Progress in Thick GEM-like
(THGEM)-based Detectors
R. Chechik, A. Breskin, C. Shalem, M. Cortesi & G. Guedes
Weizmann institute of science, Rehovot, Israel
V. Dangendorf
Physikalisch Technische Bundesanstalt, Braunschweig, Germany
D. Vartsky & D. Bar
SOREQ NRC, Yavne, Israel
MOTIVATION:
Robust, economic, large-area radiation imaging detectors
FAST, HIGH-RATE, MODERATE LOCALIZATION RESOLUTION
SNIC @ SLAC April 2006
R. Chechik et al.
Gaseous multiplication in holes
• Avalanche confined within a small volume
• Secondary effects confined/reduced -> high gains
• True pixilated structures
• Possibility to CASCADE multipliers -> further higher gains
1e- in
„Breskin,
Charpak NIM108(1973)427 Discharge in glass
capillaries
„Lum et al. IEEE NS27(1980)157,
Avalanches
& Del Guerra et al. NIMA257(1987)609
in holes
„Sakurai
et al. NIMA374(1996)341, Glass Capillary plates
& Peskov et al. NIMA433(1999)492
„Sauli
103-104 e- out
High electric field in
the holes
SNIC @ SLAC April 2006
NIMA386(1997)531
GEM
„Ostling,
Peskov et al, IEEE NS50(2003)809 G-10
“Capillary plates”
„Chechik et al. NIMA535(2004)303
& Physics/0502131
THGEM
R. Chechik et al.
Thick GEM-like multipliers: THGEM
R. Chechik et al. NIM A535 (2004) 303-308
& R. Chechik et al. NIM A553 (2005) 35-40
Manufactured by standard PCB techniques of
precise drilling in G-10 (+other materials) and Cu etching.
etching
ECONOMIC & ROBUST !
Hole diameter d= 0.3 - 1 mm
Dist. Bet. holes a= 0.7- 4 mm
Plate thickness t= 0.4 - 3 mm
A small THGEM costs ~3$ /unit.
With minimum order of 400$ Æ ~120 THGEMs.
~10 times cheaper than standard GEM.
SNIC @ SLAC April 2006
R. Chechik et al.
Standard GEM
103 gain in
single GEM
THGEM
gain in
single-THGEM
105
1mm
Cu
Important:
0.1mm G-10 rim.
reduces discharges
-> high gains!
Is THGEM an “Expanded” GEM ?
expanded
•The dimensions
does not scale up
•Electron diffusion & transport
•Electric fields
•Gain
•Timing properties
•Rate capability
•Ions transport
It is a new device that has to be studied.
SNIC @ SLAC April 2006
R. Chechik et al.
Operation mechanism
Semi-transparent
photocathode
Reflective
photocathode
(role of all fields)
Edrift
EHole
Etrans
Garfield simulation of electron multiplication in Ar/CO2 (70:30)
Need to study the effect of all fields on:
e- and ion transport in/out holes
-> efficiency to single-electron events
-> cascade THGEMs
-> ion backflow to the conversion gap
Electric fields at photocathode surface
-> operation with solid photocathode
Gain, signal rise-time, rate capability, localization, etc
SNIC @ SLAC April 2006
R. Chechik et al.
Photon detector w reflective CsI PC
deposited on the THGEM top face
Example:
(e.g. RICH photon detectors)
Electric field on photocathode surface
Electric
field
photocathode
surface
hole
dipole
field
created
byon
the
Maxwell
software
calculation
40
0.4mm thick
0.3mm holes
∆∆V
VGEM= 2200V
THGEM=2200V
∆V
=2200V
30
0.7mm pitch
∆∆V
VGEM= 1200V
=1200V
∆VTHGEM
=1200V
THGEM
E [kv/cm]
Ref PC
THGEM
e
>3kV/cm
∆∆V
V
= 800V
∆V
=800V
THGEM=800V
GEM
THGEM
20
10
5
0
-0.08
-0.04
0.00
0.04
0.08
Distance from center [mm]
Require: • High field on the PC surface (for high QE).
• Good e- focusing into the holes (for high detection efficiency).
• Low sensitivity for ionizing background radiation (e.g. RICH).
SNIC @ SLAC April 2006
R. Chechik et al.
Ref PC
e
E
E
E=0
Relative
Edrift
e- transfer efficiency [%]
Photon detector w reflective CsI PC
deposited on the THGEM top face (2)
Gain~10 3
1 Atm.
Ar/CH 4(95:5)
100
2.0
80
1.5
60
1.0
40
0.5
20
0
0.0
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
Edrift [kv/cm]
Focusing is done by hole dipole field.
• Maximum efficiency at Edrift =0.
• Slightly reversed Edrift (50-100V/cm)
good photoelectron collection low sensitivity to MIPS !
SNIC @ SLAC April 2006
R. Chechik et al.
Ref PC
e
E
E
E=0
Relative
Edrift
e- transfer efficiency [%]
Photon detector w reflective CsI PC
deposited on the THGEM top face (2)
Gain~10 3
1 Atm.
Ar/CH 4(95:5)
100
2.0
80
1.5
60
1.0
40
0.5
20
0
0.0
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
Edrift [kv/cm]
Focusing is done by hole dipole field.
• Maximum efficiency at Edrift =0.
• Slightly reversed Edrift (50-100V/cm)
good photoelectron collection low sensitivity to MIPS !
SNIC @ SLAC April 2006
R. Chechik et al.
Ref. PC
Electron transfer efficiency
Single-THGEM: e- transport into holes.
Role of Ehole
CH4
Ar/CH4 Ar/CO2
1.4
CF4
gain 30 gain 20 gain 6 gain 3
1.2
1.0
0.8
0.6
0.4
0.2
Edrift =0
0.0
0
400
800 1200 1600 2000 2400
∆VTHGEM [v]
Full electron transfer efficiency into the holes, already at low gain.
-> good single-electron detection efficiency
-> good energy resolution with highly ionizing radiation
SNIC @ SLAC April 2006
R. Chechik et al.
Single-THGEM: e- transport into holes.
Role of Ehole (2)
gain 100
ST PC
With ST PC require:
Edrift ~1kV/cm
Transfer efficiency
1.2
1.0
0.8
Ar/CO2(70:30)
Pure CH4
gain 10
Pulse counting mode
0.6
0.4
Current mode
0.2
Edrift =1kv/cm
0.0
0
400
800
1200
1600
∆V THGEM
(V)
∆VTGEM
2000
2400
2800
(v)
Full electron transfer efficiency into the holes, already at low gain.
-> good single-electron detection efficiency
-> good energy resolution with highly ionizing radiation
SNIC @ SLAC April 2006
R. Chechik et al.
Single-THGEM: gain
Example: THGEM photon detector with reflective CsI photocathode.
R. Chechik et al. NIMA553 (2005) 35-40
Ar/CH4(5%) Ar/CO (30%)
2
Atm. Pressure
CH4
6
10
5
10
10
4
CF4
Effective Gain
10
Ref. PC
105
3
10
10
2
10
1
10
10
0
10
-1
Single THGEM
10
10
-2
t=0.4, d=0.3, a=0.7 [mm]
10
-3
10
10
0
500 1000 1500 2000 2500 3000
∆VTHGEM [v]
• Gain up to 104-105 with single electrons (sparks)
Similar gain w ST PC.
SNIC @ SLAC April 2006
R. Chechik et al.
Single-THGEM: counting rate
5
Effective gain
10
Ref. PC
4
10
3
10
2
10
5
10
6
10
7
10
8
10
Rate [electrons / mm2 sec]
•Signal rise time < 10ns
•Rate capability: ~10MHz/mm2 (space charge)
SNIC @ SLAC April 2006
R. Chechik et al.
Double-THGEM: Cascaded operation
role of Etrans
C. Shalem et al. NIM A558 (2006) 475-489
e-
Ehole > Etrans
e- focused into hole
Ehole < Etrans
e- collected on GEM top
Require: Large Etrans for good extraction from THGEM1
Small Etrans for good focus sing into THGEM2
→ Optimization
SNIC @ SLAC April 2006
R. Chechik et al.
Double-THGEM: cascaded operation
Role of Etrans (2)
Ar/CO2(70:30) 760torr
1.4
Transfer efficiency
Both
0.4mm thick
0.3mm holes
0.7mm pitch
Pulse counting mode
1.2
1.0
0.8
0.6
0.4
∆VTGEM=1800V (Gain~10 )
3
∆VTGEM=1700V (Gain~10 )
∆VTGEM=1300V ( Gain~10 )
4
0.2
0.0
0
1
2
3
4
5
6
7
8
9
Etrans
EDrift (kv/cm)
efficient transfer to the 2nd THGEM:
•up to high Etrans (e.g. 3kV/cm).
•at relatively low THGEM gains.
SNIC @ SLAC April 2006
R. Chechik et al.
Double-THGEM multiplier: Gain, rise time
R. Chechik et al. NIMA553 (2005) 35-40
Etrans=3kV/cm
Atm. pressure Ar/CH4(5%) Ar/CO
2(30%)
10ns
Effective Gain
Fast signals, Double THGEM
(t=1.6mm d=1mm, a=1.5mm).
Atm. Pressure Ar/30%CO2
Total gain=~ 106
7
10
double
5
single
10
107
single
3
10
1
10
Double & single THGEM
-1
t=0.4, d=0.3, a=0.7 [mm]
Symmetric
operation
10
symmetric operation
-3
10
0
400 800 1200 1600 2000 2400
• Higher total gain (106-107) w single e-.
• >103 higher gain at same ∆VTHGEM
• Better stability
SNIC @ SLAC April 2006
double
∆VTHGEM [v]
0.4mm thick
0.3mm holes
0.7mm pitch
R. Chechik et al.
Double-THGEM multiplier: ion backflow
C. Shalem et al. NIM A558 (2006) 475-489
ST PC
ion
e-
Ion backflow smaller than with 4-GEM multiplier.
Prolonged PC life-time.
SNIC @ SLAC April 2006
R. Chechik et al.
2D double-THGEM detector: a flat
imaging detector
Resistive anode (e.g. C paint sprayed on PCB)
e-
.
Resistive anode:
•Signal broadening
•HV decoupling
•Electronics spark protection
C paint->3 MΩ/square
-> ~70% charge transmission
evaporated Ge-> 30 MΩ/square
->~95% charge transmission
2D
readout
anode
Radiation
(gas
Conversion)
THGEMs
Simple and economic readout scheme.
Induced-signal width matched to readout-pixel size.
SNIC @ SLAC April 2006
R. Chechik et al.
10x10 cm2 2D double-THGEM detector
0.4mm thick,
0.5mm Ø holes,
1mm pitch
• 2x 10x10cm2 THGEM
• 2-sided pad-string anode (0.5mm thick)
• Delay-line readout (SMD)
front side
back side
X & Y
DL-signals
2 mm pitch, 1.35ns/mm
SNIC @ SLAC April 2006
R. Chechik et al.
Ar/CH4 (95:5) 8keV x-rays
Conversion gap = 10 mm, 1kV/cm
Transfer Gap
= 2 mm, 3.5 kV/cm
Induction Gap = 1 mm, 4 kV/cm
∆V THGEM=1210 Volt
Gain =6x103
Gain and uniformity
Flat field
Energy resolution
L o cenergy
a l S p e c tru
m F e E n e of
rg y 6rekeV
s o lu tio
n = 21%
Local
spectrum
x-rays
55
3000
12
2500
FWHM 21%
2000
Count
No. of points
10
gain unformity
25 measurement points
8
6
1500
1000
4
2
0
500
+ - 10%
0
0
3
5x10
3
6x10
3
7x10
3
8x10
100 200 300 400 500 600 700 800 900 1000
C h a n n e ls
Gain
SNIC @ SLAC April 2006
R. Chechik et al.
Localization : linearity, resolution
Gain =6x103
5 lp/cm:
1mm wide slits,
2mm center to center
10 lp/cm:
0.5mm wide slits,
1mm center to center
1mm wide slits,
2mm center to center
Modulation transfer function
1.0
0.5mm wide slits,
1mm center to center
Modulation
0.8
50% @ 6lp/cm
0.6
0.4
0.2
0.0
0
2
4
6
8
10 12
Frequency (lines pairs per cm)
SNIC @ SLAC April 2006
0.4mm thick,
0.5mm Ø holes,
1mm pitch
Sub-mm resolution !
R. Chechik et al.
Long-term stability
10000
2
0.02kHz/mm 8 keV
8000
2
0.2 kHz/mm 6 keV
gain
6000
Our 10x10 cm2 double THGEM detector
Ar/CH4, 6 & 8 keV x-rays, low rate
~25% changes
4000
2000
0
0
5
10
15
20
25
hours
THGEM
GEM
Sauli:
double THGEM, Ar/CO2, 6keV x-rays,
1kHz/mm2.
Stabilize ~12 hours, x2 gain rise.
• Charging up?
• Insulator polarization?
results from
HV? Rate? total current? history?
SNIC @ SLAC April 2006
R. Chechik et al.
Long-term stability (2)
ST PC
•Measured currents w UV + ST PC.
ST
•Edrift=0.15kV/cm.
•Initial flux 5x105 e-/mm2 (~1kHz/mm2 x-ray).
Could not be raised due to PC decay.
•Possible additional effects from p, temp & HV
instability.
1mm
i
Double THGEM
Single THGEM
10
6
PC
1mm
i
ETRANS=3kV/cm
2x1500V (increase 50 V only) [3]
Gain
1900V i0=1pA [1]
10
4
1800V i0=1pA [2]
10
3
1700V i0=1pA [3]
0
5
10
15
20
25
10
5
10
4
10
3
10
2
2x1450V (change by +100 -50V only) [2]
2x1400V i0=1pA [1]
0
hours
• THGEMs require a few hours of stabilization.
5
10
15
20
hours
• Stabilization time depends on total gain/current.
• gain variation ~ factor 2.
• could depend on history (time after THGEM & gas introduction).
SNIC @ SLAC April 2006
25
R. Chechik et al.
Long-term stability
0.4mm thick,
0.3mm Ø holes,
1mm pitch
Double THGEM
10
6
(3)
0.4mm thick,
0.6mm Ø holes,
1mm pitch
ETRANS=3kV/cm
Double THGEM Etrans=3kV/cm
10
5
10
4
2x1500V (increase 50 V only) [3]
2x1800V [3]
2x1450V (change by +100 -50V only) [2]
104
103
1002
10
gain
Gain
105
2x1700V [2]
10
3
2x1400V i0=1pA [1]
2x1500V [1]
10
0
5
10
15
hours
20
25
2
0
5
10
15
20
25
hours
• Larger hole → shorter stabilization time. Effect of the bare insulator??
• Evidence for combined dependence on history + total gain/current
SNIC @ SLAC April 2006
R. Chechik et al.
Planned studies
1. Systematic study on long-term stability vs. rate (with x-ray).
2. Understanding the effect of the 0.1 mm rim (e.g. reduce it).
3. Studying THGEMs of different materials;
e.g. CIRLEX = polyimide (Kapton).
SNIC @ SLAC April 2006
R. Chechik et al.
Applications
LARGE-AREA DETECTORS, ROBUST, MODERATE COST
ns, sub-mm, MHz/mm2
• Particle tracking at moderate (sub-mm) resolutions.
e.g. muon-detector @LHC2.
• TPC readout.
• Sampling elements in calorimetry. (ILC??)
• Readout of light from LXe detector (XENON).
• Moderate-resolution, fast (ns) X-ray and n imaging.
• Single-photon imaging.
e.g. Ring Imaging Cherenkov (RICH) detectors (presently w GEMs).
advantages: robust,
high eff. for single photons,
low sensitivity to ionizing BG.
SNIC @ SLAC April 2006
R. Chechik et al.