Search for eV (pseudo)scalar penetrating particles in the SPS neutrino beam

Abstract

We carried out a model-independent search for light scalar or pseudoscalar particles a's (an example of which is the axion) that couple to two photons by using a photon-regeneration method at high energies allowing a substantial increase in the sensitivity to eV masses. The experimental set-up is based on elements of the CERN West Area Neutrino Facility (WANF) beam line and the NOMAD neutrino detector. The new particles, if they exist, could be produced through the Primakoff effect in interactions of high energy photons, generated by the 450 GeV protons in the CERN SPS neutrino target, with virtual photons from the WANF horn magnetic field. The particles would penetrate the downstream shielding and would be observed in the NOMAD neutrino detector through their re-conversion into real high energy photons by interacting with the virtual photons from the magnetic field of the NOMAD dipole magnet.

From the analysis of the data collected during the 1996 run with 1.08×10¹⁹ protons on target, 312 candidate events with energy between 5 and 140 GeV were found. This number is in general agreement with the expectation of 272±18 background events from standard neutrino processes. A 90 $\text{\%}\text{CL}$ upper limit on the aγγ-coupling g_aγγ< 1.5$\text{×10}{}^{\mathrm{-4}}\text{GeV}{}^{\mathrm{-1}}$ for a masses up to 40 eV is obtained.

Introduction

Neutral spin-zero scalar or pseudoscalar particles a's (this notation will be used for both cases) of nearly zero mass are predicted in many theories. The most motivated pseudo-Nambu–Goldstone bosons, for example, arise in models constraining spontaneously broken symmetry, see e.g. [1]. They couple to the divergence of the current whose charge generates the symmetry which is spontaneously broken. The most popular light pseudoscalar, the axion, postulated [2]to provide a solution of the “strong CP” problem, emerges as a consequence of the breaking of the Peccei–Quinn symmetry [3]. It is now believed that the axion has a mass much smaller than $\text{∼O(100)}\text{keV}$ that was originally expected 4, 5, 6. The axion two-photon interaction is given by the Lagrangian

$$\text{L}{}_{\mathrm{<mtext>int</mtext>}}\text{=−}\text{1}\text{4}\text{g}{}_{\mathrm{a\gamma \gamma }}\text{F}{}_{\mathrm{\mu \nu }}\text{F}\text{̃}{}^{\mathrm{\mu \nu }}\text{a=g}{}_{\mathrm{a\gamma \gamma }}\text{(}\text{E}\text{·}\text{B}\text{)·a}$$

where g_aγγ is the coupling constant, $\text{E}\text{,}\text{B}$ are the electric and magnetic fields, a is the axion field. If $\text{B}$ is an external magnetic field, a's interact with the photon electric field component parallel to $\text{B}$.

An example of a scalar particle weakly coupled to two photons is the dilaton, which arises in super-string theories and interacts with matter through the trace of the energy-momentum tensor [7]. In particular its interaction with photons is given by the Lagrangian

$$\text{L}{}_{\mathrm{a\gamma \gamma }}\text{=−}\text{1}\text{4}\text{g}{}_{\mathrm{a\gamma \gamma }}\text{F}{}_{\mathrm{\mu \nu }}\text{F}{}^{\mathrm{\mu \nu }}\text{a=}\text{1}\text{2}\text{g}{}_{\mathrm{a\gamma \gamma }}\text{(}\text{B}{}^2\text{−}\text{E}{}^2\text{)·a}$$

Here again, g_aγγ is the coupling constant and a is the dilaton field. If $\text{B}$ is an external magnetic field a's interact with the photon electric field component orthogonal to $\text{B}$. Usually it is assumed that g_aγγ=O(M⁻¹_Pl) and that the dilaton mass m_a=O(M_Pl), where M_Pl is the Planck mass. However, in recent models with large compactification radii (see e.g. [8]), the dilaton could be rather light and since there are no firm predictions for the coupling g_aγγ the searches for such particles become very interesting and actual.

Several experiments have placed limits on possible (pseudo)scalar bosons [1]. Experimental bounds on g_aγγ for light a's can be obtained from laser experiments 9, 10, from experiments on J/ψ and ϒ particles [11], and from orthopositronium decays [12]. The best direct experimental limit, $\text{g}{}_{\mathrm{a\gamma \gamma }}\text{<2×10}{}^{\mathrm{-3}}\text{GeV}{}^{\mathrm{-1}}$ for the eV mass range, has been extracted from the recent limit of the CLEO Collaboration Br(ϒ(1S)→aγ)<1.3×10⁻⁵ [13]by assuming this decay to occur through a virtual photon [14].

The best limit on the axion-photon coupling comes from astrophysical limits on anomalous energy loss by stars [15]. However, such astrophysical constraints, although more stringent, are model-dependent and have various uncertainties. For example, as has been demonstrated in [16], the inclusion of additional fermions which strongly interact with (pseudo)scalars of mass $\text{m}{}_{\mathrm{a}}\text{∼O(10)}\text{eV}$ may evade the astrophysics constraints. Hence, it is important to perform independent laboratory tests on the existence of such particles in the mass range discussed above.

Experimental techniques used for searching for light (pseudo)scalars are based either on the measurement of vacuum birefringence 17, 18, 19, or on the Helioscope method 20, 21, or on what is called “photon-regeneration” [22]. Here we describe a direct experimental search for a particles which are weakly coupled to two photons and which might be present in the SPS neutrino beam. The experiment is performed by using elements of the CERN West Area Neutrino Facility (WANF) beam line and the NOMAD neutrino detector and is based on the photon-regeneration method, used for the first time at high energy. In the analysis we do not assume any relationship between particle mass and coupling to photons but we assume that a's are rather long-lived particles. The present analysis as well as the experimental signature of the signal events are similar to those of our previous light gauge boson search [23].