Fluorine (F)

Fluorine took seventy-four years to isolate and injured or killed a significant fraction of the people who tried. The reason is a single number: fluorine is the most electronegative element there is, sitting at the top of the Pauling scale by definition. It wants an electron more than anything else in the universe, and having taken one it will not give it back to any chemical oxidising agent. The only way to extract fluorine from fluoride is electrically — and the product immediately attacks whatever apparatus generated it.

The chemists who did not get there

Humphry Davy attempted it and was injured. Thomas and George Knox, working in Dublin, were both badly poisoned; one of them was an invalid for a long period afterwards. Paulin Louyet in Belgium and Jérôme Nicklès in Nancy both died of hydrogen fluoride poisoning sustained in the attempt. Friedrich Wöhler was among those laid up for extended periods. The literature of nineteenth-century fluorine chemistry is unusually full of obituaries.

Henri Moissan succeeded on 26 June 1886, and his solution was as much engineering as chemistry: he electrolysed potassium bifluoride dissolved in rigorously dry hydrogen fluoride, at about −50 °C to slow the product down, in a vessel of platinum-iridium fitted with fluorspar windows because glass would not survive. The pale yellow-green gas he collected ignited silicon on contact, which was how he knew what he had. He received the Nobel prize in 1906, beating Mendeleev by a single vote in the committee, and died two months later at fifty-four.

Named by a man who never handled it, dated to a glass etcher

The credit line on the card above is an unusual pairing, and both halves need explaining.

The 1670 date belongs to Heinrich Schwanhard, a Nuremberg glass engraver who found that fluorspar treated with acid would etch glass. He had made hydrofluoric acid without knowing what it was, and the technique entered the decorative trades long before anyone had a theory of it.

André-Marie Ampère's contribution came in 1810, when he wrote to Davy suggesting that hydrofluoric acid was analogous to hydrochloric acid and therefore contained an undiscovered element related to chlorine. He proposed the name. He never attempted the isolation, which given the casualty list was a defensible decision.

The name itself comes from the mineral, fluorspar, which took its name from Latin fluere, to flow, because it had long been added to ores as a smelting flux to make slag run freely. There is a further step in the chain: George Gabriel Stokes coined "fluorescence" in 1852 after observing the blue-violet glow of fluorspar under ultraviolet light. The optical phenomenon is named after the mineral, and the element is named after the same mineral, so the two words are cousins rather than parent and child.

The bomb programme made it an industry

Fluorine chemistry was a laboratory speciality until the Manhattan Project needed it at scale. Separating uranium isotopes by gaseous diffusion requires a uranium compound that is a gas at manageable temperatures, and uranium hexafluoride is essentially the only candidate. Producing it meant making fluorine by the tonne.

It also meant finding materials that could contain the stuff. Every seal, gasket, valve packing and pipe liner in the enrichment plant had to survive hot uranium hexafluoride, and the answer turned out to be a polymer discovered by accident a few years earlier.

Roy Plunkett, at DuPont on 6 April 1938, opened a cylinder of tetrafluoroethylene and got no gas. The cylinder still weighed what it should, so rather than discarding it he sawed it open and found the interior coated with a waxy white solid: the gas had polymerised spontaneously. Polytetra- fluoroethylene was chemically inert, thermally stable and almost frictionless, and its first significant customer was the uranium programme. Non-stick cookware came later.

The strongest bond in organic chemistry, and its bill

PTFE's inertness comes from two things. The carbon-fluorine bond is the strongest single bond carbon forms — around 485 kilojoules per mole — and the fluorine atoms form a dense sheath around the carbon backbone that physically shields it from attack.

Those same two properties are why per- and polyfluoroalkyl substances do not break down. Nothing in the environment has the chemistry to attack a fully fluorinated carbon chain, so PFAS compounds released decades ago are still present, still mobile in groundwater, and now detectable in the blood of most of the human population. The name "forever chemicals" is a fair description of the bond energy rather than rhetoric.

The consequences have been litigated at length. Contamination around the DuPont plant at Parkersburg, West Virginia, led to a settlement funding the C8 Science Panel, whose epidemiology linked PFOA exposure to several conditions including kidney and testicular cancer. The US Environmental Protection Agency set enforceable drinking water limits for several PFAS in 2024 in the low single digits of parts per trillion — among the lowest numerical standards in any environmental regulation, which is itself a statement about how persistent these compounds are.

Why so many drugs have a fluorine in them

Somewhere between a fifth and a quarter of pharmaceuticals on the market contain fluorine, and more than half of agrochemicals do. This is not coincidence and it is not fashion.

A fluorine atom is only slightly larger than hydrogen, so substituting one for the other barely changes a molecule's shape and lets it bind to the same target. What it changes is metabolism. The liver disposes of drugs largely by oxidising carbon-hydrogen bonds; a carbon-fluorine bond in the vulnerable position simply cannot be oxidised, so the molecule survives longer and can be dosed less often. Fluorine also pulls electron density around, which medicinal chemists use to tune acidity and membrane permeability.

Fluoxetine, atorvastatin and ciprofloxacin are all built on this logic.

One stable isotope, and the most useful radioactive one in medicine

Fluorine-19 is the element's only stable isotope, and it is a gift to spectroscopy. Being 100% of natural fluorine, with nuclear spin ½ and a magnetic sensitivity close to hydrogen's, it gives NMR signals nearly as strong as protons do — and because biological tissue contains essentially no fluorine, a fluorine label produces a signal against a completely black background.

Fluorine-18 is the isotope that built positron emission tomography. Its half-life of about 110 minutes is the crucial specification. Short enough that a patient's radiation dose is modest and the activity is gone within a day; long enough that a cyclotron can produce it, a radiochemistry lab can attach it to a glucose analogue, and a courier can deliver the dose to a hospital an hour or two away. Regional PET networks are laid out geographically around that number.

The tracer itself, fluorodeoxyglucose, works because tumours consume glucose greedily and the fluorine substitution stops the molecule being metabolised past the first step, so it accumulates where the metabolic demand is. Almost every PET scan performed in the world uses it.

The public health argument that never ends

Frederick McKay, a dentist in Colorado Springs, spent the early twentieth century investigating why so many of his patients had brown-mottled teeth. The staining, eventually traced to naturally high fluoride in the water supply, turned out to come with a striking side effect: those teeth had very little decay.

H. Trendley Dean's epidemiology through the 1930s established the dose relationship — enough fluoride to protect, not enough to mottle — and Grand Rapids, Michigan, became the first city to fluoridate its water deliberately on 15 January 1945. The measured decline in childhood caries was substantial, and water fluoridation was subsequently described by the US Centers for Disease Control as one of the ten great public health achievements of the twentieth century.

The honest complication is that the counterfactual has changed. Fluoride toothpaste was not generally available in 1945 and is now near-universal, which supplies much of the same benefit topically. Studies conducted since then find a smaller incremental effect from fluoridating water, and how much smaller is a genuine and continuing scientific argument, distinct from the political one that has run alongside it since the beginning.