Lithography Pattern Gauge Selection Using AI Depth Sensitivity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in semiconductor manufacturing is the accurate reproduction of patterns with dimensions smaller than the classical resolution limit of lithographic projection apparatus, where sophisticated fine-tuning steps like optical proximity correction (OPC) are applied, but the ability to monitor and detect IC non-idealities is limited by image quality and alignment in inspection systems, particularly for complex structures.
Innovation Solution
A method involving characterizing depth variations in lithography simulations, evaluating aerial image sensitivity, and selecting patterns or gauges based on this sensitivity to optimize optical proximity correction (OPC) and stochastic modeling, using SEM images to refine patterns and improve model building.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If optical proximity correction (OPC) is applied to reproduce patterns with dimensions smaller than the classical resolution limit, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent applies preliminary action by performing lithography simulations before actual manufacturing to predict and characterize depth variations in pattern features. This allows optical proximity correction parameters to be pre-determined and optimized, reducing the complexity of subsequent fine-tuning steps while maintaining high manufacturing precision for sub-resolution patterns.
Solution Approach 2:
The patent uses simulated aerial images as copies of the actual lithography process outcomes. By characterizing depth variations from simulations rather than requiring direct measurement of every pattern feature, the system reduces device complexity while maintaining manufacturing precision through virtual modeling and analysis.
2Measurement precision
If inspection systems are used for defect detection, then manufacturing precision is improved, but image quality and throughput are limited by alignment and calibration issues
Solution Approach 1:
The patent uses simulated aerial images as virtual copies of actual lithography patterns for inspection and characterization. This eliminates the need for physical alignment and calibration of inspection systems, as the simulations can be generated directly from lithography process parameters, thereby improving throughput while maintaining measurement precision for defect detection.
Solution Approach 2:
The patent introduces lithography simulations as an intermediary between the lithography process and inspection/characterization. This intermediary layer provides depth variation characterization and aerial image analysis without requiring direct physical measurement, resolving the alignment and calibration limitations of traditional inspection systems and improving overall throughput.
3Manufacturing precision
If depth variation characterization is performed to select patterns for OPC, then manufacturing precision is improved, but loss of time increases
Solution Approach 1:
The patent performs depth variation characterization and pattern selection through lithography simulations as a preliminary action before actual manufacturing. By pre-characterizing aerial images and identifying patterns with acceptable depth variations in advance, the system optimizes OPC parameters beforehand, reducing time-consuming iterative adjustments during production while maintaining high manufacturing precision.
Data Source
AI summary
Methods, computer programs, and systems are disclosed, with one method including characterizing a depth variation of a predicted result within a feature of a pattern from a lithography simulation. The method evaluates the depth variation characterization and selects patterns or gauges based on the depth variation evaluation. In some embodiments, the evaluating can be based on an aerial image (AI) depth sensitivity having the depth variation.