Long-Distance Contribution to \Delta\Gamma s/\Gamma s of the $B_s-\bar{B}_s$ System

Abstract

We estimate the long-distance contribution to the \Delta{\Gamma}s of B_s system, based mainly on two-body D^(*)_s\Dbar^(*)_s modes and three-body D^(*)_s\Dbar^(*)\Kbar^(*) modes (and their CP conjugates). Some higher c\bar{s} resonances are also considered. The contribution to \Delta{\Gamma}s/{\Gamma}s by two-body modes is 10.2+-3.0%, slightly smaller than the short-distance result of 13.3+-3.2%. The contribution to width difference by D*_s0(2317), D_s1(2460), and D_s1(2536) resonances is negligible. For the three-body modes, we adopt the factorization formalism and model the form factors with off-shell D^(*)_s poles, the D_{sJ}(2700) resonance, and non-resonant (NR) contributions. These three-body modes can arise through current-produced or transition diagrams, but only SU(3)-related D^(*)_{u,d} \Dbar^(*)\Kbar modes from current diagram have been measured so far. The pole model results for D^*_{u,d} \Dbar^(*)\Kbar agree well with data, while D_{u,d} \Dbar^(*)\Kbar rates agree with data only within a factor of 2 to 3. All measured three-body rates can be reproduced by including NR contribution. The total \Delta{\Gamma}s/{\Gamma}s obtained is 16.7+-8.5%, which agrees with the short-distance result within uncertainties. For illustration, we also demonstrate the effect of D_{sJ}(2700) in modes with D^(*)K^*. In all scenarios, the total width difference remain consistence to the short-distance result. Our result indicates that (a) the operator product expansion (OPE) in short-distance picture is a valid assumption, (b) approximating the two-body decays to saturate width difference has a large correction, (c) the effect of three-body modes cannot be neglected, and (d) in addition to D^(*)_s poles, the D_{sJ}(2700) resonance also plays an important role in three-body modes. Future experiments are necessary to improve the estimation of \Delta{\Gamma}s from long-distance picture.