learn-dmft

Beginner material for learning how a DMFT calculation works. The series starts from scratch focusing on the 2D square-lattice Hubbard model. We use TRIQS for the Green's functions, meshes and Fourier transforms.

While the main goal of this notebook series is learning DMFT and how to implement it, we also use it as an opportunity to motivate the design of the ModEST, a TRIQS application, which is designed for the implementation of generic quantum embedding calculations, like DMFT, with electronic structure.

notebook	builds
1	the dispersion, the k-grid and its weights, the DOS
2	Brillouin zone summation, the density, the chemical potential
3	the Weiss field, the impurity levels and the hybridization
4	an impurity solver by hand: iterated perturbation theory
5	the DMFT loop and its convergence diagnostics
6	the $T$–$U$ phase diagram of the paramagnetic Mott transition
7	two sublattices, one impurity: the $\psi$-map and an AFM solution

The notebooks are committed with their outputs, so they can be read straight through without running anything.

Requirements

TRIQS, numpy, scipy, matplotlib.