How MiniBooNE works

No original scientific discovery can ever be considered as definitive. Other scientists using different equipment must corroborate and expand the finding, bringing their own skills and intelligence to the problem, probing and checking the discovery using their own methods. Such motivations are behind the creation of the MiniBooNE experiment.
Almost as soon as a second neutrino flavor had been imagined, scientists began to wonder if neutrinos could “oscillate,” or spontaneously change from one flavor to another. Other indicators also suggested the possibility of neutrino oscillations: only about 50% of the electron neutrinos expected to be coming from the sun are seen in experiments, and muon neutrinos interacting in our atmosphere do not follow their expected distribution while electron neutrinos do.

A 1995 experiment at the Los Alamos National Laboratory in New Mexico stunned the physics community when it reported a few instances where a neutrino of one flavor had indeed transformed into a neutrino of another flavor. The Liquid Scintillator Neutrino Detector (LSND) recorded neutrino events from 1993 to 1998, and saw about seven dozen events where neutrino oscillation had apparently occurred.

Such a finding would also imply that neutrinos have a tiny mass, and could indicate the existence of an additional, fourth flavor of neutrino, a flavor unaccounted for by the hallowed "Standard Model" of particle physics. Though statistically significant, the LSND finding was based on only a few events and it begged for confirmation. So in 1997 the idea for MiniBooNE was born when Columbia University physicists Janet Conrad and Michael Shaevitz decided to chase after the controversial LSND result.

Hunting for oscillations
MiniBooNE, the first phase of the "Booster Neutrino Experiment" located at the Fermi National Accelerator Laboratory in Illinois, is a large detector set up to hunt for neutrino oscillations--specifically, the change of a muon neutrino into an electron neutrino. Ironically, picking out oscillations produced by such infinitesimal particles requires a huge apparatus weighing hundreds of tons.

Detecting oscillations requires a beam of neutrinos produced through a series of steps:
1. First, bunches of very energetic protons (moving at about 99.3 percent of the speed of light) will be produced in Fermilab’s Booster ring, a small particle accelerator. Each bunch will consist of about 5 trillion protons and will last only 1.5 millionths of a second, with five such bunches produced each second.

2. These protons will strike a target made of beryllium.

3. When the protons strike the beryllium, particles known as pions and kaons will be produced, and gathered by a magnetic focusing device into a tight beam.

4. This second beam will travel inside a pipe about 50 meters (160 feet) long. Inside the pipe about half the pions and kaons will decay into muons and muon neutrinos.

5. The muons will be filtered out by a barrier made of about 5 meters of steel and 500 meters (1,600 feet) of earth. The neutrinos will zip unimpeded through the barrier to the MiniBooNE detector. If LSND is correct, a tiny fraction of the muon neutrinos will oscillate en route into electron neutrinos.
Both muon and electron neutrinos will enter the MiniBooNE detector. Most will fly through without stopping, but a small fraction will interact and be observed.

The detector
The MiniBooNE detector is a spherical tank sitting in a large underground vault at Fermilab. The tank, about 12 meters (40 feet) in diameter, is filled with 250,000 gallons of mineral oil. (The oil was delivered to the site in twelve railroad tanker cars.) A small fraction of the muon and electron neutrinos will collide with atoms in the mineral oil, at a rate of about one per 20 seconds. When they do, they will produce either a highly energetic muon or a highly energetic electron, respectively.

These muons and electrons will rush through the mineral oil, moving in fact faster than the speed of light in the oil. Any particle moving faster than the speed of light in a material causes a particular kind of light to be emitted, called Cerenkov light, a phenomenon similar to the sonic boom that supersonic aircraft can produce.
Just inside MiniBooNE’s outer tank is an inner tank, dotted with 1,280 light detectors. Called “photomultiplier tubes,” they can detect the cones of Cerenkov light after they are produced.
Sorting the light cones
The Cerenkov light cones produced by muons and electrons are significantly different--which allows MiniBooNE to tell the particles apart. Muons, being almost 200 times heavier than electrons, produce sharply defined cones of light. The much lighter electrons, however, do not travel nearly as far in the mineral oil; they scatter about and eventually come to a stop. This means that their Cerenkov light cones have fuzzy inner and outer edges. Since the light cones from electrons and muons are different, the particles producing the light and therefore the neutrinos engendering the particles can be distinguished.

The MiniBooNE experiment will receive about a billion trillion protons each year, and should observe about a thousand neutrino transformations in that time. It will collect much more data than LSND because of its more intense neutrino beam; this will produce more events and a result of greater statistical significance. 

As seen in an earlier section, confirming the LSND result would indicate the existence of an additional kind of neutrino beyond the three known flavors, requiring physicists to rewrite a large part of the theoretical framework of the Standard Model.

-- David Appell