Partial Atomic Charge Calculation
21 views
Skip to first unread message
Maryam Bahrami
Sep 27, 2025, 7:20:28 AM (5 days ago) Sep 27
to cp2k
Dear all
I am new to cp2k. I want to compute the partial atomic charges on amino acid molecule using cp2k. Is the charge computed from cp2k accurate like computed from Gaussian software? Is it possible to compute accurate charge using cp2k? Please let me know.please help me how to compute charge on atoms by CP2K.
Thanks
Konstantin Tokarev
Sep 28, 2025, 5:39:35 AM (4 days ago) Sep 28
to cp2k
Hello,
CP2K supports Mulliken, Hirshfeld and Löwdin population analysis. The former two are enabled by default at PRINT_LEVEL MEDIUM, however it's possible to enable or disable them explicitly via respective options in the input file, e.g. FORCE_EVAL/DFT/PRINT/MULLIKEN and so on. See the manual [1] for more details.
If you want to compare results with other software like Gaussian, make sure you are using the same kind of population analysis, and other conditions of the calculation are the same. E.g. if you calculate isolated molecule in vacuum, make sure the you define it in CP2K as a non-periodic system, and you use the same DFT functional and the same (probably all-electron) basis set in both programs.
[1] https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/DFT/PRINT.html
CP2K supports Mulliken, Hirshfeld and Löwdin population analysis. The former two are enabled by default at PRINT_LEVEL MEDIUM, however it's possible to enable or disable them explicitly via respective options in the input file, e.g. FORCE_EVAL/DFT/PRINT/MULLIKEN and so on. See the manual [1] for more details.
If you want to compare results with other software like Gaussian, make sure you are using the same kind of population analysis, and other conditions of the calculation are the same. E.g. if you calculate isolated molecule in vacuum, make sure the you define it in CP2K as a non-periodic system, and you use the same DFT functional and the same (probably all-electron) basis set in both programs.
[1] https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/DFT/PRINT.html
суббота, 27 сентября 2025 г. в 14:20:28 UTC+3, Maryam Bahrami:
Konstantin Tokarev
Sep 28, 2025, 6:08:05 AM (4 days ago) Sep 28
to cp2k
For example, attached input file calculates all three types of partial charges for an isolated C₂H₂ molecule with PBE functional and 6-31G(d,p) basis set. Some notable options are:
1) `&CELL` section defines fake cubic cell, and `&CENTER_COORDINATES` centers the molecule in it instead of using (0, 0, 0) as a center;
2) `PERIODIC NONE` has to be specified both in `&CELL` and `&POISSON` sections;
3) POISSON_SOLVER WAVELET is used, and `PREFERRED_FFT_LIBRARY FFTSG` is added to `&GLOBAL` because of that;
4) `&PRINT` in `&DFT` requests calculation of three types of charges;
5) `METHOD GAPW` and `BASIS_SET_FILE_NAME EMSL_BASIS_SETS` are required for using all-electron `BASIS_SET "6-31Gxx"` in atom `&KIND` sections.
1) `&CELL` section defines fake cubic cell, and `&CENTER_COORDINATES` centers the molecule in it instead of using (0, 0, 0) as a center;
2) `PERIODIC NONE` has to be specified both in `&CELL` and `&POISSON` sections;
3) POISSON_SOLVER WAVELET is used, and `PREFERRED_FFT_LIBRARY FFTSG` is added to `&GLOBAL` because of that;
4) `&PRINT` in `&DFT` requests calculation of three types of charges;
5) `METHOD GAPW` and `BASIS_SET_FILE_NAME EMSL_BASIS_SETS` are required for using all-electron `BASIS_SET "6-31Gxx"` in atom `&KIND` sections.
воскресенье, 28 сентября 2025 г. в 12:39:35 UTC+3, Konstantin Tokarev:
Reply all
Reply to author
Forward
0 new messages