Unraveling the polymorphism and optoelectronic properties of coumarin via high-throughput computational screening

Coumarin and its derivatives are privileged scaffolds in medicinal chemistry, yet their polymorphism poses significant challenges in pharmaceutical solid-form development. This work explores the crystal polymorphs and optoelectronic properties at ground state of coumarin. The crystal energy landscape of coumarin was constructed, identifying P4₃, Pccn, and P4₂bc as the dominant competing polymorphic structures. Electronic structure calculations reveal that the symmetry of crystal packing and intermolecular interactions significantly modulate the optical absorption edge and spectral response range, with the P4₃ phase exhibiting the smallest band gap (2.69 eV) due to its unique helical packing arrangement.

Crystal structures were generated using the PyXtal library across 230 space groups. High-throughput geometric optimization was performed using the universal machine learning interatomic potential DPA-3. Low-energy candidate structures were further refined via density functional theory (DFT) calculations using the Quantum ESPRESSO package with the PBE functional and DFT-D3 dispersion correction. Optical properties were derived from the complex dielectric function computed from first principles.