[CP2K-user] [CP2K:19878] Re: About the Dirichlet boundary

[Marcella Iannuzzi]

Hi Lakshmi

If you set a job time, then it is  good practice to also give a proper 

   - 
   
   WALLTIME 
   <https://manual.cp2k.org/trunk/CP2K_INPUT/GLOBAL.html#CP2K_INPUT.GLOBAL.WALLTIME>
   
   in input, such that cp2k can close up the calculation before being 
   killed by the scheduler. This is the best way to get proper restart files 
   such that one can restart the simulation if the allocated time was not 
   sufficient.  
   
   
   You did not explain why the output you posted contains data from two 
   different simulations, one that seems to converge better than the other. 
   
   One part of the output you posted shows a geo_opt running over 12 
   iterations, the gradients are not huge, but still high. I wonder whether 
   you tried to run geo_opt for the same system without the complications of 
   the implicit poisson solver first. Such an optimization can take many 
   hours, depending on many different factors 
   
   
   
   The extrapolation method is selected with 
   - FORCE_EVAL <https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL.html> 
   - DFT <https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/DFT.html> 
   - QS EXTRAPOLATION


   - there are different choices, which are better then just the atomic 
   guess.

Why are you activating the kpoints if you then require only gamma? By doing 
this, there are several restrictions to the available features. 

Regards
Marcella



On Saturday, February 3, 2024 at 6:25:38 PM UTC+1 lakshmi... at gmail.com 
wrote:

> Hi Marcella,
> I have set the time for the simulation to 10 hours. I am not aware of how 
> much time it takes to complete the simulation. I never tried the 
> extrapolation method. Can you please explain to me how the extrapolation 
> method is applied?
>
> Thanks,
> Lakshmi.
>
> On Sat, Feb 3, 2024 at 12:00 PM Marcella Iannuzzi <marci... at gmail.com> 
> wrote:
>
>> Hi Lakshmi
>>
>> The posted output file has data from two un-finished runs. The first part 
>> seems to correspond to a run where the convergence was not too bad.
>> Why was it stopped? The second seems not to be restarted from the 
>> previous one.
>> In general the Broyden mixing works better with a smaller ALPHA.
>> Have you tried to use an extrapolation method? 
>>
>> Regards
>> Marcella
>>
>>
>>
>> On Friday, February 2, 2024 at 4:48:44 PM UTC+1 lakshmi... at gmail.com 
>> wrote:
>>
>>> Dear All,
>>> I had tried with a large box as well as decreasing the intercept and 
>>> thickness. The simulation started running but it was not converging. It 
>>> almost ran for 10 hours but it didn't converge. I don't understand what is 
>>> missing in my input. Could anyone please help me to solve the problem? I am 
>>> attaching both the edited input and the output I got for your reference.
>>>
>>> Thanks,
>>> Lakshmi.
>>>
>>> On Wednesday, January 31, 2024 at 3:11:32 PM UTC-5 Lakshmi Anil wrote:
>>>
>>>> The size along z is not mentioned in the article. They are using around 
>>>> 200 atoms structure. I will try by using a smaller intercept value and also 
>>>> one with a larger box.
>>>> Thank you so much.
>>>> Lakshmi.
>>>>
>>>> On Wed, Jan 31, 2024 at 2:53 PM Marcella Iannuzzi <marci... at gmail.com> 
>>>> wrote:
>>>>
>>>>> I see, 
>>>>> but which box size do they use in z? you have 21.7 which is smaller 
>>>>> than 24, and a thickness of 0.5. 
>>>>> Have you tried with a larger box?
>>>>>
>>>>> Best
>>>>> Marcella
>>>>>
>>>>> On Wednesday, January 31, 2024 at 6:37:12 PM UTC+1 
>>>>> lakshmi... at gmail.com wrote:
>>>>>
>>>>>> Dear Marcella,
>>>>>> I was trying to repeat the same system given in the article :
>>>>>> DOIhttps://doi.org/10.1039/D2CP05052H. So in this, they are using 
>>>>>> the distance between the upper and lower surface as 25A0. I  try to do it 
>>>>>> in the same way. But the same error is shown. Do you have any suggestions?
>>>>>>
>>>>>> Thanks,
>>>>>> Lakshmi
>>>>>>
>>>>>> On Wed, Jan 31, 2024 at 4:07 AM Marcella Iannuzzi <marci... at gmail.com> 
>>>>>> wrote:
>>>>>>
>>>>>>> Dear Lakshmi
>>>>>>>
>>>>>>> I suspect that the problem might come from the intercept 25, but I 
>>>>>>> do not have experience with the method.
>>>>>>>
>>>>>>> Regards
>>>>>>> Marcella
>>>>>>>
>>>>>>> On Wednesday, January 31, 2024 at 3:41:29 AM UTC+1 
>>>>>>> lakshmi... at gmail.com wrote:
>>>>>>>
>>>>>>>> Dear All,
>>>>>>>> I am new to CP2K. I was trying to give an electric field to my Ru 
>>>>>>>> metal surface(144 atoms)with adsorbed nitrogen. For applying the electric 
>>>>>>>> field, I used an Implicit Poisson solver and Dirichlet boundary condition 
>>>>>>>> are used. Unfortunately, I got an error that the Dirichlet extents are 
>>>>>>>> outside the simulation cell even though I gave the same simulation cell 
>>>>>>>> dimension. Then, I reduced it to half to check whether it works, but it 
>>>>>>>> again gave me the same error. Could anyone please help me with that? I am 
>>>>>>>> attaching the input file as well as the output file for the reference.
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>> Lakshmi.
>>>>>>>>
>>>>>>> -- 
>>>>>>> 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+uns... at googlegroups.com.
>>>>>>> To view this discussion on the web visit 
>>>>>>> https://groups.google.com/d/msgid/cp2k/9eca8e1d-6a1d-47de-b1cb-457a81ad05e7n%40googlegroups.com 
>>>>>>> <https://groups.google.com/d/msgid/cp2k/9eca8e1d-6a1d-47de-b1cb-457a81ad05e7n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>>>> .
>>>>>>>
>>>>>> -- 
>>>>> 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+uns... at googlegroups.com.
>>>>>
>>>> To view this discussion on the web visit 
>>>>> https://groups.google.com/d/msgid/cp2k/44d4d72c-8a2c-40a8-8fa0-004d88f7baacn%40googlegroups.com 
>>>>> <https://groups.google.com/d/msgid/cp2k/44d4d72c-8a2c-40a8-8fa0-004d88f7baacn%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>> .
>>>>>
>>>> -- 
>> 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+uns... at googlegroups.com.
>>
> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/cp2k/1a9af2ac-e679-429f-9b56-bd256053b32an%40googlegroups.com 
>> <https://groups.google.com/d/msgid/cp2k/1a9af2ac-e679-429f-9b56-bd256053b32an%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>

-- 
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/e3c44fb4-7f10-4900-a3d4-abb21bb23ce8n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240204/46d6354e/attachment.htm>