Wafer Map Simulation Using Plasma and Etch Distribution Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in semiconductor manufacturing is the unintended electrical characteristics due to combined factors in the etching process, particularly the unpredictable distribution of gases in the wafer, leading to deviations from expected structures, necessitating improved modeling and simulation for more accurate product specifications.
Innovation Solution
A method and apparatus using a distribution model to generate a wafer map based on critical dimensions, incorporating a plasma model, distribution model, and etch model to simulate the distribution of flux and energy across wafer coordinates, thereby generating a wafer map that includes critical dimensions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If physical simulation is used to understand semiconductor processes, then accuracy of product specifications is improved, but computational complexity and time required increase
Solution Approach 1:
The simulation system is divided into separate functional modules: a plasma model for calculating plasma properties, a distribution model for predicting gas distribution patterns, and an etch model for computing etching results. Each module processes specific inputs and generates specific outputs, allowing the complex simulation to be segmented into manageable computational tasks that can be executed sequentially or in parallel.
Solution Approach 2:
The distribution model acts as an intermediary between the plasma model and the etch model. It receives plasma parameters from the plasma model, predicts gas distribution patterns based on wafer coordinates, and provides these distribution data to the etch model. This intermediary layer simplifies the overall computation by pre-calculating gas distribution patterns that can be reused in etching simulations.
2Manufacturing precision
If comprehensive modeling of gas distribution is implemented, then manufacturing precision is improved, but simulation time increases
Solution Approach 1:
The distribution model performs preliminary calculations of gas distribution patterns based on plasma parameters and wafer coordinates before the actual etching simulation. By pre-computing these distribution patterns and storing them as lookup tables or predictive models, the system can quickly retrieve or interpolate results during the etching simulation without performing comprehensive gas distribution calculations in real-time, significantly reducing simulation time.
3Manufacturing precision
If detailed wafer map generation is performed, then manufacturing precision is improved, but computational resources required increase
Solution Approach 1:
The distribution model focuses computational resources on calculating gas distribution patterns at specific locations on the wafer where they are most critical for etching uniformity. Rather than uniformly distributing computational effort across the entire wafer surface, the system prioritizes calculations in regions with complex pattern features or expected non-uniformity, using the plasma model and distribution model to efficiently determine where detailed analysis is most needed.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces distribution dispersion in semiconductor fabrication, improving yield, fabrication time, and reliability by accurately modeling and predicting gas distribution patterns.
Implementation Method 1
computing reaction of input data by using a plasma model
Implementation Method 2
modeling how the distribution of gases in a wafer is formed in response to the coordinates of the wafer
Implementation Method 3
generating a wafer map based on the second output and a structure of the wafer by using an etch model
Data Source
AI summary
Provided are an apparatus and method for generating a wafer map corresponding to coordinates of a wafer by using a distribution model. A method of performing a process simulation of a semiconductor device includes computing reaction of input data by using a plasma model, and based on the computed reaction, generating a first output including first flux and first energy, generating a second output including second flux and second energy, based on the first output and coordinates of a wafer, by using a distribution model, and generating a wafer map based on the second output and a structure of the wafer by using an etch model, wherein the second flux includes flux corresponding to each of the coordinates of the wafer, and wherein the second energy includes energy corresponding to each of the coordinates of the wafer.