Novel position reconstruction methods for highly granular electromagnetic calorimeters

Department of Physics and Astronomy, University of Kansas, Lawrence, KS 66045, USA

^* e-mail: brendon_madison@ku.edu

Published online: 18 December 2024

Abstract

We present work on design and reconstruction methods for sampling electromagnetic calorimeters with emphasis on highly granular designs. We use the clustered logarithmically weighted center-of-gravity estimator (lwk-means) for initial benchmarking of position resolution. We find that the θ and φ resolution for high energy photons in Si-W designs improves when increasing both sampling frequency and sampling thickness. Augmenting only one is found to have mixed results. We find that lwk-means is unable to effectively use calorimeter transverse cell sizes smaller than 2 mm. New reconstruction methods for highly granular designs are developed. We find that methods that only measure the initial particle shower and disregard the remaining shower can take advantage of cell sizes down to at least 10 µm, significantly outperforming the benchmark method. Of these, the best method and design is the initial particle shower “single hit” method using the calorimeter design with the highest sampling frequency and sampling fraction.