ATLAS Diboson Excesses from the Stealth Doublet Model

The ATLAS collaboration has reported excesses in diboson invariant mass searches of new resonances around 2 TeV, which might be a prediction of new physics around that mass range. We interpret these results in the context of a modified stealth doublet model where the extra Higgs doublet has a Yukawa interaction with the first generation quarks, and show that the heavy CP-even Higgs boson can naturally explain the excesses in the WW and ZZ channels with a small Yukawa coupling, \xi\sim 0.15, and a tiny mixing angle with the SM Higgs boson, \alpha \sim 0.06. Furthermore, the model satisfy constraints from colliders and electroweak precision measurements.


I. INTRODUCTION
Excesses in searching for diboson resonance using boson-tagged jets were recently reported by the ATLAS collaboration [1]. It shows local excesses in the W Z, W W and ZZ channels with significance of 3.4σ, 2.6σ and 2.9σ respectively. Similarly, the CMS collaboration [2,3] has reported an excess of 1.9σ significance in the dijet resonance channel and eνbb channel which may arise from W h with h decaying hadronically. These excesses may be evidences of new symmetries or new particles near 2 TeV.
The key points in explaining the excesses are the interactions of new resonance with the Standard Model (SM) gauge bosons, quarks and(or) gluons, the former of which is relevant to the branching ratio of the new resonance and the latter of which is relevant to the production of the new resonance at the LHC. One the one hand, one needs the couplings of new interactions to be large enough so as to give rise to a sizable production cross section at the LHC, on the other hand the strengths of these interactions should be consistent with current constraints of colliders and electroweak precision measurements. These two requirements are mutual restraint. A new resonance is not able to explain the ATLAS excesses if its interaction strengths are not mutually compatible with these two requirements.
In this paper, we explain the ATLAS excesses in the stealth doublet model, where the second Higgs doublet, H 2 , gets no vacuum expectation value, with mass near 2 TeV, and only the CP-even part of H 2 mixes with the SM Higgs boson. We assume H 2 has sizable Yukawa interaction with the first generation quarks, which is consistent with constraints of flavor physics. Such that the heavy CP-even Higgs boson can be produced at the LHC via the Yukawa interaction and decays into diboson states through the mixing with the SM Higgs boson. Our numerical simulations show that one has σ(pp → H → W W/ZZ) ∼ 5 fb by setting ξ ∼ 0.15 and α ∼ 0.06, where ξ is the Yukawa coupling of the H 2 with the first generation quarks and α is the mixing angle between two CP-even neutral states. This result is consistent with current constraints from colliders and electroweak precision measurements.
The remaining of the paper is organized as follows: In section II we give a brief intro-duction to the model. Section III is the study of constraints on the model. We investigate the ATLAS diboson excesses arising from this stealth doublet model in section IV. The last part is the concluding remarks.

II. THE MODEL
We work in the modified stealth doublet model [24,25], where the second Higgs doublet gets no vacuum expectation value (VEV) but its CP-even part mixes with the SM Higgs boson. In the following, we describe the modified stealth doublet model first, and then study its implications in the ATLAS diboson excesses. The Higgs potential is the same as that in the general two Higgs doublet model (2HDM), which can be written as In this paper, we assume the Higgs potential is CP-conserving, so all couplings in eq.(1) are real. Only one Higgs doublet gets nonzero VEV in the stealth doublet model, we take it be H 1 . The tadpole conditions for the electroweak symmetry breaking become where v 1 = √ 2 H 1 ≈ 246 GeV. After spontaneous breaking of the electroweak symmetry, there are two CP-even scalars h and H, one CP-odd scalar A and two charged scalars C ± , the mass eigenvalues of which can be written as [24] The mixing angle α between h and H can be calculated directly, we take it as a new degree of freedom in this paper. H interacts with dibosons through the mixing. We refer the reader to Ref. [24] for the feynman rules of Higgs interactions.
The Yukawa interactions of H 1 with SM fermions are exactly the same as Yukawa interactions of the SM Higgs with fermions in the SM. We assume H 2 has sizable Yukawa coupling with the first generation quarks: where Since H 2 = 0, there is almost no constraint on this Yukawa coupling, and H can be produced at the LHC via this interaction.

III. CONSTRAINTS
Before proceeding to study ATLAS diboson excesses, let us investigate constraints on the mixing angle α. Couplings of the SM-like Higgs to other SM particles were measured by the ATLAS and CMS collaboration. Comparing with SM Higgs couplings, couplings of h and H to all SM states (except u quark) are rescaled by cos α and sin α, respectively: where X represents SM states. Thus signal rates of the Higgs measurements relative to SM Higgs expectations are the function of cos α. Performing a global χ 2 fit to the Higgs data given by ATLAS and CMS, one has cos α ≥ 0.84 [26], at the 95% confidence level.
Another constraint comes from the oblique parameters [27,28], which are defined in terms of contributions to the vacuum polarizations of gauge bosons. The explicit expressions of ∆S and ∆T , which involve effects of all scalars, can be written as [29] where B i (x; y, z) = B i (x; y, z) − B i (0; y, z), i = (0, 2), the expressions of B i (x; y, z) and A 0 (x) can be find in Ref. [29], c = cos α and s = sin α, s W = sin θ W with θ W the weak mixing angle, M Z and M W are masses of Z and W bosons respectively.
The ∆χ 2 can be written as where O 1 = S and O 2 = T ; σ 2 ij = σ i ρ ij σ j with ρ 11 = ρ 22 = 1 and ρ 12 = 0.891. As can be seen from eqs. (8) and (9) where n C = 3, being the color index; V = W, Z respectively. We show in the left panel of be produced at the LHC, the model is constrained by other LHC experimental results. We will discuss these constraints one-by-one as follows: • The CMS collaboration [31] has reported an upper bound for the σ(pp where R is a new resonance. It gives σ(pp → R → W + h) ≤ 7 fb. The resonance can be the charged component of the heavy scalar doublet in our model. Its decay rate can be written as where λ(x, y, z) = x 2 + y 2 + z 2 − 2xy − 2xz − 2yz and g is the SU (2) gauge coupling. • The CP-odd component of the heavy scalar doublet can be the mediator of the process pp → R → Zh, which was also measured the CMS collaboration. One has σ(pp → where c W = cos θ W with θ W the weak mixing angle. A can also decay into dijet final states with the decay rate the same as eq. (12). We show in FIG. 3

V. SUMMARY
We investigated the prospects of the stealth doublet model as a possible explanation to the diboson excesses observed by the ATLAS collaboration. The mass of heavy Higgs boson was fixed at near 2 TeV in our study. We showed that excesses in the W W and ZZ channels can be interpreted as the decay of the heavy CP-even Higgs boson H, which can be produced at the LHC via its Yukawa interaction with the first generation quarks. One needs the Yukawa coupling ξ ∼ 0.15 and the mixing angle between two CP-even Higgs bosons α ∼ 0.06, which is consistent with precision measurements, so as to has a 5 fb production cross section at the LHC. Constraints on the model from the exclusion limits in W h and Zh channels given by CMS collaboration and dijet searches was also studied, which showed the limited parameter space (in FIG. 3) that can be accommodated with the interpretation of the ATLAS diboson excesses in the same model. We expect the running of the 13 TeV LHC to tell us the detail about the diboson excesses and show us more clear hints of new physics behind this phenomena.

ACKNOWLEDGMENTS
The author thanks to Huaike Guo and Peter Winslow for very helpful discussions. This work was supported in part by DOE Grant de-sc0011095