Novel and efficient Graph neural network (GNN) for accurate chemical property prediction

    公开(公告)号:US20220406416A1

    公开(公告)日:2022-12-22

    申请号:US17843341

    申请日:2022-06-17

    摘要: A method for selecting a material having a desired molecular property comprises generating a combinatorial library of molecule structures derived from a core molecular structure, splitting the library into a training set configured to train a graph neural network (GNN) machine learning (ML) model, a test set configured to test the validity of and assess accuracy of the GNN model, and a prediction set where predictions are made using the GNN model, optimizing geometries of the molecular structures, computing excited state energies of the optimized geometries, encoding molecular structure information into a matrix, determining three mutually orthogonal principal axes, transforming spatial coordinates into mutually orthogonal coordinates, constructing a molecular graph with n nodes, feeding the molecular graph into the GNN model as an input, and selecting a material having a suitable desired molecular property based on the output of the GNN model.

    Organic electroluminescent materials and devices

    公开(公告)号:US10121975B2

    公开(公告)日:2018-11-06

    申请号:US14302042

    申请日:2014-06-11

    摘要: A compound comprising a ligand LA selected from: as well as, devices and formulations containing the compound are disclosed. In the compounds, independently X1-X6 are CH or N; Y1-Y12 are independently selected from CH, N, and C-APR1′R1″; when present, exactly one of Y1 through Y12 is C-APR1′R1″ in LA8; A is selected from a single bond, —CRARB—, —NRA—, —O—, —S— and —SiRARB—; and R, R1′, R1″, R2, R3, RA, and RB are each independently a substituent selected from hydrogen, deuterium, alkyl, cycloalkyl, aryl, and combinations thereof. The P-atom of the ligand LA is bonded to a metal M having an atomic weight of at least 40.