Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Detector Requirements andSoftware for CLIC
André SailerOn behalf of the CLICdp Collaboration
CERN-EP-LCD
Software and Physics Requirements for e+e− CollidersJanuary 16, 2020
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 1 / 29
Table of Contents
IntroductionCLIC AcceleratorBeam-Beam EffectsBeam-induced Backgrounds
Detector, Software, Requirements, and PerformanceReconstruction SoftwareThe Full DetectorTrackerVertex DetectorCalorimeters and Particle Flow Clustering
Electromagnetic Calorimeter
Hadronic CalorimeterForward Calorimetersγγ→ hadron Background Mitigation
Summary
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 2 / 29
CLICdp Collaboration
CLIC detector and physics (CLICdp): 30institutes from 18 countries
CLICdp focuses on CLIC-specific studies ofI Physics prospects and simulation studiesI Detector optimisation and hardware R&D
for CLICI Together with CALICE and FCal
collaborations
CLIC collaboration developing the acceleratortechnology
50 µm thick silicon wafer CLICpix2 + C3PD glue assembly LumiCal Sensor Calorimeter Test Beam Scintillator Tile
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 3 / 29
CLIC Accelerator
The Compact Linear Collider (CLIC) is amulti-TeV electron–positron colliderI Two-beam acceleration with
room-temperature copper, gradient up to100 MV/m
I Main beam comes in 156 ns (176 ns at380 GeV) trains with 50 Hz
I StagingI First stage around 380 GeV for Higgs
and top physics, top-threshold scanI Second stage at 1.5 TeV possible with
single CLIC drive beamI 3 TeV stage with one drive beam complex
for each beam
156 ns 20 ms
0.5 ns
CLIC: trains at 50 Hz, 1 train = 312 bunches
TA
BC2
delay loop2.5 km
decelerator, 4 sectors of 878 m
446 klystrons20 MW, 48 µs
CR2
CR1
circumferencesdelay loop 73 mCR1 293 mCR2 439 m
BDS1.9 km
IPTA
BC2 BDS1.9 km
11 kmCR combiner ringTA turnaroundDR damping ringPDR predamping ringBC bunch compressorBDS beam delivery systemIP interaction point dump
BC1
drive beam accelerator2.0 GeV, 1.0 GHz
time delay line
e+ injector2.86 GeV
e+ PDR
389 m
e+ DR
427 m
booster linac 2.86 to 9 GeV
e+ main linac
e– injector2.86 GeV
e– DR
427 m
e– main linac, 12 GHz, 72 MV/m, 3.5 km (c)FT
Main Beam
Drive Beam
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 4 / 29
CLIC Accelerator
The Compact Linear Collider (CLIC) is amulti-TeV electron–positron colliderI Two-beam acceleration with
room-temperature copper, gradient up to100 MV/m
I Main beam comes in 156 ns (176 ns at380 GeV) trains with 50 Hz
I StagingI First stage around 380 GeV for Higgs and
top physics, top-threshold scanI Second stage at 1.5 TeV possible with
single CLIC drive beamI 3 TeV stage with one drive beam
complex for each beam
156 ns 20 ms
0.5 ns
CLIC: trains at 50 Hz, 1 train = 312 bunches(c)FT
TA
BC2
delay loop2.5 km
decelerator, 25 sectors of 878 m
540 klystrons20 MW, 148 µs
CR2
CR1
circumferencesdelay loop 73 mCR1 293 mCR2 439 m
BDS2.75 km
IPTA
BC2
delay loop2.5 km
540 klystrons20 MW, 148 µs
drive beam accelerator2.4 GeV, 1.0 GHz
CR2
CR1
BDS2.75 km
50 kmCR combiner ringTA turnaroundDR damping ringPDR predamping ringBC bunch compressorBDS beam delivery systemIP interaction point dump
drive beam accelerator2.4 GeV, 1.0 GHz
Drive Beam
Main Beambooster linac2.86 to 9 GeV
e+ main linace– main linac, 12 GHz, 72/100 MV/m, 21 km
e+ injector2.86 GeV
e+ PDR
389 m
e+ DR
427 me– injector
2.86 GeV
e– DR
427 m
BC1
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 4 / 29
Higgs Processes
I Higgsstrahlung dominates at smaller centre-of-massenergy: ∝ 1/s
I chose working point at√
s = 380 GeVI Trade-off between cross-section, luminosity, and jet
topology, more-boosted jets simplify separationI Can also do top physics at this energy
I WW-fusion dominates at larger energies: ∝ log(s)I Rarer decays more available at higher energyI Triple Higgs coupling in HHνeνe benefits from
highest energyI All studies summarised in a comprehensive paper [1]
[GeV]s0 1000 2000 3000
HX
) [fb
]→ - e+
(eσ
2−10
1−10
1
10
210
eνeνH
-e+He
ZH
ZHH
Htt
eνeνHH
Z
e−
e+
H
Z
W
W
e−
e+
νe
H
νe
W
W
H
e−
e+
νe
H
H
νe
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 5 / 29
Top Quark Studies
I 350 GeV and 380 GeVI Threshold scan around 350 GeVI Top-quark mass from radiative eventsI Flavour-changing neutral current top-quark decaysI Direct reconstruction of the top quark
I 1.4 TeV and 3 TeVI Vector boson fusion production of top pairsI Top Yukawa coupling
I Kinematic studies of top-pair production at all stagesI Summarised in comprehensive paper [2]
[GeV]s0 1000 2000 3000
(+X
)) [f
b]t t
→ - e+(eσ
1−10
1
10
210
310
tt
eνeνtt Htt
Ztt
Z∗/ γ∗
e−
e+
t
t
Z∗/ γ∗
e−
e+
t
H
t
W−∗W+∗
e−
e+
νe
t
t
νe
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 6 / 29
Luminosity and Beam-Beam Effects
I Large luminosities require high bunchcharge N and small beams σx/y/z (giventhe other constraints from the accelerator)
L ∝ N2
σx σy
I Leads to large electromagnetic fieldsduring bunch crossing B ∝ γN
σz (σx+σy )
I Use flat beams σy � σx
Par. Unit 380 GeV 3 TeV
N 5.2 ·109 3.72 ·109
σx nm ≈ 149 ≈ 45σy nm ≈ 2.9 ≈ 1σz µm 70 44L 1/cm2s1 1.5 ·1034 5.9 ·1034
L0.01 1/cm2s1 0.9 ·1034 2.0 ·1034
I The bunch particles are strongly deflectedby the fields and radiate Beamstrahlung
0 0.2 0.4 0.6 0.8 1s/s'=sx
1−10
1
10
210
sdxdN
N1
380 GeV
3 TeV
√s′/√
s 380 GeV 3 TeV
> 0.99 58% 36%> 0.90 87% 57%> 0.50 99.96% 88.6%
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 7 / 29
Beam-induced Backgrounds
I Beamstrahlung photonscollide with beam particles or other photons
I Incoherent e+e− pairsI qq pairs in γγ→ Hadron events
I Backgrounds strongly depend oncentre-of-mass energy
I Incoherent pairs have largestconcentration at small angles, and smalltransverse momentum
I Detector acceptance starts at 10 mrad,limited by coherent pairs
[rad]θ4−10 3−10 2−10 1−10 1
per
BX
θ d
N/d
4−10
2−10
1
210
410
610
810
e+
Incoherent e
hadrons→ γγ
CLICdp
380 GeV
>0 MeVT
p
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 8 / 29
Beam-induced Backgrounds
I Beamstrahlung photonscollide with beam particles or other photons
I Incoherent e+e− pairsI qq pairs in γγ→ Hadron events
I Backgrounds strongly depend oncentre-of-mass energy
I Incoherent pairs have largestconcentration at small angles, and smalltransverse momentum
I Detector acceptance starts at 10 mrad,limited by coherent pairs
[rad]θ
4−10 3−10 2−10 1−10 1
per
BX
θdN
/d
4−10
2−10
1
210
410
610
810
e+
Incoherent e
hadrons→ γγ
CLICdp
3 TeV
>0 MeVT
p
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 8 / 29
Beam-induced Backgrounds
I Beamstrahlung photonscollide with beam particles or other photons
I Incoherent e+e− pairsI qq pairs in γγ→ Hadron events
I Backgrounds strongly depend oncentre-of-mass energy
I Incoherent pairs have largestconcentration at small angles, and smalltransverse momentum
I Detector acceptance starts at 10 mrad,limited by coherent pairs
4−10 3−10 2−10 1−10 1 [rad]θ
4−10
2−10
1
210
410
610
810
per
BX
θdN
/d
e
+Incoherent e
hadrons→ γγ
CLICdp
3 TeV
>20MeVT
p
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 8 / 29
CLIC use of the Linear Collider Software
I Detector Model described with DD4HEP
I Event data model and persistency: LCIOI Reconstruction currently using iLCSoft
with the Marlin framework1. γγ→ hadron background overlay2. Digitisation applying sensor
resolutions(tracker), calibration factors(calorimeter)
3. ConformalTracking pattern recognitionand iLCSoft::KalmanFilter
4. Particle flow reconstruction:PANDORAPFA
5. Vertexing and Flavour Tagging: LCFIPlus6. Jet clustering with FastJet and
FastJetContrib7. Very Forward Calorimeter Reconstruction
with FCalClusterer
Generator
Detector Geometry: lcgeo (DD4hep)
AnalysisRecon-struction
Simulation
C++, Python OverlayDigitizationTracking
PFA
VertexingJet ClusteringFlavor Tagging
Persistency FrameworkEvent Data Model: LCIO
Whizard,Pythia, ...
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 9 / 29
Detector for CLIC
General purpose detector for Particle Flow reconstruction [3]
I Steel–Scintillator HCalwith 3 cm cell-size
I Silicon–Tungsten ECalwith 5 mm cell-size
I Silicon Tracker, mostly50 µm pitch strips
I Vertex Detector with25 µm pixels
6 m
I SuperconductingSolenoid of 4 T
I Iron Yoke with RPCs forMuon ID
I End-coilsI Forward calorimeters
for EM coverage downto 10 mrad
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 10 / 29
Track Reconstruction
I Full silicon tracking due to timing andoccupancy
I Momentum resolution of 2×10−5/GeV forcentral high momentum tracksI Needed for, e.g., slepton
measurements [4], Higgs to muons [1]0 0.5 1 1.5 2
0
0.5
1
1.5
z[m]
x[m]
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 11 / 29
Track Reconstruction
I Full silicon tracking due to timing andoccupancy
I Momentum resolution of 2×10−5/GeV forcentral high momentum tracksI Needed for, e.g., slepton
measurements [4], Higgs to muons [1]
[GeV]µp0 500 1000 1500 2000
dN/d
p
0
20
40
60no smearing
-510×=42T
/pT
pσ-510×=82
T/p
Tpσ
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 11 / 29
Track Reconstruction
I Full silicon tracking due to timing andoccupancy
I Momentum resolution of 2×10−5/GeV forcentral high momentum tracksI Needed for, e.g., slepton
measurements [4], Higgs to muons [1]
) [GeV]µµm(110 120 130 140
Eve
nts
/ 0.5
GeV
0
20
40
60
80 simulated databackground fitsignal + background fit
-µ+µ →; HννH = 3 TeVsCLICdp
) = -80%-
P(e
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 11 / 29
Pattern Recognition: Conformal Tracking
I Global pattern recognition for all silicontracking layers
I Conformal mapping to turn circle fittinginto straight line fittingI u = x
x2+y2 v = yx2+y2
I At least for prompt tracks
I Publication:https://doi.org/10.1016/j.nima.2019.163304[5]
x [mm]1500− 1000− 500− 0 500 1000 1500
y [
mm
]
1500−
1000−
500−
0
500
1000
1500CLICdp
µSingle
prompt, p = 100 GeV
prompt, p = 400 MeV
nonprompt, p = 100 GeV
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 12 / 29
Pattern Recognition: Conformal Tracking
I Global pattern recognition for all silicontracking layers
I Conformal mapping to turn circle fitting intostraight line fittingI u = x
x2+y2 v = yx2+y2
I At least for prompt tracks
I Publication:https://doi.org/10.1016/j.nima.2019.163304[5]
u [1/mm]0.04− 0.02− 0 0.02 0.04
v [
1/m
m]
0.04−
0.02−
0
0.02
0.04
CLICdp
µSingle
prompt, p = 100 GeV
prompt, p = 400 MeV
nonprompt, p = 100 GeV
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 12 / 29
Tracking Efficiency
I Single Muon efficiency >99.8% above200 MeV and θ ≥ 10◦
I High efficiency for displaced tracks, untilthey no longer leave enough hitsI Can be improved with adapted
reconstruction parameters
I Good efficiency for particles in jets, alsowhen including γγ→ hadron backgrounds
[GeV]T
p
1−10 1 10 210
Tra
ckin
g e
ffic
ien
cy
0.96
0.97
0.98
0.99
1
1.01CLICdp
µSingle
(forward)° = 10θ
(transition)° = 30θ
(barrel)° = 89θ
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 13 / 29
Tracking Efficiency
I Single Muon efficiency >99.8% above200 MeV and θ ≥ 10◦
I High efficiency for displaced tracks,until they no longer leave enough hitsI Can be improved with adapted
reconstruction parameters
I Good efficiency for particles in jets, alsowhen including γγ→ hadron backgrounds
vertex R [mm]
0 100 200 300 400 500
Tra
ckin
g e
ffic
ien
cy
0
0.2
0.4
0.6
0.8
1
1.2 µDisplaced single
° < 100φ, θ< °0 < y < 600 mm, 80
p = 1 GeV
p = 10 GeV
p = 100 GeV
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 13 / 29
Tracking Efficiency
I Single Muon efficiency >99.8% above200 MeV and θ ≥ 10◦
I High efficiency for displaced tracks, untilthey no longer leave enough hitsI Can be improved with adapted
reconstruction parameters
I Good efficiency for particles in jets,also when including γγ→ hadronbackgrounds
[GeV]T
p
1−10 1 10 210
Tra
ckin
g e
ffic
ien
cy
0.7
0.8
0.9
1
> 0.02 radMC
∆, vertex R < 50 mm, ° < 170θ < °10
= 3 TeVCM
, Ett
No background
hadrons background→γγ3 TeV
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 13 / 29
Momentum Resolution
I Reaching required momentum resolutionfor central high pT tracks [6]
I Small dependency on single pointresolution in the vertex detectorI More important for impact parameter, see
next slide
p [GeV]1 10 210
]1
) [G
eV
T,tru
e
2/p
Tp
∆(σ
5−10
4−10
3−10
2−10
1−10µSingle
= 10 degθ
= 30 degθ
= 89 degθ
mµ = 10 deg, VTX single point res 5θ
mµ = 30 deg, VTX single point res 5θ
mµ = 89 deg, VTX single point res 5θ
mµ = 10 deg, VTX single point res 7θ
mµ = 30 deg, VTX single point res 7θ
mµ = 89 deg, VTX single point res 7θ
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 14 / 29
Vertex Detector
I Silicon vertex detector: precise vertexreconstruction
I Double layers (0.2%X0 per detection layer)I Rin = 31 mmI Spiral geometry in endcaps for air cooling
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 15 / 29
VXD: Single Point Resolution
Transverse and longitudinal impact parameter resolutions for different single point resolutions ofthe vertex detector [6]
]° [θ20 40 60 80
m]
µ)
[0
d∆(
σ
1
10
210
310
CLICdp
µSingle m (default)µ = 3
VTXσp = 1 GeV,
m (default)µ = 3VTX
σp = 10 GeV, m (default)µ = 3
VTXσp = 100 GeV,
mµ = 5VTX
σp = 1 GeV, mµ = 5
VTXσp = 10 GeV,
mµ = 5VTX
σp = 100 GeV, mµ = 7
VTXσp = 1 GeV,
mµ = 7VTX
σp = 10 GeV, mµ = 7
VTXσp = 100 GeV,
]° [θ20 40 60 80
m]
µ)
[0z
∆(σ
1
10
210
310
410
510CLICdp
µSingle m (default)µ = 3
VTXσp = 1 GeV,
m (default)µ = 3VTX
σp = 10 GeV, m (default)µ = 3
VTXσp = 100 GeV,
mµ = 5VTX
σp = 1 GeV, mµ = 5
VTXσp = 10 GeV,
mµ = 5VTX
σp = 100 GeV, mµ = 7
VTXσp = 1 GeV,
mµ = 7VTX
σp = 10 GeV, mµ = 7
VTXσp = 100 GeV,
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 16 / 29
Calorimeters and Particle Flow Clustering
I Require excellent jet energy resolution,separation of jets from W’s or Z’s [4]
I Particle flow clustering, separate clustersfrom neutral and charged particles
I Fine grained calorimeters
Mass [GeV]60 70 80 90 100 110 120
Arb
itrar
y U
nits
0
2
4
6
/m = 1%mσ/m = 2.5%mσ/m = 5%mσ/m = 10%mσ
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 17 / 29
Electromagnetic Calorimeter
I Depth and sampling fraction (40 layers,22 X0) for high energy EM objectreconstruction [3]I Further optimisation of layer structure
possible, Silicon ECal is a cost driver
I High granularity (5×5 mm2) for good jetenergy resolution [3]I studied with ILD detector model
[GeV]γtrueE
0 500 1000 1500
)/E
[%]
HC
al+
EE
Cal
(Eσ
0
1
2
3
4
5
6
CLICdet_17_8
CLICdet_20_10
CLICdet_30
CLICdet_40_b
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 18 / 29
Electromagnetic Calorimeter
I Depth and sampling fraction (40 layers, 22X0) for high energy EM objectreconstruction [3]I Further optimisation of layer structure
possible, Silicon ECal is a cost driver
I High granularity (5×5 mm2) for good jetenergy resolution [3]I studied with ILD detector model
ECAL Cell Size [mm]0 5 10 15 20 25
) [%
]j
(E90
) / M
ean
j(E
90R
MS
0
1
2
3
4
5
45 GeV Jets
100 GeV Jets
180 GeV Jets
250 GeV Jets
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 18 / 29
Electromagnetic Calorimeter
I Depth and sampling fraction (40 layers, 22X0) for high energy EM objectreconstruction [3]I Further optimisation of layer structure
possible, Silicon ECal is a cost driver
I High granularity (5×5 mm2) for good jetenergy resolution [3]I studied with ILD detector model
nLayers15 20 25 30
) [%
]j
(E90
) / M
ean
j(E
90R
MS
0
1
2
3
4
5
45 GeV Jets
100 GeV Jets
180 GeV Jets
250 GeV Jets
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 18 / 29
Hadronic Calorimeter
I Jet energy resolution with differentHCal depths [4]
I Need 7.5λI to contain highest energy jets
's in CLIC HCALIλNumber of 4 6 8 10
/E [%
]Eσ
3
4
5
6 uds, jet energy:→Z45.5 GeV100 GeV250 GeV500 GeV1 TeV1.5 TeV
0.7≤ θcos
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 19 / 29
Jet Energy Resolution
I Reaching 3.5% jet energy resolution forhigh energy jets in the barrel [6]
I Endcap region more affected byγγ→ hadron backgrounds, which areforward peaked
|θ|cos0 0.2 0.4 0.6 0.8 1
)[%
]G j
/ER j
(E90
)/M
ean
G j/E
R j(E
90
RM
S
2
4
6
8
10
12
14 VLC7 Jets 50 GeV≈
100 GeV≈
250 GeV≈
750 GeV≈
1500 GeV≈
CLICdp
3.5%
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 20 / 29
Jet Energy Resolution
I Reaching 3.5% jet energy resolution forhigh energy jets in the barrel [6]
I Endcap region more affected byγγ→ hadron backgrounds, which areforward peaked
|θ|cos0 0.2 0.4 0.6 0.8 1
)[%
]G j
/ER j
(E90
)/M
ean
G j/E
R j(E
90
RM
S
2
4
6
8
10
12
14 VLC7 Jets, with 3TeV BG 50 GeV≈
100 GeV≈
250 GeV≈
750 GeV≈
1500 GeV≈
CLICdp
3.5%
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 20 / 29
W/Z-Separation
I Jet energy resolution good enough for≈ 2σ separation between jets from W andZ bosons [6]
I For different boson energies and includingbackgrounds
Background EW,Z σm(W)/m(W) σm(Z)/m(Z) ε Separation
[GeV] [%] [%] [%] [σ ]
no BG
125 5.5 5.3 88 2.3250 5.3 5.4 88 2.3500 5.1 4.9 90 2.51000 6.6 6.2 84 2.0
3 TeV BG
125 7.8 7.1 80 1.7250 6.9 6.8 82 1.8500 6.2 6.1 85 2.01000 7.9 7.2 80 1.7
380 GeV BG 125 6.0 5.5 87 2.2 dijet mass [GeV]60 80 100 120
A.U
.
0
200
400
600
800500 GeV bosons
W bosonsZ bosons
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 21 / 29
W/Z-Separation
I Jet energy resolution good enough for≈ 2σ separation between jets from W andZ bosons [6]
I For different boson energies and includingbackgrounds
Background EW,Z σm(W)/m(W) σm(Z)/m(Z) ε Separation
[GeV] [%] [%] [%] [σ ]
no BG
125 5.5 5.3 88 2.3250 5.3 5.4 88 2.3500 5.1 4.9 90 2.51000 6.6 6.2 84 2.0
3 TeV BG
125 7.8 7.1 80 1.7250 6.9 6.8 82 1.8500 6.2 6.1 85 2.01000 7.9 7.2 80 1.7
380 GeV BG 125 6.0 5.5 87 2.2 dijet mass [GeV]60 80 100 120
A.U
.
0
200
400
600500 GeV bosons, with 3 TeV BG
W bosonsZ bosons
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 21 / 29
Forward Calorimeters
I Integrated Luminosity measurements withLumiCal: require excellent polar angleresolution 20 µrad
I BeamCal complementing EM acceptancedown to 10 mrad
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 22 / 29
LumiCal Performance
I LumiCal reaching desired polar angleresolution for highest energyelectrons [6]
I Good reconstruction efficiencyI Low fake rate
0 500 1000 1500Energy [GeV]
0
50
100
150
200
rad
]µ
[θ
σ
LumiCal, 3 TeV, 40 BX < 75 mradθ50 mrad <
Polar Angle Resolution
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 23 / 29
LumiCal Performance
I LumiCal reaching desired polar angleresolution for highest energy electrons [6]
I Good reconstruction efficiencyI Low fake rate
40 60 80 100 120 [mrad]θ
0
0.5
1
1.5
Eff
icie
ncy
σLumiCal, 3 TeV, 40BX, 101500 GeV Electrons190 GeV Electrons100 GeV Electrons50 GeV Electrons10 GeV Electrons
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 23 / 29
LumiCal Performance
I LumiCal reaching desired polar angleresolution for highest energy electrons [6]
I Good reconstruction efficiencyI Low fake rate
50 100 150 [mrad]θ
4−10
3−10
2−10
1−10
Fa
ke
ra
te σLumiCal, 10Bkg: 3 TeV, 40BX
0 GeV < E < 10 GeV 10 GeV < E < 25 GeV 25 GeV < E
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 23 / 29
BeamCal Performance
I BeamCal dominated by incoherent pairbackgrounds
I Strong rejection of energy from thebackgrounds leads to lowerefficiency [6]
I Given near zero fake rate could tuneselection for slightly better efficiency
I Resolutions much worse than in LumiCal10 20 30 40 50
[mrad]θ
0
0.5
1
1.5
Eff
icie
ncy
σBeamCal, 3 TeV, 40BX, 31500 GeV Electrons1000 GeV Electrons500 GeV Electrons250 GeV Electrons
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 24 / 29
BeamCal Performance
I BeamCal dominated by incoherent pairbackgrounds
I Strong rejection of energy from thebackgrounds leads to lower efficiency [6]
I Given near zero fake rate could tuneselection for slightly better efficiency
I Resolutions much worse than in LumiCal
10 20 30 40 50 [mrad]θ
4−10
3−10
2−10
Fa
ke
ra
te σBeamCal, 3Bkg: 3 TeV, 40BX
0 GeV < E
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 24 / 29
BeamCal Performance
I BeamCal dominated by incoherent pairbackgrounds
I Strong rejection of energy from thebackgrounds leads to lower efficiency [6]
I Given near zero fake rate could tuneselection for slightly better efficiency
I Resolutions much worse than in LumiCal
400 600 800 1000 1200 1400 1600Energy [GeV]
0
0.1
0.2
0.3
0.4
0.5
) [m
rad
]θ
RM
S(
BeamCal, 3 TeV, 40 BX < 40 mradθ15 mrad <
Polar Angle Resolution
CLICdp
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 24 / 29
γγ→ hadron Background Mitigation
I Read out full bunch train and identify time of physicsevent
I Select hits around the event using the timeresolution of the sub-detectors
I Calculate truncated mean of hittimes and correct for time-of-flight
I Accept reconstructed particles depending on particletype, cluster time, and transverse momentum
I Selection cuts reduce background from 1.2 TeV to100 GeV.
I Further background reduction through jet-clustering
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 25 / 29
γγ→ hadron Background Mitigation
I Read out full bunch train and identify time of physicsevent
I Select hits around the event using the timeresolution of the sub-detectors
I Calculate truncated mean of hittimes and correct for time-of-flight
I Accept reconstructed particles depending on particletype, cluster time, and transverse momentum
I Selection cuts reduce background from 1.2 TeVto 100 GeV.
I Further background reduction through jet-clustering e−e+→ HH with γγ→ hadronbackground overlaid before and after
timing selection cuts.
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 25 / 29
Flavour Tagging
I Focus here on relative performance of differentvertex resolutions
I Optimising flavour tagging performance still work inprogressI Improvements in trackingI Tune machine learning parametersI Reject vertices from material interaction
Mis
identification e
ff.
3−10
2−10
1−10
1
Beauty contaminationmµ3 mµ5 mµ7
LF contaminationmµ3 mµ5 mµ7
CLICdp
° < 90θ < ° = 500 GeV, 20CM
Dijet events, E
Charm eff.0.5 0.6 0.7 0.8 0.9 1
σm
/ o
the
r µ
3
0.2
0.4
0.6
0.8
1
1.2 Beauty contamination
LF contamination
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 26 / 29
Flavour Tagging
I Focus here on relative performance of differentvertex resolutions
I Optimising flavour tagging performance still work inprogressI Improvements in trackingI Tune machine learning parametersI Reject vertices from material interaction
Mis
identification e
ff.
3−10
2−10
1−10
1
Charm contaminationmµ3 mµ5 mµ7
LF contaminationmµ3 mµ5 mµ7
CLICdp
° < 90θ < ° = 500 GeV, 20CM
Dijet events, E
Beauty eff.0.5 0.6 0.7 0.8 0.9 1
σm
/ o
the
r µ
3
0.4
0.6
0.8
1
1.2 Charm contamination
LF contamination
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 26 / 29
Jet Clustering
I γγ→ hadron background and longitudinalboost due to Beamstrahlung make LEP jetalgorithms unsuited for CLIC
I Use hadron collider jet algorithm featuresI Cluster forward particles into beam jetsI Benefit from longitudinal invariance.
Particle distance measure using∆R2 = ∆η2 + ∆φ2
I Specialised VLC jet algorithm [7]I Reconstruction parameters can and have
to be tuned to specific analyses, see thepresentation on the physics studies
144 Page 6 of 16 Eur. Phys. J. C (2018) 78:144
Fig. 3 The area or footprint ofjets reconstructed with R = 0.5with the three major families ofsequential recombinationalgorithms. The two shadedareas in each column correspondto a jet in the central detector(θ = π/2) and to a forward jet(θ = 7π/8). The jet axis isindicated with a cross
(rad.)π/φazimuth−0.5 0 0.5
(rad
.)π/θ
pola
r ang
le
0
0.2
0.4
0.6
0.8
1
/2π = θ
/8π = 7θ
-e+generalized e
1-cos Rijθ1 - cos
)2j
,E2i
= 2 min(Eijd
2i = EiBd
long. invariant
2R2 RΔ)2
Tj,p2
Ti = min(pijd
2Ti
= piBd
=1)γ=β (-e+VLC e
2Rijθ1 - cos
)2j
,E2i
= 2 min(Eijd
2Ti
= piBd
γ0.5− 0 0.5 1 1.5
β
1−
0.5−
0
0.5
1t
gen. k-e+e(~Durham)
t anti-k-e+e
Cambridge
Valencia
VLC-angular
anti-VLC
constant size ** shrinking footprint
hard
& c
oll.
first
**
angu
lar *
* so
ft &
col
l. fir
st
Fig. 4 Diagram of the parameter space spanned by exponents β andγ of the VLC algorithm. On the y-axis generalisations with beam jetsof the LEP/SLD algorithms are found, with the Cambridge algorithmwith angular ordering at the origin and the Durham or kt algorithm atβ = 1. Choosing β = -1 yields reverses the clustering order (like inanti-kt algorithm [38]). Choosing non-zero and positive values for γ
yields robust algorithms with a shrinking jet area in the forward region
slower decrease of the area when the polar angle goes to 0 orπ .
For γ = 0, diB = E2βi and we retrieve the generalised
e+e− algorithms with constant angular opening: the gener-alised Cambridge algorithm [17] for β = 0 and generalised kt
or Durham [18] for β = 1. Choosing β = -1 yields an e+e−variant of the anti-kt algorithm [38]. A schematic overviewof the algorithms in (β, γ ) space is given in Fig. 4.
4 Jet energy corrections
Before we turn to a detailed simulation including overlaidbackgrounds and a model for the detector response, we studythe perturbative and non-perturbative jet energy correctionsof the algorithms. Both types of corrections are closely con-nected to the jet area [39]. In this section we quantify theirimpact, following the analysis of Ref. [39]. This first explo-ration of the stability of the algorithms should be extended infuture work to quantify the impact of next-to-leading cor-rection, as performed for instance in Ref. [40]. Also therobustness of the conclusions for a variety of different setsof parameters (tunes) of the Monte Carlo simulation meritsfurther study.
4.1 Monte Carlo setup
The Monte Carlo simulation chain uses the MadGraph5_aMC@NLO package [23] to generate the matrix elementsof the hard scattering 2 → 2 event. Several processes arestudied, but results in this Section focus on e+e− → qq̄at
√s = 250 GeV and e+e− → t t̄ with fully hadronic top
decays at√s = 3 TeV. The four-vectors of the outgoing
quarks are fed into Pythia 8.180 [24], with the default tuneto LEP data, that performs the simulation of top-quark andW boson decays, the parton shower and hadronisation. Nodetector simulation is performed and initial-state radiationand beam energy spread are not included in the simulation.Particles or partons from the Pythia event record are clusteredusing FastJet 3.0.6 [33] exclusive clustering with N = 2.The default (“E-scheme”) recombination algorithm is usedto merge (pseudo-) jets.
123
Jet areas obtained from different types of jetclustering algorithm
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 27 / 29
Summary
I In last years, studied and documented CLICdet performanceI Re-use of existing components, and developments were needed, allowed detailed studies
to be performedI Detector and software can fulfil the requirements for physics at CLIC
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 28 / 29
References
[1] CLICdp Collaboration. “Higgs physics at the CLIC electron–positron linear collider”. In: Eur. Phys. J. C 77.7(2017). URL: https://arxiv.org/abs/1608.07538.
[2] CLICdp collaboration, H. Abramowicz, et al. “Top-quark physics at the CLIC electron-positron linear collider”. In: J.HEP 2019.11 (Nov. 2019), p. 3. DOI: 10.1007/JHEP11(2019)003. arXiv: 1807.02441.
[3] N. Alipour Tehrani et al. “CLICdet: The post-CDR CLIC detector model”. In: (Mar. 2017). CLICdp-Note-2017-001.URL: https://cds.cern.ch/record/2254048.
[4] L. Linssen et al., eds. Physics and Detectors at CLIC: CLIC Conceptual Design Report. CERN-2012-003,arXiv:1202.5940. CERN, 2012.
[5] E. Brondolin et al. “Conformal tracking for all-silicon trackers at future electron–positron colliders”. In: Nucl. Instr.Meth. A956 (2020), p. 163304. DOI: 10.1016/j.nima.2019.163304.
[6] Dominik Arominski et al. A detector for CLIC: main parameters and performance. 2018. arXiv: 1812.07337[physics.ins-det]. URL: https://cds.cern.ch/record/2649437.
[7] Ignacio Garcia Garcia et al. “Jet reconstruction at high-energy electron–positron colliders”. In: Eur. Phys. J. C 78.2(June 2017), p. 144.
A. Sailer Detector Requirements and Software for CLIC - HK DetSoft, Jan 16, 2020 29 / 29