Everything we weigh, from aircraft to atoms, ultimately depends on a small block of metal stored in a Paris laboratory. The International Prototype Kilogram is a squat cylinder of platinum-iridium alloy, created at the behest of the 1st General Conference on Weights and Measures in 1889. It's approximately 39mm tall, but it's exactly 1kg in mass, because one kilogram is defined to be precisely what this lump of metal weighs. [video]
Six copies were made, and 40 further replicas dished out amongst the countries important at the time. The UK got copy number 18, and every 40 years or so it's taken back to France to be compared against the original. The rest of the time it's kept in a basement at the National Physical Laboratory in Teddington, the UK's chief arbiters of everything to do with measurement, so is very rarely seen.
Every two years the NPL holds an Open House event to showcase their work to the general public, including research and expertise across the entire spread of SI units. This year's event was held yesterday afternoon, so as a confirmed IPK groupie I made sure to go along.
It's wasn't especially visible, being housed inside a metal casing inside a glass belljar inside a locked storeroom behind a pane of glass, but it was good enough for me. I've been longing to see the National Standard Kilogram ever since I saw it in a television programme at school in the 1970s, and now here I was finally in its presence.
Pictured below is the automatic mass comparator used to compare the UK national standard with its copies. Up to four copies can be compared simultaneously to an accuracy of one microgram, that's 0.0000001%. These copies are then used for calibration elsewhere, and so on, until ultimately the Co-Op knows precisely how many pasta shells should be in a packet.
But there is a catch, a serious one, with worldwide scientific implications. Every time the National Standard Kilograms are compared to the International Prototype, they're found to differ in mass. It's only a very tiny change, due to minor amounts of wear and tear every time the kilograms are cleaned and compared, but a troubling distortion all the same. When the entirety of science depends on measurements underpinned by mass, it's by no means ideal to define the kilogram based on an impermanent block of metal.
A similar problem once existed with the International Prototype Metre, when the set of platinum-iridium bars knocked up in 1889 were found to differ fractionally from the original. Here's the UK's original National Standard Metre, in the wooden case, being watched over by its official Open House guardian. She was only allowed to handle it using thick gloves, and even the later copy (resting on the table in front) was treated with a certain degree of reverence. [video]
The solution was to move away from physical blocks and switch to universal constants. In 1960 the official definition of the metre was changed to correspond to 1650763.73 wavelengths of the radiation from a krypton atom, and in 1983 upgraded to the length of the path travelled by light in vacuum during a time interval of 1/299792458 seconds. This switched the problem to accurately defining the duration of one second, which is now determined by using a laser to measure emissions from a caesium atom. Here's the laser the NPL uses.
But finding a more accurate, independent means of measuring a kilogram has proved more difficult. Whatever methods anyone came up with, it transpired that weighing a block of metal was better. But a solution looks to be on the cards, thanks to a machine devised here at the NPL in 1975 by Dr Bryan Kibble. The Kibble Balance uses the force produced by a current-carrying wire in a magnetic field to balance the weight of a mass. The original balance was a messy beast made from large coils and wires, but over the last few decades the design has been refined to something smaller and more accurate.
What the Kibble Balance actually determines is another physical invariant, the Planck constant, to an accuracy of nineteen parts per billion. That's not perfect, but with several Kibble balances in operation around the world an average value can be taken, and that'll do nicely. The kilogram can then be defined using the equivalent energy of a photon via the Planck constant, throwing in the definition of the second and the metre for good measure. And with the kilogram now defined entirely via universal constants, nobody needs to rely on a block of metal in a basement any more.
The International Committee for Weights and Measures are meeting in November and are expected to agree to the fundamental redefinition of the kilogram, as well as the tightening up of the other SI units. If accepted, the entire metric system would become wholly derivable from natural phenomena for the very first time, under conditions reproducible experimentally anywhere on the planet. The anticipated date of the switchover is 20th May 2019, coinciding with World Metrology Day, after which physics will never be quite the same again.
With the next NPL Open House not due until a year after that, yesterday was the last opportunity to see the National Standard Kilogram while it still has inherent meaning. The clock is ticking... and yes, the NPL regulates the nation's timekeeping as well.
Other things I saw, played with and chatted about at NPL Open House included a hydrogen-powered car, a microwave anechoic chamber, the world's most accurate thermometer, a caesium fountain clock, a linear accelerometer, machines for measuring air pollution by nanoparticles, 3D optical microscopes, an underwater acoustic tank, a functioning Computer Aided Design System and Alan Turing's employee record. Here's the full 40 page programme, if you want to see what you missed. It was an entirely brilliant afternoon/evening for anyone of a scientific bent, and as well-frequented by schoolchildren as by adults. Stick a reminder in your calendar to check for tickets in April 2020 (or wait for Ian Visits to remind you).