BEYOND THE BIN: Enzymes That Eat Plastics

Scientific accuracy review · 2 October 2026
Updated commercial status, distinguished modeled economics from operating results, and removed claims that compostability solves litter risk.

Editorial illustration; not experimental footage or evidence.

Enzymes breaking plastic into useful chemical building blocks? Now that gets my attention. Tiny molecular specialists, enormous industrial hopes, and a whole parade of headlines already halfway to the after-party. Sit down, headlines. A successful reactor experiment does not mean every mixed-plastic disaster in the bin has found salvation.

PET gets the molecular scissors

PET hydrolysis can recover terephthalic acid and ethylene glycol for purification and repolymerisation. Tournier and colleagues demonstrated an engineered enzyme route in 2020; subsequent work examined high-solids processing. Reported conversions depend on pretreatment, crystallinity, particle size, enzyme loading, temperature and reaction time. They are not a universal rate for bottles dropped into water.

Coloured PET and polyester textiles are potential feedstocks, but dyes, other fibres, barrier layers and additives still require separation or management. Mechanical recycling can also produce high-quality PET from suitable clean streams; it is not inevitably downcycling.

A financing announcement is not a working factory

Carbios reported on 3 August 2026 that its Longlaville financing deadline would be missed, delaying the anticipated production timetable. Its 24 September 2026 results still described finalising financing and resuming construction as priorities. This is a company-reported industrial project, not evidence of an operating full-scale plant.

Modeled economics deserve their own dressing room. A promising cost estimate is useful; putting it onstage as a demonstrated profit is how the production loses credibility.

A 2025 national-laboratory analysis estimated recycled PET production costs of $1.51/kg versus a $1.87/kg domestic virgin-PET comparison. Those are modeled values under study assumptions, not market quotes or demonstrated profits. Feedstock preparation, separation and utility requirements matter as much as enzyme speed.

PLA: different polymer, different plot

Hydrolysis of PLA can yield lactic acid, but composting, enzyme-assisted degradation and monomer recovery are different outcomes. Demonstrating one does not establish the others. This review does not establish widespread commercial enzymatic PLA recycling.

The decision test is practical: identify the feedstock, usable output yield, total energy and chemical demand, and a real collection route. Lower reaction temperature alone does not prove a lower life-cycle footprint.

Show me the feedstock arriving, usable material leaving, and the energy bill in between. I will cheer for a clever enzyme all night. I will not pretend the rest of the factory disappeared because the molecule had a good audition.

The receipts: sources and evidence

• Tournier et al. (2020): engineered PET depolymerase
• 2024 PET hydrolase study: high-solids loading and conversion
• 2025 laboratory report: modeled enzymatic PET recycling economics
• Carbios: financing update, 3 August 2026 (company statement)
• Carbios: half-year results, 24 September 2026 (company statement, French)

This is an editorial evidence review, not an independent laboratory test or peer review. Research findings apply to the material, conditions and measurements studied. Company statements are identified as such.

Start here: a guide to emerging plastics technology