Nitrogen oxides (NO_x) emitted from thermal power plants and automobile exhaust gases are major air pollutants. The selective catalytic reduction (SCR) of NO_x with NH₃ is the most effective technology for NO_x abatement ^[4]. Currently, V₂O₅-WO₃/TiO₂ and V₂O₅-MoO₃/TiO₂ are the most widely used catalysts, in a narrow temperature window of 300-400 °C ^[5]. Most SCR systems are installed behind an economizer that allows the flue gas temperature to exceed 300 °C to maintain high catalytic activity. However, the catalyst life is short because it is easily affected by high concentrations of flue dust, SO₂, alkali metals, and other factors. In addition, blockage by (NH₄)₂SO₄ byproducts during high-temperature processes is a problem that needs to be addressed. The preferred technique is to locate the SCR NO_x-removing (DeNO_x) unit after the desulfurization equipment in coal-fired power plants ^[6]. In this process, the flue gas temperature can drop to 250°C below, at which V-based catalysts do not maintain high NO_x conversions ^[7].

Recent studies suggest that V catalysts have excellent sulfur resistance ^[8]. Various methods such as use of additives or different supports, e.g., Sb-V₂O₅/TiO₂ ^[9], V₂O₅-CeO₂/TiO₂ ^[10], V₂O₅-WO₃/Ti-Si ^[11], V₂O₅/CeO₂-ZrO₂ ^[12], and V₂O₅/AC ^[13] catalysts, have been used to further enhance the low- temperature activities of V catalysts at low temperatures and maintain their high sulfur resistance. The low-temperature activities of these catalysts were improved, but the sulfur resistance of V₂O₅ decreased.

Improvements in V catalyst activities at low temperatures and maintenance of the sulfur resistance of efficient additive-containing catalysts are particularly important. It has been reported that the low-temperature activity of a Cr₂O₃/TiO₂ catalyst was high because more acid sites were present on the surface ^[14]. Thirupathi et al. ^[15] obtained 90% NO_x conversion at 180 °C using CrO_x-doped MnO_x/TiO₂; catalyst characterizations showed that surface acidity (Brønsted and Lewis acid sites) had significantly increased. Tong et al. ^[16] reported that the conversion of NO reached 80.7% at 330°C for a Cr-Ce/TiO₂ catalyst, but dropped from 80.7% to 62.4% after 10h of sulfur resistance test. Chen et al. ^[17] prepared Cr-Mn catalysts using the citric acid method; the catalytic activity reached 98.5% at 120 °C but decreased to about 20% after 5h of sulfur resistance test. Chromium oxide has good low-temperature activity, but its sulfur resistance needs to be improved.

In this study, Cr-V-O/TiO₂ catalysts were prepared using Cr and V as the active components and TiO₂ as the carrier. The influence of the Cr/V molar ratio and loading was investigated, the catalysts were characterized, and the catalytic mechanism was investigated.

Certain amounts of ammonium metavanadate (AR) and chromium nitrate (AR) were dissolved in deionized water. The solution was adequately stirred in a water bath at 60 °C and mixed with TiO₂ powders by magnetic stirring for 20 min. Moisture was removed at 60 °C using a rotary vacuum evaporator. The resulting materials were dried at 110 °C overnight and calcined at 450 °C for 3 h. The samples were compressed into tablets and crushed to 20-30 mesh. CrO_x/TiO₂ and VO_x/TiO₂ catalysts were prepared using the same method. The Cr-V/TiO₂ catalysts with different Cr/V molar ratios were denoted by Cr_x-V_1−x/TiO₂ (x = 0.7, 0.5, 0.3, 0.2, 0.1) and Cr_0.2-V_0.8/TiO₂ catalysts with different weight percentage loadings were denoted by a wt% Cr_0.2-V_0.8/TiO₂ (a = 5, 10, 20, 50).

Powder X-ray diffraction (XRD) patterns were obtained using a SCINTAG XTRA diffractometer (Ni-filtered Cu K_α radiation) operated at 40 kV and 50 mA. Intensity data were recorded in the 2θ range from 10° to 80°, with a step size of 0.04°.

X-ray photoelectron spectroscopy (XPS) was performed using an ESCALab 220i-XL electron spectrometer (VG Scientific Company) with 300 W Al K_α radiation. The pressure was maintained at 3 × 10⁻⁹ mbar. The C 1s line at 284.6 eV from carbon was used as a reference, and the binding energies were calculated with respect to this line.

H₂ temperature-programmed reduction (H₂-TPR) and NH₃ temperature-programmed desorption (NH₃-TPD) were performed using a Micromeritics Auto Chem FINE SORB-3010E chemisorption analyzer. Prior to H₂-TPR analysis, the samples (200 mg) were treated with Ar (30 mL/min at 150 °C for 30 min, cooled to 80 °C, and flushed for 10 min). The atmosphere was changed to 5% H₂/Ar (30 mL/min), and the reactor temperature was increased to 750 °C. For NH₃-TPD, the samples (150 mg) were heated in a pure He flow at 400 °C for 1 h, and cooled to 100 °C in a He atmosphere. The samples were treated with anhydrous NH₃ for 1 h and then purged with He for 1 h. The temperature was increased linearly at a rate of 10 °C/min, and TPD was performed in pure He over the temperature range 100-900 °C.

The SCR activity was measured in a fixed-bed quartz reactor containing 0.5 g of catalyst mixed with 0.5 g of quartz sand, with a gas hourly space velocity (GHSV) of 60000 mL/(h·g). The simulated flue gas composition was as follows: j (NO) = j (NH₃) = 0.05%, j (O₂) = 5%, j (SO₂) = 0.01% (as needed), j (water) = 10 vol% (as needed), and balance N₂. The total flow rate was 500 mL/min. The concentrations of NO, NO₂, O₂, and SO₂ were monitored continuously using a TESTO350-XL gas analyzer. To ensure stability and accuracy of the data, the data were recorded after 20 min when each test condition was changed.

The effects of Cr doping on the bulk structures of the catalysts were investigated using XRD. Fig. 1 shows the XRD patterns of the catalysts. It shows that all the peaks in the patterns were attributable to the anatase phase of TiO₂. No diffraction peaks attributed to Cr and V species were observed, suggesting that the low-loaded active phase was highly dispersed on the TiO₂ support and had an amorphous or microcrystalline structure.

Fig. 1. XRD patterns of catalysts. (1) TiO2; (2) VOx/TiO2; (3) Cr0.2–V0.8/TiO2; (4) Crx/TiO2.

The effect of loading on the catalyst bulk was determined by examining Cr_0.2-V_0.8/TiO₂ catalysts with different loadings. Fig. 2 shows that the composite was highly dispersed on the surface of the TiO₂ support for active component loadings less than 20 wt%. The composite state phase (Cr₂V₄O₁₃) was observed on the Cr_0.2-V_0.8/TiO₂ catalyst surface when the loading was 20 wt%. When the loading was increased to 50 wt%, the diffraction peaks of the active component became more apparent.

Fig. 2. XRD patterns of Cr0.2–V0.8/TiO2 catalysts with different loadings. (1) 50 wt%; (2) 20 wt%; (3) 10 wt%; (4) 5 wt%.

The surface areas and types of acid sites, determined using NH₃-TPD, for catalysts with various Cr/V molar ratios are summarized in Table 1.

Table 1 Types of acid sites and surface areas for catalysts with various Cr/V molar ratios determined using NH3-TPD. Catalyst Acid site distribution a ABET (m2/g) Weak Medium Strong Total VOx/TiO2 0.10 0.26 — 0.36 56.8 Cr0.1–V0.9/TiO2 0.13 0.34 — 0.47 53.7 Cr0.2–V0.8/TiO2 0.12 0.39 — 0.51 56.2 Cr0.3–V0.7/TiO2 0.13 0.11 0.44 0.68 54.7 Cr0.5–V0.5/TiO2 0.14 0.26 0.52 0.92 52.4 Cr0.7–V0.3/TiO2 0.11 0.30 0.58 1.00 56.2 CrOx/TiO2 0.05 0.30 0.59 0.94 83.0 a The amount of acidity of Cr0.7–V0.3/TiO2 was assigned as 1.0 and compared with the other samples.

Table 1 Types of acid sites and surface areas for catalysts with various Cr/V molar ratios determined using NH3-TPD.

NH₃-TPD experiments were performed to determine the acid site distributions on the prepared catalysts. The NH₃-TPD curves are shown in Fig. 3. The peak centered at about 200°C is caused by NH₃ desorption from weak acid sites, and that centered at about 400 °C is attributed to desorption from medium and strong acid sites. The peak above 400°C is attributed to NH₃ desorption from strong acid sites. Fig. 3 shows that the desorption peak of the VO_x/TiO₂ catalyst in the range 150-350°C arises from NH₃ desorption from weak and medium acid sites distributed on the catalyst surface. Addition of Cr caused broadening of the desorption peak and the amount of chemisorbed NH₃ increased. These results suggest that the addition of Cr significantly enhanced the concentration and acidity of acid sites ^[18]. The desorption peak of the Cr_0.2-V_0.8/TiO₂ catalyst appeared in the range 160-300°C; this catalyst had the largest amounts of weak and medium acid sites on its surface. When the Cr/V molar ratio was greater than 0.2:0.8, the NH₃ desorption peak in the range 450-550°C was strong. This finding indicates the presence of a large number of strong acid sites on the surfaces of these catalysts; this is not conducive to desorption of NH₃ and is unfavorable for SCR activity. The optimum amount of Cr on the V-based catalyst surface would give the highest amount of weak acid sites and the lowest amount of strong acid sites.

Fig. 3. NH3-TPD curves for catalysts with various Cr/V molar ratios. (1) VOx/TiO2; (2) Cr0.1–V0.9/TiO2; (3) Cr0.2–V0.8/TiO2; (4) Cr0.3–V0.7/TiO2; (5) Cr0.5–V0.5/TiO2; (6) Cr0.7–V0.3/TiO2; (7) CrOx/TiO2.

H₂-TPR was used to investigate the catalyst reducibility; the results are shown in Fig. 4. The peak at 470 °C for the VO_x/TiO₂ catalyst can be assigned to the reduction of V⁵⁺ to V^{3+ [14]}. The peak observed above 625°C is ascribed to unsupported V₂O₅, and could arise from uneven distribution of V on the TiO₂ support. The reduction peak of the catalyst shifted to lower temperatures as the amount of chromium oxide increased. When the Cr/V molar ratio was 0.2:0.8, the reduction temperature of V⁵⁺ dropped to 460 °C, and no Cr reduction peak was observed. These results indicate good dispersion of Cr species on the catalyst surface. A combination of these results with those of the catalyst textural analysis suggests that interactions occurred between the phases (amorphous chromiumoxides and vanadium oxides). These interactions could promote the reduction of V⁵⁺, which promoted the reaction. When the Cr/V molar ratio was higher than 0.2:0.8, the reduction peak of Cr species was observed. The CrO_x TPR profile reported in the literature ^[19] showed two reduction peaks, at 280and 420°C; these could be attributed to reduction of Cr⁶⁺ to Cr⁵⁺ and of Cr⁵⁺ to Cr³⁺, respectively. The results show that the reduction peak of V weakened. Excess Cr therefore occupied the sites of V species, possibly preventing the SCR reaction.

Fig. 4. H2-TPR curves of catalyst with various Cr/V molar ratios. (1) VOx/TiO2; (2) Cr0.1–V0.9/TiO2; (3) Cr0.2–V0.8/TiO2; (4) Cr0.3–V0.7/TiO2; (5) Cr0.5–V0.5/TiO2; (6) Cr0.7–V0.3/TiO2; (7) CrOx/TiO2.

XPS was performed to investigate the surface binding energies and valence states of various catalyst species. The results are shown in Table 2 and Fig. 5. As shown in Fig. 5(a), the O 1s photoelectron spectra of the catalysts indicated the presence of two main types of oxygen, namely lattice oxygen (O_α), with a binding energy of 529.0-530.0eV, and surface oxygen (O_β), with a binding energy of 530.0-531.0eV ^[20]. The proportions of surface oxygen in the Cr-doped catalysts were higher than that in the undoped catalyst (Table 2). This finding indicates that the Cr-doped catalyst had a higher proportion of oxygen vacancies than the undoped catalyst. The presence of surface oxygen improves the catalytic activity in the SCR reaction. This effect is an important reason for the increased catalytic activities of the Cr-doped catalysts ^[21].

Table 2 Binding energies and surface atomic concentrations of O, Cr, and V for the prepared catalysts determined from deconvoluted XPS spectra. Catalyst Binding energy (eV) Atomic concentration (%) V4+/V5+ Cr3+/Cr6+ Oβ/Oα O 1s V 2p Cr 2p O Ti (Cr)* V 10 wt% VOx/TiO2 530.02, 531.30 516.88, 517.57 — 68.33 25.70 5.97 0.59 — 0.61 5 wt% Cr0.2–V0.8/TiO2 529.98, 531.17 516.61, 517.26 577.42, 580.05 70.75 26.86 2.39 0.21 0.80 0.92 10 wt% Cr0.2–V0.8/TiO2 529.80, 530.63 516.88, 517.59 577.31, 579.78 68.06 26.91 5.03 0.73 0.97 1.00 20 wt% Cr0.2–V0.8/TiO2 529.82, 530.35 516.75, 517.42 576.93, 579.10 70.86 21.97 7.17 0.48 0.88 0.95

Table 2 Binding energies and surface atomic concentrations of O, Cr, and V for the prepared catalysts determined from deconvoluted XPS spectra.

Fig. 5. H2-TPR curves of catalyst with various Cr/V molar ratios. (1) VOx/TiO2; (2) Cr0.1–V0.9/TiO2; (3) Cr0.2–V0.8/TiO2; (4) Cr0.3–V0.7/TiO2; (5) Cr0.5–V0.5/TiO2; (6) Cr0.7–V0.3/TiO2; (7) CrOx/TiO2.

The V 2p XPS spectra of the catalyst samples are shown in Fig. 5(b). The peaks at 517.5 and 516.5eV can be ascribed to the 2p_3/2 binding energies of V⁵⁺ and V^{4+ [22]}. The surface atomic ratios of V⁴⁺ to V⁵⁺ are listed in Table 2; these were obtained by deconvolution. The 10 wt% Cr_0.2-V_0.8/TiO₂ catalyst had the highest V⁴⁺/V⁵⁺ ratio, and the ratio decreased in the order 10 wt% Cr_0.2-V_0.8/TiO₂>10 wt% VO_x/TiO₂>20 wt% Cr_0.2-V_0.8/TiO₂> 5 wt% Cr_0.2-V_0.8/TiO₂. A large V⁴⁺/V⁵⁺ ratio would generate more free electrons, which are formed from non-stoichiometric V ions in various valence-state-conversion processes.

Fig. 5(c) shows the XPS spectra of Cr 2p on the catalyst surfaces. The Cr atoms on the catalyst surfaces had two valences (Cr⁶⁺ and Cr⁴⁺). The data in Table 2 clearly show that the Cr-V loading on the catalyst had little effect on the Cr³⁺/Cr⁶⁺ molar ratio.

The NO_x conversions over catalysts with various Cr/V molar ratios as a function of reaction temperature are shown in Fig. 6. VO_x/TiO₂ and CrO_x/TiO₂ had poor activities at low temperatures, and CrO_x/TiO₂ had a narrow temperature window. The results show that small variations in the amount of Cr in the Cr-V/TiO₂ catalysts influenced NO_x conversion. The curves for Cr_0.1-V_0.9/TiO₂ and Cr_0.2-V_0.8/TiO₂ show that the catalytic activities were significantly improved at low temperatures with increasing Cr content because this enhanced the weak acidity of the SCR catalyst. However, excess Cr was unfavorable for the reaction because the active sites on the V-based catalyst were occupied by Cr, as shown by the curves for Cr_0.3-V_0.7/TiO₂, Cr_0.5-V_0.5/TiO₂, and Cr_0.7-V_0.3/TiO₂. The NO_x conversion rate exceeded 90% at 160°C when the Cr/V molar ratio was 0.2:0.8; this catalytic activity at low temperatures is clearly better than those of the other catalysts, and the temperature window of 160-300°C was broadened. These findings are in agreement with the NH₃-TPD and H₂-TPR results.

Fig. 6. NOx conversions over catalysts with various Cr/V molar ratios. Reaction conditions: GHSV = 60000 mL/(h·g), j (NO) = j(NH3) = 0.05%, j (O2) = 5%.

The effects of the catalyst loading on the activities of the DeNO_x catalysts were investigated; the results are shown in Fig. 7. The Cr-V content greatly influenced the NH₃-SCR activity. NO_x conversions over the catalysts decreased in the order 10wt% > 20 wt% > 50 wt% > 5 wt%. As shown in Table 2, a 5wt% loading of the Cr_0.2-V_0.8/TiO₂ catalyst contained only 2.39% V on the catalyst surface. The number of active sites affects the adsorption rate, therefore the catalyst with a 5 wt% loading decreased the SCR reaction rate. In contrast, the catalyst containing 10 wt% Cr-V had the highest V content on the catalyst surface. Vanadium has a stable oxidation state of +5; V⁴⁺ is reduced compared with V⁵⁺ and has one extra electron, which can act as a free electron because there is no constraint on the atom, which shows high mobility and activity. It is therefore reasonable to conclude that an increased V⁴⁺/V⁵⁺ ratio could lead to an increase in the SCR reaction rate, as a result of the free electrons formed in non-stoichiometric V ^[23]. The catalyst containing 10 wt% Cr-V therefore showed the highest activity and gave nearly 100% NO_x conversion above 200°C. However, excess loading could lead to aggregation of V₂O₅ and Cr₂V₄O₁₃, cover the V active sites, and reduce the DeNO_x catalytic activity.

Fig. 7. Effect of active phase loading on NOx conversion. j(NO) = j(NH3) = 0.05%, j(O2) = 5%.

Fig. 8 shows the resistance of the Cr_0.2-V_0.8/TiO₂ catalyst to SO₂ poisoning during SCR of NO_x with NH₃. The Cr_0.2-V_0.8/TiO₂ catalytic activity was high and NO_x conversion was higher than 75% in the absence of SO₂. In 12h of test with flue gas containing 100ppm SO₂, the NO_x conversion below 220°C remained at about 99%. Moreover, no signs of sulfur poisoning were observed. However, the introduction of SO₂ slightly reduced NO_x conversion at 180°C, because SO₂ on the catalyst surface was easily adsorbed by the V-based catalyst at low temperatures, covering the active sites and ultimately inhibiting the activity.

Fig. 8. Resistance to SO2 of Cr0.2–V0.8/TiO2 catalyst. Reaction conditions: 220 °C or 180 °C, GHSV = 60000 mL/(h·g), [SO2] = 100 ppm.

The influence of water was tested by injecting 10 vol% water into the reaction. As shown in Fig. 9, the presence of water decreased the NO_x conversion by the Cr_0.2-V_0.8/TiO₂ catalyst at 220 °C. Several authors have attributed the hindering effect of water vapor to competitive adsorption between water and NH₃ ^[23]. The Cr_0.2-V_0.8/TiO₂ catalyst still had sufficiently high catalytic activity to ensure that NO_x conversion was higher than 85% in the copresence of water and SO₂.

Fig. 9. Resistance to water and copresence of water + SO2 of Cr0.2–V0.8/TiO2 catalyst. Reaction conditions: 220 °C, GHSV = 60000 mL/(h·g), [SO2] = 100 ppm, [water] = 10 vol%.

Doping of Cr ions increased the amounts of weak and medium acid sites on the VO_x/TiO₂ catalyst surface, promoted the low-temperature reductive capacity of high-valence V ions, and increased the amount of surface oxygen on the catalyst. These factors enhanced the SCR rate. The catalyst with a Cr/V molar ratio of 0.2:0.8 and 10% loading had the largest amount of weak acid sites and gave the best DeNO_x performance, with NO_x conversion greater than 90% at 160-300°C. This catalyst also showed high stability in the presence of 0.01% SO₂.