Yefei Liu ,Yang Zou ,Hong Jiang ,Huanxin Gao ,Rizhi Chen ,*

1 State Key Laboratory of Materials-Oriented Chemical Engineering,Nanjing Tech University,Nanjing 210009,China

2 SINOPEC Shanghai Research Institute of Petrochemical Technology,Shanghai 201208,China

1.Introduction

Cumene is an important chemical feedstock for the production of phenol and acetone in industry.Cumene is firstly oxidized to cumene hydroperoxide,and then phenol and acetone are produced by the acid arrangement of cumene hydroperoxide.The cumene-to-phenol route contributes more than 90%of phenol capacity in the world[1,2].In recent years,the rapidly growing market of phenol increases the demand for cumene[3,4].

The production of cumene in industry is usually conducted by using the alkylation ofbenzene with propylene as the alkylating agentin fixed bed reactors.Since the solid phosphoric acid and aluminum chloride catalysts cause corrosion and environmental problems,several zeolitebased catalysts have been developed for producing cumene[5–7].At present,some cumene production processes have been commercialized with success,e.g.Dow/Kellogg process,UOP/Q-Max process,EniChem process and Mobil/Badger process[7].For example,in the Dow/Kellogg process,the dealuminated mordenite was firstly used as the alkylation catalyst,in which a large number of meso-scale pores enhanced the mass diffusion and reduced the tar formation[8].The UOP/Q-Max process used the metal ion-containing beta-zeolite as the alkylation catalyst.By adjusting the content of metal ions in beta-zeolite,then-propylbenzene formation was reduced and the catalyst functioned with good stability[9].Although the alkylation of benzene with propylene is an industrially importantreaction for the cumene production,the use of propylene as the alkylating agent readily results in the coke deposits on alkylation catalysts.Besides,the presence of acidic catalysts promotes the polymerization of propylene and thus large excess of benzene is needed to prevent propylene polymerization[10].

Instead of using propylene as the alkylating agent,the use of isopropanol to alkylate benzene has been studied as a promising route to produce cumene[11].Acetone as the byproduct in the cumene-tophenol route can be recycled by reduction reaction to isopropanol,and then isopropanol directly reacts with benzene to produce cumene.The economy of this process can be improved due to the recycling of acetone.Moreover,the alkylation of benzene with isopropanol is more energy-efficient than that using propylene.

Unfortunately,there are few reports on the alkylation of benzene with isopropanol.Girottiet al.[12]used isopropanol as the alkylating agent in the direct alkylation of benzene to cumene over the betazeolite catalyst,and achieved similar reaction performance with the propylene route.It was also found that the catalyst activity and catalyst deactivation rate were in fluenced by the water content in the reaction mixture,and high water content was not favorable for maintaining the catalytic performance.Deactivation of beta-zeolite catalyst was observed in some catalytic reactions[12–16],however,the catalytic stability and deactivation mechanism of beta-zeolite catalyst in the alkylation of benzene with isopropanol were still not investigated in detail.

In this study,the catalytic stability of the beta-zeolite catalyst for benzene alkylation was investigated by using a submerged ceramic membrane reactor operated in a semi-batch mode.The deactivation mechanism of beta-zeolite catalyst was explored based on a series of characterization techniques including XRD,SEM,TEM,TG,BET,NH3-TPD and GC–MS.And,the catalyst regeneration was carried out.

2.Experimental

2.1.Materials

The beta-zeolite catalyst was provided by SINOPEC Shanghai Research Institute of Petrochemical Technology,China.Benzene was purchased from Shanghai Lingfeng Chemical Reagent Co.,Ltd.,China.Isopropanol was obtained from Shanghai Shenbo Chemical Co.,Ltd.,China.Methanol(＞99.9%chromatography grade)was supplied by Yuwang Group,China.Ethanolwas provided by WuxiYasheng Chemical Co.,Ltd.,China.n-Hexane was purchased from Shanghai Shisihewei Chemical Co.,Ltd.,China.All materials were used without further treatment.

2.2.Benzene alkylation

A submerged ceramic membrane reactor system was developed for alkylation of benzene with isopropanol to produce cumene,as shown in Fig.1.The system mainly consisted of an autoclave,a ceramic membrane module and nitrogen source.The autoclave was made of stainless steel with a working volume of 1 L.The ceramic membrane made up of a fine layer of ZrO2(nominal pore size of 200 nm)was provided by Nanjing Jiusi High-Tech Co.,Ltd.,China.

Fig.1.Schematic diagram of the submerged ceramic membrane reactor system.

Typically,3.5 g of beta-zeolite catalyst,67 ml of benzene,19.5 ml of isopropanoland 385 mlofn-hexane were introduced into the autoclave reactor in sequence.The autoclave was sealed and purged with N2for five times to evacuate air.Subsequently,the sealed autoclave was pressurized to 1.0 MPa and heated.The stirring rate was controlled to be 800 rpm.When the temperature reached 180°C,N2was fed into the reactor to adjust the pressure to 3.0 MPa and then the benzene alkylation reaction with isopropanol got started.Each reaction lasted for 3 h.After reaction,the stirring was stopped and the reactor was cooled down to 60°C.The pressure was adjusted to 0.3 MPa to drive the membrane filtration process.The fine beta-zeolite catalyst could be separated from the liquid phase mixture[17,18]and retained in the reactor for next cycle of alkylation reaction.

To investigate the catalytic stability of the beta-zeolite catalyst,after the membrane filtration,the freshn-hexane solution of benzene and isopropanol were added to the reactor,and the processes of benzene alkylation and membrane filtration were repeated.Due to the vertical setting of membrane module in the autoclave as shown in Fig.1,parts of products were remained in the reactor and could in fluence the reaction.In further work,we will design horizontal membrane module,and make the membrane locate in the bottom of the reactor and more products be exported from the reactor.When all reaction tests were finished,the beta-zeolite catalyst used was collected for various characterizations to analyze the catalyst deactivation mechanism.

2.3.Analysis

The reaction products were analyzed by an HPLC system(Agilent 1200 Series,USA).The benzene conversion was de fined as the ratio of concentration of all products and initial concentration of benzene,while the selectivity of cumene was calculated as the concentration ratio of cumene in all products.

The X-ray powder diffraction(XRD)patterns of the beta-zeolite catalyst were recorded on a Rigaku MiniFlex600 diffractometer using CuKαradiation at 40 kV and 15 mA with the 2θ range of 5°–80°.The morphology ofthe beta-zeolite catalystwas examined by field emission scanning electron microscope(FESEM,Hitachi S-4800II,Japan).To verify the distribution and morphology of catalyst particles,the transmission electron microscope(TEM,Philips Tecnai 12,The Netherlands)was adopted.The thermogravimetric analyzer(TG,NetzschSTA449 F3 Jupiter,Germany)was used to study the thermalstability ofthe catalyst.During the TG analysis,the air temperature was increased from room temperature to 800 °C by 10 °C·min?1.The N2adsorption–desorption isotherms were measured by using an ASAP 2020 analyzer(Micromeritics,USA)at its normal boiling point(77 K).The specific surface area of beta-zeolite was calculated by the Brunauer–Emmett–Teller(BET)method and the pore volume was estimated by the Barrett–Joyner–Halenda(BJH)method.The temperature-programmed desorption of ammonia(NH3-TPD)was used to determine the acid amount and the acidities of the catalysts by a BELCAT-A equipment connected to a thermal conductivity detector(TCD).The composition of organic matters deposited in the beta-zeolite catalyst during the benzene alkylation was analyzed by gas chromatography with massspectrometric detection(GC–MS).Typically,0.06 g of used catalyst was carefully dissolved in 5 ml of 1mol·L?1HF solution.The solution was treated with 5 ml of toluene to extract the organic matters,and then the organic extract was analyzed by GC–MS(Agilent 7590B–5977A).

3.Results and Discussion

3.1.Catalytic stability of beta-zeolite catalyst

To investigate the catalytic stability of beta-zeolite catalyst,a number of catalytic reaction cycles were carried out in the submerged ceramic membrane reactor system.Fig.2 presents the relative benzene conversion and cumene selectivity during catalytic reaction cycles,which are expressed by the ratios of the benzene conversion and cumene selectivity after a certain number of reaction cycles to those during the first reaction cycle.It is seen that the cumene selectivity increased through six reaction cycles.This may be due to the fact that the di-isopropylbenzene as the byproduct further reacts with the excessive benzene to produce cumene[19].Unfortunately,the benzene conversion presented the sharp decrease by 90%through six reaction cycles.It is indicated that the severe deactivation of the beta-zeolite catalyst took place.

Fig.2.Change of benzene alkylation reaction performance with the number of catalytic reaction cycle.

Fig.3.XRD patterns of beta-zeolite catalysts:(a)fresh;(b)after the first cycle;and(c)after the sixth cycle.

3.2.Catalyst characterizations

To explore the deactivation mechanism of beta-zeolite catalyst,the fresh and used catalysts were characterized by XRD,SEM,TEM,TG,BET and NH3-TPD techniques.

The XRD technique was adopted to analyze the microscopic crystalline structures.In Fig.3,no obvious differences were observed for the crystal structure,crystalline size and peak intensity before and after reactions.The(101)and(302)diffraction peaks were presented at 2θ =7.8°and 22.5°,respectively.These peaks are characteristics of beta-zeolite thatare obtained from the tetragonaland monoclinic intergrowth crystals[20].The XRD pattern shows that the crystallinity and skeletal structure of the beta-zeolite catalyst were well-maintained and the catalyst still had good hydrothermal stability through six reaction cycles.

Figs.4 and 5 show the SEM and TEM images of the fresh beta-zeolite catalyst,the catalyst after the first cycle and the catalyst after the sixth cycle.By comparing with the fresh catalyst under different magnifications,the morphology of the catalysts after the first and sixth cycles had no obvious change.The nanoparticles and their agglomerates were observed.The TEM image with high resolution presents that the zeolite was composed of particles of approximately 20 nm in diameter,and the lamellar nanocrystals stacked randomly.This TEM finding was consistent with that reported in the literature[21].By the XRD,SEM and TEM analyses,the alkylation of benzene with isopropanol did not affectthe crystalline and particle size of the beta-zeolite catalyst.Therefore,it is concluded that the catalyst deactivation was not attributed to the changes in crystalline and particle size of the catalyst.

The thermogravimetric analysis curves of the fresh and used catalysts are presented in Fig.6.The mass loss of all catalyst samples below 150°C was due to the desorption of water.The minor mass loss of the fresh catalyst was observed when the temperature was increased from 200 to 800°C,which resulted from the residual water in the catalyst[12].However,the used catalysts presented pronounced mass loss and the mass loss of the catalystafter the sixth cycle was larger than that after the first cycle.Clearly,the difference in mass loss was caused by the decomposition of organic materials having different amount[22].Meanwhile,from Fig.7,the color of the catalyst became deepened as the number of reaction cycles increased,indicating that the organic matters like polycyclic aromatic hydrocarbons were produced in the alkylation ofbenzene with isopropanol[23].As a result,the catalyst was deactivated due to the deposition of carbonaceous materials on the catalyst.

Fig.4.SEM images of beta-zeolite catalysts:(a,d)fresh;(b,e)after the first cycle;and(c,f)after the sixth cycle.

To further verify the finding from the thermogravimetric analysis,the BET tests were carried out to check the changes in catalyst pores.Fig.8 gives the isotherms of nitrogen adsorption–desorption on different catalysts.At the low pressure,all the samples exhibited the I-type isotherms typical of microporous materials[24].At the high pressure,i.e.p/p0＞0.7,the obvious hysteresis loops appeared,which was the characteristic of mesoporous materials[25].The results showed that the fresh and used beta-zeolite catalysts had well-developed microporous and mesoporous structures.In Fig.9,the pore size distribution also indicated that the beta-zeolite had good crystalline and hydrothermal stability during the alkylation cycles.However,compared to the fresh catalyst,the catalysts after the first and sixth cycle had drastic drop in the N2adsorption amount.In contrast to the fresh catalyst,the specific surface area and pore volume of the used catalysts significantly decreased,and the decrease trend was more obvious by increasing the number of reaction cycles(Table 1).For example,after the sixth cycle,the specific surface area and pore volume decreased by 33.2%and 29.2%,respectively.These results suggested that some organic matters would deposit on the beta-zeolite catalyst during the alkylation cycles,leading to small specific surface area and few catalytic active sites.As a result,the catalyst deactivation occurred and the benzene conversion decreased.

Fig.5.TEM images of beta-zeolite catalysts:(a)fresh;(b)after the first cycle;and(c)after the sixth cycle.

Fig.6.TG curves of beta-zeolite catalysts:(a)fresh;(b)after the first cycle;and(c)after the sixth cycle.

Fig.8.Adsorption–desorption isotherms of beta-zeolite catalysts:(a)fresh;(b)after the first cycle;and(c)after the sixth cycle.

Fig.7.Powders of beta-zeolite catalysts:(a)fresh;(b)after the first cycle;and(c)after the sixth cycle.

The acidities of the fresh and used catalysts were detected by NH3-TPD,as shown in Fig.10.For the fresh catalyst,two NH3desorption peaks were observed in the range of 260–450 °C and 500–650 °C,respectively,indicating the weak and medium acid sites existed on the catalyst[26].Furthermore,the number of the weak acid sites was much larger than that of the medium acid sites.Obvious differences could be observed in the NH3-TPD patterns of the three samples.By calculating peak areas,itwas found thatthe amountofbeta-zeolite catalyst acidity gradually decreased from 14.88mmol·g?1of the fresh catalyst to 8.55mmol·g?1of the catalyst after the sixth cycle(Table 2).In particular,the decrease by 33.7%was so pronounced for the catalyst after the first cycle.In addition,the desorption peak temperature became lower when the number of reaction cycles increased as presented in Table 2.These changes mightbe related with the adsorption oforganic matters on the catalyst and/or the thermal deactivation of catalyst at high reaction temperatures[27].From the NH3-TPD tests,it was confirmed that the decrease in the amount of acidity was also responsible for the poor catalytic performance of beta-zeolite catalyst during the alkylation cycles.

To understand well the deactivation mechanism of the beta-zeolite catalyst,the exact composition of the organic matters deposited in the used catalyst after the sixth cycle was analyzed by GC–MS after dissolution of catalyst framework and extraction of the organics[28,29].As shown in Fig.11,the deposited organic matters mainly consisted of ethylbenzene,p-xylene and 1-ethyl-3-(1-methyl)benzene.These matters were possibly responsible for the deactivation ofbenzene alkylation over beta-zeolite catalyst.

Fig.9.Pore size distribution of beta-zeolite catalysts:(a)mesopores;and(b)micropores.

Table 1Specific surface area and pore volume of beta-zeolite catalysts

Fig.10.NH3-TPDdiagrams ofbeta-zeolite catalysts:(a)fresh;(b)used afterthe firstcycle;(c)used after the sixth cycle;(d)regenerated with calcination;and(e)regenerated with ethanol rinse.

Table 2NH3-TPD test of beta-zeolite catalysts

3.3.Regeneration of beta-zeolite catalyst

Since the deactivation mechanism of beta-zeolite catalyst has been found owing to the adsorption of organic matters on the catalyst surface,the catalyst regeneration was carried out by ethanol rinse and calcination,respectively to remove the organic matters[22,30].The deactivated catalyst was rinsed repeatedly for three times by ethanol solution,and then the rinsed catalyst was filtered and dried at 150°C overnight to get the regenerated catalyst.With regard to the catalyst calcination,the deactivated catalyst was directly calcinated in air atmosphere at 500°C for 10 h to get the regenerated catalyst.The BET and NH3-TPD were performed to analyze the regenerated catalysts.Also,the benzene alkylation was carried out by using the regenerated catalysts to evaluate the validity of the regeneration methods.

It is seen from Table 1 that the catalyst regenerated by calcination had similar surface area and pore volume with the fresh one.However,the ethanol rinse did not effectively recover the surface area and pore volume.Correspondingly,as shown in Fig.10,the amount of acidity could be easily recovered by calcination as compared to ethanol rinse.The catalytic properties of regenerated catalysts were presented in Fig.12.With respect to the catalyst regenerated by calcination,the benzene conversion had been increased to 23.6%and the selectivity of cumene had reached 90.5%,which were similar with the values for the fresh catalyst.However,the catalyst regenerated by ethanol rinse still had poor catalytic performance in the benzene alkylation reaction.By comparison,the calcination method could effectively regenerate the beta-zeolite catalyst,in agreement with the reported results[31].It is also observed form Table 1 that calcination could recover the micropores,and ethanolrinse could regenerate the mesopores.Together with the results in Fig.12,we might hypothesize thatthe benzene alkylation and the deactivation occurred in the micropores of beta-zeolite catalyst.

Fig.11.GC–MS analysis of the organic matters deposited in the beta-zeolite catalyst after the sixth cycle.

Fig.12.Effect of catalyst regeneration on the conversion and selectivity of benzene alkylation reaction.

4.Conclusions

The physicochemical properties of the fresh and used beta-zeolite catalysts were characterized to explore the deactivation mechanism for the alkylation of benzene with isopropanol.The organic matters mainly consisted of ethylbenzene,p-xylene and 1-ethyl-3-(1-methyl)benzene formed in the benzene alkylation deposited in the catalyst and reduced the specific surface area.Meanwhile,the amountof acidity largely decreased during the reaction.Thus,the decrease in specific surface area and acidity resulted in the catalyst deactivation and the decrease in benzene conversion.Based on the understanding on catalyst deactivation mechanism,the catalyst regeneration by the calcination method was found to be an effective way to maintain high catalyst activity and benzene conversion.

[1]Y.H.Bao,H.Jiang,W.H.Xing,R.Z.Chen,Y.Q.Fan,Liquid phase hydroxylation of benzene to phenol over vanadyl acetylacetonate supported on amine functionalized SBA-15,React.Kinet.Mech.Catal.116(2)(2015)535–547.

[2]H.Jiang,F.She,Y.Du,R.Z.Chen,W.H.Xing,One-step continuous phenol synthesis technologyviaselective hydroxylation of benzene over ultra fine TS-1 in a submerged ceramic membrane reactor,Chin.J.Chem.Eng.22(11–12)(2014)1199–1207.

[3]C.C.Huang,J.J.Peng,S.H.Wu,H.Y.Hou,M.L.You,C.M.Shu,Effects of cumene hydroperoxide on phenol and acetone manufacturing by DSC and VSP2,J.Therm.Anal.Calorim.102(2)(2010)579–585.

[4]G.D.Yadav,Z.S.Asthana,Selective decomposition of cumene hydroperoxide into phenol and acetone by a novel cesium substituted heteropolyacid on clay,Appl.Catal.A Gen.244(2)(2003)341–357.

[5]M.H.Han,S.X.Lin,E.Roduner,Study on the alkylation of benzene with propylene over Hβ zeolite,Appl.Catal.A Gen.243(1)(2003)175–184.

[6]Z.G.Lei,C.N.Dai,Y.L.Wang,B.H.Chen,Process optimization on alkylation of benzene with propylene,Energy Fuel23(6)(2009)3159–3166.

[7]W.Tecza,R.Brzozowski,Modern industrial processes for making ethylbenzene and cumene,Przem.Chem.82(11)(2003)1428–1434.

[8]J.Horsley,Producing bulk and fine chemicals using solid acids,ChemTech27(10)(1997)45–49.

[9]G.J.Gajda,R.L.Patton,S.T.Wilson,Discrete Molecular Sieve and Use in Aromatic-Ole fin Alkylation,UOP Inc.,1995(US 5434326).

[10]S.Barman,N.C.Pradhan,J.K.Basu,Kinetics of alkylation of benzene with isopropyl alcohol over Ce-exchanged NaX zeolite,Ind.Eng.Chem.Res.44(19)(2005)7313–7319.

[11]V.V.Bokade,U.K.Kharul,Selective synthesis of cumene by isopropylation of benzene using catalytic membrane reactor,Chem.Eng.J.147(2–3)(2009)97–101.

[12]G.Girotti,F.Rivetti,S.Ramello,L.Carnelli,Alkylation ofbenzene with isopropanolon zeolite in fluence ofphysicalstate and water concentration on catalystperformances,J.Mol.Catal.A Chem.204-205(2003)571–579.

[13]C.A.A.Monteiro,D.Costa,J.L.Zotin,D.Cardoso,Effect of metal-acid site balance on hydroconversion of decalin over Pt Beta zeolite bifunctional catalysts,Fuel160(2015)71–79.

[14]Y.Sugi,A.Vinu,Alkylation of biphenyl over zeolites shape-selective catalysis in zeolite channels,Catal.Surv.Jpn.19(3)(2015)188-120.

[15]M.D.Argyle,C.H.Bartholomew,Heterogeneous catalyst deactivation and regeneration:A review,Catalysts5(1)(2015)145–269.

[16]Y.S.You,J.H.Kim,G.Seo,Liquid-phase catalytic degradation of polyethylene wax over silica-modified zeolite catalysts,Polym.Degrad.Stab.72(2)(2001)329–336.

[17]C.C.Wang,H.Jiang,C.L.Chen,R.Z.Chen,W.H.Xing,A submerged catalysis membrane filtration system for hydrogenolysis of glycerol to 1,2-propanediol over Cu-ZnO catalyst,J.Membr.Sci.489(2015)135–143.

[18]H.Jiang,Z.Y.Qu,Y.Li,J.Huang,R.Z.Chen,W.H.Xing,One-step semi-continuous cyclohexanone production via hydrogenation of phenol in a submerged ceramic membrane reactor,Chem.Eng.J.284(2016)724–732.

[19]A.van Miltenburg,J.Pawlesa,A.M.Bouzga,N.?ilková,J.?ejka,M.St?cker,Alkaline modification of MCM-22 to a 3D interconnected pore system and its application in toluene disproportionation and alkylation,Top.Catal.52(9)(2009)1190–1202.

[20]J.M.Newsam,M.M.J.Treacy,W.T.Koetsier,C.B.D.Gruyter,Structural characterization of zeolite beta,Proc.R.Soc.A Math.Phys.420(1859)(1988)375–405.

[21]J.C.Kim,K.Cho,High catalytic performance of surfactant-directed nanocrystalline zeolites for liquid phase alkylation of benzene due to external surfaces,Appl.Catal.A Gen.470(2014)420–426.

[22]J.Q.Fu,T.B.Cao,In fluence of coke-burning regeneration using steam-containing air flow on dealumination ofbeta zeolite catalyst,Chin.J.Catal.24(11)(2003)811–815.

[23]H.S.Cerqueira,P.Ayrault,J.Datka,M.Guisnet,In fluence of coke on the acid properties of a USHY zeolite,Microporous Mesoporous Mater.38(2–3)(2000)197–205.

[24]C.Qin,Y.Chen,J.M.Cao,Manufacture and characterization of activated carbon from marigold straw(Tagetes erectaL)by H3PO4chemical activation,Mater.Lett.135(2014)123–126.

[25]A.J.Duan,G.F.Wan,Y.Zhang,Z.Zhao,G.Y.Jiang,J.Liu,Optimal synthesis of micro mesoporous beta zeolite from kaolin clay and catalytic performance for hydrodesulfurization of diesel,Catal.Today175(1)(2011)485–493.

[26]C.Kalamaras,D.Palomas,R.Bos,A.Horton,M.Crimmin,K.Hellgardt,Selective oxidation of methane to methanol over Cu-and Fe-exchanged zeolites:The effect of Si/Al molar ratio,Catal.Lett.146(2)(2016)483–492.

[27]A.V.Tokarev,L.V.Malysheva,E.A.Paukshtis,Effect of thermal treatment conditions on the acid properties of zeolite beta,Kinet.Catal.51(2)(2010)318–324.

[28]W.L.Dai,X.M.Sun,B.Tang,G.J.Wu,L.D.Li,N.J.Guan,M.Hunger,Verifying the mechanism of the ethene-to-propene conversion on zeolite H-SSZ-13,J.Catal.314(2014)10–20.

[29]G.J.Wu,Y.Hao,N.Zhang,N.J.Guan,L.D.Li,W.Grünert,Oxidative dehydrogenation of propane with nitrous oxide over Fe–O–Al species occluded in ZSM-5:Reaction and deactivation mechanisms,Microporous Mesoporous Mater.198(2014)82–91.

[30]Z.Z.Wen,X.H.Yu,S.T.Tu,J.Y.Yan,E.Dahlquist,Biodiesel production from waste cooking oil catalyzed by TiO2–MgO mixed oxides,Bioresour.Technol.101(24)(2010)9570–9576.

[31]Y.J.Liu,E.Lotero,J.G.Goodwin,X.H.Mo,Transesterification ofpoultry fatwith methanol using Mg–Al hydrotalcite derived catalysts,Appl.Catal.A Gen.331(1)(2007)138–148.