This directory contains the data and scripts to calculate lattice distortions and superconducting critical temperatures for electron-doped MoS₂ monolayer on large supercells, as shown in this paper:
Nina Girotto Erhardt, Jan Berges, Samuel Poncé, and Dino Novko, Understanding the origin of superconducting dome in electron-doped MoS₂ monolayer, arXiv:2412.02822
Reproducing the calculations requires the installation of the Python packages elphmod and StoryLines (e.g., in a virtual environment):
python3 -m pip install elphmod==0.29 storylines==0.15
We use the electron-phonon coupling data from Phys. Rev. X 13, 041009 (2023),
see Fig. 9, which is located in the folder data. To reduce the computational
cost, we extract a three-band model (Mo dz2, dx2-y2, and dxy orbitals)
from this five-band model (which also contains Mo dxz and dyz orbitals):
python3 523.py
To further reduce the cost, we map the coupling to a nearest-neighbor model, as described in Appendix B of Phys. Rev B 101, 155107 (2020):
python3 model_optimize.py
python3 model_plot.py
Now we are ready to relax the structure on different supercells, employing the “model III” presented in SciPost Phys. 16, 046 (2024):
python3 relax_2x2.py
python3 relax_8x8.py
Finally, we calculate the doping-dependent superconducting critical temperature as a function of doping, which yields the well-known dome structure (differences between the calculated points and the reference lines are due to insufficient k- and q-point densities):
mpirun phases_2x2.py
python3 phases_2x2_plot.py
The calculations on the largest supercell are best performed on a supercomputer. The results are plotted for Fig. 6 and Supplementary Figure 7 of the paper:
sbatch phases_18sqrt3.sh
python3 phases_18sqrt3_plot.py
A presentation of the relaxed structures for all dopings can also be created:
python3 structure_plot.py phases_18sqrt3/*.xyz
pdflatex phases_18sqrt3_overview.tex