By 1959 another crisis, two in fact, arose to tweak the physicist's mind. The center of the storm was at Columbia University, but the crisis was liberally shared and appreciated around the world. All of the data on the weak force to that time were kindly provided by particles during natural decay. Greater love hath no particle than to give its all for the edification of physicists. To study the weak force we simply watched particles, such as the neutron or the pion, decay into other particles. The energies involved were provided by the rest masses of the decaying particles—typically from a few MeV to around 100 MeV or so. Even the free neutrinos shooting out of reactors and undergoing weak-force collisions involved only a few MeV. After we had modified the weak-force theory with the experimental results of parity violation, we had one zinger of an elegant theory that fit all the available data provided by zillions of nuclear decays as well as the decays of pions, muons, lambdas, and probably, though difficult to prove, Western civilization.
THE EXPLODING EQUATION
Crisis No. 1 had to do with the mathematics of the weak force. In the equations, the energy at which the force is measured appears. Depending on the data, you stick in the rest mass energy of the decaying particle—1.65 MeV or 37.2 MeV or whatever—and out comes the right answer. You manipulate the terms, bump and grind, and, sooner or later out come predictions as to the lifetimes, decays, spectra of electrons—things that can be compared to experiment—and they are right. But if one puts in, say, 100 GeV (billion electron volts), the theory goes haywire. The equation explodes in your face. In the jargon of physics, this is called "the unitarity crisis."
Here's the dilemma. The equation was okay, but it had a pathology at high energy. Little numbers worked; big ones didn't. We didn't have the ultimate truth, only a truth valid for the low-energy domain. There had to be some new physics that modified the equations at high energy.
Crisis No. 2 was the mystery of the unobserved reaction. One could calculate how often a muon decayed into an electron and a photon. Our theory of the weak processes said that this should happen. Looking for this reaction was a favorite Nevis experiment, and several new Ph.D.'s spent godknowshowmany beam hours searching with no success. Murray Gell-Mann, the pundit on all matters arcane, is often quoted as the source of something called the Totalitarian Rule of Physics: "Anything that isn't forbidden is compulsory." If our laws do not rule out an event, it not only can happen! it must happen! Since a muon decaying into an electron and a photon was not forbidden, why weren't we seeing it? What forbade this mu-e-gamma decay? (For "gamma" read "photon.")
Both crises were exciting. Both offered up the possibility of new physics. Theoretical speculations abounded, but experimental blood boiled. What to do? We experimenters must measure, hammer, saw, file, stack lead bricks—do something. So we did.
MURDER INC. AND THE TWO-NEUTRINO EXPERIMENT
Melvin Schwartz, an assistant professor at Columbia, after listening to a detailed review of the troubles by Columbia theorist T. D. Lee in November 1959 came up with his GREAT IDEA. Why not create a beam of neutrinos by letting a high-energy pion beam drift through enough space that some fraction, say 10 percent, of the pions decayed into a muon and a neutrino. Pions, in flight, would disappear; muons and neutrinos, sharing the pion's original energy, would appear. So here, flying through space, we have muons and neutrinos from the 10 percent of pions that decayed, plus the 90 percent of pions that didn't decay, plus a bunch of nuclear debris originating from the target that produced the pions. Now, said Schwartz, let's aim it all into a big thick wall of steel, forty feet thick, as it turned out. The wall would stop everything but the neutrinos, which would have no trouble passing through forty million miles of steel. We'd have a pure beam of neutrinos on the other side of the wall, and since the neutrino obeys only the weak force, we'd have a handy way of studying both the neutrino and the weak force via neutrino collisions.
The scheme addressed both Crisis No. 1 and Crisis No. 2. Mel's idea was that this neutrino beam would allow us to study the weak force at energies of billions rather than millions of electron volts. It would give us a view of the behavior of the weak force at high energy. It might also provide some ideas on why we don't see muons decay into electrons plus photons, based on the notion that neutrinos are somehow involved.
As happens so often in science, an almost equivalent idea was published almost simultaneously by a Soviet physicist, Bruno Pontecorvo. If the name seems more Italian than Russian, it is because Bruno is an Italian who defected to Moscow in the 1950s on ideological grounds. His physics, ideas, and imagination were nevertheless outstanding. Bruno's tragedy was in trying to carry out his imaginative ideas within a system of stultifying bureaucracy. International conferences are venues for displaying the traditional warm friendship of scientists. At one such conference in Moscow, I asked a friend, "Yevgeny, tell me, which one of you Russian physicists is really a communist?" He looked around the hall and pointed to Pontecorvo. But that was in 1960.
When I returned to Columbia from a pleasant sabbatical at CERN in late 1959, I listened to the discussions about crises in the weak force, including Schwartz's idea. Schwartz had somehow concluded that no existing accelerator was powerful enough to make a sufficiently intense neutrino beam, but I disagreed. The 30 GeV AGS (for Alternating Gradient Synchrotron) was nearing completion at Brookhaven, and I did the numbers and convinced myself and then Schwartz that the experiment was, in fact, doable. We designed what was, for 1960, a huge experiment. Jack Steinberger, a colleague at Columbia, joined us and with students and postdocs we formed a group of seven. Jack, Mel, and I were well known for our gentle and kindly demeanor. Once as we were walking across the Brookhaven accelerator floor I overheard a physicist in a group exclaim, "There goes Murder; Incorporated!"
To block all the particles except the neutrinos, we made a thick wall around a massive detector, using thousands of tons of steel from outdated naval vessels. I once made the mistake of telling a reporter that we took apart the battleship Missouri to make the wall. I must have gotten the name wrong, because the Missouri is apparently still out there someplace. But we certainly had a battleship cut up for scrap. I also made the mistake of joking that if there was a war we'd have to paste the ship back together, and that story got embellished and pretty soon there was a rumor that the navy had confiscated our experiment to fight some war (what war this could have been—it was 1960—remains a puzzle).
What is also somewhat fabricated is my story about the cannon. We got a twelve-inch naval cannon with a suitable bore and thick walls—it made a beautiful collimator, a device for focusing and aiming a beam of particles. We wanted to fill it up with beryllium as a filter but the bore had these deep rifling grooves. So I sent a skinny graduate student inside to stuff steel wool into the grooves. He spent about an hour in there and crawled out all hot, sweaty, and irritated and said, "I quit!" "You can't quit," I cried. "Where will I find another student of your caliber?"
Once our preparations were finished, steel from obsolete ships surrounded a detector made from ten tons of aluminum tastefully arranged so that if neutrinos collided with an aluminum nucleus, the products of the collision would be observed. The detector idea we eventually used, called a spark chamber, had been invented by a Japanese physicist, Shuji Fukui. We learned a lot by talking to Jim Cronin of Princeton who had mastered the new technique. Schwartz won the ensuing contest as to the best design that could be scaled up from a few pounds to ten tons. In this spark chamber, nicely machined one-inch-thick plates of aluminum were spaced about a half inch apart and a huge voltage difference applied between adjacent plates. If a charged particle passed through the gap, a spark would follow the trail of the particle and could be photographed. How easily this is said! The technique was not without its technical problems. But the results! Zap—and the path of a subnuclear particle was rendered visible in the red-yellow light of glowing neon gas. It was a lovely device.
We built models of spark chambers and put them in beams
of electrons and pions to learn their characteristics. Most chambers of that day were about a foot square and had ten to twenty plates. The design we set about had one hundred plates, each four feet square. Each plate was one inch thick, pleading with the neutrinos to collide. Seven of us worked day and night as well as other times to assemble the apparatus and the electronics, inventing all sorts of devices—hemispherical spark gaps, automated gluing facilities, circuitry. We had help from engineers and several technicians.
We started the run late in 1960 and were immediately plagued by background "noise" created by neutrons and other debris from the target sneaking around our formidable forty feet of steel, crudding up our spark chambers, and skewing our results. Even if only one particle in a billion got through, it created problems. Leave it to background to know that one chance in a billion is the legal definition of a miracle. We struggled for weeks plugging cracks anywhere neutrons could sneak in. We searched diligently for electrical ducts under the floor. (Mel Schwartz, exploring, crawled into one, got stuck, and had to be hauled out by several strong technicians.) Every thin area was plugged with blocks of rusty steel from the ex-battleship. At one point, the director of Brookhaven's brand-new accelerator drew the line: "You'll pile those dirty blocks near my new machine over my dead body," he thundered. We didn't take him up on his offer as this would have made an unsightly lump in the shielding. So we compromised—only slightly. By late November the background was reduced to manageable proportions.
Here is what we were doing.
The protons from the AGS smashed into a target, producing about three pions on average for each collision. We produced about 1011 (100 billion) collisions per second. Assorted neutrons, protons, occasional antiprotons, and other debris were also generated. The debris that headed our way crossed a space of about fifty feet before smashing into our impenetrable steel wall. In that distance some 10 percent of the pions decayed so we had something like a few tens of billions of neutrinos. A much smaller number headed in the right direction, toward our forty-foot-thick steel wall. On the other side of the wall, about a foot away, our detector the spark chamber lay waiting. We estimated that if we were lucky we'd see one neutrino collision in our aluminum spark chamber per week! In that week the target would spray about 500 million billion (5 × 1017) particles in our general direction. This is why we had to reduce background so severely.
We expected two kinds of neutrino collisions: (1) a neutrino hits an aluminum nucleus, which results in a muon and an excited nucleus, or (2) a neutrino hits a nucleus, which results in an electron and an excited nucleus. Forget about the nuclei. What's important is that we expected muons and electrons to emerge from the collision in equal numbers, accompanied by occasional pions and other debris from the excited nucleus.
Virtue triumphed and in an eight-month exposure we observed fifty-six neutrino collisions, of which perhaps five were spurious. Sounds easy, but I will never never forget that first neutrino event. We had developed a roll of film, the result of a week of data taking. Most of the frames were empty or showed some obvious cosmic ray tracks. But suddenly, there it was: a spectacular collision with a long, long muon track speeding away. That first event was the mini-Eureka moment, the flash of certainty, after so much effort, that the experiment would work.
Our first task was to prove that these were indeed neutrino events since this was the first experiment of its kind ever. We pooled all of our experience and took turns playing devil's advocate in trying to pick holes in our own conclusion. But the data were in fact rock solid, and it was time to go public. We felt secure enough to present the results to our colleagues. You should have heard Schwartz's talk to a jammed Brookhaven auditorium. Like a lawyer he ruled out, one by one, all possible alternatives. There were smiles and tears in the audience. Mel's mother had to be helped out, sobbing uncontrollably.
There were three (always three) major consequences of the experiment. Remember that Pauli first posited the existence of the neutrino to explain the missing energy in beta decay, in which an electron is ejected from the nucleus. Pauli's neutrinos were always associated with electrons. In almost all of our events, however, the product of the neutrino collision was a muon. Our neutrinos refused to produce electrons. Why?
We had to conclude that the neutrinos we were using had a new specific property of "muon-ness." Since these neutrinos were born with a muon in the decay of pions, somehow "muon" was imprinted on them.
To prove this to the audience of genetically conditioned skeptics, we had to know and show that our apparatus did not more readily see muons, and that it therefore—by stupid design—was incapable of detecting electrons. Galileo's telescope problem all over again. Fortunately, we were able to demonstrate to our critics that we had built electron-detection capability into our equipment and had indeed verified this in test beams of electrons.
Another background effect came from cosmic radiation, which at sea level consists of muons. A cosmic-ray muon coming in from the back of our detector and stopping in the middle could be mistaken by lesser physicists as a muon from neutrinos going out, which is what we were looking for. We had installed a "block" against this, but how could we be sure it worked?
The key was to keep the detector going whenever the machine was shut down—which was about 50 percent of the time. When the accelerator was off, any muons that showed up would be uninvited cosmic rays. But none appeared; cosmic rays were unable to get past our block.
I mention all these technical details to show you that experimentation is not so easy and that the interpretation of an experiment is a subde affair. Heisenberg once commented to a colleague outside the entrance to a swimming pool, "These people go in and out all very nicely dressed. Do you conclude from this that they swim dressed?"
The conclusion we—and most others—drew from the experiment was that there are (at least) two neutrinos in nature—one associated with electrons (the plain vanilla Pauli neutrinos) and one associated with muons. So we call them electron neutrinos (plain) and muon neutrinos, the kind we produced in our experiment. The distinction is now known as "flavor," in the whimsical lingo of the standard model, and people began to draw a little table:
electron neutrino muon neutrino
electron muon
or in physics shorthand:
νe νμ
e μ
The electron is placed under its cousin, the electron neutrino (indicated by the subscript), and the muon under its muon neutrino cousin. Let's recall that before this experiment we knew of three leptons—e, ν, and μ—which were not subject to the strong force. Now there were four: e, νe, μ, and νμ. The experiment was forever called the experiment of the Two Neutrinos, which ignorant people think is an Italian dance team. This turns out to be the button upon which the standard model overcoat is sewn. Note that we have two "families" of leptons, pointlike particles, arranged vertically. The electron and electron neutrino are the first family, which is found everywhere in our universe. The second family consists of the muon and the muon neutrino. Muons are not found readily today in the universe, but must be manufactured in accelerators or in other high-energy collisions, such as those produced by cosmic rays. When the universe was young and hot, these particles were abundant. When the muon, a heavy brother of the electron, was first discovered, 1.1. Rabi asked, "Who ordered that?" The two-neutrino experiment provided one of the early clues to the answer.
Oh yes. The fact that two different neutrinos existed solved the crisis of the missing mu-e-gamma reaction. To review, a muon should decay into an electron and a photon, but no one was able to detect this reaction, though many tried. There should be a sequence of processes: a muon should first decay into an electron and two neutrinos—a regular neutrino and an antineutrino. These two neutrinos, being matter and antimatter, then annihilate, producing the photon. But nobody was seeing these photons. The reason why was now obvious. Clearly, the positive muon decays into a positron and two neutrinos, but these are an electron neutrino and an anti
muon neutrino. These neutrinos don't annihilate each other because they're from different families. They simply stay neutrinos, and no photon is produced, thus no mu-e-gamma reaction.
The second consequence of the Murder Inc. experiment was the creation of a new tool for physics: hot and cold running neutrino beams. These appeared, in due course, at CERN, Fermilab, Brookhaven, and Serpuhkov (USSR). Remember, previous to the AGS experiment, we weren't totally sure neutrinos existed. Now we had beams of them on demand.
Some of you might have noticed that I'm avoiding an issue here. What happened to Crisis No. 1, the fact that our equation for the weak force doesn't work at high energies? Indeed, our 1961 experiment demonstrated that the collision rate was increasing with energy. By the 1980s, the accelerator labs mentioned above—using more intense beams at higher energies and detectors weighing hundreds of tons—were collecting millions of neutrino events at the rate of several per minute (a lot better than our 1961 yield of one or two a week). Even so, the high-energy crisis of weak interactions was not solved, though it was greatly illuminated. The rate of neutrino collisions did increase with higher energy, as the low-energy theory predicted. However, the fear that the collision rate would become impossibly large was alleviated by the discovery of the W particle in 1982. This was part of the new physics that modified the theory and led to a gentler and kinder behavior. This postponed the crisis to which, yes, we will return.
The God Particle: If the Universe Is the Answer, What Is the Question? Page 37
