Perturbative Search for Fixed Lines in Large N Gauge Theories
A. Dymarsky, I. R. Klebanov, R. Roiban
TL;DR
This work investigates whether large-$N$ non-supersymmetric orbifolds of $\mathcal{N}=4$ SYM can possess a fixed line of conformal points emanating from the origin in the space of double-trace couplings. By computing one-loop beta functions for twisted double-trace operators across various orbifolds and accounting for $U(1)$ decoupling, the authors show that, in the examples studied, there is no fixed line through the origin at weak coupling; many beta functions remain positive or lack real zeros. The results illuminate how double-trace deformations propagate in orbifold theories, connect to AdS/CFT expectations about tachyon condensation, and motivate exploring special constructions (e.g., product orbifolds or $U(1)$-decoupled theories) where a fixed line might arise. The work therefore constrains perturbative non-supersymmetric large-$N$ conformality and highlights a potential link between gauge theory stability and closed-string tachyon dynamics.
Abstract
The logarithmic running of marginal double-trace operators is a general feature of 4-d field theories containing scalar fields in the adjoint or bifundamental representation. Such operators provide leading contributions in the large N limit; therefore, the leading terms in their beta functions must vanish for a theory to be large N conformal. We calculate the one-loop beta functions in orbifolds of the N=4 SYM theory by a discrete subgroup Gamma of the SU(4) R-symmetry, which are dual to string theory on AdS_5 x S^5/Gamma. We present a general strategy for determining whether there is a fixed line passing through the origin of the coupling constant space. Then we study in detail some classes of non-supersymmetric orbifold theories, and emphasize the importance of decoupling the U(1) factors. Among our examples, which include orbifolds acting freely on the S^5, we do not find any large N non-supersymmetric theories with fixed lines passing through the origin. Connection of these results with closed string tachyon condensation in AdS_5 x S^5/Gamma is discussed.
