A thermally tough, low-melting, low-viscosity glass composition is described that has been used as a fusion mixture in sample preparations for X-ray fluorescence analysis.

The fusion mixture is cast into a glass disk which, after annealing, is used directly for the X-ray measurements. Problems arising from disk surface imperfections have, in the past, been minimized by either prolonged polishing of the glass surface, or grinding the disk and pressing the powder into a peltet. This practice has been eliminated through the incorporation of appropriate internal standard elements into the glass fusion mixture. This approach also minimizes interelement effects and errors due to instrumental drift.

A variety of corrosion products have been successfully analyzed by this method. Results will be given.

The analysis of dry powder samples for magnesium, sodium, and fluorine by X-ray and electron excitation has been studied. As in the case of heavier elements, the form of chemical combination influences the elemental sensitivity; sensitivity changes due to self-absorption can be adequately predicted using published absorption coefficients. Where both absorption and enhancement effects are possible, selection of X-ray target or excitation potential can. eliminate the enhancement problem. Matrix effects were found to be extremely variable and unpredictable. Finally, X-ray and electron excitation results are compared for the three elements in a series of geological samples. Efficiency of excitation was far better for electron excitation, but limits of detectability were lower for X-ray excitation due to significantly lower backgrounds.

A method for analyzing solutions using a sample surface directly exposed to the primary X-ray beam is discussed. This method eliminates the need for the conventional Mylar covered liquid cells. The advantages of this method are the elimination of the scattering of the longer wavelength X-rays and the absorption effects due to the Mylar covering, thereby giving significant improvement in peak-to-background ratios and peak intensities for the light elements. This increased sensitivity can be used to improve the limits of detectability for light elements in solutions, broaden the range of practical elemental determinations, and reduce the counting time for any light element analysis in liquids.

A new liquid cell, developed for this technique, provides easily repeatable setting of target-to-sample distance and simplified preparation and handling of samples. A comparison between results obtained with conventional method and this uncovered sample surface method is made for typical solution applications.

In view of the importance of obtaining optimum conditions for the X-ray fluorescence analysis of light elements, an investigation has been made of the effects of varying the X-ray tube target, the target take-off angle, the tube voltage, and the tube window thickness. The effects of these parameters have been observed by measurement of the intensity of fluorescence of two light elements using the Ray max 60 demountable tube and vacuum path spectrometer. The results obtained are compared with those given by theoretical calculations based on consideration of the relevant parameters; good qualitative agreement has been obtained. It is shown that a high primary X-ray intensity is obtained with a high target take-off angle, a low angle of incidence of the electron beam on the target, and an optimum setting of tube voltage. It is further shown that the most suitable target to use for the fluorescence analysis of light elements is markedly dependent on the thickness of the X-ray tube window.

Gas flow proportional counters for the detection of soft X-rays were introduced about ten years ago. These detectors offered the advantages of high sensitivity, good energy discrimination qualities and the ability to handle high counting rates. Since that time they have been used for ultra-soft and harder X-rays as well, both as detectors in standard spectrographic instruments and as energy discriminating instruments themselves in various so-called nondispersive applications. Depending upon the particular instrumental application, however, their use has led to considerable complication of the associated electronic circuitry in order to realize their advantages. For the most effective use of these counters (and of sealed proportional counters as well) it is necessary to have a clear understanding of the effect of various design parameters and operating conditions on their performance. The dependence of the shape of the pulse height distribution on the operating voltage, pressure and counting rate is described as a function of the energy of the radiation detected and the nature of the gas. Stability requirements on. counter tube high voltage supplies and operating pressure are discussed. Shifts of pulse height distributions toward smaller pulse sizes with increasing counting rate are described and the dependence of these shifts on the various parameters and on wavelength are discussed. Techniques for eliminating the shifts and the implications of these techniques for the associated electronics are described.