Wideröe's paper; however; gave Lawrence an even better idea. Why not use a single gap with modest voltage but use it over and over again? Lawrence reasoned that when a charged particle moves in a magnetic field, its path is curved into a circle. The circle's radius is determined by the strength of the magnet (strong magnet, small radius) and the momentum of the charged particle (high momentum, large radius). Momentum is simply the particle's mass times its speed. What this means is that a strong magnet will guide a particle to move in a tiny circle, but if the particle gains energy and therefore also momentum, the radius of the circle will increase.
Picture a hatbox sandwiched between the north and south poles of a large magnet. Make the box out of brass or stainless steel, something strong but nonmagnetic. Pump the air out of the box. Inside it are two hollow D-shaped copper structures that almost fill the box: the straight sides of the D open and facing each other across a small gap, the round sides closed. Suppose one D is positively charged, the other negatively charged, with the difference of potential being, say, 1,000 volts. A stream of protons generated (never mind how) near the center of the circle is aimed across the gap from the positive D to the negative D. The protons gain 1,000 volts and their radius now increases since the momentum is higher. The protons sweep around inside the D, and when they return to the gap, thanks to clever switching, they again see a negative voltage across the gap. Again they are accelerated, and now they have 2,000 eV. The process continues. Every time the protons cross the gap, they gain 1000 eV. As they gain momentum they fight against the constricting power of the magnet, and the radius of their path continues to increase. The result is that the protons spiral out from the center of the box toward the perimeter. There they strike a target, a collision takes place, and the research begins.
The key to acceleration in the cyclotron is to make sure that the protons always see a negative D on the other side of the gap. The polarity has to flip-flop rapidly from D to D in exact synchronization with the rotation of the protons. But, you may be asking yourself, isn't it difficult to synchronize the alternating voltage with the protons, whose path continues to describe larger and larger circles as the acceleration continues? The answer is no. Lawrence discovered that by God's cleverness, the spiraling protons compensate for their longer path by speeding up. They complete each half circle in the same time, a process known as resonant acceleration. To match the proton orbits, one needs a fixed-frequency alternating voltage, a technology that was well known in radio broadcasting. Hence the name of the switching acceleration mechanism: radio-frequency generator. In this system the protons arrive at the edge of the gap just as the opposite D has maximum negative voltage.
Lawrence worked out the theory of the cyclotron in 1929 and 1930. Later he designed, on paper, a machine in which the protons made a hundred turns with a generation of 10,000 volts across the D-gap. That would give him a beam of 1 MeV protons (10,000 volts × 100 turns = 1 MeV). Such a beam would be "useful for the studies of atomic nuclei." The first model, actually constructed by Stanley Livingston, one of Lawrence's students, came up considerably short, reaching 80 KeV (80,000 volts). Lawrence then went big-time. He obtained a huge grant ($1,000!) to build a machine that could produce nuclear disintegrations. The pole pieces (the north and south pole pieces of the magnet) were ten inches in diameter and in 1932 the machine accelerated protons to an energy of 1.2 MeV. These were used to produce nuclear collisions in lithium and other elements only a few months after Cockcroft and Walton's group at Cambridge. Second place, but Lawrence still lit a cigar.
BIG SCIENCE AND THE CALIFORNIA MYSTIQUE
Lawrence was a mover and shaker of enormous energy and ability. He was the father of Big Science. The term refers to huge, centralized facilities of great complexity and expense that are shared by a large number of scientists. In its evolution, Big Science created new ways of carrying out research with teams of scientists. It also created exquisite sociological problems, about which more later. The likes of Lawrence had not been seen since Tycho Brahe, the Lord of Uraniborg, the laboratory on Hven. In the experimental arena, Lawrence established the United States as a serious player in world physics. He contributed to the California mystique, the love of technological extravaganzas, complex and expensive undertakings. These were alluring challenges for young California and, indeed, for the young United States.
By 1934 Lawrence was producing beams of 5 MeV deuterons with a thirty-seven-inch cyclotron. The deuteron, a nucleus consisting of one proton and one neutron, had been discovered in 1931, and had proved to be a more efficient projectile than the proton for producing nuclear reactions. In 1936 he had an 8 MeV deuteron beam. In 1939 a sixty-inch machine operated at 20 MeV. A monster started in 1940 and completed after the war had a magnet that weighed 10,000 tons! Cyclotrons were built all over the world because of their ability to unravel the mysteries of the nucleus. In medicine they were used to treat tumors. The beam of particles, directed at a tumor, deposits enough energy in the malignancy to destroy it. In the 1990s over a thousand cyclotrons are in use in hospitals across the United States. Basic research in particle physics, however, has abandoned the cyclotron in favor of a new type of machine.
THE SYNCHROTRON: AS MANY TURNS AS YOU WANT
The drive to create ever higher energies intensified and spread worldwide. At each new energy domain new discoveries were made. New puzzles were also created, increasing the desire to attain even higher energies. Nature's richness seemed to be hidden in the nuclear and subnuclear microworld.
The cyclotron is limited by its design. Because the particles spiral outward, the number of orbits is obviously limited by the circumference of the device. To get more orbits and more energy, you need a bigger cyclotron. The magnetic field must be applied to the entire spiral area, so the magnets must be large ... and expensive. Enter the synchrotron. If the particles' orbit, instead of spiraling out, could be kept to a constant radius, then the magnet would be needed only along the narrow path of the orbit. As the particles gained energy, the magnetic field could be increased synchronously to keep them imprisoned in an orbit of constant radius. Clever! Tons and tons of iron could be saved, because the magnetic pole pieces, transverse to the path of the beam, could be reduced to inches instead of feet.
Two important details must be mentioned before we proceed to the 1990s. In a cyclotron the charged particles (protons or deuterons) travel through what became thousands of turns in a vacuum chamber clamped between the poles of a magnet. To keep the particles from spreading out and striking the chamber walls, it was absolutely essential to have some kind of focusing process. Just as a lens focuses the light from a flashlight into a (nearly) parallel beam, magnetic force is used to squeeze the particles into a tight beam.
In the cyclotron this focusing action is provided by the way the magnetic field changes as the protons move toward the outer edge of the magnet. Robert R. Wilson, a young student of Lawrence's and later the builder of the Fermilab accelerator, was the first to understand the subtle but crucial effect the magnetic forces had in keeping the protons from spraying out. In the early synchrotrons, the pole pieces were shaped to provide these forces. Later, specially designed quadrupole magnets (with two north poles and two south poles) were used to focus the particles, while separate dipole magnets steered them in a fixed orbit.
Fermilab's Tevatron, a trillion-electron-volt machine completed in 1983, is a good example. The particles are steered into a circular orbit by powerful superconducting magnets, much as tracks guide a train around a turn. The highly evacuated beam pipe is a stainless steel (nonmagnetic) oval-shaped tube about 3 inches wide and 2 inches high, centered between the north and south poles of the magnets. Each dipole (steering) magnet is 21 feet long. The "quads" are 5 feet long. More than a thousand magnets are needed to cover the length of the tube. The beam pipe and magnet combination complete a circle that has a radius of 1 kilometer, or 0.6 miles—quite a change from Lawrence's first 4-inch model. You can see the advantage of the sync
hrotron design here. One needs a lot of magnets, but they're relatively skinny, just wide enough to cover the vacuum pipe. If the Tevatron were a cyclotron, we would need a magnet with pole pieces 1.2 miles in diameter to cover the 4-mile-around machine!
Particles make 50,000 orbits in one second around this 4-mile track. In 10 seconds the particles have traveled 2 million miles. Each time they pass a gap—actually a series of specially constructed cavities—a radio-frequency voltage kicks up the energy by about 1 MeV. The magnets that keep the particles focused allow them to deviate from their appointed rounds by less than one eighth of an inch over the entire trip. It's not perfect, but it's good enough. Like aiming a riñe at a mosquito sitting on the moon but hitting it in the wrong eye. To keep the protons in the same orbit while they are being accelerated, the strength of the magnets must increase in precise synchronism with the proton's gain in energy.
The second important detail has to do with the theory of relativity: protons become detectably heavier as their energy rises above 20 or so MeV. This increase in mass destroys the "cyclotron resonance" that Lawrence had discovered, in which the spiraling protons exactly compensate for their longer path by speeding up. This allows the rotation to be synchronized with a fixed frequency of the accelerating voltage across the gap. At higher energy the rotation time increases, and one can no longer apply a constant radio-frequency voltage. To counter the slowdown, the applied frequency has to decrease, so frequency-modulated (FM) accelerating voltages are used to track the increasing mass of the protons. The synchrocyclotron, a frequency-modulated cyclotron, was the earliest example of the influence of relativity on accelerators.
The proton synchrotron solves the problem in an even more elegant way. It is a little complicated but depends on the fact that the speed of the particle (99 point whatever percent of the speed of light) is essentially constant. Suppose the particle crosses the gap at that part of the radio-frequency cycle when the accelerating voltage is zero. No acceleration. We now increase the magnetic field a bit. The particle bends in a tighter circle and arrives a bit early at the gap, and now the radio frequency is in a phase to accelerate. Thus the mass grows, the orbit radius increases, and we are back to where we started but with higher energy. The system is self-correcting. If the particle gains too much energy (mass), its radius will increase and it will arrive later at the gap and see a decelerating voltage, which will correct the error. Raising the magnetic field has the effect of increasing the mass energy of our hero-particle. This method depends on "phase stability," which is discussed later in this chapter.
IKE AND THE PIONS
One early accelerator was near and dear to me—Columbia University's 400 MeV synchrocyclotron, built on an estate in Irvington-on-Hudson, New York, within commuting distance of Manhattan. The estate, named after the ancestral Scottish mountain Ben Nevis, was established in colonial times by Alexander Hamilton. Later it was owned by a branch of the Du Pont family and then by Columbia University. The Nevis cyclotron, built between 1947 and 1949, was one of the most productive particle accelerators in the world during its twenty-some years of operation (1950–1972). It also produced more than a hundred and fifty Ph.D.'s, about half of whom stayed in the field of high-energy physics and became professors at Berkeley, Stanford, Cal Tech, Princeton, and many other such fly-by-night institutions. The other half went everywhere: small teaching institutions, government labs, science administration, industrial research, investment banking ...
I was a graduate student when President (of Columbia) Dwight Eisenhower dedicated the new facility in June of 1950, in a small ceremony on the lawn of the lovely estate—magnificent trees, shrubbery, a few red brick outbuildings—sloping down to the stately Hudson River. After appropriate speechifying, Ike threw a switch and out of the loudspeakers came the amplified "cheeps" of a Geiger counter, indicating radiation. The cheeps were produced by a radioactive source I held near a particle counter because the machine had chosen that moment to crash. Ike never found out.
Why 400 MeV? The hot particle of 1950 was the pion, or pi meson, as it's also called. The pion had been predicted in 1936 by a Japanese theoretical physicist, Hideki Yukawa. It was thought to be the key to the strong force, which in those days was the big mystery. Today we think of the strong force in terms of gluons. But back then pions, which fly back and forth between the protons and neutrons to hold them together tightly in the nucleus, were the key, and we needed to make and study them. To produce pions in nuclear collisions, the particle coming in from the accelerator must have an energy greater than m(pion)c2, that is, greater than the pion's rest mass energy. Multiplying the pion's rest mass by the speed of light squared, we get 140 MeV, its rest mass energy. Since only a fraction of the collision energy goes into the production of new particles, we needed extra energy, and we settled on 400 MeV. The Nevis machine became a pion factory.
BEPPO'S LADIES
But wait. First a word on how we found out about pions in the first place. In the late 1940s, scientists at the University of Bristol in England noticed that when an alpha particle passes through a photographic emulsion coated on a glass plate, it "activates" the molecules in its path. After developing the film, you can see a track defined by grains of silver bromide. The track is easily discerned through a low-power microscope. The Bristol group sent batches of very thick emulsion up in balloons almost to the top of the atmosphere, where the intensity of cosmic rays is much higher than at sea level. This source of "naturally" occurring radiation far exceeded in energy Rutherford's puny 5 MeV alphas. It was in these emulsions exposed to cosmic rays in 1947 that the pion was first discovered by Cesare Lattes, a Brazilian, Giuseppe Occiallini, an Italian, and C. F. Powell, the resident professor in Bristol.
The most colorful of the above trio was Occiallini, known as Beppo to his friends. An amateur speleologist and compulsive practical joker Beppo was the driving force of the group. He trained a bevy of young women to do the painstaking work of studying the emulsions under a microscope. My thesis supervisor Gilberto Bernardini, a close friend of Beppo's, visited him one day in Bristol. Following directions given to him in unbroken English, a language he found very difficult, Bernardini quickly got lost. Finally he stumbled into a lab where several very proper English ladies were staring into microscopes and cursing in Italian argot that would be outlawed on the docks of Genoa. "Ecco!" cried Bernardini. "Dissa is Beppo's lab!"
What the tracks in those emulsions showed was a particle, the pion, entering at high speed, gradually slowing down (the density of the grains of silver bromide increases as the particle slows), and coming to rest. At the end of the track a new, energetic particle appears and races off. A pion is unstable, decaying within one hundredth of a microsecond into a muon (the new particle at the end of the track) and something else. The something else turned out to be a neutrino, which doesn't leave a track in the emulsion. The reaction is written π → μ + ν. That is, a pion (eventually) gives rise to a muon and a neutrino. Since the emulsion provides no time sequence information, it took careful analysis of the tracks on half a dozen of these- rare occurrences to understand what the particle was and how it decayed. The new particle had to be studied, but using cosmic rays yields only a handful of such events per year. As with nuclear disintegrations, accelerators with high enough energy were required.
At Berkeley, Lawrence's 184-inch cyclotron began to produce pions, as did the Nevis machine. Soon synchrocyclotrons in Rochester Liverpool, Pittsburgh, Chicago, Tokyo, Paris, and Dubna (near Moscow) were studying the pion in its strong interactions with neutrons and protons as well as the weak force in the pion's radioactive decay. Other machines at Cornell, Cal Tech, Berkeley, and the University of Illinois used electrons to produce pions, but the most successful machines were the proton synchrocyclotrons.
THE FIRST EXTERNAL BEAM: PLACE YOUR BETS!
So there I was in the summer of 1950 with a machine going through birth pains and me needing data so I could get a Ph.D. and earn a living. Pions were the n
ame of the game. Hit a piece of something—carbon, copper, anything containing nuclei—with the 400 MeV protons from the Nevis machine and you'd generate pions. Berkeley had hired Lattes, who showed the physicists how to expose and develop the very sensitive emulsions used so successfully in Bristol. They inserted a stack of emulsions into the beam vacuum tank and allowed the protons to hit a target near the stack. Remove the emulsions through an air lock, develop them (a week of effort), and then subject them to microscopic study (months!). All this effort had given the Berkeley team but a few dozen pion events. There had to be an easier way. The trouble was that the particle detectors had to be installed inside the machine, in the region of the strong accelerator magnet, to record the pions, and the only device that was practical was the stack of emulsions. In fact, Bernardini was planning an emulsion experiment on the Nevis machine similar to what the Berkeley folks had done. The large, elegant cloud chamber I had built for my Ph.D. project was a much better detector, but it would never fit between the poles of a magnet inside an accelerator. Nor would it survive as a particle detector in the intense radiation inside the accelerator. Between the cyclotron magnet and the experimental area was a ten-foot-thick concrete wall to confine the stray radiation.
A new postdoc, John Tinlot, had arrived at Columbia from Bruno Rossi's famed cosmic ray group at MIT. Tinlot was the quintessential physicist. In his late teens he had been a violinist of concert quality, but he had put his violin away after an agonizing decision to study physics. He was the first young Ph.D. I had ever worked with, and I learned enormously from him. Not only physics. John was a genetically infected horse player and gambler: long shots, blackjack, craps, roulette, poker—lots of poker. We played during experiments while the data were being collected. We played on vacation, on trains and airplanes. It was a moderately expensive way to learn physics, my losses being moderated by the other players—students, technicians, and security guards whom John would recruit. He had no pity.
The God Particle: If the Universe Is the Answer, What Is the Question? Page 28
