Joint Etch Exposure Mask Model Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mask process models in integrated circuit manufacturing are calibrated based on single effects such as etch or exposure, failing to accurately account for combined etch and exposure effects, leading to defects in smaller feature sizes due to pattern-dependent effects.
Innovation Solution
A method for jointly calibrating a mask process model by identifying and combining etch and exposure models, followed by a global optimization process using measured data to minimize the difference between simulated and actual mask contours, with a cost function representing the norm of this difference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If mask process models are calibrated based on single effects (etch or exposure), then the calibration process is simpler and faster, but the accuracy of simulating mask contours deteriorates due to inability to account for combined effects
Solution Approach 1:
The patent merges separate etch model calibration and exposure model calibration into a single joint calibration process. The cost function combines both etch and exposure effects, allowing simultaneous optimization of both models using the same measured data, thereby improving accuracy while managing complexity through integration rather than separate processes
Solution Approach 2:
The joint calibration model serves multiple functions by simultaneously accounting for both etch effects and exposure effects in a single framework. The unified cost function and optimization process handle both effects, making the calibration system more versatile and accurate without requiring separate specialized calibration procedures
2Reliability
If separate etch and exposure models are used, then the model calibration is more straightforward, but the fidelity of the optical lithographic process simulation deteriorates due to pattern-dependent effects
Solution Approach 1:
The patent combines etch and exposure models into a unified joint calibration framework where both effects are simultaneously optimized. The combined cost function integrates both effects, allowing the models to account for their interactions and pattern-dependent effects, thereby improving reliability through comprehensive modeling
Solution Approach 2:
The joint calibration process uses measured mask contour data as feedback to simultaneously adjust both etch and exposure model parameters. This feedback mechanism allows the models to converge on accurate representations of both effects, improving simulation fidelity through iterative optimization based on actual measurements
Data Source
AI summary
Methods for jointly calibrating etch and exposure mask process models from etch only data are described. Initially, an etch model and an exposure model may be identified. Subsequently, a combined etch/exposure model may be generated based upon the etch model and the exposure model. Following which, a global optimization process may be performed to calibrate the combined etch/exposure model based upon measured data representing the etch and the exposure effects. With some implementations, the global optimization process is based in part upon a cost function representing the norm of the difference between the simulated mask contours and the measured mask contours. Furthermore, in some implementations, the optimization variable set is the union of the parameter sets corresponding to the etch model and the exposure model individually. Further still, with various implementations, the optimization of based upon the etch parameter set is “nested” inside an optimization of the exposure parameter set, or, vice versa.