According to the formal definition of the SI Base unit of mass, the kilogram, it is stated that "The kilogram is the unit of mass; it is equal to the mass of the international prototype of the kilogram."


What was the measured mass of the prototype compared to?

Don't the chemical processes impacting this prototype make it obsolete for physicists today?

What I mean to say is how was it decided that this prototype is equivalent to the kilogram? Was this process arbitrary?

  • 2
    $\begingroup$ The original kilogramme was linked to the unit of length with 1 gramme being the mass of 1 cm$^3$ of water. Fairly soon it was found that this was not an accurate enough definition and a lump of metal was designated as being exact equal to one kilogramme. Much more here en.wikipedia.org/wiki/Kilogram $\endgroup$
    – Farcher
    Jul 17, 2016 at 13:38
  • $\begingroup$ According to Wikipedia, link "By definition, the error in the measured value of the IPK's mass is exactly zero; the IPK is the kilogram. However, any changes in the IPK's mass over time can be deduced by comparing its mass to that of its official copies stored throughout the world, a rarely undertaken process called "periodic verification"." Isn't this a massive drawback, considering quite a lot of things - chemical calculations, accurate mass measurement for experiments depend on the kilogram? There has to be a better standard that a 'metal lump' $\endgroup$
    – Zwolf
    Jul 17, 2016 at 14:08
  • $\begingroup$ That is why the CIPM have commissioned research into alternate definitions of the IPK. When these units were originally defined by French scientist they tried to use units based on systems which were eternal like the density of water (kilogrammes), the length of a quadrant of the Earth (metre), the time taken for the Earth to orbit the Sun (second), etc. $\endgroup$
    – Farcher
    Jul 17, 2016 at 14:14
  • $\begingroup$ A more fundamental definition would probably incorporate relative atomic masses of atoms. (1 atomic mass unit = 1.66053886 * 10^-27 kg). Something along this lines sounds more relevant in today's world. $\endgroup$
    – Zwolf
    Jul 17, 2016 at 14:23
  • $\begingroup$ Related: physics.stackexchange.com/q/32120/2451 and links therein. $\endgroup$
    – Qmechanic
    Jul 17, 2016 at 14:31

2 Answers 2


The kilogram is currently defined, as you well note, as exactly the mass of the International Prototype Kilogram, and by definition the mass of the IPK is $1\:\mathrm{kg}$ with zero error.

This definition took over from the previous one (the mass of $1\:\mathrm{dm}^3$ of water at $4°\mathrm C$ and sea-level pressure) because the previous one was hard to implement in a reproducible and stable manner, and it was easier to simply have a set reference standard. At the time the kilogramme des Archives was instituted, it coincided with the water-based definition to within the available experimental precision - it was just more stable and easier to use. (Later on there was a change to a second standard, the IPK, which simply reflected better technology to make the standard stable.)

However, as you note again, the mass of the IPK and its replicas is subject to change over time: they lose mass when they're polished, and they gain mass through corrosion from air. This is in fact a problem: when sister copies are compared to each other and to the IPK, their mass changes by a few tens of micrograms over the decades:

Image source.

This has some pretty profound consequences for all of metrology, because it means that mass standards are only stable to roughly the $10^{-8}$ level on decade-long timescales. In turn, that affects every measurement that depends on the mass standard to a finer precision.

Does this sort of thing bother metrologists? You bet it does. Unfortunately, until recently there simply wasn't any suitable alternative - there wasn't any suitable measurement of mass that was precise and stable enough to compete with the IPK. That feels weird (surely we can do better than $10^{-8}/\mathrm{decade}$) but that's how it is.

Note, however, the "until recently" - because now we do have better measurements available, and there is a process in motion to overhaul most of the existing SI. In particular, this includes changing the artifact-based SI to a definition that explicitly sets a value for the reduced Planck constant $\hbar$, and uses this to calibrate mass standards, primarily through the use of a watt balance. For more details on the proposed implementations for this, see What are the proposed realizations in the New SI for the kilogram, ampere, kelvin and mole?.

That said, the IPK is definitely not obsolete at the moment, despite its shortcomings - it will only become obsolete once it's replaced by a renewed definition of the kilogram in ~2018.


This definition isn't obsolete, by definition, because we are still using it.

There are, however, efforts underway to redefine the kilogram. One proposed definition is based on the watt balance, basically defining the kilogram in terms of the electrical units volt and ampere. Another proposed definition, called the Avogadro Project, is in terms of the mass of a specified number of atoms of silicon. The reason that this element was chosen is because an infrastructure with processes for creating defect-free, ultra-pure monocrystalline silicon already exists to service the semiconductor industry.

  • $\begingroup$ From my understanding, the Avogadro project will not be used to re-define the kilogram - it will be used for the mole. For more details see my linked answer here and the linked new-SI mises en pratique therein. $\endgroup$ Jul 17, 2016 at 19:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.