Enter the formulation of a potential MOF using the SMILES representation of the linker and the metal(s) to be combined, to predict the likelihood of pores of different sizes forming in the resultant structur. Due to limitations in the training dataset only some elements are supported as metal, find the list of elements supported here. Predictions were found to have an accuracy of 80.5% on a test set. This application was reported in The Liverpool Materials Discovery Server: A suite of tools for the collaborative discovery of materials. and is based on the model first reported in Machine-Learning Prediction of Metal–Organic Framework Guest Accessibility from Linker and Metal Chemistry. Please consider citing these papers if you use this in your work.
Click here for information about accessing this model using our API