[CP2K-user] [CP2K:15013] Re: The temperature I get is different from what I set when do massive and global thermostat.

[Travis]

Hi,

Yes, use the configuration from geometry optimization to start MD. Though 
in your original post, the thermostat coupling constant is much greater 
than the length of the trajectory you ran, whether massive or global - it's 
not doing anything here. The velocities are randomized and scaled to the 
target temperature by default as well so you should expect differences in 
the initial temperature fluctuations. After running for some time and 
reaching equilibrium, the average temperature from both thermostat settings 
should be equal. The massive thermostat controls every degree of freedom so 
it should get to the target temperature in fewer MD steps, making it ideal 
for equilibration. You should exercise caution in using MASSIVE for a given 
thermostat and coupling constant when you are interested in dynamic 
properties during data collection (aka production) runs. See, J. Chem. 
Theory Comput. 2013, 9, 2887−2899.

-T

On Friday, March 26, 2021 at 10:15:18 PM UTC-4 tang... at gmail.com wrote:

> Hello Julian,
>      Thanks for your suggestion. I will do geometry optimization 
> immediately. However, I'm a little confused about the last sentence  in 
> your reply, whether I should use the structure gotten by geometry 
> optimization to run massive and global thermostat firstly and than run 
> others processing? I hope i have expressed my question clearly. 
>       Tangyu.
>
> Julian Heske <j... at gmail.com> 于2021年3月26日周五 下午11:03写道：
>
>> Hello,
>>
>> in order to run a MD calculation at a specific tempertare, you have to 
>> equilibrate the system first to reach your target temperature. The time you 
>> need to equilibrate your system is dependent on your initial configuration. 
>> Since you are starting with a structure which has a very high potential 
>> energy (-157,5 vs. ~ -168 a. u. of the others, see out.txt), this energy 
>> remains as kinetic energy in the system and the thermostat will need a long 
>> time to remove the energy from the system. You can first do a geometry 
>> optimization ( CP2K_INPUT <https://manual.cp2k.org/trunk/CP2K_INPUT.html> 
>> / MOTION <https://manual.cp2k.org/trunk/CP2K_INPUT/MOTION.html> / GEO_OPT 
>> <https://manual.cp2k.org/trunk/CP2K_INPUT/MOTION/GEO_OPT.html> ), which 
>> leads to an optimized structure. Then start your MD run with your obtained 
>> structure as initial configuration. However, you should then still take a 
>> look at the temperature and not use the equilibration steps for your 
>> analysis.
>>
>> Kind regards,
>> Julian
>> tang... at gmail.com schrieb am Freitag, 26. März 2021 um 15:16:33 
>> UTC+1:
>>
>>> Dears cp2k's users,
>>>       I'm doing a vibration analysis exercise.When I did massive and 
>>> global thermostat, the temperature I get is different from what I set. 
>>> Could you tell me what causes this error? I would appreciate you generous 
>>> help. The input and output file are listed as follow.
>>>
>> -- 
>> You received this message because you are subscribed to a topic in the 
>> Google Groups "cp2k" group.
>> To unsubscribe from this topic, visit 
>> https://groups.google.com/d/topic/cp2k/MnB_Ejspnpw/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to 
>> cp... at googlegroups.com.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/cp2k/ac99e428-7df6-4e03-8b6e-cbf7c16009dbn%40googlegroups.com 
>> <https://groups.google.com/d/msgid/cp2k/ac99e428-7df6-4e03-8b6e-cbf7c16009dbn%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210326/f845763f/attachment.htm>