Geometric aspects of particle segregation

Area of Science:

• Physics
• Materials Science
• Computational Modeling

Background:

• Size segregation is a prevalent phenomenon in both natural and industrial settings.
• Existing research primarily focuses on mechanistic explanations for segregation.
• A need exists for alternative predictive models, especially for complex particle shapes.

Purpose of the Study:

• To demonstrate that particle segregation can be explained and predicted using a geometric approach.
• To develop and validate a computational algorithm for simulating segregation.
• To compare simulation results with physical experiments for arbitrary particle shapes.

Main Methods:

• A digital simulation algorithm incorporating random walks, a rebounding probability, and a non-overlap constraint.
• Implementation on a regular lattice grid to handle arbitrary particle shapes.
• Comparison of simulation outcomes with laboratory experiments using pseudo-two-dimensional containers and mixtures of nonspherical particles.

Main Results:

• The geometric algorithm successfully simulated shaking-induced segregation comparable to physical tests.
• Segregation trends were accurately predicted without explicit consideration of particle interaction forces.
• The relative motion between particles of different sizes and shapes was identified as a key geometrical factor.

Conclusions:

• Particle segregation can be adequately explained from a geometrical perspective.
• The developed geometric algorithm provides a fast and qualitative prediction tool for segregation likelihood in arbitrary mixtures.
• This approach offers a novel, computationally efficient method for understanding and predicting segregation phenomena.