Improved molecular conductance predictions using wavefunction-in-DFT quantum embedding
Year: 2026
Type: article
Abstract: A novel electronic structure methodology to describe electron transport in single-molecule junctions (SMJs) within non-equilibrium Green's function theory is presented. The approach is based on a formally exact, projection-based quantum embedding technique that combines correlated many-electron wavefunction models for the molecular region with a density functional theory (DFT) description of the metallic electrodes. This is achieved by constructing a specialized Hamiltonian for the molecular domain, leveraging Dyson orbitals corresponding to the ionized and electron-attached states of the embedded molecule. The effectiveness of this wavefunction-in-DFT embedding scheme is demonstrated through transport calculations for SMJs containing benzene-1,4-diamine and its substituted derivatives, employing Hartree-Fock, SOS-ADC(2), and CCSD methods for the molecular subsystem. The results show a marked improvement in the predicted zero-bias conductance compared to conventional DFT-based transport modeling employing the PBE functional. The proposed methodology provides a systematic way to select the most suitable electronic structure methods for the different parts of the system, maintaining a balance between accuracy and computational cost, while ensuring a proper description of electronic correlation within the molecule, which may notably impact electron transport in certain systems.
Source: The Journal of Chemical Physics
DOI: 10.1063/5.0313307
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
Awards: G1262964987, G5212758639, G6581278598
OpenAlex ID: W7127369869