[CP2K-user] [CP2K:20072] Re: The treatment of atom mass in the mixing force_evals calculation

[Xu Wang]

Dear Prof. Iannuzzi,

Thank you very much for your detailed reply! Please kindly accept my 
apology for not seeing your reply until now due to the network blockade. I 
will try to understand these example input files.

Sincerely,
Xu

在2024年1月29日星期一 UTC+1 13:58:00<Marcella Iannuzzi> 写道：

> Dear Xu
>
> The way to do this type of simulation is to apply the alchemical change.
> You can find an example input in 
>
> cp2k/tests/FE/regtest-1/Solv_alch_chng.inp
>
> In practice, you need a full set of coordinates that contains both ions, 
> this will be the subsystem for the mixed force_eval, which takes care of 
> the propagation of the equations of motion, where each atom maintains its 
> own mass. Then you prepare other coordinates set for the subsystems of the 
> individual force evals. The mixed force_eval needs to have the mapping, 
> indicating which atoms of the full set belong to which individual 
> force_eval.
> I think it is more difficult to explain it in words than understanding  it 
> directly from the example input. 
>
> Regards
> Marcella
>
>
>  
>
> On Monday, January 29, 2024 at 11:10:36 AM UTC+1 wangx... at gmail.com wrote:
>
>> Dear CP2K developers, 
>>     I am sorry to bother you for a technical detail regarding the 
>> treatment of atom mass in the mixing force_evals calculation.
>>     I intend to investigate the free energy difference between Na-Cl and 
>> K-Cl ion pairs in aqueous solutions through FPMD-based thermodynamic 
>> coupling constant integration, i.e., transitioning Na(+) gradually into 
>> K(+) using the &MIXED section:
>> [image: CP1.png]
>> together with the following settings in the &SUBSYS section:
>> [image: CP2.png]
>>     During the simulation, energies and forces are mixed via lambda to 
>> solve the equation of motion. On the other hand, the masses of atoms are 
>> also required in the equation of motion. I am wondering how CP2K handles 
>> the mass of “Na” mentioned above in simulations with 0<lambda<1. 
>> Specifically, is it calculated as 0.5*(mass_Na+mass_K), lambda*mass_Na+(1- 
>> lambda)*mass_K, or utilizing another algorithm? Thanks in advance!
>>
>> Sincerely,
>> Xu
>>
>

-- 
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/46bbf67c-f044-4a33-a96b-61398cc6eb32n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240326/bd391031/attachment.htm>