how can I optimize a monoclinic structure?

Message

magnones · #1 Post by **magnones** » Fri Mar 16, 2007 9:57 pm

Hi Let me explain my problem I want to simulate a solid which has a monoclinic simetry whit the POSCAR as follow

monoclinic oxide
1
5.14540000 0.00000000 0.00000000
0.00000000 5.20750000 0.00000000
-0.85213078 0.00000000 5.24188970
4 8
Directas
0.27580000 0.04040000 0.20890000
-0.27580000 0.54040000 0.29110000
-0.27580000 -0.04040000 -0.20890000
0.27580000 0.45960000 0.70890000
0.06900000 0.34200000 0.34500000
-0.06900000 0.84200000 0.15500000
-0.06900000 -0.34200000 -0.34500000
0.06900000 0.15800000 0.84500000
0.45100000 0.75800000 0.47900000
-0.45100000 0.25800000 0.02100000
-0.45100000 -0.75800000 -0.47900000
0.45100000 -0.25800000 0.97900000

this is an oxide whit 12 base atoms well i was working whit Cubic Structures before but never whit this kind of structures. Reading the http://cms.mpi.univie.ac.at/vasp-workshop/
i find how to optimize strcutures but it is not clear how i can optimize strcutures whit more than one degree of freedom. I read my literature and the very helpful tutorial from the Vasp workshop and i do the following

I first optimize the Kpoitns number after that i optimize the Encut energy, whit this parameters i found the optimal volumen using a Congugate Gradient energy (CG) minimization varying the first degree of freedrom ("a" in this case, the remainigs are b/a and c/a) i did this optimizing the strucuture in each step whit a relaxation allowing relax the ion positions and cell shape keeping the volumen cosntant in each step using a INCAR file as follow

SYSTEM= monoclinic
ISTART = 0
ICHARG = 2
ENCUT = 500
ISMEAR = 0
NSW = 10
IBRION = 2
ISIF = 4
SIGMA = 0.2

my question is, how can i optimize the other parameters b/a and c/a? how can i optimize the ions postions? my steps before are well? do I need some other kind of relaxation?

i dont know if someone would be so kindness to give some advices or the steps to make this kind of optimization, i will be very thankfull

#2 Post by **admin** » Mon Mar 19, 2007 2:14 pm

please have a look into the online handbook (ISIF)
to optimize the cell shape and the ionic positions, set ISIF=4
to optimize the cell volume in addition, set ISIF=3

please also increase PREC to accurate amd ENCUT to at least 1.3*max(enmax) of POTCAR