Around 40 years ago, in the mid-1980s, Swedish government approved the KBS-3 method for the direct disposal of spent nuclear fuels (SNF) in Sweden. Since then, this method has become a reference for many countries including Korea, Republic of. The main ideas of the KBS-3 method are to locate SNF at 500 m below the ground surface using a copper disposal canister and a bentonite buffer. In 2016, our government announced the National Plan (NP 2016) regarding the final management of high-level waste (HLW) in Korea. In 2019, new committee were organized to review the NP 2016, and they submitted the final recommendations to the government in 2021. Finally, the government announced the 2nd National Plan in December, 2021. So far, KAERI has developed the technologies related to the final management of SNF in two directions. One follows ‘direct disposal’ based on the KBS-3 concept, and the other ‘recycling’ based on ‘pyroprocessing-and-SFR’ (PYRO-SFR). Even though Posiva and SKB obtained the construction permits with the KBS-3 method in Finland and Sweden, respectively, there are still several technical obstacles to applying directly to our situations. Some examples are as follows: high burnup, huge amounts of SNF, and high geothermal gradient in Korean peninsula. In this work, we try to illustrate some limits of the KBS-3 method. Within our country, currently, the most probable disposal option is the KBS-3 type geological disposal, but no one knows what the best option will be in 20 or 30 years if those kinds of drawbacks are considered. That is, we compare the effects of the drawbacks using our geological data and characteristics of spent fuels. Last year, we reviewed alternative disposal concepts focusing on the direct disposal of SNF and compared the pros and cons of them in order to enhance the disposal efficiency. We selected four candidate concepts. They were multi-level disposal, deep borehole disposal, sub-seabed disposal and mined deep borehole matrix. As mentioned before, KAERI has developed a pyroprocessing technology based on the SFR to reuse fissile radionuclides in SNF. Even though we can consume some fissile nuclides such as 239Pu and 241Pu using PYRO-SFR cycle, there still remain many long-lived radionuclides such as 129I and 135Cs waiting for the final disposal. The authors review and propose several concepts for the future final management of the long-lived radionuclides.

Many countries plan to dispose of spent nuclear fuel through deep geological disposal system. In Korea, a plan is being established for the construction of a deep disposal facility to dispose of highlevel radioactive waste (or spent nuclear fuel). For construction of a deep geological repository, the NSSC (Nuclear Safety and Security Commission) stipulate that detailed technical standards for location, structure, and disposal system of deep geological repository are determined and announced by the Nuclear Safety and Security Commission Notification. Therefore, the regulatory body should carry out the process of regulatory review whether the technical standards developed by the implementer are suitable for the IAEA’s recommendations and guidelines and domestic conditions. In this process, there are many difficulties and uncertainties in terms of time and cost to independently develop safety factors in Korea by referring to the IAEA reports. So, this study intends to investigate and analyze regulatory cases for important safety factors through cases of overseas leading countries in deep geological disposal project. There are two regulatory cases intensively investigated in this study. The first is a regulatory case of regulatory bodies and external experts on the safety case, and the second is a regulatory review case in the process of site selection factor selection. In case of regulatory review of safety case, Sweden and France were selected as the representative target countries. In Sweden, safety cases such as SR-97, SR-Can, and SR-Site have been developed and there are cases of active regulatory review by regulatory agencies in the RD&D process. In France, several safety cases based on sedimentary rocks were developed and the OECD/NEA IRT (International Review Team) was inquired for review for each safety case. The site selection process is divided into a preliminary site selection stage, a site investigation stage, and a site selection and application stage. In each stage, evaluation to select a safe site is carried out using allocated siting factors of that stage. The IAEA SSG-14 report describes aspects that implementers consider in the site selection process and, with this reference, many countries are developing various siting factors and assessment methodologies in consideration of their domestic bedrock condition and geological positions. As a representative example, in Japan which is highly affected by earthquakes and igneous activities, the siting factor is classified into EF (Evaluation Factors) and FF (Favoulable Factors). So, site assessment is conducted preferentially using EF related to earthquakes and igneous activity.

Corrosion cells that simulates engineering barrier system have been stored in an aerobic KURT environment for 10 years, which were recovered and dismantled in 2021. The test specimens were compressed copper (Com. Cu), Cold spray copper (CSC Cu), Ti Gr.2, STS 304, and Cast nodular iron. The specimens were buffered by compact Ca-type Gyeongju bentonite (KJ-I) and compact Na-type Wyoming bentonite. And the corrosion cells were exposed to KURT groundwater at 30°C for about 10 years (3,675 days). As a result of the long-term experiment in aerobic environment, it was confirmed that Na-bentonite is more advantageous for inhibiting corrosion than Ca-bentonite. The corrosion thickness of the most specimens in Ca bentonite was slightly lower than in Na bentonite until the initial 500 days, but after 10 years, the corrosion thickness of copper and cast iron specimens in Na bentonite was clearly lower. The corrosion thickness of the copper specimen in Na bentonite was very low about 0.5 um in both Com. Cu and CSC Cu. Moreover, the corrosion thickness in Ca bentonite was very high about 4 um for Com. Cu and 6 um for CSC Cu. In the case of cast iron, the corrosion thickness in Na bentonite was about 13 um, and 15 um in Ca bentonite. The common feature of copper and cast iron specimens in Ca bentonite, which showed a high corrosion thickness, is the forming of a white mineral deposition layer on the specimen surface, which was presumed to be some kind of feldspar. On the other hand, it was found that the STS304 and Ti specimens were hardly corroded even after 10 years. In conclusion, when a white mineral deposition layer was formed on the specimen surface, the corrosion thickness always increased sharply than before, and thus it was estimated that the generation of the mineral deposition layer cause the increase of bentonite permeability, and rather the weakening of existing passive corrosion film.

It can take hundreds of thousands of years for decreasing radiological effects of high-level radioactive wastes to those of natural background radiation. Therefore, long-term time scale should be considered in order to demonstrate performance and safety of deep geological disposal of the radioactive wastes. The changes of surface environment for these long-term time scale can have influence on safety analysis by changing transport path of radionuclides from the radioactive wastes. Changes in climate is considered as one of main factors causing the long-term changes of the surface environment. The own effects and interactions of climate with other components of the geological disposal system are organized in features, events, and processes (FEPs). In this study, some natural processes occurred by changes of climate were suggested and the connectivity between each process is proposed based on the relation of the FEPs concerned with the changes of climate and surface environment. The processes were classified into global and regional/local scales and was analyzed, respectively. Then, the influences of the processes on shallow groundwater and surface water body environment, which might be transport path of radioactive nuclides in local/site scales, were expected. As the proposed connection demonstrate the order or hierarchical relations of the natural processes, it can shows that some output by a certain process may be input of other process connected the former process in numerical simulations for interpreting the processes. If the connection may be considered to be suitable to represent longterm changes of the surface environment, it can be evaluated that the expected scenarios based on the connection is also proper. In addition, it can be helpful in selecting factors to be studied more detailed in terms of climate change for expecting long-term changes in the surface environment by analysis on the input and output data. The results of this study can be used as basic approaches to represent the long-term changes in the surface environment caused by specific natural processes from changes of climate. It will be also helpful for formulating scenarios related to long-term evolution in the surface environment required for performance and safety assessments of the deep geological disposal.

The radioactive waste disposal systems should consist of engineering and natural barriers that limit the leakage of radionuclide from spent nuclear fuel and fundamentally block groundwater from contact with radioactive waste. These considerations and criteria for designing a disposal system are important factors for the long-term stability evaluation of deep geological repository. Colloids and gases that may occur in the near-field and groundwater infiltrated from outside can be means to accelerate the behavior of radionuclide. The gas produced and infiltrated in the disposal system is highly mobile in the porous medium, and reactive gases in particular can affect the phase and behavior of radionuclide. A free gas phase (bubble) can be formed inside the canister if the partial pressure of the generated gas exceeds the hydrostatic pressure. If the gas pressure exceeds the critical endurance pressure of canister and buffer, then a gas bubble may push through the canister perforation and the buffer. It is also known that when gas bubbles are formed, radionuclide or colloids are adsorbed on the surface of the bubbles to enable accelerated movement. An experimental setup was designed to study the acceleration of nuclide behavior induced by gas-mediated transport. A high temperature and pressure reaction system that can simulate the deep disposal environment (500 m underground) was designed. It is also designed to install specimens to simulate gas flow in engineered barriers and natural barriers. The experimental scenario was set based on 1,000 years after the closure of the repository. According to the previous modeling results, the surface temperature of the canister is about 30 to 40 degrees and the gas pressure can be generated between the canister and the buffer is 5 MPa or more. In the experimental conditions, the saturation time of compacted bentonite was measured and the gas permeability of the compacted bentonite according to the dry density was also measured. Further studies are needed on the diffusion of dissolved gas into the compacted bentonite and the permeation phenomenon due to gas overpressure.

FTIR (Fourier Transform Infrared) and Raman spectra of KJ-II bentonite provided by Clariant Korea were compared with those of MX-80 bentonite. The FTIR spectra were obtained using a Nicolet 5 FTIR spectrometer (Fisher Scientific) equipped with a diamond ATR (Attenuated Total Reflection) module. The spectra were collected for 64 scans with a resolution of 4 cm−1. Raman spectra were obtained using an optical microscope (Olympus, BX43) and a spectrometer (Andor, SR- 500). The laser beam was focused using an objective lens with a magnifying power of 50. The backscattered light from the sample was collected into an optical fiber with a core diameter of 0.4 mm. The Raman signals were recorded with CCDs (Andor, DV401A-BV for 532 nm laser wavelength and DV420A-OE for 638 and 785 nm laser wavelengths). Each pixel of CCD received the signal for 1 s and its 1000 times accumulated data were collected. The FTIR spectra of the two bentonite samples are very similar. The FTIR spectra of both bentonites showed absorption bands at 3623, 3399, 3231 cm−1 in the higher wavenumber region and at 1637, 1443, 1117, 997, 914, 887, 847, 797, 611, 515, 414 cm−1 in the lower wavenumber region. A sharp band at 3623 cm−1 and the strong band at 997 cm−1 correspond to the OH stretching of structural hydroxyl groups and the Si-O stretching vibration, respectively. In addition to these clear bands, several absorption bands observed in this experiment are well matched with the results reported in various literatures. Unlike the FTIR spectrum, it is not easy to observe the Raman bands of bentonite. The reason is that strong fluorescence interferes with the Raman spectrum. The two bentonite samples showed different fluorescence intensities. In the case of MX-80 bentonite, no clear Raman band was observed due to the influence of very strong fluorescence. KJ-II bentonite showed a relatively weak fluorescence intensity and Raman bands were partially visible at around 147, 260, 397, 709, and 1279 cm−1. In particular, the band at 1279 cm−1 is wide and sturdy. It was observed that the non-powder samples showed a better quality spectra. The Raman characteristics of KJ-II bentonite, which depend on the incident laser wavelength and the sample pretreatment, are discussed in detail.

Various models have been proposed to describe the swelling behavior of buffer in high level waster repository. One of the most notable models, the Barcelona Basic Model (BBM), is a mechanical model that simulates the behavior of unsaturated ground and is widely applied to soils that undergo large expansion due to water. Among the BBM parameters of Kyeongju bentonite, which is found in Korea, there are no experimental data for parameters that describe the unsaturated state. Such hydromechanical properties should be characterized through experimental programs. However, such experiments are highly complicated and require long periods of time to produce an unsaturated state through different methods according to the suction range. Although there are several studies in which geotechnical parameters were obtained through a back analysis instead of direct experiments, few studies have employed machine learning methods for the identification of geotechnical parameters. In this study, instead of direct experiments, the results of a relatively simple swelling pressure experiment was compared to the numerical analysis results to propose a method of determining some of BBM parameters. Influential factors were identified by a sensitivity analysis and the values of the factors were estimated using an artificial neural network and optimization method. The obtained parameters were applied to the numerical model to estimate the swelling pressure growth, which was subsequently compared to the experimental value. As a result, it was found that there was no significant difference between the two swelling values.

Garnet is one of the promising ceramic waste forms for immobilizing radioactive wastes. It has an A3 [VIII]B2 [VI]T3 [IV]O12 structure, so it can accommodate various cations of different sizes and coordination. Silicon usually occupies the centers of the tetrahedron structural site (T[IV]O4) in natural garnet. However, substitution of the T-site with iron, which has a relatively large ionic radius, causes the expansion of a unit cell volume of garnet and allows the incorporation of large cations such as actinides at other sites. Relatively few leaching data have been reported for ferrite garnet waste forms to date. In this study, we synthesized gadolinium-iron-garnet and evaluated the leaching property using cerium as a surrogate for actinide elements. The test specimens were made by cold pressing and sintering process. Three different standard leaching tests were performed as follows. The PCT-A (ASTM C1285) was performed for 7 days at 90°C to the crushed sample (0.149 to 0.074 mm). The ANSI/ANS-16.1 standard leach test was performed at ambient conditions for 5 days with constant replacement of leachate. Finally, the MCC-1 (ASTM C1220) test was performed for 28 days at 90°C with different types of leachants such as ultrapure water, brine, and silicate water. The last two leaching tests were conducted on monolithic specimens. After the end of the test, leachate was analyzed by inductively coupled plasma mass spectroscopy (Agilent, ICP-MS 7700S).

To dispose of spent nuclear fuel, the most promising method is disposal in a deep geological repository with a multi-barrier system. Among the multi-barrier system, canisters are used to contain the spent nuclear fuel. A role of the canister is to withstand corrosion load from the deep geological environment as possible as long. Corrosion processes consist of corroding agents transport to the canister surface and electrochemical reactions between the corroding agents and the canister surface. According to previous King’s electrochemical experiments, the mass-transport rate of corroding agents is slower than the electrochemical reaction rate with copper when the canister is surrounded by dense bentonite blocks. Therefore, the mass-transport rate is a rate-determining step for the whole corrosion process. Despite of the importance of transportation of oxidizing agents in bentonite, the transportation process was not paid attention. For example, existing models which are called continuum models assumed that the corroding agents pass through the pore in the porous medium because the continuum model does not consider the fracture networks in the bentonite. Here we develop a dualpermeability and dual-porosity model. In this model, the transport of corroding agents is considered that they pass through fracture within the porous medium. The difference between the dual-permeability and dual-porosity model is whether the corroding agents can pass through the pore. The dual-permeability model assumed that the mass-transport occurs within both fracture and porous medium. On the other hand, the dual-porosity model assumed that the mass-transport occurs only within fractures. Through both models, we found that the transport rate in the fractures is much higher than through the pores, and the canister lifetime at a point where contacting the fracture tip is much shorter than other parts when the canister lifetime is calculated by the transport-governed condition. In addition, the temperature distributions in the fracture are different compared to the existing continuum model. Our results show the effect of fractures in terms of not only corroding agents transport but also the canister lifetime. We anticipate our model to be a first step for the corrosion estimation model coupled with fracture networks.

Numerous low-and intermediate level radioactive wastes were generated from the decommissioning processes of nuclear power plants. Radionuclides such as Co and Cs contained in decommissioning wastes should be immobilized to prevent the release of radionuclides from the wastes due to its harmful impacts on ecosystem by high radioactivity and long half-life. Ethylenediaminetetraacetic acid (EDTA) used as decontamination agent can be contained in cement waste during decommissioning process of nuclear power plants. In addition, EDTA can be stably and strongly bound with radionuclides, resulting in the acceleration of the nuclide release from solidified cement matrix. Here, we investigated the effects of EDTA on leaching behaviors of Co and Cs immobilized in the cement specimen. The leaching tests were performed according to the ANS 16.1 “Measurement of the leachability of solidified low-level radioactive wastes by a short-term test procedure”. From the results, an increase in the EDTA content in the cement specimen led to an increase in Co leaching, whereas a decrease in Cs leaching. Leaching of Cs was dominantly controlled by diffusion from the pore space of the cement specimen to the solution. The effective diffusion coefficient and leachability index of nuclide were determined using the diffusion-release models of ANS 16.1. The results of present study can be used in the safety assessment for disposal of the radioactive waste generated by decommissioning of nuclear power plants.

Bentonite is considered as buffer of engineered barrier for retardation of radionuclide migration. Bentonite has low permeability, high swelling and high sorption capacity for radioactive nuclides. Properties have been widely investigated under various geochemical conditions simulating deep geological environments. The chemical stability of bentonite is an important factor in evaluating the long-term stability of the bentonite buffer. However, the presence of impurities in bentonite clays can reduce the retention capacity for retardation of radionuclide migration value of bentonite. Therefore, the bentonite purification is necessary. In the present study, grade improvement of montmorillonite was conducted using ultrasonic and froth flotation methods. As a result of confirming the grade of montmorillonite according to the optimal ultrasonic intensity for ultrasonic irradiation is 1.0 kHz of bentonite in Gyeongju (KJ-II) increased from 60% to 78%. In case of froth flotation method using PSS (0.1 mM) as a reagent, the grade of montmorillonite increased up to 90%.

In recent years, the importance of the thermo-hydraulic-mechanical-chemical coupled processes is increasing in the performance assessment (PA) of the high-level radioactive waste repository. In the case of mechanical behavior, it is very important because it can affect fluid flow and radionuclide transport by changing the porosity and permeability of the medium. In particular, Excavation Damaged Zone (EDZ) should be considered essential in PA because the migration of radionuclide is affected by the enhanced hydraulic transmissivity and altered geomechanical behavior of EDZ. Furthermore, due to various thermo-hydraulic behaviors such as decay heat generated from radioactive waste, pore water pressure increase, and swelling pressure of bentonite buffer material, mechanical evolution is occurred which may change the size and physical properties of EDZ. Therefore, to solve this problem, analysis of coupled thermal-hydraulic-mechanical (THM) processes with the effect of long-term evolution of EDZ due to the mechanical behavior should be accompanied. In this study, numerical model for the long-term evolution due to mechanical behavior considering EDZ using the Adaptive Process-based total system performance analysis framework for a geological disposal system (APro) proposed by the Korea Atomic Energy Research Institute (KAERI). In the case of EDZ, the concept of Mazars’ damage evolution model was applied to simulate the behavior using the continuum model, and the change in hydraulic properties according to the degree of damage was considered. To investigate the importance of mechanical behavior in PA, the results were compared by performing numerical analysis according to the presence or absence of mechanical analysis. Finally, numerical analysis considering the mechanical evolution of EDZ was conducted using the model developed in this study to investigate the effect of EDZ.

The deep geological repository consisting of a multi-barrier system (engineered and natural barriers) is generally designed to isolate the high-level radioactive waste. The natural barrier is outermost portion to secure safety of the disposal. Crystalline rocks are considered for potential geological repository media to retard and inhibit the migration of radionuclides when the radionuclides leak from the canister and break through the engineered barrier. Sorption and diffusion processes play a major role in retardation of the radionuclides in deep underground environment. In order to evaluate the migration of radionuclides in the safety assessment or geochemical modelling, distribution coefficient and diffusivity of radionuclides are required as input data. In this study, we performed mineralogical and geochemical analysis for a crystalline rock (e.g., granite) to use the sorption and diffusion experiment. The fresh rock samples are obtained from a deep core samples (DB-2) drilled up to 1 km from the surface at KURT (KAERI Underground Research Tunnel) site. For the optical and microscopic examination, thin sections of the rock sample were provided. The rock samples were crushed into powder size to analyze major and trace elements of the whole-rock aliquots. The powdered specimens also used for mineral identification and measurement of specific surface area. The major constituent minerals of the granite are plagioclase, quartz, and K-feldspar and the minor minerals are phlogopite, biotite, and chlorite. According to the results of geochemical analysis, the granite specimens generally contain more than 70wt% of SiO2 and 8wt% of total alkali oxides (Na2O + K2O). The trace elements normalized to primitive mantle compositions show positive Cs, Rb, U, K, and Pb anomalies and negative Nb and Ti anomalies. The rock samples have an average density of 2.62 g·cm−3 and an average porosity of 0.222%. The crushed samples represent the specific surface area of 0.2087 m2·g−1 for the 75–150 μm fraction and 0.1616 m2·g−1 for the 150–300 μm fraction by BET method, respectively. The granite specimens will be used for the sorption and diffusion experiments to evaluate the radionuclides’ geochemical behaviors. The mineralogical and geochemical properties provided in this study can be useful in understanding the sorption and diffusion processes of significant radionuclides under the geological disposal environments.

The design of the high-level radioactive waste (HLW) repository is made for isolating the HLW from the groundwater system by using artificial and natural barriers. Granite is usually considered to be a great natural barrier for the HLW repository in various countries including Sweden, Canada, and Korea due to its low hydraulic permeability. However, many fractures that can act as conduits for groundwater and radionuclides exist in granite. Furthermore, the decay heat generated by the HLW can induce groundwater acceleration through the fracture. Since the direction, magnitude, and lasting time of the heat-induced groundwater flow can be differed depending on the fracture geometry, the effect of fracture geometry on the groundwater flow around the repository should be carefully analyzed. In this study, groundwater models were conducted with various fracture geometries to quantify the effect of various properties of fractures (or fracture networks) on the heat-induced groundwater flow. In all models, the pressure around the repository only lasted for a short period after it peaked at 0.1 years. In contrast, the temperature lasted for 10,000 years after the disposal inducing the convective groundwater flow. Single fracture models with different orientations were conducted to evaluate the variations in groundwater velocities around the repository depending on the fracture slope. According to the results, the groundwater velocity on the fracture was the fastest when the regional groundwater flow direction and the fracture direction coincided. In double fracture models, various inclined fractures were added to the horizontal fracture. Due to the intersecting, the groundwater flow velocity showed a discontinuous change at the intersecting point. Lastly, the discrete fracture network models were conducted with different fracture densities, length distributions, and orientations. According to the modeling results, the groundwater flow was significantly accelerated when the fracture network density increased, or the average fracture length increased. However, the effect of the fracture orientation was not significant compared to the other two network properties.

The timescale for the post-closure safety assessment of a deep geological repository ranges from ten thousand to a million year. In such a long period of time, the biosphere inevitably undergoes changes. Therefore, the long-term evolution of a biosphere is recognized as an important issue in the post-closure safety assessment of a deep geological repository for spent fuels. In this study, we reviewed the approaches to address the long-term evolution of a biosphere. The major drivers of longterm evolution of a biosphere are the climate change and the resulting landscape development. They can affect the hydrogeological and hydrogeochemical characteristics of a biosphere, and then the radionuclide migration through the biosphere followed by the exposure doses for the critical groups. In addition, human activities and the social developments can affect the climate change resulting in the long-term evolution of a biosphere. To make a biosphere assessment, the long-term evolution scenarios for the biosphere should be formulated considering these climate change, landscape development, and human activities. In addition, features, events, and processes (FEPs) that affect the long-term evolution of a biosphere should be used. According to the Safety Case reports of Finland, the major long-term evolution scenario drivers of a biosphere are local sea-level change due to climate change and land use related to crop type, irrigation procedures, livestock, forest management, construction of a well, and demographics. The climate change causing the local sea-level change can be simulated using various earth system models such as CLIMBER-2, MPI/UW, and UVic and an icesheet model such as SICOPOLIS. The review results of this study and FEPs related to the climate change, the landscape development, and human activities will be used to formulate long-term evolution scenarios for the safety assessment of a deep geological repository for spent fuels.

APro, developed by KAERI as a process-based total system performance assessment model, can simulate the radionuclide transport affected by thermal, hydraulic, mechanical and geochemical changes that may occurs in the engineering and natural barriers of a geological disposal system. APro targets a large-scale and heterogeneous 3D system that includes more than 10,000 boreholes located about 500 m underground and hundreds of fractures of different sizes distributed within an area of several km2. Simulating transport and reaction phenomena for such a system through the global implicit approach (GIA) may require considerable computational resources or be intractable in some cases. Therefore, APro adopts the sequential non-iterative approach (SNIA), one of the operator splitting (OS) methods, to separate the mass transport and reaction phenomena into independent problems. By using SNIA, the parallel computation performance in APro with multiple cores is expected to be improved. In this study, the effect of SNIA on the parallel computation performance was analyzed through a simple 1D reactive transport problem. Without SNIA, finite difference equations, discretized from the partial differential equations (PDEs) describing the reactive transport problem, have to be solved at once because all dependent variables are nonlinearly and spatially interconnected through reaction and mass transport terms. When the reaction and mass transport terms are separated through SNIA, the mass transport problem can be converted into independent linear equations for each chemical and the efficient linear system solver can be applied to each linear equation. In particular, since the reaction problem is changed to independent nonlinear equations for each node, the parallel computation performance can be greatly improved. To verify this, the 1D reactive transport problem was implemented in MATLAB, and SNIA and GIA were applied to solve the problem. As a result, there was no significant difference in results between SNIA and GIA for proper spatial and temporal discretization, which verified the accuracy of SNIA. In order to see the parallel computation performance, the calculation times for SNIA and GIA with increasing number of cores were measured and compared. As the number of cores increased, the SNIA calculation speed became faster than that of GIA, which verified that SNIA could improve parallel computation performance in APro. In the future, the effect of SNIA on the parallel computation performance will be verified for the numerical analysis of large-scale geological disposal systems.

The research for the safe management of high-level waste in Korea has been conducted by the Korea Atomic Energy Research Institute since 1997, and the results have formed the basis of the national basic plan for the high-level waste management and the revised national basic plan. In the future, it is evolving and developing R&D focusing on securing technologies for demonstration of the disposal technologies and R&D to develop disposal concepts that increase safety and improve efficiency. Efficient management of heat generated from high-level radioactive waste, including spent nuclear fuel, is an important factor in establishing the disposal concepts because it must be in harmony with key factors such as repository layout, waste disposal container specifications, and design and operation for the barriers of the disposal system. For safe and complete isolation of highlevel radioactive waste in the deep geology, the disposal systems that meet the thermal requirements for the disposal system design have been developed by harmonizing the thermal characteristics of engineered and natural barriers in Korea. These disposal systems were based on low burn-up spent nuclear fuel characteristics generated in the early stages of nuclear power generation, and next, based on the high-level wastes from recycling process of the high burn-up spent nuclear fuels, and were the direct disposal systems for the high burn-up spent nuclear fuels. So, it is necessary to track and analyze the change process in the decay heat characteristics of the high-level waste to be disposed of in order to improve the disposal concept, which enhances the safety of disposal and the utilization of the national land. Therefore, in this paper, the process of change in decay heat of reference spent nuclear fuels for disposal applied to the disposal concepts from the initial stage of development of high-level waste disposal technology to the present in Korea is analyzed.

Domain decomposition method (DDM) has been widely employed for the numerical analysis of large-scale problems due to its applicability to parallel computing. DDM divides the modeling domain into a set of subdomains and obtains the entire solution iteratively until the values of each subdomain which are shared with other subdomains, such as boundary values, are converged. Therefore, in general, DDM is a memory-efficient iterative algorithm with inherent parallelism on the geometric level. APro, the process-based total system performance assessment model, aims for simulating the radionuclide transport considering coupled multi-physics phenomena occurring in large-scale geological disposal system, which are inevitably accompanied by huge memory burden. Therefore, DDM is applicable for the large-scale problem of APro and its performance in parallel computing needs to be examined. The DDM solvers provided by COMSOL which constitute APro can be classified into two methods. One is the overlapping Schwarz method that each subdomain overlaps its neighboring domains and the other is the Schur complement method that subdomains are non-overlapping and separated by boundary domains. For the Schwarz method, the additive, hybrid, multiplicative and symmetric methods can be selected according to the solution update scheme. And for the Schur method, the additive and multiplicative ordering options can be chosen for solving Schur complement system. In this study, the calculation efficiency of the DDM solvers in COMSOL and the applicability to the cluster environment were examined. In aspect of efficiency, the memory requirements with different number of subdomains and calculation schemes were compared in a single node. Then, the memory requirements with increasing number of disposal tunnels and deposition holes were investigated in multiple nodes. As a result, on the cluster environment, with the help of distributed memory architecture which enables efficient memory usage, the applicability of DDM solvers to the large-scale problem of APro was confirmed.

APro, a modularized framework of the process-based total system performance assessment, has been developed by KAERI to simulate the radionuclide transport in geological disposal system considering multi-physics phenomena. However, the target problem including more than 10,000 boreholes and over 100,000 years of simulation time is computationally challenging to deal with numerical solvers provided by COMSOL Multiphysics constituting APro. To alleviate the computational burden, machine learning (ML) techniques have been studied to develop a surrogate model replacing the heavy computation part. In recent studies, attempts have been made to integrate the knowledge of physics and numerical methods into the ML model for partial differential equations (PDEs). Unlike conventional ML approaches solely relying on data-driven method, the integration can help to make the ML model more specialized for solving PDEs. The hybrid neural network (NN) solver method is one of the strategies to develop more efficient PDE solver by interleaving NN with numerical solvers like finite element method (FEM). The hybrid NN model on the premise of numerical solver is easier to train and more stable than the purely data-driven model. For example, one previous study has used the hybrid NN model as a corrector for an incomplete numerical solver for the advection-diffusion problem. In every time step of simulation, NN corrects the error of incomplete solution obtained by a relaxed numerical solver with coarse meshing. The simulation in the next time step starts from the corrected solution, so NN interacts with the numerical solver iteratively. If the corrector is successfully trained, the incomplete but fast solver with corrector can provide reliable results comparable to the original massive solver. This study adopts the hybrid concept to develop a surrogate model for the near-field region, which is the heavy computation part in the simulation of geological disposal system. Various incomplete models such as coarse meshing or emptying the borehole domain are studied to construct a hybrid NN solver. This study also covers how to embed the hybrid NN in COMSOL Multiphysics to train and use it during the simulation.

Deep geological disposal (DGD) of spent nuclear fuels (SNF) at 500 m–1 km depth has been the mainly researched as SNF disposal method, but with the recent drilling technology development, interest in deep borehole disposal (DBD) at 5 km depth is increasing. In DBD, up to 40SNF canisters are disposed of in a borehole with a diameter of about 50 cm, and SNF is disposed of at a depth of 2–5 km underground. DBD has the advantage of minimizing the disposal area and safely isolating highlevel waste from the ecosystem. Recently, due to an increasing necessity of developing an efficient alternative disposal system compared to DGD domestically, technological development for DBD has begun. In this paper, the research status of canister operation technology and plans for DBD demonstration tests, which subjects are being studied in the project of developing a safety-enhancing high-efficiency disposal system, are introduced. The canister operation technology for DBD can be divided into connection device development and operation technology. The developing connection device, emplacing and retrieving canisters in borehole, adopted the concept of a wedge thus making replacement equipment at the surface unnecessary. The new connection device has the advantage of being well applied with emplacement facilities only by simple mechanical operation. The technology of operating a connection device in DBD can be divided into drill pipe, coiled tubing, free-drop, and wireline. The drill pipe is a proven method in the oil industry, but requiring huge surface equipment. The coiled tubing method uses a flexible tube and shares disadvantages as the drill pipe. The free-drop is a convenient method of dropping canister into a borehole, but has a weakness in irretrievability in an accident. Finally, the wireline method can be operational on a small scale using hydraulic cranes, but the number of operated canisters at once is limited. The test facility through which the connection device is to be tested consists of dummy canister, borehole, lifting part, monitoring part, and connecting device. The canister weight is determined according to the emplacement operation unit. The lifting part will be composed following wireline consisting of a crane, a wire and a winding system. The monitoring part will consist of an external monitoring system for hoists and trolleys, and an internal monitoring system for the connection device’s location, pressure, and speed. In this project, a demonstration test will be conducted in a borehole with 1km depth, 10 cm diameter provided by KAERI to verify operation in the actual drilling environment after design improvement of the connecting device. If a problem is found through the demonstration test, the problem will be improved, and an improved connection device will be tested to an extended borehole with a 2 km depth, 40 cm diameter.