Study of M-ZSM-5 nanocatalysts(M:Cu,Mn,Fe,Co…)for selective catalytic reduction

Parvaneh Nakhostin Panahi*,Darush Salari,Aligholi Niaei*,Seyed Mahdi Mousavi

1 Department of Chemistry,Faculty of Science,University of Zanjan,Zanjan,Iran

2 Department of Applied Chemistry,Faculty of Chemistry,University of Tabriz,Tabriz,Iran

3 Department of Chemical Engineering&Petroleum,University of Tabriz,Tabriz,Iran

4 Faculty of Chemistry,University of Kashan,Kashan,Iran

Keywords:NO SCR ZSM-5 Transition metal Taguchi method

ABSTRACT A series of different transition metals(V,Co,Cr,Mn,Fe,Ni,Cu and Zn)promoted H-ZSM-5 catalysts were prepared by impregnation method and characterized by X-ray diffraction(XRD),scanning electron microscopy(SEM)and transmission electron microscopy(TEM).The catalytic activity of these catalysts was evaluated for the selective catalytic reduction(SCR)of NO with NH3 as reductant in the presence of oxygen.The results revealed that the catalytic activity of Cu-ZSM-5 nanocatalyst for NO conversion to N2 was 80%at 300°C,which was the best among various promoted metals.Design of experiments(DOEs)with Taguchi method was employed to optimize NH3-SCR process parameters such as NH3/NO ratio,O2 concentration,and gas hourly space velocity(GHSV)over Cu-ZSM-5 nanocatalyst at 250 and 300°C.Results showed that the most important parameter in NH3-SCR of NO is O2 concentration;followed by NH3/NO ratio and GHSV has little importance.The NOconversion to N2 of 63.1%and 94.86%was observed at250 °C and 300 °C,respectively under the obtained optimum conditions.

1.Introduction

Nitrogen oxides,NOx(NO+NO2),remain among major sources of air pollution.They contribute to photochemical smog,acid rain,ozone depletion,and greenhouse effects[1].Due to the increasing threat of NOxto our survival,many approaches have been developed to reduce its emission;among them the selective catalytic reduction(SCR)technique was proven to be an effective way compared with other NOxabatement technologies,such as nonselective catalytic reduction technique,storage,and thermaldecomposition[2,3].The mostcommon reductants for SCR are ammonia,urea,CO,H2and hydrocarbons such as methane,ethane and propylene[4].Ammonia is still found to be a suitable reductant for NO in the presence of oxygen,in spite of being dif ficult to handle and also toxicity,because ammonia reacts selectively with NOxto produce N2[5].

In the past few decades,numerous researches have been carried out to develop SCR catalysts such as noble metals[6],transition metals[7],Zeolites[8]and others[9,10].The NOxconversion ef ficiency in the catalytic reactions mostly depends on the support and the nature of active sites.Therefore,the choice of support and metal type critically is important[11].Zeolite-based catalysts have received a greatattention because of their high activity for the reduction of NOx.Among the zeolites,ZSM-5 prompted by noble and transition metals(Co,Pt,Pd,Fe,Ni and Cu)have shown very high performance in SCR of NO[12,13].

The Taguchi method is a design of experiments(DOEs)method dedicated to produce or process performance optimization.This method also proposes an approach to analyze the experimental results in order to indentify the optimal value of each tested parameter.The NH3-SCR of NO process could be affected by NH3/NO ratio,concentration of O2,reaction temperature and gas hourly space velocity(GHSV).Boyano et al.[14]investigated the effects GHSV and NH3/NO ratio on NO reduction and their results showed that the NO conversion can be increased by using high NH3/NO ratio and low GHSV.Based on ourstudies,there has been no published work related to optimize NH3-SCR process parameters over ZSM-5 catalysts.

To achieve the goal of the highest conversion of NO,we first screened a number of different transition metals(V,Co,Cr,Mn,Fe,Ni,Cu and Zn)supported on H-ZSM-5 for comparison of the activity of various metals in SCR of NO with NH3.Then for the most active catalyst(Cu/ZSM-5),the design of experiments with Taguchi method was employed to evaluate the in fluence of process parameters such as NH3/NO ratio,O2concentration,and GHSV in NH3-SCR of NO,and tried to find the optimumvalues ofthese parameters formaximizing NO conversion.

2.Experimental

2.1.Nanocatalysts preparation and characterization

H-ZSM-5 obtained from Zeochem Int.with Si/Al2=50 was used for preparation of the catalysts.All nanocatalysts were prepared through the excess-solution impregnation method.The aim of this procedure was to deposit metal oxide on the zeolite surface,whereas a proton exchange is not explicitly intended and with this technique an accurate load of metal can be adjusted.The precursors of different metals were NH4VO3,Zn(NO3)2·6H2O,Cu(NO3)2·3H2O,Ni(NO3)2·6H2O,Co(NO3)2·6H2O,Cr(NO3)3·9H2O,Fe(NO3)3·9H2O,and Mn(NO3)2·4H2O,and the total metal content was kept at 5%(by mass).First,a desired amount of the metal precursor was dissolved in 30 ml of 50%volume ratio mixture of distilled water and ethanol,and then H-ZSM-5 was added into the solution and stirred at 45°C on a magnetic stirrer until the solvent was totally evaporated.Subsequently,the catalysts were dried in an oven at 100°C overnight,followed by calcination in air for 4 h at 500°C;underthese conditionscomplete decomposition ofmetalnitrate into metal oxide occurs.

The structure of the catalysts was analyzed by powder XRD at room temperature with a D500 Siemens diffractometer using CuKαradiation(λ=0.154050 nm).The X-ray tube was operated at 35 kV and 30 mA and the X-ray pattern was scanned with a step size of crystallites sizes of 0.016°(2θ)from 5°to 50°(2θ)and counting time of 1 s per step.The morphology of the nanocatalysts was observed by scanning electron microscopy(SEM)JEOL JSM-840 and for particle size observation,TEM images were obtained using JEOL 2000 electron microscope operated at 200 kV.

2.2.Catalytic activity tests

The activity of nanocatalysts in NO reduction was studied at atmospheric pressure in a fixed bed reactor.The reactorconsisted ofa tubular glass of 8 mm inner diameter located inside a furnace which is electrically heated.In each run,a measured amount of the prepared powder catalyst was dispread between quartz wool.The reactant gas feed,consisting of NO(1000 mg·L-1),NH3(1000 mg·L-1),O2(5%,by volume)and Ar as balanced was introduced to the reactor at a total flow rate of 200 ml·min-1,yielding a GHSV of 12,000 h-1.Different space velocities were obtained by changing the volume of catalyst bed.Before starting each run,catalysts were pre-treated with 5%O2in Ar for 30 min at 150°C in order to eliminate possible compounds adsorbed on the zeolite surface.After the pretreatment,the reactor was cooled to 100 °C and activity tests were performed from 100 to 400 °C.The concentration of N2(as selective product)and N2O(as nonselective product)in the outlet of the reactor was measured by a gas chromatograph(SHIMADZU model2010 plus)equipped with a thermal conductivity detector(TCD)and a Molecular sieve column to separate N2and N2O.

Fromthe concentration ofthe gases atsteady state,NOconversion to N2was calculated according to the following equation:

The subscripts in and outindicate the inletand outletconcentrations at steady state,respectively.

2.3.Fundamental of Taguchi method

In Taguchimethod,there are three design stages;system,parameter and tolerance designs.In this work,parameter design stage of Taguchi method was adopted.In the parameter design stage,the speci fic values for the system parameters are determined then the parameter design gives the best values for the parameters of the system.The Taguchi method employs standard tables known as orthogonal arrays for the design of experiments.In addition,it uses a special design of orthogonal arrays to study the entire parameter space with a small number of experiments only[15,16].Depending on the number of process parameters and setting levels,a suitable array is selected.Each column of the orthogonal array designates a process parameter and its setting levels in each experiment,and each row designates an experiment with the level of different process parameters in that experiment[16-18].Taguchi analysis was performed with Minitab software package which is a computer program designed to perform basic and advanced statistical functions.

In the present study,three of important parameters of NH3-SCR process were selected for optimization with Taguchi method:NH3/NO ratio,O2concentration,and GHSV and the interaction between the design parameters were neglected.

3.Results and Discussion

3.1.Effect of different metals supported on H-ZSM-5 in NH3-SCR of NO

The activity of H-ZSM-5 and various transition metals supported on H-ZSM-5 in NH3-SCR of NO reaction is presented in Fig.1.The activity of the pure H-ZSM-5 is quite low in the whole temperature range,with NO conversion less than 20%.When the transition metals are introduced,the catalytic activity is signi ficantly enhanced over the whole range of temperature,indicating that the metal species are very important to the SCR activities of the M-ZSM-5 catalysts.All the nanocatalysts were highly selective to N2and only produced trace amounts of by-product N2O at 300 °C and 400 °C(N2O selectivity was about 2%-4%at 300 °C and about 3%-5%at 400 °C for all catalysts).Under identical operating conditions,Cu-ZSM-5 nanocatalyst showed excellent performance giving 80%NO conversion to N2at 300°C,which was the best among all catalysts.High activity of Cu-ZSM-5 in NO reduction was also shown by Sultana et al.[19].Sultana et al.[20]studied SCR of NOxwith NH3over different copper exchanged zeolites and showed that Cu/ZSM-5 and Cu/ERI had an excellent performance for NOxconversion.

3.2.Characterization of catalysts

Fig.2 shows the XRD patterns of the H-ZSM-5 and M-ZSM-5 series(M:Cu,Mn,Fe,Co…).XRD patterns of the all prepared M-ZSM-5 samples are similar to H-ZSM-5 and all characteristic peaks of H-ZSM-5 are observed in M-ZSM-5 samples,which suggest that the original structure of H-ZSM-5 remains unaffected during the process of impregnation and calcination.Also there are no other peaks in XRD patterns of M-ZSM-5 catalysts,demonstrating that metal species(i.e.oxide and cations)were well dispersed through the zeolites structure.

The SEM images of H-ZSM-5 and M-ZSM-5 nanocatalysts are shown in Fig.3.According to SEMgraphs,structure and morphology ofH-ZSM-5 zeolite have not changed during impregnation of metal and calcination process.Fig.4 shows TEM micrographs of H-ZSM-5 and Cu-ZSM-5.The image of Cu-ZSM-5 catalyst clearly shows that the copper nanoparticles were more highly dispersed on the support,which can be caused by the increase of catalytic activity.

3.3.Modi fication of Cu-ZSM-5 nanocatalyst

Fig.1.NO conversion as a function of temperature on M-ZSM-5(M:Fe,Cr,Co…)catalysts(mass fraction of M is 5%)in the NH3-SCR process.

Since Cu-ZSM-5 showed the most activity,the additional studies performed on Cu-ZSM-5.To further improve the performance of the Cu-ZSM-5 nanocatalyst,the composition of the Cu-ZSM-5 was optimized by varying copper loading and calcination temperature.Fig.5(a)compares the NOconversion to N2as a function oftemperature over Cu-ZSM-5 nanocatalysts with differentcopper loadings(1%,3%,5%,7%,9%,by mass).According to Fig.5(a),NO conversion increases with increasing Cu content to 5%(by mass).Further increase in Cu loading declines the NO conversion,therefore Cu loading of 5%(by mass)was found to be the optimum loading.The low activity of Cu-ZSM-5 at loadings of 1%and 3%(by mass)can be attributed to the low quantity of copper species.The decrease of NO conversion at loadings more than 5%(by mass)can be attributed to excessive metal agglomeration,leading to the formation of larger metal particles.Moreover,very further content of copper loading blocks the pores and active sites of catalyst so the catalytic activity decreases[21].

Fig.5(b)shows the NO conversion to N2over Cu-ZSM-5 nanocatalysts(5%,by mass)prepared at different calcination temperatures.According to this figure,it can be concluded that calcination temperature is an important parameter on catalytic activity of Cu-ZSM-5.Different calcination temperatures result in different oxidation states of copper species,so the calcination temperature affects the Cu-ZSM-5 activity.Cu2+species show fast rates of NH3-SCR reactions and are active sites.Generally the catalysts with more facile reduction of Cu2+to Cu+are expected to be more active as this is a key step generating the reaction intermediates,for example,NO oxidation to NO2[22,23].Cu+species are stable and reduce at 500 °C and 770 °C[24].As shown in Fig.5(b),the Cu-ZSM-5 calcinated at550°Chasthe highestactivity(85%NO conversion to N2at 300°C).Very further increase of calcination temperature causes sintering to some extent which led to declined activity[25,26].

Fig.2.XRD of parent H-ZSM-5 and M-ZSM-5(M:Fe,Cr,Cu…).

Fig.3.SEM images of(a):parent H-ZSM-5 and(b):Fe-ZSM-5(c):Mn-ZSM-5(d):Cr-ZSM-5(e):Cu-ZSM-5(f):Co-ZSM-5.

3.4.Process parameters study using design of experiment

The design of experiment with Taguchi method was used to obtain the optimum value of important parameters in the NH3-SCR process of NO over Cu-ZSM-5(5%,by mass,Tc=550°C)at 250 and 300°C.

3.4.1.Plan of experiment

Chosen factors included NH3/NO ratio,concentration of O2,and GHSV and theirlevels are given in Table 1.The mostappropriate orthogonal array de fining the experiment schedule was chosen on the basis of the degrees of freedom of the experiments.The selected orthogonal array was the L9(33)which have 9 rows corresponding to the number of tests with three columns at three levels.The experiment schedule and NO conversion to N2values of experiments at 250 and 300°C are listed in Table 2.

Fig.4.TEM images of(a):H-ZSM-5,(b):Cu/ZSM-5.

The NO conversion to N2as a response for each level of the process parameters was created in an integrated manner as given in Table 3.The effect of process parameters on NO conversion resulting from the optimization process is plotted in Fig.6.This figure plots NO conversion in order to find the optimal levelof process parameters at temperatures of 250 and 300°C.Based on this graph,if each parameter is separately evaluated,the NO conversion takes its local maximum value at the second level for parameter A(5%,by volume,O2concentration)at 250 °C and also at 300 °C.The NO conversion decreases at the third level of O2concentration(10%,by volume).The O2can be needed to activate NO by oxidation to nitrites and/or nitrates,and also may be needed to maintain the proper oxidation state of the copper ions[27].Ham etal.[28]reported thatoxygen plays an importantrole in the reoxidation of cuprous ions to cupric ions in copper exchanged mordenite.Nevertheless,a very high concentration of O2causes NH3oxidation and consequently,NO conversion decreases.For parameter B(NH3/NO ratio),optimum value is 1 at 250 °C and 1.25 at 300 °C.Some NH3is oxidized at 300°C,therefore more NH3is needed for high NO conversion at this temperature.For parameter C(GHSV),the NO conversion increases with the decrease of the GHSV(at 250°C and also at 300°C),but changes are slight.This result suggests that the Cu-ZSM-5 nanocatalyst is highly effective for NO reduction within a wide range of GHSV(from 8000 to 24,000 h-1).The optimum values of the three process parameters(O2volume concentration=5%,NH3/NO ratio=1 at250°C and 1.25 at300 °C,GHSV=8000 h-1)were used to calculate the maximal NO conversion.

Fig.5.NO conversion as a function of temperature on Cu-ZSM-5 with(a)different Cu loadings and(b)different calcination temperatures.

Table 1 The process parameters and their levels

Table 2 Experimental plan of L9(33)orthogonal array and experiment results

Table 3 Response for the Taguchi analysis of NO conversion to N2 data

Fig.6.Effect of SCR process parameters on NO conversion(a)250 °C,(b)300 °C.

3.4.2.Analysis of variance(ANOVA)

The analysis of variance(ANOVA)was used to discuss the relative importance of chosen parameters on NO conversion.ANOVA results are illustrated in Table 4.The F-test was used to determine which process parameters have signi ficant effect on NO conversion.The F-ratio corresponding to the 95%con fidence level of the accurate calculation of process parameters is F0.05,2,8=4.459.F-values of O2concentration and NH3/NOratio parameters are greater of4.459,hence they are significant parameters,but GHSV with the F-value less than 4.459 is not signi ficant parameter.

In addition,the percentage contribution of each parameter which was calculated as the percentage of the ratio of sum of squares obtained from individual parameter to total sum of squares was given in Table 4.The percentage contribution of O2concentration was the greatest,43.15%at 250 °C and 53.47%at 300 °C,the NH3/NO ratio and GHSV being 29.17%and 0.6%at 250 °C and 29.23%and 1.56%at 300 °C,respectively.

3.4.3.Determining the maximum NO conversion and con firmation test

By using the optimallevels of process parameters,the maximum NO conversion was estimated based on the following equations:

4.Conclusions

In this study,the catalytic activity of a series of transition metals(V,Co,Cr,Mn,Fe,Ni,Cu and Zn)promoted H-ZSM-5 catalysts was evaluated for NH3-SCRofNO.Among them,Cu-ZSM-5 nanocatalystshowed the best NO conversion to N2of 80%at 300°C.Taguchi method was applied to optimize SCR process parameters(O2concentration,NH3/NO ratio and GHSV)on Cu-ZSM-5 at 250 and 300°C for maximizing NO conversion.The optimum values of the O2concentration,NH3/NO ratio and GHSV were found to be 5%,(by volume),1 at 250°C and 1.25 at 300°C,and 8000 h-1,respectively.ANOVA results showed that O2concentration and the NH3/NO ratio are signi ficant parameters on SCR of NO.The NO conversion to N2was increased to 63.1%at 250°C and 94.86%at 300°C within obtained optimization conditions and these results were in good agreement with predicted results by Taguchi method.This study clearly showed that Taguchi methodology is a suitable approach to optimize the operating conditions of NH3-SCR to maximize the NO conversion.

Acknowledgments

The authors would like to acknowledge the financial support from University of Tabriz and Iranian Nanotechnology Initiative.

Table 4 Analysis of variance of main effects of parameters

Table 5 Optimum conditions suggested by statistical calculations for maximum NO conversion to N2