The X-ray technique for analyzing diffusion couples employs a monochromatic X-ray beam which impinges on a very fine slit, a holder for aligning the diffusion zone with the slit, and a scintillation counter for the high counting rates which have to be measured. Sample preparation, experimental instrumentation, and the application of the technique to the aluminum-nickel system will be discussed.

The applications of X-ray emission or fluorescent spectrography to chemical analysis have increased spectacularly in recent years, but little attention has been paid to the potentialities of X-ray absorption techniques. Monochromatic X-ray absorption-edge spectrometry, in particular, is most promising. The ultimate sensitivity of absorption-edge spectrometry probably will be less than that of fluorescent analysis, but this disadvantage may be outweighed by the convenience, economy, and absence of matrix effects with the former method. Both methods appear limited in application only to certain elements.

A pulse height analyzer coupled with scintillation and proportional counter detection has been found to permit an increase in sensitivity of absorption-edge spectrometry, primarily because controlled window widths may be utilized in determining transmitted X-ray intensities with a sealer. Further work has led to the development of a new rapid, convenient technique known as “differential pulse amplitude distribution (PAD) peak height analysis.” Work carried out during the development of the new method is described.

The camera was constructed to fill the need for a low-temperature instrument capable of yielding high spectral resolution and accurate lattice parameters. Specimen temperatures in the range -190 to 300°C are attainable.

A special feature of the camera is the film holder which is split along the beam axis into two symmetrical halves and which can easily be removed from the base without disturbing the cooling system or the specimen. The specimenholderis mounted on a cross-slide to enable specimens of large radius to be rotated off-center with the beam at glancing incidence to their surface. Finally, diffraction patterns from large metallurgical samples may be obtained by removing one of the film holders and employing the apparatus as a “half camera.”

Some applications of the camera to the thermal expansion of SiC and MnSe are described.

Equipment is described whereby comparative measurements were made on the relative absorption of air.helium, and vacuum for the softer radiations which are of interest in furthering the range of sensitivities of the X-ray spectrograph. Data on magnesium, aluminum, and silicon and shorter wavelengths are presented.

A light-weight portable X-ray spectrograph is described in some detail. The analyzing head, weighing only 18 lb, contains a goniometer, ratemeter, and Geiger counter detector. It also contains a special window-type X-ray tube with a gold target which has been developed specifically for this spectrograph, and which provides a very high efficiency of excitation of the sample. An associated power supply contains an rf source providing 40 kv at 60 μa together with the necessary voltages for the rectifier filaments, the X-ray filament, and the Geiger counter voltage. It will detect wavelengths from atomic number 23, vanadium, to atomic number 50, tin, and then using the L series from atomic number 58, cerium, to atomic number 92, uranium. Sensitivity and some applications are discussed.

A back-reflection focusing camera has been constructed in which instrumental errors, including those due to uneven film shrinkage, have been eliminated or very much reduced. This has been achieved by employing metrological techniques both in construction and calibration, The temperature of the camera can be controlled to 0.05°C. The form of the specimen is such that horizontal and vertical divergence cause no line shift. The X-ray source is a demountable semimicrofocus tube with an adjustable focus. The radiation is monochromatized by a bent and ground quartz crystal.

Procedures and methods employed to develop anX-ray emission spectrometer technique for determining the quality of decorative chromium plate are presented. Individual plate quality factors and their significance are discussed:

1. 1. Plate thickness

2. 2. Plate density

3. 3. Plate corrosion resistance

Geometric instrumentation changes to the General Electric XRD-5S spectrometer are detailed. These alterations have made it possible to take more complete advantage of the nondestructive nature of the fluorescent method, and have also yielded an increase in useful radiation intensity of 50%.

The unsuccessful attempts to employ commercially plated Hull-cell-type test panels as “standards” are described, along with the satisfactory application of fabricated standards of vacuum vapor-deposited high-purity chromium, laminated to spectrographically pure nickel foils. The merits and limitations of the resultant “absolute purity — absolute density” standard curve are shown and are related to other measurement techniques and accelerated corrosion test data.

The statistical fluctuations of photon counts arediscussedas a factor limiting the precision of the analytical result. Assuming a Poisson distribution, the theoretical standard deviation of the result can be calculated. While this prediction does not consider causes of variation other than the count statistics, it is useful in developing methods and checking instrument reliability. Practical examples using experimental results are given.

Line intensity and background measurements were made on the K lines for the elements with atomic numbers 16 to 60, L lines for the elements above atomic number 42, and M lines for elements above atomic number 80. Three general classes of samples were investigated: (1) infinitely thick, (2) microgram deposits, and (3) thin layers.

These studies show that longer-wavelength L radiation may be preferable to the K series lines from the same element in the range of elements with atomic numbers 42 to 60, In particular the Lα lines are more intense than the K series lines from Class 2 and 3 samples. With Class 1 samples the Lα lines are weaker than the K series but their line-to-background ratio is superior to the K series.

M series lines show little promise for spectrochemical analysis except for elements with atomic numbers 90 to 92; for example, with uranium samples in Class 2 and 3, the very high line-to-background ratio of the UMβ1 line may have limited application.

Elements with atomic numbers from 16 to 22 are more sensitive than expected due to the very high line-to-background ratios and the reduced collimation requirements in this longwavelength region.

A rapid method has been developed for the determination of vanadium and nickel in crude oils in concentrations above 5 ppm. Duplicate analyses for both nickel and vanadium require only one hour, and excellent precision is obtained. At a concentration level of 100 ppm of vanadium, the standard deviation is 3.1 ppm, and at a level of 30 ppm of nickel, the standard deviation is 1.6. The results are comparable to those obtained by spectrophotometric and optical emission spectrographic analysis.

Conversion of the crude oils to greases by the addition of aluminum stéarate minimizes the usual difficulties of handling liquid samples in the conventional-geometry fluorescent X-ray spectrometer. Scattered tungsten emission lines from theX-ray tube serve as internal standards to correct for instrumental and matrix variations.

A new sample tray that permits the insertion of four samples into the instrument at a time has been constructed. Each sample can be moved into position for analysis without opening the sample chamber, making it necessary to waitfor the helium to sweep the trapped air from the X-ray path only once for each four samples.

Semiquantitative results are adequate for the satisfactory solution of many problems involving mineral analyses, andfluorescent X-ray spectrography is gaining more recognition as a satisfactory method for performance of these analyses.

Successful applications of the method, in various instances are discussed to demonstrate a system involving minimum s am - pie preparation and the use of instrumental factors in establishing multielement calibration curves.

A method is described for the analysis of microscopic samples that utilizes a commercial electron diffraction unit for the excitation of fluorescent X-rays in a specimen and photographic recording of the resulting spectra by a small fluorescenceanalysis camera placed within the electron diffraction unit. Applications of this method to the analysis of precipitates extracted from steels are included in the discussion.

Manganese nodules obtained from the ocean bottom are a vast potential source of minerals. These nodules, whose composition depend on their source, have an average composition of about 20% manganese, 14% iron, and 0.5% each of cobalt, nickel, and copper.

X-ray spectroscopy was investigated as a method for the analysis of the nodules, and analytical control during the process of extracting and recovering the minerals.

The basic borax fusion technique of Fernand Claisse (R. P, No. 327, Department of Mines, Quebec) was used. The influence of matrix elements was determined and corrections applied. The binary systems in borax were compared with the curves calculated from pure metals using the methods and calculations of Beattie and Brissey.

The results indicate that the fusion method is sufficiently rapid and accurate for use in the control analysis of manganese nodules.

Since the development of the first practical electron probe microanalyzer by Castaing and Guinier in 1951, approximately 14 such instruments have been placed in operation throughout the world and at least 15 more are under construction. All of the existing instruments contain the following basic components: (1) An electron beam system, (2) a movable specimen stage, (3) a means for viewing the surface of the specimen, and (4) an X-ray analyzer. The integration of these components into a practical instrument requires certain compromises in the performance capability of the instrument. A critical evaluation of the various approaches to this problem is given and a new electron objective lens is described which results in fewer restrictions on the over-all performance. The basic design of an instrument employing this lens, which will be manufactured by the Applied Research Laboratories, is discussed.

The electron probe microanalyzer has found numerous applications in physical metallurgy. Some recent problems which have been studied with the microanalyzer constructed by the author at the California Institute of Technology include (1) determination of the penetration of chromium in chromized iron, (2) determinations of the concentration of nickel in taenite bands occurring in a fragment of meteorite from Canyon Diablo, (3) identification of intermediate phases in the as-cast structure of an alloy of titanium and copper, (4) identification of phases in the titanium—silver binary system, (5) determination of concentration fluctuations in a gold—copper alloy resulting from slow solidification, and (6) determination of the diffusion profile of nickel in single crystal sapphire.

The reflection X-ray microscope is a microscope in the conventional sense. X-rays are actually focused in the formation of magnified images of minute objects. Its principal component is a pair of concave glass mirrors whichhave been polished and figured with great care. In common with other optical devices, the reflection X-ray microscope suffers from various aberrations which limit its resolution. Until recently its narrow field and lack of contrast have made it impractical for the examination of histological or metallurgical specimens.

The X-ray microscope has the theoretical advantage of resolution when compared with the optical microscope. While it cannot compete with the ultimate resolution of the electron microscope, it has the advantage of being able to reveal the internal details of objects which are opaque to the electron microscope- Although there is no doubt that the X-ray microscope can contribute to our knowledge of structural details, perhaps its greatest potential lies in its analytical-chemistry aspects. Already a technique exists at unit magnification (contact microradiography) for weighing histological specimens whose cellular structures range from 10−12 to 10−14 g. The enlarged image provided by X-ray microscopy should greatly aid in a more extensive application of the technique.

The above advantages and possible areas of application are a source of continuous stimulation in the development of a reflection X-ray microscope which will operate close to its theoretical limit of resolution.

For some vacuum tube work, it was desired to establish a reliable and moderately rapid means of differentiation among several related ceramic bodies, mostly among the high aluminas. Matured composition from a crystallographic standpoint was of primary interest; source and batch constituents were sought only as additional information for interpreting results on the matured ware.

Since Al2O3 represented the major crystal phase in most bodies, the indicators of difference were the minor crystal phases. A number of X-ray diffraction techniques were tried, as were machines of different manufacture. Camera studies with powdered samples were generally ineffective. Useful data were obtained using the Geiger tube—goniometer setup with experimentally determined settings for sensitivity, circuit time constant, sweep rate, beam and receivingslitwidths, etc. Fixedposition peak-height counting was much slower, not practically better for peak-height determination, and introduced chances of missing peaks unless they were first determined by scanning and recording, thus further lengthening the analysis time considerably.

Satisfactory results have been obtained on a number of the bodies of interest. Generally, those bodies which are not differentiated by the X-ray technique outlined are not markedly differentiated in the vacuum tube application.

The advantages and accuracy of counter equipment for the measurement of the intensity of scattered X-rays from small single crystals are discussed and illustrated. Several precautions necessary to obtain reliable intensity measurements are discussed. The equi-inclination method of inspecting all accessible reflections is treated in detail. The practicality of the stationary crystal method for determining integrated intensities, in principle the most accurate and rapid method, is demonstrated. Suggestions for improving the accuracy and speed of collecting data are made.