K. Roland Akselsson

REX - A Computer Programme for PIXE Analysis

The implementation of a physics-based model for non-linear least-squares analysis of proton-induced X-ray emission spectra via a Fortran programme REX is discussed. The modelling of distinct spectral components and of the physical effects involved is briefly discussed. Results are presented in graphical and tabular form. The limitations of the present model are discussed and future refinements indicated.

Rapid determination of major and trace elements in geological material with proton-induced X-ray and gamma-ray emission☆

The simultaneous detection of proton-induced X-rays and gamma-rays enables rapid determination of both major and trace elements in thick homogeneous samples. Using protons of energy 2.55 MeV, the atomic numbers of the elements detectable with this technique were Z=3, 5, 9, 11–13 from gamma-ray analysis and Z=13–92 from X-ray analysis. Thick target concentrations were determined from a thoroughly made thin target calibration and from matrix-dependent conversion factors calculated from fundamental parameters. The major compositions of samples needed in these calculations are derived in an iterative procedure using the spectra recorded. The method proposed has been tested on six USGS rock standards. Detection limits for one of the standards are presented. Approximately 20 elements were detected in each sample 5 min irradiations. For elements well above their detection limits, the results are typically accurate to within 5%. The precision of repeated day-to-day analysis is about 2% (one standard deviation). (Less)

Calibration and long-term stability of a PIXE set-up

A general mass calibration procedure of a PIXE set-up is described. As an example, the results of the calibration of the PIXE set-up in Lund are given. For the calibration commercially available standards were used. The parameters of a physical model were adjusted to fit a calibration curve. To decrease errors due to inhomogeneities of the standard foils, the average from four different pieces of each foil was taken. The accuracy of the calibrated system in Lund is estimated to be better than 5% for individual elements and still less for ratios of adjacent elements. For long-term quality control of the results, a special sample has been designed. This consists of a thick silver plate with a small copper spot in the centre. It is very sensitive to small changes in a PIXE set-up. By analysing this sample periodically, mistakes in the arrangement are promptly detected-a feature which is much appreciated in a mixed research, development and routine analysis laboratory. Also a measure of the long-term stability is obtained. For the Lund system, the long-term stability is 2.5% for a homogeneous sample and 4% for a small (inhomogeneous) sample.

Performance of an On-Demand Beam Excitation System for PIXE Analysis

A system for electrostatic on-demand beam excitation in PIXE analysis has been built and carefully tested. Significantly increased count rate capability without any need for corrections for electronic dead time and with less target deterioration makes the system superior to a traditional system with electronic pile-up rejection. By adding a fast external electronic pulse pile-up rejection system, a pile-up interval of about 250 ns is obtained. The detailed behaviour of the particle beam during deflection and experience from using the system in routine analysis are reported.

PIXE Analysis of Samples of Intermediate Thickness

A procedure for making accurate matrix corrections in PIXE analyses of samples of intermediate thickness has been developed. The transmission of a collimated X-ray beam through different parts of the sample is measured with a Si(Li) detector to determine the thickness and shape of the sample. Experiments have been performed using uniform polymer foils doped with known concentrations of different elements and with thicknesses ranging from 1.5 to 11 mg/cm2. The results from these samples indicate that the accuracy of the correction procedure is better than 5%. The correction procedure has been applied to, e.g., samples obtained in single orifice cascade impactors.

Variation of Elemental Composition with Particle Size in the St. Louis Aerosol

Detailed particle size distribution of elemental constituents have been determined in the St. Louis aerosol by sampling with cascade impactors for particles >0.25 μm aerodynamic diameter and by analyzing using proton-induced X-ray emission, PIXE, for S, Cl, K, Ca, Ti, Mn, Fe, Zn, Br, and Pb. Two rooftop sites were selected for intensive sampling in industrial and residential locations in St. Louis during February 1974, every 12 h for a 7 d period which included clear, rainy, and snowy weather regimes. The data could be compared with the results of a similar study of the St. Louis aerosol during August 1973 when the weather was generally warm and without rain. Both data sets could also be compared with results of studies of Tallahassee, situated in coastal north Florida, and the north Atlantic marine atmosphere of Bermuda. Although variation in elemental concentrations and their size distributions were observed which could be associated with local weather conditions during the February sampling period, the data set was sufficiently large for meaningful average characteristics to be computed. As examples, two elements of pollution importance, sulfur and iron, show the following comparison between St. Louis and the Florida and Atlantic locations: Sulfur in all data sets was most abundant in the smallest particle size classes. Winter and summer concentration in St. Louis air were the same within a factor of two but generally exceeded north Florida concentrations. Bermuda concentrations were intermediate between St. Louis and north Florida, except for largest particles where sea spray derived material was present. The results reflect the importance of gas phase conversion to particulate forms of sulfur, e.g. sulfate, in all three locations. Iron concentrations in St. Louis generally exceeded those in north Florida for smallest particles and to a lesser extent for larger sizes. Concentrations in Bermuda were about 10-fold lower than in St. Louis for all particle sizes. The results reflect the terrestrial dust origin of iron and also suggest that iron is partially pollution-derived in the St. Louis atmosphere, especially in smaller size ranges. (Less)

Feasibility Study of Gel Filtration as a Separation Method for Blood Serum Proteins in Combination with PIXE Analysis

The combination between a protein separation technique and the PIXE method has a great potential for large surveys, including thousands of samples, in which multielemental analysis is required. Gel filtration with a Sephadex G-200 gel and a TRIS-acetate buffer was used for separating proteins in human serum. The fractions were doped with an yttrium/vanadium standard and then concentrated and pipeted onto Kimfol™ backing foils. Using the PIXE technique, the distributions of Fe, Cu, Zn, and Se, with respect to molecular size, were found, indicating binding to specific proteins. Sulfur and phosphorus were found to correlate well with the protein content measured by UV-absorption at 280 nm. Further developments and tests on the protein separation technique is required, taking restrictions imposed by the PIXE method into consideration.

Monitoring aerosol elemental composition in particle size fractions for predicting human respiratory uptake

Abstract Proton induced X-ray emission analysis is useful in measuring changes in particle size distribution of trace elements on exposure of aerosol to the human lung environment and deposition in the air-ways. PIXE analysis is also useful in detecting variations in particle size distributions of elements as they occur in the atmosphere. For the case of lead aerosol along a highway, it is shown that the extent and site of respiratory deposition of particles depend on their size, and the size distribution may vary under urban atmospheric conditions sufficiently to affect respiratory deposition efficiency.

Part XII D. Work environment aerosolsMeV protons for the characterization of aerosols in work environments

In this paper the advantages of accelerator-based chemical analysis of work environment aerosols are discussed. Use of the PIXE method is central, but in order to take full advantage of the possibilities of an accelerator laboratory, it should be combined with nuclear methods such as particle-induced nuclear reactions and/or particle scattering for simultaneous measurements of light elements. It is also important to develop optimum sampling techniques for the analytical methods required. Some information about the current set-up in Lund is given. In a companion paper, the potential of PIXE for work environment tasks is illustrated by a summary of the design of two major studies. Some results are also included.