The God Particle
Page 41
The news came to me, hard at work on a Fermilab experiment with the exotic designation E-70. Can I now, writing in my study seventeen years later recall my feelings? As a scientist, as a particle physicist, I was overjoyed at the breakthrough, a joy tinged, of course, with envy and even just a touch of murderous hatred for the discoverers. That's the normal reaction. But I had been there—Ting was doing my experiment! True, the kinds of chambers that made Ting's experiment sharp weren't available in 1967–68. Still, the old Brookhaven experiment had the ingredients of two Nobel Prizes—if we had had a more capable detector and if Bjorken had been at Columbia and if we had been slightly more intelligent ... And if my grandmother had had wheels—as we used to taunt "iffers"—she would have been a trolley car.
Well, I can only blame myself. After spotting the mysterious bump in 1967, I had decided to pursue the physics of dileptons at the newer high-energy machines coming on the air. CERN, in 1971, was scheduled to inaugurate a proton-proton collider, the ISR, with an effective energy twenty times that of Brookhaven's. Abandoning my Brookhaven bird in hand, I submitted a proposal to CERN. When that experiment started taking data in 1972, I again failed to see the J/psi, this time because of a fierce background of unexpected pions and our newfangled leaded-glass particle detector which was, unknown to us, being irradiated by the new machine. The background turned out to be a discovery in itself: we detected high-transverse-momentum hadrons, another kind of data signifying the quark structure inside protons.
Meanwhile, also in 1971, Fermilab was getting ready to start a 200 GeV machine. I gambled on this new machine too. The Fermilab experiment turned on in early 1973, and my excuse was ... well, we really didn't get down to doing what we had proposed to do, being diverted by curious data several groups had been seeing in the brand-new Fermilab environment. It turned out to be a red herring or a blue shrimp, and by the time we got around to dileptons, the November Revolution was in the history books. So not only did I miss the J at Brookhaven, I missed it at both new machines, a new record of malpractice in particle physics.
I haven't yet answered the question, what was the big deal? The J/psi was a hadron. But we have discovered hundreds of hadrons, so why blow a gasket over one more, even if it has a fancy name like J/psi? It has to do with its high mass, three times heavier than the proton, and the "sharpness" of the mass, less than 0.05 MeV.
Sharpness? What that means is the following. An unstable particle cannot have a unique, well-defined mass. The Heisenberg uncertainty relations spell it out. The shorter the lifetime, the wider the distribution of masses. It is a quantum connection. What we mean by a distribution of masses is that a series of measurements will yield different masses, distributed in a bell-shaped probability curve. The peak of this curve, for example 3.105 GeV, is called the mass of the particle, but the spread in mass values is in fact a measurement of the particle's lifetime. Since uncertainty is reflected in measurement, we can understand this by noting that for a stable particle, we have infinite time to measure the mass and therefore the spread is infinitely narrow. A very short lived particle's mass cannot be determined precisely (even in principle), and the experimental result, even with superfine apparatus, is a broad spread in the mass measurements. As an example, a typical strong-interaction particle decays in 10−23 seconds and has a mass spread of about 100 MeV.
One more reminder. We noted that all hadron particles are unstable except the free proton. The higher the mass of a hadron (or any particle), the shorter its lifetime because it has more things into which it can decay. So now we find a J/psi with a huge mass (in 1974 it was the heaviest particle yet found), but the shock is that the observed mass distribution is exceedingly sharp, more than a thousand times narrower than that of a typical strong-interaction particle. Thus it has a long lifetime. Something is preventing it from decaying.
NAKED CHARM
What inhibits its decay?
Theorists all raise their hands: a new quantum number or, equivalently, a new conservation law is operating. What kind of conservation? What new thing is being conserved? Ah, now all the answers were different, for a time.
Data continued to pour in, but now only from the e+ e− machines. SPEAR was eventually joined by a collider in Italy, ADONE, and later by DORIS, in Germany. Another bump showed at 3.7 GeV. Call it ψ (psi prime), no need to mention J, since this was Stanford's baby entirely. (Ting and company had gotten out of the game; their accelerator had been barely capable of discovering the particle and not capable of examining it further.) But despite feverish effort, attempts to explain the surprising sharpness of J/psi were at first stymied.
Finally one speculation began to make sense. Maybe J/psi was the long-awaited bound "atom" of c and , the charm quark and its antiquark. In other words, perhaps it was a meson, that subclass of hadron consisting of quark and antiquark. Glashow, exulting, called J/psi "charmonium." As it turned out, this theory was correct, but it took another two years for the speculation to be verified. The reason for the difficulty is that when c and are combined, the intrinsic properties of charm are wiped out. What c brings, cancels. While all mesons consist of quark and antiquark, they don't have to consist of a quark with its own particular antiquark, as does charmonium. A pion, for example, is .
The search was on for "naked charm," a meson that was a charm quark tethered with, say, an antidown quark. The antidown quark wouldn't cancel the charm qualities of its partner, and charm would be exposed in all its naked glory, the next best thing to what is impossible: a free charm quark. Such a meson, a was found in 1976 at the Stanford e+ e− collider by a SLAC-Berkeley group led by Gerson Goldhaber. The meson was named D0 (D zero), and studies of D's were to occupy the electron machines for the next fifteen years. Today, mesons like cd, cs, and cd are grist for the Ph.D. mill. A complex spectroscopy of states enriches our understanding of quark properties.
Now the sharpness of J/psi was understood. Charm is a new quantum number, and the conservation laws of the strong force did not permit a c quark to change into a lower-mass quark. To do this, the weak and electromagnetic forces had to be invoked, and these are much slower to act—hence the long lifetime and narrow width.
The last holdouts against the idea of quarks gave up about this time. The quark idea had led to a far-out prediction, and the prediction had been verified. Probably even Gell-Mann began to give quarks elements of reality, although the confinement problem—there can be no such thing as a free quark—still differentiates quarks from other matter particles. With charm, the periodic table now was balanced again:
Quarks
up (u) charm (c)
down (d) strange (s)
Leptons
electron neutrino (νe) muon neutrino (νμ)
electron (e) muon (μ)
Now there were four quarks—that is, four flavors of quarks—and four leptons. We now spoke of two generations, arranged vertically in the above table. The u-d-νe-e is the first generation, and since the up and down quarks make protons and neutrons, the first generation dominates our present world. The second generation, c-s-νμ-μ is seen in the intense but fleeting heat of accelerator collisions. We can't ignore these particles, exotic as they may seem. Intrepid explorers that we are, we must struggle to figure out what role nature had planned for them.
I have not really given due attention to the theorists who anticipated and helped to establish the J/psi as charmonium. If SLAC was the experimental heart, Harvard was the theoretical brain. Glashow and his Bronx High School of Science classmate Steve Weinberg were aided by a gaggle of young whizzes; I'll mention only Helen Quinn because she was in the thick of the charmonium euphoria and is on my role-model team.
THE THIRD GENERATION
Let's pause and step away. It's always more difficult to describe recent events, especially when the describer is involved. There is not enough of the filter of time to be objective. But we'll give it a try anyway.
Now it was the 1970s, and thanks to the tremendous magnification of t
he new accelerators and the matching ingenious detectors, progress toward finding the a-tom was very rapid. Experimenters were going in all directions, learning about the various charmed objects, examining the forces from a more microscopic point of view, poking at the energy frontier addressing the outstanding problems of the minute. Then a brake on the pace of progress was applied as research funds became increasingly difficult to find. Vietnam, with its drain on the spirit and the treasury, as well as the oil shock and general malaise resulted in a turning away from basic research. This hurt our colleagues in "small science" even more. High-energy physicists are in part protected by the pooling of efforts and sharing of facilities in large laboratories.
Theorists, who work cheap (give them a pencil, some paper, and a faculty lounge), were thriving, stimulated by the cascade of data. We still saw the same pros: Lee, Yang, Feynman, Gell-Mann, Glashow, Weinberg, and Bjorken, but other names would soon appear: Martinus Veltman, Gerard 't Hooft, Abdus Salam, Jeffrey Goldstone, Peter Higgs, among others.
Let's just quickly touch on the experimental highlights, thereby unfairly favoring the "bold salients into the unknown" over the "slow steady advance of the frontier." In 1975, Martin Perl, almost singlehandedly and while dueling, d'Artagnan-like with his own colleague-collaborators, convinced them, and ultimately everyone, that a fifth lepton lurked in the SLAC data. Called tau (x), it, like its lighter cousins the electron and the muon, comes in two signs: τ+ and τ−.
A third generation was in the making. Since both the electron and the muon have neutrinos associated with them, it seemed natural to assume that a neutrino-sub-tau (ντ) existed.
Meanwhile, Lederman's group at Fermilab finally learned how to carry out the dimuon experiment correctly, and a new, vastly more effective organization of apparatus exploded open the mass domain from the J/psi peak at 3.1 all the way to pretty nearly 25 GeV, the limit allowed by Fermilab's 400 GeV energy. (Remember, we're talking about stationary targets here, so the effective energy is a fraction of the beam energy.) And there, at 9.4, 10.0, and 10.4 GeV sat three new bumps, as clear as the Tetons viewed on a brilliant day from Grand Targhee ski resort. The huge mass of data multiplied the world's collection of dimuons by a factor of 100. Christened the upsilon (it was the last Greek letter available, we thought), the new particle repeated the story of the J/psi, and the new thing that was conserved was the beauty quark—or, as some less artistic physicists call it, the bottom quark. The interpretation of the upsilon was that it was an "atom" made of a new b quark bound to an anti-b quark. The higher mass states were simply excited states of this new "atom." The excitement over this discovery nowhere near matched that of J/psi, but a third generation was indeed news and raised an obvious question: how many more? Also, why does nature insist on Xerox copies, one generation replicating the previous one?
Let me offer a brief description of the work that led to the upsilon. Our group of physicists from Columbia, Fermilab, and Stony Brook (Long Island) included some crackerjack young experimenters. We had constructed a state-of-the-art spectrometer with wire chambers, magnets, scintillation hodoscopes, more chambers, more magnets. Our data acquisition system was "dernier cri," based on electronics designed by genius engineer William Sippach. We had all worked in the same domain of Fermilab beams. We knew the problems. We knew one another.
John Yoh, Steve Herb, Walter Innes, and Charles Brown were four of the best postdocs I have seen. The important software was reaching the state of sophistication required for work at the frontier. Our problem was that we had to be sensitive to reactions that happened as rarely as only once in every hundred trillion collisions. Since we needed to record many of these rare dimuon events, we needed to harden the apparatus to a huge rate of irrelevant particles. Our team had developed a unique understanding of how to work in a high-radiation environment and still have survivable detectors. We had learned how to build in redundancy so that we could ruthlessly suppress false information no matter how cleverly nature tried to fool us.
Early in the learning process, we ran in the dielectron mode and obtained about twenty-five electron pairs above 4 GeV. Strangely, twelve of these were clustered around 6 GeV. A bump? We debated and decided to publish the possibility that there was a particle at 6 GeV. Six months later, after the data had increased to three hundred events, poof—no bump at 6 GeV. We had suggested the name "upsilon" for the fake bump, but when better data contradicted the earlier data, the incident became known as oops-leon.
Then came our new setup, with all of our experience invested in a rearrangement of target, shielding, placement of magnets, and chambers. We began taking data in May of 1977. The era of month-long runs of twenty-seven events or three hundred events were over; thousands of events per week were now coming in, essentially free of background. It isn't often in physics that a new instrument permits one to survey what amounts to a new domain. The first microscope and the first telescope are historic examples of far greater significance, but the excitement and joy when they were first used cannot have been much more intense than ours. After one week, a wide bump appeared near 9.5 GeV, and soon this enhancement became statistically solid. John Yoh had, in fact, seen a clustering near 9.5 GeV in our three-hundred-event run, but having been burned at 6 GeV, he merely labeled a bottle of Mumm's champagne "9.5" and hid it in our refrigerator.
In June we drank the champagne and broke the news (which had leaked anyway) to the laboratory. Steve Herb gave the talk to a packed and excited auditorium. This was Fermilab's first major discovery. Later that month we wrote up the discovery of a broad bump at 9.5 GeV with 770 events in the peak—statistically secure. Not that we didn't spend endless man-hours (unfortunately we had no women collaborators) looking for a malfunction of the detector that could simulate a bump. Dead regions of the detector? A software glitch? We ruthlessly tracked down dozens of possible errors. All of our built-in security measures—testing the validity of the data by asking questions to which we knew what the answers should be—checked out. By August, thanks to additional data and more sophisticated analysis, we had three narrow peaks, the upsilon family: upsilon, upsilon prime, and upsilon double prime. There was no way to account for these data on the basis of the known physics of 1977. Enter beauty (or bottom)!
There was little resistance to our conclusion that we were seeing a bound state of a new quark—call it the b quark—and its antiparticle twin. The J/psi was a cc meson. Upsilon was a bb meson. Since the mass of the upsilon bump was near 10 GeV, the b quark must have a mass near 5 GeV. This was the heaviest quark yet recorded, the c quark being near 1.5 GeV. Such "atoms" as cc and bb have a lowest-energy ground state and a variety of excited states. Our three peaks represented the ground state and two excited states.
One of the fun things about the upsilon was that we experimentalists could handle the equations of this curious atom, composed of a heavy quark circling a heavy antiquark. Good old Schrodinger's equation worked fine, and with only a brief look at our grad school notes, we raced the professional theorists to calculate the energy levels and other properties that we had measured. We had fun ... but they won.
Discoveries are always quasi-sexual experiences, and when John Yoh's "bicycle-on-line" quick analysis first indicated the existence of the bump, I experienced the now (for me) familiar feeling of intense euphoria, but tinged with the anxiety that "it can't really be true." The most obvious impulse is to communicate, to tell people. Who? Wives, best friends, children, in this case Director Bob Wilson, whose lab badly needed a discovery. We telephoned our colleagues at the DORIS machine in Germany and asked them to see if they could reach the energy required to make upsilons with their e+ e− collider. DORIS was the only other accelerator that had a chance at this energy. In a tour de force of machine magic, they succeeded. More joy! (And more than a little relief.) Later you think about rewards. Will this do it?
The discovery was made traumatic by a fire that interrupted data taking after a good week of running. In May 1977 a device that measures th
e current in our magnets, supplied no doubt by a low bidder, caught fire, and the fire spread to the wiring. An electrical fire creates chlorine gas, and when your friendly firemen charge in with hoses and spray water everywhere, they create an atmosphere of hydrochloric acid. The acid settles on all the transistor cards and slowly begins to eat them.
Electronic salvage is an art form. Friends at CERN had told me about a similar fire there, so I called to get advice. I was given the name and telephone numbers of a Dutch salvage expert working for a German firm and living in central Spain. The fire occurred on Saturday, and it was now 3 A.M. on Sunday. From my room at Fermilab, I called Spain and reached my man. Yes, he'd come. He'd get to Chicago Tuesday, and a cargo plane from Germany filled with special chemicals would arrive Wednesday. But he needed a U.S. visa, which usually takes ten days. I called the U.S. embassy in Madrid and spouted, "Atomic energy, national security, millions of dollars at stake..." I was connected to an assistant to the ambassador who was not impressed until I identified myself as a Columbia professor. "Columbia! Why didn't you say so? I'm class of fifty-six," he shouted. "Tell your fellow to ask for me."
On Tuesday, Mr. Jesse arrived and sniffed at 900 cards, each carrying about 50 transistors (1975 technology). On Wednesday the chemicals arrived. Customs gave us more heartburn, but the U.S. Department of Energy helped. By Thursday we had an assembly line: physicists, secretaries, wives, girlfriends, all dipping cards in secret solution A, then B, then drying with clean nitrogen gas, then brushing with camel's-hair brushes, then stacking. I half expected that we'd be required to accompany the ritual with a low moan of Dutch incantation, but this was not necessary.