Vol. 24, No. 2
The Bureau International
des Poids et Mesures: Establishing Standards in the Physical Sciences
In a vault
on the outskirts of Paris, a cylinder of platinum-iridium sits in a
safe under three layers of glass. It is the kilogram, kept by the Bureau
International des Poids et Mesures (BIPM), which is the international
home of metrology. Metrology is the science of measurement, and it is
of fundamental importance to us all. It is essential for trade, commerce,
navigation, transport, communication, surveying, engineering, and construction.
It is essential for medical diagnosis and treatment, health and safety,
food and consumer protection, and for preserving the environmente.g.,
measuring ozone in the atmosphere. Many of these applications are of
particular relevance to chemistry and thus to IUPAC. In all these activities
we need to make measurements reliablyto an appropriate and known
level of uncertainty. The financial implications of metrology are enormous.
In the United States, for example, some 15 % of the gross domestic product
is spent on healthcare, involving reliable quantitative measurements
for both diagnosis and treatment.
The Convention of the Metre
> Defining Our Units in Terms of Fundamental
> Tomorrow's Kilogram
headquartered in the Pavillon de Breteuil, a historic and beautiful
chateau overlooking the Seine at Sèvres, on the outskirts of
Paris. Surrounding the chateau is a number of laboratories in which
the world's measurement standards are maintained and researched, and
in which results are collated between the National Metrology Institutes
of the world. BIPM recently opened a new chemical laboratory, which
specializes in the chemistry of the atmosphere.
Convention of the Metre
system was born at the time of the French revolution. At that time a
Commission on Metrication was established (of which the mathematician
Joseph Lagrange was the president and the chemist Antoine Lavoisier
was a member) to consider the best way to define the base units. The
commission recommended definitions for the metre as 10-7
of the distance from the pole to the equator, and the kilogram as the
mass of a cubic decimetre of water at 4 °C. Two members
of the commission were entrusted to "realize" the definition of the
metre by measuring the distance from the pole to the equator. This they
did, but it took them five years and led them through many adventures.
By climbing church towers and measuring the angle to neighboring towers
(with revolutionary battles taking place beneath them) and then walking
on to the next tower, they surveyed and determined the distance over
the ground from Dunkirk to Barcelona. Dunkirk is due north of Barcelona,
and from the latitude of each city determined astronomically, they were
then able to scale up this distance to obtain the arc distance from
the pole to the equator. They returned with a metre stick, which they
claimed was "according to the definition." The definition was never
realized a second time!
Bureau International des Poids et Mesures at Pavillon de Breteuil,
of the kilogram as the mass of a cubic decimetre of water was easier
to realize, but proved difficult to measure reproducibly, with accuracy
comparable to that of comparing masses using a balance. Thus, neither
of these definitions proved satisfactory. In 1875, the government of
France invited every technologically advanced nation to send representatives
to Paris to establish and to sign the Metre Convention, which is now
the basis of our internationally agreed upon standards of measurement.
The French government then presented the Pavillon de Breteuil, with
the small park that surrounds it, as international land with the status
of an embassy, for the use of the BIPM. In the years that followed,
the kilogram and the metre were redefined in terms of platinum-iridium
prototypes, which were kept at BIPM. Copies of these were distributed
around the world. Thus the first lesson of establishing a standard was
learned: it must be easy to reproduce, with an accuracy comparable to
that with which the quantity can be measured.
are now 51 countries that are signatories to the Convention of the Metre.
They support and govern the BIPM through the Conférence Générale
des Poids et Mesures, held once every four years. They appoint the Comité
International des Poids et Mesures, the International Committee, which
meets regularly to supervise the operation of the BIPM, and to appoint
the Directorpresently Dr. Terry Quinn from England. The International
Committee also takes advice from 10 Consultative Committees with expertise
in the various specialized fields. Members of the Consultative Committees
are drawn from metrology institutes around the world, and the Consultative
Committee for Units (CCU), in particular, has representatives from most
of the international unions such as IUPAC. CCU is responsible for advising
the International Committee on changes and developments to be made to
the International System of Units (SI), and particularly on choosing
the standards that define the base units of the SI.
Our Units in Terms of Fundamental Physical Constants
of the base units are continually in need of revision as we develop
more precise methods of measurement. The kilogram is, today, the only
base unit still defined in terms of a material artifact. We now strive
to define units in terms of fundamental constants or the properties
of atoms, whichas an act of faithwe believe to be invariant.
Thus, for example, the metre, defined in 1889 as the distance between
two scratches on the prototype metre stick, was redefined in 1960 to
be a multiple of the wavelength of the red krypton atomic line. This
change was made because it was found that distances could be measured
interferometrically in terms of the atomic wavelength more accurately
and reproducibly than the distance between the scratches on the prototype
stick could be measured. Similarly, the base unit of time, the second,
used to be defined as 1/86,400 of the mean
solar day. But even in the 19th century it was known that the rotation
of the earth was slowing down, due to tidal friction, and when quartz
clocks and then atomic clocks were developed, other irregularities in
the rotation of the earth were discovered. Thus, in 1967 the second
was redefined as the duration of 9,192,631,770
periods of the hyperfine transition in the caesium atom. This was both
more stable and easier to realize than the earlier definition, and it
can be realized with much higher precision using an atomic clock. Finally,
the metre was redefined again in 1983 to be the distance that light
travels in 1/299,792,458
of a second, for exactly similar reasons.
kilogram, kept by the Bureau International des Poids et Mesures.
these changes was made because the precision with which we could make
the measurements became greater than the precision with which we could
realize, or make use of, the previous definition. Today, we choose definitions
based on fundamental physical constants and atomic properties because
they provide more stable and universally available standards. They may
also generally be measured relatively easily and with high precision.
Each change of definition is chosen to maintain unchanged the previous
value of the standard, to the accuracy with which it had previously
important to note that the effect of each new definition is to set a
fixed value for some relevant constant. Thus, the present definition
of the second sets the value of the frequency of the caesium transition
to be exactly 9,192,631,770
Hertz, and the present definition of the metre sets the speed of light
to be exactly 299,792,458
metres per second. The 1793 definition of the metre set the distance
from the pole to the equator, but it is a definition that is difficult
to realize! The 1889 definition of the metre set the distance between
the scratches on the metre stick, and the 1960 definition set the wavelength
of the red krypton line. The present definitions of the base units,
and the general specification of the International System of units,
is presented in the SI Brochure. (The SI Brochure, 7th Edition,
T.J. Quinn and I.M. Mills, BIPM 1998, ISBN 9282221547)
definition of the metre can be realized to approximately one part in
1012 , and the present definition of the second to approximately
five parts in 1016 which is equivalent to one second
in 60 million years. You may well ask, what is the point? Why do we
need all this precision? One answer is that you will never stop scientists
from developing more accurate methods of measurement any more than you
will stop men wishing to climb Mount Everest or travel to the moon.
It then becomes necessary to develop standards that reflect the precision
of our measurements. A more pragmatic answer is that more precise methods
of measurement are invariably followed by new applications not previously
example is the development of the global positioning system (GPS) based
on signals from satellites. This is entirely dependent on the fact that
each satellite carries an atomic clock, used to broadcast its position
and time with the highest precision. From the difference between the
time signals received by the observer, the differences between the distances
from the satellites can be calculated, and from the known positions
of the satellites the position of the observer can then be calculated.
our ability to make more precise measurements has been advancing rapidly
in the last hundred years, metrology is not a new subject. Many of the
foremost 19th century physicists concerned themselves with this subject.
James Clerk Maxwell wrote in 1870, "If, then, we wish to obtain standards
of length, time, and mass which shall be absolutely permanent, we must
seek them not in the dimensions, or the motion, or the mass of our planet,
but in the wavelength, the period of vibration, and the absolute mass
of these imperishable and unalterable and perfectly similar molecules."
Today we accept the truth of this statement, but in making it Maxwell
showed great foresight, because at that time we had not yet developed
methods of measuring the wavelengths, frequencies, and masses of the
the race is on to find an acceptable new definition of the kilogram.
This is because we know that the prototype kilogram can change in mass
at the level of a few parts in 108 , due to surface chemistry
and wear and tear, over a period of months. However, we can actually
weigh the kilogram (i.e., compare the mass of two kilogram artifacts)
with a precision of better than one part in 1011 using the
best modern balances. A possible new definition would be to say that
the kilogram is the mass of a specified number of carbon-12 atoms, which
would set the value of the Avogadro constant. However, at present we
cannot quite realize this definition (i.e., weigh the carbon atom, or
determine the Avogadro constant) with the necessary accuracy. The present
best estimate of the Avogadro constant is uncertain to about one part
in 107, but this is obtained indirectly from other fundamental
constants. Direct measurements are based on the silicon crystal density
method, in which the spacing of the atoms in a single crystal sphere
of silicon is determined by X-ray interferometry, and the mass and volume
of the sphere is measured, so that the number of atoms in the sphere
may be calculated. The uncertainty in the Avogadro constant needs to
be reduced by at least one order of magnitude, and preferably two, before
we adopt an atomic mass as the definition of the kilogram. There is
also a possible electrical definition of the kilogram that may prove
to be preferable, realized by a Watt balance, in which a mass is balanced
against the force on a coil in a magnetic field, so that an electrical
energy is balanced against a mechanical energy. The decision on a new
definition of the kilogram will probably be made sometime in the next
10 years. The present best estimates of all the fundamental constants,
and further information on this subject, may be found in the paper by
Mohr and Taylor. (P.J. Mohr and B.N. Taylor, J. Phys. Chem. Ref.
Data 28, pp. 1715-1852 )
I was appointed to be the IUPAC representative on the CCU, and in 1995
I was appointed President of the CCU. It is a job that I find fascinating
meeting and discussing with some of the world's best scientists and
evaluating the latest experiments that improve our ability to make measurements.
The results impact upon every activity of IUPAC, most particularly on
the importance of making valid estimates of the uncertainty in the measurements
upon which we all depend. Another aspect of the work is its international
nature: the need for us all to speak the same language of science so
that we understand one another without ambiguity and use internationally
agreed upon names and symbols for the quantities and units involved.
Like many of the activities of the BIPM, these are also problems that
are close to the heart of IUPAC's existence, so that IUPAC and the BIPM
have strong common interests.
Mills has been the IUPAC Representative on BIMP since 1996. He is
a professor at the University of Reading.