[gmx-users] Can't observe ion separation under the influence of electric field

[Man Hoang Viet]

Hi Live King,

I think the temperature coupling strongly effect on your observation since
the velocity of atoms (ions) will be rescale after every tau-t value. It
means the movement direction of ion under EEF will be mostly canceled for
every tau-t (ps). If you set tau-t to be larger than default value (.i.e
10, 100 or 1000) you can see the effect of EEF on your system.

Viet Man,
University of Pittsburgh School of Pharmacy

>
> Dear David and Harutyun,
>
> Thanks for your reply. Density plots were the first check I did. They show
> a very minor difference. I imagined that the electric field is not
> sufficient, however,  presence of high electric field ( > 1 V or 1000mV)*
> disrupts the membrane and my simulation crashes*. Is there any solution to
> this problem? I will have a look at compel.
>
> On Thu, Feb 13, 2020 at 6:21 PM Harutyun Sahakyan <h_sahakyan at mb.sci.am>
> wrote:
>
>> You can use compel and build a double bilayer system to separate ions.
>> However, if you separate positive and negative ions, you will have very
>> high membrane potential, event two ions difference can raise potential
>> about 200 mV. In your case, I suppose 500 mV is not enough to separate
>> all
>> ions in a PBC box.
>>
>> https://www.mpibpc.mpg.de/grubmueller/compel
>>
>>
>>
>> Harutyun Sahakyan
>> Senior Assistant at Laboratory of
>> Computational Modeling of Biological Processes,
>> Institute of Molecular Biology, NAS, Armenia, Yerevan
>> Tel: +374 93 323990
>> h_sahakyan at mb.sci.am
>> sahakyanhk at gmail.com
>>
>> February 13, 2020 2:07 AM, "Live King" <vikasdubey1008 at gmail.com> wrote:
>>
>> > Dear all,
>> >
>> > I am running a simple test case with a lipid bilayer (DMPC), water,
>> and
>> > ions (150mM KCL) under the influence of a constant electric field (
>> 300mV,
>> > 500mV, and 700mV). *I expected positive and negative ions to separate
>> in
>> > the presence of an external electric field*. However, I am not
>> observing
>> > such behavior despite running the all-atom simulation for 400ns in any
>> > case. I am using gromacs 2018.2, my mdp file is standard charmm-gui
>> > production with electric field option.
>> >
>> > for example 500mV electric field in Z-direction :
>> > electric-field-z = 0.060 0 0 0 ; *500**mv = 8.25nm* 0.060 =
>> > 500 mV *
>> >
>> > Are there any settings to run the electric field simulation that I
>> miss?
>> > Any help will be greatly appreciated.
>> >
>> > Thank you,