Reactor performance and safety characteristics of two-phase composite moderator concepts for modular high temperature gas cooled reactors

https://doi.org/10.1016/j.nucengdes.2020.110824Get rights and content

Highlights

  • Beryllium- and hydride-based moderators have favorable moderating power and the potential for improved in-service lifetime as compared to graphite.

  • Advanced moderators have the potential for enhanced cycle performance to that of the graphite reference case.

  • Analysis of DBAs show that the high volumetric heat capacity of the beryllium-based moderator grants them a greater margin to fuel failure than a graphite moderated system.

Abstract

Graphite moderators have an extensive historical performance record, but also feature inherent challenges for modular High Temperature Gas-Cooled Reactors (mHTGRs). Challenges with graphite include non-uniform expansion and contraction under irradiation and build-up of potential energy during the bombardment of high energy neutrons that results in a large energy release under annealing. These challenges have led to the investigation and development of alternative moderators to be utilized in mHTGRs, including beryllium- and hydride-based concepts with compositions selected for favorable moderating power and the potential for improved in-service lifetime as compared to graphite. The proposed moderators are fabricated as two-phase composites with magnesium oxide, MgO, as the radiation-stable host matrix and beryllium metal, Be, beryllium oxide, BeO, or zirconium hydride, ZrHx=1 (to account for hydrogen loss from the hydride phase during processing), as the entrained moderating phase. Here, we evaluate the reactor performance and safety characteristics of these moderator concepts relative to a graphite reference using a Ft. Saint Vrain-style fuel block. We assessed the cycle length, discharge burnup, natural resource utilization, neutron flux spectra, moderating power, moderating ratio, critical size, moderator and fuel temperature feedback, fuel cycle cost, spent nuclear fuel and high level waste radioactivity per unit energy generated, and environmental impact per unit energy generated. The results demonstrate that the advanced moderators have the potential for comparable or enhanced cycle performance to that of the graphite reference case with significantly improved performance for an optimized moderator-to-fuel ratio design. These advanced moderators are also assessed from a reactor safety standpoint for Design Basis Accidents (DBAs) including Pressurized Loss of Forced Cooling and Depressurized Loss of Forced Cooling accidents for a 350 megawatt thermal prismatic-type mHTGR. The full core thermohydraulic analysis of DBAs show that the high volumetric heat capacity of the beryllium-based moderator grants them a greater margin to fuel failure in these analyses than a conventional graphite moderated system, but the lower thermal conductivity of the beryllium-based moderators leads to longer times at elevated temperatures.

Introduction

The modular High Temperature Gas-Cooled Reactor (mHTGR) is a gas-cooled reactor that has attractive inherent safety features such as negative temperature feedback, high thermal inertia, and passive decay heat removal systems (Carlson and Ball, 2016). Experiments at the 10 MW high temperature gas-cooled reactor test module (HTR-10) demonstrated the inherent safety features of mHTGRs during two simulated anticipated transients without SCRAM for both an overpower and undercooling event (Hu and Wang, 2006). mHTGRs have helium coolant outlet temperatures of greater than 700 °C, suggesting possible non-electricity-generating applications such as hydrogen and synfuel production, desalination, or as process heat for metallurgy and petrochemistry related industrial processes (Brown and Revankar, 2012a, Brown and Revankar, 2012b). Conventional mHTGR designs utilize TRi-structural ISOtropic (TRISO) fuel particles embedded into a graphitic compact or pebble, which provides added measures of protection from fission product release during operation.

While the base technology for mHTGR’s has existed for many decades, the conventional mHTGR has challenges associated with its current design. The low power density associated with an mHTGR equates to a costlier construction and operation. Additionally, the mHTGR has unique mechanical challenges associated with the graphite moderator. Graphite experiences a significant swing in volumetric change under a neutron fluence along with non-uniform expansion and contraction in the axial and transverse directions (Campbell et al., 2016). This anisotropic swelling ultimately leads to unsupportable internal stresses and microcracking, which defines the graphite lifetime and the corresponding need for costly replacement of core internals. Moreover, measurements on irradiated graphite reveal large strains and increases in stored energy within the lattice structure as well as changes in mechanical properties such as elastic constants, strength, and creep behavior (Telling et al., 2003). The issue of stored energy is unique to the graphite system as compared with essentially all other ionically or covalently bonded ceramics (Snead et al., 2019a, Snead et al., 2019b).

Global interest has recently expanded in micro-reactors, which are small, compact, low-power (less than 20 megawatts thermal) nuclear reactors capable of providing electricity or process heat in remote locations. These reactor concepts should be able to operate for several years without refueling. It is also notable that a number of historical reactor demonstrations have focused on aircraft propulsion as this application requires compact reactors and high temperatures. These demonstrations used beryllium-based moderators with molten salt coolant and hydrogen-based moderators with gas coolant. Beryllium-based moderators have historically been used in the Aircraft Reactor Experiment (ARE), which utilized hexagonal beryllium oxide (BeO) blocks as both the moderator and reflector (Cottrell, et al., 1955). Hydrogen-based moderators were used within the Heat Transfer Reactor Experiment No. 3 (HTRE-3) (Linn, 1962), a prototype aircraft power plant with zirconium hydride (ZrH) as its primary moderator and beryllium (Be) as its reflector. These moderators were selected based on their potential to enable a compact design that is essential for aviation and were operated for brief durations. Additionally, BeO and the structural matrix material considered herein, magnesium oxide (MgO), have been shown to be chemically stable and have high melting temperatures of 2800 and 3000 K, respectively (Victor and Douglas, 1963).

The Advanced Research Projects Agency-Energy (ARPA-E) under the Modeling-Enhanced Innovations Trailblazing Nuclear Energy Reinvigoration (MEITNER) Program aims to identify alternatives to graphite moderators that will enable transformational design improvements for mHTGRs. Ongoing efforts at Stony Brook University (Snead et al., 2019a, Snead et al., 2019b) and the University of Tennessee-Knoxville (Ang et al., 2019) are focused on the fabrication of two-phase moderators with the moderating phase entrained within a fully-dense MgO host matrix as it has been reported (albeit with only limited data) to be radiation tolerant and chemically stable (Clinard et al., 1982). Concepts for the moderating phase include beryllium- and hydrogen-containing materials for their increased moderating power relative to graphite. In this study, the advanced moderators are assessed from a reactor physics standpoint with the following parameters considered: reactor cycle length, discharge burnup, natural resource utilization, neutron flux spectra, moderating power, critical size, moderator and fuel temperature feedback, fuel cycle costs, waste radioactivity, and environmental impact. Additionally, a thermal–hydraulic analysis of MgO-Be, MgO-BeO, and graphite reference cases was conducted to evaluate steady-state axial temperature profiles as well as Pressurized Loss Of Forced Cooling (PLOFC) and Depressurized Loss Of Forced Cooling (DLOFC) Design Basis Accidents (DBAs) peak and average moderator block temperatures. The scope of this paper is to provide a preliminary assessment of the validity and performance of the composite moderators by leveraging a graphite optimized prismatic fuel block design available in the open literature (Pope, 2012) and a moderator-to-fuel ratio optimized design to provide an initial comparison of reactor performance. Our results demonstrate comparable neutronics performance for unoptimized cases and improved neutronics performance for optimized lattice pitch cases. The model assessed other potential host matrixes, MgAl2O4 and SiC, and entrained moderator, Be2C, but these options were not fully investigated because of decreased or marginal improvements to fuel cycle performance compared to the graphite reference case. The approach to establishing and verifying the models is shown in Fig. 1, which illustrates the different sources of the design information used in the study and relationships between these information sources.

Section snippets

Neutronics model

The continuous-energy Monte Carlo reactor physics burnup calculation code Serpent (Leppänen et al., 2014) was utilized to compare the neutronics performance of the proposed neutron moderators against the graphite moderated reference case. These calculations utilized ENDF/B-VII.0 cross-section libraries and S(α,β) libraries for thermal neutron scattering of ZrH and BeO. The neutronics model employs a Ft. Saint Vrain-style single fuel block with a periodic boundary condition, single-batch,

Thermal-hydraulics model

The thermal–hydraulic safety analysis for PLOFC and DLOFC DBAs was performed by utilizing a RELAP5-3D ring model of the 350-MWt modular mHTGR that was adapted from an international benchmark available in the open literature (Strydom et al., 2016) in which a coupled code suite was utilized to determine a thermal fluids solution and power distribution. The RELAP ring model used in this study uses the power distribution from this international benchmark and related studies (Lu, et al., 2019). The

Neutronics and fuel cycle performance

The effectiveness of these advanced moderators is illustrated in Fig. 5 where the infinite multiplication factor, Kinf, is plotted as a function of the amount of time that the fuel is spent within the reactor at full power in units of days, namely the fuel residence time in Effective Full Power Days (EFPD). The fuel residence time is a measure of the reactor operation time for a given system. The model used does not incorporate control rods or burnable absorbers for any calculations. For the

Summary, conclusions, and future work

Using a simplified 350-MWth mHTGR fuel block Serpent neutronics model, we showed that advanced beryllium- and hydride-based moderators provide increases in cycle length and reduced overall fuel costs relative to a graphite moderated reference mHTGR. The 60% MgO 39% BeO volume fraction moderator achieved similar fuel cycle performance to the graphite reference case for a graphite optimized design and a 60% MgO 39% Be or 84% MgO 15% ZrHx=1 vol fraction moderator exhibited improved fuel cycle

CRediT authorship contribution statement

Edward M. Duchnowski: Formal analysis, Investigation, Software, Writing - original draft. Robert F. Kile: Formal analysis, Investigation, Software, Writing - review & editing. Lance L. Snead: Conceptualization, Writing - review & editing. Jason R. Trelewicz: Conceptualization, Funding acquisition, Writing - review & editing. Nicholas R. Brown: Conceptualization, Software, Formal analysis, Writing - review & editing, Supervision, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the Advanced Research Projects Agency – Energy (ARPA-E) Modeling Enhanced Innovations Trailblazing Nuclear Energy Reinvigoration (MEITNER) program under contract DE-AR000087. We would also like to acknowledge and thank Seokbin Seo of the University of Tennessee, Knoxville for his efforts in providing thermal properties for the thermal-hydraulics evaluation of the proposed moderators found in the literature and Gerhard Strydom of Idaho National Laboratories for

References (36)

Cited by (0)

View full text