RE-POLY.AI

PROJECT

RE-POLY.AI

Experimental study and AI-based modeling of solvent separation and recovery processes for the sustainable and innovative recycling of polyamide-based textile waste
PROJECT LEAD

Claudia Castelli

CONCEPT

RE-POLY.AI is an industrial research project aimed at scaling up the solvent recovery phase used in an innovative pretreatment process for recycling mixed textile and/or plastic waste—both post-industrial and post-consumer—based on polyamides. Currently, there are no viable end-of-life recovery solutions for such waste.

This process enables the recovery and reuse of different components, valorizing them in either textile (closed-loop) or plastic (open-loop) applications. The development of a digital twin prototype focused on the distillation phase of the green solvent mixture will allow for process optimization through artificial intelligence techniques.

CONTEXT

RE-POLY.AI targets mixed textile and/or plastic waste that is currently considered unrecoverable and is usually sent to landfills or incineration as general waste. The project aims to recover and revalorize polymers for new textile and plastic applications.

Achieving the project’s goals will contribute to a sustainable approach that increases the recycling rate of mixed textile and plastic waste, with positive environmental and climate impact. The digital twin simulation will also bring significant advantages to industrial scale-up, enabling the creation of best practices applicable to the recycling of other types of polymers

Beneficiari

OBJECTIVES AND EXPECTED RESULTS

The RE-POLY.AI project aims to study the effect of impurities accumulated in the solvents used for recycling mixed waste materials, focusing on their impact on distillation performance and recovered polyamide quality.

It also seeks to optimize distillation conditions and validate solvent separation and recovery on a pilot scale. The integration of digital expertise will support the development of a digital twin prototype of the distillation phase, enabling process optimization through AI-driven models.

Key expected outcomes:

  • Pilot-scale validation of a distillation system capable of regenerating the solvent used in the recycling process.
  • Definition of purge solvent volume (containing impurities) required to maintain polyamide quality.
  • Identification of the best technology to recover solvent from the purge stream.
  • Development and fine-tuning of a digital twin prototype for the distillation unit operation.
  • Demonstration of the benefits of AI in managing and optimizing the distillation process.

KEY FIGURES

16

RESEARCHERS INVOLVED

12

PROJECT DURATION

7

NUMBER OF WORK PACKAGES (WPS)

3-4

STARTING TRL

5-6

FINAL TRL