Ανάπτυξη διάταξης άμεσης ανάλυσης με νετρονική ενεργοποίηση για βιοϊατρικές εφαρμογές

Abstract

The thesis concerns the development of a Prompt Gamma Neutron Activation Analysis(PGNAA) facility for the measurement of nitrogen and thus total protein in large volumebiological samples or the whole body of small animals. The method is based on detection ofthe 10.83 MeV nitrogen prompt γ-rays, which are emitted following the neutron capture from 14N nuclei in the sample. The experimental device consists of a 185 GBq 239Pu-Be neutronsource, a graphite neutron collimator, radiation shielding, a shielded NaI(Tl) γ-ray detector anda sample holder. The device design aimed on the optimization of the detected nitrogen signal.Experimentally, optimal signal is achieved with minimization of the statistical error ofmeasurement of nitrogen at a set measurement time of the sample (or radiation dose).The design of the device was based on Monte Carlo simulations using code MCNP-4C2.Computations aimed to maximize thermal neutron flux over the volume of sample, minimizethe absorbed dose in the sample as well as the fluence rate incident on the gamma ray detector.The selection of the neutron source (239Pu-Be , 241Am-Be, 252Cf ) in combination with thecollimator length and material (graphite, polyethylene, polyethylene with 5% boron impurities,‘heavy water’ and beryllium) were studied as well as the possibility of using graphite belowthe source as a reflector. Furthermore, different detector position arrangements and shieldingconfigurations were studied. The computational results were verified against measurements forselected configurations. The calculation of the equivalent dose of neutrons and photons wasperformed using code MCNP-4C2.The production of 10.83 MeV prompt γ-rays depends on sample elemental composition andhomogeneity, the experimental configuration and the source type. The quantity ‘CombinedSensitivity’ (CS) was used to represent the number of γ-rays of nitrogen emitted from thesample per unit of time and volume of the sample which were detected by the detector. Thedistribution of CS over sample volume provides a measure of the non-uniformity of nitrogenmeasurement to small animal body and therefore is an index of the accuracy of themeasurement. The study included cylindrical phantoms, filled with ‘equivalent tissue’ liquidand different nitrogen concentration values inside internal cavities. As an extreme value, thepresence of air cavities within the sample was also studied as it representing small animal’slungs.The PGNAA calibration procedure was based on measurements performed using a 1 L involume cylindrical calibration standard of known nitrogen concentration. As a result, thedetected count ratio of nitrogen per unit of nitrogen mass (Fst) represent a standard index forthe PGNAA facility. MCNP-4C2 was used to derive appropriate correction factors for thesystem response in order to account for differences in size between measured samples and thecalibration standard. The accuracy of nitrogen determination by PGNAA in phantoms ofknown composition (2.7% - 10.9% N mass concentration of was studied. Total nitrogen wasmeasured in four raw ground meat samples using the PGNAA facility. Nitrogen was alsoassessed in the meat samples independently by the Kjeldahl and Dumas combustiontechniques with similar results.This study showed that prompt gamma ray neutron activation analysis is a non-destructivenuclear analytical technique which allows determination of nitrogen concentration inbiological samples, 0.25 – 1.00 kg in mass and the whole body of small animals.