Rope Hadronization, Geometry and Particle Production in pp and p A - - PowerPoint PPT Presentation

Rope Hadronization, Geometry and Particle Production in pp and pA Collisions

Christian Bierlich

Advisors: G¨

• sta Gustafson, Leif L¨
• nnblad, Torbj¨
• rn Sj¨
• strand.

Other collaborators: Jesper Roy Christiansen, Andrey Tarasov.

Lund University

Jan 27, 2017 Defense of Thesis for the degree of Doctor of Philosophy

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 1 / 21

Introduction

Quarks and gluons are the building blocks of protons and neutrons. They are ”confined” by the strong nuclear force. The Quark Gluon Plasma (QGP) is a hypothesized state of deconfined quarks and gluons existing at high pressure/temperature. QGP investigated in heavy ion experiments since 1980s. Recent data from LHC has revealed QGP–like behaviour in pp and pA. Question: Can ”QGP–effects” be modeled without assuming a thermalized liquid?

1

Find a method for extrapolating pp to pA and AA.

2

Develop a microscopic description of small system QGP behaviour.

3

Combine and compare to data.

4

Compare to existing, macroscopic predictions.

This thesis is concerned with the two first steps.

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 2 / 21

Overview

DIPSY and pp extrapolations (arXiv:1607.04434 [hep-ph]).

1

Geometry of a pA collision.

2

Parametrization of colour fluctuations.

3

Particle production and comparison to data.

The Rope Hadronization model (arXiv:1412.6259 [hep-ph] and arXiv:1507.02091 [hep-ph]).

1

Corrections to hadronization in dense environments.

2

Effects on strangeness.

3

Comparison to data.

String shoving (arXiv:1612.05132 [hep-ph]).

1

”The ridge”.

2

Pressure from string overlaps.

3

Results.

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 3 / 21

Extrapolation from pp to pA

Big Picture goal: A general purpose MC event generator for HI built

• n microscopic models only.

Obvious difference from small → large systems: Geometry. A good description of basic event properties is essential. We need space–time description of event structure. Extrapolation = geometry + colour fluctuations (CF). CF based on the DIPSY initial state model. Result: A model for extrapolating Pythia pp to pA events. From CB, Gustafson and L¨

• nnblad, arXiv:1607.04434 [hep-ph].

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 4 / 21

Glauber models and the wounded cross section

Reproduce ”centrality” ∝ forward particle production Standard Glauber, absorptive channels only: dσw d2b = dσabs d2b = 2 T(b) − T(b)2 Diffraction contributes in the forward direction. Wounded nucleons updated with CF (SD + DD in Good–Walker). dσw d2b = dσabs d2b +dσSD,t d2b +dσDD d2b = 2 Tp,t −

• T2

t

• p .

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 5 / 21

Parametrization of colour fluctuations

Pioneered in Glauber-Gribov formalism (Alvioli and Strikman: arXiv:1301.0728 [hep-ph]): Much faster than full DIPSY – feasible to do HI simulation. Modified parametrization: Fitted with pp data only.

Ptot(σ) = σ2 σ + σ0 exp

• −

(σ/σ0 − 1)2 Ω2

• → Ptot(σ) =

1 Ω √ 2π exp

• −

log2(σ/σ0) 2Ω2

• 50

100 150 200

σ [mb]

0.000 0.005 0.010 0.015 0.020 0.025 0.030

Pwinc(σ)

DIPSY GG Ω =0.37 GG Log-normal Ω =0.25 GG Log-normal Ω =0.33

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 6 / 21

Types of wounded nucleons

We obtain:

1

The number of wounded nucleons incl. diffractive excitation.

2

Given a T(b) assumption, which are which!

We now have input to a model for particle production – FritiofP8.

10 20 30 40 50 60

N t

w inc

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

10

P(Nw)

Black Disk GG Ω =0.82 GG Log-normal Ω =0.43 2x2 model

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 7 / 21

Results (Data: ATLAS)

Very good agreement with centrality observable. ”Absorptive” overshoots. Measuring the exact region where diffractive excitation is important.

20 40 60 80 100 120 140 160 180

ΣE ⟂ [GeV]

10-6 10-5 10-4 10-3 10-2 10-1

dN/(NdΣE ⟂) [GeV−1 ]

Sum E ⟂, −4.9 <η <−3.2, p ⟂ >0.1 GeV FritiofP8 Absorptive DIPSY

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 8 / 21

Multiplicity Data: ATLAS

Reproducing central collisions well. Does better than DIPSY in central collisions. Future: Implementation by ATLAS would be better.

b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b

Data

b

FritiofP8 Absorptive DIPSY 20 40 60 80 100 Centrality 0-1% dN/dη

• 2
• 1

1 2 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 MC/Data

b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b

Data

b

FritiofP8 Absorptive DIPSY 10 20 30 40 50 60 Centrality 10-20% dN/dη

• 2
• 1

1 2 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 MC/Data

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 9 / 21

What to do in busy events?

Events as just described are quite busy. Strings are fluxtubes i.e. confined fields. Interference in overlap regions must be treated. From CB, Gustafson, L¨

• nnblad and Tarasov, arXiv:1412.6259

[hep-ph] and CB and Christiansen, arXiv:1507.02091 [hep-ph].

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 10 / 21

Experimental motivation

Hadronic flavour description works for e+e− (Data: SLD, LEP and PDG avg.). Not even inclusively in pp (Data: ATLAS, CMS, ALICE and LHCb).

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Ratio of integrated yields

Data DIPSY Rope DIPSY PYTHIA 8

p/π K/π φ/K Λ/K0

s

Ξ/Λ Ω/Ξ 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 MC/data

LEP 91.2 GeV

0.0 0.1 0.2 0.3 0.4 0.5 0.6 Ratio of integrated yields

Data Dipsy Pythia 8 Def.

p/π K ± /π Λ/K 0

s

Ξ/Λ Ω/Ξ 0.2 0.4 0.6 0.8 1.0 1.2 1.4 MC/data

7000 GeV

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 11 / 21

String Hadronization (See e.g. hep-ph/0603175)

Non-perturbative phase of final state. Confined colour fields ≈ strings with tension κ ≈ 1 GeV/fm. Breaking/tunneling with P ∝ exp

• − πm2

⊥

κ

• gives hadrons.

Longitudinal components from: f (z) ∝ z−1(1 − z)a exp −bm2

⊥

z

• .

a and b related to total multiplicity. Flavours determined by relative probabilities: ρ = Pstrange Pu or d , ξ = Pdiquark Pquark Probabilities are related to κ via tunneling equation.

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 12 / 21

Dipole coherence effects

Overlapping gives coherence effects. The simplest example: Two q¯ q pairs act coherently. Two distinct possibilities:

c1 ¯ c1 c2 ¯ c2 r ⊕ r ¯ r ⊕ ¯ r Case (a), c1 = c2 : Case (b), c1 = c2 : r ⊕ b ¯ r ⊕ ¯ b ¯ g g Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 13 / 21

Rope formation

Multiplet structure from SU(3) Random Walk procedure (Biro et al.

Nucl.Phys.B 245 (1984) 449–468)

Highest multiplet gets larger effective string tension: κ → ˜ κ = hκ from number of overlapping strings. Calculable as secondary Casimir operator of multiplet. κ ∝ C2 ⇒ h = ˜ κ/κ = C2(multiplet) 1 GeV/fm Confirmed on the lattice, static case (Bali: arXiv:hep-lat/0006022).

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 14 / 21

Effect on hadronization parameters

Strange quark breakup supression: ρ = exp

• −π(m2

s − m2 u)

κ

• .

1 2 3 4 5 6 7 h (Enhancement of string tension) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Effective parameter

˜ ρ ˜ ξ ˜ a ˜ b

Large effect on hadronic flavours. Baryons also affected by junctions. Smaller effect on hadron p⊥ and multiplicity (tunable).

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 15 / 21

Effect in pp I (Data: STAR and CMS)

Improvement inclusively. Tail of p⊥ spectrum not fully understood. Linked to ”flow” effects. Better observable which isolates strangeness and baryons.

b b b b b b b b b b b b b b b b b b b b

Data

b

Rope DIPSY Pythia 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 p/π+ Ratio p⊥ distribution √s = 200 GeV N(p)/N(π+) 1 2 3 4 5 6 7 0.6 0.8 1 1.2 1.4 p⊥ [GeV] MC/Data

b b b b b b b b b b b b b b b b b b b b b b b b

Data

b

Rope DIPSY Pythia

• 0.2

0.2 0.4 0.6 0.8 1 1.2 1.4 Λ/K0

S versus transverse momentum at √s = 7000 GeV

N(Λ) / N(K0

S)

2 4 6 8 10 0.6 0.8 1 1.2 1.4 pT [GeV/c] MC/Data

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 16 / 21

Effect in pp II (ALICE: arXiv:1606.07424 [nucl-ex])

Strangeness enhancement for central events. Signal linked to Quark Gluon Plasma in Heavy Ion Physics. Not shown: (lack of) baryon enhancement in data.

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 17 / 21

Shoving: An experimental motivation (Data: CMS)

Ridges linked to flow seen in AA, pA and pp. Very well described by hydrodynamics. From CB, Gustafson and L¨

• nnblad, arXiv:1612.05132 [hep-ph].

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 18 / 21

Shoving: A microscopic model for expansion

Overlapping regions generate a transverse pressure, ”shoving” the strings apart.

(Abramovsky et al. Pisma Zh. Eksp. Teor. Fiz. 47, 281 (1988)). t = t1 t = t2 t = t3 t = t4 by bx

In each time–step dt, each string will get a kick from other strings: dp⊥ dydt = C0td R2 exp

• − d2

2R2

• .

Momentum conservation is observed: Transverse kicks resolved pairwise, p± recoil in kicking dipole.

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 19 / 21

Two–particle correlations

Shoving produces a ”ridge”. Currently for events consisting of long, soft strings only. Working towards a complete description.

−2 −1 1 2 3 4 5

∆φ

−0.05 0.00 0.05 0.10 0.15 0.20

S(∆η,∆φ)/B(∆η,∆φ)

2 < ∆η < 4 0.5 GeV < p ⟂ < 3 GeV

No Shove

Peripheral Central

−2 −1 1 2 3 4 5

∆φ

−0.05 0.00 0.05 0.10 0.15 0.20

S(∆η,∆φ)/B(∆η,∆φ)

2 < ∆η < 4 0.5 GeV < p ⟂ < 3 GeV

Shove and Rope

Peripheral Central Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 20 / 21

Conclusions

Rope hadronization is a microscopic model for soft-QCD collectivity.

1

Good results for strangeness.

2

Baryons are outstanding problem.

3

Promising results for The Ridge.

4

Implemented in the DIPSY generator.

FritiofP8 is a promising method for extrapolation of pp to pA and maybe AA.

1

Multiplicity well reproduced.

2

High-p⊥ less well, PDFs bring large uncertainty.

3

Fast enough to do AA (as opposed to DIPSY).

4

Implemented on top of Pythia8.

Future perspective

1

Putting it all together!

2

The smallest system: e+e−.

3

High-p⊥ observables (Jet quenching, RAA).

Thank you for your attention!

Christian Bierlich (Lund) Ropes and Particles Jan 27, Thesis defense 21 / 21