How to select 'ANDERSON_PROB' or 'SMASS' in AIMD?

[Seunghwan_Kwon]

Dear all,

I am trying to execute ab initio molecular dynamics (AIMD) using VASP.

However, there is a problem in setting specific parameters of thermostat.

I have searched various documents regarding 'ANDERSON_PROB' or 'SMASS',

but I didn't get satisfying answer and there weren't exist enough documents.

So, my question is how to decide 'ANDERSON_PROB' or 'SMASS' parameter?

Thank you.

[jonathan_lahnsteiner2]

Dear Seunghwan_Kwon,

I guess you want to run a molecular dynamics simulation in the NVT ensemble or maybe the NpT ensemble.
If you want to do a NpT ensemble you have to use the Langevin thermostat:
wiki/index.php/NpT_ensemble

If you want to do a NVT ensemble, you can either use the Andersen, Nose-Hover or the Langevin thermostat within vasp.
So you do not have to set SMASS and ANDERSEN_PROB simultaneously.
To get an overview of the ensembles available in vasp I would like to guide you the ensembles page in the wiki:
wiki/index.php/Category:Ensembles
After you have chosen an ensemble you can check which thermostats are available and how to activate them in this wiki page
wiki/index.php/NVT_ensemble

I hope this resolves your problem, otherwise contact us again

All the best Jonathan

[Seunghwan_Kwon]

Thank you for replying.

I wanted to do NVT ensemble,

but the determination of specific parameters that I have to set in order to do NVT simulation such as "SMASS" is quite ambiguous.

Do I have to decide "SMASS" empirically?

I mean, do I have to determine the "SMASS" by changing the number of "SMASS" in a row and compare the temperature fluctuation?

For example, changing "SMASS" 0, 1, 2, 3 sequentially and compare the results and choose the optimized one?

[jonathan_lahnsteiner2]

Dear Seunghwan_Kwon,

Setting the Nose-Hover mass is described on the following wiki page:
wiki/index.php/SMASS

From the wiki: The Nosé-mass should be set such that the induced temperature fluctuation show approximately the same frequencies as the typical 'phonon'-frequencies for the specific system.

A low SMASS value leads to rapid fluctuations of the heat bath (temperature) and a high value of SMASS leads to slow fluctuations of the heat bath. If you need more information on the Nose-Hover thermostat I would recommend you to take a look in the book
Daan Frankel Understanding Molecular Simulation.
So if you know the phonon frequencies of your system I would recommend you to set SMASS such that the temperature fluctuates
according to your lowest phonon frequency. You have to adjust SMASS by a trial and error procedure.

I hope this answers your question


All the best Jonathan