A Palette of Particles

Jeremy Bernstein

• Publisher: Belknap Press
• Year: 2013
• Print ISBN:
  9780674072510
• eBook ISBN:
  9780674073647
• Topic: Physical Sciences
• Subtopic: Science History

From molecules to stars, much of the cosmic canvas can be painted in brushstrokes of primary color: the protons, neutrons, and electrons we know so well. But for meticulous detail, we have to dip into exotic hues—leptons, mesons, hadrons, quarks. Bringing particle physics to life as few authors can, Jeremy Bernstein here unveils nature in all its subatomic splendor.

In this graceful account, Bernstein guides us through high-energy physics from the early twentieth century to the present, including such highlights as the newly discovered Higgs boson. Beginning with Ernest Rutherford's 1911 explanation of the nucleus, a model of atomic structure emerged that sufficed until the 1930s, when new particles began to be theorized and experimentally confirmed. In the postwar period, the subatomic world exploded in a blaze of unexpected findings leading to the theory of the quark, in all its strange and charmed variations. An eyewitness to developments at Harvard University and the Institute for Advanced Study in Princeton, Bernstein laces his story with piquant anecdotes of such luminaries as Wolfgang Pauli, Murray Gell-Mann, and Sheldon Glashow.

Surveying the dizzying landscape of contemporary physics, Bernstein remains optimistic about our ability to comprehend the secrets of the cosmos—even as its mysteries deepen. We now know that over eighty percent of the universe consists of matter we have never identified or detected. A Palette of Particles draws readers into the excitement of a field where the more we discover, the less we seem to know.

Appendix 1

Accelerators and Detectors

The purpose of this appendix is to bring together and amplify a theme that has recurred throughout this book. To probe deeper and deeper into the interior of matter, the tools used have become ever more sophisticated. I began working in this field a half century ago, and when I began this sort of evolution would have been disregarded as wildly speculative science fiction. I wrote my Ph.D. thesis in the mid-1950s. The largest part of the work that I had to do was numerical. This was done with a Marchant calculator, which was electromechanical and performed its arithmetic operations by moving gears. If you divided a number by zero, the machine kept grinding until the gears burned up. It took me literally months to carry out this work. When I had finished it I discovered that a fellow at MIT had done a rather similar thesis. He too had numerical work to do. But he had had the good sense to cultivate a group at MIT that was building the latest version of the Whirlwind computer, which was the state of the art. It used vacuum tubes and could process about 40,000 instructions a second. As I recall, it took about an afternoon to do his work. Of course, by present-day standards this machine was a dinosaur. Your laptop can probably process a hundred million instructions a second.

In experimental elementary particle physics, computers are used for much more than numerics. Perhaps the most important use is in pattern recognition. In my day at the Harvard Cyclotron there were a few women employed to scan tracks made on photos and sort out the interesting events. I suppose that there might have been substantially less than a hundred photos involved in any experiment. Later, as I will discuss, the Gargamelle bubble chamber at CERN, where an important discovery was made, produced more than 100,000 tracks that had to be scanned. This was done by computer.

In the body of the text I have avoided introducing the energy and mass units that are standard for this subject. I did this because these units are so far outside our daily experience. Instead I compared masses of particles without ever telling you what these masses are. But in this appendix I want to bite the bullet and introduce these units. I will begin with a unit we all know—the watt. The watt is a unit of power, energy per unit of time—per second, for example. When your meter is read it gives numbers in kilowatt-hours. This is the electrical energy you have purchased during the period between meter readings. The energy unit that is used in defining the watt is the joule, named after the nineteenth-century British physicist James Prescott Joule. The watt is named after James Watt, who lived a bit before Joule. One watt is one joule per second. A kilowatt-hour is 3.6 × 10⁶ joules.

Voltage is the measure of the work you have to do to move a unit of electric charge between two points in an electric field. Equivalently, it is the energy gained if you reverse this process. For our subject the important unit is the electron volt (eV). This is the energy gained if an electron moves across a voltage drop of one volt. It has a unit of energy—for example, the joule. But in terms of joules it is an absurdly small number. One electron volt equals 1.6/10,000,000,000,000,000,000 joules (1.6 × 10⁻¹⁹ joules). That is why I did not introduce it before. It makes sense only in the context of elementary particle or atomic physics, where things are naturally measured in these units. We will run into meV (which is equivalent to 10⁻³ eV), keV (10³ eV), MeV (10⁶ eV), GeV (10⁹ eV), and TeV (10¹² eV). (In my day a GeV was called a BeV, so the Bevatron was designed to accelerate protons to this kind of energy.)

It is very convenient to use Einstein’s E = mc² to convert masses into energies. Let me give an example so you will see why. In grams the mass of the electron is 9.109383 × 10⁻³¹ kilograms. The speed of light in meters per second is 299,792,458 m/s. You must square this and multiply by the mass in kilograms to get joules. Then you use the conversion of joules to electron volts to get the mass-energy in electron volts. If you do this, you will find that the mass-energy of the electron is 0.510998928 MeV. It is much easier to keep track of a half an MeV than the absurd number in grams. In the future I will simply, as is customary in our business, refer to mass-energy as the “mass.”

In this spirit the mass of the proton is 938.272046 MeV, while the mass of the neutron is 939.565379 MeV. You will notice that the neutron mass exceeds that of the proton by enough that the decay into an electron is allowed by the conservation of energy. Since the positive and negative muons are antiparticles of each other, they have the same mass, which is 105.6583715 MeV. The positive and negative pions are also antiparticles of each other, so they have the same mass, which is 139.57018 MeV. The neutral pion is its own antiparticle and has a mass of 134.9766 MeV. It is slightly less massive. The positive and negative K-mesons are also antiparticles of each other and have a mass of 493.677 MeV. The masses of the neutral kaons are tricky. As we have seen, the neutral kaons are not antiparticles of each other. The states that have definite mass are not the states that are created but the states that have evolved in time. There is a small mass difference between these states, and if we ignore it, then the mass of the neutral kaon is 497.614 MeV.

I do not intend to run through the whole list; the reader can refer to the tables in the text. But I want to give the mass of at least one nonmesonic strange particle, the Λ^o. Its mass is 1115.68 MeV. Finally I will give the masses of the W and Z mesons. The charged W particles are antiparticles of each other and have a mass of 80.399 GeV. (Note that we are now in the GeV range.) The mass of the Z is 91.1876 GeV. All the masses that I have given have an uncertainty attached to them, which I have not included.

It is not my intention to write anything like a complete history of particle accelerators. That would require another book. Instead I will make a few points that will give the general flavor. I would argue that the first particle accelerators were the cathode ray tubes used by J. J. Thomson to discover the electron. Remember that these were evacuated glass tubes with the negatively charged cathode at one end and the positively charged anode at the other. When the cathode is heated electrons are emitted, and these are accelerated by electric fields, so they strike the anode with an enhanced velocity. Recall that Thomson measured the change in temperature of the anode due to the absorption of this kinetic energy.

In Thomson’s experiments a potential difference of a few hundred volts was established between the cathode and the anode. By the very early 1930s it was possible to produce a few hundred kilovolts to use in the acceleration process. One of the devices was invented by an American engineer named Robert Van de Graff. This consisted of a metal sphere that could be charged up to millions of volts. It is not clear what physics Van de Graff had in mind. But it is very clear what physics the two British physicists John Cockcroft and Ernest Walton had in mind. They wanted to accelerate protons to an energy at which quantum mechanics predicted they would penetrate nuclei. There is an electrical energy barrier that repels the protons, but quantum mechanics tells us that it can be penetrated by tunneling through it. Cockcroft and Walton built up a potential difference of some 800 kilovolts, and in 1932 they produced protons that were energetic enough to penetrate the lithium nucleus, producing two alpha particles. All of these devices were linear accelerators. This was about to change.

In 1929 the American physicist Ernest Lawrence read a paper by the Norwegian engineer Rolf Wideröe. To say he “read” the paper is a bit of a euphemism, as Lawrence could not read German, but he got the gist by looking at a diagram. Wideröe was suggesting that accelerating could be achieved in stages by alternating the polarity of the electric fields. Lawrence recruited a young electronics expert who built a working model that in 1931 managed to accelerate mercury ions to an energy of 1 million electron volts. This was still a linear accelerator, but about this time Lawrence had an idea of genius that changed everything. Why not make the charged particles move in circles, which could be done by applying a magnetic field at right angles to the orbital plane? After a semicircle one can apply a suitable electric field to give the particles a little boost of speed. After the next semicircle the field has to be reversed to give it a new boost. Lawrence’s observation of genius was that the time it took for any circular orbit under these conditions was the same for all orbits. As the orbits got larger the particle speeded up in just such a way as to make the time the same. This vastly simplified the electronics.

The first successful model was built by one of Lawrence’s postdoctoral students, M. Stanley Livingston. The first one that Livingston built was about four and a half inches in diameter. He was able to use vacuum tube technology to supply the periodic changes in the electric field. It cost about $25 to build. It may have looked like a toy, but it produced protons of 80 keV. Lawrence kept improving and expanding his model, but the principles remained the same. This included the Harvard cyclotron. By the time I left in 1957 the energy had been boosted to about 165 MeV, too low to do any real pion physics, but it facilitated some precision nuclear physics experiments, and several generations of physicists got their degrees using it.

A 100 MeV proton moves with a speed that is somewhat less than half the speed of light. This means that the effects of the theory of relativity, which depend on the square of this number, can largely be ignored. However, once you are interested in protons that have 1 billion electron volts of kinetic energy, relativity can no longer be ignored. A particle that moves past you with a speed approaching the speed of light has an effective mass that is larger than the mass of the same particle when brought to rest. This changes everything when it comes to accelerator design. You can no longer use a magnetic field of a fixed strength to guide the particles nor a fixed-frequency electric field. These quantities must be synchronized with the increasing speeds to take into account the changes in mass. The new generation of accelerators that were designed with these effects in mind were appropriately called synchrotrons.

The first of these machines to come on line was the Cosmotron at Brookhaven, in the building where I used to play the trumpet when the machine was not running. It had a seventy-five-foot-diameter ring and reached its full proton beam energy of 3.3 GeV in January 1953. It ran until 1968. The next was the Bevatron in Berkeley, which went online in 1954 producing protons of 6.2 GeV—just enough energy to make antiprotons. It was decommissioned in 1970.

I do not intend to run through the whole list, although it is not that long. Each machine was very expensive—the Cosmotron cost $8 million in 1950s dollars—and they took years to build. The step to the next-level TeV accelerators took new technology. For example, the Tevatron at Fermilab, which went online in 1983 and was finally shut down for lack of funds in 2011, took advantage of superconducting magnets, which replaced the earlier iron magnets. It produced protons of 980 GeV.

An entirely new kind of accelerator was introduced in the 1980s. The prime example was the Large Electron Positron (LEP) Collider, which went online at CERN in 1989. Electrons and positrons were injected into a ring that was 27 kilometers in circumference. These had been preaccelerated in a linear accelerator. The two beams went in opposite directions around the ring. The beams were focused so that they collided in four places around the ring where detectors were in place. By the time the machine was shut down in 2000 it had produced colliding beams of some 209 GeV. Its successor, the Large Hadron Collider, collides protons and is expected to reach an individual beam energy of 7 TeV when it is at full strength. These machines are all marvels of technology and bear about as much resemblance to Lawrence’s first cyclotron as the drawings of Leonardo’s flying machines have to a jet airliner.

Along with the advance in accelerator technology there was a parallel advance in detector technology. It is interesting to recall Rutherford’s experiment in which the atomic nucleus was discovered. Alpha particles—helium nuclei—produced in the decay of radon were made to collide with thin metal foils. One usually reads that these were gold foils, but several other metals were used as well. After the alpha particle penetrated the foil it struck a zinc sulfide screen, causing a flash of light. Rutherford’s assistants, Geiger and Marsden, sat in a darkened room an hour or so before the experiment started, to accustom their eyes. They then looked at the flashes through a microscope. That is collecting data retail. In the 1930s, as I have mentioned in the text, Wilson cloud chambers were used to detect the tracks of charged particles produced in cosmic rays. This way both the positron and the muon were discovered. After the war photographic emulsions were used to observe the strange particles in cosmic rays. This was also collecting data retail. Once accelerators such as the Cosmotron began producing beams of these particles, a new method of detection was called for, and indeed, just at this time one was found. This was the invention of the bubble chamber by the physicist Donald Glaser.

To understand the workings of the bubble chamber, think of the physics of boiling. Say you put water in a kettle and turn on the stove. As you increase the heat, the bonds that hold the liquid together begin to give way and the water begins to vaporize. Small water vapor bubbles are formed. If you look at the forces that act on the bubbles, you can identify three. Inside the bubble there is a vapor pressure that acts to try to expand the bubble. Opposing this is the surface tension of the bubble and the pressure from the ambient environment. The temperature at which these forces equalize is called the boiling point. The bubbles grow in size and are visible as they rise to the surface. The boiling point depends on the environment in which the heating is taking place. If you have ever camped at high altitude, you will have observed how hard it is to boil an egg. There is less atmosphere pressing down on the water and the temperature of the boiling point is lowered. What Glaser made use of is the fact that some liquids can be superheated—they can be maintained at least temporarily at temperatures above their nominal boiling points. In the bubble chamber the pressure on the liquid is reduced just as the particles to be detected enter the chamber. Such a particle deposits its energy on a bubble, causing it to expand and become visible. What one sees is a track made by these particles. An early application was a liquid hydrogen bubble chamber at the Bevatron that was used to detect the antiproton.

The mother of all bubble chambers was the Gargamelle. (Gargamelle, invented by Rabelais, was the giantess who was the mother of Gargantua.) The Gargamelle, which went into operation at CERN in 1970, was filled with Freon, a relatively dense liquid. It was designed to detect neutrinos that were produced by the Proton Synchrotron (PS), which was the predecessor to the LEP. The PS accelerated protons to 25 GeV, producing massive numbers of pions, which produced neutrinos when they decayed. The great discovery made at the Gargamelle took place in 1973, when researchers produced evidence for what became known as the weak neutral current. This was prior to the actual discovery of the W and Z mesons, but it was essential for the standard model that these exist. The W mesons carry electric charge and mediate processes such as the decay of the neutron, where it decays into a positively charged proton. The Z meson mediates weak processes in which the charge is not changed. For example, a neutrino can collide with an electron, transferring its momentum to it. This type of event—the appearance of energetic electrons—is what the Gargamelle was looking for. After 83,000 events were analyzed there were 102 neutral current events discovered.

The Gargamelle represented the high-water mark of the bubble chambers. This technology had some disadvantages, including the fact that the superheated phase had to be ready at precisely the time of the particle collisions, which made it impossible to detect the reactions of very short-lived particles. Bubble chambers have been replaced by detectors such as the wire chamber, in which one or more electric wires are suspended in a gas. When an ionizing particle enters the chamber it produces a cascade of charged particles, which gather on the wire, making an electric current. The magnitude of this current is a measure of the energy of the particle being detected.

The LHC illustrates every modern detector type. The circular tunnel in which they are contained has a circumference of 17 miles. The tunnel at some points is more than 500 feet belowground. There are six detectors sited arou...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright
• Contents
• Introduction
• I. Primary Colors
• II. Secondary Colors
• III. Pastels
• L’Envoi
• Appendix 1: Accelerators and Detectors
• Appendix 2: Grand Unification
• Appendix 3: Neutrino Oscillations
• Acknowledgments
• Index