Multi-Mode Plasma Endpoint Detection via Trajectory Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing endpoint detection systems for multi-mode pulse processes are inaccurate due to variable conditions, leading to over-etching and potential damage to features being formed, as they rely on optical emission spectrum analysis which is ineffective in multi-mode pulse processes with varying chemistry, pressure, temperature, and RF parameters.
Innovation Solution
A system and method for accurately detecting the endpoint of a multi-mode pulse process by collecting and analyzing process output variables such as optical emission spectra, RF harmonics, and other plasma parameters, using multivariate analysis and template matching to identify specific trends and thresholds, allowing for precise control and real-time detection of process points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If optical emission spectrum analysis is used for endpoint detection in multi-mode pulse processes, then the detection system can operate with standard equipment, but the detection accuracy deteriorates due to variable process conditions
Solution Approach 1:
The patent transforms the variable process conditions (chemistry, pressure, temperature, RF parameters) from obstacles into detection parameters. By monitoring how these parameters change over time and comparing them to reference patterns, the system achieves accurate endpoint detection despite the inherent variability of multi-mode pulse processes
Solution Approach 2:
The system continuously monitors process output variables and compares real-time data against reference patterns stored in memory. This feedback mechanism allows the system to dynamically adjust and identify endpoint conditions by recognizing characteristic parameter trajectories, thereby maintaining high detection accuracy in variable conditions
2Adaptability or versatility
If variable multi-mode pulse process conditions are used to achieve process flexibility, then process adaptability improves, but endpoint detection accuracy deteriorates
Solution Approach 1:
The patent embraces the dynamic nature of multi-mode pulse processes by continuously adapting the detection algorithm to the current process state. The system monitors parameter trajectories and compares them against dynamic reference patterns, allowing accurate endpoint detection that accommodates the inherent variability and flexibility of changing process conditions
Solution Approach 2:
The detection process is segmented into distinct phases corresponding to different process modes. By analyzing parameter changes within each phase and comparing them to phase-specific reference patterns, the system can accurately detect endpoints even as the process transitions between different operational states with varying conditions
3Device complexity
If standard optical spectrum analysis is used, then device complexity remains low, but detection reliability deteriorates in multi-mode pulse processes
Solution Approach 1:
The patent makes the optical emission spectrum analyzer universal by programming it to perform both standard spectral analysis and advanced endpoint detection functions. The system can operate in different detection modes depending on the process type, eliminating the need for specialized equipment while maintaining high reliability across various multi-mode pulse processes
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 enables more accurate and rapid differentiation of endpoints and other process points, providing precise control over multi-mode pulsing plasma processes, reducing overall process time, and increasing production throughput.
Implementation Method 1
The etch byproducts emit a corresponding signature optical spectrum
Data Source
AI summary
A system and method of identifying a selected process point in a multi-mode pulsing process includes applying a multi-mode pulsing process to a selected wafer in a plasma process chamber, the multi-mode pulsing process including multiple cycles, each one of the cycles including at least one of multiple, different phases. At least one process output variable is collected for a selected at least one of the phases, during multiple cycles for the selected wafer. An envelope and/or a template of the collected at least one process output variable can be used to identify the selected process point. A first trajectory for the collected process output variable of a previous phase can be compared to a second trajectory of the process output variable of the selected phase. A multivariate analysis statistic of the second trajectory can be calculated and used to identify the selected process point.