Pressure coupling

User discussions
1

Dear all,

I have used deform option as in the following:
deform = 0 0 0 7.0 0 0

But, for pressure coupling I am not sure if I have written it correctly.
pcoupl = Parrinello-Rahman
pcoupltype = anisotropic
tau_p = 2.0
ref_p = 0.0 0.0 0.0 1.0 0.0 0.0
compressibility = 0.0 0.0 0.0 46e-5 0.0 0.0

Could you please tell me if it is correct?
And, if not, could you please direct me to write the correct one?

I appreciate your assistance in advance.
Best

2

Hi MOH4,

it’s correct syntax-wise: have you tried to run it? Does it work?

Best,
Michele

3

Hi Michele,

Not yet. I am using another option along with this.
But, I was not sure if what I had written (particularly zero values) were correct.

Could you please tell me what the zero values of compressibility mean?

Best

4

The definition of isotropic compressibility is the ratio between the derivative of volume w.r.t. pressure and total volume.
In a non-isotropic case you may have a different compressibility value in each direction (or rather: for each component of the pressure tensor), however the physical interpreation kinda remains the same (‘how much’ the specific deformation is sensitive to pressure differences).
Saying that compressibility is zero means that the box does not get deformed in that direction. For example, if you have a cubic box and compressibilty is non-zero only along ZZ, then the box would only be strained/compressed along z.

5

Many thanks for your informative response.

In this regard, if I write deform = 0 0 0 7.0 0 0, it means the box will be deformed along angle between xy directions as it is LyXY. Is it correct?

6

Hi,

the off-diagon terms refer to the vectors b(x) c(x) c(y) in case of a triclinic box; please have a look at Periodic boundary conditions — GROMACS 2021 documentation and Molecular dynamics parameters (.mdp options) — GROMACS 2021 documentation.
I personally never utilized off-diagonal deformations, so I am sorry but that is the only indication I am able to provide.

Cheers,
Michele

7

Hi Michele,
Many thanks for your all responses.
They are really helpful.

For the last question, could you please tell me if you know the range of value that I can give to deform?
I mean what is the maximum value for deform input components?
In the other words, is this possible to write e.g. deform = 0 0 0 77.0 0 0 or values more than this?

I apologize for asking many questions.
Many thanks in advance,
Best