Problem with SAPT calculations for platinum complexes

SAPT
1

Dear Psi4 community,
I employed following input to perform SAPT0 for platinum complex with iodine derivative but the error message “Failed to converge” is returned in console. Futhermore, I have error in output file

Traceback (most recent call last):
File “/mnt/d/QC/psi4conda/bin/psi4”, line 287, in
exec(content)
File “”, line 138, in
File “/mnt/d/QC/psi4conda/lib//python3.7/site-packages/psi4/driver/driver.py”, line 556, in energy
wfn = procedures[‘energy’][lowername](lowername, molecule=molecule, **kwargs)
File “/mnt/d/QC/psi4conda/lib//python3.7/site-packages/psi4/driver/procrouting/proc.py”, line 3404, in run_sapt
e_sapt = core.sapt(dimer_wfn, monomerA_wfn, monomerB_wfn)

RuntimeError:
Fatal Error: PSIO Error
Error occurred in file: /scratch/psilocaluser/conda-builds/psi4-multiout_1557940846948/work/psi4/src/psi4/libpsio/error.cc on line: 128
The most recent 5 function calls were:

psi::PSIO::rw(unsigned long, char*, psi::psio_address, unsigned long, int)

Printing out the relevant lines from the Psithon --> Python processed input file:
return basstrings
qcdb.libmintsbasisset.basishorde[‘ANONYMOUS3641F703’] = basisspec_psi4_yo__anonymous3641f703
core.set_global_option(“DF_BASIS_SAPT”, “anonymous3641f703”)
core.set_global_option(“GUESS”, “sad”)
core.set_global_option(“FREEZE_CORE”, “true”)
–> energy(‘sapt0’)

Psi4 stopped on: Sunday, 20 September 2020 08:29PM
Psi4 wall time for execution: 3:53:37.48

*** Psi4 encountered an error. Buy a developer more coffee!
*** Resources and help at github.com/psi4/psi4.

Ptbtpyt.dat (18.4 KB)

Replacing platinum with palladium gives the correct result. expanding disk storage and RAM didn’t help. It should be noted that this happens when using def2-tzvp
Please let me know what is the solution. Many thanks for your kindly attention.

Best regards
Eugene

2

The output file that you provided seems to be incomplete. Please provide a complete output file. If it does not include the error message that you posted (about psi::PSIO::rw), it is incomplete.

I suspect that your scratch space may not be set up correctly, but having the full output file would help.

3

Sorry, I confused the file. I am sending the correct

Ptbtpyt1.dat (149.0 KB)

Best regards
Eugene

4

I’ll ping @hokru, but my first thought is that you should double-check that you have set up your scratch directory properly.

Also, can you post the contents of your standard error file? Based on the error message, there’s more information there that will be valuable.

5

Dear Jmisiewicz,
Thank you for your help.
I will check the scratch directory.
The terminal displayed the following lines:
Failed to converge:
PSIO_ERROR: unit = 194, errval = 12
PSIO_ERROR: 12 (error writing to file)
Traceback (most recent call last):
File “/mnt/d/QC/psi4conda/bin/psi4”, line 287, in
exec(content)
File “”, line 138, in
File “/mnt/d/QC/psi4conda/lib//python3.7/site-packages/psi4/driver/driver.py”, line 556, in energy
wfn = procedures[‘energy’][lowername](lowername, molecule=molecule, **kwargs)
File “/mnt/d/QC/psi4conda/lib//python3.7/site-packages/psi4/driver/procrouting/proc.py”, line 3404, in run_sapt
e_sapt = core.sapt(dimer_wfn, monomerA_wfn, monomerB_wfn)

RuntimeError:
Fatal Error: PSIO Error
Error occurred in file: /scratch/psilocaluser/conda-builds/psi4-multiout_1557940846948/work/psi4/src/psi4/libpsio/error.cc on line: 128
The most recent 5 function calls were:

psi::PSIO::rw(unsigned long, char*, psi::psio_address, unsigned long, int)

Printing out the relevant lines from the Psithon --> Python processed input file:
return basstrings
qcdb.libmintsbasisset.basishorde[‘ANONYMOUS57DE73E0’] = basisspec_psi4_yo__anonymous57de73e0
core.set_global_option(“DF_BASIS_SAPT”, “anonymous57de73e0”)
core.set_global_option(“GUESS”, “sad”)
core.set_global_option(“FREEZE_CORE”, “true”)
–> energy(‘sapt0’)

Please specify where the standard error file is located?

Best regards
Eugene

6

The standard error was printed to the terminal; that was the information I was looking for, thank you.

This error happened when attempting to write the integrals needed for your computation (developers, this is file PSIF_SAPT_AB_DF_INTS). I advise you check the scratch directory.

7

Dear Jmisiewicz,
I checked the PSI_SCRATCH directory, it is written in the bashrc file
export PSI_SCRATCH=/mnt/d/Ubuntu/Calc_Psi4/scratch
All necessary files are created in the scratch folder (Including PSIF_SAPT_AB_DF_INTS)
image
Writing in the input file
psi4_io.set_default_path(’/mnt/d/Ubuntu/Calc_Psi4/scratch/user’)
does not write files to the scratch / user folder
Interestingly, replacing the platinum atom with palladium results in a complete SAPT calculation without errors.
Maybe it’s important I’m using the psi4 version for WSL on Windows 10

Many thanks for your kindly attention.
Best regards
Eugene

8

How much disk space do you have available for this scratch file?

9

Dear Jmisiewicz,
For SAPT calculations, 114 GB is free on the hard disk. When you start SAPT calculations, the working folder (Scratch folder) is filled with 25 GB. I checked on a disk with 200 GB, but the error still pops up.
What is very strange, when you start calculating platinum, many lines appear in the terminal
“Failed to converge:”
, this is not observed for calculating palladium complexes
Best regards
Eugene

10

The “failed to converge” indicates a problem with the ECP integrals. While I’m not familiar with ECPs, that seems unrelated to PSIO problems.

I’ll see if I can bring this to the attention of SAPT specialists, who would have a better idea of the memory requirements for a computation like this.

11

Dear Jmisiewicz,
Thank you for your attention to this issue. You are correct about the basis. Replacing the basic function def2-TZVP with aug-cc-pvtz-pp for the atoms I and Pt did not issue this error. When using def2-QZVPP as a basis, an error is also observed: Failed to converge
Please tell me if it is correct to use the basis aug-cc-pvtz-pp for Pt and I and for light atoms (H, C, N) to use the basis jun-cc-pvdz, while Psi4 for calculating SAPT0 chooses def2-QZVPP for DF_BASIS_SAPT and DF_BASIS_SCF jkfit. Is it correct to calculate weak interactions with SAPT0 so?
Best regards
Eugene

12

Per conference call discussion, pinging @sherrill.