Optical Model Adjustment for OPC Effect Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor design lacks accurate modeling of downstream processes like optical proximity correction (OPC), leading to manufacturing problems being identified late in the design flow, and existing model-based OPC techniques do not effectively capture the effects of subsequent processes.
Innovation Solution
A system that receives a test layout and an optical model not accounting for OPC effects, generates datasets before and after applying OPC, and adjusts the model using these datasets to create an enhanced optical model that includes OPC effects, allowing for the prediction of patterns post-photolithography and subsequent processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a standard optical model is used that does not account for OPC effects, then the model is simpler and faster to compute, but manufacturing problems cannot be accurately predicted at design time
Solution Approach 1:
The patent introduces an intermediary matrix that captures OPC effects separately from the base optical model. This matrix acts as a mediator that translates standard optical model outputs into predictions that account for OPC, allowing accurate manufacturing outcome prediction without requiring a completely complex custom model.
Solution Approach 2:
The patent transforms the optical model by applying parameter changes through a matrix that encapsulates OPC effects. Instead of fundamentally redesigning the optical model, the approach modifies its output parameters to reflect OPC-influenced manufacturing outcomes, maintaining model simplicity while improving prediction accuracy.
2Reliability
If downstream processes like OPC are not modeled at design time, then the design flow is faster, but manufacturing problems are identified too late
Solution Approach 1:
The patent enables preliminary action by incorporating OPC effects into the optical model at the design stage. This allows manufacturing problems to be identified and addressed before fabrication, rather than discovering them later in the process. The matrix-based approach makes this preliminary analysis feasible without excessive time cost.
3Measurement precision
If an enhanced optical model that includes OPC effects is created, then manufacturing prediction accuracy improves, but computational intensity increases
Solution Approach 1:
The patent segments the modeling task into two parts: the base optical model and the OPC effects matrix. This segmentation allows the computationally intensive OPC modeling to be pre-computed and stored as a matrix, which can then be efficiently applied to design layouts without repeating full OPC simulations for each case, reducing overall computational energy requirements.
4Manufacturing precision
If full OPC modeling is performed at design time, then manufacturing issues can be caught early, but processing time increases significantly
Solution Approach 1:
The patent creates a simplified copy or representation of OPC effects through the matrix, which captures the essential manufacturing outcomes without requiring full OPC simulation. This matrix copy can be quickly applied to evaluate manufacturing precision for design layouts, maintaining high conformity prediction capability while preserving design flow productivity.
Data Source
AI summary
One embodiment provides a system that can enable a designer to determine the effects of subsequent processes at design time. During operation, the system may receive a test layout and an optical model that models an optical system, but which does not model the effects of subsequent processes, such as optical proximity correction (OPC). The system may generate a first dataset using the test layout and the optical model. Next, the system may apply OPC to the test layout, and generate a second dataset using the corrected test layout and the optical model. The system may then use the first dataset and the second dataset to adjust the optical model to obtain a second optical model that models the effects of subsequent processes.