The Width of X-ray Spectrum Lines

Arthur H. Compton
1922 Physical Review  
Width of X-ray Spectrum Lines.-(i) Four causes are discussed: (a) the width of the slit employed, .(&) angular imperfections of the crystal, (c) finite resolving power of the crystal grating, and (d) lack of homogeneity of the x-rays. The effect of the first two causes is independent of the angle of reflection and therefore the same for all orders and all lines; the effect of the third is shown to vary as i/sin 0 cos 6 and therefore to decrease as 6 increases up to 45 0 ; while the effect of
more » ... le the effect of the last |is equal to (<2X/X) tan 6 and therefore increases with 6. (2) Measurements of two tungsten lines, X 1.242 and 1.279 A, in the first four orders from the cleavage faces of calcite and rock-salt, are given. The fact that in both cases the width is least for the second order and greatest for the fourth order shows that both the third and fourth causes contribute measurably to the observed width, as well as the first two. Lack of homogeneity of tungsten x-ray lines computed from the above measurements comes out greater than 0.0007 X =*= .00013 \ which is about 0.5' for the first order. This cannot be explained as a Doppler effect or as due to the damping of the electronic motion; however it is in accord with the complexity of the lines as predicted by Sommerfeld's theory of elliptic electronic orbits. Mag., 41, 309 (1921).
doi:10.1103/physrev.19.68 fatcat:2akwzrgynje5pkhqkz7mjmw7fu