Semiconductor Target Location Correction Using Temperature Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor manufacturing processes face challenges in accurately determining defect locations and measuring characteristics on specimens due to temperature-induced position errors, leading to inefficiencies and inaccuracies in metrology and inspection processes.
Innovation Solution
A system and method utilizing deep learning models to predict and correct target location errors based on measured temperatures, integrating temperature sensors and a computer subsystem to improve specimen placement accuracy in metrology and inspection tools.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If temperature-induced position errors are not corrected, then the system operates without additional complexity, but measurement precision and manufacturing precision deteriorate
Solution Approach 1:
The system performs preliminary temperature measurement and error prediction before the actual metrology or inspection process. By measuring temperature in advance and predicting position errors using a deep learning model, the system can correct target locations beforehand, eliminating the need for complex real-time compensation mechanisms during measurement.
Solution Approach 2:
The patent introduces temperature sensors and a deep learning model as intermediary components between the temperature field and the target location correction. These intermediaries translate temperature variations into predicted position errors, enabling accurate correction without requiring direct complex mechanical or optical compensation systems.
2Measurement precision
If site-by-site alignment is performed, then measurement precision improves, but productivity decreases due to time-consuming sequential alignment
Solution Approach 1:
The patent replaces the traditional mechanical or manual site-by-site alignment process with a computational approach using deep learning models. Instead of physically adjusting each target location sequentially, the system uses temperature data and neural network predictions to automatically calculate and apply corrections to all target locations, dramatically reducing alignment time while maintaining precision.
Solution Approach 2:
The system performs preliminary temperature measurement and error prediction for all target locations before the actual measurement process. This advance preparation allows simultaneous or parallel processing of multiple target corrections, eliminating the sequential nature of traditional alignment and significantly improving productivity.
3Manufacturing precision
If temperature measurement and correction processes are implemented, then manufacturing precision improves, but device complexity increases due to additional sensors and processing
Solution Approach 1:
The patent makes the temperature measurement system multi-functional by using the same temperature sensors and deep learning model for multiple purposes: predicting position errors, correcting target locations, and potentially monitoring other thermal effects. This universal approach maximizes the value of added components, reducing the relative impact of increased complexity while achieving improved precision.
4Measurement precision
If deep learning models are used for error prediction, then measurement precision improves, but loss of time increases due to computational processing requirements
Solution Approach 1:
The system performs temperature measurements and deep learning predictions in advance, before the actual metrology or inspection process begins. This preliminary computational work allows the system to have corrected target locations ready, eliminating time-consuming calculations during the critical measurement phase and thus reducing the perceived loss of time.
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
Enhances the accuracy of specimen placement and measurement processes by compensating for temperature-induced errors, reducing the need for site-by-site alignment and improving the efficiency of metrology and inspection operations.
Implementation Method 1
temperature-induced position errors
Implementation Method 2
a deep learning (DL) model configured for predicting error in at least one of the one or more target locations based on at least one of the one or more measured temperatures
Data Source
AI summary
Methods and systems for determining information for a specimen are provided. One system includes an output acquisition subsystem configured to generate output for a specimen at one or more target locations on the specimen and one or more temperature sensors configured to measure one or more temperatures within the system. The system also includes a deep learning model configured for predicting error in at least one of the one or more target locations based on at least one of the one or more measured temperatures input to the deep learning model by the computer subsystem. The computer subsystem is configured for determining a corrected target location for the at least one of the one or more target locations by applying the predicted error to the at least one of the one or more target locations.