Fix deform with non periodic boundary conditions

Hello Dr Steve, Dr Axel and everyone.

Can someone please guide me on how to use the fix deform with non-periodic boundary conditions? I’ve already gone through past forums and the help documentation, maybe I am missing something.

I have a graphene-metal nanocomposite and I am trying to peel graphene off iron. My current boundary conditions are non periodic in Z direction and I am using fix deform followed by nve for displacement controlled loading.

Thanks in advance,
Jane

Hello Dr Steve, Dr Axel and everyone.

Can someone please guide me on how to use the fix deform with non-periodic
boundary conditions? I've already gone through past forums and the help
documentation, maybe I am missing something.

I have a graphene-metal nanocomposite and I am trying to peel graphene off
iron. My current boundary conditions are non periodic in Z direction and I
am using fix deform followed by nve for displacement controlled loading.

there are not enough details here and it is not clear what exactly
your problem is and what you have tried so far.
the documentation is very straightforward as to how fix deform can be
used with respect to periodic/non-periodic boundaries.

axel.

Hello Dr Axel,

Thank you for your response.
I am trying to simulate the pull out test of graphene from iron. I wish to plot the force displacement curve and obtain the interfacial properties of graphene and iron. Currently, my code has periodic boundaries in X,Y and Z.

Here is my code so far:
https://we.tl/htnnz9mWAY

Thanks!

Hello Dr Axel,

Thank you for your response.
I am trying to simulate the pull out test of graphene from iron. I wish to
plot the force displacement curve and obtain the interfacial properties of
graphene and iron. Currently, my code has periodic boundaries in X,Y and Z.

Here is my code so far:
https://we.tl/htnnz9mWAY

this is a rather complex and not very self explanatory input.
and again, you are still not really explaining what your problem is or
what specifically you need help with.
i don't quite see, how fix deform should be used for what you are doing.

it almost looks like you want to compute the potential of mean force
for moving your graphene away from the surface.
there are well established ways to do that, e.g. umbrella sampling or
steered MD or metadynamics, which all are possible with LAMMPS.
if you just want to forcibly move the graphene sheet away, you might
also use fix move.
i don't quite see, how with fix deform you'd get a consistent
displacement vs. force graph in the first place.

that said, i also don't quite see, why you would have to have
non-periodic boundaries in z, if you make certain that your box in z
is large enough, so there are no interactions across the periodic
boundaries, which you control with the cutoff for the lj/cut
potential.

axel.

There are a lot of obvious problems with this script. For example, fix deform with remap v and a free surface will cause the box to expand in z but will have no effect on the atom structure. You also seem to be defining region fix containing Fe atoms with the graphene lattice. Clearly you have bitten off more than you can chew.

Dear all,

Thank you Axel & Aidan for your helpful inputs. I tried Dr Axel’s suggestion of increasing the size of the sim box in Z dir and continue using the ppp boundary conditions.

In both cases, fix move or fix deform, I visualize the following:
https://we.tl/o4mmR5vVr8

In either case, after correcting the group definitions, only the graphene layer is moving up and not truly getting “peeled off” from iron substrate. I think its because of what Aidan said, the peel strength will be sensitive to Fe/C interaction. However, I cannot think of any other alternative. I obtained the values from literature for LJ interaction.

Any advice?

Thanks a lot!
Jane

Dear all,

Thank you Axel & Aidan for your helpful inputs. I tried Dr Axel's
suggestion of increasing the size of the sim box in Z dir and continue
using the ppp boundary conditions.

In both cases, fix move or fix deform, I visualize the following:
https://we.tl/o4mmR5vVr8

In either case, after correcting the group definitions, only the graphene
layer is moving up and not truly getting "peeled off" from iron substrate. I
think its because of what Aidan said, the peel strength will be sensitive to
Fe/C interaction. However, I cannot think of any other alternative. I
obtained the values from literature for LJ interaction.

Any advice?

i disagree. if you move the whole layer, it will move as a whole.
both, fix move and fix deform will pay no attention at all to forces
and just move particles as prescribed. in other words, you're
obviously getting what you asked LAMMPS to do. if this is not what you
want to do, then you'd have to design you input differently. LAMMPS is
a computer program and computers are extremely narrowminded: they do
exactly what they get told.

so, if you want to rip of the layer like a band aid, then you'll have
to do the same thing: grab a *piece* of it and rip it off (i.e. define
a group and then pull that group away). this can be done in different
ways and depends on what kind of energy/force exactly you are after
and how it is defined.

axel.

Also, when it comes to “peeling,” size matters. What looks like peeling on a scale of mm’s looks like pure tension on a scale of nm’s. I really think you need to take a few steps back, understand MD better, and also read the literature to see what other people have been able to do in this area.

Thank you Aidan and Dr Axel for your extended inputs. I will work on it and as you both suggested, I will take a few steps back and get a better understanding and rework on my code.

Thanks again!