Fly me to the moon [and back]

C&I Issue 11, 2013

‘If you want to go to the Moon and bring the astronauts back you must carry enough oxygen to power the rocket on the way back,’ explains materials scientist Derek Fray of the University of Cambridge, UK. ‘It costs a million dollars per tonne to ship material to the Moon, so it would be much cheaper to generate the oxygen on the Moon for breathing and for propulsion,’ he adds. Rocks on the Moon are mainly oxides, so it should be possible to build a reactor on site that provides a structural metal and oxygen gas.

Planetary geochemist Larry Taylor, of the University of Tennessee, Knoxville, US, gave advice to astronauts as they collected samples on the Moon during what turned out to be the last Apollo mission in 1972. He spent years studying lunar rocks and in 1993 reviewed the various options for their potential to produce oxygen.

At that time, two oxygen-generation schemes were in the limelight: the chemical reduction of ilmenite (FeTiO3) with hydrogen, though carbon monoxide and methane were also being considered as reductants; and molten magma electrolysis. The reduction of ilmenite with hydrogen depends on the purity of the feedstock and, if sulphides are present, a purification step is required to remove the toxic hydrogen sulphide. Reduction with methane requires high operating temperatures and a multi-step process.

But it is the electrolysis route that Taylor now favours. ‘It has come a long way and has great potential for the production of oxygen on the Moon today,’ he says. The process could produce oxygen from any metal oxide. It is derived from the FFC-Cambridge process, named for the inventors, Fray, Farthing and Chen, for the electro-deoxidation of metals and metal oxides, which involves placing a metal-oxide cathode into a molten calcium chloride salt bath (CaCl2).

In the conventional set-up, a porous metal oxide body is the cathode and a carbon-based material such as graphite is the anode; the electrolytic cell is run at around 900°C. Oxygen is stripped off from the oxide cathode in its ionic form, and transported though the molten salt electrolyte. The negatively charged oxygen erodes the carbon anode and produces carbon dioxide.

To generate oxygen by this process, not carbon dioxide, the Cambridge scientists needed an inert anode. For this, they used a mixture of calcium titanate and calcium ruthenate. Because this anode barely erodes, the reaction between the oxygen ions and anode produces oxygen. The cathode is regolith from Moon rocks.

‘NASA sent us some simulants and we were able to generate oxygen through this process,’ says Fray. ‘At the start you get oxygen and iron, then the reduction of other oxides and more oxygen, so one advantage of the process is that you can get a mix of metals, an alloy.’

Fray envisages the power for the process being supplied by solar panels, or a small nuclear power reactor. The relatively low temperature of 900°C, the simplicity of the process and possibility to use robotic technology are advantages, he says (Planetary Space Science, 2012, 74, 49).

Another plus, says Fray, is that the FFC process has been refined and developed by the spin-out UK company Metalysis to provide an alternative to the Kroll process for extracting titanium metal and tantalum. Here the metal oxides serve as the cathode in the electrolytic reaction, with a carbon anode. The process should ultimately benefit any lunar process.

One slight disadvantage of this approach is that you have to bring the electrolyte calcium chloride to the Moon, Fray acknowledges, and you must also consolidate the Moon’s loose regolith before processing. But he envisages robots collecting, transporting and consolidating lunar material, something NASA is keen on.

Another competing process – high temperature molten oxide electrolysis (MOE) – is emerging from the Massachusetts Institute of Technology (MIT) in the US. High temperature electrolysis is used to make liquid metal and oxygen from a metal oxide feedstock but requires rare and consumable anode materials like iridium. It is being investigated as a way of producing iron without the colossal carbon dioxide emissions generated by traditional steel smelting.

As well as more environmentally friendly steel making (C&I, 2013, 6, 6), material scientist Donald Sadoway at MIT has also used this method to produce oxygen and metal from Moon rock simulants. This involves heating regolith up to 1600°C, turning it into a molten mass. When potential is applied across the electrodes, oxygen evolves at the anode and metal is deposited at the cathode.

The successful deployment of MOE thus hinges upon finding an inert anode capable of sustained oxygen evolution. According to Sadoway, iridium is the ideal anode metal because its oxide cannot form at above 1200°C so it remains as a metal during electrolysis, rather than corroding away because of oxidation (Journal of the Electrochemical Society, doi:10.1149/1.3560477).

But it is expensive. ‘NASA could justify the cost of iridium because it costs $100,000 per kilo to move mass from Earth’s gravity to the Moon, so it matters little what the cost of the precious metal is. Making oxygen on the Moon has a very different price point [to steel production on Earth],’ says Sadoway. ‘Initially you would use iridium, though you might make some improvements in terms of precious metal alloys. You can wait on making metal on the Moon but you need six pounds of oxygen per day or you won’t be around tomorrow. You can’t take chances with that.’

Not to be thwarted, Sadoway went on to discover a chromium alloy anode that is up to the task (Nature, doi: 10.1038/nature12134). Fray acknowledges this discovery as having implications in the future for extraterrestrial oxygen and iron generation, and could be a boon to Solar System exploration (Nature, doi: 10.1038/nature12102).

While some chemists see the high temperatures in Sadoway’s approach as a deterrent, he disagrees. ‘What I found is that the higher temperatures give you higher throughput and the energetic cost is quite small,’ he explains. ‘When you get beyond a certain cell size, the current generates enough heat to keep the cell at operating temperatures. That is how aluminium electrolysis works. To run that cell at 960°C, there are no external heaters; it just uses the current between the electrodes.’

Sadoway explains that the maximum temperature is in the centre of the cell, while at the walls the temperature can be below the melting point of the electrolyte. ‘You can have liquid electrolyte and liquid metal lying inside a frozen skull of electrolyte and metal. You don’t have containment issues, which is really elegant. People look at this and say it’s insane, you are operating at 1600°C. But it is not an isothermal reaction.’

Sadoway adds that when you look at the composition of lunar regolith it varies little. Scientists hypothesise that the Moon formed when a big mass of Earth was flung into orbit. ‘As this liquid blob solidified it did so in a way that it is graded in terms of melting points. So the outer surface of the Moon is nearly of uniform composition. I am willing to bet that as you drill down into the Moon you would find that there is a gradation in composition from the outer surface to the core,’ says Sadoway. ‘The composition at the surface is a few per cent iron, some silicon dioxide, some aluminium oxide and a little bit of titanium dioxide.’

The basicity of the material – that is the amount of free oxygen versus chemically bonded oxygen – varies little across the Moon, based on samples analysed so far. ‘For processes proposed for reducing regolith into oxygen with hydrogen or carbon monoxide, such variations have a huge impact on their efficiency and yields. With MOE, it’s not an issue.’

Sadoway reasons that a plurality of metals results the longer the electrolysis is run. ‘Initially you make iron, then once iron oxide has been depleted, the reaction on the negative electrode switches to production of silicon. We determined that the most efficient operation would deplete about half of the charge, and that would leave you with a liquid ferro-silicate alloy,’ Sadoway explains. The silicon could be used to make photovoltaic (PV) arrays and iron can be used as a structural metal. Since the Moon has no atmosphere and no moisture, ‘plain iron is tantamount to stainless steel,’ says Sadoway, ‘as there are no agents on the Moon to corrode the iron’. That gives you PV arrays for energy and metals for building.

Taylor comments: ‘These processes being developed will become important and I think they are the way of the future. But I don’t think we will be able to take what we want to the Moon. We need to get a gas station there. You spend 85% of the mass of a rocket just to get away from Earth’s gravity and it’s all fuel. If we could stop on the Moon, we could then go on to Mars. We could make water, oxygen and other valuable resources.’

However, the cuts to NASA’s budget coupled with the success and interest in robotic explorers have led to a decline in interest by NASA for these lunar reactors. Sadoway comments: ‘We made really good progress and just as it seemed we were getting to successful output at a large scale, the funding dried up.’ Fray is more optimistic: ‘Enthusiasm for the programme diminished, but I think they are starting to look at it again.’

In response, NASA says it ran a competition in the mid-2000s for teams to generate breathable oxygen from simulate lunar soil. The challenge expired in June 2009, with the purse of $1m unclaimed. There simply wasn’t enough interest in the challenge and the funds were moved to another competition.

In a recent project, NASA is seeking to use water on the Moon and Mars to produce oxygen and other materials ( but Fray points out that much of the water on the Moon is buried as ice in the polar regions, places where it would be extremely difficult to generate energy from solar cells. Some of the water lies in craters at temperatures of -230°C or less, which will make its mining and melting energy-consuming.

Fray and Sadoway believe space policy has shifted; it is no longer fashionable to view the Moon as a jump off point. Space rovers and robots reign supreme, leaving some disappointed chemists in their contrails. If the winds shift again, ideas for oxygen reactors will be of interest. Whether they will come from the US or emerging space exploring nations like India or China remains to be seen.

Anthony King is a freelance science writer based in Dublin, Ireland

Become an SCI Member to receive benefits and discounts

Join SCI