On 9 May, the KEKB accelerator achieved a major breakthrough by being the first colliding-beam facility to reach a peak luminosity above 1034 cm-2 s-1, a long-sought milestone in accelerator physics. This accomplishment will boost KEK’s programme of investigating CP violation and searching for beyond-the-Standard-Model effects in the Bbar-B system. Almost all such studies require the largest possible samples of B-meson pairs, so the single most important factor in their success is high luminosity. A luminosity of 1034 cm-2 s-1 corresponds to a B-meson pair production rate of 10 per second, and under normal operating conditions this would yield approximately 100 million B-meson pairs per calendar year.
In order to increase the luminosity, the intensity of both the electron and positron beams must be increased, and each must be squeezed to the smallest possible size. Achieving these two conditions simultaneously presents a severe technological challenge to the accelerator design team. To tackle these problems, the KEKB group, which started construction of their machine in 1994, incorporated a number of new technologies. Among them are finite-angle crossing for the collision region, a lattice design with a 2.5 π phase advance per cell, superconducting RF cavities that can tolerate large beam currents, and normal RF cavities coupled with attachments that have 10 times more energy-storage capacity than the accelerating cavity proper (called ARES cavities). The luminosity target set for KEKB at the outset was 1034 cm-2 s-1, which at the time was considered by many to be an unrealistically ambitious goal.
The commissioning of KEKB was reasonably smooth, with the luminosity reaching 20% of its design value in one and a half years. This was already highly successful in comparison with many past accelerator projects, where progress in the early stages could be laboriously slow. KEKB’s luminosity has steadily increased ever since, as the KEKB team have worked to solve the many problems they have encountered along the way.
Three categories of problems have been the most persistent and difficult to overcome. The first class of problems is related to the high beam currents. KEKB has frequently experienced serious trouble such as the breakdown and heating of the vacuum components. Vacuum chambers in the interaction region and beam abort sections, movable masks (for removing beam tails) and bellows, etc, were broken several times due to higher order mode (HOM) power from the beams, synchrotron radiation, or simply being hit by the beams themselves. The KEKB team solved these problems one by one, by developing revised versions of the components, reinforcing the cooling power and protection mechanisms, and by taking other counter-measures.
The second category of problems was a nagging blowup of the positron beam, believed to be caused by a photoelectron cloud. In the end, the KEKB team covered 2300 m (more than 90% of the free region) along the 3000 m positron ring with a solenoid winding that was designed to produce a small axial magnetic field to disperse the cloud.
The third problem category was beam blowup due to the beam-beam effect. This is a well known and common problem in colliding beam accelerators, and a huge amount of effort has been devoted since the design phase of KEKB to mitigate this blowup. The most important progress at KEKB on this issue is a special choice of betatron tunes. It turned out that making the horizontal tunes of both rings approach half-integer resonance was very effective in raising the luminosity. This effect is explained as a dynamic focusing effect by the beam-beam interaction, and is well reproduced by the beam-beam simulations. To enable this very close approach to the half-integer resonance, corrections for machine errors were crucially important.
Reaching the 1034 design luminosity in four years is a remarkable achievement. KEKB operates with 1284 bunches in both electron and positron rings, with averaged bunch spacing of 3.77 RF buckets. The currents are 1.5 and 1.1 A for the positron and electron beams, respectively, and these correspond to 58 and 100% of the design currents. Vertical beta functions at the interaction point are 5.8 and 7 mm for the positron and electron rings, respectively, which are even smaller than the design value of 10 mm. The vertical beam-beam parameters are 0.066 and 0.050 for the positron and electron beams, respectively, which are very close to (or higher than) the design value of 0.052. The fractional part of the horizontal tunes are 0.506 and 0.513 for the positron and electron beams, respectively.
The remarkable success of the KEKB project is an indication of not only the highest level of achievement in KEK’s accelerator technology, but also of the highest levels in numerous branches of industry that support KEK. The experience gained in the KEKB project will be of great value in the development of a future linear collider.
