Muon Pair Production in ep Collisions at HERA

Muons are elementary particles, very similar to electrons, but with a mass which is 200 times larger. In contrast to electrons, their large mass allows them to penetrate great distances through matter. Muons with an energy greater than a few GeV thus tend to escape from particle physics detectors, but they leave tracks behind them which provide a clear signature. Muons therefore offer a unique testing ground for rare and exotic processes, provided that the standard model backgrounds are well understood. In electron proton scattering, muon pairs are mainly produced in two-photon interactions, where the incoming photons are radiated from the beam particles. The Feynman diagram for this dominant production mechanism is shown on the right. The cross section becomes large when the momentum transfer to the scattering particles is small and the photons are quasi-real. Since the photon-photon interaction and the photon coupling to the electron are governed by electroweak interactions, they can be precisely calculated using the Standard Model of particle physics. The only large uncertainty in the calculation arises from the photon flux associated with the proton.

The data collected by the H1 collaboration at an electron proton centre of mass energy of 319 GeV have been used to extract cross sections for muon pair production, in order to test our understanding of the production mechanisms and search for new physics beyond the Standard Model. Differential cross sections have been measured for the production of muon pairs (ep -> e mu mu X) as a function of the di-muon invariant mass, the transverse momenta of the muons and the transverse momentum of the hadronic recoil, X. On the basis of the level of observed hadronic activity, the cross sections have also been separated into an elastic component, where the proton remains intact, and an inelastic component, in which the proton dissociates to form a higher mass system X. All measured cross sections were found to be in good agreement with the Standard Model predictions. As an example, the figure on the left shows the cross section as a function of the di-muon mass. The data agree well with the electroweak prediction (EW GRAPE), which contains the dominant two-photon process together with other smaller contributions from Bremsstrahlung processes in which radiated photons convert to muon pairs and contributions in which heavy Z⁰ bosons replace the photons. The predictions for other contributions to two muon production, due to Upsilon production, heavy flavour and tau lepton decays are also shown. These contributions are small and are only present at low di-muon masses.

In another recent H1 paper ( DESY-03-082 ), the closely related topic of multi-electron production was studied. Excesses over the Standard Model predictions were observed in the numbers of events containing two electrons and three electrons for which the invariant mass M₁₂ of the two electrons with the largest transverse momentum was larger than 100 GeV. The well established principle of lepton universality implies that any such anomaly in multi-electron production is likely to be visible also in muonic final states. It is therefore worthwhile to analyse the muon pair data in a similar way as is done in the multi-electron analysis. In order to optimise the statistics for this part of the analysis, the full HERA-I data, including early data at an ep centre of mass energy of 301 GeV, are included. Final states containing two muons (mu-mu) or two muons and an electron (mu-mu-e) are studied. For masses M₁₂ > 100 GeV, one mu-mu event is found, while 0.08 ± 0.01 are expected from the Standard Model predictions. No event classified as mu-mu-e with M₁₂ > 100 GeV is observed. The prediction is 0.05 ± 0.01. With such small event numbers, a comparison of the findings in the electron and muon channels is inconclusive. We therefore look forward eagerly to the large increases in statistics expected in the second phase of HERA operation.