Multi-Mode TDI Sensor Layout for Low-SNR Defect Inspection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Inspection systems face challenges in detecting small defects in semiconductor fabrication due to shrinking feature sizes, which result in weaker measurable signals and lower signal-to-noise ratios, limiting the effectiveness of existing TDI sensors.

Innovation Solution

A multi-mode TDI sensor with alternating active and inactive pixel rows that accumulate and buffer charge respectively, allowing for simultaneous image generation in multiple optical modes, enhancing signal-to-noise ratio and reducing registration errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If feature sizes are shrunk to increase device density, then device integration is improved, but signal strength deteriorates and defect detection precision worsens

Engineering Contradiction:
Improvedevice integrationVSAvoiddefect detection precision
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The sensor is divided into multiple pixel rows that can operate in different modes simultaneously. Some pixel rows are configured for UV detection while others are configured for visible light detection, allowing multi-modal inspection without requiring multiple separate sensors. This segmentation enables the system to maintain high detection precision across different wavelength ranges while supporting high device integration.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The TDI sensor is designed with multi-functional capability by incorporating pixel rows that can detect different wavelengths (UV and visible light) within the same sensor array. This universal design allows a single sensor to perform multiple inspection functions, improving defect detection precision across various feature sizes while maintaining device integration efficiency.

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

2Device complexity

If single optical mode is used to simplify sensor design, then device complexity is reduced, but measurement precision deteriorates due to lower signal-to-noise ratio

Engineering Contradiction:
Improvesensor design complexityVSAvoidsignal-to-noise ratio
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

Multiple optical detection modes (UV and visible light) are merged into a single TDI sensor array. The sensor combines pixel rows with different spectral responses in one integrated structure, achieving high signal-to-noise ratio through multi-modal detection while avoiding the complexity of using separate sensors for each wavelength range.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The sensor achieves multi-functionality by incorporating both UV-sensitive and visible light-sensitive pixel rows within the same device. This universal design enables the sensor to operate in multiple optical modes simultaneously, improving measurement precision without significantly increasing device complexity.

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

3Measurement precision

If multiple separate sensors are used to achieve multi-optical mode detection, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improvemulti-mode detection precisionVSAvoidsensor array complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges multiple optical detection functions into a single TDI sensor array by configuring different pixel rows for different wavelength ranges. This consolidation achieves multi-mode detection precision equivalent to using separate sensors while reducing device complexity by eliminating the need for multiple independent sensor assemblies.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The sensor utilizes the spatial dimension of the pixel array to differentiate optical modes. By assigning specific pixel rows to detect specific wavelengths (UV vs. visible light), the system encodes multiple detection functions along the spatial dimension of the sensor array, achieving multi-mode precision without the complexity of multiple separate sensors.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

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

The multi-mode TDI sensor improves defect detection by increasing signal-to-noise ratio and enabling simultaneous image generation across multiple optical modes without registration errors, enhancing defect inspection capabilities.

Implementation Method 1

Each of the pixels may include a photodiode... the photodiode may be configured to accumulate electrical charge in response to incident light

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Data Source

PatentUS12563316B2Multi-modality with TDI sensor
Publication Date: 2026.02.24 ORBOTECH LTD
  • US12563316B2 patent drawing
  • US12563316B2 patent drawing
  • US12563316B2 patent drawing

AI summary

A TDI sensor may be a multi-mode TDI sensor. The multi-mode TDI sensor may include active pixel rows, inactive pixel rows, and a readout circuit. The active pixel rows and the inactive pixel rows may alternate in a sequence defining the number of modes for which the multi-mode TDI sensor is configured. The length to which the sample light is scanned before switching between the optical modes is based on a pixel length divided by the number of optical modes to synchronize the active and inactive rows with the optical modes. The active pixel rows may accumulate charge in response to receiving light. The charge may be transferred between the active pixel rows and the inactive pixel rows. The inactive pixel rows may buffer the charge before transferring the charge back to the active pixel rows for further accumulation. The readout circuit may readout the charge.