PFI-TT: Developing an Efficient Computation Scheme for Modeling Low-Pressure Plasmas

点击翻译按钮获取中文摘要

The broader impact/commercial potential of this Partnerships for Innovation - Technology Translation (PFI-TT) project is the significant reduction of low-pressure-plasma modeling time using innovative computation schemes. Low-pressure plasmas are widely used for the manufacturing of semiconductor integrated circuits, displays, solar panels, thin film batteries, and coatings. The annual market of these industry segments exceeds $600 billion. Computer simulation is an essential approach to understanding the complex plasma characteristics and developing efficient plasma processing technologies. Unfortunately, current commercial plasma simulation software requires huge computational resources; it takes months to model the plasmas in practical scales. The lack of a practical tool for modeling low-pressure plasmas has resulted in improperly designed sources, leading to the inefficient processes used today. This PFI-TT project aims to develop new computation schemes for efficiently modeling certain plasmas at low pressures. These tools will allow the users to simulate plasmas of practical scales, using even desktop computers that generate feedback in 24 hours. This project will also train and broaden the participation of graduate and undergraduate students, and prepare them as future leaders in technology innovation and entrepreneurship. The proposed project combines an implicit algorithm of the motions of charged particles with an energy conservation scheme to eliminate the spatial and temporal constraints in the state-of-the-art explicit particle-in-cell/Monte Carlo collision algorithms. The key advantage of the implicit algorithm is that the high frequency oscillations of plasmas are damped in the time domain. The energy conservation scheme ensures that the total energy of the plasma is conserved and the self-heating due to a large grid spacing (i.e. significantly greater than the plasma Debye length) is eliminated. The new computation scheme is over 30 times faster than the state-of-the-art explicit particle-in-cell/Monte Carlo collision algorithms. The developed modeling tools also have the potential to address situations where a complete kinetic analysis is not yet available due to the high plasma density, such as high-power impulse magnetron sputtering and the plasma instabilities in magnetized discharges.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.