[gmx-users] Surface tension Coupling

[gromacs user]

I want to impose a surface tension of ~ - 100mN/m on a
bilayer. As I understand it, there are two ways to do
it in the mdp file. I can use  

(A)
pcoupltype = semiisotropic
tau_p              1 1
compressibility    4.5e-5 4.5e-5
ref_p              -100       1

if the system size in the z-direction is ~ 10nm, then
this should give a surface tension of (Pz-Plat)*Lz ==
(1+100)bar*10 nm ~ 10^3 bar nm = 10^3*10^5 N/m2 *10^-9
m
=100mN/m assuming the z-dimension Lz remains
approximately 10 nm throughout the window of
calculation.

The other way to do it would be

(B)
pcoupltype = surface-tension
tau_p              1 1
compressibility    4.5e-5 4.5e-5
ref_p              -1000 1

and directly specifying the surface tension.

Is my understanding correct? Which is the better way
to do it? As I see it, method B imposes a surface
tension that would automatically take the
instantaneous z-dimension into the calculations. This
should presumably give an accurate surface-tension
coupling. Whereas, method A imposes a pressure
coupling and we can extract a surface tension from it
, which would be accurate if the z-dimension does not
fluctuate by much. Am I correct?

Is there a right way to do it, or are both methods
valid?

Thanks.
Jessica Bloom


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com