AbinitioDGA is an implementation of a DGA approach for electronic structure calculations of materials. It allows the inclusion of non-local correlations beyond DMFT and the calculation of momentum-dependent susceptibilities. The method and the program are described in PRB:95.115107, arXiv:1610.02998 and arXiv:1710.06651. Please cite arXiv:1710.06651 if you use this package in one of your publications.
- HDF5 library (>= 1.8)
- LAPACK library
- BLAS library
- Fortran90 Compiler (with MPI support)
There are 3 custom library include files provided which work on the according clusters. These can be found in make_configs/make_config_* and are detected by make_configs/make_config_auto.sh. You can add your custom cluster include file to the make_configs folder and extend the detection script or simply add your include file in the root folder of the repository under the name make_config.
For more detailed instructions please visit documentation/README.pdf and documentation/configspec.
The default output of ADGA is a partially compressed HDF5 file. For that reason we strongly recommend using Python for post-processing purposes. Our recommended Python libraries are:
- h5py
- numpy
- scipy
- matplotlib
These packages are however not necessary for the purely computational part of ADGA. If one does not want to deal with Python it is still possible to output all data in form of text-files (see documentation/configspec). The instructions in documentation/hdf5_intro.pdf and the scripts found in documentation/scripts/ provide a first starting point for extracting data and making plots.
The test data and results within ADGA for our usual testbed SrVO3 (strongly correlated metal) can be found in srvo3-testdata/ and documentation/README.pdf respectively.
Anna Galler*, Patrik Thunström, Josef Kaufmann, Matthias Pickem, Jan M. Tomczak and Karsten Held.
* Corresponding author. E-mail address: [email protected]
This project is licensed under the GNU General Public License v3 which can be found in LICENSE.
Ab initio dynamical vertex approximation - PRB
Ab initio dynamical vertex approximation - arXiv
Towards ab initio Calculations with the Dynamical Vertex Approximation - arXiv
If you use this package please cite the following paper: The AbinitioDGA Project v1.0: Non-local correlations beyond and susceptibilities within dynamical mean-field-theory - arXiv
This project has been supported by European Research Council under the European Union's Seventh Framework Program (FP/2007-2013) through Grant agreement No. 306447.
