Computational prediction of the formation of microporous aluminophosphates with desired structural features

A better understanding of the relationship between the synthetic factors and the resulting structures is of great importance to rationalizing the synthesis of the target zeolitic materials. We present a computational study for predicting the formation of (6,12)-ring-containing microporous aluminophosphates (AlPOs), using a data classification approach. Through analyses of a database of AlPO synthesis with ca. 1600 reaction data, we identified a number of synthetic parameters such as three gel molar ratios of Al₂O₃, P₂O₅ and the organic amine template, as well as eleven parameters associated with the geometric and electronic characteristics of the templates deemed to be useful in distinguishing (6,12)-ring-containing AlPOs from the other AlPOs. Using these parameters, we have trained a support vector machine (SVM)-based classifier on a training dataset containing 360 (6,12)-ring-containing AlPOs and 1069 AlPOs without such rings. Analysis results revealed that the geometric size of the organic template, particularly the second longest distance of the template along with the gel molar ratios, has good predictive power for microporous aluminophosphates containing (6,12)-rings. This work demonstrates the general feasibility in establishing a relationship between the synthetic parameters and the structural features of the synthesized microporous materials, providing a useful guidance to the rational design and synthesis of such materials as well as other inorganic crystalline materials.