Mechanism of the Selective Catalytic Oxidation of Slip Ammonia over

Sep 17, 2014 - ABSTRACT: The slip ammonia from selective catalytic reduction. (SCR) of NOx in coal-fired flue gas can result in deterioration of the u...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/est

Mechanism of the Selective Catalytic Oxidation of Slip Ammonia over Ru-Modified Ce−Zr Complexes Determined by in Situ Diffuse Reflectance Infrared Fourier Transform Spectroscopy Wanmiao Chen, Yongpeng Ma, Zan Qu,* Qinghang Liu, Wenjun Huang, Xiaofang Hu, and Naiqiang Yan* School of Environmental Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China S Supporting Information *

ABSTRACT: The slip ammonia from selective catalytic reduction (SCR) of NOx in coal-fired flue gas can result in deterioration of the utilities or even the environmental issues. To achieve selective catalytic oxidation (SCO) of slip ammonia, Ru-modified Ce−Zr solid solution catalysts were prepared and evaluated under various conditions. It was found that the Ru/ Ce0.6Zr0.4O2(polyvinylpyrrolidone (PVP)) catalyst displayed significant catalytic activity and the slip ammonia was almost completely removed with the coexistence of NOx and SO2. Interestingly, the effect of SO2 on NH3 oxidation was bifacial, and the N2 selectivity of the resulting products was as high as 100% in the presence of SO2 and NH3. The mechanism of the SCO of NH3 over Ru/Ce0.6Zr0.4O2(PVP) was studied using various techniques, and the results showed that NH3 oxidation follows an internal SCR (iSCR) mechanism. The adsorbed ammonia was first activated and reacted with lattice oxygen atoms to form an −HNO intermediate. Then, the −HNO mainly reacted with atomic oxygen from O2 to form NO. Meanwhile, the formed NO interacted with −NH2 to N2 with N2O as the byproduct, but the presence of SO2 can effectively inhibit the production of N2O.



INTRODUCTION

In our previous research, we found that employing Ru as the dopant to modify SCR catalysts can significantly enhance the oxidation of elemental mercury and the mechanism was discussed accordingly.9 However, the selective catalytic oxidation (SCO) of NH3 to N2 over Ru catalysts has rarely been involved. Ru has been used as a catalyst for high concentration NH3 synthesis and decomposition,10 which indicates that Ru has the potential to be a promising active component for NH3−SCO. Recently, ceria(Ce)−zirconia(Zr) solid solutions with cubic fluorite phases (Ce−Zr > 1:1) have served as catalyst carriers in many studies and drawn more attention due to their outstanding oxygen storage capacities and unique redox properties.11 Moreover, Ce−Zr solid solution supported metal oxide catalysts exhibited excellent catalytic activity and N2 selectivity for NO reduction via NH3.12 Although the pathways of NH3−SCO and NH3−SCR are not identical, Ce−Zr solid solutions could be an attractive catalyst support for NH3−SCO. Since the catalytic oxidation activity and mechanism of mercury oxidation over Ru-modified catalysts has already been studied in detail,9 the SCO of NH3 over Ru catalysts was the main focus of this study. Ru/CexZr1−xO2 was used as the

As major air pollutants, nitrogen oxides (NOx), which contribute to photochemical smog, acid rain, ozone depletion, and greenhouse effects, are mainly generated from the combustion of fossil fuels from mobile and stationary pollution sources.1 Selective catalytic reduction (SCR) of NOx with NH3 is a mature technology for reducing NOx emission from coalfired power plants. However, the unreacted ammonia slip from SCR units, especially for aged catalysts, has always drawn wide attention. Because slip ammonia can cause a dust plug in the downstream air-preheater and form secondary inorganic aerosols that contribute significantly to PM2.5 after it is released into the atmosphere, it has been regarded as a more sensitive air pollutant than NOx.2 Therefore, the allowable slip ammonia concentration of SCR units in coal-fired power plants has been tentatively regulated to less than 3 ppm in China.3 Meanwhile, the mercury emitted from coal-fired power plants is also a worldwide concern, especially for China as the largest mercury emitting country.4 An international treaty regarding mercury pollution (the Minamata Convention on Mercury) was officially signed in October, 2013,5 and it is a significant challenge for China to reduce excessive mercury emission because of the high consumption of coal.6,7 Simultaneously removing the slip ammonia and elemental mercury using one multiple-effect catalyst has been proposed for installation at the tail-end of SCR units in our previous research.8 © 2014 American Chemical Society

Received: Revised: Accepted: Published: 12199

May 13, 2014 August 30, 2014 September 17, 2014 September 17, 2014 dx.doi.org/10.1021/es502369f | Environ. Sci. Technol. 2014, 48, 12199−12205

Environmental Science & Technology

Article

Figure 1. (a) The NH3−SCO efficiencies over various catalysts at 350 °C. (b) N2 selectivity of NH3−SCO with or without SO2 over various catalysts at 350 °C. Space velocity (SV) was approximately 3 × 105 h−1. The temperature was 623 K. Thirty ppm of NH3, 30 ppm of NO, and 500 ppm of SO2. The other gas compositions were 4% O2 and N2.

NH3, NO, NO2, and N2O were continually monitored with FTIR Multigas analyzer (IMACC, E-3200-C). The NH3-TPO was performed from 50 to 400 °C with a 1 °C/min ramping rate. The NH3 removal efficiency and N2 selectivity were calculated using eqs 1 and 2:

multiple-effect catalyst, and polyvinylpyrrolidone (PVP) was employed to enhance the dispersion of the low concentration Ru. The NH3 oxidation activity and N2 selectivity over different catalysts at various conditions was tested. Although several studies have examined the NH3−SCO process, the mechanism of NH3 oxidation and N2 formation is not clearly understood. Three major reaction pathways have been proposed for the SCO of NH3 to N2 over different catalysts.13,14 However, the mechanism of NH3−SCO over Ru doping catalysts has not been studied in detail. To rationally develop the process of NH3 oxidation to N2 over Ru-modified catalysts, the reaction mechanism must be clarified. Temperature-programmed oxidation (TPO), temperature-programmed desorption (TPD), and in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFT) were used to determine the possible reaction mechanisms.



NH3 removal efficiency =

[NH3]inlet − [NH3]outlet × 100% [NH3]inlet

(1)

⎡ [NO]outlet + [NO2 ]outlet + 2[N2O]outlet ⎤ ⎥ × 100% N2 selectivity = ⎢1 − [NH3]inlet − [NH3]outlet ⎣ ⎦

(2)

NH3-TPD. The NH3-TPD curves were collected using a chemisorption analyzer (2920, AutoChem II, Micromeritics). Prior to the measurement, the samples were first treated in a He stream at 350 °C to remove any physically adsorbed matter. The NH3 adsorption was then performed at 100 °C with 10% NH3 for 1 h. Next, the samples were purged in a He stream for 30 min to remove physically adsorbed NH3 and then heated to 600 °C with a temperature ramp of 10 °C/min with a 50 sccm He flow. Desorbed NH3 was detected with a thermal conductivity detector (TCD). In Situ DRIFT Studies. In situ DRIFT spectra were recorded on a Fourier transform infrared spectrometer (FTIR, Nicolet 6700) equipped with a smart collector and an MCT detector cooled by liquid N2. The diffuse reflectance FTIR measurements were carried out in a high-temperature cell with KCl windows. Mass flow controllers and a sample temperature controller were used to simulate the real reaction conditions. Prior to each experiment, the catalyst was heated to 400 °C in N2 with a total flow rate of 250 sccm for 3 h. The IR spectra were recorded by accumulating 100 scans at a resolution of 4 cm−1.

EXPERIMENTAL SECTION

Materials and Catalyst Preparation. The Ru doping catalysts were synthesized according to a literature method.15 Polyvinylpyrrolidone (PVP, Mw = 58 000) and quantitative RuCl3·xH2O was dissolved in 30 mL of ethanol. The mixture was refluxed at 100 °C for 3 h. Next, an Ru colloidal solution was added to CexZr1−xO2 (Supporting Information), and the mixture was dried at 60 °C with electromagnetic stirring. Then, the catalysts were calcined at 400 °C for 5 h in air with a 1 °C/ min ramping rate from room temperature and labeled as Ru/ CexZr1−xO2(PVP). The Ru/CexZ1−xO2 catalysts were prepared using the normal wet impregnation method. The Ru oxide content of CexZr1−xO2 was set at 0.2 wt % for all catalysts. All chemicals used for the preparation of catalysts were of analytical grade and were purchased from Sigma-Aldrich Co. or Sinopharm Chemical Reagent Co. The SO2, NH3, and NOx gases were supplied by Dalian Date Gas Co. Catalytic Activity Measurement. The performances of the catalysts for NH3 oxidation were tested in a fixed-bed quartz reactor. The catalyst particles were placed in the reactor with quartz wool under atmospheric pressure and were heated using a vertical electrical furnace. The feed gases consisting of 30 ppm of NOx, 30 ppm of NH3, 4% O2, and 500 ppm of SO2 in N2 were adjusted by mass flow controllers and introduced into the reactor at a total flow rate of 500 sccm. The concentrations of



RESULT AND DISCUSSION Catalytic Activity. The catalytic oxidation efficiencies of the modified Ru/Ce0.6Zr0.4O2 and Ru/Ce0.6Zr0.4O2 (PVP) were tested and evaluated under various conditions as shown in Figure 1 (preliminary experiments showed the Ce0.6Zr0.4O2 supported catalysts displayed superior catalytic activity). The NH3 oxidation over Ce0.6Zr0.4O2 was tested, and the efficiency is only approximately 17%. The NH3 oxidation efficiency over 12200

dx.doi.org/10.1021/es502369f | Environ. Sci. Technol. 2014, 48, 12199−12205

Environmental Science & Technology

Article

NH3−SCO with the coexistence of NOx and inhibited the NH3−SCO in the absence of NOx with a 100% N2 selectivity. NH3-TPD. The profiles of the NH3-TPD analysis of various catalysts are given in Figure S2, Supporting Information. One major desorption peak was observed with the Ce0.6Zr0.4O2 catalyst at 200 °C, which might be attributed to ammonia weakly adsorbed on acid sites including Lewis and Brønsted acids. After doping with Ru, the main peak at 200 °C was much weaker, and a board shoulder starting at 300 °C was observed over Ru/Ce0.6Zr0.4O2, which indicated that the doping of Ru would bate the adsorption ability of NH3. Although less NH3 was adsorbed to Ru/Ce0.6Zr0.4O2 catalyst, it displayed a better NH3 oxidation efficiency because of the excellent catalytic activity of Ru. In addition, the PVP promoted Ru/Ce0.6Zr0.4O2 displayed superior NH3 adsorption capacity than that of Ru/ Ce0.6Zr0.4O2, which resulted in even better NH3 catalytic oxidation activity, as shown in Figure 1. The profile of the sulfated catalyst was very different. Two major NH3 desorption peaks were observed at approximately 350 and 500 °C. Less NH3 desorbed from the sulfated catalyst at lower temperatures (