PPAP: Overview of Calorimetry R

1
PPAP Overview of Calorimetry RD for the ILC

• Paul Dauncey
• Imperial College London

2
Physics motivation

• Need for calorimetry driven by hadronic jet
  energy resolution
• Several important physics channels need to be
  reconstructed from quark final states, e.g. H?WW
  vs. ZZ, ee?nnWW
• Need resolutions 30/?Ejet, around a factor two
  better than Aleph/Atlas
• General agreement that particle flow jet
  reconstruction is needed
• Single particle energy resolution is not the most
  important measure
• Particle flow measures jet components where
  resolution is best
• Charged particles in the tracker
• Photons in the electromagnetic calorimeter (ECAL)
• Neutrons and KLs in the hadronic calorimeter
  (HCAL)
• Requires separation of charged particle, photon
  and neutral hadron energies in the calorimeters
  through pattern recognition
• A naïve calculation shows the HCAL dominates the
  jet resolution
• But realistically, resolution may be dominated by
  separation confusion

3
Tracking calorimetry

• Calorimeters must allow detailed shower
  reconstruction
• Need very fine granularity to reduce confusion
• Make explicit shower clusters, splitoffs and
  secondary particle identification in the ECAL and
  HCAL
• Effectively need to reconstruct tracks in the
  calorimeters
• Reconstruction becomes a significant software
  problem
• Many techniques being pursued to do this task

• Hard to quantitatively compare the physics impact
  of calorimeter designs definitively
• New clever ideas can improve resolution
• Different calorimeters may be optimised with
  different reconstruction algorithms

4
ILC detector status

• Decision on cold machine this summer gave big
  impetus to ILC
• Effort now to globalise the regional
  (Europe/US/Asia) detector concepts
• N.B. Generally, cold decision makes calorimeter
  electronics designs easier
• Three basic detector concepts being pursued
• Small ( Small), main idea is an all-silicon
  tracker, high B field,
• Medium ( Large), like TESLA detector, TPC
  tracker
• Large ( Huge), TPC (or jet chamber) tracker,
  moderate B field
• The overall schedule is to develop several
  detector designs so as to be ready for first beam
  in 2015
• LoI by 2008, used to choose two designs
• Two TDRs by 2009
• Whether we believe the 2015 date or not, we have
  to work to the 2009 TDR schedule to stay in the
  game

5
ECAL concepts

• General agreement silicon-tungsten sampling
  calorimeter is best

• Si-W uses 1?1cm2 diode pads as sensitive layer
• Very thin (i.e. compact) with fine granularity
  good for particle flow
• High channel count around 20M pads in total
  (its a tracker!)
• Tungsten has small Moliere radius, keeps showers
  narrow
• BUT.. cost is very high the silicon wafers alone
  would be over 100M today

6
ECAL concepts

• Other concepts are mainly based on scintillator,
  either tiles or strips, with tungsten (or
  sometimes lead) converter layers
• Recognised to have significantly worse pattern
  recognition how much worse in terms of physics
  performance is hard to quantify now
• Several concepts use a hybrid of scintillator
  with a few silicon layers in the body of the ECAL
  to improve pattern recognition

• ECAL choice could be seen as a continuous
  spectrum
• From all-silicon to hybrid to all-scintillator
• Optimisation point will depends on
  cost/performance balance needed

7
HCAL concepts

• Two main streams analogue or digital (binary)
• Both tend to assume steel (or lead) as converter
  layers
• Analogue HCAL (AHCAL)
• Scintillating tiles 3?3cm2 as sensitive layer
  5M channels
• Favoured analogue signal readout method is
  silicon-PM 1?1mm2 solid-state device which can
  be attached to tile fibre directly.
• Also looking at APDs/HPDs, etc, but cost may then
  require ganging and B field might be a problem

• Digital HCAL (DHCAL)
• RPCs (or GEMs) with pads 1?1cm2 as sensitive
  layer
• Channel count 40M pads
• Binary readout only threshold on each pad output
  
• No fibres, optical connections much simpler to
  build
• Loss of analogue information compensated by
  increase in granularity

140mm
70mm
8
Regional interests

• Europe
• Czech Rep/France/Russia/UK (CALICE) Si-W ECAL
• Italy/Poland Silicon/scintillator-Pb hybrid ECAL
• Czech Rep/Germany/Russia (CALICE) Tile AHCAL
• US
• BNL/Oregon/SLAC Si-W ECAL
• Others (individual Univs) Scintillator-W hybrid
  ECALs, crystal (!) ECAL
• ANL/FNAL/various Univs (CALICE) RPC and GEM
  DHCAL
• Asia
• Korea (CALICE) Si-W ECAL
• Japan Strip/tile scintillator-W ECAL, tile
  scintillator-Pb HCAL
• N.B. CALICE is more than an order of magnitude
  bigger than any other group in calorimetry, has
  collaborators from all three regions, and is
  involved in all three of ECAL, AHCAL and
  (uniquely) in DHCAL

9
UK perspective

• Decided to get involved in 2001, following TESLA
  TDR
• Experience of interested people was in
  electronics, ECALs and silicon detectors
• Not a hard choice joined CALICE Si-W ECAL effort
• Now providing readout electronics for the silicon
  wafers
• Main aim of CALICE is to validate the simulation
• Pre-prototype calorimeters in test beams from
  2004 to 2006
• This will allow believable detailed optimisation
  of calorimeter design later
• Longer term proposal (Jan 2005) will be for
  DAQ/electronics
• Several tricky potential bottlenecks identified
  which need RD
• Effectively zero work worldwide in this area at
  present
• UK could easily lead this and project could match
  feasible UK funding
• N.B. Very little HCAL work outside of CALICE
• Large parts of these DAQ studies would be common
  to both calorimeters
• UK keeps foot in both camps in case Si-W ECAL not
  chosen

10
New concept within the UK

• Possibility to make a digital ECAL (DECAL)
• Similar idea to DHCAL have small pixels and
  threshold the output of each
• Threshold set for single MIP probability of more
  than one must be small
• Particle densities are very high in 500 GeV EM
  shower core 100MIPs/mm2
• Would need pixels at most 100?100mm2 total
  pixel count 1011!
• Previously thought to be unfeasible to get so
  many signals out
• Monolithic active pixel sensors (MAPS) becoming
  mature
• Have intelligent readout electronics integrated
  on same wafer as pixels
• Allows huge zero suppression of non-hit channels
  on silicon wafer
• Data volumes read out from wafer are within order
  of magnitude of AECAL
• All the advantages of Si-W, but potentially
  further improvements
• Performance better granularity, more compact,
  better ECAL resolution
• Mechanical easier cooling and large scale
  construction
• Cost standard CMOS process, not high resistivity
  silicon factor of 2?
• Jan 2005 proposal will contain bid for MAPS
  development

11
UK future plans

• Current CALICE programme
• Simulation validation and tuning
• Test beams, data analysis and papers
• Approved to March 2005
• Jan proposal bid for two year extension to March
  2007 to finish
• Future programme (still within CALICE framework)
• DAQ/electronics RD
• MAPS detector concept validation
• Mechanical/thermal studies
• Simulation and physics
• Jan proposal bid for three year programme to
  March 2008
• Should then be in a position to contribute to the
  TDR in 2009

12
My opinions ( uneducated guesses!)

• Global detectors studies will converge on two
  designs
• At present, only Huge design does not have Si-W
  ECAL as baseline
• Very unlikely both of the chosen designs will not
  have Si-W ECAL
• Assuming a Si-W ECAL is proposed
• CALICE has lead the Si-W ECAL concept from the
  start (before the UK!)
• CALICE (including the UK) will lead, or at least
  be heavily involved
• Only possible alternative scenario
• US grabs Small design and does Si-W ECAL,
  excluding other collaborators
• AND
• Other design is Huge and goes for a pure
  (non-hybrid) scintillator ECAL
• There is always the HCAL
• My guess based on physicists being conservative
  one AHCAL, one DHCAL
• CALICE will be heavily involved in both
• DAQ gives UK insurance to join this, just in case