X-ray Metrology Parameter Optimization for Semiconductor Structures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional x-ray metrology tools face challenges in measuring complex semiconductor structures due to limited adaptability, precision, and time efficiency, particularly with small-scale, opaque materials, and high-aspect ratio three-dimensional geometries, leading to difficulties in resolving critical parameters like line edge roughness and line width roughness.
Innovation Solution
The optimization of x-ray metrology system parameters through iterative refinement of machine settings, such as angle of incidence, beam energy, and measurement duration, using statistical filters like Kalman filters, to enhance precision, accuracy, and reduce measurement time, while minimizing parameter correlation and computation time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional x-ray metrology tools use fixed measurement system configurations, then the system is simple to operate, but the system lacks adaptability to measure complex semiconductor structures with diverse geometries and materials
Solution Approach 1:
The patent implements dynamic adaptability by enabling the x-ray metrology system to automatically adjust measurement configurations based on the specific target structure being measured. The system dynamically selects appropriate illumination angles, detector positions, and measurement parameters rather than requiring manual configuration for each measurement type.
Solution Approach 2:
The system changes measurement parameters such as x-ray illumination angles, detector distances, and measurement durations based on the characteristics of the target structure. This allows the same physical system to adapt to different measurement scenarios by modifying operational parameters rather than requiring physical reconfiguration.
2Measurement precision
If the measurement system uses a broader range of system parameter values to improve measurement precision, then measurement accuracy improves, but measurement time becomes excessively long
Solution Approach 1:
The system performs preliminary characterization of the target structure using initial measurements or simulation data before conducting the full measurement sequence. This preliminary action allows the system to pre-determine optimal measurement parameters and focus measurements only on the most informative configurations, reducing total measurement time while maintaining precision.
Solution Approach 2:
The patent replaces exhaustive physical measurement sampling with computational simulation and modeling to identify optimal measurement configurations. By using simulations to predict which measurement parameters will provide the most information, the system avoids performing unnecessary measurements, thereby reducing measurement time while maintaining or improving precision.
3Measurement precision
If evenly spaced angle sampling is used for periodic targets, then the measurement setup is simple, but the system cannot resolve particular parameters of interest with sufficient precision
Solution Approach 1:
The system applies local quality by concentrating measurement sampling density in specific angular regions where the measurement signal is most sensitive to the parameters of interest. Rather than uniform sampling, the system performs measurements at non-uniformly spaced angles that are optimized for the specific target structure and measurement objectives, improving parameter resolution where it matters most.
Solution Approach 2:
The patent extends the measurement approach from simple angular sampling to multi-dimensional parameter space optimization, considering combinations of illumination angles, detector positions, and measurement energies simultaneously. This allows the system to resolve parameters that cannot be decoupled with single-parameter variations.
4Reliability
If multiple measurement parameters are varied to improve measurement accuracy, then parameter decoupling improves, but the number of measurements and computation time increase significantly
Solution Approach 1:
The system performs preliminary simulation-based analysis to identify which measurement parameter combinations will provide the most information for decoupling the parameters of interest. This preliminary action allows the system to select a minimal set of measurements that achieve reliable parameter decoupling without requiring exhaustive sampling of all possible parameter combinations.
Solution Approach 2:
The patent uses computational optimization and simulation to replace exhaustive physical measurement campaigns. By using models to predict parameter correlations and measurement information content, the system can determine the minimal measurement set required for reliable decoupling, significantly reducing both measurement and computation time while maintaining accuracy.
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 enables more precise and accurate measurement of structural and material characteristics in semiconductor fabrication, improving the ability to resolve complex parameters and reducing measurement time, thereby addressing the limitations of traditional x-ray metrology systems.
Implementation Method 1
x-ray based metrology system to collect measurement data associated with measurements of one or more target structures
Data Source
AI summary
Methods and systems for optimizing measurement system parameter settings of an x-ray based metrology system are presented. X-ray based metrology systems employing an optimized set of measurement system parameters are used to measure structural, material, and process characteristics associated with different semiconductor fabrication processes with greater precision and accuracy. In one aspect, a set of values of one or more machine parameters that specify a measurement scenario is refined based at least in part on a sensitivity of measurement data to a previous set of values of the one or more machine parameters. The refinement of the values of the machine parameters is performed to maximize precision, maximize accuracy, minimize correlation between parameters of interest, or any combination thereof. Refinement of the machine parameter values that specify a measurement scenario can be used to optimize the measurement recipe to reduce measurement time and increase measurement precision and accuracy.