The Lachance-Traill, and Lucas-Tooth-Price matrix correction equations/functions for XRF determined concentrations can be graphically interpreted with the help of three dimensional graphics. Statistically derived Lachance-Traill and Lucas-Tooth-Price matrix correction equations can be represented as follows:

1

where:

Ci -elemental concentration of element “i”

Ij -X-Ray intensity representing element “i”

Ai0 -regression intercept for element “i”

Ai -regression coefficient

Zj -correction term defined below

2

Ai0, Aj , and Zi together represent the results of a multi-dimensional contribution.

li, Ci, and Zi can be represented in three dimensional Cartesian space by X, Y and Z. These three variables are connected by a matrix correction equation that can be graphed as the function Y = F(X, Z), which represents a plane in three dimensional space. It can be seen that each chemical element will deliver a different set of coefficients in the equation of a plane that is called here a calibration plane. The commonly known and used two dimensional calibration plot is a “shadow” of the three dimensional real calibration points. These real (not shadow) points reside on a regression calibration plane in this three dimensional space. Lachance-Traill and Lucas-Tooth-Price matrix correction equations introduce the additional dimension(s) to the two dimensional flat image of uncorrected data. Illustrative examples generated by three dimensional graphics will be presented.

For the decomposition of complex multiplets, which consist of the fluorescent K-series of the lanthanides in the energy range from 30 keV to 60 keV, the procedure of digital filtering of the spectrum data was elaborated using the libraries of real-shape lines. This elaboration was applied to the software for energy-dispersive x-ray fluorescent spectrometers based on Si(Li) and HP Ge planar detectors.

Taking into account the fact that the real shape of line is changed under different count rate, the procedure of spectrum processing has the following principle peculiarities:

1. - three levels of library of real-shape lines, i.e. the level of high content of fluorescent elements, middle and low contents;

2. - the regulation of parameters of digital filters;

3. - use of the nonlinear least squares technique;

Also the estimation of quality of this investigation is given.

On the basis of experimental data and theoretical calculations, the contributions of scattering to the intensities of the K-lines of the lanthanides are considered.

This effect studied for the “one-element plus buffer” targets of medium thickness when these targets were excited by gam ma-radiation with energy 59.54 keV of radioisotope 241Am or 88.009 keV of radioisotope 109Cd.

The experimental dependencies were improved by theoretical corrections calculated by the formula for a medium thickness target. This formula takes into account the contributions to the measured intensities which arc caused by the subexcitation of fluorescent atoms by primary radiation scattered in the sample. Good approximation of the experimental data by the theoretical calculations was obtained.

It was found that contributions to the measured intensities of a lanthanide's atom fluorescence, which can be as much as a few percents, can lead to the subexcitation by scattering radiation of the targets with low concentrations of the fluorescent elements.

Spectrum processing is a critical step in energy-dispersive X-ray fluorescence analysis, with very strong influence on the final performance of the method. Spectrum processing includes two major procedures: (i) preliminary processing (which deals mainly with reduction of random and systematic noise); and (ii) evaluation of peak parameters (most often peak area, and in some cases also peak position, amplitude and width). Preliminary processing solves two main problems: (i) search and correction of erroneous channel contents (outlier values); and (ii) random noise reduction. Several procedures for search and correction of outlier values have been proposed for Raman, gamma and X-ray spectra.

Accurate and precise evaluation of the peak parameters of the recorded X-ray spectra requires the use of the correct peak shape. Usually, it is assumed that the shape of the X-ray peaks is a Voigt function, i.e., convolution of a Lorentzian distribution function of the emitted X-ray lines with a Gaussian response of the spectrometer, with additional terms for description of the tailing. This assumption is not exactly correct, because it employs a probability density function for description of a discrete signal, while, actually, the measurable parameter is the probability, i.e., the integral of the probability density function over the channel width. In trie present work, the distortion in the peak shape caused by the discrete sampling of the X-ray spectrum is studied and the significance of the introduced errors is evaluated. An improved fitting procedure for evaluation of the peak parameters is proposed.

Total reflection X-Ray FKiorescence (TXRF) originally was developed for trace analysis of small residues but has become a widespread method for measuring trace surface metal contamination on semiconductor substrates. It is estimated that approximately 100 TXRF instruments are in use in the semiconductor industry worldwide, and approximately half that for residue analysis in analytical laboratories. TXRF instrumentation is available today for reaching detection limits of the order of 109 atoms/cm2. This review emphasizes some of the more recent developments in TXRF for trace analysis, in particular with the use of synchrotron x-ray sources (SR-TXRF). There is some promise of reaching 107 atoms/cm2 detection limits for surface analysis of semiconductor substrates.

Advancements in trace clement analysis require improvements in both the signal-to-noise ratio and accurate background correction. With a sequential spectrometer, one can obtain detection limits of around 0.1 ppm for medium to heavy Z elements. Conditions can be individually optimized for each element, for example, selection of filters, collimators, crystals and background subtraction. The disadvantage is that the analysis time may become “long” if many elements are to be analyzed. This long exposure time can lead to the deterioration of some samples.

The cone penetrometer is a well-established method for exploring soil types whereby a smalldiameter pipe with a hardened cone tip is pushed hydraulically into the ground. The system can be equipped with sensors to detect soil contamination. We have investigated the development of an x-ray fluorescence (XRF) sensor to be deployed via this system.

The principal uncertainties in the application of XRF to metals detection in the cone penetrometer are the detection limits and the accuracy of a sensor built in such a confined geometry. A laboratory mock-up was constructed to investigate the performance of such a sensor. An x-ray tube operated at very low power was coupled to an electrically cooled Si(Li) detector. The x-ray path lengths were kept short and an aperture and incident x-ray filter similar to those appropriate for a penetrometer sensor were used.

Spectra were collected on a series of Standard Reference Material soils from the National Institute of Standards and Technology consisting of soils with various levels of metal contamination. The detection limits were determined for the metals present in the soils and were comparable to the solid waste regulatory limits. Quantitative measurements were compared to the NIST certified values both without matrix correction and using the fundamental parameters correction method. Results were generally within 10% of the certified values.

Using x-ray reflectometry the surface of float glass samples of different Fe content and thickness were studied. The patterns needed an oil/water layer to be modeled correctiy. The maximum Sn concentration was found to reside in the top 10-20Å of the material and the Sn concentration decreased into the bulk. It was also found that increased Fe content decreased the Sn concentrations in the surface and bulk of the glass.

In the early years, the element range covered by the Wavelength Dispersive X-Ray fluorescence (WDXRF) technique was restricted by the quality of vacuum, crystals and detectors in the system. The lightest element detectable was Aluminium (Z = 13). The area of light element analysis had obvious potential for extension over the years. It created the barrier for XRP coverage of the Periodic Table because the naturally softer radiations lay out of range of the technique. In the middle of the 1980's, new technology became available for the manufacture of synthetic crystals with artificially created 2d spacings to suit the X-ray wavelength range of interest. These hyered synthetic microstructures (also called “multilayer crystals”) allowed more sensitivity to be achieved in this part of the spectrum, and the bottom element limit soon moved down towards lighter elements.

Field portable x-ray fluorescence (FPXRF) has been used as a tool in the environmental industry to characterize metals contaminated soils for about 10 years. Initially, due to the level of sophistication of the equipment (ambient temperature detectors and empirical calibration spectral processing), and the complexity of soil matrices, it was generally believed that the instrumentation could only produce semiquantitative data. As the level of instrument sophistication increased (cooled solid-state detectors and Fundamental Parameters), that perception of FPXRF changed. However, not all users have access to the latest, state-of-the-art instrumentation. Work performed at Leadville, Colorado provides an example of how properly used older instrumentation can produce data of good quality (Kuharic, et al., 1993).

The influence of Quality in the laboratory is now all pervading. Increasingly there is a climate which requires laboratories of any reasonable standing to be accredited to the appropriate Quality standard. Within ICI, a number of quality initiatives had manifest themselves- such as GLP1 (Good Lab Practice), BS-57502,TSO-90023, or NAMAS4 (National Measurement Accreditation Service). The latter is a UK body whose accreditation is particularly appropriate to research, analytical and contract laboratories.

Specimens for x-ray fluorescence analysis (XRFA) were prepared by adding dry gelatine (10%) to the analysis solution, homogenizing the mixture and cooling for 20 minutes. Thus, a compact resilient mass could be formed with the required shape and size; the roughness of the surface was determined by the roughness of the surface on which the specimen was formed, much the same as highly polished. Various calibration methods can be applied in the XRFA of a variety of materials if such specimens are used.

The volume compression of BaNd2CuO5 has been measured to 8.7 GPa utilizing a diamond anvil high pressure eel and energy dispersive x-ray powder diffraction. The pressure dependence of the volume of the tetragonal unit cell is linear and follows the equation V = -1.6I5P + 260.622 determined by a least-squares fit of the data. Compression of the unit cell parameters of BaNd2CuO5 exhibit isotropic behavior within our experimental error. From the observed data, the bulk modulus of BaNd2CuO5 is determined to be 161 ± 6 GPa and the Young's modulus is estimated to be 193 ± 6 GPa. No evidence of a pressure-induced phase transformation in BaNd2CuO5 was found in the pressure range studied.

Pyroaurite and sjogrenite belong to the group of sandwiched lamellar metal hydroxides which have a fixed metallic ions MII:MIII ratio for a particular class. Their crystal structure consists of positively charged metal hydroxide blocks intercalated with negatively charged interlayers. The atomic positions for the interlayer are definite for a particular class. The exact chemical formula of the pyroaurite class is determined from crystal structure analysis to be MII6MIII2(OH)16-CO3-4.5H2O; it crystallizes in the space group Rm with a = 12.4376 Å and c = 23.4126 Å. Sjogrenite, MII6 MIII2(OH)16-CO3-4H2O crystallizes in the space group P63/mcm. The crystallogiraphy and structural relationship between these classes are discussed. Previous discussion on these compounds did not give any conclusion on the exact chemical formula and the atomic positions.

An in situ XRD technique employing a diffractometer equipped with a high temperature camera was used to investigate the annealing behavior of nanoerystalline copper powder produced by mechanical milling. Specimens were annealed isothermally for 12 h at temperatures between 480 and 770 K. The diffraction data was analyzed using a single-profile Fourier analysis technique. The activation energy for diffracting particle growth was determined to be 0.45 eV/atom.

The mechanical alloying process continually deforms, cold welds, and breaks apart metal powder particles. During the process of mechanical alloying elemental crystalline powders can produce an amorphous alloyed powder. Consolidation of these powders by powder metallurgy techniques can produce amorphous bulk metals.

Two Alloys 62.24 Zr-10.89 Ti-9.71 Ni-13.14 Cu-4.02 B and 64.84 Zr-11.35 Ti-11.12 Nt-13.69 Cu weight percent were mechanically alloyed for 45 hours by a SPEX 800 high energy ball-mill. The changes in structure were monitored by X-ray diffraction after every 5 hours of milling. Both powder compositions became amorphous after 15 hours of milling. New compounds began to form during milling to 35 hours. Milling for longer times produced no further structure changes. The milled samples were annealed at 950°C for 1 hour which produced a complex set of crystalline materials. The crystalline phases containing boron have larger lattice parameters and less tendency for grain growth.

The diffractograms of syn-gypsum and of fine gas desulfurization products indicate that CaSO4.2H2O is converted to other phase(s) when heated to 100°C. Syn-hannebachite (CaSO30.5H2O) is unaffected by similar thermal treatment.