Nitrous oxide (N2O) destroys the ozone in the stratosphere and it is a strong greenhouse gas [1, 2]. The continuous increase of its concentration in the atmosphere is mainly from the tail gas of adipic acid and nitric acid plants [3]. This calls for the developing of efficient catalysts for its decomposition into N2 and O2.
In recent years, 3d transition metal oxides such as Co3O4 [4, 5, 6, 7, 8, 9, 10, 11, 12], NiO [13, 14], Fe3O4 [15], and CuO [16] have been shown to exhibit high activity for N2O decomposition. Fundamental and applied research has focused on these materials, in particular on Co3O4 based materials. Many additives have been studied to improve the catalytic activity of Co3O4. The alkali metals Na+, K+, Cs+ [4, 5, 9, 10, 11] and alkaline earth metals Mg2+ [8], Ca2+, Ba2+[12] and some other transition metals like Ce [6], Ni, Zn [7, 8] have been reported to be effective. On the other hand, although NiO has a comparable catalytic activity to Co3O4 [11], fewer additives for it have been investigated, apart form Cs [13] and Ce [14]. Concerning the mechanism of the additive promotion of the NiO catalyst, it was reported that Cs [13] significantly facilitated the desorption of the oxygen produced from N2O decomposition by weakening the Ni-O bond. In the case of Ce as additive, the promotional effect was attributed to a significant increase of the catalyst surface area since it did not much strengthen the Ni-O bond [14]. This influence of Ce on NiO was confirmed in our previous work [17], and on this basis, we suggested that BaCO3 not only played the same role as Cs, but also increased the surface area of the catalyst.
In the present work, a Co-modified NiO-BaCO3 catalyst was studied, and a significant promotional effect by Co on the activity of NiO in the presence of BaCO3 was found.
All catalysts were prepared by the co-precipitation method with the following procedure. A Na2CO3 aqueous solution (0.2 mol/L) was added dropwise to a mixed solution containing known amounts of Ni(NO3)2·6H2O and Ba(NO3)2 at 40 °C with strong stirring until the pH of the solution was 9.3. The slurry was stirred for an additional 2 h before it was filtered. Then the resultant precipitate was washed with distilled water until the pH of the filtrate reached 7. This sample was dried at 100 °C overnight, followed by calcination at 500 °C in air for 3 h. The catalysts were labeled as CoxNi9 or CoxBa1.5Ni9 according to the mole ratios of Co(NO3)2·6H2O and Ba(NO3)2 to Ni(NO3)2·6H2O in the mixed solution used for the catalyst preparation.
The characterization of the catalysts with X-ray diffraction, H2 temperature-programmed reduction, O2 temperature- programmed desorption, N2 adsorption-desorption for measuring the BET surface area, and the activity tests were the same as our previous work [17].
Figure 1 presents the XRD patterns of the catalysts. No diffraction peak belonging to Co3O4 or other cobalt oxide was observed on the Co1.0Ni9 and Co1.0Ba1.5Ni9 catalysts, indicating that the cobalt was well dispersed on them. Moreover, the diffraction peaks of NiO for the Co1.0Ni9 and CoxBa1.5Ni9 catalysts were obviously shifted to lower angles as compared to pure NiO and Ba1.5Ni9, respectively. This resulted from the larger crystalline interplanar spacing of the (111), (200), (220), and (311) planes for Co1.0Ni9 and CoxBa1.5Ni9 with respect to their counterparts (Table 1). This is consistent with the radius of Co2+ (72 pm) > Ni2+ (69 pm). The result indicated that Co2+ was incorporated into the NiO phase and partially replaced Ni2+ in the Co1.0Ni9 and Co1.0Ba1.5Ni9 catalysts. Table 1 shows the BET surface areas of the catalyst samples. The surface area drastically decreased due to the formation of a solid solution (Co dissolved into NiO) in Co1.0Ni9. Interestingly, when BaCO3 was in the catalyst, the opposite occurred. For Ba1.5Ni9 and Co1.0Ba1.5Ni9, the specific surface areas were 32 and 41 m2/g, respectively, indicating that Co was effective for increasing the surface area of the NiO-BaCO3 catalyst.
Figure 2 exhibits the H2-TPR profiles of the NiO, Co1.0Ni9, Ba1.5Ni9, and Co1.0Ba1.5Ni9 catalysts. Three H2 consumption peaks at 360, 408, and 435 °C appeared for pure NiO. The two peaks at the lower temperature were associated with the NiO reduction steps of NiO→Niδ+→Ni0.33 [14], while that at the higher temperature was due to the reduction of NiO and Niδ+ inside larger NiO crystallines. Over Ba1.5Ni9, the last peak nearly did not appeared, which was associated with the small crystallite size of NiO in the sample (Table 1). Compared to the pure NiO, all of the H2 consumption peaks of Co1.0Ni9 were shifted to higher temperature, which reflects that Co in the solid solution suppressed the reduction of NiO. Nevertheless, the shift of the reduction peaks on Co1.0Ba1.5Ni9 with respect to Ba1.5Ni9 was less than that due to the presence of BaCO3. This means that the Ni-O bond strengthening caused by Co was effectively reduced by BaCO3 in the catalyst.
Figure 3 shows the O2-TPD profiles of the catalysts. O2 desorption started at 165 °C over pure NiO, while the starting temperature for O2 desorption over Co1.0Ni9 was much increased to 217 °C. For the O2-TPD profiles of Co1.0Ba1.5Ni9 and Ba1.5Ni9 catalysts, their temperatures were quite close. Clearly, all the phenomena were consistent with the H2-TPR results. Both indicated that the unfavorable effect of Co in suppressing O2 desorption was greatly reduced when there was BaCO3 in the catalyst. For N2O decomposition, the catalytic cycle can be represented as shown in Fig. 4. The desorption of oxygen is widely accepted to be the rate determining step [19]. Thus the oxygen produced by N2O decomposition on the active nickel sites (*) must be released at or below the reaction temperature, otherwise they would poison the active sites for the reaction. Hence, the active sites on each catalyst at 300 °C were calculated from the O2 desorption area below 300 °C. As shown in Table 2, the active sites over NiO, Co1.0Ni9, Ba1.5Ni9, and Co1.0Ba1.5Ni9 at the reaction temperature were 65.07, 52.17, 95.67, and 117.47 μmol/g, respectively, which was basically in proportion to the surface area of the catalysts (29, 9, 32, and 41 m2/g). This indicated again that Co gave the NiO based catalyst a special property, i.e. increasing its active sites, only when BaCO3 also existed in the catalyst. As shown, the active sites decreased with the surface area due to the Co incorporation when BaCO3 was absent from the NiO catalyst.
The conversion of N2O at each reaction temperature over the CoxBa1.5Ni9 (x = 0, 0.5, 1.0, 1.5, 2.0) catalysts are shown in Fig. 5. Interestingly, the Co1.0Ba1.5Ni9 catalyst was much more active than Ba1.5Ni9 in the temperature range of 200-350 °C, regardless of whether the feed gas was with or without 5% O2. Clearly, the superior catalytic activity of Co1.0Ba1.5Ni9 can be attributed to the large increase of active sites caused by Co. On the other hand, comparing the conversion of N2O over pure NiO and the CoxNi9 (x = 1.0, 2.0, 3.0, 4.0) catalysts at 350 °C (Fig. 6) showed that all the CoxNi9 catalysts displayed inferior activity to pure NiO, and inferior BET surface area and active site amounts.
When an appropriate amount of cobalt as additive was incorporated into NiO-BaCO3, a promotion of the catalytic activity for N2O decomposition was observed. This was due to the increased surface area and active sites of the catalyst. However, introducing Co into pure NiO without BaCO3 on the contrary decreased the surface area, amount of active sites, and activity of the catalyst. Hence, for the promotional effect of Co on the NiO-BaCO3 catalyst to occur, BaCO3 played a significant role.
2015, Vol. 36 Issue (3): 344-347
PDF (520 KB)
2015, Vol. 36 Issue (3): 344-347
PDF (520 KB)