Hybrid Modeling Method for a DEP Based Particle Manipulation

Mohamad Sawan, Fieee, Frsc, Fcae, Feic, O.q., Mohamed Amine Miled, Antoine Gagne, Mohamad SawanSensors 2013 · DOI

TL;DR

This paper introduces a novel hybrid modeling approach that bridges the gap between finite element modeling (FEM) and biological behavior for Dielectrophoresis (DEP)-based particle manipulation. By coupling ANSYS for electrical field simulation with MATLAB for particle motion calculation, this method offers a more comprehensive and versatile tool for designing BioMEMS devices, particularly for applications requiring precise particle manipulation.

What Problem Does This Paper Address?

Designing and simulating microelectromechanical systems (MEMS), especially BioMEMS for medical and biological applications, is crucial for reducing development time and costs. However, existing modeling tools often struggle to accurately capture the complex interplay between electrical fields, fluid dynamics, and the unique properties of biological particles. Specifically, simulating particle behavior within microfluidic devices under applied electrical fields, like DEP, presents challenges. Traditional FEM approaches can be computationally intensive and may not adequately account for dynamic changes in particle and medium properties during operation, which are critical for biological systems. This paper aims to address this by developing a more integrated and flexible modeling framework.

Methods

The authors propose a hybrid modeling approach that leverages the strengths of two powerful software packages: ANSYS and MATLAB.

  • ANSYS: Used for building the microfluidic channel and electrode geometry, and for simulating the propagation of the electrical field and fluid flow. This part of the process involves setting up the physical environment and applying the necessary loads (voltage, pressure, velocity).
  • MATLAB: Serves as the central hub for particle manipulation and calculation. It imports the geometry from ANSYS, defines particle properties (mass, charge, radius, permittivity), and then iteratively calculates the forces acting on the particles (dielectrophoretic, electrophoretic, and hydrodynamic). Based on these calculated forces, MATLAB updates the particle's position and sends this new information back to ANSYS for the next iteration.

This iterative coupling allows for a dynamic simulation where particle motion influences the overall simulation environment, a key improvement over static models. The method considers particle shape, weight, and electrical properties, and importantly, it decouples the meshing process from particle movement, which significantly enhances computational efficiency.

Datasets

Not discussed in the paper. The paper focuses on a simulation methodology and its validation against experimental results, rather than training on specific datasets.

Key Insights and Ideas

  • Bridging the Simulation Gap: The core innovation is the successful integration of ANSYS and MATLAB to create a unified modeling environment for BioMEMS. This hybrid approach overcomes the limitations of using either software in isolation for complex multiphysics simulations involving biological particles.
  • Dynamic Particle Interaction: By enabling an iterative feedback loop between ANSYS and MATLAB, the model accounts for how particle movement can influence the simulation environment, leading to more realistic predictions of particle behavior.
  • Computational Efficiency: Decoupling the meshing process from particle movement is a significant advancement. This allows for faster simulations compared to traditional FEM methods that require constant re-meshing as particles move.
  • Versatility in Electrode Design: The method is not limited to specific electrode architectures, allowing researchers to explore various designs and their impact on particle manipulation, including potential for 3D separation.
  • Experimental Validation: The simulation results are validated against experimental data, demonstrating the accuracy and reliability of the proposed hybrid modeling technique.

Key Results

The proposed hybrid modeling method successfully simulated particle motion in microfluidic channels under DEP forces.

  • Accurate Trajectory Prediction: The simulations accurately predicted particle trajectories, including helical paths observed in experiments, for both 4-electrode and 8-electrode architectures.
  • Z-axis Separation Potential: The model demonstrated the ability to predict particle behavior in the z-axis, suggesting potential for 3D particle separation, which is difficult to achieve with 2D experimental methods. For instance, the 4-electrode architecture showed particles suspended in the middle of the channel depth (10 µm), while the 8-electrode architecture pushed particles towards the top.
  • Comparison with Experimental Data: The simulation results showed good agreement with experimental data for particle displacement in the X and Y axes, particularly in terms of motion patterns. The modeling results were close to experimental curves, with minor differences attributed to the experimental method approximating motion via area centroid tracking.

Limitations and Future Work

  • Computational Resources: The authors acknowledge that the proposed approach requires significant CPU resources.
  • Limited to DEP: Currently, the model is primarily focused on DEP manipulation.
  • Biological Behavior Simplification: While the model considers particle properties, it currently assumes steady-state charge and conductivity for medium and particles. The authors note that biological cells exchange ions, which changes medium conductivity and affects DEP forces, a factor not fully captured in the current steady-state assumption.
  • Future Work: The authors plan to:
  • Enlarge the model to include other manipulation techniques like magnetophoresis.
  • Further investigate the biological aspects, such as considering the particle and its surrounding medium as a single entity.
  • Explore alternative solutions, such as using COMSOL, to compare results and investigate wall effects.
  • Extend the model to handle non-steady state medium and particle properties.

Summary updated

BibTeX
@article{oq2013hybrid,
  author = {Mohamad Sawan, FIEEE, FRSC, FCAE, FEIC, O.Q. and Mohamed Amine Miled and Antoine Gagne and Mohamad Sawan},
  title = {{Hybrid Modeling Method for a DEP Based Particle Manipulation}},
  journal = {Sensors},
  year = {2013},
  doi = {10.3390/s130201730},
}

Loading related content...

Figure