Collectivity in EPOS

Transcription

1 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 1 Collectivity in EPOS (flow, non-flow, and parton saturation) Klaus Werner in collaboration with T. Pierog, Y. Karpenko, B. Guiot, G. Sophys

2 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 2 To understand collectivity in small systems we should understand the evolution of collective phenomena from pp to pa to AA even better its evolution as a function of multiplicity (the generalization of the concept of centrality in AA) from low multiplicity pp to central AA

3 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 3 Many aspects of collectivity... Here : Radial flow, mean pt Hadron chemistry (statistical production or string decay) Not in this talk: Asymmetries

4 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 4 Particle ratios vs dnch dη (0) for pp, ppb, PbPb Average transverse momenta vs for pp, ppb, PbPb dnch dη (0) Charmed meson production vs dn ch (0) for pp dη dnch dη (0) for multiplicity classes defined via forward multiplicities

5 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 5 First step to get a global view, to see where EPOS works and where not, in pp, pa, AA, same version Status 2015: Two parallel developments EPOS LHC: Gribov Regge approach, parameterized flow as in EPOS1.99, tuned to LHC data (2012), very much used (and tested) by LHC pp groups, UE, forward physics etc, and used for air shower simulations EPOS 3.0xx: Gribov Regge approach, viscous hydro, parton saturation, mainly used for HI and collectivity in pp 2015/2016/2017: Fusion, to accommodate basic pp and HI features, public version; Currently: EPOS3.2xx

6 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 6 Particle ratios to pions vs dn ch dη (0) ratios to π K p Λ Ξ x3 ALICE data circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV) 10-3 x8 Ω <dn ch /dη(0)> Refs: next slide

7 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 7 Mean p t vs dn ch dη (0) <p t > <p > t 1 Ω Ξ Λ Κ π x2 x1.5 ALICE data <dn ch /dη(0)> circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV) Data partly collected by A. G. Knospe Refs: <dnch/deta> in Pb+Pb: Phys. Rev. Lett (2011) pi+-, K+-, and (anti)protons in Pb+Pb: Phys. Rev. C (2013) Lambda in Pb+Pb: Phys. Rev. Lett (2013) Xi- and Omega in p+pb: Phys. Lett. B (2016) pi+-, K+-, (anti)protons, and Lambda in p+pb: Phys. Lett. B (2014) <dnch/deta> in p+pb: Eur. Phys. J. C (2016) Xi- and Omega in p+pb: Phys. Lett. B (2016) <dnch/deta> in p+p 7 TeV: Eur. Phys. J. C (2010) pi+-, K+-, and (anti)protons in p+p 7 TeV: Eur. Phys. J. C (2015) Xi- and Omega in p+p 7 TeV: Phys. Lett. B (2012) and data points from Rafael Derradi de Souza, SQM2016

8 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 8 D or J/Ψ multiplicity vs dn ch (0) in pp dη dn D /dy(0) / <dn D /dy(0)> MB p ALICE pp 7TeV t GeV/c diagonal dn ch /dη(0) / <dn ch /dη(0)> MB dn J/Ψ /dη(0) / <dn J/Ψ /dη(0)> MB pp 200GeV data STAR J/Psi p t >4GeV/c dn ch /dη(0) / < dn ch /dη(0) > MB ALICE JHEP 09 (2015) 148, STAR, shown at MPI2016 arxiv: v1 strongly nonlinear increase

9 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 9 Core-corona picture in EPOS (details later) Gribov-Regge approach => (Many) kinky strings => core/corona separation (based on string segments) central AA peripheral AA high mult pp,pa low mult pp core => hydro => flow + statistical decay corona => string decay

10 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 10 Pion yields: core / corona contribution yield EPOS π full co+co corona core thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeV) full = with hadronic cascade (UrQMD) <dn ch /dη(0)>

11 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 11 Proton to pion ratio ratio to π EPOS p ALICE (black) full co+co corona core <dn ch /dη(0)> core hadronization: T = 164MeV, µ B = 0 statistical model fit (horizontal black line) A. Andronic et al., arxiv: T = 156.5MeV,µ B = 0.7MeV thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

12 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 12 Omega to pion ratio ratio to π 10-2 EPOS ALICE (black) Ω(x8) 10-3 full co+co corona core <dn ch /dη(0)> thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

13 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 13 Xi to pion ratio ratio to π 10-2 EPOS ALICE (black) Ξ(x3) 10-3 full co+co corona core <dn ch /dη(0)> thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

14 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 14 Lambda to pion ratio ratio to π 10-1 EPOS Λ ALICE (black) 10-2 full co+co corona core <dn ch /dη(0)> thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

15 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 15 Kaon to pion ratio ratio to π EPOS K ALICE (black) full co+co corona core <dn ch /dη(0)> thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

16 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 16 Ratios h/π for h = p,k,λ,ξ,ω vs dn dη (0) : Core and corona contributions separately roughly constant Difference (core - corona) increasing for p K,Λ Ξ Ω => inceasing slope

17 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 17 Average p t of protons <p t > EPOS p ALICE (black) 1 full co+co corona core <dn ch /dη(0)> thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

18 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 18 Average p t of Omegas <p t > EPOS Ω ALICE (black) 1 full co+co corona core <dn ch /dη(0)> thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

19 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 19 Average p t of lambdas <p t > EPOS Λ ALICE (black) 1 full co+co corona core <dn ch /dη(0)> thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

20 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 20 Average p t of kaons <p t > 1 EPOS K ALICE (black) full co+co corona core <dn ch /dη(0)> thin lines = pp (7TeV) intermediate lines = ppb (5TeV) thick lines = PbPb (2.76TeVVV) circles = pp (7TeV) squares = ppb (5TeV) stars = PbPb (2.76TeV)

21 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 21 Average p t of K,p,Λ,Ξ,Ω vs dn dη (0) : Moderate increase of core contribution (same for pp and ppb, similar to PbPb) Strong increase of corona contribution (stronger for pp than for ppb, much stronger than for PbPb) Slope(pp) > slope(ppb) >> slope(pbpb) K, π : pp-ppb splitting The multiplicity dependence of the corona contribution is crucial

22 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 22 Why such a strong mean pt increase with multiplicity for corona particles? EPOS: Gribov-Regge approach nucleon nucleon S-Matrix based on Pomerons Pomerons : Parton ladders (initial and final state radiation, DGLAP) Cutting rules to get inelastic cross sections. Same principle for AA

23 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 23 Explicite formulas for cross sections (even partial cross sections) σ tot = cutp uncut P A cut G (squared amplitude) uncut G = nucleon nucleon B } {{ } dσ exclusive => kinky strings

24 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 24 Based on string segments: core-corona separation central AA peripheral AA high mult pp low mult pp core: string segments melt => fluid corona: strings survive (ordinary kinky strings from parton ladders)

25 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 25 Parton-ladders (1) are perfectly fitted (2) as G = α(x + x ) β. G depends on the vituality cutoff: G = G(Q 0 ). To mimic the effects of gluon fusion, the fits are modified (for pp) as α(x + x ) β+ε, referred to as G eff. The exponent ε = ε(s) is chosen to reproduce the energy dependence of cross sections. nucleon ladder partons Procedure employed in EPOS LHC (1) Imaginary part G of the corresponding amplitude in b space (2) x +,x : light cone momentum fractions of the Pomeron end

26 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 26 But adding an exponent ε must be accompanied by a corresponding modification of the internal structure of the Pomeron This can be done by defining a saturation scale Q s via G eff = A (N Pom ) B G(Q s ) and then considering the parton ladder with the cutoff Q s (thus changing the internal structure! => consistent!) We find Q s = Q s (x + x ) (x + x ) 0.30

27 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 27 Parton distributions dn/dp t small Q s large Q s Increasing dn/dη(0) corresponds to increasing N Pom => Increasing Q s => harder Pomerons => harder strings => p t => more high pt particles => Strong increase of p t with dn/dη(0)

28 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 28 Very closely related to this discussion: The multiplicity dependence of charm production (D, J/Ψ,...) The ultimate tool to test multiple scattering (and the implementation of Q S )

29 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 29 EPOS 3 compared to ALICE data dn D /dy(0) / <dn D /dy(0)> MB p ALI EPOS t CE full GeV/c diagonal dn ch /dη(0) / <dn ch /dη(0)> MB hadronic cascade on/off has no effect hydro on/off has small effect

30 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 30 EPOS 3 compared to RHIC data dn D /dη(0) / <dn D /dη(0)> MB EPOS pp 200GeV data STAR J/Psi p t >4GeV/c Calculations: D mesons Data: J/Ψ Increase stronger than at LHC dn ch /dη(0) / < dn ch /dη(0) > MB

31 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 31 Multiplicity at FB rapidity (LHC) dn D /dy(0) / <dn D /dy(0)> MB p ALI EPOS t CE full GeV/c diagonal dn ch /dη(fb) / <dn ch /dη(fb)> MB FB = forward/backward rapidity range: 2.8 < η < 5.1 and 3.7 < η < 1.7 Smaller increase

32 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 32 Low multiplicity (LM) Small N Pom High multiplicity (HM) many hard IP s on avg few soft IP s (A) more IP s, but less hard IP = Pomeron Hardness increases with N Pom (larger Q s ) (B) fewer IP s, but harder

33 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 33 LM HM: Pomerons get harder (larger Q s ) favors high pt or large mass production in particular due to case B (fewer IP s, but harder) for highest pt bins! Bigger effect at RHIC due to much narrowern Pom distribution (harder IP s are needed) Smaller effect for dn (FB) as multipl. variable dη (case B is replaced by case C: fewer IP s, but more covering the FB rapidity range)

34 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 34 Summary To understand flow in small systems, we have to understand the non-flow part ( corona ). The latter one dominates low multiplicity pp, but its relative weight decreases continuously with multiplicity (but is never zero) Investigating the multiplicity dependence of particle ratios and mean pt in pp, pa: EPOS s core-corona picture describes the trend Strong increase of corona pt due to the N Pom dependence of the saturation scale... which explains also the strong lonlinearity of the D (J\Psi) multiplicity vs charged one

35 9-11 May 2017 Niels Bohr Institute Klaus Werner Subatech Nantes 35 Core => Hydro evolution (Yuri Karpenko) Israel-Stewart formulation, η τ coordinates, η/s = 0.08, ζ/s = 0 ;ν T µν = ν T µν +Γ µ νλ Tνλ +Γ ν νλ Tµλ = 0 γ( t +v i i )π µν = πµν π µν NS +I π µν τ π γ( t +v i i )Π = Π Π NS τ Π +I Π T µν = ǫu µ u ν (p + Π) µν + π µν, ;ν denotes a covariant derivative, µν = g µν u µ u ν is the projector orthogonal to u µ, π µν, Π shear stress tensor, bulk pressure π µν NS = η( µλ ;λ u ν + νλ ;λ u µ ) 2 3 η µν ;λ u λ Π NS = ζ ;λ u λ I µν π = 4 3 πµν ;γu γ [u ν π µβ + u µ π νβ ]u λ ;λ u β I Π = 4 3 Π ;γuγ Freeze out: at 164 MeV, Cooper-Frye E dn d 3 p = dσ µ p µ f(up), equilibrium distr Hadronic afterburner: UrQMD Marcus Bleicher, Jan Steinheimer