Compared to gases in primitive meteorites, Earth’s present day atmosphere is depleted in xenon. A new model, published in Nature, suggests that the mineral perovskite holds the key to understanding the riddle.

Early last week I was contacted by Andy Extance, a journalist working for Chemistry World, who asked if I would be willing to give my thoughts on a Letter to Nature which was coming out shortly. The article in question was: “The origin of the terrestrial noble-gas signature” by Svyatoslav Shcheka and Hans Keppler of the University of Bayreuth, Germany. “Oh, ummm, well, eh, it’s not really my area of expertise” I ventured initially, but Andy was not put off. “ Well they do reference one of your papers, Dr Greenwood, that’s how I got your name”, and he followed up with something along the lines of “and what would be great would be to get your views on how their model works in the context of an early magma ocean on the Earth”. Well, I couldn’t really refuse, it sounded like a very interesting piece of work and I was flattered to have been asked to provide my opinion on its merits.

What I realised quite quickly, as I glanced through the pdf supplied by Andy, was that my own paper had been cited out of context. We hadn’t, as suggested in the Nature article, discussed magma oceans “very early in Earth’s history”. In fact our paper wasn’t about Earth at all, but looked at the evidence from oxygen isotopes that magma oceans may have existed on asteroids such as 4 Vesta. Ah well, never mind, a citation is a citation and you need all the help you can get in the current environment. So, I guess in a way I was off the hook and could have decided at this point to let Andy know that I wasn’t the right chap for the job. However, while the noble gas geochemistry of early Earth is certainly not my area, I could call on a vast amount of relevant expertise within out group in the person of Dr Sasha Verchovsky. I had a quick chat with Sasha and he agreed to look at the article.

Andy only needed our comments by the following afternoon, so there was a bit of time to play with. But Sasha got stuck in straight away and came back to me with some general observations within a couple of hours. I was a lot slower off the mark. It can get fairly hectic in my household during the early evening, what with chores and homework to sort out. It was only later, when things had calmed down that there was a chance to ponder the merits of the new Shcheka and Keppler model. So just before midnight, with a cup of tea for company, I got back to Andy with some thoughts about the paper. I also forwarded Sasha’s comments from earlier in the day.

OK so what’s the paper all about?

The work discussed in the paper relates to the so-called “Missing Xenon Problem”, which crudely stated, refers to the fact that xenon in some planetary atmospheres (including Earth and Mars) is relatively depleted with respect to argon, when compared to the distribution seen in certain types of chondritic meteorites. The origin of this depletion is not well understood and a wide range of mechanisms have been put forward to account for the “missing xenon”. Some models invoke hiding the xenon in various near-surface reservoirs (glacial ices, clatherates, sediments, silicates), it has also been suggested that the xenon was concentrated in the Earth’s core.

Shcheka and Keppler claim that all these previous models have serious flaws and instead suggest that xenon was trapped in the lower mantle phase perovskite. Their model is based on high-pressure experimental work, in which they demonstrate that argon, and to a lesser extent krypton, are significantly more soluble in perovskite than xenon. They suggest that their experimental results provide “a simple explanation for the depletion of xenon relative to argon observed in Earth’s atmosphere”.

So what’s the explanation?

Shcheka and Keppler suggest that the original atmosphere of Earth was essentially solar in composition i.e. there was no depletion of xenon compared to argon. At this time the Earth would have been nearly molten, the so-called magma ocean. Perovskite crystallizing in the lower mantle would have trapped argon and to a lesser extent krypton, but very little xenon. Meanwhile, the Earth’s primary atmosphere was lost due to hydrodynamic escape as a result of high surface temperatures. Once the magma ocean had crystallized mantle convection was able to replenish argon and krypton in the atmosphere from the perovskite reservoir, but not xenon.

Both Sasha and I thought the model had merits. It was a new look at an old problem and was backed up by the results from some very challenging experiments. If nothing else, the authors clearly demonstrate that perovskite under lower mantle conditions has the potential to retain significant amounts of argon and krypton, but only traces of xenon. Our main reservation about the work was the extent to which the experimental conditions were relevant to those that would have existed in the early Earth’s lower mantle.

The work was undertaken using multi-anvil apparatus, with glass charges loaded together with 40-80 bar of argon or xenon. Sasha’s worry was that noble gases might enter the perovskite structure under such extreme conditions, but not in the natural case where the gas phase is dissolved in molten silicate magma. To be fair, what the authors appear to be suggesting is that as the silicate magma crystallizes volatiles will be concentrated in residual melt until saturation is reached and this would then lead to the formation of a residual gas phase. It is the interaction between this residual gas phase and perovskite that the experiments seek to model.

Like all new ideas in science, the model of Shcheka and Keppler is going to generate significant controversy and debate. It will certainly require a lot of additional work to fully test the idea. However, it is an interesting example of how an innovative set of experiments can provide a new way of looking at an old problem.

Links: Andy Extance’s Chemistry World article