The Telescope in the Ice Page 9
This form of radiation is named for the Russian physicist Pavel Alekseyevich Cherenkov, who won a share of the 1958 physics Nobel for discovering it. It is produced whenever a charged particle, such as a muon, speeds through a refractive medium faster than light can. The most common example is the eerie blue light given off by pool-type fission reactors or spent nuclear fuel rods immersed in water. In this case, the light is produced by electrons emitted in the beta decay of the many radioactive byproducts of the uranium reactor fuel.
If you’ve heard of the postulate in Einstein’s special theory of relativity that nothing can travel faster than the speed of light, don’t worry; we’re not violating that rule here. For Einstein was speaking of the speed of light in a vacuum. In a refractive medium such as water, ice, glass, or even air, light will travel less quickly, so that it is possible for a different kind of particle to go faster than light can in that medium and not break the rule.
Cherenkov radiation is the optical equivalent of the sonic boom that occurs when a jet “breaks the sound barrier,” or goes faster than the speed of sound: since the sound can’t keep up with the jet, it is dragged along behind it in the same way that the waves in the wake of a speedboat ride off at an angle because they can’t keep up with the boat. In a three-dimensional situation like that of the jet or a speeding muon, the sound or light waves take the shape of a cone, rather than the V of a boat wake. The muon, in other words, drags a cone of Cherenkov light along behind it. If it were to pass through a projection screen, a spot of light would appear on the screen at the moment it passed through, and the spot would change immediately into a tiny circle, which would expand and gradually fade as the muon zoomed away.
The difference between Greisen and Markov’s concepts is basically one of geometry. Markov’s idea was to place a three-dimensional grid of light detectors in a natural body of water and watch neutrino-born muons pass through the grid. We’ll call this the Markov or “plum pudding” design, because the detectors are located inside the so-called detection volume. Greisen’s idea was to surround a manmade tank of water with a shell of detectors, in which case the detectors would obviously be outside the detection volume. This is the “shell” or Greisen design. In both cases, the bigger the detection volume, the more sensitive the detector, since this increases the likelihood that a neutrino or neutrino-born muon will pass through it.
Stepping a few decades ahead for a moment, let’s visualize the cone of Cherenkov light being dragged along by a muon as it passes through IceCube. Think of the cone as a three-dimensional version of the waves produced by that boat as it skims along the surface of a calm lake. The three-dimensional grid of light detectors that the scientists have placed in the ice is analogous to a two-dimensional grid of pontoons floating on the surface of the lake. As the boat drags its wake through the grid, the pontoons will bob up and down. If you know the speed of the waves, then with the help of a little algebra and geometry you can reconstruct the speed and path of the boat using the timing of the wave fronts as they hit the different pontoons. The IceCube scientists reconstruct the direction and speed of an invisible muon zooming through their 3D detector in precisely the same way. Francis Halzen says, “It’s like flying over the lake in an airplane. You might not be able to see the boat, but the waves tell you where it is and where it’s going.”
The reason Markov and Greisen focused on muons rather than electrons was that they knew they would be easier to detect. For the muon has enough mass and therefore momentum to travel in a straight line through whatever medium it finds itself in; whereas an electron, being two hundred times lighter, will be swayed by the electric fields of nearby nuclei and begin to zig and zag within a few meters of its birth. With each zig the electron produces bremsstrahlung radiation: photons, which, if they are energetic enough, will produce electron-positron pairs. These secondary pairs will in turn zig and zag and produce more bremsstrahlung, which will produce more pairs, and so on and so forth. The result, in IceCube parlance, is a cascade: a short, cigar-shaped flash of light, proportional in volume to the energy of the electron neutrino that created it and pointing in the direction that the neutrino was traveling in.
Not only is the long, straight track of a muon much easier to see than a cascade, its direction is much more precisely defined, so that it yields a more accurate measure of the direction of its parent neutrino. This makes a track much more useful for astronomy, since it points back more accurately at whatever cosmic object the neutrino came from. And highly energetic muons (which come from highly energetic neutrinos) have the added advantage of flying through ice or bedrock for several kilometers before they decay. This is what Markov and Zheleznykh meant by a cushion: the plum pudding design can detect a muon even if it’s born a good distance outside the detector grid. Remember that we’re looking for up-going muons created by neutrinos that have entered the planet somewhere north of the South Pole. IceCube will detect a muon born in the ice or bedrock to its side or beneath it as long as the muon’s path eventually takes it into the grid. This enhances the effective volume of the detector and makes this type of instrument more sensitive to muon neutrinos than electron neutrinos, generally speaking.
Two up-going muon neutrinos (νμ) detected by an instrument of the Markov or “plum pudding” design. The neutrino on the left interacts with a nucleon in the bedrock or “cushion” below the instrument, giving rise to a muon (μ) that passes through it dragging along a cone of Cherenkov light. The neutrino on the right interacts inside the detector. The tracks of the muons indicate the directions of the corresponding neutrinos.
* * *
The reason a neutrino detector needs to be placed deep in the earth (or water or ice) is to shield it from cosmic rays raining down from above. The primary cosmic rays racing toward our planet—mostly protons and other charged nuclei—collide with nitrogen, oxygen, and other nuclei in the upper atmosphere and create showers of down-going pions and other “secondary” cosmic rays, which then decay into other particles, such as muons, or collide with the atmosphere themselves, to create so-called air showers. And this serves as a good example of the similarity between cosmic accelerators and the manmade variety.
The principle behind all accelerators is to employ powerful electromagnetic fields to accelerate charged particle beams to high energies and then slam them into targets or “beam dumps.” In the case of a manmade accelerator this might be a block of carbon, and for cosmic rays it’s our atmosphere. The “primary” cosmic rays, protons and nuclei, have been accelerated by some cataclysm in the cosmos, launched into interstellar space, and then steered by interstellar electromagnetic fields in a winding course to our planet. They hit the beam dump of our atmosphere and produce secondary particles in the same way that the beam in an earthbound accelerator does when it hits its beam dump.
The way accelerator physicists generate rare or new particles is to start off with a relatively easily produced charged particle, such as a proton; accelerate it either in a circle or a straight line, employing focused electromagnetic fields; divert it into a beam dump; and examine the resulting shards and debris with special detectors. In some designs, two beams are aimed at each other. (Richard Feynman compared this to smashing together two Swiss watches in order to find out what’s inside.)
The standard unit of energy in particle physics is the electron-volt (eV), the amount of kinetic energy acquired by an electron when it is accelerated across a potential difference of one volt. This is a miniscule quantity by everyday standards—a one-hundred-watt light bulb puts out almost 1021 (one billion-trillion) electron-volts every second—but it’s a convenient unit for expressing the masses of elementary particles. The electron, for example, has a so-called rest mass (its mass-energy in its “rest frame,” exclusive of its energy of motion) of about 510 eV. And since Einstein has shown that mass and energy are equivalent, the way to generate more massive particles in the debris of a particle beam that has just hit its dump is to generate higher and higher bea
m energies, either by making the accelerator larger, so that it can accelerate the beam over longer distances, or by employing stronger electromagnetic fields.
The present record holder is CERN’s $10 billion Large Hadron Collider, where the Higgs boson was discovered in 2012. The LHC is designed to cycle opposing beams of either protons or lead nuclei around in a circular tunnel twenty-seven kilometers, or seventeen miles, in circumference and smash them together at energies, in the case of the proton, of about fourteen trillion electron-volts.
Since cosmic accelerators are unconstrained by international science budgets and real estate considerations, however, such as the circumference of the Earth or even the Solar System, they generate energies far greater than anything humans can dream of replicating. The record thus far is held by the so-called Oh-My-God particle, which was observed by an instrument called the Fly’s Eye in a Utah desert in 1991. (Kenneth Greisen, incidentally, invented this instrument.) This single subnuclear spec packed as much punch as a baseball going about sixty miles an hour—about three hundred thousand times the capability of the LHC. It seems to have been either a proton, a heavy nucleus, or even a neutrino; but it’s impossible to know, since it died when it hit the atmosphere, giving birth to a shower of about two hundred billion secondary particles and decay products.
The mile of frozen overburden above IceCube serves as a shield from such down-going cosmic rays; however, many still punch through, penetrating deeply enough to reach the array. In fact, about a million atmospheric muons streak down through the detector for every more interesting muon, born of a neutrino, that streaks up from below. Thus one of the major challenges for this technology is to distinguish up-going from down-going muon tracks. Picking the up-going needles from the down-going haystack is a tricky business, as you might imagine.
* * *
Greisen’s shell design is typically realized as a large tank of some ultra-pure, clear liquid, usually water, completely surrounded by walls of cheek-by-jowl light detectors, situated a mile or more deep in a mine of some sort. The idea here is to detect only those muons that are born inside the tank, and to aid in this purpose Greisen suggested that the primary detector “be … enclosed in a shell of scintillating material to distinguish neutrino events from those caused by [muons].” In other words, there were actually two shells: an inner one of primary light detectors and an outer one of scintillators, which would be used to exclude or “veto” atmospheric and other muons born outside the detector. Since these particles will pass all the way through the instrument, they will light up scintillators as they enter and as they leave, while muons that are born inside the volume will only cause a signal on their way out.
Muon neutrinos (νμ) passing through an instrument of the Greisen or shell design. The upper neutrino, entering from the left, interacts inside the detector, so that the light cone from the resulting muon (μ) illuminates sensing and veto detectors only on its way out. The lower neutrino, entering from the right, interacts outside the detector, giving rise to a muon that illuminates detectors both as it enters and as it exits. Thus the shell of veto detectors helps reject muons born outside the detector.
Any muon born in the tank (or passing through it) will light up a ring of primary detectors on the outer wall as it exits: one detector at first, followed by an expanding ring around it. And the muon’s direction can be inferred from the shape of the ring: if it’s passing through at an angle, it will produce an oval rather than a circle.
The resolution of the shell design is limited to twenty or thirty degrees—good enough to tell left from right and up from down, but not much better than that—while the plum pudding design (as embodied in IceCube) can resolve to less than half a degree, roughly the angular spread of a full moon.
* * *
John Learned, who would lead DUMAND twenty years later, the first attempt to make these dreams a reality, observes that “Greisen never did anything in this direction. Reines grabbed the ball and was running with it in the States, Reines and his gang from early on.”
In a 1960 review article, Reines discussed the detection of both “neutrinos produced extra terrestrially (cosmic) and in the earth’s atmosphere (cosmic ray).” But he comes across as being much more conservative than Greisen and Markov, possibly owing to the scars from his vexing attempt to detect manmade neutrinos in the first place: he doesn’t even touch the first problem, and he pronounces the second “most formidable.” Nevertheless, he began visiting mines to scout out locations for an instrument in the Greisen tradition as early as 1963 and was considering a Markov instrument in the deep ocean at least as early as 1966.
He must have concluded that Cherenkov detection was too much of a long shot, however, since he ended up attempting to detect atmospheric (or, in his words, cosmic ray) neutrinos by a different method. This was still a big step: the first attempt to detect neutrinos produced by nature, rather than a manmade reactor or bomb.
Atmospheric neutrinos are created in the same way as atmospheric muons: through the decay of secondary particles, including muons, born in the collision of primary cosmic rays with the atmosphere. They’re useful for some kinds of particle physics, as we’ll see, but have nothing to do with astronomy, because they’re born so close by.
Sometime in 1963, Reines became aware of a doctoral thesis at the University of Bombay, suggesting that certain mines in India’s Kolar Gold Fields might be deep enough to provide sufficient shielding against atmospheric muons to permit the detection of atmospheric neutrinos. He visited the Indian scientists who had come up with the idea and the mines themselves, but eventually chose the deepest mine in the world, the East Rand Proprietary Gold Mine near Johannesburg, South Africa. By this time, he had left Los Alamos to become head of the physics department at the Case Institute of Technology, now Case Western Reserve University, in Cleveland, Ohio.
In collaboration with a group from the University of Witwatersrand in Johannesburg, Reines installed the largest particle detector ever built at the time, comprising twenty tons of liquid scintillator, in a laboratory two miles underground. Evidently, the local miners referred to the scientists as “goggafangers,” meaning “bug catchers,” and to Reines himself as “makulu bass goggafanger,” or “big boss bug catcher.”
The idea was to detect muons traveling horizontally through two parallel walls of scintillators. Since there was significantly more than two miles between the detector and the surface of the Earth in the horizontal direction, any muon created in the atmosphere would decay before reaching the detector from that direction, so those that did reach it must have been created by a neutrino interacting somewhere in the earth between the detector and the surface.
But this time around, Reines had some direct competition. The Indian group whose idea he had borrowed, led by M. G. K. Menon, had gone ahead and installed their own detector in one of the Kolar gold mines. The Case-Witwatersrand group detected their very first naturally born neutrino on February 23, 1965 (a date that recurs in neutrino astronomy), and the Indians detected theirs about a month later. But Menon’s group is accorded formal priority, because they published their results about two weeks before Reines’s did. (Publication is important, as it usually indicates that a rigorous analysis has been done and uncertainties and potential misinterpretations addressed.) The two sides tended to argue about it, of course (discord tended to follow Reines), but pretty much everyone else chalked it up as a tie.
Such pettiness aside, the neutrino again demonstrated its shyness: all told, the Case-Witwatersrand group detected only 167 atmospheric neutrinos over the six years that they took data.
* * *
Meanwhile, the seeds planted by Markov and Greisen gestated in fertile ground. Around the world, particularly in the United States and Russia, solitary individuals began visiting remote lakes in order to drop short strings of light detectors into the water and fish for muons. One was the young John Learned, a grad student at the University of Washington.
John was born in 1
940 in Plattsburgh, New York, on the edge of the vast Adirondack State Park, which takes up a large portion of the northern part of the state. His grandparents lived in a sort of ancestral home there, which had been in the family for about a hundred and fifty years. When he was six, his journalist father moved the family to Staten Island in New York City, but John and his brother kept in touch with their rural roots by spending all their summers “pretty much unsupervised” in the north country: “We hunted, fished, hiked around, and camped out as we pleased. We built projects in Grandpa’s shop (including cannons which we shot at each other), built models and tied flies … Indeed it was quite idyllic.” Thus he developed a passion for exploring, tinkering, and messing around in the wilderness, which is basically what cosmic ray physicists do for a living.
John’s years at Brooklyn Technical High also proved useful, as they gave him access to printing, pattern making, tin and machine shops, and even a foundry. He majored in physics at Columbia, and then worked in astronautics and aerospace engineering for a few years, first in the east at General Dynamics and then at Boeing in Seattle. After a few years, when Boeing began grooming him for management, he left, because he liked the technical stuff better. Too bad. A little management experience might have come in handy later on.
John first heard about the odd sport of neutrino fishing in his very first class at the University of Washington. He ended up pursuing the idea for his Ph.D. thesis, with the teacher of that class, Howard Davis, as his adviser.
“It was just wonderful fun, wonderful fun, because essentially I worked alone. I built a detector, which I took to Lake Chelan in the Cascade Mountains, and built a raft and had a boat and went out there and sank detectors down and counted cosmic rays as a function of depth and so on. Had a lot of interesting experiences. Didn’t do any science of any particular merit.…