پروژه "تحقیق و توسعه" حسگر پیکسلی دو وجهی ATLAS

Within the R&D project on Planar Pixel Sensor Technology for the ATLAS inner detector upgrade, the use of planar pixel sensors for highest fluences as well as large area silicon detectors is investigated. The main research goals are optimizing the signal size after irradiations, reducing the inactive sensor edges, adjusting the readout electronics to the radiation induced decrease of the signal sizes, and reducing the production costs. Planar n-in-p sensors have been irradiated with neutrons and protons up to fluences of View the MathML source2×1016neq/cm2 and View the MathML source1×1016neq/cm2, respectively, to study the collected charge as a function of the irradiation dose received. Furthermore comparisons of irradiated standard View the MathML source300μm and thin View the MathML source140μm sensors will be presented showing an increase of signal sizes after irradiation in thin sensors. Tuning studies of the present ATLAS front end electronics show possibilities to decrease the discriminator threshold of the present FE-I3 read out chips to less than 1500 electrons. In the present pixel detector upgrade scenarios a flat stave design for the innermost layers requires reduced inactive areas at the sensor edges to ensure low geometric inefficiencies. Investigations towards achieving slim edges presented here show possibilities to reduce the width of the inactive area to less than View the MathML source500μm. Furthermore, a brief overview of present simulation activities within the Planar Pixel R&D project is given.

The present pixel detector of the ATLAS experiment constitutes the innermost part of the tracking system. Distributed on three barrel layers at radii between 50.5 and 122.5 mm and six discs a total of 1744 pixel modules are mounted allowing for a three hit track reconstruction of charged secondary particles [1]. Each module contains a View the MathML source250μm thick large scale n-in-n pixel sensor of 62.6×18.6 mm2 with pixel implants of View the MathML source50×400μm2. Connected to each sensor are 16 7.4×11.0 mm2 ATLAS FE-I3 [2] chips with a total of 46 080 readout channels. Both the sensors and the electronics of the present ATLAS pixel modules are specified to work up to a fluence of 1015 neq/cm2 (1 MeV neutrons). While the nominal luminosity of the present LHC accelerator is 1034 cm−2/s, an upgrade to increase the luminosity by a factor of 10 is planned to be carried out in a two phase process [3]. In Phase 1 an upgrade to View the MathML source(2–3)×1034cm−2/s is foreseen without any changes to the machine hardware up to the year 2014. The Phase 2 upgrade to the Super LHC (SLHC) around 2018 is expected to reach a maximum luminosity of 1035 cm−2/s with modifications to the insertion quadrupoles and changes in the main machine parameters. In this scenario the innermost layer of the ATLAS pixel system will have to sustain fluences reaching View the MathML source(3–4)×1015neq/cm2 during the Phase 1 upgrade and View the MathML source(1–2)×1016neq/cm2 during the SLHC running period [4]. The resulting defects in the semiconductor sensors cause high leakage currents and a reduced charge collection distance. Consequently it will be a challenge to develop semiconductor sensors that will yield a sufficient signal to noise ratio.

Within the Planar Pixel Sensor project the suitability of using planar silicon pixel sensors for the ATLAS upgrade towards increased luminosities and large scale pixel systems are investigated. First characterizations of n-in-p strip sensors were carried out before and after proton and neutron irradiations up to fluences expected at the SLHC. The collected signal sizes after View the MathML source(1–2)×1016neq/cm2 at a bias voltage of 1000 V reach around 5000 electrons. This is significantly more than expected from charge trapping models based on lower irradiation fluences. Thin planar sensors irradiated to the same fluences and operated at the same bias show a larger signal compared to sensors of standard thicknesses. Furthermore they cause less multiple scattering which is beneficial for the spatial resolution of the detector. To cope with decreased signal sizes tuning procedures were studied to reduce the discriminator threshold of the front end electronics. On single FE-I3 chip modules a reduction from 4000 to less than 1500 electrons seems possible. The threshold performance of the succeeding FE-I4 readout chip is a key ingredient for using planar pixel sensors for the LHC upgrades.