Create a conda env and install the dependencies
conda create -n mof-topology
conda install python=3.8 -y
conda install -c conda-forge ipykernel -y
python -m ipykernel install --user --name=mof-topology
pip install git+https://github.com/Sangwon91/PORMAKE.git
This version allows the partial charges assignment on the building nodes.
pip install lammps-interface
pip install lammpsio
Originally cif2lammps and ToBaCCo are used for MOF generation (see folder simulation_utilities); for the automation purposes, the combination of pormake and lammps-interface seems to work better.
LAMMPS version 22 Jun 2022 (LAMMPS)
wget https://github.com/lammps/lammps/archive/refs/tags/stable_23Jun2022.tar.gz
tar -xvf stable_23Jun2022.tar.gz
The following additional packages are required:
cd lammps-stable_23Jun2022/src
make yes-molecule
make yes-extra-fix
make yes-extra-compute
make yes-fep
make yes-body
make yes-kspace
make yes-manybody
make yes-rigid
make yes-extra-molecule
Replace the fix_ti_spring.cpp in the LAMMPS/src folder with the provided one, which is slightly modified for the final NES step calculation, and compile.
cp ../../fix_ti_spring.cpp .
make mpi -j8 or make seriel -j8
export LAMMPS_PATH=`pwd`/lmp_mpi
There are three tutorial notebooks in the example folder showing the procedures for building MOF polymorphs and computing free energies. example1, example2, and example3 include a ZJU-28 series MOF, a ZIF series MOF, and an Fe₄S₄-BDT series MOF, respectively. You can adapt the notebooks or use the command lines in the next section to run the pipeline.
The simulation_utilities folder contains all the scripts for running free energy calculations on Fe₄S₄-BDT MOFs with different cations using the ToBaCCo codes and cif2lammps. This folder also includes guidelines for running simulations using this set of tools.
The databank_iron-sulfur-MOFs folder includes all the relaxed structures of the iron-sulfur MOFs studied in the paper, along with computed free energies, potential energies, and geometric properties. The PDF files are stored in H5 files that can be accessed using h5py.
The gen command is used to generate a MOF polymorph, which requires a linker and a metal node in XYZ format. You can generate bare frameworks or specify a small molecule to add to an ionic MOF. If the name of the small molecule is available in the small_molecule folder (DMA, TMA, TEA, TPA, TPP, MPA, MNP), use those files. Otherwise, you need to provide a small molecule in CIF format with atomic charges.
The run_equi command generates input files for small molecule deposition and equilibration. After equilibration, a CIF file of the equilibrated structure is provided, along with the potential energy at 300 K and 0 K for preliminary screening.
Example:
python main.py gen --node 4c_In --linker 3c_BTB --topos pto --mol DMA
python main.py run_equi --mof pto-4c_In-3c_BTB --mol DMA --n_mol 6 --nvt --npt --equi_time 50000 --temp 300 --pressure 0.0
This is a checkpoint where you can stop and use the equilibrated structures and potential energies to rank the thermodynamic stability of the MOFs.
The equilibrated structures can be used for subsequent free energy calculations. The run_fe function is used to generate the input files for calculating the free energies of the MOFs. If a small molecule is used, it is required to specify the atom type of the molecules from the LAMMPS emin data file.
Example:
python main.py run_fe --mof pto-4c_In-3c_BTB --mol DMA --center 7 --nproc 8
This will create a list of input files for running free energy simulations and create a folder for storing the output data.
The codes for integrating the free energy after the calculation is available in example 1.