Quaternary ammonium acetate: an efficient ionic liquid for the dissolution and regeneration of cellulose

Abstract

Cellulose (8%) can be dissolved in tetrabutylammonium acetate (TBAA) with dimethyl sulfoxide (DMSO) and crown ether (18-crown-6) within 5 min at 40 °C without any pretreatment.

---

Cellulose, as the most abundant of all natural substances, attracts much attention for its high performance, environmentally friendly characteristics, and having extensive applications. However, due to the inter- and intra-molecular hydrogen bonds, cellulose is difficult to dissolve in common solvents. Solvent systems¹ currently used for cellulose suffer drawbacks such as the generation of poisonous gas and high-energy cost.

Ionic liquids (ILs) are attracting increasing attention as a new class of solvent² because of its high thermostability, low viscosity, nonvolatile, and easy to recycle. Except for imidazolium ILs, quaternary ammonium ILs were reported to dissolve cellulose rapidly, such as tetrabutylammonium fluoride trihydrate (TBAF), tetraethylammonium chloride (TEAC), and 40% tetrabutyl-ammonium hydroxides (TBAH) with cosolvent.³ Hugo Cruz⁴ found that the greater basicity of anion resulted of the faster rate of dissolution of cellulose in imidazolium ILs. We envisioned that the rule could be applied to anions in quaternary ammonium ILs. The pK_a values of conjugate acids are respectively: −7 for HCl, −1.73 for H₂O, 3.18 for HF, 4.76 for CH₃COOH. However, the lower viscosity, and thus relatively higher ion diffusivity, in ethanoate ionic liquids cannot be discounted in our consideration.⁴ So we selected CH₃COO⁻ instead of F⁻ or OH⁻ as anion in tetrabutylammonium ILs and tried to dissolve cellulose in tetrabutylammonium acetate (TBAA) alone. But unfortunately, it did not work. However, for TBAF and TEAC, DMSO is an excellent cosolvent as one of the best swelling agents for the polymer.⁵ It is a polar, aprotic solvent that acts as both a soft base and a hard base.⁶ As we expected, tetrabutylammonium acetate (TBAA) was a powerful solvent for cellulose with DMSO. Then, we wanted to add LiCl into TBAA/DMSO to increase the solubility of cellulose. But different from the report about LiCl addition on the dissolution of cellulose in [BMIM]Cl,⁷ the existence of the alkali metal ions in TBAA hampered the dissolution of cellulose. And we found that the solubility of cellulose depended on the commercial source of TBAA because of the existence of K⁺, as the same as 40% TBAH reported by Tadashi Ema.⁸ The addition of crown ether (18C6) which can bind the K⁺ and be hydrogen-bonded with cellulose hydroxyl group, could promote the dissolution of cellulose in TBAA/DMSO. Different from the [AMIM]Cl,⁹ TBAA/DMSO/18C6 is a solvent for not only cellulose but also lignin. But in this communication, we mainly presented the details of the dissolution process of cellulose and the structure of the regenerated cellulose materials prepared from TBAA/DMSO/18C6.

Dissolving process of cellulose in TBAA/DMSO/18C6 at ambient temperature was real time monitored by Confocal Laser Scanning Microscope and was shown in Fig. 1. It was observed that cellulose with the degree of polymerization as high as 830 can be dissolved in TBAA/DMSO/18C6 within 25 min at room temperature without stirring (ESI†). However, remarkable swelling was not observed in the dissolution, which was the same as the [AMIM]Cl system.¹⁰ It was also observed, at initial stage the dissolution occurred very rapidly. Then dissolution rate decreased at last 20 min, which might be the result of more perfect crystalline structure in residual cellulose and the increased viscosity. With the increasing of cellulose, the dissolution time was extended. Despite this, in the present work, a solution containing down to 8% cellulose (powder, DP_w = 830) in TBAA/DMSO/18C6 was formed as a clear and viscous solution within 5 min under string at 40 °C and 12% within 30 min.‡ The solubility of microcrystalline cellulose (DP = 200) can reach to 20% in the new solvent as high as that in [BMIM]Cl. Cotton linter can also be dissolved in the new solvent as Fig. 2 showed.

Fig. 1 CLSM images of cellulose (powder, DP_w = 830) dissolved in TBAA/DMSO/18C6 at 25 °C at different time: 0; 2.5; 5; 10; 15; 25 min.

Fig. 2 The dissolution of cellulose powder (a), filter paper (b) and cotton (c) in TBAA/DMSO/18C6.

The cellulose solution showed typical viscosity–shear rate properties at the different concentration in Fig. 3. The viscosity of cellulose solutions exhibited shear-thinning behaviour and the increasing concentration of cellulose solution led to the rheological behaviour degraded due to the more chain entanglements. Clearly, the values of viscosity of 6% and 12% cellulose (powder, DP_w = 690) solution in the new solvent at 25 °C were about 140 and 1700 Pa s, which was tenth of the viscosity of cellulose in [AMIM]Cl.¹⁰ Due to its low viscosity and good formability in EtOH at room temperature, the regenerated cellulose fibres prepared from TBAA/DMSO/18C6 were easily produced under mild conditions by wet spinning.‡

Fig. 3 Viscosity–shear rate curves of cellulose solution of different concentration at 25 °C (DP_w = 690).

Fig. 4 and Table 1 showed the thermal decomposition of original cellulose, regenerated cellulose and the cellulose solution prepared from TBAA/DMSO/18C6. For regenerated cellulose prepared from [BMIM]Cl, the T_d was 70 °C lower than that for original cellulose because of the decrease in molecular weight of cellulose during dissolution.¹¹ However, the T_d values for regenerated cellulose and cellulose in TBAA/DMSO/18C6 were not significantly different to that of original cellulose. This might demonstrate that the DP_w of cellulose had no great decrease after cellulose dissolution and regeneration by TBAA/DMSO/18C6. The T_{d max} values for cellulose after dissolution and regeneration were 30 °C lower than that before, which indicated the thermal stability slightly decreased. It might result from the change of crystal of cellulose, as ¹³C NMR and XRD measured. It should be noticed that T_d at the first stage of TG scan of cellulose solution was about 70 °C because of the existence of DMSO. It meant that the dissolution of cellulose in TBAA/DMSO/18C6 should be prepared below 70 °C and the cellulose solution showed relatively poor thermal stability.

Fig. 4 TG scans of original cellulose, regenerated cellulose (b) and cellulose solution (10%, c) prepared from TBAA/DMSO/18C6.

Table 1 The pronounced decomposition onset (T_d), the maximum decomposition rate temperature (T_{d max}) of samples

Samples	Analysis of TG		Decomposed compound^a
	Td	Td max
a Cell I, Cell II and Amor represent cellulose I, cellulose II, and amorphous cellulose, respectively.
Original cellulose	310.8	362	Cell I and Amor
Regenerated cellulose	301.6	333	Amor and Cell II
Cellulose solution	314.0	339.6	Amor

---

Solid-state ¹³C-NMR spectra of cellulose exhibited strong singlets corresponding to ¹³C chemical shifts of cellulose carbons, C₁ (105 ppm), C₄ (79–92 ppm), C_2,3,5 (70–80 ppm), C₆ (60–69 ppm) of anhydroglucose units of cellulose.¹² CP/MAS ¹³C solid-state spectra of original cellulose and regenerated cellulose prepared from TBAA/DMSO/18C6 shown in Fig. 5 indicated that no derivatives occurred during the process of dissolution and regeneration. The NMR spectrum of cellulose solution also implied that no decomposition of the new solvent occurred during the dissolution process. The broad multiplicities of the C₄, C₆ carbons are assumed to be associated with the crystalline and amorphous components.¹³ C₄ downfield and upfield components are the contributions from the crystalline and amorphous for the cellulose samples, respectively. Moreover, the C₆ overlapping broad components are also the contributions from amorphous components of cellulose. The spectrum of original cellulose in Fig. 5 showed strong signals at 88.6 and 64.9 ppm and broad signals at 83.0 and 61.9 ppm, indicating the original cellulose contained both crystalline and amorphous fraction. In comparison with original cellulose, the NMR spectrum of regenerated cellulose showed almost total absence in signals at 88.6 and 64.9 ppm, suggesting a decrease in crystalline fractions and possible disruption of orderly hydrogen-bonding networks in cellulose dissolution and regeneration. It indicated that the regenerated cellulose prepared from TBAA/DMSO/18C6 and coagulated in EtOH was more that of amorphous cellulose than crystalline cellulose based on the strong amorphous signals at 83.5 and 62.2 ppm, which was further confirmed by XRD. This transformation from cellulose I to amorphous cellulose was strong evidence that complete cellulose (involving crystalline and amorphous region) dissolution occurred in a non-derivatizing solvent for cellulose.

Fig. 5 CP/MAS ¹³C solid-state spectra of original cellulose (a) and regenerated cellulose (b); ¹³C NMR spectrum of cellulose (3%, w/w) dissolved in DMSO-d₆/TBAA/18C6 (c).

The X-ray diffraction patterns of the original cellulose and the regenerated cellulose were shown in Fig. 6. Three diffraction peaks at 2θ of 15.3, 22.6 and 34.7 degrees corresponding to the typical diffraction peaks for cellulose I with the crystal planes, (101), (002) and (040) were observed in the XRD patterns of the original cellulose.¹⁴ There appeared one broad peak at 2θ = 20.3° in the XRD patterns of the regenerated cellulose indicating that cellulose II existed but was less than amorphous cellulose.¹⁵ The crystals of original and regenerated cellulose were shown in Table 1.

Fig. 6 X-ray spectra of the original cellulose (a) and the regenerated cellulose prepared from TBAA/DMSO/18C6 (b).

The SEM images of the surface, cross-section of the regenerated cellulose fibres prepared from cellulose dissolved in the TBAA/DMSO/18C6 were shown in Fig. 7. It was deserved that the novel fibres were homogeneous with smooth surfaces and circular cross-sections, whose diameter was about 40 μm.

Fig. 7 SEM images of novel cellulose fibre prepared from TBAA/DMSO/18C6: (a) side surface section; (b) magnified side surface section; (c) cross-section and the diameter of the novel fibre; (d) magnified cross-section.

In conclusion, the multi-solvent TBAA/DMSO/18C6 is a new efficient non-derivatizing solvent for the dissolution and regeneration of cellulose. It successfully solubilised cellulose at high concentrations rapidly and produced cellulose solution with low viscosity at room temperature. Moreover, the regenerated cellulose materials prepared by coagulation in EtOH exhibited good properties. On the basis of the fact that TBAA is non-volatile and easy-recycle, the process of dissolution and regeneration of cellulose seems to be a promising “green process” for the preparation of regenerated cellulose fibres. And it may take place of current industrial processes for viscose rayon and Lyocell because of green and low-energy cost.

Acknowledgements

This study was supported by the financial support of “948” Project of the State Forestry Administration (no. 2013-4-03).

Notes and references

1. (a) C. L. McCormick, P. A. Callais and B. H. Hutchinson Jr, Macromolecules, 1985, 18, 2394–2401; (b) H. Chanzy, A. Peguy, S. Chaunis and P. Monzie, J. Polym. Sci., Polym. Phys. Ed., 1980, 18, 1137–1144; (c) J. Zhou and L. Zhang, Polym. J., 2000, 32, 866–870; (d) M. Yanagisawa, I. Shibata and A. Isogai, Cellulose, 2004, 11, 169–176.
2. P. Wasserscheid and T. Welton, Ionic liquids in synthesis, Wiley Online Library, 2008.
3. (a) T. Heinze, R. Dicke, A. Koschella, A. H. Kull, E. A. Klohr and W. Koch, Macromol. Chem. Phys., 2000, 201, 627–631; (b) X. Changfa, Spectrosc. Spectral Anal., 1994, 4, 51–54; (c) M. Abe, Y. Fukaya and H. Ohno, Chem. Commun., 2012, 48, 1808–1810.
4. H. Cruz, M. Fanselow, J. D. Holbrey and K. R. Seddon, Chem. Commun., 2012, 48, 5620–5622.
5. B. Philipp, H. Schleicher and W. Wagenknecht, The influence of cellulose structure on the swelling of cellulose in organic liquids, 1973.
6. Å. Ostlund, D. Lundberg, L. Nordstierna, K. Holmberg and M. Nydén, Biomacromolecules, 2009, 10, 2401–2407.
7. A. Xu, J. Wang and H. Wang, Green Chem., 2010, 12, 268–275.
8. T. Ema, T. Komiyama, S. Sunami and T. Sakai, RSC Adv., 2014, 4, 2523–2525.
9. W.-Z. Li, M.-T. Ju, Y.-N. Wang, L. Liu and Y. Jiang, Carbohydr. Polym., 2013, 92, 228–235.
10. H. Zhang, J. Wu, J. Zhang and J. He, Macromolecules, 2005, 38, 8272–8277.
11. H. Z. Chen, N. Wang and L. Y. Liu, J. Chem. Technol. Biotechnol., 2012, 87, 1634–1640.
12. R. Dudley, C. Fyfe, P. Stephenson, Y. Deslandes, G. Hamer and R. Marchessault, J. Am. Chem. Soc., 1983, 105, 2469–2472.
13. F. Horii, A. Hirai and R. Kitamaru, Polym. Bull., 1982, 8, 163–170.
14. (a) Y. Guo, X. Wang, D. Li, H. Du, X. Wang and R. Sun, Polym. Bull., 2012, 69, 389–403; (b) Y. Guo, X. Wang, X. Shu, Z. Shen and R.-C. Sun, J. Agric. Food Chem., 2012, 60, 3900–3908.
15. Y. Guo, Q. Liu, H. Chen, X. Wang, Z. Shen, X. Shu and R. Sun, Cellulose, 2013, 20, 873–884.

---

Footnotes

† Electronic supplementary information (ESI) available: CLSM video of cellulose (powder, DPw = 830) dissolution in the TBAA/DMSO/18C6. See DOI: 10.1039/c4ra06258b

‡ (1) Dissolution of cellulose: cellulose powder (Futaba Chemical Co. Ltd; DP_w = 830, 690) was added to TBAA/DMSO/18C6 (weight: 2/7/1) under stirring at 40 °C. The clear cellulose solution (3.0, 6.0, 9.0, 12.0 wt%) was obtained within 5–30 min. (2) Preparation of cellulose fibres (regenerated cellulose): the cellulose solution was transferred to a glass syringe at room temperature and fibre spinning of the cellulose solution was carried out with a small and simple homemade wet spinning equipment. The length of the syringe needle was 35 mm and the inner diameter was 0.45 mm. The flow rate of the spinning dopes was controlled by pump and kept at 0.3 ml min⁻¹. The spinning dopes were spun into a coagulation bath of EtOH and the fresh fibres were washed in a second bath of EtOH. The novel fibres from syringe needle were passed through coagulation bath, washing bath and rolled on a roller with the rate at 100 rad min⁻¹ connected to a motor. The novel fibres were washed in running water for 30 min, and then were dried at room temperature.

---