首页 金钛交联

金钛交联

举报
开通vip

金钛交联 ~ APPLIED CATALYSIS A: GENERAL ELSEVIER Applied Catalysis A: General 142 (1996) 61-71 Catalytic activity of Ti- and Al-pillared montmorillonite and beidellite for cumene cracking and hydrocracking R. Swarnakar a, Kerstin B. Brandt b, Ronald A. Kydd ...

金钛交联
~ APPLIED CATALYSIS A: GENERAL ELSEVIER Applied Catalysis A: General 142 (1996) 61-71 Catalytic activity of Ti- and Al-pillared montmorillonite and beidellite for cumene cracking and hydrocracking R. Swarnakar a, Kerstin B. Brandt b, Ronald A. Kydd b,. a Departamento de Engenharia Quimica, Universidade Federal da Paraiba, Campina Grande, P.B., Brazil 58109-970 b Department of Chemistry, University of Calgary, 2500 University Dr. N.W, Calgary, Alberta, Canada T2N 1N4 Received 4 October 1995; revised 24 January 1996; accepted 25 January 1996 Abstract Titanium- and aluminum-pillared beidellite and montmorillonite clays were prepared from Mg-beidellite synthesized hydrothermally, and Mg-montmorillonite obtained by Mg z÷ ion ex- change of a natural montmorillonite (STx-1, Source Clay Minerals Repository). The pillaring solutions were prepared by hydrolysis of aluminium chloride and titanium tetraethoxide solutions, to produce the tridecameric "Al l3" polyoxocation, A1OaAII2(OH)24(H 20)72 +, and "Tix" poly- oxocations (structure not known), respectively. X-ray diffraction analysis of the basal spacing (d00 ~) of the samples showed that the Ti-pillared beidellite is thermally more stable than analogous montmorillonite. The order of the overall activity for c~mene conversion of the clay samples is found to be: Mg-montmorillonite < Al~s-montmorillonite = Tix-montmorillonite < Mg-beidellite < Tix-beidellite < Alls-beidellite. While the conversions were very different, the selectivity of beidellite based catalysts for cracking (measured by benzene produced) was consistently in the range of 86-93% and was not influenced by the presence of aluminium or titanium oxide pillars. However this was not the case for montmorillonite. Mg-STx-1 had a relatively lower benzene selectivity (36%), and both A120 3 and TiO 2 pillars increased this significantly. The difference probably arises because the beidellite surface is more acidic due to its tetrahedral layer charge, hence making the effect of pillars less critical. Keywords: Montmorillonite; Beidellite; Pillared clays minerals; Titanium pillars; Aluminum pillars; Cumene cracking; Hydrocracking * Corresponding author. Tel. (+ 1-403) 220 5341, fax. (+ 1-403) 289 5992, e-mail kydd@chem.ucalgary.ca 0926-860X/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved. PI1 S0926-860X(96)0006 1-0 62 R. Swarnakar et al. / Applied Catalysis A: General 142 (1996) 61-71 1. Introduction The desired properties of an efficient cracking catalyst are accessible acid sites, high surface area, and high thermal and hydrothermal stabilities. Cracking catalysts which are zeolite based possess pore sizes less than about 10 A, which makes them inefficient for the cracking of larger hydrocarbons, for example those in heavy oils [1]. The search for alternative materials with high cracking activities comparable to zeolites, but with larger pores, has motivated different research groups to investigate various metal oxide pillar-interlayered clays (abbreviated as PILCs) in recent years. These research activities have been directed towards understanding the process of pillaring, developing new pillar- ing species, and finding the role of clay sheet reactivity [2,3]. Deactivation of the pillared clay catalysts has been found to be a problem in high temperature cracking reactions [4], and the thermal stabilities of PILCs have also been questioned [5,6]. The pillaring species reported in the literature are usually polymeric, cationic compounds of A1, Ga, Zr, Fe, Cr, Si, and Ti [2], or mixtures of these elements [7,8]. The most commonly studied clay mineral is montmoril- lonite, due to its availability and its ability to intercalate large cations to create the pillared clay. The most common intercalating species has been the AIO 4 All2 (OH) 24 (H20)12 7+ ion (abbreviated All3); this ion creates interlayer spaces of about 9 A (d00 ~ -~ 18 ,~). Titanium polyoxocation intercalation of montmorillonite also has been reported to achieve relatively large basal spacings (d00 ~ = 25 A) [9-11]; the exact composition of the pillaring titanium polyoxoca- tion is not known, and it is abbreviated "Tix". Although it is known that the catalytic properties of the PILCs are influenced by the nature of the clay, relatively few works in the literature address this. Diddams et al. compared the activities of Al~3-pillared montmorillonite and beidellite (which were also A13+-exchanged) for low-temperature (< 210°C) reactions involving amine, ester, and ether synthesis [12]. They found that the pillared clays were much less active than their unpillared precursors (perhaps because the A13+ exchange made them non-acidic?). Kojima et al. also studied aluminum-pillared montmo- rillonite and beidellite, comparing their activities for the isomerization (at 220°C) of 1,2,4-trimethylbenzene [13]. The present study was, therefore, undertaken to investigate the catalytic properties for cracking and for hydrocracking of two types of clays, beidellite (in which the charge sites are in the exposed tetrahedral layers) and montmoril- lonite (charges in the hidden octahedral layers). The diagram below shows, schematically, how the two tetrahedral sheets "sandwich" the octahedral sheet in these 2:1 clay minerals; the location of the cations in the interlayer spaces is also indicated. R. Swarnakar et al. / Applied Catalysis A: General 142 (1996) 61-71 63 T~ :':~ ~:,-7"v -'772"7~'75"7~'" 'X7 . . . . . . . . 7 "7 ...... "~-5:"- ~;~7~ >'-77 ";73~:¢7?:~:;7~7~ -: :p __Td O , Pillaring ions Td Oh Td Both types of clay were pillared (separately) with All3 and Ti x polyoxoca- tions, and their activities compared with those of the unpillared species. Com- parisons of the initial activities with those after 4 hours on stream were made, in an attempt to discern some of the factors which affect the deactivation of these materials. 2. Experimental Mg-beidellite was synthesized using a modified form of the method used by P.A. Diddams [14]. Initially, a gel was obtained by dropwise addition of 37.2 g colloidal silica (Ludox AS 40) to one litre of solution prepared by dissolving 46.52 g AI(NO3) 3 • 9H20 and 3.97 g Mg(NO3) 2 • 6H20. These quantities were calculated to get a molar ratio of 8:2.0:0.5 for SIO2: A1203: MgO. The gel thus obtained was evaporated to dryness, then calcined at 700°C for 8 h to decom- pose the nitrates. The calcination temperature of 700°C was achieved in three programmed steps: room temperature to 200°C, 200 to 500°C, and 500 to 700°C, with hold times of 4, 6, and 8 h, respectively, after each step. This calcined precursor was ground to fine powder, dispersed in distilled water and treated hydrothermally at 350°C under autogenic water vapour pressure (ca. 2400 p.s.i., 1 p.s.i. = 6894.76 Pa) in a 300 cm 3 autoclave for a period of seven days. The fine fraction of Mg-beidellite formed was separated by centrifuging. Ca- montmorillonite (STx-1) from the Source Clay Minerals Repository, University of Missouri, was used for obtaining Mg-montmorillonite. The < 2 micron fraction was collected by sedimentation of an alkalized suspension of STx-1, washed three times with 2 M HC1, neutralized with NaOH and finally washed with sodium ion using 0.5 M NaC1. The Na ÷ was then replaced by Mg 2÷ by repeated washings with 0.5 M MgC12. The titanium polyoxocation pillaring solution was prepared by slow addition of titanium tetraethoxide to a 5 M HC1 solution under vigorous stirring. The molar ratio of titanium tetraethoxide to HC1 was 2:1, as it was with this ratio that the relative intensity of the do01 X-ray diffraction line and the surface area of the product has been reported to be optimum [11]. The titanium polyoxoca- tion pillaring solution was added to a clay suspension (1 g/litre) at room 64 R. Swarnakar et al./Applied Catalysis A: General 142 (1996) 61-71 temperature and stirred for four hours. The proportion of the Ti:Clay was 25 mol:l equivalent (CEC) of the clay. The Ti-pillared clay was separated by centrifugation, and washed to remove chloride ions. These samples were subjected to elemental analysis. Approximately 0.2 g of each sample was digested with 3 ml HNO 3 (70%), 1 ml HC1 (37%), and 7 ml HF (48%) in a sealed container, using a CEM Microwave Sample Preparation system, model MDS-2000. The heating rate was adjusted to give a maximum pressure of 80 p.s.i. Then 50 ml of saturated boric acid was added to remove F-, and prevent loss of Si as SiF 4. The digested solutions were analysed using a Thermo Jarrell Ash Inductively Coupled Plasma instrument. For comparison purposes, unsupported TiO 2 was also prepared, using the titanium pillaring solution under similar conditions to those used for clay pillaring. The aluminium polyoxocation pillaring solution was prepared by dropwise addition of equal volumes of 0.44 M NaOH to 0.2 M A1CI 3 under vigorous stirring, hence keeping the final OH-/A13+ ratio equal to 2.2; the final concen- tration of aluminium was 0.1 M, and the pH was 4.0. This aluminium polyoxocation solution was added to 1% clay suspension at room temperature and stirred for four hours. The ratio of mol of aluminium to the cation exchange capacity of one gram of the clay was equal to 10. The pillared clay was separated as described above. X-ray powder patterns (2® = 2 to 30 degrees, CuK~) were obtained on a Scintag X-ray diffractometer for preferentially-oriented thin films produced by spreading a few drops of the clay suspension on a silica microscope slide, and allowing the sample to dry at room temperature at a relative humidity of 15%. The thermal stabilities of the pillared clays were studied by measuring the d0o j spacings of these samples after four hour treatments at each of the successively higher temperatures of 400, 500, 600 and 700°C (heating rate between different temperatures approximately 6°C per min.). The BET surface areas were determined, after degassing the catalyst samples at 200°C for half an hour, from the nitrogen adsorption isotherm obtained using an Advanced Scientific Designs Model RXM-100 Catalyst Characterization Instrument. For cumene cracking experiments, the catalyst sample was activated at 500°C for 1.5 h in He gas. Then helium carrier gas was passed, at a flow rate of 30 cm3/min, through a saturator containing cumene maintained at a constant temperature of 15°C. The cumene vapour was fed into a continuous flow, fixed bed reactor (15 mm I.D.) containing 150 mg of ground ( < 100 mesh) catalyst sample sandwiched between quartz wool plugs; the reaction temperature was 400°C. The products were detected with a Hewlett Packard 5890 gas chromato- graph equipped with a thermal conductivity detector; sampling was carried out every 20 min for a period of 4 h. The column used contained 50% phenylmethyl polysiloxane and was 10 m long X 0.53 mm I.D. X 2 ~m film thickness. With R. Swarnakar et al. /Applied Catalysis A: General 142 (1996) 61-71 65 this column, propane and propene (or any gases produced) come through very quickly and can not be resolved; therefore they are not reported. The conversion of cumene was calculated as [15] EArea of aromatics Conversion = E(Areaofunconvertedcumene + area of aromatics) (where "Area" signifies the response-corrected peak area.) Cumene hydrocracking was carried out in the same manner as cumene cracking, except that the carrier gas was hydrogen rather than helium. 3. Results and discussion The expansion in the interlayer spacing of a clay due to the incorporation of large pillaring ions is shown by the increase in the basal spacings (d0o 1) measured by X-ray powder diffraction. Subsequent decreases in d0o 1 upon calcining at higher temperatures helps in evaluating the thermal stabilities of the samples. X-ray diffraction patterns of beidellite and STx-1 montmorillonite with Mg, Ti x- and Al13 interlayer ions are shown in Fig. 1. The d0m basal spacings for o beidellites are 13.2, 21.7 and 18.8 A, while the corresponding basal spacings for o montmorillonites are 13.9, 22.3 and 18.8 A, as shown in Table 1. Clearly the intercalation was successful, and the Ti-polyoxocation is larger than the Al-pol~- oxocation, as expected [9-11]. Estimating the silicate layer thickness as 9.5 A, the gallery height for Tix-montmorillonite is slightly larger (22.3 -9 .5 = 12.8 A) than that of Ti~-beidellite (21.7 - 9.5 = 12.2 A). In Fig. 2 the effect of calcination temperature on the powder diffraction patterns of both the titanium-pillared clay samples is presented. The thermal stability of titanium pillared beidellite is slightly greater than that of titanium pillared montmorillonite, as the decrease in peak sharpness for beidellite after 500°C is less than for montmorillonite, and some evidence of a d001 peak is seen even after heating to 600 and 700°C, while the pillared montmorillonite col- lapses between 500 and 600°C. Cumene cracking over these catalysts produced benzene and a-methyl styrene; no significant amount of other aromatic products was observed. Benzene is produced on the acid sites associated with tetrahedrally coordinated aluminium and it is thought that octahedrally coordinated aluminium is not effective [16]. Further, the benzene formation activity of the catalyst has been found to be proportional to its solid acidity per unit surface area [17]. Our results for cumene conversion and the corresponding benzene selectivities for the catalyst samples investigated are interpreted on this basis. The total conversions of cumene over unpillared and pillared beidellite and 66 R. Swarnakar et al. / Applied Catalysis A: General 142 (1996) 61-71 A ]/132A (a) 18.8A 2t .?A I! ~eide l lde . . . . lb . . . . f5 . . . . ~ . . . . :~ . . . . ~n 2-There 13.9A (b) 189A / 22.34 I Mg--s'r'x,-1 3o 2-There Fig. 1. X-ray powder patterns for (a) Mg-, Ti~-, and All3-beidellite and (b) Mg-, Tix-, and Alt3-STx-1 montmorillonite. Table 1 Surface areas a and basal spacings of materials Catalyst Montmorillonite Beidellite TiO 2 Mg Ti ~ AI 13 Mg Ti x A113 S.A.(m 2/g) 78 255 126 38.2 192 251 55.6 doo i 13.9 22.3 18.8 13.9 21.7 18.8 a After pretreating for 30 min in vacuum at 200°C; basal spacings for Mg-clays vary slightly with degree of hydration. R. Swarnakar et al. / Applied Catalysis A: General 142 (1996) 61-71 67 ~ ,~ , A T1(x)-Be,deUrte ( a ) A ~ 600 C 4ooc 23 ¢ • " 5 . . . . 1"0 . . . . 15 . . . . 20 . . . . ~ . . . . 30 2 -Theto • " S . . . . lb . . . . 15 . . . . 2o . . . . ~ . . . . 3o 2-Thela Fig. 2. X-ray powder patterns for (a) Tix-beidellite and (b) Tix-STx-1 montmorillonite after calcination at different temperatures. montmorillonite as a function of time on stream are illustrated in Fig. 3. The conversions after 20 min and after 4 h on stream, along with the corresponding benzene and c~-methyl styrene selectivities, are shown in Table 2. The conversion for pure TiO 2 (no clay mineral) was ca. 20%, and the non-acidity of this surface is demonstrated by the fact that the product was exclusively e-methylstyrene: no benzene was produced. For the unpillared clays 68 R. Swarnakar et al. / Applied Catalysis A: General 142 (1996) 61-71 80 70 80 s0 e- ._o 40 ¢m c 30 o o 20 10 Alls-Beldelllte TIx'Beidelllte ~ o - ~ - - Mg-Beldellite 0 _ _ _ [ i I I 0 40 80 120 180 200 240 Time on stream (rain) 40 A o~ c- o om q) > t- o O 30 20 10 All 3-Montmorlllonlte Tix.Montmodllonlte Mg-Montmorlllonlte 0 I I I I I 0 40 80 120 180 200 240 Time on stream (rain) Fig. 3. Cumene cracking (helium carrier gas) over beidellite and montmorillonite samples as a function of time on stream. (i.e. Mg 2+ exchangeable cation), beidellite was much more active than montmo- rillonite (37% initial conversion vs. 10%), and the products were very different. The selectivity for benzene over Mg-montmorillonite was only 36%, while over Mg-beidellite it was 86%; clearly the beidellite surface is much more acidic. The deactivation after 4 h on stream was substantial in both cases (16-20% of the initial activity), and it is worth noting that, for Mg-montmorillonite, the selectiv- ity for benzene increased with time on stream. For the pillared clays, the initial conversions for both Ti x- and Al13- montmorillonite were virtually identical at 29%; the titanium pillared clay had higher selectivity for e~-methylstyrene. Initial conversions over pillared beidellite R. Swarnakar et al. / Applied Catalysis A: General 142 (1996) 61-71 69 Table 2 Catalytic cumene stream conversion a and product selectivity for various catalysts at 20 minutes and 4 hours on Catalyst % Cumene conversion Deact. b % Benzene selectivity % a-Methylstyrene % selectivity 20 min 4 h 20 min 4 h 20 min 4 h Mg-mont 10.5 8.8 16 36.1 44.3 63.9 55.7 A113-mont 29.0 25.5 12 82.1 87.4 17.9 12.6 Ti ~-mont 28.6 27.2 5 63.9 60.2 36.0 39.8 Mg-beid 37.3 29.4 21 86.3 85.9 13.9 14.1 AI 13 -beid 71.6 69.4 3 92.9 93.9 7.1 7.1 Tix-beid 56.1 44.3 21 90.9 88.0 9.1 12.0 TiO: 21.9 20.2 8 0.0 0.0 100.0 100.0 a Carrier gas is helium. b Deactivation is defined as (conversion after 20 min - conversion after 4 h) / conversion after 20 min. were larger (72% for A113 pillars, and 56% for the Ti x pillars). Both show very high selectivity for benzene (< 90%), but the Tix-beidellite deactivates more rapidly, losing 21% of its initial activity after 4 h on stream. The A113-beidellite is surprisingly resistant to deactivation, for a surface which apparently is so acidic. The results of cumene hydrocracking studies (H 2 carrier gas) are reported in Table 3. Not surprisingly, pure TiO 2 loses most of its activity under hydrocrack- ing conditions, since et-methylstyrene formation is suppressed in the presence of H2(g). Mg-montmorillonite is also less active in hydrocracking (6.4% conver- sion, vs. 10.5% during cracking), but the reactivity over Mg-beidellite is virtually unchanged, presumably because the production of ct-methylstyrene is not a major reaction on this surface even under cracking conditions, so suppress- ing its production by adding hydrogen has little effect. Table 3 Catalytic cumene hours on stream hydrocracking a and product selectivity for various catalysts after 20 minutes and after 4 Catalyst Cumene conversion Deact. b % Benzene selectivity % a-methylstyrene % selectivity 20 m 4 h 20 m 4 h 20m 4 h Mg-mont 6.4 6.1 4.7 79.7 78.7 20.3 21.3 A113-mont 28.9 28.9 0.0 96.8 96.5 3.2 3.5 Tix-mont 18.2 17.5 3.8 90.1 89.7 9.9 10.3 Mg-beid 38.3 37.5 2.1 97.4 96.2 2.6 3.7 A113-beid 69.3 69.6 0.0 99.1 99.1 0.9 0.9 Ti x-beid 64.8 63.4 2.2 98.6 98.5 1.4 1.5 TiO 2 2.4 2.8 0.0 23.5 25.7 76.5 74.2 a Carrier gas is hydrogen. b Deactivation as defined in Table 1. 70 R. Swarnakar et al. / Applied Catalysis A: General 142 (1996) 61-71 Table 4 Elemental analyses of the titanium-pillared montmorillonite and beidellite Wt. % Ti AI Si Ti .~-mont 20.1 5.5 23.4 Tix-beid 15.5 10.8 20.7 For A113-pillared montmorillonite, the conversion during hydrocracking (29%) is almost the same as during cracking, but for Tix-montmorillonite, conversion decreases (29 to 18%) as the hydrogen causes o~-methylstyrene production to be reduced. For Al~3-beidellite, cumene conversion is unchanged at about 70%. This is different from the situation for Tix-beidellite, where the conversion during hydrocracking is larger than during cracking, and the significant deactiva- tion that occurred after 4 h on stream during cracking was no longer observed. Comparison of the catalytic results with the results of elemental analyses (Table 4) of the titanium
本文档为【金钛交联】,请使用软件OFFICE或WPS软件打开。作品中的文字与图均可以修改和编辑, 图片更改请在作品中右键图片并更换,文字修改请直接点击文字进行修改,也可以新增和删除文档中的内容。
该文档来自用户分享,如有侵权行为请发邮件ishare@vip.sina.com联系网站客服,我们会及时删除。
[版权声明] 本站所有资料为用户分享产生,若发现您的权利被侵害,请联系客服邮件isharekefu@iask.cn,我们尽快处理。
本作品所展示的图片、画像、字体、音乐的版权可能需版权方额外授权,请谨慎使用。
网站提供的党政主题相关内容(国旗、国徽、党徽..)目的在于配合国家政策宣传,仅限个人学习分享使用,禁止用于任何广告和商用目的。
下载需要: 免费 已有0 人下载
最新资料
资料动态
专题动态
is_314274
暂无简介~
格式:pdf
大小:558KB
软件:PDF阅读器
页数:11
分类:理学
上传时间:2013-02-08
浏览量:11