Describe the bug
when opening an ORCA.out freq calculation with a QM1/QM2 level, the software cannot generate an IR spectra but shows
"ValueError: x and y must have same first dimension, but have shapes (4232,) and (1,)"
this is caused by the fact in a Partial Hessian calculation Orca outputs:
-----------
IR SPECTRUM WARNING: Spectrum is based on Partial Hessian
-----------
Mode freq eps Int T**2 TX TY TZ
cm**-1 L/(mol*cm) km/mol a.u.
----------------------------------------------------------------------------
29: 5.01 0.000000 0.00 0.000000 (-0.000061 0.000011 -0.000032)
30: 5.02 0.000000 0.00 0.000000 (-0.000016 0.000010 0.000028)
[...]
removing "WARNING: Spectrum is based on Partial Hessian" allows Chimera to load an IR spectra.
However, because the partial Hessian doesn't start at 0 everything gets shifted (in my case by 29)
This means everything clusters on a frequency at 1.29cm-1 in the full hessian, located only in the "VIBRATIONAL FREQUENCIES" and "THERMOCHEMISTRY AT 298.15K" sections of the output (see screenshots)
To Reproduce
Steps to reproduce the behavior:
- remove "WARNING: Spectrum is based on Partial Hessian" line from ORCA output
- load output.out
- go to "tools -> Quantum Chemistry -> IR spectrum"
- see that the spectra makes no sense
- go to "Tools -> Quantum Chemistry -> Visualise Normal Modes"
- see the mismatch between IR intensity and Frequency compared to "IR SPECTRUM" section of orca output
and I get the same issue with the Raman spectra
Expected behavior
I expect to see the IR spectra that matches the "IR SPECTRUM" section of the Orca output file (which is the case for the calculation of just the QM1 region with no hybrid hessian)
Screenshots
side by side comparison of Chimera Normal modes and ORCA output. see how the "Int" column is the same as "IR intensity", but the frequencies are wrong

instead, frequencies line up with the frequencies from the "VIBRATIONAL FREQUENCIES" section

and the IR spectra obtained looks like this:
Structure:
not structure dependent, but here's my orca.out (I've already removed the "WARNING: Spectrum is based on Partial Hessian" lines and put a .txt suffix to be able to upload it)
Gly_water_freq_warningRemoved.out.txt
Thank you for your help and have a nice day !
Leo
Describe the bug
when opening an ORCA.out freq calculation with a QM1/QM2 level, the software cannot generate an IR spectra but shows
"ValueError: x and y must have same first dimension, but have shapes (4232,) and (1,)"
this is caused by the fact in a Partial Hessian calculation Orca outputs:
removing "WARNING: Spectrum is based on Partial Hessian" allows Chimera to load an IR spectra.
However, because the partial Hessian doesn't start at 0 everything gets shifted (in my case by 29)
This means everything clusters on a frequency at 1.29cm-1 in the full hessian, located only in the "VIBRATIONAL FREQUENCIES" and "THERMOCHEMISTRY AT 298.15K" sections of the output (see screenshots)
To Reproduce
Steps to reproduce the behavior:
and I get the same issue with the Raman spectra
Expected behavior
I expect to see the IR spectra that matches the "IR SPECTRUM" section of the Orca output file (which is the case for the calculation of just the QM1 region with no hybrid hessian)
Screenshots

side by side comparison of Chimera Normal modes and ORCA output. see how the "Int" column is the same as "IR intensity", but the frequencies are wrong
instead, frequencies line up with the frequencies from the "VIBRATIONAL FREQUENCIES" section

and the IR spectra obtained looks like this:
Structure:
not structure dependent, but here's my orca.out (I've already removed the "WARNING: Spectrum is based on Partial Hessian" lines and put a .txt suffix to be able to upload it)
Gly_water_freq_warningRemoved.out.txt
Thank you for your help and have a nice day !
Leo