Apart from signs of water ice in permanently shadowed areas of some polar craters, the Moon’s surface has generally been considered to be very dry. Rocks returned by the various Apollo missions contain minute traces of water by comparison with similar rocks on Earth. They consist only of anhydrous minerals such as feldspars, pyroxenes and olivines. But much of the lunar surface is coated by regolith: a jumble of rock fragments and dust ejected from a vast number of impact craters over billions of years. It is estimated to be between 3 and 12 m deep. Much of the finer grained regolith is made up of silicate-glass spherules created by the most powerful impacts.

The scientific and economic (i.e. mining) impetus for the establishment of long term human habitation on the lunar surface hangs on the possibility of extracting water from the Moon itself. It is needed for human consumption and as a source through electrolysis of both oxygen and hydrogen for breathing and also for rocket fuel. The stupendous cost, in both monetary and energy terms, of shifting mass from Earth to the Moon clearly demands self-sufficiency in water for a lunar base occupied for more than a few weeks.
Remote sensing that focussed on the ability of water molecules and hydroxyl (OH–) ions to absorb solar radiation with a wavelength of 2.8 to 3.0 micrometres was deployed by the Indian lunar orbiter Chandrayaan-1 that collected data for several months in 2008-9. The results suggested that OH– and H2O were detectable over a large proportion of the lunar surface at concentrations estimated at between 10 parts per million (ppm) up to about 0.1%. Where did these hydroxyl ions and water molecules come from and what had locked them up? There are several possibilities for their origin: volcanic activity that tapped the Moon’s mantle (magma could not have formed had some water not been present at great depths); impacts of icy bodies such as comets; even the solar wind that carries protons, i.e. hydrogen atoms stripped of their electrons. Conceivably, protons could react with oxygen in silicate material at the surface to produce both OH– and H2O to be locked within solid particles. To assess the possibilities a group of researchers at Chinese and British institutions have examined in detail the 1.7 kg of lunar-surface materials collected and returned to Earth by the 2020 Chinese Chang’e 5 lunar sample return mission (He, H. and 27 others 2023. A solar wind-derived water reservoir on the Moon hosted by impact glass beads. Nature Geoscience, online article; DOI: 10.1038/s41561-023-01159-6)
He et al. focussed on glass spherules formed by impact melting of lunar basalts, whose bulk composition they retain. The glass ‘beads’ contain up to 0.2 % water, mainly concentrated in their outermost parts. This alone suggests that the water and hydroxyl ions were formed by spherules being bathed in the solar wind rather than being of volcanic or cometary origin and trapped in the glass. An abnormally low proportion of deuterium (2H) relative to the more abundant 1H isotope of hydrogen in the spherules is consistent with that hypothesis. Indeed, the high temperatures involved in impact melting would be expected to have driven out any ‘indigenous’ water in the source rocks. The water and OH– ions seem to have built up over time, diffusing into the glass from their surfaces rather than gradually escaping from within.
An awful lot of regolith coats the lunar surface, as many of the images taken by the Apollo astronauts amply show. So how much water might be available from the lunar regolith? The Chinese-British team reckon between 3.0 × 108 to 3.0 × 1011 metric tons. But how much can feasibly be extracted at a lunar base camp? The data suggest that a cubic metre (~2 t) of regolith could yield enough to fill 4 shot glasses (~0.13 litres). Using a solar furnace and a condenser – the one in full sunlight the other in the shade – is not, as they say, ‘rocket science’. But for a minimum 3 litres per day intake of fluids per person, a team of 4 astronauts would need to shift and process roughly 100 m3 of regolith every day. Over a year, this would produce a substantial pit. But that assumes all the regolith contains some water, yet the data are derived from the surface alone …See also:Glass beads on moon’s surface may hold billions of tonnes of water, scientists say. The Guardian, 27 March 2023.