How to make AIMD simulation for proton transition (NVT)

Rolf David rolf.d... at
Thu Jun 25 10:02:23 UTC 2015


Just some random thoughts about the temperature,

Regarding your input geometry you have,
One proton alone (!)
One water molecule
two molecules which I'm guessing is your "membrane"

You should transform your proton as hydronium or zungel cation especially 
if your membrane is hydrated. 
Also you should solvate your membrane along the axis you want:

Along X/Y : membrane only (like it's now, but maybe add some water)
Along Z: water-membrane-water-proton-water

Maybe this will work better (solvated condition and not void) or not but 
that is what I would try next

Also MD can't overcome more than KbT barrier (over short time) so if your 
proton doesn't move a lot (just vibration motion) or not in the "path you 
want", try to lean towards biased MD.

Kind regards,

On Wednesday, June 24, 2015 at 8:23:34 PM UTC+2, Linda Sundarti wrote:
> Dear cp2k expert,
> I'm newbi user cp2k, I get the problem with my MD simulation. 
> I have run MD simulation to get proton conductivity or proton transition 
> in hydrated membrane, but I get some problems which are the proton can not 
> move (translation motion was not occur), just vibration motion that 
> occured. On the other hand, the tempertaur becomes increase high, whereas 
> I used ensemble NVT. I make condition in the constant temperature, number 
> of particle and volume. 
> From my problem, is there any mistake in my input file so I don't get the 
> transition of proton and the constant of temperature?
>  Is there specific keyword in input file in MD simulation to get it? 
> I hope anybody can help me to make it clear and better. My goal simulation 
> is make transition of proton and the constant of temperature, number of 
> particle and volume. I attach my input and output file below.
> Any suggestions are appreciate. Thank you.
> Regards,
> Linda
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the CP2K-user mailing list