Enhanced CH 4 Production from CornStalk Pyrolysis Using Ni-5 CeO 2 / MCM-41 as a Catalyst

Production of syngas from lignocellulosic biomass though pyrolysis is a promising solution for the large-scale utilization of biomass. However, current pyrolysis approaches suffer from the relative low product yield and selectivity, limiting their practical application. To solve this problem, a series of nickel-based catalysts including Ni/MCM-41, Ni-5CeO2/MCM-41, and Ni-5La2O3/MCM-41 were prepared and characterized by transmission electron microscopy (TEM), N2 adsorption–desorption, X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and H2-temperature-programmed reaction (TPR) analysis. It was observed that the simultaneous addition of Ni and CeO2 to MCM-41 could increase the reducibility of Ni and the number of active Ni0 sites on the surface of the catalyst. Consequently, Ni-5CeO2/MCM-41 gave a CH4 yield of 14.6 mmol/g, which is remarkably higher than that (10.5 mmol/g) obtained in the absence of the catalyst. Meanwhile, the CO and H2 yields increased slightly, while the CO2 yield decreased slightly. Therefore, the improved CH4 yield and selectivity was mainly due to the increased decomposition of tarry compounds catalyzed by Ni/MCM-41 with the assistance of CeO2.


Introduction
According to Ministry of Agriculture of China, over 900 million tons of biomass stalk are produced every year in China, and around 60% of stalk is arbitrarily stacked, discarded, or burned as solid waste [1,2].The improper dispositions of lignocellulosic biomass caused serious air, soil, and water environmental problems.In order to alleviate these problems, the conversion of the biomass resource into alternative fine chemicals, bio-fuels, and novel materials is a sustainable and promising solution [3,4].Numerous approaches, such as combustion, pyrolysis, pyro-gasification, fermentation, and catalytic conversion, were developed to convert biomass into value-added products [5][6][7][8][9][10].Pyrolysis is a promising way to transform biomass into renewable energy in the form of H 2 , CO, and CH 4 [11].
Nickel-based catalysts were proven to be effective to enhance the biomass pyrolysis via promoting the C-C bond cleavage of biomacromolecules during biomass pyrolysis [12,13].Nickel-based catalysts are also capable to reform tars and, thus, decrease the need for their removal from producer gas [14].In addition, nickel-based catalysts are more economically viable than noble metals [15,16].The challenges in the application of nickel-based catalysts are their rapid deactivation due to coke deposition, sintering, and morphological transformations [14,17], especially when the temperature exceeds 600 • C [18].Therefore, much efforts were devoted to improve the applicability and lifetime of nickel-based catalysts in a series of catalytic processes, such as water-gas shift reaction, tar cracking, light hydrocarbon reforming, and methanol steam reforming [19].
When Ni species are loaded onto the appropriate support, the positive metal/support synergistic effect will remarkably improve their catalytic performance and stability.For example, Choi et al. demonstrated that the addition of CeO 2 to Al 2 O 3 prior to loading Ni species increased the dispersion of the Ni species on the support due to improved CeO 2 -Ni interaction, leading to increased hydrogen yield and longer lifetime of the catalyst during auto-thermal reforming of iso-octane [20].Sánchez-Sánchez et al. showed that the existence of Ni-Ce species Ni/CeO 2 -Al 2 O 3 could enhance adsorption and dissociation of water molecules, thus improving their catalytic activity for ethanol reforming [21].MCM-41 has a large specific surface area and uniform large pore size, which can aid the easy diffusion of reactant molecules in and out of its mesopores [22].Du et al. reported that Ni/MCM-41 gives a CH 4 selectivity of 96.0% from the methanation of CO at 400 • C under atmospheric pressure [23].Zhang et al. reported that Ni/MCM-41 is also effective to produce substitute natural gas from the methanation of syngas [24].Recent studies showed that the use of CeO 2 as an additive could effectively promote the adsorption and the activation of CO 2 molecules on the catalysts, owing to the oxygen exchange capacity of CeO 2 [25,26].Moreover, the addition of CeO 2 could promote the interaction between support and metal active component, so as to adequately control the growth of the Ni particles and then improve the catalytic performance of Ni-based catalysts [27].Although the addition of CeO 2 and La 2 O 3 was demonstrated to be effective to promote the CO 2 methanation over Ni/Al 2 O 3 and Ni/MCM-41, limited research was conducted on their application for reforming the gases from biomass pyrolysis.
In this study, a series of a series of nickel-based catalysts including Ni/MCM-41, Ni-5CeO 2 /MCM-41, and Ni-5La 2 O 3 /MCM-41 were prepared via the deposition-precipitation method.The synthesized catalysts were analyzed by transmission electron microscopy (TEM), N 2 adsorption-desorption, X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and H 2 -temperature-programmed reaction (TPR), and then used to reform the gases from corn-stalk pyrolysis.The effect of CeO 2 and La 2 O 3 on the catalytic performance of Ni-5CeO 2 /MCM-41 and Ni-5La 2 O 3 /MCM-41 for the reforming of the gases from corn-stalk pyrolysis was investigated.

Materials and Methods
Corn stalk was obtained from a crop processing company (Jiangsu, China).The corn stalk was sieved with a 180-380-µm mesh and then dried at 105 • C for 7 h before use as feedstock.The proximate (air-dry base) and ultimate analyses (dry-ash-free base) of the pine sawdust sample were conducted in a muffle furnace (SX-11-4, China) and an elemental analyzer (Vario EL cube, Elementaranalyse, Germany), respectively.The volatile matter, fixed carbon, and ash contents of the corn stalk were 77.4%, 16.86%, and 6.02%, respectively, and the contents of carbon, hydrogen, oxygen, nitrogen, and sulfur in the corn stalk were 41.18%, 4.98%, 39.51%, 0.78%, and 0.19%, respectively.MCM-41 silica was purchased from Beijing Huawei Ruike Corp. (Beijing, China).Al 2 O 3 was purchased from Shanghai Macklin Biochemical Co.,Ltd.(Shanghai, China).Both MCM-41 and Al 2 O 3 were calcined at 600 • C for 4 h to eliminate the impurities adsorbed onto the surface prior to the preparation of the catalysts.Ni-5CeO 2 /MCM-41 was prepared according to the following steps: firstly, 4.94 g of Ni(NO 3 ) 2 •6H 2 O and 2.52 g of Ce(NO 3 ) 3 •6H 2 O were added to 250 mL of deionized water to get a uniform solution, and then 4.0 g of MCM-41 was added to the solution, followed by stirring for 1 h at ambient temperature.Subsequently, 0.2 mol/L K 2 CO 3 solution was added dropwise into the suspension as a precipitant until the pH value reached 9-10, followed by stirring for 2 h at 90 • C. The suspension was filtered and washed with deionized water until reaching a neutral state.The obtained precipitate was dried at 110 • C overnight and calcined at 600 • C in air for 4 h.Ni/5La 2 O 3 /MCM-41 was prepared using a similar procedure, except for the replacement of Ce(NO 3 ) 3 •6H 2 O with 2.658 g of La(NO 3 ) 3 •6H 2 O. Ni-5CeO 2 /Al 2 O 3 was prepared using the following steps: firstly, 2.47 g of Ni(NO 3 ) 2 •6H 2 O and 1.26 g of Ce(NO 3 ) 3 •6H 2 O were added to 5.4 mL of deionized water to get a uniform solution, and then the solution was dropwise added to 4.5 g Al 2 O 3 and mixed well.Next, the mixture solution was stirred at 35 • C for 4 h, and then dried at 80 • C overnight and calcined at 600 • C in air for 4 h.Ni-5La 2 O 3 /Al 2 O 3 was prepared using a similar procedure, except for the replacement of Ce(NO 3 ) 3 •6H 2 O with 1.329 g of La(NO 3 ) 3 •6H 2 O.
The XPS spectra were recorded on a PHI 5000 VersaProbe III-Scanning XPS Microprobe equipped with a focused monochromatic aluminum X-ray source with photon energy of 1486.6 eV and X-ray beam size of 100 µm.The binding energy (BE) was corrected by aligning the peak positions with reference to the standard adventitious carbon peak (C1s) at 284.6 eV.The N 2 adsorption-desorption isotherms were recorded at −196 • C using a Quadrasorb SI apparatus.Before the measurement, the sample was degassed at 300 • C for 3 h.The specific surface area and pore size were obtained using the Brunauer-Emmett-Teller (BET) method, while the pore volume was calculated using the Barrett-Joyner-Halenda (BJH) model using the desorption isotherm branch.The XRD pattern was measured on a Rigaku Miniflex diffractometer operating at 40 kV and 20 mA.Each sample was scanned at a rate of 2 • /min every 0.02 • for 2θ angles between 10 • and 90 • .The morphology of catalysts was observed by transmission electron microscopy (TEM, Philips-FEI, Tecnai G2F30).Temperature-programmed reaction of hydrogen (H 2 -TPR) was carried out on an AutoChem II 2920 instrument (Micromeritics Instrument).Each sample (30 mg) was pretreated with argon at 200 • C for 1 h, and then cooled down to 50 • C. The reduction was carried out with a heating rate of 10 • C/min up to 900 • C. The H 2 consumption amount was recorded with a thermal conductivity detector (TCD).
The pyrolysis of corn stalk was performed in a fixed-bed horizontal tubular furnace (length: 100 cm; inner diameter: 2.5 cm), as shown in Figure 1.Initially, 2.3 g of corn stalk was placed on the left side of the feedstock bin and then covered by an asbestos-free wire gauze.A layer of silica sand (30 g) was placed in the middle of the feedstock bin and then covered by another asbestos-free wire gauze to separate biomass and catalyst.Then, 0.2 g of Ni-based catalyst was added to the right side of the feedstock bin.Nitrogen gas was employed to maintain an anaerobic environment in the reactor.Typically, the feedstock bin was placed out of the tubular furnace to keep the feedstock at low temperature during the heating process of the tubular furnace.The blank experiment demonstrated that no gas products were detected when the feedstock bin was placed out of the tubular furnace.When the temperature of the furnace reached 600 • C, the feedstock bin was immediately pushed into the reactor in the middle of the furnace.The pyrolysis of biomass was conducted at 600 • C for 30 min.After the reaction, the gas product was pushed out by a nitrogen gas (40 L/h) for 8 min and collected by a 10-L air bag.The absorption and filtration devices were used to cool down the formed bio-oil and tar, and they retained these products in the liquid phase to avoid their adverse effect on the measurement of gas composition.The concentrations of H 2 , CO, CO 2 , and CH 4 were determined by gas chromatography (GC; Thermo Fisher TRACE 1310) and then their yields and CH 4 selectivity were calculated.The XPS spectra were recorded on a PHI 5000 VersaProbe III-Scanning XPS Microprobe equipped with a focused monochromatic aluminum X-ray source with photon energy of 1486.6 eV and X-ray beam size of 100 μm.The binding energy (BE) was corrected by aligning the peak positions with reference to the standard adventitious carbon peak (C1s) at 284.6 eV.The N2 adsorptiondesorption isotherms were recorded at −196 °C using a Quadrasorb SI apparatus.Before the measurement, the sample was degassed at 300 °C for 3 h.The specific surface area and pore size were obtained using the Brunauer-Emmett-Teller (BET) method, while the pore volume was calculated using the Barrett-Joyner-Halenda (BJH) model using the desorption isotherm branch.The XRD pattern was measured on a Rigaku Miniflex diffractometer operating at 40 kV and 20 mA.Each sample was scanned at a rate of 2°/min every 0.02° for 2θ angles between 10° and 90°.The morphology of catalysts was observed by transmission electron microscopy (TEM, Philips-FEI, Tecnai G2F30).Temperature-programmed reaction of hydrogen (H2-TPR) was carried out on an AutoChem II 2920 instrument (Micromeritics Instrument).Each sample (30 mg) was pretreated with argon at 200 °C for 1 h, and then cooled down to 50 °C.The reduction was carried out with a heating rate of 10 °C/min up to 900 °C.The H2 consumption amount was recorded with a thermal conductivity detector (TCD).
The pyrolysis of corn stalk was performed in a fixed-bed horizontal tubular furnace (length: 100 cm; inner diameter: 2.5 cm), as shown in Figure 1.Initially, 2.3 g of corn stalk was placed on the left side of the feedstock bin and then covered by an asbestos-free wire gauze.A layer of silica sand (30 g) was placed in the middle of the feedstock bin and then covered by another asbestos-free wire gauze to separate biomass and catalyst.Then, 0.2 g of Ni-based catalyst was added to the right side of the feedstock bin.Nitrogen gas was employed to maintain an anaerobic environment in the reactor.Typically, the feedstock bin was placed out of the tubular furnace to keep the feedstock at low temperature during the heating process of the tubular furnace.The blank experiment demonstrated that no gas products were detected when the feedstock bin was placed out of the tubular furnace.When the temperature of the furnace reached 600 °C, the feedstock bin was immediately pushed into the reactor in the middle of the furnace.The pyrolysis of biomass was conducted at 600 °C for 30 min.After the reaction, the gas product was pushed out by a nitrogen gas (40 L/h) for 8 min and collected by a 10-L air bag.The absorption and filtration devices were used to cool down the formed bio-oil and tar, and they retained these products in the liquid phase to avoid their adverse effect on the measurement of gas composition.The concentrations of H2, CO, CO2, and CH4 were determined by gas chromatography (GC; Thermo Fisher TRACE 1310) and then their yields and CH4 selectivity were calculated.Energies 2019, 12, 774 5 of 12 N 2 adsorption-desorption isotherms of the synthesized catalyst were performed to analyze their textural properties.As displayed in Figure 3a, the isotherms exhibited type IV curves with H4 hysteresis loops (IUPAC International Union of Pure and Applied Chemistry).The pore size distribution displayed in Figure 3b indicated that these catalysts have a typical mesoporous structure.Based on the nitrogen adsorption-desorption data, the specific surface area (S BET ), pore volume (V pore ), and average pore size (D meso ) of these materials were computed and are listed in Table 1.Compared with Al 2 O 3 -based catalysts, the MCM-41-based materials exhibited remarkably higher specific surface area and pore diameter, while their pore volumes were comparable.The specific surface areas of the synthesized catalysts increased in the following order: N2 adsorption-desorption isotherms of the synthesized catalyst were performed to analyze their textural properties.As displayed in Figure 3a, the isotherms exhibited type IV curves with H4 hysteresis loops (IUPAC International Union of Pure and Applied Chemistry).The pore size distribution displayed in Figure 3b indicated that these catalysts have a typical mesoporous structure.Based on the nitrogen adsorption-desorption data, the specific surface area (SBET), pore volume (Vpore), and average pore size (Dmeso) of these materials were computed and are listed in Table 1.Compared with Al2O3-based catalysts, the MCM-41-based materials exhibited remarkably higher specific surface area and pore diameter, while their pore volumes were comparable.The specific surface areas of the synthesized catalysts increased in the following order: 5CeO2/MCM-41 < Ni/MCM-41 < Ni-5CeO2/MCM-41 < 5La2O3/MCM-41 < Ni-5La2O3/MCM-41.The specific surface areas of Ni-5La2O3/MCM-41 and Ni-5CeO2/MCM-41 were higher than that of Ni-5CeO2/MCM, indicating that the simultaneous loading of Ni and metal oxide was effective to maintain the relatively high specific surface area of the catalyst.Both the pore volume and the average pore size of Ni-5CeO2/MCM and Ni-5La2O3/MCM were lower than those of the corresponding 5CeO2/MCM and 5La2O3/MCM materials, suggesting that the loading of Ni species was responsible for the decrease in pore volume and pore diameter.

XRD
The XRD patterns of the synthesized catalysts were determined to analyze their crystal structure.As displayed in Figure 4, only one broad diffraction peak attributed to MCM-41 was clearly observed in the XRD patterns of 5La 2 O 3 /MCM-41 and 5CeO 2 /MCM-41, while the characteristic peaks of La 2 O 3 and CeO 2 were not observable owing to the relative low loading (less than 5 wt.%), suggesting the highly dispersibility of La 2 O 3 and CeO 2 in 5La 2 O 3 /MCM-41 and 5CeO 2 /MCM-41.When Ni species were introduced to MCM-41, 5CeO Ni-5La 2 O 3 /MCM-41, respectively [28,29].The diffraction peaks were very weak and broad, suggesting that the nickel-oxide particles formed in the synthesis process were small and sufficiently dispersed.The XRD pattern of Ni-5La 2 O 3 /MCM-41 also exhibited a diffraction peak at 2θ = 44.4• , corresponding to metallic nickel (Ni 0 ), indicating that some metallic nickel was formed during the synthesis process.Moreover, the diffraction peak at 29.6 • corresponding to La 2 O 3 became obvious in the XRD pattern of Ni-5La 2 O 3 /MCM-41, indicating that the introduction of Ni species may have resulted in the agglomeration of small La 2 O 3 particles [30].
As displayed in Figure 4, only one broad diffraction peak attributed to MCM-41 was clearly observed in the XRD patterns of 5La2O3/MCM-41 and 5CeO2/MCM-41, while the characteristic peaks of La2O3 and CeO2 were not observable owing to the low loading (less than 5 wt.%), suggesting the highly dispersibility of La2O3 and CeO2 in 5La2O3/MCM-41 and 5CeO2/MCM-41.When Ni species were introduced to MCM-41, 5CeO2/MCM-41, and 5La2O3/MCM-41, the characteristic peaks of NiO appeared at 2θ = 35.6°and 60.8° in the XRD patterns of Ni-5CeO2/MCM-41 and Ni-5La2O3/MCM-41, respectively [28,29].The diffraction peaks were very weak and broad, suggesting that the nickeloxide particles formed in the synthesis process were small and sufficiently dispersed.The XRD pattern of Ni-5La2O3/MCM-41 also exhibited a diffraction peak at 2θ = 44.4°,corresponding to metallic nickel (Ni 0 ), indicating that some metallic nickel was formed during the synthesis process.Moreover, the diffraction peak at 29.6° corresponding to La2O3 became obvious in the XRD pattern of Ni-5La2O3/MCM-41, indicating that the introduction of Ni species may have resulted in the agglomeration of small La2O3 particles [30].

XPS
The XPS spectra of the synthesized catalysts were measured to analyze their surface elemental composition and chemical state of the transition metal.As displayed in Figure 5a, the wide XPS spectra of all the materials exhibited the characteristic peak of Si 2p and O 1s, centered at 104.4 and 533.8 eV, respectively.The presence of Ce in 5CeO2/MCM-41 and Ni-5La2O3/MCM-41 was confirmed by the peak at 902 eV, attributed Ce 3d [31]; the presence of La in 5La2O3/MCM-41 and Ni-5La2O3/MCM-41 was confirmed by the peak at 836.0 eV, attributed La 3d [28,32].Eight peaks were observed on the high-resolution Ce 3d XPS spectra (Figure 5b).The four peaks of Ce 3d5/2 at 883.9, 886.0, 890.2, and 899.4 eV were attributed to V, V', V'', and V''', respectively [33].The four peaks of Ce 3d3/2 at 902.7, 905.5, 909.1, and 917.7 eV were attributed to U, U', U'', and U''', respectively [34].The characteristic peaks attributed to La 3d2/3 and La 3d5/3 appeared at 851.7 and 835.0 eV, respectively, consistent with La-O bonding in La2O3 (Figure 5c) [35,36].As shown in Figures 5d and 5e

XPS
The XPS spectra of the synthesized catalysts were measured to analyze their surface elemental composition and chemical state of the transition metal.As displayed in Figure 5a, the wide XPS spectra of all the materials exhibited the characteristic peak of Si 2p and O 1s, centered at 104.4 and 533.8 eV, respectively.The presence of Ce in 5CeO 2 /MCM-41 and Ni-5La 2 O 3 /MCM-41 was confirmed by the peak at 902 eV, attributed Ce 3d [31]; the presence of La in 5La 2 O 3 /MCM-41 and Ni-5La 2 O 3 /MCM-41 was confirmed by the peak at 836.0 eV, attributed La 3d [28,32].Eight peaks were observed on the high-resolution Ce 3d XPS spectra (Figure 5b).The four peaks of Ce 3d5/2 at 883.9, 886.0, 890.2, and 899.4 eV were attributed to V, V', V", and V"', respectively [33].The four peaks of Ce 3d3/2 at 902.7, 905.5, 909.1, and 917.7 eV were attributed to U, U', U", and U"', respectively [34].The characteristic peaks attributed to La 3d 2/3 and La 3d 5/3 appeared at 851.7 and 835.0 eV, respectively, consistent with La-O bonding in La 2 O 3 (Figure 5c) [35,36].As shown in Figure 55d,e

H 2 -TPR
The H 2 -TPR profiles of Ni-MCM-41, 5CeO 2 /MCM-41, 5La 2 O 3 /MCM-41, Ni-5CeO 2 /MCM-41, and Ni-5La 2 O 3 /MCM-41 were measured to investigate their reducibility.As shown in Figure 6, only one weak reduction peak at around 493 • C was observed on the H 2 -TPR profiles of 5CeO 2 /MCM-41, while almost no reduction peak was observed on the H 2 -TPR profiles of 5La 2 O 3 /MCM-41.The predominant peak corresponding to the reduction of NiO particles interacting strongly with the support appeared at 675, 648, and 637 • C in the H 2 -TPR profiles of Ni-MCM-41, Ni-5CeO 2 /MCM-41, and Ni-5La 2 O 3 /MCM-41, respectively.In addition, the shoulder peaks between 450 and 520 • C in the H 2 -TPR profiles of Ni-5CeO 2 /MCM-41 and Ni-5La 2 O 3 /MCM-41 were attributed to the reduction of NiO particles interacting weakly with the support.The NiO reduction peak shifted to lower temperature compared with that of other tested catalysts, indicating the high dispersity of metal components on the support.The enhanced dispersity in Ni-5CeO 2 /MCM-41 was mainly attributed to that CeO 2 phase, which could weaken the strong NiO-support interactions [38].These results indicated that the addition of CeO 2 could increase the reducibility of Ni and the number of active Ni 0 sites on the catalyst surface, in accordance with the conclusion of the XRD and XPS analysis.compared with that of other tested catalysts, indicating the high dispersity of metal components on the support.The enhanced dispersity in Ni-5CeO2/MCM-41 was mainly attributed to that CeO2 phase, which could weaken the strong NiO-support interactions [38].These results indicated that the addition of CeO2 could increase the reducibility of Ni and the number of active Ni 0 sites on the catalyst surface, in accordance with the conclusion of the XRD and XPS analysis.

Effect of Temperature and Time on the Gas Yield
The pyrolysis of corn stalk (2.30 g) was performed under different temperatures in the absence of catalyst, with a constant pyrolysis time of 30 min.As shown in Figure 7a, the CO yields were between 1.9 and 3.3 mmol/g, and the CO2 yields were between 3.0 and 4.0 mmol/g when the reaction temperature lower than 500 °C.H2 started to form at 500 °C, while CH4 started to form at 550 °C.With the increase of reaction temperature from 500 to 700 °C, the H2 yield increased continuously from 0.1 to 1.8 mmol/g, the CH4 yield increased from 5.8 to 14.0 mmol/g, and the CO yield increased slightly from 2.6 to 3.3 mmol/g, while the CO2 yield ranged between 3.5 and 4.0 mmol/g.Since CH4 was identified as the main pyrolysis product, the CH4 selectivity was also analyzed in the subsequent process.It is interesting to note that the CH4 selectivity also increased with the increasing temperature except that the CH4 selectivity at 600 °C was slightly higher than that obtained at 650 °C.To avoid the effect of the decomposition of asbestos-free wire gauze, we calcined the asbestos-free wire gauze prior to use.When the asbestos-free wire gauze was calcined at a temperature lower than 650 °C, the gas yields were not affected by the asbestos-free wire gauze.However, when the asbestos-free wire gauze was calcined at a temperature higher than 700 °C, the asbestos-free wire gauze broke down after calcination and then could not be used to separate the catalyst and feedstock.The control experiment indicated that the increased CH4 yield obtained at 700 °C was mainly attributed to the decomposition of asbestos-free wire gauze at 700 °C.Since the effect of asbestos-free wire gauze could not be precluded at 700 °C, we only investigated the effect of the catalysts on biomass pyrolysis at 600 °C in the subsequent experiment.
In the subsequent experiment, the influence of reaction time on the pyrolysis of corn stalk was investigated to confirm the appropriate pyrolysis time.The H2, CO, CO, and CH4 yields were low when the corn stalk was loaded into the reactor before heating.In this manner, the heating process required about 120 min, owing to the relatively low heating rate of the reactor, thus leading to the low product yields [28].To solve this problem, rapid loading of biomass was used in the subsequent  The pyrolysis of corn stalk (2.30 g) was performed under different temperatures in the absence of catalyst, with a constant pyrolysis time of 30 min.As shown in Figure 7a, the CO yields were between 1.9 and 3.3 mmol/g, and the CO 2 yields were between 3.0 and 4.0 mmol/g when the reaction temperature lower than 500 • C. H 2 started to form at 500 • C, while CH 4 started to form at 550 • C. With the increase of reaction temperature from 500 to 700 • C, the H 2 yield increased continuously from 0.1 to 1.8 mmol/g, the CH 4 yield increased from 5.8 to 14.0 mmol/g, and the CO yield increased slightly from 2.6 to 3.3 mmol/g, while the CO 2 yield ranged between 3.5 and 4.0 mmol/g.Since CH 4 was identified as the main pyrolysis product, the CH 4 selectivity was also analyzed in the subsequent process.It is interesting to note that the CH 4 selectivity also increased with the increasing temperature except that the CH 4 selectivity at 600 • C was slightly higher than that obtained at 650 • C. To avoid the effect of the decomposition of asbestos-free wire gauze, we calcined the asbestos-free wire gauze prior to use.When the asbestos-free wire gauze was calcined at a temperature lower than 650 • C, the gas yields were not affected by the asbestos-free wire gauze.However, when the asbestos-free wire gauze was calcined at a temperature higher than 700 • C, the asbestos-free wire gauze broke down after calcination and then could not be used to separate the catalyst and feedstock.The control experiment indicated that the increased CH 4 yield obtained at 700 • C was mainly attributed to the decomposition of asbestos-free wire gauze at 700 • C. Since the effect of asbestos-free wire gauze could not be precluded at 700 • C, we only investigated the effect of the catalysts on biomass pyrolysis at 600 • C in the subsequent experiment.
In the subsequent experiment, the influence of reaction time on the pyrolysis of corn stalk was investigated to confirm the appropriate pyrolysis time.The H 2 , CO, CO, and CH 4 yields were low when the corn stalk was loaded into the reactor before heating.In this manner, the heating process required about 120 min, owing to the relatively low heating rate of the reactor, thus leading to the low product yields [28].To solve this problem, rapid loading of biomass was used in the subsequent pyrolysis experiment.The feedstock was placed outside during the heating of the reactor.Once the temperature of the reactor reached the stated value, the biomass in the feedstock bin was pushed into the middle of the reactor immediately to start the pyrolysis process.With pyrolysis times from 10 to 30 min, the H 2 , CO, CO, and CH 4 yields, as well as CH 4 selectivity, increased continuously (Figure 7b).However, the further increase of pyrolysis time from 30 to 40 min did not lead to an obvious increase in product yields and CH 4 selectivity.Based on the above results, the pyrolysis was performed at 600 • C for 30 min in the subsequent experiment to compare the catalytic performance of different catalysts.
Energies 2019, 12, 774 9 of 12 30 min, the H2, CO, CO, and CH4 yields, as well as CH4 selectivity, increased continuously (Figure 7b).However, the further increase of pyrolysis time from 30 to 40 min did not lead to an obvious increase in product yields and CH4 selectivity.Based on the above results, the pyrolysis was performed at 600 °C for 30 min in the subsequent experiment to compare the catalytic performance of different catalysts.

Effect of Catalyst on the Gas Yield
A series of monometallic catalysts, including Ni-MCM-41, 5CeO2-MCM-41, Ni-Al2O3, 5CeO2/ Al2O3, and 5La2O3/Al2O3, were investigated for the catalytic reforming of pyrolysis gases.As shown in Figure 8a, the gas yields and CH4 selectivity obtained with Ni/MCM-41 and Ni/Al2O3 were slightly higher than those obtained in the absence of the catalyst, indicating the catalytic activity of the Ni species.However, both the gas yields and CH4 selectivity obtained with 5CeO2/MCM-41 and 5La2O3/Al2O3 were not obviously improved when compared to the control group, indicating that these materials were incapable of improving the quality of gas products.In addition, a series of bimetallic catalysts, including Ni-5CeO2/MCM-41, Ni-10CeO2/MCM-41, Ni-5La2O3/MCM-41, and Ni-10La2O3/MCM-41, were tested to further improve the catalytic performance.When Ni-5CeO2/MCM-41 was used as a catalyst (Figure 7b), the CH4 yield increased considerably from 10.5 to 14.6 mmol/g, and the CO and H2 yields increased slightly from 1.4 and 2.9 to 1.7 and 3.2 mmol/g, respectively, while the CO2 yield reduced slightly from 3.4 to 2.8 mmol/g.As a consequence, the CH4 selectivity increased from 58.2 to 65.6%.When the CeO2 loading was increased to 10 wt.%, the gas yields were slightly higher than those obtained with Ni/MCM-41, but lower than those obtained with Ni-5CeO2/MCM-41.In contrast, the gas yields obtained with Ni-5La2O3/MCM-41 and Ni-10La2O3/MCM-41 were even lower than those obtained with Ni/MCM-41.The H2 yield increased slightly when Ni-5CeO2/MCM-41 was used as a catalyst, suggesting that CO2 methanation did not occur in this condition.Many recent studies showed that nickel-based catalysts are also capable of reforming tars [14].Therefore, the increased CH4 yield was mainly due to the decomposition of tarry compounds on the catalyst, instead of the methanation of CO2.7b), the CH 4 yield increased considerably from 10.5 to 14.6 mmol/g, and the CO and H 2 yields increased slightly from 1.4 and 2.9 to 1.7 and 3.2 mmol/g, respectively, while the CO 2 yield reduced slightly from 3.4 to 2.8 mmol/g.As a consequence, the CH 4 selectivity increased from 58.2 to 65.6%.When the CeO 2 loading was increased to 10 wt.%, the gas yields were slightly higher than those obtained with Ni/MCM-41, but lower than those obtained with Ni-5CeO 2 /MCM-41.In contrast, the gas yields obtained with Ni-5La 2 O 3 /MCM-41 and Ni-10La 2 O 3 /MCM-41 were even lower than those obtained with Ni/MCM-41.The H 2 yield increased slightly when Ni-5CeO 2 /MCM-41 was used as a catalyst, suggesting that CO 2 methanation did not occur in this condition.Many recent studies showed that nickel-based catalysts are also capable of reforming tars [14].Therefore, the increased CH 4 yield was mainly due to the decomposition of tarry compounds on the catalyst, instead of the methanation of CO 2 .

Conclusions
In the present work, a series of Ni-based catalysts including Ni/MCM-41, Ni-5CeO2/MCM-41, and Ni-5La2O3/MCM-41 were investigated to improve the product yield and selectivity from biomass pyrolysis.The textural properties, surface element composition, chemical state, crystal structure, and

3. 2 .
Gasification of Corn Stalk 3.2.1.Effect of Temperature and Time on the Gas Yield

Figure 7 .
Figure 7. Influence of temperature and time on the gas yields.

Figure 7 .
Figure 7. Influence of temperature and time on the gas yields.(a) Temperature, (b) Time.

3. 2 . 2 .
Effect of Catalyst on the Gas Yield A series of monometallic catalysts, including Ni-MCM-41, 5CeO 2 -MCM-41, Ni-Al 2 O 3 , 5CeO 2 / Al 2 O 3 , and 5La 2 O 3 /Al 2 O 3 , were investigated for the catalytic reforming of pyrolysis gases.As shown in Figure 8a, the gas yields and CH 4 selectivity obtained with Ni/MCM-41 and Ni/Al 2 O 3 were slightly higher than those obtained in the absence of the catalyst, indicating the catalytic activity of the Ni species.However, both the gas yields and CH 4 selectivity obtained with 5CeO 2 /MCM-41 and 5La 2 O 3 /Al 2 O 3 were not obviously improved when compared to the control group, indicating that these materials were incapable of improving the quality of gas products.In addition, a series of bimetallic catalysts, including Ni-5CeO 2 /MCM-41, Ni-10CeO 2 /MCM-41, Ni-5La 2 O 3 /MCM-41, and Ni-10La 2 O 3 /MCM-41, were tested to further improve the catalytic performance.When Ni-5CeO 2 /MCM-41 was used as a catalyst (Figure