Year
2013
Abstract
Organic scintillation materials are important for many applications because they are sensitive to fast neutrons and photons. However, the usefulness of the measured data is limited by the ability to accurately distinguish between neutron and photon events. A commonly used method is a two-dimensional charge integration, which classifies incident particles based on the ratio between the pulse tail and total integrals. This method works reasonably well for energy depositions above approximately 600 keV for neutrons and 70 keV for photons. However, at lower energies, the separation becomes poor, leading to higher rates of misclassified particles. This deficiency is further exacerbated by the fact that tail-to-total discrimination curves are commonly determined manually at ranges of good visual separation and extrapolated to lower energies. A significantly more accurate and robust method has been developed to assist the user in choosing the ideal points through which the discrimination curve is fitted. This is a multi-step process that includes segmenting the full data set into smaller slices, fitting the data in each slice as a sum of two Gaussians, and choosing the optimal discrimination point for each slice based on the acquired fit. After the optimal points have been determined, they can be used as the basis for fitting a discrimination curve. This process has been automated with a graphical user interface that visualizes the data and provides important analyses, including the final discrimination curve. This computer-aided, visual, pulse-shape-discrimination method provides an optimized discrimination curve based on the two-dimensional, charge-integration method by eliminating a common source of human error from the data-analysis process.