Future Colliders

Abstract:

The MALTA2 sensor is the second prototype in the MALTA family of Depleted Monolithic Active Pixel Sensors, developed using a modified 180nm CMOS imaging process and optimized for operation in the high-radiation and high hit-rate conditions of future collider experiments. With a matrix of 224 × 512 pixels and a 36.4µm pitch, MALTA2 is designed for fast charge collection, low noise, and low power consumption. MALTA2 incorporates a low doped n − layer extending across the pixel matrix, which enhances the lateral depletion and charge collection by drift. This work evaluates a MALTA2 variant with very high n − layer doping for improved radiation hardness. The sensor has been neutron irradiated up to a fluence of 5 × 1 0 15 1 MeV n eq cm −2 and characterized in test beam campaigns at the CERN SPS using a 120 GeV/c mixed hadron beam. Hit detection efficiencies above 95% were achieved up to a fluence of 3 × 1 0 15 1 MeV n eq cm −2 . At 5 × 1 0 15 1 MeV n eq cm −2 , the hit detection efficiency measured was 64.9 ± 0.2 % at a substrate bias voltage of -60 V. To investigate this degradation, the effective active depth has been extracted as 20.3 ± 0.2 μ m from cluster size measurements as a function of the beam incident angle. The results from the very high doping will guide the development of prototypes featuring ultra-high n − layer doping, aiming to achieve greater radiation hardness in next-generation silicon tracking detectors.

Abstract:

This paper demonstrates a novel method to extract photomultiplier tube (PMT) calibration timing constants in large liquid scintillation detectors from physics data using the machinery of unsupervised deep learning. The approach uses a simplified physical model of optical photon transport in the loss function, with PMT calibration constants treated as free parameters, and the simple assumption that individual events represent point-like emission. The problem is, thus, effectively reduced to that of regression on a very large scale, made tractable by deep learning architectures and automatic differentiation frameworks. Using data from the 9300 PMTs in the SNO+ detector, the method has been shown to reliably extract 3 calibration constants for each of the over 7500 online PMTs using radioactive background events. We believe that this basic approach can be straightforwardly generalized for a wide range of applications.