Ten Years of the CDF Silicon Vertex Detector: Ten Years of the CDF - - PowerPoint PPT Presentation

ten years of the cdf silicon vertex detector ten years of
SMART_READER_LITE
LIVE PREVIEW

Ten Years of the CDF Silicon Vertex Detector: Ten Years of the CDF - - PowerPoint PPT Presentation

Ten Years of the CDF Silicon Vertex Detector: Ten Years of the CDF Silicon Vertex Detector: Performance and Status Performance and Status Miguel N. Mondragon Fermilab On behalf of the CDF Silicon Group May 31, 2011 New Perspectives 2011


slide-1
SLIDE 1

Ten Years of the CDF Silicon Vertex Detector: Ten Years of the CDF Silicon Vertex Detector: Performance and Status Performance and Status

Miguel N. Mondragon

Fermilab On behalf of the CDF Silicon Group

May 31, 2011 New Perspectives 2011 Fermilab

slide-2
SLIDE 2

Overview Overview

1

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Overview of the Silicon Detectors at CDF Radiation Damage Global Performance Conclusions

slide-3
SLIDE 3

Tevatron and CDF Tevatron and CDF

Tevatron

2

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Scheduled end of operations:

September, 2011 Present delivered luminosity: ~11 fb -1 Expected delivered luminosity at the end of operations: ~12 fb -1

CDF

Much of the physics program

depends on Heavy-Flavor tagging (top- , b-quark physics, low-mass Higgs boson search) Silicon detectors essential for vertex positioning of HF decays

slide-4
SLIDE 4

The Silicon Detector The Silicon Detector

Three microstrip subdetectors, “p” strips on “n” bulk

SVX II: the main subdetector (~60% of the channels) ISL: longer in z for forward coverage L00: mounted on the beam pipe

} }

3

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

7-8 concentric layers 7 m2 of silicon 722,432 channels

Beam line

slide-5
SLIDE 5

The L00 Subdetector The L00 Subdetector

1.1 cm

Silicon sensors

cables SVX inner bore Cooling tube channel C fiber support

Mounted directly on the beam pipe Designed to improve track impact parameter resolution Single sided sensors (axial strips). Readout strip pitch 50μm Modules by three manufacturers:

4

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011. SVX innermost layer

Beam pipe

Hamamatsu , 36 “wide” modules at r=1.62 cm SGS Thomson, 10 “narrow” modules at r=1.35 cm Micron, 2 “narrow” modules (oxygenated – more radiation tolerant)

slide-6
SLIDE 6

The SVX II Subdetector The SVX II Subdetector

Double sided sensors (built-in strips on both sides) for φ and z positioning Radial segmentation: 5 layers φ segmentation: 12 wedges Axial segmentation: 6 bulkheads Strip pitch 60-140 µm Radius: 2.5 to 10.7 cm Hamamatsu, Layers 0, 1, 3. Axial (φ)/z strips Micron, Layers 2, 4. Axial/small-angle stereo strips SVX-Layer 0 (72 sensors)

5

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

wedge

slide-7
SLIDE 7

The ISL Subdetector The ISL Subdetector

6

1 m

Located between SVX II and the central drift chamber

(COT) Designed to improve track reconstruction and linking to the central drift chamber (especially at forward coverage) Three layers central (|η| < 1) and r = 22 cm Forward (1 < |η| < 2) and r = 20 cm Forward (1 < |η| < 2) and r = 28 cm Double sided sensors (axial and small-angle stereo strips)

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

slide-8
SLIDE 8

Basic Principle of Silicon Strip Sensors Basic Principle of Silicon Strip Sensors

7

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Bulk (n-doped)

Backplane (n+-doped)

Implant strips (p+-doped)

300 m μ

Insulator (SiO2)

Metal strips Reversed bias voltage Integration, amplification, digitization,... (SVX3D chip) Charged particle

+ + + + – – – –

Electrical signal

Single sided sensor

  • Vdep

Full depletion gives the maximum collection of charges (minimizes recombination)

E-field

slide-9
SLIDE 9

Radiation Damage Radiation Damage

slide-10
SLIDE 10

Tracking Volume Radiation Field Tracking Volume Radiation Field

8

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Locations of 916 Thermal Luminescent

Dosimeters (TLDs) in the CDF tracking volume.

The radiation field has been measured and modeled at CDF using 916 dosimeters (TLDs),

R.J. Tesarek et al., NIM A514, 188 (2003)

slide-11
SLIDE 11

A Model of Radiation Damage A Model of Radiation Damage

In addition to ionization, part of the radiation produces crystal defects

9

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

∆NC(Φ) = NC0(1−e−cΦ)+gCΦ, ∆NY (Φ, t,T ) = gYΦ[1−1/(1+t/τY (T))] ∆Neff (Φ, t,T ) = ∆NC(Φ) + ∆NY (Φ, t,T )

With sufficient radiation the “n” bulk turns into a “p-like” bulk Type inversion

Depletion voltage follows the concentration of “dopants”

Vdep ~ |Neff| d2

Fluence Φeq [1012 cm -2]

Depletion Voltage

M.Moll, PhD Thesis, (1992) Uni Hamburg; papers by RD2, RD50 Collaborations,... M.Moll (1992)

Data from test beam

Some complex crystal defects with vacancies can trap electrons and change the electrical properties by removing donors and creating acceptors

Parametrization:

slide-12
SLIDE 12

Radiation Damage on Operations Radiation Damage on Operations

Observable effects of Radiation Damage

10

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Increased depletion voltage (above a minimum) Lower Signal-to-Noise ratio (smaller collected Signal, larger Noise) Increased reversed-bias current Increased bias voltage required Limitations on voltage: power supply limit, sensor breakdown Signal-to-Noise ratio must be kept good enough for physics analyses

Operational Implications

The CDF Silicon Detector was designed to withstand radiation up to ~ 3 fb -1

  • f integrated luminosity
slide-13
SLIDE 13

Evolution of Depletion Voltage Evolution of Depletion Voltage

11

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Measured Vdep at different integrated luminosities: Fit a 3rd degree polynomial around minimum Linear fit above inversion point

Inversion Point (minimum)

Linear rise

As radiation damage increases: Required depletion voltage rises Need to increase reversed-bias V to keep good efficiency The linear rise can be extrapolated to make predictions...

slide-14
SLIDE 14

Outlook of Depletion Voltage for L00 Outlook of Depletion Voltage for L00

12

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Operational limit at ~500 V A few sensors are reaching the

  • perational limit for a full Vdep and

becoming under-depleted Linear fits and extrapolations for ALL sensor modules of L00:

slide-15
SLIDE 15

Conversion of Luminosity to Radiation Dose for L00 Conversion of Luminosity to Radiation Dose for L00

13

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

The (oxygenated) Micron sensors inverted after the others, as expected

Average Vdep (over all sensors) [V]

CDF Preliminary

Hamamatsu (wide) SGS Thomson (narrow) Micron oxygenated (narrow)

Using the measured radiation field and knowing the position of the silicon sensors, it is possible to convert luminosity to radiation dose

slide-16
SLIDE 16

14

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Depletion Voltage for SVX - L0 Depletion Voltage for SVX - L0

SVX-L0 Operational limit: sensors may break down at Vbias > 170 V Some ladders may become under-depleted soon

Average Vdep (over all sensors) [V]

CDF Preliminary

slide-17
SLIDE 17

Global Performance Global Performance

slide-18
SLIDE 18

Signal-to-Noise ratios for L00 and SVX Signal-to-Noise ratios for L00 and SVX

15

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

b-tagging requirement (S/N > 4)

Signal measured from data events J/Ψ to muons Noise estimated from calibration runs with beam S/N will be still good for b-tagging SVX-L0 degrades the fastest

11 12

End of collisions Now

b-tagging

11 12

φ side z side

slide-19
SLIDE 19

How much does a dead SVX-L0 impact b-tagging? How much does a dead SVX-L0 impact b-tagging?

16

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Wedge Number Wedge Number

b-tagging efficiency

Plot depicts b-tagging efficiencies estimated with data, for jets whose axis passes

through SVX -L0 Blue circles indicate wedge where sensor in layer L0 is not taking data Green circles indicate whole wedge (all layers) is not taking data

The dead of a layer L0 sensor would not degrade the efficiency so much!

Bulkhead 0 Bulkhead 1 Bulkhead 2 Bulkhead 3 Bulkhead 4 Bulkhead 5 Average (excluding encircled wedges)

slide-20
SLIDE 20

Active Modules 2002-2011 Active Modules 2002-2011

Powered in black, good in green, bad in red and error rate in pink

17

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

The silicon system is getting good data from ~86% of the sensors and running ~92.5% of the detector.

3 f b-1 7 f b-1

slide-21
SLIDE 21

Conclusions Conclusions

18

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

The CDF Silicon Detectors have outlasted their expected lifespan... and are still in good shape! Some sensors may become under-depleted soon in L00 and SVX-L0, but will continue in operation S/N ratio will still be good for b-tagging until the end

  • f online operations

Tomorrow in the Poster Session: Kyle Knoepfel, “Aging Studies of the CDF Run II Silicon Detector” Tim Harrington-Taber, “Operational Experience of the CDF Run II Silicon Detector” Aging studies Operational experience

slide-22
SLIDE 22

Back-up Slides Back-up Slides

slide-23
SLIDE 23

Radiation Damage on Silicon Crystals Radiation Damage on Silicon Crystals

Radiation produces crystal defects

i

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Vacancies Interstitials (displaced atoms) Complexes: divacancies, P-vacancy,... E> 12 KeV, clusters of defects (~100 lattice displacements) Creation of multiple energy levels between Valence and Conduction Bands Trapping of signal charges Donor removal and acceptor creation (mainly) Density of space charge is changed

Defects change electrical properties:

sensor bulk “n”

slide-24
SLIDE 24

Type Inversion Type Inversion

Before inversion

ii

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

After inversion “n” bulk turns into “p-like” type

An under-depleted single sided sensor looks:

slide-25
SLIDE 25

Extraction of Depletion Voltage of a Sensor Extraction of Depletion Voltage of a Sensor

iii

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Depletion V

Sigmoid fit

100 V

Landau fit (gaussian convoluted) of charge distribution Extract the Most Probable Value (MPV) of the charge Collect data at several Vbias settings. For each setting: Plot Charge vs Vbias for each settings Sigmoid fit Define Vdep as 95% of plateau

Efficiency

Efficiency follows the trend of the Vbias

Charge Landau fit for each point

slide-26
SLIDE 26

Evolution of Depletion Voltage Evolution of Depletion Voltage

iv

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Extract the Vdep from measurements at different integrated luminosities: Fit a 3rd degree polynomial around minimum Fit a line beyond inversion point

Inversion point

Oxygenated Micron sensors invert after the others

(SGS Thomson)

Average luminosity (over all sensors

  • f each kind) at inversion point

Linear rise

slide-27
SLIDE 27

DAQ System DAQ System

v

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

slide-28
SLIDE 28

Resolution Resolution

vi

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

The impact parameter d0 is defined as the shortest distance in the r-phi plane between the beamline and the trajectory of the particle obtained by the tracking algorithm fit.

slide-29
SLIDE 29

A Model of Radiation Damage A Model of Radiation Damage

Effective doping concentration

  • f “n”-type silicon

vii

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

∆NC(Φ) = NC0(1−e−cΦ)+gCΦ, ∆NY (Φ, t,T ) = gYΦ[1−1/(1+t/τY (T))] ∆Neff (Φ, t,T ) = ∆NC(Φ) + ∆NY (Φ, t,T )

Increases with Φ (time independent) “Reverse Annealing”, (increases with Φ, time and Temperature)

Vdep ~ |Neff| d2 Changes of depletion voltage with total fluence Φ (incident particles/area) There is a type inversion Increases after type inversion

Fluence Φeq [1012 cm -2]

Depletion Voltage

M.Moll, PhD Thesis, (1992) Uni Hamburg; papers by RD2, RD50 Collaborations,...

Two mechanisms: Donor Removal and Acceptor Creation

M.Moll (1992)

Data from test beam

slide-30
SLIDE 30

TLD packages

viii

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.

Package containing six TLDs

slide-31
SLIDE 31

Slide

ix

“The CDF Silicon Detector: Performance and Status” – MN Mondragon – New Perspectives 2011.