Electrolyte Complexity at Electrochemical Interfaces - Explicit solvation, PCET, Ion effects
Support curvature, high-metal loading, MNC asymmetry, mesopores
Detailed surface atomic structure visualization
Angew. Chem. Int. Ed. 64 (2025) - Rh/CeO2 CO-SCR
Catal. Today 425 (2024) - IF=5.2
Catal. Today 411-412 (2023) - IF=5.3
J. Mater. Chem. A 11 (2023) - IF=11.9
import bpy
from pathlib import Path
from math import pi
def setup_scene():
"""Apply camera, light, and material settings"""
# Camera setting
bpy.data.objects['Camera'].location.x = -0
bpy.data.objects['Camera'].location.y = -25
bpy.data.objects['Camera'].location.z = 13
bpy.data.scenes["Scene"].camera.rotation_mode = 'XYZ'
bpy.data.scenes["Scene"].camera.rotation_euler[0] = 65*(pi/180.0)
bpy.data.scenes["Scene"].camera.rotation_euler[1] = 0*(pi/180.0)
bpy.data.scenes["Scene"].camera.rotation_euler[2] = -5*(pi/180.0)
# Light setting
bpy.data.objects['Light'].data.type = 'SUN'
bpy.data.objects['Light'].data.energy = 2
bpy.data.objects['Light'].location.x = -10
bpy.data.objects['Light'].location.y = -10
bpy.data.objects['Light'].location.z = 20
bpy.data.objects['Light'].rotation_mode = 'XYZ'
bpy.data.objects['Light'].rotation_euler[0] = 30*(pi/180.0)
bpy.data.objects['Light'].rotation_euler[1] = -30*(pi/180.0)
bpy.data.objects['Light'].rotation_euler[2] = 0*(pi/180.0)
# Materials setting (only if materials exist)
if "Nitrogen" in bpy.data.materials:
bpy.data.materials["Nitrogen"].node_tree.nodes["Principled BSDF"].inputs[0].default_value = (1, 0.401978, 0.08022, 1)
if "Cobalt" in bpy.data.materials:
bpy.data.materials["Cobalt"].node_tree.nodes["Principled BSDF"].inputs[0].default_value = (0, 0, 0.428691, 1)
if "Molybdenum" in bpy.data.materials:
bpy.data.materials["Molybdenum"].node_tree.nodes["Principled BSDF"].inputs[0].default_value = (0.527112, 0.527116, 0.527115, 1)
# Render setting
bpy.context.scene.render.engine = 'BLENDER_EEVEE'
bpy.context.scene.eevee.taa_render_samples = 32
# Image format settings
bpy.context.scene.render.image_settings.file_format = 'JPEG'
bpy.context.scene.render.image_settings.quality = 95
def clear_mesh_objects():
"""Remove all mesh objects from scene"""
bpy.ops.object.select_all(action='DESELECT')
for obj in bpy.data.objects:
if obj.type == 'MESH':
obj.select_set(True)
bpy.ops.object.delete()
# Directories to process
input_dirs = [
Path('/Users/sean/Downloads/supported'),
Path('/Users/sean/Downloads/structures_optimized')
]
output_dir = Path('/Users/sean/Downloads/rendered_images')
output_dir.mkdir(exist_ok=True)
# Collect all .xyz files from both directories
all_xyz_files = []
for input_dir in input_dirs:
xyz_files = list(input_dir.glob('*.xyz'))
all_xyz_files.extend(xyz_files)
print(f"Found {len(xyz_files)} files in {input_dir.name}")
print(f"\nTotal: {len(all_xyz_files)} .xyz files to render\n")
# Process each file
for i, xyz_file in enumerate(all_xyz_files):
print(f"Processing {i+1}/{len(all_xyz_files)}: {xyz_file.name}")
# Clear previous mesh objects
clear_mesh_objects()
# Import the .xyz file
try:
bpy.ops.import_mesh.xyz(filepath=str(xyz_file))
except Exception as e:
print(f"Error importing {xyz_file.name}: {e}")
continue
# Apply settings
setup_scene()
# Render and save
output_path = output_dir / f"{xyz_file.stem}.jpg"
bpy.context.scene.render.filepath = str(output_path)
bpy.ops.render.render(write_still=True)
print(f"Saved: {output_path.name}")
print(f"\nDone! Rendered {len(all_xyz_files)} files to {output_dir}")
Example batch-rendered nanorod structures with varying dimensions
1. The Knowledge Condenser
A vast, chaotic nebula of glowing particulate nodes and faint, tangled edges—representing a massive "teacher" graph neural network—fills the left side of the frame. This turbulent cloud is drawn into the flared aperture of a crystalline, lens-like construct in the center. As the information passes through, it is filtered and compressed, emerging from the right as a single, coherent, intensely bright beam. This beam illuminates a complex, rotating molecular structure, projecting precise, glowing vectors onto each atom, visualizing the distilled model's accurate force estimations. The background is a dark, abstract space, subtly gridded to imply a coordinate system.
2. The Distillation Forge
In a dark, reflective environment, a massive, intricately faceted, and ancient-looking crystal floats, representing the "teacher" model. Its internal structure is a dense lattice of light, and complex graph patterns shift across its surfaces. It projects a focused, complex beam of light onto a much smaller, simpler, polished gem—the "student" model—held in an elegant, minimalist armature below.
CO2 utilization flowchart
Generative GNN pipeline visualization