Transient Thermal Barcode

Background:

Plastic pollution is one of the most profound environmental issues faced, as plastic production has increased from 2.3 million tons in 1950 to an anticipated 1 billion tons by 2050.  Of this, only 8.7% of plastics are recycled due to issues with cross-contamination of plastic types, requiring an engineering solution.  Although using recycled plastics offers many economic and business opportunities, cross-contamination with other plastic types reduces the quality of recycled plastic products and diminishes their value over virgin plastics.  In addition, plastic production contributes approximately 4.5% of global greenhouse gas emissions, so implementing a more efficient recycling platform contributes to a circular economy approach to climate change.

Technology Overview:

This University at Buffalo invention proposes an automated system to rapidly screen large volumes of plastics at the rate required by Material Recycling Facilities (MRF), allowing reliable separation of the 6 primary types of plastic (PET, HDPE, PVC, LDPE, PP and PS) for increased recycling efficiencies.  The cornerstone of the invention is the sorting and identification of plastics using a proprietary Transient Thermal Barcode technique which uses machine learning as the foundation for the automated separation of plastics by chemical composition.  This high sensitivity, high selectivity technique even sorts polymer mixtures of black plastics, the most challenging separations to achieve correctly.

Advantages:

• High throughput with negligible false identifications
• Able to handle the volumes required in a materials recycling facility
• Increased signal to noise ratio as compared to current marketed technologies, offering higher sensitivity
• Machine learning for accurate identification of polymer types
• Able to be automated for identification and sorting
• Non-invasive technique 

Applications:

• Material recycling facilities
• Secondary sorting facilities
• Expansion to additional identification capabilities, such as composites

Intellectual Property Summary:

US Provisional Patent Application 63/588,800 filed on October 9, 2023

Stage of Development:

Laboratory demonstration and analytical chemical analysis.

Licensing Status:

Available for licensing or collaboration.

Relevant Links:

• (154ab) Real-Time Plastic Waste Recognition with Mid-IR Standoff Detection and Advanced Machine Learning | AIChE
• Optical Sensors and machine learning algorithms in sensor-based material flow characterization for mechanical recycling processes: A systematic literature review - Science Direct
• The Chemical Recycling of Polyesters for a Circular Plastics Economy: Challenges and Emerging Opportunities - Paynd>e- 2021 - ChemSusChem - Wiley Online Library