Quantifying Polymer Composition in Recyclates Using Machine Learning and DSC

Recyclates often contain complex and variable polymer mixtures - especially dark-colored, multilayer, or contaminated plastic waste. For manufacturers using recycled polymers in automotive, technical, and other performance-driven applications, knowing the exact material composition is essential for consistent processing, dependable quality, and compliance with specifications.

In this webinar, discover a data-driven approach that combines Differential Scanning Calorimetry (DSC) with Proteus^® Now Quantify, NETZSCH’s machine-learning-based analysis solution. Proteus^® Now Quantify analyzes characteristic melting and CrystallizationCrystallization is the physical process of hardening during the formation and growth of crystals. During this process, heat of crystallization is released.crystallization behavior in DSC data and compares these thermal fingerprints with a structured reference dataset. This enables the identification and quantitative determination of individual polymer components in unknown recyclates and polymer blends - going beyond a simple indication of which polymers may be present.

You will learn how composition-resolved recyclate data can help you:

• assess and improve sorting efficiency,
• optimize processing parameters for variable feedstocks,
• qualify recycled materials for demanding applications, and
• create more reliable technical data sheets for polymer blends.

The session also highlights the importance of standardized test conditions and includes an industrial case study demonstrating the practical value of ML-assisted DSC analysis.

Who should attend?
This webinar is designed for professionals in recycling and compounding, polymer and materials scientists, quality-control teams, product developers, and anyone responsible for qualifying recyclates or integrating recycled content into technical products.

Register now to learn how Proteus^® Now Quantify and DSC can transform variable recyclate streams into actionable, composition-resolved material data.