The MiniBooNE Collaboration at Fermilab has revealed its first findings. The results announced on 11 April resolve questions that were raised in the 1990s by observations of the LSND experiment at Los Alamos, which appeared to contradict findings of other neutrino experiments. MiniBooNE now shows conclusively that the LSND results could not be due to simple neutrino oscillation.

The observations made by LSND suggested the presence of neutrino oscillation, but in a region of neutrino mass vastly different from other experiments. Reconciling the LSND observations with the other oscillation results would have required the presence of a fourth, or "sterile" type of neutrino, with properties different from the three standard neutrinos. The existence of sterile neutrinos would indicate physics beyond the Standard Model, so it became crucial to have some independent verification of the LSND results.

The MiniBooNE experiment took data for this analysis from 2002 until the end of 2005 using muon neutrinos produced by the Booster accelerator at Fermilab. The detector consists of a 250,000 gallon tank filled with ultrapure mineral oil, located about 500 m from the point at which the muon neutrinos were produced (CERN Courier July/August 2002 p5). A layer of 1280 light-sensitive photomultiplier tubes, mounted inside the tank, detects collisions between neutrinos and carbon nuclei in the oil.

For this analysis the collaboration looked for electron neutrinos created by the muon neutrinos in the region indicated by the LSND observations, using a blind-experiment technique to ensure the credibility of their analysis and the results. While collecting the data, the researchers did not permit themselves access to data in the region, or "box," where they would expect to see the same signature of oscillations as LSND. When the team opened the box and "unblinded" its data, the telltale oscillation signature was absent.

Although this work has decisively ruled out the interpretation of the LSND results as being due to oscillation between two types of neutrinos, the collaboration has more work ahead. Since January 2006, the MiniBooNE experiment has been collecting data using beams of antineutrinos instead of neutrinos and expects further results from these new data.

Future studies also include a detailed analysis of an apparent discrepancy in data observed at low energy, for which the source is currently unknown, together with investigations of more exotic neutrino-oscillation models.