Heteropoly acid promoted V₂O₅/TiO₂ catalysts for NO abatement with ammonia in alkali containing flue gases

Abstract

V₂O₅/TiO₂ and heteropoly acid promoted V₂O₅/TiO₂ catalysts were prepared and characterized by N₂ physisorption, XRPD and NH₃-TPD. The influence of the calcination temperature from 400 to 700 °C on crystallinity and acidic properties was studied and compared with the activity for the selective catalytic reduction (SCR) of NO with ammonia. The SCR activity of heteropoly acid promoted catalysts was found to be much higher than for unpromoted catalysts. The stability of heteropoly acid promoted catalysts is dependent on calcination temperature and there is a gradual decrease in SCR activity and acidity with increase in calcination temperatures. Furthermore, the heteropoly acid promoted V₂O₅/TiO₂ catalysts showed excellent alkali deactivation resistance and might therefore be alternative deNO_x catalysts in biomass fired power plants.

---

1. Introduction

Heteropoly acids (HPAs) and their salts are a class of compounds that have attracted significant scientific interest. Because of their unique structure, acidic and redox properties, they have in particular been studied as possible catalysts for a variety of reactions.^1–5 Acid catalysis and selective oxidation are the most important areas of catalytic application of HPAs. The majority of catalytic applications use the most stable and easily available Keggin HPAs, especially for acidic and redox catalysis.^6,7 Most typical Keggin HPAs such as H₃PW₁₂O₄₀, H₄SiW₁₂O₄₀ and H₃PMo₁₂O₄₀ are commercially available. HPAs possess stronger (Brønsted) acidity than conventional solid acid catalysts such as acidic oxides and zeolites and the acid strength of Keggin HPAs decreases in the order: H₃PW₁₂O₄₀ > H₄SiW₁₂O₄₀ > H₃PMo₁₂O₄₀ > H₄SiMo₁₂O₄₀.⁵ The acid sites in HPA are more uniform and easier to control than those in other solid acid catalysts. Usually, tungsten HPAs are the catalysts of choice because of their stronger acidity, higher thermal stability and lower oxidation potential compared to molybdenum acids.^1–5

It has been found that the 12-tungstophosphoric acid (TPA) can effectively absorb NO at flue gas temperatures, and upon rapid heating, the absorbed NO is effectively decomposed into N₂.^8,9 The results showed that the quantities of NO₂ lost from the gas phase follow the order: H₃PW₁₂O₄₀ > H₄SiW₁₂O₄₀ > H₃PMo₁₂O₄₀ and the quantity of NO₂ retained on TPA is strongly dependent on the temperature: it increases from 25 °C, reaches a maximum in the range from 150 to 300 °C, and decreases to small amounts from 500 to 600 °C. Supplementary experiments showed that the maximum quantity of NO taken up by the solid is approximately equal to those of NO₂.⁹ The adsorption of NO occurs via replacement of the structural water present between the Keggin units of heteropoly acids. NO_x adsorption/desorption capacities of TPA were measured under a representative exhaust lean gas mixture containing, for example, CO₂, H₂O and hydrocarbons and a mechanism was proposed for both NO_x adsorption and desorption on TPA.¹⁰

Later Pt/TPA and TPA supported metal oxides were also used extensively for the abatement of NO_x, primarily for mobile applications.^11,12 Recently, Pd was loaded on dispersed H₃PW₁₂O₄₀ (TPA) over the SiO₂ surface, and the catalyst was applied for selective reduction of NO with aromatic hydrocarbons for stationary applications. The catalyst exhibited high activity for NO reduction when branched aromatic hydrocarbons, such as toluene and xylene, were used as reductants.^13,14

Several authors have reported the deactivation of V₂O₅–WO₃/TiO₂ catalysts by alkaline metals in biomass fired power plants.^15–17 Most conclude that poisonous elements (e.g.potassium, barium) are affecting the Brønsted acid sites, which are responsible for the ammonia adsorption, thus decreasing both their number and activity for NO reduction. One of the possible ways to increase catalyst resistance to alkaline poisons is the use of supports, revealing high or super-acidic properties, which would interact stronger with alkali than the vanadium species. Such super-acidic properties are available in heteropoly acids.¹⁸ As mentioned above, heteropoly acids are strong Brønsted acids and catalyze a wide variety of reactions of both homogeneous and heterogeneous nature offering efficient and cleaner processes.

For practical applications, it is important to improve the physical properties of HPA (e.g., mechanical and thermal resistance). This could be reached by depositing HPA on a suitable support while preserving its chemical properties (adsorption capacity). Dispersing HPA on solid supports is important for catalytic application because the specific surface area of the unsupported HPA usually is low in the range 1–10 m² g⁻¹ (although interstitial voids are created by the terminal oxygen atoms linking the hydrated protons these are not interconnected). In general, HPA strongly interacts with supports at low loading levels, while the bulk properties of HPA prevail at high loading levels. To overcome these disadvantages the HPAs are usually supported on a suitable carrier that not only increases the available surface area but also improves the catalytic performance.

The selection of proper support material has to take into account the strong acidity of HPAs.¹⁹ If a support is moderate to strongly basic (e.g., Al₂O₃, MgO), the interaction with HPA is too strong and leads to an acid–base reaction with loss of crystallinity of HPA with a complete degradation of its storage properties. If the support is acidic (e.g., SiO₂), the structure of HPA exists, but the anchoring is not secured. In the case of lower acidity (e.g., ZrO₂, TiO₂ and SnO₂), the structural properties are retained and the activity remains high. Consequently, oxide supports can be selected from their isoelectric point (around 7). From those results, and in order to improve the performance of the simple titanium and zirconium oxides, we have chosen TiO₂ materials as supports for HPA.

In the present work, we have studied the promotional effect and alkali resistance of TiO₂ supported heteropoly acid with 3 wt% V₂O₅ as the active material on the activity of the SCR reaction with NH₃ as a reducing agent. Thus, the influence of potassium oxide additives on the activity was studied and compared with traditional V₂O₅/TiO₂ and commercial V₂O₅–WO₃/TiO₂ SCR catalysts. All the catalysts were characterized by various techniques to allow detailed discussion of the compositional effects on the SCR performance.

2. Experimental

2.1 Catalyst preparation and characterization

Anatase-supported heteropoly acids H₃PW₁₂O₄₀ (TPA), H₄SiW₁₂O₄₀ (TSiA), and H₃PMo₁₂O₄₀ (MPA) (Aldrich, 99.9%) were prepared by suspending a known amount of dried TiO₂ anatase powder (Aldrich, 99.9%) in aqueous solution of corresponding heteropoly acids. The suspension mixture (optimum heteropoly acids loading, 15 wt%) was dried at 120 °C for 12 h.^20,21 3 wt% V₂O₅ modified catalysts were prepared by wet impregnation by dissolving the required amount of ammonium meta-vanadate (Aldrich, 99.9%) as a precursor in 2 M oxalic acid of the pure TiO₂ and heteropoly acid–TiO₂ supports. For comparison a 3%V₂O₅–7%WO₃/TiO₂ (VWTi) commercial catalyst is also considered.

The potassium-doped catalysts were prepared by co-impregnation with a solution of KNO₃ (Aldrich, 99.9%) to obtain a potassium loading of 100 μmol g⁻¹catalyst corresponding to a K/V molar ratio of 0.3. Each impregnated catalyst was oven dried at 120 °C for 12 h followed by calcination at 400–700 °C for 4 h prior to use.

X-Ray powder diffraction (XRPD) measurements were performed on a Huber G670 powder diffractometer using CuK_α radiation within a 2θ range of 10–60° in steps of 0.02°. BET surface area of the samples was determined from nitrogen physisorption measurements on about 100 mg of the sample at liquid nitrogen temperature with a Micromeritics ASAP 2010 instrument. The samples were heated to 200 °C for 1 h prior to the measurement.

NH₃-TPD experiments were conducted on a Micromeritics Autochem-II instrument. In a typical TPD experiment, 100 mg of the dried sample was placed in a quartz tube and pretreated in flowing He at 500 °C for 2 h. Then, the temperature was lowered to 100 °C and the sample was treated with anhydrous NH₃ gas (Air Liquide, 5% NH₃ in He). After NH₃ adsorption, the sample was flushed with He (50 ml min⁻¹) for 100 min at 100 °C. Finally, the TPD operation was carried out by heating the sample from 100 to 700 °C (10 °C min⁻¹) under a flow of He (25 ml min⁻¹).

2.2 Catalytic activity measurements

The SCR activity measurements were carried out at atmospheric pressure in a fixed-bed quartz reactor loaded with 20 mg of fractionized (180–300 μm) catalyst samples positioned between two layers of inert quartz wool. The reactant gas composition was adjusted to 1000 ppm NO, 1100 ppm NH₃, 3.5% O₂, 2.3% H₂O and balance N₂ by mixing 1% NO/N₂ (±0.1% abs.), 1% NH₃/N₂ (±0.005% abs.), O₂ (≥99.95%) and balance N₂ (≥99.999%) (Air Liquide) using Bronkhorst EL-Flow F-201C/D mass-flow controllers. The total flow rate was maintained at 500 ml min⁻¹ (ambient conditions). Further studies with SO₂ were not performed on these catalysts since straw or wood chips have a very low content of sulfur. During the experiments the temperature was increased stepwise from 200 to 540 °C while the NO and NH₃ concentrations were continuously monitored by a Thermo Electron Model 17C chemiluminescent NH₃–NO_x gas analyzer. The catalytic activity is represented as the first-order rate constant (cm³ g⁻¹s⁻¹), since the SCR reaction is known to be first-order with respect to NO under stoichiometric NH₃ conditions.²² The first-order rate constants under plug flow conditions were obtained from the conversion of NO as:

k = −(F_NO/(m_catC_NO))ln(1 − X) (1)

where F_NO denotes the molar feed rate of NO (mol s⁻¹), m_cat the catalyst weight (g), C_NO the NO concentration (mol cm⁻³) in the inlet gas and X the fractional conversion of NO.

3. Results and discussion

To evaluate the thermal stability of the catalysts it is very convenient to study the crystalline phase transformations of the materials. For pure titania an amorphous behaviour was observed below 350 °C consisting of a mixture of anatase, brookite and rutile phases. When increasing the calcination temperature, the amount of the anatase phase increased and became predominant at 500 °C. Upon heating to 700 °C the anatase phase of titania was completely transformed into the rutile phase.²³ The XRD patterns of the catalysts with 15 wt% TPA loading calcined at 700 °C showed the role of TPA which strongly influences the crystallization of titanium hydroxide into titania and the development of new textural properties with temperature compared to pure titania.^24,25

The XRPD patterns of VTPATi, VTSiATi, and VMPATi samples calcined at various temperatures are shown in Fig. 1. At 400 °C no diffraction lines attributed to crystalline V₂O₅ or HPAs were observed, only support TiO₂ patterns can be observed indicating that the vanadium and HPAs are highly dispersed on the support. Both anatase (2θ = 25.3°, 37.9°, 47.8° and 54.3°) and a very small amount of the rutile (2θ = 27.4°, 36.1° and 54.2°) phases are present in the catalysts. Partial transformation into rutile can be seen at 600 °C and transformation into a rutile rich phase happens at 700 °C. The intensity of the rutile phase is varying for the catalysts in the order VMPATi > VTSiATi > VTPATi. At a calcination temperature of 700 °C some decomposition products of HPAs like MoO₃ (2θ = 23.3° and 27.3°) and WO₃ (2θ = 23° and 24°) can be observed along with the rutile rich support. Potassium doped catalysts showed further increase in the rutile phase in the order KVMPATi > KVTSiATi > KVTPATi. This confirms that the presence of vanadium or potassium promotes the transformation of anatase into the rutile phase with increasing calcination temperature.

Fig. 1 XRPD patterns of fresh and potassium doped VTPATi, VMPATi, and VTSiATi catalysts at various calcination temperatures.

The results of the N₂-BET surface area measurements are summarized in Table 1 for fresh and potassium doped catalysts calcined at 400 °C. The surface area of the VTi catalyst was found to be 128 m² g⁻¹ while those of the HPA promoted catalysts were slightly lower. The surface area of the commercial VWTi catalyst was found to be 68 m² g⁻¹. Potassium deactivated catalysts showed further decrease in surface area, possibly due to pore blocking. Temperature-programmed desorption (TPD) of ammonia or pyridine is a frequently used method for determining the surface acidity of solid catalysts. Ammonia is often applied as a probe molecule because of its small size, stability and high basic strength (pK_a = 9.2).²⁶ In the present investigation, the acidity measurements have been carried out by the NH₃-TPD method. The total amount of adsorbed ammonia, which is determined from the area under the TPD curve, corresponds to weakly adsorbed ammonia (temperature of desorption below 200 °C) and strongly adsorbed ammonia (temperature of desorption above 300 °C). These sites originate from the TiO₂ support, heteropoly ions and vanadium acidic sites present in the catalyst.

Table 1 Surface area and NH₃-TPD results of catalysts calcined at 400 °C

Catalyst	Surface area/m^2 g^−1	Acidity/μmol g^−1
VTi	128	571
VWTi	68	656
VTPATi	112	839
VTSiATi	114	809
VMPATi	96	787
KVTi	120	108
KVWTi	63	161
KVTPATi	102	503
KVTSiATi	104	463
KVMPATi	84	491

---

Fig. 2 shows the NH₃-TPD desorption patterns of pure HPAs, HPATi, VHPATi and KVHPATi catalysts in the temperature range of 100–650 °C. The pure HPAs exhibit sharp NH₃ desorption peaks between 300–500 °C, which could be ascribed to the desorption of NH₃ from HPA accompanied by the decomposition of HPA, because the NH₃ decomposition temperature is above the decomposition temperature of HPA. The order of the NH₃ desorption (or HPA decomposition) temperature is TPA > TSiA > MPA. A similar order was reported for these compounds by Kozhevnikov.^3,5 Pure TPA, TSiA and MPA HPAs showed acidity values of 1642, 1322 and 2647 μmol g⁻¹, respectively. Similar NH₃-desorption patterns with a NH₃ adsorption value of 1390 μmol g⁻¹ were reported on TSiA.²⁷ These acidity values indicate that the HPAs are super acidic in nature comparable to zeolites and acidic oxides.¹ Such a super acidic nature of HPAs is due to their discrete and mobile ionic structure, tunable by the chemical composition. TiO₂-supported HPAs calcined at 400 °C showed broad NH₃-desorption patterns from 150–500 °C. The low temperature desorption peak at around 150 °C is attributed to the weak acid sites of the support TiO₂, while the high temperature broad desorption peaks that appear at 300–550 °C could be due to the strong interaction of NH₃ with well-dispersed HPA on TiO₂. The TPATi, TSiATi and MPATi samples showed acidity values of 788, 765 and 755 μmol g⁻¹, respectively.

Fig. 2 NH₃-TPD profiles of pure HPA, HPATi, VHPATi, and KVHPATi catalysts calcined at 400 °C.

The VHPATi catalysts show intense NH₃-desorption peaks compared to the HPATi materials. The NH₃-TPD results in the temperature range of 100–650 °C for the VTPATi, VMPATi, and VTSiATi catalysts calcined at 400 °C are summarized in Table 1. The acidity of the pure VTi catalyst without promoters was found to be 571 μmol g⁻¹ while VMPATi, VTPATi and VTSiATi impregnated catalysts revealed increased acidity in the presence of vanadium. It is known that acidity of the catalysts is enhanced in the presence of vanadium on the support.²⁸ The acidity of the commercial VWTi catalyst revealed 656 μmol g⁻¹. The total acidity of the VHPATi catalysts followed the order VTPATi > VTSiATi > VMPATi.

Fig. 2 also shows the NH₃-TPD profiles of potassium doped catalysts calcined at 400 °C and the results are summarized in Table 1. Overall a significant decrease in total acidity and decrease in the intensity of high temperature desorption peaks are observed. It is rather obvious to assume that potassium oxide first occupies the strongest acid sites and then weakens the remaining acid sites due to electron donation. Especially, the KVTi catalyst acidity dropped from 571 to 108 μmol g⁻¹ (81%) while those of the KVWTi (75%), KVTPATi (40%), KVTSiATi (43%) and KVMPATi (38%) catalysts showed a moderate drop in acidity after doping. It is known that surface modified or promoted V₂O₅ catalysts show increased alkali resistance.^16,17

The influence of the calcination temperature on the acidity of the VTPATi, VMPATi, and VTSiATi catalysts is shown in Fig. 3. A gradual decrease in acidity of the catalysts can be seen with higher calcination temperature. The loss in acidity could be due to support phase transformations and HPAs decomposition which is also evident from the XRPD patterns. It is known that the HPAs are sensitive to high temperatures and they lose acidic protons with increasing temperature where usually, the tungsten containing HPAs are the most stable.^3,5

Fig. 3 Effect of calcination temperature on total acidity of VTPATi, VMPATi, and VTSiATi catalysts.

The catalytic activity of the pure HPAs (TPA, TSiA and MPA), and the TPATi, TSiATi and MPATi materials was measured in the temperature range 200–540 °C (Fig. 4). The catalytic activities obtained are shown as the first-order mass based rate constant k (eqn (1)). While measuring the rate constant the catalyst amount was chosen in such a way that the NO conversion values remained well below 90% to allow determination of k. All measurements were recorded after a steady state was obtained. Pure HPAs showed very little SCR activity. TiO₂-supported MPA, TPA, and TSiA catalysts calcined at 400 °C were more active following the order MPATi > TPATi > TSiATi. At higher reaction temperatures comparable activity was, however, obtained with the catalysts.

Fig. 4 Temperature dependency of the first-order rate constant for SCR of NO with pure HPAs, TPATi, MPATi, and TSiATi catalysts calcined at 400 °C. Reaction conditions: 1000 ppm NO, 1100 ppm NH₃, 3.5% O₂, 2.3% H₂O, balance N₂.

The catalytic activity of the VHPATi catalysts calcined between 400–700 °C is shown in Fig. 5. The VMPATi, VTPATi, and VTSiATi catalysts provided maximum activity when calcined at 400 °C whereas the catalysts calcined at 500–700 °C were relatively less active. From the calcination effect it is thus evident that the HPAs are sensitive to the calcination conditions. Lower calcination temperatures were not studied since the inactive amorphous TiO₂ phase is predominant and the optimal SCR reaction conditions are around 400 °C. When calcined at 400 °C the catalyst is rich in the anatase phase (most active support for the SCR reaction) and there is no crystalline V₂O₅ formed. Further increase in calcination temperature results in partial transformation of anatase into the rutile phase and formation of less active HPA decomposition products (WO₃ or MoO₃). It is evident that WO₃ and MoO₃ are excellent promoters, in the present case when they are in stable HPA form having high acidity and SCR activity as well.

Fig. 5 Temperature dependency of the first-order rate constant for SCR of NO with VTPATi, VMPATi, and VTSiATi fresh and deactivated catalysts calcined at the indicated temperature. Reaction conditions: 1000 ppm NO, 1100 ppm NH₃, 3.5% O₂, 2.3% H₂O, balance N₂.

The present catalysts are highly active and the operating temperature regions are similar. For the VMPATi, VWTi and VTi catalysts the k_max value is observed at 440 °C and by further increase in reaction temperature the activity decreases. Similarly the VTPATi and VTSiATi catalysts showed k_max at 460 °C and by further increase in reaction temperature the activity also decreases due to predominant ammonia oxidation (SCO) over SCR of NO.²⁹ Overall the VTi, VWTi, VMPATi, VTPATi and VTSiATi catalysts showed k_max values of 500, 905, 803, 966 and 963 cm³ g⁻¹s⁻¹, respectively, at their optimal operating temperature. The rate constants observed for VTPATi and VTSiATi are higher than those for the commercial VWTi catalyst. Hence comparison with the mass based rate constant gives clear idea about the potential of HPAs ability to enhance the SCR of NO.

The thermal stability of the catalysts can be best compared after calcination at 700 °C with the help of XRD measurements and the SCR activity. For these catalysts the relative SCR activity and the anatase phase XRD intensities are in the order VTPATi > VTSiATi > VMPATi. Overall the WO₃ containing VTPA and VTSiA catalysts are more active than the MoO₃ containing VMPA catalysts.

Doping the optimum catalysts (calcined at 400 °C) with potassium (100 μmol g⁻¹ or K/V = 0.3) resulted in decrease in activity and a small shift of k_max towards lower temperature (Fig. 5). A possible explanation for such a temperature shift is that the potassium loading reduced the activity of the main NO-SCR reaction while the rate of the side reaction of ammonia oxidation remained constant or even increased. All the potassium doped HPA catalysts showed similar profiles as that of undoped catalysts. The KVMPATi catalyst had a k_max value at 400 °C while the KVTPATi and KVTSiATi catalysts showed a k_max at 440 °C. Especially the KVTi catalyst exhibited a decrease in k_max from 500 to 155 cm³ g⁻¹s⁻¹ implying the severe poisoning by alkali in the absence of HPAs. In a similar way, the commercial VWTi catalyst also exhibited decrease in k_max from 905 to 320 cm³ g⁻¹s⁻¹. On the VTi and VWTi catalysts, potassium seems to coordinate preferably with the vanadium sites and make them inactive for the SCR reaction, whereas HPA promoted V₂O₅/TiO₂ catalysts showed much better deactivation resistance. The super acidic sites of HPAs can host the potassium poisons whereby the SCR active vanadium species are protected from deactivation.

The KVTi and KVWTi catalysts possessed 30–33% of their initial activity after poisoning while the KVMPATi, KVTPATi and KVTSiATi catalysts performed with 88%, 81% and 71% of the initial activity, respectively, at 400 °C. Consequently, the potassium deactivation was significantly lower in the present catalysts compared to traditional SCR catalysts. Also highly active V₂O₅–WO_x/ZrO₂ catalysts reported in the literature for biomass fired applications exhibited severe deactivation.³⁰

4. Conclusions

Heteropoly acid promoted V₂O₅/TiO₂ catalysts showed excellent alkali deactivation resistance compared to V₂O₅/TiO₂ and V₂O₅–WO₃/TiO₂ catalysts. The promoted catalysts are sensitive to high calcination temperature where acidity is lost and less SCR active materials are formed. When WO₃ or MoO₃ are in a stable heteropoly acid matrix they showed higher activity than in the decomposed state. Heteropoly acid promoted V₂O₅/TiO₂ catalysts are thus promising catalysts for coal fired as well as biomass fired power plant SCR applications.

Acknowledgements

Energinet.dk is thanked for financial support of this work through the PSO project FU7318.

References

1. T. Okuhara, N. Mizuno and M. Misono, Adv. Catal., 1996, 41, 113.
2. C. L. Hill (ed.), Polyoxometalates, Chem. Rev., 1998, 98.
3. I. V. Kozhevnikov, Chem. Rev., 1998, 98, 171.
4. N. Mizuno and M. Misono, Chem. Rev., 1998, 98, 199.
5. I. V. Kozhevnikov, Catalysts for Fine Chemicals, Wiley, Chichester, England, 2002, vol. 2.
6. M. T. Pope, Heteropoly and Isopoly Oxometalates, Springer, Berlin, 1983.
7. Polyoxometalate Chemistry from Topology via Self assembly to Applications, ed. M. T. Pope and A. Muller, Kluwer, Dordrecht, 2001.
8. R. T. Yang and N. Chen, Ind. Eng. Chem. Res., 1994, 33, 825.
9. R. Belanger and J. B. Moffat, J. Catal., 1995, 152, 179.
10. S. Hodjati, C. Petit, V. Pitchon and A. Kiennemann, J. Catal., 2001, 197, 324.
11. K. Vaezzadeh, C. Petit and V. Pitchon, Catal. Today, 2002, 73, 297.
12. M. A. Gomez-Garcia, V. Pitchon and A. Kiennemann, Catal. Today, 2005, 107–108, 60.
13. K. Okumura, R. Yoshimoto, K. Suzuki and M. Niwa, Bull. Chem. Soc. Jpn., 2005, 78, 361.
14. R. Yoshimoto, T. Ninomiya, K. Okumura and M. Niwa, Appl. Catal., B, 2007, 75, 175.
15. Y. Zheng, A. D. Jensen, J. E. Johnsson and J. R. Thøgersen, Appl. Catal., B, 2008, 83, 186.
16. A. L. Kustov, M. Yu. Kustova, R. Fehrmann and P. Simonsen, Appl. Catal., B, 2005, 58, 97.
17. J. Due-Hansen, S. Boghosian, A. Kustov, P. Fristrup, G. Tsilomelekis, K. Ståhl, C. H. Christensen and R. Fehrmann, J. Catal., 2007, 251, 459.
18. R. S. Drago, J. A. Dias and T. Maier, J. Am. Chem. Soc., 1997, 119, 7702.
19. S. Hodjati, K. Vaezzadeh, C. Petit, V. Pitchon and A. Kiennemann, Top. Catal., 2001, 16/17, 151.
20. B. M. Devassy and S. B. Halligudi, J. Catal., 2005, 236, 313.
21. A. Alsalme, E. F. Kozhevnikova and I. V. Kozhevnikov, Appl. Catal., A, 2008, 349, 170.
22. R. Q. Long and R. T. Yang, J. Catal., 2000, 196, 73.
23. S. M. Kumbar, G. V. Shanbhag, F. Lefebvre and S. B. Halligudi, J. Mol. Catal. A., 2006, 256, 324.
24. E. Ortiz-Islas, T. Lopez, R. Gomez, M. Picquart, D. H. Aguilar and P. Quintana, Appl. Surf. Sci., 2005, 252, 853.
25. S. Eibl, B. C. Gates and H. Knoziger, Langmuir, 2001, 17, 107.
26. A. Satsuma, Y. Kamiya, Y. Westi and T. Hattori, Appl. Catal., A, 2000, 194–195, 253.
27. A. Miyaji, T. Echizen, K. Nagata, Y. Yoshinaga and T. Okuhara, J. Mol. Catal. A., 2003, 201, 145.
28. S. S. R. Putluru, A. Riisager and R. Fehrmann, Appl. Catal., B, 2010, 97, 333.
29. M. Yates, J. A. Martín, M. Ángeles Martín-Luengo, S. Suárez and J. Blanco, Catal. Today, 2005, 107–108, 120.
30. J. Due-Hansen, A. L. Kustov, S. B. Rasmussen, R. Fehrmann and C. H. Christensen, Appl. Catal., B, 2006, 66, 161.

---