Elsevier

Applied Catalysis B: Environmental

Volume 181, February 2016, Pages 692-698
Applied Catalysis B: Environmental

Comparison of MoO3 and WO3 on arsenic poisoning V2O5/TiO2 catalyst: DRIFTS and DFT study

https://doi.org/10.1016/j.apcatb.2015.08.030Get rights and content

Highlights

  • Arsenic decrease number of Lewis and strength of Brønsted acid sites.

  • NH3 adsorbed on As–OH sites shows low activity.

  • Improvement of MoO3 is due to the good dispersion.

  • Mo–O–Mo and Mo = O groups provide higher reactivity than WO3 does.

Abstract

The mechanisms of arsenic poisoning on MoO3- and WO3-doped V2O5/TiO2 catalysts are studied and the poisoning resistance effect of MoO3 are revealed. The arsenic at certain amount (above 10% of arsenic from XPS) decreases the surface area and the number of active sites where reduction takes place. From the results of DRIFTS and DFT, arsenic lowers the surface acidity by decreasing the quantity of Lewis acid sites and the stability of Brønsted acid sites. We also found newly formed As–OH groups of poisoned catalysts (at 1440 cm−1) have quite low activity at high temperature.

MoO3 provides better activity on poisoned catalysts than WO3 does on V2O5/TiO2 catalysts. This may be due to the good dispersion of MoO3 on the support. The reactivity of surface W–O–W and W = O groups is considerably lower than the corresponding Mo–O–Mo and Mo = O groups from the DFT calculations models. This could be one of the explanations on the decrease of surface acidity of poisoned catalysts.

Introduction

NOx emitted from coal-fired power plants and industrial boilers is the main air pollutants, which produce particulate matter and has a serious environmental and human health impacts. Accordingly, the abatement of NOx is one of the key steps for improving air quality. Selective catalytic reduction (SCR) of NOx with NH3 is an effective method. NH3 (or urea solutions) is used in the presence of excess O2, and reacts with NOx to form H2O and N2 [1], [2]. Commercial SCR catalysts are V2O5–WO3(MoO3)/TiO2, where V2O5 is active sites with uniform monolayer dispersion. The addition of WO3 or MoO3 increases the catalytic activity and thermal durability by stabilizing TiO2 preventing a phase change from anatase to rutile. The TiO2 anatase provides good resistance to SO2 fouling and sufficiently large surface area for high dispersion of V2O5 [3], [4]. Nevertheless, there are still several problems on catalysts, for example toxicity of V2O5 to human, high conversions of SO2 to SO3 and deactivation by alkali metals or arsenic in the flue gas. These poisons not only decrease SCR performance, shorten catalysts working lifetime, but also plug the module or monolith pores of catalytic convertor, even corrode the pipeline and increase the backpressure [5], [6], [7], [8], [9].

Arsenic (As) is a serious poison for SCR catalysts, and can be found as As2O3 with concentrations from 1 μg/m3 to 10 mg/m3 [10]. Previous studies reported that arsenic had a valence of +5 and was bonded to both V2O5 and TiO2 [11]. Our recent works on potassium [12], [13], [14] and arsenic [15], [16] poisoning show that K2O decreases both Brønsted acid sites and reducibility of catalysts by tightly bonding to active sites, whereas arsenic decreases Lewis acid sites and slightly improves reducibility by forming unstable arsenic hydroxyls (As–OH). In addition, nearly half of arsenic could be effectively removed by 4% H2O2 solution, which could be a possible method of catalysts regeneration [15].

Compared with WO3, MoO3 has a promotion effect on arsenic poisoned V2O5/TiO2, but the specific mechanism of this inhibition is still unclear [17], [18]. Some researchers propose that the structural and morphological characteristics of MoO3 and WO3 on TiO2 are similar [19], yet, two questions still need to be answered: First, why do we find more N2O formed on MoO3/TiO2 than WO3/TiO2? Secondly, whether the surface species of poisoned catalysts are the same or similar as fresh catalysts [20]? Therefore, in order to design arsenic poisoning resistance catalysts, it is necessary to elucidate the different poisoning behaviors of WO3 and MoO3 on V2O5/TiO2. Moreover, from theoretical point view, many published works have successfully constructed the V2O5 cluster or slab models supported on low index planes of TiO2 [21], [22], [23]. Some studies have investigated poisoning mechanisms of alkali metals and lead on active sites [24], [25]. However, few works have been focused on arsenic [26].

In this work, V2O5–MoO3/TiO2 and V2O5–WO3/TiO2 catalysts are prepared to investigate the poisoning mechanism of arsenic by DRIFTS and density functional theory (DFT) methods. We try to clarify the relationship between surface species and catalytic activity, and the universal DFT models for SCR are also provided.

Section snippets

Catalyst preparation and poisoning

The catalysts are prepared by wet impregnation methods. The content of V2O5 is 1 wt.%. MoO3 loadings are 3 and 1.9 wt.% and WO3 loading is 3 wt.%, which is the same molar ratio as that with 1.9 wt.% of MoO3. TiO2 anatase is used as support. Samples are first dried and calcined at 500 °C for 5 h, and then sieved within 40–60 meshes for activity measurement and with more than 60 meshes for physi–chemical characterizations. They are denoted as fresh samples: F3%Mo, F3%W and F1.9%Mo. The corresponding

SCR performance

Fig. 1 shows the NOx conversion and N2O formation of fresh and poisoned catalysts for a GHSV of 120,000 mL/g h. Fresh catalysts reach 100% NOx conversion above 350 °C, and the activity order at relatively low temperatures (250 °C) is: F3%Mo > F1.9%Mo > F3%W. Poisoned catalysts exhibit lower activities than the corresponding fresh catalysts, nevertheless, their activity order is not changed: 60% and 55% NOx conversion is obtained for P3%Mo and P1.9%Mo at 450 °C, respectively, while it is only 40% for

Conclusion

Based on the experiment and DFT results, two major issues are attempted to propose.

Arsenic at certain loading (1) reduces the surface area and the reducibility of catalysts active sites; (2) the number of Lewis acid sites and the strength of Brønsted acid sites also decrease; (3) new formed As–OH sites only show weak activity above 350 °C. These factors lead to catalysts deactivation.

MoO3 provides better arsenic poisoning resistance than WO3 on V2O5/TiO2 catalysts. This could be attributed to

Acknowledgements

The authors gratefully acknowledge the financial supports from National Natural Science Foundation of China (21407088), and National High-Tech Research and the Development (863) Program of China (2013AA065401 and 2013AA065304, the outstanding young scientist funding (21325731) and the international Postdoctoral Exchange Fellowship Program of China (20130032). The authors also appreciate the support from the Brook Byers Institute for Sustainable Systems (BBISS), Hightower Chair and Georgia

References (44)

  • G. Ramis et al.

    Appl. Catal.

    (1990)
  • G. Ramis et al.

    Catal. Today

    (1996)
  • G. Busca et al.

    Appl. Catal. B

    (1998)
  • L. Lietti et al.

    Appl. Catal. B

    (1993)
  • R. Khodayari et al.

    Appl. Catal. B

    (2001)
  • L. Lisi et al.

    Appl. Catal. B

    (2004)
  • C. Senior et al.

    Combust. Flame

    (2006)
  • E. Hums

    Catal. Today

    (1998)
  • Y. Peng et al.

    Appl. Catal. B

    (2015)
  • F. Lange et al.

    Appl. Catal. B

    (1996)
  • Z. Liu et al.

    Appl. Catal. B

    (2014)
  • P. Hejduk et al.

    J. Mol. Catal. A

    (2010)
  • X. Du et al.

    Appl. Surf. Sci.

    (2013)
  • Z. Wei et al.

    Appl. Surf. Sci.

    (2011)
  • C. Wang et al.

    Chem. Eng. J.

    (2013)
  • C. Sun et al.

    J. Mol. Catal. A

    (2011)
  • L. Wang et al.

    J. Catal.

    (2012)
  • Y. Peng et al.

    Appl. Catal. B

    (2013)
  • F. Liu et al.

    Catal. Today

    (2010)
  • F. Liu et al.

    Appl. Catal. B

    (2009)
  • N. Topsoe

    Science

    (1994)
  • W. Kijlstra et al.

    Appl. Catal. B

    (1998)
  • Cited by (123)

    View all citing articles on Scopus
    View full text