SMRs-what are their prospects?

As nuclear power generation has become established since the 1950s, the size of reactor units has grown from 60 MWe to more than 1600 MWe, aimed at achieving corresponding economies of scale in operation. At the same time there have been many hundreds of smaller power reactors built both for naval use (up to 190 MW thermal) and as neutron sources, yielding significant expertise in the engineering of small units. Today, due partly to the high capital cost of large power reactors generating electricity via the steam cycle, and partly to the need to service small electricity grids, there is a renewed move to develop smaller units. These may be built independently or as modules in a larger complex, with capacity added incrementally as required. Economies of scale are essentially provided by the numbers produced. There are also moves to develop similar small units for remote sites.

The International Atomic Energy Agency (IAEA) defines ‘small’ reactors as under 300 MWe, and ‘medium’ as up to 700 MWe. The classification of these capacities as alternatives to the mainstream is a modern thing; the categories would include many operational units from the 20th century. Together they are now referred to as small and medium reactors (SMRs). The abbreviation ‘SMR’ is sometimes also, rather confusingly, used to refer to small modular reactors, but SMRs are not necessarily of modular construction. Modern small reactors for power generation are expected to have greater simplicity of design, economy of mass production, and reduced siting costs. Most are also designed for a high level of passive or inherent safety in the event of malfunction. It is possible that many safety provisions necessary, or at least prudent, in large reactors may not be necessary in the small designs forthcoming.

Some existing operating reactors are already within the SMR category, including the Indian 220 MW pressurised heavy water reactors (PHWRs) based on Canadian technology, and the Chinese 300-325 MW PWR as built at Qinshan Phase I and at Chashma in Pakistan (now called CNP-300). The Nuclear Power Corporation of India (NPCIL) is now focusing on 540 MW and 700 MW versions of its PHWR, and is offering both 220 and 540 MW versions internationally. There are also four small 62 MW units operating at the Bilibino co-generation plant in a remote area of Siberia, of an unusual graphite-moderated boiling water design.

Looking ahead, the most advanced small (and modular) reactor project is in China, where Chinergy is preparing to build the 210 MW HTR-PM. In South Africa, the Pebble Bed Modular Reactor (Pty) Limited and Eskom were developing the pebble bed modular reactor (PBMR), but funding for this project has been stopped.

Looking further ahead, there is currently a lot of publicity and evident enthusiasm for a variety of new SMRs. Many are PWRs, such as the Russian floating nuclear power plant, the CAREM reactor being developed by INVAP in Argentina, South Korea’s SMART, Westinghouse’s IRIS, Babcock & Wilcox’s mPower and the NuScale reactor. Others are fast neutron reactors like the Hyperion Power Module and the Travelling Wave Reactor (TWR). There is obviously an interface between these latter units and the Generation IV and INPRO reactor designs, generally not expected to come into operation until the mid-2020s.

Despite the proliferation of proposed SMR designs, none have yet applied for certification by the Nuclear Regulatory Commission (NRC) in the United States and it says that it expects to receive its first SMR design certification application in 2012.

In principle, the idea that small size may allow mass manufacture in a factory, enabling considerable savings relative to field construction and assembly that is typical of large reactors, is very attractive. SMRs, particularly those of modular construction, could be cheaper because they will be more like assembly-line cars than hand-made Rolls Royces. From the financing viewpoint, it is also preferable not to have to put forward so much money at an early stage in the project, and the financial returns from the initial units can help finance subsequent reactors.

SMR developers outline the economic case for modular construction of their designs and point out that the small size and simple design of their reactors are ideally suited for modular construction in the sense of progressively building a large power plant with multiple small operating units. The economies of scale are replaced here with the economy of serial production of many small and simple components and prefabricated sections. They generally expect that construction of units will be completed in under three years, with subsequent reduction to only two years.

Site layouts have been developed with multiple single units or multiple twin units. In each case, units will be constructed so that there is physical separation sufficient to allow construction of the next unit while the previous one is operating and generating revenue. In spite of this separation, the plant footprint can be very compact. A site with several single modules providing 1000 MWe capacity is similar to or smaller in size than one with a comparable total power single unit.

Eventually SMRs should have a capital cost and production cost comparable with larger plants. But any small unit such as these will potentially have a funding profile and flexibility otherwise impossible with larger plants. As one module is finished and starts producing electricity, it will generate positive cash flow for the next module to be built. Westinghouse estimates that 1000 MWe delivered by three IRIS units built at three year intervals financed at 10% for ten years require a maximum negative cash flow less than $700 million (compared with about three times that for a single 1000 MWe unit). For developed countries, small modular units offer the opportunity of building as necessary; for developing countries it may be the only option, because their electric grids cannot take 1000+ MWe single units.

There may, however, be some factors that neutralise these advantages and make the costs per kilowatt of small reactors higher than large reactors. First, in contrast to cars or smart phones or similar widgets, the materials cost per kilowatt of a reactor may go up as the size goes down. This is because the surface area per kilowatt of capacity, which dominates materials cost, goes up as reactor size is decreased. Similarly, the cost per kilowatt of secondary containment, as well as independent systems for control, instrumentation, and emergency management, increases as size decreases. Cost per kilowatt also increases if each reactor has dedicated and independent systems for control, instrumentation, and emergency management. For these reasons, the nuclear industry has been building larger and larger reactors in an effort to try to achieve economies of scale and make nuclear power economically competitive.

Proponents also argue that because these nuclear projects would consist of several smaller reactor modules instead of one large reactor, the construction time will be shorter and therefore costs will be reduced. Assuming the intention is to build a series of SMRs on one site, the realisation of economies of scale would depend on the construction period of the entire project, possibly over an even longer time span than present large-reactor projects. The cash flow profile would, however, be more favourable. But if the later-planned units are not built, for instance due to a slower-than-expected demand growth, the earlier units could be more expensive than present reactors.

These possible cost increases may be offset if the entire reactor is manufactured at a central facility and some economies are achieved by mass manufacturing compared to large reactors assembled on site. Critics will argue, however, that estimates of low prices must be regarded with scepticism due to the history of past cost escalations for nuclear reactors, and the potential for cost increases due to requirements arising in the process of licensing. Some SMR designers are proposing that no prototype be built and that the necessary licensing tests be simulated. Whatever the process, it will have to be rigorous to ensure safety.

Mass manufacturing will also raise a host of new safety, quality, and licensing concerns that the regulators have yet to address. For instance, they may have to devise and test new licensing and inspection procedures for the manufacturing facilities, including inspections of welds and the like. Other issues that will affect safety are regulator’s requirements for operating and security personnel, which have yet to be determined. To reduce operating costs, some SMR vendors are advocating lowering the number of staff in the control room so that one operator would be responsible for controlling several modules.

Proponents claim that with longer operation on a single fuel charge and with less production of spent fuel per reactor, waste management would be simpler. Used fuel management for SMRs could, however, be more complex, and therefore more expensive, because the waste would be located in many more sites. The infrastructure that we have for spent fuel management is geared toward light-water reactors at a limited number of sites. In some proposals, the reactor would be buried underground, possibly making waste retrieval even more complicated.

Despite these possible downsides, it is clear that the nuclear industry has to investigate seriously alternative business models to the large light water reactors. Their escalating costs, together with the weaker world financial situation, have led to a slowing of the nuclear renaissance in most of the developed nuclear world. While China races ahead with a huge reactor programme, seemingly unaffected by the negatives experienced elsewhere, expectations of new reactors in the United States and parts of Europe have been downgraded. The hope is that the experience of mass construction of large reactors in China, and then in India, allows the costs of new reactor projects to come down in the remainder of the world, hopefully at the same time as there is a firmer economic recovery.

The key advantage of the SMRs may eventually turn out to be merely their small size and suitability for smaller grids, factors which will determine how they are likely to fare in developed nuclear markets. The key factor in these may turn out to be how the regulatory process works out. If it cannot be streamlined for SMRs, it is likely that the regulatory cost will effectively be a fixed cost which will be difficult to bear for smaller units. Most of these markets require a large amount of power, either to meet incremental demand or replace retired units, and will therefore still try to maximise reactor size, maintaining the strong trend of the past 10-20 years.

The IAEA now has a list of 65 potential new nuclear countries, many of which are unlikely to have electricity grids suitable for large reactors even in ten years’ time. Even those new countries with well-established infrastructure, such as Singapore, are looking at SMRs because their characteristics may be more favourable for markets that have previously rejected the large units.

Once more development work is completed with the SMRs, it will be interesting to see if many new countries can pass through all the necessary regulatory, human capital and technical hoops that are necessary to obtain nuclear power. Some will argue that it is necessary for the developed nuclear nations to start building large numbers of new traditional reactors before nuclear can spread to many of these new countries, but the SMRs offer an alternative way which may prove attractive.

Steve Kidd is director of strategy & research at the World Nuclear Association, where he has worked since 1995 (when it was still the Uranium Institute). Any views expressed are not necessarily those of the World Nuclear Association and/or its members.

Post to: Delicious Digg reddit Facebook StumbleUpon Tweet Linked In