non-silicon solid state detectors harris kagan

Non-Silicon Solid State Detectors

Harris KaganOhio State University

42nd INFN Eloisatron WorkshopSep 29, 2003 - Erice, Italy

Outline of the Talk

• Introduction

• Status of Diamond Research

• Status of SiC Research

• Radiation Monitoring - a new application

• The Future

• Summary

42nd INFN Eloisatron Workshop

Sep 29, 2003 - Erice, Italy

Non-Silicon Solid State Detectors (page 1) Harris Kagan

Ohio State University

Introduction

Motivation: Tracking Devices Close to Interaction Region of Experiments

LHC + SLHC Issues:→ Inner tracking layers must survive!

→ Inner tracking layers must provide high precision tracking to tag b, t, Higgs, . . .

→ Annual replacement of inner layers perhaps?

Material Properties:• Radiation hardness

• Low dielectric constant → low capacitance

• Low leakage current → low readout noise

• Room temperature operation, Fast signal collection time → no cooling

Material Presented Here:• Chemical Vapor Deposition (CVD) Diamond

• Silicon Carbide

Reference → http://rd42.web.cern.ch/RD42→ http://rd50.web.cern.ch/RD50





Introduction

Comparison of Various Materials

Property Diamond 4H-SiC Si

Band Gap [eV] 5.5 3.3 1.12Breakdown field [V/cm] 107 4×106 3×105

Resistivity [Ω-cm] > 1011 1011 2.3×105

Intrinsic Carrier Density [cm−3] < 103 1.5×1010

Electron Mobility [cm2V−1s−1] 1800 800 1350Hole Mobility [cm2V−1s−1] 1200 115 480Saturation Velocity [km/s] 220 200 82

Mass Density [g cm−3] 3.52 3.21 2.33Atomic Charge 6 14/6 14Dielectric Constant 5.7 9.7 11.9Displacement Energy [eV/atom] 43 25 13-20

Energy to create e-h pair [eV] 13 8.4 3.6Radiation Length [cm] 12.2 8.7 9.4Spec. Ionization Loss [MeV/cm] 4.69 4.28 3.21Ave. Signal Created/100 µm [e] 3600 5100 8900Ave. Signal Created/0.1% X0 [e] 4400 4400 8400





Diamond

Characterization of Diamond:

Signal formation

e-h Creation

Charged Particle

Electrodes

Diamond

Vbias

Amplifier

• Q=dtQ0 where d = collection distance = distance e-h pair move apart

• d=(µeτe + µhτh)E

• d=µEτ

with µ = µe + µh

and τ = µeτe+µhτh

µe+µh





Diamond

Diamond Properties:

Signal formation

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

10-4

10-3

10-2

10-1

1 10 102

Signal versus applied electric field

0

5 0

100

150

200

250

0 0 .2 0 .4 0 .6 0 .8 1 1 .20

2000

4000

6000

8000

Col

lect

ion

Dis

tanc

e ( µ

m)

Electric Field (V/µm)

Mean C

harge (e)

• Metalization was typically Cr/Au or Ti/Au or Ti/W → new

• Polycrystalline CVD diamond typically “pumps” by a factor of 1.5-1.8

• Usually operate at 1V/µm → drift velocity saturated

• Test Procedure: dot → strip → pixel





Diamond

Growth side of a recent polycrystalline CVD (pCVD) diamond.

(Courtesy of Element Six)





Diamond

In 2000 RD42 entered into a Research Program with Element Six to increasethe charge collected from pCVD diamond.

Latest Diamonds Measured with a 90Sr Source:

• System Gain = 124 e/mV• QMP = 62mV = 7600e• Mean Charge = 79mV = 9800e

• Source data well separated from 0• Collection Distance now 275µm• Most Probable Charge now ≈ 8000e• 99% of PH distribution now above3000e

• FWHM/MP ≈ 0.95 — Si has ≈ 0.5• This diamond available in large sizes

The Research program worked!





Diamond

History of Diamond Progress

*

Charge Collection in DeBeers CVD Diamond

0

100

200

1994 1995 1996 1997 1998 1999 2000 2001 2002

RD42 Goal

Time (year)

Co

llect

ion

Dis

tan

ce (

mic

ron

s)





Diamond

Recent pCVD diamond wafer ready for test:





Diamond - Tracking Studies

CERN Testbeam Setup for Diamond Telescope:

4 diamond stripsin frame 1 in frame 2

3 diamond strips

plastic

V V

30 mm

scintillationtrigger

150 mm

V HV H H V H V V H H H VV Hdirection of strips

H=horizontal

V=vertical

pion beam100 GeV/c

4 x Si4 x Si

CDS-88 CDS-90 UTS-23 CDS-82

P2P1P2

UTS-24CDS-83

P2P1 P2 P1

• 100 GeV/c pion/muon beam

• 7 planes of CVD diamond strip sensors each 2cm × 2cm

• 50µm pitch, no intermediate strips → new metalisation procedure

• 2 additional diamond strip sensors for test

• Several silicon sensors for cross checks

• Strip Electronics (2 µsec) → ENC ≈ 100e + 14e/pF






Photograph of Two Planes of the Telescope:






PH Distribution on each Strip

channel number [ ]0 20 40 60 80 100 120

sig

nal

ch

arg

e, t

wo

str

ips

[e]

-5000

0

5000

10000

15000

20000

25000CDS-90

mean values on channels

Diamond Tracker CDS-90, Signal Charge on Strips

Residual versus Track Position

track position between two strips [ µ m] -25 -20 -15 -10 -5 0 5 10 15 20 25

spat

ial r

eso

luti

on

, sq

rt( ⟨

(u h -

u t ) 2 ⟩ )

, [ µ m

]

0

5

10

15

20

25

30

CDS-69

Diamond, 2-Strip Non-Linear Eta Residuals

• Uniform signals on all strips → new metalisation

• Pedestal separated from “0” on all strips

• 99% of entries above 2000 e

• Mean signal charge ∼ 8640 e → new metalisation

• MP signal charge ∼ 6500 e






Residuals

]2mµ

0.5

×

entr

ies

[per

120

0

5

10

15

20

25

30

35

40

m]µ, [track - Uhit(2cg)

residuals, U-100 -50 0 50 100

m]

µ [

trac

kU

-6000

-4000

-2000

0

2000

4000

6000

CDS-83-P1

Nent = 23513

Mean x = -0.965

Mean y = 1494

RMS x = 20.33

RMS y = 1867

215 10 2

1232 20399 1432

0 1 222

Integ = 2.04e+04

Diamond Detector PlaneCDS-83-P1

Residuals perpendicular to Strips

CDS-83-P1

Nent = 23513

Mean x = -0.965

Mean y = 1494

RMS x = 20.33

RMS y = 1867

215 10 2

1232 20399 1432

0 1 222

Integ = 2.04e+04]2

mµ 0

.5

×en

trie

s [p

er 1

20

0

5

10

15

20

25

30

35

40

]2mµ

0.5

×

entr

ies

[per

120

0

5

10

15

20

25

30

35

m]µ, [track - Uhit(2cg)

residuals, U-100 -50 0 50 100

m]

µ [

trac

kV

-6000

-4000

-2000

0

2000

4000

6000

CDS-83-P1

Nent = 23513

Mean x = -0.9702

Mean y = 190.9

RMS x = 20.31

RMS y = 1906

119 13 34

1291 20389 1420

37 8 202

Integ = 2.039e+04

Diamond Detector PlaneCDS-83-P1

Residuals along Strips

CDS-83-P1

Nent = 23513

Mean x = -0.9702

Mean y = 190.9

RMS x = 20.31

RMS y = 1906

119 13 34

1291 20389 1420

37 8 202

Integ = 2.039e+04

]2mµ

0.5

×

entr

ies

[per

120

0

5

10

15

20

25

30

35






Next advance → take advantage of charge sharing:

Use intermediate strips to force charge sharing.






Radiation Hard Diamond Tracking Modules:

20 40 60 80 100 120 140 160 1800

500

1000

1500

2000

2500

htempNent = 37904 Mean = 50.2RMS = 16.38

max abs(tim-346)<2 && sig<170 htempNent = 37904 Mean = 50.2RMS = 16.38

• Large (2cm × 4cm) Module constructed with new metalisation

• Fully radiation hard SCTA128 electronics → 25ns peaking time

• Tested in a 90Sr → ready for beam test and irradiation

• Charge distribution cleanly separated from the noise tail → S/N > 8/1

• Efficiency will be measured in test beams at 40 MHz clock rate





Diamond Pixel Detectors

ATLAS FE/I Pixels (Al)

• Atlas pixel pitch 50µm × 400µm• Over Metalisation: Al• Lead-tin solder bumping at IZM in Berlin

CMS Pixels (Ti-W)

• CMS pixel pitch 125µm × 125µm• Metalization: Ti/W• Indium bumping at UC Davis

→ Bump bonding yield ≈ 100 % for both ATLAS and CMS devices

New radiation hard chips produced this year.






Results from a CMS pixel detector

Efficiency Resolution

• Results with 200µm collection distance diamond

Efficiency ∼ 94%

Spatial resolution ∼ 31µm for 125µm pitch






Results from a CMS pixel detector

Efficiency vs Pixel

• Inefficient pixels due to bump bonding and/or electronics - shown in pulser tests

• Excellent correlation between beam telescope and pixel tracker data!





Diamond Radiation Hardness Studies with Trackers

Proton Irradiation Studies with Trackers:

Signal to Noise

2 strip transparent signal to single strip noise [ ]0 50 100 150 200 250

entr

ies/

bin

[1/

100]

0

100

200

300

400

500

600

700

800before irradiation

mean 57, most prob. 41FWHM 54

after 1 E15 protons/cm2

mean 49, most prob. 35FWHM 41

Diamond CDS-69 at 0.9 V/µm

after 2.2 E 15 protons/cm2

and re-metalizationmean 47, most prob. 35FWHM 36

Signal from Irradiated Diamond Tracker

• Dark current decreases with fluence• S/N decreases at 2× 1015/cm2

• Resolution improves at 2× 1015/cm2

Resolution

entr

ies/

2µm

[ ]

0

100

200

300

400

500

600

700

800

Residual Distributions, Proton Irradiated Diamond

Diamond CDS-69

before irradiationσ = 11.5 µm

2-strip center ofgravity method

0

200

400

600

800

1000

CDS-69

after 1E15 p/cm2

σ = 9.1 µm

residual, uh-ut, [µm]-100 -80 -60 -40 -20 0 20 40 60 80 100

0

200

400

600

800

1000

1200

CDS-69

after 2.2 E 15 p/cm2

and re-metalization

σ = 7.4 µm





Diamond Radiation Hardness Studies with Trackers

Pion Irradiation Studies with Trackers:

Signal to Noise

2-strip transparent signal to single strip noise [ ]0 50 100 150 200 250

entr

ies

0

100

200

300

400

500

600

700

800

CDS-38 at 0.6 V/µmbefore irradiation

mean 66most prob. 44FWHM 58

cm2⁄after 2.9 E 15 π

mean 28

and re-metalization

most prob. 21FWHM 27

Signal Distributions from Diamond Tracker

Resolution

entr

ies/

2µm

0

50

100

150

200

250CDS-38before irradiation

σ = 13.7 µm

2-strip center of gravity

method

Residual Distributions in Diamond Tracker

residual, uh-ut [µm]-100 -80 -60 -40 -20 0 20 40 60 80 100

entr

ies/

2µm

0

50

100

150

200

250

300CDS-38

cm2⁄after 2.9 E 15 π

σ = 10.6 µm

and re-metalization

2-strip center of gravitymethod

• Dark current decreases with fluence

• 50% loss of S/N at 2.9× 1015/cm2

• Resolution improves 25% at 2.9× 1015/cm2





Diamond Future: Single Crystal CVD Diamond

Could we make a CVD diamond with improved characteristics?

• Remove the grain boundaries, defects , etc.

• Lower operating voltage.

• Eliminate pumping.

This is single crystal CVD (scCVD) diamond: [Isberg et al., Science 297 (2002) 1670].

CD135 - Both Sides - Pumped (Sr-90 source)

0 10000 20000 30000 40000 50000

Signal Size (Electrons)

0

100

200

300

400

500

Num

ber

per

400

Ele

ctro

ns





Single Crystal CVD Diamond

HV characteristics

E-Field (V/micron)0 0.2 0.4 0.6 0.8 1 1.2

Co

llect

ion

Dis

tan

ce (

mm

)

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

scCVD HV Curve

High quality scCVD diamond collects all the charge at E=0.2V/µ!





Single Crystal CVD Diamond

Pumping characteristics

Time (min)1 10 10

210

3

Co

llect

ion

Dis

tan

ce (

um

)

0

100

200

300

400

500

600

700

071415 Pump-up Curve (HV=500V)

High quality scCVD diamond does not pump!





Silicon Carbide

Structures in 4H-SiC:

The properties of silicon carbide are in some sense the geometric meanbetween silicon and diamond. As a result one hopes to take advantage of thestrengths of both. Two types of SiC structures have been studied:

In SemiInsulating material the charge collection depends on native defects;Epitaxial material has low native defects but only exists in thin layers.





Silicon Carbide

Characterization of SiC:

Source Setup





Silicon Carbide

Charge Distributions from Semi-insulating 4H-SiC:

Semi-insulating SiC works but has problems with defects, full chargecollection and stability.





Silicon Carbide

Charge Distributions from Epitaxial 4H-SiC:

Signal to Noise of 7:1 attained with Epitaxial SiC. Signal just separated fromthe pedestal.





Silicon Carbide

Charge Distributions from Epitaxial 4H-SiC:

Epitaxial SiC has been shown for thin layers to collect all the charge atelectric fields of ∼ 1.5V/µm.





Radiation Monitoring - A New Application - BaBar, Belle, CMS

Motivation:

→ Radiation monitoring crucial for silicon operation/abort system

→ Abort beams on large current spikes

→ Measure calibrated daily and integrated dose

→ BaBar/Belle presently use silicon PIN diodes, leakage current increases 2nA/krad

→ After 100fb−1 signal≈10nA, noise≈ 1-2µA

→ Large effort to keep working, BaBar/Belle PIN diodes will not last past 2004-05





Radiation Monitoring

The BaBar/Belle Diamond Radiation Monitor Prototypes:

• Package must be small to fit in allocated space

• Package must be robust

Schematic View

Diamond

Au Contact

In SolderHV Insulation

Ground BraidKapton Insulation

Copper Shield







Photo of Belle Prototype Device Photo of Packaged Belle Prototype







Photo of Installed BaBar Device Photo of Installed Belle Device

BaBar device inside the silicon vertex detector.

Belle device just outside the silicon vertex detector.






The CMS Diamond Radiation Monitor Program:

• Diamond activity has begun!

• Test beam emulating beam accident in Autumn 2003

• Possible location in the CMS detector:

Monitors

Simulation of a Beam Accident in CMS






Results on Calibration in BaBar:

• In BaBar during injection relative to silicon diodes: 5.9mrad/nC (Feb)

• In BaBar during injection relative to silicon diodes: 5.8mrad/nC (Apr)

• Correlation coefficient unchanged over several months

Diamond Current/nA0 5 10 15 20 25 30 35 40 45 50

Dio

de

Rat

e/m

Rad

/s

0

50

100

150

200

250

300

0.2552 ±p0 = -4.897

0.01263 ±p1 = 5.916

February

BE:MID rate vs BE diamond current 0.2552 ±p0 = -4.897

0.01263 ±p1 = 5.916

Diamond Current/nA0 5 10 15 20 25 30 35 40 45 50

Dio

de

Rat

e/m

Rad

/s0

50

100

150

200

250

300

0.2533 ±p0 = -7.085

0.01236 ±p1 = 5.801

April

BE:MID rate vs BE diamond current 0.2533 ±p0 = -7.085

0.01236 ±p1 = 5.801

Calibration repeatable but so far limited by systematics






Data Taking in BaBar:

0

5

10

15

20

25

0 2500 5000 7500 10000 12500 15000 17500 20000 225000

200

400

600

800

1000

1200

1400

1600

1800

20000 2500 5000 7500 10000 12500 15000 17500 20000 22500

BE diamond current

BE:MID diode signal current (Sig 11)

LER current

Time/seconds

Dia

mon

d/D

iode

cur

rent

/nA

Time/secondsH

ER

cur

rent

/mA

Fast Abort Soft Abort

System operating for 4 months in BaBar and works well!






Leakage Current in BaBar

• Diamonds have received 250kRad 60Co plus 250kRad while installed

• No observed change in leakage current (<0.1nA) or fluctuations (30pA)

• Data directly from BaBar SVTRAD system

• Electronic noise (≈ 0.5nA) substracted off

0

0.02

0.04

0.06

0.08

0.1

0.12

20 40 60 80 100 120 140February 5, No Beams Time (s)

Cur

rent

(nA

)

April 17, No Beams Time (s)

Cur

rent

(nA

)

0

0.02

0.04

0.06

0.08

0.1

0.12

20 40 60 80 100 120 140






Very Fast Time Scale (ns) in BaBar

• Use a fast amplifier to look at PIN-diode and diamond signals

• Trigger on the PIN-diode signal

• Look at fast spikes: red = diamond, black = PIN-diode

Diamond is fast enough for Fast Abort






An attempt at final packaging

Ceramic Package





The Future

• Diamond and silicon carbide have very promising futures

• Diamond work is being pursued by RD42 pCVD → scCVD

• SiC work is being pursued by RD50 epi layers → 100µm

• Present pCVD diamonds should surpass the performance of present silicon at around1015p/cm2

• Semi-insulating SiC will require lots of engineering





Summary

• Charge Collection in Diamond

270 µm collection distance diamond attained in pCVD research contract

MP signal ≈ 8000 e

99% of charge distribution above 3000 e

Attained S/N=60/1 with 2µs shaping time; 8/1 at 25ns

FWHM/MP ∼ 0.95 – Working with manufacturers to increase uniformity

This diamond process now in production reactors

Single crystal CVD diamond is here: >450 µm collection distance attained

MP signal ≈ 13000 e

99% of charge distribution above 10000 e

FWHM/MP ∼ 0.30

• Charge Collection in Silicon Carbide

40 µm collection distance epitaxial SiC attained

Full charge collection at E ∼ 1.5V/µm

Attained S/N of 7/1 with 2µs shaping time using a source

Wafer diameters up to 3 cm and thicknesses up to 100µm soon

Tracking devices now being fabricated





Summary

• Radiation Hardness of Large Bandgap SemiConductors

Using trackers allows a correlation between S/N and Resolution

Dark current decreases with fluence

Some loss of S/N with fluence

Resolution improves with fluence

Tests must be repeated with more trackers and latest pCVD and scCVD

diamonds and Epitaxial 4H-SiC

• Radiation Monitoring

Successfully tested BaBar and Belle devices

CMS performing tests this summer

Radiation monitoring should lead to the development of the next levelradiation hard devices





Future Plans for RD42

• Charge Collection

Continue research program to improve pCVD material:

collection distance → 300µm (Q = 10, 800e)

→ improved uniformity

→ identification of trapping centers

Begin research program on scCVD diamond

• Radiation Hardness of Diamond Trackers and Pixel Detectors

Continue tracker irradiations this year, add pixel irradiations

With Protons:

→ 5× 1015/cm2

With Pions:

→ 5× 1015/cm2

With Neutrons:

→ 5× 1015/cm2

• Beam Tests with Diamond Trackers and Pixel Detectors

→ trackers with intermediate strips, SCTA128 electronics

→ pixel detectors with ATLAS and CMS radhard electronics now available!

→ construct the first full ATLAS diamond pixel module

• Material Research

→ Florence, OSU, Paris, Rome





Future Plans for RD50

Goals: Define optimal materials and device structures to ensure bestradiation tolerance.

• Defect Engineering of Si

Oxygen, Oxygen dimmers, etc

• New Materials

SiC, GaN

• New Geometries

3D, thin detectors

• Defect Modeling and Device Simulation

Detectors should (soon) be able to handle the highest luminosities of theSLHC!





non-silicon solid state detectors harris kagan

Documents