Perceiving Zeolite Self-Assembly: A Combined Top-Down and Bottom-Up Approach within the Tiling Model

Abstract

We present an extended tiling model of the assembly of the zeolite frameworks, where the building units can be natural tiles (minimal cages of the framework) as well as parts of the tiles. We have enumerated all possible methods of splitting each of the 400 natural tiles found in 248 known zeolite frameworks into halves and revealed 199 half-units that can play a role of building units in the zeolite assembly. We have showed that the observed intimate or epitaxial intergrowth of zeolite minerals can be perceived as a result of several possible ways of crystal growth with different methods of connection of the half-units from the same set. We have considered many zeolite pairs, for which the intergrowth was either observed or not revealed yet, and demonstrated that our model properly predicts and interprets the intergrowth phenomena.

Pore topology analysis in porous molecular systems

“…In this context, the face-to-face orientation of porous molecules is reminiscent of the tiling representation of porous networks [64] and assembling porous structures (e.g. zeolites) from tiles (cages fused by faces) [65]. For example, the hexagonal faces of the chair-shaped adamantane-like cages in the structure FOXLAG (figure 2) can be superimposed onto each other and their fusion would lead to the dia net, composed of the tiles of the same topology, 6 4 , as the 2,3M10-1 skeletons (figure 10).…”

Controlling Nucleation Pathways in Zeolite Crystallization: Seeding Conceptual Methodologies for Advanced Materials Design

“…Their detailed investigation into graph similarity highlights that framework similarity among zeolites is likely more complex than the identification of common rings, building units, or (half) tiles. 143 Okubo and co-workers 139 used machine learning (ML) to create a similarity network of zeolite frameworks (Figure 6A). Rather than using mathematically modeled crystal lattices, however, their ML algorithm was trained based on synthesis gel components and compositions.…”

“…They demonstrated that a high degree of graph similarity is an excellent predictor of which parent–daughter pairs can participate in IZT classes that require two complementary structures, such as diffusionless (topotactic) or (epitaxial) intergrowth, whereas for recrystallization the parent largely decomposes prior to daughter nucleation. Their detailed investigation into graph similarity highlights that framework similarity among zeolites is likely more complex than the identification of common rings, building units, or (half) tiles . Okubo and co-workers used machine learning (ML) to create a similarity network of zeolite frameworks (Figure A).…”

Pore Topology Analysis in Porous Molecular Systems

“…zeolites) from tiles (cages fused by faces). 65 For example, the hexagonal faces of the chair shaped adamantane-like cages in the structure FOXLAG (Fig. 2) can be superimposed onto each other and their fusion would lead to the dia net, composed from the tiles of the same topology, 6 4 , as the 2,3M10-1 skeletons (Fig.…”