This paper describes various high-level nuclear researches including nuclear-fuelled pebbles that are being conducted across South Africa. The pebbles are ingenious industrial products, designed to passively limit the amount of heat unleashed by the nuclear fission reactions that drive the reactor. The spheres that give the pebble bed reactor its name enclose fissionable uranium inside layers that serve various roles, such as moderating fission, containing pressure, and accommodating deformation of the core. Nuclear-fuelled pebbles are introduced at the top of the reactor vessel and slowly wend their way down through the annular-packed bed under the action of gravity to the bottom of the reactor vessel. In a towering building at the headquarters of Nesca in Pelindaba, South Africa, reactor components are being tested for their ability to work with high-pressure helium. Those parts will go in the pebble bed modular reactor power plant to be constructed at Koeburg, near Cape Town. The plan of the pebble bed reactor power plant will use the helium coolant to run the turbine directly rather than heat a secondary fluid, as in a water reactor.


High-level nuclear research is conducted at a number of sites across South Africa, each astonishing in its own way. The state-owned Nuclear Energy Corp. of South Africa, for instance, is headquartered in Pelindaba, about 25 miles west of Pretoria. As much as the rolling veldt in Pelindaba looks like the Texas hill country, it has a momentous history unlike anything in America. Just to the west lies the Cradle of Humankind, a World Heritage Site consisting of a series of limestone caves that contain the world's richest concentrations of hominid remains, some dating back over 2.8 million years. One of the major discoveries there was evidence of the controlled use of fire dating back more than a million years.

Looking out over the Pelindaba landscape from the top of a nuclear test facility, one also remembers that this is where six atomic bombs were assembled back in the 1980s. (In a grim coincidence, the place name "Pelindaba" originates from Zulu words meaning "the matter is settled.") Later, in the 1990s, South Africa dismantled all of its nuclear weapons, thus becoming the only government to date that has voluntarily given up its home grown arsenal. The African Nuclear Weapon-Free Zone Treaty, an international agreement among African states not to manufacture or acquire control of nuclear weapons, is also known as the Treaty of Pelindaba.


Today, Pelindaba is one of the places that is developing a technology that may not be as revolutionary as fire, but is certainly one of the most promising emerging energy sources. The pebble bed modular reactor is a power plant design that uses heat from fission to "fuel" a gas turbine to generate electricity. Now under development and less than a year from prototype construction, the PBMR will have a gas turbine-driven electrical generator, which will convert some 41 percent of the reactor's thermal output into 165 MW of electricity. That compares quite favorably to an average of 32 percent for conventional light water nuclear power plants.

If all goes according to plan, the PBMR will become something like. the Boeing 747 of nuclear reactors: a reliable, exportable high-tech product that transforms the South African economy.

The development and subsequent manufacture of the reactors is being carried out by Pebble Bed Modular Reactor (Pty.) Ltd. in South Africa, with participation by teams from Eskom (South Africa's state-owned utility), Westinghouse from the U.S., Mitsubishi Heavy Industries of Japan, and Nukem of Germany.

According to Tom Ferreira of PBMR Ltd., as of last July about 2,000 support staff and engineers in various countries were working on the PBMR project, with about 700 at PBMR's headquarters in Centurion, just to the west of Pretoria.

Although South Africa is the country most closely identified with the concept, PBMR Ltd. credits the German physicist Rudolf Schulten, who died in 1996, as the father of the pebble bed reactor. He proposed the idea in 1956, when he joined Brown Boveri & Cie. in Mannheim, Germany, after leaving Gottingen, where he took his doctorate under Werner Heisenberg. The first pebble bed reactor power plant was built at the Nuclear Research Center of Julich, close to Aachen, where Schulten subsequently became a director and a chaired professor at the Technical University of Aachen. With a thermal output of 40 MW to a steam Rankine cycle, the research reactor became fully operational in 1966, and ran successfully for 21 years, producing utility grid electrical power for 70 percent of its operating life.

the pebble bed

Nuclear-fueled pebbles are introduced at the top of the reactor vessel and slowly wend their way down through the annular packed bed under the action of gravity to the bottom of the reactor vessel. There, a pebble falls into a defueling chute leading into a tubing system external to the reactor vessel. The pebble is then moved pneumatically by helium gas pressure up to assaying equipment above the reactor vessel, much as pneumatic mail was. delivered through pressurized air tubes in department Stores and banks during the last century.

Pebbles are individually assayed for burn-up-the nu clear neologism for reactor fuel consumption-and physical integrity. Depending on the results, pebbles are pneumatically routed either back to the top of the reactor core for reuse or to a spent fuel tank. It is expected that a typical pebble will make six loops through the reactor in the three years of its fuel life.

Conventional light-water nuclear power plants shut down completely every one to two years for refueling, an expensive operation during which the plant is not producing revenue.

A PBMR plant has a continuous fueling process occurring during full load operation. Since fresh pebbles can be added as needed, there's no need for such a shutdown. Plans do call for a 30-day shutdown for maintenance every six years, which amounts to about six or seven shutdowns over the planned 40-year life of a PBMR. That compares favorably to the 20 or 30 shutdowns expected over the life of a conventional nuclear power plant.

working with helium

The pebble bed modular reactor uses helium gas, rather than water, as working fluid. At a PBMR power output of 165 MW, helium gas flow as 190 kilograms per second down through the reactor. In the interstices formed by the spherical carbon surfaced pebbles. The gas is heated by convection radiation from the nuclear flssion heat generation in the kernels and conducted to pebble surfaces.

The PBMR design does not put the helium in direct contact with radioactive substances, since the fissile material and fission products are sealed in the kernels and pebbles. What's more, helium is inert, so neutron activity does not cause it to become radioactive. Thus, the helium may be used directly as the working fluid in gas turbine turbomachinery, without the need for an isolating heat exchanger and another, separate fluid loop.

The part load characteristics of a closed-cycle gas turbine are remarkably good, quite unlike open-cycle operation. If the PBMR gas turbine is at the design output of 165 MW, load reduction is achieved by bleeding helium from the closed loop. This reduces the helium mass flow rate, reducing power output, and lowering the mean gas density, but maintaining constant gas velocities at constant rpm. In gas turbine designer terminology, the turbomachinery velocity triangles remain the same, so that the PBMR design efficiency of 41 percent will remain the same over a wide range of load operation.

To get the most efficient load reduction, one should bleed helium from a location of high pressures in the cycle- that is, at the exit of the high compressor. In that way, when power output is to be increased, helium can be injected back into the closed loop at a location of lower pressure, such as the entrance of the low compressor. Wise management of helium inventory can obviate the need to use costly power to compress the injected helium.

This first reactor had many similarities with the design PBMR Ltd. is now developing. The nuclear fuel consisted of coated kernels that were formed into spherical pebbles. Helium flowed around and between the pebbles that were stacked in the reaction chamber, and the gas carried away heat that could then be used to generate electricity. Tests demonstrated that the plant was completely safe in the event of a total shutdown of the helium coolant.

Schulten most assuredly deserves the credit as the inventor of the first working pebble bed reactor used to generate electrical power. But the idea of the pebble bed reactor actually goes back to the very first nuclear reactor work at the University of Chicago in the 1940s. It's a story of a technology pathway not taken.

Potomac Pebbles

Enrico Fermi 's carefully arranged pile of uranium and graphite bricks ran on Dec. 2, 1942, in the university's squash courts, producing a thermal output of about 200 W from a sustained nuclear chain reaction. At the time, the physicists weighed the potential for using helium as a coolant for future nuclear power plants, but the idea was rejected as too complicated and water was favored instead.

In 1944, Farrington Daniels, a chemistry professor from the University of Wisconsin, joined the Chicago Metallurgical Laboratory group as associate director of the chemistry division. For some years, Daniels had been working on a process of fixing nitrogen from air using a novel pebble bed heated furnace. Percy R oyster, working in the 1920s with Frederick Cottrell, director of the United States Fixed Nitrogen Research Laboratory in Washington, DC, had perfected the technology. Royster had found that gravel pebbles dredged from the Potomac River, graded for size and put into a steel furn ace, provided the best means of heating the air for the nitrogen fixing process.

Building upon the legacy of Potomac pebbles and his own pebble bed furnace research, Daniels proposed that a chain-reacting pile be constructed along similar lines. The pile would consist of uranium oxide and carbide pebbles whose heat of fission would be removed by a flow of a cooling gas. Daniels filed a patent on his idea on Oct. 11 , 1945. In the patent (2,809,931, U.S.) he calls the pile a "pebble bed reactor," claiming that the cooling gases be used to generate steam (to power a steam turbine), or " .... the heated gases can be used directly in gas turbines." The next year, design work started at Oak Ridge on the Daniels power pile, a helium-cooled reactor based on Farrington Daniels' concept.

The Atomic Energy Commission was also formed that year, and one of its first acts was to cancel the project. Instead, according to Alvin Weinberg, who at that time was director of the physics division at Oak Ridge (and ultimately headed the entire lab), the Daniels power pile team became the nucleus of the group that designed the first naval nuclear power plant. Funded by the Navy under the direction of Hyman Rickover, the reactor for the submarine Nautilus was cooled by pressurized water. Originally chosen to fit into the cramped confines of a submarine hull, water-cool ed reactors are used today in most of the world's 440 operating nuclear plants.

Although light water reactors-both pressurized and boiling water-dominate, there are gas-cooled reactors in operation in the United Kingdom. The first was at Calder Hall in 1956. High-temperature, gas-cooled reactors have been attempted in the past, notably at Fort St. Vrain in Colorado and a 300 MWe pebble bed unit built in Germany. Both units were shut down in the 1980s after a few years of operation. The high-temperature, gas-cooled reactor projects now under way include not only the South African effort, but also the General Atomics GT-MHR unit, a high-temperature test reactor in Japan and a 10 MWt pebble bed reactor in China.

Of those projects, the South African PBMR is the one closest to commercial testing and manufacturing. Recently, the U.S. Department of Energy awarded a $3 million first-phase engineering contract to PBMR Ltd. for the Next Generation Nuclear Plant at the Idaho National Laboratory. There is widespread interest in PBMR, and many in the world's electric power community will wait to see test results in the coming years from the first unit.


The pebble bed reactor will be constructed at Koeburg, a site on the Atlantic Coast about 20 miles north of Cape Town, and will run slightly cooler than Schulten's test reactor. The helium-gas coolant will flow at a little more than 900°F and 1,300 pounds per square inch into the reactor assembly. A capsule 90 feet high and 20 feet wide is packed with about 450,000 heat-producing " nuclear pebbles," each the size of a tennis ball.

The pebbles are ingenious industrial products, designed to 'passively limit the amount of heat unleashed by the nuclear fission reactions that drive the reactor. The fissionable material is divided into groups of 100 million billion uranium-235 atoms scattered within a 0.5 mm sphere of uranium dioxide. That core is coated with four containment layers. The first is a porous carbon buffer layer that can accumulate gaseous fission products and accommodate core deformation , which might occur during the nuclear fuel lifetime. The second, third , and fourth coatings are pyrolytic carbon, silicon carbide, and pyrolytic carbon, respectively, all layered to provide radiation, gas, and mass diffusion barriers and to act in concert as a pressure containment vessel. (A feature, "Hot Case," in Mechanical Engineering's January 2004 issue discussed kernel manufacturing.)

Some 15,000 carbon-coated uranium dioxide kernels, looking like nuclear poppy seeds, are embedded in a tennis ball-size graphite and resin matrix. Graphite acts to slow the fission neutrons and so acts as an embedded moderator. The matrix is then encased in a 5 mm pure carbon shell, sintered, annealed, and machined to a uniform diameter of 6 cm.

Because the fissile material and fission products are so well sealed-and because neutron activity doesn't make helium radioactive in any case-the coolant gas that flows through the stack of pebbles will be used directly as a working fluid in a gas turbine. Exiting the reactor at some 1,650°F, the helium enters the turbine section of the horizontal single- shaft drivetrain. Operating at 6,000 rpm, the turbine drives both the gas turbine compressor and a speed reduction gearbox, connected to the 165 MW electrical generator.

In the more usu al open-cycle gas turbine, the turbine exhaust would exit to the atmosphere. But the PBMR runs in a closed loop, so the helium leaving the turbine first passes through a recuperator to transfer some of its heat, then into a pre- cooler and on to the low-pressure compressor. From there, it passes through an intercooler to lower its temperature and its need for compressive work, before entering the high compressor. At the exit to the high compressor, helium pressure is back up to 1,300 psi, having experienced an overall 3.1 pressure ratio from high and low compressors. Helium then completes one Brayton cycle circuit as it passes through the recuperator (picking up heat from the turbine exhaust) and back into the top of the reactor vessel.

Some of the specific design details of the turbine are not publicly available, due to contractual agreements between PBMR Ltd. and Mitsubishi Heavy Industries , which has the design, development, and manufacturing responsibilities for the gas turbine turbomachinery. It is known that the gas turbine components will have oil lubricated bearings (instead of gas or electromagnetic) and the shafts will use dry gas seals. Other details will undoubtedly emerge as prototype plant testing takes place in Koeburg.

Ambitious Plans, High Stakes

My nuclear tour of South Africa took my wife and me to Potchefstroom, a town on the veldt two hours south west of Johannesburg. "For watch dogs at our home, we use three ostriches," reported a woman at breakfast in the hotel there. "They are big, tall-and quiet."

Potchefstroom is the home of North West University, where Professor Pieter Rousseau , with his students and colleagues, has for some years done experimental and computer modeling of the PB Ml system to predict its dynamic behavior. A 165-foot-long micromodel of the PBMR has been constructed and tested extensively, using an electric heater to emulate the nuclear reactor and off the- shelf components to model compressors, turbine, and heat exchangers. With the help of Flownex, a Potchefstroom engineering firm, two large, elaborate test facilities have been constructed at the university to carefully determine detailed design information on the high- temperature heat transfer and high-pressure flow characteristics of helium in pebble bed configurations.

Another facility in Pelindaba tests full-scale structures and components for their ability to work with high-pressure helium. Project engineer Kevin Chetty told me that handling helium gas at high pressures in the 10- story rig is a challenge; helium is second only to hydrogen in its ability to leak. The PBMR power plant Will have welded piping, but the helium facility test rig has many bolted flanges, all of which provide opportunities for leaks.

All reactor project work is coordinated at PBMR Ltd. headquarters in Centurion. During my visit in July, Johan Slabber, P13MR's chief technology officer, pointed out the built-in safety features of the pebble bed modular reactor for a loss-o f- coolant accident. In the event of a c0111plete shutdown of helium flow in the pebble bed, the temperature would rise at most to 2,900°F, a level well below the thermal limit of the graphite pebbles. At the higher temperature, the more plentiful uranium-238 nuclei absorb mo re neutrons (due to an effect call ed Doppler broadening) and the reactor output decreases, lowering the reactor temperature until an equilibrium is reached. T he reactor heat is transferred passively by radiation, conduction, and natural convection to the steel reactor vessel, which is designed to reject the heat without human intervention.

The plans are ambitious, and the stakes are high. If the pebble bed reactor lives up to its promise, it would provide safe, carbon- free energy with high reliability at a relatively low cost. It would also be a posthumous vindication to some of the first generation of nuclear power developers who saw great promise in the gas- cooled pebble bed design.

After his work on nuclear reactors at the University of Chicago, Farrington Daniels returned to the University of Wisconsin and, in the 1950s, became one of America's first experts on the practical utilization of solar energy. One can say that once Daniels was unsuccessful in promoting his invention of a new type of fission nuclear reactor, he turn ed his attention and talents to practically harnessing fusion-at least the fusion reactions at the core of the sun, mankind's first nuclear reactor.