How SIU’s µBites Turn PET Plastic and Corn Waste Into Cookies

A plastic bottle does not go directly into a cookie. In SIU Carbondale’s µBites system, PET plastic and discarded agricultural biomass are first chemically processed into carbon compounds. Engineered yeasts then use those compounds to make proteins, fats, acids, vitamins and flavor molecules that can be blended and 3D-printed into a protein-rich food prototype. 81011

The idea was developed for environments where conventional food supply chains are difficult to maintain—especially future deep-space missions. Its potential is significant, but the available evidence describes an early-stage platform rather than a finished food product.

How the µBites process works

1. PET plastic and plant waste are chemically processed

The researchers use polyethylene terephthalate, or PET—the plastic commonly used in many beverage bottles—alongside discarded corn stalks, leaves and other biomass. The materials are ground and treated using oxidative hydrothermal dissolution, a process that combines water, oxygen, heat and pressure to break them down into water-soluble carbon molecules. 101113

This step matters because intact PET and fibrous plant waste are not readily usable as food-production inputs for ordinary microbes. The treatment creates a carbon-rich slurry that engineered microorganisms can metabolize. 1011

2. Engineered yeasts turn carbon into ingredients

Different yeast strains are programmed to convert the processed compounds into specific products. The reported outputs include proteins, fats, organic acids, vitamins and flavoring molecules. Baker’s yeast is among the organisms being engineered, including for vanilla flavor production; other strains are intended to produce compounds such as beta-carotene. 571013

The core concept is microbial “rebuilding”: instead of trying to eat plastic or plant waste directly, the system uses their carbon as feedstock for fermentation. The resulting microbial products can then be combined into a more conventional food formulation.

3. The formulation is 3D-printed

The microbial product slurry is mixed with fiber, starch and sweetener before being deposited by a 3D food printer. The prototype cookies are shaped like the Greek letter µ, giving the product its name: µBites, pronounced “micro-bites.” 101114

The wider platform is designed to be adjustable. Supplements, spices and other components can be added, while the final product could potentially range from a semisolid food to a liquid rather than being limited to a cookie. 1011

What is known about safety, aroma and taste?

The evidence is promising but preliminary—and it is important not to confuse a tested prototype with a food approved for general sale.

SIU reported evaluating an earlier cookie prototype for safety and nutritional parameters before conducting human sensory analysis. That prototype received an overall acceptability score of 6.5 out of 9 on a hedonic scale. Aroma was its highest-rated attribute, at 7.33 out of 9, while color, shape and texture each scored above 5. 11

Those results indicate that at least one prototype was acceptable to the panel involved in the test. They do not establish that the current multi-ingredient version is safe for unrestricted consumption. Public reporting does not provide all the details needed to independently assess the work, such as the full panel size, contaminant results, toxicology data or independent peer-reviewed safety validation. Other reporting says formal taste testing remained pending and that the cookies had not received clearance for human consumption. 11

The most defensible conclusion is that the project has produced a promising prototype with encouraging aroma and sensory results, not a commercially approved snack. Large, controlled and independently documented consumer testing would be needed before making stronger claims about taste or acceptance.

How NASA and the NSF supported the work

NASA provided the application-focused context through its Deep Space Food Challenge. In 2021, NASA selected SIU’s µBites design as one of 18 promising U.S. concepts and awarded the team $25,000 to continue its Phase 2 work. The proposed system was aimed at producing food for four people during a three-year Mars mission while using limited water, energy and resupply. 110

The challenge framed the central problem: astronauts on long missions would need food systems that are nutritious, acceptable and efficient with resources and waste. µBites addresses that goal by treating discarded carbon as a potential input for microbial food production rather than relying only on stored ingredients. 110

The NSF CAREER grant supported related scientific research into hydrothermal biomass liquefaction. The award, titled “CAREER: Manipulating Polarity to Enhance Hydrothermal Liquefaction of Biomass for Biofuels,” concerns the underlying hydrothermal-processing science; it should not be interpreted as proof that the µBites food system is market-ready. 8

Where the technology could be useful

Deep-space missions

For astronauts, a compact system that converts selected waste streams and stored packaging into tailored food ingredients could potentially reduce dependence on resupply. It could also support a more closed-loop approach to food production during missions where water, energy, storage and crew time are constrained. 110

Extreme and resource-limited environments

On Earth, the same type of modular food-production system could be relevant in places where conventional farming, refrigeration or regular supply deliveries are difficult. Reported possibilities include disaster zones, submarines, polar stations and other resource-constrained settings. 1113

These are potential use cases, not demonstrated deployments. The technology would need to show that it can operate reliably, safely and economically in each environment before those benefits could be claimed in practice.

The biggest obstacles are cost and scale

The reported production cost is approximately $60 per kilogram, making µBites far more expensive than ordinary staple foods at its current prototype stage. That figure is an estimate rather than a demonstrated mass-production price, and the available reporting does not establish whether it includes every cost associated with energy, collection and sorting, equipment, labor, quality control, regulatory compliance and waste handling.

Scaling the process would require several difficult systems to work together:

• high-throughput oxidative hydrothermal processing;
• reliable fermentation with engineered microbial strains;
• water recovery and nutrient balancing;
• consistent control of contaminants and unwanted byproducts;
• automated formulation and 3D-printing equipment; and
• independent safety testing and a workable regulatory pathway. 11

Consumer acceptance is another challenge. Even if the chemistry and fermentation are validated, people may be reluctant to eat food whose starting materials included plastic waste. The researchers have also identified the need for a fully integrated, automated system, additional development and policy work. 11

Is µBites coming to stores soon?

Not based on the evidence currently available. SIU has described the possibility of µBites reaching dinner tables within the next few years, but that is a development goal rather than a confirmed launch date. 11

Before public availability, the project would need stronger documentation of contaminant control, nutritional consistency, toxicological safety, regulatory approval, production economics and consumer acceptance. Until those steps are completed, µBites is best understood as a NASA-linked research prototype that demonstrates a striking possibility: waste carbon can be processed and biologically converted into ingredients for food.

That makes the project scientifically notable—but it does not yet make a plastic-derived cookie an everyday food.