VASP 自旋轨道耦合计算错误汇总
静态计算时, 报错 :
VERY BAD NEWS! Internal 内部 error insubroutine 子程序 IBZKPT:
Reciprocal 倒数的 lattice and k-lattice belong todifferent class oflattices. Often results are still useful... 48
INCAR 参数设置:
对策:根据所用集群,修改 INCAR 中NPAR 。将 NPAR=4 变成 NPAR=1 ,已解决!
错误: sub space matrix 类错误
报错: 静态和能带计算中出现警告: WARNING: Sub-Space-Matrix isnot hermitian 共轭 inDAV
结构优化出现错误:
WARNING: Sub-Space-Matrix isnot hermitian inDAV 4 -4.681828688433112E-002
对策: 通过将默认 AMIX=0.4 ,修改成 AMIX=0.2 (或 0.3),问题得以解决。
以下是类似的错误:
WARNING: Sub-Space-Matrix isnot hermitian inrmm -3.00000000000000
RMM: 22 -0.167633596124E+02 -0.57393E+00 -0.44312E-01 1326 0.221E+00BRMIX:
very serious problems the old and the new charge density differ old charge density: 28.00003 new 28.06093 0.111E+00
错误:
WARNING: Sub-Space-Matrix isnot hermitian inrmm -42.5000000000000
ERROR FEXCP: supplied Exchange-correletion table istoo small, maximal index : 4794
错误: 结构优化 Bi2Te3时, log 文件:
WARNING inEDDIAG: sub space matrix isnot hermitian 1 -0.199E+01
RMM: 200 0.179366581305E+01 -0.10588E-01 -0.14220E+00 718 0.261E-01
BRMIX: very serious problems the old and the new charge density differ old charge density: 56.00230 new 124.70394
66F= 0.17936658E+01 E0= 0.18295246E+01 dE=0.557217E-02
curvature: 0.00 expect dE= 0.000E+00 dE for cont linesearch 0.000E+00
ZBRENT: fatal error inbracketing
please rerun with smaller EDIFF, orcopy CONTCAR toPOSCAR and continue
但是,将 CONTCAR 拷贝成 POSCAR ,接着算静态没有报错,这样算出来的结果有问题吗?
对策 1:用这个 CONTCAR 拷贝成 POSCAR 重新做一次结构优化,看是否达到优化精度!
对策 2:用这个 CONTCAR 拷贝成 POSCAR ,并且修改 EDIFF (目前参数 EDIFF=1E-6 ),默认为 10-4
错误:
WARNING: Sub-Space-Matrix isnot hermitian inDAV 1 -7.626640664998020E-003
网上参考解决方案:
对策 1:减小 POTIM: IBRION=0 ,标准分子动力学模拟。通过 POTIM 控制步长。
POTIM
VASP自旋轨道耦合计算错误汇总.pdf