Sterile neutrinos are among the most intriguing BSM (beyond-standard-model) ideas around, with a long history of data hinting that something unusual is going on. The idea that neutrinos might oscillate into a new type of invisible (or nearly invisible) neutrinos has attracted much theoretical interest, along with a large number of experiments. Unfortunately, even after 25 years of study, we still don't know if sterile neutrinos exist or not.
The sterile neutrino story starts in the mid 1990s, when the LSND (liquid Scintillator Neutrino Detector) studied neutrinos produced by the decay-at-rest of pions from LAMPF (the Los Alamos Meson Production Facility). They observed a significant excess of electron-flavored neutrinos (henceforth electron-neutrinos) over the expectations. The excess could naturally be explained via neutrino oscillations, but the oscillation parameters required to explain the data were inconsistent with the known neutrino masses and mixing. It could, however, be explained by positing a fourth neutrino flavor (beyond those connected with the electron, muon and tau leptons). These models are sometimes called 3+1 models, denoting three conventional neutrinos plus one sterile neutrino.
The result was immediately controversial, even within the LSND collaboration. The concern was that there could be some type of unmodelled background, such as neutrons sneaking into the detector. The KARMEN experiment at Rutherford Appleton Laboratory (in England) searched for similar oscillations, but did not find anything anomalous.
LSND was followed by the MiniBoone experiment at Fermilab, which ran from 2002 to about 2008. MiniBoone was designed to confirm or refuse the LSND excess. Unfortunately, it also found an anomalous result, but one which was in some tension with the LSND result, at least within a 3+1 model.; their 2013 paper used the phrases 'have some overlap with' and 'marginally compatible.' This led to a profusion of more complex models, with a 3+2 model, with two sterile flavors gaining some popularity. Of course, one expects that a model with more free parameters to do a better job of fitting the data. Models with unstable (decaying) sterile neutrinos were also proposed.
By now, a good number of experiments were in position to search for sterile neutrinos. Unfortunately, they found a wide spectrum of results, with some favoring sterile neutrinos, and some not. IceCube was in the latter category, reporting results consistent with the standard model.
There were also dedicated experiments, most notably MicroBoone at Fermilab. MicroBoone is a liquid argon time projection chamber, a follow-on to MiniBoone. The collaboration recently released their results late this year, with four analyses (of the same data) finding that their data was consistent with the standard model.
Of course, many theorists have pointed out ways that the MicroBoone null results could be compatible with the previous LSND and MiniBoone positive results. So, we still don't really know if sterile neutrinos exist or not. However, MicroBoone is a strong experiment, designed to avoid MicroBoone's weak spots (e. g. the ability to distinguish photons and electrons). It did not, however, cover exactly the same range of parameters that MiniBoone did. It would be a bit of a coincidence that nature provided sterile neutrinos with the right characteristics to elude MicroBoone. So, although sterile neutrinos are not impossible, but they seem less likely than they did on New Years Day in 2021.
