Industrial Chemistry
The Thermite Reaction: Molten Iron From Rust and Aluminum
The thermite reaction is a self-oxidizing redox reaction in which powdered aluminum tears the oxygen out of iron(III) oxide — ordinary rust — releasing so much heat that the iron product pours out white-hot and molten. The balanced equation is deceptively simple, 2 Al + Fe₂O₃ → 2 Fe + Al₂O₃, yet it liberates roughly 850 kJ per mole of rust and drives the mixture to about 2500 °C, far past iron's 1538 °C melting point.
What makes it remarkable is not that it burns but how: it carries its own oxidizer locked inside the oxide, so it needs no air, cannot be smothered, and cannot be put out with water. Understanding the thermodynamics behind that is exactly what tells you why a controlled thermite demonstration is dazzling — and why the same chemistry is used to weld railway tracks and once filled incendiary bombs.
- Balanced equation2 Al + Fe₂O₃ → 2 Fe + Al₂O₃
- EnthalpyΔH ≈ −851 kJ / mol Fe₂O₃ (~4 kJ/g)
- Adiabatic temp~2500 °C (theoretical cap ~2860 °C)
- Iron melts at1538 °C — product runs molten
- IgniterMg ribbon burning ~3100 °C
- OxidizerBuilt-in — oxygen from Fe₂O₃, no air
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
What thermite actually is
A thermite is any mixture of a reactive metal fuel and a metal-oxide oxidizer that reacts to swap oxygen from the weaker oxide to the stronger one. The textbook case pairs aluminum powder with iron(III) oxide — the same red-brown Fe₂O₃ that is rust — in roughly the stoichiometric mass ratio the balanced equation demands:
2 Al(s) + Fe₂O₃(s) → 2 Fe(l) + Al₂O₃(l) ΔH ≈ −851 kJ
This is fundamentally a redox reaction. Aluminum is oxidized from the 0 oxidation state to +3, giving up three electrons per atom: Al → Al³⁺ + 3e⁻. Iron is reduced from +3 in the oxide back to metallic 0: Fe³⁺ + 3e⁻ → Fe. The oxide ions simply migrate from iron to aluminum. Aluminum is the reducing agent; iron(III) oxide is the oxidizing agent. Because the oxygen needed to burn the aluminum is already chemically bound inside the Fe₂O₃, the reaction is self-oxidizing — the single most important fact about it, and the one that separates thermite from an ordinary fire.
The thermodynamics: why aluminum steals oxygen from rust
Whether one metal can pull oxygen from another's oxide is settled by bond strength, and the definitive map of that is the Ellingham diagram, which plots the free energy of oxide formation against temperature. Aluminum's line sits far below iron's across the whole range, meaning 4/3 Al + O₂ → 2/3 Al₂O₃ is much more thermodynamically favorable per mole of oxygen than 4/3 Fe + O₂ → 2/3 Fe₂O₃. Whenever one metal's oxide line lies below another's, that metal can reduce the other's oxide. Aluminum wins handily, so it reduces the rust — a process called aluminothermic reduction.
The numbers come straight from the enthalpies of formation: ΔHf(Al₂O₃) ≈ −1676 kJ/mol versus ΔHf(Fe₂O₃) ≈ −824 kJ/mol. The reaction enthalpy is the difference, ΔH ≈ −1676 − (−824) ≈ −851 kJ per mole of Fe₂O₃. Spread over the ~214 g of reactants that represents, that is only about 4 kJ per gram — modest next to TNT's ~4.6 kJ/g.
So why the extreme temperature? Because the products are condensed — solid-then-liquid metal and oxide, with virtually no gas to carry heat away. In a wood fire, most of the released energy flows off in hot CO₂ and steam. In thermite, all 850 kJ is dumped into a small mass of iron and alumina that has nowhere to put it but its own temperature. Divide the enthalpy by the heat capacities and latent heats of the products and you overshoot iron's melting point (1538 °C) and alumina's (2072 °C) easily, reaching an adiabatic temperature near 2500 °C. The theoretical limit runs close to iron's boiling point (2862 °C), so a little iron actually vaporizes, which caps how hot it can get.
Why it needs a spark: activation energy and kinetics
If thermite is so exothermic, why can you store the mixture on a shelf indefinitely? Because thermodynamics tells you a reaction can happen, not how fast. At room temperature this one is agonizingly slow, for two reasons. First, every aluminum grain wears a tough, self-healing skin of Al₂O₃ only a few nanometers thick — the same passivating oxide that keeps aluminum foil from bursting into flame — and it physically walls the metal off from the oxidizer. Second, a solid–solid reaction requires atoms to diffuse across that interface, which has a very high activation energy. The reactants are trapped in a deep kinetic well.
To start it you must locally heat a patch to the ignition temperature, roughly 1000–1500 °C — hot enough to melt the aluminum (mp 660 °C), rupture the passivating shell, and let liquid aluminum wet the oxide. An ordinary match cannot do this; the classic igniter is a strip of magnesium ribbon, which burns at about 3100 °C, or a magnesium-and-barium-peroxide starter. Once one patch reacts, its −851 kJ heats the neighboring material past the ignition threshold, which reacts and heats the next — a self-propagating high-temperature synthesis (SHS). The glowing reaction front then crawls through the mixture at the order of centimeters per second, needing no further outside help.
Step by step: from ribbon to molten iron
- Ignition. The magnesium igniter reaches ~2000–3100 °C, delivering the activation energy to a small volume of the mixture.
- Wetting. Aluminum melts, its alumina passivation layer cracks, and liquid Al makes intimate contact with the Fe₂O₃ particles.
- Electron transfer. At the interface, aluminum atoms give up electrons (oxidation to Al³⁺) while iron(III) accepts them (reduction to Fe⁰); oxide ions shift from the iron lattice to form Al₂O₃.
- Self-propagation. The intense local heat pushes adjacent material above the ignition temperature, and a bright reaction front sweeps through the charge.
- Phase separation. Both products melt. Molten iron (density ~7 g/cm³) is far denser than molten alumina slag (~4 g/cm³), so the iron sinks to the bottom while the alumina floats on top as a glassy slag. This gravity separation is precisely what makes thermite useful for depositing a puddle of nearly pure iron exactly where you want it.
Where the heat goes to work: welding, metallurgy, incendiaries
The most enduring industrial use is exothermic (thermite) welding, patented by the German chemist Hans Goldschmidt around 1895 and still called the Goldschmidt or Thermit process. A refractory crucible of Fe₂O₃/Al is ignited over a mold clamped around the gap between two rail ends; the molten iron pours into the mold and fuses the rails into one continuous piece as it solidifies. It is how the world's continuous-welded railway track is joined in the field, with no external power — just a crucible and an igniter. The same principle, using copper-oxide thermite, underlies the “Cadweld” exothermic connections that bond copper grounding conductors in electrical substations.
Beyond welding, aluminothermic reduction is a genuine metallurgical route for metals that carbon reduces poorly, including chromium (from Cr₂O₃), manganese, vanadium, and various ferroalloys — the reaction gives them without carbon contamination. On the destructive side, thermite's oxygen-independence made it a WWII incendiary: the U.S. AN-M14 thermate grenade burns at roughly 2200 °C and can melt through an engine block or a gun breech, and thermite filled countless incendiary bombs precisely because water cannot extinguish it.
Variants: copper thermite, nano-thermite, and the oxidizer zoo
“Thermite” is a family, not a single recipe. Swap the oxidizer or the fuel and the character changes:
- Copper(II) oxide thermite: 2 Al + 3 CuO → Al₂O₃ + 3 Cu. Far faster and more violent than the iron version — bordering on explosive when confined — which is why it is used for the fast, molten-copper Cadweld joints rather than slow rail welding.
- Other oxidizers: magnetite (Fe₃O₄), MnO₂, Cr₂O₃, and PbO all serve; each shifts the energy release and product melting points. Even the classic iron thermite often uses Fe₃O₄ mill scale in practice.
- Other fuels: magnesium, calcium, and silicon can replace aluminum, trading temperature for cost or reactivity.
- Nano-thermite (metastable intermolecular composites): milling the aluminum and oxidizer to nanometer scale slashes the diffusion distance, so the reaction that took seconds in a crucible can complete in microseconds — a research-grade material studied in propellants and pyrotechnics.
Common misconceptions — and why it is genuinely dangerous
“Thermite explodes.” Classic Fe₂O₃/Al thermite does not detonate; it deflagrates — it burns very hot but produces almost no gas and no shockwave. The hazard is not a blast but blinding light (the molten iron radiates like a small sun and can damage unprotected eyes) and a puddle of ~2500 °C metal that ignites or melts almost anything it touches.
“Just pour water on it.” This is the dangerous mistake. Molten iron flash-boils water to steam explosively and can even reduce it, liberating hydrogen — so water causes violent spattering of molten metal, not extinguishment. Because the oxidizer is built in, you also cannot smother thermite by cutting off air; it will burn underwater or in a vacuum. “Rust is inert waste.” is the last misconception the reaction demolishes: relative to aluminum, rust is a loaded energy reservoir waiting for a partner that bonds oxygen even harder. For all these reasons thermite is only ever run as a controlled demonstration or industrial process — behind shielding, over sand or a firebrick base, with eye protection — and the value of understanding the chemistry is knowing exactly why those precautions exist.
| Property | Thermite (Fe₂O₃/Al) | Ordinary combustion (wood, fuel) | High explosive (TNT) |
|---|---|---|---|
| Oxidizer source | Bound O in the metal oxide (self-contained) | Atmospheric O₂ from the air | Bound O in the molecule (self-contained) |
| Main products | Molten metal + molten oxide slag (condensed) | CO₂ + H₂O gas + soot | Hot gases (N₂, CO, CO₂, H₂O) |
| Peak temperature | ~2500 °C, iron runs liquid | ~1000–1500 °C flame | ~3000–5000 °C, but fleeting |
| Propagation | Deflagration front, cm/s (self-sustaining) | Diffusion-limited flame front | Detonation, km/s shockwave |
| Gas / blast | Almost none — no shockwave | Modest hot gas, no blast | Enormous — a supersonic pressure wave |
| Can you smother it? | No — remove air and it still burns | Yes — starve it of O₂ | No — but it self-consumes in microseconds |
Frequently asked questions
What is the balanced equation for the thermite reaction?
The classic reaction is 2 Al + Fe₂O₃ → 2 Fe + Al₂O₃. Two moles of aluminum reduce one mole of iron(III) oxide to give two moles of iron and one mole of aluminum oxide. It releases about 851 kJ of heat per mole of Fe₂O₃, driving the iron out molten.
How hot does thermite get?
The adiabatic reaction temperature is roughly 2500 °C, and theoretically it can approach iron's boiling point near 2862 °C — at which point some iron vaporizes and caps the temperature. That is well above iron's 1538 °C melting point, which is why the iron product runs liquid and white-hot.
Why does thermite not need air to burn?
The oxidizer is chemically built into the mixture: the oxygen is already bound inside the iron(III) oxide. The aluminum simply takes that oxygen, so no atmospheric O₂ is required. This is why thermite burns underwater, in a vacuum, and cannot be extinguished by cutting off air.
Why does thermite need a magnesium ribbon to start?
The reaction has a very high activation energy — the aluminum is sheathed in a passivating oxide layer and the atoms must diffuse across a solid interface. You must locally heat the mix to roughly 1000–1500 °C to ignite it, and magnesium ribbon (burning near 3100 °C) supplies that. Once lit, the reaction's own heat sustains it.
What is thermite actually used for?
Its biggest legitimate use is exothermic (thermite) welding of railway track and copper grounding conductors, where molten iron or copper fuses joints with no external power. Aluminothermic reduction also produces metals like chromium and manganese. Historically, its oxygen-independence made it a military incendiary.
Can you put out a thermite reaction with water?
No — and it is dangerous to try. Water hitting ~2500 °C molten iron flash-boils to steam and can be reduced to hydrogen, causing violent spattering of molten metal. Because thermite carries its own oxidizer, smothering it does not work either; it must simply burn out over a fire-safe base.