Stable non-BPS D-branes in Type I string theory

TL;DR

This work develops a boundary-state description of stable non-BPS D-branes in Type I string theory by embedding Sen's tachyon condensation into closed-string boundary states. Starting from a Type IIB D1–anti-D1 bound state with a Z2 Wilson line, the authors show that tachyon condensation yields a stable non-BPS D-particle, whose mass agrees with prior results and whose RR sector vanishes at the tachyon maximum. Applying the Omega projection via a crosscap, the Type I realization of the D-particle is shown to be tachyon-free in the open-string channel and to possess nine translational bosonic zero-modes plus a fermionic zero-mode structure, confirming complete stability and a spinor SO(32) representation. The analysis is extended to a broader set of non-BPS Type I branes, yielding a K-theory-consistent classification in agreement with previous mathematical results, and highlighting how boundary-state methods illuminate nonperturbative dualities in string theory.

Abstract

We use the boundary state formalism to study, from the closed string point of view, superpositions of branes and anti-branes which are relevant in some non-perturbative string dualities. Treating the tachyon instability of these systems as proposed by A. Sen, we show how to incorporate the effects of the tachyon condensation directly in the boundary state. In this way we manage to show explicitly that the D1 -- anti-D1 pair of Type I is a stable non-BPS D-particle, and compute its mass. We also generalize this construction to describe other non-BPS D-branes of Type I. By requiring the absence of tachyons in the open string spectrum, we find which configurations are stable and compute their tensions. Our classification is in complete agreement with the results recently obtained using the K-theory of space-time.