ε′/ε in 331 models

Motivated by the recent findings that the ratio epsilon'/epsilon in the Standard Model (SM) appears to be significantly below the data we investigate whether the necessary enhancement of epsilon'/epsilon can be obtained in 331 models, based on the gauge group SU(3)(C) x SU(3)(L) x U(1)(X). In these models new physics (NP) contributions to epsilon'/epsilon and other flavour observables are dominated by tree-level exchanges of a Z' with non-negligible contributions from tree-level Z exchanges generated through the Z - Z' mixing. NP contributions to epsilon'/epsilon in these models are governed by the electroweak penguin operator Q(8) for which the hadronic matrix element is already rather well known so that our analysis of NP contributions is subject to much smaller theoretical uncertainties than within the SM. In particular strong cancelations between different contributions do not take place. The size of NP effects depends not only on M-Z' but in particular on a parameter beta, which distinguishes between various 331 models. Also a parameter tan (beta) over bar present in the Z - Z' mixing plays a role. We perform the analysis in seven 331 models characterized by different beta, tan (beta) over bar and fermion representations under the gauge group that have been selected in our earlier analysis on the basis of electroweak precision data. Imposing the constraints from Delta F = 2 transitions we find that only three of these models can provide a significant positive shift in epsilon'/epsilon up to 6 x 10(-4) for M-Z' = 3TeV. Two of them allow simultaneously a supression of the rate for B-s -> mu(+)mu(-) by 20% thereby bringing the theory closer to the data without any significant impact on the Wilson coefficient C-9. The third model provides simultaneous shift Delta C-9 = -0.6, softening the anomalies in B -> K*mu(+)mu(-), without any significant impact on B-s -> mu(+)mu(-). NP effects in rare K decays, in particular in K+ -> pi(+)nu(nu) over bar, turn out to be small. This is also the case of B -> K (K*)nu(nu) over bar. Both predictions could be challenged by NA62 experiment and Belle II in this decade. The special flavour structure of 331 models implies that even for M-Z' = 30TeV a shift of epsilon'/epsilon up to 8 x 10(-4) and a significant shift in epsilon(K) can be obtained, while the effects in other flavour observables are small. In this manner epsilon'/epsilon and epsilon(K) appear to be unique flavour observables in these models which provide the possibility of accessing masses of M-Z' far beyond the LHC reach.