Running constant surface tension simulations

User discussions
1

GROMACS version: 2020.5
GROMACS modification: No

Hello all,
I am planning to simulate a trilayer system. It is arranged as lipid-monolayer/organic-solvent/lipid-monolayer, with water and ions surrounding it. I am trying to find out how to run at constant surface tension values between 1-10 mN/m.

I saw this helpful discussion post that helped me to understand that I should use pcoupletype=surface-tension. However, I want to make sure I understand what the ref-p setting should be. It says to set ref-p as “the reference surface tension times the number of surfaces bar nm”. Is this the number of types of interfaces in the system (i.e. 2 = water-lipid and lipid-solvent), or the total number of interfaces (i.e. 4 = water/lipid, lipid/solvent, solvent/lipid, lipid/water)?

I also plan to run with the NP(gamma)T (gamma for surface tension) ensemble, since apart from testing surface tension effects, I want to run at atmospheric pressure and room temperature. I would like to hear opinions about if this is appropriate, or if there are any considerations/issues I should be careful of that could warrant the use of a different ensemble.

Lastly, I have attached a sample of the mdp settings I plan to use. I appreciate any help, thanks!

Thanks,
Toyin

integrator = md
dt = 0.002
nsteps = 500000
nstxout = 50000
nstvout = 50000
nstfout = 50000
nstcalcenergy = 100
nstenergy = 1000
nstlog = 1000
;
cutoff-scheme = Verlet
nstlist = 20
rlist = 1.2
vdwtype = Cut-off
vdw-modifier = Force-switch
rvdw_switch = 1.0
rvdw = 1.2
coulombtype = PME
rcoulomb = 1.2
;
tcoupl = Nose-Hoover
tc_grps = MEMB SOLV
tau_t = 1.0 1.0
ref_t = 298.15 298.15
;
pcoupl = Parrinello-Rahman
pcoupltype = surface-tension
tau_p = 5.0
compressibility = 4.5e-5 4.5e-5
ref_p = 10.0 1.0
refcoord_scaling = com
;
constraints = h-bonds
constraint_algorithm = LINCS
continuation = yes
;
nstcomm = 100
comm_mode = linear
comm_grps = TRILAYER SOLV

2

The barostat couples to the target total surface tension given by ref-p. mdrun can’t know how many interfaces you have in the system, it can only measure the sum of the tension of all interfaces perpendicular to the z-axis.

3

Alright, this clears things up for me. Given the experiments I am going off of, then I need to consider the surface tensions as being applied to the whole system. Thank you!