Nudged Elastic Band¶

NEBAnalyzer ¶

NEBAnalyzer(
    n_images: int = 5,
    spring_constant: float | list[float] = 0.1,
    climb: bool = False,
    remove_rotation_and_translation: bool = False,
    method: Literal[
        "aseneb",
        "improvedtangent",
        "eb",
        "spline",
        "string",
    ] = "improvedtangent",
    interpolate_lattices: bool = False,
    pbc: bool = True,
    autosort_tol: float = 0.5,
    end_amplitude: float = 1,
    calculator: BaseCalculator | None = None,
    neb_transformation: NEBTransformation | None = None,
)

A class used to perform the Nudged Elastic Band (NEB) calculation.

Interpolates a series of images between two endpoint structures (via NEBTransformation) and optimizes them into a minimum energy path (MEP) using the NEB method.

Initializes the NEBAnalyzer object.

Parameters:

Name	Type	Description	Default
`n_images`	`int`	Number of images to use in the NEB calculation. Defaults to 5.	`5`
`spring_constant`	`float \| list[float]`	Spring constant(s) for the NEB calculation. Defaults to 0.1 eV/Ang^2.	`0.1`
`climb`	`bool`	Whether to use the climbing image method. When `True`, the path is first optimized to convergence without the climbing image, then re-optimized with it enabled, following standard NEB practice. Defaults to False.	`False`
`remove_rotation_and_translation`	`bool`	Whether to remove rotation and translation of images. Defaults to False.	`False`
`method`	`str`	Method to use for the NEB calculation. Options are 'aseneb', 'improvedtangent', 'eb', 'spline', or 'string'. Defaults to 'improvedtangent'.	`'improvedtangent'`
`interpolate_lattices`	`bool`	Whether to interpolate lattices between images. Defaults to False.	`False`
`pbc`	`bool`	Whether to apply periodic boundary conditions. Defaults to True.	`True`
`autosort_tol`	`float`	Tolerance for autosorting images. Defaults to 0.5.	`0.5`
`end_amplitude`	`float`	Amplitude for the end images. Defaults to 1.	`1`
`calculator`	`BaseCalculator \| None`	The calculator object used for energy calculations. Defaults to a lazily constructed default calculator.	`None`
`neb_transformation`	`NEBTransformation \| None`	The transformation object used to generate the intermediate images. If not provided, a new instance is initialized from `n_images`, `interpolate_lattices`, `pbc`, `autosort_tol`, and `end_amplitude`.	`None`

calculate ¶

calculate(
    initial_structure: Structure | Atoms,
    final_structure: Structure | Atoms,
    is_relaxed: bool = False,
    **kwargs: Any,
) -> dict[
    str, list[Structure] | list[float] | float | bool
]

Calculates the minimum energy path (MEP) between two structures.

Interpolates intermediate images between initial_structure and final_structure (via NEBTransformation), then optimizes the resulting band with the NEB method. If the endpoints are not already relaxed, each is relaxed with self.calculator first.

Parameters:

Name	Type	Description	Default
`initial_structure`	`Structure \| Atoms`	The initial structure for the NEB calculation.	required
`final_structure`	`Structure \| Atoms`	The final structure for the NEB calculation.	required
`is_relaxed`	`bool`	Whether `initial_structure` and `final_structure` are already relaxed. Defaults to False.	`False`
`**kwargs`	`Any`	Additional keyword arguments passed to the calculator's optimizer.	`{}`

Returns:

Type Description

dict[str, list[Structure] | list[float] | float | bool]

dict[str, list[Structure] | list[float] | float | bool]: Dictionary with keys: - images: The optimized path images as pymatgen Structure objects, including the endpoints as the first and last elements. - energies: Potential energy (eV) of each image after optimization. - barrier: Forward energy barrier, from a cubic-spline fit through the images' energies and forces (ase.mep.NEBTools.get_barrier), relative to the initial (first) image. - reverse_barrier: Reverse energy barrier, i.e. the fitted barrier minus the reaction energy. - reaction_energy: Reaction energy, i.e. the energy of the final (last) image minus the energy of the initial (first) image. - converged: Whether the last optimizer run converged within self.calculator.steps.

Raises:

Type	Description
`ValueError`	If the calculator does not implement the 'energy' property.

neb_transformation ¶

neb_transformation() -> NEBTransformation

Returns the transformation object used to generate the interpolated NEB images.

Returns:

Name	Type	Description
`NEBTransformation`	`NEBTransformation`	The transformation object used to generate the interpolated images.