This transcript was generated automatically and lightly edited for readability.

Oliver Strimpel: This is Geology Bites with Oliver Strimpel. In a recent episode, Nadja Drabon spoke about newly discovered zircon crystals that formed during the late Hadean and early Archean, when the Earth was between 500 million and a billion years old. The zircons revealed information about processes occurring in the Earth's nascent crust, casting light on when and how modern-day plate tectonics may have started. In this episode, we talk about a very different source of information about the early Earth, namely the abundances of noble gases occurring within present-day basalts. It turns out that these can probe the Earth's mantle and atmosphere even further back in time, to the first hundred million years of Earth history.

Sujoy Mukhopadhyay leads a team of researchers who have developed new techniques for measuring the abundances of noble gas isotopes in a variety of Earth materials. By combining the results of these measurements with geochemical models, he has shed light on questions about the very early Earth and planet formation that have challenged researchers for decades. Here, we focus on one of these: do any structures originating from the very early Earth survive in today's mantle? Sujoy Mukhopadhyay is Professor of Geochemistry at the University of California, Davis. Sujoy Mukhopadhyay, welcome to Geology Bites.

Sujoy Mukhopadhyay: Thank you, Oliver. I'm delighted to join you, and I am really excited to chat with you.

Oliver Strimpel: How is it that we are able to use noble gases to probe the very early Earth? Is it to do with the fact that they are chemically inert?

Sujoy Mukhopadhyay: Yes, you're absolutely right. It has got to do with their chemical inertness. So in that respect, they keep a memory of these very early events. And likewise, we can combine the variety of isotopes from the different noble gases to probe the Earth on different time scales.

Oliver Strimpel: The noble gases form a group in the periodic table, from helium, the lightest, through neon, argon, krypton, xenon and radon. How were these elements produced?

Sujoy Mukhopadhyay: In the noble gases, we have isotopes that we call primordial, which means these are the isotopes that were produced in stars, and then the solar system acquired these elements from neighboring stars. And then there is a set of isotopes that are produced by radioactive decay of elements like uranium and thorium and potassium. The really fascinating aspect about this is that the half-lives of the isotopes that ultimately end up producing these noble gases range from something like 14 billion years for thorium, all the way down to something like 16 million years for an isotope of iodine that produces a particular isotope of xenon. So in many respects, this very wide range of half-lives gives us the means to probe the Earth on very different time scales.

Oliver Strimpel: You focused on a particular isotope of xenon to probe the early Earth, and that was xenon-129. Can you explain why?

Sujoy Mukhopadhyay: Yes. Xenon-129 comes from the radioactive decay of iodine-129. This particular isotope was present when the solar system formed, but its half-life was 16 million years, and typically the useful time range for any particular radioisotope is roughly six times its half-life. So that means that after about 100 million years, iodine-129 is no longer around. All of it has been converted to xenon-129. Therefore the Earth is able to keep a memory of this radioactive decay happening within the first 100 million years, because after 100 million years, there was not going to be any new production of xenon-129. So any kind of variability that people see on the Earth would have to relate back to this first hundred million years.

Oliver Strimpel: I said in my introduction that you can use such noble gas abundance measurements to shed light on whether today's mantle retains any of the original structure from which it first differentiated. How do you connect measurements of xenon isotopes with the mantle?

Sujoy Mukhopadhyay: What we do is we start probing the products of melting of the modern-day mantle. For example, I go to places like Hawaii or Iceland, which are really nice places to visit in their own right, but also happen to be locations of mantle plumes, or hotspots. For these mantle plumes, we now have evidence from seismology that they're derived from material very near the core-mantle boundary. So in that respect, these plumes are acting as chemical probes of the composition of the deepest parts of the Earth's mantle. Likewise, when we go to a mid-ocean ridge spreading center, where new oceanic crust is being created, the partial melting that we see is derived from material that's present in the shallower part of the mantle. So then, by comparing the composition of basalts from the mantle plumes with the composition of the basalts from the mid-ocean ridges, we're effectively probing different depths within the Earth's mantle.

And so if we see differences in the xenon-129 composition between these two different regions of the mantle, we can immediately say, aha, the system could only have changed within the first 100 million years. The fact that we see differences points to the fact that they had to have come from that first 100 million years. So that's how we connect our measurements of mantle-derived basalts to what's actually within the Earth's interior.

Oliver Strimpel: That's fascinating. I'm struck by the fact that you can actually measure the abundances of noble gases in various kinds of basalts. So how do noble gases get into the rock in the first place? I mean, being inert, they can't really form minerals or react with the minerals in any way, can they?

Sujoy Mukhopadhyay: You're right. They can't really chemically react with the minerals, and because of that, they don't actually like to be in a mineral, because they are not forming any kind of chemical bond with their neighboring constituents. So often they tend to exist in what might be distorted sites within a mineral lattice, or they might be present in very large ring structures in certain kinds of minerals, like amphiboles. But those sites are rare, and because of that, they would rather be in a liquid phase, where there are more opportunities for these noble gases to be present. And compared to a liquid phase, a vapor phase is their preferred host, because then they're really happy.

So what happens is, when the mantle starts partially melting, the noble gases are among the first elements to leave the mineral lattice and go into the melt. But there are other compounds, such as CO2 or water, which also typically do not like to be in minerals, and they effectively go from the mineral into the melts that form as a parcel of mantle starts to rise towards the surface and the pressure release causes partial melting. Then the melt starts rising to the very surface of the Earth, and as it rises, ultimately a CO2 vapor bubble might form. And then the noble gases will right away go from the melt into the CO2 vapor bubble. When this melt erupts onto the sea floor, it immediately quenches, because it is in contact with really cold ocean water, and it forms a glass, and the glass effectively encapsulates these vapor bubbles. We then take these glass samples into our laboratory, and we crush them under vacuum and release the vapor bubbles, which releases the noble gases. And that's what we end up measuring in our laboratory.

Oliver Strimpel: That's fascinating, that they come out into the gas phase. But that makes me wonder why these noble gases went into the solid phase at all to begin with. When the Earth was formed, why didn't they just all stay in the atmosphere?

Sujoy Mukhopadhyay: That's a really fascinating question, and we can speculate as to why. In the early Earth, at some point there was a magma ocean, where maybe most of the mantle was molten. This magma ocean would eventually degas to form the very earliest atmosphere, and the noble gases are going to partition between this magma ocean and the atmosphere. Their solubility in this magma is very low, but it's not zero. So there's still going to be a finite amount in the magma ocean. Then, as the planet cools by losing heat to space, the magma ocean is going to slowly start to crystallize. And basically the noble gases are prisoners at this point, because as the minerals are forming, they are being shoved into these minerals, because they can't escape. So they are present in very low abundances in these melts, and then when the melt starts to solidify, they enter, or are forced to enter, the minerals, but again in very low abundances. And then they wait for the first opportunity to try and escape, and that first opportunity is remelting of the mantle.

Oliver Strimpel: So now we have these gas bubbles, which, in the case of the mid-ocean ridge basalts, are spewing out onto the sea floor, and in the case of the ocean islands, are coming up in volcanoes like in Hawaii. But at that point they must start being affected, in one case by contamination by seawater, and in the other case by weathering with the atmosphere on the islands. The atmosphere itself also has some noble gases. We hear quite a bit about the helium in the atmosphere, and we know that radon is sometimes generated in our basements if our houses are built on granite, and we've got to watch out for that. So do these gases contaminate the gas bubbles, and how do you get at the actual mantle abundances?

Sujoy Mukhopadhyay: Yes, this is one of the frustrations that we have to deal with in these measurements: we often don't get a sample that is completely devoid of contamination by atmospheric gases. And as you point out, when the basalt erupts onto the sea floor, it might get contaminated by the ocean water, which has some dissolved noble gases in it. For the mantle plumes, we also try to collect the samples that erupt underwater, as opposed to the lavas that erupt straight into the atmosphere, subaerially. The reason is that if they erupt under water pressure, there's still a chance that the gas bubbles will be trapped in the liquid itself. Whereas if they erupt straight into the atmosphere, the chances are that the bubbles will be lost very quickly to the atmosphere, and the solidified basalt sample won't have a lot of gas bubbles. But independent of this, both the plume basalts and the mid-ocean ridge basalts will have a little bit of atmospheric contamination.

The way we try and tease this out is by crushing the sample in multiple steps, because the expectation is that there are going to be points of weakness in these basalt samples, like fractures, through which the contamination enters the sample. So gas bubbles that are close to a microfracture might have a lot more of this atmospheric contamination, whereas little tiny gas bubbles that are much further away from the fractures may have very little or none. So when we start cracking the sample open, the first set of bubbles released will be the ones that are very close to the fractures, because the samples break along these pre-existing planes of weakness. So typically what we see when we first start crushing the sample is that the measured noble gas compositions look largely similar to what you would expect in seawater, which is the atmospheric signature dissolved in seawater. And then as we progressively crack the sample into smaller and smaller bits, we start to see the composition move away from the atmospheric composition towards the mantle composition.

And very interestingly, what we find is that both the mid-ocean ridge basalts and the mantle plume basalts start out at the same point in their initial crushes, which is the atmospheric composition. But then, as we progressively crush the samples, the trends start diverging in two different directions, which indicates that whatever composition is being contaminated, it's not the same. If it were the same, both sets of rocks would have defined a single trend, because it would be a single mantle composition contaminated by a single atmospheric composition. But we see two distinct trends, and that's telling us there are two distinct sources within the Earth's interior.

Oliver Strimpel: How do you actually measure the noble gas abundances? You crush the basalt in a vacuum chamber, and then how do you get the abundance of a noble gas? I'm curious, because it's so inert.

Sujoy Mukhopadhyay: Again, that's a fascinating question. When we crush the basalt and we release the gas bubbles, most of the gas is actually CO2, and dissolved in that CO2, at the level of parts per trillion or parts per quadrillion, are these noble gases. So one of the first things we have to do is get rid of the CO2, and get rid of any nitrogen, any little bits of methane or water that might be present. The way we do this is we expose all of the gases to a highly reactive surface, and that highly reactive surface absorbs all of these constituents, like CO2, water, nitrogen and so on. But the noble gases are left behind, because they're chemically inert. So that's how we purify the noble gases. Then we can separate them from each other using a very cold cryogenic trap, where the noble gases get trapped onto a metallic surface. Then we progressively increase the temperature of that metal surface, and that slowly releases first helium, then neon, then argon, and so on. As each of the gases gets released from this very cold cryogenic trap, we take that gas and put it into a mass spectrometer to measure how many atoms of helium we have and what its isotopic composition is. And then we repeat it for all of the other noble gases, like neon and argon and krypton and xenon.

Oliver Strimpel: So let's come back to the progressive crushing. You start out with the first bubbles being essentially reflective of the atmosphere, and then as you crush more and more, the abundances diverge, until, I guess, the very, very last crushes are more reflective of the actual mantle abundances, on the one hand of mid-ocean ridge basalt, and on the other of ocean island basalt. So what do those results show?

Sujoy Mukhopadhyay: That's where the half-life of the parent radioisotope, iodine-129, comes in. What we are detecting are different signatures of xenon-129 in the ocean island basalts and in the mid-ocean ridge basalts, and the difference could be created in only two ways. One is through differences that were inherently present in the mantle early on, because of differences in, let's say, the degree of degassing of the Earth's interior. Or they might be created by subduction of atmosphere. What these two different trends tell us is that subduction of atmosphere can't possibly explain this difference between the mid-ocean ridge basalts and the ocean island basalts, because they don't lie on the same mixing trend. So our only option in terms of an interpretation is that the differences in xenon-129 come from the decay of iodine-129 in the mantle. And as I pointed out earlier in the podcast, iodine-129 was only around for the first hundred million years of Earth's history. So that means that this difference in xenon-129 could only have been produced within that first hundred million years, and that's the maximum range.

Oliver Strimpel: Which of the two kinds of basalt has the lower amount of xenon-129?

Sujoy Mukhopadhyay: It turns out that the lower amount of xenon-129 is present in the plume mantle, or the ocean island basalt mantle.

Oliver Strimpel: That's the one, then, that retains an identity that goes all the way back to the first 100 million years of Earth history. Is that right?

Sujoy Mukhopadhyay: That's right. We think that its identity is related to its history within that first hundred million years, and that it's telling us something about the differences in the degree of gas loss from these two different regions of the Earth's mantle. Because, again, one critical point to remember is that xenon, or any of the other noble gases, doesn't just escape out of the mantle on its own. They usually take a ride on some other gas species, like CO2 or maybe water. So their history is tied to the history of these other gases, like carbon dioxide or water.

Oliver Strimpel: OK. So the ocean island basalt, which we think was sampled from much, much deeper in the Earth than the mid-ocean ridge basalt, which we think comes from the middle or perhaps upper levels of the mantle, has less of the xenon-129, and therefore it is what retains the memory of the very, very earliest formation of the Earth. So does that tell us that, in the mantle, it has not in fact experienced any mixing with the material that forms the mid-ocean ridge basalt?

Sujoy Mukhopadhyay: I don't think it indicates that it has had no mixing. I think what it indicates is that it has had a limited amount of mixing. Another way of saying it is that after 100 million years, the Earth has never been homogenized. Because if it had been homogenized, such as by mixing, then we wouldn't see this difference in xenon-129. Once we homogenize it and mix away this difference after 100 million years, I don't have a mechanism to regenerate the difference. So the way I think about it is that, yes, there has been mixing between the mid-ocean ridge basalt source and the ocean island basalt source, but it's limited. If I were able to dial back time all the way to 4.4 billion years ago, I would predict that the differences I'd see compared to today would have been much, much larger. So over time that difference has been whittled away by mixing, but it has not been erased.

Oliver Strimpel: So how might such a separate source material at the bottom of the mantle have formed so early on, and maintained its separate identity, at least in part, throughout four and a half billion years of Earth's history?

Sujoy Mukhopadhyay: That is one of the million-dollar questions. One potential mechanism for creating these differences might be related to Earth's violent birth, when the planet was being bombarded by other embryos, or protoplanets, generating very deep magma oceans. More recent studies have shown that among the products of crystallization of these magma oceans could be iron- and magnesium-rich phases, which settled to the bottom of the mantle. Those phases are going to be inherently denser than the rest of the mantle. So one could imagine a scenario where there was a giant impact during Earth's growth that melted the planet, maybe all the way down to the core-mantle boundary, and you crystallized this iron- and magnesium-rich layer, which was inherently denser. Then subsequent magma oceans created by other impacts never reached down to that depth, and so that inherently denser material has effectively survived all through the planet's history. So that's one potential explanation. I don't think we have a definite answer at this point to how exactly this reservoir was produced. And I think that is one of the challenges: figuring out how it was produced, and why it is still around after such a long time. Why didn't later giant impacts, such as maybe the Moon-forming giant impact, wipe it out?

Oliver Strimpel: It reminds me of a couple of earlier podcasts in which two speakers, in fact, Allen McNamara and Matt Jackson in particular, spoke about these very large structures we see when we make seismic images of the Earth, called large low-shear-velocity provinces, where it appears that the shear waves are traveling a little bit slower, which one can interpret as being because they're a little warmer than the surrounding mantle, or alternatively, or maybe better, both, that they're made of a different material. Is it possible that these reservoirs you're talking about, which have the separate identity, correspond to these large low-shear-wave-velocity provinces that we can actually see?

Sujoy Mukhopadhyay: Most likely the answer to that is yes. Because, again, as we look at recent seismic evidence, we see that these ocean island basalts, these mantle plumes, are being derived from these large low-shear-wave-velocity provinces. And I'm of the group that believes that these large low-shear-wave-velocity provinces are not just warmer, but are actually also denser than the surrounding mantle. They're not made entirely of material that has existed in pristine form for over four and a half billion years, but they have at least chunks of material that go all the way back to 4.5 billion years. There might also be other material that has been recirculated from the surface into these large low-shear-wave-velocity provinces, and we actually do see some chemical evidence that that has happened.

Oliver Strimpel: And have people built geodynamic models? I think Allen McNamara showed some of that with the simulations that he did, which can generate stable patterns that are consistent with what we know about how slabs subduct and sink to very great depths, and are then possibly recycled from somewhere quite far down, near the core-mantle boundary.

Sujoy Mukhopadhyay: Yes, I think Allen's work does show that these features could be stable for the entire time of Earth's history, about four and a half billion years, and there are other groups that have likewise shown that. But I'm not sure that at this point there is a general consensus that these provinces, which are maybe a little bit bigger than the size of continents on the surface, actually go all the way back four and a half billion years. There's another school of thought that says they are created entirely through subduction of Earth's tectonic plates.

Oliver Strimpel: But now we know that can't really be the case, right? Because of the work you just described.

Sujoy Mukhopadhyay: That's correct. I would argue that when we put together the seismic evidence that these mantle plumes are coming from these large low-shear-wave-velocity provinces with the xenon composition that differs between mid-ocean ridge basalts and these ocean island basalts, there must be a component in this material that goes all the way back four and a half billion years. It doesn't rule out that maybe the vast majority is recycled, but there are chunks of material that are really, really old.

Oliver Strimpel: I'm struck by the coincidence that there happen to be noble gas radiogenic isotopes produced on a time scale that's very short compared to the age of the Earth, but still of the order of the time over which geodynamic processes can occur, i.e. a few tens of millions of years. That is what enables us to distinguish a reservoir that was isolated so early on in Earth history, within a time less than a few half-lives of the corresponding decay. Are there any other isotopes whose abundances we can measure that are produced by radiogenic decay with still shorter half-lives, specifically half-lives shorter than about 10 million years, which might enable us to probe still closer to the formation of the Earth?

Sujoy Mukhopadhyay: Yes, there are, and one of these isotopes is hafnium-182, which decays to tungsten-182. This system is particularly interesting because hafnium likes being in the mantle, while tungsten, the decay product, likes being in the core. So hafnium-tungsten has been used extensively to look at the differentiation between a planet's metallic core and its silicate mantle. Interestingly, there are some recent studies that show differences in tungsten isotopic composition between ocean island basalts and mid-ocean ridge basalts. Again, this is an indication that that signature might have been created within the time span over which hafnium-182 was alive. Because the half-life of hafnium-182, about 8.9 million years, is shorter than that of iodine-129, the total amount of time available is about 60 million years. Now there is some growing debate as to what the implications of this tungsten isotope difference between ocean island basalts and mid-ocean ridge basalts might be. Could it potentially also be an indicator of chemical interaction between the Earth's core and its silicate mantle? The core-mantle boundary is a fascinating region. It's probably one of the most reactive boundaries we have on the Earth. It is super hot, and on one side we have pure iron metal, or iron metal with a little bit of other stuff, and on the other side we have silicate material. So one should expect that there are some chemical reactions going on.

And there are other isotopes, such as palladium-107, which decays to silver. The half-life associated with that is even shorter, six and a half million years. There is an isotope of manganese, manganese-53, that decays to chromium and has a half-life of three million years, and recently this manganese-chromium system was used to look at the time scales of magma ocean crystallization on Mars. So there are other isotopes that might allow us to probe the Earth's formation much further back in time. And one of the ways forward, in my view, is to make all of these measurements on the same set of samples, so we can build a much more complete picture of what might be going on. So far, we are still measuring these different isotopes on different samples, so we are not necessarily always comparing an apple to an apple. It might be an apples-to-oranges comparison.

Oliver Strimpel: I'd like to ask you a final question. If you could command a limitless budget for either space-based or terrestrial research into the early Earth, how would you use it?

Sujoy Mukhopadhyay: I think I would design a mission to a comet, to bring some comet samples back to Earth, so we can study comets in much more detail. The reason I say that is because I'm fascinated by the question of what the composition of the early atmosphere was, and where the early atmosphere came from. In the noble gases, we now see evidence that the Earth's early atmosphere cannot have been generated simply by outgassing of its interior. But we also don't have a meteorite, or a carbonaceous chondrite necessarily, that we can put our finger on and say that this particular meteorite has all of the necessary characteristics to explain the atmospheric fingerprint. The noble gases are extremely useful for fingerprinting what kinds of materials might have contributed to the early atmosphere. And if we know that, we have a much better sense of what the overall composition of the atmosphere might have been. Was it CO2-dominated, or was it CO-dominated? What was the proportion of hydrogen in the atmosphere compared to H2O? We know so little about cometary noble gases, and comets in general, that I would argue that's where I would focus my attention: to bring a little bit of a comet back to my laboratory.

Oliver Strimpel: Do you have your eye on any specific comets that are likely to come into our solar neighborhood anytime soon?

Sujoy Mukhopadhyay: I cannot give you a direct answer in terms of a specific comet, other than that we would want to sample both the icy part of the comet as well as the rocky part. A lot of our observations of comets so far are actually of the icy part, which sublimates as comets come into the inner solar system. We have very little information about the rocky part. We do have some material that was returned by the Stardust mission, but not enough for somebody like me to make very precise measurements of all of the noble gases.

Oliver Strimpel: Sujoy Mukhopadhyay, thank you very much.

Sujoy Mukhopadhyay: Thank you, Oliver, for having me. This was wonderful. It was a delight chatting with you.

Oliver Strimpel: For more about Geology Bites, as well as pictures and illustrations that support this podcast, go to geologybites.com.