Dynamical stability of the baryon vertex in AdS5 × RP5: decay energetics across discrete-torsion sectors and the odd-N Clifford-algebra ambiguity of fivebrane-threebrane intersections

The AdS/CFT dual of $\mathcal{N}=4$ super Yang-Mills theory with orthogonal or symplectic gauge group is type IIB string theory on $\mathrm{AdS}_5\times\mathbb{RP}^5$, where the four possible choices of discrete torsion $(\theta_{NS},\theta_{RR})$ of the two two-form gauge fields correspond to the gauge groups $SO(2k)$, $SO(2k+1)$, $Sp(k)$, and a fourth choice related by $SL(2,\mathbb{Z})$ duality. The baryon vertex of the boundary theory — a D5-brane whose worldvolume $S^5$ covers $\mathbb{RP}^5$ two-to-one, with $N$ attached fundamental strings — and its wrapped-brane relatives were introduced and classified topologically in Witten's 1998 study, which explicitly left open the dynamical questions this problem addresses. First: which of the topologically allowed decays of the baryon vertex are energetically realized in each discrete-torsion sector — decay to meson-like pairwise string pairings (allowed for $Sp(k)$) and decay to a state containing a D3-brane wrapped on $\mathbb{RP}^3$ plus strings (allowed for $SO(N)$) — and how do the answers change in deformations that give the boundary theory a mass gap, where the wrapped branes become genuine particles? Second: in the fat-string configuration (a fivebrane wrapped on $\mathbb{RP}^4$), the fermion zero modes of strings stretched to the $N$ threebranes generate a Clifford algebra that for odd $N$ has two inequivalent irreducible representations, distinguished by the sign of $\psi_1\psi_2\cdots\psi_N$; which representation is realized in $\mathrm{AdS}_5\times\mathbb{RP}^5$ and to what boundary-theory object does it correspond?

Background

The AdS/CFT correspondence equates type IIB superstring theory on the product spacetime $\mathrm{AdS}_5\times S^5$ (five-dimensional anti-de Sitter space times a five-sphere) with four-dimensional $\mathcal{N}=4$ super Yang-Mills theory with gauge group $SU(N)$, where $N$ counts units of five-form flux through the $S^5$. If the five-sphere is replaced by real projective five-space, $\mathbb{RP}^5=S^5/\mathbb{Z}_2$, via an orientifold (a quotient combined with a reversal of the string worldsheet orientation), the dual gauge theories have orthogonal or symplectic gauge groups. The four possible backgrounds are distinguished by the discrete torsion $(\theta_{NS},\theta_{RR})$ of the Neveu-Schwarz and Ramond-Ramond two-form gauge fields $B_{NS}$ and $B_{RR}$: $(0,0)$ corresponds to $SO(2k)$, $(\theta_{NS}\neq 0,0)$ to $Sp(k)$, $(0,\theta_{RR}\neq 0)$ to $SO(2k+1)$, and the remaining choice is related to these by the $SL(2,\mathbb{Z})$ duality of type IIB string theory, which permutes the three nonzero torsions.

In this setting Witten's 1998 paper constructed the holographic avatars of the nonperturbative objects of the boundary gauge theory, with tension of order $N$: the baryon vertex, a Dirichlet fivebrane (D5-brane — a solitonic hypersurface in string theory on which open strings can end) whose worldvolume is a five-sphere $S^5$ mapping two-to-one onto $\mathbb{RP}^5$ (a fivebrane wrapped only once on $\mathbb{RP}^5$ is topologically forbidden), with $N$ fundamental strings attached, coupling $N$ external quarks into a gauge-invariant baryon; a D3-brane wrapped on $\mathbb{RP}^3\subset\mathbb{RP}^5$, dual to Pfaffian operators such as $\mathrm{Pf}(\Phi)$ — the square root of the determinant of an antisymmetric matrix of adjoint fields — built from $k=N/2$ antisymmetrized adjoint fields (which exist for $SO(2k)$ but not for $SO(2k+1)$ or $Sp(k)$); a 'fat string' made of a D5-brane wrapped on $\mathbb{RP}^4$, dual to external charges in a spinor representation of $SO(N)$; and domain walls from D5-branes wrapped on $\mathbb{RP}^2$.

Which wrappings are consistent is governed by the condition that the cohomology class $[H_{NS}]$ of the $B$-field restricts trivially to the brane worldvolume $Y$, $i^*[H]=0$ for the inclusion $i:Y\to M$, corrected by a Poincaré-dual term $[D]$ when lower-dimensional branes end on $Y$. Witten observed that global gauge invariance of the $B$-field requires a further torsion correction $i^*[H]=W$, where $W$ is the image of the second Stiefel-Whitney class $w_2(Y)$ (the obstruction to a spin structure) under the Bockstein connecting map of $0\to\mathbb{Z}\to\tilde{\mathbb{Z}}\to\mathbb{Z}_2\to 0$ (coefficients twisted by the orientation bundle); this is now called the Freed-Witten anomaly condition and was derived from worldsheet anomalies shortly afterwards, with the correct form $i^*\zeta_B=W_3$ given by the third integral Stiefel-Whitney class.

With these rules the decay pattern of the baryon vertex is fixed topologically. For symplectic gauge group ($\theta_{NS}\neq 0$) the baryon vertex can decay to a state with no fivebrane in which the $N$ external quarks are joined pairwise by elementary strings, meson-like; for orthogonal gauge group ($\theta_{NS}=0$) the baryon is odd under the generator $\tau$ of $O(N)/SO(N)$ (the orientation-reversing component of $O(N)$, acting as a global $\mathbb{Z}_2$ symmetry) while mesons are even, so decay to mesons alone is forbidden, but decay to a state containing a D3-brane wrapped on $\mathbb{RP}^3$ (a $\tau$-odd, Pfaffian-like object) plus strings is topologically allowed, with an odd number of strings running from the wrapped threebrane to the boundary when $N$ is odd. What the topological analysis cannot decide is dynamics: the relative energies of these channels, whether the baryon vertex is stable, metastable, or unstable in each sector, and whether the conclusions are consistent under $SL(2,\mathbb{Z})$. The question is sharpest in deformations of the background in which the function $f(x_0)$ relating boundary energy to bulk mass is bounded below: the undeformed conformal theory has no mass gap, so a wrapped brane of mass $m$ at radial position $x_0$ has boundary energy $m\sqrt{f(x_0)}$ with $f(x_0)\to 0$ and corresponds to an operator creating arbitrarily soft states rather than a particle, whereas with $f\geq f_0>0$ the wrapped branes become genuine particles and their decays become ordinary particle decays.

A second leftover concerns the fermionic zero modes of strings stretched between the wrapped fivebrane and the $N$ threebranes (5-3 strings) in the flat-space orientifold description $\mathbb{R}^4\times(\mathbb{R}^6/\mathbb{Z}_2)$. Their quantization supplies $N$ fermion operators $\psi_1,\ldots,\psi_N$ generating a Clifford algebra (the algebra of gamma matrices), all odd under the $\mathbb{Z}_2$ remnant of the fivebrane worldvolume gauge symmetry. For even $N$ this $\mathbb{Z}_2$ acts as chirality and the ground states form the two spinor representations of $SO(N)$, which is how the fat string carries spinor-representation external charges. For odd $N$ there are instead two inequivalent irreducible representations of the Clifford algebra, distinguished by the sign of the central element $\psi_1\psi_2\cdots\psi_N$, and the physical meaning of the appearance of both — which is realized in $\mathrm{AdS}_5\times\mathbb{RP}^5$ and to what boundary-theory object it corresponds — was left uninterpreted in the source, which handled only the case of even $N$.

Three sibling questions raised in the same paper were later resolved: the precise torsion correction to B-field quantization (Freed-Witten, 1999); the interpretation of the two-brane obtained by wrapping a D3-brane on $\mathbb{RP}^1\subset\mathbb{RP}^5$, identified in 2022 as the holographic dual of a non-invertible topological symmetry operator of the boundary theory (a symmetry defect whose action cannot be inverted, so that the symmetries form a structure more general than a group); and the holographic status of the $k$-instanton moduli space, settled by the multi-instanton saddle-point analysis of 1999. These are not part of the problem stated here.

Problem Statement

This problem concerns the dynamical and quantum completion of the topological analysis of branes wrapped on $\mathbb{RP}^5$ in type IIB string theory on $\mathrm{AdS}_5\times\mathbb{RP}^5$, in the framework and notation of Witten, JHEP 07 (1998) 006, where the topological part is settled.

(1) Decay energetics of the baryon vertex. For each of the four discrete-torsion sectors — $(\theta_{NS},\theta_{RR})=(0,0)$ with gauge group $SO(2k)$; $(0,\theta_{RR}\neq 0)$ with $SO(2k+1)$; $(\theta_{NS}\neq 0,0)$ with $Sp(k)$; and the fourth choice, related to the others by $SL(2,\mathbb{Z})$ — determine which of the topologically allowed decays of the baryon vertex (a fivebrane whose worldvolume $S^5$ wraps $\mathbb{RP}^5$ twice, with $N$ attached fundamental strings) are energetically realized: decay to a state with no fivebrane in which the $N$ external quarks are joined pairwise by elementary strings (topologically allowed for $\theta_{NS}\neq 0$), and decay to a state containing a D3-brane wrapped on $\mathbb{RP}^3$ plus strings pairing or connecting the remaining external charges (topologically allowed for $\theta_{NS}=0$, with an odd number of strings from the wrapped threebrane to the boundary when $N$ is odd). Determine, at least at leading order in the $1/N$ and string-coupling expansions, whether the baryon-vertex configuration is stable, metastable, or unstable against each channel in each sector and whether the conclusions are consistent under the $SL(2,\mathbb{Z})$ action on $(\theta_{NS},\theta_{RR})$; the same questions apply in deformations of the background whose boundary dual develops a mass gap, where the wrapped $\mathbb{RP}^3$ threebrane is a Pfaffian particle and the decays become ordinary particle decays whose rates and final states are well defined.

(2) The odd-$N$ Clifford-algebra ambiguity. In the flat-space orientifold description of the fat string (a fivebrane wrapped on $\mathbb{RP}^4$ meeting the $N$ threebranes), identify the physical meaning of the two inequivalent irreducible representations of the Clifford algebra generated by the 5-3-string fermion zero modes $\psi_1,\ldots,\psi_N$ when $N$ is odd, which are distinguished by the sign of the central element $\psi_1\psi_2\cdots\psi_N$: determine which representation(s) occur for the wrapped branes of $\mathrm{AdS}_5\times\mathbb{RP}^5$, and to what boundary-theory object — operator, external charge, or symmetry defect — each corresponds in the $SO(2k+1)$ theory.

References

1. E. Witten, "Baryons and branes in anti de Sitter space," JHEP 07 (1998) 006. DOI: 10.1088/1126-6708/1998/07/006. arXiv:hep-th/9805112. URL: https://arxiv.org/abs/hep-th/9805112
2. D. S. Freed and E. Witten, "Anomalies in string theory with D-branes," Asian J. Math. 3 (1999) 819. DOI: 10.4310/ajm.1999.v3.n4.a6. arXiv:hep-th/9907189. URL: https://arxiv.org/abs/hep-th/9907189
3. S. Gukov, M. Rangamani and E. Witten, "Dibaryons, strings and branes in AdS orbifold models," JHEP 12 (1998) 025. DOI: 10.1088/1126-6708/1998/12/025. arXiv:hep-th/9811048. URL: https://arxiv.org/abs/hep-th/9811048
4. P. Bouwknegt and V. Mathai, "D-branes, B-fields and twisted K-theory," JHEP 03 (2000) 007. DOI: 10.1088/1126-6708/2000/03/007. arXiv:hep-th/0002023. URL: https://arxiv.org/abs/hep-th/0002023
5. O. Bergman, E. G. Gimon and S. Sugimoto, "Orientifolds, RR torsion, and K-theory," JHEP 05 (2001) 047. DOI: 10.1088/1126-6708/2001/05/047. arXiv:hep-th/0103183. URL: https://arxiv.org/abs/hep-th/0103183
6. P. Bouwknegt, J. Evslin, B. Jurčo, V. Mathai and H. Sati, "Flux compactifications on projective spaces and the S-duality puzzle," Adv. Theor. Math. Phys. 10 (2006) 345. DOI: 10.4310/ATMP.2006.v10.n3.a3. arXiv:hep-th/0501110. URL: https://arxiv.org/abs/hep-th/0501110
7. N. Dorey, T. J. Hollowood, V. V. Khoze, M. P. Mattis and S. Vandoren, "Multi-instanton calculus and the AdS/CFT correspondence in N=4 superconformal field theory," Nucl. Phys. B 552 (1999) 88. DOI: 10.1016/S0550-3213(99)00193-5. arXiv:hep-th/9901128. URL: https://arxiv.org/abs/hep-th/9901128
8. O. Aharony, Y. E. Antebi, M. Berkooz and R. Fishman, "'Holey sheets': Pfaffians and subdeterminants as D-brane operators in large N gauge theories," JHEP 12 (2002) 069. DOI: 10.1088/1126-6708/2002/12/069. arXiv:hep-th/0211152. URL: https://arxiv.org/abs/hep-th/0211152
9. C. G. Callan, Jr., A. Guijosa and K. G. Savvidy, "Baryons and string creation from the fivebrane worldvolume action," Nucl. Phys. B 547 (1999) 127. DOI: 10.1016/S0550-3213(99)00057-7. arXiv:hep-th/9810092. URL: https://arxiv.org/abs/hep-th/9810092
10. K. Sfetsos and K. Siampos, "Stability issues with baryons in AdS/CFT," JHEP 08 (2008) 071. DOI: 10.1088/1126-6708/2008/08/071. arXiv:0807.0236. URL: https://arxiv.org/abs/0807.0236
11. E. M. Sahraoui, "The decay of the five-brane in AdS5 x RP5," arXiv:hep-th/0301247 (2003; unpublished, unrefereed preprint treating only the orthogonal sector, giving the worldvolume energy of the decayed threebrane with the radius only in a BPS configuration; it does not settle the energetic selection across the four discrete-torsion sectors, the $SL(2,\mathbb{Z})$ consistency, or the mass-gap case). URL: https://arxiv.org/abs/hep-th/0301247
12. S. Giombi and B. Offertaler, "Wilson loops in N=4 SO(N) SYM and D-branes in AdS5 x RP5," JHEP 10 (2021) 016. DOI: 10.1007/JHEP10(2021)016. arXiv:2006.10852. URL: https://arxiv.org/abs/2006.10852
13. I. García Etxebarria, "Branes and non-invertible symmetries," Fortsch. Phys. 70 (2022) 2200154. DOI: 10.1002/prop.202200154. arXiv:2208.07508. URL: https://arxiv.org/abs/2208.07508
14. O. Bergman and S. Hirano, "The holography of duality in N=4 super-Yang-Mills theory," JHEP 11 (2022) 069. DOI: 10.1007/JHEP11(2022)069. arXiv:2208.09396. URL: https://arxiv.org/abs/2208.09396