Recipe 3: Color coding grain boundary atoms

How to Script with OVITO

Author

Affiliation

Stefan Bringuier

Materials Scientist

When you have grain boundaries it is convenient to be able to distinguish those atoms from the bulk. One way to do so is using a order or structural parameter that captures some information about the local environment. OVITO has a few different options built in that enable characterize whether an atom is in an FCC, BCC, HCP, or other structure. Here I’ll be working with the LAMMPS dump file from the example in (Bringuier 2015) which reproduces the results from (Cahn, Mishin, and Suzuki 2006).

%%capture
! pip install -U ovito

Import OVITO modules

from ovito.io import import_file
from ovito.vis import Viewport
from ovito.modifiers import CommonNeighborAnalysisModifier, CalculateDisplacementsModifier
from ovito.modifiers import ExpressionSelectionModifier, DeleteSelectedModifier, InvertSelectionModifier
from ovito.modifiers import AssignColorModifier
from ovito.vis import VectorVis
from ovito.vis import OSPRayRenderer

Step 1: Download and import file

Note

This LAMMPS dump file was from a 2D simulation but the symmetry is 3D so we reset the PBC for all directions. I’m doing this because I want to maintain the correct structural analysis results.

%%capture
! wget 'https://drive.google.com/uc?id=1Id0D6rfHxJOTuPx2RVAYkm2yRBBgIzSE&export=download' -O dump.Cu_Bicrystal_Shear_298K.gz
pipeline = import_file('dump.Cu_Bicrystal_Shear_298K.gz',input_format='lammps/dump')
pipeline.source.data.cell_.pbc = (True, True, True)

Step 2: Perform structural analysis and calculate displacement vectors

I will use the CommonNeighborAnalysisModifier which provides standard structural analysis to identify the closed-packing of the crystal (e.g., FCC, BCC, HCP, ICO).

Once the structure modifier is used, I will color code just the atoms that are of StructureType == which corresponds to “Other”. To do this we can use the modifier to select these atoms and then color code.

pipeline.modifiers.append(CommonNeighborAnalysisModifier())

# Select particles based on structure type and position
expression = 'StructureType != 0 || (Position.Z > CellSize.Z-5 || Position.Z < 5.0 )'
pipeline.modifiers.append(ExpressionSelectionModifier(expression=expression))

# Color the selected particles
pipeline.modifiers.append(AssignColorModifier(color=(0.0, 0.5, 1.0))) 
pipeline.modifiers.append(InvertSelectionModifier())
# Now color the GB atoms
pipeline.modifiers.append(AssignColorModifier(color=(0.8, 0.5, 0.2)))
pipeline.modifiers.append(InvertSelectionModifier())

pipeline.add_to_scene()

Step 3: Render first animation

vp = Viewport(type=Viewport.Type.Ortho)
vp.camera_dir = [-1, 0, 0]
vp.zoom_all()
renderer = OSPRayRenderer(
    ambient_light_enabled=True,
    denoising_enabled=True,
    direct_light_enabled=True,
    direct_light_intensity=1.0,
    material_shininess=10.0,
    material_specular_brightness=0.02
)

fname = "Cu_Bicrystal_Shear_298K_ColorCoding.gif"
vp.render_anim(size=(600,400), 
               filename=fname, 
               renderer=renderer)

Note

The code below is to allow for displaying rendered images in Google Colab.

try:
    import google.colab
    from IPython.display import Image
    Image(open(fname, 'rb').read())
except ImportError:
    "Assuming local run."

Figure 1: Grain boundary color coded.

Step 4: Displacement vectors

We may want to show how the atoms at the grain boundary are moving with respect to the shear force. We can do this with the displacement vectors. The displacement vectors are taken with respect to an previous frame offset of 25. This is used to showcase how the grain boundary atoms are moving, what you’ll see is that its in a cork-screw like manner.

Since we already have a pipeline and all we want to do is add the displacement vectors and then disable visualizing the particles themselves. I also just want to so the displacement vectors for the grain boundary atoms, so I need to delete the selected particles and then we just need to add a few additional lines of code and then re-render to a new animation file.

# Visualization object for vector arrows
vec_vis = VectorVis(width=0.1,
                    scaling=1.5,
                    flat_shading=False,
                    color=(1,0,0))

displacement_mod = CalculateDisplacementsModifier(frame_offset=-25,
                                                  use_frame_offset=True,
                                                  vis=vec_vis)
displacement_mod.vis.enabled = True
pipeline.modifiers.append(displacement_mod)

# Need this line to turn off particle visibility
# Delete elected particles
pipeline.modifiers.append(DeleteSelectedModifier(operate_on={'particles'}))
pipeline.compute(frame=25).particles.vis.enabled = False

Step 5: Render second animation

vp2 = Viewport(type=Viewport.Type.Ortho)
vp2.camera_dir = [-1, 0, 0]
vp2.zoom_all()
fname = "Cu_Bicrystal_Shear_298K_DisplacementVectors.gif"
nframes = pipeline.source.num_frames
vp2.render_anim(size=(600,400), 
               filename=fname, 
               range=(25,nframes),
               renderer=renderer)

try:
    import google.colab
    from IPython.display import Image
    Image(open(fname, 'rb').read())
except ImportError:
    "Assuming local run."

Figure 2: Visualization of displacement vectors for grain boundary.

References

Bringuier, Stefan. 2015. “Reproducing Cu Bicrystal Grain Boundary Coupling Motion to Shear Deformation,” May. https://doi.org/10.5281/zenodo.3900344.

Cahn, John W., Yuri Mishin, and Akira Suzuki. 2006. “Coupling Grain Boundary Motion to Shear Deformation.” Acta Materialia 54 (19): 4953–75. https://doi.org/10.1016/j.actamat.2006.08.004.