White Phosphorus: The Element Kept Under Water

The Molecule: Four Atoms Forced Into a 60° Cage

White phosphorus is a molecular solid built from P₄ units. Four phosphorus atoms sit at the corners of a regular tetrahedron, each bonded to the other three, giving six P–P bonds of 2.21 Å. But the geometry is a trap: a tetrahedron forces every P–P–P angle to 60°, while phosphorus — bonding through orbitals with heavy 3p character and very little s–p hybridisation — wants roughly 93–100°. The H–P–H angle in phosphine is 93.5°; in trimethylphosphine, about 98.6°.

Squeezing a ~95° preference down to 60° means the bonds cannot lie along the lines joining the nuclei. Electron density bows outward into bent “banana” bonds, overlap is poor, and the molecule carries the shortfall as stored energy: ~96 kJ/mol (22.8 kcal/mol) of strain per P₄. Spread over six bonds that is about 16 kJ/mol per bond — and the textbook P–P enthalpy of ~200 kJ/mol is no yardstick for it, because that figure is itself derived from P₄ (atomising the gas costs 1,207 kJ/mol, or 201 kJ/mol per bond) and so already carries the strain. A strain-free P–P increment sits nearer 217 kJ/mol, so every bond in the cage runs at roughly a 7% energy penalty before anything touches it.

Two things follow. First, P₄ is the least stable common allotrope: converting it to red phosphorus releases about 70 kJ per mole of P₄ (ΔH_f° of red P is −17.6 kJ per mole of P atoms), and to black phosphorus about 157 kJ. White phosphorus is nonetheless the thermodynamic reference state, ΔH_f° ≡ 0, chosen for reproducibility — the textbook zero is not the stable form. Second, P₄ exists only because phosphorus is bad at multiple bonds: diffuse 3p orbitals overlap sideways poorly, so P≡P (~490 kJ/mol) loses to three P–P single bonds (~600 kJ/mol), the reverse of nitrogen, where N≡N at 945 kJ/mol beats three 160 kJ/mol singles. Nitrogen ends up an inert gas; phosphorus ends up a strained cage already halfway to reacting.

The Glow: A Branched Chain Running in the Vapour

The green-white glow does not come from the waxy solid but from a thin layer of P₄ vapour above it: white phosphorus has a vapour pressure of only about 4 Pa at 20 °C, and all the interesting chemistry happens in that whisper of gas.

Worked example. Convert that pressure to a concentration: n/V = p/RT = 4 Pa ÷ (8.314 J mol⁻¹ K⁻¹ × 293 K) = 1.6 × 10⁻³ mol/m³. Multiply by the molar mass of P₄, 123.9 g/mol, and the saturated headspace holds ≈ 0.20 g/m³ — about 200 mg/m³. The OSHA permissible exposure limit is 0.1 mg/m³, so the air just above an uncovered lump sits roughly 2,000× the workplace limit and 40× the NIOSH IDLH of 5 mg/m³. A negligible vapour pressure is not a negligible hazard.

Oxygen attacks that vapour as a branched chain: initiation makes radicals, propagation carries them, and — crucially — some steps hand back more carriers than they consume, so the radical population multiplies rather than merely persisting. The carriers are PO•, PO₂• and O atoms, but they are not what you see — PO's own band systems sit in the ultraviolet. The green light comes from two short-lived products born electronically excited, the dimer (PO)₂ and HPO, radiating a broad continuum that peaks at 510–560 nm; HPO is why rigorously dried phosphorus and oxygen barely glow. The light is manufactured fresh by each reaction event, not released from storage. Termination, meanwhile, happens largely on the vessel walls.

Semenov's Critical Pressure, and Why the Glow Becomes a Fire

In 1926 two of Nikolay Semenov's students, Yuli Khariton and Zinaida Val'ta, found something ordinary kinetics forbids. Below a sharp lower critical oxygen pressure the phosphorus vapour did not glow at all; a hair above it, the glow switched on. Rates are supposed to vary smoothly with pressure, not step. Stranger still, adding an inert gas lowered the critical pressure — exactly what you expect if radicals die by diffusing to the walls and the inert gas slows that diffusion.

Semenov's 1927 interpretation founded branched-chain theory: chains survive when branching outruns termination, and that ratio crosses one abruptly, producing an ignition threshold rather than a gradual onset. Cyril Hinshelwood reached the same framework independently on the hydrogen–oxygen reaction; the two shared the 1956 Nobel Prize in Chemistry. Phosphorus also shows an upper critical pressure: raise the oxygen partial pressure far enough and three-body gas-phase termination outruns branching, so a sample that glows in air can sit dark in pure oxygen near one atmosphere, then relight as the pressure is pumped down.

The chain also builds its own trigger. Every cycle is exothermic, and a warmer sample has a higher P₄ vapour pressure, which feeds the chain more fuel, which releases more heat. That loop closes at about 30 °C — below 37 °C body temperature, and below white phosphorus's own melting point of 44.1 °C. The element ignites before it melts. A lump warmed in a bare hand is being held above its autoignition temperature.

Combustion Arithmetic, and Why the Jar Is Full of Water

Once runaway begins the reaction is P₄ + 5 O₂ → P₄O₁₀, releasing ~2,984 kJ per mole of P₄. At 123.9 g/mol that is ~24 kJ/g — roughly five times the ~4.6 kJ/g TNT delivers on detonation, and about half of gasoline's 46 kJ/g. It is not an explosive — it needs outside oxygen and burns as a deflagration — and per gram it merely matches magnesium (~25 kJ/g) and trails aluminium (~31 kJ/g). What sets it apart is that nothing has to light it.

The product is a dense white smoke of P₄O₁₀ molecules — the familiar “P₂O₅” is an empirical formula, not a structure. P₄O₁₀ is ferociously hygroscopic and hydrolyses to phosphoric acid, P₄O₁₀ + 6 H₂O → 4 H₃PO₄, so 124 g of phosphorus pulls water from the air to build 392 g of acid aerosol: an excellent obscurant, and a corrosive one.

Water storage attacks this on four fronts at once, and the numbers explain each. Solubility: P₄ dissolves at only 3 mg/L at 20 °C, so the water cannot carry the sample away. Density: at 1.82 g/cm³ it sinks and stays covered rather than floating up to meet the air. Oxygen supply: air-saturated water holds ~9 mg/L of O₂, about a thirtieth of air's molar concentration, and oxygen diffuses through water some 10⁴ times more slowly than through gas, so the branched chain is starved at the source. Temperature: water's heat capacity and evaporative cooling pin the sample below the ~30 °C threshold. Industry uses the same trick at scale — molten P₄ is held and shipped a little above its 44.1 °C melting point, typically around 50–60 °C, under a blanket of water, with the vapour space usually padded with nitrogen.

Making It, Shipping It, Measuring It

Commercially, white phosphorus comes from the thermal (electric-furnace) process: phosphate rock, silica and coke go into a submerged-arc furnace at ~1,400–1,500 °C, where 2 Ca₃(PO₄)₂ + 6 SiO₂ + 10 C → 6 CaSiO₃ + 10 CO + P₄. The P₄ leaves as vapour and is condensed under water, from the first seconds of its existence. Almost all of it goes onward to PCl₃, P₄S₁₀ and high-purity thermal phosphoric acid, and thence to herbicides, flame retardants and lubricant additives.

Standards. It ships as UN 1381 (Phosphorus, white or yellow, dry or under water or in solution), hazard class 4.2, spontaneously combustible, subsidiary risk 6.1, toxic, packing group I, GHS hazard statement H250, catches fire spontaneously if exposed to air; molten material travels as UN 2447. The OSHA PEL and NIOSH REL are both 0.1 mg/m³ as an 8-hour TWA, with an IDLH of 5 mg/m³. Pyrophoric classification itself comes from UN Manual of Tests and Criteria Test N.2: a 1–2 mL portion of the solid is dropped onto a non-combustible surface, and it counts as pyrophoric if it ignites during the fall or within five minutes of settling.

Measurement. The 2.21 Å bond length and the tetrahedral geometry came from gas-phase electron diffraction. Solution ³¹P NMR is unmistakable: P₄ gives one sharp line near δ −520 ppm against 85% H₃PO₄, among the most upfield ³¹P shifts known. Raman spectroscopy picks out the totally symmetric breathing mode near 606 cm⁻¹. For contaminated soil and water, EPA SW-846 Method 7580 extracts white phosphorus into solvent and quantifies it by gas chromatography on a phosphorus-selective detector, down to parts-per-billion levels. One such detector closes the circle: the flame photometric detector identifies phosphorus by burning it and watching the HPO* emission band at 526 nm. The same green light, turned into an instrument reading.

From Hennig Brand to Phossy Jaw to Protocol III

In 1669 the Hamburg alchemist Hennig Brand, boiling down vats of urine in search of the philosopher's stone, distilled out a waxy solid that glowed in the dark without being heated. It was the first element with a named discoverer, and its light named it: Greek phosphoros, light-bearer.

The commercial consequence arrived in the 1830s, when Charles Sauria and others put white phosphorus into match heads. The matches struck anywhere; the workers who dipped them breathed the vapour and developed “phossy jaw”, a necrosis of the mandible that contemporary accounts describe as glowing greenish-white in the dark, and that was usually treated by removing the bone. The Bryant and May matchgirls' strike in London in 1888 turned it into a public scandal, and the Berne Convention of 1906 banned white phosphorus in matches internationally. Britain legislated in 1908; the United States, a non-signatory, taxed the matches out of existence with the Esch–Hughes Act of 1912.

The same properties made it a standard military smoke and incendiary filler — “Willie Pete”, in rounds such as the 155 mm M825 felt-wedge projectile. Its use at Fallujah in 2004 and over populated parts of Gaza in 2008–09 drew sustained controversy: US officials first said the Fallujah rounds served only to illuminate and screen, then acknowledged their use as an incendiary against combatants, while the Gaza shelling was documented by Human Rights Watch and examined by the UN's 2009 fact-finding mission. Legally it is not a scheduled chemical weapon under the Chemical Weapons Convention, which covers agents acting by toxicity. Incendiary use falls instead under Protocol III of the 1980 Convention on Certain Conventional Weapons, which restricts incendiary weapons near concentrations of civilians — though Protocol III defines an incendiary weapon by design intent and expressly excludes munitions built primarily to make smoke or illuminate, which is how white phosphorus shells are normally classified, whatever they set alight. There is an environmental legacy too: at Eagle River Flats in Alaska, particles from decades of smoke rounds settled into estuarine sediment and killed on the order of a thousand to two thousand dabbling ducks a year until the ponds were drained.

What It Is Not: Phosphorescence, Red Phosphorus, and Other Look-Alikes

It is not phosphorescence. The element named the phenomenon it does not perform. Phosphorescence is delayed re-emission of light a material has already absorbed, released through a long-lived triplet state — charge a glow-in-the-dark star at a lamp and watch it fade. White phosphorus absorbs nothing: its energy comes from making P–O bonds, so the glow is chemiluminescence and lasts exactly as long as there is phosphorus and oxygen. Nor is it fluorescence, which stops with the lamp, or incandescence, which needs a hot body; this glow runs at room temperature.

• Not red phosphorus. Same element, different architecture: red P is an amorphous polymer of opened-up P₄ cages, ignites near 260 °C, does not glow, is insoluble in CS₂ and is not acutely toxic.
• Not radioactive. ³¹P is a stable nuclide; the light is chemistry, not decay.
• Not permanently put out by water. Water suppresses a burning fragment only while it stays wet; dried residue re-ignites. Hence contaminated wounds are irrigated and kept wet, and the rule is to submerge, not douse.
• Not a low-dose nuisance. The oral lethal dose is on the order of 1 mg/kg, and poisoning brings hypocalcaemia, liver failure and cardiac arrhythmia. The old copper-sulfate wound wash, used to coat particles black for removal, is now discouraged because absorbed copper causes haemolysis.

The final irony sits in a kitchen drawer. Strike a safety match and friction converts a trace of the striker's red phosphorus into a microscopic quantity of white phosphorus, which does what white phosphorus does and sets off the potassium chlorate in the head. The allotrope banned from match heads in 1906 still lights every match.

The common allotropes of phosphorus — one element, four architectures, wildly different behaviour.

Allotrope	Structure	Ignites in air at	Density and handling
White (yellow) P	Discrete P₄ tetrahedra; 60° angles, 2.21 Å bonds, ~96 kJ/mol strain	~30 °C (glows well below it)	1.82 g/cm³ — sinks; stored under water; lethal dose ~1 mg/kg
Red P	Amorphous polymer of opened-up P₄ cages	~260 °C	~2.16 g/cm³ — air-stable powder; the striker strip on safety matches
Violet (Hittorf's) P	Crystalline tubular polymer; the ordered form of red	~300 °C	~2.36 g/cm³ — made by slow crystallisation from molten lead
Black P	Puckered graphite-like layers; the thermodynamically stable form	above ~400 °C	2.69 g/cm³ — a semiconductor; exfoliates to phosphorene