Beshir Awol of Mekdela Amba University introduced DeePTB-NEGF, a new framework designed to accelerate quantum transport simulations in two-dimensional materials. The method integrates a deep learning-based tight-binding Hamiltonian, derived from first-principles calculations, with the DPNEGF simulation package. This approach addresses the high computational costs traditionally associated with combining Non-Equilibrium Green’s Function (NEGF) formalism and Density Functional Theory (DFT), which have historically restricted large-scale or high-throughput studies.

Validation of the DeePTB-NEGF framework was performed on five prototypical 2D materials: graphene, hexagonal boron nitride (h-BN), MoS2, WS2, and black phosphorus. The model demonstrated excellent agreement with conventional DFT-NEGF results regarding transmission spectra and band structures. Beyond basic validation, the researchers utilized the tool to explore complex scenarios including strain engineering, such as uniaxial strain on graphene and biaxial strain on MoS2, as well as substitution doping and current-voltage characteristics of a graphene field-effect transistor (FET).

A performance scaling analysis indicates that DeePTB-NEGF can successfully simulate systems containing hundreds of atoms in just minutes. For complex heterostructures, such as graphene/h-BN/graphene, the framework achieves a performance speed-up of over 700x compared to standard DFT-NEGF methods. These capabilities position DeePTB-NEGF as a viable tool for the autonomous, high-throughput design of quantum transport in microscopic heterostructures, facilitating the rapid prototyping of next-generation nanoelectronic devices.

Beshir Awol of Mekdela Amba University introduced DeePTB-NEGF, a new framework designed to accelerate quantum transport simulations in two-dimensional materials. The method integrates a deep learning-based tight-binding Hamiltonian, derived from first-principles calculations, with the DPNEGF simulation package. This approach addresses the high computational costs traditionally associated with combining Non-Equilibrium Green’s Function (NEGF) formalism and Density Functional Theory (DFT), which have historically restricted large-scale or high-throughput studies.

Validation of the DeePTB-NEGF framework was performed on five prototypical 2D materials: graphene, hexagonal boron nitride (h-BN), MoS2, WS2, and black phosphorus. The model demonstrated excellent agreement with conventional DFT-NEGF results regarding transmission spectra and band structures. Beyond basic validation, the researchers utilized the tool to explore complex scenarios including strain engineering, such as uniaxial strain on graphene and biaxial strain on MoS2, as well as substitution doping and current-voltage characteristics of a graphene field-effect transistor (FET).

A performance scaling analysis indicates that DeePTB-NEGF can successfully simulate systems containing hundreds of atoms in just minutes. For complex heterostructures, such as graphene/h-BN/graphene, the framework achieves a performance speed-up of over 700x compared to standard DFT-NEGF methods. These capabilities position DeePTB-NEGF as a viable tool for the autonomous, high-throughput design of quantum transport in microscopic heterostructures, facilitating the rapid prototyping of next-generation nanoelectronic devices.