计算化学公社

标题: 求助：在NVT系综的AIMD计算中出现SCF不收敛的问题 [打印本页]

作者
Author: anrushan 时间: 2021-3-7 15:15
标题: 求助：在NVT系综的AIMD计算中出现SCF不收敛的问题
在已经完成了在npt的弛豫的构象内正式计算AIMD时出现了scf不收敛的问题，应该不是构象的问题，几个参数也都用的默认数值。尝试增添DIIS项后依然是无法收敛。请各位老师指教。
这次计算分子为一多氮物。
SCF WAVEFUNCTION OPTIMIZATION

  Step    Update method    Time Convergence       Total energy Change
  ------------------------------------------------------------------------------
   1 NoMix/Dav.  0.40E+00 0.5    3.65652626    -185.3385542749 -1.85E+02
   2 Broy./Dav.  0.40E+00 0.5    5.04097556    -178.4647054514  6.87E+00
   3 Broy./Dav.  0.40E+00 0.5    0.65877679    -196.9598045613 -1.85E+01
。。。。。。
125 Broy./Dav.  0.40E+00 0.5    0.02160860    -196.5311668555 -4.15E-02
126 Broy./Dav.  0.40E+00 0.5    0.01407403    -196.5717251976 -4.06E-02
127 Broy./Dav.  0.40E+00 0.5    0.02355139    -196.6118522437 -4.01E-02
128 Broy./Dav.  0.40E+00 0.5    0.00619748    -196.6601445349 -4.83E-02

  Leaving inner SCF loop after reaching 128 steps.

  Electronic density on regular grids:    -107.9999999999       0.0000000001
  Core density on regular grids:             107.9999999950    -0.0000000050
  Total charge density on r-space grids:    -0.0000000049
  Total charge density g-space grids:       -0.0000000049

  Overlap energy of the core charge distribution:             1.18503093169915
  Self energy of the core charge distribution:             -503.47211874021690
  Core Hamiltonian energy:                                  156.76882654704423
  Hartree energy:                                           203.45762976821723
  Exchange-correlation energy:                               -54.59951304167091

  Total energy:                                           -196.66014453492721

*** WARNING in qs_scf.F:576 :: SCF run NOT converged ***
ENERGY| Total FORCE_EVAL ( QS ) energy [a.u.]:          -196.809196806938473

MD_INI| MD initialization
MD_INI| Potential energy [hartree]                         -0.196809196807E+03
MD_INI| Kinetic energy [hartree]                            0.218510258687E+00
MD_INI| Temperature [K]                                           2000.000000
MD_INI| Pressure [bar]                                     8.912970522431E+05
MD_INI| Cell volume [bohr^3]                               1.858398232120E+03
MD_INI| Cell volume [ang^3]                               2.753862022160E+02
MD_INI| Cell lengths [bohr]    1.22945582E+01  1.22945582E+01  1.22945582E+01
MD_INI| Cell lengths [ang]    6.50600000E+00  6.50600000E+00  6.50600000E+00
MD_INI| Cell angles [deg]       9.00000000E+01  9.00000000E+01  9.00000000E+01

Number of electrons:                                                       108
Number of occupied orbitals:                                              54
Number of molecular orbitals:                                              54

Number of orbital functions:                                              296
Number of independent orbital functions:                                  296

Extrapolation method: ASPC

以下是部分输入文件

&GLOBAL
  PROJECT 01
  PRINT_LEVEL LOW
  RUN_TYPE MD
&END GLOBAL

&FORCE_EVAL
  METHOD Quickstep
  &SUBSYS
&CELL
   A    6.50600000    0.00000000    0.00000000
   B    0.00000000    6.50600000    0.00000000
   C    0.00000000    0.00000000    6.50600000
   PERIODIC XYZ #Direction of applied PBC (geometry aspect)
&END CELL
&COORD
  C       5.7512427511    14.6247805356    -3.2379190427
  C       3.5879314640    13.6057951729    -4.1422006332
  N       6.7446849278    14.0749270321    -2.4919660132
  N       7.7454339226    14.9329172329    -2.7696084834
  H       8.6506119723    14.8338896903    -2.3040760447
&END COORD
# &VELOCITY #You can set initial atomic velocities in this section
# &END VELOCITY
&KIND C
   ELEMENT C
   BASIS_SET DZVP-MOLOPT-SR-GTH
   POTENTIAL GTH-PBE
&END KIND
&KIND N
   ELEMENT N
   BASIS_SET DZVP-MOLOPT-SR-GTH
   POTENTIAL GTH-PBE
&END KIND
&KIND H
   ELEMENT H
   BASIS_SET DZVP-MOLOPT-SR-GTH
   POTENTIAL GTH-PBE
&END KIND
  &END SUBSYS

  &DFT
BASIS_SET_FILE_NAME  BASIS_MOLOPT
POTENTIAL_FILE_NAME  POTENTIAL
# WFN_RESTART_FILE_NAME to_be_specified
#Net charge
MULTIPLICITY  1 #Spin multiplicity
&QS
   EPS_DEFAULT 1E-10 #This is default. Set all EPS_xxx to values such that the energy will be correct up to this value
   EXTRAPOLATION ASPC #Extrapolation for wavefunction during e.g. MD. ASPC is default, PS also be used
   EXTRAPOLATION_ORDER 3 #Order for PS or ASPC extrapolation. 3 is default
&END QS
&POISSON
   PERIODIC XYZ #Direction(s) of PBC for calculating electrostatics
   PSOLVER PERIODIC #The way to solve Poisson equation
&END POISSON
&XC
   &XC_FUNCTIONAL PBE
   &END XC_FUNCTIONAL
&END XC
&MGRID
   CUTOFF 300
   REL_CUTOFF 40
&END MGRID
&SCF
   MAX_SCF 128
   EPS_SCF 1E-5 #Convergence threshold of density matrix during SCF process
#    SCF_GUESS RESTART #Use wavefunction from WFN_RESTART_FILE_NAME file as initial guess
   &DIAGONALIZATION
      ALGORITHM  DAVIDSON
      #Algorithm for diagonalization. DAVIDSON is faster for large systems
   &END DIAGONALIZATION
   &MIXING #How to mix old and new density matrices
      METHOD BROYDEN_MIXING
      ALPHA 0.4 #Default. Mixing 40% of new density matrix with the old one
      NBROYDEN 8 #Default is 4. Number of previous steps stored for the actual mixing scheme
   &END MIXING
   &PRINT
      &RESTART #Use "&RESTART OFF" can prevent generating wfn file
      BACKUP_COPIES 0 #Maximum number of backup copies of wfn file
      &END RESTART
   &END PRINT
&END SCF
  &END DFT
   STRESS_TENSOR ANALYTICAL
  &END FORCE_EVAL

&MOTION
&MD
   ENSEMBLE NVT             #constant temperature and pressure using an isotropic cell
   STEPS 14000
   TIMESTEP 1
   TEMPERATURE 2000
  &THERMOSTAT
&NOSE
   TIMECON 1000
&END NOSE
&END THERMOSTAT
&END MD
  &PRINT
&TRAJECTORY
   &EACH
      MD    1 #Output frequency of geometry
   &END EACH
   FORMAT xyz
&END TRAJECTORY
&VELOCITIES
   &EACH
      MD    1 #Output frequency of velocity
   &END EACH
&END VELOCITIES
&RESTART
   BACKUP_COPIES 0 #Maximum number of backup copies of MD restart file
   &EACH
      MD 10 #Frequency of updating restart file
   &END EACH
&END RESTART
  &END PRINT
&END MOTION

作者
Author: wzkchem5 时间: 2021-3-7 18:30
你确定初始构象合理？光是弛豫过不能保证合理，还要看过弛豫后的结构，确定是你想要的结构才能保证合理

作者
Author: anrushan 时间: 2021-3-7 18:58

wzkchem5 发表于 2021-3-7 18:30
你确定初始构象合理？光是弛豫过不能保证合理，还要看过弛豫后的结构，确定是你想要的结构才能保证合理

确定合理，上面的in文件坐标部分不全，在弛豫过后检查过输出的XYZ轨迹，并没有发生结构变化，原子间的距离也合理，我曾经用初始结构跑了一下上面in文件里的nvt收敛的可以，不明白为什么跑了弛豫之后会出现不收敛的状况，谢谢老师的提醒

作者
Author: anrushan 时间: 2021-3-7 20:21
升高了收敛的上限，观察输出文件发现结构已经散了，是方法有问题。

作者
Author: sobereva 时间: 2021-3-7 23:53
没你的体系细节信息没法说，起码给个当前体系结构图，并且显示出盒子
如果是单个分子，用ORCA甭用CP2K
使用ORCA做从头算动力学(AIMD)的简单例子
http://sobereva.com/576（http://bbs.keinsci.com/thread-20800-1-1.html）

作者
Author: anrushan 时间: 2021-3-8 20:50

sobereva 发表于 2021-3-7 23:53
没你的体系细节信息没法说，起码给个当前体系结构图，并且显示出盒子
如果是单个分子，用ORCA甭用CP2K
使 ...

卢老师，如果只考虑分子在动力学中的轨迹问题是不是不用关注SCF的收敛（分子在高温条件下发生的分解反应是否也会影响SCF方程的收敛呢）？

作者
Author: sobereva 时间: 2021-3-9 04:19

anrushan 发表于 2021-3-8 20:50
卢老师，如果只考虑分子在动力学中的轨迹问题是不是不用关注SCF的收敛（分子在高温条件下发生的分解反应 ...

需要关注
如果某一步没收敛，而且最后一轮SCF的时候收敛程度很差，那么这一步的受力就会算得很糟糕，轻则不准确，重则造成动力学不稳定

收敛成功几率和这一步结构与上一步结构差异有关，差异越小收敛成功几率越大。高温下如果发现中途难收敛，尝试降低步长

作者
Author: anrushan 时间: 2021-3-9 08:46

sobereva 发表于 2021-3-9 04:19
需要关注
如果某一步没收敛，而且最后一轮SCF的时候收敛程度很差，那么这一步的受力就会算得很糟糕，轻 ...

谢谢卢老师，注意休息

作者
Author: LLYYTT 时间: 2022-4-13 16:01
我也遇到了这个问题，请问应该怎么解决呢

欢迎光临计算化学公社 (http://bbs.keinsci.com/)