JINR, June 06, 2012 Turbulent nonabelian matter in high energy nuclear collisions A. Leonidov P.N. Lebedev Physical Institute, Moscow A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
◮ Elliptic flow in heavy ion collisions ◮ Emergence of Kolmogorov spectrum in glasma ◮ Emergence of Kolmogorov spectrum in the toy CGC model ◮ Emergence of Kolmogorov spectrum in QGP ◮ Turbulent instability in QED plasma A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Elliptic flow Y X Spectators Ψ RP b Spectators Definition of the reaction plane A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Elliptic flow Spatial asymmetry of the reaction zone ◮ � y 2 − x 2 � ǫ s , part = � y 2 + x 2 � ◮ � = 1 dxdy ( y 2 − x 2 ) dN p � y 2 − x 2 � N p dx dy Momentum asymmetry: elliptic flow ◮ � p 2 � X − p 2 Y v 2 ≡ p 2 X + p 2 Y ◮ 1 dN 1 dN = (1+ p T dydp T d φ 2 π p T dydp T 2 v 2 ( p T ) cos 2( φ − Ψ RP ) + . . . ) A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Elliptic flow Y � � � � X Hydrodynamic origin of the elliptic flow: anisotropic pressure converts spatial anisotropy is into momentum one A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Elliptic flow 0.08 2 v 0.06 0.04 0.02 ALICE STAR 0 PHOBOS PHENIX -0.02 NA49 CERES -0.04 E877 EOS -0.06 E895 FOPI -0.08 2 3 4 1 10 10 10 10 s (GeV) NN Average elliptic flow as a function of √ s A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Elliptic flow ALICE at p T = 1.7GeV/c at p T = 0.7GeV/c 10 3 10 4 Differential elliptic flow as a function of √ s A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Elliptic flow ε / 2 0.25 HYDRO limits v 0.2 0.15 0.1 E /A=11.8 GeV, E877 lab E /A=40 GeV, NA49 lab 0.05 E /A=158 GeV, NA49 lab s =130 GeV, STAR NN s =200 GeV, STAR Prelim. NN 0 0 5 10 15 20 25 30 35 (1/S) dN /dy ch Hydro limit for ideal liquid for v 2 reached at RHIC A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Elliptic flow Glauber 25 STAR non-flow corrected (est.) -4 η /s=10 STAR event-plane 20 η /s=0.08 v 2 (percent) 15 η /s=0.16 10 5 0 0 1 2 3 4 p T [GeV] CGC -4 η /s=10 25 STAR non-flow corrected (est). η /s=0.08 STAR event-plane 20 v 2 (percent) 15 η /s=0.16 10 5 η /s=0.24 0 0 1 2 3 4 p T [GeV] Dependence of v 2 on viscosity for Glauber and CGC initial conditions A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Initial conditions Initial transverse energy density for AuAu collisions at √ s = 200 GeV A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Little Bang: collision stages t freeze out hadrons in eq. hydrodynamics gluons & quarks in eq. gluons & quarks out of eq. kinetic theory strong fields classical EOMs z (beam axis) A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Little Bang: before the collision Initial state at t < 0 A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Nuclei before collision: degrees of freedom fields sources k + Λ + P + 0 ◮ Characteristic evolution time for parton modes k − ∼ 2 k + = 2 P + ∆ x + ∼ 1 x k 2 k 2 ⊥ ⊥ ◮ Static modes (sources): x ∼ 1 ◮ Fluctuational modes (fields): x ≪ 1 QCD physics at high energies is that of fields with x ≪ 1 A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Nuclei before collision: fields The fields A a µ and the source J a µ are related by the equation [ D µ , F µν ] = J ν ⇔ J µ = δ µ + ρ 1 ( x ⊥ , x − ) Solution of classical equations: A − = 0 A + = 0 , i A i g U ( x ⊥ , x − ) ∂ i U † ( x ⊥ , x − ) = where � � � x − U ( x ⊥ , x − ) dy − α ( x ⊥ , x − ) = P exp ig −∞ α ( x ⊥ , x − ) − ρ ( x ⊥ , x − ) / ∇ 2 = ⊥ A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Nuclei before collision: observable quantities ◮ Charge density ρ ( x ⊥ , x − ) is random. Event-by-event averaging with respect to ρ ( x ⊥ , x − ) is described by some functional W Λ + [ ρ ] ◮ For the simplest Gaussian ensemble A δ ab δ 2 ( x ⊥ − y ⊥ ) δ ( x − − y − )) � ρ a ( x ⊥ , x − ) ρ b ( y ⊥ , y − ) � = g 2 µ 2 ◮ Structure function: 2 k + dN (2 π ) 3 � A i a ( k , x + )) A i a ( − k , x + )) � W Λ+ = d 3 k 1 � � �� � A i a (0)) A i − x 2 ⊥ Q 2 S ln( x 2 ⊥ µ 2 ) a ( x )) � ∼ 1 − exp x 2 ⊥ ◮ Q 2 S - saturation scale, 0 e λ s Y , Q 2 S ( Y ) ≃ Q 2 Q 2 0 ∼ A 1 / 3 A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Nuclei before collision: quantum evolution x = k + 1 δ S ⊥ ∼ P + Q 2 A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Nuclei before collision: quantum evolution fields sources k + Λ + Λ + P + 1 0 δ T NLO T LO ◮ Structure function, classical approximation � < AA > = [ d ρ ] W Λ + [ ρ ] A cl . ( ρ ) A cl . ( ρ ) ◮ Arbitrary observable, classical approximation � �O� Y = [ d α ] O [ α ] W Y [ α ] ◮ Quantum evolution: JIMWLK equation: � ∂ �O [ α ] � Y = � 1 δ δ Y ( x ⊥ ) χ ab x ⊥ , y ⊥ [ α ] Y ( y ⊥ ) O [ α ] � Y δα a ∂ Y 2 δα b x ⊥ , y ⊥ A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Nuclei before collision: quantum evolution ◮ The JIMWLK equation is Hamiltonian: ∂ �O [ α ] � Y = �H JIMWLK O [ α ] � Y ∂ Y ◮ Kernels of JIMWLK equation: � d 2 z ⊥ ( x ⊥ − z ⊥ )( y ⊥ − z ⊥ ) χ ab x ⊥ y ⊥ [ α ] = 4 π 3 ( x ⊥ − z ⊥ ) 2 ( y ⊥ − z ⊥ ) 2 �� � � �� 1 − U † 1 − U † x ⊥ U z ⊥ z ⊥ U y ⊥ ◮ Nonlinear dependence on sources � � � x − U † ( x ⊥ , x − ) = P exp dy − α a ( x ⊥ , x − ) T a ig −∞ ◮ In the limit of small α JIMWLK turns into BFKL A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Nuclear collision: classical solution [ D µ , F µν ] = J ν J µ = δ µ + ρ 1 ( x ⊥ , x − ) + δ µ + ρ 2 ( x ⊥ , x − ) Look for a solution in all orders in ρ 1 , 2 A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Boost-invariant classical solution ◮ Coordinates τ, η x 0 + x 3 = τ e η , x 0 − x 3 = τ e − η ◮ For a single source one uses gauges A ± = 0 ◮ For the two-source problem it is convenient to use the mixed gauge A τ = 0 A τ = A τ ≡ 1 τ ( x + A − + x − A + ) ◮ Boost-invariant solution does not depend on η A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Boost invariant classical solution t x + x − η = cst. τ = cst. (3) A µ = ? (1) (2) z A µ = pure gauge 1 A µ = pure gauge 2 (4) A µ = 0 ◮ Look for the η - independent solution of the form: θ ( − x + ) θ ( x − ) A i (1) + θ ( x + ) θ ( − x − ) A i (2) + θ ( x + ) θ ( x − ) A i A i = (3) θ ( x + ) θ ( x − ) A η A η = (3) ◮ Matching conditions at τ = 0 : A i A i (1) + A i (3) | τ =0 = (2) ig � � A η A i (1) , A i (3) | τ =0 = (2) 2 A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Immediately after collision there form longitudinal chromoelectric and chromomagnetic fields - glasma : . . . . . . . . . . . . . . . . . . . . . . . . . . � � E z A i (1) , A i = ig (2) ig ǫ ij � � (1) , A j B z A i = (2) A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Initial conditions: hydrodynamics? ◮ Equations of motion ∂ µ T µν = 0 ◮ Equation of state p = f ( ǫ ) ◮ Initial conditions set at some τ = τ 0 T µν ( τ = τ 0 , η, x ⊥ ) ◮ Generic structure of T µν : ǫ ǫ T µν = 3 ǫ 3 ǫ 3 A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Initial conditions, Color Glass Condensate For a configuration E a µ = λ B a µ ǫ ǫ � T µν ( τ = 0 + , η, x ⊥ ) � = ǫ − ǫ Does not look as hydro at all but is very similar to QCD string models (negative p z !) Glasma flux tubes strings Negative p z string tension Glasma instabilities string breaking Isotropisation mechanism? A. Leonidov Turbulent nonabelian matter in high energy nuclear collisions
Recommend
More recommend