Nuclear power in Brazil

New-build plans

Brazil’s national energy plan to 2030 has proposed five different options for the development of electrical power. They predict total nuclear power in 2030 ranging from 7.3GW to 11.3GW, up from 2.0GW currently. Even the most conservative case calls for the completion of Angra 3, and the construction of four 1000MW new nuclear power plants, two in the northeast, and two in the southeast.

Although construction of Angra 3 began in 1983, the only site work carried out before construction halted was some excavation. Many key components were bought with Angra 2, and remain in storage. This stockpile includes the reactor vessel, steam generators, pressurizers and cooling water pumps, turbine-generator sets and pipework.

The project was resurrected in 2005, and the Areva NP 1400MWe PWR is now scheduled to be connected to the grid in 2015. The project has received two partial licences for construction, the latest for class 2 (non-nuclear building construction) in March 2010. As of January 2009, BRL1.6 billion ($760 million) has been spent, 40% on equipment and site work, the rest on equipment storage, financial costs and other charges related to stopping the construction. The total estimated cost of building Angra 3 is BRL8.3 billion.

Existing contracts were with Areva for NSSS, Andrade Gutierrez for civil works, Engevix for part of civil design structures, and a consortium of Confab, Bardella and Nuclep for national supplies. Services still required to finish the project are engineering, construction management and commissioning. Supplies still required include instrumentation and control, diesel generators and switchgear, process equipment for auxiliary systems, and piping material. It estimated that about seven months would be spent on site preparation, and 66 months on construction and commissioning.

Eletronuclear, a wholly-owned subsidiary of Brazilian state electricity company Eletrobrás, is carrying out Angra 3 development. For the selection of new-build sites, it is following EPRI’s ‘Siting Guide: Site Selection and Evaluation Criteria for an Early Site Permit Application’ (2002, no. 1006878). This proposes a four-stage process: excluding inappropriate sites, avoiding areas of high risk, identifying suitable areas (from 20 potential sites as of December 2009 to 10 primary sites), and a final selection process of five sites open to local communities and government. Risks that rule out sites fall into four categories: health and safety, environment, socioeconomic factors, engineering/cost factors.

Four coastal states are in the running for the northeast new-build locations: Pernambuco, Alagoas, Sergipe and Bahia. Each site would host up to six units, for economies of scale in licensing, infrastructure, construction, operation, maintenance and waste management. Site selection for the southeast is set to begin in 2010, although media reports in January pointed to inland state Minas Gerais.

One of four selected states in the northeast of the country will host a new nuclear power plant

Reactor types in the frame include the Atomstroyexport VVER 1000, Areva/MHI Atmea (1000MWe), Areva EPR (1700MWe), KHNP APR-1400, MHI US-APWR (1700MWe) and the Westinghouse-Toshiba AP1000 (1200MWe).

The government will share investment in new power stations with the private sector, as it has done with other units. Angra 2 is 50.4% state-owned; Angra 3 is 54%. Future units would be 70% state-owned, according to Eletronuclear. Its estimates for new plant production range from BRL9.9 billion ($5.6bn) to BRL14.2 billion ($8.05bn).

Fuel plans

Brazil boasts of 309 million tonnes of U3O8 in actual reserves, making it the sixth-largest reserve worldwide. lt currently produces about 400t/year of U3O8, extracted and refined mostly from the Bahia-state Caetité mine, according to government body Indústrias Nucleares do Brasil (INB). Planned expansions there would increase output to 800t/year by 2012, and to 1200t/year by 2014. At the same time, a new phosphate mine at Santa Quitéria in northeastern Ceará state would open in 2012, producing 1100t U3O8/year in 2012, rising to 1600t by 2017. In total, Brazil aims to produce 2700t/year by 2017. At the same time, it estimates that national consumption will be 500t-1400t per year between 2011-2025 (depending on first fuel loads and refuelling), so there is a significant potential for export market development. According to Eletronuclear, the expansion is being funded by BRL72 million ($40m) of state money and BRL640 million ($360m) of private investment.

Brazil's U3O8 supply and demand predictions show export potential

Brazil currently has no commercial capacity to convert yellowcake into uranium hexafluoride for enrichment – which is why it recently signed a five-year conversion services contract with Areva. INB is planning to open its first industrial plant in 2014, with a capacity of 1200t UF6/year, which would make the country self-sufficient. It plans to double this capacity by 2018.

Currently Brazil has little commercial enrichment capacity, but a massive BRL3.4 billion spending plan in the 2009-25 period would boost the capacity of uranium centrifuges at the Angra site to about 100 million SWU in 2012. An additional 100 million SWU would be added each year thereafter, bringing the total capacity to 1400 million SWU in 2025. The plans include building a new centrifuge factory, assembly and testing facility and new supporting infrastructure, according to Eletronuclear.

The fuel cycle is controlled by INB, a subsidiary of the Brazilian nuclear energy commission (CNEN), itself part of the government’s ministry of science and technology, a sister department to the unit that holds Eletrobrás, the government’s ministry of mines and energy. The third nuclear branch of the Brazilian government, the ministry of defence, controls research institutes, including Navy facilities where much of the country’s fuel cycle research is being carried out.

Waste plans

Brazil is in the midst of plans for dealing with radioactive wastes. According to environment ministry CONAMA resolution no. 31 of 2001, and national council of energy policy CNPE Resolution no. 5 of 2001, plants are granted an environmental license assuming ‘a long-term solution for ILW and LLW generated by the Angra 1, 2 and 3 plants shall be in effect by the time Angra 3 goes online’.

A location for a national waste repository for ILW and LLW is due to be chosen in 2011, and start operation in 2018. Responsibility for site selection, design, as well as for the construction and operation of the final repository is the responsibility of national nuclear energy commission CNEN. Two options are being considered: the construction of a repository exclusively for waste from Angra or a facility that would accept material from all nuclear and radioactive installations in Brazil.

The final destination for high-level waste will depend on a future decision about fuel reprocessing. This decision will take into account the contemporary political and economic situation in 20-50 years when the useful life of these plants has expired.

High-level wastes would initially be stored in pools near the reactors, and then moved to an intermediate storage location for 500 years. This intermediate site is expected to begin operation in 2026; a proposed plan was due to be finished by 2009, and a prototype validated by 2013, according to Eletronuclear.