A series of FORTRAN IV programs have been written to aid in qualitative x-ray fluorescence (XRF) analysis. These programs have been implemented on a DEC PDP 11/34 computer with an RSX-11M operating system and access the NIH elemental data base (1). The programs and their application to XRF analysis in our laboratory will be described.

Results of an experimental X-ray study on 96 synthetic glasses show that the 2θ positions of glass diffraction maxima have an inverse relation to SiO2 concentration in silicate glasses. This relationship is the basis of a new technique for semiquantitative determinations of SiOa in silicate materials by X-ray diffraction methods. Samples to be examined are fused and the resulting glass scanned from 12 to 40° 2θ using CiiKa radiation. The mean 2θ position of the diffraction maximum is a measure of the SiOs content of the glass. Calibration curves for both weight and molecular percent SiO2 vs. 2θ are presented in this report. The technique requires only small, unweighed amounts of sample for analysis ; it is simple, rapid, and utilizes standard diffraction equipment without modification. Its accuracy, at present, allows SiOa determinations to within ±1 to 4% of the actual concentration.

A very accurate 8X8 matrix approach to dynamical theory of X-ray diffraction in which fewer approximations are made than in the classic vonLaue approach, is described here. The method is related to the very general matrix method of Kokushima and Yamakito, and is particularly suited to numerical solution with a computer. It can be used to solve problems in ideal, undistorted crystals with high precision even at near grazing incidence without special consideration of refraction or external reflection. It is also easy to apply to problems where periodicity of oblique Bragg planes varies in the direction normal to the surface. Such strain may be induced, for example, by variation of composition with depth. Certain problems wherein simultaneous diffraction by two sets of Bragg planes occurs can also be treated by this approach.

Recently aluminum nitride (A1N) has been intensively studied as a promising material for production of hybrid integrated circuit substrates because of its high thermal conductivity, high fjexural strength, and nontoxic nature. The estimated theoretical value of its thermal conductivity at room temperature is 320 W/mK, but it is strongly degraded by the introduction of oxygen. The measured values vary from 30 to 260 W/mK, Therefore, in production of this material the reduction of oxygen contamination is of paramount importance.

The penetration depth of 1-12 keV electrons in most materials is less than one micron and the characteristic soft x-rays that are produced can be used to identify the elements present in the surface. Varying the energy of the incident electron beam enables the depth of analysis to be controlled.

Soft x-rays often exhibit large 'chemical effects' (changes in peak profile and peak position) which can he correlated with chemical changes. A study of such effects for each element present in the sample surface, as a function of electron-beam energy, can in some cases, permit changes in the chemical state (valency - coordination number-spin state etc.) to be determined as a function of depth.

Such analyses can be carried out either in a conventional x-ray spectrometer in which the x-ray tube has been replaced by a gas-discharge source, or in a spectrometer in which the sample is bombarded with electrons from a normal electron gun. In this paper these techniques are outlined and some applications reviewed:- the analysis of oxide layers on aluminium and steel, the analysis of aluminium-nitride layers produced by MOCVD on gallium arsenide, the analysis of silica fiims (with added boron and phosphorus oxides) on silicon and the analysis of zinc-oxide films on glass.

It is shown that the integrated intensities of diffraction lines from calcite and aragonite powders prepared by precipitation vary markedly, due to variations in sample packing efficiency arising from different degrees of polydispersity in the particle size distributions. Since prolonged grinding to equalize initially divergent distributions changes the polymorph composition, the packing effect imposes the principal limitation on the precision of the x-ray method for certain types of calcium carbonate preparations.

This paper examines some of the concerns regarding the development of an algorithm for the refinement of X-ray diffraction profiles. The object of the algorithm is to provide a time efficient method of refinement through the choice of a suitable profile function and optimization technique.

Seven profile models were tested using a least-squares error criterion for refinement. Profile parameters were refined using non-linear Gauss-Newton, Marquardt and Simplex algorithms. The profiles were refined on a pattern digitally collected from an NBS 640A silicon sample.

The results of this study indicate the repetitive function evaluations are not necessarily the time consuming step in the profile fitting process. As the number of parameters needed to evaluate the profile and the number of points in the profile increases, the time required to perform the mathematics in the Gauss-Newton and Marquardt algorithms increases. Although the Simplex was most memory and time efficient, our Gauss-Newton optimization algorithm provided a more consistent set of refined values which were not as dependent on the initial estimates of the parameters.

The most favorable results were obtained by using the split Pearson VII profile with the alpha 2 reflection fixed in position and intensity with respect to the alpha 1 reflsction. This method generated the lowest residual error and was found to avoid some problems resulting from the alpha 1, alpha 2 line overlap.

Texture means the orientation distribution of the many crystallites in a polycrystalline sample with respect to a reference system on the sample, for example the rolling direction in steel. Texture has to be distinguished from preferred orientation observed in laboratory-made powder samples. Preferred orientation is just a nuissance in the experiment and must be corrected in the ensuing analysis, for example by the formula given by March and published by Dollase Preferred orientation we find commonly in samples containing plate-like crystal flakes. This is again to be distinguished from a non-random particle distribution in the powder samples prepared for diffraction experiments. This effect is always present, also for example in NBS silicon, even if only to a. very small degree (see for example Will et. al). It must and can easily be corrected for in the final least-squares calculation.

Photoelectric diodes with aluminum photocathod.es have been calibrated in the energy range from 0.1 keV to 1 keV with a multielement Henke X-ray tube system using regenerative monochromatic filters. The calibration is performed at six energies: (1) Be - (0.109 keV) , (2) C - Kα (0.277 keV), (3) Ti – Lα1, 2 (0.452 keV) , (4) Fe – Lα1, 2 (0.705 keV), (5) Cu – Lα1, 2 (0.930 keV), and (6) Al – Kαl, 2 (1.487 keV). Several different 99.9%, (1000 series) aluminum foils were calibrated for use as cathodes. In addition, the average energy per electron ion pair, W, has been measured for methane at these same energies.

A compact X-ray energy spectrometer has been developed consisting essentially of a radioisotope X-ray source, a lithium-drifted silicon (or germanium) detector and a small computer. Interchangeable sources enable efficient excitation of K X-rays from Na to U and L X-rays from about Ag to U. Energy resolution of K X-rays from adjacent elements down to Na is possible. Depending on the source and the part of the spectrum examined, the characteristic X-rays from up to about 15 elements can be simultaneously excited and measured, for either qualitative or quantitative multi-element analysis. The computer stores detected spectra and performs simple data processing such as peak recognition, background subtraction, peak integration, ratioing and solution of linear equations.

The analysis reported in this paper is the determination of V, Cr, Fe, Co, W and Mo in tool steels and is intended to illustrate the capabilities of the radioisotope X-ray fluorescence analysis technique, and the instrument, for multi-element analysis of a system having fairly complex interelement effects.

A 100 mCi Pu-238 source was used to excite the K X-rays of V, Cr, Fe, Co and Mb and the L X-rays of W. The count time used was five minutes per sample. Data reduction consisted essentially of peak integration, background subtraction and solution of sixth order linear matrices of a modified Criss-Birks type. The 36 matrix coefficients were determined using six standards, and were then used to analyze seven other analyzed specimens which were treated as unknowns. The measured values of concentration were in very good agreement with the quoted values. An iteration technique was employed to reduce errors in the matrix inversioiis.

To study the relationship between crystal growth conditions and resultant crystalline perfection, large nickel single crystals more than 12 cm long and 2 to 3 cm in diameter have been grown from the melt by the Czochralski method. Unlike semiconducting materials, one cannot easily thin metal crystals, without straining them, for the purpose of applying ordinary Lang topography with μL∼l where kinematical scattering in imperfect crystals is a good approximation. This situation with metal crystals necessitates the use of dynamical diffraction effects in imperfect crystals to permit sample crystals to be thick enough to demonstrate their imperfections as in the bulk.

Stress gradients in the direction of the surface normal influence the shape of the interplanar spacing “d” vs. sin2ψ (where ψ is the specimen tilt) plot obtained from the surface layers of a specimen.(1-3) If the gradients are caused by the shear stresses σ13, σ23, the “d” vs. sin2ψ plot exhibits “psi-splitting”, that is the “d” values measured at positive ψ tilts are different from the values measured at negative if tilts. (2) If the shear stresses σ13, σ23, are zero, but the normal stress σ33 exists in the layets penetrated by the x-ray beam, “d” vs. sin2ψ, plot exhibits curvature. (3) Various methods have been proposed to obtain the complete stress tensor from split or curved “d” vs. sin2ψ data, and all of these methods require the “unstressed” lattice spacing “d0” for their calculations.

Analysis of serai conductor material and associated integrated circuits (IC) is imperative for ensuring quality products. Currently, routine circuit testing is dominated by measurement of the optical and electrical material/device properties through final device performance and parametric testing.

Characterization of the crystal microstructure still is not considered a routine process test. Structural characterization usually is based on double-crystal rocking curves, x-ray topography, or a combination of these techniques.

The phase and elemental distributions on zirconium alloys containing 5, 10, 25, 35, and 50 at.%. yttrium, oxidized in dry air at 1200 and 1600°C, have been studied by X-ray diffraction, fluorescence, and electron-microprobe techniques. Zirconium-yttrium alloys, which are entirely solid at 1200°C, and zirconium 5 and 25 at.% yttrium alloys, which aire two-phase solid plus liquid at 1600°C, form a multiphase scale of ZrO2, Y2O3, and ZrN. The phase distribution varies with yttrium content, while the average metal composition remains constant throughout the oxide scale and base metal. In contrast, zirconium 35 and 50 at.% yttrium alloys, which are entirely liquid at 1600°C, form a single-phase Y2O3 layer.

The mechanism of air oxidation of solid zirconium-yttrium alloys, up to 50 at. % yttrium, is one of anion diffusion to the oxide-metal interface with little or no diffusion of the metallic elements. The oxidation of liquid alloys differs in that, as a result of yttrium diffusion to the oxide-liquid interface, Y2O3 is the only oxide formed. Ternary Zr-Y-O isotherms are presented to illustrate the phase distribution and mode of formation.

X-ray spectrochemical analysis is a versatile technique that can be used to determine considerably more than just the elemental composition of a sample. X-ray lines and bands are, in many cases, influenced by the state of chemical combination of the element whose spectrum is being investigated. Significant effects due to changes in bonding are seen in the K and L spectra of the low atomic number elements which fall in the 20 to 90 Å region. Results using primary excitation, a stearate crystal, and a flow proportional counter are shown for the K spectra of boron, carbon, and nitrogen. Chlorine and sulfur L spectra from some simple metal compounds of these elements are also shown.

We describe an adaptation of photon excited x-ray fluorescence analysis which is optimized for the analysis of small samples. A fine focus x-ray tube is used in conjunction with small diameter detector collimators in order to focus on a small sample volume with as high sensitivity as possible. Sample areas of less than 1 mm diameter can be analyzed with ppm detectability. In applications involving the analysis of human hair samples, a minimum detectable limit of 10 ppm Hg can be realized in a 1 mm long segment of a single hair in a counting time of 200 seconds. Simultaneous measurements of the sample mass can be obtained from the intensity of the incoherent scattering. An automated x-ray fluorescence analysis system using the technique for the scanning of elemental profiles in such hair samples will be described.

The relation of the X-ray small-angle scattering parameters to the results of transmission electron microscopy for spherical Guinier-Preston zones was investigated. Experimentally, the Guinier radius is found to equal (<R7>/<R5>1/2) and the Porod radius to equal <R3>/<R2>. These results were theoretically predicted by Baur and Gerold, A log normal distribution was found to be in close agreement with the experimental data. The use of this distribution allowed the size-distribution curve of spherical Guinier-Preston zones to be constructed from the X-ray parameters and from a knowledge of the metastable miscibility gap. This approach is used to follow the evolution of size distribution in both Al-Ag and Al-Zn alloys. The application of this approach to nncleation and growth as well as redistribution processes is discussed. Growth paths are established from the size distribution evolution curves through an Nv> construction. A brief discussion of the technique is included.

The purpose of this work was to develop accurate calibration standards which were fully characterized in terms of uniformity and concentration using fundamental measuring methods. Three similar sets of vacuum deposits were commercially made, each set containing the single deposits CuS, KCl, CaF2, Cr, Fe, Cu, RbNO3, SrF2, MoO3 , BaF2, and Pb. Thickness variations in each deposit were measured with PIXEA (proton induced x-ray excitation analysis) measurements taken at 6 to 8 positions along the deposit diameters. Relative elemental concentrations on corresponding deposits from each set were measured using multiple XRF intercomparisons. One set of deposits was destructively analyzed at the National Bureau of Standards with isotope dilution thermal ionization mass spectrometry (IDMS) in order to calibrate the remaining sets of vacuum deposits. The calibrated deposits were compared with standards from two commercial sources. For seven elements heavier than chlorine there was an average deviation of 13.5% between the calibrated deposits and the commercial standards. Disagreements as large as 15% were observed between standards from the two commercial suppliers.

This paper deals with some recent investigations on beryllides, borides, carbides, nitrides, aluminides, and germanides of transition metals as well as with combinations of these with special reference to hafnium and uranium compounds. The results will be discussed within the general field of refractories excluding the class of typical oxides. An intense study has revealed the influence of metalloid impurities (for instance carbon, nitrogen, and oxygen on high-melting borides and silicides or nitrogen and oxygen or carbides, respectively). Very often small amounts of such metalloids introduced by preparation techniques, like hot-pressing, sintering, and powder-reaction methods may change the constitution of the system.

The analysis of mixtures of phases which produce complicated composite x-ray powder patterns is greatly facilitated by use of our profile fitting method and the technique of applying it is illustrated with a five-compound mixture. Profile fitting gave higher precision in the determination of the reflection angles and Intensities and resolved overlaps in a much shorter time than with other methods. If the reference standards are obtained with the same precision, a smaller error window width can b e used in the search/match procedure.