Beyond the copper surface2 April 2019
Before emplacement in a deep geological repository, Canada, Finland and Sweden all plan to encapsulate their spent fuel in copper canisters. Kristina Gillin drills into the countries’ spent fuel programmes and explores similarities and differences.
THREE COUNTRIES WITH ADVANCED PLANS for long-term management of spent nuclear fuel have opted for the same multi-barrier concept. Canada, Finland and Sweden plan to encapsulate fuel bundles in reinforced copper canisters, surrounded by bentonite clay and place them in a repository located approximately 500m deep in the bedrock.
The three countries have many prerequisites for implementing a spent fuel repository in common, so this is hardly surprising. The rock types are similar. Seismicity is low. They are at latitudes that require the repositories to withstand the next ice age. And all have vast experience in mining and metal industries.
At the same time, there are differences. In all three countries, separate waste management organisations have been established. Canada’s Nuclear Waste Management Organization (NWMO) and Finland’s Posiva are responsible only for long-term management of spent fuel, whereas the Swedish Nuclear Fuel and Waste Management Co (SKB) also has responsibility for interim storage and managing low- and intermediate-level waste. Funding is, in all three cases, secured through segregated trust funds, which are funded by the reactor owners.
Spent fuel inventory and interim storage
Power production in full-scale reactors began in all three countries in the 1970s. Canada has 22 large heavy-water reactors, while the light-water reactors in Sweden and Finland total 12 and five, respectively (including Finland’s Olkiluoto 3, which is undergoing commissioning).
Given the differences in reactor fleets, significant differences emerge in the countries’ spent fuel inventories. Canada’s inventory is many times greater, largely due to having Candu reactors, which use natural uranium in small fuel bundles. But the low fuel burnup makes it easier to accommodate decay heat and the short fuel assembly length offers more flexibility in terms of handling and storage configuration.
In all three countries there are storage pools directly adjacent to the reactors, and separate facilities for interim storage have been built. In Canada and Finland, these are located at the reactor sites. In Sweden, spent fuel is shipped to central storage, located in rock caverns not far from the Oskarshamn plant.
Deep geological repository
All three countries are pioneers in research and development of long-term management of spent fuel and have extensive records of international collaboration, with each other and with other countries.
Each country has constructed a deep geological research facility for experiments and testing of methods under realistic conditions. Finland’s underground rock characterisation facility, Onkalo, is at the site of the planned repository. In contrast, Sweden’s Äspö Hard Rock Laboratory and Canada’s (now closed) Underground Research Laboratory were built as stand-alone facilities and these countries also have separate laboratories to facilitate in-house technology development related to the spent fuel canisters and bentonite buffer.
The reference concepts in the three countries are strikingly similar, although all are still being refined. One difference is that, in Canada, the canister will be enclosed in bentonite clay boxes prior to being transferred down to the repository. Laurie Swami, president and chief executive officer of NWMO, explains: “A unique feature of our programme is that a lot of the work can be done at surface, in an environment that is more like a factory.”
Once emplaced, the bentonite clay will absorb water and reduce transport of both water and other substances near the canister. It will also act as a mechanical buffer, in case of rock movement. Prior to closure of the repositories, tunnels and shafts will be backfilled.
Despite Canada’s significantly larger current and projected spent fuel inventory, the anticipated footprint at repository depth will only be about twice the size of Finland’s and Sweden’s. This due to the lower burnup and decay heat of Candu fuel, which enables a more compact design (since heat load is a limiting factor for the distance required between canisters in the repository).
Site selection and licensing
To identify a suitable site, a systematic process has been outlined in all three countries. Common denominators include transparency and emphasis on a knowledgeable and willing host community. But the Canadian process stands out, as it was developed collaboratively with the Canadian public prior to beginning site selection.
In the Finnish programme, the siting process was completed 20 years ago. The government has made a decision in principle’ to implement the proposed method at the proposed site. A construction licence – the first in the world for a repository for spent fuel – was issued in 2015. Today, construction is well underway. Pasi Tuohimaa, head of communications at Posiva, says: “The shaft for transfer of canisters is nearing completion and what remains is raise boring the last 100m. We estimate that emplacement of canisters can begin in the mid-2020s, after a licence to operate has been received.”
In Canada, the federal government has also selected a deep geological repository as the method for long-term management of spent fuel. Site selection began in 2010 and 22 communities expressed interest. These have been narrowed down to five, all in Ontario, where most of the Canadian reactors are located. Laurie Swami says: “By 2023, NWMO will have selected a preferred site. After that, we will enter into the regulatory approval phase.”
The siting process in Sweden was completed ten years ago. A licence application for the proposed method and site was submitted in 2011 and is pending government approval. Anders Ström, who is responsible for the development of SKB’s deep geological repository for spent fuel, says: “In a few specific areas, it was deemed necessary to complement our application once more. In particular, related to potential effects of copper corrosion. If deemed sufficient and the government then approves the method and site selection, our next step will be to apply for construction start.”
In Finland and Sweden, the selected repository site is adjacent to a nuclear power plant (Olkiluoto and Forsmark, respectively). Public support is high in both local communities. Posiva’s Pasi Tuohimaa says: “The less you know, the more you fear. But here in Olkiluoto, people understand the nuclear safety culture.”
Runners-up in both Nordic countries were also existing nuclear host communities. In both countries, there was a sense of competition between the last two candidates before the final site selection was announced. In Sweden, this led to a unique agreement while the last two were being assessed: both municipal governments would receive funding for community development, but the value of initiatives would be unevenly distributed; the community selected receives 25% and the runner-up 75%.
Canister design and manufacturing
It was determined early on that copper would be suitable as a corrosion barrier in a spent fuel repository since there is no oxygen at depth in the rock. In Canada, titanium was considered, but copper was selected.
Spent fuel canisters have to withstand high mechanical loads, because during glacial periods, the water-bearing rock formations at these latitudes are subject to pressure from several kilometres of overlaying ice. Proposed designs and manufacturing methods have evolved significantly since the early days.
Tomas Rosengren is responsible for encapsulation at SKB and he says: “At first, our canister design had a copper thickness of 20cm. The void inside the copper tube was, at one time, to be filled with lead. Filling with sand or glass beads has also been the plan. But since we arrived at the design with a cast iron insert inside a 50mm copper tube about 20 years ago, it has stood firm.” The Finnish programme has adopted the same canister design as the Swedish – the only differences being length and configuration of the cast iron insert (to accommodate different fuel types).
In Canada, a similar canister type was considered but a Canadian design has evolved. Given the short length of Candu bundles, NWMO’s reference design is much smaller and lighter than SKB’s and Posiva’s. Key advantages with the Canadian canister include the use of standard steel components and the lack of any gap between the load- bearing structure and the corrosion-resistant layer. Alan Murchison, who is manager fuel handling and sealing system design at NWMO, explains: “Mechanically, the steel tube and copper coating act as one assembly. So even if severely deformed, the canister will retain its corrosion resistance.”
Canisters require high-integrity closure welding. In the Canadian case, the load-bearing steel structure will be welded using a hybrid laser-arc technique. In Finland and Sweden, the welding will be on the copper shell (whereas the cast iron inserts will be bolted). Welding of thick copper had not been done before, so SKB and Posiva had to develop new methods. At first, electron beam welding emerged as the reference method, but getting consistent results proved to be a challenge.
A better option was found in friction stir welding. It had previously been used for thin aluminium plates, so there was extensive testing of different tool designs and materials to adapt the method to welding 50mm-thick copper. Lars Cederqvist, who is responsible for friction stir welding at SKB’s canister laboratory, says: “The welding process is much easier to control. So friction stir welding is very robust, reliable and repeatable.”
Non-destructive testing is another cornerstone of the development activities in SKB’s canister laboratory. Methods include digital radiography, phased array ultrasonics and eddy current arrays. SKB’s Ulf Ronneteg is leading those activities and says: “Welding and non- destructive testing have been developed iteratively, so it has been a major advantage to have the equipment and expertise for both in-house, under one roof.”
Encapsulation and transportation
Apart from the repository, each country will need a plant to encapsulate spent fuel in copper canisters. In shielded cells inside these plants, the fuel bundles will be transferred into canisters, and the canisters welded shut and tested.
The total number of canisters that are projected in the Canadian programme is significantly larger than in the other two so the Canadian encapsulation plant will have a much higher throughput than the others.
In Canada and Finland, the encapsulation plant will be at the repository site, at surface level. The Swedish equivalent will be built at a different nuclear site, as an expansion of the existing central storage facility.
Transport plans vary. In Sweden, spent fuel has been transported routinely since the 1980s from reactor sites to the central storage facility. The ship currently used for these transports, M/S Sigrid, will be used for the transport from the encapsulation plant in Oskarshamn to the repository in Forsmark. In Finland and Canada, new transport systems will be required. They will differ in terms of size and complexity, but in both cases, the transport to the repository is assumed to occur on land.
Author information: Kristina Gillin, Principal consultant at Lloyd’s Register