Study of neutral baryon production at the very forward region of the - - PowerPoint PPT Presentation

Kentaro Kawade

STE laboratory, Nagoya University, Japan 2 July 2013 @ Rio de Janeiro, Brazil

Study of neutral baryon production at the very forward region of the LHC

Contents

• Introduction
• Motivation
• LHCf experiment
• Detector performance
• Data analysis
• Summary

Hadronic interaction in cosmic-ray showers

• Forward particle production is

quite important to understand the cosmic-ray shower

• Forward baryons

(Shower core)

• Mesons

(EM shower)

• Verifying the hadronic

interaction models → measurement of forward particles by using “accelerator” = LHCf

UHECR

leading baryon (shower core)

EM shower μ+/-(Meson)

π0→2γ(EM)

LHCf detector

neutral baryons

(neutron~94%, Λ ~6% (DPM3))

The LHCf experiment

• LHCf measures neutral particles

emitted to forward region of LHC

• Two independent detectors “Arm1”

and “Arm2” are installed in both side of the LHC IP1

• The LHCf detectors are composed
• f sampling and imaging

calorimeters

• Calorimeter; Tungsten &

scintillator

• position sensor; SciFi or Silicon

strip sensor Arm1 Detector

LHC Interaction point 1

Holizontal(mm)

• 40
• 30
• 20
• 10

10 20 30 40 Vertical(mm)

• 20
• 10

10 20 30 40 50 60 70 80

Arm1

=10.15 η 11mm, =9.46 η 22mm, =8.77 η 44mm,

Small tower η > 10.15 Large tower 8.77<η<9.46

More details about LHCf → 4 july Menjo’s talk ()

155~310µrad

0~80µrad

• 140m

140m

Motivations; Forward baryons

• Very large difference in neutral

baryon spectra among the models is expected

• Direct measurement of

inelasticity

• Muon excess
• Muon excess in CR
• bservation is found relative

to the MC predictions ( +30% than MC)

• Forward baryon production

is important [T. Pierog, K. Werner PRL 101, 171101 (2008)]

Energy[GeV] 1000 2000 3000 4000 5000 6000 Entries 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000

Energy[GeV] 1000 2000 3000 4000 5000 6000 Entries 500 1000 1500 2000 2500 3000 3500 4000

Expected neutron energy spectra LHCf Small tower LHCf Large tower

100 101 102 500 1000 1500 2000 S [VEM] Radius [m] Proton Sim Iron Sim Data

[ J.Allen, et al. ICRC2011 Proceedings]

Analysis

Detector performance

• The performance of the

LHCf detector was studied using MC simulations

• Detection efficiency
• Energy reconstruction
• Position determination
• The energy scale was

checked comparing with the beam experiment at SPS (350GeV proton)

Small Large Detection efficiency ~70% ~70% Non linearity ±1% ±4% Energy resolution 37~42% 36~48% Position resolution 2.5~0.5 mm 4.0~1.0 mm

(Details are summarized in my proceedings) icrc2013 #0850

Data statistic

• Data
• Taken in May 2010 with √s = 7TeV (≒2.5x1016eV @ ELab)
• Integrated luminosity 0.68nb-1 (calculated in previous study)
• Ninel = 0.68nb-1 * 71.5 [mb] ≒ 4.8 x 107 [collisions] (50M col)
• MC
• QGSJET2 : 1x107 collisions
• EPOS : 1x107 collisions
• PYTHIA : 0.8x107 collisions
• SYBILL : 0.5*107 collisions

Analysis method

Event selection Reconstruct hit position and shower shape (L20% and L90%) Reconstruct energy and pT

dE in calorimeter, SciFi

Reconstruct energy (1st)

• Same analysis procedure

was used for the experimental data and MC

• Calorimeter was well

calibrated by using SPS beams (for photon analysis)

• No correction (PID,

detection efficiency) was applied

Input

3 fold coincidence in calorimeter

PID (energy, L20%, L90%)

Particle identification (PID)

Sample)

Shower development in a calorimeter

5 10 15 20 25 30 35 40 1000 2000 3000 4000 5000 6000 7000 8000

hTC_Small_tower

L90%

L20[r.l] 5 10 15 20 25 30 35 40 L90[r.l.] 5 10 15 20 25 30 35 40

signal

L90 - 0.25*L20 5 10 15 20 25 30 35 40 200 400 600 800 1000 1200 1400

L90 - 0.25*L20

L20% L90%

L20%

Layer[r.l.]

hadron photon

projection along the sloped line

• To perform PID with higher

efficiently and less contamination is essential

• 2D method using L20% and

L90% are used for PID in this study

• L20% (L90%) is the depth

containing 20% (90%) of total deposited energy

• 2D cut parameter L2D is
• btained as below

L2D = L90% - 1/4*L20%

L2D

Reconstructed Energy [GeV] 1000 2000 3000 4000 5000 6000 7000 8000 /GeV

inel

events/N 0.5 1 1.5 2 2.5 3 3.5 4

• 6

10 ×

Data EPOS QGSJET2 PYTHIA SYBILL

LHCf Preliminary

hData

Reconstructed Energy [GeV] 1000 2000 3000 4000 5000 6000 7000 8000 /GeV

inel

events/N 0.5 1 1.5 2 2.5 3 3.5

• 6

10 ×

Data EPOS QGSJET2 PYTHIA SYBILL

LHCf Preliminary

Small tower

Preliminary result

• No model can explain our result perfectly

LHCf Small tower LHCf Large tower

No rapidity selection No efficiency correction Only statistical error No rapidity selection No efficiency correction Only statistical error

Next steps

The systematic errors

• The major parts are listed in the table
• Studies are ongoing

% Luminosity +-5% PID +-10% Multi-hit event study ongoing Beam center position study ongoing Energy scale study ongoing

• The measured spectra were

smeared by the detector response ⇔ to determine the inelasticity, unsmeared spectra are important

• Study of unfolding method

based on the bayesian statistics is ongoing

• The detector response trained

by QGSJET2 spectra unfolded EPOS spectra

measured (MC) true energy (MC)

• unfolded spectra

Next challenge; Spectra unfolding

LHCf preliminary

Emeasure = AEtrue Etrue = A-1Emeasure A; Response matrix Emeasure; Measured energy spectra Etrue; True energy spectra

• LHCf measured neutral baryons at the very

forward region of the LHC with √s=7TeV p-p collision

• The energy spectra were compared with the

known models (EPOS, QGSJET2, SYBILL, PYTHIA)

• No model can explain our result perfectly
• Study about systematic uncertainties is ongoing
• Unfolding study is also ongoing

Summary and plans

Spare slides

PID systematic error

• PID efficiency and purity are

estimated from MC simulation

• Scaling Photon/Hadron ratio to

reproduce the experimental data → some disagreement between data and MC

• PID efficiency and purity was

estimate by another method (artificial method)

• Difference between the two methods

is considered as systematic error → ～10% (depending on energy)

5 10 15 20 25 30 35 40 5000 10000 15000 20000 25000 30000

500-1000GeV

MC; Hadron MC; Photon

×; Data ■; MC fit

5 10 15 20 25 30 35 40 5000 10000 15000 20000 25000 30000 35000

500-1000GeV

L2D L2D

500 1000 1500 2000 2500 3000 3500 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Entries 80134 Mean x 2089 Mean y 0.08944 RMS x 872.3 RMS y 0.05328

100 200 300 400 500 600 700 800

Entries 80134 Mean x 2089 Mean y 0.08944 RMS x 872.3 RMS y 0.05328

hEvsPthadron_Small_True

500 1000 1500 2000 2500 3000 3500 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Entries 43829 Mean x 1226 Mean y 0.3085 RMS x 784.3 RMS y 0.1957

100 200 300 400 500 600 700 800

Entries 43829 Mean x 1226 Mean y 0.3085 RMS x 784.3 RMS y 0.1957

hEvsPthadron_Large_True

pT-Energy coverage

• Expected pT-Energy spectra (EPOS)
• Detector response not include

LHCf Small tower LHCf Large tower

Vertex resolution

• The position resolution of

large tower seems worse

• Reason;

because the position resolution is defined as standard deviation of hitposition distribution

• Width itself is almost same

Incident Energy[GeV] 500 1000 1500 2000 2500 3000 3500 Fitting RMS[mm] 0.5 1 1.5 2 2.5 3 Incident Energy[GeV] 500 1000 1500 2000 2500 3000 3500 Fitting RMS[mm] 0.5 1 1.5 2 2.5 3

Position resolution 2.5~0.5 mm 4.0~1.0 mm

LHCf Small tower LHCf Large tower

↓FWHM