In collider physics at the TeV scale, there are many important processes which involve six or more jets. The sensitivity of the physics analysis depends critically on the performance of the jet clustering algorithm. We present a full detector simulation study for the ILC of our new algorithm which makes use of secondary vertices which improves the reconstruction of b jets. This algorithm will have many useful applications, such as in measurements involving a light Higgs which decays predominantly into two b quarks. We focus on the measurement of the Higgs self-coupling, which has so far proven to be challenging but is one of the most important measurements at the ILC.
Jet clustering is an essential technique in high energy physics experiments in which the multitude of produced particles are combined into jets which represents an attempt to reconstruct the originating quarks and gluons in the final state. The development of jet clustering algorithm has a long history ever since QCD jets have been produced in particle collisions; we name a few examples in lepton colliders such as the Jade algorithm [1], the Durham algorithm [2], and the Cambridge algorithm [3]. These algorithms have dealt with the challenging question of how to deal with gluon emissions of various energies. With the advent of future lepton colliders at the TeV scale, the number of quarks in the final state increases roughly with the collision energy, which makes even more challenging to correctly group the resulting hadrons into their originating partons. There is further complication arising from the imbalance of the parton energies due to the difference in their origin, such as whether they directly come from e + e -collisions or from W or Z boson decays, and also because of initial state radiation which adds a boost to the system.
We will focus on the physics application of jet clustering at a future lepton collider, such as the International Linear Collider (ILC), although applications to hadron colliders should be possible with minor adjustments.
The ability to group the particles according to their originating parton is particularly important in the analysis of physics processes involving multi-jet final states, such as the measurement of the Higgs self-coupling, which uses the e + e -→ ZHH channel for √ s = 500 GeV, or the top Yukawa coupling, which uses the e + e -→ ttH → bW + bW -H channel. Depending on the decay modes of the W, Z, and the Higgs, the number of jets in the final state can be as high as 6 for ZHH and 8 for ttH. This is relevant especially in the case of a light Higgs particle as motivated by electroweak precision measurements, whose branching ratio of H → bb is 68% for a Higgs mass of 120 GeV. These channels have major background processes with similar number of jets; particularly important is e + e -→ tt whose has a large cross section. Many such processes can be greatly reduced if the flavor of the originating quark can be identified; by requiring the correct number of reconstructed jets originating from b quarks (“b jets”), the tt background could be eliminated. In reality, flavor identification itself is a challenging task, which results in a leakage of tt background even with an efficient flavor identification algorithm, which results in significant background due to the sheer size of the cross section. Other backgrounds include those in which the Higgs decay H → bb is replaced by the Z decay Z → bb, which becomes an irreducible background.
Flavor identification can be accomplished by looking for signs of secondary decays of b hadrons whose proper lifetime is typically 400-500 µm/c. This results in, for example, a heavy tail in the impact parameter distributions of charged tracks, secondary vertices which are displaced from the primary vertex, increased transverse momentum relative to the jet direction due to the heavy b hadron, as well as the presence of leptons due to semileptonic decays of the W boson. Such signatures are typically combined using a multivariate analysis technique [4] into a single variable which can be used to discriminate b jets from jets originating from lighter quarks. Similar techniques can be applied to identify c jets.
Traditionally, the jet clustering procedure is performed first, after which the flavor identification algorithm is applied to each of the resulting jets. The search for secondary vertices is restricted to combination of particles within the jet, which reduces the computing cost arising from combinatorial effects. This method has the consequence that mistakes in jet clustering, such as particles originating from the same vertex being associated into separate jets (vertex splitting), or the inclusion of multiple vertices of b origin into a single jet (vertex merging), cannot be fixed at a later stage. As computing resources grow inexpensive, performing the vertex finding procedure using all particles in the event can be performed in a reasonable amount of computing time. Our methods exploit this fact and use it to improve the jet clustering procedure. In this study, we show that, in multi-jet environment, the accuracy of jet clustering is significantly improved by this method.
The software framework used in this study is based on LCIO [5]. The detector simulation is performed by Mokka, a Geant4 based program. Collisions of electron and positron beams are simulated with the International Large Detector (ILD) Concept [6] at a center-of-mass energy of 500 GeV. Initial state radiation and beamstrahlung effects are included. The event reconstruction is done using the Marlin framework, which consists of a series of modules which perform hit
This content is AI-processed based on open access ArXiv data.
