"An ESQ accelerator with its exceptionally high-currents will reduce patient treatment time (because of the high yield of neutrons) which means (among other advantages) that a greater number of patients can be treated."

According to Kwan, it has been estimated that patient treatment time would be less than one hour with a 100 milliamp proton beam at 2.5 MeV. BNCT treatments involving nuclear reactors generally last much longer.

In addition to the accelerator itself, an accelerator-based BNCT system also needs a neutron production target, and a moderator and filter assembly for shaping the neutrons into a clinically useful beam. Although there are a number of good candidates, Berkeley researchers have chosen lithium as their target because it offers the highest yield of suitable neutrons.

The electrical neutrality that enables neutrons to penetrate atomic nuclei also makes shaping them into a beam a tricky proposition. Unable to use the electromagnetic fields that focus and direct protons or electrons, scientists must depend upon more elaborate gimmicks in which neutrons are directed into certain materials that can influence their flight paths and energies. These materials, however, can create their own sets of problems.

Explains Chu, "Neutrons can lose energy very fast in water or plastic, scatter a lot in other materials, or stop altogether in still another kind of material. Not only must we shape the beam energy and profile, we must do so without making too many extraneous photons (x-rays) or leaving residual fast (high energy) neutrons."

To accomplish this, LSD physicist Bernhard Ludewigt, along with Rick Donahue, from the Environmental Health and Safety Division, and UC Berkeley nuclear engineering grad student Darren Bleuel have come up with a three-stage system involving a moderator, a reflector and a collimator. The moderator, a block of material such as aluminum fluoride, is placed downstream from the target; the reflector, a block of aluminum oxide, is placed upstream. When the proton beam bombards the target, neutrons fly off in all directions at energies ranging from 200 to 800 keV. Forward-moving neutrons directly strike the moderator; backwards-moving neutrons are reflected into it. The moderator slows these neutrons down or "moderates" them to the 20 keV or lower energies required for safe BNCT. Further downstream from the moderator is the collimator, a block of plastic with a hole that faces the patients brain. Epithermal neutrons swarm into the collimator but emerge out of the hole in an orderly fashion as a beam that shines

"The neutron beam energy can be shaped to reach deep-seated tumors," says Ludewigt, "so that we can place 50-percent more dose in the middle of the tumor than what can be achieved with neutrons from reactors."

As stated earlier, the success of BNCT also depends upon the selectivity of the boron-carrying compound. The clinical trials now taking place at BNL have been approved by the Food and Drug Administration (FDA) to be conducted with two compounds that were developed in the late 1950s and are not viewed as ideal. Recently, UCSF Professor Stephen Kahl developed a compound called "boronated (proto)-porphyrin" or BOPP, which has shown promise in various animal studies. For any such compound to be effective, the crucial requirements are that it be able to slip past the brains blood barrier, that it concentrates in tumors, and that it rapidly clears from healthy brain cells and the blood system.