Characterization of Irradiated Depleted Monolithic Active Pixel Sensors in 150 nm and 180 nm CMOS Technology for High-Rate and High-Radiation Environments

Abstract

Modern tracking detectors in high-energy physics experiments target a minimum material budget and precise spatial and time resolution for optimized vertexing capabilities. Monolithic active pixel sensors (MAPS) combine sensitive volume and active readout electronics within a single entity of silicon. This eliminates the complex interconnection step and additional material required for the hybrid pixel detector approach. Recent advances in commercial CMOS technologies and increased availability of high-resistivity silicon substrates facilitate developments of MAPS with large depleted volume and, thus, fast charge collection primarily via drift. These so-called depleted monolithic active pixel sensors (DMAPS) form a viable alternative to hybrid pixel detectors, even in high-rate and high-radiation environments. The Monopix2 chips characterized in this thesis constitute two DMAPS with different design approaches for application in such environments.<br/>

For LF-Monopix2, all in-pixel electronics are implemented within a large charge-collection electrode relative to the pixel pitch that yields a homogeneous electric field across the sensor. Combined with the resulting short drift distances within a pixel cell, this design approach facilitates high radiation tolerance. Within the scope of this thesis, the radiation tolerance of LF-Monopix2 is examined up to levels of 5 x 10¹⁵ n_eq cm⁻² NIEL fluence and 400 Mrad total ionizing dose (TID). Due to the high bias voltage capabilities, 100 µm thin samples can be fully depleted up to NIEL fluences of 3 x 10¹⁵ n_eq cm⁻². A performance degradation between 1 and 10 Mrad TID and a subsequent recovery characteristic for this technology feature size is observed during X-ray irradiation. Beam tests of LF-Monopix2 verify hit-detection efficiencies above 99 % after irradiation.<br/>

In the case of TJ-Monopix2, the in-pixel electronics are separated from the small charge-collection electrode, which facilitates minimizing the pixel pitch. The resulting small detector capacitance reduces the sensor noise and, thus, requires less power consumption of the electronics. Due to the complex electric field configuration of this design approach, additional process modifications are required to improve fast charge collection via drift, especially after irradiation. As part of this thesis, the origin and extent of a periodic threshold fluctuation relative to the hit arrival time is investigated. The functionality of TJ-Monopix2 after irradiation to 1 x 10¹⁵ n_eq cm⁻² NIEL fluence and 100 Mrad TID is verified and beam tests confirm hit-detection efficiencies above 99 %.