Search for Fourth Generation Quarks

It is still a mystery why the Standard Model as we know it has only three families. At new high energy colliders it is worthwhile to search for a new additional family which obviously would have a heavy neutrino to avoid the LEP bounds. This paper discusses new studies made with the CMS detector for the search of new heavy b‐like quarks in several different decay modes and for different possible mass regions. These studies are based on detailed detector simulation, including all Standard Model backgrounds. Particular emphasis is given to possible early discoveries, i.e. with 100 pb−1 or less. Projected 95% CL exclusion limits as a function of luminosity are presented as well.


INTRODUCTION
The Standard Model (SM) describes successfully most of the physics phenomena with three generations of quarks. The precision measurements at Z pole in the Large Electron-Positron Collider (LEP) experiment exclude light fourth generation neutrino with mass less than half of Z mass. However, it does not exclude the existence of fourth generation neutrino.
The existence of an irreducible phase in the quark mixing matrix provides the only source of CP violation in the SM. This source of CP violation is too small to account for the asymmetry between matter and antimatter in the university. The existence of fourth generation quarks may rescue the model [1].
CDF at the Tevatron has performed searches for the fourth generation quarks. For b' quark decays into tW, CDF has announced their result [2]. The study consists of searching for same-sign dilepton pair. The limit on the b' mass is 325 GeV at 95% confidence level. CDF also performs a search for t'→qW and the lower limit on t' mass is 311 GeV at 95% confidence level [3]. For b'→bZ channel, the lower limit on the b' mass (> 268 GeV at 95% confidence level) is provided by CDF [4]. The result is obtained by assuming 100% b'→bZ decay branch fraction.
With different assumptions of the relative strength of |V cb' | and |V tb' |, both b'→cW and b'→t (*) W (or tW (*) ) could be the major b' decays if M b' < 255 GeV. For the case of M b' > 255 GeV, the decay channel b'→tW is expected to be dominant and the transition b'→cW would be suppressed except for |V cb' | >> |V tb' |. The decay b'→t' should be suppressed if M t < M t' < M b' [5] and b'→b mode only occurs through second order loop diagrams. This paper only focuses on bottom-like fourth generation quark (b') with M b' > 300 GeV and the branch ratio of b'→tW is assumed 100%. When both b' quarks decay into top quark and W boson, the final state contains four W bosons. Each boson can either decay leptonically or hadronically. The same-sign dileptonic ( Figure 1) and trileptonic channels are selected in final state. The signature of same-sign dileptonic channels is very rare for SM processes, and therefore the backgrounds are negligible small. The branch ratio of trileptonic mode is much smaller than the dileptonic channels and thus is free from most of the SM backgrounds.
A dataset corresponding to 100 pb -1 integrated luminosity at 14 TeV is assumed for the search and exclusion limits. The Monte Carlo (MC) samples, produced with a fast simulation of the CMS detector [6], are used in this analysis.

EVENT SELECTIONS AND EVENT YIELD
The events are selected to pass either single non-isolated electron trigger or single non-isolated muon trigger. Four types of objects are used in this analysis: electrons, muons, jets and missing transverse energy (MET). These selections are optimized by Genetic Algorithm and the optimization is based on 300 GeV b'. The detail selections could be found in Ref. [7].
Signal events can be characterized with the variable, H T , which is defined as: H T = ∑P T (jets) +∑P T (leptons)+MET. H T is the most effective variable that carries the information of b' mass, sine b' mass cannot be reconstructed in this analysis. The distributions of H T with 100 pb -1 integrated luminosity are shown in Figure 2. The signal is significant for the low b' mass (300 ~ 400 GeV). The expected signal and background yields are listed in Table 1. The background to signal is expected to be small, but a data-driven method is still introduced for the background estimation.

EXCLUSION LIMIT
The exclusion limit on the b' mass is estimated using Bayesian statistics for null hypothesis. Comparing to the Pythia LO cross section, we are able to exclude the production of pp→b'b' up to 480 (420) GeV at 95% confidence level with 100 (30) pb -1 integrated luminosity (see Figure 3).