Detecting Neutrinos

extremeexperiments Detecting MASSIVENeutrinos BY EDWARD KEARNS, TAKAAKI KAJITA AND YOJI TOTSUKA

68 Updated from the August 1999 issue COPYRIGHT 2003 SCIENTIFIC AMERICAN, INC. SUPER-KAMIOKANDE DETECTOR resides in an active zinc mine inside Mount Ikenoyama. Its stainless-steel tank contains 50,000 tons of ultrapure water so transparent that light can pass through 70 meters of it before losing half its intensity (for a swimming pool that figure is a few meters). The water is monitored by 11,000 photomultiplier tubes that cover the walls, floor and ceiling. Each tube is a handblown, evacuated glass bulb half a meter in diameter. The tubes register conical flashes of Cherenkov light, each of which signals a rare collision of a high- energy neutrino and an atomic nucleus in the water. Technicians in inflatable rafts clean the bulbs while the tank is filled (right). INSTITUTE FOR COSMIC RAY RESEARCH (ICRR), UNIVERSITY OF TOKYO

ne man’s trash is another man’s treasure. For a physicist, the trash is “background”—some A giant detector Ounwanted reaction, probably from a mundane and well-understood process. The treasure is “signal”— in the heart of a reaction that we hope will reveal new knowledge about the way the universe works. Case in point: over the past two decades, several groups have been hunt- Mount Ikenoyama ing for the radioactive decay of the proton, an exceed- ingly rare signal (if it occurs at all) buried in a back- in Japan has ground of reactions caused by elusive particles called neutrinos. The proton, one of the main constituents of demonstrated that the atom, seems to be immortal. Its decay would be a strong indication of processes described by the Grand neutrinos Uniﬁed Theories that many believe lie beyond the extremely successful Standard Model of particle physics. metamorphose Huge proton-decay detectors were placed deep under- ground, in mines or tunnels around the world, to escape the constant rain of particles called cosmic rays. in flight, strongly But no matter how deep they went, these devices were still exposed to penetrating neutrinos produced by the suggesting that these cosmic rays. The ﬁrst generation of proton-decay detectors, op- ghostly particles erating from 1980 to 1995, saw no signal, no signs of proton decay—but along the way have mass the researchers found that the supposedly mundane

COPYRIGHT 2003 SCIENTIFIC AMERICAN, INC. posited, physicists have assumed that they LIGHT are massless. But if they can change from KOV REN one flavor to another, quantum theory in- CHE dicates that they most likely have mass. Muon-neutrino And in that case, these ethereal particles could collectively outweigh all the stars in Muon the universe. A Bigger Neutrino Trap AS IS SO OFTEN the case in particle physics, the way to make progress is to build a bigger machine. Super-Kamio- kande, or Super-K for short, took the basic design of Kamiokande and scaled it up by about a factor of 10 [see illustration on preceding pages]. An array of light-sensitive detectors looks in toward the center of 50,000 tons of water whose protons may decay or get struck by a neutrino. In either Electron-neutrino case, the reaction creates particles that are spotted by means of a flash of blue light Electron known as Cherenkov light, discovered by shower Pavel A. Cherenkov in 1934. Much as an aircraft flying faster than the speed of sound produces a shock wave of sound, an electrically charged particle (such as an electron or a muon) emits Cherenkov CONES OF CHERENKOV LIGHT are emitted when high-energy neutrinos hit a nucleus and produce light when it exceeds the speed of light in a charged particle. A muon-neutrino (top) creates a muon, which travels perhaps one meter and the medium in which it is moving. This projects a sharp ring of light onto the detectors. An electron, produced by an electron-neutrino motion does not violate Einstein’s theory (bottom), generates a small shower of electrons and positrons, each with its own Cherenkov cone, of relativity, for which the crucial veloci- resulting in a fuzzy ring of light. Green dots indicate light detected in the same narrow time interval. ty is c, the speed of light in a vacuum. In neutrino background was not so easy to Neutrinos are amazing, ghostly parti- water, light propagates 25 percent slow- understand. cles. Every second, 60 billion of them, er than c, but other highly energetic par- One such experiment, Kamiokande, mostly from the sun, pass through each ticles can still travel almost as fast as c it- was located in Kamioka, Japan, a mining square centimeter of your body (and of self. Cherenkov light is emitted in a cone town about 250 kilometers (155 miles) everything else). But because they seldom along the flight path of such particles. from Tokyo (as the neutrino flies). Scien- interact with other particles, generally all In Super-K, the charged particle gen- ) tists there and at the IMB experiment, lo- 60 billion go through you without so erally travels just a few meters and the

cated in a salt mine near Cleveland, Ohio, much as nudging a single atom. In fact, Cherenkov cone projects a ring of light this page used sensitive detectors to peer into ultra- you could send a beam of such neutrinos onto the wall of photon detectors [see il- pure water, waiting for the telltale ﬂash of through a light-year-thick block of lead, a proton decaying. and most of them would emerge unscathed EDWARD KEARNS, TAKAAKI KAJITA and Such an event would have been hid- at the end. A detector as large as Kami- YOJI TOTSUKA are members of the Super- den, like a needle in a small haystack, okande catches only a tiny fraction of the Kamiokande Collaboration. Kearns, pro- among about 1,000 similar ﬂashes caused neutrinos that pass through it every year. fessor of physics at Boston University,

by neutrinos interacting with the water’s Neutrinos come in three flavors, cor- and Kajita, professor of physics at the ); ICRR, UNIVERSITY OF TOKYO ( atomic nuclei. Although no proton decay responding to their three charged partners University of Tokyo, lead the analysis THE AUTHORS was seen, the analysis of those 1,000 re- in the Standard Model: the electron and team that studies proton decay and actions uncovered a real treasure—tanta- its heavier relatives, the muon and the tau atmospheric neutrinos in the Super- lizing evidence that the neutrinos were un- particle. An electron-neutrino interacting Kamiokande data. Totsuka recently be- preceding pages expectedly fickle, changing from one spe- with an atomic nucleus can produce an came the director of KEK, Japan’s na- cies to another in midflight. If true, that electron; a muon-neutrino makes a muon; tional particle physics laboratory, after phenomenon was just as exciting and the- a tau-neutrino, a tau. For most of the sev- serving as spokesperson for Super-K

ory-bending as proton decay. en decades since neutrinos were ﬁrst since its inception. DAVID FIERSTEIN (

lustration on opposite page]. The size, shape and intensity of this ring reveal the properties of the charged particle, which θ in turn tell us about the neutrino that pro- INCOMING duced it. We can distinguish the Cheren- COSMIC RAYS θ kov patterns of electrons from those of O muons: the electrons generate a shower of IN SUPER-K R particles, leading to a fuzzy ring quite un- EUT N like the crisper circle from a muon. From G the Cherenkov light we also measure the TIN ATMOSPHERE energy and direction of the electron or mu- OSCILLA on, which are decent approximations of the energy and direction of the neutrino. θ Super-K cannot easily identify the third type of neutrino, the tau-neutrino. θ Such a neutrino can interact with a nu- ZENITH cleus and make a tau particle only if it has enough energy. A muon is about 200 times as heavy as an electron; the tau, about 3,500 times. The muon mass is well within the range of atmospheric neutrinos, but only a tiny fraction are at tau energies, so most tau-neutrinos in the mix will pass through Super-K undetected. One of the most basic questions ex- HIGH-ENERGY COSMIC RAY striking a nucleus in the perimenters ask is “How many?” We COSMIC atmosphere (below) generates a shower of particles, have built a beautiful detector to study RAY mostly pions. The pion’s sequence of decays produces neutrinos, and the ﬁrst task is simply to two muon-neutrinos for every electron-neutrino. Equal count how many we see. The related neutrino rates should be seen from opposite directions (above) because both result from cosmic rays hitting question is “How many did we expect?” the atmosphere at the same zenith angle, θ. Both these To answer that, we must analyze how the ratios are spoiled when muon-neutrinos traveling long neutrinos are produced. distances have time to change flavor. Super-K monitors atmospheric neu- AIR NUCLEUS trinos, which are born in the spray of particles when a cosmic ray strikes the top of our atmosphere. The incoming projectiles (called primary cosmic rays) are mostly PIONS protons, with a sprinkling of heavier nuclei such as helium or iron. Each collision MUON generates a shower of secondary particles, mostly pions and muons, which decay ELECTRON during their short ﬂight through the air, 2 MUON- creating neutrinos [see illustration at NEUTRINOS 1 ELECTRON-NEUTRINO right]. We know roughly how many cosmic rays hit the atmosphere each second and roughly how many pions and muons are made in each collision, so we can predict how many neutrinos to expect. Tricks with Ratios UNFORTUNATELY, this estimate is only accurate to 25 percent, so we take advantage of a common trick: often the ratio of two quantities can be better de- SUPER-K

DAVID FIERSTEIN termined than either quantity alone. For

www.sciam.com THE EDGE OF PHYSICS 71 COPYRIGHT 2003 SCIENTIFIC AMERICAN, INC. Which flavor is HOW QUANTUM WAVES MAKE A NEUTRINO OSCILLATE detected depends on the interference Muon-neutrino created in Two wave packets of different the upper atmosphere mass travel at different velocities Muon Pion (decays) or Muon

Interference pattern of wave packets determines probability of the neutrino flavor Not enough 100% muon- 0% muon- 100% muon- energy neutrino neutrino neutrino to make tau

0% tau- 100% tau- 0% tau- neutrino neutrino neutrino WHEN A PION DECAYS (top left), it produces a neutrino. chance, with the chances heavily favoring a muon-neutrino Described quantum-mechanically, the neutrino is apparently close to where it was produced. But the probabilities oscillate a superposition of two wave packets of different mass (purple back and forth, favoring the tau-neutrino at just the right and green). The wave packets propagate at different speeds, distance and returning to favor the muon-neutrino farther on. with the lighter wave packet getting ahead of the heavier one. When the neutrino finally interacts in the detector (top right), As this proceeds, the waves interfere, and the interference the quantum dice are rolled. If the outcome is muon-neutrino, pattern controls what flavor neutrino—muon (red) or tau a muon is produced. If chance favors the tau-neutrino, and the (blue)—is most likely to be detected at any point along the neutrino does not have enough energy to create a tau particle, flight path (bottom). Like all quantum effects, this is a game of Super-K detects nothing. —E.K., T.K. and Y.T. LAURIE GRACE

Super-K, the key is the sequential decay assumptions about the flux of neutrinos, descend deep into the atmosphere, and of a pion to a muon and a muon-neutri- how they interact with the nuclei and how these can develop differently from those no, followed by the muon’s decay to an our detector responds, we cannot explain that plunge straight in from above. electron, an electron-neutrino and an- such a low ratio—unless neutrinos are But geometry saves us: if we “look” other muon-neutrino. No matter how changing from one type into another. up into the sky at some angle from the many cosmic rays are falling on the We can play the ratio trick again to vertical and then down into the ground earth’s atmosphere, or how many pions test this surprising conclusion. The clue at the same angle, we should “see” the they produce, there should be about two to our second ratio is to ask how many same number of neutrinos coming from muon-neutrinos for every electron-neu- neutrinos should arrive from each possi- each direction. Both sets of neutrinos are trino. The calculation is more complicat- ble direction. Primary cosmic rays fall on produced by cosmic rays hitting the at- ed than that, but the final predicted ratio the earth’s atmosphere almost equally mosphere at the same angle; it is just that is accurate to 5 percent, providing a from all directions, with only two effects in one case the collisions happen over- much better benchmark. spoiling the uniformity. First, the earth’s head and in the other they are partway After counting neutrinos for almost magnetic field deflects some cosmic rays, around the world. To use this fact, we se- two years, the Super-K team has found especially the low-energy ones, skewing lect neutrino events of sufficiently high that the ratio of muon-neutrinos to elec- the pattern of arrival directions. Second, energy (so their parent cosmic ray was tron-neutrinos is about 1.3 to 1 instead of cosmic rays that skim the earth at a tan- not deflected by the earth’s magnetic the expected 2 to 1. Even if we stretch our gent make neutrino showers that do not field) and then divide the number of neu- 956 1930 1969 1933 1 1962 Wolfgang Pauli rescues Enrico Fermi formulates Frederick Reines (center) and At Brookhaven, the first Raymond Davis, Jr., first conservation of energy by the theory of beta-decay Clyde Cowen first detect the accelerator beam of neutrinos measures neutrinos from hypothesizing an unseen particle incorporating Pauli's particle, neutrino using the Savannah proves the distinction the sun, using 600 tons of that takes away energy missing now called the neutrino River nuclear reactor. between electron-neutrinos cleaning fluid in a mine in from some radioactive decays. (“little neutral one” in Italian). and muon-neutrinos. Homestake, S.D.

72 SCIENTIFIC AMERICAN THE EDGE OF PHYSICS COPYRIGHT 2003 SCIENTIFIC AMERICAN, INC. 200 Prediction without neutrino oscillation trinos going up by the number going Prediction with neutrino oscillation down. This ratio should be exactly 1 if Super-Kamiokande measurement none of the neutrinos are changing ﬂavor. 150 We saw equal numbers of high-energy electron-neutrinos going up and going down, as expected, but only half as many upward muon-neutrinos as downward ones. This ﬁnding is the second indication 100 that neutrinos are changing identity. Moreover, it provides a clue to the meta- morphosis. The upward muon-neutrinos cannot be turning into electron-neutrinos, 50 umber of Muon-Neutrino Events Muon-Neutrino of umber ) because there is no excess of upward elec- N sun

Anglo-Australian tron-neutrinos. That leaves the tau-neutrino. The muon-neutrinos that become tau-neutrinos pass through Super-K with- 0 out interaction, without detection.

); AIP EMILIO SEGRÈ VISUAL ARCHIVES, Arrival Angle and Distance Traveled by Neutrino ); DAVID MALIN Pauli Davis Fickle Flavor THE ABOVE TWO RATIOS are good 12,800 km 6,400 km 500 km 30 km 15 km evidence that muon-neutrinos are trans- forming into tau-neutrinos, but why NUMBER OF HIGH-ENERGY MUON-NEUTRINOS seen arriving on different trajectories at Super-K clearly should neutrinos switch flavor at all? matches a prediction incorporating neutrino oscillations (green) and does not match the no- oscillation prediction (blue). Upward-going neutrinos (plotted toward left of graph) have traveled far Quantum physics describes a particle enough for half of them to change flavor and escape detection. moving through space by a wave: in addi- tion to properties such as mass and charge, trino with a probability of 100 percent. when it is traversing an optically active the particle has a wavelength, can diffract, After traveling a certain distance, it looks material its polarization will rotate (that and so on. Furthermore, a particle can be like a tau-neutrino with 100 percent prob- is, oscillate) from vertical to horizontal the superposition of two waves. Now ability. At other positions, it could be ei- and so on, as its two circular components suppose that the two waves correspond to ther a muon-neutrino or a tau-neutrino. go in and out of sync. slightly different masses. Then, as the This oscillation sounds like bizarre be- For neutrino oscillations of the type ); COURTESY OF SOHO/EXTREME ULTRAVIOLET IMAGING TELESCOPE CONSORTIUM ( waves travel along, the lighter wave gets havior for a particle, but another familiar we see at Super-K, no “optically active” ahead of the heavier one, and the waves particle performs similar contortions: the material is needed; a sufficient mass dif- interfere in a way that fluctuates along the photon, the particle of light. Light can oc- ference between the two neutrino com-

); COURTESY OF THE INSTITUTE FOR ADVANCED STUDY, PRINCETON, N.J. ( particle’s trajectory [see box on opposite cur in a variety of polarizations, including ponents will cause ﬂavor oscillations

Super-Kamiokande page]. This interference has a musical vertical, horizontal, left circular and right whether the neutrino is passing through Reines analogue: the beats that occur when two circular. These do not have different air, solid rock or pure vacuum. When a

); ICRR ( notes are almost but not exactly the same. masses (all photons are massless), but in neutrino arrives at Super-K, the amount ); TIMELINE BY HEIDI NOLAND; AIP EMILIO SEGRÈ VISUAL ARCHIVES, COLLECTION ( In music this effect makes the volume certain optically active materials, light it has oscillated depends on its energy and graph SN1987A

( oscillate; in quantum physics what oscil- with left circular polarization moves the distance it has traveled. For down- lates is the probability of detecting one faster than right circular light. A photon ward muon-neutrinos, which have trav- type of neutrino or another. At the out- with vertical polarization is actually a su- eled at most a few dozen kilometers, only PHYSICS TODAY COLLECTION ( Observatory LAURIE GRACE ( set the neutrino appears as a muon-neu- perposition of these two alternatives, and a small fraction of an oscillation cycle has

picture of SNO detector,

or the sun or... 2001 1998 1987 1989 2000 Neutrino astronomy: the IMB The Z 0 decay rate is Super-K assembles Tau-neutrino events The solar neutrino puzzle is solved, and Kamiokande proton-decay precisely measured at evidence of neutrino are detected at Fermilab, with key evidence from the experiments detect 19 neutrinos SLAC and CERN, showing oscillation using completing the distinction Sudbury Neutrino Observatory from Supernova 1987A in the there are only three active atmospheric neutrinos. of neutrinos by flavor experiment in Ontario. Large Magellanic Cloud. neutrino generations. begun in 1962.

taken place, so the neutrinos’ flavor is trinos make high-energy muons, which that is what determines the oscillation slightly shifted, and we are nearly certain can travel through many meters of rock wavelength. It is not sensitive to the mass to detect their original muon-neutrino fla- and enter our detector. We count such of either one alone. Super-K’s data give a vor. The upward muon-neutrinos, pro- muons from upward-traveling neutri- mass-squared difference somewhere be- duced thousands of kilometers away, nos—downward muons are masked by tween 0.001 and 0.01 electron volt (eV) have gone through so many oscillations the background of cosmic-ray muons squared. Given the pattern of masses of that on average only half of them can be that penetrate Mount Ikenoyama. other known particles, it is likely that one detected as muon-neutrinos. The other We can count upward-traveling mu- neutrino is much lighter than the other, half pass through Super-K as undetectable ons arriving on trajectories that range which would mean that the mass of the tau-neutrinos. from directly up to nearly horizontal. heavier neutrino is in the range of 0.03 to This description is just a rough pic- These paths correspond to neutrino trav- 0.1 eV. What are the implications? ture, but the arguments based on the ra- el distances (from production in the atmo- First, giving neutrinos a mass does not tio of flavors and the up/down event rate sphere to the creation of a muon near Su- wreck the Standard Model. The mis- are so compelling that neutrino oscilla- per-K) as short as 500 kilometers (the dis- match between the mass states that make tion is now widely accepted as the most tance to the edge of the atmosphere when up each neutrino requires the introduc- likely explanation for our data. We have looking horizontally) and as long as tion of a set of so-called mixing parame- also done more detailed studies of how 13,000 kilometers (the diameter of the ters. A small amount of such mixing has the number of muon-neutrinos varies ac- earth, looking straight down). We find long been observed among quarks, but cording to the neutrino energy and the ar- that the numbers of muon-neutrinos of our data imply that neutrinos need a rival angle. We compare the measured lower energy that travel a long distance much greater degree of mixing—an im- number against what is expected for a are more depleted than higher-energy portant piece of information that any suc- wide array of possible oscillation scenar- muon-neutrinos that travel a short dis- cessful new theory must accommodate. ios (including no oscillations). The data tance. This behavior is exactly what we Second, 0.05 eV is still very close to look quite unlike the no-oscillation ex- expect from oscillations, and careful pectation but match well with neutrino analysis produces neutrino parameters oscillation for certain values of the mass similar to those from our first study. MINOS difference and other physical parameters If we consider just the three known detector (Soudan, [see illustration on preceding page]. neutrinos, our data tell us that muon-neu- Minn.) With about 5,000 events from our first trinos are changing into tau-neutrinos. two years of experimentation, we were Quantum theory says that the underlying Fermilab able to eliminate any speculation that the cause of the oscillation is almost certain- accelerator anomalous numbers of atmospheric neu- ly that these neutrinos have mass—al- (Batavia, Ill.) trinos could be just a statistical fluke. But though it has been assumed for 70 years it is still important to confirm the effect by that they do not. (The box on the oppo- looking for the same muon-neutrino os- site page mentions some other scenarios.) cillation with other experiments or tech- Unfortunately, quantum theory also niques. Detectors in Minnesota and Italy limits our experiment to measuring only have provided some verification, but be- the difference in mass-squared between cause they have measured fewer events the two neutrino components, because they do not offer the same certainty. LONG-BASELINE neutrino Corroborating Evidence oscillation experiments are planned in Japan and the U.S. FURTHER SUPPORT comes from stud- Beams of neutrinos from ies of a different atmospheric neutrino in- accelerators will be detected teraction: collisions with nuclei in the hundreds of kilometers Super-K KEK accelerator away. The experiments rock around our detector. Electron-neu- Tokyo trinos again produce electrons and subse- should confirm the oscillation phenomenon and quent showers of particles, but these are precisely measure the absorbed in the rock and never reach Su- constants of nature that

per-K’s cavern. High-energy muon-neu- control it. LAURIE GRACE

74 SCIENTIFIC AMERICAN THE EDGE OF PHYSICS COPYRIGHT 2003 SCIENTIFIC AMERICAN, INC. zero, compared with other particles. (The OTHER PUZZLES, OTHER POSSIBILITIES lightest of those is the electron, with a mass PARTICLE PHYSICISTS have been busy sorting out other indications of neutrino mass. of 511,000 eV.) So the long-held belief that For more than 30 years, scientists have been capturing electron-neutrinos generated neutrinos have zero mass is understand- by nuclear fusion processes in the sun. These experiments have always counted fewer able. But theoreticians who wish to build neutrinos than the best models predict. a Grand Unified Theory, which would el- Super-K has also counted solar neutrinos and finds only about 50 percent of the egantly combine all the forces except grav- number expected. If solar neutrinos are changing flavor, this deficit is understandable, ity at enormously high energies, also take because at solar energies Super-K responds to the electron flavor and mostly ignores note of this relative lightness of neutrinos. those transformed into the muon or tau flavors. The Sudbury Neutrino Observatory They often employ a mathematical device (SNO) in Ontario, however, which uses 1,000 tons of heavy water, has recently called the seesaw mechanism, which ac- achieved a breakthrough in proving this change. The heavy water allows SNO to tually predicts that such a small but nonze- measure the total number of neutrinos (electron, muon and tau) as well as the number ro neutrino mass is natural. Here the mass of electron-neutrinos alone, and it shows that the total is much greater. The of some extremely heavy particle, perhaps accounting seems to balance. at the Grand Unified mass scale, provides It appears that the mass splitting associated with solar neutrinos is much smaller the leverage to separate the very light neu- than that for atmospheric neutrinos. This fits a picture in which the three flavors of trinos from the quarks and leptons that neutrinos are spread over three distinct neutrino masses. But the picture doesn’t are a billion to a trillion times heavier. allow for the hint of neutrino oscillation, suggesting much larger masses, detected at Another implication is that the neu- Los Alamos National Laboratory. Some exotic explanations are waiting in the wings trino should be considered in the book- while Fermilab checks this signature. keeping of the mass of the universe. For Physicists are also checking the theory that transforms solar neutrinos. In a cavern some time, astronomers have been trying in the same zinc mine as Super-K, a detector has been built that uses 1,000 tons of to tabulate how much mass is found in mineral oil doped with a chemical that emits light in response to the neutrino reaction. luminous matter, such as stars, and in or- This detector counts electron-neutrinos from more than two dozen Japanese nuclear dinary matter that is difficult to see, such power reactors, from 80 to 400 kilometers away. The results are being compared with a as brown dwarfs or diffuse gas. The mass precise model of how many neutrinos are expected from each reactor. This experiment can also be measured indirectly from the should pin down the detailed particle physics revealed by solar neutrinos. orbital motion of galaxies and the rate of Overall, our picture of neutrinos is just coming into focus. More clarity will rely on more expansion of the universe. The direct ac- ambitious projects. Later in this decade Super-K will be exposed to a beam of neutrinos counting falls short of these indirect mea- from a much more intense accelerator being built near Japan’s Pacific coast. The goal is to sures by a factor of 20. The neutrino mass verify that muon-neutrinos change flavor to tau- and electron-neutrinos in a proportion suggested by our result is too small to re- that fits our newfound expectations. Follow-up measurements may reveal the role of solve this mystery by itself. Nevertheless, neutrinos in the matter-antimatter imbalance of the universe. Or we may be presented neutrinos created during the big bang per- with new puzzles to solve. —E.K., T.K. and Y.T. meate space and could account for a mass nearly equal to the combined mass of all the stars. They could have affected the Of course, a good atmospheric neu- terize the muon-neutrinos before they os- formation of large astronomical struc- trino detector is also a good accelerator cillate. We are essentially using the ratio tures, such as galaxy clusters. neutrino detector, so in Japan we are us- (again) of the counts near the source to Finally, our data have an immediate ing Super-K to monitor a beam of neutri- those far away to cancel uncertainty and implication for two new experiments. nos created at the KEK accelerator labo- verify the effect. Since 1999, neutrinos Based on the earlier hints from smaller de- ratory 250 kilometers away. Unlike at- from the first long-distance artificial neu- tectors, many physicists have decided to mospheric neutrinos, the beam can be trino beam have passed under the moun- stop relying on the free but uncontrollable turned on and off and has a well-defined tains of Japan, with 50,000 tons of Super- neutrinos from cosmic rays and instead energy and direction. Most important, we K capturing a small handful. Exactly how are creating them with high-energy accel- have placed a detector similar to Super- many are being captured will be the next erators. Even so, the neutrinos must trav- K near the origin of the beam to charac- chapter in this story. el a long distance for the oscillation effect to be observed. So the neutrino beams are MORE TO EXPLORE aimed at a detector hundreds of kilome- The Search for Proton Decay. J. M. LoSecco, Frederick Reines and Daniel Sinclair in Scientific American, Vol. 252, No. 6, pages 54–62; June 1985. ters away. One detector, MINOS, is be- The Elusive Neutrino: A Subatomic Detective Story. Nickolas Solomey. Scientific American Library, ing built in a mine in Soudan, Minn., to W. H. Freeman and Company, 1997. study neutrinos sent from the Fermilab Official Super-Kamiokande Web site: www-sk.icrr.u-tokyo.ac.jp/doc/sk/ accelerator near Batavia, Ill., 730 kilome- K2K Long Baseline Neutrino Oscillation Experiment Web site: neutrino.kek.jp/ ters away on the outskirts of Chicago. Super-Kamiokande at Boston University Web site: hep.bu.edu/~superk/ www.sciam.com THE EDGE OF PHYSICS 75 COPYRIGHT 2003 SCIENTIFIC AMERICAN, INC.