Steven
May 11, 2024, 7:28pm
#1
Hi there,
When I did the nonadiabatic coupling with an extra keyword $external charges,
"
NCharge is 392
get_w_dc time: CPU 298.52 s wall 15.52 s*
… ESP will be evaluated on a read-in grid points from file ESPGrid.*
… The data will be saved in the file plot.esp*"
On top of that, the error message indicates that
" Fortran runtime error: Cannot open file ‘ESPGrid’: No such file or directory "
May I know what’s going wrong with this?
Thanks in advance.
can you please provide a minimal input file that exhibits this problem?
Steven
May 11, 2024, 8:09pm
#3
Hi there,
Thank you for attention.
Despite omitting the details of coordinates and charges (the position unit of external charge is A),
$molecule … H -4.658 0.087 3.438 H 2.511 -1.022 -4.204 C 3.797 -0.064 -2.778 … $end
$rem jobtype sp basis 6-31G **method pbe0 cis_n_roots 4 cis_triplets true rpa true CIS_DER_NUMSTATE 3 CALC_NAC true set_quadratic false set_iter 100 $end
$derivative_coupling comment 0 1 2 $end
$external_charges … -2.150 -4.349 0.012 0.594 -0.195 -3.584 0.690 0.062 -3.059 -4.814 -0.723 -0.497 … $end
Please post a complete input file that exhibits the problem, and I will look into it. If it’s a problem with $external_charges then you might only need one charge, and small molecules are easier for debugging and testing.
Steven
May 12, 2024, 8:25pm
#5
Thank you for the feedback.
$molecule 0 1 C -0.959750 0.241930 -0.011920 C 0.446990 0.242050 -0.011870 C 1.151170 1.459990 -0.012690 C 0.446830 2.676860 -0.013680 C -0.958870 2.676710 -0.013970 C -1.663570 1.459890 -0.012720 H -1.506840 -0.705280 -0.012170 H 0.994350 -0.704990 -0.011840 H 2.244630 1.457710 -0.011690 H 0.987530 3.625980 -0.013980 H -1.499650 3.625710 -0.015380 H -2.757010 1.458500 -0.013820 $end
$rem jobtype sp basis 6-31G **method pbe0 cis_n_roots 3 cis_singlets true cis_triplets false rpa true CIS_DER_NUMSTATE 3 CALC_NAC true set_quadratic false set_iter 100 $end
$derivative_coupling comment 0 1 2 $end
$external_charges 7.715147 7.223384 1.330103 -0.094319 7.779379 6.982973 -1.301776 -0.094733 6.658450 8.829292 -2.824952 -0.099668 5.460383 10.909124 -1.717081 -0.093927 5.397606 11.150537 0.914911 -0.101064 6.532443 9.311701 2.437614 -0.095811 8.579319 5.793390 2.509443 0.094976 8.693213 5.366879 -2.159485 0.095589 6.715350 8.646858 -4.860829 0.097792 4.582151 12.330652 -2.896289 0.096258 4.471149 12.759091 1.773262 0.097696 6.492589 9.502828 4.473038 0.097209 -1.559137 2.831074 6.516853 -0.096564 1.082341 2.928887 6.501338 -0.103902 2.317919 5.261413 6.358985 -0.090988 0.911358 7.496355 6.229898 -0.098731 -1.729629 7.398637 6.247529 -0.099416 -2.965415 5.065883 6.392169 -0.096897 -2.514904 1.025668 6.613701 0.098222 2.170067 1.199182 6.586470 0.099086 4.361337 5.334546 6.338576 0.095612 1.864309 9.301610 6.113170 0.097996 -2.814823 9.128417 6.145465 0.098029 -5.008738 4.987649 6.396005 0.097553 -8.667550 7.121376 1.197047 -0.089526 -8.605473 7.284724 -1.440803 -0.094618 -7.416268 9.335436 -2.609939 -0.099384 -6.272644 11.213503 -1.142661 -0.095371 -6.333682 11.049437 1.495321 -0.090760 -7.538439 9.007436 2.664608 -0.108992 -9.585900 5.534063 2.102396 0.094393 -9.475692 5.823815 -2.577076 0.095267 -7.377925 9.466980 -4.650068 0.097437 -5.341046 12.792841 -2.048463 0.095469 -5.448969 12.501597 2.631633 0.095322 -7.593430 8.884623 4.704946 0.100762 -2.048879 2.780732 -6.540607 -0.098389 0.590917 2.645087 -6.554742 -0.096001 2.028867 4.861302 -6.464772 -0.096021 0.826415 7.213368 -6.357152 -0.095485 -1.812871 7.348937 -6.340504 -0.097613 -3.251104 5.132515 -6.434291 -0.094529 -3.161115 1.065541 -6.599018 0.096727 1.520852 0.824960 -6.623868 0.096230 4.070886 4.753965 -6.467663 0.096493 1.935987 8.929013 -6.283661 0.095831 -2.739914 9.169235 -6.252631 0.096362 -5.293255 5.235033 -6.415299 0.096394 $end
Thanks. This looks like a bug, reproducible with the latest version of Q-Chem. I will post a ticket.
1 Like
Steven
May 14, 2024, 7:36pm
#7
Hi there,
Hard to say. In this case there is someone at Q-Chem looking at it.
Hi Steven, the Q-Chem Office is already working on this ticket. We will notify you as soon as we have an update. Thank you.
Steven
June 25, 2024, 10:27am
#10
Hi there,
Sorry for taking your time.
A fix for this was committed on June 12, ticket #3381 . Is available to developers now (in the trunk) and will ship with Q-Chem v. 6.2.1.
Steven
July 2, 2024, 8:02am
#12
Hi, sorry for taking your time again.
I requested one developer of Q-Chem to test whether this bug has been fixed (same input file as before).
Unfortunately, the task was not successfully completed yet, and the error message is
qcprog.exe: malloc.c:4036: _int_malloc: Assertion `(unsigned long) (size) >= (unsigned long) (nb)’ failed.
Could you please check again?
I also get a crash with the current Q-Chem trunk and your original input file. I will reopen the ticket.
Steven
August 2, 2024, 1:55pm
#14
Hi there,
The ticket is open but I don’t know the specifics. It is ticket #3381 if you have access to Q-Chem Trac site.
There has been a new fix checked in for this, and your sample seems to work. Please check it out and see if it works for you.
Steven
August 9, 2024, 9:48pm
#17
Thank you very much for the effort.
No problem, but this one’s on Q-Chem staff, I’m just the reporter.
Steven
August 19, 2024, 7:48pm
#19
Hi, I am truly sorry for taking your time again.
I requested one developer of Q-Chem to confirm this bug has been fixed (same input file as before).
However, when we examined the results carefully, the output seemed not to be correct.
Because the number of total atoms is only 12, but the CIS deriativce coupling with ETF also included the “dummy atoms” (positions of the 48 point charges), together 60 atoms (12+48), see below.
So there are supposed to be some minor errors left.
Could you please double check again?
Thank you.
why don’t you just have that developer communicate directly with us on Trac? It’s ticket #3381 . This would eliminate me having to act as go-between.