Plastic-eating enzymes are proteins that cut the chemical bonds holding PET together, splitting it into terephthalic acid and ethylene glycol. The best known are PETase, found in a bacterium at a Japanese bottle recycling site in 2016; LCC-ICCG, engineered by Carbios and TBI Toulouse, which broke down 90 percent of PET in 10 hours; and FAST-PETase, engineered at the University of Texas at Austin with help from machine learning. The hard parts now are crystalline plastic, enzyme cost and scale.
How an enzyme breaks down PET
PET is a long chain of two building blocks joined by ester bonds. Certain enzymes, called PET hydrolases, use water to cut those bonds. In the bacterium where PETase was found, the work is split between two enzymes: PETase cuts the plastic mostly into an intermediate called MHET, and a second enzyme, MHETase, splits MHET into terephthalic acid and ethylene glycol. Those two molecules are the same ones PET is made from in the first place.
A short history
| Year | Milestone |
|---|---|
| 2016 | Researchers screen 250 samples from a PET bottle recycling site in Sakai, Japan, and find Ideonella sakaiensis 201-F6. It almost completely breaks down a thin, low-crystallinity PET film in six weeks at 30 °C using PETase and MHETase (Yoshida et al., Science). |
| 2018 | A team from the University of Portsmouth and NREL solves PETase’s structure at 0.92 Å and finds that narrowing its active site makes it work better, a sign nature had not finished optimizing it (Austin et al., PNAS). |
| 2020 | Carbios and TBI Toulouse engineer a compost cutinase into LCC-ICCG. It depolymerizes at least 90 percent of PET into monomers in 10 hours at 72 °C, and the recovered monomers are made into new bottles (Tournier et al., Nature). |
| 2022 | Hal Alper’s group at the University of Texas at Austin uses a machine-learning model to design FAST-PETase. It almost completely degrades untreated post-consumer PET from 51 different products within a week, working between 30 and 50 °C (Lu et al., Nature). |
| 2023 | A Carbios-led comparison under industrial-style loading reports LCC-ICCG reaching 98 percent conversion in 24 hours at 68 °C with a third of the enzyme, while other engineered PETases struggle at that scale (Arnal et al., ACS Catalysis). |
Why crystalline plastic is the hard part
PET is partly crystalline: some of its chains are packed in tight, ordered regions that enzymes struggle to reach. The more crystalline the PET, the slower enzymes break it down. Packaging film can be about 10 percent crystalline; bottles and textiles often run 30 to 40 percent, though individual bottles vary.
Temperature helps, up to a point. In water, PET’s glass transition is around 65 to 70 °C. Below it the plastic is glassy and hard to attack; near it the chains loosen. But PET also starts to recrystallize around 70 °C, so the best working temperature sits just below that line. That is why Carbios lowered its process from 72 to 68 °C.
Today’s leading processes still pretreat the plastic. Carbios melts PET at 265 °C, cools it fast to lock in the amorphous form, and grinds it finer than 500 micrometers. That works, but it adds cost, and researchers estimate a process needs at least 90 percent conversion, ideally 95, to pay for itself.
Why reuse the enzyme
An enzyme dissolved in the reaction is used once and washed out with the product. Reusing it spreads its cost over far more plastic, so researchers are attaching PET hydrolases to solid supports that can be recovered and reused:
- FAST-PETase and MHETase fixed together on magnetic nanoparticles released 2.5 times more product than the free enzymes and kept over 80 percent of their activity for four cycles (Kotnis et al., 2025).
- PETase bound to silica-coated magnetic nanoparticles kept over 50 percent of its activity after seven reuse cycles and broke down post-consumer bottles (Zhai et al., 2025).
- Cross-linked PETase aggregates were reused for five cycles and lasted longer at 35 to 50 °C than the free enzyme (Lee et al., 2024).
Results vary widely with the support. One study of silica-bound FAST-PETase kept only about 15 percent activity after five cycles. All of this work is still at lab scale.
Reuse is where PlastiBioFuel works. We hold engineered PET-degrading enzymes inside a proprietary porous framework so the same batch can run again and again, inside a continuous spiral-flow reactor. The enzymes themselves are published science and we don’t claim them. Our target is 50 reuse cycles from one batch, which is a goal we are testing toward, not a result yet. See our lab results on real bottle PET.
Who is commercializing it
- Carbios (France) is building a plant at Longlaville designed for 50,000 tonnes of PET waste a year. Construction was postponed in December 2024 pending financing. It had targeted production in the first half of 2028, and in August 2026 it said financing would not close by its September 30 target.
- Samsara Eco (Australia) opened its first enzymatic recycling plant in September 2025, focused on nylon 6,6 and polyester.
- Protein Evolution (Connecticut) received $3 million from the U.S. Department of Energy in 2026 for recycling PET textile waste.
- Birch Biosciences (Oregon) licensed NREL’s enzymatic PET recycling patents in 2025.
Nearly all of these aim to turn old plastic into new plastic. We are pointing the same chemistry at fuel. Read how the plastic-to-fuel routes compare.
What comes next
The next gains will come from enzymes that handle crystalline PET without heavy pretreatment, from supports that keep enzymes active across many cycles, and from reactors designed to keep solid plastic, catalyst and liquid in contact. Each one cuts the cost per ton, and cost per ton is what decides whether enzymatic PET processing leaves the lab.
Common questions
Do plastic-eating enzymes come from bacteria?
The first PET-eating enzyme system, PETase and MHETase, was found in the bacterium Ideonella sakaiensis at a Japanese bottle recycling site in 2016. Most enzymes used today are engineered versions of natural enzymes, redesigned in the lab for speed and heat tolerance.
How fast can enzymes break down a plastic bottle?
In a 2020 Nature study, the engineered enzyme LCC-ICCG broke down at least 90 percent of pretreated PET in 10 hours at 72 degrees Celsius. FAST-PETase almost completely degraded untreated post-consumer PET from 51 products within a week. Speed depends heavily on crystallinity and pretreatment.
Are plastic-eating enzymes used commercially yet?
Not yet at full commercial scale for PET. Carbios’s first plant in France is still closing financing as of September 2026, and other companies are at demonstration or early commercial stages.
Sources
- Yoshida et al., A bacterium that degrades and assimilates poly(ethylene terephthalate), Science (2016), doi:10.1126/science.aad6359
- Austin et al., Characterization and engineering of a plastic-degrading aromatic polyesterase, PNAS (2018), doi:10.1073/pnas.1718804115
- Tournier et al., An engineered PET depolymerase to break down and recycle plastic bottles, Nature (2020), doi:10.1038/s41586-020-2149-4
- Lu et al., Machine learning-aided engineering of hydrolases for PET depolymerization, Nature (2022), doi:10.1038/s41586-022-04599-z
- UT News, Plastic-eating enzyme could eliminate billions of tons of landfill waste (2022)
- Arnal et al., Assessment of four engineered PET degrading enzymes considering large-scale industrial applications, ACS Catalysis (2023), doi:10.1021/acscatal.3c02922
- Thomsen, Almdal and Meyer, Significance of PET substrate crystallinity on enzymatic degradation, New Biotechnology (2023), doi:10.1016/j.nbt.2023.11.001
- Kawai, Iizuka and Kawabata, Engineered PET hydrolases: perspectives and limits, Appl Microbiol Biotechnol (2024), doi:10.1007/s00253-024-13222-2
- Kotnis, Gulati and Sun, Biotechnology and Bioengineering (2025), doi:10.1002/bit.28963
- Zhai et al., ACS Applied Materials and Interfaces (2025), doi:10.1021/acsami.5c15692
- Lee et al., International Journal of Biological Macromolecules (2024), doi:10.1016/j.ijbiomac.2024.130284
- Carbios, update on the financing of its Longlaville plant project (Aug 3, 2026)
- Carbios, postponement of Longlaville construction (Dec 19, 2024)
- Packaging News, Samsara Eco opens first enzymatic recycling plant (2025)
- Ecotextile News, US backs enzymatic PET textile recycling project (2026)
- Business Wire, Birch Biosciences license agreement with NREL (2025)