cp2k geometry optimization energy not decreasing

[Matt W]

Nothing is badly wrong at a quick look. But there are a some things in your 
input that could be improved. 

I think the most important thing is to improve your forces  - use a larger 
CUTOFF value, this 
<https://www.cp2k.org/events:2018_summer_school:converging_cutoff> might be 
helpful. 

Use a better minimizer, especially if you restart from near the minima 
already found - I'd suggest removing all the GEO_OPT options to use the 
default BFGS optimizer.

Just for efficiency, you don't need such a large box, I don't think. 5 
angstrom away from any atom should be enough.

Matt 


On Wednesday, June 27, 2018 at 7:13:02 PM UTC+1, Raghav wrote:
>
> Dear cp2k users,
>
>                         I am doing geometry optimization of a 
> configuration but the ENERGY is stuck at a value and is not decreasing and 
> the calculation is running like this and not getting finished. Am I doing 
> something wrong?? Need help.
>
>
>
>  *******************************************************************************
>  ***                 BRENT   - NUMBER OF ENERGY EVALUATIONS :       
> 1        ***
>
>  *******************************************************************************
>
>  --------  Informations at step =   253 ------------
>   Optimization Method        =                   SD
>   Total Energy               =      -361.6413293878
>   Real energy change         =         0.0000000001
>   Decrease in energy         =                   NO
>   Used time                  =              185.503
>
>   Convergence check :
>   Max. step size             =         0.0000000000
>   Conv. limit for step size  =         0.0030000000
>   Convergence in step size   =                  YES
>   RMS step size              =         0.0000000000
>   Conv. limit for RMS step   =         0.0015000000
>   Convergence in RMS step    =                  YES
>   Max. gradient              =         0.0006486901
>   Conv. limit for gradients  =         0.0004500000
>   Conv. for gradients        =                   NO
>   RMS gradient               =         0.0001373171
>   Conv. limit for RMS grad.  =         0.0003000000
>   Conv. in RMS gradients     =                  YES
>  ---------------------------------------------------
>
> With regards
> Raghav
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20180627/3d2a9b56/attachment.htm>