Earth’s outer core is a roughly 2,260-kilometer-thick shell of liquid metal sitting between the solid inner core and the rocky mantle, and the single most striking fact about it is that its churning, molten iron generates the magnetic field that shields life on the surface from solar radiation. That liquid state is what makes everything else about the outer core interesting: it allows convection currents that act as a planetary dynamo, it creates a density puzzle that scientists are still working to solve, and it drives interactions with both the inner core below and the mantle above that shape the planet’s thermal and magnetic evolution. The outer core is one of the most inaccessible places on Earth, yet we know a surprising amount about it from seismic waves, lab experiments, and computer simulations.
A Sea of Liquid Iron and Nickel
The outer core is made primarily of iron alloyed with roughly 5 to 10 percent nickel by weight.1Scientific Reports. Experimental constraints on light elements in the Earth’s outer core That may sound simple, but identifying the exact recipe has been one of the longest-running detective stories in geophysics. Nobody can drill down to the outer core; it begins about 2,900 kilometers beneath your feet, where pressures exceed a million times atmospheric pressure and temperatures climb to roughly 4,000–5,000 °C near its top. Instead, scientists rely on seismic waves from earthquakes, which change speed and direction when they pass through different materials. Shear waves, which travel only through solids, cannot pass through the outer core at all, which is how geophysicists first confirmed it was liquid more than a century ago.
Researchers also study the outer core’s velocity structure by comparing the arrival times and amplitudes of different types of seismic waves that travel along overlapping paths through the core.2Journal of Geophysical Research: Solid Earth. Seismic velocity structure in the Earth’s outer core Those measurements, combined with high-pressure laboratory experiments on iron alloys, have built up a detailed picture of the outer core’s density and sound speed at various depths. The picture is detailed enough that scientists can now identify where the outer core deviates from what pure molten iron would look like under the same conditions, and those deviations point to a major unresolved question.
The Density Deficit and the Hunt for Light Elements
If the outer core were made of pure liquid iron, it would be about 8 percent denser than seismic observations say it actually is.3PubMed Central. Experimental constraints on light elements in the Earth’s outer core That gap is too large to explain away with measurement error. Something lighter than iron has to be mixed in, and working out what that something is remains one of the biggest open questions in deep-Earth science. The leading candidates are sulfur, silicon, oxygen, carbon, and hydrogen, often referred to collectively as the core’s “light elements.”4PubMed Central. A seismologically consistent compositional model of Earth’s core
Each candidate behaves differently under extreme pressure. Oxygen, for example, can account for the density deficit if present at around 5 to 7 percent by weight, but experiments show that no amount of oxygen alone produces sound speeds high enough to match what seismologists actually measure in the outer core.5Journal of Geophysical Research: Solid Earth. The Density of Liquid FeO and the Oxygen Contents of the Cores of Earth and Mars That finding effectively rules out oxygen as the dominant light element, though it could still be present alongside silicon or sulfur. The current best guess among many researchers is that the outer core contains a cocktail of several light elements rather than just one, making it a complex multicomponent alloy.6Journal of Geophysical Research: Solid Earth. A Density Model for Multicomponent Iron‐Rich Alloys Under Earth’s Outer Core Conditions
Why does the exact mix matter? Because the identity and proportion of light elements control how heat moves through the core, how fast the inner core grows, and how vigorous the convection currents are that generate our magnetic field. Getting the composition wrong means getting the planet’s thermal history wrong.
How Liquid Metal Generates a Magnetic Field
The outer core is the engine room of Earth’s magnetic field, a process geophysicists call the geodynamo. Electrically conducting liquid iron is in constant motion, driven by a combination of heat escaping the core, the solidification of iron onto the inner core’s surface (which releases lighter elements and latent heat), and the planet’s rotation. Those flows stretch and twist magnetic field lines, sustaining the dipole field that compasses respond to at the surface.
Getting a computer model to reproduce the real magnetic field has been a long-standing challenge. Simulations of the outer core have only recently reached the point where they can match magnetic-field patterns recorded in ancient rocks, which is a major step forward in validating our understanding of core dynamics.7Eos. How Geodynamo Models Churn the Outer Core The difficulty is partly computational: the outer core’s fluid motions span scales from thousands of kilometers down to centimeters, and modeling all of those scales simultaneously pushes even modern supercomputers to their limits.
A key constraint on the geodynamo is how much heat the core can dump into the mantle. Current estimates suggest the core needs to lose more than about 10 terawatts across the core-mantle boundary to sustain the kind of convection that drives the observed magnetic field.8Physics of the Earth and Planetary Interiors. The high conductivity of iron and thermal evolution of the Earth’s core If heat flow drops below a certain threshold, thermal convection weakens, and the dynamo has to rely more heavily on chemical convection, the buoyancy released when light elements are expelled from the growing inner core.
Thermal Conductivity Is Much Higher Than Scientists Expected
For decades, geophysicists worked with relatively low estimates of the outer core’s thermal conductivity, around 30 to 50 watts per meter per kelvin. Starting around 2012, a series of studies based on new experimental techniques and advanced calculations revised those numbers dramatically upward. One influential study found that both thermal and electrical conductivity of iron at core conditions are two to three times higher than the old estimates, pushing thermal conductivity well above 90 W/m/K at the top of the core.9PubMed. Thermal and electrical conductivity of iron at Earth’s core conditions Later work on iron-silicon alloys, which better approximate the real core composition, found values of roughly 100 to 110 W/m/K near the topmost outer core.10PubMed Central. Thermal conductivity of Fe-Si alloys and thermal stratification in Earth’s core
This revision sent ripples through the field. Higher conductivity means heat moves through the core metal more efficiently by conduction alone, which makes it harder for convection to get started. It also implies the core has been cooling faster than previously thought, which in turn suggests the inner core is younger than the roughly 3 to 4 billion years many older models assumed. Some analyses now place the inner core’s age at under a billion years, a startlingly recent feature for a planet that is 4.5 billion years old.8Physics of the Earth and Planetary Interiors. The high conductivity of iron and thermal evolution of the Earth’s core That raises a puzzle: the geological record suggests a strong magnetic field going back over 3 billion years, so the dynamo had to be running long before the inner core existed. Heat-driven convection alone must have powered it initially, and the onset of inner core crystallization later gave the dynamo an extra energy boost.
A Stratified Layer at the Top of the Outer Core
One of the more surprising findings in recent years is that the outermost few hundred kilometers of the outer core may not be convecting at all. If the thermal conductivity is as high as newer estimates suggest, the heat flowing out of the core at the core-mantle boundary could be less than what conduction alone can carry. When that happens, the top of the core becomes “thermally stratified,” meaning the temperature gradient is too gentle to drive upwelling, and the liquid sits in stable layers rather than churning.
Research on iron-silicon alloys at core conditions suggests this stratified layer could be 400 to 500 kilometers deep if the core contains a large proportion of silicon.10PubMed Central. Thermal conductivity of Fe-Si alloys and thermal stratification in Earth’s core Seismological observations support the existence of some kind of anomalous layer at the top of the outer core: both body-wave travel times and normal-mode measurements show that seismic velocities in the outermost core are lower than standard reference models predict, with a steeper velocity gradient than expected.11Geophysical Journal International. The signal of outermost-core stratification in body-wave and normal-mode data That pattern is consistent with a layer whose composition or temperature profile differs from the well-mixed bulk of the outer core below it.
If a thick stratified layer really exists, it has implications for how efficiently the core’s magnetic field can reach the surface and for how chemical elements exchange between the core and the mantle. It would also mean that the vigorous convective dynamo operates in a somewhat smaller volume than the full outer core.
The Core-Mantle Boundary
The boundary between the outer core and the mantle above it, roughly 2,900 kilometers down, is one of the most extreme interfaces anywhere in the solar system. On one side sits solid rock; on the other, liquid iron alloy. Temperature, density, and chemical composition all change sharply across it. This boundary is not a passive dividing line. Vigorous thermal and chemical interactions take place there, shaped by a complex layer of rock on the mantle side known as the D″ (D-double-prime) region, which facilitates heat transfer and chemical exchange between the two realms.12PubMed Central. The emerging picture of a complex core-mantle boundary
Seismologists have found that the D″ layer is far from uniform. It contains patches of ultra-low-velocity zones where seismic waves slow dramatically, suggesting pockets of partial melt or iron-enriched material. Some researchers think these patches are places where core material has chemically reacted with the base of the mantle, while others suggest they are remnants of ancient subducted ocean floor that sank all the way to the bottom of the mantle. Either way, the core-mantle boundary is an active zone of exchange, not just a fence between two very different materials.
Heat flow across this boundary is the ultimate throttle on the geodynamo. As noted earlier, new conductivity estimates imply the core needs to export more than 10 terawatts through this interface to sustain magnetic-field generation. At the same time, the mantle above controls how much heat it can accept, because rock convects far more slowly than liquid iron. The result is a long-running feedback loop: the mantle’s sluggish overturn limits how fast the core can cool, which in turn controls how vigorously the outer core convects and how strong the magnetic field is at any given epoch.
Geomagnetic Reversals Start in the Outer Core
Earth’s magnetic field has flipped hundreds of times over geological history, with north and south magnetic poles swapping places. These polarity reversals are recorded in volcanic rocks and ocean-floor basalts, and they are ultimately driven by changes in the flow patterns of the outer core. The reversals are not periodic or predictable; they happen at irregular intervals ranging from tens of thousands to tens of millions of years.
There is no single accepted model for why reversals happen, but recent work frames them as a natural product of the turbulent, chaotic flow inside the core rather than a sign that something dramatic has gone wrong. One approach treats reversals and excursions (partial flips that do not fully complete) as outcomes of chaotic turbulence, in which no major rearrangement of the large-scale flow structures is needed for a reversal to occur.13Journal of Geophysical Research: Solid Earth. Geomagnetic Reversals and Excursions as an Outcome of Non‐Equilibrium Wave‐Turbulence and Beating MAC Waves in the Core An older but still influential model proposes that the outer core’s two main energy sources, heat loss at the core-mantle boundary and energy released by inner-core solidification, generate opposing patterns of fluid helicity in each hemisphere. Random fluctuations in the balance between these two sources can occasionally cause the net helicity to flip sign, triggering a reversal.14Physics of the Earth and Planetary Interiors. Geomagnetic polarity reversals in a turbulent core
In practical terms, a reversal takes a few thousand years to complete, during which the field weakens and becomes more complex, with multiple magnetic poles appearing at the surface. Life on Earth has survived every reversal so far, though the weakened field during a transition would allow more cosmic radiation to reach the surface.
How the Outer Core Flows
Calling the outer core “liquid” might conjure an image of something sloshing around like water in a bowl, but the reality is more nuanced. The liquid iron in the outer core has a viscosity, meaning a resistance to flow, that varies depending on where you look and how you measure it. Molecular-dynamics simulations find that, at the extreme pressures and temperatures of the lowermost outer core near the inner-core boundary, the zero-frequency viscosity could be as high as about a billion pascal-seconds.15Journal of Geophysical Research: Solid Earth. Viscoelasticity of Liquid Iron at Conditions of the Earth’s Outer Core That is far thicker than liquid iron at the surface, which flows about as easily as water. The lowermost outer core, by contrast, has a viscosity closer to that of thick honey or warm glass, at least on certain timescales.
This matters because viscosity affects how tightly the planet’s rotation can organize the flow into columns and jets, which in turn shapes the magnetic field the dynamo produces. Earth’s rotation exerts a powerful organizing influence on the outer core, tending to align convective flows into north-south columns parallel to the rotation axis. But where viscosity is higher or where stratification suppresses convection, that tidy columnar flow can break down into more complex patterns.
What the Outer Core Tells Us About Other Planets
Earth is not the only planet with a metallic core, and comparing cores across the solar system helps clarify what makes ours special. Mars, for instance, has a core that is proportionally larger relative to its size but appears to contain a higher fraction of lighter elements, possibly including a lot more oxygen than Earth’s core. Interestingly, the same experiments that ruled out oxygen as the dominant light element in Earth’s outer core found that at Martian core pressures, iron-oxygen mixtures behave differently: the non-ideal mixing effects are larger, causing oxygen to raise the sound speed of liquid iron rather than leaving it too low, as happens at Earth’s higher pressures.5Journal of Geophysical Research: Solid Earth. The Density of Liquid FeO and the Oxygen Contents of the Cores of Earth and Mars
Mars does not currently have a global magnetic field, which tells us something went wrong with its dynamo. The leading explanation is that Mars’s core cooled enough to shut down vigorous convection, either because it lost heat too efficiently early on or because its inner core, if it ever formed, did not grow fast enough to supply the chemical buoyancy needed to sustain a dynamo. Earth’s outer core, by contrast, is still losing heat and still growing its inner core, keeping the dynamo alive. That ongoing activity is not guaranteed to last forever, but based on thermal models, the outer core should remain liquid and convecting for billions of years yet, meaning the magnetic shield will persist well beyond the timescales that matter for life on the surface.
Venus presents a different puzzle: it is nearly the same size as Earth and presumably has a similar iron core, yet it too lacks a global magnetic field. One hypothesis is that Venus’s mantle does not convect efficiently enough to draw heat out of the core, leaving the core too hot and too uniformly heated to convect. The outer core’s behavior, in other words, depends not just on its own properties but on the planet around it. A liquid iron core is a necessary but not sufficient condition for a dynamo; you also need the right thermal gradient and the right pattern of energy release to keep the fluid stirring.