Palladium (Pd)

Palladium drinks hydrogen. A piece of the metal at room temperature will absorb several hundred times its own volume of hydrogen gas, taking the atoms into the spaces between its own, swelling by a few per cent, and giving most of it back again on warming. Thomas Graham described the effect in 1866 and called it occlusion. Nothing else does it on remotely the same scale, and it is the property from which palladium's most useful applications and its most embarrassing episode both descend.

Fifteen guineas for the entire world supply

In April 1803 handbills appeared in London shops advertising "Palladium; or, New Silver", a new noble metal, available from a mineral dealer in Gerrard Street. The bills were unsigned. No paper accompanied them, no institution vouched for the material, and the price was set by the dealer.

Richard Chenevix, an Irish chemist and Fellow of the Royal Society, smelled a swindle and bought the shop's entire stock for about fifteen guineas in order to expose it. He analysed the material and announced that it was not an element at all but an alloy of platinum and mercury, and the Royal Society awarded him the Copley Medal for the work.

An anonymous letter then appeared in Nicholson's Journal offering twenty pounds to anyone who could manufacture palladium from platinum and mercury in front of three named chemists. Nobody collected. Vauquelin, Klaproth and Gehlen examined the substance and none of them could support Chenevix's conclusion. In 1805 William Hyde Wollaston finally read a paper to the Royal Society admitting that the handbills, the reward and the metal were all his.

His reasons appear to have been partly commercial — he had a profitable business refining platinum and no wish to publish his methods — and partly a test of whether the scientific establishment would judge a substance on its properties rather than its provenance. It did not come out of the experiment well. Wollaston named the metal after 2 Pallas, an asteroid found by Heinrich Olbers only months before.

Cold fusion, and why palladium was the obvious candidate

On 23 March 1989 Martin Fleischmann and Stanley Pons announced at a press conference in Utah that they had achieved nuclear fusion at room temperature by electrolysing heavy water at a palladium cathode.

The choice of palladium was not arbitrary, which is precisely why the claim was taken seriously for as long as it was. Deuterium loaded into the palladium lattice reaches densities comparable to those in a liquid, and it was not absurd on its face to ask whether that could bring nuclei close enough to interact. What sank it was the absence of the neutrons and gamma rays that any real fusion at the claimed power would have produced in lethal quantity, and the failure of serious replication attempts. The episode is now a standard case study in announcing by press conference rather than by peer review.

The hydrogen affinity has genuine uses. Thin palladium-silver alloy membranes pass hydrogen and essentially nothing else, because the hydrogen dissolves atomically into the metal and diffuses through while other molecules cannot enter at all; this is how ultra-pure hydrogen is made for semiconductor fabrication. Hydrogen sensors exploit the resistance change as the metal takes up gas, and palladium-based storage has been investigated repeatedly, though the metal's weight per kilogram of hydrogen stored has always been prohibitive.

Diesel's disgrace, and a metal worth more than gold

Palladium and platinum can substitute for each other in oxidation catalysis, and for years the industry split them by fuel: platinum in diesel exhaust, palladium in petrol. That equilibrium broke in 2015, when the Volkswagen emissions scandal and the regulatory reaction to it drove buyers out of diesel across Europe. Petrol and hybrid vehicles took the volume, palladium demand rose against a supply that could not move, and the price passed gold in late 2018 and stayed above it for more than three years, peaking above $3,400 an ounce in March 2022 as the invasion of Ukraine raised the prospect of Russian supply being cut off.

That last point is the structural risk. Norilsk Nickel's operations in the Russian Arctic produce roughly 40% of the world's palladium as a byproduct of nickel and copper mining; most of the remainder comes from the South African platinum group mines, where palladium is a co-product of platinum. There is no third significant source, and no palladium mine anywhere exists for palladium alone.

Autocatalysts account for the large majority of demand. The second use was for years multi-layer ceramic capacitors, where palladium formed the internal electrodes; capacitor makers switched most production to nickel electrodes as palladium prices rose, which is a rare example of a precious metal being successfully designed out at scale. Dentistry, electrical contacts, and white gold alloys — where palladium replaces the nickel that causes contact allergies — take the rest.

The reaction that built the modern pharmacy

If palladium disappeared tomorrow, the greatest disruption would probably be to organic synthesis. Palladium catalyses the joining of two carbon atoms from separate molecules under conditions mild enough to leave the rest of the molecule intact, and it does so across an enormous range of substrates.

The Heck, Negishi and Suzuki reactions — recognised together with the 2010 Nobel Prize in Chemistry — differ in what the partner molecule carries, but all rely on palladium cycling between two oxidation states, picking up one fragment, then the other, and expelling the joined product. Before these reactions, building a biaryl bond meant harsh conditions and poor yields. After them, it became a step a graduate student runs on a Tuesday. A very large share of drugs approved in the last three decades has at least one palladium-coupled bond in its synthesis, and the same chemistry makes most organic light-emitting and agrochemical molecules.

Palladium is also the workhorse of hydrogenation. Palladium on carbon reduces almost anything; the Lindlar catalyst, palladium on calcium carbonate deliberately poisoned with lead, stops halfway and converts an alkyne to the cis alkene rather than the alkane, which is how a great many fragrance and vitamin intermediates are made.

Seventeen days, implanted

Palladium has six natural isotopes and one manufactured one that matters clinically. Palladium-103 decays by electron capture with a half-life of about 17 days, emitting low-energy X-rays that are absorbed within a couple of centimetres of tissue.

Sealed into titanium seeds the size of a grain of rice and implanted permanently into the prostate, it delivers its entire dose over a couple of months and then becomes inert metal that is simply left in place. Its shorter half-life than the iodine-125 alternative means a higher initial dose rate, which suits faster-growing tumours; the two are chosen between on exactly that basis.