Reconstructing Cluster Information from Silicon Pixel Detectors

This project was completed through the Research Experience for Undergraduate (REU) program at the University of Kansas. Current particle detectors containing billions of pixels process images taken millions of times a second to obtain measurements with precise position and time resolution. These silicon pixel detectors can also be used to monitor a beam of thermal neutrons by inserting a boron conversion layer. We utilized Allpix Squared, a generic simulation framework to simulate 100,000 neutrons impinging on a 14x14mm silicon sensor with 55 μm² pixels. From this, we reconstructed the simulated cluster data using python code to find the spatial resolution by measuring the difference between reconstructed positions and the original, simulated position. This resulted in a 1 μm cluster position resolution. With the large number of pixels in each cluster, one can find the angle of incidence, with respect to the sensor and ultimately “track” the position of the beam. There will be too much data collected for the present readout electronics to handle at speed. This project lays the groundwork for future innovations in imaging technology that will employ an algorithm that will mimic the human brain, using local processing and storage within the detectors becoming more energy efficient.