J.L. Campbell

Quantitative PIXE microanalysis of thick specimens

Abstract Methods of standardization in quantitative micro-PIXE analysis are reviewed and various issues that bear on analytical accuracy are explored; pertinent recent work on Si(Li) X-ray detector response is included and some geochemical examples are drawn upon. Extension of the GUPIX software to deal with multilayer targets, including secondary fluorescence within and between layers, is reported; analytical examples include alloy foils and multilayer solar cell structures.

Analytic fitting of monoenergetic peaks from Si(Li) X-ray spectrometers

Abstract Monoenergetic X-ray lines produced by diffraction from a curved crystal monochromator have been used to test the applicability of the Hypermet function for Si(Li) detectors. Excellent fits are achieved; the functions parameters vary smoothly with energy, and the intensity of the non-Gaussian peak components correlates with the thickness of the frontal dead layer where charge collection is incomplete. Accurate silicon escape peak intensities are presented and the contribution to the lineshape from Compton-scattered events is discussed qualitatively.

Si(Li) Detector Lineshapes: Contributions from Atomic Physics and Detector Properties

The detailed response of an Si(Li) detector to the manganese K x-rays from a 55 Fe source was studied by least-squares fitting various spectrum models using a generalized variant of the GUPIX code. The peak-to-background ratio (>10 6 ) observed for the manganese Kα x-rays is superior in the detector used here to that observed in a crystal spectrometer, and very large spectrum intensities are recorded. In addition to well recognized effects such as the intrinsic Lorentzian distribution of an x-ray line and the approximately Gaussian detector response function, the Kα and Kβ peak shapes are influenced by multiplet structure. The Kβ shape is affected by 3p3d exchange coupling, and both Kα and Kβ are influenced by 3d spectator vacancy satellites. The flat shelf-like features below the peaks appear to be associated with escape of Auger and photoelectrons, and show a step-like structure that has not hitherto been discussed. Using an accurate energy calibration provided by simultaneously recorded rubidium K x-rays, the escape peak energy shift was shown to have the theoretical value of 1.74 keV. The escape peak width was significantly augmented, relative to the value expected for a diagram line at 4.2 keV energy, by shake-off satellites and by an additional component on its low-energy side; this observation explains the widely reported escape energy values in excess of 1.75 keV. Approximate determinations were made of the shape and intensity of the KLL radiative Auger feature and of the intensity of the KMM feature.

X-ray spectrometers for PIXE

Abstract The accuracy of PIXE analysis is enhanced by having available a complete characterization of the Si(Li) X-ray detector as regards both efficiency and resolution function (lineshape) in the 1–35 keV X-ray energy range. Work in both these areas is reviewed, and the dead-layer concept, together with its role in low-energy response, is critically examined. The potential of other spectrometer types including high-purity germanium detectors and diffraction spectrometers is briefly considered.

Effects of random surface roughness in pixe analysis of thick targets

Abstract The ratio of proton-induced X-ray yields from thick targets with perfectly smooth and randomly rough surfaces is computed as a function of proton energy, target-detector geometry and X-ray attenuation coefficient. The rough surfaces are obtained by a Monte Carlo technique. Attenuation coefficient is shown to be the dominant parameter governing the effect of roughness.

Deviation of the Kβ /Kα intensity ratio from theory observed in proton-induced X-ray spectra in the 22⩽Z⩽32 region

Measurements of I(K beta )/I(K alpha ) for proton-induced X-rays from thin targets in the 22<or=Z<or=32 region confirm a previously suspected systematic discrepancy between experimental data and the theoretical predictions for de-excitation of a single K vacancy in a neutral atom. The present paper is the first proton work to pay specific attention to the KMM radiative Auger satellite and to use a measured resolution function in unfolding K alpha -K beta peak overlaps. A discrepancy is common to electron-, photon- and proton-induced ionisation but sufficient precision is not yet available to reveal any small differences among the three processes, which might arise from differing multiple ionisation.

Non-statistical population of magnetic substates of the erbium L3 subshell in photoionization

Angular distributions of the Ll, L alpha 1,2 and L beta 2,15 transitions of erbium were measured following photoionization by 8.904 keV photons. A Johansson-type monochromator was used to select the CuK beta 1 line for ionization. The angular distributions were well described by the function W( theta )=W0(1+ beta P2(cos theta )), with anisotropy parameters beta (Ll)=0.052+or-0.016, beta (L alpha 1,2)=0.016+or-0.022 and beta (L beta 2,15)=0.012+or-0.015. These values are slightly higher than the theoretical values of Berezhko and co-workers (1978). The effect of partially polarized photons on the angular distribution is discussed.

MICRO-PIXE STUDIES OF SR ZONING IN ARCTIC CHARR OTOLITHS : MIGRATORY BEHAVIOUR AND STOCK DISCRIMINATION

Abstract Micro-PIXE has been used to determine the content of strontium in the growth zones of Arctic charr otoliths. A well-defined correspondence is observed between the optically observed growth zones and the strontium zoning pattern. Line-scan data indicate whether or not the fish concerned is anadromous, and they define the years in which migration to the sea has taken place; they also are sensitive to the geographic origin of the fish. These results suggest that micro-PIXE can be a useful tool in the management and conservation of the Arctic charr.

Extension of gupix to 2H, 3He and 4He excitation

Abstract The ionization and stopping power databases in gupix are extended to deal with deuterons and helium ions. The new databases are tested through analysis of standard reference materials using pure elemental targets as standards. Good accuracy is demonstrated for elements in a NIST steel reference material. Deficiencies in the L cross-section database are demonstrated through measurements on BCR standard gold films on nickel substrates.

Assessment of a self-consistent theoretical data base for L X-Ray relative intensities in proton-induced X-Ray emission analysis

Abstract Recent Dirac-Hartree-Slater and Dirac-Hartree-Fock calculations of ionization cross-sections, fluorescence and Coster-Kronig yields and X-ray emission rates offer a “self-consistent” theoretical data base for relative intensities of proton-induced L X-rays. Interpolation schemes for use with tables of these quantities are described. Theoretical intensity ratios for major line groups agree with compiled experimental data to within a few percent, justifying use of the data base in PIXE.