I am trying to construct a semi-crystalline polyethylene system using non-periodic boundary conditions in the x and y directions (periodic only in z). The boundary conditions are specified correctly in the simulation setup.
However, I have observed that during EMC moves (e.g., displacement, reptation, rebridge), the non-periodic boundary conditions in x and y do not seem to be enforced. Molecules appear to move across the x–y boundaries as if periodic boundary conditions were still applied.
Is it possible to restrict EMC moves (such as displacement, reptation, rebridge, etc.) so that they are applied only to molecules or segments located inside a specified region?
If this is not directly supported, are there any recommended workarounds (e.g., selection rules, constraints, or staged simulations) to effectively achieve region-limited EMC sampling?
First off: I see that the build file you use is taken from the provided example in $EMC_ROOT/examples/build/polymer/polyethylene/, which allows for building semicrystalline polyethylene configurations according to the formalism as proposed by Rutledge et al. This formalism assumes 3D periodicity in order to be technically executable due to the existence of loops and bridges.
Having said that, uncoupling periodicity as you did when building the crystal does not influence the periodicity of the simulation. It merely cuts whatever bonds in the crystal cross the x- and y-boundaries. To obtain your objective, I would suggest building and equilibrating a 3D periodic structure. Once equilibrated, you could unwrap all polymers and enlarge your simulation box in both y and z directions. You most likely will see some bridges and/or loops to have long crazy bonds. Cut these bonds and replace their ends with a ch3 type. The result would come closest to what I think you are trying to accomplish.
While waiting for your reply, I also tested a possible solution on my own.
Since regions that are fixed do not participate in Monte Carlo (MC) moves, I added a dummy layer to the direction where I wanted non-periodic boundary condition and kept it fixed during the MC moves.
After the MC move, I removed the dummy layer to obtain a non-periodic semi-crystalline PE.
I would appreciate it if you could share your opinion on the physical validity of the structure generated using this approach.