Effective Cell Approximation for Non-Periodic Logic Metrology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional optical metrology techniques struggle to accurately model and measure complex, non-periodic structures like random logic cells in semiconductor fabrication due to the lack of periodic patterns, which limits their ability to provide precise characterization.
Innovation Solution
The use of an effective cell approximation, which represents the density of lines and spaces in a standard logic cell using periodic patterns, allowing for the creation of a smaller model or metrology target that can be used to acquire measured data and determine desired characteristics, enabling accurate characterization without requiring precise positioning or rigorous modeling of the entire cell.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional optical metrology techniques are used to model and measure complex non-periodic structures like random logic cells, then measurement capability is maintained for simple periodic structures, but measurement accuracy and applicability deteriorate for complex non-periodic structures due to the lack of periodic patterns required by conventional modeling techniques
Solution Approach 1:
The patent creates a simplified periodic copy (effective cell approximation) of the complex non-periodic structure that replicates the key density characteristics. This copy can be measured using conventional periodic modeling techniques while representing the essential features of the original non-periodic logic cell, thereby extending measurement capability to complex structures without sacrificing accuracy
Solution Approach 2:
The patent transforms the non-periodic structure into a periodic representation by changing the structural parameter from arbitrary layout to regular repeating pattern, while preserving the density parameter (lines and spaces density) that is critical for optical measurement. This parameter transformation enables the use of conventional periodic modeling techniques on non-periodic structures
2Measurement precision
If the entire standard logic cell is modeled and measured with high precision, then accurate characterization is achieved, but measurement time and computational complexity increase significantly
Solution Approach 1:
The patent extracts only the essential density characteristic (lines and spaces density) from the complex logic cell structure, creating a simplified effective cell approximation. This extraction process removes unnecessary geometric details while retaining the critical measurement parameter, thereby reducing measurement time and computational complexity while maintaining characterization accuracy
Solution Approach 2:
Instead of measuring the entire complex logic cell structure, the patent applies partial action by measuring only the essential density representation. This partial measurement approach achieves sufficient characterization accuracy without the time and computational burden of complete structure analysis
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 allows for efficient and accurate measurement of standard logic cells by approximating non-periodic patterns with periodic ones of the same density, facilitating the use of existing modeling techniques and improving measurement throughput and accuracy.
Implementation Method 1
optical metrology equipment to provide non-contact evaluation of substrates during processing. With optical metrology, a sample under test is illuminated with light, e.g., at a single wavelength or multiple wavelengths. After interacting with the sample, the resulting light is detected and analyzed
Data Source
AI summary
Characteristics of a standard logic cell, e.g., a random logic cell, are determined using an effective cell approximation. The effective cell approximation is smaller than the standard logic cell and represents the density of lines and spaces of the standard logic cell. The effective cell approximation may be produced based on a selected area from the standard logic cell and include the same non-periodic patterns as the selected area. The effective cell approximation, alternatively, may represent non-periodic patterns in the standard logic cell using periodic patterns having a same density of lines and spaces as found in the standard logic cell. A structure on the sample, such as a logic cell or a metrology target produced based on the effective cell approximation is measured to acquire data, which is compared to the data for the effective cell approximation to determine a characteristic of the standard logic cell.