Study of the internal quantum efficiency of FBK sensors with optimized entrance windows

Abstract

Abstract Single-photon detection of X-rays in the energy range of 250 eV to 1 keV is difficult for hybrid detectors because of the low quantum efficiency and low signal-to-noise ratio. The low quantum efficiency is caused by the absorption of soft X-rays in the entrance window of the silicon sensors. The entrance window consists of an insensitive layer on the surface and a highly doped layer, which is typically from a few hundred nanometers to a couple of micrometers thick and is comparable to the absorption depth of soft X-ray photons (e.g. the attenuation length of 250 eV X-ray photons is ∼100 nm in silicon). The low signal-to-noise ratio is mainly caused by the small signal amplitude (e.g. ca. 70 electrons for 250 eV X-ray photons in silicon) with respect to the electronic noise. To improve the quantum efficiency, the entrance window must be optimized by minimizing the absorption of soft X-rays in the insensitive layer, and reducing charge recombination at the Si-SiO2 interface and in the highly doped region. Low gain avalanche diodes (LGADs) with a multiplication factor between 5 and 10 increase the signal amplitude and therefore improve the signal-to-noise ratio for soft X-rays, enabling single-photon detection down to 250 eV. Combining LGAD technology with an optimized entrance window technology can thus allow hybrid detectors to become a useful tool also for soft X-ray detection. In this work we present the optimization of the entrance window by studying the internal quantum efficiency of eight different process technology variations. The sensors are characterized using light emitting diodes with a wavelength of 405 nm. At this wavelength, the light has an absorption depth of 125 nm, equivalent to that of 276 eV X-rays. The best variation achieves an internal quantum efficiency of 0.992 for 405 nm UV light. Based on this study, further optimization of the quantum efficiency for soft X-rays detection is planned.