Particle physicists know of the existence of three types (or "flavours") of neutrino: the electron neutrino, the muon neutrino, and the tau neutrino, each with their corresponding antineutrino. However, there is no reason why these should be the only neutrino types in the universe. One possibility for an unobserved neutrino flavour is the "sterile neutrino".
Neutrinos, being neutral leptons, can only interact via the weak nuclear force; they carry neither charge nor colour. Sterile neutrinos take this one further; they do not interact via the weak force either. This leaves them no avenue to interact through the standard particle physics forces, hence the name 'sterile'. A sterile neutrino can presumably interact via gravity, and it should participate in neutrino oscillation, where neutrinos spontaneously transform into different flavours as they propagate through space. The latter is more important, as it presents the only means of actually detecting a sterile neutrino. Many of the methods that have been used to determine the number of neutrino flavours are only sensitive to neutrinos that interact via the weak force, and thus are useless for detecting a sterile neutrino.
The rate of oscillation between two neutrino flavours is inversely related to the difference in mass between the two flavours. Since we have not yet seen neutrino oscillation into sterile neutrinos, we must conclude that any sterile neutrino must be very heavy, at least by neutrino standards. Present experimental limits require a heavy neutrino to be above 40 GeV in mass, or at least 40 times the mass of the proton. This contrasts with the wispy electron neutrino, which is known to be less than 3 eV, ten billion times smaller. The other two known neutrino flavours are believed to have comparable mass but have only been established to be less than the mass of the electron at 511,000 eV, still five orders of magnitude difference. Oscillation between such highly-disparate masses is only possible due to a high total energy.
In addition to the 'ordinary' sterile neutrino, which is identical to the known neutrinos outside of its mass and sterility, there is another class of neutrino that is effectively sterile, the 'wrong-handed' neutrino. One of the properties of the weak nuclear force is that it only couples to 'left-handed' particles and 'right-handed' antiparticles. (For more details, see the helicity and parity violation nodes.) Thus all interactions involving neutrinos only involve left-handed neutrinos and right-handed antineutrinos. This does not rule out the existence of right-handed neutrinos and left-handed antineutrinos, which may collectively be called 'wrong-handed', and, because of the weak force's selectivity in handedness, will be sterile.
The possibility of wrong-handed neutrinos is only tenable if the neutrino belongs to the class of particles known as Dirac particles. All known elementary fermions are thought to be Dirac particles, but since the neutrino has no electric charge, it is possible that it is instead a Majorana particle. Unlike Dirac particles, which have distinct antiparticles, a Majorana particle is identical to its antiparticle. If the neutrino is in fact a Majorana particle, then we call left-handed neutrinos 'particles' and right-handed neutrinos 'antiparticles' only because the weak force treats them differently, not because they are intrinsically different. This model does have some unusual features, particularly violation of lepton number conservation.
The ongoing search for sterile neutrinos is just one of the possible modifications of the Standard Model of particle physics presently being investigated by experiments. Although sterile neutrinos generally do not interact with other matter, they may be detectible through neutrino oscillations or by having very weak interactions with other particles, in particular, the property of the weak force that selects one particular handedness may not be absolute. At this point, the existence of a fourth, sterile neutrino flavour is considered unlikely, but the wrong-handed neutrinos are the subject of considerable ongoing speculation.
Many thanks to rootbeer277 for his comprehensive critique of this writeup's original form.
This writeup is copyright 2005 D.G. Roberge and is released under the Creative Commons Attribution-NoDerivs-NonCommercial licence. Details can be found at http://creativecommons.org/licenses/by-nd-nc/2.0/ .