For nearly 130 years the definition of the kilogram has been tied to a lump of metal inside a Parisian vault that varies in mass from decade to decade. Now all that is about to change
Deep underneath the Parisian suburb of Saint-Cloud, in a vault that can only be opened by three people wielding three different keys, there lies a hunk of metal that is so crucial to the world of measurements that it only ever leaves its tightly-controlled environment to be cleaned and weighed.
For the last 129 years, this small cylinder of platinum-iridium has defined the weight of the kilogram. The International Prototype of the Kilogram, or IPK, does not just weigh one kilogram. It is one kilogram. If its mass goes up or down – and in the past century it has done both – then the definition of a kilogram goes up or down too.
Such is the importance of the IPK, there is even a sixteen-page manual devoted to the task of cleaning the mass. First, it’s wiped with a chamois leather that has been soaked for 48 hours in a mixture of ethanol and ether to remove any impurities. It’s then sprayed with steam from double-distilled water. The remaining water is then blotted away with filter paper, or blown off with clean gas, before the IPK is flipped over and the process starts again.
In total, it takes 50 minutes to clean the IPK, which measures just 39 millimetres in both height and diameter. “It's the world's mass standard and you don't muck around with it, it's really rather important,” says Ian Robinson, a researcher at the National Physical Laboratory who has been working on the redefinition of the kilogram for nearly four decades.
So what’s wrong with the IPK?
The problem, Brown explains, is that tying a definition to a physical object is nowhere near as useful as using physical constants. “Ultimately the traceability of the kilogram is through this single material artefact which is held at this international facility,” he says, but the IPK’s mass actually fluctuates by millionths of a kilogram as it accrues grime or loses mass when it gets cleaned. The aim of the conference was to switch the definition to something that can never change.
For most people, this precise definition of the kilogram isn’t going to change how we measure out flour or weigh ourselves, but for organisations such as the NPL, which promote accurate measurements across industries, getting precise about the kilogram is a huge deal.
All countries that are part of the Metre Convention – the 1875 treaty that established the created the international Bureau of Weights and Measures (known by its French initials, BIPM) – have their own copies of the IPK. Every 40 years these copy prototypes are carried – usually by hand – to Paris, where they are weighed to see how they compare to the IPK. The last weighing-in happened in 2014 to lay the groundwork for the upcoming redefinition, but there had only been three prior calibrations before then.
The UK’s copy is called Kilogram 18, and rests in a safe at the NPL’s headquarters in Teddington, South London. “It’s got a pretty nice life,” says Robinson, who, despite leading work on the redefinition of the kilogram, rarely works with Kilogram 18 itself. Like the IPK, Kilogram 18 – so-called because it was the 18th copy to be alloted by ballot and distributed – is kept under two bell jars filled with filtered air.
It’s these copies that national measurement laboratories, such as the NPL, use to set measurement standards and correctly calibrate delicate instruments. But the problem is that no copy is exactly the same mass as the IPK. Kilogram 18 is around 60 micrograms heavier than the IPK – overweight by an amount equivalent to a couple of small grains of sands. And that’s before you account for fluctuations in the mass of the IPK itself, as it picks up pollutants from the air or loses some of its alloy during cleaning.
“You're looking at stability of the mass, and in no circumstances can you get a mass that's exactly one kilogram, they all drift slightly,” says Robinson. “So what you're bothered about is that your own [mass] is the same difference as the IPK or you can predict that difference.”
This is a problem when you’re setting standards for industries such as pharmaceuticals, which regularly deal with measurements right down to micrograms. All metric mass measurements are derived from the kilogram and it’s extremely tricky to put in an accurate shot when your goalposts are moving every forty years or so. And that’s why Robinson and his colleagues have been working on a device that will allow scientists to leave behind the IPK for good.
Nearly 40 years ago, Robinson started working with his colleague, the late Bryan Kibble, on an instrument called a Kibble balance – a complex mess of scaffolding and wires that from a distance looks like it could be a spacecraft instead of an instrument that’s redefining one of our most important units of measurement.
Originally called the watt balance, but renamed the Kibble balance after Bryan Kibble’s death in 2016, the device was intended to solve a rather arcane problem with how the ampere – a unit that measures the flow of electrical current – is calculated. In 1988 Robinson and Kibble published a paper showing it was possible to use the Kibble balance to fix the definition of the ampere without having to use a complex instrument composed of two parallel lengths of wire.
Soon cracking the ampere, Robinson and Kibble realised that they could use the same device to leave the IPK in the dust. It could let them redefine the kilogram itself.
Here’s how it works. First you need to know that when you pass a current through a coil of wire, it generates a magnetic field. That’s exactly how loudspeakers work – they use electrical signals to switch an electromagnet on and off, causing a speaker cone to vibrate. What Kibble and Robinson realised is that you could balance this magnetic force against a physical mass – a little like resting a mass on your speaker cone and measuring how much electrical current you need to run through your electromagnetic to get it to move.
Through the clever deployment of a couple of physical laws – including one that won its creator the 1985 Nobel Prize in physics – the pair worked out that it was possible to express mass in terms of electromagnetic forces. Armed with this information, the pair worked out they could use the Kibble balance to calculate the Planck constant – an important number in quantum physics that relates to the amount of energy carried in a single particle of light.
Since Kibble and Robinson knew that they could calculate the Planck constant by starting with a mass, they also knew they could work backwards and derive a mass by starting with the Planck constant. And, unlike the IPK, the Planck constant does not fluctuate with time.
“You can then use any equation you like in physics to get yourself from the Planck Constant to mass,” says Robinson. “If you have the value of the Planck constant you can get a milligram, a kilogram, an atomic mass anything you like.”
But the kilogram isn’t the first of the seven base metric units to switch from a physical measure to a mathematical constant. A metre used to be defined as one ten millionth of the distance between the North Pole and the Equator. In 1983, the BIPM voted to instead derive the metre from the distance that light travels through a vacuum in a fraction of a second.
The kelvin, too, is getting an upgrade. Currently defined as a fraction of the temperature at which water can exist as a liquid, solid and gas, it will soon be defined in terms of the Boltzmann constant – a number that relates to the energy of particles in gas. It will join the kilogram along with the ampere and mole as the latest metric units to be defined in terms of fundamental constants.
If everything goes to plan, Brown says, no one outside of the scientific world will notice a difference when the kilogram switch is made next year. “It's very important that when we make the change, actually nothing changes,” he says. “We have to be absolutely sure that we're going to replace it with something that's better and something that is effectively the same size.”
That’s why Kilogram 18 is heading back to Paris for a final weighing-in session to compare it with the IPK in preparation for the switch over to the Planck constant. After that, standards will be derived from Kibble balances instead of from kilogram prototypes. This will spur a whole new industry of people working to refine the Kibble balance, Brown hopes.
For Robinson, who has spent nearly four decades working on the device that will replace that IPK, today’s vote was the conclusion of an extremely long and complicated process. After all, when you’re redefining one of the world’s most fundamental units of measurements, people tend to move extremely slowly.
“We were aiming at the kilogram in 1988-90, it's just that it's an extraordinarily difficult job,” Robinson says. But now, as he sets on his new task helping the world’s measurement bodies getting to grips with the Kibble balance, he’ll do so with the knowledge that behind the scenes of every weight measurement made in metric, his elegant solution to the kilogram problem is being put into flawless motion.