SEarch for the Standard Model Higgs boson in the H→tau tau→ lepton + jet channel in the Vector Boson Fusion production

Maiko Takahashi, Alexandre Nikitenko and Costas Foudas

Introduction

The primary objective of the CMS experiment is to observe the Higgs boson as predicted in the Standard Model. While results from the LEP experiments leave a strong indication that the Higgs boson mass is below the threshold to allow decays into vector bosons, a decay channel with a pair of tau leptons becomes particularly important. The CMS detector with its excellent lepton identification performance and tracking system enables the reconstruction of such events to a high resolution. However, the characteristics of one or two high PT lepton(s) in an event is not enough to distinguish the Higgs boson signals among other high PT physics events at the LHC. The Vector Boson Fusion (VBF, qq→qqH) Higgs production process opens the possibility of selecting the signal events using additional signatures from the leading quark jets. The production cross section of the VBF process is the second highest at the LHC, and the H→tau tau decay channel has a moderate branching ratio, which enables the observation of the Higgs boson within a first few years of physics data taking at the LHC.

In the following, the discovery potential of the Higgs boson, produced via VBF and decaying to a tau lepton pair with one leptonic and the other hadronic final state, is discussed. Using a full simulation of the signal and a number of background processes in low luminosity environment, the event reconstruction methods and the selection and background rejection criteria are evaluated. The statistical significance of observable events is calculated for an integrated luminosity of 30 fb^-1, which corresponds to roughly three years of LHC low luminosity runs.

Analysis strategy

Particularity of the vector boson fusion:

Despite the fact that, at low Higgs mass, the cross section from the gluon fusion is at least an order of magnitude higher than the rest of the channels, the vector boson fusion (VBF) is an equally important production channel, since it creates two outgoing jets which can be tagged providing a characteristic signature of the Higgs event. The characteristics of the VBF process are that the two outgoing jets are mainly in the forward direction, and that the hadronic activity is heavily suppressed in the central region. This is due to the absence of colour exchange between the initial and the final state quark-jets (see Fig .1). The rapidity gap allows observation of the Higgs decay products in an isolated environment, and can also be used to distinguish VBF events from the background QCD processes which often generate central jets.

Figure 1: An illustration of the η-Φ distribution of the objects in a signal H→tau tau event (the location of each object drawn with a circle), superimposed on top of the statistical η distribution of the objects. The large separation between two forward jets from the VBF production process leaves the central region with very low hadronic activities

As shown in Fig .1, additional jets in the QCD events are centrally distributed with respect to the tagging jets, whereas for the EW events jets are mostly at high rapidity. This indicates that a veto based on the additional jets in the central region can be an effective method to reject background events with colour exchange.

Lepton tau:

The particularities of the tau leptons are excessively described in the reconstruction and selection of tau leptons (link to Sasha’s stuff). In this analysis, at least one of the tau produced from the Higgs decay must go into a lepton to allow an easier selection of these events while the other decays hadronically.

Reconstruction of the Higgs mass:

The discovery of the Higgs boson from the experimental data requires a reconstruction of its mass from its decay products which should form a resonance peak.

background processes

Irreducible Backgrounds

Irreducible physics background processes involve two correlated tau leptons and two high PT forward jets which have similar kinematics to that of the VBF process. The neutral Z bosons decaying into two tau leptons have topologically the same behaviour, except for the intrinsic difference in the tau polarization arising from the fact that the Z boson has spin = 1, and the Higgs has spin = 0. Hence, selections based on the behaviour of jets are particularly important. The two types of Z production processes considered are shown on Fig .2 and Fig .3.

Figure 2: Feynman diagrams for the main QCD Z+2jets background processes

Figure 3: Feynman diagrams for the main EW 2 tau+2jets background processes.

Reducible Background

A reducible background can be any process with two energetic jets, a high PT isolated lepton, a tau-jet or a narrow jet which can be misidentified as a tau-jet, and some missing ET in the event. It is relatively easy to distinguish these events from the signal process because of the absence of a kinematical constraint for a large rapidity gap between two high ET jets. Furthermore, the invariant mass of the two tau-like objects do not form a resonance peak, so the contamination from such events in the Higgs mass region should be uniform.

The reducible background processes include the W + multi-jet production with a leptonic decay of the W boson, and the ttb production where the top quarks decay weakly to bottom quarks producing two b-jets which may act as the tagging jets. Although reducible, both processes have sizable cross sections at the LHC, and may contribute substantially to the background of this Higgs production/decay channel. Hence, a large reduction factor from the event selection and a low fake rate in the tau-jet reconstruction are required.

Analysis Result

The invariant mass of the reconstructed tau's is shown in Fig .4 for the signal sample with M_H= 135 GeV/c² and for the background samples. The number of entries are normalized to the expected number of events at an integrated luminosity of 30 fb^-1 with all the selection cuts applied. A gaussian distribution is used to fit the signal distribution, and a Breit-Wigner and a second order polynomial function are used for the resonance peak from the Z/gamma* decays and for the reducible backgrounds respectively. 

Figure 4: Invariant mass distribution of the 2-tau system.

After all the selection cuts, the resolution of the Higgs mass distribution is of the order of 10%, and roughly 10 events are expected to be observed at an integrated luminosity of 30 fb^-1. An observation of the Higgs boson signal is possible with 3.9 σ significance at 30 fb^-1 in the mass region M_H = 135 GeV/c², and over 5 σ significance is achieved at 60 fb^-1.