Added tutorial 8

[simongravelle]

The tutorial follows:

• 1. mixing.lmp : placing nylon and CNT in the same box and equilibrating it in NPT
• 2. reacting.lmp : making the nylon react and recording the amount of water molecules in time

[simongravelle]

@jrgissing I have written a base for the tutorial 8. I still need to generate a snapshot showing the system, as well as a graph showing the evolution of the reaction with time. I hope to do it tomorrow.

[simongravelle]

On the side note, with the last stable version of LAMMPS, I got the error:

• ERROR: Invalid range string: c2

I did not look into yet, but may be the origine of the error is obvious to you?

[jrgissing]

@simongravelle I am unable to reproduce the error using the LAMMPS 'stable' branch (29 Aug 2024), with either the current mixing.lmp or reacting.lmp input files in this PR

[jrgissing]

by the way, this tutorial is looking great. I am taking a closer look and will probably have some suggestions. For example, do you think it would be better if we used a single periodic CNT? I think I added an example of this in examples/replicate that I can adapt

[simongravelle]

by the way, this tutorial is looking great. I am taking a closer look and will probably have some suggestions. For example, do you think it would be better if we used a single periodic CNT? I think I added an example of this in examples/replicate that I can adapt

My idea with having a dispersed suspension of CNT was that we could then perform tests on the final polymer/CNT mixture (tensile or other), or simply make a measurement of the adsorption of the polymer and the water within the CNTs. With a single CNT that would not be possible anymore, and also the entire equilibrating would have to be made different due to the anisotropy of the system (which is not a bad thing given that its tutorial number 8, we can do complicated stuffs) What would be the advantage to have a single periodic CNT? What was the idea behind it?

[simongravelle]

@simongravelle I am unable to reproduce the error using the LAMMPS 'stable' branch (29 Aug 2024), with either the current mixing.lmp or reacting.lmp input files in this PR

Good to know, I will try again.

[jrgissing]

What would be the advantage to have a single periodic CNT? What was the idea behind it?

Here are my thoughts: These CNTs are too short to be physically realistic (even ultra-short CNTs are tens of nms). Also, I am used to seeing significant degrees of polymerization. Without that, I do not think the usefulness of the method gets across. I think the high packing fraction of the CNTs, small system, and long monomer units are mostly to blame (along with short run time, which we can't really get around for a tutorial). So, I am thinking that a single CNT with higher percentage of smaller monomers might do the trick. For example, we could try polystyrene.

For analysis, perhaps in a separate tutorial or add-on section, we could look at inclusion energy by comparing energy of composite with sum of energies of neat polymer and lone CNT.

I can take a first pass at these updates, if you agree with them. Please let me know what you think.

[simongravelle]

These CNTs are too short to be physically realistic

I don't think that the realism of the system should be a big concern here, since the goal is for the users to learn something new about LAMMPS compared to the previous tutorial.

• With a single periodic CNT, the system is more difficult to equilibrate due to the anisotropy of the system, which is may be completely fine if that's something we think is meaningful for the user to learn, and would indeed be a novelty compared to the previous tutorials.
• With small multiple CNTs, the system is easier to equilibrate but may offer more possibilities for the analysis.

I am fine with both choices.

@akohlmey Do you have a strong argument against one or the other?

[akohlmey]

@akohlmey Do you have a strong argument against one or the other?

Not really. If explained properly, I see a small benefit in the option to go with a single periodic CNT and discussing the benefits and limitations of such a setup.

[simongravelle]

OK, OK, let's do a single CNT then. :)

[jrgissing]

@simongravelle do you want to merge this branch, and I will take a pass at a single periodic CNT and using polystyrene as matrix?

[simongravelle]

@simongravelle do you want to merge this branch, and I will take a pass at a single periodic CNT and using polystyrene as matrix?

Its up to you. I've let it open in case you want to work from there, but it can be merged.