Redefinition of the kilogram based on a physical constant

par Richard Davis (BIPM)

mardi 6 janv. 2009, 14:00 → 15:00 Europe/Paris

Auditorium

The International System of Units (SI) is similar to the MKSA system. The SI adopts the MKSA base units metre, kilogram, second and ampere but also includes three additional units: kelvin, mole and candela. There is a formal mechanism for amending the definitions of SI units to keep them current with technology. The SI units presuppose corresponding quantities (length, mass, time, electrical current, etc.). There has always been a synergy between the SI and the values of the fundamental constants of physics as recommended by CODATA [1]. Precision measurements of fundamental constants demand the utmost of the SI and, in turn, the fundamental constants can be exploited to improve definitions of the base units. At present, the second is based on a fixed value for the hyperfine splitting of a caesium atom. The speed of light in vacuum c is a fixed quantity (one can no longer measure the speed of light—at least not within the SI framework) and this fixed value of c effectively defines the metre. However, the definition of the kilogram has been unchanged since 1889: “The kilogram is the mass of the international prototype of the kilogram.” This means that the kilogram is the mass of a particular object that was manufactured in the late 19th century, and whose stability is by no means guaranteed. Mass metrology has traditionally involved high-accuracy comparisons—usually indirect—between objects of interest and the defining artifact kilogram. Because the ampere definition links electrical and mechanical quantities, the present artifact-based definition of the kilogram influences SI measurements not only of fundamental mechanical quantities such as the Planck constant h, the electron mass me, etc. but also of electrical quantities such as the elementary charge e, the Josephson constant 2e/h, the von Klitzing constant h/e2, etc. Ideally, one should redefine the kg in terms of a fundamental constant of mass such as the electron mass me, or in terms of a fundamental conversion factor between mass and frequency, such as h/c2. The impediment to doing this has been practicality: the experimental mismatch between 1 kg and me has been unbridgeable; the de Broglie-Compton frequency nu defined by 1 kg = (h/c2)nu seems to be an experimental nonsense. Nevertheless, precise experimental realizations of both these routes to the kilogram are nearly complete. Prospects for a redefinition of the kilogram in 2011 will be discussed. [1] CODATA 2006