I. Orlic

The National University of Singapore nuclear microscope facility

Abstract The National University of Singapore nuclear microscope facility is based around a HVEC AN2500 single ended Van de Graaff accelerator and an Oxford Microbeams coupled quadrupole triplet focusing system. Particle induced X-ray emission (PIXE), nuclear or Rutherford backscattering spectrometry (RBS) and scanning transmission ion microscopy (STIM) can be carried out simultaneously. Data acquisition is carried out using a simple but flexible PC based system (Oxford Microbeams DAQ) and the data is analysed using a combined RUMP and GUPIX PC based interactive package (NUSDAN) acting under WINDOWS. Resolution tests using a calibration grid and a multi layer integrated circuit have shown the facility to be capable of 600 nm spot sizes for 2 MeV protons at currents suitable for microanalysis.

The 2000 IAEA intercomparison of PIXE spectrum analysis software

Seven programs for analysis of PIXE spectra were compared using the 2000 IAEA test spectra, i.e. Geopixe, Gupix, Pixan, Pixeklm, Sapix, Winaxil and Witshex. A systematic statistical study of the analysis results was performed based on z-scores. The results indicate that most of the programs perform reasonably well with respect to peak areas. Except for a very rare exception, the statistical analysis shows that the participants generally reported smaller uncertainties than would have been expected from the reference uncertainty values. The results show that all the participants reported a number of statistically significant “false hits” and “misses” in their reports. All this indicates that programs in general still leave room for further improvements.

TTPIXAN — A package of computer programs for quantitative thick target PIXE analysis

Abstract We present the computer program TTPIXAN, developed recently in our group as a tool for quantitative thick target PIXE analysis. Our intention was to develop an accurate and general computer code which could be useful in all aspects of PIXE analysis, i.e. either for choosing optimal working conditions when planning an experiment, or afterwards, when performing quantitative analysis for a wide range of sample types and thicknesses. The program is suitable for proton excitation as well as for deuterons and He ions in the common range of excitation energies (0.5–4 MeV).

Two years of aerosol pollution monitoring in Singapore: a review

Abstract An aerosol sampling campaign was initiated more than two years ago in Singapore. The aim was to determine the average elemental concentrations in fine and coarse aerosol fractions as well as to identify major pollution sources and their impact. For that purpose, two air samplers were employed at two different sampling locations; one sampler was a fine particulate aerosol sampler (PM2.5) located at the vicinity of a major industrial area. The other was a stacked filter unit (SFU) sampler designed for collection of fine and coarse fractions (PM2.5 and PM10) and installed in the residential area. Samples were taken typically twice a week and in several occasions daily. During the period of two years more than 700 aerosol samples were collected and analyzed using PIXE and RBS techniques. All samples were analyzed for 18 elements ranging between Na, Mg, Al, etc. up to As and Pb. Large daily and seasonal variations were found for most of the elements. These variations are attributed mainly to meteorological changes, in particular changes in wind speed and direction. On several occasions, short term sampling was performed to identify fingerprints of major pollution sources such as road traffic, refineries, as well as the rain-forest fires in neighboring countries. A summary of our findings is presented and discussed.

Parametrization of the total photon mass attenuation coefficients in the energy range 0.1–1000 keV

Abstract It is convenient to generate mass attenuation coefficients using semi-empirical schemes. The validity of most of the existing schemes is limited to a relatively narrow energy interval (1–40 keV) and their accuracies are poor in some energy regions. In this work, a semi-empirical scheme flexible enough to give a good fit to data in a very wide photon energy range (0.1–1000 keV) was employed. Fitting coefficients for the entire range were obtained by utilizing mass attenuation data from two sources: (1) semi-empirical data of Henke et al. in the low photon energy region, and (2) theoretical values generated with the XCOM code for fitting in the high energy region. The root mean square of the fit is generally less than 0.2% except for energies below 1 keV where the available data are scattered. A computer code for generating mass attenuation coefficients based on the proposed scheme has been developed.

Long-term accuracy and precision of PIXE and PIGE measurements for thin and thick sample analyses

This paper describes PIXE/PIGE measurements on thin Micromatter Standard (±5%) foils run over a period of 10 years. The selected foils were typically 50 μg/cm2 thick and covered the commonly used PIXE X-ray energy range 1.4–20 keV and the light elements F and Na for PIGE studies. For the thousands of thick obsidian and pottery samples analysed over a 6-year period, the Ohio Red Clay standard has been used for both PIXE and PIGE calibration of a range of elements from Li to Rb. For PIXE, the long-term accuracy could be as low as ±1.6% for major elements with precision ranging from ±5% to ±10% depending on the elemental concentration. For PIGE, accuracies were around ±5% with precision ranging from ±5% in thick samples to ±15% in thin samples or for low yield γ-ray production.

New parameters for the calculation of L subshell ionization cross sections

Abstract New parameters for the calculation of L subshell ionization cross sections using a semiempirical expression are presented. The parameters were obtained by fitting an analytical function to the experimentally obtained values of L subshell ionization cross sections. The experimental data included compilation of the published data from 1975 to 1982 by Sokhi and Crumpton as well as our newly compiled data published in 1982–1991. The number of data employed for the fitting was 2295, 3 1 2 times more than that used in Miyagawas work. The data were divided into six groups according to the Z values: 14 ≤ Z ≤ 42, 43 ≤ Z ≤ 50, 51 ≤ Z ≤ 60, 61 ≤ Z ≤70, 71 ≤ Z ≤ 80 and 81 ≤ Z ≤ 92. Each was fitted by the analytical function to obtain the parameters for the calculation of L1, L2 and L3 subshell ionization cross sections. For data in the group with Z ≤ 42, only parameters for the calculation of L total ionization cross sections were determined. A comparison of the various theoretical predictions with the experimental data and the fits is also presented.

SEMIEMPIRICAL FORMULAS FOR CALCULATION OF L SUBSHELL IONIZATION CROSS SECTIONS

Presented are new parameters for the calculation of L subshell ionization cross sections for proton impact using a semiempirical expression. A similar paper was published by our group in early 1993 but the fitting parameters were obtained by using only 2295 experimental L shell cross section data. Since then a large number of experimental data have become available and therefore a new fitting has been performed employing more than 5000 data points. All available data were fitted separately for L1, L2 and L3 subshells. For targets with low atomic numbers (14≤Z≤42), only coefficients for Ltot were obtained. Because of the slight Z dependence of the universal function, data were also devised into five sub-groups according to their atomic numbers and fitted separately within each group for L1, L2 and L3 subshells. To extend the energy range of validity of the new fitting function theoretical values were used in the high energy region where experimental data were lacking. Results are compared with ECPSSR predictions and discussed.

Present status of the experimental L-shell ionization cross sections for light ion impact

Abstract Compilations of all available experimental L-shell X-ray production and ionization cross sections for proton and light ion impact have recently been produced by the group from the National University of Singapore (I. Orlic, C.H. Sow and S.M. Tang, At. Data Nucl. Data Tables 57, in press; I. Orlic, Experimental L-shell x-ray production and ionization cross sections for light ion impact, internal report, National University of Singapore, Department of Physics, 1993 [1,2]). Together with the previous tabulations more than 9700 experimental data points for proton impact and approximately the same number of data points for heavier ion impact have become available. An overview of all available experimental data is given in this work. Discussed are annual growth and decline of published data, distribution of L-shell cross section data vs. target atomic number as well as distribution of number of data vs. incident ions. Experimental data are compared with the corresponding ECPSSR predictions in the usual manner: plotting the ratio S = σexper./σtheory vs. reduced velocity parameter for each individual subshell L1, L2 and L3 separately for proton and heavier ion impact. Mean experimental cross sections are obtained using the statistical procedure recommended by the Particle Data Group (Particle Data Group, Phys. Lett. B 204 (1988) 5 [3]). Statistical errors of mean values are significantly smaller than errors of individual experimental results allowing for some generalization. A brief review of the obtained results is presented.

The nuclear microscope: A review of applications in the environmental sciences

Abstract It is only recently that nuclear microscopy has been applied in different fields of environmental science, for example studies of single airborne particles and analysis of concentration depth profiles in shellfish, tree rings, ancient ice, etc. A review of a number of environmental applications is given with the emphasis on analysis of single aerosol particles. The results are briefly compared with the results of other microanalytical techniques such as electron microprobe and laser microanalysis. Possibilities for optimization of experimental setup and also some problems still existing in quantitative analysis of nuclear microscopy images are discussed.