Subsystem Health Check Model for Plasma Chamber Etch Uniformity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current virtual metrology methods for qualifying plasma processing chambers are expensive, time-consuming, and lack accuracy in determining uniformity, relying on expensive film substrates and requiring frequent updates due to changing chamber conditions, and are limited by the granularity of data collected by sensors.
Innovation Solution
The development of a subsystem health check (SSHC) predictive model that uses data from less expensive non-film substrates, such as bare silicon, to correlate sensor data and construct a robust predictive model capable of accounting for chamber drift and variations, employing highly granular sensors and a fast processing computing engine for data analysis, allowing for reduced maintenance costs and faster qualification of processing chambers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a metrology method using actual metrology tools is employed to qualify a processing chamber, then measurement accuracy is improved, but the cost and time consumption increase significantly
Solution Approach 1:
The patent creates a virtual metrology system that generates synthetic measurement data through computer simulations, replacing the need for actual physical metrology tools. The virtual system copies the essential measurement functions while eliminating the time-consuming nature of physical measurements, allowing rapid chamber qualification without sacrificing measurement accuracy.
Solution Approach 2:
The patent substitutes physical metrology tools and mechanical measurement processes with a computational model-based system. By using virtual metrology that relies on process physics models and sensor data rather than physical measurement instruments, the system eliminates the time consumption associated with actual tool measurements while maintaining measurement accuracy through sophisticated algorithms.
2Measurement precision
If actual metrology tools are used to measure substrate properties, then measurement accuracy is improved, but the cost increases
Solution Approach 1:
The virtual metrology system creates synthetic measurements through computational models, replacing expensive physical metrology tools. This copying approach maintains measurement accuracy by using validated process models while eliminating the high costs associated with purchasing, maintaining, and operating actual metrology equipment.
Solution Approach 2:
The patent uses inexpensive virtual measurement models instead of expensive physical metrology tools. The computational models can be rapidly updated and discarded if needed, replacing the need for costly, long-lived physical instruments while maintaining measurement accuracy through continuous model validation and refinement.
3Measurement precision
If measurements are taken after a substrate lot is processed, then measurement accuracy is improved, but productivity decreases due to delayed problem identification
Solution Approach 1:
The virtual metrology system performs measurements and chamber qualification in advance, before actual substrate lots are processed. By using process models and real-time sensor data to predict chamber performance, the system identifies problems beforehand, allowing preventive actions to be taken and avoiding delays in production while maintaining measurement accuracy.
Solution Approach 2:
The system implements continuous feedback loops where sensor data from the processing chamber is fed into the virtual metrology models in real-time. This allows the system to monitor chamber conditions continuously, detect deviations from expected performance immediately, and trigger alerts or corrective actions without waiting for post-processing measurements, thereby maintaining both accuracy and productivity.
Data Source
AI summary
A method for predicting etch rate uniformity for qualifying health status of a processing chamber during substrate processing of substrates is provided. The method includes executing a recipe and receiving processing data from a first set of sensors. The method further includes analyzing the processing data utilizing a subsystem health check predictive model to determine calculated data, which includes at least one of etch rate data and uniformity data. The subsystem health check predictive model is constructed by correlating measurement data from a set of film substrates with processing data collected during analogous processing of a set of non-film substrates. The method yet also includes performing a comparison of the calculated data against a set of control limits as defined by the subsystem health check predictive model. The method yet further includes generating a warning if the calculated data is outside of the set of control limits.