Before picking up a kilogram with a pair of widemouthed forceps called lifters, Davis flicks off suspected specks of dust with a fine-tipped brush. (“My wife paints.”) He has modified the artist’s brush for his purposes by degreasing its fibers and covering its metal ferrule with plastic, “so if you accidentally hit the kilogram, you won’t scratch it.” On a balance precise to 10 decimal places, a scratch counts.
Davis tests the Irish kilogram in a sealed chamber against three BIPM working standards that are also made of stainless steel. He doesn’t weigh it against the platinum-iridium standard, since stainless weights are only one-third as dense, and therefore three times as large, displacing a much greater quantity of air. “You’d have to make an air buoyancy correction that would amount to almost a tenth of a gram,” he explains. “That is huge.”
Although Davis serves as the IPK’s official guardian, even he rarely sees the original prototype, which is too precious and vulnerable to damage to remain in constant use. Over the course of its century-plus lifetime, the IPK has emerged only three times to serve “campaigns” of active duty, most recently in 1988–1992, when it participated in a formal verification of all kilogram prototypes belonging to the 51 Meter Convention member states. On that occasion, however, the IPK itself was found wanting. Despite all the protective protocols and delicate procedures, it had mysteriously changed. No one can say whether the IPK has lost weight (perhaps by the gradual escape of gases trapped inside it from the start) or if most of the prototypes have gained (possibly by accumulating atmospheric contaminants). The difference is approximately 30 micrograms —30 billionths of a kilogram—in a hundred years. (Imagine 30 cents out of a $10 million stack of pennies.)
This alarming show of instability is driving global efforts to redefine the kilogram, so that mass need not depend on the safety or stability of some manufactured item stored in a safe. In fact, more than mass hangs in the balance, for the kilogram is tied to three other base units of the International System of Units (SI), namely the ampere, the mole, and the candela. Several more quantities—including density, force, and pressure—are in turn derived from the kilogram.
Other 19th-century artifacts of measurement have long since been retired in favor of fundamental constants of nature. In 1983, for example, the platinum-iridium bar that described the length of the meter yielded to a new benchmark: A meter is now defined as the distance light travels in a vacuum in 1/299,792,458 second (a second being the time it takes an atom of cesium-133 to vacillate 9,192,631,770 times between the two hyperfine levels of its ground state). These figures fail to give the average person any real feel for the quantities in question, but to a metrologist —one who specializes in the science of measurement—such equivalences rooted in physics have the advantage of permanence and reproducibility.
One invariant vying to replace the IPK is Planck’s constant, which could be determined via an experimental device called a watt balance. Alternatively, researchers may successfully express mass in terms of Avogadro’s number (which is tied to the unchanging mass of individual atoms), provided they can count the atoms in a crystal of silicon-28.
But neither of these complex, costly endeavors is likely to yield a new standard in time for the next meeting of the General Conference of Weights and Measures, scheduled for 2011. For now, the International Prototype Kilogram stands firm on metrology’s last frontier.
