How to construct non-self-consistent density matrix and pass it to the RT-TDDFT module

[Zhubin HU]

Dear all,

I am trying to perform a real-time TDDFT (RT-TDDFT) simulation in NWChem 7.2.3 where the initial state is prepared by removing an electron from a specific molecular orbital (e.g., HOMO–1) of a neutral molecule, without performing a subsequent SCF optimization.

According to some literature (e.g., J. Phys. Chem. Lett. 2017, 8, 3991−3996), this can be done by:

1. Performing a ground-state SCF calculation for the neutral molecule.

2. Removing an electron from a specific orbital (e.g., HOMO–1) by using OCCUP directive.

3. Propagating the nonstationary density using the rt_tddft module, starting from this ionized configuration.

What I’ve done:

• I can use OCCUP directive  and "noscf" keywords in DFT  directive to get a initial state with removing an electron from a specific orbital (e.g., HOMO–1) and save the **.moves file.

• I load the .moves in the  rt_tddft module by "load vectors **.moves"

Problem:

Even though I provide the correct **.movecs , the initial density at time 0 in the RT-TDDFT run appears to correspond to the ground-state HOMO-ionized density, not the manually HOMO–1-ionized density I intended.  This suggests that NWChem may recompute or override the density internally, ignoring the load density directive, or perhaps I'm missing a step to ensure the non-self-consistent density is honored.

Questions:

1. How can I ensure that RT-TDDFT truly uses the non-ground-state density matrix generated from my modified .movecs?

2. Is there an internal flag or keyword required to prevent density recomputation at time step 0?

3. Is there a more reliable way to construct and pass a non-self-consistent density matrix to RT-TDDFT in NWChem?

I’ve attached a minimal test case using a water molecule, with HOMO–1 ionized, if needed.

Any guidance or insights would be greatly appreciated!

Best regards,
Zhubin HU

[Edoardo Aprà]

Could you upload the output file, too? Please use the "echo" keyword in the input file so that the input file content appears in the output file, too.

https://nwchemgit.github.io/ECHO.html

[Zhubin HU]

Thank you Edoardo.

To clarify my issue:
I attempted to prepare an initial state where an electron is removed from the HOMO–1 orbital . 

Prior to the RT-TDDFT simulation, I used the "OCCUP" directive in the DFT section to manually modify the occupation numbers, specifically removing one beta electron from the HOMO–1 orbital. 

I also used the "noscf" keyword to avoid any subsequent SCF optimization, so that the resulting electron density would reflect the non-stationary configuration. 

After this step, I used the DPLOT module to visualize the density difference (ρ_cation – ρ_neutral), and confirmed that the electron density indeed corresponds to a hole in the HOMO–1 orbital, as expected.

However, once I launched the RT-TDDFT simulation using this modified state (by loading the .movecs file from the prior noscf DFT calculation), I found that the step-0 total energy (which is significantly lower than the energy obtained in the noscf step), as well as the electron density, were identical to the results of a separate RT-TDDFT calculation where the hole was placed in the HOMO instead. 

This led me to suspect that something in the RT-TDDFT setup might be overwriting or re-projecting the initial density — perhaps reverting to a ground-state configuration — although I am not certain.

To help clarify, I’ve attached output files (with the "echo" keyword enabled so that input details are included):
water-HOMO.log (removing one beta electron from HOMO)
water-HOMO-1.log (removing one beta electron from HOMO–1)

[Niri Govind]

Hi Zhubin,

Thank you for reporting this issue and for the detailed description. We will take a look at it.

Best,

Niri 

Niri Govind

PNNL

--
You received this message because you are subscribed to the Google Groups "NWChem Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nwchem-forum...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/nwchem-forum/f8eab16e-766e-440e-a1b5-e04c5484b51an%40googlegroups.com.

[Zhubin HU]

Hello Dr. Niri Govind,

I hope this message finds you well.

I just wanted to kindly follow up on the RT-TDDFT issue I previously reached out about (regarding the initial density when removing an electron from the HOMO–1 orbital). I realize you're very busy, but I was wondering if there has been any update on the case I sent.

I'm still very interested in understanding whether the RT-TDDFT module may be overriding or recalculating the density at step 0, and any insight from your team would be greatly appreciated.

Thank you again for your time and consideration!

Best regards,
Zhubin