ABINIT Discussion Forums

Dear Forum,

i want to calculate elastic constant for a semiconductor compound what is the good way for this reason. I tried through elastic tutorial (Tutorespon) for calculating elastic constant of Aluminium metal, but i didn't get a good result, so any one have another way ?




Kind Regards

Hi!

Well, the response function method (described in the tutorial) is the best way to calculate elastic constants as far as I know. You can try finite difference method, which is also described in elastic tutorial (part 4). This method is much simpler to understand, but you have to do a lot of things by hand.

Concerning the bad results, they can be because of lot of things, like bad pseudopotential, unrelaxed structure, unconverged ecut, ngkpt,... (did you check all of this?) You would have to be more specific if you want to analyze the source of these bad results.

There are no many differences between the two cases: semicond or metal. Just change occopt and erase tsmear is you want to adopt the Al input for your material.

Best regards,

Igor

Dear ilukacevic,

Thank you for your answer, but how i can relax the structure?

Kind regards

Dear Forume,

I am trying to calculate the elastic constant for a semiconductor compound, but i got this error
( The eigenvector number1 has zero norm. )
Plz one can show me why i got this and here is my input file: