ABINIT Discussion Forums

Dear all,

How is it possible to make a convergence study with respect to both a and c for an hexagonal structure?
I have a method but it doesn't look very practical:
To vary "a". And for every value of "a", you vary "c "and the adequate value of "c" is the one for which you get the minimum total energy.

Or is right to to vary only "a" and keep "c" fixed or use the known value of c/a = (8/3)^(1/2)?
I'd appreciate your help!
Omar

Dear Omar,

your method is good. But you could also use the automated algorithm for optimization (BFGS) by including the combination of optcell and ionmov variables. This method could be even more useful if you have atoms which need to relax their positions.

Igor L.