qm9_conj_OT-w

This database comprises a curated collection of 21,085 conjugated molecules, filtered from the original QM9 dataset. For each molecule, calculations were performed using the LC-ωPBE/6-31G* method. To account for the system dependence of the range-separated parameter ω in the LC-ωPBE approach, the IP method was employed to fine-tune ω, ensuring an optimal value for each molecule. Based on these refined parameters, further calculations were performed to determine molecular properties, extracting key data such as the Hamiltonian matrix, overlap matrix, eigenvalues, and eigenvectors.

The dataset is stored in a DB file format, with each entry containing the following information:

i: Index of the molecule

SMILES: SMILES representation of the molecule

omega: Value of the range-separated parameter ω

j2: Equation value constructed when tuning ω using the IP method

coordinates: Three-dimensional coordinates of atoms in the molecule

Z: Atomic numbers of the atoms in the molecule

n_atoms: Number of atoms in the molecule

hamiltonian: Hamiltonian matrix of the molecule

eigenvalues: Eigenvalues of the molecule

overlap: Overlap matrix of the molecular orbitals

eigenvectors: Eigenvectors of the molecule

Usage example :

import sqlite3

conn = sqlite3.connect('qm9_conj_OT-w.db') cursor = conn.cursor()

cursor.execute("SELECT * FROM qm9_data WHERE i = 0")molecule_data = cursor.fetchone()

smiles = molecule_data[1]omega = molecule_data[2] 

conn.close()