Investigation of the promotion effect of WO₃ on the decomposition and reactivity of NH₄HSO₄ with NO on V₂O₅–WO₃/TiO₂ SCR catalysts

Abstract

In this study, we systematically investigated the interaction between NH₄HSO₄ and WO₃-promoted V₂O₅/TiO₂ catalysts in the selective catalytic reduction of NO with NH₃, along with the promotion effect of WO₃ on the decomposition and reactivity of NH₄HSO₄ with NO. In the NH₄HSO₄-deposited samples, WO₃ addition increased the electron cloud density around the S atoms in SO₄²⁻, which was beneficial for the reduction of S atoms with +6 formal oxidation number, as in NH₄HSO₄, to those in the +4 oxidation state, as in SO₂. Consequently, WO₃-doping led to an increase in the amount of SO₂ released at low-temperature regions during the heating process as well as an obvious enhancement in the decomposition behavior of NH₄HSO₄, as illustrated by decreases in temperatures attributed to the weight loss peaks. Meanwhile, the introduced WO₃ had a slight promotion effect on the reactivity of NH₄HSO₄ with NO, together with an inhibitory effect on the production of N₂O during the reaction process. Finally, WO₃-promoted catalysts exhibited enhanced SO₂-resistance.

---

1. Introduction

The selective catalytic reduction of NO with NH₃ has been proved to be a mature technology for NO_x removal.¹ Recently, developing low-temperature SCR systems is significant, because flue gas becomes much cleaner after passing through desulfurized and particulate control devices, which could protect catalysts from being deactivated by high concentrations of ash and SO₂ in the flue gas, thereby prolonging the catalyst lifespan. Several types of low-temperature SCR catalysts, which exhibit excellent SCR activity in the temperature region of 100–300 °C, have been developed.^2–5 However, certain amount of SO₂ and H₂O in the flue gas leads to the deposition of NH₄HSO₄ on the catalysts, which adversely affects the SCR reaction; formation of that product is the main barrier to the commercialization of low-temperature SCR systems.⁶ Therefore, investigating the key factor in the NH₄HSO₄ decomposition and reactivity behaviors on the catalysts constitutes the vital step in industrializing low-temperature SCR systems.

Our previous study⁷ has investigated the decomposition and reactivity of the main poisoner, NH₄HSO₄, on the V₂O₅/TiO₂ catalysts and claimed that V₂O₅/TiO₂ catalysts promote the decomposition and reactivity behaviors of NH₄HSO₄ with NO. And crystallite ammonium sulfate salts are more stubborn than amorphous salts. Therefore, it seems that the best way to protect catalysts from being deactivated by the deposition of NH₄HSO₄ is to promote the NH₄HSO₄ decomposition and reactivity behaviors to avoid the formation of crystallite ammonium sulfate salts on the catalysts, on which scanty research has been conducted, even though V₂O₅/TiO₂ catalyst system is typically used for stationary applications. By contrast, Liu et al.^8–11 found NH₄HSO₄ decomposes and reacts with NO more easily on the V₂O₅/activated carbon (V/AC) catalysts than on the V₂O₅/TiO₂ catalysts. Interaction between AC and V₂O₅ is considered to be the main reason for the enhanced NH₄HSO₄ decomposition and reactivity behaviors on the AC-based catalysts. Knowledge of such effects lays a solid foundation for developing novel catalysts with superior activity and sulfur resistance, even if variations in the physicochemical properties between TiO₂ and AC exist.

The typical promoter for V₂O₅/TiO₂ catalysts is the metal oxide WO₃, which is widely applied in industrial catalysts. For the behaviors of NH₄HSO₄ on the WO₃-promoted catalysts, Baltin¹² claimed that NH₄HSO₄ formed on the V₂O₅–WO₃/TiO₂ catalysts during SCR reactions can react with NO to produce H₂SO₄ at 170 °C. However, some crucial aspects of the NH₄HSO₄ behaviors on the WO₃-promoted catalysts still lack investigation. (a) It is clear that WO₃ addition would decrease the catalyst surface area due to partial plugging of the pore structures,¹³ which might cause the appearance of crystallite ammonium sulfate salts, thereby inhibiting the NH₄HSO₄ decomposition and reactivity behaviors to some extent. (b) It has been proved that SO₄²⁻ are only bonded to the TiO₂ surface instead of the V₂O₅ surface in the NH₄HSO₄-deposited V₂O₅/TiO₂ samples.⁷ For WO₃-promoted catalysts, SO₄²⁻ might be also linked to the WO₃ surface, because of the high W content in the V₂O₅–WO₃/TiO₂ catalysts (WO₃ content: ca. 5%). Thus, variations in the NH₄HSO₄ decomposition and reactivity behaviors would be expected to occur. (c) Provided that NH₄HSO₄ exist in the amorphous state on the WO₃-promoted catalysts and SO₄²⁻ are only bonded to TiO₂ sites rather than WO₃ sites, WO₃-doping might have a positive effect on the decomposition and reactivity of NH₄HSO₄ with NO, for WO₃-promoted catalysts exhibit an enhanced SO₂-resistant ability at lower temperatures.^14,15 In this study, attention was focused on the interaction between NH₄HSO₄ and WO₃-promoted catalysts, together with the effect of WO₃ on the decomposition and reactivity of NH₄HSO₄ with NO.

For the purposes above, NH₄HSO₄ was deposited on V₂O₅/TiO₂ and V₂O₅–WO₃/TiO₂ catalysts. As a reference, NH₄HSO₄-deposited TiO₂ and WO₃/TiO₂ samples were also adopted. X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), and X-ray photoelectron spectroscopy (XPS) were performed to explore the interaction between NH₄HSO₄ and catalysts. Temperature-programmed methods were applied to study the promotion effect of WO₃ on the NH₄HSO₄ decomposition and reactivity behaviors. Finally, catalyst sulfur resistance tests were conducted to illustrate the improved NH₄HSO₄ decomposition and reactivity behaviors could protect catalysts from being deactivated by the excess deposition of NH₄HSO₄, which leads to an enhanced catalyst SO₂-resistant ability.

2. Experimental

2.1 Chemicals

P25 (85%: anatase, 15%: rutile; S_BET = ca. 55 m² g⁻¹) was purchased from Degussa. NH₄VO₃ (AR: 99.0%), H₄₀N₁₀O₄₁W₁₂·xH₂O (AR, WO₃ content: 85–90%) and NH₄HSO₄ (AR, H₂SO₄ content: 41.5–43.5%) were available from Sinopharm Chemical Reagent Co., Ltd. N₂, O₂ and NO (0.5% NO + 99.5% N₂) were obtained from Hangzhou New Century Mixed Gas Co., Ltd.

2.2 Preparation of the samples

According to the formula of commercial SCR catalysts (V₂O₅ content: 1 wt%, WO₃ content: 5, 6.5, 8 wt%), V₂O₅–WO₃/TiO₂ (denoted as VW/Ti) catalyst in this study was prepared through the wet impregnation method using the mixture of ammonium metavanadate–oxalic and ammonium tungsate–oxalic solutions. As for the preparation of TiO₂, WO₃/TiO₂ and V₂O₅/TiO₂ (denoted as Ti, W/Ti, V/Ti) catalysts, the same procedures were adopted, except that the mixture of ammonium metavanadate–oxalic and ammonium tungstate–oxalic solutions was replaced by deionized water, ammonium tungsate–oxalic solution and metavanadate–oxalic solution, respectively. All of the prepared samples were dried at 110 °C overnight and calcined at 500 °C for 5 h. The as-prepared catalyst was crushed and sieved into 100 mesh for NH₄HSO₄ deposition.

The NH₄HSO₄-deposited samples were prepared using a previously reported wet impregnation method.⁸ The samples were named according to the Ti-based catalysts. For example, ABS-VW/Ti means that NH₄HSO₄ was deposited on the VW/Ti (WO₃ content: 5 wt%) catalyst surface; ABS-VW8/Ti is related to the sample containing 8 wt% WO₃ and 10 wt% NH₄HSO₄, respectively. All of the samples were dried at 110 °C in air overnight. The as-prepared samples were crushed and sieved to 40–60 mesh for temperature-programmed decomposition (TPDC) and temperature-programmed surface reaction (TPSR).

2.3 Reaction systems

Generally, 0.3 g catalyst was used in the sulfur resistance tests at 250 °C. The reaction conditions were as follows: 1000 ppm NO, 1000 ppm NH₃, 5% O₂, 1000 ppm SO₂, 10% H₂O, and N₂ as balance gas. The total gas flow rate was 1.2 L min⁻¹, corresponding to a gas hourly space velocity (GHSV) of 240 000 mL g⁻¹ h⁻¹. The NO and NO_x concentrations were monitored online by a gas analyzer (Testo 350). The NO_x conversion was calculated according to the following equation:

NO_x conversion (%) = ([NO_x]_out/[NO_x]_in) × 100% (1)

The relative activity X/X₀, where X is the NO_x conversion at a certain time in the deactivation process and X₀ is the initial NO_x conversion for fresh catalysts, was calculated to diagnose the degree of catalyst deactivation resulting from the introduction of SO₂ and H₂O.

TPDC method was conducted in the reaction systems. Briefly, an NH₄HSO₄-deposited sample of 0.3 g was preheated to 100 °C in N₂ for 30 min in order to remove the physically adsorbed water and other impurities. The sample was then heated to 450 °C at a heating rate of 5 °C min⁻¹ in N₂. The outlet SO₂ signal, which was used to identify the thermal stability of sulfate species, was monitored using an FTIR flue gas spectrometer (Gasmet, gx4000).

The NH₄HSO₄ reactivity behavior on the Ti-based catalysts was measured via TPSR with NO. First, 0.4 g or 0.3 g NH₄HSO₄-deposited catalyst (40–60 mesh) was exposed to a stream containing 1000 ppm NO, 3% O₂ and balance N₂ at a flow rate of 1.2 L min⁻¹ or 0.5 L min⁻¹, corresponding to a GHSV of 180 000 mL g⁻¹ h⁻¹ or 100 000 mL g⁻¹ h⁻¹, respectively. Then the temperature was ramped from 100 °C to 450 °C at a heating rate of 5 °C min⁻¹. The inlet and outlet concentrations of NO and N₂O were monitored using an FTIR spectrometer (Gasmet, gx4000).

The reaction rate constant k was calculated according to previous study of Baltin.¹² As the change of the concentration of the second reaction partner, NH₃, during a single measuring operation (about 5 seconds) is differentially small for every experimental value, the pseudo-first order reaction rate constant k can be calculated using the following equation:

where AV is attributed to the area velocity.

2.4 Characterizations of the samples

The XRD patterns were recorded using an X-ray diffractometer (RIGAKU D/MAX 2550, Japan), which was equipped with Cu Kα radiation at 40 kV and 100 mA, at a step size of 0.0167° in the 2θ range of 10–80°. N and S elemental analysis was measured on an elemental analyzer (EA1112, Italy).

XPS spectra were obtained with a Thermo ESCALAB 250 electron spectrometer, which was equipped with an Al Kα X-ray source and a hemispherical electron analyzer. The binding energy of C 1s was calibrated to 284.5 eV.

FTIR was conducted using a Nicolet 6700 spectrometer. The sample-KBr mixtures in a mass ratio of 1 : 50 were ground. The spectrum was collected 32 times, followed by the automatic subtraction of the background line obtained. The spectra were recorded from 4000 to 400 cm⁻¹.

Thermogravimetric analysis (TGA) was conducted in the system of TG (TG-Q500 TGA). Approximately 6 mg of the sample was pretreated at 100 °C in a flow of N₂ for 30 min to remove physically adsorbed water and other impurities. The sample was then heated to 650 °C at a heating rate of 10 °C min⁻¹ in flowing N₂.

In situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) experiments of the NH₄HSO₄ decomposition and reactivity behaviors were performed using an FTIR spectrometer (Nicolet Nexus 6700), which was equipped with an MCT/A detector. The background spectra were obtained in N₂ at 100 °C, 150 °C, 250 °C, 350 °C and 450 °C, respectively. Then the tested sample was heated to certain temperatures in N₂ or N₂ + NO (1000 ppm) + O₂ (5%) to investigate the dynamic changes in the catalyst surface functional groups.

3. Results and discussions

3.1 Catalyst surface functional groups

As is shown in Table 1, the deposition amount of NH₄HSO₄ on each catalyst is almost the same because of the existence of errors during the catalyst preparation process and elemental analysis experiments. The XRD patterns of the samples illustrate that NH₄HSO₄, V₂O₅ and WO₃ are highly dispersed and exist in the amorphous state on the catalyst surface (Fig. S1†). In the case of the sulfate-free samples, no obvious characteristic peaks can be detected in the range between 1000 and 2000 cm⁻¹, except for the band at 1631 cm⁻¹ attributed to the vibrations of H₂O on the catalyst surface (Fig. S2†). This observation suggests that the bands assigned to the S=O and S–O vibrations can only be detected in the NH₄HSO₄-containing samples. Given the deposition of NH₄HSO₄ on the TiO₂ catalyst surface, the characteristic peaks of bidentate sulfate anions at 1224, 1139, 1054 cm⁻¹, instead of the band attributed to free HSO₄⁻ at 1174 cm⁻¹,^16,17 are generated (Fig. 1).^18,19 Based on previous studies, the characteristic peaks at 1224, 1139, and 1054 cm⁻¹ can be assigned to the v₃ vibrations of bidentate SO₄²⁻ in C_2v symmetry.²⁰ In addition, this form of S-containing functional groups is stable and continues to exist after the calcination process of sulfated TiO₂ at 500 °C.²¹ The bands at 1224, 1139 cm⁻¹ are related to the asymmetrical and symmetrical stretching of the S=O vibrations. At 1054 cm⁻¹, this band is assigned to the asymmetrical stretching of the S–O vibrations.^19,22,23 These results suggest that chemical interaction occurs when NH₄HSO₄ is deposited on the Ti catalyst surface, with the transformation of sulfate species from HSO₄⁻ to bidentate SO₄²⁻, which are subsequently bonded to TiO₂ sites.

Table 1 Elemental analysis of series NH₄HSO₄-deposited samples

Samples	S content (wt%)	N content (wt%)
ABS-Ti	2.30	0.82
ABS-W/Ti	2.31	0.82
ABS-V/Ti	2.39	0.86
ABS-VW/Ti	2.41	0.86
ABS-VW6.5/Ti	2.36	0.83
ABS-VW8/Ti	2.37	0.84

---

Fig. 1 FTIR spectra: NH₄HSO₄-deposited catalysts and pure NH₄HSO₄.

Given the addition of WO₃, characteristic peaks assigned to NH₄⁺, adsorbed H₂O and bidentate SO₄²⁻, are still presented, demonstrating the occurrence of the interaction between NH₄HSO₄ and TiO₂ support in the W-doped samples. At the same time, no extra peaks could be detected; the absence of vanadyl sulfate or tungstate sulfate compound indicates that SO₄²⁻ are only bonded to the TiO₂ surface rather than the V₂O₅ or WO₃ surfaces.²⁴ Although no V–SO₄ or W–SO₄ species are detected from the IR spectra obtained, chemical interactions between V species (or W species) and NH₄HSO₄ still need to be explored.

Considering the existence of chemical interaction between NH₄HSO₄ and catalysts, it is related to the electron deviation between catalyst atoms and ammonium ions or sulfate anions, which is still unknown from the XRD patterns and IR spectra obtained. Therefore, the XPS method was conducted to extensively explore this chemical interaction.

3.2 Catalyst atom environment

The XPS method was conducted to investigate the variations in the catalyst atom environment before or after the introduction of NH₄HSO₄. For the V/Ti sample, as shown in Fig. 2(b), the binding energies of Ti 2p_3/2 (458.35 eV) and Ti 2p_1/2 (464.05 eV) are typical characteristics of Ti⁴⁺ ions.²⁵ Given the deposition of NH₄HSO₄ on the catalyst surface, these Ti 2p photoelectron bands shift towards a high binding energy (increased by about 0.28 eV), indicating that Ti atoms exist in an electron-deficit state in the NH₄HSO₄-loaded samples.²⁴ While in the VW/Ti sample, greater increase in the binding energy values corresponding to the Ti 2p photoelectron peaks occurs with the introduction of NH₄HSO₄ (increased by about 0.48 eV). And combined with the binding energy values attributed to the Ti 2p photoelectron peaks of the NH₄HSO₄-deposited samples (Fig. 2), it seems that Ti species in the W-containing samples show more severe deviation of electron cloud by interacting with NH₄HSO₄.²⁶

Fig. 2 XPS spectra of Ti 2p: (a) V-free samples; (b) V-containing samples.

For V atoms, the peak ranging between 516.40 eV and 517.00 eV can be indexed to V⁵⁺. V⁴⁺ and V³⁺ are in the range of 515.70–516.40 eV and 515.00–515.70 eV, respectively.²⁷ As shown in Fig. 3(a), the deconvolution of V 2p photoelectron peaks indicates the co-existence of V⁵⁺, V⁴⁺ and V³⁺ in the V/Ti and VW/Ti samples. After the introduction of NH₄HSO₄, a dramatic increase in the ratio of V⁵⁺/(V⁴⁺ + V³⁺ + V⁵⁺) can be seen, indicating the increase in the valence state of V atoms, as shown in Table 2.

Fig. 3 XPS spectra: (a) V 2p; (b) W 4f; (c) O 1s: V-free sample; (d) O 1s: V-containing samples; (e) N 1s; (f) S 2p.

Table 2 XPS data of series Ti-based samples

Samples	Ti atom %	O atom %	V atom %	W atom %	N atom %	S atom %	V^5+ %	Oα/(Oα + Oβ) %
Ti	32.12	67.88	—	—	—	—	—	20.44
ABS-Ti	23.79	66.91	—	—	4.51	4.79	—	41.88
V/Ti	27.96	70.86	1.18	—	—	—	44.58	27.00
ABS-V/Ti	22.9	66.25	0.39	—	5.11	5.34	59.27	43.17
W/Ti	28.44	67.88	—	3.68	—	—	—	25.44
ABS-W/Ti	20.09	69.17	—	2.59	4.11	4.04	—	51.13
VW/Ti	26.69	68.85	0.71	3.74	—	—	54.24	26.19
ABS-VW/Ti	20.16	68.1	0.31	2.53	4.42	4.48	70.42	47.56

---

For W atoms, the peaks in the range of 36.36–36.74 eV and 37.41–37.68 eV can be assigned to the spin–orbit splitting components of the W 4f named W 4f_7/2 and 4f_5/2 respectively, as shown in Fig. 3(b).²⁸ This result illustrates the existence of W atoms in the +6 oxidation state. Given the introduction of NH₄HSO₄, these W 4f photoelectron peaks shift toward higher binding energies, indicating that the electron cloud density around the W atoms is lower than that in the NH₄HSO₄-free samples.

According to Fig. 3(c) and (d), the O1s spectra can be deconvoluted into two peaks, which are related to various oxygen-containing chemical bonds. Based on previous studies, the first peak centered in the range of 529.44–530.06 eV is assigned to lattice oxygen O²⁻ (O_β), and the second peak (531.00–531.62 eV) may be attributed to chemical adsorbed oxygen O_α, including the defect-oxide or hydroxyl-like group.²⁹ The concentration ratio of O_α/(O_α + O_β) is quite low for the sulfate-free samples; this ratio increases with the deposition of NH₄HSO₄ (as shown in Table 1). Saur et al.³⁰ claimed that the structure of sulfate species is presented as (M–O)₃S=O under dry conditions. By contrast, the sulfate species are bridged bidentate with H, forming a Bronsted site, (M₂SO₄)H, under wet conditions. Thus, Bronsted sites increase and more hydroxyls come out with the deposition of NH₄HSO₄, which leads to a relatively high ratio of O_α/(O_α + O_β). Besides, the binding energies assigned to the O 1s photoelectron peaks increase with the deposition of NH₄HSO₄, revealing a decreased electron cloud density around the O atoms.³¹

As is previously mentioned, NH₄HSO₄ introduction leads to a decline in the electron cloud density around the catalyst atoms. Considering the charge balance, it can be concluded that electrons around the catalyst atoms in the NH₄HSO₄-loaded samples might deviate toward sulfate species or ammonium species. As shown in Fig. 3(e), the N 1s spectra demonstrate that N species exist as NH₄⁺.³² No obvious variations in the binding energies corresponding to the N 1s photoelectron peaks could be detected with the addition of WO₃ (401.50 eV in V-free sample, 401.62 eV in V-loaded sample). Based on Fig. 3(f), after the curve fitting of S 2p spectra, the S doublet can be attributed to the S 2p_3/2 (between 169.82 and 169.52 eV) and S 2p_1/2 (ranging from 168.32–168.62 eV) transitions of SO₄²⁻.³³ Compared with W-free samples, a higher electron cloud density around the S atoms in the W-added samples can be detected, as is illustrated by the decrease in the binding energies of the S 2p photoelectron peaks. Considering the alkalinity of NH₃, it seems that SO₄²⁻ possess higher electronegativity in comparison to NH₄⁺, although no direct evidence about the exact electronegativity values of SO₄²⁻ and NH₄⁺ is available. Generally, S=O covalent bond is responsible for the generation of acidic sites on the sulfate-containing catalysts and has a strong ability to attract electrons from basic molecules.³⁴ Consequently, electrons around the catalyst atoms in the NH₄HSO₄-deposited samples deviate to sulfate species resulting from the stronger electron affinity by S⁶⁺ in S=O structures, thereby leaving S atoms exist in an electron-enrichment state; and WO₃ introduction leads to an increase in the electron cloud density around the S atoms with +6 formal oxidation number as in SO₄²⁻.

3.3 The promotion effect of WO₃ on the NH₄HSO₄ decomposition behavior

According to previous studies, the decomposition behavior of NH₄HSO₄ was closely related to the release of SO₂ gas.³⁵ That is to say, the NH₄HSO₄ decomposition behavior on the catalyst surface during the heat process is attributed to the reduction of S atoms in the +6 oxidation state as in SO₄²⁻ to those in the +4 oxidation state as in SO₂, which is promoted in a reduction atmosphere.³⁶ Based on the XPS results, the addition of WO₃ increases the electron cloud density around the S atoms in SO₄²⁻; it seems that SO₄²⁻ on the WO₃-promoted catalysts would be easily reduced to SO₂ during the heat process in comparison to those on the WO₃-free catalysts.

In order to prove this assumption, TPDC method was conducted to explore the behaviors of sulfate species during the heat process.⁸ In the case of the ABS-Ti sample, SO₂ begins to release at a temperature of ca. 250 °C, as is shown in Fig. 4. With increasing temperature, the outlet SO₂ concentration almost keeps constant, which starts to increase at a temperature higher than 400 °C. As for the SO₄²⁻ consumption behavior on the W/Ti catalyst surface, an obvious increase in the outlet SO₂ concentration can be detected, indicating that more S-containing functional groups are consumed during the heat process. Similar phenomenon could be observed in the V-containing samples that the addition of WO₃ promotes the SO₄²⁻ consumption behavior in the lower temperature region, which is consistent with the assumptions drawn from the XPS results.

Fig. 4 SO₂ profiles of the NH₄HSO₄-deposited samples.

Fig. 5 illustrates the NH₄HSO₄ decomposition behavior on the W-added or W-free catalysts; all the samples exhibit two weight loss peaks during the heat process. Compared with ABS-Ti sample (at ca. 368 and 467 °C, respectively), temperatures (at ca. 337 °C and 454 °C, respectively) corresponding to the weight loss peaks decrease for the ABS-W/Ti sample. Similar trend could be detected in the V-containing samples that the addition of WO₃ exerts a positive effect on the NH₄HSO₄ decomposition behavior in the lower temperature region, as the decrease in the temperatures assigned to the weight loss peaks illustrates. In comparison with W-free samples, the introduced WO₃ lead to a high decomposition rate of NH₄HSO₄ at low-temperature regions, as the increased derived weight value assigned to the first weight loss peak reveals. That is to say, the addition of WO₃ adversely affects the stability of the deposited NH₄HSO₄, which might be an important reason for the enhanced sulfur resistance of WO₃-promoted catalysts at lower temperatures to some extent.^1,14 Fig. 6 illustrates the effect of high WO₃ contents on the decomposition behavior of the deposited NH₄HSO₄. The increase in the WO₃ content leads to an enhancement in the NH₄HSO₄ decomposition behavior in the lower temperature region, as the obvious decrease in the temperatures attributed to the weight loss peaks suggests.

Fig. 5 DTG profiles of the NH₄HSO₄-deposited samples.

Fig. 6 DTG profiles of the ABS-VW/Ti samples containing 5, 6.5 and 8 wt% WO₃.

In situ DRIFTS and elemental analysis were conducted to investigate the detailed NH₄HSO₄ decomposition behavior over the VW/Ti catalyst surface. As shown in Fig. 7, characteristic IR peaks at 1244 and 1434 cm⁻¹, which can be assigned to the asymmetrical stretching vibrations of S=O in bidentate SO₄²⁻ and NH₄⁺ bonded to Bronsted acid sites, appear when the temperature is 100 °C.⁷ Meanwhile, obvious N–H stretching vibration bands from ionic NH₄⁺ (3250, 3024 and 2857 cm⁻¹) also come out.³¹ The surface bidentate sulfate species undergo a blue shift (from 1244 to 1286 cm⁻¹) with the increase in the temperature.³⁷ It should be noted that there is a drastic decrease in the intensity of the peaks assigned to NH₄⁺ at 1434 cm⁻¹ and sulfate species at 1286 cm⁻¹ at the temperature of 450 °C, which is attributed to the decomposition behavior of the deposited NH₄HSO₄ during the heat process. Additionally, characteristic IR peak at 1375 cm⁻¹, the asymmetrical stretching vibrations of O=S=O species in tridentate SO₄²⁻, starts to appear.³¹ That might be attributed to the transformation of sulfate species from (M₂SO₄)H structure to (M–O)₃S=O one.³⁸ Elemental analysis was conducted to identify the S and N contents of the ABS-VW/Ti samples after the calcination process at various temperatures (250, 350 and 450 °C), which facilitated estimation of the behaviors of ammonium and sulfate species during the heat process. These temperatures are selected based on the results of TG-DTG. As is shown in Table 3, the sample remains almost unchanged at a calcination temperature lower than 250 °C. The N content starts to decrease at a calcination temperature higher than 350 °C, revealing the decomposition of ammonium ions. Meanwhile, a slight decrease in the S content is observed at a calcination temperature of 350 °C. Sulfate species begin to be greatly consumed when the calcination temperature is increased further. This finding confirms that the first weight loss peak at ca. 337 °C is mainly attributed to the decomposition of NH₄⁺ along with a small part of SO₄²⁻, while the latter one might be attributed to the simultaneous consumption of NH₄⁺ and SO₄²⁻.

Fig. 7 In situ DRIFTS study of the NH₄HSO₄ decomposition behavior over VW/Ti catalyst surface.

Table 3 Elemental analysis of ABS-VW/Ti sample under different calcination temperatures

Samples	S content (wt%)	N content (wt%)
Uncalcined	2.41	0.86
250 °C	2.39	0.82
350 °C	2.25	0.46
450 °C	1.24	0

---

Since it has been proved that ammonium species are bonded to sulfate sites according to the FTIR spectra, NH₄⁺ accommodated on sulfate sites would be gradually consumed with increasing temperature, as illustrated by the decreased N content at a calcination temperature higher than 250 °C.⁷ Provided that a decrease in the amount of sulfate sites occurs during the heat process, a decline in the surface ammonium species would be expected to occur. In the case of W-added samples, the addition of WO₃ increases the electron cloud density around the S atoms in SO₄²⁻, which is beneficial for the reduction of S atoms in the +6 oxidation state in SO₄²⁻ to those in the +4 oxidation state in SO₂. As a result, the addition of WO₃ has a promotion effect on the SO₄²⁻ consumption behavior at lower temperatures, as is illustrated by the increased SO₂ release amount in the lower temperature region for the WO₃-promoted samples during the heat process. The decreased surface sulfate sites for the accommodation of NH₄⁺ promotes the NH₄⁺ consumption behavior at low-temperature regions. Consequently, the first weight loss peak mainly attributed to the decomposition of NH₄⁺ would shift to lower temperatures. As previously mentioned, electron deviation from catalyst atoms to sulfate anions would be the key factor in the NH₄HSO₄ decomposition behavior on the catalysts and WO₃ addition increases the electron cloud density around the S atoms with +6 formal oxidation number as in SO₄²⁻. Therefore, the decomposition behavior of NH₄HSO₄ would be promoted on the WO₃-added catalysts.

3.4 The promotion effect of WO₃ on the NH₄HSO₄ reactivity behavior with NO

Based on our previous studies, NH₄⁺ belonging to NH₄HSO₄ acts as a reductant and reacts with NO to produce N₂ and H₂O.⁷ Similar with SCR mechanism, it seems that NH₄⁺ activation process might be the key factor in the reactivity behavior of NH₄HSO₄ with NO, which catalyst redox property determines. It has been proved that the addition of WO₃, the typical promoter in industrial catalysts, would enhance the low-temperature SCR activity through tuning the catalyst acid and redox properties,³⁹ which might lead to the variations in the NH₄HSO₄ reactivity behavior. In this section, TPSR and in situ DRIFTS experiments would be performed to prove these assumptions.

The NO TPSR profiles with NH₄HSO₄ indicate that the addition of WO₃ leads to a slight enhancement in the reactivity behavior of NH₄HSO₄ with NO, as the higher reaction rate constant below 330 °C and a decline in the temperature assigned to the maximum reaction rate constant reveal (Fig. S3(a)†). Meanwhile, an obvious decrease in the outlet concentration of N₂O (the unwanted byproduct) is observed with the addition of WO₃ (Fig. S3(c)†). Combined with the fairly low outlet concentration of NH₃ (less than 1 ppm, Fig. S3(b)†), it can be concluded that the introduced WO₃ exert a promotion effect on the catalytic selectivity to N₂ during the TPSR process. In addition, the reaction of NH₄HSO₄ with NO is an important reason for the consumption of the surface ammonium species during the TPSR process (Fig. S3(d)†). Fig. 8 illustrates the effect of high WO₃ contents on the reactivity of NH₄HSO₄ with NO. Given increases in WO₃ content, a slight decline in the temperature attributed to the maximum reaction rate constant can be detected. This reaction suggests a little higher reactivity of NH₄HSO₄ with NO on the high W content catalyst surfaces.

Fig. 8 TPSR profiles of NH₄HSO₄ with NO on the VW/Ti catalysts containing 5, 6.5, 8 wt% WO₃; condition: 1000 ppm NO + 3% O₂ in N₂ at 0.5 L min⁻¹; heating rate: 5 °C min⁻¹; 0.3 g samples.

As for the detailed reactivity behavior of NH₄HSO₄ with NO at a constant temperature over the VW/Ti catalyst surface, in situ DRIFTS experiments were conducted. It seems that the reaction rate of NH₄HSO₄ with NO at 250 °C over the VW/Ti catalyst is quite low, because NH₄HSO₄ remains almost unchanged after the exposure in an NO-containing flue gas in those first three hours, which begins to be obviously consumed after 200 min (Fig. S4†). Combined with NO TPSR profiles, it seems that the low reactivity of NH₄HSO₄ with NO at low-temperature regions might be an important reason for the serious catalyst deactivation in the SO₂- and H₂O-containing flue gas below 300 °C to some extent. In addition, no characteristic peaks assigned to nitrate and nitrite species are detected, indicating the reaction pathway between NH₄⁺ species and gaseous NO dominates in the reactivity behavior of NH₄HSO₄ with NO over the VW/Ti catalyst surface. The dehydrogenation of NH₄⁺ species by adjacent redox sites may be the rate determining step, as the subsequent reaction with NO to produce N₂ and H₂O follows.¹

3.5 The promotion effect of WO₃ on the catalyst sulfur resistance

The effects of SO₂ and H₂O addition on the SCR of NO with NH₃ over the V/Ti, VW5/Ti, VW6.5/Ti and VW8/Ti catalysts were detected as a function of time on stream. As is shown in Fig. 9, in the case of the V/Ti catalyst, an obvious decrease in the SCR activity occurs upon the addition of SO₂ and H₂O. The ratio of X/X₀ reaches 0.8 after 9.5 h. In the case of the VW5/Ti catalyst, the duration required for the relative activity X/X₀ to reach 0.8 is 12.5 h. Further increasing the WO₃ addition amount, a significant enhancement in the catalyst sulfur resistance occurs; for the VW6.5/Ti and VW8/Ti catalysts, the X/X₀ ratio reaches 0.8 after 14.5 and 19.5 h, respectively, which are longer than V/Ti and VW5/Ti catalysts. It seems that catalyst sulfur resistance is greatly enhanced with the addition of WO₃ at lower temperatures.

Fig. 9 Relative activity in the presence of SO₂ and H₂O for the SCR of NO by NH₃ over V/Ti, VW5/Ti, VW5/Ti, VW8/Ti at 250 °C. Conditions: 1000 ppm + 1000 NH₃ + 5% O₂ + 1000 ppm SO₂ + 10% H₂O in N₂ at 1.2 L min⁻¹; 0.3 g catalyst.

According to the V₂O₅/TiO₂-based catalyst sulfur poisoning mechanism, it is clearly confirmed that catalyst sulfur resistance is mainly related to the rate difference between the NH₄HSO₄ formation and consumption behaviors on the catalyst surface.^6,10 Considering the inevitable NH₄HSO₄ formation behavior in the SO₂- and H₂O-containing flue gas at lower temperatures,⁴⁰ it seems that catalyst sulfur resistance is closely attributed to the decomposition and reactivity behaviors of NH₄HSO₄ with NO. Combined with thermogravimetric analysis and TPSR profiles, WO₃ addition leads to an enhancement in the NH₄HSO₄ decomposition and reactivity behaviors, which could protect SCR catalysts from being deactivated by the excess deposition of NH₄HSO₄. Therefore, WO₃-promoted V/Ti catalysts exhibit an enhanced SO₂-resistant ability in the lower temperature region.

For the design of catalysts possessing high catalytic activity and sulfur resistance, the addition of promotors may be an effective method. As is mentioned above, electron deviation from catalyst atoms to sulfate anions might play an important role in the NH₄HSO₄ decomposition behavior; NH₄⁺ activation process might be the key factor in the NH₄HSO₄ reactivity behavior with NO. Based on the DFT calculations, Sb₂O₅, as a typical promoter in SCR catalysts, has the higher HOMO energy than WO₃, MoO₃ and Nb₂O₅, indicating the easier electron donation of Sb₂O₅.⁴¹ Considering the strong electronegativity of SO₄²⁻, it seems that the NH₄HSO₄ decomposition behavior would be improved with the addition of Sb₂O₅, which has not been studied. And in comparison to WO₃, MoO₃ and Nb₂O₅, Sb₂O₅ has the lower LUMO energy, indicating the higher oxidation ability of Sb₂O₅. A high reactivity of NH₄HSO₄ with NO on the Sb₂O₅-doped catalysts would be expected to occur. In the next study, extensive investigation would be conducted on the promotion effect of Sb₂O₅ on the decomposition and reactivity of NH₄HSO₄ with NO.

4. Conclusion

Catalyst crystal phases and surface functional groups remain unchanged regardless of WO₃ addition. Compared with W-free samples, the introduction of WO₃ leads to an increased electron cloud density around the S atoms in SO₄²⁻, which might have a positive effect on the reduction of S atoms in the +6 oxidation state as in NH₄HSO₄ to those in the +4 oxidation state as in SO₂. Consequently, WO₃-doping increases the SO₂ release amount at lower temperatures during the heat process; the decomposition behavior of NH₄HSO₄ is promoted by the introduction of WO₃, as the decrease in the temperatures for the weight loss peaks illustrates. Meanwhile, the addition of WO₃ has a mild promotion effect on the reactivity of NH₄HSO₄ with NO together with an inhibitory effect on the production of N₂O during the reaction process. Finally, WO₃-promoted catalysts exhibit an enhanced SO₂-resistant ability.

Acknowledgements

This work is supported by the National Science Foundation for Distinguish Yong Scholars of China (No. 51125025) and the National High-Tech Research and Development (863) Program of China (No. 2013AA065401).

References

1. G. Busca, L. Lietti, G. Ramis and F. Berti, Appl. Catal., B, 1998, 18, 1–36.
2. F. Liu, W. Shan, Z. Lian, L. Xie, W. Yang and H. He, Catal. Sci. Technol., 2013, 3, 2699.
3. R. Qu, X. Gao, K. Cen and J. Li, Appl. Catal., B, 2013, 142–143, 290–297.
4. F. Liu, H. He, Z. Lian, W. Shan, L. Xie, K. Asakura, W. Yang and H. Deng, J. Catal., 2013, 307, 340–351.
5. M. Kobayashi, R. Kuma and A. Morita, Catal. Lett., 2006, 112, 37–44.
6. Y. Xi, N. A. Ottinger and Z. G. Liu, Appl. Catal., B, 2014, 160–161, 1–9.
7. D. Ye, R. Qu, H. Song, X. Gao, Z. Luo, M. Ni and K. Cen, Chem. Eng. J., 2016, 283, 846–854.
8. Z. Zhu, H. Niu, Z. Liu and S. Liu, J. Catal., 2000, 195, 268–278.
9. Z. Zhu, Z. Liu, H. Niu, S. Liu, T. Hu, T. Liu and Y. Xie, J. Catal., 2001, 197, 6–16.
10. Z. Huang, J. Catal., 2003, 214, 213–219.
11. P. Li, Q. Liu and Z. Liu, Chem. Eng. J., 2012, 181–182, 169–173.
12. G. Baltin, H. Köser and K.-P. Wendlandt, Catal. Today, 2002, 75, 339–345.
13. D. W. Kwon and S. Chang Hong, RSC Adv., 2016, 6, 1169–1181.
14. D. W. Kwon, K. H. Park and S. C. Hong, Chem. Eng. J., 2016, 284, 315–324.
15. D. W. Kwon and S. C. Hong, Appl. Surf. Sci., 2015, 356, 181–190.
16. A. Goypiron, J. De Villepin and A. Novak, J. Raman Spectrosc., 1980, 9, 297–303.
17. M. Mamlouk, P. Ocon and K. Scott, J. Power Sources, 2014, 245, 915–926.
18. J. P. Chen and R. T. Yang, J. Catal., 1993, 139, 277–288.
19. J. L. Ropero-Vega, A. Aldana-Pérez, R. Gómez and M. E. Niño-Gómez, Appl. Catal., A, 2010, 379, 24–29.
20. J. R. Sohn and H. W. Kim, J. Mol. Catal., 1989, 52, 361–374.
21. S. T. Choo, Y. G. Lee, I.-S. Nam, S.-W. Ham and J.-B. Lee, Appl. Catal., A, 2000, 200, 177–188.
22. L. K. Noda, R. M. de Almeida, L. F. D. Probst and N. S. Gonçalves, J. Mol. Catal. A: Chem., 2005, 225, 39–46.
23. K. Nakamoto, Infrared and Raman spectra of inorganic and coordination compounds, Wiley Online Library, 1986.
24. X. Guo, C. Bartholomew, W. Hecker and L. L. Baxter, Appl. Catal., B, 2009, 92, 30–40.
25. X. Du, X. Gao, Y. Fu, F. Gao, Z. Luo and K. Cen, J. Colloid Interface Sci., 2012, 368, 406–412.
26. F. Liu, H. He, C. Zhang, Z. Feng, L. Zheng, Y. Xie and T. Hu, Appl. Catal., B, 2010, 96, 408–420.
27. J. Haber, Catal. Today, 2009, 142, 100–113.
28. R. Camposeco, S. Castillo, V. Mugica, I. Mejía-Centeno and J. Marín, Chem. Eng. J., 2014, 242, 313–320.
29. J.-C. Dupin, D. Gonbeau, P. Vinatier and A. Levasseur, Phys. Chem. Chem. Phys., 2000, 2, 1319–1324.
30. O. Saur, M. Bensitel, A. B. M. Saad, J. C. Lavalley, C. P. Tripp and B. A. Morrow, J. Catal., 1986, 99, 104–110.
31. F. Liu, K. Asakura, H. He, W. Shan, X. Shi and C. Zhang, Appl. Catal., B, 2011, 103, 369–377.
32. W. Xu, H. He and Y. Yu, J. Phys. Chem. C, 2009, 113, 4426–4432.
33. S. Liu and X. Chen, J. Hazard. Mater., 2008, 152, 48–55.
34. T. Yamaguchi, T. Jin and K. Tanabe, J. Phys. Chem. C, 1986, 90, 3148–3152.
35. J. Li and G.-E. Zhang, Acta Phys.-Chim. Sin., 1992, 8, 123–127.
36. H. Tian, Q. Guo, X. Yue and Y. Liu, Fuel Process. Technol., 2010, 91, 1640–1649.
37. B. Q. Jiang, Z. B. Wu, Y. Liu, S. C. Lee and W. K. Ho, J. Phys. Chem. C, 2010, 114, 4961–4965.
38. A. Kayo, T. Yamaguchi and K. Tanabe, J. Catal., 1983, 83, 99–106.
39. C. Wang, S. Yang, H. Chang, Y. Peng and J. Li, Chem. Eng. J., 2013, 225, 520–527.
40. S. Matsuda, T. Kamo, A. Kato, F. Nakajima, T. Kumura and H. Kuroda, Ind. Eng. Chem. Prod. Res. Dev., 1982, 21, 48–52.
41. X. Du, X. Gao, W. Hu, J. Yu, Z. Luo and K. Cen, J. Phys. Chem. C, 2014, 118, 13617–13622.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra09072a

---