Deactivation mechanism of potassium additives on Ti_0.8Zr_0.2Ce_0.2O_2.4 for NH₃-SCR of NO

Abstract

Potassium additives slightly decreased catalytic activity for NH₃-SCR of NO when the mass amount of K₂O was less than 1.0%, but dramatically decreased the catalytic activity when the mass amount of K₂O was more than 3.0%. The deactivation mechanism was that potassium additives directly decreased the specific surface area, the total acid amount and the acid strength of the strong acid sites, and induced the KCl-rich crystallization and the formation of non-active K₂ZrO₃, K₂TiO₃ and K₂CeO₃.

---

In light of worse air pollution and global warming threats, it has been a worldwide topic to build a low carbon society.¹ NO_x is a major air pollutant that causes acid rain, photochemical smog and the greenhouse effect.^2–4 Stringent emission regulations, rigorous emission control standards and growing environmental awareness are being imposed or raised to reduce NO_x pollution. Denitrification has become a very important research subject worldwide. Selective catalytic reduction (SCR) with merits of high efficiency, perfect stability, mature technology and large amount of flue gas treatment has been regarded as one of the most promising approaches for removing NO_x from stationary sources.^5–7 As the technological core of SCR, denitrification catalysts have been attracting wide attention. In particular, transition metal oxides are probably the best candidates for SCR of NO_x due to efficient catalytic activity, high N₂ selectivity, perfect anti-poisoning ability, superior hydrothermal stability, and low cost.^8–11 Of all the metal oxide catalysts, a commercial catalyst of V₂O₅(WO₃, MoO₃)/TiO₂ is the most attractive for NH₃-SCR of NO_x.¹² But vanadium pentoxide possesses toxicity, the spent catalyst has been a secondary solid pollutant,¹³ and the cost of nano-titania is very expensive.¹⁴ Consequently, the V–Ti-based catalyst will be limited for industrial application in the near future. Thus, developing a novel metal oxide catalyst with advantages of high catalytic performance, perfect anti-poisoning ability, low cost and being environmentally friendly has become an emerging research topic in denitrification. A novel complex oxide of Ti_0.8Zr_0.2Ce_0.2O_2.4 (TZC-4) possesses perfect catalytic performance and is considered to be a potential catalytic material for future SCR of NO_x.¹⁵

In actual denitrification projects, catalyst deactivation directly decreases deNO_x efficiency due to poisoning, fouling, surface masking, pore blocking and over-sintering.^16,17 In order to avoid catalyst poisoning induced by ammonium sulfites or ammonium sulfates, Efstathiou and coworkers developed a novel Pt/MgO–CeO₂ catalyst for SCR of NO with H₂ instead of NH₃.^18–22 For the factors of catalyst deactivation, besides the bad influences of atmospheres including SO₂ and water vapor, fly ash is another major deactivation factor because the fly ash contains many physical and chemical deactivating species such as alkaline metals, CaO, As and so on. In particular, the alkaline metal is one of the worst poisons for SCR catalysts. Chen et al.²³ and Larsson et al.²⁴ both found that the inhibitory effect of alkaline metals on catalytic performance followed the order of Cs₂O > Rb₂O > K₂O > Na₂O > Li₂O. In view of real alkaline metal contents in coal-fired fly ash, potassium has been regarded as the most poisonous alkaline metal for SCR catalysts.^25,26 For deactivation of V₂O₅/TiO₂ poisoned by potassium, potassium preferentially coordinates to the hydroxyl groups of V₂O₅ as well as the co-catalysts, consequently decreasing the surface acid amount and acid strength of the Brønsted acid sites and directly decreasing the catalytic activity.²⁷ Anyway, potassium decreased surface acidity and inhibited effective activation of NH₃ on the catalyst surface.^23,28 In the present work, with the objective of understanding the deactivation mechanism of potassium additives on the TZC-4 complex oxide for NH₃-SCR of NO, the effects of potassium additives on catalytic performance, specific surface area, micro-structure and surface acidity of the TZC-4 complex oxide were systematically studied.

We synthesized catalysts by sol–gel method with potassium mass amount from 0 to 10.0%, as shown in Table 1. The catalytic performances of the catalysts for NH₃-SCR of NO were tested under identical conditions (NH₃/NO = 1, GHSV = 50 000 h⁻¹, catalyst particle sizes = 1.25–8.00 mm). The effect of reaction temperature on catalytic activity for NH₃-SCR of NO in the presence of oxygen was shown in Fig. 1. As the reaction temperature increased, all catalytic activities of the catalysts increased at first and decreased at the highest temperatures. The effect of reaction temperature on catalytic activity may be caused by the chemical structure and surface concentration of the active NO_x intermediate species. The chemical structure of the active NO_x species was found to depend on reaction temperature.²⁰ Particularly, the chemical structure was that of bidentate or monodentate nitrate (NO₃⁻) at T < 200 °C and that of chelating nitrite (NO₂⁻) at T > 200 °C. The concentration of the active NO_x intermediates that lead to N₂ formation was found to be practically independent of reaction temperature (120–300 °C).²⁰ The maximum NO conversion for NH₃-SCR of NO was a result of two competing reactions: NO reduction by NH₃ and oxidation of NH₃ by oxygen.²⁹ The oxidation of NH₃, which reduced the amount of reductant, became dominating at higher temperatures. Consequently the high temperature catalytic activities decreased quickly. The TZC-4 blank sample exhibited a stable catalytic activity in temperature range of 200–350 °C, and the maximum catalytic activity was obtained 96.9% at 300 °C. Moreover, the catalytic activity increased at the low temperature range of 150–200 °C and decreased dramatically at more than 350 °C. Compared with the catalytic activity of the TZC-4 blank sample, potassium additives slightly decreased catalytic activity when the mass amount of K₂O was less than 1.0%, and the maximum catalytic activity of the TZCK₁ was still able to maintain 72.04% at 300 °C. Kamata et al.²⁷ and Zhu et al.³⁰ both demonstrated that the catalytic activity decreased sharply as the amount of K₂O increased in the V₂O₅–WO₃/TiO₂ catalysts, and the catalytic activity was almost completely lost when the mass amount of K₂O reached 1.0%. As seen in Fig. 2, Zhu and coworkers³¹ found that the decline rate of NO removal is almost proportional to the K₂O content. When n_K/n_V is equal to 2, herein the mass content of K₂O is almost equal to 1.0%, the catalyst of 1%V₂O₅–10%WO₃/TiO₂ obtains its maximum activity loss. In contrast, the TZC-4 complex oxide possesses stronger anti-potassium poisoning ability than that of V₂O₅–WO₃/TiO₂, and the V₂O₅–WO₃/TiO₂ catalyst deactivates sooner than TZC-4 for NH₃-SCR of NO in the presence of K₂O. However, potassium additives also dramatically decreased the catalytic activity when the mass amount of K₂O was more than 3.0%; even the catalysts almost completely lost their catalytic activity. Additionally, potassium additives had more influence on catalytic performance at low temperature than that at high temperature, but potassium additives have little effect on the active temperature range of 200–350 °C. Anyway, potassium additives have a bad influence on the TZC-4 complex oxide for NH₃-SCR of NO.

Table 1 Summary of the synthesized catalysts

Sample	Abbreviation	K2O mass amount (%)	Specific surface area (m^2 g^−1)
1	TZC-4	0	108.70
2	TZCK0.1	0.1	95.36
3	TZCK0.3	0.3	70.84
4	TZCK0.5	0.5	—
5	TZCK0.8	0.8	—
6	TZCK1	1	—
7	TZCK3	3	57.38
8	TZCK5	5	—
9	TZCK8	8	—
10	TZCK10	10	—

---

Fig. 1 Effect of reaction temperature on catalytic activity of the catalysts.

Fig. 2 Effect of K₂O content on denitrification efficiency for NH₃-SCR of NO. (This work was cited from ref. 21.)

We detected the specific surface area and pore diameter of the representative catalysts by N₂-BET method, as shown in Table 1. The BET specific surface areas of TZC-4, TZCK_0.3 and TZCK₃ were 108.70 m² g⁻¹, 70.84 m² g⁻¹ and 57.38 m² g⁻¹, respectively. It seems that the more potassium additive the smaller the BET specific surface area, indicating that the decrease of specific surface area directly decreased the catalytic performance. Furthermore, potassium additives increased pore diameters of the TZCK_x complex oxides because the radius of K⁺ (1.38 Å) is bigger than the radii of Zr⁴⁺ (0.72 Å), Ti⁴⁺ (0.61 Å) and Ce⁴⁺ (1.03 Å). The average pore diameters of TZC-4 and TZCK_0.3 were 8.8 nm and 10.0 nm, respectively.

The micro-structure of the TZC-4 complex oxide was observed by TEM, as presented in Fig. 3. Particle agglomerates were presented in Fig. 3(A), and a micro-particle size of 6–8 nm was estimated. For deeper insight, the analysis of high-resolution image was undertaken to establish the structure of the nano-particles. Fig. 3(B) showed the HR-TEM image of the TZC-4 complex oxide subjected to the same thermal treatment, it could be seen that the solid phase of the nano-particles was amorphous. The micro-structure of the representative potassium-doped TZCK₃ was observed by SEM, as shown in Fig. 4. Compared with the micro-structure of the TZC-4 complex oxide, the micro-particles also gathered into groups, but potassium additives dramatically increased micro-particle size of the TZCK₃ complex oxide, and a micro-particle size of 0.1–0.5 μm was estimated. Moreover, there were some big particles on the surface of the TZCK₃ complex oxide. The element chemical compositions of the A and B micro-regions in Fig. 4(a) were detected by EDX; the results are shown in Fig. 4(b) and (c), respectively. According to the results of element quantitative analysis, the big particles on the surface of the TZCK₃ complex oxide were mainly made from K and Cl, and the molar ratio of K/Cl was almost equal to 1, indicating KCl-rich crystallization on the catalyst surface. Zheng et al.^32,33 also found that potassium in the form of chloride was a strong poison for the catalyst. The micro-particles in the B micro-region were mainly made from the elements of Ti, Zr, Ce and K, suggesting that some solid solutions were made of the above elements. Subsequent XRD analysis confirmed the formations of the K₂ZrO₃, K₂TiO₃, and K₂CeO₃.

Fig. 3 TEM (A) and HR-TEM (B) images of the TZC-4 complex oxide.

Fig. 4 SEM micrograph and EDX spectra of the TZCK₃. (a) SEM micrograph of the TZCK₃; (b) EDX spectrum of A micro-region; (c) EDX spectrum of B micro-region.

The solid phase structures of the potassium-doped catalysts were characterized by means of XRD, as shown in Fig. 5. Previously we have demonstrated that the solid phase of the TZC-4 complex oxide was amorphous,¹⁵ and the TEM characterization also confirmed this finding. After doping with potassium, the major solid phase was still amorphous, but the original solid-phase equilibrium of the TZC-4 complex oxide was broken and the degree of crystallinity increased with the increasing amount of K₂O. All the weak X-ray diffraction peaks of the potassium-doped catalysts exhibited many types of young crystals, including anatase TiO₂ (PDF-ICDD 1-562), cubic CeO₂ (PDF-ICDD 1-800), orthorhombic ZrTiO₄ (PDF-ICDD 7-290), orthorhombic Ce₂TiO₅ (PDF-ICDD 49-1606), cubic Ce₂Zr₂O₇ (PDF-ICDD 8-221), cubic KCl (PDF-ICDD 1-786), orthorhombic K₂ZrO₃ (PDF-ICDD 72-824), K₂TiO₃ (PDF-ICDD 1-1016) and cubic K₂CeO₃ (PDF-ICDD 31-989). These results not only illustrated the KCl-rich crystallization, but also signified the occurrence of the following three solid state reactions:

K₂O + CeO₂ → K₂CeO₃ (1)

K₂O + TiO₂ → K₂TiO₃ (2)

K₂O + ZrO₂ → K₂ZrO₃ (3)

Yun et al.³⁴ and Onodera et al.³⁵ both demonstrated that the poisoning mechanism of K₂O for V-based catalyst was caused by the formation of non-active KVO₃ that directly decreased acid amount and acid strength of the Brønsted acid sites. Moreover, potassium significantly inhibited the reduction of the surface vanadia species.³⁶ Anyway, the TZC-4 complex oxide possessed stronger anti-potassium poisoning ability than that of V₂O₅–WO₃/TiO₂ in NH₃-SCR of NO. Due-Hansen et al.³⁷ found that the resistance of the catalysts towards model poisoning with potassium depended dramatically on the crystallinity of the zirconia. Kustov et al.³⁸ also reported that the vanadia, copper and iron oxide catalysts based on sulphated-ZrO₂ revealed the highest resistance towards alkali poisoning. Herein the zirconia component may also improve the resistance towards potassium poisoning in NH₃-SCR of NO.

Fig. 5 X-ray diffraction patterns of the potassium doped catalysts.

The total acidities of the catalysts were evaluated by NH₃-TPD, as shown in Fig. 6. All the ammonia desorption profiles of the representative TZC-4, TZCK₁ and TZCK₃ complex oxides proceeded in two stages: the ammonia desorption peaks centered at low temperature corresponded to weak acid sites, the other ammonia desorption peaks centered at high temperature corresponded to strong acid sites. For the weak acid sites, all the low temperature ammonia desorption peaks of the three catalysts were centered at about 200 °C, indicating that the three catalysts had the same acid strength of weak acid sites. However, the peak area of the ammonia desorption profile at the weak acid sites decreased as the potassium additives increasing. As mentioned above, potassium additives had little effect on the acid strength of the weak acid sites, but dramatically decreased the amount of weak acid sites. For the strong acid sites, potassium additives leaded to a shift of ammonia desorption center towards to lower temperature about 50 °C, which illustrated that potassium additives decreased the acid strength of the strong acid sites. Furthermore, the peak area of the ammonia desorption profile at the strong acid sites decreased as the potassium additives increasing, indicating that potassium additives also decreased the amount of strong acid sites. Previously we have already confirmed that the TZC-4 complex oxide simultaneously possessed Lewis acid sites and Brønsted acid sites, and that the Brønsted acid sites are the strong acid sites while the Lewis acid sites are the weak acid sites.¹⁵ Hence, herein the inhibitory influence of potassium additives on catalytic performance of the TZC-4 for NH₃-SCR of NO was majorly caused by decrease of the total acid amount and the acid strength of the strong Brønsted acid sites. This result is consistent with the research of Kamata et al.²⁷

Fig. 6 NH₃-TPD profiles of catalyst samples.

In summary, the TZC-4 complex oxide possessed stronger anti-potassium ability than that of V₂O₅–WO₃/TiO₂ but potassium additives still decreased its catalytic performance for NH₃-SCR of NO, especially when the mass additive of K₂O was more than 3.0%. The deactivation mechanism of potassium additives on catalytic performance indicates that potassium additives decreased the specific surface area, the total surface acid amount and the acid strength of the strong Brønsted acid sites; increased micro-particle size and the solid phase transformation from amorphous to crystalline; and induced the KCl-rich crystallization and the formation of non-active K₂ZrO₃, K₂TiO₃ and K₂CeO₃. The deactivation mechanism is depicted in Fig. 7.

Fig. 7 The deactivation mechanism of potassium additives on TZC-4 complex oxide for NH₃-SCR of NO.

Experimental

Synthesis of catalysts

The catalysts were separately synthesized by thermal decomposition of aged Ti–Zr–Ce–(K) composite gels calcined at 500 °C for 2 h. The Ti–Zr–Ce–(K) sols were synthesized using Ti(OC₄H₉)₄, ZrOCl₂·8H₂O, Ce(NO₃)₃·6H₂O and KNO₃ as precursors. Composite solution A was synthesized by dissolving Ti(OC₄H₉)₄ in ethanol (CH₃CH₂OH) under continuous stirring for 10 min, and composite solution B was synthesized by mixing glacial acetic acid, deionized water and CH₃CH₂OH. Solution B was added dropwise into solution A under continuous stirring for 20 min, and then the required amounts of Ce(NO₃)₃·6H₂O (Ti/Ce molar ratio was 4), ZrOCl₂·8H₂O (Ti/Zr molar ratio was 4) and KNO₃ were dissolved in AB mixed solution under vigorous stirring for 30 min to obtain the resulting Ti–Zr–Ce–(K) composite sols. The aged gels were obtained by aging of the sols.

Catalyst characterization

The specific surface area and average pore diameter (BJH method) of the samples were measured by N₂ adsorption/desorption isotherms at 77 K using a surface area analyzer (Micromeritics, 2020M V3.00H). All the samples were degassed at 350 °C under vacuum for 3 h prior to the adsorption experiments. Transmission electron microscopy (HR-TEM) was conducted by a JEM-2010 UHR transmission electron microscope, using a 200 kV accelerating voltage. The samples were dispersed in methanol using an ultrasonicator for 5 min and fixed on a carbon-coated copper grid (LC200-Cu, EMS) prior to the TEM observation. Scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) were performed on a JSM −5900 instrument to investigate surface morphology as well as the element compositions of the catalysts. X-ray diffraction (XRD) patterns were obtained on a D/MAX-RB X-ray diffractometer (Rigaku, Japan) using Cu Ka radiation. The 2θ scans covered the range 5–80°, and the accelerating voltage and applying current were 40 kV and 40 mA, respectively. Total acidity measurement was evaluated by a temperature-programmed desorption of ammonia (NH₃-TPD) using an Autochem 2910 (Micromeritics). The samples were preheated to 450 °C under helium flow for 1 h, and then cooled to 100 °C for the ammonia adsorption. A stream containing 10 vol% NH₃ in helium (30 mL min⁻¹) was introduced for 1 h to achieve the adsorption equilibrium. Then the samples were purged with helium for 1 h to remove the weakly adsorbed ammonia. Finally ammonia was desorbed in a flow rate of 30 mL min⁻¹ helium from 100 °C to 800 °C at a heating rate of 10 °C min⁻¹. The ammonia desorption was monitored online by Thermo ONIX ProLab mass spectrometer (m/z = 15).

Measurement of catalytic activity

Catalytic activity measurements of the catalysts for NH₃-SCR of NO were carried out in a fixed-bed quartz reactor (10 mm inner diameter), with 10.0 mL catalyst (particle sizes of 1.25–8.00 mm) and 8333.3 mL min⁻¹ gas flow rate corresponding to a GHSV of 50 000 h⁻¹. The simulated flue gas compositions were 700 parts per million by volume (ppmv) NO, 700 ppmv NH₃, 5 vol% O₂, and balance N₂. The NO concentrations at the inlet and outlet of the reactor were monitored on-line by a flue gas analyzer (KM9106, Kane, British). The catalytic activity for NH₃-SCR of NO was expressed by the equation:

X_NO = ([NO]_inlet − [NO]_outlet)/[NO]_inlet × 100% (4)

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 21106071), the New Teachers' Fund for Doctor Stations sponsored by the Ministry of Education of China (No. 20113221120004), the National High-Tech Research and Development Program of China (863 Program, No. 2009AA05Z313), the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD), and the Research Subject of Environmental Protection Department of Jiangsu Province of China (No. 201016).

Note and references

1. Y. G. Wang, P. He and H. S. Zhou, Energy Environ. Sci., 2011, 4, 805–817.
2. P. Forzatti, L. Lietti and I. Nova, Energy Environ. Sci., 2008, 1, 236–247.
3. G. Busca, L. Lietti, G. Ramis and F. Berti, Appl. Catal. B: Environ., 1998, 18, 1–36.
4. Z. Q. Tong, Air pollution control engineering, Machinery Industry Press, Beijing, 2006.
5. D. Nicosia, I. Czekaj and O. Krocher, Appl. Catal. B: Environ., 2008, 77, 228–236.
6. K. Asa, C. Andersson, M. Ase, D. Eskilsson and S. G. Jaras, Appl. Catal. B: Environ., 2007, 69, 240–251.
7. S. Roy, M. S. Hegde and G. Madras, Appl. Energy, 2009, 86, 2283–2297.
8. L. Chmielarz, P. Kustrowski, R. Dziembaj, P. Cool and E. F. Vansant, Appl. Catal. B: Environ., 2006, 62, 369–380.
9. C. He, M. Paulus, W. Chu, J. Find, J. A. Nickl and K. Koehler, Catal. Today, 2008, 131, 305–313.
10. S. Bennici, A. Gervasini, N. Ravasio and F. Zaccheria, J. Phys. Chem. B, 2003, 107, 5168–5176.
11. H. He, C. Zhang and Y. Yu, Catal. Today, 2004, 90, 191–197.
12. P. Forzatti, Appl. Catal. A: Gener., 2001, 222, 221–236.
13. A. M. Chen, Y. H. Zhang and C. H. Hu, Chin. J. Ind. Hygiene Occupational Diseases, 2006, 24, 287–288.
14. Z. X. Wang, China Power Enterprise Management, 2009, 10, 16–19.
15. Y. S. Shen, Y. F. Ma and S. M. Zhu, Catal. Sci. Technol., 2012, 2, 589–599.
16. F. Castellino, A. D. Jensen, J. E. Johnsson and R. Fehrmann, Appl. Catal. B: Environ., 2009, 86, 196–205.
17. Y. J. Zheng, A. D. Jensen and J. E. Johnsson, Appl. Catal. B: Environ., 2005, 60, 253–264.
18. C. N. Costa, P. G. Savva, J. L. G. Fierro and A. M. Efstathiou, Appl. Catal. B: Environ., 2007, 75, 147–156.
19. G. G. Olympiou and A. M. Efstathiou, Chem. Eng. J., 2011, 170, 424–432.
20. P. G. Savva and A. M. Efstathiou, J. Catal., 2008, 257, 324–333.
21. C. N. Costa and A. M. Efstathiou, Appl. Catal. B: Environ., 2007, 72, 240–252.
22. C. N. Costa and A. M. Efstathiou, J. Phys. Chem. C, 2007, 111, 3010–3020.
23. J. P. Chen and R. T. Yang, J. Catal., 1990, 125, 411–420.
24. A. C. Larsson, J. Einvall, A. Andersson and M. Sanati, Energy Fuels, 2006, 20, 1398–1405.
25. R. Khodayari, Doctoral dissertation, Department of Chemical Engineering, Lund Institute of Technology, Sweden, 2001.
26. F. Moradi, J. Brandin, M. Sohrabi, M. Faghihi and M. Sanati, Appl. Catal. B: Environ., 2003, 46, 65–76.
27. H. Kamata, K. Takahashi and C. U. I. Odenbrand, J. Mol. Catal. A: Chem., 1999, 139, 189–198.
28. L. Lietti, P. Forzatti, G. Ramis, G. Busca and F. Bregani, Appl. Catal. B: Environ., 1993, 3, 13–35.
29. S. Suarez, S. M. Jung, P. Avila, P. Grange and J. Blanco, Catal. Today, 2002, 75, 331–338.
30. C. B. Zhu, B. S. Jin, Z. P. Zhong, F. Li and J. X. Zhai, Boiler Tech., 2009, 40, 63–67.
31. C. B. Zhu, B. S. Jin, Z. P. Zhong, F. Li, L. X. Chen and J. X. Zhai, J. Southeast University, 2008, 38, 101–105.
32. Y. J. Zheng, A. D. Jensen and J. E. Johnsson, Ind. Eng. Chem. Res., 2004, 43, 941–947.
33. Y. J. Zheng, A. D. Jensen, J. E. Johnsson and J. R. Thøgersen, Appl. Catal. B: Environ., 2008, 83, 186–194.
34. D. Yun, Q. Song and Q. Yao, Coal Conv., 2009, 32, 91–96.
35. S. Onodera and Y. Ikegami, Inorg. Chem., 1980, 19, 615–618.
36. A. J. van Hengstrum, J. Pranger, J. G. van Ommen and P. J. Gellings, Appl. Catal., 1984, 11, 317–330.
37. J. Due-Hansen, A. L. Kustov, S. B. Rasmussen, R. Fehrmann and C. H. Christensen, Appl. Catal. B: Environ., 2006, 66, 161–167.
38. A. L. Kustov, S. B. Rasmussen, R. Fehrmann and P. Simonsen, Appl. Catal. B: Environ., 2007, 76, 9–14.

---