| Abstract | Portable x-ray fluorescence spectrometers (PXRF) have been greatly improved over the last few decades such that they now can find far-reaching application in geology and geochemistry. Herein, we review
features of available PXRF, discuss their benefits and limitations, and give practical suggestions and guidelines for their selection and use in mineral exploration. Finally, we present several case-studies from our recent work in the Abitibi
Greenstone Belt, Quebec and Ontario and the Bathurst Mining Camp, northern New Brunswick.
In energy dispersive x-ray fluorescence spectrometry, a sample is bombarded by x-rays that cause the atoms within the sample to fluoresce (i.e., give off
their own characteristic x-rays) and this fluorescence is then measured, identified and quantified. The energy of the characteristic x-rays identify the elements present in the sample and, in general, the intensities of the x-ray lines are
proportional to the concentration of the elements in the sample, allowing quantitative chemical analysis.
Over the last few years, miniaturized x-ray tubes have largely replaced previously used radioisotope sources for most applications as these
provide higher x-ray fluxes, shorter count times, and better precision. Silicon positive intrinsic negative (SiPIN) detectors are most commonly used today, and these convert incoming x-ray signals to voltage that is proportional to the energy of the
incoming x-rays; these voltages are then sorted by a multichannel analyzer and fed to a miniature computer. Their energy resolution is too low to permit detection and quantification of many of the key light elements (LE; Mg, Al, Si). However, recent
improvements that have placed the analytical path in a vacuum rather than air (most presently available PXRF) are offered as options on some models. Because low energy x-rays generated by LE are attenuated in air, its removal maximizes the x-rays
that are detected. Within the last year, silicon drift detectors (SDD) have become an available option in many PXRF; they have a higher energy resolution and count rates, providing better precision and shorter analytical times, thus making them much
more suitable for measurement of LE in an air path.
Hand-held PXRF are available in a lightweight pistol and bench-top forms. Bench-top PXRF are slightly larger and heavier, and have a more powerful tube than most hand-held PXRF, which provides
better precisions and detection limits for many elements; both can be operated by battery power. Numerous accessories are available for PXRF, and several manufacturers have optimized their products for geological and geochemical applications,
including Thermo Fisher Scientific and Innov-X Systems.
Geological and geochemical applications of PXRF generally require multi-element analysis; however, the more elements that are included within an analytical test, the greater the likelihood of
problems such as peak overlaps or interferences, and manufacturers typically will provide machine calibrations for 20-30 elements in a particular analytical mode (see below). Our instruments have been calibrated for a range of elements for
characterization of lithological units, different mineralization types and associated hydrothermal alteration, and other geochemical exploration vectors.
Fundamental Parameters (FP) are universal standardless, factory built-in calibration programs
that describe the physics of the detector's response to pure elements, correction factors for overlapping peaks, and a number of other parameters to estimate element concentration while theoretically correcting for matrix discrepancies. FP should be
used for accurately measuring samples of unknown chemical composition in which concentrations of light and heavy elements may vary from ppm to high percent levels. Compton Normalization (CN) is an 'internal' standard, in which spectra are normalized
to the Compton peak, which is produced by incoherent backscattering of the source radiation and is present for every sample. The intensity of the incoherent radiation backscatter reflects both the composition of LE in the sample matrix as well as the
overall concentration of detectable elements. LE-dominant matrices produce a larger Compton peak, so this method provides the best results for measurement of sub-percent concentrations of heavy elements in samples composed mainly of LE.
Our
instruments have 4 analytical modes, with the first two being most useful as they are relatively insensitive to sample matrix composition: 1) Process Analytical: employs FP and should be used for analysis of 'ore grade or style' mineralization with
some gangue; this provides 'assay' level data and typical detection limits (DL) of 0.5 wt%. 2) 'Soils': employs CN and should be used for analysis of soils, rocks, and other materials with a predominantly LE matrix; gives 'geochem' level data at DL
varying between 1 and 100 pm. 3) Empirical Analysis: uses reference standards to establish an empirical calibration line, and can be useful for samples where all major elements present cannot be analyzed. The standards should be matrix-matched and
contain a range of element concentrations bracketing the desired level of quantification. 4) 'Analytical' or 'Alloys': employs FP and should be used for native metals and provides typical DL of 0.1-0.5%.
PXRFs are well-suited to myriad
applications, spanning the entire exploration, mining, and remediation cycle. Among these are: 1) geological surface and underground mapping; 2) geochemical exploration (rock, soil and stream sediment surveys); 3) determination of metal contents of
mineralized samples, mineral identification during prospecting and logging of drill core and cuttings; 4) mining and mineral processing grade control; and 5) environmental baseline and monitoring studies.
PXRF analyses were used to assist in
distinguishing between different volcanic units of the Chibougamau area on the basis of high field strength elements (HFSE; Ti, Zr, Y), and to characterize the base metal contents of mineralization intersected in drillcore in the footwall of the
Lemoine VMS deposit, Quebec (Fig. 1). Two major units can be recognized on the basis of HFSE contents: high Ti-low Zr, and high Zr-low Ti. PXRF Zr abundances are very similar to ICP-ES analyses of fewer selected samples, but an applied correction
factor of 1.5 was necessary for PXRF Ti to match ICP values. However, data for both methods readily distinguished the units. The Cu-Zn-enriched nature of mineralization is correctly identified by both methods.
PXRF analyses of drillcores of
sulfidic black shales intercalated with volcanic rocks of the Kidd-Munro Assemblage were used to differentiate between sedimentary horizons that contain a hydrothermal component and those that do not. The characterization of the element enrichment
suite and metal abundance data of the former are being used to vector toward concealed hydrothermal vent sites and mineralization. Figure 2 depicts geochemical profiles for Cu and Zn in 3 drillcores situated stratigraphically along 2.3 km of strike
length and located about 8 km north of the Kidd Creek Zn-Cu-Pb-Ag mine, Ontario, as determined by hand-held and bench-top PXRF. The figure shows that of the 3 broadly spaced argillite-bearing intervals, only the stratigraphically lowermost contains a
hydrothermal signature whereas the upper 2 are barren. Within this horizon, diagnostic decoupled distribution of Cu and Zn is not evident in conventional assay data but is clearly recognized in PXRF analyses. Vectoring along this lower horizon may
guide further exploration.
Powdered samples of banded iron formation exhalites that are spatially and temporally associated with VMS mineralization in the Ordovician Bathurst Mining Camp of northern New Brunswick were analyzed by PXRF. These data
were then compared with previously obtained high-quality laboratory bulk compositional data in order to evaluate the efficacy of PXRF analyses in lieu of laboratory data and test applicability of PXRF in vectoring. There is generally good agreement
between PXRF and laboratory data (except for high-Fe samples, as noted above; a few elements required application of correction factors). Elements of hydrothermal origin that can be analyzed by PXRF and which vary with distance from known VMS
deposits include: Fe, Mn, Ca, Sr, Ba, Pb, Zn, Bi, Cu, Sb, and Sb. Proportional symbol maps for various elements and element ratios highlight anomalies similar to laboratory data (e.g., Fig. 3), and show that PXRF analyses serve as a primary vectoring
tool that can replace time consuming, more costly laboratory analyses.
Preliminary testing of the use of PXRF in analysis of glacial till was conducted in the vicinity of the Halfmile Lake VMS deposit area, also located in the Bathurst Mining
Camp. PXRF data were collected for dry, <63 µm (silt+clay) fraction till that were previously analyzed by high-quality laboratory methods. These laboratory data (Cu, Zn, Pb, Au, Sn, In, As, Sb, Hg) were previously used to define an eastward trending
glacial dispersal train. Bulk, unprocessed (moist, coarse+fine) till was also analyzed by PXRF from 35 of the 55 original sample sites. In general, there is remarkably good agreement between PXRF data (e.g., Cu, Pb, Zn, As) for dry and moist tills,
and the laboratory data. These results indicate that PXRF can be used to detect metal-rich till, either to guide sampling for laboratory analyses or direct detection of glacial dispersal trains in the field.
PXRFs provide rapid, in-situ, low cost,
non-destructive, quantitative and/or qualitative multi-element analyses of many different sample media that require little or no sample preparation and instrument calibration for varied applications. The examples presented here illustrate the use of
PXRF in mapping rock units and mineralized zones, and also vectoring toward mineralized sources using rocks and surficial material. |