A dedicated readout ASIC for Time-of-Flight Positron Emission Tomography using Silicon Photomultiplier (SiPM)

Abstract

STiC is a mixed-mode Application Specific Integrated Circuit (ASIC) for Silicon Photomultiplier (SiPM) readout with very high timing resolution. It is designed for the Time-of-Flight (ToF) measurement in Positron Emission Tomography (PET) and in high energy physics experiments. STiC is dedicated particularly to the EndoTOFPET-US project, which is developing and building a multi-modal instrument combing ToFPET and ultrasound endoscopy for the development of new biomarkers for pancreas and prostate oncology. A 64-channel prototype (STiC v3) has been developed and produced in 0.18µm UMC CMOS technology. A fully differential analog front-end has been developed to suppress the common-mode noise from both the on-chip digital part and the outer sources, while it allows for both differential and single-ended readout connection schemes with SiPM. A digital-to-analog converter on the input stage tunes the SiPM bias voltage with a range of ∼ 900 mV, providing the possibility to compensate the variation on the breakdown voltage of the connected SiPMs. A linearized time-over-threshold method has been implemented to provide better energy resolution than the conventional ToT method. The time and energy information are converted into two time stamps, which are digitized by a build-in TDC with time binning of 50.2 ps. A dedicated digital part has been developed for chip configuration with a Serial Peripheral Interface and for data transmission to an external Data Acquisition System using a 160 MBit/s LVDS serial link with 8/10-bit encoding. Measurements have shown a time jitter < 20 ps on the analog front-end and a time jitter of ∼ 37 ps on the TDC and digital part. A Coincidence Time Resolution of ∼ 214 ps FWHM has been obtained by STiC v3 with 3.1 × 3.1 × 15 mm2 LYSO:Ce crystals and Hamamatsu MPPCs (S12643-050CN(x)). Here we present the details of the ASIC design, measurement results as well as the 128-channel front-end module for EndoTOFPET-US project.