Look, I know that the Perseverance rover has brought some flashy stuff to Mars. It’s got lasers, it’s got a big robot arm, it’s got a little robot arm, and it even launched a helicopter. That’s all great, but tucked up inside of Perseverance is another instrument about the size of a car battery that doesn’t move even a little bit and in fact spends most of its time not functioning at all. It’s the appallingly bacronymed Mars Oxygen ISRU (In-Situ Resource Utilization) Experiment, or MOXIE, and I’m going to try to convince you that it’s the most exciting thing happening on Mars right now.
MOXIE’s job on Mars is to demonstrate that it’s possible to break the carbon dioxide in the Martian atmosphere into carbon monoxide and oxygen through solid oxide electrolysis. The carbon monoxide is returned to the atmosphere, while the oxygen is stored and can be used in a variety of ways. MOXIE has already run successfully a couple of times, producing 5.4 grams of 98% pure oxygen over the course of an hour on April 20. 5.4 grams isn’t much, only enough to provide an astronaut with about 10 minutes of breathable air, but it proved that the system worked.
It’s a little bit strange to have MOXIE on Perseverance at all; MOXIE would be perfectly happy to remain completely stationary and derives no benefit from being hauled all around Jezero crater. The fact that it ended up on a rover (potentially taking the place of a science instrument that could have taken advantage of Perseverance’s mobility) seems to have been the result of shifting priorities at NASA with a history that goes back to the 1970s. After the successful Apollo missions, some folks at NASA (including Wernher von Braun, chief architect of the Saturn V) advocated for the development of a crewed mission to Mars. NASA decided to focus instead on low Earth orbit, starting work on the Space Shuttle, followed by the International Space Station. There simply weren’t all that many resources left over for any other major initiatives.
However, in the late 1990s, there was a small window where NASA felt like the Shuttle and the ISS were stable programs, and the agency was willing to start thinking about Mars exploration again. The window closed as the ISS rapidly got much more expensive, but one Mars mission squeaked through, sort of—Mars Surveyor 2001. What was unique about this mission was it wasn’t really about Mars science in the sense that it wasn’t doing investigative geology or chemistry the way that missions before and since have. Instead, Mars Surveyor 2001 was intended to run experiments on the Martian surface in the context of future crewed missions. Its payloads included experimental solar cells, dust characterization and mitigation, and perhaps most importantly, the OGS, or Oxygen Generator Subsystem, NASA’s attempt at in-situ propellant production (ISPP), now more generally known as ISRU (in-situ resource utilization).
Well before Mars Surveyor 2001, NASA understood that getting astronauts to Mars and back is all about mass. Really, getting anything into space period is all about mass. For example, the SpaceX Falcon Heavy can put 16,800 kg into Mars transfer orbit, but doing so requires over 1.4 million kg of launch vehicle and fuel, meaning that the payload makes up just over 1% of the mass of the rocket. These single-digit payload fractions are the norm for sending things into space, which is why there’s such an intense focus on finding ways of reducing the amount of stuff that you have to haul out of a gravity well like Earth. This is the reason why NASA wanted to put the Oxygen Generator Subsystem on Mars Surveyor 2001: if we could make our own oxygen on the Martian surface, we’d have to send way, way, waaay less stuff to Mars in the first place. It would be easier, it would be cheaper, and it would be safer.
Unfortunately, after both the Mars Climate Orbiter and Mars Polar Lander were lost in 1998 (the latter because sensor errors led to engine shutdown 40m above the Martian surface and the former because of an embarrassing miscommunication involving metric units), NASA cancelled the Mars Surveyor 2001 mission and the lander was later repurposed for the successful Phoenix mission in 2008, with a different and much more sciency payload that didn’t include ISPP.
NASA’s focus throughout the early 2000s continued to be the Shuttle and the ISS, but by the mid-2010s with the Shuttles winding down and the ISS mostly put together, NASA again found itself with another small budgetary window that could be leveraged towards human exploration of Mars. That window would shut again when the agency decided to refocus on the Moon, but it was open long enough for MOXIE to sneak through. “The Mars community recognized that the single most important thing we didn’t get done [after Mars Surveyor] was ISPP,” says Michael Hecht, MOXIE’s principal investigator at the MIT Haystack Observatory. According to Hecht, it took three different NASA directorates, including Science, Space Technology, and Human Exploration to get MOXIE onto the Mars Perseverance rover: “They all got together, which doesn't happen every day, and came up with this plan to say let's just go nail this one, let’s hit it out of the park.” And they did: MOXIE is on Mars.
How MOXIE generates O 2 from CO 2Image: Michael Hecht
MOXIE is, fundamentally, a fuel cell. Here on Earth, we use fuel cells to generate energy by combining a gas like hydrogen with ambient oxygen in the air to produce electricity along with water as a byproduct. If you run that same fuel cell backward (in what’s called regenerative mode), you instead consume electricity to split water into hydrogen and oxygen. This is what MOXIE does on Mars: it pulls in CO2 from the Martian atmosphere (which is 96% CO2), filters it, compresses it, and then uses solid oxide electrolysis to break off some of those Os and store them, while returning the leftover carbon monoxide to the atmosphere. It takes a significant amount of power to do this—by the time MOXIE heats itself up to operating temperature (800 degrees Celsius) and compresses the Martian atmosphere and runs for an hour, MOXIE has sucked down about 1,000 watt-hours of energy, which it takes Perseverance’s radioisotope thermal generator about 10 hours to produce. So when MOXIE runs, it takes up all of the power available for all of the rover’s payloads for that day. In exchange, MOXIE produces about as much oxygen as a smallish tree, between six and 10g of O2 per hour, less than half what it would take to keep a human alive.
But keeping humans alive is not really what MOXIE is about. A mission to Mars with four crew members that spends a year on the surface will only need about a ton of oxygen for breathing. Getting back to Earth, though, will require at least 25 tons of oxygen, along with seven tons of methane or another kind of rocket fuel. The astronauts will probably have to bring the methane or whatever with them to Mars, but if they can produce all of the oxygen they need, that’ll be an enormous amount of mass that they won’t have to worry about. This is why MOXIE is so important: it’ll generate the bulk of the resources that astronauts will need to get back home.
This is why MOXIE is so important: It’ll generate the bulk of the resources that astronauts will need to get back home.
One important question to ask is why we’re even bothering with MOXIE at all—we know there’s water ice at the poles, and we may be able to find it underneath the surface as well. And while water is very appealing because you can split it into both oxygen and hydrogen, it’s also a lot more complicated than what MOXIE is doing. First, you’ve got to find the ice. If it’s underground, that’s a problem, and even if it’s not underground, then you’ve got to somehow harvest it, with robots or something like that, plus it’s almost certainly going to be full of dust. It’s totally doable, but when you compare all of that complexity to just plopping a bunch of MOXIEs on the surface and letting them sit there, MOXIE seems like a much more straightforward solution.
The other problem with ice, Michael Hecht says, is that the places where we know we can get at it are, for lack of a better word, boring. “Most Mars scientists are geology-driven. They don’t want to have anything to do with those icy areas, because once you have ice, you have erosion, while if you stay near the lower latitudes you're looking at four and a half billion years of history.” And if we’re going to go all the way to Mars, we may as well make it worthwhile, right?
“In spaceflight, we have to take everything with us that we need. If we could instead utilize resources we find at the destination, that would make our exploration efforts more efficient. MOXIE is actually the first ever ISRU experiment on another planet.”
—Jeff Sheehy, Chief Engineer, NASA Space Technology Mission Directorate
Producing the 25 tons of oxygen required to get humans off of the Martian surface isn’t something that MOXIE itself is capable of, but the fundamental technology is scalable. Essentially, you’d just send the equivalent of 200 MOXIEs to Mars, in the form of something about the size of a small chest freezer weighing around 1000 kg. It would be able to produce 3 kg of O2 per hour, but you’d send it far in advance of any astronauts, the idea being that MOXIE could chug away on its own for a year or two, slowly but steadily harvesting oxygen from the Martian atmosphere such that by the time NASA was ready to send a crewed mission, the oxygen would be already all taken care of and waiting for them on the surface.
As you might expect, scaling MOXIE up is slightly more complicated than just stapling 200 of them together. For example, MOXIE has to be very careful about how it splits up the CO2, because otherwise the reaction will produce wayward carbon atoms that gum everything up in the form of soot. And each of MOXIE’s components need to be scaled up as well, including the compressor, control system, oxygen storage system, and the filtration system. That last one was of particular concern, but it turns out that thanks to favorable dust particle size and very low atmospheric pressure, filters work much better on Mars that they do on Earth. “When I think of what we have learned,” says Hecht, “that to me is probably one of the most valuable lessons.”
Beyond just making MOXIE bigger, it also has to run reliably, because if it doesn't do what it needs to do, then there won’t be a human mission to Mars. And when you need to spend years on Mars with no maintenance, reliability alone is not enough, explains Hecht. “If you're sending a system to Mars to run for a year and create the return oxygen for a human mission, it’s not going to fail. It can’t fail. So you want some redundancy.” Redundancy means several (probably four or five) independent MOXIE-like systems, each of which will produce about a kilogram of O2 per hour for about 10,000 hours.
The fundamental pieces of these scaled-up MOXIE-like systems are already complete, and there’s a NASA-funded company called OxEon Energy that’s developing commercial versions of the fuel cell technology. It seems like we have the technology to do this, it’s just a question of when, and how much of a difference MOXIE’s success on Mars will mean for the potential for human exploration—is MOXIE a big step forward, or will Mars remain 15 years away, just like it was 15 years ago? Michael Hecht is optimistic:
I think this is different because of the scale of the investment. When NASA puts this much money and resources on the line and takes up a valuable space on a high profile flagship mission to do this, I think they’re serious. And while priorities change and governments change, the fact that three different countries went to Mars this time around tells you that it’s no longer NASA going on its own. And there’s Elon Musk, of course, with SpaceX. This is becoming a broad-based enterprise, and someone’s going to do it. So there’s a real reason for optimism that we are taking the first step in a way that we haven’t in the past.
With that in mind, we asked Jeff Sheehy, Chief Engineer at NASA’s Space Technology Mission Directorate, to describe what a MOXIE-based oxygen generation system might look like on Mars:
What people envision generally is that you’d land several tons of MOXIE-based oxygen production capability—a little oxygen production plant with an integrated oxygen storage facility. The oxygen production capability would be set up a few years before the astronauts ever got there. You’d land a fission power plant, and some sort of robotic rover that would attach cables from the power system to the oxygen plant. And ultimately you’d have a Mars ascent vehicle, and some sort of plumbing that goes from the oxygen storage system to the vehicle’s oxidizer tank.
The plan is to test some of these components on the Moon. Not MOXIE, obviously, but a fission power plant and the robots that you’d need to get everything set up and plugged in. “We’re establishing a sustained presence on the Moon by setting up all that capability and designing it in a way that it can be used on Mars,” says Sheehy. “So that the Moon really does become a stepping stone where we will have demonstrated the technologies and capabilities we need on Mars to provide for the first human expeditions.”
As for MOXIE, the plan is to run it up to ten times over the course of the Perseverance mission, characterizing how the system responds to different inputs. All it has left to prove now is that it can survive over the long term, giving us confidence that we can send this technology to Mars and rely on it to get us home.
Evan Ackerman is a senior editor at IEEE Spectrum. Since 2007, he has written over 6,000 articles on robotics and technology. He has a degree in Martian geology and is excellent at playing bagpipes.