MaginnGroup/Cassandra

Insertion of a huge number of molecules in a GCMC simulation

Closed this issue · 5 comments

camjjr commented

Hi everyone,

I am facing some troubles with GCMC in Cassandra and I want to check if it is a bug in the software.
First, I am trying to simulate the insertion of gas molecules (n2/co2) inside different CNTs. In this case, the reservoir is composed of 85% of N2 and 15% of CO2 (the chemical potentials were obtained using the widom method). The problem is that I would expect as I increase the radius of the CNT, the proportion of N2/CO2 inside the CNT would reach the value of the reservoir. However, for all CNT that I tested in Cassandra, only N2 is inside the CNTs at the end of the simulations. I tested the same simulations in lammps and, with this software, I see the gas inside the CNT reaching the limit of bulk gas. Is there a problem with my input? I will attach one that I used here. The bulk simulation in Cassandra seems fine.
with_restriction.inp.txt
without_restriction.inp.txt

Second, I am also trying to obtain the structure of the gas outside the CNTs. So, I performed a GCMC simulation without the restriction of gas inside the CNT. In this case, the gas seems to condensate outside the CNT in the whole simulation box, which seems unphysical. Instead of gas, the simulations seem to be a liquid. I also tested it in lammps and it shows no condensation. Can you check my inputs?

Thank you

WhatsApp Image 2022-11-07 at 14 54 05

romanielo commented
camjjr commented

Hi Lucienne,

Thank you for taking the time to read my input files. I will enumerate the problems that you mentioned.

  1. Should I create a fragment file for the cnt? The documention says "Since the solid structure will be rigid, no bond distances, angle parameters, or dihedral parameters are needed in the MCF. The library_setup.py script will not create a fragment library since the MCF lists zero fragments." Link: https://cassandra-mc.readthedocs.io/en/latest/guides/solids.html

  2. About the cross interactions.
    I am using the same label for the atoms in the mcf file in this section. For example, in n2.mcf the label of N is N_s1, and in co2.mcf, the label of C_s1. Is this wrong? Does Cassandra use different labels from the mcf file?

  3. About the chemical potential.

I corrected it in my input files. I used none as described in the manual. Now, there is co2 molecules inside the CNT. However, the number of molecules from cassandra is still not matching my results from LAMMPS

Can you help me?

Carlos

romanielo commented
camjjr commented

Hi Lucienne,

I made a new directory and run the library script before running Cassandra. The results are the same as the previous simulation. However, I noticed the program I used to compare Cassandra's results with those from Lammps had a problem. I corrected it and now the simulations are matching.

Thank you very much,

Carlos

romanielo commented