The fist-sized lump of platinum alloy that defines the kilogram turns 125 this month, but it may be its last significant birthday.

The kilogram is the last SI unit still defined by an artefact rather than a physical constant, represented since September 1899 by a 4cm cylinder made of 90 per cent platinum and 10 per cent iridium. Called the International Prototype Kilogram (IPK), it is stored in the basement of the International Bureau of Weights and Measures in Sèvres, France.

Over the last 125 years, however, comparisons between the IPK and its six official copies have shown their masses slowly diverging – presumed to be due to wear and chemical interaction with the atmosphere – prompting calls to redefine the unit in terms of a fundamental constant of nature.

This November, at the General Conference on Weights and Measures (CGPM), metrology experts are expected to ratify a final roadmap to redefine the kilogram in terms of Planck’s constant – used to describe the size of quanta, or units of energy, in quantum mechanics – by 2018.

According to Stuart Davidson, head of mass metrology at the UK’s National Physical Laboratory (NPL) in Teddington, the benefits of such a redefinition are twofold: “One, the guaranteed long-term stability of the unit and two, everyone, theoretically, has access to it.”

Aside from determining mass, the kilogram is essential for defining the newton, which in turn directly or indirectly defines the pascal, the joule and the watt among other SI-derived units. But unlike other SI units such as the second or the metre, it has proved tricky to nail down.

“There isn’t a simplistic equivalent in the kilogram’s case,” said Davidson. “Most have been to do with weighing atoms, but that’s obviously not an easy thing to do.”

The breakthrough has come from advances in the watt balance, a device consisting of a set of scales in a vacuum with a coil on one side to which a current is applied to support a weight on the other. If gravity is known then it is possible to use measurements from the experiment to determine the value of Planck’s constant.

“The difficulty is it’s all state of the art measurements,” said NPL’s Ian Robinson, who has dedicated his career to the development of the watt balance. Progress in measurement technology means the uncertainty of 20 parts per billion required to fix the value of the constant is close to being achieved.

By linking the experiment to the IPK and fixing the value of the constant, the uncertainty will be transferred to the artefact and the watt balance can be used to determine the weight of a kilogram.

Confirming the accuracy of the measurement requires comparison with a parallel effort to measure the Avogadro constant – the number of atoms in a mole and an indirect measurement of Planck’s constant – using highly polished spheres of purified silicon with a mass of one kilogram.

According to Robinson, it was only last year that these two methods began to agree at the level needed to be confident in the results, but with the two efforts now converging a redefinition looks imminent.

“At the moment there is a single point of failure in the system with the kilogram in Paris; there isn’t another one,” said Robinson. “In the new system anyone who wants to join in can, as long as they can afford to build a watt balance or do the Avogadro experiment.”

The change will benefit scientists who require very low uncertainties such as calibration laboratories and pharmaceutical companies, but it will also have implications for electrical engineers. The redefinition process will involve fixing another constant known as the elementary charge – the electric charge carried by a single proton – which will result in the redefinition of the ampere.

Combined with fixing Planck’s constant, this will also fix several dependent constants used to define the volt and ohm since the late 1980s, a move that will return these units to the SI system rather than the conventional electrical unit system.

“A lot of other constants become fixed because the Planck and elementary charge will have zero uncertainty. It will help a lot of people out across science,” added Robinson.