Ytterbium (Yb)

The second is currently defined by a microwave transition in cesium, and it is on its way out. Optical transitions oscillate roughly a hundred thousand times faster, and dividing a second into more ticks makes each tick easier to count precisely. Two elements lead the field of candidates to replace cesium: strontium and ytterbium.

The clock transition, and what it can measure

Ytterbium-171 has a transition at 578 nanometres that is very nearly forbidden — which is exactly what a clock wants, because a transition that hardly ever happens has an extremely narrow linewidth. Trap tens of thousands of ytterbium atoms in a lattice made of laser light, at a wavelength chosen so that the lattice shifts both energy levels identically and therefore does not disturb the transition being measured, and you have an oscillator whose frequency is reproducible to around one part in 10¹⁸.

Ionised ytterbium offers something even sharper. The Yb⁺ ion has an octupole transition at 467 nanometres whose excited state survives for something like a year and a half, giving a natural linewidth in the nanohertz range — the narrowest transition employed in any clock anywhere.

At those accuracies a clock stops being only a timekeeper. General relativity says a clock runs slower deeper in a gravitational potential, and at one part in 10¹⁸ the effect is measurable over a height difference of about a centimetre. Two ytterbium clocks connected by a fibre link can therefore compare the gravitational potential at two locations — relativistic geodesy, in which the instrument reads height by reading time. Comparing ytterbium's two ion transitions against each other, meanwhile, constrains whether the fine-structure constant is drifting, because the two respond to it differently.

Marignac's ytterbia was two elements

The element's own history is one of repeated subdivision. Jean Charles Galissard de Marignac took Mosander's erbia apart in 1878 and found a new component in it, which he named ytterbia after the Swedish village that had already supplied three element names — the quarry story is told on the terbium page.

Ytterbia held for twenty-nine years. In 1907 Georges Urbain in Paris demonstrated that it too was a mixture of two elements, and Carl Auer von Welsbach in Austria demonstrated the same thing independently and almost simultaneously. Urbain called his pair neoytterbium and lutecium; Welsbach called them aldebaranium and cassiopeium. The commission that settled it kept Urbain's naming, dropped his prefix, and the heavier half became lutetium — a decision that started an argument lasting forty years.

A full shell, and a metal that reports pressure

Ytterbium's 4f subshell is completely full, and a full subshell is stable enough that ytterbium holds on to a second oxidation state the way its lighter relative europium does. In the metal, that means only two electrons per atom go into the conduction band rather than three, and ytterbium ends up notably larger, less dense and lower-melting than the heavy lanthanides on either side of it.

The interesting part is what happens when you squeeze it. Compression narrows the gap between the 4f level and the conduction band, and beyond a certain pressure a 4f electron is promoted into it — ytterbium changes valence under load. Its electrical resistance rises sharply and predictably with applied stress as a result, which is the basis of the ytterbium foil stress gauge: a thin element embedded in a target whose resistance reports the pressure passing through it during a shock wave, on a timescale of microseconds. Few materials give an electrical signal that clean under those conditions.

The laser that cuts most of the steel

Ytterbium's biggest commercial outlet is as a dopant — a fraction of a per cent of it dissolved in the core of a glass fibre. Doped silica of that kind lases at around 1030 to 1080 nanometres and is pumped by diodes at 915 or 976 nanometres, and the gap between those two numbers is small: the quantum defect is under ten per cent.

Almost all of the pump energy therefore ends up in the output beam rather than as heat. Ytterbium fibre lasers routinely convert 30 to 50% of the electricity going in into usable light, against a few per cent for the lamp-pumped and carbon dioxide lasers they replaced. The fibre geometry also sheds waste heat over its whole length and guides the beam, so the output stays diffraction-limited at kilowatt powers.

Ytterbium's electronic structure is what allows the doping levels needed. It has essentially one excited manifold and one ground manifold, with no intermediate levels for an excited ion to absorb another photon into, so the parasitic processes that limit other rare-earth lasers simply have nowhere to happen.

The consequence is that industrial laser processing has been rebuilt around this element. Sheet metal cutting, welding, cleaning, marking, and the laser powder bed fusion machines that print metal parts are overwhelmingly ytterbium fibre systems. In the research world, ytterbium-doped crystals have also displaced titanium-sapphire for high-average-power ultrafast lasers, for the same reason: they can be pumped directly by diodes.

A nuclear spin of one half

Ytterbium-171 is unusual among heavy atoms in having a nuclear spin of exactly ½. That means its nuclear spin is a clean two-level system, well isolated from the electronic state and therefore from most sources of noise.

Quantum computing has taken full advantage. Trapped ytterbium ions provide the hyperfine qubit used in several commercial ion-trap machines, and neutral ytterbium atoms held in optical tweezer arrays encode qubits in that nuclear spin, with the same clock transition the timekeepers use available for readout. Ytterbium is one of very few elements that is simultaneously a leading candidate for the definition of the second and a leading platform for quantum computation, and it is the same nuclear and electronic structure serving both.

Seven isotopes, and a feedstock

Natural ytterbium comprises seven isotopes, and one of them is quietly important to medicine without ever appearing in a patient. Ytterbium-176, the heaviest, captures a neutron to become ytterbium-177, which beta-decays within hours to lutetium-177 — the therapeutic isotope behind two widely used cancer drugs.

Because the product is a different element from the target, it can be chemically separated to give lutetium-177 with no non-radioactive lutetium mixed in, which matters a great deal for the potency of the finished drug. Enriched ytterbium-176 is consequently a strategic material for the radiopharmaceutical industry, and demand for it is set by something that happens two decays downstream.

Ytterbium-169, made by irradiating the lighter isotope, has a 32-day half-life and emits soft gamma rays; it serves as a portable source for radiographing thin steel and light alloy where harder sources would overexpose the film.