In a previous paper it was shown that 0·0007 per cent, of 29 Cu and 0·0003 per cent, of 26 Fe could be detected in 30 Zn by atomic analysis by X-ray spectroscopy. This sensitivity is greater than that which was claimed by Noddack, Tacke, and Berg, who set the limit at about 0·1 per cent, for non-metals, and by Hevesy, who stated it to be about 0·01 per cent, for an element present in an alloy. It was later suggested by Hevesy that the high value of the sensitivity which we found might result from the fact that some of the alloys we had used were composed of elements of almost equal atomic number, and that the sensitivity would be smaller for a constituent of low atomic number mixed with a major constituent of high atomic number. To elucidate these disagreements we have made further observations of the sensitivity with elements of different atomic number and have investigated the conditions which can influence the sensitivity.
The Factors Determining Sensitivity
. The detection of one element in a mixture of elements depends upon the identification of its K or L lines in the general spectrum emitted by the mixture under examination. The intensity with which these lines are excited in the target (“excited intensity”) is proportional to the number of atoms of the constituent element excited,
i. e
., to its concentration and to the volume of the target in which the cathode ray energy is absorbed. The depth of penetration of the cathode rays is determined by the density of the target material and by their velocity (
i. e
., by the voltage applied to the X-ray tube). Schonland has shown that the range of homogeneous cathode rays in different elements, expressed as a mass per unit area, is approximately constant and is independent of the atomic number of the absorbing element. When their velocity is increased, the cathode rays will penetrate to a greater depth, and therefore a greater number of atoms of all constituents will be ionised. This will increase the “excited intensity” of the lines due to the particular constituent sought equally with those lines of the other elements present. The intensity of a line further depends upon the difference between the voltage applied to the X-ray tube and that necessary to excite the series. For these reasons, a high applied voltage is required for a high sensitivity.
The sensitivity of atomic analysis by X-rays
