Cundy Dan Cox, 2005

Microporous and Mesoporous Materials 82 (2005) 1–78
www.elsevier.com/locate/micromeso
Review
The hydrothermal synthesis of zeolites: Precursors,

intermediates and reaction mechanism
a,* b
Colin S. Cundy , Paul A. Cox
a
Centre for Microporous Materials, School of Chemistry, University of Manchester, P.O. Box 88, Sackville Street,
Manchester M60 1QD, United Kingdom
b
School of Pharmacy and Biomedical Sciences, University of Portsmouth, St Michael’s Building, White Swan Road,
Portsmouth PO1 2DT, United Kingdom
Received 5 November 2004; accepted 11 February 2005

Available online 8 April 2005
Abstract
An account is presented of the mechanistic aspects of hydrothermal zeolite synthesis. The introduction provides a historical and
experimental perspective and is followed by a summary of proposed mechanisms and associated modelling studies. The central sec-
tion of the review contains a description of the most probable mechanistic pathways in zeolite formation. In this, the reaction stages
of the induction period, nucleation and crystal growth are examined in chronological sequence. Finally, particular aspects of the
synthesis process such as the constitution of growth species, template–framework interactions and the nature of zeolite solubility
are treated in more detail.
Emphasis is placed upon the chemical basis of zeolite synthesis. Fundamental to this are the TAOAT bond-making and bond-
breaking reactions which establish the equilibration between solid and solution components. The consequent generation of order,
driven by energy differences and strongly moderated by kinetic limitations, is essentially one of continuous evolution. However, the
discreet step of nucleation provides a discontinuity in which isolated regions of local order are superceded by the establishment of a
periodic crystal lattice, capable of propagation. Crystal growth occurs through an in-situ, localised construction process from small,
mobile species ordered by the participating cations.
The process of hydrothermal zeolite synthesis can be most adequately explained by a mechanism based upon the solution–medi-
ation model, whether or not there is a visible liquid phase. The common presence of mobile species emphasises the overall similarity
of zeolite synthesis reactions so that the need to distinguish any separate ‘‘gel rearrangement’’ or ‘‘solid-phase transformation’’
mechanism becomes unnecessary.
2005 Elsevier Inc. All rights reserved.
Keywords: Hydrothermal synthesis; Nucleation; Crystal growth; Modelling; Mechanism
Contents
Part I: Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1. History of hydrothermal zeolite synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
*
Corresponding author. Tel.: +44 161 200 4512; fax: +44 161 200 4559.
E-mail address: colin.cundy@manchester.ac.uk (C.S. Cundy).
1387-1811/$ - see front matter 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.micromeso.2005.02.016
2 C.S. Cundy, P.A. Cox / Microporous and Mesoporous Materials 82 (2005) 1–78
1.2. Scope and structure of this review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2. Experimental observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3. Summary of proposed mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.1. Richard Barrer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.2. The early work of Breck and Flanigen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.3. Kerrs recirculation experiment and the work of Ciric. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.4. Studies at Leningrad . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.5. Overview—1959 to 1971 and beyond . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.6. Introduction of organic templates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.7. Chang and Bell and after . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4. Modelling the processes of zeolite synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4.1. Mathematical models of synthesis reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4.2. Molecular modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.2.1. Modelling zeolite-template pairs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.2.2. Cluster calculations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Part II: Synthesis mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5. The induction period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
6. The evolution of order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
6.1. The nature of the amorphous material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
6.2. Primary and secondary amorphous phases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
6.2.1. The Montpellier study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
6.2.2. Related investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
6.3. Further evidence for pre-crystalline order from synthesis studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
6.4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
7. Nucleation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
7.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
7.2. General considerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
7.3. Determination of zeolite nucleation patterns from measurements on the resulting crystals . . . . . . . . . . . . . . . . . 16
7.3.1. Studies under isothermal conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
7.3.2. Ageing studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
7.4. The use of seed crystals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
7.5. Autocatalytic nucleation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
7.6. Nucleation in zeolite systems—The nature of the reaction sol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
7.7. Nucleation in zeolite systems—Homogeneous or heterogeneous? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
7.8. Nucleation in zeolite systems—Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
7.9. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
8. Crystal growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
8.1. Experimental methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
8.2. Experimental observations—Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
8.3. Experimental observations—Studies of macrocrystalline systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
8.4. Experimental observations—Studies of nanocrystalline systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
8.5. Size-dependent growth of nanocrystals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
8.6. Growth models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
8.7. Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
8.8. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Part III: Key topics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
9. The nature of growth species and the role of aggregation processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
9.1. Growth from soluble, pre-fabricated units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
9.2. Growth from simple species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
9.3. Mineralising agents other than hydroxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
9.4. Growth from particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
9.4.1. Particle aggregation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
9.4.2. Chemical and physical consequences of an aggregation mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
9.5. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
10. Solid state transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
10.1. Hydrothermal synthesis in the presence of a liquid phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
10.2. Hydrothermal synthesis in the apparent absence of a liquid phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
C.S. Cundy, P.A. Cox / Microporous and Mesoporous Materials 82 (2005) 1–78 3
10.3. Non-aqueous syntheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

10.4. The role of water in apparent solid state transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
10.5. Solid state transformations at high temperatures and pressures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
10.6. Summary and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
11. Ageing effects in zeolite synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
11.1. Ageing as a means to control product phase purity and crystal size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
11.2. Rationalisation of ageing effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
11.3. Detailed analyses of ageing-related effects in silicalite synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
11.4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
12. X-ray amorphous zeolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
12.1. XRD evidence for ‘‘X-ray amorphous zeolites’’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
12.2. IR evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
12.3. Evidence from other physical measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
12.4. Evidence from catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
12.5. Evidence from the synthesis process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
12.6. Other amorphous materials related to zeolites—zeolite degradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
12.7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
13. Template–framework interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
13.1. Geometric matching. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
13.2. Template classification and versatility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
13.3. Structure blocking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
13.4. Variations induced by heteroatoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
13.5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
14. Solubility and supersaturation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
14.1. Zeolite solubility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
14.2. Zeolite solubility as a function of base concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
14.3. Supersaturation in relation to zeolite crystal growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
14.4. Thermodynamic vs. kinetic factors in zeolite synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
14.5. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
15. Zeolite dissolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
15.1. The kinetics of dissolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
15.2. Morphological and compositional changes on dissolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
15.3. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
16. Metastability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
16.1. Precursors, intermediates and co-products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
16.2. Layer structures as transients and precursors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
16.3. Conversion of one zeolite into another . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
16.4. Ostwald ripening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
16.5. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
17. Optimisation of zeolite syntheses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
17.1. Comparisons of zeolite synthesis reaction rates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
17.2. Procedures for improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
17.2.1. Reaction optimisation (and its limitations). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
17.2.2. Addition of seed crystals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
17.2.3. Additives and ‘‘promoters’’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
17.2.4. Microwave synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
17.3. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
18. Relationship of zeolite synthesis mechanism to that of other porous materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
18.1. Zeolites and clathrate hydrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
18.2. Zeotypes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
18.3. Microporous vs. mesoporous structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
18.4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
19. Summary and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Acknowledgements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Appendix A. A chemical model for the crystal growth of zeolite molecular sieves . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Part I: Background [26,27] series, have periodic structures with far larger
pore sizes (up to 200 Å) but are not conventionally
1. Introduction crystalline [28–30].
Investigative work aimed at gaining an understanding
1.1. History of hydrothermal zeolite synthesis of the synthesis process has its origins in the 1960s.
These studies have continued up to the present day,
The history of man-made zeolites can be traced back spurred on at various points by discoveries of new mate-
to the claimed laboratory preparation of levynite by St rials, advances in synthetic techniques, innovations in
Claire Deville in 1862 [1]. However, zeolite synthesis as theoretical modelling methods and, especially, by the
we know it today had its origins in the work of Richard development of new techniques for the investigation of
Barrer and Robert Milton, commencing in the late reaction mechanisms and the characterisation of prod-
1940s. Barrer (principally at Imperial College, London) ucts. It is the purpose of the present Review to offer,
began his work by investigating the conversion of for the case of zeolites, an account of such exploratory
known mineral phases under the action of strong salt and background work.
solutions at fairly high temperatures (170–270 C).
Among the products, species P and Q [2–4] (isostruc- 1.2. Scope and structure of this review
tural variants) displayed unique characteristics and rep-
resented the first synthetic zeolite unknown as a natural In an earlier survey [31], a summary was given of the
mineral. These materials were later found to have the main discoveries and advances in thinking in the field of
KFI structure [5] determined subsequently for zeolite zeolite synthesis from the 1940s up to 2002. That ac-
ZK-5 [6,7]. Robert Milton (in the Linde Division of count was principally concerned with the pattern of
the Union Carbide Corporation, Tonawanda, New discovery and the consequent progression of ideas.
York) pioneered the use of more reactive starting mate- Discussion of the mechanism of zeolite synthesis was
rials (freshly precipitated aluminosilicate gels), enabling limited to this evolutionary context. This present review
reactions to be carried out under milder conditions and attempts to expand this critical argument and to de-
leading to the discovery of zeolites A [8] and X [9]. By scribe in detail the most probable steps by which amor-
1953, Milton and his colleagues had synthesised 20 zeo- phous aluminosilicate reagents are converted to
lites, including 14 unknown as natural minerals [10]. crystalline molecular sieves. In addition to summarising
Following the foundations laid in the 1950s, the next earlier proposals, it will be necessary to bring forward
decade saw many significant developments. Earlier work some further ideas which are perhaps new in the current
on zeolite synthesis had utilised only inorganic reaction context. This should clarify the link between nucleation
components but in 1961 the range of reactants was ex- and crystal growth by considering the chemical steps
panded to include quaternary ammonium cations [11– which are common to both. Fortunately, when this is
13]. The introduction of organic constituents was to done the picture becomes simpler rather than more com-
have a major impact upon zeolite synthesis and the plex and the need to make any differentiation between,
key step followed quite rapidly with the disclosure in for example, ‘‘solution-mediated growth’’ and ‘‘gel rear-
1967 of the first high-silica phase, zeolite beta [14], whilst rangement’’ finally disappears. The text is confined to
the archetypal high-silica zeolite, ZSM-5, was discov- hydrothermal methods of synthesis and concentrates
ered in 1972 [15]. on aluminosilicate zeolites, mentioning alternative zeo-
There has subsequently been a large rise in the num- types or other porous materials only when this is neces-
ber of known synthetic zeolites [16] and also the discov- sary to illustrate or broaden the main argument.
ery of new families of zeolite-like or zeolite-related However, it seems very probable that the main features
materials [17]. The latter ‘‘zeotypes’’ may be represented observed for zeolites will also be found in the synthesis
by the microporous alumino- and gallo-phosphates (Al- of closely related materials, modified in some cases
POs and GaPOs) [18–20] and titanosilicates (such as through the differences in composition, structure, polar-
ETS-10) [21–23]. Such materials display great composi- ity and solution chemistry.
tional diversity and frequently have frameworks un- The sections of this survey fall into three groups. In
known for zeolites. This increased structural flexibility Part I and following the above brief historical introduc-
has its origins in the available spectrum of heteroelement tion, the experimental observations associated with a
atomic radii, bond lengths and bond angles, and in the typical hydrothermal zeolite synthesis are outlined and
emergence of coordination numbers greater than four. the various interpretations which have been advanced
Even greater divergence from the norm of microporous to explain them summarised (Sections 1–3). Section 4 re-
aluminosilicates is seen in a major new class of zeolite- views work on modelling the processes of zeolite synthe-
related phases discovered in the early 1990s. Mesopor- sis. Part II represents an attempt to put forward a
ous materials, synthesised with the aid of surfactant detailed and self-consistent view of the most probable
molecules and typified by the M41S [24,25] and SBA mechanistic pathways in zeolite formation, presented
within the overall chronology of the hydrothermal syn- framework are imported in an oxide form. These oxidic
thesis reaction, i.e. the induction period and the nature and usually amorphous precursors contain SiAO and
of the amorphous material (Sections 5 and 6), the mech- AlAO bonds. During the hydrothermal reaction in the
anism by which zeolite crystals are nucleated (Section 7) presence of a ‘‘mineralising’’ agent (most commonly an
and the mechanism of zeolite crystal growth (Section 8). alkali metal hydroxide), the crystalline zeolite product
In Part III (Sections 9–18), some key problems and (e.g. zeolite A) containing SiAOAAl linkages is created.
questions associated with zeolite synthesis are addressed Since the bond type of the product is very similar to that
in more detail, leading, finally (Section 19), to some present in the precursor oxides, no great enthalpy
overall conclusions. change would be anticipated. In fact, the overall free en-
For further general information on the subject of zeo- ergy change for a zeolite synthesis reaction is usually
lite synthesis, the reader is referred to the standard text- quite small, so that the outcome is most frequently
books [17,32–35] and recent reviews [31,36–43]. kinetically controlled [31,43–46].
Kinetic control is a pervading influence throughout
zeolite synthesis, where the desired product is frequently
2. Experimental observations metastable. Much of the know-how in this industrially
important area centres around choice of the exact condi-
A typical hydrothermal zeolite synthesis can be de- tions for product optimisation, so that the required
scribed in briefest terms as follows: material can be prepared reproducibly and to the same
specification [47]. Such considerations will often influ-
1. Amorphous reactants containing silica and alumina ence the choice of starting reagents. Whilst these may in-
are mixed together with a cation source, usually in clude the simple oxides or hydroxides mentioned above
a basic (high pH) medium. (e.g. precipitated silica or alumina trihydrate), it is also
2. The aqueous reaction mixture is heated, often (for very common for the reagents to represent some degree
reaction temperatures above 100 C) in a sealed of pre-combination, as for example in sodium silicate
autoclave. solution or solid sodium aluminate. These materials
3. For some time after raising to synthesis temperature, may represent advantages in cost or ease of processing
the reactants remain amorphous. but may also offer optimum routes to particular materi-
4. After the above ‘‘induction period’’, crystalline zeo- als, since flexibility in the choice of reagents enables
lite product can be detected. equilibria to be approached from different directions.
5. Gradually, essentially all amorphous material is This may offer kinetic benefits, such as the preferred
replaced by an approximately equal mass of zeolite nucleation of one phase over another in situations where
crystals (which are recovered by filtration, washing mixtures may otherwise co-crystallise. A good general
and drying). appreciation of the field of practical zeolite synthesis
can be obtained from the handbook issued by the Syn-
This is illustrated schematically in Fig. 1. The ele- thesis Commission of the International Zeolite Associa-
ments (Si, Al) which will make up the microporous tion (IZA) [48].
Fig. 1. Hydrothermal zeolite synthesis. The starting materials (SiAO and AlAO bonds) are converted by an aqueous mineralising medium (OH
and/or F) into the crystalline product (SiAOAAl bonds) whose microporosity is defined by the crystal structure.
3. Summary of proposed mechanisms 3.2. The early work of Breck and Flanigen
The brief summary given above (Section 2) outlines In 1960, Flanigen and Breck reported [51,52] a study
the transformation of an amorphous, aqueous in which XRD measurements were employed to follow
aluminosilicate gel under the action of heat into a crys- the crystallisation with time of zeolite Na-A (at
talline zeolite product. In the present section, an over- 100 C) and Na-X (at 50 and 100 C). They showed
view is presented of the main suggestions which have the now-familiar S-shaped growth curves and described
been put forward to explain these experimental observa- an induction period followed by a sudden rapid growth.
tions. The historical aspects of this topic have been dis- The morphological changes observed [53] were inter-
cussed more fully in our earlier account [31] and some of preted as a successive ordering of the gel as crystallisa-
the main mechanistic proposals are summarised in Ta- tion proceeds, leading to a conclusion that crystal
ble 1. The details of the synthesis mechanism form the growth takes place predominantly in the solid phase.
subject of Part II (Sections 5–8). Their conclusions may be summarised as follows
[31,51,52]:
3.1. Richard Barrer 1. Extensive heterogeneous nucleation occurs during

formation of the highly supersaturated gels.
The first consideration of synthesis mechanism was 2. The nuclei do not necessarily represent a unit cell but
that given by Barrer, Baynham, Bultitude and Meier may consist of more preliminary building units of
in 1959. The discussion section of a paper [49] in which polyhedra (e.g. the hexagonal prism) as suggested
a wide variety of alumino-, gallo- and germano-silicates by Barrer et al. [49].
were synthesised begins as follows: 3. During the induction period, the nuclei develop to a
‘‘The formation of diverse kinds of structural frame- critical size and then grow rapidly to small and uni-
work leads to questions as to the mechanism of growth. form sized crystals.
The phases are often obtained reproducibly in yields 4. Growth of the crystal proceeds through a type of
nearing 100% and the free-energy balance between the polymerisation and depolymerisation process (break-
many possible aluminosilicate nuclei must be delicate. ing and remaking Si,AlAOASi,Al bonds), catalysed
The development of elaborate and continued space pat- by excess hydroxyl ion and involving both the solid
terns by progressive additions of single (Al,Si)O4 tetrahe- and liquid phases (although the solid phase appears
dra is difficult to imagine, particularly in the case of very to play the predominant role).
open zeolite structures. The formation of these frame-
works is, however, much more easily visualised if in the In a subsequent review [54], Breck described zeolite
aqueous crystallising magma there are secondary build- formation in the following terms: the gel structure is
ing units in the form of rings of tetrahedra or polyhedra. depolymerised by hydroxide ions; rearrangement of
These may pack in various simple coordinations to yield the aluminosilicate and silicate anions present in the
different aluminosilicates.’’ hydrous gel is brought about by the hydrated cation
Examples of some possible ions were then tabulated: species present; tetrahedra re-group about hydrated so-
rings of 3–6 tetrahedra, the double-4-ring, the double-6- dium ions to form the basic polyhedral units (24-hedra);
ring (or 3 4-rings). It was further pointed out that such these then link to form the massive, ordered crystal
units could give rise to more complex ones, such as the structure of the zeolite.
linking of six 4-ring anions to give the nosean-sodalite
cubo-octahedral unit. Barrer returned to this theme in 3.3. Kerr’s recirculation experiment and the work of Ciric
a later review, considering that the growth of alumino-
silicate crystals from alkaline media was unlikely to pro- A paper published by George Kerr in 1966 describes
ceed by the capture of single monomeric silicate and [55] an experiment carried out to test the hypothesis that
aluminate tetrahedral ions TOn 4 since ‘‘in the elaborate [56] ‘‘a zeolite could be formed via dissolution of gel by
porous crystalline structures of the zeolites, for instance, sodium hydroxide solvent followed by deposition of zeo-
it would seem difficult for the lattice to persist in its very lite crystals from gel-derived species in solution.’’ In the
open pattern when rapidly adding such small units’’ [50]. experiment, a sodium hydroxide solution at 100 C was
He felt that ‘‘a plausible process would be the accretion circulated through two filters, the first of which con-
in simple coordination of polygonal or polyhedral an- tained a specially prepared amorphous sodium alumino-
ions by condensation polymerisation’’, giving as exam- silicate, whilst the second held crystals of zeolite Na-A.
ples the 4-ring, 6-ring, cube and hexagonal prism and When the experiment was terminated after about 4 h,
the formation of the ‘‘crankshaft’’ double chain (found nearly all of the amorphous solid had been dissolved
in feldspars and the phillipsite-harmotome zeolites) by and the zeolite sample (estimated to be essentially
linkage of 4-rings. 100% zeolite A by water sorption) had approximately
Table 1
Summary of principal proposals for zeolite synthesis mechanism, 1959–2004
Author(s) [Ref.] Principal system studied Main features of mechanism Schematic summary
Barrer [49,50] Various low-silica Condensation polymerisation
phases of polygonal and polyhedral anions
Flanigen and Breck [51–54] Na-A, Na-X Linkage of polyhedra (formed by
C.S. Cundy, P.A. Cox / Microporous and Mesoporous Materials 82 (2005) 1–78
M+-assisted arrangement of anions):
crystal growth mainly in the solid phase
fast
Kerr [55,56] Na-A Crystal growth from solution species Amorphous solid ! soluble speciesðSÞ
slow
ðSÞ þ nucleiðor zeolite crystalsÞ ! zeolite A
Zhdanov [57] Na-A, Na-X Solid M liquid solubility equilibrium,

nuclei from condensation reactions, Amorphous solid phase Liquid phase
crystal growth from solution
Accumulation of zeolite crystals Formation of nuclei
Derouane, Detremmerie, Na,TPA-ZSM-5 Synthesis ‘‘A’’: liquid phase ion

Gabelica and transportation. Synthesis ‘‘B’’:
Blom [58–62] solid hydrogel phase transformation
Chang and Bell [63] Na,TPA-Si-ZSM-5 Embryonic clathrate TPA-silicate units,

ordered into nuclei through OH-mediated
SiAOASi cleavage/recombination
Burkett and Davis [64–66] TPA-Si-ZSM-5 Pre-organised inorganic–organic composites,

nucleation through aggregation, crystal
growth layer-by-layer
Leuven Group [67–73] TPA-Si-ZSM-5 Oligomers ! precursor ‘‘trimer’’ (33 Si)

! ·12 ! ‘‘nanoslabs’’, growth by
aggregation
7
doubled in mass. From these (and other [55]) observa- polyhedral building units. Similar precursor units were
tions, the mechanism was perceived to be that of rapid envisaged by Flanigen and Breck, although their think-
dissolution of the amorphous solid to yield soluble spe- ing was focused largely upon the solid phase. In this
cies. The rate-determining step was then the combina- view, the initial, random aluminosilicate gel structure
tion of these soluble species with nuclei or zeolite was dis-assembled into its constituent tetrahedra by
crystals to yield the zeolitic product. the action of OH ions and new, oligomeric polyhedral
Working for the same company (Mobil) but not in units were formed through the ordering influence of the
the same laboratories, Julius Ciric presented in 1968 cations. Crystal growth proceeded by an OH-catalysed
the most detailed study of zeolite synthesis published polymerisation and depolymerisation process, involving
at that date [74]. Kinetic curves were determined from predominantly the solid phase but with some contribu-
water sorption and chemical analyses were carried out tion from solution species. The studies of Zhdanov
on reaction filtrates. In addition, data were obtained and of Kerr provided a more solution-oriented perspec-
by particle counter, optical microscopy and BET surface tive. The original amorphous gel was seen as a dynamic
area methods. The work adds much to the ideas set out entity, in equilibrium (or coming to equilibrium) with
in the Kerr report [55], pointing to a solution-mediated the liquid phase. Dissolving under the action of heat
growth mechanism modified by the presence of the gel and base, the gel released active soluble species into
phase (so that transport of growth species to crystals the solution from which nuclei formed and grew, from
embedded in gel is restricted by diffusion through the solution, into crystals, although the detailed nature of
gel). In addition, Ciric pointed out that his kinetic re- the migrating units was unspecified.
sults were consistent with Barrers ideas of anionic During the later 1970s, the significance of the solution
blocks [49,50] as well as with the Flanigen–Breck view phase in zeolite synthesis was to become increasingly
[51] on the catalysis by OH ions. apparent, as demonstrated by two Raman spectroscopic
studies on the formation of zeolite A. Observing no
3.4. Studies at Leningrad changes with time other than the appearance of crystal-
line product, McNicol et al. concluded that crystallisa-
Some striking advances in thinking and technique tion occurred within the solid phase of the gel [75,76].
were reported by Zhdanov at the Second International However, by using a combination of chemical analyses,
Zeolite Conference in 1970 [57]. Measurements on crys- Raman spectroscopy, XRD, sorption and particle size
tal linear growth rates for zeolite A showed directly for measurements, Angell and Flank [77] reached the oppo-
the first time the effect of temperature in increasing site conclusion. They demonstrated that the mechanism
growth rate and that the crystals grew at a near-constant involved formation and subsequent dissolution of an
rate over the majority of the synthesis period. From this amorphous aluminosilicate intermediate, with solution
latter observation and the product crystal size distribu- transport from the gel to the growth surface of the crys-
tion, Zhdanov was also able to deduce the nucleation tallite. This view was reinforced by two further synthetic
rate profile over the course of the reaction. These con- studies. Culfaz and Sand examined crystallisation rates
siderations, together with measurements of chemical for mordenite, zeolite X and zeolite A [78]. From consid-
changes in the solution phase of the reaction mixture erations of rate limitations by diffusion and seed crystal
and detailed consideration of such phenomena as the surface area, they deduced that crystal growth in these
induction period and seeding effects, led to a more chem- cases occurred from solution. Kacirek and Lechert [79]
ically detailed picture of zeolite crystallisation. In this used detailed kinetic studies on seeded faujasite synthe-
view, the solid and liquid phases are connected by the ses to develop further the solution growth model, con-
solubility equilibrium. Condensation reactions give rise cluding that the rate-determining step was the
to ‘‘primary aluminosilicate blocks (4- and 6-membered connection of silicate species to the surface of the crys-
rings)’’ and crystal nuclei. Crystal growth occurs from tal. They also pointed out that, under their conditions,
solution until dissolution of the amorphous phase is the solution phase would contain essentially only mono-
complete. Analytical data supported the proposition mers and dimers during the crystallisation of zeolite X,
that the composition of the crystals depended on that with higher oligomers (perhaps up to Si20) present in
of the liquid phase from which they crystallised. the synthesis of the more siliceous Y-types.
3.5. Overview—1959 to 1971 and beyond 3.6. Introduction of organic templates
It may be useful at this point to summarise the main In 1961, two groups of workers disclosed the effect of
opinions expressed up to the year 1971. Barrer had con- introducing quaternary ammonium cations into zeolite
cluded that zeolite crystallisation was a solution-medi- synthesis. Barrer and Denny described amine-associated
ated process, the structure being formed by the routes to zeolites A and X [11] whilst Kerr and Koko-
condensation polymerisation of anionic polygonal or tailo published [12,13] data on a tetramethylammonium
(TMA) silica-rich version of zeolite A named ZK-4 (Si/ ‘‘lattice’’ of the water clathrate and surround the hydro-
Al up to 1.7). At first, no new structures resulted from phobic organic guest molecules. In this way, the struc-
this pioneering work, but in 1967 Wadlinger, Kerr and tural chemistry of water below room temperature is
Rosinski reported [14] the discovery of the first high-sil- translated to that of silica near 200 C. This concept
ica zeolite, zeolite beta (5 < Si/Al < 100), made using the was developed and extended in a landmark paper by
tetraethylammonium cation. ZSM-5 followed in 1972 Chang and Bell [63].
[15], the original syntheses being based on a tetrapropyl-
ammonium (TPA)–sodium mixture. All these materials 3.7. Chang and Bell and after
were formed as crystalline products containing the
encapsulated organic cations, leading to the idea of The work of Chang and Bell [63] was based upon
‘‘templated’’ synthesis with the organics acting as struc- studies of the formation of ZSM-5 from Al-free precur-
ture-directing agents (SDAs). In terms of the mechanis- sor gels at 90–95 C using XRD, 29Si MAS NMR spec-
tic alternatives discussed above (Section 3.5), the troscopy and ion exchange. The NMR results suggested
introduction of these organic reactants provided new that major changes in gel structure occur during the
possibilities for probing the chemistry of the synthesis early stages of reaction. This was confirmed by the dem-
reaction. Investigating the synthesis of zeolite A and onstration of ion sieve effects indicating that, in the
other aluminous zeolites, McNicol et al. were able to de- tetrapropylammonium (TPA) system, embryonic struc-
tect the clathration of TMA units by a shift in the tures with Si/TPA = 20–24 are formed rapidly upon
754 cm1 Raman band, supported by results from heating. These first-formed units may resemble ZSM-5
Eu3+ phosphorescence spectroscopy [76]. They found channel intersections (4 per unit cell of 96 tetrahedral
no evidence for cage-like building blocks in either solu- atoms), each containing essentially one TPA+ cation,
tion or solid before the onset of crystallisation. and thus provide a possible mechanism for ZSM-5
In time, the attractive idea of a strong ‘‘lock and key’’ nucleation. In this scheme, the hydrophobic effect and
relationship between framework and template would the isomorphism between water and silicate structure
come to dominate much of the thinking on synthesis lead to (i) formation of a clathrate-like water structure
mechanism and two reviews published in the early around the template, and then (ii) conversion of the
1980s focused attention on this area of host-guest sci- clathrate-like hydrate to a clathrate-like silicate by iso-
ence [80,81]. However, a scheme [58] introduced by Der- morphous substitution of silicate for water in the embry-
ouane and co-workers at about the same time was aimed onic units. Such units are initially randomly connected
mainly at explaining experimental observations and con- but in time become ordered (‘‘annealed’’) through re-
centrated on the inorganic gel chemistry [58–62]. Based peated cleavage and recombination of siloxane bonds,
on investigations using a wide variety of techniques, mediated by hydroxide ion. Thus, nucleation occurs
they proposed two pathways for ZSM-5 formation. through progressive ordering of these entities into the fi-
The use of Al-rich ingredients and polymeric silica was nal crystal structure. This dynamic-assembly argument
pictured as generating a small number of nuclei which is very reminiscent of that originally put forward by
grew by a liquid phase ion transportation process to Flanigen and Breck for an inorganic system [31,51] (Sec-
yield large ZSM-5 single crystals (synthesis A). This as- tion 3.2).
pect of the suggested reaction mechanism therefore The principal concepts advanced by Chang and Bell
bears many resemblances to the solution-mediated [63] have been extended in a series of papers in which
scheme of Zhdanov (Section 3.4). For syntheses of type Burkett and Davis [64–66] examine the role of TPA as
B (typified by high Si/Al ratios and the use of ‘‘mono- structure-directing agent in silicalite synthesis, primar-
meric’’ Na silicate), the results were interpreted in terms ily by MAS NMR spectroscopy. 1HA29Si CP MAS
of numerous nuclei which rapidly yielded very small NMR results provide direct evidence for the existence
ZSM-5 microcrystallites directly within the hydrogel of pre-organised inorganic–organic composite struc-
in a process described as a solid hydrogel phase tures in which the TPA molecules take up a conforma-
transformation. tion similar to that adopted in the zeolite product. The
The first suggestion as to how the presence of an or- initial formation of the inorganic–organic composite is
ganic template molecule might modify the physical initiated by overlap of the hydrophobic hydration
chemistry of the synthesis medium was put forward by spheres of the inorganic and organic components. Sub-
Flanigen and co-workers [82,83]. It was proposed that sequent release of ordered water enables favourable
the crystallisation mechanism of siliceous zeotypes in- van der Waals interactions to be established. Nucle-
volves clathration of the hydrophobic organic cation ation is then brought about through aggregation of
in a manner analogous to the formation of crystalline these composite species. Crystal growth occurs through
water clathrates of alkylammonium salts. Thus, under diffusion of the same species to the surface of the grow-
synthesis conditions, the silica tetrahedra assemble into ing crystallites to give a layer-by-layer growth
a framework in place of the hydrogen-bonded water mechanism.
Broadly similar ideas have also been developed in For example, predictions of crystal size and size distri-
what has become a very extensive study by a team at bution can be developed to reflect changes in nucleation
Leuven. Using a wide variety of experimental tech- and growth behaviour brought about by gel ageing (Sec-
niques, the work has concentrated on a detailed charac- tion 11.2).
terisation of the MFI precursor material originally The only significant themodynamics-based model of
described by Schoeman [84,85]. The first papers in the zeolite synthesis to have emerged is the equilibrium
series [67–71] identified constituent ‘‘nanoslabs’’ having model of Lowe [92]. This was initially developed to pro-
dimensions 1.3 · 4.0 · 4.0 nm with nine intersections vide insight into the pH changes which occur in the
per particle, each of these containing a TPA cation. course of high-silica zeolite syntheses [93]. The model
Aggregation of such nanoslabs leads to larger particles considers the zeolite synthesis process as a series of
measuring up to 15.6 · 8 · 8 nm and ultimately to the pseudo-equilibria:
crystalline colloidal MFI-type material which forms amorphous solid $ solution species $ crystalline zeolite
the final product of the synthesis. More recently
— progress of reaction !
[72,73], specific silicate oligomers (particularly a penta-
cyclic dodecamer) were identified as intermediates in At the start, amorphous solid is in equilibrium with
nanoslab evolution. However, the elaborately detailed solution species. This initial equilibrium is then main-
interpretations adopted in these studies are the subject tained while product crystals grow from the supersatu-
of increasing criticism [86–88]. rated solution. Finally, when all the amorphous
Some of the principal ideas from the period described precursor has been consumed, the crystalline zeolite
in Sections 3.1–3.7 are summarised in Table 1. equilibrates with its mother liquor. Lowes original
sketch of this process is reproduced in Fig. 2.
This simple analysis enables the solution chemistry,
4. Modelling the processes of zeolite synthesis and in particular the effects of solubility and pH, to be
understood at a fundamental level [92]. Computer mod-
As computing capability has mushroomed, modelling elling of the pH function provides a good simulation of
methods have become an increasingly important adjunct the types of pH curve observed experimentally [94]. The
to experimental studies. It is convenient to consider most notable feature is the sharp rise in pH which occurs
‘‘reaction models’’ and ‘‘molecular models’’ under sepa- when all of the solid gel phase has been consumed and
rate headings, although it is hoped that in due course the control of the solubility is transferred to the crystalline
two branches of the subject will grow together. product. The difference between initial and final pH val-
ues is directly related to the difference in solubility be-
4.1. Mathematical models of synthesis reactions tween zeolite product and gel precursor, providing a
measure of the strength of the templating effect for a ser-
Reaction models as used by chemists and engineers ies of organic additives [95]. The most effective template
are of two basic types: (i) those using a kinetic approach gives the most stable (least soluble) product and hence
and (ii) those founded on a thermodynamic approach. the largest pH rise. Perhaps the key relationship is the
Those in the first category [89] range from simple empir-
ical correlations to complex computer programs. Of the
more complex treatments, the most important are those
based on particle numbers, such as the population bal-
ance model [90,91] developed extensively by Thompson
and co-workers and built upon the basic equation (for a
well-mixed reactor)
on on n
þQ ¼ ;
ot oL s
where n is a number density function (characterising the
crystal size distribution at any time), t is time, L is crys-
tal length, Q is the crystal linear growth rate and s is res-
idence time. Further relationships set boundary
conditions and the material balance. Solutions for the
resulting cohort of equations can be developed to pro-
Fig. 2. Conceptual basis for the Lowe equilibrium model [92] of the
vide simulations covering a wide variety of conditions.
zeolite synthesis process (B.M. Lowes original sketch). Control of
In this way, hypothetical reactions can readily be ex- solubility passes from the initial equilibrium between amorphous solid
plored to assess the effect of changing reaction variables and solution species to the final equilibration between the crystalline
and introducing other components such as seed crystals. zeolite product and its mother liquor.
ratio of precursor and product solubility constants (Ks),

since this is directly related to the solution supersatura-
tion (Section 14) and hence the driving force for the
crystallisation reaction (DG):

K s;gel
DG ¼ RT ln :
K s;zeolite
4.2. Molecular modelling
In terms of zeolite synthesis, molecular modelling

methods have provided insight into (i) the determination
of the location and energies of the templating agents oc-
cluded within zeolite structures during synthesis and (ii)
the detailed investigation of small framework fragments.
4.2.1. Modelling zeolite-template pairs

The key advantage modelling methods offer over
experimental techniques is that they permit investigation
of the energetics of template–framework interactions.
However, experimental studies, most notably X-ray dif-
fraction, have yielded crucial data on template location,
enabling modelling methods to be validated. Modelling
studies of zeolite-template interactions have usually fo-
cused on the use of molecular mechanics calculations.
This basic methodology has been employed in several
different ways to investigate the relationship between Fig. 3. The energy minimised location for dibenzyldimethylammo-
template molecules and the zeolite product. Moini nium ions shown relative to the EU-1 channel system.
et al. [96] optimised the geometry of the template mole-
cules using the molecular mechanics approach prior to
docking them ‘‘by eye’’ into the framework structure. ergy in all its products, and that levels of void filling
In this way, they demonstrated the excellent void filling are also similar in each case.
properties exhibited by EU-1 (EUO) templates, as Zones et al. have made effective use of modelling
shown here in an alternative representation (Fig. 3). methods to aid their search for bulky molecules that
This type of technique was also successfully employed are too large, or have the wrong geometries, to synthe-
by Schmitt and Kennedy to derive new templates for sise common default products. This has resulted in the
ZSM-18 (MEI) based on their geometric match with discovery of several new structures [100]. An automated
the framework [97]. ‘‘De Novo’’ approach for tailored template design has
The molecular mechanics approach has been en- been developed by Lewis and co-workers [101]. In this
hanced by the addition of a Monte Carlo algorithm to method, a template molecule is ‘‘grown’’ computation-
dock the guest molecule inside the framework prior to ally inside the zeolite host in order to maximise the
the application of an energy minimisation or simulated non-bonded interactions between template and the sur-
annealing routine. The application of these techniques rounding lattice. This has resulted in the successful de-
has shown convincingly that the relative non-bonded en- sign of a new template for DAF-5 (a CoAPO material)
ergy between the template and framework can be used [102].
as a measure of the efficacy with which a selected tem-
plate can form a particular product. Shen and co-work- 4.2.2. Cluster calculations
ers [98] have demonstrated the importance of including The properties of small silica fragments critical to the
the Gibbs Free Energy in order to fully rationalise the synthesis process are difficult to study via experimental
synthesis of ZSM-11 by TBA ions. The addition of a methods. In this respect, cluster calculations, usually
simulated annealing protocol to the Monte Carlo rou- based on quantum mechanical approaches, have proved
tine is particularly useful when several templates are highly valuable as a tool to probe the detailed structure,
optimised within the simulation box. Stevens et al. [99] geometry and reactions for a wide range of fundamental
successfully used this method to show that a template silica fragments likely to be of importance in the synthe-
which can be used to synthesise different products under sis process (see for example Refs. [103,104]). Pereira
different synthesis conditions has a similar binding en- et al. [105,106] have developed these procedures to
examine the interaction between solvated zeolite frag- individual reactants; after it, the reaction mixture con-
ments and template molecules. The results from this tains a myriad of small zeolite crystals, already formed
study demonstrate that the organic molecule plays a and in the course of growing larger. However, the
key role in stabilizing clusters of framework material, growth process once underway displays no discontinu-
preserving their structure under the influence of water. ity, so that the mechanism of further growth is essen-
Methods for investigating nucleation processes have re- tially that already established in the early stages. This
cently been extended by the work of Wu and Deem point is reinforced by the analysis of the induction per-
[107]. A Monte Carlo procedure has been developed in iod made by Subotić and co-workers [109], where it is
order to model a silicate solution on the atomic scale shown that the activation energy determined from s is
in the absence of a template molecule. This study has essentially that of the entire crystallisation process. A
yielded an insight into some of the fundamental pro- further modelling study demonstrates the importance
cesses associated with nucleation such as estimation of of including the lag time (tr) in analyses of zeolite crys-
the nucleation barrier and the critical cluster size. tallisation and confirms the importance of gel/solution
rearrangements to this element of the induction period
[110]. It is also shown that size-dependent crystal solu-
Part II: Synthesis mechanism bility (the Kelvin effect) is not a significant contributor
to the crystal growth-rate function (see also Sections
In Sections 5–8, suggestions for the most probable 8.5 and 16.4).
mechanistic pathways in zeolite formation are described In Section 6, we consider further the significance of tr
in sequence: induction period, nucleation, crystal and the nature of the equilibration processes which
growth. In order to balance coherence with conciseness, transform the amorphous phase before zeolite product
some of the background material necessary to support appears. These prove to be of prime importance in set-
the main arguments is noted only briefly. Fuller acting the stage for the unique event of nucleation (Section
counts of such subjects as the nature of growth species 7). It will then be demonstrated (Section 8) that such
and the role of aggregation processes are given later, processes also provide the link connecting the chemistry
in the ‘‘key topics’’ part of this review (Sections 9–18). of nucleation with that of crystal growth.
5. The induction period 6. The evolution of order
The induction period is the time (t) between the no- 6.1. The nature of the amorphous material
tional start of the reaction and the point at which crys-
talline product is first observed. It will therefore depend It is often convenient to treat the amorphous phase as
on the moment chosen for setting t = 0 (often taken as a constant quantity which, apart from depletion, re-
the time at which the reactants reach the working tem- mains essentially unchanged throughout the synthesis.
perature) and upon the method of analysis used to de- Such an approach is usually adopted for the purposes
tect the product (most usually X-ray diffraction). For of reaction modelling [89–92] (Section 4). However,
precipitation reactions, classical nucleation theory the dynamic nature of this material was recognised in
[108] divides this period (s) into a number of subunits: the earliest mechanistic studies (Section 3). Flanigen
and Breck envisaged [51,52] a transformation through
s ¼ tr þ tn þ tg ;
polymerisation and depolymerisation catalysed by ex-
tr is referred to as the relaxation time and is said to be cess hydroxyl ion, whilst Zhdanov [57] and Kerr [55]
the time required ‘‘for the system to achieve a quasi- saw the initial amorphous gel as coming to equilibrium
steady-state distribution of molecular clusters’’. In zeo- with the liquid phase and releasing active soluble species
lite terms, this can be equated with the equilibration into the solution, thus changing with time. As will be
reactions taking place on mixing the reagents and allow- shown, this transformation in the nature of the amor-
ing them to reach reaction temperature, during which phous phase during the early stages of the reaction is
period the observable distribution of silicate and alumi- very significant.
nate ions (and other species) is established (see Section
6.2). The subunits tn (the time for the formation of a sta- 6.2. Primary and secondary amorphous phases
ble nucleus) and tg (the time for the nucleus to grow to a
detectable size) are directly translatable into zeolite At the point where the synthesis reactants are initially
chemistry. mixed together, a visible gel is frequently formed. This
It will be apparent from the above that the induction will be referred to as the primary amorphous phase. In
period encompasses all the significant events of zeolite some cases (‘‘clear solution’’ syntheses), this primary
formation. Before this time, there existed only the phase is colloidal, and thus invisible to the naked eye,
but its function and behaviour are essentially the same SiO2 source Al2O3 source
(see Section 7.6). The primary amorphous phase
represents the initial and immediate product from the
reactants and is a non-equilibrium and probably hetero- M+ OH – Q+ OH–
geneous product containing (for example) (a) precipi-
tated amorphous aluminosilicates, (b) coagulated silica partly-reacted, heterogeneous
non-equilibrium mixture:
and alumina precipitated from starting materials desta-
bilised by the change in pH and increase in salt content solution + solid
↑
and (c) unchanged reactants. The pH of such a mixture
is not usually a useful characteristic measurement, since PRIMARY AMORPHOUS PHASE
it depends on particular circumstances and will change
with age. time temperature
After some time, either on standing, or—more rap-
idly—on heating at reaction temperature, the above
– H2O
mixture undergoes changes due to the equilibration
reactions which occur (Section 7.8) and is converted Si(OH)4 SixOy(OH)z
EQUILIBRATED
into a pseudo-steady-state intermediate, the secondary OH– SOLUTION OH–
amorphous phase. Concurrently, the relationship be-
tween the solid and solution phases approaches an – H2O
equilibrium and a characteristic distribution of silicate ionised monomers ionised polymers
and aluminosilicate anions is established (Fig. 4). A
pH measurement will now provide a useful reference
point from which, in high-silica zeolite synthesis, the Al(OH)4–
progress of the reaction can be monitored by recording
subsequent changes [92,93]. In the final stage of the colloidal and gel colloidal and gel
synthesis reaction (usually at elevated temperature for silicates
↑ aluminosilicates
↑
a prolonged period), the secondary amorphous phase
is converted into the crystalline zeolite product (Fig. SECONDARY AMORPHOUS PHASE
5).
The concept of an equilibrated intermediate phase Fig. 4. Equilibration of the starting mixture to establish a partly
ordered intermediate (secondary amorphous phase) and a character-
is clearly expressed in Zhdanovs representation of
istic distribution of solution species.
the synthesis process [57] (Section 3.4) and is also im-
plied in the type-A and type-B schemes of Derouane
et al. [58] (Section 3.6—see also Table 1). In order Solution
to show that such equilibration has occurred, the bal-
ance of solution species and the partition of compo-
nents between the solid phase and solution can be
Reactants
nucleation
probed by a variety of analytical and spectroscopic Initial Equilibrated
techniques (e.g. Refs. [61,111–116]). The structuring gel gel growth
action of cations is apparent from their role in the
organisation of the solid phase [117], as described Crystalline
random increasingly ordered product
below. In some cases, there are also strong interactions
with solution species—as demonstrated, for example,
by the correlation between the occurrence of the cubic
octamer [Si8O20]8 and the presence of the TMA cat-
ion [118].
Several groups of authors have commented upon the (a) (b) (c)
existence of the secondary amorphous phase. Angell and
Fig. 5. The evolution of order, from the primary amorphous phase (a)
Flank [77] report (for zeolite A synthesis) that ‘‘. . . the through the secondary amorphous phase (b) to the crystalline product
initially formed sodium aluminosilicate gel is converted (c).
via solution transport to an apparently amorphous alu-
minosilicate intermediate. It is this latter material which
is converted to crystalline zeolite via dissolution by the precipitation of a secondary gel having a similar Si/Al
basic medium.’’ Fahlke et al. in 1987 observed (for zeo- ratio to that of the zeolite product [119]. However, a
lite Y) an immediate precipitation of a silica-rich pri- study by the Montpellier group in 1993 [117] is particu-
mary gel, followed by its slow dissolution and then the larly informative and merits a more detailed description.
6.2.1. The Montpellier study the preparation of aluminous (inorganic) zeolites

This investigation [117] is the clearest demonstration [77,119].
yet published of the multi-stage nature of the zeolite syn-
thesis reaction as outlined in Fig. 5. In this work, Nicolle 6.2.2. Related investigations
et al. prepared a series of Na, TEA-aluminosilicate gels There appears at first sight to be a discrepancy be-
of varying composition (TEA = tetraethylammonium) tween the observations of the Montpellier group [117]
and heated samples of these to 150 C (i) for a relatively and those described in a detailed investigation of zeolite
short period and (ii) for a longer time. The solid prod- beta synthesis by other workers. On the basis of thermal
ucts from each reaction period were isolated and ana- analysis and charge-balance data, Perez-Pariente et al.
lysed. The actual heating times varied with the [120] concluded that almost no TEA was associated with
reactivity of the particular composition but were aimed the amorphous solid. However, in later work, Camblor
at achieving in the first instance (i) an equilibrated gel, and Perez-Pariente reported the presence of TEA cat-
and in the second case (ii), a crystalline zeolite product. ions associated with the aluminate sites in pre-crystalline
The four compositions chosen (A–D) crystallised in the samples [121], concluding nevertheless that the overall
second stage to (A) zeolite beta, (B) mordenite, (C) crystallisation pattern (by a solution transport mecha-
ZSM-12 and (D) ZSM-5. However, the most significant nism) was the same in both cases. The most obvious dif-
findings came from the analysis of the initial, equili- ferences between the two studies were the use of
brated products. tetraethyl silicate and a low crystallisation temperature
X-ray diffraction revealed only broad features indica- (100 C) in the earlier work, whereas the later investiga-
tive of local ordering but no crystal lines. All samples tion employed amorphous silica (Aerosil) and a reaction
showed aluminium in tetrahedral coordination only temperature of 135 C. However, a more significant dis-
(27Al MAS NMR) and no zeolitic bands in their infrared parity lies probably not with these variables but in the
spectra. Chemical analysis demonstrated that the hydro- cation balance. The 1991 study centred around synthesis
thermal treatment had greatly modified the composition compositions which are fairly typical of routine zeolite
of the solid phase (i.e. compared to that before heating), beta preparations, namely
with TEA+ replacing Na+ as the preferred cation. The x½yK; ð1 yÞNa
2 O 12:5ðTEAÞ2 O Al2 O3 50SiO2 750H2 O
TEA content was nearly constant, corresponding to
one organic cation per 16–20 T-atoms, whatever the where x was varied from 0 to 4.5 and y from 0 to 1, with
method of preparation or the Si/Al ratio. Other, specific OH/SiO2 constant at 0.56. The range of compositions
tests were carried out on selected samples. Equilibrated used for the 1987 study was generally similar—with
samples A and B had (after calcination) cation exchange the exception of the gel used for the analysis of the
capacities in close agreement with their aluminium con- amorphous intermediate (B2 at 12 h), which was
tents (85 mmol NHþ 4 /100 g). For equilibrated samples 7:8Na2 O 0:11K2 O 1:5ðTEAÞ2 O Al2 O3 30SiO2 360H2 O
(A)–(C), calcination generated a micropore system hav-
ing about 60% of the capacity shown by the zeolite beta The very high M/TEA ratio will have a major effect
crystallised from composition A. Amorphous sample B upon the partition of cations between the liquid and
showed the same micropore volume for cyclohexane as gel phases, so that the low TEA content of the isolable
for nitrogen but was impermeable to trimethylbenzene. solid phase is perhaps not unexpected.
The authors concluded that the amorphous, equili-
brated samples differed from the gels initially precipi-
tated at room temperature and had been formed by 6.3. Further evidence for pre-crystalline order from
dissolution and reprecipitation. These secondary prod- synthesis studies
ucts were silicoaluminates sharing several properties
with high-silica zeolites. In both cases, clusters of silica Tsuruta et al. studied the initial product in the syn-
or alumina tetrahedra had been organised (templated) thesis of zeolite A from concentrated solutions in the
around a bulky molecule, whose decomposition left a presence of an anionic surfactant [122]. Using X-ray
micropore system. The essential difference between the and electron diffraction, they found that the amorphous
two classes of solids was represented by the dispersion aluminosilicate isolated under mild reaction conditions
of the cluster geometry. Zeolites featured well-defined (60 C, 15 min) possessed a short-range order of Si
and repeatable site geometries whereas the equilibrated, and Al atoms similar to that in crystalline zeolite A.
amorphous products presented irregular and hence ape- Electron diffraction, thermal analysis and FTIR spec-
riodical, local organisation. From data discussed else- troscopy were used by Subotić and co-workers in a
where (Sections 6.3 and 12) it is clear that these are study of the structural properties of X-ray amorphous
general observations and not, for example, associated sodium and potassium aluminosilicate gels [123,124].
only with the synthesis of zeolites templated by quater- The gels were found to contain structurally ordered re-
nary ammonium ions. Similar phenomena occur during gions, or particles of a partly crystalline phase inside
an amorphous matrix. Subsequent hydrothermal treat- (primary amorphous phase). However, over a period of
ment in 2 M NaOH at 80 C yielded zeolites A and X. time (tr), and especially on heating, silicate and alumino-
Walton and OHare [125] examined amorphous gallium silicate equilibration reactions will occur, leading to a
silicates isolated in the early stages of the hydrothermal re-distribution of species and a repartition of reaction
synthesis of Ga-hydroxysodalite by means of EXAFS components between the solid and liquid phases as they
and XANES spectroscopy. Both Ga and Si K-edge EX- approach equilibrium with one another. Cations play a
AFS indicated some degree of medium-range order (i.e. structuring role in the organisation of the solid phase,
beyond the first coordination shell), indicative of an ex- which is generated as a new or reconstructed material
tended network of alternating Ga(OSi)4 and Si(OGa)4 having a similar chemical composition to that of the
units resembling the structure of the crystalline product. eventual zeolite product but lacking in long-range, peri-
A related multi-nuclear MAS NMR and neutron diffrac- odical organisation and hence amorphous to X-rays (sec-
tion study of amorphous zeolite A precursors by Yang ondary amorphous phase). However, elements of local
et al. [126] detected changes in medium range order prior order will be present and will impart to the amorphous
to crystallisation as Al (accompanied by Na+ cations) intermediate some of the chemical and physical proper-
was incorporated into the silicate network. However, ties associated with the final, crystalline, ‘‘isomeric’’ zeo-
both spectroscopic and diffraction techniques indicated lite. This is shown diagrammatically in Fig. 5 and is
that there were no well-defined structural units (e.g. discussed further when the topic of ‘‘X-ray amorphous
SBUs) present before the formation of the zeolite crys- zeolites’’ is considered (Section 12). In Section 7, this
tals. Rather equivocal results were obtained from a semi-organised solid phase is shown to be pivotal in
129
Xe NMR study of samples taken during the course the transforming step of zeolite nucleation.
of various zeolite syntheses [127]. Evidence of gel struc- In the above discussion, a firm distinction is made be-
turing (large cavities) was found for Na,TPA-ZSM-5 tween primary and secondary amorphous phases in order
and Na-Y but not for Na,TEA-ZSM-20. However, only to emphasise the importance of the changes taking place.
in the latter case did the NMR measurement show any However, the situation in some zeolite syntheses may be
appreciable microcrystallite formation in the absence less clear cut, since the different stages in the process may
of XRD crystallinity. overlap, e.g. nucleation may begin whilst the bulk of the
There have been several other investigations of the pre- reaction mixture is still far from attaining a steady state
crystalline state of MFI-type synthesis compositions. In (this is more likely to occur in high temperature synthe-
the work of Burkett and Davis, 1HA29Si CP MAS ses). This ‘‘relative kinetics’’ effect is an important factor
NMR spectroscopy demonstrated that the TPA mole- in the dependence of the outcome of a synthesis reaction
cules and silicate species were in close proximity in the upon the nature and mode of mixing of the reactants.
heated gel before the characteristic XRD and IR finger- Certainly, in cases where no intermediate samples are
prints of crystalline silicalite became apparent [64]. No taken for physico-chemical characterisation, the re-
such correspondence was found in the unheated gel. This searcher may be unaware that the progression from reac-
result was supported by an in-situ multinuclear NMR tants to product is anything other than a continuum.
study at 80 C [128]. Hydrogen bonds between the organic The extent to which the secondary amorphous phase
and water-clathrated molecules are progressively re- can be identified or isolated will also depend upon the
placed by hydrophobic interactions between the organic nature of the synthesis mixture. In some cases, partition
and silica species. Similar evidence of TPA occlusion of the reactants between the liquid and solid phases may
comes from SAXS [129] and SANS [130,131] measure- be such that little or no material may be identifiable by
ments. Several groups of workers have reported FTIR conventional methods of solid–liquid separation, so that
bands at 550–560 cm1 in isolated material which is X- colloid chemical techniques may be necessary to observe
ray amorphous [84,130–134], these being interpreted as it (see Sections 7.6 and 7.7).
indicative of the presence of material having the MFI Finally, it should be noted that in Fig. 5 gross heteroge-
structure. However, some caution is necessary in this neity (e.g. the presence of some unchanged starting com-
analysis since such an absorption is not unique to ZSM- ponents) in the primary amorphous phase has been
5. It is advisable to compare the bands which should ap- ignored for the sake of simplicity and the material illus-
pear at both 550 and 450 cm1: for a well-crystallised trated as a single homogeneous phase of random structure.
ZSM-5, the 550:450 intensity ratio should be 0.7, a fig-
ure which decreases with decreasing crystallinity [135].
7. Nucleation
6.4. Summary
7.1. Introduction
When first prepared, the reaction mixture for a zeolite
synthesis consists of a non-equilibrium combination of From the above (Section 6), it is clear that the equil-
components—heterogeneous and at best partly reacted ibration which allows the conversion of primary to
secondary amorphous phases (Fig. 5 and Section 7.8)

forms an extremely important part of the zeolite synthe- ∆GS
sis process. By this means, the initial gel reacts with + ve
solution species to establish (or approach) a pseudo-
Free energy (∆G)

steady-state distribution of liquid phase species, pre-
4
dominantly ionic but with a wide range of charge/mass ∆Gc = πσrc2
3
ratios (Fig. 4 and e.g. Refs. [111–116]). The changes in 0 rc r
the amorphous phase involve an increase in structural
ordering but without the establishment of the periodic zeo-
lite lattice itself. For this, a discreet nucleation event has - ve
to occur. In this step, a statistical selection of the recon- ∆GV ∆G
structed areas reach a critical nuclear size and degree of
order such that a periodic structure is able to propagate,
i.e. crystal growth can begin. Fig. 6. The energetics of nucleation, illustrating the concept of a
In the following treatment, nucleation is first consid- critical nucleus of radius rc; beyond this size, the net energy gain from
ered from a classical standpoint (Section 7.2), after the resultant (DG) of cohesive (DGv) and surface (DGs) terms is
favourable to growth (after Mullin [108]).
which experimental methods for the determination of
nucleation patterns are described (Section 7.3). Next,
brief reviews are given on the topics of seeding (Section
7.4) and autocatalytic nucleation (Section 7.5). Finally, It seems reasonable to suppose that the construction
the special features of zeolites are examined, leading to of the nucleus of a zeolite crystal may involve a more
a consideration of the underlying mechanism (Sections complex assembly process than is necessary for simpler
7.6–7.8). substances whose unit cells are smaller and contain far
fewer atoms, although there should be no difference in
7.2. General considerations basic principles. One possible special feature in the for-
mation of microporous crystals has however been iden-
Amongst the components (tr, tn, tg) comprising the tified by Pope, who has pointed out that the energetics
induction period (s) (Section 5), tn is the time taken to of zeolite nucleation may be greatly modified from that
form a stable nucleus. Fig. 6 illustrates the classical con- of dense phases because of the presence of the large
cept of a critical nuclear size [108]. Beyond this point internal surface area [136].
(where nuclear radius (r) = rc), sufficient matter has The nomenclature for the most usually recognised
come together in an ordered way for the cohesive energy types of nucleation can be summarised as follows
(DGv) to outweigh the energy expended in creating a sur- [108]: primary nucleation may be homogeneous (from
face dividing the nucleus from the continuum (DGs). solution) or heterogeneous (induced by foreign parti-
This enables a stable unit to be formed: a viable nucleus, cles); secondary nucleation (induced by crystals) may
capable of growth. For zeolite systems, a critical size of be considered as a special case of heterogeneous nucle-
1–8 unit cells has been estimated by Thompson and ation in which the nucleating particles are crystals of
Dyer [90]. However, the chemical and physical processes the same phase. Good general accounts of nucleation
leading up to critical nucleus formation are governed by in zeolite systems (along with further references) will
the structure being formed and by the experimental con- be found in reviews by Barrer [137] and Thompson
ditions. If these conditions change, then rc will also vary, [37,91].
as can be seen from the general form of the expression
for r below [108]:
7.3. Determination of zeolite nucleation patterns from
2rv measurements on the resulting crystals
r¼ ;
kT ln S
Obtaining direct experimental data on nucleation
where r is the interfacial tension (surface energy per processes is very difficult due to the extremely small frac-
unit area), v is the molecular volume and S is the super- tion of the total mass involved and the problems of dis-
saturation ratio. A notional nucleation rate (J) can be tinguishing this from surrounding reactants of very
defined as the rate of production of units having radius similar nature and composition. Consequently, much
r P rc. A simple measure of J is often taken by assum- of the information available is at best that of early
ing J / 1/s [31,78]. However, since s is a compound growth behaviour rather than of nucleation itself. Per-
term, the danger in this procedure is apparent, unless haps only the recent cryo-TEM [129] and HRTEM stud-
tr and tg are known to be small compared with tn ies [134,138,139] (Section 7.7) genuinely fall outside this
[109,110]. category and can be considered as direct evidence of
nucleation phenomena. However, much useful data can

be obtained from ‘‘circumstantial evidence’’, i.e. mea-
surements of the appearance, growth and size distribu-
tion of the crystals resulting from the earlier, hidden
patterns of nucleation.
7.3.1. Studies under isothermal conditions

It is still fairly common practice to estimate nucle-
ation rates in zeolite synthesis from the reciprocal of
the induction times [78], despite the acknowledged pit-
falls (Section 7.2) [31]. This is largely because of the dif-
ficulty of making more appropriate and accurate
measurements, particularly if temperature variation is
to be included. An improvement on this can be found
in the procedure of Zhdanov and Samulevich, which en-
ables the calculation of isothermal nucleation rate pro-
files from determinations of growth rate and crystal
size distribution [140,141]. Originally implemented in
analyses of zeolite Na-A [57] and Na-X [140] crystallisa-
tion, the method has subsequently been applied to other
zeolite systems, including silicalite [142,143]. If it is sup-
posed that all the crystals in a batch have the same
(known) growth rate behaviour, the total growth time
of each crystal can be calculated. Assuming also that
the nucleation point for each crystal can be obtained
by linear extrapolation to zero time, the nucleation pro-
file for the whole batch can be determined, as illustrated
in Fig. 7. (Although Zhdanov and Samulevich were the
first to apply the above type of analytical procedure to
zeolite synthesis, Giaya and Thompson, whilst research-
ing their own numerical technique for determining the
crystal size distribution function [144], discovered that
the approach had been discussed in a wider context
much earlier by Bransom and Dunning [145].)
7.3.2. Ageing studies

Although the types of investigation mentioned above
yield useful data, far more detailed information on
nucleation behaviour can be obtained from ageing stud-
ies. In these cases, a reaction mixture is allowed to age at
one temperature and is then crystallised at a second
(usually higher) temperature. By correlating changes
(principally of size) in the product crystals with the
length of the ageing period, aspects of the nucleation
behaviour can be deduced. In particular, the evolution
of order (Section 6) in the zeolite synthesis mixture Fig. 7. Determination of nucleation rate profiles by the method of
Zhdanov and Samulevich [140] as applied to silicalite synthesis [142].
(equivalent to the concept of accumulation of proto- From final product crystal size distribution (a), crystal linear growth
nuclei) can be determined quantitatively. A detailed ac- rate (b) allows calculation of nucleation rate as a function of reaction
count of ageing processes is given in Section 11. time (c). Data consistency is shown by comparison of calculated
mass% growth rate ((d), curve) with experimental values ((d), marked
points). Note that the length axis in (a) becomes a time axis in (b)–(d).
7.4. The use of seed crystals
It is a common and very useful practice to add seed two main effects anticipated are (i) a reduction in the
crystals to zeolite reaction mixtures and excellent ac- synthesis time and (ii) a ‘‘direction’’ of the synthesis
counts of this technique are available [37,91,146]. The towards a desired phase with consequent reduction in
impurities. It is also possible to exert control over prod- (or, more correctly, surface area) of added seed material
uct crystal size distribution [147–150]. The bases for is reduced, then the natural supersaturation and conse-
these valuable influences are worthy of brief quent self-nucleation will no longer be suppressed and
consideration. the product crystal size will deviate more and more from
The basic action of seed crystals (when added effec- that predicted on the basis of linear growth on seed crys-
tively) is to provide surface area on which the required tals only (Table 2).
product can grow. This removes the necessity for such Secondary nucleation certainly occurs in zeolite sys-
surface to be self-generated by primary nucleation and tems but has been the subject of few detailed studies
thus reduces the induction time (s), since the tn compo- [37,91]. If also assisted by agitation, then some form of
nent of s is eliminated (see Section 7.2). In some cases collision breeding [37,91,108,153] is likely to be opera-
(equivalent to the traditional scratching with a glass tive on a macro- or micro-scale. From a practical point
rod in organic chemistry), almost any intervention of view, secondary nucleation has been observed in syn-
may produce the desired effect. However, in others there theses which for various reasons (sometimes unknown)
is a definite element of surface recognition by which one are very reluctant to self-nucleate even though the com-
phase can be given a kinetic advantage whilst potential positions are believed to be near the optimum. Possibly
competitor products are relatively unresponsive. ‘‘active’’ impurities [37,154] are absent, or nucleation/
If seed crystals are added to a synthesis mixture, they growth poisons are inadvertently present [155,156]. In
may behave in a number of ways, i.e. the crystals may such cases, the use of seed crystals can be very effective
(a) remain inert, (b) dissolve, (c) act as pure seeds, in since the response to secondary nucleation is a geometric
that mass is deposited upon them and they grow, or function of the quantity added. This is demonstrated in
(d) give rise to secondary nuclei and hence a new crop the synthesis of siliceous forms of EUO-type zeolite
of crystals. In general, it is necessary to provide sufficient (EU-1, ZSM-50) [16] (Fig. 8). The reactions [157] were
surface area to achieve an effect, so that it is rare for a run at 150 C for three days using the systems
very small quantity of large crystals to cause any signif- 10Na2O : xAl2O3 : 60SiO2 : Template : 3000H2O, where
icant change in the normal course of the reaction. Such either x = 0, Template = 13 dibenzyldimethylammo-
crystals can usually be found at the end of the synthesis, nium chloride (DMDBACl), or x = 0.05, Template = 10
either intact (i.e. case (a)) or showing signs of attack hexamethonium bromide (HEXBr2). In the absence of
(case (b)). If the added material is unstable in the synthe- seed crystals, reactions gave either very low yields or
sis medium, it will normally dissolve and only influence products heavily contaminated with impurity phases
the course of the synthesis if added in sufficient quantity (mainly EU-2 and cristobalite). With DMDBA tem-
to alter significantly the reaction stoichiometry (limit of plate, crystallisation was greatly assisted by the inclusion
case (b)). of seed crystals (5 mass% based on silica). Large ‘‘par-
The balance between seed growth (c) and secondary ent’’ and small ‘‘daughter’’ crystals are clearly distin-
nucleation (d) depends on the nature of the system, guishable. The greater available surface area provided
the quantity of material added and the degree of agita- by calcined seed makes this a more effective nucleant
tion. If the surface area of seed crystals present is suffi- (Fig. 8b) than the uncalcined material (Fig. 8a), result-
cient to absorb most of the available flux of growth ing in a larger number of smaller product crystals. The
species and thus prevent the effective solution supersat- spawning of daughter crystals from a partly dissolving
uration from reaching high levels, then most of the calcined seed crystal is shown in Fig. 8c. A similar effect
growth in the system will take place on the seed crystals, was found for the HEX preparation of EU-1, where the
whose size and growth rate can be closely controlled and product crystal size from calcined seed (Fig. 8d) was
predicted by kinetic modelling [79,89–91,147,148]. Very again smaller (1 lm) than that from as-made material
small (colloidal) zeolite crystals are, as would be ex- (2 lm, not shown). The micrographs reveal a further
pected, particularly effective [149–152]. As the quantity interesting point. The same batch of Na,DBDMA-
EUO seed was used in all cases but it is clearly the syn-
thesis reaction template which controls the product
morphology.
Table 2 A further aspect of seeding has been demonstrated by
Calculated and observed product crystal sizes in seeded silicalite
Bronić and co-workers. The syntheses of zeolites Y and
synthesis [152]
P [158] and of zeolite A [159] have been investigated
Seeding level (%) Product crystal size (nm)
using a 2-compartment reactor in which seed crystals
Calculated Observed (±1 nm) and reactant gel were separated by a submicron mem-
5 111 98 brane. Crystal growth occurred in both compartments,
10 89 83 in some cases giving different zeolite phases (Y on seed
20 74 70
side, P on gel side). However, it was found that the seed
Seed size 43 nm. crystals modify the crystallisation process only if they
Fig. 8. Seed crystals (large) seen amongst the induced secondary crystal population in siliceous EUO-type zeolite synthesis. Templates were: (a–c)
DMDBA, (d) HEX. Seed crystals were: (a) as-made, (b–d) calcined. Note that the micrographs vary in scale. See text for discussion.
are in physical contact with the gel. Thus it can be Subotić and co-workers [143,160–165]. However, it has
deduced that growth species in the liquid phase travel been pointed out [166] that the nucleation periods pre-
freely through the membrane, whereas nucleation is con- dicted by the model are unrealistically long. The idea
trolled more locally. These studies are thus reminiscent was therefore modified [167] by postulating that the dor-
of the Kerr recirculation experiment [55,56] (Section mant nuclei were located preferentially near the periph-
3.3). ery of the gel particles and therefore became activated
much earlier in the crystal growth/gel dissolution pro-
7.5. Autocatalytic nucleation cess. A subsequent study by Falamaki et al. [168] on
ZSM-5 crystallisation using this modified approach gave
One of the sources of nuclei proposed in the classic excellent agreement between the model and experiment.
paper by Zhdanov [57] was that some germs lie dormant However, there appears to be no chemical justification
in the amorphous phase until activated by release into for the assumption and any self-consistent set of model
the solution through gel dissolution. Since the rate of parameters which generated nuclei as a suitably limited
gel dissolution must increase with the rate of consump- function of synthesis time would presumably produce an
tion of growth species by the increasing cumulative crys- equivalent result. This is demonstrated by the work of
tal surface area, the process was seen as ‘‘autocatalytic’’. Nikolakis et al. [169] who provide an alternative to the
This approach has been elaborated and modelled by hypothesis of nuclei release from the dissolving gel by
treating gel dissolution and nucleation as interfacial phe-

nomena controlled by the gel–solution boundary. Their
model for the gel microstructure gives rise to a maxi-
mum in the temporal surface area function which corre-
sponds to the observed early maximum in the nucleation
rate.
7.6. Nucleation in zeolite systems—The nature of the

reaction sol
In Sections 7.3–7.5, nucleation has been considered

largely from an experimental standpoint. The mechanis-
tic basis for these phenomena will now be discussed. As
a first step, it is necessary to expand upon the consider-
ation of the synthesis reaction mixture introduced in the
original summary of experimental observations (Section
2). In a typical situation (from which there will be vari- Fig. 9. The colloid problem: the liquid phase resulting from conven-
ants), there is likely to be present (i) amorphous mate- tional solid–liquid separation is likely to contain both true solution
species and colloidal amorphous material.
rial, (ii) a solution phase and (iii) one or more
crystalline phases. It is relatively easy to identify compo-
nent (iii), even if the crystals are of colloidal dimensions. liquid phase by filtration or (standard) centrifugation.
The distinction between constituents (i) and (ii) is much The result of this is that separation of solid and liquid
less clear cut and has been the source of much phases in syntheses of this type does not remove amor-
misunderstanding. phous material from the liquid phase (Fig. 9). Indeed,
The simplest case is that in which most of the amor- in most so-called ‘‘clear solution’’ syntheses (see e.g.
phous material is visibly solid in nature, as in traditional Refs. [130,142,175–177]), all the amorphous feedstock
syntheses of aluminous zeolites such as A or X. At the is stored by the system in this way, forming the nutrient
low Si/Al ratio and high pH of these preparations, which provides the driving force for the crystallisation
nearly all the Si,Al-derived nutrients are present as a of the zeolite product. Thus, any consideration of nucle-
solid gel phase, with the balance existing as low molec- ation and crystal growth in hydrothermal zeolite synthe-
ular weight species (monomers, oligomers) in a strongly sis must take into account the role of amorphous
alkaline (Na+OH) solution phase [170]. Separation of components at all levels of subdivision.
solid and liquid phases by normal laboratory methods
(filtration, centrifugation) will achieve a reasonably 7.7. Nucleation in zeolite systems—Homogeneous or
effective division of the reaction mixture into a liquid heterogeneous?
phase containing (largely) true solution species
[111,171] and a solid phase from which the proportion In zeolite synthesis, there is evidence for the partici-
of crystalline to amorphous material can be determined pation of both primary and secondary (crystal-cataly-
by XRD measurement. sed) nucleation, although the former is believed to be
At higher Si/Al ratios and lower pH, the situation be- the most prevalent [37,91]. Even so, there is often dis-
comes complicated by an increasing proportion of high agreement on whether the predominant mechanism is
molecular weight silicate and aluminosilicate compo- homogeneous or heterogeneous. For example, a recent
nents. The distinction between solution-related poly- study combining both theoretical and experimental
mers and colloidal material is a problem which lies at work on the nucleation of zeolite A concluded that
the heart of the understanding of silicate chemistry nucleation was homogeneous except in the case of seed-
[171–174]. However, the sequence of polymeric species ing with certain high surface-area titanias [178]. Related
Si1, Si2, Si3 ,. . ., Sin cannot be regarded as a continuum, comparisons of calculated and observed behaviour for
since at some point the larger components behave as if zeolites A and Y by Bronić and Subotić found that the
they are particulate and phase-separated, resembling contribution of homogeneous nucleation was negligible
an invisible precipitate [171–173]. Thus, whereas one [179]. However, it can be seen from the above discussion
part of the silicate and aluminosilicate loading of the of the nature of the liquid phase (Section 7.6) that there
liquid phase behaves as true solution species, a second, is a more subtle ambiguity within the primary classifica-
colloidal fraction resembles a separate amorphous tion as to whether an apparently homogeneous nucle-
phase. This latter is essentially equivalent, in terms of ation event may be genuinely so—or in reality
energetics, to the solid, visible, amorphous gel but is heterogeneous—depending on whether or not the col-
invisible to optical detection and inseparable from the loid phase is involved.
A study on (initially) clear solution synthesis of alu- such systems seems always to be associated with the
minous zeolites by Aiello et al. [180] showed that heter- gel phase, so that the product crystals, when observed
ogeneous nucleation appeared to occur on gel lamellae in mid-reaction, are found to be intimately dispersed
which separated at an early stage from the solution within a matrix of amorphous material. From observa-
phase. A similar phenomenon was later observed in tions on dilute reaction mixtures [142,180], it is reason-
the clear solution crystallisation of silicalite, where the able to assume that these crystals have indeed
first visible crystals were seen to nucleate on traces of nucleated heterogeneously within the gel phase. Such a
amorphous material (‘‘gel rafts’’) which could be seen situation is credible since nutrient concentration gradi-
using optical microscopy [142]. The authors of the latter ents will be greatest at the solid–solution interface and
paper suggested that ‘‘the heterogeneous process directly the surface of the amorphous phase will provide sites
observable from the small amount of gel precipitated at which the free energy change necessary for nucleus
early in the reaction. . . may be indicative of a more gen- formation is lower than would be the case for homoge-
eral heterogeneous process involving the much smaller, neous nucleation (equivalent to a reduced number of de-
invisible, colloidal gel particles which represent the prin- grees of freedom).
cipal reservoir of silica in the system.’’ The presence of A more complex type of heterogeneous nucleation
these particles was at that time inferred from indirect has been observed by Zandbergen [182]. A Na,TPA-sil-
measurements. More recently, this has been confirmed icalite synthesis mixture was heated quiescently at 65 C
by the application of improved scattering techniques and the solid phase which began to form after two days
[31,85,181] and electron microscopy (see below). was examined from time to time by HREM. After four
For syntheses in which a visible gel phase is present days, two types of morphologies were observed within
(probably the most familiar case), examination of the the solids: (A) large (2 lm) amorphous, irregularly
reaction mixture under an optical microscope usually re- shaped particles, and (B) agglomerates of small, partly
veals gel particles of around 1000–50,000 nm (1–50 lm) crystalline particles (0.1 lm). The ratio B/A + B and
in size. These dimensions are system-dependent and also the fraction of crystalline particles in B both increased
liable to change through processes of agglomeration or with reaction time. The d-spacings detected in B
agitation-induced fission. However, crystallisation in suggested the initial formation of a dense phase. Large
Fig. 10. Direct observation of zeolite Y nucleation on colloidal gel particles by HRTEM [139]: (a) amorphous particles in freshly prepared
aluminosilicate solution; (b–d) development of crystallinity upon hydrothermal treatment at 100 C after (b) 28 h, (c) 48 h and (d) 75 h. (Reproduced
from Ref. [139] with permission.)
d-values in the ED patterns were observed only after for most ‘‘clear solution’’ syntheses is colloidal in
25 days and only in the vicinity of the dense crystalline nature).
phase, after which the presence of the MFI phase in-
creased rapidly. Very small zeolite crystals (<10 nm) 7.8. Nucleation in zeolite systems—Mechanism
were rare, indicating very rapid growth once the critical
size had been reached. This section considers the implications of the equil-
Clear solution synthesis is a special case of gel synthe- ibration reactions which lead to the formation of the
sis in which the amorphous particles remain of colloidal semi-ordered secondary amorphous phase and how
or sub-colloidal dimensions (650 nm), i.e. invisible by related processes then give rise to zeolite nuclei. In
optical examination. However, the considerations of Fig. 5, the primary amorphous phase was illustrated
crystal nucleation and growth are essentially the same. diagrammatically as a glass-like network having no
This has recently been beautifully demonstrated using elements of order. Where there is a driving force
electron microscopy for the clear solution syntheses of for zeolite formation, i.e. negative DG in the
zeolites A [138], Y [139] and Si-MFI [134] by Mintova equation
and co-workers (Fig. 10). Using high-resolution trans-
K s;gel
mission electron microscopy (HRTEM) in conjunction DG ¼ RT ln ðfrom Section 4.1Þ;
with in situ DLS, X-ray diffraction and other tech- K s;zeolite
niques, Mintova et al. have imaged the development of then there will be an overall tendency for this phase to
crystalline structure within amorphous gel particles of become more ordered, as represented in the figure by
nanometre dimensions, i.e. the process depicted sche- the appearance of grouped hexagons for the secondary
matically in Fig. 5. Single zeolite A crystals were ob- amorphous phase (equilibrated gel) and a hexagonal
served [138] to nucleate in amorphous gel particles of network to signify the periodic structure of the crystal-
40–80 nm in size within three days at room temperature. line product. (Note that, although the hexagon does rep-
The embedded zeolite A nanocrystals grew at room tem- resent a 6-T-atom ring, there is no chemical or structural
perature, consuming the gel particles and forming a col- significance in the choice of this specific unit: it is simply
loidal suspension of 40–80 nm crystals. A broadly a convenient geometric shape with which to represent
similar picture was found for the zeolite Y [139] and the development of an ordered region. It does not imply
Si-MFI [134] phases. any particular role or importance for 6-rings in general.)
It is worth noting that the only case in which a ‘‘clear The chemical route by which such progressive order-
solution’’ synthesis does not involve some amorphous ing is brought about is that highlighted by Flanigen and
material is that in which all components are in true solu- Breck [51,52] and by Chang and Bell [63], namely the
tion and crystallisation is driven by differences in solu- breaking and remaking of Si,AlAOASi,Al (TAOAT)
bility only (see Section 14.3). This is very rare and can bonds catalysed by hydroxyl ion, the related condensa-
yield only small amounts of product—a very dilute sys- tion reaction also playing a part:
tem of moderate Si/Al ratio could perhaps provide an
example. TAOH þ OAT TAOAT þ OH
A further comment can here be made concerning the
TAOH þ HOAT TAOAT þ H2 O
interpretation of the phenomenological and practical
differences observed between the ‘‘A’’ and ‘‘B’’ type However, the anions do not work in isolation and a cru-
pentasil syntheses of Gabelica and co-workers [58–62] cial role in the ordering process is played by the cations
(Section 3.6). Dissimilarities in reagent sources and present in the synthesis system. These will act to attract
reaction composition lead to changes in the partition around themselves energetically favourable coordina-
of reaction components between solution, macroscopic tion spheres of oxy-species and in so doing will generate
gel and colloid phases. These in turn give rise to the certain preferred geometries. In this way, cation-depen-
observed differences, particularly in nucleation behav- dent elements similar to those present in the eventual
iour. In case ‘‘A’’, nucleation occurs mainly on colloi- zeolite product will gradually be assembled. The notion
dal particles in the liquid phase. In ‘‘B’’, the lower base of cation-assisted ordering goes back to the ideas of
level, lower silica/water ratio and higher Na/TPA ratio Breck [54] (Section 3.2) and was more recently general-
lead to a much higher proportion of amorphous solids ised in terms of ordering through minimisation of the
and a very high rate of nucleation within the solid gel potential energy of molecular assemblies by Brunner
phase. [183]. The structuring of silicate and aluminosilicate
In summary, it is concluded that (i) the most common units through electrostatic and (for organo-cations)
process of zeolite nucleation relies on a primary nucle- van der Waals forces as proposed by Flanigen [82,83],
ation mechanism and (ii) the most probable primary Chang and Bell [63] and Burkett and Davis [64–66]
nucleation mode is heterogeneous and centred upon has also been discussed earlier (Sections 3.6 and 3.7).
the amorphous phase of the reaction mixture (which The following argument builds on these ideas to suggest
a specific mechanism for the construction of the zeolite ilarly elsewhere. Statistically, some areas of the overall
lattice. structure will be more ordered than others and in due
In Fig. 11, a cation is envisaged as migrating to a tem- course, particular ‘‘islands of order’’ will be established
porary site in a developing structure at which the coor- randomly throughout the now semi-ordered network
dination geometry is suitable but not ideal. Whilst in (as (b) in Fig. 5). These constitute the ‘‘proto-nuclei’’
place, the cation mediates the acquisition of new T-units discussed later on (Section 11). The whole system is dy-
from solution (through the condensation reactions given namic and at any one moment such islands are being
above), guiding them into a more favourable coordina- created, up-graded or (by dissolution) destroyed. How-
tion geometry and in so doing generating a periodically ever, the overall trend will be towards an increasing de-
regular local structure. In due course, the cation may gree of order. Eventually, some areas of the structure
have sufficient of its original hydration shell replaced become sufficiently ordered that a periodic lattice can
by lattice oxygen donors to become a fixture at the propagate, i.e. nucleation has occurred. This discontinu-
newly created site. Otherwise (and depending upon the ity takes the form of a topological rearrangement akin
charge balance), the cation may migrate to function sim- to an isomerisation. Such a transformation can be
viewed as a first order phase transition [184] and corre-
sponds to the achievement of critical radius described in
Section 7.2. From this point, the kinetics of accretion
sufficiently outweigh the kinetics of dissolution to result
in net growth. Thus, the reversible equilibration reac-
tions characteristic of the tr component of the induction
period (s) are superceded by the initial net growth repre-
sented by tg. However, the underlying chemistry remains
the same and the reactions still maintain a significant re-
verse component.
Zeolite nucleation is therefore a discreet event which
could be defined as ‘‘a phase transition whereby a criti-
cal volume of a semi-ordered gel network is transformed
into a structure which is sufficiently well ordered to form
a viable growth centre from which the crystal lattice can
propagate.’’ Two comments may be added. The critical
volume refers to the requirement that the germ structure
needs to be of a certain minimum size, corresponding to
the critical radius (rc) of Section 7.2. Following this, it
may be surmised that the potential nucleus does not
have to be perfect in its structure – rather, it needs to
be ‘‘good enough’’ to function as a nucleus under the
prevailing circumstances. This reflects the condition that
rc is not a fixed quantity but varies, depending on the
values of other key variables such as temperature and
supersaturation. Just as the critical size is situation-
dependent, then a variable ‘‘perfection index’’ can also
be envisaged, the necessary degree of precise periodicity
modulating with the growth environment. Thus, for
example, a rather defective potential nucleus might re-
main dormant (or dissolve) under quasi-equilibrium
conditions but may grow into a crystal under a high
supersaturation driving force. An analogy for the grad-
ual evolution of such a nucleus would be the cutting of a
new key for a tumbler (‘‘Yale’’-type) lock. Initially, there
is no chance that the key will function. However, as the
machining progresses, the new key will gradually be-
come more perfect and at some definite point will per-
Fig. 11. The basic mechanism for the cation-mediated assembly of
form the discreet action of opening the lock.
ordered regions: (a) nomenclature and symbolism; (b) details of in-situ
construction process by addition of solution units to a surface site. The Comparison of the new key with an original master
same mechanism can be applied to zeolite crystal growth. See text for key may reveal errors and imperfections but within a
full description. certain tolerance it will perform the required function
in the existing circumstances. It is ‘‘good enough’’. How- upon the method of detection—most commonly a com-
ever, if circumstances change (e.g. the lock becomes hot bination of visual inspection or microscopy with X-ray
and expands), it is possible that the ‘‘imperfect’’ key may diffraction. Thereafter, crystal growth can be monitored
no longer open it. by the same techniques and the resulting S-shaped
growth curve of bulk crystallinity against time is by
7.9. Summary far the most commonly reported measurement of zeolite
crystallisation kinetics.
As can be seen from the above, the overall process of
zeolite nucleation is a summation of a complex chain of
events. This is necessarily so since it encompasses the en-
tirety of the transformation from an initially random
structure to the beginnings of a regular, periodic crystal
lattice. However, it is suggested that the basic steps are
quite simple, namely: (i) mixing of the reactants to give
a non-equilibrated, inhomogeneous starting material
(primary amorphous phase), (ii) equilibration to form
a semi-organised precursor (secondary amorphous
phase) containing ‘‘islands of order’’ or ‘‘proto-nuclei’’,
(iii) the establishment of sufficient regular structure
within a statistical distribution of ordered sites to enable
such structure to propagate (the nucleation step itself),
and (iv) the beginnings of crystal growth on the estab-
lished nuclei. The reversible condensation reactions
which constantly make and break TAOAT bonds in
the dynamic reaction medium of hydrothermal synthesis
provide the chemical mechanism by which all these
changes are accomplished. The accumulation of T-units
(e.g. T(OH)3O) from solution is mediated by the asso-
ciated cations, which provide structural organisation for
the assembling architecture.
8. Crystal growth
In Sections 5–7, the pathway of zeolite synthesis has

been followed from the assembling of reaction compo-
nents through to the inception of crystal growth. Crucial
intermediate stages can be identified: the development of
local order and the enabling discontinuity of the nucle-
ation step. Thus, nearly all the main features of zeolite
formation have already been described. However, the
remaining stage—the growth of a nascent microcrystal
into a material entity—is of great significance, not least
because it is this final step which we most commonly
monitor in the laboratory or synthesis plant and which
gives us the tangible zeolite products, sorbents and cat-
alysts with which we are familiar. The present section
discusses the experimental observations on this process
and their mechanistic basis.
8.1. Experimental methods

Fig. 12. A cautionary tale in particle size analysis. The monodisperse
but anisotropic silicalite crystals (a) have caused particle counters
In a typical zeolite synthesis, the first substantive evi-
operating on conductivity (b) (Coulter Multisizer) and diffraction (c)
dence for a successful reaction is the appearance of crys- (Malvern 3300 Laser Particle Sizer) to report a multimodal distribu-
tals of the product. As noted earlier (Section 5), this tion based on the angularity of the crystals and their hydrodynamic
signal for the end of the induction period is dependent behaviour in suspension.
More closely related to the growth mechanism itself is [193,194]. In this technique, a mechanical probe with a
the crystal linear growth rate and to determine this some tip of atomic dimensions is rastered over the surface of
form of temporal size measurement is necessary. Unfor- the sample. The derived signals can be processed into
tunately, the crystals of synthetic zeolites are usually too images which have a lateral resolution of 20 nm in
small (<1–20 lm) and the experimental conditions too the principal (x, y) plane of the specimen but which are
severe for the application of many of the traditional sensitive to features 61 nm in size in the vertical (z)
crystal growth measurement techniques [108,153,185]. dimension. The results provide information on zeolite
Nevertheless, data can be obtained, usually from sam- crystal surfaces of astonishing detail (see below). In a
ples taken during the course of small-scale experiments, further new development, Cr-sputtered SEM images
through the use of particle counters relying on (for showing unprecedented resolution of similar surface
example) light scattering, conductivity across an orifice, topography have recently been reported [195,196].
or diffraction effects. However, unless the product crys-
tal size range is narrow, the result will be some form of 8.2. Experimental observations—Introduction
size distribution, which must be very carefully inter-
preted. Problems are likely to arise the more the sample The majority of reports on zeolite crystallisation
deviates from a spherical shape (Fig. 12). In the example kinetics are based upon bulk crystallinity measurements,
illustrated [186], the instruments have provided results usually as determined by X-ray diffraction. In these
showing multimodal distributions, based on the angu- cases, it is not possible to estimate crystal growth rates
larity of the crystals and their hydrodynamic behaviour in the absence of further information on nucleation
in suspension. (When analysed by the measurement of behaviour or crystal size distributions. However, there
SEM micrographs, the sample in fact showed a more are now in the literature a significant number of investi-
uniform size distribution (±3%) than the glass micro- gations of zeolite synthesis in which crystal linear
sphere standard supplied for calibration purposes.) growth rates have been reported (Table 3). From this
More directly, the sizes of individual crystals can be information, it is possible to draw some conclusions
measured by optical or electron microscopy. These on the nature of zeolite crystal growth in relation to that
methods were used in the first determinations of linear of other ionic, or partially ionic, substances.
growth rates (for zeolites A and X) by Zhdanov and The linear growth rates (0.5Dl/Dt) of zeolite crystals
co-workers [57,140] and were later applied to ZSM-5 vary from about 0.1 lm h1 for zeolites A and X to
synthesis by Nastro and Sand [187]. In these cases, mea- around 0.02 lm h1 for ZSM-5 (for near-optimum com-
surements were carried out on syntheses sampled or ter- positions at 90 C), whilst the values for some zeolites
minated at a series of different reaction times. The first such as EU-1 may be an order of magnitude lower [188].
in-situ growth rate determinations were made by Lowe These growth rates are 2–4 orders of magnitude smaller
and associates [94,188], who used optical microscopy than typical values for simple ionic salts (such as alums
to study the growth behaviour of ZSM-5 and other or alkali halides) or simple molecular compounds (e.g.
high-silica zeolites by direct observation of reaction mix- sucrose) [208,209]. This large difference reflects the nat-
tures contained in glass capillaries. This work was ex- ure of the zeolite synthesis reaction, in which a largely
tended by Sano, Iwasaki and co-workers using a covalent, polymeric structure is being assembled piece
specially constructed cell in which a sample of ZSM-5 by piece as TAOAT bonds are formed one after an-
reaction mixture maintained at temperatures up to other. This is clearly a more complex and intricate pro-
170 C could be directly observed using an optical reflec- cess than the packing of relatively simple units in ionic
tion microscope [189,190]. The procedure was later fur- or molecular crystals where bonding is electrostatic or
ther refined by the addition of an interferometric van der Waals and relatively non-directional.
technique which allowed the detailed observation of The plots of crystal size against time typically show a
growth behaviour on all three faces of silicalite crystals constant linear growth rate for most of the reaction with
[191]. a final tailing-off due to nutrient depletion, as illustrated
The above methods permit the growth kinetics of zeo- in Fig. 7b and also later in Section 11.3 (Fig. 22a). From
lite crystals in one, two or (in favourable circumstances) this and also from the magnitude of the activation ener-
three dimensions to be studied as a function of reaction gies (45–90 kJ mol1), it is very probable that the rate
conditions. In this way, inferences can be indirectly controlling process is the surface integration step of
drawn concerning the manner in which the crystal is crystal growth itself [33,37,188]. If the reactions were
being assembled from the components available in the generally dissolution-rate limited (as found for unreac-
synthesis mixture (Section 8.2). Most recently, data from tive silica sources [210,211]), the linear growth plots
atomic force microscopy (AFM) and high-resolution would be intrinsically curved as the reagent sources were
TEM (HRTEM) have provided more detailed evidence called upon to provide ever increasing nutrient fluxes at
of the surface construction process [192]. AFM is one diminishing surface areas. The activation energies are
of a new family of scanning probe microscopies typical of those for making and breaking TAO bonds
Table 3
Some published values for zeolite macrocrystal linear growth rates
System investigated Range of measured linear growth ratesa (lm h1) Ref.
Zeolite Temperature range (C) Other parameters investigated
Na-A 50–100 Alkalinity, seeding 0.011–0.13 [57]
Na-A 60–80 Alkalinity, SiO2 source, Na/K 0.02–0.06 [197]
Na-A 60–80 Ageing, seeding 0.14 (1.1b) [198]
Na-X 98 SiO2 source, ageing, seeding 0.083 [199]
Na-X 70–100 Ageing 0.02–0.11 [140]
NaX,Y 88 Si/Al variation 0.0005–0.2 [79]
Na-analcime 130–160 SiO2 source 1.2–5.5 [200]
Dodecasil 1H 160–200 Seeding, stirring 0.47–0.98 [201]
-ZSM-5
M,NH4-TPA- 177 (Si/Al = 30 throughout) 0.02 (zero M), [187]
M = Li, Na, K 0.11–0.72
Na,TPA- 170 Na, TPA, OH, SiO2 0.5–1.1 [202–204]
Na,TPA- 170 Si/Al = 50–400 0.6–1.4 (2.3c) [143]
Na,TPA- 170 pH, TPA/Na (F system) 0.08–1.0 [205]
Na,TPA- 150–200 – 0.4–3.2 [206]
Na,TPA- 90–175 Composition and conditions 0.01–1.17 [142,188]
Na,TPA- 150–170 Si/Al P 50 0.2–0.6 [190,191]
a
0.5Dl/Dt.
b
Anomalous value (see text).
c
Anomalous value, but similar figure also suggested elsewhere (Ref. [207, Fig. 10]).
[212,213] and much higher than would be expected for a minium compounds, tetramethylammonium salts or
process controlled by diffusion from the bulk liquid ethanol reduced the observed value.
phase to the crystal surface [214]. This conclusion is fur- Several groups of workers have investigated the effect
ther supported by the work of Schoeman and co-work- of dilution and base level or pH upon growth rate. As
ers who applied Nielsens method of chronomal analysis part of their in-situ study of silicalite crystallisation,
[215] to the synthesis of colloidal silicalite [216,217]. The Iwasaki et al. [191] measured the growth rates at
best linear fit to their data was obtained by assuming a 150 C of the (0 0 1), (1 0 0) and (0 1 0) crystal faces for
first order surface reaction to be rate limiting. values of H2O/SiO2 from 75 to 300 at final crystal
As indicated above, zeolite crystal linear growth rates lengths of 40 lm. Rates progressively decreased in
show a strong temperature dependence. However, even each case, the c-axis value declining from 0.6 to
for a given structure type, the rates are also influenced 0.2 lm h1. In approximately the same dilution range,
by other synthesis variables, particularly reaction com- other workers [188] concluded that growth rate was
position. In a few cases, systematic studies have been re- not a sensitive function of the concentration parameters,
ported, although most of these relate to the ZSM-5 although the overall tendency was for the rate to in-
system. crease with concentration and with increasing base.
Tavolaro et al. [205] studied silicalite crystallisation in
8.3. Experimental observations—Studies of a fluoride system at 170 C as a function of the pH of
macrocrystalline systems the initial gels. For crystals of 300–500 lm in size, lin-
ear growth rates were found to vary from 0.08 lm h1 at
Kacirek and Lechert [79,147] provided the first quan- pH 2.6 to 1.0 lm h1 at pH 6.7. A plot of [log (growth
tification of the observation that the formation rate of rate)] against pH was found to be linear. These results
faujasite decreases with increasing silica content of the are in contrast to those determined for some colloidal
product. The limits of the series shown in Table 3 corre- systems (see Section 8.4).
spond to Si/Al = 1.4 (growth rate = 0.2 lm h1) and Si/ An interesting observation relates to the variation in
Al = 3.4 (0.00053 lm h1). A detailed in-situ investiga- ZSM-5 crystal aspect ratio with reaction conditions,
tion of the MFI system by Cundy et al. [188] examined showing that each unique crystal face may respond dif-
crystal length and width growth rates as a function of ferently to changes in the system. At constant composi-
temperature, concentration, cation type, aluminium tion, major increases in the length/width ratio are seen
content and addition of salts (Na and TMA halides) with increasing reaction temperature [188,206] and sim-
or ethanol. At a given temperature (the most sensitive ilar changes are seen as a function of reaction composi-
parameter), the highest growth rates were found for sil- tion [188,191,202–204]. For high-pH systems, the
icalite at moderate base levels, whilst addition of alu- crystals elongate as the pH is reduced [207,218–220]
whereas for a fluoride-mediated system, the aspect ratio served growth behaviour is similar to that found for lar-
increases with pH in the range pH 3–7 [205]. Crystal as- ger (>1 lm) crystals. However, different trends are
pect ratio may also vary during the course of a synthesis observed in the effects of dilution and virtually no data
[148,186,205], an indication of changes in the growth beyond radial growth rates are available in view of the
environment as the reaction proceeds. very small size of the particles. The vast majority of pub-
Even if the reaction product and its morphology do lished work has been carried out on MFI systems, in
not change, variations in reaction conditions can still ex- view of the convenience of the available ‘‘clear solution’’
ert a powerful influence upon the course of a synthesis syntheses.
by altering the nucleation behaviour. Cundy et al. Pioneering work on colloidal zeolite crystals and
[142] studied the crystallisation behaviour of a single determination of their growth behaviour using dynamic
synthesis composition for silicalite at a fixed tempera- light scattering (DLS) has been carried out by Scho-
ture (95 C) by varying the order of mixing of reagents, eman and co-workers, including a detailed in-situ
the type of reaction vessel, the stirring speed and the de- DLS study at 70 C [181]. For colloidal silicalite
gree of ageing. In all cases, pure crystalline silicalite was [216,217], radial growth rates (e.g. 1.9 nm h1 at
obtained in high yield at the same linear growth rate 98 C) were much lower than those reported earlier
[(1.95 ± 0.05) · 102 lm h1] but there were major for macrocrystals (e.g. 20 nm h1 at 95 C [142]) and
changes in particle size distribution. Analysis by the were essentially independent of alkalinity. However,
method of Zhdanov and Samulevich [140] gave the cor- the growth rates approached the existing values upon
responding nucleation profiles, which demonstrated dilution (e.g. growth rate ! 5.0 nm h1 for a factor
how both the yield and the size distribution of the prod- of three dilution and ! 9.3 nm h1 for a factor of 12
uct at any given time are strongly linked to the treatment [85]). A similar dilution effect was observed by Tsay
of the reaction mixture. In a separate study, the chemi- and Chiang following a determined growth rate value
cal changes surrounding the growth and dissolution of of 1.6 nm h1 at 115 C [221].
Na,TPA-ZSM-5 crystals were investigated in detail Twomey et al. [222] have also investigated the nucle-
[111]. The zoned composition of the product crystals ation and growth of silicalite using an ex-situ DLS
was shown to be a natural consequence of the evolving method. In addition to temperature variation, the effect
reaction environment and to reflect the composition of of ageing and change of cation was also investigated.
intermediate samples taken during crystal growth. At 96 C, the observed growth rate of 18 nm h1 was
One of the very few zeolite crystal growth studies to found to increase to 42 nm h1 after an ageing period
have been applied to non-batch preparations concerns of nine days. The probable reason for this difference
the synthesis of MFI-type materials under semi-continu- is discussed below. Substitution of K for Na in the
ous conditions [148,186]. It is demonstrated that the reaction composition gave an almost identical radial
process is controlled by an equation of the form growth rate. A similar DLS measurement of the radial
dx u growth rate for colloidal silicalite at 100 C by Cundy
¼ constant ; and co-workers gave a value of 11.7 nm h1, whilst fur-
dt A
ther growth in a different reaction mixture using the
where the crystal linear growth rate dx/dt is governed by initial colloidal product as seed gave 25 nm h1 [150].
the ratio of the nutrient mass feed rate (u) to the instan- Other workers have employed X-ray scattering tech-
taneous total surface area of the crystal population (A). niques. De Moor et al. [223] have used in-situ time-re-
This relationship was explored for a wide variety of solved ultra-small angle X-ray scattering (USAXS) to
reaction conditions. Two effects are observed as the measure the growth rates of Si-TPA-MFI crystals in
nutrient feed rate is increased: (i) the product crystal lin- the 50–400 nm size range. Linear growth rates at
ear growth rate (dx/dt)obs increases in accordance with 125 C of 36 ± 3 nm h1 were found for Si/OH values
the above expression, up to a fixed value (dx/dt)max cor- of both 2.4 and 3.0.
responding to the maximum observed in a batch synthe- It is notable that the growth rates found in the
sis, and (ii) there is an increasing tendency to nucleate experiments of Twomey, Cundy and de Moor are very
new crystals as ratio (dx/dt)obs/(dx/dt)max (the ‘‘effective close to the expected macrocrystal values, although the
supersaturation’’) rises. reaction compositions are at first sight very similar to
that employed by Schoeman and co-workers. The prin-
8.4. Experimental observations—Studies of cipal difference lies in cation content of the reaction
nanocrystalline systems mixture. The preparation used by Schoeman and col-
leagues is very low in alkali metals (Na/TPA = 0.02,
In recent years, the increased use of scattering tech- K much lower), whereas that of Twomey et al. has
niques has led to a considerable number of investiga- Na/TPA = 0.2 and the others are comparable or
tions on the formation and growth of nanosized higher. It seems probable that the steric stabilisation
zeolite particles (usually <100 nm). In general, the ob- conferred by the envelope of TPA cations surrounding
the colloidal particles may also cause mass transport where cr is the solubility of small particles of size r and c*
limitations. Dilution, or the introduction of smaller is the equilibrium (saturation) solubility of large parti-
and more mobile cations may alleviate these, as well cles (i.e. r ! 1); M is the molecular mass of the solid
as facilitating the process of crystal growth [150]. in solution, r is the surface energy of the solid particle
Some comparison with the MFI system is possible in contact with the solution and q is its density.
from two studies on the crystallisation of zeolite A at The significance of such a size-dependent solubility
sizes up to 1 lm. In a very significant and carefully for zeolite synthesis would be to bring about a reduced
considered paper, Gora et al. [198] found radial growth crystal growth rate at early reaction times when the crys-
rates of 139 nm h1 at 60 C and 1100 nm h1 at 80 C. tals are very small. The consequences of this have been
As with the related study on silicalite [222], the growth discussed by Warzywoda et al. [110], working from the
rate of the crystallites was strongly influenced by ageing, moment equations contained in the Thompson popula-
increasing at 60 C to 334 nm h1 after 10 h of ageing at tion balance model [89–91] (Section 4.1). However, it
25 C. In this case, further investigation showed that the is possible to achieve a similar result algebraically as
particles formed from the aged solution were agglomer- follows:
ates. The anomalous apparent increase in growth rate is For a spherical crystal of radius r, the radial growth
thus explained, since polycrystalline particles have a rate is given by
higher surface area per particle and can therefore assim-
dr hc c in
ilate nutrient faster than single crystals with regular pla- ¼k
eq
;
nar faces. Otherwise, it is difficult to see why the ageing dt c
of a reaction mixture should change the linear growth
rate of the resulting crystals, unless some physico- where k is the radial growth rate constant, c is the actual
chemical change has occurred to remove a kinetic solution concentration, ceq is the equilibrium solubility,
restraint—for example, the solubilisation of a nutrient c* is the saturation solubility for a large crystal and n is
source having a very slow dissolution rate. The surpris- an exponent (normally 1–2) [108].
ingly large increase in growth rate between 60 and 80 C For large crystals, ceq = c*, but for small crystals
has not been accounted for, although the authors note where the solubility is size-dependent, ceq = cr and the
that the data at this higher temperature refer to the lar- supersaturation [(c cr)/c*] is not constant. Thus,
ger component of a bimodal population of product obtaining cr from the Gibbs–Thomson equation
crystals. (above),
Using a very similar reaction composition to that of n
dr cgel 2Mr
Gora and colleagues, Singh et al. [224] also used DLS ¼ k exp ;
dt c RT qr
to measure the growth rate of zeolite A at temperatures
of 30–60 C. Once again, ageing was found to shorten where, for zeolite synthesis, c = cgel, the zeolite gel
the induction period and increase the growth rate. Val- solubility.
ues found for radial growth rates from fresh solutions If reasonable values for silicalite synthesis are in-
and those aged at room temperature for a few hours serted into this equation (with n = 1 and r taken as
were—at 40 C: 15.5 nm h1 (fresh), 28.8 nm h1 (aged); 5 · 106 J cm2), then numerical solutions show that
at 60 C: 94 nm h1 (fresh), 165 nm h1 (aged). The the dependence of growth rate upon size becomes insig-
growth rates at 60 C therefore compare fairly well with nificant (<2%) at values for r of around 5–10 nm, i.e. for
those determined by Gora et al. and also with values crystals greater than 10–20 nm in size. This agrees
determined for the growth of larger crystals quite well with the more accurate calculations of War-
(Table 3). zywoda et al. (cf. Fig. 5 in Ref. [110]) and also with
the experimental observations of Schoeman [181], who
found that the growth rate of colloidal silicalite at
8.5. Size-dependent growth of nanocrystals
70 C was non-linear only below a mean crystal size of
20 nm. However, whilst such correlations are reassur-
Very small particles have an unusually high surface
ing, the agreement may be fortuitous. There are no
free energy associated with their extreme curvature
established values for the surface energy, and both sets
and as a consequence display anomalous properties.
of calculations are based upon a simple model of radial
Nanosized liquid droplets show increased vapour pres-
growth with no allowance made for particulate addition.
sure. The enhanced solubility of very small solid parti-
In reality, the growth units are ‘‘quantised’’ in mass and
cles is given by a relationship variously described as
size, so that, at the nanometre and sub-nanometre level,
the Kelvin, Gibbs–Kelvin, Gibbs–Thomson or Ost-
the increase in size is not continuous. Nevertheless, it
wald–Freundlich equation [108,110]:
h c i 2Mr does seem probable that the Kelvin effect is only signif-
ln ¼
r
; icant in zeolite synthesis at the very earliest stages of
c RT qr crystal growth.
8.6. Growth models
The slow linear growth rates of zeolite crystals at

fairly high supersaturations [148] and the rarity of the
observation of growth spirals [196,225,226] suggests that
the predominant growth mode is that of the adsorption
layer type [108], controlled by surface nucleation rather
than by the dislocations necessary for the Burton–Cab-
rera–Frank mechanism [227]. A simple picture of this is
given in Fig. 13. Following Gibbs–Volmer theory [228],
a growth unit is adsorbed on the growing crystal face
and migrates to its optimum location (a). Following
completion of a given layer (b), no further growth can
occur in the absence of a dislocation. This hiatus is over- Fig. 14. Kossel-type model of a growing crystal surface, showing a
come by the creation of monolayer island nuclei (two- surface-adsorbed growth unit (D) on a flat surface (A) and also steps
(B), kinks (C), an edge vacancy (E) and a surface vacancy (F). The
dimensional nucleation) (c) from which the growth kink site (C) has the highest binding energy and is thus the most likely
process can continue on a layer-by-layer basis. Clearly, point of attachment for a growth unit (after Kossel [229] and Mullin
[108]).
this is a greatly oversimplified representation. As one

of many authors who have worked to make the picture
more realistic [108], Kossel [229] has provided a model
showing the sites of high binding energy at which
growth units are most easily incorporated (Fig. 14).
Direct evidence for the layer growth model is emerg-
ing from AFM studies (Fig. 15), the growth terraces
sketched in Figs. 13 and 14 being clearly identifiable.
In Fig. 15a, square terraces from different nucleation
points on the surface of zeolite A are seen to merge, giv-
ing rise to scalloped growth fronts progressing across
the crystal. The elliptical terraces on the surface of zeo-
lite SSZ-42 (Fig. 15b) also merge to give cusp-shaped
growth fronts but in this case the shape persists, rather
than growing out. This is probably linked to a reduced
growth rate at kink sites, a factor also responsible for
the elliptical nature of the terraces themselves. In both
these cases, layer growth is proceeding simultaneously
from many nucleation points distributed across the crys-
tal surface. However, for the small (1 lm) crystal of zeo-
lite Y illustrated in Fig. 15c, the step pattern appears to
arise from a much reduced rate of surface nucleation.
Detail from a large (60 lm) silicalite crystal (Fig. 15d)
demonstrates the complex terrace structure in the vicin-
ity of one of the ‘‘ramps’’ on the (0 1 0) face. The circular
terraces on the bulk (0 1 0) face show the radial decrease
of terrace spacing typical of the closing stages of batch
growth [230]. However, terrace distortion in the vicinity
of the ramp suggests that the ramp corners have acted as
nucleation centres. It is also possible that the intruding
Fig. 13. Zeolite crystal growth via a layer-by-layer mechanism. A surface is that of a (1 0 0) face, i.e. a 90 intergrowth
growth unit is adsorbed on to the growing crystal face and migrates to [231]. In all the above micrographs, the step heights cor-
a high-energy kink site at which the number of points of attachment
are maximised (a). Following completion of a layer (b), further growth
respond to identifiable macro-features of the unit cell,
can occur only after the creation of monolayer island nuclei (c) (after for example the thickness of a pentasil chain (1.0 nm)
Mullin [108]). in the silicalite image.
Fig. 15. AFM images of zeolite crystal surfaces: (a) zeolite A, (b) SSZ-42, (c) zeolite Y and (d) silicalite. Scale bar below each image = 1 lm. See text
for discussion. (We thank Itzel Meza and Dr J.R. Agger of the University of Manchester Centre for Microporous Materials for these images.)
Using a representation of the Gibbs–Volmer type, lite A corresponding to the sizes of a double-4-ring
Agger and colleagues have modelled the growth of zeo- (D4R) [196,236]. These have been interpreted [236] as
lite A in terms of the features seen in AFM images [232– suggesting that the D4R unit is the key building unit
234]. Growth probabilities are assigned to the potential for crystal growth in zeolite A. The authors speculate
growth sites (edge, kink, etc.) and the relative values for that ‘‘first Si, Al and O bond together to form the
these probabilities are adjusted until the computed unit- D4R, and then D4R units bond together, like building
by-unit growth pattern matches that observed experi- blocks, to form the LTA structure’’—an idea strongly
mentally. The information gained by this procedure at reminiscent of the early suggestions of Barrer and of
present provides no indication of the nature of the Breck (Section 3.2). An alternative possibility is put for-
growth units in the real system. However, it is found ward below (Section 8.2), where this case would provide
that, to match the observed features, growth at kink one example of the cation-mediated ‘‘in-situ’’ construc-
sites has to be 500 times faster than growth on the clean tion of a microporous framework. The observation of
surface and 15 times faster than growth at an edge site. a surface terminated by double-4-rings would suggest
It is also apparent that the overall advance of a growth that this configuration represents a metastable interme-
face must be controlled by the surface nucleation rate, diate in the layer-by-layer construction process, i.e. (i)
the lateral spreading rate being much faster. The chal- slow 2D nucleation of a new surface layer, (ii) relatively
lenge for this type of approach is to forge a link between rapid lateral spread of that layer, (iii) slow nucleation of
the parameters used to fit the experimental data and a further surface layer, etc. On a time-averaged basis,
their chemical basis, as has been achieved in the simula- the growing structure would be most often observed in
tion of ice crystal growth by Wathen et al. [235]. An state (ii), which would correspond to the AFM result,
early attempt to describe zeolite crystal growth by a although this does not explain why the surface should
chemically significant model is summarised in Appendix apparently terminate in D4R units rather than sodalite
A. cages. A possible rationalisation for this would be that
A very interesting finding from recent AFM studies the cations are more tightly held in the square prisms,
has been the observation of steps on the surface of zeo- rendering these units relatively more stable (less soluble)
in the reaction medium. Similar work by Wakihara et al. units. In the zeolite synthesis process, the first important
[237] was carried out on faujasite by observing crystals transformation brought about by this means is the con-
isolated from seeded growth in dilute solution. From a version of the initial random gel structure (primary
seed at near equilibrium with the solution, the initially amorphous phase) into a semi-ordered network (second-
rough surface was found to have become a well-ordered ary amorphous phase). In the next stage, the generation
(1 1 1) face. For the solution-seed growth environment, of heterogeneous nucleation sites is brought about in
most of the top surfaces of the crystals were found to the same way. The third manifestation of the equilibra-
be terminated by D6R units, while some presented com- tion/cation-structuring mechanism is in crystal growth.
plete or incomplete sodalite cages. These results were This can follow an essentially identical pattern to that
taken to show that aluminosilicate species equal to or shown in Fig. 11 except that the process must be capable
smaller than a 6-ring contributed to the crystal growth. of self-perpetuation. However, this is a case of ‘‘success
breeding success’’ since the more perfect the (n 1)th
8.7. Mechanism structure, the easier it is to construct without error the
nth repeat.
Some of the main observations relating to the growth The sequence is therefore as follows. A cation is
of zeolite crystals have been summarised above. The envisaged as docking at a surface site on an amorphous
question now arises as to the nature of the processes particle. Monomer units from solution then become at-
occurring at an atomic and molecular level which are tached to the growth site and are assembled in an or-
responsible for the observed behaviour. In general, par- dered array around the cation to create a cyclic unit.
ticles may increase in size in two ways: (i) by addition of (Note again that the depiction of hexagonal units in
growth units, and (ii) by aggregation. The latter is a spe- Fig. 11 is illustrative only and has no particular chemical
cial case of the former in which the particles added are significance.) It is a requirement that, in the course of
of comparable size to the existing particle, whereas this process, a new cation site is generated so that the
growth units are assumed to be incremental in terms coordination steps will iterate and the cycle continue.
of size. Whilst there is abundant evidence for the aggre- Gel equilibration, nucleation and crystal growth thus
gation and coalescence of zeolite macrocrystals during share a common mechanism. However, as the structure
their growth [146,148], the predominant growth mecha- develops, different stages of the overall growth process
nism clearly involves the step-by-step addition of much may have different activation energies and rates.
smaller units. In recent years, far more work has been
carried out on the growth of zeolite crystals of colloidal 8.8. Summary
size than on their larger counterparts. For these nano-
scale situations, the distinction between the two modes Zeolite crystals appear to grow rather slowly, com-
of size increase is far less clear cut. In the following dis- pared to ionic crystals (such as common salt) or molec-
cussion, it is assumed that growth is taking place on ular crystals (such as sugar). The reason for this lies in
macroscopic zeolite crystals, i.e. those of maximum the necessity to construct a semi-covalent three-dimen-
dimension J 0.5 lm, and that the growth mechanism sional lattice: a polymer of ‘‘TO2’’. The predominant
is predominantly one of unitary addition. Furthermore, growth mode is that of the adsorption layer type [108],
the growth species are taken to be small solution units, where the overall rate is controlled by the surface inte-
predominantly the silicate and aluminate monomeric gration step and nucleation of a new layer is slower than
anions Si(OH)3O and AlðOHÞ 4 . Other possibilities the lateral spread of such a layer.
are later treated in more detail (Section 9). At present, The overt features of equilibration, nucleation and
the primary concern is with the chemistry of the growth crystal growth have been discussed in terms of an overall
process. model for zeolite crystallisation. The pre-requisites for
In Section 7.8, there is described a long-recognised, this mechanism to operate are:
very simple but fundamental equilibration mechanism
of TAOAT bond making and bond breaking, based 1. The interactions are mediated by small solution units
upon condensation and nucleophilic displacement reac- such as silicate monomer (see Section 9).
tions. This facilitates dynamic structural modification 2. The mechanism relies on an in-situ, localised con-
through a hydroxl ion catalysed polymerisation–depoly- struction process rather than the assembly of large,
merisation of the silicate or aluminosilicate gel network. pre-fabricated units as required by some growth the-
Further structuring is achieved principally through the ories. Also, not all the growth units forming the final
absorption of cations at the surface and their reaction crystal may have arrived by solution transport. In a
with small growth units from the solution phase to con- situation in which the amorphous phase is both the
struct ordered regions. During this cation-mediated nutrient reservoir and host of nucleation sites, some
assembly process, some or all of the water in the coordi- local reconstruction of the gel particle may become
nation sphere of the cation becomes replaced by silicate built in to the final product (Fig. 16). If there are
Fig. 16. Progress from a gel particle to crystalline zeolite (see also Figs. 5, 10 and 11). From an initially amorphous structure (a), areas of local order
are established (b), some of which develop into crystal nuclei (c) and grow by acquisition of building units from solution (d). To provide such growth
units, amorphous material is dissolved (e) to supply nutrient to both distant (e-i) and nearby (e-ii) growth sites. The distant site could be located on
another gel particle. For the nearby site, crystal growth may reduce in the limit to a local reconstruction of the host gel particle. Eventually, all
amorphous material is converted into an approximately equal mass of zeolite crystals (f).
several nucleation sites on a single gel particle, this any particular units do have especial significance for
situation must inevitably occur where 2 or more the genuine chemical construction of certain zeolites,
growth areas converge and unite. then it may be because such units are ordered and assem-
3. A key feature of the postulated mechanism is that the bled at the growth site by the participating cations. Thus,
assembly process at the growth points is subject to the mere occurrence of such units in the solution phase
the organising influence of polar structuring agents, (e.g. as detected by NMR spectroscopy) is unlikely to be
usually cations. significant.
4. The system must be dynamic, with the chemical reac- A complicating factor is the possible duality of
tions reversible. In this way, errors which do occur in growth by (i) unitary addition (as above) and (ii) particle
the overall process can be corrected and the structure aggregation. However, this is unlikely to introduce any
can propagate. Similarly, there must also be a mech- new chemical features, being more a variant on the basic
anism for redissolution, although this may not process described above. A full discussion is given in
necessarily be the simple reverse of the synthesis Section 9.
reaction.
Part III: Key topics
The above proposal is compatible with the known
experimental data and also with many of the published The explanation given in Part II (Sections 5–8) is con-
ideas for synthesis mechanism. It additionally provides sidered to represent the most likely mechanism for zeo-
an explanation for the formation of zeolites from poly- lite formation in the majority of hydrothermal zeolite
meric templates [31,238–241], where macromolecules syntheses. It is self-consistent and chemically plausible.
extend through, and are encapsulated by, many frame- The detail of individual steps is already well established
work unit cells. With a localised construction process, and no new or special conditions have to be invoked.
a single crystal can be assembled around surface- However, some degree of qualification may be required
adsorbed polymer molecules, convergent growth proba- for unusual reaction situations. Also, in setting out these
bly occurring at several sites simultaneously. The suggestions, other views and some associated topics
suggested mechanism also does not completely exclude relating to the nature of the synthesis mechanism have
earlier proposals of key roles for particular building often been mentioned only in passing. It is the purpose
units (e.g. specified rings or polyhedra) although these of Part III of this Review (Sections 9–18) to deal more
are regarded with caution (see Section 9). However, if fully with these issues.
9. The nature of growth species and the role of the real structure of silicate and (where known) alumi-
aggregation processes nosilicate anions in solution differs considerably from
that of most SBUs, (ii) there is no correspondence be-
In Section 8.6, mass increase by the unitary addition tween the occurrence of particular solution anions and
of simple solution species was taken as the most likely the nature or formation rate of a particular zeolite,
growth mechanism for macrocrystals (J 0.5 lm). How- (iii) zeolites of widely different structure can be formed
ever, it was recognised that aggregation processes may from solutions containing apparently similar distribu-
also play a significant role, especially for very small crys- tions of anions, (iv) complex zeolite structures are
tallites. In this case, the growth units are not soluble spe- formed in situations (high dilution, high temperature,
cies but particles of comparable size to the existing high levels of Na and K) where the solution anions
particle. The ‘‘in-situ construction’’ process proposed are known to revert to the simplest species, (v) com-
for macrocrystal growth provides a realistic mechanism ponents can be added which change the course of
for the generation of the periodic crystal lattice. It is less the synthesis but do not materially alter the anion dis-
easy to see how such a regular, and essentially perfect, tribution (e.g. the appearance of MAZ-structured
structure can be formed by the addition of large blocks products in zeolite Y synthesis upon addition of
of new matter, especially if such material is wholly or TMA cations), (vi) structural predictions (e.g. on the
partly amorphous. Grain boundaries and residual glassy density of defects) made on the basis of an assumed
areas would presumably be observable in the product. D5R precursor in MFI synthesis are found to be
The nature of the species which carry mass to the incorrect.
growing crystal is not known with certainty and the res- It is interesting that the 29Si NMR assignments of
olution of this issue presents a challenging problem. The double-ring structures which gave rise to some of the
reason for the difficulty lies partly in the dearth of in-situ original speculation [245] are now believed to be errone-
micro-techniques for probing the chemistry of the ous [246]. However, such a theory can never be com-
boundary layer at the crystal surface but also in the pletely rejected since there is (as in catalysis) always
complexity of silicate sol chemistry, which offers such the possibility of an active component present at a low
an array of candidate species. The following paragraphs (and conceivably undetectable) level.
examine first the arguments for crystal growth by the Szostak [247] has introduced an interesting variant
addition of both simple and also more complex solution on the ZSM-5 synthesis mechanism proposed by van
species. Since it is tacitly assumed in these discussions Santen et al. [242]. The original suggestion considered
that the synthesis in question is of the traditional formation of ZSM-5 via ring opening and polymerisa-
OH-mediated type, there follows a short discussion tion of double-5-ring silicate anions (Fig. 17a). In the
of other possibilities, for example systems based on fluo- modified version, emphasis is placed on the role of uni-
ride ion. Finally, the features of growth by aggregation versally present free silicate monomer species (Q0),
are evaluated and the consequences of such a mecha- which are largely responsible for further growth once
nism are considered. a single or dimerised D5R unit has been formed
(Fig. 17b). The depletion of silicate monomer by the
9.1. Growth from soluble, pre-fabricated units growing chains and sheets would be made up from
the solution phase equilibrium of silicate species, as
There exists a structural similarity between some in Thompsons ‘‘tugging chain’’ mechanism (Section
solution species and the building units which appear 9.2). This modified scheme has much in common with
in zeolite frameworks. Several groups of workers have that outlined in Section 8 and is altogether more
accordingly adopted the idea that a zeolite containing credible.
(for example) double-5-rings is constructed by the A more recent revival of the idea that absolutely spe-
extraction and polymerisation of these units from cific units are required for the synthesis of particular
the distribution of such species available in solution, zeolites is seen in the nanoblock hypothesis of the Leu-
the equilibrium then rapidly shifting to make up the ven research group [67–73] (Section 3.7). Both the exper-
deficit. Although attractive at first sight, this theory imental observations and their interpretation are now
does not survive close examination. Prominent sup- the subject of active debate [86–88]. Whilst it is not
porters of this idea, especially as applied to MFI syn- inconceivable that such a prescribed mechanistic course
thesis [242], have given an excellent account of its may be followed under a particular set of experimental
genesis and subsequent demise [243], whilst the more conditions, it seems overwhelmingly unlikely that such
general conjecture of zeolite growth ‘‘from secondary a situation represents the generality. Pre-nucleation
building units’’ (SBUs) has likewise been thoughtfully building units (PNBUs) are also the key feature of
analysed and rejected [244]. mechanisms proposed for the formation of new open-
The above papers [243,244] should be consulted for framework materials discovered recently by Férey
their detailed analysis but the main points are that: (i) et al. (Section 18).
Fig. 17. Suggested mechanisms for the formation of zeolite ZSM-5: via (a) ring opening and polymerisation of double-5-ring silicate anions [242],
(b) nucleation by the SBU followed by growth through addition of monomeric TO2 units [247]. (Adapted from Ref. [247].)
Variants of the above hypothesis include those and other aluminous types by the assembly of ring units
founded on topological considerations. One such are based on spectroscopic studies of synthesis solutions,
suggestion (by Melchior et al. [248]) is based primarily for example by 29Si and 27Al NMR spectroscopy [249],
on 29Si MAS NMR spectroscopic data and argues Raman spectroscopy [250–252] and UV Raman spec-
from aluminium-ordering patterns that the fauja- troscopy [253]. Although such arguments are thought-
sites examined must have been assembled by the provoking, they lack chemical detail and suffer from
pathway similar objections to those mentioned above since there
is not necessarily any direct link between the existence of
4R ! D6R ! FAU; species in solution and the nature of solid phases which
crystallise from those solutions. The hypothesis of in-
where the single four-ring (4R) is seen as the secondary situ construction at the crystal surface in a reaction-crys-
building unit leading to the double six-ring (D6R), tallisation process (Section 8.7) does not require the
which itself is the immediate precursor to the FAU lat- existence of pre-fabricated building blocks. What is
tice. Similar schemes for the formation of zeolites A, Y needed is made as the structure is generated and many
of the arguments for the presence of particular precursor solution phase would contain essentially only mono-
units can be adapted to this reaction-at-growth-site mers and dimers during the crystallisation of zeolite
model. As Thompson [37] has pointed out, ‘‘the fact that X, with higher oligomers (perhaps up to Si20) present
the solution has the thermodynamic tendency to make in the synthesis of the more siliceous Y-types. Since
small oligomeric structures similar to the final crystal these two zeolites share a common structure (FAU),
structure also suggests that material having the same it is unlikely that the construction process differs
structure can be assembled at the crystal–solution inter- markedly in the two instances. Hence, the likelihood
face, i.e. that the surface reaction would be expected to is that simple units are the growth species in both
create similar crystalline material.’’ cases. On the same basis, it could be argued that
the growth rate should be proportional to the frac-
9.2. Growth from simple species tion Sin/Sitotal of simple units in the reaction mixture
(where n = 1 or 2). This is at least qualitatively true,
A more credible idea for zeolite crystal growth is that since the calculated linear growth rate decreases as
it occurs predominantly from simple units, especially the Si/Al ratio (and hence degree of silica polymerisa-
monomer, although all solution species may contribute tion) increases [79].
to some extent [37,94,142,188,222,244,254,255]. Whilst 5. For the case of aluminate, there is no doubt that the
it is impossible (at least for the present) to provide building block is AlðOHÞ 4 monomer since under the
unequivocal proof of this theory, the circumstantial evi- normal conditions of synthesis this is the only alumi-
dence in its favour can be summarised as follows: nate species present [258,259] (with the assumption
that there is no direct participation of aluminosilicate
1. The in-situ construction hypothesis (Section 8.7) does polymers).
not demand the presence of any specialised building 6. In a simple, two-dimensional computer model of zeo-
blocks, requiring only that the simple solution silicate lite growth from poly-TO2 solution species (see
species re-equilibrate very rapidly—as is known, pre- Appendix A), it has been shown that only simple spe-
dominantly from NMR spectroscopy, to be the case cies are viable for network propagation at realistic
[256,257]. rates. Larger and more complex units incurred a time
2. From (1), it is apparent that solution species which penalty and led to potentially fatal faults as all the
are not used directly are consumed indirectly in the statistically possible docking combinations were
re-equilibration process. This is very well expressed tried. Only a few of these (depending on the symme-
in Thompsons ‘‘tugging chain’’ model [37] which, try and conformation of the unit) were successful in
in common with the mechanism suggested earlier propagating the structure. With monomer as the
(Sections 7.8 and 8.7), proposes that elements of the growth unit, every docking that becomes chemically
final crystal structure are assembled at the crystal– locked must be successful. Every potential growth
solution interface from solution species: ‘‘. . . of the unit will have (in given circumstances) a sticking fac-
myriad silicate oligomer species present at the crys- tor and an error probability. In the case of monomer,
tal–solution interface, there may be only a few which the latter is zero.
are incorporated, e.g. monomers and dimers. . . The
other, larger, more complex species are continually
unraveled, at a relatively fast rate, to maintain the 9.3. Mineralising agents other than hydroxide
equilibrium distribution of oligomers, or very near
to it. Thus, it is quite possible that the whole process This account of the background to the synthesis
is governed by the ordering of silicates around the chemistry of zeolites has concentrated almost exclusively
pertinent template species adsorbed at the crystal on OH-based systems. This is partly because they are
surface.’’ the most common but principally because so little de-
3. Under semi-continuous synthesis conditions in which tailed information is available on any of the other con-
silicate monomer is by far the dominant solution spe- tenders. For a synthesis system to operate successfully,
cies [148], crystal linear growth rates are directly pro- it must contain some agent (the ‘‘mineraliser’’) which
portional to the rate of supply of nutrient solution performs the following functions, acting in many ways
and the crystals formed are of a very high degree of as a form of catalyst: (a) it must convert the starting
crystallinity. materials into mobile forms (for example by taking them
4. As noted earlier (Section 3.5), Kacirek and Lechert into solution or the vapour phase), (b) it must convey
[79] interpreted their detailed kinetic studies on chemical reactivity such that the mobile units (e.g. sili-
seeded faujasite syntheses in terms of a solution cate anions) can react together to form new chemical
growth model, the rate-determining step being the bonds and generate the zeolite framework, and (c) it
connection of silicate species to the surface of the crys- must de-complex from the mobile units during or after
tal. They noted that, under their conditions, the stage-(b) so that the new structure can exist as a stable
solid phase. The hydroxide ion performs these tasks very silicalite system, using advanced scattering and diffrac-
effectively. tion techniques. The basic argument is that primary
The most viable alternative to OH as a mineraliser is sub-colloidal particles with diameter 3–4 nm aggregate
the fluoride ion. Introduced into zeolite synthesis by and ‘‘densify’’ in a series of steps to give clusters with
Flanigen and Patton [260], it came to prominence only a size of 6–7 nm and amorphous to X-rays. After this,
more recently, largely due to the pioneering research the densified clusters aggregate to particles of around
programme of the Mulhouse school [261–265]. Unlike 50 nm in size which display X-ray crystallinity. The pri-
the traditional synthesis in basic media, the fluoride- mary particles correspond to the hydrated tetrapropy-
mediated route can operate over a very wide range of lammonium silicate complexes identified by various
pH values. The ability to function under acidic condi- authors [63–66,128,272]. Aggregation models are de-
tions has proved advantageous for the introduction of scribed by Dokter [273] de Moor [274,275] and co-work-
heteroelements which form unreactive precipitates in ers, Regev et al. [129], Nikolakis et al. [276], Corkery
alkaline solutions. Although many zeolites and zeotypes and Ninham [277] and the Leuven group [67–73]. A rep-
can be prepared by either synthetic pathway, in some resentative schematic illustration of one proposed aggre-
cases the fluoride method provides the only route. There gation model is given in Fig. 18.
is a tendency for such products to contain some struc- Much of the evidence for growth by aggregation in
tural fluoride and this may perhaps account for the the studies outlined above is based upon the detection
importance of its presence in the reaction mixture. A of variously sized particle populations at different stages
further advantage of the fluoride route is its tendency of the synthesis process, although one group has also
to furnish large (in zeolite terms) crystals of the product. made AFM measurements of interactive forces [276].
This suggests that the mineralising power of F is less Extensive theoretical analyses have also been under-
than that of OH so that solubilities and effective super- taken by three different research groups but with dis-
saturations are lower, a point also indicated by the agreement in their conclusions. DLVO calculations
higher temperature and longer reaction times usually were carried out to ascertain whether or not particles
necessary for fluoride-based syntheses. Useful parallels migrating to a crystal surface have sufficient energy to
between the hydroxide and fluoride routes have been overcome the repulsive energy barrier at the liquid–solid
set out by Guth et al. [261], who have also provided interface and attach themselves. Schoeman [278] deter-
some insight into the synthesis chemistry. Using an elec- mined that the thermal energy of the colloidal particles
trode specific to F, they concluded that fluoride syn- at crystallisation temperature (373 K) was not sufficient
thesis media probably contained the species Si(OH)3F to overcome the net repulsive energy between the nega-
and Al(OH)3F, with octahedral coordination (SiF2 6 , tively charged particles. Since similar results were found

AlF36 ) becoming more prominent as F concentration for particles of both comparable (3 nm) and very differ-
increased [263]. More detailed investigations, particu- ent (3 nm, >10 nm) sizes, it was concluded that the
larly using multinuclear NMR spectroscopy, have been aggregation model was not a reasonable description of
carried out on the syntheses of zeotype materials (Sec- molecular sieve growth. Kirschhock et al. [71] calculated
tion 18.2). that the energy barriers were net repulsive at room tem-
The only other class of mineralising agents to have re- perature but furnished a net attractive secondary energy
ceived any significant attention are chelating oxygen li- minimum at 7 Å from the surface at 373 K. In the
gands such as o-dihydroxy aromatic compounds and work of Nikolakis and co-workers [276], the potential
glycols. These are well-known complexing agents for sil- energy was net repulsive but a fraction of the particles
icon [172,266,267]. Pyrocatechol (benzene-1,2-diol) has were estimated to have sufficient diffusional energy to
been used to control the crystal size in silicalite synthe- overcome the energy barrier and subsequently attach
sis, the presence of a silicon–benzene-1,2-diol complex rapidly to the crystal surface.
being detected by 29Si-NMR spectroscopy [268]. The Most recently, Erdem-Sß enatalar and Thompson [279]
same pattern was found in the silica-sodalite [269] sys- have used the Matijević model for the aggregation of
tem [270], although in this case the ethylene glycol sol- colloidal particles [280,281] to assess the likelihood of
vent alone is also known to form 5-coordinate silicalite growth by ‘‘nanoslab’’ addition in clear solu-
silicoglycollate complexes in the reaction solution [271]. tion synthesis. It was found that the original, diffusion-
limited model predicted a very much faster process than
9.4. Growth from particles that observed experimentally. Modifications to account
for surface-reaction growth limitations alone also failed
9.4.1. Particle aggregation to simulate the system, showing slowed growth as the
Several authors have postulated mechanisms for zeo- nanoslabs were depleted. A further modification to al-
lite nucleation and growth which involve the aggrega- low for a distribution of nanoslab structures gave some
tion of particles. Much of this speculation arises from improvement but did not reproduce the observed sud-
recent work on ‘‘clear solution’’ syntheses in the TPA- den termination of crystal growth. All simulations pre-
dicted a broadening of the particle size distribution over

an extended period, a feature which is not observed
experimentally. The authors concluded that MFI
growth by nanoslabs could not be ruled out, although
none of the simulations considered were able to capture
all the features observed by various research groups.
9.4.2. Chemical and physical consequences of an

aggregation mechanism
No detailed explanations have been provided as to
how a ‘‘growth by aggregation’’ model operates in
chemical terms. As regards the particulate behaviour
of the reaction sol, a degree of agglomeration of sub-
colloidal or colloidal amorphous matter seems very
probable. However, the aggregation of amorphous
with crystalline particles [276] as a growth mechanism,
or the agglomeration of non-viable nuclei with other
growing nuclei [222], are more problematic. If some
sort of direct amalgamation is implied, there has to
be a mechanism for the ordered formation of the nec-
essary chemical bonds, or the resulting product would
contain amorphous areas or extensive defects. There
are at least two possibilities for the case in which an
amorphous particle adheres to one that is partly or
wholly crystalline. Firstly, the aggregate may unify
and become more crystalline overall by the usual equil-
ibration mechanism of TAOAT bond making and
Fig. 18. An aggregation model for the formation of silicalite from a
breaking (Section 7.8). Alternatively, the aggregate
clear sol: (a) silicate/TPA clusters in solution, (b) primary fractal
may behave as a nutrient-recipient pair, with the amor- aggregates formed from the silicate/TPA clusters, (c) densification of
phous particle depolymerising to primary growth units the primary fractal aggregates, (d) combination of the densified
which then feed growth sites on the adjacent (or con- aggregates into a secondary fractal structure and crystallisation, (e)
joined) crystallite. The latter situation may form part densification of the secondary aggregates and crystal growth [37,273].
(Redrawn from Ref. [273] with permission.)
of the normal route for mass transfer in some systems
and would be consistent with Thompsons ‘‘tugging
chain’’ mechanism [37] (see Section 9.2). It will be
appreciated that these cases share many features with product was believed to have grown from an agglomer-
the generalised picture of nucleation and growth illus- ation of 5–10 nm particles (partially ordered TPA+–sil-
trated earlier in Fig. 16. ica–water ‘‘cubosomes’’) that crystallised to become
Direct experimental evidence relating to the signifi- syntaxial only after aggregation. There were no signs
cance of aggregation processes comes largely from of defects. Some of the 50 nm particles were fused to-
HRTEM studies. Thus, a weakness of an aggregation gether in open aggregates of 50–100 units with no pre-
growth mechanism involving additive entities of ferred orientation. The fused regions showed a poorly
5 nm in size has been highlighted by Davis [282]. This diffracting glassy structure a few tens of Ångstroms
is its inability to account for the existence of ordered thick. It thus seems probable that many of the possibil-
crystalline intergrowths—a fact which is readily ex- ities outlined above do occur in practice, namely (a)
plained by a normal layer-by-layer mechanism of crystal amalgamation and subsequent crystallisation of partly
growth (Section 8). A high-resolution TEM study on ordered particles, (b) grain boundary repair to give over-
nanoclusters of zeolite L [283] has verified that each all crystallographic alignment, and (c) some random fu-
crystallite has a single nucleation centre as its origin. sion of crystallites, with glassy boundaries.
Larger crystals possess grain boundaries, each one a Other experimental observations which add further
new domain resulting from an isolated nucleation event to the understanding of the processes taking place are
and not via the agglomeration of previously discrete col- found in synthetic studies. In the preparation of colloi-
loidal crystallites. A similar study on colloidal silicalite dal MFI-type materials, a very noticeable transition in
[277] was interpreted rather differently. In this case, morphology occurs as the synthesis temperature is
the particles were 50 nm in size but composed of 5– raised [150,284]. At temperatures of around 90 C,
10 nm single-domain sub-particles. The 50 nm silicalite the typical morphology is that of poorly defined
spheroids which, although seeming to behave as single the same general principles should apply to every
entities (e.g. in the size analysis of sols), appear at high system.
resolution to consist of aggregates of smaller particles At early synthesis times in colloidal systems, discrete
(Fig. 19a). Materials made by others are very similar nanoparticles may agglomerate and then grow to-
[134,181,216,221,222,277,285–287]. However, as the gether, in some cases leaving detectable grain bound-
synthesis temperature is raised, the morphology of aries in the final crystalline particle. However, a
the products becomes better defined. At 150–175 C, crystal growth mechanism involving growth by addi-
the nanoparticles are clearly discreet crystals with tion of amorphous or semi-ordered colloidal particles
well-developed facets (Fig. 19b). As far as can be ascer- is not, by itself, realistic: a depolymerisation or in-situ
tained from limited data at intermediate temperatures, reconstruction step involving mobile growth units is
there is no abrupt discontinuity in behaviour. Thus, it also necessary for the generation of an ordered periodic
seems likely that the observed pattern is the result of structure. The alternative growth mechanisms (directly
two competing processes: (a) an aggregation of small from solution or through aggregation and subsequent
and mainly amorphous particles into larger units, and ordering) are linked by the common chemistry of the
(b) the sintering together of agglomerated particles into reversible TAOAT bond-making and -breaking equili-
a macrocrystal, which is all the time becoming more bration reactions. Investigation of the nature of the
crystalline and also growing larger. Process (a) is prob- species which carry mass to the growing crystal is a
ably governed by the aggregation processes of colloid challenging problem, which may in due course be re-
chemistry and is driven by interparticle forces rather solved by (for example) a combination of AFM and
than the formation of chemical bonds. Process (b) re- Raman microscopies.
lies on the ubiquitous reversible mechanism of TAOAT
bond formation discussed earlier and will include the
acquisition of further growth units from solution.
The aggregation processes will occur at low tempera- 10. Solid state transformations
ture and may even become less effective as the temper-
ature is raised. Conversely, the true crystallisation In earlier parts of this work, zeolite synthesis has
reaction will be strongly encouraged by increasing been described in terms of a solution-mediated reac-
temperature. tion-crystallisation process, as shown in Fig. 2. Amor-
The temperature-dependent morphology observed in phous precursor material is dissolved and the zeolite
the above MFI syntheses is clearly closely related to product crystallises from the resulting solution. This
the HRTEM results mentioned earlier. A 2-component model has been directly demonstrated by the Kerr
growth mechanism was discussed by Schoeman in his experiment [55,56] (Section 3.3) in which the dissolu-
detailed study of aggregation and growth processes tion, solution transport and crystallisation stages are
[278]. Both a quantitative treatment of colloid stability physically separated. It provides the most generally sat-
and also extensive experimental results on the synthesis isfactory mechanism by which the experimental obser-
and properties of colloidal zeolites were considered. It vations of zeolite nucleation and crystal growth can
was concluded that a reasonable postulate for the be explained. However, there are circumstances in
nucleation of silicalite is via a cluster-cluster aggrega- which this model appears unrepresentative. For exam-
tion mechanism until a certain size (in this case possi- ple, zeolites can be synthesised by exposing a suitable
bly less than 1 nm) is attained, whereafter the growth solid precursor mixture to a vapour source (steam or
mechanism is best described as being via the addition amine) at elevated temperature (see Section 10.3). In
of low molecular weight species, most likely the this technique, a solid precursor is converted to zeolite
monomer. product in the absence of a bulk liquid phase. The exis-
tence of this and other, less extreme, cases of appar-
9.5. Summary ently direct conversion of amorphous components
into crystalline products has led to the postulation of
The most generally applicable crystal growth model an alternative ‘‘solid state transformation’’ mechanism
for zeolite systems is that based on mass gain from sim- for zeolite synthesis.
ple species (predominantly monomers) in solution. Fol- The implication is of an internal, bond-switching
lowing nucleation on or near the surface of an rearrangement from amorphous to crystalline material,
amorphous particle, the crystalline region is extended although the nature of this is frequently ill-defined and
by the acquisition of growth units from solution. These has never been the subject of a detailed, chemically
are replenished by the adjustment of solution equilibria specific description or illustration. In Sections 10.2–
and the dissolution of amorphous (or less ordered) 10.6, the justification for such a postulate is investi-
material. The identity of the growth species will depend gated. A useful reference document for the general
on the reaction chemistry (e.g. OH- or F-based) but topic of heterogeneous events in zeolite crystallisation
10.1. Hydrothermal synthesis in the presence of a liquid

phase
The original suggestion by Flanigen and Breck [51]

of a predominant role for the solid phase in a normal,
liquid slurry zeolite synthesis has been examined else-
where [31] and has been shown to reflect a participa-
tive, but not exclusive, function for the solid phase
(Section 3.2). The results of McNicol et al. [75,76]
were taken to be indicative of a zeolite crystallisation
in the solid gel phase on the basis that changes could
be seen in the isolated and dried solid phase but not
in the separated liquid. However, as is clear from
the subsequent and more detailed study of Angell
and Flank [77], there is no conflict between the
observations of McNicol et al. and either the views
of Flanigen and Breck or the reaction model given
in Part II, so that either scheme could account for
them.
More recently, other groups of workers have
proposed significant roles for solid-state transforma-
tions or gel rearrangements in zeolite synthesis [58–
62,288,289]. Derouane et al. base their argument for
the existence of identifiably distinct ‘‘liquid phase ion
transport’’ and ‘‘solid hydrogel transformation’’ pro-
cesses on a wide variety of measurements (Section 3.6)
but principally upon chemical and thermal analyses
and SEM results [58–62]. However, they are comparing
two (and only two) syntheses from different starting
materials and with widely divergent compositions: sub-
stantial dissimilarities are not unexpected and can be
rationalised in terms of differences in the partition of
reaction components between solution, macroscopic
gel and colloid phases (Section 7.7). The conclusions
of Gittleman and co-workers for silicalite synthesis
[288,289] were similarly based on extensive spectro-
scopic and chemical studies, affording a picture in which
the condensing silica network encapsulated hydrophobic
TPA cations in cages having only short range order and
inaccessible to ion exchange. Upon heating, the cages
were believed to rearrange through the breaking and ref-
ormation of siloxane bonds into the more stable silicalite
structure. However, the presence of ion exchange selec-
tivity in amorphous zeolite precursors has been observed
previously [122,290] and does not necessarily lead to the
conclusion of a ‘‘hydrogel-solid transformation mecha-
nism’’. Quaternary ammonium templates can certainly
be strongly bound in amorphous precursor complexes
Fig. 19. Morphological differences seen in particles of colloidal [117,291] and this is compatible with a subsequent solu-
silicalite synthesised using conventional heating at (a) 90 C (scale
tion-mediated growth mechanism [64–66]. In the latest
bar 50 nm) and (b) 150 C (scale bar 100 nm). The same batch of
reaction mixture (age 5 days) was used for both experiments. [292] of a series [39,293] of studies in which solid–solid
transformations are postulated to play a prominent role,
Serrano and associates conclude that the synthesis of
TS-2 from clear liquid gels occurs by a dual mechanism
has recently been provided by Serrano and van Grie- (cf. Section 9.4.2). The initial stage operates by a reorga-
ken [39]. nisation-agglomeration mechanism and only when a
crystallinity of around 80% is reached do soluble species of water vapour only [297] since they lie within the com-
contribute. positional range in which ZSM-5 can be formed in the
Whilst the interpretation of these and similar experi- absence of organic additives. The crystallisation to
ments is subject to differences of opinion, more initially ZSM-5 of TPA-containing pelleted gels in a water va-
misleading are the cases in which the entire synthesis pour atmosphere was also reported by Crea and co-
reaction appears to take place in the solid phase. workers [298]. It should also be pointed out that some
of the reaction compositions used by Mostowicz and
10.2. Hydrothermal synthesis in the apparent absence Sand for pentasil synthesis were reported as having the
of a liquid phase consistency of dry flowing powders [207], whilst a simi-
lar ‘‘dry’’ powder was observed to convert to zeolite X
In 1990, Xu et al. reported the first examples of a zeo- over a period of weeks at ambient temperature [299].
lite synthesis in which some of the reaction components Thus, the phenomenon of zeolite crystallisation under
were supplied from the vapour phase [294]. In these hydrothermal conditions but in the absence of a visible
cases, a pre-prepared, damp or dried sodium aluminosil- liquid phase is well established.
icate gel was suspended above liquid in an autoclave and The apparent lack of a liquid phase does not mean
subjected to the mixed vapour of ethylenediamine, tri- that there is no water (or other polar solvent) present.
ethylamine and water at elevated temperature and pres- The ‘‘dry’’ powder reaction mixtures mentioned above
sure (Fig. 20). Under these conditions, ZSM-5 [207,299] contained 20–44% water by weight. It is appar-
crystallised in 5–7 days at 180–200 C. Not surprisingly, ent that such mixtures contain sorbed and surface-
it was stated that ‘‘the crystallisation process of ZSM-5 bonded water so that they present a ‘‘solution-like’’
belongs to the solid-phase transformation process’’. phase in which the same chemistry can occur as that
Subsequent work by several groups has provided many in more open solution. Consequently, there is no reason
more examples of this type of process, which is described to view these reactions as fundamentally altered from
variously as the dry gel conversion (DGC) or vapour those having a more accessible liquid phase. However,
phase transport (VPT) technique and has recently been there may be differences due to mass transport limita-
reviewed [295]. A useful distinction can be made be- tions, high concentration gradients and the very high
tween the amine vapour procedure of the original report surface to volume ratio of the mobile phase. Such dis-
by Xu et al. [294] and methods in which only steam is similarities may manifest themselves in changes of rate,
supplied from the gas phase. The latter method is desig- crystal morphology and state of aggregation [300,301],
nated ‘‘steam-assisted conversion’’ (SAC) by Matsukata consequent changes in the pattern of intermediates or
and co-workers [295]. metastable phases [302,303], or differences in the course
As expected [154,296], sodium aluminosilicate gels of reactions heated by microwave and thermal methods
having compositions near to those used by Xu et al. [304,305]. Very useful synthetic procedures may result.
[294] will crystallise to MFI products in the presence For example, the ‘‘dry’’ route to zeolite Beta leads to
the more siliceous compositions which are not obtain-
able by the more traditional wet gel route from similar
starting materials [295,306]. Again, the synthesis of zeo-
litic mesoporous materials can be accomplished by the
controlled steaming of dried precursor gels [307].
Two recent studies of solid-to-solid conversions merit
special mention in view of the detailed and informative
nature of the accompanying characterisation work. Ar-
nold and co-workers [308] investigated the dry-gel syn-
thesis of gallium-rich zeolite [Ga]Beta using SEM,
XRD and multi-nuclear MAS NMR spectroscopy. They
found significant changes in particle size and morphol-
ogy showing that the synthetic method did not operate
solely by breakage and reformation of chemical bonds
within the dry-gel particles but also caused a redistribu-
tion of the solid particles, i.e. a ‘‘recrystallisation’’. The
XRD and 71Ga NMR data proved that the major part
Fig. 20. The vapour phase transport method for zeolite synthesis, as of the crystallisation reaction (including gallium frame-
originally demonstrated for ZSM-5 by Xu et al. [294]. Inside the special
work incorporation) occurred over a relatively short
autoclave, the amorphous gel (b) rests on a porous sieve plate (c) in the
reactant container (a), which itself is supported by a stainless steel period (16 h), whilst other NMR measurements indi-
stand (d) above the bulk solution phase (e). (Reproduced from Ref. cated that local structural rearrangement (e.g. silanol
[294] by permission of The Royal Society of Chemistry.) group condensation) continued over a longer timescale
(65 h). In the second case, a careful study of the conver- tem based on tetraethylene glycol, varying amounts of
sion of an alumina-pillared montmorillonite into soda- water (from a trace up to six times the molar quantity
lite was undertaken by Choy, Lee and colleagues of the glycol) produced a considerable range of alumino-
[309,310]. The pillared clay was mixed with 25 mass% phosphate products [315], giving rise to considerable
water and powdered NaOH (OH/Si = 10), and the speculation about their derivation from soluble chain
resulting solid gel mixture heated at 80 C under an precursors through a hydrolysis-condensation self-
ambient atmosphere. Samples were removed periodi- assembly pathway [315,316].
cally and analysed by XRD, FTIR, MAS NMR, SEM
and HRTEM. The XRD and spectroscopic data show 10.4. The role of water in apparent solid state
a smooth transition from dominant pillared clay to transformations
dominant sodalite, whilst the SEM images show a major
change in particle size and morphology. The HRTEM Matsukata and colleagues have studied the role of
images initially display the layered clay diffraction lines. water in the steam-assisted conversion (SAC) synthesis
After 30 min, areas of sodalite lattice fringes appear ran- of zeolite Beta materials [295], concluding that keeping
domly throughout the original pattern. Eventually the water vapour saturated is essential for zeolite-form-
(150 min), a long range sodalite fringe pattern becomes ing reactions. A related study by Serrano and co-work-
dominant. The authors conclude that ‘‘the sodalite crys- ers on the crystallisation of TS-1 and TS-2 from
tallisation takes place by the rearrangement of deloca- wetness-impregnated xerogels highlights a ‘‘contribution
lised sodalite nuclei formed through the solid–solid of solid–solid transformations’’ [317]. However, it is
transformation’’. It is also noted that ‘‘in contrast to clearly stated that ‘‘the process cannot be viewed as a
the conventional solution-mediated transport processes pure solid–solid mechanism since the liquid phase plays
for crystal growth, the absence of a distinct X-ray amor- also several roles’’ and ‘‘the origin of this anomalous
phous stage implies that the transformation does not mechanism is probably related to the high concentration
take place through long-range diffusion of dissolved alu- of solids present compared to most conventional zeolite
minium and silicon species but through a short-range syntheses’’. Similarly, the conversion of kanemite to sil-
molecular-level rearrangement.’’ However, neither the icalite in sealed glass ampoules by ‘‘solid state transfor-
role of water nor the requirement for NaOH is men- mation’’ [318] showed many similarities to a standard
tioned in the discussion. hydrothermal route (although the ‘‘solid state’’ route
Unlike the above, vapour-assisted syntheses in which was faster). However, the reaction did not occur in an
the reaction components are in reality hydrated, or open vessel or if the precursor had been dried at
otherwise solvated, the reaction system described by Alt- P100 C, showing that ‘‘adsorbed water plays an
hoff and co-workers [311] appears to be genuinely important role during the transformation’’.
‘‘dry’’. ZSM-5 is synthesised at 180 C from amorphous, Outside the realm of zeolite synthesis, a limiting case
SiO2 Æ Al2O3 precursors (dried at 650 C) in the presence of apparent solid state conversion can be found in the
of dried NH4F and TPABr. The reaction is believed to ammonium nitrate (IV) ! (III) phase transformation.
involve a vapour phase mass transfer process with Amongst the five polymorphs of ammonium nitrate,
SiF4 as the mobile species. the transition between Phases IV and III which occurs
near 32 C is of particular importance because it lies
10.3. Non-aqueous syntheses within the range of likely storage temperatures and is
a significant factor in the caking (and therefore safe stor-
Zeolite synthesis reactions carried out in non-aque- age) of the product. Studies of the mechanism of this
ous solvents (e.g. organic compounds, liquid ammonia) phase transition by Davey and co-workers [319] are rel-
[312] are in many ways analogous to the ‘‘dry’’ syntheses evant to the discussion of ‘‘dry’’ zeolite crystallisation
described above. It is often difficult to assess from the given above. The ammonium nitrate (IV) ! (III) transi-
published information whether the reaction in question tion does not occur in the absence of water under the
is brought about by the presence (perhaps accidental) experimental conditions employed. However, on wetting
of small quantities of water, or whether the non-aqueous a Phase (IV) crystal with a saturated aqueous solution, a
solvent is playing a positive role rather than acting only dark line was seen to move through the crystal as it
as a diluent or heat transfer medium. Clearly, the more transformed to the Phase (III) structure. Further studies
polar the solvent, the more the likelihood of genuine [320] revealed the situation illustrated in Fig. 21, where
participation. The reactions reported by Kuperman it is shown that the phase transition occurs in a micro-
et al. [313,314] involving the use of the pyridine-HF scopic layer of ammonium nitrate solution which travels
complex in the presence of (at most) stoichiometric as an interfacial zone along the transforming crystal.
quantities of water are probably in the participative cat- Thus, the transformation is a solvent-mediated re-crys-
egory. Very large (mm-size) zeolite crystals were formed tallisation in which a minute amount of water catalyses
in this system. Interestingly, in an AlPO4-n synthesis sys- the transition between the two phases. In zeolite
do not transform under comparable conditions as the

required pathway is blocked by the manner of K-coordi-
nation to the 8-rings.
A B C
10.6. Summary and conclusions

Fig. 21. One possible model for a localised phase transformation by a
solution-mediated mechanism. Initial phase A dissolves into solution B As evidenced by the examples given above, there is a
and crystallises as less soluble phase C, so that the boundary film tendency for reactions to proceed by a solvothermal
migrates across the reaction volume (which may not remain constant). mechanism, if a pathway exists, rather than by a true so-
In the case of ammonium nitrate [320], A = Phase IV NH4NO3, lid state route, presumably because the activation energy
B = NH4NO3 solution, C = Phase III NH4NO3 and the transforma-
will usually be lower in the former case. Even at fairly
tion occurs within a single crystal. In a zeolite system, A could be
amorphous precursor, B the reaction solution and C crystalline zeolite. high temperatures, there are relatively few zeolite trans-
formations which can be unequivocally classified as so-
lid–solid or gas-solid reaction systems. In so many
synthesis, a similar mechanism could facilitate an amor- cases, there is the possibility of participation by a liquid
phous-to-crystalline conversion, or even (see Section phase, if only as an adsorbed layer. For solvothermal
10.5) the analogous transition from one crystalline zeo- zeolite synthesis at normal temperatures (up to, say,
lite phase to another. 230 C and certainly below 300 C), reaction mixtures
containing no water or other polar solvent as a liquid
10.5. Solid state transformations at high temperatures phase, or at least as an adsorbed, liquid-like phase, do
and pressures not undergo the dissolution–re-crystallisation reactions
normally essential to zeolite formation. However, such
Although not normally considered in terms of synthe- reactions do occur in mixed liquid–solid phase media
sis, zeolites undergo many solid state transformations. where most of the observable changes take place in the
Most of these relate to decomposition reactions, or, at ‘‘solid’’ phase. In this case, ‘‘solid’’ may refer to filter-
lower temperatures, changes in symmetry rather than able or centrifugable intimate mixtures of amorphous
overall structure (e.g. for ZSM-5 [321,322], ZSM-39 and crystalline material. In other instances, the solid
[322] and AlPO4-5 [323]). However, in some cases a particles may be of colloidal dimensions. For ‘‘dry’’ syn-
more complete metamorphosis can take place, such as theses, there is no apparent or separable liquid phase at
the transformation of Ba-A to celsian at P600 C all. However, if all traces of bulk and adsorbed polar
[324,325]. There are also several examples of the high- solvent are removed, transformation to zeolite does
temperature (500 C) conversion of layer phases to not occur. It is therefore suggested that any arbitrary
zeolites, although in all cases this change occurs only separation of hydrothermal synthesis reactions carried
upon removal of the organic template by calcination out under these conditions into ‘‘solution-mediated’’
(see Section 16.2). and so-called ‘‘solid-state transformation’’ is not only
TPA-containing ZSM-5 has been reported to be con- unnecessary but is actually misleading since it implies
verted into ZSM-11 through the action of pressure alone a fundamental difference in mechanism where there is
(4 GPa/15 min) [326], which seems surprising in view of no convincing evidence that any exists.
the larger unit cell volume (5563 Å3) and lower frame- Perhaps the key point in these considerations is the
work density (17.6 T/1000 Å3) of the MEL structure issue of mobility. In a given circumstance, a potential
over that of the MFI type (5187 Å3, 17.9 T/1000 Å3) growth unit may detach (dissolve) from an amorphous
[16]. In other cases, the application of very high pres- particle and migrate through the solution to a growth
sures renders zeolites amorphous (see Section 12.6). So- site on a remote developing crystal. At the other ex-
dium forms of CHA- and EAB-type zeolites were found treme, the growth unit may never escape into solution
to transform topotactically to pseudomorphic SOD-type but may be transferred in a concerted process from an
products above 600 C in dry nitrogen or at lower tem- amorphous region to an immediately adjacent centre
peratures (130–550 C) in the presence of moisture [327– of crystal growth (an idea implied, although not specif-
329]. The mechanism for these ‘‘isomerisations’’ involves ically illustrated, in Fig. 16). In this view, the ‘‘solid
cleavage of selected TAOAT bridges followed by bond hydrogel phase transformation’’ [58] is seen as a limiting
rotation to give TO4 inversion. In a way reminiscent case of the more general solution-mediated process, the
of the ammonium nitrate transformation (Section main difference being in the distance travelled by the
10.4), there is a catalytic role for water, which in the mass-carrying ions. Once again, the underlying mecha-
present case activates the critical TAOAT bridges to- nism is that of the OH-catalysed TAOAT bond-mak-
wards rupture, presumably by acting as a nucleophile. ing and bond-breaking process (Section 7.8). Whilst
Significantly, the potassium versions of these materials there is sufficient ionic mobility for this equilibration
process to continue, the mechanism of zeolite formation crystallisation in order to encourage the formation of
will remain basically the same. the faujasite phase and avoid contamination by
Although the above discussion has concentrated impurities such as zeolites P (gismondine-type) or S
upon the amorphous ! crystalline transformation, (gmelinite-type). Sometimes, the synthesis composition
similar considerations will apply to the situation in is inoculated with an aged nucleating sol or slurry,
which the precursor is also crystalline. The apparent rather than ageing the entire reaction mixture
crystal–crystal reconstruction demonstrated by Choy [332,333]. Similar procedures have been described for
and co-workers [309,310] (Section 10.2) is no different other zeolites, for example zeolite L [334], zeolite omega
in principle to the amorphous–crystal conversion within [335] and ZSM-5 [336]. This behaviour is not confined to
gel particles reported (with similar HRTEM evidence) systems containing a visible gel phase. In a study of
by Mintova et al. [134,138,139] (Section 7.7). Both sets analcime synthesis, Brock et al. showed [337] that bimo-
of processes would stop if the alkalinity and/or all the dal crystal size distributions could result from ageing the
water were removed from the system. In this respect, it clear reaction solutions at room temperature. For clear-
is worth recalling the demonstration of the hydrother- solution syntheses of zeolite A, Gora and Thompson
mal lability of the zeolite crystal lattice by von Ballmoos found [338] that additions of aged reaction solution to
and Meier [330]. Isotopic 18O exchange was employed to the fresh material resulted in the formation of a popula-
show that, even at the relatively low temperature of tion of crystals which in some cases was clearly distin-
95 C, both hydroxyl groups and TAOAT bridges are guishable from the population nucleated by the fresh
cleaved in reversible equilibration reactions between li- solution alone.
quid water and NH4-ZSM-5. Although the nucleating mixture is usually found to
It is suggested that such terms as solid-state transfor- be X-ray amorphous, its seeding potential is time-depen-
mation should be restricted (in the usage of zeolite syn- dent, suggesting the development of structured compo-
thesis) to conversions involving only solid or solid and nents. The changes in chemical composition which
gas phases, the participation of solvates and ‘‘solids’’ occur during the ageing of faujasite hydrogel precursors
having sorbed solvent being excluded. Most other reac- were the subject of early work by Polak and Cichocki
tion systems may then be regarded as, in principle, solu- [339]. The product-promoting effects of ageing were
tion-mediated, although there may be major differences attributed to the action of polycondensation reactions
in appearance and in the partition of components bein the formation of groups facilitating crystallisation.
tween solid and liquid phases. It may be helpful and A similar but more recent investigation by Ginter
convenient for some purposes to classify reactions into et al. has taken advantage of the wide range of analytical
types—such as the ‘‘Type A’’ and ‘‘Type B’’ syntheses techniques now available to provide detailed physico-
described by Derouane and associates [58–62] and dis- chemical data on room-temperature ageing in the Na-
cussed earlier. Such classifications highlight divergences Y system [340,341].
in detail but should not necessarily be taken to signify The first studies to link these time-dependent effects
differences in reaction mechanism at a fundamental with control of crystal size were those of Freund [199]
level. and of Zhdanov and co-workers [140]. These investiga-
tions showed that room temperature ageing of alumino-
silicate Na-A and Na-X gels resulted in both an
11. Ageing effects in zeolite synthesis acceleration of the crystallisation and a decrease in the
crystal size of the product, thus linking the ageing pro-
Ageing effects have been mentioned earlier in refer- cess directly to the number of nuclei formed. A more re-
ence to the investigation of nucleation processes (Section cent example for zeolite A concerned the effect of ageing
7.3.2). In the present section, this aspect is explored in on microwave-mediated synthesis [342]. It was found
greater depth, together with its relationship to the use that products from short ageing times (5 min) were
of the ageing technique in controlling zeolite product amorphous after 5 min microwave heating at 100 C,
properties. whereas zeolite A could be crystallised in 1 min at
120 C after overnight ageing of the synthesis mixture.
11.1. Ageing as a means to control product phase purity Product crystal size decreased with ageing time, again
and crystal size suggesting that, during the ageing process, structured
elements related to the nucleation process are being
It has been known for many years that the crystallisa- formed.
tion behaviour of a synthesis reaction mixture is often Ageing of reaction mixtures is (mainly for conve-
dependent upon the time–temperature history of the nience) most frequently carried out at room tempera-
sample in advance of the principal heating step. Brecks ture. However, ageing has sometimes been carried out
original synthesis of zeolite Y from colloidal silica at more elevated temperatures, in some cases at levels
[32,331] requires an ageing step prior to hydrothermal higher than the main synthesis temperature. Thus, a
method has been described for preparing impurity-free assumed that every such nucleus grows into a crystal, we
high-silica zeolites in a 2-stage crystallisation process may take as an example the situation of 2 lm crystals
[343,344] where the first stage is conducted at 245– (density 2 g cm3) with a unit cell size of 20 · 20 ·
325 C for 1–20 min and the second stage at 50–225 C 10 Å growing from a 5 unit cell critical nucleus. In this
for 40 h or more. Dodwell and co-workers have used a case, the ratio of total mass of nuclear material to mass
2-stage heating regime to reduce the crystallite size for of product would be only 2 · 108, although represent-
ZSM-5 and EU-1 [345]. In the latter case, their synthesis ing 1011 nuclei per gram of nutrient. However, in cases
gave 30 lm crystals of EU-1 in a static system at a con- where the nucleation rate is extremely high, recent scat-
stant temperature of 200 C. However, carrying out the tering experiments have begun to reveal the evolution of
reaction in a stirred autoclave at 150 C for 24 h, fol- particulate precursors but, even within the MFI synthe-
lowed by a temperature increase to 200 C for 48 h, still sis system, there are considerable variations both in the
gave pure EU-1 but the crystal size was submicron. experimental scattering behaviour and in its interpreta-
Reduction of crystallite size is also one aspect of a dis- tion [131,181,275,350–352]. In turn, there is essentially
closure on catalysts with improved performance (activ- no correlation between these results and data from zeo-
ity and lifetime) by the Mobil company, which also lite A synthesis [224]. This suggests that the observations
concerns the 2-stage synthesis of zeolites [346]. In this are more closely connected with differences in colloid
case, it is specifically stated to comprise a lower-temper- and gel chemistry than to the actual process of zeolite
ature nucleation stage and a higher temperature crystal- formation, i.e. that they relate to the particles and their
lisation stage. This procedure is related to that of the aggregation rather than the chemistry going on within
‘‘precursor heating method’’ discussed for ZSM-34 by them to assemble the zeolite lattice.
Inui et al. [347]. In this latter case, the reaction mixture Modelling studies have allowed the simulation of gel
is maintained at 100 C for three days before being ageing by population balance methods [353]. Assuming
raised at a constant rate to a higher temperature (130– that the ageing step results in the formation of viable nu-
220 C). The idea was later extended to the method of clei that effectively lie dormant until the temperature is
‘‘temperature programmed crystallisation’’ in which raised, it has been shown that the number of nuclei gen-
the temperature profile of the synthesis would be tai- erated during ageing strongly influences subsequent
lored to provide a more rapid crystallisation time and crystallisation behaviour, whilst their size is relatively
a product with more uniform properties [348]. A similar unimportant. Other types of kinetic analysis and simula-
concept, designated ‘‘temperature programmed synthe- tion, together with supporting experimental data, have
sis’’ was developed at about the same time by the present led to similar conclusions [160,165].
authors to control the crystal size of zeolitic sorbents
and catalysts [349]. 11.3. Detailed analyses of ageing-related effects in
silicalite synthesis
11.2. Rationalisation of ageing effects
Recent studies on the silicalite-1 system have pro-
As can be seen from most of the examples given vided more detailed insights into nucleation processes.
above, the dominant effect of ageing and related two- Čižmek et al. [354] examined the influence of time and
temperature treatments is the separation, or partial temperature of gel ageing on the particulate properties
separation, of the nucleation and growth stages of the of silicalite-1 synthesised at 170 C. At ambient temper-
synthesis. The nucleating mixtures contain structured ature, gels were aged from 0 to 192 h, whilst at ageing
elements (crystal proto-nuclei), whilst the batch-ageing temperatures of 20, 35, 50, 65 and 80 C the ageing per-
treatments encourage the self-generation of similar spe- iod was for 12 and 24 h. In no case could crystalline
cies in situ. The net effect in both cases is to introduce material be detected by XRD, FTIR or DTG before
into the reaction mixture entities which would not be the final heating step. The twinned-lozenge morphology
present (at least in such numbers) if the composition of the product was not affected by the ageing treatments
were merely made up and heated to the operating tem- but the number of crystals formed was found to increase
perature in the normal way. The modified reaction mix- linearly with ageing time at ambient temperature. This
ture will have a different nucleation profile from that of result was interpreted in terms of the formation of Burk-
the unmodified composition, resulting in different ki- ett–Davis type inorganic–organic composite species
netic behaviour and changes in product crystal size [65,66] and their aggregation to structural globular units
distribution. [129] equivalent to germ nuclei. At higher temperatures,
Attempts [339–341] to detect nucleating entities in crystal numbers were found to go through a maximum,
ageing reaction mixtures by chemical or spectroscopic which was taken as the point at which already-formed
means have usually yielded little direct information germ nuclei started to decompose.
since, although great in number, the ‘‘nuclei’’ may repre- In an interesting series of papers [287,355,356], Li et al.
sent only a very small fraction of the total mass. If it is have utilised a two-stage-varying-temperature synthesis
the ultimate crystal size decreased (Fig. 22). At the point

where the values were approximately equal to those ob-
tained for a synthesis run entirely at the initial tempera-
ture, it was concluded that the nucleation stage was
completed. In this way, nucleation and growth were sep-
arated, since essentially all of the nucleation occurred
during an induction period when little or no crystal
growth was observed. By changing the temperature and
duration of the initial stage, the nucleation period for col-
loidal TPA-silicalite-1 crystallisation was determined at
60, 80 and 100 C. If the temperature change was made
after the completion of the inferred nucleation period,
the higher synthesis temperature controlled only the lin-
ear growth rate and yield of the product crystals.
In an extension of this work, Li and co-workers have
investigated the additional effect of ambient temperature
ageing on their reaction mixtures before subjecting them
to the two-stage varying-temperature synthesis [356].
For the ageing periods examined (up to 30 days), only
a mild influence on nucleation and crystallisation kinet-
ics was found for reaction mixtures based on tetraethyl
orthosilicate (TEOS), but for a commercial colloidal sil-
ica sol (Ludox TM) the changes were more marked,
leading to reductions in product mean crystal size of
around 50% over the 30 day period. The final crystal size
for the Ludox system after 30 days ageing was similar to
that for the TEOS reactions after an ageing time of only
one day. These variations clearly reflect the differing
reactivity of the silica sources, the very labile TEOS sys-
tem having equilibrated to a much greater extent before
the thermal treatment. As the authors point out, two
processes could be occurring during ageing: (i) depoly-
merisation of colloidal silica and (ii) formation of nuclei.
For TEOS, the first process was not significant under the
conditions employed.
Cundy et al. investigated the effect of room-tempera-
ture ageing on colloidal silicalite synthesis compositions
[150,357]. Sol precursors were aged for periods of up to
several years and then crystallised at temperatures be-
tween 90 and 175 C using both thermal and microwave
heating. Analysis of the product particle sizes as a func-
tion of the different ageing times and heating rates al-
lowed the nucleation behaviour of the system to be
Fig. 22. The 2-stage varying-temperature procedure of Li et al. for interpreted. The largest particle sizes were seen at the
colloidal silicalite synthesis [287]: (a) comparison of 1-stage crystalli-
shortest ageing times and fastest heating rates. As either
sations at 60 and 100 C with 2-stage crystallisations starting at 60 C
and ending at 100 C; effect of the duration of the first stage at 60 C the ageing time increased, or the heating ramp rate and
on (b) the ultimate crystal size and (c) the ultimate crystal concentra- final temperature were reduced, the particle size de-
tion. (Reprinted from Ref. [287], copyright 1999, with permission from creased. Sufficient data points were collected to define
Elsevier.) the overall form of the crystal numbers curve for the first
time (Fig. 23). The curve is seen initially to be linear (as
procedure to investigate the nucleation period in colloi- found by Čižmek et al. [354]) but then to level off. At this
dal TPA-silicalite synthesis. The basic method involves point (200 days ageing), no crystalline material could
a rapid increase in temperature at some point during be detected by conventional XRD methods in samples
the course of an otherwise conventional crystallisation. isolated from the unheated sol.
As the duration of the lower-temperature period was ex- The first level of interpretation of the above
tended, the crystal number concentration increased and result considers the curve to chart the development of
200 350
180
300
(# (g crystal mass)-1 x 1014)

160
Crystal size (nm)

250
Crystal numbers
140
120
200
100
150
80
60 100
40
50
20
0 0
0 100 200 300 400 500 600 700 800
Sol age (days)
Fig. 23. Colloidal silicalite synthesis: the effect of ageing the precursor sol at 22 C upon subsequent syntheses at 90–175 C. With increasing age of
the precursor sol, the final crystal size (h) decreases and crystal numbers (s) increase, eventually reaching constant values. Data from thermal and
microwave syntheses at 90, 100, 150, 175 C [150,357].
accumulating numbers of ‘‘proto-nuclei’’ (Section 7.8). temperatures and on a short timescale. Conversely, a
These lie dormant under the prevailing conditions, slow-growing phase such as Na,hexamethonium-EU-1
maturing into viable nuclei and initiating crystal growth will show ageing effects only if its precursor composition
only on heating to crystallisation temperature, or spon- is treated at relatively high temperatures for extended
taneously upon prolonged ageing under ambient condi- periods of time.
tions. A more mechanistically detailed interpretation Why does the crystal numbers curve of Fig. 23 reach
would equate the generation of proto-nuclei to the for- a plateau? One possible reason is that any further in-
mation of the ordered intermediates shown in Fig. 5, crease in numbers has been curtailed by the onset of
the number of proto-nuclei corresponding to the mean crystal growth and the consequent drop in supersatura-
concentration of such ‘‘islands of order’’ (and to their tion. If so, however, the failure to detect any crystalline
average degree of organisation). The conversion of pro- product by XRD is surprising. A further possibility is
to-nucleus to viable nucleus (nucleation) corresponds to that the equilibration process leading to the intermediate
the establishment of a sufficiently ordered and suffi- semi-ordered state has reached a thermodynamic
ciently extensive element of the zeolite structure by the pseudo-equilibrium, i.e. the concentration of proto-nu-
process shown in Fig. 11 to initiate systematic periodic clei has reached a steady state. Therefore, the resulting
propagation, i.e. crystal growth. In all cases, the changes nucleation on heating the aged reaction mixture after
are brought about by the constant action of the revers- these extended time periods is no longer time-dependent.
ible TAOAT bond-making and bond-breaking equili- An interesting effect of ageing on the apparent
bration reactions described in Sections 7.8. A similar growth rates of submicron zeolite crystals has been men-
interpretation could be applied to the results of the tioned earlier (Section 8.4). The observed growth rate
2-stage-varying-temperature procedure of Li and co- was found to increase with ageing time at ambient tem-
workers described above, their ‘‘nucleation stage’’ corre- perature for both silicalite [222] and zeolite Na-A [198].
sponding to the generation of proto-nuclear local order The cause of the apparent acceleration was ascribed to
in the present analysis. Conversion of proto-nuclei to the agglomeration of nuclei during ageing and the
developing nuclei (the true nucleation step) occurs resulting enhancement of mass gain through the higher
subsequently, either on prolonged ageing at the lower surface area of the polycrystalline particles.
temperature, or upon raising to the higher temperature
of the second stage. 11.4. Summary
It should be noted that there is a strong link between
the rate of crystal growth (as measured approximately Ageing phenomena in zeolite synthesis have long
by the development of X-ray crystallinity and more spe- been recognised. Early applications of ageing proce-
cifically by linear growth rate measurements) and the dures were aimed largely at directive effects for improv-
rates of the equilibration processes underlying ageing ing the reproducibility and economy of industrially
behaviour. This is to be expected because they are both important syntheses. However, following the establish-
controlled by the same basic chemistry. Therefore, the ment of the link between gel ageing and particle size,
reaction mixture of a rapidly growing zeolite such as ageing procedures have been used increasingly both
zeolite Na-A will be sensitive to ageing effects at low to control crystal size and as an experimental probe
to investigate indirectly the development of order in

zeolite synthesis compositions. In this way, they are a
valuable tool in studies on the mechanism of
nucleation.
12. X-ray amorphous zeolites
Earlier (Section 6), we have considered some of the

indications of pre-crystalline order in zeolite synthesis.
For example, there is evidence (Section 6.2) for the for-
mation of a secondary amorphous phase which has
many compositional and chemical similarities to the fi-
nal crystalline product but is lacking in long-range
order. The relationship between this semi-ordered pre-
cursor and the final zeolite product is therefore compa-
rable to that between a glass and a crystalline solid of
similar composition. This raises many interesting ques-
tions, which were highlighted in an important paper
by Jacobs, Derouane and Weitkamp, published in
1981 [132]. The authors presented evidence for the exis-
tence of entities which were amorphous to XRD but,
based on the appearance of a vibration at 550 cm1, ap-
peared to be fully ‘‘crystalline’’ by infrared spectroscopy
and showed catalytic activity similar to macrocrystalline
ZSM-5 in the hydroconversion of n-decane. These enti-
ties were described as ‘‘X-ray amorphous zeolites’’ and
their nature will now be considered in detail. Such an Fig. 24. Computed XRPD patterns for ultra-small ZSM-5 zeolite
examination is particularly relevant at the present time crystals [361]: (a) crystals with edges 10 unit cells in length, (b) crystals
in view of the increasing interest in the incorporation with edges 4 unit cells in length. In both diagrams, the lower pattern is
of elements of zeolite microstructure into mesoporous that calculated for crystals of 50 · 50 · 50 unit cells. (Redrawn from
Ref. [361] with permission.)
composites (see e.g. Refs. [358–360]).
12.1. XRD evidence for ‘‘X-ray amorphous zeolites’’

the quantity of such material present (of whatever crys-
The problems associated with XRD patterns of ultra- tal size) was below their limit of detection.
small zeolite crystals have been elegantly treated by
Schlenker and Peterson in a paper [361] which also dis- 12.2. IR evidence
cusses specifically the original report of ‘‘X-ray amor-
phous zeolites’’ [132]. These authors point out that the The observation of an IR band at 550 cm1 has
decisive factor is concentration and not size. Hence, all been used by Jacobs et al. [132] and other authors
materials will have a diffraction pattern of some sort [84,130,131,133,134] as an indicator of the presence of
but there is a concentration limit below which it cannot the ZSM-5 structure. Although it has been shown that
normally be observed. The expected patterns for crystals this is not an unambiguous test [135], it would seem to
of any size can be realistically simulated. Theoretical be a reasonable analytical probe in a system where
patterns for ZSM-5 crystals whose edges are of length ZSM-5 is known to be the eventual product. However,
from 50 to 4 units cells are given in Fig. 24. From these the data presented do give some cause for concern.
it can be appreciated that, whereas the pattern from a The variation of the 550 cm1 band with time does not
103 unit cell crystallite is still recognisable as ZSM-5, follow a regular pattern and its relation to conventional
that from a 43 unit would be effectively indistinguishable XRD crystallinity is not clear. This can be seen from the
from the amorphous pattern which would be obtained following summary, which is taken from the Figure in
from a dried but unaged and unheated reaction mixture. Ref. [132]: (a) no zero-time value is given, (b) the lowest
Therefore, the reason that Jacobs et al. did not observe time value (2 h) shows an IR crystallinity of 61%, (c) at
an XRD pattern for the material they studied could in- 10 h, the IR crystallinity was 66%, (d) at subsequent
deed be due to the very small size of the ZSM-5-related times (P2 days), the IR crystallinity was P94%,
component. However, there is also the possibility that although XRD crystallinity was detectable only at
P3 days. In contrast, an IR study of macrocrystal- ity with that of 6 lm ZSM-5 crystals, USY zeolite and
line Na,TPA-ZSM-5 crystallisation (SiO2/Al2O3 = 81) amorphous silica-alumina (Table in Ref. [132]). In partic-
showed a typical S-shaped growth curve which corre- ular, they highlight the percentage of 2-methylnonane
lated well with observations from XRD and electron formed, since this is a characteristic size- and shape-selec-
microscopy [362]. The data were obtained by plotting tive ZSM-5 product. From their results, the authors con-
the 550 cm1/450 cm1 IR band ratio against time for clude that ‘‘(i) the XRA [X-ray Amorphous]/ZSM-5
a large number of samples taken during crystallisation zeolites are fairly active materials in the present reaction;
at 100 C. In the present context it is of interest to note the activity is definitely not due to the amorphous matrix
that whilst the IR band ratio gave the usual [135] value in which the small crystals are embedded; (ii) the XRA/
of 0.73 for 100% crystallinity (140 h), the zero-time sam- ZSM-5 zeolites exhibit bifunctional characteristics very
ple (0% crystallinity, 0 h) showed a value of 0.26, i.e. similar to those of crystalline ZSM-5.’’ However, it is
there was appreciable IR intensity at 550 cm1 at the notable from the data presented that whilst the IR crys-
start of the experiment where it would be unreasonable tallinity increases from 61% to 97% (at zero XRD crystal-
to expect any crystalline material to be present. linity), the conversion increases from 8.0 to only 9.2%.
The conversion for the large-crystal sample of ZSM-5 it-
12.3. Evidence from other physical measurements self was 35%, although its composition was not given.
These results should be compared with those of work-
Following the original report by Jacobs and co-workers at the CSIR laboratories in South Africa [291,365]
ers [132], Gabelica et al. presented further evidence for which also suggest size and shape-selectivity in a product
X-ray amorphous ZSM-5 zeolites based on data from which, in this case, is amorphous by both XRD and IR cri-
thermal analysis and MAS NMR spectoscopy teria. In a detailed report, Howden [291] describes alumi-
[363,364]. Working on both type-A and type-B prepara- nosilicates prepared using TPA/Na ratios of >10 at
tions as delineated by Derouane and co-workers [58] temperatures up to 80 C for several days. A product re-
(Section 3.6), they contrast the evolution in analytical acted at room temperature for three days followed by
parameters from samples taken during the course of three days at 70 C contained 10 mass% of bound
the syntheses. In the type-A synthesis, the temporal indi- TPAOH but lacked both XRD crystallinity and any
cators of crystallinity (XRD, DTA, NMR) agree within absorption band at 550 cm1 in its IR spectrum. However,
experimental error, whilst the Si/Al ratio of the solid in a comparison of catalytic properties with a sample of
phase increases from 1.8 (48 h) to a final value of 13.2 ZSM-5 of near-identical composition (SiO2/Al2O3 = 82),
(312 h). In type-B, XRD crystallinity (12%) appears first the C-number product distribution of the two catalysts
at a reaction time of 49 h, following a zero value at 42 h. was similar, although the amorphous catalyst was much
Between 42 h and the first reported data at 14 h, the nor- less active and had a tendency to produce more branched
malised values from other indicators are 34–43% (DTA) molecules. The amorphous sample had only 60% of the
and 30–32% (NMR), i.e. these measurements are at lev- acid sites present in ZSM-5 although their strength (as
els corresponding to P30% of their final values even at measured by ammonia release) was the same.
the earliest recorded time. In the case of zeolite Y, Fahlke et al. [119] tested the
The type-A synthesis is interpreted as the solution- activity for cumene cracking of intermediate samples ta-
mediated formation of relatively large single crystals, ken during the course of a 72 h synthesis at 90 C. The
so that the TPA content of the solid phase is directly pro- sample taken at 48 h had 69% of the catalytic activity
portional to its XRD crystallinity. Conversely, in the of the final fully crystalline sample (72 h) but was X-
type-B preparation, the initial composition of the solid ray amorphous. Samples taken up to 10 h showed only
phase reflects that of the starting reagents. Extensive 4% of the final activity.
nucleation is said to occur rapidly to yield a large number The catalytic results given above all indicate the pres-
of XRD-invisible ZSM-5 crystallites by hydrogel phase ence in XRD amorphous materials of structured entities
transformation and it is the resulting strongly bound which possess to some degree the activity and selectivity
TPA species which are detected by DTA or 13C-NMR associated with zeolite catalytic sites. These properties
spectroscopy. This latter conclusion is not in dispute are most probably derived from a semi-ordered overall
but the anomalous time-trace of the derived ‘‘crystallin- structure. However, the superimposition upon back-
ities’’ is unreasonable. The most probable explanation is ground of activity resulting from genuinely crystalline
that the early 30% values reflect the presence of partially material present below the level of XRD detection can-
ordered TPA aluminosilicate complexes. not be ruled out.
12.4. Evidence from catalysis 12.5. Evidence from the synthesis process
Jacobs et al. compare the catalytic performance of In the paper by Jacobs et al. [132], the procedure used
samples taken before the appearance of XRD crystallin- for the synthesis was based on Example 1 of the original
Mobil patent [15] and is given as follows: ‘‘A mixture the structure, reducing base-catalytic activity to the level
with the following molar composition: ðNPrn4 OHÞ125 - of that found for amorphous silica-alumina. However,
ðNa2 OÞðAl2 O3 ÞðSiO2 Þ200 ðH2 OÞ5500 was allowed to react using toluene alkylation as a probe reaction, intermedi-
at 160 C. Crystallisation times varied between 0.25 ate levels of milling were concluded to have a differential
and 10 days. The materials were then air-dried, calcined effect, moderately decreasing both Lewis base and Lewis
at 550 C and ammonium exchanged.’’ Such a reaction acid site concentration whilst strongly decreasing Bron-
crystallised ZSM-5 in four days (as measured by the sted acid site density [371]. In the case of benzene oxida-
conventional XRD data) under what cannot be tion to phenol with ZSM-5 ferrisilicate, practically
described as mild conditions. complete amorphization caused the disappearance of
Calculations have been carried out to estimate the catalytic activity, so that in this case destruction of the
early growth rate of very small zeolite crystals [110] periodic channel structure also destroyed the active sites
(see also Section 8.5). The results show that the influence [374].
of the Kelvin effect (enhanced solubility, and hence Amorphisation under high pressure is a special case
reduced growth rate, at very small crystal size) is slight. in which the process of structural transformation and
Therefore, if a synthesis mixture contains entities accu- atom displacement might perhaps be compared to the
rately describable as ZSM-5 crystals (however small), compression of a spring. Up to a point, the dimensions
these entities will very rapidly grow into detectable mac- deform but the intact structure returns on relieving the
roscopic crystals. The linear growth rate (0.5Dl/Dt) of pressure. Progress beyond that point leads to irreversible
Na,TPA-ZSM-5 at 160 C is in the range 0.1–1.0 lm h1 destruction. Thus, Rutter et al. [375] carried out high-
[188,206] (see Table 3). Thus, any genuinely crystalline pressure synchrotron X-ray diffraction measurements
material present within the three-day X-ray amorphous at room temperature on samples of hydrated zeolite
period should have reached a crystal size of several Li-A and Na-A in a large volume press. After a rapid
microns. initial loss in long-range ordering at pressures up to
2.2 GPa, there was a gradual, continued decrease up
12.6. Other amorphous materials related to zeolites— to 4.1 GPa where the samples appeared to be X-ray
zeolite degradation amorphous. However, after decompression, diffraction
patterns at ambient pressure indicated that the samples
It has been known for many years that the crystalline reverted back to their initial crystal structure, demon-
structure of zeolites can be destroyed by the input of strating ‘‘structural memory’’. This was ascribed to the
suitable non-thermal energy, such as a TEM electron presence of ‘‘non-deformable units’’ [376] which act as
beam [366–368], mechanical treatment [369–374] or high templates about which the original crystal structure
pressure [375–381]. Two groups of researchers have car- can be reformed. However, in related experiments on
ried out detailed studies on the effect of ball milling (see hydrated zeolite Na-X, the sample became irreversibly
e.g. Refs. [369–371] and references therein). An interest- amorphous at 9 GPa [377]. The mineral zeolites scolecite
ing aspect of this work is the evidence from a variety of and mesolite were subjected to high pressure by Gillet
investigative techniques that not all structural order is and co-workers [378]. Progressive structural changes
destroyed, although the products show no X-ray diffrac- were observed below 10 GPa before complete amorph-
tion lines. For example, Kosanović et al. rendered zeo- ization. The transition from the crystal to the amor-
lites A, X and mordenite amorphous by ball milling phous phase was irreversible, the pressure-quenched
and then carried out hydrothermal syntheses on the res- samples showing Raman spectra similar to those of
idues to yield zeolites A, P and hydroxysodalite [372]. aluminosilicate glasses. However, in some cases, decom-
From powder XRD and particle size distribution mea- pression from 15 GPa was accompanied by a rim re-
surements, it was concluded that the nuclei for zeolite crystallisation and preservation of an amorphous phase
crystallisation originated from residual nano-sized with a structure different from that quenched in other
quasi-crystalline particles. These were envisaged as pos- experiments. The pressure-induced amorphisation of
sessing short-range ordering of Si and Al atoms inside zeolites has recently been the subject of further, in-depth
amorphous regions that had not been completely de- experimental and theoretical investigation [379–381].
stroyed during milling. Similarly, in a study of the dry
grinding of natural and synthetic zeolites, disappearance 12.7. Conclusions
of the XRD powder patterns indicated a major loss of
structure but analysis of the radial distribution function From the information given earlier (Section 6) and
suggested that the channel system was not completely above (Sections 12.2–12.7), it can be seen that both syn-
destroyed at the microscopic level [373]. thetic and degradative processes can lead to zeolite-re-
In some cases, catalytic properties of zeolites have lated materials which lack the regular periodicity of a
been investigated as a function of grinding. For KNaX crystalline lattice but which nevertheless do contain ele-
zeolites, high-energy ball milling eventually amorphized ments of short-range order similar to that found locally
in zeolite structures. These materials can display, to a fore follows that the better the template fills the void
limited extent, some of the spectroscopic and size-exclu- space, the greater the non-bonded interaction will be.
sive characteristics normally associated with zeolites. As a result, many examples are observed both directly
The ‘‘X-ray amorphous zeolites’’ of Jacobs et al. [132] from experiment and from theoretical calculations that
are members of this family. They are most probably re- show there is an excellent match between the template
lated to the semi-ordered, pre-nucleation phase de- and the framework. Highlighted here is the further
scribed earlier (Section 6.2), being neither crystalline in observation that this relationship often extends to a
terms of long-range, XRD-detectable order nor in terms strong match in symmetry between the template and
of viability for growth in a suitable nutrient medium. It the host framework.
is not credible that entities definable as crystals would Schmitt and Kennedy used the symmetry of the tem-
remain dormant under the synthesis conditions em- plate as a criterion for successfully designing new SDAs
ployed or would behave as described in the reported for ZSM-18 (MEI) [97]. Here, the ZSM-18 cage that
NMR/TGA experiments [363,364]. hosts the template molecule has the same threefold
The early literature on ‘‘X-ray amorphous zeolites’’ rotational symmetry as the template (Fig. 25). Further-
gives the impression that these materials are, in effect, more, Stucky and co-workers have shown [387,388]
genuine zeolites of very small crystal size. The data sum- that the symmetry of the framework can be dictated
marised above demonstrates that this cannot be the by the template molecule used. The reduction of a
case. The term ‘‘X-ray amorphous zeolites’’ is therefore sodalite cage from cubic to triclinic symmetry was dem-
both misleading and a misnomer. To paraphrase the onstrated using ethylene-diamine as structure directing
Summary section of the original report [132], it would agent.
be more accurate to say that the materials ‘‘contain
some ZSM-5-like domains of less than 8 nm in size in
an amorphous matrix’’. 13.2. Template classification and versatility
Although templates appear to require a specific shape

13. Template–framework interactions in order to successfully synthesise ZSM-18, there are cer-
tain frameworks that are successfully synthesised with
In Sections 9–12, attention has been focused upon an apparently diverse set of structure directing agents.
interactions between the synthesis chemistry and the This led Davis and Lobo [36] to classify templates into
development and transformation of zeolite structure. three different types: (i) ‘‘true’’ templates, (ii) structure-
In subsequent pages (Sections 14–16), this discussion directing agents and (iii) space-filling species. In this
will be extended to the stability relationships between classification, because many different organic molecules
the zeolite and its surrounding solution. In the present
section, we look within the zeolite crystal to examine
the correlation between the shape and symmetry of the
zeolite framework and those of an encapsulated organic
template.
13.1. Geometric matching
Molecular modelling calculations based on a molecu-

lar mechanics type approach have been widely used to
investigate the relationship between zeolite frameworks
and the structure directing agents (SDAs) used in their
synthesis (see for example Refs. [99,382–384]). One fea-
ture of these calculations is that they highlight the
importance of the non-bonded energy between template
and framework in determining the zeolite product that is
formed. In addition, an excellent correlation has been
observed between the non-bonded energy and crystalli-
zation time for a series of different templates that suc-
cessfully synthesise nonasil (NON) [385] and other
structures [386].
The non-bonded energy term in these calculations is
dominated by the van der Waals interactions between Fig. 25. The close shape matching observed between ZSM-18 and the
the framework atoms and those of the template. It there- triquat template molecule used in its synthesis.
side two different products, LEV and NON. Despite the

difference in conformation for the template molecule in
these cases, the calculated non-bonded energy is found
to be very similar.
13.3. Structure blocking
The structure directing agent (SDA) is usually added

to the synthesis mixture in at least stoichiometric
amounts in order to ensure complete crystallisation
and maximum void filling of the product. At lower levels
of addition, the most usual result in single-phase systems
is a reduced yield of fully SDA-loaded product, rather
than any significant reduction in the degree of void fill-
ing. However, in reactions where several product phases
can crystallise, other notable effects may be observed at
low template concentrations, in some cases showing a
major influence on the product formed. The most strik-
ing example of this has been provided by Casci [155]. In
this study, it was shown that the addition of only very
small quantities of the hexamethonium ion (CH3)3N+-
(CH2)6N+(CH3)3 (HEX) to a standard inorganic ZSM-
5 reaction mixture was sufficient to prevent the
formation of the MFI structure. At high HEX/SiO2 ra-
tios (P2.0), EU-1 was the only product formed. As tem-
plate levels fell to HEX/SiO2 = 0.004 and below,
crystallisation of EU-1 ceased, the products isolated
being mordenite and quartz. However, only when the
template was omitted entirely did the default product
Fig. 26. The N,N 0 -bis-dimethylpentanediyldiammonium template (ZSM-5) appear. This powerful ‘‘structure blocking’’ ef-
adopts different conformations when used to synthesise (a) the LEV fect has been attributed to the fact that the cation binds
and (b) the NON zeolite structures. However, despite the difference in strongly to the growing surface of ZSM-5, leaving bulky
conformation for the template molecules within the different frame- (CH3)3 groups in such a position that they can prevent
works, the calculated non-bonded energy is found to be very similar.
the formation of subsequent layers of the structure
[156]. Therefore, in this case, the key factor in determin-
were known to synthesise structures such as ZSM-5 and ing the product formed is not so much the strong geo-
ZSM-48, these templates were considered simply to be metric match with the potential new product as the
‘‘space fillers’’. ‘‘Structure directing agents’’ were defined geometric ‘‘mismatch’’ with the default product. This ef-
as those that direct a specific product to form, whereas fect has also been observed in a detailed investigation of
‘‘true templates’’ direct the framework to adopt the geo- two related aluminophosphonate structures [391]. An
metric and electronic configurations unique to the tem- associated case of synthesis inhibition has been de-
plating molecule. However, although some simple scribed by Zones and Nakagawa for preparations start-
organic compounds such as amines appear to display a ing from boron-beta as reactant source [392]. In some
non-specific relationship with their product, extensive experiments, adsorption of the organo-cation stabilised
experimental and theoretical work by Rollmann et al. the reactant framework, retarding or preventing its
[383,384] shows that even simple amines play a pore-sta- dissolution and subsequent transformation into new
bilizing role. Within a series of reactions, changes in product.
amine–pore interaction energies were consistent with
the changes in the zeolite products observed. 13.4. Variations induced by heteroatoms
In many cases, the same organic molecule can be used
to generate different products under different synthesis One effective route to synthesise new framework
conditions. Hong and co-workers [389,390] have shown types using the same organic agent is to incorporate dif-
that molecular flexibility is one factor which enables a ferent atom types into the network. This approach has
template molecule to synthesise more than one product been highlighted in a recent review by Corma [42].
as experimental conditions are varied. Fig. 26 shows the Substituting different atom types for Si or Al into the
two distinct conformations adopted by one template in- structure will alter the bond lengths and angles as well
as the stabilities of different secondary building units. In solid phases. In such a situation, it is not so easy to spec-
their work, Corma and associates have extensively stud- ify or measure the supersaturation parameters which are
ied the structure directing effects of replacing Si by Ge. driving the reaction. There is in theory a direct relation-
Theoretical calculations established that D4R secondary ship between a definite (but incompletely defined) super-
building units were stabilized when Si4+ was substituted saturation function and the equilibrium solubility value
by Ge4+ [393]. This has been used to improve crystallisa- of the crystalline zeolite product. However, this direct
tion times for ITQ-7, a structure containing D4Rs, from solution–crystal relationship is rarely seen, because in
seven days to 12 h by introducing Ge into the synthesis nearly all cases an amorphous precursor of intermediate
mixture. Moreover, this approach has been used to syn- solubility intervenes. This amorphous material is only
thesise a range of new structures, including ITQ-17, slightly more soluble than the crystalline product—but
ITQ-21, ITQ-22 and ITQ-24, all with D4R units present is able to precipitate more quickly than the zeolite can
[394–397]. crystallise, i.e. it is the first kinetic product in the Ost-
wald ‘‘successive transformation’’ sequence.
13.5. Conclusions
14.1. Zeolite solubility
A strong geometric match is often observed between
Early work by the Mulhouse group was aimed at
the framework and the template. In many cases, this ex-
establishing rules for zeolite solid–solution equilibria
tends to a match in symmetry also. High void filling
and solubility relationships [398–400]. In further papers
within the pore system maximises the short-range inter-
[401,402], the same workers showed that the silicate oli-
actions between the template and the framework, thus
gomer distributions in silicate solutions could be calcu-
stabilizing the product relative to the structure that
lated to reasonable agreement with experiment from
would be made in the absence of the SDA. Thus, a
known or estimated equilibrium constants. A classical
sound basis for the tailored design of novel templates
approach to zeolite solubility was adopted [398], so that
is provided by combining these features. However, such
the solid zeolite and the final solution species could be
an approach is not always necessary in order for the or-
related by one or more equilibrium constants (K). Thus,
ganic to influence the product formed. Small amounts of
for zeolites A and X:
a strongly interacting SDA can alter the relative stability
of one framework with respect to another through a NaAlSix O2ðxþ1Þ þ 2ðx þ 1ÞH2 O
structure-blocking mechanism. K
xSiðOHÞ4 þ AlðOHÞ4 þ Naþ

* via further equilibria with intermediate species such as

14. Solubility and supersaturation aluminosilicate complexes.
In an important series of recent papers, Šefčik and
In a conventional crystallisation, it is usually a fairly McCormick review the thermochemistry of aqueous sil-
straightforward matter both to define and to measure icate solutions [403], discuss what is meant by the solu-
supersaturation [108]. If we consider a simple salt such bility of zeolite A [404] and extend the modelling of
as sodium chloride, it will have an equilibrium solubility solution equilibria to the prediction of zeolite crystallisa-
(equilibrium saturation concentration, c*) at a given tion diagrams [405]. It is established [404] that simple
temperature, so that any excess concentration (Dc) con- concentration products for zeolite A such as [Si][Al] or
stitutes a measure of the supersaturation prevailing in [Na][Si][Al] do not provide a consistent description of
the system: saturation with respect to zeolite A across different li-
Dc ¼ c c ; quid phase compositions. However, the solubility prod-
uct calculated using a second-order model showed very
where c is the actual solution concentration. good consistency with the experimental data for mother
In thermodynamic terms, it is more convenient to de- liquor compositions and did not change with varying
fine a supersaturation ratio (S), since this is can be more liquid composition. This extended model allows for
directly related to the free energy driving the subsequent the presence in solution of Si(OH)4, AlðOHÞ4 ,

crystallisation: n +
SiOn ðOHÞ4n (n = 1, . . ., 4), OH , Na , and also de-pro-
c tonated silicate and aluminosilicate dimers with a nega-
S ¼ ;
c tive charge of up to 3 per Si (e.g. Si2 O6 7 and
2
AlSiO2 ðOHÞ5 ). Using the second-order model, the sol-
DG ¼ RT ln S:
ubility product of zeolite A at 80–90 C for [Na+] =
Hydrothermal zeolite synthesis is a multi-phase reac- 1–4 M was tentatively found to be:
tion-crystallisation process, commonly involving at least
one liquid phase and both amorphous and crystalline ½SiðOHÞ4
½AlðOHÞ4
½Naþ
¼ 1:2ð0:3Þ 108 M3 :
Experimental data for the (overall) solubility of zeo- Consider the simple relationship (Section 14.1) which
lite A and its amorphous precursor have been published defines silica solubility:
by Ejaz and co-workers. Data for both Si and Al com- K s ¼ ½SiðOHÞ4
:
ponents are given as a function of temperature and
NaOH concentration. The precursor was found to have The monomeric silicic acid is linked to a complex chain
a much higher solubility than the zeolite A crystals of oligomers and polymers, all of which will de-proton-
under similar conditions [406]. Related work based on ate as the pH is raised (Fig. 4). When base is added to
experimental investigations is described by Zhang the system, the equilibria are shifted towards the anionic
et al., who have developed semi-empirical expressions silicates and more solid silica dissolves. If sufficient base
linking silica solubility in supersaturated aluminate solu- is added, all the solid silica will pass into solution up to a
tions with the precipitation of sodium aluminosilicates value (for NaOH) of SiO2/Na2O ’ 3.5. It is sometimes
[407]. A further relevant observation has been reported incorrectly implied that this dissolution of silica alters
by Grujić et al. [408] as part of their study of the trans- the prevailing solubility and supersaturation relation-
formation of zeolite A into hydroxysodalite. Increasing ships. However, it can be seen from the above that these
the reaction temperature accelerated the transformation, are in fact tied to constant quantities (the Ks values) and
as would be expected. However, analysis of the effect of therefore unaffected (to a first approximation) by any in-
the two contributing components (increase in kinetic en- crease in base. The work of Šefčik and McCormick (Sec-
ergy and increase in supersaturation) showed that the tion 14.1) [404] can now be more fully appreciated, since
influence of the kinetic effect was about ten times more it proves the same point for aluminous zeolites: the true
important than the thermodynamic contribution. Exper- solubility product for zeolite A is arrived at only when
imental work by Young and Lowe to obtain accurate corrections are made for the contributions of higher spe-
values for the solubilities of high-silica zeolites in water cies to the overall solution concentrations of silica and
highlighted the difficulty of this task. In particular, alumina.
traces of residual non-crystalline material lead to
erroneously high values—to the extent that such mea- 14.3. Supersaturation in relation to zeolite crystal growth
surements represent a very sensitive technique for the
detection of amorphous impurities [409]. In view of the complexities outlined above (Sections
The equilibrium model of Lowe [92] was based on 14.1 and 14.2), it is not surprising that very few investi-
high-silica zeolite synthesis and took as its starting point gations of zeolite synthesis attempt to link supersatura-
the solubility (Ks) of pure silica: tion to growth rate, as would be the norm for studies on
SiO2ðsÞ þ 2H2 OðlÞ SiðOHÞ4ðsolnÞ simpler systems (e.g. ionic salts). Pioneering work on
zeolites has been carried out by the Zagreb group
K s ¼ ½SiðOHÞ4
: [161,410–413]. However, as pointed out by Šefčik and
McCormick [404], the crystal linear growth rates are
The driving force for zeolite crystallisation (Section 4.1)
linked by these authors to the total aluminium and sili-
was then related to the ratio of the solubility constants
con concentrations in the solution phase:
for the gel and crystalline phases:
growth rate ¼ constant ð½Al
½Al
sat Þ ð½Si
½Si
sat Þ;
K s;gel
DG ¼ RT ln :
K s;zeolite
so that the relationship is subject to error. As noted ear-
These studies were concerned with the establishment of lier (Section 8.3), Cundy et al. have suggested [148,186] a
guiding principles at the semi-quanitative level. There- pragmatic solution to the problem of measuring the real
fore, no direct allowance was made for the presence of supersaturation by relating the observed crystal growth
aluminium species nor for the possibility of changes in rate to a dimensionless ‘‘effective supersaturation’’ based
composition of the solid phases during the course of on the measured maximum growth rate value for the
the synthesis, although both these points were consid- same, or a comparable, reaction system:
ered [92,94].
dx=dtðobsÞ
effective supersaturation ¼ :
dx=dtðmaxÞ
14.2. Zeolite solubility as a function of base concentration
In terms of the Lowe equilibrium model [92], the super-
There is a dilemma over the concepts of solubility and saturation may to a first approximation be envisaged as
of supersaturation in zeolite systems in terms of the rela- changing across the reaction coordinate as shown in Fig.
tionship between these quantities and the concentration 27. The ‘‘effective supersaturation’’ represents the frac-
of base present. This will be discussed in the context of tion of the total DKs which is applying under the reac-
high silica zeolites, for which the problem is particularly tion conditions. This value for the concentration
acute. driving force cannot exceed the unit value of DKs but
Ks(gel) nary ammonium complexes of zeolites ZSM-5 and beta

Solution Si(OH)4 concentration
amorphous (gel) precursor to be in the range 4.9 to 8.5 kJ mol1 SiO2. No single
thermodynamic factor dominated the overall Gibbs free
energies and the small energetic differences were taken to
suggest that kinetic factors were of major importance in
∆Ks molecular sieve preparation. It is possible to reach an
analogous conclusion from calculations based on the
Ks(precursor zeolite)
Lowe equilibrium model for zeolite synthesis [31,92,94]
crystalline and the similarity in template binding energy between
zeolite different products of a common organic guest molecule
x Ks(zeolite) has already been noted (Section 4.2.1).
Reaction coordinate Kinetic effects, rather than any augmentation in
supersaturation, are also the reason behind the rise in
Fig. 27. Silica monomer (Si(OH)4) concentration during the course of
zeolite crystallisation rate usually found for an increase
a high-silica zeolite synthesis according to the concepts of the Lowe
equilibrium model [92]. The difference (DKs) between gel solubility and in the base level (Sections 8.3 and 17.1). Adding further
zeolite solubility is the true supersaturation and represents the base will cause more silica to go into solution so that the
concentration driving force for the reaction (Ks(gel) Ks(zeolite)). The concentration of dissolved species will increase. This in-
interval x represents the period over which the pH will rise, creased flux of nutrient units will potentially cause an in-
corresponding to the transfer of control of solubility from precursor
crease both in nucleation rate and in crystal growth rate.
to product. If the starting material is another zeolite (or other
crystalline precursor), the supersaturation is consequently reduced
(Ks(precursor zeolite) Ks(zeolite)). 14.5. Summary
The topic of zeolite solubility is not a simple one. It is

may be less than this because of kinetic restrictions due complicated both by experimental difficulty in measure-
to (for example) mass transport limitations or reactant ment and also in the interpretation of the results. Accu-
dissolution rate. rate expressions for solubility product have to be made
An important point follows from the increase in over- in terms of basic equilibria. Allowance must be made
all silica solubility which occurs with rising alkalinity for the presence of more complex species which add to
(Section 14.2). Typical values for true silica solubilities the total solution (and sol) concentrations but do not
in water are 2.5 · 103 and 1.8 · 104 mol dm3 for contribute to the key solubility relationships. Similar
amorphous and crystalline silica, respectively [92]. reservations apply to reagent-induced effects such as
Hence if Fig. 27 is imagined as applying to a (very slow) the high-level of silica ‘‘solubility’’ in strongly basic sili-
crystallisation of silica in pure water, the value of DKs cate solutions. These same considerations have to be
would be of the order of mmol dm3. Since silica con- taken into account when defining the concept of super-
centrations in alkaline zeolite synthesis sols can be of saturation in zeolite synthesis systems. To a first approx-
molar magnitude, it follows that changes associated imation, this is dependent only upon temperature, the
with variation of pH and composition will far exceed nature of the starting materials (silica and alumina
the very small difference related to DKs, which will there- sources) and the nature of the product phase, and is thus
fore be masked. essentially constant over the major part of the zeolite
formation process. This is the reason why crystal linear
14.4. Thermodynamic vs. kinetic factors in zeolite growth rates are themselves essentially constant over the
synthesis same period [111]. For a given zeolite, differences in
growth rate between one synthesis system and another
Throughout this survey, many references have been are largely kinetic (as opposed to thermodynamic) in
made to the impact of thermodynamic and kinetic fac- origin.
tors in the synthesis of zeolites. It has been noted that,
in general, the products of such reactions are most likely
to be kinetically determined. Recent papers by Petrovic 15. Zeolite dissolution
et al. and Piccione et al. support this observation by pre-
senting data on the thermochemistry of high-silica zeo- Having considered the topic of solubility and,
lites [414,415] and the thermodynamics of their through this, the relationship between supersaturation
synthesis [46]. The observed range of enthalpies of tran- and crystal growth in zeolite synthesis, it is now neces-
sition is quite narrow at only 6.8–14.4 kJ mol1 above sary to examine the processes by which zeolite crystals
that of quartz. Correspondingly, Gibbs free energy dissolve. Although closely related, the mechanisms of
changes for the crystallisation of microporous silica growth and dissolution are not necessarily just the re-
phases from amorphous silica are estimated for quater- verse of one another. Solution compositions, and the
associated chemical driving forces, can be markedly dif- the dissolution of the thermal decomposition products
ferent. However, there are also interesting parallels: for occurred in two stages. After a rapid dissolution of a
example, just as nuclei have to be created in the process thin (sub)surface layer and slower consumption of the
of crystal formation, so holes have to be nucleated to bulk particles, the reaction became one of solution-med-
initiate the process of dissolution—and both processes iated transformation into zeolite Na-A. The dissolution
tend to show an induction period. of amorphous aluminosilicate zeolite precursors has also
The zeolite synthesis process is one of dynamic equi- been the subject of further research by the same group
librium between crystal growth and crystal dissolution. [424–426].
As noted earlier, the reversible nature of the growth Some investigations on the dissolution of the high-sil-
reaction provides the key mechanism for the equilibra- ica zeolite ZSM-5 have also been reported. Čižmek and
tion reactions which enable order to be established as co-workers have studied the dissolution of a range of
the synthesis proceeds (Sections 6 and 7). As zeolite MFI materials of varying alumina content in 5 M
crystals dissolve, new information about the zeolite– NaOH solution at 60 C. For samples containing the
solution interaction can be revealed, especially in cases TPA template, the initial rate of dissolution was found
of non-congruent dissolution, i.e. those in which the to be directly proportional to the surface area of the dis-
composition of the dissolving solid and that of the sur- solving solids, the rate constant being a linear function
rounding solution are not the same. As might be ex- of the solids Si/Al ratio. Later behaviour was more com-
pected from considerations of solubility and crystal plex due to preferential dissolution of silica, inhomoge-
growth (Section 14), the rate of dissolution is in general neous distribution of aluminium in the ZSM-5 crystals
proportional to the surface area of the dissolving solid themselves and interactions between solution species
and the degree of undersaturation of the solution. [420,421]. After calcination, dissolution of the tem-
plate-free samples was found to be much faster than be-
15.1. The kinetics of dissolution fore, although the analysis was complicated by the
precipitation of a solid product in the course of the reac-
The most detailed study of zeolite dissolution is to be tion. Plots of dC/dt against dissolution time showed that
found in the notable series of papers published by Sub- the rate of dissolution (expressed in these terms) in-
otić and associates [416–422]. These reports present creased with time for the templated samples but de-
extensive experimental data and the corresponding creased with time for those which had been calcined
mathematical analyses of the dissolution process. Disso- [422]. A different aspect was shown by Cundy et al. in
lution of the aluminous zeolites A, X and mordenite in a study in which both the synthesis and the dissolution
hot sodium hydroxide solutions [416–418] took place of ZSM-5 were investigated in detail [111]. In this case,
in accordance with the model of Davies and Jones the results were analysed in terms of two-thirds order
[423] which implies the presence of a monolayer of sol- plots (Q vs. t), which would be linear with time if disso-
vated ions with constant composition at the dissolving lution occurred by a shrinking core process with reten-
crystal surface: tion of shape:
dC dx 2=3
b
¼ kSðC Al C Al ÞðC Si C Si Þ ; ¼ kð1 xÞ ;
dt dt
where dC/dt is the differential change in the concentra- so that, on integration:
tion, CX (X = Al or Si) is the liquid phase concentration, 1=3
Q ¼ ½1 ð1 xÞ
¼ kt;
C X the equilibrium concentration, k the dissolution rate
constant, S the total surface area of dissolving particles where x is the molar fraction of tetrahedral atoms dis-
and b the Si/Al ratio of the zeolite. (Note that the con- solved at time t and Q is the extent of dissolution.
centration terms are the total measured values and thus After some initial preferential dissolution of (proba-
(Section 14.3) do not accurately represent the true bly amorphous) silica, this model was followed quite
undersaturations.) For zeolites with Si/Al > 1, the disso- well for the TPA-containing sample, and allowed an
lution was found to be incongruent, with a gradual in- analysis of the inhomogeneous radial Si/Al profile to
crease in the Si/Al ratio of the liquid phase. In an be carried out. Values for Si/Al of 200–300 near the
interesting extension of this work [419], similar dissolu- core, reducing to about 40–50 towards the surface,
tion studies were conducted on zeolites Na-A, K-A agreed well with analytical data obtained during the ori-
(from ion exchange) and two thermal decomposition ginal growth of the crystals and also with XPS measure-
products of K-A: an amorphous material obtained by ments on the final product. Comparison of the
heating K-A at 962 C for 30 min and a re-crystallised dissolution rate of the templated ZSM-5 with that of a
mixture of kalsilite and kaliophilite produced by more comparable calcined (i.e. template-free) silicalite (Si/
severe treatment (962 C for 180 min). Whilst the origi- Al > 800) showed the calcined material to dissolve about
nal zeolites followed the previously observed behaviour, 40 times faster. The data of Čižmek et al. [421,422]
suggest that the contribution of template removal to this ZSM-5 five times more slowly than the silicalite and,
increase far outweighs the additional enhancement due after perforation of an (amorphous) aluminous surface
to the lower Al content. The non-linear (concave) Q-plot layer, extracted relatively Si-rich material from the inner
for the open-pore sample also demonstrated the auto- parts of the particle to leave an aluminium-rich crust.
acceleration which occurs in specific dissolution rate as Several groups of investigators have examined the
the original shape is lost and the sample fragments. dissolution of calcined ZSM-5 samples. Young and
Lowe studied the dissolution of piperazine,TPA-ZSM-
15.2. Morphological and compositional changes on 5 in water at 90–180 C for a wide range of Si/Al values
dissolution [409]. Detailed SEM observations on siliceous samples
revealed preferential dissolution of the interior of the
In some of the studies mentioned above, useful in- crystals. Solubility measurements as a function of
sights into the dissolution process are obtained from solid–liquid ratio allowed an accurate estimation of
the morphology of the dissolving crystals. The twinned the quantity of trace amorphous contaminant to be
ZSM-5 crystals examined by Cundy et al. [111] dissolved made. Hydrothermally modified aluminous products
in hot caustic soda solution as shown in Fig. 28. Surface were tested catalytically in butene isomerisation, and en-
pitting was followed by attack at the twin boundaries, hanced activity and selectivity were demonstrated. A
revealing the curved junctions of the 90 intergrowths similar approach was adopted by workers at Mobil
described by Hay and co-workers [427] and discussed [430] where calcined ZSM-5 samples were treated with
in detail elsewhere [231]. The minor component of the refluxing 0.5 M Na2CO3 solution. Silica-rich material
twin was eventually removed completely to leave the was removed from the interior of the crystals and the
remaining crystal as a torus. In a related investigation, resulting alumina-rich materials showed enhanced cata-
Le Fêbre [428] prepared crystals of silicalite and ZSM- lytic activity in hexane cracking. Le Van Mao et al. [431]
5 (Si/Al = 36) by the ammonia/TPA route [429] and treated several zeolites including ZSM-5 with 0.8 M so-
studied their dissolution in refluxing 2 M NaOH and dium carbonate solutions at 80 C. Silica was removed
in 1 M HF at 60 C. Fairly uniform attack by pitting from the ZSM-5 crystals so that their Si/Al ratio was
was observed for NaOH upon silicalite and for the HF approximately halved and additional mesoporosity was
in both cases. However, the caustic alkali dissolved the created at the expense of micropore volume. More re-
Fig. 28. Dissolution of Na-TPA-ZSM-5 crystals in 1 M NaOH at 80 C [111]. Reaction times (h) are: (a) 0.5, (b) 3.0, (c) 6.0, (d) 9.0. Scale bar = 1 lm
throughout.
cent work has developed this latter effect as a means of heterogeneity and interactions between solution species.
modifying the overall porosity of high-silica zeolites The process of crystal dissolution can be utilised both to
[432,433]. probe these factors and to modify crystal morphology,
An interferometric technique was used to study the in particular to introduce secondary porosity. Detail of
dissolution of single crystals of Na,TPA-MFI materials changes at the atomic level during crystal dissolution
(Si/Al in synthesis mixture = 1, 1000, 500) [434,435]. can be investigated using AFM techniques. However,
The effect of dissolution on all three crystalline faces intimate chemical insight remains a future prospect,
was examined. The dissolution rate was strongly depen- since in no case so far have the dissolving units been
dent upon the framework aluminium content (even at identified.
the very low levels used) and upon the concentration
of the NaOH solution. It was found that the crystals
were convex in shape with the inclination of the (0 1 0) 16. Metastability
face in ZSM-5 greater than that for silicalite. The activa-
tion energies for dissolution of the three crystal faces of 16.1. Precursors, intermediates and co-products
silicalite were found to be comparable, and similar to
that of quartz. In the course of a zeolite synthesis, the first product
The dissolution of the zeolite heulandite has been to appear may not be the target material. This is as ex-
studied in detail by a range of spectroscopic and micro- pected from the Ostwald Law of Successive Transfor-
scopic techniques. Ragnarsdottir et al. examined the mations (or Rule of Stages) which states that an
crystal surface by SIMS, XPS and SEM before and after unstable system does not necessarily transform directly
treatment with solutions of pH 2–12 [436]. At acidic pH, into the most stable state but into one which most clo-
the surface became depleted in aluminium, being re- sely resembles its own, i.e. into another transient state
duced at pH 2 to a state essentially equivalent to amor- whose formation from the original is accompanied by
phous silica. At all values of pH, there appeared to be an the smallest loss of free energy [108,439,440]. In zeolite
ion exchange between Na+/K+ and Ca2+, with an addi- terms, this means that a series of kinetic products may
tional charge-balance exchange between Al3+ and H+ form as transient phases on the way to the final ther-
(hydrolysis). More direct morphological information de- modynamic product. In some cases, such preliminary
rives from an AFM study by Yamamoto and co-work- phases may represent necessary mechanistic steps in
ers [437] in which natural heulandite was immersed in the construction of the product framework and may
either 0.1 N NaOH or 0.2 N H2SO4 aqueous solutions truly be called precursors (although it is inevitable that
at room temperature. The alkaline treatment caused at- this term will also be applied less strictly). In others,
tack at the uppermost aluminosilicate layer of the (1 0 0) the initially observed phases have no primary mecha-
surface, leaving isolated or agglomerated islands. The nistic significance and can be termed intermediates.
acidic solution attacked the (0 1 0) surface, forming pits. These are only indirectly involved in the synthesis,
Step retreat did not occur in either case, so that the crys- essentially as nutrient-storing transients. However, if
tal dissolved from terraces layer-by-layer as the reagents such materials persist long enough to appear in the
penetrated the pores. Dissolution at low pH took place product as impurity phases, they are best regarded as
at a significantly faster rate than at higher pH values, co-products.
this being ascribed to the more rapid hydrolysis of In most zeolite syntheses, the structured amorphous
AlAOASi linkages in acidic solution compared with that phase which precedes nucleation, the secondary amor-
of SiAOASi bonds in alkaline solution [438]. In both of phous phase (Section 6.2) is an essential stage on the
the above studies, evidence was obtained for weakening way to zeolite formation and is therefore a true precur-
of framework bonds on the hydrated mineral surface, sor. Similarly, the crystalline, as-made product from an
thus presumably facilitating the dissolution of the organically templated synthesis is a true precursor to the
structure. open-pore zeolite formed by subsequent calcination (or
other means of removing the organic component).
15.3. Summary However, many of the transformations seen in zeolite
synthesis represent the formation and subsequent
Although the chemistries of zeolite crystal growth decomposition of metastable intermediates. Indeed,
and dissolution are closely related, there are consider- many of the target zeolites are themselves metastable
able mechanistic and morphological differences between in the reaction medium and not true thermodynamic
the two processes. The kinetics of dissolution have been products. Much of the art of zeolite synthesis has cen-
investigated in detail, the rate being in general propor- tred around the discovery of optimum conditions for
tional to the surface area of the dissolving solid and the isolation of such products by determining the best
the degree of undersaturation of the solution. Complex compromise between reaction time, yield and phase pur-
behaviour may be due to factors such as compositional ity [47].
It should not be assumed that an intermediate will and pentasil-group zeolites have been discussed by
have no effect on the more stable product which follows Garcés et al. [446].
it. It is possible that the more stable phase will nucleate Tuel [447] has made a detailed study of the synthesis
on the surface of the intermediate—which could then be of zeolites ZSM-48 and beta with hexamethonium and
regarded as a catalyst to product formation. Similarly, tetraethylammonium cations, respectively. In both
the morphology of the secondary product may by influ- cases, highly crystalline layered organo-silicates or org-
enced by that of the intermediate—in extreme cases fol- ano-aluminosilicates were formed at short synthesis
lowing the latter as a pseudomorph. times but gradually decomposed to amorphous materi-
als. The zeolite products started to form only after the
complete degradation of the layered structures. The
16.2. Layer structures as transients and precursors two layered compounds, although from different reac-
tion systems, seemed to possess very similar structures,
There have been several different instances of the whereas there were no elements common to the XRD
appearance of lamellar materials during the course of patterns of the final zeolite products. It was concluded
zeolite syntheses. Materials with layer structures such that the layered materials dissolved in the reaction med-
as the silicates kanemite, magadiite or ilerite (sodium ium, providing nutrients which were reused in the crys-
‘‘octasilicate’’) have been used as starting materials tallisation of the zeolites, the structure of which
(see Ref. [441] and references therein). A number of such depended only on the nature of the quaternary ammo-
syntheses fall into the category of apparent ‘‘solid state nium template cation present in the liquid phase.
transformations’’ (Section 10). In some cases, amor- A very different situation exists with a family of
phous micro-precipitates have been observed in the form materials which do not convert to zeotype phases hydro-
of lamellae, upon which the zeolite product appeared to thermally but which do so upon isolation and high-tem-
nucleate [142,180]. In other situations, crystalline inter- perature calcination. In general, the initial structure is
mediates having layer structures have been observed be- that of a two-dimensional, layered complex which, on
fore zeolite crystallisation and have been regarded as removal of the organic ‘‘spacers’’, becomes converted
precursors to the subsequent three-dimensional micro- into a three-dimensional framework by the condensa-
porous networks [442–444]. In the crystallisation of tion of silanol groups. The first example of a zeolite
TEA-beta, a similar observation led Lohse et al. to con- formed by calcination of a layer-structured precursor
clude that the zeolite was formed via a direct transfor- appears to be that of NU-6 [448]. Later examples in-
mation of a layered aluminosilicate [445]. In none of clude the silica polymorph EU-20 [449] and the borosil-
these cases is there proof of the suggested 2D ! 3D icate ERB-1 [450]. Most detail is available in the case of
transformation. Some possible structural relationships ‘‘PREFER’’, where the transformation of this two-
between layered alkali metal silicates and the mordenite- dimensional aluminosilicate into the zeolite ferrierite
Fig. 29. The family of materials derived by delamination and reassembly from the laminar precursor PREFER (CTMA+ = cetyltrimethylammo-
nium). (Adapted from Ref. [454].)
on calcination at 550 C has been carefully studied version of zeolite LTA into zeolite Li-A(BW) using
[451,452]. The XRD pattern of PREFER contains dif- aqueous LiCl at temperatures of 180–260 C [459]. The
fraction lines similar to those of the derived zeolite, crystallisation was followed by in-situ time resolved syn-
reflecting the structural conformity between the two chrotron XRPD combined with ex-situ SEM of partially
materials. The precursor appears to consist of ferrie- converted samples. These combined techniques clearly
rite-type sheets in the bc plane, separated by molecules showed that the inter-zeolite conversion was a solu-
of the bulky template (4-amino-2,2,6,6-tetramethylpi- tion-mediated process, the Li-A(BW) product growing
peridine). On burning out the occluded organic mole- radially from nucleation sites on the parent LTA
cule, the ferrierite layers progressively link together in crystals.
the a-direction as new TAOAT linkages are formed by A complex and interesting system of transformations
condensation reactions. However, it is also possible to surrounds the synthesis of zeolite omega, the synthetic
modify this process. PREFER and related materials analogue of the mineral zeolite mazzite and also iso-
can be delaminated and reassembled in a number of structural with the synthetic phase ZSM-4 [16]. Many
ways to give a series of high surface area derivatives with syntheses of zeolite Na-Y will over-run to give zeolite
catalytically active sites accessible to bulky reactants. Na-P. However, in the presence of tetramethylammo-
Examples from the Valencia school are ITQ-2 [453], nium (TMA) ions, this transition does not occur, the
ITQ-6 and ITQ-36 [454] (Fig. 29). Similar relationships faujasite decomposing instead to zeolite ZSM-4. In the
characterise the Mobil MCM-22 family [455,456] which study by Dwyer and Chu [460], no ZSM-4 was formed
already form the basis for a new and commercially oper- until the faujasite crystallisation was complete and there
ating process for the manufacture of ethyl benzene [457]. was no evidence for any amorphous or crystalline inter-
mediate, the ZSM-4 appearing to nucleate on the sur-
16.3. Conversion of one zeolite into another face of the dissolving faujasite crystals. Broadly similar
results were reported for zeolite omega [461]. A more re-
Just as lamellar phases are sometimes transiently ob- cent investigation found that faujasite appeared as an
served during the course of zeolite preparations, the for- intermediate in Na,TMA-omega synthesis at 100 C
mation of one zeolite phase may be preceded by but was not observed at 135 C. Depending upon the
another. In early work on the synthesis of ZSM-5, choice of reagents and the reaction conditions, sodalite,
Erdem and Sand observed the intermediate appearance analcime, mordenite and cristobalite were also observed
of mordenite and analcime during the synthesis of Na, as co-products [462]. Other transformations which have
TPA-ZSM-5 as the TPA/Na ratio was lowered [458]. been reported include clinoptilolite to faujasite [463],
For the TPA,K system, metastable harmotome was ob- Na-A to Na-Y [464], faujasite into hydroxycancrinite
served. However, zeolite metastability is most com- [465] and a pseudomorphic transformation of zeolite
monly encountered in the phenomenon of reaction Na-A into nepheline by the action of water vapour at
over-run, in which the zeolite product decomposes 600 C [466].
under the reaction conditions to afford a more dense Two research groups have made particular studies
phase. In this way, aluminous zeolites such as A or X of zeolite–zeolite transformations. Subotić and co-
give way to sodalite, whilst siliceous zeolites transform workers carried out a detailed investigation of the con-
into dense silica phases, usually quartz or cristobalite. version of zeolite A into hydroxysodalite which occurs
Intermediate cases maintain the materials balance by in concentrated (>10%) sodium hydroxide solutions
undergoing a type of disproportionation, so that alumi- [408,467–469]. Working in the temperature range 70–
nous ZSM-5 products decompose to quartz and a more 85 C, they monitored the phases and particle popula-
aluminous zeolite such as mordenite or (for high-Al tions present in the reaction mixture and measured the
samples) analcime. These latter cases provide examples concentrations of silica and alumina in solution. From
of the conversion of one zeolite into another. Whilst a detailed analysis of the reaction kinetics, they con-
these instances are more likely to be troublesome than cluded that the reaction was a solution-mediated pro-
helpful, there are cases in which zeolite interconversions cess in which the zeolite A dissolved and the sodalite
can be of practical use, or interesting from a mechanistic product nucleated heterogeneously. The growth of
point of view. the sodalite crystals (by surface addition of monomeric
One of the earliest successful zeolite syntheses—and silicate and aluminate ions) was the rate-limiting step
the first synthesis of a zeolite not found in nature— of the conversion. No intermediate phases were de-
was the conversion of analcime into two KFI-type tected and the transformation rate increased strongly
phases by Barrer in 1948 [2–4]. This synthesis was not with increasing NaOH concentration. Similar conclu-
of the OH-mediated type but involved heating the sions were reached in a parallel study of the conver-
starting phase in strong solutions of barium salts at sion of zeolite A into zeolite P which occurs under
170–270 C. The same generic procedure was the subject comparable conditions but at lower base concentra-
of a more recent detailed study of the hydrothermal con- tions [470].
The other team to investigate zeolite–zeolite transfor- decomposition of zeolite LSX [477]. The pure phase
mations in detail is that of Zones et al. The work initially was synthesised in a mild environment (20–90 C) and
set out along the lines of the faujasite ! omega conver- then subjected to more forcing conditions (e.g. 150 C,
sion (see above), exploiting the unusual organo-cations several hours) under thermal or microwave heating in
which were to become such a characteristic feature of its mother liquor. In circumstances where the faujasite
this groups work. In this way, cubic zeolite P (Na-Pc) appeared to be stable, or to undergo only very slow
was transformed into zeolite SSZ-13 (a high-silica decomposition, seeds of the decomposition product
chabazite) by the action of the N,N,N-trimethylammo- (sodalite) were added to the reaction mixture. The deg-
nium-1-adamantane cation, although the presence of radative transition was then found to occur very rapidly.
further silica was necessary to satisfy the mass balance It is thus clear that the pure phase will be stable in its
[471]. Starting as before from Na-Pc, other templates mother liquor whereas a product already containing
gave sodalite (from TMA) and (from quaternaries con- some sodalite impurity will decompose. The stability
taining cyclic units) the zeolites SSZ-16, SSZ-17 (NU-3) of the pure product will ultimately depend on the possi-
and SSZ-19 [472]. In all cases, the organocations were bility of nucleating the decomposition phase under the
incorporated into the crystalline products. With the prevailing conditions. The difference between a success-
extension of this work to faujasites as starting reagents ful synthetic route and a closely similar but problematic
[473], an interesting synthetic method began to develop one may therefore be only that the latter co-produces a
[474], since it was found that the reactivity of the system minor amount of a critical impurity—the decomposition
could be modified by treating the primary zeolites in dif- product of the major phase. In this case, the first (pure)
ferent ways, for example by steaming, calcination or ion synthesis may allow the desired metastable product to
exchange. The synthetic utility of inter-zeolite conver- reside unchanged in its mother liquor for a prolonged
sions was subsequently well demonstrated by the versa- period. The alternative synthesis will at best show a
tility discovered for boron-beta zeolite [475], which gave strong sensitivity to reaction time and at worst may ren-
enhanced nucleation, increased reaction rates, greater der the isolation of a tolerably pure product impossible.
flexibility in choice of template and new possibilities
for lattice substitution. In later studies, it was also found 16.4. Ostwald ripening
that, for a given template, product selectivities could be
altered by changing the initial B/SiO2 ratio, for example Related to the topics of solubility (Section 14), disso-
by adding extra borate to the reaction mixture [392]. lution (Section 15) and the present consideration of
These new synthesis strategies for zeolite–zeolite metastability is the phenomenon of Ostwald ripening
transformations have subsequently been adopted by [108]. This coarsening of particle size is driven by the
other workers (e.g. Ref. [476]). Their utility stems from tendency of solid particles dispersed in saturated or
the additional degree of control attainable in these crys- supersaturated solution to achieve a minimum in total
tal-to-crystal routes compared to the more usual path- surface free energy. Thus the smaller particles dissolve,
way from amorphous starting materials. This results in the larger grow, the size distribution narrows and the
a reduction of supersaturation level, since the starting mean crystal size increases.
phase will have a solubility intermediate between that The Gibbs–Thomson relationship (Section 8.5) quan-
of a normal amorphous precursor and that of the crys- tifies the difference in solubility between small and large
talline product (see Fig. 27). In addition, a crystalline particles:
precursor may have unusual surface reactivity, e.g. for h c i 2rv
r
the generation of heterogeneous nucleation sites. ln ¼ ;
c RTr
A further point which does not appear to have been
addressed concerns the overall rate limiting step for where cr is the solubility of small particles of size r and c*
the above phase transformations. This is clearly of some is the equilibrium (saturation) solubility of large parti-
importance since control of the stability of the primary cles (i.e. r ! 1), r is the surface energy of a solid parti-
phase under reaction conditions may be critical to the cle in contact with the solution and v is the molecular
success of the synthesis. In some cases, the decomposi- volume of the solute. From this is derived the
tion may be initiated by the attack of solution species expression:
upon the primary phase. However, in the case of the 2rvc
transformation of zeolite A into sodalite, Grujić et al. r¼ ;
RT ðc c Þ
have shown [408] that the reaction in a seeded system,
once underway, is limited by the growth rate of the where c is the average bulk solution concentration. All
sodalite decomposition product. This suggests that particles of size r will be in equilibrium with the bulk
the stability of the primary product is governed by the solution (dr/dt = 0); those of a larger size will grow
nucleation of the decomposition product. This has re- (dr/dt > 0), whilst those smaller than r will dissolve
cently been confirmed in studies of the formation and (dr/dt < 0). In general, the Ostwald ripening effect is
only likely to be significant for very small particles 17. Optimisation of zeolite syntheses
(r 6 1 lm). The kinetics are also very dependent upon
solubility. Comparing K2SO4 (solubility 103 mol m3) A great deal of effort has been expended in estab-
and BaSO4 (solubility 102 mol m3) and assuming lishing reliable and reproducible synthetic routes to
the effect to be growth rate limited, the relative ripening known zeolites (see e.g. Ref. [48]). Clearly, there are
rates would be at least 105:1 and may in any case be many different aspects to this goal and a variety of
dominated by factors affecting the surface reaction [108]. approaches to its attainment. For example, an indus-
Ostwald ripening is frequently cited as playing a part trial production team will usually be most concerned
in zeolite crystal growth and a model based on this effect with overall productivity (mass yield per reactor per
has been used to relate crystallisation rate to product sta- unit time), product purity and reproducibility. The
bility [242]. However, as far as the present authors are product must also remain within specification, which
aware, no well-authenticated example of Ostwald ripen- will include further criteria such as particle size and
ing in zeolite chemistry has ever been provided. In a re- colour. There are also likely to be application stan-
cent study of nanoscale zeolite TMA-A crystallisation, dards such as catalytic performance. However, con-
Mintova and co-workers report a system which crystal- cealed within these practical measures are more
lises completely in seven days at room temperature fundamental parameters, such as those relating to
[138]. On subsequent hydrothermal treatment at 80 C the solubility, nucleation rates and growth rates of
for one day, this material (40–80 nm) was found to have the product crystals.
re-crystallised to a much larger (200–400 nm) product. In There are many reports in the literature which discuss
our own studies [477], similar synthesis compositions possible improvements to zeolite synthesis systems. In
crystallised to nanosized TMA-A over a comparable some cases, the bases for such claims are not well
timescale but the particle size distribution remained con- defined. It is the purpose of this section to examine some
stant on subsequent or prolonged heating unless the pre- of the possible approaches and review the criteria
cursor contained a residue of amorphous material which necessary to establish the case for a significant advance.
could act as nutrient for further growth. The effect was The discussion will concentrate upon kinetic
also specifically investigated during a study of colloidal considerations.
silicalite synthesis [150] and no evidence for its
occurrence could be found. Such dissolution-growth
17.1. Comparisons of zeolite synthesis reaction rates
combinations very probably occur amongst amorphous
or semi-ordered particles and in the earliest stages of zeo-
The statement is frequently made that one zeolite syn-
lite crystal growth. It is quite possible that a clear demon-
thesis is ‘‘faster’’ than another. Some examples of the
stration of the operation of Ostwald ripening in zeolite or
expression of reaction rate might be (i) how long a reac-
zeotype synthesis will in due course emerge, but until
tion time appears to be required, (ii) the time taken to
then the significance of this effect in the chemistry of
achieve maximum crystallinity, or (iii) a measurement
microporous materials must remain uncertain.
of the crystal linear growth rate. Only the last mentioned
is likely to yield fundamental data, whilst the measure-
16.5. Summary
ment criteria for (i) may be extremely loosely defined.
Often, there is no distinction between the induction time
Zeolite syntheses frequently display an element of
and the growth period, so that it is impossible to tell
metastability. In some cases, transient phases are ob-
whether a reaction is ‘‘slow’’ because it takes a long time
served before the appearance of the desired product,
to nucleate or because the crystals grow slowly in the gi-
whilst in others the product itself may tend to decom-
ven circumstances.
pose into further species. The observed intermediates
A comparison of some related synthesis reactions is
may be true chemical or structural precursors to the fi-
shown in Table 4. This is derived from a computer sim-
nal product but in most cases they lack this mechanistic
ulation of zeolite growth based on a simple crystal
significance. In zeolite-to-zeolite (or similar) transforma-
growth model [79,147,148,150], i.e.
tions, the ability to modify the starting material in a
controlled manner (e.g. by ion exchange) may offer the dr
radial growth rate ¼
possibility of new and useful synthetic routes. dt
The rate-limiting step in such conversions is likely to
be the nucleation and subsequent growth of the product and
phase. Therefore, the stability of a zeolite product in its
4npqr3
mother liquor (i.e. its resistance to reaction over-run) m¼ ;
3
may depend crucially on its purity. There appears to
be no convincing evidence for the operation of Ostwald where m is the mass of product crystals (density q,
ripening in zeolite crystallisation reactions. equivalent spherical radius r) per unit volume and n is
Table 4 ried out elsewhere [478], the nucleation periods and

Zeolite growth simulations: demonstration of comparative rate criteria growth times can be estimated. Knowing the growth
Parameter settings Run 1 Run 2 Run 3 periods and the final crystal sizes, approximate values
Crystal radial growth rate (lm h ) 1
0.01 0.02 0.01 for the growth rates and number of crystals nucleated
Nucleation rate (# cm3 h1)a 109 109 1012 can then be determined. The result (Fig. 30) provides a
Induction period (h) 0 5 15 rare demonstration of the way in which both nucleation
Simulation data and growth rates vary as a function of the alkalinity of
Time to reach 100% yield, tf (h) 20 15 17 the reaction mixture. The maximum in the growth curve
Yield at 0.25 tf (%) 1.6 0 0 corresponds to a linear growth rate (0.5Dl/Dt) at 95 C
Yield at 0.50 tf (%) 12.5 1.8 0 of about 2.3 · 102 lm h1. This agrees well with values
Yield at 0.75 tf (%) 42 25 0
of around 2 · 102 lm h1 obtained in more detailed
Crystallisation rate (% h1)b 9.6 19.5 100
Crystal size (lm) 4 4 0.4 studies [142,188]. Interestingly, the maxima for the
a growth and nucleation curves are not coincident, that
Applied during first hour of reaction only.
b
Gradient at 50% conversion. for nucleation occurring at a considerably higher base
level. This is consistent with established practice for
the formulation of ‘‘nucleating mixtures’’ (Section
their number per unit volume. The data are not specific 11.1), which are almost invariably made up at a higher
to any particular system but are chosen to be reasonably pH than that of the final reaction composition. As a
realistic and could, for example, reflect compositional rationalisation, nucleation may be considered in terms
variations in MFI-type syntheses at around 100 C. Fol- of a collision mechanism in which the constituent frag-
lowing an initial single nucleation event, the monodis- ments are constantly coming together and separating
perse crystal population is allowed to grow at the in a dynamic process. In this view, the nucleation rate
stated rate up to a nominal 100% conversion of starting will be proportional to the flux of solution species, which
materials. It can be seen that the three simulations illus- itself will increase with alkalinity. (This argument is still
trated have similar total reaction times (15–20 h) but valid for the more likely case (Section 7.7) in which nu-
that their growth profiles are completely different. clei are assembled within colloidal particles, since the ex-
Depending on the criterion chosen (e.g. yield at t1/2, change with solution species which facilitates the
total reaction time, crystallisation rate), each one of necessary rearrangement reactions will be related to
the three could be claimed as the fastest synthesis! the flux of such species.) At low pH, the deficit of
Most of the ambiguities can be overcome by specify- OH-related units leads to reduced nucleation and
ing (a) some form of growth curve (determined from the growth [unless other types of mass carriers (e.g. fluoride
evolution with time of a specific product property, e.g. species) are present], whilst the maximum of the growth
XRD crystallinity, micropore volume, characteristic IR curve presumably signals the optimum balance between
bands, or reaction pH) and (b) some form of reliable
product crystal size distribution (e.g. from a calibrated
SEM). From these basic data, an estimate can be made
of (i) how long the reaction took to start, (ii) when the
first crystals appeared, (iii) the time at which crystallisa-
tion stopped and (iv) how to interpret the growth period
in the light of the likely nucleation profile. Good exam-
ples of the detailed application of this type of detective
work can be found in the application of the Zhdanov
method for the determination of growth rates and nucle-
ation profiles [57,140–143] (Section 7.3.1). However,
useful information can also be derived in other ways.
In some cases, new light can be shed on existing experi-
mental data.
In a paper by Fegan and Lowe [219], the behaviour of
the silicalite synthesis system xNa2O : 2TPABr : 20-
SiO2 : 1000H2O at 95 C is determined as a function of
alkalinity (x is varied from 0.25 to 6.5). The published
results are given primarily in the form of (a) XRD crys-
tallinity and pH curves compiled from samples taken
Fig. 30. Effect of alkalinity on silicalite synthesis in the system
during the course of the reactions and (b) SEM measure- xNa2O : 20SiO2 : 2TPABr : 1000H2O at 95 C, calculated from data
ments of the final products. On the basis of these results, given in Ref. [219]. The maxima estimated for growth rate (solid curve)
and guided by comparisons with similar reactions car- and nucleation rate (dashed curve) are not coincident.
competing addition and dissolution reactions. Both systems which are reluctant to nucleate, for controlling
nucleation and growth will fall away as the dissolution crystal size and for reducing the formation of crystalline
limit is approached at the highest value of x. impurity phases. It follows that the method is likely to
have a major influence on reaction kinetics. Further dis-
17.2. Procedures for improvement cussion will be found in Section 7.4.
17.2.1. Reaction optimisation (and its limitations) 17.2.3. Additives and ‘‘promoters’’
A standard method for the improvement of a zeolite In cases where a synthesis system has been investi-
synthesis is to test the sensitivity of the reaction to vari- gated over a range of conditions, the usual mode of ac-
ations in the synthesis parameters and so optimise the tion of an additive is to reduce the crystal linear growth
procedure. A very large number of publications are cen- rate. For MFI systems, this has been seen for the addi-
tred around such a strategy. The time profile of a synthe- tion of competing organic cations or amines (e.g. TMA
sis reaction will be influenced by the reaction variables [188,479], triethylenetetramine [479]), charge-carrying
(temperature, reaction composition, mass transfer ef- heteroelements such as aluminium [188,479] and iron
fects, etc.). As conditions vary (at a fixed temperature), [479], complexing agents such as citric acid [479] and dil-
a reaction may appear to occur ‘‘faster’’ (progress to- uents such as ethanol [188]. In the limit, very small
wards an optimum) or ‘‘slower’’ (movement away from amounts of certain reaction components (e.g. hexame-
an optimum) depending upon the starting point and thonium cations) appear to block MFI structure forma-
the criteria applied. Detailed measurements provide an tion completely [155,156] (Section 13.3).
objective assessment of the changes taking place. Exam- An unusual case is that of oxyanions such as perchlo-
ples of this have been given earlier (Section 8.3), as in the rate, phosphate and bromate, which are claimed
inverse relation of faujasite linear growth rate to Si/Al ra- [480,481] to act as ‘‘promoters for enhancing nucleation
tio shown by Kacirek and Lechert [79,147] and the direct and crystallisation’’. Data are quoted mostly as crystal-
correlation of linear growth rate with pH for fluoride-sil- lisation times, although some crystallisation curves and
icalite described by Tavolaro et al. [205]. particle sizes are given. The additives (P) are included
Let us now assume that a synthesis system has been in significant quantities (usually P/SiO2 = 0.1) and many
investigated under a series of experimental conditions. are such as would influence the reaction pH (e.g. HClO4,
Once conditions for phase purity have been established, NaHCO3, Na2HPO4). It is not clear to what extent
there may also be scope for variation of the balance be- allowance has been made for this. Reductions in synthe-
tween nucleation and growth, giving rise to changes in sis times to 20–60% of the original values are reported.
overall synthesis time and particle size (e.g. as illustrated The effect is rationalised in terms of polarising ability
by Table 4). However, these phenomena are themselves and an inverse correlation between the synthesis time
subject to limitations. There is evidence, at least for the and the charge/radius ratio (Z/r) of the central cation
MFI system, that the number of crystals nucleated by a of the promoter oxyanion. Kumar et al. describe
given reaction composition may be subject to an upper [480,481] a linear decrease in crystallisation time up to
limit [150,287,355–357]. For each zeolite system, there an optimum additive concentration, after which the time
also appears to be a maximum value for the linear remained the same. Luo and co-workers [482] have
growth rate at a given temperature. It has been suggested reported a different pattern on adding sodium
(Section 8.2) that this maximum value represents a limit phosphate (P = Na3PO4 Æ 12H2O) to the synthesis
set by the surface integration step of crystal growth itself, mixture for the hexagonal faujasite phase EMT. Using
all other possible resistances (e.g. limitations by nutrient a Na2O/18-crown-6 ratio of 5, the crystallisation time
dissolution rate or mass transfer) being at a minimum. at 100 C was found to be 30 days. With the composi-
Assuming no fundamental change in growth mechanism, tion 2.1Na2O : 10SiO2 : Al2O3 : 0.33(18-crown-6) : zP :
further acceleration of the growth rate would only be 140H2O, the times were 25 days (z = 0.2), 20 days (z =
possible if this rate-limiting step could itself be catalysed 0.1), 6 days (z = 0.05), 8 days (z = 0.03) and 12 days
in some way, a feat which has so far proved elusive. (z = 0.02). This unusual result is ascribed to the phos-
Therefore, any procedure which appears to accelerate phate acting as an emulsifying and water-structure-
the overall rate of synthesis must be acting either by (i) breaking agent. In all these investigations, there is no
shortening the induction period, (ii) moving a non-ideal indication of reproducibility nor how other changes of
system towards its optimum, or (iii) altering the balance conditions (optimisation) might affect the outcome. It
between nucleation and growth. is clear that any conclusions on the effects reported must
await more detailed studies.
17.2.2. Addition of seed crystals
The use of crystal seeding to a large extent by-passes 17.2.4. Microwave synthesis
normal nucleation requirements and can be a very A technique which has been shown to offer significant
effective method for encouraging product formation in improvements in a variety of chemical syntheses is that
of microwave dielectric heating [483–487]. The benefits purity requirements. Shortening of synthesis times can
to zeolite synthesis have been discussed in a review often be achieved through seeding procedures and by
[305]. In conventional (thermal) heating, reactions may the use of microwave heating. However, any claims
be strongly influenced by the rate of heat transfer for chemical ‘‘promoters’’ should be viewed with
through the walls of the containing vessel and the result- caution.
ing thermal gradients across the reaction volume. These
limitations are modified when a different form of energy
input is applied. The most apparent advantage of micro- 18. Relationship of zeolite synthesis mechanism to
wave heating is the significant shortening of reaction that of other porous materials
time which is found in virtually every case. The factors
underlying the reduced timescale have been analysed In order to keep this survey within reasonable
[151,305] in terms of the contributions from (i) thermal bounds, the subject matter has been restricted to the
lag, (ii) the induction period and (iii) crystal growth. field of aluminosilicate zeolites and their related zeolitic
In summary, the major accelerations derive from (i) silica polymorphs. However, it is necessary to place this
and (ii) by virtue of the rapid heating rate to the working work in perspective by making some reference to the
temperature and from non-equilibrium (e.g. differential synthesis of other porous materials. The most significant
heating) contributions to equilibration reactions and distinctions relate to the different assembly mechanisms
the nucleation process. Under pseudo-steady state con- and are reflected in differences in rates of formation.
ditions, there is no difference between crystal linear It has already been pointed out (Section 8.2) that the
growth rates for comparable thermal- and microwave- crystal linear growth rates of zeolites are several orders
heated preparations [150,152]. of magnitude lower than those of simple inorganic salts.
A further potential advantage of microwave heating In the latter case, the building blocks of the crystal are
is that of selectivity, in cases where some component already present in solution (e.g. Na+ and Cl ions for
of the system is selectively sensitive to microwave energy common salt), so that the process of crystal assembly
input, or at least to heating rate. Such a situation might is relatively facile, depending on a sequence of (i) migra-
arise in a heterogeneous reaction mixture containing tion to the growth site, (ii) desolvation and (iii) surface
components of different dielectric properties, or through integration. Depending on the circumstances, any one
differences in the rival rates of nucleation and growth of of these (fairly comparable) steps might be rate limiting.
competing phases. Probable instances of the former are Although the same general considerations apply to zeo-
provided by the synergistic effect of microwave heating lites, there is a major difference in step (iii), which
and nanocrystal seeding in MFI synthesis [151] and by involves the making of covalent bonds as the key oper-
the failure of zeolite formation in a ‘‘dry’’ system ation in the in-situ construction of a polymeric inorganic
through immobilisation of key reagents through differ- structure. This is a relatively slow process and is usually
ential heating [304]. Phase selectivity in zeolite Y synthe- rate controlling, the main exceptions deriving from the
sis has been demonstrated by the suppression of use of unreactive starting materials (dissolution limita-
undesired co-products in microwave-mediated preparation) [210] or the controlled addition of nutrients (sup-
tions [488], even under conditions (150 C) where zeolite ply limitation) [148,186].
P would normally dominate in a thermal synthesis [489].
Microwave heating thus offers an alternative to 18.1. Zeolites and clathrate hydrates
conventional thermal methods in zeolite synthesis. It is
often more convenient and may offer practical advanta- The extremes of mechanism and rate noted above
ges (e.g. in the synthesis of surface films [490]). A signif- are exemplified in the contrast between the syntheses
icant reduction in overall synthesis time is usually of members of the zeolite family [48] and those of
observed and there may be the possibility of selectivity the water-soluble, complex-cation silicate hydrates
in heat input or phase purity. However, there is no [491]. The former require hydrothermal syntheses at
evidence for special (i.e. athermal) ‘‘microwave effects’’. elevated temperature for hours or days, whereas the
latter can be rapidly crystallised from aqueous solu-
17.3. Summary tions under ambient conditions. An interesting com-
parison is that between the A-type zeolite ZK-4
In balancing the respective merits of different zeolite [12,13] and the tetramethylammonium octasilicate hy-
syntheses, it is important that realistic and sufficient drate [Me4N]8[Si8O20] Æ 65H2O [492]. The synthesis
criteria are chosen for the comparisons. Some form mixtures will in both cases contain a distribution of
of quantitative reaction profile and product size mea- species, including both the TMA cation and the cubic
surement are essential. Changes in reaction parameters octasilicate anion [116,493,494]. Both materials can be
should normally permit some measure of optimisation, considered as being constructed entirely from D4R
although the scope for change may be limited by phase silicate or aluminosilicate rings associated with TMA
cations (and additional sodium cations in the case of

ZK-4). However, in the case of the zeolite, the D4R
rings are bound covalently in an infinite three-dimen-
sional network with the TMA cations locked into
sodalite cages. The silicate hydrate is also a host-guest
compound but a far more labile one. The host struc-
ture consists of a three-dimensional heteronetwork
made up of cubic octasilicate anions and water mole-
cules linked via hydrogen bonds, the guest TMA cat-
ions being sited within large, irregular cage-like
voids. This polyhydrate melts incongruently at 72 C
to form a liquid phase and a second crystalline hetero-
network clathrate, the host structure of which pos-
sesses the topology of the zeolite structure type AST.
This therefore provides an even more direct compari-
son between two materials having similar topologies
but different bonding regimes, namely the fragile ionic
and H-bonded network of the AST-type clathrate hy-
drate and the robust, covalent framework of the ‘‘iso-
structural’’ AST-zeolite.
Related double-four-ring units have recently formed
the basis of a synthesis strategy discussed by Villaescusa
et al. [495]. Using a variety of quaternary cations (Q+), it
is found that compounds of the type Q+[Ge8O12-
(OH)8F] are formed under mild conditions. Within
every cube resides a fluoride ion, giving each germanate Fig. 31. The D4R unit and one of the interrupted germanate zeolite
unit its negative charge. At higher temperatures and frameworks containing structurally related components [495]: (a) the
longer reaction times, three-dimensional networks are Ge8O12(OH)8F anion, (b) layer B of the STAG-1 structure (in the a, b
plane) showing one of each of the two SBUs.
formed. Since these products, some with very open
frameworks (Fig. 31), contain complete or ring-opened
D4R units linked together in various ways, the authors 18.2. Zeotypes
have adopted the premise of pre-nucleation building
units (PNBUs) [496,497], suggesting that the materials The many different ideas on synthesis mechanism ex-
have been formed by condensation of such units pre- pressed for zeolites (Section 3) have also been applied to
existing in the solution. The respective merits of this the synthesis of zeotype materials. However, as men-
hypothesis and possible alternative explanations have tioned above (Section 18.1), there has been far greater
been discussed earlier (Section 9) and are revisited below emphasis on the view of the crystallisation process as
(Section 18.2). The effect of Ge-substitution in altering an assembly of pre-fabricated units [496,497]. Complex
relative stabilities and thus opening pathways to new models of zeotype formation by linkage of one- to
structures has been noted earlier (Section 13.4). three-dimensional precursor units have been proposed
Outside the realm of zeotypes and in a rather different [316,501,502]. Based largely on NMR evidence of solu-
system, cubic Si8O20 building blocks have been cross- tion species and the commonality of similar structural
linked into a porous vanadosilicate solid matrix by units in the products, a recent review of metallophos-
reacting the metallated (Bu3Sn)8Si8O20 with VOCl3 phate formation discusses a synthesis mechanism in
[498]. In this case, there is no doubt about the direct par- which significant elements of the structure exist in solu-
ticipation of the D4R precursor. However, the chemistry tion as identifiable building blocks before being
of this non-aqueous route is completely different from ‘‘clipped’’ together to from the precipitated product
that of the hydrothermal synthesis, being in essence an [503]. However, there seems no reason at present to re-
alkoxide-transfer process brought about by a specific ject the alternative possibility (Section 9.1) that much
organotin reagent. In a similar context, it is worth not- of this chemistry goes on at the liquid–solid interfacial
ing the existence of some structural parallels between growth points rather than (as is implied) independently
zeolites and certain coordination polymers. Although of the growing crystal. A suggestion for the detailed
far removed in composition and chemical properties, crystallisation mechanism of AlPO4-21 [504] could again
some self-assembled metal complexes share a common be reinterpreted in this manner.
topology with microporous aluminosilicates such as zeo- It is undeniable, however, that many zeotype systems
lite A and sodalite [499,500]. behave very differently from their zeolite counterparts.
One of the most striking demonstrations of this eral ‘‘cooperative assembly’’ model by Stucky and co-
concerns the framework zinc phosphates, which can be workers [514,515]. This emphasises the role of specific
crystallised very rapidly under very mild conditions interactions between surfactant headgroups and inor-
[505–509], suggesting a more nearly ionic assembly ganic species and identifies three key factors: (i) multid-
route. For example, the sodalite analogue Na6(Zn- entate binding of silicate oligomers to the cationic
PO4)6 Æ 8H2O can be made from a mixture of ZnO, surfactant, (ii) preferential silicate polymerisation in
H3PO4, NaOH and water in 24 h at room temperature the interface region and (iii) charge density matching
[506], whilst Na,TMA-ZnPO4-faujasite can be synthes- between the surfactant and the silicate.
ised in 30 min at 4 C [509]. There is also evidence that It is interesting to compare the basic stages in the syn-
some AlPO-type products may be much more soluble thesis of microporous and mesoporous materials and to
in their mother liquors than is the case for zeolites, in note the differences in the kinetic limitations. Firouzi
some cases exhibiting a retrograde temperature solubil- et al. have pointed out the importance in mesophase
ity which causes a marked reduction in yield unless the synthesis of separating the effects of self-assembly from
product is filtered hot [510]. Even further along the the kinetics of silicate polymerisation [514]. For
polarity scale are the inorganic–organic hybrid materials MCM-41, the spontaneous hexagonal ordering of micel-
[511] such as the complex metal carboxylates. In an lar rods on encapsulation by silicate ions was demon-
intriguing recent report [512], a single reaction mixture strated by 14 N NMR [513] and EPR [516], whilst the
produced five different cobalt succinates, the structure latter also indicated the hardening of the inorganic
obtained depending only upon the reaction temperature. phase in a second step as the silicate ions polymerised
As the temperature was increased from 60 to 250 C, the at the interface. In an in-situ ATR FTIR study of
products became less hydrated and more dense, the MCM-41 formation by Holmes et al. [517], slow dissolu-
structures changing from chains to sheets to frameworks tion of the silica source highlighted the relative rates of
across the series. subsequent reactions. Under these conditions, the co-
It seems unlikely that any completely new concepts operative assembly of the embryonic MCM-41 structure
will be necessary to explain the formation patterns of occurred rapidly, with further crosslinking through con-
zeotypes. However, just as the range of behaviour ob- densation reactions being clearly identified as the slow
served for zeolites requires a flexible (but coherent) step in the main synthesis sequence. These observations
mechanistic scheme, this spectrum will need to be fur- allow some parallels to be drawn between the syntheses
ther extended in order to allow for the differences in of two types of ordered, porous silica polymorphs such
composition, structure, polarity and solution chemistry as (for example) the zeolitic silicalite and the mesopor-
found in zeotype synthesis. It should also be noted that ous MCM-41 (Fig. 32).
the structures of many of the as-made zeolite-related The initial silica dissolution step (1) is not normally
materials synthesised under mild conditions do not sur- rate-limiting in zeolite synthesis, although it may be-
vive calcination to remove the organic component. They come so for very unreactive sources of silica [210]. How-
thus have no microporosity and should be regarded as ever, in MCM-41 synthesis, even the dissolution of a
inorganic–organic complexes having zeolite-like topolo- reactive silica source such as fumed silica may control
gies rather than true zeotypes. the initial reaction rate since the formation of the initial
mesophase complex (IM) is so rapid (step 2M) [517].
18.3. Microporous vs. mesoporous structures This complex (IM) possesses the overall hexagonal sym-
metry of the final hydrothermal product (IIM) but is less
The classic synthesis of mesoporous materials em- crosslinked (i.e. higher Q3/Q4 ratio) and less well or-
ploys cationic surfactant molecules to generate the char- dered. In a further step (3M) under hydrothermal condi-
acteristic structure of the products. For example, RMe3 tions, the condensation reactions necessary to crosslink
N+ cations (R = CnH2n+1, n = 8–16) are used in the syn- the structure are slowly brought about as IM is con-
thesis of the MCM-41 and MCM-48 members of the verted into IIM. The final crosslinking is completed as
M41S family [24,25]. Initially, two possible formation the surfactant template is lost on calcination (step 4M).
pathways were proposed by the Mobil group [25]. In The zeolite equivalent of the MCM-41 precursor
the first, silicate species interact with a pre-existing li- (IM), complex IZ, is formed to a greater or lesser extent,
quid crystal phase to form the composite mesophase. depending on the reaction conditions, and is the second-
The alternative model is based on the cooperative inter- ary amorphous phase discussed earlier (Section 6.2). In
action of inorganic species with surfactant micelles in contrast to the preparation of the mesophase, the hydro-
solution to generate the ordered composite structure. thermal crystallisation step (3Z) is usually the rate-
Since most syntheses are carried out at surfactant con- limiting reaction in zeolite synthesis (Section 8.2).
centrations well below those necessary for the formation Compounds I and II are in both cases related by their
of liquid crystalline phases [513], the second route is the relative degrees of order. For the zeolite, complex IZ is
more probable and has been developed into a more gen- likely to have roughly the same composition as the
ions and cations which already exist in the synthesis

SiO2 source solution. It seems that aluminosilicate zeolites fall lar-
gely into the first category whilst at least some of the
aluminophosphates display a decidedly more polar char-
step 1 OH – acter. Such a wide spectrum of reactivity suggests differ-
ences also in mechanism. Thus, while the concept of
Silicate species pre-nucleation building units probably does not apply
in solution to the situation of zeolite synthesis, it may prove to be
more relevant to the assembly mode of some zeotypes.
A more fundamental difference exists between the
construction mode of zeolite-family members and that
2Z Q+ S+ 2M
of mesoporous materials. This is largely a case of syn-
thesis being the mirror of structure since the mesoporous
Organic-inorganic
OOOooooOrganic Initial mesophase materials are non-crystalline. Zeolites do not develop
I complex complex
IZ IM their most characteristic properties until the crystal has
been formed and this is usually a slow process. Con-
versely, mesoporous materials (or at least their pore-
3Z hydrothermal hydrothermal 3M filled precursors) derive their characteristic periodic
structure from the assembly of a surfactant-silicate (or
Crystalline organic- Crosslinked mesophase other surfactant-oxide) complex. Since this does not ini-
inorganic complex structure tially involve the formation of covalent bonds, it can oc-
II Z II M cur very rapidly.
4Z calcination calcination 4M
19. Summary and conclusions
Zeolite (crystalline Ordered mesoporous
silica polymorph) silica The objective of this review has been to provide an in-
III Z III M sight into the hidden machinery underlying the overt
experimental procedures of zeolite synthesis. As with
Fig. 32. Schematic comparison between the synthesis of a crystalline most practical enterprises, it is not necessary to under-
zeolitic polymorph (e.g. silicalite) (Z) and that of an ordered stand the machinery in order to operate the process—
mesoporous silica (e.g. MCM-48) (M), using a quaternary ammonium
but such expediency is unlikely to provide a future for
template (Q+) or surfactant (S+) respectively. Reaction steps are shown
in Arabic numerals and reaction intermediates by Roman numerals. the technology! No attempt will be made in this conclud-
ing section to provide a complete digest of all that has
gone before. Rather, a few of the most salient points will
derived zeolite (IIZ), but has only local order. Material be revisited and their importance summarised.
IIZ is a crystalline compound with fully developed long- The overall requirements for the synthesis process to
range periodicity. The mesoporous intermediates IM operate can be outlined in a simple scheme (Fig. 33).
and IIM differ to a much smaller extent, sharing similar This highlights the fundamental importance of mobility,
composition and status as ordered amorphous solids. the generation of which is the basic function of the so-
The extent of ordering increases in the sequence IM < called mineralising component in zeolite synthesis. Such
IIM < IIIM as the degree of silicate condensation a component, most usually (but not exclusively)
becomes greater. For the zeolite, the sequence is
IZ IIZ IIIZ. Compounds IIZ and IIIZ are both
crystalline and differ essentially only in composition. AX– anions →
crystallisation,
release of X–
18.4. Summary
REACTANTS ZEOLITE
The syntheses of zeolite-related crystalline com- “A” PRODUCT
pounds can be considered as being on a continuous scale
between the extremes of quartz-like and salt-like materi- attack
by X–
als. In the first case there is the slow, in-situ assembly ←X –
under autoclave conditions of a largely covalent poly-
Fig. 33. Simplified scheme illustrating the cyclic role of the mineraliser
meric structure from simpler building blocks. In the lat- X in zeolite synthesis (where X may be, for example, OH or F). In
ter situation, there is a more rapid precipitation under many cases, some X anions or their derivatives (e.g. AlðOHÞ 4 ) will be
mild conditions of an essentially ionic structure from an- retained in the product.
hydroxide ion, converts the reactants into mobile and the discreet step of nucleation, at which point the peri-
reactive forms, thus enabling the synthesis to proceed. odic crystal lattice is energetically established and subse-
Acting in some ways as a catalyst, much of the mineral- quently able to propagate freely.
iser is released at the end of the cycle. Recent work has Much of the published discussion on the mechanism
provided new perspectives on some of the components of zeolite synthesis has been concerned with location,
within this overall scheme. The induction period, by nat- and in particular as to whether key events occur in the
ure and definition the least visibly active stage of the solid phase or in solution. It is hoped that the evidence
synthesis process, paradoxically has been shown to and reasoning presented in this survey have demon-
incorporate all the main steps in zeolite formation, since strated that this is a sterile and unnecessary argument.
it covers the transition between totally amorphous mate- Once the overall equilibration mechanism is accepted,
rial and the establishment of crystallinity. Also, the rate it can be seen that the differing individual situations
limiting step for layer-by-layer crystal growth in zeolites are all variations upon a theme, the exact situation
has for some time been identified as the chemical reac- depending upon the experimental circumstances, and
tion in which growth units are integrated into the crystal in particular upon reaction composition and concentra-
surface. Experimental and modelling studies have now tion. Thus, the two extremes of ‘‘solution-mediated’’
allowed a further dissection of the growth mechanism, and ‘‘gel rearrangement’’ mechanism mentioned at var-
revealing the importance of the surface nucleation step: ious points (e.g. Sections 3.6 and 10) and depicted by
this provides the dominant rate limitation within the Caullet and Guth [38,518] may be replaced by the more
overall growth process, the lateral spreading of a nucle- realistic and generally applicable scheme shown in Fig.
ated layer being relatively rapid. 34. This mechanism should apply universally to all
It appears increasingly probable that the mobile units hydrothermal zeolite syntheses, with a minimum of
responsible for structural evolution and crystal growth modification (for example, in the case where the starting
are small, mobile solution species, with aggregation pro- material is itself crystalline, as in a zeolite-to-zeolite
cesses significant only at very small particle sizes. The transformation).
most likely growth mechanism functions through an In earlier sections, we have described the progress in
in-situ, localised construction process in which the understanding which is being brought about by develop-
building units are ordered and assembled at the growth ments in both theoretical and experimental techniques.
site by the participating cations. By acting as coordina- Notable practical advances include the proliferation of
tion centres for water molecules, silicate anions and new families of materials and consequent discovery of
other polar units, these provide much of the blueprint novel types of frameworks. Amongst instrumental meth-
for the spatial architecture of the developing structure ods, the potential of HRTEM and AFM is outstanding
(Fig. 11). The overall process at a growth point on a for the elucidation of structural and surface detail. The
crystal surface is in some ways analogous to that of continuing development of environmental cells will
weaving or knitting. The different building materials transform microscopy into a technique for the investi-
(separate lines of thread) are taken in, assembled gation of synthesis mechanism. Computational ap-
(woven) according to a definite plan (fabric pattern) into proaches should more and more facilitate the
the finished article (cloth) which issues from the assem- investigation of experimental situations which would
bly point (loom) as a continuous network. be difficult or impossible to realise in the laboratory.
Whilst the myriad examples of synthesis reactions Overall, it is hoped to have shown in this survey that,
display a broad spectrum of diversity, it can be helpful although the extent of knowledge in the realm of hydro-
to consider the similarities rather than the differences. thermal zeolite synthesis is far from complete, a sound
The apparent complexity should not be allowed to ob- scientific basis for most of the mechanistic details is al-
scure the pervading influence of a small number of key ready well established.
reactions which enable the entire process to function. To conclude, we speculate on a possible new ap-
Predominant amongst these are the TAOAT bond-mak- proach to the understanding of zeolite formation. Start-
ing and bond-breaking reactions: ing from the biological viewpoint of the very precise
transcription mechanism for protein manufacture within
TAOH þ OAT TAOAT þ OH
a living cell, the question arises as to the nature of the
TAOH þ HOAT TAOAT þ H2 O equivalent (but far more diffuse) coding for a particular
phase within the realm of zeolite synthesis. In a recent
These reactions establish the equilibration between solid review, Corma and Davis [43] point out that ‘‘. . . all
and solution components which is fundamental to the the intelligence must be contained in the starting re-
ordering process of zeolite synthesis (Fig. 5). This pro- agents to allow them to self-assemble properly in high
cess is essentially one of continuous evolution, driven yield.’’ Such intelligence takes the form not only of the
by energy differences and powerfully moderated by ki- chemical compounds but also their physical form. The
netic limitations. The single discontinuity is found in encapsulated energetic and kinetic ‘‘programme’’ also
Fig. 34. A generalised mechanism for zeolite synthesis. A fragment or domain of amorphous material (a) equilibrates with solution species (anions
and cations) to develop elements of local order (b). In due course, the equilibration process leads to an area of sufficient order for a periodic structure
to become established—i.e. nucleation has occurred (c). The same equilibration reactions (TAOAT bond-making and bond-breaking) then allow the
nascent crystal to grow and the amorphous areas to dissolve (d). The self-assembly process is mediated by the associated solvated cations, which act
as coordination centres (templates) for the construction of the framework (central insert). These transformations most usually take place via a bulk
solution phase, but may occur within a solvated layer at the surface of a ‘‘dry’’ solid (apparent solid-phase transformation).
has external input from variables such as temperature Chemistry in providing partial support. They are also
and degree of mixing. These larger instruments will indebted to the authors of cited papers for their permis-
eventually bear upon the detailed mechanism of the sion to reproduce original diagrams. Thanks are due
assembly process itself through the medium of ionic also to R.W. Thompson for preprints of forthcoming
and intermolecular forces. Amongst this complexity lies work and to the family of the late Dr B.M. Lowe for
the critical information. Some individual correlations allowing the use of unpublished material.
are readily demonstrated—for example in the work of
Zones et al. [519] who show in a ‘‘pyramid’’ plot that
there is an inverse relationship between the size and Appendix A. A chemical model for the crystal growth of
complexity of an organo-cation template and the range zeolite molecular sieves
of zeolites that it will synthesise. However, this is only
one aspect of a much larger problem. A solution may B.M. Lowe (University of Edinburgh), C.M. Lowe
lie nearer to the emerging discipline of cheminformatics
rather than to the traditional approach of the chemist. A summary of the paper presented at the 13th Annual
Meeting of the British Zeolite Association, Chislehurst,
July, 1990.
Acknowledgements
The late Dr Barrie Lowe was prevented by illness
The authors would like to acknowledge the helpful from publishing or extending this preliminary computa-
contributions of J.R. Agger and colleagues in improving tional study. It was carried out using one of the earliest
the manuscript and of the UMIST Department of types of personal computer available in the UK, the
BBC home computer. A summary is given here since the [4] R.M. Barrer, C. Marcilly, J. Chem. Soc. (A) (1970) 2735.
present authors are aware of no subsequent work which [5] Ch. Baerlocher, W.M. Meier, D.H. Olson, Atlas of Zeolite
Structure Types, fifth ed., Elsevier, Amsterdam, 2001, p. 156, and
similarly attempts to describe zeolite crystal growth in references therein.
chemically significant terms. Regrettably, no copy of [6] G.T. Kerr, Science 140 (1963) 1412.
the original source code has survived and this account [7] G.T. Kerr, Inorg. Chem. 5 (1966) 1539.
has been compiled from the material presented in the [8] R.M. Milton, US Patent 2,882,243 (1959).
1990 lecture. [9] R.M. Milton, US Patent 2,882,244 (1959).
[10] R.M. Milton, in: M.L. Occelli, H.E. Robson (Eds.), Zeolite
The computer program started by initiating the Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 1.
growth of a predetermined two-dimensional layer, e.g. [11] R.M. Barrer, P.J. Denny, J. Chem. Soc. (1961) 971.
a hexagonal network or a layer from a known zeolite [12] G.T. Kerr, G.T. Kokotailo, J. Am. Chem. Soc. 83 (1961) 4675.
structure. A growth unit such as monomer, dimer, or [13] G.T. Kerr, Inorg. Chem. 5 (1966) 1537.
cyclic pentamer was then chosen and the layer allowed [14] R.L. Wadlinger, G.T. Kerr, E.J. Rosinski, US Patent 3,308,069
(1967).
to grow from the designated unit, selecting growth sites [15] R.J. Argauer, G.R. Landolt, US Patent 3,702,886 (1972).
either at random from all possible sites (‘‘unsecured’’ [16] Ch. Baerlocher, W.M. Meier, D.H. Olson, Atlas of Zeolite
growth) or at random from sites which gave the maxi- Structure Types, fifth ed., Elsevier, Amsterdam, 2001.
mum bonding to the existing growth layer (‘‘secured’’ [17] R. Szostak, Molecular Sieves—Principles of Synthesis and
growth). Growth layer T-atoms attached to only one Identification, Van Nostrand Reinhold, New York, 1989, second
ed., Blackie, London, 1998.
or two others were allowed to move or become [18] S.T. Wilson, B.M. Lok, C.A. Messina, T.R. Cannan, E.M.
detached. Flanigen, J. Am. Chem. Soc. 104 (1982) 1146.
After a sufficient number of unit additions (usually [19] E.M. Flanigen, R.L. Patton, S.T. Wilson, in: P.J. Grobet, W.J.
3000), growth was stopped and the results (displayed Mortier, E.F. Vansant, G. Schulz-Ekloff (Eds.), Innovation in
as a network) analysed. Growth layer T-atoms attached Zeolite Materials Science, Studies in Surface Science and
Catalysis, vol. 37, Elsevier, Amsterdam, 1988, p. 13.
to three others were defined as ‘‘crystalline’’ (number [20] J.A. Martens, P.A. Jacobs, in: J.C. Jansen, M. Stöcker, H.G.
NX) and T-atoms with fewer lattice bonds as ‘‘amor- Karge, J. Weitkamp (Eds.), Advanced Zeolite Science and
phous’’ (number NA). The function k ¼ N A =ðN X Þ1=2 Applications, Studies in Surface Science and Catalysis, vol. 85,
was found to be characteristic of the growth layer type Elsevier, Amsterdam, 1994, p. 653.
chosen (linearly proportional to the framework density) [21] S.M. Kuznicki, US Patents 4,853,202 (1989), 4,938,939 (1990).
[22] M.W. Anderson, O. Terasaki, T. Ohsuna, A. Philippou, S.P.
and always greater for unsecured growth. Particularly MacKay, A. Ferreira, J. Rocha, S. Lidin, Nature 367 (1994) 347.
interesting was the defect analysis. The number of de- [23] J. Rocha, M.W. Anderson, Eur. J. Inorg. Chem. (2000) 801.
fects increased with the size and complexity of the cho- [24] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S.
sen growth unit and was always much larger for Beck, Nature 359 (1992) 710.
unsecured growth. Growing the structure with a mixture [25] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T.
Kresge, K.D. Schmitt, C.T.-W. Chu, D.H. Olson, E.W. Shepp-
of a complex unit (e.g. cyclic pentamer) and monomer ard, S.B. McCullen, J.B. Higgins, J.L. Schlenker, J. Am. Chem.
substantially reduced the number of defects. All defects Soc. 114 (1992) 10834.
could be healed by further growth with monomer only. [26] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Frederickson, B.F.
The conclusions (given here in full) were: Chmelka, G.D. Stucky, Science 279 (1998) 548.
[27] M. Kruk, M. Jaroniec, S.H. Joo, R. Ryoo, J. Phys. Chem. B 107
(2003) 2205.
1. The growth layer of a molecular sieve has a signifi- [28] A. Corma, Chem. Rev. 97 (1997) 2373.
cant amount of amorphous character even when [29] U. Ciesla, F. Schüth, Micropor. Mesopor. Mater. 27 (1999) 131.
monomeric growth units are used. [30] F. Di Renzo, A. Galarneau, P. Trens, F. Fajula, in: F. Schüth,
2. Monomer is a suitable growth unit for all framework K.S.W. Sing, J. Weitkamp (Eds.), Handbook of Porous Solids,
structures. vol. 3, Wiley-VCH, Weinheim, 2002, p. 1311.
[31] C.S. Cundy, P.A. Cox, Chem. Rev. 103 (2003) 663.
3. Larger growth units are more likely to produce mate- [32] D.W. Breck, Zeolite Molecular Sieves, Wiley, New York, 1974.
rial with defects. The more defects, the less likely the [33] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic
crystal is to grow. Press, London, 1982.
4. Larger units may attach to the growth layer in unsuit- [34] H. van Bekkum, E.M. Flanigen, P.A. Jacobs (second ed. only),
able places. This slows down the growth process until J.C. Jansen (Eds.), Introduction to Zeolite Science and Practice,
Studies in Surface Science and Catalysis, first ed., vol. 58,
the unit is detached. Elsevier, Amsterdam, 1991, second ed., vol. 137, 2001.
[35] J.C. Jansen, M. Stöcker, H.G. Karge, J. Weitkamp (Eds.),
Advanced Zeolite Science and Applications, Studies in Surface
Science and Catalysis, vol. 85, Elsevier, Amsterdam, 1994.
References [36] M.E. Davis, R.F. Lobo, Chem. Mater. 4 (1992) 756.
[37] R.W. Thompson, in: H.G. Karge, J. Weitkamp (Eds.), Molec-
[1] H. de St Claire Deville, Compt. Rend. Séances Acad. Sci. 54 ular Sieves, vol. 1, Springer-Verlag, Berlin, 1998, p. 1.
(1862) 324. [38] J.-L. Guth, H. Kessler, in: J. Weitkamp, L. Puppe (Eds.),
[2] R.M. Barrer, J. Chem. Soc. (1948) 127. Catalysis and Zeolites, Springer-Verlag, Berlin, 1999, p. 1.
[3] R.M. Barrer, L. Hinds, E.A. White, J. Chem. Soc. (1953) 1466. [39] D.P. Serrano, R. van Grieken, J. Mater. Chem. 11 (2001) 2391.
[40] K.J. Balkus, Progr. Inorg. Chem. 50 (2001) 217. [73] C.E.A. Kirschhock, S.P.B. Kremer, P.J. Grobet, P.A. Jacobs,
[41] K. Nishi, R.W. Thompson, in: F. Schüth, K.S.W. Sing, J. J.A. Martens, J. Phys. Chem. B 106 (2002) 4897.
Weitkamp (Eds.), Handbook of Porous Solids, vol. 2, Wiley- [74] J. Ciric, J. Colloid Interface Sci. 28 (1968) 315.
VCH, Weinheim, 2002, p. 736. [75] B.D. McNicol, G.T. Pott, K.R. Loos, J. Phys. Chem. 76 (1972)
[42] A. Corma, in: E. van Steen, L.H. Callanan, M. Claeys (Eds.), 3388.
Proc. 4th Int. Zeolite Conf., Catal. Soc. S. Africa, Cape Town, [76] B.D. McNicol, G.T. Pott, K.R. Loos, N. Mulder, in: W.M.
2004, p. 25. Meier, J.B. Uytterhoeven (Eds.), Molecular Sieves, ACS Adv.
[43] A. Corma, M.E. Davis, ChemPhysChem 5 (2004) 304. Chem. Ser., vol. 121, 1973, p. 152.
[44] R.A. van Santen, G. Ooms, C.J.J. den Ouden, B.W. van Beest, [77] C.L. Angell, W.H. Flank, in: J.R. Katzer (Ed.), Molecular
M.F.M. Post, in: M.L. Occelli, H.E. Robson (Eds.), Zeolite Sieves—II, ACS Symp. Ser., vol. 40, 1977, p. 194.
Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 617. [78] A. Culfaz, L.B. Sand, in: W.M. Meier, J.B. Uytterhoeven (Eds.),
[45] A. Navrotsky, I. Petrovic, Y. Hu, C.-Y. Chen, M.E. Davis, Molecular Sieves, ACS Adv. Chem. Ser., vol. 121, 1973, p. 140.
Micropor. Mater. 4 (1995) 95. [79] H. Kacirek, H. Lechert, J. Phys. Chem. 80 (1976) 1291.
[46] P.M. Piccione, S. Yang, A. Navrotsky, M.E. Davis, J. Phys. [80] B.M. Lok, T.R. Cannan, C.A. Messina, Zeolites 3 (1983) 282.
Chem. B 106 (2002) 3629. [81] E. Moretti, S. Contessa, M. Padovan, Chim. Ind. (Milan) 67
[47] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic (1985) 21.
Press, London, 1982, p. 175. [82] E.M. Flanigen, J.M. Bennett, R.W. Grose, J.P. Cohen, R.L.
[48] H. Robson (Ed.), Verified Syntheses of Zeolitic Materials, second Patton, R.L. Kirchner, J.V. Smith, Nature 271 (1978) 512.
rev. ed., Elsevier, Amsterdam, 2001. [83] E.M. Flanigen, in: L.V.C Rees (Ed.), Proc. 5th Int. Conf.
[49] R.M. Barrer, J.W. Baynham, F.W. Bultitude, W.M. Meier, J. Zeolites, Heyden, London, 1980, p. 760.
Chem. Soc. (1959) 195. [84] B.J. Schoeman, in: H. Chon, S.-K. Ihm, Y.S. Uh (Eds.), Progress
[50] R.M. Barrer, Chem. Brit. (1966) 380. in Zeolite and Microporous Materials, Studies in Surface Science
[51] E.M. Flanigen, D.W. Breck, 137th Meeting of the ACS, Division and Catalysis, vol. 105, Elsevier, Amsterdam, 1997, p. 647.
of Inorganic Chemistry, Cleveland, OH, April 1960, Abstracts, p. [85] B.J. Schoeman, O. Regev, Zeolites 17 (1996) 447.
33-M. [86] C.T.G. Knight, S.D. Kinrade, J. Phys. Chem. B 106 (2002)
[52] E.M. Flanigen, D.W. Breck, 137th Meeting of the ACS, Division 3329.
of Inorganic Chemistry, Cleveland, OH, April 1960, Paper No. [87] C.E.A. Kirschhock, R. Ravishankar, F. Verspeurt, P.J. Grobet,
82: Crystalline zeolites, V—Growth of zeolite crystals from gels P.A. Jacobs, J. Martens, J. Phys. Chem. B 106 (2002) 3333.
(unpublished—see Ref. [31]). [88] D.D. Kragten, J.M. Fedeyko, K.R. Sawant, J.D. Rimer, D.G.
[53] See D.W. Breck, Zeolite Molecular Sieves, Wiley, New York, Vlachos, R.L. Lobo, M. Tsapatsis, J. Phys. Chem. B 107 (2003)
1974, p. 342–343; D.W. Breck, J. Chem. Ed. 41 (1964) 684. 10006.
[54] D.W. Breck, J. Chem. Ed. 41 (1964) 678. [89] R.W. Thompson, A. Dyer, Zeolites 5 (1985) 202.
[55] G.T. Kerr, J. Phys. Chem. 70 (1966) 1047. [90] R.W. Thompson, A. Dyer, Zeolites 5 (1985) 292.
[56] G.T. Kerr, Zeolites 9 (1989) 451. [91] R.W. Thompson, in: C.R.A. Catlow, R. Vetrivel (Eds.), Mod-
[57] S.P. Zhdanov, in: E.M. Flanigen, L.B. Sand (Eds.), Molecular elling and Reactivity in Zeolites, Academic Press, London, 1992,
Sieve Zeolites—I, ACS Adv. Chem. Ser., vol. 101, 1971, p. 20. p. 231.
[58] E.G. Derouane, S. Detremmerie, Z. Gabelica, N. Blom, Appl. [92] B.M. Lowe, Zeolites 3 (1983) 300.
Catal. 1 (1981) 201. [93] J.L. Casci, B.M. Lowe, Zeolites 3 (1983) 187.
[59] Z. Gabelica, E.G. Derouane, N. Blom, Appl. Catal. 5 (1983) [94] B.M. Lowe, in: P.J. Grobet, W.J. Mortier, E.F. Vansant, G.
109. Schulz-Ekloff (Eds.), Innovation in Zeolite Materials Science,
[60] Z. Gabelica, N. Blom, E.G. Derouane, Appl. Catal. 5 (1983) Studies in Surface Science and Catalysis, vol. 37, Elsevier,
227. Amsterdam, 1988, p. 1.
[61] Z. Gabelica, E.G. Derouane, N. Blom, in: T.E. Whyte, Jr., R.A. [95] A. Araya, B.M. Lowe, Zeolites 6 (1986) 111.
Dalla Betta, E.G. Derouane, R.T.K. Baker (Eds.), Catalytic [96] A. Moini, K.D. Schmitt, E.W. Valyocsik, R.F. Polomski,
Materials, ACS Symp. Ser., vol. 248, 1984, p. 219. Zeolites 14 (1994) 504.
[62] P. Bodart, J.B. Nagy, Z. Gabelica, E.G. Derouane, J. Chim. [97] K.D. Schmitt, G.J. Kennedy, Zeolites 14 (1994) 635.
Phys. 83 (1986) 777. [98] V. Shen, K. Watanabe, A.T. Bell, J. Phys. Chem. B 101 (1997)
[63] C.D. Chang, A.T. Bell, Catal. Lett. 8 (1991) 305. 2007.
[64] S.L. Burkett, M.E. Davis, J. Phys. Chem. 98 (1994) 4647. [99] A.P. Stevens, A.M. Gorman, C.M. Freeman, P.A. Cox, J.
[65] S.L. Burkett, M.E. Davis, Chem. Mater. 7 (1995) 920. Chem. Soc. Faraday Trans. 92 (1996) 2065.
[66] S.L. Burkett, M.E. Davis, Chem. Mater. 7 (1995) 1453. [100] S.I. Zones, Y. Nakagawa, L.T. Yuen, T.V. Harris, J. Am.
[67] R. Ravishankar, C. Kirschhock, B.J. Schoeman, P. Vanoppen, Chem. Soc. 118 (1996) 7558.
P.J. Grobet, S. Storck, W.F. Maier, J.A. Martens, F.C. De [101] D.W. Lewis, D.J. Willock, C.R.A. Catlow, J.M. Thomas, G.J.
Schryver, P.A. Jacobs, J. Phys. Chem. B 102 (1998) 2633. Hutchings, Nature 382 (1996) 604.
[68] R. Ravishankar, C.E.A. Kirschhock, P.-P. Knops-Gerrits, [102] D.W. Lewis, G. Sankar, J.K. Wyles, J.M. Thomas, C.R.A.
E.J.P. Feijen, P.J. Grobet, P. Vanoppen, F.C. De Schryver, G. Catlow, D.J. Willock, Angew. Chem. Int. Ed. 36 (1997) 2675.
Miehe, H. Fuess, B.J. Schoeman, P.A. Jacobs, J.A. Martens, J. [103] J. Sauer, Chem. Rev. 89 (1989) 199.
Phys. Chem. B 103 (1999) 4960. [104] J.R. Hill, J. Sauer, J. Chem. Phys. 98 (1994) 1238.
[69] C.E.A. Kirschhock, R. Ravishankar, F. Verspeurt, P.J. Grobet, [105] J.C.G. Pereira, C.R.A. Catlow, G.D. Price, J. Phys. Chem. A
P.A. Jacobs, J.A. Martens, J. Phys. Chem. B 103 (1999) 4965. 103 (1999) 3252.
[70] C.E.A. Kirschhock, R. Ravishankar, L. Van Looveren, P.A. [106] J.C.G. Pereira, C.R.A. Catlow, G.D. Price, J. Phys. Chem. A
Jacobs, J.A. Martens, J. Phys. Chem. B 103 (1999) 4972. 103 (1999) 3268.
[71] C.E.A. Kirschhock, R. Ravishankar, P.A. Jacobs, J.A. Martens, [107] M.G. Wu, M.W. Deem, J. Chem. Phys. 116 (2002) 2125.
J. Phys. Chem. B 103 (1999) 11021. [108] J.W. Mullin, Crystallisation, fourth ed., Butterworth Heine-
[72] C.E.A. Kirschhock, V. Buschmann, S. Kremer, R. Ravishankar, mann, Oxford, 2001.
C.J.Y. Houssin, B.L. Mojet, R.A. van Santen, P.J. Grobet, P.A. [109] B. Subotić, J. Bronić, T. Antonić, in: M.M.J. Treacy, B.K.
Jacobs, J.A. Martens, Angew. Chem. Int. Ed. 40 (2001) 2637. Marcus, M.E. Bisher, J.B. Higgins (Eds.), Proc. 12th Int. Zeolite
Conf., Materials Research Society, Warrendale, PA, 1999, p. [139] S. Mintova, N.H. Olson, T. Bein, Angew. Chem. Int. Ed. 38
2057. (1999) 3201.
[110] J. Warzywoda, R.D. Edelman, R.W. Thompson, Zeolites 9 [140] S.P. Zhdanov, N.N. Samulevich, in: L.V.C. Rees (Ed.), Proc. 5th
(1989) 187. Int. Conf. Zeolites, Heyden, London, 1980, p. 75.
[111] C.S. Cundy, M.S. Henty, R.J. Plaisted, Zeolites 15 (1995) [141] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic
342. Press, London, 1982, p. 145–147.
[112] F. Roozeboom, H.E. Robson, S.S. Chan, Zeolites 3 (1983) 321. [142] C.S. Cundy, B.M. Lowe, D.M. Sinclair, J. Cryst. Growth 100
[113] P.K. Dutta, M. Puri, J. Phys. Chem. 91 (1987) 4329. (1990) 189.
[114] G. Engelhardt, B. Fahlke, M. Mägi, E. Lippmaa, Zeolites 5 [143] G. Golemme, A. Nastro, J.B. Nagy, B. Subotić, F. Crea, R.
(1985) 49. Aiello, Zeolites 11 (1991) 776.
[115] J. Shi, M.W. Anderson, S.W. Carr, Chem. Mater. 8 (1996) 369. [144] A. Giaya, R.W. Thompson, AIChE J. 50 (2004) 879.
[116] A.T. Bell, in: M.L. Occelli, H.E. Robson (Eds.), Zeolite [145] S.H. Bransom, W.J. Dunning, Discuss. Faraday Soc. 5 (1949)
Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 66. 96.
[117] M.A. Nicolle, F. Di Renzo, F. Fajula, P. Espiau, T. des [146] S. Gonthier, R.W. Thompson, in: J.C. Jansen, M. Stöcker, H.G.
Courieres, in: R. von Ballmoos, J.B. Higgins, M.M.J. Treacy Karge, J. Weitkamp (Eds.), Advanced Zeolite Science and
(Eds.), Proc. 9th Int. Zeolite Conf., Butterworth-Heinemann, Applications, Studies in Surface Science and Catalysis, vol. 85,
Boston, 1993, p. 313. Elsevier, Amsterdam, 1994, p. 43.
[118] I. Hasegawa, S. Sakka, in: M.L. Occelli, H.E. Robson (Eds.), [147] H. Kacirek, H. Lechert, J. Phys. Chem. 79 (1975) 1589.
Zeolite Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 140. [148] C.S. Cundy, M.S. Henty, R.J. Plaisted, Zeolites 15 (1995) 353.
[119] B. Fahlke, P. Starke, V. Seefeld, W. Wieker, K.-P. Wendlandt, [149] J.P. Verduijn, World Patent 9308124 (1993).
Zeolites 7 (1987) 209. [150] C.S. Cundy, J.O. Forrest, R.J. Plaisted, Micropor. Mesopor.
[120] J. Perez-Pariente, J.A. Martens, P.A. Jacobs, Appl. Catal. 31 Mater. 66 (2003) 143.
(1987) 35. [151] C.S. Cundy, R.J. Plaisted, J.P. Zhao, Chem. Commun. (1998)
[121] M.A. Camblor, J. Perez-Pariente, Zeolites 11 (1991) 202. 1465.
[122] Y. Tsuruta, T. Satoh, T. Yoshida, O. Okumura, S. Ueda, in: Y. [152] C.S. Cundy, J.O. Forrest, in: A. Galarneau, F. DiRenzo, F.
Murakami, A. Iijima, J.W. Ward (Eds.), New Developments in Fajula, J. Vedrine (Eds.), Zeolites and Mesoporous Materials at
Zeolite Science and Technology, Kodansha-Elsevier, Tokyo- the Dawn of the 21st Century, Studies in Surface Science and
Amsterdam, 1986, p. 1001. Catalysis, vol. 135, Elsevier, Amsterdam, 2000, p. 192, Paper No.
[123] B. Subotić, A.M. Tonejc, D. Bagović, A. Čižmek, T. Antonić, 02-P-08.
in: J. Weitkamp, H.G. Karge, H. Pfeifer, W. Hölderich (Eds.), [153] A.D. Randolph, M.A. Larson, Introduction to the Theory of
Zeolites and Related Microporous Materials: State of the Art Crystallisation Processes, second ed., Academic Press, London,
1994, Studies in Surface Science and Catalysis, vol. 84, Elsevier, 1988.
Amsterdam, 1994, p. 259. [154] B.M. Lowe, J.R.D. Nee, J.L. Casci, Zeolites 14 (1994) 610.
[124] I. Krznarić, T. Antonić, B. Subotić, V. Babić-Ivančić, Thermo- [155] J.L. Casci, in: J. Weitkamp, H.G. Karge, H. Pfeifer, W.
chim. Acta 317 (1998) 73. Hölderich (Eds.), Zeolites and Related Microporous Materials:
[125] R.I. Walton, D. OHare, J. Phys. Chem. Solids 62 (2001) State of the Art 1994, Studies in Surface Science and Catalysis,
1469. vol. 84, Elsevier, Amsterdam, 1994, p. 133.
[126] H. Yang, R.I. Walton, S. Antonijevic, S. Wimperis, A.C. [156] P.A. Cox, J.L. Casci, A.P. Stevens, J. Chem. Soc. Faraday
Hannon, J. Phys. Chem. B 108 (2004) 8208. Trans. 106 (1997) 473.
[127] T. Ito, J. Fraissard, J.B. Nagy, N. Dewaele, Z. Gabelica, A. [157] S.R. Chiverton, C.S. Cundy, R.J. Plaisted, unpublished work.
Nastro, E.G. Derouane, in: P.A. Jacobs, R.A. van Santen [158] P.K. Dutta, J. Bronić, Zeolites 14 (1994) 250.
(Eds.), Zeolites: Facts, Figures, Future, Studies in Surface [159] J. Bronić, B. Subotić, M. Škreblin, Micropor. Mesopor. Mater.
Science and Catalysis, vol. 49, Elsevier, Amsterdam, 1989, p. 579. 28 (1999) 73.
[128] R. Gougeon, L. Delmotte, D. Le Nouen, Z. Gabelica, Micropor. [160] J. Bronić, B. Subotić, I. Šmit, Lj.A. Despotović, in: P.J. Grobet,
Mesopor. Mater. 26 (1998) 143. W.J. Mortier, E.F. Vansant, G. Schulz-Ekloff (Eds.), Innovation
[129] O. Regev, Y. Cohen, E. Kehat, Y. Talmon, Zeolites 14 (1994) in Zeolite Materials Science, Studies in Surface Science and
314. Catalysis, vol. 37, Elsevier, Amsterdam, 1988, p. 107.
[130] J.N. Watson, L.E. Iton, R.I. Keir, J.C. Thomas, T.L. Dowling, [161] B. Subotić, J. Bronić, in: R. von Ballmoos, J.B. Higgins, M.M.J.
J.W. White, J. Phys. Chem. B 101 (1997) 10094. Treacy (Eds.), Proc. 9th Int. Zeolite Conf., Butterworth-Heine-
[131] J.N. Watson, A.S. Brown, L.E. Iton, J.W. White, J. Chem. Soc. mann, Boston, 1993, p. 321.
Faraday Trans. 94 (1998) 2181. [162] B. Subotić, A. Graovac, in: Recent Progress Reports and
[132] P.A. Jacobs, E.G. Derouane, J. Weitkamp, J. Chem. Soc. Chem. Discussion, 5th Int. Conf. Zeolites, Giannini, Naples, 1980, p. 54.
Commun. (1981) 591. [163] B. Subotić, A. Graovac, in: B. Držaj, S. Hočevar, S. Pejovnic
[133] R. Ravishankar, C. Kirschhock, B.J. Schoeman, D. De Vos, (Eds.), Zeolites: Synthesis, Structure, Technology and Applica-
P.J. Grobet, P.A. Jacobs, J.A. Martens, in: M.M.J. Treacy, tion, Studies in Surface Science and Catalysis, vol. 24, Elsevier,
B.K. Marcus, M.E. Bisher, J.B. Higgins (Eds.), Proc. 12th Int. Amsterdam, 1985, p. 199.
Zeolite Conf., Materials Research Society, Warrendale, PA, [164] B. Subotić, in: M.L. Occelli, H.E. Robson (Eds.), Zeolite
1999, p. 1825. Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 110.
[134] S. Mintova, N.H. Olson, J. Senker, T. Bein, Angew. Chem. Int. [165] A. Katović, B. Subotić, I. Šmit, Lj.A. Despotović, M. Ćurić, in:
Ed. 41 (2002) 2558. M.L. Occelli, H.E. Robson (Eds.), Zeolite Synthesis, ACS Symp.
[135] G. Coudurier, C. Naccache, J.C. Vedrine, J. Chem. Soc. Chem. Ser., vol. 398, 1989, p. 124.
Commun. (1982) 1413. [166] R.W. Thompson, Zeolites 12 (1992) 837.
[136] C.G. Pope, Micropor. Mesopor. Mater. 21 (1998) 333. [167] S. Gonthier, L. Gora, I. Guray, R.W. Thompson, Zeolites 13
[137] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic (1993) 414.
Press, London, 1982, p. 133. [168] C. Falamaki, M. Edrissi, M. Sohrabi, Zeolites 19 (1997) 2.
[138] S. Mintova, N.H. Olson, V. Valtchev, T. Bein, Science 283 [169] V. Nikolakis, D.G. Vlacho, M. Tsapatsis, Micropor. Mesopor.
(1999) 958. Mater. 21 (1998) 337.
[170] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic Perspectives in Molecular Sieve Science, ACS Symp. Ser., vol.
Press, London, 1982 (Chapters 3–5). 368, 1988, p. 277.
[171] K.R. Andersson, L.S. Dent Glasser, D.N. Smith, in: J.S. Falcone [205] A. Tavolaro, R. Mostowicz, F. Crea, A. Nastro, R. Aiello, J.B.
(Ed.), Soluble Silicates, ACS Symp. Ser., vol. 194, 1982, p. 115. Nagy, Zeolites 12 (1992) 756.
[172] R.K. Iler, The Chemistry of Silica, Wiley, New York, 1979. [206] N.N. Feoktistova, S.P. Zhdanov, W. Lutz, M. Bülow, Zeolites 9
[173] R.K. Iler, in: J.S. Falcone (Ed.), Soluble Silicates, ACS Symp. (1989) 136.
Ser., vol. 194, 1982, p. 95. [207] R. Mostowicz, L.B. Sand, Zeolites 3 (1983) 219.
[174] J.-E.A. Otterstedt, M. Ghuzel, J.P. Sterte, J. Colloid Interface [208] J. Garside, M.B. Shah, Ind. Eng. Chem. Process Des. Dev. 19
Sci. 115 (1987) 95. (1980) 509.
[175] S. Ueda, H. Murata, M. Koizumi, H. Nishimura, Am. Mineral. [209] J.W. Mullin, Crystallisation, fourth ed., Butterworth Heine-
65 (1980) 1012. mann, Oxford, 2001, p. 237.
[176] R.W. Thompson, A. Dyer, Zeolites 5 (1985) 302. [210] R.A Cournoyer, W.L. Kranich, L.B. Sand, J. Phys. Chem. 79
[177] B.J. Schoeman, J. Sterte, J.-E. Otterstedt, Zeolites 14 (1994) 110. (1975) 1578.
[178] T. Brar, P. France, P.G. Smirniotis, J. Phys. Chem. B 105 (2001) [211] R. Mostowicz, L.B. Sand, Zeolites 2 (1982) 143.
5383. [212] R.K. Iler, The Chemistry of Silica, Wiley, New York, 1979, p.
[179] J. Bronić, B. Subotić, Micropor. Mater. 4 (1995) 239. 248.
[180] R. Aiello, R.M. Barrer, I.S. Kerr, in: E.M. Flanigen, L.B. Sand [213] C.J. Creswell, R.K. Harris, P.T. Jageland, J. Chem. Soc. Chem.
(Eds.), Molecular Sieve Zeolites—I, ACS Adv. Chem. Ser., vol. Commun. (1984) 1261.
101, 1971, p. 44. [214] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic
[181] B.J. Schoeman, Zeolites 18 (1997) 97. Press, London, 1982, p. 152.
[182] H.W. Zandbergen, Instit. Phys. Conf. Ser., No. 93, vol. 2 (1988) [215] A.E. Nielsen, Kinetics of Precipitation, Pergamon, Oxford, 1964.
363. [216] A.E. Persson, B.J. Schoeman, J. Sterte, J.-E. Otterstedt, Zeolites
[183] G.O. Brunner, Zeolites 12 (1992) 428. 14 (1994) 557.
[184] F. Taullelle, personal communication (2001). [217] B.J. Schoeman, J. Sterte, J.-E. Otterstedt, Zeolites 14 (1994)
[185] J. Garside, A. Mersmann, J. Nyvlt (Eds.), Measurement of 568.
Crystal Growth Rates, Europ. Fed. Chem. Eng. Working Party [218] S.G. Fegan, B.M. Lowe, J. Chem. Soc. Chem. Commun. (1984)
on Crystallisation, Druckhaus Deutsch, München, 1990. 427.
[186] C.S. Cundy, M.S. Henty, R.J. Plaisted, Zeolites 15 (1995) 400. [219] S.G. Fegan, B.M. Lowe, J. Chem. Soc. Faraday Trans. I 82
[187] A. Nastro, L.B. Sand, Zeolites 3 (1983) 57. (1986) 785.
[188] C.S. Cundy, B.M. Lowe, D.M. Sinclair, J. Chem. Soc. Faraday [220] G.H. Kühl, Eur. Pat. Appl. 93519 (1983).
Discuss. 95 (1993) 235. [221] C.S. Tsay, A.S.T. Chiang, Micropor. Mesopor. Mater. 26 (1998)
[189] T. Sano, Y. Kiyozumi, F. Mizukami, A. Iwasaki, M. Ito, M. 89.
Watanabe, Micropor. Mater. 1 (1993) 353. [222] T.A.M. Twomey, M. Mackay, H.P.C.E. Kuipers, R.W. Thomp-
[190] A. Iwasaki, M. Hirata, I. Kudo, T. Sano, S. Sugawara, M. Ito, son, Zeolites 14 (1994) 162.
M. Watanabe, Zeolites 15 (1995) 308. [223] P.-P.E.A. de Moor, T.P.M. Beelen, B.U. Komanschek, O. Diat,
[191] A. Iwasaki, M. Hirata, I. Kudo, T. Sano, Zeolites 16 (1996) 35. R.A. van Santen, J. Phys. Chem. B 101 (1997) 11077.
[192] M.W. Anderson, Curr. Opin. Solid State Mat. Sci. 5 (2001) [224] P.S. Singh, T.L. Dowling, J.N. Watson, J.W. White, Phys.
407. Chem. Chem. Phys. 1 (1999) 4125.
[193] H. Takano, J.R. Kenseth, S.-S. Wong, J.C. OBrien, M.D. [225] M. Aoki, H. Abe, in: New Developments in Zeolite Science and
Porter, Chem. Rev. 99 (1999) 2845. Technology, Preprints of Poster Papers, 7th Int. Zeolite Conf.,
[194] J.I. Paredes, A. Martı́nez-Alonso, J.M.D. Tascón, Micropor. Tokyo, 1986, Jap. Assoc. Zeolite, Tokyo, 1986, Abstract 4B-1, p.
Mesopor. Mater. 65 (2003) 93. 157.
[195] S. Bazzana, S. Damrul, J. Warzywoda, L. Hsiao, L. Klass, M. [226] G. Binder, L. Scandella, A. Schumacher, N. Kruse, R. Prins,
Knapp, J.A. Rains, E.M. Stein, M.J. Sullivan, C.M. West, J.Y. Zeolites 16 (1996) 2.
Woo, A. Sacco, in: R. Aiello, G. Giordano, F. Testa (Eds.), [227] W.K. Burton, N. Cabrera, F.C. Frank, Phil. Trans. A 243 (1951)
Impact of Zeolites and Other Porous Materials on the New 299.
Technologies at the Beginning of the New Millennium, Studies in [228] M. Volmer, Kinetik Der Phasenbildung, Steinkopff, Dresden,
Surface Science and Catalysis, vol. 142, Elsevier, Amsterdam, 1939.
2002, p. 117. [229] W. Kossel, Annln. Phys. 21 (1934) 457.
[196] S. Damrul, S. Bazzana, J. Warzywoda, R.R. Biederman, A. [230] C.S. Cundy, I. Meza, J.R. Agger, paper in preparation.
Sacco, Micropor. Mesopor. Mater. 54 (2002) 79. [231] J.R. Agger, N. Hanif, C.S. Cundy, A.P. Wade, S. Dennison,
[197] W. Meise, F.E. Schwochow, in: W.M. Meier, J.B. Uytterhoeven P.A. Rawlinson, M.W. Anderson, J. Am. Chem. Soc. 125
(Eds.), Molecular Sieves, ACS Adv. Chem. Ser., vol. 121, 1973, (2003) 830.
p. 169. [232] J.R. Agger, N. Hanif, M.W. Anderson, Angew. Chem. Int. Ed.
[198] L. Gora, K. Streletzky, R.W. Thompson, G.D.J. Phillies, 40 (2001) 4065.
Zeolites 18 (1997) 119. [233] M.W. Anderson, J.R. Agger, N. Hanif, O. Terasaki, T. Ohsuna,
[199] E.F. Freund, J. Crystal Growth 34 (1976) 11. Solid State Sci. 3 (2001) 809.
[200] G.S. Wiersema, R.W. Thompson, J. Mater. Chem. 6 (1996) [234] M.W. Anderson, J.R. Agger, N. Hanif, O. Terasaki, Micropor.
1693. Mesopor. Mater. 48 (2001) 1.
[201] M.D. Grebner, A. Reich, F. Schüth, K.K. Unger, K.D. Franz, [235] B. Wathen, M. Kuiper, V. Walker, Z. Jia, J. Am. Chem. Soc.
Zeolites 13 (1993) 139. 125 (2003) 729.
[202] F. Crea, A. Nastro, J.B. Nagy, R. Aiello, Zeolites 8 (1988) 262. [236] S. Sugiyama, S. Yamamoto, O. Matsuoka, H. Nozoye, J. Yu,
[203] D.T. Hayhurst, A. Nastro, R. Aiello, F. Crea, G. Giordano, G. Zhu, S. Qiu, O. Terasaki, Micropor. Mesopor. Mater. 28
Zeolites 8 (1988) 416. (1999) 1.
[204] D.T. Hayhurst, R. Aiello, J.B. Nagy, F. Crea, G. Giordano, A. [237] T. Wakihara, A. Sugiyama, T. Okubo, Micropor. Mesopor.
Nastro, J.C. Lee, in: W.H. Flank, T.E. Whyte, Jr. (Eds.), Mater. 70 (2004) 7.
[238] L.D. Rollmann, in: R.B. King (Ed.), Inorganic Compounds with Science and Applications, Studies in Surface Science and
Unusual Properties—II, ACS Adv. Chem. Ser., vol. 173, 1979, Catalysis, vol. 85, Elsevier, Amsterdam, 1994, p. 75.
p. 387. [266] R.M. Laine, K.Y. Blohowiak, T.R. Robinson, M.L. Hoppe, P.
[239] R.H. Daniels, G.T. Kerr, L.D. Rollmann, J. Am. Chem. Soc. Nardi, J. Kampf, J. Uhm, Nature 353 (1991) 642.
100 (1978) 3097. [267] S.D. Kinrade, J.W. Del Nin, A.S. Schach, T.A. Sloan, K.L.
[240] P.A. Wright, P.A. Cox, G.W. Noble, V. Patinek, in: M.M.J. Wilson, C.T.G. Knight, Science 285 (1999) 1542.
Treacy, B.K. Marcus, M.E. Bisher, J.B. Higgins (Eds.), Proc. [268] C. Shao, X. Li, S. Qiu, F.-S. Xiao, O. Terasaki, Micropor.
12th Int. Zeolite Conf., Materials Research Society, Warrendale, Mesopor. Mater. 39 (2000) 117.
PA, 1999, p. 1603. [269] D.M. Bibby, M.P. Dale, Nature 317 (1985) 157.
[241] L.W. Beck, M.E. Davis, Micropor. Mesopor. Mater. 22 (1998) [270] C. Shao, X. Li, S. Qiu, F.-S. Xiao, Micropor. Mesopor. Mater.
107. 33 (1999) 215.
[242] R.A. van Santen, J. Keijsper, G. Ooms, A.G.T.G. Kortbeek, in: [271] B. Herreros, S.W. Carr, J. Klinowski, Science 263 (1994) 1585.
Y. Murakami, A. Iijima, J.W. Ward (Eds.), New Developments [272] R.M. Barrer, in: M.L. Occelli, H.E. Robson (Eds.), Zeolite
in Zeolite Science and Technology, Kodansha-Elsevier, Tokyo- Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 11.
Amsterdam, 1986, p. 169. [273] W.H. Dokter, H.F. van Garderen, T.P.M. Beelen, R.A. van
[243] J.J. Keijsper, M.F.M. Post, in: M.L. Occelli, H.E. Robson Santen, W. Bras, Angew. Chem. Int. Ed. 34 (1995) 73.
(Eds.), Zeolite Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 28. [274] P.-P.E.A. de Moor, T.P.M. Beelen, B.U. Komanschek, L.W.
[244] C.T.G. Knight, Zeolites 10 (1990) 140. Beck, P. Wagner, M.E. Davis, R.A. van Santen, Chem. Eur. J. 5
[245] G. Boxhoorn, O. Sudmeijer, P.H.G. van Kasteren, J. Chem. Soc. (1999) 2083.
Chem. Commun. (1983) 1416. [275] P.-P.E.A. de Moor, T.P.M. Beelen, R.A. van Santen, L.W.
[246] C.T.G. Knight, R.T. Syvitski, S.D. Kinrade, in: L. Bonneviot, Beck, M.E. Davis, J. Phys. Chem. B 104 (2000) 7600.
S. Kaliaguine (Eds.), Zeolites: A Refined Tool for Designing [276] V. Nikolakis, E. Kokkoli, M. Tirrell, M. Tsapatsis, D.G.
Catalytic Sites, Studies in Surface Science and Catalysis, vol. 97, Vlachos, Chem. Mater. 12 (2000) 845.
Elsevier, Amsterdam, 1995, p. 483. [277] R.W. Corkery, B.W. Ninham, Zeolites 18 (1997) 379.
[247] R. Szostak, Molecular Sieves—Principles of Synthesis and [278] B.J. Schoeman, Micropor. Mesopor. Mater. 22 (1998) 9.
Identification, Van Nostrand Reinhold, New York, 1989, p. 197. [279] A. Erdem-S ß enatalar, R.W. Thompson, J. Colloid Interface Sci.,
[248] M.T. Melchior, D.E.W. Vaughan, C.F. Pictroski, J. Phys. Chem. submitted for publication.
99 (1995) 6128. [280] V. Privman, D.V. Goia, J. Park, E.J. Matijević, J. Colloid
[249] A. Thangaraj, R. Kumar, Zeolites 10 (1990) 117. Interface Sci. 213 (1999) 36.
[250] P.K. Dutta, D.C. Shieh, J. Phys. Chem. 90 (1986) 2331. [281] J. Park, V. Privman, E. Matijević, J. Phys. Chem. B 105 (2001)
[251] P.K. Dutta, D.C. Shieh, M. Puri, J. Phys. Chem. 91 (1987) 11630.
2332. [282] M.E. Davis, in: L. Bonneviot, S. Kaliaguine (Eds.), Zeolites: a
[252] P.K. Dutta, D.C. Shieh, M. Puri, Mater. Res. Soc. Symp. Proc. Refined Tool for Designing Catalytic Sites, Studies in Surface
111 (1988) 101. Science and Catalysis, vol. 97, Elsevier, Amsterdam, 1995, p.
[253] G. Xiong, Y. Yu, Z.-C. Feng, Q. Xin, F.-S. Xiao, C. Li, 35.
Micropor. Mesopor. Mater. 42 (2001) 317. [283] M. Tsapatsis, M. Lovallo, M.E. Davis, Micropor. Mater. 5
[254] J.-P. Gilson, in: E.G. Derouane, F. Lemos, C. Naccache, F.R. (1996) 381.
Ribeiro (Eds.), Zeolite Microporous Solids: Synthesis, Structure [284] C.S. Cundy, R.J. Plaisted, J.O. Forrest, unpublished work.
and Reactivity, NATO ASI Series, No. C352, Kluwer, Dordr- [285] H. Kalipcilar, A. Culfaz, Cryst. Res. Technol. 35 (2000) 933.
echt, 1992, p. 19. [286] G. Reding, T. Mäurer, B. Kraushaar-Czarnetzki, Micropor.
[255] H. Lechert, H. Kacirek, Zeolites 11 (1991) 720. Mesopor. Mater. 57 (2003) 83.
[256] L. Griffiths, C.S. Cundy, R.J. Plaisted, J. Chem. Soc. Dalton [287] Q. Li, D. Creaser, J. Sterte, Micropor. Mesopor. Mater. 31
Trans. (1986) 2265. (1999) 141.
[257] A.V. McCormick, A.T. Bell, C.J. Radke, in: Y. Murakami, A. [288] C.S. Gittleman, A.T. Bell, C.J. Radke, Micropor. Mater. 2
Iijima, J.W. Ward (Eds.), New Developments in Zeolite Science (1994) 145.
and Technology, Kodansha-Elsevier, Tokyo-Amsterdam, 1986, [289] C.S. Gittleman, S.S. Lee, A.T. Bell, C.J. Radke, Micropor.
p. 247. Mater. 3 (1995) 511.
[258] R.J. Moolenaar, J.C. Evans, L.D. McKeever, J. Phys. Chem. 74 [290] S.P. Zhdanov, M.A. Shubaeva, M.M. Piryutko, Bull. Acad. Sci.
(1970) 3629. USSR 32 (1983) 1764.
[259] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic [291] M.G. Howden, Preparation and Evaluation of an Amorphous
Press, London, 1982, p. 106. Silica-Alumina Catalyst Synthesised in the Presence of Tetra-
[260] E.M. Flanigen, R.L. Patton, US Patent 4,073,865 (1978). propylammonium Hydroxide, CSIR Report CENG 441, CSIR,
[261] J.L. Guth, H. Kessler, J.M. Higel, J.M. Lamblin, J. Patarin, A. Pretoria, 1982 (ISBN 0 7988 2620 7).
Seive, J.M. Chezeau, R. Wey, in: M.L. Occelli, H.E. Robson [292] D.P. Serrano, M.A. Uguina, R. Sanz, E. Castillo, A. Rodrı́guez,
(Eds.), Zeolite Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 176. P. Sánchez, Micropor. Mesopor. Mater. 69 (2004) 197.
[262] H. Kessler, in: J. Klinowski, P.J. Barrie (Eds.), Recent Advances [293] R. Van Grieken, J.L. Sotelo, J.M. Menéndez, J.A. Melero,
in Zeolite Science, Studies in Surface Science and Catalysis, vol. Micropor. Mesopor. Mater. 39 (2000) 135.
52, Elsevier, Amsterdam, 1989, p. 17. [294] W. Xu, J. Dong, J. Li, J. Li, F. Wu, J. Chem. Soc. Chem.
[263] J.L. Guth, P. Caullet, A. Seive, J. Patarin, F. Delprato, in: D. Commun. (1990) 755.
Barthomeuf, E.G. Derouane, W. Hölderich (Eds.), Guidelines [295] M. Matsukata, M. Ogura, T. Osaki, P.R.H.R. Rao, M. Nomura,
for Mastering the Properties of Molecular Sieves, NATO ASI E. Kikuchi, Topics in Catalysis 9 (1999) 77.
Series, No. B221, Plenum, New York, 1990, p. 69. [296] R.W. Grose, E.M. Flanigen, US Patent 4,257,885 (1981).
[264] H. Kessler, in: R. von Ballmoos, J.B. Higgins, M.M.J. Treacy [297] M.-H. Kim, H.-X. Li, M.E. Davis, Micropor. Mater. 1 (1993)
(Eds.), Proc. 9th Int. Zeolite Conf., Butterworth-Heinemann, 191.
Boston, 1993, p. 73. [298] F. Crea, R. Aiello, A. Nastro, J.B. Nagy, Zeolites 11 (1991) 521.
[265] H. Kessler, J. Patarin, C. Schott-Darie, in: J.C. Jansen, M. [299] H. Khatami, E.M. Flanigen, unpublished results, quoted in Ref.
Stöcker, H.G. Karge, J. Weitkamp (Eds.), Advanced Zeolite [32], p. 340.
[300] A.K. Patel, L.B. Sand, in: J.R. Katzer (Ed.), Molecular Sieves II, [332] J. Ciric, US Patent 3,433,589 (1969).
ACS Symp. Ser., vol. 40, ACS, Washington, 1977, p. 207. [333] S. Kasahara, K. Itabashi, K. Igawa, in: Y. Murakami, A. Iijima,
[301] K.R. Franklin, B.M. Lowe, Zeolites 7 (1987) 135. J.W. Ward (Eds.), New Developments in Zeolite Science and
[302] E. Moretti, G. Leofanti, M. Padovan, M. Solari, G. De Alberti, Technology, Kodansha-Elsevier, Tokyo-Amsterdam, 1986, p.
F. Gatti, in: Proc. 8th Int. Conf. Catal., vol. 4, 1985, p. IV 713 185.
(Chem. Abs. 10514 121564g). [334] D.E.W. Vaughan, Eur. Patent Appl. 142,347 (1985).
[303] M. Padovan, G. Leofanti, M. Solari, E. Moretti, Zeolites 4 [335] S. Nicolas, P. Massiani, M. Vera Pacheco, F. Fajula, F.
(1984) 295. Figueras, in: P.J. Grobet, W.J. Mortier, E.F. Vansant, G.
[304] J.P. Zhao, C.S. Cundy, R.J. Plaisted, J. Dwyer, in: M.M.J. Schulz-Ekloff (Eds.), Innovation in Zeolite Materials Science,
Treacy, B.K. Marcus, M.E. Bisher, J.B. Higgins (Eds.), Proc. Studies in Surface Science and Catalysis, vol. 37, Elsevier,
12th Int. Zeolite Conf., Materials Research Society, Warrendale, Amsterdam, 1988, p. 115.
PA, 1999, p. 1591. [336] H. Kacirek, A. Meyer, Eur. Patent Appl. 150,256 (1985).
[305] C.S. Cundy, Collect. Czech. Chem. Commun. 63 (1998) 1699. [337] A.A. Brock, G.N. Link, P.S. Poitras, R.W. Thompson, J. Mater.
[306] M. Matsukata, T. Osaki, M. Ogura, E. Kikuchi, Micropor. Chem. 3 (1993) 907.
Mesopor. Mater. 56 (2002) 1. [338] L. Gora, R.W. Thompson, Zeolites 18 (1997) 132.
[307] S.P. Naik, A.S.T. Chiang, R.W. Thompson, J. Phys. Chem. B [339] F. Polak, A. Cichocki, in: W.M. Meier, J.B. Uytterhoeven
107 (2003) 7006. (Eds.), Molecular Sieves, ACS Adv. Chem. Ser., vol. 121, 1973,
[308] A. Arnold, S. Steuernagel, M. Hunger, J. Weitkamp, Micropor. p. 209.
Mesopor. Mater. 62 (2003) 97. [340] D.M. Ginter, G.T. Went, A.T. Bell, C.J. Radke, Zeolites 12
[309] J.-H. Choy, S.-R. Lee, Y.-S. Han, M. Park, G.-S. Park, Chem. (1992) 733.
Commun. (2003) 1922. [341] D.M. Ginter, A.T. Bell, C.J. Radke, Zeolites 12 (1992) 742.
[310] S.-R. Lee, Y.-S. Han, M. Park, G.-S. Park, J.-H. Choy, Chem. [342] P.M. Slangen, J.C. Jansen, H. van Bekkum, Micropor. Mater. 9
Mater. 15 (2003) 4841. (1997) 259.
[311] R. Althoff, K. Unger, F. Schüth, Micropor. Mater. 2 (1994) 557. [343] T. Roland, H. Schmidt, A. Tissler, D. Prescher, Eur. Patent
[312] R.E. Morris, S.J. Weigel, Chem. Soc. Rev. 26 (1997) 309. Appl. 402,639 (1990).
[313] A. Kuperman, S. Nadimi, S. Oliver, G.A. Ozin, J.A. Garcés, [344] T. Roland, A. Tissler, Eur. Patent Appl. 402,801 (1990).
M.M. Olken, Nature 365 (1993) 239. [345] G.W. Dodwell, R.P. Denkewicz, L.B. Sand, Zeolites 5 (1985)
[314] S. Nadimi, S. Oliver, A. Kuperman, A. Lough, G.A. Ozin, J.A. 153.
Garcés, M.M. Olken, P. Rudolph, in: J. Weitkamp, H.G. Karge, [346] P. Chu, D.J. Klocke, G.W. Kirker, J.P. McWilliams, D.O.
H. Pfeifer, W. Hölderich (Eds.), Zeolites and Related Micropo- Marler, J.C. Vartuli, Eur. Patent Appl. 306,181 (1989).
rous Materials, Studies in Surface Science and Catalysis, vol. 84, [347] T. Inui, N. Morinaga, Y. Takegami, Appl. Catal. 8 (1983) 187.
Elsevier, Amsterdam, 1994, p. 93. [348] T. Inui, in: M.L. Occelli, H.E. Robson (Eds.), Zeolite Synthesis,
[315] S. Oliver, A. Kuperman, A. Lough, G.A. Ozin, J.A. Garcés, ACS Symp. Ser., vol. 398, 1989, p. 479.
M.M. Olken, P. Rudolph, in: J. Weitkamp, H.G. Karge, H. [349] C.S. Cundy, M.S. Henty, R.J. Plaisted, unpublished work.
Pfeifer, W. Hölderich (Eds.), Zeolites and Related Microporous [350] J. Dougherty, L.E. Iton, J.W. White, Zeolites 15 (1995) 640.
Materials, Studies in Surface Science and Catalysis, vol. 84, [351] P.-P.E.A. de Moor, T.P.M. Beelen, R.A. van Santen, J. Appl.
Elsevier, Amsterdam, 1994, p. 219. Cryst. 30 (1997) 675.
[316] S. Oliver, A. Kuperman, G.A. Ozin, Angew. Chem. Int. Ed. 37 [352] P.-P.E.A. de Moor, T.P.M. Beelen, R.A. van Santen, J. Phys.
(1998) 46. Chem. B 103 (1999) 1639.
[317] D.P. Serrano, M.A. Uguina, G. Ovejero, R. Van Grieken, M. [353] J.D. Cook, R.W. Thompson, Zeolites 8 (1988) 322.
Camacho, J. Chem. Soc. Chem. Commun. (1996) 1097. [354] A. Čižmek, B. Subotić, D. Kralj, V. Babić-Ivančić, A. Tonejc,
[318] M. Salou, Y. Kiyozumi, F. Mizukami, P. Nair, K. Maeda, S. Micropor. Mater. 12 (1997) 267.
Niwa, J. Mater. Chem. 8 (1998) 2125. [355] Q. Li, B. Mihailova, D. Creaser, J. Sterte, Micropor. Mesopor.
[319] R.J. Davey, P.D. Guy, B. Mitchell, A.J. Ruddick, S.N. Black, J. Mater. 40 (2000) 53.
Chem. Soc. Faraday Trans. I 85 (1989) 1795. [356] Q. Li, B. Mihailova, D. Creaser, J. Sterte, Micropor. Mesopor.
[320] R.J. Davey, A.J. Ruddick, P.D. Guy, B. Mitchell, S.J. Maginn, Mater. 43 (2001) 51.
L.A. Polywka, J. Phys. D: Appl. Phys. 24 (1991) 176. [357] C.S. Cundy, J.O. Forrest, R.J. Plaisted, in: A. Galarneau, F. Di
[321] B.F. Mentzen, M. Sacerdoteperonnet, Mater. Res. Bull. 28 Renzo, F. Fajula, J. Vedrine (Eds.), Zeolites and Mesoporous
(1993) 1017. Materials at the Dawn of the 21st Century, Studies in Surface
[322] C.A. Fyfe, G.T. Kokotailo, H. Strobl, H. Gies, G.J. Kennedy, Science and Catalysis, vol. 135, Elsevier, Amsterdam, 2000, p.
C.T. Pasztor, G.E. Barlow, in: H.G. Karge, J. Weitkamp 143 (Paper No. 02-O-02).
(Eds.), Zeolites as Catalysts, Sorbents and Detergent Builders, [358] W.P. Guo, L.M. Huang, P. Deng, Z.Y. Xue, Q.Z. Li, Micropor.
Studies in Surface Science and Catalysis, vol. 46, Elsevier, Mesopor. Mater. 44 (2001) 427.
Amsterdam, 1989, p. 827. [359] Z.T. Zhang, Y. Han, F.S. Xiao, S.L. Qiu, L. Zhu, R.W. Wang,
[323] N. Ohnishi, S. Qiu, O. Terasaki, T. Kajitani, K. Hiraga, Y. Yu, Z. Zhang, B.S. Zou, Y.Q. Wang, H.P. Sun, D.Y. Zhao,
Micropor. Mater. 2 (1993) 73. Y. Wei, J. Am. Chem. Soc. 123 (2001) 5014.
[324] G. GellAgli, C. Ferone, M.C. Mascolo, M. Pansini, Solid State [360] Y. Liu, T.J. Pinnavaia, Chem. Mater. 14 (2002) 3.
Ionics 127 (2000) 309. [361] J.L. Schlenker, B.K. Peterson, J. Appl. Cryst. 299 (1996) 178.
[325] N.J. Clayden, S. Esposito, C. Ferone, M. Pansini, J. Mater. [362] We thank P.J. Hogan for providing these results.
Chem. 13 (2003) 1681. [363] Z. Gabelica, J.B. Nagy, G. Debras, E.G. Derouane, Acta Chim.
[326] L. Xiaoyang, S. Wenhui, W. Yifeng, Z. Xudong, J. Chem. Soc. Hung. 119 (1985) 275.
Chem. Commun. (1992) 902. [364] Z. Gabelica, J.B. Nagy, G. Debras, J. Catal. 84 (1983) 256.
[327] W.M. Meier, M. Groner, J. Solid State Chem. 37 (1981) 204. [365] M.R.S. Manton, J.C. Davidtz, J. Catal. 60 (1979) 156.
[328] S. Cartlidge, W.M. Meier, Zeolites 4 (1984) 218. [366] J.M. Thomas, L.A. Bursill, Angew. Chem. Int. Edn. 19 (1980)
[329] S. Cartlidge, E.B. Keller, W.M. Meier, Zeolites 4 (1984) 226. 745.
[330] R. von Ballmoos, W.M. Meier, J. Phys. Chem. 86 (1982) 2698. [367] L.A. Bursill, J.M. Thomas, Nature 289 (1981) 157.
[331] D.W. Breck, US Patent 3,130,007 (1964). [368] L.A. Bursill, J.M. Thomas, J. Phys. Chem. 85 (1981) 3007.
[369] C. Kosanović, B. Subotić, I. Šmit, Thermochim. Acta 317 (1998) [402] P. Caullet, J.-L. Guth, in: M.L. Occelli, H.E. Robson (Eds.),
25. Zeolite Synthesis, ACS Symp. Ser., vol. 398, 1989, p. 83.
[370] C. Kosanović, A. Čižmek, B. Subotić, I. Šmit, M. Stubičar, A. [403] J. Šefčik, A.V. McCormick, AIChE J. 43 (1997) 2773.
Tonejc, Zeolites 15 (1995) 632. [404] J. Šefčik, A.V. McCormick, Micropor. Mater. 10 (1997) 173.
[371] J. Xie, S. Kaliaguine, Appl. Catal. A 148 (1997) 415. [405] J. Šefčik, A.V. McCormick, Chem. Eng. Sci. 54 (1999) 3513.
[372] C. Kosanović, B. Subotić, I. Šmit, Croat. Chem. Acta 74 (2001) [406] T. Ejaz, A.G. Jones, P. Graham, J. Chem. Eng. Data 44 (1999)
195. 574.
[373] E. Kasai, H. Mimura, K. Sugiyama, F. Saito, K. Akiba, Y. [407] Y. Zhang, Y. Li, Y. Zhang, J. Cryst. Growth 254 (2003) 156.
Waseda, Adv. Powder Technol. 5 (1994) 189. [408] E. Grujić, B. Subotić, L.A. Despotović, in: P.A. Jacobs, R.A.
[374] A.S. Kharitonov, V.B. Fenelonov, T.P. Voskresenskaya, N.A. van Santen (Eds.), Zeolites: Facts, Figures, Future, Studies in
Rudina, V.V. Molchanov, L.M. Plyasova, G.I. Panov, Zeolites Surface Science and Catalysis, vol. 49, Elsevier, Amsterdam,
15 (1995) 253. 1989, p. 261.
[375] M.D. Rutter, T. Uchida, R.A. Secco, Y. Huang, Y. Wang, J. [409] D. Young, Ph.D. Thesis, University of Edinburgh, 1988.
Phys. Chem. Solids 62 (2001) 599. [410] S. Bosnar, J. Bronić, B. Subotić, in: I. Kiricsi, G. Pál-Borbély,
[376] J.S. Tse, D.D. Klug, J.A. Ripmeester, S. Desgreniers, K. J.B. Nagy, H.G. Karge (Eds.), Porous Materials in Environ-
Lagarec, Nature 369 (1994) 724. mentally Friendly Processes, Studies in Surface Science and
[377] H.J. Liu, R.A. Secco, Y.N. Huang, Phys. Chem. Comm. 4 Catalysis, vol. 125, Elsevier, Amsterdam, 1999, p. 69.
(2001) 37 (art. no. 8). [411] T. Antonić, B. Subotić, N. Stubičar, Zeolites 18 (1997) 291.
[378] P. Gillet, J.M. Malezieux, J.P. Itie, Amer. Mineral. 81 (1996) [412] S. Bosnar, B. Subotić, Micropor. Mesopor. Mater. 28 (1999)
651. 483.
[379] Y. Huang, E.A. Havenga, Chem. Phys. Lett. 345 (2001) 65. [413] S. Bosnar, B. Subotić, Croat. Chem. Acta 75 (2002) 663.
[380] J. Gulı́n-González, G.B. Suffritti, Micropor. Mesopor. Mater. 69 [414] I. Petrovic, A. Navrotsky, M.E. Davis, S.I. Zones, Chem. Mater.
(2004) 127. 5 (1993) 1805.
[381] G.N. Greaves, F. Meneau, A. Sapelkin, L.M. Colyer, I. [415] P.M. Piccione, C. Laberty, S. Yang, M.A. Camblor, A.
ap Gwynn, S. Wade, G. Sankar, Nature Mater. 2 (2003) 622. Navrotsky, M.E. Davis, J. Phys. Chem. B 104 (2000) 10001.
[382] D.W. Lewis, C.M. Freeman, C.R.A. Catlow, J. Phys. Chem. 99 [416] A. Čižmek, L. Komunjer, B. Subotić, M. Široki, S. Rončević,
(1995) 11194. Zeolites 11 (1991) 258.
[383] L.D. Rollmann, J.L. Schlenker, C.L. Kennedy, G.J. Kennedy, [417] A. Čižmek, L. Komunjer, B. Subotić, M. Široki, S. Rončević,
D.J. Doren, J. Phys. Chem. B 103 (1999) 7175. Zeolites 11 (1991) 810.
[384] L.D. Rollmann, J.L. Schlenker, C.L. Kennedy, G.J. Kennedy, [418] A. Čižmek, L. Komunjer, B. Subotić, M. Široki, S. Rončević,
D.J. Doren, J. Phys. Chem. B 104 (2000) 721. Zeolites 12 (1992) 190.
[385] T.V. Harris, S.I. Zones, in: J. Weitkamp, H.G. Karge, H. [419] C. Kosanović, B. Subotić, V. Kaučič, M. Škreblin, Phys. Chem.
Pfeifer, W. Hölderich (Eds.), Zeolites and Related Microporous Chem. Phys. 2 (2000) 3447.
Materials: State of the Art 1994, Studies in Surface Science and [420] A. Čižmek, L. Komunjer, B. Subotić, R. Aiello, F. Crea, A.
Catalysis, vol. 84, Elsevier, Amsterdam, 1994, p. 29. Nastro, Zeolites 14 (1994) 182.
[386] G. van de Goor, C.C. Freyhardt, P. Behrens, Z. Anorg. Allg. [421] A. Čižmek, B. Subotić, R. Aiello, F. Crea, A. Nastro, C. Tuoto,
Chem. 621 (1995) 311. Micropor. Mater. 4 (1995) 159.
[387] X. Bu, P. Feng, G.D. Stucky, J. Am. Chem. Soc. 120 (1998) [422] A. Čižmek, B. Subotić, I. Šmit, A. Tonejc, R. Aiello, F. Crea,
1338. A. Nastro, Micropor. Mater. 8 (1997) 159.
[388] X. Bu, P. Feng, G.D. Stucky, Chem. Mater. 12 (2000) 1505. [423] C.W. Davies, A.L. Jones, Trans. Faraday Soc. 51 (1955)
[389] W.C. Paik, C.-H. Shin, J.M. Lee, B.J. Ahn, S.B. Hong, J. Phys. 812.
Chem B 105 (2001) 9994. [424] T. Antonić, A. Čižmek, C. Kosanović, B. Subotić, J. Chem. Soc.
[390] B. Han, S.-H. Lee, C.-H. Shin, P.A. Cox, S.B. Hong, Chem. Faraday Trans. 89 (1993) 1817.
Mater. 17 (2005) 477. [425] T. Antonić, A. Čižmek, B. Subotić, J. Chem. Soc. Faraday
[391] P.A. Cox, P. Grewal, P.A. Wright, J. Mater. Chem., submitted Trans. 90 (1994) 1973.
for publication. [426] T. Antonić, A. Čižmek, B. Subotić, J. Chem. Soc. Faraday
[392] S.I. Zones, Y. Nakagawa, Micropor. Mater. 2 (1994) 543. Trans. 90 (1994) 3725.
[393] T. Blasco, A. Corma, M.J. Dı́az-Cabañas, F. Rey, J.A. Vidal- [427] D.G. Hay, H. Jaeger, K.G. Wilshier, Zeolites 10 (1990) 571.
Moya, C.M. Zicovich-Wilson, J. Phys. Chem. B 106 (2002) 2634. [428] R.A. Le Fêbre, Ph.D. Thesis, TU Delft, 1989.
[394] A. Corma, M.T. Navarro, F. Rey, J. Rius, S. Valencia, Angew. [429] M. Ghamami, L.B. Sand, Zeolites 3 (1983) 155.
Chem. Int. Ed. 40 (2001) 2277. [430] R.M. Dessau, E.W. Valyocsik, N.H. Goeke, Zeolites 12 (1992)
[395] A. Corma, M.J. Dı́az-Cabañas, L.J. Martı́nez-Triguero, F. Rey, 776.
J. Rius, Nature 418 (2002) 514. [431] R. Le Van Mao, S. Xiao, A. Ramsaran, J. Yao, J. Mater. Chem.
[396] R. Castañeda, A. Corma, V. Fornés, F. Rey, J. Rius, J. Am. 4 (1994) 605.
Chem. Soc. 125 (2003) 7820. [432] T. Suzuki, T. Okuhara, Micropor. Mesopor. Mater. 43 (2001)
[397] A. Corma, F. Rey, S. Valencia, J.L. Jordá, J. Rius, Nature 83.
Mater. 2 (2003) 493. [433] J.C. Groen, L.A.A. Peffer, J.A. Moulijn, J. Pérez-Ramı́rez,
[398] J.-L. Guth, P. Caullet, R. Wey, in: L.V.C Rees (Ed.), Proc. 5th Micropor. Mesopor. Mater. 69 (2004) 29.
Int. Conf. Zeolites, Heyden, London, 1980, p. 30. [434] T. Sano, Y. Nakajima, Z.B. Wang, Y. Kawakami, K. Soga, A.
[399] P. Caullet, J.-L. Guth, G. Hurtrez, R. Wey, Bull. Soc. Chim. Fr. Iwasaki, Micropor. Mater. 12 (1997) 71.
(Part I), (1981) I-253. [435] A. Iwasaki, T. Sano, Zeolites 19 (1997) 41.
[400] P. Rouet, A. Wolff, J.-L. Guth, R. Wey, Rev. Chim. Min. 22 [436] K.V. Ragnarsdottir, C.M. Graham, G.C. Allen, Chem. Geol.
(1985) 226. 131 (1996) 167.
[401] J.-L. Guth, P. Caullet, R. Wey, in: B. Držaj, S. Hočevar, S. [437] S. Yamamoto, S. Sugiyama, O. Matsuoka, K. Kohmura, T.
Pejovnic (Eds.), Zeolites: Synthesis, Structure, Technology and Honda, Y. Banno, H. Nozoye, J. Phys. Chem. 100 (1996) 18474.
Application, Studies in Surface Science and Catalysis, vol. 24, [438] W.H. Casey, H.R. Wetrich, G.W. Arnold, Geochim. Cosmo-
Elsevier, Amsterdam, 1985, p. 183. chim. Acta 52 (1988) 2785.
[439] W. Ostwald, Lehrbuch der Allgemeinen Chemie, vol. 2, Engel- [471] S.I. Zones, R.A. Van Nordstrand, Zeolites 8 (1988) 166.
mann, Leipzig, 1896, p. 444. [472] S.I. Zones, R.A. Van Nordstrand, Zeolites 8 (1988) 409.
[440] W. Ostwald, Z. Phys. Chem. 22 (1897) 289. [473] S.I. Zones, J. Chem. Soc. Faraday Trans. 87 (1991) 3709.
[441] F. Feng, K.J. Balkus Jr., Micropor. Mesopor. Mater. 69 (2004) [474] S.I. Zones, Y. Nakagawa, in: L. Bonneviot, S. Kaliaguine
85. (Eds.), Zeolites: a Refined Tool for Designing Catalytic Sites,
[442] U. Oberhagemann, P. Bayat, B. Marler, H. Gies, J. Rius, Studies in Surface Science and Catalysis, vol. 97, Elsevier,
Angew. Chem. Int. Ed. 35 (1996) 2869. Amsterdam, 1995, p. 45.
[443] S. Vortmann, J. Rius, S. Siegmann, H. Gies, J. Phys. Chem. B [475] S.I. Zones, L.T. Yuen, Y. Nakagawa, R.A. Van Nordstrand,
101 (1997) 1292. S.D. Toto, in: R. von Ballmoos, J.B. Higgins, M.M.J. Treacy
[444] S. Vortmann, J. Rius, B. Marler, H. Gies, Eur. J. Mineral. 11 (Eds.), Proc. 9th Int. Zeolite Conf., Butterworth-Heinemann,
(1999) 125. Boston, 1993, p. 163.
[445] U. Lohse, B. Altrichter, R. Fricke, W. Pilz, E. Schreier, Ch. [476] R.K. Ahedi, Y. Kubota, Y. Sugi, J. Mater. Chem. 11 (2001)
Garkisch, K. Jancke, J. Chem. Soc. Faraday Trans. 93 (1997) 2922.
505. [477] R.J. Plaisted, M.Phil. Thesis, UMIST, 2003.
[446] J.M. Garcés, S.C. Rocke, C.E. Crowder, D.L. Hasha, Clays [478] C.S. Cundy, unpublished work.
Clay Miner. 36 (1988) 409. [479] A. Iwasaki, T. Sano, Y. Kiyozumi, Micropor. Mesopor. Mater.
[447] A. Tuel, Chem. Mater. 11 (1999) 1865. 25 (1998) 119.
[448] T.V. Whittam, Eur. Patent 0054364 (1982). [480] R. Kumar, A. Bhaumik, R.K. Ahedi, S. Ganapathy, Nature 381
[449] A.J. Blake, K.R. Franklin, B.M. Lowe, J. Chem. Soc. Dalton (1996) 298.
Trans. (1988) 2513. [481] R. Kumar, P. Mukherjee, R.K. Pandey, P. Rajmohanan, A.
[450] R. Millini, G. Perego, W.O. Parker, Jr., G. Bellussi, L. Carluccio, Bhaumik, Micropor. Mesopor. Mater. 22 (1998) 23.
Micropor. Mater. 4 (1995) 221. [482] Y. Luo, J. Sun, W. Zhao, J. Yao, Q. Li, Chem. Mater. 14 (2002)
[451] L. Schreyeck, P. Caullet, J.-C. Mougenel, J.-L. Guth, B. Marler, 1906.
J. Chem. Soc. Chem. Commun. (1995) 2187. [483] D.M.P Mingos, D.R. Baghurst, Chem. Soc. Rev. 20 (1991) 1.
[452] L. Schreyeck, P. Caullet, J.-C. Mougenel, J.-L. Guth, B. Marler, [484] C. Gabriel, S. Gabriel, E.H. Grant, B.S.J. Halstead, D.M.P.
Micropor. Mater. 6 (1996) 259. Mingos, Chem. Soc. Rev. 27 (1998) 213.
[453] A. Corma, V. Fornes, S.B. Pergher, Th.L.M. Maesen, J.G. [485] N. Elander, J.R. Jones, S.-Y. Lu, S. Stone-Elander, Chem. Soc.
Buglass, Nature 396 (1998) 353. Rev. 29 (2000) 239.
[454] A. Corma, U. Diaz, M.E. Domine, V. Fornes, J. Am. Chem. [486] A. Loupy (Ed.), Microwaves in Organic Synthesis, Wiley-VCH,
Soc. 122 (2000) 2804. Weinheim, 2002.
[455] M.K. Rubin, P. Chu, US Patent 4,954,325 (1990). [487] C.R. Strauss, Aust. J. Chem. 52 (1999) 83.
[456] S.L. Lawton, A.S. Fung, G.J. Kennedy, L.B. Alemany, C.D. [488] A. Arafat, J.C. Jansen, E.R. Ebaid, H. van Bekkum, Zeolites 13
Chang, G.H. Hatzikos, D.N. Lissy, M.K. Rubin, H.-K.C. (1993) 162.
Timken, S. Steuernagel, D.E. Woessner, J. Phys. Chem. 100 [489] J.P. Zhao, C.S. Cundy, J. Dwyer, in: H. Chon, S.-K. Ihm, Y.S.
(1996) 3788. Uh (Eds.), Progress in Zeolites and Microporous Materials,
[457] J.C. Cheng, T.F. Degnan, J.S. Beck, Y.Y. Huang, M. Kalyan- Studies in Surface Science and Catalysis, vol. 105, Elsevier,
araman, J.A. Kowalski, C.A. Loehr, D.N. Mazzone, in: H. Amsterdam, 1997, p. 181.
Hattori, K. Otsuka (Eds.), Science and Technology in Catalysis [490] J.H. Koegler, A. Arafat, H. van Bekkum, J.C. Jansen, in: H.
1998, Studies in Surface Science and Catalysis, vol. 121, Chon, S.-K. Ihm, Y.S. Uh (Eds.), Progress in Zeolites and
Kodansha-Elsevier, Tokyo-Amsterdam, 1999, p. 53. Microporous Materials, Studies in Surface Science and Catalysis,
[458] A. Erdem, L.B. Sand, J. Catal. 60 (1979) 241. vol. 105, Elsevier, Amsterdam, 1997, p. 2163.
[459] P. Norby, J. Am. Chem. Soc. 119 (1997) 5215. [491] Y.I. Smolin, in: J.S. Falcone (Ed.), Soluble Silicates, ACS Symp.
[460] F.G. Dwyer, P. Chu, J. Catal. 59 (1979) 263. Ser., vol. 194, 1982, p. 329.
[461] A.J. Perrotta, C. Kibby, B.R. Mitchell, E.R. Tucci, J. Catal. 55 [492] M. Wiebcke, M. Grube, H. Koller, G. Engelhardt, J. Felsche,
(1978) 240. Micropor. Mater. 2 (1993) 55.
[462] F. Fajula, M. Vera-Pacheco, F. Figueras, Zeolites 7 (1987) 203. [493] C.T.G. Knight, R.J. Kirkpatrick, E. Oldfield, J. Am. Chem. Soc.
[463] H.E. Robson, K.L. Riley, D.D. Maness, in: J.R. Katzer (Ed.), 109 (1987) 1632.
Molecular Sieves—II, ACS Symp. Ser., vol. 40, 1977, p. [494] S.D. Kinrade, J.C.H. Donovan, A.S. Schach, C.T.G. Knight, J.
233. Chem. Soc. Dalton Trans. (2002) 1250.
[464] Q.-H. Xu, S.-L. Bao, J.-L. Dong, in: Recent Progress Reports [495] L.A. Villaescusa, P.S. Wheatley, R.E. Morris, P. Lightfoot,
and Discussion, 5th Int. Conf. Zeolites, Naples, Italy, 1980, p. Dalton Trans. (2004) 820.
49. [496] G. Férey, J. Solid State Chem. 152 (2000) 37.
[465] K. Katsuki, K. Majima, Y. Yoshino, in: New Developments in [497] F. Taulelle, M. Pruski, J.P. Amoureux, D. Lang, A. Bailly, C.
Zeolite Science and Technology, Preprints of Poster Papers, 7th Huguenard, M. Haouas, C. Gerardin, T. Loiseau, G. Férey, J.
Int. Zeolite Conf., Tokyo, 1986, Jap. Assoc. Zeolite, Tokyo, Am. Chem. Soc. 121 (1999) 12148.
1986, Abstract 3C-7, p. 31. [498] N.N. Ghosh, J.C. Clark, G.T. Eldridge, C.E. Barnes, Chem.
[466] J. Richter-Mendau, W. Lutz, A. Grossmann, Cryst. Res. Commun. (2004) 856.
Technol. 23 (1988) 1245. [499] J. Galy, A. Mosset, I. Grenthe, I. Puigdomènech, B. Sjöberg, F.
[467] B. Subotić, D. Škrtić, I. Šmit, L. Sekanović, J. Cryst. Growth 50 Hultén, J. Am. Chem. Soc. 109 (1987) 380.
(1980) 498. [500] P.V. Solntsev, J. Sieler, A.N. Chernega, J.A.K. Howard, T.
[468] B. Subotić, L. Sekanović, J. Cryst. Growth 75 (1986) 561. Gelbrich, K.V. Domasevitch, Dalton Trans. (2004) 695.
[469] B. Subotić, N. Mašić, I. Šmit, in: B. Držaj, S. Hočevar, S. [501] M. Eddaoudi, D.B. Moler, H. Li, B. Chen, T.M. Reineke, M.
Pejovnic (Eds.), Zeolites: Synthesis, Structure, Technology and OKeeffe, O.M. Yaghi, Acc. Chem. Res. 34 (2001) 319.
Application, Studies in Surface Science and Catalysis, vol. 24, [502] C. Serre, F. Taulelle, G. Férey, Chem. Commun. (2003)
Elsevier, Amsterdam, 1985, p. 207. 2755.
[470] B. Subotić, I. Šmit, O. Madžija, L. Sekanović, Zeolites 2 (1982) [503] F. Taulelle, Curr. Opin. Solid State Mater. Sci. 5 (2001)
135. 397.
[504] Z. Liu, W. Xu, G. Yang, R. Xu, Micropor. Mesopor. Mater. 22 [513] C.-Y. Chen, S.L. Burkett, H.-X. Li, M.E. Davis, Micropor.
(1998) 33. Mater. 2 (1993) 27.
[505] T.E. Gier, G.D. Stucky, Nature 349 (1991) 508. [514] A. Firouzi, D. Kumar, L.M. Bull, T. Besier, P. Sieger, Q. Huo,
[506] T.M. Nenoff, W.T.A. Harrison, T.E. Gier, G.D. Stucky, J. Am. S.A. Walker, J.A. Zasadzinski, C. Glinka, J. Nicol, D. Margo-
Chem. Soc. 113 (1991) 378. lese, G.D. Stucky, B.F. Chmelka, Science 267 (1995) 1138.
[507] M. Wallau, J. Patarin, I. Widmer, P. Caullet, J.L. Guth, L. [515] G.D. Stucky, A. Monnier, F. Schüth, Q. Huo, D. Margolese, D.
Huve, Zeolites 14 (1994) 402. Kumar, M. Krishnamurty, P. Petroff, A. Firouzi, M. Janicke,
[508] N.Z. Logar, N. Rajič, V. Kaučič, Lj. Golič, J. Chem. Soc. Chem. B.F. Chmelka, Mol. Cryst. Liq. Cryst. 240 (1994) 187.
Commun. (1995) 1681. [516] J. Zhang, Z. Luz, D. Goldfarb, J. Phys. Chem. B 101 (1997)
[509] W.T.A. Harrison, T.E. Gier, K.L. Moran, J.M. Nicol, H. 7087.
Eckert, G.D. Stucky, Chem. Mater. 3 (1991) 27. [517] S.M. Holmes, V.L. Zholobenko, A. Thursfield, R.J. Plaisted,
[510] E.M. Flanigen, personal communication (1990). C.S. Cundy, J. Dwyer, J. Chem. Soc. Faraday Trans. 94 (1998)
[511] Special issue on metal-organic open frameworks, Micropor. 2025.
Mesopor. Mater. 73 (1–2) (2004). [518] J.-L. Guth, P. Caullet, J. Chim. Phys. 83 (1986) 155.
[512] P.M. Forster, A.R. Burbank, C. Livage, G. Férey, A.K. [519] S.I. Zones, Y. Nakagawa, G.S. Lee, C.Y. Chen, L.T. Yuen,
Cheetham, Chem. Commun. (2004) 368. Micropor. Mesopor. Mater. 21 (1998) 199.

Cundy Dan Cox, 2005

Uploaded by

Copyright:

Available Formats

Cundy Dan Cox, 2005

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cundy Dan Cox, 2005

Uploaded by

Copyright:

Available Formats

Microporous and Mesoporous Materials 82 (2005) 1–78

The hydrothermal synthesis of zeolites: Precursors,

Received 5 November 2004; accepted 11 February 2005

Keywords: Hydrothermal synthesis; Nucleation; Crystal growth; Modelling; Mechanism

1.2. Scope and structure of this review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

10.3. Non-aqueous syntheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

3.1. Richard Barrer 1. Extensive heterogeneous nucleation occurs during

Zhdanov [57] Na-A, Na-X Solid M liquid solubility equilibrium,

Derouane, Detremmerie, Na,TPA-ZSM-5 Synthesis ‘‘A’’: liquid phase ion

Chang and Bell [63] Na,TPA-Si-ZSM-5 Embryonic clathrate TPA-silicate units,

Burkett and Davis [64–66] TPA-Si-ZSM-5 Pre-organised inorganic–organic composites,

Leuven Group [67–73] TPA-Si-ZSM-5 Oligomers ! precursor ‘‘trimer’’ (33 Si)

3.5. Overview—1959 to 1971 and beyond 3.6. Introduction of organic templates

ratio of precursor and product solubility constants (Ks),

4.2. Molecular modelling

In terms of zeolite synthesis, molecular modelling

4.2.1. Modelling zeolite-template pairs

5. The induction period 6. The evolution of order

6.2.1. The Montpellier study the preparation of aluminous (inorganic) zeolites

secondary amorphous phases (Fig. 5 and Section 7.8)

Free energy (∆G)

nucleation phenomena. However, much useful data can

7.3.1. Studies under isothermal conditions

7.3.2. Ageing studies

treating gel dissolution and nucleation as interfacial phe-

7.6. Nucleation in zeolite systems—The nature of the

In Sections 7.3–7.5, nucleation has been considered

In Sections 5–7, the pathway of zeolite synthesis has

8.1. Experimental methods

8.6. Growth models

The slow linear growth rates of zeolite crystals at

this is a greatly oversimpliﬁed representation. As one

dicted a broadening of the particle size distribution over

9.4.2. Chemical and physical consequences of an

10.1. Hydrothermal synthesis in the presence of a liquid

The original suggestion by Flanigen and Breck [51]

do not transform under comparable conditions as the

10.6. Summary and conclusions

the ultimate crystal size decreased (Fig. 22). At the point

(# (g crystal mass)-1 x 1014)

Crystal size (nm)

to investigate indirectly the development of order in

12. X-ray amorphous zeolites

Earlier (Section 6), we have considered some of the

12.1. XRD evidence for ‘‘X-ray amorphous zeolites’’

Although templates appear to require a speciﬁc shape

13.1. Geometric matching

Molecular modelling calculations based on a molecu-

side two diﬀerent products, LEV and NON. Despite the

13.3. Structure blocking

The structure directing agent (SDA) is usually added

* via further equilibria with intermediate species such as

Ks(gel) nary ammonium complexes of zeolites ZSM-5 and beta

The topic of zeolite solubility is not a simple one. It is

Table 4 ried out elsewhere [478], the nucleation periods and

cations (and additional sodium cations in the case of

ions and cations which already exist in the synthesis

You might also like