Purification Influence of Synthesis Gas Derived from Methanol Cracking on the Performance of Cobalt Purification Influence of Synthesis Gas Derived from Methanol Cracking on the Performance of Cobalt

Purification Influence of Synthesis Gas Derived from Methanol Cracking on the Performance of Cobalt

  • 期刊名字:天然气化学(英文版)
  • 文件大小:165kb
  • 论文作者:Wei Zhou,Shengying Liu,Yulan W
  • 作者单位:State Key Laboratory of Coal Conversion,Graduate School of Chinese Academy of Sciences
  • 更新时间:2020-07-08
  • 下载次数:
论文简介

Journal of Natural Gas Chemistry 14(2005)193 -198SCIENCE PRESSPurification Influence of Synthesis Gas Derived fromMethanol Cracking on the Performance of CobaltCatalyst in Fischer- Tropsch SynthesisWei Zhou12,Shengying Liu',Yulan Wang',Kegong Fang',Jiangang Chen',Yuhan Sun'1. State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyruan030001, China 2. Graduate School of Chinese Acaderny of Sciences, Bejing 1009, China[Manuscript received September 29, 2005; revised November 17, 2005]Abstract: Synthesis gas derived from methanol cracking (SGMC) was applied as simulating feedstock ofFischer-Tropsch synthesis (FTS) in laboratory. With MS and GC detector, a trifle of sulfur compounds,a small amount of oxygenates including H2O, CH3OH, DME and CO2 as well as a few of low carbonalkanes were found in the SGMC. After purification, the sulfur compounds, H2O, CHzgOH and DME couldbe eliminated efficiently from the SGMC while CO2 and the low carbon alkanes were partly removed.When the unpurified SGMC, the desufurized SGMC and the totally purified SGMC were sequentiallyapplied in cobalt-based FTS, the catalytic performance of Co/ZrO2/SiO2 catalyst was gradually improvedcorresponding to the degree of purification. The untreated SGMC led to the serious deactivation of thecobalt catalyst, the partially treated SGMC slowed down the deactivation rate and the totally purifiedSGMC resulted in lttle deactivation of the catalyst, which was similar to what the pure synthesis gas(the mixture of pure H2 and CO) did. The results indicated that the SGMC should be purified and thepurification course used in this paper was effective for the SGMC. Furthermore, the totally purified SGMCcould substitute for the pure synthesis gas in cobalt FTS.Key words: Fischer-Tropsch synthesis, synthesis gas, methanol cracking, purification1. Introductiondistillates as products [1,2].Among the above three stages, the manufacture ofFischer- Tropsch synthesis (FTS) has gained in-synthesis gas is by far the most capital-intensive stagedustrial attention for converting synthesis gas to high-[2]. Generally, synthesis gas is manufactured by par-boiling points waxes that can be further converted totial oxidation of natural gas or gasification of coal insulfur-free motor fuels by hydrogenation or hydroc-industry, while it is produced by mixing the pure COracking as alternative resources to the limited crudend H2 or decomposing methanol in laboratory [3-8].oil. In the Fischer Tropsch synthesis processes, syn-Thereof, synthesis gas derived from methanol crack-:hesis gas is first generated, then the F-T reactioning (SGMC) has particular advantages such as higherforms in the second step, in which synthesis gas iscontents of efective components (CO+H2>95%) rela-converted into the hydrocarbons over the catalysts.tive t(+H2>85%), moreIn the final step, the hydrocarbons undergo furthersuitab中国煤化工to co (2:1) thanhydrogenation or hydrocracking to yield clean middle coal-b:TYHCN M H G-0.5-1.5) and nat-* Corresponding author. Tel: (0351)4121877; Fax: (0351)4041153; E-mail: yhsun@sxicc.ac.cn.Financial supported from National Natural Foundation of China (20590361 and 20303026) and State Key FoundationProgram for Development and Research of China (2005cb221402).194Wei Zhou et al./ Journal of Natural Gas Chemistry Vol. 14 No. 4 2005ural gas- based synthesis gas (H2/C0>2) [3] and much2.3. F-T reactioncheaper than the mixture of pure CO and H2. There-fore, the SGMC is applied widely in many laboratoriesA series of experiments were carried out on the[6- 8] and even was used in AMSTG course, which wasSGMC, the partially purified SGMC, the totally pu-one modified FT technology in Japan several yearsrified SGMC and the pure synthesis gas (the mix-ago [3].ture of pure H2 and CO), respectively. 1 ml cata-With respect to the catalyst for FTS, supportedlyst was diluted with 4 ml SiO2 to minimize the tem-cobalt catalysts are highly attractive due to their highperature gradients. Then the catalyst activation wasactivity, high yields of long-chain paraffins and lowconducted in pure hydrogen at 673 K, 0.5 MPa andactivity for the water-gas shift reaction [9- 11]. How-2000 h-1 for 6 h in a fixed bed reactor of id. 10 mm.ever, when the SGMC is applied in cobalt-based FTS,After the activation period, the reactor temperaturethe feed gas must be kept very pure for the activewas decreased to the room temperature and the syn-metallic cobalt was inclined to be poisoned. Thoughthesis gas (H2/CO=2.0) was introduced. The F-T re-the SGMC is used widely, the purifying course has notaction proceeded at 493 K, 2.0 MPa and 1500 h- 1.been reported so far. Therefore, this paper mainly in-Liquid products and wax were obtained through avestigated the purifying course of the SGMC on thecold trap keeping at 273 K and a hot trap keeping atreaction behavior of cobalt-based FTS.413 K, respectively. The products analysis was justthe same as previous work [16]. All reported data were2. Experimentalmeasured after 72 h time-on-stream in order to ensuresteady-state behavior of the reaction. The mass and2.1. Analysis systemcarbon balance of all the reactions both maintainedbetween 95% and 105%.The analysis system consisted of one mass spec-trometer (MS, HPR-20) and two sets of gas chro-3. Results and discussionmatograph (GC). One GC was equipped with thermalconductivity detector (TCD) and a carbon molecular3.1. Analysis of the SGMCsieve column, while the other was set with flame ion-ization detector (FID) and a Porapak-Q column. ArThe SGMC was admitted into mass spectrome-was the carrier gas for GC. The sulfur contents wereter and Sulfur Analyzer (see Figure 1 and Table 1)monitored by the HC-2 Sulfur Analyzer. The SGMCThe result suggested that there were a trifle of sul-was routed on-line into the MS, GC and Sulfur Ana-fur compounds including H2S and COS and a smalllyzer, separately.amount of oxygenates including H2O, CO2, CHgOH,and DME in the SGMC (see Figure 1).2.2. Catalyst preparationTable 1. MS analyses results of the synthesisThe Co/ZrO2/SiO2 catalyst was applied in thisgas by cracking of methanol beforepaper for its good catalytic performance in FTS [1,12-and after purifying14]. A 15% Co/10% ZrO2/SiO2 catalyst was pre-Compound inBeforeAfterpared by sequentially incipient wetness impregnationthe SGMCpurifying (%)method with drying at 303 K after each impregna-H263.8764.22tion. Sol-gel derived SiO2 (278 m2/g) was used asCo31.1631.33the support. The zirconium and cobalt precursorsCO21.161.20were zirconium (IV) nitrate and cobalt (I) nitrate,CH40.880.86respectively [15]. The catalyst impregnated with zir-H2S*0.00020.00001conium was first calcined at 773 K for 4 h in theCOS*0.0005air and then impregnated with the cobalt solution,redried and then calcined with the same procedure as中国煤化工mentioned above. The cobalt loading was verified byMHCNMHG0.004using an inductively coupled plasma atomic emission2.38spectrometer (ICP-AES) analysis method.* The contents were monitored by the sulfur analyzerJournal of Natural Gas Chemistry Vol. 14 No.42005195(aH2O(bC)_CO,CO2CFCHCOS(6c)HsS.66cos.4EHHS.2日.2小(a1.0 E0.80.6CHOH.4DME2年CHUR0.004 E0.002CHJOH2 4681012141602468101214160246810121416Time (min)Figure 1. MS detection results of the SGMC(@) Unpurifed SGMC, (b) Purifed SGMC after desulfurizing, (c) Purifed SGMC after desulfurizing and oxygenates-removingGC results (see Figure 2 and Table 2) proved CO2 low carbon alkanes was also detected in the SGMCand DME existed in the SGMC. Moreover, a trace ofthrough MS and GC.(间)b)支会(2)1)16 18 20中国煤化工56Retention time (min)TYHCNMHGFigure 2. GC detection results of the SGMC(1) Unpurified synthesis gas, (2) Purified synthesis gas196Wei Zhou et al./ Journal of Natural Gas Chemistry Vol. 14 No. 42005Table 2. GC analyses results of the synthesiscould not induce side reactions. ZnO catalyst (T305)gas by cracking of methanol beforewas a good choice for removing sulfur compounds forand after purifyingit was widely used in the purification of the synthe-Compound inBeforesis gas when the content of the sulfur compounds wasthe SGMCpurifying (%)small [27,32]. In addition, blue silica gel was good atH285.4466.44absorbing water while 5A molecular sieve was a goodCO32.0332.31oxide adsorbent for its uniform reticulate structureCO21.210.34[33]. Therefore, T305, blue silia gel and 5A mole-CH40.760.81cular sieve were applied in the SGMC purificationC2H60.09courses. T305 catalyst was put into a desulfurizingC3Hg0.01container that was set in a heating furnace. The mix-C4H1o0.00DME0.46ture of the blue silica gel and 5A molecular sieve (mix-ing ratio was 1/1) was sent into the other vessel. Thedesulfurizing course was carried out at 500 h-1 and3.2. Purification of SGMC230。C and the second course was at 500 h- 1 androom temperature.It was well known that the sulfur compoundsThe purification effects for the SGMC are showncould poison metallic cobalt and lead to the deacti-in Figures 1, 2 and Tables 1, 2. The MS and Sulfurvation of the catalyst [17]. So the removal of sulfurAnalyzer results exhibited that the amount of the sul-compounds from the SGMC was crucial. H2O wasfur compounds decreased by one order of magnitudealso one species that could make cobalt catalyst de-after desulfurizing and the amount of the CHgOH andactivate in FTS [4,5,18- 30]. It was found that wa-DME decreased by two orders of magnitude after totalter could result in the reoxidation of cobalt [18 -23],purification. However, the amount of H2O and CO2formation of interacting species [24- -26] and sinteringdecreased not significantly according to the MS re-[27-30)],Schanke et al. [18] thought that the ox-sults. Furthermore, GC results showed that the DMEidation of surface cobalt atoms or highly dispersedcould not be detected any more and the amount ofcobalt phases by water on alumina supported cobaltCO2 as well as the low carbon alkanes decreased incatalysts was responsible for the catalyst deactiva-part after the purification courses. It demonstratedtion. Krishnarmoorthy et al. [23] proved that waterthat the removal of the sulfur compounds, CHgOHcould reoxidize the surface cobalt of Co/SiO2 catalystand DME from the SGMC were efficient, while CO2at H2O/H2 ratio greater than 0.8. In addition, theand the low carbon alkanes could be partly absorbedformation of interaction species between cobalt andin the purification course. To assess the water absorp-Al2O3 or SiO2 supports under hydrothermal condi-tion in the mixture of blue silica gel and 5A moleculartions had been reported[24 26]. As for the sintering,sieve, the SGMC was introduced at 500 h-1 and am-Bartholomew [27] pointed that water vapor would in-bient temperature. After one month, about 1/5 bluecrease the sintering rate of supported metals. Thesilica gel turned pink. It proved that the existencesintering had been proved on cobalt catalysts at highof water in the SGMC and the validity of the waterwater partial pressure, which was thought from theremoval in the purification course. Thus it could beoxidation-reduction cycles induced by water in F-Tseen that a part of water and carbon dioxide detectedreaction [28- -30]. Therefore, the synthesis gas must beby MS might come from the residual atmosphere inkept dry. As for other oxygenates, such as, CH3OHthe ultra high vacuum house of MS. In brief, the sulfurand DME, no report is published whether they werecompounds, methanol, DME and water in the SGMCharmful to cobalt catalyst. Moreover, CO2 was foundcould be eliminated efficiently and CO2 and the lowto incorrelate with the deactivation of cobalt in FTScarbon alkanes could be removed partly through theif CO2 content was lower than 19% [31]. Consider-purified courses.ing the oxidation effects, the oxygenates should be中国煤化工、cleaned up. However, a small amount of low carbon3.nce of SGMC overalkanes, which would be produced in FTS, could be igthe:MYHCNMHGnored for having no report about the negative effectsof the alkanes on the cobalt catalyst.The reaction results of F-T synthesis wereThe purification courses should be simple andgraphed in Figure 3. For the experiment that wasJournal of Natural Gas Chemistry Vol. 14 No.42005197run in the SGMC, the catalyst activity and the secarbon dioxide and low carbon alkanes did not affectlectivities for CH4 and Cs+ decreased rapidly, imply-the performance of the cobalt catalyst in F-T synthe-ing that the impurities in the SGMC were harmfulsis. Figure 3 also iluminated that the catalyst showedto the cobalt catalyst in the F-T reaction. Whennearly the same reaction performance whether in thethe SGMC was desulfurized, the catalyst activity andpure synthesis gas or in the purified SGMC, implyingthe selectivities for CH4 and C5+ were improved ev-that the purified SGMC could substitute for the pureidently. Still, after the SGMC was purified by T305synthesis gas in the evaluation of the cobalt catalyst.and the mixture of blue silica and molecular sieves, se-quentially, the catalyst didn't deactivate for at least4. Conclusions550 h, indicating that the SGMC should be purifiedand the purification method for the SGMC in this Pa-A trifle of sulfur compounds, a small amount ofper was efficient. In addition, the catalyst showed anoxygenates consisting of water, methanol, DME andexcellent performance though a few of carbon dioxideCO2 a8 well a8 a trace of low carbon alkanes wereand low carbon alkanes existed in the totally purifieddetected in the synthesis gas derived from methanolSGMC, which implied that the existence of a few ofcracking. After two step purification, the sulfur com-pounds, water, methanol and DME were eliminated00 refficiently, while CO2 and low carbon alkanes were re-r (4moved partly. When the unpurified synthesis gas was80 t(3(a)applied in the cobalt F-T reaction, the Co/ZrO2/SiO2(2)catalyst deactivated severely. Next when the desul-furized synthesis gas was used, the cobalt catalystshowed an amendatory performance. At last, whenthe totally purified synthesis gas through the desulfu-20 Frizing and oxygenates-removing courses was employedin the F-T reaction, the cobalt catalyst exhibited goodperformance and stability, which was nearly the sameas in the pure synthesis gas. It proved that: (1)((bthe SGMC should be purified; (2) the purification8method for the SGMC in this paper was eficient; (3)2)the purified SGMC could substitute for the pure syn-thesis gas in cobalt-based FTS.ofAcknowledgementsThe authors acknowledge the financial sup-port from National Natural Foundation of China90 F(4)(C(20590361 and 20303026) and State Key Founda-tion Program for Development and Research of China30 t(3)(2005cb221402).70 5Referencesc50 E[1 Dry M E. Catal Today, 2002, 71: 227[2] GeerlingsJJ C, WilsonJ H, Kramer G J et al. Appl96 144 192 240 288 336 384 432 480Catal A, 1999, 186: 27Time (h)(3]中国煤化工J Fuel Chem Tech,Figure 3. F-T synthesis performance in different syn-194thesis gases over Co/ZrO2/SiO2 catalyst4LiTYHC N M H GAppl Catal A, 2002,(1) Unpurified SGMS, (2) Purified SGMS after desulfuriz-236: 67ing, (3) Purifed SGMS after desulfurizing and oxygenates-[5]LiJL,ZhanXD,ZhangYQetal.ApplCatalA,removing, (4) Pure synthesis gas2002, 228: 203198Wei Zhou et al./ Journal of Natural Gas Chemistry Vol. 14 No. 42005[6)] QiHJ,LiD B, Ma Y G et al. J Fuel Chem Tech,(20] Berge P J van, Loosdrecht J van de, Barradas s et al.2003, 31(2): 119Catal Today, 2000, 58: 321[7] Liu Y, Zhong B, PengS Y et al. Cuihua Xuebao (Chin[21] Jacobs G, Patterson P M, Zhang Y Q et al. ApplJ Catal), 1998, 19(2): 107Catal A, 2002, 233: 215[8] Zhang Y Q, Zhong B, Wang Q. Feizi Cuihua (J Mol [22] LiJ L, Davis B H. Stud Surf Sci Catal, 2004, 147:Catal), 1997, 1(3): 204307[9] ChenJ G, Xiang H w, LiY w et al. Huagong Xuebao23] Krishnamoorthy S, Tu M, Ojeda M et al. J Catal,(J Chem Ind Eng), 2003, 54(4): 5162002, 211: 422[10] YinD H, Li W H, Yang WS et al. Prog Chem, 2001,[24] Zhang Y L, Wei D G, HammacheS et al. J Catal,13(2): 1181999, 188: 281[11] ZhangJ L, ChenJ G, Li Y W et al. J Natural Gas[25] Kiss G, Kliewer C E, DeMartinG J et al. J Catal,Chem, 2002, 11(3-4): 99 .2003, 217: 127[12] Post, Martin F M. ESP 8604428, 198626]ChenJG,WangXz,XiangHWetal.StudSurfSci[13] Ali S, Chen B, Goodwin J G Jr. J Catal, 1995, 157:Catal, 2001: 5253527] Bartholomew C H. Appl Catal A, 2001, 212: 17[14] ChenJ G, Xiang H W, Wang X Z et al. Cuihua Xue-28] Bertole C J, Mims C A, Kiss G. J Catal, 2002, 210:bao (Chin J Catal), 2000, 21(4): 359[15] Chen J G, Xiang H W, Sun Y H. Cuihua Xuebao [29] Das T K, Jacobs G, PattersonP M et al. Fuel, 2003,(Chin J Catal), 2000, 21(2): 16982: 805[16] ChenJ G, XiangH W, GaoH Y et al. React Kinet [30] Jacobs G, Patterson P M, Das T K et al. Appl CatalCatal Lett, 2001, 73(1): 169A, 2004, 270: 65[17] Espinoza R L, Steynberg A P, Jager B et al. Appl [31] Freide J J H M F, Gamlin T D, Hensman J R. JCatal A, 1999, 186: 13Natural Gas Chem, 2004, 13(1): 1[18] Schanke D, Hilmen A M, Bergene E et al. Catal Lett,32] Zhang B J. Coal-Based Liquid Fuel. Taiyuan: Shanxi1995, 34: 269Science and Technology Press, 1993, 30[19] Hilmen A M, Schanke D, Hanssen K F et al. Appl [33] Hersh C K. Molecular Sieves. Chen E transl. Bejing:Catal A, 1999, 186: 169Chinese Industry Press, 1965, 46中国煤化工MYHCNMHG

论文截图
版权:如无特殊注明,文章转载自网络,侵权请联系cnmhg168#163.com删除!文件均为网友上传,仅供研究和学习使用,务必24小时内删除。