Towards a 10μs, thin high resolution pixelated CMOS sensor for future vertex detectors

Size: px

Start display at page:

Download "Towards a 10μs, thin high resolution pixelated CMOS sensor for future vertex detectors"

Justina Greene
5 years ago
Views:

1 Towards a 10μs, thin high resolution pixelated CMOS sensor for future vertex detectors Yorgos Voutsinas IPHC Strasbourg on behalf of IPHC IRFU collaboration CMOS sensors principles Physics motivations MIMOSA26 and applications evolutions Integration studies Developing a 10μs readout sensor 1

noise electronics Use of only NMOS transistors for on pixel signal processing Undepleted sensitive volume non ionising radiation tolerance, charge

2 CMOS sensors principles Signal created by mips: ~80e--h pairs/μm Electrons diffuse thermally at epi layer collected by an Nwell-p-epi junction Advantages High granularity O(10μm) Low material budget (<50μm) Signal processing on substrate Cost Limitations Small signal O(1000e-) calls for low noise electronics Use of only NMOS transistors for on pixel signal processing Undepleted sensitive volume non ionising radiation tolerance, charge collection Exploration of high resistivity epitaxial layers partially depleted sensitive volume Faster charge collection More radiation tolerant sensors 2

16 % X0) Power dissipation << 100W Readout time ~25μs inner layers, ~100μs outer layers Radiation tolerance ~0.

3 Motivations CMOS sensors appropriate for high precision tracking devices Vertex detectors Beam telescopes R&D mostly driven by An option for ILC vertex detector s.p. resolution ~ 3μm Extra low material budget ( % X0) Power dissipation << 100W Readout time ~25μs inner layers, ~100μs outer layers Radiation tolerance ~0.3MRad, few 1011neq/cm2 Unprecedented impact parameter resolution Mid-term applications RHIC, FAIR, EUDET beam telescope Considered as an option for ALICE experiment upgrade 3

MIMOSA26 Fast full scale sensor 10kframes/s, 2cm2 active area Binary output & integrated zero suppression Column parallel readout each column ends with a discriminator binary output Pixel pitch 18.

4 MIMOSA26 Fast full scale sensor 10kframes/s, 2cm2 active area Binary output & integrated zero suppression Column parallel readout each column ends with a discriminator binary output Pixel pitch 18.4μm 665k pixels readout time~110μs (80MHz freq.) In pixel preamplfication & CDS Spatial resolution ~ 4.5μm Data sparsification ( reduced data flow) Power dissipation 280mW/cm2 Fabricated in 0.35μm technology MIMOSA26 final sensor for EUDET beam telescope Forerunner for STAR HFT, CBM MVD and ILC applications 4

5 MIMOSA26 test beam results Beam test at CERN SPS, September 2009 with pion beam, E=120GeV 6 MIMOSA26 sensors mounted on TAPI (Strasbourg BT) running at 80MHz Efficiency 99.5 ± 0.1 (stat) ± 0.3(prel)% for below 10-4 fake hit rate Resolution ~ 4.5μm 5

MIMOSA26 applications - evolutions MIMOSA26 equips the reference planes of

at CERN SPS at 2009 Extrapolated resolution ~ 2μm Upto 106particles/cm2/s beam

data expected at 2013 CBM MVD More severe radiation tolerance requirements

6 MIMOSA26 applications - evolutions MIMOSA26 equips the reference planes of EUDET BT EUDET FP6 project - infrastructure for ILC detectors R&D Commissioned at CERN SPS at 2009 Extrapolated resolution ~ 2μm Upto 106particles/cm2/s beam intensity Baseline sensor for STAR HFT 1M pixels 200μs integration time First data expected at 2013 CBM MVD More severe radiation tolerance requirements Double sided ro (20μs int. time) Prototyping 2012 ILC VXD (option) ALICE upgrade (option) 6

Sensors with high resistivity epitaxial layer MIMOSA26 with high resistivity epitaxial layer partially depleted sensitive volume Faster charge collection Shorter path for the charge carriers => more

7 Sensors with high resistivity epitaxial layer MIMOSA26 with high resistivity epitaxial layer partially depleted sensitive volume Faster charge collection Shorter path for the charge carriers => more tolerant to non ionising radiation Preliminary results: Seed S/N MPV vs radiation for different epitaxial layers (room temperature) Twice larger seed S/N Expected improved tolerance to non ionising radiation > 1014neq/cm2 (for T<0 & nominal readout) Radiation (neq/cm2) Std epi layer 14μm High resistivity 400Ωxcm, 15μm High resistivity 400Ωxcm, 10μm x x

Integration studies: PLUME project Pixel Ladder with Ultra-low Material Embedding Bristol - DESY - Oxford - Strasbourg Double sided ladder equipped with 2x6 thinned down to 50 μm MIMOSA-26 (ILC DBD

8 Integration studies: PLUME project Pixel Ladder with Ultra-low Material Embedding Bristol - DESY - Oxford - Strasbourg Double sided ladder equipped with 2x6 thinned down to 50 μm MIMOSA-26 (ILC DBD 2012 target: 0.3 % X0) Explore feasibility, performances and added value of double-sided ladders Allows for improved time resolution (outer layer with longer and fewer pixels) First prototype at reduced scale tested at CERN November 2009 Alignment studies (AIDA EU FP7) 8

Developing a 10μs readout sensor MIMOSA26 can be operated up to a 110MHz

o of the sensor => 40μs Moving to smaller feature size (0.

ogy (MIMOSA27) or less) r.o. time 35μs Double sided ladders with one

9 Developing a 10μs readout sensor MIMOSA26 can be operated up to a 110MHz clock frequency => 80μs int. time Double sided r.o of the sensor => 40μs Moving to smaller feature size (0.18μm technology (MIMOSA27) or less) r.o. time 35μs Double sided ladders with one side equipped with elongated pixels on 1 dimension for timestamping 15μm pitch r.o. time ~ 35-40μs 60μm pitch r.o. time < 10μs Depleted epi layer larger sensing diode spacing 9

Further developments 3D integrated sensors Combine different fabrication processes per layer Faster readout (<2μs) Decrease of inactive surface To be assessed: power dissipation &

.) Ultimate goal: CLIC sensor Integration studies:serwiete project (HP2 EU FP7) Sensors wrapped in thin polymerized film it may match cylindrical surfaces Material budget < 0.

10 Further developments 3D integrated sensors Combine different fabrication processes per layer Faster readout (<2μs) Decrease of inactive surface To be assessed: power dissipation & material budget 3D consortium (Fermilab,CNRS,INFN..) Ultimate goal: CLIC sensor Integration studies:serwiete project (HP2 EU FP7) Sensors wrapped in thin polymerized film it may match cylindrical surfaces Material budget < 0.15% X0 Proof of principle in 2011 Prototype made of 3 MIMOSA26 sensors Others Explore high resistiviy epitaxial layers from different vendors & different processes (including VDSM in collaboration with CERN) Stitching AIDA Large Area Telescope MIMOSA26 sensors will equip the vertex detector of FIRST exp. (see Eleuterio Spiriti's talk)... 10

11 Conclusion and future perspectives MIMOSA 26 first fast real scale sensor with binary output and integrated signal processing Developed for EUDET telescope Serve as a forerunner for other applications (STAR, CBM, ALICE, ILC vertex detectors) Availability of high resistivity epitaxial layers Improved radiation tolerance to non ion. radiation Better CCE Increase of intrinsic speed of the sensor Development of a 10μs readout sensor Integration studies PLUME SERWIETE Future perspectives Smaller feature size 3D integrated sensors 11

12 BACKUP SLIDES 12

13 MIMOSA 25 Exploration of the high resistivity epi-layers: MIMOSA-25 (0.6 µm) Tested in CERN 2009 before and after irradiation Cluster size 2x2 pixels (3x3 for low resistivity epi) S/N 60 at seed 30 after 3x1013neq/cm2 Eff = 99.99% for non-irradiated sensor after 3x1013neq/cm2 Improved tolerance to non ionising radiation (1-2 OoM) 13

Towards a 10 μs, thin high resolution pixelated CMOS sensor system for future vertex detectors

Towards a 10 μs, thin high resolution pixelated CMOS sensor system for future vertex detectors Rita De Masi IPHC-Strasbourg On behalf of the IPHC-IRFU collaboration Physics motivations. Principle of operation