Historical research guide · 0765 · 2014
First results from the new PVLAS apparatus: A new limit on vacuum magnetic birefringence
F. Della Valle · E. Milotti · A. Ejlli · G. Messineo · L. Piemontese · G. Zavattini · U. Gastaldi · R. Pengo · G. Ruoso
In one minute
Quantum electrodynamics says a strong magnet turns empty space into something like a crystal: light polarised along the field should travel very slightly slower than light polarised across it. The predicted difference is about four parts in a thousand million million million million per tesla squared, and nobody has yet measured it directly. Federico Della Valle, Guido Zavattini, Giuseppe Ruoso and their INFN colleagues report the first results from the rebuilt PVLAS apparatus in Ferrara: a laser at 1064 nanometres locked into a 3.3-metre cavity of finesse 670,000, so the light crosses the field 430,000 times, passing through the bores of two rotating permanent magnets. Calibrating against the known magnetic birefringence of helium, and running 210 hours, they find a value compatible with zero and set a new limit — which cuts the gap to the predicted quantum effect to a factor of fifty. The same data tighten laboratory bounds on axion-like particles and on hypothetical fractionally charged particles, neutrinos included.
Connection
Chapter 2 says the vacuum is a real, structured medium, and vacuum magnetic birefringence is its cleanest optical signature: bend light with a magnet and nothing else, and the medium stops being a figure of speech. This paper is the moment the gap to that measurement came down to a factor of fifty, and it is chapter 1’s evidence ladder in miniature — a calibration against a gas whose birefringence is already known, a cross-check with the two magnets deliberately spun at different rates, and a named limiting noise source. The collaboration’s later account of the same apparatus is on this site at /library/stm-e58fdfd01c, and the twenty-five-year review of the whole programme, with the final limits and the LHC-dipole proposal that would close the gap, is at /library/stm-08601764f6.
From the source
Vacuum magnetic birefringence is predicted by quantum electrodynamics and has never yet been observed directly. In the Euler-Heisenberg framework the unitary field magnetic birefringence of vacuum is 3.97 times ten to the minus twenty-four per tesla squared, and observing it would be the first direct detection of light-by-light interaction at low photon energy — the only evidence for zero point quantum fluctuations to date coming from the Casimir effect, which applies to photons alone.Sect. I, Introduction, Eqs. (1) to (3)
The new Ferrara apparatus is a heterodyne polarimeter on a 4.5 tonne granite bench: a 1064 nanometre Nd:YAG laser locked to a Fabry-Perot cavity 3.303 metres long with finesse 670,000, the beam crossing the bores of two rotating permanent dipole magnets in Halbach configuration, each contributing a field-squared-times-length integral of 5.12 tesla squared metres. The cavity multiplies the single-pass ellipticity of 1.2 times ten to the minus sixteen by 430,000, giving a signal of about 5 times ten to the minus eleven to be measured at twice the magnet rotation frequency.Sect. II, Experimental method and setup; Tables II and IV
The apparatus was first calibrated on the Cotton-Mouton effect of helium at 32 microbar, giving a unitary birefringence of 2.2 plus or minus 0.1 times ten to the minus sixteen per tesla squared per atmosphere, in agreement with previously published values and with no spurious peaks at other harmonics. From 210 hours of vacuum data with the magnets rotating between 2.4 and 3 hertz — 40 of those hours with the two magnets deliberately spun at slightly different frequencies, so that neither could be generating a spurious signal — the measured unitary vacuum magnetic birefringence is 4 plus or minus 20 times ten to the minus twenty-three per tesla squared. It is compatible with zero and stands as a new limit, a factor of 50 from the predicted quantum electrodynamic value.Sect. III, Calibration; Sect. IV, Results; Sect. V.A, Eq. (8)
The same limit bounds the elastic photon-photon scattering cross section for unpolarised light at 1064 nanometres to less than 4.6 times ten to the minus sixty-six square metres, against a quantum electrodynamic prediction of 1.84 times ten to the minus sixty-nine square metres.Sect. V.A, Eqs. (9) and (10)
The measurement also yields improved model-independent bounds on the coupling of axion-like particles to two photons for masses greater than one millielectronvolt, and a factor of two improvement in the fractional charge limit on millicharged fermions and scalars — including neutrinos, whose charge is limited to about ten to the minus seven of the electron charge for masses below 20 millielectronvolts.Sect. V.B and V.C; Figs. 6 and 7
The sensitivity is still far from the shot noise limit, and the excess noise is clearly due to the presence of the Fabry-Perot cavity — without the cavity, shot noise is achieved. The authors suspect variations in the intrinsic birefringence of the reflective coating caused by thermal effects, and say that effort will now go into improving the sensitivity.Sect. V.A, final paragraph
The way in
https://doi.org/10.1103/PhysRevD.90.092003How to cite it
F. Della Valle, E. Milotti, A. Ejlli, G. Messineo, L. Piemontese, G. Zavattini, U. Gastaldi, R. Pengo, G. Ruoso (2014) First results from the new PVLAS apparatus: A new limit on vacuum magnetic birefringence doi:10.1103/PhysRevD.90.092003