24小时热门版块排行榜

返回列表

当前只显示满足指定条件的回帖，点击这里查看本话题的所有回帖

junfeiyan

新虫 (小有名气)

应助: 2 (幼儿园)
金币: 52.2
散金: 30
红花: 1
帖子: 149
在线: 229.1小时
虫号: 2419400
注册: 2013-04-15
专业: 固体力学

[求助] 求助大神， VASP 使用杂化泛函结构优化一直报错已有5人参与

小弟最近使用杂化泛函进行结构优化，原因是我算的是含f电子稀土化合物， vasp 给的赝势用GGA+U 收敛极其困难，用杂化泛函好些，已经有发表在nature 子篇的论文，做过类似的体系，用的杂化泛函优化，求助各位大神，看看是小弟的输入文件有问题，还是天河超算中心的VASP 编译问题，感谢大神
小弟的INCAR

Name of system
SYSTEM =    a

Start parameter
ISTART =    0
INIWAV =    1
ICHARG =    2

Write flags
NWRITE =    2
LWAVE    =    T
LCHARG =    T

Parallelization
NPAR    =    72
LPLANE =    T
LSCALU =    F
LSCALAPACK = F

Electronic Relaxation 1
ENCUT    =    500.00  eV
EDIFF    =    1.0E-05  eV
PREC    =    High
NELM    =    200
NELMIN =    2

Electronic Relaxation 2
ALGO    =    Damped
TIME    =    0.4
LREAL    =    Auto

Ionic Relaxation
NSW    =    150
IBRION =    2
ISIF    =    3
EDIFFG =    -0.05
POTIM    =    0.3
ISYM    =    3
IWAVPR =    1
LCORR    =    T

DOS related values
ISMEAR =    0
SIGMA    =    0.025
NEDOS    =    3000
# NBANDS =    3958

Magnetic parameter
LORBIT =    11
Exchange-correlation functional
GGA    =    PS
ISPIN = 2
  MAGMOM = 4*8 4*4 12*0
HSE06 parameter
LHFCALC =    T
HFSCREEN  =    0.2
PRECFOCK  =    Fast
# NKRED    =    2

还有一个问体就是我想用HSE 考虑稀土元素的f 电子，那么我选择赝势的时候是否可以用PBE 里面不含f 电子的赝势，不知道杂化泛函会不会自己考虑f 电子
前提是我用GGA+U 算不含f电子的赝势，很快收敛，用含f电子的赝势极难收敛
报错一
ROT:  1 0.159495398721E+04 0.17831E+04BRMIX: very serious problems
the old and the new charge density differ
old charge density: 160.00000 new***********
0.603E+06
ERROR FEXCF: supplied exchange-correlation table
is too small, maximal index :    912695

这个报错有人说应该减小POTIM。有成功的经验吗
报错2
ROT:  1  -0.155058592914E+03 -0.30390E+02 0.284E+01
ROT:  2  -0.141575035880E+03 0.13484E+02 0.160E+01
ROT:  3  -0.139046123851E+03 0.25289E+01 0.114E+01
ROT:  4  -0.139495088906E+03 -0.44897E+00 0.453E+00
ROT:  5  -0.139199449050E+03 0.29564E+00 0.133E+00
gam= 0.382 g(H,U,f)=  0.305E+03 0.218E+02 0.320E+00 ort(H,U,f) =-0.188E+03-0.426E+02-0.465E+00
DMP: 7 -0.112154452901E+03 -0.50536E+02 -0.95574E+02  2160 0.327E+03-0.231E+03
      N    E                   dE          rms(c)
ROT:  1  -0.159501125680E+03 -0.20006E+02 0.795E+01
ROT:  2  -0.157770040931E+03 0.17311E+01 0.495E+01
ROT:  3  -0.155549182836E+03 0.22209E+01 0.173E+01
ROT:  4  NaN                   NaN          NaN
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image             PC             Routine          Line       Source
vasp             00000000005A769A  Unknown             Unknown  Unknown
vasp             000000000070A100  Unknown             Unknown  Unknown

U31~GV(QLRG}V{D75RS6[]8.png

回复此楼