EquipmentStatus: FOUNDATIONALUpdated 2026-08-24

Electron-Beam Defect Review and Inspection System

A scanning electron-beam system used to image, classify, and sometimes detect wafer defects at higher spatial resolution than production optical inspection.

Evidence status

Checked against 1 inspected source

One source was retrieved, identified and read, and the claims below are tied to specific passages at the scope those passages state. Each source also records what it cannot establish.

Rely on this page for

The specific claims that carry a cited passage, at the scope that passage states.

Definition

An electron-beam defect review or inspection system scans selected wafer locations with a focused electron beam, detects emitted signals, and produces high-resolution images and classifications that connect inspection candidates to physical defect evidence.

Process position

03 · FEOL04 · MEOL05 · BEOL06 · Wafer test

Inputs

  • Wafer and defect coordinates
  • Review recipe
  • Inspection map
  • Classification taxonomy

Outputs

  • High-resolution defect images
  • Classification labels
  • Defect mechanism evidence

How it works

  1. 01Load and align wafer
  2. 02Navigate to candidate coordinates
  3. 03Optimize beam and imaging condition
  4. 04Acquire images or scan area
  5. 05Classify and correlate results

Role in the production flow

E-beam review adds morphological evidence to faster inspection signals and can reveal smaller features, but its throughput and possible beam interaction usually require targeted sampling rather than blanket substitution for optical inspection.

Interested partyCombines sources[1][2]

A scanning electron-beam system used to image, classify, and sometimes detect wafer defects at higher spatial resolution than production optical inspection.

Boundary: The cited manufacturers document the equipment category and its intended uses; this record does not independently validate vendor performance claims or rank products.

Qualification and production boundaries

A review flow must define coordinate accuracy, target defect types, landing conditions, charging mitigation, acceptable beam interaction, image and classification standards, sampling rules, and correlation back to process history.

Bounded inferenceMaha inference[1][2]

A review flow must define coordinate accuracy, target defect types, landing conditions, charging mitigation, acceptable beam interaction, image and classification standards, sampling rules, and correlation back to process history.

Boundary: Actual acceptance limits, recipes, throughput, availability, and ownership economics are product-, process-, site-, and contract-specific.

Sources

Citations support the tagged claims above. Access dates record when Maha Strategies last checked the public source.

  1. [1]Annual Report: Process Control and Yield Management · KLA · 2024 · accessed 2026-08-13
  2. [2]Annual Report: Inspection, Metrology, and Yield Analysis · KLA · 2019 · accessed 2026-08-13

Continue through the graph

Concept

Metrology & inspection

Concept

Reliability & failure analysis

Direct answer

  • An electron-beam defect review or inspection system scans selected wafer locations with a focused electron beam, detects emitted signals, and produces high-resolution images and classifications that connect inspection candidates to physical defect evidence.

Mechanism and method

  • Load and align wafer
  • Navigate to candidate coordinates
  • Optimize beam and imaging condition
  • Acquire images or scan area
  • Classify and correlate results

What is measured

  • Landing energy
  • Beam current
  • Resolution
  • Charging control
  • Coordinate accuracy
  • Review throughput
  • Image-resolution check
  • Coordinate correlation
  • Classification repeatability
  • Beam-damage study
  • Optical-to-e-beam capture analysis

Limitations

  • Beam-induced change
  • Charging artifact
  • Coordinate miss
  • Low-contrast image
  • Contamination deposition
  • Sampling bias

Boundaries declared by the cited sources

  • A preprint evaluating research models on imec datasets. It reports no tool throughput, no production yield impact, and nothing about how any commercial inspection system performs. Its abstract carries no numeric accuracy figures, so no quantitative performance claim may rest on it. (boundary declared by Deep Learning-Based Defect Classification and Detection in SEM Images)

Related records