NEB calculations - energy ?

[Marcella Iannuzzi]

Hi

did you check whether the geometries of the end-points have actually 
changed?
The keyword is 
BAND&OPTIMIZE_BAND&OPTIMIZE_END_POINTS
but its default  is .false.
 
Did you set ALIGN_FRAME and ROTATE_FRAME to false? 
You better extract the end points' geometries and optimise them again to 
see what happens.
Kind regards
Marcella


On Thursday, March 23, 2017 at 2:14:28 PM UTC+1, Mathilde IACHELLA wrote:
>
> Hi Marcella,
>
> Thank you for your quick answer ! 
> I did not fix the first and last images, in fact. Is there a keyword that 
> allows it ? 
> Thank you again 
>
> Mathilde 
>
> Le jeudi 23 mars 2017 11:17:20 UTC+1, Marcella Iannuzzi a écrit :
>>
>> Hi Mathilde,
>>
>> Are the end geometries fixed or allowed to change?
>> If they are allowed to change, this would  explain why the energy also 
>> changes. The optimisation of the NEB's beads is affected by the NEB's 
>> conditions.
>> Kind regards 
>> Marcella
>>
>> On Wednesday, March 22, 2017 at 2:49:49 PM UTC+1, Mathilde IACHELLA wrote:
>>>
>>> I'm using CP2K since 4 months now and I performed NEB calculations. I 
>>> wondered that the energy for the initial and final replica are different 
>>> from the energies of the same structures found by geometric optimizations 
>>> (with the same pseudopotential, same basis set, same functional ...). The 
>>> difference is around 0.3 eV (~0.01 a.u.)
>>> Does someone have an explanation to this ? Is it just because of the 
>>> convergence/accuracy of the system ? 
>>>
>>> Thank you in advance, 
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170323/13e92b84/attachment.htm>