Rotating Substrate Liquid Film Analysis Using Luminance Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semiconductor cleaning processes lack a quantitative method to analyze the state of the liquid membrane on a rotating substrate, which affects the quality and yield of semiconductor manufacturing.
Innovation Solution
A method involving the acquisition of luminance values from video images of the substrate surface, calculating first and second statistics such as luminance and dissimilarity averages, and analyzing the liquid membrane state based on these statistics to quantify its behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a video camera is installed to observe the substrate operation, then the liquid membrane state can be visualized, but there is no quantitative evaluation method available
Solution Approach 1:
The patent replaces complex mechanical observation systems with optical imaging and computational analysis. By using a video camera to capture luminance information and processing it through statistical calculations (mean luminance, standard deviation, histogram analysis), the system achieves quantitative measurement without requiring complex physical sensors or mechanical measurement devices直接接触 the liquid membrane.
Solution Approach 2:
The patent introduces luminance values as an intermediary parameter to bridge the gap between visual observation and quantitative measurement. The luminance data from video images serves as a mediator that can be processed through statistical methods to extract meaningful information about liquid membrane state, such as uniformity, stability, and transition characteristics.
2Loss of information
If statistical analysis of luminance values is performed, then quantitative analysis of liquid membrane state is achieved, but the analysis complexity increases
Solution Approach 1:
The patent segments the complex liquid membrane state into multiple quantifiable parameters: mean luminance (overall brightness), standard deviation (uniformity), histogram distribution (intensity frequency), and temporal changes (stability). This segmentation allows each aspect to be analyzed independently through specific statistical metrics, making the overall complex state measurable and controllable.
Solution Approach 2:
The patent transforms the physical state of the liquid membrane into changes in luminance parameters. By monitoring how luminance values, their distributions, and statistical properties change over time or under different conditions, the system quantitatively captures liquid membrane behavior without directly measuring physical properties like thickness or composition.
3Productivity
If the substrate rotates at high speed, then cleaning efficiency is improved, but the observation and analysis of liquid membrane becomes more difficult
Solution Approach 1:
The patent uses high-speed video imaging to continuously capture luminance data during substrate rotation. By maintaining continuous observation throughout the rotation cycle and processing the time-series luminance data, the system can extract statistical parameters that represent the average liquid membrane state, effectively freezing the dynamic process into measurable data points.
Solution Approach 2:
The patent leverages the periodic nature of substrate rotation to sample liquid membrane states at consistent phases. By analyzing luminance data collected over multiple rotation cycles and computing temporal statistics, the system identifies stable patterns and variations that represent the true liquid membrane behavior, filtering out rotational noise.
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
Enables the quantitative analysis of the liquid membrane state, allowing for the determination of switching completion times and evaluation of the liquid membrane's stability, thereby improving process control and quality assurance in semiconductor manufacturing.
Implementation Method 1
acquiring luminance values of individual pixels in a region that is predetermined, from a video image obtained by imaging a surface of the substrate
Data Source
AI summary
An analysis method for analyzing a state of a liquid membrane, on a substrate, of a liquid supplied to the substrate that is rotating in a substrate processing apparatus includes: a process (S303) of calculating a first statistic from luminance values of individual pixels in a predetermined region from a video image obtained by imaging a surface of the substrate; a process (S304) of calculating at least one second statistic that is a statistic of time-series data of the first statistics; and a process (S305) of analyzing a state of the liquid membrane on the substrate based on at least one of the first statistic or the second statistic. The first statistic includes a luminance average value obtained by averaging the luminance values and a dissimilarity average value obtained by averaging dissimilarities of the luminance values. The second statistic includes at least one of a luminance statistic which is a statistic of the luminance average value or a dissimilarity statistic which is a statistic of the dissimilarity average value. As a result, a state of the liquid membrane on the substrate can be quantitatively analyzed.