New Prospects in Fixed Target Searches for Dark Forces with the SeaQuest Experiment at Fermilab

TL;DR

SeaQuest E906 beam-dump extension offers ultrasensitive probes of light hidden-sector gauge bosons, including Abelian dark photons and non-Abelian hadronic portals, by exploiting copious light-meson decays and initial-state radiation. The paper develops radiative meson-decay channels (π^0, η) and hadronic portals (ρ′, B′) within a unified framework, deriving decay widths and branching ratios, and proposes realistic experimental sensitivities for a 200-day run with a downstream pair spectrometer. It also discusses the impact of invisible decays and polarized beams on the attainable parameter space, and compares SeaQuest reach to existing experiments. Overall, the work outlines concrete strategies for SeaQuest to either discover hidden vector bosons or constrain broad classes of hidden-sector models.

Abstract

An intense, 120 GeV proton beam incident on an extremely long, iron target generates enormous numbers of light-mass particles that also decay within that target. If one of these particles decays to a final state with a hidden gauge boson, or if such a particle is produced as a result of the initial collision, then that weakly interacting, hidden-sector particle may traverse the remainder of the target and be detected downstream through its possible decay to an $e^+e^-$, $μ^+μ^-$, or $π^+π^-$ final state. These conditions can be realized through an extension of the SeaQuest experiment at Fermilab, and in this initial investigation we consider how it can serve as an ultrasensitive probe of hidden vector gauge forces, both Abelian and non-Abelian. A light, weakly coupled hidden sector may well explain the dark matter established through astrophysical observations, and the proposed search can provide tangible evidence for its existence --- or, alternatively, constrain a "sea" of possibilities.

Figures (8)

• Figure 1: Schematic of the SeaQuest spectrometer layout Isenhower:2001zz. The 120 GeV proton beam from the Fermilab Main Injector approaches the spectrometer through a 25-cm-long hole of 2.5 cm in diameter in the 5-m-long solid iron magnet. An $A'$ generated in the first meter of the beam dump traverses the Focusing Magnet (FMAG) without being affected by the magnetic field and can decay in the fiducial region into a lepton pair, or a pion pair (upon upgrade). Stations 1, 2, and 3 comprise a series of drift chambers and an array of scintillator hodoscope paddles used for track reconstruction and triggering purposes. The 3-m-long air-gap KTeV Magnet (KMAG) is used to focus the muons back into the spectrometer to facilitate momentum measurements. The 1-m long iron absorber wall is followed by an array of proportional tubes used for muon identification.
• Figure 2: Plot shows the projection contours of the coupling constant $\varepsilon$ as a function of dark photon mass $m_{A'}$ for four different processes that could be used to search for dark photons at SeaQuest. Regions I and II are bounded by the contour plots for $\eta \to \gamma A' \to \gamma e^{+}e^{-}$ and $\eta \to \gamma A' \to \gamma \mu^{+}\mu^{-}$, respectively, whereas regions III and IV refer to the limits inferred from use of the proton bremsstrahlung production mechanism, followed by $A'\to e^+ e^-$ (III) and $A'\to \mu^+ \mu^-$ (IV) decay. The area excluded by electron beam dump experiments E137 Bjorken:1988asBjorken:2009mm, E141 E141, and the searches by BABAR Lees:2014xha, CHARM Gninenko:2011uvGninenko:2012eq, NA48/2 Batley:2015lha are bounded by solid lines at 90 $\%$ CL, whereas those excluded by $\nu$-Cal I ($\pi^{0}$)Blumlein:2011mv and $\nu$-Cal I (p-Brem)Blumlein:2013cua are bounded by solid lines at 95 $\%$ CL. Also, the planned sensitivities of APEX (full run) APEX_FULL, HPS HPS, DarkLight Freytsis:2009bh (all at 90$\%$ CL) and LHCbLHCb (at 95$\%$ CL) are shown as dotted lines for comparison. We omit the anticipated limits from VEPP-3 Wojtsekhowski:2012zq, Refs. Gninenko:2013rkaAndreas:2013lya, Mu3e Echenard:2014lma, and MESA Beranek:2013yqa, which all probe lighter masses, as well as Ref. Alekhin:2015oba, for visual clarity. The region above $\varepsilon = 10^{-3}$ (not shown in the figure) has been excluded by several experiments such as E774 Bross:1989mp, APEX (test run)APEX_TEST, HADES Agakishiev:2013fwl, KLOE KLOE, PHENIX PHENIX, MAMI MAMI, along with the 2$\sigma$ exclusion limit obtained from $(g-2)_{e}$Davoudiasl:2014kua. Approximate limits at still weaker mixing angles from the LSND experiment Batell:2009diEssig:2010guLSND_expt and from astrophysical considerations Bjorken:2009mmDent:2012mxDreiner:2013muaFradette:2014szaFoot:2014ubaKazanas:2014mca have been omitted. Note that the limits shown all assume that decays of the $A'$ to the invisible sector are nonexistent.
• Figure 3: The branching ratios for ${\cal B}(A'\to e^+ e^-)$ (solid), ${\cal B}(A'\to \mu^+ \mu^-)$ (dotted), and ${\cal B}(A'\to \pi^+ \pi^-)$ (dashed) as per Eqs. (\ref{['pi2ee']}), (\ref{['Ap2pi']}), and (\ref{['fpi']}), assuming that decays to the hidden sector do not occur, as a function of the $A'$ mass in GeV.
• Figure 4: The $\eta$ (solid) and $\pi^{0}$ (dotted) yield/proton as a function of the energy of the particles as obtained from GEANT4 Monte Carlo simulations of 120 GeV protons interacting with the Fe beam dump.
• Figure 5: The $\eta$ (solid) and $\pi^{0}$ (dotted) yield/proton as a function of the transverse momentum $p_{T}$ obtained from GEANT4 Monte Carlo simulations of 120 GeV protons interacting with the Fe beam dump.