Scatterometry Overlay Metrology Using Raw Signal Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing overlay metrology methods rely on asymmetry measurements, which are prone to errors due to coupling with other asymmetries, and often use specialized targets that do not conform to design rules, leading to inaccurate measurements of overlay errors in semiconductor devices.
Innovation Solution
The development of a measurement model based on raw scatterometry data that is trained using sites with known overlay variations, allowing for direct measurement of overlay errors and additional parameters like focus and critical dimensions, using design rule targets that adhere to semiconductor manufacturing processes, and employing infrared optical measurement systems to penetrate opaque structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If asymmetry-based scatterometry overlay measurement is used, then overlay error can be characterized, but measurement accuracy deteriorates due to coupling with other asymmetries such as line profile asymmetry or beam illumination asymmetry
Solution Approach 1:
The patent extracts and removes the harmful asymmetry components from the measurement signal. By separating the overlay-induced asymmetry from other asymmetry sources (line profile asymmetry, beam illumination asymmetry), the method isolates the true overlay error signal, thereby improving measurement accuracy without being contaminated by other asymmetries.
Solution Approach 2:
The patent intentionally introduces known asymmetry through asymmetric target structures with programmed overlay offsets. By using asymmetric targets designed with specific known asymmetries, the method creates a controlled asymmetry signal that can be differentiated from unwanted asymmetries, enabling accurate overlay measurement through differential comparison.
2Measurement precision
If specialized target structures are used for overlay measurement, then overlay error can be measured, but measurement accuracy deteriorates because the targets do not conform to design rules and do not represent actual device structures
Solution Approach 1:
The patent applies different qualities to different parts of the measurement system. The target structures are designed to have specific local properties (asymmetric patterns with programmed offsets) that are optimized for measurement sensitivity, while the overall target design conforms to global design rules. This allows the measurement region to have enhanced measurement capability without compromising the representativeness of actual device structures.
Solution Approach 2:
The patent changes the structural parameters of the target structures to create asymmetric patterns with specific geometric properties. By adjusting parameters such as line widths, spacing, and pattern symmetry, the targets are designed to produce measurable asymmetry signals while still adhering to design rules and representing actual device structures, thereby resolving the contradiction between measurement sensitivity and design rule conformity.
3Power
If large pitch targets are used to generate sufficient signal, then signal strength is improved, but measurement applicability deteriorates because actual device pitches are much smaller
Solution Approach 1:
The patent transitions from relying solely on pitch size to generate signal to using asymmetric pattern design in a different dimension (geometric configuration) to enhance signal strength. By introducing asymmetry as a new dimension of signal generation, the method achieves sufficient signal strength without requiring large pitch targets, enabling measurement on targets with pitch sizes comparable to actual device structures.
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 reduces errors and approximations associated with traditional methods, provides more accurate predictive results, and streamlines the measurement process by using only raw scatterometry data, improving computation and user time efficiency while minimizing sensitivity to systematic errors and asymmetries.
Implementation Method 1
In some examples, the metrology system employed to perform overlay measurements includes an infrared optical measurement system. In these examples, the infrared illumination light penetrates opaque structures disposed between layers of patterned structures employed to evaluate overlay.
Data Source
AI summary
Methods and systems for creating a measurement model based only on measured training data are presented. The trained measurement model is then used to calculate overlay values directly from measured scatterometry data. The measurement models receive scatterometry signals directly as input and provide overlay values as output. In some embodiments, overlay error is determined from measurements of design rule structures. In some other embodiments, overlay error is determined from measurements of specialized target structures. In a further aspect, the measurement model is trained and employed to measure additional parameters of interest, in addition to overlay, based on the same or different metrology targets. In some embodiments, measurement data from multiple targets, measurement data collected by multiple metrologies, or both, is used for model building, training, and measurement. In some embodiments, an optimization algorithm automates the measurement model building and training process.