My Community

I am in the geometric optimization work for bilayer graphene.
PBE-D was used for this work. Many papers using VASP showed successful result for it.
In my case, however, a fluctuating energy profile than a monotoneous decrease was observed and it does not converge well.
I use the same condition of the paper as written in below.
The rectangular cell was considered in it.

INCAR

PREC = Normal
ENCUT = 600
IBRION = 2
NSW = 100
ISIF = 4
LVDW = .true.
ALGO = Normal (blocked Davidson)
NELM = 60
NELMIN = 2
EDIFF = 1.0e-05
EDIFFG = -0.01
VOSKOWN = 1
NBLOCK = 1
ISPIN = 1
INIWAV = 1
ISTART = 0
ICHARG = 2
LWAVE = .FALSE.
LCHARG = .FALSE.
ADDGRID = .FALSE.
ISMEAR = 1
SIGMA = 0.1
LREAL = .FALSE.
RWIGS = 0.77


POSCAR
(C)8 (P1) ~ 2L_PC (VASP)
1.0
4.24700000 0.00000000 0.00000000
0.00000000 2.45200000 0.00000000
0.00000000 0.00000000 20.00000000
8
Direct
0.83335000 0.49996000 0.35000000
0.33339000 0.00000000 0.35000000
0.50004000 0.50004000 0.35000000
0.00000000 0.00000000 0.35000000
0.81993000 0.50298000 0.49014000
0.65328000 0.00294000 0.49014000
0.31997000 0.00302000 0.49014000
0.15332000 0.50298000 0.49014000


KPOINTS
Automatic mesh
0
Monkhorst Pack
12 18 1
0. 0. 0.



I don't know what's the problem with it.
The works using ISIF = 2 or POTIM = 0.1 showed still the same problem.
In addition, it showed a similar result for monolayer graphene.
May the incorrect compiling of VASP program exhibit such a behavior?

Thanks for your helps in advance.

Young

The total energy obtained using your input files decreases monotoneously:
...
110 F= -.73749305E+02
111 F= -.73749316E+02
112 F= -.73749436E+02
113 F= -.73749522E+02
114 F= -.73749525E+02
115 F= -.73749562E+02
116 F= -.73749573E+02
117 F= -.73749576E+02
118 F= -.73749632E+02
119 F= -.73749621E+02
120 F= -.73749639E+02
121 F= -.73749638E+02
122 F= -.73749668E+02
123 F= -.73749654E+02
124 F= -.73749659E+02
125 F= -.73749661E+02
126 F= -.73749661E+02
127 F= -.73749679E+02
128 F= -.73749675E+02
129 F= -.73749685E+02
130 F= -.73749688E+02
131 F= -.73749709E+02