Readout electronics for high-rate high-resolution energy-dispersive X-ray detection system

My doctoral thesis focuses on the preliminary study, design, and characterization of readout electronics, both in circuit-level and system-level, adapted for high-rate high-resolution energy-dispersive x-ray detection system. X-ray spectroscopy is a technique widely used in applied physics, e.g. X-ray Fluorescence (XRF), X-ray Absorption Spectroscopy (XAS), X-ray Microprobe (µXRF), and many other beamline-enabled X-ray experiments, as well in many industrial instruments, e.g. energy-dispersive system for SEMs, portable XRF analyzers, macro XRF scanner, X-ray diffraction, etc. Therefore, the possible applications could range from fundamental research to cultural heritage analysis, mining and mineralogy measurements, as well as safety, food and pharmaceutical studies. In the past few years, detection module for X-ray spectroscopy based on Silicon Drift Detector (SDD) with a concentric anode, coupled with CMOS monolithic Charge Sensitive Amplifier (CSA), has shown to be very competitive in terms of performances and ease of fabrication. CMOS-based CSA is characterized by high transconductance, when coupled to SDD that has relatively small anode capacitance, it results in excellent noise performance and thus, good energy resolution. The combination is also known to be compatible with high-rate operation, provided several conditions such as small detector unit dimension, low-temperature operation, and fast detector pulse processing system. High-rate operation is essential to some applications, for example, in energy-dispersive detector in synchrotron beamlines performing experiments such as XRF, XAS, and µXRF. The requirement is driven by the ever-increasing flux of x-ray incident beams in upgraded synchrotron facilities, leading to higher fluorescence flux available for the detector, potentially improving the quality of the acquired data and/or reduce significantly experiment time. The first part of the thesis focuses on this question: can we develop an X-ray spectrometer, adapted for experiments in synchrotron beamline, that can provide good energy resolution and high throughput capability, with immunity to electrical disturbances typically present in the beamlines? The conclusions for these questions have been achieved through the milestones in the ARDESIA project. ARDESIA (Array of Detectors for Synchrotron Radiation Applications), is an SDD-based, 4-channel X-ray spectrometer, optimized for synchrotron experiments such as X-ray fluorescence (XRF) and X-ray absorption fine structure (XAFS). Its detection module, has been established previously, is able to achieve high-count rate (>1Mcps/channel) and high-resolution (≈ 125eV of FHWM Mn-Kα line at optimum shaping time, ≤ 200eV at short shaping times) X-ray fluorescence detection. During the course of my doctoral study, the complete instrument, including internal and auxiliary electronics, has been assembled and tested successfully in both laboratory and synchrotron beamline settings. In DAфNE-Light DXR1 beamline, satisfactory results of XRF and XANES acquisition have been achieved in soft X-ray regime. In LISA BM-08 ESRF beamline, measurements including long-duration XAFS measurements and XAFS measurement on trace elements confirm the qualification and performance of the instrument, in terms of energy resolution, throughput capability, immunity against external disturbances, and stability over time. High-rate spectroscopy can only be realized with SDD-CSA based detection system, only if the following stages provide fast pulse processing. In a system with a limited number of detection channels (<10 channels), Digital Pulse Processor (DPP) is the most commonly adopted pulse processing solution, thanks to its satisfactory spectroscopic performance, high throughput, and robustness of digital system. On the other hand, when high-density multichannel detection is needed, as in future synchrotron-based experiments, or when the power and/or area constraints apply, Analog Pulse Processor (APP), implemented in an integrated circuit, is still an attractive solution, since it can provide compact and cost-effective solution compared to DPPs. ASIC-based APPs, if exist, with relatively high-throughput capability, appear to be a desirable solution to equip a compact instrument with high-density multichannel readout electronics (50-100 channels) within a reasonable cost and power budget. The second part of the thesis raises these questions: is it conceptually feasible to create an analog pulse processor with throughput capability above 1Mcps output count rate with satisfactory energy resolution and spectrum quality? If yes, can we implement it in ASIC and demonstrate experimentally such performance? In my doctoral study, the questions are first tackled by theoretical study of a fast APP solution. Systematic assessment of various shaping amplifiers has been conducted, considering series noise, ballistic deficit immunity, and pile up immunity. Estimation and validation on its ballistic deficit effect to additional spectrum broadening are also explored. The study is concluded with analysis of analog-based pile up rejection (PUR) strategies and maximum throughput estimation taking into account several parameters on analog processing channel. The assessment provides some relevant insights to define the specification of TERA (Throughput Enhanced Readout ASIC), an APP implemented as an ASIC with a purpose to demonstrate experimentally the results obtained in the study. The ASIC development has been started from its predecessor, SFERA (SDD Front-End Readout ASIC), designed in the same 0.35um technology node. Each pulse processing channel includes a 7th-order semi-Gaussian shaping amplifier with controllable shaping times and full scale ranges, followed by a peak stretcher and a switched-capacitor analog memory. Each channel is also equipped with a dedicated peak detector and a novel pile up rejection (PUR) logic. Each pair of channels can be optionally digitized by a 12-bit on-chip SAR ADC. The architecture enables to achieve high throughput and satisfactory energy resolution. In fact, in Fe-55 spectroscopy measurements, when using the shortest shaping pulse width of 200 ns, 2.5 times shorter than the smallest shaping time implemented in SFERA, a FWHM Mn-Kα line of 159.4 eV was obtained at low rates (10 kcps). This energy resolution can be achieved thanks to the minimization of the effect of the ballistic deficit, achieved by optimizing the SDD detector in terms of optimum biasing, low operating temperature (-37oC), and small size (1 mm diameter). At 1.61 Mcps/channel input rate, an output count rate of 1.09 Mcps/channel and a resolution of 205.1 eV were achieved with a 4mm-diameter SDD and optimum PUR settings. Such high-rate performances are, to our knowledge, the best ones reported so far for an APP, based on a spectroscopy-grade ASIC, and very close to those achievable by a standard DPP. Therefore, TERA can represent an attractive, compact, and scalable pulse processing solution for high-rate multichannel energy-dispersive X-ray detection systems.

La mia tesi di dottorato si concentra sullo studio preliminare, la progettazione e la caratterizzazione dell'elettronica di lettura per sistemi di rivelamento di raggi X. In particolare, mi concentro sulla spettroscopia energetica ad alta risoluzione e ad alti conteggi, sia a livello di circuito che a livello di sistema. La spettroscopia a raggi X è una tecnica ampiamente utilizzata nella fisica sperimentale, e.g. X-ray Fluorescence (XRF), X-ray Absorption Spectroscopy (XAS), X-ray Microprobe (µXRF) e molti altri esperimenti di raggi X alla linea di luce di sincrotrone, nonché in molti strumenti industriali, e.g. sistemi di dispersione energetica per SEM, analizzatori XRF portatili, scanner XRF macro, diffrazione di raggi X, ecc. Pertanto, le possibili applicazioni vanno dagli esperimenti di fisica all'analisi del patrimonio culturale, alle misurazioni minerarie e di mineralogia, nonché alla sicurezza, alimentare e farmaceutica.