View
12
Download
0
Category
Preview:
Citation preview
Universita degli Studi di Milano–Bicocca
Istituto Nazionale di Fisica Nucleare – Sezione di Milano–Bicocca
The Performance of the LHCb Pixel Hybrid Photon Detectors
in a 25ns Structured Test-Beam
D. L. Perego
on behalf of the LHCb RICH Collaboration
2007 IEEE Nuclear Science Symposium – October 29th, 2007Honolulu – Hawaii (U.S.A.)
Outline
1. introduction
– the LHCb experiment at the LHC
– RICH detectors
– Pixel Hybrid Photon Detectors (HPD)
2. September 2006 Test Beam
3. photoelectron yield analysis
– the model
– N2 and C4F10 photoelectron yields
4. Cherenkov angle resolution (early study)
5. conclusions
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 2 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
The LHCb Experiment
250mrad
100mrad
M1
M3M2
M4 M5
RICH2HCAL
ECALSPD/PS
Magnet
T1T2T3
z5m
y
5m
− 5m
10m 15m 20m
TTVertexLocator
RICH1
1. detector construction nearing completion
2. detector commissioning in progress
3. LHCb ready to collect data at the LHC start–up
RICH 1
RICH 2
p p
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 25 50 75 100 125 150 175 200
p (GeV/c)
θ (r
ad)
RICH 1
RICH 2
Radiator aerogel C4F10 CF4
L 5 cm 85 cm 167 cmn 1.03 1.0014 1.0005
θmaxC 242 mrad 53 mrad 32 mrad
σtot(θC) 2.5 mrad 1.4 mrad 0.7 mrad
Npe 5.3 24.0 18.4
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 3 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
The Pixel Hybrid Photon Detectors (HPD)
A total of 484 pixel HPDs produced to detect Cherenkov photons
1. vacuum tube, diameter 83 mm, height 120 mm
2. S20 multi–alkali photocathode
3. 200 − 600 nm sensitive wavelength range
4. average QE improved during production up to ∼ 35%
5. cross focusing optics (∆V = 20kV), demag factor ∼ 5
6. photoelectrons focused on the pixelized silicon anode
7. ∼ 500k pixels over total active area ∼ 3.3m2
Further details by T. Blake (N53–1)
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 4 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
September 2006 Test Beam (1)
1. first test with LHC time structure
– complete RICH detector, prototype placed at SPS Hall H6 (CERN)
– final versions of the read–out electronics and DAQ
– analysis using LHCb software
2. tilted parabolic mirror to focus Cherenkov photons
3. three RICH2 columns with 48 HPDs in total
4. N2 and C4F10 gas as Cherenkov radiators
– L ∼ 1 m
5. beam properties
– 25 ns bunch spacing
– 80 GeV/c beam from CERN–SPS
– mixture of π (80%), e (10%), K (7%), p (3%)
– average < 1 particle per bunch train
6. trigger
– two scintillators in coincidence
– no veto on event with more than one particle
7. tracking
– two HPD silicon sensors as tracking stations
Radiator Vessel
Detector Housing
Quartz Window
Mirror
HPDs
HPD column
Beam
Aluminum Foil
First test withLHC bunch spacing
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 5 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
September 2006 Test Beam (2)
Beam
Tracker HPDs
Gas Enclosure
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 6 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Radiators and Data Taking
1. N2 and C4F10 gas radiator used
– N2 data: n = 1.000297 =⇒ Cherenkov ring on single HPD
– C4F10 data: n = 1.0014 =⇒ Cherenkov ring on 3 or 4 HPD
2. LED and dark count runs (−→ noise studies)
3. goals of this test beam
– both hardware and software global tests
– data acquisition and analysis checks
– photoelectron yield studies (N2 and C4F10 data)
– Cherenkov angle resolution evaluation
test the full electronic chain
at the same time
X (mm)-750 -700 -650 -600 -550 -500 -450
Y (
mm
)
-100
-50
0
50
100
150
0
500
1000
1500
2000
2500
3000
3500
4000
4500
RUN0008 - N2
X (mm)-750 -700 -650 -600 -550 -500 -450
Y (
mm
)
-100
-50
0
50
100
150
0
20
40
60
80
100
310×RUN0027 - C4F10
X (mm)-750 -700 -650 -600 -550 -500 -450
Y (
mm
)-100
-50
0
50
100
150
0
10
20
30
40
50
60
70
310×RUN0028 - C4F10
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 7 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Photoelectron Yield Analysis
1. fit N2 and C4F10 rings on an event–by–event basis
– require at least 5 hits in each event
– fit a circle (3 free parameters)
2. determine a road around the average ring centre
– select signal and reject background events with large number of fired hits outside the signal region
– road defined as R ± ∆R (∆R = 3 pixels for N2; ∆R ∼ 1.7 pixels for C4F10)
3. selection of events with 4 or more hits inside the road and less than 3 outside
4. histogram the number of hits in the road for selected events
Pixel Column0 5 10 15 20 25 30
Pix
el R
ow
0
5
10
15
20
25
30
0
200
400
600
800
1000
1200
1400
1600
1800
2000
RUN0008_HPD_117
Pixel Column0 5 10 15 20 25 30
Pix
el R
ow
0
5
10
15
20
25
30
5000
10000
15000
20000
25000
30000
35000
40000
RUN0027_HPD_222
Number of Hits in the Road0 10 20 30 40 50
0
2
4
6
8
10
12
310×RUN_0027_HPD_222N2 C4F10 C4F10
11,601 events 108,216 events
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 8 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Modelling the Photoelectron Yield
1. photoelectron yield from a constrained fit to the N(n) hit distribution
2. fit model
– sum of Poisson contributions modelling Cherenkov emission
– beam composition: 80% π and 10% e (saturated), 7% K and 3% p (relevant for N2 only)
– describe one, two (and three) particle contributions in the beam
3. HPD effects (measured in dedicated LED and dark count runs)
– account for pixel–to–pixel charge sharing (∼ 3% on average)
– double hits contamination
4. vary the µ parameter in the fit model
Preliminary Results
N1 = 14880 ± 124N2 = 351 ± 29
µ = 12.42 ± 0.04s = 0.016 (fixed)
χ2/NDF = 31.52/23
µexpected ∼ 13
Pixel Column0 5 10 15 20 25 30
Pix
el R
ow
0
5
10
15
20
25
30
0
200
400
600
800
1000
1200
1400
1600
1800
RUN0006_HPD_117
Number of Hits in the Road
0 5 10 15 20 25 30 35
200
400
600
800
1000
1200
1400
RUN0006_HPD_117
Full fit
contributionπK contribution
p contribution
2-particle contribution
HPD Measured Expected
117 12.32±0.12 12.20±0.62264 13.14±0.13 14.09±0.70265 12.56±0.12 12.81±0.65
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 9 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
C4F10 Photoelectron Yield
Pixel Column0 5 10 15 20 25 30
Pix
el R
ow
0
5
10
15
20
25
30
5000
10000
15000
20000
25000
30000
35000
φ∆
RUN0029_HPD_117
Number of Hits in the Road
0 5 10 15 20 25 30 35 40 45 50
-110
1
10
210
310
410
RUN0029_HPD_117
Full fit
contributionπ2-particle contribution
3-particle contribution
N1 = 105401.6 ± 336.0N2 = 1395.1 ± 104.6N3 = 1008.3 ± 72.1µ = 12.72 ± 0.04s = 0.016 (fixed)
χ2/NDF = 19.0/21
1. include and fit 1, 2 and 3 particle contributions
2. only one–particle species assumed (saturated π)
3. reasonable χ2/NDF =⇒ N(n) accurate model of data
4. 2 and 3 particle fraction ∼2% of the total
5. stability of the dµ/d∆φ ratio checked: flattish distribution along φ
6. 10% spread of µ/∆φ in agreement with the QE of the HPDs involved
7. expected yield in good agreement with full Monte–Carlo simulation
8. systematic uncertainties contribute at 5% level
/ ndf 2χ 1.27 / 2
Constant 4.06± 23.01
Mean 0.084± 8.915
Sigma 0.0857± 0.5932
(photoelectrons/radian)φ∆/µ5 6 7 8 9 10 11 12 13 14 15
0
5
10
15
20
25
/ ndf 2χ 1.27 / 2
Constant 4.06± 23.01
Mean 0.084± 8.915
Sigma 0.0857± 0.5932
/ ndf 2χ 1.27 / 2
Constant 4.06± 23.01
Mean 0.084± 8.915
Sigma 0.0857± 0.5932
Preliminary Results
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 10 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Cherenkov Angle Resolutions
1. Cherenkov angle, single photon resolution studies started recently
2. encouraging results using ray tracing procedure
3. N2 data
– σ(θC) ∼ 1.6 mrad
– work ongoing to better understand the various contributions to the resolution
4. C4F10 data
– σ(θC) ∼ 1.6 mrad
– excellent resolution reached due to iterative alignment procedure
– shift in absolute θC: refractive index not calibrated (5 − 10% N2 contamination)
(rad)CθCherenkov angle 0.016 0.018 0.02 0.022 0.024 0.026 0.028 0.03 0.032
0
2
4
6
8
10
12
14
16
18310×
2N
RUN0006_HPD_117
(rad)CθCherenkov angle 0.042 0.044 0.046 0.048 0.05 0.052 0.054 0.056 0.058
0
5
10
15
20
25
310×
10F4C
RUN0027_HPD_222
Preliminary Results
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 11 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Conclusions
1. September 2006 Test Beam
– first test with 25 ns LHC time structure
– electronics and DAQ working at 40 MHz
– three RICH2 columns with 48 HPDs in total
– N2 and C4F10 Cherenkov radiators
– 80 GeV/c mixed beam from CERN–SPS
2. data collected using an early version of the official LHCb online framework
3. analysis done using the official LHCb reconstruction software
– photoelectron yield analysis
– Cherenkov angle resolutions
4. LHCb photon detection system successfully tested
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 12 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Spectral Analysis: the Lomb Method
Back–up slides
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 13 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
B Physics at LHC
1. the aim of the LHCb experiment
– study CP Violation in B decays through precise measurement of the CKM parameters
– hints of New Physics in rare B–decays
2. the Large Hadron Collider
– start–up in fall 2008
– proton–proton at√
s = 14 TeV
– 40 MHz frequency
– L = 2 × 1032 cm−2s−1 – mostly single interaction
– σbb
' 500 µb
3. Nbb ' 1012/year expected
4. all b–hadron species produced: Bd, Bs, Bc and b–baryons
5. very high statistics available for several decay channels
6. bb pairs correlated in the forward/backward cones
– LHCb: single arm forward spectrometer
– acceptance: 300 mrad (H) and 250 mrad (V)
– 1.9≤ η ≤4.9 region
01
23
0
1
2
3
0
2000
4000
6000
8000
10000
x 10 3
θb (rad)
θb
– (rad)
bb angular distr
ibution
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 14 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
RICH Detectors
Particle identification and B0s → K+K− decay channel (as an example)
Clear correlation between momentum and polar angle of particles to be
identified
– 2 RICH detectors
– 3 radiators, one solid (aerogel) and two gaseous (C4F10 and CF4)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 25 50 75 100 125 150 175 200
p (GeV/c)
θ (r
ad)
RICH 1
RICH 2
Radiator aerogel C4F10 CF4
L 5 cm 85 cm 167 cmn 1.03 1.0014 1.0005
θmaxC 242 mrad 53 mrad 32 mrad
σtot(θC) 2.5 mrad 1.4 mrad 0.7 mrad
Npe 5.3 24.0 18.4
250 mrad
Track
Beam pipe
Photon Detectors
Aerogel
VELO exit window
Spherical Mirror
Plane Mirror
C4F10
0 100 200 z (cm)
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 15 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Modelling the Photoelectron Yield
1. photoelectron yield: constrained fit to the N(n) hit distribution
2. properties of the fit function
– account for pixel–to–pixel charge sharing (s) and double hits (d)
– describe one, two (and three) particle contributions in the beam
– beam composition: 80% π and 10% e (saturated), 7% K and 3% p (relevant for N2 only)
3. measured values of charge sharing (from dedicated LED and dark count runs) fixed in the minimization
4. vary the parameter in the fit model
N(n) =
pion/electron termz }| {
Nπ × P (n; µπ, s, d)+
kaon termz }| {
NK × P (n; µK , s, d) +
proton termz }| {
Np × P (n; µp , s, d) +
two−particle termz }| {
N2π × P (n; 2µπ, s, d)
5. probability to observe n fired pixel hits given a photoelectron yield µ as product of Poisson distributions
P (n; µ, s, d) =nX
i=0
∞X
j=0
P (n − i + j; µ)| {z }
number of hits
P (i; (n − i)s)| {z }
charge sharing
P (j; (n − i + j)(n − i + j − 1)d)| {z }
double hits
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 16 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Charge Sharing Studies
1. photoelectron signal optimally contained in a single pixel
2. signal over threshold in two or more adjacent pixels, probability around 3% =⇒ charge sharing
3. important to know its fraction for each HPD to control the photoelectron yield
4. dedicated LED data and toy model
– estimate number of photoelectrons 6= number of hits due to charge sharing
– estimate number of adjacent hits due to two genuine photoelectrons
5. ion feedback and micro discharges, main background contributions of this study
– account for 5% of hits in the HPDs
– clustered in large groups of adjacent pixels
– ion feedback: shower of photoelectrons created by residual ion inside the vacuum tube colliding with
the photocathode
– micro discharge: spark from near the photocathode due to electrical breakdown of the gas around
the tube
6. charge sharing fraction measured for each HPD
– 0.1% precision
– main inaccuracy still from ion feedback
– numbers to be used in the N2 and C4F10 photoelectron analyses
2007 IEEE NSS – October 29th, 2007 D. L. Perego – 17 The Performance of the LHCb Pixel Hybrid Photon Detectors . . .
Recommended