On the Particle Interpretation of the PVLAS Data: Neutral versus Charged Particles

TL;DR

The paper investigates the PVLAS anomaly in vacuum polarization by comparing two low-energy explanations: axion-like particles (ALPs) and millicharged particles (MCPs). It derives and contrasts the rotation and ellipticity signals, including new results for millicharged scalars, and demonstrates how signal signs and parametric dependences on $B$, the laser frequency, and the apparatus length can distinguish scenarios. A global likelihood analysis of BFRT, PVLAS, and Q&A data yields constrained regions in ALP and MCP parameter spaces, highlighting target regions for upcoming experiments and showing that near-future data may rule out some possibilities. The study emphasizes that, while current data do not decisively select a single model, the distinct signatures—especially birefringence signs and regeneration channels—provide a clear path to discriminating ALP versus MCP origins of the PVLAS signal. Overall, the work provides a concrete framework for using optical polarization measurements to probe new light particles and guides future experimental tests.

Abstract

Recently the PVLAS collaboration reported the observation of a rotation of linearly polarized laser light induced by a transverse magnetic field - a signal being unexpected within standard QED. Two mechanisms have been proposed to explain this result: production of a single (pseudo-)scalar particle coupled to two photons or pair production of light millicharged particles. In this work, we study how the different scenarios can be distinguished. We summarize the expected signals for vacuum magnetic dichroism (rotation) and birefringence (ellipticity) for the different types of particles - including new results for the case of millicharged scalars. The sign of the rotation and ellipticity signals as well as their dependencies on experimental parameters, such as the strength of the magnetic field and the wavelength of the laser, can be used to obtain information about the quantum numbers of the particle candidates and to discriminate between the different scenarios. We perform a statistical analysis of all available data resulting in strongly restricted regions in the parameter space of all scenarios. These regions suggest clear target regions for upcoming experimental tests. As an illustration, we use preliminary PVLAS data to demonstrate that near future data may already rule out some of these scenarios.

Figures (4)

• Figure 1: Schematic view of a "light shining through a wall" experiment. (Pseudo-)scalar production through photon conversion in a magnetic field (left), subsequent travel through a wall, and final detection through photon regeneration (right).
• Figure 2: Dependence of the rotation and ellipticity signals on the strength of the magnetic field $B$, the wavelength $\lambda$ of the laser, and the length $L$ of the magnetic region inside the cavity. For ALPs (dark green) and MCPs (light red). The crossing of the blue dotted lines corresponds to the PVLAS published rotation and preliminary ellipticity signal for $B=5$ T, $\lambda=1064$ nm, and $L=1$ m.
• Figure 3: ALP: The $5\sigma$ confidence level of the model parameters (red). The blue shaded regions arise from the BFRT upper limits for regeneration (darkblue), rotation (blue) and ellipticity (lightblue). The gray shaded region is the Q&A upper limit for rotation. The bands show the PVLAS $5\sigma$ C.L.s for rotation (coarse-hatched) and ellipticity (fine-hatched) with $\lambda=532$ nm (left-hatched) and $\lambda=1064$ nm (right-hatched), respectively. The darkgreen band shows the published result for rotation with $\lambda=1064$ nm. The lightgreen bands result from an inclusion of preliminary data from PVLAS. The upper panels show the fit to the published data; the center panels include also the preliminary data from PVLAS, and the lower panels depict the fit using only PVLAS data. The preliminary data is only used to demonstrate the potential to distinguish between the different scenarios.
• Figure 4: MCP: The $5\sigma$ confidence level of the model parameters (red). The blue shaded regions arise from the BFRT upper limits for rotation (blue) and ellipticity (lightblue). The gray shaded region is the Q&A upper limit for rotation. The bands show the PVLAS $5\sigma$ C.L.s for rotation (coarse-hatched) and ellipticity (fine-hatched) with $\lambda=532$ nm (left-hatched) and $\lambda=1064$ nm (right-hatched), respectively. The darkgreen band shows the published result for rotation with $\lambda=1064$ nm. The lightgreen bands result from an inclusion of preliminary data from PVLAS. The upper panels show the fit to the published data; the center panels include also the preliminary data from PVLAS, and the lower panels depict the fit using only PVLAS data. The preliminary data is only used to demonstrate the potential to distinguish between the different scenarios. The preliminary PVLAS value for the sign of the ellipticity singles out the large-$\chi$ (small-mass) branch of the fermionic MCP $\frac{1}{2}$ and the small-$\chi$ (large-mass) branch of the scalar MCP 0, cf. Table \ref{['tab1']}, as is visible in the center and lower panels.