Eigenvector problems using Zheev

Post here if you have a question about LAPACK or ScaLAPACK algorithm or data format

Eigenvector problems using Zheev

Postby ginetom » Tue Apr 09, 2013 3:51 pm

Hello Dear all,

My doubt is quite simple (I think)...I am diagonalizing an Hermitian matrix (Hamiltonian) from which I need to get the eigenvectors to calculate the spin expectation as a function of some parameter that is changed in the Hamiltonian. The problem is that when I have a degeneracy in the eigenvalues the eigenvectors do not agree with the expected behavior (analitical), in fact it mixes the spin components in the eigenfunction wich is expected not to happen as the elements in the Hamiltonian are not mixing them. I am using the Zheev subroutine.

If anyone has faced this issue before or knows how to solve it, I will really appreciate!!!
thank you so much in advance!
Ginetom Diniz
ginetom
 
Posts: 1
Joined: Tue Apr 09, 2013 3:41 pm

Re: Eigenvector problems using Zheev

Postby error5772 » Tue May 07, 2013 2:12 am

hmm....

If you multiply out the resulting eigenvector(s) with your
Hamilton-matrix and the result is okay the problem is solved!

Then you should look back on the analytical expectation...
Analytical expectations based on pertubation theory typically fail
when degeneracy occurs...?

Can you calulate a simple analytical solvable special-case with your code
and check this first?

Can you reduce the size of you Hamilton-matrix to something controllable and
check this with mathlab or else.

If all this does not help - find the structure of the failing problem and send
a mini example matrix here.

That's the best I have!

M.
error5772
 
Posts: 19
Joined: Thu Sep 01, 2011 5:02 am


Return to Algorithm / Data

Who is online

Users browsing this forum: No registered users and 1 guest