Huge energy oscillation in Molecular Anchoring process on a slab
Posted: Thu Jun 21, 2012 1:06 pm
Dear all,
this is a very general request of help regarding Molecular Anchoring on a slab.
My problem is that whichever the approach I use with VASP, at the end I always obtain huge oscillations in energy of the coupled system. In my case I am trying to attack an olephin on top of a Transition metal oxide slab.
Moreover I find this problem only when using Plane Waves.
MO approaches indeed always lead to stable structures I am not able to reproduce at PW level. Is there a physical reason (methodological) for such behaviour?
If not, could you suggest me a "safe" procedure to get this mixed systems? I attach my INCAR. Maybe you can suggest me which parameters I should have to better "tune". Consider that as initial step the Force convergence threshold is chosen high just to facilitate the initial convergence, but once I refine the convergence parameters, the final system (the anchored molecule) tends to "explode".
Any comment and suggestion is extremely welcome.
Best,
Giacomo
SYSTEM = Si_Vac_Td
# xc
ISTART = 1
GGA=PE
#surface dipole
#IDIPOL = 3
#LDIPOL = .TRUE.
#DIPOL = 0.2500000 0.5000156037602940 0.4034361849150227
#electronic relaxation
#IALGO = 48
LREAL =.False.
PREC= High
ENCUT=500.0
ENAUG=605.0
#ISMEAR = -5 for tetrahedron k-smearing w/ Blochl corrections
#ISMEAR = N for Methfessel-Paxton order N
#SIGMA = 0.05 eV of Temperature smearing
ISMEAR = -5
SIGMA = 0.005
#electronic RELAXATION
NELMIN = 4
NELM = 200
# ISIF = 7 Change cell vol without shape
# ISIF = 6 Change cell vol + shape, keeping ions fixed
# ISIF = 3 Change everything
# ISIF = 0 Relax ions only
ISIF = 2
# IBRION =1 for close to equilibrium RMM-DIIS
# IBRION =2 for OK guess CG
# IBRION =3 for bad guess Damped Molecular Dynamics
IBRION = 1
ADDGRID=.TRUE.
#accuracy of the calculation
# EDIFF is break condition for energy within electronic loop
# EDIFFG is break condition for forces between ionic moves
EDIFF = 1.E-3 ! 1.E-5
EDIFFG = -0.4 ! -0.04
#ionic steps
NSW =500
NGX=52
NGY=32
NGZ=199
this is a very general request of help regarding Molecular Anchoring on a slab.
My problem is that whichever the approach I use with VASP, at the end I always obtain huge oscillations in energy of the coupled system. In my case I am trying to attack an olephin on top of a Transition metal oxide slab.
Moreover I find this problem only when using Plane Waves.
MO approaches indeed always lead to stable structures I am not able to reproduce at PW level. Is there a physical reason (methodological) for such behaviour?
If not, could you suggest me a "safe" procedure to get this mixed systems? I attach my INCAR. Maybe you can suggest me which parameters I should have to better "tune". Consider that as initial step the Force convergence threshold is chosen high just to facilitate the initial convergence, but once I refine the convergence parameters, the final system (the anchored molecule) tends to "explode".
Any comment and suggestion is extremely welcome.
Best,
Giacomo
SYSTEM = Si_Vac_Td
# xc
ISTART = 1
GGA=PE
#surface dipole
#IDIPOL = 3
#LDIPOL = .TRUE.
#DIPOL = 0.2500000 0.5000156037602940 0.4034361849150227
#electronic relaxation
#IALGO = 48
LREAL =.False.
PREC= High
ENCUT=500.0
ENAUG=605.0
#ISMEAR = -5 for tetrahedron k-smearing w/ Blochl corrections
#ISMEAR = N for Methfessel-Paxton order N
#SIGMA = 0.05 eV of Temperature smearing
ISMEAR = -5
SIGMA = 0.005
#electronic RELAXATION
NELMIN = 4
NELM = 200
# ISIF = 7 Change cell vol without shape
# ISIF = 6 Change cell vol + shape, keeping ions fixed
# ISIF = 3 Change everything
# ISIF = 0 Relax ions only
ISIF = 2
# IBRION =1 for close to equilibrium RMM-DIIS
# IBRION =2 for OK guess CG
# IBRION =3 for bad guess Damped Molecular Dynamics
IBRION = 1
ADDGRID=.TRUE.
#accuracy of the calculation
# EDIFF is break condition for energy within electronic loop
# EDIFFG is break condition for forces between ionic moves
EDIFF = 1.E-3 ! 1.E-5
EDIFFG = -0.4 ! -0.04
#ionic steps
NSW =500
NGX=52
NGY=32
NGZ=199