Radical Anion Error/Bug in 1.6

1

I recently upgraded to 1.6 and as a test case, ran an optimization that was successful under 1.3.2 (and seems to be under 1.4.1 – waiting for it to fully finish, but it has already progressed further than under 1.6)

The Input file (for a given radical anion):

memory 102400 mb

molecule mol {
-1 2
C 0.00000000 0.00000000 0.00000000
C 1.21064700 0.71690500 0.00000000
C 1.20587000 2.10633100 0.00000000
C 0.00000000 2.80649500 0.00000000
C -1.20587000 2.10633100 0.00000000
C -1.21064700 0.71690500 0.00000000
H -2.14648500 0.17133400 0.00000000
H -2.14658900 2.64534700 0.00000000
H 0.00000000 3.89046700 0.00000000
H 2.14658900 2.64534700 0.00000000
H 2.14648500 0.17133400 0.00000000
C 0.00000000 -1.42321300 0.00000000
C 0.00000000 -2.63387100 0.00000000
C 0.00000000 -4.05708400 0.00000000
C -1.21064700 -4.77398900 0.00000000
C -1.20587000 -6.16341500 0.00000000
C 0.00000000 -6.86357900 0.00000000
C 1.20587000 -6.16341500 0.00000000
C 1.21064700 -4.77398900 0.00000000
H 2.14648500 -4.22841800 0.00000000
H 2.14658900 -6.70243100 0.00000000
H 0.00000000 -7.94755100 0.00000000
H -2.14658900 -6.70243100 0.00000000
H -2.14648500 -4.22841800 0.00000000

}

set scf_type df
set basis aug-cc-pVTZ
set reference uks
set_num_threads(64)

mol.update_geometry()
mol.symmetrize(1e-3)

e, wfn = optimize('b3lyp-d3bj', return_wfn=True)
oeprop(wfn, "MULLIKEN_CHARGES")
mol.print_out()
print_variables()

Under version 1.6, I get the error (note there is line that says: “This is a bug. Please file a report.”):

  ==> Iterations <==

                           Total Energy        Delta E     RMS |[F,P]|

   @DF-UKS iter SAD:  -538.66710083028875   -5.38667e+02   0.00000e+00 

Traceback (most recent call last):
  File "/data/Apps/anaconda3/envs/psi4env/bin/psi4", line 338, in <module>
    exec(content)
  File "<string>", line 54, in <module>
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/driver.py", line 1211, in optimize
    G, wfn = gradient(lowername, return_wfn=True, molecule=moleculeclone, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/driver.py", line 619, in gradient
    wfn = procedures['gradient'][lowername](lowername, molecule=molecule, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 90, in select_scf_gradient
    return func(name, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 2550, in run_scf_gradient
    ref_wfn = run_scf(name, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 2455, in run_scf
    scf_wfn = scf_helper(name, post_scf=False, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 1670, in scf_helper
    e_scf = scf_wfn.compute_energy()
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/scf_proc/scf_iterator.py", line 86, in scf_compute_energy
    self.iterations()
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/scf_proc/scf_iterator.py", line 429, in scf_iterate
    self.form_C()

RuntimeError: 
Fatal Error: Count 49 alpha electrons, expected 48.
This is a bug. Please file a report.
Error occurred in file: /scratch/psilocaluser/conda-builds/psi4-multiout_1652994828216/work/psi4/src/psi4/libscf_solver/hf.cc on line: 482
The most recent 5 function calls were:




Printing out the relevant lines from the Psithon --> Python processed input file:
    core.set_global_option("BASIS", "aug-cc-pVTZ")
    core.set_global_option("REFERENCE", "uks")
    set_num_threads(64)
    mol.update_geometry()
    mol.symmetrize(1e-3)
--> e, wfn = optimize('b3lyp-d3bj', return_wfn=True)
    oeprop(wfn, "MULLIKEN_CHARGES")
    mol.print_out()
    print_variables()

!----------------------------------------------------------------------------------!
!                                                                                  !
! Fatal Error: Count 49 alpha electrons, expected 48.                              !
! This is a bug. Please file a report.                                             !
! Error occurred in file: /scratch/psilocaluser/conda-                             !
!     builds/psi4-multiout_1652994828216/work/psi4/src/psi4/libscf_solver/hf.cc on !
!     line: 482                                                                    !
! The most recent 5 function calls were:                                           !
!                                                                                  !
!----------------------------------------------------------------------------------!

    Psi4 stopped on: Friday, 03 June 2022 12:13PM
    Psi4 wall time for execution: 0:05:00.69

*** Psi4 encountered an error. Buy a developer more coffee!
*** Resources and help at github.com/psi4/psi4.
Traceback (most recent call last):
  File "/data/Apps/anaconda3/envs/psi4env/bin/psi4", line 338, in <module>
    exec(content)
  File "<string>", line 54, in <module>
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/driver.py", line 1211, in optimize
    G, wfn = gradient(lowername, return_wfn=True, molecule=moleculeclone, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/driver.py", line 619, in gradient
    wfn = procedures['gradient'][lowername](lowername, molecule=molecule, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 90, in select_scf_gradient
    return func(name, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 2550, in run_scf_gradient
    ref_wfn = run_scf(name, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 2455, in run_scf
    scf_wfn = scf_helper(name, post_scf=False, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 1670, in scf_helper
    e_scf = scf_wfn.compute_energy()
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/scf_proc/scf_iterator.py", line 86, in scf_compute_energy
    self.iterations()
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/scf_proc/scf_iterator.py", line 429, in scf_iterate
    self.form_C()

RuntimeError: 
Fatal Error: Count 49 alpha electrons, expected 48.
This is a bug. Please file a report.
Error occurred in file: /scratch/psilocaluser/conda-builds/psi4-multiout_1652994828216/work/psi4/src/psi4/libscf_solver/hf.cc on line: 482
The most recent 5 function calls were:



Printing out the relevant lines from the Psithon --> Python processed input file:
    core.set_global_option("BASIS", "aug-cc-pVTZ")
    core.set_global_option("REFERENCE", "uks")
    set_num_threads(64)
    mol.update_geometry()
    mol.symmetrize(1e-3)
--> e, wfn = optimize('b3lyp-d3bj', return_wfn=True)
    oeprop(wfn, "MULLIKEN_CHARGES")
    mol.print_out()
    print_variables()


!----------------------------------------------------------------------------------!
!                                                                                  !
! Fatal Error: Count 49 alpha electrons, expected 48.                              !
! This is a bug. Please file a report.                                             !
! Error occurred in file: /scratch/psilocaluser/conda-                             !
!     builds/psi4-multiout_1652994828216/work/psi4/src/psi4/libscf_solver/hf.cc on !
!     line: 482                                                                    !
! The most recent 5 function calls were:                                           !
!                                                                                  !
!----------------------------------------------------------------------------------!


ERROR conda.cli.main_run:execute(34): Subprocess for 'conda run ['/data/Apps/anaconda3/envs/psi4env/bin/psi4', '-s', '/archive/WebMO/webmo-54951/887', 'input.inp', 'output.out']' command failed.  (See above for error)
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image              PC                Routine            Line        Source             
libgdma.so         00007F035C2660DC  for__signal_handl     Unknown  Unknown
libpthread-2.17.s  00007F0371279630  Unknown               Unknown  Unknown
python3.8          000056259F6885D3  Unknown               Unknown  Unknown
core.cpython-38-x  00007F03686A45C0  Unknown               Unknown  Unknown
core.cpython-38-x  00007F036872EAEB  Unknown               Unknown  Unknown
core.cpython-38-x  00007F0368A47700  _ZNSt15_Sp_counte     Unknown  Unknown
core.cpython-38-x  00007F03687E9A67  _ZN3psi7Process11     Unknown  Unknown
libc-2.17.so       00007F0370ED5CE9  Unknown               Unknown  Unknown
libc-2.17.so       00007F0370ED5D37  Unknown               Unknown  Unknown
python3.8          000056259F732608  Unknown               Unknown  Unknown
python3.8          000056259F73263B  Unknown               Unknown  Unknown
python3.8          000056259F732680  Unknown               Unknown  Unknown
python3.8          000056259F735604  PyRun_SimpleFileE     Unknown  Unknown
python3.8          000056259F735A29  Py_RunMain            Unknown  Unknown
python3.8          000056259F735C29  Py_BytesMain          Unknown  Unknown
libc-2.17.so       00007F0370EBE555  __libc_start_main     Unknown  Unknown
python3.8          000056259F6D8AD7  Unknown               Unknown  Unknown
2

My initial attempts to reproduce this are not working.

How many iterations do you need to get this issue?
What happens if you get rid of the -d3bj correction? What if you downgrade from B3LYp to HF? What if you go from an optimization to an energy computation?

3

Here’s my replies:

How many iterations do you need to get this issue?
→ Appears after the first Iteration:

   ==> Iterations <==

                           Total Energy        Delta E     RMS |[F,P]|

   @DF-UKS iter SAD:  -538.66710083028840   -5.38667e+02   0.00000e+00 

Traceback (most recent call last):
  File "/data/Apps/anaconda3/envs/psi4env/bin/psi4", line 338, in <module>
    exec(content)
  File "<string>", line 54, in <module>
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/driver.py", line 1211, in optimize
    G, wfn = gradient(lowername, return_wfn=True, molecule=moleculeclone, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/driver.py", line 619, in gradient
    wfn = procedures['gradient'][lowername](lowername, molecule=molecule, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 90, in select_scf_gradient
    return func(name, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 2550, in run_scf_gradient
    ref_wfn = run_scf(name, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 2455, in run_scf
    scf_wfn = scf_helper(name, post_scf=False, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 1670, in scf_helper
    e_scf = scf_wfn.compute_energy()
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/scf_proc/scf_iterator.py", line 86, in scf_compute_energy
    self.iterations()
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/scf_proc/scf_iterator.py", line 429, in scf_iterate
    self.form_C()

RuntimeError: 
Fatal Error: Count 49 alpha electrons, expected 48.
This is a bug. Please file a report.
Error occurred in file: /scratch/psilocaluser/conda-builds/psi4-multiout_1652994828216/work/psi4/src/psi4/libscf_solver/hf.cc on line: 482
The most recent 5 function calls were:

Printing out the relevant lines from the Psithon --> Python processed input file:
    core.set_global_option("BASIS", "aug-cc-pVTZ")
    core.set_global_option("REFERENCE", "uks")
    set_num_threads(64)
    mol.update_geometry()
    mol.symmetrize(1e-3)
--> e, wfn = optimize('b3lyp-d3bj', return_wfn=True)
    oeprop(wfn, "MULLIKEN_CHARGES")
    mol.print_out()
    print_variables()

What happens if you get rid of the -d3bj correction?
→ Same issue. Error appears after first iteration.

What if you downgrade from B3LYp to HF?
→ It completes successfully using this input:

memory 102400 mb

molecule mol {
-1 2
C 0.00000000 0.00000000 0.00000000
C 1.21064700 0.71690500 0.00000000
C 1.20587000 2.10633100 0.00000000
C 0.00000000 2.80649500 0.00000000
C -1.20587000 2.10633100 0.00000000
C -1.21064700 0.71690500 0.00000000
H -2.14648500 0.17133400 0.00000000
H -2.14658900 2.64534700 0.00000000
H 0.00000000 3.89046700 0.00000000
H 2.14658900 2.64534700 0.00000000
H 2.14648500 0.17133400 0.00000000
C 0.00000000 -1.42321300 0.00000000
C 0.00000000 -2.63387100 0.00000000
C 0.00000000 -4.05708400 0.00000000
C -1.21064700 -4.77398900 0.00000000
C -1.20587000 -6.16341500 0.00000000
C 0.00000000 -6.86357900 0.00000000
C 1.20587000 -6.16341500 0.00000000
C 1.21064700 -4.77398900 0.00000000
H 2.14648500 -4.22841800 0.00000000
H 2.14658900 -6.70243100 0.00000000
H 0.00000000 -7.94755100 0.00000000
H -2.14658900 -6.70243100 0.00000000
H -2.14648500 -4.22841800 0.00000000

}

set scf_type df
set basis 6-31G(d)
set reference uhf


mol.update_geometry()
mol.symmetrize(1e-3)

e, wfn = optimize('scf', return_wfn=True)
oeprop(wfn, "MULLIKEN_CHARGES")
mol.print_out()
print_variables()

What if you go from an optimization to an energy computation?
→ Still fails after first Iteration:

 ==> Iterations <==

                           Total Energy        Delta E     RMS |[F,P]|

   @DF-UKS iter SAD:  -538.66710083175246   -5.38667e+02   0.00000e+00 

Traceback (most recent call last):
  File "/data/Apps/anaconda3/envs/psi4env/bin/psi4", line 338, in <module>
    exec(content)
  File "<string>", line 53, in <module>
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/driver.py", line 501, in energy
    wfn = procedures['energy'][lowername](lowername, molecule=molecule, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 2455, in run_scf
    scf_wfn = scf_helper(name, post_scf=False, **kwargs)
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/proc.py", line 1670, in scf_helper
    e_scf = scf_wfn.compute_energy()
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/scf_proc/scf_iterator.py", line 86, in scf_compute_energy
    self.iterations()
  File "/data/Apps/anaconda3/envs/psi4env/lib//python3.8/site-packages/psi4/driver/procrouting/scf_proc/scf_iterator.py", line 429, in scf_iterate
    self.form_C()

RuntimeError: 
Fatal Error: Count 49 alpha electrons, expected 48.
This is a bug. Please file a report.
Error occurred in file: /scratch/psilocaluser/conda-builds/psi4-multiout_1652994828216/work/psi4/src/psi4/libscf_solver/hf.cc on line: 482
The most recent 5 function calls were:

Printing out the relevant lines from the Psithon --> Python processed input file:
    core.set_global_option("SCF_TYPE", "df")
    core.set_global_option("BASIS", "aug-cc-pVTZ")
    core.set_global_option("REFERENCE", "uks")
    core.set_global_option("S_TOLERANCE", 1e-9)
    set_num_threads(64)
--> energy('b3lyp-d3bj')
    mol.print_out()
    print_variables()

Once again, under 1.3.2 and 1.4.1 – not an issue.

The key error seems to be → 

RuntimeError: 
**Fatal Error: Count 49 alpha electrons, expected 48.**
**This is a bug. Please file a report.**

One thing I noticed is that during pre-iteration in 1.3.2, you get → 

 ==> Pre-Iterations <==

   -------------------------------------------------------
    Irrep   Nso     Nmo     Nalpha   Nbeta   Ndocc  Nsocc
   -------------------------------------------------------
     Ag       160     160       0       0       0       0
     B1g      130     130       0       0       0       0
     B2g       63      63       0       0       0       0
     B3g       84      84       0       0       0       0
     Au        63      63       0       0       0       0
     B1u       84      84       0       0       0       0
     B2u      160     160       0       0       0       0
     B3u      130     130       0       0       0       0
   -------------------------------------------------------
    Total     874     874      48      47      47       1
   -------------------------------------------------------

But in 1.6, you get:

  ==> Pre-Iterations <==

  SCF Guess: Superposition of Atomic Densities via on-the-fly atomic UHF (no occupation information).

   -------------------------
    Irrep   Nso     Nmo    
   -------------------------
     Ag       160     160 
     B1g      130     130 
     B2g       63      63 
     B3g       84      84 
     Au        63      63 
     B1u       84      84 
     B2u      160     160 
     B3u      130     130 
   -------------------------
    Total     874     874
   -------------------------

Ultimately, it seems the number of electrons (95 total) is not being counted correctly in 1.6 for this molecule — ???

4

I can reproduce this now, which is a start. I’ve added an issue to bring this to the attention of other developers.

Normally, I’d take this one myself, but I have too many obligations to take this one as well.

5

Thanks. I’ll look forward to it. Until this is fixed there isn’t a way to calculate the polarizability tensor for this molecule under 1.6 – correct? . I recall you had noted that the polarizability tensor issue (for radical anion) had been fixed in 1.5 (?).

6

Correct: if you can’t get an energy, you certainly can’t get a polarizability tenor.

Yes, you should be able to get this polarizability through 1.5. The important code is this commit.

7

Yes, can confirm optimization and polarizability worked in 1.5 for this radical anion.