Study of a Compensating

Calorimeter

for a e+ e- Linear Collider

at Very High Energy

30 Aprile 2007

Vito Di Benedetto

ILCA future project for a

e+ e- Linear Collider

electron-positron collider;

ILC's design consist of

two facing linear

accelerators, each 20

kilometers long;

c.m. energy 0.5 - 1 TeV;

ILC target luminosity:

500 fb-1 in 4 years.

Fourth Concept Detector (“4th”)

Basic conceptual design: 4 subsystems• Vertex Detector 20-micron pixels• Time Projection Chamber

Drift Chamber as alternative to overcome

known limitations of the TPC technology• Double-readout calorimeters

Fibers hadronic calorimeter:

scintillation/Čerenkov Crystals EM calorimeter

• Muon dual-solenoid spectrometer

Requirements for ILC Detectors

Physics goal of ILC Wide variety of processes

Energy range: Mz<ECM<1 TeV

Basic detectors requirements Efficient identification and precise 4-momentum measurement

of the particles Extremely good jet energy resolution to separate W and Z

Efficient jet-flavor identification capability

Excellent charged-particle momentum resolution

Hermetic coverage to veto 2-photon background

Calorimetry at ILC

Most of the important physics processes to be studied in the ILC

experiment have multi-jets in the final state

Jet energy resolution is the key in the ILC physics

The world-wide consensus of the performance goal for the

jet energy resolution is:

)GeV(/%30/ EEE

Problems in Hadron Calorimeters

The most important fluctuation is in the em shower

fraction, fem

LESSONS FROM 25 YEARS OF R&D

Energy resolution determined by fluctuations

To improve hadronic calorimeter performance

reduce/eliminate the (effects of) fluctuations that dominate the

performance

Measurement of fem value event by event by comparing

two different signals from scintillation light and

Ĉerenkov light in the same device.

Solution: Dual Readout Calorimeter

Unit cell

Back end of

2-meter deep

module

Physical

channel

structure

Dual REAdout Module (DREAM)http://www.phys.ttu.edu/dream/

From DREAM to the 4th Concept HCAL

Cu + scintillating fibers

+ Ĉerenkov fibers

~1.5° aperture angle

~ 10 int depth

Fully projective geometry

Azimuth coverage

down to 3.8°

Barrel: 13924 cells

Endcaps: 3164 cells

Simulation/Reconstruction Steps

inside ILCRoot FrameworkMC Simulation Energy Deposits in Detector

Digitization Detector response combined

Pattern Recognition Recpoints

Track Finding Tracks

Track Fitting Track Parameters

ILCRoot: summary of features

CERN architecture (based on Alice’s Aliroot)

Full support provided by Brun, Carminati, Ferrari, et al.

Uses ROOT as infrastructure– All ROOT tools are available (I/O, graphics, PROOF,

data structure, etc)– Extremely large community of users/developers

Six MDC have proven robustness, reliability and portability

Single framework, from generation to reconstruction through simulation. Don’t forget analysis!!!

Calibration

Energy of HCAL calibrated in 2 steps:

Calibrate with single 40 GeV e-

EC and E

S

Calibrate with single 40 GeV

C

and S

Sh

eS

Ch

eC

Reconstructed energy

Once HCAL calibrated, calorimeter energy:

SC

SCCCSSHCAL

EEE

11

HCAL Resolution Plots

40 GeV e-

40 GeV π-

S

S

C

C

EHCAL

EHCAL

Reconstructed vs Beam Energy

Total Energy

Pattern

Recognition

cc & & ss

Independent Independent

on Energyon Energy

Pions data

all HCAL energy

single recpart energy

Visible Visible

energy fully energy fully

measuredmeasured

Resolution for hadrons

Low statistics

Pattern

Recognition

Pions data

all HCAL energy

single recpart energy

/ndf 1.351e-05/4

P0 0.3545± 0.01041

P1 0.001335±0.001704

Total Energy

/ndf 1.435e-05/4

P0 0.3803± 0.01072

P1 0.0002627±0.001756

Particle Identification

e

e

40 GeV particles

Jets Studies

e+ e- -> q q (uds)

The Jet Finder Algorithm Look for the jet axis using a Durham algorithm

Charged tracks

Calorimeter cells

Calorimeter Clusters

Jet core

Open a cone increasingly bigger around the jet axis (< 60°)

Run a Durham j.f. on the cells of the calorimeter inside the cone

Jet outliers

Check leftover/isolated calo cluster/cells for match with a track from TPC+VXD

Add calorimetric or track momentum

Add low Pt tracks not reaching the calorimeter

Muons

Add tracks reconstructed in the MUD

Total Energy PlotsNo jet finderEnergy calibration with

no material in front

Energy Resolution

Total visible

Energy (no jet

finding)

Single jet (jet

finding included)

Physics Studies

e+e- -> ZoHo -> cc

Jet Finder Performance

Angular resolution < 2°

Energy resolution = 4 GeV

Jet-Jet Mass Plot

ConclusionsThe 4th Concept has chosen a Calorimeter with

Dual Readout

The technology has been tested at a test beam,

but never in a real experiment

Performance of Calorimeter is expected to be

extremely good:

σE/E = 38%/√E (single particles)

σE/E = 39%/√E (jets)An ECAL design with Dual Readout crystal

technology is under way

Bottom view of single cell

Bottom cell size: ~4.8 × 4.8 cm2

Top cell size: ~ 8.8 × 8.8 cm2

Prospective view of clipped cell

Cell length: 150 cm

Number of fibers inside each cell: 1980equally subdivided between Scintillating and CerenkovFiber stepping ~2 mm

Hadronic Calorimeter CellsHadronic Calorimeter Cells

Simulation (1)

Light production in the fibers simulated through 2 separate

steps:

1. Energy deposition (hits) in active materials calculated by

the tracking algorithm of the MC

2. Conversion of the energy into the number of S and C

photons by specific routins taking account several

factors: energy of the particle, angle between the particle

and the fiber, etc. Poisson uncertaintity introduced in the

number of photon produced

Simulation (2)

Response function of the electronics not

yet simulated (digits)

Random noise generated to test the ability

of reconstruction algorithm to reject such

spurious “hits”

Reconstruction

Clusterization ( pattern recognition)

cluster = collection of nearby “digits” Build Clusters from cells distant no more than two

towers away

Unfold overlapping clusters through a Minuit fit to cluster shape

Reconstructed energy E adding separately ES

and EC of all the cells belonging to the

reconstructed cluster

e+e- -> ZoHo -> cc Pandora-Pythia (Ecm=350 GeV, MH=140 GeV) + Fluka

No MUD (use MC truth) Cut recoil mass 20 GeV around Zo mass

Maximize j.f. efficiency through yt cut (ff=97%)