This transcript was generated automatically and lightly edited for readability.

Oliver Strimpel: This is Geology Bites with Oliver Strimpel. Imagine forcing a piece of material no bigger than a pencil eraser into a vise that generates pressures and temperatures equivalent to those at the center of the Earth. This is what happens at the Z Machine, a facility at Sandia National Laboratories, and my guest today uses it to understand what materials are really like deep inside planets. Most of the material inside the Earth and other rocky bodies exists under conditions utterly unlike anything on our surface. Descend just a short distance below the surface, and pressures and temperatures quickly reach values orders of magnitude beyond our everyday experience. To understand how different planets work, why some are geologically dynamic and others are static, and how their interiors have evolved since they formed, we really want to replicate those conditions in the lab and see how familiar materials respond. And that is the focus of Steve Jacobsen's research. He is a mineralogist and materials scientist, and a professor in the Department of Geological Sciences at the University of Colorado Boulder. What makes his work especially striking is the range of extremes he tackles. In this episode, we'll explore three very different environments he has sought to replicate: the crushing pressures at the core–mantle boundary, nearly 3,000 kilometers beneath our feet; the mantle transition zone, where minerals can lock away vast quantities of water; and a completely different kind of extreme, the harsh vacuum and radiation of the lunar surface, where NASA's Artemis program hopes to build a permanent base. Steve Jacobsen, welcome to Geology Bites.

Steve Jacobsen: It's great to be here, Oliver. Thank you for inviting me.

Oliver Strimpel: Let's start at the very bottom of the mantle, the core–mantle boundary. Why is it important to study this region, and how can we try to comprehend the pressures we're dealing with down there?

Steve Jacobsen: The reason these are interesting environments to study materials in is that, in most of the universe, those are the conditions under which materials persist, including in planets and stars. Now, when we look at our own planet, the Earth, the biggest boundary below the crust is really the core–mantle boundary. As you mentioned in the introduction, it's at around 3,000 kilometers depth, like flying from Chicago to Hawaii, but straight down. The unit of pressure that we use is called a gigapascal, so a billion pascals. To put that into perspective, one gigapascal translates to about 10,000 bars, and a bar is about an atmosphere. So you could think of one gigapascal as being around 10,000 times the squeezing force that the air is putting on your forearm. Pressure is force over area, and so we'll be looking at how, using just the very tiny tip of a diamond, a tip about the diameter of a human hair, we're able to generate these pressures in static compression experiments. The pressure at the core–mantle boundary is around 120 gigapascals. It's quite a lot.

Temperatures we know much less well. That is to say, we can calculate the variation of density or pressure within the Earth pretty easily, just using mechanics, but the temperature is unknown. We have estimates based on experiments and theory. For example, if you were able to determine the melting temperature of iron at 120 gigapascals, you'd have a pretty good idea of what the temperature might be. We estimate the temperature at the core–mantle boundary to be somewhere between 3,000 and 4,000°C.

Oliver Strimpel: So, as I mentioned, you use the Z Machine at Sandia National Laboratories, a facility originally designed to simulate thermonuclear detonations. Can you walk us through what actually happens during an experiment?

Steve Jacobsen: It's a very large storage device for electricity, and it uses a concept called pulsed power. That is to say, if we can compress electrical energy in space and time to the point where there's a very high density of energy, we can transfer that energy to a material and basically blow it up, and by doing so, measure in situ, during a very brief experiment, what some of the thermodynamic properties of those materials are. So it's plugged into the wall and it charges up. The capacitors are able to store around 20 megaamps of electricity. That's 20 million amps, and the target in the center of the machine is meant to be a short, an electrical short. Imagine if you were to short out two wires: you would see a spark. That's what the Z Machine does. But along the way, it's compressing those 20 megaamps into a very, very short pulse, just a few nanoseconds in duration. And by shorting out that current at around 100,000 volts, we're able to push on the sample to many, in fact thousands, of gigapascals.

Oliver Strimpel: Traditional shock compression follows something that's known as the Hugoniot curve, where both pressure and temperature shoot up together. But I gather the temperatures that you get that way are much higher than what we'd actually expect inside the Earth. So you've found a way around this. Can you explain how you do that?

Steve Jacobsen: Yeah, you brought up the Hugoniot. That's a very important concept in shock physics. But as you mentioned, the states of pressure, density and temperature along the Hugoniot don't quite match what we find in the Earth. The pressure–temperature profile that we experience when we go down into the Earth is called the geotherm, and that's basically a measured temperature in a borehole, for example. But once you go below the crust, the temperature–pressure path follows what we call an adiabat. Adiabatic compression means there's no flow of heat in or out during compression. So what we want to try to do is mimic an adiabat, not the Hugoniot.

Oliver Strimpel: So how does the experimental setup enable you to depart from what you get if you just whack the sample with a hammer?

Steve Jacobsen: Yeah. There are actually two stages of compression that we're using in our pulsed-power experiments on the Z Machine. The first step is sort of like a conventional shock experiment, where an impactor traveling at many kilometers per second hits the sample to produce a shock wave. So you want to think of stage one as a shock, and that brings the pressure–temperature state in the sample up to some point on the Hugoniot. What's unique about the Z Machine is that we can control the release of current in a way that leaks out the energy. Instead of producing a single sharp pulse, which one might want to use for a fusion experiment, we can pulse it in, but then let the rest of the energy slowly creep into the sample. So what we're going to do is shock up to some state close to the core–mantle boundary condition and then what we call ramp compress the rest of the way. It's the ramp compression that's actually following a pressure–temperature curve that's very close to the Earth's adiabat. In order to do that, it comes down to the design of the target, which is the piece of metal that the sample is sitting on. It's actually the short that produces the lightning bolt that shocks everything, so the design of the target itself is what allows us to ramp compress following the shock. We call this technique shock-ramp compression.

Oliver Strimpel: And that enables the pressure to go up without the temperature shooting up beyond what we expect to prevail down there.

Steve Jacobsen: Yeah, that's right.

Oliver Strimpel: You also use a completely different approach, diamond anvil cells, where you squeeze tiny samples between diamonds under static compression. How does that compare with what you do on the Z Machine, and when would you use one technique rather than the other?

Steve Jacobsen: Yeah, the diamond anvil cell is a very versatile tool. These static compression experiments differ principally in that they are isothermal. That is to say, if I put a sample in the diamond anvil cell and I increase the pressure, the temperature remains at room temperature. And that's because the heat due to compression, unlike in adiabatic compression, does whisk away into the environment. So you might have a sample sitting at core–mantle boundary pressures in a diamond anvil cell, but it's at room temperature. Now, the advantage of the diamond anvil cell is that there are a lot of different measurements we can make during compression. For instance, we might use synchrotron X-ray diffraction to determine the structure of the material at high pressure. But the diamonds are also transparent to most of the electromagnetic spectrum, so we might use optical techniques: lasers, UV, visible light, infrared light. There are a lot of different tools with which we can perform in situ experiments on materials under static compression. Therein lies the real power of the diamond anvil cell.

Now, that's not to say that you can't heat the sample. Many of my colleagues do heat samples in diamond anvil cells, with lasers or even with resistors. You know, you can wrap the diamond anvil cell in a little tungsten wire and heat up the whole thing. Of course, you can't go to very high temperatures that way, because if you're not careful, the diamonds will burn. But what happens is that the quality and the types of physical property measurements you can make while heating a diamond anvil cell become very limited. So let's go back to the Z Machine for a minute. The advantage of that is that we can simultaneously achieve a pressure and temperature equivalent to those found at the center of the Earth while making a density measurement. And that's one of the key parameters that we want to know about materials at the conditions of planetary interiors, because what do we know about the core–mantle boundary? Well, there are really only a couple of things that we know about it, based on the propagation of seismic waves through the Earth. We know what the compressional velocity, the seismic P-wave velocity, is, what the seismic S-wave velocity is, and what the density is. Those are the three principal things that we know about the interior of the Earth from seismology. So that's what we're after: the sound velocities and the density.

Oliver Strimpel: That reminds me of a previous episode of Geology Bites I did with Barbara Romanowicz, who talked about how seismic tomography has revealed a couple of enormous regions near the core–mantle boundary in which seismic waves travel anomalously slowly, the so-called large low-shear-velocity provinces. And at their edges, there are even more extreme regions, the so-called ultra-low-velocity zones. Has your work shed light on what might be going on in those regions?

Steve Jacobsen: Not yet, but that's what we're working on. The goal of this study is to try to test a couple of different theories on why those ultra-low-velocity zones are there and what they are. I'll say a few more things about them. You mentioned already that they're slow; that is why they're called ultra-low-velocity zones. But another thing about them that I think is kind of strange is that they're very dense, and normally seismic velocities follow density in a way where lower density produces lower velocity. So this is sort of the opposite of the trend that one would expect for a material. Initially, there were a couple of theories about why that might be. Maybe partial melt, right? Because liquids, molten rock, might be quite slow. Of course, we don't know exactly what silicate melt sound velocities are at those conditions; we're working on that now. But we estimate that they're similar to melts in the Earth's crust near the Earth's surface, and that they're relatively slow compared to rock. The problem with that is that melts are also typically lower density. So unless something weird happens at very high pressures and melts become denser than rock, which might be true, that theory has probably become a little less intriguing than another one, which we're testing on Z.

And that is to modify the composition of the rock in a way that is so extreme it produces very low velocities but very high density. There's one element on the periodic table, very, very common in the Earth, that tends to do that, and that's iron. Now, iron exists in most minerals. For example, everyone might know the minerals olivine, pyroxene and garnet. These are all sort of common mantle minerals, and they all contain iron. What we have found in laboratory experiments is that when you increase the fraction of iron relative to magnesium in a silicate, the density goes up, but the shear velocities tend to go down. So the material that we're putting on the Z Machine is called magnesiowüstite, or ferropericlase, and it's just an iron magnesium oxide. What we want to do is crank up the iron content in the ferropericlase and measure the speed of sound and the density on Z as a function of iron content, to see if we can line up the physical properties of ferropericlase with what seismologists see in the ultra-low-velocity zones.

Oliver Strimpel: That's interesting. So density and seismic wave velocity usually go hand in hand, but you're suggesting that these ultra-low-velocity zones could be not melted but iron-enriched.

Steve Jacobsen: Yes. The theory that we're testing, iron enrichment, really goes back to plate tectonics itself. We know that the core is full of iron, mostly iron, and the lowermost mantle is made of iron magnesium silicates. So this boundary is very interesting. You know, are they reacting with one another? Are they in equilibrium? And another thing that could be going on at the core–mantle boundary is a piling up of subducted oceanic crust. Where the oceans form crust, it's also destroyed at subduction zones along convergent plate boundaries, and over the course of geologic time, we speculate that some of these slabs go all the way to the core–mantle boundary. We can see that in seismic tomography. Once subducted slabs reach the core–mantle boundary, what do they do? Do they just kind of pile up, or does that material get mixed and come back up in convective cycles? Well, if the basaltic piece of the subducted oceanic crust were to pile up at the large low-shear-velocity provinces, we would have a mechanism for creating iron enrichment in the ultra-low-velocity zones. Some people like to call this the slab graveyard theory. It's where old slabs maybe go to be buried. Maybe that's what we're seeing.

Oliver Strimpel: Let's move up to another extreme environment, the mantle transition zone. Can you remind us where the zone sits and what defines it?

Steve Jacobsen: The transition zone of the mantle, so called because there's a steep change in the gradient of seismic velocities, is what we consider to be the change from the upper mantle to the lower mantle. The top of the transition zone occurs at 410 kilometers depth, and the base of the transition zone occurs at around 660 kilometers depth. The reason I use such specific numbers is that it's defined by seismic discontinuities. That is to say, when seismic waves pass through these boundaries, there are strong reflections and strong changes in sound velocity, so they're probably associated with some sort of change in the mineralogy.

Oliver Strimpel: Can you tell us a bit about the minerals that we think are down there that are changing the velocity of the seismic waves?

Steve Jacobsen: We have a pretty good idea. Even though it's 400 kilometers away, that's within the range of experiments that are readily available in the laboratory. 400 kilometers to 600 kilometers depth corresponds to pressures in the range of 12 to 16 gigapascals. Remember, when we were down at the core–mantle boundary, the pressure was 130 gigapascals. Up in the transition zone, we can perform what are called large-volume experiments, which use a big hydraulic press, the same kind of press that you might see car parts being made with, I don't know. So rather than a very tiny little diamond anvil cell, picture now a 5,000-ton hydraulic press, where you're putting in not micrometer-sized samples but millimeter-sized samples. Over time, we as a field, not me, have kind of mapped, if you will, the mineralogy, the phase diagram, against what we observe seismically.

What we can tell is that olivine, the most abundant silicate in the uppermost mantle, transforms to a mineral we call wadsleyite, a modified spinel structure, at conditions that correspond exactly to 410 kilometers depth. Within the transition zone, there's another phase, another high-pressure polymorph of olivine, because wadsleyite becomes unstable at around 520 kilometers depth, and it forms a mineral we call ringwoodite, which is one of my favorite minerals. It's blue, and it has a very large water storage capacity. Below 660 kilometers depth, those polymorphs are gone, and every mineral becomes either what we call bridgmanite, which has a pyroxene composition and a perovskite structure, or what's left over, the oxide we talked about a minute ago, ferropericlase, (Mg,Fe)O. So there are really just two principal minerals in the lowermost mantle, which means the transition zone from the uppermost mantle to the lower mantle is a place where there are some pretty unusual minerals: wadsleyite, ringwoodite, majoritic garnet.

Oliver Strimpel: Have you been able to study the physical properties of these strange minerals, wadsleyite and ringwoodite, in experiments that you've done in the lab?

Steve Jacobsen: Yeah. The reason I got interested in wadsleyite is that in the late 1980s, Joe Smyth, who was a professor right here at the University of Colorado Boulder, predicted that its structure ought to incorporate large amounts of hydrogen defects. By that I mean that, although it's normally just a magnesium silicate, it's very likely that the structure would substitute hydrogen for magnesium, as defects, and the reason why has to do with electrostatic potentials. In other words, if we look down in gory detail at the structure of wadsleyite, there's an oxygen atom which is a little bit underbonded. It's kind of looking for a little bit more cation charge than it has. When he realized this, Professor Smyth suggested that hydrogen atoms might balance out that charge imbalance in the structure. And it wasn't long before people actually made the stuff in the lab, very easily, with water putting in several thousands of parts per million, or even weight percent, of H2O. Not liquid water, but the compositional ingredients of water were just going right into the material, sort of like it was acting like a sponge. So these minerals can actually effectively store large quantities of water in the lab.

So now the question is, what about the Earth? And I want to bring us back to the core–mantle boundary. Many years ago, I think it was Quentin Williams and Ed Garnero who noticed that there was a very rough geospatial correlation between the ultra-low-velocity zones and the ocean islands, that is, places like Hawaii and Iceland, where deep-sourced magmas are coming probably from the core–mantle boundary. So if the ultra-low-velocity zones are the source of deep melts in the Earth, those must pass through the transition zone. And this gets me back to why water might be important in the transition zone. Geophysical research has had the perennial problem that magmas that come from different depths have different compositions, especially with respect to isotopes. For example, mid-ocean ridge basalts come from very shallow depths, and ocean island basalts come from very deep. Now, that might not sound like a problem, but when you remember that convection is supposed to be occurring on a whole-mantle scale, how then do you produce different magma types in an Earth, or a planet, that is mixed through convection? It's a dichotomy that the fields of geophysics, geochemistry and mineral physics have been grappling with for decades and more.

Well, the transition zone might fit in very nicely, because if the transition zone were acting like a sponge for incompatible elements, as we call them, like hydrogen, it's possible that it might be stripping away some of the incompatibles as melts ascend through the entire mantle from the core–mantle boundary, almost like trapping them there, if you will. I want to associate that idea with Shun-ichiro Karato. He and David Bercovici at Yale wrote a paper some years ago that they called the transition zone water filter model. It was a very nice theory on how the transition zone might not only be acting as a large reservoir of H2O in the Earth, with implications for the origin of Earth's water, but also acting as one of the ways in which we can produce different compositions of magma in a whole-mantle-convection planet.

Oliver Strimpel: Wow, that's fascinating. So a kind of dual role for this mantle transition zone, as a kind of sponge but also as a filter.

Steve Jacobsen: Yeah. I think that you can't have oceans on a planet until the mantle is saturated. Imagine trying to put a little lake on a sponge. As you add water to the sponge, it would just get absorbed by the sponge until the sponge is saturated, and only then would you be able to put a film of liquid water on top of it. It's possible planets might act the same way, and that planets that have liquid water on their surfaces might need something like the transition zone inside them, and therefore only planets of a certain size might be able to have liquid water on their surfaces.

Oliver Strimpel: That's interesting. So does that mean that a planet really needs to have this kind of a sponge system to regulate how much water it has, or to allow water to exist in liquid form on the surface? I mean, are those two things connected at all?

Steve Jacobsen: It's a theory that people are working on from modeling perspectives and experimental perspectives. Let's look at Mars. Mars is not as big as the Earth. The Martian mantle barely gets to a pressure high enough to support ringwoodite near the core, so there is no transition zone in Mars. And going one step further, Earth's water was long supposed to have come from comets, during and after accretion. But this idea that at extreme conditions minerals can incorporate the components of water, hydrogen, leads us to think about that again. It's quite possible that planets could retain quite a lot of their original water. Now that we have more appreciation for the properties of melts and minerals at the conditions of deep planetary interiors, they seem to be quite happy holding on to water, and they might not have degassed all their hydrogen during accretion, which is what we thought previously.

Oliver Strimpel: That's interesting. It reminds me of previous episodes about the origin of the Earth's water: the meteorite origin, which Sara Russell talked about; Lindy Elkins-Tanton, who talked very much about the fact that the Earth had its water from the start; and Anat Shahar, about a way in which the Earth could make its water from a magma ocean and a hydrogen atmosphere. So it seems that there are many, many different theories as to how the Earth got its water, and understanding the mineral properties of the transition zone plays into that discussion. Just to step back to the experimental side for a moment, I'm reminded of the episode I did with David Kohlstedt, who also does high-pressure experiments. Can you say how your approach differs from his?

Steve Jacobsen: Yeah. Professor Kohlstedt is a rock mechanics expert, and he studies what we call rheology, the way solids deform plastically beyond their elastic limit. All of the deformation that I'm performing in the diamond anvil cell under static compression is elastic. When we squeeze a sample, a crystal of olivine, say, to 10 gigapascals at room temperature and bring it back, that's a fully elastic process. The propagation of seismic waves is also an elastic process, in the sense that as the wave moves through the material, there's compression and there's dilation, but those changes are completely reversible, or elastic. What David does is push rocks to their elastic limit and then study what happens thereafter. Now, that's very important in the context of mantle convection, because there are places in the uppermost mantle and crust where shear stresses are so high that there's plastic deformation going on. We can see that. You can go outside and look at metamorphic rocks and fabrics forming in very high-stress shear zones. So there's a contrast between work in rock mechanics, which deals with those rheological properties, and mineral physics, which deals with more fundamental, intensive physical and chemical properties, such as density, thermal conductivity, heat capacity and so on.

Oliver Strimpel: So far, we've been talking about the extreme pressures deep inside planets. Let's now shift to a completely different kind of extreme, the surface of the Moon. What makes that such a challenging environment for materials?

Steve Jacobsen: The challenge posed by doing geology even just on the surface of the Moon comes mainly from low temperatures and high radiation, and it's very far away. One of the goals of Artemis is to learn how to get people onto the Moon and off to Mars, and so we have to practice everything on the Moon. The first thing you need to be able to do is land a rocket in the same place multiple times. We need a base. And the regolith, the lunar soil, if you will, is just basically crushed rock. It's very harmful; it's very brittle. I mean, it's just nasty, glassy shards of minerals, so that if you were set up at a camp already and I were to arrive with a rocket a bit later and land without a landing pad, I would sandblast the camp. So we're working in the area of additive manufacturing. Developing the technology for building a landing pad or a base on the Moon requires using fake lunar soil; we can't really work with the real thing. So one of the things we're doing is trying to understand simulants. Those simulants are then melted using a laser to produce a rock. I'm actually holding one in my hand right now and looking at one of these lunar bricks, if you will. It's pretty dark, and it's kind of layered; the layers have slightly different shades. The rock has basically the composition of a mixture of basalt and anorthosite. But it's layered because the process we're using is called powder bed fusion. Imagine a sandbox of lunar regolith that you then scan with the laser and melt. Where those melt pools form, we can spread more material in and build up a solid material one layer at a time. This is how we're working towards developing 3D printing on the Moon.

Oliver Strimpel: Now, obviously the lunar surface is a totally different environment. As you said, it's a vacuum, and it's subject to extreme radiation. Does that change the material? Is it something that makes it really different from what we can simulate here on Earth?

Steve Jacobsen: It does. We call that space weathering. On Earth, when we talk about weathering, we're talking about the reaction of rocks with water and the atmosphere, and the transport of that material by gravity. On the Moon, when we talk about weathering, or space weathering, we're really talking about modifications due to radiation from the Sun. Those may include bombardment by high-energy particles, hydrogen and helium ions, for example. There is also a continuous stream of micrometeorite bombardment on the surface of the Moon. And all of these processes modify the regolith, because there is no weathering in the Earth sense. That is to say, the regolith sitting on the surface of the Moon has been there for four and a half billion years, so it's highly modified through ion bombardment and micrometeorite bombardment. The regolith simulant, the fake stuff that we're working with to 3D print, is not space weathered, and we are very concerned about what that might mean for the engineering. One of the things we see in real lunar regolith is little tiny particles, actually nanoscale particles, of iron on the surfaces of grains. This nanophase iron would certainly have a big impact on the laser heating properties in the 3D printing process. So we'd like to try to simulate space weathering in the lab. We've actually built a very large vacuum chamber, inside which we're doing this laser 3D printing. What we find, for example, is that a lot of the elements that are volatile, potassium, sodium, immediately boil off during this process. So there's almost a whole new petrology that we have to explore under vacuum in order to work with these materials on the surface of the Moon.

Oliver Strimpel: Some of the experiments you do to understand planetary interiors have created materials that have never existed before. Can you give us an example?

Steve Jacobsen: Sure. Let me just start with one that is in full-scale production. The field of superhard materials has to do with abrasives, or materials that might be able to withstand very high pressures and temperatures without degradation. We think of diamond as an extreme material, but in fact diamond burns pretty easily. Diamond will start to burn in air at around 700°C, even though it's the hardest known substance. So in the field of superhard materials, a material doesn't necessarily have to be harder; it just has to have higher thermal stability. Cubic boron nitride is a good example of a material that was discovered through high-pressure, high-temperature synthesis. It has a hardness very close to that of diamond, but it's stable up to about 1,300°C. Now, you need 5 or 6 gigapascals of pressure and 1,200°C to make it, and so commercially it's not easy, but it's at a scale where you can buy cubic boron nitride parts. The idea that we can study materials in extreme environments of pressure, temperature, radiation and so on to learn about the universe can be flipped around: we can use those extreme conditions to create new materials with functional properties. There's a whole suite of ferroelectrics, thermoelectrics, superhard materials, materials with exotic magnetic states, superconductors and so on that are being generated at these very high pressures and temperatures. So there's a whole branch of materials science that comes out of materials in extreme environments.

Oliver Strimpel: What are you most excited about working on at the moment?

Steve Jacobsen: I'm excited by the fact that we continue to make breakthrough discoveries in science. When I look back and I think about how much has changed in our knowledge about the Earth, the mantle and the composition of the mantle, and the technology we've developed, everything from materials discovery to the stuff we just talked about, 3D printing bricks or conducting compression experiments where we measure sound velocities directly at core–mantle boundary conditions, I mean, in hindsight, that would have looked like science fiction to me. And I think that's of course just going to continue. So graduate students today can think about that for a minute: what they'll be doing maybe 20 or 25 years from now might sound like science fiction to us now. That's what I get excited about.

Oliver Strimpel: Steve Jacobsen, thank you very much.

Steve Jacobsen: Thank you, Oliver. It's been really fun to talk to you.

Oliver Strimpel: To see pictures and illustrations that support this podcast, go to geologybites.com, where you'll also find a subject matter index of all the episodes. There you can also give me feedback, which I welcome, as well as sign up to get my emails about new episodes.