Molcas Forum

Anugraha

Member

Registered: 2025-12-17

Posts: 22

Aborted casscf optimization after frequency calculation

Dear Molcas Developers,

I am performing an excited-state CASSCF optimization using a floating point point  with  s and p diffuse functios having an exponent of 0.02 .

  Basis set
    X....1s1p / Inline
    0.00 1
    * s functions
    1 1
    0.02
    1.0
    * p functions
    1 1
    0.02
    1.0
X        2.132062791     -0.539432129      0.000000000 angstrom
End of basis

However, my calculation was aborted after the frequency calculation. In the log file, I can see that polarizabilities and isotopic shifts were also printed. I have a few questions regarding this unexpected termination:

This is from the log file…

                      ** On entry to DGEBAL parameter number  3 had an illegal value
                         --- Stop Module: mclr at Tue Feb  3 15:47:04 2026 /rc=-1 ---

                      --- Module mclr spent 13 minutes 6 seconds ---

                        .########################.
                        .# Non-zero return code #.
                        .########################.


                                  Aborting...
                         Timing: Wall=9619.02 User=1503.13 System=431.63

The status file showed:

MCLR: Solving CP(CAS)HF equations for perturbation 45.

From the log file, it appears that perturbation 45 is related to the floating point.

------------------------------------
      No.    Symmetry    Center Direction
     -------------------------------------
       1       a'           N1         x
       2       a'           N1         y
       3       a'           C2         x
       4       a'           C2         y
       5       a'           C3         x
       6       a'           C3         y
       7       a'           C4         x
       8       a'           C4         y
       9       a'           C5         x
      10       a'           C5         y
      11       a'           H6         x
      12       a'           H6         y
      13       a'           H7         x
      14       a'           H7         y
      15       a'           H8         x
      16       a'           H8         y
      17       a'           H9         x
      18       a'           H9         y
      19       a'           H10        x
      20       a'           H10        y
      21       a'           O11        x
      22       a'           O11        y
      23       a'           H12        x
      24       a'           H12        y
      25       a'           H12        z
      26       a'           X          x
      27       a'           X          y
      28       a'           MLTPL  1 X
      29       a'           MLTPL  1 Y
      30       a"           N1         z
      31       a"           C2         z
      32       a"           C3         z
      33       a"           C4         z
      34       a"           C5         z
      35       a"           H6         z
      36       a"           H7         z
      37       a"           H8         z
      38       a"           H9         z
      39       a"           H10        z
      40       a"           O11        z
      41       a"           H12        x
      42       a"           H12        y
      43       a"           H12        z
      44       a"           X          z
      45       a"           MLTPL  1 Z
     -------------------------------------


 ****************************************
 *                                      *
 *           WARNING!!                  *
 * Elements in the E^[2] matrix small!! *
 * The calculation might diverge.       *
 *                                      *
 * Check your active space!!!!          *
 *                                      *
 * Make sure degenerate orbitals do not *
 * belong to different spaces.          *
 * Note that no LR code can handle      *
 * 2.0 occupancy in active orbitals!!   *
****************************************

Also, I have found warnings as given above in the output for perturbations 28, 29, and 45, which appear to be associated with the floating point. I have carefully checked my active space and the natural occupation numbers — none of the occupations are close to 2.0.

My question is: is this abrupt termination of the calculation related to the floating point ? Is there any way to obtain a result without encountering this termination issue?

2. My next question is: since the optimization has finished and the frequency calculation was performed, can I consider this structure as the minimum-energy structure of the molecule?
I obtained six imaginary frequencies — does this mean the optimized structure is not a true minimum?

symmetry a'

                    1         2         3         4         5         

     Frequency:       i41.45     i0.01     i0.01      9.49     13.38    


symmetry a" 

                    1         2         3         4         5         

     Frequency:       i47.50     i3.88      0.03     16.81     96.10   

3.Is it correct to perform a CASPT2 energy calculation at this finally optimized geometry using the Project.rasorb file even though the MCLR module calculation (frequency analysis) did not complete?

I would be really thankful if anyone could comment on these issues.

Anugraha C P