Plasma Fluid Modeling Transient Stochastic Transformation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current plasma modeling and simulation methods for complex plasma systems, such as dual-frequency CCP etching systems, require extensive computational time, often taking several hours to days, due to the need to solve multiple frequency scales and complex chemistry, limiting real-time process optimization and hardware development.

Innovation Solution

A computer method that approximates transient sheath behavior using a probability function-based time-domain simulation, allowing for the simulation of plasma density distribution in response to variables with different time bases, reducing computational time by employing stochastic transformation and random sampling to achieve a steady-state plasma density distribution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If full transient numerical simulation is performed to model plasma density distribution in dual-frequency CCP systems, then accuracy and reliability of plasma behavior prediction is improved, but computational time increases to several hours or days

Engineering Contradiction:
Improveaccuracy of plasma behavior predictionVSAvoidcomputational time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the plasma simulation problem into two distinct parts: (1) a simplified steady-state model that provides accurate plasma density distribution, and (2) a separate sheath model that accounts for transient effects. This segmentation allows each model to be optimized independently, achieving both accuracy and computational efficiency without requiring full transient simulation of the entire plasma system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the temporal parameter of the simulation by solving the plasma bulk equations under steady-state conditions rather than transient conditions. The time-dependent terms are retained only in the sheath region equations, effectively separating the time scales of the plasma bulk and sheath regions. This parameter change reduces computational time while maintaining accuracy for the relevant plasma properties.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If multiple frequency scales (e.g., 100 MHz and 1-2 MHz) are solved simultaneously in DF-CCP etching system, then completeness of plasma process modeling is improved, but computational complexity and time increase drastically

Engineering Contradiction:
Improvecompleteness of plasma process modelingVSAvoidcomputational complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the multi-frequency plasma system into a plasma bulk region and a sheath region, where each region is modeled with appropriate physics. The plasma bulk is modeled with time-averaged equations that capture the effect of multiple frequencies without resolving each frequency explicitly. The sheath region is modeled separately with transient equations that capture the high-frequency dynamics. This segmentation reduces computational complexity while maintaining modeling completeness.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from a time-domain multi-frequency solution to a spatial-domain solution by solving the plasma bulk equations in a time-averaged steady state. The temporal dimension is effectively removed from the bulk plasma equations, and the effects of multiple frequencies are captured through spatial distributions of plasma parameters. This dimensional change significantly reduces computational complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Manufacturing precision

If high resolution reactor geometry and complex chemistry are modeled, then manufacturing precision of plasma process control is improved, but computational time increases to several days

Engineering Contradiction:
Improveprecision of plasma process controlVSAvoidcomputational time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent segments the complex plasma chemistry and transport equations into a steady-state plasma bulk model and a transient sheath model. The complex chemistry reactions are solved in the steady-state bulk plasma where time derivatives are neglected, reducing the computational burden. High-resolution geometry is maintained in the spatial discretization, but the removal of time-dependent terms from the bulk equations dramatically reduces computational time while preserving manufacturing precision for plasma process control.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS8103492B2Plasma fluid modeling with transient to stochastic transformation
Publication Date: 2012.01.24 TOKYO ELECTRON LTD
  • US8103492B2 patent drawing
  • US8103492B2 patent drawing
  • US8103492B2 patent drawing

AI summary

The invention relates to the simulation method and apparatus used in plasma modeling. It includes a method to transform transient formulations of the phenomenological plasma model into a quasi-stochastic spatial formulation. Specifically, the invention aids in decreasing computational time for the modeling of plasma in a plasma processing system, particularly those involving two different time-based parameters. The invention is particularly described in connection with plasma simulations used for the optimization dual-frequency capacitively-coupled plasma etching systems.