Progress on a Preconceptual Supercritical Water-Cooled Small Modular Reactor Open Access

Harnessing energy has been a key contributor to human evolution, development, and physical wellbeing. Initially, the sources of energy were materials readily found in nature, and the exploitation of these had minimal or no impact on the environment. With an ever-increasing demand for energy, new sources of energy were developed, sometimes to the detriment of the environment. For example, during the industrial revolution, an exponential growth in the use of coal took place. Human advancements also took off but at the expense of polluting our environment. Today's use of fossil fuels has a significant impact on our climate on a global scale, and it is recognized that a replacement or alternatives are needed.

Nuclear energy offers clean energy and is available at the scale of today's energy consumption levels. First introduced in the 1950s, the nuclear industry generates about 9% of the world's electrical energy [1]. From the earliest designs to today, each succeeding generation of nuclear reactors seek to incorporate lessons learned from all sources, with a goal of ever-improved safety and economy of operation. The small modular reactor, also known as SMR, in theory, offers a lower initial capital cost by reducing the scale of the reactor (i.e., small), allowing serially produced shop-fabricated modules to be shipped to site rather than the reactor being erected from components in the field (i.e., modular).

A project under the Horizon 2020 platform is called Joint European Canadian Chinese Development of Small Modular Reactor Technology, denoted as (ECC-SMART) [2]. The ECC-SMART project, which began in September 2020, is oriented toward assessing the feasibility and identification of safety features of an intrinsically and passively safe supercritical water-cooled small modular reactor (SCW-SMR), taking into account specific knowledge gaps related to the future licensing process and implementation of this technology. The main objectives of the ECC-SMART project are to define the design requirements for the future SCW-SMR technology, to develop the prelicensing study and guidelines for the demonstration of the safety in the further development stages of the SCW-SMR concept including the methodologies and tools to be used, and to identify the key obstacles for the future SMR licensing and propose a strategy for this process.

Furthermore, Canada is a signatory of the supercritical water-cooled reactor (SCWR) system arrangement under the Generation IV International Forum (GIF). Similarly, GIF [3] identified the SCWR as one of the Gen IV nuclear reactor technologies.

Fossil-fired plants have been operating under supercritical² (SC) conditions for several decades (since 1950–1960), whereas nuclear power plants are still working under subcritical conditions. The reason fossil-fired plants moved to SC conditions is the increase in thermal efficiency. Increasing this efficiency translates to more power for the same amount of fuel, in other words, more revenue and fewer by-products. This is advantageous provided the cost of operating or maintaining the plant does not offset the gains. Current supercritical fossil-fired plants have efficiencies of up to 50%, whereas nuclear power plants are around 30–35% [4,5].

Therefore, SCW-SMRs are in a good position to achieve both goals, i.e., reduce the financial risk, and fight climate change by producing energy with a near-zero greenhouse gas source of energy.

This paper presents an overview that covers the current progress on various intertwined disciplines, such as a market study, thermodynamics and energy conversion, reactor physics, thermal hydraulics, materials and chemistry, and safety assessments.

Small reactors (SR)³ have been proposed since the beginning of the nuclear industry. However, owing to technological gaps of earlier times and elevated costs, their deployment did not succeed as expected. Fortunately, a few concepts were developed and built, and hence, some experience and knowledge in this area are available. The following excerpt from the Canadian book Canada Enters the Nuclear Age [6] gives a high overview of SRs in the past:

However, there are numerous SRs and SMRs design concepts nowadays. A comprehensive literature review was performed to gain more knowledge on concepts that are similar to the Canadian SCW-SMR.

In this paper, only five concepts are presented because of their characteristics and relevance. The selected concepts are GE-SCWR [7], CANDU-BLW (boiling light water) reactor [8], CANDU-3 [9], NuScale [10], and the BWRX-300 [11].

From the five concepts reviewed, it was observed that cost is a key element for the successful deployment of nuclear reactors. Nuclear power plants must be a competitive source of energy. If this requirement is not satisfied, it will be very challenging to convince investors to finance and support this technology. Another important observation made by CANDU-3 [9], NuScale [10] and BWRX-300 [11] designers and vendors is the importance of using well-proven technology, rather than highly innovative designs. In other words, SMRs should rely on Gen I, Gen II, and Gen III technology predecessors.

From the GE-SCWR [7] concept, the pivotal message is that the use of supercritical water is a normal evolution of water-cooled reactors as fossil-fired plants already operate at these conditions. Furthermore, the use of water as a coolant offers practicability. Interestingly, the Canadian SCWR [4,12] concept is similar to the GE-SCWR concept in the sense that it uses pressure tubes and is vertically oriented.

The CANDU-BLW [8] was also developed to decrease costs. To do that, the heavy water coolant was replaced with light water. A direct cycle was selected, eliminating the costs associated with the steam generators. However, these changes created new challenges for the CANDU-BLW, such as reactor instabilities and corrosion growth. Reactor operation under boiling conditions favors the production of crud, and density variations make the reactor difficult to control.

Interestingly, the vendors and designers of the CANDU-3 [9], the NuScale [10] and the BWRX-300 [11] concepts, which are based on experience with their predecessors, stressed the importance of using components that are already qualified, and of relying on a supply chain that is already available.

The use of existing supply chain in an evolutionary manner is expected to greatly decrease timelines for SMR deployment. Some of the drivers for SMRs require relatively near-scale deployment timelines to fully take advantage of market and policy opportunities.

The methodology used in this study relies on the knowledge of previous water-cooled SMRs and of various SCWR technologies developed by GIF member countries, in particular the Canadian SCWR [12]. The first step in the methodology was to identify high-level design requirements that are aligned with the ECC-SMART design targets [2]. The ECC-SMART design targets alongside GIF's technology goals are used to establish a list of high-level requirements for the conceptualization of an SCW-SMR. However, as pointed out by Sesonske [13], one of the most important steps is to start by defining the problem to be solved. The Canadian SMR roadmap [14] defines in detail the reasons Canada is embarking on the development of SMRs, which is associated with economic incentives. A condensed study of SMRs for the Canadian market is, therefore, provided in the  Appendix. An outcome of this study is that 300 MW_e is feasible from a Canadian perspective.

The second step was to identify the pertinent specifications, plant characteristics, strategies, and figures of merit that will allow us to reach the high-level design requirements.

Once the high-level design requirements are established, more specific technical values are needed to meet these requirements. This process is multidisciplinary and requires several iterations to obtain a feasible solution that will meet the imposed constraints (design requirements). To minimize the number of iterations, it is recommended to take advantage of the available experience with water-cooled SMRs; some parameters or values have been selected in this study to start the iteration process, such as the number of channels or assemblies and the core size. Table 1 presents relevant reference design values for the concepts reviewed.

Furthermore, to ensure that important design variables that are common in different reactor technology areas are considered globally, a preliminary study was carried out on the intertwining of key variables ensuring that critical design decisions made by one design team will not have adverse effects on the objectives of other design teams. At this point, five technical areas were selected to start the identification of design characteristics, limits, and parameters. These were (1) thermodynamics and energy conversion, (2) reactor physics, (3) thermal hydraulics of the fuel bundle, (4) materials, and (5) safety.

The design targets for the proposed concept are summarized in Table 2.

Some of the most important variables in the development of a power plant are the thermodynamic operating conditions. The selection of the optimal nuclear reactor outlet pressure and temperature for the purpose of power generation remains a topic of discussion. While it is widely accepted that an increase in operating temperature generally leads to an increase in conversion efficiency, the influence of operating pressure within the system is less obvious. To determine the response of the reactor across the pressure and temperature range contemplated for an SCW-SMR design, a study was conducted. Specifically, the influence the thermodynamic properties of water may have on the nominal operation of the reactor was examined, as well as the suitability of the steam for supply to a turbine set. The pressure and temperature ranges examined include operating pressures ranging from the critical point up to 30 MPa and outlet temperatures ranging from 400 °C to 650 °C. However, the temperatures are based on the limitations of the in-core materials. From this, an optimal space solution can then be found for future SCW-SMR designs intended for power generation.

The availability of a practical turbine set operating at pressures and temperatures above the critical point is first examined. A water Rankine cycle is chosen as this remains the power conversion cycle of choice due to its widespread adoption and availability, scalability, and nontoxic working fluid. Within the Rankine cycle, high-pressure steam is expanded by means of turbine stages, extracting thermal energy from the steam and converting this to mechanical energy. The conservation of mass and energy within the system dictates that a proportion of the steam will condense during this expansion process. Free liquid within the turbine, beyond a certain limit, is problematic as the water droplets impinge on the fast-moving blades, leading to premature wear or failure of the turbine. Thus, the expansion of steam is done in stages, and measures such as steam reheat in the case of fossil-fueled generating plants or moisture separator and reheater combination for nuclear generating plants are applied interstage to eliminate excess moisture. Turbomachinery designers can also incorporate features or geometries within the turbine that would allow greater tolerance to moisture.

To better understand the effects of the thermodynamic operating conditions on the proposed SCW-SMR, a notional turbine set operating at 90% conversion efficiency and rejecting steam to a condenser at 0.005 MPa is considered. Steam conditions are considered at the turbine inlet and outlet only, without reheat cycles or moisture separation. This exercise is conducted to determine trends in steam quality, and hence liquid volume, within the rejected steam across the operating range. Figure 1 demonstrates that for a given outlet temperature, increasing the outlet pressure results in an increasingly lower steam quality at rejection into the condenser. Hence, for the pressure and temperature ranges studied, a greater liquid load within the turbines can be expected at increasing reactor operating pressures, and thus a greater potential for turbine damage.

where $Tc$ is the absolute temperature of the cold reservoir and $TH$ is the absolute temperature of the hot reservoir. Thus, it can be expected that for a given steam inlet and outlet temperature, the conversion efficiency of a practical Rankine cycle would remain similar across the pressure range under consideration, with steam temperature having a much greater effect on conversion efficiency than steam pressure.

As seen in Fig. 2, plotting specific enthalpy of water as a series of isotherms for pressures ranging from 22 to 30 MPa reveals that for increasing pressures, specific enthalpy decreases slightly. This effect is most pronounced at temperatures just above the critical point. From this, it can be expected that when increasing operating pressures while maintaining a fixed outlet coolant temperature and reactor thermal output, the overall coolant mass flow delivered to the turbines would increase because of the lower specific enthalpy of the coolant. Feedwater pumping power would increase slightly with operating pressure due to the increase as well as the increase in mass flowrate, serving to reduce the overall plant efficiency.

For the reasons stated above, the following should be considered when developing high-level requirements for an SCW-SMR:

• For the pressure and temperature ranges assessed, no compelling advantage was found for an increase in operating pressure, given a constant outlet temperature. Increasing the operating pressure would result in a greater moisture load within the turbines, increase the feedwater pumping power, have minimal impact on conversion efficiency, and increase the coolant temperature within the fuel channel while marginally increasing the coolant density within the fuel channel. An increase in outlet temperature would increase the conversion efficiency of the system.

• It is recommended that the highest feasible operating temperature first be selected when elaborating a concept, based on in-core material limitations. Scoping calculations can be performed using a nominal reactor outlet pressure of 24–25 MPa. Based on this analysis, the proposed operating conditions are an inlet temperature set to 280–290 °C and an outlet temperature set to 450 °C with the option to increase it to 500 °C. The system pressure is set to 25 MPa.

Knowing the operating conditions and proposing an 800 MW_th reactor (assuming a thermodynamic efficiency of 37.5% as a first approximation), together with the design characteristics listed in Table 2, reactor physics calculations were performed in order to define a viable SCW-SMR core configuration, which could be used as a reference basis for further studies. The calculations were based on the 64-element fuel bundle design [15] (depicted in Fig. 3). The key parameters used in the calculations are listed in Table 3. The Monte Carlo reactor physics code serpent 2 [16] (Version 2.1.32) was used for calculations.

Some key variables for the design requirements are fuel type, burnup, and maximum enrichment allowed. In this analysis, the proposed fuel is the conventional fuel, that is, UO₂. The burnup is set to a minimum of two years without refueling, with a maximum fuel enrichment of 19.9%.

The analysis first assumed a fuel enrichment of 5%. The core was first divided into eight radial zones (see Fig. 4) and a burnup calculation was performed to determine its excess reactivity. Material properties among the 10 axial segments, i.e., the densities and temperatures of the coolant and the temperatures of the fuel and cladding, were modeled individually and obtained through thermal hydraulics calculations. The result indicated that the excess reactivity allowed for an operation of only ∼200 days and the axial power distribution was skewed toward the outlet, which was not desirable. In order to remedy the situation, a profile was implemented so that the enrichment varied axially along the 10 segments. The bottom segment was assigned an enrichment of 13% (arbitrarily chosen as the value similar to that presented by Ultra-Safe Nuclear Corporation for the Micro-Modular Reactor to be built at Canadian Nuclear Laboratories (CNL) [17]), and for four separate calculations, the enrichment of each subsequent segment was decreased by 0.1%, 0.25%, 0.5%, and 1%, respectively. The results are shown in Fig. 5, in which the 5% flat enrichment is also shown for comparison. Based on these results, it was recommended that the enrichment with 0.25% decrement, henceforth referred to as the reference core, be adopted for further studies. In contrast to the core with flat 5% enrichment (∼200 days), the reference core would increase the excess reactivity to allow for ∼700 days of operation.

The coolant void reactivity (CVR) phenomenon is one of the most important parameters for reactor safety as it relates to the loss-of-coolant accident (LOCA). A k_eff value of 1.41644 was obtained with the reference core with the nominal fuel coolant properties. Another calculation was performed in which the fuel coolant density was set to a constant 1000 kg/m³ while keeping the central coolant unchanged, resulting in a k_eff value of 1.42707. That is, the CVR for this core is estimated to be +5.3 mk (1 mk = 100 pcm). In contrast, the CVR for the CANDU equilibrium core is approximately +15 mk. Note that, other than the fact that the CVR is shown to be positive, the magnitude should only be taken as a first estimate as the calculation is based on a core that is far from criticality.

The fuel assembly for the proposed SCW-SMR is the same as the Canadian SCWR [12]. It consists of the fuel elements, central flow tube, encapsulated insulator, upper and lower fuel element supports, inlet/outlet flow exchanger, and outlet flow tube. The arrangement is illustrated in Fig. 6. Inlet coolant enters the fuel assembly from the inlet plenum and initially flows through the periphery of the fuel assembly. Above the fuel elements and upper fuel element support, a flow exchanger transfers the inlet coolant to the central flow tube. The same flow exchanger transfers the outlet coolant from the periphery of the fuel assembly to the outlet flow tube where it proceeds to the outlet header. Inlet coolant flows down the central flow tube to the bottom of the fuel assembly. The coolant reverses direction at the bottom of the fuel assembly and flows up the periphery of the fuel assembly over the fuel elements to the flow exchanger-outlet flow tube.

The 64-element fuel bundle concept was developed for the Canadian SCWR and optimized for supercritical conditions (depicted in Fig. 7). For that reason, it was decided to continue using the 64-element bundle for the proposed SCW-SMR. The heated length and inner diameter of the fuel channel remain identical to that of the Canadian SCWR, i.e., 5 m long and 14.4 cm in inner diameter. However, it is anticipated that the heated length will decrease as the concept evolves.

Selecting the reactor inlet and outlet coolant temperatures is of critical importance as these two temperatures impact all the design requirements. The performance of the plant, safety margins, reliability, and lifetime of associated system components are intertwined with the operating conditions, especially the outlet coolant temperature as it is related to safety limits.

Fortunately, a large body of knowledge is available to support the selection of the operating conditions for an SCW-SMR. For example, several SCWR concepts have been developed worldwide [48]. Some of these concepts are presented in Ref. [4] alongside the operating conditions. Based on the review of these concepts, the proposed fuel channel conditions are an inlet fuel channel temperature set to 290 °C and an outlet coolant temperature set to 450 °C with the option to increase it to 500 °C. The operating pressure is set to 25 MPa, similar to the Canadian SCWR [12].

All nuclear reactors generate a nonuniform cosine-like power distribution. This is mainly attributed to the shape of the reactor, the heterogeneity of the reactor components, and the change in coolant properties along the reactor core. The heterogeneity of the fuel also contributes to the nonlinear profiles. To study the impact of different burnup options, several enrichment configurations or options were proposed and assessed from a neutronics perspective. Understandably, each enrichment option results in a different axial power distribution because the fuel is “burning” at different rates. The resulting axial power distributions are shown in Fig. 5.

The resulting power distributions were assessed from a thermal hydraulics perspective using the subchannel code ASSERT-PV SC [18]. The maximum cladding temperature (MCT) is used as a key variable to evaluate the performance of each enrichment option [19].

The base case model was based on the set of recommended options for SC conditions [15]. These model options are presented in Table 4.

The ASSERT-PV SC [18] predictions of MCT are presented in Table 5. Figures 8(a)–8(e) presents the heat transfer coefficient (red line), the cladding temperature (blue line), and the coolant temperature (black line) along the length of the rod and subchannel where the MCT was predicted to occur. The axial and radial location are italicized in Tables 5 and 6. Figure 9 shows the subchannel and rod numbering of one quadrant of the 64-element fuel bundle.

In all simulated cases, the heat transfer coefficient presents two maxima. The reason is that the fluid crosses the pseudo-critical point (temperature in this case) at two different locations. First, the outer cladding will reach the pseudo-critical temperature. At this location the fluid near the boundary layer effectively removes the heat generated from the fuel elements (first peak). Second, as the bulk of the coolant flows through the fuel channel absorbing the heat produced by the fuel elements, it crosses the pseudocritical temperature as well (second peak).

Based on the current MCT analysis, the simulation results points toward selecting option 2 for further analysis as it results in the lowest MCT when compared with all the options analyzed. Furthermore, additional sensitivity cases were conducted to support the selection.

In all sensitivity cases the power remained unchanged, and only one variable or model was changed in each case. The model options used for the base case are listed in Table 4. The sensitivity cases are described below, and the results are presented in Table 6.

Case A heated length: This case studies the impact of decreasing the heated length from 5 m to 4 m. The goal of this case is to quantify the impact of a shorter reactor on the MCT. The inlet and outlet coolant temperature remains the same as the base case. The mass flow is adjusted to match the power of the base case. The results show that there is still room for improvement. For example, for Option 2, the increase in temperature is 22.1 °C for a decrease of 1 m of heated length.

Case B inlet temperature: This case studies the impact of decreasing the inlet coolant temperature from 290 °C to 280 °C. The outlet coolant temperature remains the same, and the mass flow is adjusted to match the power of the base case. The decrease of the inlet temperature does not result in a change in the MCT compared to the base case.

Case C outlet coolant temperature: This case studies the impact of increasing the outlet coolant temperature from 450 °C to 500 °C. The inlet coolant temperature remains the same. The mass flow is adjusted to match the power of the base case. The increase in the outlet temperature does not result in a linear increase in the MCT. In all cases, the MCT increases but at different rates. Option 2 resulted in the lowest MCT.

Case D thermal resistance: This case studies the impact of increasing the thermal resistance of the fuel element assuming oxide growth on the fuel elements. One of the knowledge gaps of any SCWR is the impact of corrosion on the performance of the reactor. The corrosion, or oxide buildup on the cladding, creates an additional layer that will increase the thermal resistance of the cladding. In this case, assuming a 20 μm oxide layer with a thermal conductivity of 13.55 to 16.85 W/m K, the results did not show any significant effect on the MCT predictions.

Case E heat transfer coefficient: This case studies the impact of changing the heat transfer correlation for supercritical conditions. In this case, the Jackson correlation [20] is changed to the Bishop correlation [23]. The results shows that the Bishop correlation systematically leads to higher MCT values. In previous heat transfer correlation assessments, the Jackson correlation [20] gave the best results, followed by the Bishop correlation [23]. In this case, the Bishop correlation predicts much higher MCT values for Option 3 and Option 4.

Case F intersubchannel turbulent mixing: One of the most important variables in subchannel analysis is turbulent mixing between interconnected subchannels. This case studies the impact of reducing the turbulent mixing by halving the coefficient of the mixing model. In this case the decrease in the mixing coefficient does not have a significant impact on MCT.

Case G decreased cladding roughness height: This case studies the impact of reducing the roughness height of the cladding material from 8.0 × 10^–3 to 8.0 × 10^–3 cm. The roughness height of the material affects directly the Reynolds number and, thus, affects the hydraulic resistance and heat transfer coefficient. These variations affect the mass, momentum, and energy distributions among subchannels. In this case, the decrease in the roughness height did not show any important variation.

Case H increased cladding roughness height: This case studies the impact of increasing the roughness height of the cladding material from 8.0 × 10^–3 to 8.0 × 10^–1 cm. increasing the cladding roughness height resulted in changes in all directions and the magnitude of the MCT. Interestingly the MCT is predicted to occur always in Rod 2 facing Subchannel 1.

Based on these results, a power distribution that is skewed to the inlet is recommended. However, the peak of the axial power distribution should be below a threshold to ensure no heat transfer deterioration close to the inlet of the channel.

In brief, this analysis supports the following considerations:

• The 64-element fuel bundle is recommended for the SCW-SMR.

• The recommended inlet coolant temperature is 290 °C.

• The recommended outlet coolant temperature is 450–500 °C.

• An enrichment strategy that results in an inlet-skewed cosine power distribution is recommended, provided the maximum local power does not result in heat transfer deterioration.

In the areas of materials and chemistry for SCW-SMR applications, knowledge gaps pertaining to microstructure changes under extreme conditions of temperature, pressure, and high neutron radiation field have been identified. The most significant materials challenge remains the selection and qualification of a fuel cladding material for the SCW-SMR concept that can withstand the supercritical conditions. The selection of the SCW-SMR cladding candidate materials is also a multidisciplinary problem as there are multiple constraints that need to be satisfied. These constraints can be grouped into three major categories, namely, performance, safety, and economics. Extensive out-of-pile oxidations of seven alloys were tested at CNL in the Refresh Supercritical Water Autoclave Facility. The detailed description of the loop used to conduct the out-of-pile oxidation tests, and the experimental procedure are described in detail elsewhere. These seven materials were selected based on consideration of the existing data. Table 7 summarizes the rationale for the selection of these materials. The first three materials, an alloy 800H (A800H), stainless steel 310 (310S) and alumina-forming alloy (AFA), were selected by the ECC-SMART collaboration program [2].

It is known that zirconium alloys such as Zr-2 and Zr-4 degrade rapidly under supercritical water conditions attributed to hydrogen embrittlement from aggressive hydrogen pickup during corrosion. The chromium-coated Zr–Nb alloys have been developed as accident-tolerant fuel (ATF) and have shown promising corrosion resistance at temperatures above 500 °C. The chromium-coated Zr–2.5Nb pressure tube and coated Zr–1.2Cr–0.1Fe could be considered as a potential cladding material for SCW-SMRs. One alternative that has been proposed is to employ Ti-50 enriched titanium alloys. Natural titanium contains approximately 74 wt.% Ti-48, which has a high neutron cross section, and as a result natural titanium alloys would not be suitable for in-core use. The neutron absorption cross section of Ti-50 is comparable to that of natural zirconium, but its natural abundance is only about 5 wt.%. However, because isotopes behave the same chemically, it is possible to test commercially available titanium alloys to examine their corrosion behavior. If they are found to be superior to other candidate materials, then we can begin to explore the cost of isotope enrichment [35–40].

In the first batch of Ni alloys, all three materials (A800H, 310S, and AFA) were oxidized in the same autoclave at two temperatures, 380 °C and 500 °C, and at 23 MPa. The feedwater was slightly oxygenated to about 150 μg/kg. The temperature near the supercritical point of water was chosen in order to observe corrosion during the startup before the system reaches its full power and during the cooldown of the SCW-SMR. Target operating conditions for the SCW-SMR were 500 °C and ∼25 MPa. In the second batch, materials were coated zirconium- and titanium-based alloys. Again, all coated coupons were exposed all together in the same autoclave tested at 500 °C and ∼25 MPa in 630 μg/kg oxygenated SCW. After oxidation tests, weight gains were used as a first approximation to assess the performance of the selected candidate fuel-cladding materials. The weight gain per unit area (in mg/dm²) of each coupon was obtained by an analytical balance with an accuracy of ±0.001 mg for samples weighing less than 10 g. The standard deviation between measurements of the same mass was 0.02 mg.

Results from the weight change measurements after approximately 1000 h of exposure time were normalized (assuming linear kinetics) and are shown in Fig. 10. At lower-temperature supercritical water, A800H has a weight gain lower than that of 310S and one-nineteenth of the weight gain of itself measured at 500 °C. At a full-power operating temperature of 500 °C, both 310S and AFA have good corrosion resistance. Note that the AFA coupons were tested only at 500 °C; no data of AFA weight change at 380 °C were available. Among oxidized coupons coated by approximately 10 μm of metallic chromium, titanium alloys show weight gain similar to that of Zr–1.2Cr–0.1Fe.

A systematic study was performed on coupons made of the Zr–2.5Nb pressure tube, currently CANDU's in-core component. As expected, the coupons made of as-received Zr–2.5Nb pressure tube show the highest weight gain when compared with others. The weight gain decreases by a factor of 15 with an approximately 10 μm coating thickness of chromium added via physical vapor deposition. The coated Zr–2.5Nb pressure tube specimens were dual-beam (proton and heavy ion) irradiated for a nominal dose between 5 to 15 dpa at Michigan Ion Beam Laboratory at the University of Michigan. The coated and irradiated specimens were oxidized in an SCW autoclave at CNL. The weight gains were measured and are included in Fig. 10. It shows that the weight gain increases by a factor of eight when compared with the coated but nonirradiated pressure tube specimens, when oxidized at 500 °C and ∼25 MPa. In addition to weight change measurements, microstructure and elemental mapping analyses of the corroded coupons are ongoing.

where $WG$ is the weight gain in mg/dm², $fo/ox$ is the weight fraction of oxygen in oxide, and $ρalloy$ is the density of the material in kg/m³. The design life of SCW-SMR fuel cladding is expected to be three to five years; therefore, the cladding tube made of 310S and AFA will be thinner by 6 μm, which is considered relatively thin when compared with the original thickness of few hundred micrometers. Currently, 800H is one of a few code-qualified materials for fabricating in-core and out-of-core components attributed to its superior creep resistance and rupture strength during extended high-temperature operations. Therefore, it is reasonable to include this alloy as one of the SCW-SMR candidate cladding materials.

Long-term corrosion tests would be needed to confirm the assumed linear kinetics. A thickening oxide of these candidate materials would have implications for the heat transfer and neutron economy of the fuel cladding, and would require further investigation.

A preliminary safety assessment was performed to quantify the safety performance and demonstrate the effectiveness of the safety system for two postulated accident events in the Canadian SCW-SMR concept. The events of interest include the following:

• Loss-of-coolant accident, cold-leg, and hot-leg.

• Loss-of-flow accident (LOFA), which is also referred to as the loss-of-feedwater accident (LOFA) for the direct thermodynamic cycle.

Based on experience with current water-cooled reactors, these events are treated as design-basis accidents and must be analyzed during all phases (from conceptual design to licensing) in the development of the Canadian SCW-SMR concept. The thoroughness of the analysis increases as the phase approaches licensing.

As indicated in the literature review section, several reactor-core concepts, fuel-assembly concepts, and fuel-channel concepts were explored in establishing the Canadian SCW- SMR concept. This concept evolved from the Canadian SCWR, and the safety analyses performed for the Canadian SCWR were used as a platform for the Canadian SCW-SMR. Experience and knowledge acquired from these preliminary analyses have been applied in the current analysis.

The analysis focuses on the heat transport system coupled with the proposed safety systems. Since the development of all systems is still in the conceptual phase, the detailed configuration of piping (e.g., sizes, lengths, and angles) and components (such as valves, bends, and elbows) has not been dimensioned properly. Simplified layouts are used in the analysis as this is the first iteration of the design. For those reasons, a large break LOCA was omitted in this study. Instead, two break sizes LOCA are presented: a 3% pipe break size for the hot-leg (small LOCA) and 10% for the cold-leg LOCA (medium LOCA).

As the concept matures, additional design-basis accident (DBA) analyses will be needed to examine any impact on safety for the advanced concept. Additional safety systems may be implemented as needed.

The Canadian algorithm for thermal hydraulic network analysis (CATHENA) computer code was used in the preliminary safety assessment [41]. CATHENA was developed primarily for safety analyses of power and research reactors at subcritical pressures. A new version of the code was developed for system thermal hydraulics analyses of the Canadian SCWR concept and has been used for the Canadian SCW-SMR.

Table 8 summarizes key models selected in the analysis. The Jackson correlation [20] has been implemented into CATHENA for calculating heat transfer between the cladding and the supercritical coolant. The Colebrook–White equation for friction factor [21] is generally applicable for all fluids and takes into account the roughness effect. It was applied for the pressure loss calculation in channels. The homogeneous model for orifices was applied in the calculation of the break-discharge flow. It has been shown to provide better agreement than the separate-flow model with recent critical-flow data at supercritical pressures [42]. However, modifications have been introduced to the homogeneous model enhancing further the prediction accuracy [43].

The CATHENA core channel grouping nodalization of the Canadian SCW-SMR model is shown in Fig. 11. Figure 12 presents the associated systems.

In the CATHENA model, the reactor core is grouped in channel groups named CG-X, where X represents the group number from 1 to 8. The total number of channels per channel group is not uniform, and have the following distribution: 4, 10, 16, 28, 28, 32, 36, and 32, respectively. Each channel group defines the respective flow areas, hydraulic diameters, and powers. There are 103 wall heat transfer models used to capture the heat transfer from the fuel elements to the coolant, moderator, and reactor vessel environment. Radiative heat transfers among fuel elements and fuel channel walls are also considered in modeling.

Thermal hydraulic components to be modeled include core channels, inlet plenum, outlet header, moderator calandria, circulation pump, LOCA break blowdown lines, and the isolation condenser system (ICS), as shown in Fig. 12.

In this analysis three key thermal hydraulic variables were used to assess the safety of the reactor, namely, the maximum fuel temperature (MFT), MCT, and maximum averaged coolant temperature (MACT).⁴ The thermal hydraulics constitutive models and correlations are listed in Table 8.

In view of the simplified equipment configurations and component layouts at the conceptual phase, several representative (and mostly limiting) cases have been selected in the analysis. The scenario and sequence of each case in the event are described below.

Before performing the transient and/or safety assessment, it is important to ensure that the simulation is initialized from a proper steady-state operational condition. In this state, the Canadian SCW-SMR core generates 800 MW of thermal power. The coolant mass flow through the core is 453.3 kg/s pressurized at 25 MPa. The outlet coolant temperatures for all the modeled groups are shown in Fig. 13(a). An outlet skewed power distribution is used for this steady-state simulation (option 5 in Fig. 5).

Figure 13(b) is the core coolant axial density profile. It clearly shows that the coolant flows enter as liquid-like phase and exits as gas-like phase. The simulation predicted an MFT of 1356 °C and an MCT of 552 °C.

Two additional steady-state cases were performed: case I, to assess the impact of the axial power distribution, and case II, to assess the thermal hydraulics performance of a shorter reactor core. In case I, an inlet-skewed cosine axial power distribution was used (Fig. 14(a)) instead of the outlet-skewed distribution. This case also aims to support the subchannel analysis results in the previous section. In case II, the inlet-skewed cosine axial power distribution is used again with a reduction of one meter of the effective core length, i.e., from 5 m to 4 m (Fig. 14(b)). Both analyses aim to assess the possibility of reducing the core size.

The CATHENA predictions support the selection of an inlet-skewed cosine power distribution from a thermal hydraulics perspective. The predicted MFT changed from 1356 °C to 1230 °C, and the predicted MCT decreased from 552 °C to 506 °C (as shown in Fig. 15(a)). However, applying the inlet-skewed cosine power distribution and reducing the core effective length from 5 m to 4 m increased the predicted MFT from 1356 °C to 1545 °C, but the predicted MCT is reduced from 552 °C to 524 °C (as shown in Fig. 15(b)). This supports the viability of reducing the core length from 5 m to 4 m. The safety simulations presented hereafter use the recommended inlet-skewed axial power distribution (option 2 in Fig. 5) and a core length of 4 m.

The logic and timings of the simulated DBAs are summarized in Table 9. These DBAs were selected based on terms of occurrence probability and frequency during the lifespan of nuclear reactors [46,47].

For each DBA analyzed, two simulations were performed: (1) without using the ICS and (2) the same simulation is performed, but the ICS is credited.⁵

The detailed accident progression is still preliminary at this stage of the preconceptualization. For that reason, only analyses of the three key variables are presented. The behavior of the MFT and MCT variables in the cold-leg LOCA, hot-leg LOCA, and LOFA accidents are shown, respectively, in Figs. 16–18.

Both the hot-leg and cold-leg LOCAs were simulated for the analysis. The hot-leg LOCAs were assumed to occur on one of the steam line pipes, and the cold-leg LOCAs were assumed to occur on one of the feedwater (FW) pipes. The inner diameter of the line pipes is 0.56 m for the steam line and 0.31 m for the feedwater line. The break size diameter for the hot- and cold-leg LOCA is 0.1 m.⁶

For the same break size, the cold-leg LOCA produced higher fuel cladding temperatures than the hot-leg LOCA because of a reduction in the coolant flow through the reactor core at the initiation of the cold-leg LOCA event. By comparison, the coolant flowing through the core increased in the hot-leg LOCA event. The reactor may be tripped by a low steam line flow, a low FW line flow, or a high coolant temperature.

The cold-leg LOCA is the worst scenario of the three analyzed scenarios, with respect to MCT and MFT. It is also noted that crediting ICS does not significantly affect the predicted peak core temperatures.

Figure 16 shows that, with or without crediting the ICS, the peak maximum fuel centerline temperature in the cold-leg LOCA transient is 1523 °C, the peak maximum fuel cladding temperature in the cold-leg LOCA transient is 1003 °C, and the peak core outlet temperature is 575 °C. However, crediting ICS does significantly reduce the predicted core outlet temperature after about 35 s.

This design-basis accident is less severe than the cold-leg LOCA, similar to BWRs. This is because of the lower density of the coolant at the outlet, which leads to choking of the flow at the discharge and thus a reduction in the discharge flowrate. Figure 17 shows that, with or without crediting the ICS, the predicted maximum fuel centerline temperature, maximum fuel cladding temperature, and core outlet temperature decrease consistently.

Loss-of-flow accident simulation results showed an increase in the core temperatures. Figure 18 shows that, with or without crediting the ICS, the predicted MFT is 1718 °C, the MCT is 827 °C, and the MACT is 640 °C. Crediting the ICS significantly reduces the predicted fuel and coolant temperatures after about 35 s.

Furthermore, it is noted that after the power trip and core isolation (after about 40 s), there are still natural circulation flows through the core channels. It is also noted that natural circulation flow is enhanced by crediting the ICS after about 60 s, which improves long-term cooling of the reactor core.

Based on the CATHENA simulation results, it is concluded that:

• Among the three accident scenarios analyzed, the cold-leg LOCA results in higher MCT and MFT. The peak maximum fuel cladding temperature is predicted to be 1003 °C.

• The ICS does not affect the predicted maximum fuel cladding temperatures in all three accident scenarios; however, it significantly reduces long-term core temperatures in the simulated accident scenarios.

This study presents the progress in the conceptualization of a supercritical water cooled small modular reactor (SCW-SMR). This process is a multi-objective and multidisciplinary exercise. An iterative approach narrows down the solution space. This document presents the first iteration of the conceptualization of the proposed SCW-SMR. To start, a market study is used as a base to identify the needs and thus the high-level requirements that the concept must meet. A Canadian market study is presented. Three different options, called streams, are identified: stream 1, on-grid reactor concepts with powers around 300 MW_e; stream 2 on-grid advanced reactors; and stream 3, off-grid concepts. Furthermore, different market opportunities are identified for the Canadian landscape. Based on the market study, it was decided to pursue the feasibility of a SCW-SMR as a replacement for fossil-fired plants, with a reactor power of 300 MW_e.

An analysis of thermodynamics and energy conversion cycle of the proposed SCW-SMR is presented. The outlet reactor coolant temperature is the key variable for increasing the thermodynamic efficiency and is based on the maximum temperature allowed by the materials. Furthermore, the system pressure should be selected to minimize the risk of damaging the secondary side, such as the turbine. The recommended operating conditions are an inlet coolant temperature of 290 °C, an outlet coolant temperature of 450–500 °C, and a system pressure between 24 and 25 MPa.

Found on the recommended reactor power, operating conditions, and cladding candidates, reactor physics calculations are performed to estimate the fuel enrichment to achieve the desired burnup. Five enrichment profiles are analyzed, resulting in five different axial power distributions. Preliminary analysis of the reactivity coefficients is discussed.

Using the findings from the thermodynamic and energy conversion analysis, alongside the axial power distributions derived from the reactor physics calculations, a thermal hydraulics analysis of the fuel bundle is presented. Using the subchannel code ASSERT-PV SC, several sensitivity cases are carried out to assess the impact of the operating conditions and the axial power distribution on the maximum outer cladding temperature, and to support the recommendations. From this analysis, it is recommended to continue using the 64-element bundle and an enrichment option that has an inlet-skewed cosine-like power distribution.

The selection of the cladding candidate for an SCWR and SCW-SMR reveals one of the most critical knowledge gaps. Although there are suitable materials that withstand the supercritical water environment, these materials have large neutron absorption cross section compared to Zr-based claddings. To compensate for the neutron losses, higher levels of enrichment are needed. This results in higher costs and supply disadvantages compared with the current enrichments for light water reactors. However, new promising materials are under study worldwide. For this analysis, two cladding materials are proposed, namely, alloy 800H and stainless steel 310S. Oxidation experimental campaigns using refreshed autoclaves operating at the recommended supercritical water operating conditions are ongoing. Preliminary weight gain results are presented in this document.

Based on the results and recommendations from this multidisciplinary work, three design-basis accidents are assessed as part of the conceptualization process. This analysis aims to identify any potential impasse from the safety perspective. The selected cases are cold-leg LOCA, hot-leg LOCA and LOFA. Three key variables are used to quantify the impact of the accident, , namely, maximum outer cladding temperature, fuel centerline temperature, and coolant average temperature. The preliminary analysis shows that the concept is feasible from the safety perspective.

• Office of Energy Research and Development (OERD) at Natural Resources Canada (Funder ID: 10.13039/501100007178).

• Natural Sciences and Engineering Research Council (NSERC) (Funder ID: 10.13039/501100000038).

• Atomic Energy of Canada Limited (AECL) (Funder ID: 10.13039/501100004953).

AECL =

Atomic Energy of Canada Limited

AFA =

alumina-forming alloy

ASSERT-PV =

Advanced solution of subchannel equations in reactor thermal hydraulics—pressure velocity

ATF =

accident-tolerant fuel

BLW =

boiling light water

BWR =

boiling water reactor

CANDU =

Canadian Deuterium

CANFLEX =

CANDU flexible

CATHENA =

Canadian algorithm for thermal hydraulic network analysis

CNL =

Canadian Nuclear Laboratories

CVR =

coolant void reactivity

DBA =

design-basis accident

dpa =

displacement per atom

ECC-SMART =

Joint European Canadian Chinese Development of Small Modular Reactor Technology

FW =

feedwater

GE =

general electric

GIF =

generation-IV International Forum

HALEU =

high-assay low-enriched uranium

IAEA =

International Atomic Energy Agency

ICS =

isolation condenser system

LEU =

low-enriched uranium

LOCA =

loss-of-coolant accident

LOFA =

loss-of-flow accident

MACT =

maximum average coolant temperature

MCT =

maximum outer cladding temperature

MFT =

maximum fuel centerline temperature

NBP =

New Brunswick Power

OPG =

Ontario Power Generation

ORNL =

Oak Ridge National Laboratory

PWR =

pressurized water reactor

SaskPower =

Saskatchewan Power

SC =

supercritical

SCW =

supercritical water

SCWR =

supercritical water-cooled reactor

SCW-SMR =

supercritical water-cooled small modular reactor

SMR =

small modular reactor

SR =

small reactor

A facet of defining design requirements for future SCW-SMR technology proposed in the ECC-SMART project is to assess whether the design target for electrical output of around 100–300 MW_e for an SMR is feasible. Such an assessment requires accounting for the opportunities and challenges in developing a potential pathway that supports a strategy for future SCW-SMR deployment. To support this facet from a Canadian market perspective, two recent studies that identified Canadian market segments for SMRs are reviewed and assessed. In the review, the essence is to present the market opportunities insofar as they intertwine with technical aspects to consider for the SCW-SMR design. While several parameters presented in the review are dynamic and could change over time, the strategy is expected to be adapted to such changes in future research.

In the first study [14], the Canadian market for SMRs was divided into remote communities, mines, oil sands, heavy industry, and utilities (Table 10). In the second study [48], which represented the Canadian utilities' perspectives of market segments, Ontario power generation (OPG) and Bruce Power in Ontario, New Brunswick Power (NB Power), and Saskatchewan Power Corporation (SaskPower) proposed three market streams of SMR project proposals (Table 11) to the governments of Ontario, New Brunswick, and Saskatchewan:

1. Stream 1 represents a market for on-grid ready-now SMRs in the short term.

2. Stream 2 represents a market for on-grid advanced SMRs.

3. Stream 3 represents a market for off-grid smaller SMRs or micro modular reactors (MMRs).

The primary differences between streams 1 and 2 is twofold. First, the SMR technology in stream 1 targets first power in 2028, and stream 2 focuses on Gen IV designs that have a deployment potential beyond 2030. In addition, stream 1 requires the use of existing infrastructure to take advantage of recent refurbishment experience. Stream 1, thus, implies the use of conventional technology as much as possible. Second, the primary type of fuel cycle that motivates stream 1 is a once-through fuel cycle, whereas the primary type of fuel cycle that motivates stream 2 is recycling. Stream 3 focuses on reactor sizes that are smaller than in streams 1 and 2. In addition, in stream 3 heat could be a primary product. In this stream, the current focus is on reactors that use high temperature gas as the coolant and use a once-through fuel cycle.

In relation to the first study, stream 3 maps onto the first four market segments where there is a potential to use an SMR for either heat or power, whereas Stream 1 and 2 subdivide the utilities market segment where the primary product is power.

Irrespective of market segments, Canadian [14,48] and international [50] studies recognize that a global interest in deploying nuclear reactors stems from reducing greenhouse gas emissions. The time horizon for this opportunity is expected to be available for mature technology from now to 2050, whereas for a less mature reactor technology it is beyond 2050 [50]. The maturity, therefore, limits the timing of the opportunity. How can a strategic pathway be developed to address market segments and technological maturity?

Typically, purchasers of power plants are concerned with economics, safety, availability, and reliability [51]. These concerns map into the general (or broad) design requirements for utilities: economics, safety, and performance [52–56]—where performance includes availability and reliability. Since these broad requirements are likely highly valued by all market segments, additional criterion is needed to establish a strategy for developing a commercialization pathway.

In adopting new products, the elements of familiarity and surprise are embedded in people's preference for complexity—up to the point that they stop understanding something [57,58]. Given these aspects, the utilities sector is the only sector amongst the various market segments familiar with nuclear reactors. For adopting nuclear reactors in segments with less nuclear experience, it would help to then “look at institutions with access to many potential buyers with common interests” [58], like those institutions supporting utilities. Thus, Streams 1 and 2 from the second study should be prioritized for a pathway. Furthermore, since the site for stream 1 provides a definite size per reactor, the recommended starting point for a unit size is 300 MW_e.

In these segments, an economic value for power plant owners is the potential cost savings from using nuclear instead of fossil fuel. Current costs and projected estimates by utilities and grid operators indicate that fossil fuel costs are approximately between three and six times larger than nuclear fuel costs [59,60]. A value proposition is, thus, to save fuel costs while reducing emissions, and maintaining safety and performance. To achieve this value proposition within a feasible amount of cost and time is a challenge.

To understand how to potentially reduce the development cost and time, it is useful to consider the infrastructure and technological base of an industrial ecosystem for SCW-SMRs. An “industrial ecosystem can be thought as a matrix where heterogeneous organizations operating in one (or more) sectoral value chain(s) draw on one (or more) capability domains (and the different types of technologies they include) to perform a number of production and technology functions” [61]. Such a matrix for SCW-SMRs is shown in Table 12. This table shows horizontally the infrastructure status for a capability used in a sectoral value chain that can be incorporated by an SCW-SMR technological architecture. To minimize infrastructure costs, this architecture will require recombining existing components, e.g., years of experience of operating water-cooled nuclear reactors and SC fossil-fired plants. In addition, successful innovation is partly characterized by replicative-integrative capability [62]. The key capabilities for integrating several sectoral chains are design and engineering, and materials and chemistry testing. This capability is required, for instance, to overcome the design challenges arising from technological discontinuities in scaling from large to small reactors [63]—threshold effects. In addressing these challenges, a complementary cost cutting strategy to incorporate in the integration capability could be to eliminate entire components or systems [63,64]. Furthermore, the integration capability will require testing submodules and modular components when determining the feasibility of materials and configurations. While new infrastructure is required in developing an SCW-SMR, its evolutionary aspect will enable to minimize infrastructure costs.

As conditions change, the strategy for deploying an SCW-SMR may be adapted. Other aspects that may affect this strategy for future study are the expected demand for power and/or heat, energy policies, planning horizons, design requirements for each market segment, uncertainty, and financing. Given the market opportunities and challenges, the next step is the establishment of a list of design requirements that will satisfy the identified market needs. The design requirements section presents the methodology and the preliminary design requirements of the proposed SCW-SMR.