1. LHCf and High Energy Cosmic Rays
Yoshitaka Itow
STE Lab / Kobayashi-Maskawa Inst.
Nagoya University
and on behalf of the LHCf collaboration
“HCPS 2012”
Nov 12-16, 2012, Kyoto

2. Connection to Ultra High Energy Cosmic Rays
Thanks for exciting HCP2012, various exciting results in Higgs, BSM, rere dacays, QGP, etc..
And now, something completely different point of view...
Connection to Ultra High Energy Cosmic Rays

3. Hadron interactions at ultra high energy Accelerator n Cosmic rays
1020eV
1017eV
Hadron interaction data
Hint for interactions at
ultra-ultra high energy
ECM ~ ( 2 × Elab × Mp ) 1/2
√s=14 TeV n 1017eV cosmic rays
√s=447TeV n 1020eV cosmic rays

4. 1017 eV :Crossroad of accelerators and UHECRs
AUGER TA
HEAT
Air shower experiments
AUGER, TA
TALE
GZK
Ankle
Knee
2nd Knee?
14TeV
Colliders
0.5
0.9
2.2 7
Cosmic rays
1014
1017
1020 eV
LHC, Tevatron, SppS and RHIC can verify interactions at 1014 ~ 1017 eV
Low E extension (TALE, HEAT) plan can verify 1017eV shower

5. Air shower observation
Air Florescence telescope (FD)
EM component
(>90% of collision energy)
EM component
Total calorimeter
Shower max altitude
Had component
surface
detectors
Surface Detectors (SD)
# of particles at given altitude
(EM or Muon component)

6. GZK cut-off confirmed ? But…
UHECR2012
Need identify UHECR is
“proton”
Too many Auger, if proton
TA prefers proton
Auger prefers Fe
ICRC2011
ICRC2011

7. ① Inelastic cross section
If large s
rapid development
If small s
deep penetrating
④ 2ndary interactions
nucleon, p
② Forward energy spectrum
If softer
shallow development
If harder
deep penetrating
③ Inelasticity k= 1-plead/pbeam
If large k
(p0s carry more energy)
rapid development
If small k
( baryons carry more energy)
deep penetrating
(relevant to Nm )

8. Feedback from LHC to UHECRs
Astropart.Phys. 35 (2011)
　sinel
Central multiplicity
EPL, 96 (2011) 21002
CMS PAS FWD
Forward energy flow
Forward energy flow 3
3.5 4
4.5 5 h
CERN-PH-EP/

9. Very forward : Majority of energy flow (√s=14TeV)
Multiplicity
Energy Flux
All particles
neutral
8.4 < h <　∞
Most of the energy flows into very forward
( Particles of XF > 0.1 contribute 50% of shower particles )

10. very forward ~ very low pTN 1
String　fragment
pT (GeV/c)
2000
Energy (GeV)
remnant
NxE 1
pT (GeV/c)
T.Pierog
2000
Energy (GeV)

11. The LHCf collaboration
T.Iso, Y.Itow, K.Kawade, Y.Makino, K.Masuda, Y.Matsubara, E.Matsubayashi, G.Mitsuka, Y.Muraki, T.Sako
Solar-Terrestrial Environment Laboratory, Nagoya Univ.
H.Menjo Kobayashi-Maskawa Institute, Nagoya Univ.
K.Yoshida Shibaura Institute of Technology
K.Kasahara, T.Suzuki, S.Torii Waseda Univ. T.Tamura Kanagawa University
M.Haguenauer Ecole Polytechnique, France
W.C.Turner LBNL, Berkeley, USA
O.Adriani, L.Bonechi, M.Bongi, R.D’Alessandro, M.Grandi,
P.Papini, S.Ricciarini, G.Castellini
INFN, Univ. di Firenze, Italy
K.Noda, A.Tricomi INFN, Univ. di Catania, Italy
A-L.Perrot CERN, Switzerland
~30 physicists from 5 countries

12. LHCf site ATLAS 96mm LHCf/ZDC TAN D1 magnet IP 140m Protons
Charged particles (+)
Neutral particles
Beam pipe
Protons
Charged particles (-)
96mm IP
ATLAS
140m
LHCf/ZDC
140m
TAN absorber

13. The LHCf detectors IP1 g n p0 Arm2 Arm1 44X0, 1.6 lint 140m
calorimeter
calorimeter
IP1 g
Arm2 n
Arm1 p0
Front Counter
Front Counter
Arm2
Arm1
44X0,
1.6 lint
16 tungsten + pl.scinti. layers
25mmx25mm+32mmx32mm
4 Silicon strip tracking layers
16 tungsten + pl.scinti. layers
20mmx20mm+40mmx40mm
4 SciFi tracking layers

14. Calorimeter performance
Gamma-rays (E>100GeV, dE/E<5%)
Neutral Hadrons (E>a few 100 GeV, dE/E~30%)
Neutral Pions (E>700GeV, dE/E<3%)
Shower incident position (170mm / 40mm for Arm1/Arm2) p0
g-like
Had-like

15. Brief history of LHCf May 2004 LOI Feb 2006 TDR
Jul 2006
construction
May 2004 LOI
Feb 2006 TDR
June 2006 LHCC approved
Jan 2008
Installation
Aug 2007
SPS beam test
Sep 2008
1st LHC beam
Mar 2010
1st 7TeV run
Dec 2009
1st 900GeV run
(2nd 900GeV in May2010)
Jul 2010
Detector removal

16. LHCf single g spectra at 7TeV
0.68 (0.53)nb-1 on 15May2010
DPMJET QGSJETII-03 SIBYLL 2.1 EPOS 1.99 PYTHIA 8.145
Gray hatch : Sys+stat errors
Magenta hatch: Stat errors of MC
h>10.94
8.81<h<8.99
8.81<h<8.99
h>10.94
Arm1
8.81<h<8.99
h>10.94
PLB 703 (2011)
Arm2
None of the models agree with data
Data within the range of the model spread

17. LHCf single g spectra at 900 GeV
PLB 715 (2012)
May GeV data ( 0.3nb-1 , 21% uncertainty not shown )
DPMJET QGSJETII-03 SIBYLL 2.1 EPOS 1.99 PYTHIA 8.145
h>10.15
8.77 <h<9.46
8.77 <h<9.46
h>10.15 6 6
MC/Data 5
8.77 <h<9.46 5 4 4 3 3 2 2 1 1
h>10.15 50
E(GeV)
450 50
E(GeV)
450

18. Comparison of Data/MC ratio at two energies
DPMJET QGSJETII-03 SIBYLL 2.1 EPOS 1.99 PYTHIA 8.145
High h
low h
MC/Data
MC/Data 2 2
7TeV 1 1
MC/Data
MC/Data
900GeV 2 2 1 1

19. LHCf 7TeV p0 analysis Type-I Type-II Type-I sM=3.7% Type-II1 1
Type-II
PTp0(GeV/c)
Ep0(GeV)
3500
Ep0(GeV)
3500

20. DPMJET 3.04 QGSJETII-03 SIBYLL 2.1 EPOS 1.99 PYTHIA 8.145
LHCf p0 PT spectra at 7TeV
PRD 86 (2012)
DPMJET QGSJETII-03 SIBYLL 2.1 EPOS 1.99 PYTHIA 8.145
PT[GeV]
0.6
PT[GeV]
0.6
PT[GeV]
0.6
PT[GeV]
0.6
PT[GeV]
0.6
PT[GeV]
0.6

21. LHCf p0 PT spectra at 7TeV (data/MC)
DPMJET QGSJETII-03 SIBYLL 2.1 EPOS 1.99 PYTHIA 8.145
EPOS gives the best agreement both for shape and yield.
MC/Data
PT[GeV]
0.6
PT[GeV]
0.6
PT[GeV]
0.6
MC/Data
PT[GeV]
0.6
PT[GeV]
0.6
PT[GeV]
0.6

22. Feed back to UHECR composition
Retune done for cosmic ray MC (EPOS, QGSJET II) with all the LHC input.
(cross section, forward energy flow, LHCf, etc.)
Uncertainty reduced from 50 gcm2 to 20gcm2
( p-Fe difference is 100gcm2 )
Detal reanlaysis of UHECR is also needed for conclusion.
T.Pierog, S.Ostapchenko, ISVHECRI2012
Before LHC
After LHC

23. LHCf future plan
2010
2011
2012
2013
2014
2015
0.9, 7 TeV
8TeV
14TeV
LHC LS1
Detector upgrade
Detector upgrade
Reinstall
Reinstall
LHCf
p-Pb
LHCf II
LHCf I
RHICf?
Analysis ongoing for 2010 data
Neutron energy spectra g inelasticity.
Reinstall Arm2 for p-Pb in early 2013
Very important information for nuclear effect.
Under discussion of common triggers for
combined analysis w/ ATLAS detector.
Reinstall Arm1+2 for 14TeV in 2014
Now upgrading detectors w/ rad-hard GSO.
A new measurement at RHIC 0 degree
Under discussions for 500GeV p+p and d + light-A.
Far future (>2020?) p-N and N-N collisions at LHC ?

24. Expct’d E spectra (p-remnant side )Pb p
Small tower
Large tower  n

25. Summary
UHECR needs accelerator data to solve the current enigma, and may also hint QCD at beyond-LHC energy.
eV is an unique overlap region for colliders and UHECRs
Various LHC data already implemented for cosmic rays interaction models.
LHCf provides dedicated measurements of neutral particles at 0 deg to cover most of collision energy flow.
E spectra for single gamma at 7TeV and at 900GeV. Agreement is “so-so”, but none of models really agree.
PT spectra for 7TeV p0. EPOS gives nice agreement.
Future
2004 LHC p-Pb run to study nuclear effect at 0 degree.
Revisit “14TeV” at ~2014 with a rad-hard detector.
Possible future RHIC run is under discussion.
Possible LHC light ion runs is under discussion.

26. Backup

27. LHCf calorimeters
Arm#2 Detector
Arm#1 Detector
90mm
290mm

28. Setup in IP1-TAN (side view)
BRAN-IC
BRAN-Sci
ZDC
type2
LHCf Calorimeter
LHCf Front Counter
ZDC
type1
Beam pipe
TAN
Neutral
particles
Side view
IP1
Distance from center
Beam pipe

29. Event sample (p0 g 2g ) Total Energy deposit Energy Shape PID X-view
Longitudinal development measured by scintillator layers
25mm Tower
32mm Tower
Total Energy deposit
Energy
Shape
PID
600GeV 　　photon
420GeV 　　photon
Lateral distribution measured by silicon detectors
X-view
Hit position,
Multi-hit search.
Y-view
π0 mass reconstruction from two photon.
Systematic studies

30. Forward production spectra vs Shower curve
XF = E/Etot
Half of shower particles comes from large XF g
Measurement at
very forward region
is needed

31. Parent p0 pseudorapidity producing ground muons

32. The single photon energy spectra at 0 degree at 7TeV
(O.Adriani et al., PLB703 (2011)
DATA
15 May :45-21:23, at Low Luminosity 6x1028cm-2s-1, no beam crossing angle
0.68 nb-1 for Arm1, 0.53nb-1 for Arm2 MC
DPMJET3.0４, QGSJETII03, SYBILL2.1, EPOS1.99 PYTHIA with the default parameters.
107 inelastic p-p collisions by each model.
Analysis
Two pseudo-rapidity, η>10.94 and 8.81<η<8.99.
No correction for geometrical acceptance.
Luminosity by FrontCounter (VdM scan)
Normalized by number of inelastic collisions
with assumption as s inela = 71.5mb. (c.f ± mb by TOTEM )
Arm1
Arm2
+1.8
-1.3

33. New 900 GeV single g analysis
0.3nb-1 data (44k Arm1 and 63k Arm2 events ) taken at 2,3 and 27 May, 2010
Low luminosity (L~1028 typical,1 or 4 xing), negligible pile up ( 0.05 int./xing ).
Relatively less h-dependence in the acceptance. Negligible multi-incidents at a calorimeter (~ 0.1 g (>50GeV) /int. )
Higher gain operation for PMTs. Energy scale calibration by SPS beam, checked with p0 in 7TeV data.
g-like
hadron-like
PID (L90)
Arm1 small GeV
Mp0 in 7TeV
w/ normal&high gain

34. XF spectra for single g: 900GeV/ 7TeV comparison
(sys error not included)
Arm1-Data Preliminary
Arm1-EPOS Preliminary XF XF
Comparing XF for common PT region at two collision energies.
Less root-s dependence of PT for XF ?

35. LHCf g / p0 measurement η ∞ p0 @ 7TeV p0 g gg y=8.9 PT [GeV] y=10.0
1.0
h=8.40
h=8.77
h=7.60
h=6.91
h=5.99
7TeV
PT [GeV]
Energy [GeV]
p0 g gg η ∞
8.7 θ
[μrad]
310
Viewed from IP1
(red:Arm1, blue:Arm2)
Projected edge of beam pipe

36. LHCf type-I p0 analysis
Low lumi (L~5e28) on 15-16May, 2.53(1.91) nb-1 at Arm1 (Arm2). About 22K (39K) p0 for Arm1(Arm2) w/ 5%BG.
For Eg>100GeV, PID (g selection), shower leakage correction, energy rescaling (-8.1% and -3.8% for Arm1&2).
(E, PT) spectra in +-3s p0 mass cut w/ side band subtracted.
Unfolding spectra by toy p0 MC to correct acceptance and resolution 1
PTp0(GeV/c)
1/Nint Ed3N/dp3
y=8.9
y=10.0
10-4
Mp0(MeV)
0.6
　Acceptance of p0
Rapidity
PTp0(GeV/c)

38. Average PT of p0 Comparison w/ UA7@630GeV Extend to higher h regions
1. Thermodynamics (Hagedron, Riv. Nuovo Cim. 6:10, 1 (1983))
Comparison w/
Extend to higher h regions
Less energy dependence of <PT>?

39. Next target: Inelasticity~ 0 degree neutrons
Important for Xmax and also Nm
Measurement of inelasticity at LHC energy
Neutral hadrons at 14 TeV
(LHCf acceptance, 30% resolution)
Neutral hadrons at 14 TeV
(LHCf acceptance, no resolution)

40. Nuclear effects for very forward region
Air showers take place via p-N or Fe-N collisions !
Nuclear shadowing, final state interaction, gluon saturations
Nuclear modification factor at 0 degree may be large.
Phys. Rev. Lett. 97 (2006)
p-p
QGSJET II-04
p-N
All s
p-Pb
8.81<<8.99
>10.94
Courtesy of S. Ostapchenko

41. LHCf p – Pb runs at √sNN=4.4TeV (Jan2013)
IP8
IP2
IP1
Arm2
2013 Jan / a month of p-Pb opportunity.
3.5TeV p +1.38TeV/n Pb (√sNN=4.4TeV)
Expected luminosity: 3×1028cm-2s-1, sAA=2b
Install only Arm2 at one side (Si good for multiplicity)
Trig. exchange w/ ATLAS g centrality tagging
　Requested statistics : Ncoll = 108 ( Lint = 50 mb-1 )
　2106 single γ
　 0
　Assuming L = 1026 cm-2s-1
(1% of expected lumi)
t = 140 h (6 days) !
Arm2
Si Elec.
Preamp

42. “RHICf” : h acceptance for 100GeV/n d-N MC
h>5.8 is covered d N
All particles
No acceptance forπ0
at 900GeV
Neutrals
Acceptance for p0
Energy flow

43. Future p-N and Fe-N in LHC ?
LHC 7TeV/Z p-N and N-N collisions realize the laboratory energy of 5.2x1016eV and 3.6x1017eV, respectively (N: Nitrogen)
Suggestions from the CERN ion source experts:
LHC can in principle circulate any kind of ions, but switching ion source takes considerable time and manpower
Oxygen can be a good candidate because it is used as a ‘support gas’ for Pb ion production. This reduces the switching time and impact to the main physics program at LHC.
According to the current LHC schedule, the realization is not earlier than 2020.
New ion source for medical facility in discussion will enable even Fe-N collisions in future