|
[求助]
QE添加范德瓦尔兹力计算原子受力结果为NaN, 求教应该怎么办已有1人参与
目的是弛豫获得CsPbI3的立方相晶体结构
材料的范德瓦尔兹力比较强,对晶格影响比较大,在qe中选的是DFT-d3的方法
计算结果报错,CRASH文件里这样显示:
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
task # 0
from checkallsym : error # 1
some of the original symmetry operations not satisfied
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
尝试关闭对称性( nosym = .true.)之后仍然有相同的报错,检查out文件发现:
atom 1 type 1 force = NaN NaN NaN
atom 2 type 2 force = NaN NaN NaN
atom 3 type 3 force = NaN NaN NaN
atom 4 type 3 force = NaN NaN NaN
atom 5 type 3 force = NaN NaN NaN
atom 1 type 1 force = NaN NaN NaN
atom 2 type 2 force = NaN NaN NaN
atom 3 type 3 force = NaN NaN NaN
atom 4 type 3 force = NaN NaN NaN
atom 5 type 3 force = NaN NaN NaN
Total force = NaN Total SCF correction = 0.000002
猜测是由于受力没有成功计算才导致CRASH中的报错,但不知为何会有这样的错误(去掉范氏力参数则没有报错,但不能得到真实的晶格常数)
求各位大佬给出宝贵指导!
以下是in文件:
&CONTROL
calculation='vc-relax', prefix='CsPbI3',
outdir='./', verbosity='high'
pseudo_dir = '/home/zhaizh/2-CsPbI3/qe/CON_PBE_rel/pseudo',
tprnfor=.true., tstress=.true., forc_conv_thr=1.0d-4
/
&SYSTEM
ibrav= 0,
nat= 5, ntyp= 3,
ecutwfc= 60, ecutrho = 300,
! noncolin = .true.
! lspinorb=.true.
vdw_corr = 'DFT-D3'
dftd3_version = 4 !D3 with BJ damping
nosym= .true.
/
&ELECTRONS
electron_maxstep = 500
conv_thr = 1.0d-12
/
&IONS
ion_dynamics='bfgs'
/
&CELL
press_conv_thr=0.1
/
&IONS
ion_dynamics='bfgs'
/
&CELL
press_conv_thr=0.1
/
ATOMIC_SPECIES
Cs 132.90545 Cs_ONCV_PBE_FR-1.0.upf
Pb 207.2 Pb_ONCV_PBE_FR-1.0.upf
I 126.90447 I_ONCV_PBE_FR-1.1.upf
ATOMIC_POSITIONS {crystal}
Cs 0.5000000000 0.500000 0.500000
Pb 0.0000000000 0.000000 0.000000
I 0.5000000000 0.000000 0.000000
I 0.0000000000 0.500000 0.000000
I 0.0000000000 0.000000 0.500000
CELL_PARAMETERS {angstrom}
6.273000000 0.000000000 0.000000000
0.000000000 6.273000000 0.000000000
0.000000000 0.000000000 6.273000000
K_POINTS {automatic}
11 11 11 0 0 0 |
|