Starship runs on liquid methane and liquid oxygen, which SpaceX describes on its Mars page as “natural resources that can be mined and refined on Mars.” How much hardware that refining would take is the question behind a Mars propellant plant study from engineers working at NASA’s Johnson Space Center in Houston.<\/p>\n\n\n\n
“The SIMA team conducted analyses on a full-scale Martian ISRU production system for oxygen and methane cryopropellant, with the goal of recycling as much material as possible,” the authors write in their paper for the AIAA ASCEND conference<\/a>, held in July 2025. SIMA is the Systems Engineering and Integration ISRU Modeling and Analysis project, and ISRU stands for in-situ resource utilization, or living off local materials.<\/p>\n\n\n\n The work was led by Avery L. Carlson, a chemical engineer in modeling and simulation at Amentum, which supports Johnson Space Center under NASA’s JETS II contract, with Kyle Ostendorp of Amentum, Noah Andersen of HX5 and Jacob Collins of NASA Johnson.<\/p>\n\n\n\n “Returning crew from Mars to the Earth will require hundreds of tons of rocket fuel,” the team’s slides say, and “transporting fuel to the Martian surface is expensive.”<\/p>\n\n\n\n “A major challenge identified in DRA 5.0 was focused on the difficulty of resupplying exploratory missions,” the paper adds, referring to NASA’s Mars Design Reference Architecture 5.0, because “the 26-month synodic window between Earth and Mars will not easily permit direct delivery methods, such as those employed for low Earth orbit (LEO), for longer missions.”<\/p>\n\n\n\n “The origins of a Martian ISRU architecture can be traced back to the late 1970’s,” the authors note, when the idea was “to react Carbon Dioxide (CO2) from the Martian atmosphere, and Hydrogen (H2) via electrolyzed H2O from the soil, to produce a yield of CH4 and O2.” Their version splits steam in a solid oxide electrolyzer, a ceramic cell that runs on electricity, and sends the hydrogen to reactors “in which CO2 and H2 are subjected to high-temperature equilibrium conditions to preferentially form CO and H2O,” before a last step makes methane.<\/p>\n\n\n\n The study looks at 33 tons (30 metric tons) and 331 tons (300 metric tons) of liquid oxygen and methane “produced during the course of a Martian year (~687 Earth days), at a ratio of approximately 3.5:1, respectively.” That works out to 25.7 tons (23.3 metric tons) of oxygen and 7.4 tons (6.7 metric tons) of methane in the small case, and ten times as much in the large one.<\/p>\n\n\n\n “This is the correct blend of propellants to service a Martian Ascent Vehicle (MAV),” the paper says, the MAV being the rocket that would lift astronauts off the surface.<\/p>\n\n\n\n For the 30-metric-ton case, “the system model estimated over 50% mass savings compared to baseline resupply-only strategy,” and at 300 metric tons the savings climbed to nearly 75 percent, “strongly implying that the subsystem operations become economical at higher production rates.”<\/p>\n\n\n\nWhy the fuel has to be made on Mars<\/h2>\n\n\n\n
Two plant sizes over one Martian year<\/h2>\n\n\n\n
MOXIE already made oxygen on Mars<\/h2>\n\n\n\n