The sulfur tolerance of Ca modified /TiO2 catalysts in low-temperature SCR process was investigated. Experimental results revealed that the durability of developed catalysts in the presence of SO2 could be improved by Ca modification. After being subjected to a range of analytical techniques, it was found that the surface Ca species could act as a SO2 trap by preferentially reacting with SO2 to form bulk-like CaSO4, inhibiting the sulfation of active phase. Furthermore, the introduction of SO2 had also preserved part of Lewis sites over the Both of these are conducive to NH3 adsorption and activation at low temperature, hence improving the sulfur tolerance of Ca doped catalysts.

1. Introduction

Selective Catalytic Reduction (SCR) of NOx with NH3 in excess O2 is one of the most effective ways to eliminate NOx emissions from stationary sources [1]. However, by using commercialized V2O5/TiO2-based catalysts, the SCR reactors have to be located upstream of the particulate control device due to their high operation temperature. This brings inherent problems like SO2 oxidation, installation difficulty, and high deactivation risk. Therefore, the development of low-temperature SCR catalysts has attracted great research interests in recent years [2].

Up to now, lots of transitional metal modified low-temperature catalysts have been reported, among which, manganese oxides have been studied extensively due to their superior low-temperature SCR activity [39]. However, the Mn-based catalysts still suffer from SO2 poisoning due to the presence of residual SO2 (even after desulphurization), whose deactivation mechanism has been reported due to the sulfation of active phase or/and the blockage of catalysts’ micropores by the deposition of (NH4)2SO4 and NH4HSO4 [10]. As such, improvement of sulfur tolerance of the low-temperature SCR catalysts was also needed to widely concern. One of the effective ways is the modifications by adding some metal oxides into the catalysts. Chang et al. [6] had reported that the addition of Sn into MnOx-CeO2 catalysts could greatly enhance the SO2 resistance owing to the enhanced Lewis acid sites. Shen et al. [11] concluded that the iron doping would improve SO2 tolerance of Mn-Ce/TiO2 due to the inhibition of surface sulfate formation. Our previous work had demonstrated that adding Ce to Mn/TiO2 could inhibit the active phase sulfation, contributing to the promotion in SO2 resistance [12].

In our previous study [13], we reported that the Ca doping could greatly improve the low-temperature SCR activity of Mn/TiO2 catalysts due to its positive effects on MnOx dispersion and adsorptive capacity of NOx, but the work did not look insightfully into the reaction behaviors of these catalysts in the presence of SO2. As an alkali earth metal, Ca might have stronger chemical affinity to SOx, which could lessen the sulfation of MnOx, thereby enhancing the sulfur tolerance of catalysts. Similar finding was also observed by Du et al. [14] for Cu-Ce/TiO2 system. Therefore, in this paper, the effects of Ca doping on SO2 tolerance of Mn/TiO2 catalysts were performed and the detailed mechanisms were analyzed by using XRD, BET, XPS, and DRIFT.

2. Experimental

2.1. Catalyst Preparation

The Ca-modified Mn/TiO2 catalysts were prepared via a sol-gel method using butyl tetratitanate (0.1 mol), ethanol (0.8 mol), water (0.6 mol), acetic acid (0.3 mol), manganese nitrate (0.04 mol), and certain amount of calcium nitrate as we reported in the previous work [13]. The catalysts were hereafter denoted by Ca(x)Mn(0.4)/Ti, where x, 0.4 represented the molar ratio of CaO to TiO2 and MnOx to TiO2, respectively.

2.2. Catalytic Activity Measurements

SCR activity measurements were carried out in a fixed-bed, quartz tubular flow reactor within the 60–200°C range of temperature. The typical reactant gas composition was NO of 600 ppm, NH3 of 600 ppm, SO2 of 50 ppm, O2 of 3%, and balanced N2, and the GHSV (gas hourly space velocity) was 40,000 h−1. The concentrations of NO, NO2, and N2O were monitored by nondispersive infrared- (NDIR-) based gas analyzer (Photon-PGD-100 Madur Electronics).

2.3. Characterization

X-ray diffraction patterns (XRD) were recorded on a Rigaku D/Max-RA powder diffractometer using Cu Kα radiation (40 kV and 150 mV). X-ray photoelectron spectroscopy (XPS) was recorded with Al Kα X-rays (Thermal, ESCALAB 250). FT-IR spectra were acquired using an in situ DRIFT cell equipped with a gas flow system (Nicolet 6700 FTIR spectrometers). Samples were pretreated at 400°C in a He environment for 2 h and then cooled to 160°C. The background spectrum was recorded with flowing He and was subtracted from the catalyst spectrum.

3. Results and Discussion

3.1. Catalytic Activity in the Presence of SO2

The effects of SO2 on the SCR activities of Mn/TiO2 and Ca doped Mn/TiO2 catalysts were illustrated in Figure 1, which showed that SO2 had obvious poisoning effect on SCR activity over all the catalysts. The Mn(0.4)/TiO2 catalyst was much more susceptible to SO2 than the Ca doped ones, where the NO conversion of Mn(0.4)/TiO2 catalyst declined from 100% to 30% after introducing 50 ppm SO2 over 4 h, but about 80% NO conversion was preserved over the Ca(0.1)-Mn(0.4)/TiO2 catalyst. This indicated that the Ca doping could enhance the SO2 tolerance of Mn/TiO2 catalyst to some extent. However, the deactivation of catalysts could not be avoided in long-term running.

3.2. XRD and XPS Results
3.2.1. XRD Analysis

Figure 2 showed the XRD spectra of Mn(0.4)/TiO2 and Ca(0.1)-Mn(0.4)/TiO2 catalysts before and after SO2 poisoning (fresh catalysts and used catalysts after SCR reaction in the presence of SO2). After SO2 poisoning, no obvious differences were observed on Mn(0.4)/TiO2 catalyst, but peaks for CaSO4 (PDF-no. 30-0279) were detected on Ca(0.1)-Mn(0.4)/TiO2 catalysts, indicating the formation of bulk-like CaSO4 during the SCR reaction. There were no peaks for (NH4)2SO4 and NH4HSO4 that existed for both catalysts, suggesting that the deactivation under SO2 atmosphere could be due to the sulfation of active phase. The BET results (see Table in supplementary material available online at http://dx.doi.org/10.1155/2014/904649) also confirmed this assumption as only minor changes of pore volume and surface area were detected.

3.2.2. XPS Analysis

The photoelectron spectra of Mn 2p for different catalysts before and after SO2 poisoning were displayed in Figures 3(a) and 3(b), respectively. After being pretreated by SO2, Mn 2p3/2 peak of Mn(0.4)/TiO2 obviously shifted to about 0.5 eV higher binding energy, which indicated that some new surface species were formed. Similar phenomena were reported that the XPS peaks of metal-oxides catalysts could migrate to higher binding energy due to the active phase sulfation [15, 16]. Therefore, we could conclude that the MnOx was sulfated during the SCR in the presence of 50 ppm SO2. In contrast, for Ca doped Mn(0.4)/TiO2 catalyst, there were no noticeable changes in the binding energy of Mn 2p3/2 being observed after SO2 introduction, which illustrated that the Ca addition could efficiently restrain the sulfation of MnOx on catalysts surface, shielding the active phases. Figure 3(c) revealed the binding energies of Ca 2p photoelectron peaks. For fresh Ca(0.1)-Mn(0.4)/TiO2, the binding energy of Ca 2p3/2 was 346.8 eV, which was close to the value of CaO (347.2 eV) reported in the literature [17]. After SO2 treated, the XPS peaks of Ca 2p3/2 shifted to higher binding energy range. Combined with the XRD results, the shift in binding energy was mainly ascribed to the formation of sulfated calcium (Ca 2p peaks for CaSO4 at about 348.1 eV) [18]. Figure 4(d) also represented the photoelectron spectra of S 2p for Mn/TiO2 and Ca(0.01)-Mn/TiO2 after SO2 poisoning. The peaks at 168.9 and 169.8 eV all could be attributed to the S(VI), indicating the formation of sulfate species after SO2 poisoning [19, 20].

Additionally, it can be also observed that the surface sulfur content of Ca(0.1)-Mn(0.4)/TiO2 was lower than that of Mn(0.4)/TiO2 after SCR reaction (see Table ), which suggested that the deposition of sulfate species on catalyst surface was inhibited after Ca doping. The possible reason is that Ca dopants would preferentially react with SO2 to form CaSO4 and thereby weaken the sulfation of MnOx. However, it should be noted that the sulfate species could also migrate into the bulk phase as CaSO4 (see XRD results) would also lead to the decline of surface sulfur content.

3.3. In Situ DRIFT Study
3.3.1. SO2 Adsorption on Mn/TiO2 and Ca-Mn/TiO2

Figure 4 showed the DRIFT spectra of adsorbed species over Mn(0.4)/TiO2 and Ca(0.1)-Mn(0.4)/TiO2 catalysts in flowing SO2 + O2/He at 160°C as a function of time. As shown in Figure 4(a), several bands at 1166, 1250, 1350, 1442, and 1626 cm−1 were attributed to adsorbed sulfate species. The bands at around 1200 cm−1 (1166 and 1250 cm−1) were caused by bulk-like sulfated species and assigned to the vibrations of S–O and S=O [2123], and the bands at 1442 and 1350 cm−1 were caused by the bonds vibrations of surface sulfate [22]. Moreover, the bands around 1626 cm−1 were assigned to the adsorbed H2O due to the reaction between SO2 and surface hydroxyl groups [24]. In contrast, the IR spectra of Ca(0.1)-Mn(0.4)/TiO2 were somewhat different from that of Mn(0.4)/TiO2 as shown in Figure 4(b), which is the characterized peaks of bulk sulfate [22]. According to the XRD and XPS results, these bulk sulfate species should be attributed to bulk-like CaSO4. This indicated that after Ca addition, the sulfate species tended to form in bulk rather than on surface, protecting the active phases of the catalysts.

3.3.2. NH3 Adsorption after SO2 Pretreatment on Catalysts’ Surface

DRIFT spectra of NH3 adsorption on Mn(0.4)/TiO2 and Ca(0.1)-Mn(0.4)/TiO2 catalysts at 160°C after SO2 pretreatment for 30 min were presented in Figures 5(a) and 5(b), respectively. And as a comparison, the spectra of the fresh catalysts treated with NH3 for 30 min were also shown in Figure 5 as the first curve. As shown in Figure 5(a), the spectrum taken after 30 min of NH3 adsorption on Mn(0.4)/TiO2 was characterized by bands at 930, 964, 1160, and 1598 cm−1 [3]. The bands located in 930 and 964 cm−1 were attributed to the weakly adsorbed NH3 or gas phase NH3. In the NH stretching region, the bands at 1160, 1598 cm−1 were indicative of coordinative adsorbed NH3 on Lewis sites [2527]. As for SO2 treated catalyst, new bands at 1430, 1342, and 1105 cm−1 emerged. According to the reports [8, 28], the band at 1430 cm−1 could be attributed to the bending vibrations of ions formed on Brönsted acid sites and the band at 1105 cm−1 could be assigned to the hydrogen bonds of adsorbed NH3 [21]. Compared with the fresh sample, SO2 pretreatment would significantly promote the formation of Brönsted acid sites, where the Lewis acid sites almost disappeared after introducing SO2. Moreover, the negative peak at 1342 cm−1 was ascribed to S=O band, which tended to bond with NH3 on catalyst surface according to the previous study [10].

As for Ca(0.1)-Mn(0.4)/TiO2 catalysts (Figure 5(b)), similar bands were observed for NH3 adsorption after SO2 pretreatment. The band at 1103 cm−1 was assigned to hydrogen-bonded NH3 adspecies. The band at 1440 cm−1 was due to the ions formed on Brönsted acid sites. The negative peaks at 1294 cm−1 could be attributed to (S=O) bond due to the interaction between sulfate species and NH3. However, the IR spectra also showed some differences with that for Ca-free catalyst. It was found that the band at 1598 cm−1 due to NH3 adsorption on Lewis acid sites was preserved although the intensity was reduced, and the peak at 1160 cm−1 (NH3 adsorption on Lewis acid sites) still can be detected, which was probably overlapped by the bands at 1103 cm−1. Thus, since the NH3 adsorption and activation on Lewis acid sites played a critical role in the low-temperature SCR process [25, 29], we can conclude that the lessen in MnOx sulfation and the preservation of part of Lewis acid sites could be the key reason for the great enhancement in SO2 tolerance of the catalysts after Ca addition.

4. Conclusion

Ca modifications of MnOx/TiO2 catalysts using a sol-gel method would bring an obvious enhancement in SO2 tolerance for low-temperature SCR of NO with NH3. Experimental results showed that around 80% NO conversion could be remained for Ca(0.1)-Mn(0.4)/TiO2 catalyst in the presence of 50 ppm SO2 for 4 h, while that was less than 30% for Ca-free catalyst. Based on the characterizations by XRD, XPS, and IR, it was concluded that the sulfation of MnOx was greatly inhibited by Ca doping, which was assumed to be due to the fact that the Ca dopants would preferentially react with SO2 to form bulk-like sulfate species. Furthermore, DRIFT results also indicated that part of Lewis acid sites could be preserved on Ca doped catalysts that was pretreated with SO2, which was beneficial to the fulfillment of the low-temperature SCR cycle.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.


This work is financially supported by Key Project of Zhejiang Provincial Science and Technology Program (no. 2012C03003-4), National High-Tech Research and Development Program (863) of China (2011AA060801), and Zhejiang Provincial Natural Science Foundation of China (no. LQ12E08011).

Supplementary Materials

“Details regarding the BET surface areas and XPS analyses of Mn/TiO2 and Ca(0.1)-Mn/TiO2 catalysts before and after SCR reaction” were present in supplementary section.

  1. Supplementary Material