The first indications arose about 30 years ago. In 1968, Ray Davis and his colleagues published the initial results of an experiment that measured neutrinos from the Sun using a large container of cleaning fluid as a neutrino detector in the Homestake Mine in South Dakota. The number of electron-neutrinos detected was less than the figure predicted from theoretical solar models. The experiment was continued for two decades and the theory of the Sun was improved, but the deficit in the observed number of neutrinos persisted thereby creating the solar neutrino problem.
This discrepancy could be explained if some of the electron-neutrinos that are created in the energy-producing core of the Sun mutate into one of the two other known types of neutrinos: the muon-neutrino and the tau-neutrino. Such a process is called neutrino oscillation because eventually the transformed neutrino will change back into its original form only to undergo the transformation again and again.
It is a fundamental result that neutrino oscillations cannot occur unless neutrinos have mass. So the radiochemical experiment of Davis indicated that neutrinos are massive. However, the situation was not conclusive. Many astrophysicists did not believe that the solar models were sufficiently accurate and that they were predicting too large a solar output. Others felt that Ray Davis and his co-workers were not measuring all the neutrinos that they should have been. Such an error in experiment would not be surprising because neutrinos are extremely weakly interacting particles rending them the most difficult particles in nature to detect. (See What Are Neutrinos? below.) So for several decades, scientists "lived" with the solar neutrino problem, not knowing whether there was a problem (1) with the theory, (2) with the experiment, or (3) with the Standard Model of Particle Physics for which neutrinos are massless.
The elusive nature of the neutrino and the inherent difficulty in its detection has created many false signals. In an experiment involving nuclear beta decay, a team of Russian physicists announced that electron-neutrinos had a mass in the several electron-Volt range (An electron-Volt or eV is actually a unit of energy; particle physics measure masses in terms of energy using E=mc2; An electron-Volt is 1.78 x 10-36 kilograms). However, subsequent beta decay experiments did not confirm the result.
In 1985, a new neutrino mystery arose. Cosmic rays are elementary particles such as protons and electrons or electromagnetic radiation such as X-rays and gamma rays that travel throughout space. They are produced by many kinds of astrophysical objects such as stars including our Sun, supernovae, black holes, neutron stars, active galactic nuclei and quasars. When such cosmic rays strike the Earth's atmosphere, pions are often produced. The charged pion decays into a muon and a muon-neutrino. The muon then decays into an electron, a muon-neutrino and an electron-neutrino. Thus, there should be about twice as many muon-neutrinos as electron-neutrinos streaming down on Earth from cosmic rays.
Earlier measurements of atmospheric neutrinos had error bars sufficiently large as to be compatible with this 2-1 ratio. In 1985, two experiments in deep mines one in the United States called IMB and one in Japan called Kamiokande observed that the ratio was less than two-to-one. Either less muon-neutrinos or more electron-neutrinos were being detected. This became known as the atmospheric neutrino anomaly. It again could be explained by neutrino oscillations. However, the oscillations could not be of the same type as those that would explain the solar neutrino problem. Furthermore, two other solutions to the atmospheric neutrino anomaly were possible: The theory of cosmic ray showers could be wrong, or the experimental difficulty in detecting neutrinos might be producing an incorrect result.
Then, an experiment in England in the late 1980's announced that a neutrino had a mass of 17,000 eV. This became known as the 17 KeV neutrino (kilo-electron-Volts). Other physicists were unable to reproduce the result and eventually the original experiment was found to be faulty.
In the early 1990's, two new solar neutrino experiments SAGE (in Russia) and GALLEX (in Italy) that used gallium as a target confirmed the experimental deficit in solar neutrinos. It seemed clear to most physicists at this point that an error in experiment could be ruled out. Few were ready to announce that neutrinos had masses because the possibility still existed that the theoretical solar models were in error in regard to solar neutrino production.
In 1996, an experiment called LSND at Los Alamos surprised the world by announcing that it had seen the oscillation of muon anti-neutrinos to electron anti-neutrinos: 22 events had been observed whereas 4 had been expected. However, the result was near the border of a region of neutrino parameters already ruled out by experiments in Europe. In addition, subsequent other accelerator experiments did not support the LSND findings. A majority of scientists now believes that the Los Alamos experiment was a fluke.
Until the late 1990's, the "official" values for the masses of the three neutrinos were zero. However, the combined analysis of a variety of experiments actually favored the bizarre result the masses squared were negative!? Few scientists believed that the neutrinos could really be tachyonic (that is, have negative m2) because this is not physical: Tachyonic particles should travel faster than the speed of light in violation of Einstein's special theory of relativity.
Quite a few theoretical extensions of the Standard Model of Particle Physics predicted that neutrinos had masses. With these theoretical suggestions and the hints from experiments, many theorists believed the neutrinos oscillated. They felt that too many unusual things were happening for neutrinos to be massless particles.
In 1996, Super K was completed and began taking data. Super K is not a store: It is a scaled up version of Kamiokande and it began detecting atmospheric neutrinos at rates much higher than its predecessors. In 1998, after analyzing more than 500 days of data, the experimentalists at Super-Kamiokande announced that the atmospheric neutrino anomaly was not a statistical aberration: The deficit in muon-neutrinos is due to the upward traveling ones. Oscillations easily explain this: Those muon-neutrinos raining down on the mine do not have sufficient time to oscillate while those traveling through the Earth do. In early June of 1998, the press reported that neutrinos had mass. At this point, a majority of physicists became convinced that neutrino oscillations were a reality. However, given all the false signals of the past, the difficult nature of neutrino experiments, and the remote possibility that the theory of cosmic ray neutrino production was incorrect, there remained a small chance that neutrinos did not have mass.
This year that lingering uncertainty was eliminated. In a paper submitted on June 18, 2001, a team of almost 200 scientists released the first results from the Sudbury Neutrino Observatory (SNO), an experiment which uses a kiloton of heavy water (that is, water containing significant amount of deuterium) that is located in Sudbury, Ontario. With greater than 99-percent confidence, solar neutrinos are undergoing changes on their way from the Sun to the Earth. When combined with data from Super Kamiokande, SNO was able to demonstrate that the electron-neutrinos are changing into muon-neutrinos or tau-neutrinos. If neutrino oscillations are occurring, then the solar neutrino flux is consistent with the theoretical models of the Sun. Of the three possible solutions to the solar neutrino problem mentioned above, only one was viable: Neutrinos had to have masses and had to undergo oscillations.
It has taken more than 30 years to establish this result. It is easy to modify the Standard Model of Particle Physics to incorporated neutrino masses, but additional experimental information is needed to determine precisely how this is done.
The most likely scenario is that some electron-neutrinos from the Sun oscillate into muon-neutrinos and perhaps tau-neutrinos, and some atmospheric muon-neutrinos oscillate into tau-neutrinos. The mass of the heaviest neutrino is roughly 0.05 eV, the mass of the next is somewhere between 0.01 eV and 0.0001 eV, and the third has a mass significantly less than these two values. For comparative purposes, the lightest non-neutrino particle is the electron with a mass of 511,000 eV.
Because neutrino masses are less than a few electron-Volts, neutrinos can only account for a small fraction of the dark matter that permeates the Universe. The mystery of dark matter thus remains unresolved.
In short, a major discovery has taken place in physics. However, it has taken 30 years to achieve the result.
John Updike has written a wonderful poem about neutrinos. It begins
Neutrinos: they are very small
They have no charge; they have no mass;
they do not interact at all.
The Earth is just a silly ball
to them, through which they simply pass
like dustmaids down a drafty hall
or photons through a sheet of glass.
The second line of that poem now needs to be modified to
"They have no charge; they almost have no mass;"
The following is a pedestrian introduction to neutrinos:
Neutrinos are fundamental, subatomic particles that are similar to electrons but are neutral meaning that they possess no electric charge. The lack of charge renders the behavior of neutrinos very different from the behavior of electrons. Most of the time, an electron is attracted to or repelled by another object because of electromagnetism. Being neutral, a neutrino experiences no electromagnetic force. For this reason, neutrinos are extremely weakly interacting, meaning that hardly deflect off other matter. When they do deflect, it is due to the subnuclear weak force, a feeble interaction that is, among things, responsible for certain decay of nuclei. To give you an idea of how weakly interacting neutrinos are, consider the following. A neutrino released in a nuclear decay will travel on average a distance of roughly one light-year through lead before changing its direction of motion! Despite the fact that about 100 cosmic-ray-generated neutrinos pass through your body every second, there is only about one chance in ten that a neutrino will bounce off a nucleus in your body! John Updike was being accurate when he said in his poem, "The Earth is just a silly ball to them, through which they simply pass." Roughly, all but one among a million neutrinos will travel through the Earth undeflected.
There are three types, or "flavors," of neutrinos, one of which is associated with the electron, one of which is associated with the muon, and one of which is associated with the tau lepton. The muon and tau are heavier versions of the electron (The mass of the muon (respectively, tau) is about 200 (respectively, 3500) times the mass of an electron). When electrons are involved in a high-energy scattering process, only electron-neutrinos are initially produced. Ditto for muon- and tau-neutrinos.
When neutrinos are initially produced, they spin in a very specific way: If you take your left hand and point your thumb in the direction that the neutrino is moving and let your fingers curl naturally, the fingers indicate the spinning direction. For this reason, neutrinos are called left-handed particles. The Standard Model of Particle Physics has only left-handed neutrinos. The simplest way to incorporate massive neutrinos in the Standard Model is to add in right-handed neutrinos. (Neutrinos like electrons can possible spin in only two directions; Such particles are called spin-one-half fermions). Right-handed neutrinos do not experience any forces except through the effects of neutrino masses. If neutrinos have mass, then it is expected that they should also oscillate: Electron-neutrinos can transform into muon-neutrinos or tau-neutrinos and then go back into electron-neutrinos. Similarly, muon-neutrinos and tau-neutrinos can undergo oscillatory flavor changes. The evidence is now convincing that the solar neutrino problem and the atmospheric neutrino anomaly are solved by neutrino oscillations.