ovito.modifiers

This module contains all modifiers available in OVITO. See the introduction to learn more about modifiers and their role in the data pipeline system. The following table lists the Python names of all modifier types that can be instantiated. Please consult the OVITO user manual for a more in-depth description of what each of these modifiers does.

Python class name	User interface name
AcklandJonesModifier	Ackland-Jones analysis
AffineTransformationModifier	Affine transformation
AmbientOcclusionModifier	Ambient occlusion
AssignColorModifier	Assign color
AtomicStrainModifier	Atomic strain
BondAnalysisModifier	Bond analysis
CalculateDisplacementsModifier	Displacement vectors
CentroSymmetryModifier	Centrosymmetry parameter
ChillPlusModifier	Chill+
ClearSelectionModifier	Clear selection
ClusterAnalysisModifier	Cluster analysis
ColorCodingModifier	Color coding
CombineDatasetsModifier	Combine datasets
CommonNeighborAnalysisModifier	Common neighbor analysis
ComputePropertyModifier	Compute property
ConstructSurfaceModifier	Construct surface mesh
CoordinationAnalysisModifier	Coordination analysis
CoordinationPolyhedraModifier	Coordination polyhedra
CreateBondsModifier	Create bonds
CreateIsosurfaceModifier	Create isosurface
DeleteSelectedModifier	Delete selected
DislocationAnalysisModifier	Dislocation analysis (DXA)
ElasticStrainModifier	Elastic strain calculation
ExpandSelectionModifier	Expand selection
ExpressionSelectionModifier	Expression selection
FreezePropertyModifier	Freeze property
GenerateTrajectoryLinesModifier	Generate trajectory lines
GrainSegmentationModifier	Grain segmentation
HistogramModifier	Histogram
IdentifyDiamondModifier	Identify diamond structure
InvertSelectionModifier	Invert selection
LoadTrajectoryModifier	Load trajectory
PolyhedralTemplateMatchingModifier	Polyhedral template matching
PythonScriptModifier	Python script
ReplicateModifier	Replicate
SelectTypeModifier	Select type
SliceModifier	Slice
SmoothTrajectoryModifier	Smooth trajectory
SpatialBinningModifier	Spatial binning
SpatialCorrelationFunctionModifier	Spatial correlation function
TimeAveragingModifier	Time averaging
UnwrapTrajectoriesModifier	Unwrap trajectories
VoronoiAnalysisModifier	Voronoi analysis
VoroTopModifier	VoroTop analysis
WignerSeitzAnalysisModifier	Wigner-Seitz defect analysis
WrapPeriodicImagesModifier	Wrap at periodic boundaries

Note that some modifiers of the graphical version of OVITO are missing from this list and are not accessible from Python scripts. That is because they perform simple operations that can be accomplished equally well or even easier using other means in Python.

class ovito.modifiers.AcklandJonesModifier

Base class

ovito.modifiers.StructureIdentificationModifier

This modifier analyzes the local neighborhood of each particle to identify simple crystalline structures. The structure identification is performed using the bond-angle classification method proposed by Ackland and Jones. See the corresponding user manual page for more information on this modifier.

Note that this class inherits several important parameter fields from its StructureIdentificationModifier base class.

Modifier inputs:

Particle properties
Position	The input coordinates of the particles.
Selection	The selection state of the input particles. Only needed if only_selected is True.

Modifier outputs:

Particle properties
Structure Type	The structure type computed by the algorithm for each particle, encoded as an integer value: Numeric value Python constant 0 AcklandJonesModifier.Type.OTHER 1 AcklandJonesModifier.Type.FCC 2 AcklandJonesModifier.Type.HCP 3 AcklandJonesModifier.Type.BCC 4 AcklandJonesModifier.Type.ICO
Color	Particle coloring to indicate the identified structure type for each particle; only if color_by_type is True. See the structures array on how to customize the colors.

Attributes
AcklandJones.counts.OTHER	Number of particles not matching any of the recognized structure types.
AcklandJones.counts.FCC	Number of particles identified as face-centered cubic.
AcklandJones.counts.HCP	Number of particles identified as hexagonal close packed.
AcklandJones.counts.BCC	Number of particles identified as body-centered cubic.
AcklandJones.counts.ICO	Number of particles identified as icosahedral.

Data tables
structures	A bar chart with the particle counts for each structure type identified by the modifier. You can retrieve this DataTable from the DataCollection.tables dictionary.

class ovito.modifiers.AffineTransformationModifier

Base class

ovito.pipeline.Modifier

This modifier applies an affine transformation to data elements in order to move, rotate, shear or scale them. See also the corresponding user manual page for more information.

Inputs:

The modifier can operate on any combination of the following data elements:

Data element specifier	Description
particles	Transforms the Position particle property.
vector_properties	Transforms the vectorial particle properties Velocity, Force, and Displacement.
cell	Transforms the SimulationCell.
surfaces	Transforms the vertices of all SurfaceMesh objects.
dislocations	Transforms all dislocation lines in a DislocationNetwork.
voxels	Transforms the spatial domain of all VoxelGrid objects.

The modifier will act on all of these by default. You can restrict the transform to a subset of data objects by setting the operate_on field.

Examples:

The following code applies a simple shear transformation to all particle coordinates. The shear transformation is specified as a 3x3 matrix with ones on the matrix diagonal and an off-diagonal element that is non-zero. No translation is applied to the particles. Thus all elements in the fourth column of the extended 3x4 affine transformation matrix are set to zero:

xy_shear = 0.05
mod = AffineTransformationModifier(
        operate_on = {'particles'}, # Transform particles but not the box.
        transformation = [[1, xy_shear, 0, 0],
                          [0,        1, 0, 0],
                          [0,        0, 1, 0]])

Note that the modifier itself only supports static transformations, which remain constant over the entire trajectory. However, is possible to employ the AffineTransformationModifier in a user-defined modifier function, which calculates the transformation matrix dynamically at each animation frame:

import numpy as np

def rotate(frame, data):
    theta = np.deg2rad(frame * 5.0)  # time-dependent angle of rotation
    tm = [[np.cos(theta), -np.sin(theta), 0, 0],
          [np.sin(theta),  np.cos(theta), 0, 0],
          [            0,              0, 1, 0]]
    # Execute AffineTransformationModifier as a sub-operation:
    data.apply(AffineTransformationModifier(transformation = tm))

pipeline.modifiers.append(rotate)

property only_selected

Controls whether the modifier should transform only the subset of currently selected elements (e.g. selected particles). For data object types that do not support selections, e.g. the simulation cell, this option has no effect.

Default

False

property operate_on

A set of strings specifying the kinds of data elements this modifier should act on. By default the set includes all types of data elements supported by the modifier.

Default

{'particles', 'vector_properties', 'cell', 'surfaces', 'dislocations', 'voxels'}

property relative_mode

Selects the operation mode of the modifier.

If relative_mode == True, the modifier transforms data elements by applying the specified transformation matrix.

If relative_mode == False, the modifier determines the transformation dynamically from the current shape of the SimulationCell and the specified target_cell matrix. The former will be mapped to the latter by the transformation.

Default

True

property target_cell

This 3x4 matrix specifies the target cell shape. It is only used if relative_mode == False.

The first three columns of the matrix specify the three edge vectors of the target cell. The fourth column specifies the origin vector of the target cell.

The following code shows how to scale the simulation box, whose shape may vary with simulation time, back to the initial shape at frame 0, including the cell’s contents. As a result, the output dataset generated by the modifier will have a constant simulation cell size.

pipeline = import_file("input/simulation.*.dump")
modifier = AffineTransformationModifier(
        relative_mode = False,
        target_cell = pipeline.compute(0).cell[...]
)
pipeline.modifiers.append(modifier)

property transformation

The 3x4 transformation matrix to apply to input elements. The first three matrix columns define the linear part of the transformation. The fourth column specifies the translation vector.

Note that this matrix describes a relative transformation and is used only if relative_mode == True.

Default

[[ 1.  0.  0.  0.] [ 0.  1.  0.  0.] [ 0.  0.  1.  0.]]

class ovito.modifiers.AmbientOcclusionModifier

Base class

ovito.pipeline.Modifier

Performs a quick lighting calculation to modulate the brightness of particles according to the degree of occlusion by other particles. See the corresponding user manual page for more information.

This modifier should always be inserted at the end of a data pipeline, after particle filtering and coloring is done. It can help to improve the visual appearance of complex particle structures which are rendered with the default OpenGLRenderer. It is not needed when using other renderer, which perform the ambient occlusion calculation on the fly during image rendering.

Inputs:

Particle properties
Position	The input coordinates of the particles.
Color	The original per-particle colors.

Outputs:

Particle properties
Color	The new per-particle colors, which have been modulated by the occlusion factor computed by the modifier for each particle.

property buffer_resolution

A positive integer controlling the resolution of the internal render buffer, which is used to compute how much light each particle receives. For large datasets, where the size of particles is small compared to the simulation dimensions, a higher buffer resolution should be used.

Valid range

[1, 4]

Default

3

property intensity

Controls the strength of the shading effect.

Valid range

[0.0, 1.0]

Default

0.7

property sample_count

The number of light exposure samples to compute. More samples give a more even light distribution but take longer to compute.

Default

40

class ovito.modifiers.AssignColorModifier

Base class

ovito.pipeline.Modifier

This modifier assigns a uniform color to all selected data elements. See also the corresponding user manual page for more information. The modifier can operate on different data elements:

Data element
particles	Sets the Color property of selected particles.
vectors	Sets the Vector Color property of selected particles.
bonds	Sets the Color property of selected bonds.
surface_vertices	Sets the Color property of selected vertices of a SurfaceMesh structure.
surface_faces	Sets the Color property of selected faces of a SurfaceMesh structure.
surface_regions	Sets the Color property of selected spatial regions of a SurfaceMesh structure.

By default the modifier will act on particles. You can change this by setting the operate_on field. The modifier uses the input Property named Selection to determine which subset of data elements should be assigned the specified color. Data elements whose Selection property is zero, keep their current color. In case the Selection property does not exist at all, the modifier assigns the color to every data element.

OVITO provides various modifiers for creating selections: SelectTypeModifier, ExpressionSelectionModifier, SliceModifier, InvertSelectionModifier.

Examples

Select some particles with the SelectTypeModifier and give them a new color:

pipeline.modifiers.append(SelectTypeModifier(types={1,3}))
pipeline.modifiers.append(AssignColorModifier(color=(1.0, 0.3, 0.3)))

Use the CalculateDisplacementsModifier to calculate the atomic displacement vectors and visualize them as arrows in rendered images. Particularly large displacement arrows can be highlighted using the AssignColorModifier:

# Calculate and visualize atomic displacement vectors:
modifier = CalculateDisplacementsModifier()
modifier.vis.enabled = True
pipeline.modifiers.append(modifier)

# Select particles with large displacement magnitudes:
pipeline.modifiers.append(SelectExpressionModifier(expression='DisplacementMagnitude > 1.2'))
# Highlight large displacement vectors using a special color: 
pipeline.modifiers.append(AssignColorModifier(operate_on='vectors', color=(0.0, 0.9, 1.0)))

property color

The RGB color that will be assigned to all selected elements by the modifier.

Default

(0.3, 0.3, 1.0)

property operate_on

Selects the kind of data elements this modifier should operate on. Supported values are: 'particles', 'bonds', 'vectors', 'surface_vertices', 'surface_faces', 'surface_regions'.

Default

'particles'

class ovito.modifiers.AtomicStrainModifier

Base class

ovito.pipeline.ReferenceConfigurationModifier

Computes the atomic-level deformation with respect to a reference configuration. See the corresponding user manual page for more information.

The modifier is a subclass of ReferenceConfigurationModifier, which provides the programming interface for specifying the reference configuration and how particle displacements get calculated. By default, frame 0 of the processed simulation sequence is used as static reference configuration.

Outputs:

Particle properties
Shear Strain	The von Mises shear strain invariant of the computed atomic Green-Lagrangian strain tensor.
Volumetric Strain	One third of the trace of the computed atomic Green-Lagrangian strain tensor.
Strain Tensor	The six components of the symmetric Green-Lagrangian strain tensor. Only if output_strain_tensors was set to True.
Deformation Gradient	The nine components of the atomic deformation gradient tensor. Only if output_deformation_gradients was set to True.
Stretch Tensor	The six components of the symmetric right stretch tensor U in the polar decomposition F=RU. Only if output_stretch_tensors was set to True.
Rotation	The atomic microrotation obtained from the polar decomposition F=RU as a 4-component quaternion. Only if output_rotations was set to True.
Nonaffine Squared Displacement	The D2min measure of Falk & Langer, which describes the non-affine part of the local deformation. Only if output_nonaffine_squared_displacements was set to True.
Selection	Set to a non-zero value for particles for which the modifier failed to determine a local deformation tensor, because they do not have enough neighbors within the specified cutoff distance. Only if select_invalid_particles was set to True. The selected particles without valid deformation values can subsequently be removed using a DeleteSelectedModifier.

Attributes
AtomicStrain.invalid_particle_count	Number of particles for which the modifier could not compute a deformation tensor, because they do not have enough neighbors within the specified cutoff distance. You typically should increase the cutoff distance if this value is non-zero.

property cutoff

The spatial range up to which neighboring atoms will be taken into account to calculate the local strain measure.

Default

3.0

property output_deformation_gradients

Controls the output of the per-particle deformation gradient tensors. If False, the computed tensors are not output as a particle property to save memory.

Default

False

property output_nonaffine_squared_displacements

Enables the computation of the squared magnitude of the non-affine part of the atomic displacements. The computed values are output in the "Nonaffine Squared Displacement" particle property.

Default

False

property output_rotations

Controls the calculation of the per-particle rotations.

Default

False

property output_strain_tensors

Controls the output of the per-particle strain tensors. If False, the computed strain tensors are not output as a particle property to save memory.

Default

False

property output_stretch_tensors

Controls the calculation of the per-particle stretch tensors.

Default

False

property select_invalid_particles

If True, the modifier selects the particle for which the local strain tensor could not be computed (because of an insufficient number of neighbors within the cutoff).

Default

True

class ovito.modifiers.BondAnalysisModifier

Base class

ovito.pipeline.Modifier

Computes the bond angle distribution and the bond length distribution for the particle system with bonds. See the corresponding user manual page for more information on this modifier.

The input for this modifier must contain a Bonds data object with the bond topology information, which may either be loaded from the simulation file or by first inserting the CreateBondsModifier into the pipeline.

The modifier outputs the computed bond length and bond angle histograms as two DataTable objects, which may be exported to an output text file using the export_file() function or retrieved from the pipeline’s output data collection, see the DataCollection.tables dictionary and the code examples following below.

Inputs:

Particle properties
Position	The coordinates of the input particles.
Particle Type	Required if partition is set to ByParticleType.
Selection	Required if partition is set to ByParticleSelection.

Bond properties
Topology	The input bonds list.
Bond Type	Required if partition is set to ByBondType.
Selection	Required if partition is set to ByBondSelection.

Outputs:

DataTable
bond-angle-distr	The bond angle distribution histogram computed by the modifier.
bond-length-distr	The bond length distribution histogram computed by the modifier.

Examples:

The following Python script demonstrates how to load a particle system (without bond topology) from a file, create the bond topology using OVITO’s CreateBondsModifier, and then compute the bond angle distribution. Finally, the histogram is written to an output text file using the export_file() function.

from ovito.io import import_file, export_file
from ovito.modifiers import BondAnalysisModifier, CreateBondsModifier

# Set up data pipeline:
pipeline = import_file('input/simulation.dump')
pipeline.modifiers.append(CreateBondsModifier(cutoff = 3.2))
pipeline.modifiers.append(BondAnalysisModifier(bins = 100))

# Export bond angle distribution to an output text file.
export_file(pipeline, 'output/bond_angles.txt', 'txt/table', key='bond-angle-distr')

# Convert bond length histogram to a NumPy array and print it to the terminal.
data = pipeline.compute()
print(data.tables['bond-length-distr'].xy())

The script above computes the instantanous distribution for the initial frame in the simulation file only. To compute an average bond angle distribution for the entire MD trajectory, you can make use of the TimeAveragingModifier:

from ovito.io import *
from ovito.modifiers import *

# Load a LAMMPS topology data file containing bond information:
pipeline = import_file('input/input.data', atom_style='bond')

# Load trajectory data from separate LAMMPS dump file.
traj = LoadTrajectoryModifier()
traj.source.load('input/trajectory.dump')
pipeline.modifiers.append(traj)

# Calculate instantaneous bond angle distribution. 
pipeline.modifiers.append(BondAnalysisModifier())

# Perform time averaging of the DataTable 'bond-angle-distr'.
pipeline.modifiers.append(TimeAveragingModifier(operate_on='table:bond-angle-distr'))

# Compute and export the time-averaged histogram to a text file.
export_file(pipeline, "output/avg_bond_angles.txt", 'txt/table', key='bond-angle-distr[average]')

property bins

The number of bins in the length and angle histograms generated by the modifier.

Default

180

property cosine_mode

If set to True, the modifier will calculate the distribution of the cosines of the bond angles instead of the angles themselves, and the bond angle histogram output by the modifier will then extend over the value interval [-1, +1] instead of [0, 180] degrees.

Default

False

property length_cutoff

Maximum bond length at which the bond length distribution gets truncated. Together with the bins parameter, this value determines the bin size of the bond length histogram output by the modifier.

Note that this parameter does not affect the computation of the bond angle distribution. All bonds are included in the angle distribution, even if their length exceeds the cutoff.

Default

4.0

property partition

This mode parameter instructs the modifier to split the bond length and bond angle distributions into partial distributions, one for each particle or bond type combination. Available partitioning modes are:

• BondAnalysisModifier.Partition.Off

• BondAnalysisModifier.Partition.ByBondType

• BondAnalysisModifier.Partition.ByBondSelection

• BondAnalysisModifier.Partition.ByParticleType

• BondAnalysisModifier.Partition.ByParticleSelection

If partitioning is turned on, the y property array of the two DataTable histograms computed by the modifier will have multiple vector components, one for each partial distribution. The follow code example shows how to access the partial histograms:

from ovito.io import import_file
from ovito.modifiers import BondAnalysisModifier

# Load LAMMPS data file with bonds and compute bond angle distribution partitioned by bond type:
pipeline = import_file('input/input.data', atom_style='bond')
pipeline.modifiers.append(BondAnalysisModifier(partition=BondAnalysisModifier.Partition.ByBondType))
data = pipeline.compute()

# Retrieve y-property array from output DataTable:
histogram = data.tables['bond-angle-distr'].y

# Print the individual columns of the vector-valued property array:
for column, name in enumerate(histogram.component_names):
    print("Angle distribution for bond types:", name)
    print(histogram[:,column])

Default

BondAnalysisModifier.Partition.Off

class ovito.modifiers.CalculateDisplacementsModifier

Base class

ovito.pipeline.ReferenceConfigurationModifier

Computes the displacement vectors of particles with respect to a reference configuration. See the corresponding user manual page for more information.

The modifier is a subclass of ReferenceConfigurationModifier, which provides the programming interface for specifying the reference configuration and how particle displacements get calculated. By default, frame 0 of the processed simulation sequence is used as static reference configuration.

Outputs:

Particle properties
Displacement	The computed displacement vectors.
Displacement Magnitude	The length of the computed displacement vectors.

property vis

A VectorVis element controlling the visual representation of the computed displacement vectors. Note that the computed displacement vectors are hidden by default. You can enable the visualization of arrows as follows:

modifier = CalculateDisplacementsModifier()
modifier.vis.enabled = True
modifier.vis.color = (0.8, 0.0, 0.5)

class ovito.modifiers.CentroSymmetryModifier

Base class

ovito.pipeline.Modifier

Computes the centro-symmetry parameter (CSP) of each particle, which is a measure of the local lattice disorder around a particle in centrosymmetric crystal lattices. See the corresponding user manual page for more information.

Inputs:

Particle properties
Position	The input coordinates of the particles.

Outputs:

Particle properties
Centrosymmetry	The non-negative CSP value computed for each particle. Values close to zero mean neighboring particles are in a perfect centrosymmetric arrangement.

property mode

Selects how pair-wise opposite neighbors should be picked during the CSP calculation. Valid modes are:

• CentroSymmetryModifier.Mode.Conventional

• CentroSymmetryModifier.Mode.Matching

See the user manual for more information on these two modes.

Default

CentroSymmetryModifier.Mode.Conventional

property num_neighbors

The number of nearest neighbors to take into account for the computation, e.g.

• 12 for FCC crystals

• 8 for BCC crystals

Default

12

class ovito.modifiers.ChillPlusModifier

Base class

ovito.modifiers.StructureIdentificationModifier

Analyzes the local neighborhood of each particle to identify different structural arrangements of water molecules. The structure identification is performed using the CHILL+ algorithm. See the corresponding user manual page for more information.

Note that this class inherits several important parameter fields from its StructureIdentificationModifier base class.

Modifier inputs:

Particle properties
Position	The input coordinates of the particles.
Selection	The selection state of the input particles. Only needed if only_selected is True.

Modifier outputs:

Particle properties
Structure Type	The structure type computed by the algorithm for each particle, encoded as an integer value: Numeric value Python constant 0 ChillPlusModifier.Type.OTHER 1 ChillPlusModifier.Type.HEXAGONAL_ICE 2 ChillPlusModifier.Type.CUBIC_ICE 3 ChillPlusModifier.Type.INTERFACIAL_ICE 4 ChillPlusModifier.Type.HYDRATE 5 ChillPlusModifier.Type.INTERFACIAL_HYDRATE
Color	Particle coloring to indicate the identified structure type for each particle; only if color_by_type is True. See the structures array on how to customize the colors.

Attributes
ChillPlus.counts.OTHER	Number of particles not matching any of the known structure types.
ChillPlus.counts.HEXAGONAL_ICE	Number of particles identified as hexgonal ice.
ChillPlus.counts.CUBIC_ICE	Number of particles identified as cubic ice.
ChillPlus.counts.INTERFACIAL_ICE	Number of particles identified as interfacial ice.
ChillPlus.counts.HYDRATE	Number of particles identified as hydrate.
ChillPlus.counts.INTERFACIAL_HYDRATE	Number of particles identified as interfacial hydrate.

Data tables
structures	A bar chart with the particle counts for each structure type identified by the modifier. You can retrieve this DataTable from the DataCollection.tables dictionary.

property cutoff

The cutoff distance for bonds between water molecules.

Default

3.5

class ovito.modifiers.ClearSelectionModifier

Base class

ovito.pipeline.Modifier

This modifier clears the current selection by removing the Selection property from a PropertyContainer such that subsequent modifiers in the pipeline won’t see it. See also the corresponding user manual page for more information. The modifier can operate on different kinds of data elements:

Data element
particles	Removes the Selection property of particles.
bonds	Removes the Selection property of bonds.
voxels	Removes the Selection property of voxel grid cells.

By default the modifier will act on particles. You can change this by setting the operate_on field.

property operate_on

Selects the kind of data elements this modifier should operate on. Supported values are: 'particles', 'bonds', 'voxels'.

Default

'particles'

class ovito.modifiers.ClusterAnalysisModifier

Base class

ovito.pipeline.Modifier

This modifier groups particles into disconnected clusters based on a selectable connectivity criterion. See the corresponding user manual page for more information.

Outputs:

Particle properties
Cluster	The 1-based numeric ID of the cluster each particle was assigned to by the modifier.

Attributes
ClusterAnalysis.cluster_count	Total number of clusters produced by the modifier. Cluster IDs range from 1 to this number.
ClusterAnalysis.largest_size	Number of particles in the largest cluster (cluster ID 1). Only computed if sort_by_size is set to True.

DataTable

clusters

The list of clusters identified by the modifier. You can retrieve this DataTable from the DataCollection.tables dictionary. It consists of several data columns, each being a separate Property object: Cluster Identifier: The numeric ID of each identified cluster (1-based). Cluster Size: The number of particles in each identified cluster. Center of Mass: The XYZ coordinates of the center of mass of each cluster (only if compute_com is set). Radius of Gyration: The radius of gyration of each cluster (only if compute_gyration is set) in simulation units of length. Gyration Tensor: The gyration tensor of each cluster (only if compute_gyration is set) in simulation units of length squared. The tensors are stored as vectors with six components [XX, YY, ZZ, XY, XZ, YZ].

Example:

The following example code demonstrates how to export the data table generated by the modifier to a text file and how to access the information from Python.

from ovito.io import import_file, export_file
from ovito.modifiers import ClusterAnalysisModifier
import numpy

pipeline = import_file("input/simulation.dump")
pipeline.modifiers.append(ClusterAnalysisModifier(
    cutoff=2.8, 
    sort_by_size=True, 
    compute_com=True))

# Export results of the clustering algorithm to a text file:
export_file(pipeline, 'output/clusters.txt', 'txt/table', key='clusters')

# Directly access information stored in the DataTable:
data = pipeline.compute()
cluster_table = data.tables['clusters']
print("Centers of mass:")
print(cluster_table['Center of Mass'][...])

property cluster_coloring

Enables the coloring of particles based on their assignment to clusters. Each cluster is represented by a unique random color.

Default

False

property compute_com

Enables the computation of the center of mass of each cluster. The center coordinates will be output as an extra column named Center of Mass in the clusters data table.

Default

False

property compute_gyration

Enables the computation of the radius of gyration and the gyration tensor of each cluster. Both quantities will be output as auxiliary properties to the clusters data table, see above.

Default

False

property cutoff

The cutoff distance used by the algorithm to form clusters of connected particles. This parameter is only used when neighbor_mode is set to CutoffRange; otherwise it is ignored.

Default

3.2

property neighbor_mode

Selects the neighboring criterion for the clustering algorithm. Valid values are:

• ClusterAnalysisModifier.NeighborMode.CutoffRange

• ClusterAnalysisModifier.NeighborMode.Bonding

In the first mode (CutoffRange), the clustering algorithm treats pairs of particles as neighbors which are within a certain range of each other given by the parameter cutoff.

In the second mode (Bonding), particles which are connected by bonds are combined into clusters. Bonds between particles can either be loaded from the input simulation file or dynamically created using for example the CreateBondsModifier or the VoronoiAnalysisModifier.

Default

ClusterAnalysisModifier.NeighborMode.CutoffRange

property only_selected

Lets the modifier perform the analysis only for selected particles. In this case, particles which are not selected are treated as if they did not exist and will be assigned cluster ID 0.

Default

False

property sort_by_size

Enables the sorting of clusters by size (in descending order). Cluster 1 will be the largest cluster, cluster 2 the second largest, and so on.

Default

False

property unwrap_particles

Enables the “unwrapping” of particle coordinates in order to make clusters contiguous that are cut by a periodic simulation cell boundary.

Default

False

class ovito.modifiers.ColorCodingModifier

Base class

ovito.pipeline.Modifier

Colors elements to visualize one of their properties. See also the corresponding user manual page for more information. The modifier can operate on different data elements:

Data element
particles	Sets the Color property of particles.
vectors	Sets the Vector Color property of particles.
bonds	Sets the Color property of bonds.
voxels	Sets the Color property of voxel grid cells.
surface_vertices	Sets the Color property of the vertices of a SurfaceMesh structure.
surface_faces	Sets the Color property of the faces of a SurfaceMesh structure.
surface_regions	Sets the Color property of the spatial regions of a SurfaceMesh structure.

By default the modifier will act on particles. You can change this by setting the operate_on field.

Example:

pipeline.modifiers.append(ColorCodingModifier(
    property = 'Potential Energy',
    gradient = ColorCodingModifier.Hot()
))

If the start_value and end_value parameters are not explicitly specified during modifier construction, the modifier will automatically set them to the minimum and maximum of the input property (at time of insertion into the data pipeline).

The ColorLegendOverlay may be used in conjunction with a ColorCodingModifier to insert a color legend into rendered images.

property end_value

This parameter defines, together with the start_value parameter, the normalization range for mapping the input property values to colors.

property gradient

The color gradient used to map normalized property values to colors. Available gradient types are:

• ColorCodingModifier.BlueWhiteRed()

• ColorCodingModifier.Grayscale()

• ColorCodingModifier.Hot()

• ColorCodingModifier.Jet()

• ColorCodingModifier.Magma()

• ColorCodingModifier.Rainbow() [default]

• ColorCodingModifier.Viridis()

• ColorCodingModifier.Gradient(Nx3 array)

• ColorCodingModifier.Image("<image file path>")

The Gradient constructor lets you define your own coloring scheme and takes an array of dimensions N x 3 containing a table of colors (RGB values in the range [0-1]). The color coding modifier will linearly interpolate between the N colors of the table. The Image constructor expects the path to an image file on disk, which will be used to create a custom color gradient from the first row of pixels in the image.

Code example:

color_table = [
    (1.0, 0.0, 0.0),  # red
    (1.0, 1.0, 0.0),  # yellow
    (1.0, 1.0, 1.0)   # white
]
modifier = ColorCodingModifier(
    property = 'Position.X',
    gradient = ColorCodingModifier.Gradient(color_table)
)

property only_selected

If True, only selected elements will be affected by the modifier and the existing colors of unselected elements will be preserved; if False, all elements will be colored.

Default

False

property operate_on

Selects the kind of data elements this modifier should operate on. Supported values are: 'particles', 'bonds', 'vectors', 'voxels', 'surface_vertices', 'surface_faces', 'surface_regions'.

Default

'particles'

property property

The name of the input property that should be used to color elements.

When the input property has multiple components, then a component name must be appended to the property base name, e.g. "Velocity.X".

property start_value

This parameter defines, together with the end_value parameter, the normalization range for mapping the input property values to colors.

class ovito.modifiers.CombineDatasetsModifier

Base class

ovito.pipeline.Modifier

This modifier loads a set of particles from a separate simulation file and merges them into the primary dataset. See also the corresponding user manual page for more information.

Example:

from ovito.io import import_file, export_file
from ovito.modifiers import CombineDatasetsModifier

# Load a first set of particles.
pipeline = import_file('input/first_file.dump')

# Insert the particles from a second file into the dataset. 
modifier = CombineDatasetsModifier()
modifier.source.load('input/second_file.dump')
pipeline.modifiers.append(modifier)

# Export combined dataset to a new file.
export_file(pipeline, 'output/combined.dump', 'lammps/dump',
            columns = ['Position.X', 'Position.Y', 'Position.Z'])

property source

A FileSource that provides the set of particles to be merged. You can call its load() function to load a data file as shown in the code example above.

class ovito.modifiers.CommonNeighborAnalysisModifier

Base class

ovito.modifiers.StructureIdentificationModifier

This modifier analyzes the local neighborhood of each particle to identify simple crystalline structures. The structure identification is performed using the Common Neighbor Analysis (CNA) method. See the corresponding user manual page for more information on this modifier and the structure identification algorithm it implements.

Note that this class inherits several important parameter fields from its StructureIdentificationModifier base class.

Modifier inputs:

Particle properties
Position	The input coordinates of the particles. They are not used if mode is set to BondBased.
Selection	The selection state of the input particles. Only needed if only_selected is set to True.

Bond properties
Topology	The input bond topology; required only if mode is set to BondBased.

Modifier outputs:

Particle properties
Structure Type	The structure type computed by the algorithm for each particle, encoded as an integer value: Numeric value Python constant 0 CommonNeighborAnalysisModifier.Type.OTHER 1 CommonNeighborAnalysisModifier.Type.FCC 2 CommonNeighborAnalysisModifier.Type.HCP 3 CommonNeighborAnalysisModifier.Type.BCC 4 CommonNeighborAnalysisModifier.Type.ICO
Color	Particle coloring to indicate the identified structure type for each particle; only if color_by_type is True. See the structures array on how to customize the colors.

Attributes
CommonNeighborAnalysis.counts.OTHER	Number of particles not matching any of the recognized structure types.
CommonNeighborAnalysis.counts.FCC	Number of particles identified as face-centered cubic.
CommonNeighborAnalysis.counts.HCP	Number of particles identified as hexagonal close packed.
CommonNeighborAnalysis.counts.BCC	Number of particles identified as body-centered cubic.
CommonNeighborAnalysis.counts.ICO	Number of particles identified as icosahedral.

Data tables
structures	A bar chart with the particle counts for each structure type identified by the modifier. You can retrieve this DataTable from the DataCollection.tables dictionary.

property cutoff

The cutoff radius used for the conventional common neighbor analysis. This parameter is only used if mode is CommonNeighborAnalysisModifier.Mode.FixedCutoff.

Default

3.2

property mode

Selects the algorithm type. One of the following constants:

• CommonNeighborAnalysisModifier.Mode.FixedCutoff

• CommonNeighborAnalysisModifier.Mode.AdaptiveCutoff

• CommonNeighborAnalysisModifier.Mode.IntervalCutoff

• CommonNeighborAnalysisModifier.Mode.BondBased

Default

CommonNeighborAnalysisModifier.Mode.AdaptiveCutoff

class ovito.modifiers.ComputePropertyModifier

Base class

ovito.pipeline.Modifier

Evaluates a user-defined math expression for every input element and stores the results in an output property. See also the corresponding user manual page for more information. The modifier can operate on different data elements:

Data element
particles	Computes a particle property.
bonds	Computes a bond property.
voxels	Computes a voxel grid property.

By default the modifier will act on particles. You can change this by setting the operate_on field.

Example:

pipeline.modifiers.append(ComputePropertyModifier(
    output_property = 'Color',
    expressions = ['Position.X / CellSize.X', '0.0', '0.5']
))

Note that a PythonScriptModifier may sometimes be a better choice than this modifier to set properties, in particular when the computation involves complex element indexing or conditions.

property cutoff_radius

The cutoff radius up to which neighboring particles are visited to compute neighbor_expressions. This parameter is only used if operate_on is set to 'particles' and the neighbor_expressions field has been set.

Default

3.0

property expressions

A list of strings containing the math expressions to compute, one for each vector component of the selected output property. If the output property is scalar, the list must comprise one expression string.

See the corresponding user manual page for a description of the expression syntax.

Default

["0"]

property neighbor_expressions

The list of strings containing the math expressions for the per-neighbor terms, one for each vector component of the output particle property. If the output property is scalar, the list must comprise one string only.

The neighbor expressions are only evaluated for each neighbor particle and the value is added to the output property of the central particle. Neighbor expressions are only evaluated if operate_on is set to 'particles'.

Default

[]

property only_selected

If True, the property is only computed for currently selected elements. In this case, the property values of unselected elements will be preserved if the output property already exists.

Default

False

property operate_on

Selects the kind of data elements this modifier should operate on. Supported values are: 'particles', 'bonds', 'voxels'.

Default

'particles'

property output_property

The output property that will be assigned the computed values.

Default

"My property"

class ovito.modifiers.ConstructSurfaceModifier

Base class

ovito.pipeline.Modifier

Constructs the geometric surface of the solid region formed by point-like particles. The modifier generates a SurfaceMesh, which is a closed manifold consisting of triangular elements. It can further compute the surface area and the volume of the region enclosed by the surface mesh as well as empty pores in the interior of the filled region. See the corresponding user manual page for more information.

Basic usage example:

from ovito.io import import_file
from ovito.modifiers import ConstructSurfaceModifier

pipeline = import_file("input/simulation.dump")
pipeline.modifiers.append(ConstructSurfaceModifier(radius = 2.9))
data = pipeline.compute()
surface_mesh = data.surfaces['surface']

Alpha-shape method:

The modifier supports two different construction methods for the surface, which are selected by setting the method parameter. The AlphaShape method uses a Delaunay tessellation constructed on the basis of the input particle coordinates to identify volume elements that cannot fully accomodate a virtual probe sphere of a given radius and which are thus classified as being part of the filled (or solid) region. The radius parameter controls the size of the virtual probe sphere and determines how much detail of the morphology are resolved during the surface construction. A larger radius leads to a surface with fewer details, reflecting only coarse features of the surface morphology. A small radius, on the other hand, will resolve finer surface features and small pores in the interior of a solid, for example.

See [A. Stukowski, JOM 66 (2014), 399-407] for a description of the alpha-shape surface construction algorithm. Please cite this paper when publishing analysis results obtained with the help of OVITO.

Gaussian density method:

The second surface construction method supported by the modifier is based on the smeared-out representation of the point-like particles as overlapping Gaussian distribution functions centered on each particle site. The density field has local maxima at the particle centers and decays to zero far away from any particles. The surface mesh representation is then constructed as an iso-surface of the Gaussian density field, with the threshold isolevel chosen such that the resulting surface roughly represents the finite diameters of the input particle spheres.

pipeline.modifiers.append(ConstructSurfaceModifier(
    method = ConstructSurfaceModifier.Method.GaussianDensity,
    radius_scaling = 1.2,
    isolevel = 0.5))

Volume analysis:

The AlphaShape method provides the option to compute volume and surface area of the solid region formed by the input particles. Furthermore, it allows identifying pores inside the solid and to compute their respective volumes and interior surface areas. To enable this extra calculation step, set identify_regions to True. The modifier will output various aggregate values such as the total surface area and the pore volume fraction as global attributes, see the table below.

pipeline.modifiers.append(ConstructSurfaceModifier(
    method = ConstructSurfaceModifier.Method.AlphaShape,
    radius = 2.9,
    identify_regions = True))

data = pipeline.compute()
print("Surface area: %f" % data.attributes['ConstructSurfaceMesh.surface_area'])
print("Solid volume: %f" % data.attributes['ConstructSurfaceMesh.filled_volume'])

Outputs:

Surface
SurfaceMesh	The surface mesh computed by the modifier. You can access it through the surfaces dictionary in the output DataCollection under the lookup key "surface", see the code example below.

The following output attributes will only be computed if identify_regions is turned on:

Attributes
ConstructSurfaceMesh.surface_area	The total surface area in squared simulation units of length. Includes only interfaces between empty and filled regions if they modifier identifies internal boundaries (filled-filled interfaces).
ConstructSurfaceMesh.specific_surface_area	The total surface area divided by the simulation cell volume (in reciprocal units of length).
ConstructSurfaceMesh.filled_volume	The total volume of the filled region(s) bounded by the surface in cubic simulation units of length.
ConstructSurfaceMesh.empty_volume	The total volume of the empty region(s) in cubic simulation units of length. Includes interior pores as well as the overlap of the infinite exterior region with the simulation box volume in case of a finite system with open boundary conditions.
ConstructSurfaceMesh.filled_fraction	Total volume of filled regions divided by the simulation box volume (unitless).
ConstructSurfaceMesh.empty_fraction	Total volume of empty regions divided by the simulation box volume (unitless).
ConstructSurfaceMesh.filled_region_count	Integer number of disconnected volumetric regions filled with particles.
ConstructSurfaceMesh.empty_region_count	Integer number of disconnected empty regions (voids).
ConstructSurfaceMesh.cell_volume	The total volume of the simulation box in cubic simulation units of length. Equal to SimulationCell.volume.

The following particle property will only be assigned if select_surface_particles is turned on:

Particle properties
Selection	Will be set to 1 by the modifier for particles right at the surface of a filled region, and 0 for interior particles or isolated particles not forming any solid. Computed only if select_surface_particles is turned on.

property grid_resolution

Specifies the number of grid cells along the longest dimension of the simulation cell when generating the Gaussian density grid. This parameter thus also controls the level of detail of the of the final surface mesh, which is constructed as an isosurface from the density grid data.

The parameter is only used if GaussianDensity is selected as construction method.

Default

50

property identify_regions

This option lets the modifier identify individual volumetric regions (filled with particles or empty) and compute their volumes and surface areas.

Identifying volumetric regions is only supported by the AlphaShape surface construction method. For other construction methods, the setting will be ignored.

Default

False

property isolevel

The threshold value used for constructing the isosurface from the Gaussian density field. This parameter is only used if the selected construction method is set to GaussianDensity.

Default

0.6

property method

Selects the algorithm for constructing the surface mesh from the input particles. The following methods are supported:

• ConstructSurfaceModifier.Method.AlphaShape

• ConstructSurfaceModifier.Method.GaussianDensity

If the default alpha-shape method is selected, you should set the radius parameter to specify the size of the virtual probe sphere, which determines the level of detail of the resulting surface.

Default

ConstructSurfaceModifier.Method.AlphaShape

property only_selected

If True, the modifier acts only on selected particles and ignores other particles; if False, the modifier constructs the surface around all particles.

Default

False

property radius

The radius of the virtual probe sphere used in the alpha-shape surface construction algorithm. This parameter is only used by the modifier if method is set to the default mode AlphaShape.

A rule of thumb is that the radius parameter should be slightly larger than the typical distance between nearest neighbor particles.

Default

4.0

property radius_scaling

Scaling factor applied to the input particle radii when constructing the Gaussian density field for surface generation. This parameter is only used if the selected construction method is set to GaussianDensity. The optional scaling serves as a way to increase the spatial extent of the Gaussian function centered on each particle site and to increase the overlap between the Gaussians, yielding a more connected isosurface.

Default

1.0

property select_surface_particles

This option lets the modifier select those particles that are part of the constructed geometric surface. This provides a way to identify surface atoms or particles. If the flag is set to True, the modifier will create the Selection particle property, assigning a value of 1 to surface particles and 0 to bulk particles. This particle selection may then be used in subsequent operations in the data pipeline.

Selecting surface particles is only supported by the AlphaShape construction method. For other construction methods, the setting is ignored.

Default

False

property smoothing_level

The number of times the smoothing procedure is applied to the generated surface mesh. This parameter is only used by the modifier if method is set to the default mode AlphaShape.

Note that the smoothing level does only affect the computed surface area but not the solid volume. That is because the solid volume is computed before smoothing the mesh. (Smoothing is supposed to be volume preserving.)

Default

8

property transfer_properties

This option lets the modifier copy the property values of the particles over to the vertices of the generated surface mesh.

Note: If the Gaussian density method is used, only particle properties of data type Float will be transferred to the surface. Integer properties will be skipped, because the algorithm has to blend the property values of several particles in order to compute the value at each surface vertex. In case of the alpha-shape method, all properties are transferred, including integer properties, because there is a one-to-one mapping between input particles and output mesh vertices.

Default

False

property vis

The SurfaceMeshVis element controlling the visual appearance of the generated surface in rendered images and animations.

class ovito.modifiers.CoordinationAnalysisModifier

Base class

ovito.pipeline.Modifier

Computes coordination number of each particle and the radial distribution function (RDF) for the entire system. See the corresponding user manual page for more information.

Inputs:

Particle properties
Position	The input coordinates of the particles.
Particle Type	Required if partial is set to True.

Outputs:

Particle properties
Coordination	The number of neighbors of each particle.

DataTable
coordination-rdf	The RDF computed by the modifier. You can retrieve the RDF data through the tables dictionary of the output DataCollection under the lookup key "coordination-rdf", see the code examples below.

Examples:

The following batch script demonstrates how to load a particle configuration, compute the RDF using the modifier and export the data to a text file:

from ovito.io import import_file
from ovito.modifiers import CoordinationAnalysisModifier
import numpy

# Load a particle dataset, apply the modifier, and evaluate pipeline.
pipeline = import_file("input/simulation.dump")
modifier = CoordinationAnalysisModifier(cutoff = 5.0, number_of_bins = 200)
pipeline.modifiers.append(modifier)
data = pipeline.compute()

# Export the computed RDF data to a text file.
numpy.savetxt("output/rdf.txt", data.tables['coordination-rdf'].xy())

The next script demonstrates how to compute the RDF for every frame of a simulation sequence and build a time-averaged RDF histogram from the data:

from ovito.io import import_file
from ovito.modifiers import CoordinationAnalysisModifier
import numpy

# Load a simulation trajectory consisting of several frames:
pipeline = import_file("input/simulation.dump")

# Insert the modifier into the pipeline:
modifier = CoordinationAnalysisModifier(cutoff = 5.0, number_of_bins = 200)
pipeline.modifiers.append(modifier)

# Initialize array for accumulated RDF histogram to zero:
total_rdf = numpy.zeros((modifier.number_of_bins, 2))

# Iterate over all frames of the sequence.
for frame in range(pipeline.source.num_frames):
    # Evaluate pipeline to let the modifier compute the RDF of the current frame:
    data = pipeline.compute(frame)
    # Accumulate RDF histograms:
    total_rdf += data.tables['coordination-rdf'].xy()

# Averaging:
total_rdf /= pipeline.source.num_frames

# Export the average RDF to a text file:
numpy.savetxt("output/rdf.txt", total_rdf)

property cutoff

Specifies the cutoff distance for the coordination number calculation and also the range up to which the modifier calculates the RDF.

Default

3.2

property number_of_bins

The number of histogram bins to use when computing the RDF.

Default

200

property partial

Setting this flag to true requests calculation of element-specific (partial) RDFs.

Default

False

class ovito.modifiers.CoordinationPolyhedraModifier

Base class

ovito.pipeline.Modifier

Constructs coordination polyhedra around selected particles for visualization purposes. See the corresponding user manual page for more information. A coordination polyhedron is the convex hull constructed from the neighbor atoms of some central atom.

In order to tell the modifier which input particles should be surrounded by a coordination polyhedron, the central particles must be selected first. The particle selection can be created by inserting a SelectTypeModifier into the data pipeline prior to this modifier.

The modifier furthermore requires input bonds connecting a central particle with its neighbors to define which atoms the algorithm should use in the construction of the convex hulls. You can insert the CreateBondsModifier prior to this modifier to let OVITO dynamically generate neighbor bonds between particles.

Example:

# Import simulation dataset and add it to the scene:
pipeline = import_file("input/simulation.0.dump")
pipeline.add_to_scene()

# Select all atoms of species type 1. They will form the centers of the polyhedra.
pipeline.modifiers.append(SelectTypeModifier(types={1}))
# Create bonds between nearby atoms.
pipeline.modifiers.append(CreateBondsModifier(cutoff=3.0))
# Let the modifier construct the coordination polyhedra around selected atoms.
modifier = CoordinationPolyhedraModifier()
pipeline.modifiers.append(modifier)

# Optional: Configure visual appearance of the polyhedra and make them semi-transparent.
modifier.vis.surface_transparency = 0.4
modifier.vis.highlight_edges = True

Modifier inputs:

Particle properties
Selection	Determines the subset of particles for which coordination polyhedra should be constructed. You can select all particles of certain chemical type(s) by first inserting a SelectTypeModifier into the pipeline.

Bond properties
Topology	The bonds list, which is used to determine the set of bonded neighbor particles serving as vertices for the construction of a coordination polyhedron around some central particle. You can let OVITO create the bond topology by inserting a CreateBondsModifier into the pipeline first.

Modifier outputs:

Polyhedral mesh
coord-polyhedra	The polyhedral SurfaceMesh generated by the modifier. You can retrieve it from the surfaces dictionary of the DataCollection.

property vis

The SurfaceMeshVis element controlling the visual appearance of the generated polyhedra in rendered images and animations.

class ovito.modifiers.CreateBondsModifier

Base class

ovito.pipeline.Modifier

Creates bonds between nearby particles. See the corresponding user manual page for more information.

Inputs:

Particle properties
Position	The xyz coordinates of the input particles.
Particle Type	The particle type information, which is used only if mode is set to Pairwise.
Molecule Identifier	The assignment of atoms to molecules, which is used only if intra_molecule_only is set to True.

Outputs:

Bond properties
Topology	The modifier will create new bond topology entries and append them to the property arrays in an existing Bonds object; or it creates a new Bonds instance if necessary.
Periodic Image	Stores the transitions of each bond through the faces of a periodic simulation cell if the bond connects two particles from different periodic images of the system.
Bond Type	The type ID information that is assigned to newly created bonds according to the modifier’s bond_type field.

Attributes
CreateBonds.num_bonds	The number of bonds that exists after the modifier’s operation.

property bond_type

The BondType that will be assigned to the newly created bonds. This lets you control the display color of the new bonds.

property cutoff

The upper cutoff distance for the creation of bonds between particles. This parameter is only used if mode is set to Uniform.

Default

3.2

get_pairwise_cutoff(type_a, type_b) → float

Returns the pair-wise cutoff distance that was previously set for a specific pair of particle types using the set_pairwise_cutoff() method.

Parameters

• type_a (str,int) – The name or numeric id of the first particle type

• type_b (str,int) – The name or numeric id of the second particle type

Returns

The cutoff distance set for the type pair. Returns zero if no cutoff has been set for the pair.

property intra_molecule_only

If this option is set to true, the modifier will create bonds only between atoms that belong to the same molecule (i.e. which have the same molecule ID assigned to them).

Default

False

property lower_cutoff

The minimum bond length. No bonds will be created between particles whose distance is below this threshold.

Default

0.0

property mode

Selects the mode of operation. Valid modes are:

• CreateBondsModifier.Mode.Uniform

• CreateBondsModifier.Mode.Pairwise

In Uniform mode one global cutoff is used irrespective of the particle types. In Pairwise mode a separate cutoff distance must be specified for all pairs of particle types between which bonds are to be created.

Default

CreateBondsModifier.Mode.Uniform

set_pairwise_cutoff(type_a, type_b, cutoff)

Sets the cutoff range for creating bonds between a specific pair of particle types. This information is only used if mode is set to Pairwise.

Parameters

• type_a (str,int) – The name or numeric id of the first particle type

• type_b (str,int) – The name or numeric id of the second particle type

• cutoff (float) – The cutoff distance to be used by the modifier for the type pair

If you want no bonds to be created between a pair of types, set the corresponding cutoff radius to zero (which is the default).

property vis

The BondsVis object controlling the visual appearance of the bonds created by this modifier.

class ovito.modifiers.CreateIsosurfaceModifier

Base class

ovito.pipeline.Modifier

Constructs an isosurface for a scalar field defined on a three-dimensional VoxelGrid. See the corresponding user manual page for more information on this modifier.

The modifier takes an existing VoxelGrid as input, for example a charge density field loaded from an input simulation file or a three-dimensional voxel grid dynamically computed by the SpatialBinningModifier. The input voxel grid for the isosurface must be specified by setting the operate_on field to the string 'voxels:' followed by the identifier of the input voxel grid. Furthermore, you need to specify the name of the field Property for which to construct the isosurface, because multiple fields may be defined on the same voxel grid.

# Import a charge density field from a VASP calculation:
pipeline = import_file("input/CHGCAR.nospin.gz")
pipeline.add_to_scene()

# The identifier of the imported VoxelGrid is 'charge-density',
# and the field property defined on the grid is named 'Charge density'.
print(pipeline.source.data.grids['charge-density'])

# Set up the isosurface modifier:
modifier = CreateIsosurfaceModifier(
    operate_on = 'voxels:charge-density', 
    property = 'Charge density',
    isolevel = 0.00014)
pipeline.modifiers.append(modifier)

# Adjust visual appearance of the isosurface in rendered images:
modifier.vis.show_cap = False
modifier.vis.surface_transparency = 0.4

The following example demonstrates how to construct an isosurface for a dynamically computed field by the SpatialBinningModifier:

pipeline.modifiers.append(SpatialBinningModifier(
    property = 'Particle Type',
    direction = SpatialBinningModifier.Direction.XYZ, 
    bin_count = (30, 30, 30),
    reduction_operation = SpatialBinningModifier.Operation.Mean
))
pipeline.modifiers.append(CreateIsosurfaceModifier(
    operate_on = 'voxels:binning[Particle Type]', 
    property = 'Particle Type',
    isolevel = 1.5
))

Modifier outputs:

SurfaceMesh
isosurface	The SurfaceMesh constructed by the modifier. You can retrieve it from the surfaces dictionary of the pipeline’s output DataCollection.

DataTable
isosurface-histogram	A histogram of the input field values. You can retrieve this DataTable from the tables dictionary of the pipeline’s output DataCollection.

property isolevel

The value at which to create the isosurface.

Default

0.0

property operate_on

Specifies the VoxelGrid this modifier should operate on. Set this to the prefix string 'voxels:' followed by the identifier of the grid.

Note: You can use the Python statement print(pipeline.compute().grids) to find out what the identifiers of the voxel grids in your data pipeline are.

Default

'voxels:'

property property

The name of the Property in the selected input VoxelGrid for which to construct the isosurface. Note that this must be a scalar property. If the input grid property is a vector property, you need to explicitly specify which vector component to use, e.g. 'Dipole Orientation.Z'.

property vis

The SurfaceMeshVis controlling the visual appearance of the generated isosurface in rendered images.

class ovito.modifiers.DeleteSelectedModifier

Base class

ovito.pipeline.Modifier

This modifier deletes the currently selected elements. See also the corresponding user manual page for more information.

Inputs:

The modifier can operate on any combination of the following data elements:

Data element
particles	Deletes all particles with a non-zero value of the Selection property.
bonds	Deletes all bonds with a non-zero value of the Selection property.
surface_regions	Deletes all selected spatial regions (including the corresponding mesh faces) from a SurfaceMesh.

By default the modifier will act on all data element types simultaneously. You can restrict it to a subset by setting the operate_on field.

property operate_on

A set of strings specifying the kinds of data elements this modifier should operate on. By default the set contains all data element types supported by the modifier.

Default

{'particles', 'bonds', 'surface_regions'}

class ovito.modifiers.DislocationAnalysisModifier

Base class

ovito.modifiers.StructureIdentificationModifier

This analysis modifier extracts all dislocations in a crystal and converts them to continuous line segments. The computational method behind this is called Dislocation Extraction Algorithm (DXA) and is described in the paper [MSMSE 20 (2012), 085007]. See also the corresponding user manual page for this modifier.

The extracted dislocation lines are output as a DislocationNetwork object by the modifier and can be accessed through the DataCollection.dislocations field after the modification pipeline has been evaluated. This is demonstrated in the example script below.

Furthermore, you can use the export_file() function to write the dislocation lines to a so-called CA file. The CA file format is described on this page of the OVITO user manual.

Inputs:

Particle properties
Position	The coordinates of the input particles.
Selection	The selection state of the input particles. Only needed if only_selected is True.

Outputs:

Dislocations
DislocationNetwork	The dislocation lines found by the modifier. You can access this data object through the dislocations field of the output DataCollection, see the code example below.

Attributes
DislocationAnalysis.total_line_length	Total length of all dislocation lines found by the DXA (in simulation units).
DislocationAnalysis.length.1/n<ijk>	A set of attributes indicating the length of dislocations broken down by Burgers vector type. For example, the attribute DislocationAnalysis.length.1/6<112> specifies the total line length of Shockley partials found by the DXA.
DislocationAnalysis.length.other	Total length of dislocation lines with an unusual Burgers vector that do not belong to any of the predefined standard dislocation types.
DislocationAnalysis.cell_volume	The volume of the input simulation cell. You can use it to calculate the dislocation density from the line length.
DislocationAnalysis.counts.OTHER	Number of particles not matching any of the known structural types.
DislocationAnalysis.counts.FCC	Number of particles identified as face-centered cubic structure.
DislocationAnalysis.counts.HCP	Number of particles identified as hexagonal close packed structure.
DislocationAnalysis.counts.BCC	Number of particles identified as body-centered cubic structure.
DislocationAnalysis.counts.CubicDiamond	Number of particles identified as cubic diamond structure.
DislocationAnalysis.counts.HexagonalDiamond	Number of particles identified as hexagonal diamond structure.

Particle properties
Structure Type	The structure type assigned to each particle by the CNA sub-algorithm, encoded as an integer value: Value Python constant 0 DislocationAnalysisModifier.Lattice.OTHER 1 DislocationAnalysisModifier.Lattice.FCC 2 DislocationAnalysisModifier.Lattice.HCP 3 DislocationAnalysisModifier.Lattice.BCC 4 DislocationAnalysisModifier.Lattice.CubicDiamond 5 DislocationAnalysisModifier.Lattice.HexagonalDiamond
Color	A per-particle color representing the identified structure type (only if color_by_type is True).
Cluster	The ID of the cluster each atom has been assigned to. A “cluster” is a contiguous group of atoms, all being of the same crystalline structure type. Non-crystalline atoms are assigned to cluster ID 0.

Example:

from ovito.io import import_file, export_file
from ovito.modifiers import DislocationAnalysisModifier
from ovito.data import DislocationNetwork

pipeline = import_file("input/simulation.dump")

# Extract dislocation lines from a crystal with diamond structure:
modifier = DislocationAnalysisModifier()
modifier.input_crystal_structure = DislocationAnalysisModifier.Lattice.CubicDiamond
pipeline.modifiers.append(modifier)
data = pipeline.compute()

total_line_length = data.attributes['DislocationAnalysis.total_line_length']
cell_volume = data.attributes['DislocationAnalysis.cell_volume']
print("Dislocation density: %f" % (total_line_length / cell_volume))

# Print list of dislocation lines:
print("Found %i dislocation segments" % len(data.dislocations.segments))
for segment in data.dislocations.segments:
    print("Segment %i: length=%f, Burgers vector=%s" % (segment.id, segment.length, segment.true_burgers_vector))
    print(segment.points)

# Export dislocation lines to a CA file:
export_file(pipeline, "output/dislocations.ca", "ca")

# Or export dislocations to a ParaView VTK file:
export_file(pipeline, "output/dislocations.vtk", "vtk/disloc")

property circuit_stretchability

The number of steps by which a Burgers circuit can stretch while it is being advanced along a dislocation line.

Default

9

property color_by_type

Controls whether the modifier assigns a color to each particle based on the identified structure type.

Default

True

property defect_mesh_smoothing_level

Specifies the number of iterations of the surface smoothing algorithm to perform when post-processing the extracted defect mesh.

Default

8

property defect_vis

The SurfaceMeshVis element controlling the visual representation of the generated defect mesh.

property disloc_vis

The DislocationVis element controlling the visual representation of the generated dislocation lines.

property input_crystal_structure

The type of crystal to analyze. Must be one of:

• DislocationAnalysisModifier.Lattice.FCC

• DislocationAnalysisModifier.Lattice.HCP

• DislocationAnalysisModifier.Lattice.BCC

• DislocationAnalysisModifier.Lattice.CubicDiamond

• DislocationAnalysisModifier.Lattice.HexagonalDiamond

Default

DislocationAnalysisModifier.Lattice.FCC

property line_coarsening_enabled

Flag that enables the coarsening of extracted dislocation lines, which reduces the number of sample points along the lines.

Default

True

property line_point_separation

Sets the desired distance between successive sample points along the dislocation lines, measured in multiples of the interatomic spacing. This parameter controls the amount of coarsening performed during post-processing of dislocation lines.

Default

2.5

property line_smoothing_enabled

Flag that enables the smoothing of extracted dislocation lines after they have been coarsened.

Default

True

property line_smoothing_level

The number of iterations of the line smoothing algorithm to perform.

Default

1

property only_perfect_dislocations

This flag controls whether the algorithm should extract only perfect dislocations (and no partial dislocations, which is normally done for FCC/HCP and diamond lattices). Make sure you set the circuit_stretchability parameter to a high value when activating this option, because large Burgers circuits are needed to identify dissociated dislocations with a wide core.

Default

False

property only_selected

Set this to True to perform the analysis on selected particles only. Particles that are not selected will be treated as if they did not exist. Use the SelectTypeModifier, for example, to restrict the crystal structure identification to a sub-lattice formed by one species of particles in a multi-component system.

Default

False

property trial_circuit_length

The maximum length of trial Burgers circuits constructed by the DXA to discover dislocations. The length is specified in terms of the number of atom-to-atom steps.

Default

14

class ovito.modifiers.ElasticStrainModifier

Base class

ovito.pipeline.Modifier

This modifier computes the atomic-level elastic strain and deformation gradient tensors in crystalline systems. See also the corresponding user manual page for this modifier.

The modifier first performs an identification of the local crystal structure and stores the results in the Structure Type particle property. Possible structure type values are listed under the input_crystal_structure property. Atoms that do not form a crystalline structure or which are part of defects are assigned the special type OTHER (=0). For these atoms the local elastic deformation cannot be computed.

If calculate_deformation_gradients is set to true, the modifier outputs a new particle property named Elastic Deformation Gradient, which contains the per-atom elastic deformation gradient tensors. Each tensor has nine components stored in column-major order. Atoms for which the elastic deformation gradient could not be determined (i.e. which are classified as OTHER) will be assigned the null tensor.

If calculate_strain_tensors is set to true, the modifier outputs a new particle property named Elastic Strain, which contains the per-atom elastic strain tensors. Each symmetric strain tensor has six components stored in the order XX, YY, ZZ, XY, XZ, YZ. Atoms for which the elastic strain tensor could not be determined (i.e. which are classified as OTHER) will be assigned the null tensor.

Furthermore, the modifier generates a particle property Volumetric Strain, which stores the trace divided by three of the local elastic strain tensor. Atoms for which the elastic strain tensor could not be determined (i.e. which are classified as OTHER) will be assigned a value of zero.

Inputs:

Particle properties
Position	The coordinates of the input particles.
Selection	The selection state of the input particles. Only needed if only_selected is True.

Outputs:

Particle properties
Elastic Deformation Gradient	The computed per-atom elastic deformation gradient tensors. Each tensor consists of 9 components stored in column-major order. All components will be set to zero for atoms for which no elastic deformation tensor could be determined (because they were classified as OTHER). This output property is only generated if calculate_deformation_gradients is set to True.
Elastic Strain	The computed per-atom elastic strain tensor. Each symmetric strain tensor has six components stored in the order XX, YY, ZZ, XY, XZ, YZ. All components will be set to zero for atoms for which no elastic strain tensor could be determined (because they were classified as OTHER). This output property is only generated if calculate_strain_tensors is set to True.
Volumetric Strain	Scalar particle property which is set to one third of the trace of the computed local elastic strain tensor. Atoms for which no elastic strain tensor could be determined (because they were classified as OTHER) will have a volumetric strain value of zero.
Structure Type	The structure type determined by the algorithm for each particle, encoded as an integer value: Value Python constant 0 ElasticStrainModifier.Lattice.OTHER 1 ElasticStrainModifier.Lattice.FCC 2 ElasticStrainModifier.Lattice.HCP 3 ElasticStrainModifier.Lattice.BCC 4 ElasticStrainModifier.Lattice.CubicDiamond 5 ElasticStrainModifier.Lattice.HexagonalDiamond
Color	A per-particle color representing the identified structure type (only if color_by_type is True).
Cluster	The ID of the cluster each atom has been assigned to. A “cluster” is a contiguous group of atoms, all being of the same crystalline structure type. Non-crystalline atoms are assigned the invalid cluster ID 0.

property axial_ratio

The c/a ratio of the ideal unit cell for crystals with hexagonal symmetry.

Default

sqrt(8/3)

property calculate_deformation_gradients

Flag that enables the output of the calculated elastic deformation gradient tensors. The per-particle tensors will be stored in a new particle property named Elastic Deformation Gradient with nine components (stored in column-major order). Particles for which the local elastic deformation cannot be calculated, are assigned the null tensor.

Default

False

property calculate_strain_tensors

Flag that enables the calculation and out of the elastic strain tensors. The symmetric strain tensors will be stored in a new particle property named Elastic Strain with six components (XX, YY, ZZ, XY, XZ, YZ).

Default

True

property input_crystal_structure

The type of crystal to analyze. Must be one of:

• ElasticStrainModifier.Lattice.FCC

• ElasticStrainModifier.Lattice.HCP

• ElasticStrainModifier.Lattice.BCC

• ElasticStrainModifier.Lattice.CubicDiamond

• ElasticStrainModifier.Lattice.HexagonalDiamond

Default

ElasticStrainModifier.Lattice.FCC

property lattice_constant

Lattice constant (a₀) of the ideal unit cell.

Default

1.0

property push_strain_tensors_forward

Selects the frame in which the elastic strain tensors are calculated.

If true, the Eulerian-Almansi finite strain tensor is computed, which measures the elastic strain in the global coordinate system (spatial frame).

If false, the Green-Lagrangian strain tensor is computed, which measures the elastic strain in the local lattice coordinate system (material frame).

Default

True

class ovito.modifiers.ExpandSelectionModifier

Base class

ovito.pipeline.Modifier

Expands the current particle selection by selecting particles that are neighbors of already selected particles. See the corresponding user manual page for more information.

Inputs:

Particle properties
Selection	The selection state of the input particles.
Position	The coordinates of the input particles (only used if mode is Cutoff or Nearest).

Bond properties
Topology	The list of bonds (only used if mode is Bonded).

Outputs:

Particle properties
Selection	The output particle selection.

property cutoff

The maximum distance up to which particles are selected around already selected particles. This parameter is only used if mode is set to ExpansionMode.Cutoff.

Default

3.2

property iterations

Controls how many iterations of the modifier are executed. This can be used to select neighbors of neighbors up to a certain recursive depth.

Default

1

property mode

Selects the mode of operation, i.e., how the modifier extends the selection around already selected particles. Valid values are:

• ExpandSelectionModifier.ExpansionMode.Cutoff

• ExpandSelectionModifier.ExpansionMode.Nearest

• ExpandSelectionModifier.ExpansionMode.Bonded

Default

ExpandSelectionModifier.ExpansionMode.Cutoff

property num_neighbors

The number of nearest neighbors to select around each already selected particle. This parameter is only used if mode is set to ExpansionMode.Nearest.

Default

1

class ovito.modifiers.ExpressionSelectionModifier

Base class

ovito.pipeline.Modifier

Selects data elements based on a user-defined Boolean expression. See also the corresponding user manual page for more information. The modifier can operate on different data elements:

Data element
particles	Selects particles according to a user-defined criterion.
bonds	Selects bonds according to a user-defined criterion.

By default the modifier will act on particles. You can change this by setting the operate_on field.

Outputs:

Attributes
ExpressionSelection.count	The number of data elements (particles/bonds) that have been selected by the modifier.

Example:

# Select all atoms with potential energy above -3.6 eV:
pipeline.modifiers.append(ExpressionSelectionModifier(expression = 'PotentialEnergy > -3.6'))
data = pipeline.compute()
# Demonstrating two ways to get the number of selected atoms:
print(data.attributes['SelectExpression.num_selected'])
print(numpy.count_nonzero(data.particles.selection))

property expression

A string containing the Boolean expression to be evaluated for every element. The expression syntax is documented in OVITO’s user manual.

property operate_on

Selects the kind of data elements this modifier should operate on. Supported values are: 'particles', 'bonds', 'surface_regions'.

Default

'particles'

class ovito.modifiers.FreezePropertyModifier

Base class

ovito.pipeline.Modifier

This modifier copies the values of a source property from some reference animation frame (frame 0 by default) to the current animation frame. It allows preserving a particle or bond property that would otherwise change with time. See also the corresponding user manual page for more information. The modifier can operate on different data elements:

Data element
particles	Freezes a particle property.
bonds	Freezes a bond property.
voxels	Freezes a voxel grid property.

By default the modifier will operate on a particle property. You can change this by setting the operate_on field.

Example:

from ovito.io import import_file, export_file
from ovito.modifiers import (CoordinationAnalysisModifier, FreezePropertyModifier,
                            ExpressionSelectionModifier)

# Load a input simulation sequence. 
pl = import_file("input/simulation.*.dump")

# Add modifier for computing the coordination numbers of particles.
pl.modifiers.append(CoordinationAnalysisModifier(cutoff = 2.9))

# Save the initial coordination numbers from frame 0 under a new name.
modifier = FreezePropertyModifier(source_property = 'Coordination', 
                                  destination_property = 'Coord0',
                                  freeze_at = 0)
pl.modifiers.append(modifier)

# Select all particles whose coordination number has changed since frame 0
# by comapring the dynamically computed coordination numbers with the frozen ones.
pl.modifiers.append(ExpressionSelectionModifier(expression='Coordination != Coord0'))

# Write out number of particles exhibiting a change in coordination number.
export_file(pl, 'output/changes.txt', 'txt', 
            columns = ['Timestep', 'SelectExpression.num_selected'], 
            multiple_frames = True)

property destination_property

The name of the output property that should be created by the modifier. It may be the same as source_property. If the destination property already exists in the modifier’s input, the values are overwritten.

property freeze_at

The animation frame number at which to freeze the input property’s values.

Default

0

property operate_on

Selects the kind of properties this modifier should operate on. Supported values are: 'particles', 'bonds', 'voxels'.

Default

'particles'

property source_property

The name of the input property that should be evaluated by the modifier on the animation frame specified by freeze_at.

class ovito.modifiers.GenerateTrajectoryLinesModifier

Base class

ovito.pipeline.Modifier

This modifier periodically samples the time-dependent positions of particles to produce a TrajectoryLines object. The modifier is typically used to visualize the trajectories of particles as static lines. See the corresponding user manual page for more information.

The generation of trajectory lines must be explicitly triggered by a call to generate() as shown in the following example:

from ovito.io import import_file
from ovito.modifiers import GenerateTrajectoryLinesModifier
from ovito.vis import TrajectoryVis

# Load a particle simulation sequence:
pipeline = import_file('input/simulation.*.dump')

# Insert the modifier into the pipeline for creating the trajectory lines.
modifier = GenerateTrajectoryLinesModifier(only_selected = False)
pipeline.modifiers.append(modifier)

# Now let the modifier generate the trajectory lines by sampling the 
# particle positions over the entire animation interval.
modifier.generate()

# Configure trajectory line visualization:
modifier.vis.width = 0.4
modifier.vis.color = (1,0,0)
modifier.vis.shading = TrajectoryVis.Shading.Flat

# Insert pipeline into the scene to make the particles and 
# the trajectory lines visible in rendered images.
pipeline.add_to_scene()

property frame_interval

The animation frame interval over which the particle positions are sampled to generate the trajectory lines. Set this to a tuple of two integers to specify the first and the last animation frame; or use None to generate trajectory lines over the entire animation sequence.

Default

None

generate()

Generates the trajectory lines by sampling the positions of the particles from the upstream pipeline in regular animation time intervals. Make sure you call this method after the modifier has been inserted into the pipeline.

property only_selected

Controls whether trajectory lines should only by generated for currently selected particles.

Default

True

property sampling_frequency

Length of the animation frame intervals at which the particle positions should be sampled.

Default

1

property unwrap_trajectories

Controls whether trajectory lines should be automatically unwrapped at the box boundaries when the particles cross a periodic boundary.

Default

True

property vis

The TrajectoryVis element controlling the visual appearance of the trajectory lines created by this modifier.

class ovito.modifiers.GrainSegmentationModifier

Base class

ovito.pipeline.Modifier

This analysis modifier identifies the individual grains in a polycrystalline microstructure. See the corresponding user manual page for more information on this modifier.

Inputs:

Particle properties
Position	The input coordinates of the particles.
Structure Type	The local structure types computed by the PolyhedralTemplateMatchingModifier.
Orientation	The local lattice orientations computed by the PolyhedralTemplateMatchingModifier.

Outputs:

Particle properties
Cluster	The numeric identifier of the grains the particles have been assigned to by the algorithm.
Color	A per-particle color representing either the grain (if color_particles is True) or the identified PTM structure type.

Attributes
GrainSegmentation.grain_count	Number of grains that have been found by the algorithm.

DataTable
grains	A data table containing the list of grains found by the algorithm.
grains-merge	A scatter plot of the ordered cluster merge steps computed by the grain segmentation algorithm.

property algorithm

Selects the cluster merge algorithm to use. One of the following constants:

• GrainSegmentationModifier.Algorithm.GraphClusteringAuto

• GrainSegmentationModifier.Algorithm.GraphClusteringManual

• GrainSegmentationModifier.Algorithm.MinimumSpanningTree

Default

GrainSegmentationModifier.Algorithm.GraphClusteringAuto

property color_particles

Controls whether the modifier colors atoms according to the grains they have been assigned to.

Default

True

property handle_stacking_faults

Controls whether the algorithm should handle stacking faults and coherent twin boundaries explicitly. If turned off, atoms with cubic and hexagonal crystal structure will simply be treated as separate phases and will never be merged into the same grain.

Default

True

property merging_threshold

Specifies the maximum graph edge contraction distance and determines the resulting number and sizes of grains. A lower threshold produces more (and smaller) grains; a larger threshold produces fewer (and larger) grains.

This parameter is ignored if algorithm is GraphClusteringAuto, in which case the merging threshold is automatically determined by the algorithm to give optimal segmentation results.

Default

0.0

property min_grain_size

Minimum number of atoms in a grain. Crystallite clusters smaller than this threshold get dissolved during the grain segmentation and their atoms are merged into neighboring grains.

Default

100

property orphan_adoption

Controls the merging of non-crystalline atoms (e.g. grain boundary atoms) into adjacent grains.

Default

True

class ovito.modifiers.HistogramModifier

Base class

ovito.pipeline.Modifier

Generates a histogram of a property, i.e. the distribution of its per-element values. See also the corresponding user manual page for more information. The modifier can operate on different types of data elements:

Data element
particles	Computes the histogram for a particle property.
bonds	Computes the histogram for a bond property.
voxels	Computes the histogram for a voxel grid property.

By default the modifier will operate on a particle property. You can change this by setting the operate_on field.

The value range of the histogram is determined automatically from the minimum and maximum values of the selected property unless fix_xrange is set to True. In this case the range of the histogram is controlled by the xrange_start and xrange_end parameters.

Example:

from ovito.modifiers import HistogramModifier
from ovito.io import import_file, export_file

pipeline = import_file("input/simulation.dump")

modifier = HistogramModifier(bin_count=100, property='peatom')
pipeline.modifiers.append(modifier)

export_file(pipeline, "output/histogram.txt", "txt/table", key="histogram[peatom]")

property bin_count

The number of histogram bins.

Default

200

property fix_xrange

Controls how the value range of the histogram is determined. If set to False, the interval and bin widths are chosen automatically by the modifier to include all input values. If set to True, the histogram’s value range may be specified explicitly by setting xrange_start and xrange_end.

Default

False

property only_selected

If True, the histogram is computed only on the basis of currently selected particles or bonds. You can use this to restrict histogram calculation to a subset of input particles/bonds.

Default

False

property operate_on

Selects the kind of data elements this modifier should operate on. Supported values are: 'particles', 'bonds', 'voxels'.

Default

'particles'

property property

The name of the input property for which to compute the histogram. For vector properties a component name must be appended in the string, e.g. "Velocity.X".

Default

''

property xrange_end

If fix_xrange is true, then this parameter controls the upper end of the value interval covered by the histogram. Input values higher than this range limit will be discarded from the computed histogram.

Default

0.0

property xrange_start

If fix_xrange is true, then this parameter controls the lower end of the value interval covered by the histogram.Input values lower than this range limit will be discarded from the computed histogram.

Default

0.0

class ovito.modifiers.IdentifyDiamondModifier

Base class

ovito.modifiers.StructureIdentificationModifier

Analyzes the local neighborhood of each particle to identify cubic or hexagonal diamond lattices. See the corresponding user manual page for more information.

Note that this class inherits several important parameter fields from its StructureIdentificationModifier base class.

Modifier inputs:

Particle properties
Position	The input coordinates of the particles.
Selection	The selection state of the input particles. Only needed if only_selected is True.

Modifier outputs:

Particle properties
Structure Type	The structure type computed by the algorithm for each particle, encoded as an integer value: Numeric value Python constant 0 IdentifyDiamondModifier.Type.OTHER 1 IdentifyDiamondModifier.Type.CUBIC_DIAMOND 2 IdentifyDiamondModifier.Type.CUBIC_DIAMOND_FIRST_NEIGHBOR 3 IdentifyDiamondModifier.Type.CUBIC_DIAMOND_SECOND_NEIGHBOR 4 IdentifyDiamondModifier.Type.HEX_DIAMOND 5 IdentifyDiamondModifier.Type.HEX_DIAMOND_FIRST_NEIGHBOR 6 IdentifyDiamondModifier.Type.HEX_DIAMOND_SECOND_NEIGHBOR
Color	Particle coloring to indicate the identified structure type for each particle; only if color_by_type is True. See the structures array on how to customize the colors.

Attributes
IdentifyDiamond.counts.OTHER	Number of particles not matching any of the recognized structure types.
IdentifyDiamond.counts.CUBIC_DIAMOND	Number of particles identified as fully coordinated cubic diamond.
IdentifyDiamond.counts.CUBIC_DIAMOND_FIRST_NEIGHBOR	Number of particles identified as partially coordinated cubic diamond.
IdentifyDiamond.counts.CUBIC_DIAMOND_SECOND_NEIGHBOR	Number of particles identified as partially coordinated cubic diamond.
IdentifyDiamond.counts.HEX_DIAMOND	Number of particles identified as fully coordinated hexagonal diamond.
IdentifyDiamond.counts.HEX_DIAMOND_FIRST_NEIGHBOR	Number of particles identified as partially coordinated hexagonal diamond.
IdentifyDiamond.counts.HEX_DIAMOND_SECOND_NEIGHBOR	Number of particles identified as partially coordinated hexagonal diamond.

Data tables
structures	A bar chart with the particle counts for each structure type identified by the modifier. You can retrieve this DataTable from the DataCollection.tables dictionary.

class ovito.modifiers.InvertSelectionModifier

Base class

ovito.pipeline.Modifier

This modifier inverts the current data element selection. See also the corresponding user manual page for more information. The modifier can operate on different kinds of data elements:

Data element
particles	Inverts the values of the Selection particle property.
bonds	Inverts the values of the Selection bond property.
voxels	Inverts the values of the Selection voxel grid property.

By default the modifier will act on particles. You can change this by setting the operate_on field.

property operate_on

Selects the kind of data elements this modifier should operate on. Supported values are: 'particles', 'bonds', 'voxels'.

Default

'particles'

class ovito.modifiers.LoadTrajectoryModifier

Base class

ovito.pipeline.Modifier

This modifier loads trajectories of particles from a separate simulation file. See also the corresponding user manual page for this modifier.

The typical use case for this modifier is when the topology of a molecular system (i.e. the definition of atom types, bonds, etc.) is stored separately from the trajectories of atoms. In this case you should load the topology file first using import_file(). Then create and apply the LoadTrajectoryModifier to the topology dataset, which loads the trajectory file. The modifier will replace the static atom positions from the topology dataset with the time-dependent positions from the trajectory file.

Example:

from ovito.io import import_file
from ovito.modifiers import LoadTrajectoryModifier

# Load static topology data from a LAMMPS data file.
pipeline = import_file('input/input.data', atom_style='bond')

# Load atom trajectories from separate LAMMPS dump file.
traj_mod = LoadTrajectoryModifier()
traj_mod.source.load('input/trajectory.dump')
print("Number of frames: ", traj_mod.source.num_frames)

# Insert modifier into data pipeline.
pipeline.modifiers.append(traj_mod)

Furthermore, it is possible to let the modifier load varying bond topologies for each trajectory frame from a LAMMPS dump local file:

bonds_mod = LoadTrajectoryModifier()
bonds_mod.source.load('input/bonds.dump.local', 
    columns = [None, 'Bond Type', 'Topology.A', 'Topology.B', 'Length', 'Energy'])
pipeline.modifiers.append(bonds_mod)

Here, the columns function parameter specifies the mapping of data columns in the imported dump file to corresponding target bond properties within OVITO. The dump local file bonds.dump.local contains six data columns and has been produced by the following LAMMPS commands:

compute 1 all property/local btype batom1 batom2
compute 2 all bond/local dist engpot
dump bonds all local 100 bonds.dump.local index c_1[1] c_1[2] c_1[3] c_2[1] c_2[2]

property source

A FileSource that provides the trajectories of particles. You can call its load() function to load a simulation trajectory file as shown in the code example above.

class ovito.modifiers.PolyhedralTemplateMatchingModifier

Base class

ovito.modifiers.StructureIdentificationModifier

This modifier analyzes the local neighborhood of each particle to identify common structural motives and crystalline structures. The structure identification is based on the Polyhedral Template Matching (PTM) algorithm. See the corresponding user manual page for more information. The PTM algorithm is able to compute local crystal orientations, elastic lattice strains, and can identify local chemical orderings in binary compounds.

Note that this modifier inherits several important parameter fields from the StructureIdentificationModifier base class.

Modifier inputs:

Particle properties
Position	The input coordinates of the particles.
Particle Type	The chemical types of the input particles; only used if output_ordering is True.
Selection	The selection state of the input particles; only used if only_selected is True.

Modifier outputs:

Particle properties
Structure Type	The structure type computed by the algorithm for each particle, encoded as an integer value: Numeric Value Python constant Initial state 0 PolyhedralTemplateMatchingModifier.Type.OTHER 1 PolyhedralTemplateMatchingModifier.Type.FCC enabled 2 PolyhedralTemplateMatchingModifier.Type.HCP enabled 3 PolyhedralTemplateMatchingModifier.Type.BCC enabled 4 PolyhedralTemplateMatchingModifier.Type.ICO disabled 5 PolyhedralTemplateMatchingModifier.Type.SC disabled 6 PolyhedralTemplateMatchingModifier.Type.CUBIC_DIAMOND disabled 7 PolyhedralTemplateMatchingModifier.Type.HEX_DIAMOND disabled 8 PolyhedralTemplateMatchingModifier.Type.GRAPHENE disabled The algorithm only identifies enabled structure types; see structures array for details.
RMSD	The per-particle RMSD values computed by the PTM algorithm. Only if output_rmsd is set.
Interatomic Distance	The per-particle local atomic distances computed by the PTM algorithm. Only if output_interatomic_distance is set.
Orientation	The local lattice orientations computed by the PTM algorithm, encoded as quaternions. Only if output_orientation is set.
Elastic Deformation Gradient	The per-particle elastic deformation gradient tensors computed by the PTM algorithm (3x3 components). Only if output_deformation_gradient is set.
Ordering Type	The local chemical ordering type determined by the PTM algorithm, encoded as an integer value. Only if output_ordering is set. Numeric value Python constant 0 PolyhedralTemplateMatchingModifier.OrderingType.OTHER 1 PolyhedralTemplateMatchingModifier.OrderingType.PURE 2 PolyhedralTemplateMatchingModifier.OrderingType.L10 3 PolyhedralTemplateMatchingModifier.OrderingType.L12_A 4 PolyhedralTemplateMatchingModifier.OrderingType.L12_B 5 PolyhedralTemplateMatchingModifier.OrderingType.B2 6 PolyhedralTemplateMatchingModifier.OrderingType.ZINCBLENDE_WURTZITE 7 PolyhedralTemplateMatchingModifier.OrderingType.BORON_NITRIDE
Color	Particle coloring to indicate the identified structure type for each particle; only if color_by_type is True. See the structures array on how to customize the colors.

Attributes
PolyhedralTemplateMatching.counts.OTHER	Number of particles not matching any of the recognized structural types.
PolyhedralTemplateMatching.counts.FCC	Number of particles identified as face-centered cubic structure.
PolyhedralTemplateMatching.counts.HCP	Number of particles identified as hexagonal close packed structure.
PolyhedralTemplateMatching.counts.BCC	Number of particles identified as body-centered cubic structure.
PolyhedralTemplateMatching.counts.ICO	Number of particles identified as icosahedral structure.
PolyhedralTemplateMatching.counts.SC	Number of particles identified as simple cubic structure.
PolyhedralTemplateMatching.counts.CUBIC_DIAMOND	Number of particles identified as cubic diamond structure.
PolyhedralTemplateMatching.counts.HEX_DIAMOND	Number of particles identified as hexagonal diamond structure.
PolyhedralTemplateMatching.counts.GRAPHENE	Number of particles identified as 2d graphene structure.

Data tables
structures	A bar chart with the particle counts for each structure type identified by the modifier. You can retrieve this DataTable from the DataCollection.tables dictionary.
ptm-rmsd	A histogram with the RMSD value distribution computed by the modifier. You can retrieve this DataTable from the DataCollection.tables dictionary.

property output_deformation_gradient

Boolean flag that controls whether the modifier outputs the computed per-particle elastic deformation gradients as a new particle property named Elastic Deformation Gradient.The elastic deformation gradient describes the local deformation and rigid-body rotation of the crystal with repect to an ideal reference lattice configuration. See the OVITO user manual for details.

Default

False

property output_interatomic_distance

Boolean flag that controls whether the modifier outputs the computed per-particle interatomic distance as a new particle property named Interatomic Distance.

Default

False

property output_ordering

Boolean flag that controls whether the modifier should identify local ordering types and output them as a new particle property named Ordering Type.

Default

False

property output_orientation

Boolean flag that controls whether the modifier outputs the computed per-particle lattice orientations as a new particle property named Orientation. The lattice orientation is specified in terms of a quaternion that describes the rotation of the crystal with repect to a reference lattice orientation. See the OVITO user manual for details.

Default

False

property output_rmsd

Boolean flag that controls whether the modifier outputs the computed per-particle RMSD values as a new particle property named RMSD.

Default

False

property rmsd_cutoff

The maximum allowed root mean square deviation for positive structure matches. If this threshold value is non-zero, template matches that yield a RMSD value above the cutoff are classified as “Other”. This can be used to filter out spurious template matches (false positives).

Default

0.1

class ovito.modifiers.PythonScriptModifier

Base class

ovito.pipeline.Modifier

Allows you to insert a user-defined Python function into a data pipeline.

This class makes it possible to implement new modifier functions in the Python language which can participate in OVITO’s data pipeline system and which can be used just like OVITO’s built-in modifiers. You can learn more about the usage of this class in the User-defined modifiers section.

Example:

from ovito.io import import_file

# Load some input data:
pipeline = import_file("input/simulation.dump")

# Define our custom modifier function, which assigns a uniform color
# to all particles, similar to the built-in AssignColorModifier.
def assign_color(frame, data):
    color_property = data.particles_.create_property('Color')
    color_property[:] = (0.2, 0.5, 1.0)

# Insert the user-defined modifier function into the data pipeline.
pipeline.modifiers.append(assign_color)
# Note that appending the Python function to the pipeline is equivalent to
# creating a PythonScriptModifier instance and appending it:
#
#   pipeline.modifiers.append(PythonScriptModifier(function = assign_color))

# Evaluate data pipeline. This will make the system invoke assign_color().
data = pipeline.compute()
print(data.particles.colors[...])

property function

The Python function to be called each time the data pipeline is evaluated by the system.

The function must have a signature as shown in the example above. The frame parameter specifies the current animation frame number at which the data pipeline is being evaluated. The DataCollection data initially holds the input data objects of the modifier, which were produced by the upstream part of the data pipeline. The user-defined modifier function is free modify the data collection and the data objects stored in it.

Default

None

class ovito.modifiers.ReplicateModifier

Base class

ovito.pipeline.Modifier

This modifier replicates all particles, bonds and other elements of a system to visualize periodic images. See also the corresponding user manual page for more information.

Inputs:

The modifier can operate on any combination of the following data elements:

Data element
particles	Duplicates Particles and Bonds.
surfaces	Duplicates the mesh geometry of SurfaceMesh objects.
voxels	Duplicates the voxel elements of VoxelGrid objects.
dislocations	Duplicates dislocation lines in a DislocationNetwork.

By default the modifier will operate on all of these. You can restrict it to a subset by setting the operate_on field.

property adjust_box

Controls whether the simulation cell is resized. If True, the simulation cell is accordingly extended to fit the replicated data. If False, the original simulation cell size (containing only the primary image of the system) is maintained.

Default

True

property num_x

A positive integer specifying the number of copies to generate in the x direction (including the existing primary image).

Default

1

property num_y

A positive integer specifying the number of copies to generate in the y direction (including the existing primary image).

Default

1

property num_z

A positive integer specifying the number of copies to generate in the z direction (including the existing primary image).

Default

1

property operate_on

A set of strings specifying the kinds of data elements this modifier should operate on. By default the set contains all data element types supported by the modifier.

Default

{'particles', 'voxels', 'surfaces', 'dislocations'}

property unique_ids

If True, the modifier automatically generates new unique IDs for each copy of particles. Otherwise, the replica will keep the same IDs as the original particles, which is typically not what you want.

Note: This option has no effect if the input particles do not already have numeric IDs (i.e. the 'Particle Identifier' property does not exist).

Default

True

class ovito.modifiers.SelectTypeModifier

Base class

ovito.pipeline.Modifier

Selects data elements of a certain type or types (e.g. all atoms of a chemical species). See also the corresponding user manual page for more information. The modifier can operate on different data elements:

Data element
particles	Selects all particles of a certain type.
bonds	Selects all bonds of a certain type.

By default the modifier will act on particles. You can change this by setting the operate_on field.

Outputs:

Attributes
SelectType.num_selected	The number of data elements (particles/bonds) that have been selected by the modifier.

Example:

from ovito.io import import_file
from ovito.modifiers import SelectTypeModifier, CommonNeighborAnalysisModifier

pipeline = import_file("input/simulation.dump")

# Insert a CNA modifier to determine the structural type of each atom:
pipeline.modifiers.append(CommonNeighborAnalysisModifier())

# Apply the SelectTypeModifier to select all atoms of FCC and HCP type:
pipeline.modifiers.append(SelectTypeModifier(
    operate_on = "particles",
    property = "Structure Type",
    types = { CommonNeighborAnalysisModifier.Type.FCC,
              CommonNeighborAnalysisModifier.Type.HCP }
))

# The SelectTypeModifier reports the number of selected elements as an attribute:
data = pipeline.compute()
print("Number of FCC/HCP atoms: %i" % data.attributes['SelectType.num_selected'])

property operate_on

Selects the kind of data elements this modifier should select. Supported values are: 'particles', 'bonds'.

Default

'particles'

property property

The name of the property to use as input; must be an integer property.

For selecting particles, possible input properties are 'Particle Type' and 'Structure Type', for example. For selecting bonds, 'Bond Type' is a typical input property.

Default

'Particle Type'

property types

The set of types to select. You can add numeric type IDs or type names to this set. Type names will automatically be translated into corresponding numeric type IDs by the modifier. Thus, it is not necessary for you to look up the numeric ID for a type name using Property.type_by_name(). For example, to select all atoms belonging to the type named ‘Cu’:

modifier = SelectTypeModifier(property = 'Particle Type', types = {'Cu'})

When using the SelectTypeModifier to select structural types, you can directly use the predefined numeric constants for the structures, e.g.:

# Let the CNA modifier identify the structural type of each particle:
pipeline.modifiers.append(CommonNeighborAnalysisModifier())
# Select all FCC and HCP particles:
modifier = SelectTypeModifier(property = 'Structure Type')
modifier.types = { CommonNeighborAnalysisModifier.Type.FCC,
                   CommonNeighborAnalysisModifier.Type.HCP }
pipeline.modifiers.append(modifier)

Default

set([])

class ovito.modifiers.SliceModifier

Base class

ovito.pipeline.Modifier

Deletes or selects data elements located within a semi-infinite region bounded by a plane or within a slab bounded by a pair of parallel planes. See also the corresponding user manual page for more information.

Inputs:

The modifier can operate on any combination of the following data elements:

Data element
particles	Deletes (or selects) particles.
surfaces	Cuts the mesh geometry of a SurfaceMesh.
dislocations	Cuts dislocation lines of a DislocationNetwork.

By default the modifier will act on all of these. You can restrict it to a subset by setting the operate_on field. Furthermore, you can restrict the operation to only selected particles by setting the only_selected option.

The select option lets you select all elements on one side of the plane instead of deleting them. Currently, the selection mode works only for Particles or SurfaceMesh vertices, which get selected by setting their Selection particle property to 1.

property distance

The distance of the slicing plane from the origin (along its normal vector).

Default

0.0

property inverse

Reverses the sense of the slicing plane.

Default

False

property normal

The normal vector of the slicing plane. Does not have to be a unit vector.

Default

(1,0,0)

property only_selected

Controls whether the modifier should act only on currently selected data elements (e.g. selected particles).

Default

False

property operate_on

A set of strings specifying the kinds of data elements this modifier should operate on. By default the set contains all data element types supported by the modifier.

Default

{'particles', 'surfaces', 'dislocations'}

property plane_vis

A TriangleMeshVis object controlling the visual appearance of the cutting plane in rendered images. It is only used when render_plane has been set to True.

property render_plane

Controls whether the modifier should produce renderable geometry to visualize the cutting plane. The visual appearance of the plane can be adjusted through the plane_vis element.

Default

False

property select

If True, the modifier selects data elements instead of deleting them.

Default

False

property slab_width

The thicknes of the slab to cut. If zero, the modifier cuts away everything on one side of the cutting plane.

Default

0.0

class ovito.modifiers.SmoothTrajectoryModifier

Base class

ovito.pipeline.Modifier

This modifier smoothens the particle motion by averaging the particle positions from several successive snapshots of a simulation trajectory. It can be used to create smoothly looking animations from a relatively coarse sequence of simulation snapshots and reduce fluctuations or thermal vibrations in particle trajectories. See also the corresponding user manual page for this modifier.

Note: Make sure you insert this modifier at the beginning of a data pipeline, in particular in front of any other modifiers that delete or filter particles, because it requires access to the complete particle system in order to perform the trajectory smoothing calculation.

property minimum_image_convention

If this option is set, the modifier will automatically detect when particles cross a simulation box boundary in between two successive simulation frames and computes the unwrapped displacements correctly. You should leave this option activated unless the input particle coordinates are already in unwrapped form.

Default

True

property window_size

Controls how many input animation frames to take into account when calculating the time-averaged particle coordinates. The modifier uses a sliding time window of the given size that is centered around the current animation time. A window size of 1 disables smoothing.

Default

1

class ovito.modifiers.SpatialBinningModifier

Base class

ovito.pipeline.Modifier

This modifier maps the particles to a bin grid and applies a reduction operation (mean, sum, min, max, etc.) to some property of the particles located within each spatial bin. The modifier outputs a one-, two- or three-dimensional grid of field values. See also the corresponding user manual page for more information on this modifier.

The direction parameter of the modifier selects the dimensionality of the bin grid and the spatial direction(s) along which the grid cells are oriented. The bin_count parameter controls the resolution of the generated grid, which always covers the entire SimulationCell. Finally, you have to specify the input particle property which should be transferred to the grid cell by setting the modifier’s property field to the name of an existing particle property.

If the dimensionality of the selected output grid is one-dimensional, then the modifier will generate a DataTable object containing the computed bin values along the selected spatial direction. The data table may later be retrieved from the tables dictionary of the pipeline’s output DataCollection under the look-up key 'binning[<PROPERTY>]', where <PROPERTY> is the name of the input particle property from which the bin values have been calculated.

If the dimensionality of the selected output grid is two- or three-dimensional, then the modifier will generate a VoxelGrid object. The generated voxel grid may later be retrieved from the grids dictionary of the pipeline’s output DataCollection under the look-up key 'binning[<PROPERTY>]', where <PROPERTY> is the name of the input particle property for which the grid was built.

Examples

The following example shows how to calculate the gradient of the Velocity.X field along the spatial Z axis. For this, we use the first_derivative option, which lets the modifier compute the first derivative of the data points using a finite differences scheme.

pipeline.modifiers.append(SpatialBinningModifier(
    property = 'Velocity.X',
    direction = SpatialBinningModifier.Direction.Z, 
    bin_count = 80,
    reduction_operation = SpatialBinningModifier.Operation.Mean,
    first_derivative = True
))

The output DataTable can then be exported to an output file using OVITO’s export_file() function or printed to the console:

export_file(pipeline, 'output/velocity_profile.txt', 'txt/table', key='binning[Velocity.X]')
data = pipeline.compute()
print(data.tables['binning[Velocity.X]'].xy())

The following example demonstrates how to compute a three-dimensional VoxelGrid of the particle number density. Since the SpatialBinningModifier always requires an input particle property, we first employ the ComputePropertyModifier to give all particles a new property having the uniform value 1, which can then serve as input property for computing the number density on the grid:

pipeline.modifiers.append(ComputePropertyModifier(expressions=['1'], output_property='Unity'))

pipeline.modifiers.append(SpatialBinningModifier(
    property = 'Unity',
    direction = SpatialBinningModifier.Direction.XYZ, 
    bin_count = (100, 100, 100),
    reduction_operation = SpatialBinningModifier.Operation.SumVol
))

The resulting VoxelGrid may now be exported to an output file, for example in the VTK format:

export_file(pipeline, 'output/density.vtk', 'vtk/grid', key='binning[Unity]')

Or it can serve as input for subsequent modifiers in your data pipeline, for example the CreateIsosurfaceModifier.

property bin_count

Specifies the number of bin cells to generate along each axis of the binning grid. You should assign a tuple containing one, two, or three positive integers to this parameter field, depending on the grid’s dimensionality set by the direction parameter.

Note that the entries in the tuple specify the number of bins along the grid’s first, second, and third dimension and not along the spatial axes. For example, if the binning direction is Direction.YZ, setting bin_count to (100, 50) will let the modifier generate a two-dimensional grid with 100 bins along the second simulation cell vector (spatial y-axis) and 50 bins along the third cell vector (z-axis) of the 3-dimensional simulation box. Examples:

SpatialBinningModifier(direction=SpatialBinningModifier.Direction.Z, bin_count=100)
SpatialBinningModifier(direction=SpatialBinningModifier.Direction.XZ, bin_count=(40, 80))
SpatialBinningModifier(direction=SpatialBinningModifier.Direction.XYZ, bin_count=(80, 80, 80))

If you assign just a single number, or a tuple with fewer entries than required for the selected grid dimensionality, the bin count in any of the remaining grid dimensions will be implicitly set to 1.

Default

(200, 200, 200)

property direction

Selects the alignment of the bins and the dimensionality of the grid. Possible values:

• SpatialBinningModifier.Direction.X

• SpatialBinningModifier.Direction.Y

• SpatialBinningModifier.Direction.Z

• SpatialBinningModifier.Direction.XY

• SpatialBinningModifier.Direction.XZ

• SpatialBinningModifier.Direction.YZ

• SpatialBinningModifier.Direction.XYZ

For modes X, Y, and Z, the modifier will generate a one-dimensional grid with bins aligned perpendicular to the selected simulation cell vector. For modes XY, XZ, and YZ, the modifier will generate a two-dimensional grid with bins aligned perpendicular to both selected simulation cell vectors (i.e. parallel to the third vector). In the last case (XYZ), the modifier generates a three-dimensional voxel grid.

Default

SpatialBinningModifier.Direction.X

property first_derivative

Set this to True to let the modifier numerically compute the first derivative of the binned data points using a finite differences approximation. This works only if binning is performed in a single direction (Direction.X, Direction.Y or Direction.Z).

Default

False

property only_selected

This option lets the modifier take into account only the currently selected particles. Unselected particles will be excluded from the binning process. You can use this option to restrict the calculation to a subset of particles.

Default

False

property property

The name of the input particle Property which the reduction operation should be applied to. This can be one of the standard particle properties or a some user-defined particle property. For vector properties, a specific vector component must be appended, e.g. "Velocity.X", because the modifier only supports the calculation of scalar field in the current program version.

property reduction_operation

Selects the reduction operation to be carried out. Supported parameter values are:

• SpatialBinningModifier.Operation.Mean

• SpatialBinningModifier.Operation.Sum

• SpatialBinningModifier.Operation.SumVol

• SpatialBinningModifier.Operation.Min

• SpatialBinningModifier.Operation.Max

The operation SumVol first computes the sum and then divides the result by the volume of the respective bin. It is intended to compute pressure (or stress) within each bin from the per-atom virial.

Default

SpatialBinningModifier.Operation.Mean

class ovito.modifiers.SpatialCorrelationFunctionModifier

Base class

ovito.pipeline.Modifier

This modifier calculates the spatial correlation function between two particle properties. See also the corresponding user manual page for this modifier.

The algorithm uses the FFT to compute the convolution. It then computes a radial average in reciprocal and real space. This gives the correlation function up to half of the cell size. The modifier can additionally compute the short-ranged part of the correlation function from a direct summation over neighbors.

Usage example:

from ovito.modifiers import SpatialCorrelationFunctionModifier
from ovito.io import import_file, export_file

pipeline = import_file("input/simulation.dump")

mod = SpatialCorrelationFunctionModifier(property1='Position.X', property2='peatom')
pipeline.modifiers.append(mod)
data = pipeline.compute()

# Export RDF and correlation functions to text files:
C = data.tables['correlation-real-space']
rdf = data.tables['correlation-real-space-rdf']
export_file(C, 'output/real_correlation_function.txt', 'txt/table')
export_file(rdf, 'output/rdf.txt', 'txt/table')

# Compute normalized correlation function (by co-variance):
C_norm = C.xy()
mean1 = data.attributes['CorrelationFunction.mean1']
mean2 = data.attributes['CorrelationFunction.mean2']
covariance = data.attributes['CorrelationFunction.covariance']
C_norm[:,1] = (C_norm[:,1] - mean1*mean2) / (covariance - mean1*mean2)
import numpy
numpy.savetxt('output/normalized_real_correlation_function.txt', C_norm)

property apply_window

This flag controls whether nonperiodic directions have a Hann window applied to them. Applying a window function is necessary to remove spurios oscillations and power-law scaling of the (implicit) rectangular window of the nonperiodic domain.

Default

True

property direct_summation

Flag controlling whether the real-space correlation plot will show the result of a direct calculation of the correlation function, obtained by summing over neighbors.

Default

False

property grid_spacing

Controls the approximate size of the FFT grid cell. The actual size is determined by the distance of the simulation cell faces which must contain an integer number of grid cells.

Default

3.0

property neighbor_bins

This integer value controls the number of bins for the direct calculation of the real-space correlation function.

Default

50

property neighbor_cutoff

This parameter determines the cutoff of the direct calculation of the real-space correlation function.

Default

5.0

property property1

The name of the first input particle property for which to compute the correlation, P1. For vector properties a component name must be appended in the string, e.g. "Velocity.X".

Default

''

property property2

The name of the second particle property for which to compute the correlation, P2. If this is the same as property1, then the modifier will compute the autocorrelation.

Default

''

class ovito.modifiers.StructureIdentificationModifier

Base class

ovito.pipeline.Modifier

Abstract base class for all modifiers in OVITO that analyze the local neighborhood of particles to identify structural motives or crystalline structures. It contains parameter fields that are common to all these modifiers:

• AcklandJonesModifier

• ChillPlusModifier

• CommonNeighborAnalysisModifier

• IdentifyDiamondModifier

• PolyhedralTemplateMatchingModifier

• VoroTopModifier

property color_by_type

Controls the coloring of particles by the modifier to indicate their identified structure type. See the documentation of the structures list on how to change the colors representing the different structure types recognized by the modifier.

Default

True

property only_selected

Set this to True to perform the analysis on selected particles only. Particles that are not selected will be treated as if they did not exist and are assigned to the “OTHER” structure category. Use a SelectTypeModifier in your pipeline, for example, to restrict the structure identification to a sub-lattice formed by one species of particles in a multi-component system.

Default

False

property structures

A list of ElementType instances managed by this modifier, one for each recognized structural type. You can modify the type objects in this list to adjust the coloring and to turn the identification of certain structural types on or off. The ordering of the ParticleType objects in this list corresponds to the numeric type IDs defined by the concrete structure identification modifiers. In the following code snippets, the CommonNeighborAnalysisModifier serves as an example.

Your can change the color of a type by setting its color property to a new RGB value:

modifier = CommonNeighborAnalysisModifier()
modifier.structures[CommonNeighborAnalysisModifier.Type.FCC].color = (0.2, 1.0, 0.8)
modifier.structures[CommonNeighborAnalysisModifier.Type.HCP].color = (0.0, 0.4, 1.0)

To turn the identification of a particular structure type on or off, you set its enabled property:

modifier.structures[CommonNeighborAnalysisModifier.Type.BCC].enabled = False
modifier.structures[CommonNeighborAnalysisModifier.Type.ICO].enabled = True

.

class ovito.modifiers.TimeAveragingModifier

Base class

ovito.pipeline.Modifier

This modifier can compute the time average of some time-dependent input quantity by sampling it over all frames of the loaded trajectory. The input quantity to be averaged can be any time-varying attribute or DataTable generated on a frame-by-frame basis by another modifier in the OVITO data pipeline. See also the corresponding user manual page for more information on this modifier.

Averaging scalar quantities:

The following code example demonstrates how to use the TimeAveragingModifier to calculate the mean value of a scalar quantity that changes with simulation time. In this example, the CreateBondsModifier is added to the pipeline to dynamically create bonds between the moving particles (thus, the number of created bonds will change with time). The modifier reports the number of bonds that have been created in the current simulation frame as an output attribute named 'CreateBonds.num_bonds'. This varying attribute is then selected as the input quantity to be averaged by the following TimeAveragingModifier, which in turn outputs the time-averaged value it cmoputes as a new attribute named 'CreateBonds.num_bonds (Average)'. This value can be retrieved from the DataCollection produced by Pipeline.compute(). Note that is does not matter at which simulation time we evaluate the data pipeline – the mean number of bonds is a static quantity and remains constant over the entire trajectory.

pipeline = import_file('input/trajectory.dump')
pipeline.modifiers.append(CreateBondsModifier(cutoff=2.9))
pipeline.modifiers.append(TimeAveragingModifier(operate_on='attribute:CreateBonds.num_bonds'))
data = pipeline.compute()
print(data.attributes['CreateBonds.num_bonds (Average)'])

Averaging data tables:

Some analysis modifiers of OVITO output their results as DataTable objects, which can represent histograms or distributions functions. For example, the CoordinationAnalysisModifier computes the radial pair distribution function (RDF) of a particle system for the current simulation time and outputs it as a data table named 'coordination-rdf'. We can have the TimeAveragingModifier compute the time average of this varying distribution function over all frames of the loaded trajectory:

pipeline = import_file('input/trajectory.dump')
pipeline.modifiers.append(CoordinationAnalysisModifier(cutoff=6.0, number_of_bins=50))
pipeline.modifiers.append(TimeAveragingModifier(operate_on='table:coordination-rdf'))
data = pipeline.compute()
print(data.tables['coordination-rdf[average]'].xy())

Note that, in case of data tables, the output table’s identifier is given the suffix [average] by the TimeAveragingModifier.

An average of a DataTable can only be computed if the table’s x-axis interval does not vary with time and the x-coordinates of the data points remain fixed. That’s because the modifier simply averages the time-varying y-coordinate of each data point. Thus, to generate a histogram DataTable using the HistogramModifier that is suitable for time averaging, you will have to activate the HistogramModifier.fix_xrange option.

Averaging properties:

The modifier can time-average properties that belong to a Particles object, VoxelGrid, or other type of PropertyContainer. When setting the operate_on field, you need to specify the container’s identifier and the name of the property to average:

pipeline.modifiers.append(TimeAveragingModifier(operate_on='property:particles/Coordination'))
data = pipeline.compute()
print(data.particles['Coordination Average'][...])

Note that the output Property is given the suffix 'Average' by the TimeAveragingModifier.

property interval

The animation frame interval over which the input trajectory is sampled to compute the average. You can set this to a pair of integers specifying the first and the last frame of the averaging interval; or assign None to let the modifier compute the average over the entire trajectory.

For example, to restrict the average to the second half of the loaded simulation trajectory, you can specify the frame interval based on the FileSource.num_frames value:

modifier.interval = (pipeline.source.num_frames//2, pipeline.source.num_frames-1)

Default

None

property operate_on

Selects the input quantity to be averaged by this modifier. Supported values for this field are:

• 'attribute:<NAME>'

• 'table:<ID>'

• 'property:<CONTAINER>/<PROPERTY>'

Here, <NAME> refers to the name of a global attribute to be averaged. <ID> is the identifier of the DataTable to be averaged, <CONTAINER> is the identifier of a PropertyContainer and <PROPERTY> the name of the Property in that container, which should be averaged. Note that the Particles property container has the standard identifier 'particles'. Its Bonds child container is referenced by the hierarchical identifier 'particles/bonds'.

Default

''

property sampling_frequency

Animation step interval at which the quantity to be averaged is sampled from the input trajectory. You can set this to a larger value to perform a coarser sampling and reduce the total number of trajectory frames that need to be loaded into memory.

Default

1

class ovito.modifiers.UnwrapTrajectoriesModifier

Base class

ovito.pipeline.Modifier

This modifier determines when particles cross through the periodic boundaries of the simulation cell and unwraps the particle coordinates in order to make the trajectories continuous. As a result of this operation, particle trajectories will no longer fold back into the simulation cell and instead lead outside the cell.

Note that, to unwrap the particle coordinates, the modifier may have to step through all frames of the input simulation trajectory to detect jumps in the trajectories. This will not be necessary, however, if the Periodic Image particle property has been loaded from the input simulation file, because then the modifier can directly use this information to unwrap the particle coordinates.

from ovito.io import import_file
from ovito.modifiers import UnwrapTrajectoriesModifier

# Load a simulation trajectory:
pipeline = import_file('input/simulation.*.dump')

# Insert the unwrap modifier into the pipeline. 
# Note that the modifier should typically precede any other modifiers in the pipeline. 
pipeline.modifiers.append(UnwrapTrajectoriesModifier())

# For demonstration purposes, request last frame of the trajectory. 
# The returned DataCollection will contain modified particle coordinates, which are computed 
# by the unwrap modifier by tracing the trajectories of the particles and unfolding
# them whenever crossings of the periodic cell boundaries are detected.
data = pipeline.compute(pipeline.source.num_frames - 1)

class ovito.modifiers.VoroTopModifier

Base class

ovito.modifiers.StructureIdentificationModifier

“This modifier uses the Voronoi cell topology of particles to characterize their local structural environments “[Lazar, Han, Srolovitz, PNAS 112:43 (2015)]. See the corresponding user manual page for more information on this modifier. Note that this modifier inherits several important parameter fields from its StructureIdentificationModifier base class.

The Voronoi cell of a particle is the region of space closer to it than to any other particle. The topology of the Voronoi cell is the manner in which its faces are connected, and describes the manner in which a particle’s neighbors are arranged. The topology of a Voronoi cell can be completely described in a vector of integers called a Weinberg vector [Weinberg, IEEE Trans. Circuit Theory 13:2 (1966)].

This modifier requires loading a filter, which specifies structure types and associated Weinberg vectors. Filters for several common structures can be obtained from the VoroTop website. The modifier calculates the Voronoi cell topology of each particle, uses the provided filter to determine the structure type, and stores the results in the Structure Type particle property. This allows the user to subsequently select particles of a certain structural type, e.g. by using the SelectTypeModifier.

This method is well-suited for analyzing finite-temperature systems, including those heated to their bulk melting temperatures. This robust behavior relieves the need to quench a sample (such as by energy minimization) prior to analysis. Further information about the Voronoi topology approach for local structure analysis, as well as additional filters, can be found on the VoroTop webpage.

property filter_file

Path to the filter definition file used by the modifier. Filters files are available from the VoroTop website.

Default

''

property use_radii

If True, the modifier computes the poly-disperse Voronoi tessellation, which takes into account the radii of particles. Otherwise a mono-disperse Voronoi tessellation is computed, which is independent of the particle sizes.

Default

False

class ovito.modifiers.VoronoiAnalysisModifier

Base class

ovito.pipeline.Modifier

Computes the atomic volumes and coordination numbers using a Voronoi tessellation of the particle system. See the corresponding user manual page for more information. See Example B1: Computing Voronoi indices for a code example demonstrating the use of this modifier.

Inputs:

Particle properties
Position	The coordinates of the input particles.
Radius	Per-particle radii are used if use_radii is set to True.
Particle Type	Per-type radii are used if use_radii is set to True and Radius property is not present.
Selection	The selection state of the input particles. Only needed if only_selected is set to True.

Outputs:

Particle properties
Atomic Volume	The computed volume of each particle’s Voronoi polyhedron.
Coordination	The number of faces of each particle’s Voronoi polyhedron.
Voronoi Index	The index vector of each Voronoi polyhedron. Only computed if compute_indices is set to True.
Max Face Order	The maximum number of edges in any face of a particle’s Voronoi polyhedron. Only if compute_indices is set to True.

Attributes
Voronoi.max_face_order	Indicates the maximum number of edges of any face in the computed Voronoi tessellation (ignoring edges and faces that fall below the area/length thresholds).

Bond properties
Topology	The connectivity information of newly created Bonds (one bond for each Voronoi face). Only if generate_bonds is set to True.

property bonds_vis

The BondsVis object controlling the visual appearance of the bonds generated by the modifier if generate_bonds is set to True.

property compute_indices

If True, the modifier calculates the Voronoi indices of particles. The modifier stores the computed indices in a vector particle property named Voronoi Index. The i-th component of this property will contain the number of faces of the Voronoi cell that have i edges. Thus, the first two components of the per-particle vector will always be zero, because the minimum number of edges a polygon can have is three.

Default

False

property edge_threshold

Specifies the minimum length an edge must have to be considered in the Voronoi index calculation. Edges that are shorter than this threshold will be ignored when counting the number of edges of a Voronoi face. The threshold parameter is an absolute value in units of length of your input data.

Default

0.0

property face_threshold

Specifies a minimum area for individual Voronoi faces in terms of an absolute area. The algorithm will ignore any face of a Voronoi polyhedron with an area smaller than this threshold when computing the coordination number and the Voronoi index of a particle. The threshold parameter is an absolute area given in units of length squared (in whatever units your input data is given).

Note that this absolute area threshold and the relative_face_threshold are applied simultaneously.

Default

0.0

property generate_bonds

Controls whether the modifier outputs the nearest neighbor bonds. The modifier will generate a bond for every pair of adjacent atoms that share a face of the Voronoi tessellation. No bond will be created if the face’s area is below the face_threshold or if the face has less than three edges that are longer than the edge_threshold.

Default

False

property generate_polyhedra

Controls whether the modifier outputs the computed Voronoi cells as a polyhedral SurfaceMesh object.

Default

False

property mesh_vis

The SurfaceMeshVis object controlling the visual appearance of the polyhedral mesh generated by the modifier if generate_polyhedra is set to True.

property only_selected

Lets the modifier perform the analysis only for selected particles. Particles that are currently not selected will be treated as if they did not exist.

Default

False

property relative_face_threshold

Specifies a minimum area for Voronoi faces in terms of a fraction of total area of the Voronoi polyhedron surface. The algorithm will ignore any face of a Voronoi polyhedron with an area smaller than this threshold when computing the coordination number and the Voronoi index of particles. The threshold parameter is specified as a fraction of the total surface area of the Voronoi polyhedron the faces belong to. For example, a threshold value of 0.01 would remove those faces from the analysis with an area less than 1% of the total area of the polyhedron surface.

Note that this relative threshold and the absolute face_threshold are applied simultaneously.

Default

0.0

property use_radii

If True, the modifier computes the poly-disperse Voronoi tessellation, which takes into account the radii of particles. Otherwise a mono-disperse Voronoi tessellation is computed, which is independent of the particle sizes.

Default

False

class ovito.modifiers.WignerSeitzAnalysisModifier

Base class

ovito.pipeline.ReferenceConfigurationModifier

Performs the Wigner-Seitz cell analysis to identify point defects in a crystal. See the corresponding user manual page for more information.

Defects are identified with respect to a perfect reference crystal configuration. By default, frame 0 of the current simulation sequence is used as reference configuration. The modifier inherits from the ReferenceConfigurationModifier class, which provides further settings that control the definition of the reference configuration.

Outputs:

Particle properties
Occupancy	The computed site occupation numbers, one for each particle in the reference configuration.

Attributes
WignerSeitz.vacancy_count	The total number of vacant sites (having Occupancy == 0).
WignerSeitz.interstitial_count	The total number of of interstitial atoms. This is equal to the sum of occupancy numbers of all non-empty sites minus the number of non-empty sites.

Example:

The Occupancy particle property generated by the Wigner-Seitz algorithm allows to select specific types of point defects, e.g. antisites, using OVITO’s selection tools. One option is to use the ExpressionSelectionModifier to pick sites having a certain occupancy. The following script exemplarily demonstrates the use of a custom PythonScriptModifier to select and count A-sites occupied by B-atoms in a binary system with two atom types (A=1 and B=2).

from ovito.io import *
from ovito.data import *
from ovito.modifiers import *
from ovito.pipeline import *
import numpy as np

pipeline = import_file("input/simulation.*.dump")

# Perform Wigner-Seitz analysis:
ws = WignerSeitzAnalysisModifier(
    per_type_occupancies = True,
    affine_mapping = ReferenceConfigurationModifier.AffineMapping.ToReference)
pipeline.modifiers.append(ws)

# Define a modifier function that selects sites of type A=1 which
# are occupied by exactly one atom of type B=2.
def modify(frame, data):

    # Retrieve the two-dimensional array with the site occupancy numbers.
    # Use [...] to cast it to a Numpy array.
    occupancies = data.particles['Occupancy']

    # Get the site types as additional input:
    site_type = data.particles['Particle Type']

    # Calculate total occupancy of every site:
    total_occupancy = np.sum(occupancies, axis=1)

    # Set up a particle selection by creating the Selection property:
    selection = data.particles_.create_property('Selection')

    # Select A-sites occupied by exactly one B-atom (the second entry of the Occupancy
    # array must be 1, and all others 0). Note that the Occupancy array uses 0-based
    # indexing, while atom type IDs are typically 1-based.
    selection[...] = (site_type == 1) & (occupancies[:,1] == 1) & (total_occupancy == 1)

    # Additionally output the total number of antisites as a global attribute:
    data.attributes['Antisite_count'] = np.count_nonzero(selection)

# Insert Python modifier into the data pipeline.
pipeline.modifiers.append(modify)

# Let OVITO do the computation and export the number of identified
# antisites as a function of simulation time to a text file:
export_file(pipeline, "output/antisites.txt", "txt/attr",
    columns = ['Timestep', 'Antisite_count'],
    multiple_frames = True)

# Export the XYZ coordinates of just the antisites by removing all other atoms.
pipeline.modifiers.append(InvertSelectionModifier())
pipeline.modifiers.append(DeleteSelectedModifier())
export_file(pipeline, "output/antisites.xyz", "xyz",
    columns = ['Position.X', 'Position.Y', 'Position.Z'],
    multiple_frames = True)

property output_displaced

Specifies whether the modifier should output the atoms of the current configuration or replace them with the sites from the reference configuration.

By default, the modifier throws away all atoms of the current configuration and outputs the atomic sites from the reference configuration instead. Thus, in this default mode, you will obtain information about how many atoms occupy each site from the reference configuration. If, however, you are more insterested in visualizing the physical atoms that are currently occupying the sites (instead of the sites being occupied), then you should activate this modifier option. If set to true, the modifier will maintain the input atoms from the current configuration. The Occupancy property generated by the modifier will now pertain to the atoms instead of the sites, with the following new meaning: The occupancy number now counts how many atoms in total are occupying the same site as the atom the property refers to does. Furthermore, the modifier will in this mode output another property named Site Type, which reports for each atom the type of the reference site it was assigned to by the W-S algorithm.

Default

False

property per_type_occupancies

A flag controlling whether the modifier should compute occupancy numbers on a per-particle-type basis.

If false, only the total occupancy number is computed for each reference site, which counts the number of particles that occupy the site irrespective of their types. If true, then the Occupancy property computed by the modifier becomes a vector property with N components, where N is the number of particle types defined in the system. Each component of the Occupancy property counts the number of particles of the corresponding type that occupy the site. For example, the property component Occupancy.1 contains the number of particles of type 1 that occupy a site.

Default

False

class ovito.modifiers.WrapPeriodicImagesModifier

Base class

ovito.pipeline.Modifier

This modifier maps particles located outside of the simulation cell back into the cell by “wrapping” their coordinates around at the periodic boundaries of the SimulationCell. See also the corresponding user manual page for this modifier. This modifier has no configurable parameters.