The KLT-40S is a barge-mounted floating SMR producing 35 MWe per module. The design is a compact loop configuration with most primary system components external to the reactor pressure vessel. The length of the hot and cold leg pipes connecting the reactor vessel with the two steam generator vessels is kept very short to reduce the likelihood of a large-break loss-of-coolant accident and because of space constraints on the barge. The design is based substantially on the successfully operated KLT-40 reactors that provide propulsion to a fleet of commercial ice-breakers.

The 1.2 m diameter by 1.2 m tall reactor core consists of 121 hexagonal fuel

assemblies with uranium enrichment of less than 20%. Control rods and burnable poison rods are used to control reactivity. The four external steam generators are a helical coil type. The four primary reactor coolant pumps are also external to the reactor vessel.

The KLT-40S is normally configured with two units on a 30 m wide by 140 m long barge and will be delivered to the operating site fully constructed and ready for operation. Initial customers include the isolated communities along Russia’s Arctic coast. To extend operations in more remote locations, the barge will be dispatched with multiple core loadings and the reactors will be refueled on-board after three to four years. It is expected that the barge will return to a centralized facility every 10-12 years for major overhauls.

The first KLT-40S floating power plant began construction in 2007 and the barge was launched in the Baltiysky Zavod shipyards in June 2010, prior to installation of the reactor units. The first barge, named the Akademik Lomonosov, is expected to be operational at its deployment site in 2016. Key parameters and a representative graphic for the KLT-40S design are given in Figure 2.7. [7]

A significant number of potential nuclear power plant customers have constraints on the size of allowable and needed increments of power capacity additions, which are smaller than the 1000 MWe and larger ratings of currently offered advanced reactors. The allowable size of additions reflects the somewhat contorted grid layout and interconnections in several US regions. Needed size increments reflect anticipated growth in load demand and incentives to replace older, small generating stations, mostly coal burning, with those using other fuels. As well, since the smaller SMRs should take less time to build than 1000 MWe units, demand forecasts need to be projected for fewer years out than are presently needed. Further markets for small nuclear units are emerging in smaller developing countries which have not previously embarked on nuclear power utilization. In developed countries with well-established nuclear power programs, remote regions and sites vital for national security exist which have power needs that can ideally be supplied by SMRs. Additionally, SMRs in these countries can supply process heat on the scale appropriate to commercial chemical processing plant needs.

These major incentives for SMRs are buttressed by several other desirable factors deriving from the small SMR characteristics:

• effective protection of plant investment from the potential to achieve a reactor design with enhanced safety characteristics;

• possible reduction of the current 10-mile emergency planning zone by virtue of the smaller core inventory and potential for added safety design features;

• reduction of transmission requirements and a more robust, more reliable grid;

• use of components which do not require the ultra-heavy forgings of today’s gigawatt-scale nuclear power plants and are rail shippable which could be supplied by a reinvigorated US heavy industry; and

• suitability for the district heating mission.

The Gas Turbine Modular High-temperature Reactor (GT-MHR) has a long history and got its start in the mid-1980s during the push in the USA to develop smaller reactor designs with assured safety. Since that time, variants of the GT-MHR have been considered for the New Production Reactor program, the Weapons Material Disposition program, and most recently the Next Generation Nuclear Plant program. As such, information sources for the GT-MHR tend to be quite varied with respect to specific parameters of the design, which reflect the different program priorities and constraints. Fundamentally, the design uses TRISO-coated uranium-oxicarbide particle fuel that is dispersed into cylindrical pellets and stacked into vertical fuel channels within hexagonal graphite moderator blocks that are 0.36 m wide (across flats) by 0.79 m tall. The moderator blocks are stacked 10 rows high in 66 columns forming an annular ring. Additional non-fueled moderator blocks fill the center of the annulus and surround the core as an outer reflector. The uranium contained in the uraniun oxycarbide (UCO) fuel kernels is enriched to 15.5% 235U. The heated helium flows to the vertically mounted gas turbine via a concentric cross-duct.

Although the GT-MHR design is relatively mature and has been considered for multiple programs, there is no current commitment to construct a demonstration or

Compressors

Figure 2.18 GT-MHR (United States) — General Atomics (GA).

commercial unit. The most promising project is a joint program between the United States and the Russian Federation to build a demonstration GT-MHR for disposition of weapons-grade material. Key parameters and a representative graphic for the GT-MHR design are given in Figure 2.18. [16]

The LOCA challenges reactor safety by raising peak cladding and fuel temperatures from stored core energy and decay heat generation. The response to this challenge differs among the SMR types as follows — however, for fully integral SMRs no large primary coolant diameter piping exists, thus no classic large LOCAs can occur:

• For the water reactors the primary coolant released flashes in the containment, creating steam expansion which pressurizes the containment and can cause mechanical damage to equipment. While equipment can be secured from this threat and containment can be sized both in volume and wall thickness to survive this threat, the threat of exposure of fuel, even after shutdown of the fission process, requires a means to replenish core coolant inventory. Passive, gravity-driven core reflood systems are the current design vehicle. They must be sized both in delivery head and volume sufficient to rewet the cladding if the fuel is uncovered or simply maintain the cladding wet if the reactor system can be designed to prevent core exposure even during a design basis-LOCA as are all integral PWR-type SMRs. For both situations the sufficiency of core coolant inventory reverses the trend of increasing temperature before the zirconium-based cladding reaches the regulatory limit, now 1204 °C, at which its ductility, and hence its integrity, is threatened. Ultimate removal of decay heat is achieved by means of dedicated decay heat removal loops which transfer heat to the environment or passive conduction heat removal through the reactor containment.

• For gas reactors timely replacement of coolant inventory at pressure is impractical. However, the use of high conductivity graphite as the core moderating material offers a radial conduction path for core energy to an ex-vessel heat sink. The graphite core material provides a significant heat sink which maintains temperatures at allowable levels until passive heat removal capability can match the decay heat level. For cores of modest dimensions the length of this path is short enough and the heat storage capacity of the graphite moderator is large enough to allow steady-state power ratings of hundreds of MWe. These ratings are made possible by the use of coated particle fuel with its high 1600 °C limit for onset of significant fission product diffusion or leakage through disrupted fuel coatings.

• For liquid metal reactors the very low vapor pressure of coolant even at the high operating temperature allows the NSSS to be housed in a pool of coolant within a thin-walled reactor vessel which itself is surrounded by a close fitting thin-walled guard vessel. Even upon loss of integrity of the reactor vessel, the coolant inventory is retained in the guard vessel keeping the cladding covered with coolant and the decay heat is removed by a dedicated in-vessel natural circulation coolant loop and/or radial heat flow through the guard vessel to a dedicated air chimney system, both of which discharge heat outside the containment.

• All three design solutions are satisfactory, although they operate on different principles and have configurations of differing passive safety responses.

In December 2009 the winner of a competitive tender by the United Arab Emirates for a large LWR was announced. The proponent of a higher priced, and losing, entry was quoted as blaming the loss on the fact that superior safety is expensive. Au contraire! In a plant properly designed according to the Safety-by-Design approach the increased level of safety is accompanied by a decrease in cost, as elaborated by Carelli.9 In fact, such a simpler plant is safer (there are fewer ‘things’ which can go wrong) and at the same time cheaper (there are fewer ‘things’).

This inverse connection between safety and cost has not been fully exploited by reactor designers, not going much further than the elimination of large LOCAs and associated piping systems. Rather, relying on proven technology, supply chain and infrastructure of the large PWRs has been taken by traditional LWR vendors as the sure path to economic competiveness. This conventional, sure-footed approach completely misses the fact that the iPWR is not just a smaller PWR, but it is a completely different design, with its own individual challenges and rewards. If properly identified, addressed and fully exploited the iPWR does indeed yield both increased safety and decreased cost. The systematic approach adopted in the IRIS design is in principle applicable to all iPWRs in general. It starts by eliminating the following major components and systems present in large LWRs:

• all large piping to/from the reactor vessel;

• steam generator pressure vessel;

• canned motors and seals of primary pumps;

• pressurizer vessel and pressurizer spray system;

• vessel head penetrations due to external control rods drive mechanisms (CRDMs);

• vessel bottom penetrations and seals due to in-core instrumentation;

• all active safety systems;

• high-pressure emergency core cooling system;

Also, the following major components have been reduced:

• shielding;

• number and complexity of passive safety systems;

• number of valves;

• size of containment and nuclear building;

• number of NSSS buildings (from two or more to one);

• number of large forged components (from approximately a dozen to one).

The only added major component/system is the seismic isolators.

In most iPWR designs the control rod drive mechanisms are located above the core, with the steam generators arranged outside in an annular configuration against the vessel. This leaves a coolant downcomer between the core and the vessel which is a quite effective shielding of the vessel wall and provides another iPWR intrinsic cost reduction. In fact, in some designs like IRIS the downcomer is large enough to reduce by several orders of magnitude the neutron fluence to the vessel wall, practically eliminating the vessel embrittlement and allowing increased plant lifetime. Also, the routine personnel exposures, and thus the ALARA costs, are reduced because the radiation level outside the iPWRs vessel is significantly decreased in respect to traditional LWRs.

The economy of scale, where the capital cost per unit power increases as the plant size decreases obviously applies and does favor larger plants. But traditional LWRS and iPWRS are on different, roughly parallel, cost versus power curves, with the iPWR being, even significantly, below because of the simpler and more economic design. In addition, SMRs in general enable economy of multiples versus single monolithic plants by relying on bulk/serial components fabrication (e. g. many small serial steam generators versus a few one-of-a-kind), accelerated learning and multiple units savings.

Modular construction and multiple modules deployment yield shorter construction schedule, module deployment tailored to demand with reduced spin reserve (i. e. reduced requirement for purchase power).

Quantification of the various factors is shown in Table 3.2, which compares the IRIS SMR (335 MWe) as part of a four-unit deployment against a single large PWR of 1340 MWe. The SMR starts with a large (1.7) penalty factor due to the economy of scale, but the other factors unique to SMRs in general (factors 2 through 5) and iPWRs in particular (factor 6) bring the estimated penalty for the iPWR down to 1.05, which is within the uncertainty level.

Two other factors have a significant financial impact favoring the SMRs in general: improved cash flow and reduced capital at risk. Construction of large plants takes a long time, with no income until completion. On the other hand, staggered deployment of the smaller SMRs enables ‘bootstrapping’ with the first unit generating income to support construction of the second unit and so on. As a result, the maximum cash outflow, or capital at risk, is significantly reduced as shown in Figure 3.2, which compares the cash flow of the same four 335 MWe IRIS plants (1340 MWe total) deployed every three years against a single 1340 MWe large PWR. An in-depth discussion will be found in Carelli et al.10 and Boarin and Ricotti11 iPWR economics is the subject of Chapter 10.

Finally, a critical consideration for iPWRs, which has a very significant impact in terms of both economics and public acceptance: a direct consequence of reducing the probability of catastrophic accidents to the order of E-8 is that the plant can

be licensed with a drastically reduced emergency planning zone, in principle to the plant boundary, although in practice to a few kilometers.12

Although potentially deployed as a stationary or floating power plant, the Russian RITM-200 is primarily intended to provide propulsion for the next generation of Russian ice-breakers. It is a 50 MWe compact integral system reactor that uses a

Figure 2.8 RITM-200 (Russian Federation) — OKBM Afriantov © Joint Stock Company ‘Afriantov OKB Mechanical Engineering’.

fuel type and configuration similar to the KLT-40S. The vertically mounted four main reactor coolant pumps are external to the reactor pressure vessel, as are the control rod drives mechanisms and a gas pressurizer. Four internal steam generators extract heat from the primary coolant system. Because of the external main coolant pumps and pressurizer, additional water injection systems are provided to mitigate the consequences of a large break loss-of-coolant accident.

By moving the steam generator into the reactor pressure vessel, the reactor system and containment is very compact compared to the KLT-40S. With overall dimensions of 6.4 m X 6.4 m X 15.5 m and weighing approximately 1200 ton, the RITM-200 occupies 45% less volume and is 35% lighter while producing 40% more power than the KLT-40S. It is expected to operate continuously for 26 000 hours (3.5 years) and be refueled after 7 years. Key parameters and a representative graphic for the RITM-100 design are given in Figure 2.8. [6]

The three major challenges for SMR deployment are as follows.

1.2.2.1 Sufficient reduction of financial risk

The investor-perceived financial risk arises from three key factors:

• NRC licensing requirements which could affect the capital as well as operating cost of these SMRs regarding plant staffing, security requirements, insurance and licensing fees, and decommissioning funding;

• the validity of the expected learning curve to reduce capital costs through factory manufacture;

• the more typical nuclear construction concerns, such as:

о construction and commercial operation schedule delay due to regulatory related delays,

о construction cost overrun due to constructor inexperience such as the current EPR Finnish and French construction activities and unforeseen mandated design enhancements such as those arising from the Fukushima accident, and о loss of investment due to operational and maintenance cost escalation or occurrence of a severe reactor accident.

All reactors are equally designed to a top level set of regulatory requirements, which however are not fully harmonized internationally. In the US these requirements have been made much more explicit for water-cooled reactors, since among the other coolants only the Fort St. Vrain helium-cooled reactor received a US Nuclear Regulatory Commission (NRC) (commercial) operating license. The explicit existing definition of water-cooled reactor regulatory requirements is a major benefit to light — water reactor (LWR) SMRs in comparatively assessing the licensability of other SMR coolant types. However, even for LWR SMRs the following factors significant to regulatory acceptance will need to be resolved:

• the reactor control strategy leading to reduction in the number of required operators;

• the reactivity control issues related to the desired long duration of the irradiation cycle to be accomplished by some designs without the use of soluble poisons;

• definition of the mechanistic source term for fission product release in a severe accident; and

• multi-module interactions.

Finally, LWR plant vendors are assuming that their designs will be accepted in a timely manner by the regulator. They base their optimism on the contention that their designs employ proven, current licensed concepts using proven components and systems configurations at power levels sufficiently low to allow the enhanced use of passive safety features which have already been reviewed and approved for the larger Generation Ш+ advanced light-water reactor (ALWRs). This assumption, even if proven correct, needs to reflect regulatory acceptance of at least some of the factors noted above cast in a manner yielding economic benefit to the SMR.

For SMRs using non-traditional coolants such as helium, sodium, lead-bismuth, or molten salts, the regulatory challenge is more difficult since the NRC staff lack familiarity with these reactor designs. Additionally, given the still largely prescriptive nature of light-water-based regulations in the US, the licensing process is not amenable to the newer more innovative designs. There have been calls for using a technology-neutral licensing process to license these new reactor concepts such that the inherent design features can be recognized by the regulator. The development of such a process is underway but is proceeding very slowly.

After decades of being focused on the GT-MHR and its variants, General Atomics recently introduced a new gas-cooled SMR called the Energy Multiplier Module (EM2). Although leveraging some of the GT-MHR technologies, the design is a significant departure from the thermal-spectrum GT-MHR. The EM2 is a fast spectrum reactor with its primary purpose to be a consumer of spent nuclear fuel. Specifically, it is a ‘breed and burn’ type reactor that converts fertile elements contained within the special fuel assemblies into fissile elements to sustain the process, while also consuming minor actinides produced in precursor once-through reactors such as LWRs. The reactor core uses uranium-carbide fuel clad in silicon-carbon composite material. The fuel is in the form of porous annular pellets and stacked within SiC-SiC tubes to form pins. The first reactor is loaded with nearly equal portions of low-enriched uranium fuel (approximately 6% 235U average enrichment) and discharged LWR

Upper

plenum

Upper

reflector

Side

reflector

Reactor

core

Control drum — Lower reflector

к

Reactor

vessel

Figure 2.19 EM2 (United States) — General Atomics (GA).

fuel. Subsequent EM2 reactors are loaded entirely with the fuel discharged from preceding EM2 reactors. The nearly 50% power conversion efficiency is achieved by operating with an outlet helium temperature of 850 °C and by using a vertically mounted, variable-speed direct Brayton-cycle power conversion unit.

It is expected that the reactor will operate for 30 years without refueling or fuel shuffling. After 30 years, the core will be replaced with a fresh load of recycled used fuel and the discharged fuel will be processed using a dry oxidation process to produce feedstock for subsequent reactors. Units are expected to be deployed in twin-unit plants with up to four plants per site. Fuel development and qualification tests are underway but are expected to be the longest lead challenge for eventual deployment. Despite the longer-term deployment schedule for the EM2, it is included in this survey because of the significant level of commercial investment by General Atomics. Key parameters and a representative graphic for the EM2 design are given in Figure 2.19. [17]

The US, Russia, South Korea, China, Japan, Argentina, and France all have concepts under design and component/system testing is underway in several cases. The most advanced situations are in the US, Russian and Chinese programs. US LWR SMR vendors have well advanced design and testing programs and have all announced deployment objectives with commercial utilities or regional state partnerships. The US Department of Energy (DOE) has launched a two-staged program offering $452 million in government grants over five years to support the design and certification of SMRs leading to reactor deployment by 2025. The SMRs to be developed would be less than 300 MWe, with scalable designs that could be manufactured in factories and shipped to utilities. Selected projects will be granted under a cost-shared agreement which requires that industry match at least half of the project’s costs. The first stage was awarded to Babcock and Wilcox in November, 2012, while the second stage funding was awarded to NuScale in December, 2013. Russian activities are numerous, as detailed in Section 1.1.3, and are centered on the delivery of electricity and co­generated heat and/or desalinized water to remote locations, through vessel-mounted reactors or terrestrial installations. The fleet of nuclear-powered icebreakers is being expanded as well. China has a two-unit commercial helium-cooled pebble bed plant (100 MWe per unit) under construction.

Although derived from marine propulsion systems, the VBER-300 also incorporates features from the larger Russian VVER reactor designs. The reactor core is composed of 85 hexagonal fuel elements similar in design to VVERs. Each element contains 312 pins of less than 5% enriched UO2 fuel. Soluble boron in the primary coolant

is used in addition to the 48 control rods and distributed gadolinium pins to control reactivity. Each steam generator is composed of 55 individual once-through coil-type cassettes with the feed water and outlet steam partitioned into three independent headers. The primary coolant flows over the shell-side of the steam generator cassettes. The external vessels are flanged very close to the reactor pressure vessel to reduce the likelihood of leaks or large-break loss-of-coolant accidents.

The VBER-300 is being designed for maximum flexibility as either a ground — based stationary plant or a floating plant and can be used for electricity generation only or as a co-generation plant for electricity and heat for water desalination, district heating or industrial applications. It is also being designed to be scaled to different capacities by using between two and six heat exchanger loops. If used for co-generation, steam is extracted from the low-pressure turbine with a commensurate loss of electrical output. The containment for the ground-based plant is a large-volume steel-lined concrete structure that is 34 m diameter by 42 m tall. Key parameters and a representative graphic for the VBER-300 design are given in Figure 2.9. [6]