Ben Liu,Nangui Lv,Chan Wang,,Hongwei Zhang,Yuanyuan Yue,, ,Jingdong Xu,Xiaotao Bi,4,Xiaojun Bao,,5

1 National Engineering Research Center of Chemical Fertilizer Catalyst,College of Chemical Engineering,Fuzhou University,Fuzhou 350002,China

2 Qingyuan Innovation Laboratory,Quanzhou 362801,China

3 Sinochem Quanzhou Energy Technology Co,Ltd,Quanzhou 362000,China

4 Department of Chemical &Biological Engineering,University of British Columbia,2360 East Mall,Vancouver,BC V6T 1Z3,Canada

5 State Key Laboratory of Photocatalysis on Energy &Environment,College of Chemistry,Fuzhou University,Fuzhou 350116,China

Keywords:Catalyst Zeolite Cu-SSZ-13 Ion-exchange Redistribution of Cu species Selective catalytic reduction (SCR)

ABSTRACT The nature and distribution of Cu species in Cu-SSZ-13 play a vital role in selective catalytic reduction of NO by NH3(NH3-SCR),but existing methods for adjusting the Cu distribution are complex and difficult to control.Herein,we report a simple and effective ion-exchange approach to regulate the Cu distribution in the one-pot synthesized Cu-SSZ-13 that possesses sufficient initial Cu species and thus provides a “natural environment” for adjusting Cu distribution precisely.By using this proposed strategy,a series of Cu-SSZ-13x zeolites with different Cu contents and distributions were obtained.It is shown that the dealumination of the as-synthesized Cu-SSZ-13 during the ion-exchange generates abundant vacant sites in the double six-membered-rings of the SSZ-13 zeolite for relocating Cu2+ species and thus allows the redistribution of the Cu species.The catalytic results showed that the ion-exchanged Cu-SSZ-13 zeolites exhibit quite different catalytic performance in NH3-SCR reaction but superior to the parent counterpart.The structure–activity relationship analysis indicates that the redistribution of Cu species rather than other factors (e.g.,crystallinity,chemical composition,and porous structure) is responsible for the improved NH3-SCR performance and SO2 and H2O resistance.Our work offers an effective method to precisely adjust the Cu distribution in preparing the industrial SCR catalysts.

1.Introduction

Nitrogen oxides (NOx) emitted from mobile and stationary sources are one kind of the most environmentally harmful compounds resulting in the formation of photo-chemical smog,haze and acid rain as well as the direct damage to human respiratory system [1–3].To meet the increasingly stringent emission standards,numerous efforts have been devoted to the development and application of NOxconversion technologies[4].Among various strategies available,the selective catalytic reduction (SCR) of NOxwith NH3or urea as the reducing agent has been regarded as the best technique for controlling NOxemissions[3–5].In recent years,the successful synthesis and commercialization of Cu-SSZ-13 catalyst became a new milestone in the development of SCR technique due to its outstanding activity,selectivity,and durability [6].

It has been revealed that the nature and distribution of Cu species in Cu-SSZ-13 play vital roles in NH3-SCR reaction,although it is still a much debated topic [7].Korhonen et al.[8] proposed that isolated Cu2+ions were the catalytically active sites in SCR reaction.SSZ-13 zeolite with CHA topology can provide multifarious sites including cha cages,double six-membered-rings (D6Rs),and eight-membered-rings (8MRs) for loading Cu2+ions [9–12].Deka et al.[13] used in-situ X-ray absorption fine structure (XAFS) and X-ray diffraction (XRD) techniques to investigate the nature of Cu species under realistic NH3-SCR conditions,and confirmed that the isolated Cu2+ions located in the D6Rs of SSZ-13 zeolite were the active sites.Additionally,Paolucci et al.[14] prepared Cu-SSZ-13 via aqueous ion-exchange with a Cu(NO3)2solution and found that the Cu2+ions first populated two Al (2Al) sites before populating remaining unpaired or 1Al sites in six-memberedrings (6MRs),demonstrating the presence of some preferred sites for locating Cu2+ions in SSZ-13 zeolite.Interestingly,Beale et al.[15] employed the synchrotron-based in-situ time-resolved XRD technique to study the conversion behavior of Cu ions during activation of Cu-SSZ-13 prepared by aqueous ion-exchange with an aqueous solution of copper sulfate,and they concluded that the Cu2+ions in D6Rs and 8MRs would migrate during the NH3-SCR reaction,which was affected by the initial Cu2+distribution in the zeolite and further influenced the catalytic performance.Although there is no agreement reached,it is generally accepted that the isolated Cu2+ions located in the D6Rs of Cu-SSZ-13 are the desired active sites in NH3-SCR reaction.

Various attempts (including the use of different synthesis strategies and templates,and the adjustment of Si/Al ratios and topologies)have been made to manipulate the nature and distribution of Cu species in Cu-SSZ-13 zeolite.For example,Jiang et al.[16]reported that synthesis strategy could influence the distribution of Cu species,and they found that the Cu-SSZ-13 synthesized via onepot route possessed a higher content of Cu2+species than that obtained through ion-exchange method.Corma and co-workers[17] synthesized different Cu-SSZ-13 zeolite with controllable Cu/(Si+Al) ratios and active Cu sites using two templates(Cu-tetraethylenepentamine (Cu-TEPA) complex and the organic N,N,N-trimethyl-1-adamantammonium).Fan et al.[18] found that the varying Si/Al ratios could adjust the nature of Cu species and further influenced their NH3-SCR activity,because the suitable Al distribution can benefit the distribution of high stable Cu species.Sun et al.[19] fabricated Cu-CHA composites with two zeolites(Cu-SAPO-34 and Cu-SSZ-13) and adjusted Cu species distribution using Cu-SSZ-13 as Cu,Si and Al sources.However,the abovementioned methods are complex and cannot precisely regulate the nature and distribution of Cu species in the resulting Cu-SSZ-13 zeolites to obtain abundant isolated Cu2+ions located in the D6Rs.It is known that,unlike in the Cu-SSZ-13 zeolite obtained by post-synthesis method (such as wet or solid-state ionexchange and wetness impregnation) which needs additional ion-exchange or impregnation and calcination steps and has the drawback of controlling the nature and distribution of Cu species,in an as-synthesized Cu-SSZ-13 zeolite via one-pot strategy,excessive Cu species introduced by Cu-TEPA that is used as both Cu source and template are already present,providing a “natural environment” for adjusting Cu species distribution.Therefore,it is highly interesting to adjust the initial distribution of Cu species in one-pot synthesized Cu-SSZ-13 during an indispensable ionexchange step,which may be a simple and effective method to regulate the active sites for preparing the promising SCR catalyst.

In this contribution,a series of Cu-SSZ-13 zeolites were prepared by the ion-exchange of a one-pot synthesized parent Cu-SSZ-13 zeolite with a HNO3solution for different times before the template was removed,and their physiochemical properties especially the distribution of Cu species and NH3-SCR performance were systematically characterized and evaluated,respectively,which leads to the establishment of the relationship between Cu species distribution and NH3-SCR performance and the successful development of a Cu-SSZ-13 based NH3-SCR catalyst system.

2.Experimental

2.1.Preparation of Cu-SSZ-13 zeolites with different Cu amounts

The parent Cu-SSZ-13 zeolite was synthesized via the one-pot hydrothermal method reported by Xiao et al [20].In a typical run,NaAlO2and NaOH were first dissolved into a suitable amount of deionized water resulting in a clear solution;then,Cu(NO3)2-·3H2O and TEPA were added into the clear solution,followed by stirring for 4 h to yield a homogeneous mixture;subsequently,Ludox AS-40 was added into the above mixture dropwise under vigorous stirring;after stirring for 4 h,the obtained aluminosilicate gel with a molar ratio of 4.8 Na2O:10 SiO2:1 Al2O3:2 Cu-TEPA:200 H2O was transferred into a Teflon-lined stainless steel autoclave and crystallized at 140 °C for 4 days;once the crystallization was finished,the solid product was collected by filtration,washing with deionized water,and drying at 120°C for 12 h to obtain the parent Cu-SSZ-13 zeolite;finally,the resulting parent Cu-SSZ-13 zeolite was calcined at 600°C for 5 h yielding a template-free sample designated as Cu-SSZ-130.

To obtain Cu-SSZ-13 zeolites with different Cu distributions,the as-synthesized parent Cu-SSZ-13 zeolite without the removal of the template was ion-exchanged with a 0.1 mol·L-1HNO3solution for 4,8,12,and 16 h,respectively,followed by filtration,washing with deionized water,drying at 120 °C for 12 h and calcination at 600 °C for 5 h.The resultant samples were denoted as Cu-SSZ-13xwhere subscript “x” stands for the ion-exchange time.The Cu and Na contents of these samples are listed in Table 1.

Table 1 Physicochemical properties of Cu-SSZ-13x

2.2.Characterizations

The XRD patterns were collected on a D/max Ultima IV diffractometer (Rigaku,Japan) with a Cu Kα radiation source (40 kV and 40 mA) in the 2θ range of 5°–70° and operated at a scan rate of 8(°)·min-1to determine the structures and the phase compositions of the samples.N2adsorption–desorption experiments were conducted at -196 °C on a Micromeritics ASAP 2046 M instrument(USA) to obtain the specific surface areas (SBET) and pore volumes(Vpore)of the samples.An inductively coupled plasma optical emission spectrometer (ICP-OES,Varian ICP-720A,USA) was used to determine the sodium,copper,silicon,and aluminum contents of the obtained samples.Prior to analysis,a certain amount of one sample to be measured was dissolved in a mixture of HNO3+HCl in a microwave oven.High-resolution transmission electron microscopy(HRTEM)characterization was conducted on a Tecnai G2 F20 transmission electron microscope (Switzerland) to study the microstructures of the zeolites.X-ray photoelectron spectroscopy(XPS) measurements were carried out on a Thermo Fisher KAlpha spectrometer (USA) with an Al Kα X-ray source,and the binding energies were calibrated using C 1s peak at 284.6 eV as an internal standard.Temperature programmed desorption of ammonia (NH3-TPD) experiments were performed on an Auto-Chem 2920 apparatus(USA) with a thermal conductivity detector.Temperature-programmed reduction of hydrogen (H2-TPR) experiments were conducted on a Quantachrome Chem BET Pulsar TPR instrument (USA).Ultraviolet–visible diffuse reflectance spectroscopy (UV–vis DRS) spectra were recorded using a Lambda950 ultraviolet–visible photometer (Perkin-Elmer,USA).Ultraviolet(UV) Raman spectra were measured by using an InVia-Reflex Raman spectrometer (British Renishawn Company) with a resolution of 2 cm-1.A He/Cd laser with wavelength of 532 or 325 nm and output power of 50 mW was used as an exciting source.Solid-state27Al and29Si magic angle spinning nuclear magnetic resonance (MAS NMR) spectra were measured on a Bruker Advance III 600 and 500 MHz instruments(Germany),respectively.Electron paramagnetic resonance (EPR) experiments were conducted on a Bruker ESP320 spectrometer (USA) operated in the X-band microwave frequency range.

2.3.Catalytic activity tests

The NH3-SCR performance assessments of the different samples were carried out under atmospheric pressure in a fixed bed quartz microreactor operating in a steady flow mode in the temperature range from 100 to 700 °C.The Cu-SSZ-13xzeolites were tableted,crushed,and sieved to obtain a series of particles of 420–840 μmin size.The reactant gas mixtures contained 500 ml·m-3NO,500 ml·m-3NH3,5% (vol) O2,10% (vol) H2O (when used),100 ml·m-3SO2(when used),and balanced N2.The total flow rate was 600 ml·min-1and thus a gaseous hourly space velocity(GHSV)of 180,000 was obtained.The concentrations of NO,NH3,and NO2in the outlet gas mixture were measured by a Testo340 flue gas analyzer (Germany).The concentration of N2O in the outlet gas mixture was analyzed by a Nicolet iS50 Fourier-transform infrared(FT-IR)spectrometer(Thermo Fisher,USA)equipped with a heated,2 m path-length gas cell.The NO conversion and N2selectivity were calculated as described elsewhere [21].

3.Results and Discussion

3.1.Physicochemical properties of Cu-SSZ-13x

Fig.1 shows the XRD patterns of the different Cu-SSZ-13xzeolites.Cu-SSZ-130exhibits the typical diffraction peaks at 2θ=9.5°,14.0°,16.1°,17.9°,20.7°,25.1° and 30.7° associated with the CHA structure(Fig.1(a)),indicating the successful synthesis of a highly crystalline SSZ-13 zeolite[22].Gratifyingly,the intensities of these diffraction peaks in the XRD patterns of the ion-exchanged Cu-SSZ-13x(x ＞0) samples have no obvious difference (Fig.1(a)),confirming that the framework structures of the zeolites are well preserved even after the ion-exchange with the HNO3solution.As shown in Fig.1B,additional diffraction peaks at 2θ=35.6°,38.7°,and 38.9° ascribed to the CuO phase (PDF #48-1548) are observed in the XRD pattern of Cu-SSZ-130.Here it should be pointed out that no diffraction peak assigned to CuO phase can be observed in the as-synthesized Cu-SSZ-13 zeolite and the ionexchanged zeolites before the template was removed (Fig.S1 in Supplementary Material).These results suggest that the formation and migration of CuO phase on the external surface of Cu-SSZ-130are occurred during calcination process,which is due to excessive Cu species in Cu-SSZ-130.The XRD patterns of the ion-exchanged samples Cu-SSZ-134and Cu-SSZ-138have significantly decreased intensities of the diffraction peaks ascribed to the CuO phase,indicating the depressed formation of the CuO phase [23,24].In the XRD patterns of Cu-SSZ-1312and Cu-SSZ-1316that were obtained by increasing the ion-exchange time to 12 and 16 h,respectively,the diffraction peaks associated with the CuO phase totally disappear,indicating the highly dispersed Cu species in them.

According to the ICP-OES results listed in Table 1,we can see that the Cu content of Cu-SSZ-130reaches up to 11.27% (mass),resulting in the migration of Cu cations outside of the cha cage and the formation of the bulk CuO phase during the followed calcination [4,22],consistent with the XRD result.As compared with that of Cu-SSZ-130,the Cu content of Cu-SSZ-134obtained by ion-exchange for 4 h dramatically decreases from 11.27% to 4.95% (mass),but that of Cu-SSZ-1316obtained by further extending the ion-exchange time to 16 h only slightly drops from 4.95%to 4.69% (mass).The presence of CuO has been regarded as a major reason for the deteriorated NH3-SCR performance of Cu-zeolites at high temperatures (＞400 °C) due to the occurrence of unselective catalytic oxidation of NH3to NOx[25,26].Therefore,it is necessary to treat the one-pot synthesized Cu-SSZ-13 via the ionexchange with a HNO3solution to remove the excessive Cu species.In addition,as co-cations in zeolites,Na+ions also impact the catalytic activity of Cu-zeolites.As seen from Table 1,the Cu-SSZ-13xzeolites after ion-exchange with the HNO3solution for different lengths of time also have the decreased Na contents,from 3.15%(mass) of Cu-SSZ-130to 0.6%–0.7% (mass) of Cu-SSZ-134,Cu-SSZ-138and Cu-SSZ-1312and finally to 0.41% (mass) of Cu-SSZ-1316.Interestingly,an increasing trend of bulk SiO2/Al2O3ratio can also be observed in Table 1,signifying that the dealumination occurs during ion-exchange.

Fig.1.XRD patterns of the different zeolites at 2θ=5°–40° (A) and 35°–40° (B).

Fig.2.N2 adsorption–desorption isotherms of Cu-SSZ-13x.

Fig.2 shows the N2adsorption–desorption isotherms of the Cu-SSZ-13xzeolites and Table 1 lists the derived textural parameters,from which the changes in pore structure caused by the ionexchange can be observed.According to the IUPAC classification,all the samples have the type-I isotherms,indicating that they are characteristic microporous materials[27,28].It is apparent that the samples Cu-SSZ-13x(x ＞0) obtained after the ion-exchange have largely increased N2adsorption amounts both at relative pressure P/P0below 0.02 and above 0.9 as compared with Cu-SSZ-130,suggesting the more open pore channels.Besides,Table 1 also shows that Cu-SSZ-130has the lowest values of SBET(375 m2·g-1),Vpore(0.20 cm3·g-1),microporous specific surface area (Smicro,370 m2·g-1),and microporous volume (Vmicro,0.19 cm3·g-1) [27,28],whereas the samples obtained after the ion-exchange with the HNO3solution have significantly increased values of the corresponding parameters,attributed to the removal of the excessive Cu species from the parent Cu-SSZ-13 zeolite[29].Combined with the results of XRD,ICP-OES,and N2adsorption–desorption,it is reasonable to declare that the ion-exchange with HNO3solution before calcination is an effective route to eliminate the excess Cu species in the one-pot synthesized Cu-SSZ-13 zeolite.

Table 2 Surface atom compositions and oxidation state of Cu 2p species over the different zeolites obtained by XPS analysis

3.2.NH3-SCR performance of Cu-SSZ-13x

The catalytic activities of Cu-SSZ-13xzeolites were evaluated under the standard SCR conditions,and the results are shown in Fig.3.As seen in Fig.3(a),with the increase in reaction temperature,the NO conversion over Cu-SSZ-130rapidly increases first,then keeps stable between ca.200 and 400 °C,and finally drops drastically,giving a narrow reaction temperature window(defined as the temperature interval for NO conversion ＞90%) of 200–400 °C.It was previously reported [18,21,29] that the presence of CuO had little impact on the NH3-SCR activity at low temperatures,but severely degraded the activity at high temperatures because of the overoxidation of NH3to NOx.Therefore,the worsen activity above 400 °C is due to the existence of CuO on the surface of Cu-SSZ-130.This can be also proved by the sharp decline in N2selectivity and the abrupt increase in NO2yield over Cu-SSZ-130in the high temperature range (＞400 °C,as shown in Fig.3(b) and(c)).These facts emphasize that the removal of CuO from the surface of Cu-SSZ-13 is essential to obtain a NH3-SCR catalyst with satisfactory high temperature activity.This is further supported by the significantly widen reaction temperature windows of Cu-SSZ-13x(x ＞0) with much lower Cu contents,as reflected in Fig.3(a).Additionally,it is noted that the reaction temperature window of Cu-SSZ-13xcatalysts widens first and then slightly narrows as the ion-exchange time is extended (Fig.3(a)).It is also noted that over Cu-SSZ-13xthe undesired product NO2yields are similar when the reaction temperature is lower than 450 °C but differ from each other when the temperature is higher than 450 °C (Fig.3(c)).Although having different NO conversions and NO2yields,all of the ion-exchanged samples gave very high N2selectivity(nearly 100%)and very low unwanted NO2yield(below 6 ml·m-3)in the temperature range of 200–500°C(Fig.3(b)and 3(c)).Among the five catalysts,Cu-SSZ-1312without CuO possesses the widest reaction temperature window(145–690°C)and lowest NO2yield.The above results lead to the conclusion that the NH3-SCR performance of the ion-exchanged Cu-SSZ-13xzeolites are closely related to the ion-exchange time.Specifically,the prolonged ion-exchange time leads to the decreased CuO content and thereby the widened reaction temperature window.Importantly,here we demonstrate that Cu-SSZ-1312obtained via ion-exchange for 12 h has the unprecedented catalytic performance at temperatures of as high as 700°C and thus can be operated under the harsh conditions as encountered in a real diesel exhaust [30].

It is known that the diesel engine exhaust usually contains a significant amount of H2O and SO2generated from the combustion of diesel fuel that can poison Cu-SSZ-13.Therefore,the antipoisoning performance of Cu-SSZ-13xwas further tested in the presence of H2O and/or SO2,and the results are presented in Fig.4.It can be seen that the presence of 10%(vol)H2O in the feed gases has no influence on NO conversion,indicating the good resistance of the different zeolites to H2O;however,the presence of 10%(vol)H2O and 100 ml·m-3SO2leads to the drastic drop in NO conversion,with Cu-SSZ-13xof much lower Cu contents showing the much better tolerance to the coexistence of H2O and SO2than Cu-SSZ-130of a higher Cu content.More importantly,after switching off the H2O and SO2streams from the feeding gas,the NO conversion over Cu-SSZ-13xrestores to～100% within about 30 min,whereas that over Cu-SSZ-130only goes back to about 90%.These results demonstrate that the ion-exchange of Cu-SSZ-130with the HNO3solution can dramatically enhance the tolerance of the resulting Cu-SSZ-13xzeolites to H2O and SO2poisoning,with Cu-SSZ-1312showing the best SO2and H2O resistance.

The above characterization and catalytic performance results show that the ion-exchange treatment of parent Cu-SSZ-13 zeolite by HNO3solution for different lengths of time yields Cu-SSZ-13xzeolites with significantly decreased CuO and Na contents and more open pore channels (i.e.,larger SBETand Vpore) [25].To our surprise,however,the different zeolites Cu-SSZ-13xobtained after ion-exchange for different lengths of time(4–12 h)exhibited quite different catalytic performance despite their almost identical physicochemical properties such as crystallinity,Cu and Na contents,and SBETand Vporevalues.This raises the question:what is the underlying reason for this phenomenon?

3.3.Redistribution of Cu species in Cu-SSZ-13x during ion-exchange

To elucidate how the different zeolites Cu-SSZ-13xwith the almost identical physiochemical properties perform differently in the NH3-SCR reaction,the microstructure of the Cu-SSZ-13xzeolites and the distribution of copper species in them were studied.The TEM images and corresponding elemental mappings were acquired and are illustrated in Figs.5 and 6.As seen in Fig.5A–E,all of the samples are in the form of irregular agglomerates consisting of many small particles.It is also seen that,in the TEM images of Cu-SSZ-130(Fig.5A and a),many black dots ascribed to CuO nanoparticles are observed,indicating that the Cu species are heterogeneously distributed on the surface of Cu-SSZ-130.The average size of these nanoparticles is 2.78 nm based on the statistical analysis of more than 100 particles.Differently,as illustrated in Fig.5B–E and b–e,no CuO nanoparticle is clearly detected on the surface of the ion-exchanged samples.This further indicates that the ion-exchange avoids the formation of CuO nanoparticles.

Fig.3.NO conversion (A),N2 selectivity (B),NO2 (C) and N2O (D) yields over the Cu-SSZ-13x catalysts.

Fig.4.NO conversion over Cu-SSZ-13x in the presence of H2O and/or SO2.Reaction conditions:500 ml·m-3 NO,500 ml·m-3 NH3,5%(vol)O2,10%(vol)H2O,100 ml·m-3 SO2 (balance N2),GHSV=180000 h-1,and 200 °C.

The TEM images together with the corresponding elemental mappings of Na,Al,Si,and Cu are displayed in Fig.6(a)-(e).It can be seen that small particles attributed to Cu species in bright contrast can be found on the surface of Cu-SSZ-130,indicating the aggregation of CuO species.Interestingly,no enrichment of Na,Al,Si,and Cu is observed on the surface of the samples treated with the HNO3solution for different lengths of time.Besides,the Na,Al,and Cu atoms are homogeneously distributed on the surfaces of Cu-SSZ-13x(x ＞0).This phenomenon may be attributed to the partial removal of the Al and Cu species during the ionexchange [29].However,the corresponding elemental mappings of the samples treated with the HNO3solution for 4 to 16 h are similar,inconsistent with their differences in SCR performance.Based on the above results,it is reasonable to conclude that the nature and distribution of the Cu species play an important role in regulating the SCR performance,therefore further research is needed.

Fig.5.TEM images of Cu-SSZ-130 (A and a),Cu-SSZ-134 (B and b),Cu-SSZ-138 (C and c),Cu-SSZ-1312 (D and d),and Cu-SSZ-1316 (E and e).

Fig.6.TEM images and corresponding elemental mappings of Cu-SSZ-130 (A),Cu-SSZ-134 (B),Cu-SSZ-138 (C),Cu-SSZ-1312 (D),and Cu-SSZ-1316 (E).

Fig.7.Survey (A) and Cu 2p (B) XPS spectra of Cu-SSZ-13x zeolites.

Fig.7(a) shows the survey XPS spectra of the Cu-SSZ-13xzeolites,which can provide information about the surface environment of Cu in Cu-SSZ-13 zeolite.By integrating the peak areas,the surface atomic concentrations of the zeolites were calculated and are listed in Table 2.The surface Si/Al ratios of Cu-SSZ-13x(x ＞0) are higher than that of Cu-SSZ-130,signifying that partial Al atoms are removed from the surfaces of the zeolites during the ion-exchange.Gao et al.[31] proposed that the location and redox property of the Cu species in Cu-SSZ-13 zeolite could be affected by varying Si/Al ratios.Regulating the surface Si/Al ratio of SSZ-13 zeolite via the HNO3treatment for different lengths of time may pave the way to adjusting the distribution and redox property of Cu species,which further affects the SCR performance of the resulting catalysts [32].Besides,the Cu and Na contents were also dramatically decreased after the ion-exchange,suggesting that some Cu and Na species on the surface the zeolites were also removed during the HNO3treatment [26,33].The removal of partial Cu species can avoid their aggregation.According to Xie et al.[34],excessive Na species decrease the stability of Cu species and further degrade the catalytic performance of Cu species.Therefore,decreasing the content of Na species on the zeolite surface may contribute to the widened reaction temperature window.As known,the Cu 2p XPS spectrum can provide the information on the valence state of Cu species on the zeolite surface.From the Cu 2p XPS spectra of Cu-SSZ-13xin Fig.7(b),the Cu 2p3/2and 2p1/2peaks appear in the ranges of 930–948 and 948–966 eV,respectively [35,36].The binding energies of isolated Cu2+ions and CuO species are located at around 935 and 933 eV,respectively[32].According to the quantification results calculated from Fig.7(b),the contents of isolated Cu2+ions on the surfaces of Cu-SSZ-13x(x ＞0) are much higher than that of Cu-SSZ-130,whereas those of CuO on the surfaces of Cu-SSZ-13xare lower than that of Cu-SSZ-130.These results strongly evidence that less CuO species were formed on the surfaces of Cu-SSZ-13x(x ＞0)during the calcination after the acid treatment.This can be attributed to the removal of partial Cu and Al species during the ion-exchange in the HNO3solution,with the stable Cu species remaining on the zeolite surfaces.

Table 3 Calculated areas of the different types of Si(nAl) species from the deconvolution of 29Si MAS NMR spectra of Cu-SSZ-13x

UV–vis DRS characterization,which is sensitive to the coordination environment of Cu2+species,was used to distinguish the different copper species [37],and the resultant UV–vis DRS spectra of Cu-SSZ-13xzeolites are shown in Fig.8.The band at about 210 and 800 nm are ascribed to the ligand charge transfer O2-→Cu2+from lattice oxygen to isolated Cu2+ions and the d-d transition of isolated Cu2+ions in an octahedral environment[26,33,38,39].The band at 250 nm is associated with the d-d transition and charge transfer transition of Cu with octahedral environment in CuOx[26,33,38].According to the literature [40],sufficiently large CuOxnanoparticles cannot be detected by UV–vis technique.The spectra of Cu-SSZ-13xwere fitted into several peaks based on Gaussian deconvolution.As seen in Fig.8(b),the CuOxspecies take a share of up to 91%,which are the dominant Cu species in the Cu-SSZ-130zeolite.However,the isolated Cu2+ions of the Cu-SSZ-134zeolite obtained after ion exchange for 4 h account for 78%,becoming the major Cu species in the catalyst.With increasing ion-exchange time,the total relative content of isolated Cu2+ions increase dramatically from 78% of Cu-SSZ-134to ca.88% of Cu-SSZ-138and Cu-SSZ-1312and then slightly decreases to 84% of Cu-SSZ-1316.This demonstrates that the relative content of isolated Cu2+ions is strongly affected by the ionexchange time in the HNO3solution [41] and thus can be easily controlled by simply varying the ion-exchange time.

Fig.8.UV–vis DRS spectra (a) and the corresponding deconvolution results (b) of Cu-SSZ-13x zeolites.

Fig.9.EPR spectra (A) and their magnification (B) of Cu-SSZ-13x zeolites.

Electron paramagnetic resonance(EPR)is an effective technique to quantify the amount of isolated Cu2+ions,because CuOxand Cu+species are EPR silent,whereas isolated Cu2+ions are EPR active[42].Therefore,EPR characterization was carried out to further verify the nature and distribution of Cu species in Cu-SSZ-13x,and the results are shown in Fig.9.All of the samples have the same gIIand AIIvalues,indicating the existence of Cu2+ions in the same coordination environment[43].The spectrum of Cu-SSZ-130exhibits the lowest peak intensity due to its least amount of isolated Cu2+ions.Differently,those of the ion-exchanged samples have the stronger EPR signals of Cu2+species,indicating that their higher amounts of isolated Cu2+species which may be either retained or newly formed from the ion-exchange.To further identify the Cu2+species,the spectra at low magnetic field were magnified and are displayed in Fig.9(b).It can be clearly seen that the spectra show gII=2.39 and AII=130 G ascribed to the Cu2+species active for SCR reaction[35,42–44].The amounts of isolated Cu2+ions were calculated by setting the value of Cu-SSZ-130as a reference(1.00)and then integrating the EPR spectra in Fig.9(b).As shown in Fig.S2 Fig.S2(Supplementary Material),the amount of the isolated Cu2+species in different samples increased with the increase of ion-exchange time.Specifically,the amount of isolated Cu2+species jumps from 1.00 in Cu-SSZ-130to 1.62 in Cu-SSZ-134that was treated with the HNO3solution for 4 h;however,the amounts of isolated Cu2+species in the other samples obtained by further increasing the ionexchange time remain almost unchanged.

Fig.10.H2-TPR profiles of Cu-SSZ-13x (a) and the relative amounts of each copper species in Cu-SSZ-13x based on H2-TPR (b).

To understand the location of various Cu species in Cu-SSZ-13x,H2-TPR characterization was carried out and the results are given in Fig.10.According to the literature [9–12],different types of Cu species possessing diverse reduction properties are located at different positions in CHA zeolites;furthermore,the distribution of Cu species is a function of Cu loading and operation environment[28,42].The reduction of dispersed bulk CuO to Cu0occurs below 500 °C in a single step [42],and the reduction of Cu2+ions occurs at different temperatures in two steps:Cu2+→Cu+(at low temperature)and subsequent Cu+→Cu0(at high temperature)[35,36].In Fig.10,the peaks below 300°C are assigned to the reduction of isolated Cu2+species in cha cage [28,35],those between 300 and 400°C belong to the reduction of CuO species to Cu0[23,35],those centered between 400 and 500 °C are ascribed to the reduction of isolated Cu2+ions located in D6Rs [35,36],and those appearing at higher temperatures (＞500 °C) correspond to the reduction of Cu+ions to metallic Cu0[28,35,36].According to the literature[25,26],two coordination sites for Cu+ions coexist in CHA structure:site I is in the center of the hexagonal prism and site II is sited from the 6MRs into the cha cage(Fig.S3,Supplementary Material).Because of the steric hindrance,the reduction temperature for Cu+ions in site I is much higher than that for Cu+ions in site II.It is reasonable to conclude that the peaks between 500 and 700°C are due to the reduction of Cu+in site II and the highest peaks above 700°C are assigned to the reduction of Cu+in site I.The relative contents of the different copper species in Cu-SSZ-13xwere obtained by integrating the peaks,and the results are plotted in Fig.10(b).The relative content of CuO in Cu-SSZ-134decreases dramatically from 40% of Cu-SSZ-130to 10%,and that in Cu-SSZ-138further drops to 9%,and those in Cu-SSZ-1312and Cu-SSZ-1316remain at a nearly constant value (ca.3%).Differently,the relative content of Cu+in site II first decreases with increasing ion exchange time,from 26% of Cu-SSZ-130to 6% (Cu-SSZ-1312),and then goes back to 10% of Cu-SSZ-1316;while with the increase of ion exchange time,the relative content of Cu+in site I shows the tendency of moderate increase at first and then slow decline.Notably,the relative amount of Cu2+ions in D6Rs rises sharply with increasing ion-exchange time,reaching at a maximum value of 81% when the ion-exchange time was 12 h and decreasing to 76% when the ion exchange time was 16 h.These results demonstrate that prolonging ion-exchange time can lead to the redistribution of Cu species in the resulting Cu-SSZ-13 zeolites.Without experiencing the ion exchange,some copper species in the parent zeolite Cu-SSZ-130may occupy cha cages and block pore channels,giving rise to degraded catalytic performance [45].By extending the ionexchange time,more Cu species are redistributed in D6Rs,while the total Cu amounts in the resulting zeolites keep almost at a constant.According to the literature [6,46],Cu2+ions located in D6Rs are considered as the highly active sites for catalyzing NH3-SCR reaction.Moreover,previous reports have also demonstrated that the SO2-tolerance of the isolated Cu2+species (especially located at the D6Rs) is higher than that of CuO species [47],and the presence of CuO species is more likely to cause catalyst irreversible deactivation [48].In our work,without experiencing the ion exchange,the existence of CuO species in Cu-SSZ-130results in the irreversible deactivation except reversible deactivation and thereby the activity of Cu-SSZ-130is only partly restored after switching off the H2O and SO2streams.However,as except,after ion-exchange,CuO species in Cu-SSZ-13x(x ＞0)has been removed and more isolated Cu2+species are redistributed in D6Rs.Therefore,the activity of Cu-SSZ-13x(x ＞0) can be completely recovery after removing SO2and H2O.Fig.S4 (Supplementary Material)shows the UV Raman spectra of the Cu-SSZ-13xzeolites.As shown in Fig.S4A,the corresponding spectra exhibits two intense bands at 330 and 475 cm-1and another weak band at 350 cm-1.The intense band at 330 cm-1is assigned to the T-O-T (T＝Si or Al) vibration mode of the 6-membered rings of CHA structure;and the characteristic Raman band at about 475 cm-1is ascribed to the T-O-T vibration mode of CHA zeolite (4-member ring of CHA structure)[49,50].The presence of a band at 350 cm-1demonstrates that Cu-SSZ-130zeolites contains the Cu-O-Cu species,while the absence of corresponding band in the spectra of other samples indicates the improving Cu species distribution [51].According to the literature [51],the bands relate to Cu-O-Cu vibrations(350–600 cm-1region),which cannot be detected by 325 nm laser excitation.This is in line with our results (Fig.S4B).Interestingly,the intensity of the band at 475 cm-1in the spectrum of Cu-SSZ-130is quite lower when compared with other samples (treated with HNO3solution),which can be related the aggregation of CuO species in the cage or the surface of CHA zeolite[51].Besides,the intensity of the band at 330 cm-1decreases first and then increases with the increasing HNO3treatment only detected by 532 nm laser excitation,which may relate to the Cu species redistributed in D6Rs during the ion-exchange process [49–51].The results of Raman are quite in line with the above characterizations and further confirm the existence of various Cu species in different zeolite crystals.

Fig.11.29Si (a) and 27Al (b) MAS NMR spectra of Cu-SSZ-13x zeolites.

Fig.12.NH3-TPD profiles (a) and the corresponding deconvolution results (b) of Cu-SSZ-13x zeolites.

MAS NMR spectroscopy is commonly used to characterize the coordination environment of Si and Al species in zeolitic materials because these atoms are MAS NMR sensitive [52].For this reason,the29Si and27Al MAS NMR spectra of Cu-SSZ-13xwere collected to further probe the subtle changes in the coordination environments of Si and Al species during the ion-exchange process,which cannot be detected by XRD.As shown in Fig.11(a),the peaks at ca.-101,-107 and -113 ppm are assigned to Si(2Al),Si(1Al) and Si(0Al)groups,respectively [52].For analyzing the Si species in Cu-SSZ-13x,the collected spectra were carefully deconvoluted using the Gaussian fitting function and the corresponding results are given in Table 3.A clear downward trend can be observed on the relative contents of both Si(2Al) and Si(1Al) species.On the contrary,the relative content of Si(0Al) species increases monotonously with increasing ion-exchange time,and reaches up to 45.5% when the ion-exchange time was 16 h.In addition,the framework SiO2/Al2O3ratio of Cu-SSZ-130is 9.2,which is quite close to that of the initial aluminosilicate gel.With the extended ion-exchange time,the framework SiO2/Al2O3ratios of the resulting Cu-SSZ-13xincrease to 11.7,12.2,12.4,and 12.5,respectively,strongly evidencing the occurrence of dealumination during ion-exchange in the HNO3solution.This result is in well agreement with the change of bulk SiO2/Al2O3ratio of these samples (Table 1).As shown in Fig.11(b),the spectrum of Cu-SSZ-130has an exclusively strong peak at around 56 assigned to tetrahedral Al species [53],indicating that the Al atoms are fully incorporated into the zeolitic framework.However,two new peaks at ca.30 and -3,attributed to pentacoordinated and octahedral extra-framework Al atoms,respectively [35,42–44],emerge in the27Al MAS NMR spectra of the ion-exchanged Cu-SSZ-13xzeolites.The presence of extraframework Al atoms also demonstrates that dealumination really occurs,well in line with the29Si MAS NMR results.Janas et al.[54]proposed that isolated Cu2+ions could migrate to vacant sites generated by the removal of Al atoms from the framework.By combing the H2-TPR and EPR results,we can safely draw a conclusion that the removal of Al atoms provides suitable sites for redistribution of isolated Cu2+ions,leading to the relatively higher content of isolated Cu2+ions located in the D6Rs of Cu-SSZ-13 and thereby improving the SCR performance.Besides,the acidity of the zeolite is caused by the coordination imbalance of Al and O atoms,and the reduction of framework Al atoms will decrease the total acidity of the zeolite [55,56].

Since the acidity of zeolite-based catalysts significantly influences their catalytic performance,NH3-TPD tests were further carried out to examine the acidity change of the Cu-SSZ-13xzeolites with the increase in ion-exchange time [37].Fig.12(a) shows the NH3-TPD profiles of the Cu-SSZ-13xzeolites,and each profile was deconvoluted into several different sub-peaks.The desorption peak centered approximately at 240 °C is mainly related to the desorption of ammonia from weak Lewis acid sites [37].Those peaks between 300 and 500 °C are assigned to NH3desorption from Lewis acid sites related to Cu species such as [Cu(NH3)2]+and [Cu(OH)(NH3)3]+[55],and those at higher temperatures (500–800°C)are attributed to NH3desorption from strong Br?nsted acid sites [55,56].Fig.12(b) shows the acid amounts of the Cu-SSZ-13xzeolites calculated from Fig.12(a).Interestingly,Cu-SSZ-130has the lowest acid amount,which may be due to the formation of abundant CuOxspecies that cover the acid sites [1,57].With the increase in HNO3treatment time,the total acid amount of Cu-SSZ-13xzeolites first increases and then slightly decreases.Among these samples,Cu-SSZ-138owns the highest acid amount.This phenomenon may be caused by the first removal of excessive Cu species which makes more acid sites exposed after the HNO3treatment.Then,the removal of Al species from the framework was becoming the predominant event when further extending HNO3treatment time,leading to reduce total acid sites.Particularly,the total Lewis acid sites related to Cu species and the strong Br?nsted acid sites increase with increasing ion exchange time,and then reach their maxima when the ion-exchange time is 8 h.However,the total strong Br?nsted acid sites of Cu-SSZ-138are quite high.According to the literature [30],a high content of Br?nsted acid sites would lead to the occurrence of side reactions,reducing SCR activity at high temperatures.When the ionexchange time is longer than 8 h,the total amount of Br?nsted acid sites decrease,which would benefit for broadening temperature windows of Cu-SSZ-1312and Cu-SSZ-1316.The total Lewis acid sites related to Cu species have a slight decrease.According to the literature[55,56],the Cu species located at different sites exhibit diverse storage capacity.The NH3-TPD results of Cu-SSZ-13xshow that the ion-exchange time has a different influence on NH3storage capacity of Cu species,which is caused by the dealuminization and transformation of Cu sites during the ionexchange process [27,28].Combing the increasing of SiO2/Al2O3ratio confirmed by ICP-OES and MAS NMR with the results of NH3-TPD,the varying acidity of zeolites caused by coordination of Si and Al can further contribute to stabilizing of Cu species redistributed at different sites during ion exchange process [58].

4.Conclusions

In summary,we have reported an effective strategy to regulate the nature and distribution of Cu species of a one-pot synthesized Cu-SSZ-13 zeolite via ion exchange with a HNO3solution before removing the template.The obtained catalyst samples were comprehensively characterized by various techniques and tested in the NH3-SCR reaction.The results show that when the ionexchange time was 4 h,the obtained zeolite(Cu-SSZ-134)has dramatically decreased Cu and Na contents but increased specific surface,pore volume,Si/Al ratio and isolated Cu2+content,and thus shows a much wider reaction temperature window,a higher N2selectivity,and a lower NO2yield than the parent zeolite Cu-SSZ-130.When the ion-exchange time was further increased from 4 h to 16 h,the crystallinities,chemical compositions,and textural parameters of the resulting samples kept almost unchanged,but their catalytic activities were greatly enhanced.Among the various samples,Cu-SSZ-1312obtained by ion-exchange for 12 h exhibited the widest reaction temperature window(145–690°C),the lowest NO2yield,and the best SO2and H2O resistance.The systematical characterization results reveal that the redistribution of Cu species occurred during the ion-exchange,caused by the dealumination from outside to inside providing the sites for Cu2+ions location.It is exactly due to the redistribution of Cu species which makes the huge enhancement of both the SCR activity and antipoisoning performance of Cu-SSZ-13 zeolites.

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

We gratefully acknowledge the financial supports from National Natural Science Foundation of China (Nos.22178059 and 91934301),Natural Science Foundation of Fujian Province,China (2020J01513),Sinochem Quanzhou Energy Technology Co.,Ltd.(ZHQZKJ-19-F-ZS-0076),Qingyuan Innovation Laboratory(No.00121002),and Fujian Hundred Talent Program.

Supplementary Material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cjche.2021.10.027.