Tổng hợp, cấu trúc, công nghệ, ứng dụng của zeolites b držaj,s hočevar and s pejovnik

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 1

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 2

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 3

Volume 4

Volume

5

Volume 6
Volume

7

Volume 8
Volume 9
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 3--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 3 to 8 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

8. 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.
1. 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-


2
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


3
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".
2. 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-


4
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) 2

+

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


5
not be discussed further. The systematic study of zeolite formation involves
other important areas such as:
Precursor species in the parent magma
Nucleation kinetics
Kinetics of crystal growth

Definition of crystallisation fields
Stabilisation of porous crystals by inclusion of guests
Templating
Habit modification
Growing larger crystals
Compositional zoning
Successive transformations
Isomorphous replacements
AI-Si ordering
Lattice defects
In later sections stabilisation of host crystals by inclusion, templating,
and the analysis of growth and nucleation kinetics will be referred to. First,
however, one may attempt to relate synthesis conditions with the Si/Al r atios in the zeolites.
3. Si/Al RATIOS IN ZEOLITES
Three dimensional frameworks based on fully linked (AI,

suo 4

tetrahedra

cover the range 0 < Si/A 1<00. A convenient but arbitrary sub-division of
phases in this range is indicated below (ref.25, 26).

1
OAluminate
sodalites;
bicchul ite

3-dimensional frameworks based only

on tetrahedra
(aluminate, aluminosilicate, silica)
4
3
2
Si/Ah 12
1Si/Ah5
Zeolites
Zeolites
Felspars
(porous)
(non-porous)
(ferrierite,
svetlozarite;
Felspathoids
porous)
(porous and
non-porous)
Zeol ites
(porous)

5
Si/Al =
Crysta11 i ne
silicas
(porous and
non-porous}

6
Much synthesis research has been concerned with the aluminous zeolites
of category 2. However there is increasing emphasis upon syntheses yielding
zeolites in categories 3 and 4, and porous silicas of category 5. Such work
has been stimulated by the discovery of shape-selective catalysts based on
ZSM-5, and other zeolites often rich in silica. To some extent the "rules"
governing synthesis of aluminous zeolites may differ from those for highly
siliceous ones. Examples of compositional ranges of each kind are illustrated
by Table 1. Special methods such as 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 1 may be
2!A1203
capable of extension by direct synthesis, as shown recently by the preparation of mordenite with Si0

equal to 18 (ref.27a). That for KZ-l,
2!AI20 3
based on one analysis (ref. 28) does not indicate the composition range but
serves to place this zeolite clearly in category 4. Also some zeolites such
as ZSM-5, -11 and -39 can be made with so little Al that they may properly

be considered as impure porous silicas, corresponding respectively with
Silicalites I and II and dodecasil-3 C .
TABLE 1
Compositional ranges in some aluminous and siliceous zeolites (ref. 25, 26)

Name

Category 2
Range in ratios
Si02/A1203

Chabazite
Faujasite
Zeolite A
Zeolite L
Soda1ite
Paulingite
Levynite(27)

2.16-7.8
2.2-6.8
2.0-6.8
2.08-7.0
2.0-10.0
2.9
3.35-4.43

Name

Category 4
Range in ratios
Si02/A1203

Beta
ZSM-5
ZSM-ll

ZSM-12
ZSM-23(EU-13)
ZSM-48
Nu-1
Fu-l
Zeta-l
Zeta-3
KZ-1(28)
KZ-2 (28)(Nu-l 0)

30-75
25-1000
78-1000
45-160
55-217
870-1340
40-120
20-40
25-32
60-74
108
25-492


7
3.1. Synthesis behaviour of aluminous zeolites
Observations on the synthesis of zeolites which have some generality are
of much interest. In this section several observations of this type are made
with aluminous zeolites particularly in mind. In chapter 3.2. some further
observations relating to siliceous zeolites and porous crystalline silicas will

be made.
(j) Zeolite formation has been observed over a considerable range of

temperatures. Based on geological evidence an upper limit of about 350

0C

has been suggested (ref. 29 ). In direct synthesis analcime has been made up
to at least 366°C, mordenite to 430
to 375

0C

0C,

clinoptilolite to 370

0C

and ferrierite

(ref.30). The higher the synthesis temperature the smaller tends

to be the water content and intracrystalline porosity of any zeolites which
crystallise. Yields of the zeolite also decline as the temperature increases
beyond an optimum for the given zeolite type. The most porous zeolites
(faujasite, chabazite and zeolites A, RHO and ZK-5) with pore volumes in
3
3
the range NO. 45 to 0.53 cm per cm of crystal do not form at temperatures much above 100

0C.

Silica-rich phases such as mordenite ZSM-5, -11

and -39, and KZ-l and -2 or porous crystalline silicas such as melanophlogite and dodecasil -3C, all with rather low intracrystalline porosities, are
usually best made at temperatures in the range 100 to 200

0C.

It is not al-

ways the most porous zeolites which are required but rather those with the
right size of channel, window or cavity for the applications envisaged.
(ii) For aluminous zeolites the grater the concentration of alkali the

smaller tends to be the ratio Si0

The lower limit is 2.0. This
2/A1203•
behaviour is shown in Fig. 1 for the synthesis of chabazite- and edingtonitetype zeolites (ref.3l), respectively K-G and K-F, and in Fig. 2 for a range
of zeolites (ref.32). In Fig. 1 the horizontal dashed lines give the ratios
SiO

of the parent gels. For these initial compositions the crystal
Z/AI 203
phases are much richer in Al than the gels, so that it is largely Si0 which
2
is retained in the mother liquor.
(iii) When the concentration of alkali hydroxide (KOH) was constant but

the volume of the mineralising alkali solution was increased keeping the
amount of gel constant, the yields both of K-G (from 1 IVI KOH) and of K-F
(from 6 M KOH) diminished, as shown in Fig. 3 (ref.3l). This indicates a


8

o
c:::

101------------ -

-

-- ---- ---

8-

u
~

CD

-0

e0.

.~

"
c:::

6---------- -- -----------------

OL-_.L.I
1

--L.
I

3

--L
I

5

Molality of KOH solution

.L..I

1

........
I

~

9

Fig.'
Effect of concentration of KOH solution upon the ratio
Si0 21 AI203 in the resultant crystals. The ~roducts
are primarily chabazite type K-G and edingtonite type K-F.' lJ

type of solubility of the zeolite in the mother liquor but one in which the
dissolved species are not necessarily the lattice-forming units of the crystal.
(iv) Ostwald's law of successive transformations is obeyed in that the
species first produced may be progressively consumed and replaced by a
second species which is more stable and the second by a third which is still
more stable until the final most stable phase appears. This tendency, to
move down the ladder of thermodynamic stabilities appropriate to the experimental conditions rung by rung, is illustrated in Fig. 4 (ref.33) for the


9

5------------------------,
41-

•

II>

~>-

3"-

...
0

s


......

u;

•I
•

~
+

.~

a

2-

0

'<>

~B·O

0
o~lC9q" o\j.,

o~

~%

0

0"

0

.,

0

Q

Q

"

~

1-

I

I

08

04

I

12

o

0

0.,

I

16

I

20

I

24

I

28

o

()

I

32

( No 20 - A1 20 3) 15.°2 In gels

Fig. 2. Si/AI ratios in various zeolite crystals correlated with excess
alkali. expressed as ttle ratio (Na2o-AI203)/Si02 in the parent gels~l~)
• = mordenite
+ = zeolite L
O. () = faujasite
0 = Na-P
= Na-chabazite
• = K-M
I2l = K-chabazite
-. = Na-A
•
= K, Na-chabazite
• = analcime
6, • = erionite
x = K-F

o

crystallisation of faujasite from gel in presence of some N( CH

40H followed
3)
by its replacement by mazzite-type ZSM-4. The behaviour indicates that the

less stable the phase the greater the chance that initially it will nucleate and
grow, but the less its chance of subsequent survival.
(v) A tectosilicate grown from aluminosilicate gels having a wide range
of compositions is often intermediate in composition between the extremes of
those of the gel. This behaviour is illustrated in Fig. 5 (ref.34) for zeolites
K-F, K-G and K-M (the latter is probably a merlinoite-type zeolite) and the
felspathoid K-N (kalsf lite}. If the ratio Si0

in the parent gel is less
2/A1203
than 2 (the Lowenstein or AI-O-AI avoidance limit) then this ratio increases

to at least the Lowenstein limit in the crystals. This means a mother liquor


10

14f--

02~

I

20

I

I

60

I

I

I

I

100
140
cm 3 KOH

180

220

Fig. 3 Effect of the volume of KOH solution of constant concentration upon the yield of crystallization products from metakaolinite. 0,
1 M KOH, .giving chabazite type K-G; 6, 6 M KOH, giving edingtonite
type K-F. (:lI1)

._------_.,-100

'i

~80

z-c

~ 60

o

10

20

30

40

50

Crystallization lime ( h )

60

70

80

Fig. 4- Yields of crystals against time at about 1000C. The ratio
TMA+I(TMA + + Na +) is such as to yield ultimately only ZSM-4
Fauiasite (0) appears first and is then progressively replaced by
ZSM-4. ('3~)


II
with lower ratios Si0

than in the parent gel. Conversely, for such
2/A120 3
aluminous tectosilicates, if the parent gel is silica-rich, the crystals are
more aluminous and the mother liquor is more siliceous than is the initial
mixture (see Fig. 1). As yet there is not much information on the nature of

species co-existing in the mother liquors after crystallisation. Some information might be obtained by 29Si and 27 Al nv rn r-,
s

S,02 / A1203

Fig. 5 Composition changes when alkaline aqueous potassium
aluminosilicate gels were crystallized.
Horizontal scales give the
ratios Si0 2 / AI 203 in parent gels (lower) and in products (upper).(34-)
K-F = edingtonite type zeolite
K-M = phillipsite type zeolite
K-G = chabazite type zeolite
K-N = felspathoid kalsilite


12
3.2. Synthesis behaviour of siliceous zeolites
Most studies of silica-rich zeolites and porous crystalline silicas have
been concerned with zeolite ZSM-5 and its variants, with more limited studies of other phases. Several aspects of their synthesis are summarised
below:
(l) As noted in chapter 3.1. (I ) the temperatures needed to prepare sili-

ceous zeolites and porous crystalline silicas in a convenient time are higher
than those required to make aluminous zeolites.

(ji) At least for ZSM-S, crystallising as the (NH

NPr
(Pr =
4,
4)-form
propyl), the more siliceous the parent mixture the smaller the induction
period and the greater the rate of crystallisation (ref. 35). It is desirable
to know how general this behaviour is for other siliceous phases like those
in Table 1. It is certainly the case for them all that they nucleate and grow
only from siliceous parent magmas.
(iii) Successful syntheses have nearly all required the addition of organic

compounds to the crystallising magmas. These organic compounds are frequently bases or salts of bases. Sometimes the only base needed is the organic one without any requirement for an inorganic base such as NaOH, KOH
etc. For example, ZSM-5 needs as base only NPr 40H and ZSM-ll needs only
NBu
(Bu = butyl). This suggests a templating action of the organic base
40H
in nuc leation of zeolite (chapter 4). However, a surprising number of other
organic compounds have also been reported to aid the crystallisation of silicarich zeolites, as illustrated for several zeolites in Table 2. These compounds
are of different shapes and sizes which makes douptful a universal explanation in terms of templating.
(tv) The synthesis from silica gel of melanophlogite, a porous crystalline
silica having the framework structure of clathrate hydrate of type 1, was
achieved (ref.38) under high pressures of CH

0C

+ CO
(150 bar at 170
for

4
2
0
6 weeks). High pressures of such gases, around 0 C are also required to

prepare clathrate gas hydrates (ref. 39 ) .
The guest molecules (here CH 4 and CO

by occupying a sufficient frac2),
tion of the cavities, form a solid solution in the growing crystals and thereby
lower the chemical potential of the Si0

2

in the host lattice. This stabilisation


13
TABLE 2

Organic species used to aid synthesis of some zeolites

Zeolite ZSM-S(36}
NPr 40H

(n-C 4H g)2 NH

NPr 3

NH 2(CH2}SNH 2

NH 2C 2H4NH2

NH 2(CH 2}3 NH2

OHC 2H4NH2

CH

Zeolite ZSM-S (cont.)

n-C 3H7NH2

NEt 40H

OHC 3H6NH

Zeolite ZSM-S (cont.)

Hexanediol

f\NH

0"-.-/

2

-'9

C(CH 20H)4

Dipropyl enetri amine

3

Triethylenetetramine

NH 3+C 2H SOH

C2H SOH
H2- r H- yH2
OH OH OH

Diethylenetriamine

r

Zeolite KZ_l(28)

Zeolite KZ-2(28)

Pyrrolidine.
CH -CH-CH
3

I

NH 2

3

(CH 3)2NH
Zeolite ZSM-39(37)

(C

2HS}NH
CH 3(CH2}3- NH2
NH 2CH2CH2

NH

NH 2CH 2CH 2

Ferrierite-type Zeolites(36)

(CH 3)4NOH

Cho1i ne

(C2HS}4NOH

Pyrrolidine

Pyrrolidine

NH -CH2CH 2-NH2
2
NH2--CH2CH2CH2--NH2

(CH 3 ~ikH2cH20H

(continued)


14

Table 2 (continued)
Ferrierite-type Zeolites(36)(cont.)
NH2--CH2CH2CH2CH2--NH2
2,4-pentanedione
N-methylpyridinium hydroxide
Piperidine
Alkylpiperidine

of the porous crystals by guest molecules has a thermodynamic basis (ref.
40) which is given in the next section (Mineralising catalysts), and can also
be modelled in a statistical thermodynamic treatment (ref.41, 42).

4. MINERALISING CATALYSTS
It has been recognized for a long time that water is an excellent mineral-

iser, assisted in this role by OH

ions in particular. This mineralising prop-

erty is the basis of hydrothermal chemistry, and is exerted in the following
ways (ref. 43) .
(i) Water may be incorporated into anhydrous glasses, melts and solids
by chemisorption which breaks Si-O-Si and Si-O-Al bonds which may then

reform with different partners. Chemical reactivity is thereby enhanced and
magma viscosity lowered.
(ii) Water is a good solvent which assists disintegration of solid components and facilitates transport and mixing.
(iii) High pressures of water can modify phase equilibrium temperatures.
(iv)

In zeolite synthesis zeolitic water is essential as a guest species

which stabilises the host lattice. The zeolitic water may then be removed
leaving the unchanged anhydrous zeolite so that in this respect the water
conforms with the classic definition of a catalyst.
Hydroxyl ion greatly augments factors (i) and (Li ) above; indeed in alka-


15
line media the solubility of silica increases nearly exponentially with concentration of alkali, and according to the ratios M

20/Si02

in the resultant mix-

ture a range of silicate anions may appear of various degrees of oligomerisation (ref.44). Alumina and other amphoteric oxides are also soluble in alkaline aqueous media.
In the case of alumina there is at high pH minimal oligomerisation, the
dominant anion always being Al(OH)4

(ref.45). Alkaline media thus enable

ready mixing of reactants and so facilitate nucleation and crystal growth.
It was mentioned in (tv) above and in chapter 3.2. (iv ) that guest mole-

cules can stabilise porous host lattices, by occupying cavities and channels
within the host. If the gas phase behaves ideally, the lowering in chemical
potential, II

p, of structural units composing the host lattice, H, by a guest

molecule, A, is given by (ref.46)

(lJ

where VH = partial molar volume of lattice-forming units of the host;
P = total pressure; M
rnA = molecular weights respectively of the lattice
H,
forming unit and of the guest molecule, A; p A = partial pressure of A;
xA = the amount of A in the host crystals in gig at pressure pA •
The term VHP is normally very small and can be neglected compared with
the integral, the value of which increases for a given pA the more rectangular the sorption isotherm. The integral may be evaluated graphically as the
area under plots of x AlpA against the partial pressure of A up to pA' II P
is itself given by,
(2)

where ,uH is the chemical potential of lattice-forming units of the host in
presence of the uptake xA of guest and

u~

is this chemical potential for the

guest-free host.

Equation 1 may be examined with the synthesis of meIanophIogite in mind
o

(ref. 38 ), at 170 C and under a pressure of 150 bar of CH4 as guest mole-


16
cule , At this temperature the pressure of water vapour is 7.226 bar. When
both dodecahedral and tetradecahedral cavities are fully occupied the satura,is 0.04638 g per g of Si0 for CH or 0.05217 g per g
2
4
sat
0
for H
In a mixture of CH at 150 bar and H
at 7.226 bar at 170 C
4
20
20.
and assuming that Langmuir's isotherm equation is valid
tion uptake, x

aPCH

X
CH
4

4

SCH

~sat

4

+

77sat

2

SH 0"

+

2

apCH

+

4

bP H

xH 0
2

apCH

4

bP H 0
2

(3)

0
+

bP H 0
2

When p is in bars a and b are in bars -1. Then under the above conditions
of pressure and temperature the values of B
and B
CH
H
. 4
2
if a = b for these two guest molecules, as glven below:
a

=b

(bar

-1

)

B

CH

4

0.948
0.953
0.954
0.954

1

10
100
1000

B H2

° are respectively,

°

0.0457
0.0459
0.0460
0.0460

If each of these two guest molecules was sorbed in the absence of the other,
0

again at 170 C and at a pressure of 150 bar for CH

and 7.226 bar for H
4
20,
eqn, 1 assuming Langmuir's isotherm equation for sorption, gives the following values of b. p. for different values of a or b:
(a) CH

4:

1

a (bar

10
100
1000

-1

)

-f::,;U

(cals per' mole
Si0
2)

765
1116
1467
1818

(cals per mole
Si0
2)
1

10
100
1000

321
655
1004
1355


17
o

The CH

at 150 bar and 170 C for a = 1 stabilises the melanophlogite
4
o
structure more than H

at 7.226 bar and 170 C for b = 10. In the non20
polar structure of melanophlogite one could anticipate comparable values of
the constants a and b , in which case for CH 4 + H
under the

~buve

the calculation indicates
20
conditions that little water but much methane would be en-

capsulated during growth of guest-stabilised melanophlogite.
In the synthesis of melanophlogite, cristobalite was often the initial product, but was gradually replaced by melanophlogite (ref.38). Thus, the
lowering of chemical potential of Si0

by encapsulation of (CH + CO + H
2
4
2
20
rendered the melanophlogite more stable than cristobalite. The replacement
accords with Ostwald's rule of successive transformations referred to in
chapter 3. i ,
In the syntheses of all the zeolites in Table 2 the added organic species
are, like the water, potential guest molecules within the porous zeolite
frameworks, and for some of these species at least their affinity for the
zeolite will exceed that of water. They can therefore be the effective stabilisers of the host crystals when their sorption isotherms are more rectangular
than those of water. The thermodynam ic measure of stabilisation when a
number of guest species are sorbed simultaneously is given by the following
extension of eqn. 1:

X

_A_
PA

dp

A

Thus the stabilisation measured by

1
+ -m-

S
B

x

B
p -pdp B + ..• ]

BOB

D.

,,u

(4)

depends on the sum of a series

of integrals each involving the sorption isotherm of one guest in the mixture
of guest species. Accordingly a thermodynamic basis exists for part at least
of the role of various organic additives in promoting zeolite crystallisation.
The highly siliceous zeolites become progressively more organophilic and
more hydrophobic as the silica content increases. This means in line with
the above treatment, that zeolitic water stabilises them less and intracrystalline organic species more. It

i~

usual in these very silica-rich zeolites to

find organic species within the structure in amounts grater than that required
as cations to neutralise framework charge (eqs , ZSM-5 and -11). For silica-


18
rich near-hydrophobic zeolites the use in synthesis of high pressures of
"help-gases" deserves further study.
Among aluminous zeolites on the other hand the crystals are hydrophilic
and zeolitic water gives very rectangular isotherms and so, on its own is a
very effective stabi Iiser . This does not mean that organic bases do not sometimes play an important part in promoting their formation, but their role is
based less upon displacing zeolitic water as primary stabi Iiser- and more
upon specific effects such as templating, considered below.
5. TEMPLATING
It has been found that the kind of zeolite which crystallises from alkaline

aqueous gels can be strong ly influenced by the type of cation present (ref. 47) .

(il For example, sodic environments favoured sodalite and cancrinite

hydrates, gismondine-types (Na-P), gmelinites, faujasites and zeolite A.
(ii) On the other hand mordenite formed in sodic , calcic or strontiumcontaining environments. Edingtonite and phillipsite-type zeolites and the
analcime framework topology were also obtained from various cationic environments.
(iii) Chabazite-types and zeolite L syntheses were promoted by potassic
and zeolites Li-AB\v and Li-H by lithic environments.
(iv) Some zeolites grew best when two or more cations were present.
Thus' zeolite EAB (T]\iIA-E), offretite and mazzite are favoured by mixtures of
Na + and TMA + (TMA

=

tetramethyl ammonium).

Quaternary ammonium ions are regarded as structure-making in aqueous
solutions in that the organic ions are believed to serve as the centres of
tiny hydrogen-bonded water "icebergs". Small inorganic cations on the other
hand are regarded as structure-breaking in aqueous solutions. However, they
are energetically hydrated and so may hold minute water clusters. One may
envisage situations in which the water associated with cations is at least in
part displaced by silicate, aluminate or aluminosilicate species in solution.
The cations may thus serve as centres or templates around which these
species aggregate, forming, with the cations, precursors to specific germ
nuclei. If so, the cations are tending to function as templates with a structure-directing role.


19
While the above evidence is qualitative, there are examples where templating appears to be the only logical explanation. Thus, soda lite was made
from an aluminosilicate hydrogel in which TMA was the only cation (ref.48).

There was one TMA + cation per soda lite cage, and the Si/A I ratio was 5 (as
required for one cation per cage). The TMA + ion is much too large to enter
or leave a soda lite cage after the cage has been formed, so that the cage
must form around the cation, either as a species in solution, or when the
cation is adsorbed on the surface of a growing crystal.
A templating role has also been recorded for TMA + when zeolite A is
made in its silica-enriched forms (often termed ZK-4) by adding TMAOH to
the crystallising mixture. 13C nv rn r , served to differentiate between TMA +
c

in soda lite cages and TMA + in the large 26-hedral cavities (ref. 49). The ion
was located only in the soda lite cage for materials containing one TMA + per
unit cell. As the TMA

+

content increases, the sodalite cages remain com-

pletely filled (one TMA + per cage) and the remainder spill over into the
large cavities. As with TMA-sodalite the TMA + ion can be incorporated in
the soda lite cage only if the cage grows around it because the ion is too
large to enter or leave completed cages.
As a further example, the role of the polyelectrolyte

in promoting the formation of the one-dimensional channel structures gmelinite and mordenite can be mentioned (ref. 50). For appropriate contents of
the polyelectrolyte in the reaction mixture at 90

0C

synthesis temperature,

o

chabazite (a cage structure) was replaced by gmelinite, while at 170-180 C
the compact analcime structure was replaced by mordenite. Moreover, the
organic nitrogen content per unit cell was near that calculated for the polyelectrolyte chain stretched out along each linear channel of either zeolite.
The situation strongly recalls the clathration of long n-paraffin chains by
urea which forms a hydrogen-bonded framework around stretched out paraffin
chains occupying parallel one-dimensional channels.
Anions may also have a structure-directing influence (ref. 51). Thus when
salt-bearing sodalites and cancrinites were made from kaolinite (2 g), 200