Adding value to renewables: a one pot process combining microbial cells and hydrogen transfer catalysis to utilise waste glycerol from biodiesel production

Abstract

Waste glycerol was converted to secondary amines in a one pot reaction, using Clostridium butyricum and catalytic hydrogen transfer-mediated amination.

---

New strategies to utilise renewable feedstocks are important to the future competitiveness of the chemical industry.¹ One versatile approach is to use a microbial process to digest the feedstock and form fermentation products, which can then be used as renewable platform chemicals. However, the fermentation products are polar and produced in dilute aqueous solutions containing mixtures of other metabolic products and cell material. This poses a difficult separation problem.^2,3 It would be desirable, therefore, to avoid the need for expensive and wasteful intermediate isolation, by developing chemistry that can be integrated directly with the microbial fermentation.^1,4

In this paper, we describe the first one pot microbial and chemocatalytic reaction to transform a renewable feedstock into higher value chemical intermediates. Adding value to glycerol-containing wastes is an important target, since glycerol is now being produced in significant quantities as a side product from biodiesel manufacturing.^1,5Fermentation of glycerol to form 1,3-propanediol is attractive since 1,3-propanediol is a valuable precursor for manufacturing high value polymers and platform chemicals.^2,6–10 However, the isolation of 1,3-propanediol from fermentation media is particularly challenging.^2,11 Therefore, integrating chemical catalysis with the 1,3-propanediol fermentation represents an especially attractive target to exemplify our approach.

We describe the conversion of waste glycerol to 1,3-propanediol coupled with amination mediated by hydrogen transfer catalysis (Fig. 1). Hydrogen transfer catalysed by a transition metal complex enables low temperature oxidation of alcohols to reactive aldehydes. Furthermore, water tolerant catalysts are available that do not require aggressive oxidants.^12,13 The aldehyde intermediate is then intercepted by reaction with an amine. This is particularly attractive because the resulting imine can serve as the electron acceptor for alcohol oxidation and yield alkylated amine products.¹⁴ This avoids side reactions such as aldol condensations, and ensures high atom efficiency. Most importantly, amination provides access to valuable secondary amines, important intermediates in manufacturing pharmaceuticals and other bioactives.¹⁵

Fig. 1 Amination of an alcohol by hydrogen transfer.

Clostridium butyricum is a suitable biocatalyst for fermentation of glycerol to 1,3-propanediol as it produces a relatively clean mixture of fermentation products compared with the enteric, glycerol-fermenting organisms.¹⁶ Furthermore, C. butyricum is an obligate anaerobe , so the stream of fermentation products is oxygen-free. This enables the use of potentially air-sensitive catalysts and reagents in downstream chemical reactions. Not all strains of C. butyricum can ferment waste glycerol from biodiesel manufacturing,^6–9 so the ability of C. butyricum DSM10703 to ferment pure and crude glycerol was evaluated (Table 1). There was little difference in growth rate, biomass production, substrate consumption or 1,3-propanediol formation with either substrate, thus confirming that this strain of C. butyricum is suitable for conversion of crude glycerol to 1,3-propanediol.

Table 1 Fermentation of pure and crude glycerol using Clostridium butyricum¹⁷

	Substrate
	Pure glycerol	Crude glycerol
Growth rate/h^−1	0.37	0.4
Final biomass concentration (OD660nm)	5.37	5.85
Glycerol consumed/mM	204	207
Final 1,3-propanediol concentration/mM	127	134

---

A range of hydrogen transfer catalysts were screened for the amination of 1,3-propanediol by aniline in toluene (Table 2, Fig. 2).

Fig. 2 Amination of 1,3-propanediol with aniline.

Table 2 Amination of 1,3-propanediol with aniline^a

			Proportion of products^b (%)
Catalyst	Modifier	Conv.^c (%)	1	2	3
a Reactions conducted in a sealed tube, aniline (1.0 mmol), 1,3-propanediol (0.5 mmol), toluene (0.5 mL), K2CO3 (10 mol%), catalyst (1 mol% of –OH) at 115 °C, N2, 24 h. b By ^1H NMR. c Based on 1,3-propanediol. d Double scale reaction. e Catalyst (5 mol%). ND = none detected. dppf = bis-diphenyl phosphino ferrocene; dppm = bis-diphenyl phosphino methane; dppp = bis-diphenyl phosphine propane.
[Cp*IrCl(μ-Cl)]2^de	—	84	48	35	17
[Ru(p-cymene)Cl2]2	dppf	59	93	7	ND
[Ru(p-cymene)Cl2]2	dppm	72	86	4	10
[Ru(p-cymene)Cl2]2	dppp	79	81	14	5
a	—	48	84	16	ND
b	—	>99	ND	65	35

---

The most efficient catalyst screened was b¹⁹ (Fig. 3), previously introduced for the dynamic kinetic resolution of alcohols to chiral esters. The action of b yielded the diaminated product 2 and, unexpectedly, N-propyl aniline 3, the result of monoamination (to 1), dehydration and reduction. No reaction was observed in the absence of catalyst. We are currently investigating the reaction mechanism of N-propyl aniline formation.

Fig. 3 NHC hydrogen transfer catalystsa¹³ and b.¹⁹

The next step was to test the use of b for amination of 1,3-propanediol produced by fermentation. The fermentation broth was produced by growing C. butyricum on pure glycerol (204 mM) until growth and fermentation had stopped.¹⁷ The broth was collected, centrifuged to remove the cells, and the supernatant was collected by decanting it, leaving the pelleted cells behind. The aqueous culture supernatant was analysed by HPLC, and contained 1,3-propanediol (127 mM), acetate (10.8 mM), butyrate (10.7 mM), and lactate (3.7 mM). It also contained residual culture medium and other microbial products (not identified, but assumed to be secreted proteins and cell lysis products). The cells had consumed all of the glycerol and the final pH was 6.70.

This culture supernatant was used directly in the catalytic transformation without further purification. b is insoluble in water and the chemocatalytic reactions were conducted in aqueous fermentation broth/organic biphasic media. Initially, aniline, base and catalystb were dissolved in toluene and added directly to the aqueous culture supernatant. The volume of culture supernatant was adjusted so that it provided 0.5 mmol of the 1,3-propanediol produced during the fermentation (Table 3). The biphasic mixture was incubated at 115 °C for 24 h, and then analysed by HPLC and NMR. The reaction was less efficient than in the absence of water (Table 2), but, nevertheless, 20% consumption of 1,3-propanediol was observed. 1 was formed as the major product, with 2 and 3 as minor products. The rate of amination in the biphasic system (Table 3) was lower than in pure toluene (Table 2). This is not surprising as the biphasic reaction is dependent on partitioning of 1,3-propanediol into the organic phase. 1,3-Propanediol is hydrophilic and extraction into organic solvents is, therefore, inefficient.^2,11 The concentration of the diol in the toluene phase was, therefore, much lower in the biphasic system than in the non-aqueous system. Despite this substrate limitation, the conversion and product yields are reasonable. Therefore, we propose that the catalytic reaction shifted the equilibrium for partitioning 1,3-propanediol into the solvent phase by forming the water-insoluble, N-alkylated products. In effect, this drove a reactive separation.

Table 3 Biphasic amination of 1,3-propanediol with aniline in the presence of fermentation broth^a

					Proportion of products^d (%)
Glycerol source for fermn^b	Solvent	T/°C	Time/h	Conv.^c (%)	1	2	3
a Reactions conducted in a sealed tube, aniline (1.0 mmol), solvent (0.5 mL), K2CO3 (10 mol%), catalystb (1 mol% of –OH), N2 and using aqueous culture supernatants as the source of 1,3-propanediol (0.5 mmol). The culture supernatants were produced by fermentation^17 of pure glycerol or crude glycerol from biodiesel manufacturing. b 3.96 mL or 3.74 mL were used, respectively. c Based on 1,3-propanediol. d By ^1H NMR of products extracted into hexane. ND, none detected.
Pure	Toluene	115	24	20	68	20	12
Crude	Toluene	115	24	15	82	18	ND
Pure	N1,8,8,8NTf2	115	24	11	ND	24	76
Pure	N1,8,8,8NTf2	60	48	7	ND	ND	100
Pure	N1,8,8,8NTf2	42	48	2	ND	ND	100
Crude	N1,8,8,8NTf2	60	48	12	ND	ND	100
Crude	N1,8,8,8NTf2	42	48	10	ND	ND	100

---

The amination of 1,3-propanediol produced by C. butyricum grown on crude glycerol from biodiesel manufacturing (207 mM¹⁷) was also demonstrated. The culture supernatant was prepared as before and contained 1,3-propanediol (134 mM), acetate (25 mM) and butyrate (19.5 mM) at pH 6.58. The glycerol had been consumed completely and lactic acid was not produced. The supernatant was mixed with catalyst, base and aniline in toluene. After incubation at 115 °C, 15% of the 1,3-propanediol was converted to 1 as the major product and 2 as the minor product. N-propyl aniline (3) was not detected. The lower conversion and change in the products formed might be due to the presence of chemical contaminants in the crude glycerol used in the fermentation, since it contained unidentified, non-glycerol organic materials (1.72%) and sulfated ash (6.72%), in addition to the glycerol (91.5%). Alternatively, the acetate and butyrate concentrations were higher after growth on crude glycerol compared with cultures grown on pure glycerol, and the increased acid concentration may have affected the hydrogen transfer, which is dependent on base.

In conclusion, the biphasic reaction could be operated using aqueous 1,3-propanediol produced by fermentation of either pure or crude glycerol after removal of the cells but without further purification of the fermentation broth. The product distribution depended on the purity of the glycerol used as the initial fermentation substrate.

It would be desirable to operate a fully integrated bio-chemocatalytic process. Toluene is a biocidal solvent,²⁰ and is unsuitable for use in an integrated one pot cascade. Toluene was substituted with the ionic liquid, methyl trioctylammonium bistriflamide (N_1,8,8,8NTf₂), since the latter solvent has been used in other biphasic microbial processes.^21–25 When the amination was performed in N_1,8,8,8NTf₂ at 115 °C using the supernatant from fermentation of pure glycerol, 11% conversion in 24 h afforded products 2 and 3. Thus, the selectivity was shifted towards dehydration and reduction of the monoaminated product 1.

Since the biocatalyst operates only at temperatures <42 °C, the catalytic reaction was tested at lower temperatures. When the 1,3-propanediol had been produced by fermentation of pure glycerol, only 7 and 2% of the substrate was converted after reacting for 48 h at 60 and 42 °C, respectively. However, the selectivity was shifted entirely towards N-propyl aniline formation. The conversion was improved when the culture supernatant from fermentation of crude glycerol was used as the source of 1,3-propanediol. Conversions of 12 and 10% were achieved at 60 and 42 °C, respectively (Table 3). Again, the only detected product was N-propyl aniline. Overall 3 was the favoured product at lower conversion and lower temperatures in the aqueous/ionic liquid biphasic reaction medium. Further studies are underway to assess the scope of this transformation, and to optimise the reaction conditions. We are also investigating methods to recover and re-use the catalyst.

We have demonstrated that a one pot bio- and chemo-catalytic process can be used for direct conversion of crude glycerol to valuable secondary amines in a biphasic system without intermediate separation of 1,3-propanediol. The reaction conditions for the biological and chemocatalytic reactions are remarkably convergent with respect to solvents, temperature and pressure, offering excellent prospects for future development of a fully integrated process.

We thank the EPSRC Life Science Interface Programme for financial support via grants EP/E010636/1 and EP/E010687/1.

Notes and references

1. (a) P. Gallezot, Green Chem., 2007, 9, 295; (b) R. Luque, L. Herrero-Davila, J. M. Campelo, J. H. Clark, J. M. Hidalgo, D. Luna, J. M. Marinas and A. A. Romero, Energy Environ. Sci., 2008, 1, 542.
2. Z.-L. Xiu and A.-P. Zeng, Appl. Microbiol. Biotechnol., 2008, 78, 917.
3. H.-J. Huang, S. Ramaswamy, U. W. Tschirner and B. V. Ramarao, Sep. Purif. Technol., 2008, 62, 1.
4. A. Bruggink, R. Schoevaart and T. Kieboom, Org. Process Res. Dev., 2003, 7, 622.
5. (a) M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi and C. D. Pina, Angew. Chem., Int. Ed., 2007, 46, 4434; (b) A. Behr, J. Eilting, K. Irawadi, J. Leschinski and F. Lindner, Green Chem., 2008, 10, 13; (c) C.-H. Zhou, J. N. Beltramini, Y.-X. Fana and G. Q. Lu, Chem. Soc. Rev., 2008, 37, 527.
6. E. Petitdemange, C. Durr, S. A. Andaloussi and G. Raval, J. Ind. Microbiol., 1995, 15, 498.
7. M. Gonzalez-Pajuelo, J. C. Andrade and I. Vasconcelos, J. Ind. Microbiol. Biotechnol., 2004, 31, 442.
8. S. Papanikolaou, P. Ruiz-Sanchez, B. Pariset, F. Blanchard and M. Fick, J. Biotechnol., 2000, 77, 191.
9. S. Papanikolaou, M. Fick and G. Aggelis, J. Chem. Technol. Biotechnol., 2004, 79, 1189.
10. C. E. Nakamura and G. M. Whited, Curr. Opin. Biotechnol., 2003, 14, 454.
11. J. J. Malinowski, Biotechnol. Tech., 1999, 13, 127.
12. R. V. Oppenauer, Recl. Trav. Chim. Pays-Bas, 1937, 56, 137.
13. F. Hanasaka, K. Fujita and R. Yamaguchi, Organometallics, 2004, 23, 1490.
14. (a) C. F. Winans and H. Adkins, J. Am. Chem. Soc., 1932, 54, 306; (b) M. V. Klyuev and M. L. Khidekel, Russ. Chem. Rev., 1980, 49, 14; (c) R. Grigg, T. R. B. Mitchell, S. Sutthivaiyakit and N. Tongpenyai, J. Chem. Soc., Chem. Commun., 1981, 611; (d) Y. Tsuji, R. Takeuchi, H. Ogawa and Y. Watanabe, Chem. Lett., 1986, 293; (e) D. M. Roundhill, Chem. Rev., 1992, 92, 1; (f) K.-i. Fujita, Z. Li, N. Ozeki and R. Yamaguchi, Tetrahedron Lett., 2003, 44, 2687; (g) H. S. A. Hamid, P. S. Slatford and J. M. J. Williams, Adv. Synth. Catal., 2007, 349, 1555 , and reference therein; (h) A. P. Da Costa, M. Viciano, M. Sanaú, S. Merino, J. Tejeda, E. Peris and B. Royo, Organometallics, 2008, 27, 1305.
15. (a) R. N. Salvatore, C. H. Yoon and K. W. Jung, Tetrahedron, 2001, 57, 7785; (b) M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Kesseler, R. Stuermer and T. Zelinski, Angew. Chem., Int. Ed., 2004, 43, 788.
16. H. Biebl, K. Menzel, A. P. Zeng and W. D. Deckwer, Appl. Microbiol. Biotechnol., 1999, 52, 289.
17. C. butyricum DSM10703 was grown anaerobically in modified Y5 medium¹⁸ at 35 °C and pH 6.5. Pure or crude glycerol was added to 20 g L⁻¹. Initial and final glycerol concentrations and concentrations of fermentation products were measured by HPLC. Further details are provided in ESI†.
18. M. Rebros, N. Gunaratne, J. Ferguson, K. R. Seddon and G. Stephens, Green Chem., 2009, 11, 402.
19. (a) R. Corbera, M. Sanau and E. Peris, J. Am. Chem. Soc., 2006, 128, 3974; (b) A. C. Marr, C. L. Pollock and G. C. Saunders, Organometallics, 2007, 26, 3283.
20. C. L. Woldringh, J. Bacteriol., 1973, 114, 1359.
21. H. Pfruender, M. Midjojo, U. Kragl and D. Weuster-Botz, Angew. Chem., Int. Ed., 2004, 43, 4529.
22. H. Li, P. Williams, J. Micklefield, J. M. Gardiner and G. Stephens, Tetrahedron, 2004, 60, 753.
23. M. F. Rodriguez Martinez, N. Kelessidou, Z. Law, J. Gardiner and G. Stephens, Anaerobe, 2008, 14, 55.
24. H. Korbekandi, P. Mather, J. Gardiner and G. Stephens, Enzyme Microb. Technol., 2008, 42, 308.
25. R. J. Cornmell, C. L. Winder, S. Schuler, R. Goodacre and G. Stephens, Green Chem., 2008, 10, 685.

---

Footnote

† Electronic supplementary information (ESI) available: Protocols for fermentation, amination and characterisation of the products. See DOI: 10.1039/b820657k

---