M.A. Reis

Distribution in Portugal of some pollutants in the lichen Parmelia sulcata

During the months of July and August 1993 a lichen (Parmelia sulcata Taylor) collection campaign was held in Portugal where samples were obtained from olive tree bark at 228 sites, following a grid of 10 x 10 km along the Atlantic coast and 50 x 50 km in the interior of the country. The samples were analysed by instrumental neutron activation analysis and proton induced X-ray emission techniques. Concentration data patterns for the pollutants As, Cr, Hg, Ni, Pb, S, Sb, Se, and V were obtained for the whole country surface by making use of an extinction rule of 1/r(3), preventing any cut-off distance from being artificially introduced. Some pollution sources were identified: (1) oil-powered plants on the Lisbon-Setúbal axis (V, Ni), (2) coal-power plants in Porto and Sines (S, Se), (3) traffic in the northern area and the Lisbon-Setúbal axis (Pb), (4) a chemical industry south of Porto (Hg, As), and (5) soil influence (Cr, Sb). Arsenic and chromium results largely exceed, in a few areas, the concentrations acceptable to plants, and in a few spots also Hg, Ni, and Pb data.

DATTPIXE, a computer package for TTPIXE data analysis

Abstract Thick target PIXE (TTPIXE) data analysis requires the calculation of an integral (which we call penetration integral) which accounts for the various effects that take place during proton stopping in the target and lead to X-ray emission. This is done by numerical integration and in most TTPIXE work target homogeneity both along the surface and in depth is necessary. This reduces the applicability of the technique in cases as ion implanted, film coated, or other kinds of samples which are by their own nature nonhomogeneous in depth while preserving their surface homogeneity (we call them layered targets). The DATTPIXE main program, which enables the conversion of spectra peak area data into sample concentration values, is being developed so as to be able to face these kinds of problems. Secondary X-ray fluorescence is accounted for and the correction is calculated for each of the layers into which the target is divided for numerical integration. The program was tested by applying it to the analysis of three metal alloy certified reference materials from the British Chemical Standards. The need for secondary X-ray fluorescence corrections and the correctness of the method used are shown.

Bio-monitoring of trace-element air pollution in Portugal: Qualitative survey

During the months of July and August 1993 a lichen collection campaign was held in Portugal where about 250 samples were collected and analysed by thick target TIPIXE and INAA. Results for 44 different elements were obtained and a data base was built and subjected to Monte Carlo Aided Target Transform Factor Analysis (MCATTFA), a method developed at IRI (Delft). 10×10 km2 coast and 50×50 km2 far from coast sampling grids were used for collection. In this work we will present the results which were obtained based on a reduced data set of 36 elements measured by INAA. The results for 22 elements obtained by TTPIXE were published elsewhere. In this work we will also present the conclusions considering all the 44 elements determined by both techniques.

Quality assurance in elemental analysis of airborne particles

Abstract Instrumental neutron activation analysis (INAA) and particle induced X-ray emission (PIXE) are commonly used to yield multielemental data in atmospheric studies. In this work the quality assurance of the analytical data is discussed based on two studies. The first is a precision study, involving sampling and analysis of air particulate matter in the size ranges of the aerodynamic diameter (AD)

Lichens (Parmelia sulcata) time response model to environmental elemental availability

Parmelia sulcata transplants, collected in a non-polluted area, were exposed to new atmospheric conditions at six stations, of which five were located near power plants and one at an unpolluted area. Data were collected for a 1-year period, on rainfall, airborne particulates, elemental deposition and on lichen elemental composition. The study was carried out to quantify the relationships between elemental availability and elemental accumulation in lichens. A mathematical model was developed, involving a simple representation of the lichen, and a memory loss function which describes the time relations for the lichens progressive representation of its new ambient conditions. The collected data set was used to test the model. The results indicate that the environmental data on Na, Al and Pb could be reproduced and predicted by the model, including the variability patterns. Half-memory times were calculated for Na, Al and Pb as 200, 500 and 600 days, respectively, based on using all station data. However, further improvements are still needed, including the clarification of differences in model outcomes between stations. Copyright (C) 1998 Elsevier Science B.V.

Matrix effects correction for quantitative TTPIXE analysis

Abstract The use of thick targets in PIXE analysis has many advantages, namely the smaller probability of significative contamination, easy target handling and preparation. Also in some applications as archaeometry, materials sciences and many others, the use of thick targets is imperative. A correct calculation of the thick target yield is then needed in order to obtain good results. In this work, corrections for the enhancement effects due to secondary X-ray fluorescence in metal alloys and due to radiation damage in biological samples are proposed. Numerical integration problems in critical matrix absorption situations are also presented.

Quality assessment on airborne particulate matter of k0-INAA

The analysis of airborne particulate matter (APM) by k0-NAA was assessed using: (1) BCR reference material (RM) simulated air-filters, (2) synthetic air-filters prepared by spiking blank filters with standard solutions, and (3) real APM filters. k0-INAA is a suitable technique for the analysis of APM, delivering accurate and precise results. However, the quality assessment of APM analysis appears to be a difficult task.

Simultaneous PIXE and RBS data analysis using Bayesian inference with the DataFurnace code

The Rutherford backscattering spectroscopy (RBS) and particle induced X-ray emission (PIXE) techniques can be used to obtain complementary information about the characteristics of a sample but, traditionally, a gap has separated the available computer codes for analyzing data from each technique, being hard to simultaneously analyze data from the same sample. The recent development of a free and open source library, LibCPIXE, for PIXE simulation and analysis of arbitrary multilayered samples, has permitted to integrate this technique into the DataFurnace code which already handles many other IBA techniques such as Rutherford and non-Rutherford backscattering, elastic recoil detection and non-resonant nuclear reaction analysis. The fitting capabilities of DataFurnace can therefore now be applied to PIXE spectra as well, including the Bayesian inference analysis and the simultaneous and coherent fitting of multiple spectra from different techniques. Various examples are presented in which the simultaneous RBS and PIXE analysis allows us to obtain consistent results that cannot be obtained by independent analysis of the data from each technique.

LibCPIXE : A PIXE simulation open-source library for multilayered samples

Abstract Most particle induced X-ray emission (PIXE) data analysis codes are not focused on handling multilayered samples. We have developed an open-source library called “LibCPIXE”, for PIXE data analysis. It is written in standard C and implements functions for simulating X-ray yields of PIXE spectra taken from arbitrary samples, including multilayered targets. The library is designed to be fast, portable, modular and scalable, as well as to facilitate its incorporation into any existing program. In order to demonstrate the capabilities of the library, a program called CPIXE was developed and used to analyze various real samples involving both bulk and layered samples. Just as the library, the CPIXE source code is freely available under the General Public License. We demonstrate that it runs both under GNU/Linux systems as well as under MS Windows. There is in principle no limitation to port it to other platforms.

Elemental analysis of particulate matter and source identification in Lisbon

A Gent PM10 air sampler with stacked filter units was used to collect size-fractionated aerosol samples in Lisbon between October 1993 and January 1995. Sampling was made on a twice-a-week basis (one weekend day and one midweek day). The filter loads were measured by gravimetry and elemental analyses were carried out by PIXE and INAA techniques. The complementarity of the techniques allowed the determination of the concentrations of 22 elements in more than 30% of all samples. Monitoring of NO, NO 2 , CO and SO 2 concentrations was also performed. Time variations for both total particulate mass and elemental concentrations are presented. Weekend and midweek data are compared. Enrichment factors were calculated for the fine and coarse fractions and the ratios between the two fractions were derived. Principal components analysis was applied to the identification of aerosol sources Correlations were found between the gases and coarse particle elemental concentrations. A tendency for a concentration increase in the total suspended particle mass was observed at the end of each year. The concentration pattern for the rare earth elements measured was altered owing to the activity of a nearby refinery.