This C++ code uses tensor networks Matrix Product State ansatz to study ground state properties of a two-component Bose-Hubbard model in 1D (DMRG/imaginary time TDVP using ITensor v3 C++ implementation).
In particular, you can study exotic phases such as Bose-Bose lattice quantum droplets.
The code provides (details in Jupyten notebook "model.ipynb")
- MPS for the ground state of the system for given set of Hamiltonian parameters. MPS are obtained within within desired accuracy of the entropy at the central bond
- The expectation value of the total Hamiltonian, and its separated parts in the ground state.
- Atom densiteis and one-body correlation matrix.
- Two-body correlation matrix for atomic pairs
$a-b$ . - Density-density correlations for atoms
$a$ and$b$ .
Using:
- Install ITensor v3 library (C++ implementation): https://itensor.org/docs.cgi
- Install TDVP library for ITensor: https://github.com/ITensor/TDVP
- Enter Bose-Hubbard model parameters in bash script run.sh (number of bosons, number of sites, interactions strenght, hopping, etc.).
- Execture "bash run.sh".
Repo contains:
- Source code.
- Jupyter notebook with two-component Bose-Hubbard Hamiltonian and calculated quantities in "model.ipynb".