A Numerical Study of an Ellipsoidal Nanoparticles under High Vacuum Using the DSMC Method

Jinwoo Jang, Young‐Woo Son, Sanghwan LeeMicromachines 2023 · DOI

TL;DR

This paper tackles the critical challenge of predicting nanoparticle behavior in high-vacuum environments, crucial for semiconductor manufacturing. By employing a GPU-accelerated Direct Simulation Monte Carlo (DSMC) method, the researchers accurately calculated drag forces on both spherical and ellipsoidal nanoparticles, extending predictions to previously inaccessible high Knudsen number regimes.

What Problem Does This Paper Address?

The semiconductor and display manufacturing industries rely on ultra-precise processes that must occur in high-vacuum conditions to minimize contamination. However, ironically, these high-vacuum environments are precisely where controlling and predicting the movement of tiny impurity particles becomes extremely difficult. Conventional analytical tools struggle to accurately model particle flow in these rarefied gas conditions, especially at high Knudsen numbers (where the mean free path of gas molecules is comparable to or larger than the characteristic length scale). This lack of predictive capability hinders the ability to mitigate particle-induced defects, impacting production yield and quality. This paper aims to bridge this gap by providing a robust numerical method to simulate and understand nanoparticle behavior in these challenging environments.

Methods

The core of this research lies in the application and enhancement of the Direct Simulation Monte Carlo (DSMC) method. DSMC is a powerful technique for simulating rarefied gas flows by tracking the statistical behavior of a large number of representative particles.

Key methodological aspects include:

  • GPU Acceleration (CUDA): The DSMC method is computationally intensive. To overcome this, the researchers leveraged Nvidia's CUDA platform to harness the parallel processing power of GPUs. This significantly speeds up the simulations, making it feasible to analyze complex scenarios.
  • Lagrangian Particle Tracking: The DSMC implementation uses a Lagrangian approach to track individual particles and their trajectories.
  • Drag Force Calculation: The study focuses on calculating the drag force acting on nanoparticles. This involves incorporating slip correction factors to account for the non-continuum nature of the flow at high Knudsen numbers. The paper reviews and utilizes established slip correction formulations (e.g., Cunningham, Millikan, Weber).
  • Ellipsoidal Particle Modeling: Moving beyond idealized spherical particles, the study introduces the analysis of ellipsoidal nanoparticles. This involves varying the aspect ratio (AR) of the ellipsoid to understand its impact on drag force.
  • Validation: The DSMC code was first validated against theoretical solutions for microchannel flow (Poiseuille flow with slip boundary conditions) to ensure its accuracy.

Datasets

Not discussed in the paper. The study relies on theoretical models and empirical formulas for validation, rather than pre-existing datasets for training or evaluation.

Key Insights and Ideas

  • GPU-Accelerated DSMC is Essential: The computational demands of DSMC for high-vacuum nanoparticle flow are immense. The paper strongly advocates for and demonstrates the necessity of GPU computing (CUDA) to make such simulations practical.
  • Ellipsoidal Shape Matters: Real-world nanoparticles are often not perfectly spherical. The study reveals that the drag force on ellipsoidal particles increases significantly with their aspect ratio, especially in rarefied gas conditions. This is attributed to the dominance of direct collision effects over frictional forces in high-vacuum.
  • Extending to High Knudsen Numbers: The research successfully extends drag force predictions to very high Knudsen numbers (up to 10,000), a regime that is experimentally challenging and where traditional methods falter. This opens up new possibilities for understanding and controlling particle behavior in advanced manufacturing.
  • Drag Force Dominance by Collision: In high-vacuum (rarefied gas) regimes, the drag force on particles is primarily driven by direct molecular collisions rather than viscous friction, a key difference from continuum flow.

Key Results

  • Microchannel Flow Validation: The DSMC code's accuracy was confirmed by achieving good agreement with theoretical solutions for microchannel flow at moderate Knudsen numbers (around 0.04).
  • Spherical Particle Drag: The simulation results for spherical particles showed good agreement with experimental data and empirical formulas for Knudsen numbers up to 83. Crucially, the study extended these predictions to Knudsen numbers as high as 10,000.
  • Ellipsoidal Particle Drag: For ellipsoidal particles, the drag force was found to increase linearly with the aspect ratio (AR) for a given Knudsen number. This increase was observed to be independent of the Knudsen number itself within the studied range.
  • Normalized Drag Force: Figure 7 demonstrates that the normalized drag force (relative to a spherical particle) increases with aspect ratio, with the increase being consistent across different Knudsen numbers.

Limitations and Future Work

  • Experimental Verification: While the study validates against existing experimental data for spherical particles, direct experimental validation for ellipsoidal particles at high Knudsen numbers is lacking due to experimental difficulties.
  • Particle-Particle Interactions: The current study primarily focuses on single-particle drag. Future work could explore the effects of particle-particle interactions, which become more significant in denser particle flows.
  • Complex Geometries: The analysis was performed in simplified geometries (microchannels and cylinders). Extending the DSMC model to more complex industrial geometries would be valuable.
  • Surface Interactions: While slip is considered, more detailed modeling of surface interactions (e.g., adhesion, electrostatic forces) could further enhance the predictive capabilities for specific applications.
  • Material Properties: The study focuses on the fluid dynamics. Incorporating specific material properties of nanoparticles could provide more nuanced insights.

Summary updated

Abstract

The semiconductor and display manufacturing process requires high precision. Therefore, inside the equipment, fine impurity particles affect the yield rate of production. However, since most manufacturing processes are performed under high-vacuum conditions, it is difficult to estimate particle flow with conventional analytical tools. In this study, high-vacuum flow was analyzed using the direct simulation Monte Carlo (DSMC) method, and various forces acting on fine particles in a high-vacuum flow field were calculated. To compute the computationally intensive DSMC method, GPU-based computer unified device architecture (CUDA) technology was used. The force acting on the particles in the high-vacuum rarefied gas region was verified using the results of previous studies, and the results were derived for the difficult-to-experiment region. An ellipsoid shape with an aspect ratio rather than a spherical shape was also analyzed. The change in drag force according to various aspect ratios was analyzed and compared with the results of the spherical shape under the same flow conditions.
BibTeX
@inproceedings{jang2023numerical,
  author = {Jinwoo Jang and Young‐Woo Son and Sanghwan Lee},
  title = {{A Numerical Study of an Ellipsoidal Nanoparticles under High Vacuum Using the DSMC Method}},
  booktitle = {Micromachines},
  year = {2023},
  doi = {10.3390/mi14040778},
}
Figure