Figure 1. Fuel element design for PBMR
Click here to enlarge image

In 1993, South African utility Eskom began investigating Pebble Bed Modular Reactor (PBMR) technology as a possible option for future South African electricity supply. In April 2000, the country’s government gave its approval for a detailed feasibility study, an environmental impact assessment (EIA) and a public consultation process for a PBMR demonstration project.

The detailed feasibility study (DFR), designed to determine the technical, commercial and economic feasibility of the PBMR, is nearing completion. If the project runs to plan, construction of the 120 MWe thermal demonstration module at Koeberg, near Cape Town, will begin in mid-2002.

When complete, the DFR will be submitted to the South African government for review by an independent team of experts. The project will also require shareholder approval, an EIA, a construction license by the National Nuclear Regulator, and government consent. If construction begins on schedule in 2002, the first PBMR will be completed by the beginning of 2005 and operational by 2006. Commercial operation is forecast for about four years later.

Project progress

Figure 2. The PBMR power conversion unit is based on the standard Brayton (gas turbine) cycle
Click here to enlarge image

Eskom began to look at the possibility of developing PBMR technology as part of its Integrated Electricity Planning process in the early 1990s. While the country currently has excess power generating capacity, the utility estimates that new capacity will have to be commissioned by 2008, and that reserve margins could be depleted by 2010.

In addition, Eskom’s older power stations will reach the end of their design life after 2025, and many of them are located close to inland coal seams while the growing demand centres are located on the coast.

Eskom therefore believes that PBMR technology could play a part in the country’s future electricity supply. In 1995, Eskom commissioned a pre-feasibility study followed by a techno-economic study in 1997. By mid-1998 the project had progressed to the point at which it had entered the full-scale engineering design phase.

The first phase of the project, involving the DFR and the EIA, was given the go-ahead by the South African government in April 2000.

The PBMR project is being developed by PBMR (Pty) Ltd., an independent, unincorporated research and development company in which Eskom is the major shareholder with a 30 per cent stake. The other stakeholders are the IDC (25 per cent), the UK’s BNFL (22.5 per cent) and Exelon of the USA (12.5 per cent). Ten per cent is reserved for black empowerment investment.

In spite of growing opposition to nuclear power in some parts of Europe, PBMR Ltd. believes that there is a strong market for the technology around the world.

Driving a revival of the nuclear industry are government policy in the USA, concerns over fuel diversity and security in Europe, recent oil price volatility and the need to reduce greenhouse gas emissions.

The involvement of Exelon, a US utility with a market capitalization of $28 billion, could accelerate the marketing process. Exelon has already indicated that it may want to apply for a license to construct PBMRs in the USA as early as 2004.

Modular design

Click here to enlarge image

The design of the PBMR is based on the philosophy that the new generation of nuclear reactors should be small, modular and inherently safe. Their modularity allows flexibility, making it possible to build plants to serve local needs and expand them as demand grows. The modular design also allows operational flexibility, i.e., power output can be matched to grid requirements and daily load patterns.

The flexible output design criteria call for each unit being able to move from half power to full power to half power in a matter of minutes. When the possible variations for a single module are combined into a multi-unit plant, it is easy to see how the PBMR could closely emulate a gas-fired unit for purposes of meeting demand ‘peaks’ and ‘shoulders’.

Plant modularity also has great significance in terms of possible market penetration. Even in ‘big grid’ places like Japan, North America and Northern Europe, there are always pockets where transmission assets are slim. While it may not be feasible to build small-scale light water reactors (LWR) near these underserved load pockets, PBMRs could fit these local load requirements very well.

High temperature design

The Pebble Bed concept uses coated uranium granules as a fuel in a high temperature helium cooled reactor coupled to a gas turbine cycle to convert heat energy to electric power. The concept was first developed in Germany, where a research reactor was constructed and operated for 21 years.

The PBMR consists of a vertical steel pressure vessel, 6 m in diameter and 20 m high. It is lined with a 100 cm-thick layer of graphite bricks, which serves as a reflector and a passive heat transfer medium. The graphite brick lining is drilled with vertical holes to house the control rods.

The graphite reflector encloses the core, which is 3.5 m in diameter and 8.5 m high. Its outer zone contains 330 000 fuel spheres and its inner zone contains 110 000 solid nuclear grade graphite spheres which serve as an additional nuclear moderator.

The nuclear reaction in the outer core is controlled by the control rods housed in the side reflector. Helium gas flows through the pebble bed and removes the heat generated and transfers it to the closed gas turbine generator cycle.

Helium coolant enters the reactor vessel at a temperature of about 500°C and a pressure of 70 bar. It then moves down between the hot fuel spheres, after which it leaves the bottom of the vessel having been heated to a temperature of about 900°C.

The gas then enters the first of three gas turbines, the first two of which drive compressors and the third of which drives the electrical generator. The coolant leaves the last turbine at about 530°C and 26 bar, after which it is cooled, recompressed, reheated and returned to the reactor vessel.

The process cycle used is a standard Brayton cycle with a closed circuit water-cooled inter-cooler and pre-cooler. A high efficiency recuperator is used after the turbine generator to recuperate thermal energy. Lower energy helium is passed through the pre-cooler and intercooler and the low and high pressure compressors before it is returned through the recuperator to the reactor core.

A heat efficiency of over 40 per cent is anticipated in the basic PBMR design. Increases in fuel performances, leading to higher operating temperatures, offer the prospect of up to 50 per cent efficiency.

Fuel spheres

The PBMR uses silicon carbide and pyrolitic carbon coated particles of enriched uranium oxide encased in graphite to form a fuel sphere or pebble about the size of a tennis ball. The pyrolytic carbon and silicon carbide layers provide an impenetrable barrier designed to contain the fuel and the radioactive decay products resulting from the nuclear reactions.

The fuel elements are machined to a uniform thickness of 60 mm. Each fuel sphere contains 9 g of uranium, which means that the total uranium in one fuel load is 2.79 t. The total mass of a fuel sphere is 210 g.

When a fuel sphere has reached a burn-up of 80 000 MWd/T of uranium metal, it is removed and sent to the spent fuel storage facility. Each fuel pebble passes through the reactor about ten times and a reactor will use ten to 15 total fuel loads in its design lifetime.

The extent to which the enriched uranium is used to depletion is much greater in the PBMR than in conventional power reactors. There is therefore minimal fissile material that could be extracted from depleted PBMR fuel. This, coupled with the level of technology and cost required to break down the barriers surrounding the spent fuel particles, protects the PBMR fuel against the possibility of nuclear proliferation.

The fuel is transported to the spent fuel storage facility in the reactor building by means of a pneumatic fuel handling system. The spent fuel storage consists of ten tanks, and one tank can store up to 500 000 spheres. Storage is dry at all times, with decay heat carried away by natural convection. There is no requirement for pools or pumps, or a risk of contaminating water supplies.


The design target construction cost for a commercial scale PBMR plant, consisting of ten modules with a total output of about 1200 MWe, is about $1000 per KWe installed, compared to $900 per KWe for a new coal-fired power station in South Africa.

If this target is achieved, PBMR’s output cost should be lower than a coal-fired plant at the South African coast and well below the world average cost of $0.034/kWh.

The EIA and public participation process started in June 2000. The current target date for a record of decision from the South African government on the EIA is early February 2002. Environmental challenges centre largely on the disposal of spent fuel.