Optimization of an anionic MOF in VASP

[shelaka_15]

Dear All,
I am facing an issue while trying to optimize a 3D MOF structure using VASP 5.4.1. The original structure is Orthorhombic in nature, with alpha=beta=gamma=90 deg. However after my optimization these angles deviate a lot from the 90 deg angle. The MOF is anionic in nature, meaning it carries an overall -1 charge on whole lattice. To compensate for it, I put the reported counterion NH4+ in the unit cell. So I get the result attached below.
alpha=97.19958, beta=85.04839, gamma=75.40539

My input is:
Electronic structure
  PREC    = ACCURATE
  ADDGRID = .TRUE.
  LREAL   = Auto
  ENCUT   = 520
  ISYM    = 2
  ISPIN   = 1
Electronic minimization
  IALGO   = 48
  EDIFF   = 1e-6
  NELM    = 100
  MAXMIX  = 80
  ISMEAR  = 0
  SIGMA   = 0.05
Density functional
  GGA     = PE
  IVDW    = 12
Job parameters
  ISTART  = 1
Ionic relaxation
  ISIF    = 3
  EDIFFG  = -0.05
  NSW     = 1000
  IBRION  = 2
  POTIM   = 0.1
Parallelization
  NSIM    = 4
  NPAR    = 2
  LPLANE  = .TRUE.
  LSCALU  = .FALSE
Other parameters

Please suggest, if I am missing on something.

[christopher_sheldon1]

Dear Shelaka,

Thank you for your question. Could you attach the POSCAR and OUTCAR files? If you want to keep the structure orthorhombic, I would recommend fixing these angles so that they cannot change. You can do this by using a different value for ISIF. I.e. you can keep the cell shape constant while changing the volume and positions using ISIF=8.

Does this answer your question?

Best wishes,

Chris