The finding that magmatic material from the Moon is more enriched in the heavy isotopes of zinc than its terrestrial and Martian analogues prompts fresh thinking about the origin of our natural satellite. See Letter p.376
Like the perfect Martini, the Moon has a reputation for being dry. Most obviously, it lacks oceans on its surface, other than those of crystallized magma. Recent evidence suggests, however, that the Moon has a touch of damp. Several studies have found concentrations of water much higher than expected in lunar minerals1,2 and quenched magma3,4. From these observations, it has been suggested that the interior of the Moon may contain similar concentrations of water to those in Earth's mantle. Yet it is difficult to make accurate measurements of water, and rather few samples are amenable to robust analysis. Instead, the abundances of less analytically challenging volatile elements can be studied as proxies for water, allowing investigation of a wider and probably more representative range of lunar materials. In this issue, Paniello and colleagues5 provide a new perspective on the history of the Moon's volatile elements, based on zinc and, in particular, its isotopic composition. They show that the Moon has a distinctly heavier Zn isotopic composition than Earth, consistent with the satellite having experienced a marked depletion in volatiles.
From our everyday experience, we would not consider Zn to be volatile: there is generally little swooning from its vapours. However, in cosmochemistry, elements that condense at relatively low temperatures (less than about 1,000 °C) from a tenuous gas, as Zn does, are termed volatile. The abundance of these volatile elements varies greatly between planetary bodies — which can be sampled by means of meteorites — but the origin of this variability remains poorly understood. From the first analyses of samples returned by the Apollo missions, it has been clear that the Moon has very low abundances of volatile elements. Yet it is a considerable challenge to translate elemental concentrations measured in the erupted, but now cooled, magmas (melts) at the surface, into a global volatile inventory for the Moon. The power of isotopic measurements, such as those made by Paniello et al., is that isotopes, unlike elements, are not significantly fractionated from each other during magmatic processes, and so erupted melts should faithfully record the composition of their deeper sources.
Using this isotopic approach, the authors show that the abundance ratio of Zn isotopes with mass numbers 66 and 64 (66Zn/64Zn) is nearly constant for a range of lunar magma types, including those thought to be most representative of the interior. Therefore, the results provide a well-constrained value of this ratio for the Moon as a whole. Most significantly, Paniello et al. find that this value is markedly higher — by about 1.5 parts per 1,000 — than that of Earth, Mars and primitive meteorites, which are believed to be the building blocks of the rocky planets. This difference may seem minor, but in terms of isotopic differences between planetary bodies, it is huge.
The exact mechanism by which the Zn isotopic composition of the Moon became heavier than that of Earth is unclear; however, most processes of volatile loss that can be envisaged predict this as an outcome. For example, during condensation of the Moon from a cloud of gas and melt, the heavy isotopes of Zn should have preferentially partitioned into the melt phase (Fig. 1). If the gas phase never fully accreted onto the Moon, the Moon would be left poor in volatiles, but also enriched in the heavy isotopes of volatile elements such as Zn. This diagnostic isotopic signature of volatile depletion has long been sought in lunar samples, and thus its eventual documentation by Paniello et al. is notable. Indeed, it was a major surprise that such an effect was not found in a high-precision isotope study6 of the volatile element potassiumalmost 20 years ago.
More recently, researchers have reported7 distinct isotopic differences between lunar and terrestrial chlorine, perhaps a more familiar volatile element. However, the lunar chlorine isotopic composition is not constant, and its trend to heavy values is argued to reflect variable loss of gas from magmas during magma eruption, rather than the interior composition of the Moon7. A more complete understanding of planetary volatile evolution requires an explanation of the contrasts in the isotopic signatures of these supposed chemical brethren. Of potential importance in this quest are better estimates of element volatilities under the specific conditions in which the Moon formed. Paniello et al. follow a traditional approach of assuming elemental condensation temperatures calculated for a hydrogen-rich, 'nebular' environment8. These values are appropriate for the conditions of the earliest Solar System but not for the accretion of the Moon from a silicate-rich debris disk (Fig. 1).
Paniello and colleagues' new Zn isotope data may also contribute to the reawakened discussion of the very mechanism by which the Moon formed. A theory known as the giant-impact hypothesis — according to which our natural satellite is the outcome of reassembled debris from a collision between the proto-Earth and another planetary body — has dominated recent thinking about lunar genesis. In particular, an oblique-impact scenario was shown9 to reproduce many of the physical and chemical attributes of the Moon. However, this model predicts that the Moon should be comprised predominantly of material from the colliding impactor rather than the target proto-Earth. Instead, the Moon has proven to be embarrassingly similar to Earth in several isotopic characteristics10,11,12, seemingly requiring it to be derived almost wholly from Earth. Models have thus investigated mechanisms by which the isotopic composition of the Earth and Moon could have become homogenized after the impact13. The notable differences in Zn isotopic composition between Earth and Moon documented by Paniello et al. may be more readily reconciled with an entirely new model for Moon formation, in which the impactor hits a rapidly spinning proto-Earth14. This flurry of recent developments emphasizes a waxing interest in our ever-puzzling satellite.
Boyce, J. W. et al. Nature 466, 466–469 (2010).
McCubbin, F. M. et al. Proc. Natl Acad. Sci. USA 107, 11223–11228 (2010).
Saal, A. E. et al. Nature 454, 192–195 (2008).
Hauri, E. H., Weinrich, T., Saal, A. E., Rutherford, M. C. & Van Orman, J. A. Science 333, 213–215 (2011).
Paniello, R. C., Day, J. M. D. & Moynier, F. Nature 490, 376–379 (2012).
Humayun, M. & Clayton, R. N. Geochim. Cosmochim. Acta 59, 2131–2148 (1995).
Sharp, Z. D., Shearer, C. K., McKeegan, K. D., Barnes, J. D. & Wang, Y. Q. Science 329, 1050–1053 (2010).
Lodders, K. Astrophys. J. 591, 1220–1247 (2003).
Canup, R. M. & Asphaug, E. Nature 412, 708–712 (2001).
Wiechert, U. et al. Science 294, 345–348 (2001).
Touboul, M., Kleine, T., Bourdon, B., Palme, H. & Wieler, R. Nature 450, 1206–1209 (2007).
Zhang, J., Dauphas, N., Davis, A. M., Leya, I. & Fedkin, A. Nature Geosci. 5, 251–255 (2012).
Pahlevan, K. & Stevenson, D. J. Earth Planet. Sci. Lett. 262, 438–449 (2007).
Ćuk, M. & Stewart, S. T. Early Solar System Impact Bombardment II Poster 4006 (Lunar & Planetary Inst., 2012).