Joseph Nicolosi

Full spectrum calculations of EDXRF spectra

Rapid and accurate methods are becoming available to calculate all of the relevant physical effects that contribute to an EDXRF spectrum, rather than just the characteristic line intensities given by the traditional fundamental parameters method. To evaluate the utility of such methods, we have calculated the full spectra of several compounds covering a wide range of compositions. The calculated spectra are compared directly with measured spectra. They include scattering of the X-ray tube lines and continuum, the Compton profile, and the detector response. Our results indicate that it is now possible to compute the full spectrum from an EDXRF system with very good accuracy.

F-15 IMPROVING TRACE ELEMENT DETECTION IN EDXRF BY REDUCING PILEUP ARTIFACTS

Pileup artifacts appear in energy-dispersive X-ray spectra at high count rates when X-rays arrive at the detector with time separations less than the resolving time of the pulse processor. These artifacts often appear as extra peaks in the spectrum and can mask (or be mistaken for) weak peaks of trace elements. In X-ray fluorescence (XRF) the background is very small compared to charged-particle excitation techniques. This makes it ideal for trace element quantification but also makes it particularly susceptible to pileup artifacts. Recent improvements in high-speed digital discrimination have improved rejection of near-simultaneous events leading to pile-up at very high count rates. This new capability also improves the predictability of pileup rejection, which is essential for accurate modeling and reliable removal of the inevitable events that get past the pileup inspection. We have successfully reduced pileup artifacts by a combination of hardware changes and software correction. We will report the results for a variety of spectra to demonstrate their effectiveness. To evaluate the effect of pileup artifacts on trace element determination, we will show spectra and quantitative analysis for trace amounts of Zr in a Ni alloy and for trace levels of Cu in glasses, with and without the new pileup rejection methods.

MINIMAL USE OF STANDARDS FOR QUANTIFYING TRACE ELEMENTS IN PLASTICS

The minimum requirements for standards when quantifying trace elements in plastics by X-ray fluorescence (XRF) have been investigated. The fundamental parameters method, backed up by one or more standards, is used to translate XRF intensities into quantitative information about trace element concentrations. The method includes accounting for variations in sample thickness using scattered radiation. Of course, the method works best if the standards contain certified values of all of the trace elements of interest in a matrix identical to the unknown. However, we have had reasonable success using only a single element from a plastic standard as a calibrant. The presence of uncertified and uncalibrated elements in both the standards and the unknowns can also have a significant affect on the results, depending on how such elements are treated in the fundamental parameters method. We have also investigated using standards based on a different matrix than the unknown, such as polyvinyl chloride standards for analyzing various other plastic resins and water-based standards for analysis of trace elements in oil. Standard deviations are typically 10% to 15% relative when all elements are calibrated from a standard, and up to 20% when minimum use is made of standards.

Fundamental parameters analysis of RoHS elements in plastics

European Community Directive 2002/95/EC restricts the use of certain hazardous substances in electrical and electronic equipment. In particular, restrictions are placed on lead, mercury, cadmium, hexavalent chromium, and bromine (in polybrominated biphenyls or polybrominated diphenyl ethers). XRF is a convenient method for detecting the presence and measuring the amounts of these elements. Reliably quantifying all of these elements in plastics typically requires a large number of standards that are not yet readily available. Because of the light element matrix, using a “standardless” fundamental parameters method requires some reliance on the primary beam scatter, complicating the analysis algorithm and increasing the uncertainty. We have tested a simplified fundamental parameters method that determines the matrix via difference, requiring only one standard. The method was tested on a series of reference materials containing all of the regulated elements in a variety of plastic resins. One multi-element reference standard was used. It was necessary to include all of the additives in the specimens to achieve good quantitative accuracy. In addition, the scattered primary intensity was used in one set of tests to compensate for variations in specimen thickness. This thickness compensation was necessary to get acceptable results for Cd. Results were very promising, with average relative errors and relative standard deviations of about 10%.

F-26 MICRO-XRF ANALYSIS OF THIN FILMS AND COATINGS USING FUNDAMENTAL PARAMETERS

The fundamental parameters (FP) method can be applied to thin films and multiple layer coatings by means of the Mantler equations[1] for predicting XRF intensities. However, in the case of microspot X-ray fluorescence (micro-XRF) the primary spectrum is modified by the optic used to focus the incident beam. This talk will describe a scatter ratio method for measuring the effect of the optic, correcting the primary spectrum, and performing FP calculations for microspot XRF of thin films and layered systems. The calculations were performed using the corrected primary spectrum together with an updated set of atomic parameters[2] and the closed form intensity equations of DeBoer[3]. The composition of each individual layer (weight percent of constituents) and the thickness for a variety of systems were determined both with and without reference to standards.

F-13 ADVANTAGES AND DISADVANTAGES OF BAYESIAN METHODS FOR OBTAINING XRF NET INTENSITIES

A method of extracting net intensities for the element peaks in an X-ray fluorescence (XRF) spectrum based on Bayes Theorem will be reviewed and discussed. The method has several advantages, among them that the total counts are preserved and that the a priori information is incorporated in a natural manner. These make it, for example, less sensitive to errors in the peak position, widths, and alpha/beta ratios. However, there are also disadvantages. The preservation of total counts means that spurious peaks, such as those from crystallite diffraction, will have their counts included in one or more of the element peaks. We present the results of applying a Bayesian deconvolution method to several XRF spectra and compare them to conventional methods. The spectra are selected to illustrate both the advantages and disadvantages of this method. Errors introduced by variations in the functions used to model the peak shapes are investigated. Quantitative comparisons are also made by using the net intensities from each method in standardless analysis.