Studies on Catalyst Deactivation Rate and Byproducts Yield during Conversion of Methanol to Olefins

Scientific ResearchStudies on Catalyst Deactivation Rate andByproducts Yield during Conversion ofMethanol to OlefinsYan Dengchao; Munib Shahda; Weng Huixin(College of Chemical Engineering, East China University of Science & Technology;Shanghai 200237)Abstract: The conversion of methanol to olefins (MTO) over the SAPO-34 catalyst in fixed-bed microreactorwas studied. The effect of reaction temperatures for methanol conversion to olefins and byproducts wasinvestigated. A termperature of 425 C appeared to be the optimum one suitable for conversion of methanol toolefins. Since the presence of water could increase the olefins seclectivity, the methanol conversion reactionswith mixed water/methanol feed were also studied. The effect of weight hourly space velocity on conversionof methanol was also studied. The results indicated that the olefins selectivity was significantly increased asWHSV increased till approximately 7.69 h' then it began to level off. Different factors afecting the catalystdeactivation rate was studied. showing that the catalyst deactivation time was dependent on reactionconditions, and temperatures higher and lower than the optimal one made the catalyst deactivation faster.Adding water to methanol could slow down the catalyst deactivation rate.Key words: methanol conversion; catalyst deactivation: SAPO-34; MTO1 Introductionucts yield and the ethylene/propylene ratio. Also our studyincluded the investigation of the main factors, which couldThe conversion of methanol to olefins has altracted greataffct the olefins selectivily over the SAPO-34 catalyst inaattention nowadays. Before the emergence of the molecularbid to compiete our previous studies with this kind of acidsieve silicoaluminophosphates (SAPOs). a lot of catalysts catalys. In all of our experiments a fixed-bed reactor and athat were used to promote the conversion reactions encom-GC equipment were used.passed the small-pore zeolites such as chabazite, erionite,zeolite TI- +, and ZK-5 I.6, and the medium-pore zeolites, a2 Experimentaltypical one of which was the ZSM-5 zeolite'l.2.1. Experimental procedureThe SAPOs series zeolites have weak acidity compared withthe ZSM-5 zeolite. and show low activity for promoting chemi- The whole series of experiments were carried out in a stain-cal reactions, such as hydrogen transfer. oligomerization andless steel fixed-bed microreactor loaded with fresh SAP0-34coking reactions that need the presence of strong acidic sites.catalyst. A DWT-702 temperature controller was used to con-Among the SAPOs seres zeolies the SAPO-34 zcolite is be- trol the reaction temperature. The feed was introduced onlineing most frequently studied for methanol conversion to hy-into the reactor. The reactor was operated under atmosphericdrocarbons(., because its pore opening and acidity seem to pressure. At the end of cach run. prior to replacement of thebe most suitable for ethene and propene production.catalyst the reactor was purged with nitrogen to eliminate theresidual products. Figure I shows the sketch of the experi-The main objective of this paper was to investigate the ef-men中国煤化工fects of space velocity, reaction temperature and methanolconcentration on the catalyst deactivation tine, the by-prod- The.TYHCNMHGwereanalyzedbyagas33China Petroleum Processing and Petrochemical TechnologyNo.3, September 2006= t图Preheatermal conductivity detector (TCD). which was used in our ex-periments to analyze the conversion process. Chemical com-Reactor GCRota-meter/pounds were identified based on the residence time: hydro-但EracationSitogen Valvecarbon products were quantitatively analyzed by measuring两一G-port \alveEvacuationthe peak area. This was also used to analyze DME, which wasFeedquantitatively analyzed by measuring the peak area also. Table+Q -2 presents the analytical conditions of the gas chromatograph.Flow pumpFigure I Sketch of the experimental unit3 Results and Discussionchromatograph, which was equipped with a thermal conduc-3.1 Effect of reaction temperaturetivity detector (TCD). A GDX101 Porapak QS packed column,2m in length and 3 mm in inside diameter, was used to sepa-3.1.1 Effect of reaction temperature on catalyst deactiva-rate the products. Hydrogen was used as the carrier gas.tion time2.3 Description of the catalyst and gasIn our experiments the catalyst deactivation time was the timechromatographafter which the methanol conversion rate decreased to belowThe whole series of our experiments were performed over the50%.calcined SAPO-34 catalyst, which was provided by the Re-Figure 2 & 3 show the effect of reaction temperature on cata-search Institute of Petroleum Processing. SINOPEC.lyst deactivation time (T) and the selectivity (propylene+Table I lists the physical properties of the SAPO-34 catalyst.ethylene), respectively, for pure methanol fed and the metha-The gas chromatographic analysis was performed with a ther-nol/water mixed feed with the weight hourly space velocity(WHSV) equating to 3.29 h'. It can be seen from Figure 2 & 3Table 1 Physical properties ofthat the catalyst deactivation time and the selectivitySAPO-34 catalyst(propylene+ ethylenc) increased with an increasing reactionCatalyst designationSAP0-34temperature till approximately 425 'C then started to decline.Acidity, mmol/gAt higher and lower temperatures. the catalyst deactivation0.0036time was less than that achieved at medium temperature, be-B0.0306cause the coke formation was faster at a higher temperature.SiO,/AI,O0.46At a lower temperature the catalyst was deactivated becauseSpecific surface area, m2/g575of the formation of oligomers inside the zeolite channels. TheCapacity, mLg0.26olefins were generally prone to oligomerization and were notMicro-activity (800"C/4h)2likely to form aromatics. At a lower temperature. since thebulky oligomers could not be cracked, they were incapable ofTable 2 Analytical conditions for gasdiffusing outside of the channels!9. and consequently weretrapped in the cages of SAPO- 34 zeolite, making the internalGC-columnPorapakQacid sites inaccessible to reactants, which could lead to cata-Carrier gasHlyst deactivation. Likewise. it can be seen from Figure 2 thatColumn inlet pressure, MPa0.1adoption of the methanol/water mixed feed could slow downFlow rate, mL/min20the catalyst deactivation as compared to the case using pureColumn temperature, Cmetha. good choice to deterEvaporation temperature, C140the cat中国煤化工: of water in the feed,Thermal conductivity bridge current, mA150someTYHC N M H G surface were occu-34Scientific Research340.5 t29024.3 t昌190.1 t* 140.9 t9).7 L375400475 500425 450175 500Temperature. (Temperaure, CFigure 4 Effect of reaction temperature onFigure 2 Effect of reaction temperature onC,HC,H。ratio at WHSV=3.29h'catalyst deactivation timie (WHSV=3.29 h')◆- pure methanol feed:■- methanol/water ratio=l◆一pure methanol feed;■- -methanol/water ratio=I6(3.1.2 Effect of reaction temperature on byproducts5:5 50selectivity45房40The presence of by-products such as Co, CO2 CH. andCH,3:would significantly affect the downstream separation and3(purification processes, which were important factors in de-42545075500termining the economics of the MTO process.Temperature, CFigure 3 Effect of reaction temperature onFigure 5 shows the effect of reaction temperature on by-prod-selectivity of ( ethylene +propylene) atucts selectivity. It can be seen from Figure 5 that the yields ofWHSV=3.29h'by- products Co, CO2 and CH, increased while the CH yield◆- pure methanol feed;■- -methanol/water ratio=ldecreased with an increasing reaction temperature. At lowtemperature the CH, selectivity reached 40 % thanks to thepied by polar water molecules and thus were unavailable tocatalyst acidity, leading to occurrence of hydrogen transferolefinsl"ol, which could prevent olefins from undergoing oli- reactions among light olefins at low temperature. Furthermore,gomerization and coking on the active sites, and might keepa higher temperature could give rise to dehydration process.the channels inside the catalyst unblocked. Therefore, lesscoke was formed on the catalyst.The ethylene to propylene ratio is related to reaction tem-perature as shown in Figure 4. This ratio increased with anincreasing reaction temperature, because propylene was more0treactive than ethylene and could undergo oligomerizationmore easily to form bigger oligomers that would be trapped inthe cages of the SAPO- 34 zeolite. As the reaction tempera-37400 425450475 500ture increased further, these oligomers could be cracked toform ethylene. At a higher temperature, more coke was formedTemperature. Con the catalyst. Therefore, the ethylene to propylene ratio isFigure 5 Effect of reaction temperature onrelated with the coke content or the degree of catalyst products selectivity with pure methanol feed atdeactivation. The parially deactivated catalyst favored the中国煤化工”surge in the ethylene to propylene ratio.MHCNMHG:●-CH,35China Petroleum Processing and Petrochemical TechnologyNo.3, September 20063.2 Effect of methanol concentration1.25 r1.20 tThe effect of feed dilution on the catalyst deactivation time,1.1the (ethylene +propylene) selectivity, the ethylene to propy-.10 tlene ratio and the by- products distribution is presented in.05上Figures 6, 7, 8 and 9. When the catalyst loading was kepl at0.5g. the flow rate was adjusted to keep the space velocity0.331.003.00identical (3.29 h") at the reaction temperature of 450C andWater/MeOH ratio475"C, respectively. It can be seen from Figures 6. 7 and 8 thatdiution of the feedstock with waler could slow down the Figure 8 Effet of water/methanol ratio in feedcatalyst deactivation, enhance the catalyst selectivity (foron C,H /C,H。ratio at WHSV=3.29 h'ethylene +propylene) and increased the ethylene to propy-◆- 450C;■- -475Clene ratio. The water dilution of methanol feedstock reducedthe feed concentration and also reduced the influent metha-nol quantity, resulting in an increase of the number of activesites. The water has been proved to be a good choice for导1070602.0050冒400Figure 9 Effect of water/methanol ratio in feed”30200on the by-products selectivity.100(Reacion temperalure=450'C at WHSV=3.29h')◆-CH;■--C0;▲-CH;●- CH,reducing the partial pressure of methanol to enhance the ole-fin selectivily and deter the rate of catalystFigure 6 Effect of methanol/water ratio in feed dectivato.711.1 because of the following two factors.on catalyst deactivation time at WHSV= =3.29 h' Firstly, the reduced partial pressure of methanol favored a▲-450C:■- -475Chigher olefins yield. Secondly. because of the presence ofwater, the catalyst deactivation was greatly depressed.60 tFigure 9 shows the by-products yield decreased as the waterdilution of feedstock increased because of increased MTO50上conversion and decreased coke deposition. After the addi-40 5tion of water to methanol the methanol molecules were notdecomposed to form by-products. Also we did not verify0Lthat the strong acidic sites on the catalyst would catalyze thedecomposition of methanol and DME into CO, CO、and CH;otherwise, the amount of CO, CO. and CH, would decrease inFigure 7 Effect of methanolwater ratio in feed the course of catalyst deactivation.on the (ethylene+ propylene) selectivity atWHSV=3.29 h'The res中国煤化工c hydrocatons were◆- . 450C;■1- 475CdetecteHC N M H G1 were in agreement36Scientific Researchwith the case using pure methanol as the feed.lene +ethylene increased W ith an increasing space velocityand was stabilized at 7.69 h I because of the increased amount3.4 Effect of space velocityof methanol, leading to a faster catalyst deactivation resultedfrom increased light olefins yield as evidenced by Figure 10.A series of experiments were performed to investigate the The effect of space velocity on the ethylene to propyleneinfluence of space velocities at a reaction temperature of 450ratio was quite insignificant as shown in Figure 12. indicatingC. Figures 10, II. 12 & 13 show the effects of space velocitythat there was no significant change in the ethylene to pro-on the catalyst deactivation time, the (ethylene +propylene)pylene ratio in relation to varying space velocity. The high-selectivity. the ethylene to propylene ratio and the by-prod-est selectivity for CO.. CH, and C.H、in the products wasucts distribution, respectively. The space velocity in our ex-identified at a lowest space velocity while the selectivity forperiments was calculated based on the amount of catalyst CH, increased with an increasing space velocity. The de-loading (0.5 g) with the methanol flow rate being adjusted to composition of methanol took place at a much slower rategenerate different space velocities. Figurel0 has shown thethan the MTO conversion, since a lower space velocity couldtime of catalyst deactivation to be declining with an increas-be conductive lo the formation of more CO, CH, than at aing space velocity, because the coke formation rate was faster higher space velocity. Also some ethane products were be-when the methanol flow increased. The selectivity for propy-lieved to be emanated from hydrogen transfer reactions..4「300.2 -200冒.0 t100).8 L_13.295.497.699.88WHSV. h'Figure 10 Effect of space velocity on the timeFigure 12 Effect of space velocity on theethylene to propylene ratioof catalyst deactivation(methanol/water ratio=I; reaction temperature=450'C)(methanol/water ratio=l: reaction temperature=450C)2:62 r60 t58 te 1:56 t54↓_ L_50 I48 l.49.88WHSV, h' .Figure 13 Effect of space velocity onFigure 11 Effect of space velocity on thehv-products selectivity(ethylene+ propylene) selectivity中国煤化工peraure=450C)(methanol/water ratio= 1; reaction temperature: =450C )YHCNMHG- CH.37China Petroleum Processing and Petrochemical TechnologyNo.3, September 20064 ConclusionsReferences[I]Anthony R G, Singh B B. Chem Eng Commun. 1980, (6):Our study has shown that the reaction temperature could 215have a signification infuence on the catalyst deactivation [2] Lin F N, Chao J C, Anthony R G. Coal Processingtime. Higher temperatures and lower temperatures made the Technology, 1978.4:.73catalyst deactivation time faster than that achieved at me- [3] Singh B B, Lin F N, Anthony R G. Chem. Eng. Commun.dium reaction temperatures. The optimum reaction tempera-1980,5: 749ture for a fixed-bed reactor appeared to be approximately 425 [4] LiuL, Tobias RG. Mclaughlin K,Anthomy RG In: Hermanc over the SAPO-34 catalys. Also the reaction temperature R G(Ed.), Catalytic Conversion of Synthesis Gas to Alcoholscould aft the by- products distribution and the ethylene to andChemicals, Plenum Press, New York, 1984.323- -360propylene ratio.[5] Chang C D, Lang W H, Silvestri A J. US Patent 40629051972Adding water to methanol as a diluent has proved to be a [6]ChangC D. Cata Rev Sci Eng.1983.25(1):1good choice to increase the olefins slectiviy and reduce the [7] Dehertog W J H, FromentGF. Appl Cala, 1991.71, 153-coke deposition on the catalyst, which could deter catalyst 165deactivation compared to the case using pure methanol as [8] Munib Shahda , Yan Dengchao, Wang Zhihe, Wen Huixin.the feed.Study on Olefins Yield from Methanol Conversion over Dif-ferent Catalysts. China Petroleum Processing & Petrochemi-Space velocity could have a strong impact on the MTO cal Technology, 2006,(1):49- -54reaction; the catalyst deactivation time decreased with an [9] Derouane EG, GilsonJ P, Nagy J B.J Mol Catal, 1981. 10:increasing space velocity. However, the olefins selectivity 331- 340increased because the amount of methanol increased at a [10} Marchi AJ, Froment G F. App! Catal 1991, 71:139fixed catalyst amount that made the coke formation faster on [11] Wu X, Anthony R G. Appl Catal A: Gen, 2001,218: 241-the catalyst. The ethylene to propylene ratio also increased 25but the by-products yield decreased with an increasing space [12] Niekerk MJ V, Flecher JCQ. O'ConnorCT. Appl Catalvelocity.A:Gen, 1996, 138: 135- -145Bayer's MDI Unit in Shanghai BecameOperationalThe MDI unit in the Shanghai integrated base of Bayer Matc-missioned in June 2006 in order to separate and produce therial Science located in Shanghai Chemical Industry Park has pure and polymerization grade MDI from the MDI mixture.Itbeen put on stream and more than 700 tons of qualified prod- is learned that the MDI product is designated to serve theucts have been delivered after three days of operation.Asia Pacific customers with its quality in full compliance withthe global quality standard of Bayer Corporation.This MDI unit is rated at 80 kv/a, and the unit was pre-com-中国煤化工MYHCNMHG38