(197ai) Effect of Molecular Structure on the HOMO-LUMO GAP Using Atomic Signatures As Molecular Descriptors

Compounds with low HOMO‒LUMO gap E_gap are in need as they are kinetically unstable making them reactive and desirable for different applications. However, the search for new organic compounds with a low E_gap is an expensive endeavor due to the exponentially increasing pool of virtual compounds. Thus, there is a need to construct a robust, predictive, and computationally inexpensive quantitate structure property relationship (QSPR) to identify and screen for structures possessing low E_gap. Herein, a QSPR is proposed that utilizes atomic Signatures as molecular descriptors correlating the molecular structure with the E_gap. A dataset consisting of 112 organic compounds was used to construct the QSPR using forward stepping multilinear regression (FSMLR) with leave-one-out cross validation (LOOCV), achieving a regression coefficient (r²) and predictability (q²) of 0.86 and 0.76, respectively. The QSPR was able to infer relations between different structural motifs (i.e. atomic Signatures) and the E_gap. It was found that the atomic Signature encapsulating aromatic bonds, [C](p[C]p[C][H]), was able to explain around 50% of the variance in the data, showing its significance in affecting the property of interest. The five most significant Signatures were found to negatively affect the E_gap, due to them encapsulating π‒bonds, heteroatoms, and aromatics. This is attributed to high electron density, making these structural motifs reactive sites, specifically due to Ï€ ‒ Ï€ interactions, lone pair electrons, and/or delocalized π‒bonds. In constant the presence of terminal nitrogen or internal alkane chain groups causes an increase in the E_gap, which can be attributed to steric hindrance and non‒conjugated bonding. Finally, an external test set were used to further elucidate the predictive ability of the model ‒ an r² of 0.91 was achieved. Overall, the constructed QSPR can be used as a reliable initial screening tool to identify potential candidates possessing low E_gap values