The influence of the crust layer on RPV structural failure under severe accident condition

https://doi.org/10.1016/j.nucengdes.2017.02.033Get rights and content

Highlights

  • The crust layer greatly affects the RPV structural behavior.

  • The RPV failure is investigated in depth under severe accident.

  • The creep and plastic damage mainly contribute to RPV failure.

  • An elastic core in RPV wall is essential for ensuring RPV integrity.

  • The multiaxial state of stress accelerates the total damage evolution.

Abstract

The so called ‘in-vessel retention (IVR)’ is regarded as a severe accident (SA) mitigation strategy, which is widely used in most of advanced nuclear power plants. The effectiveness of IVR strategy is to employ the external water flooding to cool the reactor pressure vessel (RPV). The RPV integrity has to be maintained within a required period during the IVR period. The degraded melting core is assumed to be arrested in the lower head (LH) to form the melting pool that is bounded by upper, side and lower crusts. Consequently, the existence of the crust layer greatly affects the RPV structural behavior as well as failure process. In order to disclose this influence caused by the crust layer, a detailed investigation is conducted by using numerical simulation on the two RPVs with and without crust layer respectively. Taking the RPV without crust layer as a basis for the comparison, the present study assesses the likelihood and potential failure location, time and mode of the LH under the loadings of the critical heat flux (CHF) and slight internal pressure. Due to the high temperature melt on the inside and nucleate boiling on the outside, the RPV integrity is found to be compromised by melt-through, creep, elasticity, plasticity as well as thermal expansion. Through in-depth investigation, it is found that the creep and plasticity are of vital importance to the final structural failure, and the introduction of crust layer results in a significant change on field parameters in terms of temperature, deformation, stress(strain), triaxiality factor and total damage.

Introduction

A severe accident (SA) of core meltdown is a huge threat to the reactor pressure vessel (RPV) integrity, because the temperature of the melting pool formed by decay heat may exceed the melting point of RPV material, resulting in the release of a large amount of core material into the containment (Mao et al., 2016a). In overcoming this difficulty, a mitigation strategy known as ‘in-vessel retention (IVR)’ is widely adopted in most of advanced nuclear power plants (NPP) (Kulkarni et al., 2013). During the IVR period, the primary system is assumed to be depressurized by the pressure relief facility, meanwhile the lower head (LH) is fully submerged in cavity water prior to the arrival of melting pool on the inside (Mao et al., 2016b). Moreover, a crust is formed between the yet-molten material and RPV material, which greatly affects the RPV failure mechanism under the IVR-ERVC (in-vessel corium retention through external reactor vessel cooling) condition (Lomperski and Farmer, 2007). However, the structural behavior had not been appropriately assessed, especially for the era before Fukushima accident on 2011 (Mao et al., 2016c). The traditional concept of IVR without consideration of internal pressure and corium crust effects is seriously challenged nowadays. In dealing with this issue, the in-depth understanding of the RPV structural failure is highly desirable by incorporating the above factors.

Actually, the safety margin of the RPV had been conservatively evaluated with the thermal-mechanical loads created during the SA of core meltdown, as indicated in most previous studies (Kim and Jin, 1999, Kim et al., 2015). For simplicity, in most cases, the effect of crust layer on the RPV failure prediction had been ignored (Jung et al., 2015), the main reason for which was assumed that the crust was likely to be too weak to support itself above the inner surface of the LH (Almyashev et al., 2016). However, this simplified methodology may results in a totally different prediction on failure site, time and mode, which has been regarded as improper prediction in some accident events (Government of Japan, 2011). As pointed out previously, the strength of a corium crust depended upon its thickness, temperature, chemical composition and morphology (Carénini et al., 2014). Accordingly, the crust configuration was found to be a strong function of melt composition and cooling rate by some researchers (Wang et al., 2013). In accounting for the crust effect on RPV failure, someone assumed that the melt pool was a two-layer one with light metallic crust of Fe-Zr on top and heavy metallic crust of UO2-ZrO2-MxOy on bottom (Kang et al., 2014). In order to accurately describe the crust configuration, a stratified molten pool consisting of a heavy metallic layer (in the bottom), a ceramic layer (in the middle), and a light metallic layer was assumed. As for the crust formation, it was quite complicated. The continuously melted material streamed downwards in the melting pool which was bounded by upper and lower crusts, as shown in Fig. 1. A variety of technical approaches had been proposed for the cooling and stabilization of an in-vessel melt (Park et al., 2013). Usually, they employed water to facilitate crust formation. From the structural integrity perspective, the melting pool attacked on the RPV lower head (LH), and the various heat flux among the crust layer led to a significant change on temperature distribution (Sang and Kune, 2013), indirectly affecting the RPV deformation. In considering the most dangerous thermal failure, the critical heat flux (CHF) was defined as the coolability limit by the cooling water at the sudden transition of the flow regime from nucleate to film boiling (Mao et al., 2016d). As discussed previously, the inner surface of the crust was hot (∼1327 °C) due to it's in contact with the melt pool, and the utmost outer surface was cool (∼150 °C) due to nucleate boiling. Since the crust consisted of mostly of poorly-conducting UO2, the heat transfer between the melt and water was found to be particularly low for thick crust (Sumit et al., 2015). Consequently, the high temperature gradient must result in a various thermal expansion between crust and RPV.

Besides, a strong coherent crust anchored to the RPV wall significantly changes the stress distributions. For some regions, the presence of crust layer prevented the RPV wall from further plastic yielding and creep damage, compared to the one without consideration of crust effect (Theofanous et al., 1997a). Some researchers (Park et al., 2015) pointed out that the crust layer could allow the yet-molten material to fall away from the structure, thereby thermally and mechanically decoupling the melt from the RPV material and sharply reducing the temperature and stress on the RPV. However, it was worth noting that some localized regions still suffered very high stress concentration, especially for the geometric discontinuity (Mao et al., 2016d). In order to take the crust effect into account, the models without dynamic formation of the layer for steady state were developed and used in US NPP (Theofanous et al., 1997b). Actually, the crust formation in homogeneous pool was first introduced by O’Brien and Hawkes for analysis of external cooling of PWR-1000 (Tellier et al., 2015). Subsequently, several numerical and experimental investigations had been performed (Kim et al., 2012, An et al., 2016), and the crust strength tests had been an element of many programs (Zhdanov and Baklanov, 2005). In order to illustrate the structural behavior of the crust layer, the nugget-sized samples and specimens were extracted from the thick crust of several large-scale IVR test (Lomperski and Farmer, 2009, Magallon, 2006). Overall speaking, despite of the fact that the experimental investigation is very important for understanding the influence of the crust layer on RPV failure, the experiment cost is always prohibitively high, and this kind of research works is scarcely found as well as numerical studies. Therefore, the prediction on the failure site, time and mode by using finite element method (FEM) is still highly desirable for ensuring the RPV integrity under core meltdown accident.

Accordingly, the main objective of the present study is to use FEM based on existing continuum damage mechanics (CDM) to investigate the influence of the crust layer on the structural behaviors of the highly-eroded RPV under core meltdown condition. In achieving this goal, two 2D nonlinear FE-models were developed on the platform of ABAQUS software. Taking the RPV without crust as a basis for comparison, the RPV with crust layer was carefully modeled with a characteristic of two-layered structures. In order to reduce some conservatism, the critical heat flux loading was applied on the inner surface of the FE-models, which was considered as a limit boundary thermal condition before the melt-through failure. Besides, the slight internal pressure of 0.3 MPa measured in some experimental tests was also applied. Since the RPVs suffered high temperature gradient from 1327 °C on the inside to 150 °C on the outside, the temperature-dependent material properties were considered in the FE calculation. Moreover, due to the high temperature melt pool, the inner RPV wall of T > 425 °C suffered creep damage during the IVR condition, so the time-dependent creep was also considered in FEM. In the present study, the CHF loading resulted in a local material melting to create a typical RPV configuration with geometric discontinuity. Therefore, the highly-eroded region was found to be performed in a multiaxial manner. Actually, the introduction of crust layer significantly changed the field parameters in terms of temperature, displacement, stress(strain), triaxiality factor and damage. Finally, the present paper provided clear evidence that the elastic layer at around the midpoint was essential for maintaining the RPV integrity, and the creep and plastic damage were the major contributors to total damage.

Section snippets

Creep and plastic damage approaches

As is well known, the severe accident condition leads to the formation of high temperature gradient across the wall thickness from 150 °C on the outside to 1327 °C on the inside. With the further effect from pressurized melting pool, the RPV wall failure is governed by the creep and plastic damage accumulation. In general form, the damage approaches first proposed by Lemaitre are adopted to investigate the RPV structural failure, and the expression is given as follows,Ḋ=FDY,p,D,Yṗ·1-Dwhere Ḋ

The temperature-and time-dependent deformation

As for the severe accident of core meltdown scenario, the so called ‘IVR mitigation’ is assumed to be able to arrest the degraded melting core and maintain the RPV integrity within a period of prescribed time. The melt pool may reach high temperatures due to the significant decay heat, while the outside of RPV is submerged in the water flooding. Usually, the RPV failure may not occur immediately after the occurrence of core meltdown accident. Actually, the RPV failure experiences a temperature-

Conclusions

In the event of core meltdown accident, the essence of the so-called ‘IVR’ mitigation strategy is at least to ensure the RPV integrity under thermal-mechanical loads during first 72 h. Actually, the RPV failure is a very complex process, the failure mechanisms can span a wide range of structural behaviors, such as melt-through, creep damage, plastic yielding as well as thermal expansion. In order to maintain the RPV integrity, the water cooling at the external is usually adopted as a mitigation

Acknowledgements

This work supported by National Natural Science Foundation of China (Grant Nos. 51505425; 51575489), Zhejiang Provincial Natural Science Foundation of China under Grant Nos. LQ15E050007, LY16E050012.

References (32)

Cited by (9)

  • Study on structural integrity of reactor pressure vessel with various ablation under core meltdown accident

    2022, Nuclear Engineering and Design
    Citation Excerpt :

    As previously mentioned, the RPV expands outward and the material bends inward. At this time, the existence of bending moment makes the inner wall of RPV bear compressive stress, while the outer wall bears tensile stress (Mao et al., 2017). Furthermore, the stress concentration is also serious at this site, as shown in Fig. 8.

  • Development of a lumped-parameter code for efficient assessment of in-vessel melt retention strategy of LWRs

    2021, Progress in Nuclear Energy
    Citation Excerpt :

    Therefore, modelling of melt pool heat transfer (which predicts the thermal load) is a key step in the IVR assessment. Researches also focus on the RPV integrity assessment with detailed mechanistic analysis, e.g. (Mao et al, 2017a, 2017b; Fichot et al., 2018). One reasonable concern is that the ‘hot part’ of the vessel would have a negative effect on the overall resistance because of creep and dilatation, thus it could be a serious deficiency of the original safety demonstration in particular for “long term” retention of corium in RPV (Fichot et al., 2018).

  • Investigation on the RPV structural behaviors caused by various cooling water levels under severe accident

    2017, Engineering Failure Analysis
    Citation Excerpt :

    In most cases, the RPV integrity should be ensured during the first 72 h, while the operator may have sufficient time to take some emergence measures [9]. However, due to the complexity of the thermal-mechanical loadings, some of the failure mechanisms are not fully understood due to the low possibility of the occurrence [10,11]. In dealing with the issue of structural safety under SA, numerous efforts have been made by researchers and engineers.

  • The studies on multiaxial creep behavior of inconel 718 notched bar at 700 °c

    2021, Journal of Pressure Vessel Technology, Transactions of the ASME
  • In-vessel core melt retention strategy applied for the Rivne VVER-440 unit

    2020, International Conference on Nuclear Engineering, Proceedings, ICONE
View all citing articles on Scopus
View full text