3D Wafer Reconstruction From Diagonal SEM Cross-Sections

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current 3D volumetric sampling methods for semiconductor wafers are either limited in application scope, time-consuming, or destructive, and cannot be conducted in-line, hindering efficient high-resolution inspection of 3D structural elements.

Innovation Solution

An algorithm that converts a 2D SEM image of a diagonally cut semiconductor structure into a 3D reconstruction, allowing for high-resolution, non-destructive, and efficient 3D volumetric inspection by combining layers of the structure exposed at different depths through a single diagonal cut, with optional sequential delayering for enhanced resolution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If TEM or CD-SAXS is used for 3D volumetric sampling, then measurement precision is improved, but loss of time increases and the process becomes destructive

Engineering Contradiction:
Improve3D volumetric sampling precisionVSAvoidtime-to-results
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The wafer is divided into multiple discrete cut locations where individual cuts are made to expose different depth layers. Each cut provides a 2D cross-section that is then used to reconstruct the 3D structure, breaking down the complex 3D measurement into manageable 2D segments that can be processed faster

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of using the entire wafer for measurement (destructive), only small discrete cut locations are sacrificed. The 3D structure is reconstructed by copying and assembling information from multiple 2D cross-sectional images, creating a virtual 3D model from 2D copies of different depth slices

Inventive Principle:
Principle #26Copying

2Manufacturing precision

If FIB is used for high-resolution imaging, then manufacturing precision is improved, but the process becomes destructive to the specimen

Engineering Contradiction:
Improveimaging resolutionVSAvoidspecimen damage
Core Design Contradiction:
Manufacturing precisionVSObject-affected harmful factors

Solution Approach 1:

High-resolution FIB imaging is applied only at specific local cut locations rather than across the entire wafer. Low beam currents are used for imaging at these discrete locations, minimizing material removal and damage while achieving 5 nm resolution where needed

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

Instead of performing extensive FIB milling across the whole wafer, the method uses partial action by making only the necessary discrete cuts at selected locations. This partial approach achieves the required imaging resolution while significantly reducing the harmful effects of FIB on the overall specimen

Inventive Principle:
Principle #16Partial or excessive action

3Adaptability or versatility

If multiple imaging methods are used to cover different application spaces, then adaptability is improved, but device complexity increases

Engineering Contradiction:
Improveapplication space coverageVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent creates a universal 3D reconstruction algorithm that can process images from multiple different imaging methods (SEM, FIB, TEM, CD-SAXS, scatterometry). This single multi-functional algorithm replaces the need for separate specialized analysis tools for each imaging method, reducing system complexity while maintaining broad adaptability across logic and memory applications

Inventive Principle:
Principle #6Universality (Multi-functionality)

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 high-resolution 3D reconstruction and measurement of semiconductor structures with minimal wafer area sacrifice, suitable for in-line implementation, reducing time-to-results and maintaining structural integrity.

Implementation Method 1

Scanning electron microscope (SEM) produces images of a sample (such as wafer samples) by scanning the surface with a focused beam of electrons. The electrons interact with atoms in the sample, producing various signals that contain information about the surface topography and composition of the sample.

Methodology Applied
Scientific EffectElectron beam interaction: Electron Beam

Implementation Method 2

At higher primary currents, a great deal of material can be removed by sputtering, allowing precision milling of the specimen down to a sub micrometer or even a nano scale.

Methodology Applied
Scientific EffectSputtering: Sputtering

Data Source

PatentUS20250391114A1Algorithm for 3D reconstruction of diagonally cut semiconductor logic structure
Publication Date: 2025.12.25 APPL MATERIALS ISRAEL LTD
  • US20250391114A1 patent drawing
  • US20250391114A1 patent drawing
  • US20250391114A1 patent drawing

AI summary

A method for generating a 3D reconstruction of structural elements of a wafer including: obtaining a 2D image of a top view of the plurality of structural elements, sectioned diagonally in a compound angle allowing 3D volumetric sampling, such that each of the plurality of structural elements is cut at a different height thereof, and generating a representative 3D reconstruction of the structural elements, based on the images.