CHEMSUSCHEM COMMUNICATIONS DOI: 10.1002/cssc.201300964
Selective Conversion of Biorefinery Lignin into Dicarboxylic Acids Ruoshui Ma, Mond Guo, and Xiao Zhang*[a] The emerging biomass-to-biofuel conversion industry has created an urgent need for identifying new applications for biorefinery lignin. This paper demonstrates a new route to producing dicarboxylic acids from biorefinery lignin through chalcopyrite-catalyzed oxidation in a highly selective process. Up to 95 % selectivity towards stable dicarboxylic acids was obtained for several types of biorefinery lignin and model compounds under mild, environmentally friendly reaction conditions. The findings from this study paved a new avenue to biorefinery lignin conversions and applications.
Lignin is a ubiquitous component in almost all plant biomass.[1] Large quantities of industrial lignin are already produced annually as waste product of the pulp and paper industry. The vast majority is burned as low-cost fuel to provide energy for the chemical pulping process.[2] The emerging biomass refinery industry will further introduce enormous amounts of lignin.[3] Thus, there is an urgent need to develop technologies that can create new applications for biorefinery lignin. Lignin is the largest source of renewable material with an aromatic skeleton. Depolymerizing lignin to low-molecularweight aromatic compounds (LMWAC) offers a promising route to generating value-added products and has been the focus of much research.[2, 4] However, among the oxidative conversion products one potential group of chemicals has been overlooked: that of the dicarboxylic acids (DCAs). DCAs such as muconic, maleic, and succinic acids are important and highly valuable industrial chemicals and intermediates used in many industries, including the biopolymer, pharmaceutical, and food additives industries.[5] Current commercial dicarboxylic acids are all produced from petroleum-based feedstocks.[6] Developing a green route to produce DCAs from renewable biomass lignin will be of a prime interest to both the chemical and biomass conversion industries.[7] Oxidative aromatic ring cleavage to yield muconic acid and its derivatives is a well-recognized reaction.[8] Although DCAs and their derivatives have been detected in a reaction mixture from oxidative delignification of wood pulp,[9] there has not been a successful attempt to produce DCAs from biomass lignin. Discovering a selective route
[a] R. Ma, M. Guo, Prof. Dr. X. Zhang Voiland School of Chemical Engineering and Bioengineering Bioproducts, Science & Engineering Laboratory Washington State University 2710 Crimson Way, Richland, WA, 99354 (USA) E-mail: [email protected] Supporting Information for this article is available on the WWW under http://dx.doi.org/10.1002/cssc.201300964.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
to convert lignin into DCAs will provide a new avenue towards lignin utilization. Herein, we report the selective production of DCAs through chalcopyrite (CuFeS2)-catalyzed oxidation of biorefinery lignin including diluted-acid corn stover lignin (DACSL) and steam-exploded spruce lignin (SESPL; for details see the Supporting Information 1.2.), in the presence of hydrogen peroxide (H2O2). Chalcopyrite is a ditransition metal catalyst that belongs to the family of transition-metal dichalcogenides.[10] The unique structure of chalcopyrite with complex metal oxidation states, including CuI, CuII, FeII, and FeIII, allows for highly effective electron transfer on the surface of, and inside the, chalcopyrite crystals.[11] This opens a potential for application of chalcopyrite as an effective mediator of catalytic oxidation reactions. Although H2O2 can generate oxidation products on its own, chalcopyrite-aided catalysis gives rise to successful reactions at conditions mild enough that control samples with only H2O2 exhibited no catalytic activity. Figure 1 shows the products generated from DACSL during a 5 h reaction at 60 8C with chalcopyrite in the presence of
Figure 1. Product distribution of DACSLs after 5 h oxidation.
H2O2 at pH 4 (acetic acid buffer). DACSL is a GSH-type lignin that contains all three phenylpropane units: guaiacyl (G), syringyl (S), and p-hydroxylphenyl (H). The G:S:H ratio for DACSL is 57/19/24 as determined by 13C NMR analysis (for details see the Supporting Information, 1.2.2a). Significant amounts of LMWAC including p-coumaric acid, p-hydroquinone, p-hydroxybenzoic acid, and vanillic acid were produced after the first hour of reaction. After 2 h of oxidation, ring cleavage of LMWAC became the predominant reaction, leading to the formation of the following DCAs: malonic acid (35 %), succinic acid (50 %), malic acid (6 %), and maleic acid (3 %). The selectivChemSusChem 2014, 7, 412 – 415
412
CHEMSUSCHEM COMMUNICATIONS
www.chemsuschem.org
ity (Figure 1) and yield (Supporting Information Figure S4) tocaused by a transition metal ion-catalyzed Fenton-like reaction, wards these acids continued to rise as the oxidation time inwhich generates hydroxyl radicals (HOC) that disrupt lignin linkcreased. DCA represented approximately 94 % of the total soluages and produce monolignols.[14] The formation of these HOC ble products after 5 h. as reactive species is catalyzed by the Cu + /Cu2 + and Fe2 + /Fe3 + SESPL contains predominantly guaiacyl units, approximately redox pairs on the surface of the chalcopyrite catalyst under 95 % as determined by 13C NMR analysis (for details see the mildly acidic conditions (acetate buffer, pH range 3–5). Based on XRD patterns, only small structural changes occurred on Supporting Information 1.2.2b). Similar to DACSL oxidation, chalcopyrite during the 5 h reaction. The appearance of a peak conversion of steam-exploded spruce lignin (SESPL) by chalcoat 2 q = 47.5 is probably due to a small leaching of the transipyrite/H2O2 led to the initial production of LMWACs, such as tion metal ion from the catalyst surface (for details see the benzoic acid, vanillin, and vanillic acid, followed by ring cleavSupporting Information Figure S1). age of these compounds to DCAs. After 5 h, malonic acid To verify and understand the subsequent ring cleavage reac(36 %), succinic acid (48 %), maleic acid (2 %), and malic acid tion, the model compounds guaiacol, catechol, and vanillin (3 %) represented about 89 % of the soluble products. were used to represent different LMWACs. Chalcopyrite/H2O2 Although the intermediate LMWAC products differ, the selectivity towards the final DCAs was very similar to that of the oxidation resulted in the selective formation of DCA products DACSL. The mass conversion yields from DACSL and SESPL to that were identical to those observed in both biorefinery lignin DCAs under the tested conditions were 14 % and 11 %, respecreactions (for details see the Supporting Information Figurtively. From the change in product distribution over time es S6–12). All reactions achieved high selectivity, 83–95 %, to(LMWAC to DCA), it can be seen that chalcopyrite-catalyzed wards DCA, with conversion yields between 18 % and 33 % lignin oxidation occurred in a two-stage reaction: initial lignin (Table 1). The high product selectivity would help facilitate depolymerization during short reaction times followed by aroDCA separation by either recrystallization or distillation typicalmatic ring cleavage at longer reaction times. A variety of multily used for diacid separation.[15] This product distribution not hydroxylated/methoxylated phenols and phenolic acids were only confirms that DCAs originate from the aromatic rings in the main products from depolymerization, whereas the reacthe biorefinery lignin, but also reaffirms the convergence of intion converged to two primary DCA products during the ring cleavage. Table 1. Primary DCA distribution after 5 h oxidation of lignin and 3 h oxidation of The cleavage of b-aryl ether bonds during mild monolignols. acid hydrolysis is well understood;[12] however, little is known about the kinetics and mechanism of carbon– Compounds DCA distribution [%] Yield[a] Total selectivity carbon bond cleavage between phenylpropane units malonic acid succinic acid malic acid [%] [%] at mild reaction conditions. To better understand the DACSL 35 50 6 14 91 mechanism of the initial lignin depolymerization, hySESPL 36 48 3 11 85 guaiacol 51 30 14 18 95 droxymatairesinol (HMR), representing a phenylprocatechol 49 25 18 33 92 panyl dimer with b–b linkages commonly observed vanillin 51 19 13 25 83 in lignin, was reacted with chalcopyrite/H2O2. Approx[a] Yield calculations of DACSL and SESPL are based on mass yield. Yield calculations imately 66 wt % of HMR was converted to vanillin, vaof monolignols are based on carbon yield. nillic acid, and vanillyl alcohol after 1 h (Scheme 1;
termediate products to DCAs as seen in the lignin reactions. This observation is the key to understanding the reaction mechanisms leading to this high selectivity. Detailed examination of the guaiacol reaction pathway reScheme 1. Oxidation of hydroxymatairesinol as a model compound for understanding lignin depolymerization. vealed that two consecutive events occurred during ring cleavage. First, after 30 min of reaction, 54 % of the initial see the Supporting Information Figure S5). These HMR degraguaiacol was converted to a mixture of catechol (79 %) and hydation results confirmed the occurrence of depolymerization in droquinone (12 %) with small amounts of 2-methoxyresorcinol the initial stages of the biorefinery lignin reactions, with hyand 3-methoxy-1,2-benzenediol. Second, after 3 h of reaction, droxylation and oxidative cleavage of side chains of the lignin all aromatic rings were transformed into a mixture of malonic contributing to the formation of phenols and phenolic acids. acid (51 %), succinic acid (30 %), maleic acid (4 %), and malic This corresponds with the previously reported use of copperacid (14 %). These results suggest that an initial ring hydroxylbased catalysts in the selective removal of lignin side groups ation of monolignol occurred prior to ring cleavage to DCAs. to reveal the aromatic nuclei structure.[13] This reaction is likely 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 412 – 415
413
CHEMSUSCHEM COMMUNICATIONS
www.chemsuschem.org
Scheme 2. Proposed ring-opening pathways using guaiacol as model compounds and the hindering effect with a-tocopherol.
The ring hydroxylation reactions are postulated to be the result of a combination of two reactive species: HOC and hydroxonium ions (HO + ). The formation and role of HOC has already been well explored,[9, 16] and HO + is known to be formed by the heterolytic cleavage of the peroxy bond (O O) in H2O2.[17] The strong reaction preference for catechol and hydroquinone as the predominant ring hydroxylation intermediates suggests the presence of HO + and its subsequent electrophilic attack on the o-methoxy moiety of guaiacol. To further understand the role of HOC and HO + on ring hydroxylation and cleavage, a-tocopherol was added during guaiacol oxidation in the presence of chalcopyrite/H2O2 at a dosage of 2.0 mol per mol of H2O2 to inhibit radical formation. a-Tocopherol addition led to the formation of hydroxylation and demethoxylation products preferentially, leaving the aromatic ring intact (Scheme 2; for details see the Supporting Information Figure S8). Only small amounts of DCAs were observed even after 3 h of reaction. This result substantiates the essential role of radicals towards ring opening reactions and confirms that HO + contributes to ring hydroxylation and demethoxylation. As reported previously for phenol oxidation (especially on cuprous catalysts), o-quinone structures are typically cleaved into muconic acid, whereas p-quinone structures are transformed into maleic/fumaric acid and oxalic acid.[18] Maleic acid was detected only in small amounts during guaiacol oxidation (4 %) as well as during biorefinery lignin oxidation, indicating its brief presence as an unstable intermediate before ring cleavage. However, no muconic acid was detected in the reaction products by GC–MS, likely due to its low solubility in aqueous solutions. Oxidation of a pure muconic acid suspension by chalcopyrite/H2O2 was conducted as a control experiment; despite low conversions, malonic acid, succinic acid, and maleic/fumaric acid were detected as products. Thus, it is possible that muconic acid contributed to the formation of the final end products in all reactions. The results from the oxidation of model compounds show that the chalcopyrite/H2O2 system can promote a mild Fenton reaction, contributing to the selective conversion of biorefinery lignin to DCAs. Both HOC and HO + were present as the primary reactive species. The main reaction pathways involved in chalcopyrite/H2O2 oxidation of lignin are illustrated in Scheme 3. The first reaction step is the depolymerization of lignin by side-chain substitution (ring hydroxylation by HO + ) and/or oxidative side-chain cleavage to produce LMWACs that are further oxidized to p- and o-quinone derivatives (Scheme 3 A). Aromat 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Scheme 3. Proposed mechanisms for A) lignin depolymerization and aromatic nuclei oxidation, B) aromatic ring cleavage, and C) formation of final products.
ic ring cleavage of quinone to maleic/fumaric acid and muconic acid derivatives is the second reaction step (Scheme 3 B). It is likely that muconic acid then quickly converts to maleic/fumaric acid. The third reaction step is the formation of malonic and succinic acids from the hydrolysis of maleic/fumaric acid, as observed in all reaction results (Scheme 3 C). To verify this last step, control experiments using only maleic acid produced both malonic and succinic acid. Small amounts of malic acid seen in all reaction results suggest that further oxidation of malic acid accompanied by decarboxylation is likely the mechanism leading to the formation of malonic acid. The exact reaction mechanism describing succinic acid formation is still under investigation. The observation of small amounts of formic acid during the reaction suggests a mechanism of intermolecular electron transfer between maleic/fumaric acid and ChemSusChem 2014, 7, 412 – 415
414
CHEMSUSCHEM COMMUNICATIONS formic acid to form succinic acid. Similar reaction mechanisms using formic acid derivatives as electron donors have been reported.[19] It is speculated that the saturated nature of succinic and malonic acid makes them stable endpoints in the oxidation process, contributing to their high selectivity in every reaction. According to the reaction mechanism, a theoretical carbon yield of 40 % can be achieved from biorefinery lignin conversion to DCAs based on a guaiacyl-type (C10) monolignol unit. This yield would be significant considering lignin as a renewable-carbon-neutral feedstock. However, relatively low lignin-to-DCA conversion yields were obtained from the conditions used in this study. There is a significant potential to improve the DCA yield by optimizing the reaction process condition. In conclusion, chalcopyrite-catalyzed H2O2 oxidation of biorefinery lignin demonstrated a new, highly selective conversion route to produce valuable dicarboxylic acid products from biomass lignin. In addition, the mild reaction conditions offer competitive advantages over reductive aromatic conversions to linear-chain hydrocarbons. The findings from this study will pave a new avenue to biorefinery lignin conversions and applications.
Experimental Section The oxidation reactions were performed in 20 mL batch reactors (clear glass headspace vials). Substrates, buffer (pH 4), catalyst, and H2O2 were mixed well prior to reaction. The mixtures were reacted in a preheated incubator (60 8C) for 0–5 h. The reaction products were extracted from the reaction mixtures through liquid–liquid extraction using ethyl acetate as the solvent. The ethyl-acetate extracts were dried under nitrogen and silylated with trimethylchlorosilane (TMCS) (10 % in BSTFA, N,O-bis(trimethylsilyl)trifluoroacetamide) prior to GC–MS analysis. All selectivity data was based on the results from GC–MS (for details see the Supporting Information).
Acknowledgements Authors are thankful to the Northwest Advance Renewable Alliance (NARA) for the financial support to this project through the United States Department of Agriculture Grant no. 2011-6800530416. 13C NMR analysis was performed using EMSL, a national scientific user facility sponsored by the Department of Energy’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. We would also like to thank Drs. Melvin Tucker and Xiaowen Chen from the National Renewable Energy Laboratory for providing the corn stover lignin and Dr. James Deng at FPInnovations for providing steam-exploded spruce samples. Keywords: chalcopyrite · cleavage reaction · dicarboxylic acid · lignin · oxidation [1] a) D. Fengel, G. Wegener, Wood: Chemistry, Ultrastructure, Reactions, Walter de Gruyter, Berlin, 1984; b) E. Sjostrom, Wood Chemistry, Fundamentals and Applications, 2nd ed., Academic Press, Orlando, 1993.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org [2] X. Zhang, M. B. Tu, M. G. Paice, Bioenerg. Res. 2011, 4, 246 – 257. [3] J. E. Holladay, J. J. Bozell, J. F. White, D. Johnson, Top Value-Added Chemicals from Biomass, Volume II: Results of Screening for Potential Candidates from Biorefinery Lignin, U. S. Department of Energy, 2007. [4] a) J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius, B. M. Weckhuysen, Chem. Rev. 2010, 110, 3552 – 3599; b) H. Lange, S. Decina, C. Crestini, Eur Polym J. 2013, 49, 1151 – 1173; c) S. K. Hanson, R. L. Wu, L. A. Silks, Angew. Chem. 2012, 124, 3466 – 3469; Angew. Chem. Int. Ed. 2012, 51, 3410 – 3413; d) A. Rahimi, A. Azarpira, H. Kim, J. Ralph, S. S. Stahl, J. Am. Chem. Soc. 2013, 135, 6415 – 6418. [5] a) S. Y. Lee, S. H. Hong, S. H. Lee, S. J. Park, Macromol. Biosci. 2004, 4, 157 – 164; b) K. Sato, M. Aoki, R. Noyori, Science 1998, 281, 1646 – 1647; c) M. Dugal, G. Sankar, R. Raja, J. M. Thomas, Angew. Chem. 2000, 112, 2399 – 2402; Angew. Chem. Int. Ed. 2000, 39, 2310 – 2313; d) H. Yu, F. Peng, J. Tan, X. W. Hu, H. J. Wang, J. A. Yang, W. X. Zheng, Angew. Chem. 2011, 123, 4064 – 4068; Angew. Chem. Int. Ed. 2011, 50, 3978 – 3982. [6] a) B. Cornils, P. Lappe in Ullmann’s Encyclopedia of Industrial Chemistry, 2006; b) A. Castellan, J. C. J. Bart, S. Cavallaro, Catal. Today 1991, 9, 237 – 254. [7] a) D. R. Dodds, R. A. Gross, Science 2007, 318, 1250 – 1251; b) T. Werpy, G. Petersen, A. Aden, J. Bozell, J. Holladay, J. White, A. Manheim, D. Eliot, L. Lasure, S. Jones, Top Value-Added Chemicals from Biomass, Volume I—Results of Screening for Potential Candidates from Sugars and Synthesis, GasDTIC Document, 2004; c) P. Gallezot, Chem. Soc. Rev. 2012, 41, 1538 – 1558; d) K. Niemel, J. Chem. Technol. Biotechnol. 1990, 48, 17 – 28. [8] a) T. R. Demmin, M. D. Swerdloff, M. M. Rogic, J. Am. Chem. Soc. 1981, 103, 5795 – 5804; b) G. Lin, G. Reid, T. D. H. Bugg, J. Am. Chem. Soc. 2001, 123, 5030 – 5039; c) N. Q. Ran, L. S. Zhao, Z. M. Chen, J. H. Tao, Green Chem. 2008, 10, 361 – 372. [9] J. Gierer, Wood Sci. Technol. 1986, 20, 1 – 33. [10] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, M. S. Strano, Nat. Nanotechnol. 2012, 7, 699 – 712. [11] a) M. Fujisawa, S. Suga, T. Mizokawa, A. Fujimori, K. Sato, Phys. Rev. B 1994, 49, 7155 – 7164; b) K. P. Jayadevan, T. Y. Tseng, J. Nanosci. Nanotechnol. 2005, 5, 1768 – 1784. [12] a) K. Lundquist, T. Kirk, Acta Chem. Scand. 1971, 25, 889 – 894; b) K. Lundquist, Acta Chem. Scand. 1972, 26, 2005 – 2023; c) K. Lundquist, Studies on lignin hydrolysis products and their relations to lignin structures, Chalmers University of Technology, 1973; d) S. Yasuda, N. Terashima, T. Ito, J. Jpn. Wood Res. Soc. 1981, 27, 478 – 483; e) S. Yasuda, N. Terashima, H. Kaneko, J. Jpn. Wood Res. Soc. 1982, 28, 570 – 576; f) T. Ito, N. Terashima, S. Yasuda, J. Jpn. Wood Res. Soc. 1981, 27, 383 – 387. [13] a) K. Kaiser, R. Benner, Anal. Chem. 2012, 84, 459 – 464; b) M. A. GoÇi, S. Montgomery, Anal. Chem. 2000, 72, 3116 – 3121. [14] a) C. Crestini, M. Crucianelli, M. Orlandi, R. Saladino, Catal. Today 2010, 156, 8 – 22; b) T. D. H. Bugg, M. Ahmad, E. M. Hardiman, R. Rahmanpour, Nat. Prod. Rep. 2011, 28, 1883 – 1896. [15] a) A. Orjuela, A. Orjuela, C. T. Lira, D. J. Miller, Bioresour. Technol. 2013, 139, 235 – 241; b) C. R. Campbell, (Monsanto) US 3180878, 1965. [16] J. A. Buswell, E. Odier, T. K. Kirk, Crit. Rev. Biotechnol. 1987, 6, 1 – 60. [17] a) W. Nam, H. J. Han, S.-Y. Oh, Y. J. Lee, M.-H. Choi, S.-Y. Han, C. Kim, S. K. Woo, W. Shin, J. Am. Chem. Soc. 2000, 122, 8677 – 8684; b) H. Tan, R. Yang, W. Sun, S. Wang, Ind. Eng. Chem. Res. 2010, 49, 1473 – 1479; c) X. Zhao, K. Cheng, D. Liu, Appl. Microbiol. Biotechnol. 2009, 82, 815 – 827. J. Gierer, Wood Sci. Technol. 1986, 20, 1 – 33; d) D. C. Johnson in Chemistry of Delignification with Oxygen, Ozone, and Peroxides (Eds.: J. S. Gratzl, J. Nakano, R. P. Singh), 1980, pp. 217 – 228. [18] a) G. Speier, Z. Tyeklar, J. Mol. Catal. 1980, 9, 233 – 235; b) J. G. Walsh, P. J. Furlong, L. A. Byrne, D. G. Gilheany, Tetrahedron 1999, 55, 11519 – 11536. [19] a) R. Shen, T. Chen, Y. Zhao, R. Qiu, Y. Zhou, S. Yin, X. Wang, M. Goto, L.B. Han, J. Am. Chem. Soc. 2011, 133, 17037 – 17044; b) J. M. Brunel, Synlett 2007, 0330 – 0332.
Received: September 10, 2013 Revised: November 25, 2013 Published online on January 24, 2014
ChemSusChem 2014, 7, 412 – 415
415
Selective Conversion of Biorefinery Lignin into Dicarboxylic Acids Ruoshui Ma, Mond Guo, and Xiao Zhang*[a] The emerging biomass-to-biofuel conversion industry has created an urgent need for identifying new applications for biorefinery lignin. This paper demonstrates a new route to producing dicarboxylic acids from biorefinery lignin through chalcopyrite-catalyzed oxidation in a highly selective process. Up to 95 % selectivity towards stable dicarboxylic acids was obtained for several types of biorefinery lignin and model compounds under mild, environmentally friendly reaction conditions. The findings from this study paved a new avenue to biorefinery lignin conversions and applications.
Lignin is a ubiquitous component in almost all plant biomass.[1] Large quantities of industrial lignin are already produced annually as waste product of the pulp and paper industry. The vast majority is burned as low-cost fuel to provide energy for the chemical pulping process.[2] The emerging biomass refinery industry will further introduce enormous amounts of lignin.[3] Thus, there is an urgent need to develop technologies that can create new applications for biorefinery lignin. Lignin is the largest source of renewable material with an aromatic skeleton. Depolymerizing lignin to low-molecularweight aromatic compounds (LMWAC) offers a promising route to generating value-added products and has been the focus of much research.[2, 4] However, among the oxidative conversion products one potential group of chemicals has been overlooked: that of the dicarboxylic acids (DCAs). DCAs such as muconic, maleic, and succinic acids are important and highly valuable industrial chemicals and intermediates used in many industries, including the biopolymer, pharmaceutical, and food additives industries.[5] Current commercial dicarboxylic acids are all produced from petroleum-based feedstocks.[6] Developing a green route to produce DCAs from renewable biomass lignin will be of a prime interest to both the chemical and biomass conversion industries.[7] Oxidative aromatic ring cleavage to yield muconic acid and its derivatives is a well-recognized reaction.[8] Although DCAs and their derivatives have been detected in a reaction mixture from oxidative delignification of wood pulp,[9] there has not been a successful attempt to produce DCAs from biomass lignin. Discovering a selective route
[a] R. Ma, M. Guo, Prof. Dr. X. Zhang Voiland School of Chemical Engineering and Bioengineering Bioproducts, Science & Engineering Laboratory Washington State University 2710 Crimson Way, Richland, WA, 99354 (USA) E-mail: [email protected] Supporting Information for this article is available on the WWW under http://dx.doi.org/10.1002/cssc.201300964.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
to convert lignin into DCAs will provide a new avenue towards lignin utilization. Herein, we report the selective production of DCAs through chalcopyrite (CuFeS2)-catalyzed oxidation of biorefinery lignin including diluted-acid corn stover lignin (DACSL) and steam-exploded spruce lignin (SESPL; for details see the Supporting Information 1.2.), in the presence of hydrogen peroxide (H2O2). Chalcopyrite is a ditransition metal catalyst that belongs to the family of transition-metal dichalcogenides.[10] The unique structure of chalcopyrite with complex metal oxidation states, including CuI, CuII, FeII, and FeIII, allows for highly effective electron transfer on the surface of, and inside the, chalcopyrite crystals.[11] This opens a potential for application of chalcopyrite as an effective mediator of catalytic oxidation reactions. Although H2O2 can generate oxidation products on its own, chalcopyrite-aided catalysis gives rise to successful reactions at conditions mild enough that control samples with only H2O2 exhibited no catalytic activity. Figure 1 shows the products generated from DACSL during a 5 h reaction at 60 8C with chalcopyrite in the presence of
Figure 1. Product distribution of DACSLs after 5 h oxidation.
H2O2 at pH 4 (acetic acid buffer). DACSL is a GSH-type lignin that contains all three phenylpropane units: guaiacyl (G), syringyl (S), and p-hydroxylphenyl (H). The G:S:H ratio for DACSL is 57/19/24 as determined by 13C NMR analysis (for details see the Supporting Information, 1.2.2a). Significant amounts of LMWAC including p-coumaric acid, p-hydroquinone, p-hydroxybenzoic acid, and vanillic acid were produced after the first hour of reaction. After 2 h of oxidation, ring cleavage of LMWAC became the predominant reaction, leading to the formation of the following DCAs: malonic acid (35 %), succinic acid (50 %), malic acid (6 %), and maleic acid (3 %). The selectivChemSusChem 2014, 7, 412 – 415
412
CHEMSUSCHEM COMMUNICATIONS
www.chemsuschem.org
ity (Figure 1) and yield (Supporting Information Figure S4) tocaused by a transition metal ion-catalyzed Fenton-like reaction, wards these acids continued to rise as the oxidation time inwhich generates hydroxyl radicals (HOC) that disrupt lignin linkcreased. DCA represented approximately 94 % of the total soluages and produce monolignols.[14] The formation of these HOC ble products after 5 h. as reactive species is catalyzed by the Cu + /Cu2 + and Fe2 + /Fe3 + SESPL contains predominantly guaiacyl units, approximately redox pairs on the surface of the chalcopyrite catalyst under 95 % as determined by 13C NMR analysis (for details see the mildly acidic conditions (acetate buffer, pH range 3–5). Based on XRD patterns, only small structural changes occurred on Supporting Information 1.2.2b). Similar to DACSL oxidation, chalcopyrite during the 5 h reaction. The appearance of a peak conversion of steam-exploded spruce lignin (SESPL) by chalcoat 2 q = 47.5 is probably due to a small leaching of the transipyrite/H2O2 led to the initial production of LMWACs, such as tion metal ion from the catalyst surface (for details see the benzoic acid, vanillin, and vanillic acid, followed by ring cleavSupporting Information Figure S1). age of these compounds to DCAs. After 5 h, malonic acid To verify and understand the subsequent ring cleavage reac(36 %), succinic acid (48 %), maleic acid (2 %), and malic acid tion, the model compounds guaiacol, catechol, and vanillin (3 %) represented about 89 % of the soluble products. were used to represent different LMWACs. Chalcopyrite/H2O2 Although the intermediate LMWAC products differ, the selectivity towards the final DCAs was very similar to that of the oxidation resulted in the selective formation of DCA products DACSL. The mass conversion yields from DACSL and SESPL to that were identical to those observed in both biorefinery lignin DCAs under the tested conditions were 14 % and 11 %, respecreactions (for details see the Supporting Information Figurtively. From the change in product distribution over time es S6–12). All reactions achieved high selectivity, 83–95 %, to(LMWAC to DCA), it can be seen that chalcopyrite-catalyzed wards DCA, with conversion yields between 18 % and 33 % lignin oxidation occurred in a two-stage reaction: initial lignin (Table 1). The high product selectivity would help facilitate depolymerization during short reaction times followed by aroDCA separation by either recrystallization or distillation typicalmatic ring cleavage at longer reaction times. A variety of multily used for diacid separation.[15] This product distribution not hydroxylated/methoxylated phenols and phenolic acids were only confirms that DCAs originate from the aromatic rings in the main products from depolymerization, whereas the reacthe biorefinery lignin, but also reaffirms the convergence of intion converged to two primary DCA products during the ring cleavage. Table 1. Primary DCA distribution after 5 h oxidation of lignin and 3 h oxidation of The cleavage of b-aryl ether bonds during mild monolignols. acid hydrolysis is well understood;[12] however, little is known about the kinetics and mechanism of carbon– Compounds DCA distribution [%] Yield[a] Total selectivity carbon bond cleavage between phenylpropane units malonic acid succinic acid malic acid [%] [%] at mild reaction conditions. To better understand the DACSL 35 50 6 14 91 mechanism of the initial lignin depolymerization, hySESPL 36 48 3 11 85 guaiacol 51 30 14 18 95 droxymatairesinol (HMR), representing a phenylprocatechol 49 25 18 33 92 panyl dimer with b–b linkages commonly observed vanillin 51 19 13 25 83 in lignin, was reacted with chalcopyrite/H2O2. Approx[a] Yield calculations of DACSL and SESPL are based on mass yield. Yield calculations imately 66 wt % of HMR was converted to vanillin, vaof monolignols are based on carbon yield. nillic acid, and vanillyl alcohol after 1 h (Scheme 1;
termediate products to DCAs as seen in the lignin reactions. This observation is the key to understanding the reaction mechanisms leading to this high selectivity. Detailed examination of the guaiacol reaction pathway reScheme 1. Oxidation of hydroxymatairesinol as a model compound for understanding lignin depolymerization. vealed that two consecutive events occurred during ring cleavage. First, after 30 min of reaction, 54 % of the initial see the Supporting Information Figure S5). These HMR degraguaiacol was converted to a mixture of catechol (79 %) and hydation results confirmed the occurrence of depolymerization in droquinone (12 %) with small amounts of 2-methoxyresorcinol the initial stages of the biorefinery lignin reactions, with hyand 3-methoxy-1,2-benzenediol. Second, after 3 h of reaction, droxylation and oxidative cleavage of side chains of the lignin all aromatic rings were transformed into a mixture of malonic contributing to the formation of phenols and phenolic acids. acid (51 %), succinic acid (30 %), maleic acid (4 %), and malic This corresponds with the previously reported use of copperacid (14 %). These results suggest that an initial ring hydroxylbased catalysts in the selective removal of lignin side groups ation of monolignol occurred prior to ring cleavage to DCAs. to reveal the aromatic nuclei structure.[13] This reaction is likely 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 412 – 415
413
CHEMSUSCHEM COMMUNICATIONS
www.chemsuschem.org
Scheme 2. Proposed ring-opening pathways using guaiacol as model compounds and the hindering effect with a-tocopherol.
The ring hydroxylation reactions are postulated to be the result of a combination of two reactive species: HOC and hydroxonium ions (HO + ). The formation and role of HOC has already been well explored,[9, 16] and HO + is known to be formed by the heterolytic cleavage of the peroxy bond (O O) in H2O2.[17] The strong reaction preference for catechol and hydroquinone as the predominant ring hydroxylation intermediates suggests the presence of HO + and its subsequent electrophilic attack on the o-methoxy moiety of guaiacol. To further understand the role of HOC and HO + on ring hydroxylation and cleavage, a-tocopherol was added during guaiacol oxidation in the presence of chalcopyrite/H2O2 at a dosage of 2.0 mol per mol of H2O2 to inhibit radical formation. a-Tocopherol addition led to the formation of hydroxylation and demethoxylation products preferentially, leaving the aromatic ring intact (Scheme 2; for details see the Supporting Information Figure S8). Only small amounts of DCAs were observed even after 3 h of reaction. This result substantiates the essential role of radicals towards ring opening reactions and confirms that HO + contributes to ring hydroxylation and demethoxylation. As reported previously for phenol oxidation (especially on cuprous catalysts), o-quinone structures are typically cleaved into muconic acid, whereas p-quinone structures are transformed into maleic/fumaric acid and oxalic acid.[18] Maleic acid was detected only in small amounts during guaiacol oxidation (4 %) as well as during biorefinery lignin oxidation, indicating its brief presence as an unstable intermediate before ring cleavage. However, no muconic acid was detected in the reaction products by GC–MS, likely due to its low solubility in aqueous solutions. Oxidation of a pure muconic acid suspension by chalcopyrite/H2O2 was conducted as a control experiment; despite low conversions, malonic acid, succinic acid, and maleic/fumaric acid were detected as products. Thus, it is possible that muconic acid contributed to the formation of the final end products in all reactions. The results from the oxidation of model compounds show that the chalcopyrite/H2O2 system can promote a mild Fenton reaction, contributing to the selective conversion of biorefinery lignin to DCAs. Both HOC and HO + were present as the primary reactive species. The main reaction pathways involved in chalcopyrite/H2O2 oxidation of lignin are illustrated in Scheme 3. The first reaction step is the depolymerization of lignin by side-chain substitution (ring hydroxylation by HO + ) and/or oxidative side-chain cleavage to produce LMWACs that are further oxidized to p- and o-quinone derivatives (Scheme 3 A). Aromat 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Scheme 3. Proposed mechanisms for A) lignin depolymerization and aromatic nuclei oxidation, B) aromatic ring cleavage, and C) formation of final products.
ic ring cleavage of quinone to maleic/fumaric acid and muconic acid derivatives is the second reaction step (Scheme 3 B). It is likely that muconic acid then quickly converts to maleic/fumaric acid. The third reaction step is the formation of malonic and succinic acids from the hydrolysis of maleic/fumaric acid, as observed in all reaction results (Scheme 3 C). To verify this last step, control experiments using only maleic acid produced both malonic and succinic acid. Small amounts of malic acid seen in all reaction results suggest that further oxidation of malic acid accompanied by decarboxylation is likely the mechanism leading to the formation of malonic acid. The exact reaction mechanism describing succinic acid formation is still under investigation. The observation of small amounts of formic acid during the reaction suggests a mechanism of intermolecular electron transfer between maleic/fumaric acid and ChemSusChem 2014, 7, 412 – 415
414
CHEMSUSCHEM COMMUNICATIONS formic acid to form succinic acid. Similar reaction mechanisms using formic acid derivatives as electron donors have been reported.[19] It is speculated that the saturated nature of succinic and malonic acid makes them stable endpoints in the oxidation process, contributing to their high selectivity in every reaction. According to the reaction mechanism, a theoretical carbon yield of 40 % can be achieved from biorefinery lignin conversion to DCAs based on a guaiacyl-type (C10) monolignol unit. This yield would be significant considering lignin as a renewable-carbon-neutral feedstock. However, relatively low lignin-to-DCA conversion yields were obtained from the conditions used in this study. There is a significant potential to improve the DCA yield by optimizing the reaction process condition. In conclusion, chalcopyrite-catalyzed H2O2 oxidation of biorefinery lignin demonstrated a new, highly selective conversion route to produce valuable dicarboxylic acid products from biomass lignin. In addition, the mild reaction conditions offer competitive advantages over reductive aromatic conversions to linear-chain hydrocarbons. The findings from this study will pave a new avenue to biorefinery lignin conversions and applications.
Experimental Section The oxidation reactions were performed in 20 mL batch reactors (clear glass headspace vials). Substrates, buffer (pH 4), catalyst, and H2O2 were mixed well prior to reaction. The mixtures were reacted in a preheated incubator (60 8C) for 0–5 h. The reaction products were extracted from the reaction mixtures through liquid–liquid extraction using ethyl acetate as the solvent. The ethyl-acetate extracts were dried under nitrogen and silylated with trimethylchlorosilane (TMCS) (10 % in BSTFA, N,O-bis(trimethylsilyl)trifluoroacetamide) prior to GC–MS analysis. All selectivity data was based on the results from GC–MS (for details see the Supporting Information).
Acknowledgements Authors are thankful to the Northwest Advance Renewable Alliance (NARA) for the financial support to this project through the United States Department of Agriculture Grant no. 2011-6800530416. 13C NMR analysis was performed using EMSL, a national scientific user facility sponsored by the Department of Energy’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. We would also like to thank Drs. Melvin Tucker and Xiaowen Chen from the National Renewable Energy Laboratory for providing the corn stover lignin and Dr. James Deng at FPInnovations for providing steam-exploded spruce samples. Keywords: chalcopyrite · cleavage reaction · dicarboxylic acid · lignin · oxidation [1] a) D. Fengel, G. Wegener, Wood: Chemistry, Ultrastructure, Reactions, Walter de Gruyter, Berlin, 1984; b) E. Sjostrom, Wood Chemistry, Fundamentals and Applications, 2nd ed., Academic Press, Orlando, 1993.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org [2] X. Zhang, M. B. Tu, M. G. Paice, Bioenerg. Res. 2011, 4, 246 – 257. [3] J. E. Holladay, J. J. Bozell, J. F. White, D. Johnson, Top Value-Added Chemicals from Biomass, Volume II: Results of Screening for Potential Candidates from Biorefinery Lignin, U. S. Department of Energy, 2007. [4] a) J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius, B. M. Weckhuysen, Chem. Rev. 2010, 110, 3552 – 3599; b) H. Lange, S. Decina, C. Crestini, Eur Polym J. 2013, 49, 1151 – 1173; c) S. K. Hanson, R. L. Wu, L. A. Silks, Angew. Chem. 2012, 124, 3466 – 3469; Angew. Chem. Int. Ed. 2012, 51, 3410 – 3413; d) A. Rahimi, A. Azarpira, H. Kim, J. Ralph, S. S. Stahl, J. Am. Chem. Soc. 2013, 135, 6415 – 6418. [5] a) S. Y. Lee, S. H. Hong, S. H. Lee, S. J. Park, Macromol. Biosci. 2004, 4, 157 – 164; b) K. Sato, M. Aoki, R. Noyori, Science 1998, 281, 1646 – 1647; c) M. Dugal, G. Sankar, R. Raja, J. M. Thomas, Angew. Chem. 2000, 112, 2399 – 2402; Angew. Chem. Int. Ed. 2000, 39, 2310 – 2313; d) H. Yu, F. Peng, J. Tan, X. W. Hu, H. J. Wang, J. A. Yang, W. X. Zheng, Angew. Chem. 2011, 123, 4064 – 4068; Angew. Chem. Int. Ed. 2011, 50, 3978 – 3982. [6] a) B. Cornils, P. Lappe in Ullmann’s Encyclopedia of Industrial Chemistry, 2006; b) A. Castellan, J. C. J. Bart, S. Cavallaro, Catal. Today 1991, 9, 237 – 254. [7] a) D. R. Dodds, R. A. Gross, Science 2007, 318, 1250 – 1251; b) T. Werpy, G. Petersen, A. Aden, J. Bozell, J. Holladay, J. White, A. Manheim, D. Eliot, L. Lasure, S. Jones, Top Value-Added Chemicals from Biomass, Volume I—Results of Screening for Potential Candidates from Sugars and Synthesis, GasDTIC Document, 2004; c) P. Gallezot, Chem. Soc. Rev. 2012, 41, 1538 – 1558; d) K. Niemel, J. Chem. Technol. Biotechnol. 1990, 48, 17 – 28. [8] a) T. R. Demmin, M. D. Swerdloff, M. M. Rogic, J. Am. Chem. Soc. 1981, 103, 5795 – 5804; b) G. Lin, G. Reid, T. D. H. Bugg, J. Am. Chem. Soc. 2001, 123, 5030 – 5039; c) N. Q. Ran, L. S. Zhao, Z. M. Chen, J. H. Tao, Green Chem. 2008, 10, 361 – 372. [9] J. Gierer, Wood Sci. Technol. 1986, 20, 1 – 33. [10] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, M. S. Strano, Nat. Nanotechnol. 2012, 7, 699 – 712. [11] a) M. Fujisawa, S. Suga, T. Mizokawa, A. Fujimori, K. Sato, Phys. Rev. B 1994, 49, 7155 – 7164; b) K. P. Jayadevan, T. Y. Tseng, J. Nanosci. Nanotechnol. 2005, 5, 1768 – 1784. [12] a) K. Lundquist, T. Kirk, Acta Chem. Scand. 1971, 25, 889 – 894; b) K. Lundquist, Acta Chem. Scand. 1972, 26, 2005 – 2023; c) K. Lundquist, Studies on lignin hydrolysis products and their relations to lignin structures, Chalmers University of Technology, 1973; d) S. Yasuda, N. Terashima, T. Ito, J. Jpn. Wood Res. Soc. 1981, 27, 478 – 483; e) S. Yasuda, N. Terashima, H. Kaneko, J. Jpn. Wood Res. Soc. 1982, 28, 570 – 576; f) T. Ito, N. Terashima, S. Yasuda, J. Jpn. Wood Res. Soc. 1981, 27, 383 – 387. [13] a) K. Kaiser, R. Benner, Anal. Chem. 2012, 84, 459 – 464; b) M. A. GoÇi, S. Montgomery, Anal. Chem. 2000, 72, 3116 – 3121. [14] a) C. Crestini, M. Crucianelli, M. Orlandi, R. Saladino, Catal. Today 2010, 156, 8 – 22; b) T. D. H. Bugg, M. Ahmad, E. M. Hardiman, R. Rahmanpour, Nat. Prod. Rep. 2011, 28, 1883 – 1896. [15] a) A. Orjuela, A. Orjuela, C. T. Lira, D. J. Miller, Bioresour. Technol. 2013, 139, 235 – 241; b) C. R. Campbell, (Monsanto) US 3180878, 1965. [16] J. A. Buswell, E. Odier, T. K. Kirk, Crit. Rev. Biotechnol. 1987, 6, 1 – 60. [17] a) W. Nam, H. J. Han, S.-Y. Oh, Y. J. Lee, M.-H. Choi, S.-Y. Han, C. Kim, S. K. Woo, W. Shin, J. Am. Chem. Soc. 2000, 122, 8677 – 8684; b) H. Tan, R. Yang, W. Sun, S. Wang, Ind. Eng. Chem. Res. 2010, 49, 1473 – 1479; c) X. Zhao, K. Cheng, D. Liu, Appl. Microbiol. Biotechnol. 2009, 82, 815 – 827. J. Gierer, Wood Sci. Technol. 1986, 20, 1 – 33; d) D. C. Johnson in Chemistry of Delignification with Oxygen, Ozone, and Peroxides (Eds.: J. S. Gratzl, J. Nakano, R. P. Singh), 1980, pp. 217 – 228. [18] a) G. Speier, Z. Tyeklar, J. Mol. Catal. 1980, 9, 233 – 235; b) J. G. Walsh, P. J. Furlong, L. A. Byrne, D. G. Gilheany, Tetrahedron 1999, 55, 11519 – 11536. [19] a) R. Shen, T. Chen, Y. Zhao, R. Qiu, Y. Zhou, S. Yin, X. Wang, M. Goto, L.B. Han, J. Am. Chem. Soc. 2011, 133, 17037 – 17044; b) J. M. Brunel, Synlett 2007, 0330 – 0332.
Received: September 10, 2013 Revised: November 25, 2013 Published online on January 24, 2014
ChemSusChem 2014, 7, 412 – 415
415
