[DFTB-Plus-User] comparison of hamsqr1.dat, oversqr1.dat and band.dat

[Bálint Aradi]

> 
> I am trying to compare the H and S from the hamsqr1.dat and oversqr.dat 
> for the Gamma point and noticing that the eigenvalues given in band.dat 
> do not correspond to the eigenvalues I get from solving hamsqr1.dat and 
> oversqr.dat in MATLAB. Am I missing something? Or should these both match?

Yes, they should match. Please keep in mind, though, that with WriteHS = 
Yes, DFTB+ would only builds the Hamiltonian and quits then immediately, 
so that no SCC-cycles are done. If you want to get the Hamiltonian for a 
fully converged SCC-calculation, then

* Run the SCC-calculation normally

* Start a follow-up calculation with ReadInitialCharges = Yes and 
WriteHS = Yes.

Diagonalizing the hamsqr1.dat and oversqr.dat files you obtained this 
way, should reproduce the eigenvalues you got with the SCC-calculation.

I've just tried it for H2O with the following Python-snippet, and it 
seems to work. ;-)

import numpy as np
import scipy.linalg as linalg

over = np.loadtxt('oversqr.dat', skiprows=5)
ham = np.loadtxt('hamsqr1.dat', skiprows=5)

eigvals, eigvecs = linalg.eigh(ham, over)
print(eigvals)

Best regards,

Bálint


-- 
Dr. Bálint Aradi
Bremen Center for Computational Materials Science, University of Bremen
http://www.bccms.uni-bremen.de/cms/people/b-aradi/