|
首先,感谢各位大佬百忙中点进来关注这个问题。
我在用VASP得到MnSi的弛豫后的结构后,用VASP计算了能带没有发现问题。想用QE来后续计算,但是自洽始终无法收敛。输入文件和相关信息贴在下方了:
scf.in
&control
calculation = 'scf'
prefix = 'MnSi',
pseudo_dir = 'pp/',
outdir = './tmp'
verbosity = 'medium',
/
&system
ibrav = 0,
nat = 8,
ntyp = 2,
ecutwfc = 70,
ecutrho = 800,
occupations = 'smearing',
smearing = 'gaussian',
degauss = 0.01
nspin = 2
starting_magnetization(1) = 1.0
/
&electrons
conv_thr = 1.0d-8
/
ATOMIC_SPECIES
Mn 54.938 Mn.pbe-spn-rrkjus_psl.1.0.0.UPF
Si 28.0855 Si.pbe-n-rrkjus_psl.1.0.0.UPF
ATOMIC_POSITIONS (crystal)
Mn 0.3632081590620636 0.1367918259379387 0.8632081890620661
Mn 0.1367918259379387 0.8632081890620661 0.3632081590620636
Mn 0.8632081890620661 0.3632081590620636 0.1367918259379387
Mn 0.6367918109379339 0.6367918109379339 0.6367918109379339
Si 0.6547480253453379 0.8452519746546621 0.1547480403453356
Si 0.8452519746546621 0.1547480403453356 0.6547480253453379
Si 0.1547480403453356 0.6547480253453379 0.8452519746546621
Si 0.3452519746546620 0.3452519746546620 0.3452519746546620
K_POINTS {automatic}
11 11 11 0 0 0
CELL_PARAMETERS {angstrom}
4.5105628384197418 0.0000000000000000 -0.0000000000000000
-0.0000000000000000 4.5105628384197418 0.0000000000000000
0.0000000000000000 -0.0000000000000000 4.5105628384197418
HUBBARD (ortho-atomic)
U Mn-3d 0.5
EOF
收敛信息如下:
estimated scf accuracy < 10.63587605 Ry
estimated scf accuracy < 2725661.81405824 Ry
estimated scf accuracy < 2523025.30321596 Ry
estimated scf accuracy < 2523456.63290837 Ry
estimated scf accuracy < 2523333.64125067 Ry
estimated scf accuracy < 2523403.89969335 Ry
estimated scf accuracy < 2523515.17646111 Ry
estimated scf accuracy < 2524910.26830625 Ry
estimated scf accuracy < 2523829.20939331 Ry
estimated scf accuracy < 2525194.86959638 Ry
estimated scf accuracy < 2532227.61909879 Ry
estimated scf accuracy < 2522045.87774801 Ry
estimated scf accuracy < 2526694.21906833 Ry
estimated scf accuracy < 2520163.08909798 Ry
estimated scf accuracy < 2528056.83484624 Ry
estimated scf accuracy < 2515712.09155797 Ry
这也太离谱了。我增大了能量截断到100多Ry也还是一样。还减小了mixing_beta到0.1还是一样。由于自洽无法收敛,也没办法用QE来优化结构。
|
|