Mechanism of SO₂/H₂O enhanced rare earth tailings catalysts in NH₃-SCR at medium and high temperature

Abstract

Rare earth tailings (RET) NH₃-SCR catalysts were prepared by mechanical and microwave activation of a large amount of rare earth tailings after beneficiation of Bayan Ebo rare earth ore. The effects of SO₂/H₂O on the denitrification performance of the RET catalysts were evaluated by conducting denitrification activity tests, SO₂/H₂O tolerance tests and in situ DRIFTs mechanistic analysis. The results showed that the denitrification activity was significantly increased in the presence of SO₂/H₂O. And in situ DRIFTs analysis showed that in the presence of SO₂/H₂O, SO₂ could be adsorbed as SO₃²⁻ groups by the hydroxyl groups on the catalyst surface and react with SO₄²⁻ to form S₂O₇²⁻ species. And in the presence of NH₃, S₂O₇²⁻ would decompose into unstable SO₄²⁻ species and SO₃²⁻ and continue to react cyclically to form S₂O₇²⁻ species, providing the RET catalyst provides more acid sites, facilitating the SCR reaction.

---

1 Introduction

NO_x is considered to be one of the serious air pollutants and its main source is industrial combustion emissions from fossil fuels in thermal power plants. NH₃-SCR technology is an effective method for NO_x removal.^1–3 V₂O₅–WO₃(MoO₃)/TiO₂ has been used industrially for many years,^4,5 but its applicable denitrification temperature range is narrow. To broaden or reduce the applicable temperature range, researchers have carried out work on the preparation of NH₃-SCR catalysts using transition metal elements,^6–13 which mainly focused on Mn-based, Fe-based and rare earth-based catalysts. It was found that Mn-based catalysts have a good De-NO_X activity in the low temperature band but poor SO₂/H₂O tolerance.^6–8 Fe-based catalysts have some activity (lower than Mn-based catalysts) in the medium to high temperature band (150–450 °C) and have better SO₂ tolerance.^9–13 Single metal oxide catalysts usually have certain defects, so they are usually loaded with other elements to form composite metal oxides for the preparation of the NH₃-SCR catalysts to compensate for each other's defects.

Bayan Ebo has a large amount of rare earth tailings after beneficiation, which contains a rich variety of NH₃-SCR active elements (Fe, Ce, Mn),¹⁴ of which rare earth elements are excellent NH₃-SCR catalyst active components.¹⁵ In addition, there are congenial relationships such as adjacency, leaching and encapsulation of various active minerals in rare earth tailings, and the interactions that exist between these congenial minerals can also facilitate the SCR reaction process,^16,17 so rare earth tailings are naturally excellent materials that can be used to prepare NH₃-SCR catalysts.

SO₂/H₂O poisoning phenomenon seriously affects the activity of SCR catalysts. Due to the increase of HSO₄⁻ and SO₄²⁻ on the catalyst surface, it often leads to accumulation of ammonium sulfate on the catalyst surface and irreversible sulfation of the active components of the catalyst leading to decreased activity. So improving the SO₂/H₂O tolerance of catalysts is directly related to the practical application of catalysts and is an unavoidable indicator in the evaluation of NH₃-SCR catalysts. Composite oxide catalysts doped with rare earth elements exhibit better SO₂/H₂O tolerance and denitrification efficiency compared to Fe-based and Mn-based denitrification catalysts.^18–21 In contrast, rare earth tailings with multiple active components (Fe, Ce, Mn) are natural composite oxides, and rare earth tailings catalysts have better SO₂/H₂O tolerance performance than most composite oxide catalysts.

Therefore, it is necessary to investigate the SO₂/H₂O tolerance characteristics of the Bayan Ebo rare earth tailings catalysts (RET) in the NH₃-SCR denitrification process and to investigate the SO₂/H₂O tolerance mechanism in the denitrification process. This paper explores the process and tolerance mechanism of SO₂/H₂O on the denitrification performance of the RET catalysts and expands the research field of NH₃-SCR denitrification with rare earth tailings catalysts from Bayan Ebo.

2 Materials and methods

2.1 Catalysts preparation

The rare earth tailings catalyst was prepared by ball mills. And the rare earth tailings come from Bayan Ebo (Baotou, China). The beneficiated rare earth tailings material was fixed in a canister on a planetary ball mill and ball milled at 300 rpm for 2 h. The obtained material was passed through a 200 mesh sieve and microwave roasted at 1100 w, 250 °C for 20 min to obtain the catalyst powder. The elemental content of the catalyst was quantified by XRF. As shown in Table 1, the results of the XRF analysis, in which Fe, Ca, Si, Mg, Ce and Al were the major elements with a total content of 50.4%, and the remaining 49.6% are Mn and other trace elements, and the content of each trace element is <1%. The Bayan Ebo rare earth tailings catalyst is indicated as RET.

Table 1 Rare earth tailings catalyst (RET) XRF results

Element	Fe	Ca	Si	Mg	Ce	Al	Mn	Other
Percentage (%)	17.394	17.798	8.804	3.104	1.783	1.485	0.926	48.706

---

2.2 Catalytic performance experiment

The NH₃-SCR performance experiments with the RET catalysts were carried out in a SCR denitrification evaluation unit, as shown in Fig. 1. The denitrification evaluation unit is consisted of gas distribution system (1–4), reaction system (5–8) and analysis system (9–11), and is equipped with a tail gas treatment unit (12). The sample is fixed in a 9 mm inner diameter quartz fixed bed reaction tube. The reaction gas mixture consists of NO (500 ppm), NH₃ (500 ppm), H₂O (6 vol% when in use), SO₂ (500 ppm when in use), O₂ (6 vol%) and N₂ are controlled by the gas distribution system at a flow rate (100 mL min⁻¹), and the GHSV are set to 30 000 h⁻¹. The gas mixture were measured by on-line multi-component flue gas analyzer (HP-OMGA, China), the measurement accuracy is ≤± 2% FS, and the NO_X conversion is obtained from the formula (1):

Fig. 1 Schematic diagram of the NH₃-SCR evaluation unit ((1) gas flow control system; (2) heating water tank; (3) gas distribution tank; (4) heating pipes; (5) heating furnace; (6) quartz tubes; (7) RET catalyst; (8) quartz wool; (9) rotameter; (10) condensation drying device; (11) flue gas online detection device; (12) tail gas treatment device).

2.3 Catalysts characterization

The RET catalysts were characterization by the X-ray diffraction (XRD), temperature programmed reduction by H₂ and NH₃ (H₂-TPR and NH₃-TPD), and in situ Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS).

3 Results and discussion

The RET catalysts after denitrification under different operating conditions were characterised as sample 0, 1, 2 and 3. The total flow rate for all operating condition cases is 100 mL min⁻¹, as shown in Table 2.

Table 2 Working conditions for the preparation of the four RET catalysts

Catalysts	NH3/ppm	NO/ppm	O2/%	H2O/%	SO2/ppm
Sample 0	500	500	6	—	—
Sample 1	500	500	6	6	—
Sample 2	500	500	6	—	500
Sample 3	500	500	6	6	500

---

3.1 Structural changes of catalysts

As shown in Fig. 2, the diffraction peaks of fluorite (CaF₂), quartz (SiO₂), aluminium trioxide (Al₂O₃), iron dolomite {Ca(MgFe)[(CO₃)₂]}, hematite (Fe₂O₃) and cerium fluoride (CeCO₃F) were mainly detected on the surface of sample 0, where the Ca(MgFe)[(CO₃)₂], Fe₂O₃ and CeCO₃F diffraction peaks are overlap, it is due to the natural action of these reactive minerals to form a solid solution structure.^18,19 In comparison to sample 0, no significant crystalline phase changes were observed for sample 1 when the catalyst was reacted under H₂O conditions for 24 hours. However, when the catalyst was reacted under SO₂ or SO₂ and H₂O conditions for 24 hours, a decrease in the intensity of the Ca(MgFe)[(CO₃)₂] and Fe₂O₃ diffraction peak intensities was found for samples 2 and 3, but no characteristic metal sulphate peaks were detected, indicating that the SO₂ sulphide phase was present in an amorphous form and could not be observed by XRD. However, in the presence of SO₂ or SO₂/H₂O, new diffraction peaks were observed for samples 2, 3 at 18.43°, 24.17°, 24.84°, 33.14°, 37.02° and 50.68°, which can be attributed to sulphonate (NH₃SO₃) (PDF#08-0483). And these diffraction peaks overlap with or are similar to the active mineral diffraction peaks of Ca(MgFe)[(CO₃)₂], Fe₂O₃ and CeCO₃F. It can be inferred that the newly formed crystalline phase NH₃SO₃ is produced because the acid groups formed on the catalyst surface after SO₂/H₂O is involved in the denitrification process and during the drop to low temperature interact with the NH₃ adsorbed species and a reduction reaction occurs to reduce the active metal centres in the active minerals to their original state.^22–24

Fig. 2 Rare earth tailings catalyst (RET) XRD results.

3.2 Catalyst performance change

To examine the effect of SO₂/H₂O on the NH₃ adsorption of the RET catalyst as shown in Fig. 3. The peak temperatures of the desorption peaks of sample 0 were 147 °C (weak acid centers), 356 °C (medium centers) and 480 °C (strong centers).²⁵ The NH₃ adsorption capacity of samples 1, 2 and 3 was significantly higher than sample 0, while the NH₃ adsorption capacity of samples 2 and 3 was slightly higher than sample 1. Sample 0 showed a desorption peak at 147 °C corresponding to the weak acid site, a desorption peak at 356 °C corresponding to at the medium to strong acidic site, and a desorption peak at 480 °C corresponding to the strong acidic site. At 480 °C corresponds to the desorption peak from the strongly acidic site. Samples 1, 2 and 3 also showed three peaks in the low and medium-high temperature sections, each acidic site being shifted back and enhanced compared to sample 0, indicating that the presence of a suitable amount of SO₂/H₂O can strengthen the acidic sites on the RET catalyst. The peak at 150 °C for sample 3, which was influenced by the combined presence of SO₂/H₂O, was attributed to the weakly acidic sites, with a significantly greater ability to adsorb NH₃ on the weakly acidic sites relative to the RET catalysts influenced by the presence of SO₂ or H₂O alone with larger desorption peaks at 365 °C and 489 °C attributed to the moderately strong and strong acidic sites. Significantly larger peak areas in the mid to high temperature fractions compared to samples 1 and 2, which were affected by SO₂/H₂O alone. It means that co-modification of SO₂ and H₂O can also substantially enhance the ability of medium to strong acidic sites to adsorb NH₃. Overall, the RET catalyst with the influence of SO₂/H₂O both increased the acidic sites and improved NH₃ adsorption capacity. Combined with XRD analysis, it was found that the increase in acidic sites after the catalyst was affected by SO₂/H₂O can be attributed to the formation of NH₃SO₃, which is readily decomposed by heat, thus providing more acidic sites.

Fig. 3 Rare earth tailings catalyst (RET) NH₃-TPD results.

H₂-TPR tests were using to investigate the effect of the SO₂/H₂O on the redox ability of the RET catalyst. In Fig. 4, the O_a peak located below 400 °C is the reduction peak of oxygen species adsorbed on the catalyst surface. The O_b1 peak located between 400 and 600 °C is the peak of H₂ consumption (Fe₂O₃ to Fe₃O₄ and the reduction of the CeO₂).^26,27 The O_b2 peak between 600 and 700 °C is the peak of H₂ consumption (Fe₃O₄ to FeO).²⁶ The O_b3 peak above 700 °C is the peak of H₂ consumption (FeO to Fe).²⁶ The reduction peaks of the RET catalysts with SO₂ or SO₂/H₂O are shifted towards a lower temperature region and the peak area is slightly increased compared to the RET catalysts without the influence of SO₂ or H₂O. This means that SO₂ and H₂O can influence the RET catalysts to induce a beneficial change in redox capacity towards a more beneficial catalytic reduction reaction occurring, with new shoulder peaks are due to the reduction of the CeO₂ at 693 °C and 695 °C,²⁷ and the introduction of SO₂/H₂O leads to a more pronounced shoulder peak shape at 695 °C. Which combined with XRD analysis, is presumed to be due to the influence of SO₂, which induces the reduction of RET catalysts with the natural co-occurrence of active minerals associated properties of the RET catalyst resulted from the joint participation of multiple substances in the redox reaction.²⁸ From the H₂-TPR results, electron transfer paths can be derived (Fe³⁺ ↔ Fe²⁺ and Ce⁴⁺ ↔ Ce³⁺).

Fig. 4 Rare earth tailings catalyst (RET) H₂-TPR results.

The surface of the rare earth tailings catalyst (Fig. 5a) showed an uneven and pitted morphology, and the surface of the rare earth tailings catalyst (Fig. 5b) did not change much after the H₂O tolerance test. After being affected by SO₂ alone and SO₂/H₂O tolerance, the catalysts (Fig. 5c and d) showed some mobility, which was in the form of stacked lamellar fragmentation with a few fine particles attached to the surface, after being affected by SO₂/H₂O tolerance, there was a large volume of lamellar particles stacked on the surface of the catalysts, and after the catalysts were affected by the simultaneous presence of SO₂ and H₂O, the lamellar particles were attached to the surface of the catalysts.

Fig. 5 SEM and EDS surface sweeps before and after the effect of SO₂/H₂O on RET catalysts ((a) fresh RET catalyst; (b) after H₂O; (c) after SO₂; (d) after SO₂/H₂O).

3.3 In situ DRIFTs analysis

3.3.1 Effect of H₂O on the adsorption of SO₂ on the catalyst surface. The adsorption of SO₂ in the presence of H₂O at 350 °C was investigated by in situ DRIFTS, as shown in Fig. 6.

Fig. 6 Effect of H₂O on the adsorption of SO₂ by the RET catalyst.

As shown in Fig. 6, the RET catalyst was exposed to SO₂/O₂ for 60 min and treated in a pure N₂ gas stream (100 mL min⁻¹) at 350 °C for a further 30 min. 5 min after the addition of H₂O, an absorption peak of molecular water (1647 cm⁻¹) appeared, with little change in peak intensity with increasing time, meaning that the H₂O tolerance of the RET catalyst was good. The absorption peak of sulfite (1434 cm⁻¹) disappeared with increasing time and the absorption peak of pyrosulfate (1396 cm⁻¹) gradually became stronger. The absorption peaks of bridged tri-dentate sulphate (1326 cm⁻¹), double-dentate sulphate (1076 cm⁻¹), bridged double-dentate sulphate (1197 cm⁻¹) and single-dentate sulphate (1261 cm⁻¹) showed little change.^31,32,35,36 The crystalline water variable angle vibrational peak shifted from 1535 to 1581 cm⁻¹, it is due to the chemisorption of gaseous SO₂ onto the catalyst surface via hydroxyl groups.³⁷ The increase in persulphate is due to the presence of H₂O, which causes the formation of hydroxyl groups on the RET catalyst surface, and adsorb more SO₂ to form SO₃²⁻, which in turn reacts with the catalyst surface sulphate species to form S₂O₇²⁻, so H₂O promotes the formation of S₂O₇²⁻ species on the RET catalyst surface.

3.3.2 Effect of SO₂/H₂O on surface acidity. Fig. 7 shows that NH₃ was first adsorbed without SO₂ for 60 min at 350 °C and then treated in a pure N₂ gas stream (100 mL min⁻¹) for a further 30 min.

Fig. 7 Effect of SO₂/H₂O on catalyst NH₃ adsorption.

Fig. 7 shows the absorption peaks of NH₄⁺ adsorption at the Brønsted acid site (1392, 1427, 1461, 1500 cm⁻¹), NH₃ adsorption at the Lewis acid site (1025, 1157, 1228, 1620 cm⁻¹) and it causes NH₃ form –NH₂ (1332 cm⁻¹).^29,30 After continuing the ammonia stream and 5 min of SO₂, H₂O and O₂, the absorption peaks of NH₄⁺ adsorption (1392, 1461 cm⁻¹) and NH₃ adsorption at the Lewis acid site (1157, 1228, 1620 cm⁻¹) disappeared. The intermediate product –NH₂ (1529 cm⁻¹)^29,30 and NH₃ adsorption (1600 cm⁻¹)^29,30 with increasing peak intensity with time, meaning that the presence of SO₂/H₂O promoted NH₃ adsorption. Bidentate sulfate (1068 cm⁻¹), bridged bidentate sulfate (1191 cm⁻¹), and monodentate sulfate (1265 cm⁻¹) absorption peaks were also present after 40 min.^31,35,36 The intensity of the bidentate and bridged bidentate sulfate absorption peaks frequently changes but no pyrosulfate species were found at 1391 cm⁻¹. In combination with XRD analysis, was attributed to the formation of NH₃SO₃ from NH₃ adsorbed species bound to the catalyst surface. After 40 min of addition of SO₂, H₂O and O₂, the intermediate product –NH₂ (1300, 1572 cm⁻¹) appeared, resulting from the creation of a new Lewis acid site, which is consistent with NH₃-TPD results. SO₂/H₂O caused the formation of an unstable sulphate species, which adsorbed on the active metal sites are strongly acidic, so this unstable sulphate species provides more acidic sites for the NH₃-SCR process and is less likely to form ammonium sulphate species with NH₃. The combination of NH₃ with catalyst surface species to form NH₃SO₃ resulted in the disappearance of persulphate species from the RET catalyst surface, indicating that the presence of NH₃ promoted the decomposition of S₂O₇²⁻.

3.3.3 Effect of SO₂ on the adsorbed NO species. As in step with Fig. 8, the in situ DRIFTS spectra showed the effect of SO₂/H₂O on NO adsorption on the RET catalyst at 350 °C.

Fig. 8 Effect of SO₂/H₂O on NO adsorption by RET catalysts.

In Fig. 8, the addition of NO and O₂ gas streams followed by a pure N₂ gas stream shows monodentate nitrite (1027, 1091, 1159 cm⁻¹), bridged nitrite (1234 cm⁻¹), intermediate product-NO₂ (1338, 1399, 1458 cm⁻¹), intermediate product-NO₃ (1521 cm⁻¹) and adsorbed NO₂ (1593, 1650 cm⁻¹).^33,34 After simultaneous introduction of NO, O₂, H₂O and SO₂ for 5 min, the absorption peak of bridged nitrite shifted from 1234 to 1218 cm⁻¹ with little change in peak intensity with increasing time, with bridged nitrite still occupying the active site. After 10 min of addition of H₂O and SO₂, the absorption peaks of –NO₂ (1399 cm⁻¹) and adsorbed NO₂ (1650 cm⁻¹) disappeared and absorption peaks of bidentate sulfate (1056 cm⁻¹), pyrosulfate (1393 cm⁻¹) and NO₂ (1697 cm⁻¹) generated by the reaction of adsorbed NO on the surface appeared.^33,34,38 In the presence of SO₂/H₂O, the absorption peak of S₂O₇²⁻ species gradually increased with time and the –NO₂ disappeared, due to the competition between SO₂ and NO adsorption on this active site to produce S₂O₇²⁻ species. It occupies part of the NO adsorption site, and the increase in surface active oxygen species due to SO₂ acidification of the RET catalyst, converting more NO to NO₂ and hence the absorption peak of NO₂.

3.3.4 Transient reactions of NH₃ and NO on the catalyst surface. As shown in Fig. 9a, the in situ DRIFTs spectra of the reaction of NO with pre-adsorbed NH₃ species on the RET catalyst surface under the influence of SO₂/H₂O and Fig. 9b shows the in situ DRIFTs spectra of the reaction of NH₃ with pre-adsorbed NO species on the RET catalyst surface under the influence of SO₂/H₂O.

Fig. 9 RET Catalytic transient reaction in the presence of SO₂/H₂O ((a) NH₃/NO; (b) NO/NH₃).

As shown in Fig. 9a, the RET catalysts were firstly exposed to NH₃ gas flow for 60 min in the addition of SO₂/H₂O, and then purged with N₂ at 350 °C for 20 min.

Fig. 9a shows the absorption peaks of NH₄⁺ adsorption (1392, 1438, 1485, 1513, 1727 cm⁻¹), NH₃ adsorption (1150, 1215, 1616 cm⁻¹) and –NH₂ (1334, 1560 cm⁻¹).^29,30 After 5 min of the addition of NO and O₂, the NH₃ adsorbed species disappeared and peaks appeared for monodentate nitrite (1030, 1097 cm⁻¹), bridged nitrite (1230 cm⁻¹), intermediate product –NO₂ (1355, 1415, 1618 cm⁻¹), bidentate nitrate (1569 cm⁻¹) and NO₂ (1699 cm⁻¹).^33,34 The peak of –NO₂ (1415 cm⁻¹) disappeared rapidly with increasing time. The peak pattern of bridged nitrite shifted from 1230 to 1240 cm⁻¹ with little change in peak intensity. Monodentate nitrite (1097 cm⁻¹) and bidentate nitrate (1569 cm⁻¹) showed little change. Absorption peaks for intermediate product-NO₂ (1305, 1398 cm⁻¹) and intermediate product-NO₃ (1502 cm⁻¹) appeared after 20 min of the addition of NO/O₂. Absorption peaks of –NO₃ at 1519, 1508, 1485 and 1533 cm⁻¹ appeared, with significant changes in peak pattern with increasing time. The absorption peaks of –NO₂ did not change much after 20 min of NO/O₂ addition, and the peak intensity increased. This is due to the fact that when NO/O₂ were first added, the NH₃ adsorbed species mainly combined with –NO₂ to form the intermediate product, and after 20 min as NO was oxidized to more NO₂, the –NO₃ group was generated to participate in the reaction. The results show that –NO₂ is easier to combine with NH₄⁺ than –NO₃ to form the intermediate product NH₄NO₂. And when NH₃ is present, the NO species mainly form the –NO₂ group directly to participate in the reaction, without oxidation between the catalyst and NO₂ to form the –NO₃ group, shortening the reaction process in favor of the NH₃-SCR reaction.

In Fig. 9b, the samples were first exposed to NO and O₂ for 60 min in the presence of SO₂/H₂O, and then treated in a pure N₂ gas stream (100 mL min⁻¹) at 350 °C for a further 30 min. The appearance of monodentate nitrite (1030, 1090, 1147 cm⁻¹) was observed in Fig. 9b, with bridging nitrite (1230 cm⁻¹), –NO₂ (1285, 1340, 1400, 1460 cm⁻¹), –NO₃ (1521 cm⁻¹) and adsorbed NO₂ (1594, 1654 cm⁻¹) absorption peaks.^33,34 After 5 min of the addition of NH₃, the peaks of the nitrate-like adsorbed species, adsorbed state NO₂, intermediate products –NO₂ and –NO₃ disappeared, along with the peaks of adsorbed NH₄⁺ (1370, 1425, 1494 cm⁻¹), adsorbed NH₃ (1080, 1257 cm⁻¹) and –NH₂ (1309, 1530, 1560 cm⁻¹),^29,30 where the absorption peak of –NH₂ increased in intensity with time. The peak of adsorbed NH₃ was less variable, and the peak of adsorbed NH₄⁺ disappeared quickly. Absorption peaks for NH₄⁺ (1385, 1419, 1456, 1467, 1495 cm⁻¹) and –NH₂ (1625 cm⁻¹) adsorbed at the Brønsted acid site appeared after 20 min of the addition of NH₃.^29,30 When NH₃ was present, the –NO₂ and –NO₃ groups as well as the adsorbed NO₂ disappeared from the catalyst surface and NH₄⁺ and –NH₂ groups were generated at subsequent time periods of the reaction, indicating that intermediate products NH₄NO₃, NH₄NO₂ and NH₂NO₂ were generated on the RET catalyst surface to remove NO adsorption products, suggesting that both the E–R mechanism and the L–H mechanism, which acted together to remove NO_X through both pathways.

4 Denitrification mechanism of SO₂/H₂O enhanced rare earth tailings catalyst

Combining the results of in situ DRIFTs analysis and SO₂ characterization,^22,23,38,39 We can surmise the following reaction mechanism (Fig. 10) as well as reaction equations (eqn (2)–(16)) can be derived for the effect of SO₂/H₂O on the RET catalysts. As shown in Fig. 10, combined with the analysis of the H₂-TPR results, it is found that the conversion between Fe³⁺ ↔ Fe²⁺ and Ce⁴⁺ ↔ Ce³⁺ on the RET catalyst surface provides the active sites for SCR catalytic denitrification, and the conversion between Ce³⁺ and Ce⁴⁺ mainly acts as oxygen storage and release (eqn (2)–(5)). When SO₂/H₂O is added, SO₂ is adsorbed by the hydroxyl groups formed by H₂O and reacts with O₂ to form SO₃²⁻ groups, while combining with SO₄²⁻ groups to form S₂O₇²⁻ (eqn (7)–(9)). When NH₃ and NO are present, they are converted to NH₄⁺ and NO₂⁻ respectively, which are then reduced to N₂ and H₂O (eqn (10)–(15)). The adsorption of H₂O by the catalyst surface increases the –OH sites on the catalyst surface, leading to easier adsorption and conversion of SO₂ to form SO₃²⁻, while the S₂O₇²⁻ group is converted to NH₃SO₃ and the unstable SO₄²⁻ group by the reduction of NH₃ (eqn (16)), allowing the catalyst reactive sites on the surface are restored, inhibiting the continued sulphation of the catalyst surface. NH₃ also adsorbs to the Brønsted and Lewis acidic sites on the RET catalyst surface as –NH₂ and NH₄⁺, and perform NH₃-SCR reaction.

Fe³⁺ ↔ Fe²⁺ + △ (2)

O_2(g) + 2△ → 2O − △ (3)

Ce³⁺ + O_2(g) ↔ Ce⁴⁺ + 2O⁻ (4)

O⁻ + △ → O − △ (5)

H₂O_(g) + △ → △ − OH + H⁺ (6)

△ − OH + SO_2(g) → △ − SO²⁻₃ + H₂O (7)

2SO_2(g) + 3/2O_2(g) + H₂O → 2SO₄²⁻ + 2H ⁺ (8)

SO₄²⁻ + △ − SO₃²⁻ → △ − S₂O₇²⁻ (9)

△ − S₂O₇²⁻ + H⁺ → △ − S₂O₇^2–H⁺ (10)

△ − S₂O₇²⁻H⁺ + NH_3(g) → △ − S₂O₇²⁻NH₄⁺ (11)

△ − S₂O₇²⁻NH₄⁺ + NO_(g) + 1/2O_2(g) → △–NO₂⁻S₂O₇^2–NH₄⁺ (12)

△ − NO₂⁻S₂O₇^2–NH₄⁺ → △ − S₂O₇²⁻ + N_2(g) + 2H₂O (13)

△ − S₂O₇²⁻ + NH_3(g) → △ − H₂NS₂O₇²⁻H⁺ (14)

△ − H₂NS₂O₇²⁻H⁺ + NO_(g) → △S₂O₇²⁻H⁺ + N_2(g) + H₂O (15)

△ − S₂O₇²⁻H⁺ + NH_3(g) + H⁺ → NH₃SO₃ + SO₄²⁻ + △ (16)

where △ represents the surface oxygen vacancy in the active metal center. The symbol (g) represents NO, NH₃, SO₂, NO₂, N₂, and O₂ in the gaseous state. The symbol (a) represents the adsorbed state of NH₃, NH₄⁺ and SO₃ on the catalyst surface.

Fig. 10 Mechanism of SO₂/H₂O influence on the surface of rare earth tailings (RET) catalysts (△ represents the active metal sites on the RET catalyst surface; blue arrows represent the gas absorption and release process; black arrows represent the internal electron transfer process of the catalyst; red arrows represent the reaction cycle process of the S₂O₇²⁻ group).

5 Catalyst performance

5.1 Effect of H₂O and SO₂ on catalytic NO_X reduction

The effect of SO₂/H₂O on the catalytic denitrification activity of the RET catalyst was studied in the temperature range of 50–450 °C with φNO, φNH₃ and φSO₂ of 500 × 10⁻⁶, φO₂ of 6 × 10⁻² and φH₂O of 6 × 10⁻². After the SCR reactor outlet φNO was stabilized for 10 min, the NH₃ was started to be introduced and the denitrification rate for each temperature section was calculated as the average of the NO_X outlet concentration values within 180 min of stabilization were calculated and the results, as shown in Fig. 11.

Fig. 11 RET catalyst NO_X removal rate.

Fig. 11 shows that the NO_X conversion of the RET catalyst in the absence of SO₂/H₂O and SO₂ is much lower than that in the presence of SO₂ and H₂O, and H₂O has almost no effect on the NO_X conversion of the RET catalyst. At relatively low temperatures (50 to 200 °C), the enhancement of catalytic activity in the presence of SO₂ and H₂O was similar, about 10 ± 1%. However, when the reaction temperature is in the medium to high temperature range (250–450 °C), the activity enhancement under SO₂/H₂O can reach 16 ± 1%, and the activity enhancement of the RET catalyst in SO₂ only condition is only 10 ± 1%. Furthermore, at 350 °C, SO₂/H₂O in the gas supply leads to a significant increase in SCR activity of about 18 ± 1%, with a NO_X conversion of 66 ± 1% for the RET catalyst. This enhancement is due to SO₂ gas phase acidification of the active metal oxides under SO₂.³⁷ It suggests that the appropriate conditions of SO₂/H₂O presence are favorable to improve the catalytic activity of the RET catalysts.

With the same steps, the stability of the effect of SO₂/H₂O on the NO_X conversion in the temperature range of 350 °C of the RET catalyst over time is shown in Fig. 12 (the data points in the figure are the average values of NO_X conversion over 30 min).

Fig. 12 Stability of RET catalyst NO_X removal rates under the influence of SO₂/H₂O.

Fig. 12 showed stable catalytic activity at 350 °C during the 72 h test. When SO₂ was added to the gas stream, the NO_X conversion increased from a stable 49 ± 1% in the first 3 h to obtain a stable NO_X conversion of 62 ± 1% in 24 h, indicating that SO₂ has a significant effect on the catalyst activity. The addition of SO₂/H₂O increases the catalytic activity from 49 ± 1% to 66 ± 1%. However, when H₂O is removed from the gas, the activity can be restored, implying that the enhancement effect of H₂O in promoting SO₂ influence on the activity is reversible. The NO_X conversion of the RET catalyst can be recovered when there is no SO₂ or H₂O in the reaction gas mixture, which proves that the RET catalyst has a good effect on SO₂/H₂O, and the SCR activity of the RET catalyst can be obtained at 350 °C with a 17 ± 1% enhancement under the conditions of both SO₂ and H₂O.

6 Conclusion

In summary, this paper prepared rare earth tailings (RET) catalysts from Bayan Ebo rare earth tailings by ball milling and microwave roasting, and analyzed the SO₂/H₂O tolerance mechanism of the RET catalysts by activity testing, characterization and in situ DRIFTs experiments. And revealed the mechanism of SO₂/H₂O to enhance the denitrification performance of catalysts by promoting the cyclic reaction of S₂O₇²⁻ groups on the RET catalysts surface. Although RET catalysts are less active, they can be used as a catalyst active carrier through their excellent SO₂ tolerance properties.

(1) The effect of SO₂/H₂O on the RET catalyst was facilitated, and the NO_X conversion of the RET catalyst increased from 48% to 66 ± 1% at 350 °C under SO₂/H₂O, which was higher than that of SO₂ or H₂O alone.

(2) Rare earth tailings catalyst NH₃-SCR denitrification followed the combined L–H and E–R mechanism. The Feⁿ⁺–Ceⁿ⁺ coactivation of the RET catalyst provides Brønsted and Lewis acid sites.

(3) The effect of SO₂ on the RET catalyst is mainly the formation of SO₄²⁻ and SO₃²⁻ species, while H₂O promotes the formation of hydroxyl groups and facilitates SO₂ adsorption to form SO₃²⁻, which promotes the reaction of SO₃²⁻ with SO₄²⁻ to form S₂O₇²⁻ groups, which provide acid sites for the RET catalyst and improve the adsorption of NH₃. At the same time, the S₂O₇²⁻ group decomposes into SO₃²⁻ and unstable SO₄²⁻ groups, thus continuing the cyclic reaction to form S₂O₇²⁻groups, thereby increasing the denitrification activity of the RET catalyst activity and inhibit the continued sulphation of the catalyst surface.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The research was financially supported by the Natural Science Foundation of Inner Mongolia (Grant No. 2019ZD13).

References

1. K. Skalska, J. S. Miller and S. Ledakowicz, Trends in NOx abatement: A review, Sci. Total Environ., 2010, 408(19), 3976–3989.
2. G. L. Manney, L. Froidevaux, J. W. Waters, R. W. Zurek, W. G. Read, L. S. Elson, J. B. Kumer, J. L. Mergenthaler, A. E. Roche, A. O'Neill, R. S. Harwood, I. Mackenzie and R. Swinbank, Chemical depletion of ozone in the Arctic lower stratosphere during winter 1992-93, Nature, 1994, 370(6489), 429–434.
3. D. Maric and J. P. Burrows, Formation of N₂O in the photolysis/photoexcitation of NO, NO₂ and air, J. Photochem. Photobiol., A, 1992, 66(3), 291–312.
4. H. Bosch and F. Janssen, Catalytic reduction of nitrogen oxides-A review on the fundamental and technology, Catal. Today, 1988, 2(4), 369–532.
5. G. Busca, L. Lietti, G. Ramis and F. Berti, Chemical and mechanistic aspects of the selective catalytic reduction of NOx by ammonia over oxide catalysts: A review, Appl. Catal., B, 1998, 18(1–2), 1–36.
6. F. Kapteijn, L. Singoredjo and A. Andreini, Activity and selectivity of pure manganese oxides in the selective catalytic reduction of nitric oxide with ammonia, Appl. Catal., B, 1994, 3(1), 173–189.
7. T. S. Park, S. K. Jeong, S. H. Hong and S. C. Hong, Selective catalytic reduction of nitrogen oxides with NH₃ over natural manganese ore at low temperature, Ind. Eng. Chem. Res., 2001, 40(21), 4491–4495.
8. L. Singoredjo, R. Korver and F. Kapteijin andJ.Moulijn, Alumina supported manganese oxides for the low-temperature selective catalytic reduction of nitric oxide with ammonia, Appl. Catal., B, 1992, 1(4), 297–316.
9. C. Liu, S. Yang, L. Ma, Y. Peng, A. Hamidreza, H. Chang and J. Li, Comparison on the Performance of alpha-Fe₂O₃ and gamma-Fe₂O₃ for Selective Catalytic Reduction of Nitrogen Oxides with Ammonia, Catal. Lett., 2013, 143(7), 697–704.
10. A. Kato, S. Matsuda, T. Kamo, F. Nakajima, H. Kuroda and T. Narita, Reaction between NO_X and NH₃ on iron oxide-titanium oxide catalyst, J. Phys. Chem., 1981, 85(26), 276–287.
11. F. Liu, H. He and C. Zhang, Novel iron titanate catalyst for the selective catalytic reduction of NO with NH₃ in the medium temperature range, Chem. Commun., 2008, 164(17), 2043–2045.
12. F. Liu, H. He, C. Zhang, Z. Feng, L. Zheng, Y. Xie and T. Hu, Selective catalytic reduction of NO with NH₃ over iron titanate catalyst: Catalytic performance and characterization, Appl. Catal., B, 2010, 96(3), 408–420.
13. F. Liu, K. Asakura, H. He, Y. Liu, W. Shan, X. Shi and C. Zhang, Influence of calcination temperature on iron titanate catalyst for the selective catalytic reduction of NO_X with NH₃, Catal. Today, 2011, 164(1), 520–527.
14. L. Hou, S. Fu, X. Yan, C. Qiao and W. Wu, Effects of sulfate on denitration performance of rare earth tailings for selective catalytic reduction of NO with NH₃, Mol. Cryst. Liq. Cryst., 2023, 755(1), 91–106.
15. P. Zhang, Study on the process mineralogy of large-sized Bayan Obo Fe-REE-Nb deposit, J. Chin. Rare Earth Soc., 1991, 9(4), 350–353.
16. J. Wang, C. Zhu, B. Li, Z. Gong, Z. Meng, G. Xu and W. Wu, Prepare a catalyst consist of rare earth minerals to denitrate via NH₃-SCR, Green Process. Synth., 2019, 9(1), 191–202.
17. X. Bai, J. Lin, Z. Chen, L. Hou and W. Wu, A Study on the Effect of Different Ball Milling Methods on the NH₃-SCR Activity of Aluminum-Laden Bayan Obo Tailings, Catalysts, 2021, 11(5), 568.
18. G. Qi and R. Yang, Low-temperature selective catalytic reduction of NO with NH₃ over iron and manganese oxides supported on titania, Appl. Catal., B, 2003, 44(3), 217–225.
19. Z. Chen, F. Wang, H. Li, Q. Yang, L. Wang and X. Li, Low-Temperature Selective Catalytic Reduction of NO_X with NH₃ over Fe-Mn Mixed-Oxide Catalysts Containing Fe₃Mn₃₀s Phase, Ind. Eng. Chem. Res., 2012, 51(1), 202–212.
20. B. Shen, T. Liu, N. Zhao, X. Yang and L. Deng, Iron-doped Mn-Ce/TiO₂ catalyst for low temperature selective catalytic reduction of NO with NH₃, J. Environ. Sci., 2010, 22(9), 1447–1454.
21. F. Wang, B. Shen, S. Zhu and Z. Wang, Promotion of Fe and Co doped Mn-Ce/TiO₂ catalysts for low temperature NH₃-SCR with SO₂ tolerance, Fuel, 2019, 249, 54–60.
22. E. Hartley and M. J. Matteson, Sulfur dioxide reactions with ammonia in humid air, Ind. Eng. Chem. Fundam., 1975, 14(1), 67–72.
23. Z. Gao, W. Yang, X. Ding, G. Lv and W. Yang, Support effects in single atom iron catalysts on adsorption characteristics of toxic gases (NO₂, NH₃, SO₃ and H₂S), Appl. Surf. Sci., 2018, 436, 585–595.
24. L. Liu and Z. Yuan, Ordered Mesoporous Carbon Materials Synthesized by Organic-Organic Self-Assembly, Prog. Chem., 2014, 26(05), 756–771.
25. X. Zhu, Y. Wang, Y. Huang and Y. Cai, Selective catalytic reduction of NO with NH₃ over Ce-W-Ti oxide catalysts prepared by solvent combustion method, Appl. Sci., 2018, 8(12), 2430.
26. F. Liu and H. He, Structure-Activity Relationship of Iron Titanate Catalysts in the Selective Catalytic Reduction of NO_X with NH₃, J. Phys. Chem. C, 2010, 114(40), 16929–16936.
27. X. Huang, P. Wang, J. Tao and Z. Xi, CeO₂-modified Mn-Fe-O composites and their catalytic performance for NH₃-SCR denitrification, J. Inorg. Mater., 2020, 35(05), 573–580.
28. Q. Fu, J. Wen, L. Li, Y. Xiao and K. Gao, The Study of REE Occurrence State in Bayan Obo Tailings, Chin. Rare Earths, 2017, 38(05), 103–110.
29. J. Sun, Y. Lu, L. Zhang, C. Ge, C. Tang, H. Wan and L. Dong, Comparative Study of Different Doped Metal Cations on the Reduction, Acidity, and Activity of Fe₉M₁O_X (M= Ti⁴⁺, Ce^4+/3+, Al³⁺) Catalysts for NH₃-SCR Reaction, Ind. Eng. Chem. Res., 2017, 56(42), 12101–12110.
30. G. Ramis and M. A. Larrubia, An FT-IR study of the adsorption and oxidation of N-containing compounds over Fe₂O₃/Al₂O₃ SCR catalysts, J. Mol. Catal. A: Chem., 2004, 215(1–2), 161–167.
31. A. A. Ioffe, V. P. Bulatov, V. A. Lozovsky, M. Y. Goldenberg, O. M. Sarkisov and S. Ya, Umansky. On the reaction of the NH₂ radical with SO₂ at 298–363 K, Chem. Phys. Lett., 1989, 156(5), 425–432.
32. D. Ye, R. Qu, C. Zheng, K. Cen and X. Gao, Mechanistic investigation of enhanced reactivity of NH₄HSO₄ and NO on Nb- and Sb- doped V-W/Ti SCR catalysts, Appl. Catal., A, 2018, 549, 310–319.
33. F. Liu and H. He, Selective catalytic reduction of NO with NH₃ over manganese substituted iron titanate catalyst: Reaction mechanism and H₂O/SO₂ inhibition mechanism study, Catal. Today, 2010, 153(3–4), 70–76.
34. L. Chen, X. Yao, J. Cao, F. Yang, C. Tang and L. Dong, Effect of Ti⁴⁺ and Sn⁴⁺ co-incorporation on the catalytic performance of CeO₂-MnO_X catalyst for low temperature NH₃-SCR, Appl. Surf. Sci., 2019, 476, 283–292.
35. F. Gao, X. Tang, H. Yi, S. Zhao, J. Wang and T. Gu, Improvement of activity, selectivity and H₂O&SO₂-tolerance of micro-mesoporous CrMn₂O₄ spinel catalyst for low-temperature NH₃-SCR of NO_X, Appl. Surf. Sci., 2019, 466, 411–424.
36. S. Pan, H. Luo, L. Li, Z. Wei and B. Huang, H₂O and SO₂ deactivation mechanism of MnO_X/MWCNTs for low-temperature SCR of NO_X with NH₃, J. Mol. Catal. A: Chem., 2013, 377, 154–161.
37. T. Gu, L. Yue, H. Wang and Z. Wu, The enhanced performance of ceria with surface sulfation for selective catalytic reduction of NO by NH₃, Catal. Commun., 2010, 12(4), 310–313.
38. W. S. A. El-Yazeed, M. Eladl, A. I. Ahmed and A. A. Ibrahim, Sulfamic: acid incorporated tin oxide: Acidity and activity relationship, J. Alloys Compd., 2021, 858, 158–192.
39. E. Hayon, A. Treinin and J. Wilf, Electronic spectra, photochemistry, and autoxidation mechanism of the sulfite bisulfite pyrosulfite systems. SO2-, SO3-, SO4-, and SO5-radicals, J. Am. Chem. Soc., 1972, 94(1), 47–57.

---