High Resolution X-Ray Spectroscopy with Compound Semiconductor Detectors and Digital Pulse Processing Systems

An 8.72 µW Low-Noise and Wide Bandwidth FEE Design for High-Throughput Pixel-Strip (PS) Sensors

“…A typical thickness for Si-sensor is about 300 µm; the limiting irradiation energy, which would penetrate protons through the sensor, is about 6.2 MeV [ 5 , 7 ]. With moderate cooling by means of small Peltier cells, silicon drift detectors and Si-PIN sensors show particularly excellent spectroscopic performances and good detection efficiency below 15 keV [ 5 , 11 , 12 ]. In contrast to the spectroscopy amplifier, the major concern for a fast amplifier is the preservation of the charge collection process while keeping a wide bandwidth, which in turn optimizes the signal rise time [ 4 , 5 , 9 , 12 ].…”

“…With moderate cooling by means of small Peltier cells, silicon drift detectors and Si-PIN sensors show particularly excellent spectroscopic performances and good detection efficiency below 15 keV [ 5 , 11 , 12 ]. In contrast to the spectroscopy amplifier, the major concern for a fast amplifier is the preservation of the charge collection process while keeping a wide bandwidth, which in turn optimizes the signal rise time [ 4 , 5 , 9 , 12 ]. The improvement of energy resolution leads to optimization of the charge collection process by designing the lowest possible rise time of the charge sensitive amplifier (CSA) compared to the peaking time of the shaping amplifier; this would prevent ballistic deficit, which involves loss of resolution.…”

“…Therefore, the energy sensitivity of the readout module should be high enough to minimize the energy loss and guarantee a high rate collection process, which is characterized by its rise time (tr) and can be performed in less than 10 ns to guarantee high counting rate operations [ 4 ]. Moreover, for multi-channel readout electronics, the spatial resolution should be more than 2 µm [ 4 , 5 , 12 ].…”

“…This voltage is filtered and digitized by an Analog to Digital Converter (ADC) as shown in Figure 2 . The resulting data are then coded into an appropriate format so that pixel address, time, amplitude or transverse momentum [ 5 , 12 ] can be extracted through an FPGA module for further processing [ 11 , 15 ].…”

“…For a given source, the generated preamplifier noise and the input impedance of the amplifier influence the front-end noise performance. The impact of radiations on the devices exacerbates the situation [ 9 , 11 , 12 , 16 , 17 ]. Therefore, the front-end input stage must ensure that optimum noise matching is achieved for the source impedance [ 11 , 12 , 17 ].…”