DMFT code

This repository contains code for the dynamical mean field theory (DMFT) calculation.

This code was used to obtain the results in the following paper:

Dynamical Mean-Field Theory of Moiré Bilayer Transition Metal Dichalcogenides: Phase Diagram, Resistivity, and Quantum Criticality (Phys. Rev. X 12, 021064)

Code author: Jiawei Zang (jiaweizang)

Description

We use the single-site DMFT with the continuous-time hybridization expansion solver as implemented in the TRIQS software library to perform the calculation. The results are obtained first on the imaginary axis, and we do an analytical continuation to get the results on the real frequency axis.

Paramagnetic state:

• paramagnetic.py runs the DMFT calculation for the paramagnetic state.
• para_tail_fitting.ipynb performs a tail fitting of the self energy.
• para_AC_1.py does the first step of the analytical continuation for the self energy, after it has been tail fitted.
• para_AC_2.py does the second step of the analytical continuation for the self energy.

Magnetic state:

• magnetic.py runs the DMFT calculation for the magnetic state.
• mag_tail_fitting.ipynb performs a tail fitting of the self energy.
• mag_AC_1.py does the first step of the analytical continuation for the self energy, after it has been tail fitted.
• mag_AC_2.py does the second step of the analytical continuation for the self energy.

note: except for para_AC_1.py and mag_AC_1.py, all the other files utilize MPI library, thus they can be submitted to the cluster to significantly reduce the running time. job.sh provides an example for the submission script.

After we get the self energy on the real frequency, we use Eq.2 and 8 in the paper to calculate the resistivity.

Required Packages

• TRIQS
• Maxent