Waste Material Classification Based on a Wavelength-Sensitive Ge-on-Si Photodetector

Area of Science:

• Materials Science
• Optical Engineering
• Waste Management Technology

Background:

• Efficient waste material classification is crucial for effective recycling and resource management.
• Current methods may be costly or lack the ability to differentiate a wide range of materials.
• Spectral analysis offers a promising avenue for non-destructive material identification.

Purpose of the Study:

• To develop and demonstrate a novel, low-cost material classification system.
• To leverage a voltage-tunable photodetector for spectral information extraction.
• To achieve accurate classification of diverse waste materials using photocurrent signal processing.

Main Methods:

• Utilized a single, voltage-tunable Germanium-on-Silicon (Ge-on-Si) photodetector.
• Operated the system across visible and short-wave infrared (SWIR) spectral regions.
• Employed a broadband illuminator and analyzed photocurrent signals at varying bias voltages with classification algorithms.

Main Results:

• Successfully demonstrated the system's functionality for material classification.
• Achieved effective distinction between seven different material types (plastics, aluminum, glass, paper, etc.).
• Showcased near real-time classification capabilities.

Conclusions:

• The developed Ge-on-Si photodetector system offers a viable, low-cost solution for waste material identification.
• Tunable spectral sensing combined with signal processing enables robust classification.
• This technology has significant potential to enhance recycling and waste management efficiency.