Decoupling and Near-Horizon Limits

A near-horizon limit zooms into a throat; a decoupling limit additionally proves that the retained throat excitations stop exchanging finite energy with asymptotic bulk modes. The distinction matters: deleting the constant in a harmonic function can produce a useful local metric without defining an autonomous quantum system. The D3-brane provides the clean model, while D1–D5 illustrates how compact radii and multiple charges must also be scaled.

Required background. Gauge Theory and String-Theory Preparation supplies D-branes, open/closed strings, and the gauge coupling whose fixed scaling defines the limit.

Helpful background. Quantum Gravity clarifies what decoupling establishes; Quantum Information prepares later Hilbert-space statements; and Black-Hole Geometry and Mechanics provides the horizon and throat geometry being isolated.

The D3-brane throat

The string-frame extremal D3 solution has constant dilaton and

$$ds^2=H^{-1/2}(-dt^2+d\mathbf x_3^2) +H^{1/2}(dr^2+r^2d\Omega_5^2), \qquad H(r)=1+\frac{L^4}{r^4},$$

where flux quantization gives $L^4=4\pi g_sN\alpha'^2$. At $r\ll L$, the constant in $H$ is negligible and the geometry becomes AdS$_5\times S^5$ with common radius $L$. This is a near-horizon approximation at fixed string parameters.

The stronger Maldacena limit sends

$$\alpha'\to0, \qquad U=\frac{r}{\alpha'}\ \text{fixed}, \qquad g_{\mathrm{YM}}^2N\ \text{fixed}, \qquad N\ \text{fixed or large as specified}.$$

In the weakly coupled open-string description, oscillator masses diverge, leaving the brane gauge theory. In the asymptotically flat region, the surviving finite-energy closed-string modes are free massless ten-dimensional fields. The throat sector is different: it retains the full string theory, including locally massive string excitations. For example, with $\lambda=4\pi g_sN$, an excitation of proper energy $E_{\rm proper}\sim1/\sqrt{\alpha'}$ at fixed $U$ has boundary energy

$$E=H^{-1/4}E_{\rm proper} \simeq\frac{r}{L}E_{\rm proper} \sim\frac{U}{\lambda^{1/4}},$$

which stays finite. Here $U$ is an energy-valued radial coordinate, not the energy of every state at that radius. This redshift is why the decoupling limit yields string theory on the throat before any additional supergravity approximation. See Aharony et al. 2000, §3.1, Eqs. (3.3)–(3.6).

Why the sectors decouple

There are two complementary descriptions. At weak coupling, low-energy open strings on the branes interact through the gauge theory, while closed strings in the bulk couple through powers of $\kappa_{10}E^4$. Sending $\alpha'\to0$ at fixed gauge parameters drives that interaction to zero. At strong effective coupling, the same low-energy sector is represented by modes trapped in the infinite redshift throat.

The absorption probability supplies an operational check: for a low-frequency bulk quantum, the D3 absorption cross section is suppressed by a positive power of $\omega L$. Thus a finite throat excitation cannot efficiently escape to infinity in the limit. The general D$p$-brane scaling, including where curvature or dilaton control fails, is derived in Itzhaki et al. 1998, §§2–3, pp. 3–8.

Decoupling does not say that the two sectors are absent from the original finite-$\alpha'$ theory. The discarded coupling is an irrelevant deformation. Restoring the constant in $H$ or moving the cutoff outward couples throat data back to asymptotically flat gravity.

Retained and discarded data

A well-defined limit lists what is fixed:

• energy relative to the brane ground state, conjugate to the retained boundary time $t$;
• integer charges such as $N$ and any compact momenta;
• the gauge coupling or dimensionless effective coupling;
• rescaled radial position and compactification radii needed to keep finite tensions;
• boundary conditions at the throat entrance.

It also lists what is removed: massive oscillators from the low-energy open-string action and the asymptotically flat sector, finite-energy exchange between the two sectors, and irrelevant open-string operators whose coefficients vanish with $\alpha'$. Massive string modes within the throat are subsequently omitted only in a low-energy supergravity approximation. The string-loop and curvature conditions remain separate. In the gauge normalization $S_{\rm YM}=-(2g_{\rm YM}^2)^{-1}\int\operatorname{Tr}F^2$, with $\operatorname{Tr}(T^aT^b)=\delta^{ab}/2$, D3-branes have

$$\frac{L^4}{\alpha'^2}=4\pi g_sN=\lambda,$$

so large $\lambda$ suppresses curvature corrections and large $N$ at fixed $\lambda$ suppresses bulk quantum loops. Perturbative type-IIB string theory also asks for $g_s=\lambda/(4\pi N)\ll1$. The familiar sufficient window is $1\ll\lambda\ll4\pi N$. The gauge theory itself exists outside that window; only the simple geometric description is lost.

D1–D5 and the compactification check

For D1–D5 on $S^1\times X_4$, the throat radius depends on two harmonic charges, while the six-dimensional Newton constant depends on $g_s$, $\alpha'$, and the volume of $X_4$. The decoupling limit must retain the circle radius, charge integers, and an appropriate scaled compact volume so that the AdS$_3\times S^3$ region and its finite-energy excitations survive. Merely taking $r\to0$ would not specify the dual energy or eliminate asymptotic exchange.

An independent check is to reverse one scaling. Keep $\alpha'$ small but hold $r$ rather than $U=r/\alpha'$ fixed. Then $U\to\infty$: the observer is driven to the throat boundary and the full harmonic function retains its asymptotically flat region. Alternatively, restore a finite $\alpha'$; massive strings and irrelevant open/closed couplings return. Both tests identify which physical assumptions, rather than which metric terms, created decoupling.

Exercises

Substitute $r=\alpha'U$ and $L^4=4\pi g_sN\alpha'^2$ into the near-horizon D3 metric. Show that $ds^2/\alpha'$ remains finite and determine the two independent parameters controlling curvature and loops.

Solution

In the throat, $H=L^4/r^4$. The longitudinal factor becomes $H^{-1/2}=r^2/L^2=\alpha'U^2/\sqrt{4\pi g_sN}$, whereas $H^{1/2}dr^2=L^2dr^2/r^2=\alpha'\sqrt{4\pi g_sN}\,dU^2/U^2$. The sphere term has the same finite coefficient after division by $\alpha'$. Hence $ds^2/\alpha'$ is finite with radius squared $L^2/\alpha'=\sqrt{4\pi g_sN}$. Large $g_sN$ suppresses alpha-prime curvature corrections. Bulk quantum loops are instead suppressed by the large flux rank, conventionally $1/N^2$ at fixed ’t Hooft coupling, so the two controls are independent.

References

• Aharony, Ofer, Steven S. Gubser, Juan Maldacena, Hirosi Ooguri, and Yaron Oz. “Large N Field Theories, String Theory and Gravity.” Physics Reports 323 (2000): 183–386. Open manuscript.
• Itzhaki, Nissan, Juan Maldacena, Jacob Sonnenschein, and Shimon Yankielowicz. “Supergravity and the Large N Limit of Theories with Sixteen Supercharges.” Physical Review D 58 (1998): 046004. Open manuscript.