[Paper Review] Searches for Charged Higgs Bosons at LEP

André Holzner|ArXiv.org|May 17, 2001

Particle physics theoretical and experimental studies3 citations

TL;DR

This paper reports on a search for charged Higgs bosons in e⁺e⁻ collisions at LEP, using 210 pb⁻¹ of data collected in 2000. While the L3 experiment observed a 4.4σ excess at ~68 GeV in the H⁺→τν decay channel, no such signal was seen by other LEP collaborations, leading to a combined 95% confidence level lower limit of 78.5 GeV on the charged Higgs mass, despite the preliminary nature of the results.

ABSTRACT

The four LEP experiments Aleph, Delphi, L3 and Opal updated their searches for pair production of charged Higgs bosons using more than 210/pb luminosity collected per experiment in the year 2000. Combining it with previously collected data, a significant deviation from background (equivalent to 4.4 sigma) is found by the L3 collaboration for low values of the branching ratio H+/- -> tau nu around masses of about 68 GeV. This excess is however not seen by the other LEP collaborations and thus a lower limit on the charged Higgs mass is set at 78.5 GeV at 95% confidence level. All results reported here are still preliminary.

Motivation & Objective

• To search for pair-produced charged Higgs bosons (H⁺H⁻) in e⁺e⁻ collisions at LEP using increased luminosity data.
• To test the existence of charged Higgs bosons predicted in Two Higgs Doublet Models (2HDMs), particularly in the H⁺→τν decay mode.
• To resolve discrepancies between LEP experiments by combining results and setting a robust lower limit on the charged Higgs mass.
• To investigate the nature of a 4.4σ excess observed by L3 at ~68 GeV, which was not confirmed by other collaborations.

Proposed method

• The analysis used e⁺e⁻ collision data at center-of-mass energies up to ~200 GeV, focusing on H⁺H⁻ pair production via Z*/γ* exchange.
• Three final-state channels were analyzed: fully leptonic (τ⁺ντ⁻ν̄), semileptonic (c s̄ τ⁻ν̄), and fully hadronic (c s̄ c̄ s) with four jets and no missing energy.
• Each experiment applied distinct event selection techniques: likelihood methods (Delphi), linear discriminants (Aleph), and neural networks (L3) to separate signal from background.
• Backgrounds were dominated by W⁺W⁻, q q̄, and γγ events; signal regions were optimized using kinematic variables like τ polarization, Higgs production angle, and jet mass pairing.
• The L3 four-jet analysis used a mass-maximizing jet pairing algorithm and a cut on the Higgs production polar angle to reduce W⁺W⁻ background.
• Results were combined across Aleph, Delphi, L3, and Opal using a profile-likelihood approach to set upper limits on branching ratios and set a lower mass limit at 95% confidence level.

Experimental results

Research questions

• RQ1 Does the L3 experiment's observed 4.4σ excess at ~68 GeV in the H⁺→τν decay channel represent a genuine charged Higgs signal?
• RQ2 Why is the excess observed by L3 not confirmed by the other LEP experiments (Aleph, Delphi, Opal)?
• RQ3 What is the combined lower limit on the charged Higgs boson mass when all LEP data are considered together?
• RQ4 How do the limits on the charged Higgs mass depend on the branching ratio BR(H⁺→τν)?
• RQ5 Can the observed discrepancy between L3 and the other LEP experiments be explained by statistical fluctuations or detector/systematic effects?

Key findings

• The L3 collaboration observed a 4.4σ excess in the H⁺H⁻ pair production search at a Higgs mass of approximately 68 GeV, primarily in the H⁺→τν decay channel.
• No significant excess was observed by Aleph, Delphi, or Opal in their respective analyses of the same final states.
• When combining data from all four LEP experiments, the significance of the excess is reduced, and a 95% confidence level lower limit of 78.5 GeV is set on the charged Higgs boson mass.
• The observed lower limit on m(H⁺) is 78.0 GeV for Aleph, 73.8 GeV for Delphi, 67.1 GeV for L3 (individual), and 72.2 GeV for Opal, with the combined limit being 78.5 GeV.
• The expected limit (median background-only) for the combined analysis is 80.1 GeV when BR(H⁺→τν) = 0, and 91.7 GeV when BR(H⁺→τν) = 1.
• The excluded region in the m(H⁺) vs. BR(H⁺→τν) plane shows that the signal region around 68 GeV is inconsistent with the combined data at more than 3σ significance.