Every permanent magnet has a temperature at which it completely loses its magnetism, known as its Curie temperature. For the strong neodymium magnets stuck to your refrigerator, that point is around 310 °C (590 °F). For a plain iron magnet, it is roughly 770 °C (1,418 °F). But the practical answer is more nuanced than a single number, because magnets start losing useful strength long before they hit that threshold, and the type of magnet, its size, and even the pressure it is under all shift the temperature at which trouble begins.
What Happens Inside a Magnet as It Heats Up
A permanent magnet works because billions of atomic magnetic moments inside the material are aligned in the same direction. At low temperatures, these moments hold their alignment against small disturbances. As temperature climbs, thermal energy rattles those moments more and more aggressively, nudging some out of line. The magnet’s net field weakens gradually with each degree gained. This is not an on/off switch; it is a slow fade that accelerates as the temperature rises.
At the Curie temperature, thermal energy finally overwhelms the forces holding the moments in alignment. The material transitions from a ferromagnetic state, where moments cooperate, to a paramagnetic state, where they point in random directions and produce no net field. Cool the material back down, and the ferromagnetic phase returns, but the aligned domain structure that made it a useful magnet does not automatically reappear. You would need to re-magnetize it with a strong external field.
Common Magnet Types and Their Temperature Limits
Different magnet materials have wildly different Curie temperatures, and their practical operating ceilings are always lower. Here are the most widely used types:
- Neodymium (NdFeB): Curie temperature around 310–340 °C, but most grades are rated for a maximum operating temperature of only 80–200 °C depending on their specific composition. These are the strongest permanent magnets commercially available, but also the most heat-sensitive.
- Samarium cobalt (SmCo): Curie temperature around 700–800 °C, with operating limits typically in the 250–350 °C range. Favored in aerospace and military applications precisely because they tolerate heat far better than neodymium.
- Ferrite (ceramic): Curie temperature roughly 450 °C, operating range up to about 250–300 °C. Weaker than rare-earth magnets but cheap and corrosion-resistant, which is why they show up in speakers and basic motors.
- Alnico (aluminum-nickel-cobalt): Curie temperature around 800–860 °C, with very high operating limits. Alnico magnets handle heat well but are easily demagnetized by opposing fields, so they’ve been largely replaced by rare-earth types in applications where field strength matters.
Pure elemental magnets illustrate the range even more starkly. Nickel loses its magnetism at about 358 °C, iron at 770 °C, and cobalt holds out to roughly 1,115 °C. Researchers studying dozens of ferro- and ferrimagnetic materials have developed computational models that can predict these Curie temperatures from a material’s electronic structure, typically getting within about 126 K of the measured value.1American Physical Society (Physical Review Materials). Predicting the Curie temperature of magnetic materials with automated calculations across chemistries and structures The wide spread across materials is why choosing the right magnet for a hot environment matters so much.
Why Magnets Fail Well Before the Curie Point
If neodymium magnets have a Curie temperature above 300 °C, you might wonder why manufacturers often rate standard grades for only 80 °C. The gap exists because permanent magnets don’t just need to be ferromagnetic; they need to resist demagnetization under the specific conditions of their application. Two properties drive this: remanence (how strong the field is after the external magnetizing field is removed) and coercivity (how well the magnet resists losing that field when exposed to opposing forces or heat).
As temperature rises, both remanence and coercivity decline. Long before the Curie temperature, coercivity can drop enough that the magnet’s own internal demagnetizing field, or a modest external opposing field from the device it sits in, starts flipping magnetic domains. Once domains flip, the magnet permanently loses some of its strength. This is irreversible demagnetization, and it can begin at temperatures well within the range a motor or sensor might encounter during normal operation.
That is why the maximum operating temperature specification on a datasheet is always lower, sometimes dramatically so, than the Curie temperature. The operating limit is the temperature below which the magnet can be heated and cooled repeatedly without permanent loss. Go above it and some fraction of the magnetization does not come back when the magnet cools.
Reversible Versus Irreversible Loss
This distinction is the single most practical thing to understand about magnets and heat. Reversible losses are the normal, gradual weakening that happens as any magnet warms up. Pull a neodymium magnet off a hot engine, let it cool, and it returns to full strength. No harm done.
Irreversible losses happen when the magnet crosses a critical thermal threshold. At that point, enough domains flip that the magnet follows a different magnetic “path” when it cools, ending up weaker than it started. Research on permanent magnets under stress confirms that there is a definable critical temperature beyond which remanence drops along a new trajectory, meaning the original magnetization cannot be recovered simply by cooling.2Defence Technology. Irreversible demagnetization mechanism of permanent magnets during electromagnetic buffering Below that critical temperature, the magnet bounces back. Above it, re-magnetization with an external field is the only fix.
In real applications, engineers build in a thermal safety margin. If the worst-case temperature inside an electric vehicle motor is 150 °C, the designer picks a neodymium grade rated to at least 180 or 200 °C, not one rated to 80 °C. This margin accounts for hot spots, transient spikes, and the fact that an irreversibly weakened magnet in a sealed motor is expensive to replace.
Can You Restore a Heat-Damaged Magnet?
If the magnet never exceeded the Curie temperature, the answer is usually yes, but not by just cooling it down. The material is still ferromagnetic; it simply lost its organized domain alignment. Placing it inside a strong enough magnetizing field will re-align the domains and restore most or all of the original performance. Industrial magnetizers used in manufacturing do exactly this.
If the magnet was heated above the Curie temperature and its microstructure survived (no melting, no oxidation, no grain-boundary damage), re-magnetization still works in principle. But extreme heat can cause chemical and structural changes, particularly in neodymium magnets, whose rare-earth-rich grain boundaries are reactive at high temperatures. Heat treatment studies on NdFeB alloys show that carefully controlled annealing can dramatically improve magnetic properties, but these processes require precise temperature control and rapid heating and cooling rates.3European Journal of Science and Technology. Effects of Heat Treatment and on Magnetic Properties of NdFeB Based Permanent Magnet Alloys A magnet that was simply overheated in service without any such control is unlikely to return to its original performance without being reprocessed from scratch.
What Extreme Cold Does to Magnets
Heat gets all the attention, but cold can cause problems too, especially for neodymium magnets. As temperature drops, coercivity generally increases, which sounds like good news. And it is, down to a point. In NdFeB-based ribbons, coercivity climbs steadily from about 14 kOe at room temperature to 24 kOe at 150 K (-123 °C).4Journal of Applied Physics. Spin reorientation in melt-spun neodymium-iron-boron ribbons
Below about 135 K (-138 °C), though, something unexpected happens. The crystal structure of the neodymium-iron-boron phase undergoes a so-called spin reorientation: the preferred magnetization direction shifts from a single easy axis to an easy cone. The demagnetization curve, which describes how well the magnet holds its field under opposing stress, develops a two-step shape with a large drop in magnetization at low reverse fields. The result is that the magnet’s effective resistance to demagnetization actually decreases at very low temperatures, even though the intrinsic coercivity number remains high.4Journal of Applied Physics. Spin reorientation in melt-spun neodymium-iron-boron ribbons For most everyday uses this does not matter, since few environments outside cryogenic labs or space reach those temperatures. But for engineers designing satellite components or superconducting-magnet assemblies, it is a real concern.
Samarium cobalt and ferrite magnets handle cold differently. SmCo magnets are largely stable across a huge temperature range, which is one reason they dominate in aerospace. Ferrite magnets actually become harder to demagnetize as they warm up slightly from room temperature, but their coercivity drops in cold, making them more vulnerable to demagnetization in winter conditions if they are operating near their limits.
Why Tiny Magnets Are More Vulnerable to Heat
Shrink a magnet down to nanometer scale and the rules change. A nanoparticle so small that it consists of a single magnetic domain does not demagnetize the way a bulk magnet does. Instead, thermal fluctuations can spontaneously flip the entire particle’s magnetization direction. Above a certain temperature, called the blocking temperature, these flips happen so fast that the particle’s time-averaged magnetization is effectively zero. The particle becomes superparamagnetic: still magnetic in a fundamental sense, but useless as a permanent magnet because its orientation drifts constantly.
The blocking temperature depends on particle size, shape, and how closely neighboring particles are packed together. Studies on iron-oxide nanoparticles have shown that stronger dipole interactions between closely spaced particles push the blocking temperature higher, essentially stabilizing each particle’s magnetization against thermal fluctuation for a bit longer.5PubMed Central. Effect of Dipole Interactions on Blocking Temperature and Relaxation Dynamics of Superparamagnetic Iron-Oxide (Fe3O4) Nanoparticle Systems This matters in fields like biomedicine, where magnetic nanoparticles are used for targeted drug delivery and MRI contrast. If those particles lose their magnetic orientation at body temperature, they are useless for guidance. Controlling size distribution and surface chemistry keeps the blocking temperature safely above 37 °C.
The key takeaway for practical purposes: the smaller the magnet, the lower the temperature at which thermal energy can overpower its magnetization. A bulk neodymium magnet laughs off 50 °C. A five-nanometer iron-oxide particle might already be superparamagnetic at that temperature if it is isolated.
Engineering Around the Heat Problem
Because neodymium magnets deliver the strongest fields per unit volume, engineers have strong motivation to push their heat tolerance higher rather than simply switching to weaker, more heat-resistant materials. One widely used strategy is adding the rare-earth element dysprosium, which boosts coercivity and raises the operating ceiling. The catch is that dysprosium is expensive and geopolitically concentrated, so manufacturers have developed grain-boundary diffusion processes in which a thin layer of dysprosium is applied to the magnet’s surface and diffused inward at high temperature. This concentrates the dysprosium at the grain boundaries where it is most effective, rather than diluting it throughout the bulk. Research into optimizing this process has found that reducing grain size or increasing the intergranular phase volume accelerates diffusion, making it possible to use less dysprosium while still achieving a meaningful coercivity boost.6Materials Science Forum. Influence of the Grain Size on the Dysprosium Diffusion in NdFeB Magnets
Other engineering approaches include designing cooling systems around the magnets (liquid-cooled motor housings are common in electric vehicles), choosing magnet grades with inherently higher coercivity even if that sacrifices some peak field strength, and shaping the magnetic circuit so the magnet operates at a lower internal stress point. None of these eliminate the thermal ceiling entirely, but together they allow neodymium magnets to work reliably in environments that would have destroyed them a generation ago.
Technology That Deliberately Uses Thermal Demagnetization
Not every application treats heat-induced demagnetization as a problem. Heat-assisted magnetic recording (HAMR) is a data storage technology that intentionally heats a tiny spot on a magnetic disk to near its Curie temperature, writes data into that softened spot with a magnetic field, and then lets it cool into a stable, high-coercivity state. The recording medium is designed with a very high room-temperature coercivity so that data stays put indefinitely, but a carefully focused laser momentarily lowers that coercivity so writing is possible.
In one study modeling this process, researchers used a magnetic film with a median Curie temperature of about 686 K (roughly 413 °C). At 700 K, without any applied field, about half the grains in a written track spontaneously demagnetized, confirming that the film was right at the transition point.7Nature / Scientific Reports. Model of advanced recording system for application in heat-assisted magnetic recording With a magnetic field applied during heating, writing could occur at an even lower temperature of about 661 K. HAMR is one of the clearest examples of thermal demagnetization being harnessed rather than feared, and it is already appearing in commercial hard drives aimed at data centers.
Magnets Deep Underground
The relationship between temperature and magnetism plays out on a planetary scale too. Earth’s crust contains ferromagnetic minerals, primarily magnetite, that contribute to the magnetic anomalies geologists use to map subsurface structures. But the deeper you go, the hotter it gets. At some depth, the temperature exceeds the Curie point of the dominant magnetic mineral, and the rock loses its magnetic signal entirely. The surface connecting all those demagnetization points underground is called the Curie depth (or Curie isotherm).
Different minerals lose their magnetism at different temperatures under burial conditions. Pyrrhotite and titanomagnetite typically demagnetize around 300–350 °C, magnetite at about 575–585 °C, and some cobalt- or nickel-bearing minerals can hold on up to 760–800 °C.8Geoscience Frontiers. Analysis of Curie point depth characteristics based on the tectonic framework of continental China Mapping the Curie depth tells geologists about the thermal structure of the crust. In volcanically active regions, the Curie depth can be shallow because the rock is hot near the surface. In old, stable continental shields, it can be tens of kilometers deep.
This geological perspective puts everyday magnet worries into context. The same physics that threatens a motor magnet at 200 °C is actively shaping which parts of Earth’s crust produce detectable magnetic fields and which do not. Temperature is arguably the most important single variable controlling where magnetism exists on the planet.
Pressure, Radiation, and Other Complicating Factors
Temperature is the dominant factor, but it does not act alone. Mechanical shock can demagnetize a magnet by physically jarring domains out of alignment. Strong opposing magnetic fields do the same, and the threshold for that kind of demagnetization drops as temperature rises because coercivity is already reduced. In devices like electromagnetic buffers, both temperature rise and field exposure happen simultaneously, creating a compounding effect that can trigger irreversible loss at a temperature lower than heat alone would require.2Defence Technology. Irreversible demagnetization mechanism of permanent magnets during electromagnetic buffering
Pressure also shifts the Curie temperature. Under high hydrostatic pressure, the atomic lattice compresses, changing the exchange interactions between magnetic atoms. For most common ferromagnetic metals, moderate pressure raises the Curie point slightly, but the effect is small compared to everyday temperature swings. It becomes relevant mainly in deep-earth geophysics, where both temperature and pressure climb together.
Radiation is another consideration in space and nuclear environments. High-energy particle bombardment can displace atoms in the magnet’s crystal lattice, degrading magnetic properties over time in a way that mimics some aspects of heat damage. Samarium cobalt magnets are generally preferred over neodymium types in radiation-heavy settings, not just for their thermal stability but for their somewhat better radiation resistance.
For most people, though, temperature remains the concern. If you leave a neodymium magnet on a dashboard in summer, it will probably be fine, since interior car temperatures rarely exceed 80 °C and standard-grade neodymium magnets tolerate that. Leave one on an engine block or near an exhaust manifold, and you may find it noticeably weaker afterward. The Curie temperature is the absolute ceiling, but the practical ceiling, the one that actually matters for your magnets, is always well below it.