Skip to main content
NCBI home page
iScience logoLink to iScience
. 2020 May 18;23(6):101173. doi: 10.1016/j.isci.2020.101173

Tailored Alkali Resistance of DeNOx Catalysts by Improving Redox Properties and Activating Adsorbed Reactive Species

Mehak Nawaz Khan 1,2, Lupeng Han 1,2, Penglu Wang 1, Dengsong Zhang 1,3,∗
PMCID: PMC7262565  PMID: 32480128

Summary

It is still challenging to develop strongly alkali-resistant catalysts for selective catalytic reduction of NOx with NH3. It is generally believed that the maintenance of acidity is the most important factor because of neutral effects of alkali. This work discovers that the redox properties rather than acidity play decisive roles in improving alkali resistance of some specific catalyst systems. K-poisoned Fe-decorated SO42−-modified CeZr oxide (Fe/SO42−/CeZr) catalysts show decreased acidity but reserve the high redox properties. The higher reactivity of NHx species induced by K poisoning compensates for the decreased amount of adsorbed NHx, leading to a desired reaction efficiency between adsorbed NHx and nitrate species. This study provides a unique perspective in designing an alkali-resistant deNOx catalyst via improving redox properties and activating the reactivities of NHx species rather than routinely increasing acidic sites for NHx adsorption, which is of significance for academic interests and practical applications.

Subject Areas: Chemical Engineering, Catalysis, Environmental Chemistry, Environmental Chemical Engineering

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    Fe-decorated SO42−-modified CeZr catalysts exhibit superior alkali resistance

  • •

    Improved redox properties compensate for the loss of the acidity

  • •

    Higher reactivity of NHx species makes up their decreased quantity

  • •

    Alkali resistance is enhanced via improving the redox and reactivity of NHx species

Chemical Engineering; Catalysis; Environmental Chemistry; Environmental Chemical Engineering

Introduction

Concerns about the severe acid rain and haze problems caused by NOx excessive emission have triggered extensive researches on effective abatement controls of NOx (deNOx) via NH3 selective catalytic reduction (NH3-SCR) (Han et al., 2019, Han et al., 2019, Paolucci et al., 2017, Qu et al., 2020). Since the ultra-low emissions of NOx have been almost attained for the power plants, it is more pressing to reduce NOx emissions in some non-electrical industries such as steel plants, biomass burning boilers, and waste incinerators. There is an increasing demand for novel alkali-resistant SCR catalysts because the commercial vanadia-based catalysts tend to be poisoned by the alkali metals such as K and Na released from the flue gas (Hao et al., 2019, Huang et al., 2013, Marberger et al., 2016, Peng et al., 2016). A generally recognized deactivation mechanism is the proton exchange of alkali-metal ions onto the active Brønsted acid sites, which results in the loss of acidity over the neutralized acid sites (Hao et al., 2019, Hu et al., 2015b). A feasible approach to improve the alkali tolerance is adopting strongly acidic supports that supply sufficient acidic sites to interact with alkali ions and thus protect the active sites. Sulfated metal oxides supports such as TiO2 and ZrO2 as well as sulfated titanate nanotubes have effectively improved alkali resistance (Due-Hansen et al., 2007, Gao et al., 2014, Putluru et al., 2012, Wang et al., 2015). However, this approach impairs the catalytic activity to some extent because of the sacrificial acidity of acidic supports. Another efficient strategy of alkali resistance is constructing the alkali-trapping sites that separate active sites and alkali-poisoned sites. Such a measure could reserve the original activity of catalysts on account of the intact acidic and redox sites. Hollandite manganese oxide and hexagonal WO3 have been demonstrated to be effective for strengthening alkali resistance owing to the trapping effects of internal tunnels without destroying active sites (Hu et al., 2015b, Zheng et al., 2016). These strategies could protect active sites from alkali poisoning through precise structural designs. However, the stringent specification of acidity strength, pore tunnels, and type of oxides restrict the alkali-resistant applications. Actually, the alkali-resistant strategy is still lacking to develop highly efficient catalysts.

So far, less research effort has been devoted to the variation of redox sites and the reactivity of adsorbed nitrate and NHx species associated with alkali poisoning, which are decisive to SCR activity. In this study, we discover an unexpected result that redox properties rather than acidity play decisive roles in improving alkali resistance of Fe-decorated SO42−-modified CeZr oxide (Fe/SO42−/CeZr) catalysts. Herein, Ce-Zr mixed oxides combining the highly refractory property of ZrO2 with the superior oxygen storage capacity of CeO2 are used as the model catalyst, which are usually served as supports for SCR catalysts (Ding et al., 2015, Li et al., 2008, Sánchez Escribano et al., 2009). SO42− modification is performed on Ce-Zr mixed oxides to increase the acidic sites and improve the SCR activity (Zhang et al., 2017), whereas the SCR activity of SO42−/CeZr is largely decreased after alkali poisoning because of the loss of acidity and redox properties associated with the decreased reactivity of adsorbed nitrate and NHx species. Via decorating Fe on SO42−/CeZr, the redox properties are enhanced after K poisoning owing to the promoted electron transfer between K, Fe, Ce, and Zr as well as sufficient Ce3+ and active oxygen species. Although the acid amount of K/Fe/SO42−/CeZr catalysts decreases more severely than K/SO42−/CeZr, the enhanced redox properties compensate for the loss of acidity, which notably improves the reactivity of adsorbed NHx and meanwhile maintains the high reactivity of adsorbed nitrate species. Therefore, the K-poisoned Fe/SO42−/CeZr catalyst exhibits satisfactory SCR activity. This finding is of significance in revealing novel alkali-resistant mechanisms and paves a novel way for developing alkali-resistant catalysts in the future research.

Results

Originally, one strategy of increasing acidity was attempted to improve the SCR activity via introducing SO42− onto the surface of CeZr oxides. As expected, SO42−/CeZr shows higher SCR activity (300°C–480°C, NO conversion above 90%, Figure 1) compared with the pristine CeZr mixed oxides (the highest NO conversion of 80% at 330°C, Figure S1). K-poisoned CeZr catalyst is almost deactivated with the highest NO conversion of only 16.7% at 360°C (Figure S1) and K-poisoned SO42−/CeZr catalyst also displays poor activity with the maximum 64.8% NO conversion at 360°C (Figure 1). This result indicates that introducing more acidic sites is not enough to maintain the SCR activity of CeZr catalysts after K poisoning. Besides, Fe-decorated CeZr (without SO42− modification) shows lower activity than SO42−/CeZr and inferior K resistance because of the less acidity (Figure S2). Via tuning the amount of Fe decoration on SO42−/CeZr, 1.5Fe/SO42−/CeZr exhibits the optimum SCR activity with a broad temperature window (270°C–450°C, NO conversion above 90%) (Figure S3). After K poisoning, 1.5Fe/SO42−/CeZr still exhibits more than 85% NO conversion within 270°C–420°C (Figure 1), indicating that Fe decoration notably improves the alkali resistance of SO42−/CeZr. Besides, the fresh/K-poisoned SO42−/CeZr and 1.5Fe/SO42−/CeZr catalysts all show good N2 selectivity above 90% within 150°C–480°C (Figure S4).

Figure 1.

Figure 1

Open in a new tab

NO Conversion during the SCR Reaction over Various Catalysts.

Reaction conditions: 500 ppm NO, 500 ppm NH3, 5 vol % O2, N2 as the balance gas, and GHSV of 100,000 h−1.

To probe the effects of Fe decoration on the alkali resistance of SO42−/CeZr, the structural and textural features of fresh and K-poisoned SO42−/CeZr and 1.5Fe/SO42−/CeZr were first investigated. The X-ray diffraction patterns (Figure S5) and Raman spectra (Figure S6) both evidence the formation of Ce-Zr solid solution over fresh/K-poisoned SO42−/CeZr and 1.5Fe/SO42−/CeZr catalysts. Besides, no FeOx-related X-ray diffraction peaks or Raman bands are observed, indicating the FeOx is highly dispersed on the surface of SO42−/CeZr. The scanning electron microscope mapping of the representative K-poisoned 1.5Fe/SO42−/CeZr catalyst shows that Fe, Ce, and Zr active components are highly dispersed with each other and K is also uniformly dispersed on the surface of catalysts (Figure S7). The Brunauer-Emmett-Teller (BET) surface areas of fresh SO42−/CeZr and 1.5Fe/SO42−/CeZr are 60.06 and 44.42 m2/g, whereas they increase to 70.37 and 66.29 m2/g after K poisoning, respectively. The increase of surface area after K poisoning is likely due to the contribution of K2O nanoparticle on catalyst surface (Table S1). There is no correlation between surface area and SCR activity, indicating that the surface area of catalysts is not decisive to the SCR activity of catalysts.

The exposure of catalysts to alkali metals could reduce the acid sites for NH3 adsorption/activation, thus causing a severe decrease of catalytic activity. It is necessary to probe the changes in acidity of SO42−/CeZr and 1.5Fe/SO42−/CeZr catalysts before/after K poisoning. Herein, NH3 temperature-programmed desorption combining with mass spectrum (NH3-TPD-MS) was performed to study the acidic properties of catalysts (Figure 2A). The NH3 desorption peaks below 300°C on all the catalysts are attributed to the weakly acidic sites, and the peaks above 300°C are attributed to the strongly acidic sites (Park et al., 2016). K poisoning obviously decreases the amount of weakly acidic sites over SO42−/CeZr and 1.5Fe/SO42−/CeZr. The quantitative analysis of NH3-TPD-MS reveals that the total acid amount of SO42−/CeZr decreases from 292.2 to 182.9 μmol/g (decrease by 37.4%), whereas that of 1.5Fe/SO42−/CeZr decreases from 257.6 to 115.2 mmol/g (decrease by 55.3%) after K poisoning. The fresh 1.5Fe/SO42−/CeZr (257.6 μmol/g) owns less acidity than the fresh SO42−/CeZr (292.2 μmol/g) likely because FeOx species occupy some SO42− acidic sites. The acid amount of 1.5Fe/SO42−/CeZr decreases more severely than SO42−/CeZr (decrease by 55.3% versus 37.4%) after K poisoning, which is likely because more weakly acidic sites derived from Fe-OH are lost (Sugawara et al., 2007). This is also evidenced by that K/1.5Fe/SO42−/CeZr almost lost all the weak acid but K/SO42−/CeZr still possesses some weak acid (Figure 2A). It is notable that the total acid amount of K/1.5Fe/SO42−/CeZr is even less than K/SO42−/CeZr, implying that the acid amount is not the essential reason for the strong K resistance of 1.5Fe/SO42−/CeZr catalysts. Besides, SO2 signal (m/z = 64) was observed above 600°C over all fresh and K-poisoned catalysts during NH3-TPD-MS (Figure S8), indicating that SO42− strongly bonds on CeZr catalysts and is thermally stable during the whole SCR active temperature region (<500°C).

Figure 2.

Figure 2

Open in a new tab

Acidity and Redox Properties of Different Catalysts

(A) NH3-TPD-MS, (B) H2-TPR, (C) XPS spectra of Ce 3d, and (D) XPS spectra of O 1s over different catalysts. Catalysts: SO42−/CeZr (a), K/SO42−/CeZr (b), 1.5Fe/SO42−/CeZr (c), and K/1.5Fe/SO42−/CeZr (d).

As mentioned above, the acid amount is notably reduced for K-poisoned 1.5Fe/SO42−/CeZr. One question arises: Does the increased redox property associated with Fe decoration improve the alkali resistance? To check the changes of redox properties along with Fe decoration, H2 temperature-programmed reduction (H2-TPR) and X-ray photoelectron spectroscopy (XPS) were carried out for Fresh and K-poisoned catalysts. As shown in Figure 2B, SO42−/CeZr shows two fitted reduction peaks around 600°C, which are related to the reduction of CeOx that is not interacted with ZrOx (low-temperature peak) and CeOx that is strongly interacted with ZrOx (high-temperature peak), respectively. After K poisoning, the two reduction peaks of SO42−/CeZr shift to lower temperatures (around 540°C), which implies K as an electron donating promoter facilitates the CeO2 reduction. Compared with SO42−/CeZr, the two fitted reduction peaks of 1.5Fe/SO42−/CeZr shift to much lower temperatures (around 450°C), in which the low-temperature peak is likely related to the FeOx reduction and the high-temperature one is attributed to the CeOx reduction (Liu and He, 2010). With increasing the amount of Fe to 3 wt % Fe and 5 wt % Fe, the reduction peaks further shift to lower temperatures of ∼440°C and ∼420°C, respectively (Figure S9). These results indicate that the strong interaction between Fe and Ce facilitates the reduction of CeO2. The reduction peaks of FeOx/CeOx of 1.5Fe/SO42−/CeZr catalysts after K poisoning shift from ∼450°C to ∼440°C, and the third reduction peak around 510°C appears, indicating that K poisoning improves the reduction of FeOx/CeOx species but meanwhile impairs the interaction between Fe and Ce to some extent, which restrains the reduction of CeOx. The reduction of CeOx over K poisoned 1.5Fe/SO42−/CeZr is still more reducible compared with K-poisoned SO42−/CeZr according to the highest reduction temperature of CeOx (510°C for K/1.5Fe/SO42−/CeZr versus 550°C for K/SO42−/CeZr). These indicate that the reducibility is improved over Fe-decorated SO42−/CeZr catalysts and K poisoning further enhances the reducibility to some extent. Via analyzing X-ray photoelectron spectroscopy (XPS) of Ce 3d, K-poisoned SO42−/CeZr and 1.5Fe/SO42−/CeZr both show higher Ce3+/(Ce3++Ce4+) ratio than fresh ones (Figure 2C) (Han et al., 2019b), which is attributed to that K as an electron donating promoter reduces the valence of Ce species. Although the fresh 1.5Fe/SO42−/CeZr has less Ce3+ fraction than SO42−/CeZr, the K-poisoned 1.5Fe/SO42−/CeZr shows the highest Ce3+ fraction of 29.2%, which indicates that the electron-donating effects of K also facilitate the electron transfer from Fe to Ce. Generally, the formation of Ce3+ species brings out more oxygen vacancies, and the higher Ce3+ ratio with the more oxygen vacancies improves the oxidizability. Moreover, the surface oxygen species also deliver a different evolution after K poisoning. It is generally recognized that the surface-adsorbed oxygen species (denoted as Oα) are much more reactive in SCR reactions than the lattice oxygen species (denoted as Oβ). As seen in Figure 2D, the Oα/(Oα+Oβ) ratio of SO42−/CeZr significantly decreases from 87.1% to 42.6% after K poisoning, whereas the Oα/(Oα+Oβ) ratio of K-poisoned 1.5Fe/SO42−/CeZr keeps at 93.7% that is almost unchanged with the fresh one (94.0%). This indicates that Fe decoration maintains the surface adsorption oxygen species that are active for the oxidation process for the SCR reaction. The O2-TPD results also evidence that the chemically adsorbed oxygen molecule anion (O2–) and oxygen anion (O–) species are notably reduced over K-poisoned SO42−/CeZr but almost unchanged over K-poisoned 1.5Fe/SO42−/CeZr (Figure S10). The electron states of Fe, Zr, and S for alkali-poisoned SO42−/CeZr and 1.5Fe/SO42−/CeZr were also investigated. As the weak Fe 2p XPS signals of 1.5Fe/SO42−/CeZr and K/1.5Fe/SO42−/CeZr (Figure S11A), the Fe 2p XPS spectra of 5Fe/SO42−/CeZr before/after K poisoning were analyzed (Figure S11B). The fresh 5Fe/SO42−/CeZr possesses more Ce3+ fraction (22.97%) than the fresh 1.5Fe/SO42−/CeZr (14.64%), indicating more electron transfer from Fe to Ce. Additionally, K-poisoned 5Fe/SO42−/CeZr possesses higher Fe2+/(Fe2+ + Fe3+) ratio and Ce3+ fraction than the fresh one owing to the electron-donating effects of K. 1.5Fe/SO42−/CeZr has higher binding energy of Zr 3d5/2 than SO42−/CeZr (Figure S12), indicating Fe likely gets electron from Zr owing to the strong interaction between Fe and Zr. After K poisoning, the Zr 3d5/2 binding energy of both catalysts shifts to a lower value due to the electron-donating effects of K. Additionally, K poisoning does not change the valence of S species, which exist in SO42− species on all fresh and K-poisoned catalysts (Figure S13). Based on the above results, K-poisoned 1.5Fe/SO42−/CeZr improves the reducibility and maintains the high oxidative capacity because of the facilitated electron transfer between K, Fe, Ce, and Zr as well as adequate Ce3+ and active oxygen species. As a comparison, the reducibility of K-poisoned SO42−/CeZr is not as good as K-poisoned 1.5Fe/SO42−/CeZr and the oxidative capacity of K-poisoned SO42−/CeZr is largely impaired owing to the notable decrease of active oxygen species. Therefore, the K-poisoned 1.5Fe/SO42−/CeZr shows stronger redox properties than K-poisoned SO42−/CeZr.

It has been demonstrated that the acidity decreases but the redox properties reserve over K-poisoned 1.5Fe/SO42−/CeZr catalysts. Why can the K/1.5Fe/SO42−/CeZr catalyst maintain high activity in spite of the decreased acidity? Differences in the reactivity of adsorbed NHx and nitrate species may be the possible reason. Therefore, the adsorption and activation characteristics of NH3 and NO need to be probed. From the in situ DRIFTs of NH3 desorption under various temperatures (Figure S14), the NH4+ and NH3 species adsorbed on both SO42−/CeZr and 1.5Fe/SO42−/CeZr catalysts are not stable and easy to desorb above 250°C. The NH4+ (1,698, 1,472, and 1,438 cm−1) (Huang et al., 2016, Huang et al., 2017, Wei et al., 2016) and NH3 (1,369 cm−1) (Zhang et al., 2018) species still exist on K-poisoned SO42−/CeZr at 300°C (Figures 3A and 3A1), indicating that the adsorption strength of these NHx species becomes stronger after K poisoning. By comparison, fewer NH4+ and NH3 species adsorb on K/1.5Fe/SO42−/CeZr at 300°C (Figures 3B and 3B1), implying that these NHx species are likely more reactive because of their appropriate bonding strength. Additionally, in situ DRIFTs of NO + O2 desorption under various temperatures were studied to investigate the adsorbed strength of NOx species. It can be seen that the N2O4, bidentate nitrate, monodentate nitrite, and metal-NO2 species adsorb on SO42−/CeZr (Figure S15A), whereas the N2O4, bridged nitrate, and bidentate nitrate species adsorb on Fe-decorated SO42−/CeZr catalysts (Figure S15B). With increasing temperatures, the NOx species adsorbed on 1.5Fe/SO42−/CeZr are easier to desorb than on SO42−/CeZr, implying these species could react with NHx species more easily. After K poisoning, K/SO42−/CeZr catalysts show adsorbed NOx species including N2O4 (1,713 cm−1) (Davydov, 2003), adsorbed NO2 (1,629 cm−1) (Liu et al., 2017), bidentate nitrate (1,498 cm−1) (Davydov, 2003), and metal-NO2 (1,367 and 1,333 cm−1) (Davydov, 2003) (Figures 3C and 3C1). It is notable that the bidentate nitrate and metal-NO2 species still adsorb on K/SO42−/CeZr at 300°C and the strong bonding with catalyst likely reduces their reactivity. By comparison, the N2O4 (1,707 cm−1), bidentate nitrate (1,517 cm−1) (Davydov, 2003), and metal-NO2 (1,330 cm−1) (Davydov, 2003) species on K-poisoned 1.5Fe/SO42−/CeZr desorb more easily than those on K-poisoned SO42−/CeZr catalysts (Figures 3D and 3D1). These results indicate that the reactivities of NOx species over K-poisoned 1.5Fe/SO42−/CeZr are likely higher than those over K-poisoned SO42−/CeZr. Moreover, NO-TPD-MS (Figure S16A) and NO + O2-TPD-MS (Figure S16B) results show that the amount of nitrate species adsorbed on SO42−/CeZr reduces after K poisoning, whereas those on 1.5Fe/SO42−/CeZr increases after K poisoning. This indicates that Fe decoration helps to improve the formation of nitrite species over K-poisoned catalysts, which is likely due to the improvement of redox properties after K poisoning.

Figure 3.

Figure 3

Open in a new tab

In situ DRIFTs of NHx/NOx Species Desorption over Catalysts as a Function of Temperature

In situ DRIFTs of NH3 desorption and the corresponding mapping results over K/SO42−/CeZr (A and A1) and K/1.5Fe/SO42−/CeZr (B and B1) catalysts after exposure to a flow of 500 ppm NH3 for 1 h at 30°C; in situ DRIFTs of NO + O2 desorption and the corresponding mapping results over K/SO42−/CeZr (C and C1) and K/1.5Fe/SO42/CeZr (D and D1) catalysts after exposure to a flow of 500 ppm NO + 5% O2 for 1 h at 30°C.

In order to further reveal the changes in the reactivity of adsorbed NHx and NOx, in situ DRIFTs transient reactions of adsorbed species were investigated at 250°C. Before K poisoning, the bidentate nitrates and metal-NO2 species adsorbed on SO42−/CeZr and the N2O4, bidentate nitrates, and metal-NO2 species adsorbed on 1.5Fe/SO42−/CeZr are all reactive upon introducing NH3 (Figures S17A and S17B). Notably, the reactivity of nitrate species on 1.5Fe/SO42−/CeZr is much higher than that on SO42−/CeZr. The adsorbed NH4+ and NH3 species on fresh SO42−/CeZr and 1.5Fe/SO42−/CeZr decrease slowly upon introducing NO + O2, indicating the reactivity of NHx species over both catalysts are all relatively inactive, whereas the reactivity of NHx species on 1.5Fe/SO42−/CeZr is slightly higher than that on SO42−/CeZr (Figure S18A and S18B). These results imply that Fe decoration relatively improves the reactivities of adsorbed nitrate and NHx species. After K poisoning (Figures 4A and 4A1), the N2O4 (1,696 cm−1), bidentate nitrates (1,562 cm−1) (Liu et al., 2017), and metal-NO2 (1,384 cm−1) (Davydov, 2003) species adsorbed on K/SO42−/CeZr decrease slowly and meanwhile the NH4+ (1,680 and 1,423 cm−1) (Ma et al., 2014) and NH3 (1,583 cm−1) (Weng et al., 2016) species gradually increase with introducing NH3, indicating the reactivity of nitrate species largely decreases compared with the fresh one. In contrast, the adsorbed bidentate nitrates (1,573, 1,559 cm−1) (Hu et al., 2015a, Liu et al., 2018) and metal-NO2 (1,381, 1,368 cm−1) (Davydov, 2003) species on K/1.5Fe/SO42−/CeZr reduce notably with the introduction of NH3, and meanwhile amide-like species (1,724 cm−1) (Jeong et al., 2017) and NH4+ species (1,678 cm−1) (Weng et al., 2016) emerge (Figures 4B and 4B1). These results indicate that K poisoning decreases the reactivity of nitrate species over SO42−/CeZr but has less impact on that over 1.5Fe/SO42−/CeZr. In terms of the reactivity of NHx species on K/SO42−/CeZr (Figures 4C and 4C1), the adsorbed amide-like species (1,724 cm−1) (Jeong et al., 2017), NH4+ (1,680 and 1,630 cm−1) (Yu et al., 2014) and NH3 species (1,362 and 1,332 cm−1) (Zhang et al., 2018) are almost unchanged upon introducing NO + O2, and meanwhile the metal-NO2 species that overlap with the NH3 species gradually increase with the introduction of NO + O2. By comparison, the adsorbed NH4+ (1,678, 1,654, and 1,434 cm−1) (Weng et al., 2016) and NH3 (1,375 cm−1) (Wang et al., 2016) species on K-poisoned 1.5Fe/SO42−/CeZr notably reduce with the introduction of NO + O2 (Figures 4D and 4D1). Meanwhile, the metal-NO2 (1,367 cm−1) species increase with continuously introducing NO + O2 (Figures 4D and 4D1). These indicate that the adsorbed NHx species on SO42−/CeZr become more inactive after K poisoning; however, the adsorbed NHx species on K-poisoned 1.5Fe/SO42−/CeZr become more reactive with the adsorbed nitrate species compared with the fresh one. Overall, both fresh SO42−/CeZr and 1.5Fe/SO42−/CeZr conduct the SCR reaction through the reaction between adsorbed nitrate species and adsorbed NHx species following the Langmuir-Hinshelwood (L-H) mechanism. In this reaction pathway, the reactivity of NHx species over both fresh catalysts are relatively inactive. After K poisoning, the reactivity of NHx and nitrate species largely decreases on SO42−/CeZr, leading to an inferior alkali resistance. However, K-poisoned 1.5Fe/SO42−/CeZr reserves high reactivity of adsorbed nitrate species and especially largely improves the reactivity of adsorbed NHx species. Consequently, 1.5Fe/SO42−/CeZr exhibits a satisfactory alkali resistance in spite of the decreased acidic sites.

Figure 4.

Figure 4

Open in a new tab

In situ DRIFTs of the Transient Reactions over K/SO42−/CeZr and K/1.5Fe/SO42−/CeZr

In situ DRIFTs of the transient reactions between NH3 and preadsorbed NO + O2 over K/SO42−/CeZr (A and A1) and K/1.5Fe/SO42−/CeZr (B and B1) at 250°C as a function of time; in situ DRIFTs of the transient reactions between NO + O2 and preadsorbed NH3 over K/SO42−/CeZr (C and C1) and K/1.5Fe/SO42−/CeZr (D and D1) at 250°C as a function of time.

Discussion

Nowadays, it is generally believed that the decrease in acidity is the dominant reason for the deactivation of catalysts after alkali poisoning (Hu et al., 2015b, Putluru et al., 2011, Wang et al., 2015). Additionally, the impaired redox properties resulting from alkali poisoning also lead to the decline of activity, such as the reduced reducibility for K-poisoned V2O5-WO3/TiO2 (Wang et al., 2019) and decreased oxidative capacity for K-poisoned Mn/TiO2 (Wei et al., 2018). Besides the reduced acidity and redox properties, the absence of active NOx species at low temperatures and the formation of inactive nitrate species at high temperatures result in the decreased activity of alkali-poisoned V2O5/CeO2 (Peng et al., 2014). In this study, we found that the redox properties rather than acidity play decisive roles in improving the alkali resistance of some specific catalyst systems. It is demonstrated that the acid amount of K-poisoned Fe/SO42−/CeZr is lower than that of K-poisoned SO42−/CeZr; however, the SCR activity of the former one is much higher than that of the latter one. The K-resistant mechanism of F/SO42−/CeZr is revealed as shown in Figure 5. The essential reason is that the enhanced redox properties compensate for the reduced acidity of the K-poisoned Fe/SO42−/CeZr catalysts. K-poisoned Fe/SO42−/CeZr facilitates the electron transfer between K, Fe, Ce, and Zr while reserving plenty of Ce3+ and active oxygen species, which improves the reducibility and reserves the high oxidative capacity. As a result, K/Fe/SO42−/CeZr maintains the high reactivity of adsorbed nitrate species and notably improves the reactivity of adsorbed NHx species. The higher reactivity of NHx species makes up for the loss in quantity of NHx species, which is the key for the strong K resistance of Fe/SO42−/CeZr. Consequently, Fe/SO42−/CeZr maintains the high reaction efficiency between adsorbed NHx species and adsorbed nitrate species (L-H mechanism). Conversely, the acidity decreases and the redox circle is locked for the decreased oxidative capacity on K/SO42−/CeZr, which largely reduces the reactivity of nitrate species and NHx, leading to decreased activity. This work sheds light on a novel alkali-resistant mechanism via improving redox properties and activating the reactivities of NHx species rather than routinely increasing acidic sites for NHx adsorption. This study provides a unique perspective in designing an alkali-resistant deNOx catalysts, which is beneficial for their commercial, environmental, and industrial applications.

Figure 5.

Figure 5

Open in a new tab

Schematic Diagram of the Deactivation Mechanism over K/SO42−/CeZr and Alkali-Resistant Mechanism over K/Fe/SO42−/CeZr Catalysts.

Limitations of the Study

More catalyst systems should be probed that can improve their alkali resistance via improving redox properties and activating the reactivities of NHx and nitrate species.

Resource Availability

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Dengsong Zhang (dszhang@shu.edu.cn).

Materials Availability

All unique/stable reagents generated in this study are available from the Lead Contact without restriction.

Data and Code Availability

All relevant data are available from the corresponding author (dszhang@shu.edu.cn) upon reasonable request.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Acknowledgments

We acknowledge the support of the National Natural Science Foundation of China (21722704; 21976117; 21906102), the National Key R&D Program of China (2017YFE0132400), the Shanghai Sailing Program (19YF1415300), and the China Postdoctoral Science Foundation (2018M630426).

Author Contributions

D.Z. designed the experiments, supervised the projects, and contributed to the revision of this paper. M.N.K. and L.H. contributed equally to this work. They performed catalyst preparation and catalyst characterizations, prepared the figures, and co-wrote the manuscript. P.W. analyzed the experimental results. All authors discussed the results, drew conclusions, and commented on the manuscript.

Declaration of Interests

The authors declare no competing interests.

Published: June 26, 2020

Footnotes

Supplemental Information can be found online at https://doi.org/10.1016/j.isci.2020.101173.

Supplemental Information

Document S1. Transparent Methods, Figures S1–S18, and Table S1
mmc1.pdf (1.5MB, pdf)

References

  1. Davydov A. John Wiley & Sons Ltd; 2003. Molecular Spectroscopy of Oxide Catalyst Surfaces; p. 124. Chapter 2. [Google Scholar]
  2. Ding S., Liu F., Shi X., Liu K., Lian Z., Xie L., He H. Significant promotion effect of Mo additive on a Novel Ce–Zr mixed oxide catalyst for the selective catalytic reduction of NOx with NH3. ACS Appl. Mater. Interfaces. 2015;7:9497–9506. doi: 10.1021/acsami.5b00636. [DOI] [PubMed] [Google Scholar]
  3. Due-Hansen J., Boghosian S., Kustov A., Fristrup P., Tsilomelekis G., Ståhl K., Christensen C.H., Fehrmann R. Vanadia-based SCR catalysts supported on tungstated and sulfated zirconia: influence of doping with potassium. J. Catal. 2007;251:459–473. [Google Scholar]
  4. Gao S., Wang P., Chen X., Wang H., Wu Z., Liu Y., Weng X. Enhanced alkali resistance of CeO2/SO42-–ZrO2 catalyst in selective catalytic reduction of NOx by ammonia. Catal. Commun. 2014;43:223–226. [Google Scholar]
  5. Han L., Cai S., Gao M., Hasegawa J., Wang P., Zhang J., Shi L., Zhang D. Selective catalytic reduction of NOx with NH3 by using novel catalysts: state of the art and Future prospects. Chem. Rev. 2019;119:10916–10976. doi: 10.1021/acs.chemrev.9b00202. [DOI] [PubMed] [Google Scholar]
  6. Han L., Gao M., Feng C., Shi L., Zhang D. Fe2O3–CeO2@Al2O3 nanoarrays on Al-mesh as SO2-tolerant monolith catalysts for NOx reduction by NH3. Environ. Sci. Technol. 2019;53:5946–5956. doi: 10.1021/acs.est.9b01217. [DOI] [PubMed] [Google Scholar]
  7. Han L., Gao M., Hasegawa J., Li S., Shen Y., Li H., Shi L., Zhang D. SO2-tolerant selective catalytic reduction of NOx over meso-TiO2@Fe2O3@Al2O3 metal-based monolith catalysts. Environ. Sci. Technol. 2019;53:6462–6473. doi: 10.1021/acs.est.9b00435. [DOI] [PubMed] [Google Scholar]
  8. Hao Z., Shen Z., Li Y., Wang H., Zheng L., Wang R., Liu G., Zhan S. The role of alkali metal in α-MnO2 catalyzed ammonia-selective catalysis. Angew. Chem. Int. Ed. 2019;58:6351–6356. doi: 10.1002/anie.201901771. [DOI] [PubMed] [Google Scholar]
  9. Hu H., Cai S., Li H., Huang L., Shi L., Zhang D. Mechanistic aspects of deNOx processing over TiO2 supported Co–Mn oxide catalysts: structure–activity relationships and in situ DRIFTs analysis. ACS Catal. 2015;5:6069–6077. [Google Scholar]
  10. Hu P., Huang Z., Gu X., Xu F., Gao J., Wang Y., Chen Y., Tang X. Alkali-resistant mechanism of a hollandite deNOx catalyst. Environ. Sci. Technol. 2015;49:7042–7047. doi: 10.1021/acs.est.5b00570. [DOI] [PubMed] [Google Scholar]
  11. Huang Z., Gu X., Wen W., Hu P., Makkee M., Lin H., Kapteijn F., Tang X. A “Smart” hollandite deNOx catalyst: self-protection against alkali poisoning. Angew. Chem. Int. Ed. 2013;52:660–664. doi: 10.1002/anie.201205808. [DOI] [PubMed] [Google Scholar]
  12. Huang L., Zha K., Namuangruk S., Junkaew A., Zhao X., Li H., Shi L., Zhang D. Promotional effect of the TiO2 (001) facet in the selective catalytic reduction of NO with NH3: in situ DRIFTS and DFT studies. Catal. Sci. Technol. 2016;6:8516–8524. [Google Scholar]
  13. Huang L., Hu X., Yuan S., Li H., Yan T., Shi L., Zhang D. Photocatalytic preparation of nanostructured MnO2-(Co3O4)/TiO2 hybrids: the formation mechanism and catalytic application in SCR deNOx reaction. Appl. Catal. B. 2017;203:778–788. [Google Scholar]
  14. Jeong Y.E., Kumar P.A., Huong D.T., Ha H.P., Lee K. In situ-DRIFTS study of Sb–V–CeO2/TiO2 catalyst under standard and fast NH3-SCR conditions. Top. Catal. 2017;60:755–762. [Google Scholar]
  15. Li Y., Cheng H., Li D., Qin Y., Xie Y., Wang S. WO3/CeO2-ZrO2, a promising catalyst for selective catalytic reduction (SCR) of NOx with NH3 in diesel exhaust. Chem. Commun. 2008;44:1470–1472. doi: 10.1039/b717873e. [DOI] [PubMed] [Google Scholar]
  16. Liu F., He H. Structure−activity relationship of iron titanate catalysts in the selective catalytic reduction of NOx with NH3. J. Phys. Chem. C. 2010;114:16929–16936. [Google Scholar]
  17. Liu J., Meeprasert J., Namuangruk S., Zha K., Li H., Huang L., Maitarad P., Shi L., Zhang D. Facet–activity relationship of TiO2 in Fe2O3/TiO2 canocatalysts for selective catalytic reduction of NO with NH3: in situ DRIFTs and DFT studies. J. Phys. Chem. C. 2017;121:4970–4979. [Google Scholar]
  18. Liu N., Wang J., Wang F., Liu J. Promoting effect of tantalum and antimony additives on deNOx performance of Ce3Ta3SbOx for NH3-SCR reaction and DRIFT studies. J. Rare Earths. 2018;36:594–602. [Google Scholar]
  19. Ma L., Cheng Y., Cavataio G., McCabe R.W., Fu L., Li J. In situ DRIFTS and temperature-programmed technology study on NH3-SCR of NOx over Cu-SSZ-13 and Cu-SAPO-34 catalysts. Appl. Catal. B. 2014;156-157:428–437. [Google Scholar]
  20. Marberger A., Ferri D., Elsener M., Kröcher O. The significance of lewis acid sites for the selective catalytic reduction of nitric oxide on vanadium-based catalysts. Angew. Chem. Int. Ed. 2016;55:11989–11994. doi: 10.1002/anie.201605397. [DOI] [PubMed] [Google Scholar]
  21. Paolucci C., Khurana I., Parekh A., Li S., Shih A., Li H., Di Iorio J., Albarracin-Caballero J., Yezerets A., Miller J. Dynamic multinuclear sites formed by mobilized copper ions in NOx selective catalytic reduction. Science. 2017;357:898–903. doi: 10.1126/science.aan5630. [DOI] [PubMed] [Google Scholar]
  22. Park J.H., Pang C., Chung C.-H., Bae J.W. Methyl acetate synthesis by esterification on the modified ferrierite: correlation of acid sites measured by pyridine IR and NH3-TPD for steady-state activity. J. Nanosci. Nanotechnol. 2016;16:4626–4630. doi: 10.1166/jnn.2016.11013. [DOI] [PubMed] [Google Scholar]
  23. Peng Y., Li J., Huang X., Li X., Su W., Sun X., Wang D., Hao J. Deactivation mechanism of potassium on the V2O5/CeO2 catalysts for SCR reaction: acidity, reducibility and adsorbed-NOx. Environ. Sci. Technol. 2014;48:4515–4520. doi: 10.1021/es405602a. [DOI] [PubMed] [Google Scholar]
  24. Peng Y., Si W., Li X., Chen J., Li J., Crittenden J., Hao J. Investigation of the poisoning mechanism of lead on the CeO2–WO3 catalyst for the NH3–SCR reaction via in situ IR and Raman spectroscopy measurement. Environ. Sci. Technol. 2016;50:9576–9582. doi: 10.1021/acs.est.6b02307. [DOI] [PubMed] [Google Scholar]
  25. Putluru S.S.R., Riisager A., Fehrmann R. Alkali resistant Cu/zeolite deNOx catalysts for flue gas cleaning in biomass fired applications. Appl. Catal. B. 2011;101:183–188. [Google Scholar]
  26. Putluru S.S.R., Kristensen S.B., Due-Hansen J., Riisager A., Fehrmann R. Alternative alkali resistant deNOx catalysts. Catal. Today. 2012;184:192–196. [Google Scholar]
  27. Qu W., Liu X., Chen J., Dong Y., Tang X., Chen Y. Single-atom catalysts reveal the dinuclear characteristic of active sites in NO selective reduction with NH3. Nat. Commun. 2020;11:1532. doi: 10.1038/s41467-020-15261-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Sánchez Escribano V., Gallardo Amores J.M., Fernández López E., del Hoyo Martínez C., Busca G., Resini C. Surface and structural characterization of ZrO2–CeO2-supported molybdenum oxide catalysts. Mater. Chem. Phys. 2009;114:848–853. [Google Scholar]
  29. Sugawara K., Nobukawa T., Yoshida M., Sato Y., Okumura K., Tomishige K., Kunimori K. The importance of Fe loading on the N2O reduction with NH3 over Fe-MFI: effect of acid site formation on Fe species. Appl. Catal. B. 2007;69:154–163. [Google Scholar]
  30. Wang P., Wang H., Chen X., Liu Y., Weng X., Wu Z. Novel SCR catalyst with superior alkaline resistance performance: enhanced self-protection originated from modifying protonated titanate nanotubes. J. Mater. Chem. A. 2015;3:680–690. [Google Scholar]
  31. Wang P., Sun H., Quan X., Chen S. Enhanced catalytic activity over MIL-100(Fe) loaded ceria catalysts for the selective catalytic reduction of NOx with NH3 at low temperature. J. Hazard. Mater. 2016;301:512–521. doi: 10.1016/j.jhazmat.2015.09.024. [DOI] [PubMed] [Google Scholar]
  32. Wang X., Cong Q., Chen L., Shi Y., Shi Y., Li S., Li W. The alkali resistance of CuNbTi catalyst for selective reduction of NO by NH3: a comparative investigation with VWTi catalyst. Appl. Catal. B. 2019;246:166–179. [Google Scholar]
  33. Wei L., Cui S., Guo H., Ma X., Zhang L. DRIFT and DFT study of cerium addition on SO2 of Manganese-based Catalysts for low temperature SCR. J. Mol. Catal. A. 2016;421:102–108. [Google Scholar]
  34. Wei L., Cui S., Guo H., Zhang L. The effect of alkali metal over Mn/TiO2 for low-temperature SCR of NO with NH3 through DRIFT and DFT. Comput. Mater. Sci. 2018;144:216–222. [Google Scholar]
  35. Weng X., Dai X., Zeng Q., Liu Y., Wu Z. DRIFT studies on promotion mechanism of H3PW12O40 in selective catalytic reduction of NO with NH3. J. Colloid. Interface Sci. 2016;461:9–14. doi: 10.1016/j.jcis.2015.09.004. [DOI] [PubMed] [Google Scholar]
  36. Yu Y., Wang J., Chen J., Meng X., Chen Y., He C. Promotive effect of SO2 on the activity of a deactivated commercial selective catalytic reduction catalyst: an in situ DRIFT study. Ind. Eng. Chem. Res. 2014;53:16229–16234. [Google Scholar]
  37. Zhang H., Zou Y., Peng Y. Influence of sulfation on CeO2-ZrO2 catalysts for NO reduction with NH3. Chin. J. Catal. 2017;38:160–167. [Google Scholar]
  38. Zhang Y., Yue X., Huang T., Shen K., Lu B. In situ DRIFTS studies of NH3-SCR mechanism over V2O5-CeO2/TiO2-ZrO2 catalysts for selective catalytic reduction of NOx. Materials. 2018;11:1307. doi: 10.3390/ma11081307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Zheng L., Zhou M., Huang Z., Chen Y., Gao J., Ma Z., Chen J., Tang X. Self-protection mechanism of hexagonal WO3–based deNOx catalysts against alkali poisoning. Environ. Sci. Technol. 2016;50:11951–11956. doi: 10.1021/acs.est.6b03203. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Transparent Methods, Figures S1–S18, and Table S1
mmc1.pdf (1.5MB, pdf)

Data Availability Statement

All relevant data are available from the corresponding author (dszhang@shu.edu.cn) upon reasonable request.

Articles from iScience are provided here courtesy of Elsevier

RESOURCES