Wafer Thickness Estimation via Transfer Robot Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor production lines, the existing inspection methods for determining the completion of processes in process chambers are time-consuming and space-intensive, leading to decreased productivity and increased costs due to the need for separate inspection processes after each semiconductor process.
Innovation Solution
A thickness estimating apparatus that uses a transfer robot, light source, camera, and machine learning-based thickness predicting model to quickly estimate the thickness of wafers and element layers by capturing images and applying pixel data to predict the thickness, allowing for real-time process feedback and improved yield.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a separate inspection process is used to measure wafer thickness, then measurement precision is improved, but productivity decreases and loss of time increases
Solution Approach 1:
The inspection apparatus merges the thickness measurement function with the transfer robot system by integrating a light source and camera onto the transfer path. This allows wafer thickness to be measured during the transfer process itself, eliminating the need for separate inspection equipment and processes, thereby improving productivity while maintaining measurement precision
Solution Approach 2:
The system performs thickness measurement preliminarily during the transfer process before the wafer reaches the final processing stage. By capturing images during transfer and immediately processing them through the trained model, the system obtains thickness data in advance, enabling real-time feedback and process adjustment without delaying production
2Measurement precision
If a separate inspection process is used to measure wafer thickness, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The inspection apparatus merges the thickness measurement function with the transfer robot system by integrating a light source and camera onto the transfer path. This allows wafer thickness to be measured during the transfer process itself, eliminating the need for separate inspection equipment and processes, thereby improving productivity while maintaining measurement precision
Solution Approach 2:
The system maintains continuous operation by performing thickness measurement during the transfer process without interrupting the workflow. The light source continuously illuminates the wafer during transfer, and the camera captures images in real-time, allowing measurement to occur continuously alongside the transfer operation rather than requiring separate inspection time
3Manufacturing precision
If machine learning model is trained with comprehensive data set, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The system employs a self-service approach where the inspection apparatus uses its own captured images and the transfer robot's operational data to train the thickness estimation model. By leveraging readily available data from the existing transfer and imaging systems, the model is trained without requiring additional complex measurement equipment or external data sources, thus improving accuracy while limiting complexity growth
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 solution enables rapid and accurate determination of process completion, reducing production delays and costs by integrating the inspection process into the transfer path, thereby enhancing productivity and yield in semiconductor production.
Implementation Method 1
a light source configured to emit light onto the wafer while on the transfer path; a camera configured to acquire an original image of the wafer based on the light reflected from the wafer
Data Source
AI summary
A thickness estimating apparatus includes a transfer robot, a light source, a camera, a memory and a controller. The memory stores a thickness predicting model generated based on a data set including a thickness of at least one of a test wafer corresponding to the wafer or a test element layer formed on the test wafer, and the thickness predicting model being trained to minimize a loss function of the data set. The controller applies pixel data, which is acquired from at least one pixel selected from a plurality of pixels included in a captured image, to the thickness predicting model, to predict a thickness of at least one of the wafer or an element layer formed on the wafer in a position corresponding to a position of the selected pixel.