Shiyue Li,Shouying Huang,Kai Cai,Ying Li,Jing Lv,Xinbin Ma

Key Laboratory for Green Chemical Technology of Ministry of Education,Collaborative Innovation Center of Chemical Science and Engineering,School of Chemical Engineering and Technology,Tianjin University,Tianjin 300350,China

Keywords:Dimethyl ether Carbonylation Silver modification Metal location

ABSTRACT Mordenite (MOR) has shown great potential to catalyze dimethyl ether (DME) carbonylation to methyl acetate (MA) in industry.The synergy between metal species and Br?nsted acid sites accelerates DME conversion.Here we designed and prepared two catalysts with different Ag locations by seed-directed growth method and two-step impregnation method (named as Ag@HMOR and Ag/HMOR-out,respectively),to explain the effect of Ag location on catalytic performance.The results of TEM,XPS,CO-IR and UV–Vis showed that Ag species mainly presented as Ag0 species over both Ag@HMOR and Ag/HMOR-out.Meanwhile,Ag0 species mainly located in the micropores of Ag@HMOR,while as for Ag/HMOR-out,Ag0 mainly existed on external surface.After comparing the performance of the catalysts with different Ag positions,we confirmed that the Ag0 species encapsulated in the channels of HMOR promoted the DME carbonylation,which revealed the importance of spatial adjacency on the acid-metal catalysts.

1.Introduction

Ethanol,as a fuel alternative or additive,has attracted much attention due to its environmental friendliness.Recently,dimethyl ether (DME) carbonylation into methyl acetate (MA),which is a crucial step in ethanol production from syngas,is gaining more interest in both academic and industrial fields.H-form mordentite(HMOR),due to its unique spatial confinement in the 8-MR channel[1,2],is commonly applied and is considered as the most promising catalyst for DME carbonylation.It has been accepted that the intrinsic active sites are the Br?nsted acid sites in the 8-membered ring of MOR [1,3],and existing researches have recognized the critical role of transition metals in promoting HMORcatalyzed DME carbonylation.When metals (e.g.Cu,Zn,Co,Pt,Ag,etc.) are employed,the catalytic performance can be significantly enhanced[4–6].With regard to Ag,Ag0species with suitable size have been proved the metal active species that facilitate the reaction,while Ag+species do not contribute to the improvement of catalytic performance [7].Since the Ag species cannot convert DME to MA solely,in other words,a synergy between Ag and the Br?snted acid sites exists.Hence,the relative position or distance between the dual-active sites is crucial to the promotion effect.The similar synergism among metal species and zeolites has been reported in previous studies [8–11],but is still unclear over Ag modified HMOR catalyst on DME carbonylation.Therefore,through rationally designing and preparing HMOR catalysts with different Ag positions,exploring the impact of Ag location on DME carbonylation is of great significance for understanding the synergistically catalytic mechanism between Ag0active species and the Br?nsted acid sites.

Thus far,several studies have provided novel strategies for preparing metal clusters or nanoparticles encapsulated in zeolite pores.Iglesia et al.[12–14],employing metal ligand as precursor,hydrothermally synthesized zeolites like Linde Type A(LTA),Sodalite (SOD) with metal clusters (Pt,Pd,Ru,Ag,etc.) located in the cages or small pores.Corma et al.[15]used 2D MWW-type precursors and a surfactant to stable subnanometric Pt clusters in MCM-22.Xiao’s group[16]encapsulated metal nanoparticles(Pt,Pd,Rh,Ag)in zeolite using a controllable seed-directed growth method,in which the nanoparticles show great sinter resistance.These encapsulation approaches avoid metal species to deposit on the external surface of zeolite and enable them close to acid sites in the pores.On the other hand,some researches aim at placing metal clusters outside the pores of supports,which can be achieved by using a two-step impregnation strategy.For example,following by filling the inside with organic solvent,carbon nanotubes are impregnated with metal precursor solution.As a result,the metal nanoparticles can be selectively decorated on the outer surface of the nanotubes[17,18].Similarly,when the TS-1 zeolite is primarily stuffed with water and then mixed with Au complexes solution,Au nanoparticles would deposit on the external surface of TS-1 [19].

Herein,the seed-directed growth and two-step impregnation method were applied to prepare Ag modified HMOR with Ag species located in the pores or on the outer surface.TEM,XPS,COIR,and UV–Vis were used to compare the physical and chemical properties of Ag catalysts with different locations.Combining with the characterizations and catalytic test,the synergistic function among the Br?nsted acid sites and metal active species on DME carbonylation was further proposed.

2.Materials and Methods

2.1.Catalyst preparation

2.1.1.Preparation of Ag@HMOR catalyst

The Ag@HMOR sample was prepared by one-pot method via seed-directed hydrothermal synthesis[16].The commercial parent NH4-mordenite (NH4-MOR,Yangzhou Zhonghe Petrochemical Institute Co.,Ltd) zeolite was impregnated with 2 mL 0.2 mol·L-1AgNO3solution.After drying at room temperature overnight and drying at 353 K for 4 h,the sample was calcined at 673 K for 4 h,and then reduced at 573 K for 2 h in a 10 % H2/Ar atmosphere.The metal loading content of Ag/MOR seed was 2%(mass)as determined by ICP-OES.

After that,NaAlO2,NaOH,silica and H2O were mixed and the molar ratio of the gel components was 11 Na2O:1 Al2O3:26 SiO2:270 H2O.Then 0.4 g prepared Ag/MOR seed was added into the gel,and stirred until evenly.After crystallization at 413 K for 4 days,the product was filtered,washed,dried overnight and calcined at 823 K for 4 h.Then the calcined sample was mixed with 75 mL 0.2 mol·L-1NH4NO3solution for ion exchange under the temperature of 353 K.After the filtration,washing and drying,the sample was further calcined at 773 K and reduced under a flow of 10 % H2/Ar at 383 K.The obtained sample was labelled as Ag@HMOR.

2.1.2.Preparation of HMOR-s catalyst

0.64 g NaAlO2,1.12 g NaOH,4 g silica and 13.2 g H2O were mixed and stirred at room temperature for 12 h.The molar ratio of the gel components was 11 Na2O:1 Al2O3:26 SiO2:270 H2O,then added with 0.4 g MOR seed.The crystallization was carried out in Teflon-lined stainless-steel autoclave at 413 K for 4 days.The as-synthesized product was washed by deionized water,dried at 383 K overnight and calcined at 823 K for 4 h.The calcined Nazeolite was transformed into H-zeolite via ion exchange method as well.The obtained solid product was dried and further calcined at 773 K.The obtained sample was labelled as HMOR-s.

2.1.3.Preparation of Ag/HMOR-out catalyst

The prepared HMOR-s sample was impregnated with ethanol first,and then was mixed with AgNO3solution under stirring.Then the mixture was dried at room temperature overnight and transferred into a heating oven at 353 K for 10 h.The process should be careful to protect from exposure to light.The Ag/HMOR-out sample was synthesized after the calcination at 673 K for 3 h and the reduction in 10 % H2/Ar flow at 573 K for 2 h.

2.2.Catalyst characterization

Powder X-ray diffraction (XRD) patterns were conducted on a Rigaku D/max-2500 diffractometer.The device was equipped with a Cu Kα(λ=0.154 nm)radiation source,and the working condition was at 40 kV and 200 mA.All patterns were operated from 3° to 90° (2θ) with a scanning pace of 8 (°)·min-1.

Nitrogen adsorption–desorption isotherms were performed at 77 K on a Micromeritics ASAP 2460 device.Prior to the adsorption,all samples were reduced at 673 K for 3 h under a flow of 10 %H2/Ar and were outgassed at 623 K for 24 h.Brunauer-Emmett-Teller (BET) method,t-plot method and Horvath-Kawazoe equation were applied to determine the porous structure including the specific surface area,volume and pore size.

The elemental composition was determined by Inductively Coupled Plasma Optical Emission Spectrometer(ICP-OES,Varian Vista-MPX).All the samples were digested into the HF solution,and then coordinated by over-saturated H3BO3solution.

27Al MAS NMR was characterized on Varian Infinityplus 300 MHz spectrometer,under 78.1 MHz at a rate of 8 kHz.Excess Al(NO3)3aqueous solution was taken as a reference for the chemical shift.

Transmission electron microscopy(TEM)images were obtained using a JEM-2100F electron microscope.The reduced samples were dispersed into ethanol solution under sonication for 30 min and supported on carbon-coated copper grids.And the scanning electron microscopy (SEM) images were collected on Hitachi S-4800 equipment at 5 kV.

Diffuse reflectance ultraviolet–visible (UV–Vis) measurements were carried out on a UV–Vis Spectrometer (Shimadzu UV-2600),using BaSO4as a reference.

X-ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy (AES) were collected by a Thermo Fisher Scientific ESCALAB Xi+with monochromatic Al Kα (hν=1486.6 eV) X-ray radiation source.The binding energy of spectra were normalized by Si 2p peak at (103.5 ± 0.1) eV.

In-situ Fourier transform infrared (FTIR) spectra were carried out on Thermo Scientific Nicolet 6700 equipment with MCT/A detector.NH3and CO were used as probe molecules to quantify the total amounts of Br?nsted acid sites and Ag+species,respectively.The framework T–O–T vibration(T=Si,Al)in different channels was also recorded by FTIR.Prior to measurements,Ag modified HMOR samples (～20 mg) were reduced at 673 K with a flow of 10 % H2/Ar in the in-situ cell,and then were cooled to 423 K (for NH3) and 298 K (for CO).IR spectra were collected by scanning 32 times per spectrum at a resolution of 4 cm-1.

DME dissociation experiments were carried out on a flow reactor connected to HIDEN HPR20 quadrupole mass spectrometer.The samples were pretreated at 723 K for 1 h and cooled down to 473 K,then exposed to 5% DME/He flow (10 ml·min-1) for 1.5 h,during which the signals of methanol (m/e=32) and DME(m/e=45) were recorded.

2.3.Catalysis reaction

The DME carbonylation reaction was carried out in a fixed-bed reactor with 8 mm inner diameter.The catalysts of～500 mg,250–380 μm were placed in the reactor tube with silica sand above and below.The catalyst was activated at 673 K for 3 h with a flow of 10%H2/Ar or N2(only for HMOR-s sample).Subsequently the reaction was performed under the reaction conditions of 473 K and 1.5 MPa with CO/DME gases(CO/DME=1/49,volume ratio)blowing.The composition of products was analyzed online using a coupled gas chromatograph (Shimadzu GC-2014C Gas Chromatograph) equipped with a thermal conductivity detector(TCD)and a flame ionization detector(FID).Based on the previous calculations,the reactant conversion (XDME),product selectivity(SMA),product space–time yield (STYMA),and MA formation rate on per Br?nsted acid sites in 8-MR (named FRB) were calculated[4,7,20].

3.Results and Discussion

3.1.Structure and composition of catalysts

The crystal structures of HMOR-s and Ag modified samples were verified by XRD.Fig.1a shows the diffraction profiles of all samples,which exhibit the characteristics of MOR (JCPDS no.29-1257).SEM images (Fig.1c-e) showed that the particle size of all samples was approximately 1.5 μm.Meanwhile,the peaks of Ag0particles at 38 ° and 44.5 ° are not present in the XRD patterns,indicating that the Ag species were highly dispersed on external surface or in channels of MOR.The composition of samples is listed in Table 1.The Si/Al ratio of all samples is around 4,and the Ag content for Ag promoted samples is around 0.2% (mass).The adsorption and desorption isotherms of the different samples at 77 K(Fig.1b) are all type Ⅰisotherms.This confirms that all samples mainly consist of micropores.Some small hysteresis loops are also present at higher relative pressure P/P0,which means the existence of few mesopores.The detailed structural parameters including specific surface area and pore volume are listed in Table 1.The specific surface area and pore volume of Ag@HMOR sample are slightly larger,but overall,both pore properties of the three samples do not differ notably.The addition of Ag does not affect the pore structure of HMOR zeolite,meaning that introduction of Ag species neither block the channels of MOR nor damage the zeolitic framework.This agrees with the well-structured crystallization of samples displayed in XRD,which is probably because of the low Ag loading on HMOR.

Acid properties especially Br?nsted acid,as the essential and critical identity of HMOR over DME carbonylation,can be quantified by in-situ FTIR technique.NH3with the small kinetic diameter(0.26 nm),is able to diffuse into both 12-MR and 8-MR channels of MOR and interact with the acid sites.Thus,the total amount of Br?nsted acid sites (Btotal) of all samples can be determined by NH3-adsorbed IR.As shown in Fig.2a,the band at 1430 cm-1is accredited to the N–H vibration of NH3molecules adsorbed on Br?nsted acid site.It seems that the shape and position of the band for the Ag modified HMOR and unmodified HMOR-s are similar.We calculated the Btotalof the samples by using the reported extinction coefficients[21]and summarized the results in Table 2.The Btotalof Ag@HMOR is 1604 μmol·g-1,which is slightly higher than 1509 μmol·g-1of HMOR-s and 1469 μmol·g-1of Ag/HMORout.

The bands within the range of 3500–3800 cm-1provides information about different types of hydroxyl groups in MOR.As presented in Fig.2b,there are four overlapped peaks in the hydroxyl region,which center at 3739 cm-1,3660 cm-1,3607 cm-1and 3587 cm-1.Previous studies have attributed these bands to the O–H vibration of terminal silanol and extra-framework Al species,and the O–H groups bonded to framework Al (H+–Si(O)Al-) in the 12-MR channel and 8-MR channel,respectively[22,23].After peak fitting and integral,the relative amount of each hydroxyl species can be obtained.Further,combining with the Btotalobtained from the NH3-adsorbed IR,the amounts of different hydroxyl species were calculated and listed in Table 2.Compared with the HMORs sample,both the amounts of Br?nsted acid sites in 8-MR channel(B8-MR) and in 12-MR channel (B12-MR) of Ag@HMOR sample increased.This is in agreement with the variation trend of Btotaland might result from the deviation caused by hydrothermal synthesis.While those of Ag/HMOR-out slightly decreased in comparison with HMOR-s,which might be caused by the calcination treatment after impregnation.However,the B8-MR/B12-MRratio of each sample is similar.This indicates that Ag modification and the location of Ag species do not affect the distribution of Al in different channels of MOR.27Al MAS NMR were further employed to characterize the structure of Al atoms,as shown in Fig.3.The chemical shift at 55 with an intense peak is attributed to tetrahedrally coordinated framework Al [24],while the chemical shifts at 0 and 38 are assigned to the octahedrally extra-framework Al and penta-coordinated Al respectively[25].The results of peak fitting are summarized in Table 2.The content of framework Al in the three samples is basically around 78%,meaning that the addition of Ag species via the hydrothermal crystallization has no significant effect on the formation of topological structure,and the two-step impregnation treatment does not abstract Al atoms from the zeolite framework.The results of27Al MAS NMR coincide with those observed from IR spectra in the O-H stretching region(see Table 2,the columns of AlF,Al’F,AlEFand Al’EF).Overall,Ag modified samples prepared by two methods have the approximate ratio of framework Al to extra-framework Al,as well as the basically same distribution between B8-MRand B12-MR,showing that the structure of HMOR can be maintained during Ag modification.

Fig.1.XRD profiles (a),N2 adsorption and desorption isotherms (b) of prepared samples,and SEM images for HMOR-s (c),Ag@HMOR (d) and Ag/HMOR-out (e).

Table 1 Textural properties of HMOR-s and Ag modified samples

Fig.2.NH3-IR (a) and IR spectra in the O-H stretching region (b) of HMOR-s and Ag modified samples.

Table 2 The quantity of acid sites and Al species of HMOR-s and Ag modified samples

Fig.3.27Al MAS NMR of HMOR-s and Ag modified samples.

3.2.Position and chemical state of silver species

Fig.4 provides the TEM images of Ag modified samples,displaying that Ag particles were well dispersed with the average particle size of (2.4 ± 0.5) nm for Ag@HMOR,and (3.3 ± 0.5) nm for Ag/HMOR-out.It is noteworthy that some Ag particles can be obviously observed outside the Ag/HMOR-out surface,but not found on Ag@HMOR sample.This suggests that compared with Ag@HMOR sample,Ag/HMOR-out made by two-step impregnation method has a considerable amount of Ag species dispersed outside the surface.Conjointly,it also shows that the catalysts with different Ag positions are successfully acquired through different preparation methods.

Fig.4.TEM images of Ag@HMOR (a) and Ag/HMOR-out (b),and particle size distribution of Ag@HMOR (c) and Ag/HMOR-out (d).

Fig.5.XPS spectra of Ag@HMOR (a) and Ag/HMOR-out (b) samples (Ag 3d core level transitions).

With the aim of further determining the states of Ag species,Ag 3d spectra are shown in Fig.5.Sputtering is used to detect the Ag species at different depths.It is apparent from Fig.5a that the peaks of Ag@HMOR at Ag 3d3/2and Ag 3d5/2[26,27]without sputtering are negligible,which demonstrates that there are few Ag species on the surface.In contrast,as for Ag/HMOR-out in Fig.5b,the peaks of Ag 3d3/2and Ag 3d5/2with the binding energies of 375.4 eV and 369.2 eV can be evidently noticed,indicating the existence of a certain number of Ag species on the surface of Ag/HMOR-out.After sputtering the surface for 1 min,the peak intensity of Ag@HMOR at Ag 3d3/2and Ag 3d5/2becomes pronounced,and the binding energy is 375.7 eV and 369.5 eV respectively.On the contrary,the peaks of Ag/HMOR-out in this region declined after sputtering for 1 min.The element composition of Ag with and without sputtering is summarized in Fig.5.The Ag atomic content of Ag@HMOR on the surface was not detectable,but it raised to 0.01 % (atom) after sputtering,revealing that Ag species were mainly dispersed inside the pores of Ag@HMOR during the hydrothermal synthesis.On the other hand,regarding Ag/HMORout,the Ag atomic content without and with sputtering is 0.15 %(atom) and 0.03 % (atom),respectively.It verifies that Ag species preferentially exist on the external surface of the impregnated sample(Ag/HMOR-out).The above results are well consistent with the position of Ag species observed by TEM,confirming that the different synthesis methods of Ag modified catalysts effectively regulated the location of Ag species.

Fig.6.CO desorption IR spectra of Ag modified samples:Ag@HMOR (a),Ag/HMOR-out (b).

Fig.7.UV–Vis profiles of HMOR-s and Ag modified samples.

To further investigate the nature of Ag species in Ag@HMOR and Ag/HMOR-out catalysts,CO was used as a probe molecule in IR experiments.Fig.6 shows the IR spectra of CO adsorption under the flow of helium at room temperature.The excessive gaseous CO shows the intense and broad peaks in the region from 2040 cm-1to 2240 cm-1[28].With He purging,the bands decrease rapidly.As previous study reported[7],the chemisorbed CO on Ag+species in Ag modified HMOR results in a band located at 2189 cm-1,and no adsorbed CO on Ag0species could be observed due to the weak interaction between CO and Ag0at room temperature.It can be seen from Fig.6a that the characteristic peaks of CO decline with time and almost completely disappear,which indicates that there is no Ag+species interacting with CO over Ag@HMOR.That is,there only exist Ag0species in Ag@HMOR synthesized by the seed-directed growth method.Similarly,there is also no Ag+species in Ag/HMOR-out sample.In our previous work,Ag catalysts prepared by ion exchange method are difficult to be completely reduced especially at low Ag content,so the Ag modified HMOR comprises Ag0species as well as Ag+species [7].Compare to ion exchange method,the two methods here are productive to prepare Ag-HMOR catalyst with Ag0species even at low metal loading.This is because the sample was firstly impregnated with ethanol,which fully occupied the pores of zeolite and prevented the metal species from depositing on or ion-exchanging inside the pores.Since the exchangeable Br?nsted acid protons on the external surface are very limited,the quantity of ionexchange species in oxidation state prepared by two-step impregnation method is negligible at low metal loading.Therefore,it provides possibility to investigate the promotion effect of Ag0species by eliminating the interference of inactive Ag+species,and the utilization efficiency of metal is improved.Combining TEM and XPS results towards the position of Ag species,there can be concluded that Ag species in Ag@HMOR are mainly encapsulated in the pores of HMOR as Ag0species,while that in Ag/HMOR-out are located on the external surface of HMOR.

Furthermore,Ag0species were further subdivided by using UV–Vis spectroscopy.As shown in Fig.7,the absorption bands of the three samples appear at 200–500 nm.As for HMOR-s sample,two main peaks located at 221 nm and 275 nm,are attributed to the characteristics of the MOR framework.It should be noted that the former peak at around 221 nm is also overlapped with the characteristic band of Ag+ions [29].Compared with HMOR-s,the bands of Ag modified samples are more complex,with new peaks emerging at 270 nm,356 nm and 397 nm.The band at 270 nm is attributed toclusters,and the bands at 356 nm and 397 nm are assigned to Ag0nanoparticles with larger size [30].Compared with Ag/HMOR-out,the band of Ag@HMOR at 270 nm is much more obvious,which suggests that there are moreclusters in Ag@HMOR sample than Ag/HMOR-out.This is because Ag clusters were encapsulated in the pores of MOR,which inhibited the aggregation and growth of Ag species.Moreover,both the Ag modified samples show similar band intensity at 356 nm and 397 nm,indicating Ag0nanoparticles are present on the two samples and have the similar particle size.The results of UV–Vis spectra verify that Ag species exist mainly asclusters and Ag0nanoparticles in Ag@HMOR and Ag/HMOR-out,which is accordant with CO-IR.Note that Ag@HMOR comprises moreclusters with smaller size,so the average size of Ag species is smaller in Ag@HMOR than that of Ag/HMOR-out.This is consistent with the particle size statistics of TEM.Accordingly,we conclude that the agglomeration of Ag species can be effectively limited in the channels of MOR.

Fig.9.DME dissociation on HMOR and Ag@HMOR (1 Torr=133.3 Pa).

3.3.Promotion on catalytic performance

Fig.8 compares the DME conversion,MA yield and MA selectivity among HMOR-s and Ag modified catalysts.Initially,the DME conversion of all samples shows an upward trend and reaches the highest activity at 1.5 h.It is clear that the Ag@HMOR catalyst outperform the other two catalysts.The rapid deactivation is due to the coke deposition [31,32].Fig.8b presents that the MA yield of Ag@HMOR at 1.5 h is 209 g·(kg·cat)-1·h-1,while that of Ag/HMOR-out and HMOR-s is 117 g·(kg·cat)-1·h-1and 140 g·(kg·cat)-1·h-1.This shows that Ag@HMOR encapsulated Ag0species accelerates the DME carbonylation,while Ag/HMOR-out prepared after two-step impregnation of HMOR-s exhibits slightly lower MA yield than HMOR-s.It suggests the crucial role of Ag0location.The Ag0species on the outer surface hardly promote the DME carbonylation.Instead,Ag0particles might cover the Br?nsted acid sites or hinder the diffusion of molecules such as DME,resulting in a slight decrease in catalytic activity.Meanwhile,all catalysts maintain high and equivalent MA selectivity(＞99%),implying that the addition of Ag does not damage the structure and change the confinement effect in 8-MR channel of MOR.This is consistent with the conclusion obtained from structure characterization.

Since the Br?nsted acid sites are the active sites on DME carbonylation,it is necessary to exclude the effect of acid property to clarify the role of Ag species during the reaction.The MA yield was normalized based on the amount of Br?nsted acid in 8-MR,and labeled as FRB(Fig.8b).The FRBof Ag@HMOR is 3.0 h-1,which is significantly higher than that of HMOR-s (2.1 h-1) and Ag/HMOR-out (1.8 h-1).This demonstrates that Ag0clusters located inside the pores improve the DME conversion,while Ag0species on the outer surface have no contribution to the reaction.This proves that the synergetic effect occurs only if the metal sites are spatially adjacent to the Br?nsted acid site [9].When Ag0species on the external surface of MOR is far away from the Br?nsted acid sites,the synergetic promotion on DME carbonylation would not happen over the metal modified HMOR.In some published work,the extra-framework metal species with oxidation state (e.g.Cu+,Cu2+,Ni2+,Co2+) were proposed to act as Lewis acid sites those facilitate CO adsorption [33].Our previous work claimed that Cu0promoted the reaction by changing the pathway of DME dissociation and CO insertion[9].Here,we compared DME dissociation on HMOR and Ag@HMOR and found that DME dissociation is accelerated on Ag@HMOR (Fig.9).Combining with the weak interaction between CO and Ag0(Fig.6),it is speculated that Ag0species work together with the neighboring Br?nsted acid sites and change the pathway of DME dissociation rather than strengthen CO adsorption.But the detailed promotion mechanism needs more experimental and theoretical study.

4.Conclusions

Through seed-directed hydrothermal synthesis and two-step impregnation method,Ag@HMOR and Ag/HMOR-out catalysts with different Ag locations were prepared.XRD and physical adsorption showed that Ag modification retained the wellstructured crystallization and pore structure of HMOR.The27Al MAS NMR and in-situ FTIR proved that Ag addition by the two methods had a slight effect on the total amount of Br?nsted acid sites,but did not alter the distribution of Al in different channels.TEM,XPS,CO-IR and UV–Vis revealed that Ag species of Ag@HMOR and Ag/HMOR-out were well dispersed mainly as Ag0species.Ag0species over Ag@HMOR were present in the pores of MOR with smaller particle size,while Ag0species over Ag/HMOR-out existed on the external surface.Combining with the reactivity of DME carbonylation and the formation rate of MA per B8-MR(FRB) over Ag@HMOR,we concluded that Ag0species encapsulated in pores instead of on external surface can promote the DME carbonylation,which supports the significance of the spatial adjacency on the dual active sites.This provides an insight into the synergistic promotion mechanism of metal species and Br?nsted acid over metalmodified 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

This work was supported by the National Natural Science Foundation of China(21978209,21325626),the Program of Introducing Talents of Discipline to Universities (BP0618007).