application of diamond detectors to proton beam

35
Application of Diamond Detectors to proton beam measurements and ultrafast timing Marcin Jastrząb The Henryk Niewodniczański Institute of Nuclear Physics PAN Kraków, Poland Joint LIA COLL - AGAIN COPIGAL POLITA WORKSHOP 26 - 29 April 2016, LNS Catania

Upload: others

Post on 26-Feb-2022

4 views

Category:

Documents


0 download

TRANSCRIPT

Application of Diamond Detectors

to proton beam measurements

and ultrafast timing

Marcin Jastrząb

The Henryk Niewodniczański Institute of Nuclear Physics PAN

Kraków, Poland

Joint LIA COLL-AGAIN COPIGAL POLITA WORKSHOP 26-29 April 2016, LNS Catania

2

Diamond detectors for particle detection and TOF measurements

Specifications of two cyclotrons operating at IFJ PAN. Requirements for measurement systems to be used with CVD Diamonds.

Performance and challenges for front-end electronics for Diamonds. Emphasis on low-noise performance and wide bandwidth (>1GHz).

Multiple particle detection (pile-up)

TOF measurement preliminary results

Outlook and future plans

Outline

3

Parameter Diamond 4H-SiC Si

Bandgap eV (nm) 5.5 3.26 1.1

Dielectric constant 5.7 9.7 11.8

Thermal conductivity W/um*K*1e-6

22 4.9 11.8

electron mobility um2/V*ns

450 100 150

Hole mobility um2/V*ns 380 11.5 45

Saturated electon mobility um/ns

270 220 100

Saturated hole mobility um/ns

270 90

Dielectric breakdown field V/um

1000 250 30

electron-hole pair creation [eV]

13.6 3.26 3.6

intrinsic carrier ni concentration at 300 K cm-3

~1e-27 ~1e-5 1.07E+07

atoms/cm3 1.77E+23 5.01E+22

Element Six (main material

supplier) scCVD diamond

material:

Typically 100% single sector {100},

[N] < 5 ppb,

[B] < 1 ppb

Orientation {100} faces, <110>

edges (+/-3° crystallographic

orientation)

Surface finish Ra<5nm on two

polished {100} faces.

Diamond detectors offer

high radiation resistance,

especially working in

transmission mode (no

Bragg peak in the material)

Diamond material as particle detector.

Mostly created artificially by Chemical Vapor Deposition process.

Single (scCVD)- and poly (pcCVD)- crystaline. CCE ~ 100% for scCVD.

4

From planar detectors to 3D diamonds

F. Bachmair et al. A 3D diamond detector for particle tracking,

NIM A Volume 786, 21 June 2015, pp. 97–104

3D technology advantages:

- even higher radiation resistance

- higher S/N

- signal pulse duration depends on

distance between electrodes

rather than detector thickness

5

1. Single e-h pair generated in the middle of det. volume

0 100p 200p 300p 400p 500p

0,0

0,1

0,2

0,4

0,5

0,6

0,7

0,9

1,0

1,1

Charg

e(e

)

Time (ps)

charge

Total charge

333ps = d/ve=v

h

Qe=Q

h=e/2

QTotal

=e

0 100p 200p 300p 400p 500p

0

200p

400p

600p

800p

1n

333ps = d/ve=v

h

Single charge

created in the middle

Thickness 50 um

ve=v

h150 m/ns

Cu

rre

nt

(pA

)

Time (ps)

0,0 100,0p 200,0p 300,0p 400,0p 500,0p0,0

2,0µ

4,0µ

6,0µ

8,0µ

10,0µ

12,0µ

14,0µ

Cu

rre

nt

(uA

)

Time (ns)

Current in ideal diamond

no capacitance, 50 um

70 MeV proton - single

no of pairs 11010

SRIM calculation

ih = i

e

iTot

T=v/d

eN/T

0 100p 200p 300p 400p0

200

400

600

800

1000

1200

T=v/d

ch

arg

e e

time s

Single proton 70 MeV

ScCDV 50 m S

continous charge along track

Qe=Q

h

eN

QTotal

=Qe+Q

h

2. Continuous charge distribution along particle track+-

d

x

ER

V

C G

f -3dB

Signal generation in Diamond Detector

Typical E : 1V/μm

of det. thickness.

Carrier drift velocities

however saturate only

at ~ 10 V/μm (ve=vh).

by Tomasz Nowak

6

Diamond detector signal. The effect of the detector

capacitance.

50Cdet

0,0 200,0p 400,0p 600,0p 800,0p 1,0n

-1,0µ

0,0

1,0µ

2,0µ

3,0µ

4,0µ

5,0µ

6,0µ

7,0µ

8,0µ

9,0µ

10,0µ

11,0µ

12,0µSingle proton 70 MeV

ScCVD 50 m Selectroda

~3.13 mm2

no capacitance, R=50, charge along track

C=4.02 pF, R=50, charge along track

Cu

rre

nt@

50

[A

]

Time ps

FWHM ~261 ps

90%

10%

tr=71 ps

f-3db

~5 GHz

Where is the limit in terms of rise time when the bandwidth f-3dB = 1 GHz is used? Tr =~ 350 ps.

Capacitance C=e*S/d= 4.04 pF@50 um

by Tomasz Nowak

50μm scCVD Diamond

and 70MeV protons

registered with Tektronix

DPO5104 1GHz 10GS/s

oscilloscope

Sig

na

l A

mp

litu

de

[V

]

Sample no. (10 GS/s)

5 ns

Signal rise time (90-10%) ~500 ps | FWHM ~800 ps

single proton

three protons

7

Two cyclotrons at IFJ: AIC-144 & IBA Proteus-235

AIC-144

IBA Proteus-235

Beam energy 60 MeV

Beam current up to 80 nA

Magnet Leg Diameter 144 cm

Magnetic Structure 4 spiral sectors

Magnetic Field 0,85 ÷ 1,8 T

Main Coil Current 0 ÷ 650 A

Number of Harmonic Coils 4

Trim Coils Current ±400 A

Number of Dees 1 (α=180º)

RF Generator Frequency 10 ÷ 27 MHz

Beam energy 70 - 230 MeV

Ion beam current @ 230 MeV: 1 - 500 nA

Magnet yoke outside diameter 4,34 m

Magnet leg diameter 2,1 m

Magnetic structure 4 spiral sectors

Maximum magnetic field 3,1 T

Maximum current in main magnet coil 800 A

Number of dees 2 (45°)

RF system operation frequency 106 MHz

Dee voltage 50 - 100 kV

Extraction Factor 70%

8

The AIC-144 & IBA Proteus-235 beams: time structures.

IBA Proteus-235

0 10m 20m 30m 40m

0,0

0,2

0,4

0,6

0,8

1,0

Cu

rre

nt

[uA

]

Time [ns]

Repetition 20 ms

Duty cycle 0.460 ms

Fill factor ~2.3%

AIC-144

0,0 1,6 3,1 4,7 6,3 7,9 9,4 11,0 12,6 14,1

-1,0

-0,5

0,0

0,5

1,0

0 60 120 180 240 300 360 420 480 540

HF

vo

ltag

e r

ela

tive

Time ns

HF frequency=106 MHz

T=9.43 ns

30

780 ps

Degrees

Beam microstructure: Time distance

between adjacent micropulses = 38.08 ns

0,0 9,6n 19,1n 28,7n 38,2n 47,8n 57,4n 66,9n 76,5n

-56

-28

0

28

56

HF

vo

lta

ge

[k

V]

Time [ns]

HF= 26.26 MHz

T = 38.08 ns

B ~ 1.8 T

Micropulses

time distance

<38.08 ns width

<3.4 ns

12 041 micropulses

in 460us macropulse

Time ns

Beam macrostructure: Time distance

between adjacent macropulses = 20 ms

Beam microstructure: Time distance

between adjacent micropulses = 9.43 ns

No beam macrostructure.

Quasi-continuous beam.

9

38.08 ns

600 μs

Time structure of the AIC-144 @ 50μm scCVDD

10

The AIC-144 beam time structure measured by scCVDD

38.08 ns

protons impinging on Diamond

~3.7 ns

11

proton interaction

electronic (preamplifier) noise

The Proteus-235 beam time structure with 50μm scCVDD

Semi-continous beam (no macrostructure).

Distance betwee beam micropulses = 9.43 ns

10 ns

12

Performance and challenges for front-end electronics for diamond detectors.

13

Fast (Pulse) preamplifier

1.5 - 2 Ghz of Bandwidth

Rise time ~150 ps

Gain min. 45 dB ~180xDiamonddetector

Digital oscilloscope

with min. 10 GS/s

sample rate and

min. 2 GHz of

Bandwidth

Applications at IFJ PAN

Particle counting and TOF measurements:

• Single particle detection with high efficiency for each micropulse separately

• Mechanism of dealing with pile-up within micropulses. Is it possible to distinguish between

multiple, contemporary protons within single micropulse?

• TOF resolution < 50 ps for measurements of beam monochromaticity

Because of very low intrinsic noise of Diamond material, one of the most critical

elements, often setting limits in performance of detection systems with CVD Diamond

detectors is the preamplifier.

• RF technology >1.5 GHz bandwidth

• Signal amplification ~150-300 V/V for low-LET particles and/or thin detectors (single 70MeV proton loses

(the most probable value) ~150 keV (~11 000 electrons) in 50μm scCVD, which turns into

~20 mV pulse amplitude after 45dB of amplification). This is very comparable to one MIP in 300μm

(~10 800 electrons) in scCVD Diamond.

• Extremely low noise RMS requirement @ high amplification

Diamond detector front-end electronics

Wideband Electronics Highlights:

Wideband Electronics for MIP-level signals

14

DBA IV – the reference preamplifier (2 GHz of bandwidth)

The reference broadband (2 GHz) preamplifier with the gain of up to 50 dB,

designed and developed at GSI Darmstadt, was used. DBA IV was originally

optimized for measurements of heavier ions with Diamond detectors.

DBAIV main characteristics

15

scCVDD signal amplitude

spectrum assymetry = signal.

Electronic noise has

a gaussian distribution.

Signal pulse amplitude [V]

Sig

nal am

plitu

de [

V]

The DBAIV preamplifier cont.Signal spectrum of 70MeV protons measured @ Proteus-235 with 50μm

scCVD detector by DDL (Diamond Detectors Ltd.)

16

CVD front-end electronics design R&D: PA-10 preamp.

A new design of the low-noise and high-gain

has been carried out in Krakow.

The design minimum requirements:

Improvement of S/N ratio @ minimum bandwidth

of 1.5 GHz and gain >= 45 dB.

The result is the PA-10 wideband preamplifier.

The performance of the PA-10 allows to continue

with signals by particles generating >~10 000

electrons of the signal as for single particle

detection.

An exemplary gain measurement of a single stage

prototype preamplifier by R&S network analyzer

PA-10 - 45dB preamplifier

In terms of TOF performance, the improvement

of the S/N may be required and optimized,

depending on timing resolution needed in

application.

17

Random Jitter 1.87 ps

Test signal – 2GHz sinusoidal wave

Wideband amplifier PA-10 tests

and qualification.

Noise level and jitter.

CVD front-end electronics: PA-10 laboratory tests

Noise RMS: 2.49 mVGAIN: 45 dBOscilloscope bandwidth: 4GHz

Random jitter < 2ps

18

CVD front-end electronics: PA-10 beam testSignal spectrum of 70MeV protons measured @ Proteus-235 with 50μm scCVD

detector by DDL (Diamond Detectors Ltd.)

Log scale Linear scale

Sig

nal am

plitu

de [

V]

Signal pulse amplitude [V]

Signal of ~11000 e-Signal of ~11000 e-

(MIP@300μm scCVD

~ 10 800 e-)

19

R² = 0.9985

0.00E+00

2.00E+03

4.00E+03

6.00E+03

8.00E+03

1.00E+04

1.20E+04

1.40E+04

0 100 200 300 400 500 600

avera

gep

roto

ns/1

ms*d

ete

cto

r are

a

Proteus Beam current - before degrader [nA]

GANTRY Beam Current measurements with DIAMOND D05 Detector

ScCVD 50 um bias 150 V@isocenter 70 MeV protons

Diamond 15mV

proton energy 70 MeV 150 MeV 225 MeV

average energy loss [eV] 151097 85900 66059

SD eV 39432 32644 28350

SD% 26.1 38.0 42.9

no of pairs e-h 11110 6316 4857

normalized to 70 MeV [%] 100 56.9 43.7

CVD front-end electronics: PA-10. Beam intensity

measurements.

Signal amplitude

comparison for different

proton beam energies

20

DBA IV & PA-10 technical summary

DBAIV PA-10

Bandwidth(-3db) 3 MHz - 2 GHz 1 MHz - 1.5 GHz

Gain max. 50 dB (~316) fixed 45 dB (~178)

Input impedance 50 Ω 50 Ω

Output impedance 50 Ω 50 Ω

Max bias voltage +/- 2 kV +/- 500 V

Power supply +12 V, 150 mA +12 V, 75 mA

Noise r.m.s. 5.9 mV@45 dB 2.5 mV@45 dB

Preamplifiers for CVD Diamonds: DBAIV vs PA-10

21

Main parameters to be defined

according to application

Spatial resolution

Dimensions (total area)

Fill Factor

Signal (from single pixel) time duration

Time resolution

Concept of a segmented CVD detector for time-and space- beam profilometry

22

Proteus-235 beam profiles measured in Experimental Hall

of CCB (Cyclotron Centre Bronowice).

Distance from Ion Guide End (IGE): Z=0 Distance from IGE: Z=0.7 m

Distance from IGE: Z=2 m

85mm

65

mm

Measurements have been performed

in the air by ProBImS measurement

system developed at IFJ PAN and

composed of scintillating screen and

high-resolution ATIK 383 L+ digital

camera.

by Marzena Rydygier

23

AIC-144 cyclotron beam diagnostics at IFJ PAN. Pile-up resolving.

24

Signal amplitudes from 60MeV protons with 500μm scCVD are ~ 3x higher than ones measured with 50μm scCVD and have a duration of ~4.5 ns FWHM instead of ~1.2 ns in the case of 50μm detector.

scCVD detector 500μm @ 350V bias voltage

AIC-144 beam micropulses

Sig

na

l a

mp

litu

de

[V

]

AIC-144: micropulse by micropulse proton beam diagnostics with 500μm scCVD.

25

Digitally integrated signal – charge spectrum

Amplitude spectrum

500μm scCVD

digital measu-

rements with

10Gs/s sample

rate.

The charge spec-

trum (upper)

profits from an

increased S/N.

Signal amplitude [V]

AIC-144: multiple (pile-up) proton separation within micropulse

26

TOF preliminary measurements with heavy ions and 70MeV protons

27

Start signal: 32S at 50deg.

Test of performance of scCVD at Heavy Ion Laboratory (HIL), Warsaw.

Ag target

Collaboration:

HIL UW, IFJ Krakow, CEA Saclay, Univ. of Huelva

STOP signal: Ag at 58deg.

scCVD 50μm

Beam 32S – 90MeV

scCVD 100μm

28

TOF test setup at HIL. Configuration of detectors during the experiment.

100μm

50μm

29

Measurement of kinematics with heavy ions (up to 100 MeV) at HIL.

TOF was measured with 5GHz analog bandwidth

LeCroy SDA5000 oscilloscope @ 20GS/s

PA-20 preamplifiers

Time

resolutions

30

TOF measurements in Krakow CCB (Bronowice Cyclotron Centre) – hadrontherapy site

31

TOF preliminary measurements with 70MeV protons at Gantry site with two scCVD’s: 100μm and 300μm set in series (with respect to beam direction).

Detector’s signals were measured

and stored by 5GHz analog

bandwidth LeCroy SDA5000 5 GHz

(analog bandwidth) oscilloscope /

SDA @ 20GS/s.

Beam direction. Intensity profile σ ~3mm

Wideband preamplifiers

with 20dB gain

32

TOF preliminary results at Gantry: scCVD pulses.

Signals from 100μm and 300μm detectors were amplified by single PA-10

preamplifier and two PA-20 preamplifiers (in series) for the second detector.

Signal amplitudesapprox. 35 mVRMS noise < 3mV

Pulse width ~1ns FWHM

33

TOF preliminary results with 70MeV protons at GantryDIGITAL CFD + CUBIC INTERPOL. OF ZERO CROSSING

FWHM for the set

of two detectors

TOF results with 70MeV protons is ~125 ps (FWHM) for the set of two detectors.

Assuming both detectors equal (not actually the case) the distibution σ < 40 ps for

single scCVD Diamond detector.

σ2time = [trise/(S/N)]2 + σ2

t,quantization + σ2t,CVDD

34

Summary and perspectives

Proteus 235 produced protons (70-230 MeV) are challenging to detect because of the low signal (~150 keV for 70 MeV) and high bandwidth.

One of the most important Nuclear Physics (and medical) application at IFJ PAN is the precise measurement of Proteus-235 proton energies before they arrive to the „target”. The required precision <0.3 % can only be reached by TOF technique. To accomplish the task, the system TOF precision has to be <50 ps, assuming the distance between detectors to be ~2m.

Preliminary TOF test beam provide ~125 ps FWHM of resolution (σ < 40 ps per detector, assuming two identical scCVD’s).

For systems with diamond detectors, one of the major limiting factor is the front-end electronics. State-of-the-art preamplifiers are capable of separating signal from noise (for 50μm scCVD) for proton energies up to ~100 MeV. New developments may be necessary in order to cope with higher particle energies, thus lower energy loses.

35

Our collaborators:

CEA-SaclayMichał Pomorski

University of Huelva

Ismael Martel Bravo

Heavy Ion Laboratory (HIL) WarszawaPaweł NapiorkowskiMichalina KomorowskaKatarzyna Wrzosek-LipskaJędrzej Iwanicki Paweł MatuszczakAgnieszka Trzcińska

IFJ PAN KrakówPiotr BednarczykTomasz NowakMarzena RydygierMichał CiemałaKrzysztof DrozdowiczJan Dankowski Witold Męczyński