Effects of NO 2 and SO 2 on selective catalytic reduction of nitrogen oxides by ammonia

The selective catalytic reduction (SCR) characteristics of NO and NO2 over V2O5–WO3–MnO2/TiO2 catalyst using ammonia as a reducing agent have been determined in a fixed-bed reactor at 200–400 °C. The presence of NO2 enhances the SCR activity at lower
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Effects of NO 2  and SO 2  on selective catalytic reductionof nitrogen oxides by ammonia Jeong Hoi Goo  a , Muhammad Faisal Irfan  a , Sang Done Kim  a,* , Sung Chang Hong  b a Department of Chemical and Biomolecular Engineering, Energy and Environment Research Center, Korea Advanced Institute of Scienceand Technology, 305-701 Daejeon, Republic of Korea b Department of Environmental Engineering, Kyonggi University, 442-760 Suwon, Republic of Korea Received 19 July 2006; received in revised form 25 October 2006; accepted 25 October 2006Available online 20 December 2006 Abstract The selective catalytic reduction (SCR) characteristics of NO and NO 2  over V 2 O 5  –WO 3  –MnO 2 /TiO 2  catalyst using ammonia as areducing agent have been determined in a fixed-bed reactor at 200–400   C. The presence of NO 2  enhances the SCR activity at lower tem-peratures and the optimum ratio of NO 2 /NO x  is found to be 0.5. During the SCR reactions, there are some side reactions occurred suchas ammonia oxidation and N 2 O formation. At higher temperatures, the selective catalytic oxidation of ammonia and the nitrous oxideformation compete with the SCR reactions. The denitrification (DeNO x ) conversion decreases at lower temperatures but it increases athigher temperatures with increasing SO 2  concentration. The presence of SO 2  in the feeds inhibits N 2 O formation.   2006 Elsevier Ltd. All rights reserved. Keywords:  SCR; NO x  removal; NH 3  oxidation; N 2 O formation 1. Introduction Emission of nitrogen oxides (NO, NO 2  and N 2 O) is aglobal environmental problems since it causes photochem-ical smog, acid rain, ozone depletion and greenhouse effects(Bosch and Janssen, 1988). In recent years, many tech-niques have been developed to reduce the emission of nitro-gen oxides. The most widely used technology to reducenitrogen oxide emissions from stationary sources is knownto be the selective catalytic reduction (SCR) process (Liettiand Forzatti, 1996) having the main reaction of 4NO +4NH 3  + O 2  = 4N 2  + 6H 2 O.Efficient SCR reaction can be achieved by the conven-tional catalysts such as V 2 O 5 /TiO 2  mixed with WO 3  orMoO 3  with an addition of ammonia as a reducing agent(Janssen et al., 1987; Chen and Yang, 1992; Ramis et al.,1993; Parvulescu et al., 1998). This reaction requires ahigher temperature in the range of 300–400   C to obtainthe desired removal efficiency of NO x  (Bahamonde et al.,1996; Beretta et al., 1998). Therefore, reheating step forexhaust gas passing through a desulfurizier and/or theparticulate control device is needed since it cannot beconducted at lower temperatures below 250   C. In addi-tion, most of the existing power plants do not have anenough space to install a De–NO x  catalyst unit (Ertlet al., 1994) so that highly effective catalysts should bedeveloped for NO x  removal. In recent years, Koebelet al. (2002a) and Madia et al. (2002) reported that the fastSCR reaction with equimolar of NO and NO 2  using thecommercial catalyst exhibits a reaction rate at least tentimes higher than that of the well-known SCR reactionwith pure NO as: 2NO + 2NO 2  + 4NH 3  = 4N 2  + 6H 2 Oat lower temperatures. In this reaction, it is suggested thatNO 2  can reoxidize the reduced catalyst faster than oxygen(Koebel et al., 2002a). But, NO 2  fraction should be below0.5 because the SCR reaction rate becomes lower withexcess NO 2  as of 3NO 2  + 4NH 3  = 3.5N 2  + 6H 2 O.During the SCR reactions, side reactions also takeplace such as ammonia oxidation and ammonium nitrate 0045-6535/$ - see front matter    2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.chemosphere.2006.10.070 * Corresponding author. Tel.: +82 42 8693913; fax: +82 42 8693910. E-mail address: (S.D. Kim). Chemosphere 67 (2007) 718–723  formation. At higher temperatures, the selective catalyticoxidation of ammonia competes with the SCR reactions.By these reactions, ammonia as a reducing agent is con-sumed unnecessarily and also produce other nitrogen oxi-des; NO, NO 2 , N 2 O. Below 180   C, ammonium nitrate isformed that may cause catalyst poisoning (Madia et al.,2002). It is known that SO 2  inhibits the activity of manycatalysts whereas some studies reported that SO 2  promotesthe NO x  removal efficiency since the formation of surfacesulphated species would increase surface acidity (Chenand Yang, 1993; Long et al., 2001). It is known that strongacidity favours ammonia adsorption for the SCR reaction.In the present study, the effect of NO 2  fraction on SCRreaction was determined over V 2 O 5  –WO 3  –MnO 2 /TiO 2  cat-alyst with ammonia. MnO 2  was added as a promoter in thevanadium oxide catalyst since MnO 2  is a well-known oxi-dizing agent so that it will reoxidize the V 4+ state to V 5+ state more easily and consequent increase in the catalyticactivity. Also, the effect of the side reactions by ammoniaoxidation on the SCR performance was determined asfunction of temperature and NO 2  fraction. In addition,the effect of SO 2  on the fast SCR was determined. 2. Experimental The V 2 O 5  –WO 3  –MnO 2 /TiO 2  catalyst was prepared bythe wet impregnation method. The supporting titania (ana-tase; Millennium) was sieved and calcined in air for 4 h at500   C. Subsequently, the sample was impregnated with asolution of ammonium metavanadate (NH 4 VO 3 ; Aldrich)and then dried at 110   C and calcined in air for 4 h at400   C. The properties of the catalyst were determined byX-ray fluorescence (XRF), XRD and Brunauer–Emmett– Teller (BET) methods. The catalyst contains approximately4% V 2 O 5 , 13% WO 3  and 3% MnO 2  on TiO 2  (anatase) as asupporter. Size and BET surface area of the catalyst arerespectively 0.318–0.425 mm and 65 m 2 g  1 .The SCR activities were measured in a fixed-bed quartzreactor (8 mm-ID ·  460 mm-high). The experimental equip-ment consists of three sections: reactor, gas feeding systemand gas analyzer. The catalyst powders were loaded withan aid of quartz wool and heated to the desired reactiontemperature by IR heater with PID controllers. The reac-tion conditions were as follows: 300 ppm NO x  (NO andNO 2 ), 300 ppm NH 3 , 10% O 2 , 8% H 2 O, 50–200 ppm SO 2 ,and balanced N 2  at the total flow rate of 1500 ml min  1 .The gas flow rates were regulated by mass flow controllers(MKS 1179). Water was dosed into the gas stream bypass-ing N 2  and O 2  through a water bottle at 52   C. All the gaslines were heated to 200   C to prevent formation anddeposition of ammonium nitrate and ammonium sulphate.Concentrations of NO, NO 2  (using NO x  converter), SO 2 and O 2  were continuously measured by non-dispersiveinfrared (NDIR) absorption type gas analyzer (Fuji Elec-tric Instruments Co., ZKJ-2) in which the principle of infralight absorption where gas absorbs radiation of a knownwavelength and this absorption is a measure of the gas con-centration. To avoid errors caused by unreacted ammonia,an ammonia trap that containing 4% boric acid solutionwas installed at the upstream of gas analyzer. Concentra-tions of N 2 O and NH 3  were measured by NDIR type gasanalyzer (Hartmann and Braun Co., Urea 10E) and porta-ble gas detector (Gastec, model 801), respectively. 3. Results and discussion 3.1. SCR activity The SCR reactions with three different NO x  feeds (pureNO, equimolar NO and NO 2 , pure NO 2 ) were tested overV 2 O 5  –WO 3  –MnO 2 /TiO 2  catalyst with ammonia as a reduc-ing agent. The results are shown in Fig. 1a where theDe–NO x  conversion is defined as: DeNO x  = {[NO x ] in   [NO x ] out }/[NO x ] in . As can be seen, the De–NO x  conver-sions in all the SCR reactions increase with increasingtemperature upto 300   C, and then decrease slightly, exceptSCR reaction with pure NO which is present in the com-mercial SCR process. The fast SCR exhibits the highestSCR activity over all the temperature ranges studied. Thedifferences in the conversion between the fast and standardSCRs are 22% and 94% at lower temperatures (200   C),respectively, but they do not produce any significant differ-ences at higher temperatures. The SCR reaction with pureNO 2  exhibits higher SCR activity than the standard SCRbelow 250   C, but the reverse is true at higher tempera-tures. Many studies proposed mechanisms of the SCRreaction such as the Eley–Rideal mechanism (Inomataet al., 1980; Miyamoto et al., 1982) in which ammoniafirstly adsorbs on active site of the catalyst and then reactswith gaseous nitrogen oxide, and the reduced catalyst, V 4+ ,is reoxidized to V 5+ by oxygen. In the fast SCR reaction,this reoxidizing role by oxygen is substituted by NO 2  whichis a more effective reducing agent than oxygen. Koebelet al. (2002b) proposed a mechanism for the fast SCR reac-tion similar to the one suggested by Ramis et al. (1990) forthe standard SCR reaction. They reported that NO 2  makesreoxidation of the vanadium sites that is faster than oxygenby the Raman experiments and it makes the reaction rateof the fast SCR higher. Also they confirmed that one mol-ecule of NO is formed for each molecule of NO 2  consumedin the reoxidation of the vanadium sites through the tran-sient experiments.The effect of space velocities (SV = 60000–300000 h  1 )on the De–NO x  conversions in the fast and standardSCR reactions at 200   C and 300   C is shown in Fig. 1bwhere the conversions decrease with increasing SV. AtSV = 60000 h  1 , both the fast and standard SCR reactionsexhibit nearly 100% De–NO x  conversion at 300   C. Withincreasing SV, De–NO x  of the standard SCR decreases sig-nificantly, but that of the fast SCR decreases marginally.At SV = 300000 h  1 , De–NO x  conversions of the standardand fast SCRs are 47% and 95%, respectively. The differ-ence between both SCRs becomes larger at 200   C depend-ing on space velocities. The fast SCR shows more than 90% J.H. Goo et al. / Chemosphere 67 (2007) 718–723  719  of NO x  removal upto SV of 200000 h  1 , but the standardSCR exhibits less than 40%. The performance of the fastSCR is more efficient than that of the standard SCR overall the space velocities employed even at lower tempera-tures. It means that the amount of catalyst and operatingcost for the NO x  removal operation can be reduced consid-erably if the fast SCR process is employed to the exhaustsystem.The effect of NO 2 /NO x  ratio on the De–NO x  conversionis shown in Fig. 1c, where the maximum conversion exhib-its at NO 2 /NO x  = 0.5. The conversion increases even atlower temperatures with increasing NO 2  fraction upto0.5. Thereafter, the conversion decreases with increasingthe ratio of NO 2 /NO x . With the same amount of NOand NO 2 , the fast SCR reaction takes place whereas, theNO 2  fraction below 50%, both fast SCR and standardSCR reactions occur sequentially. First, all the availableNO 2  are consumed by the fast SCR reaction, and thenthe remaining NO are participated in the standard SCRreaction. At NO 2  fraction above 50%, the fast SCR reac-tion with NO and NO 2  also takes place initially and theremaining NO 2  reacts with ammonia.As can be seen in Fig. 1d, the presence of water inhibitsthe activity of both fast and standard SCRs reactions. Inthe presence of water, the De–NO x  conversion in the stan-dard SCR decreases more than 10% at the given tempera-ture ranges because water competes with NH 3  to adsorbthe active site. However, in the fast SCR, the effect of waterpresence is marginal because of its high reaction rate asreported by Madia et al. (2002). Temperature ( ˚ C) 200250300350    D  e  -   N   O      x   c  o  n  v  e  r  s   i  o  n   (   %   ) 020406080100 Standard SCR (pure NO)Fast SCR (50% NO+ 50% NO 2 )Low SCR (pure NO 2 ) Space velocity (x10,000 h -1 ) 051015202530 Standard SCR (T=200 ˚ C)Fast SCR (T=200 ˚ C)Standard SCR (T=300 ˚ C)Fast SCR (T=300 ˚ C) Temperature ( ˚ C) 200250300350 Fast SCR (8% water)Fast SCR (without water)Standard SCR (8% water)Standard SCR (without water) (a) (b)(d) NO 2  /NO x   ratio (-)    D  e  -   N   O      x   c  o  n  v  e  r  s   i  o  n   (   %   ) 020406080100 T=200 ˚ CT=250 ˚ CT=300 ˚ C (c) Fig. 1. De–NO x  conversion as function of (a) reaction temperature in three different SCRs, (b) space velocities, (c) NO 2 /NO x  ratio and (d) water presence/absence. (Total [NO x ] = 300 ppm, [NH 3 ] = 300 ppm, [O 2 ] = 10%.)720  J.H. Goo et al. / Chemosphere 67 (2007) 718–723  The effects of oxygen on De–NO x  conversion in thestandard and fast SCR were also determined since oxygenis needed in the standard SCR to reoxidize the reduced cat-alyst. The results show that the absence of oxygen in thestandard SCR decreases the De–NO x  conversion consider-ably, whereas that in the fast SCR do not affect on the con-version due to reoxidation of NO 2 .The kinetic parameters of the standard and fast SCRsare also determined by the Arrhenius plot. By the well-known equation in a plug flow reactor, the conversioncan be converted into the rate constants  k  r  with an assump-tion of first order reaction of the total NO x  as k  r  ¼  V  W    ln ð 1    DeNO  x Þ where  V   is the gas flow rate (cm 3 s  1 ) and  W   the catalystweight (g). The obtained values of   k  r  were used in theArrhenius plot and then the activation energies of the stan-dard and fast SCRs are derived. As anticipated, the activa-tion energy ( E  a ) of the fast SCR (30.67 kJ mol  1 ) is lowerthan that of the standard SCR reaction (55.21 kJ mol  1 ).The difference in the two reaction rates is more pronouncedat the lower temperatures. 3.2. Side reactions in the SCR reactions During the SCR reactions of NO x , side reactions cantake place in the SCR, but the major product with V 2 O 5  – WO 3  –MnO 2 /TiO 2  catalyst in this experiment is N 2 O. TheN 2 O formation as a function of temperature in three differ-ent SCR reactions is shown in Fig. 2a. As can be seen, N 2 Oformation with pure NO increases with increasing temper-ature since N 2 O can be formed by the direct oxidation of ammonia, 4NH 3  + 4O 2  = 2N 2 O + 6H 2 O, and by thereaction between NH 3  and NO, 4NH 3  + 4NO + 3O 2  =4N 2 O + 6H 2 O, at higher temperatures. In both reactionsthe major source of N 2 O formation is ammonia oxidation,which causes unnecessarily ammonia consumption. In gen-eral, SCR activity with pure NO increases with increasingtemperature but the reaction of ammonia oxidation is alsopronounced at higher temperatures. The competitive reac-tion between SCR and ammonia oxidation may decreasethe activity and selectivity of the reaction. With the feedcontaining pure NO 2 , the N 2 O formation exhibits a maxi-mum value at a temperature around 250   C. Thereafter, itdecreases with temperature but the N 2 O formation is muchhigher than that in the standard and fast SCRs. Madiaet al. (2002) reported that N 2 O in the SCR with pureNO 2  can be formed by following each other reactionpathways, NH 4 NO 3  = N 2 O + 2H 2 O by reaction tempera-tures below 250   C and 6NH 3  + 8NO 2  = 7N 2 O + 9H 2 Oand 4NH 3  + 4NO 2  + O 2  = 4N 2 O + 6H 2 O reactions above250   C. By Blanco et al. (2000), the N 2 O formation overV 2 O 5 /TiO 2  catalyst with pure NO 2  and NH 3  is activelytakes place at the temperature range of 180–230   C. Withthe feed containing equimolar of NO and NO 2 , the amountof N 2 O formation is comparatively smaller than those inother cases studied and is not affected at the temperaturerange of 200–350   C. This may indicate that the fast SCRreaction has better selectivity to the desired reaction thanthe other side reactions.The effect of NO 2 /NO x  ratio on the nitrous oxide forma-tion at different temperature is shown in Fig. 2b where theminimum N 2 O formation occurs at NO 2 /NO x  = 0.5. Ascan be seen, NO 2  fraction less than 0.5 in the feeds inhibitsN 2 O formation significantly but NO 2  fraction higher than0.5 accelerates N 2 O formation. The NO 2 /NO x  ratio below0.5, the amount of N 2 O formed is constant at 200   C. Therate of N 2 O formed at the ratio of NO 2 /NO x  = 0.5 is thefastest at 250   C. The N 2 O formation at NO 2  fraction Temperature (˚C) 200250300350    N    2    O   f  o  r  m  a   t   i  o  n   (  p  p  m   ) 0510152025 Pure NOEquimolar NO and NO 2 Pure NO 2 NO 2  /NO x   ratio T = 200 ˚ CT = 250 ˚ CT = 300 ˚ CT = 350 ˚ C (a) (b) Fig. 2. N 2 O formation as a function of (a) reaction temperature in three different SCRs and (b) NO 2 /NO x  ratio. (SV = 100000 h  1 , total[NO x ] = 300 ppm, [NH 3 ] = 300 ppm, [O 2 ] = 10%, [H 2 O] = 8%.) J.H. Goo et al. / Chemosphere 67 (2007) 718–723  721  below 0.5 is caused by the direct oxidation of ammonia andthe reaction between NH 3  and NO. With increasing NO 2 /NO x  ratio, the amount of N 2 O formed becomes lessbecause the feeds (NO x , NH 3 ) follow the fast SCR reactionthat has much more selectivity than the other reactions. Onthe other hand, N 2 O formed at NO 2  fraction higher than0.5 increases due to the decomposition of ammoniumnitrate and the reaction between NH 3  and NO x .The effect of oxygen on N 2 O formation was determinedin the standard and fast SCR reactions. In the standardSCR, the amount of N 2 O increases upto 1% O 2  and lev-els-off upto 15% O 2 . On the other hand, the N 2 O formationin the fast SCR is not affected by O 2  in the range of 0–15%.In the standard SCR, oxygen is needed to reoxidize thereduced catalyst but oxygen causes the side reactions byammonia oxidation. In the fast SCR, NO 2  is more favour-able than oxygen to reoxidize the reduced catalyst, so itmay inhibit ammonia oxidation. 3.3. Effect of SO  2  on the fast SCR reaction The effect of SO 2  concentration on the DeNO x  conver-sion in the fast SCR reaction is shown in Fig. 3 whereDe–NO x  conversion decreases slightly (2%–3%) at lowertemperatures, but it remains nearly constant at higher tem-peratures with increasing SO 2  concentration. In the stan-dard SCR reaction, the De–NO x  conversion decreasesapproximately 10% by SO 2  at lower temperatures but itincreases with SO 2  at higher temperatures as reported byAmiridis et al. (1996) who tested the SCR reaction at tem-perature above 350   C over V 2 O 5 /TiO 2  catalyst. Sulphurdioxide is chemisorbed on the active site of metal oxideand then forms instable sulphite ion, which react with thechemisorbed oxygen to form sulphate. The formationand stability of sulphate on the transition metal oxidedepends on the reaction temperature. Also, Chen and Yang(1990) reported that the sulphated TiO 2  has a solid super-acid property that may increase the catalyst surface acidity. 4. Conclusions The effects of NO 2  and SO 2  on selective catalytic reduc-tion with ammonia over V 2 O 5  –WO 3  –MnO 2 /TiO 2  catalystwere determined. The presence of NO 2  in NO x  promotesperformance of the SCR system at lower temperatures,and the optimum ratio of NO 2 / NO x  for the SCR reactionis found to be 0.5. The N 2 O formation in the fast SCR islower than that in the standard SCR. The amount of N 2 O formation in the standard SCR increases with increas-ing reaction temperature, but in the fast SCR it remainsconstant. With increasing the ratio of NO 2 /NO x  up to0.5, the amount of N 2 O formation decreases but N 2 O for-mation increases with increasing the NO 2 / NO x  ratio above0.5. The presence of SO 2  inhibits SCR reactions at lowertemperatures, otherwise improves at higher temperatures.SO 2  inhibits N 2 O formation at all the reaction tempera-tures employed. Acknowledgements The authors acknowledge the financial support from theKorea Energy Management Corporation (KEMCO) andthe Korea Power Engineering Company (KOPEC). Also,this work is supported by the Brain Korea 21 project. References Amiridis, M.D., Wachs, I.E., Deo, G., Jehng, J.M., Kim, D.S., 1996.Reactivity of V 2 O 5  catalysts for the selective catalytic reduction of NOby NH 3 : influence of vanadia loading, H 2 O, and SO 2 . J. Catal. 161,247–253.Bahamonde, A., Beretta, A., Avila, P., Troconi, E., 1996. An experimentaland theoretical investigation of the behavior of a monolithic Ti–V–W– sepiolite catalyst in the reduction of NO x  with NH 3 . Ind. Eng. Chem.Res. 35, 2516–2521.Beretta, A., Orsenigo, C., Ferlazzo, N., Troconi, E., Forzatti, P., 1998.Analysis of the performance of plate-type monolithic catalysts forselective catalytic reduction DeNO ( x )  applications. Ind. Eng. Chem.Res. 37, 2623–2633.Blanco, J., Avila, P., Suarez, S., Martin, J.A., Knapp, C., 2000. Alumina-and titania-based monolithic catalysts for low temperature selectivecatalytic reduction of nitrogen oxides. Appl. Catal. B Environ. 28,235–244.Bosch, H., Janssen, F., 1988. Formation and control of nitrogen oxides.Catal. Today 2, 369–379.Chen, J.P., Yang, R.T., 1990. Mechanism of poisoning of the V 2 O 5 /TiO 2 catalyst for the reduction of NO by NH 3 . J. Catal. 125, 411–420.Chen, J.P., Yang, R.T., 1992. Role of WO 3  in mixed V 2 O 5  –WO 3 /TiO 2 catalysts for selective catalytic reduction of nitric oxide with ammonia.Appl. Catal. A Gen. 80, 135–148.Chen, J.P., Yang, R.T., 1993. Selective catalytic reduction of NO withNH 3  on SO  24  = TiO 2  superacid catalyst. J. Catal. 139, 277–288.Ertl, G., Kno¨zinger, H., Weitkamp, J., 1994. Handbook of HeterogeneousCatalysis. Wiley-VCH, Weinheim. SO 2  concentration (ppm) 050100150200    D  e  -   N   O      x   c  o  n  v  e  r  s   i  o  n   (   %   ) 020406080100    N    2    O   f  o  r  m  a   t   i  o  n   (  p  p  m   ) 01020304050 T = 200 ˚ CT = 250 ˚ CT = 300 ˚ CT = 350 ˚ CN 2 O formation (T = 250 ˚ C) Fig. 3. Effect of SO 2  concentration on the De–NO x  conversion in the fastSCR reaction. (SV = 100000 h  1 , total [NO x ] = 300 ppm, [NH 3 ] = 300ppm, [O 2 ] = 10%, [H 2 O] = 8%.)722  J.H. Goo et al. / Chemosphere 67 (2007) 718–723
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks