U235 / U238 izotopik oranının pasif gama ışını spektroskopisi ile tayini

Abstract

ÖZET U235/U238 izotopik oranının belirlenmesi gelişen nükleer teknolojiyle önem kazanmıştır. Bu oranın belirlenmesinde en çok kullanılan metodlar, gama spektroskopisi, nötron aktivasyonu, alfa ve kütle spektrometresidir. Bu çalışmada izotopik oranın belirlenmesinde, mevcut imkanlar gözönüne alınarak pasif gama spektroskopisi yön temi seçilmiş ve elde edebildiğimiz bazı numunelerde bu oran belirlenmiştir. U235/U238 izotopik oranının belirlenmesinden önce, gerek dedektör gerekse sayım sisteminin performanslarını belirleyici bir seri ölçümler yapılmıştır. özellikle bunlar arasında dedektörün özelliklerini ve backgroundun etkisini belirleyen hususlar incelenmiştir. Ü235/U238 oranının belirlenmesi için U238 çekirdeklerinin sayısı, U238 'in alfa bozunması sonrasında meydana gelen Th234 'ün beta bozunması sonrasında gönderilen 63 keV ve 93 keV enerjilerdeki dubletler çözülerek örnek içinde ki ü238 çekirdeklerinin sayısı ise bu çekirdeğin alfa bozunmasını takiben gönderilen 186 keV gama ışınının şiddetinden gidilerek bulunmuştur. Ölçülen örneklerde U235/U238 izotopik oranı uranilnitrat ihtiva eden numune için %» 4.3, depleted uran ihtiva eden peletlerde ise %> 0.6 değerleri bulunmuştur. iv

SUMMARY DETERMINATION OF U235/U238 ISOTOPIC RATIO BY PASSIVE GAMMA-RAY SPECTROMETRY The determination of u235/U238 isotopic ratio is of great importance in nuclear technolgy, because in many nuclear power reactors enriched fuel is needed. There are some convenient methods to determine this ratio, like thermal neutron activation analysis followed by high resolution gamma-ray, mass, passive gamma-ray, alpha spectrometries and delayed neutron counting. In this work, the passive gamma-ray spectrometry was chosen as the most convenient method which could be made in nuclear physics labratory of Istanbul Technical Uni versity, Department of Physics. This method is based on the measuring of the gamma- rays originate from the decay of u235,U238 and daughters of them. To obtain U235/U238 isotopic ratio, strong gamma-ray at 185.7keV from U235 and doublets at 63 keV and 93 keV of Th234 daughter of U238 decay were used. Activation calculation had been made for doublet in 93 keV (92.35 and 92.78) of Th234 and also for 63.29 peak of 63 doub let. In secular equilibrium the number of U238 atoms can be calculated from the number of Th234 atoms because half life of uranium is much longer than the half life of tho- urium. Samples that we used, were mostly natural uranium and depleted uranium sources. Isotopic ratio determina tion were done for uranylnitrat (U02(N03)2+6H20) from Fluka Inc., Sweden and depleted uranium plates from Reac tor Experiments Inc, U.S.A. and one thick pellet with un known composition. The self absorbtion effect was neg lected for uranylnitrat samples because of it's low den sity and thin thicknesses. Figure 1 shows the spectrum of uranylnitrat sample.4E*3n 3E«-3- £2E*3- 1E*3 0E*8- CO 3 OD s* 4- *-*^K*<A^A CD 400 i i i i - r -t-i- i i 1 I 1200 KANAL i r i i i 1600 t i i t r i 2000 Figure 1 : The gamma spectrum of an uranylnitrat sample. However self absorption for depleted uranium samples was seen quite high in some cases and not to be neglec ted. Hence in depleted uranium samples self absorption effect depend on thicknesses had been measured. Main parts of experimental system are detector, mul tichannel analyzer Canberra System 100, interactive fit ting program for gamma -ray spectra taken with high reso lution solid state detectors (Sampo 90) and Pb shilding for the absorption of the background radiation. Detector used in experiments is an extended range (XTRA) HpGe with high efficiency (%20) and has 0.5 mm thick Be window. Because of berilium's very low Z value it is possible to measure the gamma-rays nearly down to 5 keV with this detector. In this work first of all some tests of the whole experimental system were undertaken to find out the per formance of the system compared to the values given by the manufacturer. viEspecially for the detector efficiency, energy reso lution, peak-compton ratio, etc.. tests had been done. By using standart calibration sources (Am241, Ba133, Cs137, Co60) the absolute efficiency of the detector had been done for 6 cm detector-source distance. In addition to these standart calibration sources an Eu132 source (activity has not known) had been used to de termine efficiency curve better. To see how good the efficiency curve had been done, a Cs137 source with well known activity (1.033 uCi) was used to check the efficiency values. The activity was found as 1.024 uCi with an error less than 6 % which is in good agreement with given value. Also the effect of the Be-windowed detector to the low energy part of the gamma spectra was studied. For Al -windowed detector, the efficiency curve fit ting was possible with a suitable fitting function. For Be-windowed detector, no function could have been found to fit the efficiency values at all. All the given fit ting functions in literatures have very fast decreasing on the low energy part, on the contrary, our curve has smooth region between 20 keV-100 keV. Errors of experimental values in efficiency measure ments was calculated due to errors in measuring of dedec- tor-source distance, uncertainities of activities and intensities of the gamma-rays. In worst case total error in efficiency values were less than 9 %. Figure 2 shows the efficiency of the detector between 10 keV and 1500 keV. Gamma-ray spectra had been taken with a PC based multichanel analyzer (Canberra System 100). System 100 has it's own microprocessor (Intel 80186) so it is pos sible to make it work independently in one way from per sonal computer. It has many easinesses mentioned in the work for user like using the screen with groups, energy calibration, live or true time settings, peak information options and region of interest processes. To obtain numerical results from the measured spect ra, spectra should be analyzed and the peaks have to be fitted with a computer program. For this reason we use new version of a fitting program called Sampo 90 which could be run by PC. vii.o to o2 10 I/' Be window » ! i 1 1 LI I I, I l ' i i i i i i I i io too : 1000 ENERGY (keV) Figure 2 : Absolute efficiency of the detector at 6 cm. One of the most convenient program among the other commercial gamma spectroscopy programs like Spectran AT, ND Hypermet, Apogee etc... The main superiority of this program is interactive working. The program uses least square fitting procedure and shows the residual distrubi- tion in the fitting area on the screen. The goodnees of the fitting procedure which have been made, can be seen easily by checking %2 and the residuals. In Sampo 90, background can be chosen as linear or second degree func tion. Before making any process in Sampo 90 shape, ener gy and efficiency calibration files have to be created to fit any spectrum. Shape calibration procedure makes a relation between peak shape and the energy. It uses neccessarly three pa rameters to express peak shape. These parameters are low tail, high tail and center of the peak. viiiIn sampo 90, efficiency values for various energies had been done before it can be used directly. For every measurement, peak shape and energy calib rations should be made separately for standart gamma sources which have well known energies and many isolated peaks with comparable intensities. For the low level counting measurements in the la boratory, the effect of the background has to be conside red also. Hence a Pb shilding was used and absorption capacity of it was measured in the work. In our background spectrum, it was seen many of the gamma lines from various natural radioactive decaying sources and the effect of them were subtracted from the counts belong to measuring samples. ix