Feb/March_AMP_Digital

A D V A N C E D M A T E R I A L S & P R O C E S S E S | F E B R U A R Y / M A R C H 2 0 2 1 1 9 X -ray fluorescence (XRF) is a well established analytical tool that is used for determining the ele- mental composition of materials. It is a quantitative microanalysis technique in which x-rays from a primary source are impinged on a material of interest to cause emission of secondary x-rays from the material. The energies of these secondary x-rays are characteristic to the material, and this can be used to determine the relative amounts of ele- ments present in the material. Over the years, XRF has found a wide range of applications across many industries. Some examples where it is used include: positive materials iden- tification of metals and alloys, quali- ty control in materials fabrication and component manufacturing, geological and environmental analysis, drug man- ufacturing in the pharmaceutical in- dustry, and as a diagnostic tool in the medical industry. XRF units come in var- ious forms, and with varying elemental sensitivity and quantification capabil- ity—from a portable handheld unit for quick, semi-quantitative analysis to benchtop analyzers capable of high spatial resolution elemental mapping, to a wavelength dispersive XRF sys- tem for high sensitivity and quantifica- tion capability. Micro-XRF (µ-XRF) operates under the same principles as conventional XRF. An important difference between conventional XRF analysis and µ-XRF is that in conventional XRF, large sample areas of several millimeters are usually analyzed with a broad beam size rang- ing from hundreds of micrometers to several millimeters with the purpose of quantification. Wider beam size results in difficulties analyzing the composi- tions of smaller regions within a mate- rial (such as dilution effects in welds) or to distinguish spatial heterogeneities in chemical compositions at closely spaced sites (e.g., segregation). On the other hand, µ-XRF uses a small spot cre- ated by polycapillary x-ray optics to ex- cite a much smaller, predefined sample location. The resulting beam spot size measures in the tens of microns for the same depth of penetration as a conven- tional XRF. In addition, samples for con- ventional XRF are prepared to optimize the conditions for quantification and preparation steps often involve homog- enization, among other modifications. Such a preparation procedure ensures a standardized sample evaluation rou- tine and high reproducibility of results. However, inhomogeneity of the sample can be an important aspect of the ana- lytical question. The improved spatial resolution achieved in µ-XRF, with the option for little to no sample prepara- tion, can play a key role in establish- ing local compositional variations for many applications. When this capabil- ity is coupled with a large, motorized stage having precise control in the XYZ directions, it becomes possible to col- lect maps by reproducibly traversing the stage across the defined area of in- terest. The results from such an analysis are displayed as elemental distribution plots that allow for the interrogation of elemental changes and/or selected locations. The outcomes are similar to el- emental distribution mapping per- formed using energy dispersive spec- troscopy (EDS) in a scanning electron microscope (SEM), although there are some key differences regarding sensi- tivity and spatial resolution. The sen- sitivity of µ-XRF for many elements is, in principle, higher than what can be achieved in SEM-EDS due in part to the lack of background of Bremsstrahlung radiation in the former. This leads to manufacturers of µ-XRF specifying high- er detectability limits for elements with Z > 20 than in SEM-EDS, for a given ma- trix composition. These theoretical im- provements in sensitivity for coarse grained crystalline materials can be off- APPLICATION OF MICRO-XRF MAPPING: POWER GENERATION INDUSTRY CASE STUDIES Through a series of case studies, the benefits of micro x-ray fluorescence (XRF) over traditional XRF are demonstrated in the power generation industry for informing weld procedures, identifying root cause, and providing materials or component specifications. Tapasvi Lolla and John Siefert, Electric Power Research Institute, Charlotte, North Carolina Geoff West, WMG, University of Warwick, Coventry, England Tina Hill, Bruker AXS Inc., Madison, Wisconsin