YU Wenju 1, XIANG Liping 1, HUANG Minzhong 2, LIU Yi 3, YANG Fan 2, FENG Yaxin 3
(1. School of Nuclear Science and Technology, University of South China, Hengyang, 421001, Hunan, China; 2. Xiamen Institute of Rare Earth Materials, Haixi Institutes, Chinese Academy of Sciences, Xiamen 361021, Fujian, China; 3. China Institute of Atomic Energy, Beijing 102413, China)
Extended Abstract:[Background and purpose] Supercritical water-cooled reactors (SCWR), one of the representatives of Generation Ⅳ nuclear energy systems, have attracted much attention, due to their high thermal efficiency and simplified design systems. The operating principle of SCWR is based on the unique physical properties of water above the critical point (374 ℃, 22.1 MPa), where the density, specific heat capacity, and thermal conductivity of water undergo a significant change to give it the solvency of a liquid, as well as the mobility and heat transfer properties of a gas. These properties enable SCWR to provide highly efficient heat transfer and thermal efficiencies in single-phase coolant cycles, typically reaching 45%. However, the extreme operating conditions of supercritical water (high temperatures, high pressures, and strongly oxidizing or reducing environments) pose significant challenges to materials concerning their mechanical properties, radiation resistance, and chemical stability. Therefore, in the design of supercritical water-cooled reactors (SWCR), selecting appropriate structural materials is crucial to ensure the safety and economy of the reactor. This study was aimed to prepare silicon nitride ceramics (HP-Si3N4) by hot press sintering method, with the corrosion behavior and the influence of material properties in a supercritical water environment (450 ℃, 25 MPa) to be evaluated, in order to provide a theoretical basis and a design reference for the application of this material in extreme conditions, such as nuclear energy and supercritical water oxidation.[Methods] Silicon nitride ceramics (HP-Si3N4) were prepared using hot press sintering method. Silicon nitride powder (α-Si3N4, 0.6 μm, purity 93%) was used as the raw material, while yttrium oxide powder (Y2O3, 0.5 μm, purity 99.99%) and magnesium oxide powder (MgO, 0.8 μm, purity 99.9%) were used as sintering aids. The powders with designed composition were mixed by using ball milling with zirconia balls as the grinding media, in anhydrous ethanol, for 6 h. The mixed powder was loaded into a graphite mould pretreated with boron nitride and sintered in a vacuum hot press sintering furnace. The sintering was conducted at 1750 ℃ for 120 min in nitrogen atmosphere. Before corrosion experiments, the sintered specimens were pretreated to sizes of 40 mm×4 mm×3 mm and 10 mm×10 mm×5 mm, followed by grinding and polishing to a mirror finish. The samples were ultrasonically cleaned in ethanol and then placed in a C276 Hastelloy multi-station autoclave, under supercritical conditions of 450 ℃ and 25 MPa, with ultrapure water as medium for 50 h, 100 h, 150 h and 200 h. Before and after test, the samples were weighed with an accuracy of 0.0001 mg, the mass change before and after corrosion was recorded. An X-ray diffractometer (XRD, Miniflex 600, RIGAKU, Japan) was used to examine phase composition of the samples. In addition, field emission scanning electron microscopy (SEM, FEI Talos F200X G2, FEI, USA) was utilized to characterize surface micromorphology of the samples. To evaluate the degree of deterioration in mechanical strength, three-point bending tests were carried out at room temperature, using a universal testing machine (Innovatest Falcon 507), with a fixture span of 30 mm, at a speed of 0.5 mm·min−1. The micro Vickers hardness of the samples was measured using a Vickers hardness tester (Falcon 507), while fracture toughness was obtained by the indentation method.[Results] The main crystalline phase of the uncorroded sample is β-Si3N4, while the sintering aids were not detected, but an intermediate phase Y2Si2O7 was present. Compared with the uncorroded samples, no new phase was found in the XRD patterns after corrosion for different times. Meanwhile, with increasing corrosion time, the intensity of the characteristic peaks of the samples gradually decreased, because only surface layer of the samples was penetrated by the X-ray. With increasing corrosion time, the corroded layer became gradually amorphous, while the depth was gradually increased, thus leading to weakened diffraction signal. According to the surface and cross-sectional SEM images, after 50 h of corrosion, the sample surface was gradually covered by the SiO2 corrosion layer. Some areas were corroded more seriously, evidenced by the large corrosion layer. In the early stage, this SiO2 layer acted at protective film to slow the further oxidation process. As the corrosion time was extended to 100 h, the high solubility of OH⁻ in supercritical water led to gradual dissolution and possible redeposition of the SiO2 layer. This dissolution and redeposition led to the inhomogeneous surface structure, whereas the corrosion layer gradually losed the protective effect on the substrate. When the specimen was corroded for 150 h, the SiO2 layer was delaminated from the substrate and extensively peeled off. As the SiO2 layer was peeled off, the substrate was exposed to the supercritical environment. Due to the high solubility of supercritical water, the grain boundary phases on the surface of the samples were preferentially and selectively corroded, resulting in the exposure of rod-like Si3N4 grains. As the corrosion time was prolonged to 200 h, the corrosion layer will be completely detached, so that the substrate was entirely exposed to the supercritical environment. Therefore, after the matrix was corroded along the crystal, the Si3N4 grains were corroded, resulting in partials dissolving of the exposed Si₃N₄ grains. Pits were present on surface of the matrix. Fracture toughness and hardness of the HP-Si3N4 samples after corrosion for 200 h were not significantly changed. The fracture toughness was increased slightly from 4.1 MPa·m1/2 to 4.5 MPa·m1/2, while the hardness was decreased from 15.2 GPa to 14.9 GPa. In contrast, the bending strength was decreased from 650 MPa to 523 MPa, with a decrease of 18%.[Conclusions] The initial silica (SiO2) corrosion layer formed in the HP-Si3N4 ceramics in supercritical water has a protective effect on the substrate. With increasing corrosion time, the layer gradually dissolved and falled off, due to the high solubility of supercritical water, resulting in direct exposure of the material surface to the corrosive medium, which accelerated the corrosion process. Fracture toughness and hardness of the HP-Si3N4 varied very slightly after corrosion for 200 h. However, the bending strength was decreased significantly, indicating that corrosion substantially affected mechanical properties of the materials. An in-depth exploration of the corrosion behaviour and corrosion mechanism of HP-Si3N4 in supercritical water was explored, providing a theoretical basis for understanding the application of HP-Si3N4 under extreme conditions, which is of great significance for the selection of materials and design optimization in extreme environments, such as the nuclear energy industry and supercritical water oxidation technology.
Key words: Si3N4 ceramics; supercritical water; corrosion; microstructure; mechanical properties