The dry reforming of methane (DRM) has attracted considerable attention because CO2 is the primary greenhouse that has brought about significant climate change, and methane as the main component of natural gas used in combustion leads to a great deal of CO2 emission. Thus, the DRM reaction converts two greenhouse gases to the industrially important synthesis gas (H2 and CO), which serves as the platform for downstream liquid fuels and valuable chemicals [1-8].
Although the noble metal catalysts have exhibited high catalytic performance and low carbon deposition, their high cost and low availability limit their large scale application in industry [9-13]. Ni-based catalysts have been established as potential catalysts for industrial application in synthesis gas production from the DRM reaction owing to their low cost and comparable activity and selectivity to noble metal catalysts. Unfortunately, Ni-based catalysts have a bottleneck issue of rapid deactivation due to the carbon deposition as well as the sintering and oxidation of metallic Ni under the harsh reaction conditions [14-22]. Among these, carbon deposition has been accepted as the main reason of catalyst deactivation [23-25]. Many reports have focused on retarding coke deposition [7, 26-31]. Typically, the coke can be efficiently suppressed by Ni-support interface tuning [32] and enhancing heat transfer [33, 34]. There is agreement in the references that the coke originates mainly from the two reactions of the decomposition of methane and the Boudouard reaction of CO, which are strongly dependent on the nature of support.
In the references [35-39], ZrO2 has been widely used as the catalyst support for diverse transformations including DRM, owing to its high thermal stability, unique acid-base properties and surface oxygen mobility. It was found that the crystalline phase, crystalline size, and pore structure of the ZrO2 support significantly affected the catalytic performance of the supported catalysts [40-42]. Zheng et al. [43] compared the catalytic performance of supported Ni catalysts on a ZrO2 prepared by flowing nitrogen (ZrO2-AN) and air (ZrO2-CP). It was found that Ni/ZrO2-AN exhibited much higher stability than Ni/ZrO2-CP, which was ascribed to the different nature but not to the amount of coke. Change et al. [44] confirmed that the simultaneous alteration of the support and Ni surface of Ni/ZrO2 with a Ce modifier and a Ca promoter led to a high performance catalyst, mainly ascribed to a synergy effect of the Ca promoter and Ce modifier for coke resistance and high temperature catalyst aging. Gonzalez-Delacruz et al. [3] found that the size of metallic Ni of the Ni/ZrO2 catalyst can be efficiently tuned by a H2/CO treatment, which significantly affects the catalytic performance for DRM. Xu et al. [45] explored the size effect of ZrO2 support particles on the catalyst and catalytic performance of a Ni/ZrO2 catalyst for the utilization of natural gas through dry reforming by CO2, and found that the supported Ni catalyst on the ZrO2 support with a smaller particle size exhibited much higher catalytic stability, ascribed to the enhancing effect of reducing the particle size on oxygen transfer ability. It was found that the shape of the supports (e.g., CeO2, ZnO, and La2O3) strongly affected their catalytic performance [47-49]. In many cases, the morphology effect of CeO2 and doped CeO2 has been widely used on a catalyst support to promote the activation of CO2 and oxidative removal of surface carbon species [7, 46, 49-51]. However, to the best our knowledge, few reports regarding the morphology effect of the ZrO2 support on Ni-based catalysts for DRM reaction can be found.
Herein, we successfully prepared two m-ZrO2 supports with different morphologies including an immature pinecone shaped hierarchically structured zirconia (ZrO2-ipch) and a cobblestone-like zirconia nanoparticulate (ZrO2-cs), and tested the catalytic performance of the supported Ni catalysts (Ni/ZrO2-ipch and Ni/ZrO2-cs) on the two supports for the DRM reaction. Ni/ZrO2-ip was also prepared and tested for comparison. It was found that the Ni/ZrO2-ipch catalyst exhibited much higher catalytic activity than Ni/ZrO2-cs or Ni/ZrO2-ip, and it also exhibited higher coke-resistant stability. Characterization techniques including X-ray diffraction (XRD), N2 adsorption measurement (BET), transmission electron microscopy (TEM), temperature-programmed reduction of hydrogen (H2-TPR), CO chemisorption, temperature-programmed desorption of CO2 (CO2-TPD), X-ray photoelectron spectroscopy (XPS) and thermogravimetric analysis (TGA) were employed to investigate the morphology effect of the ZrO2 support on the supported Ni catalyst for DRM. The origin of the high activity and coke-resistant stability of Ni/ZrO2-ipch was explored.
All chemicals (Sinopharm Chemical Reagent Co., Ltd) were used as purchased without further purification. For ZrO2-ipch, 3.22 g (7.5 mmol) of Zr (NO3)4.5H2O was dissolved in de-ionized water under magnetic stiring, followed by adding 1.5 g (10.7 mmol) of hexamethylenetetramine aqueous solution to the Zr (NO3)4.5H2O aqueous solution. Then NH4F was introduced into the above mixture. The resulting solution was transferred into a Teflon-lined autoclave and hydrothermally treated for 24 h. After the autoclave was cooled down to room temperature, the suspension was separated by filtration, washed with de-ionized water several times, and then dried at 105 ℃ overnight. The solid was subsequently calcinated at 550 ℃ for 5 h in a muffle furnace, and ZrO2-ipch was otained. For ZrO2-cs, asolution containing 3.22 g (7.5 mmol) of Zr (NO3)4.5H2O was transferred into a Teflon-lined autoclave and was hydrothermally treated at the same conditions as above. After the autoclave was naturally cooled down to room temperature (RT), the solid was recovered by filtration, washed with de-ionized water several times, and then dried at 105 ℃ overnight. The solid was calcined at 550 ℃ for 5 h in a muffle furnace, and the ZrO2-cs sample was obtained. For comparison, a traditional ZrO2-ip support was also prepared by the common precipitation method by using 35% ammonia as precipitant.
The supported Ni catalysts on the ZrO2-ipch, ZrO2-cs and ZrO2-ip supports (10% Ni loading) were prepared by using the previously established LA-IWI method [52]. The NiO/ZrO2- ipch, NiO/ZrO2-cs and NiO/ZrO2-ip samples were reduced at 850 ℃ for 2 h under a mixture of 20% H2 in 80% N2 at a flow rate of 30 mL/min to obtain the reduced samples, which were denoted as Ni/ZrO2-ipch, Ni/ZrO2-cs and Ni/ZrO2-ip, respectively.
TEM images were obtained by using a Tecnai F30 HRTEM instrument (FEI Corp.) at an acceleration voltage of 300 kV. The sample was dispersed in ethanol with ultrasonic treatment for 10 min, and a drop of the suspension was placed on a copper grid for TEM observation. FE-SEM experiments were performed on a JEOL JSM-5600LV SEM/EDX instrument.
Nitrogen adsorption and desorption isotherms were determined on a Beishide apparatus Model 3H-2000PS1 at -196 ℃. The specific surface areas were calculated by the Brunauer-Emmett-Teller (BET) method, and the mesopore size distribution was calculated from the adsorption branch of the isotherm by the BJH model.
XRD patterns were collected from 10° to 80° at a step width of 0.02° using a Rigaku Automatic X-ray Diffractometer (D/Max 2400) equipped with a Cu Kα source (λ=0.15406 nm). The average crystallite size was estimated on the basis of the Scherrer formula from the characteristic diffraction peaks by the MDI Jade 5 software.
TGA analysis was conducted to study the amount of coke deposited on the spent ZrO2-ipch and ZrO2-ip catalysts using a Perkin-Elmer STA 6000 with a heating rate of 10 ℃/min from 30 to 800 ℃ in an air stream.
The Ni dispersion of the two catalysts was measured by CO titration at 35 ℃ using a Builder PCA-1200 Pulsar TPR/TPD equipment by pulse chemisorption (300 mL for each pulse, 10% CO/90% He, purity of He is 99.999%). The catalyst (~100 mg) was reduced in situ with H2 at 850 ℃ for 1.5 h, then flushed at 400 ℃ with He for 50 min. After pre-reduction, the catalyst was cooled to 30 ℃ and CO chemisorption was carried out. The Ni dispersion was calculated by
where DM, VS, SF, SW and MW were metal dispersion (%), volume of active gas chemisorbed (cm3 at STP), stoichiometric factor, catalyst mass (g), and molecular mass of metallic Ni (g/mol), respectively [53].
H2-TPR experiments were performed on a Builder PCA-1200 automated system. 100 mg of catalyst was placed in a U-shape quartz tube in a temperature-controlled oven. The catalyst was first purged under 30 cm3/min of Ar flow at 300 ℃ for 30 min (using a ramp rate of 10 ℃/min) and then cooled down to RT. After that, it was reduced with a 10% H2-90% Ar mixture (30 mL/min) by heating to 800 ℃ at a ramp rate of 10 ℃/min. The amount of hydrogen consumption was measured using a thermal conductivity detector.
CO2-TPD was performed using a Builder PCA-1200 automated instrument. The catalyst was pre-reduced under a 10% H2/Ar mixture gas flow (30 mL/min) by ramping to 800 ℃ (10 ℃/min). Then, the system was purged with He at 800 ℃ for 30 min, and then cooled down to RT. The adsorption of CO2 was performed at 50 ℃ in a pure CO2 (99.999%) flow with a rate of 30 mL/min, and the catalyst was purged with a He stream at 50 ℃. After that, the system was cooled down to RT in a He stream until the baseline was steady. Finally, CO2-TPD was performed with a ramp of 10 mL/min from the RT to 800 ℃ in a 30 mL/minHe stream.
The XPS experiments were carried out on an ESCALAB 250 XPS system with a monochromatized Al Ka X-ray source (15 kV, 150 W, 500 μm, pass energy=50 eV). The binding energy was calibrated by the C 1s photoelectron peak at 284.6 eV. The binding energy of O 1s was analyzed using the Shirley baseline correction method.
The catalytic performance measurement of the catalysts was performed in a quartz tube fixed-bed reactor (6 mm O.D. and 4 mm I. D.) at atmospheric pressure. Typically, 30 mg catalyst with the 40-60 mesh particle size was loaded between two quartz wool plugs. The temperatures were measured using K Type thermocouples and controlled by a PID controller. Before the reaction, the catalyst was reduced at 850 ℃ for 2 h using a mixture of 20% H2 in 80% N2 at a flow rate of 30 mL/min. The reaction feed contained CH4, CO2 and N2 (N2 was used as the internal standard gas), which was controlled by mass flow controllers. The reforming reaction was performed at 650, 750 and 850 ℃, and the N2 was used an internal standard to calculate the CH4 and CO2 conversion. The effluent gas was passed through a trap and then analyzed by a gas chromatograph online with a molecular sieve column and a Porapak Q column. The stability was tested by using the following conditions: mcat=30 mg, CH4:CO2:N2=1:1:3, GHSV=100000 mL h-1 g-1, T=850 ℃ and atmospheric pressure. The conversions of CH4 (XCH4) and CO2 (XCO2), the selectivity of H2 (SH2), as well as H2/CO ratio were calculated as follows:
The morphology of the three synthesized ZrO2 supports was examined by FE-SEM. Fig. (1a-c) displays typical FE-SEM images. From Fig. 1(a), a panoramic view revealed particles with different shapes including rice-like ones for ZrO2-ipch. The enlarged FE-SEM image of ZrO2-ipch (inset in Fig. 1(a)) further suggested that ZrO2-ipch has an immature pinecone shaped hierarchical structure composed of ZrO2 nanoparticles. However, ZrO2-cs has cobblestone-like agglomerated ZrO2 nanoparticles (Fig. 1(b)). Fig. 1(c) shows that ZrO2-ip has an aggregate shape of iregular particles. The TEM images of ZrO2-ipch, ZrO2-cs and ZrO2-ip (Fig. 1(d-f)) present further views of their morphology. From Fig. 1(d), the hierarchical structure feature with mesopores of ZrO2-ipch can be further confirmed. More interestingly, in Fig. 1(e), the nanopores (less than 20 nm) of the cobblestone-like agglomerated ZrO2 nanoparticles can be clearly seen. From Fig. 1(f), the structural characteristic of aggregation of ZrO2-ip composed of irreggular particles can be further confirmed. From the above, it was confirmed that an immature pinecone shaped hierarchical structure with mesopores (ZrO2-ipch) and a cobblestone-like agglomerated ZrO2 nanoparticles with nanopores in the nanoparticles (ZrO2-cs) were successfully prepared. The textural properties of the two synthesized ZrO2 supports were measured by nitrogen adsorption.
Fig. 2 presents the nitrogen adsorption of ZrO2-ipch, ZrO2-cs and ZrO2-ip (the insets in Fig. 2(a) are specific surface area and pore volume). The most probable pore size are listed in Table 1. In Fig. 2(a), the isotherm of ZrO2-ipch is of type IV with a H3 hysteresis loop, indicating the presence of narrow slit-shaped mesopores (less than 20 nm) [54, 55], which agreed with the TEM observation (Fig. 1(a)). The isotherm corresponding to the ZrO2-cs support showed the type IV feature with a H3 hysteresis loop. However, in comparison with the isotherm of ZrO2-ipch, there was a capillary condensation step at 0.9-0.95 in the high relative pressure region, indicating the formation of macropores [56]. In Fig. 2(b) and in Table 1, the mesopores with the size 4 nm can be ascribed to the pores of the cobblestone-like nanoparticulates shown in Fig. 1(b), whereas the macropores with the size of 20-70 nm with bimodal pores (12.4 and 42.0 nm) resulted from the pores of agglomerated cobblestone-like nanoparticles. ZrO2-ip has the most probable pore diameter of 12.8 nm. The hierarchically structured ZrO2-ipch has a lower surface area (15.3 m2/g, the inset in Fig. 2(a)) and smaller pore volume (0.07 cm3/g) than those of ZrO2-cs (33.2 m2/g and 0.29 cm3/g, respectively) and ZrO2-ip (39.5 m2/g and 0.16 cm3/g, respectively), which might be ascribed to the denser accumulation of nanoparticles in the hierarchical structure.
The crystalline structure of the three synthesized supports was investigated by XRD analysis. Fig. 3 presents XRD patterns of the ZrO2-ipch, ZrO2-cs and ZrO2-ip supports. From Fig. 3, the m-phase ZrO2 can be confirmed for both of them (JCPDS 37-1484) [57]. The ZrO2-ipch demonstrated a higher intensity of the XRD peaks than ZrO2-cs, indicating higher crystallinity. The weaker and broader XRD peaks for ZrO2-cs than those for ZrO2-ipch suggest a smaller average crystalline size. The narrowest XRD peaks were observed with ZrO2-ip, indicating the largest average crystallite size. The average crystalline sizes of m-ZrO2 for the two supports were estimated by the Scherrer equation from the peaks for the (111) and (1111-) planes of monoclinic ZrO2 (Table 1). The average crystalline size of m-ZrO2 for the ZrO2-ipch support (13.6 nm) is larger than that for the ZrO2-cs support (12.1 nm). However, these two crystallite sizes were smaller than that of ZrO2-ip (14.0 nm). A high thermal stability is indispensable for the catalyst support to satisfy the requirement imposed by the harsh conditions for the DRM reaction. Therefore, we compared the thermal stability of the two ZrO2 supports by performing an XRD analysis on the ZrO2 supports after calcinating at 850 ℃. The estimated average crystalline sizes of the m-ZrO2 for the two supports calcined at 850 ℃ are listed in Table 1. From Table 1, the crystalline size of m-ZrO2 for ZrO2-cs increased from 12.1 to 21.9 nm (81.0% growth, the percentage of the increased size over the original size), suggesting its poor thermal stability. However, only 26.5% growth (from 13.6 to 17.2 nm) for the ZrO2-ipch support and 20.0% of growth (from 14.0 to 16.8 nm) for ZrO2-cs were observed. This revealed that both ZrO2-ipch and ZrO2-ip have superior thermal stability in comparison to the ZrO2-cs sample.
The initial catalytic performance evaluation of Ni/ZrO2-ipch, ZrO2-cs and Ni/ZrO2-ip catalysts was performed to explore the effect of the morphology of the support (Fig. 4). From Fig. 4(a, b), it can be observed that the CH4 and CO2 rates of all the supported Ni catalysts were strongly dependent on the reaction temperature, indicating the thermodynamic control of the DRM. The order of the CH4 and CO2 rates was Ni/ZrO2-ipch > Ni/ZrO2-cs > Ni/ZrO2-ip. Moreover, it can be observed that the CO2 rate was higher than CH4 rate over the catalysts, especially at lower reaction temperatures, which might be due to the simultaneous occurrence of the reverse water gas shift (RWGS) reaction [58]. The RWGS reaction results in the H2/CO ratio being less than 1 (Fig. 4(d)). From Fig. 4(c, d), an increasing reaction temperature does not result in a visible change in H2 selectvity, but led to an increase in the H2/CO ratio, which might be attributed to an enhanced water gas shift (WGS) reaction by the increased temperature.
Table 2 shows that the Ni/ZrO2-ipch catalyst has a lower surface area (17 m2/g) than Ni/ZrO2-cs (20 m2/g) and Ni/ZrO2-ip (32 m2/g). Generally, supported Ni catalysts show a lower surface area than their corresponding support. Interestingly, the Ni/ZrO2-ipch catalyst showed a higher specific surface area than the ZrO2-ipch support (15.3 m2/g), which was ascribed to a high Ni dispersion and higher thermal stability of the ZrO2-ipch support (Table 1). On correlating the surface area data to reaction results, in comparison with Ni/ZrO2-cs or Ni/ZrO2-ip, the lower surface area of the Ni/ZrO2-ipch catalyst did not reduce the rate for CH4 and CO2 transformation, implying that there were other aspects that resulted in its higher catalytic activity. Therefore, XRD, CO chemisorption, H2-TPR, CO2-TPD and XPS experiments were further performed to reveal the origin for the higher catalytic activity of the Ni/ZrO2-ipch catalyst in comparison with Ni/ZrO2-cs or Ni/ZrO2-ip for the DRM reaction.
Fig. 5 presents the XRD patterns of the Ni/ZrO2-ipch, Ni/ZrO2-cs and Ni/ZrO2-ip catalysts. In Fig. 5, the XRD peaks appearing at 44.5°, 51.8° and 76.4° can be well resolved, and were assigned to the reflections of the (111), (200) and (220) planes of the face centered cubic (fcc) Ni phase (JCPDS No. 65-2865), respectively. The weaker and broader XRD peaks corresponding to the (111), (200) and (220) planes of Ni for the Ni/ZrO2-ipch catalyst in comparison with those for Ni/ZrO2-cs or Ni/ZrO2-ip can be clearly seen, which suggested a better dispersion of Ni on ZrO2-ipch. The average crystalline size of Ni for the two catalysts was estimated by the Scherrer equation based on the above three typical reflections planes. The results are listed in Table 2. From Table 2, the 12.7, 13.3 and 20.4 nm of the average crystalline size of Ni for Ni/ZrO2-ipch, Ni/ZrO2-cs and Ni/ZrO2-ip, respectively, can be seen. The smaller average crystalline size of Ni for Ni/ZrO2-ipch may be one reason for its higher catalytic acivity for DRM in comparison with Ni/ZrO2-cs or Ni/ZrO2-ip catalyst. CO chemisorption was performed to measure the Ni dispersion. The chemisorption stoichimetry of CO to Ni was assumed to be 1 [52]. The results are listed in Table 2. The higher Ni dispersion (5.3%) of the Ni/ZrO2-ipch catalyst than that of Ni/ZrO2-cs (3.8%) and Ni/ZrO2-ip (3.1%) can be observed, revealing the effect of the morphology of the support on Ni dispersion of a supported Ni catalyst. The higher Ni dispersion of the Ni/ZrO2-ipch catalyst endowed it with a higher catalytic activity for the DRM by the increased active sites.
The H2-TPR technique is a powerful tool for unveiling the reducibility and metal-support interaction of supported metal catalysts. Fig. 6 presents the H2-TPR profiles for the Ni/ZrO2-ipch, Ni/ZrO2-cs and Ni/ZrO2-ip catalysts. The quantitative analytical results are listed in Table 2. In Fig. 6, the H2-TPR profiles for the two catalysts can be roughly divided into two reduction regions (region Ⅰ and region Ⅱ). Region Ⅰ can be assigned to the reduction of free NiO on the support surface and the reduction of NiO with a weak metal-support interaction [59-61], whereas region Ⅱ can be assigned to the reduction of NiO with a strong Ni-ZrO2 interaction [52]. The initial reduction temperature for region Ⅰ corresponding to the Ni/ZrO2-ipch catalyst was shifted to a lower temperature, which might be ascribed to the reduction of surface oxygen species. From Table 2, the H2 uptake for region Ⅱ of the Ni/ZrO2-ipch catalyst was much higher than that of the Ni/ZrO2-cs catalyst, indicating more NiO with a strong metal-support interaction. Interestingly, no peak in region Ⅱ from Ni/ZrO2-ip was observed, suggesting the weakest Ni-support interaction. A strong Ni-ZrO2 interaction is favorable for high Ni sintering resistance, which would improve the catalytic stability for DRM at high temperature. Moreover, from Table 2, the Ni/ZrO2-ipch catalyst exhibited a higher reduction degree (the percentage of total H2 consumption to the nominal H2 uptake for NiO complete reduction, 38.1%) in comparison with that of the Ni/ZrO2-cs (35.8%) and Ni/ZrO2-ip (31.6%) catalysts, which allowed the Ni/ZrO2-ipch catalyst to exhibit superior catalytic activity to the other two catalysts.
CO2-TPD was conducted to investigate the surface basic properties of the catalysts, which signifiantly affects the adsorption and activation of CO2 [62-64]. Fig. 7 presents the CO2-TPD profiles, and the quantitative results are listed in Table 2. In Fig. 7, besides the desorption peak at 129 ℃ (weak basic sites) in the CO2-TPD profile of the Ni/ZrO2-ipch catalyst, another peak at 247 ℃ can be clearly observed. However, only one peak occurred at 112 ℃ in the CO2-TPD profile of Ni/ZrO2-cs and at 113 ℃ in the CO2-TPD profile of the Ni/ZrO2-ip catalyst. This result illustrated that the Ni/ZrO2-ipch catalyst has strong basic sites in addition to the existing weak ones. From Table 2, the total desorbed CO2 from the Ni/ZrO2-ipch catalyst (64.6 µmol/g) was more than that of the Ni/ZrO2-cs (57.2 µmol/g) and Ni/ZrO2-ip catalysts (36.7 µmol/g). In comparison with the Ni/ZrO2-cs catalyst, more and stronger basic sites can promote the adsorption and activation of CO2, which can efficiently enhance the DRM reaction. From the references [65, 66], lattice oxygen (O2-) can activate the C-H bond and also promote coke gasification.
The surface oxygen species were monitored by XPS analysis. Fig. 8 displays the O 1s region of the XPS spectra for the Ni/ZrO2-ipch, Ni/ZrO2-cs and Ni/ZrO2-ip catalysts. In Fig. 8, the O 1s spectra of the two catalysts can be deconvoluted into two peaks. The peaks at 530.1 and 531.4 eV can be assigned to lattice oxygen (O2-) and adsorbed oxygen, respectively [55, 56]. The percentage of lattice oxygen and adsorbed oxygen to the total amount of O, Ni, and Zr is summarized in Table 3. It can be observed that the percentage of the lattice oxygen on Ni/ZrO2-ipch (49.7%) was much higher than that on the Ni/ZrO2-cs (30.7%) and Ni/ZrO2-ip (23.4%) catalysts. The higher percentage of lattice oxygen allows Ni/ZrO2-ipch to exhibit higher catalytic activity and coke resistant stability [65, 66].
Deactivation by coke is still the key problem to be solved for the industrial application of Ni-based catalysts for the DRM reaction. Therefore, the development of stable and coke resistant catalysts is a major concern. Here, the morphology effect of the support on the catalytic stability of the supported Ni catalysts for the DRM reaction was investigated. The results are displayed in Fig. 9. From Fig. 9, the Ni/ZrO2-ipch catalyst exhibited higher catalytic activity and stability in comparison with the Ni/ZrO2-ip catalyst during 30 h of time on stream reaction. From the above analysis, a higher activity of the Ni/ZrO2-ipch catalyst in comparison with Ni/ZrO2-ip can be ascribed to the higher Ni dispersion, reducibility, percentage of lattice oxygen, and more basic sites with higher strength. From Fig. 9(c, d), the two catalysts have similar H2 selectivity and H2/CO ratio. The Ni/ZrO2-ipch catalyst exhibited outstanding catalytic stability including the converison of CH4 and CO2, H2 selectivity and H2/CO for DRM reaction, which makes it a promosing catalyst for synthesis gas production through the DRM reaction.
The XRD, TEM and TGA techniques were used to unveil the reason for the outstanding catalytic stability of the Ni/ZrO2-ipch catalyst. Fig. 10 displays the XRD patterns of the fresh and spent Ni/ZrO2-ipch and Ni/ZrO2-ip catalysts. The average crystalline size of Ni was estimated by the Scherrer equation on the basis of the Ni (111), (200) and (220) planes (Table 2). From Fig. 10 and Table 2, no visible Ni sintering on the two catalysts can be observed, suggesting the existence of a strong Ni-ZrO2 interaction on the two catalysts. In addition, compared with the fresh catalysts, the spent Ni/ZrO2-ipch and Ni/ZrO2-ip catalysts show an obvious diffraction peak ascribed to graphitic carbon located at 26° [67]. The stronger diffraction peak corresponding to graphitic carbon from Ni/ZrO2-ip than from Ni/ZrO2-ipch can be observed. Therefore, it can be proposed that the poorer stability of Ni/ZrO2-ip should be ascribed to its weaker coke resistance. TEM and TGA measurements were performed to further investigate the coking behavior of the two catalysts.
Fig. 11 displays typical TEM images of the two spent catalysts. In Fig. 11, coke with a nano-tubular feature on the two catalysts can be observed. Moreover, the amount of this kind of tubular carbon on the spent Ni/ZrO2-ip catalyst was much more than that on the Ni/ZrO2-ipch catalyst. The amount of deposited coke on the two spent catalysts was further investigated by TGA analysis.
Fig. 12 presents the TGA/DTG curves of the spent Ni/ZrO2-ipch and Ni/ZrO2-ip catalysts. The coking rate calculated on the basis of the TGA analysis is summarized in Table 2. From Fig. 12 and Table 2, the Ni/ZrO2-ipch demonstrated a lower mass loss (9.0 wt%) and coking rate (1.95 mg g-1 h-1) than the Ni/ZrO2-ip catalyst (35.1%, 6.19 mg g-1h-1). In comparison with the previously reported Ni/ZrO2-AN catalyst, the developed supported Ni catalyst on ZrO2 exhibited obviously better coking resistance [43]. Even if when comapred with a supported Ni nanosheet catalyst, Ni/ZrO2-ipch still exhibited a much lower coke formation rate [68]. According to the litearure [65, 69], lattice oxygen can react with deposited carbon on the surface of catalysts, and the residual vacancies can be supplied by CO2. Moreover, the promoted adsorption and activation by the more basic sites of Ni/ZrO2-ipch catalyst also can eliminate deposited coke [70, 71]. Furthermore, from Figs. 1 and 2, one can clearly see that the ZrO2-ipch sample contains mesopores with a narrow pore size distribution in the region of 4 nm, which may endow it with good catalytic stability through the stabilization effect on nickel particles against sintering and coke deposition by the confinement effect of the mesopores [33, 69, 72, 73]. The much smaller crystallite size of Ni on Ni/ZrO2-ipch was also one reason for its higher coke tolerance than Ni/ZrO2-ip [17, 62, 74-76]. From Fig. 12(b), only one peak in the DTG curves of the two catalysts can be observed, suggesting only one type of coke on the two catalysts. The lower temperature for the DTG peak on Ni/ZrO2-ipch in comparison with that on Ni/ZrO2-ip indicated the more active coke species on Ni/ZrO2-ipch. The more active the coke is, the easier the coke gasification is. This might also be one reason for the higher coke resistant stability of Ni/ZrO2-ipch during the DRM reaction. Correlating the XPS and CO2-TPD to the TGA results, the more lattice oxygen and surface basic sites on the Ni/ZrO2-ipch catalyst than those on Ni/ZrO2-ip significntly improve the coke resistant ability of Ni/ZrO2-ipch, which endowed Ni/ZrO2-ipch to be a promising catalyst for synthesis gas production through the DRM reaction.
An immature pinecone shaped hierarchically structured zirconia (ZrO2-ipch), agglomerated porous cobblestone-like zirconia (ZrO2-cs) and traditional irregular ZrO2 particles (ZrO2-ip) as supports were synthesized by a facile hydrothermal method. The hierarchically structured Ni/ZrO2-ipch catalyst exhibited much higher activity and stability in comparison with Ni/ZrO2-ip. The catalytic performance was significantly dependent on Ni crystalline size, Ni dispersion, reducibility, lattice oxygen percentage, and basic properties, which were notably affected by the morphology of the support. The higher activity and Ni-and coke resistant stability of the Ni/ZrO2-ipch catalyst makes it a potential candidate for the DRM reaction. Owing to its unique microstructure and high thermal stability, the immature pinecone shaped hierarchically structured zirconia would be a promising support for synthesizing other supported catalysts for diverse transformations, especially for high temperature reactions.
2016,
Vol. 37 Issue (12): 2122-2133 
PDF
2016,
Vol. 37 Issue (12): 2122-2133 PDF