ALPHACAL: A new user-friendly tool for the calibration of alpha-particle sources.

In this work, we present and describe the program ALPHACAL, specifically developed for the calibration of alpha-particle sources. It is therefore more user-friendly and less time-consuming than multipurpose codes developed for a wide range of applications. The program is based on the recently developed code AlfaMC, which simulates specifically the transport of alpha particles. Both cylindrical and point sources mounted on the surface of polished backings can be simulated, as is the convention in experimental measurements of alpha-particle sources. In addition to the efficiency calculation and determination of the backscattering coefficient, some additional tools are available to the user, like the visualization of energy spectrum, use of energy cut-off or low-energy tail corrections. ALPHACAL has been implemented in C++ language using QT library, so it is available for Windows, MacOs and Linux platforms. It is free and can be provided under request to the authors.

Influence of source composition and particle energy on the determination of gross alpha activity.

The influence of different source compositions and α-particle energies on the detection efficiency of a gas-flow proportional counter was examined using experimental measurements and Monte Carlo simulations. Efficiency variation with alpha-particle energy was very marked, being less significant with the substrate composition. These results show that the determination of gross alpha activity in an unknown sample must be carried out very carefully in order to give a correct estimation of its activity.

Improving the trade-off between simulation time and accuracy in efficiency calibrations with the code DETEFF.

Quick and relatively simple procedures were incorporated into the Monte Carlo code DETEFF in order to consider the escape of Bremsstrahlung radiation and secondary electrons. The relative bias in efficiency values was thus reduced for photon energies between 1500 and 2000 keV, without any noticeable increment of the simulation time. A relatively simple method was also included to consider the rounding of detector edges. The validation studies showed relative deviations of about 1% in the energy range 10-2000 keV.

Assessment of the suitability of different random number generators for Monte Carlo simulations in gamma-ray spectrometry.

The Monte Carlo method has become a valuable numerical laboratory framework in which to simulate complex physical systems. It is based on the generation of pseudo-random number sequences by numerical algorithms called random generators. In this work we assessed the suitability of different well-known random number generators for the simulation of gamma-ray spectrometry systems during efficiency calibrations. The assessment was carried out in two stages. The generators considered (Delphi's linear congruential, mersenne twister, XorShift, multiplier with carry, universal virtual array, and non-periodic logistic map based generator) were first evaluated with different statistical empirical tests, including moments, correlations, uniformity, independence of terms and the DIEHARD battery of tests. In a second step, an application-specific test was conducted by implementing the generators in our Monte Carlo program DETEFF and comparing the results obtained with them. The calculations were performed with two different CPUs, for a typical HpGe detector and a water sample in Marinelli geometry, with gamma-rays between 59 and 1800 keV. For the Non-periodic Logistic Map based generator, dependence of the most significant bits was evident. This explains the bias, in excess of 5%, of the efficiency values obtained with this generator. The results of the application-specific assessment and the statistical performance of the other algorithms studied indicate their suitability for the Monte Carlo simulation of gamma-ray spectrometry systems for efficiency calculations.

A method to reproduce alpha-particle spectra measured with semiconductor detectors.

A method is proposed to reproduce alpha-particle spectra measured with silicon detectors, combining analytical and computer simulation techniques. The procedure includes the use of the Monte Carlo method to simulate the tracks of alpha-particles within the source and in the detector entrance window. The alpha-particle spectrum is finally obtained by the convolution of this simulated distribution and the theoretical distributions representing the contributions of the alpha-particle spectrometer to the spectrum. Experimental spectra from (233)U and (241)Am sources were compared with the predictions given by the proposed procedure, showing good agreement. The proposed method can be an important aid for the analysis and deconvolution of complex alpha-particle spectra.

Efficiency transfer in the calibration of a coaxial p-type HpGe detector using the Monte Carlo method.

Monte Carlo simulation was applied to the efficiency transfer exercise described in the EUROMET428 project (Appl. Radiat. Isot. 55 (2001) 493), evaluating the peak efficiencies in the energy range 60-2000 keV for a typical coaxial p-type HpGe detector and several types of source configuration: point sources located at various distances from the detector and a cylindrical box containing three matrices. The efficiency values were derived in two ways: (a) by direct calculation taking into account the physical dimensions of the detector provided by the supplier, and (b) by means of relative computation (efficiency transfer) taking also into consideration the known efficiency values for a reference point source. As expected, some significant discrepancies between the calculated and experimental values were found when a direct computation was made using the data provided by the supplier. On the contrary, the results for the peak efficiency derived by relative calculation by means of an efficiency transfer were in good agreement with the experimental values. The deviations found with this last procedure were generally below 5% for all the geometries considered, which is entirely satisfactory for the purposes of routine measurements.

Monte Carlo simulation of the self-absorption corrections for natural samples in gamma-ray spectrometry.

Gamma-ray self-attenuation corrections in the energy range 60-2000 keV were evaluated by means of Monte Carlo calculations for environmental samples in a cylindrical measuring geometry. The dependence of the full-energy peak efficiency on the sample density was obtained for some particular photon energies and, as a result, the corresponding self-attenuation correction factors were obtained. The calculations were performed by assuming that natural materials have mass attenuation coefficients very similar to those of water in the energy range studied. Three different HpGe coaxial detectors were considered: an n-type detector with 44.3% relative efficiency and two p-type detectors of relative efficiencies 20.0% and 30.5%. Our calculations were in very good agreement with the self-attenuation correction factors obtained experimentally by other workers for environmental samples of different densities. This work demonstrates the reliability of Monte Carlo calculations for correcting photon self-attenuation in natural samples. The results also show that the corresponding correction factors are essentially unaffected by the specific coaxial detector used.