WORRIED about your weight? Spare a thought for metrologists, who are desperately embarrassed about theirs. There are other unhealthy signs for them to fret about, too. They have a dodgy-looking mole and their temperature is all over the place – and don’t even mention their lack of spark.
Metrologists are the folks who take care of the units and measures scientists use to size up, record and verify nearly everything around us – the international system of units (). It is the health of these units that is keeping many metrologists awake at night.
The problems start with a lump of metal in a vault in the suburbs of Paris, France. By international agreement, it is the only object in the universe with a mass of exactly 1 kilogram. Trouble is, the lump of metal is changing and nobody knows by how much. There is more at stake here than an unruly piece of metal. An errant kilogram also makes for an iffy mole, the unit chemists use to measure amounts of the particular elements and compounds they are working with.
Advertisement
To add to their woes, metrologists’ unit of temperature becomes rather vague above 1000 kelvin and the official unit of electric current, the ampere, is so impractical researchers have abandoned it completely. In short, the majority of today’s units are simply not accurate enough to meet the exacting demands of modern science.
This is anathema to metrologists, who pride themselves on measuring things with exquisite precision. So at a meeting in Paris in November they plan to discuss the biggest overhaul of the SI system for decades. The aim is to link the wayward units to the most dependable aspects of the universe; the properties of atoms, and fundamental constants such as the speed of light. This will be far from easy, though, and getting it wrong could undermine all of science.
The SI system is the hook upon which we hang all other measures. Whether it is the accuracy of your bathroom scales, the amount of aspirin in a pill or the reliability of the altimeter in the cockpit of a plane, there is a chain of safety laws and calibration certificates that lead back to just seven “base” units: the kilogram, metre, second, kelvin, ampere, mole and candela (which measures light intensity).
Metrology’s job is to keep the SI system in good shape, and to stay ahead of science’s insatiable appetite for precision, says Ian Mills, a metrologist at the University of Reading, UK. “You will never stop scientists devising more accurate methods of measuring things.”
Earth girdle
The metre is a case in point. It was first enshrined in a platinum bar, whose length was chosen so that 40 million such bars would match the Earth’s circumference. Establishing the correct length for the bar took French scientists Jean Delambre and Pierre M�chain seven years in the late 1700s. They did it by climbing a series of church towers, from Dunkirk to Barcelona, triangulating the distance between them and observing the position of the pole star from these points to work out the distance from pole to equator.
But there are no prizes for effort in metrology and even before the definition had been internationally adopted in 1889, some were already grumbling that the measure was too vague. Over a decade earlier, physicist James Clerk Maxwell had argued that units defined according to the dimensions of the Earth were inherently unstable because the planet’s surface is constantly wearing away and reforming. A better choice, he said, would be the features of “imperishable and unalterable and perfectly similar” molecules. This planted the idea of rooting measurement in the fundamental constants of physics and the properties of atoms, which to the best of our knowledge are the only truly stable and enduring features of the universe.
Nearly a century passed before Maxwell’s words were acted upon. The platinum metre-bar was shelved in 1960. In its place came a metre based on the wavelength of a line in the emission spectrum of krypton gas, which was in turn superseded in the 1980s by defining it as the distance that light travels in a specific time interval. In the 1960s, the SI unit of time, the second – originally linked to the Earth’s rotation – was refined by fixing it to the regular “tick-tock” energy states of a caesium atom. Not much has changed since.
Enter Mills and metrologists, Terry Quinn, Peter Mohr, Edwin Williams and Barry Taylor, who two years ago decided that metrology should finish what Maxwell had started. In a paper published in the journal Metrologia (vol 43, p 227), (metrologists have agreed that, for now, the candela can wait). To grasp why, the first place to look is inside that French vault. Housed at metrology’s spiritual home, near Paris, the vault contains one of science’s most valuable and most derided objects: the kilogram. Since 1889, this platinum cylinder has defined the mass of 1 kilogram. “Physicists find it completely incongruous that we’re using such ancient technology,” says Mohr of the in Gaithersburg, Maryland.
Gentle surprise turns to concern when researchers realise what this metallic curiosity means for their experiments and calculations. On occasions when the prototype has been compared with its closest relatives – the 40 or so duplicate kilograms kept in various labs worldwide – it has emerged that its mass has drifted by at least 100 micrograms, the mass of roughly half a sugar crystal.
Quest for perfection
It gets worse. The prototype and its duplicates could be drifting by much more, because there is no external benchmark against which to compare them: the prototype is the benchmark. “It’s sure to be changing. But we don’t know by how much,” says Quinn, ex-director of the BIPM and former custodian of the prototype.
Now metrologists are finally close to fixing the kilogram problem, and the solution could also bring two other errant SI units into line. Over the past 30 years, two possible ways to redefine the kilogram have emerged. The first is to link it to atomic mass by defining it as a fixed number of silicon atoms.
Led by , the experiments involve counting atoms inside perfectly polished silicon spheres. To do this, you measure the sphere’s volume and divide it by the volume of a single atom, which can be determined by peering inside using X-rays. Because each silicon-28 atom has a fixed, unchanging mass, a kilogram could then be redefined as a specific number of atoms. “The theory behind it is simple,” says Peter Becker, who leads the Avogadro Project at Germany’s metrology laboratory, the Physical Technical Agency (PTB) in Braunschweig.
Another advantage of this approach would be the unshackling of the mole, which relies on the kilogram for its present definition (see Table). The mole measures amounts of substances – in other words, the number of atoms or molecules. When it was defined in the late 1960s, scientists had no means of counting atoms accurately, so a single mole was defined as the same number of atoms as are found in 12 grams of carbon-12. This number (about 6 � 1023) is known as the Avogadro constant. Redefining the kilogram as the mass of a particular number of silicon atoms would mean the mole could simply be defined as a particular number of silicon atoms.
The second possible route to redefining the kilogram is less intuitive than atom-counting but is the clear front-runner. It involves measuring mass in terms of its equivalent in energy, using a sophisticated device called a . In essence, this is a sensitive pair of scales with the kilogram on one side and an electromagnetic field on the other.
Today metrologists use the Watt balance to measure precise values for the Planck constant, a fundamental constant of quantum mechanics that relates the energy of electromagnetic radiation to its frequency. However, the experiment can be turned on its head, allowing the Planck constant to become the benchmark to redefine the kilogram. “The Watt balance is a beautiful experiment to realise that definition,” says Mills.
The Watt balance can perform other tricks, too. Mills and colleagues point out that slotting the Planck constant into known equations yields a precise value for the Avogadro constant and so fixes the mole. What is more, it could bring the ampere in from the cold.
Till now, electrical researchers have simply sidestepped the SI system. To calculate current and other electrical measurements they use two quantum processes called the Josephson and quantum Hall effects. While these effects are fundamental and very precise, the units they deliver are independent of the SI system, though this isn’t widely known. “If you buy a high-precision voltmeter today, nobody will tell you it is not calibrated by SI volts,” says Mohr. In fact, “the electrical system is an island”.
Happily, the equations behind these quantum effects also define the relationship between the Planck constant and another fundamental constant, the “elementary charge”. This is the charge carried by one proton or one electron. So metrologists could take the Planck constant value from the Watt balance and use it to calculate a fixed value for the elementary charge. Counting the electrons that go by in 1 second tells you the current, says Mohr. At a stroke, this would anchor the definition for current to the rest of the SI units.
That leaves only the kelvin to fix. Temperature is currently defined using the triple point of water: the temperature and pressure at which ice, liquid water and water vapour coexist. Although the triple point can be measured to within 0.05 millikelvin, that precision is quickly lost when scaling up from the triple point to very high or very low temperatures.
The big idea with the kelvin’s redefinition is to ditch its reference to water and a single point on the temperature scale, and tie it to the Boltzmann constant, which relates thermal energy to mechanical energy. “The redefinition won’t change the world,” admits Michael de Podesta at the UK’s National Physical Laboratory in Teddington, “but it’s the right thing to do.”
Perhaps not surprisingly, big changes like these involve a plethora of committees and consultations. Ultimately, any change will mean winning the approval of hundreds of metrologists at an international summit that meets every four years at the BIPM. At the next meeting, in November, Mills and his colleagues plan to put their proposal to their fellow metrologists, with the aim of getting final approval for the changes at the 2011 meeting. “All of this is done by consensus,” says Quinn. “It’s very civilised – that’s why it takes so long.”
Reaching such a consensus will not be easy. Mills points to an ingrained conservatism in some corners of metrology. Right now, the main resistance is coming from mass metrologists charged with looking after their country’s copy of the kilogram, he says. “All around the world, the national metrology institutes are rather devoted to their local kilogram. They feel we are taking it away.”
Other metrologists say they can’t understand the big rush. Michael Gl�ser, a member of the Avogadro Project at the PTB, thinks scientists and wider society can live with the current SI system for at least another decade. “I don’t know why they are urging for new definitions in 2011,” he says. “We should wait until the experiments are more reliable.”
Even Mills recognises that there is good reason to tread carefully. “We don’t want a feeling in 10 years’ time that we made the wrong choice.” Ernst G�bel, president of the PTB, is even more cautious. “Making a mistake would blot metrology’s reputation and undermine experiments across science,” he says. “We have to be very conservative.”
Quinn, however, confesses to frustration at such foot-dragging. “It’s all a question of vision,” he says. “You can always think of arguments for not doing it now, but as regards science, it’s an obvious change to make. Not to redefine these units when you have the possibility of defining them in absolute terms seems to me to be perverse.”
“Not to redefine these units when we can do it in absolute terms seems perverse”
Any decision to overhaul the SI system in 2011 will hinge on the Avogadro and Watt experiments to redefine the kilogram. The bad news is that there is a major discrepancy between them. They are radically different experiments, but there is a known relationship between the two constants involved that can be used to determine whether their results agree. “One of them is wrong, and at the moment we do not know which,” says Mills. “We can’t redefine the kilogram until this discrepancy is resolved. We all agree on that.”
The hold-up will also stall any changes to the other units because, apart from the kelvin, the new definitions of the mole and ampere depend on these experiments, too. Besides, changing one unit at a time could be confusing for scientists outside metrology. “If we are going to make a major change, let’s do it all at once,” says Mills.
Right now, all eyes are on the Avogadro Project. In around two years, the collaboration will release an eagerly anticipated result that could dispel all doubt. A possible reason for the discrepancy between the experiments is that the silicon spheres used to count atoms last time round contained a mix of silicon isotopes of different atomic masses, which could have skewed the results. In the next few months, Becker and his colleagues will begin experiments on the purest silicon balls yet, taking atom-counting to a new level of precision (Â鶹´«Ă˝, 28 July, p 52).
If the tally of atoms in these spheres confirms the results for the Watt balance experiment, it would give the green light for a revolution that many metrologists have been waiting for since the 19th century. “I’m not sure what will happen if it doesn’t match,” says Becker. “We have to wait, we have to measure, that’s the only thing we can do.” A metrologist’s job, it seems, is never done.
