The feasibility of the INAA of samples in the kg range has been demonstrated in 1994 byOverwater et al. In his studies, however, he demonstrated only the agreement between the corrected
-ray spectrum of large samples and that of small samples of the same material. In this paper, thek0-calibration of the IRI facilities for large samples is described, and some ofOverwater's results for homogeneous materials are presented again, this time in tems of (trace) element concentrations. It is concluded that large sample INAA can be as accurate as ordinary INAA.
In the year 2000, at the MARC V conference, the first results obtained at constant count rates with so-called "zero dead time counting" (ZDT) as implemented in ORTEC's DSPECPLUS® were presented. In this paper, further experiments are described that were performed to establish how the DSPECPLUS® performs at varying count rates. At the same time, the experiments were designed to demonstrate the possible inadequacy of the dual spectrum approach sometimes used to solve the problem of non-Poisson counting statistics encountered in loss-free counting, and to test the "variance spectrum" alternative offered by the DSPECPLUS® . It is concluded that the DSPECPLUS® performs with good accuracy at dead times lower than 90%, even when count rates vary. It is also concluded that the dual spectrum approach indeed is inadequate. Finally, it is shown that the "variance" spectrum approach provides the correct uncertainties to be used in the treatment of LFC or ZDT data.
By employing neutron (or X-ray) diffraction, the structure of crystalline materials can be determined. However, if an impurity in the crystal is present in concentrations below, say, 1·10–4, its influence cannot be observed in the diffraction patterns. If the impurity present at low concentrations is to be localized, a signal uniquely attributable to the impurity must be obtained. In this paper, two such methods, based on the same principles as the "X-ray standing wave" technique, are proposed for neutrons.
The germanium semiconductor detector revolutionized NAA in the late sixties. Software was used from early on to analyze the spectra obtained and compute the concentrations. Since then, our understanding of the physics playing a part has improved, and desktop computing power has increased enormously. Highly accurate nuclear data, procedures and algorithms have become available. However, the software available for gamma-ray spectrometry, in general, and INAA specifically has not changed essentially since the early 1970s. In this paper, a brief overview of the history of such software is given, current developments enumerated and the ultimate, not-yet-existing software for gamma-ray spectrometry and INAA is sketched.
The k1-method for standardization in INAA specifically tackles the problem of the interpretation of gamma-ray spectra as obtained with highly efficient detectors, as opposed to the k0-method. In this paper, results obtained from three NIST reference materials, measured after neutron activation with a gold-lined well-type detector, are presented. It is concluded that the accuracy of the method is better than 1%.
The independently measured catalogues of two single comparator methods for standardization in INAA, i.e., thekZn-and thek0-method, were compared. Many reactions were listed only in thekZn-catalogue; in these cases other literature data were used to supplement thek0-catalogue. The agreement between thekZn-and thek0-catalogue was found to be much better than the agreement between thekZn-catalogue and other literature data. It is therefore suggested to use the convertedkZn-catalogue as a supplement of the previously publishedk0-catalogue.
As early as 1973, an extensive -ray catalogue for INAA with Ge(Li) detectors was created at IRI based exclusively on experiments performed at IRI. This catalogue was perfected and extended over the years and now covers all radionuclides and peaks observed in practice, escape- and sum peaks included. The human-readable form of this catalogue can be a great help in solving interpretation problems of gamma-ray spectra obtained with highly efficient detectors.
Measurements performed in the past to determine sensitivity enhancements (later identified as neutron density increases) in PGNAA as a function of hydrogen concentration in slab-shaped samples are described. The results are compared to the results of Monte Carlo computations. It is concluded that, like H2O, D2O can also cause substantial neutron density increases. In one concentrated salt solution, however, D2O seems to cause a neutron density decrease that cannot be explained from the macroscopic neutron scattering and absorption cross sections in the model used.
Two samples of one litre each of mercury contaminated soil were analyzed by Big Sample INAA. Using this method, sample preparation procedures can be omitted and the number of sample size reduction steps are decreased. Therefore, the representativity of the sample is improved. Afterwards, Big Sample INAA is compared to standard INAA by analyzing ten subsamples taken from each one litre sample. It is concluded that multi-element analysis results of Big Sample INAA are valid even for inhomogeneous large samples. This paper presents the first concentrations measured with BSINAA.
The definitions used for the k0-constant, the coincidence correction factorc and the detector efficiency
f in the k0-method for NAA provide no means to interprete or correct for interference by artificial peaks. In this paper, extended descriptions for the detector efficiency are proposed to deal with escape peaks. For sum peaks caused by true coincidence, a k1-constant is defined as an alternative for the k0-constant, by separating the k0 into a part related to activation and a part related to spectrometry. The k1-constant is based on experimental data, just like the k0-constant.