Has any one know how to set up the poscar/incar for the calculation of formation energy of isotropic biaxial stress? The sizes of crystal in x and y directions are fixed, only in z direction that can be relax.
Is VASP 4.6 work on this problem?
sutatch
LDA of formation energy
Moderators: Global Moderator, Moderator
-
sutatch
LDA of formation energy
Last edited by sutatch on Tue Jul 07, 2009 3:32 pm, edited 1 time in total.
-
forsdan
- Sr. Member
- Posts: 339
- Joined: Mon Apr 24, 2006 9:07 am
- License Nr.: 173
- Location: Gothenburg, Sweden
LDA of formation energy
By default you can't put an arbitrary constraint on the length of two of the lattice vectors and relax the third lattice vector, all within 1 INCAR file. The simple solution is to use a small script (e.g. bash, python or C) that loops over a set of supercell sizes in the z-direction in order to find the optimal geometry. ISIF = 2 may be an appropriate choice to reoptimize the atomic positions at each supercell size (if necessary).
An informal alternative is that you can modify the source code "constr_cell_relax.F" according to the example given there and then recompile vasp. This will give you the possibility to perform the relaxation you're after by using the ISIF = 3 option. In this case, please notice that modifications made to the code is most likely not supported so you'll have to know what you're doing. So if you don't know, stick to the script approach...
Best regards,
/Dan
PS. The Selective Dynamics tag is not applicable to this problem since the option locks direct coordinates and not Cartesian coordinates, so it can't be used to perform this type of lattice vector optimization in one direction. DS
<span class='smallblacktext'>[ Edited Wed Jul 08 2009, 05:04PM ]</span>
An informal alternative is that you can modify the source code "constr_cell_relax.F" according to the example given there and then recompile vasp. This will give you the possibility to perform the relaxation you're after by using the ISIF = 3 option. In this case, please notice that modifications made to the code is most likely not supported so you'll have to know what you're doing. So if you don't know, stick to the script approach...
Best regards,
/Dan
PS. The Selective Dynamics tag is not applicable to this problem since the option locks direct coordinates and not Cartesian coordinates, so it can't be used to perform this type of lattice vector optimization in one direction. DS
<span class='smallblacktext'>[ Edited Wed Jul 08 2009, 05:04PM ]</span>
Last edited by forsdan on Tue Jul 07, 2009 8:23 pm, edited 1 time in total.