The titanium dioxide support was prepared by the sol-gel method. Briefly, 12 mL tetrabutyl orthotitanate (98%, TBOT), 48 mL absolute ethanol (AR), 3.8 mL acetylacetone (AR, Hacac) and a certain amount of aluminium acetylacetonate (98%, Al(acac)₃) were mixed and stirred for 1 h to obtain solution A while solution B consisted of 60 mL absolute ethanol and 7 mL H₂O. Solution B was dropwise added into solution A by titration funnel under magnetic stirring. After stirring for 1 h at room temperature, the sol was heated at 50˚C for 4 h in water bath and then subsequently dried at 80˚C for 6 h. After that the gel was calcined at 350˚C for 2 h in air. Finally, the titanium dioxide support was obtained.

The loading of V₂O₅ was prepared by impregnation with the requisite amount of ammonium metavanadate (the V₂O₅ content in all catalysts is 1% by weight). The NH₄VO₃ was dissolved in 20 mL deionized water at 60˚C, and impregnated by contacting the Al-Ti support. The mixture was heated for 4 h at 60˚C in a water bath. Then obtained mixture was dried at 120˚C for 6 h and calcined at 350˚C for 4 h. Samples with different ratios of Ti³⁺ were sighed as x-VTi which x represented the molar percentage of Al(acac)₃ to TBOT.

Powder X-ray diffraction (XRD) patterns of all samples were obtained on a XD-3 diffractometer with Cu Ka radiation (k = 0.15418 nm) (Beijing Purkinje General Instrument Co., Ltd., China) operated at 36 kV and 30 mA. Intensity data were reported in the 2θ range from 10˚ to 80˚, with a step size of 0.04˚.

Scanning electron microscopy (SEM) produced by Hitachi was used to observe the morphologies of catalysts under the accelerating voltage of 15 kV.

Experiments to investigate the catalytic activity of the catalyst were carried out in a fixed-bed flow reactor at 120˚C - 330˚C containing 0.30 g catalyst under atmospheric pressure. The reactor was heated by a temperature-controlled furnace. They were used to control the gas flow of mass flowmeters. The total gas flow rate was 100 mL/min. The reaction gas components were as follows: 500 ppm NO, 500 ppm NH₃, 5% O₂ and balanced N₂. The gas hourly space velocity was 23,885 h⁻¹. Different space velocities were obtained by changing the gas total flow or the volume of catalyst used. The mixed gas went into the reactor and the NO and NO₂ (NO_X = NO + NO₂) concentrations were monitored by a Testo350 flue gas analyzer (Testo, Germany). The reaction system was kept for 1 h at each reaction temperature to reach a steady state before the analysis of the outlet gas was performed. The NO_X conversion could be defined as Equation (1):

X NO X = [ NO X ] in − [ NO X ] out [ NO X ] in (1)

There [ NO X ] in and [ NO X ] out represent the concentrations of NO_X in the inlet and outlet gas stream respectively.

The activity of different Ti³⁺ doping amount is tested, and the results are shown in Figure 1. The denitration activity of all Ti³⁺ self-doped V₂O₅/TiO₂ catalysts increased with the increase of temperature. Compared with the 0-VTi catalyst, the activity efficiency is significantly improved. The catalytic activity of 0.2-VTi was the highest, and reached 88% at 210˚C. Therefore, when the Al(acac)₃/TBOT ratio is 0.2%, Ti³⁺ self-doping has the best effect on the catalyst.

The XRD patterns of the prepared catalysts are shown in Figure 2. It can be seen from Figure 2 that the sample after Ti³⁺ self-doping (the 0.2-VTi catalyst) shows the major crystalline phase of anatase TiO₂. However, the 0-VTi catalyst shows trace amount of rutile phase. This indicates that the Ti³⁺ doping can inhibit the formation of rutile phase. In addition, there is no characteristic peak of V₂O₅ was detected, indicating that V element may be well dispersed on the catalyst surface [15] [16].

Figure 3 shows the SEM images of the synthesized 0.2-VTi catalyst sample. SEM image shows that the composite consists of irregular shaped aggregates. Figure 3 shows there is no obvious change of structure between fresh catalyst (Figure 3(a)) and used catalyst (Figure 3(b)). Moreover, no single V₂O₅ particles or clusters were found on the surface of TiO₂. This indicates that V₂O₅ species may cover the surface of TiO₂ particles in the form of a single thick layer. This is consistent with the results of XRD [17].

Figure 4 shows the NO_x conversion as a function of temperature (from 120˚C to 330˚C) in the NH₃-SCR reaction over Ti³⁺ self-doped V₂O₅/TiO₂ catalysts of the different calcined temperature of the support. From the result, it was clear to be found that the NO_x conversion is the highest when the support calcinations temperature is 350˚C. It presented an over 80% NO_x conversion within the temperature range of 200˚C - 330˚C, especially at 240˚C - 330˚C where the NO_x conversion was increased to a maximum of exceed 99%. However, the catalytic activity at low temperature calcinations is relatively low. This may be because the low temperature calcined catalyst has a low degree of crystallinity [18]. And in appearance, the support prepared at a low temperature appeared black, because of incomplete calcinations and residual carbon, resulting in low denitrification efficiency. It can also be seen from Figure 4 that the catalytic activity declines when the support was calcined at high temperatures.

It was found through previous studies that oxygen is very important for SCR of

NO with NH₃ at low temperature [19]. Figure 5 shows the effect of oxygen concentration on SCR activity at 210˚C. It was clear that the NO_x conversion over the catalyst increased with the increasing of oxygen concentration, especially when the oxygen concentration was less than 3%. The 36% NO_x conversion was obtained in the presence of without O₂ at 210˚C. This indicates that the catalyst has the ability to catalyze and reduce in the absence of O₂. However, the NO_x conversion increases significantly as the O₂ concentration increased to 3%. This shows that O₂ has a great influence on the catalytic activity. The NO_x conversion increased to 88.1% when O₂ reached 5%. However, the catalytic activity remained basically stable with the continued increase of O₂ concentration. Therefore, the optimal O₂ concentration is 5%.

It is well known that the [NH₃]/[NO] molar ratio is a key parameter in SCR of NO with NH₃ [14]. Figure 6 shows the effect of different [NH₃/NO] molar ratio on catalytic activity at 210˚C. It can be clearly seen that the catalytic activity changes significantly when the molar ratio of [NH₃]/[NO] is less than one, and the activity rose rapidly as the [NH₃]/[NO] molar ratio increases. When the [NH₃]/[NO] molar ratio is 1, the catalytic activity reached over 88%. However, the NO_x conversion varied more gradually when a higher molar ratio of [NH₃]/[NO] was used. The continued increase of the NH₃ concentration will not increase the NO_x conversion. Therefore, the [NH₃]/[NO] molar ratio of 1 is advisable for this experiment.

The GHSV is an important parameter that has a great influence on the denitrification efficiency of SCR catalysts. This work also measured the effect of GHSV on SCR activity, as shown in Figure 7. It can be clearly seen that the catalytic activity decreases with the increasing of GHSV. And the catalytic activity was the best at 23,885 h⁻¹, in particular, the catalytic activity reached 88.1% at 210˚C and

the activity exceeded 99% between 240˚C - 330˚C. This is because when the GHSV is too high, the contact time between the reactants and the catalyst is short and the reaction is not sufficient, the conversion rate is low.

In order to evaluate the stability of the catalytic activity, a comprehensive experiment was carried out at a standard reaction of catalyst activity test at 210˚C. The results are shown in Figure 8. It was found that the NO conversion was always kept over 80% for the catalyst continued to react for over 1750 min·s, indicating that the catalyst had better activity and stability.

Sulfur dioxide as another important pollution was in the flue gas inevitably [12]. The effect of H₂O and SO₂ on the activity of the catalyst was investigated experimentally at 210˚C, and the result is exhibited in Figure 9. When the 500 ppm SO₂ was added in firstly, the obvious decrease of NO conversion was not observed. The NO_x conversion decreased from 88.1% to about 78% after a 250 min·s 'SCR reaction process. When the supply of SO₂ was cut off at 250 min, NO_x concentration in the exhaust recovered gradually to the original level. Then 5% H₂O was introduced at 450 min, and the experimental results showed that the NO_x conversion decreased rapidly. When the H₂O is turned off at 600 min, the NO_x conversion is quickly restored. Therefore, Ti³⁺ self-doped V₂O₅/TiO₂ catalyst has superior SO₂ and H₂O resistance.