The solar neutrino problem is one mystery that has remained unsolved by both astrophysicists and nuclear physicists alike for more than 30 years. Nuclear physics predicts that, assuming we understand stellar physics, we should see neutrinos streaming from the Sun at a particular rate. Unfortunately current observations using neutrino telescopes show this not to be the case, in fact the actual number observed turns out to be about a third of that predicted by the current standard solar models (SSMs).

This discrepancy has caused much head-scratching over the years and cast some doubt over the solar models. Possible solutions could have implications not only for the SSMs themselves, but for particle physics models and the question of the ultimate fate of the Universe as well. Recent data from the Sudbury Neutrino Observatory (SNO) in Ontario, Canada, has put us several steps closer to a conclusive answer, but what are the implications, and where do we go from here?

What are Neutrinos?

Neutrinos are fundamental particles that have zero charge, very low mass and travel at very high velocities. Their interactions with matter are only very weak; they were not actually detected until 1956 (in an experiment which used 615 tonnes of cleaning fluid in the Homestake mine in South Dakota), 25 years after their existence was proposed by Pauli to explain the apparent non-conservation of energy and momentum during beta decay processes (see radioactivity).

Beta decay is the process by which a neutron decays to produce a proton, an electron (also known as a beta particle) and an anti-electron neutrino (the opposite particle to an ordinary electron neutrino). Normally, in any physical process, energy is conserved - it can be transformed from one type of energy to another, such as electrical into movement, heat and sound in a hairdryer for example, but it cannot be destroyed. Before we knew about the neutrino, the beta decay process appeared to violate this fundamental principle¹.

There are three types of neutrino: the electron neutrino (ν_e), the muon neutrino (ν_μ) and the tau neutrino (ν_τ), plus their associated antiparticles, making a total of six.

According to the SSMs, the Sun produces only electron neutrinos. There are many processes by which this occurs within the extreme environments of stellar cores - one such mechanism is the beta decay of boron-8 (⁸B) deep in the Sun's core. This reaction is particularly important as the energies of the neutrinos produced are above the minimum threshold at which they become detectable with current technology (other reactions produce larger quantities of neutrinos, but as yet we are unable to detect them due to their low energies). This type (or flavour) of neutrino is produced in vast quantities in the cores of all stars and in even larger numbers in supernova explosions - a total of 19 neutrinos were detected coming from the direction of SN1987A, a supernova explosion that was observed in the Large Magellanic Cloud in February of 1987. The data collected showed that in actual fact, 10⁵⁸ neutrinos (that's 1 followed by 58 zeros!) were emitted during the first second of the explosion alone². In a star, the actual quantity of neutrinos produced depends on the conditions within the star itself.

Because they only interact weakly, neutrinos produced in a star's core are able to escape without colliding with, and losing energy to, matter in the star's outer layers. Imagine a ball flying through the air. Air resistance will slow it down a bit, but it still travels a fair distance. Now imagine the same ball, thrown with the same force, through water instead of air. The water will cause more of an obstruction and will cause the ball to lose more energy, slowing it down³.

So because they interact weakly, by the time they reach the Earth, the chances are that each neutrino has the same energy as it had when it left the star. This implies that we can learn about the conditions in the core of the Sun by looking at the solar neutrino flux that arrives at detectors here on Earth.

How Do We Detect Them?

So how do we actually go about detecting them? Several detectors have been built around the world for this purpose, the Sudbury Neutrino Observatory (SNO) in Ontario is one of the most recent; it consists of 1,000 tonnes of heavy water surrounded by nearly 10,000 photomultiplier tubes. The detector has to be so large because neutrinos only interact very weakly - of the several billion neutrons that pass through the SNO detector every second, only a few interact each day.

The situation is further complicated by the presence of cosmic rays. A shower consists mainly of hydrogen and helium nuclei but small numbers of other particles such as electrons, positrons and neutrinos are also present. Cosmic rays are absorbed when they reach the Earth's surface so in order to prevent them affecting the number of events recorded, the SNO was built in an old mine 2km underground.

In the detector three reactions can occur:

• ν_e + d —> p + p + e^- a charged current reaction (CC)
• ν_x + d —> p + n + ν_x a neutral current reaction (NC)
• ν_x + e^- —> ν_x + e^- an elastic scattering reaction (ES)

where e represents an electron, n is a neutron, p is a proton, d is a deuteron (a nucleus composed of a proton and a neutron) and x can represent any of the three types of neutrino⁴.

When a deuterium nucleus absorbs a neutrino it can decay to produce either a proton and a neutron (a neutral current reaction) or two protons and an electron (a charged current reaction). Electrons produced in charged current reactions carry away most of the energy and travel at such high velocities through the water that they emit Cerenkov radiation - the electromagnetic equivalent of a sonic boom which occurs when charged particles move through a transparent medium at a speed greater than that of light in the same medium⁵. This is the type of reaction that the SNO was designed to detect - an event is recorded in the detector when the emitted photons are picked up by the photomultiplier tubes.

While charged current reactions only occur with electron neutrinos, neutral current reactions can involve any of the three types of neutrino. Similarly, elastic scattering reactions can occur with all three flavours, but this kind of reaction is more sensitive to electron neutrinos than to either of the other two types and is not specific to deuterium; it can occur in other fluids⁶.

Missing neutrinos?

So we can detect the neutrinos that are produced in the Sun's core when they reach the Earth, then that's it surely, where's the problem? Unfortunately, the actual number detected is about a third of what the SSMs say it should be and, coincidentally, there are three types of neutrino. In the strange world of particle physics there are some types of particle that can decay into other members of the same family and it is thought that neutrinos may be one of these. If so, then this would explain the discrepancy between the theoretical values and the actual rates that have been detected by the various detectors over the years.

So what have the physicists at the SNO discovered from their data? They compared the flux of solar electron neutrinos from charged current reactions (φ^CC) in their detector with measurements of the more inclusive flux of solar neutrinos obtained from elastic scattering reactions (φ^ES) made by Super-Kamiokande (SK), another solar neutrino detector, located in Japan, which detects solar neutrinos via elastic scattering off atomic electrons⁷. If solar neutrinos do not change flavour on their journey to Earth, then the ν_e flux determined from the charged current reaction (in the SNO) should be the same as the ν_x flux from the electron scattering reaction (in SK), ie φ^CC = φ^ES. If, however φ^ES turns out to be greater than φ^CC, then the implication is that solar neutrinos do indeed change type⁸.

The Results

When the SNO team analysed their data, they found that, according to their detector, φ^CC is 1.75 million neutrinos per square centimetre per second⁹. In comparison, the SK team found φ^ES to be 2.32 million ¹⁰. This is a discrepancy of 0.57 million neutrinos per square centimetre per second and, taking the uncertainties of both results into account, the SNO team have shown that there is strong evidence that other flavours of neutrino are present in the solar flux; ie it is likely that solar neutrinos do change type on their journey through space. Using their results they calculated the total solar neutrino flux to be 5.44 million neutrinos per square centimetre per second, which is in close agreement with the predictions of current solar models¹¹.

More recently, the SNO team have measured the neutral current flux by adding two tonnes of salt to the water in the detector tank. This measurement tells us the total number of neutrinos reaching the Earth from the Sun as it is sensitive to all three types of neutrino. In this experiment they found the total flux to be 5.21 million neutrinos per square centimetre per second, in agreement with the previous estimate from the CC measurement.

The decay process that neutrinos go through has an equal chance of producing any of the three types as a product, so the fact that observations of the number of solar electron neutrinos only gives a result that is one third what is expected from the models makes sense. Over a path length the size of an astronomical unit, the distance between the Sun and the Earth, the distribution of neutrinos settles down to a more or less even distribution so that approximately one-third are of each type by the time they reach our detectors here on Earth.

A Twist in the Tail?

That's not quite where the story ends though. Originally it was thought that neutrinos were massless particles that travelled at the speed of light, but massless particles do not decay into other members of the same family, so proof that they do indeed alter their flavour confirms that they have mass and, therefore, must travel significantly slower. This has implications for particle physics; the fact that neutrinos do have mass now has to be incorporated into the Standard Model.

This result also has consequences for cosmological models. If neutrinos have mass - even if it is absolutely minuscule - they could account for a significant proportion of the dark matter, or 'missing mass', in the Universe. This would help explain why observations of galaxies show them to be rotating as though they contain far more matter than we can actually see. There are however many more candidates for dark matter and neutrinos are only likely to make up a small percentage of the missing mass. But at least we know that they are there!

The puzzle is not yet completely solved, the research is continuing, more results from the SNO are due to be published in the near future and no doubt much more data will be obtained by the SNO and Super-Kamiokande¹² teams in the years to come. The mystery appears close to being solved, but the story is not finished yet.

Further Reading

The Sudbury Neutrino Observatory, The Sudbury Neutrino Observatory Institute (SNOI) (including some nice images of the SNO detector)

Katie Pennicott, 'Solar neutrino puzzle is solved', Physics World, Vol. 14, No. 7, (July 2001)

John Gribbin, The case of the missing neutrinos. Penguin, 2000

Adrian Cho, 'Lost and found', New Scientist, 23 June 2001