A method for measuring the atomic number effect is developed using a sandwich sample technique. The depth distributions of x-ray production, ϕ(ρz) curves, have been measured for a zinc tracer in aluminum, copper, silver and gold matrices at 30, 25, 20 and 15 keV. The ϕ(ρz) curves were measured using a Cambridge Microscan 5 in which the electron beam is normal to the sample surface and the x-ray take-off angle is 75°.

Samples of the low concentrations of copper (∼1 Weight %) in aluminum, nickel, silver and gold were prepared. For each alloy system (for example, Cu - Al), three different concentrations of copper were prepared. The intensity ratios from the sample to the pure element (standard) for each system have been plotted against concentration. At such low concentrations of copper the relation between this ratio and concentration is linear. The slopes of the curves have been compared to the equivalent factors obtained as ratios of the area under F(ρz) curves for aluminum, silver and gold to the area under F(ρz) curve for copper, respectively. The F(ρz) curves are obtained from ϕ(ρz) curves; F(ρz) = ϕ(ρz) exp(-μρz csc ψ) where μ is mass absorption coefficient.

Comparisons are made between these experimental data and the current methods of calculating the atomic number effect.

A theoretical model and experimental measurements of the X-ray emission from thin films generated by an electron beam were analyzed and compared. Ionization correction factors for the thin films and for substrate effects were developed to calculate intensity ratios of film and film on substrate to pure bulk intensities.

The fraction of electrons transmitted through gold, silver, and copper with energy greater than the critical excitation energy was measured as a function of thickness and primary electron energy. The characteristic X-ray intensities from gold on bismuth, palladium, nickel and aluminum Substrates; copper on palladium, nickel and aluminum substrates; and silver on aluminum substrates were measured and intensity ratios to pure bulk intensities were compared to the model prediction. The prediction of the X-ray intensity ratios were approximated by a simple electron balance model for the ionization correction factors and the consideration of electron energy distributions did not alter the predictions significantly.

An empirical conversion parameter, following the bulk procedures of Ziebold and Ogilvie, was determined experimentally for alloy thin films of gold and silver on oxidized silicon substrates for films between 500 and 2000 Å. The prediction of the parameter from the intensity model is in fair agreement with the measured values. However, the empirical conversion parameter could prove to be useful for thin film analysis.

An electron probe Auger emission microanalyzer has been constructed. The instrument is composed of an electromagnetic focussing primary probe column and a cylindrical mirror electron energy analyzer. By using this instrument, Auger electron spectroscopy studies have been carried out in the modes of both emission microanalysis and emission micrograph. The feasibility o£ this method is investigated through its application to the study of iron surface.

Facilities for energy-dispersive x-ray fluorescence analysis have been added to an advanced version of the high-sensitivity ESCA (electron spectroscopy for chemical analysis) instrument described at the 1972 Denver Conference. Since the excitation mechanisms for electron emission and x-ray fluorescence are the same, the instrument's powerful source of primary x-rays is an asset to both types of spectroscopy. The geometrical arrangement of source, electron analyzer, and x-ray detector permits easy change from one mode of operation to the other without change of sample. While ESCA is valuable for the analysis of light elements and of surfaces apart from the bulk, x-ray fluorescence is useful for the analysis of bulk or substrate. The high excitation power makes the instrument useful also for trace analysis in solid or liquid samples. Modifications that could further enhance the usefulness of the instrument are described.

The design and performance of a spherically bent crystal x-ray spectrometer with variable curvature are given. A thin crystal with the diffracting planes parallel to the face is mounted on a vacuum chuck consisting of an O-ring in a brass mounting. A controlled partial vacuum is applied behind the crystal to cause spherical deformation of the lattice. Thus, rays from a point source on the focusing circle are diffracted to a line image also on the focusing circle. The differential pressure is automatically varied such that the source-to-crystal and crystal-to-image distances are equal and constant for all Bragg angles and hence the simple θ-2θ motion of a one flat crystal spectrometer is used.

The data are accumulated by a scanning proportional counter tube placed behind a vertical slit (perpendicular to the scattering plane) located at the image line. The fixed chord length is 22 cm and the instrument is designed to scan from zero up to 120° 2θ. Crystals are easily interchanged and the automatic vacuum regulator has sufficient flexibility to allow tailoring the spherical bending to crystals of materials of various thicknesses. The resolution is easily adjusted by either the size of the x-ray source or the width of the detector slit. The performance of the spectrometer has been evaluated by characteristic x-rays produced by various samples placed in a demountable x-ray tube. The main advantages of this three-dimensional focusing instrument are the very high signal-to-noise ratio and the very low levels of x-ray flux required.

Planar, Silicon diodes have been calibrated in the energy region 1.8 to 5.0 keV using low-energy, continuous-mode x rays. The spectral energy distribution of each of six x-ray spectra used in the calibration was measured using a non-dispersive Si(Li) x-ray spectrometer. X-ray spectral fluxes were measured using a xenon-filled, parallel-plate ionization chamber. A chopper-wheel and phase-sensitive detector combination was used to enhance the signal- to-noise of the silicon diode signals in order to measure the x-ray induced diode currents, ∼10-10) A., in the presence of diode dark currents of order, 10-5 A. Measured data were consistent with diode silicon dead layers of from 0.3 to 0.4 μm. Average agreement to within five percent was found between measured diode sensitivities and calculated diode sensitivities. The measured diode x-ray sensitivities are compared with independently-measured diode x-ray sensitivities of similar diodes.

In any petroleum product, a high sulphur content is considered undesirable. Sulphur in diesel fuels is claimed to aggravate problems of wear and deposit. With residual oils used in some metallurgical operations, the sulphur combustion products can have an adverse effect on the metal. In combustion equipment, the oxides of sulphur together with water vapour can lead to undesirable acid condensatlon on cool metal surfaces.

The more general problem is air pollution. Local and national legislation present the oil industry with a continuing demand for fuels of lower and lower sulphur content. Best use must therefore be made of crude oils of inherently low sulphur content. Desulphurising processes are comnonly used for distillates and are available for residual products. Such processes however tend to add significantly to produetion costs.

A PDF 15/40 computer wich ADC and CAMAC interfaces is used to control data collection apparatus, acquire data, and reduce data to determine the elemental composition of aerosol samples. The background is subtracted from each energy spectrum, peak centers are located automatically using a Gaussian correlation technique, peak multiplets are resolved with Gaussian fits, peak energies are compared with entries in a table of x-ray lines for possible identification, multiple identification of peaks and line interferences are resolved, and the elemental amounts are determined from the areas of the Gaussian fits.

An energy-dispersive X-ray spectrometer that (1) uses as the primary excitation source the power supply and tungsten X-ray tube from a conventional crystal spectrometer (General Electric XRD-6) and (2) uses as the secondary excitation source elemental metal foils that are readily interchangeable has been built and operated. The use of an X-ray tube with a high-voltage capability, 75 kilovolts max, enables the determination of elements with atomic numbers as high as 66 (terbium) to be based on the K series of X-rays; the highpower capability, 3.7 kilowatts max, enables a particularly intense beam of X-rays to be generated by the secondary source and hence, provides a particularly high detection capability for trace elements in a sample. An instrument that uses interchangeable secondary sources to irradiate the samples has several advantages over those instruments in which excitation is accomplished by direct irradiation with an X-ray tube: (1) the background radiation in the energy range where the X-rays of interest are measured is several orders of magnitude lower and is very uniform and (2) the energy of the excitation radiation can be closely matched to the absorption edges of the elements of interest in the sample.

In the application of the instrument, particular emphasis has been placed on the development of tectmiques that will enable an energy-dispersive X-ray spectrometer to be used as the detection instrument for quantitative elemental analysis. Methods for the determination of the individual rare earths, plutonium and uranium at the microgram level with an accuracy of ± 1% are outlined and for the determination of plutonium and uranium at the milligram level with an accuracy of ± 0.1% are proposed.