Calorimetry at the Linear Collider

1
Calorimetry at the Linear Collider
Colloquium at Fermilab 9 March 2005
José Repond Argonne National Laboratory
2
Calorimeter
calor heat (Latin)
metron measure (Greek)
3
Calorimetry
Historical milestones for particle physics
Based on K.Pretzls CALOR02 review talk
1930
First calorimetric measurement Mean
energy of continuous ß spectrum from 210Bi
L. Meitner and W. Orthmann Zeitschrift für Physik
60 (1930) 143
Telescope counters
Hodoscopes
Ionization chambers
Absorber (Iron)
1954
First sandwich calorimeter
Measure cosmic rays with E gt 1014 eV N.L.
Grigorov et al. Zh.Exsp.Teor.Fiz. 34(1954) 506
4
Calorimetry
First total absorption calorimeter
Using large NaI(Tl) or CsI Crystals for p0
spectroscopy E.B.Hughes et al., IEEENS 17
(1970) 14
1968
First hadron calorimeter GARGAMELLE (bubble
chamber) at CERN with 5 ?I Discovery of
neutral currents
1970
Mark II
1980s
First 4p calorimeters at
colliders SPEAR, PETRA, PEP, SppS
5
Calorimetry
1982
First compensating calorimeter with e/h
1 Axial field spectrometer at the ISR
H.Gordon et al., NIM 196 (1982) 303
1990
First application of
Energy Flow Algorithms
ALEPH detector searching for Higgs
Limits on Higgs coupling
Now Particle Flow Algorithms
6
Outline
I Set the stage The Linear Collider Project II
Motivate RD for calorimetry Particle Flow
Algorithms III Developments in
calorimetry Silicon-Tungsten Electromagnetic
Calorimeters Digital Hadron Calorimeters IV
Timescales Testbeam program V Conclusions
7
TESLA GLOBAL NEXT FUTURE
International
The Linear Collider
Baseline Machine ECM of operation 200 500
GeV Luminosity and reliability for 500 fb-1 in
4 years Energy scan capability with lt10
downtime Beam energy precision and stability
below 0.1 Electron polarization of gt80 Two
interaction regions desirably with detectors
ECM down to 90 GeV for calibration
Upgrades ECM about 1 TeV Capability of
running at any ECM lt 1 TeV L and reliability
for 1 ab-1 in 3 4 years
Options Extend to 1 ab-1 at 500 GeV in 2
years e-e-, ??, e-? operation e polarization
50 Giga-Z with L several 1033 cm-2s-1 WW
threshold scan with L 1033 cm-2s-1
As defined in
International Scope Document
See www.fnal.gov/directorate/icfa/LC_parameters.pd
f
8
The Linear Collider
Technology
Traditionally Circular electron
machines Synchrotron Radiation as
dE 8.85 10-5 E4/? MeV/turn

with E beam energy in GeV

? Radius of machine in km Assume E
500 GeV ? 4.3 km (LEP)
Assume E 500 GeV W
2.5 GeV/27km (LEPII)
LEP
W 1.3 TeV/turn
? 100 km
Warm cavities Cold cavities
Linear Machine
9
The Linear Collider
Cavities
Warm
Cold
US - Japan
TESLA
Frequency of RF system L-Band (1.3 GHz)
Operating at superconducting temperature
Status Technology demonstrated All
subsystems prototyped Industrial production
assured High gradients (gt40 MeV/m) achieved
Reliable operation of structures
RF system X-Band (11.4 GHz) Operating at
room temperature Status Technology
demonstrated All subsystems prototyped High
gradients (65 MeV/m) achieved Reliable
operation of structures
10
The Linear Collider
Technology Decision
Two mature technologies available Resources for
RD limited Pursuit of two technologies waste of
resources
Need to make a choice
International Technology Recommendation Panel
Recommendation announced at ICHEP04 (Beijing)
While both technologies can achieve the goals,
the SC technology has features that tipped the
balance in its favor
Main arguments
Large cavity aperture and long bunch interval
more robust Lower risk with main LINAC and RF
system Construction of EU XFEL (approved) with
same technology Industrialization of most LINAC
components underway Reduced power consumption
11
The Linear Collider
The next steps
Creation of a Involving all 3 regions
(N America, EU, Asia) Director
search by 6 member committee ? 13 nominations
? 3 candidates ? 1 director (by now) Find a
9 proposals KEK
DESY, RAL/Daresbury
LBL, BNL, Cornell, FNAL, SLAC,
Triumph Evaluation by 6 member
committee
Global Design Initiative
Host for GDI Central Team
12
The Linear Collider
Z Boson line shape
The Physics
The Standard Model of Particle Physics -
Unchallenged for decades - Describes all
experimental data - Is known to be incomplete
EW interactions have symmetry SU(2)xU(1)
This symmetry is spontaneously broken SM
does not explain this
New Physics - Expected at the 100 GeV mass
scale Higgs(es), SUSY, extra dimensions
13
The Linear Collider
LHC LC Complementarity
Discovery potential greater at LHC (vs 14
TeV) Reaching higher mass scales
Interpretation of signals often ambiguous LC
provides precision measurements Sparticle
masses to O(1) at LC compared to O(10) at LHC
Higgs branching ratios to few compared to few
10 at LHC Interpretation of signals thru
measurement of particle properties LHCLC data
reach higher precision Improved measurement
of Higgs coupling Improved measurement of
SUSY parameters Distinguish universal extra
dimensions from SUSY Determine parameters of
Large extra dimensions
Concurrent running important
14
The Linear Collider
Measuring WW and Z0Z0
Many final states involve WW or ZZ pairs
ee- ? WW?? or ee- ? ZZ?? Hadronic decay of
W or Z Branching ratio 70 Results in
two hadronic jets Requires excellent to
resolve ?mZ-W 9.76 GeV
60/vEjet
ALEPH
Jet Energy Resolution
30/vEjet
15
Traditional Jet Measurement
Uses calorimeter alone ? Example of CDF
live event Sandwich design Used by most
calorimeters at colliders ? Alternating
layers of Absorber plates to incite
shower and Active medium (detector)
counting charged particles traversing it

ET
e
16
Traditional jet measurement
Calorimeter measures photons and hadrons in jet
Typically with different response e/h ? 1
Leads to poor jet energy resolution of gt
100/vEjet ZEUS tuned Scintillator and
Uranium thickness to achieve e/h 1 ? Best
single hadron energy resolution ever At the
Linear Collider Goal of
35/vE 50/vE Jet Energy Resolution
s/Ejet 30/vEjet
New approach
17
Need new approach
Particle Flow Algorithms
KL
The idea
HCAL
ECAL
Charged particles
Tracker
measured with the Neutral particles
Calorimeter
p
?
18/vE
Required for 30/vE
Requirements on detector ? Need excellent
tracker and high B field ? Large RI of
calorimeter ? Calorimeter inside coil ?
Calorimeter with extremely fine segmentation
Figure of merit BRI2
18
Particle Flow Algorithms
Do they work?
Applied to existing detectors ALEPH, CDF,
ZEUS ? Significantly improved
resolution Goal for the Linear Collider
Detector Huge simulation effort
underway ? England, France, Germany,
Argonne, Iowa, Kansas, NIU, SLAC
YES! But that is not the issue
Design a detector optimized for the application
of PFAs
19
Particle Flow Algorithms
How do they work?
Many different approaches ? Example from
ANL/SLAC development
I Track calorimeter cluster matching
Identify calorimeter cells belonging to shower
from charged track Algorithm based on
tubes Use track momentum and eliminate
cells II Photon identifier Identify
photons in ECAL Use longitudinal and lateral
shower profile III Measure neutral hadron
energy Apply cone jet algorithm
Identify remaining cells as energy from neutral
hadrons
20
Particle Flow Algorithms
Track-cluster Matching
I Pick up all seed cells close to extrapolated
track Can tune for optimal seed cell
definition For cone size lt 0.1 get 85 of
energy II Add cells in a cone around each seed
cell Apply through n layers III Linked
seed cells in subsequent cones Form the
reconstructed shower IV Discard all cells linked
to track
21
Particle Flow Algorithms
Photon Identification
Definition of EM cluster I Cluster of EM
energies within a cone of 0.04 II No
requirements on EECAL/EHCAL or EHCAL Cuts to
select photon cluster and reject anything else
I EM clusters rejected within 0.03 of
extrapolated track
within 0.01 if track MIP in all 30
layers II EM clusters required to have shower
maximum energy gt 30 MeV (SME
is sum of layers 8,9 and 10) III Require
Ecluster /ptrack gt 0.1, if EM cluster within 0.1
of track
Total Photon Candidate Energy
Total Hadron Level Photon Energy (GeV)
22
Particle Flow Algorithms
First Results
Applied to ee- ? Z0 ? q q events Two Gaussian
fit Future improvements to - Tube
algorithm - Photon finding - Neutral hadron
energy measurement
2 Gaussian fit µ1 88 GeV s1 4.0 GeV
Jet Energy Resolution still factor 2 from goal
µ2 84 GeV s2 7.0 GeV
Lots of effort needed!!!
(before being useful for detector design)
23
Detector concepts
Identical concepts Different dimensions Different
technologies
SiD LDC HD
Silicon Tracking
Gaseous Tracking
24
Calorimeter Developments
Requirements for the
LCD Highly segmented readout
Compact design Layer by layer
longitudinally Short radiation length X0 for
ECAL O(1 cm2) laterally Short
interaction length ?I for HCAL

Minimal Molière radius RM
Molière Radius Definition RM X0ES/EC
with X0 Radiation length
Electron looses all but 1/e of its energy by
Bremsstrahlung Scale for
longitudinal development of EM showers
ES Scaled energy 21 MeV EC
Critical energy Energy
where shower development dies Meaning
90 of energy contained in cylinder with R RM
25
Calorimeter developments
CALICE Collaboration
CAlorimeter for the LInear Collider with Electrons
170 Physicists 27 Institutes 8
Countries 3 Regions
Argonne National Laboratory University of
Iowa Northern Illinois University University of
Texas at Arlington
http//polywww.in2p3.fr/flc/calice.html
26
Calorimeter developments
EM Calorimeter
Choices of absorber
the smaller the better
Known development efforts
27
EM Calorimeter
Why Silicon?
Advantages of Silicon Can be finely
segmented Provides analog information S/N gt
20 Allows for thin active gap Proven
technology
Cost 1/cm2 ???
Readout electronics
Need 2000 4000 m2
28
Tungsten Silicon
Structure 2.8 (21.4mm of W plates)
Structure 4.6 (31.4mm of W plates)
CALICE Prototype
Structure 1.4 (1.4mm of W plates)
Metal insert
Prototype for test beams 30 layers
Active area 18 x 18 cm2 1 x 1 cm2 Silicon
pads 9720 channels Motivation for
prototype Detailed measurement of
showers Simulated energy resolution

Detector slabs
ACTIVE ZONE
1010 mm2
60 mm
Si Wafer with 66 pads
60 mm
29
CALICE Prototype
Detector slabs
PCB Boards 14 layer boards Located outside
of module Thickness of 2.4 mm Molière
Radius 9 mm (W) ? 24 mm (W Gap)
Will do better for LCD
30
CALICE Prototype
Test beam results
3 GeV electrons from DESY
First digital photographs of showers
31
Tungsten Silicon
SLAC Oregon - BNL
6 Silicon Wafers 5 mm Silicon pads 1024
channels Readout chip 1 per wafer Only
300 µm thick Not packaged Bump bonded to
G-10 board Molière Radius 9 mm (W) ? 14 mm
(W gap)
32
Calorimeter developments
Hadron Calorimeter
Assuming geometry of the TESLA detector
Steel best choice for absorber Long
radiation length Short interaction length
Assuming 1 X0 sampling
38 layers 10 mm for active medium
Assumption of 1 X0 sampling being revisited
Segmentation of readout Layer-by-layer
longitudinally O(1 cm2) laterally 5107
readout channels
Challenge
33
Hadron Calorimeter
Known development efforts
Analog Readout
O(10) bit resolution
Digital Readout
1 bit resolution Trading resolution with
granularity of readout
members of CALICE
34
Hadron Calorimeter
5 GeV p
Analog versus Digital
Single Particle Resolution
Analog
Erec (GeV)
1 cm2 pads
Analog
s/µ 22
Digital
Etrue (GeV)
s/µ 19
Digital Readout
Etrue (GeV)
Works due to low density of hadronic
showers Linear response to single hadrons Single
particle resolution preserved Landau tail reduced
35
Hadron Calorimeter
Prototype Section
Dimensions etc 1 m3 (to contain most of
hadronic showers) 40 layers with 20 mm steel
plates as absorber Scintillator Gas
Electron Multipliers (GEMs) as active
medium Resistive Plate Chambers (RPCs)
Motivation for construction and beam tests
Validate techniques Validate concept of the
electronic readout Measure hadronic showers
with unprecedented spatial resolution Validate
MC simulation of hadronic showers Compare
Analog and Digital HCAL
Comparison of hadron shower simulation codes by
G Mavromanolakis
36
Hadron Calorimeter
Analog HCAL
Scintillator tiles (for prototype section) 3 x
3 cm2 ? 12 x 12 cm2 Wavelength shifting fiber
imbedded in tiles 8000 channels in
total Silicon Photomultiplier readout New
development Located directly on tile
37
AHCAL
RD at MEPHI (Moscow) together with PULSAR
(Russian industry)
Silicon PM
Depletion Region 2 ?m
Some features Sensitive area 1 x 1 mm2 Gain
2 106 Ubias 50 V Recovery time 100
ns/pixel Number of pixels 1000/mm2 Dynamic
range gt 200
38
DHCAL
ANL Boston Chicago FNAL Iowa ITEP, Moscow
Resistive Plate Chambers
No ageing ever observed with glass RPCs
are Simple, robust, cheap, quiet, well
understood, reliable Adaptable to different
requirements (TOF, high efficiency, large area)
Graphite
Pick-up pads
Signal
HV
Resistive plates
Gas
µ
Prototypes Built O(20) chambers so far
Glass as resistive plates Dimensions
20 x 20 cm2, 30 x 92 cm2, 1 x 1 m2
Different designs single/double gas
gaps, varying surface resistivity of
graphite layer, exotic designs
39
Hadron Calorimeter
Comparison of RPCs and Scintillator
Studies based on GEANT4 Studies of lateral
shower sizes with 1 cm2 readout pad
sizes and digital readout EM
showers narrower in RPCs Hardonic
showers narrower in RPCs
Clear advantage for separating components of
hadronic jets (PFA)
40
RPC vs Scintillator
Hadronic shower size
Shower radius by particles
Scintillator
Scintillator Shower size dominated by
contribution from protons Protons
knocked-out from neutrons Sensitive to slow
components of shower (nuclear reactions) RPCs
Density of gas too low for sizeable neutron
cross section Energy deposition has no slow
component
Scintillator
RPCs
But needs to be confirmed by data
41
RPCs
Measurements with RPCs
Virtually all RD on chambers completed
Chamber technology in hand
42
RPCs
How do they work?
With increasing HV Efficiency increases to
98 Inefficiency around spacers 2
loss Streamer fraction increases
Above 8.5 kV gt95 efficiency
few streamers
Measurements with Cosmic Rays
Plateau of 1 kV
Streamers
Avalanches
43
RPCs
Digital Readout
Low noise 0.1 Hz/cm2
VME readout system (for tests only) Handles
64 channels Provides hit pattern and time
stamp Tests with RPCs in avalanche mode
64 pad array
Hit multiplicity 1.6 1.7 for efficiency
95 1.4 1.5 for efficiency 90
1 cm2
Trigger area
44
DHCAL
Electronic Readout System
Real challenge Cheap (1 2 /channel)
Low cross-talk, noise
40 layers à 1 m2 1 cm2 readout pads
400,000 readout channels
Conceptual design of system I Front-end
ASIC II Data concentrator III VME
data collection IV Trigger and timing system
45
DHCAL
Front-end ASIC
64 inputs with choice of input gains RPCs
(streamer and avalanche), GEMs Triggerless or
triggered operation 100 ns clock cycle Output
hit pattern and time stamp
Abderrezak Mekkaoui James Hoff
Ray Yarema
Design work at FNAL Design work started in
June, 2004 Digital section completed
First submission on March 21st 2005
46
Calorimeters
Mechanical Structures
Movable table for HCAL Being designed at
DESY To be used by all HCALs Gap for
active medium adjustable 6 10 mm Can
rotate plates keeping same gap Can support ECAL
Movable table for ECAL Being used in test beams
at DESY
47
Time Scales for the ILC
48
Time scales
Potential sites
Germany
Ideally not a green site Easy access for all
nationalities Expected to contribute 40 50
of cost Collisions with high energy protons?
California
Japan
Switzerland
Illinois
49
Time scales
Detector RD and construction
Assuming T 2015
Now is the time to initiate test beam effort
50
Time scales
FNAL MT6 Test Beam
Beam parameters matched to our needs
Momentum between 5 and 120 GeV Protons,
pions, muons, electrons Resonant
extraction implemented Intensity can be
reduced Up to 6 m in lateral space
available
Technical note submitted to FNAL management
35 institutions 14 different projects
worldwide Program to start in 2005
To last until 2008?
51
Conclusions
LC Detector is the first detector being
designed with the application of PFAs in mind
- The goal is 30/vE - This is
the most important detector RD for LC!
Digital hadron calorimetry with very fine
segmentation is going to be a major
breakthrough in jet measurement techniques The
proposed 1 m3 test section will improve our
knowledge of hadronic showers and prove the
concept of digital calorimetry -
The challenge is the electronic readout system
- Funding still needs to be sorted
out The time scale for test beams is 2005
8 - Assuming completion of the LC in
2015
52
RPC DHCAL
Cost estimate (MS only)
500,000 50 contingency