The automation of analytical equipment is proceeding at a rapid pace, particularly since the introduction of inexpensive microcomputer systems. Most of this equipment has one characteristic in common, that is, they produce digital spectral data. The usual method of recording spectral data has been the strip chart recorder. Strip charts require the hand encoding of position and intensities of the spectral lines. This requires that all of the lines be on scale or that the sample be run several times in order to amplify weaker lines. This problem is eliminated by recording the data digitally. Digital data can then be rapidly plotted on a cathode ray terminal to any desired scale. The user of digital data has access to a great variety of automatic data reduction programs.

This paper outlines the use of an IBM Series/1 small computer for instrument automation and data reduction for X-ray polycrystalline diffractometry and wavelength dispersive X-ray fluorescence spectrometry. The profile fitting method is used to determine 2θ, d and relative peak and integrated intensities in diffraction, and the fundamental parameters method (LAMA program) is used for quantitative analysis of bulk and thin film samples. The methods are precise and rapid.

Plastic deformation necessarily accompanies fracture even in high hardness steels such as ball bearing steel, though it is within a very shallow layer just under the fracture surface. The depth of the plastic deformation zone can he determined by X-ray measurement of half value breadth. However, in the case of high hardness steel, the half value breadth is not changed significantly by this kind of plastic deformation. On the contrary, residual stress on the fractured surface was found to change remarkably depending on the mode of fracture such as static, fatigue or delayed types.

A position-sensitive proportional counter (PSPC) with high counting efficiency has been made for stress analysis with low intensity X-rays such as microbeam X-rays.

This PSPC system has made it possible to measure the residual stress in a small region such as a fatigue crack tip in very short time compared with the measurement by standard diffractometer or film methods.

X-rays can be used to measure residual as well as applied stresses in polymers and reinforced polymeric composites, by diffracting from filler particles that are embedded before curing. We have investigated various fillers that exhibit suitable stress induced shifts of diffraction angle when embedded in uniaxial graphite fiber/epoxy composites. X-ray measurements of elastic strains in the particles are proportional to the corresponding composite strains, in agreement with the model of H. T. Hahn. Results indicate that the stress sensitivity (change in X-ray diffraction angle per MPa applied to the composite) increases in the order W, CdO, Hi, Ag, Nb, Al. With Al 333 + 511 reflections of CuKα1, the diffraction angle shifts 8.52 x 10-4 deg 2θ per MPa. Theoretical predictions are in reasonable agreement with this figure. The elastic range terminates at a yield point beyond which little or no further shift of the diffracted beams is seen for metallic fillers, but no such yield point has appeared with CdO particles. Oppositely directed shifts are seen after unloading from stress levels that have exceeded the yield point.

Since the measurement of residual stress by X-ray diffraction techniques is dependent on the difference in angle of a diffraction peak maximum when the sample is examined consecutively with its surface at two different angles to the diffracting planes, it is important that these diffraction angles be obtained precisely, preferably with an accuracy of ± 0.01 deg. 2θ. Similar accuracy is desired in precise lattice parameter determination. In such measurements, it is imperative that the diffractometer be well-aligned. It is in the context of diffractometer alignment with the aid of a silicon powder standard free of residual stress that the diffraction peak analysis techniques described here have been developed, preparatory to residual stress determinations.

Peeling is a surface fatigue failure of a roller bearing that consists of many shallow pits less than 10 pm in depth and cracks that link the pits. Peeling occurs rather easily on a smooth Surface when in contact with a rough surface under insufficient thickness of the lubricating oil film.

X-ray residual stress measurements on and under the contact surface after a peeling test revealed that the 2θ versus sin2ψ curve is not linear and that it curves depending upon the rolling contact condition and especially upon the existence of slip. Nonlinearity of the 2θ-sin2ψ) curve has been reported by Wakabayashi in a study of residual stress accompanying the grinding of soft steel and by Faninger in a study of residual stress due to rolling contact with annealed steel, but hot in the case of high hardness steel such as ball bearing steel. No complete explanation of this non-linearity has been made as yet.

A so-called “Vector Method” for the texture analysis of cubic materials was presented for the first time at this conference in 1976. Since then this method has been refined and applied successfully to non cubic-materials. It is shown in this paper that the Vector Method provides several advantages over series methods of texture analysis, the most important of which being the relatively small amount of experimental data which are needed for the determination of the entire crystallite orientation distribution.

Powder diffraction profiles of well crystallized compounds can be fitted to distributions of the type Iθ=I°(1 + k2x2)-n where k is a scale factor related to the half width of the profile. The value of n varies with the diffraction angle, 2θ, and is generally different for the low-angle and high-angle sides of the same profile. Limiting values of n for a specific Guinier camera-micro-densitometer combination are 1.2 ≤ n ≤ 2.3. Similar values are obtained for diffracto meter profiles after Ktt2 stripping. Line broadening due to departure from perfect crystallinity in the specimen affects the value of n a swell as that of k.

The above observations are interpreted interms of the convolution of a Gaussian with a Lorentzian distribution, the exponent n of the convolute being dependent upon the relative half widths of these two functions, expressed as the ratio bL/bG.

Synthetically prepared silicon nitride is one of the more promising ceramic materials for structural components of gas turbines. Typical material may contain a-silicon nitride, Si3N4 (which is believed to always contain oxygen and therefore, according to Grievson, Jack and Wild, is more properly written as Si11.5N15O0.5), β-silicon nitride, Si3N4, silicon oxynitride, Si2ON2, silicon metal, Si, and α-cristobalite, SiO2. Because the physical properties of the ceramic parts are dependent on their phase composition, it is essential that a technique be available for performing a phase analysis. An X-ray diffraction procedure has been, developed for the quantitative phase analysis of synthetically prepared silicon nitride. This procedure converts experimentally measured intensities of selected X-ray diffraction peaks to weight fractions of components using empirically determined intensity coefficients.