Active Materials & Manufacturing Food & Agriculture

SustaPack Partnership: AI and Modelling for the Next Generation of Sustainable, High-Performance Packaging to Meet Net Zero Targets

In plain English

AI plain-English summary

A factory in Cambridgeshire is using AI to spot invisible defects in the spray-on coatings that keep paper bottles from leaking. Pulpex Ltd makes bottles from wood fibres for products like shampoo and detergent. These bottles have a carbon footprint 30% lower than plastic, but their interior barrier coating—which prevents liquid leakage and oxygen entry—can develop imperfections during manufacturing. Current quality-control methods cannot detect these flaws quickly enough on opaque bottles, leading to waste and unreliable products. This project combines novel dry-spray coating processes, thermal imaging, computer vision, and AI to solve that problem. Researchers will build mechanistic models of how coatings form, then use machine learning to spot defects as they happen and adjust the process in real time. The goal is 100% reliable, energy-efficient packaging production. If successful, the work could help brand owners meet Net Zero targets and comply with UK Plastic Pact regulations and the UN Global Plastics Treaty. For consumers, it means more sustainable packaging choices. For the environment, it means less plastic pollution and lower carbon emissions from manufacturing.

View original technical description
Context Fibre-based packaging and bottles are a critical step in reducing the detrimental legacy associated with other packaging materials, which persist in the natural environment and are challenging to recycle. Pulpex Ltd. has in the past four years developed production processes and technologies to deliver sustainable packaging at scale – made from natural wood fibres – that shows promise for liquids in the fast-moving consumer goods sector. The Pulpex bottle uses sustainable materials, can be recycled in existing paper waste-streams, can naturally degrade if not recycled, and has a carbon footprint 30% less than poly(ethylene terephthalate). An essential feature is its multi-layered barrier coating on its interior that prevents the leakage of liquid contents and inward oxygen permeation. However, for Pulpex to design the next generation of production technology and materials, novel and fundamental research (TRL 1-3) is necessary to find missing analytical techniques for product quality improvements, to resolve performance challenges, and to decrease in-process imperfections. Challenges Challenges faced by Pulpex include a need for improved analytical methods to identify imperfections rapidly in barrier coatings on opaque bottles, to ensure product reliability and to minimise wastage. Another challenge is to deliver and automate real-time next generation quality control to adapt materials and manufacturing processes to maximise yield. This is technology that is not commercially available today. Presently, Pulpex technology uses barrier coating materials that are deposited from water and followed with a heating step to drive evaporation and film formation. They now seek next-generation innovative coatings processes that use less energy and water. At the same time, there are continued demands for a longer shelf-life of packaged goods and light weight barriers, which will both only be possible with higher performing barrier systems delivered at scale. Aims and Objectives The aim of this project is to harness a combination of novel coating processes, mechanistic modelling, computer vision, and AI to address the challenges faced in manufacturing sustainable packaging by Pulpex and to underpin the next generation of the process technology. These are not solutions that are commercially available. The specific objectives are: Develop an alternative “dry” spray coating process to deposit layers that are food-safe and degradable. Investigate the feasibility of using thermal imaging to identify defects in wet coatings as they occur – allowing a quick remedy by drawing on AI. Employ advanced multi-scale mechanistic models of the coatings process to identify ways to optimise it to eliminate all defects and non-uniformities in coatings. Apply new modes of computer vision with AI to identify defects in industrial production and optimise the materials and processing to ensure 100% reliability of the sustainable packaging. Benefits By joining forces with Surrey’s eminent scientists and engineers, Pulpex will see benefits in their technologies and processes. In turn, brand owners of fast-moving consumer goods will have confidence in using their products, which will lead to expanded sales and more jobs at Pulpex’s Cambridgeshire site. Brand owners will be able to meet their Net Zero targets and to comply with Extended Producer Responsibility and UK Plastic Pact regulation, along with any forthcoming legislation arising from the UN Global Plastics Treaty. Consumers will have a greater packaging choice and reduce their personal environmental impact. There will be benefits to the environment from a reduced amount of plastic pollution and lower carbon emissions from packaging manufacturing.

View the original record at the funder ↗

Researchers

Chuan-Yu Wu (Co-Investigator)Hui Luo (Co-Investigator)Joseph Keddie (Principal Investigator)Robert Dorey (Co-Investigator)Simon Hadfield (Co-Investigator)

Related Research

Grants with similar aims, by meaning.

Next-generation sustainable liquid packaging using only bio-based feedstock derived from non-food waste - Pulpex UK BioScience Initiative
Potential for BioBased Waxes for Paper Bottle Barrier Coatings
Develop and scale-up (TRL4-TRL6) the manufacture of a biobased, recyclable and biodegradable coating for paperboard packaging as a replacement for current petrochemical-based or fluorochemical materials
Co-creating a digital platform for the rapid development of sustainable polymer products
GREENBOTTLE - Development of an innovative high speed paper moulding machine

Original classification

Research and Innovation

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.