Acousto-optic Depth Metrology for Semiconductor Specimens

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current non-destructive metrology techniques for semiconductor specimens lack the accuracy and speed required for characterizing increasingly smaller structures, particularly in depth measurement.

Innovation Solution

The method employs acousto-optic depth-metrology using pulsed pump and probe beams, where the pump beam induces acoustic pulses within the specimen, and the probe beam undergoes Brillouin scattering to probe depths, with an optimization algorithm processing the signals to determine structural parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If scanning electron microscopy, optical critical dimension scatterometry, or small angle X-ray scattering are used for metrology, then non-destructive measurement is achieved, but measurement precision and speed are insufficient for increasingly smaller structures

Engineering Contradiction:
Improvedepth measurement precisionVSAvoidmeasurement speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces traditional mechanical/scanning-based metrology systems with an acousto-optic system that uses light-acoustic interactions. The pump-probe beam configuration with Brillouin scattering eliminates the need for mechanical scanning while achieving high depth resolution through acoustic pulse propagation timing, thereby improving both measurement precision and speed simultaneously

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the measurement parameter from direct optical reflection to acoustic pulse propagation characteristics. By measuring the time-resolved Brillouin scattering signal from acoustic pulses generated at different depths, the system achieves high depth precision without sacrificing measurement speed, as the acoustic pulses naturally provide depth encoding through their propagation time

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If traditional non-destructive techniques are used, then measurement speed is maintained, but accuracy for depth measurement of small structures is insufficient

Engineering Contradiction:
Improvedepth measurement accuracyVSAvoidmetrology technique complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces acoustic pulses as an intermediary between the light source and the depth information. The pump beam generates acoustic pulses that propagate through the specimen, and the probe beam detects these acoustic pulses via Brillouin scattering. This intermediary mechanism translates depth information into detectable optical signals with high precision, while the overall system remains relatively simple compared to other high-precision metrology techniques

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS20250116597A1Model-based acousto-optic depth-metrology of specimens
Publication Date: 2025.04.10 APPL MATERIALS ISRAEL LTD
  • US20250116597A1 patent drawing
  • US20250116597A1 patent drawing
  • US20250116597A1 patent drawing

AI summary

Disclosed herein is a method for non-destructive depth-profiling including projecting a pulsed pump beam into a specimen, projecting a pulsed probe beam thereinto, and sensing light returned therefrom to obtain a measured signal. Each probe pulse is configured to undergo Brillouin scattering off a primary acoustic pulse induced by the directly preceding pump pulse, so as to be scattered there off at a respective depth within the specimen. The method further includes executing an optimization algorithm configured to receive as inputs the measured signal, and/or a processed signal obtained therefrom, and output values of structural parameter(s) characterizing the specimen through minimization of a cost function indicative of a difference between the measured signal and a simulated signal obtained using a forward model simulating the scattering of a pulsed probe beam off at least the primary acoustic pulses.