Fission Gas Research Articles

Analytical local stress model for UMo/Al dispersion fuel

The stress evolution occurring in UMo/Al dispersion fuel is important since it affects the fuel performance during irradiation. In this study, a new analytical model was developed to predict the local stresses in UMo/Al dispersion fuel. In the model, a hypothetical unit sphere composed of a UMo fuel particle, interaction layer (IL), and Al matrix was considered, and the governing equations for the stress-strain relationship, strain-displacement, and mechanical equilibrium were established using a spherical coordinate system. The mathematical derivations were obtained for local stresses in the radial and circumferential direction using a thick-walled sphere model. This analytical model employed the stress distribution as the boundary condition, which is calculated using a finite element model with homogenized fuel meat. The developed model’s solution scheme was verified against Abaqus solutions obtained for two irradiated plates with heterogeneous meat. The model calculated consistent results for interfacial stresses in the IL and Al matrix, indicating that the newly developed model was reliable when simultaneously calculating fission gas pressurization on the UMo/IL/Al composite, and the use of the stress distribution in the homogenized fuel meat as the boundary condition for the analytical model was acceptable.

Modelling intra-granular bubble movement and fission gas release during post-irradiation annealing of UO2 using a meso-scale and spatialized approach

The effective diffusion theory, which is generally used in the modelling of base-irradiation of nuclear fuel, cannot predict the intra-granular fission gas release during post-irradiation annealing tests. From this discrepancy between experiments and usual theory, several alternative scenarios emerged. The purpose of this work is to model these scenarios, as mechanistically as possible, and to distinguish those that could really explain the observations. The difficulty is that the fission gas bubbles in irradiated UO2 are extremely small and numerous (mean distance between nano-bubble centers is only of the order of 10–20 nm) while the grain radius is about 5μm. A new spatialized mesoscale model was developed where individual bubbles are described, along with the diffusion of vacancies from each bubble to the other, as well as from the free surface. Random movement and coalescence of the bubbles have also been included in the model. Based on this principal, two scenarios, and the combination of those, could be assessed: (a) the movement of bubbles in a vacancy gradient, and (b) the Brownian movement of bubbles. It was demonstrated that neither of these two scenarios, nor the combination of them, could explain the large fission gas release obtained during post-irradiation annealing in our reference experiment. This encourages us to consider additional mechanisms, involving dislocations for instance, that could explain the high fission gas release.

Optical system for real-time monitoring of nuclear fuel pellets at high temperature

Understanding the behavior of nuclear materials regarding fission gas release in relation to the different thermal loads to which they can be subjected requires appropriated annealing tests in order to measure both the absolute level and the time dependence of the released fission products together with the corresponding fuel micro-structural changes during representative thermal transients. In this context, we describe in this paper the development and qualification of an experimental platform coupling heat treatment, gas release analysis and optical systems, for the study of fuel pellets at high temperature. This system, which is mainly devoted to power transient and LOCA (Loss Of Coolant Accident) type simulation, is based on an induction furnace to heat the pellet at high temperatures (up to 2000 °C with up to 50 °C/s temperature ramps) in controlled atmosphere. The device is coupled with dedicated analysis loops that are designed to identify and quantify on line the gaseous fission products released during the annealing test. The purpose of the optical system is the real time monitoring of the sample surface to provide additional information (for instance on the micro structure evolution and fuel fragmentation). It is coupled with non-contact temperature measurements by thermal radiation analysis to monitor and control the fuel temperature. Based on experiments conducted on non-irradiated Uranium dioxide samples, we demonstrate the performance of the system in the range 20–1800 °C with 10 μm resolution on a field of 1 cm2, with simultaneous temperature measurements obtained by multispectral pyrometry. We also discuss the limitations of the system as well as further developments for integration in a hot cell for application to irradiated fuels.

CFD study on mechanisms of fission gas burst release from defective fuel rods of a typical PWR

During the reactor operation, it is necessary to estimate the amount of fission gas release from the fuel-cladding gap to the coolant. The most common method for predicting the fission gas release process is based on the first-order kinetic model, which is used for analysis of the steady-state stage of the fission gas release.The purpose of this paper is to reveal the fission gas burst release mechanisms under PWR fuel defective conditions. The fission gas release process has been simulated using the CFD method. A validation calculation has been performed to validate the method. The simulation results are discussed in the paper including the two-phase flow evolution and the amount of fission gas release over time. For short stage after the equilibrium, the gas phase is released in the coolant by the form of jet. While for long stage after the equilibrium, the gas phase is entrained into the coolant by the vortex in the defect site. Sensitivity analysis of short-term fission gas release has been performed to evaluate which parameter critically influenced the release process. The results show that steam generation rate in the gap is the most relevant parameter for the short-term fission gas release rate.

First-principles study of surface properties of crystalline and amorphous uranium aluminides

Surface energy is a crucial material property required for the modeling of fission gas bubble behavior in nuclear materials. However, there is no information available regarding the surface properties of uranium aluminides. In this work we systematically investigate the surface properties of uranium aluminides (UAl3 and UAl4) by first-principles density functional theory (DFT) calculations. A total of 13 surface orientations with a maximum Miller index (MMI) up to 3 are studied for cubic UAl3, while 19 surfaces with a MMI of 2 for orthorhombic UAl4. Using the surface energies predicted by DFT, the surface properties of equilibrium single crystal UAl3 and UAl4, including surface area weighted surface energy, dominant surface orientations and surface anisotropy, are obtained from Wulff construction. To understand gas bubble behaviors in amorphous uranium aluminides under irradiation, we study the surface properties of amorphous UAl3 and UAl4. Compared to the crystalline phases, the amorphous phases are found to have lower surface energies due to the reduced number of breaking U-Al bonds close to the surface. The currently obtained surface properties of UAl3 and UAl4 can be used for the future modeling of gas bubble behaviors in both crystalline and amorphous uranium aluminide fuels.

The use of U3Si2/Al dispersion fuel for high power research reactors

In order to investigate the irradiation performance of U3Si2/Al dispersion fuel at relatively aggressive reactor conditions, compared to how this fuel is typically used in a research or test reactor, post irradiation examination was performed on two fuel plates that were irradiated at ∼270 W/cm2 average surface heat flux and to a maximum local burnup greater than 8 × 1021 fissions/cm3 in the Advanced Test Reactor as part of the RERTR-8 experiment. As part of the non-destructive examinations that were performed on both fuel plates, detailed thickness measurements using a high-fidelity measurement bench were performed on one fuel plate and a segment of the second fuel plate. This paper provides the first reported results of such high-resolution thickness measurements for U3Si2/Al dispersion fuel plates, containing primarily the U3Si2 fuel phase, irradiated at such power levels to high burnup. For destructive examination using optical metallography, samples were generated from both irradiated plates. Localized regions of high swelling (characteristic of internal blistering) were observed on the surface of both fuel plates, suggesting that the performance limits had been reached for each plate. Destructive examinations showed that the fuel particles that are primarily U3Si2, which become amorphous during irradiation, in both plates contained relatively large fission gas bubbles with evidence of fission gas bubble interconnection.

Effects of U-Mo irradiation creep coefficient on the mesoscale mechanical behavior in U-Mo/Al monolithic fuel plates

Abstract The fuel foils in U-Mo/Al monolithic fuel plates will evolve into porous media owing to the formed gas bubbles. Three-dimensional finite element simulations are implemented for the thermo-mechanical behavior in U-Mo/Al monolithic fuel plates with different U-Mo irradiation creep coefficients, and the results of mesoscale normal stress, bubble volume fraction and bubble pressure in the fuel foil are obtained. In the simulations, the mechanistic fission gas swelling model is adopted, considering the dependence of external hydrostatic pressure. The mesoscale normal stress refers to the effective normal stress in the skeleton of porous fuel foil, which relates to the fuel fracture. The numerical results indicate that an irradiation creep coefficient of 2000 × 10−31 m3/MPa can be identified for the current U-10Mo fuel foil, because the calculated fuel-foil thickness increments match well with the experimental data. The influences of U-Mo irradiation creep coefficient on its bubble volume fraction and pore pressure are significant in the vicinity of fuel foil edges and corners. The dangerous regions locate near the fuel foil corners, surrounding the points with maximum mesoscale normal stresses. The enlarged irradiation creep coefficient will result in stress relaxation on the whole, but the through-thickness creep strain components can be heavily magnified. When the irradiation creep coefficient increases from 10 × 10−31 m3/MPa to 1000 × 10−31 m3/MPa, the maximum mesoscale normal stresses will decrease from 58.04 MPa to 28.04 MPa, with a reduction of ∼52%; while the corresponding through-thickness creep strain components increase by ∼5%. With an increase of irradiation creep coefficient from 1000 × 10−31 m3/MPa to 2000 × 10−31 m3/MPa, the maximum mesoscale normal stresses increase by ∼9%, and the corresponding through-thickness creep strain components increase by ∼219%. The irradiation creep performance should be optimized and an extreme large creep coefficient is not a good choice, because the creep-induced damage might degrade the mesoscale strength of the fuel foil.

Fission gas release in the micro-cell fuel pellet under normal operating conditions: A simplified approach based on UO2 pellet experience

Following the Fukushima accident in March 2011, Korea Atomic Energy Research Institute (KAERI) has developed micro-cell fuel pellets for the purpose of reducing gas release, both stable and radioactive, not only under accident conditions, but during normal operation as well. The micro-cell pellet is composed of many cells covered with a ceramic or metallic wall material, whose function is either to lower or block the movement of gas atoms out of the cells, thereby reducing gas release out of the pellet. Based on both a Halden irradiation test of the micro-cell pellets up to 16 MWd/kgU and experience of fission gas release (FGR) in the conventional UO2 pellet, a simplified approach is proposed to estimate FGR in the micro-cell pellets when it is irradiated with a constant linear power of 30 kW/m up to 60 MWd/kgU under normal operating conditions. Considering the lack of measured data on diffusivity in the micro-cell pellets, sensitivity studies for FGR were performed for two cases: diffusivity in the micro-cell pellet is the same as and 10 times higher than in the UO2 pellet, respectively. Even for the 10 times higher diffusivity, FGR in the ceramic micro-cell pellet is at least comparable to or lower than in the UO2 pellet, and FGR in the metallic micro-cell pellet is clearly lower than in the UO2 pellet, due to its lower fuel temperature and the impact of its walls on gas movement. In real situations, the degree of reduction of FGR in the micro-cell pellets would be determined by the fraction of perfect walls and the magnitude of gas diffusivity. Until PIE results become available for the irradiated micro-cell pellets, the current approach provides only rough estimates and therefore its results need to be taken as preliminary.

Cluster dynamics simulation of uranium self-diffusion during irradiation in UO2

As fission fragments pass through UO2 nuclear fuel, a considerable concentration of Frenkel pair defects (i.e. vacancies and interstitials) are created. The steady-state concentration of these defects leads to enhanced uranium self-diffusion, one of several fundamental kinetic parameters that control key engineering properties such as creep and fission gas swelling in UO2 nuclear fuel. A cluster dynamics method to track point defects and defect clusters has been implemented in the MARMOT phase-field code in order to predict as-measured out-of-pile and irradiation enhanced thermal diffusivity. The calculated uranium self-diffusion coefficient compares well with non-irradiated fuel measurements, and shows similar trends to those observed in irradiated fuel, which is a good result given the complexities introduced by non-stoichiometric compositions.

Characterization of PVD Cr, CrN, and TiN coatings on SiC

SiC ceramic matrix composites are a potential replacement for current light water nuclear reactor fuel cladding material. However, loss of fission gas via micro-cracks and corrosion remain an issue. Cathodic arc Cr, CrN, and TiN coatings were deposited on SiC tubes and plates to provide hermeticity and corrosion resistance. These coatings were characterized to determine as-deposited quality. Cross-sectional microscopy, X-ray diffraction, glow discharge optical emission spectroscopy, and scratch tests were performed to evaluate the purity, structure, and mechanical performance of the coatings. Nitride coatings had stable interfaces, but larger defects in the coatings as compared to the Cr coatings which showed cracking at the interface, but less deposition-induced defects. Despite the local state of the interface, the mechanical properties of the metallic coatings versus ceramic coatings enabled the Cr coatings to resist loads three times that of the nitride coatings during scratch tests. Glow-discharge optical emission spectroscopy showed that improvement in elemental purity is needed for future coatings.

Development of a thermo-mechanical coupling code based on an annular fuel sub-channel code

To simulate the thermal hydraulic performance of the annular fuel more accurately, some mechanical models have been incorporated into the sub-channel code SACAF, including pellet displacement model, gap heat transfer model, fission gas release model and fuel-cladding mechanical interaction model. The results calculated by the developed code are in good agreement with that by FROBA-ANNULAR. The developed code can consider the change of the heat transfer coefficient of the inner and outer gaps, the release of the fission gas release, the inner pressure of the fuel, the stress and strain of the cladding, which cannot be calculated by the conventional sub-channel code. Based on the calculation results, a great change is found in the gap heat transfer coefficient and MDNBR during the operation, which implies that the coupling of the sub-channel code and the mechanical performance models is necessary.

Fuel performance optimization of U3Si2-SiC design during normal, power ramp and RIA conditions

Benefited from its higher mechanical strength and much-improved oxidation resistance at very high temperatures (>1200 °C), SiC cladding is one of the promising accident tolerant fuel (ATF) concepts. Considering the thermal conductivity degradation of SiC cladding after irradiation and its negligible creep rate, adopting conventional UO2 with SiC cladding can lead to aggravated fission gas release and elevated fuel centerline temperatures. In this work, U3Si2 fuel with its high thermal conductivity is considered in place of UO2, when adopting SiC cladding. First, the preliminary full core simulation for U3Si2-SiC was conducted to show its holistic fuel performance compared to the current UO2-Zircaloy fuel. Then the most limiting fuel rod was selected based on the calculated chemically vapor deposited (CVD) SiC failure risk from the full core simulation. For the limiting rod, fuel-clad gap optimization was conducted to minimize CVD SiC failure risk. Subsequently, power ramp history was appended to the normal power history and full core simulation analysis for updated U3Si2-SiC design was re-performed to ensure the effectiveness of the chosen gap width. For the limiting rod, reactivity-initiated accident (RIA) analysis at hot zero power (HZP) was analyzed to evaluate its performance and infer its potential failure modes. The results from normal operation and power ramp simulations implied that CVD SiC failure risk was mainly induced by the strong pellet-cladding mechanical interaction (PCMI) and the failure probability of CVD SiC can be effectively minimized to 2.0×10-7 through enlarging the fuel-clad as-fabricated gap width. The leading failure mode for U3Si2-SiC fuel system during RIA changed from fuel melting to Ceramic Matrix Composite (CMC) failure with increasing burnup.

Overview of the SAS4A Metallic Fuel Models and Extended Analysis of a Postulated Severe Accident in the Prototype Gen-IV Sodium-Cooled Fast Reactor

The SAS4A safety analysis code, originally developed for the analysis of postulated severe accidents in oxide fuel sodium-cooled fast reactors (SFRs), has been significantly extended to allow the mechanistic analysis of severe accidents in metallic fuel SFRs. The SAS4A metallic fuel models simulate the metallic fuel thermomechanical and chemical behavior and track the evolution and relocation of multiple fuel and cladding components during the pretransient irradiation and during the postulated accident, allowing an accurate description of the changes in the local fuel composition. The local fuel composition determines the fuel thermophysical properties, such as freezing and melting temperatures, which in turn affect the fuel relocation behavior and ultimately the core reactivity and power history during the postulated accidents. Models describing the fuel-cladding interaction and eutectic formation, the effects of the in-pin sodium on the in-pin fuel relocation, and the postfailure reentry of the molten fuel and fission gas from the pin plenum have also been added. This paper provides an overview of the SAS4A key metallic fuel models emphasizing the postfailure metallic fuel relocation models included in the LEVITATE-M module of SAS4A. The capabilities of the SAS4A metallic fuel models are illustrated through an extended SAS4A analysis of a postulated unprotected loss-of-flow and transient-overpower accident in the metallic fuel prototype Gen-IV sodium fast reactor. The results show that the maximum relative power reached during the postulated accident is 1.19 P0. The favorable characteristics of the metallic fuel cause a significant decrease in net reactivity and relative power due to prefailure in-pin fuel relocation. Negative net reactivity values persist after cladding failure, and the postfailure fuel relocation events occur at low and decreasing power levels.

A molecular dynamics study of the behavior of Xe in U3Si2

Uranium-silicide (U-Si) fuels are being pursued as a possible accident tolerant fuel (ATF). This uranium alloy fuel benefits from higher thermal conductivity and higher fissile density compared to uranium dioxide (UO2). The role of fission gas swelling on the operational performance of U-Si fuels remains an open question, however, fission gas swelling is a critical phenomenon in UO2, U-Zr and U-Mo nuclear fuels. Given the lack of experimental data, in order to study the fundamentals of bubble formation and evolution in U-Si, it is critical that there be an atomistic description of Xe within the U-Si system. In this work, a recently developed U-Si MEAM interatomic potential is leveraged to generate a description of the U-Si-Xe system fit to density functional theory data. The point defect energies of Xe in U3Si2 are determined, in addition to the point defect segregation energy for Xe with respect to two grain boundary orientations. Finally, the properties of small Xe bubbles are analyzed and an equation of state is developed.

Effects of porous fuel structure on the irradiation-induced thermo-mechanical coupling behavior in monolithic fuel plates

A UMo/Al monolithic fuel plate is composed of UMo fuel meat and Al cladding. These fuel plates with a high uranium density have a promising prospect to be used in advanced research and test reactors. Under the neutron irradiation environments, nuclear fissions occur to result in heat generation in the fuel meat; accumulation of fission-induced solid and gaseous products will lead to irradiation swelling; irradiation creep of UMo will affect the mechanical interactions between the fuel meat and the cladding. Especially, intragranular and intergranular gas bubbles will make the fuel meat evolve into a porous structure, and the fuel porosity and pore pressure will change continuously with the irradiation time. As a result, a complex irradiation-induced multi-scale thermo-mechanical coupling behavior needs to be simulated for optimal design and safety evaluation of UMo/Al monolithic fuel plates. In this study, a theoretical model for fuel porosity evolution is developed, based on a fission gas swelling model with the grain recrystallization, the resolution of intergranular gas atoms and the dependence of hydrostatic pressure involved. Moreover, the computation method of microscopic interfacial normal stress is established, with consideration of fuel porosity and pore pressure. The evolution model of fuel porosity is introduced into three-dimensional finite element simulation of the irradiation-induced multi-scale thermo-mechanical coupling behavior, which correlates the current thermal conductivity of UMo fuel with the fuel temperature and porosity. The distribution and evolution disciplines of thermo-mechanical variables in the fuel plate are obtained at different irradiation time. In addition, the effects of fuel porosity on the temperature, the main deformation and microscopic interfacial normal stresses are investigated. Simultaneously, the effects of surface heat transfer coefficient on the thermo-mechanical behavior are analyzed. The research results indicate that the porous fuel structure and fission-gas induced pore pressure become the main fracture mechanisms of monolithic fuel plates.

A new method for solving the fission gas diffusion equations with time-varying diffusion coefficient and source term considering recrystallization of fuel grains

• A new time-discrete method is proposed for solving the intergranular gas atom concentration and flux with gas atom resolution and grain recrystallization considered. • Finite element simulation of the irradiation-induced multi-scale thermo-mechanical coupling behavior in a composite fuel pellet is performed. • It is demonstrated that the proposed new method is an efficient method with high accuracy and acceptable computational cost. • The enlarged fission gas diffusion coefficient resulting from the elevated temperature leads to a considerable increase in fuel swelling. • This new method is necessary to be adopted for in-pile behavior analysis of fuel elements under transient and accident conditions. The fission gas swelling of nuclear fuels depends on the fission gas behavior in fuel grains, which will be accelerated by grain recrystallization occurred in some fuels. Under transient and accident conditions, the temperature and fission rate could vary greatly, which will affect the fission gas behavior in fuel grains significantly. It is necessary to develop a new method to solve the fission gas diffusion equations with these effects involved. In this study, for fission gas diffusion equations with time-dependent diffusion coefficient and source term, a new time-discrete method for solving the intergranular gas atom concentration and flux is proposed. With the grain recrystallization considered, the corresponding time-discrete solutions are obtained respectively for the grains in the recrystallized and un-recrystallized areas. The fission gas swelling can be subsequently calculated out. These new mechanistic models are imbedded in the finite element (FE) simulation of the irradiation-induced multi-scale thermo-mechanical coupling behavior in a composite fuel pellet. FE simulation results demonstrate that: (1) the proposed new method is an efficient method with high accuracy and acceptable computational cost; (2) the enlarged fission gas diffusion coefficient will lead to a considerable increase in fuel particle swelling; (3) for composite pellets with high volume fraction of fuel particles, their radial deformation will be heavily affected when the temperature dependence of diffusion coefficient is involved. It can be concluded that the fission gas swelling computation based on this new solution method is necessary to be adopted, when the in-pile behavior simulation is performed for fuel elements under transient and accident conditions.

Radial microstructural evolution in low burnup fast reactor MOX fuel

Examination of irradiated fast reactor mixed-oxide (MOX) fuel was conducted to analyze microstructural behavior and defect evolution as a function of radial position. The fuel analyzed in this work was irradiated to 3.4% fissions per initial metal atom (FIMA) in the Fast Flux Test Facility. Scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS) were used to analyze microstructural features and fission product secondary phases. A dual-beam focused ion beam (FIB)/SEM was used to prepare several samples along the fuel radius for transmission electron microscopy (TEM) defect analysis. The microstructure near the center of the fuel pellet showed separated grain boundaries accompanied by fission gas bubbles. The fuel pellet did not reach the threshold linear heating rate (LHR) needed for restructuring, formation of the columnar or equiaxed regions was not observed. Metallic fission products aggregated in pores along grain boundaries near the pellet center, but no metallic precipitates were visible in the outer region. Moreover, no appreciable amount of insoluble perovskite phase was formed. The dislocation loop density is highest near the fuel central void but decreases past the mid-point of the fuel radius likely due to the self-irradiation effects. Dislocation lines follow an inverse trend, with the lower density observed near the hottest regions and increasing in the pellet outer radius. These results indicate that the thermal gradient in MOX fuels heavily influences the localized fission product morphology and defect behavior.

Uranium dioxide - Molybdenum composite fuel pellets with enhanced thermal conductivity manufactured via spark plasma sintering

Fuel for nuclear reactors with an increased thermal conductivity offers the potential for lower fuel operating temperatures and reduced fission gas release rates. Uranium dioxide (UO2) based composites could provide a method of achieving a higher thermal conductivity and molybdenum (Mo) has been identified as a potential candidate for use in such composites. Uranium dioxide composites containing Mo were manufactured via spark plasma sintering with concentrations of up to 10 vol% Mo. The composites were characterised on their microstructural properties and the thermal conductivity was determined by laser flash analysis. The composites achieved densities greater than 95 %TD but lower than typical grain size. The microstructure contained channel like structures of Mo, due to the use of an agglomerated UO2 precursor powder. An increased thermal conductivity, proportional to the concentration of Mo, was determined for the composites. At the maximum measurement temperature of 800 ∘ C the increase was found to be 65% and 117% in the 5 and 10 vol% composites respectively.

A study of constituent redistribution in U–Zr fuels using quantitative phase-field modeling and sensitivity analysis

Constituent redistribution in U–Zr fuels produces radially-distributed phase fields, each with different material properties and behaviors. The location and composition of the phase fields evolve dynamically due to the influence of swelling, fission gas release, and sodium infiltration on the thermal conductivity of the fuel. Gaps in the understanding of these processes limit the ability to model higher-level behaviors such as fuel-cladding chemical interaction and perform design and safety analyses with confidence. In the current work, we developed a quantitative phase-field model of macroscale constituent redistribution in the U–Zr system. Model parameters were optimized, and the model was validated against an independent dataset. No modification of the phase transition temperatures was necessary, but diffusion needed to be enhanced to reproduce the observed behaviors. Sensitivity analysis was then used to investigate gaps in the understanding of the system and identify the most impactful parameters and behaviors. The mobile interface between the γ and β+γ phase fields was found to be particularly influential in forming the Zr bathtub. The β and γ kinetic parameters scale the overall magnitude of constituent redistribution, while parameters associated with swelling and sodium infiltration influence the size, shape, and location of the Zr bathtub. These findings were used to justify specific recommendations for improving U–Zr modeling capability, which will expedite fuel development and qualification.

Numerical and experimental study on an isolated bubble in the swirling separator

In the online fission gas removal system applied for thorium molten salt reactor, the vane-type gas-liquid separator was developed to separate the dilute dispersed bubbles from the reactor coolant. For the efficient separation, it’s essential to quantitatively describe the bubble trajectory and the separation length of bubble that for the design of the separator. The separation trajectory of sized spherical bubble ranging from 0.22 mm to 1.05 mm produced by the customized isolated bubble generator were preliminary investigated numerically and experimentally in surrogate molten salt of five concentrations of CaCl2 aqueous solution up to 32.91 wt% covering initial inlet velocity of the separator from 0.842 m/s to 2.526 m/s. A promising numerical simulation method was used to effectively and fast predict the isolate bubble motion in the swirling flow, which combines bubble mechanical model with analytical expressions of liquid velocity field. The visualization experiments using high-speed camera and image processing were implemented to record the bubble trajectory during separation under the same conditions. Both numerical and experimental results indicate that the bubble has a convergent spiral movement during separation before captured by the air core. The separation length Ls increases as the bubble diameter decreases or the initial inlet velocity and the concentration of CaCl2 solution increase. The comparison between numerical and experimental data of Ls shows good agreement and the relative errors of Ls between them are bounded to ±10%.