Insight into SO2 poisoning and regeneration of one-pot synthesized Cu-SSZ-13 catalyst for selective reduction of NOx by NH3

Xin Yong,Hong Chen*,Huawang Zhao,Miao Wei,Yingnan Zhao,Yongdan Li,,*

1 School of Environmental Engineering,Tianjin University,Tianjin 300072,China

2 Collaborative Innovation Center of Chemical Science and Engineering (Tianjin),Tianjin Key Laboratory of Applied Catalysis Science and Technology,State Key Laboratory of Chemical Engineering (Tianjin University),School of Chemical Engineering,Tianjin University,Tianjin 300072,China

3 Department of Chemical and Metallurgical Engineering,School of Chemical Engineering,Aalto University,Espoo 02150,Finland

Keywords: Emission control SCR One-pot synthesis of Cu-SSZ-13 SO2 poisoning Activity regeneration

ABSTRACT The effects of SO2 on an one-pot synthesized Cu-SSZ-13 catalyst for selective reduction of NOx by NH3 were examined.The addition of SO2 inhibited NOx conversion significantly below 300°C,while no effect on NOx conversion was observed above 300 °C.TGA,TPD,and XPS results showed that the deactivation was caused by the formation of(NH4)2SO4,SO2 chemisorption on the isolated Cu2+ion sites,as well as the formation of CuSO4-like species.Among them,the site-blocking effect of (NH4)2SO4 on Cu2+was the primary reason for deactivation.Fortunately,89% of deNOx activity of the poisoned catalyst was recovered after thermal treatment at 500°C in air,where(NH4)2SO4 was completely decomposed.The activity was further recovered with regeneration temperature increasing to 600°C,removing the adsorbed SO2 on the Cu2+sites.The regeneration at 600 °C could not recover the activity completely,because of the high stability of CuSO4-like species.

1.Introduction

NOxare the main air pollutants and selective catalytic reduction with NH3(NH3-SCR)has been commercially applied as an efficient technique to eliminate NOxemission from mobile sources [1–6].Cu-SSZ-13 catalyst was developed and commercialized in the past decade,which showed excellent deNOxactivity,superior selectivity to N2,and high hydrothermal stability[1,2,7–10].Nevertheless,Cu-SSZ-13 catalyst is highly sensitive to the SO2,which will affect the low temperature NH3-SCR activity of the catalyst and limit its applications.Therefore,the effect of sulfur is still a worthwile topic for a catalyst in NH3-SCR reaction.

The sulfur poisoning (including SO2and SO3) has been examined with Cu/zeolite [11–35].For sulfur poisoning on Cu-SSZ-13,SO2mainly inhibits the SCR activity at low-temperatures [18],while recently we proved that SO2even has positive effects if present during hydrothermal treatment at 800 °C [35].At lowtemperatures,two sulfur-containing species,ammonium-sulfate and Cu-sulfate species,are formed upon sulfur exposure,causing the deactivation of Cu-SSZ-13 [15,16].

Fortunately,sulfur poisoning of Cu-SSZ-13 is largely reversible with a high temperature treatment in lean conditions [15,26].However,the regeneration procedure of the poisoned sample should adapt to the sulfation condition,e.g.,the reactant gas composition [15,18],the reaction temperature [13,15],as well as the properties of the catalyst (including Cu loading,Cu2+ions type and position,as well as Si/Al ratio,etc.) [16],because those factors affect the nature of the formed sulfur intermediates.Jangjouet al.[16]proposed that two types of active sites Z2Cu(Cu2+)and ZCuOH(Cu(OH)+) coexisted in Cu-SSZ-13 and SO2only adsorbed on Z2Cu sites when it was cofed with NH3,viaformation of ammonium sulfate,while,reactation of SO2and ZCuOH would form copper bisulfite species at low temperature,which was further oxidized to form copper bisulfate with the increase of temperature.

One-pot synthesis,using copper-tetraethylenepentamine (Cu-TEPA)as a template,and ion-exchange post zeolite synthesis were the two most common techniques to prepare Cu-SSZ-13 with highly dispersed isolated Cu2+ions [36–38].Interestingly,the Cu-SSZ-13 catalysts prepared with those two routes show different intrinsic properties.The Cu-SSZ-13 synthesized with one-pot process typically shown lower hydrothermal stability than the samples prepared with the ion-exchange route [39].Nevertheless,the one-pot Cu-SSZ-13 exhibits higher activity and selectivity for NH3oxidation to N2,while the Cu ion-exchanged catalyst shown low NH3oxidation activity even with high Cu loading [40–42].The difference in the stability and activity was probably due to the use of Cu-TEPA in the one-pot synthesis,serving as both the Cu2+source and the template[36,37].Cu-TEPA influenced the type and location of the Cu2+ions,as well as the distribution of Si and Al in the framework of the Cu-CHA zeolite [37,43].

Regarding SO2poisoning on Cu-SSZ-13,the previous works focused on the influence of SO2on ion-exchange prepared samples[12,15,16,18,19,21].For the Cu-SSZ-13 preparedviaone-pot route,only the effect of SO2exposure on the hydrothermal stability of Cu-SSZ-13 had been reported previously [13].However,the effect of SO2poisoning on the one-pot Cu-SSZ-13 at low-temperature is still not investigated in detail.On the other hand,in the previous works regarding SO2poisoning on Cu-SSZ-13 [12,15,16,18],around 5% H2O was added in the feed to mimic the real-world exhaust gas composition.However,the SCR reactions over Cu-SSZ-13 also produce H2O,as shown in Eqs.(1) and (2).

The standard SCR:

The fast SCR:

In this work,we investigated the nature of SO2poisoning over the one-pot prepared Cu-SSZ-13 catalyst without adding H2O in the feed.

2.Experimental

2.1.Chemicals

Copper(II) sulfate (CuSO4>99% (mass)) was purchased from Guangfu (Tianjin).Tetraethylenepentamine (TEPA >98% (mass)),sodium hydroxide (NaOH >98% (mass)) and sodium aluminate(NaAlO2>98% (mass))were supplied by Aladdin Industrial Corporation(Shanghai).Ammonium nitrate(NH4NO3>99% (mass)) was ordered from Yuanli (Tianjin).Colloidal silica (JN-30,SiO2=30% (mass)) was from Haiyang (Qingdao).All the aqueous solutions were prepared using ultra-pure water bought from Yongqingyuan(Tianjin).Reactant mixed gas(2% NO,2% NH3,1% SO2)was bottled by Air Liquide.High-purity N2and O2were obtained from Liufang(Tianjin).

2.2.Synthesis of catalyst

The Cu-SSZ-13 sample was synthesized with a one-pot hydrothermal method using Cu-TEPA as a template [36,37].The gel was prepared with a mole ratio of 2.5 Na2O:1 Al2O3:10 SiO2:150 H2O:2 Cu-TEPA,then transferred into an autoclave and crystallized at 140 °C for four days.The Si/Al ratio of the synthesized sample is around 4 according to ICP-OES [44].The solid product was washed with water and dried at 100 °C overnight.15 g of the dried powder was immersed in 300 ml NH4NO3solution(1 mol?L-1) at 80 °C and kept for 6 h for ion exchange.The sample was then filtrated,washed and dried at 100 °C,followed by calcination at 600 °C in synthetic air for 6 h.

2.3.Ammonia selective catalytic reduction

2.3.1.Activity test

The NH3-SCR tests were carried out in a fixed-bed quartz tube reactor (8 mm inner diameter).Two K-type thermocouples were located upstream and downstream of the catalyst bed to monitor and control the temperature.300 mg catalyst sample (250–420 μm) was diluted with 100 mg silica gel in the same size,before loading into the tube reactor.The feeding gas is a synthetic gas mixture containing 0.05% (vol) NO,0.05% (vol) NH3,5% (vol) O2,0.01% or 0.02% (vol)SO2,and balanced with N2.The total flow rate was 500 ml?min-1,yielding a gas hourly space velocity (GHSV) of 50,000 h-1.

Before the test,the fresh Cu-SSZ-13 sample was first activated in 5% (vol)O2/N2flow at 550°C for 1 h.During the SCR activity test with SO2exposure,a waiting time of 60 min was applied at every 50°C from 100°C to 550°C,before collecting the products for analysis.The gas content(NH3,NO,NO2and N2O)was recorded by FTIR(Thermo Nicolet iS10) online with a 2 m gas cell.The conversions of NOx(NOx=NO+NO2) and NH3were calculated respectively with the following equations:

2.3.2.900 min activity test with poisoning

The 900 min SCR activity test was carried out in the same fixedbed reactor mentioned above using 300 mg (250–420 μm) of Cu-SSZ-13 sample.At the target temperature (200 °C or 300 °C),the catalyst was first stabilized in the gas mixture with 0.05% (vol)NO,0.05% (vol) NH3,5% (vol) O2and N2balance for 60 min.Then,0.01% (vol) of SO2was switched into the reaction stream for 840 min.In comparison,the activity of fresh Cu-SSZ-13 catalyst was also measured at 200°C under the same condition but without SO2,keeping 100% NOxconversion in 900 min.

2.3.3.The regeneration of the catalyst

After 900 min SO2poisoning test as described in Section 2.3.2,the poisoned samples were regenerated in air with a flow rate of 500 ml?min-1at 500 or 600 °C for 2 h.After the regeneration,50 mg (250–420 μm) of catalyst mixed with 350 mg (250–420 μm) of silica gel was exposed to the condition of 0.05% (vol) NO,0.05% (vol) NH3,5% (vol) O2,5% (vol) H2O,balanced with N2.The fresh and S poisoning samples also were remeasured under the same conditions.The corresponding GHSV is 300,000 h-1.The concentration of the products was also measured with the FTIR when the reaction reached a steady state (the NO,NO2,N2O,and NH3concentration change less than 1×10-6(vol) in 10 min).Within the context of this article,the sample notations will be used as shown in Table 1.

Table 1 Sample notations and details of preparation procedure

Table 2 Experimental conditions used in transition experiments

2.3.4.Transient experiments

Transient reactions were carried out stepwise in prescribed atmospheres at 200 °C with 150 mg catalyst.Table 2 summarizes the experimental conditions.The effluent gases were analyzed with FTIR.The fresh Cu-SSZ-13 sample was activated in 5% (vol)of O2/N2flow for 1 h at 550 °C before the test.

2.4.Catalyst characterization

N2physisorption at-196°C was used to determine the specific surface area and pore volume of the samples on Quantachrome Autosorb-1,with pre-degassing at 200 °C for 8 h.The specific surface area was calculated using the BET equation,and the pore volume was estimated using thet-plot method.Thermo-gravimetric analysis (TGA) was performed on NETZSCH STA 449 F3 Jupiter(Germany) in a flow of air (50 ml?min-1),and the heating rate was 10 °C?min-1from 50 to 950 °C.All the samples were pretreated in a flow of air (50 ml?min-1) at 50 °C for 1 h before the tests.The chemical composition of the samples was determined by inductively coupled plasma optical emission spectrometry(ICP-OES)with a VISTA-MPX apparatus from Varian.Scanning electron microscopy(SEM) and elemental mapping images of samples were recorded on a Regulus 8100 field emission scanning electron microscopy.The samples were fixed onto a sample holder using carbon tape and then spray with Au nanolayer to make them conductive.The electron paramagnetic resonance (EPR) spectra of the samples were acquired on a Bruker A300 instrument with 50 mg of sample loaded in a quartz tube.The spectra were recorded at-150°C with the magnetic field swept from 2000 to 4000 G.Bruker BioSpin WinEPR software was used for the data analysis.The X-ray photoelectron spectroscopy (XPS) was measured with a PHI-1600 ESCA SYSTEM spectrometer using Mg Kαas X-ray source(1253.6 eV) under a residual pressure of 5×10-6Pa.The error of the binding energy was ±0.2 eV using C 1s at 284.6 eV as the standard.The temperature-programmed desorption (TPD) experiments were done with 150 mg of the poisoned sample with a heating rate of 10 °C?min-1to the final temperature of 600 °C,the concentrations of the outlet gases (NO,NO2,N2O,SO2,and NH3)were measured with FTIR.

3.Results

3.1.Deactivation by SO2

Fig.1 illustrates the NOxconversion with or without SO2cofeeding with the fresh Cu-SSZ-13 sample in 100–550 °C.The Cu-SSZ-13 sample exhibited high activity,with over 90% NOxconversion in 175–500 °C.However,the deNOxactivity of Cu-SSZ-13 decreased dramatically with 0.01% (vol) SO2in the feed below 300 °C,with only 35% and 85% NOxconversion remaining at 200°C and 250 °C,respectively.The catalytic activity showed a slight further decrease with the SO2concentration increase to 0.02% (vol) below 300 °C.In contrast,no apparent negative effect was observed above 300 °C with 0.01% or 0.02% (vol) SO2in the feed,but positive effect was recorded in the temperature range >450°C.The above results indicate that the SO2poisoning mainly inhibit the deNOxactivity of one-pot Cu-SSZ-13 below 300°C,and the difference of 0.01% or 0.02% (vol) SO2in the feed only makes little chage of the SCR performance.In order to distinguish the catalytic activity of catalysts clearly,the higher space velocity(300,000 h-1:50 mg catalyst sample (250–420 μm) and 350 mg silica) were applied.The results were showed in Fig.S1.When the GHSV increased from 50,000 h-1to 300,000 h-1,the deNOxactivity of the fresh Cu-SSZ-13 catalyst showed a decreased by 8% ,62% and 12% at 100,150,and 200°C,respectively.While the deNOxactivity increased above 500 °C compared to at the space velocity of 50,000 h-1.The catalytic activity showed a apparent negative effect when the temperature was under 350 °C with 0.01% or 0.02% (vol) SO2in the feed at 300,000 h-1.And a decrease by 20% with the addition of 0.01% or 0.02% (vol) SO2at 300 °C was observed.

Fig.1.NOx conversion curves of the fresh and SO2 poisoned Cu-SSZ-13 samples as a function of reaction temperature.Reaction conditions:0.05 % (vol) NO,0.05 % (vol)NH3,0.01% or 0.02% (vol) SO2 (when used),5 % (vol) O2 and balance N2;flow rate:500 ml?min-1;GHSV:50,000 h-1.

To further test the stability of Cu-SSZ-13 during SO2exposure,the NOxconversion was measured with 0.01 % (vol) SO2cofeeding for 900 min,at 200°C and 300°C,respectively.The results are presented in Fig.2.Initially,100% NOxconversion was achieved in the standard SCR condition for 60 min without SO2poisoning,as shown in Fig.2.Subsequently,0.01 % (vol) SO2was added into the reactant gases.At 200 °C,the NOxconversion decreased rapidly in 120 min after the SO2added in the feed,from 100% to around 30% ,confirming the significant poisoning of the catalyst by SO2at 200°C.From 180 to 900 min,only minor NOxconversion decline was observed,decreased from 30% (180 min) to 20% (900 min).

Fig.2.NOx and NH3 conversion curves at 200 and 300°C over Cu-SSZ-13.Reaction conditions:0.05% (vol)NO,0.05% (vol)NH3,0.01% (vol)SO2,5% (vol)O2 and balance N2;flow rate:500 ml?min-1;GHSV:50,000 h-1.

The decrease of NH3conversion without NOxat 200 °C showed an apparent delay as compared to NOxconversion in Fig.2.The NOxconversion decreased to 60% at 120 min,whereas NH3conversion reached 60% at 210 min.At 300 °C,both NOxand NH3conversion was kept as 100% during 900 min,confirming that SO2poisoning had a minor influence on Cu-SSZ-13 when the temperature reached 300 °C.

3.2.Transient reactions

Fig.3.Transient reaction results at 200°C over the Cu-SSZ-13 catalyst.The reaction details are summarized and shown in Table 2.

In addition to the reaction with NH3in the feed,SO2could also be adsorbed on isolated Cu2+ions in Cu-CHA catalysts[16,28].The transient study was carried out to verify the influence of SO2exposure to isolated Cu2+ions in Cu-SSZ-13 on the SCR performance.The reaction condition of each step was summarized in Table 2,and the results were given in Fig.3.In Experiment 1,Step 1 (E1,S1),Cu-SSZ-13 sample was first stabilized in standard SCR reaction conditions for 60 min.At S2 in E1,the NOxand NH3were switched off and 0.01% (vol) SO2was added to the flow.The SO2concentration began to break through at around 71 min and slowly increased to their inlet values (0.01% (vol)) at 88 min.This indicates that around 0.053 mmol?g-1SO2was adsorbed on the isolated Cu2+ions during SO2exposure,because the zeolite could not adsorb SO2[22,28].SO2was switched off after the saturation,and the poisoned catalyst was exposed to the standard SCR condition again at the S3 in E1.With the addition of the standard SCR gas,the concentration of NO gradually increased to 0.014% (vol)and then decreased,and NH3was detected after 11 minutes and then gradually increased.NH3may be adsorbed on the sulfur poisoning site and form ammonium sulfate species.The After stabilization,around 0.005% (vol)NH3and NOxremained in the outlet gases,corresponding to 88% NOxconversion,indicating that SO2adsorption on the isolated Cu2+ions deactivates the Cu-SSZ-13.

3.3.Regeneration temperature

The Cu-SO2-200 sample was regenerated in air at 500 °C and 600°C,and the effectiveness of desulfation(deSOx)was compared.The NOxconversion increased to 76.2% and 79.2% at 200 °C,with the Cu-SO2-Re-500 and Cu-SO2-Re-600 samples,respectively,as shown in Fig.4(a) and 4(b),indicating the benefit of increasing the regeneration temperature from 500°C to 600°C.Nevertheless,there was still 6% NOxconversion loss of Cu-SO2-Re-600 (79.2%)compared to that of the Cu-SSZ-13 (85.2%) at 200 °C,indicating the regeneration in air at 600 °C cannot recover the activity completely.For the N2selectivities,only the sulfur-poisoned samples had a decrease of about 4% compared to the fresh samples at 150°C,and recovered after regeneration as shown in Fig.S2.Meanwhile,all samples showed nearly 100% N2selectivity within the tested temperature range,which means that the sulfur poisoning has no effect on the nitrogen selectivity of the catalyst.

3.4.Structure and chemical content

Table 3 lists the data of surface area and the pore volume of the samples.The surface area and pore volume decreased significantly,with only around 25% surface area and pore volume remaining in the Cu-200-SO2,as compared to Cu-SSZ-13 sample.After the regeneration in air at 500 °C and 600 °C,around 89% and 95% of surface area and the pore volume were recovered in Cu-SO2-Re-500 and Cu-SO2-Re-600,respectively.The Cu content was almost constant after SO2poisoning according to ICP,as shown in Table 4.

Table 3 Textural properties of the catalysts

Table 4 The amount of NH3,SO2 desorbed from the catalysts obtained from SO2,NH3-TPD,the Cu content and the sulfur coverage

Fig.4.(a)NOx conversion curves of the fresh,SO2 poisoned Cu-SSZ-13 samples and regenerated Cu-SSZ-13 samples.The dash line rectangle highlights the conversion at 200°C.(b) NOx convention activities of the catalysts at 200 °C.Cu-Re-500 and Cu-Re-600 in Fig.4(b) are the Cu-SO2-Re-500 and Cu-SO2-Re-600 for short.Reaction conditions:0.05% (vol) NO,0.05% (vol) NH3,5% (vol) O2,5% (vol) H2O and balance N2;flow rate:500 ml?min-1;GHSV:300,000 h-1.

3.5.Thermogravimetric and desorption analysis

TGA was utilized to distinguish the different sulfate species formed on the sulfated Cu-SO2-200 sample.As shown in Fig.5,two weight loss peaks at around 200°C and 925°C were observed with the fresh Cu-SSZ-13 sample,ascribed to the H2O evaporation and O2desorption from SSZ-13 framework due to structure collapse,respectively [29].For the Cu-SO2-200 sample,except for the peaks appeared in Cu-SSZ-13,two additional mass losses in 250–600 °C and 650–800 °C with two main DTG peaks at 430 and 735 °C as well as a shoulder peak at 310 °C appeared with the Cu-SO2-200 sample.Those three peaks were related to the decomposition of sulfate species deposited in Cu-SO2-200 sample.The mass loss content in 250–600 °C (around 6.3% (mass)) was much higher than that in 650–800 °C (around 0.89% (mass)),indicating most of the sulfate species can be decomposed below 600°C.

TPD was carried out to track the gas products from the decomposition of sulfur-containing species in Cu-SO2-200 sample.Cu-200 sample was chosen as a reference sample,the results are illustrated as in Fig.6(a),much more NH3was desorbed from the Cu-SO2-200 sample (1.65 mmol?g-1) than that from the Cu-200 sample (0.27 mmol?g-1),with a main peak at 410 °C and a shoulder peak at 300 °C,indicating the interaction of NH3and SO2on Cu-SO2-200.As for the SO2-TPD results in Fig.6(b),no SO2emission was detected for the Cu-200 sample.However,for the Cu-SO2-200 sample,around 0.64 mmol?g-1of SO2was desorbed below 500°C with a desorption peak at 423°C,and around 0.04 mmol?g-1of SO2was desorbed above 500 °C.

3.6.SEM and element mapping

The SEM and element mapping images of Cu-SSZ-13 and Cu-SO2-200 were shown in Fig.7(a) and (b),respectively.Both samples showed a typical cubic crystal.There is no S element detected in the Cu-SSZ-13 sample as shown in Fig.7(a).While,about 2.81% (mass)S was detected on the sulfated catalysis surface as shown in Fig.7(b).which means that some SO2may be adsorbed on the surface of the catalyst.This should be ascribed to the formation of sulfur-containing species in Cu-SO2-200 during SO2exposure,implying the coverage of sulfur-containing species on Cu-SSZ-13 during SCR reaction with SO2exposure.

Fig.5.TG-DTG curves of samples Cu-SSZ-13(upper)Cu-SO2-200(bottom).The TGA measurements were carried out in air with the ramping rate of 10 oC?min-1.

3.7.Electron paramagnetic resonance

EPR experiment was made to probe the state of cupric sites,shown in Fig.8.Only isolated Cu2+ions give EPR signal,whereas other Cu species are EPR silent[1,45].The strength of the EPR spectrum decreased after SO2poisoning and around 31% isolated Cu2+content was reduced for the Cu-SO2-200 sample,as compared to Cu-SSZ-13,shown in Fig.8(a)and 8(b).This indicates a part of isolated Cu2+ions were poisoned by SO2.

3.8.X-ray photoelectron spectroscopy

Fig.9(a) shows the XPS spectra of the S 2p regions for and Cu-SO2-200 samples.The S 2p binding energies on the Cu-SO2-200 mainly appeared at 168.8 and 166.6 eV,indicating the presence of S6+and S4+in Cu-SO2-200 [43,44].The binding energy at168.8 eV was assigned to,while binding energy at 166.6 eV was attributed toor adsorbed SO2[46,47,48].Jangjouet al.[16] reported that SO2would react with ZCuOH (Cu(OH)+) and form ZCuHSO3,which could be further oxidized into copper bisulfate ZCuHSO4.Therefore,the S in the Cu-SO2-200 should exist asand chemisorbed SO2[46,47,48].

Fig.6.TPD profiles of the Cu-200 and Cu-SO2-200 samples.(a)NH3 curves and(b)SO2 curves.No NO,NO2,N2O and SO3 were detected during TPD.The TPD measurements were conducted in 20% O2/N2 with a ramping rate of 10 oC?min-1.

Fig.9(b)shows the XPS spectra of the Cu 2p regions for Cu-SSZ-13 and Cu-SO2-200 samples.The peaks centered at 933.3 eV for Cu 2p3/2are shown with Cu-SSZ-13,assigned to CuO [49–51].Compared to the Cu-SSZ-13,an additional peak at 935.5 eV appeared in the Cu-SO2-200 sample,which was assigned to the CuSO4[52].

4.Discussion

4.1.The sulfur-containing species deposited

Large amount of NH3(1.65 mmol?g-1) and SO2(around 0.64 mmol?g-1) were desorbed from Cu-SO2-200,as shown in the TPD results in Fig.6.This clearly indicates that the ammoniasulfate species formed in the Cu-SO2-200 sample during SO2exposure in SCR reaction.However,NH3release started at around 250°C,while SO2desorption was only observed when the temperature was above 350°C for Cu-SO2-200 sample during TPD in Fig.6.This suggests that the decomposition of the ammonia-sulfate species should follow two-steps;the first step was the NH3release below 300 °C,and residues were subsequently decomposed to NH3and SO2in the second step at higher temperatures.This evidence is consistent with the nature of the (NH4)2SO4decomposition [30],which follows the equations below:

Therefore,the ammonia-sulfate species deposited on Cu-SO2-200 might be ascribed to (NH4)2SO4.The amount of the NH3and SO2decomposed from ammonia-sulfate species during TPD(Fig.6) further confirm this hypothesis,as the mole ratio of the NH3:SO2is around 2,the data as shown in Table 4.The formation of (NH4)2SO4on Cu-SO2-200 during SO2poisoning probably follows the reaction (7):

where H2O is produced during the reaction according to Eqs.(1)and (2).

As shown in Fig.6,0.04 mmol?g-1SO2was desorbed in 500–600°C during TPD,while no corresponding NH3desorption was observed,indicating that this SO2release peak was not from the decomposition of (NH4)2SO4.In the previous SO2+O2-TPD results of Cu-SSZ-13 and Cu-SAPO-34 preparedviaion-exchange method[16,30],the release of SO2chemisorbed on isolated Cu2+ions was exclusively appeared above 500 °C.Therefore,the SO2desorption in 500–600 °C (Fig.6(b)) indicates that a part of SO2was chemisorbed on isolated Cu2+ions in Cu-SO2-200 sample,consistent with the XPS results in Fig.9(a).The TG-MS results on Cu-SO2-200 showed the results,and there was a desorption peak of SO2at 590 °C and no NH3signal above 500 °C as shown in Fig.S4,which further proves a part of SO2was chemisorbed on isolated Cu2+ions in Cu-SO2-200 sample.

As shown in Fig.S3,although the samples only sulfcated in SO2+O2conditions showed higher SCR activity at low temperature (<350 °C) compared to Cu-SO2-200 sample,it still had a severe deactivation compared to the fresh samples.The above results indicate both (NH4)2SO4formation and chemisorption of SO2could cover the isolated Cu2+ions in Cu-SSZ-13 catalyst[15,16,18].Therefore,the S/Cu ratio,based on moles of SO2released during the TPD (Fig.6(b)) per mole of Cu measured by ICP-OES,was calculated to show the S coverage on cupric sties in the Cu-SO2-200,shown in Table 4.The S/Cu ratio (0.88) in Cu-SO2-200 is close to 1,indicating most of the cupric sites were covered by the sulfur species,probably causing the deactivation of Cu-SSZ-13 (Fig.2).Moreover,it is interesting to point out that one-pot Cu-SSZ-13 shows a higher S/Cu ratio than those of both SO2poisoned Cu-SSZ-13 (0.62) and Cu-SAPO-34 catalysts (0.59)prepared with ion-exchanged method [15,27].This indicates that the formation of (NH4)2SO4and chemisorption of SO2on Cu2+ions in Cu-CHA catalyst are influenced by the preparation method,which may be related to the different type and dispersion of Cu2+ions in the samples [37,38].One-pot Cu-SSZ-13 usually contains a large number of ZCuOH sites,compared to ionexchange Cu-SSZ-13 [37,38,40].Jangjouet al.[16] proposed that the SO2only adsorbed on Z2Cu sites when it was cofed with NH3,viaformation of ammonium sulfate,while,reaction of SO2with ZCuOH would form copper bisulfite species at low temperature,which was further oxidized to copper bisulfate with the increase of temperature.So the influence of sulfur at low temperature is mainly the formation of (NH4)2SO4,which will cover two types of active copper sites,regardless of whether it is Cu-SSZ-13 obtained by one-pot method or ion exchange method.However,more ZCuOH sites in one-pot Cu-SSZ-13 will lead to form more stable bisulfite/bisulfate species.

Fig.7.(a) SEM and elemental mapping images of Cu-SSZ-13 sample and (b) SEM and elemental mapping images of Cu-SO2-200 sample.

As shown in EPR results in Fig.8,31% isolated Cu2+ions decreased in the Cu-SO2-200,as compared to the fresh Cu-SSZ-13.This indicateds that a part of isolated Cu2+ions had been poisoned by SO2.XPS results (Fig.9) further confirm the formation of CuSO4-like species in Cu-SO2-200 sample.TPD (Fig.6),EPR (Fig.8),XPS (Fig.9) and TG-MS (Fig.S4) results show that there were three types of sulfate species coexisting on the catalyst after the SO2exposure,i.e.,(NH4)2SO4,chemisorbed SO2on Cu2+ions and CuSO4-like species.Recently,Jangjouet al.[16]and Bergmanet al.[53] put forward that the highly stable CuSO4-like species were ZCuHSO4rather than CuSO4by DFT calculations.

Fig.8.(a) EPR spectra of the Cu-SSZ-13 and Cu-SO2-200 samples.(b) Concentration of isolated Cu2+ ions.The normalized Cu2+ ions concentration was determined with double integrating the EPR spectra,using the isolated Cu2+ ions content in Cu-SSZ-13 as the reference.Spectra were collected at -150 °C.

Fig.9.XPS spectra of (a) S 2p and (b) Cu 2p for the Cu-SSZ-13 and Cu-SO2-200 samples.

4.2.The temperature effect on regeneration

According to TPD (Fig.6),EPR (Fig.8) and XPS (Fig.9) results,there are three types of sulfate species coexist on the catalyst after the SO2exposure,i.e.,(NH4)2SO4,chemisorbed SO2on Cu2+ions and CuSO4-like species.The different regeneration temperatures recovered the activity to different degrees,as shown in Fig.4,indicating that the different regeneration temperature could remove different sulfur in the sample.

(NH4)2SO4was decomposed completely after regeneration at 500 °C,as confirmed by the TPD results.At the same time,the NOxconversion was mostly recovered (76.2% for Cu-SO2-Re-500 vs.79.2% for fresh Cu-SSZ-13)at 200°C,as shown in Fig.4(b).This indicates that the suppression of SCR activity during SO2exposure at low temperatures was mainly ascribed to the formation of the(NH4)2SO4,explained by the pore-blocking effect of (NH4)2SO4in Cu-SO2-200,as confirmed with the recovery of most of the pore volume,surface area (Table 4) as well as the deNOxactivity(Fig.4) after regeneration at 500 °C.

Cu-SO2-Re-600 showed a higher NOxconversion (79.2%) as compared to that (76.2%) of Cu-SO2-Re-500 at 200 °C,as shown in Fig.4(b).This can be explained by the removal of chemisorbed SO2on Cu2+ions in 500–600 °C (Fig.6(b)) because the SO2adsorbed on isolated Cu2+ions also deactivate the Cu-SSZ-13,as shown in the transient reaction result in Fig.3,that NH3may been adsorbed on the sulfur poisoning site and form ammonium sulfate species,furthermore cause the decrease of NH3-SCR activity.

The Cu-SO2-Re-600 still showed about 6% activity loss at 200°C,as compared to the Cu-SSZ-13,shown in Fig.3(b).This should be related to the isolated Cu2+transformation to CuSO4-like species,required higher desulfation temperatures (>600 °C) (Figs.8,9 and S4).

Based on the discussion above,the regeneration of the Cu-SO2-200 sample at 500 °C and 600 °C was proposed as in Fig.10.After being treated at 500 °C,(NH4)2SO4in the Cu-SO2-200 sample was decomposed,recovering most of the deNOxactivity.When the regeneration temperature was increased to 600 °C,both the(NH4)2SO4species and the SO2adsorbed on the Cu2+sites were removed,resulting in a further activity improvement for the Cu-SO2-Re-600 sample as compared to Cu-SO2-Re-500.However,the formation of rather stable CuSO4-like species caused the incomplete activity recovery for Cu-SO2-200 even after treatment at 600 °C.

5.Conclusions

The activity of the one-pot synthesized Cu-SSZ-13 decreased significantly with 0.01% (vol) SO2co-feeding under standard SCR conditions below 300 °C,while no apparent activity loss detected in the temperature range above 300°C with the same SO2concentration in the feed.The deactivation of the Cu-SSZ-13 was caused by the formation of (NH4)2SO4,the SO2chemisorption on isolated Cu2+ion sites,as well as the formation of CuSO4-like species.

Fig.10.The effect of different regeneration temperatures on regenerating the Cu-SO2-200 sample.

The formation of (NH4)2SO4blocking the active sites was considered as the primary reason for the deactivation of Cu-SSZ-13.

Increase the reaction temperature to 500°C efficiently removes the(NH4)2SO4and recover most of the activity of the poisoned Cu-SSZ-13 catalyst.After being treated at 600 °C,where both the(NH4)2SO4and the SO2adsorbed on the Cu2+sites in the Cu-SO2-200 were removed,further improvement of the activity in the Cu-Re-SO2-600 than that in the Cu-Re-SO2-500 was observed.The incomplete activity recovery of the Cu-Re-SO2-600 sample was due to the formation of the highly stable CuSO4-like species.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

Financial supports from the Natural Science Foundation of Tianjin 19JCTPJC42300.

Supplementary Material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cjche.2021.06.012.