Semiconductor Metrology With Mechanical Stress Modulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current metrology systems face challenges in accurately measuring complex semiconductor structures due to increasing complexity, small resolution requirements, multi-parameter correlation, and the use of opaque materials, particularly in characterizing high-k dielectric layers, leading to unsatisfactory compromises in computational tractability and accuracy.
Innovation Solution
Measure structural parameters by inducing mechanical stress in the measurement target, quantifying changes in electrical and optical properties, and using these derivatives as input to a measurement model to break correlations among different materials, enabling improved sensitivity and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a complex measurement model with multiple parameters is used to represent semiconductor structures, then measurement accuracy is improved, but computational tractability deteriorates due to parameter correlations and large number of floating parameters
Solution Approach 1:
The patent applies parameter changes by introducing mechanical stress modulation as a new control variable that dynamically alters the optical properties of the semiconductor structure. This allows the system to differentiate between parameters that would otherwise be correlated, thereby improving measurement accuracy without requiring an excessive number of fixed parameters in the measurement model.
Solution Approach 2:
The patent introduces dynamic modulation of the measurement target through mechanical stress applied at a known frequency. This dynamic approach transforms static measurement problems into dynamic ones, allowing the system to extract multiple parameter values by analyzing the frequency-dependent response, thus reducing the number of simultaneous parameters that need to be solved.
2Productivity
If parameters are fixed in the measurement model to reduce computational effort and parameter correlations, then calculation speed is improved, but measurement accuracy deteriorates due to errors in parameter value estimates
Solution Approach 1:
The patent applies preliminary action by pre-characterizing the mechanical stress-optical property relationships through separate calibration measurements. These pre-obtained derivative values are stored and used during actual measurements, allowing the system to quickly estimate parameter changes without performing complex real-time optimization, thus maintaining both speed and accuracy.
Solution Approach 2:
The patent uses mechanical stress as an intermediary that couples the measurement system to the semiconductor structure. By modulating the stress at a known frequency, the system creates a controlled perturbation that can be used to extract optical property changes without requiring direct measurement of multiple parameters simultaneously, thereby avoiding the accuracy-loss trade-off.
3Measurement precision
If fill material is applied to the wafer surface to improve measurement accuracy by reducing parameter correlation, then measurement accuracy is improved, but system complexity and contamination risk increase
Solution Approach 1:
The patent replaces the mechanical approach of applying fill material to the wafer surface with a non-contact mechanical stress modulation approach. The stress is applied through a controlled mechanical actuator that induces optical property changes without physical contact with the wafer surface, thereby eliminating contamination risks and fill material handling complexity while achieving the same parameter differentiation effect.
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
Enhances measurement sensitivity and reduces parameter correlations, allowing for precise characterization of complex semiconductor structures such as 3D VNAND and Gate-All-Around structures without contact or destruction, using optical, electron-based, and x-ray based systems.
Implementation Method 1
A mechanical wave excitation source excites a mechanical wave propagating through a measurement target. The mechanical wave induces changes in electrical and optical properties of the measurement target.
Implementation Method 2
The electrical and optical properties of a measurement target are perturbed by exciting a mechanical wave within the measurement target
Data Source
AI summary
Methods and systems measuring structural parameters characterizing a measurement target based on changes in measurement signal values and estimated changes in electrical properties, optical properties, or both, of the measurement target due to variation of mechanical stress are presented herein. The electrical and optical properties of a measurement target are perturbed by inducing a mechanical wave within the measurement target under measurement. In preferred embodiments, the mechanical wave is excited by an ultrasonic actuator in contact with a back side of a wafer under measurement. Both the changes in the measurement signal values and estimated changes in the electrical, properties, optical properties, or both, of the measurement target are quantified and provided as input to a measurement model. In this manner, the measurement is based on the derivatives of measurement signals with respect to electrical properties, optical properties, or both.