The Atomic Weight of Fluorine

Abstract

MESSRS. H. S. Patterson, R. Whytlaw-Gray, and W. Cawood have published¹ some preliminary results on a new revision of the molecular weight of methyl fluoride which seems to lead to the value 19.010 for the atomic weight of fluorine. I should like to point out that the references given in favour of this value are by no means trustworthy. In the first place, the work of McAdam and Smith² on the analytical ratio, sodium fluoride to sodium chloride, was a preliminary one. The only two transformations carried out give, after correction to a vacuum, fluorine = 19.019 (instead of 19.010). Smith and van Hagen³ conclude more recently that “the experience of these authors shows that the reaction in question did not proceed smoothly”, and that “the errors of experience—traces of water in sodium fluoride and conversion incomplete to sodium chloride—would have increased the value of the atomic weight of fluorine”. Germann and Booth⁴ have determined the normal density of silicon tetrafluoride, but, as Clarke has pointed out,⁵ “do not go so far as to compute from this figure the molecular weight of the fluoride”. By comparison with the deductions of Jacquerod and Tourpaian,⁶ Clarke obtains for the molecular weight 104.47 and hence for the atomic weight of fluorine 19.09 (not 19.010). Van Laar, on the other hand, assuming that the critical constants for silicon tetrafluoride as determined by Moissan are incorrect, has calculated the molecular weight as 104.13 and hence they find 18.995 for fluorine.