[gmx-users] Dielectric Constant

[Mark Abraham]

Hi,


On Tue, Jun 21, 2016 at 5:22 PM Life Sciences Inc <
contact.lifesciences.inc at gmail.com> wrote:

> Hi
>
> I am getting the value of Epsilon from the output of gmx dipoles as
> approximately close to 60, in my mdp file I used the default parameters for
> epsilon-r (1) and for epsilon-rf(0), the output as below
>

These are totally different things - the resulting epsilon of your whole
simulation system, vs the effective dielectric of your model physics that
produces the simulation.

Clearly you haven't read the .mdp descriptions


> Select a group: 12
> Selected 12: 'Water'
> There are 4531 molecules in the selection
> Last frame      50000 time 100000.000
> Average volume over run is 149.579
>
> Dipole moment (Debye)
>  ---------------------
> Average  =   2.2740  Std. Dev. =   0.0001  Error =   0.0000
>
> The following averages for the complete trajectory have been calculated:
>
> Total < M_x > = 3.03347 Debye
> Total < M_y > = -12.361 Debye
> Total < M_z > = -19.1034 Debye
>
> Total < M_x^2 > = 28871.2 Debye^2
> Total < M_y^2 > = 29326.8 Debye^2
> Total < M_z^2 > = 29241.6 Debye^2
>
> Total < |M|^2 > = 87439.6 Debye^2
> Total |< M >|^2 = 526.935 Debye^2
>
> < |M|^2 > - |< M >|^2 = 86912.7 Debye^2
>
> Finite system Kirkwood g factor G_k = 3.7096
>
> Infinite system Kirkwood g factor g_k = 2.49376
>
> Epsilon = 59.7615
>
>
> But I want to calculate the dielectric constant with the imaginary surface
> around my system or it can also be said as with PBC boundaries, for which
> there is an option in gromacs for mdp file which is epsilon-surface(0), but
> whenever I am using this option with other than default value with
> epsilon-r = 1 or epsilon-rf = 0 or epsilon-r = 1 or epsilon-rf = 1,  I am
> getting LINCs error. For electrostatistics I am using PME.


Hmm, if you'd done what I already asked and checked the .mdp documentation,
then you'd know that varying epsilon-rf is useless with PME.
epsilon-surface could well be buggy (since it is not automatically tested),
but AFAICR it doesn't affect the forces, so your LINCS errors must have
some other origin - eg whatever ad hoc changes you are making to your model
physics.


> As I have used
> SPC water model I tried to use the value of epsilon-surface = 70, and also
> used 60 so that these values may correct the lincs error.


Why are those sensible values? Have you read the .mdp documentation for
what this means?

Mark


> I have also made
> my input molecule as a whole using gromacs. Simulation runs if I use
> epsilon-surface = 0 but other than this simulation i not running. I am
> using gromacs version 5.
>
> I am not getting where is the problem and how it can be solved.
>
> Kindly help.
>
> On Thu, Jun 16, 2016 at 10:05 PM, Mark Abraham <mark.j.abraham at gmail.com>
> wrote:
>
> > Hi,
> >
> > These are all documented in the mdp sections. What isn't clear?
> >
> > Mark
> >
> > On Thu, 16 Jun 2016 18:06 Life Sciences Inc <
> > contact.lifesciences.inc at gmail.com> wrote:
> >
> > > Hi all
> > >
> > > Exactly in Gromacs what and where is Dielectric constant option , If
> for
> > > example I have to run simulations to study behavior of water around
> > > molecules, and I am using experimental value of SPC/E water model as 71
> > for
> > > Dielectric constant. But in my input file for gromacs where I will put
> > > Dielectric constant is it epsilon-surface, or epsilon-r or
> epsisilon-rf,
> > it
> > > is little confusing. Kindly someone elaborate .
> > >
> > > Thank you
> > > --
> > > Gromacs Users mailing list
> > >
> > > * Please search the archive at
> > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > > posting!
> > >
> > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > >
> > > * For (un)subscribe requests visit
> > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > > send a mail to gmx-users-request at gromacs.org.
> > >
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>