Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 98
5 Accelerators for Elementary-Particle Physics INTRODUCTION TO ACCELERATORS The Why and How of Accelerators Accelerators are the essential tools in most elementary-particle physics research. They provide the high-energy particles used in experiments; the costs of their construction and operation command the major portion of the support budget for particle physics; and a sizable fraction of the community of high-energy physicists is primarily concerned with accelerator technology. Particle physics has always been characterized by the fact that a part of this scientific community has devoted its professional energy and ingenuity to the continuing development of these tools of research. Accelerators thus exemplify imaginative ideas at the frontier of technical complexity and sophisti- cation. The spinofffrom accelerator research and development has had applications ranging from radar to controlled thermonuclear fusion and to high-intensity x rays for biological research. Figure 5.1 shows how an accelerator works. A bunch of electrically charged particles, either electrons or protons, passes through an electric field. The particles gain energy because they are accelerated by the electric field, hence the name accelerator. The energy gained by each particle is given by the voltage across the electric field. Thus an electron passing through a voltage of I volt gains an energy of I 98
OCR for page 99
ACCELERATORS FOR ELEMENTARY-PARTICLE PHYSICS 99 Bunch of Electrons 1 Negative High Voltage Plate Positive F1 igh Voltage Plate FIGURE 5.1 Accelerators work by exerting an electric force on a charged particle. In the example here a negative plate repels the bunch of electrons and a positive plate attracts them. The electrons thus gain energy in moving from the negative plate to the positive plate. By the time they reach the positive plate they are traveling so quickly that they pass through the hole in the plate and can be used for experiments. electron volt, abbreviated 1 eV. And an electron passing through I million volts gains an energy of I million electron volts, abbreviated 1 MeV. In scientific notation 1 MeV = 106 eV. (Since protons have the same electric charge as electrons, a proton passing through a million volts also gains an energy of ] MeV.) The highest-energy accelerator in the world is the Tevatron proton accelerator at Fermilab, which is designed to produce an energy of I TeV, which is 106 MeV or 10'2 eV. Accelerators are either linear or circular (Figure 5.21. In the linear accelerator the particle is propelled by strong electromagnetic fields to gain all of its energy in one pass through the machine. In the circular accelerator, the particles are magnetically constrained to circulate many times around a closed path or orbit, and the particle energy is increased on each successive orbit by an accelerating electric field. Until the 1960s, experiments in particle physics had been conducted using only stationary (fixed) targets. In this case, the beam of acceler- ated particles is extracted from the accelerator and directed at a fixed target that may consist of a gas, a liquid, or a solid. Usually the target material is the simplest element, hydrogen, whose nucleus is a single proton. A wide variety of proton-proton and electron-proton experi- ments have been performed that study the absorption or scattering of the beam particles in the target material, the production of new
OCR for page 100
100 ELEMENTARY-PARTICLE PHYSICS Accelerating Plates A I I I I I r ~ 1 1 1 1 1 1 ~ . 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Li near Accelera tor ;\~\ ~ Circular Accelerator ~~ Accelerating FIGURE 5.2 Very high energies cannot be obtained by using just one pair of plates, as in Figure 5.1. There are two ways to solve this problem. In the linear accelerator, many pairs of plates are lined up, and the particles being accelerated are given more and more energy as they pass through each pair of plates. In a circular accelerator, only one pair of plates is used, but the particles are made to travel in a circle, thus passing through that pair of plates again and again. Each time they pass through the pair of plates they are given more energy. secondary particles during the collision, and the transformation of the incident and target particles into new kinds of matter. Not only are the primary reactions of the accelerated particles on fixed targets studied, but also in many experimental situations the secondary particles (such as pions, muons, and K mesons) are them- selves selected and collimated to produce beams of projectiles that interact with other targets. As efforts were made to increase the energy in the primary interac- tion in fixed-target experiments, it was recognized that a large fraction of the energy of the incident particles was not available for the interaction itself but was rather retained as the energy of motion of the recoiling products of the collision. At relativistic energies (i.e., ener- gies that are large compared with the rest energy of the accelerated particles) the collision between a projectile particle and a similar particle at rest makes available for interaction only an amount of energy that is proportional to the square root of the energy of the projectile. That is, J E = V2mEparticle.
OCR for page 101
ACCELERATORS FOR EfEMENTARY-PARTICLE PHYSICS 101 E is the usable or center-of-mass energy, Ep~jC~e is the energy of the accelerated particle, and m is the mass (in energy units) of the target particle. Thus as the energy of the accelerated particle increases, more and more of it is wasted, since only E is usable. For example, if the energy of the incident particle is increased by a factor of 100, the energy available in the center of mass is increased by only a factor of 10. Eventually it becomes economically and technically impractical to continue to increase the usable energy in fixed-target accelerators. Hence for very high energies we have gone to a different and newer accelerator concept: the particle collider. Particle Colliders A simplified example of a particle collider is shown in Figure 5.3. In a circular machine, a bunch of electrons and a bunch of positrons circulate in opposite directions, the particle bunches being held in the machine by a magnetic guide field. (These machines are also called storage flings.) At two opposite places in the machine, the bunches , Interaction Point O~ _ Bunch of Positrons Bunch of Am_ Electrons ~ ~ I nteract ion Point \ FIGURE 5.3 This colliding-beam storage-ring accelerator has two bunches of particles moving in opposite directions. The bunches collide at the two interaction points. Even though the bunches collide, most of the particles in the bunch pass right through the other bunch; therefore the bunches continue to rotate again and again around the orbits.
OCR for page 102
102 ELEMENTARY-PARTICLE PHYSICS collide head on. The usable energy is now E _ where Epanjc~e is the energy of a particle in either bunch. Thus all the particle energy is usable. (This is the usual case, where both colliding particles have the same energy. If that is not the case, as in an electron-proton collider, then not all the particle energy is usable.) When the bunches come together, most of the particles in one bunch simply pass through the other bunch without actually colliding. Thus they continue to rotate around the storage ring. The bunches may rotate for hours or even days, making thousands or even millions of rotations per second. The particles are put into the storage ring by an auxiliary accelerator called an injector. In lower-energy storage rings the particles are usually injected with their full energy. In higher-energy storage rings, the particles are accelerated after injection to their full energy. The following combinations of particles are now used or will be used in colliders: e+- e~ P - P P - P e~- p e+- P electrons colliding with positrons protons colliding with protons protons colliding with antiprotons electrons colliding with protons positrons colliding with protons A critical property of colliders is called luminosity, which is a measure of the rate at which particle collisions occur. Since particle collisions are the essence of particle experiments, the more collisions per second, the more useful the collider. A quantity called the cross section, S. measures the relative probability of two particles colliding. In a collider the rate, R. of collisions per second is R = LS, where L is the collider luminosity. Since the cross section S has units of centimeters squared, the units of L are collisions centimeters2 second (This is abbreviated as cm~2 sol, the numerator's unit being omitted.) Existing colliders have luminosities in the range of 1029 to 1032 cm-2 s I. An alternative to storage rings for particle colliders is the use of colliding beams produced by linear accelerators (Figure 5.41. The colliding bunches of particles pass through each other just once. Much
OCR for page 103
ACCELERATORS FOR ELEMENTARY-PARTICLE PHYSICS 103 Interaction Point 1 ,,o-- - ~ ~ ~ Bunch of Bunch of Positrons Electrons FIGURE 5.4 In this sketch of a linear colliding-beam accelerator the two bunches collide only once. To make full use of that single collision the bunches have to be much denser than in a circular collider. denser bunches must be used to compensate for the absence of repeated collisions. One form of such a device is currently being con- structed that will accelerate in close succession bunches of electrons and positrons in a single linear accelerator. In this case, the charges of opposite sign are separated by magnets and then brought into a head-on collision in a single pass. Superconducting Magnets in Accelerators In circular fixed-target accelerators or in circular colliders, the particles are kept moving in a curved path by strong magnetic fields. Those fields are generated by electromagnets that fill most of the circumference of the ring. One of the practical limitations on the achievement of higher energies with circular proton machines has been the size of the ring and the cost of electric power to operate the magnets. The present largest accelerators have a four-mile circumfer- ence and consume many tens of megawatts of power. An innovation that has led to much higher available energy for circular proton accelerators and storage rings has been the develop- ment of superconducting magnets. Superconducting metals, such as a niobium-titanium (NbTi) alloy, have zero electrical resistance when cooled to liquid helium temperature. This is a temperature just a few degrees above absolute zero. Since the electrical resistance is zero, no power is consumed in operating electromagnets whose coils are made of a superconductor, although some power must be used for refriger- ation to keep the magnets cold. This is one advantage of super- conducting magnets. There is also a second advantage. Superconducting magnet coils can carry extremely high currents. These can give magnetic fields two to four times stronger than ordinary magnets. The circumference of a circular proton machine depends on the strength of the magnetic field for a fixed energy. Hence the use of superconducting magnets allows a smaller circumference to be used or, conversely, a higher energy can
OCR for page 104
104 ELEMENTARY-PARTICLE PHYSICS be achieved in the same circumference. This has been done during the last several years at Fermilab, where the change from ordinary to superconducting magnets has doubled the energy of the proton accel- erator. Progress in Accelerators and the Energy Frontier Over the last 50 years there has been a continuous development of new accelerator ideas and engineering achievements. It is remarkable that as each set of concepts appeared to reach a dead end, a new idea, a new technology, has evolved to continue to roll back the frontiers of energy and luminosity. This is most strikingly illustrated in Figure 5.5, which was first published over 20 years ago but is still a good repre- sentation of our progress on the energy frontier. Here we have only adjusted our definitions to represent particle colliders in terms of the equivalent energy of the particle striking a stationary target. ELEMENTARY-PARTICLE PHYSICS AND THE VARIETY OF ACCELERATORS In the last section we described how accelerators work. We now turn to the reasons for the variety of accelerators used in elementary- particle physics: fixed-target accelerators and particle colliders, proton accelerators and electron accelerators, low-energy accelerators and high-energy accelerators. This variety exists to serve the many dif- ferent purposes of elementary-particle physics experiments. We will outline these purposes and give some illustrations. Study of the Properties of Known Particles Often we know that a particle exists, but we know little about its properties. An example in present-day research is the B meson, which contains a b or bottom quark and has a mass of about 5 GeV. The B meson can decay in many different ways through the weak interaction, and we would like to know much more about these different modes of decay. The cleanest way to study those decay modes at present is to produce a single B meson and a single anti-B meson (B) in an electron-positron collision using an electron-positron collider. Since the total mass to be created is about 10 GeV, an electron-positron collider that has its maximum luminosity at about 10 GeV is best. Such a collider is the CESR facility at Cornell University. Lower-energy electron-positron colliders do not have enough energy to create the BB
OCR for page 105
ACCELERATORS FOR ELEMENTARY-PARTICLE PHYSICS 1000 TeV 100 TeV I O TeV I TeV co cr As 100 GeV m `,: 10 GeV o I GeV v 100 MeV 10 MeV I MeV 105 o // i// / Proton Storage Ring _; ( Equiv. Energy) / / Proton Synchrotron Weak Focusing \ AG / / Electron Synchrotron f Weok Focusing /f'/ ~ ~~ q Proton Li nac ~Sector- Focused - /~r i/~ /// PI ~ _% ~ — Betatron ~ ~0 into Electron Linac _ Synchrocyclotron /~ n \ E lectrostati c Generotor Rectifier Generotor 100 KeV I l l l l l l 1 930 1940 1950 1960 1970 1980 1990 FIGURE 5.5 The maximum energy achievable by an accelerator has increased exponentially with time over the last 50 years. This exponential increase has been maintained by a succession of new inventions in accelerator technology. The highest energies have been achieved by storage rings, the latest invention in accelerator technology. In this figure the energy of storage rings is denoted by the equivalent energy that a fixed-target accelerator would have to possess to give the same useful energy.
OCR for page 106
106 ELEMENTARY-PARTIC~E PHYSICS pair, while higher-energy colliders have less luminosity at the required energy. On the other hand, to measure the lifetime of the B meson rather than its decay modes, the meson should have high velocity. Then it is best to produce it at higher energy, and the PETRA and PEP electron- positron colliders have that higher energy. Thus the first measurements of the lifetime of the B meson were made by experiments at PEP. The recently discovered Z° particle is another example. The discovery of the Z° was made at the CERN proton-antiproton collider because that was the only existing collider or accelerator with enough energy to create the 93-GeV mass of this particle. But electron-positron colli- sions should provide the cleanest and easiest way to create Z° particles in great numbers so that their properties can be studied in great detail. Indeed, studying the physics of the Z° is the first purpose of two electron-positron colliders now under construction, the Stanford Lin- ear Collider (SLC) and LEP at CERN (see the section below on Accelerators We Are Using or Building). Existing electron-positron colliders do not have enough energy to create Z° particles. The study of the decays of K mesons provides another example. The puzzling phenomenon of CP violation is observed only in such decays. To study these decays in detail we need a large number of K mesons, which are best produced in fixed-target proton accelerators. Thus the Tevatron, the Alternating Gradient Synchrotron (AGS), and the SPS machines are all used to produce K beams for various studies of K-meson decays. Study of the Known Forces Three of the four known forces, the electromagnetic force, the weak force, and the strong force, can be studied using accelerators. But the most suitable accelerator depends on the force to be studied and how it is to be studied. An old but still interesting example is the discovery that the total cross section (that is, the total rate) for the interaction of protons with protons through the strong force increases as the energy increases. The increase is not large, but it is a clear increase. This is called the rising total cross-section effect, and we do not understand why it occurs. To make progress on this problem we need more data on proton-proton interactions at yet higher energy. These data can only come from a higher-energy proton-proton collider. Further studies of the weak force at higher energy require a different facility. The weak interaction can only be studied in a collision if the
OCR for page 107
ACCELERATORS FOR EfEMENTARY-PARTICLE PHYSICS 107 strong force is not present; otherwise the strong force masks the weak force. Therefore one of the particles in the collision must be a lepton, because leptons do not feel the strong force. The classic way to study the weak interaction has been to collide neutrinos with protons or with neutrons in a fixed-target experiment. The neutrinos must come from a secondary neutrino beam produced at a proton accelerator. However, as we discussed in the last section, fixed-target experi- ments are more limited in their maximum energy than are collider experiments. Thus the highest-energy weak-force studies will have to be done using an electron-proton collider. No such collider exists, but the knowledge and technology needed to build such a facility do exist. The DESY laboratory in Germany is now building such a collider, called HERA. Tests of New Ideas and Theories It is rare that a new idea or theory can be tested with experimental data that already exist. More commonly it is necessary to carry out new experiments to test the new ideas or theory. Such experimental tests often stretch the capabilities of the accelerator being used. For example, the principle of lepton conservation states that the decay muon ~ electron + photon cannot occur. This decay has been looked for but has not been found to a precision of about 1 part in 10"'. To test some theories that say that this decay should in fact occur at a level of I part in 10~2 the experimenter needs a great number of muons. The best source for such muons is the secondary muon beam from a high-intensity proton accelerator. High intensity, not high energy, is important. Therefore experimenters use a relatively low-energy but high-intensity proton accelerator such as the 800-MeV LAMPF machine at Los Alamos. Other tests of new ideas and theories require high energies. For example, in Chapter 4 the technicolor theory was mentioned; this theory predicts new particles in the mass range of I TeV. No existing collider can produce particles with such a large mass. Therefore a higher-energy proton-proton or proton-antiproton collider is needed. The proposed Superconducting Super Collider (SSC), discussed below in the section on The Superconducting Super Collider, A Very-High- Energy Proton-Proton Collider, would have sufficient energy to pro- duce these massive new particles.
OCR for page 108
108 ELEMENTAR Y-PARTICLE PHYSICS The Search for New Particles and the Mass Scale The need for higher-energy colliders to search for new particles is so fundamental to our goals that we will discuss this in more detail. There are two questions involved in the search for a new particle. How much energy is needed? How much intensity or luminosity is needed? The answer to the energy question depends on the type of collider used to produce the particle. In electron-positron colliders, when the electron and positron annihilate, they can give all their energy to the production of the particle. If a single particle is to be produced, then the total energy of the collider need only be equal to the mass of the single particle. Proton-proton colliders or proton-antiproton colliders require more total energy than the mass of the particle that is to be produced. This is because the production process actually occurs through the collision of a single quark or gluon in one proton with a single quark or gluon in the other proton (or antiproton). On the average a single quark or gluon in a proton only carries about 1/6 of the total energy. Therefore the total collision energy needed to produce a particle of a certain mass is about 6 times that mass. This is a rough rule, because the second question how much luminosity is required is also important. If the production process for ~ new particle is rare, then a high luminosity is required. The range of masses that can be produced at a collider should overlap the mass range or mass scale of the theory that is to be tested. To achieve this mass range both high energy and high luminosity are necessary. To reach the mass scales of the theories discussed in Chapter 4, colliders should have the following general properties: 10-TeV minimum total energy proton-proton or proton-antiproton electron-positron 1032 cm~2 so minimum luminosity 1-TeV minimum total energy 1032 cm~2 s~ ~ minimum luminosity As will be described in the section below on The Superconducting
OCR for page 121
ACCELERATORS FOR ELEMENTARY-PARTICLE PHYSICS 121 range of several TeV: · What is the origin of mass, and what sets the masses of the different elementary particles? Is the Higgs hypothesis correct, and can the Higgs particles be found? If the Higgs hypothesis is wrong, what replaces it? · Are there more quark or lepton generations? Why do these particles form generations? · Are the quarks and leptons truly elementary? · Are new theoretical ideas like technicolor or supersymmetry correct? Can the strong and electroweak interactions be unified? -- Are there undiscovered fundamental forces? The mass range needed to study these problems is illustrated in Figure 5.12. This mass range cannot be reached with fixed-target accelerators; it requires a hadron-hadron collider. As mentioned ear- lier, when hadrons collide, the full energy of the hadrons is not available for conversion into mass, even in a colliding-beam accelera- tor. This is because the hadron-hadron collision really consists of a quark-quark, quark-gluon, or gluon-gluon collision; and these constit- uents only carry a fraction of the total energy of the hadron. The rough rule is that 1/6 of the total energy is available, on the average, for conversion into large masses. We emphasize that this is an average. There is a large probability that 1/10 to 1/20 of the energy can be converted into large masses and a small probability that 1/3 can be converted. Collider Goals These physics goals, searching for answers to fundamental questions and exploring new physics in the several-Ted mass range, require a hadron-hadron collider of very high energy and large luminosity. Our knowledge and experience in accelerator technology enables us to set the practical goals for the collider of a maximum energy of 40 TeV and a maximum luminosity of 1033 cm~2 S-~. To achieve this luminosity a proton-proton collider is favored. The richness and range of the particle physics that can be done at such a facility dictates that there be multiple interaction regions for particle detectors. Six or more inter- action regions are desirable. Summarizing, the practical goals for the Superconducting Super Collider are as follows: Maximum total energy Maximum luminosity Number of interaction regions 40 TeV 1033 Cm-2 S-] 6 or more
OCR for page 122
22 ELEMENTAR Y-PARTICLE PHYSICS 4 3 - J ~ 2 In _ Cl) In ~ L 1 o Domain I of SSC~ Posi ble t Quark Lepton I nterna I 0 Structure Domain of | Tevatron \; I* `lWz' ~ I In - TOP ? New Quarks and Leptons Other Ws ~ Z's Technicolor Higgs Super Boson Symmetry Pointlike Weak Breakdown ) ~ FIGURE S. 12 The mass scale at which physicists believe that a number of fundamental new phenomena may appear. The SSC would extend this scale beyond the mass of a few tenths of a TeV that can be probed by facilities now under construction to the regime of 2 to 3 TeV and above. Design Studies Since 1982 the U.S. elementary-particle physics community has been developing a plan for the construction of a high-luminosity proton-proton collider in the energy range of 40 TeV. The work began in the summer of 1982 at a meeting in Snowmass, Colorado (see Proceedings of the 1982 Division of Plasma and Fluids Summer Study on Elementary Particle Physics and Future Facilities, June 28-July 16, 1982, Snowmass, Colorado, R. Donaldson, R. Gustafson, and F. Paige,
OCR for page 123
ACCELERATORS FOR El EMENTARY-PARTICLE PHYSICS 123 eds.~. The result of this and other studies was that a recommenda- tion for the construction of such a facility was made to the U.S. Department of Energy by the 1983 High Energy Physics Advisory Panel of the Department of Energy (HEPAP) Subpanel on New Facilities. This collider has been named the Superconducting Super Collider (SSC) because it requires the use of superconducting magnets to keep its size and its operating power costs within reasonable bounds. Although the basic technology for the machine is at hand, the scale is unprecedented. Therefore an intensive series of design studies has been carried out. In April 1983 an informal one-week workshop (see Report of the 20 TeV Hadron Collider Technical workshop, Newman Laboratory, Cornell University, Ithaca, New York) was held at Cornell to study the design problems and to make initial estimates of feasibility, time scale, and costs. This was followed by meetings and workshops on hadron collider detectors, on the physics that can be done at the SSC, on accelerator issues related to the SSC, and on cryogenic issues related to superconducting magnets for accelerators. During this period a subpanel of HEPAP was set up to provide advice on the content and implementation of a preliminary research and development (R&D) effort. The most intensive design work at present is the National SSC Reference Designs Study (see SSC Reference Designs Study Group Report, May 1984), which was conducted from February through May 1984. This study addressed three areas: · Technical feasibility: the designs of 40-TeV total-energy proton- proton colliders were explored using three of several possible superconducting magnet styles as study models. · Economic feasibility: the likely cost range was estimated using preliminary engineering designs for the three magnet styles and the other hardware and conventional facilities required to construct and operate technically feasible colliders. · Required R&D: the R&D needed to verify design calculations and technical assumptions was identified. It was not intended' however, that the Reference Designs Study Group Report be either a design proposal or a site preference study. Some of the material in this section is based on this study. Superconducting Magnets The feasibility of constructing the SSC has been substantially enhanced by the recent success in accelerating protons to high energy
OCR for page 124
124 ELEMENTARY-PARTICLE PHYSICS in a superconducting accelerator, the Tevatron at Fermilab. This machine uses about 1000 superconducting magnets in a circumference of about 4 miles. It now operates at 800 GeV for physics experiments, and it has also operated in the beam-storage mode preliminary to its use as a proton-antiproton collider. Although the full operating energy of 1000 GeV and full beam intensity are yet to be attained, the perform- ance of the Fermilab machine is a definitive verification of the practicality of using superconducting technology for obtaining beams of very high energy. The Tevatron has opened the door to a new era. In the course of its construction a great deal has been learned. Great strides have been made in the development of superconducting, niobium-titanium cables, so that high-quality materials are now avail- able in large quantities at reasonable cost. The technique for constrain- ing the superconducting cable in the magnet with the required high precision has been well demonstrated, and protection systems have been developed to cope with the inevitable magnet quenches. (A superconducting magnet quenches when a part of its coil becomes too warm to maintain zero electrical resistance.) During the same period, the industrial capability for producing refrigeration equipment has grown rapidly, and large machines with much higher reliability are now available. The ability to transport large volumes of helium liquid over long distances has been demonstrated. Automatic control over the refrigeration and cryogenic systems has been remarkably successful, and the capability for beam location and control has been demon- strated. Preliminary Collider Designs and Considerations The design studies, particularly the National SSC Reference Designs Study, have shown that a conservative extension of existing or near-term technology can lead to the successful achievement of an SSC. Several design options exist' and the selection of a particular design to optimize the cost is one of the most important considerations. The final cost will depend on the results of the R&D program that will be carried out before initiating construction. One of the principal factors determining the detailed design of the collider is the strength of the magnetic guide field. The options cover a broad range of magnetic- field values. The Reference Designs Study has considered the three superconducting, niobium-titanium magnet designs (a), (b), and (c) listed next. Other work has considered the design (d). As shown in Figure S.13(a), the diameter of the collider decreases as the magnetic field increases.
OCR for page 125
A CCELERA TORS FOR E! EMENTAR Y-PARTIC~E PH YSICS 125 (b) / , 1 Col l l der and I n ~ actors , ~J, to Scole (a) Lo ~ 20 at 10 o Nigh Ene: Enlorgement ~ Booster tog/ of \> 1 TeV By In jectors }/ - - Totol Energy - 40 TeV 0 2 4 6 8 10 . Ll noc MAGNETIC FIELD (Teslo) O to 0.001 TeV ~ Low Energy Booster to 0.07 TeV - Interoc:\ colons —~D: FIGURE 5.13 (a) The diameter of a proton-proton or proton-antiproton collider depends on the total energy desired and the magnetic field used. (b) Schematic layout of the SSC indicating the injector complex and the main ring, where protons are accelerated to 20 TeV in counterrotating bunches that collide at six points around the circumference. The total collision energy is 40 TeV. (a) A high-field magnet design has a 6.5-tesla field, with both beam tubes and both coil sets side by side in a common iron yoke contained in a single cryostat. This approach is referred to as the 2-in-1 design. Intrinsic to this approach is magnetic coupling, limiting the extent to which the field strengths in the two apertures may differ. This results in an SSC main ring about 18 miles in diameter. (b) A medium-field dipole magnet has a 5-tesla field, with each beam tube and coil in its own cryostat. Each cryostat has only enough iron to shield one coil from the magnetic field of the other. This is referred to as the 1-in-1 no-iron design. This results in an SSC main ring about 22 miles in diameter. (c) A low-field magnet has a 3-tesla field, with each beam tube and each coil set in separate iron yokes, one above the other, in a single cryostat. In this design, although the iron is driven well into saturation, the field is determined primarily by the iron pole faces, and the magnetic fields of the two rings are not strongly coupled. This magnet is referred to as the superferric design. This results in an SSC main ring about 32 miles in diameter.
OCR for page 126
126 ELEMENTARY-PARTICLE PHYSICS (d) The designs listed above use a niobium-titanium superconductor, with which we have a great deal of experience. Very high magnetic fields, 8 teslas or more, can be achieved with a niobium-tin supercon- ductor, but there is little experience at present with such magnets. a The choice among these systems is complex. The medium-field technology and to some extent the high-field technology have already been proven in the Tevatron and in the Brookhaven CBA design. Although such magnets could simply be copied and manufactured in quantity, without cost-saving design and production changes the over- all cost of the installation would be great. The accelerator tunnel in this case would have a moderate length. The low-field design is expected to be reliable because of the low values of the forces and low field strengths in the superconductor. It also has the advantage that since the iron profile largely determines the field accuracy, it should be less sensitive to the placement of conduc- tors. It has the disadvantage of requiring a larger tunnel perimeter. The use of very high field magnets would minimize the tunnel length. However, suitable superconducting cable has not yet been produced in quantity, and the appropriate technology has not yet been developed. This type of magnet might therefore require much more R&D than lower-field designs. No matter what magnetic-field strength is chosen, the cost of the collider can be reduced if magnets with a smaller aperture can be developed and used. This requires R&D both in magnet design and in the accelerator physics of the collider. Finally, advantage must be taken of the increased scale of production. New fabrication methods suited to mass production will have to be developed. The SSC facility is shown schematically in Figure 5.13(b). The collider itself sets the size of the site. The injector complex would lie against one portion of the collider ring. The six interaction regions would be distributed around the ring. An important question is the site required for such a machine. A large number of factors must be taken into account in the site selection. These include ring diameter; environmental considerations; availability of water' power, and roads; and proximity to airports, villages, and cities. Stating first with the technical considerations, it is clear that the number of suitable sites will be strongly dependent on the radius of the machine. From the point of view of beam dynamics, gentle deviations from flatness of the ring might be tolerated. One might be able to take advantage of this in order to locate the interaction halls and service buildings near the surface; and it may permit the use of contour- following, cut-and-cover techniques for the machine closure instead of
OCR for page 127
ACCELERATORS FOR ELEMENTAR Y-PARTICLE PHYSICS 127 the more expensive mode of tunneling. A cursory search for suitable sites has suggested that several can be found that would be suitable for even the largest of the rings. one of 100-mile perimeter or more. Schedule and Cost Research and development will be needed before beginning the construction of the collider. About 2 years will be required before a working design can be established. It will be necessary to learn how to mass produce low-cost' high-quality magnets and how to handle. mount' and survey the magnets into position with a high degree of precision. It is likely that a full-scale prototype of a relatively long tunnel section and guide field will be constructed in order to test the practicality and integration of the system. It may even be necessary to work on the design of more than one of these systems in parallel in order to determine the minimum-cost system. Such R&D activity is essential to carry out the design. The scale of this project far exceeds any of our existing high-energy physics facilities. It is obvious that an administrative organization will be required that is responsible to a broadly based national representa- tion of the elementary-particle physics community. The federal funding agencies must indicate that they are receptive to a proposal to build such a machine. International cooperation with respect to building some of the detectors or other costs should be explored. The Reference Designs Study has considered the construction schedule, as follows: .'In this study, we have assumed a six-year construction period, which would lead to completion in early 1994 if construction were to begin in PY 1988. The optimum duration of the construction period should itself be an object of study. . . It will depend on many factors, such as the detailed scope of the facility that is ultimately proposed, the technical means devised for its construction. and the spending pattern needed. Finding ways for minimizing the delay between start of construction and first use for physics research must be given great emphasis." The same study has estimated the construction costs of the SSC. These costs, based on the three magnet technologies (a), (b), and (c) listed above, range from $2.70 billion to $3.05 billion in fiscal year 1984 dollars. (The costs of research equipment, preconstruction R&D, and possible site acquisition are not included.) Quoting the study, ''The contingencies are intended to be sufficiently conservative that these totals represent our best estimate today for an upper bound on the SSC cost. With intense R&D and effective planning, lower costs could re- sult.
OCR for page 128
128 ELEMENTARY-PARTICLE PHYSICS RESEARCH AND DEVELOPMENT FOR VERY-HIGH-ENERGY LINEAR COLLIDERS Physics Motivation In the section above on Elementary-Particle Physics and the Variety of Accelerators we saw that hadron-hadron and electron-positron colliders largely complement each other in the physics that they explore. As mentioned earlier, there is a rule of thumb that an electron-positron collision has the same available energy as a proton- proton collision when the actual energy of the electron plus positron is about 1/6 of the actual energy of the two protons. Thus to reach the same available energy as the planned 40-TeV proton-proton collider, an electron-positron collider would require a total energy in the several-Ted range. This is beyond the reach of the known technology of circular electron-positron colliders; thus a new electron-positron collider technology such as the linear collider is needed. incidentally, although most of the thought and work on linear colliders is for electron-positron machines, the concept may also be applicable to electron-proton colliders. Present Technology and Concepts As described above in the section on Accelerators We Are Using and Building, the first application of linear collider principles is now being made in the construction at SLAC of the Stanford Linear Collider, a facility with a maximum total energy of 100 to 140 GeV. Starting from this machine, we now consider what R&D is needed in order to build a much larger TeV machine. In linear accelerators and colliders, the critical parameter is the accelerating gradient, i.e., the energy gained per meter of length. In the SLC, it will be about 20 GeV per kilometer. A 2-TeV collider based on the present SLAC accelerating structure would consist of two conventional linear accelerators each 50 kilome- ters in length. With 12 electron-positron bunches per pulse, there could be a magnetic switchyard that would feed the bunches to 6 parallel interaction regions, each with a luminosity of the order of 1032 cm~2 s~'. Using the electrical efficiency of today's pulsed radio-frequency power sources, the total power consumed would be approximately 300 MW. These numbers are quite large. It is desirable to reduce the length and hence the construction cost of such a machine, and also its power consumption. Research and development work aimed toward these
OCR for page 129
ACCELERATORS FOR ELEMENTARY-PARTICLE PHYSICS 129 goals is now beginning. Of course experience with the operation of SLC will also stimulate progress toward these goals. One of the directions for improving the technology of linear colliders is to reduce the wave length (increase the operating frequency) of the accelerator structure. A reduction to 5 cm (SLAC uses 10 cm) doubles the accelerating gradient and doubles the electrical efficiency. Re- search and development is required to produce high-power klystrons at this higher frequency, and several ideas exist for ultra-relativist~c or laser-driven klystrons that could provide not only the requisite power but also much higher electrical efficiency. Alternative accelerator structures that promise much higher accelerating gradients will also be explored. (At these shorter wavelengths there is increased energy spread in the accelerated beam, and further development in chromatic corrections of the final focusing systems is required to handle this energy spread.) The repetitive nature of linear accelerators naturally suggests auto- mated production techniques to reduce construction costs. Also, ener- gy-recovery schemes, perhaps using superconducting microwave ac- celerator units, need to be explored to increase overall electrical efficiency further. As these technologies advance, the design of a linear collider facility can be optimized, and the construction and operating costs can be reduced. RESEARCH ON ADVANCED CONCEPTS FOR ACCELERATORS AND COLLIDERS To conclude this chapter we discuss some advanced ideas for accelerators and colliders. We do not know if any of these ideas can be reduced to practice. But if we are to move substantially beyond the energy range of present accelerator technologies, we must find new ways to accelerate particles. This section describes some of the ideas now being explored. Linear Accelerators and Colliders Calculations and research are being carried out in the United States and abroad on a variety of new and advanced concepts for obtaining higher accelerating gradients, which is the energy gain per unit of accelerator length. There is reason to believe that accelerator struc- tures can be built to handle up to 200 GeV per kilometer, ten times the currently available gradients. What is needed is a suitable high- efficiency, high-power source of short-wavelength electromagnetic
OCR for page 130
130 ELEMENTARY-PARTICLE PHYSICS . radiation that can provide a relatively large amount of energy per unit length. We list some of the possibilities: 1. Very-high-power, very-short-pulse-length, high-frequency klys- trons suitable for this purpose may be developed. 2. A special case of a source of short-wavelength radiation is the wake field of a high-energy beam passing through a cavity system. This idea is being pursued theoretically and shows considerable promise and a special simplicity since the wake-field source cavity can be combined with a beam-accelerating cavity within a single structure. 3. In the two-beam accelerator concept, a high-power, low-energy electron beam travels parallel to the desired high-energy particle beam. Using a principle such as that of the free-electron laser, the high- power, low-energy beam radiates its power to the high-energy beam, thus providing the acceleration. 4. A more radical approach is to use the very short wavelength obtainable from a laser. In this case one cannot consider accelerating structures of conventional design; the dimensions are far too small. It appears possible, however, to use a suitable optical grating in place of a conventional cavity. The most extreme case would be obtained if the periodic grating were replaced by a periodic plasma, possibly formed over a grating surface. In this case gradients as high as I TeV per kilometer could theoretically be attained. Such high and obviously desirable gradients can only exist in or near a plasma and not in or near any solid conductor or dielectric. S. A particularly interesting solution occurs when a plasma is exposed to two laser beams of suitably close frequency. The beat frequency between the two lasers can be matched to the natural plasma frequency, and a strong periodic and moving charge modulation can be induced. Large electrostatic fields are generated by this modulation, and these could be used to accelerate suitably injected beams. Accel- erating fields as high as 2 TeV per kilometer have been discussed, but there remains great uncertainty about the stability, energy efficiency, and suitability of such a mechanism to the construction of a high- energy linear collider. Many such ideas have been suggested. Some of them may not work. Others may work but not have application for high-energy physics. It is clear, however, that without some such idea, no great further step in energy will be possible. On the other hand, with gradients of the order of I TeV per kilometer theoretically possible, an accelerator of 100 TeV is not unthinkable. It is thus important to the future of the field that these ideas are followed up.
OCR for page 131
ACCELERATORS FOR ELEMENTARY-PARTICLE PHYSICS 131 Ultrahigh-Energy Circular Colliders We have a great deal of knowledge and experience with the technology to be used to build a 40-TeV proton-proton circular collider. The primary limitation of that technology is that we do not know how to increase substantially the magnetic field that guides the particles in a circle. and hence we do not know how to decrease sub- stantially the circumference of the collider. Some size and cost reduction can be obtained in guide-field magnets by the use of new superconducting materials such as niobium-tin. While such develop- ments are important, they do not promise a radical saving or access to much higher energies. Mechanical forces will limit the usable magnetic fields no matter what conductors become available. Even if we could substantially decrease the circumference of a proton-proton collider, we would then reach a second limitation: the protons would begin to lose large amounts of energy via synchroton radiation, as occurs at much lower energies in circular electron-pos- itron colliders. Indeed, no ideas have yet been proposed to enable an increase of the energy of a circular collider beyond the 100-TeV range. The Need for Advanced Research on Accelerators and Colliders Thus new accelerator ideas need to be developed and explored. In the past, new ideas have indeed occurred, resulting in the enormous increases in accelerator energy that have been achieved in the past 50 years. However, the present scale of R&D in accelerator technology is small and certainly not commensurate with its importance. Part of the problem is the reluctance of individuals to commit themselves to tasks whose possible fruition seems quite distant. Another problem is the lack of suitably trained multidisciplinary experts. A third may be traced to the mechanisms for supporting accelerator physics. Encouragement to universities to expand training in accelerator physics is needed. Possibly, too, it would be desirable to have a funding mechanism that would allow laboratories to pursue such work with an assurance that such funding was truly an addition to that for more immediate goals. There is a strong and natural tendency for internal priorities to cut back on such long-range activities. Despite these reservations, it is encouraging to note that there are still many people working on new ideas and that advanced accelerator workshops and schools take place regularly. We can hope and expect to see significant new activity in the coming decade.
Representative terms from entire chapter: