Earth’s core stays searingly hot because of two reinforcing heat sources: the energy left over from the planet’s violent formation roughly 4.5 billion years ago, and the steady trickle of heat produced by radioactive elements decaying deep inside the planet. Together, these keep the boundary between the core and the mantle at temperatures around 5,500 K, and the processes that might drain that heat away work extraordinarily slowly thanks to the immense insulating blanket of rock overhead. The story is richer than just “leftover heat plus radiation,” though, because the core is also actively generating new thermal energy as it freezes from the inside out.
Where the Original Heat Came From
When Earth assembled from the disk of dust and gas orbiting the young Sun, each collision between planetesimals converted kinetic energy into heat. The bigger Earth grew, the harder incoming material slammed into it, and by the late stages of accretion the impacts were energetic enough to melt vast stretches of rock. The most dramatic single event was the Moon-forming impact, a collision with a roughly Mars-sized body that almost certainly liquefied a large fraction of Earth’s surface and interior. All of that thermal energy did not simply radiate away into space. Much of it was buried beneath fresh layers of impacting debris and retained inside the growing planet.
As the young Earth heated and partially melted, dense iron-rich metal sank toward the center while lighter silicates floated upward. This gravitational separation released enormous additional energy, converting potential energy into heat as heavy material fell inward. An early analysis of this process concluded that by now, core formation is virtually complete and the energy still being released from ongoing metal settling is hundreds of times smaller than the heat generated by radioactive decay.1Geophysical Journal International. Thermal History of the Earth: I. The Formation of the Core In other words, differentiation was a one-time bonanza that loaded the core with heat early on but is not a meaningful source of new heat today.
Radioactive Decay as a Slow-Burning Furnace
The second pillar of the core’s heat budget is radioactivity. Certain isotopes of uranium, thorium, potassium, and a few other elements are scattered throughout Earth’s interior. As they decay, they convert tiny amounts of mass into energy, warming the surrounding rock and metal. Current estimates put the total radiogenic heat production of the entire Earth at about 20 terawatts.2Geochemistry, Geophysics, Geosystems. Radiogenic Power and Geoneutrino Luminosity of the Earth and Other Terrestrial Bodies Through Time Most of that heat is generated in the crust and mantle, where uranium and thorium are concentrated. But a long-running question is how much radioactive heat the core itself produces.
Potassium-40 has been the leading candidate for a significant core-hosted radiogenic source. The idea was proposed decades ago on theoretical grounds, but experiments trying to determine whether potassium actually dissolves into molten iron at core-forming pressures have yielded conflicting results.3PubMed. Experimental evidence that potassium is a substantial radioactive heat source in planetary cores If potassium is present in the core in meaningful quantities, it would add a direct internal heat source that slows the core’s cooling. If it is largely absent, the core’s heat budget leans more heavily on primordial energy and the heat released by ongoing solidification.
Thermal-history models that track how Earth’s interior temperature has changed over billions of years indicate that radioactive heat production today accounts for somewhere between about 30% and 85% of Earth’s total surface heat loss, with the most plausible range sitting around 45% to 65%.4Journal of Geophysical Research: Solid Earth. Thermal histories of convective Earth models and constraints on radiogenic heat production in the Earth The remainder comes from the planet gradually surrendering its stored primordial heat. If you took away all the radioactive isotopes tomorrow, the core would eventually cool down, but the process would stretch over billions of years because there is simply so much stored thermal energy and so little ability for that energy to escape quickly.
How Hot the Core Actually Is
Measuring the temperature at Earth’s center directly is impossible. Instead, researchers squeeze tiny samples of iron and iron-nickel alloy between diamond anvils at pressures mimicking those thousands of kilometers underground, then heat them with lasers until they melt. By pinpointing the melting temperature at each pressure, they can trace a melting curve outward to the conditions at the boundary between the inner and outer core. One set of these experiments, using synchrotron techniques sensitive to atomic motion, estimated an upper bound for the temperature at the inner-core/outer-core boundary of about 5,500 K, give or take 200 degrees.5Earth and Planetary Science Letters. Temperature of Earth’s core constrained from melting of Fe and Fe0.9Ni0.1 at high pressures That is comparable to the temperature of the Sun’s surface.
The temperature rises further toward the very center of the inner core, though by how much depends on assumptions about the core’s composition. Light elements like sulfur, oxygen, silicon, hydrogen, and carbon are almost certainly mixed into the outer core’s liquid iron, because seismology shows the outer core is less dense than pure iron would be at those pressures.6Physics of the Earth and Planetary Interiors. Light elements in the Earth’s outer core: A critical review Which elements dominate that mix remains debated, and the answer affects melting behavior and therefore the temperature profile. The core is not simply a ball of pure metal; it is a complex alloy whose exact recipe is still being worked out.
The Inner Core as a Heat Engine
Earth’s inner core is solid, and the outer core surrounding it is liquid. The boundary between them is not static. As the planet slowly cools, liquid iron at the top of the outer core loses heat to the mantle, and eventually iron at the base of the outer core crystallizes onto the growing inner core. This freezing process does two things that keep the core’s thermal engine running.
First, when iron crystallizes, it releases latent heat, the same kind of energy you would feel if you watched water freeze and held your hand near it, except on a planetary scale. That latent heat warms the surrounding liquid and helps sustain convection in the outer core. Second, the light elements that were dissolved in the liquid iron do not fit neatly into the solid crystal structure. They get expelled back into the liquid, making the fluid near the inner-core boundary lighter and more buoyant. That buoyancy drives compositional convection, fluid rising because of its chemistry rather than just its temperature.
Estimates suggest that at present, compositional convection supplies roughly 80% of the power driving the outer-core dynamo, with thermal convection contributing about 20%.7Physics of the Earth and Planetary Interiors. The strength and efficiency of thermal and compositional convection in the geodynamo Early in Earth’s history, before the inner core existed, thermal convection had to do all the work alone. That shift over time from a thermally driven dynamo to a compositionally driven one turns out to be important for understanding whether and how the magnetic field has persisted.
The Mantle as an Insulating Blanket
Almost 2,900 kilometers of silicate rock separate the core from the surface. Rock is a terrible conductor of heat compared with metal, so this mantle acts as a massive insulator. Heat does escape from the core into the mantle, but it does so slowly, primarily through convection, where hot mantle rock creeps upward over millions of years and cooler rock sinks. The net outward heat flow from the core through the mantle and eventually to the surface is only a few terawatts, a fraction of what the core stores.
Recent seismic work has identified mysterious patches right at the base of the mantle, sitting directly on top of the core, that appear to slow heat transfer even further. These structures, known as ultra-low velocity zones, have unusually low seismic velocities, high temperatures, and high densities. Modeling suggests they behave like localized thermal insulation blankets, significantly impeding the flow of heat from the core into the mantle and altering where and how fast the core loses energy.8Nature Communications. Mysterious, thermally insulating patches at the base of Earth’s mantle Think of them as hot, dense blobs of material that sit on the core-mantle boundary like a quilt, trapping heat underneath. Their patchy distribution means the core does not cool uniformly; some regions lose heat faster than others, and that unevenness could influence convection patterns in both the outer core and the lower mantle.
The New Core Paradox
The magnetic field is Earth’s most visible product of core heat. Convection in the liquid outer core generates the geodynamo, the self-sustaining electrical dynamo that produces the field. Paleomagnetic evidence shows that a global magnetic field has existed for at least 3.5 billion years, and possibly longer. Here is the problem: recent laboratory measurements suggest that iron at core conditions conducts heat much better than older estimates assumed. If the core is highly thermally conductive, heat can leak out through simple conduction without needing convection. And without vigorous convection, there is no dynamo.
This tension is called the “new core paradox.” High thermal conductivity in the core makes it much harder to sustain convective motions, and therefore dynamo action, during the long period before the inner core began to crystallize.9Journal of Geophysical Research: Solid Earth. The “New Core Paradox”: Challenges and Potential Solutions Once the inner core nucleated, the latent heat and compositional buoyancy from freezing provided a powerful boost. But if the inner core is younger than previously thought, the period without that boost was longer, and the paradox is sharper.
Some paleointensity measurements from rocks formed during the Neoproterozoic era, roughly 600 to 800 million years ago, hint at an extremely weak magnetic field at that time, as though the dynamo was on the brink of collapse right before the inner core started to form.10Geophysical Journal International. First palaeointensity data from the cryogenian and their potential implications for inner core nucleation age If the inner core nucleated only around 600 to 700 million years ago, that is dramatically younger than earlier estimates that placed it at one to two billion years old. A younger inner core implies the core was hotter for longer and cooled faster in recent geological time, reshaping our picture of Earth’s entire thermal history.
Several proposed solutions to the paradox exist. One is that the core’s thermal conductivity is not actually as high as the newest lab measurements suggest. Another is that radioactive potassium in the core provided extra heat that kept convection going. A third is that dissolved light elements like magnesium precipitated out of the cooling core and drove their own form of compositional convection, propping up the dynamo before inner-core freezing began. None of these explanations is settled, and the paradox remains one of the liveliest debates in deep-Earth science.
How Long the Inner Core Has Been Growing
The rate at which the inner core grows depends on how fast the core loses heat to the mantle. One thermal evolution model found that the inner core could have grown to its present size in about 2.8 billion years assuming a certain average heat flux from the core.11Journal of Geophysical Research: Solid Earth. On the thermal evolution of the Earth’s core But different assumptions about thermal conductivity, the core’s composition, and heat flux at the core-mantle boundary give wildly different ages. That same model showed that the balance between compositional and thermal contributions to the dynamo shifts depending on the heat flux: at higher heat fluxes, thermal convection’s share of the work rises.
The inner core is currently about 1,220 kilometers in radius, roughly the size of the Moon. It grows by perhaps a fraction of a millimeter per year, a rate so slow it is negligible on human timescales. But on geological timescales, the growth matters enormously because it is the primary mechanism topping up the energy supply to the dynamo and, by extension, the magnetic field that shields the surface from solar wind.
What Happened When Mars Lost Its Heat
Mars offers a sobering case study in what happens when a planet’s core cools too much. Widespread crustal magnetization on Mars shows that the planet once had a global magnetic field, probably generated by a dynamo in its liquid core. But that field disappeared roughly 3.6 to 4 billion years ago. Mars is smaller than Earth, so it had less primordial heat to begin with and a proportionally larger surface area through which to lose it.
Two complementary lines of evidence help explain the Martian dynamo’s death. One study found that a series of giant impacts during early Martian history heated the deep mantle unevenly, reducing the heat flow across the core-mantle boundary by 10 to 40% and potentially shutting down convection in the core.12Journal of Geophysical Research: Planets. Giant impacts on early Mars and the cessation of the Martian dynamo Another found that Mars’s iron-sulfur core has a much higher thermal conductivity than previously thought, allowing the core to cool efficiently through conduction alone and forming a thermally stratified layer that chokes off convection.13PubMed Central. A thermally conductive Martian core and implications for its dynamo cessation
Without a magnetic field, Mars lost much of its atmosphere to the solar wind, its surface water evaporated or froze, and the planet became the cold desert we see today. Earth has so far avoided that fate because its larger size, radioactive inventory, and actively freezing inner core collectively keep the dynamo running. But the Martian example is a reminder that planetary heat budgets have direct consequences for surface habitability.
Could a Rocky Exoplanet Keep Its Core Hot?
The question of what keeps a core hot extends well beyond our own planet. As astronomers discover growing numbers of rocky exoplanets, especially so-called super-Earths with masses several times that of our planet, researchers want to know which ones could sustain magnetic fields and volcanism. Both depend heavily on how hot the core stays and how efficiently it loses that heat.
Modeling suggests that massive super-Earths should have crystallizing cores over a large temperature range, meaning they are likely to benefit from the same latent-heat and compositional-buoyancy engine that powers Earth’s dynamo.14Journal of Geophysical Research: Planets. Super‐Earth Internal Structures and Initial Thermal States Bigger planets also retain more primordial heat from accretion, and their larger mantles contain more radioactive material. One recent study concluded that partitioning radiogenic heat into the core of a super-Earth significantly raises the core-mantle boundary temperature and total heat flow, making long-lived volcanism and strong magnetic dynamos likely.15PubMed Central. Radiogenic heating sustains long-lived volcanism and magnetic dynamos in super-Earths
There is a catch, though. If a super-Earth’s mantle is hot enough to stay partially molten at the base, a layer of magma ocean can form between the mantle and the core. That magma ocean could actually reduce the heat flow across the boundary and smother the dynamo, even if the core itself has plenty of thermal energy.16Journal of Geophysical Research: Planets. Energetic Requirements for Dynamos in the Metallic Cores of Super‐Earth and Super‐Venus Exoplanets So size alone does not guarantee a magnetic field. The interplay between core heat, mantle state, and the boundary between them determines the outcome, just as it does on Earth.
Why the Core Will Stay Hot for Billions of Years
Earth’s core is not cooling quickly by any human standard. The planet is roughly 4.5 billion years old, and the inner core may have started forming only in the last billion years or so, meaning it took the majority of Earth’s lifetime for the core to cool enough for iron to begin freezing at the center. Radioactive isotopes with half-lives measured in billions of years will continue producing heat for a comparable stretch into the future. The mantle remains a formidable insulator. And each increment of inner-core growth releases latent heat and compositional energy that partially offset the cooling.
The practical upshot is that the core’s heat engine, and the magnetic field it generates, should persist for billions of years to come. Long before the core solidifies completely, the Sun will have exhausted its own hydrogen fuel and expanded into a red giant, rendering the question moot for life on Earth’s surface. In the meantime, the planet’s deep interior continues to churn away as it has since before the first microbes appeared: a slow, self-regulating heat engine powered by the echoes of planetary formation and the quiet disintegration of ancient atoms.