A search for R-parity violating Supersymmetric top decays at CMS with \(\sqrt{s} = 8\) TeV


A search for a supersymmetric top decay assuming a 100% branching ratio of \(\widetilde{t}\widetilde{t}^* \rightarrow \mu^+ \mu ^- b \overline{b}\) is presented using a minimally supersymmetric model at an integrated luminosity of \(19.5\ \mathrm{fb}^{-1}\). The datasets were recorded with the CMS detector at the LHC. Using Baysian marginalization, an upper limit on the cross section of this process is computed and a cut off point is calculated below which the data does not support the presence of the target decay. This cut off point was calculated to be around 780 GeV.



With the discovery of the Higgs, the Standard Model seems to offer a complete tally of the fundamental particles with which all of the matter around us is made.(Aad 2012) However, with this answer comes more questions as phenomena such as the origin of dark matter and the seemingly unnatural Higgs mass remain unanswered, and cause theorists to look for physics beyond the Standard Model.

Supersymmetry is one such theory and proposes an extension to space-time which allows for a symmetry relating the two groups of fundamental particles: fermions and bosons, where each of the particles from one group have a corresponding supersymmetric partner differing by half-integer spin. These supersymmetric partners provide opposite quantum corrections to the mass of the Higss and thus yield a natural explanation for the Higss mass. Associated with this extension is a quantum number \(R\), defined as

\[R = (-1)^{(3B + L + 2S)}\]

where S, B, and L, are the spin, baryon, and lepton quantum numbers of the particle.(Huh 2009)

It is important to note that due to the definition of this quantum number, particles in this theory must be created in a pair-wise fashion. Because of this, this quantum number is refered to as ’R-parity.’ It is also important to note that Supersymmetric particles have an R value of -1 whereas particles that are of the Standard Model have R = 1.

In order for energy and R-parity to be conserved, Supersymmetric particles must decay pair-wise through a series of intermediate processes and have an eventual final state of lighter Standard particles and some number of the lightest supersymmetric particle (LSP).(Berger 2013) However, if R-parity is not conserved, a Supersymmetric particle can have a final state which consists only of particles found in the Standard Model.

While there have been a few searches for R-parity violating decays in the past, R-parity conserving theories are more widely researched as they provide an explanation for the massive prevalence of seemingly weakly interacting dark matter.(Trotta 2007)(Lahanas 2007) However, there is little reason to believe a priori that spacetime would behave in such a way as to conserve \(R_p\). Coupling this with the increased capacity for data collection available to the Compact Muon Solenoid (CMS) at LHC at CERN, it is necessary to perform searches for these decays using new techniques which might uncover possibly interesting results.

This paper describes a search for such a decay of the form:

\[\widetilde{t}\widetilde{t}^* \rightarrow \mu^+ \mu ^- b \overline{b}\]

Since the signature of this decay resembles that of many common Standard Model vertices, various data-based background estimation techniques were implemented in conjunction with Monte Carlo simulations to obtain a number of expected events. For a detailed description of these techniques see section \ref{sect:eventSelection}.

The CMS detector

\label{sect:CMS} FIX ORDER The primary feature of the CMS detector is a superconducting solenoid that is 6m in diameter, 13m long, and generates a 3.8T field which is used to determine the sign of the charge of the particle, as well as it’s momentum. On the inside of the magnet is an electromagnetic calorimeter (ECAL) which is made of lead tungstate crystals and collects and measures electromagnetic deposits. This primarily detects electrons and photons, muons are more massive and therefore have more energy to deposit before being stopped, allowing them to travel past the ECAL and further into the detector. The ECAL surrounds a silicon tracker which matches particle interactions with other areas of the detector in order to reconstruct trajectories. Inside the tracker is the beam pipe in which particles are accelerated by the superconducting solenoid. After the products of the decay have passed through the tracker and ECAL they reach a hadronic calorimeter (HCAL) made of dense brass and steel designed to stop the massive energy deposit from hadronization, this absorbs most of the energy left in the collision. The particles that do make it through the HCAL are either neutrinos or muons. The muons are detected and collected in a separate configuration around the magnet composed of a drift tube and cathode-strip detector and since the effects of baryonic matter on neutrinos is taken to be negligible by the cross section, neutrinos appear in the data as missing energy in the transverse direction (MET).(AUFFRAY 2002)(Pooth 2010)(Khachatryan 2014)

CMS uses a right-handed coordinate system whose origin is at the point of interaction between the two proton beams, \(x\)-axis pointing to the center of the LHC, \(y\)-axis perpendicular to the plane created by the beam, and \(z\)-axis in the anticlockwise-beam direction. \(\theta\) measured from the positive \(z\)-axis.(Khachatryan 2014) For a detailed description of the CMS detector see (Collaboration 2008).

Event selection and Monte Carlo simulation


Each event is processed using a global event reconstruction algorithm which reconstructs and indentifies the particles using information from all of the subdetectors as described in section \ref{sect:CMS}. It is important to note that this algorithm relies heavily on the identifacation of the particle type (photon, electron, muon, hadron) to reconstruct the trajectory of the particle.(Khachatryan 2014) After events are reconstructed, they are selected from an online dilepton trigger (events only include those whose final state constain \(ee\), \(e\mu\), \(\mu e\), and \(\mu\mu\)) which requires the tranverse momentum (\(p_T\)) of the largest lepton to be greater than 17 GeV and the smaller to be greater than 8 GeV.

The reconstructed trajectories (known as hypotheses) are then grouped together according to the location in the LHC where the interaction occured. This is done to prevent particles that originated in other interactions (pileup) from being considered as part of the final state. Each hypothesis is restricted to non \(ee\) events with \(p_T >\) 20 GeV, oppositely charged leptons, \(|\eta| < 2.4\) where \(\eta \equiv - \ln{\frac{\theta}{2}}\), and have a total invariant mass greater than 20 GeV. A minimum isolation is also required according to the loose working points of the combined secondary vertex (CSV-L) algorithm. (Khachatryan 2014) The hypothesis with the highest combined \(p_T\) bewtween the two leptons that passes these criteria is then taken to be the primary interaction at the corresponding LHC bunch crossing.

After the primary hypothesis has been found, the hadronic jets corresponding to the event are reconstructed using the anti-\(k_T\) algorithm.(Cacciari 2008) The jet momentum is determined as the vector sum of all particle momenta in the jet. For this analysis, the jets considered are required to be inside the tracker acceptance (\(|\eta|\) \(<\) 2.4) as well as a scaled \(p_T > 20\) GeV. The primary jets are taken to be the two jets in the bundle which when paired with the leptons in the primary hypothesis, minimize the total invariant mass. This is done under the assumption that the most likely process to occur is the one with the smallest amount of \(p_T\). Once the event jets have been reconstructed, the collection of primary hypotheses are then restricted to those events with a \(p_T > 20\) GeV as well as requiring at least 2 jets to have passed the criteria above. A minimum combined invariant mass of 20 GeV as well as a minimum difference in mass between the two pairings of 100 GeV is set to prevent events containing fakes and photons to be considered in the analysis. In order to prevent events originating from a process similar to Drell-Yan, a z-veto was added to the selection criteria by restricting events to having a combined invariant momentum that does not fall within 15 Ge