Differences Between Mulliken and Hirshfeld Charge Analysis
Kha Trinh
Dear CP2K Users,
I’m working on the geometry optimization of a transition state featuring an O--Li--O structure using CP2K and ASE. As part of the analysis, I calculated Mulliken and Hirshfeld charges, but I noticed significant discrepancies between the two methods.
Here are the results for two cases of the transition state:
Mulliken Charges:
Oxygen (O): -0.361
Lithium (Li): 0.565
Oxygen (O): -0.367
Hirshfeld Charges
Oxygen (O): -0.261
Lithium (Li): -0.470
Oxygen (O): -0.265
The results from Mulliken and Hirshfeld methods show opposite trends regarding the charge on the Li atom. The Hirshfeld analysis, in particular, seems counterintuitive because I would expect lithium, located between two oxygen atoms, to have a partially positive charge. However, in my field, it is more common to report Mulliken charges.
Could this discrepancy arise from the charge partitioning methods or my calculation settings? My system includes C, H, O, N, and Li atoms, using the DZVP-GTH basis set and GTH-PBE functional. Any advice on interpreting or validating these results would be greatly appreciated. I have attached the relevant input and output files for reference.
Thank you so much in advance
Best regards,
Kha
Thomas Kühne
--
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+uns...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/3417dede-bb67-46a0-85ff-91a941b8dc81n%40googlegroups.com.
<last_frame.xyz><PROJECT-hirshfeld_charges.hirshfeld><geo_opt.py><PROJECT-mulliken_charges.mulliken>