The discovery of neutrino oscillations provided conclusive evidence of lepton flavor violation. Lepton flavor violation in the neutrino sector also results in charged lepton flavor violation (CLFV) through loop-suppressed processes such as µ → eγ. However, the resulting predicted CLFV rates are highly suppressed due to the small neutrino masses -Br(µ → eγ) < 10 -54 -and are far beyond the reach of any current or planned experiments. On the other hand, many Beyond Standard Models (BSM) scenarios predict CLFV rates that are both much larger and within reach of ongoing or near-future experiments. For example, supersymmetry-based models predict rates as high as Br(µ → eγ) ∼ 10 -15 [1], while the current experimental limit on the µ → eγ process already reached Br(µ → eγ) < 4.2 × 10 -13 [2]. On the other hand, while there have been extensive searches for CLFV processes between the first and second lepton generations, denoted as CLFV (1,2) for brevity, the constraints on CLFV (1,3) processes that involve e ↔ τ transitions are weaker by several orders of magnitude. These constraints on CLFV (1,3) [3,4] were obtained through searches for e + p → τ + X, τ → eγ, and p+ p → e+τ+X at HERA [5,6], BaBar [7], and the LHC [8] respectively. However, there are many BSM scenarios such as grand unified theories with leptoquarks and Rparity violating supersymmetry that predict CLFV (1,3) rates that are enhanced compared to CLFV(1,2) processes, motivating continued searches dedicated for e ↔ τ transitions.
We carry out the simulation analysis based on the design of the ECCE Detector (recommended as Detector 1 by the EIC Detector Proposal Advisory Panel [9]), for determining the sensitivity to the CLFV(1,3) process e + p → τ+X in the leptoquark framework [3,4], though such analysis could also be performed in the SMEFT framework [4].
Leptoquarks are color triplet bosons that carry both lepton (L) and baryon (B) numbers, coupling leptons to quarks and mediating the e + p → τ + X CLFV (1,3) process at tree-level, as shown in Fig. 1. The leptoquarks are classified into 14 types [10] based on their fermion number provide complementary information [4] to the constraints
Figure 1: From [3]: Representative Feynman diagrams for e → τ scattering processes via one-leptoquark mediator. The fermionic number F is assumed to be conserved, as in the BRW effective model [10]. The partonic cross section is convoluted with the PDF of the initial state (anti)quark of each diagram, and depends on the parameter λ 1α λ 3β /M and can be studied in the near future. In this study, we fo-134 cus only on searching for the τ "3-prong" decay events.
135
The features of "3-prong" leptoquark events include: To reconstruct the secondary vertex, we first look for 3π candidate events. In the following, the charged pion's tracking information is from the simulated tracking and the vertex detector responses, though particle identification (PID) is based on generator information (namely, perfect PID is assumed). In our algorithm, one track is matched to a second track and the middle point at the closest approach is the candidate secondary vertex position which will be further justified based on the topological structure. For a given 3-π candidate event, there are three such pair-combinations and we can reconstruct three "intermediate" vertices and the candidate vertex will be the average of all three. Cross section sensitivity for leptoquark search vs number of residual background events for 100 fb -1 integrated luminosity. The grey line corresponds to the scenario that only "3-prong" decay modes are detected. The blue line corresponds to the scenario where electron and pion "1-prong" decay modes could be detected with 50% efficiency of the "3-prong" case. And the red line shows the scenario if all decay modes were detected at the same efficiency as the "3-prong" case. ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ---------