Dual-Wavelength Reflectance Layer Thickness Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the thickness of layers in double-layer stacks, such as in semiconductor wafers, are limited by their reliance on neglecting the influence of the second layer, leading to systematic errors and inability to achieve high accuracy and resolution simultaneously.
Innovation Solution
The method involves measuring the intensity of reflected light in two different wavelength ranges, allowing for the determination of both layer thicknesses by minimizing the influence of one layer at each wavelength, and using calibration curves to accurately calculate the thicknesses without neglecting the combined influence of both layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ellipsometric measurements are used to determine layer thickness, then measurement accuracy is improved, but measurement time increases to several tens of seconds per point
Solution Approach 1:
The patent segments the spectral measurement into multiple discrete wavelength points rather than performing a full spectral scan. By selecting specific wavelength points where the derivative of reflectivity with respect to layer thickness is maximized, the measurement achieves high accuracy while reducing the number of measurement points required, thus decreasing measurement time.
Solution Approach 2:
The patent changes the measurement parameter from a full spectral range to specific discrete wavelength points. By identifying wavelengths where the optical response is most sensitive to thickness changes (maximizing the derivative dR/dt), the method achieves ellipsometry-level accuracy with reduced measurement time, as only critical wavelength points need to be measured rather than the entire spectrum.
2Device complexity
If a small wavelength range is selected to minimize the influence of the second layer, then the measurement of the first layer is simplified, but systematic errors occur due to neglecting the combined influence of both layers
Solution Approach 1:
The patent uses a feedback approach by iteratively adjusting the estimated thickness values of both layers until the calculated reflectivity spectrum matches the measured spectrum. This iterative refinement process accounts for the combined influence of both layers, eliminating systematic errors while maintaining a relatively simple measurement model that can be efficiently implemented.
Solution Approach 2:
The patent transitions from measuring only the first layer thickness to simultaneously determining both layer thicknesses by utilizing the spectral information at multiple wavelength points. This dimensional expansion in the parameter space (from one thickness parameter to two) allows the method to account for the combined influence of both layers and resolve the systematic errors that arise when only one layer is considered.
3Productivity
If reflectance measurements are performed to determine layer thickness, then the method is faster than ellipsometry, but spatial resolution is restricted to a few tens of μm
Solution Approach 1:
The patent employs dynamic focusing capabilities to achieve high spatial resolution in reflectance measurements. By implementing focus variation techniques that allow rapid adjustment of the focal plane, the system can achieve lateral resolutions down to a few micrometers while maintaining the speed advantages of reflectance measurement, thus overcoming the spatial resolution limitation of conventional reflectance methods.
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 accurate and simultaneous measurement of layer thicknesses with high accuracy and resolution, reducing systematic errors and improving measurement speed and lateral resolution.
Implementation Method 1
measuring a reflectivity of the sample stack in two different wavelength ranges
Implementation Method 2
the intensity of the reflected light versus thickness variation is nearly zero within the selected small wavelength range. This idea is based on the fact, that the intensity of the light reflected by a two layer stack on top of a substrate varies strongly with the wavelength range due to the interference of the optical waves
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for determining the thickness of a layer in a sample stack of at least two layers with an assembly comprising a light source for illuminating a stack of layers and a detector for detecting light reflected by the stack of layers in a defined first wavelength range, the method comprises a first step of obtaining a calibration curve by the calibrating steps of providing two or more reference stacks of layers, where each layer of the reference stacks has a known thickness, the same material as the sample stack and the layers occur in the same order as in the sample stack; illuminating the reference stacks with light from the light source; and detecting the intensity of light reflected by the reference stacks with the detector in the first wavelength range. Further steps of the method comprise illuminating the sample stack of layers with light from the light source; detecting the intensity of the light reflected by the sample stack of layers with the detector in the first wavelength range; and determining the thickness of the layer comprised in the sample stack of layers from the intensity detected by the detector by means of the calibration curve. The method is characterized in that further calibration curves are obtained of reference stacks, where the thickness of another, different layer is known also, thereby providing a first series of calibration curves in the first wavelength range; a second plurality of calibration curves is obtained in the same way as the first series of calibration curves for a further wavelength range; and the thickness of the layer of the sample stack is determined from the intensity detected by the detector by means of the first and second series of calibration curves.