As regenerated feedstock, biomass has attracted much attention recently for conversion to chemicals or liquid fuels in replacement of oil and coal in the near future [1-4]. Generally, cellulose, hemicellulose, and lignin are major contents of plant biomass. The key points of the conversion of biomass include the degradation of cellulose into top valuable platform molecules with multi-functional groups, such as lactic acid, formic acid, and oxalic acid, etc. [5-7]. Among them, formic acid has attracted much attention due to its potential application in various fields, such as chemical, agricultural, pharmaceutical, textile, and rubber industries . In addition, formic acid has also been recognized as a source of hydrogen, which makes it potential applications in the hydrogeneration of oxygen-containing compounds to alkanes, and in fuel cells to efficiently generate electricity .
The conversion of cellulose or monosaccharide into formic acid is usually carried out by oxidation of relative carbohydrates under catalysis. The reported catalysts for efficient conversion of carbohydrates to formic acid include alkali hydroxide with hydrogen peroxide as the oxidant , Ru nanoparticles on reduced graphite oxide [11, 12], vanadium-substituted phosphomolybdic acids with molecular oxygen as the oxidant , polyoxometalate in aqueous phase or water-organic biphase with oxygen or air as the oxidant , and NaVO3/H2SO4 aqueous solution with molecular oxygen as the oxidant. However, the processes are usually heterogeneous, and cellulose powder are directly used during the process. The degradation of cellulose aggregations is usually step-by-step from the surface to the inner, resulting in a slow degradation rate. Dissolving of cellulose in a solution will be helpful for the degradation since the cellulose chains can be accessed by the active sites of catalyst, especially when solid catalyst is utilized. So the degradation of cellulose in its green solvent is especially important for the conversion of biomass .
It is well known that cellulose is difficult to be dissolved in a simple solvent for the existence of crystalline structure of complex hydrogen-bonding network. Recently, two kinds of green solvents for cellulose has been reported, i.e. room temperature ionic liquids and alkaline aqueous solution [16-19]. Plenty of researches were reported using ionic liquid as solvent to convert cellulose in its homogeneous solution over solid or liquid catalysts, such as conversion of cellulose to 5-hydroxymethylfurfural (HMF) or lactic acid etc. [20-22]. However, the high-cost of ionic liquids and recovery and reutilization of ionic liquid limit their application in large scale by now. In addition, the bio-toxicity of ionic liquids in the long term is still a problem that has not been resolved. On the other hand, alkaline aqueous solution such as NaOH and urea/PEG/thiourea are also the green solvents for cellulose, and it is easy to form homogeneously cellulose aqueous solution by a freezing-thawing process . So it provides a chance to degrade cellulose in the homogeneously aqueous solution, and our previous study revealed that cellulose can be efficiently degraded to lactic acid and malonic acid in its homogeneously aqueous solution in the absence of any catalysts . Recently Jiang et al. found that oxalic acid can be efficiently synthesized from the homogeneously aqueous solution of cellulose under the catalysis of metal oxides, especially CuO . All the above reports revealed that the alkaline aqueous solution of cellulose can be used as the homogeneous system for the efficient degradation of cellulose over proper catalysts.
Catalytic conversion of biomass with metal-organic frameworks (MOFs) as catalysts was investigated owning to their high surface area, tunable pore sizes and controllable structures [25, 26]. However, the size of MOF usual in nanometers or micrometers scale, and the recovery of the catalyst from the reactive system are still difficult. Fixation of MOF on a solid support, especially on a 3D porous support is highly attracting. Graphene, a famous single or a few layers of carbon, has been paid much attention recently for a wide application by fabricating various kinds of architectures, including catalysis. So the combination of 3D graphene architectures with MOF nanoparticles will be fascinating for catalysis, especially as catalysts for biomass conversion. Recently, some nanocomposites of MOF and graphene has been prepared, and efficient catalytic activity was reported [27-29]. However, there is still no report on the preparation of 3D MOF/graphene architecture for biomass catalytic conversion. Zeolitic imidazolate framework (such as ZIF-8) has been reported an efficient catalyst for photocatalysis or carbon dioxide conversion because the external surface of ZIF-8 has not only Lewis acid and basic sites, but also Br
In this work, a new 3D MOF/graphene catalyst (3D ZIF-8/graphene hydrogel) has been synthesized, and it was used for the catalytic conversion of cellulose in its alkaline homogeneously aqueous solution for formic acid, and high conversion and selectivity can be obtained.Ⅱ. EXPERIMENTS A. Materials
Chemicals including analytical standard formic acid, oxalic acid, acetic acid, analytical grade methyl alcohol, sodium hydroxide, urea, zinc nitrate hexahydrate (ZnNO3·6H2O, 98 %) and 2-methylimidazole (MeIM, 98 %) were all purchased from Sinopharm Chemical Reagent Co. Ltd. Microcrystalline cellulose powder (
Graphene oxide (GO) was synthesized by a modified Hummer's method . 0.18 g GO powder was dissolved in 30 mL water under the assistance of strong ultrasound to get GO solution, and then 0.55 g ascorbic acid and 0.13 g NaOH were added under mild ultrasonic for 30 min, and the solution was heated in a test tube in water bath at 70 ℃ for 1 h to get a graphene hydrogel. 4.38 g of zinc nitrate hexahydrate and 5.1 g of 2-methylimidazole dispersed in 50 mL methyl alcohol were added respectively, and the system retained for 24 h for the diffusion of the reactants inside the network of rGO hydrogel, and during the process ZIF-8 MOF nanoparticles could be formed therein. After reaction, the product was washed using methyl alcohol and distilled water for three times, and the dried hydrogel of 3D ZIF-8/rGO can be obtained after it was freeze dried (Fig. 1).C. Characterization
Wide-angle X-ray diffraction (XRD) analyses were carried out on an X-ray diffractometer (D/MAX-1200, Rigaku Denki Co. Ltd., Japan). X-ray photoelectron spectroscopy (XPS) was recorded on an Escalab MK II photoelectron spectrometer (VG Scientific Ltd., United Kingdom). Fourier transform infrared spectra of the samples were recorded by a Bruker vector-2 spectrophotometer (German) using KBr-disk method. The morphologies of the hydrogel were characterized by SEM (Superscan SSX-550, Shimazu). TEM images were carried out on a Hitachi H-800 microscope at 200 kV. The thermal properties of the samples were recorded by a thermogravimeter (TGA, DTA-50, Shimazu), and all of the measurements were performed in air over a temperature range of 30-800 ℃ with a ramp rate of 5 ℃/min. Nitrogen sorption measurement was conducted using a Micromeritics ASAP 2020 system at 77 K. The catalysts were investigated by NH3-TPD (temperature programmed desorption of ammonia), the catalysts were at 573 K under helium flow (ultrahigh purity, 100 mL/min) for 2 h, and adsorption of ammonia was carried out at 393 K for 1 h. After the catalysts were flushed with He at 393 K for 1 h, the programmed desorption of NH3 was run from 393 K to 973 K with a heating rate of 10 K/min. The desorbed ammonia was measured by a gas chromatograph (GC-SP6890, Shandong Lunan Ruihong Chemical Instrument Co., Ltd., Tengzhou China) with a thermal conductivity detector.D. Cellulose aqueous solution preparation
3 g NaOH was dissolved in 45 mL deionized water, and 2 g microcrystalline cellulose was then added very slowly at 273 K under vigorous stirring, 15 min later, the mixture was cooled down to 243 K in a refrigerator for 4 h, and it was thawed under strong stirring at room temperature. At the end the homogeneously aqueous cellulose solution with concentration of about 0.04 g/mL was obtained . When the mixture of cellulose and NaOH aqueous solution was thawed twice, a clear solution was obtained, which can stable at least 6 h. Urea is always used as a stabilizer to stabilize the solution for a longer time, and cellulose can also be dissolved in the alkaline aqueous solution even in the absence of urea when the pro-cooling temperature is -32 ℃.E. Catalysis activity test
Conversions of cellulose in its homogeneously aqueous solution over the as-prepared catalysts were performed in a 100 mL stainless steel pressure vessel. Typically, 8 mL cellulose aqueous solution and 0.15 g dried ZIF-8/rGO hydrogel catalyst were added into the pressure vessel, and then the reactor was heated up to 323-573 K at a rate of 5 K/min under vigorous stirring. After a desired period of time, the reaction was quickly terminated by cooling the reactor to room temperature, followed by neutralizing the product by 1 mol/L hydrochloric acid aqueous. The samples were then analyzed by a high-performance liquid chromatography (HPLC, LC-20AD, Shimadzu) on a C-18-column equipped with an UV-Vis detector, and the retention time for oxalic acid, formic acid, and acetic acid are 2.9, 3.3, and 4.3 min (by the standard), respectively.Ⅲ. RESULTS AND DISCUSSION
In our previous work, rGO hydrogel can be easily prepared by the in situ reduction and self-assembly of GO nanosheets by various chemical reducing agents [30-32]. Here, the rGO hydrogel was prepared by the same method at first, and the microstructure of the dried hydrogel can be measured by means of SEM, as shown in Fig. 2(a). The hydrogel is porous with plenty of inter-connective pore in size of micrometers, which makes the feedstocks of ZIF-8 easily enter the inside of the network and in situ growth therein. Figure 2(b) shows the SEM image of as-prepared dried 3D ZIF-8/rGO hydrogel, and clearly there formed plenty of ZIF-8 nanocrystalline inside the network, and Fig. 2(c) shows that the size of the ZIF-8 crystalline is in size of hundreds of nanometers. Figure 2(d) shows the TEM image of the nanocomposite at the edge, and clearly the nanocrystalline of ZIF-8 can be found in size of about 500-800 nm, on the surface of the single layer graphene nanosheets.
Figure 3(a) shows the TGA curves of the dried rGO and ZIF-8/rGO hydrogels. For the dried rGO hydrogel, it is thermally stable below 723 K, and the mass loss is about 20 % at the range of 298 K to 723 K, corresponding to the removal of the oxygen-containing groups. At high temperature, all the carbon will be oxidized to CO2 in air. For the dried 3D ZIF-8/rGO hydrogel, the TGA curve is different. At the range of 298 K to 573 K, the mass losing is mainly attributed to the oxygen-containing groups of rGO, while that at the range of 573 K to 673 K corresponds to the mass losing of dimethyl imidazole residues, that at 673-853 K corresponds to the mass losing of the organic ligand of ZIF-8, and the remaining about 18 % of the weight contributed to the residual Zn. Figure 3(b) shows the XRD patterns of the pure ZIF-8 and 3D ZIF-8/rGO composite. For the pure ZIF-8, the enhanced peak intensities at about 7.9°, 11.5°, 14.3°, 16.7°, 18.8°, and 20.7° correspond to the (011), (002), (112), (022), (013), and (222) planes in a typical ZIF-8. For the as-prepared dried 3D ZIF-8/rGO hydrogel, both the intensities and positions of the XRD peaks of all catalysts correspond to the typical known topology of ZIF-8, and a new wide peak at around 28° corresponds to the stacking of rGO nanosheets in the hydrogel, confirming the formation of the target nanocomposite. The surface area and pore distribution of the sample were analyzed by N2 adsorption and desorption isotherms. Figure 3(c) shows the pore size distributions of the dried 3D ZIF-8/rGO hydrogel, and the ZIF-8/rGO nanocomposite mainly hold the micropores less than 2 nm, while there are some big pores in size of about 10 nm, indicating that the as-formed 3D hydrogel owns both mesopores and micropores, which might originate from the inheritance of porous ZIF-8 and the stacking of graphene nanosheets, and the specific surface area is 1341.2 m2/g. Figure 3(d) shows the NH3-TPD profiles for the dried 3D ZIF-8/rGO hydrogel. There shows two acid sites with different acid strength, corresponding to Lewis and Brϕnsted acid sites, respectively, indicating that the as-prepared nanocomposite is an acidic catalyst.
Figure 4 shows typical XPS spectra of the dried 3D ZIF-8/rGO hydrogel, and the C 1s peak is mainly attributed to the C=C, indicating that the efficient reduction of rGO in the nanocomposite. In addition, the N1s, O1s, and Zn2p patterns also support the successful synthesis of the ZIF-8/rGO hydrogel.
The as-prepared 3D ZIF-8/rGO hydrogel has been directly used as the catalyst for the degradation of cellulose in its alkaline aqueous solution, and the homogeneous dispersion of the cellulose chain makes it possible to enter the inner pore of rGO hydrogel, and a series of degradation could take place therein under hydrothermal condition. After reaction, the products were analyzed by means of HPLC, and Fig. 5 shows the typical HPLC pattern of the degraded product. Clearly, the product is relativly simple, and the main products are oxalic acid, formic acid, and acetic acid.
Figure 6 shows the effect of degradation condition, especially catalyst, on the conversion of cellulose in its aqueous solution at 573 K. As a control, catalyst-free and only rGO hydrogel as catalyst was also studied at 573 K. Clearly, in all cases, cellulose can be degraded completely, indicating that the hydrothermal condition is a key point for cellulose degradation. However, the products and their distribution have a directly relationship with the catalysts. In free of any catalyst, the total yield of formic acid, oxalic acid, and acetic acid is only 35.11 % without formic acid. When rGO hydrogel was used as the catalyst, the total yield of the three acids only 19.07 %, also no formic acid formed. In addition, when pure ZIF-8 nanoparticles was used for the catalysis, the total yield of the three acids reached 58.63 %, with 48.52 % of formic acid, indicating that ZIF-8 favors the formation of formic acid conversion from the polymeric cellulose chains, and the TOF of the catalyst for formic acid is 10.40. However, when the as-prepared 3D ZIF-8/rGO hydrogel was used as the catalyst, the yield of formic acid increased hugely up to 91.12 % with TOF of 78.34, and the total yield of the three acids reached 99.91 %, indicating that the combination of the ZIF-8 MOF and the rGO hydrogel can promote the conversion of cellulose into formic acid efficiently. In addition, the yield of both oxalic acid and acetic acid decrease, indicating that the 3D ZIF-8/rGO hydrogel catalyst increases the selectivity of formic acid. Since cellulose is formed by glucose units, the molecular weight of cellulose is calculated as 162X (X is the number of glucose units). When Y g cellulose was used, it would be degraded to (Y/162X)×6×46X=1.70 Y g formic acid completely. Z g formic acid was obtained after the experiment, so the yield was calculated as Z/1.7Y.
Scheme 1 shows the possible mechanism of the degradation process, and the macromolecular cellulose chain was hydrolysis into glucose at first, and then it was continuously degraded into oxalic acid by oxidation route or levulinic acid by hydrolysis route, respectively. Next the as-formed oxalic acid or levulinic acid were degraded to formic acid or acetic acid .
The effect of reaction time has also been studied. Table Ⅰ lists the results of the yield of formic acid on the conversion of cellulose at various reaction time, and the yield of formic acid increases gradually from 0.5 h to 2 h with yield of 93.66 %, but it decreases to 85.38 % when the reaction time is 3 h, indicating that longer reaction time may result in the excess oxidation of formic acid.
The content of ZIF-8 inside the 3D ZIF-8/rGO nanocomposites is another key factor for the catalytic activity. When the ratio of ZIF-8 to rGO hydrogel in the nanocomposites changed from 0.01, 0.06, 0.12 to 0.23, the yield of formic acid was 0, 43.67 %, 91.12 %, and 56.57 %. Clearly, when in the low ZIF-8 context with ratio of 0.01, no formic acid was obtained while the yield increases to 91.12 % when the ratio of ZIF-8 is 0.12. However, the yield of formic acid decreased when the ratio of ZIF-8 increased up to 0.23, indicating that much ZIF-8 will result in the decreasing of the target product.
Temperature is another key factor for the reaction, and the reactions were carried out from 323 K to 573 K. Figure 7 shows the effect of temperature on the conversion of cellulose and the yield of relative acids, and the maximum conversion (100 %) and yield of formic acid (91.12 %) were obtained when the reaction was carried out at 573 K. At low temperature, the conversion of cellulose is only 50 % with 18.76 % yield of formic acid, and the conversion of cellulose and the yield of formic acid increase with the increase temperature.
The stability of the catalyst is another key point needed to be studied. The catalyst can be recovered easily from the products by filtration. After it was dried, it can be reused for the next cycle of reaction. Figure 8 shows results of five cycles. Clearly, the yield of formic acid decreases gradually with the increasing of catalyst cycles. However, the total yield of the three acids still remains very high with nearly no change, indicating the high catalytic activity.Ⅳ. CONCLUSION
A 3D hydrogel of reduced graphene oxide was synthesized via a reduction-induced self-assembly and it worked as the porous support for MOF nanoparticles, zinc ion and the ligand 2-methylimidazole were then dispersed inside the inner surface of the rGO hydrogel, and the nanoparticles of ZIF-8 grew therein to form a 3D ZIF-8/rGO hydrogel nanocomposite, which is in size of centimeters owning to both micropores and mesopores pores with acidic sites. The as-prepared 3D ZIF-8/rGO nanocomposite is stable in alkaline solution, and shows excellent catalysis activity for conversion of cellulose in its homogeneously aqueous solution, and the yield of formic acid reaches 93.66 %. It provides a new green method to convert cellulose to chemicals.Ⅴ. ACKNOWLEDGMENTS
This work is supported by the National Natural Science Foundation of China (No.51373162), and the Natural Science Foundation of Anhui Province (No.1408085MKL03).
|||A. Corma, S. Iborra, and A. Velty, Chem. Rev. 107 , 2411 (2007). DOI:10.1021/cr050989d|
|||J. J. Bozell, Science 329 , 522 (2010). DOI:10.1126/science.1191662|
|||D. R. Dodds, and R. A. Gross, Science 318 , 1250 (2007). DOI:10.1126/science.1146356|
|||C. O. Tuck, E. Perez, I. T. Horvath, R. A. Sheldon, and M. Poliakoff, Science 337 , 695 (2012). DOI:10.1126/science.1218930|
|||M. G. Adsul, A. J. Varma, and D. V. Gokhale, Green Chem. 9 , 58 (2007). DOI:10.1039/B605839F|
|||M. S. Holm, S. Saravanamurugan, and E. Taarning, Science 328 , 602 (2010). DOI:10.1126/science.1183990|
|||E. J. Cho, S. Jung, H. J. Kim, Y. G. Lee, K. C. Nam, H. J. Lee, and H. J. Bae, Chem. Commun. 48 , 886 (2012). DOI:10.1039/C2CC16661E|
|||M. Niu, Y. Hou, S. Ren, W. Wu, and K. N. Marsh, Green Chem. 17 , 453 (2015). DOI:10.1039/C4GC01440E|
|||M. Grasemann, and G. Laurenczy, Energy Environ. Sci. 5 , 8171 (2012). DOI:10.1039/c2ee21928j|
|||F. Jin, J. Yun, G. Li, A. Kishita, K. Tohji, and H. Enomoto, Green Chem. 10 , 612 (2008). DOI:10.1039/b802076k|
|||J. Xu, H. Zhang, Y. Zhao, Z. Yang, B. Yu, H. Xu, and Z. Liu, Green Chem. 16 , 4931 (2014). DOI:10.1039/C4GC01252F|
|||J. Xu, Y. Zhao, H. Xu, H. Zhang, B. Yu, L. Hao, and Z. Liu, Appli. Cat. B 154 , 267 (2014).|
|||J. Zhang, M. Sun, and Y. Han, Rsc Adv. 4 , 35463 (2014). DOI:10.1039/C4RA05424E|
|||J. Reichert, B. Brunner, A. Jess, P. Wasserscheid, and J. Albert, Energy Environ. Sci. 8 , 2985 (2015). DOI:10.1039/C5EE01706H|
|||J. L. Song, H. L. Fan, J. Ma, and B. X. Han, Green Chem. 15 , 2619 (2013). DOI:10.1039/c3gc41141a|
|||S. D. Zhu, Y. X. Wu, Q. M. Chen, Z. N. Yu, C. W. Wang, S. W. Jin, Y. G. Ding, and G. Wu, Green Chem. 8 , 325 (2006). DOI:10.1039/b601395c|
|||J. P. Zhou, and L. N. Zhang, Polym. J. 32 , 866 (2000). DOI:10.1295/polymj.32.866|
|||J. P. Zhou, L. N. Zhang, and J. Cai, J. Polym. Sci. Part B 42 , 347 (2004). DOI:10.1002/(ISSN)1099-0488|
|||N. Sun, H. Rodriguez, M. Rahman, and R. D. Rogers, Chem. Commun. 47 , 1405 (2011). DOI:10.1039/C0CC03990J|
|||C. Z. Li, Q. Wang, and Z. K. Zhao, Green Chem. 10 , 177 (2008). DOI:10.1039/B711512A|
|||L. L. Zhou, R. J. Liang, Z. W. Ma, T. H. Wu, and Y. Wu, Bioresour. Technol. 129 , 450 (2013). DOI:10.1016/j.biortech.2012.11.015|
|||F. J. Liu, R. K. Kamat, I. Noshadi, D. Peck, R. S. Parnas, A. M. Zheng, C. Z. Qi, and Y. Lin, Chem. Commun. 49 , 8456 (2013). DOI:10.1039/c3cc44703k|
|||L. Yan, and X. Qi, Acs Sustainable Chem. Eng. 2 , 897 (2014). DOI:10.1021/sc400507s|
|||Z. Jiang, Z. Zhang, J. Song, Q. Meng, H. Zhou, Z. He, and B. Han, ACS Sustainable Chem. Eng. 4 , 305 (2016). DOI:10.1021/acssuschemeng.5b01212|
|||J. Chen, R. Liu, Y. Guo, L. Chen, and H. Gao, Acs Catalysis 5 , 722 (2015). DOI:10.1021/cs5012926|
|||A. Corma, H. Garc, F. X. Llabr, and és í Xamena, Chem. Rev. 110 , 4606 (2010). DOI:10.1021/cr9003924|
|||M. Jahan, Z. Liu, and K. P. Loh, Adv. Funct. Mater. 23 , 5363 (2013). DOI:10.1002/adfm.v23.43|
|||Y. J. Min, W. Z. Li, G. Lin, L. S. Jia, W. H. Li, Z. W. Tao, and J. Qing, Adv. Energy Mater. 5 , 1500107 (2015). DOI:10.1002/aenm.201500107|
|||Y. Z. Chen, G. Cai, Y. Wang, Q. Xu, S. H. Yu, and H. L. Jiang, Green Chem. 18 , 1212 (2016). DOI:10.1039/C5GC02530C|
|||W. Chen, and L. Yan, Nanoscale 3 , 3132 (2011). DOI:10.1039/c1nr10355e|
|||L. Zhang, Y. Mao, J. P. Zhou, and J. Cai, Ind. Eng. Chem. Res. 44 , 522 (2005). DOI:10.1021/ie0491802|
|||W. Chen, S. Li, C. Chen, and L. Yan, Adv. Mater. 23 , 5679 (2011). DOI:10.1002/adma.v23.47|