Tổng hợp, cấu trúc, công nghệ, ứng dụng của zeolites Zeolite có thể được phân loại theo tỉ lệ SiAl . Cách tổng hợp dựa vào thành phần các hợp chất chứa Al và Si. Zeolit là khoáng chất silicat nhôm (aluminosilicat) của một số kim loại có cấu trúc vi xốp với công thức chung: Me2xO.Al2O3.nSiO2.mH2O Trong đó: Me là kim loại kiềm như Na, K (khi đó x = 1) hoặc kim loại kiềm thổ như Ca, Mg... (khi đó x = 2). Tên gọi zeolit được nhà khoáng vật học người Thụy Điển là Axel Fredrik Cronstedt nghĩ ra năm 1756, khi ông quan sát thấy khi nung nóng nhanh stilbit thì nó sinh ra một lượng lớn hơi nước bị vật liệu này hấp phụ trước đó. Hiện nay có khoảng 150 loại zeolit đã được tổng hợp và khoảng 48 loại có trong tự nhiên đã được biết đến. Zeolit có cấu trúc mở vì vậy nó có thể kết hợp với các ion kim loại khác nhau như Na+, K+, Ca2+, Mg2+. Zeolit tự nhiên được hình thành từ sự kết hợp giữa đá và tro của núi lửa với các kim loại kiềm có trong nước ngầm. Zeolit được dùng với nhiều mục đích khác nhau trong các lĩnh vực như công nghiệp hóa học, kỹ thuật môi trường như là các chất hấp phụ, xúc tác, chiết tách... Zeolit có thể gặp ở trạng thái tự nhiên hoặc nhân tạo. Để tổng hợp zeolit có thể thực hiện theo 2 cách: Trực tiếp từ các nguồn nguyên liệu tự nhiên, biến tính các aluminosilicat là các khoáng phi kim loại như cao lanh, bentonit. Tổng hợp trực tiếp từ các silicat và aluminat.
Studies in Surface Science and Catalysis 24 ZEOLITES Synthesis, Structure, Technology and Application Proceedings of an International Symposium organized by the "Boris Kidri~" Institute of Chemistry, Ljubljana on behalf of the International Zeolite Association, Portoroz-Portorose, September 3-8, 1984 Editors B Driaj, S Ho~evar and S Pejovnik »Boris Kidric« Institute of Chemistry, 61000 Ljubljana, Yugoslavia ELSEVIER Amsterdam - Oxford - New York - Tokyo 1985 ELSEVIER SCIENCE PUBLISHERS B.V Sara Burgerhartstraat 25 P.O Box 211, 1000 AE Amsterdam, The Netherlands Distributors for the United States and Canada: ELSEVIER SCIENCE PUBLISHING COMPANY INC 52, Vanderbilt Avenue New York, NY 10017 ISBN 0-444-42568-3 (Vol 24) ISBN 0-444-41801-6 (Series) © Elsevier Science Publishers B V., 1985 Produced in collaboration with Mladinska knjiga International, Ljubljana, Yugoslavia All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the publisher, Elsevier Science Publishers B.V.lScience & Technology Division, P.O Box 330, 1000 AH Amsterdam, The Netherlands Special regulations for readers in the USA - This publication has been registered with the Copyright Clearance Center Inc (CCC), Salem, Massachusetts Information can be obtained from the about conditions under which photocopies of parts of this publication may be made in the USA All other copyright questions, including photocopying outside of the USA, should be referred to the copyright owner, Elsevier Science Publishers B.V., unless otherwise specified cec Printed in Yugoslavia XI Studies in Surface Science and Catalysis Volume Preparation of Catalysts I Scientific Bases for the Preparation of Heterogeneous Catalysts Proceedings of the First International Symposium held at the Solvay Research Centre, Brussels, October 14-17, 1975 edited by B Delmon, P.A Jacobs and G Poncelet Volume The Control of the Reactivity of Solids A Critical Survey of the Factors that Influence the Reactivity of Solids, with Special Emphasis on the Control of the Chemical Processes in Relation to Practical Applications by V.V Boldyrev, M Bulens and B Delmon Preparation of Catalysts II Scientific Bases for the Preparation of Heterogeneous Catalysts Proceedings of the Second International Symposium, Louvain-Ia-Neuve, September 4-7,1978 edited by B Delmon, P Grange, P Jacobs and G Poncelet Growth and Properties of Metal Clusters Applications to Catalysis and the Photographic Process Proceedings of the 32nd International Meeting of the Societe de Chimie physique, Villeurbanne, September 24-28,1979 edited by J Bourdon Catalysis by Zeolites Proceedings of an International Symposium organized by the Institut de Recherches sur la Catalyse - CNRS - Villeurbanne and sponsored by the Centre National de la Recherche Scientifique, Ecully (Lyon); September 9-11, 1980 edited by B Imelik, C Naccache, Y Ben Taarit, J.C Vedrine, G Coudurier and H Praliaud Catalyst Deactivation Proceedings of the International Symposium, Antwerp, October 13-15,1980 edited by B Delmon and G.F Froment New Horizons in Catalysis Proceedings of the 7th International Congress on Catalysis, Tokyo, 30 June-4 July 1980 edited by T Seiyama and K Tanabe Catalysis by Supported Complexes by Yu.1 Yermakov, B.N Kuznetsov and V.A Zakharov Physics of Solid Surfaces Proceedings of the Symposium held in Bechylle, Czechoslovakia, September 29-0ctober 3,1980 edited by M Lazni~ka Adsorption at the Gas-Solid and Liquid-Solid Interface Proceedings of an International Symposium held in Aix-en-Provence, September 21-23,1981 edited by J Rouquerol and K.S.W Sing Volume Volume Volume Volume Volume Volume Volume Volume 10 Volume 11 Volume 12 Volume 13 Metal-Support and Metal-Additive Effects in Catalysis Proceedings of an International Symposium organized by the Institut de Recherches sur la Catalyse - CNRS Villeurbanne and sponsored by the Centre National de la Recherche Scientifique, Ecully (Lyon), September 14-16, 1982 edited by B Imelik, C Naccache, G Coudurier, H Praliaud, P Meriaudeau, P Gallezot, G.A Martin and J.C Vedrine Metal Microstructures in Zeolites Preparation - Properties - Applications Proceedings of a Workshop, Bremen, September 22-24, 1982 ed ited by P.A Jacobs, N.1 Jaeger, P Jiru and G Schulz-E kloff Adsorption on Metal Surfaces An Integrated Approach edited by J Benard Volume 14 Vibrations at Surfaces Proceedings of the Third International Conference, Asilomar, California, U.S.A., 1-4 September 1982 edited by C.R Brundle and H Morawitz Volume 15 Heterogeneous Catalytic Reactions Involving Molecular Oxygen by G.I Golodets Volume 16 Preparation of Catalysts III Scientific Bases for the Preparation of Heterogeneous Catalysts Proceedings of the Third International Symposium, Louvain-Ia-Neuve, September 6-9,1982 edited by G Poncelet, P Grange and P.A Jacobs XII Volume 17 Spillover of Adsorbed Species Proceedings of the International Symposium, Lyon·Villeurbanne, September 12-16,1983 edited by G.M Pajonk, S.J Teichner and J.E Germain Volume 18 Structure and Reactivity of Modified Zeolites Proceedings of an International Conference, Prague, July 9-13, 1984 edited by P.A Jacobs, N./ Jaeger, P Jrr~, V.B Kazansky and G Schulz·Ekloff Volume 19 Catalysis on the Energy Scene Proceedings of the 9th Canadian Symposium on Catalysis, Quebec, P.Q., September 30-0ctober 3, 1984 edited by S Kaliaguine and A Mahay Volume 20 Catalysis by Acids and Bases Proceedings of an International Symposium organized by the Institut de Recherches sur la Catalyse-CNRS-Villeurbanne and sponsored by the Centre National de la Recherche Scientifique, Villeurbanne (Lyon), September 25-27,1984 edited by B Imelik, C Naccache, G Coudurier, Y Ben Taarit and J.C Vedrine Volume 21 Adsorption and Catalysis on Oxide Surfaces Proceedings of a Symposium, Brunei University, Uxbridge, June 28-29,1984 edited by M Che and G.C Bond Vo.urne 22 Unsteady Processes in Catalytic Reactors by Yu.Sh Matros Volume 23 Physics of Solid Surfaces 1984 edited by J Koukal Volume 24 Zeolites Synthesis, Structure, Technology and Application Proceedings of an Internation Symposium, Boris Kidric Institute, Ljubljana, September 8, 1984 edited by B Drzaj, S Hoeevar and S Pejovnik XIII I?REFACE At the Sixth International Zeolite Conference in Reno (U.S.A.), 1983, the Council of the International Zeolite Association entrusted the organization of the 1984 International Symposium on synthesis of zeolites, their structure determination and their technological use, to Yugoslavia This symposium was organized by the Department of Catalysis and Substances with Well-Developed Surfaces of the "Boris Kidric" Institute of Chemistry, Ljubljana, and held in Portoroi-Portorose from to September 1984 The symposium ran in three successive sections The section on synthesis began with a plenary lecture prepared by R.M Barrer and given by D:E.W Vaugha~ New directions in the synthesis of molecular sieves were outlined in the plenary lectures of E.M Flanigen and Xu Ruren The mechanisms of aluminosilicate formation with emphasis on the synthesis of zeolites were presented in the plenary lecture of W Wieker, which was given by K.-H Bergk The plenary lectures of W.M Meier and G.T Kokotailo introduced the work in the section on structure determination of zeolites The symposium work in the third section, that on the technology and application of zeolites, followed the introductory thoughts presented by R Sersale and Liang Juan in their respective plenary lectures The synthesis of zeolites with desired structure and properties is of great importance for the preparation of highly active and selective catalysts for inorganic and organic reactions The zeolite matrix offers unique possibilities for carrying out molecular shape-selective catalysis and this places the zeolite matrices among tne most successful tools used in molecular engineering on a large scale The papers presented in this book concentrate on the possible ways of synthesising the silica-high types of zeolites, like ZSM-5, ZSM-11, NU-3, erionite, offretite, L-type zeolite, etc., from the four-or five-component system On the one hand, detailed explanations are given of the processes in solutions, during the nucleation period and during crystallization; on the other hand, special emphasis is, given to the use of modern physical techniques, e.g.neutron diffraction, X-ray diffraction using synchrotron radiation, nuclear magnetic resonance of samples spinned at "magic" angle (MAS NMR), in the determination of the structure and distribution of framework and exchangeable ions Descriptions are given of the possible technological use of synthetic zeolites in the fields of adsorption, catalysis, the production of laundry XIV detergents,the removal of radioactive wastes, and the technological use of natural zeolites in the fields of municipal water treatment, paper and cement production and energy storage In the name of the Organizing Committee we wish to thank first the Council of IZA for giving us the opportunity to organize this symposium in Yugoslavia We are indebted to all the plenary lecturers, all of whom did their best to provide a thorough overview of the present state-of-the-art in selected fields of zeolite chemistry, to the members of the International Scientific Committee, to the chairpersons of sessions and to some other renowned scientists who reviewed the symposium contributions Special thanks are due to Mr D Gabrovsek for revising the language of all the contributions We are very grateful to our sponsors: Federal Committee for Energy and Industry Slovene Academy of Arts and Sciences University of Edvard Kardelj, Ljubljana Federation of Chemists and Technologists of Yugoslavia Union of Yugoslav Chemical Societies Committee for Research Activities and Technology of Slovenia Research Community of Slovenia for their understanding and for the unfailing support they gave in the organization of the symposium, and to the contributors: Council of the Association of Self-Managing Communities of Interest of the Republics and Provinces for Research Activities, Ljubljana Birac Alumina Factory, Zvornik for their generous financial support We also express our gratitude to the Auditorium Congress Center, PortorozPortorose for their hospitality, and to all other institutions and individuals who helped either financially or otherwise, thus enabling the symposium to take place smoothly We hope that this symposium may encourage-at least partly-new research on the synthesis, structure determination and technological use of zeolites, and may help create a favorable atmosphere for a fruitful Seventh International Zeolite Conference, to be held in Tokyo in 1986 B DRzAJ S HOcEVAR S PEJOVNIK Dtfaj, S Hocevar and S Pejovnik (Editors) Zeoliu:» Q 1985 Elsevir Science Publishers B V Amsterdam - Printed in Yugoslavia SYNTHESIS OF ZEOLITES R.M BARRER Chemistry Dept , , Imperial College, London SW7 (Great Britain) ABSTRACT Zeolite synthesis has been considered in terms of factors influencing the species formed and aspects important for systematic study Zeolites Can be divided into categories according to the Si/Al ratios Synthesis behaviour varies to some extent between the most aluminous and most siliceous groups Synthesis of the most siliceous zeolites, which tend to be hydrophobic, is promoted by many organic species, usually basic in nature Two roles of such compounds have been indicated Firstly, the host crystal is stabilised by inclusion of guest molecules This has a thermodynamic basis and is considered specifically in connection with the synthesis of melanophlogite Secondly, a component of the mixture may act as a template assisting nucleation and crystal growth Finally, attention has been directed to the method of analysis of curves of yield of crystals against time, developed by Zhdanov and Samulevich This requires measurements of linear growth rates of the largest crystals and of the final size diatr ibution of crystals INTRODUCTION Zeolite mineralogy began around 1756 with the discovery of stilbite by Cronstedt (ref 1) Zeolites were first observed in basalts which had been altered by hydrothermal action that sometimes formed crystals of museum display quality Reports in the late 19th century began to record their occurrence in sedimentary tuffs (ref 2) and marine sediments (ref 3) These reports have become numerous in the periods both before and especially after the second world war, with zeolite formations often in very large amounts, in vitric tuffs, dry sa line lake beds, in association with bentonites, and also in low-grade metamorphic rocks (ref.4) The sedimentary deposits are sometimes nearly monomineralic, but compared with crystals found in basalts the crystals in sediments and tuffs are very small The role of water as a mineralising catalyst, aided by alkaline conditions, drew the attention of mineralogists to hydrothermal reactions and syntheses The first of these may have been the growth of hydrothermal quartz by Schaf- heut (ref.5) in 1845 from silica gel and water Reviews by Niggli and Morey (ref.6) and Morey and Ingerson (ref 7) summarise much of hydrothermal chemistry and mineralogy up to about 1937 The first claim to have made a named zeolite, levynite, was that by St Claire Deville in 1862 (ref.8) Solutions of K and Na silicates heated to 170 0C in sealed glass tubes gave hex- agonal, tabular, uniaxial crystals having the composition In 1850-52 the nature of ion-exchange in soils was clarified by Way and Thompson (ref 9), and this ion exchange property in zeolites led to ear ly investigations from 1870 onwards (ref 10) Also Grandjean in 1910 (ref 11) made some notable early studies of the sorption of heavy vapours (1 Br2 ' 2, HgS) in chabazite, following quantitative measurements S, Hg, Hg C1 2C12, 2, of water vapour-zeolite equilibria by Tamman (ref 12) About 1932 McBain (ref.13) introduced the term "molecular sieve" to describe the selectivity of some micro-porous carbons and zeolites in uptake of molecules according to size Molecules too large to enter the micropores were sorbed less than smaller ones which could enter Through these illustrations one sees a little of the early history of zeolite science My interest began in the mid-thirties It seemed to me that, because they were both porous and crystalline, zeolites should act as almost perfect Maxwell demons in barring entry to molecules of the wrong shape and size to pass the mesh of the lattice, while freely admitting and sorbing large amounts of molecules having the right shape and size to permeate the crystals Hence they should be able to separate appropriate mixtures quantitatively in a single step This behaviour was demonstrated abundantly in the early and mid 1940's with examples of nearly all the kinds of separation now of interest (ref.14, 15, 16) Three and later five (ref.17) categories of molecular sieve zeolite were specified From the early 1940's onwards one also began systematic synthesis studies on zeolites, the first made being analcime, mordenite (ref 18) zeolites with the edingtonite framework and also with the framework of that subsequently termed ZK-5 (ref.19, 20, 20a) ZK-5 was thus the first zeolite without a natural counterpart to be synthesised Hydrogen zeolites were also made for the first time in 1949, by heating the ammonium exchanged forms of mordenite (ref.21) It was my intention to publish in chemistry-oriented rather than mineral- ogical journals in order to bring to the attention of chemists and chemical engineers an area, so far neglected by them, that seemed to have great potential It was however about 10 years after my first paper before interest was roused among a small group of scientists at Linde Air Products in Tonawanda From that point on, as my own research group expanded and industrial interest at first slowly and then rapidly increased, zeolite science and technology burgeoned until it became the present great tree, with many branches and still growing vigorous ly Zeolites may, with justification, be termed what Eitel, that grand old man of silicate science, referred to in a private letter to me, as "the pride of mineralogists" ZEOLITE SYNTHESIS One limb of the zeolite tree embodies our experience of synthesis and chemical knowledge which this has given Ideally this knowledge should tell of events at the molecular level leading to nucleation and growth, and should enable one to design and synthesise lattices of novel types for possible new applications This stage has not yet been achieved, but on a more empirical basis zeolite synthesis is rich in chemical interest and discovery In the formation of zeolites the results may be influenced by such factors as: The nature of the reactants and their pre-treatment The way in which the reactant mixture is made and pre-treated and its overall chemical composition Homogeneity or heterogeneity of the mixture pH of the mixture Low temperature ageing of gels Seeding Addition of special additives Temperature and pressure The first three of these in particular can result in history-dependent factors which indicate that nucleation may not be controlled by such thermodynamic variables as composition, temperature or pressure but by environ- mentally sensitive kinetic factors For example, kaolinite (oxide formula AI and metakaolinite (oxide formula AI obtained 203.2Si02.2H20) 203.2Si02, 0C), by heating kaolinite to 1\1 500 both with and without added silica, were heated with aqueous Ba( OH) + LiOH The crystalline products were entirely different (ref 22), as seen be low: From metakaolinite From kaolinite u.r, (Ba, LiJ -Q (yugawaralite type) Ba)-ABW, zeolite (Ba, LiJ -edingtonite (Ba, LiJ-G, L (like zeolite LTL) (Ba, LiJ-M (phillipsite group) (Ba, LiJ_PJ( (non-zeolite, like cymrite ) (Ba, LiJ-T (silicate) (Ba, LiJ-N (unidentified) Conversion of kaolinite to metakaolinite has altered dramatically the dominant nucleation process under otherwise similar conditions of temperature, pressure and gross chemical composition Seeding of aluminosilicate gels with crystals of the desired zeolite can be successful in directing crystallisation provided other conditions are appropriate for that particular zeolite to grow Thus, from otherwise suitable 0C, Na-aluminosilicate gels at 200 seeding with faujasite will not be successful but seeding at 90 to 1000C can induce growth of faujasite on the seeds as well as fresh nucleation of faujasite Even here, however, the result may be history-dependent Thus, when sodium aluminosilicate gels were formed around the seeds, little extra growth on them was observed but when the seeds were added after gel formation their addition successfully promoted faujas ite growth on them (ref 23) This behaviour supports the view that growth of crystals involves reaction of dissolved chemical nutrients with the crystal surface rather than reaction involving gel and crystal directly Seeding may shorten or even eliminate the induction period This period may also be reduced by a preliminary low temperature ageing of the parent gel In addition, one may sometimes instead of seeding with pre-formed crystals, add as seed material a little of the mother liquor from a previous batch to the new gel The literature (ref.24) contains many examples of the importance of all the factors referred to at the beginning of this section and these factors will 676 concerned with a bifunctional catalyst, containing both metallic and acidic active sites Therefore, the preparation of bifunctional catalysts is considered very critical and requires careful adjustment of all preparation parameters involved, in order to avoid the preparation of catalysts reacting as monofunctional acidic or metallic catalysts Accordingly, the object of this study isroXa) identify the type of n-hexane isomerization mechanism taking place over the prepared Pt/La-faujasite catalysts and (b) investigate the effect of both, Si0 ratio and the extent of Na 2/AI 20 removal, on the isomerization activity of prepared catalysts (ref.l, 10, 11) EXPERII.IENTAL Catalyst preparation Six different Pt loaded La-faujasite catalysts were prepared The procedure fOrtheir preparation was the same, but the levels assigned to their preparation parameters were varied, as indicated in Table Catalysts under investigation were prepared from NaX and NaY type zeolites, which possess the same structure as that of faujasite, but differ in their Si0 ratios; 2.5 and 5.0 respectively Na cations, in each zeolite type 2/AI 203 were exchanged with La cations to three different levels This waS realized through the number of La( N03)3 exchange batches which varied from one to three according to the extent of exchange required The exchange reaction was control3+ led by the ratio of g.ion equivalent of La / g.ion equivalent of Na+, which 3+ was calculated with respect to the 3Na~ La equilibrium isotherm (ref.12) The third exchange batch was always preceded by an intermediate calcination step, carried out at 450 0C in a stream of very dry air for hours This step aimed at decreasing the residual Na in the treated zeolite, since it helps the Na cations, encaged in the small-orifice sodalite cages, to migrate to ne~ posi- tions, from where they could be easily removed by further exchange batches (ref 12) La-exchanged samples were then loaded with 0.5 Pt, using the competitive wt~ cation exchange technique (ref.9, 13) For each sample, a competition curve was traced in order to determine the optimum zone of competition, which corresponds to the best dispersion and maximum homogeneous distribution of the Pt particles over the zeolite This zone is represented by 'x/~' which is the ratio of the total number of competitor cations to that of Pt cations in the medium Selected values of the competition ratio 'x/~' are given in Table The Pt loaded La-faujasite samples were then calcined at 475 0C for hours in a stream of very dry air flowing at a rate of lit/h/g sample Prepared catalysts were subjected to x-ray analysis of Na, La and Pt contents as well as measurements 677 Activity testing of prepared catalysts Catalysts prepared were tested in the isomerization of n-hexane The reaction was carried out under high pressure in a "Catatest unit" operating at different conditions of temperature and space velocity (LHSV) according to the conversion level required Outputs of the reaction were analysed in a gas chromatographic apparatus connected to a microprocessor The catalyst bed, containing 30 cc of the calcined catalyst, in the form of mm extrudates was activated at 450 0C and atm for hour in a stream of pure nitrogen It was then reduced with pure hydrogen at 450 0C and 30 bars for hours, before being ready for catalytic activity testing The reaction operating conditions were adjusted at two stages each corresponding to a certain level of conversion The first, a high conversion stage aims at investigating the repartition of cracking products, as well as the effect of catalyst cation exchange level and Si0 ratio on its activity The operating conditions 2/Al 203 o -1 were thus adjusted at: 330 C, 30 bars, LHSV = h a n d H (molar) = 2/n-hexane The second stage is one of low c011l.e'sion, during which the reaction temperature l oC and LHSV were adjusted between 260and300 and to 12 h- respectively, in order to obtain a total conversion of ~ 16 mole% This low conversion acquires the reaction a zero order with respect to the hydrocarbon (ref.14) and avoids the OCCu~ of side reactions, such as cracking RESULTS AND DISCUSSION Preparation of Pt/La-faujasite catalysts Catalysts prepared are described in Table from which it that for each of the two Si0 can be observed ratios (2.5 and 5.0) three different levels 2/A1 203 of cation exchange were realized The first level (0.742 for NaX- and 0.670 for NaY-based catalysts) was below that of the thermodynamic equilibrium isotherm, which approaches 0.82 for NaX and 0.7 for NaY (ref.12) These thermodynamic values correspond to the removal of 69 and 39 Na+ cations out of 85 and 55 Na+ cations per unit cell of NaX and NaY, respectively These removed cations are located in different accessible sites in the eight supercages per unit cell On the other hand, the remaining non-exchanged, Na+ cations (16 per unit cell for each zeolite) are located in the network of small cages, consisting of 16 hexagonal prisms and sodalite cages per unit cell (ref.15) which are not easily accessible The second level of cation exchange, 0.935 for NaX and 0.750 for NaY, has exceeded the corresponding thermodynamic equilibrium It corresponds to the removal of 92.5 and 76.8 wt% Na, originally present in both zeolites respectively As for the third exchange level, which was realized by three exchange batches including an intermediate calcination step, it affected significa~lythe ~d 678 residual Na, still encaged in both zeolites after the second exchange batch This residual Na thus removed corresponds to 6.3 and 16 wt% Na originally present in NaX and NaY zeolites respectively From Table 1, it can be also observed that Pt content in the six prepared catalysts is between 0.24 and 0.37 wt%, which is within the suitable range for n-hexane isomerization over bifunctional metal loaded zeolite catalysts (ref 1) TABLE Main features of catalysts prepared Catalyst No Extent of cation Exchange 'E' 2/A1 20 = 2.5 2/A1 203 = 5.0 a Catalysts based on Si0 0.742 0.935 0.989 b Catalysts based on Si0 0.670 0.750 0.935 Compet i t ion ratio 'x/oc I Pt wt % 3.50 1.10 0.18 0.29 0.24 0.24 3.55 2.30 0.71 0.36 0.37 0.29 (NaX) 130 160 160 115 140 140 Na wt % (NaY) Activity testing of prepared catalysts Catalysts prepared were tested in the isomerization of n-hexane, which leads to the formation of four different isomers, namely 2,2-dimethylbutane (2,2-DMB), 2,3-dimethylbutane (2,3-DMB), 2-methylpentane (2-MP) and 3-methylpentane (3-MP) The aim of the reaction was realized through two main stages, each of which had different operating conditions, as previously described in the Experimental part High conversion stage Through this stage, the effect of both the extent of cation exchange 'E' and Si0 ratio of prepared catalysts, on their iso2/A1 20 merization activity could be demonstrated At the same time, the repartition of cracking products produced from the reaction on the different catalysts could be examined The effect of 'E' on the isomerization activity of type X-and type Y-based catalysts, is illustrated in Fig For the former type (samples 1, and 3), 'E' had a significant influence Its increase from 0.742 to 0.935, which represents the decrease of Na content from 3.5 to 1.1 wt% (i.e 16.5% of the total Na originally present in NaX) , increased the isomerization activity at 330 0C from 10% to 69% Further removal of residual Na from 1.1 to 0.18% (i.e increase of 'E' from 0.935 to 0.989), increased the isomerization activity from 69% to 73%, which approaches that of the thermodynamic equilibrium at 330 0C by 92% (ref.16) At the same time, the increase of 'E' caused a slight decrease in 679 catalyst selectivity due to the increase in cracking • 2· a-Si~/AI203 80 b- Si02 AI203 60 Total Conversion 60 - o o ~o 40 =5 80 - - - -Thermodynamic limit Thermodynamic limit Isomerization ~ 20 20 Cracking Cracking 0.7 0.8 0.9 0.8 0,7 \ Extent of cation exchange £ 0.9 1.0 I Fiy, Effect of extent of cation exchange on n-hexane isomerization over prepared catalysts Operating conditions: 330 oC, 30 bars, LHSV ~ 2h-l,H2/nC6~ As for type V-based catalysts (samples 4, and 6), it is obvious from Fig.lb that 'E' exhibited a remarkable effect only up to 'E' ~ 0.750 Beyond this value and up to the maximum exchange practically realized (0.935), no significant change was observed The maximum isomerization activity was obtained at 'E' ~ 0.75 and reached a value of 53.2%, which corresponds to 67% only of the thermodynamic equilibrium This activity is inferior to that realized by type X_ -based catalyst (73%) However, when comparing both types at the same 'E' of 0.75, it appears that the activity exhibited by the type V-based catalysts is 2.65 times higher Actually, the effect of 'E' on the activity of prepared catalysts is considerably dependent on the catalyst Si0 not be neglected For 'E' < ratio Their interaction effect could 2/AI 203 0.87, the type V-based catalyst exhibited a higher activity than the corresponding type X, while for 'E' a significantly higher acti vi ty On the other hand, > 0.87, the latter showed when 'E'was in the range of 0,87, the Si0 ratio had no significant effect, and both types of catalysts 2/AI 203 possessed the same activity of about 50% Examining the repartition of cracking products, given in Table 2, it could be observed, for type X-based catalysts, that (C + C + C and (C + C 2)/(C4 5) l l 2)/C3 ratios are < 1.0 at 'E' ~ 0.742, which is almost in accordance with the simple cracking of n-hexane and its isomers However, when 'E' increased to 0.989, the 680 previously mentioned ratios decreased, indicating a decrease in (C + C with I 2) respect to both C and (C + C At the same time, a significant increase in 5) both iC and iC was observed These changes in cracking products indi5 5/nC 4/nC4 cate that a certain reaction, involving two or more molecules, may have taken place, leading to the formation of C as well as iC and iC (ref.9) The absence of (C + C with respect to C + C could also be attributed to the occurrence 2) 5, of -rapid secondary alkylation reactions (ref.17) TABLE Isomerization of n-hexane over prepared catalysts Repartition of cracking oC, 1, products Operating conditions: 330 30 bar, LHSV = hH = 2/nC6 Catalyst No Extent of cation exchange 'E' a Catalysts based on Si0 0.742 0.935 0.989 b Catalysts based on Si0 0.670 0.750 0.935 (C + C 2) (C + C 5) 2/A1 203 2/A1 203 (C + C 2) C = 2.5 (NaX) 0.80 0.36 0.31 0.80 0.32 0.24 0.0 0.89 1.47 0.0 1.39 1.89 0.65 0.62 0.55 0.67 0.23 0.54 0.63 0.44 0.62 = 5.0 (NaY) 0.52 0.60 0.42 On the other hand, the repartition of cracking products on type Y-based catalysts are not significantly influenced by changes in 'E' However, their (C + C + C and (C + C ratios still show low values, as those pro2)/(C4 2)/C3 1 5) duced on the type X-based catalysts, thus indicating the absence of hydrogenolysis which occurs on monofunctional metallic catalysts Accordingly, it could be mentioned that the transformation of n-hexane and its isomers into light products has proceeded through monofunctional acidic, as well as bifunctional cracking reactions Low conversion stage Through this stage, the kinetics of n-hexane isomeri- zation could be investigated, since for a low conversion of ~ 16 mole%, the re- action corresponds to a zero order with respect to the hydrocarbon, and this simplifies the calculations of isomerization reaction rate constant 'k' and the apparent activation energy 'A' At the same time, it avoids the occurrence of side reactions, such as cracking The repartition of the various isomers for all prepared catalysts are des- cribed in Table 3, in which the corresponding thermodynamic values are also given for comparison 681 TABLE Kinetic data of n-hexane isomerization over prepared catalysts l, 0C, Operating conditions: 300 30 bars, LHSV = 2-12 hH = 2/nC6 Catalyst Extent of exch 'E' No Reaction rate 'k' constant (h- l) a Catalysts based on Si0 2/A1 203 5.1 78.9 246.8 0.742 0.935 0.989 b Catalysts based on Si0 0.670 0.750 0.935 *Calculated ~iC~ 2/A1 203 11 , 43.0 = IiC * Repartition of isomers App Activ 2-MP 3-MP 2,2-DMB 2,3-DMB energy 'A' I iC IiC~ IiC~ I.iC~ (Kcal/mole) 2.5 (NaX) 38.4 38.8 38.2 0.0 0.015 0.034 0.0 0.022 0.044 0.569 0.572 0.562 0.431 0.407 0.394 0.012 0.013 0.014 0.018 0.015 0.018 0.573 0.581 0.579 0.409 0.403 0.403 0.219 0.133 0.516 0.352 5.0 (NaY) 37.8 39.3 39.8 from thermodynamics (ref.16) - 2,2-DMB For the type Y-based catalysts, as well as for the low La-exchanged type X, the repartitions of 2-~~ and 3-MP are very near to their corresponding thermo- dynamic values, while the two other repartitions are considerably inferior Such repartitions of isomers correspond to a typical bifunctional mechanism with the skeletal isomerization of carbonium ion as the rate controlling step On the other hand, for the high La-exchanged type X-based catalyst (sample 3), the repartition of 2,3-DMB increased to 34% of its equilibrium value, indicating that two mechanisms may exist: a monofunctional acidic with the formation of intermediate carbonium ion as the controlling step, and a bifunctional mechanism, with the dehydrogenation of n-hexane or the hydrogenation of n-hexene as the rate controlling step The isomerization reaction rate constant 'k' at 300 0C, over the prepared catalysts, are given in Table For type X-based catalysts, the increase of 'E' l, from 0.742 to 0.935 increased 'k' from to 79 hindicating that sample is almost 16 times more active than sample Further decrease of Na from 1.1 l to 0.18 wt% increased 'k' to 247 h(sample 3) As for type Y-based catalysts, it is obvious from Table that 'E' does not exhibit a similar significant effect as in the case of type X-based catalysts The apparent activation energy 'A' of prepared ~atalysts ± 1, in n-hexane isomeri- zation at 270-300 0C, reaction through a bifunctional mechanism (ref.l, 9) occu~ was found to be within 38.8 which indicates that the 682 CONCLUSIONS - The decrease of Na content in the zeolitic support had a pronounced effect on the n-hexane isomerization activity of prepared catalysts, especially for those based on zeolite type X - Investigation of the ekt of Si0 ratio on prepared catalysts' acti vities 2/AI 203 revealed the presence of its strong interaction with the extent of cation excha- nge Catalysts based on a ratio of 2.5 showed higher activities than those based on a ratio of 5.0, when the extent of exchange was ~ 0.9 - The most active prepared catalyst (sample 3) possessed a high reaction rate l constant of 247 h- at 300 oC It was prepared from NaX, for which the Na content waS reduced from 14.6 to 0.18 wt%, by three successive La-exchange batches inclu- ding an intermediate calcination Pt (0.24 wt%) was then introduced by competitive cation exchange, and the catalyst was finally calcined at 475 being reduced with hydrogen at 400 oC 0C before - The kinetic investigation of n-hexane isomerization over prepared catalysts, revealed that the reaction proceeded through a bifunctional mechanism rather than an acidic or metallic monofunctional one REFERENCES 10 11 12 13 14 15 16 17 Kh.M Minachev and Y.I Isakova, Zeolite Chemistry and Catalysis, Am Chern Soc., Chap 10, 1976 A Roumegous, These Docteur Ingenieur, Univ Paris VI, June 1978 J.A Rabo, P.E Pickert, D.N Stamires and J.E Boyle, Actes ern Congr Intern Catalyse Paris 1960, (1961) 2055 Kh.M Minachev, V.I Garanin, V.V Kharlamov and T.A Isakova, Kinet Catal., 13 (1972) 1104 Kh.M Minachev, V.I Garanin, T.A Isakova, V.V Kharlamov and V.I Bogmolov, Adv Chern Ser No.102 (1971) 441 M.A Lanewata, P.E Pickert and A.P Bolton, J Catal., (1967) 95 J.E Cole and H.W Kouwenhoven, Adv Chern Ser No.12l (1973) 583 F.G Ciapetta, R.M Dobres and R.W Baker, Catalysis Vol.6, Reinhold, New York, Chap 6, 1958 F Ribeiro, These Docteur Science Physique, Univ de Poitier, Mars 1980 H.W Kouwenhoven, Adv Chern Ser No.12l (1973) 529 L.P Shirinskaya, V.S Komarov, 1.1 Urbanovich and T.M Korneeva, Kinet Catal., 19, (1978) 645 Ch Marcilly and E Freund, Rev Inst Francais du Petrole,27, (1972) 247 H.A Benesi, U.S Patent 3,527,835, Sept (1970) G.M Panchenkov and V.P Lebedev, Chemical Kinetics and Catalysis, Mir Publishers, IIoscow 1976 D.H Olson, J Phys Chern 74 (1970) 2758 J.A Ridgway and W Schoen, Ind Eng Chern., 51 (1959) 1023 M.L Poutsma, Zeolite Chemistry and Catalysis, Am Chern Soc., Chap 8, 1976 683 B Drzaj, S Hocevar and S Pejovnik (Editors), Zeolires © 1985 Elsevir Science Publishers B V Amsterdam - Printed in Yugoslavia LOW AROMATIC SOLVENTS THROUGH DEAROMATIZATION ON MOLECULAR SIEVE 13X D.AHMETOVIC and S.SVEL-CEROVECKI INA, Research and Development, Proleterskih brigada 78, Zagreb (Yugoslavia) ABSTRACT The possibility of preparing low aromatic solvents by passing some of refinery streams over molecular sieve 13X has been studied From experimental data dynamic and equilibrium capacity were calculated and S-curves constructed and used to calculate the height of mass transfer zone and height of unused mass transfer zone Laboratory experiments data gave us optimal temperature and liquid velocity for each crude oil fraction at which the height of mass transfer zone had the lowest value The correlation of zone height and linear liquid velocity for gasoline, white spirit and kerosene was found INTRODUCTION There are two basic types of adsorption systems differing in quantities of material we want to separate from a mixture (ref.l) if up fo % of the component that we want to separate is present in the mixture the process is called purification, and if there is 50% it is bulk separation Purification through adsorption on zeolites is largely applied in industry today Adsorption as a separation process has become more and more commercially interesting in the course of last years (ref.2) There are many processes in which through adsorption much better separation is obtained than by means of other methods In adsorption processes the adsorbent is present in the form of granules and it is subjected to a large number of cycles of adsorption! desorption (ref 3) We have performed a series of experiments in a small laboratory apparatus with the fixed bed of the adsorbent to estimate the suitability of molecular sieve 13X for the preparation of low aromatic solvents from gasoline, white spirit and kerosene and also the optimal conditions for performing dear-omattzat ion in a large laboratory apparatus 684 Rates of adsorption in fixed bed dynamic systems may be described in two ways: with mass-transfer coefficients and with mass transfer zone concept The mass transfer zone concept has been suggested by Michaels (rei'.4) for the interpretation of kinetic data on ion exchange while the stable zone of mass transfer is being formed In order that this concept might be applied it is necessary in the course of the experiment to satisfy the following requirements: the fixed bed should be uniformly packed; the temperature and the adsorbent activity through the bed should be balanced; the velocity, the temperature and the composition of raw material should be constant; there should be no radial gradient of temperature, concentration and velocity; the temperatures of adsorbent and raw material should be equal; there should be no phase changes; there should be no chemical reactions From the experimental data on linear velocity through zeolite column and concentration the S curve is being constructed From the S-curves the height of the mass transfer zone is calculated It depends on temperature of the adsorbent bed, the size of particles, type of the raw material and the concentration of aromatics in it, as well as on the velocity of raw material through the bed (em) (1) h z = the height of mass transfer zone; h o = the height of the adsorbent column; t u = time of column saturation when the percentage of aromatics has reached the desired value in the efi'luent; t z = time needed to change the aromatics concentration from the breakthrough concentration to the concentration oi' column saturation (l-i') h z (em) (2) h NS = the unused zone height; f = exploitation grade of mass transfer zone In ideal cases it is 0.5 but usually it is less than 0.5 and it is obtained from the suri'ace under and above the S curve 685 From experimental data a specific dynamic capacity: ml aromatics adsorbed to the breakthrough 100 g of zeolite 13X and specific equlibrium capacity are determined: ml aromatics adsorbed to the saturation 100 g of zeolite 13X (4) The ratio 0(= exploitation grade of adsorbent column capacity EXPE~U1ENTAL Experiments have been carried out in a glass column of 2.5 cm diameter and 40 cm height for gasoline and 70 cm height for kerosene and white spirit The column has been packed with fresh activated molecular sieve 13X (8 to 12 mesh) Aromatics were separated from the following fractions: gasoline boiling point: 50-160 00 with % by vol aromat white spirit boiling point: 158-206 00 with 1?7~ by vol aromat kerosene boiling point: 170-260 with 18.,52% by vol aromat Experiments were carried out at different temperatures (15-50 00 ) and at different liquid velocities (0.3-1.5 em/min) Dearomatization procedure was controlled by refractivity index change UV-spectroseopy and gas chromatography were used to determine aromatics content in dearomatized products The dearomatized gasoline had less than 200 ppm per vol of aromatics, white spirit had less than 0.39%per vol of aromatics and kerosene less than 1.0% per vol of aromatics Results of experiments are presented in tables 1, 2, and figures 1, 2, 686 TABLE Experimental data on gasoline dearomatization on molecular sieve 13X c;aract;tiC ~ Column temperature t,OC 15 15 15 20 20 20 25 25 Linear velocity through zeolite column ULS' em/min 0.29 0.4 0.48 0.26 0.33 0.46 0.26 0.46 0.4 0.48 0.48 0.39 0.42 0.42 0.42 0.32 0.6 0.52 0.52 0.52 0.61 0.58 0.58 0.68 Height of mass transfer zone hz,cm 11.6 12 6.6 9.9 10.9 9.6 16.7 Unused zone height hNS,cm 5.4 6.0 4.0 5.7 6.3 11.3 Specific dynamic capacity,cm3/100 g 13X 12.8 13.5 13.4 13.6 13.2 12.2 13.5 Exploitation grade of zone height f bf = f 6.2 5.5 Equilibrium specific capacity,cm3/100 g 13X Exploitation grade of adsorbent capacity riJ,% §2 I 84 84 89 86 85 82 78 687 TABLE Experimental data on white spirit dearomatization on molecular sieve 13X C~l 20 30 30 40 40 40 50 50 Column temperature t,OC 20 Linear velocity through zeolite column ULS,cm/min 0.34 0.57 0.49 0.64 0.49 0.54 0.61 0.63 0.71 Exploitation grade of zone height 0.21 0.18 0.17 0.24 0.30 0.33 0.24 0.33 0.38 f = f' 0.79 0.82 0.83 0.76 0.70 0.67 0.76 0.67 0.62 Height of mass transfer zone hz, cm 19.6 18.7 17.3 17.2 16.7 16.7 16.9 21.8 19.2 Unused height hNS,cm 15.4 15.4 14.3 13.0 11.7 11.2 12.8 14.6 11.9 1-f Specific dynamic capacity,cm3 A/loo g 11.8 9.8 13X 12.4 12.6 13.6 12.9 13.8 12.3 13.4 Equilibrium specific capacity,cm3 A/loo g 13.8 11.5 14.2 14.5 16.7 16.0 16.2 16.6 18.3 13X Exploitation grade of adsorbent capacity rIv,% 85 86 88 84 81 80 85 74 73 If for measure of the dearomatization efficiency the height of the mass transfer zone is supposed to be as low as possible, the specific dynamic capacity as large as possible and the grade of exploitation the largest, then this is achieved for the tested gasoline at the temperature of 20 0C and linear velocity through the zeolite column of 0.26 cm/min,and for white spirit at the 688 temperature of 40 00 and the linear velocity through the zeolite column 0.49 to 0.61 em/min 0.17 6.1 0.26 6.8 0.29 6.8 0.34 0.46 9.2 10.9 (6) 0.2 0.4 0.8 0.8 em/min Fig Diagram showing relationship between the height of mass transfer zone and linear velocity through zeolite column for gasoline b.p 50-160°0 at the temperature of 20°0 For velocities from 0.2 to 0.5 em/min h z can be calculated according to equation (6) Experimental data differ from calculated ones by less than 10 "!o E u U12 h 0.24 0.'9 0.49 0.54 0.61 0.67 0.71 0.8~ 19.2 14.1 16.8 16.8 16.9 19.2 19.6 21.9 o (7) 0.2 0,4 Q6 em/min Fig Diagram showing relationship between the height of mass transfer zone and linear velocity through zeolite column for white spirit b.p 158-206 at the temperature of 40°0 689 For the velocities from 0.4 to 0.8 em/min h z can be calculated according to ~quation (7) Experimental data differ from the calculated ones by less than 10 "/0 0.32 10.9 0.67 21.4 (8) em/min Fig Diagram showing relationship between the height of mass transfer zone and linear velocity through zeolite column for kerosene b.p 170-260 0C For velocities from 0.3 to 0.7 em/min the height of the mass transfer zone can be calculated according to (8) and experimental data differ from the calculated ones by less than 10 "/0 CONCLUSION Experiments on laboratory scale have shown that successful dearomatization of gasoline 50/160, white spirit 158/206 and kerosene 170/260 is possible on molecular sieve 13X Data of laboratory experiments interpreted in terms of the height of mass transfer zone have enabled us to determine the optimal conditions of velocity and temperature for carrying out large-scale laboratory experiments, i.e for gasoline 0.26 em/min and temperature of 20 0C, and for white spirit 0.4 to 0.6 em/min and temperature of 40 0C Within the range of rates from 0.2 to 0.8 em/min the linear 690 dependence between the height of mass transfer zone and linear velocity was found: for gasoline at 20°C hZ for white spirit at 40°C hz 17.3 ULS + 17.4 ULS + 7.5 for kerosene at 40°C hz 27 ULS + Changes of veloqities through the zeolite column influence equally the change of the height of mass transfer zone for gasoline at 20°C and white spirit at 40°C REFERENCES D.W D.B G.M A.S Breck, CEP, 73 (9) (1977) 44-48 Broughton, CEP, 73 (9) (1977) 49-51 Lukchis, Chem Eng., 80 (13) (1973) 111-116 Michaels, Ind Eng Chem 44 (8) (1952) 1922-30 ... ratios Synthesis behaviour varies to some extent between the most aluminous and most siliceous groups Synthesis of the most siliceous zeolites, which tend to be hydrophobic, is promoted by many... organic species, usually basic in nature Two roles of such compounds have been indicated Firstly, the host crystal is stabilised by inclusion of guest molecules This has a thermodynamic basis and is... ultrastabilisation can further increase the ratios Si0 but these are treatments subsequent to the syntheses 2!Al203, and will not be discussed Also the Si0 ratios in Table may be 2!A1203 capable