计算化学公社

标题: 求助orca计算出错，该怎么改输入文件？ [打印本页]

作者
Author: 掌上明珠 时间: 2020-9-23 15:19
标题: 求助orca计算出错，该怎么改输入文件？
! PBE0 def2-SV(P) def2/J RIJCOSX grid4 gridx4 tightSCF opt def2-SVP/C noautostart miniprint nopop
%basis
   newGTO Ti "ZORA-SV(P)" end
end
%pal nprocs 4 end
%maxcore  2000
%tddft
nroots 10
TDA false
end
* xyz 0 1
B    -1.51697800 -0.39564352 2.07516869
B    -1.55233255 0.61666126 1.37342501
B    -1.85712012 0.68212330 -0.01120169
B    -2.20235029 -0.58850755 -0.75844419
B    -2.31189137 -1.80372847 -0.12637826
B    -2.06911269 -1.79719164 1.34718643
B    -4.99852588 -0.47454811 3.10271991
B    -4.07464710 0.48056937 3.32676931
B    -2.56650749 -0.15308555 2.89742290
B    -2.59182184 -1.21504050 2.35277378
B    -3.37581881 -1.79248190 1.97322342
B    -4.77805660 -1.56512038 2.18704222
B    -4.93560136 0.02830702 -1.12432051
B    -5.07071221 1.17490133 -0.81719765
B    -4.30537651 1.89901400 -0.28612399
B    -2.78574951 1.65046428 -0.33805970
B    -2.72457240 0.53478028 -1.19261379
B    -3.55667946 -0.40734451 -1.35837644
B    -5.93250165 -0.34094842 0.68246786
B    -5.67519774 -0.37377065 1.92325534
B    -5.26087442 0.89745878 2.61329919
B    -5.55838454 1.80133390 0.39056657
B    -5.86439824 0.55128219 -0.05500881
B    -3.26532525 2.64635385 0.39278361
B    -2.99156252 1.27801577 3.04008243
B    -1.91943066 1.69730266 1.04936441
B    -5.10714292 2.03080405 1.85888281
B    -4.27631054 1.90795327 2.67316710
B    -4.59572314 2.56097077 1.31442501
B    -1.76768629 1.06020404 2.28074371
B    -4.34924019 -2.46502265 1.19041034
B    -4.50345475 -1.31968089 -1.34933242
B    -5.82360105 -0.92682624 -0.44491546
B    -5.54942753 -1.53441480 0.78018903
B    -3.10704076 -2.40092666 0.42624180
B    -3.10965818 -1.80420009 -0.88196025
Ti -3.95920965 0.09444185 0.90795837
*
错误提示如下
--- error analysis -----

[3] on SC-202008041509
C:\Orca\orca_cis_mpi.exe ended prematurely and may have crashed. exit code 12

---- error analysis -----

ORCA finished by error termination in CIS
Calling Command: mpiexec -np 4  C:\Orca\orca_cis_mpi.exe 100.cisinp.tmp 100
[file orca_tools/qcmsg.cpp, line 458]:
  .... aborting the run

作者
Author: wzkchem5 时间: 2020-9-23 17:12
这些错误信息太少了，不足以得出结论，建议贴一下之前的大概100行
不过我先盲猜一个，可能是遇到了基态不稳定性。。。这个配合物的电子结构看起来太复杂了，波函数很容易不稳定

作者
Author: 喵星大佬 时间: 2020-9-23 17:19
你这个B团簇中间包个Ti，团簇本身都没优化过，感觉像乱点的。。。。

作者
Author: chrinide 时间: 2020-9-23 18:16
既然用了ZORA相对论基组，主keyword里面也应该加ZORA。
输入文件不伦不类，如果考虑相对论效应，那么全部元素都用ZORA相对论基组最好，虽然轻元素的基组问题不大。

作者
Author: wzkchem5 时间: 2020-9-23 18:32

chrinide 发表于 2020-9-23 18:16
既然用了ZORA相对论基组，主keyword里面也应该加ZORA。
输入文件不伦不类，如果考虑相对论效应，那么全部 ...

电子结构这么奇怪的体系，又是第一过渡周期金属，相对论效应估计比泛函误差小多了。。。我建议Ti改成def2-TZVP。当然无论怎么说这个计算的结果可信度都很堪忧，CAS又跑不动

作者
Author: 掌上明珠 时间: 2020-9-24 10:29

wzkchem5 发表于 2020-9-23 18:32
电子结构这么奇怪的体系，又是第一过渡周期金属，相对论效应估计比泛函误差小多了。。。我建议Ti改成def2 ...

好的我试试

作者
Author: 掌上明珠 时间: 2020-9-24 10:29

chrinide 发表于 2020-9-23 18:16
既然用了ZORA相对论基组，主keyword里面也应该加ZORA。
输入文件不伦不类，如果考虑相对论效应，那么全部 ...

好的，修改下试试

作者
Author: 掌上明珠 时间: 2020-9-24 15:00
本帖最后由掌上明珠于 2020-9-24 15:30 编辑

chrinide 发表于 2020-9-23 18:16
既然用了ZORA相对论基组，主keyword里面也应该加ZORA。
输入文件不伦不类，如果考虑相对论效应，那么全部 ...

输入文件改为如下，计算还是出错，该怎么改？
! PBE0 opt ZORA-def2-SV(P) tightSCF def2-SVP/C noautostart miniprint nopop
%basis
   newGTO Ti "def2-TZVP" end
end
%pal nprocs 4 end
%maxcore  3000
%tddft
nroots 10
TDA false
end
* xyz 0 1
坐标省略

输出文件出现如下
---- error analysis -----

ORCA finished by error termination in SCF
Calling Command: mpiexec -np 4  C:\Orca\orca_scf_mpi.exe 101.gbw b 101
[file orca_tools/qcmsg.cpp, line 458]:
  .... aborting the run

作者
Author: chrinide 时间: 2020-9-24 18:01
输入文件直接发上来吧

作者
Author: chrinide 时间: 2020-9-24 18:21

wzkchem5 发表于 2020-9-23 18:32
电子结构这么奇怪的体系，又是第一过渡周期金属，相对论效应估计比泛函误差小多了。。。我建议Ti改成def2 ...

PBE0用在B原子簇加第一行过渡金属，算是个比较合理的选择。如果关心中芯poperties，比如NMR，对第一行过渡金属，标量相对论影响有时候高达25%，甚至更高。不过对闭壳层几何结构而言，标量相对论主导，一般影响很小。开壳层的话，如果SO影响不可忽视，甚至体系的磁结构是非共线的（Fe、Co、Ni等元素），极大可能几何结构和最终的平衡结构会完全改变。直接用ZORA相对论比较安全，实际上并不会增加计算量。

作者
Author: chrinide 时间: 2020-9-24 18:28

掌上明珠发表于 2020-9-24 15:00
输入文件改为如下，计算还是出错，该怎么改？
! PBE0 opt ZORA-def2-SV(P) tightSCF def2-SVP/C noautos ...

直接贴个输入文件吧，文本很小的。

作者
Author: 掌上明珠 时间: 2020-9-24 19:59
! PBE0 ZORA-def2-SV(P) def2/J tightSCF opt def2-SVP/C noautostart miniprint nopop
%basis
   newGTO Ti "def2-TZVP" end
end
%pal nprocs 4 end
%maxcore  3000
%tddft
nroots 10
TDA false
end
* xyz 0 1
B    -1.51697800 -0.39564352 2.07516869
B    -1.55233255 0.61666126 1.37342501
B    -1.85712012 0.68212330 -0.01120169
B    -2.20235029 -0.58850755 -0.75844419
B    -2.31189137 -1.80372847 -0.12637826
B    -2.06911269 -1.79719164 1.34718643
B    -4.99852588 -0.47454811 3.10271991
B    -4.07464710 0.48056937 3.32676931
B    -2.56650749 -0.15308555 2.89742290
B    -2.59182184 -1.21504050 2.35277378
B    -3.37581881 -1.79248190 1.97322342
B    -4.77805660 -1.56512038 2.18704222
B    -4.93560136 0.02830702 -1.12432051
B    -5.07071221 1.17490133 -0.81719765
B    -4.30537651 1.89901400 -0.28612399
B    -2.78574951 1.65046428 -0.33805970
B    -2.72457240 0.53478028 -1.19261379
B    -3.55667946 -0.40734451 -1.35837644
B    -5.93250165 -0.34094842 0.68246786
B    -5.67519774 -0.37377065 1.92325534
B    -5.26087442 0.89745878 2.61329919
B    -5.55838454 1.80133390 0.39056657
B    -5.86439824 0.55128219 -0.05500881
B    -3.26532525 2.64635385 0.39278361
B    -2.99156252 1.27801577 3.04008243
B    -1.91943066 1.69730266 1.04936441
B    -5.10714292 2.03080405 1.85888281
B    -4.27631054 1.90795327 2.67316710
B    -4.59572314 2.56097077 1.31442501
B    -1.76768629 1.06020404 2.28074371
B    -4.34924019 -2.46502265 1.19041034
B    -4.50345475 -1.31968089 -1.34933242
B    -5.82360105 -0.92682624 -0.44491546
B    -5.54942753 -1.53441480 0.78018903
B    -3.10704076 -2.40092666 0.42624180
B    -3.10965818 -1.80420009 -0.88196025
Ti -3.95920965 0.09444185 0.90795837
*

作者
Author: 掌上明珠 时间: 2020-9-24 20:05

chrinide 发表于 2020-9-24 18:28
直接贴个输入文件吧，文本很小的。

好的

作者
Author: wzkchem5 时间: 2020-9-24 20:09

掌上明珠发表于 2020-9-24 15:00
输入文件改为如下，计算还是出错，该怎么改？
! PBE0 opt ZORA-def2-SV(P) tightSCF def2-SVP/C noautos ...

。。。叫你多贴几行输出文件你怎么不听呢，这几行错误信息根本没有任何信息量
另外如果你决定不用ZORA，那么B原子的基组就不要用ZORA-def2-SV(P)，应该用def2-SV(P)

作者
Author: wzkchem5 时间: 2020-9-24 20:12

chrinide 发表于 2020-9-24 18:21
PBE0用在B原子簇加第一行过渡金属，算是个比较合理的选择。如果关心中芯poperties，比如NMR，对第一行过 ...

确实，加相对论有益无害，我只是想说益处其实也不大。主要还要考虑到ORCA里写ZORA只能考虑标量相对论效应，SOC只能在CI/TDDFT水平下以变分形式考虑，或者以微扰形式考虑，而且不支持非共线自旋

作者
Author: 掌上明珠 时间: 2020-9-24 20:26

wzkchem5 发表于 2020-9-24 20:09
。。。叫你多贴几行输出文件你怎么不听呢，这几行错误信息根本没有任何信息量
另外如果你决定不用ZORA， ...

要用zora了

                        ORCA TD-DFT CALCULATION
------------------------------------------------------------------------------

Input orbitals are from       ... 100.gbw
CI-vector output             ... 100.cis
Tamm-Dancoff approximation    ... deactivated
RPA-Integral strategy       ... AO-integrals
Integral handling             ... AO integral Direct
Max. core memory used       ... 2000 MB
Reference state             ... RHF
Generation of triplets       ... off
Follow IRoot                ... off
Number of operators          ... 1
Orbital ranges used for CIS calculation:
Operator 0:  Orbitals  41...100  to 101...537
XAS localization array:
Operator 0:  Orbitals  -1... -1
-------------------
XC-INTEGRATION GRID
-------------------

General Integration Accuracy    IntAcc    ...  3.670
Radial Grid Type                RadialGrid  ... Gauss-Chebyshev
Angular Grid (max. acc.)       AngularGrid ... Lebedev-50
Angular grid pruning method    GridPruning ... 3 (G Style)
Weight generation scheme       WeightScheme... Becke
Basis function cutoff          BFCut    ... 1.0000e-11
Integration weight cutoff       WCut       ... 1.0000e-14
Grids for H and He will be reduced by one unit

# of grid points (after initial pruning)    ...  27510 ( 0.0 sec)
# of grid points (after weights+screening) ...  22457 ( 0.2 sec)
nearest neighbour list constructed          ... 0.0 sec
Grid point re-assignment to atoms done    ... 0.0 sec
Grid point division into batches done       ... 0.0 sec
Reduced shell lists constructed in 0.7 sec

Total number of grid points                ... 22457
Total number of batches                   ...    368
Average number of points per batch          ...    61
Average number of grid points per atom    ...    607
Average number of shells per batch          ... 177.12 (76.34%)
Average number of basis functions per batch  ... 432.92 (80.47%)
Average number of large shells per batch    ... 139.13 (78.55%)
Average number of large basis fcns per batch ... 324.46 (74.95%)
Maximum spatial batch extension             ...  23.73, 16.68, 21.96 au
Average spatial batch extension             ... 1.14,  0.91,  1.12 au

---------------------
COSX-INTEGRATION GRID
---------------------

General Integration Accuracy    IntAcc    ...  2.670
Radial Grid Type                RadialGrid  ... Gauss-Chebyshev
Angular Grid (max. acc.)       AngularGrid ... Lebedev-50
Angular grid pruning method    GridPruning ... 3 (G Style)
Weight generation scheme       WeightScheme... Becke
Basis function cutoff          BFCut    ... 1.0000e-11
Integration weight cutoff       WCut       ... 1.0000e-14
Grids for H and He will be reduced by one unit

# of grid points (after initial pruning)    ...  14322 ( 0.0 sec)
# of grid points (after weights+screening) ...  11694 ( 0.1 sec)
nearest neighbour list constructed          ... 0.0 sec
Grid point re-assignment to atoms done    ... 0.0 sec
Grid point division into batches done       ... 0.0 sec
Reduced shell lists constructed in 0.3 sec

Total number of grid points                ... 11694
Total number of batches                   ...    200
Average number of points per batch          ...    58
Average number of grid points per atom    ...    316
Average number of shells per batch          ... 180.71 (77.89%)
Average number of basis functions per batch  ... 443.88 (82.51%)
Average number of large shells per batch    ... 144.69 (80.07%)
Average number of large basis fcns per batch ... 339.82 (76.56%)
Maximum spatial batch extension             ...  19.87, 13.35, 17.88 au
Average spatial batch extension             ... 1.27,  1.10,  1.26 au

---------------
TD-DFT XC SETUP
---------------

DFT calculation                      ... on
Name of the grid file                ... 100.grid_cis.tmp
Exchange functional (TD-DFT)          ... PBE
PBE kappa parameter  XKappa (TD-DFT) ...  0.804000
PBE mue parameter XMuePBE (TD-DFT) ...  0.219520
Correlation functional (TD-DFT)       ... PBE
PBE beta parameter CBetaPBE (TD-DFT) ...  0.066725
LDA part of GGA corr.  LDAOpt (TD-DFT) ... PW91-LDA
Hybrid DFT                            ... on
Exchange mixing (TD-DFT)          ...  0.250
GGA exch. scaling (TD-DFT)          ...  0.750
GGA corr. scaling (TD-DFT)          ...  1.000

Building densities          ... done
Calculating rho(r) on the grid ... done
Building xc-kernel on the grid ... done
                  *** TD-DFT CALCULATION INITIALIZED ***

-------------------
RPA-DIAGONALIZATION
-------------------

Dimension of the (A+B)(A-B) eigenvalue problem ...  26220
Number of roots to be determined             ...    10
Maximum size of the expansion space          ... 300
Convergence tolerance for the residual       ... 2.500e-07
Convergence tolerance for the energies       ... 2.500e-07
Orthogonality tolerance                      ... 1.000e-14
Constructing the preconditioner             ... o.k.
Building the initial guess                   ... o.k.
Number of trial vectors determined          ... 300

                     ****Iteration 0****

Memory handling for direct AO based RPA:
Memory per vector needed    ... 11 MB
Memory needed                ... 345 MB
Memory available             ...  2000 MB
Number of vectors per batch ... 173
Number of batches          ...    1
Time for densities:    2.333
Time for XC-Integration: 96.139
[file orca_tools/qcmem.cpp, line 931, Process 3]:  OUT OF MEMORY ERROR!

job aborted:
[ranks] message

[0-2] terminated

[3] process exited without calling finalize

---- error analysis -----

[3] on SC-202008041509
C:\Orca\orca_cis_mpi.exe ended prematurely and may have crashed. exit code 12

---- error analysis -----

ORCA finished by error termination in CIS
Calling Command: mpiexec -np 4  C:\Orca\orca_cis_mpi.exe 100.cisinp.tmp 100
[file orca_tools/qcmsg.cpp, line 458]:
  .... aborting the run

出现out of memory error 是内存有问题吗

作者
Author: 喵星大佬 时间: 2020-9-24 21:49
OUT OF MEMORY ERROR! 当然是让你多给点内存咯

欢迎光临计算化学公社 (http://bbs.keinsci.com/)