Positive Polarization energy with SCCS model

Janson Shi

unread,

Mar 13, 2023, 3:19:45 AM3/13/23

to cp2k

Dear sir,

Sorry to bother you again. Recently, I have been bothered with SCCS model using cp2k software. Today, I have used the SCCS model to simulate the implicit solvent effect on 2D COF. However, the polarization energy was jumped from negative value to positive value as marked in the figures below. Although the calculation has been converged at last with large positive polarization energy.

Herein, is the positive polarization energy normal? As far as I know, the solvent should decrease the potential surface of the system, so it should be negative.

If the simulation using SCCS model has some problem with the parameter setting, please give me some instructions to make it right.

The corresponding files have been attached, and many thanks for your help all the time.

Best regards,

Janson Shi

scf_N3-COF_PBE0_vdw_sol.out

N3-COF.cif

scf_N3-COF_PBE0_vdw_sol.inp

Krack Matthias

unread,

Mar 13, 2023, 9:25:18 AM3/13/23

to cp...@googlegroups.com

Hi

Convergence does not mean that you obtained a physically meaningful result necessarily. A positive solvation energy is possible and simply means that the solute is not solvable in the selected solvent based on your simulation (model).

As suggested earlier, try the default derivative method FFT instead of CD5, especially when you have a non-orthorhombic cell like in your case. You can also recast your unit cell as orthorhombic.

HTH

Matthias

--
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 on the web visit https://groups.google.com/d/msgid/cp2k/d03f3645-c2a4-4518-8e64-e554dff28afbn%40googlegroups.com.

施张胜

unread,

Mar 13, 2023, 9:44:00 AM3/13/23

to cp...@googlegroups.com

Dear sir,

Still doesn't make sense, since the COF has polar atoms like N. We simulated deformed COF induced by H2O from AIMD with the same parameters, and the polarization energy is negative as expected.

Following your suggestions, trying the default derivative method FFT still get the positive polarization energy. Should I try other derivative methods?

------------------ 原始邮件 ------------------

发件人: "cp2k" <matthia...@psi.ch>;

发送时间: 2023年3月13日(星期一) 晚上9:25

收件人: "cp...@googlegroups.com"<cp...@googlegroups.com>;

主题: Re: [CP2K:18538] Positive Polarization energy with SCCS model

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ZRAP278MB08270E02952FA15BFE9DABF3F4B99%40ZRAP278MB0827.CHEP278.PROD.OUTLOOK.COM.

D8575246@5B85A132.98280F64.jpg

3C91EFA8@64829054.98280F64.jpg