The high-luminosity upgrade of the ATLAS and CMS experiments includes dedicated sub-detectors to perform the time-stamping of minimum ionizing particles (MIPs). These detectors will be exposed up to fluences in the range of 1.5-2.5 x 10(15) eta(eq)/cm(2) and require a time resolution per detecting layer of 30 ps, for non-irradiated sensors, to 50-70 ps (depending on the exposed fluences) for sensors at the end of their lifetime. To cope with these requirements, the low-gain avalanche diode (LGAD) has been chosen as the baseline detection technology. In this article, an in-depth radiation tolerance study on LGADs manufactured at IMB-CNM using a so-called shallow junction is presented. Proton irradiation at CERN-PS up to fluences of 3 x 10(15)eta(eq)/cm(2) and neutron irradiation at JSI-Ljubljana up to 2.5 x 10(15) eta(eq)/cm(2) were performed. Two different active thicknesses were studied: 35 mu m and 50 mu m. Gain degradation, operation stability, and timing performance were evaluated.

LGAD technology is established within the field of particle physics, as the baseline technology for the timing detectors of both the ATLAS and CMS upgrades at the HL-LHC. Pixelated LGADs have been proposed for the High Granularity Timing Detector (HGTD) and for the Endcap Timing Layer (ETL) of the ATLAS and CMS experiments, respectively. The drawback of segmenting an LGAD is the non-gain area between pixels and the consequent reduction in the fill factor. In this sense, inverse LGAD (iLGAD) technology has been proposed by IMB-CNM to enhance the fill factor and to reach excellent tracking capabilities. In this work, we explore the use of iLGAD sensors for X-Ray applications by developing a new generation of iLGADs. The periphery of the first iLGAD generation is optimized by means of TCAD tools, making them suitable for X-Ray irradiations thanks to the double side optimization. The fabricated iLGAD sensors exhibit good electrical performances before and after an X-Ray irradiation. The second iLGAD generation is able to withstand the same voltage, as contrary to the first iLGAD generation after irradiation.

LGAD detectors on 300 mu m thick high resistivity p-type substrates were proposed for the first time by IMB-CNM-CSIC. They are customized Avalanche Photodiodes (APD) to obtain a high electric field region confined close to the reversed junction. Therefore, only electrons generated by an incident particle passing through the detector and drifting to the n+ contact, start the impact ionization process. Thus, the collected charge is multiplied. The basic difference between APDs and LGADs is the gain. LGADs have a moderate gain in order to avoid the inherent problems due to high multiplication: cross talk and high noise. In that way, the detector signal can be kept high without increasing the noise. These devices have been successfully fabricated and extensively characterized, before and after irradiation. Unfortunately, neutron and proton radiation cause the degradation of the gain and the creation of bulk traps, degrading the timing resolution. One way to reduce the radiation induced degradation is to minimize the substrate thickness, thus improving the timing resolution of LGAD detectors. Two technology approaches have been contemplated: the use of SOI (Silicon on insulator) substrates and Silicon to Silicon bonding substrates, both with a very thin active silicon layer of 50 mu m. As a consequence, drifting distances of generated electrons and holes are significantly reduced, resulting in a decrease in the number of electrons and holes trapped by radiation induced bulk defects. A new family of thin detectors, produced in 2x2 arrays prototypes, for the ATLAS experiment High Granularity Timing Detector (HGTD) is proposed. These detectors are suitable for timing applications with time resolution in the range of 30 ps at 20 degrees C. Optimization of the LGAD structures for the HGTD experiment and the detector experimental performances are presented and discussed.