[CP2K-user] [CP2K:21768] Re: Single point energy diverges in large box, but converges in smaller box? P_Mix problem?

[Marcella Iannuzzi]

you need to find yourself what the right box is
sorry, we cannot help with that
Marcella

On Friday, August 22, 2025 at 6:43:26 PM UTC+2 andreas... at gmail.com wrote:

> you are 100% right, my mistake! 
>
> But why is it then exploding in the original box with 16 A, but works in 
> the smaller box?
>
> Marcella Iannuzzi schrieb am Freitag, 22. August 2025 um 17:53:27 UTC+2:
>
>> Hi Andreas, 
>>
>> This is a structure problem. Having a triclinic cell with c vector  1.625 
>>     1.123    10.481 
>> you cannot expect to add something to the z component and maintain the 
>> same symmetry. 
>> The angles alpha and beta are change and probably the box is not anymore consistent 
>> with the coordinates.
>> Regards
>> Marcella
>>
>> CELL_TOP| Vector a [angstrom    37.453     0.000     0.000   |a| =    
>> 37.452754
>>
>>  CELL_TOP| Vector b [angstrom   -20.749    29.823     0.000   |b| =    
>> 36.331227
>>
>>  CELL_TOP| Vector c [angstrom     1.625     1.123    10.481   |c| =    
>> 10.665429
>>
>>  CELL_TOP| Angle (b,c), alpha [degree]:                                
>> 90.034600
>>
>>  CELL_TOP| Angle (a,c), beta  [degree]:                                
>> 81.235440
>>
>>  CELL_TOP| Angle (a,b), gamma [degree]:                               
>> 124.828009
>>
>>
>>
>> CELL_TOP| Vector a [angstrom    37.453     0.000     0.000   |a| =    
>> 37.452754
>>
>>  CELL_TOP| Vector b [angstrom   -20.749    29.823     0.000   |b| =    
>> 36.331227
>>
>>  CELL_TOP| Vector c [angstrom     1.625     1.123    16.481   |c| =    
>> 16.598865
>>
>>  CELL_TOP| Angle (b,c), alpha [degree]:                                
>> 90.022232
>>
>>  CELL_TOP| Angle (a,c), beta  [degree]:                                
>> 84.381360
>>
>>  CELL_TOP| Angle (a,b), gamma [degree]:                               
>> 124.828009
>>
>> On Friday, August 22, 2025 at 2:53:43 PM UTC+2 andreas... at gmail.com 
>> wrote:
>>
>>> Hallo,
>>>
>>> I want to calculate RESP charges for a COF structure with pbc in x and y 
>>> direction. For ease I put vacuum on top to get the charges of a singular 
>>> layer. This procedure worked out for another structure with 66 atoms. 
>>>
>>> For this structure which has 132 atoms, the single point energy 
>>> calculation FAILS! The energy quickly diverges if I consider a large box in 
>>> z direction, but converges if I have a smaller box. What is the reason for 
>>> that? At the P_Mix step it diverges, I cannot even get to DIIS minimizer!
>>>
>>> I have attached the input file resp.inp, the only thing changed between 
>>> runs is the cell in z direction from 16 A to 10 A as well as the xyz 
>>> structure and the output for the large cell. The working output.dat is also 
>>> attached. Note again that the only thing which I changed is the smaller 
>>> cell.
>>>
>>> Any input is appreciated, as I do not understand why the cell size makes 
>>> the energy diverge!
>>>
>>> Sincerely, 
>>> Andreas Doell
>>>
>>

-- 
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 visit https://groups.google.com/d/msgid/cp2k/cf488817-8081-4f1b-8f06-5520217b391bn%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20250822/78e1c97f/attachment.htm>