[en] Highlights: • Angular heat load on reactor vessel by natural convection of oxide pool was measured. • High Ra was achieved by using mass transfer experiments based on analogy concept. • Measured Nusselt numbers agreed reasonably with the other existing studies. • Three different types of volumetric heat sources were compared. • They didn’t affect the heat flux of the top plate but affected those of the reactor vessel. - Abstract: We measured the heat load imposed on reactor vessels by natural convection of the oxide pool in severe accidents. Based on the analogy between heat and mass transfer, mass transfer experiments were performed using a copper sulfate electroplating system. A modified Rayleigh number of the order 10"1"4 was achieved in a small facility with a height of 0.1 m. Three different types of volumetric heat sources were compared and the average Nusselt number of the curved surface was 39% lower, whereas in the case of the top plate was 6% higher than in previous studies with a two-dimensional geometry due to the high Sc value of this study. Reliable experimental data on the angular heat flux ratios were reported compared to those of the BALI and SIGMA CP facilities in terms of fluctuations and consistency.

[en] This study is focusing on the angle dependent heat flux distribution at the reactor vessel plenum due to mixture layer natural convection experiment. We simulated heat transfer using a sulfuric acid - copper sulfate (H_2SO_4 - CuSO_4) electroplating system based on the heat and mass transfer analogy concept. An S-bend shaped copper is used as the volumetric heat source, which is simulated as a heater in previous heat transfer studies. The advantage of mass transfer experiment is the achievement of the high buoyancy condition similar to reactor vessel because of high Pr. This study performed mass transfer experiment using a sulfuric acid - copper sulfate (H_2SO_4 - CuSO_4) electroplating system based on the heat and mass transfer analogy concept. The experimental result was compared with previous 2D study (SIGMA CP)

[en] Highlights: • Natural convection of oxide pool in 3-layer configuration during IVR was investigated. • High Ra was achieved by using mass transfer experiments based on analogy concept. • Heat ratio to light metal layer was 14% higher for 3-layer configuration than 2-layer one. • Heat transfer to heavy metal layer was poor and hence heat load to side wall increased. • Angular heat loads to side wall showed strengthened heat focusing at uppermost location. - Abstract: We investigated the natural convection of the oxide layer in a three-layer configuration of core melts in a severe accident. In order to achieve high modified Rayleigh numbers of 10"1"2–10"1"3, mass transfer experiments were performed using a copper sulfate electroplating system based upon the analogy between heat and mass transfer. Four different cooling conditions of the top and the bottom plates were tested. The upward heat ratios were 14% higher for three-layer than for two-layer due to the reduced heights and the downward heat ratios were lower the same amount. The local Nusselt numbers for the top and the bottom plates were measured and compared with the two layer configuration. To explore the heat load to the reactor vessel, the angle-dependent heat fluxes at the side wall, were measured and compared with the two-layer configuration. Heat load to the side wall and peak heat at the uppermost location were intensified for the three-layer configuration.

[en] Highlights: • Natural convection of oxide pool depending on the Rayleigh number was studied. • High buoyancy was achieved with compact test rig adopting mass transfer experiment. • Trends of total Nusselt number were similar to existing studies. • Split ratio of the heat flux between lower head and top plate was affected by Prandtl number. • Presence of top cooling did not yield significant change on heat flux distribution. - Abstract: The experiments were performed on natural convection heat transfer of an oxide pool using a three-dimensional hemispheric test rig. To achieve a high Ra, mass transfer experiments were conducted based on the analogy between heat and mass transfers. A CuSO_4-H_2SO_4 electroplating system was used. The range of modified Ra was from 8.63 × 10"1"2 to 1.45 × 10"1"5 with a cooling top plate and from 2.44 × 10"1"2 to 4.73 × 10"1"4 with an insulating top plate. The Sc was 2014. The mean and local Nu values of the lower head and top plate were measured and compared with previous reports. The total Nu values agreed well with existing studies, within 10%. The split ratio of the heat fluxes between the lower head and top plate was affected by the difference in the Pr. Heat flux distributions at the lower head and top plate were not affected significantly by the modified Ra, except for the uppermost location of the lower head. The heat flux distribution at the lower head was not affected by the top cooling condition. Heat transfer correlations for mean and local Nu values were developed from the experimental results.

[en] Experimental techniques for serial sectioning and reconstruction were investigated in the three-dimensional imaging of microstructures in metallic materials. Mechanical polishing equipped with a light optical microscope and a digital image processor was used for the serial sectioning and reconstruction. Different processing conditions were investigated for three different alloys: Al-Si alloy, Al-SiC composite, and AZ31 magnesium alloy. Proper selection of polishing cloths and polishing abrasives was pre-requisite for obtaining good slice images. The thicknesses of the slices were precisely controlled for each alloy by changing the mechanical polishing and chemical etching conditions. Two-step polishing was a very effective method for serial sectioning of multi-phase alloys with a soft matrix and hard particles such as Al-SiC composites. Image filtering with a delineate and median filter was very useful for successful image segmentation of multi-phase alloys.

[en] The integrity of reactor vessel is one of the prime concern in a severe accident condition. When the InVessel core melts Retention by External Reactor Vessel Cooling (IVR-ERVC) strategy is adopted as the design concept, the local heat load imposed on the reactor vessel should be identified in order to confirm the integrity of the reactor vessel. There are several studies simulating the natural convection of the oxide pool experimentally. In them, modified Ra (Ra') substitutes conventional Ra in order to represents decay heat of the core melts, due to the self-exothermic condition of the oxide pool. Difficulties in those experiments were the realization of the homogeneous self-exothermic volumetric heat sources. For this reason, the experiments using semicircular facility were also carried out instead of those of hemisphere facility [5-8]. The mean and local Nu of the lower head and the top plate were measured and correlations of the mean Nu were developed in existing studies. However, the comparisons between 2D and 3D results and phenomenological analyses have not been sufficiently performed. In this study we measured and compared the mean and local Nu using 2D and 3D Mass Transfer Experimental Rig for Oxide Pool (MassTER-OP). The experiments were carried out using cupric acid copper sulfate (H_2SO_4-CuSO_4) electroplating system based on the analogy between heat and mass transfer system. The Pr was 2,014 and Ra'H were varied from 7.15X10"1"2 to 3.05X10"1"5.

[en] This study simulated oxide pool using 3D and 2D facilities and performed mass transfer experiments based on the heat and mass transfer analogy concept. We found the existing definition of Ra' is not proper for 2D facility so as to compare with 3D results and redefined Ra'_2_D for 2D facility. There are several studies simulating the IVR-ERVC phenomena with 3D and 2D experimental facilities. Since the mixture layer generates the decay heat continuously, the volumetric heat source should be established. The modified Rayleigh number, Ra' was defined as the product of Ra and Damköhler number to incorporate the internal heat generation. We discovered that the existing definition of Ra' is improper for the 2D tests. This study discusses the problem of the existing definition of Ra'for the 2D experiments and redefines it. We performed mass transfer experiments based on the analogy between heat and mass transfer with 2D and 3D facilities. The experimental facilities are semi-circular for 2D and hemi-spherical for 3D (MassTER-OP3 and MassTER-OP2: Mass Transfer Experimental Rig for a 3D and 2D Oxide Pool). We applied the redefined Ra'_2_D to the all 2D results including MassTER-OP2, and compared with MassTER-OP3 results and 3D correlations.

[en] This paper describes the three-layer phenomena and preliminary plan to simulate the oxide layer experimentally. We will perform the mass transfer experiments using a copper sulfate-sulfuric acid (CuSO_4-H_2SO_4) electroplating system based on the heat and mass transfer analogy concept. By performing the mass transfer experiments, we can achieve the high buoyancy condition with small facilities. The test facility is semicircular whose bottom is chopped, simulating the oxide pool above the heavy metal layer in a three-layer configuration. We will measure the heat flux at the top plate, side wall and bottom plate, and compare these results with those for a two-layer pool. In a three-layer configuration, the upper light metal layer becomes thinner, increasing the focusing effect. Thus, it is important to evaluate the heat flux from the oxide pool to the upper metallic layer. However, there is few heat transfer studies for a three-layer configuration. This paper is to discuss and to make a plan for the heat transfer experiments of oxide pool in a three- layer system. We will perform the mass transfer experiments based on the heat and mass transfer analogy concept. The test results will be analyzed phenomenologically and compared with two-layer results

[en] To investigate the heat loads imposed on a reactor vessel through the natural convection of core melts in severe accidents, mass transfer experiments were performed based on the heat transfer/mass transfer analogy, using two- (2-D) and three-dimensional (3-D) facilities of various heights. The modified Rayleigh numbers ranged from 10¹² to 10¹⁵, with a fixed Prandtl number of 2,014. The measured Nusselt numbers showed a trend similar to those of existing studies, but the absolute values showed discrepancies owing to the high Prandtl number of this system. The measured angle-dependent Nusselt numbers were analyzed for 2-D and 3-D geometries, and a multiplier was developed that enables the extrapolation of 2-D data into 3-D data. The definition of Ra_H^′ was specified for 2-D geometries, so that results could be extrapolated for 3-D geometries; also, heat transfer correlations were developed

[en] This study simulated the IVR-ERVC phenomena using 2D and 3D facilities: MassTER-OP2 and MassTER-OP3 (Mass Transfer Experimental Rig for a 2D and 3D Oxide Pool). The heights were varied: 0.042 m, 0.1 m and 0.167 m. Based upon analogy concept between heat and mass transfer, we performed the mass transfer experiments using a copper sulfate-sulfuric acid (CuSO_4–H_2SO_4) electroplating system instead of heat transfer experiment. The modified Rayleigh numbers ranged from 10"1"2 to 10"1"5 with fixed Prandtl number of 2,014. The measured mean Nu was different comparing with existing studies due to difference of Pr. But the trends were similar. The measured local Nu for the curved surface and the top plate was compared between 2D and 3D geometries. Also, we specified the definition of the modified Rayleigh number allowing the proper comparison of 2D and 3D results. (author)