LSC BETA SPECTRA ANALYSIS BY STANDARDIZATION BASED ON SHIFTING AND MULTI-WINDOW COMPUTING

LSC BETA SPECTRA ANALYSIS BY STANDARDIZATION BASED ON SHIFTING AND MULTI-WINDOW COMPUTING M. Buzinny Radioactivity calculations using spectrum windows were widely used in the LSC technique since the beginning especially for measuring low levels of 3H, 14C and Cherenkov counting. Thus for standardization of spectra were used one of the methods based on external standard. For a precise standardization of the spectra we use sets of calibration spectra measured under different quenching conditions, count the sample for a long time, we introduce a controlled shifting of sample spectra for processing and we apply multi-window calculation of high energy part of the spectra. Our approach to the processing of the spectra we use both for the old and the new LS spectrometers Quantulus 1220TM when measuring 14C using two types of 7 ml and 0.8 ml Teflon vials. To prove the method we analyzed results given by our approach when measuring samples prepared from the one quenched sample of benzene by various dilutions. We used six high energy windows that partially overlap. Sample spectra were controlled shifted until coincidence of spectra end-point. Visual comparison of spectra was performed using automatic selection of scale for each spectrum. Controlled sample spectra shift and a common visualization of set of spectra: calibration, sample and background allow achieving an acceptable coincidence of spectra endpoint. Another way of optimization is to perform set of calculation for each of several shifts of sample spectrum and chose an optimal based on standard deviation. Spectra processing program Easy View enables in one-step loading of calibration and background spectra and choose preset of windows and then save the processing parameters of the spectrum of the sample in a spreadsheet format for subsequent calculations. During the processing of all spectra we apply one step smoothing. However, it is not applicable in Easy View to make spectra shift by channel after applying several smoothing steps that hinder achievement matches end-points of spectra. The use of specially developed software code which compares net spectra: sample-background to calibration-background provides a much better coincidence for spectra endpoint, which is only attainable that way for low active samples. Performing calculations in multiple windows for shifted spectra allows obtaining optimal result and simultaneous assessing its uncertainty (discrepancy)