静态计算时出现警告
静态计算时的INCAR如下:
SYSTEM =
ENCUT =
ISTART = 0
ICHARG = 2
ISMEAR = -5
ISPIN = 2
NELM = 2000
EDIFF = 1E-4
IALGO = 48
LREAL = Auto
GGA = 91
VOSKOWN = 1
但是在计算中出现警告如下:
RMM: 22 -0.310967594429E+03 0.21293E+00 -0.24248E+00 10244 0.439E+00 0.422E+01
WARNING: DENTET: can't reach specified precision
Number of Electrons is NELECT = 476.996768664236
RMM: 23 -0.310376643978E+03 0.59095E+00 -0.83636E-01 10268 0.266E+00 0.375E+01
RMM: 43 -0.309736141363E+03 0.53076E-02 -0.50929E-02 10165 0.574E-01 0.356E+00
RMM: 44 -0.309811626871E+03 -0.75486E-01 -0.48571E-02 10210 0.597E-01 0.369E+00
WARNING: DENTET: can't reach specified precision
Number of Electrons is NELECT = 477.041023713897
RMM: 45 -0.309698348515E+03 0.11328E+00 -0.13902E-02 10260 0.310E-01BRMIX: very serious problems
the old and the new charge density differ
old charge density: 476.99930 new 477.04032
0.319E+00
WARNING: DENTET: can't reach specified precision
Number of Electrons is NELECT = 476.985853390457
RMM: 46 -0.310016610388E+03 -0.31826E+00 -0.42439E-02 10215 0.447E-01BRMIX: very serious problems
the old and the new charge density differ
old charge density: 477.04032 new 476.98515
0.296E+00
RMM: 47 -0.309997772396E+03 0.18838E-01 -0.27267E-02 10259 0.413E-01BRMIX: very serious problems
the old and the new charge density differ
old charge density: 476.98515 new 476.99930
0.198E+00
RMM: 48 -0.310064999937E+03 -0.67228E-01 -0.44514E-02 10200 0.397E-01 0.171E+00
这些警告对电荷密度的结果有影响吗?应该如何消除? 返回小木虫查看更多
the reason for this error message (DENTET) is that the Fermi level
cannot be determined accurately enough by the tetrahedron method (i.e, the integrated DOS stays different from the number of valence electrons)
please try (one of) the following:
--choose a different BZ integration scheme (ISMEAR)
--if you want or have to keep the tetrahedron integration,
-- increase the number of k-points (mind that Gamma has to be included in the k-mesh)
-- increase NEDOS
以上是vasp论坛对此类问题的回复
大意是费米能级不能精确确定。解决办法有如下三条:
1. 改变ISMEAR(如使用ISMEAR=0)
2. 增加K点
3. 增加NEDOS(DOS的采点个数)
这三种方法都试了,只有第一种方法把ISMEAR改为0才可以消除这些警告?但是在静态计算的时候可以设置ISMEAR=0,SIGMA=0.2吗?期待你的回复,谢谢。
你是不是用的VASP.4.6,编译器版本是ifort10 ?
我用的是VASP.4.6,编译器版本是ifort9
静态计算时候我个人觉得可以用ISMEAR=0
但是不同系统对于ISMEAR的设置是不同的,=0的设置对大多数系统都是有效的。SIGMA取值需要测试,用多个SIGMA计算单点能,取一个合理的entropy。具体做法我不太清楚了,不妨查查论坛里类似的帖子。
比如你是小系统绝缘体或半导体,用-5和较多的KPOINTS
如果是大系统绝缘体,用=0并取小的SIGMA值,如0.05。
比如你是金属体系,用>0,调试合适的SIGMA值
,
我的体系时金属间化合物,算的是有关晶界的性质,在静态计算的时候ISMEAR是取0还是应该取2呢?
金属化合物如果是绝缘体,有带隙,就用0
如果是导体,考虑用2