Cellulase-Inspired Solid Acids for Cellulose Hydrolysis: Structural

Jan 17, 2018 - This work presents a detailed structure–activity analysis of a polymeric solid acid catalyst used in cellulose hydrolysis. In contras...
5 downloads 8 Views 2MB Size
Download PDF
Letter Cite This: ACS Catal. 2018, 8, 1464−1468

pubs.acs.org/acscatalysis

Cellulase-Inspired Solid Acids for Cellulose Hydrolysis: Structural Explanations for High Catalytic Activity Maksim Tyufekchiev,‡ Pu Duan,¶ Klaus Schmidt-Rohr,¶ Sergio Granados Focil,§ Michael T. Timko,*,‡,∥ and Marion H. Emmert*,†,∥ †

Department of Chemistry and Biochemistry and ‡Department of Chemical Engineering, Worcester Polytechnic Institute, 100 Institute Road, Worcester, Massachusetts 01609, United States § Gustaf H. Carlson School of Chemistry, Clark University, 950 Main Street, Worcester, Massachusetts 01610, United States ¶ Department of Chemistry, Brandeis University, 415 South Street, Waltham, Massachusetts 02453, United States S Supporting Information *

ABSTRACT: This work presents a detailed structure−activity analysis of a polymeric solid acid catalyst used in cellulose hydrolysis. In contrast to previous work, our studies show that the high catalytic activity is likely not due to hydrogen bonding between C−Cl moieties at the polymer surface and cellulose fibers. Instead, we report that such C−Cl bonds hydrolyze readily under polymer functionalization conditions to produce C−OH groups on the exterior of the solid acid beads. Furthermore, continued C−Cl to C−OH substitution under cellulose or cellobiose hydrolysis conditions releases HCl from the resin, which contributes to cellulose hydrolysis. Overall, the presented studies stress the need for detailed, quantitative analysis of polymer structures and spatial distribution of functional groups in order to correctly interpret the catalytic results obtained with polymer-based solid acids. KEYWORDS: biomass, catalysis, cellulose hydrolysis, polymers, cooperative effects, sustainable chemistry

T

the cellulose surface.25,26 Interestingly, aqueous glucan hydrolysis catalyzed by carbonaceous solid acids benefits from association of glucans to the graphitic domains (binding), which is likely driven by entropically favored, hydrophobic effects and enthalpically favored C−H-π interactions.27,28 Similarly, glucan adsorption to mesoporous and microporous carbon materials has been documented.29 In contrast, for polymeric solid acids, similar physicochemical principles of glucan adsorption are not as clearly established. The polymer-based solid acids with the highest activity for cellulose hydrolysis to date have been reported by Shuai and Pan13 and produce up to 93% glucose from cellulose under relatively mild conditions (H2O, 120 °C, 10 h). Pan’s catalyst consists of an aromatic-rich, styrenic polymer decorated with C−Cl moieties (originally referred to as “binding groups”) that are believed to enable hydrogen bonding to cellulose while the sulfonic acid moieties catalyze the glycosidic bond hydrolysis (Scheme 1). Unfortunately, no information on the quantitative composition or functional group distribution has been provided for Pan’s catalyst, thus weakening the arguments for catalyst− cellulose interactions. Despite these issues, follow-up work by several different investigators has described similar design principles for polymer-based solid acids; all designs incorporate hydrogen

he controlled and selective hydrolysis of cellulose has the potential to provide abundant access to carbon-based building blocks such as ethanol, glucose, hydroxymethylfurfural (HMF), and levulinic acid (LA) from renewable, underutilized resources.1 However, cellulose recalcitrance leads to slow conversion into more desirable small molecules.2−5 Cellulose can be hydrolyzed by enzymes at low temperatures (50 °C) or by liquid acids at elevated temperatures. However, enzyme hydrolysis rates are slow, while acid-catalyzed hydrolysis requires high acid loadings or high temperatures, leading to side reactions that limit the yields of useful products.6−10 Furthermore, both types of treatment are costly, as acids and enzymes employed are typically not recoverable. This is one of the reasons why industrial-scale production of glucose from cellulose (on the pathway toward second generation bioethanol) is often challenging.11 Alternatives that have been discussed widely in recent years include solid acid catalysts for cellulose hydrolysis, as these acids might be recoverable after cellulose conversion and have shown promise for direct conversion of cellulose to glucose in high yields and with good selectivity.12 So-called “cellulase-mimetic” solid acids have provided especially remarkable results.13−21 The hypothesized mechanism of action of these acids has been formulated in analogy to the design principles of cellulases (cellulose-cleaving enzymes), which exhibit a cellulose-binding domain and a catalytic domain for cellulose hydrolysis.22−24 In enzymatic hydrolysis, these structural features allow cellulases to catalyze glycosidic bond hydrolysis more efficiently through binding to © 2018 American Chemical Society

Received: December 1, 2017 Revised: January 16, 2018 Published: January 17, 2018 1464

DOI: 10.1021/acscatal.7b04117 ACS Catal. 2018, 8, 1464−1468

Letter

ACS Catalysis Scheme 1. Solid Acid Design Based on Pan’s Catalyst:13 PreCoordination of Sugar Polymer Through “Binding Sites” X Acting as Hydrogen Bond Acceptors

cellobiose and cellulose (Scheme 2B,C; Table S2 and S3 in the SI) were in close agreement with literature data,16 instilling confidence that the obtained material was suitable for more detailed structural analysis. After the successful activity tests, CMP-SO3H-0.3 and its precursor resin CMP were both analyzed using quantitative solid-state 13C nuclear magnetic resonance (NMR) spectroscopy (Figure 1) to determine the concentration of functional

bonding functionalities (e.g., C−Cl, C−CO2H) in addition to strongly acidic moieties,14,16−21 but none of these catalysts match the reported activity of Pan’s catalyst. As such, the question of whether the presence of “binding groups” indeed leads to an increase in hydrolysis activity has not been unambiguously answered. Another drawback of Pan’s catalyst and related catalysts is the undesirably low substrate/catalyst mass ratio (1:1 or lower) required for high hydrolysis activity. The work described in this manuscript focused on identifying and quantifying the role of such binding groups in glycosidic bond hydrolysis through detailed structural and catalytic characterization of a representative solid acid catalyst. We focused on synthesizing a solid acid (called CMP-SO3H0.3 in this manuscript) reported after the initial description of the Pan catalyst had appeared in the literature (Scheme 2A).16 Scheme 2. Synthesis and Catalytic Activity of Solid Acid Catalyst CMP-SO3H-0.3a Figure 1. 13C NMR spectra of polymer precursor (CMP) (top) and CMP-SO3H-0.3 (bottom). Thick black line: all C; thin red line: nonprotonated or mobile C.

a

groups present within the modified polymer beads. NMR confirmed the desired partial substitution of the benzylic C−Cl groups in CMP-SO3H-0.3. Specifically, the intensity of the signal at 138 ppm attributed to the carbon bonded to the chloromethyl groups decreased and a new signal appeared at around 58 ppm, consistent with the formation of the expected benzyl sulfonic acid moiety (for peak assignments, see Figure S7). Quantification of the NMR signals indicated that ∼30% of the C−Cl functionalities in CMP had been converted to C− SO3H moieties in CMP-SO3H-0.3. In addition to the signals of the benzylic C−Cl and C−SO3H groups, the MAS NMR spectrum of CMP-SO3H-0.3 contains a band at 62 ppm, consistent with the presence of benzylic CH2−OH groups constituting ∼14% of the total benzylic functional groups. The presence of C−OH groups is further supported by a weak band observed in the ATR-IR spectrum at 3400 cm−1 (see SI, Figure S1). These observations indicate that the benzylic C−Cl functionalities, previously hypothesized to act as “binding groups“, hydrolyze partially to produce benzylic alcohols under typical polymer modification conditions.30 Notably, after employing CMP-SO3H-0.3 to catalyze cellobiose hydrolysis, the C−OH signal intensity further increased, while the C−Cl signal intensity decreased (Figures 2 and S14), consistent with the general instability of benzylic C−Cl groups under catalytic conditions. Another conclusion suggested by these data is that the release of HCl through hydrolysis of benzylic C−Cl bonds32,33 may be a factor relevant for catalytic hydrolysis reactivity. However, the spectra in

See also ref 16.

The synthesis leading to CMP-SO3H-0.3 is adapted from wellknown procedures for modifying polymers.16,30,31 Furthermore, CMP-SO3H-0.3 does not bear any additional functional groups (such as the amine substructure in Pan’s catalyst) other than C−Cl and C−SO3H moieties. We reasoned that the relative simplicity of CMP-SO3H-0.3 would enable a more straightforward elucidation of structure−activity relationships. The bulk characterization of CMP-SO 3 H-0.3 by ATR-FTIR and elemental analysis (Figure S1 and Table S1 in the SI) was in agreement with the literature.16 Furthermore, the observed catalytic activities of CMP-SO3H-0.3 in the hydrolysis of 1465

DOI: 10.1021/acscatal.7b04117 ACS Catal. 2018, 8, 1464−1468

Letter

ACS Catalysis

Figure 2. 13C NMR spectrum of CMP-SO3H-0.3 before (blue, dotted line) and after (red line) cellobiose hydrolysis. Catalysis conditions: 5 h, 175 °C, 0.2 g of catalyst, 0.1 g of cellobiose, 2 mL of H2O. Figure 4. Cross-sectional Raman analysis of CMP-SO3H-0.3. Marked bands: 1265 cm−1 (CH2−Cl; decreasing from inside to outside of the bead), 1040 cm−1 (CH2−SO3H, increasing from inside to outside of the bead). The value R signifies the distance of the measurement from the center of the polymer bead.

Figures 2 and S13 also raise additional questions: what is the spatial distribution of C−OH and C−Cl groups and which of these two groups (if either) is available for binding carbohydrates during hydrolysis? To investigate, polymer beads of CMP-SO3H-0.3 (particle sizes ∼500 μm) were sectioned in half; Raman and energy-dispersive spectroscopy (EDS; for details, see Figures S5−S9 in the SI) were then performed on the obtained cross sections to gain insight into the spatial distribution of functional groups. Figure 3 indicates the locations of EDS spectra acquired along the bead cross-section and the obtained Cl/S ratios.

saccharides and the benzyl chloride groups seems inconsistent with the spatially resolved structural data. The SI contains additional data obtained from activity tests of catalyst powder (Table S8), which examines the question of surface effects in more detail. Despite these new insights, the structural analysis does not clearly establish the source of the significant catalytic activity exhibited by CMP-SO3H-0.3 in cellulose hydrolysis. To test whether the measured activity in cellulose hydrolysis is simply due to the presence of the sulfonic acid groups, the catalytic activity of CMP-SO3H-0.3 was compared to that of a similarly prepared polymer resin, CMP-SO3H-1.2 (see SI for synthesis and characterization). In CMP-SO3H-1.2, the C−Cl groups are completely substituted with C−SO3H groups, as verified by solid-state 13C NMR, FTIR, Raman, and EDS analysis (for details, see Figures S2, S5, S9, S10, S11, and S12 in the SI). If sulfonic acid groups are the sole source of catalytic activity, and if neither the chloride nor the hydroxyl groups contribute to the observed activity, then CMP-SO3H-1.2 would be expected to show greater activity than CMP-SO3H-0.3, simply due to the greater amount of C−SO3H groups present. However, yields of glucose, LA, and formic acid (3%, 7%, and 12%, respectively) obtained using CMP-SO3H-1.2 for cellulose hydrolysis were lower than those with CMP-SO3H-0.3 (6%, 38%, and 51%, respectively; Scheme 3). Furthermore, the reaction solution using CMP-SO3H-1.2 as catalyst clearly contained residual cellulose as a white powder (Figure 5B), while CMP-SO3H-0.3 yielded a yellow solution suggesting nearly complete cellulose hydrolysis (Figure 5A). These differences in activity of the two catalysts contradict the hypothesis that only the sulfonic acid groups in CMP-SO3H-0.3 are responsible for hydrolysis activity. Interestingly, cellulose hydrolysis experiments employing the precursor polymer CMP with nonmodified benzylic C−Cl moieties also resulted in appreciable cellulose conversion: The use of CMP as catalyst produced less LA (13%) and formic acid (18%) than what was obtained with CMP-SO3H-0.3 (Scheme 3; 38% LA and 51% formic acid, respectively), but more glucose (16%) than when using CMP-SO3H-0.3 as catalyst (6%). These results with the parent resin CMP suggest as a new hypothesis that in situ HCl generation is likely responsible

Figure 3. Locations of EDS measurement on CMP-SO3H-0.3 polymer bead cross-section (left) and obtained Cl/S ratios (right).

Interestingly, the Cl/S ratio is not uniform, with the values measured near the polymer bead’s exterior being smaller than those observed in its interior. Raman analysis (Figure 4) of the cross-section shows an intense band attributable to CH2Cl at 1265 cm−1, which decreases toward the outside of the bead. The opposite trend is observed for the SO3− band at 1040 cm−1. Together, these data indicate that the C−Cl moieties in the center of the polymer beads remain mostly intact, while the chemical modifications occur more completely in the outer regions of the beads than in the interior. EDS and Raman cross-sectional analyses further suggest that access to the inside of the beads is limited for the aqueous reagent mixtures used for polymer modification as well as the polysaccharide hydrolysis reaction mixture. This is most likely due to the hydrophobic environment in the beads’ interior. Notably, with more C−Cl bonds in the interior of the particle, the previously postulated13,16 hydrogen-bonding contact between C−Cl moieties and cellulose fibers during hydrolysis seems less attainable than if the C−Cl groups were to remain intact on the polymer surface. Thus, for the resin under investigation, attributing their superior catalytic activity to the presence of supramolecular interactions between the poly1466

DOI: 10.1021/acscatal.7b04117 ACS Catal. 2018, 8, 1464−1468

Letter

ACS Catalysis Scheme 3. Comparison of Catalytic Activity of CMP-SO3H0.3, CMP-SO3H-1.2, and Catalyst Precursor CMP in Cellulose Hydrolysisa

Figure 5. Visual comparison of cellulose hydrolysis suspensions for (A) CMP-SO3H-0.3, (B) CMP-SO3H-1.2, and (C) leachate from CMP-SO3H-0.3.

measurements, showed a similar trend (0.042 and 0.141 M, respectively). These data confirm that HCl is formed from CMP-SO3H-0.3 when using this material as solid acid catalyst. In a next step, we investigated if HCl formed by leaching CMP-SO3H-0.3 can affect cellulose hydrolysis at similar catalytic levels as the polymer beads. To this end, CMPSO3H-0.3 was treated with H2O at 175 °C for 10 h, mimicking cellulose hydrolysis conditions without the presence of the catalytic substrate. The polymer beads were then removed, and the resulting leachate was used for cellulose hydrolysis studies. In a first set of experiments, cellulose was hydrolyzed only with the leachate as reagent (Scheme 4 and Table S6 in the SI). Importantly, these experiments resulted in significantly higher yields of levulinic acid (57%) and formic acid (60%) than those determined for the tests using CMP-SO3H-0.3 as catalyst. This suggests that HCl formed through hydrolysis of benzylic C−Cl sites on CMP-SO3H-0.3 is a powerful catalyst for cellulose hydrolysis. Furthermore, gradual HCl release from CMPSO3H-0.3 under cellulose hydrolysis conditions would lead to a lower average HCl concentration over the reaction time than found in the HCl-containing leachate, providing a reason for the overall lower activity observed for CMP-SO3H-0.3, compared to the activity of the leachate. Interestingly, high activity (46% levulinic acid, 52% formic acid) of the leachate was still observed after the solution was combined with polymeric CMP-SO3H-1.2 and employed in cellulose hydrolysis (Scheme 4 and Table S7 in the SI). However, both CMPSO3H-0.3 and CMP-SO3H-1.2 appear to play an additional role during cellulose hydrolysis (see Figure 5A/B and Figure S15 in the SI): The polymer beads undergo a color change, possibly consistent with adsorption of humin side products. Such adsorption possibilities are not present in the reaction mixtures employing the leachate, which leads to humin precipitation (Figure 5C). In conclusion, the studies presented herein suggest that benzylic chloride functionalities are hydrothermally labile and mostly absent from the surface of the polymer beads used as solid acid catalysts. As such, benzylic C−Cl groups are unlikely to be involved in supramolecular interactions with polysaccharide substrates such as cellulose. Alternatively, our data suggest that residual benzylic chloride moieties within the polymer beads release HCl under the hydrolysis reaction conditions, providing a homogeneous acid source to catalyze cellulose hydrolysis. Furthermore, our analyses indicate that previously unreported functional groups−specifically benzylic C−OH moieties−are present in the polymer and that their concentration increases upon use of the catalyst in cellobiose hydrolysis. Overall, our results question the commonly accepted role of benzylic C−Cl groups acting as hydrogen bond acceptors and present an alternative explanation for high catalytic activities through a combination of sulfonic acid catalysis and in situ release of HCl; notably, humin adsorption on the polymer surface was observed, which in turn may affect

a

Conditions: 0.100 g of cellulose, water (2.00 mL), polymer catalyst (0.200 g), 175 °C, 10 h, sealed pressure glass vial.

for part of the observed hydrolysis activity with CMP-SO3H0.3. Since NMR analysis indicated formation of benzylic alcohol groups during polymer modification, the activity of a polymer with only benzylic hydroxyl groups (CMP−OH; Scheme 4 and Table S9 in the SI) for cellulose hydrolysis was tested to determine whether the alcohol moieties are capable of catalyzing the hydrolysis reaction. Notably, none of the typical products of cellulose hydrolysis (glucose, levulinic acid, formic acid, HMF) were observed in these tests, indicating that CMP− OH is catalytically inactive in the absence of additional stronger acid groups. Scheme 4. Comparison of Catalytic Activity of CMP-SO3H0.3, CMP−OH, Leachate from Treating CMP-SO3H-0.3 with H2O (175 °C, 10 h), and Leachate + CMP-SO3H-1.2 in Cellulose Hydrolysisa

a Conditions (unless otherwise specified): 0.100 g of cellulose, water (2.00 mL), polymer catalyst (0.200 g), 175 °C, 10 h, sealed pressure glass vial.

To directly test whether formation of HCl occurs in reaction mixtures employing CMP-SO3H-0.3 as catalyst, chloride concentrations were measured in these mixtures after hydrolysis of cellobiose and cellulose (see Table S5 in the SI). As expected, both reaction mixtures showed significant concentrations of chloride (0.051 M for cellobiose hydrolysis at 150 °C after 5 h; 0.195 M for cellulose hydrolysis at 175 °C after 10 h). Moreover, the H+ concentrations, as determined by pH 1467

DOI: 10.1021/acscatal.7b04117 ACS Catal. 2018, 8, 1464−1468

Letter

ACS Catalysis

62498; National Renewable Energy Laboratory: Golden, CO, 2015. DOI: 10.2172/1176746 (12) Huang, Y.-B.; Fu, Y. Green Chem. 2013, 15, 1095−1111. (13) Shuai, L.; Pan, X. Energy Environ. Sci. 2012, 5, 6889−6894. (14) Geremia, J. M.; Baynes, B. M. Polymeric Acid Catalysts and Uses Thereof. Patent No. WO2012118767A1, Feb. 27, 2012. (15) Li, X.; Jiang, Y.; Shuai, L.; Wang, L.; Meng, L.; Mu, X. J. Mater. Chem. 2012, 22, 1283−1289. (16) Zuo, Y.; Zhang, Y.; Fu, Y. ChemCatChem 2014, 6, 753−757. (17) Teong, S. P.; Yi, G.; Cao, X.; Zhang, Y. ChemSusChem 2014, 7, 2120−2124. (18) Yang, Q.; Pan, X. BioEnergy Res. 2016, 9, 578−586. (19) Yu, F.; Thomas, J.; Smet, M.; Dehaen, W.; Sels, B. F. Green Chem. 2016, 18, 1694−1705. (20) Parveen, F.; Gupta, K.; Upadhyayula, S. Carbohydr. Polym. 2017, 159, 146−151. (21) Shen, F.; Smith, R. L.; Li, L.; Yan, L.; Qi, X. ACS Sustainable Chem. Eng. 2017, 5, 2421−2427. (22) Coutinho, J. B.; Gilkes, N. R.; Kilburn, D. G.; Warren, R. A. J.; Miller, J. R. C. FEMS Microbiol. Lett. 1993, 113, 211−217. (23) McCarter, J. D.; Withers, G. S. Curr. Opin. Struct. Biol. 1994, 4, 885−892. (24) Boraston, A. B.; Kwan, E.; Chiu, P.; Warren, R. A. J.; Kilburn, D. G. J. Biol. Chem. 2003, 278, 6120−6127. (25) Bornscheuer, U.; Buchholz, K.; Seibel, J. Angew. Chem., Int. Ed. 2014, 53, 10876−10893. (26) Liu, Z.; Ho, S.-H.; Sasaki, K.; den Haan, R.; Inokuma, K.; Ogino, C.; van Zyl, W. H.; Hasunuma, T.; Kondo, A. Sci. Rep. 2016, 6, 24550. (27) Kobayashi, H.; Yabushita, M.; Komanoya, T.; Hara, K.; Fujita, I.; Fukuoka, A. ACS Catal. 2013, 3, 581−587. (28) (a) Yabushita, M.; Kobayashi, H.; Hasegawa, J.-y.; Hara, K.; Fukuoka, A. ChemSusChem 2014, 7, 1443−1450. (b) Yabushita, M.; Kobayashi, H.; Fukuoka, A. Appl. Catal., B 2014, 145, 1−9. (29) (a) Chung, P. W.; Charmot, A.; Gazit, O. M.; Katz, A. Langmuir 2012, 28, 15222−15232. (b) Chung, P. W.; Charmot, A.; Click, T.; Lin, Y.; Bae, Y.; Chu, J. W.; Katz, A. Langmuir 2015, 31, 7288−7295. (c) Chung, P.-W.; Yabushita, M.; To, A. T.; Bae, Y.; Jankolovits, J.; Kobayashi, H.; Fukuoka, A.; Katz, A. ACS Catal. 2015, 5, 6422−6425. (30) Hwang, M.-L.; Choi, J.; Woo, H.-S.; Kumar, V.; Sohn, J.-Y.; Shin, J. Nucl. Instrum. Methods Phys. Res., Sect. B 2014, 321, 59−65. (31) Fréchet, J. M. J.; de Smet, M. D.; Farrall, M. J. Polymer 1979, 20, 675−680. (32) Beste, G. W.; Hammett, L. P. J. Am. Chem. Soc. 1940, 62, 2481− 2487. (33) Mabey, W.; Mill, T. J. Phys. Chem. Ref. Data 1978, 7, 383−415.

glucan-polymer interactions. Generally, this work highlights the need for caution when interpreting catalytic results with polymer-based solid acids without a detailed, quantitative analysis of polymer structure and spatial distribution of functional groups.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscatal.7b04117. Detailed results of catalytic runs in table format; NMR, FTIR, EDS, Elemental Analysis, pH, and Raman characterization of polymers, experimental procedures, determination of ion content after catalysis (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Klaus Schmidt-Rohr: 0000-0002-3188-4828 Marion H. Emmert: 0000-0003-4375-8295 Author Contributions ∥

M.T.T. and M.H.E. contributed equally.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Massachussetts Clean Energy Center and NSF CBET 1554283 (to M.T.T.). S.G.F. gratefully acknowledges funding from the Sherman-Fairchild foundation to purchase the EDS-capable SEM instrument.

■

ABBREVIATIONS CMP, chloromethyl polystyrene; CMP-SO3H-0.3, chloromethyl polystyrene modified with 0.30 equiv of reagent; CMPSO3H-1.2, chloromethyl polystyrene modified with 1.2 equiv of reagent

■

REFERENCES

(1) Alonso, D. M.; Bond, J. Q.; Dumesic, J. A. Green Chem. 2010, 12, 1493−1513. (2) Zhang, J.; Zhang, B.; Zhang, J.; Lin, L.; Liu, S.; Ouyang, P. Biotechnol. Adv. 2010, 28, 613−619. (3) Hall, M.; Bansal, P.; Lee, J. H.; Realff, M. J.; Bommarius, A. S. FEBS J. 2010, 277, 1571−1582. (4) Zhao, X.; Zhang, L.; Liu, D. Biofuels, Bioprod. Biorefin. 2012, 6, 465−482. (5) Zhao, H.; Kwak, J. H.; Wang, Y.; Franz, J. A.; White, J. M.; Holladay, J. E. Energy Fuels 2006, 20, 807−811. (6) Zhao, H.; Kwak, J. H.; Conrad Zhang, Z.; Brown, H. M.; Arey, B. W.; Holladay, J. E. Carbohydr. Polym. 2007, 68, 235−241. (7) Rinaldi, R.; Schüth, F. ChemSusChem 2009, 2, 1096−1107. (8) Bobleter, O.; Schwald, W.; Concin, R.; Binder, H. J. Carbohydr. Chem. 1986, 5, 387−399. (9) Silveira, M. H. L.; Morais, A. R. C.; da Costa Lopes, A. M.; Olekszyszen, D. N.; Bogel-Łukasik, R.; Andreaus, J.; Pereira Ramos, L. ChemSusChem 2015, 8, 3366−3390. (10) Zhang, J.; Zhang, J.; Lin, L.; Chen, T.; Zhang, J.; Liu, S.; Li, Z.; Ouyang, P. Molecules 2009, 14, 5027−5041. (11) Davis, R.; Tao, L.; Scarlata, C.; Tan, E.; Ross, J.; Lukas, J.; Sexton, D. Process Design and Economics for the Conversion of Lignocellulosic Biomass to Hydrocarbons, Report No. NREL/TP-51001468

DOI: 10.1021/acscatal.7b04117 ACS Catal. 2018, 8, 1464−1468