Thermal shock fracture of hot silicon carbide immersed in water

https://doi.org/10.1016/j.jnucmat.2015.09.020Get rights and content

Abstract

High purity CVD-SiC, considered as a nuclear grade cladding material, exhibits thermal shock tolerance ∼1260 °C in room temperature water and beyond it (>1260 °C) in saturated water. Being thinner than the tested specimen thickness (1.5  mm × 2.0  mm), the actual cladding (0.57 mm) is anticipated to exhibit enhanced thermal shock tolerance. This implies that thermal shock alone may not shatter the SiC cladding in reflood. Level of fuel rod internal pressure will be a decisive factor in predicting cladding fracture during reflood. Decreasing water subcooling significantly reduces thermal shock fracture danger of ceramic materials. Thermal shock experiments showed strength retention for both pressureless sintered-SiC and CVD SiC, as well as Al2O3 samples quenched from temperatures up to 1260 °C in saturated water. Solid–liquid contacts during nucleate and transition boiling, and boiling incipience upon water bath entering are a highly probable origin of thermal shock fracture in water quenching.

Introduction

Silicon Carbide (SiC) is being considered as a potential replacement for the current zirconium-based cladding of LWR fuel to make the fuel more accident-tolerant [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32]. SiC shows excellent high temperature and irradiation tolerance [11], [12], [20], [21], [22], [23], [24], [25], [26], [27] as well as 2–3 orders of magnitude slower oxidation rate compared to zirconium-based alloy at 1200 °C [9], [14], [18]. In addition, SiC absorbs fewer neutrons than Zirconium-based cladding [4], [5]. Those identified characteristics make SiC a promising cladding candidate for the development of an accident tolerant fuel for LWRs under beyond design basis accidents (DBAs) without a need for large changes in the current core design.

Yet, SiC is a brittle ceramic material that lacks ductility. This implies that the safety evaluation of SiC clad LWR fuel should be considered in a drastically different framework from the current practice for the metallic and ductile Zirconium-based cladding. Retention of cladding ductility is a key safety metric for Zirconium-based clad fuel, given that not enough load-bearing capability is expected by elastic deformation of embrittled phase alone. The SiC fracture, initiated by a local critical flaw, is inherently statistical whereas the Zirconium-based alloy failure occurs in a deterministic manner, thanks to the plasticity-mediated dispersed fracture. Recognizing such a fundamental difference in strength-failure between SiC and Zirconium-based alloy, increasing attention is being made to structural failure analysis of SiC cladding [7], [17] at steady-state operation as a starting point.

Under accident conditions, a fuel cladding may experience significant stresses beyond the level of usual steady-state operation due to thermal shock in the reflooding phase of a loss of coolant. In general, thermal shock fracture often sets the maximum service temperature for high temperature ceramic applications below the melting point. Indeed, thermal shock fracture is also a limiting failure mode for the Zirconium-based alloy cladding in accidents involving reflood quenching [33]; brittle fracture caused by oxygen-affected brittle phase and oxide layer (ZrO2) in the cladding serves as a basis for the current emergency core cooling system (ECCS) criteria for loss of coolant accidents (LOCA). That is, the current 17% equivalent cladding reacted (ECR) limit and 1204 °C of peak cladding temperature are considered to ensure retention of cladding ductility and strength even after thermal shock [34]. In case of SiC, which inherently lacks ductility, occurrence of brittle fracture under reflood quenching should be carefully addressed. The fuel, ECCS, and reactor core designs should ensure that the reflood quenching in loss of coolant accidents do not fracture the cladding. It should be noted that in GENIII + reactors, the core is expected to remain covered with water after a LOCA, and thus the fuel does not reach very high temperatures, and no quenching need to be addressed.

In this early stage of the concept development and feasibility study, basic phenomenological aspects of silicon carbide fracture behavior subject to water quenching should be explored. The goals of this study are to (1) experimentally explore thermal shock tolerance of SiC in different subcoolings of water bath relevant to probable reflooding scenarios (saturated and room temperature) and (2) investigate a heat transfer origin (responsible heat transfer modes) of thermal shock fracture.

Section snippets

Experiments

An experimental facility, which was originally designed to run pool boiling experiments [35] was modified to bring specimens up to 1500 °C and drive them into a pool of water, as illustrated in Fig. 1. Specimens were suspended in air inside a quartz tube located at the center of a furnace. A K-type thermocouple wire connected to an air-pressure driven alumina rod was used to hold the specimen. By employing bottom-flooding, this experiment has similar experimental designs/conditions to what was

Strength degradation by quenching

Fig. 3 and Fig. 4 show the sample strength with respect to different quenching temperatures and subcooling (room temperature water, 22 °C or saturated water 100 °C) for tubular pressureless-sintered SiC and bar-shape CVD-SiC, respectively. The first point in the graphs shows the original strength of the material. For those sintered-SiC samples that were quenched in room temperature water, substantial strength degradation was observed. The strength degradation starts at around 350 °C as far as

Subcooling effects

The heat transfer coefficient on a solid surface varies very rapidly upon quenching, imposing different levels of stresses during a short period of time. In that sense, the heat transfer origin of a brittle fracture may be appropriately addressed with detailed understanding of transient heat transfer modes. With this enhanced understanding of transient heat transfer modes, the brittle fracture of a solid material upon quenching can be addressed in terms of presence of a critical pre-existing

Conclusions

The following points summarize key conclusions of this work:

  • High purity CVD-SiC, considered as a nuclear grade cladding material, exhibits thermal shock tolerance ∼1260 °C in room temperature water and beyond it in saturated water (>1260 °C). Being thinner than the tested specimen thickness (1.5  mm × 2.0  mm), the actual cladding (0.57 mm) is anticipated to exhibit enhanced thermal shock tolerance beyond the experimentally confirmed temperature in this study.

  • Saturated water, in comparison with

Acknowledgments

Professor Mujid S. Kazimi suffered a heart attack and passed away while this article was being prepared. The authors acknowledge Professor Kazimi for his insights and guidance that have pervaded every corner of this work.

References (46)

  • Y. Lee et al.

    J. Nucl. Mater.

    (2015)
  • Y. Lee et al.

    Nucl. Eng. Technol.

    (2013)
  • L.L. Snead et al.

    J. Nucl. Mater.

    (2007)
  • L.L. Snead et al.

    J. Nucl. Mater.

    (2011)
  • L.L. Snead et al.

    J. Nucl. Mater.

    (2014)
  • K.A. Terrani et al.

    J. Nucl. Mater.

    (2015)
  • S.J. Zinkle et al.

    J. Nucl. Mater.

    (2014)
  • M. Ben-Belgacem et al.

    J. Nucl. Mater.

    (2014)
  • B.A. Pint et al.

    J. Nucl. Mater.

    (2013)
  • Y. Katoh et al.

    J. Nucl. Mater.

    (2014)
  • Y. Katoh et al.

    J. Nucl. Mater.

    (2011)
  • Y. Katoh et al.

    J. Nucl. Mater.

    (2007)
  • Y. Katoh et al.

    J. Nucl. Mater.

    (2010)
  • Y. Katoh et al.

    J. Nucl. Mater.

    (2013)
  • Y. Katoh et al.

    J. Nucl. Mater.

    (2014)
  • A.G. Perez-Bergquist et al.

    J. Nucl. Mater.

    (2015)
  • Y. Katoh et al.

    J. Nucl. Mater.

    (2015)
  • Y. Katoh et al.

    J. Nucl. Mater.

    (2014)
  • C.M. Silva et al.

    J. Nucl. Mater.

    (2015)
  • S. Kondo et al.

    J. Nucl. Mater.

    (2015)
  • D. Kim et al.

    J. Nucl. Mater.

    (2015)
  • H. Kim et al.

    Int. J. Heat. Mass Transf.

    (2010)
  • T.J. Lu et al.

    Acta Mater.

    (1998)
  • Cited by (18)

    • Effect of cooling rate on the residual ductility of Post-LOCA Zircaloy-4 cladding

      2020, Journal of Nuclear Materials
      Citation Excerpt :

      Yet, the effects of resulting microstructures under different cooling rates on the cladding’s post-LOCA mechanical behavior have not been investigated. As for the second effect, recent advances on thermal shock fracture studies revealed the importance of heat transfer rate on brittle materials’ thermal shock fracture upon water quenching [19–21]. Chemical Vapor Deposited (CVD) SiC quenched in saturated water (∼100 °C) exhibited superior thermal shock tolerance compared to specimens quenched in the room temperature water, owing to the promotion of film boiling that lowers transient temperature gradient within the solid [19].

    • Emergency core cooling system performance criteria for Multi-Layered Silicon Carbide nuclear fuel cladding

      2019, Nuclear Engineering and Design
      Citation Excerpt :

      SiC has a higher decomposition temperature and better oxidation resistance compared to Zr (Lee et al., 2015). However, being a ceramic material, SiC is susceptible to brittle fractures which can cause the cladding structure to fail stochastically (Lee et al., 2015). To overcome this limitation, researchers proposed multi-layered SiC clad structures (Lee et al., 2017).

    • Experimental and analytical investigation into boiling induced thermal stress: Its impact on the stress state of oxide scales of nuclear components

      2019, Nuclear Engineering and Design
      Citation Excerpt :

      Although boiling under various conditions has been studied intensively in the last three decades (Giraud et al., 2016), knowledge about the thermal stress caused by temperature fluctuation during the boiling process is scarce. The significant wall temperature fluctuations inducing thermal stress oscillations may cause material thermal fatigue damage (Giraud et al., 2015; Klevtsov and Crane, 1994; Lee et al., 2016; Lee et al., 2013, 2015). Oxide scales protect underlying metallic materials from corrosion.

    View all citing articles on Scopus
    1

    The first author's current affiliation is Korea Advanced Institute of Science and Technology (KAIST), Department of Nuclear and Quantum Engineering, Daejeon, 305–701, South Korea.

    2

    Deceased, 30th of June 2015. Professor Mujid S. Kazimi passed away while this article was being prepared with his inputs. This work was conducted under his supervision.

    View full text