The Wave-length of X-Rays

Abstract

IT is well known that determinations of the wavelength, λ, of X-ray spectral lines, which have recently been made by means of line gratings, do not agree with those found by crystals, the crystal values being 0.1 to 0.3 per cent less than the line grating values. In this use of line gratings the angles of incidence and diffraction have been small and have proved difficult to measure, but the accuracy attained in some of the observations of λ (for example, that of Backlin for aluminium Kα.) is probably not less than 1 in 1000. In the crystal method, relations of the form nλ = 2d sin θ and d3 = eM/ρF (e, electronic charge; F, faraday; M, molecular mass; and ρ the density of the crystal) are used to find λ, and the angle of reflection θ has been measured with high accuracy. Since M, ρ, and F are subject to smaller errors than e, the disagreement in the values of λ found by the two methods is usually attributed to an error of 0.3 to 0.9 per cent in the accepted value of e. If the precision of the grating method could be increased, these relations would be available to find the electronic charge more accurately than it is known at present.