Radiochim. Acta 99, 59–63 (2011) / DOI 10.1524/ract.2011.1788  by Oldenbourg Wissenschaftsverlag, München Determination of the activation constants for miniature neutron source reactor (MNSR) by linear least squares regression By M. Wasim ∗ , M. Arif and J. H. Zaidi Chemistry Division, Pakistan Institute of Nuclear Science and Technology, P.O. Nilore, Islamabad, Pakistan (Received June 2, 2010; accepted in revised form August 10, 2010) Activation constant / SMELS / MNSR / NAA standardization Summary. This paper presents a detailed description of the determination of experimental activation constants for 22 elements for miniature neutron source reactor (MNSR). The activation constants have been measured for the most intense γ -rays of the radionuclides commonly observed in neutron activation analysis. The experimental values have been compared with those based on mathematical expression and calculated by using f and α of the irradiation channel and k 0 and Q 0 factors from published literature. A compar- ison of the expression-based and experimentally determined activation constants shows a good agreement, except for some cases where difference is about 10% of the expression-based values. The activation constants were validated by analyz- ing synthetic multielement standards (SMELS-I and -II); the results revealed all elements within ±3 Z -score. 1. Introduction There are three main calibration methods [1] in neutron acti- vation analysis (NAA). These include relative, absolute and semi-absolute standardizations. Relative method is perhaps the oldest one, in which a sample is irradiated simultan- eously with one or more standards containing elements of interest. If the matrices of standard and sample are closely matched in their composition, the results produced by the relative method are usually accurate. Though the method is simple in concept, it requires more resources in terms of de- tector and reactor times. The main drawback of this method is the preparation of a suitable standard with the presence of all elements of interest. Since the physics of NAA can be modelled completely in mathematics, therefore, the absolute method [2] was developed to overcome the shortcomings of the relative method. The use of the absolute method re- quires an accurate determination of neutron flux, detector efficiency, gamma emission probabilities, cross sections and other parameters. The calculations in the absolute method utilise nuclear data, which are retrieved from the published literature. Since the published data are occasionally associ- ated with large uncertainties, the final results may usually contain high errors. Even today, the application of the abso- lute method does not guarantee accurate results for majority *Author for correspondence (E-mail: mwasim@pinstech.org.pk). of the elements. To overcome the problems associated with the requirement of suitable standards in the relative method and the associated high uncertainties of nuclear data in the absolute method, several versions of semi-absolute methods were developed. One such method involves the experimen- tal determination of k-factors [3], which lately changed to k 0 -NAA standardization [4]. The latter requires, for 1/v nuclides, thermal to epithermal neutron flux ratio ( f ), ep- ithermal neutron flux shape factor (α), k 0 -factors, Q 0 -factors (ratio of resonance integral to thermal neutron cross section) and full-energy peak detection efficiency (ε p ). Both k 0 and Q 0 factors are available in literature. The other variants of semi-absolute method involved the experimental determin- ation of isotope related k i -factors [5] or irradiation position related activation constants [6], an interesting concept for the reactors showing very stable neutron flux characteris- tics over a long period of time. Activation constants can be measured either experimentally or by using the k 0 concept as presented by Kennedy et al. [7]. Our laboratory implemented k 0 -NAA standardization for two research reactors [8, 9] by using bare triple monitor method [10]. The nuclide identification and elemental quan- tification was performed by our in-house written software called GammaLab [11]. We applied k 0 -NAA successfully to the determination of elemental composition of various samples in different types of matrices [12 – 14]. It is our experience over many years that MNSR shows good re- producibility in neutron flux, which allowed us to develop semi-absolute standardization, based on the use of activation constants. The main aim of the present work is to present the method in detail for the determination of experimental as well as mathematical expression-based activation con- stants for Pakistan Research Reactor-2 (PARR-2) irradiation facility. In relative standardization of NAA, usually one cal- ibration sample is used; the concept of activation constant however, utilizes more than two standards to produce good regression coefficient with minimum error. The experimen- tal method for the determination of activation constant is the same for 1/v and non 1/v nuclides. The validation of the determined activation constants was performed by ana- lysing two recently developed synthetic multi-element stan- dards (SMELS-I and SMELS-II) [15]. They were prepared by the Institute of Reference Materials and Measurements (IRMM), in cooperation with the Laboratory of Analyti- cal Chemistry, Ghent University. In literature [7] activation Brought to you by | New York University Bobst Library Technical Services Authenticated Download Date | 5/31/15 3:58 AM