OPC Knowledge Base for SRAF Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional methods for reducing the printing of sub-resolution assist features (SRAFs) in integrated circuits (ICs) are inaccurate and labor-intensive, leading to operational volatility and reduced fabrication output due to inconsistent classification and prediction techniques.

Innovation Solution

A method and system for creating a knowledge base using a predictive algorithm to identify and adjust for printed SRAFs in ICs, which involves fabricating circuits, identifying features, predicting SRAF presence, adjusting the algorithm based on correct or incorrect predictions, and storing data for training, thereby improving prediction accuracy over time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional classification techniques (edge detection filters or manual classification) are used to identify printed SRAFs, then the process can detect SRAF presence, but the accuracy is insufficient and requires high engineering hours

Engineering Contradiction:

ImproveSRAF detection accuracyVSAvoidengineering hours for classification

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual classification and edge detection filter methods with a machine learning-based predictive algorithm. The system fabricates test circuits, captures images, and uses trained ML models to automatically predict SRAF presence, eliminating the need for labor-intensive manual classification while improving detection accuracy through pattern recognition capabilities of machine learning

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a knowledge base by fabricating representative test circuits that copy the characteristics of production circuits. These test circuits are used to train predictive algorithms, which then generalize to predict SRAF presence in actual production circuits without requiring direct manual inspection of each production item

Inventive Principle:

Principle #26Copying

2Manufacturing precision

If OPC models include SRAFs to improve printability of isolated features, then the printability of isolated features is enhanced, but the risk of inadvertently printing SRAFs increases

Engineering Contradiction:

Improveprintability of isolated featuresVSAvoidinadvertent SRAF printing

Core Design Contradiction:

Manufacturing precisionVSObject-affected harmful factors

Solution Approach 1:

The patent implements a feedback mechanism where predictive algorithms analyze fabricated circuit images to determine whether SRAFs were inadvertently printed. The system uses this feedback information to refine OPC models and adjust SRAF parameters, creating a closed-loop system that continuously improves at reducing harmful SRAF printing while maintaining the benefits of SRAFs for isolated feature printability

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent performs preliminary prediction of SRAF printing risk before full production by fabricating and analyzing test circuits. The predictive algorithm identifies patterns and conditions that lead to inadvertent SRAF printing, allowing OPC models to be adjusted in advance to prevent harmful SRAF printing in production while maintaining beneficial SRAF effects

Inventive Principle:

Principle #10Preliminary action

3Loss of time

If conventional prediction techniques relying solely on layout attributes are used, then the prediction process can be performed early in design, but the prediction accuracy is insufficient

Engineering Contradiction:

Improveprediction timingVSAvoidprediction accuracy

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The patent performs preliminary prediction by fabricating test circuits and training predictive algorithms before full production. This early prediction phase uses actual fabricated circuit images to train ML models, enabling accurate prediction of SRAF printing risk in production circuits while maintaining early intervention capabilities in the manufacturing process

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent transitions from using only layout attributes for prediction to using actual fabricated circuit images as training data. This parameter change from theoretical layout data to empirical image data significantly improves prediction accuracy by capturing real-world printing variations and effects that layout attributes alone cannot predict

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10386714B2Creating knowledge base for optical proximity correction to reduce sub-resolution assist feature printing

Publication Date: 2019.08.20 TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD

AI summary

Embodiments of the present disclosure include methods, program products, and systems for creating a knowledge base for optical proximity correction (OPC). Methods according to the disclosure can include: fabricating a circuit using a proposed IC layout; identifying a plurality of features in an image of the fabricated circuit; predicting, based on the identifying and a predictive algorithm, whether the fabricated circuit includes a printed sub-resolution assist feature (SRAF) from the proposed IC layout; determining the predicting as being correct when the fabricated circuit includes the printed SRAF, or as being incorrect when the fabricated circuit does not include the printed SRAF; in response to the predicting being incorrect: adjusting the predictive algorithm, and flagging the fabricated circuit as incorrectly predicted; in response to the predicting being correct, flagging the fabricated circuit as correctly predicted; and storing the image of the fabricated circuit in a repository of training data.