Hyperspectral Plasma Imaging for Fast Etch Endpoint Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional spectrometers face a tradeoff between signal-to-noise ratio (SNR) and throughput in semiconductor manufacturing processes, particularly in low-open area plasma etching, where achieving higher SNR requires sacrificing spectral resolution or reducing acquisition rate, which is detrimental for precise endpoint detection in cyclical processes like atomic layer etching.
Innovation Solution
The implementation of hyperspectral imaging technology, utilizing a wavelength tunable filter and array detector, allows for the collection and processing of optical signals across a wide range of wavelengths without the limitations of traditional spectrometers, maintaining high SNR and spectral resolution while enabling fast acquisition rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional spectrometers are used to increase signal-to-noise ratio in low-open area plasma etching, then measurement precision improves, but throughput decreases due to reduced acquisition rate
Solution Approach 1:
The patent transitions from conventional spectrometry (measuring intensity vs. wavelength) to hyperspectral imaging (measuring intensity vs. x-y position vs. wavelength), adding a spatial dimension to the measurement. This allows simultaneous acquisition of spectral information across multiple spatial locations, effectively increasing throughput while maintaining measurement precision through parallel data collection.
Solution Approach 2:
The patent segments the measurement process by dividing the plasma etching chamber into multiple spatial regions that can be monitored simultaneously using array detectors. This segmentation allows parallel acquisition of spectral data from different locations, increasing the effective acquisition rate and throughput without sacrificing signal-to-noise ratio in any individual region.
2Measurement precision
If conventional spectrometers increase spectral resolution, then measurement precision improves, but acquisition rate decreases
Solution Approach 1:
By adding the spatial dimension through hyperspectral imaging with array detectors, the system captures full spectral information across multiple wavelengths and spatial positions simultaneously. This parallel acquisition maintains high spectral resolution while dramatically increasing the acquisition rate compared to sequential scanning in conventional spectrometers.
Solution Approach 2:
The patent merges spectral resolution capabilities with high-speed acquisition by combining wavelength-dispersive optics with spatially-resolved array detectors. This integration allows the system to maintain the spectral resolution of conventional spectrometers while achieving the acquisition speeds necessary for real-time monitoring of fast transient plasma phenomena.
3Productivity
If conventional spectrometers maintain high acquisition rate, then productivity improves, but signal-to-noise ratio decreases
Solution Approach 1:
The hyperspectral imaging approach captures spectral information across the entire field of view simultaneously, allowing high acquisition rates to be maintained while collecting sufficient photons across multiple spatial and spectral channels. This multidimensional data collection increases the effective signal-to-noise ratio even at high speeds.
Solution Approach 2:
The hyperspectral imager serves multiple functions simultaneously: it provides spatial mapping of plasma properties, spectral identification of chemical species, and temporal resolution of dynamic processes. This multi-functionality allows the system to maintain high acquisition rates while achieving superior signal-to-noise ratio through the combined information from all measurement dimensions.
4Loss of information
If hyperspectral imaging is implemented, then full spectral information is obtained, but device complexity increases
Solution Approach 1:
While hyperspectral imaging does add complexity, the patent justifies this by demonstrating that the additional spatial and spectral dimensions provide comprehensive information about plasma composition, temperature, and dynamics that cannot be obtained with conventional spectrometry. The increased information completeness enables new capabilities in process monitoring and control.
Solution Approach 2:
The patent employs intermediary optical elements such as imaging spectrometers and wavelength-tunable filters that bridge the gap between the plasma source and detectors. These intermediaries enable the complex hyperspectral measurement function while managing the inherent system complexity through well-established optical design principles.
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 enhances the ability to accurately diagnose and control semiconductor manufacturing processes by providing full spectral information without compromising throughput or spectral resolution, improving endpoint detection and process monitoring in plasma and non-plasma etching processes.
Implementation Method 1
a wavelength tunable filter for tunably selecting a wavelength of the transmitted optical signal
Implementation Method 2
an array detector for detecting the wavelength-filtered optical signal
Data Source
AI summary
Disclosed are embodiments of an improved apparatus and system, and associated methods for optically diagnosing a semiconductor manufacturing process. A hyperspectral imaging system is used to acquire spectrally-resolved images of emissions from the plasma, in a plasma processing system. Acquired hyperspectral images may be used to determine the chemical composition of the plasma and the plasma process endpoint. Alternatively, a hyperspectral imaging system is used to acquire spectrally-resolved images of a substrate before, during, or after processing, to determine properties of the substrate or layers and features formed on the substrate, including whether a process endpoint has been reached; or before or after processing, for inspecting the substrate condition.