Coarse-Grained Model for the Hydrothermal Synthesis of Zeolites

Abstract

Zeolites are the most used solid catalysts in the chemical industry. The hydrothermal synthesis of these porous aluminosilicate crystals is a multistep process that involves the polymerization of silica from solution to form amorphous aggregates, their crystallization, growth by oriented attachment, and eventually dehydration by heating. The molecular pathways by which zeolites are formed from the precursor solution are not yet fully understood. Molecular simulations can play an important role in elucidating these microscopic processes but are limited by the lack of models able to represent both the polymerization of silica and its crystallization with feasible computational costs. Here, we present a simple, computationally efficient coarse-grained model for the hydrothermal synthesis of zeolites from aqueous solutions. Our TS + W model has three components: silica, a structure-directing agent, and water, each represented by a single particle with short-range interactions. The model quantitatively reproduces the experimental evolution of silica species during the polymerization from solution, yielding amorphous nanoparticles with shape, composition, and silica speciation in agreement with the amorphous precursors of zeolites in experiments. Importantly, the TS + W model spontaneously nucleates and grows zeolites from the amorphous precursor phase and reproduces the temperature required for their dehydration. Our simulations indicate that the preferential formation of zeolites at the amorphous–water interface does not arise from an ability of that surface to heterogeneously nucleate the zeolite. The simplicity and computational efficiency of the TS + W model make it ideal to investigate the interplay between polymerization and crystallization in zeolite synthesis.

What Is the Smallest Zeolite That Could Be Synthesized?**

“…The Gibbs‐Thomson equation assumes that the zeolite is the ordered structure: it cannot predict the particle size at which the crystal phase ends, [16] giving rise to a prevalence of more compact size‐dependent isomers. We investigate this limit using molecular dynamics simulations with the TS+W model, [17] which is first to represent all stages of hydrothermal synthesis, [17] and predicts relative energies of the most common zeolites comparable to those determined in experiments [18] (Supp.Table S4).The model reproduces the experimental structure of silica oligomers in the early stages of synthesis, the Q n speciation of silica as it polymerizes into nanoscopic amorphous precursors, and their crystallization into a zeolite [17] …”

“…Z1 has 10‐member ring channels, [19c] same as silicalite‐1, [15] with a framework density about 30 % lower, comparable to the one of faujasite. Z1 has about 15 % undercoordinated silica sites [19a] that render it 5.1 kJ mol −1 less stable than silicalite‐1, and a relatively low =560 °C [17] . We use molecular simulations to compute the zeolite to amorphous equilibrium temperature of Z1 nanoparticles (some of them shown in the middle row of Figure 1b) immersed in aqueous solution of SDA.…”

“…To assess the relative stability of small zeolites and competing structures, we investigate the nucleation of Z1 zeolite in cooling simulations of amorphous nanoparticles with diameter 9 to 2 nm (Supporting Information A.2). The ratio of silica to SDA in the aqueous synthesis mixture is the one that promotes zeolite Z1, 3 0.3 [17, 19c] and is the same ratio used in the synthesis of silicalite‐1 [20] . Figure 1b shows snapshots of perfect Z1 nanoparticles of various sizes (mid row), as well as the structures obtained after their melting and recrystallization in cooling simulations (bottom row).…”

“…The model reproduces the experimental structure of silica oligomers in the early stages of the Q n speciation of silica as it polymerizes into nanoscopic amorphous precursors, and their crystallization into a zeolite. [17] The current parameterization of SDA in TS + W stabilizes zeolite Z1, [17] which has the structure of FIR-30 metal organic framework, [19] and has not yet been experimentally realized as a silica zeolite. Z1 has 10-member ring channels, [19c] same as silicalite-1, [15] with a framework density about 30 % lower, comparable to the one of faujasite.…”

What Is the Smallest Zeolite That Could Be Synthesized?**

“…The model reproduces the experimental structure of silica oligomers in the early stages of synthesis, the Q n speciation of silica as it polymerizes into nanoscopic amorphous precursors, and their crystallization into a zeolite. [17] The current parameterization of SDA in TS + W stabilizes zeolite Z1, [17] which has the structure of FIR-30 metal organic framework, [19] and has not yet been experimentally realized as a silica zeolite. Z1 has 10-member ring channels, [19c] same as silicalite-1, [15] with a framework density about 30 % lower, comparable to the one of faujasite.…”