Industrial Chemistry

The Mond Process

Refine nickel by turning it into a gas and back again

The Mond process refines nickel by a temperature swing: crude metal reacts with carbon monoxide near 50 °C to form volatile nickel tetracarbonyl, Ni(CO)₄, which is then decomposed on hot nickel pellets at ~230 °C to deposit pure nickel and release the CO for reuse. It exploits a rare gaseous metal compound to leave iron, copper, and cobalt behind.

  • Discovered1890 (Mond & Langer)
  • Key moleculeNi(CO)₄ (b.p. 43 °C)
  • Volatilise at≈ 50 °C, 1 atm
  • Decompose at≈ 230 °C
  • Purity achieved99.9%+ nickel
  • First refineryClydach, Wales, 1902

Interactive visualization

Press play, or step through manually. The visualization is yours to drive — try it before reading on.

Open visualization fullscreen ↗

Watch the 60-second explainer

A condensed visual walkthrough — narrated, captioned, under a minute.

What the Mond process does

Almost every metal is refined by melting it, dissolving it in acid, or plating it out of solution. Nickel has a fourth option that no other common metal shares: it can be turned into a gas, floated away from its impurities, and then turned back into a solid on demand. The gas is nickel tetracarbonyl, Ni(CO)₄ — a colourless, volatile liquid (boiling point just 43 °C) that forms when carbon monoxide is passed over impure nickel at mild temperature.

The trick rests on a single reversible reaction and the fact that it flips direction with temperature:

    Ni(s) + 4 CO(g)  ⇌  Ni(CO)₄(g)     ΔH ≈ −160 kJ/mol

  forward (cool, ~50 °C):  nickel dissolves into the gas → volatiliser
  reverse (hot, ~230 °C):  the gas gives the nickel back → decomposer

Crude nickel — typically from smelting sulfide ores, and contaminated with iron, cobalt, copper, and traces of gold, silver, and platinum — enters a cool volatiliser. Only the nickel reacts with the CO; the impurities cannot form a volatile carbonyl under these conditions and are left behind as a solid residue. The Ni(CO)₄ vapour is carried in the gas stream to a hotter decomposer, where it breaks apart and plates fresh nickel onto growing pellets. The liberated CO is pumped back to the volatiliser and used again. The result is nickel at better than 99.9% purity, produced in a continuous loop.

The mechanism: two chambers, one equilibrium

The heart of the process is the same molecule assembling in one chamber and disassembling in the other. Follow the electrons.

  1. Carbonyl formation (volatiliser, ~50 °C). A carbon monoxide molecule approaches a nickel surface atom. CO is a σ-donor: its carbon lone pair points into an empty nickel orbital, forming a Ni←CO dative bond. Nickel then pushes electron density back the other way — its filled d-orbitals overlap with the empty π* antibonding orbitals of CO. This back-donation (a synergic π-bond) is what makes the metal–carbonyl bond strong enough to lift the nickel atom clean off the surface.
  2. Reaching 18 electrons. Nickel(0) brings 10 d-electrons. Each CO donates a lone pair worth 2 electrons. Four CO ligands add 8, and 10 + 8 = 18 — the closed, krypton-like shell the 18-electron rule predicts. Nickel accepts exactly four CO ligands (not three, not five) because four is what fills the shell, and the complex adopts a symmetric tetrahedral geometry.
  3. Volatilisation. The freshly built Ni(CO)₄ molecule has no net charge and only weak van der Waals attractions to its neighbours, so it boils off at 43 °C and joins the gas stream. The nickel atom has effectively evaporated, wrapped in four CO molecules.
  4. Transport. The gas — CO carrier plus Ni(CO)₄ vapour — flows to the decomposer, leaving the non-volatile iron, cobalt, copper, and precious-metal residue behind in the volatiliser.
  5. Decomposition (decomposer, ~230 °C). Heat drives the equilibrium backward. Thermal energy exceeds the metal–ligand bond strength; the four Ni–C bonds break, the back-donation collapses, and the nickel atom deposits onto a hot nickel seed pellet. The four CO molecules are released intact.
  6. Regeneration. The freed CO is recompressed and returned to the volatiliser. It is a genuine catalyst-like carrier: consumed in the forward step, regenerated in the reverse step, and reused indefinitely apart from make-up gas for leaks.

Because the two chambers share one equilibrium at two temperatures, the process is essentially a thermal pump for a single metal — nickel is ferried across on a shuttle of carbon monoxide.

Reagents, conditions, and the two operating regimes

There are two industrial versions, differing mainly in pressure:

  • Classic (low-pressure) Mond, as run at Clydach. Volatilisation at roughly 50–80 °C and near-atmospheric CO pressure; decomposition at about 230 °C. The mild conditions are what make the process so selective — iron and cobalt need far harsher conditions to form their carbonyls, so at low pressure they simply don't.
  • High-pressure carbonyl process (Vale/INCO, Copper Cliff and Sudbury). Volatilisation at around 150 °C and 70 atm (7 MPa) of CO, which drives the equilibrium hard toward the carbonyl and speeds up an otherwise slow reaction. The trade-off is that iron and cobalt carbonyls now form in small amounts, so the crude Ni(CO)₄ is fractionally distilled before decomposition.
    VOLATILISER  ~50 °C, ~1 atm        DECOMPOSER  ~230 °C
    ┌──────────────────────┐          ┌──────────────────────┐
    │ crude Ni + CO(g)     │  Ni(CO)₄ │ Ni(CO)₄ → Ni + 4 CO  │
    │  → Ni(CO)₄(g) ↑      │ ───────► │  pure Ni plates out  │
    │  Fe, Cu, Co left     │          │  CO recycled ◄────── │
    └──────────────────────┘          └──────────────────────┘
              ▲                                   │
              └──────────── CO returned ──────────┘
  • Feed. Crude nickel from sulfide-ore smelting, often reduced first with hydrogen or water gas to give a reactive, high-surface-area metal (nickel oxide reacts poorly; the metal must be freshly reduced).
  • Reagent. Carbon monoxide, typically 90–100% CO from producer gas or reformed syngas.
  • Product. Nickel pellets or a fine nickel powder (carbonyl nickel powder), 99.9%+ pure, valued for its controllable particle shape.
  • Byproduct. A residue enriched in copper, cobalt, iron, and precious metals — itself a valuable feed for further refining.

Why it is so selective

The whole process hinges on one fact: nickel forms a volatile carbonyl under mild conditions and its neighbours do not.

MetalNeutral carbonylBehaviour under mild Mond conditions
NickelNi(CO)₄, tetrahedral, b.p. 43 °CForms readily at 50 °C, volatilises — carried off
IronFe(CO)₅, trigonal bipyramidal, b.p. 103 °CNeeds high CO pressure and heat; stays put at low pressure
CobaltCo₂(CO)₈, dimeric, decomposes ~52 °CForms sluggishly, far less volatile — left in residue
CopperNone stableNo neutral carbonyl — never leaves the residue
Gold / PlatinumNoneInert to CO — concentrate in the residue

Nickel(0) is the sweet spot: it has exactly the right number of d-electrons (10) to reach 18 with four CO ligands, and its metal–CO bond strength lets the carbonyl form at low temperature yet fall apart again with modest heating. Iron needs five ligands and much stiffer conditions; cobalt's odd electron count forces it to dimerise into a heavy, non-volatile molecule; copper's filled d¹⁰ shell has little to back-donate and forms no lasting carbonyl at all. The Mond process is, in effect, the 18-electron rule turned into an industrial separation.

Worked example: refining a tonne of crude nickel

Suppose you feed 1,000 kg of crude nickel (about 95% Ni, the rest Fe/Cu/Co) into a low-pressure Mond unit.

  • Nickel to volatilise. 950 kg of Ni is 950,000 / 58.69 ≈ 16,200 mol.
  • CO required per pass. Each Ni needs 4 CO, so 4 × 16,200 ≈ 64,700 mol of CO are bound at any instant — roughly 1.45 million litres at STP. But this CO is recycled: the decomposer hands every molecule back, so the standing inventory of CO, not a continuous feed, is what matters.
  • Product. Essentially all 950 kg of nickel reappears as 99.9%+ pellets in the decomposer. The 50 kg of iron, copper, and cobalt stay in the volatiliser as residue.
  • Energy. Formation releases ~160 kJ/mol (16,200 mol → ~2.6 GJ given off in the volatiliser); the same amount, plus process heat, is supplied in the decomposer to reverse it. The temperature swing, not a net chemical consumption, is the real cost.

The elegance is that the reagent bill is almost nil. Carbon monoxide is a carrier, not a consumable — you pay for the heat to swing the equilibrium and for topping up CO lost to leaks, and in return you get a metal purer than most electrolytic routes deliver.

Mond process vs electrolytic nickel refining

Mond (carbonyl) processElectrolytic refining
PrincipleReversible volatile carbonyl, temperature swingAnodic dissolution + cathodic plating
Key reagentCO (recycled, not consumed)Sulfate/chloride electrolyte + electricity
Selectivity sourceOnly Ni forms a volatile carbonyl (mild conditions)Standard electrode potentials of the metals
Typical purity99.9%+ (99.99% for carbonyl powder)99.9%+
Product formPellets or shaped fine powderCathode sheets / squares
Main hazardNi(CO)₄ — extreme acute toxicity (0.05 ppm TLV)Electrolyte handling, energy use
Handles precious-metal-rich feed?Yes — Au/Pt concentrate in the residueYes — collect in anode slime
Continuous?Yes — closed CO loopBatch-ish, tied to electrode life

A real application: Clydach and Sudbury nickel

The first Mond refinery opened in Clydach, near Swansea in south Wales, in 1902, built by the Mond Nickel Company. It refines nickel by the carbonyl route to this day, more than 120 years later — one of the longest-running processes in industrial chemistry, now owned by Vale.

Across the Atlantic, the enormous nickel–copper sulfide deposits of the Sudbury Basin in Ontario (the eroded remains of a 1.85-billion-year-old meteorite impact crater) are refined partly by a high-pressure carbonyl process at Copper Cliff. There the crude nickel sees ~150 °C and ~70 atm of CO; the resulting Ni(CO)₄ is distilled away from small amounts of Fe(CO)₅ before being decomposed. Carbonyl nickel is prized for battery electrodes, catalysts, superalloys for jet-engine turbine blades, and coinage — the coins in your pocket often trace their nickel back through a molecule of Ni(CO)₄.

Limitations and side reactions

  • Extreme toxicity of Ni(CO)₄. The intermediate is one of the deadliest industrial gases, lethal at a few ppm. Every joint, valve, and pump must be leak-tight and continuously monitored; a Ni(CO)₄ release is a fatal emergency, not a spill to mop up.
  • Feed must be reactive metal. Nickel oxide barely reacts with CO. The feed has to be pre-reduced (with H₂ or water gas) to porous, high-surface-area metal, adding a step and consuming reductant.
  • Cobalt carry-over at high pressure. The high-pressure regime that speeds volatilisation also drags in some cobalt and iron carbonyls, forcing a distillation stage the mild Clydach process avoids.
  • CO make-up and containment. Carbon monoxide is itself toxic and flammable, and although recycled it leaks; the plant needs a constant CO top-up and rigorous gas handling.
  • Slow forward kinetics at low pressure. At 1 atm the carbonyl forms slowly, which is why throughput-hungry plants moved to 70 atm despite the loss of selectivity.

Discovery: corrosion that turned out to be a purification

The Mond process was born from an industrial nuisance. In the late 1880s Ludwig Mond, a German-born chemist and co-founder of the Brunner Mond alkali company (later a cornerstone of ICI), was running an ammonia-soda plant and noticed that carbon monoxide in the gas stream was mysteriously corroding the nickel valves. Working with his assistant Carl Langer, Mond investigated and in 1890 identified the culprit: a new, volatile compound, nickel tetracarbonyl. When the gas was heated, it deposited a bright mirror of pure nickel — the "corrosion" was picking up nickel in one place and putting it down, purified, in another.

Mond immediately grasped that a nuisance was a refining method in disguise, patented it, and built the Clydach works. Ni(CO)₄ was the first metal carbonyl ever discovered; its study helped launch the whole field of organometallic chemistry and gave chemists their clearest early example of the 18-electron rule and of metal-to-ligand back-donation.

Safety and industrial notes

Nickel tetracarbonyl earns its reputation. It is a colourless liquid at room temperature whose vapour has a threshold limit value near 0.05 ppm — thousands of times lower than carbon monoxide's already-dangerous limit. It is absorbed through the lungs and skin, decomposes inside the body to deposit metallic nickel and free CO, and produces a delayed chemical pneumonitis: a worker may feel only mildly unwell, then suffer lung damage a day or two later. It is also flammable and can decompose explosively.

Because of this, carbonyl refineries are engineered as fully sealed, negative-pressure systems with continuous atmospheric monitoring and no routine human contact with the gas. The same properties that make Ni(CO)₄ a superb purification vehicle — volatility, easy decomposition, mobility through the body — are exactly what make it deadly. The Mond process is a standing lesson that a molecule's usefulness and its hazard can spring from the very same chemistry.

Frequently asked questions

Why does carbon monoxide dissolve nickel but leave iron and copper behind?

Nickel is nearly unique among the common base metals in forming a stable, volatile carbonyl under mild conditions. At 50–80 °C and atmospheric pressure, Ni reacts spontaneously with CO to give gaseous Ni(CO)₄, which simply boils off (b.p. 43 °C) and carries the nickel away as a vapour. Copper forms no neutral carbonyl at all. Iron does form Fe(CO)₅, but far more slowly and only under much higher CO pressure and temperature, so under Mond's mild low-pressure conditions the iron and copper stay put as solids. This selectivity is the whole trick of the process.

What is the 18-electron rule and why does Ni(CO)₄ obey it?

The 18-electron rule says a stable low-oxidation-state transition-metal complex tends to have 18 valence electrons — a filled set of nine bonding orbitals (one s, three p, five d), the noble-gas configuration of krypton. Nickel(0) contributes 10 d-electrons, and each of the four CO ligands donates a lone pair (2 electrons), giving 10 + 4×2 = 18. That closed shell is exactly why the tetrahedral Ni(CO)₄ is so stable and volatile, and why nickel takes four CO ligands rather than three or five.

How does the process separate nickel from cobalt?

Cobalt does form a carbonyl, Co₂(CO)₈, but only sluggishly and at higher CO pressure, and it is far less volatile than Ni(CO)₄. Under Mond's mild low-pressure conditions the cobalt is left behind in the residue with the copper and precious metals. Modern high-pressure carbonyl plants that run at 150 °C and 70 atm do pick up some cobalt, so they add a preliminary distillation step: Ni(CO)₄ boils at 43 °C, well below the decomposition range, allowing the two carbonyls to be fractionated before decomposition.

Why is the temperature swing the key to the whole design?

The formation of Ni(CO)₄ is exothermic and reversible: Ni + 4 CO ⇌ Ni(CO)₄. Le Chatelier's principle means low temperature favours the carbonyl (forward) and high temperature favours the metal (reverse). Mond exploited this by running two chambers on the same equilibrium: a cool volatiliser near 50 °C where nickel dissolves into the gas, and a hot decomposer near 230 °C where the same molecule breaks apart and plates out. The carbon monoxide is regenerated at each cycle and pumped back, so it acts as a reusable carrier rather than being consumed.

How dangerous is nickel tetracarbonyl?

Extremely. Ni(CO)₄ is one of the most toxic industrial gases known — a colourless, volatile liquid whose vapour is lethal at a few parts per million, with a threshold limit value of about 0.05 ppm. It is absorbed through the lungs and skin, decomposes in the body to deposit nickel and release CO, and causes delayed chemical pneumonitis that can kill days after an apparently mild exposure. It is also flammable and can detonate. Carbonyl refineries are engineered as sealed, monitored systems precisely because a leak of this gas is deadly.

Who invented the Mond process and when?

Ludwig Mond, along with his assistant Carl Langer, discovered nickel tetracarbonyl in 1890 while investigating why CO gas was corroding the nickel valves in his ammonia-soda (Solvay) plant. Mond realised the corrosion was itself a purification reaction and patented it; the first commercial Mond nickel refinery opened in Clydach, near Swansea in Wales, in 1902. That plant still refines nickel by the carbonyl route today, well over a century later.