SCARLET: A Readout ASIC Bump-Bonded to SDD Array for Large Event Throughput

This work introduces silicon drift detector-ASIC array for large event throughput (SCARLET), a novel readout ASIC designed for integration with monolithic arrays of silicon drift detectors (SDDs) using a bump bonding scheme within a hybrid pixel assembly, suitable for high-rate high- density energy-dispersive X-ray detection systems in synchrotron beamlines or industrial applications. The ASIC, designed in a standard 0.35 mu m CMOS technology, is composed of four parallel readout channels, each integrating the complete pulse processing electronics, from the charge-sensitive amplifier to the analog-to- digital converter (ADC), delivering the signal in a digital format. Specifically, the pulse processing chain features a reset-type CSA with pMOS input stage, a seventh-order semi-Gaussian shaping amplifier, a peak stretcher equipped with peak detector circuit and optimum pile-up rejection (PUR) scheme, and an analog memory to enhance throughput. Each pair of channels is digitized using a priority multiplexing scheme by a 12-bit on-chip successive approximation register (SAR) ADC, providing a digital output at a maximum sampling rate of 5 Ms/s. This enables the allocation of 2.5 MHz to each channel when the multiplexer operates sequentially between them. The layout floorplan is compatible for bump bonding assembly by means of gold-stud bump technique with 2 mm pitch square SDDs. Experimental spectroscopic measurements conducted with the hybrid module yielded a resolution of 240 eV FWHM (24.4 e-rms) at the shortest filter pulsewidth of 200 ns. A best resolution of 148 eV FWHM (10.3 e-rms) was achieved at the optimum filter pulsewidth of 1 mu s. In terms of throughput capability, the ASIC reaches an average output count rate (OCR) of 1.8 Mc/s per channel. When two channels are irradiated simultaneously with an input count rate (ICR) of 4 Mc/s, a maximum count rate of 3.6 Mc/s is measured at the output of the shared ADC operated in priority modality.