Hydrogen is a clean fuel and has an important role in reducing environmental emissions for the future . Especially, the preparation of pure hydrogen is an attractive subject for fuel cell applications, which are considered to have the potential to provide a clean and alternative energy source for automobile. Hydrogen is also an important raw material for the chemical industry. For instance, it is used for ammonia synthesis, petrochemical refinery and methanol production. The use of lignocellulosic biomass as a rich, environmentally friendly and renewable resource for production of hydrogen or bio-fuels has attracted considerable attention [2-5]. Bio-oil is a black oxygenated organic liquid derived from fast pyrolysis of lignocellulosic biomass, which has been considered to be a promising platform chemical for producing hydrogen, bio-fuels or chemicals [1, 2, 6-9]. However, the raw bio-oil has some unfavorable properties such as high viscosity, high acidity, low stability, and low heating value due to its high oxygen content, which impedes its direct utilization as engine fuels . Thus, it is needed to transform oxygenated compounds in bio-oil into clean fuels to meet the requirements of the conventional transportation fuels. The catalytic reforming of bio-oil can produce a rich-hydrogen mixture gas, and further obtain pure hydrogen by gas separation process. Probably, production of hydrogen by the catalytic reforming of bio-oil is one of the most promising options because it can achieve higher yield and higher content of hydrogen, compared with direct gasification of biomass [1, 2].
The bio-oil reforming is mainly determined by the ability of a catalyst to catalyze the reforming reactions of oxygenated organic compounds and the water-shift reaction [1, 2]. Conventional commercial catalysts for the reforming of hydrocarbons are NiO/Al2O3 catalyst, usually operating at 700-850 ℃ [1, 2]. Various modified Ni-based catalysts and noble metal catalysts have been investigated for production of hydrogen via the reforming of bio-oil or its model oxygenates like acetic acid [11-13]. Noble metals generally show higher reforming reaction activity and less carbon depositing than using the Ni-based catalysts, but are not suitable for real applications due to high cost . One of the major problems for the catalytic reforming of bio-oil is the deactivation of catalysts due to coke or oligomer deposition even in the presence of an excess of steam [1, 2]. For the production of pure hydrogen from bio-oil, another hindering is that the gas products generally contain a certain amount of methane, carbon monoxide, carbon dioxide and other gaseous compounds, which increase the difficulty of hydrogen purification and its cost.
Generally, bio-oil contains hundreds of oxygenated organic compounds, including acids, alcohols, aldehydes, ketones, substituted phenolics and other oxygenates coming from the pyrolysis of biomass. In our previous work, much attention has been paid to the production of hydrogen or bio-syngas from bio-oil and its model compounds [14-16]. A low-temperature electrochemical catalytic reforming method for efficient production of hydrogen using the volatile fraction of the bio-oil and its compounds has been investigated [17-21]. An integrated catalytic transformation for production of high pure hydrogen from the bio-oil models was demonstrated, which included the catalytic reforming of oxygenated organic compounds to hydrogen-rich mixture gas followed by the water gas reaction and the removal of CO2. More attention was paid to production of hydrogen from the model compound of the sugar saccharide due to the lack of invetigation in the past. So far as we know, there is no such report regarding the production of high pure hydrogen from bio-oil and its models with very low contents of CO and CH4. This integrated catalytic transformation potentially provides a useful route for the production of pure hydrogen from lignocellulosic biomass-derived bio-oil.Ⅱ ExperimentS A Catalyst preparation and characterization
The metal and composite metal oxide catalysts with a settled molar ratio were prepared by the side-by-side co-precipitation method using respective metal nitrates as precursors, according to the same procedure introduced in our previous work [18, 19]. Briefly, the preparation mainly included the following steps: (ⅰ) the preparation of respective metal nitrates solution, (ⅱ) the preparation of a mixture of NaOH (1 mol/L) and Na2CO3 (1 mol/L) as precipitates, (ⅲ) the preparation of precipitates by the side-by-side co-precipitation of respective metal nitrates solution with the precipitators at a constant pH=9.0 and 80 ℃, (ⅳ) the precipitates was aged for 12 h at 25 ℃, washed until pH=7, and then dried at 110 ℃ overnight, (ⅴ) the precipitates were finally calcined at 550 ℃ for 6 h to obtain the corresponding metal and composite metal oxide catalysts.
The metallic element contents in the catalysts were determined by inductively coupled plasma and atomic emission spectroscopy (ICP/AES, Atomscan Advantage, Thermo Jarrell Ash Corporation, USA). The catalysts were also characterized by X-ray diffraction (XRD), N2 adsorption/desorption and temperature programmed oxidation (TPO) analyses, as the same procedures described in our previous work [22-25]. Briefly, the XRD patterns of the catalysts were obtained on an X'pert Pro Philips diffractrometer (Philips, Netherlands) using a Cu K
As shown in Fig. 1, an integrated catalytic transformation process for production of pure hydrogen from the bio-oil model compounds by coupling the catalytic reforming reactions, the water gas reaction and the CO2 adsorption were designed and operated under atmospheric pressure. The system consisted of three units: one for the production of hydrogen by the catalytic reforming of the bio-oil model compounds (part 1) together with the water gas reaction unit for the conversion of CO to CO2 (part 2) and the CO2 removal unit via the CO2 adsorbents (part 3).
For the production of hydrogen by the catalytic reforming of the bio-oil model compounds, 4 g of the Ni-based catalyst or the composite metal oxide catalyst was typically loaded in the reactor Ⅰ (inner diameter: 30 mm, length: 400 mm). Before each run, the catalysts were reduced by a 10%H2/Ar gas at 550 ℃ for 5 h. Then, the reactor was flushed with nitrogen at a flow rate of 300 cm
The water gas shift reaction unit for the conversion of CO to H2 and CO2 was conducted in the second unit (named as the second step). Typically, 4 g of the 34wt% CuO/65wt%ZnO/Al2O3 catalyst was loaded in the reactor Ⅱ (inner diameter: 30 mm, length: 400 mm). Before each run, the reactors were externally heated to the given temperatures. Then, the off-gas come from the reactor Ⅰ was fed into the reactor Ⅱ. The analysis procedures of the products obtained from the water gas shift reaction unit were the same as the steps for the catalytic reforming, as mentioned above. Finally, the CO2 removal by the CO2 adsorbents (like CaO/Na2CO3 solution or NaOH solution) was conducted in the third unit. Typically, 1 L of the CO2 adsorbent solution was loaded in the absorption bottle. The gaseous compositions were on-line analyzed using a gas chromatograph (GC-SP6890, Shandong Lunan Ruihong Chemical Instrument Co., Ltd., Tengzhou, China).
Finally, the conversion, hydrogen yield and the resulting product distribution were calculated according to the method introduced [20, 21]. All the tests were repeated three times and the reported data are the mean values of three trials.Ⅲ RESULTS AND DISCUSSION A Screening of reforming catalysts
The choice of the reforming catalysts is crucial in the production of hydrogen processes by the catalytic reforming of oxygenated organic compounds, due to its role in controlling the hydrogen yield and selectivity towards the different products. As shown in Table Ⅰ, the performance of production hydrogen by the the reforming reaction of glucose over different metals modfied Ni-baesd catalysts was tested under the low-temperature reforming (400 ℃) and the high-temperature reforming (700 ℃), respectivively. As compared with the common NiO/Al2O3 catalyst, the carbon conversion and the hydrogen yield using the the transition metals or precious metals modfied Ni-baesd catalysts (bimetallic composite oxide catalysts) obviously increased, indicating that adding the transition metals or precious metals into the NiO/Al2O3 catalyst improved the low-temperature reforming reactions at 400 ℃. Among the tested catalysts, the 20wt%NiO/20wt%CoO/Al2O3 catalysts exhibited the highest conversion of glucose along with the highest hydrogen yield. Meanwhile, the bimetallic composite oxide catalysts combined NiO with CoO, CuO or PdO significantly reduced the content of carbon monoxide in the resulting gas products, indicating that adding these transition metals or precious metals enhanced the water gas reaction during the low-temperature reforming process. However, the content of methane kept the close level over different metals modfied Ni-baesd catalysts at 400 ℃. This means the methane can not be efficiently removed by the low-temperature reforming reaction.
Futhermore, all tested Ni-baesd catalysts have good reforming activity for the high-temperature reforming of glucose. The content of methane was significantly reduced over different metals modfied Ni-baesd catalysts at 700 ℃. However, the content of CO from the high-temperature reforming was increased due to the decrease in the water gas reaction at high reaction temperatures. In view of no significant changes for high-temperature reforming characteristics (the conversion, hydrogen yield, and gas distribution) over the tested catalysts, the NiO/Al2O3 catalyst with lower cost was selected as the reforming reaction catalyst for producing pure hydrogen by the following integrated process.B Effect of temperature on production of hydrogen by direct catalytic reforming
Although the catalytic reforming of the bio-oil model compounds like acetic acid and ethylene glycol has been much studied [1, 26, 27], the catalytic reforming of sugar molecules is scarcely described in the literature. Figure 2 shows the effect of temperature on the production of hydrogen by the direct catalytic reforming of glucose over the 20wt%NiO/Al2O3 catalyst. The conversion of glucose monotonously rose from 350 ℃ to 700 ℃, and was close to a near complete conversion at 700 ℃. The highest yield of hydrogen was about 12.8wt%, corresponding to 95.9% of theoretical yield in the investigated temperature range. The above results indicate that high temperatures will facilitate the catalytic reforming reaction of glucose. Since the catalytic reforming reaction is an endothermic process, and thereby, enhance the yield of hydrogen at higher temperature.
The main gas products derived from the direct catalytic reforming of glucose mainly included H2 and CO2, together with small amount of CO and CH4. The distribution of the products observably depends on the reaction temperature. As the temperatures increased, the content of hydrogen was increased. Especially, the methane content decreased to 0.05% at 700 ℃ which was caused by the further reforming of the intermediate product of methane. Moreover, it was noticed that the content of carbon monoxide slightly increased with increasing the reforming temperature, accompanied by the decrease in the amount of carbon dioxide. This means that the water gas shift reaction was reduced at higher temperatures.C Production of pure hydrogen by coupling reforming reaction, water gas reaction and CO2 removal
To effectively remove carbon monoxide from the reforming-derived gas products, the experimental study on production of pure hydrogen with the integrated process by coupling the reforming reaction, the water gas reaction and the removal of CO2 was conducted. Firstly, the catalytic reforming of glucose into CO-containing mixture gas was run in the reactor Ⅰ over the 20wt%NiO/Al2O3 catalyst, and subsequently, CO obtained in the reactor Ⅰ were further converted into H2 and CO2 by the water gas reaction over the 34wt%CuO/65wt%ZnO/Al2O3 catalyst in the reactor Ⅱ in series. Finally, the production of pure hydrogen was conducted by the removal of CO2 by the chemical adsorption method.
Figure 3 shows the effect of the reforming reaction temperature (
To further study the water gas reaction, the reaction temperature in the reactor Ⅱ was varied in the range of 150-350 ℃, while the operating parameters in the reactor Ⅰ were fixed at the optimal reaction condition. Figure 4 shows the influence of the water gas shift reaction temperature (
In order to prepare high pure hydrogen, we tested the removal of CO2 by the CO2 adsorption method using the different adsorbents. As shown in Table Ⅱ, 25%(CaO+Na2CO3) solution or 20%NaOH solution behavoirs excellent adsorption ability for carbon dioxide. The CaO and CaOH solution present medium adsorption capacity for the removal of CO2. But molecular sieves like MCM-41 gives very lower adsorption ability for carbon dioxide. After the CO2 adsorption treatment, 99.96% of high purity hydrogen can be obtained in the integrated process via the coupling of the catalytic reforming reaction, the water gas shift reaction and the CO2 removal in series.D Stability of catalysts
As shown in Fig. 5, the stability of the catalysts was tested during the production of hydrogen by coupling the reforming reaction and the water gas reaction. The deactivation of the catalysts obviously observed in the above transformation processes, leading to the gradual decrease in the conversion of glucose from 97.1% to 86.7% along with the decrease in the yield of hydrogen from 99.2% to 85.4.0% for 35 h. With increasing the reaction time, the contents of H2 and CO2 decreased, accompanied by the increase of the CO and CH4 compositions.
In the integrated process, two catalysts of 20wt%NiO/Al2O3 and 34wt%CuO/65wt%ZnO/Al2O3, were selected and used for the reforming reaction and water gas reaction, respectively. Table Ⅲ shows main physical and chemical properties of the fresh and used catalysts. In comparison with the fresh 20wt%NiO/Al2O3 catalyst, the BET surface areas and pore volume for the used ones after the reforming slightly decreased, accompanied with the increase of the Ni particle size. The formation of the coke on the 20wt%NiO/Al2O3 used catalyst was also observed as characterized by the temperature programmed oxidation. Accordingly, the activity decay observed during the the reforming reaction process (in the first step) could be attributed to coke deactivation together with irreversible deactivation like the slight decrease in the BET surface areas and the increase of the Ni particle size, caused by the metal nanoparticle aggregation at the higher reforming temperatures. The catalyst deactivation caused by the carbon deposition on the 20wt%NiO/Al2O3 catalyst could be a main factor, since the most of activity (about 94.5%) for the used 20wt%NiO/Al2O3 catalyst was able to be recovered by regenerating the catalyst with oxygen at 600 ℃ for 5 h.
Moreover, the content of carbon monoxide in the gas products kept at a very low level ( < 0.1%) for 35 h run. This implies that the 34wt%CuO/65wt%ZnO/Al2O3 catalyst has good stability during the water gas reaction, mainly benefiting from the low-temperature operation conditions. The high stability of the 34wt%CuO/65wt%ZnO/Al2O3 catalyst was also supported by the fact that the negligible coke-deposition and no significant changes in its physical and chemical properties were observed during the water gas reaction.E Comparison of production hydrogen from different model compounds
Typically, bio-oil produced through the fast pyrolysis of biomass is a complex mixture of several hundreds of oxygenated organic compounds, which can be classified into acids, alcohols, aldehydes, ketones, sugar saccharide carbohydrates, furans, substituted phenolics and other oxygenates [1, 27]. To gain insight on the differences of the catalytic reforming reactions among different types of compounds, we performed the following comparative experiments using acetic acid, methanol, formaldehyde, acetone, glucose, furan and phenol as the selected model compounds.
Table Ⅳ shows the carbon conversion, hydrogen yield and gas product distribution for the catalytic reforming of different feedstocks over the 20wt%NiO/Al2O3 catalyst. The conversion and hydrogen yields obtained from the tested feedstocks were lined up as follows: methanol > acetone > formaldehyde > acetic acid > glucose > furan > phenol. Methanol has the highest low-temperature reforming reactivity for the production of hydrogen, along with a conversion of 91.6% and a H2 yield of 85.8% even at 400 ℃. The gas products mainly contains 72.45%H2 and 23.49%CO2 together with the low concentration of CO (0.84%) and CH4 (3.22%). This indicates that the methanol reforming reaction (CH3OH
Production of pure hydrogen from oxygenated organic compounds related to bio-oil was proved by a catalytic transformation involving the catalytic reforming of oxygenated organic compounds followed by the conversion of CO to CO2 by the water gas reaction and the removal of CO2. The highest content of H2 up to 99.96% and the conversion of 97.1% were obtained by the integrated catalytic transformation of glucose as the model compound of the sugar saccharides in bio-oil. The chioce of the reforming catalyst, the reaction conditions and various oxygenated organic compounds were investigated. The reaction pathways were discussed based on the observed characteristics of observed characteristics of decomposition catalytic reforming and the water gas reaction. The transformation potentially provides a useful way for production of pure hydrogen using biomass.Ⅴ ACKNOWLEDGMENTS
This work is supported by the National Sci-Tech Support Plan (No.2014BAD02B03), the Program for Changjiang Scholars and Innovative Research Team in University and the Fundamental Research Funds for the Central Universities (No.wk2060190040).
|||M. Balat, Int. J. Hydrogen Energy 33 , 4013 (2008). DOI:10.1016/j.ijhydene.2008.05.047|
|||S. A. Chattanathan,and S. Adhikari, N. Abdoul-moumine, Renew Sustain Energy Rev. 16 , 2366 (2012). DOI:10.1016/j.rser.2012.01.051|
|||R. Trane, S. Dahl, M. Skj th, - Rasmussen,and A. Jensen, Int. J. Hydrogen Energy 37 , 6447 (2012). DOI:10.1016/j.ijhydene.2012.01.023|
|||P. Y. Bi, J. C. Wang, Y. J. Zhang, P. W. Jiang, X. P. Wu, J. X. Liu, H. Xue, T. J. Wang,and Q. X. Li, Bioresour. Technol. 183 , 10 (2015). DOI:10.1016/j.biortech.2015.02.023|
|||Y. J. Zhang, P. Y. Bi, J. C. Wang, P. W. Jiang, X. P. Wu, H. Xue, J. X. Liu, X. G. Zhou,and Q. X. Li, Appl. Energy 150 , 128 (2015). DOI:10.1016/j.apenergy.2015.04.023|
|||J. C. Wang, P. Y. Bi, Y. J. Zhang, H. Xue, P. W. Jiang, X. P. Wu, J. X. Liu, T. J. Wang,and Q. X. Li, Energy 86 , 488 (2015). DOI:10.1016/j.energy.2015.04.053|
|||Y. N. Yuan, P. Y. Bi, M. H. Fan, Z. X. Zhang, P. W. Jiang,and Q. X. Li, J. Chem. Technol. Biotechnol. 89 , 239 (2014). DOI:10.1002/jctb.2014.89.issue-2|
|||S. Qiu, Y. Xu, T. Ye, M. Yamamoto, Y. Liu,and Q. X. Li, Energy Sources Part A 36 , 2238 (2014). DOI:10.1080/15567036.2011.567229|
|||P. Y. Bi, Y. N. Yuan, M. H. Fan, P. W. Jiang, Q. Zhai,and Q. X. Li, Bioresour. Technol. 136 , 222 (2013). DOI:10.1016/j.biortech.2013.02.100|
|||Z. X. Zhang, P. Y. Bi, P. W. Jiang, M. H. Fan, S. M. Deng, Q. Zhai,and Q. X. Li, Energy 90 , 1922 (2015). DOI:10.1016/j.energy.2015.07.009|
|||T. Hou, L. X. Yuan, T. Q. Ye, L. Gong, J. Tu, M. Yamamoto, Y. Torimoto,and Q. X. Li, Int. J. Hydrogen Energy 34 , 9095 (2009). DOI:10.1016/j.ijhydene.2009.09.012|
|||T. Q. Ye, L. X. Yuan, Y. Q. Chen, T. Kan, J. Tu, X. F. Zhu, Y. Torimoto, M. Yamamoto,and Q. X. Li, Catal. Lett. 127 , 323 (2009). DOI:10.1007/s10562-008-9683-2|
|||Z. X. Wang, Y. Pan, T. Dong, X. F. Zhu, T. Kan, L. X. Yuan, Y. Torimoto, M. Sadakata,and Q. X. Li, Appl. Catal. A 320 , 24 (2007). DOI:10.1016/j.apcata.2006.12.003|
|||Y. Xu, T. Q. Ye, S. B. Qiu, S. Ning, F. Y. Gong, Y. Liu,and Q. X. Li, Bioresour. Technol. 102 , 6239 (2011). DOI:10.1016/j.biortech.2011.02.069|
|||T. Dong, Z. X. Wang, L. X. Yuan, Y. Torimoto, M. Sadakata,and Q. X. Li, Catal. Lett. 119 , 29 (2007). DOI:10.1007/s10562-007-9148-z|
|||Z. X. Wang, T. Dong, L. X. Yuan, T. Kan, X. F. Zhu, Y. Torimoto, M. Sadakata,and Q. X. Li, Energy Fuels 21 , 2421 (2007). DOI:10.1021/ef0700275|
|||T. Kan, J. X. Xiong, X. L. Li, T. Q. Ye, L. X. Yuan, Y. Torimoto, M. Yamamoto,and Q. X. Li, Int. J. Hy-drogen Energy 35 , 518 (2010).|
|||L. X. Yuan, T. Q. Ye, F. Y. Gong, Q. X. Guo, Y. To, - rimoto, M. Yamamoto,and Q. X. Li, Energy Fuels 23 , 3103 (2009). DOI:10.1021/ef801131a|
|||Y. Q. Chen, L. X. Yuan, T. Q. Ye, S. B. Qiu, X. F. Zhu, Y. Torimoto, M. Yamamoto,and Q. X. Li, Int. J. Hydrogen Energy 34 , 1760 (2009). DOI:10.1016/j.ijhydene.2008.12.044|
|||F. Y. Gong, T. Q. Ye, L. X. Yuan, T. Kan, Y. Torimoto, M. Yamamoto,and Q. X. Li, Green Chem. 11 , 2001 (2009). DOI:10.1039/b915830h|
|||L. X. Yuan, Y. Q. Chen, C. F. Song, T. Q. Ye, Q. X. Guo, Q. S. Zhu, Y. Torimoto,and Q. X. Li, Chem. Commun. 41 , 5215 (2008).|
|||M. H. Fan, P. W. Jiang, P. Y. Bi, S. M. Deng, L. F. Yan, Q. Zhai, T. J. Wang,and Q. X. Li, Bioresour. Technol. 143 , 59 (2013). DOI:10.1016/j.biortech.2013.05.097|
|||C. G. Hong, F. Y. Gong, M. H. Fan, Q. Zhai, W. W. Huang, T. J. Wang,and Q. X. Li, J. Chem. Technol. Biotechnol. 88 , 109 (2013). DOI:10.1002/jctb.2013.88.issue-1|
|||W. W. Huang, F. Y. Gong, M. H. Fan, Q. Zhai, C. G. Hong,and Q. X. Li, Bioresour. Technol. 121 , 248 (2012). DOI:10.1016/j.biortech.2012.05.141|
|||F. Y. Gong, Z. Yang, C. G. Hong, W. W. Huang, S. Ning, Z. X. Zhang, Y. Xu,and Q. X. Li, Bioresour. Technol. 102 , 9247 (2011). DOI:10.1016/j.biortech.2011.07.009|
|||E. Gallegos, - Su, á rez, A. Guerrero, - Ruiz, M. Fern, á ndez, - Garc, I. Rodr, í guez, - Ramos,and A. Kubacka, Appl. Catal. B 165 , 139 (2015). DOI:10.1016/j.apcatb.2014.10.007|
|||S. A. Chattanathan,and S. Adhikari, N. Abdoul-moumine, Renew. Sust. Energ. Rev. 16 , 2366 (2012). DOI:10.1016/j.rser.2012.01.051|