To a particle physicist the absence of anti-matter is at first sight surprising; the fundamental constituents of matter come in pairs, for every kind of particle there is an antiparticle that is identical in mass but differs in other characteristics such as electric charge. This symmetry has been well verified by experiment. Since the discovery of the positron the catalog of antiparticles has grown in tandem with the catalog of particles. High energy collisions at particle accelerators yield matter and anti-matter in equal quantities. The early universe was filled with a great abundance of elementary particles of every kind, its evolution governed by the forces between them. Consequently, particle physics holds profound implications for cosmology and vice versa. If the evidence in the laboratory points to a balance between matter and anti-matter, naively one would expect the universe to reflect this.

The first clue that particle physics might be able to explain the imbalance, came in 1964 when it was observed that the weak interactions of certain particles and antiparticles differ. These differences, which are small and difficult to measure, amount to violations of a symmetry called charge parity reversal, or CP. The origin of CP violation and the nature of the connection between CP violation in particle physics and cosmology is one of the outstanding questions in physics. Three new particle physics experiments coming on line in 1999 offer a realistic opportunity to understand the phenomena.

Over the last 30 years the field of particle physics has seen enormous progress that is codified in the Standard Model (SM) which describes all of the hundreds of observed particles and their interactions in terms of six quarks and six leptons and their anti particles. These entities are arranged in three generations of increasing mass. Ordinary matter is composed of particles from the first generation.

The SM describes three kinds of interactions among particles: (1) the familiar electromagnetic force; (2) the weak force which is responsible for beta decay, and, at a more fundamental level, allows the more massive quarks and leptons to decay into lighter particles; (3) the strong force which gives rise to the stability of the nucleus, and, at a more fundamental level confines the quarks within composite particles such as baryons (three quarks) and mesons (one quark and one anti-quark). Each force is transmitted by a propagator, the familiar photon, the W and Z bosons and the gluon, for the electromagnetic, weak and strong forces respectively. Finally, the SM requires a Higgs particle, whose interactions are believed to give rise to the mass of quarks and leptons.

Whereas photons are massless, the W and Z bosons are almost 100 times as massive as a proton. This makes it difficult to produce them and is the reason why weak interactions are weak at low energies. However, at energies much greater than the W and Z mass, these heavy propagators are easy to produce, and the weak and electromagnetic forces become equal in strength. To account for the symmetry between the weak and electromagnetic forces at high energy, and the broken symmetry at low energies, a scalar field , the Higgs field, is introduced. The closest analog to a scalar field is the electrostatic potential . The electric and magnetic fields appear only if is inhomogeneous or time dependent. In other words if the whole universe was at the same electrostatic potential of, say, 100 V, we would be unaware of it, 100 V, 120 V or 0 V are all vacuum states. In the same way a constant scalar field looks like a vacuum state.

There is an important difference between and . The latter does not have energy associated with it, whereas may have an energy density V(). If V() has a minimum at , the whole universe is eventually filled with the field . The field is invisible, but it interacts with the W and Z in such a way that they become heavy, while the photon remains massless.

The SM is an extremely successful description of the world, but it contains 18 parameters that must be determined by experiment. There is no understanding of why three generations of leptons and quarks exist, and the Higgs particle is yet to be observed. In addition, while the SM unifies the weak and electromagnetic forces, this unification does not extend to the strong force. There are many theoretical extensions of the standard model. For example, Grand Unified Theories (GUT’s) extend the symmetry of the SM by placing quarks and leptons on the same footing. A suitable multiplet of Higgs fields ensure that the W, Z, and photon have the correct mass while the masses of the new bosons, X, in the theory, have a mass of GeV. Well above this energy the X bosons are readily produced and the strong force is unified with the electroweak force. In GUT’s, the X bosons transform quarks into leptons and vice versa. Baryon number, the number of matter particles minus the number of anti-matter particles, is not conserved.

Symmetry is a powerful tool in many branches of physics. In HEP, symmetry was used to predict the existence of a quark partner for the beauty quark long before the discovery of the top quark. Another important symmetry is parity. The parity operator, P, turns an object into its mirror reflection and rotates it 180 degrees about the axis perpendicular to the mirror. A theory has P symmetry if the laws of physics are the same in the parity reversed world as in the real world. Leptons and quarks can be classified as right or left handed depending upon the sense of their internal rotation (spin) about their direction of motion. If P symmetry holds, right handed particles behave exactly the same as left handed particles. The laws of QED and the strong interactions obey P symmetry, but not so the weak interactions. Only left handed particles can decay by a weak interaction, and only left handed neutrinos exist.

The charge conjugation operator, C, changes the quantum numbers of every particle into those of its antiparticle. C symmetry is also violated in weak interactions; antineutrinos are not left-handed, only right-handed. Combining the operation of C and P to obtain CP, a left-handed neutrino becomes a right-handed antineutrino. Both exist and have identical properties. CP symmetry appears to be obeyed by neutrinos.

Either C or P, if acting on a system twice, returns it to the original state. As a result the CP of a state can only be +1 or -1. If nature obeys CP symmetry, no state with CP number +1 can transform into a state with CP number -1. Consider neutral kaons. The consists of a down quark and an antistrange quark; the consists of an antidown quark and a strange quark. Because CP transforms a quark into an antiquark the is not an eigenstate of CP, however by superposing the and wave functions, two eigenstates of CP are obtained. By the rules of quantum mechanics, these combination kaon states correspond to real particles that have definite masses and lifetimes. The kaon with CP=+1 decays rapidly to two pions, a state also with CP = +1, because a kaon at 497 MeV/c² (about half the mass of a proton) is much heavier than two pions (135 MeV/c²). The kaon with CP = -1 decays 500 times more slowly to three pions, a state with CP = -1, because there is barely enough mass for this reaction to occur. The observation of long and short lived neutral kaons was considered evidence that the combination kaon states obeyed CP symmetry.

This simple picture was overturned in 1964 when Cronin and Fitch observed that about 1 in 500 of the CP = -1 kaons decay to two pions. If CP were an exact symmetry of nature, it would forbid such a decay. Although CP violation was not predicted by the SM, in 1972 Kobayashi and Maskawa (KM) showed that CP violation can be accommodated in the SM if three or more quark generations exist. The KM ansatz does not explain CP violation; it relates the origin of CP violation to the central mysteries of the SM! In 1975 Perl discovered the tau lepton. In 1977, 1994 and 1998 experiments at FNAL discovered the beauty quark, top quark and tau neutrino respectively.

The type of CP violation found by Cronin and Fitch, characterised by the quantity , arises because of the mixing of the wavefunctions of the and , and is known as mass matrix CP violation. In the SM, CP can be violated in a second way, known as direct CP violation. This involves a difference in the decay rate of a particle and antiparticle to the same final state. Direct CP violation is independent of particle mixing. NA31 at CERN, is so far the only experiment to have observed the phenomenon. This was achieved by building a spectrometer capable of collecting and measuring about 10⁹ decays. Direct CP violation in the kaon system is small, characterized by the quantity , it is 1,000 times rarer than mass matrix CP violation. Unfortunately this small size has made it experimentally difficult to test the SM explanation and compare it to rival theories.

There is a consensus that at early times the universe was a lot hotter and denser than today. The existence of the cosmic microwave background demands a very hot early universe and the observed expansion of the universe demands a compact early universe. The standard Big Bang theory asserts that the universe was born in a state of infinitely high temperature and density about 15 billion years ago. The temperature of the expanding universe gradually decreased as a reciprocal of its size, and finally evolved into the relatively cold universe in which we now live.

The universe possesses a variety of properties that would not be expected from the Big Bang model. These include (1) isotropy and homogeneity–the cosmic wave background radiation is much more uniform than expected; (2) flatness–the universe is close to critical density. Many cosmologists believe that the answer to these problems lies in the theory of inflation. Inflation is a period in the early history of the universe in which extremely rapid expansion occurred. In a minute fraction of a second the size of the universe increased by 10⁴² times or more (a model dependent quantity). This enormous expansion is the result of a (temporary) cosmological constant.

Scalar fields are the mechanism for inflation. The vacuum energy density associated with the scalar field acts just like a cosmological constant. In one scenario, called chaotic inflation, it is assumed that the scalar field begins in a random state in which all possible values of occur. In some regions, these values correspond to a large potential energy and inflation ensues. According to chaotic inflation, what we now see as the observable universe was contained entirely within one of the inflating regions. At the end of inflation, the energy of the vacuum-like state transformed into thermal energy, the universe became hot and its subsequent evolution can be described by the standard Big Bang theory.

The synthesis of nuclear physics, atomics physics and the Big Bang theory, starting in the 1940’s, allowed physicists to talk meaningfully about the condition of the universe in its earliest seconds, a remarkable feat compared to our lack of certainty of the Hubble age of the universe to within a few billion years.

In astrophysics nuclear reactions are primarily confined to two arenas. One, the fusion process that powers stars, stellar nucleosynthesis, is not of interest here. The second concerns the brief moment, long before stars existed, when the high density and high temperature allowed nuclear reactions to occur throughout the universe. In the process of fusing single protons to make helium nuclei, deuterons must maintain stability. But deuterons are easily photodissociated by high energy gamma rays. At an age of 100 s, the temperature of the Universe had dropped to 10⁹ K, and there were very few photons energetic enough to photodissociate deuterons; nucleosynthesis began. All isotopes of H and He with Z £ 4 were produced. By 10³ s the Universe had cooled sufficiently that fusion ceased.

The observed cosmic abundances of the light elements agree extremely well with calculations and give confidence to the assertion that we understand the state of the universe in the era 1-1000s. This success is attributable to knowing the fundamental physics well. It is then natural to ask if such a success could be repeated for an even more basic quantity, the baryon number of the universe? As we examine earlier, and more energetic times, we substitute particle physics for nuclear physics and encounter less certain territory.

The matter anti-matter asymmetry of the universe may have been an initial condition. However, such an asymmetry would have been eliminated if the early universe contained any process that could change baryon number. If GUT’s are operative, such processes would have been very common. For this reason there is a theoretical preference for an initially symmetric universe, in which matter came to dominate as the universe cooled.

In 1967 Sakaharov formulated the three conditions required for an asymmetry to develop. (1) Since the initial and final baryon numbers of the universe differ, there must exist baryon number violating transitions. (2) CP violation must exist, otherwise every process that changes the amount of matter, would be balanced by a similar process for antimatter. (3) The baryon number and CP violating transitions have to proceed out of thermal equilibrium. (When in thermal equilibrium all states of equal energy contain equal populations of particles. Since particles and anti-particles have the same mass they would be generated at the same rate.) GUT’s can meet all three conditions, and baryogenesis may have arisen in this way. At this time the universe was slightly older than 10^-35 seconds, and the temperature was 10²⁸ K. A baryon number generated at such high temperatures could be eliminated, or greatly diluted, in the subsequent evolution of the universe. Secondly, there is no experimental evidence for GUT’s, and due to the very high mass scales involved, little hope in the forseeable future of obtaining anything other than indirect information about these theories.

Baryogenesis at the electroweak scale is the most actively analyzed scenario at present. The electroweak phase transition occurs when the universe is about 10^-10 seconds old and the temperature is 10¹⁵K. At energies below the phase transition baryon number violating processes are rare. Consequently there is lttle danger of a net baryon number emerging from this phase transition being subsequently diluted. Eventually, the universe cooled to the point where particles and antiparticles could no longer be generated in collisions, but would annihilate when they met, leaving the residue of matter we see today. Significantly, current calculations show that, within the SM, baryogenesis at the electroweak scale is not capable of generating a sufficiently large asymmetry, although some extensions of the SM may be capable of doing so. This failure of the SM suggests there must be other ways in which CP violation arises and hence the SM may be incomplete. Measurements of CP violation are therefore a very sensitive probe of new physics beyond the Standard Model.

Kaon physics had been actively studied for thirty years. Experiments had produced kaons copiously, and rare decays at the 10^-9 level had been well measured, but new attempts to measure with greater precision were proving difficult. In contrast to kaon decays, large CP asymmetries are expected by the SM in rare beauty quark decays. Therefore, many in the kaon community switched to experiments with B mesons.

A meson is obtained from a kaon by substituting a beauty quark for a strange quark. Two types of meson are of immediate interest; when an anti-beauty quark is paired with an up quark a B⁺ results; when paired with a down quark a results. A neutral meson consists of a mixture of and a wavefunctions. If at time a is produced, at a later time there is a non-zero probability an observer will find a . Neutral mesons, like neutral kaons can exhibit CP violation mediated by mixing, and direct CP violation. Charged particles cannot mix and so a B⁺ meson can only exhibit direct CP violation. Within the SM B meson direct CP violation is less easy to predict than mixing mediated CP violation.

Since a beauty quark has a mass five times that of a proton, mesons are more difficult to produce than kaons. Electron positron colliders, at a center of mass energy of 10.58 GeV were found, at Cornell and DESY in Hamburg, Germany, were found to be optimal for producing beauty quarks. Each electron positron annihilation has a high probability of producing a B meson and its antiparticle. The CLEO and ARGUS experiments at Cornell and DESY efficiently measured the particles that emerged from the decays of the B mesons.

However, limited by accelerator performance, data samples of 10⁵ mesons were state of the art. While CP violating effects were expected to be large for both neutral and charged mesons, the decays that exhibit these effects are mostly very rare, and samples of at least a few 10⁷ ‘s are needed. A sample of this size was well beyond the reach of particle accelerators then in existence.

There were some significant compensating features for the ex-kaon physicists however. First, physics was virgin territory and with the data sample available it was possible to constrain two of the 18 fundamental parameters of the standard model and to demonstrate that the meson exhibits quantum mechanical mixing, just like a kaon. Secondly, when beauty quarks are manufactured large quantities of charm and strange quarks, and tau leptons are also produced, and new tests of parity violation and CP violation became possible.

Semileptonic (beta) decays of baryons containing charm quarks allow particularly simple and sensitive tests of theoretical predictions through measurements of parity violation, and the decay amplitude as a function of the momentum transfer squared in the decay. The contains a charm quark, an up quark and a down quark. Since the charm quark mass is almost twice that of a proton, while the up and down quarks have masses much less than that of a proton, the resembles a hydrogen atom, i.e., the charm quark is relatively static with the light quarks orbiting it. If this picture is correct the decay (which is the familiar beta decay with the substitution of a charm quark for a down quark) should exhibit close to 100 percent parity violation. Data confirming this is shown in Figure 1. Measurements such as this, when combined with future data on beauty baryon decay from the Fermilab Tevatron and the Large Hadron Collider at CERN will allow a very precise determination of V_ub and V_cb, the coupling strengths of the and processes, which are related to fundamental parameters of the SM.

The X baryon, and its anti-particle, are also copiously produced at Cornell. (The X contains a strange quark and two up quarks.) If the SM is correct, CP violation in X baryon decay is related to CP violation in kaon decay, but can be relatively large even if is small. CLEO searched for CP violation by comparing the decay angular distribution of protons in the sequence , , to the antimatter state . Any difference in the two angular distributions would be unequivocal evidence for direct CP violation, which would eliminate several rival theories of the SM, and may indicate new physics. No evidence for CP violation was found.

Especially since 1990 when the CLEO II detector was commissioned, CLEO has been a world leader in heavy flavor physics. Over half of all measurements of beauty mesons, charm mesons and baryons are based on CLEO results. Since 1991, 22 percent of all papers in Physical Review and Physical Review Letters originating from U.S. HEP experimental facilities have come from CLEO. The data sample, the world’s largest, now stands at 10⁷ B’s with similar number of charm and tau particles.

This year the DoE High Energy Physics Advisory Panel (HEPAP), the NSF Cahn Panel and the National Review Committee of the National Academy re-iterated that B physics a national priority. The first experiments sensitive enough to observe and measure CP violation in the B system, the "B Factory" experiments; BaBar at SLAC, Belle at KEK, Japan, and CLEO III at Cornell, will commence data taking in 1999. Each experiment expects to acquire of order 10⁸ B mesons.

The Babar, Belle and CLEO detectors are very similar, however the accelerators differ significantly. At SLAC and KEK two separate rings deliver different energies to the electrons and positrons. When the beams collide the pair of B mesons produced are boosted in the direction of the higher energy beam. Moving at half the speed of light, the B meson travels for 250 microns before decaying. This facilitates the measurement of the B meson point of production and decay, which is neccessary to observe mixing mediated CP violation in neutral B mesons. At Cornell, the particle beams are of equal energy, and so the B mesons are produced almost at rest. This configuration is ideal for searches for direct CP violation in either type of B meson. Both types of collider will provide crucial and complementary information about CP violation.

All three experiments are significantly more comprehensive than the original CLEO and ARGUS experiments. All can precisely measure the momentum, velocity, energy, and penetrating power of the charged and neutral particles produced by the B mesons as they decay.

An essential element of all three experiments is a high precision silicon microstrip tracker which detects the birth and death of the elementary particles containing beauty and charm quarks whose existence lasts for a trillionth of a second. Based on microelectronic fabrication technology similar to a computer chip or a video camera, the basic building block of the detector is a silicon wafer, slightly greater in area than two postage stamps and 300 m m thick. The silicon is depleted of free charge carriers. Incident radiation ionizes the silicon, under the influence of an internal electric field the liberated charges drift rapidly towards typically 500 electrodes (microstrips) patterned on each side of the silicon wafer. The motion of the charges generates an electrical signal on one or a few nearby strips. Once generated the signals are amplified and recorded by special electronic readout chips. For example, the CLEO III silicon detector, Si3, currently being built at Purdue, is a very precise, accurate and fast array of 447 silicon wafers arranged in four concentric cylinders centered on the point the particle beams will collide. Si3 contains 450,000 individual microstrips attached to 125,000 channels of sensitive electronics and assembled on an elaborate and very stable tripod made of synthetic diamond and copper. The detector will be installed in 1999 and will take data for approximately five years.

The B factories may tell us that the Standard Model concept of CP violation works, and then help to determine its remaining parameters. Alternatively, results could show that the Standard Model predictions are incompatible with the data, and, furthermore, rule out entire classes of extensions of the Standard Model.

A reality check will occur about 2002 when an evaluation of the state of the field will take place by national panels to decide on the future support of e⁺e^- machines in the United States. Many experimentalists, especially the ex-kaon physicists, believe that while the B Factories will be able to observe CP violation they may not be able to obtain sufficient constraints and precision to choose between a variety of theoretical alternatives to the Standard Model. A central question would then be: how large a data sample is needed and by what method do we obtain it? There are two approaches. One is to significantly increase the luminosity of the e⁺ e^- machines. This is very difficult, but is believed to be easier to achieve in the symmetric configuration. The other is to use the proton colliders at Fermilab and CERN. Groups of physicists are already in the planning and prototyping stages for new experiments at both types of facility.

Experiments at the proton facilities offer a realistic opportunity to obtain the largest data samples, up to a few 10⁹ B mesons, but with much larger backgrounds than at e⁺ e^- machines. The proton experiments also present a significant technological challenge. The very high data rates and large backgrounds require very fast and precise radiation hard detectors. Silicon pixel detectors and silicon microstrip detectors in conjunction with other, newer technologies such as microstructure gas detectors (MSGD), which are being developed at Purdue, are expected to play an important part of the proposed new experiments.

MSGD’s are a class of detectors that apply micro-electronic fabrication techniques to define microstructures around which sizeable gas amplification can be obtained. The simplest version of the detector, the Micro Strip Gas Chamber (MSGC), resembles a gas proportional wire counter in which the wires have been replaced by a series of a closely spaced metallic strips (200 m m pitch) printed using lithography onto a plastic or glass foil or onto a thin film of synthetic diamond. Radiation ionizes an inert gas mixture above the foil and the resulting electrons are accelerated by an electric field to the detector strips where they produce a current pulse.

Miniaturization produces a detector which has excellent energy resolution, is an order of magnitude more precise than pervious gas detectors, while being several orders of magnitude faster. To reduce the operating volage of an MSGC a preamplification structure can be used. A GEM consists of a kapton mesh 50 m m thick coated with 5 m m of copper on both sides of the kapton. Holes in the kapton are 100 m m in diameter at a pitch of 140 m m. Application of a voltage difference between the metal layers of the GEM produces an electric field in the holes sufficient for gas multiplication. When GEM is inserted above an MSGC, electrons liberated above the GEM drift into the GEM holes, multiply in the high field, and then continue to the MSGC which may be operated at a much lower voltage than before to achieve the same gas amplification.

The Purdue group was the first in the United States to build a MSGC entirely in-house. Recently, at Purdue, some of the best radiation hardness results in the world have been obtained with MSGC’s and MSGC’s in combination with GEMS’s.

CP violation, first observed in 1964, is still not understood. The B factories coming on line in 1999 offer a realistic opportunity to understand the phenomena. We may soon know why our universe is matter anti-matter asymmetric, an asymmetry that permits us to exist.

I thank Daniela Bortoletto, Art Garfinkel, and David Miller for reading early versions of this manuscript. I also want to thank Ikaros Bigi, and Serguei Khlebnikov for valuable discussions. The forthcoming book CP Violation by Ikaros Bigi and Tony Sanda, (in press) Cambridge University Press (1998), provides an excellent and complete account of many of the issues discussed in this article. This article has been written from the perspective of a member of the CLEO collaboration. A slightly different perspective can be obtained by reading the article by Michael Witherell, a member of BaBar, in the October 1998 issue of Scientific American.

Professor Ian Shipsey is involved in searches for CP violation and studies of parity violation in the weak interactions of elementary particles. As a graduate student he was a member of the NA31, the only experiment to have observed direct CP violation. He has been a member of the CLEO collaboration since 1986. He built the muon detector for the CLEO II, and since 1994 he has been leader of the mechanical design and fabrication group for the silicon microstrip detector for CLEO III. He is one of the leaders of CLEO IV a proposed new experiment that would succeed CLEO III in about 2004. He is also involved in the development of microstructure gas detectors for future high energy physics experiments. He collaborates at Purdue with Professors David Miller and Ed Shibata, postdocs Jim Fast, Jik Lee and Jun Miyamoto, engineer Kirk Arndt, technicians Ernie Beard and Tom Smith, current and former graduate students: Mary Bishai, Shenjian Chen, Ekkehard Gerndt, Naresh Menon, Ting Miao and Victor Pavlunin, and many undergraduate students including, most recently, Martha Neustadt and Derek Tournear.