pythia 8 status - conference.ippp.dur.ac.uk · diffraction i move from inel/nsd !inel>0 datasets...
TRANSCRIPT
PYTHIA 8 STATUS
Richard Corke
Department of Astronomy and Theoretical PhysicsLund University
August 2011
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 1 / 24
PYTHIA 8
I PYTHIA 8 is the C++ rewrite of PYTHIA 6
I Some features removed, some not yetimplemented
I Independent fragmentation and mass-orderedshowers removed
I No ep, γp and γγ beam configurations
I Focus of new development; many new features not found in PYTHIA 6.4I Fully interleaved p⊥-ordered MPI/ISR/FSR evolutionI Richer mix of underlying-event processes (γ, J/ψ, DY, ...)I Can select two hard interactions in the same eventI Hard diffractive component (S. Navin)I τ lepton polarisation in production and decay (P. Ilten)I Updated decay data and LO PDF sets (T. Kasemets)
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 2 / 24
Diffraction
I Move from INEL/NSD→ INEL>0 datasetsI Reproducible definitions!I Diffractive description more important
I Soft description same as in PYTHIA 6I Pomeron kicks out valence quark or gluon from the proton
I New high-mass diffractive framework using Ingelman-Schlein pictureI “Diffraction in Pythia,” S. Navin, arXiv:1005.3894 [hep-ph]
I Single diffractionI Proton emits Pomeron according to Pomerom PDF, fIP/p(xIP, t)I Pomeron–proton collision using the full machinery of
proton–proton collisions
ticles) and the non-dissociated proton as the pink dot in figure 2. The LHC cross-section (at√s = 14TeV) for SD is ∼ 10mb [5].
φ
η0 5 10
0
−5−10
p1
p2
P
p1′
X2
Figure 2: SD diagram and a window showing a rapidity gap between −10 < η < 3.5.
If both the colliding protons dissociate, then it is Double Diffractive (DD) (p1+ p2 → X1+X2)
as seen in figure 3. The LHC cross-section (at√s = 14TeV) for DD is ∼ 7mb [5].
φ
η0 5 10
0
−5−10
p1
p2
P
X1
X2
Figure 3: DD diagram and window showing a rapidity gap between −3.5 < η < 4.
A different topology becomes possible with two Pomerons exchanged, namely Central Diffrac-
tion (CD) (p1 + p2 → p′1 + X + p′2) or Double Pomeron Exchange. In this process, both the
protons are intact and are seen in the final state (as two pink dots seen in figure 4). The LHC
cross-section for CD is ∼ 1mb [5].
In Non-Diffractive (ND) interactions there is an exchange of colour charge and subsequently
more hadrons are produced. This is shown in figure 5. ND interactions are the dominant
process in pp interactions and are expected to be ∼60% of all interactions at the LHC with a
cross-section of ∼65mb (at√s = 14TeV) [5].
3
I Double diffraction: modelled as two Pomeron–proton collisions
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 3 / 24
Diffraction
I PYTHIA 8 and PHOJET agree quite well
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
<N
>
-410
-310
-210
-110
1 Pythia6
Phojet
Pythia8
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
<N
>
-410
-310
-210
-110
1
Pythia6
Phojet
Pythia8
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
<N
>
-410
-310
-210
-110
1 Pythia6
Phojet
Pythia8
pT Distributions (√s=0.9 TeV)
10
Softer pT spectrum in Pythia6 due to lack of high mass diffraction
Pythia8 and Phojet agree quite well
ND
DD
SD
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
<N
>
-410
-310
-210
-110
1 Pythia6
Phojet
Pythia8
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
<N
>-410
-310
-210
-110
1
Pythia6
Phojet
Pythia8
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
<N
>
-410
-310
-210
-110
1 Pythia6
Phojet
Pythia8
pT Distributions (√s=0.9 TeV)
10
Softer pT spectrum in Pythia6 due to lack of high mass diffraction
Pythia8 and Phojet agree quite well
ND
DD
SD
Beate Heinemann, MB/UE Working Group
I Choice between 5 Pomeron PDFsI Free parameter σIP needed to fix 〈nint〉 = σjet/σIP
I Framework still needs more testing and tuning!
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 4 / 24
Interleaved evolution
I Interleaved evolution for ISR, FSR and MPI, all p⊥ ordered
dPdp⊥
=
(dPMPI
dp⊥+∑ dPISR
dp⊥+∑ dPFSR
dp⊥
)× exp
(−∫ p⊥max
p⊥
(dPMPI
dp′⊥+∑ dPISR
dp′⊥+∑ dPFSR
dp′⊥
)dp′⊥
)I ISR and MPI compete for beam momentumI Hybrid approach to shower recoils:
I FSR is dipole: nearest colour-connected neighbourI ISR is traditional: whole hard-scattering system affected
(as ISR dipole gives wrong answer e.g. for p⊥Z)
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 5 / 24
Interleaved FSR
I Problems in describing the underlying event (PYTHIA 8.135)
b
b
b
b b bbb b b b b b b b b b b b b b b b b b b
bb b b b
bb
b
CDF datab
PYTHIA 8.135
0
0.2
0.4
0.6
0.8
1
1.2
Transverse region charged particle density
〈Nch〉/
dη
dφ
0 50 100 150 200 250 300 350 400
0.6
0.8
1
1.2
1.4
pT(leading jet) / GeV
MC
/d
ata
b
b
b
bb b b b b
bb b
bb b b b
b b b b b b b bb b
b bb
b
b
b
b
CDF datab
PYTHIA 8.135
0
0.5
1
1.5
2
Transverse region charged ∑ p⊥ density
〈 ∑p
tra
ckT
〉/d
ηd
φ/
GeV
0 50 100 150 200 250 300 350 400
0.6
0.8
1
1.2
1.4
pT(leading jet) / GeV
MC
/d
ata
I No longer a universal minimum bias and underlying event tuneI ISR and MPI already interleaved in PYTHIA 6.4; issue with FSR?
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 6 / 24
Interleaved FSR
I Final-state parton may have colour partner in the initial stateI How to subdivide FSR and ISR in this kind of dipole?I Large mass→ large rapidity range for emission
m ∼ p⊥
m >> p⊥
I In dipole rest frame
Final-state parton may have colour partner in the initial state.How to subdivide FSR and ISR in this kind of dipole?Large mass → large rapidity range for emission:
m ∼ p⊥
m >> p⊥
m/2
p⊥
FSR
ISR
In dipole rest frame
(think rapidity space)
Solution: suppress final-state radiation in double-counted regionI Suppress final-state radiation in double-counted region
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 7 / 24
Matrix element comparisons
I Study how well the parton shower fills the phase spaceI Eventual goal: full matching to 2→ 3 real-emission matrix elementsI Start with a comparison of the first shower emissionI Would changing the shower starting scale give better agreement?
10-1
100
101
102
0 10 20 30 40 50
dσ /
dp⊥
[nb
/ GeV
]
p⊥ [GeV]
(a) p⊥ 3
PSME
10-2
10-1
100
101
102
0 10 20 30 40 50
dσ /
dp⊥
[nb
/ GeV
]
p⊥ [GeV]
(b) p⊥ 4
PSME
10-4
10-3
10-2
10-1
100
101
102
103
0 10 20 30 40 50
dσ /
dp⊥
[nb
/ GeV
]
p⊥ [GeV]
(c) p⊥ 5
PSME
p⊥min3 = 5.0 GeV, p⊥
min5 = 5.0 GeV, Rsep = 0.10
I Good qualitative agreementI Best in soft and collinear regionsI Accuracy degrades when jets are hard and widely separatedI Large region of phase space well describedI No indication for a change in starting scale
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 8 / 24
MPI overview
I MPI framework one of the most important sources of tunable parametersI Regularise cross section with p⊥0 as free parameter
dσdp2⊥∝ α2
s(p2⊥)
p4⊥
→ α2s(p2⊥0 + p2
⊥)
(p2⊥0 + p2
⊥)2
I p⊥0 has energy dependence
p⊥0(ECM) = pref⊥0 ×
(ECM
E refCM
)EpowCM
I Impact parameter, b, with matter profileI Single Gaussian; no free parametersI Overlap function
exp(−bEpow
exp)
I Double Gaussian
ρ(r) ∝ 1− βa3
1exp
(− r 2
a21
)+β
a32
exp(− r 2
a22
)I Many partons produced close in space–time→ colour rearrangement
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 9 / 24
Tevatron tunes
I FSR and hadronisation tuned to LEP data (H. Hoeth)I Identify key parameters and start with by-hand tuneI Rivet for comparisons against dataI Tunes 2C (CTEQ6L1) and 2M (MRST LO**)
b
b
b
b b bbb b b b b b b b b b b b b b b b b b b
bb
b b bb
b
b
CDF datab
PYTHIA 8.142 Tune 2C
PYTHIA 6.422 Pro-Q20
PYTHIA 6.422 Perugia 0
0
0.2
0.4
0.6
0.8
1
Transverse region charged particle density
〈Nch〉/
dη
dφ
0 50 100 150 200 250 300 350 400
0.6
0.8
1
1.2
1.4
pT(leading jet) / GeV
MC
/d
ata
b
b
b
bb b b
b bbb b
bb b
b bb b b b b b b b
b bb b
bb
b
b
b
CDF datab
PYTHIA 8.142 Tune 2C
PYTHIA 6.422 Pro-Q20
PYTHIA 6.422 Perugia 0
0
0.5
1
1.5
2
Transverse region charged ∑ p⊥ density
〈 ∑p
tra
ckT
〉/d
ηd
φ/
GeV
0 50 100 150 200 250 300 350 400
0.6
0.8
1
1.2
1.4
pT(leading jet) / GeV
MC
/d
ata
I Unified MB/UE tune!
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 10 / 24
LHC tunes
I Comparisons against early LHC UE/MB dataI Tevatron tunes give too little activity when compared against LHC dataI Start with Tune 2C and vary only MPI parameters→ Tune 4C
10-7
10-6
10-5
10-4
10-3
10-2
10-1
100
101
102
0 10 20 30 40 50 60 70 80 90 100
1/N
ev d
Nev
/ dN
ch
Nch
ATLAS 7.00 TeV (x 100)Pythia 7.00 TeV (x 100)
ATLAS 900 GeVPythia 900 GeV
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
0 10 20 30 40 50 60 70 80 90 100
<p ⊥
>
[GeV
]
Nch
ATLAS 7.00 TeVPythia 7.00 TeV
ATLAS 900 GeV (-0.5)Pythia 900 GeV (-0.5)
I Gives reasonable agreement with LHC dataI More details: RC, T. Sjostrand, JHEP 1103 (2011) 032.I http://mcplots.cern.ch/ for more plots and generator
comparisons
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 11 / 24
LHC tunes
I Comparisons against early LHC UE/MB dataI Tevatron tunes give too little activity when compared against LHC dataI Start with Tune 2C and vary only MPI parameters→ Tune 4C
0.0
0.5
1.0
1.5
2.0
0 2 4 6 8 10 12 14 16 18 20
<d2 N
ch /
dηdφ
>
pTlead [GeV]
Towardregion
ATLAS 7.00 TeVPythia 7.00 TeV
ATLAS 900 GeVPythia 900 GeV
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
0 2 4 6 8 10 12 14 16 18 20
<d2 N
ch /
dηdφ
>
pTlead [GeV]
Transverseregion
ATLAS 7.00 TeVPythia 7.00 TeV
ATLAS 900 GeVPythia 900 GeV
I Gives reasonable agreement with LHC dataI More details: RC, T. Sjostrand, JHEP 1103 (2011) 032.I http://mcplots.cern.ch/ for more plots and generator
comparisons
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 11 / 24
An x-dependent proton size
I Normally assume PDFs factorise in longitudinal and transverse space
f (x , r) = f (x)ρ(r)
I Contradicts theoretical expectations?I BFKL, Balitsky-JIMWLK, Colour Glass Condensate ...I Mueller’s dipole cascade (e.g. Lund DIPSY, study by Avsar)I Froissart-Martin σtot ∝ ln2 s→ Gribov theory rp ∝ ln(1/x)
I Address this in inelastic non-diffractive eventsRC, T. Sjostrand, JHEP 1105 (2011) 009.
ρ(r , x) ∝ 1a3(x)
exp(− r2
a2(x)
)with a(x) = a0
(1 + a1 ln
1x
)I a1 ≈ 0.15 tuned to rise of σND
(Donnachie & Landshoff + Schuler & Sjostrand)I a0 tuned to value of σND (dependent on PDFs, p⊥0, etc..)
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 12 / 24
An x-dependent proton size
I Tune 4C + x-dependent proton size + lower p⊥0 → Tune 4Cx
1.0
1.5
2.0
2.5
3.0
3.5
4.0
-2 -1 0 1 2
<d2 ∑
p T /
dηdφ
>
[GeV
]
pTlead [GeV]
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
10-6
10-4
10-2
100
102
104
0 10 20 30 40 50 60 70 80 90 100
1/N
ev d
Nev
/ dN
ch
Nch
ATLAS 7 TeV (x 100)Log 7 TeV (x 100)
Overlap 7 TeV (x 100)ATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
0.0
0.5
1.0
1.5
2.0
2.5
0 2 4 6 8 10 12 14 16 18 20
<d2 N
ch /
dηdφ
>
pTlead [GeV]
Towardregion
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
0.0
1.0
2.0
3.0
4.0
5.0
0 2 4 6 8 10 12 14 16 18 20
<d2 ∑
p T /
dηdφ
>
[GeV
]
pTlead [GeV]
Towardregion
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0 2 4 6 8 10 12 14 16 18 20
<d2 N
ch /
dηdφ
>
pTlead [GeV]
Transverseregion
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
0.0
0.5
1.0
1.5
2.0
0 2 4 6 8 10 12 14 16 18 20
<d2 ∑
p T /
dηdφ
>
[GeV
]
pTlead [GeV]
Transverseregion
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
Figure 11: Tune 4C, using the log profile, and with a raised p⊥0 in the MPI framework,compared against an overlap profile with p = 1.6, also with a raised p⊥0, and LHC data
affect the results shown here. Just this change leads to a rise in the tail of the chargedmultiplicity distributions, with an increase in activity in all regions of the underlying event,as expected from the considerations of the previous sections. This behaviour is most closelymatched by an overlap function with p = 1.6, against which we can compare the results.The simplest way to remove this excess activity is a retuning of the p⊥0 parameter of theMPI framework, in this case achieved by raising pref⊥0 = 2.085 → 2.15GeV. This rise doesnot greatly affect the relative slope of a0, as constrained in Sec. 3.1. The results are shownin Fig. 11 for the same distributions as Fig. 10.
After this retuning, the log profile shows some promise. For the charged multiplicity
19
1.0
1.5
2.0
2.5
3.0
3.5
4.0
-2 -1 0 1 2
<d2 ∑
p T /
dηdφ
>
[GeV
]
pTlead [GeV]
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
10-6
10-4
10-2
100
102
104
0 10 20 30 40 50 60 70 80 90 100
1/N
ev d
Nev
/ dN
ch
Nch
ATLAS 7 TeV (x 100)Log 7 TeV (x 100)
Overlap 7 TeV (x 100)ATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
0.0
0.5
1.0
1.5
2.0
2.5
0 2 4 6 8 10 12 14 16 18 20
<d2 N
ch /
dηdφ
>
pTlead [GeV]
Towardregion
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
0.0
1.0
2.0
3.0
4.0
5.0
0 2 4 6 8 10 12 14 16 18 20
<d2 ∑
p T /
dηdφ
>
[GeV
]
pTlead [GeV]
Towardregion
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0 2 4 6 8 10 12 14 16 18 20
<d2 N
ch /
dηdφ
>
pTlead [GeV]
Transverseregion
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
0.0
0.5
1.0
1.5
2.0
0 2 4 6 8 10 12 14 16 18 20
<d2 ∑
p T /
dηdφ
>
[GeV
]
pTlead [GeV]
Transverseregion
ATLAS 7 TeVLog 7 TeV
Overlap 7 TeVATLAS 900 GeV
Log 900 GeVOverlap 900 GeV
Figure 11: Tune 4C, using the log profile, and with a raised p⊥0 in the MPI framework,compared against an overlap profile with p = 1.6, also with a raised p⊥0, and LHC data
affect the results shown here. Just this change leads to a rise in the tail of the chargedmultiplicity distributions, with an increase in activity in all regions of the underlying event,as expected from the considerations of the previous sections. This behaviour is most closelymatched by an overlap function with p = 1.6, against which we can compare the results.The simplest way to remove this excess activity is a retuning of the p⊥0 parameter of theMPI framework, in this case achieved by raising pref⊥0 = 2.085 → 2.15GeV. This rise doesnot greatly affect the relative slope of a0, as constrained in Sec. 3.1. The results are shownin Fig. 11 for the same distributions as Fig. 10.
After this retuning, the log profile shows some promise. For the charged multiplicity
19
I Consistent with minimum-bias and underlying-event dataI Parameter a1 more constrained than other options?
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 13 / 24
An x-dependent proton size
I Differences in the underlying event accompanying hard processesI With e.g. single Gaussian matter profile, Sudakov already saturated at
scales above ∼ 10 GeVI Same impact-parameter profile if hard process has scale 100 GeV or 1 TeV
I With x-dependence, collisions at large x likely to be at small bI Further large-to-medium-x MPIs are enhanced
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
0 0.5 1 1.5 2 2.5
(1 /
N)
dN
/ d
bM
PI
norm
bMPInorm
(a)
SGDYZ
0
Z’
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 14 / 24
An x-dependent proton size
I Differences in the underlying event accompanying hard processesI With e.g. single Gaussian matter profile, Sudakov already saturated at
scales above ∼ 10 GeVI Same impact-parameter profile if hard process has scale 100 GeV or 1 TeV
I With x-dependence, collisions at large x likely to be at small bI Further large-to-medium-x MPIs are enhanced
0
0.02
0.04
0.06
0.08
0.1
0 5 10 15 20 25
(1 /
N)
dN
/ d
NM
PI
NMPI
(a) DY
SGLog
0
0.02
0.04
0.06
0.08
0.1
0.12
0 5 10 15 20 25
(1 /
N)
dN
/ d
NM
PI
NMPI
(c) Z’
SGLog
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 14 / 24
CKKW-L merging
I Consistently combine higher-order tree-level matrix elementsand parton shower description
I Leif Lonnblad and Stefan Prestel working to include CKKW-L mergingI Reconstruct a shower history for the ME stateI Trial showers to get Sudakov factorI Consistent inclusion of interleaved MPI
0 20 40 60 80
100 120 140 160
10 20 30 40 50 60 70 80
Devia
tion [%
]
k⊥ 2
(GeV)
ME4PS 15ME1PS 30ME2PS 30ME3PS 30ME4PS 30
1.0⋅10-11
1.0⋅10-10
1.0⋅10-9
1.0⋅10-8
1.0⋅10-7
1.0⋅10-6
dσ
/dk
⊥ 2
[m
b/G
eV
]
PYTHIA8ME
31PS tMS=30 GeV
ME31PS tMS=30 GeV
ME32PS tMS=30 GeV
ME33PS tMS=30 GeV
ME3PS tMS=30 GeV
0 10 20 30 40 50 60 70
0 10 20 30 40 50
Devia
tion [%
]
k⊥ 1
(GeV)
(ckkw-l) / PYTHIA8(alternative) / PYTHIA8
1⋅10-9
2⋅10-9
4⋅10-9
8⋅10-9
2⋅10-8
3⋅10-8
6⋅10-8
dσ
/dk
⊥ 1
[m
b/G
eV
]
PYTHIA8ME2PS (ckkw-l)
ME2PS (alternative)
I Article + code coming soon!
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 15 / 24
BSM physics
I SLHA interface updated; QNUMBERS now supportedI Calculation of SUSY decay widths (N. Desai)
I R-hadronisation now integrated into PYTHIA 8I Long-lived coloured particles e.g. g or t1 → R-hadronsI R-mesons (gqq, t1q), R-baryons (gqqq, t1qq) or glueballs (gg)I A.C. Kraan, Eur. Phys. J. C37 (2004) 91
M. Fairbairn et al., Phys. Rep. 438 (2007) 1CMS, arXiv:1101.1645
I Hidden valley/secluded sector frameworkI New gauge groups at low energy scales, hidden by potential
barrier or weak couplings (Strassler & Zurek)I Can pick Abelian U(1) or non-Abelian SU(N) gauge groupI Different production mechanisms (e.g. massize Z′, kinetic mixing, ...)I Interleaved shower in QCD, QED and HV sectorsI Hadronisation in hidden SU(N) sectorI L. Carloni & T. Sjostrand, JHEP 09 (2010) 105
L. Carloni, J. Rathsman & T. Sjostrand, JHEP 04 (2011) 091
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 16 / 24
Hadron spectra
I Know that certain hadron flavours need more p⊥I Latest π/K/p p⊥ spectra from LHC
I ALICE 900 GeV - inelastic events, |y | < 0.5, arXiv:1101.4110v3I ALICE 7 TeV - read from public talks
(assume same acceptance as 900 GeV)I Compare against PYTHIA 8
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.5 1 1.5 2 2.5
(K+ +
K- )
/ (π
+ +
π- )
p⊥ [GeV]
ALICE 900 GeVALICE 7 TeV
PYTHIA 900 GeVPYTHIA 7 TeV
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.5 1 1.5 2 2.5
(p+ +
p- )
/ (π
+ +
π- )
p⊥ [GeV]
ALICE 900 GeVALICE 7 TeV
PYTHIA 900 GeVPYTHIA 7 TeV
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 17 / 24
Final-state hadron scattering
I Final-state hadron scatteringI No need to significantly change particle composition,
only (partial) collective flow?I Low-energy scatterings in CM frame and boost backI Higher masses take bigger kick, c.f. collective flow in heavy ionI Ideally assign production vertices to outgoing hadrons and follow path
I Simple model based on distance in y − φ spaceI High-p⊥ hadrons in jets formed at later times and less likely to scatterI Scattering probability
Pij = (1− e−kσelij (s)) max
(0, 1−
∆R2ij
R2max
)I Order scatterings based on “relative transverse velocity”
v⊥ij =
∣∣∣∣ p⊥i
m⊥i− p⊥j
m⊥j
∣∣∣∣I Dominated by pions, so focus on ππ, πK and πpI Most scatterings close to threshold
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 18 / 24
Final-state hadron scattering
I Isospin partial-wave parameterisations (no Columb corrections)
0
20
40
60
80
100
120
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
σel
[m
b]
ECM [GeV]
π+π
-
π+π
0
π+π
+
0
10
20
30
40
50
60
70
80
0.6 0.8 1 1.2 1.4 1.6 1.8 2
σel
[m
b]
ECM [GeV]
π+K
-
π0K
+
π+K
+
0
50
100
150
200
250
1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6
σel
[m
b]
ECM [GeV]
π+p
π0p
π-p
0
20
40
60
80
100
120
-1 -0.8-0.6-0.4-0.2 0 0.2 0.4 0.6 0.8 1
dσ
el /
d c
os(θ
)
cos(θ)
1.0 * 0.40 GeV0.5 * 0.77 GeV1.0 * 1.10 GeV
0
10
20
30
40
50
60
70
80
-1 -0.8-0.6-0.4-0.2 0 0.2 0.4 0.6 0.8 1
dσ
el /
d c
os(θ
)
cos(θ)
1.0 * 0.80 GeV0.5 * 0.89 GeV1.0 * 1.10 GeV
0
10
20
30
40
50
60
70
-1 -0.8-0.6-0.4-0.2 0 0.2 0.4 0.6 0.8 1
dσ
el /
d c
os(θ
)
cos(θ)
1.0 * 1.10 GeV0.3 * 1.23 GeV1.0 * 1.40 GeV
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 19 / 24
Final-state hadron scattering
I Results from simple model; try to boost K/π ratioI Perform scattering after first round of hadron decays (η → π)
900 GeV
0
0.1
0.2
0.3
0.4
0.5
0 0.5 1 1.5 2
(K+ +
K- )
/ (π
+ +
π- )
p⊥ [GeV]
ALICEPYTHIA Default
PYTHIA HS
0
0.1
0.2
0.3
0.4
0.5
0 0.5 1 1.5 2(p
+ +
p- )
/ (π
+ +
π- )
p⊥ [GeV]
ALICEPYTHIA Default
PYTHIA HS
I Exact shape depends on how “soft” hadrons are pickedI Protons get too much of a kick
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 20 / 24
Final-state hadron scattering
I Results from simple model; try to boost K/π ratioI Perform scattering after first round of hadron decays (η → π)
7 TeV
0
0.1
0.2
0.3
0.4
0.5
0 0.5 1 1.5 2
(K+ +
K- )
/ (π
+ +
π- )
p⊥ [GeV]
ALICEPYTHIA Default
PYTHIA HS
0
0.1
0.2
0.3
0.4
0.5
0 0.5 1 1.5 2(p
+ +
p- )
/ (π
+ +
π- )
p⊥ [GeV]
ALICEPYTHIA Default
PYTHIA HS
I Exact shape depends on how “soft” hadrons are pickedI Protons get too much of a kick
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 20 / 24
Conclusions
I PYTHIA 6 is still supported, but no longer actively developed
I PYTHIA 8 is the natural successorI Number of new features available only in PYTHIA 8 continuing to growI Starting to have competitive tunesI Welcome other tuning efforts (and reports on the outcome!)I Provide feedback; both negative and positive!
I Announcement list:http://www.hepforge.org/lists/listinfo/pythia8-announce
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 21 / 24
Backup slides
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 22 / 24
An x-dependent proton size
I Time integrated overlap
O(b; x1, x2) =1π
1a2(x1) + a2(x2)
exp(− b2
a2(x1) + a2(x2)
)I Define n(b) as average number of interactions at b
n(b) =∑i,j
∫∫∫dx1 dx2 dp2
⊥ fi(x1,p2⊥) fj(x2,p2
⊥)dσij
dp2⊥
∣∣∣∣reg
O(b; x1, x2)
I Such that
σhard =
∫n(b) d2b
σND =
∫Pint d2b =
∫ (1− e−n(b)
)d2b
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 23 / 24
Final-state hadron scattering
I Probability that a hadron is soft enough to scatter:
Psoft =N exp
(− p2
⊥4σ2
frag
)(1− k)exp
(− p2
⊥4σ2
frag
)+ k pp
⊥0
(p2⊥0+p2
⊥)p/2
I N ∼ 1, k ∼ 0.5 and p ∼ 6 for results
I Isospin partial-wave parameterisations:I ππ - Froggatt and Petersen, Nucl. Phys. B129 (1977) 89-110I πK - Estabrooks et. al., Nucl. Phys. B133 (1978) 490-524I πN - GWU SAID WI08 solution
SP06 - Phys. Rev. C74 045205 (2006)WI08 - http://gwdac.phys.gwu.edu/analysis/pin analysis.html
Richard Corke (Lund University) QCD@LHC St. Andrews 2011 August 2011 24 / 24