|
I must say that I agree neither with what was said in the original commentary by R. D. Freeman (1), nor with the letter by P. Karol (2). The definition of units is not a matter of proposals by people who have rarely, if at all, been involved in precise measurements. It requires clear knowledge of most precise measurements in order to be able to describe the realization of the unit. Thus the kilogram will be changed when we shall be able to compare masses more precisely by counting atoms than by weighing against a prototype. This is likely to happen in the not too distant future. It will be the people who are developing the precise measurements who will be able to say what number of which atoms should be used to define the kilogram. It is more likely to be silicon than carbon. For instance the meter was initially defined as a ten millionth (round integer) part of the distance between the Pole and the Equator of Earth, then as the distance between two scratches on the prototype kept at BIPM in Sèvres (Paris), then in 1960 as a certain non-integer number (1,650,763.73) of wavelengths of a particular transition in the atomic spectrum of krypton. Only when it became clear that the uncertainty in the experimental measurement of the speed of light was limited by the shape of the krypton spectral line was the definition of the meter changed to what it is now: the distance traveled by light in vacuum in a fraction of a second. There is no integer magic in the definition, but the best estimate of the speed of light happened to be 299,792,458 m/s and that was subsequently chosen to be exact, with the definition of the meter dependent on it. If by any chance we had the kilometer as the base unit of length, it would have been defined as the distance traveled by light in the time interval 1/299,792.458 of a second. Equally, if the best estimate had been 10 times more precise, we would not have defined the meter as an integer part of the second. As it is at present, the physical quantity we express in moles (i.e., amount of substance, chemical amount, Stoffmenge in German, and a plethora of suggested but unaccepted names) has a dimension of its own and is not dimensionless (or of dimension one as purist would put it) like a number. Hence amount of substance is not a number (as correctly pointed out by Freeman above). The mole as a particular amount of substance is not a number either. The next thing in Karol’s letter that I fail to understand is the value 6.0221418 × 10 23. Where does it come from? The 1998 recommended value (3) for the Avogadro constant is 6.02214199 × 10 23. In effect, Karol is suggesting that we do not need a quantity of a different dimension from number to express large numbers of atoms, molecules etc. Indeed, we could do without amount of substance and the mole. However, it would make chemistry less clear: the gas constant would be the same as the Boltzmann constant, molar energy the same as energy, molar mass the same as mass, amount concentration, c = n/V, would become the same as number density, C = N/V, etc. The distinction between properties of individual atoms and molecules and those of macroscopic samples would be less clear. I do not think that the commentary by Karol helps in clarifying the issues on amount of substance and its unit mole, nor on the redefinition of the kilogram. Furthermore this Journal is not a journal for discussing metrological problems encountered in precise definitions of units. Metrologia would be more appropriate. I am afraid that readers of this Journal might get a wrong impression of how unit definitions are chosen, and what considerations have to be taken into account. Literature Cited- Freeman, R. D. J. Chem. Educ. 2003, 80, 16–21.
- Karol, P. J. Chem. Educ. 2004, 81, 800.
- Mohr, P. J., Taylor, B. N. Rev. Mod. Phys. 2000, 72, 351–495.
See Karol's subsequent Letter re: this letter.
|
Citation
