5-HMF J CLEAN PRODUCTION 2019.pdf

Journal of Cleaner Production 213 (2019) 1096e1102 Contents lists available at ScienceDirect Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro Clean production of 5-hydroxymethylfurfural from cellulose using a hydrothermal/biomass-based carbon catalyst Qiong Wu a, Gaoyue Zhang a, Mingming Gao a, Shuangshuang Cao a, Lu Li a, Shiwei Liu a, Congxia Xie a, Lang Huang a, b, *, Shitao Yu a, **, Arthur J. Ragauskas c, *** a State Key Laboratory Base of Eco-chemical Engineering, College of Chemical Engineering, Qingdao University of Science and Technology, 53 Zhengzhou Road, Qingdao, Shandong province, 266042, PR China b Composite Material and Engineering Center, Washington State University, PACCAR Rm 112, 2001 East Grimes Way, Pullman, WA, USA, 99164 c Department of Chemical & Biomolecular Engineering, University of Tennessee Knoxville, 419 Doughterty Engineering Bldg, Knoxville, TN, USA, 37996 a r t i c l e i n f o a b s t r a c t Article history: Received 24 September 2018 Received in revised form 25 December 2018 Accepted 26 December 2018 Enhanced 5-hydroxymethylfurfural (5-HMF) production from cellulose was achieved using a novel and clean Ni-doped biomass-based carbon catalyst (Nin/CS) under hydrothermal degradation conditions. The obtained catalyst was characterized by ICP-AES, N2 adsorptionedesorption, TEM, XPS, XRD, FT-IR, NH3TPD, and pyridine IR. Abundant hydroxyl and carboxyl groups were present on the biomass-based carbon spheres (CSs), providing Lewis and Brønsted acid active sites. Furthermore, adding a Ni source increased the total acid content and the Brønsted acid strength. Ni mainly existed in its metallic form and was embedded in the porous carbon skeleton, providing active sites for cellulose adsorption and catalytic conversion. The effect of Ni dose, reaction time, and reaction conditions on 5-HMF production were investigated and optimized, acid catalyzed hydrolysis instead of hydrogenation reaction occurred when the parameters were optimized. A 5-HMF yield of 85% was obtained when cellulose (1 g) was reacted with H2 (6 MPa) at 200  C for 60 min using Ni2.0/CS (200 mg) as catalyst. © 2018 Elsevier Ltd. All rights reserved. Keywords: 5-HMF Cellulose Catalysis Hydrothermal carbonization 1. Introduction The efficient use of biomass to replace nonrenewable fossil fuel resources has become an important strategy for solving current global energy problems. Various bio-based carbon, membrane, fiber, and superhydrophobic materials have been prepared from biomass (Dai et al., 2018; Wu et al., 2015; Xu et al., 2018) and applied in catalysis, environmental science, energy production, and artificial intelligence (Gai et al., 2017; Herbert and Krishnan, 2016; Liu et al., 2018; Zhao et al., 2017). Furthermore, high value-added platform compounds, such as aldehydes and alcohols, can be obtained through pretreatment of biomass materials and subsequently converted into chemicals and transportable liquid fuels * Corresponding author.State Key Laboratory Base of Eco-chemical Engineering, College of Chemical Engineering, Qingdao University of Science and Technology, 53 Zhengzhou Road, Qingdao, Shandong province, 266042, PR China. ** Corresponding author. *** Corresponding author. E-mail addresses: [email protected] (L. Huang), [email protected] (S. Yu), [email protected] (A.J. Ragauskas). https://doi.org/10.1016/j.jclepro.2018.12.276 0959-6526/© 2018 Elsevier Ltd. All rights reserved. (such as gasoline, diesel, and activated coal) (Arni, 2017; CatalanMartinez et al., 2018). Bioethanol produced from biomass-based platform compounds is expected to replace conventional gasoline on a large scale (Sewsynker-Sukai and Gueguim Kana, 2018). Among these valuable platform compounds, 5hydroxymethylfurfural (5-HMF) is the one produced from the degradation of cellulose to levulinic acid (Rout et al., 2016). 5-HMF is an important intermediate for synthesis of the fine chemicals and furan polymers, and is an active ingredient in medicines and pesticides (Babaei et al., 2018; Li et al., 2017). 5-HMF has also been identified as a key intermediate in biomass chemistry, organic n-Leshkov et al., chemistry, and the petroleum industry (Roma 2006). Therefore, finding an effective and green route for preparing 5-HMF from biomass is of practical importance. When water at temperatures higher than 200  C is used as a reaction medium, reactants can undergo both thermal degradation and hydrolysis in a process known as hydrothermal degradation. High-temperature hydrothermal degradation has been widely used in waste polymer degradation, biomass conversion and utilization, and heavy oil conversion as an effective and efficient method for breaking down polymers (Kumar et al., 2018; Sasaki et al., 1998; Q. Wu et al. / Journal of Cleaner Production 213 (2019) 1096e1102 Suzuki et al., 1999). The hydrothermal degradation of cellulose has been investigated, with results indicating that, even without a catalyst, cellulose exhibits a high conversion rate that can be tuned by adjusting the operation temperature and pressure. However, this conversion process has some disadvantages, such as poor product selectivity. Acid catalysts have been used to enhance the conversion rate and yield of 5-HMF (Li et al., 2017; Mittal et al., 2017). Our previous study showed that biomass-based carbon materials obtained from hydrothermal carbonization contained large amounts of oxygencontaining surface groups and exhibited acidic properties (Wu et al., 2014). Adding a metal during the hydrothermal carbonization process can enhance the formation of mesoporous structures in the obtained carbon materials, which is conducive to catalytic reaction media transport, resulting in good catalytic activity (Wu et al., 2018a,b). Accordingly, employing metal-doped biomassbased carbon materials as catalysts for the catalytic degradation of biomass materials will be a meaningful development. In this study, a green route for converting biomass resources into platform compounds using a green catalyst was developed. Specifically, enhanced 5-HMF was obtained by cellulose degradation using Ni2.0/CS, a metal-doped biomass-based carbon material prepared by hydrothermal carbonization of glucose, as catalyst, under H2 at high temperature and pressure. A reaction mechanism was also proposed and analyzed. 1097 (FEI, Holland). X-ray photoelectron spectroscopy (XPS) was performed using a Physical Electronics spectrometer (PHI5700, Chanhassen, MN, USA), with binding energies referenced to the C1s line at 284.4 eV. C and O were quantitatively analyzed based on the peak intensities of the C1s and O1s signals. X-ray diffraction (XRD) measurements were performed using a Rigaku D/Max-rB diffractometer (Tokyo, Japan) with Cu Ka radiation in the scan range of 10e80 . Fourier-transform infrared spectroscopy (FT-IR) was performed using a Nicolet 510P spectrometer in the wavelength range Temperature-programmed desorption of 500e4000 cm 1. ammonia (NH3-TPD) was used to determine sample acidity (RS 232, Micromeritics). Pyridine IR spectra were obtained with a Nicolet 380 spectrometer (Thermo, USA) using a resolution of 4 cm 1 and 32 scans. 2.4. Catalytic hydrogenation of cellulose Catalytic conversion experiments were performed in a 150-mL stainless-steel reactor with catalyst (200 mg), microcrystalline cellulose (1 g), and deionized water (45 mL) as solvent. Reactions were conducted at 6 MPa of H2 and 200  C using different reaction times. Upon reaction completion, the reactor was cooled using ice water and products were collected by filtration for further analysis. All reactions were conducted in triplicate. For comparison, the reaction was also conducted under ambient pressure using N2 instead of H2. 2. Materials and methods 2.5. Product determination 2.1. Materials Glucose (Sigma-Aldrich; 99 wt%), nickel acetate (C4H6O4Ni$4H2O; Sigma-Aldrich; Ni content, >99 wt%), microcrystalline cellulose (Sigma-Aldrich), furfural (Sigma-Aldrich; 99%), and 5HMF (Sigma-Aldrich; 99%) were obtained and used directly without purification. 2.2. Catalyst preparation Glucose (3.0 g) and nickel acetate (mass ratios: 10:1, 5:1, 5:2, 5:3, 5:4, and 1:1) were dissolved in distilled water (80 mL) and ultrasonicated for 30 min. The obtained solution was transferred to a 100-mL autoclave and placed in a homogeneous reactor at 230  C for 8 h. After cooling, the solid products were separated, collected by centrifugation, washed with pure ethanol and distilled water until the solution was clear, and dried at 80  C overnight for further use. The resulting black solid was placed in a furnace under a nitrogen atmosphere and heated to 900  C at a heating rate of 5  C/ min, held at 900  C for 2 h, and then allowed to cool to room temperature. The obtained black product was denoted as Nin/CS, where n is one-tenth of the percentage of Ni added during catalyst preparation. Catalyst without added Ni, denoted as CSs, was also prepared for comparison. 2.3. Catalyst characterization The doped metal content of Nin/CS was elucidated by inductively coupled plasma-atomic emission spectroscopy (ICP-AES) using a forward power of 1200 W (Varian Vista Pro ICP-AES instrument). N2 adsorptionedesorption isotherms were obtained using an ASAP 2020 instrument to characterize the porous structure (Micromeritics, USA), with the specific surface area calculated from adsorption isotherms using the BrunauereEmmetteTeller (BET) equation. The pore morphologies were observed by transmission electron microscopy (TEM) using a JEOL 2011 microscope The obtained products were quantitatively analyzed by HPLC equipped with a C18 column and a UV detector at 284 nm. Methanol/water (3:2, v/v) was selected as the mobile phase with a flow rate of 0.5 mL/min. The column temperature was 30  C. Before analysis, all degradation products were filtered through a 0.2-mm membrane filter. The yields of 5-HMF and furfural were calculated using the following equations: 5 HMF yield ðmol%Þ ¼ Furfural yield ðmol%Þ ¼ moles of 5 HMF produced *100 moles of starting cellulose moles of furfural produced *100 moles of starting cellulose (1) (2) 3. Results and discussion 3.1. Catalyst characterization Table 1 summarizes the textural parameters of the CSs and Ni2.0/ CS catalysts. After high-temperature carbonization, the specific surface area of CSs was 462 m2/g with a pore volume of 0.24 cm3/g. When doped with Ni, the specific surface area was slightly decreased to 427 m2/g, mainly due to Ni addition being beneficial for the formation of mesoporous structures, as previously reported (Feng et al., 2016). When the theoretical additive amount was 20%, the actual Ni content doped in carbon materials was only 10.02%, as calculated by ICP-AES, indicating that a large amount of Ni was lost during preparation. After carbonization, the remaining Ni was present in a relatively stable form. TEM images of Ni2.0/CS are shown in Fig. 1 (a, b). The samples were uniformly distributed spheres with diameters of 200e400 nm and showed some adhesion. In the HRTEM images (Fig. 1(c) and (d)), an obvious ordered striated lattice structure was observed, 1098 Q. Wu et al. / Journal of Cleaner Production 213 (2019) 1096e1102 Table 1 Textural parameters of catalysts. Samples Metal contenta/wt% Metal contentb/wt% SBET(m2/g) Pore Volume(cm3/g) CSs Ni2.0/CS e 20 e 10.02 462 427 0.24 0.23 a b Metal content in the preparation process. Metal content calculated by ICP-AES. Fig. 1. TEM and HRTEM images of Ni2.0/CS. with lattice spaces measured as 0.20 nm and 0.34 nm, corresponding to metallic Ni and graphite-like carbon structures, respectively. This confirmed the successful doping of Ni into the carbon skeleton (Feng et al., 2016; Wu et al., 2015). Three obvious peaks were observed in the XPS curve of Ni2.0/CS (Fig. 2(a)), corresponding to characteristic C, O, and Ni peaks, which Fig. 2. (a,b) XPS, (c) XRD, and (d) FT-IR spectra of Ni2.0/CS and CSs. further confirmed successful Ni doping. High-resolution Ni XPS of Ni2.0/CS was also performed, with the corresponding fitting curves shown in Fig. 2(b). The broad peak centered in the range 850e885 eV was characteristic of Ni doping. The main fitting peaks were obtained at 852.5 eV and 870.4 eV, which were attributed to Ni 2p3/2 and Ni 2p1/2, respectively, indicating that Ni mainly existed in the metallic state. Meanwhile, a small peak centered at 854.9 eV, and satellite peaks at 860.1 and 880.8 eV, showed the presence of some Ni2þ cations (Naushad et al., 2017), confirming that Ni was present in two forms. XRD analysis was performed to further verify the Ni valence state. An obvious peak was observed centered at 25.2 , characteristic of graphitic carbon, which was consistent with the TEM results above. Three peaks located at 44.7, 52.1, and 77.2 were characteristic of face-centered cubic Ni. An additional small diffraction maximum centered at 43.4 was indexed to Ni oxide phases (Zubizarreta et al., 2009). This further confirmed that Ni in Ni2.0/CS existed as mainly metallic Ni with a small amount of nickel oxide. From the FT-IR spectra, both CSs and Ni2.0/CS samples obtained in this study exhibited abundant C¼C bonds, indicating that these samples contained unsaturated functional groups (mainly unsaturated aromatic structures) (Sun and Li, 2004; Titirici et al., 2008). These samples were also rich in oxygen-containing groups, such as hydroxyl, carboxyl, and carbonyl groups, on their surfaces (Liu et al., 2010). Notably, the characteristic peak of carboxyl groups (~1700 cm 1) was much larger for Ni2.0/CS than for CSs, indicating a larger carboxyl group content on the surface after Ni was added. FT-IR showed that the Ni2.0/CS surface was rich in oxygencontaining groups, especially large amounts of carboxyl groups, after Ni doping. Therefore, NH3-TPD (Fig. 3a) was used to evaluate the acidity and acid strength of both CSs and Ni2.0/CS. The peak centered below 100  C was attributed to physically adsorbed ammonia, the peak centered at 220e520  C was attributed to medium-strength acid centers, and those centered above 520  C were attributed to strong acid centers (Luo et al., 2018). For CSs, an obvious broad peak was centered at 220e520  C, implying medium-strength acid centers were present on the surface, derived from hydroxyl and carboxyl groups. When Ni was doped into the material, a sharp peak centered at 350  C appeared, which might be due to the obvious increase of carboxyl groups. This higher acid content could be beneficial for acid-catalyzed reactions. Fig. 3. (a) NH3-TPD and (b) pyridine IR profiles of Ni/C2.0 and CSs. Q. Wu et al. / Journal of Cleaner Production 213 (2019) 1096e1102 1099 The desorption curves from pyridine IR at 200  C are shown in Fig. 3(b), with the peaks at 1450 cm 1 and 1540 cm 1 characteristic of Lewis and Brønsted acids, respectively (Barzetti et al., 1996). CSs had an obvious Lewis acid peak and high Lewis acid content, reaching 150.91 mmol/g, with a total acid content of 166.01 mmol/g. CSs mainly consisted of CeC double bonds with abundant hydroxyl groups on the surface, which can form hydrogen bonds that exhibit electrophilicity, resulting in Lewis acidity (Ryczkowski, 2001). Adding Ni resulted in enhanced intensity of the characteristic peak at 1540 cm 1 in Ni2.0/CS, with the Brønsted acid content increasing to 72.20 mmol/g. In contrast, the Lewis acid content decreased, but the total acid content still increased overall, reaching 201.73 mmol/ g. Combined with the FT-IR analysis above, Ni doping caused an increase in carboxyl functional groups on the Ni2.0/CS surface, which provided additional Brønsted acid active sites, leading to an increased acid content (Table 2). 3.2. Catalyst analysis Fig. 4. HPLC chromatographs of (a) standard chemicals and (b) representative reactions. The catalytic reaction conditions are listed in detail in section 2.4. Under these condition, standard samples of 5-HMF and furfural were first injected for HPLC analysis, with the characteristic peaks of 5-HMF and furfural appearing at 6.1 min and 9.07 min, respectively (Fig. 4a). When Ni2.0/CS was used as catalyst for cellulose degradation, two obvious peaks appeared at the same position (Fig. 4b), confirming that the main degradation products were 5HMF and furfural. 3.2.1. Effect of Ni dose on catalysis Carbon materials with different Ni doses were selected as catalysts to investigate the effect of Ni content on the catalytic activity. From the results shown in Fig. 5, as the Ni/C ratio was increased from 1:10 to 1:1 (theoretical value), the yield of 5-HMF initially increased and then decreased. When Ni2.0/CS was used as catalyst, the yield of 5-HMF reached the highest value of 85%, which is higher than in many previous reports (Li et al., 2017, 2018; Xia et al., 2016). This result was mainly due to the occurrence of acidcatalyzed hydrolysis. With a lower Ni loading, the catalyst acidity was weak with fewer catalytic active sites. When the Ni loading was too high, it had a negative effect on the specific surface area of the catalyst, leading to a decrease in mesoporous structures (Wu et al., 2018a,b), which provide catalytic adsorption site and promote transportation of the reaction medium, and resulting in a decreased 5-HMF yield. The yield of furfural exhibited the same trend, but all yields were lower than 10%. The highest furfural yield was around 10% when using Ni4.0/CS as catalyst. According to these results, Ni2.0/CS was selected as the catalyst to achieve the highest yield of 5-HMF. 3.2.2. Effect of reaction time on catalysis Ni2.0/CS was selected as catalyst to further investigate the effect of reaction time on 5-HMF and furfural yields (see Fig. 6). When the reaction time was 30 min, the yield of 5-HMF was 76%. Increasing the reaction time initially increased the 5-HMF yield, eventually resulting in the highest obtained yield of 85%, while further extending the reaction time resulted in a decrease in the 5-HMF yield. Meanwhile, the yield of furfural showed a sustained downward trend with increasing reaction time. The results showed that Fig. 5. Effect of Ni/C ratio on 5-HMF and furfural production. Reaction conditions: Cellulose (1 g), catalyst (0.2 g), H2 (6 MPa), 200  C, 60 min. 60 min was the optimal reaction time, with longer reaction times promoting further reaction and leading to byproduct generation. However, as the catalysts prepared in this study had mediumstrength acid sites and lower acidity, subsequent reactions were difficult to follow or only proceeded in small quantities. 3.2.3. Effect of reaction conditions on catalysis The results above confirmed that cellulose undergoes acidcatalyzed hydrolysis rather than hydrogenation. The effects of different catalytic environments on the catalytic results were further investigated as shown in Fig. 7. Using the reaction conditions of H2 (6 MPa), 200  C, and 60 min, with cellulose (1 g) as reaction substrate and without catalyst, the yields of 5-HMF and furfural were 33% and 3%, respectively, indicating that the high reaction temperature and pressure alone could promote the reaction. The high reaction temperature effectively promoted cellulose Table 2 Catalyst acidity parameters. Samples Brønsted acid content/umol/g Lewis acid content/umol/g Total acid content/umol/g B/L (%) CSs Ni2.0/CS 15.11 72.20 150.91 129.53 166.01 201.73 0.10 0.56 1100 Q. Wu et al. / Journal of Cleaner Production 213 (2019) 1096e1102 Fig. 6. Effect of reaction time on 5-HMF and furfural production. Reaction conditions: Cellulose (1 g), Ni2.0/CS as catalyst (0.2 g), H2 (6 MPa), 200  C. obtained in this study, the main reaction was acid hydrolysis. The metallic Ni and porous carbon carrier provide active sites for cellulose adsorption and catalytic conversion (Gai et al., 2017). In an acidic environment, cellulose is first hydrolyzed into glucose, then isomerizes to form fructose, mainly involving Lewis acid sites (Zhang et al., 2017), under high temperature and pressure condition. Fructose further dehydrates to produce 5-HMF, which is mainly dependent on the catalytic effect of Brønsted acid sites (Shaikh et al., 2018), and a small amount of glucose can directly dehydrate to produce 5-HMF (Sevilla and Fuertes, 2009). As the reaction proceeds, some 5-HMF generates furfural via decarbonization. As the reaction time increases, some humins and byproducts are produced, while as the catalyst belongs to mediumstrength acidity, the consequence reaction is difficult for further going. The resulting Ni0.2/CS composite showed excellent catalytic activity due to strong interactions between Ni metal and the carbon support (Gai et al., 2017). As a result, the catalytic degradation products mainly consisted of 5-HMF and a small part of furfural. Nin/CS have magnetic properties, they can be recovered by external magnet after reaction. After each run, the recovered catalysts were washed with deionized water and ethanol repeatedly, dried at 80  C before next run. Fig. 9 showed the reusability of Ni2.0/ CS catalyst, they exhibited good reusability performance during the catalytic reaction, which can be attributed to the strong support of Ni nanoparticles from the carbon matrix and the high recovery rate of Ni2.0/CS during magnetic field recyclization. 4. Conclusion Fig. 7. Effect of reaction conditions on 5-HMF and furfural production. dissolution, while the high pressure reduced the reaction energy barrier. When using CSs as catalyst, the yields of both 5-HMF and furfural increased, reaching 58% and 11%, respectively, which were higher than those achieved using some traditional catalysts (Li et al., 2018; Liu et al., 2013), mainly due to the acidity, especially Lewis acidity, on the CSs surface promoting cellulose degradation and acid hydrolysis. Furthermore, when Ni2.0/CS was employed as catalyst using a N2 atmosphere (6 MPa) instead of a H2 atmosphere, the yields significantly decreased, to even lower than for the reaction without catalyst. This directly showed that H2 was important in the reaction, mainly for keeping the doped Ni in an active state (Liang et al., 2014). Under H2 at atmospheric pressure, the yields were also not ideal, further confirming that high pressure played an important role in this reaction process. Based on the above analysis, a possible reaction pathway for cellulose degradation was proposed (Fig. 8). High-temperature water is a good reaction medium that might participate as an active reactant and facilitate the reaction as a catalyst. High temperatures reduce the dielectric constant of water and increase the cellulose solubility (Cantero et al., 2015). As hydroxyl and carboxyl groups were present in the catalyst A Ni-doped carbon catalyst (Nin/CS) with Brønsted and Lewis acid active sites was prepared using glucose as the carbon resource. Ni2.0/CS was found to exhibit high catalytic activity for the conversion of cellulose to 5-HMF owing to acid active sites and interactions between Ni metal and the carbon support, affording an enhanced 5-HMF yield of 85% using with H2 (6 MPa) at 200  C for 60 min. The high reaction temperature and high pressure promoted cellulose dissolution and reduced the reaction energy barrier. The metallic Ni and porous carbon carrier were able to provide active sites for cellulose adsorption and catalytic conversion. Altogether, this study provides an efficient route for cellulose conversion using a biomass-based carbon catalyst and provides a new approach to the high value utilization of biomass resources. Future work will be conducted on exploring an optimized route to maximize the yield of 5-HMF production and obtain a deep understanding of the catalytic reaction mechanism on a molecular scale. Fig. 8. Possible reaction pathway for cellulose conversion to 5-HMF and furfural. Q. Wu et al. / Journal of Cleaner Production 213 (2019) 1096e1102 Fig. 9. The reusability of Ni2.0/CS catalyst. Reaction conditions: Cellulose (1 g), catalyst (0.2 g), H2 (6 MPa), 200  C, 60 min. Acknowledgements The National Natural Science Foundation of China (31700517), Shandong Provincial Natural Science Foundation, China (ZR2017BC100), Qingdao Applied Basic Research Program (18-2-24-jch), Special Grant from China Postdoctoral Fund (2018T110667), Key research and development project of Shandong Province (2017GGX70102, 2017GGX40106) and the Taishan Scholars Program of Shandong Province (ts201511033) financially supported this work. We thank Simon Partridge, PhD, from Liwen Bianji, Edanz Editing China (www.liwenbianji.cn/ac), for editing the English text of a draft of this manuscript. References Arni, S.A., 2017. Comparison of slow and fast pyrolysis for converting biomass into fuel. Renew. Energy 124, 197e201. https://doi.org/10.1016/j.renene.2017.0 4.060. Babaei, Z., Najafi Chermahini, A., Dinari, M., Saraji, M., Shahvar, A., 2018. Cleaner production of 5-hydroxymethylfurfural from fructose using ultrasonic propagation. J. Clean. Prod. 198, 381e388. https://doi.org/10.101 6/j.jclepro.2018.06. 305. Barzetti, T., Selli, E., Moscotti, D., Forni, L., 1996. Pyridine and ammonia as probes for FTIR analysis of solid acid catalysts. J. Chem. Soc. Faraday Trans. 92, 1401e1407. https://doi.org/10.1039/FT9969201401.  Cantero, D.A., S anchez Tapia, Angel, Bermejo, M.D., Cocero, M.J., 2015. Pressure and temperature effect on cellulose hydrolysis in pressurized water. Chem. Eng. J. 276, 145e154. https://doi.org/10.1016/j.cej.2015.04.076. Catalan-Martinez, D., Domine, M.E., Serra, J.M., 2018. Liquid fuels from biomass: An energy self-sustained process integrating H2 recovery and liquid refining. Fuel 212, 353e363. https://doi.org/10.1016/j.fuel.2017.10.014. Dai, J.D., Zhang, R.L., Ge, W.N., Xie, A.T., Chang, Z.S., Tian, S.J., Zhou, Z.P., Yan, Y.S., 2018. 3D macroscopic superhydrophobic magnetic porous carbon aerogel converted from biorenewable popcorn for selective oil-water separation. Mater. Des. 139, 122e131. https://doi.org/10.1016/j.matdes.2017.11.001. Feng, J., Wang, Q., Fan, D.L., Ma, L.R., Jiang, D.L., Xie, J.M., Zhu, J.J., 2016. Nickel-based xerogel catalysts: synthesis via fast sol-gel method and application in catalytic hydrogenation of p-nitrophenol to p-aminophenol. Appl. Surf. Sci. 382, 1101 135e143. https://doi.org/10.1016/j.apsusc.2016.04.125. Gai, C., Zhang, F., Guo, Y.C., Peng, N.N., Liu, T.T., Lang, Q.Q., Xia, Y., Liu, Z.G., 2017. Hydrochar-supported, in situ-generated nickel nanoparticles for sorptionenhanced catalytic gasification of sewage sludge. ACS Sustain. Chem. Eng. 5, 7613e7622. https://doi.org/10.1021/acssuschemeng.7b00924. Herbert, G.M.J., Krishnan, A.U., 2016. Quantifying environmental performance of biomass energy. Renew. Sustain. Energy Rev. 59, 292e308. https://doi.org/10. 1016/j.rser.2015.12.254. Kumar, M., Oyedun, A.O., Kumar, A., 2018. A review on the current status of various hydrothermal technologies on biomass feedstock. Renew. Sustain. Energy Rev. 81, 1742e1770. https://doi.org/10.1016/j.rser.2017.05.270. Li, W.Z., Zhu, Y.S., Lu, Y.J., Liu, Q.Y., Guan, S.N., Chang, H.M., Jameel, H., Ma, L.L., 2017. Enhanced furfural production from raw corn stover employing a novel heterogeneous acid catalyst. Bioresour. Technol. 245, 258e265. https://doi.org/10. 1016/j.biortech.2017.08.077. Li, X.C., Peng, K.H., Xia, Q.N., Liu, X.H., Wang, Y.Q., 2018. Efficient conversion of cellulose into 5-hydroxymethylfurfural over niobia/carbon composites. Chem. Eng. J. 332, 528e536. https://doi.org/10.1016/j.cej.2017.06.105. Liang, G.F., He, L.M., Cheng, H.Y., Li, W., Li, X.R., Zhang, C., Yu, Y.C., Zhao, F.Y., 2014. The hydrogenation/dehydrogenation activity of supported Ni catalysts and their effect on hexitols selectivity in hydrolytic hydrogenation of cellulose. J. Catal. 309 (1), 468e476. https://doi.org/10.1016/j.jcat.2013.10.022. Liu, B., Zhang, Z.H., Zhao, Z.B.K., 2013. Microwave-assisted catalytic conversion of cellulose into 5-hydroxymethylfurfural in ionic liquids. Chem. Eng. J. 215 (3), 517e521. https://doi.org/10.1016/j.cej.2012.11.019. Liu, L., Yang, X., Hou, Q.D., Zhang, S.Q., Ju, M.T., 2018. Corn stalk conversion into 5hydroxymethylfurfural by modified biochar catalysis in a multi-functional solvent. J. Clean. Prod. 187, 380e389. https://doi.org/10.1016/j.jclepro.2018.03.234. Liu, S.X., Sun, J., Huang, Z.H., 2010. Carbon spheres/activated carbon composite materials with high CR (VI) adsorption capacity prepared by a hydrothermal method. J. Hazard Mater. 173 (1), 377e383. https://doi.org/10.101 6/j.jhazmat. 2009.08.086. Luo, J.Y., Kamasamudram, K., Currier, N., Yezerets, A., 2018. NH3-TPD methodology for quantifying hydrothermal aging of Cu/SSZ-13 SCR catalysts. Chem. Eng. Sci. 190, 60e67. https://doi.org/10.1016/j.ces.2018.06.015. Mittal, A., Black, S.K., Vinzant, T.B., O'Brien, M., Tucker, M.P., Johnson, D.K., 2017. Production of furfural from process-relevant biomass-derived pentoses in a biphasic reaction system. ACS Sustain. Chem. Eng. 5, 5694e5701. https:// doi.org/10.1021/acssuschemeng.7b00215. Naushad, M., Ahamad, T., Al-Maswari, B.M., Alqadami, A.A., Alshehri, S.M., 2017. Nickel ferrite bearing nitrogen-doped mesoporous carbon as efficient adsorbent for the removal of highly toxic metal ion from aqueous medium. Chem. Eng. J. 330, 1351e1360. https://doi.org/10.1016/j.cej.2017.08.079. n-Leshkov, Y., Chheda, J.N., Dumesic, J.A., 2006. Phase modifiers promote Roma efficient production of hydroxymethylfurfural from fructose. Science 312 (5782), 1933e1937. https://doi.org/10.1126/science.1126337?. Rout, P.K., Nannaware, A.D., Prakash, O., Kalra, A., Rajasekharan, R., 2016. Synthesis of hydroxymethylfurfural from cellulose using green processes: a promising biochemical and biofuel feedstock. Chem. Eng. Sci. 142, 318e346. https://doi. org/10.1016/j.ces.2015.12.002. Ryczkowski, J., 2001. IR spectroscopy in catalysis. Catal. Today 68, 263e381. https:// doi.org/10.1016/S0920-5861(01)00334-0. Sasaki, M., Kabyemela, B., Malaluan, R., Hirose, S., Takeda, N., Adschiri, T., Arai, K., 1998. Cellulose hydrolysis in subcritical and supercritical water. J. Supercrit. Fluids 13, 261e268. https://doi.org/10.1016/S0896-8446(98)00060-6. Sevilla, M., Fuertes, A.B., 2009. The production of carbon materials by hydrothermal carbonization of cellulose. Carbon 47, 2281e2289. https://doi.org/10.1016/j.ca rbon.2009.04.026. Sewsynker-Sukai, Y., Gueguim Kana, E.B., 2018. Simultaneous saccharification and bioethanol production from corn cobs: process optimization and kinetic studies. Bioresour. Technol. 262, 32e41. https://doi.org/10.1016/j.biortech.2 018. 04.056. Shaikh, M., Singh, S.K., Khilari, S., Sahu, M., Ranganath, K.V.S., 2018. Graphene oxide as a sustainable metal and solvent free catalyst for dehydration of fructose to 5HMF: a new and green protocol. Catal. Commun. 106, 64e67. https://doi.org/10. 1016/j.catcom.2017.12.018. Sun, X.M., Li, Y.D., 2004. Colloidal carbon spheres and their core/shell structures with noble-metal nanoparticles. Angew. Chem. Int. Ed. 43, 597e601. https://doi. org/10.1002/ange.200352386. Suzuki, Y., Tagaya, H., Asou, T., Kadokawa, J., Chiba, K., 1999. Decomposition of prepolymers and molding materials of phenol resin in subcritical and supercritical water under an Ar atmosphere. Ind. Eng. Chem. Res. 38, 1391e1395. https://doi.org/10.1021/ie9805842. Titirici, M.M., Antonietti, M., Baccile, N., 2008. Hydrothermal carbon from biomass: a comparison of the local structure from poly-to monosaccharides and pentoses/hexoses. Green Chem. 10, 1204e1212. https://doi.org/10.1039/b807009a. Wu, Q., Liu, S.W., Xie, C.X., Yu, H.L., Liu, Y., Yu, S.T., Huang, L., 2018a. Ni-doped mesoporous carbon obtained from hydrothermal carbonization of cellulose and their catalytic hydrogenation activity study. J. Mater. Sci. 53, 7900e7910. https://doi.org/10.1007/s10853-018-2054-5. Wu, Q., Li, W., Liu, S.X., 2014. Carboxyl-rich carbon microspheres prepared from pentosan with high adsorption capacity for heavy metal ions. Mater. Res. Bull. 60, 516e523. https://doi.org/10.1016/j.materresbull.2014.09.015. Wu, Q., Li, W., Tan, J., Wu, Y.J., Liu, S.X., 2015. Hydrothermal carbonization of carboxymethylcellulose: one-pot preparation of conductive carbon 1102 Q. Wu et al. / Journal of Cleaner Production 213 (2019) 1096e1102 microspheres and water-soluble fluorescent carbon nanodots. Chem. Eng. J. 266, 112e120. https://doi.org/10.1016/j.cej.2014.12.089. Wu, Q., Zhang, G.Y., Ma, G.Q., Liu, S.W., Xie, C.X., Yu, H.L., Liu, Y., Huang, L., Yu, S.T., 2018b. Carbon composite materials with ordered mesoporous structures from straw: hydrothermal preparation and application as catalysts. Nanotechnology 29, 385604e385612. https://doi.org/10.1088/1361-6528/aacf57. Xia, H.A., Xu, S.Q., Yan, X.P., Zuo, S.L., 2016. High yield synthesis of 5hydroxymethylfurfural from cellulose using FePO4, as the catalyst. Fuel Process. Technol. 152, 140e146. In: https://doi.org/10.1016/j.fuproc.2016.06.030. Xu, M.C., Li, W., Ma, C.H., Yu, H.P., Wu, Y.Q., Wang, Y.G., Chen, Z.J., Li, J., Liu, S.X., 2018. Multifunctional chiral nematic cellulose nanocrystals/glycerol structural colored nanocomposites for intelligent responsive films, photonic inks and iridescent coatings. J. Mater. Chem. C 6, 5391e5400. https://doi.org/10.1039/ C8TC01321G. Zhang, L.X., Xi, G.Y., Chen, Z., Qi, Z.Y., Wang, X.C., 2017. Enhanced formation of 5HMF from glucose using a highly selective and stable SAPO-34 catalyst. Chem. Eng. J. 307, 877e883. https://doi.org/10.1016/j.c ej.2016.09.003. Zhao, D.W., Chen, C.J., Zhang, Q., Chen, W.S., Liu, S.X., Wang, Q.W., Liu, Y.X., Li, J., Yu, H.P., 2017. High performance, flexible, solid-state supercapacitors based on a renewable and biodegradable mesoporous cellulose membrane. Advanced Energy Materials 7 (18), 1700739e1700747. https://doi.org/10.1002/aenm.20170 0739. ndez, J.A., Pis, J.J., Arenillas, A., 2009. Improving hydrogen Zubizarreta, L., Mene storage in Ni-doped carbon nanospheres. Int. J. Hydrogen Energy 34 (7), 3070e3076. https://doi.org/10.1016/j.ijhydene.2009.01.040.