Example 002

About

This script configures and runs a custom NEB workflow that borrows features of path finding, endpoint relaxations, and more.

Key Info
Contributor	Becca Radomsky
Github User	@becca9835
Last updated	2023.12.30
Simmate Version	v0.13.2

Prerequisites

• Simmate configured with default settings will work!
• The NEB settings below use CI-NEB, which requires some of the VTST-tools installed. You modify these settings to just use a normal VASP NEB run if you'd like (see the comment Modify these if you don't want CI-NEB).

The script

Info

Run this script where you'd like the VASP calculations to run!

The script can be called with something like python myscript.py and this line can be on your desktop, inside a SLURM job, or whereever!

# -*- coding: utf-8 -*-

from simmate.apps.vasp.inputs import PBE_POTCAR_MAPPINGS
from simmate.apps.vasp.workflows.base import VaspWorkflow
from simmate.apps.vasp.workflows.diffusion import (
    NebAllPathsWorkflow,
    SinglePathWorkflow,
    VaspNebFromImagesWorkflow,
)

# -----------------------------------------------------------------------------

# BULK UNITCELL RELAXATION


class Relaxation__Vasp__WarrenLab(VaspWorkflow):
    functional = "PBE"
    potcar_mappings = PBE_POTCAR_MAPPINGS
    _incar = dict(
        ALGO="Fast",
        EDIFF=1e-06,
        ENCUT=520,
        IBRION=2,
        ICHARG=1,
        ISIF=3,
        ISPIN=2,
        ISYM=0,
        IVDW=12,
        LORBIT=11,
        LREAL="Auto",
        LWAVE=False,
        NELM=200,
        NELMIN=4,
        NSW=99,
        PREC="Accurate",
        ISMEAR=0,
        SIGMA=0.05,
        KSPACING=0.35,
        LMAXMIX=4,
    )


# -----------------------------------------------------------------------------

# BULK UNITCELL STATIC ENERGY


class StaticEnergy__Vasp__WarrenLab(VaspWorkflow):
    functional = "PBE"
    potcar_mappings = PBE_POTCAR_MAPPINGS
    _incar = dict(
        ALGO="Fast",
        EDIFF=1e-06,
        ENCUT=520,
        IBRION=-1,
        ICHARG=1,
        ISIF=3,
        ISPIN=2,
        ISYM=0,
        IVDW=12,
        LORBIT=11,
        LREAL="Auto",
        LWAVE=False,
        NELM=200,
        NELMIN=4,
        NSW=0,
        PREC="Accurate",
        ISMEAR=0,
        SIGMA=0.05,
        KSPACING=0.35,
        LMAXMIX=4,
    )


# -----------------------------------------------------------------------------

# ENDPOINT SUPERCELL RELAXATIONS


class Relaxation__Vasp__WarrenLabNebEndpoint(VaspWorkflow):
    functional = "PBE"
    potcar_mappings = PBE_POTCAR_MAPPINGS
    _incar = dict(
        ALGO="Fast",
        EDIFF=5e-05,
        EDIFFG=-0.01,
        ENCUT=520,
        IBRION=2,
        ICHARG=1,
        ISIF=2,
        ISPIN=2,
        ISYM=0,
        IVDW=12,
        LORBIT=11,
        LREAL="Auto",
        LWAVE=False,
        LCHARG=False,
        NELM=200,
        NSW=99,
        PREC="Accurate",
        ISMEAR=0,
        SIGMA=0.05,
        KSPACING=0.4,
        LMAXMIX=4,
    )


# -----------------------------------------------------------------------------

# ENDPOINT SUPERCELL STATIC ENERGY


class StaticEnergy__Vasp__WarrenLabNebEndpoint(VaspWorkflow):
    functional = "PBE"
    potcar_mappings = PBE_POTCAR_MAPPINGS
    _incar = dict(
        ALGO="Fast",
        EDIFF=5e-05,
        EDIFFG=-0.01,
        ENCUT=520,
        IBRION=-1,
        ICHARG=1,
        ISIF=2,
        ISPIN=2,
        ISYM=0,
        IVDW=12,
        LORBIT=11,
        LREAL="Auto",
        LWAVE=False,
        LCHARG=False,
        NELM=200,
        NSW=0,
        PREC="Accurate",
        ISMEAR=0,
        SIGMA=0.05,
        KSPACING=0.4,
        LMAXMIX=4,
    )


# -----------------------------------------------------------------------------

# NEB FROM IMAGES


class Diffusion__Vasp__WarrenLabCiNebFromImages(VaspNebFromImagesWorkflow):
    functional = "PBE"
    potcar_mappings = PBE_POTCAR_MAPPINGS
    _incar = dict(
        ALGO="Fast",
        EDIFF=5e-05,
        EDIFFG=-0.01,
        ENCUT=520,
        IBRION=3,
        ICHARG=1,
        ISIF=2,
        ISPIN=2,
        ISYM=0,
        IVDW=12,
        LORBIT=11,
        LREAL="Auto",
        LWAVE=False,
        LCHARG=False,
        NELM=200,
        NSW=99,
        PREC="Accurate",
        ISMEAR=0,
        SIGMA=0.05,
        KSPACING=0.4,
        LMAXMIX=4,
        NIMAGES__auto=True,  # set automatically by simmate
        # Modify these if you don't want CI-NEB
        LCLIMB=True,
        SPRING=-5,
        POTIM=0,
        IOPT=1,
    )


# -----------------------------------------------------------------------------
# The two sections below use Simmate to piece together our individual
# VASP calculations above.
# -----------------------------------------------------------------------------

# SINGLE PATH NEB


class Diffusion__Vasp__WarrenLabNebSinglePath(SinglePathWorkflow):
    endpoint_relaxation_workflow = Relaxation__Vasp__WarrenLabNebEndpoint
    endpoint_energy_workflow = StaticEnergy__Vasp__WarrenLabNebEndpoint
    from_images_workflow = Diffusion__Vasp__WarrenLabCiNebFromImages


# -----------------------------------------------------------------------------

# ALL PATHS NEB


class Diffusion__Vasp__NebAllPathsWarrenLab(NebAllPathsWorkflow):
    bulk_relaxation_workflow = Relaxation__Vasp__WarrenLab
    bulk_static_energy_workflow = StaticEnergy__Vasp__WarrenLab
    single_path_workflow = Diffusion__Vasp__WarrenLabNebSinglePath


# -----------------------------------------------------------------------------
# Now that we have our new & custom NEB workflow, we can run that
# full workflow analysis. Here, we just run it locally and one vasp calc
# at a time.
# -----------------------------------------------------------------------------

# now run the workflow!
result = Diffusion__Vasp__NebAllPathsWarrenLab.run(
    structure="example.cif",
    migrating_specie="Li",
    command="mpirun -n 10 vasp_std > vasp.out",  # make sure -n is divisible by nimages!
    # Then any extra optional settings below.
    # These parameters are automatically available thanks
    # to the configuration we did above.
    relax_bulk=True,
    relax_endpoints=True,
    nimages=5,
    min_supercell_atoms=80,
    max_supercell_atoms=240,
    min_supercell_vector_lengths=10,
    max_path_length=5,
    percolation_mode=">1d",
    vacancy_mode=True,
)