View
6
Download
0
Category
Preview:
Citation preview
Peter Krieger, University of Toronto WRNPPC 2004 1
ATLAS Calorimetry at the Large Hadron Collider
Peter Krieger, University of Toronto / IPP
• Introduction to Calorimetry / Calorimeter Design• The ATLAS Calorimeters
Peter Krieger, University of Toronto WRNPPC 2004 2
The ATLAS Detector at the Large Hadron Collider
p (7TeV)
p (7TeV)
Peter Krieger, University of Toronto WRNPPC 2004 3
What is a Calorimeter ?
CalorimeterOutput signal to electronics
Incident particle flux
A calorimeter consists of:Dense absorber material to fully absorb incident particles
Active material to produce an output signal proportional to the input energy
Peter Krieger, University of Toronto WRNPPC 2004 4
Sampling Calorimeters
Absorber and active materials can be the same (homogeneous calorimeter: e.g. CsI, NaI, Lead glass) or (more commonly) different
absorber (i.e. Pb) active layer
for example:
• Pb (ATLAS EMB,EMEC)
• Copper (ATLAS HEC)
• Tungsten (ATLAS FCal)
• Depleted Uranium (ZEUS)
for example:
• Scintillator (light output)
• Liquid argon (ionization)
Peter Krieger, University of Toronto WRNPPC 2004 5
EM Energy Loss vs. Energy (here in Pb) – Electromagnetic Showers
Bremsstrahlung
Lead (Z = 82)Positrons
Electrons
Ionization
Moller (e−)
Bhabha (e+)
Positron annihilation
1.0
0.5
0.20
0.15
0.10
0.05
(cm
2g−
1 )
E (MeV)10 10 100 1000
1 E−
dE dx
(X0−1
)
Photon Energy
1 Mb
1 kb
1 b
10 mb10 eV 1 keV 1 MeV 1 GeV 100 GeV
Lead (Z = 82)− experimental σtot
σp.e.
κe
Cros
ssec
tion
(bar
ns/a
tom
)
aσR y el igh
σCom t np o
κn cuelectron/positrons
photons
Critical energy ~ 10 MeVcE
e−
e−
γ
01 X
Peter Krieger, University of Toronto WRNPPC 2004 6
e−
e−
γ
01 X
Absorber Active medium (scintillation or ionization)
Peter Krieger, University of Toronto WRNPPC 2004 7
How does required detector size scale with energy ?
0( )X 2t
Calorimeter (use EM as example)
After t radiation lengths total number of particles present is
Average energy of a shower particle at depth t is
Shower has maximum number of particles when
So, shower maximum appears at
0( ) / 2tE t E=
( ) cE t E=
As we go to higher and higher energies, calorimetry becomes a more critical part of a detector (relative to tracking)
0max
ln( / )ln(2)
cE Et =
For comparison:
Tracking chambers: for same resolution required detector size (for same magnetic field) scales with
/p pδE
Peter Krieger, University of Toronto WRNPPC 2004 8
ATLAS Calorimetry at the Large Hadron Collider
44 m long 22 m high 7000 tons
Peter Krieger, University of Toronto WRNPPC 2004 9
Hadronic Energy Containment in Iron
99%
95%
Single Hadron Energy (GeV)
Dep
th in
Iro
n (
cm)
Dep
th in
Iro
n (
λ I)
5 10 50 100 500 1000 50
100
150
200
3
4
5
6
7
8
9
10
11
12
/ Bock param./ CDHS data/ CCFR data
Peter Krieger, University of Toronto WRNPPC 2004 10
Hadronic Calorimetry
Process Percent ofTotal
Secondary proton ionization 31.6Electromagnetic cascade 21.0Nuclear binding energy plus neutrino energy 20.6Secondary charged pion ionization 8.2Neutrons with E > 10 MeV 4.9Neutrons with E < 10 MeV 3.9Residual nuclear excitation energy 3.7Z > 1 ionization 2.4Primary proton ionization 2.3Other 1.4
λΙ
Principle the same as for electromagnetic calorimeters …… BUT
Hadronic interaction length larger than EM radiation length
Needs to be deeper (denser) than EM calorimeter
More processes contribute to hadronic shower development than to EM
Note that a sizeable fration of the energy deposited in a hadronic shower is electromagnetic (from production and decay of neutral pions )
Some processes do not lead to an observable signal (e.g. neutrinos)
!!!!!
0X
γγπ →0
A calorimeter’s response to electromagnetic particles differs from it’s response to hadronic particles
Average fractional energy deposition for 10 GeVprotons in an Fe-LAr sampling calorimeter
e/h ≠ 1 (for non-compensating calorimeters)
Peter Krieger, University of Toronto WRNPPC 2004 11
Calorimeter Design Criteria
Aspects driven by physics goals
• Signal linearity
• Resolution (energy, position)
Aspect driven by experimental conditions
• Radiation tolerance choice of technology (TileCal vs Liquid Argon)
• Collision frequency speed of response
Practical aspects
• Cost
• Ease of construction
Requirements are dictated by resolution required for desired signal (e.g. Higgs discovery )
Liquid krypton would yield improved resolution, but is more expensive and requires higher purity than liquid argon
Peter Krieger, University of Toronto WRNPPC 2004 12
Energy Resolution of a Sampling Calorimeter
Resolution of a sampling calorimeter typically takes the form:
noise term
a b cE E Eσ= ⊕ ⊕ constant term
sampling term
sampling term constant term noise term• Choice of absorber
• Choice of active material
• Thickness of sampling layers
• …..
• depth of detector
• detector non-uniformities
• cracks
• dead material …….
• electronic noise
• signal pileup0( , )X λΙ
Dominates at high energy Dominates at low energyTypically most important in 10-100 GeV energy range
Peter Krieger, University of Toronto WRNPPC 2004 13
Status of the Search for the Higgs Boson
80.2
80.3
80.4
80.5
80.6
130 150 170 190 210
mH114 300 1000
mt
mW
Preliminary
68% CL
∆α
LEP1, SLD Data
LEP2, pp− Data
0
2
4
6
10020 400
mH [GeV]
∆χ2
Excluded Preliminary
∆αhad =∆α(5)
0.02761±0.00036
0.02747±0.00012
Without NuTeV
theory uncertainty
Discovery of the Higgs bosons is one of the main purposes of the LHCLEP direct search limit > 114 GeV
Current experimental evidence points to a low mass HIggs
Peter Krieger, University of Toronto WRNPPC 2004 14
Higgs Branching Ratio as a Function of Higgs Mass
92
43
43~
125
7
7
8
1010100200
10101)8.0(10101)1(~~1010500108.0800105.15.1101515
/sec/Pr
nbGeVp
QCDjetspbTeVMHpbTeVMgg
bbbpbttnbeeZnbeW
yearEventEventsocess
T
SM
g
>
==
→→
−
−
µ
νσ
BR(H)
bb_
τ+τ−
cc_
gg
WW
ZZ
tt-
γγ Zγ
MH [GeV]50 100 200 500 1000
10-3
10-2
10-1
1
bb
In favoured low mass region decay is almost entirely to bb
QCD background to too large. Need to look for cleaner search channel
0H bb→
Favoured discovery channel for a light mass Higgs is 0H γγ→
Peter Krieger, University of Toronto WRNPPC 2004 15
Electromagnetic Energy Resolution
1
10
10 2
102
103
mH (GeV)
Sig
nal s
igni
fica
nce
H → γ γ + WH, ttH (H → γ γ ) ttH (H → bb) H → ZZ(*) → 4 l
H → ZZ → llνν H → WW → lνjj
H → WW(*) → lνlν
Total significance
5 σ
∫ L dt = 100 fb-1
(no K-factors)
ATLAS
10000
12500
15000
17500
20000
105 120 135
mγγ (GeV)
Eve
nts
/ 2 G
eV
0
500
1000
1500
105 120 135
mγγ (GeV)
Sign
al-b
ackg
roun
d, e
vent
s / 2
GeV
0H γγ→
Higgs discovery potential in the favored low-mass region relies on reconstruction of the final state
Required resolution on the reconstruction represents the most stringent constraint on ATLAS electromagnetic calorimetry
Small signal on significant background. The better the signal resolution, the better the signal to noise ratio
Require good energy resolution AND good position resolution
0H γγ→
Peter Krieger, University of Toronto WRNPPC 2004 16
ATLAS Calorimetric Coverage
|η| < 1.7|η| < 3.2
1.5 < |η| < 3.23.1 < |η| < 4.9
ln(tan )θη = −2
Total mass ~ 4000 tons
Peter Krieger, University of Toronto WRNPPC 2004 17
The ATLAS Liquid Argon and Tile Calorimeters
Tile barrel Tile extended barrel
LAr forward calorimeter (FCAL)
LAr hadronic end-cap (HEC)
LAr EM end-cap (EMEC)
LAr EM barrel
Peter Krieger, University of Toronto WRNPPC 2004 18
The ATLAS Tile Calorimeter
PMT
Plastic scintillatorinside steel absorberstructure
WLS fiber
“Conventional” Iron-Scintillating Tile Sampling Calorimeter, but with unconventional mechanical structure
Peter Krieger, University of Toronto WRNPPC 2004 19
Tile Calorimeter Construction
Iron cutting
sub-module assembly
tile manufacturing
tile insertion
Peter Krieger, University of Toronto WRNPPC 2004 20
One TileCal Barrel Module
Peter Krieger, University of Toronto WRNPPC 2004 21
TileCal Modules Before Assembly
Peter Krieger, University of Toronto WRNPPC 2004 22
TileCal Barrel Preassembly (April 2003)
Final assembly of TileCal detector is done in the ATLAS Cavern
Peter Krieger, University of Toronto WRNPPC 2004 23
ATLAS Electromagnetic Barrel Calorimeter
Detector design dictated by physics goals:
e.g.
Accordion structure chosen to ensure azimuthal uniformity (no cracks)
Liquid argon chosen for radiation hardness and speed
0 0, 4 ,H H ZZ e W e eeγγ ν′ ′→ → → → , Ζ →
|η| < 1.475
Copper/kaptonelectrode
Honeycomb spacer to maintain LAr gap
Stainless-steel-clad Pb absorber plates
Peter Krieger, University of Toronto WRNPPC 2004 24
LAr Gap Structure in Accordion EMB Calorimeter
Particle from interaction point
Peter Krieger, University of Toronto WRNPPC 2004 25
ATLAS EMB Calorimeter (Half Barrel}
Half Barrel Assembly
Note η segmentation of readout
Peter Krieger, University of Toronto WRNPPC 2004 26
Readout segmentation of accordian electrodes (in η)
Peter Krieger, University of Toronto WRNPPC 2004 27
Readout segmentation of accordian electrodes (in η)
Finer segmentation in first layers for better position resolution
Peter Krieger, University of Toronto WRNPPC 2004 28
Stacked Electromagnetic Barrel Calorimeter Module
Peter Krieger, University of Toronto WRNPPC 2004 29
Installation of EMB Half-Barrel Module
Peter Krieger, University of Toronto WRNPPC 2004 30
EMB Half-Barrel Wheel on Insertion Stand
Peter Krieger, University of Toronto WRNPPC 2004 31
ATLAS EBM Calorimeter Installed in Barrel Cryostat
Peter Krieger, University of Toronto WRNPPC 2004 32
The ATLAS Endcap Calorimeter
Electomagetic Endcap Calorimeter (EMEC)
Hadronic Endcap Calorimeter (2 Wheels)
Forward Calorimeter
Peter Krieger, University of Toronto WRNPPC 2004 33
Electomagnetic Endcap Calorimeter (EMEC)
1.375 < |η| < 3.2electrode folding outer wheel
electrode folding inner wheel
Accordion structure chose here as well. Design is more complicated as folding amplitude is a function of radius
Constructed as an inner and outer wheel
″spanish fan ″ geometry
Peter Krieger, University of Toronto WRNPPC 2004 34
Construction of 1/8 Sector of EMEC
Peter Krieger, University of Toronto WRNPPC 2004 35
Construction of an EMEC Wheel
Inner and outer wheels visible
Peter Krieger, University of Toronto WRNPPC 2004 36
Endcap C EMEC Installed and Cabled
Peter Krieger, University of Toronto WRNPPC 2004 37
ATLAS Hadronic Endcap Calorimter (HEC)
LAr-Cu sampling calorimeter covering 1.5 < η < 3.2
HEC Module Structure
Back Wheel
Front Wheel
Composed of 2 wheels per end, 32 modules per wheel
Peter Krieger, University of Toronto WRNPPC 2004 38
HEC Module
Peter Krieger, University of Toronto WRNPPC 2004 39
HEC Readout Granularity
Peter Krieger, University of Toronto WRNPPC 2004 40
HEC Readout Electrode Structure
• The read-out structure is electrostatictransformer:– Small gap avoids ion build-up– Same behaviour as a 4 mm gap with
lower HV (2 kV instead of 4 kV)
• Robust against H.V. shorts
Peter Krieger, University of Toronto WRNPPC 2004 41
HEC Module Assembly into HEC Wheel
Peter Krieger, University of Toronto WRNPPC 2004 42
Rotation of HEC Wheel to Vertical (HEC Rotator)
Peter Krieger, University of Toronto WRNPPC 2004 43
HEC Storage and Insertion into Endcap Cryostat
Peter Krieger, University of Toronto WRNPPC 2004 44
Endcap C Hadronic Endcap Calorimeter Installed and Cabled
Peter Krieger, University of Toronto WRNPPC 2004 45
ATLAS Forward Calorimeter
Provides calorimetric hermiticity: covers 3.1 < η < 4.9
Important for determination of missing transverse energy (signature of supersymmetry and other possible new physics scenarios).
Require liquid argon calorimeter with very thin gaps
Novel detector design – tubular electrodes in an absorber matrix
• one copper module (copper electrodes, copper absorber)
• two hadronic modules (tungsten electrodes, tungsten absorber)
• one uninstrumented brass plug (prevents punch through to muon system)
Peter Krieger, University of Toronto WRNPPC 2004 46
Forward Calorimeter Design Concept
Tubular electrode structure allows for very narrow LAr gaps:
• FCal1: 250 µm ~ 12K electrodes
• FCal2: 375 µm ~ 10K electrodes
• FCal3: 500 µm ~ 8K electrodes
Absorber formed by material matrix containing the electrodes as well as by the electrode rods
• FCal1 matrix: 18 copper plates
• FCal2 matrix: tungsten slugs
• FCal3 matrix: tungsten slugs
Peter Krieger, University of Toronto WRNPPC 2004 47
FCal1 Structure / Assembly
Insertion of FCal1 electrode tube
Peter Krieger, University of Toronto WRNPPC 2004 48
FCal2/3 Structure and Assembly
Endplate and inner absorber assembly
During matrix stacking (tubes and slugs)
Peter Krieger, University of Toronto WRNPPC 2004 49
Matrix Stacking of Hadronic FCal Module
Peter Krieger, University of Toronto WRNPPC 2004 50
FCal2/3 Structure and Assembled Module
Liquid Argon Gap
Tungsten Rod
Peter Krieger, University of Toronto WRNPPC 2004 51
Forward Calorimeter Modules and Final Assembly
FCal1 FCal2 FCal3
First Integrated FCal ready for delivery to ATLAS
Final assembly of the second FCal will begin this spring. Modules are completed
Peter Krieger, University of Toronto WRNPPC 2004 52
The ATLAS Endcap Calorimeters
Endcap Cryostat Forward Calorimeter Support Tube
Peter Krieger, University of Toronto WRNPPC 2004 53
Calorimeter TestbeamsBeam tests of calorimeter prototypes and production module are a critical part of calorimeter development and construction.
EMEC, HEC testbeams are covered in next two talks
Still to come (summer 2004)
• HEC/EMEC/FCal Combined Testbeam 2004
• ATLAS Barrel Testbeam (Full slice through ATLAS barrel, including tracking)
Ec
Ebaσ/E ⊕⊕=
a=(3.76±0.06)%, b=(24.5±0.84)√GeV %, c=(145.5±1.6) GeV%
FCal electron resolution
From the 2003 FCalTestbeam
Exceeds design goal
Recommended