I'm attempting to do some HSE06 calculations with VASP.
I'm checking k-point convergence at the moment and I find that VASP crashes quite often. The general problem seems to be:
LAPACK ZPOTRF fails (and then a number: 5, 11, 21, 24)
What is causing this? And how can it be solved?
The used input is:
INCAR
Code: Select all
general:
SYSTEM = fcc_bulk
ISTART = 0 ! start from 0: scratch, 1: previous WAVECAR
ICHARG = 1 ! 1: read old CHGCAR, 2: superposition from atom positions
ISMEAR = 0 ! 0: Gaussian; 1-N: MP; -5: Tetra+Blochl
SIGMA = 0.1
EDIFF = 1.0E-8
PREC = High ! to make ROPT=2.0E-4 for LREAL=Auto
ENCUT = 500 ! ENCUT for O
LWAVE = .FALSE.
LCHARGE = .TRUE.
LVTOT = .FALSE.
ISPIN = 2 ! include spin
!MAGMOM = ! put spins of Ce in same direction
NBANDS = 24 ! fix number of bands manually
VOSKOWN = 1 ! important for GGA (PW91) for interpolation of XC
HSE Calculation:
ISTART = 1
LHFCALC = .TRUE. ; HFSCREEN = 0.2
NBANDS = 24
ALGO = all ; TIME = 0.25
LMAXFOCK = 4 ! default 4, but we have f electrons=> +1
ENCUTFOCK = 0 !omit flag for High Quality Calcs
!NKRED = 2 !omit Flag for High Quality Calcs
dynamic:
IBRION = -1 ! -1: Fix atoms; 0: MD; 2: ConjGrad relax
NSW = 0 ! Number electronic steps
parallel:
LPLANE = .TRUE.
!NPAR = 1 ; NSIM = 8
KPOINTS for 3x3x3
Code: Select all
K-points
0
Gamma ! We have a cubic lattice
3 3 3
0 0 0
and POSCAR:
Code: Select all
fcc_bulk
4.85000000000000
0.0000000000000000 0.4863029426057665 0.4863029426057665
0.4863029426057665 0.0000000000000000 0.4863029426057665
0.4863029426057665 0.4863029426057665 0.0000000000000000
1
Selective dynamics
Direct
0.0000000000000000 0.0000000000000000 0.0000000000000000 F F F
0.00000000E+00 0.00000000E+00 0.00000000E+00
Thx Danny
