[CP2K-user] [CP2K:17983] Questions on LR-TDDFPT calculations with range-separated functionals using cp2k2022.1

[liu xiangyang]

Hello, everyone:

I was trying to do some excited state calculations involving some charge 
transfer states using the LR-TDDFPT method in cp2k2022.1. 

Actually, I have tried this before using cp2k7.1. 
https://groups.google.com/g/cp2k/c/PcPUy6FouGE/m/CKOUiGxUDAAJ
Even though all global hybrid functionals such as PBE0-TC and B3LYP-TC can 
give results comparable to the G09 program, the results obtained by some 
well known range separated functionals such as wB97XD-TC with MIX_CL_TRUNC 
are very problematic and the excited state energy difference between the 
results given by G09 and cp2k7.1 can be large than 1 eV. Such problem 
remain the same even in cp2k9.0.

Recently, I have updated the cp2k version from cp2k7.1 to cp2k2022.1 and I 
tried the same calculation using the same input as before. Luckily, the 
newest version of cp2k2022.1 seems give excitation energies close to those 
in G09, which excites me a lot for a period of time. However, when I do 
some tests on a donor-acceptor system composed of ZnPc and C60, I found 
that the lowest CT states corresponds to S1 state predicted by G09 is not 
the case in cp2k2022.1, in which the CT states have much higher energies 
than the LE states.

Since the MIX_CL_TRUNC interaction potential is used for range separated 
functional calculation, I guess the small CUTOFF_RADIUS in the HF section 
of about 5.0 ang might be the reason responsible for such inconsistency. So 
I increased the CUTOFF_RADIUS from original 5.0 ang to 10.0 ang. 
Unfortunately, whenever I use a larger CUTOFF_RADIUS for the truncated 
calculation, the convergence problem arises and I have tried reading a 
pre-converged wavefunction, tried to increase the CUTOFF in GPW 
calculation, switched to the OT algorithm, etc. However, none of them 
works  and the convergence remain fail.

I wonder whether anyone can give me some advices on the convergence problem 
or the reason for correcting such inconsitency between G09 and cp2k2022.1 
since it is important to correcly predict the charge transfer state in 
organic solar cells.

Thanks a lot.
PS: I have attached my input file and coordinate file for LR-TD-wB97XD 
caulations.



-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/6e6e41e2-dd2c-4f15-969a-021379ba547en%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20221104/1a783203/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: wB97XD.inp
Type: chemical/x-gamess-input
Size: 2472 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20221104/1a783203/attachment.inp>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: coord.xyz
Type: chemical/x-xyz
Size: 6440 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20221104/1a783203/attachment.xyz>