Noble Gases

Group 18 is the one family on this table that arrived as a complete surprise. Mendeleev's 1869 arrangement worked by leaving gaps where an element ought to be, and every famous vindication of it — gallium, scandium, germanium — involved something dropping into a slot he had already drawn. The noble gases dropped into no slot at all. They required an entire new column to be added on the right of the table, and the fact that the periodic law absorbed eight new elements without breaking was a far more severe test than any of the predicted-gap successes.

They were missed because they do almost nothing. A family defined by chemical inactivity leaves no salts, no ores and no reactions to be curious about, and argon makes up nearly one per cent of the atmosphere without anyone having noticed it for the whole of the chemical revolution.

Found in a discrepancy of half a per cent

Lord Rayleigh spent years measuring gas densities carefully enough to matter, and found that nitrogen extracted from air was consistently about half a per cent heavier than nitrogen made from ammonia. The difference was small, reproducible, and refused to go away.

Working with William Ramsay from 1894, he removed the oxygen, the nitrogen, the carbon dioxide and the water from a sample of air and found that something was left — about one part in a hundred that would combine with nothing. They called it argon, from the Greek for idle. The residue turned out to contain more than one gas: neon, krypton and xenon were separated out of it by fractional distillation in 1898, three elements in a matter of weeks.

Helium had already been seen, but not on Earth. During the 1868 solar eclipse Pierre Janssen and, independently, Norman Lockyer recorded a yellow spectral line in the Sun's chromosphere that matched no known element, and Lockyer named the hypothetical source after the Sun. It stayed hypothetical until 1895, when Ramsay released gas from the uranium mineral cleveite and found the same line in the laboratory — the only element identified in a star before it was identified on the planet.

Radon completed the column in 1900, when Friedrich Ernst Dorn observed that radium produced a radioactive gas. The 1904 Nobel awards recognised both ends of the work at once: physics went to Rayleigh, chemistry to Ramsay, one year, one finding cut down the middle.

Full shells, but that is only half the explanation

The textbook account is that these atoms have complete octets and therefore do not bond. That account is not wrong so much as incomplete, and it fails immediately on helium, which is stable with two electrons rather than eight.

The property that actually varies down the column is how tightly the outermost electrons are held. Helium's first ionisation energy is 2,372 kJ/mol, the highest of any element; neon's is 2,081; and by xenon it has fallen to 1,170, which is lower than that of oxygen. That last comparison is not a curiosity. It is the reason xenon has a chemistry at all.

Two other properties change smoothly down the group and both come from the same growth in electron count. Boiling points climb from helium's 4.2 K to radon's 211 K, because larger, more polarisable atoms give stronger London dispersion forces even though no atom has a permanent dipole. And the same polarisability makes the heavier members progressively easier to excite, which is why the column's discharge colours range from helium's pale peach through neon's saturated red-orange to the blue-violet of krypton and xenon.

Neil Bartlett's inference

In 1962 Neil Bartlett, then at the University of British Columbia, was working with platinum hexafluoride and observed that it oxidised molecular oxygen, producing the salt O₂⁺[PtF₆]⁻. Removing an electron from O₂ requires 1,175 kJ/mol.

Bartlett noticed that this figure is almost identical to xenon's first ionisation energy. A whole family had been written off on a principle nobody had thought to check against a number, and here was the number, sitting in his own results. He mixed the two gases, got an orange-yellow solid immediately, and ended sixty years of received opinion on the strength of a coincidence he had the presence of mind to take seriously.

Within months other laboratories had xenon tetrafluoride, xenon difluoride and xenon hexafluoride, and xenon chemistry is now extensive: oxides, oxyfluorides, xenate and perxenate salts, and compounds with xenon–carbon and xenon–nitrogen bonds. Krypton difluoride exists but is thermodynamically unstable and must be kept cold. Radon should in principle be the most reactive of all, and its fluoride has been detected, but its longest-lived isotope decays with a half-life under four days, so the chemistry is studied at tracer scale. Argon has given exactly one compound, argon fluorohydride, observed in a solid noble-gas matrix at temperatures below 40 K. Neon and helium have given nothing under any ordinary condition.

Reactivity in this column therefore increases downwards, which is the reverse of what happens among the nonmetals to their left, and it follows directly from the falling ionisation energy.

Helium is a different kind of element

Any column's first entry tends to misbehave, but helium's departures are extreme even measured against that expectation.

It is the only substance that does not solidify at any temperature under its own vapour pressure; it takes about 25 atmospheres to freeze it. Below 2.17 K liquid helium-4 becomes a superfluid, flowing with zero viscosity, climbing the walls of its container as a film and conducting heat so efficiently that it stops boiling entirely. The rare light isotope helium-3 has a superfluid phase as well, but it appears near two and a half thousandths of a kelvin and arises by an entirely different route, through paired fermions rather than a Bose condensate.

Its supply position is also unlike anything else in the group. Neon, argon, krypton and xenon are extracted from the atmosphere by fractional distillation of liquid air, and the atmosphere is not going to run out of them. Helium is not: any helium released at the surface reaches escape velocity in the upper atmosphere and is lost from the planet permanently. Terrestrial helium is radiogenic — alpha particles from uranium and thorium decay, accumulated over geological time and trapped under the same caprock formations that trap natural gas — and it is recovered only from those few gas fields with a high enough helium fraction to justify the cryogenic plant. The United States built a strategic reserve near Amarillo, Texas, in the 1920s, legislated in 1996 to sell it off, and the resulting price distortion shaped the world market for two decades. Roughly a fifth of production goes into cooling superconducting magnets, mostly in medical scanners, for which there is no substitute at all.

What the column does for the world

The applications divide neatly into two kinds, and both are consequences of doing nothing.

Where a process must exclude oxygen and nitrogen, argon is the default: welding shields, single- crystal silicon growth, metallurgical degassing, and the fill gas in double glazing, where its low thermal conductivity is a bonus. It is cheap because it is the third most abundant component of dry air, and its abundance is itself a nuclear accident — almost all of it is argon-40, produced by potassium-40 decaying inside rocks and leaking out over billions of years.

Where the requirement is an electrical discharge, the heavier gases dominate. Xenon's discharge is close to daylight in spectrum and intense enough for cinema projection, flash photography and high-intensity headlamps; it is also a general anaesthetic and an efficient ion-thruster propellant, having the highest atomic mass of any gas that is easy to store. Krypton fills some incandescent and excimer applications. Neon's contribution to signage is real but small in tonnage terms.

There is a third use that depends on the chemistry being absent rather than merely convenient. Because a noble gas forms no minerals, any of it found inside a rock must have been produced there or trapped there, and its isotopic composition is not disturbed by weathering or metamorphism. Potassium–argon and argon–argon dating exploit that to time volcanic events, and the ratio of helium-3 to helium-4 in erupted gases distinguishes helium inherited from the solar nebula from helium generated by radioactive decay — which is how deep mantle plumes are told apart from shallow melting. The elements that refuse to participate turn out to be the most reliable witnesses.

Radon is the exception in every sense. It is the only member of the column with a public-health profile: it seeps from uranium-bearing rock into buildings and its short-lived decay products lodge in the lung, making it the principal cause of lung cancer among people who have never smoked in several national assessments.

Four things worth getting straight

• "Inert" is not the same as "noble". The older name was abandoned precisely because it turned out to be false for the heavier half of the column.
• The octet rule does not explain helium. Two electrons fill the first shell because n = 1 admits only an s orbital. The rule is a consequence of shell structure, not a cause of it.
• These gases are monatomic, not diatomic. They are the only elements that are gases as single atoms at room temperature, which is why their boiling points are so low relative to their masses.
• Oganesson is placed here by position only. Relativistic calculations suggest it would not be a gas at all, and nothing about its behaviour has been measured. Its column and its family are different claims, and only the first is established.