Pixel Readout Chip Research Articles

SEU tolerant memory design for the ATLAS pixel readout chip

The FE-I4 chip for the B-layer upgrade is designed in a 130 nm CMOS process. For this design, configuration memories are based on the DICE latches where layout considerations are followed to improve the tolerance to SEU. Tests have shown that DICE latches for which layout approaches are adopted are 30 times more tolerant to SEU than the standard DICE latches. To prepare for the new pixel readout chip planned for the future upgrades, a prototype chip containing 512 pixels has been designed in a 65 nm CMOS process and a new approach is adopted for SEU tolerant latches. Results in terms of SEU and TID tolerance are presented.

Results of 65 nm pixel readout chip demonstrator array

Complex and challenging instrumentation projects (LHC upgrades, HL LHC, new Detector concepts) will require the adoption of ever more empowering and complex IC technologies. We clearly see that by moving only two nodes from the current 130 nm CMOS to 65 nm, substantial processing power could be embedded in the front-end system. We have designed and fabricated a prototype pixel readout chip to explore 65 nm CMOS as a potential integrated circuit technology for future particle physics applications. Not only the reported functional test results are very encouraging, but we also found that the process is fundamentally tolerant to radiation doses higher than 600 Mrad.

Operating the GridPix detector in dark matter search experiments

The DARWIN (dark matter WIMP search with noble liquids) design study aims to use liquid argon and liquid xenon targets to look for nuclear recoils due to weakly interacting massive particles (WIMPs). To measure the recoil energy in dual-phase noble gas time projection chambers the combination of scintillation and ionisation detection is used to discriminate nuclear from electron recoils. Current experiments use an array of photomultiplier tubes to detect the primary scintillation and the ionisation electrons via secondary scintillation in the gas phase. Within the research framework for DARWIN, one candidate for an alternative direct charge readout is GridPix, a micro-pattern gaseous detector composed of a Micromegas-like amplification grid over the Timepix 65k pixel readout chip. It can achieve a single-electron detection efficiency of up to 98% and has thus great potential to identify the ionisation electrons in dark matter search experiments. The main challenges for this application are low outgassing, thermal robustness, and operation in pure (thus quencher-free) noble gas. To investigate its applicability we operated a GridPix detector in an argon cryostat. We proved GridPix's performance in pure argon. The gas amplification was confirmed in pure argon in the broad temperature range from 300K down to 87K. Additionally, we discuss results of thermal tests of GridPix devices at liquid xenon temperature in a dry nitrogen atmosphere.

Test-beam studies of diamond sensors for SLHC

Diamond sensors are studied as an alternative to silicon sensors to withstand the high radiation doses that are expected in future upgrades of the pixel detectors for the SLHC. Diamond pixel sensors are intrinsically radiation hard and are considered as a possible solution for the innermost tracker layers close to the interaction point where current silicon sensors cannot cope with the harsh radiation environment.An effort to study possible candidates for the upgrades is undergoing using the Fermilab test-beam facility (FTBF), where diamonds and 3D silicon sensors have been studied. Using a CMS pixel-based telescope built and installed at the FTBF, we are studying charge collection efficiencies for un-irradiated and irradiated devices bump-bonded to the CMS PSI46 pixel readout chip. A description of the test-beam effort and preliminary results on diamond sensors will be presented.

Planar pixel sensors for the ATLAS upgrade: beam tests results

The performance of planar silicon pixel sensors, in development for the ATLAS Insertable B-Layer and HighLuminosity LHC (HL-LHC) upgrades, has been examined in a series of beam tests at the CERN SPS facilities since 2009. Salient results are reported on the key parameters, including the spatialresolution, the charge collection and the charge sharing between adjacent cells, for different bulk materials and sensor geometries.Measurements are presented for n+-in-n pixel sensors irradiated with a range of fluences andfor p-type silicon sensors with various layouts from different vendors. All tested sensorswere connected via bump-bonding to the ATLAS Pixel read-out chip. The tests reveal that both n-type and p-type planar sensors are able to collect significant chargeeven after the lifetime fluence expected at the HL-LHC.

Signal and noise of diamond pixel detectors at high radiation fluences

CVD diamond is an attractive material option for LHC vertex detectors mainly because of its strong radiation-hardness causal to its large band gap and strong lattice. In particular, pixel detectors operating close to the interaction point profit from tiny leakage currents and small pixel capacitances of diamond resulting in low noise figures when compared to silicon. On the other hand, the charge signal from traversing high energy particles is smaller in diamond than in silicon by a factor of about 2.2. Therefore, a quantitative determination of the signal-to-noise ratio (S/N) of diamond in comparison with silicon at fluences in excess of 1015 neq cm−2, which are expected for the LHC upgrade, is important. Based on measurements of irradiated diamond sensors and the FE-I4 pixel readout chip design and performance,we determine the signal and the noise of diamond pixel detectors irradiated with high particle fluences. To characterize the effect of the radiation damage on the materials and the signal decrease, the change of the mean free path λe/h of the charge carriers is determined as a function of irradiation fluence.We make use of the FE-I4 pixel chip developed for ATLAS upgrades to realistically estimate the expected noise figures: the expected leakage current at a given fluence is taken from calibrated calculations and the pixel capacitance is measured using a purposely developed chip (PixCap). We compare the resulting S/N figures with those for planar silicon pixel detectors usingpublished charge loss measurements and the same extrapolation methods as for diamond. It is shown that the expected S/N of a diamond pixel detector with pixel pitches typical for LHC, exceeds that of planar silicon pixels at fluences beyond 1015 particles cm−2, the exact value only depending on the maximum operation voltage assumed for irradiated silicon pixel detectors.

Simulation and laboratory test results of 3D CMS pixel detectors for HL-LHC

The CMS pixel detector is the innermost tracking device at the LHC, reconstructing interaction vertices and charged particle trajectories. The current planar sensors located in the innermost layer of the pixel detector will be exposed to very high fluences which will degrade their performances. As a possible replacement for planar pixel sensors in the High Luminosity-LHC (HL-LHC), 3D silicon technology is under consideration due to its expected good performance in harsh radiation environments. Studies are also in progress for using 3D silicon pixel detectors in near-beam proton spectrometers at the LHC. Deep Reactive Ion Etching (DRIE) plays a key role in fabricating 3D silicon detectors in which readout and ohmic electrodes are processed through the silicon substrate instead of being implanted on the silicon surface. 3D pixel devices considered in this study were processed at FBK (Trento, Italy), bump bonded to the CMS pixel readout chip, and characterized in the laboratory. Numerical simulations were also carried out. We report on selected results from laboratory measurements and TCAD simulations.

3D double sided detector fabrication at IMB-CNM

The Large Hadron Collider (LHC) recorded its first collisions during the last months of 2009. By 2020 a two-stage upgrade of the accelerator complex, the High Luminosity LHC (HL-LHC), will increase the instantaneous luminosities up to a factor of ten compared to the current design. The particle fluxes at ATLAS will increment substantially with special impact on the inner tracking detector which will be subjected to large occupancies and radiation damage. In order to cope with the higher instantaneous luminosities ATLAS will upgrade its current Inner Detector (ID) in two phases, first by introducing a new pixel layer (IBL) mounted directly on the beam pipe, and later by completely replacing the current ID with several layers of semiconductor detectors (pixels and strips). The upgrades to the ATLAS ID require the development of new silicon technologies, since the current planar pixel sensors are not suitable for the expected radiation doses at small radii. For these inner detector layers, the most promising technology is the so-called 3D sensor, while improved planar sensors are considered for the external layers. Silicon detectors with cylindrical electrodes offer advantages over standard planar sensors mainly because they are more radiation hard. 3D detectors with the double sided geometry have been fabricated at IMB-CNM clean room facilities. The layouts fits the new pixelated readout chip FE-I4 developed by the ATLAS collaboration.

Architectural modeling of pixel readout chips Velopix and Timepix3

We examine two digital architectures for front end pixel readout chips, Velopix and Timepix3. These readout chips are developed for tracking detectors in future high energy physics experiments. They must incorporate local intelligence in pixels for time-over-threshold measurement and sparse readout. In addition, Velopix must be immune to single-event upsets in its digital logic.The most important requirements for both chips are pixel size, timing resolution, low power and high-speed sparse readout. We describe the transaction level architectural models of the chips using SystemVerilog. The correctness of the models is ensured using Open Verification Methodology. We will also discuss the advantages gained from transaction level modeling.

Recent Developments of HEP Pixel Detector Readout Chips

This article reviews the development of readout integrated circuits for hybrid pixel particle physics detectors. The 250-nm feature size chips in the presently operating ATLAS and CMS experiments are compared with the current state of the art in 130-nm feature size represented by the FE-I4 chip that will be used to add a new beam pipe layer for the ATLAS experiment in 2013 and the upgrade options of the CMS pixel readout chip. This includes a discussion of the array and pixel size, analog performance, readout architecture, power consumption, power distribution options and radiation hardness. Finally, recent work in 65-nm feature size as a means to continue the evolution of readout chip technology towards smaller feature size, higher rate, and lower power is presented.

Measurement of secondary radiation during ion beam therapy with the pixel detector Timepix

In ion beam therapy the finite range of the ion beams in tissue and the presence of the Bragg-peak are exploited. Unpredictable changes in the patient`s condition can alter the range of the ion beam in the body. Therefore it is desired to verify the actual ion range during the treatment, preferably in a non-invasive way. Positron emission tomography (PET) has been used successfully to monitor the applied dose distributions. This method however suffers from limited applicability and low detection efficiency. In order to increase the detection efficiency and to decrease the uncertainties, in this study we investigate the possibility to measure secondary charged particles emerging from the patient during irradiation.An initial experimental study to register the particle radiation coming out of a patient phantom during the therapy was performed at the Heidelberg Ion Beam Therapy Center (HIT) in Germany. A static narrowly-focused beam of carbon ions was directed into a head phantom. The emerging secondary radiation was measured with the position-sensitive Timepix detector outside of the phantom. The detector, developed by the Medipix Collaboration, consists of a silicon sensor bump bonded to a pixelated readout chip (256 × 256 pixels with 55 μm pitch). Together with the USB-based readout interface, Timepix can operate as an active nuclear emulsion registering single particles online with 2D-track visualization.In this contribution we measured the signal behind the head phantom and investigated its dependence on the beam energy (corresponding to beam range in water 2–30 cm). Furthermore, the response was measured at four angles between 0 and 90 degrees. At all investigated energies some signal was registered. Its pattern corresponds to ions. Differences in the total amount of signal for different beam energies were observed. The time-structure of the signal is correlated with that of the incoming beam, showing that we register products of prompt processes. Such measurements are less likely to be influenced by biological washout processes than the signal registered by the PET technique, coming from decays of beam-induced radioactive nuclei.This work demonstrates that the Timepix detector is able to register ions emerging from the patient during the treatment by carbon ion beams. In future work it will be investigated which information about the incoming beam can be gained from the analysis of the measured data.

Application of the Medipix2 technology to space radiation dosimetry and hadron therapy beam monitoring

Abstract The Medipix2 Collaboration, based at CERN, has developed the TimePix version of the Medipix pixel readout chip, which has the ability to provide either an ADC or TDC capability separately in each of its 256 × 256 pixels. When coupled to a Si detector layer, the device is an excellent candidate for application as an active dosimeter for use in Space Radiation Environments. In order to facilitate such a development, data have been taken with Heavy Ions at the HIMAC facility in Chiba, Japan. In particular, the problem of determining the resolution of such a detector system with respect to heavy ions of differing charges and energies, but with similar dE/dx values has been explored for several ions. The ultimate problem is to parse the information in the pixel “footprint” images from the drift of the charge-cloud produced in the detector layer. In addition, with the use of convertor materials, the detector can be used as a neutron detector, and it has been used both as a charged particle and neutron detector to evaluate the detailed properties of the radiation fields produced by hadron therapy beams. The first space flight of a Medipix-based detector is currently planned for the first quarter of 2012 onboard the UK TechDemoSat-1 satellite as part of the LUCID project. The instrument to be flown is currently planned to have 5 TimePix versions of the Medipix2 detector deployed on 5 of the 6 faces of a cube with ∼3 cm on each side and enclosed by a 0.7 mm thick Al cylindrical cover. The planned orbit will have an altitude of 660 km and will be sun-synchronous with an inclination of 98°Current plans are for the instrument to be located on a vertical side of the satellite with one face upward, one downward, two opposed laterally and one facing in the forward direction with respect to the satellite’s velocity vector. As such, the instrument should be exposed to the outer belt electrons during polar passes as well as the South Atlantic Anomaly and ambient Galactic Cosmic Rays in addition to albedos from the atmosphere throughout its orbit. New versions of the basic chip design are ongoing.

Simulated performance of high-Z detectors with Medipix3 readout

The Medipix3 hybrid pixel readout chip incorporates “charge summing” circuitry, which sums together signals detected in adjacent pixels to reconstruct individual hits. This should compensate for fluorescence and charge sharing effects which degrade the performance of high-Z semiconductor detectors such as germanium, gallium arsenide and cadmium telluride. The behaviour of different high-Z sensors read out by Medipix3 is simulated, to compare their performance and to test the effectiveness of a proposed new inter-pixel communication scheme.The effects of charge sharing are eliminated in all these sensors. In Ge and GaAs sensors with 55 μm pixels, fluorescence effects only cause a small reduction in detector performance, immediately above the k-edge energy. The improvement in spectral performance makes it easier to deal with charge trapping effects in GaAs; with an electron lifetime of 10 ns, the main result of charge trapping is a tolerable reduction in energy resolution. In CdTe, fluorescence effects strongly degrade the spatial resolution above the k-edge energy of 26.7 keV, but with appropriate threshold settings the resolution can be recovered at energies above 35 keV. When CdTe is used in colour mode with 110 μm pixels, charge summing greatly reduces the escape and fluorescence peaks in the spectral response. Finally, since the charge summing circuitry cannot correct parallax effects, parallax will become the main source of blurring in the detector in many X-ray scattering experiments.

Simulations of charge summing and threshold dispersion effects in Medipix3

A novel feature of the Medipix3 photon-counting pixel readout chip is inter-pixel communication. By summing together the signals from neighbouring pixels at a series of “summing nodes”, and assigning each hit to the node with the highest signal, the chip can compensate for charge-sharing effects. However, previous experimental tests have demonstrated that the node-to-node variation in the detector's response is very large. Using computer simulations, it is shown that this variation is due to threshold dispersion, which results in many hits being assigned to whichever summing node in the vicinity has the lowest threshold level. A reduction in threshold variation would attenuate but not solve this issue. A new charge summing and hit assignment process is proposed, where the signals in individual pixels are used to determine the hit location, and then signals from neighbouring pixels are summed to determine whether the total photon energy is above threshold. In simulation, this new mode accurately assigns each hit to the pixel with the highest pulse height without any losses or double counting.

CMS pixel detector with new digital readout architecture

The plan for the CMS pixel detector is to be upgraded in 2016 to a new one with a significantly reduced material budget. The new pixel system with more layers (4 for the barrel pixel) has to operate through the existing services at double the luminosity. Therefore a new readout scheme is implemented with a new pixel readout chip (ROC). A description of the ASIC modifications of the digital readout interface of the ROC is presented as well as the results and the performance of the physics based electronic simulations of a complete pixel module consisting of 16 ROCs and a token bit manager chip.

Characterization of the Medipix3 pixel readout chip

The Medipix3 chip is a hybrid pixel detector readout chip working in Single Photon Counting Mode. It has been developed with a new front-end architecture aimed at eliminating the spectral distortion produced by charge diffusion in highly segmented semiconductor detectors. In the new architecture charge deposited in overlapping clusters of four pixels is summed event-by-event and the incoming quantum is assigned as a single hit to the summing circuit with the biggest charge deposit (this mode of operation is called Charge Summing Mode (CSM)). In Single Pixel Mode (SPM) the charge reconstruction and the communication between neighbouring pixels is disabled. This is the operating mode in traditional detector systems.This paper presents the results of the characterization of the chip with electrical stimuli and radioactive sources.

Trigger performance of the ALICE Silicon Pixel Detector

Abstract The ALICE Silicon Pixel Detector (SPD) forms the two innermost layers of the ALICE Inner Tracking System (ITS). It consists of two-layer barrel of hybrid silicon pixel detectors. The SPD contains 1200 pixel readout chips and about 107 pixel cells. The unique feature of the SPD among the vertex detectors of LHC experiments is the prompt trigger capability, called Fast-Or, which allows it to contribute to the experimental first level trigger decision. The SPD data stream includes 1200 digital signals (Fast-Or) promptly asserted by each readout chip on the presence of at least one pixel hit in the matrix. The off-detector Pixel Trigger system elaborates in real time the 1200 Fast-Or signals continuously received from the detector in order to generate a pre-processed signal for the ALICE first level trigger. All the Fast-Or signals are written in the data and available for offline analysis. In this contribution it is presented the methodology used to evaluate the performance of the Fast-Or. The methods are entirely based on offline data analysis. Results are given for a sample of data of proton–proton collisions at s = 7 TeV .

Status of the Timepix MCP-HPD development

This paper describes the design of a high-speed, single-photon sensitive, Hybrid Photon Detector (HPD). The detector consists of a vacuum tube, containing a Micro Channel Plate (MCP) and 4 CMOS pixel readout chips, sealed with a transparent optical input window with a photocathode. The design described here utilizes currently available technologies, specifically the Timepix readout chips and the Photonis Planacon MCP-PMT vacuum tubes. The aim of the project is to demonstrate the feasibility of a 4-side buttable square MCP-HPD with high packing fraction sensitive area. Presented here is the mechanical and thermal design of the prototype detector.

Reliability and performance studies of DC-DC conversion powering scheme for the CMS pixel tracker at SLHC

The upgrades of the Large Hadron Collider (LHC) introduce a significant challenge to the power distribution of the detectors. DC-DC conversion is the preferred powering scheme proposed to be integrated for the CMS tracker to deliver high input voltage levels and performing a step-down conversion nearby the detector modules. In this work, we propose a step-up/step-down powering scheme by performing voltage step up at the CAEN supply unit and voltage step down near the detector. We designed step-up converters and investigate the pixel performance and power loss on the FPIX power distribution system. Tests are performed using the PSI46 pixel readout chips on a forward pixel panel module and the DC-DC converters developed at CERN and Fermilab. Reliability studies include the voltage drop measurements on the readout chips and the power supply noise generated from the converter. Performance studies include pixel noise and threshold dispersion results. Comparison between step-down only and step-up/step-down conversion powering schemes are provided.

Application of the Medipix2 technology to space radiation dosimetry and hadron therapy beam monitoring

The Medipix2 Collaboration, based at CERN, has developed the TimePix version of the Medipix pixel readout chip, which has the ability to provide either an ADC or TDC capability separately in each of its 256×256 pixels. When coupled to a Si detector layer, the device is an excellent candidate for application as an active dosimeter for use in space radiation environments. In order to facilitate such a development, data have been taken with heavy ions at the HIMAC facility in Chiba, Japan. In particular, the problem of determining the resolution of such a detector system with respect to heavy ions of differing charges and energies, but with similar d E/d x values has been explored for several ions. The ultimate problem is to parse the information in the pixel “footprint” images from the drift of the charge cloud produced in the detector layer. In addition, with the use of convertor materials, the detector can be used as a neutron detector, and it has been used both as a charged particle and neutron detector to evaluate the detailed properties of the radiation fields produced by hadron therapy beams. New versions of the basic chip design are ongoing.