Making Plastic from Carbon Dioxide, Water, and Electricity

What if a machine could suck up carbon dioxide from the atmosphere, run it through a series of chemical reactions, and essentially spit out industrially useful plastic?

"I think that is something that we, as a society, would be interested in. After all, in addition to being a greenhouse gas, carbon dioxide is an abundant and inexpensive feedstock," says Theo Agapie (PhD '07), the John Stauffer Professor of Chemistry and the executive officer for chemistry at Caltech. "With our new work, we have taken a significant step in that direction."

Reporting in the journal Angewandte Chemie International Edition, Agapie and a team of Caltech chemists have developed a system that uses electricity from sustainable sources to carry out the chemical conversion of carbon dioxide (CO2) into molecules, such as ethylene and carbon monoxide, that are useful for making more complex compounds. When this is accomplished using light as the energy source, without plants, such a process is known as artificial photosynthesis. The new system feeds the ethylene and carbon monoxide that has been generated into a second catalytic loop that yields industrially useful plastics called polyketones, which are known for their strength, durability, and thermal stability, making them ideal for applications ranging from adhesives to car parts and from sports equipment to industrial piping.

"We have shown that one can use CO2 to make a material that is useful, without using plants as a mediator," says lead author Max Zhelyabovskiy (MS '24), a graduate student in Agapie's lab who was co-mentored on the project by Jonas C. Peters, Caltech's Bren Professor of Chemistry and director of the Resnick Sustainability Institute.

The Caltech-led team is not the first to build a system that attempts to pair CO2 reduction with a second chemical reaction to ultimately produce polymers. But previous systems have added ethylene that comes from petroleum products, rather than deriving it from carbon dioxide and water.

The conversion of CO2 all the way to plastic has been challenging for a number of reasons. Among those is the fact that previous electrochemical CO2 reduction systems have yielded very little ethylene and carbon monoxide, the reagents needed to feed the second step of the conversion to polyketones. In fact, most have produced less than 5 percent concentrations of these desired compounds, along with other undesired chemicals that can potentially harm downstream processes.

"It has been difficult, at least on the lab scale, to obtain high concentration, high purity streams of reagents that can then be converted into something like a plastic or a fuel," Zhelyabovskiy says. But the system he helped develop achieves significantly higher concentrations—11 percent ethylene and 14 percent carbon monoxide.

But that is not the only challenge. Coupling the two systems—one for the CO2 reduction and another for the catalysis step that follows—is not trivial, says Zhelyabovskiy. "Most work in the literature focuses on either the first or the second step, separately and with pure feedstocks. Not both."

A Two-Step System

Recognizing the vastly different environments needed for the CO2 reduction and the secondary catalytic step to operate with high efficiency, the Caltech team devised a system featuring two distinct loops for the separate reactions.

For the first loop, the system begins with gas diffusion electrode cells, hydrophobic polymers coated in a thin layer of copper. The scientists pump CO2 into a gas cylinder connected to the cells and flow a potassium bicarbonate electrolyte through the cells, all while applying a voltage to the electrodes. By looping the gases through this electrochemical setup multiple times, they are able to generate relatively high concentrations of ethylene and carbon monoxide.

After about an hour of building up those gases, the researchers then feed the ethylene and carbon monoxide into the second step: a closed reactor where the gases are bubbled through a solution of a palladium catalyst. Like a bubbler in a fish tank, this process enriches the solution in ethylene and carbon monoxide. And the catalyst, what is known as a co-polymerization catalyst, drives the efficient formation of a polymer—in this case, a polyketone—from the two monomers.

A Catalyst that Does Its Job Under Working Conditions

Typically, catalysts are tested under pristine conditions that do not necessarily represent the environments they are exposed to during electrochemical CO2 reduction. For example, although water vapor is very harmful to many polymerization catalysts, water is a necessary part of CO2 reduction, and thus the introduction of water vapor is inevitable. In the new work, Agapie, Peters and their colleagues have shown that the palladium catalyst can be used even in the presence of contaminants that are introduced during CO2 reduction—including not only water vapor but also hydrogen, unreacted CO2, alcohol vapors, and other chemical intermediates.

Zhelyabovskiy says that the new system and technique needs additional refinement. It does not yet produce polyketones with the same molecular weights as those made the standard way, for example. However, he says, "by demonstrating that it's possible, we might increase the amount of interest in this field, and maybe people can build upon this principle." Agapie notes that for this process to lead to a sustainable and practical technology, electricity has to come from renewable and carbon-neutral sources, and it has to be sufficiently inexpensive to compete with petroleum sources.

Additional authors of the paper, "Plastic from CO2, Water, and Electricity: Tandem Electrochemical CO2 Reduction and Thermochemical Ethylene-CO Copolymerization," are Hyuk-Joon Jung and Paula L. Diaconescu of UCLA.

The project was based on work performed by the Liquid Sunlight Alliance (LiSA), an energy hub supported by the U.S. Department of Energy that combines the expertise and efforts of scientists at Caltech, Lawrence Berkeley National Laboratory, the Stanford National Accelerator Laboratory, and the National Renewable Energy Laboratory, along with university partners at UC Irvine, UC San Diego, and the University of Oregon. Additional funding came from the National Science Foundation for the UCLA team.