计算化学公社

标题: VASP优化Ni单质SCF不收敛 [打印本页]

作者
Author: longyp 时间: 2024-12-7 19:12
标题: VASP优化Ni单质SCF不收敛
本帖最后由 longyp 于 2024-12-7 19:25 编辑

使用VASP优化Ni单质，定义了磁矩和+U，但是SCF始终不收敛，请问应该怎么调节参数呢？ALGO试过Fast,Very_Fast和Normal都不行
以下为使用的INCAR文件内容：Global Parameters
ISTART =  1          (Read existing wavefunction, if there)
ISPIN  =  2          (Non-Spin polarised DFT)
NCORE = 15
# ICHARG =  11       (Non-self-consistent: GGA/LDA band structures)
LREAL  = .FALSE.    (Projection operators: automatic)
ENCUT  =  450       (Cut-off energy for plane wave basis set, in eV)
PREC =  Accurate (Precision level: Normal or Accurate, set Accurate when perform structure lattice relaxation calculation)
LWAVE  = .TRUE.       (Write WAVECAR or not)
ALGO = Normal
LCHARG = .TRUE.       (Write CHGCAR or not)
ADDGRID= .TRUE.       (Increase grid, helps GGA convergence)

Electronic Relaxation
ISMEAR =  1          (Gaussian smearing, metals:1)
SIGMA  =  0.1       (Smearing value in eV, metals:0.2)
NELM =  200          (Max electronic SCF steps)
NELMIN =  6          (Min electronic SCF steps)
EDIFF  =  1E-05       (SCF energy convergence, in eV)
GGA  =  PE          (PBEsol exchange-correlation)

Ionic Relaxation
NSW =  300       (Max ionic steps)
IBRION =  2          (Algorithm: 0-MD, 1-Quasi-New, 2-CG)
ISIF =  2          (Stress/relaxation: 2-Ions, 3-Shape/Ions/V, 4-Shape/Ions)
EDIFFG = -3E-02       (Ionic convergence, eV/AA)
ISYM =  0          (Symmetry: 0=none, 2=GGA, 3=hybrids)
IVDW = 12

MAGMOM = 108*0.8

LDAU =  .TRUE.
LDAUTYPE  =  2
LDAUL  = 2
LDAUU  = 5.5
LDAUJ  = 0
LMAXMIX  =4

作者
Author: Farnesol 时间: 2024-12-7 23:00
NELM可以删掉，EDIFF=1e-4一般也够用，ALGO不能用veryfast，更不利于收敛

作者
Author: longyp 时间: 2024-12-8 00:49

Farnesol 发表于 2024-12-7 23:00
NELM可以删掉，EDIFF=1e-4一般也够用，ALGO不能用veryfast，更不利于收敛

好的，我尝试一下，谢谢

作者
Author: 北大-陶豫 时间: 2024-12-8 19:19

SIGMA = 0.1 (Smearing value in eV, metals:0.2)

复制代码

看来 vaspkit 生成的 INCAR 你是一点不看

金属体系本来就需要用更大的展宽来促进收敛（收敛之后把展宽降回去是另一码事）

作者
Author: longyp 时间: 2024-12-9 09:57

北大-陶豫发表于 2024-12-8 19:19
看来 vaspkit 生成的 INCAR 你是一点不看

金属体系本来就需要用更大的展宽来促进收敛（收敛之后把 ...

好的，谢谢，我用0.2试下看看

欢迎光临计算化学公社 (http://bbs.keinsci.com/)