Fusion proponents have long promised the idea of using the same reaction as the sun's energy source to produce nearly limitless energy on Earth. Today, fusion's biggest hurdle is ensuring that a fusion power plant will produce more electricity than it needs to run. The second is ensuring that there is enough fuel to run it.
Many fusion reactors are designed to run on a mixture of two isotopes of hydrogen: deuterium and tritium. (A typical hydrogen atom has no neutrons; a deuterium atom has one neutron, and a tritium atom has two.) While deuterium is abundant in ocean water, tritium is so rare that it essentially has to be manufactured.
“Right now there's only 20 kilograms of tritium in the world,” Kyle Schiller, CEO of Marathon Fusion, told TechCrunch. It takes a few kilograms to run a single commercial-scale power plant, meaning there's only enough tritium in the world for about 12 of them. His quietly operating startup thinks it has a solution to the problem.
Currently, the world's tritium supply is a waste byproduct of the few nuclear power plants that use nuclear fission, which has been used as an energy source since the mid-20th century. Assuming scientists can use nuclear fusion to produce practical electricity on Earth, the first fusion power plants will use this source. Future nuclear reactors will rely on the first fusion power plants being designed to produce additional fuel.
“Deploying a fusion device is this doubling process,” says Marathon CTO Adam Rutkowski. “You breed enough tritium to sustain the steady-state consumption of the device, but you also need to breed excess tritium to power the next reactor.”
This breeding occurs when neutrons released during the fusion reaction collide with the lithium layer, releasing helium and tritium, which are then pumped out of the core, where they are filtered out. Some of the tritium is reinjected into the reactor, while the rest is stored as fuel for other reactors.
There is existing equipment for this task, but it's only good for experimental work. Though efficient and effective, experimental reactors are run for short periods of time and don't have the processing power needed for a commercial power plant. To reach that level, the filtration system would need “orders of magnitude improvement,” Schiller said.
That's where Marathon wants to get in. The company is working to improve on a 40-year-old technology called “super permeation,” which uses solid metals to filter impurities from hydrogen.
Here's how it works: First, hydrogen and other substances that need to be filtered are turned into a plasma, but at temperatures that aren't as hot as those inside a nuclear reactor. Using the pressure of the reactor's exhaust gases, they are forced against a metal membrane, which allows hydrogen (including tritium) to pass through and blocks everything else. As a nifty side effect, the membrane also compresses the hydrogen on the other side.
“The whole idea here is to get maximum throughput as quickly as possible,” Rutkowski said.
Rutkowski and Schiller have been working on this problem for a few years, with early support from the Department of Energy's ARPA-E and Breakthrough Energy Fellows programs. Marathon recently raised $5.9 million in a seed round, the company told TechCrunch exclusively. The round was led by 1517 Fund and Anglo American, with participation from Übermorgen Ventures, Shared Future Fund and Malcolm Handley.
Marathon said it had received letters of intent from two fusion startups, Commonwealth Fusion Systems Inc. and Helion Energy Inc., which have raised $2 billion and $607 million, respectively.
Marathon's bet may seem premature, given that commercial fusion power is still years away (if at all): after all, only one fusion experiment has ever broken even in a scientific sense, after facility overhead, something a commercial power plant can't do.
Schiller disagrees that his company is too ahead of its time: “Over the last decade or so, we've been continually amazed at how quickly we've advanced. [with fusion] “It's over,” he says. “You'll wake up one morning and break even and really wish you'd started sooner.”