Fully quantified, micron-resolution, whole thin section synchrotron XRF mapping of trace elements in geological materials
Abstract
Thin sections of geological samples have been mapped for their major and trace element contents using a new fast high resolution, full spectral XRF mapping technique. Each pixel of the maps (~2 microns) contains the full XRF spectra in the energy range ~4.5 to 18 keV and the data is collected for areas ~2x1cm in 4-6 hours. This is the first time that fully quantified, whole thin section element maps have been collected using dynamic XRF spectral deconvolution and this allows us to image elements that are expected and those that are unexpected or at trace levels in a single data set. The data is collected using the new CSIRO-Brookhaven MAIA-386 detector system (Ryan et al., 2010) installed on the Australian Synchrotron XFM beamline. The Maia detector system combines a large planar silicon detector array and custom ASICs developed by Brookhaven National Laboratory and a pipelined, parallel processor with embedded Dynamic Analysis image projection developed at CSIRO. It is closely coupled with sample scanning and performs real-time processing and projection of XRF elemental images enabling pixel transit times as short as 50 μs and image sizes up to 100M pixels. The data is processed using the GeoPIXE software (CSIRO) that utilizes a dynamic analysis matrix (DA) spectral deconvolution method to calculate chemical element concentrations from the XRF spectra. In doing this we can be sure that the chemistry is correctly associated and not an artifact of simplistic Region of Interest (ROI) fitting or peak overlaps. The SXRF technique also gives greater sensitivity to element concentrations with trace element detection limits ~10-100 ppm relative to wt% for many SEM EDS mapping systems. Individual element maps vary in size but are typically 10,000x4500 pixels at full resolution. This mapping technique has also been extended to large area (200-400 microns) XANES mapping to understand the high resolution distribution of redox species (Fe2+/Fe3+) in geological materials. Here a data stack is built up at energy increments across the K or L edges of the element of interest to be able to define XANES spectra at each pixel. This has been applied to looking at small-scale redox changes in Fe and As bound in minerals associated with gold deposition, and is only possible by combining rapid, high resolution mapping with tunable energy sources. The Synchrotron XRF MAIA imaging of geological materials is providing a new way to look at fine scale spatial chemical patterns at the 1-5 cm scale, and this has also been extended to speciation mapping using XANES. This data allows for the first time to view textural chemical relationships that have been hard or impossible to see before. As an example recent work will show the mapping of the distribution of trace Au in geological materials as well as meteorites.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2011
- Bibcode:
- 2011AGUFM.V23G..04C
- Keywords:
-
- 1042 GEOCHEMISTRY / Mineral and crystal chemistry;
- 1065 GEOCHEMISTRY / Major and trace element geochemistry;
- 1094 GEOCHEMISTRY / Instruments and techniques