The upgrades of the ATLAS pixel detector€¦ · Introduction The original inner tracker of ATLAS...

29
The upgrades of the ATLAS pixel detector Stefano Terzo Max-Planck-Institut f ¨ ur Physik unchen Young Scientist Workshop Ringberg Castle 17 th July 2014 Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI f ¨ ur Physik) 1 / 15

Transcript of The upgrades of the ATLAS pixel detector€¦ · Introduction The original inner tracker of ATLAS...

The upgrades of the ATLAS pixel detector

Stefano Terzo

Max-Planck-Institut fur PhysikMunchen

Young Scientist WorkshopRingberg Castle17th July 2014

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 1 / 15

Introduction

The original inner tracker of ATLAS

144 cm

43cm

6 cm6 cm

2 cm

Composed of hybrid planar pixel detectors:I 1744 modulesI 46080 pixel channels per module:

I 50µm× 400µm cell pitchI n-in-n silicon sensors: 250µm thick

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 2 / 15

3 barrel layers2× 3 end cap disks

Introduction

Hybrid silicon pixel detectors

I Lightly n-doped silicon:I n-doped (− charge) pixelsI p-doped (+ charge) backside

I A reverse bias voltage is applied:I silicon depletion from the backside

(pn junction)I ∆Vbias > Vdep → full depletionI depleted region = sensitive volume

I A particle passing through the depletedregion creates electron-hole pairs which driftto the electrodes

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 3 / 15

Introduction

Hybrid silicon pixel detectors

I Lightly n-doped silicon:I n-doped (− charge) pixelsI p-doped (+ charge) backside

I A reverse bias voltage is applied:I silicon depletion from the backside

(pn junction)I ∆Vbias > Vdep → full depletionI depleted region = sensitive volume

I A particle passing through the depletedregion creates electron-hole pairs which driftto the electrodes

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 3 / 15

Introduction

Hybrid silicon pixel detectors

I Lightly n-doped silicon:I n-doped (− charge) pixelsI p-doped (+ charge) backside

I A reverse bias voltage is applied:I silicon depletion from the backside

(pn junction)I ∆Vbias > Vdep → full depletionI depleted region = sensitive volume

I A particle passing through the depletedregion creates electron-hole pairs which driftto the electrodes

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 3 / 15

Introduction

Hybrid silicon pixel detectors

I Lightly n-doped silicon:I n-doped (− charge) pixelsI p-doped (+ charge) backside

I A reverse bias voltage is applied:I silicon depletion from the backside

(pn junction)I ∆Vbias > Vdep → full depletionI depleted region = sensitive volume

I A particle passing through the depletedregion creates electron-hole pairs which driftto the electrodes

+V +V

-V

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 3 / 15

Introduction

Hybrid silicon pixel detectors

I Lightly n-doped silicon:I n-doped (− charge) pixelsI p-doped (+ charge) backside

I A reverse bias voltage is applied:I silicon depletion from the backside

(pn junction)I ∆Vbias > Vdep → full depletionI depleted region = sensitive volume

I A particle passing through the depletedregion creates electron-hole pairs which driftto the electrodes

+V +V

-V

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 3 / 15

Introduction

Hybrid silicon pixel detectors

I Lightly n-doped silicon:I n-doped (− charge) pixelsI p-doped (+ charge) backside

I A reverse bias voltage is applied:I silicon depletion from the backside

(pn junction)I ∆Vbias > Vdep → full depletionI depleted region = sensitive volume

I A particle passing through the depletedregion creates electron-hole pairs which driftto the electrodes

+V +V

-V

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 3 / 15

Introduction

Hybrid silicon pixel detectors

I Pixel surface segmentation:I precise two dimensional spatial

informationI Guard ring structures to smoothly drop

the potential at the edge of the sensorI Readout chip coupling (bump bonding)

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 4 / 15

Introduction

Hybrid silicon pixel detectors

I Pixel surface segmentation:I precise two dimensional spatial

informationI Guard ring structures to smoothly drop

the potential at the edge of the sensorI Readout chip coupling (bump bonding)

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 4 / 15

Introduction

Hybrid silicon pixel detectors

I Pixel surface segmentation:I precise two dimensional spatial

informationI Guard ring structures to smoothly drop

the potential at the edge of the sensorI Readout chip coupling (bump bonding)

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 4 / 15

Motivations

The inner tracker upgrade plan

Energy: 8 TeV 13/14 TeV

Luminosity: >1034cm−2s−1 ∼3×1034 cm−2s−1

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 5 / 15

5×1034 cm−2s−1

Motivations

The inner tracker upgrade plan

Energy: 8 TeV 13/14 TeV

Luminosity: >1034cm−2s−1 ∼3×1034 cm−2s−1

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 5 / 15

Insertable B-Layer (IBL): new pixellayer at 3.2 cm from the pp interaction

5×1034 cm−2s−1

Motivations

The inner tracker upgrade plan

Energy: 8 TeV 13/14 TeV

Luminosity: >1034cm−2s−1 ∼3×1034 cm−2s−1

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 5 / 15

Insertable B-Layer (IBL): new pixellayer at 3.2 cm from the pp interaction

“Phase II”: full inner detectorreplacement (4 layers)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

0.0

0.5

1.0

1.5

z (m)

r(m

)

eta = 0.0 eta = 1.0

eta = 2.0

eta = 3.0

5×1034 cm−2s−1

Motivations

Why a fourth layer?

I Mantain and improve performances lost due to current detectordegradation:

I radiation damageI module failure

I Improve tracking, vertexing and b-tagging:I proximity to the interaction pointI larger extension along the beam lineI improve track measurement

I Cope with pile-up at high luminosityI The new IBL will be tolerant to higher

occupancy without saturation

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 6 / 15

The Insertable B-Layer (IBL)

The Insertable B-Layer (IBL)

14 support structures (staves) to hold 32 new front-end chips each

I Mixed sensor technology:I 200 µm thick planar n-in-n silicon sensorsI 230 µm thick 3D n-in-p silicon sensors

I Full φ coverage by tilting staves by 14◦

I Overlap of the inactive regions

I Coverage in |η| <2.8I No overlap possible due to limited space

I Recover z coverage with slim edge sensors

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 7 / 15

The Insertable B-Layer (IBL)

The Insertable B-Layer (IBL)

14 support structures (staves) to hold 32 new front-end chips each

I Mixed sensor technology:I 200 µm thick planar n-in-n silicon sensorsI 230 µm thick 3D n-in-p silicon sensors

I Full φ coverage by tilting staves by 14◦

I Overlap of the inactive regions

I Coverage in |η| <2.8I No overlap possible due to limited space

I Recover z coverage with slim edge sensors

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 7 / 15

The Insertable B-Layer (IBL)

The Insertable B-Layer (IBL)

14 support structures (staves) to hold 32 new front-end chips each

I Mixed sensor technology:I 200 µm thick planar n-in-n silicon sensorsI 230 µm thick 3D n-in-p silicon sensors

I Full φ coverage by tilting staves by 14◦

I Overlap of the inactive regions

I Coverage in |η| <2.8I No overlap possible due to limited space

I Recover z coverage with slim edge sensors

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 7 / 15

The Insertable B-Layer (IBL)

Slim edge planar pixel sensors for IBL

I Initial designI 1.1 mm dead area

I Reduced guard rings and edge cutI 450 µm dead area

I Pixels shifted behind the guard ringsI Dead area is reduced to 200 µm

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 8 / 15

No overlap possiblealong the beam direction

for the inner layer→ new slim edge design

The Insertable B-Layer (IBL)

Slim edge planar pixel sensors for IBL

I Initial designI 1.1 mm dead area

I Reduced guard rings and edge cutI 450 µm dead area

I Pixels shifted behind the guard ringsI Dead area is reduced to 200 µm

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 8 / 15

No overlap possiblealong the beam direction

for the inner layer→ new slim edge design

The Insertable B-Layer (IBL)

Slim edge planar pixel sensors for IBL

I Initial designI 1.1 mm dead area

I Reduced guard rings and edge cutI 450 µm dead area

I Pixels shifted behind the guard ringsI Dead area is reduced to 200 µm

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 8 / 15

No overlap possiblealong the beam direction

for the inner layer→ new slim edge design

The Insertable B-Layer (IBL)

The new readout chipFE-I3 vs FE-I4

50×400 µm2 Smaller pixel pitch 50×250 µm2

1×1 cm2

160 row×

18 columns

Bigger area

2×2 cm2

336 row×

80 columns

1×1015neq/cm2 Radiation hardness 5×1015neq/cm

2

columndrain

architecturetrigger

higherhit

rate

localin pixelstorage

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 9 / 15

The Phase II upgrade at HL-LHC

From IBL to the inner layers of Phase II

A smaller pixel cell to reduce the occupancy

Now: FE-I3 chip 50×400 µm2

IBL: FE-I4 chip 50×250 µm2

Ph. II: new chip 25×100 µm2

or 50×50 µm2

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 10 / 15

preliminary studies using theFE-I4 chip and 25×500 µm2

pixel shape

Pil

e up

eve

nts 23

230

.

.

The Phase II upgrade at HL-LHC

From “n-in-n” to “n-in-p” sensors

Present pixel detector + IBL → Phase II (HL-LHC)

N-in-nI Depletion from the backsideI Guard rings on the backsideI Double sided process→ Pixel can be shifted below

the guard rings

N-in-pI Depletion from the pixel sideI Guard rings on the pixel sideI Single sided process→ Cost effective to cover

large areas

A new solution is necessary to reduce the dead area

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 11 / 15

The Phase II upgrade at HL-LHC

Planar active edge sensors for HL-LHC

p-substrateSensor

n+

100

µm

50 µm

GR GR

Edge implantationSilicon oxide Boron doping Phosphorus doping

Support wafer

50 µm

I Present IBL design:I n-in-n sensor with 200 µm edge

(pixel shifted under the guard rings)

I Phase II alternatives:I n-in-p sensor with 125 µm active edge

(only one Bias Ring)

I n-in-p sensor with 50 µm active edge(Floating Guard Ring)

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 12 / 15

125 µm

50 µm

200 µm

Backside implantationextended to the edges

The Phase II upgrade at HL-LHC

Edge efficiency

FE-I3, 50 µm active edge FE-I3, 125 µm slim edge

m]µDistance from last pixel edge [-100 0 100 200 300 400 500 600

Effic

ienc

y [%

]

6065707580859095

100

m]µDistance from the last pixel edge [

Eff

icie

ncy

[%]

60

65

70

75

80

85

90

95

100

-1000100200300400500

Active area extends outside the lastpixel implant up to the active edge

Active area only until thebias ring structure

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 13 / 15

→ last pixel implant

← end of the implant

last pixel implant←

end of the implant→

600 -1250-50 6000

The Phase II upgrade at HL-LHC

N-in-p quad modules for the outer layers

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 14 / 15

3 2

0 1

Quad module:4× readout chips connected to a bigger sensors

I Planned full module geometry for phase III No guard ring structures between the 4 chipsI Fully active area between the 4 chips

The Phase II upgrade at HL-LHC

N-in-p quad modules for the outer layers

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 14 / 15

3 2

0 1

Quad module:4× readout chips connected to a bigger sensors

I Planned full module geometry for phase III No guard ring structures between the 4 chipsI Fully active area between the 4 chips

First tests performed on pseudoquad module interconnected to

IBL FE-I4 chips

Summary and outlook

Summary and outlook

I IBL, the 4th pixel layer in ATLAS has been successfully inserted:I closer to the interaction pointI featuring the new FE-I4 chipI improved radiation hardness

I New module designs are necessary to cope with the HL-LHCenvironment:

I reduced occupancyI radiation hardnessI increased active area (for the innermost layers)

I N-in-p active/slim edge sensor are under investigation for theHL-LHC upgrade:

I 50 µm active edge sensors show efficiency even outside the pixel area

I First pseudo quad modules have been assembled and characterized

Young Scientist Workshop 16th - 19th July 2013 Stefano Terzo (MPI fur Physik) 15 / 15