1. Zeolite Crystallization

Crystallization from solution generally occurs via the sequential steps of nucleation of the phase, or phases, dictated by the composition of the solution, followed by growth of the nuclei to larger sizes by incorporation of material from the solution. Nucleation and crystal growth rates typically are governed by a driving force related to the supersaturation.Molecular sieve zeolites usually are precipitated from aluminosilicate solutions in basic media, frequently at elevated temperatures and autogenous pressures. Most commercially interesting syntheses are preceded by the formation of an amorphous gel phase which dissolves to replace reagents consumed from the solution by crystal growth. Some experimental systems which are "crystal clear" have been developed that permit certain in situ analytical techniques to be used to study the crystallization process.In hydrothermal zeolite systems it is more difficult to identify a "supersaturation," because of the myriad species present in the aluminosiicate solution, because of the role of structure directing agents in some cases, and because the relative concentrations of these in a batch system change as the crystallization proceeds. For these reasons, the issue of defining the precise driving force for zeolite nucleation and crystal growth has yet to be accomplished with any degree of certainty. However, recent progress has been reported in defining solubility products and crystallization diagrams for zeolites NaA and NaX, and further progress can be expected using the approach developed there. [1]

2. Zeolite Nucleation

It is expected that the crystallization processes occurring in hydrothermal zeolite precipitation are similar to those which are known to occur in simpler inorganic or organic crystallization systems. That being the case, one should note that nucleation mechanisms in liquid-solid systems have been divided into several categories, most notably [2,3]:

1. Primary Nucleation

Homogeneous
Heterogeneous

2. Secondary Nucleation

Initial breeding
Contact
Shear
Fracture
Attrition
Needle

Chapter 5 of the text by Randolph and Larson provides excellent background on these mechanisms. [2] The primary references contained in their bibliography also are quite informative.

Primary nucleation is characterized as being driven by the solution itself, either strictly within the solution, as in homogeneous nucleation, or catalyzed by extraneous material in the solution, as in heterogeneous nucleation. Certainly with the presence of amorphous gel in most zeolite synthesis systems, one might anticipate that heterogeneous nucleation on gel surfaces might be important. This has yet to be demonstrated unequivocally.

Secondary nucleation is catalyzed by the presence of parent crystals of the same phase, and occurs with a lower activation energy than primary nucleation. The parent crystals might be added as seed crystals at the beginning of a synthesis, or grown in the original unseeded system. Initial breeding results from the addition of seeds which will be discussed below. The other secondary nucleation mechanisms could stem from added seed crystals or crystals grown in situ.For example, Culfaz and Sand reported that new mordenite crystals appeared to grow from seed crystal surfaces and break off acicular pieces, resulting in nuclei formation as the dendritic pieces grew to macroscopic sizes. [4]

The mechanisms involving secondary nucleation induced by fluid shear, contact (or collision) breeding, and fracture all require sufficient fluid motion to cause physical damage to the parent crystals, and thereby promote formation and release of secondary nuclei from the parent crystal surface. In many zeolite systems, there is no induced fluid motion, as by stirring, and crystal settling is generally viewed as insufficient to cause secondary nucleation. A review of the evidence suggesting that agitation-induced secondary nucleation is not important in zeolite systems was published recently. [5] More recently, an interesting study by Falamaki, et al, has suggested that severe agitation during synthesis of ZSM-5 was not sufficient to either promote nucleation or to break ZSM-5 crystals. [6]

The numerous works by Subotic, et al., and references contained therein, have pointed to the possibility that nuclei form within the amorphous gel matrix, and are released to become viable growing crystals as the gel phase dissolves. [7] This mechanism is strn under review, and yet may be plausible given the new evidence that nanometer-sized particulates (or nanoparticles) have been observed in clear solution synthesis mixtures, and may be the origin of nuclei. [8] It has not yet been determined how or when these nanoparticles form, although recent progress has been made in determining that they form almost instantaneously in the MFI synthesis system. [9]

3. Crystal Growth

Most crystallization processes involve assimilation of material from solution via a growth process which can be described by the relation [2,3]:dL/dt = G = Ks^aWhere a is an exponent expressing the dependence of the linear crystal growth rate, G, on the supersaturation, s, and K is a temperature-dependent rate constant. The value of a will be 1.0 for diffusional transport limitations to a planar crystal surface, and between 1-2 for most surface reaction limited growth processes. [2,3] Schoeman, et al. have analyzed the growth rate behavior for Silicalite in clear solutions using a chronomal analysis and concluded that its growth rate is limited by a first-order surface reaction with an activation energy of 42 kJ/mol. [10] A recent summary of several reports of activation energies for zeolite crystal growth showed values to be in the range of 43-96 kJ/mol, magnitudes which certainly suggest surface kinetics limited growth rather than diffusional limitations. [11]

A debate has arisen in the literature recently regarding what species actually add to the crystal surface to promote growth and whether an agglomeration mechanism plays a role in zeolite crystal growth. Schoeman used the DLVO theory for colloidal stability to predict that nanoparticle agglomeration would not be possible in a Silicalite synthesis solution. [12] He concluded that zeolite crystal growth was supported by addition of low molecular weight species, most likely the monomer. He also concluded that the nanoparticles, observed by several research groups, would be predicted to be stable in Siicalite synthesis solutions. However, Kirschhock, et al. extended Schoemanâs work to show that growth by agglomeration of nanoparticles with crystal surfaces was possible. [13] Specifically, they noted that the nanoparticles should come to rest at about 7 Å from the crystal surface in an energy well, and have ample time to orient and chemically bond to the surface. More recently, Nikolakis, et al. analyzed Silicalite crystal growth and the energetics of nanoparticle-crystal interactions using atomic force microscopy. [14] They also extended the DLVO theory and concluded that zeolite crystal growth by nanoparticle addition was possible, even though their total potential energy curves showed no energy wells or negative values at distances greater than a fraction of an angstrom.

Nothing definitive can be said about this debate at this time. Thus, growth by addition of monomers, low molecular weight species, or nanoparticles cannot be ruled out. However, growth is suggested by the magnitudes of activation energies to be limited by surface reaction kinetics rather than diffusion from the bulk liquid to the crystal surface, and the nanoparticles would be expected to diffuse rather slowly, potentially at rate-limiting rates, due to their very large molecular weights. These observations might lead one to suspect that growth should be by addition

4. Seeding

Adding seed crystals to a crystallization system has typically resulted in increased "crystallizationâ rates. The enhanced rate might be due to simply increasing the rate at which solute is integrated into the solid phase from solution due to the increased available surface area, but also might be the result of enhanced nucleation of new crystals. Understanding the precise role of seed crystals is an area of ongoing investigation.

The secondary nucleation mechanism referred to as initial breeding results from microcrystalline dust being washed off of seed crystal surfaces in a new synthesis batch, and has been reported in zeolite systems. [15] These microcrystalline fragments grow to observable sizes, and result in greatly enhanced Îcrystallizationâ rates due to the significantly increased crystal surface area compared to the unseeded system. Consequently, it is to be expected that addition of seed crystals to a synthesis system will introduce sub-micron sized crystallites into the system which will serve as nuclei.

Finally, it is worth noting that a recent study of initial bred nuclei using a clear synthesis solution [16] suggested that the initial bred nuclei themselves may be the same nanoparticles observed by Schoeman [8], Kirschhock, et al. [9], and independently elsewhere. [17]. That is, the same particulates which appear to catalyze zeolite nucleation in unseeded systems may remain in sufficient number to catalyze nucleation in seeded systems, since they are inherently present with the seed crystal sample, and may be impossible to eliminate by typical filtration techniques.

5. References

[1] J. Sefcik, A. V. McCormick, Chem. Eng. Sci., 54 (1999) 3513
[2] A. D. Randolph, M. A. Larson, Theory of Particulate Processes, 2nd ed., Academic Press, San Diego, 1988
[3] A. G. Jones in Controlled Particle, Droplet, and Bubble Formation, D. J. Wedlock (ed.), Butterworth-Heinemann, Oxford, 1994, p. 61
[4] A. Culfaz, L B. Sand, in Adv. Chem. Ser., No. 121, W. M. Meier, J. B. Uytterhoeven (eds.), ACS, 1973, p. 140
[5] R W. Thompson in Modeling of Structure and Reactivity in Zeolites, C. R A. Catlow (ed.), Academic Press, San Diego, 1992, p. 231
[6] C. Falamaki, M. Edrissi, M. Sohrabi, Zeolites 19 (1997) 2; and personal communication, Edinburgh, July, 1996
[7] B. Subotic in ACS Sym. Ser. 398, M. L Occelli, H. E Robson (eds.), 1989, p. 110; and A. Katovic, B. Subotic, I Smit, Lj. A. Despotovic, M. Curic, ibid,, p. 124
[8] B. J. Schoeman, Zeolites 18 (1997) 97
[9] C. E. A. Kirschhock, R. Ravishankar, F. Verspeurt, P. J. Grobet, P. A. Jacobs, J. A. Martens, J. Phys. Chem. 103 (1999) 4956, and C. F. A. Kirschhock, R Ravishankar, L Van Looveven, P. A. Jacobs, J. A. Martens, J. Phys. Chem. 103 (1999) 4972
[10] B. J. Schoeman, J. Sterte, J.-E Otterstedt, Zeolites 14 (1994) 568
[11] R W. Thompson in Molecular Sieves, Science and Technology, Vol. 1, H. G. Karge, J. Weitkamp (eds.), Springer, Berlin, 1998, p. 21
[12] B. J. Schoeman, Micropor. Mesopor. Mater., 22 (1998) 9
[13] C. F. A. Kirschhock, R. Ravishankar, P. A. Jacobs, J. A. Martens, J. Phys. Chem. B, 103 (1999) 11021
[14] V. Nikolakis, F. Kokkoli, M. Tirrell, M. Tsapatsis, D. G. Vlachos, Chem. Mater. 12 (2000) 845
[15] 5. Gonthier, R W. Thompson in Stud. Surf. 56. Catal., No. 85, J. C. Jansen, M. Stoecker, H. G. Karge, J. Weitkamp (eds.), Elsevier, Amsterdam, 1994, p. 43
[16] L Gora, K. Streletzky, R W. Thompson, G. D. J. Philles, Zeolites 19 (1997) 98
[17] L Gora, K. Streletzky, R W. Thompson, G. D. J. Phillies, Zeolites 18 (1997) 132