Skip to the Main Content

Note:These pages make extensive use of the latest XHTML and CSS Standards. They ought to look great in any standards-compliant modern browser. Unfortunately, they will probably look horrible in older browsers, like Netscape 4.x and IE 4.x. Moreover, many posts use MathML, which is, currently only supported in Mozilla. My best suggestion (and you will thank me when surfing an ever-increasing number of sites on the web which have been crafted to use the new standards) is to upgrade to the latest version of your browser. If that's not possible, consider moving to the Standards-compliant and open-source Mozilla browser.

July 10, 2003

Not in Kyoto

Scheduling prevents me from being at the Strings Conference in Kyoto this week. But I can live vicariously over the web.

Here are some of the talks that struck me.


Andy Strominger’s been thinking about spacelike branes. One version of the story is to introduce a boundary interaction on the string worldsheet,

(1)S= Σ(x ix ix 0x 0)+λ Σcosh(x 0) S= \int_\Sigma (\partial x^i \partial x^i - \partial x^0 \partial x^0) + \lambda\int_{\partial \Sigma} \cosh (x^0)

Following Sen, one notes that this is the analytic continuation of Boundary Sine-Gordon theory, which is exactly soluble. For λ=±1/2\lambda=\pm 1/2, the theory is particularly interesting, as the boundary interaction is equivalent to setting Dirichlet boundary conditions for x 0x^0 (an “SD-brane”).

The boundary state for x 0x^0 is

(2)|B=(11+sin(πλ)e x 0+11+sin(πλ)e x 01)|0+oscillators | B\rangle = \left( \frac{1}{1+\sin(\pi\lambda)e^{x^0}} + \frac{1}{1+\sin(\pi\lambda)e^{-x^0}} -1\right) | 0\rangle + \text{oscillators}

For λ±1/2\lambda\to\pm 1/2, this has the limit

(3)|B={ nδ(x 02πin) λ1/2 nδ(x 02πi(n+1/2)) λ1/2 | B\rangle = \left\{ \array{ \sum_n \delta(x^0- 2\pi i n) & \lambda\to -1/2 \\ \sum_n \delta(x^0- 2\pi i (n+1/2)) & \lambda\to 1/2 }\right.

For λ1/2\lambda\to - 1/2, this is indeed sets x 0=0x^0=0 at the boundary. But the physics is a little awkward. It corresponds to a rolling tachyon which rolls up from - \infty towards the local maximum of the tachyon potential and then rolls back down. This has no analogue in the superstring. λ=+1/2\lambda=+1/2 seems to be even more boring. It looks like the boundary state vanishes (for real x 0x^0), and Sen has interpreted this as the closed string vacuum.

“Not so!” says Strominger. The closed string field configuration which is sourced by the brane satisfies

(4)(L 0+L¯ 0)|ϕ=|B (L_0+\overline{L}_0) |\phi\rangle = | B \rangle

Clearly, |ϕ |\phi\rangle is a solution, but it is not, according to Strominger, the one dictated by the usual rules of string perturbation theory. Rather, one should take the analytic continuation from Euclidean signature. Equivalently, one should solve

(5)ϕ=g iδ 4(x,tt i) \square \phi = g\sum_i \delta^4(\vec{x},t-t_i)

using the Feynman propagator and analytically-continue to imaginary t it_i. This gives

(6)ϕ=gr(tanhr+t2+tanhrt2) \phi = \frac{g}{r} \left( \tanh\frac{r+t}{2}+\tanh\frac{r-t}{2}\right)

This, indeed, vanishes near the origin, but is nonetheless nontrivial.

This is a really interesting subtlety. For static branes, finding the closed string field configuration sourced by the brane involves solving an elliptic equation, and demanding that the solution die off at infinity gives a unique solution. In the present situation, one has a hyperbolic equation to solve, and there is some extra physical input that is required. The proposal for SD branes is very nice, but I wonder what the general prescription might be like.


Raphael Bousso talked about proving Bekenstein’s original Entropy Bound for weakly-gravitating systems. The Bound states that the entropy of a system of total mass, MM, contained in as sphere of radius RR is bounded by

(7)S2πMR S \leq \frac{2\pi M R}{\hbar}

Since MM bh=R/2GM\ll M_{bh} = R/2G, the mass of a black hole whose Schwarzschild radius is RR, this is much more stringent than the Holographic Bound

(8)SπR 2G=A4G S \ll \frac{\pi R^2}{G \hbar} = \frac{A}{4 G \hbar}

of 't Hooft and Susskind.

To prove Bekenstein’s bound, Raphael invokes a version of the Covariant Entropy Bound due to Flanagan, Marolf and Wald. Let A 1A_1 and A 2A_2 be spacelike hypersurfaces, and LL be a lightsheet whose boundary is A 1A 2A_1-A_2, which is non-expanding with respect to A 1A_1 and expanding with respect to A 2A_2. Then the entropy crossing LL is bounded by

(9)S<A 1A 24G S \lt \frac{A_1-A_2}{4 G \hbar}

What’s cool about this version of the Covariant Entropy Bound is that the surfaces A iA_i don’t need to be closed. Indeed, Raphael takes them to be disks, such that the lightsheet emanating from A 1A_1 is just barely nonexpanding. The lightrays passing through our weakly-gravitating system are focussed, so A 2A_2 is a slightly smaller disk than A 1A_1. The radius of A 1A_1 is roughly πR 2\pi R^2. Because of the focussing, A 2A_2 is smaller by a fraction 2M/R2 M/R. So, sloughing over all the technical details, one obtains something like Bekenstein’s bound.

Theoretically, the bound you obtain this way could actually be tighter than Bekenstein’s, but it’s not clear that, in practice, a self-gravitating system can give a tighter bound.


Mina Aganagic gave a real tour de force talk about recent work on the open topological A-Model. In the A-Model, one has a Calabi-Yau manifold XX, and a Riemann surface Σ\Sigma with boundaries. One is interested in counting holomorphic maps from Σ\Sigma into XX such that the boundaries of Σ\Sigma get mapped into Special Lagrangian submanifolds of XX. In the case of XX a local Calabi-Yau, with special Lagrangian T 2×T^2\times\mathbb{R} fibers over 3\mathbb{R}^3, these guys have an incredibly elegant solution which, alas, defies my humble attempts at a simple summary.

More next time …

Posted by distler at July 10, 2003 11:24 AM

TrackBack URL for this Entry:

0 Comments & 0 Trackbacks

Post a New Comment