Nitrogen oxides (NOx) are pollutants that adversely affect human health and the environment, which is an important research topic for society’s current environmental ideals. Selective catalytic reduction (SCR) is a promising pathway for the efficient removal of NOx. Zeolites are SCR materials and have received much attention in research and industry [1, 2, 3, 4, 5, 6]. Cu2+-exchanged zeolites such as Cu-ZSM-5 [1] and Cu-Beta [7] exhibit high activities for SCR, but their poor hydrothermal durabilities limit their applications [3, 8]. Cu2+-exchanged chabazites (Cu-CHA) exhibit high activity, selectivity and stability, and Cu-SSZ-13 and Cu-SAPO-34 [3, 9] have received much attention. Such metal-containing zeolites have usually been prepared in several steps, such as hydrothermal synthesis, removal of structure-directing agents, metal ion exchange and calcination. Xiao group [10] and Moliner group [11] recently found that Cu-SSZ-13 and Cu-SAPO-34, each of them with high activity for SCR, could be directly prepared by introducing Cu2+ to the synthesis precursors. This greatly simplified the Cu-zeolite synthesis process. They demonstrated the fabrication and high performance of specific Cu-zeolite catalyst orientations, using a simple synthesis process.
In 1989, the Engelhard company reported titanosilicate ETS-10 (composition of (Na,K)2Si5TiO13·4H2O), with a typical microporous structure [12]. The framework of ETS-10 consisted of [SiO4] tetrahedra and [TiO6] octahedra connected with oxygen bridges. ETS-10 included 12-member and double 7-member rings with relatively larger micropores (0.76 nm x 0.49 nm), and it possessed a high specific surface area and pore volume. The [Si5TiO13]2- building blocks of ETS-10 required two positive charges to balance charge, which resulted in ETS-10 possessing a large ion exchange capacity [13]. ETS-10 has been used in adsorption [14], ion exchange [15], photocatalysis [15, 16], and DeNOx [17, 18] applications. Gervasini et al. [17, 18] reported the catalytic activity of Cu-ETS-10 for NOx reduction by C2H4, but for NH3-SCR the literatures are not seen. It is also important to understand the nature of Cu species within Cu-ETS-10. Cu-SAPO-34 and Cu-SSZ-13 [9, 10, 19, 20, 21] have received much attention in recent years. In contrast to these two catalysts, ETS-10 can be prepared without directing agents, so the removal of such agents during ETS-10 preparation was not necessary. The ETS-10 fabrication cost should therefore be lower than that of SAPO-34 and SSZ-13. The large ion exchange capacity of ETS-10 means that greater Cu loadings can be achieved.
Cu2+ species within Cu-zeolite micropores participate in charge balance, and reportedly may also be active sites for NH3-SCR, since the Cu2+ species content affects the catalytic performance [22]. Gervasini et al. [17] found that [TiO6] octahedra benefited the stabilization of isolated Cu2+ species in ETS-10 micropores. Therefore, Cu-ETS-10 might be expected to exhibit high activity for NH3-SCR.
In the current study, ion exchange method was used to fabricate Cu-ETS-10 catalysts with different Cu contents, and the NH3-SCR activity and selectivity for N2 over the catalysts were measured. The physiochemical properties of the catalysts were characterized by X-ray diffraction (XRD), N2 sorption measurements, inductively coupled plasma atomic emission spectrometry (ICP-AES), H2 temperature-programmed reduction (H2-TPR), and diffuse reflectance infrared Fourier transform spectroscopy of adsorbed CO (CO-DRIFTS).
ETS-10 was synthesized by a hydrothermal process [12, 13, 16]. Specifically, 6.6 g (165 mmol) of NaOH was dissolved in 50 mL of deionized water at 40 °C, and 7.37 g of white carbon black was added. The solution was stirred for 4 h. 1.82 mL of TiCl4 (16.6 mmol) was added to 10 mL of water at 0 °C to form a white emulsion, which was added to the above solution. Then 4.6 g of KF·2H2O (50 mmol) was added, and the solution was stirred for 0.5 h. The resulting mixture was transferred to a polytetrafluoroethylene-lined autoclave, which was sealed and heated at 180 °C in an oven for 72 h. The product was obtained after cooling, filtering, washing, and drying.
Cations including Na+ and K+ existed in the pores of ETS-10 to balance the negative charge of the framework, but these could be substituted. Samples exchanged with NH4+ and Cu2+ were denoted as NH4-ETS-10 and Cu-ETS-10, respectively. NH4-ETS-10 was synthesized using a reported procedure [8]. Specifically, 20 g of ETS-10 in 1300 mL of NH4NO3 aqueous solution (1 mol/L) was stirred for 10 h at 85 °C. The solution was cooled, and the product collected by filtration and dried. This procedure was repeated six times to obtain NH4-ETS-10 with residual of 0.27% Na2O and 0.012% K2O. The Si/Ti molar ratio of NH4-ETS-10 was 4.7. For Cu-ETS-10 preparation, 10 g of NH4-ETS-10 in 500 mL of Cu(C2H3O2)2 solution (0.1 mol/L) was stirred for 2 h at room temperature. The product Cu-ETS-10-3.9 was collected by filtration, washed, dried for 12 h at 120 °C, and calcined for 4 h in air at 500 °C. Cu-ETS-10-x (where x is the Cu wt%) catalysts were obtained when the ion exchange process in Cu(C2H3O2)2 solution was repeated.
The NH3-SCR performance was investigated using a fixed-bed quartz tube reactor, with 0.2 mL of Cu-ETS-10 (40-60 mesh) catalyst used. The total feed flow was 100 mL/min, which contained NO (0.1%), NH3 (0.1%), O2 (5.0%) and a balance of He with a space velocity (SV) of 30000 h-1. The gas mixture in the reactor outlet included NO, NO2, N2O, and N2 determined by a gas chromatography (GC-2014C, Shimadzu) and a Bruker tensor 27 Fourier transform-infrared (FT-IR) spectrometer equipped with a 2.4 m pathlength gas cell. The NOx conversion (X) [19] and N2 selectivity (SN2) were calculated by:
Powder XRD patterns were recorded using a Bruker D8 Advance diffractometer, operated at 40 kV and 40 mA using Cu Kα radiation (λ = 0.154 nm), at 2θ = 5°-45° with a step size of 0.02°/s. Specific surface areas, pore volumes, and pore size were measured on a physical adsorption instrument (Micrometritics ASAP 2020). Specific surface areas were calculated by Brunauer-Emmett-Teller (BET) method. Each sample was degassed at 250 °C under vacuum for 12 h, and N2 was adsorbed at -196 °C. Na, K, and Cu contents were determined using ICP-AES (Thermol Elemental, IRIS Intrepid ER/S). H2-TPR measurements were performed on a chemical adsorption apparatus (Micrometritics AutoChem II 2920) with a thermal conductivity detector (TCD). 50 mg of catalyst (40-60 mesh) was pretreated in O2 for 30 min at 300 °C, and cooled to 40 °C in Ar. Then 10% H2/Ar was introduced to the tube, which was then heated to 500 °C at a rate of 10 °C/min. After the baseline stabilized, signals were recorded by TCD. CO adsorption was carried out using a FT-IR spectrometer (Nicolet 6700) with a reaction chamber. Spectra were obtained from 32 scans at 4 cm-1 resolution. In a typical experiment, the catalyst was pretreated in 18% O2/He for 30 min at 300 °C, to remove any adsorbed impurities. After cooling to 40 °C, the sample was purged with He for 30 min, and a background spectrum was recorded. The sample was heated to 200 °C, and 5% CO/He was introduced to the chamber, and the spectra were recorded.
The XRD patterns of the as-prepared samples are shown in Fig. 1, in which all peaks could be attributed to the ETS-10 structure. There were no peaks indicating impurities or CuO phases. This implied that Cu species were highly dispersed on the zeolite support. The peak intensities decreased after ion exchange, perhaps associated with a partial collapse of the ETS-10 structure [13, 23].
Figure 2 shows the N2 adsorption-desorption isotherms and pore size distributions of the as-prepared Cu-ETS-10 catalysts. Table 1 summarizes the specific surface areas, pore volumes, and pore sizes of these samples. It shows that the isotherms were all of type I, indicating that all Cu-ETS-10 samples possessed microporous structure. The BET surface areas of Cu-ETS-10-3.9, Cu-ETS-10-6.6, Cu-ETS-10-9.3, and Cu-ETS-1- 12.6 were 379.5, 345.4, 334.5, and 288.4 m2/g, respectively. An increase in Cu content led to a decrease in the surface area of the Cu-ETS-10 samples, because of the partial packing of micropores and partial covering of the support surface. Similar trends were observed in the pore volumes and pore sizes. An increase in Cu content from 3.9 to 12.6 wt% decreased the pore volume from 0.23 to 0.18 cm3/g, and decreased the pore size from 0.91 to 0.87 nm.
Figure 3 shows the NOx conversion and selectivity of N2 over the Cu-ETS-10 catalysts. The Cu-ETS-10-0 catalyst exhibited no activity at 150-400 °C. The NOx conversion over the other catalysts initially increased to a maximum with increasing reaction temperature, and then decreased with further increasing temperature. The maximum conversion of NOx over the Cu-ETS-10-3.9 catalyst was 89% at 250 °C. For the Cu-ETS-10-6.6, Cu-ETS-10-9.3, and Cu-ETS-12.6 catalysts, more than 90% of the NOx could be removed at 216, 202, and 218 °C, respectively. The selectivity of N2 over Cu-ETS-10-3.9 decreased with increasing reaction temperature, while those over Cu-ETS-10-6.6, Cu-ETS-10-9.3, and Cu-ETS-12.6 remained constant, because of their higher Cu contents. Under the same reaction conditions, the Cu-ETS-10-9.3 catalyst exhibited the best catalytic performance. The specific surface areas and microporous structure of the as-prepared Cu-ETS-10 samples had little effect on the catalytic performance for NOx-SCR by NH3.
H2-TPR measurements were carried out to investigate the redox properties of the as-prepared samples. Figure 4 shows H2-TPR profiles of these catalysts. Four different Cu species were observed from the reduction peaks [6, 24] of the Cu-ETS-10 catalysts. Reduction peak at 140-160 °C could have been due to the reduction of amorphous CuOx. Amorphous CuOx may have formed from Cu2+ agglomeration during ion exchange. Such species could be easily reduced at low temperatures, but did little to promote NOx reduction [9]. Reduction peak at ~180 °C was due to the reduction of isolated Cu2+ to Cu+. Isolated Cu2+ species in Cu-zeolites can reportedly be reduced to Cu+ at temperatures < 230 °C. Putluru group [6] found that isolated Cu2+ was reduced at ~180 °C in the H2-TPR of Cu-ZSM-5 and Cu-MOR catalysts. Isolated Cu2+ species in Cu-zeolite catalysts have been reported to be active sites for NH3-SCR [9]. The peak at 200-265 °C may have been attributed to the reduction of nano CuO crystals (nano CuO species) [6, 22]. The peak at temperatures 300 °C above corresponded to the reduction of Cu+, and the interaction between Cu and Ti-O groups [22, 25].
The areas of the reduction peaks in the H2-TPR profiles were used to calculate the H2 consumption of each Cu species in the as-prepared Cu-ETS-10 catalysts. The results are shown in Table 2. No amorphous CuOx species were observed in the Cu-ETS-10-3.9 catalyst, and nano CuO species were the main species, in addition to minor isolated Cu2+ quantities. This suggested that most Cu2+ dispersed on the surface of the support, during the first ion exchange. During calcination, the dispersed species became nano CuO species. The areas of the reduction peaks attributed to the reduction of amorphous CuOx increased with increasing Cu content from 6.6% to 12.6%. This indicated an increase in the amount of amorphous CuOx. Xue et al. [9] reported that no amorphous CuOx existed in the Cu/SAPO-34 catalyst with Cu contents < 2.89%. The peak area attributed to the reduction of isolated Cu2+ species in Cu-ETS-10-6.6 was higher than that in Cu-ETS-10-3.9. Thus, there were more isolated Cu2+ species formed in the Cu-ETS-10-6.6 catalyst during the second ion exchange, and the amount of nano CuO species also increased. Further increasing the Cu content from 6.6% to 9.3% led to more isolated Cu2+ species and nano CuO species. However, there were more nano CuO species and less isolated Cu2+ species in the Cu-ETS-10-12.6 catalyst. The amount of isolated Cu2+ species in the Cu-ETS-10 catalysts may have depended on the following factors. The Cu2+ content introduced within the micropores increased with increasing Cu content. Further increasing the Cu content resulted in destruction of the Cu-ETS-10 structure (Fig. 1), which decreased the amount of exchangeable sites. The interaction of these factors resulted in a maximum possible isolated Cu2+ content. In the present study, this maximum was reached in the Cu-ETS-10-9.3 catalyst.
Figure 5 shows the relationship between the amounts of isolated Cu2+ species and nano CuO species, NOx conversions at 200 and 400 °C, and the total Cu content of the Cu-ETS-10 catalysts. It shows that the amount of nano CuO species increased with increasing Cu content in the Cu-ETS-10 catalysts. However, the isolated Cu2+ content increased to a maximum, and then decreased. The Cu-ETS-10-9.3 sample possessed the highest isolated Cu2+ content. NOx conversions over the Cu-ETS-10 catalysts are also shown in Fig. 5, and Cu-ETS-10-9.3 exhibited the highest activity. The isolated Cu2+ content and activity exhibited broad inverse-parabolic curves, as a function of Cu content. The results suggested that the activity of Cu-ETS-10 was regulated by isolated Cu2+ species [6, 9, 22, 26, 27]. A similar relationship between Cu species and activity was also found in Cu/SAPO-34 catalysts, and isolated Cu2+ species were verified as the active sites of Cu-zeolite catalysts for NH3-SCR [9].
Figure 6 shows CO-DRIFT spectra collected after 5% CO/He was introduced to the reaction chamber for 5 min at 200 °C. The interactions between CO molecules and Cu sites in the zeolites were detected. The H2-TPR results indicated that isolated Cu2+ species could easily be reduced to Cu+. The band at ~2155 cm-1 was assigned to Cu+(CO) (ion exchange sites) [24]. The intensity of this band for the Cu-ETS-10-9.3 sample was slightly stronger than those of the other samples, indicating it had the highest isolated Cu2+ species content. The band at ~2145 cm-1 red-shifted with increasing Cu content in the Cu-ETS-10 catalysts. The band at ~2128 cm-1 may have been attributed to CO adsorbed on reduced Cu+ sites of nano CuO [24, 28]. The results showed that multiple Cu species coexisted in Cu-ETS-10, and that the largest isolated Cu2+ content existed when the Cu content was 9.3%.
We also investigated the influence of SO2 and H2O on the catalytic performance of Cu-ETS-10 for NH3-SCR. The results are shown in Fig. 7. It shows that when SO2 was introduced into the feed gas, as the SO2 content (< 0.1%) and reaction time (< 27 h), the NOx conversion over Cu-ETS-10-9.3 remained at ~95%. When 0.1% SO2 + 5% H2O were added into the feed gas, the NOx conversion rapidly reduced to ~20%. When 0.1% SO2 was cut off, the NOx conversion immediately recovered to ~90%. These results indicated that separately introducing SO2 or H2O had a negligible influence on the catalytic performance of the Cu-ETS-10-9.3 catalyst for NH3-SCR. However, the simultaneous addition of SO2 and H2O caused a drastic decrease in NOx conversion. This may have been associated with competitive adsorption among SO2, H2O, and the reactant at the active sites [29].
0-12.6% Cu-ETS-10 catalysts with microporous structure and large specific surface areas were prepared by ion exchange method. The Cu content was regulated by the times of ion exchange. Cu species were highly dispersed on the zeolite supports. The isolated Cu2+ content affected the catalytic performance of Cu-ETS-10 for the NH3-SCR of NOx. Under a NO:NH3:O2 molar ratio of 1:1:50 and SV of 30000 h−1, the Cu-ETS-10-9.3 catalyst with the highest isolated Cu2+ species content exhibited superior catalytic activity: NOx conversion was > 90% at 200-350 °C. This study on Cu-ETS-10 catalysts will benefit the design and development of SCR catalysts.
2014, Vol. 35 Issue (7): 1030-1035
PDF (567KB)
2014, Vol. 35 Issue (7): 1030-1035
PDF (567KB)