Black Hole Production in Planckian Scattering*

Transcription

1 Black Hole Production in Planckian Scattering* Kyungsik Kang and Horatiu Nastase Department of Physics, Brown University, Providence, RI 02912, USA Gauge/gravity duality correspondence suggests to a new possibility of tackling non-perturbative problems in gauge theories at high energies from the gravity duals of AdS with an appropriately modified background metric in the infrared region. In particular, high energy scattering in QCD can be related to Planckian scattering of shock-waves inside AdS where black holes are produced. In this talk, we present a general formalism of black hole production cross section at high energies via the 't Hooft recipe, i.e., collision of two gravitational shock-waves of Aichelburg and Sexl type, inside AdS of higher dimensionality with a curved background metric, and discuss string corrections to the cross section. * Supported in part by the US DoE Contract DE-FG-02-91ER40688-Task A Ref. K. Kang and H. Nastase, hep-th/ and hep-th/

2 Motivation for studying Black Hole production in high energy scattering 1. Low gravitational scale M pl, e.g.) large extra dimensions; warped space-time such as Randall-Sundrum scenarios => possibility of BHP at accelerator energies! 2. Gauge/gravity duality - High energy scattering in a gravity theory can be related to high energy scattering in QCD or gauge theories in general. - Outstanding non-perturbative QCD problems can be reconsidered. e.g.) High energy behavior of hadronic scattering amplitude, Regge behavior, Froissart bound, etc. Gauge-Gravity Dualities: Prototype(Maldacena, 1997): AdS 5 S 5 N=4 Super Yang-Mills SU(N) gauge group, N

3 SO(d,2)=AdS d+1 inv = M ink,d conformal inv. In general, gravity background gauge theory (gravity dual) QCD gravity dual is AdS modified in UV and IR (high and low r-5 th coordinate): 2 Brane Randall- Sundrum model Polshinski and Strassler (2002) QCD4: Scattering of colorless object Scattering in AdS 5 S 5 (glue balls) Match high energy QCD regimes with AdS scattering regime. 3. Giddings and Thomas, others In flat space, where geometrical horizon area

4 4. Eardley and Giddings: high energy collision a la t Hooft, Collision of two gravitational shockwaves of Aichelburg-Sexl type. 5. Even though one cannot calculate precisely the metric in the future of the collision except perturbatively (D Earth and Payne, 1992), Penrose: Existence of a closed D-2 dimensional surface of convergence at (, ), - a trapped surface GR: Exist horizon outside the trapped surface. 6. Eardley and Giddings: Extended Penrose s method 7. We try to refine this calculation and consider string corrections to the geometrical cross section results.

5 The Aichelburg-Sexl wave and t Hooft scattering at high energy t Hooft has proposed that an (almost) massless particle at high energies behaves like a plane gravitational wave a shockwave. That shockwave solution is due to Aichelburg and Sexl. The Aichelburg-Sexl solution is of the pp wave type. and has Ricci tensor and the rest are zero.

6 A black hole at rest has Boosted, one gets At the limit, where (since Einstein s equation is ) For 4d gravity,, and where

7 Particles following geodesics in the A-S metric are subject to two effects. 1) Geodesics going along u at fixed v are straight except at u = 0 where there is a discontinuity 2) The second effect is a refraction (or gravitational deflection, rather), where the angles and made by the incoming and outgoing waves with the plane at an impact parameter

8 The scattering of two massless particles of very high energy yet Particle one with a free wavefunction becomes (at u=0, just after the shockwave) The scattering amplitude is

9 With and integration and The differential cross section which is like Rutherford scattering, as if a single graviton is exchanged. ( with the effective gravitational coupling replacing of QED )

10 The generalization to higher dimensions, as Amati and Klimcik did, a shockwave metric Would shift the geodesics at u=0 by and the S matrix An impact parameter transform as in D = 4 and get with ( photon energy and also) so.

11 Then one obtains (with ) where the q dependence from and

12 Black hole production via Aichelburg- Sexl wave scattering The case of the collision of two A-S waves, due to Penrose, and extended by Eardley and Giddings, there will be a black hole in the future of the collision without actually calculating the gravitational field. One can prove the existence of a trapped surface and then one knows that the future of the solution will involve a black hole whose horizon will be outside the trapped surface. An apparent horizon is the outermost marginally trapped surface. A marginally trapped surface is defined as a closed spacelike D-2 surface, the outer null normals (in both future-directed directions) of which have zero convergence. In physical terms, what this means is that there is a closed surface whose normal null geodesics (light rays) don t diverge, so are trapped by gravity.

13 Convergence in the case of a congruence of timelike geodesics characterized by the tangent vector, defining and the projector onto the subspace orthogonal to (induced metric), the convergence is. The convergence is The metric of two colliding general shockwaves; one moving in the u direction, and the other in the v direction. Take the union of the two null hypersurfaces and with a D-2 dimensional intersection, that intersects on its turn S on a D-3 surface C. The condition of zero convergence implies interior to C.

14 On the first disk, And the null geodesics through defined by are The tangent generators of the surface We choose By imposing, we get

15 On disk 2, implies Two surfaces intersect on C, thus the normal,, has to be continuous across C. This means that for the A-S wave at b=0, when implying, In D=4, For D>4, The total area of the trapped surface in D spacetime dimensions (two flat balls) is

16 From the explicit form of the Schwarzschild solution in D dimensions the horizon radius of a black hole of mass is so that the horizon area of the mass = black hole is The area of the trapped surface is smaller than the horizon area of the black hole to form (since the horizon is by definition outside the trapped surface)

17 Extension Let us now try to extend this for the case of nonzero b in any dimension. The correct solution is that, just are Green s functions for sources at which both are zero on the same curve C enclosing and. Then one imposes the condition for continuity of the null normal which gives which fixes (together with the previous conditions) the form of C.

18 If b=0 we reproduce the known result of. and the area of the trapped surface satisfies, so that at the maximum b is. In D>4, the condition implies

19 String Correcctions Amati and Klimcik first generalize the t Hooft and Dray and t Hooft calculation. A shockwave metric would shift the geodesics at u=0 by and the S matrix In string theory, the t Hooft scattering in the shockwave background gives (for an open string -> photon) the S matrix acting on creation/annihilation operators as

20 where and are nonzero-modes. We match with the S matrix obtained by resumming string diagrams, And the tree amplitude is As a first approximation, we can neglect all string oscillators in and obtain where is (impact parameter space), and becomes equal to the A-S result at large b.

21 For whereas for In D=4 we get, with.

22 Then the condition for the trapped surface, gives (for, so ) thus increases. The area of the trapped surface giving the bound on the horizon area is and, so also increases At nonzero impact parameter of the two Amati-klimcik waves, the normal condition is, so of, so we only get an extra factor to the condition,

23 Randall-Sundrum-type models The solution for an A-S type wave in the RS background is where which is a solution of Einstein s equation with. Another form for the metric is on the brane (y=0)

24 At large distances, At small distances, The solution can also be expressed as just a modification of the function. The new function is Imposing the continuity of the normal condition

25 The area of the trapped surface is the area of two disks, is the horizon radius of the formed black hole, For Thus imposing For and

26 The mass of the black hole At nonzero b, the normal continuity condition becomes, so we only get an extra factor of to the condition, so that now The trapped area,, remains the same as a function of and b,

27 Generalizing the formalism to curved higher dimensional background A curved spacetime background of the RS type: The vector normal to the surface is The continuity condition for the normal,, Necessary to calculate the coordinate transformation from the coordinate system

28 where The convergence of the normals is now again where and Therefore we write

29 The trapped surface is a surface defined by both (const) and by. The nonzero b case: find a surface C and a function that satisfies both and with.

30 From the second (continuity) equation we get

31 Conclusions We have found that string corrections increase the horizon area. For the effective shockwave metric, we have found that if we scatter head-on (at b=0) two such waves, each characterized by an impact parameter, we obtain trapped surfaces which are deformed disks of area higher than the area obtained from A-S wave scattering. For the effective shockwave metric, in the case of, we get an increase of the area of the black hole formed, as well as of the classical scattering cross section,, while in the we get that the area of the formed black hole is of the order of Y (modified string scale), not, so much larger.

32 For higher dimensions, we have found a conservative approximation scheme for the area of the horizon formed which gives us a maximum impact parameter (indicative of the scattering cross section, as we expect that ). We have thus obtained that in D=4, and in D=5 for instance. For more realistic scenarios, involving possible creation of black hole at accelerators for low fundamental scale, the one brane Randall-Sundrum case. In the case that the 5 th direction is highly curved, we have obtained just corrections to the flat 4d case, whereas for a weakly curved 5 th direction, we have corrections about the 5d flat space black hole creation. Finally, we have found a solution for an Aichelburg-Sexl wave inside an AdS background, and we have calculated the scattering amplitude for t Hooft scattering in such a wave, at small and large distances r. This was done for later use for analysis of the gravity dual of QCD high energy scattering (See Nastase s talk this afternoon)