Highly selective formation of 2,6-dimethylnaphthalene in HCl-modified triethylamine–aluminum chloride ionic liquid

Abstract

HCl-modified triethylamine aluminum chloride ionic liquid (Et₃NHCl–2.0AlCl₃–HCl) was prepared and ²⁷Al NMR measurements were carried out. The acidity of the ionic liquids (ILs) was characterized by FT-IR with using pyridine as a probe molecule and the Hammett acidity function was determined by UV-vis spectroscopy. The catalytic performance of the ionic liquids for the synthesis of 2,6-dimethylnaphthalene (2,6-DMN) with high selectivity by the transalkylation of 2-methylnaphthalene (2-MN) with 1,2,4,5-tetramethylbenzene (TeMB) was investigated. Under the optimal reaction conditions, 15.6% conversion of 2-MN and 100% selectivity of 2,6-DMN in the dimethylnaphthalenes were obtained.

---

Introduction

2,6-Dimethylnaphthalene (2,6-DMN) is a key starting compound in the synthesis of polyethylene naphthalate (PEN), a polymer with extensive applications in packaging materials, electronic components, aviation materials, spaceflight material, and so on.¹ Currently, there are three main routes for the preparation of 2,6-DMN:² (1) extracting procedures applied to refinery stream; (2) multi-step synthesis starting from o-xylene; (3) one-step synthesis by alkylation of naphthalene (NAPH) or MN. Compared to the extraction procedures and the multistep synthetic process, one-step synthesis is simple and desirable.³ However, as many as ten possible dimethylnaphthalene (DMN) isomers are obtained and the isomers are difficult to separate, especially in the case of 2,6-DMN and 2,7-DMN, whose boiling points differ by only 0.3 °C.⁴

Thus, the selective synthesis of 2,6-DMN is rather difficult using the alkylation of NAPH or MN with conventional Friedel–Crafts catalysts (such as HF, H₂SO₄ and AlCl₃), because of the similar amounts of 2,6-DMN and 2,7-DMN isomers (2,6-DMN and 2,7-DMN) in the thermodynamic equilibrium; 12.8% and 12.7%, respectively.⁵ Other problems include environmental pollution, product separation and purification.⁶ Various zeolites, such as ZSM-5,⁷ ZSM-12,⁵ mordenite and modified zeolites, have been used for the alkylation of NAPH and MN as environmentally friendly catalysts.⁸ The shape-selective properties of zeolites in the alkylation of NAPH and MN were used to increase the selectivity for 2,6-DMN,⁹ with ZSM-12 predicted to be the most promising zeolite catalyst for the selective synthesis of 2,6-DMN.¹⁰ But the maximum selectivity for 2,6-DMN was 30.5%, with a 2,6-DMN/2,7-DMN ratio of 2.3.¹¹ In the presence of zeolites, high temperatures and pressures (300–400 °C, 3–4 MPa) were necessary and these catalysts deactivated readily. In the light of this, development of a highly efficient and environmentally friendly catalytic system for the selective synthesis of 2,6-DMN under mild conditions remains challenging.

In recent years, room temperature ionic liquids have received increased attention as green solvents and catalytic media.¹² The first successful use of ionic liquids (dialkylimidazolium chloroaluminate) as the catalyst in Friedel–Crafts acylations was reported in 1986.¹³ Recently, the alkylation of 2-methylnaphthalene with long-chain alkenes has been studied in the presence of ionic liquids, and a selectivity of 100% to the monoalkyl methylnaphthalenes was obtained under the optimal reaction conditions.¹⁴ However, the distribution of the alkylate was not studied. In our laboratory, we worked to develop ionic liquid catalysts for the selective synthesis of 2,6-DMN.¹⁵ When the N-alkylpyridinium halides–aluminum chloride ionic liquids were used as catalysts in the transalkylation of 2-MN with TeMB, a selectivity of 100% to 2,6-DMN in the total DMN isomers was obtained, but the conversion of 2-MN was very low (7.9%).¹⁶ So, it remains necessary for a highly effective catalyst for synthesis of 2,6-DMN with high conversion and selectivity to be developed. Compared to N-alkylpyridinium halides–aluminum chloride, triethylamine aluminum chloride ionic liquids (Et₃NHCl–2.0AlCl₃) are easily prepared and inexpensive. In this work, HCl-modified triethylamine aluminum chloride ionic liquids (Et₃NHCl–2.0AlCl₃–HCl) were prepared and used as catalysts for the synthesis of 2,6-DMN from 2-MN and TeMB by transalkylation. A higher 2-MN conversion and a high selectivity to 2,6-DMN were successfully achieved.

Experimental section

Materials and equipment

All chemicals (analytical reagent grade) were commercially available and the solvent was dried by 4 Å molecular sieves. The ²⁷Al NMR measurements were carried out using a Bruker AV-400 NMR. The FT-IR spectra were obtained on a PerkinElmer Spectrum 100 spectrometer, and the UV-vis spectra were recorded on a Shimadzu UV3600 spectrophotometer. The reaction mixtures were qualitatively analyzed by GC-MS (Agilent 6890/5973N) and quantitatively analyzed by GC on an Agilent 6890 instrument equipped with a FID detector.

Synthesis and characterization of the ILs

Et₃NHCl–2.0AlCl₃ was prepared by slow addition of the desired amount of anhydrous AlCl₃ to Et₃NHCl under dry nitrogen. When anhydrous AlCl₃ was completely dissolved, the temperature was increased to 80 °C and the mixtures were stirred for 2 h under a dry nitrogen atmosphere. Et₃NHCl–2.0AlCl₃–HCl was prepared by stirring the ionic liquid under a HCl gas atmosphere according to the literature method.¹⁷

The ²⁷Al NMR measurements were carried out to investigate the aluminum complex qualitatively in ILs, the acidity of the ILs was characterized by the pyridine adsorbed FT-IR spectra (Py-IR) and the Hammett acidity function of ionic liquids was determined using mesitylene as the basic indicator by UV-vis spectroscopy (Hammett method).

Transalkylation of 2-MN with TeMB

In a typical reaction, ionic liquid (15 mmol), 2-MN (7.6 mmol), TeMB (7.6 mmol) and cyclohexane (10 ml) were added to a three-necked flask under dry N₂ at the reaction temperature, then the mixture was stirred vigorously. After a given reaction time, the upper layer of the reaction mixture was separated from the ionic liquid phase by decantation. The reaction mixture was qualitatively analyzed by GC-MS and quantitatively analyzed by GC with FID detector.

Results and discussion

Characterization of the ILs

The acidities of Et₃NHCl–2.0AlCl₃ and Et₃NHCl–2.0AlCl₃–HCl were characterized by FT-IR using pyridine as a probe molecule (Fig. 1).^17,18 In Fig. 1, it is shown that both Et₃NHCl–2.0AlCl₃ and Et₃NHCl–2.0AlCl₃–HCl exhibited two bands near 1450 cm⁻¹ and 1540 cm⁻¹, which indicated the existence of Lewis acid and Brønsted acid in the ionic liquids, respectively. In the FT-IR spectrum of Et₃NHCl–2.0AlCl₃–HCl, the band’s blueshift, located from 1537 cm⁻¹ to 1540 cm⁻¹, is shown, which indicated that the strength of the Brønsted acidity was enhanced after modifying Et₃NHCl–2.0AlCl₃ by HCl.

Fig. 1 FT-IR spectra of ionic liquids using pyridine as a probe molecule: a Et₃NHCl–2.0AlCl₃ b Et₃NHCl–2.0AlCl₃–HCl.

When mesitylene was used as the basic indicator to trap the dissociative proton, the Brønsted superacid properties of the protons were studied by the determination of the Hammett acidity function (H₀) with UV-vis spectroscopy (Hammett method).¹⁹ In the present case, this method consists of evaluating the protonation extent of uncharged mesitylene (named I) in cyclohexane, in terms of the measurable ratio [I]/[IH⁺]. The maximal absorbance of the unprotonated mesitylene was observed at 220 nm in cyclohexane (Fig. 2, upper) and it decreased in varying degrees when Et₃NHCl–2.0AlCl₃ and Et₃NHCl–2.0AlCl₃–HCl were added into the mixture, respectively. By taking the total unprotonated indicator as the initial reference (ILs were not added to the solution), we could determine the [I]/[IH⁺] ratio from the measured absorbances of the organic layer after treating with acidic ionic liquids. Then the Hammett functions (H₀) of Et₃NHCl–2.0AlCl₃–HCl and Et₃NHCl–2.0AlCl₃ were calculated as −16.5 and −15.7 by the equation: H₀ = pK(I)_aq + log([I]_s/[IH⁺]_s), respectively. This result indicated the Brønsted superacid properties of both ionic liquids. The Brønsted acidic strength was increased after modifying the Et₃NHCl–2.0AlCl₃ by HCl, and this was beneficial to forming active carbocations by the protonation of TeMB.

Fig. 2 UV-vis absorption spectra of mesitylene in cyclohexane.

At the same time, various Al complex anions in the ionic liquid were identified by ²⁷Al NMR. On the basis of their chemical shifts, the two peaks in the ²⁷Al NMR spectrum of Et₃NHCl–2.0AlCl₃ can be assigned to AlCl₄⁻ (98 ppm) and Al₂Cl₇⁻ (103 ppm).²⁰Fig. 3 shows that Al₂Cl₇⁻ is the dominant Al species in Et₃NHCl–2.0AlCl₃. After modifying the Et₃NHCl–2.0AlCl₃ with HCl, the peak intensity at 103 ppm decreased and the peak intensity at 98 ppm increased, indicating that the AlCl₄⁻ species increased by interaction between HCl and aluminum-containing species:^19aHCl + Al₂Cl₇⁻ ⇋ H⁺ + 2AlCl₄⁻. In this process, a “new” superacid proton was generated, which increased the acidity of the ionic liquid.

Fig. 3 ²⁷Al NMR spectra of the ionic liquids.

In the transalkylation of TeMB and 2-MN, the protonated TeMB was the active species.¹⁶ When the Et₃NHCl–2.0AlCl₃ was modified by HCl, its acidity was enhanced and the concentration of the protonated TeMB increased, which was probably beneficial to the transalkylation.

Catalytic performance of the ILs

The effects of modifying Et₃NHCl–2.0AlCl₃ ILs and its dosage. The synthesis of 2,6-DMN by the transalkylation of 2-MN with TeMB was performed in the presence of ILs Et₃NHCl–2.0AlCl₃ and Et₃NHCl–2.0AlCl₃–HCl, and the results are listed in Table 1. Compared to the Et₃NHCl–2.0AlCl₃, the Et₃NHCl–2.0AlCl₃–HCl showed higher catalytic activity and selectivity for 2,6-DMN (Table 1, entries 1 and 2). The 2,6-DMN selectivity of 100% in the total DMN isomers and the 2-MN conversion of 15.6% were obtained within 3 h at 20 °C when using Et₃NHCl–2.0AlCl₃–HCl as the catalyst. At the same time, 1-MN was found in the reaction mixture, but the percentage of 1-MN in the total MN was very low. To our knowledge, this is the best result in the synthesis of 2,6-DMN by the alkylation of 2-MN under mild reaction conditions. In addition, the synthetic rate of 2,6-DMN was accelerated in the presence of Et₃NHCl–2.0AlCl₃–HCl due to the superacid proton given up in the Et₃NHCl–2.0AlCl₃–HCl ionic liquid¹¹ and the number of active carbocations from the protonated reaction of TeMB with Et₃NHCl–2.0AlCl₃–HCl increased. In the absence of TeMB, the reaction was run using 2-MN as the sole substrate. The result showed that no DMN was formed and 1-MN was the sole product arising from the isomerisation of 2-MN (Table 1, entry 3).

Table 1 The effects of the modifying Et₃NHCl–2.0AlCl₃ and its’ dosage on the transalkylation of 2-MN with TeMB^a

Entry	Dosage of IL mmol	Conversion %	Product selectivity %		DMN distribution %			Percentage of 1-MN in total MN %
			1-MN	DMN	2,6-	2,7-	2,3-
a Reaction conditions: 2-MN (7.6 mmol), TeMB (7.6 mmol), cyclohexane 10 mL, reaction time 3 h and reaction temperature 20 °C. b Et3NHCl–2.0AlCl3. c Et3NHCl–2.0AlCl3–HCl. d Without TeMB.
1^b	15	7.5	49.6	50.4	79.3	13.1	7.6	3.8
2^c	15	15.6	48.9	51.1	100	0	0	8.3
3^d	15	12.8	100	—	—	—	—	12.8
4^c	12	12.5	50.7	49.3	100	0	0	6.7
5^c	18	17.5	45.9	54.1	80.6	11.6	7.8	8.8

---

Based on the above results and previous literature,^11–16 a possible mechanism of the transalkylation reaction of 2-MN2 with TeMB in the Et₃NHCl–2.0AlCl₃–HCl catalyst was proposed. The anionic species Al₂Cl₇⁻ was assumed to react with the dissolved HCl, and the superacid properties of protons were generated by this process, which releases protons with extremely low solvation and high reactivity.²¹ The superacid protons in ILs are the active species and the carbocation (I) was formed by the reaction of TeMB and the superacid H⁺ (Scheme 1),^15,19 and then the transalkylation was carried out between the carbocation (I) and 2-MN. So, the superacid protons played important roles in the transalkylation of TeMB and 2-MN.

Scheme 1 The possible mechanism of the transalkylation reaction.

By varying the dosage of ionic liquid Et₃NHCl–2.0AlCl₃–HCl from 12 mmol to 18 mmol, the effects of the ionic liquid dosage on the transalkylation of 2-MN with TeMB were investigated and the results were given in Table 1. The 2-MN conversion increased from 12.5% to 17.5% when the loading of the catalyst was increased from 12 mmol to 18 mmol, and the 2,6-DMN selectivity in total DMN isomers was maintained at 100% when the dosage of ionic liquid was no more than 15 mmol. The 2,6-DMN selectivity decreased when the dosage of the ionic liquid was increased to 18 mmol, and the 2-MN conversion increased (Table 1, entries 2, 4 and 5), which was probably due to the fact that the concentration of the carbocation (I) increased with increasing loading of the ionic liquid, which may have led to acceleration of the reaction. However, the selectivity for 2,6-DMN decreased (Table 1, entries 2 and 5), because an excessively high concentration of carbocation (I) increased the chance of its attacking at different positions in the 2-MN molecule.¹¹ Therefore, the optimal dosage of the ionic liquid to ensure a high 2,6-DMN selectivity is 15 mmol. So, a suitable concentration of carbocation (I) was necessary for the selective synthesis of 2,6-DMN.

The effects of the reaction time and temperature. The effects of reaction time on the 2-MN conversion and the 2,6-DMN selectivity has been studied and the results are shown in Table 2 (entries 1–4). It can be seen that the isomerization of 2-MN was predominant during the initial stage. The selectivity to 1-MN was as high as 67.9% after 2.0 h. When the reaction time was prolonged, the selectivity to 1-MN decreased and the selectivity to DMN increased. It is notable that the selectivity for 2,6-DMN in total DMN isomers remains at 100% within 3.0 h. When the reaction time was increased to 3.1 h, the selectivity to 2,6-DMN decreased and the thermodynamically equal stable isomer 2,7-DMN was found when we used Agilent GC with a WCOT PLC column to analyze the 3.1 h reaction mixture. But when we used 2,6-DMN as the sole substrate under the same reaction conditions, only the product 1,6-DMN was found. We thought that thermodynamic factors on the distribution of DMN isomers becomes important after 3.0 h and the chance of the carbocation (I) attacking C-7 positions on 2-MN increased. In order to investigate the effect of reaction time on the product distribution in the transalkylation, the reaction time was prolonged to 24.0 h (Table 2, entry 4). The conversion of 2-MN reached 48.7% and the selectivity of 2,6-DMN in the total DMN decreased to 63.5%. At the same time, NAPH was formed and other DMN isomers, 1,6-DMN and 1,7-DMN, were found. So, the product distribution was affected by the reaction time. Therefore, in order to synthesise 2,6-DMN with high selectivity, the suitable reaction time was 3.0 h.

Table 2 The effects of the reaction conditions on the transalkylation of 2-MN with TeMB^a

Entry	Reaction temperature °C	Reaction time h	Conversion %	Product selectivity %		DMN distribution %				Percentage of 1-MN in total MN %
				1-MN	DMN	2,6-	2,7-	2,3-	Other
a Reaction conditions: 2-MN (7.6 mmol), TeMB (7.6 mmol) Et3NHCl–2.0AlCl3–HCl (15 mmol), cyclohexane 10 mL. b The catalyst was used for the second time. c The catalyst was used for the third time.
1	20	3.0	15.6	48.9	51.1	100	0	0	0	8.3
2	20	2.0	10.2	67.9	32.1	100	0	0	0	7.1
3	20	3.1	16.1	48.8	51.2	89.0	11.0	0	0	8.6
4	20	24.0	48.7	19.9	67.1	63.5	15.9	0	20.6	15.9
5	15	3.0	7.4	73.6	26.4	100	0	0	0	5.5
6	25	3.0	22.7	30.2	69.8	69.0	15.7	4.2	0	8.2
7^b	20	3.0	16.4	55.6	44.4	100	0	0	0	9.8
8^c	20	3.0	19.3	26.0	74.0	80.6	11.3	0	8.1	5.9

---

Reaction temperature has an important effect on the transalkylation when the ionic liquid Et₃NHCl–2.0AlCl₃–HCl was used as a catalyst, and the results are shown in Table 2. The 2-MN conversion and the DMN selectivity increased and the isomerization of 2-MN was inhibited with increasing reaction temperature (Table 2, entries 1, 5 and 6). The 2-MN conversion was obviously increased with increasing reaction temperature, the highest 2-MN conversion of 15.6% and 2,6-DMN selectivity of 100% are reached at 20 °C within 3.0 h. To the best of our knowledge, this is better than any result reported in the literature. When the reaction temperature was increased beyond 20 °C, the selectivity to 2,6-DMN decreased and more 2,7-DMN was obtained. This was because the activity of the carbocation (I) was most suitable at 20 °C; the carbocation was able to selectively attack C-6 positions on 2-MN due to its high electron density²² and the smaller steric hindrance, and a high selectivity of 2,6-DMN was achieved (Scheme 1). However, the activity of the carbocation (I) was too high at higher temperatures, thus weakening the effect of electron density distinction between C-6 positions on 2-MN and the other positions on 2-MN at high temperature.²² Therefore, 2,7-DMN and 2,3-DMN were generated and the selectivity of 2,6-DMN was decreased. We thought that, in this transalkylation, 2,6-DMN and 2,7-DMN were obtained in the presence of ionic liquid because the steric accessibility to C-6 and C-7 were better than for C-1 and C-3. Furthermore, the selectivity of 2,6-DMN was higher than 2,7-DMN, due to the higher electron density of C-6. Thus, the high selectivity of 2,6-DMN was caused by a combination of steric and the electronic effects.

The reusability of the catalyst

Under the optimal reaction conditions, the reusability of the ionic liquid catalyst was investigated under an atmosphere of N₂ with the use of Schlenk techniques, and the results are shown in Table 2. When the ionic liquid Et₃NHCl–2.0AlCl₃–HCl was reused for the second time, the conversion of 2-MN increased, and the selectivity of 2,6-DMN in the total DMN was 100% (Table 2, entry 7). When the catalyst was reused on the third time, the selectivity of 2,6-DMN decreased to 80.6% and isomers of DMN, such as 2,7-DMN, 1,6-DMN and 1,7-DMN, were found in the reaction mixture (Table 2, entry 8).

Conclusions

In summary, 2,6-DMN was synthesized with high selectivity by the transalkylation reaction of 2-MN with TeMB in the HCl-modified Et₃NHCl–2.0AlCl₃ ionic liquid. Under optimal reaction conditions (at 20 °C, 3.0 h), a 2,6-DMN selectivity of 100% of total DMN isomers and the 2-MN conversion of 15.6% were obtained, respectively. Results showed that this ionic liquid catalyst could be at least twice without a decrease in the selectivity for 2,6-DMN. It was noteworthy that these results may be the basis for technical research of the synthesis of 2,6-DMN through a one-step reaction.

Acknowledgements

This work is supported by the Chinese National Sciences Foundation (No. 21006021, 21076065) and the Natural Science Foundation of Heilongjiang Province of China (No. ZD200820-02).

Notes and references

1. K. Tanabe and W. F. Holderich, Appl. Catal., A, 1999, 181, 399.
2. (a) L. D. Lillwitz, Appl. Catal., A, 2001, 221, 337; (b) T. Chen, N. Y. Kang, C. W. Lee, H. Kim, S. B. Hong, H. D. Roh and Y. K. Park, Catal. Today, 2004, 371, 93.
3. (a) C. Zhang, X. W. Guo, C. S. Song, S. Q. Zhao and X. S. Wang, Catal. Today, 2010, 149, 196; (b) K. Bobuatong, M. Probst and J. Limtrakul, J. Phys. Chem. C, 2010, 114, 21611; (c) X. F. Bai, K. Y. Sun, W. Wu, P. F. Yan and J. Yang, J. Mol. Catal. A: Chem., 2009, 314, 81.
4. R. Millini, F. Frigerio, G. Bellussi, G. Pazzuconi, C. Perego, P. Pollesel and U. Romano, J. Catal., 2003, 217, 298.
5. (a) S. B. Pu and T. Inui, Appl. Catal., A, 1996, 146, 305; (b) J. N. Park, J. Wang and S. I. Hong, Appl. Catal., A, 2005, 292, 68.
6. G. P. Hagen and D. T. Hung, U. S. Patent 5 670 704, 1997.
7. (a) L. J. Jin, H. Q. Hu, S. W. Zhu and B. Ma, Catal. Today, 2010, 149, 207; (b) L. Zhao, H. B. Wang, M. Liu, X. W. Guo, X. S. Wang, C. S. Song and H. M. Liu, Chem. Eng. Sci., 2008, 63, 5298.
8. L. J. Jin, Y. M. Fang and H. Q. Hu, Catal. Commun., 2006, 7, 255.
9. K. Vahteristo, K. M. Sahala and S. Koskimies, Ind. Eng. Chem. Res., 2010, 49, 4018.
10. (a) W. Wu, W. G. Wu, O. V. Kikhtyanin, L. F. Li, A. V. Toktarev, A. B. Ayupov, J. F. Khabibulin, G. V. Echevsky and J. Huang, Appl. Catal., A, 2010, 375, 279; (b) C. Zhang, X. W. Guo, C. S. Song, S. Q. Zhao and X. S. Wang, Catal. Today, 2010, 149, 196; (c) W. Wu, L. F. Li, G. Wu, X. Wang, W. G. Wu, O. Kikhtyanin, A. Toktalev and G. Echevskii, Chin. J. Catal., 2009, 30, 531.
11. G. Pazzucconi, C. Perego, R. Millini and F. Frigerio, U. S. Patent 6 147 270, 2000.
12. D. B. Zhao, M. Wu, Y. Kou and E. Z. Min, Catal. Today, 2002, 74, 157.
13. J. A. Boon, J. A. Levinsky, J. I. Pflug and J. S. Wilkes, J. Org. Chem., 1986, 51, 480.
14. (a) Z. K. Zhao, Z. S. Li, G. R. Wang, W. H. Qiao and L. B. Cheng, Appl. Catal., A, 2004, 262, 69; (b) Z. K. Zhao, W. H. Qiao, G. R. Wang, Z. S. Li and L. B. Cheng, J. Mol. Catal. A: Chem., 2005, 231, 137; (c) Z. K. Zhao, B. Yuan, W. H. Qiao, Z. S. Li, G. R. Wang and L. B. Cheng, J. Mol. Catal. A: Chem., 2005, 235, 74; (d) Z. K. Zhao, W. H. Qiao, G. R. Wang, Z. S. Li and L. B. Cheng, Appl. Catal., A, 2005, 290, 133.
15. W. Wu, X. L. Wei and G. Wu, Chin. J. Catal., 2007, 28, 572.
16. W. Wu, G. Wu and M. L. Zhang, Appl. Catal., A, 2007, 326, 189.
17. C. J. Adams, M. J. Earle, G. Roberts and K. R. Seddon, Chem. Commun., 1998, 2097.
18. Y. L. Yang and Y. Kou, Chem. Commun., 2004, 226.
19. (a) G. P. Smith, A. S. Dworkin, R. M. Pagni and S. P. Zingg, J. Am. Chem. Soc., 1989, 111, 525; (b) D. King, R. Mantz and R. Osteryoung, J. Am. Chem. Soc., 1996, 118, 11933; (c) C. Thomazeau, H. Olivier-Bourbigou, L. Magna, S. Luts and B. Gilbert, J. Am. Chem. Soc., 2003, 125, 5264.
20. (a) J. S. Wilkes, J. S. Frye and G. F. Reynolds, Inorg. Chem., 1983, 22, 3870; (b) J. L. Gray and G. E. Maciel, J. Am. Chem. Soc., 1981, 103, 7147; (c) T. Welton, Chem. Rev., 1999, 99, 2071.
21. P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed., 2000, 39, 3772.
22. Y. M. Fang and H. Q. Hu, Catal. Commun., 2006, 7, 264.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c2cy00239f

---