Collective Dynamics of the p+pb Collisions Wojciech Broniowski CEA Saclay & UJK Kielce & IFJ PAN Cracow th Workshop on Non-Perturbative QCD Paris, - June [Piotr Bo»ek & WB, PLB 78 () 557, 7 () 5, arxiv:.]
Introduction Signatures of sqgp Main signatures of sqgp in ultra-relativistic A+A collisions: Collective ow Jet quenching Flow manifests itself in harmonic components in the momentum spectra, certain features in correlation data (ridges), interferometry (femtoscopy),... Bozek : p+a and p+p in hydro Ridges discovered in small systems, p+a and high-multiplicity p+p
Introduction -stage approach Our three-phase approach: initial hydro statistical hadronization Initial phase - Glauber model Hydrodynamics - + D viscous event-by-event Statistical hadronization Main questions: Are the central p-pb collisions collective? What is the nature of the initial state? What are the limits/conditions on applicability of hydrodynamics?
-stage approach Snapshots of initial Glauber condition in central p+pb Typical transverse-plane conguration of centers of the participant nucleons in a p+pb collision generated with GLISSANDO 5% of collisions have more than 8 participants, rms.5 fm quite large! p-pb, N =8 W y y - - - - - - x - - x
-stage approach Snapshot of peripheral Pb+Pb Most central values of N w in p-pb would fall into the 6-7% or 7-8% centrality class in Pb+Pb Pb+Pb: c=6-7% N w, c=7-8% N w Pb+Pb, N w =8 y 5 - - - - -5-5 - - - - 5 x in Pb+Pb somewhat larger size than in p+pb
-stage approach Smearing Gaussian smearing with width. fm (physical eect) standard compact This is fed into e-by-e hydro as initial condition
-stage approach Size in p+pb p+pb, N w = 8 7 distribution 6 5 p Pb compact p Pb standard..5..5..5 r fm red - centers of participants, blue - center-of-mass of colliding pairs
-stage approach Size in p+pb vs Pb+Pb xed N w = 8 7 distribution 6 5 p Pb compact p Pb standard Pb Pb..5..5..5 r fm smaller size in p+pb larger entropy density more rapid expansion All in all, initial conditions in most central p+pb not very far from peripheral Pb+Pb
-stage approach Multiplicity distribution To reproduce the multiplicity distribution of the most central events in p+pb one needs to fluctuate the strength of the Glauber sources. We overlay the Gamma distribution (Gamma + Poisson = negative binomial). At statistical hadronization Poissonian fluctuations are generated y.5.5.5 -.5 - -.5 - p+pb, N =8, overlaid Γ w distribution -.5 -.5 - -.5 - -.5.5.5.5 x.5.5.5.5 P N... x x x CMS Tracks Glauber NB Glauber Poisson x x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx x xxxxx x xxxx x xxx 5 5 5 N
-stage approach Assumed factorization of the transverse and longitudinal distributions alignment of F and B event planes (can be checked experimentally) collimation of ow at distant longitudinal separations ridges!
collectivity in p-pb -stage approach Surfers - the near-side ridge
-stage approach Hydrodynamics [Bo»ek ] + D viscous event-by-event hydrodynamics τ init =.6 fm/c, η/s =.8 (shear), ζ/s =. (bulk) freezeout at T f = 5 MeV average initial temperature in the center of the reball T i = MeV (< R > / =.5 fm), or 9 MeV (< R > / =.9 fm) adjusted to t multiplicity realistic equation of state (lattice + hadron gas [Chojnacki & Florkowski 7]), viscosity necessary for small systems lattice spacing of.5 fm (thousands of CPU hours)
-stage approach < R > / =.5 fm < R > / =.9 fm 5 ppb 5GeV N part 9 5 ppb 5GeV N part 9 Τ fm c Τ fm c x y fm x y fm isotherms at freeze-out T f = 5 MeV for two sections in the transverse plane evolution lasts about fm/c - shorter but more rapid than in A+A
-stage approach Statistical Hadronization Statistical hadronization via Frye-Cooper formula + resonance decays (THERMINATOR), transverse-momentum conservation approximately imposed, local charge conservation included
Flow LHC: v vs ATLAS. p-pb 5.TeV ATLAS v v. {} v {} v.8 {PC}.6.. v {} hydro Glauber v {} hydro Glauber+NB v {} hydro Glauber+NB 6 8 Σ E GeV
Flow v {} and v {} vs p T v.. p-pb 5.TeV -% η/s=.8 τ η/s=.6 / <R > =.9fm τ =.fm/c =.6fm/c <R > / =.5fm. v.5. p-pb 5.TeV -% η/s=.8 τ η/s=.6 / <R > =.9fm τ =.fm/c Glauber+NB =.6fm/c <R > / =.5fm.5 Glauber+NB.5.5 p [GeV].5.5 p [GeV]
Flow v and v from the scalar-product method [STAR, Luzum & Ollitrault ]. v n.5. v v p side EP Pb side EP combined EP p-pb 5.TeV -%.5.5.5 p [GeV]
Flow v and v cuts: η <.5,. < p T < 5 GeV < R > / =.5 fm c=-.% c=.-7.8% Glauber+Poisson v {} [ ].7().78() v {} [ ].().95() v {} [ 6 ] -.().8(5) Glauber+NB v {} [ ] 8.8() 8.() v {} [ ].5(8).5(6) v {} [ 6 ] 5(7) 6(6) more uctuations more harmonic ow v {} very sensitive (uctuations)
Correlations Denition of the D correlation function C( η, φ) = N pairs phys ( η, φ) = S( η, φ) N pairs mixed ( η) B( η, φ) (more convenient than the per-trigger correlations)
Correlations Ridge in p-pb, ATLAS Pb Σ E T < GeV ATLAS - L µb (a) p+pb =5. TeV s NN a,b.5<p < GeV T Pb ΣE T >8 GeV (b) C( φ, η). C( φ, η). φ η - - φ η - - c=-.%,.5<p < GeV T C( η, φ).....99.98 5 - - - - -5 - η φ 5
Correlations Projection on η 5, ATLAS B( φ)d( φ) Y ( φ) = C( φ) b ZYAM N The near-side ridge from our model: Y( φ).6.5. c=-.%.5 < p <. GeV T....5.5.5 φ red - < R > / =.5 fm, blue - < R > / =.9 fm [CGC-based calculation: Dusling & Venugopalan]
Correlations Ridge in p-pb (a) N, <p <GeV trk T (a) N,.<p <GeV trk T C trig ( η, φ).6.55.5.5. - - - - - η φ 5 C trig ( η, φ).9.85 - - - - - η φ 5 (b) 9 N trk <, <p <GeV T (b) N,.<p <GeV, no bal. trk T C trig ( η, φ)... - - - - - η φ 5 C trig ( η, φ).9.85 - - - - - η φ 5
Correlations Flow from correlations (two-particle cumulants) LHC: v n {, η > GeV} vs CMS top - v, bottom - v ( η) [%] v ( η), v 8 most central,.<p <5GeV T 7 6 5.5.5.5 η yellow - CMS blue - standard (< R > / =.5 fm) red - compact (< R > / =.9 fm)
Interferometry HBT radii Interferometric radii due to Bose-Einstein correlations - measure of the size of the system at freeze-out [fm] R side 5 a) R ALICE Data hydro model side p-p 9GeV p-p 7TeV Pb-Pb.76TeV STAR Data Cu-Cu 6.GeV Cu-Cu GeV Au-Au 6.GeV Au-Au GeV [fm] b) R out c) R long long, R 5 R out p-pb 5.TeV p-p 7TeV Au-Au GeV Cu-Au GeV Pb-Pb.76TeV 5 / (dn/dη) 5 / (dn/dη)
Conclusions Conclusions In hydro there is ow! Is there collectivity in small systems? collective dynamics is compatible with high-multiplicity LHC data for p-pb v n coecients measured in p-pb reproduced semiquantitatively v large Model -D correlations exhibit the two ridges, in particular the near-side ridge (surfers) Interferometric radii for p+pb are close to the A+A line, away from the p+p line - way to distinguish, will be veried shortly by ALICE Other eects (jets, corona,...) not included p+p [Bozek ] needs structure of the proton Other models of the initial collision [Bzdak et al., CGC+hydro]