Vytautas Astromskas (U)
Pixellated CdTe semiconductor detectors provide the stopping power that is needed for experiments performed at synchrotron radiation facilities at high photon energy (over 20 keV). Previous experiments conducted with CdTe Medipix detectors with Ohmic contacts showed defects that affect the image quality, as they cannot be corrected by flat field coefficient factors. Electron collection Schottky contact CdTe sensors show better imaging performance, however, they are affected by polarization. The focus of this project is to characterise and evaluate the performance of an electron collection Schottky contact CdTe sensor in terms of stability over time, degradation of performance with regards to temperature and flux since experiments at synchrotrons have high ionizing radiation flux, as well as, to understand the physics behind the defects. An Acrorad electron collection Schottky contact CdTe sensor bump-bonded to a Medipix3RX ASICs developed by CERN and readout by the Merlin readout system developed by Diamond Light Source were tested for this study. The CdTe semiconductor sensor was segmented into a 128x128 pixel matrix with each pixel being 110μm pitch. The Medipix3 ASIC is a 256x256 pixel array with 55 μm pitch, every fourth pixel of the ASIC is bump-bonded to the sensor. Every pixel of the chip has two discriminators. Since the Medipix3 enable to cluster the discriminators of four pixels, each pixel of the sensor is equipped with eight discriminators. This allowed setting separate up to eight energy threshold windows for each of the 110μm pixels. This mode of operation is known as colour mode and gives spectral information in addition to spatial resolution. Three single type sensors of 128x128 pixel array and three quad type sensor with 256x256 pixel array (bump-bonded to 4 Medipix3 ASICs) were tested during this study. The aim of the study was to test the behaviour the aforementioned setup in terms of uniformity, linearity, polarization effects and to create a map of optimum conditions such as temperature, flux and bias refresh time in order to find the best operational settings. The detectors were tested under various temperature (from 12°C to 24°C) and flux (2.5kcps per pixel to 30kcps per pixel) conditions in order to characterise the degradation of the pixel response over time. The degradation of the pixel response was studied by looking into each pixel’s count rate variations and the changes in the total count rate inside the region of interest (ROI) over time. The tests showed that cooling the detector resulted in the reduction of the rate of degradation of the pixel response while higher flux resulted in increased degradation. Also, turning the applied bias voltage off for a short period of time and turning it back on depolarized the detector and, hence, recovered the performance of the pixels to the original state. Finding the optimum bias refresh time is crucial for optimal operation of the detector. Despite the rate of the degradation being comparable for both of the single type detectors, two types of defects appeared after the detector was polarised. The defects were studied by investigating the response of each of the pixels in the ROI with an anomalous response. In the first case, the central pixel went dead and increasingly reduced the sensitivity of the nearby pixels over time. In the second case, the central pixel went dead but the perpendicular pixels resulted in higher sensitivity. This study showed that electron collection Schottky CdTe sensors can provide good imaging quality when operated under optimized operational conditions.
Vytautas Astromskas (U)
Annika Lohstroh (University of Surrey) Dr Eva N. Gimenez (Diamond Light Source) Dr Nicola Tartoni (Diamond Light Source) Paul Sellin (University of Surrey)