methylimidazole component of mixed [emim]Cl/1-methylimidazole samples to be determined by reference to the calibration curve, and the overall composition calculated from the initial weight of sample taken.

The absorption spectra of solutions of known composition with [mim]:[emim]Cl ratio in the range 0–5 mol% and methylimidazole concentration in the range 0–30 mmol l⁻¹, (11–19) with constant copper(II) concentration (Table 2) were measured, and λ_max plotted as a function of 1-methylimidazole concentration (Fig. 3). The copper concentration used was calculated to be sufficient to complex all the 1-methylimidazole over the composition range studied [up to ca. 27 mmol l⁻¹, equivalent to ca. 8 mol% in the samples 11–19 (Table 2)]. The solutions prepared varied from yellow (in the absence of 1-methylimidazole, [mim] = 0) progressively to blue–green as the 1-methylimidazole concentration was increased, with a shift in λ_max from 1100 to 750 nm (Figs. 4 and 5).

Fig. 3 The absorption maxima (λ_max) as a function of 1-methylimidazole concentration from Table 2. Data were fitted with an exponential curve y = a + b^(−x/c), a = 742.47, b = 355.89, c = 6.46; R²=0.996.

Fig. 4 Change in the electronic absorption spectra of the calibration standards (from Table 2) with 1-methylimidazole concentration; (a) 25.0 mmol l⁻¹, (b) 6.8 mmol l⁻¹, (c) 3.9 mmol l⁻¹, (d) 0.9 mmol l⁻¹, (e) 0 mmol l⁻¹. In each case, [Cu(II)] concentration is 5.336 mmol l⁻¹ and [emim]Cl concentration is 317.8 mmol l⁻¹.

Fig. 5 The progressive, visual change in colour of test solutions, on decreasing [mim] concentration, from blue–green (8 mmol l⁻¹) to yellow (0 mmol l⁻¹).

The absorption maximum with 1-methylimidazole composition varies following an exponential curve (Fig. 3) and demonstrates a very high sensitivity to 1-methylimidazole concentration. Under the experimental conditions chosen for this study, λ_max varies by 300 nm for a change in 1-methylimidazole concentration in the range 0–10 mmol l⁻¹. By changing the concentration of [Cu(II)] used, the sensitivity of the method can be tuned to different concentration ranges of 1-methylimidazole. Under the conditions described here, the upper limit in 1-methylimidazole concentration determination in the sample is ca. 27 mmol, at which point, all of the available copper ions are complexed to 1-methylimidazole as the blue, square-planar [Cu(mim)₄]²⁺ species.¹⁶ It is notable that the sensitivity of this method for determining the purity of [emim]Cl is greatest at low [mim] concentration, which is ideal for industrial process monitoring.

These results demonstrate that, under the range of conditions used, the response of the absorption maximum to 1-methylimidazole concentration is independent of the [emim]Cl concentration. This both confirms the conclusions drawn from the Job analysis and shows that the calibration plot can be used to determine the 1-methylimidazole content of unknown (i.e. experimental) samples of [emim]Cl of accurately known, but varying, weights. This, then, is the basis of the analytical procedure described below.

Analytical procedure

A standard procedure was developed to measure the 1-methylimidazole content of 1-ethyl-3-methylimidazolium chloride samples taken from the alkylation reaction. A sample of the reaction mixture (approximately 0.5 g, assumed to contain at least 95% [emim]Cl) was accurately weighed into a 10 cm³ volumetric flask and dissolved in ethanol, taking precautions to limit absorbance of moisture from the environment. A 1 cm³ aliquot of the standard ethanolic copper(II) solution (containing 1.0 g, 5.87 mmol in 100 cm³ ethanol) was then added and the absorption spectra recorded in the range 500–1500 nm using 1 cm path length cuvettes. The maximum, λ_max, was measured and compared to the standard calibration plot (Fig. 3) to determine the total concentration of 1-methylimidazole in the test solution. The molar ratio of 1-methylimidazole to [emim]Cl in the reaction mixture can then be calculated from this result and the weight of the sample.

Under the conditions used ([emim]Cl samples of ca. 0.5 g), 1-methylimidazole content can be determined in the range 0–3 mol%, corresponding to a concentration in the range 0–10 mmol l⁻¹ in the test solution. Over this concentration range, the position of λ_max changes by almost 300 nm and provides a very sensitive technique to monitor 1-methylimidazole content to better than ±0.2 mol% in [emim]Cl.

Conclusion

In conclusion, we present a quick, simple reliable method for following the important reaction between 1-methylimidazole and chloroethane and determining the purity of the product, [emim]Cl. The general method has been shown to also be applicable in general to the formation of other 1,3-dialkylimidazolium salts by alkylation of a functionalised imidazole with chloro- and bromo-alkanes. The procedure is equally applicable to a small scale laboratory synthesis, or to a multi-ton commercial process.

Acknowledgements

We would like to thank the ERDF Technology Development Programme and the QUESTOR Centre (J. D. H.) for financial support and the EPSRC and Royal Academy of Engineering for the award of a Clean Technology Fellowship (to K. R. S.). This research was also partially funded under a Brite-Euram III framework project (BRPR-CT97-0431/BE-96-3745).

References

1. C. L. Hussey, Adv. Molten Salt Chem., 1983, 5, 185.
2. C. L. Hussey, Pure Appl. Chem., 1988, 60, 1763.
3. Y. Chauvin and H. Olivier-Bourbigou, Chemtech, 1995, 25, 26.
4. K. R. Seddon, Kinet. Catal., 1996, 37, 693.
5. K. R. Seddon, J. Chem. Tech. Biotech., 1997, 68, 351.
6. J. D. Holbrey and K. R. Seddon, Clean Prod. Proc., 1999, 1, 233.
7. T. Welton, Chem. Rev., 1999, 99, 2071.
8. W. Keim and P. Wasserscheid, Angew. Chem., Int. Ed., 2000, 39, 3772.
9. J. S. Wilkes, J. A. Levisky, R. A. Wilson and C. L. Hussey, Inorg. Chem., 1982, 21, 1263.
10. J. S. Wilkes and M. J. Zaworotko, J. Chem. Soc., Chem. Commun., 1992, 965.
11. P. Bonhôte, A. P. Dias, N. Papageorgiou, K. Kalyanasundaram and M. Grätzel, Inorg. Chem., 1996, 35, 1168.
12. A. K. Abdul-Sada, P. W. Ambler, P. K. G. Hodgson, K. R. Seddon and N. J. Stewart, World Pat., WO95/21871, 1995..
13. M. Freemantle, Chem. Eng. News, 2000, 78(May 15), 37..
14. D. Adam, Nature, 2000, 407, 938.
15. R. D. Rogers, Green Chem., 2000, 2, G94.
16. W. Clegg, J. R. Nicholson, D. Collison and C. D. Garner, Acta Crystallogr., Sect. C, 1988, 44, 453.
17. P. Job, Ann. Chim., 1928, 9, 113.
18. P. J. Baesjou, W. L. Driessen, G. Challa and J. Reedijk, J. Mol. Catal. A, 1996, 110, 195.

---