The LHCb RICH - CERN · 2018-10-12 · The LHCb RICH detectors Two RICH detectors RICH 1 (C4F10):...
Transcript of The LHCb RICH - CERN · 2018-10-12 · The LHCb RICH detectors Two RICH detectors RICH 1 (C4F10):...
The LHCb RICH
Silvia Gambetta,University of Edinburgh
on behalf of the LHCb RICH collaboration
July 30, 2018
Introduction
The LHCb experiment
General purpose single arm forward spectrometer(2 < η < 5, 4% of solid angle)
[J. Instrum. 3 (2008) S08005]
LHCb physics:
rare b- and c- hadrondecays
CP-violation in b sector
CKM parameters
indirect search for NP
spectroscopy
electroweak physics
Unprecedented collection of bottom and charm hadronsVery successful physics programme
3 / 27
The LHCb performance
LHCb designed to run at lower luminositythan ATLAS and CMS
mean number of interactions per bunchcrossing ∼ 1
pp beams displaced to reduce theinstantaneous luminosity:L ∼ 4× 1032cm−2s−1
twice the luminosity design value
Excellent performance in 2017!
∼ 3 fb−1 of pp collisions at 7-8 TeV in Run 1
∼ 5 fb−1 of pp collisions at 13 TeV in Run 2
expect to reach ∼9 fb−1 at the end of Run 2
4 / 27
The LHCb RICH detectorsTwo RICH detectors
RICH 1 (C4F10): upstream, 2 GeV/c - 60 GeV/c over 25 mrad - 300 mradRICH 2 (CF4): downstream, 30 GeV/c - 100 GeV/c over 15 mrad - 120 mrad
Charged particles produce Cherenkov radiation then focused on Hybrid PhotonDetectors (HPD) planeResponsible for charged hadrons particle identification (PID)
5 / 27
RICH 1
VELO exit
covers the low and intermediatemomentum region over the fullangular acceptance
detector close to the interactionpoint to limit its volume
two halves above and below thebeam pipe
gas enclosure sealed directly onVELO exit window to limit materialbudget
tilted spherical focusing primarymirrors (4 segments)
secondary flat mirrors to limitdetector length (16 segments)
The total material budget for RICH1 is only about 5% X0 within theexperimental acceptance
gas radiator: C4F10
6 / 27
RICH 1
7 / 27
RICH 2
7.2
m
spherical
mirrors
at
mirrors
magnetic
shieldsPhoton
Detectors
covers the high momentum region over the the angular range 15-120 mraddownstream detector since high momentum particles are less affected mymagnetic fieldtwo halves on the sides of the beam pipetilted spherical focusing primary mirrors (56 segments)secondary flat mirrors to limit detector length (40 segments)The total material budget for RICH 2 is about 15% X0 within the experimentalacceptancegas radiator: CF4
8 / 27
RICH 2
9 / 27
The Hybrid Photon Detectors
HPDs arranged incolumns with ancillaryfront-end electronics
LV and HV boards topower the column
Level-0 boards transmitdata to Level-1 boards(off detector) via opticallinks
Level-1 boards zerosuppress and transmitdata to the DAQ
photon detectors developed in collaboration withindustry for application in the LHCb RICH system:
quartz window with thin multialkaliphotocathode
cross focusing optics ⇒ photons focused onsilicon pixel array (granularity 2.5× 2.5mm2)
front-end chip encapsulated in the vacuumtube
magnetic shield installed around each HPD
active diameter 75 mm:
196 tubes in RICH 1288 tubes in RICH 2
10 / 27
RICH hitmaps
/RICH/Default Run 212325, started 2018-07-23 05:41:42, duration: 00:27:54
00 01 02 03 04 05 06 07 08 09 10 11 12 13
D0
D1
D2
D3
D4
D5
D6
0
50
100
150
200
250
300
350
400
310×HitMap for Rich1 bottom panel
13 12 11 10 09 08 07 06 05 04 03 02 01 00
U6
U5
U4
U3
U2
U1
U0
0
50
100
150
200
250
300
350
400
310×HitMap for Rich1 top panel
0 1000 2000 3000 4000 5000 6000 7000 8000 9000# hits / event
0
5000
10000
15000
20000
25000NHitInclusiveRich2Entries 543908Mean 1723Std Dev 873.6
Number of hits in Rich2
C0 C1
C2
C3
C4
C5
C6
C7
C8
15
14
13
12
11
10
09
08
07
06
05
04
03
02
01
00
0
20
40
60
80
100
120
310×
HitMap for Rich2 CSide-right panel
0 1000 2000 3000 4000 5000 6000 7000 8000 9000# hits / event
0
5000
10000
15000
20000
25000
NHitInclusiveRich1Entries 543908Mean 1995Std Dev 1008
Number of hits in Rich1
A0 A1
A2
A3
A4
A5
A6
A7
A8
15
14
13
12
11
10
09
08
07
06
05
04
03
02
01
00
0
20
40
60
80
100
120
310×
HitMap for Rich2 ASide-left panel
11 / 27
Alignment and Calibration
Intervention on RICH 1 Down box
Intervention performed during 2016-2017 winter shutdown: replacement of someHPDs, L0 and LV boards
13 / 27
Magnetic field correction (once per year)
HPD image distortion due to themagnetic field
distance from the measured lightspot to the test point before andafter correction
Dedicated setup illuminating the HPDs in RICH 1with pattern of light spots with magnet on
0 500 1000 1500 2000 2500 3000 35000
50
100
150
200
20
40
60
80
100
120
140
RICH1 - down
D0
D1
D2
D3
D4
D5
D6
12
11
10
09
08
0706
05
04
03
02
01
1300
0 50 100 150 200 250 300 350 400 4500
0
0
0
0
0
0
0
0
0
0
✒ ✑
✓ ✏
✒ ✑
✓ ✏ ✒ ✑
✓ ✏
✒ ✑
✓ ✏
14 / 27
HPD image calibration (once per run ≃1h)
The position of the Photocathode image on the anode can change due to thermal andcharging effects degrading the reconstruction of the Cherenkov angle and affect thePID performance
image cleared from ionfeedback
edges detected applyingSobel filter
fit to determine the radiusand the centre position usedthen by reconstruction
0
1
2
3
4
5
6
7
8
9
10
HPD Image Shifts (x,y) | Run 212325
2− 1.5− 1− 0.5− 0 0.5 1 1.5 2
x / mm
2−
1.5−
1−
0.5−
0
0.5
1
1.5
2
y /
mm
Entries 444
Mean x 0.05336−
Mean y 0.05805
Std Dev x 0.3001
Std Dev y 0.3173
0
1
2
3
4
5
6
7
8
9
10
Entries 444
Mean x 0.05336−
Mean y 0.05805
Std Dev x 0.3001
Std Dev y 0.3173
HPD Image Shifts (x,y) | Run 212325
15 / 27
Refractive index calibration (once per run ≃1h)
The refractive index of the radiators can vary due to temperature, pressure andvariation in the gas mixture. Monitoring of these quantities not precise enough forphysics analysis ⇒ further correction needed
Very high momentum tracks, with well defined Cherenkov angle, are used to evaluatethe difference between the measured and the expected Cherenkov angles
1100
1200
1300
1400
1500
1600
310×
En
trie
s
Rich1Gas Rec-Exp Cktheta | All photonsEntries 1.993715e+08
Mean 0.0002482
Std Dev 0.004308
/ ndf2χ 898.4 / 90
Gaus Const 5.656e+05
Gaus Mean 0.0006317
Gaus Sigma 0.001666
Bkg Par 0 1.044e+06
Bkg Par 1 9.45e+06
Bkg Par 2 1.237e+09−
Bkg Par 3 1.609−
Entries 1.993715e+08
Mean 0.0002482
Std Dev 0.004308
/ ndf2χ 898.4 / 90
Gaus Const 5.656e+05
Gaus Mean 0.0006317
Gaus Sigma 0.001666
Bkg Par 0 1.044e+06
Bkg Par 1 9.45e+06
Bkg Par 2 1.237e+09−
Bkg Par 3 1.609−
0.008− 0.006− 0.004− 0.002− 0 0.002 0.004 0.006 0.008
delta(Cherenkov Theta) / rad
900
1000
1100
1200
1300
1400
1500
1600
310×
En
trie
s
Rich1Gas Rec-Exp Cktheta | All photonsEntries 1.993715e+08
Mean 0.0002482
Std Dev 0.004308
/ ndf2χ 898.4 / 90
Gaus Const 5.656e+05
Gaus Mean 0.0006317
Gaus Sigma 0.001666
Bkg Par 0 1.044e+06
Bkg Par 1 9.45e+06
Bkg Par 2 1.237e+09−
Bkg Par 3 1.609−
Run 212325SF 1.024069
Rich1Gas Rec-Exp Cktheta | All photonsRich1Gas Rec-Exp Cktheta | All photons
600
700
800
900
310×
En
trie
s
Rich2Gas Rec-Exp Cktheta | All photonsEntries 1.044003e+08
Mean 05−2.305e
Std Dev 0.002123
/ ndf2χ 864 / 88
Gaus Const 3.543e+05
Gaus Mean 0.0001728−
Gaus Sigma 0.0007202
Bkg Par 0 5.31e+05
Bkg Par 1 6.261e+06
Bkg Par 2 3.948e+09−
Bkg Par 3 0.7025−
Entries 1.044003e+08
Mean 05−2.305e
Std Dev 0.002123
/ ndf2χ 864 / 88
Gaus Const 3.543e+05
Gaus Mean 0.0001728−
Gaus Sigma 0.0007202
Bkg Par 0 5.31e+05
Bkg Par 1 6.261e+06
Bkg Par 2 3.948e+09−
Bkg Par 3 0.7025−
0.004− 0.003− 0.002− 0.001− 0 0.001 0.002 0.003 0.004
delta(Cherenkov Theta) / rad
500
600
700
800
900
310×
En
trie
s
Rich2Gas Rec-Exp Cktheta | All photonsEntries 1.044003e+08
Mean 05−2.305e
Std Dev 0.002123
/ ndf2χ 864 / 88
Gaus Const 3.543e+05
Gaus Mean 0.0001728−
Gaus Sigma 0.0007202
Bkg Par 0 5.31e+05
Bkg Par 1 6.261e+06
Bkg Par 2 3.948e+09−
Bkg Par 3 0.7025−
Run 212325SF 0.988728
Rich2Gas Rec-Exp Cktheta | All photonsRich2Gas Rec-Exp Cktheta | All photons
Correction factor extracted and used to compute the refractive index
16 / 27
Mirror alignment (once per fill ≃12h)
mirror misalignment can cause a displacement of thecentre of the ring with respect to the track
a dependence of the variation of the Cherenkov angleon the azimuthal angle is observed ⇒ sinusoidaldistributions
distribution fitted and misalignment corrected inreconstruction
Mirror alignment monitored during 2015 and 2016 ⇒ alignment tolerances determined
⇒ full automation implemented in 2017
17 / 27
Mirror tilts
alignment constants determined for each mirror
correlation of different mirror pairs taken into account
iterative procedure on a dedicated farm allows to extract mirror alignment to beapplied by the reconstruction for each fill
18 / 27
Mirror alignment stability in 2017 (RICH 1)
αy
αz
βy
βz
Alignment number [a.u.]0 50 100 150
Y V
aria
tion [
mra
d]
∆
0.1−
0.05−
0
0.05
0.1LHCb RICH 1
Preliminary
Primary Mirrors
2017/06/02 - 2017/11/29
primary mirror 0primary mirror 1primary mirror 2primary mirror 3
Alignment number [a.u.]0 50 100 150
Z V
aria
tion [
mra
d]
∆
0.1−
0.05−
0
0.05
0.1LHCb RICH 1
Preliminary
Primary Mirrors
2017/06/02 - 2017/11/29
primary mirror 0primary mirror 1primary mirror 2primary mirror 3
Alignment number [a.u.]0 50 100 150
Y V
aria
tion [
mra
d]
∆
1.5−
1−
0.5−
0
0.5
1
1.5 LHCb RICH 1Preliminary
M������ ����������t� � ���� �� ������
2017/06/02 - 2017/11/29
S�������� �������
Alignment number [a.u.]0 50 100 150
Z V
aria
tion [
mra
d]
∆
1−
0.5−
0
0.5
1 LHCb RICH 1Preliminary
���� �! � "� ! #$$% &' (#)(*()+�, -(��.�!
2017/06/02 - 2017/11/29
/01234567 896626:
19 / 27
Performance
Cherenkov resolution stability over time
Timestamp
RICH 1 resolution [mrad]
1.9
1.7
1.5
1.3
1.1
2016-04-30 2016-11-01 2017-05-05 2017-11-06 2018-05-11
Timestamp
RICH 2 resolution [mrad]0.99
0.89
0.79
0.69
0.59
0.49
0.39
0.29
Timestamp2016-04-30 2016-11-01 2017-05-05 2017-11-06 2018-05-11
Excellent stability!!!21 / 27
PID calibration sample
PID performance studied with pure samples: each particle species is selected withkinematical information only
pions performance studied using K0S→ π+π−
kaons performance studied using D∗ → D0(
K−π+)
π+
protons performance studied using Λ0 → pπ−
1.84 1.86 1.88 1.90
m(K−π+) GeV/c2
0
50000
00000
50000
200000
250000
Candidates/028
MeV/c2
LHCb
D;+→ D0π+
D0→ K−π+
1100 1110 1120 1130
m( π−) MeV/c2
0
0000
20000
0000
0000
50000
Candidates/00
MeV/c2
LHCb
Λ0→ π−
example of mass distribution for PID samples [PoS (ICHEP2016) 295]
22 / 27
RICH PID performance (2017)
Momentum (MeV/c)0 20 40 60 80 100
310×
Eff
icie
ncy
0
0.2
0.4
0.6
0.8
1
1.2
1.4) > 0πLL(K - ∆
) > 5πLL(K - ∆
K→K
K→π
= 13 TeV 2017 validationsLHCb
Momentum (MeV/c)0 20 40 60 80 100
310×
Eff
icie
ncy
0
0.2
0.4
0.6
0.8
1
1.2
1.4) > 0πLL<= >∆
) > 5πLL<= >∆
p→p
p→π
? @A BCD EF@G HIJKNIOKPQsLHCb
Momentum (MeV/c)0 20 40 60 80 100
310×
Eff
icie
ncy
0
0.2
0.4
0.6
0.8
1
1.2
1.4RRTU V WX Y Z∆
RRTU V WX Y [∆
p→p
p→\
] ^_ `ab cd^e fghijgkilmsLHCb
Excellent PID performance!!!23 / 27
Gas purity RICH 1
gas purity monitored during the year
RICH 1 filled with CO2 during wintershutdown and back with C4F10 beforethe beginning of the run
air contamination increases during theyear due to small leak
gas cleaning performed during technicalstops to ensure the stability of theperformance
Gas contamination affects refractive index, dispersion, photon yield....
24 / 27
Impact of gas purity
two periods in 2017 selected: gas clean and contamination above the limit forcleaning
both periods with the same magnet polarity (UP)
use calibration samples to compare PID performance in two time periods25 / 27
PID dependence on gas purity
0 20 40 60 80 100
310×
Track Momentum [MeV/c]
0
0.2
0.4
0.6
0.8
1
Eff
icie
ncy
K Clean Gas Sample→K K >1% Air Contamination→K
K Clean Gas Sample→ π K >1% Air Contamination→ π
0 20 40 60 80 100
310×
Track Momentum [MeV/c]
0
0.2
0.4
0.6
0.8
1
Eff
icie
ncy
p Clean Gas Sample→p p >1% Air Contamination→p
p Clean Gas Sample→ π p >1% Air Contamination→ π
0 20 40 60 80 100
310×
Track Momentum [MeV/c]
0
0.2
0.4
0.6
0.8
1
Eff
icie
ncy
p Clean Gas Sample→p p >1% Air Contamination→p
p Clean Gas Sample→K p >1% Air Contamination→K
PID performance stable in time
26 / 27
Conclusions
LHCb RICH detectors successfully and stably operating since 2010
Key ingredient to the successful physics program delivered by LHCb
]2c[GeV/−pKm
5.2 5.4 5.6 nop
q
2 c
rsuvwvsxyz{|}~�y�{
0
200
400
600
p��
1000 LHCb−
pK→b0
Λ
−πp→b
0Λ
+π−K→
0B
−K+K→s
0B
Comb. bkg.
Part. reco. bkg.
]2c[GeV/+Kpm
5.2 5.4 5.6 5.8
)2 c
Can
did
ates
/ (
10
MeV
/
0
200
400
600
800
1000 LHCb+Kp→b
0Λ
+πp→b
0Λ
−π+K→
0B−
K+K→s0B
Comb. bkg.
Part. reco. bkg.
]2c[GeV/−πpm
5.2 5.4 5.6 5.8
)2 c
Can
did
ates
/ (
10
MeV
/
0
100
200
300
400
500
600
700
800LHCb
−πp→b
0Λ
−pK→b
0Λ
−π+K→
0B−
π+π→0B
Comb. bkg.
Part. reco. bkg.
]2c[GeV/+πpm
5.2 5.4 5.6 5.8
)2 c
Can
did
ates
/ (
10
MeV
/
0
100
200
300
400
500
600
700
800LHCb
+πp→b
0Λ
+Kp→b
0Λ
+π−K→
0B
−π+π→
0B
Comb. bkg.
Part. reco. bkg.
“Search for CP violation in Λ0b→ pK− and Λ0
b→ pπ− decays”
[LHCb-PAPER-2018-025]
27 / 27
Thank you for your attention!
Extra slides
Timing scan
3 timing scans acquiredduring 3 bunch fill
RICH1 already well timed
corrections of timing appliedfor RICH2
further scan acquired toconfirm timing
Run 205938, after timingcorrection
RICH2 gas
RICH2 gas status monitored by gasgroup and plot available online
CO2 well contempt well controlled by gasgroup (very stable)
small air contamination observed thisyear
contamination is decreasing while newgas is pushed in
Mirror alignment stability in 2017 (RICH 2)
αy
αz
βy
βz
Alignment number [a.u.]0 50 100 150
Y V
aria
tion [
mra
d]
∆
0.1−
0.05−
0
0.05
0.1LHCb RICH 2
Preliminary
Primary Mirrors
2017/06/05 - 2017/11/29
������� ���������the 56 individual mirrors
Alignment number [a.u.]0 50 100 150
Z V
aria
tion [
mra
d]
∆
0.1−
0.05−
0
0.05
0.1LHCb RICH 2
Preliminary
Primary Mirrors
2017/06/05 - 2017/11/29
������� ���������the 56 individual mirrors
Alignment number [a.u.]0 50 100 150
Y V
aria
tion [
mra
d]
∆
0.2−
0.15−
0.1−
0.05−
0
0.05
0.1
0.15
0.2
LHCb RICH 2Preliminary
������� ���������the 40 individual mirrors
��� ¡�¢¡�£ ¤ ��� ¡��¡�¥
¦§¨©ª«¬® ¯°©±
Alignment number [a.u.]0 50 100 150
Z V
aria
tion [
mra
d]
∆
0.2−
0.15−
0.1−
0.05−
0
0.05
0.1
0.15
0.2LHCb RICH 2
Preliminary²³´µ¶´· ´¶¸´¶·¶¹º
the 40 individual mirrors
»¼½¾¿¼À¿¼Á  »¼½¾¿½½¿»Ã
ÄÅÆÇÈÉÊËÌ ÍÎËËÇËÏ