In an article published on arXiv today, seven physicists have reported the observation of an unexpected new light boson, named the E(38), at the Nuclotron superconducting particle accelerator at the Joint Institute for Nuclear Research in Dubna, near Moscow. A boson is a particle whose wave function returns to its original value when rotated through 360°, and the presently known elementary particles which are bosons are the photon, the gluons which bind quarks and other gluons together into protons, neutrons, and atomic nuclei, the W and Z vector bosons that transmit the weak nuclear force, the recently discovered Higgs boson candidate, and the graviton. The wave functions of the other presently known elementary particles return to their original value times -1 when rotated through 360°, and only return to their original value when rotated through 720°.

In contrast to the Higgs boson candidate recently discovered at CERN, whose mass of about 125 GeV is about 133 times the proton mass, and roughly equal to the mass of a Caesium nucleus containing 55 protons and 78 neutrons, the mass of the new E(38) boson is only 38 MeV or 0.038 GeV, less than 1/3 of the mass of the pion, which is formed from a quark and an antiquark bound together by gluons, and is the lightest strongly interacting particle. The E(38) boson is not predicted by the Standard Model of the strong and electroweak interactions, and if the observation is confirmed, it will be the first discovery of an elementary particle not predicted by the Standard Model, since the Standard Model became established in the 1970's. The E(38) boson can not, however, be a constituent of the unknown dark matter that comprises about 83% of the mass of the matter in the universe, because it is not long-lived.

The existence of the E(38) boson was first argued for by Eef van Beveren and George Rupp, who claimed, in an article published in February this year, to find evidence for the particle in previously published data of the CMD-2 Detector Collaboration at the VEPP-2M Collider at the Budker Institute of Nuclear Physics in Novosibirsk, the CDF Collaboration at the Tevatron at Fermilab, the BABAR Collaboration at the Stanford Linear Accelerator Center (SLAC), the CB-ELSA Collaboration at the Crystal Barrel experiment in Bonn, and the COMPASS Collaboration at the Super Proton Synchrotron (SPS) accelerator facility at CERN. However the COMPASS Collaboration disputed van Beveren and Rupp's interpretation of their data, to which van Beveren and Rupp in turn replied.

The new results from the Dubna Nuclotron find evidence for the E(38) boson in three separate processes: scattering of a beam of deuterons with 2.0 GeV per nucleon energy on a carbon target, scattering of a beam of deuterons with 3.0 GeV per nucleon energy on a copper target, and scattering of a beam of protons with 4.6 GeV per proton energy on a carbon target. In each case the evidence for the E(38) boson was a small excess above background, peaked at about 38 MeV, in the distribution of the invariant mass of photon pairs produced in the reaction. The excesses are typically only about 10% to 15% above the background at their greatest and are thus sensitive to errors in the modelling of the background, as is also the case for most searches for new physics at the Large Hadron Collider at CERN.

As George Rupp has pointed out in a comment below, the statistical significance of the Dubna observation is about 5 to 6 sigma for each process. For example for the first plot in the article, for a deuteron beam on a copper target, in Fig. 2(a) on page 2 of the article, the background is about 7000 gamma pairs per 2 MeV bin in the plot of events versus invariant mass of the gamma pairs, corresponding to a standard deviation of about 84 gamma pairs per 2 MeV bin, while the number of events above background is about 500 in the 2 MeV bin at the signal peak, corresponding to a 6 sigma significance for this data point.

Barry Adams and Tommaso Dorigo have also written about the E(38) on Science 2.0, and Luboš Motl has also written about it.