Jiezi: an open-source Python software for simulating quantum transport based on non-equilibrium Green's function formalism
Description
We present a Python-based open-source library named Jiezi, which provides the means of simulating the electronic transport properties of nanoscaled devices on the atomistic level. The key feature of Jiezi lies in its core algorithm, i.e., self-consistent orchestration between the non-equilibrium Green's function (NEGF) method and a Poisson's equation solver. Beyond the construction of the tight-binding (TB) Hamiltonian with empirical parameters for conventional materials, the package offers a comprehensive framework for constructing the Wannier-based Hamiltonian matrix, enabling the investigation of novel materials and their heterostructures. To expedite the solution of NEGF systems, a methodology based on renormalization theory is proposed for reducing the dimension of the Hamiltonian matrix. Additionally, we adopt a non-linear Poisson equation solver with no analytical approximation in this software. The software facilitates seamless integration with external tools for geometry and mesh generation and post-processing. In this paper, we present the main capabilities and workflow by demonstrating with a simulation for the carbon nanotube field-effect transistor (CNTFET).