Download presentation
Presentation is loading. Please wait.
Published bySydney Lyons Modified over 8 years ago
1
Changes of the particle detection properties of irradiated silicon microstrip sensors after room and elevated temperature annealing G. Casse, A. Affolder, P. P. Allport, V. Chmill, D. Forshaw, A. Greenall, I. Tsurin, T. Huse, 1 G. Casse,8th IWORID, Zurich 3-7 July 2011
2
OUTLINE Annealing to be part of the running scenario of experiments Reverse annealing of the V FD to be avoided at the expenses of reducing the reverse current? Accelerated annealing a tool for drawing this scenario Accuracy of the prediction of accelerated annealing when translated to other temperatures Plans for the sensor upgrade of the general purpose LHC experiments G. Casse,8th IWORID, Zurich 3-7 July 2011 2
3
CONTEXT: Accelerated annealing is a valuable tool for establishing the running scenario of the sensors installed in the experiments, undergoing severe hadron irradiation. Besides, the comparison of the properties of different sensors irradiated to equal doses has to take into account the effect of the annealing, being especially important at short annealing times, where sharper changes are expected. In the past, we have used to anneal sensors for 80 minutes at 60 o C. What is the right procedure when the charge collection is taken into account? 3 G. Casse,8th IWORID, Zurich 3-7 July 2011 As usual, precious irradiation work done by CERN/PS (M. Glaser), JSI (V. Cindro) and Karlsruhe (A. Dierlamm) colleagues, many thanks!
4
G. Casse,8th IWORID, Zurich 3-7 July 2011 4 Annealing can be accelerated with “known” factors by mean of rising the temperature, e.g.: 40 o C f = 30 60 o C f = 550 80 o C f = 7400 (relative to room temperature RT = 20 o C).
5
Using fixed annealing time for comparison of the electrical properties 5 G. Casse,8th IWORID, Zurich 3-7 July 2011 Reverse current and full depletion voltage Data from M. Moll Currently, in order to take the sensors to an annealing condition where the measured quantities show little variation in order to make more reliable comparison, the sensors are annealed for 80’ at 60 o C.
6
G. Casse,8th IWORID, Zurich 3-7 July 2011 6 HPK FZ n-in-p, 1E15 n eq cm -2 Neutron irradiations in Ljubljana. Using fixed annealing time for comparison of the electrical properties 80 minutes at 60 o C (30 days at RT).
7
7 Accelerated CCE Annealing 5 and 1.5E16 n cm -2 5x10 15 n eq cm -2 Irradiated with 26MeV protons (Karlsruhe). 1.5x10 16 n eq cm -2 Irradiated with 26MeV protons (Karlsruhe). Accelerated annealing at 40, 60 and 80 o C. Alibava DAQ based on Beetle chip. G. Casse,8th IWORID, Zurich 3-7 July 2011
8
8 Accelerated Annealing of the reverse current, n-in-p sensors, 1E15 and 1.5E16 n cm -2
9
Room Temperature Annealing of the collected charge, HPK FZ n-in-p, 2E15 n cm -2 ( 26MeV p irradiation) We make large use of accelerating annealing: is this a safe and correct approach? 9 G. Casse,8th IWORID, Zurich 3-7 July 2011
10
10 It can be difficult to carry on these measurements over years...... Failure of the sensor irradiated to 2E15 n eq cm -2, due to moisture formation after the extraction from the measurement freezer.
11
Room Temperature Annealing of the collected charge, HPK FZ n-in-p, 1E16 n cm -2 (26MeV p irradiation) We make large use of accelerating annealing: is this a safe and correct approach? 11 G. Casse,8th IWORID, Zurich 3-7 July 2011
12
12 Comparison of Room Temperature and Accelerated Annealing of the collected charge, HPK FZ n-in-p, 1and 1.5E15 n cm -2 (26MeV p irradiation) Accepted acceleration factor Acceleration factor divided by 2.1 Suggestion for comparing results (awaiting for systematic studies): 120 minutes at 660 o C
13
Room Temperature Annealing of the reverse current, n-in-p sensors, 0.2 and 1E16 n cm -2 (26MeV p irradiation) 13 G. Casse,8th IWORID, Zurich 3-7 July 2011 NOISE
14
G. Casse,8th IWORID, Zurich 3-7 July 2011 14 Evidence of a charge multiplication effect: not only the whole charge is recovered, but increased by f = 1.75/2.1 140 and 300 m n-in-p Micron microstrip sensors after 5x10 15 n eq 26MeV p 14 This effect is probably crucial in annealing.
15
High fluences, high voltages: Most probable charge drops due to short term annealing: N eff drops smaller peak electric field less multiplication Most probable charge rises due to long term annealing: N eff rises larger peak electric field more multiplication Breakdown voltage is lower at 5·10 14 and 1·10 15 than at 2·10 14 and 5·10 15 for detectors irradiated to 5·10 14 and 1·10 15 breakdown voltage decreases with reverse annealing ______________________________________________________________________________________________________ I. Mandić, 17th RD50 Workshop, CERN, 17 – 18 November 2010 6 Annealing of collected charge I. Mandić, 17th RD50 Workshop, CERN, 17 – 18 November 2010 15 G. Casse,8th IWORID, Zurich 3-7 July 2011
16
Leakage current Increase of leakage current with annealing multiplication ______________________________________________________________________________________________________ I. Mandić, 17th RD50 Workshop, CERN, 17 – 19 November 2010 7 guard rings not bonded 16 G. Casse,8th IWORID, Zurich 3-7 July 2011
17
Plans in ATLAS and CMS G. Casse,8th IWORID, Zurich 3-7 July 2011 17
18
ATLAS Thick Strip Sensors Collaboration of ATLAS with Hamamatsu Photonics (HPK) to develop 9.75x9.75 cm 2 n + -strip in p-type substrate devices (6 inch wafers) for strip regions – 4 segments (2 axial, 2 stereo), 1280 strip each, 74.5 mm pitch, ~320 mm thick – FZ1 and FZ2 material studied – Miniature sensors (1x1 cm 2 ) for irradiation studies 18 Axial Stereo See N. Unno, et. al., Nucl. Inst. Meth. A, Vol. 636 (2011) S24-S30 for details See J. Bohm, et. al., Nucl. Inst. Meth. A, Vol. 636 (2011) S104-S110 for details Full-size sensor prototypes fully characterized that meet final specs –Inter-strip capacitance & resistance, coupling capacitance, depletion voltage, leakage currents and polysilicon resistors qualified
19
ATLAS Strip Sensor Charge Collection Miniature devices irradiated to strip barrel fluences with neutrons, pions, protons Charge collection measured with 90Sr b-source S/N greater than 10:1 for strip sensor types with expected noise performance – ~600-800 e- short strips, ~800-1000 e- long strips – See H. Sadrozinski, et.al., Nucl. Inst. Meth. A, doi:10.1016/j.nima.2011.04.0646 for details NIEL scaling good after annealing corrections Charge sufficient for pixel layers as long as adequate bias voltage and cooling provided!!! 19 Neutrons Unannealed 10:1 Signal-to-Noise 240 MeV Pions Annealed 80 min at 60 C Unannealed 900V
20
E2v Strip Sensors 20 Prototyping with e2v to make n + -strip in p-type substrate devices in ATLAS strip geometry –First submission of miniature sensors (1x1 cm 2 ) only to determine optimal isolation/production parameters So far, no surprises with performance before/after irradiation Currently finalizing masks for axial-only full-size sensor in ATLAS strip geometry – Designed to match with both 128 channel ABCn 250 nm ASIC and future 256 channel ABCn 130 nm ASIC
21
G. Casse,8th IWORID, Zurich 3-7 July 2011 21 Pixels for Upgrade R&D directions: Radiation tolerance Active area optimisation (reduced edges) Pursued by various geometry sensor design and simulations Rad-hard sensors: n-in-p or n-in-n, possible 3D. Active area optimisation: GR simulations, n-in-n GRs on backside for asymmetric design and active area close to edge, trenches, edge passivation, active edges. Baseline now, 450µm distance of active area to cut line. Thin devices. Various production ongoing with different geometries encompassing the above ideas (CiS, HPK, Micron, MPI planar sensors, CNM, FBK, Sintef 3D sensors). Simulation tools capable of reproducing novel effects in silicon (charge multiplication) can significantly help advance solutions.
22
Alexander Dierlamm, KIT Strategy Most of the volume of a future CMS Tracker will be equipped with planar silicon sensors (upgrade Phase 1 ≥ 2020) Survey of available silicon materials to probe their individual limits Investigation of properties of several layout options for strip, strixel and pixel sensors A well defined measurement plan has been worked out and participating institutes have been inter-calibrated to guarantee comparable measurements For the test wafers one producer was chosen, that can provide the large quantity and high quality we need “HPK campaign” Measurements are complemented by device simulations In parallel, there are R&D projects as additional options for the most inner layer(s) 3D silicon sensors (production with Sintef, FBK, CNM) Diamond sensors Irradiation plans involve p, n and mixed p,n irradiations to cover the fluences at various radii based on simulations for current CMS geometry at 3000fb -1
23
Alexander Dierlamm, KIT HPK campaign – Materials Initially ordered production of 126 wafers delivered completely “Deep diffused” wafers are about 20% cheaper than 200µm thin wafers! We will investigate this option since advantage of lower bias voltage is still there and benefits in total material budget are minor using physically 200µm thin wafers in place of 320µm Additional material ordered lately and expected Sept./Nov. n-type p-type (p-stop) p-type (p-spray) FZ3206 / 6 FZ200 deep diff.6 / 6 FZ120 deep diff.6 / 6 MCz200 physical6 / 6 Epi1002 / 66 / 6 Epi704 / 0-- Epi506 / 6 FZ200 deep diff. & 2.metal6 / 6 FZ200 physical0 / 6 0 / 4 FZ120 on carrier0 / 6 0 / 4
24
Alexander Dierlamm, KIT HPK wafer layout
25
Alexander Dierlamm, KIT Structures and goals Diodes – Material characterization and add. annealing studies – Measure IV, CV, CCE, TCT, DLTS, TSC Mini sensor I (Baby_std) – Material characterization, charge collection – Measure IV, CV, strip para., CCE, e-TCT Mini sensor II (Add_Baby) – Material characterization using different radiation sources – Measure IV, CV, Lorentz angle Sensor with integrated PA (Baby_PA) – Layout study: spare glass PA – Measure strip capacitances, CC, signal coupling Sensor with short strips and edge read-out (Baby_strixel) – Layout study: read-out lines from inner strip to outer edge – Measure strip capacitances, CC, signal coupling Test structure field (TS) – Process qualification – Measure many things incl. SIMS, SRP, SEM, … Pixel – Real size pixel sensors for CMS ROC footprint with some layout variations – Measure IV, efficiency, σ Multi-geometry strip (30mm) sensor (MSSD) – Layout study: strip width and pitch variations – Measure CV, IV, Cint, S/N, σ Multi-geometry pixel (1.25mm/2.5mm) sensor (MPix) – Layout study: pixel length and pitch variations – Measure CV, IV, Cint, R poly /R PT, S/N, σ Baby_std Baby_strixel Pixel Diode See Georg Auzinger poster!
26
T. Rohe et al. CMS-Pixel upgrade Phase I – iworid 2011 page 26 The CMS pixel detector Forward Pixel Barrel Pixel 3 barrel layers & 2 F/B disks easy & quick removal / installation in shut down designed for luminosity < 1x10 34 cm -2 sec -1 design fluence ~0.6x10 15 n/cm 2 (tolerate ~2×more) Measured resolution r = 13 m z = 25 m Pixel Thresholds mean = 2450 e
27
T. Rohe et al. CMS-Pixel upgrade Phase I – iworid 2011 page 27 Objectives of Upgrade Phase I Part of the motivation for the upgrade/replacement planned for 2016/17 is visible in this plot: Reduction of material: CO 2 cooling + shift material out of tracking -region Fill the gap between pixel and tracker inner barrel (TIB) and increase number of pixel tracking points Barrel 3 4 layers, forward 2 × 2 3×2 disks Further Reduce radius of layer1 to 3cm (new beam pipe necessary) ROC modifications: DC-DC converter, digital readout (use existing cables) + operate at L ~ 2×10 34 cm -2 s -1 Do not change the sensor Radiation hardness sufficient Cross section of the CMS pixel barrel
28
T. Rohe et al. CMS-Pixel upgrade Phase I – iworid 2011 page 28 Radiation Hardness of Sensor – Highly irradiated sensors operative up to 1kV – No signal saturation with bias for > 2×10 15 n eq /cm 2 – Sensor with = 2.8×10 15 n eq /cm 2 delivers > 7ke (at 800V) Expect “decent” particle detection efficiency High voltage limited in CMS by connectors, cables and power supplies. – Operational limit of CMS pixel detector will be given by the steady decrease of the spatial resolution caused by the decrease of the Lorentz angle – Improvement high voltage capability (sparking) of single sided n-in-p detectors is currently under investigation
29
G. Casse,8th IWORID, Zurich 3-7 July 2011 29 CONCLUSIONS The annealing of the CC(V) shows discrepancies from the accepted rescaling of the time axis with temperature. The discrepancy is evident for the measurements with sensors irradiated up to 1-2E15 n eq cm -2 (need more dense data as a function of fluence to clarify this point). At higher doses the difference seems smaller (trapping/charge multiplication dominated)? The reverse current has qualitatively a similar reduction (accelerated and RT). A more accurate measurement, making use of a better heat sink (possibly using pad sensor in thermal contact with a big mass) can give a more accurate insight. Controlled annealing (at 20 0 C) is still a very useful tool to reduce power dissipation and recover fraction of S/N in heavily irradiated silicon detectors. The discrepancies between the accelerated and RT annealing need to be studied to base the prediction for the changes of CCE with time according to the real operation and maintenance temperatures in the experiments. An annealing procedure for comparison of the detector properties could be agreed instead of the present 80 minutes at 60 o C?
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.