Semiconductor Wafer Height Difference Measurement via SEM Shadow Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for measuring height difference in patterns on semiconductor wafers require individual measurement of absolute heights, which is inefficient and not suitable for features that may not be physically adjacent, while also failing to account for width differences effectively.
Innovation Solution
A novel method using a height difference determination model expressed in terms of relative brightness or change in distance between characteristic points on SEM images, allowing for the calculation of height and width differences without separate absolute height measurements, applicable to both adjacent and non-adjacent features on the same or different wafers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If individual absolute height measurements are performed for each feature, then measurement accuracy is improved, but measurement time and process complexity increase significantly
Solution Approach 1:
The measurement process is segmented into two parts: (1) a single reference feature is measured to establish absolute height calibration, and (2) all other features are measured as relative height differences from the reference. This segmentation eliminates the need to individually measure absolute heights of all features, dramatically reducing measurement time while maintaining accuracy through the reference-based calibration approach.
Solution Approach 2:
The reference feature is extracted as a separate calibration element from the group of features to be measured. By measuring the reference feature's absolute height once and using it as a baseline, the system extracts the time-consuming absolute measurement step from the routine measurement process, converting it into a one-time calibration operation that enables rapid relative measurements of all other features.
2Measurement precision
If multiple sets of measurement data from different angular orientations are collected, then three-dimensional mapping accuracy is improved, but device complexity and measurement process complexity increase
Solution Approach 1:
The patent introduces the dimension of tilt angle variation into the measurement process. By capturing SEM images at multiple tilt angles and analyzing the变化 in shadow depth and characteristic point distances, the system achieves three-dimensional mapping information without requiring complex multi-angle physical repositioning of the sample or detector, simplifying the measurement system while maintaining 3D mapping accuracy.
3Loss of information
If conventional shadow depth analysis is used, then height information is obtained, but width differences between features are not effectively accounted for
Solution Approach 1:
The patent merges shadow depth analysis with characteristic point distance measurement into a unified measurement approach. By simultaneously analyzing both the shadow depth and the distance between characteristic points on features imaged at different tilt angles, the system extracts both height and width difference information in a single integrated process, ensuring that width variations are properly accounted for in the height difference calculations.
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
Enables accurate and efficient determination of height and width differences between pattern features on semiconductor wafers, improving process control and precision in semiconductor manufacturing by utilizing SEM images and a computer processor-based system.
Implementation Method 1
Scanning Electron Microscope (SEM) images are used in various mapping and imaging applications
Implementation Method 2
The SEM collects each set of data from a different angular orientation with respect to the device
Implementation Method 3
a measurable variable of the model is expressed in terms of a function of a change in a depth of shadow (i.e. relative brightness), wherein the depth of shadow depends on the height difference
Data Source
AI summary
An improved technique for determining height difference in patterns provided on semiconductor wafers uses real measurements (e.g., measurements from SEM images) and a height difference determination model. In one version of the model, a measurable variable of the model is expressed in terms of a function of a change in depth of shadow (i.e. relative brightness), wherein the depth of shadow depends on the height difference as well as width difference between two features on a semiconductor wafer. In another version of the model, the measurable variable is expressed in terms of a function of a change of a measured distance between two characteristic points on the real image of a periodic structure with respect to a change in a tilt angle of a scanning electron beam.