Heteropoly acids as efficient acid catalysts in the one-step conversion of cellulose to sugar alcohols

Abstract

Cellulose and even spruce can be converted efficiently into valuable platform chemicals via combined hydrolysis and hydrogenation in the aqueous phase. Thereby, heteropoly acids together with supported ruthenium catalysts show not only high activity but also remarkable selectivity to sugar alcohols reaching up to 81% yield of C₄ to C₆sugar alcohols in only 7 h at 160 °C.

---

Diminishing reservoirs of fossil fuels and global warming foster the transition from a fossil fuel based society. Therein, lignocellulose presents a promising feedstock which is not in competition with the food chain, available in large amounts and could enable closed carbon cycles. Nevertheless, selective low-temperature transformations to platform chemicals and implementation of complete value chains in the scope of biorefinery concepts remain challenging.^1–4

Cellulose presents the most abundant biopolymer and with around 40% the major fraction of lignocellulose. Consequently, its efficient valorisation, e.g. via enzymatic or acid catalyzed hydrolysis to glucose, is highly attractive. Though, the former suffers from low space-time yields, while acid hydrolysis as in traditional wood saccharification results in by-product formation due to further degradation of sugars to furfural compounds and challenging acid recovery. In line, various recent investigations focused on the development of catalysts and reaction engineering to selectively convert cellulose to sugars or potential platform molecules as 5-hydroxymethylfurfural and levulinic acid.^5–9 With regard to future applications, however, complete value chains need to be explored and successful transition technologies will be essential to facilitate integration of renewable feedstocks in the mature technology scenery of today's refineries.

Hydrogenolysis, resulting in C–C and C–O cleavage by hydrogen, could present such a transition technology allowing direct transformation of biopolymers, including not only starch, but also cellulose and hemicellulose into established platform chemicals including sugar alcohols,^10,11glycols,¹² and even alkanes.¹³ Therein, the high stability of sugar alcohols allows to avoid further degradation reactions as in sugar hydrolysis. Surprisingly, only scarce studies on the combined hydrolysis–hydrogenation of cellulose in the aqueous phase exist. At 245 °C within 30 min, 30% yield of sorbitol (62% C₄–C₆sugar alcohols) and 85.5% conversion of cellulose could be reached over supported Ru catalysts.¹⁰ Fukuoka et al. demonstrated 31% yield of sugar alcohols over supported Pt catalysts within 24 h at 190 °C.¹¹ Zhang et al. presented Ni promoted tungsten carbides and tungsten promoted Ni catalysts as efficient systems with the former supported on three dimensional mesoporous carbon as the best system.¹² Interestingly, at 245 °C up to 72% yield of ethylene glycol at complete conversion were reported. Recently, Geboers et al. showed already the conversion of microcrystalline cellulose with heteropoly acids at 185 °C reaching promising results.¹¹ However, applications in the conversion of real systems remain rare and a flexible product spectrum at moderate reaction conditions would be desirable.

Our previous studies focused on hydrogenolysis of cellulose at 160 °C combining mineral acids and supported noble metals to develop a quantitative composition–activity relationship model.¹⁴ Therein, the type of noble metal had a strong impact on product selectivity, while acid strength and concentration mainly determined the conversion indicating a two-step process viahydrolysis of cellulose to yield glucose, which undergoes further hydrogenation and hydrogenolysis.⁴ Although up to 60% yield of sugar alcohols could be reached, alternative systems to substitute mineral acids and combine efficient carbon utilization with high selectivity to certain target molecules are highly desirable.

In this work, we present that among various acids, heteropoly acids (HPA) in combination with supported Ru catalysts (Ru/C) allow not only excellent conversion of cellulose with above 80% yield of C₄–C₆sugar alcohols and 91% carbon efficiency at only 160 °C, but may even be applied effectively in the transformation of spruce as real biomass feedstock.

In a typical experiment, an autoclave was loaded with α-cellulose, Ru/C, water and acid, flushed with H₂, pressurized to 50 bar, and heated to 160 °C for 7 h. After the reaction, remaining solid including the catalyst was recovered and the liquid phase analyzed viaHPLC. Detailed descriptions of experimental procedures can be found in the ESI.‡

In the reaction, glucose and xylose resulting from hydrolysis of cellulose and hemicellulose (10% in α-cellulose) are hydrogenated to sorbitol and xylitol, respectively. Additionally, further dehydration and hydrogenolysis may yield sorbitan and isosorbide as well as erythritol, glycerol, propylene or ethylene glycol and methanol (Scheme 1 ESI‡). Obviously, various acids can be combined with Ru/C to reach reasonable conversion of cellulose, the observed product distributions, however, are rather different (Table 1). Using mineral acids, conversions of cellulose above 60% can be reached dependent on acid strength and concentration (Table 1, entries 1–3). The main products are C₅ and C₆sugar alcohols with selectivities up to 83%, while no sugars can be detected emphasizing high hydrogenation activity of the ruthenium based catalyst in the presence of mineral acids. In contrast, with p-toluenesulfonic acid (p-TSA) complete conversion of cellulose is observed, but significant amounts of glucose and xylose remain, probably due to deactivation of Ru/C in the presence of p-TSA (Table 1, entry 4).

Table 1 Properties and catalytic activity of several acidic catalysts combined with Ru/C in the hydrogenolysis of cellulose^a

Entry	Catalyst	Acid conc./mmol l^−1	Acidity, pKa	Conversion^b (%)	Yield^c^,^g, C4–C6 (%)	Yield^c (%)
						C6 (glucose)	C5 (xylose)	C4	C3	C2	C1
a Reaction conditions: cellulose (500 mg), water (10 ml), Ru/C (100 mg), solid acid (100 mg), 433 K, 50 bar H2 (298 K), 7 h. b See ESI‡ c Based on theoretical stoichiometric coefficients (ESI‡). d 3 h. e PW shows a Hammett acidity of H0 = −13.6 corresponding to a superacid (ref. 15a). HPAs exhibit acidities greater than p-TSA at the same concentration (ref. 15b). f Equivalent amount of acid sites based on ion exchange capacity with 2.55 mmol g^−1 for Amberlyst70 and 4.8 mmol g^−1 for Dowex50. g Without sugars.
1	HCl ^d	686	−7	98	25.1	15.9 (0.0)	9.2 (0.0)	0.0	1.7	0.3	0.5
2	H2SO4	255	−3.9	72	59.9	48.6 (0.0)	11.3 (0.0)	0.0	2.2	0.0	1.7
3	H3PO4	255	2.16	59	33.3	25.0 (0.0)	7.3 (0.0)	1.0	2.2	0.0	0.0
4	p-TSA	55.1	0.7	100	20.8	33.8 (24.1)	4.5 (0.6)	7.2	1.0	0.0	0.0
5	H4[Si(W3O10)4]	55.1	Strong^e	98.8	80.6	53.0 (0.1)	8.9 (0.0)	18.7	3.9	0.6	4.4
6	H3[P(W3O10)4]	55.1	Strong^e	93.8	66.4	57.6 (0.0)	8.8 (0.0)	0.0	2.7	0.0	1.8
8	Amberlyst 70	25.5^f	—	42.6	1.4	0.9 (0.1)	0.3 (0.0)	0.3	1.0	0.4	4.6
9	Dowex	48.0^f	—	61.0	8.4	5.3 (0.1)	1.5 (0.0)	1.7	3.7	0.8	0.0

---

Surprisingly, HPAs in the form of phosphotungstic (PW) and silicotungstic (SiW) acid in combination with Ru/C allow not only high conversion, but also high selectivity to C₄–C₆sugar alcohols (Table 1, entries 5 and 6). Overall, up to 80.6% yield of sugar alcohols at carbon efficiencies above 90% can be reached.¹⁶ Therein, the required acid concentrations are low compared to mineral acids and even 3.5 × 10⁻³ mol l⁻¹PW were sufficient to yield 44% of C₄–C₆sugar alcohols (Fig. 1). Interestingly, both acids are reduced under reaction conditions as indicated by a bluish color of the fresh reaction solution, but were re-oxidized at room temperature upon contact with air along with decoloration.¹⁷ No color change appeared in the absence of H₂. Recycling of the HPAs, titration and ³¹P NMR measurements of the reaction solution before and after reaction did not show any noticeable change. PW and SiW are known to exhibit higher acid strength and superior thermal and hydrolytic stability compared to other HPAs as H₄[Si(Mo₃O₁₀)₄] (SiMo) and H₃[PMo₁₀V₂O₄₀] (VMo).¹⁸ These properties may cause the high catalytic activity of PW and SiW. For PMo and VP, up to 45% conversion, but below 5% yield of products in the liquid phase could be reached (ESI‡). Nevertheless, based on these results irreversible structural changes of the HPAs under reaction conditions cannot be excluded. Future investigations will focus on this issue, also regarding water insoluble HPAs as solid acids.

Fig. 1 Conversion and yield in the hydrogenolysis of cellulose combining the heteropoly acids tungstosilicic acid (left) and phosphotungstic acid (right) in various concentrations with ruthenium supported on activated carbon (0.5 g α-cellulose, 0.1 g Ru/C, 10 ml water, 160 °C, 50 bar H₂ (25 °C)).

Previous studies indicate the rate of cellulose hydrolysis to be strongly dependent on acid concentration.¹⁹ In line, our results emphasize an increasing conversion of cellulose at higher H₃O⁺ concentrations and longer reaction times for PW and SiW, respectively. Additionally, further dehydration of sorbitol to sorbitan and isosorbide together with formation of smaller polyols appear. According to reaction mechanisms discussed in the literature, C–C and C–O cleavage may occur via retro-aldol and dehydration reactions.²⁰ Thereby, dehydrogenation to a β-hydroxyl carbonyl occurs, followed by retro-aldol condensation or dehydration and final rehydrogenation. Possible products include C₁–C₆polyols, whereby the extent of C–C cleavage and remaining hydroxyl groups appear to depend on the balance of metal and acid catalyst.¹³

Concerning product selectivity, Zhang et al. discussed the effect of W for supported tungsten carbide and bimetallic (W + Me: Ni, Ru, Pd, Pt, Ir) catalysts.¹² The authors observed high conversion and a shift in selectivity to ethylene glycol (EG) in the presence of W. Experiments emphasized a shift of selectivity to hexitol and erythritol for high Me/W ratios. A cooperative effect between Me and W was proposed. We suppose as well an influence of tungsten. Experiments combining H₂SO₄ with Ru/C simply adding W—as metal or as bimetallic catalyst—showed as well improved selectivites to C₄–C₆sugar alcohols (ESI‡), while the absence of EG may be attributed to high metal/W ratios and a low reaction temperature. A proper balance between the amount and strength of acid sites, and the activity of the hydrogenation catalyst appears to be essential. Additionally, the presence of W facilitates to reach high selectivity to certain products which may be attributed to cooperative effect between tungsten, metal sites and the substrate. In line, differences in product selectivity comparing SiW and PW could origin from differences in their interaction with the substrate. Clearly, further investigations are necessary to elucidate the underlying principles.

With regard to an industrial application, separation becomes an important issue. Compared to mineral acids, the lower concentration of HPA would result in reduced salt formation upon neutralization. Besides, HPAs may be precipitated via ion exchange with larger cations, e.g.K⁺, Cs⁺ and NH₄⁺, or in certain cases even extracted to allow direct recycle. Nevertheless, to facilitate catalyst separation, solid catalysts would be advantageous. Ion exchange resins such as Amberlyst70 and Dowex50 exhibit sulfonic groups on their external surface resembling to some extent p-TSA. Interestingly, they allow reasonable conversion of cellulose, but suffer from degradation under reaction conditions accompanied by leaching of acid sites. Hence, little selectivity to sugar alcohols could be reached, but levulinic acid proved to be a major product (Table 1, entry 8, 9). Likewise, HPAs were supported on silica following ref. 21 and zinc dodecatungstophosphate (ZnPW) as insoluble HPA was prepared following ref. 22. Therein, supported PW with 40% HPA loading gave conversions up to 45%, but exhibited pronounced leaching of active species, while ZnPW showed little activity and only 23.8% conversion. Nevertheless, optimization of solid acids improving catalyst–support interactions may overcome these obstacles.

The concept of hydrogenolysis of cellulose combining HPAs and Ru/C is not only applicable to pure cellulose but could be transferred to spruce as wooden feedstock (Table 2). Assuming the cellulose and hemicellulose fraction of spruce to be about 45 and 30%, respectively, SiW allowed complete conversion of these fractions. Again, C₄–C₆sugar alcohols present the main products with up to 65% yield in only 5 h reaction time at 160 °C. Besides, small amounts of C₁–C₃polyols including glycerol, PG, EG and methanol were formed (ESI‡).

Table 2 Properties and catalytic activity of HPA in combination with Ru/C in the hydrogenolysis of spruce^a

Entry	Catalyst	Conversion^b^,^c (%)	Yield C4–C6^c^,^d (%)	Yield C1–C3^c^,^d (%)
a 500 mg spruce chips without pretreatment: 10 ml water, 100 mg 5%Ru/C, 500 mg HPA, 433 K, 50 bar H2 (298 K), 5 h. b See ESI‡ c Conversion and yield were calculated assuming 45% cellulose and 30% hemicellulose, calculated as 75% cellulose (ESI‡). d Based on theoretical stoichiometric coefficients (ESI‡).
10	H4[Si(W3O10)4]	100	64.9	7.7
11	H3[P(W3O10)4]	87	42.0	5.2

---

In conclusion, we have shown that heteropoly acids combined with supported Ru catalysts allow direct transformation of cellulose into sugar alcohols with a yield of 81% and above 90% carbon efficiency. Additionally, this approach may even be applied to wooden feedstocks and yields of sugar alcohols of 65% could be reached starting from spruce.

We would like to thank Robert-Bosch Foundation for financial support. This work was performed as part of the Cluster of Excellence “Tailor-Made Fuels from Biomass”, funded by the Excellence Initiative by the German federal and state governments to promote science and research at German universities. We thank Mr A. Deege and Ms H. Hinrichs for HPLC analyses.

Notes and references

1. (a) G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106, 4044; (b) G. W. Huber and A. Corma, Angew. Chem., Int. Ed., 2007, 46, 7184–7201; (c) A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411.
2. (a) G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic, Science, 2005, 308, 1446–1450; (b) F. M. A. Geilen, B. Engendahl, A. Harwardt, W. Marquardt, J. Klankermayer and W. Leitner, Angew. Chem., Int. Ed., 2010, 49(32), 5510–5514.
3. (a) J.-P. Lange, R. Price, P. M. Ayoub, J. Louis, L. Petrus, L. Clarke and H. Gosselink, Angew. Chem., Int. Ed., 2010, 49(26), 4479–4483; (b) J. Q. Bond, D. M. Alonso, D. Wang, R. M. West and J. A. Dumesic, Science, 2010, 327, 1110–1114; (c) R. Palkovits, Angew. Chem., Int. Ed., 2010, 49(26), 4336–4338.
4. R. Rinaldi and F. Schüth, ChemSusChem, 2009, 2, 1096–1107.
5. (a) C. Li, Q. Wang and Z. K. Zhao, Green Chem., 2008, 10, 177–182; (b) C. Li and Z. K. Zhao, Adv. Synth. Catal., 2007, 349(11–12), 1847–1850; (c) L. Vanoye, M. Fanselow, J. D. Holbrey, M. P. Atkins and K. Seddon, Green Chem., 2009, 11, 390–396; (d) J. B. Binder and R. T. Raines, J. Am. Chem. Soc., 2009, 131, 1979.
6. (a) D. Yamaguchi, M. Kitano, S. Suganuma, K. Nakajima, H. Kato and M. Hara, J. Phys. Chem. C, 2009, 113, 3181–3188; (b) A. Onda, T. Ochi and K. Yanagisawa, Green Chem., 2008, 10, 1033–1037.
7. (a) R. Rinaldi, R. Palkovits and F. Schüth, Angew. Chem., Int. Ed., 2008, 47, 8047–8050; (b) R. Rinaldi, J. N. Maine, J. von Stein, R. Palkovits and F. Schüth, ChemSusChem, 2010, 3(2), 266–276.
8. (a) S. Suganuma, K. Nakajima, M. Kitano, D. Yamaguchi, H. Kato, S. Hayashi and M. Hara, J. Am. Chem. Soc., 2008, 130, 12787–12793; (b) M. Kitano, D. Yamaguchi, S. Suganuma, K. Nakajima, H. Kato, S. Hayashi and M. Hara, Langmuir, 2009, 25(9), 5068–5075.
9. W. Deng, X. Tan, W. Fang, Q. Zhang and Y. Wang, Catal. Lett., 2009, 133, 167–174.
10. C. Luo, S. Wang and H. Liu, Angew. Chem., Int. Ed., 2007, 46, 7636–7639.
11. (a) A. Fukuoka and P. L. Dhepe, Angew. Chem., Int. Ed., 2006, 45, 5161–5163; (b) J. Geboers, S. Van de Vyver, K. Carpentier, K. de Blochouse, P. Jacobs and B. Sels, Chem. Commun., 2010, 46, 3577–3579.
12. (a) N. Ji, T. Zhang, M. Zheng, A. Wang, H. Wang, X. Wang and J. G. Chen, Angew. Chem., Int. Ed., 2008, 47, 8510–8513; (b) M.-Y. Zheng, A.-Q. Wang, N. Ji, J.-F. Pang, X.-D. Wang and T. Zhang, ChemSusChem, 2010, 3(1), 63–66; (c) Y. Zhang, A. Wang and T. Zhang, Chem. Commun., 2010, 46, 862–864.
13. G. W. Huber, R. D. Cortright and J. A. Dumesic, Angew. Chem., Int. Ed., 2004, 43, 1549–1551.
14. R. Palkovits, K. Tajvidi, R. Rinaldi, J. Procelewska and A. Ruppert, Green Chem., 2010, 12(6), 972–978.
15. (a) T. Okuhara and M. Misono, Oxide catalysts in solid state chemistry, Encyclopedia of Inorganic chemistry, ed. R. B. King, John Wiley & Sons, 1994, ISBN 0471 93620 0; (b) Y. Izumi, K. Matsuo and K. Urabe, J. Mol. Catal., 1983, 18, 299–314.
16. Based on conversion X_Cellulose and yield of liquid phase products Y_L, carbon efficiency E_C = Y_LX⁻¹_Cellulose.
17. D. G. Barton, S. L. Soled and E. Iglesia, Top. Catal., 1998, 6, 87.
18. I. V. Kozhevnikov, Chem. Rev., 1998, 98, 171–198.
19. (a) A. Sharples, Trans. Faraday Soc., 1957, 53, 1003–1013; (b) S. Yildiz, E. D. Gezer and U. C. Yildiz, Build. Environ., 2006, 41, 1762–1766.
20. K. Wang, M. C. Hawley and T. D. Furney, Chem. Eng. Sci., 2003, 58, 4271–4285.
21. M. Arias, D. Laurenti, V. Bellière, C. Geantet, M. Vrinat and Y. Yoshimura, Appl. Catal., A, 2008, 348, 142–147.
22. J. Li, X. Wang, W. Zhu and F. Cao, ChemSusChem, 2009, 2, 177.

---

Footnotes

† This article is part of the ‘Emerging Investigators’ themed issue for ChemComm.

‡ Electronic supplementary information (ESI) available: Experimental procedures and additional tests. See DOI: 10.1039/c0cc02263b

---