What Is the Freezing Point of Iron?

Iron freezes at 1538 °C (2800 °F), or 1811 K on the Kelvin scale, under normal atmospheric pressure. That single number is what you will find in most reference tables, and for everyday metallurgy it is perfectly reliable. But iron’s freezing behavior gets far more interesting once you move beyond surface conditions, because pressure, impurities, and even particle size can shift that temperature by hundreds or thousands of degrees.

Why Iron Freezes at Such a High Temperature

Iron sits in the middle of the transition metals on the periodic table, and its atoms are bound together by strong metallic bonds. In a chunk of solid iron at room temperature, the atoms are arranged in a body-centered cubic crystal lattice, meaning each atom sits at the center of a cube formed by eight neighbors. Breaking those bonds and letting the atoms flow freely as a liquid requires a lot of thermal energy, which is why you need to reach roughly 1538 °C before the metal actually melts or, equivalently, why liquid iron must cool to that same temperature before it solidifies.

The freezing point and the melting point of a pure substance are the same temperature. The difference is just the direction you are traveling: heating a solid past 1538 °C melts it, and cooling a liquid to 1538 °C freezes it. In practice, liquid iron can sometimes be “undercooled” well below that threshold without freezing, a phenomenon that matters in both research and industrial casting, but the equilibrium transition temperature remains the same.

Undercooling and How Iron Actually Solidifies

When molten iron cools in a furnace or a mold, it does not always freeze the instant it reaches 1538 °C. Under carefully controlled laboratory conditions, researchers have pushed liquid iron hundreds of degrees below its equilibrium freezing point without triggering solidification. The liquid stays in a metastable state until something, often a tiny impurity or a vibration, nudges a cluster of atoms into forming a crystal nucleus. Once that nucleus appears, solidification can be explosive. In experiments on undercooled pure iron, dendrites (the branching, tree-like crystals that form during freezing) have been clocked growing at speeds up to 69 meters per second at an undercooling of about 280 K below the melting point.

1Acta Materialia. A comparative study of dendritic growth within undercooled liquid pure Fe and Fe50Cu50 alloy

At small amounts of undercooling, these dendrites are beautifully well-defined. At larger undercoolings, the microstructure changes: instead of obvious branching crystals, you see a network of grain boundaries with no clear dendritic shape. The degree of undercooling matters enormously for the final grain structure of a casting, which is one reason steelmakers control cooling rates so carefully.

What Happens Under Extreme Pressure

The 1538 °C figure applies at the pressures we experience on Earth’s surface, about one atmosphere. Deep inside the planet, pressure climbs to millions of atmospheres, and iron’s freezing point rises dramatically. This is not just an academic curiosity. Earth’s core is mostly iron, and knowing exactly when it freezes at those pressures tells geophysicists about the temperature profile of the planet, the age and growth rate of the solid inner core, and the convective currents that generate our magnetic field.

One landmark set of experiments used a laser-heated diamond-anvil cell to squeeze iron up to about 150 gigapascals, roughly 1.5 million times atmospheric pressure. Those results indicated that iron melts at around 6350 K (give or take 350 K) at the pressure found at Earth’s center, about 363 gigapascals.

2PubMed. Temperatures in Earth’s Core Based on Melting and Phase Transformation Experiments on Iron

A separate approach, using shock-wave experiments that compress iron with a gas gun, placed the melting temperature at the inner-core boundary at roughly 5950 K, and at the core-mantle boundary at about 4300 K.

3Geophysical Research Letters. Shock Melting Curve of Iron: A Consensus on the Temperature at the Earth’s Inner Core Boundary

Computer simulations using quantum-mechanical calculations have pushed the estimate higher still, suggesting about 6700 K at the inner-core boundary pressure.

4Nature. The melting curve of iron at the pressures of the Earth’s core from ab initio calculations

The spread among these results, from roughly 5950 K to 6700 K at inner-core pressures, is not a sign that something is wrong with the science. Measuring temperature inside a diamond anvil at millions of atmospheres is extraordinarily difficult, and shock-wave experiments are over in microseconds. The fact that different techniques agree within about 15 percent is actually remarkable given the conditions involved.

Why the Numbers Still Do Not Quite Agree

A persistent tension in high-pressure physics is that shock-wave experiments tend to yield higher melting temperatures for iron than static diamond-anvil-cell experiments at overlapping pressure ranges. Researchers have debated this for decades.

5AIP Conference Proceedings. Metastability in shocked iron: Controversy with regard to sound velocity and temperature measurements

Part of the problem is that the two methods probe iron in very different ways. In a diamond-anvil cell, iron is squeezed slowly between two diamond tips and heated by a laser beam; scientists measure whether the sample has melted by looking for changes in its X-ray diffraction pattern. In a shock experiment, a projectile slams into an iron target at kilometers per second, and temperature is inferred from the glow of the sample captured by fast optical sensors. Each technique has its own systematic uncertainties, and iron can exist in metastable states during the violent compression of a shock wave, complicating the interpretation.

Computer modeling has tried to break the tie. One study using quantum-mechanical molecular dynamics found a melting curve that agreed with the lower static-experiment temperatures at moderate pressures but swung toward the higher shock-wave temperatures at the extreme pressures relevant to Earth’s core.

6PubMed. Quasi-Ab initio molecular dynamic study of Fe melting Even so, the question is not fully settled. The crystal structure of iron at core conditions adds another variable: calculations show that a body-centered cubic phase of iron, which some models suggest could exist in the inner core, is dynamically stable at those extreme temperatures and pressures.

7PubMed Central. Dynamically stability of body center cubic iron at the Earth’s core conditions

How Impurities Lower Iron’s Freezing Point

Pure iron freezes at 1538 °C, but the iron in Earth’s core is not pure. Seismological data show the outer core is less dense than pure liquid iron at equivalent pressures, which means lighter elements such as sulfur, silicon, oxygen, carbon, and hydrogen are dissolved in it. Those dissolved elements lower the freezing point, sometimes by a lot.

Recent work modeling iron alloys at inner-core-boundary pressures found that even small concentrations of light elements cause substantial melting-point depression. Starting from a baseline of about 6230 K for pure iron at those pressures, adding hydrogen, carbon, oxygen, or sulfur to about 15 atomic percent dropped the melting temperature by roughly 450 to 1000 K depending on the element. Carbon was the strongest depressant, pulling the melting point down to around 5230 K, while sulfur had a milder effect at about 5780 K. Silicon was the oddball: a small amount lowered the melting point slightly, but higher concentrations actually raised it above that of pure iron.

8ESS Open Archive. Effect of Light Elements on the Melting Behavior of Liquid Iron Alloys under Earth’s Inner Core Boundary Conditions

Experimental work on binary iron systems at more moderate pressures has also quantified this effect, finding depressions on the order of 30 K per weight-percent for silicon, 50 K per weight-percent for oxygen, and 100 K per weight-percent for sulfur.

9Comptes Rendus Geoscience. Properties of iron alloys under the Earth’s core conditions – Section: 2. Melting temperature depression

This matters because the exact composition of the outer core sets the temperature at which the inner core’s surface is crystallizing right now. Get the impurity effect wrong, and your estimate of Earth’s core temperature shifts by hundreds of degrees. It also affects models of how fast the inner core is growing, which in turn affects estimates of how much longer Earth’s magnetic field will be sustained by the convection driven by that solidification.

Iron Snow in Smaller Worlds

Earth’s inner core grows from the center outward: iron crystals form at the inner-core boundary and the solid sphere gradually expands. But in smaller planets and moons with lower core pressures, the physics can flip upside down. If the melting curve of iron or an iron alloy slopes the right way with depth, iron crystals can form near the top of the core, at the core-mantle boundary, and then sink like snowflakes through the surrounding liquid.

10Geophysical Research Letters. A Laboratory Model for Iron Snow in Planetary Cores

This “iron snow” regime has been proposed for Mercury, Ganymede, and Mars. For Mars, the argument is motivated by experiments on the iron-sulfur system, which show that the liquidus temperature (the temperature at which crystals first appear on cooling) increases with decreasing pressure at Martian core conditions. That means the coolest part of the core, near the top, is also the part most prone to crystallization.

11Earth and Planetary Science Letters. Iron snow in the Martian core?

Iron snowflakes falling through a liquid iron-alloy core would remelt at greater depth where conditions are hotter, releasing latent heat and pure iron into the liquid. The compositional and thermal convection this drives could, in principle, generate a magnetic field, even in a body where the core is cooling and partially solidifying from the top rather than the bottom. Whether Mars once had such a dynamo, and whether its extinction is linked to the end of iron-snow convection, is still an active area of planetary science.

Nanoscale Iron Melts at Lower Temperatures

At the other end of the size spectrum from planetary cores, the freezing point of iron shifts downward when you are dealing with particles only a few nanometers across. This is a general phenomenon for metals: as particles get smaller, a larger fraction of their atoms sit on the surface, where they have fewer neighbors and are less tightly bound. The result is a lower melting point.

Molecular dynamics simulations of iron nanoparticles have explored this in detail. Using reactive force-field models, researchers computed the melting points of iron nanoparticles of various sizes and found that the melting point does not decrease linearly with particle size. Instead, the relationship is more complex, and the presence of defects in the crystal structure of the nanoparticle further affects where the transition occurs.

12Scientific Reports. Molecular Dynamics Simulations of Melting Iron Nanoparticles with/without Defects Using a Reaxff Reactive Force Field

For bulk iron, the same simulations predicted melting points of roughly 1747 K for the body-centered cubic crystal and 1726 K for the face-centered cubic crystal, both somewhat above the experimental value, which is a known limitation of that computational method.

12Scientific Reports. Molecular Dynamics Simulations of Melting Iron Nanoparticles with/without Defects Using a Reaxff Reactive Force Field The significance is not the absolute numbers but the trend: shrink a piece of iron from bulk scale to a few thousand atoms and its melting point can drop noticeably. This has practical implications for nanotechnology, catalysis, and additive manufacturing, where iron-based particles in the nanometer range are routinely used and their thermal stability is a design constraint.

Iron in the Early Solar System

Before there were planets, the solar system was a disk of gas and dust surrounding the young Sun. As that disk cooled, different materials condensed out of the gas at different temperatures. Iron-nickel alloy was among the earliest solids to appear, condensing at higher temperatures than the magnesium silicates that eventually became the rocky mantles of the terrestrial planets.

13Geochimica et Cosmochimica Acta. Condensation in the primitive solar nebula

This temperature ordering is significant. Because iron condensed first, it was available as a solid before silicate minerals formed. In models where the Earth accumulated material in roughly the order it condensed, this sequence supports the idea that the iron-rich core and the silicate-rich mantle built up as distinct layers from the start, rather than iron sinking through a fully mixed planet. The temperature gap between iron condensation and silicate condensation grows with increasing pressure, meaning the effect was strongest near the center of the early solar nebula where temperatures and pressures were highest.

Direct evidence of this condensation process has been found in primitive meteorites. Some iron-nickel grains in CH carbonaceous chondrites show chemical zoning patterns that imply they formed by condensation from the nebular gas at temperatures between roughly 1370 and 1270 K, cooling at a rate of about 0.2 K per hour.

14PubMed. Large-scale thermal events in the solar nebula: evidence from Fe,Ni metal grains in primitive meteorites Those grains are essentially frozen snapshots of conditions in the solar nebula more than four billion years ago, and their existence depends entirely on the thermodynamic properties, including the freezing point, of iron alloys at low pressures.

How Core Crystallization Leaves an Isotopic Fingerprint

When iron freezes inside a planetary core, the solid and liquid phases do not grab iron atoms in exactly the same proportions of light and heavy isotopes. Experiments simulating core crystallization at 1300 °C found that the solid metal becomes enriched in heavier iron isotopes by about 0.13 per mil relative to the remaining liquid.

15PubMed Central. Heavy iron isotope composition of iron meteorites explained by core crystallization

That tiny difference, measured as a shift in the ratio of iron-57 to iron-54, accumulates as crystallization proceeds. Iron meteorites, which are fragments of the metallic cores of small bodies that broke apart in collisions, preserve this isotopic signature. By measuring the isotope ratios across a collection of iron meteorites, researchers can reconstruct how far along crystallization had progressed in the parent body’s core before it was destroyed. The freezing point of iron under the relevant pressure and compositional conditions determines when and where that crystallization began.

First-principles calculations have extended this work to the iron-sulfur system, showing that the equilibrium isotope fractionation factors calculated for solid iron alloys serve as reasonable proxies for what happens in molten systems at the extreme pressures and temperatures of core formation.

16Earth and Planetary Science Letters. First-principles investigation of equilibrium iron isotope fractionation in Fe1−xSx alloys at Earth’s core formation conditions In other words, the isotopic bookkeeping that started when iron first froze inside these ancient bodies is legible today in the meteorites that land on Earth, and it gives us a backdoor into the thermal histories of worlds that no longer exist.

Iron Alloys in Everyday Life

Most people encounter iron not as a pure element but as steel, which is iron alloyed primarily with carbon and often with other elements like chromium, nickel, manganese, or silicon. Adding carbon lowers the freezing point of the alloy compared to pure iron, and the more carbon you add (up to a point), the lower it goes. A typical low-carbon steel might begin to solidify around 1530 °C, only slightly below pure iron. A cast iron with roughly four percent carbon starts solidifying closer to 1150 °C. The depression is governed by the same thermodynamic principles that operate in Earth’s core, just at atmospheric pressure and with carbon as the main impurity rather than sulfur or oxygen.

The temperature at which a steel fully solidifies also affects its microstructure and, therefore, its mechanical properties. Faster cooling through the freezing range tends to produce finer grain structures, while slower cooling allows larger crystals to grow. Foundry workers and steelmakers have been manipulating these variables for centuries, long before anyone understood the underlying physics. In that sense, humanity’s practical knowledge of iron’s freezing behavior predates its scientific understanding by millennia. The 1538 °C number is the anchor point around which all of that empirical knowledge was eventually organized.