Scribd Logo
Upload

Welcome to Scribd!

  • 59 views

    Casm Cli Reference

    Uploaded by

    Khaled Badawy
    This document provides an example workflow for specifying a crystal structure, constructing configurations, calculating energies, fitting an energy model, running Monte Carlo simulations, and constructing a phase diagram using the CASM software. Key steps include initializing a project, enumerating configurations, calculating DFT energies, fitting an energy model, running Monte Carlo simulations to obtain free energies, and analyzing the results to determine phase boundaries.

    Copyright:

    © All Rights Reserved
    Download as TXT, PDF, TXT or read online from Scribd

    Casm Cli Reference

    Uploaded by

    Khaled Badawy
    59 views4 pages
    This document provides an example workflow for specifying a crystal structure, constructing configurations, calculating energies, fitting an energy model, running Monte Carlo simulations, and constructing a phase diagram using the CASM software. Key steps include initializing a project, enumerating configurations, calculating DFT energies, fitting an energy model, running Monte Carlo simulations to obtain free energies, and analyzing the results to determine phase boundaries.

    Copyright

    © © All Rights Reserved

    Available Formats

    TXT, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides an example workflow for specifying a crystal structure, constructing configurations, calculating energies, fitting an energy model, running Monte Carlo simulations, and constructing a phase diagram using the CASM software. Key steps include initializing a project, enumerating configurations, calculating DFT energies, fitting an energy model, running Monte Carlo simulations to obtain free energies, and analyzing the results to determine phase boundaries.

    Copyright:

    © All Rights Reserved
    Download as TXT, PDF, TXT or read online from Scribd
    Download as txt, pdf, or txt
    59 views4 pages

    Casm Cli Reference

    Uploaded by

    Khaled Badawy
    This document provides an example workflow for specifying a crystal structure, constructing configurations, calculating energies, fitting an energy model, running Monte Carlo simulations, and constructing a phase diagram using the CASM software. Key steps include initializing a project, enumerating configurations, calculating DFT energies, fitting an energy model, running Monte Carlo simulations to obtain free energies, and analyzing the results to determine phase boundaries.

    Copyright:

    © All Rights Reserved
    Download as TXT, PDF, TXT or read online from Scribd
    Download as txt, pdf, or txt
    You are on page 1of 4

    # This is an example workflow going from

    # specifying a crystal structure to


    # constructing part of a phase diagram.
    #
    # It is assumed that CASM v1.2.0 is installed.
    #

    # in the ZrO example directory:


    root=`pwd`

    casm status -n

    ### Initialize project


    casm init

    ### Set composition axes


    casm status -n
    casm composition --calc
    casm composition --select 1

    ### Enumerate configurations


    casm status -n
    casm enum --method ScelEnum --max 4
    casm enum --method ConfigEnumAllOccupations --all

    ### Enable VESTA visualization,


    # Correct command depends on your system
    # On Mac, often: 'casm.view "open -a /Applications/VESTA/VESTA.app"'
    # On Linux: 'casm.view VESTA'
    casm settings --set-view-command 'casm.view "open -a
    /Applications/VESTA/VESTA.app"'
    casm view SCEL4_2_2_1_0_0_0/12

    ### Calculate DFT energies


    casm status -n
    casm select --set-on

    # set path to VASP pp by changing 'POTCAR_DIR_PATH' variable


    emacs training_data/settings/calctype.default/SPECIES

    # set VASP INCAR settings


    emacs training_data/settings/calctype.default/INCAR

    # set KPOINTS, either Auto, or


    # to be scaled to match density determined from
    # training_data/settings/calctype.default/POSCAR
    emacs training_data/settings/calctype.default/KPOINTS

    # set cluster submission parameters and convergence criteria


    emacs training_data/settings/calctype.default/calc.json

    # setup VASP jobs for manual checking


    # (optional, skip if no VASP pseudopotentials)
    casm calc --setup

    # Can't actually submit jobs, would use:


    # casm calc --submit
    # casm update
    # Instead, copy calculation results:
    cp -r _casm_with_properties/jsonDB/* .casm/jsonDB/
    casm status -d

    ### Set chemical reference


    casm ref --set-auto
    # Should output:
    # Global chemical reference:
    # Zr(1): -8.54698
    # O(0.5)Zr(0.5): -9.0907

    ### Specify basis set

    # Modify 'bspecs.json' as desired to specify orbits & basis functions


    # (for this demo, don't actually change)
    emacs basis_sets/bset.default/bspecs.json
    casm bset -u

    ### Select configurations


    casm select --set 'is_calculated' -o calculated.json

    ### Query properties


    casm query -c CALCULATED -k 'corr formation_energy' -o fitting_data.json

    ### ECI fitting

    # Create a directory for fitting work


    cd $root/fit_1

    # Create new eci setting


    casm settings --new-eci test

    # Select structures to fit with


    casm select --set 'lt(comp(a),0.695)' -o all_lt_0p7
    casm select -c all_lt_0p7 --set-off 'not(is_calculated)' -o train

    # Create a 'casm-learn' input file


    casm learn --exGeneticAlgorithm > fit_1_ga.json

    # For demo purposes, a quicker calculation will suffice.


    # Change "feature_selection"/"kwargs"/"evolve_params_kwargs"/"n_repetition": 5 or
    10
    emacs fit_1_ga.json

    # Run 'casm-learn'
    casm learn -s fit_1_ga.json

    # Check results
    casm learn -s fit_1_ga.json --hall
    casm learn -s fit_1_ga.json --checkhull --indiv 0
    python plot_hull.py

    # Select ECI to use


    casm learn -s fit_1_ga.json --select 0

    ### Run Monte Carlo

    # semi-grand canonical setup


    cd $root/mc/example_grand_canonical
    casm format --monte

    # edit metropolis_grand_canonical.json as desired

    # run
    casm monte -s metropolis_grand_canonical.json --verbosity verbose

    # check results
    cat results.json
    cat results.csv

    # very basic Python script showing how to plot MC data


    python plot.py

    ### Using provided Monte Carlo results to construct a phase diagram

    # For consistency, switch to 'default' ECI provided with demo


    casm settings --set-eci default

    # Low temperature expansion to get free energy reference states


    # lte1 setup
    cd $root/mc/example_lte1

    # edit lte1_grand_canonical.json as desired

    # run
    casm monte -s lte1_grand_canonical.json

    # check results
    cat results.csv
    cat results.json
    python plot.py

    # An example script for running a coarse grid of Monte Carlo calculations


    cd $root/ZrO/mc/grid
    python run.py

    # Use pre-computed monte carlo fine grid results


    cd $root/ZrO/mc/fine_grid
    tar -xzvf results.tar.gz
    python parse.py

    # The 'plot_vary_T.py' and 'plot_vary_xi.py' scripts "animate" a range of


    # Monte Carlo calculation paths by repeatedly plotting for a scan of parameters.
    # To see the next plot, close the current one.

    # check results for paths with varying T


    python plot_vary_T.py
    python plot_vary_T_all.py

    # check results for paths with varying parameteric chemical potential


    python plot_vary_xi.py
    python plot_vary_xi_all.py

    # free energy integration


    python free_energy.py

    # find phase boundary


    python plot_phi_vary_xi.py
    python plot_boundary_vary_xi.py

    You might also like