B Mikulec

Performance limits of a single photon counting pixel system

Abstract X-ray imaging using hybrid pixel detectors in single photon counting mode is a relatively recent and exciting development. The photon counting mode implies that each pixel has a threshold in energy above which a hit is recorded. Sharing of charge between adjacent pixels would therefore lead to a loss of registered hits and for medical imaging applications to a higher patient dose. This explains why the demand for high spatial resolution and consequently small pixel sizes ( 1 collaboration to study the effects of charge sharing between pixels on system performance. Two different simulation codes are used to simulate the energy loss inside the detector and the charge transport towards the pixel electrodes. The largest contribution to the lateral spreading of charge comes from diffusion and can result in a considerable loss of detection efficiency in photon counting systems for small pixel sizes.

Performance of a 4096-pixel photon counting chip

A 4096 pixel Photon Counting Chip (PCC) has been developed and tested. It is aimed primarily at medical imaging although it can be used for other applications involving particle counting. The readout chip consists of a matrix of 64 by 64 identical square pixels, whose side measures 170 micrometers and is bump-bonded to a similar matrix of GaAs or Si pixel diodes covering a sensitive area of 1.18 cm2. The electronics in each cell comprises a preamplifier, a discriminator with variable threshold and a 3-bit threshold tune as well as a 15-bit counter. Each pixel can be individually addressed for electrical test or masked during acquisition. A shutter allows for switching between the counting and readout modes and the use of static logic in the counter enables long data taking periods. Electrical test of the chip have shown a maximum counting and readout modes and the use of static logic in the counter enables long data taking periods. Electrical test of the chip have shown a maximum counting rate of up to 2 MHz in each pixel. The minimum reachable threshold is 1400 e with a variation of 350 e rms that can be reduced to 80 e rms after tuning with the 3-bit adjustment. Electrical noise at the input is 170 e rms. Several read-out chips have been bump bonded to 200 micrometers thick GaAs pixel detectors. Test with (gamma) -ray and (beta) sources have been carried out. A number of objects have been imaged and a 260 micrometers thick aluminum foil which represents a contrast to the surrounding air of only 1.9 percent has been correctly imaged.

X-ray imaging using single photon processing with semiconductor pixel detectors

More than 10 years experience with semiconductor pixel detectors for vertex detection in high energy physics experiments together with the steady progress in CMOS technology opened the way for the development of single photon processing pixel detectors for various applications including medical X-ray imaging. The state of the art of such pixel devices consists of pixel dimensions as small as 55 55 m 2 , electronic noise per pixel <100 e rms, signal-to-noise discrimination levels around 1000 e with a spread <50 e and a dynamic range up to 32 bits per pixel. Moreover, the high granularity of hybrid pixel detectors makes it possible to probe inhomogeneities of the attached semiconductor sensor.

Characterisation of a single photon counting pixel system for imaging of low-contrast objects

In the framework of the Medipix1 collaboration the PCC, a single photon counting pixel chip, has been developed with the aim of improving the contrast resolution in medical imaging applications. The PCC consists of a matrix of 64]64 square pixels with 170 lm side length, each pixel comprising a 15-bit counter and a pulse-height discriminator. The chip has been bump bonded to equally segmented 200 lm thick SI-LEC GaAs detectors showing a very high absorption energy for X-rays used in diagnostics. An absolute calibration of the system with a radioactive source and a synchrotron beam are described resulting in the value of the test input capacitance of &24.7 fF. Using this value a full characterisation of the system from electrical measurements is presented. The entire system can reach a minimum threshold of &2100e~ with &250e~ rms noise. One of the characteristic features of the PCC is the possibility to adjust the thresholds of all pixels on a pixel-by-pixel basis with 3-bit precision. The threshold distribution after adjustment is &120e~ rms. The spatial resolution of the system has been measured to be 3.6 lp/mm. A comparison of a tooth image taken with the PCC and with a screen-CCD-system demonstrates its imaging capabilities. ( 2001 Elsevier Science B.V. All rights reserved.

X-ray imaging using a hybrid photon counting GaAs pixel detector

Abstract The performance of hybrid GaAs pixel detectors as X-ray imaging sensors were investigated at room temperature. These hybrids consist of 300 μm thick GaAs pixel detectors, flip-chip bonded to a CMOS Single Photon Counting Chip (PCC). This chip consists of a matrix of 64 × 64 identical square pixels (170 μm × 170 μm) and covers a total area of 1.2 cm 2 . The electronics in each cell comprises a preamplifier, a discriminator with a 3-bit threshold adjust and a 15-bit counter. The detector is realized by an array of Schottky diodes processed on semi-insulating LEC-GaAs bulk material. An IV-characteristic and a detector bias voltage scan showed that the detector can be operated with voltages around 200 V. Images of various objects were taken by using a standard X-ray tube for dental diagnostics. The signal to noise ratio (SNR) was also determined. The applications of these imaging systems range from medical applications like digital mammography or dental X-ray diagnostics to non destructive material testing (NDT). Because of the separation of detector and readout chip, different materials can be investigated and compared.

Fixed pattern deviations in Si pixel detectors measured using the Medipix1 readout chip

Abstract Dopant fluctuations and other defects in silicon wafers can lead to systematic errors in several parameters in particle or single-photon detection. In imaging applications non-uniformities in sensors or readout give rise to fixed pattern image noise and degradation of achievable spatial resolution for a given flux. High granularity pixel detectors offer the possibility to investigate local properties of the detector material on a microscopic scale. In this paper, we study fixed pattern detection fluctuations and detector inhomogeneities using the Medipix1 readout chip. Low-frequency fixed pattern signal deviations due to dopant inhomogeneities can be separated from high-frequency deviations.

Performance potential of the injectors after LS1

The main upgrades of the injector chain in the framework of the LIU Project will only be implemented in the second long shutdown (LS2), in particular the increase of the PSB-PS transfer energy to 2GeV or the implementation of cures/solutions against instabilities/e-cloud effects etc. in the SPS. On the other hand, Linac4 will become available by the end of 2014. Until the end of 2015 it may replace Linac2 at short notice, taking 50MeV protons into the PSB via the existing injection system but with reduced performance. Afterwards, the H− injection equipment will be ready and Linac4 could be connected for 160MeV H− injection into the PSB during a prolonged winter shutdown before LS2. The anticipated beam performance of the LHC injectors after LS1 in these different cases is presented. Space charge on the PS flat-bottom will remain a limitation because the PSB-PS transfer energy will stay at 1.4GeV. As a mitigation measure new RF manipulations are presented which can improve brightness for 25 ns bunch spacing, allowing for more than nominal luminosity in the LHC.

Charge sharing studies with a Medipix1 pixel device

Abstract The Medipix1 chip is a prototype (digital) CMOS imaging chip that emerged from particle detection in high energy physics experiments. It was designed at CERN following specifications from the Medipix1 Collaboration. This contribution aims at demonstrating the applicability of Medipix1 Si and GaAs pixel devices for alpha and X-ray particle detection. A study of “charge sharing” between pixels and test of “spatial resolution” using narrow beam and “edge” image contrast was performed. For photons with energy

Applications of pixellated GaAs X-ray detectors in a synchrotron radiation beam

Hybrid semiconductor pixel detectors are being investigated as imaging devices for radiography and synchrotron radiation beam applications. Based on previous work in the CERN RD19 and the UK IMPACT collaborations, a photon counting GaAs pixel detector (PCD) has been used in an X-ray powder diffraction experiment. The device consists of a 200 μm thick SI-LEC GaAs detector patterned in a 64×64 array of 170 μm pitch square pixels, bump-bonded to readout electronics operating in single photon counting mode. Intensity peaks in the powder diffraction pattern of KNbO3 have been resolved and compared with results using the standard scintillator, and a PCD predecessor (the Ω3). The PCD shows improved speed, dynamic range, 2-D information and comparable spatial resolution to the standard scintillator based systems. It also overcomes the severe dead time limitations of the Ω3 by using a shutter based acquisition mode. A brief demonstration of the possibilities of the system for dental radiography and image processing are given, showing a marked reduction in patient dose and dead time compared with film.

LIU: EXPLORING ALTERNATIVE IDEAS

The baseline upgrade scenarios for the injector complex cover the connection of Linac4 to the PSB, the increase of the PSB-PS transfer energy from 1.4 GeV to2 GeV and the major SPS RF upgrade during LS2. The achievable beam characteristics will nonetheless remain below the expectation of the HL-LHC project. Therefore, alternative or additional options like, e.g., special bunch distributions, the use of injection optics optimized for high space charge or extra RF systems will be discussed. The expected beam parameters, possible implementation and impact on beam availability for these more exotic options will be analysed and compared to the LIU baseline plan. Moreover, the potential interest of further batch compression schemes will be evaluated.