Bader integration using weighted integration

This code takes the grid-based output from VASP as from CHGCAR or AECCAR files and performs Bader integration over the basins of attraction. The call

weight_int CHGCAR [grid1] [grid2] ...

uses the first grid to define atom-centered basins of attraction, and integrates each grid-defined quantity over those same basins of attraction. At a minimum, one grid file can be used. The grids must all be compatible: same grid dimensions defined for each. It does not check that the atomic coordinates are the same, and it only reads in the first grid found in the file; thus, it requires some hacking of the files in order to, e.g., integrate the magnetization. There are a few options available:

• -a assigns volumes to true basins (maxima) rather than explicitly to atoms.
• -s divides (scales) the integral by the total volume of the cell; required to compute a Bader charge.
• -o base output the weights on a grid, to be used for visualization; "base" is the name of the weight file: base0001
• -n N output only the weight for atom/basin N
• -V use Voronoi volumes instead of Bader (basins of attraction)
• -v/-t verbose or testing (extra verbose) modes

References

Original algorithm described in the first reference; the latter two references include examples. If you use the code, please cite the first reference as well as

• M. Yu and D. R. Trinkle, "Accurate and efficient algorithm for Bader charge integration." J. Chem. Phys. 134, 064111 (2011). doi
• M. Yu, D. R. Trinkle, and R. M. Martin, "Energy density in density functional theory: Application to crystalline defects and surfaces." Phys. Rev. B 83, 115113 (2011). doi
• M. Yu and D. R. Trinkle, "Au/TiO2(110) interfacial reconstruction stability from ab initio." J. Phys. Chem. C 115, 17799-17805 (2011). doi

Contributors

Min Yu and Dallas R. Trinkle, algorithm development and implementation

Acknowledgments

The research was supported by NSF under grant number DMR-1006077 and through the Materials Computation Center at UIUC, NSF DMR-0325939, and with computational resources from NSF/TeraGrid provided by NCSA and TACC. We also thank G. Henkelman at U. Texas for helpful discussions, and R. E. L. Deville at UIUC for helpful discussions.