Synthesis of pillar[n]arenes (n = 5 and 6) with deep eutectic solvent choline chloride 2FeCl₃

Abstract

A novel and distinct method of preparing pillar[5]arene and pillar[6]arene with high selectivity and efficiency has been achieved by condensation of 1,4-dialkoxybenzene and paraformaldehyde with the choline chloride (ChCl)/ferric chloride (FeCl₃) deep eutectic solvent in CH₂Cl₂ at room temperature. Under the optimal conditions, the yield of pillar[5]arene and pillar[6]arene is 35% and 53%, respectively. The reaction mechanism is investigated by room-temperature X-band Electron Spin Resonance (ESR), indicating that a free radical takes part in this cyclization reaction and acts as an intermediate. Our research is the first report about the application of DESs in supramolecular macrocyclic host synthesis.

---

Introduction

Deep eutectic solvents (DESs) are composed of mixing halide salts with metal salts or a hydrogen bond donor (HBD).¹ DESs can be easily prepared from low cost chemical materials, and through an atom economical approach. In 2001, Abbott² reported firstly that choline chloride and MCl₂ (M = Zn and/or Sn) could form a moisture-stable deep eutectic solvent. In recent decades, DESs have been used widely as Lewis acid catalysts and reaction medium for Diels–Alder reactions,³ O-acetylation of cellulose and monosaccharides,⁴ Kabachnik–Fields reaction⁵ and esterification reactions.⁶

Pillar[n]arenes are a new kind of supramolecular macrocyclic host which consist of 1,4-O-disubstituted hydroquinone subunits linked by methylene bridges at their 2- and 5-positions.⁷ Since they were first discovered by Ogoshi in 2008,^8a an increasing number of researches focused not only on their applications in supramolecular chemistry but also on the improvements of their synthesis methods.⁹ Generally, there are two pathways for the synthesis of pillar[n]arenes, known as the direct method and the indirect method. The direct method is the condensation of 1,4-dialkoxybenzene and paraformaldehyde catalyzed by Lewis acids BF₃·OEt₂ (ref. 8) and FeCl₃ (ref. 10) or organic acid.¹¹ While the indirect method is the cyclooligomerization of 1,4-dialkoxy-2,5-bis(alkoxy-methyl)benzenes in the presence of p-toluenesulfonic acid¹² or the cyclization of hydroxyl or bromo substituted 2,5-dialkoxybenzene¹³ with different kinds of Lewis acids. However, both of the methods have some defects. For example, the yield of pillar[6]arene is poor using the organic acid-catalysed method,^11a,c and the highest yield of pillar[6]arene is 45% by using the iron(III) chloride catalysed route.^10a Ogoshi has reported the preparation of methyl pillar[5]arene in yields of 71% (ref. 8b) and cyclohexyl pillar[6]arene in yields of 87% (ref. 8c) using BF₃·OEt₂ as catalyst with the direct method, but the reaction conditions is too harsh. When using the indirect method, the yield of pillar[n]arenes is considerable, but the reaction route is relatively intricate. Nowadays, with the rapid development of pillar[n]arenes in supramolecular chemistry, it is extremely desired to develop a new high-yield synthesis pathway for pillar[n]arenes using a new catalytic systems.

Herein, we initiate a new and distinct method of preparing pillar[5]arene and pillar[6]arene with high selectivity and efficiency by the condensation of 1,4-dialkoxybenzene and paraformaldehyde catalyzed by choline chloride (ChCl)/ferric chloride (FeCl₃) DES in CH₂Cl₂ at room temperature. Under the optimal conditions, the yield of pillar[5]arene and pillar[6]arene can be obtained with the yield of 35% and 53%, respectively. To the best of our knowledge, the research is the first report about the application of DESs in supramolecular macrocyclic host synthesis, which will stimulate the applications of DESs in different relative fields.

Results and discussion

Firstly, the cyclization of 1,4-diethoxybenzene (1a) with paraformaldehyde was chosen as a model reaction (Scheme 1).

Scheme 1 Condensation of 1,4-diethoxybenzene and paraformaldehyde with the deep eutectic solvent.

Then, a series of DESs including [ChCl][MCl]₂ (M = Fe, Zn and Sn) and [ChCl][MCl·H₂O]₂ (M = Mg, Al, Cr, Mn, Co, Ni and Cu) were selected to study their influence on the synthesis reaction of pillar[n]arenes. As shown in Table 1, it was found that the cyclization only occurred in the presence of [ChCl][FeCl₃]₂ DES (Table 1, entry 1). However, other DESs, such as [ChCl][ZnCl₂]₂ (Table 1, entry 2–10), had no effect on this reactions even if the reaction time was prolonged to 24 hours. In order to investigate the influence caused by solvents, 1,2-dichloroethane (DCE), CHCl₃ and chlorocyclohexane (Cl-CyC6) were selected as reaction mediums to carry out the model reaction. As a result, the yield of 1b in DCE was nearly the same as in CH₂Cl₂, but the yield of 1c was rather low (Table 1, entry 12). However, when CHCl₃ or Cl-CyC6 was utilized as solvent, it only gave poor yields of 1b and 1c (Table 1, entry 13 and 14). Interestingly, this result was totally different from others' works.^8c

Table 1 Preparation of 1b and 1c with different reaction conditions^a

Entry	Catalyst	Time	Solvent	Yield^b of 1b	Yield^b of 1c
a Reaction conditions: 1a (1 mmol), paraformaldehyde (3 mmol), and DESs (0.15 mmol) in CH2Cl2 at room temperature (25 °C).b Isolated yields.c The same reaction conditions as in ref. 10a.d The same reaction conditions as in ref. 8d.e DCE = 1,2-dichloroethane.
1	[ChCl][FeCl3]2	4 h	CH2Cl2	35	53
2	[ChCl][ZnCl2]2	4–24 h	CH2Cl2	0	0
3	[ChCl][SnCl2]2	4–24 h	CH2Cl2	0	0
4	[ChCl][MgCl2·6H2O]2	4–24 h	CH2Cl2	0	0
5	[ChCl][AlCl3·6H2O]2	4–24 h	CH2Cl2	0	0
6	[ChCl][CrCl3·6H2O]2	4–24 h	CH2Cl2	0	0
7	[ChCl][MnCl2·4H2O]2	4–24 h	CH2Cl2	0	0
8	[ChCl][CoCl2·6H2O]2	4–24 h	CH2Cl2	0	0
9	[ChCl][NiCl2·6H2O]2	4–24 h	CH2Cl2	0	0
10	[ChCl][CuCl2·2H2O]2	4–24 h	CH2Cl2	0	0
11	No catalyst	24 h	CH2Cl2	0	0
12	[ChCl][FeCl3]2	4 h	DCE^e	48	18
13	[ChCl][FeCl3]2	4 h	CHCl3	2	Trace
14	[ChCl][FeCl3]2	4 h	Cl-CyC6	3	Trace
15^c	FeCl3	2 h	CHCl3	30	34
16^d	BF3·OEt2	20 min	CHCl3	20	15

---

To validate the interesting results shown above, the reactions with FeCl₃ based DESs were further investigated. As shown in Table 2, the amount of [ChCl][FeCl₃]₂ significantly influences the yield of cyclic pentamers. The yield of cyclic pentamers 1b increased with the rasing amount of [ChCl][FeCl₃]₂ (Table 2, entry 1–6). From 5 mol% to 30 mol% of [ChCl][FeCl₃]₂, the yield of 1b increased from 22% to 40%. However, with the addition of [ChCl][FeCl₃]₂ (Table 2, entry 1–6), the yield of 1c increased in a different way from 1b. The yield of 1c reached the summit in the presence of 15 mol% [ChCl][FeCl₃]₂ (Table 2, entry 3). When the amount of [ChCl][FeCl₃]₂ further increased from 15 mol% to 30 mol% (Table 2, entry 4–6), the yield of pillar[6]arene declined, instead. Moreover, the influence caused by the formation of DESs was also investigated. When Lewis basic mixtures were formed with iron(III) chloride and choline chloride at the ratio of 1 : 3 or 1 : 2,¹⁴ neither [ChCl]₃[FeCl₃] (Table 2, entry 7) nor [ChCl]₂[FeCl₃] (Table 2, entry 8) could promote this reaction. Compared with [ChCl][FeCl₃]₂, the catalytic efficiency of [ChCl][FeCl₃]₃ (Table 2, entry 9) was almost the same. Furthermore, when choline chloride in the DES was replaced by acetylcholine chloride, the [AcetylCl][FeCl₃]₂ DES showed low inefficiency in synthesizing both of 1b and 1c (Table 2, entry 10). In summary, the best practical and the highest efficient method for the synthesis of pillar[n]arenes (n = 5 and 6) is in the presence of 15 mol% [ChCl][FeCl₃]₂ in CH₂Cl₂ at room temperature, giving the yield of 1b and 1c 35% and 53%, respectively. The yield of 1c is the highest yield among all the yields reported. It indicates that DES system should be a good choice in the synthesis of pillar[6]arene.

Table 2 Preparation of 1b and 1c with different FeCl₃ based DESs^a

Entry	Catalyst (mol%)	Yield^b of 1b	Yield^b of 1c
a Reaction conditions: 1a (1 mmol), paraformaldehyde (3 mmol), and DESs (0.05–0.30 mmol) in CH2Cl2 at room temperature (25 °C).b Isolated yields.
1	[ChCl][FeCl3]2 (5)	22	21
2	[ChCl][FeCl3]2 (10)	26	36
3	[ChCl][FeCl3]2 (15)	35	53
4	[ChCl][FeCl3]2 (20)	32	51
5	[ChCl][FeCl3]2 (25)	38	44
6	[ChCl][FeCl3]2 (30)	38	43
7	[ChCl]3[FeCl3] (15)	0	0
8	[ChCl]2[FeCl3] (15)	0	0
9	[ChCl][FeCl3]3 (15)	36	48
10	[AcetylCl][FeCl3]2 (15)	24	28

---

The alkoxy used in the model reaction was ethyoxyl, and a number of alkoxys with different chain length were also chosen to evaluate the scope of the reaction. The results in Table 3 showed that the cyclization with different alkoxys were all very satisfactory. Noticeably, the overwhelming majority product was 2b by the cyclization of 1,4-dimethoxybenzene (2a) and paraformaldehyde under the optimal conditions. However, the higher cyclooligomer 2c was so poor yielded that it could be only detected by mass spectrometry. Using 1, 4-dibutyloxybenzene (3a) as the starting material, the yield of cyclic pentamers 3b and cyclic hexamers 3c was 33% and 47%, respectively. The yield of them were both a little lower than that of 1b and 1c. As for other starting compounds (R = n-hexyl, n-octyl), it showed that the amount of the two major products reduced and the oligomers were detected by ¹H NMR. These results indicates that with the increase of chain length of alkoxy substituents, the rate of the cyclization of 1,4-dimethoxybenzene with paraformaldehyde decelerates, which is in accord with ref. 11a.

Table 3 Preparation of 2b–5b and 2c–5c with different alkoxy substituents R^a

Starting material	R	Yield^b of b	Yield^b of c
a Reaction conditions: 2a–4a (1 mmol), paraformaldehyde (3 mmol), and DESs (0.15 mmol) in CH2Cl2 at room temperature (25 °C).b Isolated yields.
2a	Methyl	2b 68	2c Trace
3a	n-Butyl	3b 33	3c 47
4a	n-Hexyl	4b 33	4c 39
5a	n-Octyl	5b 32	5c 31

---

Finally, the reaction mechanism was initially investigated by room-temperature X-band Electron Spin Resonance (ESR). The mixture of reactants 1,4-diethoxybenzene (1a) and paraformaldehyde showed no ESR signals. At the moment [ChCl][FeCl₃]₂ was added, the reaction started. One minute later, two strong ESR signals were observed. These signals represented that free radical and iron(III) species occurred in the reaction system. (Fig. 1, black line). The signal intensity of ESR in the reaction declined as the reaction processed (Fig. 1, red and blue line). After 60 minutes of reaction, the free radical signal disappeared and there was only Fe³⁺ signal left (Fig. 1, green line). To confirm the role that free radical played in this system, two control experiments were carried out with 1a, paraformaldehyde, [ChCl][FeCl₃]₂ and 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO), a free radical capturing reagent. In the first experiment, TEMPO was added at the beginning. As expected, no products could be obtained with the existence of TEMPO. In the second experiment, TEMPO was added to the system after an hour of reaction. As a result, poor yield of pillar[n]arenes were achieved, and the conversion of 1a was low, by contrast (Scheme 2). The ESR results and the experiments reveals that free radical takes part in this cyclization reaction and acts as an intermediate. Further studies will focus on the structure of the free radical, and the pathway of its formation and consumption.

Fig. 1 Room-temperature X-band ESR spectra: during the reaction proceeded between 1 min to 60 min.

Scheme 2 Mechanistic study on the reaction.

Conclusions

In conclusion, a novel method for the effective synthesis of pillar[n]arenes (n = 5 and 6) by condensation of 1,4-dialkoxybenzene and paraformaldehyde catalyzed by [ChCl][FeCl₃]₂ deep eutectic solvent under mild conditions have been achieved. This method makes it possible to synthesize pillar[6]arene easily and efficiently. The effect of different Lewis acids DESs and the amount of DESs on the cyclization reaction are investigated. The FeCl₃ based DESs stands out among all the catalysts reached. Furthermore, different alkoxy substituted pillar[n]arenes (n = 5 and 6) are obtained through this method, and the chain length of alkoxy substituents can obviously influence the reaction rate and yield. The reaction were carried out in accordance with the free radical mechanism. Indeed, the exploration towards the characteristics and utilization of pillar[n]arenes in supramolecular chemistry field would benefit from our findings.

Experimental

Materials

All reagents and solvents for syntheses were purchased from commercial sources and used without further purification. It is important to note that the solvents are commercially available without additional drying. 1, 4-dibutyloxybenzene (3a), 1, 4-dihexyloxybenzene (4a) and 1, 4-dioctyloxybenzene (5a) were synthesized according to the paper.¹⁵

Measurements

The ¹H and ¹³C NMR spectra were recorded on a Bruker 400 MHz NMR spectrometer at 298 K. The chemical shifts (δ) were given in part per million relative to internal tetramethylsilane (TMS, 0 ppm for ¹H), CDCl₃ (77.3 ppm for ¹³C). ESI-MS measurement was performed on Thermo Finnigan LCQ advantage at 298 K. All MALDI-TOF-MS spectra were recorded with an Axima TOF2 mass spectrometry. EPR spectra were recorded on a Bruker X-band A200 spectrometer. The solution sample was taken out into a small tube and then analyzed by EPR. EPR spectra was recorded at 298 K on EPR spectrometer operating at 9.420 GHz. Typical spectrometer parameters were: scan range, 3000 G; center field set, 3361 G; time constant, 163.84 ms; scan time, 30.00 s; modulation amplitude 2.0 G; modulation frequency 100 kHz; receiver gain 1.00*104; microwave power, 19.71 mW.

Preparation of deep eutectic solvent

A mixture of the ferric chloride (FeCl₃) and choline chloride in a molar ratio of 2 : 1 was heated to 100 °C with gentle stirring until a dark brown clear liquid formed.

General synthetic process of pillar[n]arenes (see ESI† for more details)

Representative synthetic process of pillar[n]arenes: 1b and 1c. To the solution of 1,4-diethoxybenzene (1a) (1.6620 g, 10 mmol) in dichloromethane (150 ml) was added paraformaldehyde (0.9000 g, 30 mmol). And then, [ChCl][FeCl₃]₂ (0.6970 g, 1.5 mmol) was added to the solution. After the mixture stirred at 25 °C for 4 h, the reaction was quenched by addition of water. The organic phase was separated and washed with saturated aqueous NaHCO₃, H₂O, and brine. The crude product was purified by column chromatograph to yield 1b (CH₂Cl₂/petroleum ether = 3 : 1). ¹H NMR (400 MHz, CDCl₃): δ 6.73 (s, 1H), 3.83 (q, J = 6.92 Hz, 2H), 3.77 (s, 1H), 1.27 (t, J = 6.9 Hz, 3H). ¹³C NMR (100 MHz, CDCl₃): δ 149.8, 128.4, 115.0, 63.7, 29.8, 15.1. HR ESI-MS calcd. for C₅₅H₇₀O₁₀Na [M + Na]⁺ 913.4867, found 913.4. 1c (CH₂Cl₂/petroleum ether = 100 : 1). ¹H NMR (400 MHz, CDCl₃): δ 6.69 (s, 1H), 3.88–3.74 (m, 3H), 1.29 (t, J = 5.9 Hz, 3H). ¹³C NMR (100 MHz, CDCl₃): δ 150.4, 127.8, 115.2, 64.0, 30.9, 15.2. HR ESI-MS calcd for C₆₆H₈₄O₁₂Na [M + Na]⁺ 1091.5860, found 1091.5.

Acknowledgements

This work was financially supported by the National Science Foundation of China (no. 20972120).

Notes and references

1. Q. Zhang, K. D. Vigier, S. Royer and F. Jerome, Chem. Soc. Rev., 2012, 41, 7108–7146.
2. A. P. Abbott, G. Capper, D. L. Davies, H. Munro, R. K. Rasheed and V. Tambyrajah, Chem. Commun., 2001, 2010–2011.
3. A. P. Abbott, G. Capper, D. L. Davies, R. K. Rasheed and V. Tambyrajah, Green Chem., 2002, 4, 24–26.
4. A. P. Abbott, T. J. Bell, S. Handa and B. Stoddart, Green Chem., 2005, 7, 705–707.
5. S. T. Disale, S. R. Kale, S. S. Kahandal, T. G. Srinivasan and R. V. Jayaram, Tetrahedron Lett., 2012, 53, 2277–2279.
6. Y. Yang, W. He, C. Jia, Y. Ma, X. Zhang and B. Feng, J. Mol. Catal. A: Chem., 2012, 357, 39–43.
7. (a) P. J. Cragg and K. Sharma, Chem. Soc. Rev., 2012, 41, 597–607; (b) H. Zhang and Y. Zhao, Chem.–Eur. J., 2013, 19, 16862–16879.
8. (a) T. Ogoshi, S. Kanai, S. Fujinami, T. Yamagishi and Y. Nakamoto, J. Am. Chem. Soc., 2008, 130, 5022–5023; (b) T. Ogoshi, T. Aoki, K. Kitajima, S. Fujinami, T. Yamagishi and Y. Nakamoto, J. Org. Chem., 2011, 76, 328–331; (c) T. Ogoshi, N. Ueshima, T. Akutsu, D. Yamafuji, T. Furuta, F. Sakakibara and T. i. Yamagishia, Chem. Commun., 2014, 50, 5774–5777; (d) X. B. Hu, Z. Chen, L. Chen, L. Zhang, J. L. Hou and Z. T. Li, Chem. Commun., 2012, 48, 10999–11001.
9. T. Ogoshi and T. Yamagishi, Chem. Commun., 2014, 50, 4776–4787.
10. (a) H. Tao, D. Cao, L. Liu, Y. Kou, L. Wang and H. Meier, Sci. China: Chem., 2012, 55, 223–228; (b) L. Liu, D. Cao, Y. Jin, H. Tao, Y. Kou and H. Meier, Org. Biomol. Chem., 2011, 9, 7007–7010.
11. (a) K. Wang, L. L. Tan, D. X. Chen, N. Song, G. Xi, S. X. A. Zhang, C. Li and Y. W. Yang, Org. Biomol. Chem., 2012, 10, 9405–9409; (b) T. Boinski and A. Szumna, Tetrahedron, 2012, 68, 9419–9422; (c) C. Han, F. Ma, Z. Zhang, B. Xia, Y. Yu and F. Huang, Org. Lett., 2010, 12, 4360–4363.
12. D. Cao, Y. Kou, J. Liang, Z. Chen, L. Wang and H. Meier, Angew. Chem., Int. Ed., 2009, 48, 9721–9723.
13. (a) Y. Ma, Z. Zhang, X. Ji, C. Han, J. He, Z. Abliz, W. Chen and F. Huang, Eur. J. Org. Chem., 2011, 5331–5335; (b) M. Holler, N. Allenbach, J. Sonet and J. F. Nierengarten, Chem. Commun., 2012, 48, 2576–2578.
14. A. P. Abbott, G. Capper, D. L. Davies and R. Rasheed, Inorg. Chem., 2004, 43, 3447–3452.
15. P. Paduraru, R. Popoff, R. Nair, R. Gries, G. Gries and E. Plettner, J. Comb. Chem., 2008, 10, 123–134.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, preparation of deep eutectic solvents, synthesis process of pillar[n]arenes and characterization data of all compounds. See DOI: 10.1039/c4ra15758c

---