The South African Nuclear Energy Corporation (Necsa) is this year celebrating the fiftieth anniversary of its SAFARI-1 research reactor, the country’s – and Africa’s – first nuclear reactor. It first ‘went critical’ (achieved critical mass, and so started operating) on March 18, 1965 – at 18:33 (or 6:33 pm) to be precise. Originally, SAFARI was an acronym for South African Fundamental Atomic Research Installation, but that full name has not been used for decades now.

SAFARI-1 is a pool-type reactor, in which the core is immersed in highly purified (and crystal clear) water which acts as a shield, moderator and coolant all at the same time. The sides of the reactor are constructed of concrete and lead. Its design is based on that of a research reactor built at the Oak Ridge National Laboratory, in Tennessee, in the US. Originally, SAFARI had a maximum output of 6.66 MW, but, in 1968, modifications to the cooling system allowed this to be increased to 20 MW.

Past . . .

“The context of the building of SAFARI-1 in South Africa – it was almost like a gift from the US in return for South African assistance in their uranium programme, around 1944/45, as part of the Manhattan Project. That’s my impression,” says Necsa CEO Phumzile Tshelane. (The Manhattan Project was the programme that developed nuclear weapons during the Second World War.) “[SAFARI-1] was also an outgrowth of America’s Atoms for Peace programme and of the creation of the International Atomic Energy Agency (IAEA) in 1957.”

South Africa was a founder member of the IAEA. In 1948, the country created the Atomic Energy Board (AEB), the original precursor to today’s Necsa. In 1959, the country launched its Nuclear Research and Development (R&D) Programme and, in 1961, the Pelindaba site, west of Pretoria, was chosen to host the planned nuclear research facility, centred on the then planned research reactor. The first personnel moved into offices at Pelindaba in 1963.

“Essentially, for us, it allowed South Africa to enter the realm of the application of nuclear science and nuclear reactor science and increase the skills development of our people,” he points out. When built, and for four decades afterwards, SAFARI-1, like most research reactors around the world, used highly enriched uranium (HEU) as its fuel.

The retired MD of Necsa subsidiary company NTP Radioisotopes, Don Robertson, has divided SAFARI-1’s life so far into three major and overlapping phases. These are: the R&D Era, the Strategic Era and the Commercial Era.

The R&D Era ran from the early 1960s to the mid 1970s. South African scientists and engineers used the reactor to carry out research in a wide range of areas. This included nuclear physics, reactor physics, reactor and radiation safety, materials irradiation and transmutation, industrial and medical isotope production, industrial applications of radiation, fundamental and applied materials science, geological and environmental sciences and medical and health physics.

This phase came to an end as a result of international, especially American, hostility to the then South African domestic policy of apartheid. “In 1976, the export of fuel [for SAFARI-1] from the US to South Africa was terminated,” notes Robertson. “In 1977, SAFARI-1’s power level was reduced to 5 MW and it was operated only on alternate weeks. In 1978, a project to locally manufacture fuel was initiated and, in 1982, the first locally produced fuel was loaded into the reactor, but, thereafter, its utilisation reached an all-time low.”

The Strategic Era ran from the mid 1970s to the early 1990s. In 1982, the AEB was merged with the Uranium Enrichment Corporation (created in 1970) to form the Atomic Energy Corporation (AEC). In this period, the focus was on the development of the local production of high-quality fuel for the Koeberg nuclear power plant (near Cape Town), which started delivering electricity to the national grid in 1984. Another, very secret, project was the development of nuclear weapons. SAFARI-1 was pretty much irrelevant to, and not involved with, either of these programmes. This period represented the low point in SAFARI-1’s history.

The Strategic Era ended with political transformation in South Africa. “In the early 1990s, sanctions were lifted,” points out Robertson. “But there was a drastic reduction in the AEC’s budget. There was the closure of the nuclear fuel production plants, which down-sized the organisation. However, the Nuclear Non-Proliferation Treaty was signed in July 1991.”

At that time, the nuclear infrastructure and expertise possessed by the AEC at Pelindaba and elsewhere was extensive. It possessed a fuel manufacturing plant (closed as not being economically viable once sanctions were lifted), an inventory of HEU, depleted uranium, waste disposal facilities and manufacturing workshops (subsequently down-scaled but now being redeveloped). These were in addition to SAFARI-1, which was underused, and its associated hot cell facility (hot cells allow researchers to safely manipulate radioactive materials), along with the ability to design, build and operate hot cells. Finally, there were the skills that had been developed – theoretical and experimental reactor physicists and isotope separation chemists, for example.

It was in this period that there emerged a threat to the continuing existence of the reactor. Some in government felt that SAFARI-1 was expensive to operate and that it was inappropriate to spend money on it. As a result, a SAFARI Overview Panel was set up in 1995. Fortunately, the third phase of the reactor’s life to date, the Commercialisation Era, had already started. The panel recognised this and recommended that the reactor be kept in operation but stated that 30% of its operating costs should be met from its commercial activities. The reactor made significant progress within a single year and it was decided not to shut it down.

Central to commercialisation was the production of radioisotopes (although the reactor undertakes other commercial operations as well). This activity can be traced back to 1973 when SAFARI-1 began to supply Iodine-131 in South Africa. In 1974, it also made Technitium-99M available to the local market. A dedicated isotope production facility was established at Pelindaba in 1977. In the 1980s Caesium-137 and Iridium-192 sealed sources were added to the product list. By the 1990s, the reactor was responsible for 90% of the radiopharmaceuticals used in South Africa.

The 1990s also saw the start of the Molybdenum-99 (Mo-99) project. “This is the most widely used isotope and is used in 40-million diagnostic procedures per annum worldwide,” affirms Robertson. “Mo-99 production requirements are enriched uranium, the fabrication of target plates, the design of irradiation rigs, a reactor, hot cells, chemical extraction processes, containers, waste handling and disposal facilities and the necessary expertise and skills. The project rapidly grew into a business that serves the global market and today some 20% of the global demand for Mo-99 is supplied from SAFARI-1.”

In 2009, a successful project was carried out, with US assistance, to convert SAFARI from HEU fuel to low-enriched uranium fuel. This served as the proof of concept for such a conversion, proving that SAFARI could continue to fulfil all its roles with the new fuel. Since then, the US has assisted other countries to carry out similar conversions.

. . . Present . . .

“Today, SAFARI is being used for the production of radioisotopes as well as neutron-related research and development and other neutron-related activities, such as neutron irradiation transmutation of silicon ingots, for applications in the electronics industry,” explains Tshelane. “It now operates at close to 20 MW on a continual basis. SAFARI-1 is available more than 300 days a year.”

The neutron irradiation of silicon ingots is another major commercial activity. “This process alters the resistivity of an ultrapure silicon ingot. Homogeneity throughout the ingot is critically important,” explains Robertson. “The quality of the SAFARI product is on a par with the best in the world. We have contracts with semiconductor companies in Europe and Japan.”

“It is the most highly commercially utilised research reactor in the world today,” he highlights. “It is the cornerstone of the South African isotope production programme. It has contributed largely to the success of NTP Radioisotopes.” (The reactor produces the nuclear elements used to manufacture the radiopharmaceuticals, which is done by NTP Radioisotopes, also at Pelindaba.)

Between eight- and ten-million medical procedures a year, in more than 60 countries, are carried out using radioisotopes produced by SAFARI-1. Millions of lives have been saved, worldwide. It is one of only 18 operational research reactors in the world today with an age of 50 years or more. But, thanks to 15 years of operating at a low level, it is the only one that has not suffered operational problems and even lengthy closures. SAFARI-1 has an excellent safety and operational record and has a well-funded maintenance programme.

The dedicated reactor crew is composed of some 55 to 60 scientists, engineers and operators. “The radioisotope business brings in just about R1-billion per annum and about 40% of that is accounted for by SAFARI-1,” reports Tshelane. “This business certainly justifies replacing it!”

. . . and Future

And replace SAFARI-1, before it wears out, is certainly what Necsa plans to do. “We hope SAFARI-1 will continue to operate until 2030,” Tshelane avers. “We’re not just hoping, we’re doing age management of some of the components to ensure it does. For strategic reasons, we are planning to replace SAFARI-1 and we need to find innovative ways to finance it, rather than use a conventional ‘business case’ approach. This is because the need for a large R&D component in the use of the future reactor makes a conventional business case approach difficult.”

Necsa wants the successor reactor to have greater capabilities than SAFARI-1, so that it can be used to support the country’s planned programme to build new and extra nuclear power plants. A timeline for the new research reactor has been sketched out, but it is not rigid. “Between 2023 and 2024, we should have a new multipurpose [research] reactor,” asserts Tshelane. “Assuming a four-year construction period, construction must start in 2019. So the decision must be made next year to allow the procurement of long-lead items, precise site selection (within the Pelindaba complex) and licensing.”

Necsa has developed a user specification for the new reactor (which will not be called SAFARI-2, to avoid the idea of a continuous series). One new capability, for example, for the new unit would be the ability to irradiate the fuel of pressurised water reactors (used in nuclear power plants) to test it. The corporation hopes its requirements will be met by a modified off-the-shelf design to reduce costs. Because of its unique requirements, Necsa cannot simply take a completely off-the-shelf option. “We think of it not just as a reactor. We think of it as a radiochemical complex,” elucidates Tshelane. “The core of the radiochemical complex will be an off-the-shelf reactor, but with specific systems for the complex – but, as few of these as possible, because we do not want to be guinea pigs! The more you customise, the more you add risks.”

The new reactor will be put in its own, new building. In due course, Necsa will send out requests for information to potential suppliers – Argentinian, French, Korean and Russian. Some have more than one design to offer.

There will be a period in which the two reactors will be running simultaneously. But once the new reactor is established in full operation, SAFARI-1 will be decommissioned, then totally dismantled and its building decontaminated. “We hope to reuse the building,” says Tshelane. The remaining radioactive parts will be securely contained for final disposal. Decontaminated parts will be sold as scrap. Components that are still contaminated will be treated as high-level waste. This entire process (including the reuse of a research reactor building) has been successfully carried out in Australia.