[CP2K:9803] Problems of shell model and optimization

[liuxians... at gmail.com]

Hi Prof. Matthias Krack,

is final version CP2K 4.1 and CP2K 5 implement this change?.


Liu

On Sunday, December 17, 2017 at 4:53:45 PM UTC+1, Matthias Krack wrote:
>
> Hi zuoheng199247
>
> there has been a bug in the force packing routine which I have fixed now 
> in the CP2K trunk version. The bug affected optimisation runs (geo_opt, 
> cell_opt, and possibly NEB runs) involving core-shell atoms combined with 
> intramolecular potentials. MD runs shouldn't be affected. You may try again 
> the optimisation runs which failed with the CP2K trunk version.
> Thanks for reporting that issue.
>
> Best regards
>
> Matthias
>
> On Friday, 15 December 2017 08:37:32 UTC+1, zuoh... at gmail.com wrote:
>>
>> I  am sorry. I have a mistake on my reference. The paper is T.K. Kundu et 
>> al. / Journal of Colloid and Interface Science 295 (2006) 364–373. This is 
>> the link: http://linkinghub.elsevier.com/retrieve/pii/S0021979705009574
>>
>> 在 2017年12月15日星期五 UTC+8上午12:42:54，Matthias Krack写道：
>>>
>>> Hi zuoheng199247
>>>
>>>  
>>>
>>> If I am not mistaken, bonded (intramolecular) interactions involving 
>>> core-shell atoms are not considered/implemented in CP2K. MD might run, but 
>>> with wrong forces obviously as shown by the GEO_OPT runs. So your MD runs 
>>> are presumably not OK, too.
>>>
>>>  
>>>
>>> Best regards
>>>
>>>  
>>>
>>> Matthias
>>>
>>>  
>>>
>>>  
>>>
>>> *From:* cp... at googlegroups.com [mailto:cp... at googlegroups.com] *On 
>>> Behalf Of *zuoh... at gmail.com
>>> *Sent:* 14 December 2017 14:52
>>> *To:* cp2k
>>> *Subject:* [CP2K:9803] Problems of shell model and optimization
>>>
>>>  
>>>
>>> Dear friends,
>>>
>>>  
>>>
>>> I have a problem when I use core-shell model with bond potential to do 
>>> cell(geometry) optimization at the same time. I choose classical force 
>>> field. In FORCE_EVAL / MM / FORCEFIELD, I set some buckingham potential and 
>>> morse potential bond and hamonic bend parameters.
>>>
>>>  
>>>
>>> When I only use bond potential and optimization without core-shell 
>>> model, or only use core-shell model and bond potential to do MD 
>>> simulations, all of them are OK. When I use the three modules at the same 
>>> time, the computation will interrupt after some steps, though the CPU is 
>>> still occupied and don't have any error or warning in output files.
>>>
>>>  
>>>
>>> The following table is more clear. The input files are given as three 
>>> attach files.
>>>
>>>  
>>>
>>> I want to use core-shell model with classical force field to do energy 
>>> minimization. I don't know how to fix this problem even though I have study 
>>> FORCE_EVAL / MM / FORCEFIELD section in CP2K 4.1 manual. I am confused 
>>> understand why the MD simulation is OK with shell model but optimization is 
>>> not OK.
>>>
>>>  
>>>
>>> I hope to receive your reply. Thank you very much.
>>>
>>>  
>>>
>>> Best regards,
>>>
>>>  
>>>
>>>
>>> <https://lh3.googleusercontent.com/-ja4OPffuI4k/WjKBVb6x4_I/AAAAAAAAAAY/KC0FJV43HEkQqcd0vBdzYIc8ue8H4bQgACLcBGAs/s1600/%25E5%25BE%25AE%25E4%25BF%25A1%25E5%259B%25BE%25E7%2589%2587_20171214214846.png>
>>>
>>> -- 
>>> 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+... at googlegroups.com.
>>> To post to this group, send email to cp... at googlegroups.com.
>>> Visit this group at https://groups.google.com/group/cp2k.
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20180223/8c860d2e/attachment.htm>