Design and demonstration of a laboratory-scale oxygen controlled liquid sodium facility

Abstract

The control and detection of dissolved oxygen in liquid sodium is important for the mitigation of impurity driven corrosion in containment components used in sodium-cooled fast reactor (SFR) environments. Oxygen purification systems are also necessary in smaller institutions to facilitate benchtop-scale testing of future SFR instrumentation. With interest in increasing laboratory-scale SFR research, this paper focuses on the design and demonstration of a safe and cost-effective purification loop using a cold trap (CT) and a plugging meter (PM) to control and measure oxygen levels. The demonstration of the purification system’s effectiveness to manipulate oxygen concentration was made through a series of validation experiments at several CT set-points in the range of $8–28[\mathit{wppm}]$. Agreement below $14[\mathit{wppm}]$ with a relative error less than 1.8% was observed. These results indicate that an accessible oxygen controlled sodium system can be safely designed and operated at relatively low capital investment.

Introduction

In accordance with the renewed interest in commercializing advanced nuclear reactor technologies, the Nuclear Energy Innovation Capabilities Act of 2017 directed the United States Department of Energy (U.S. DOE) to assess the mission need for a reactor-based fast neutron source to be used as a national user facility (Crapo, 2017). In 2019, the DOE launched the Versatile Test Reactor (VTR) program to help close the technical capability gap in the implementation of advanced nuclear reactor fuels, materials and instrumentation (Yvon, 2017). The experimental fast-neutron reactor’s conceptual design plans to utilize knowledge obtained from matured pool-type sodium-cooled fast reactor (SFR) technology. Since the 1950s, these reactors have acquired over 400 cumulative operating years of experience worldwide (Andrus et al., 2019). Prior to 1994, the U.S. has operated four SFR national user facilities including the Sodium Reactor Experiment, Fermi I, the Experimental Breeder Reactor II, and the Fast Flux Test Facility; in the 25+ years since, domestic advanced reactor technological progress has relied on international support for experimental testing (Koch, 1951, Furukawa et al., 2009).

Early interest in the use of liquid sodium as a primary working fluid for advanced nuclear reactors has grown due its exceptional thermophysical and neutronic properties in high-temperature, fast neutron environments. Sodium does not moderate fast neutrons as effectively as water which allows for higher burn-up through the use of fast breeder reactor designs (Kubo, 2012). Furthermore, SFRs are able to operate at low pressures because the prototypical outlet temperature ($\approx 550[C°]$) remains well below the boiling point of sodium ($\approx 883[C°]$) (Rao, xxxx). In addition, increased power density in SFRs leveraged from sodium’s high thermal conductivity and adequate heat capacity allows for smaller fuel bundle packing arrangements, thereby promoting a smaller reactor core footprint (Foust, 1972). Lastly, it has been reported that chemistry-controlled liquid sodium does not impact the mechanical properties of austenitic and ferritic steels (Furukawa and Yoshida, 2012, Hanrahan, 2007, Furukawa et al., 1966). The combination of these positive attributes has allowed SFR technology to steadily increase over the last several decades.

Nevertheless, the use of sodium for cooling applications comes with unique safety challenges. As an alkali metal, liquid sodium is highly chemically reactive with water and moisture at elevated temperatures (Eduful et al., xxxx). In the presence of air at a temperature above $200[°C]$ sodium will ignite and burn; in the presence of water, liquid sodium will violently explode (Borgstedt and Mathews, 1987). Therefore it is critical that the containment materials in contact with sodium have the structural integrity to prevent sodium reaction accidents over the service life of the system. It is not to intention of this note to discuss sodium safety in the context of large thermal hydraulic system development however it is useful to mentions information that can contribute to the safety in the advanced nuclear industry. The Nuclear Regulatory Committee is working closely with The U.S. DOE to help provide risk-informed decisions in sodium development (Doane, 2020).

The presence of nonmetallic impurities in sodium environments has been observed to negatively affect both the short-term and long-term operation of SFR systems. The impurity with the highest impact is oxygen. As a reducing agent, sodium has a powerful affinity for oxygen. Oxygen can exist in liquid sodium in several specific solid forms including sodium monoxide (${\mathit{Na}}_{2}O$), sodium peroxide (${\mathit{Na}}_2{O}_2$), sodium superoxide (${\mathit{NaO}}_2$), and sodium ozonide (${\mathit{NaO}}_3$). Also, if impurities such as hydrogen, nitrogen, carbon or moisture are present, more complex compounds such as sodium hydride (NaH), sodium hydroxide (NaOH), sodium carbonate (${\mathit{NaCO}}_3$), and/or sodium nitride (${\mathit{Na}}_x{N}_y$) will be present. All of the listed compounds can combine and interact of make larger oxide particles. In isolated systems with liquid sodium temperatures below 250 [$°C$] and oxygen concentrations below 50 [wppm] (typical SFR levels are controlled below $4[\mathit{wppm}]$) – sodium monoxide (${\mathit{Na}}_{2}O$) precipitates result as the main product in the equilibrium reaction of oxygen and liquid sodium (Foust, 1972).

Furthermore, large quantities of suspended ${\mathit{Na}}_{2}O$ (concentrations above $50[\mathit{wppm}]$) can cause permanent plugging in restricted flow regions of a sodium loop (Foust, xxxx). It should also oxygen concentration has been shown to increase steady-state corrosion of various types of structural alloys (Rivollier et al., 2019, Furukawa and Yoshida, 2012). Yvon noted that “in order to avoid issues arising from solid oxides circulating in the reactor, the oxygen concentration in the liquid Na should be kept always below its solubility limits” and that “from a corrosion and mechanical resistance point of view it is preferable to keep the amount of oxygen in liquid Na below $3[\mathit{wppm}]$ (Yvon, 2017).” Therefore, integrated chemistry controlled purification systems are required to prevent these issues from arising in long-term flowing sodium thermal hydraulic systems.

Durable instrumentation capable of making real-time determination of oxygen in alkali metal systems has been an ongoing area of research since the conception of alkali systems. Several viable physical methods previously studied include plugging meters (PMs), Blake Meters, freezing point depression techniques, spectroscopy, and electrochemical oxygen detection methods (Mausteller et al., 1967, Davis, xxxx). The wide-spread use of PMs in the sodium field stems from their simple design and reliability. Traditionally, the tandem use of a PM and CT together give sodium experimentalists the ability to leverage oxygen’s limited solubility in sodium for both oxygen control and detection.

As a component of the larger sodium system, purification bypass loops are designed to take advantage of sodium’s oxygen/sodium oxide solubility relationship with temperature. Above $2[\mathit{wppm}]$, the solubility of oxygen in sodium begins to have a distinguishable functional dependency with the sodium’s internal temperature (Borgstedt and Mathews, 1987). The thermodynamic equilibrium of dissolved oxygen and sodium to form sodium oxide can be expressed generally as,

$$4{\mathit{Na}}_{[l]}+O_{2[d]}\rightleftharpoons 2{\mathit{Na}}_2{O}_{[s]}$$

Extensive research has been conducted to determine the solubility of oxygen in sodium. Taking into account over 107 solubility determinations, Atomics International has established the following empirical relationship between oxygen solubility in sodium (S) and temperature ($T_{\mathit{sat}}$) (Eichelberger, xxxx):

$${\mathit{log}}_{10}(S)=6.239-\frac{2447}{T_{\mathit{sat}}[K]}$$

This aggregated correlation allows the calculation of oxygen concentration by the determination of the saturation temperature from the purification system. Specifically, plugging meters are devices that can measure the oxygen saturation temperature in sodium by the evaluation of flow-rate behavior during a plugging event. Furthermore, cold trapping takes advantage of sodium’s temperature dependent oxide solubility to filter impurities from the facility.

Currently only a small number of U.S.-based sodium facilities are capable of conducting liquid sodium research. This is due to both the safety risk from aforementioned sodium hazards and the perceived complexity of instrumentation required to provide oxygen control and monitorization. In collaboration with the University of Wisconsin-Madison, Fort Lewis College (FLC) has recently validated a safe and cost-effective, oxygen-controlled sodium facility which can be used to conduct a wide range of sodium-compatible instrumentation testing experiments. This report seeks to help streamline sodium system development and reduce the barrier to entry for smaller startup companies and university-sized laboratories by detailing a representative sodium facility design. In addition, insight behind the operation of this facility has been dispensed through a modern demonstration of the purification system’s functionality by assessment of agreement between its plugging meter and cold trap.