PNNL

Abstract

Metal-Organic Frameworks (MOFs) exhibit attractive characteristics for separations such as remarkable surface area and diverse porosities. However, a mechanistic understanding of their synthesis and scale-up remains underexplored due to the complicated nature of building block interactions. In this work, we investigate the collective assembly of building units that have been experimentally observed to initiate MOF nucleation, using MIL-101(Cr) as a prototypical example. We use large-scale molecular dynamics simulations under a variety of synthesis conditions and mixture compositions. We observe that the choice of solvent (water or DMF), introduction of ions (Na+, F-) and the relative population of MIL-101(Cr) half-secondary building unit (half-SBU) isomers has a strong influence on the cluster formation process. In more detail, the shape, size, nucleation and growth rates, crystallinity and short and long-range order largely vary depending on the synthesis conditions. We evaluate these properties as they naturally emerge when interpreting self-assembly of MOF nuclei as the time-evolution of an undirected graph. Solution-induced conformational complexity and ionic concentration have a dramatic effect on the morphology of clusters emerging during assembly, such diversity is captured by key features of the graph representation. More precisely, pure solvent leads to rapid formation of a small number of large clusters, while ions result in slower nucleation through smaller clusters in water. Finally, we use Principal Component Analysis (PCA) on graph properties to successfully deconvolute MOF self-assembly into a small number of molecular descriptors, such as the average coordination number between half-SBUs and fractal dimension, which can be followed by time-resolved spectroscopy. We conclude that graph theory can be used to understand complex processes such as MOF nucleation through providing molecular descriptors accessible by both simulation and experiment.