Search

A technique to monitor the permeation behavior of tritium in metals to pure water was successfully developed. A metal membrane separated two containers: one is for tritium loading as an upstream si...

The development of novel tritium breeding materials was urgently needed in order to continuously optimize the tritium breeding ratio (TBR) of thermonuclear fusion reactors. From this point of view, Li4TiO4-Li2TiO3 core-shell breeding materials with more reasonable structure and theoretical Li density of 0.464 g/cm3 were prepared in this work. Notably, the mass transfer experiment at 900 °C in 1% H2/Ar shows that the theoretical Li density of this core-shell material after heating for 30 days was significantly higher than that of other breeding materials, indicating that it can provide more stable and efficient TBR. Specifically, the Li mass loss of the sample after 30 days heating was 3.4%, resulting in a decrease of Li density to 0.415 g/cm3. The mechanism of Li mass loss in Li4TiO4-Li2TiO3 core-shell breeding materials was investigated in detail. Moreover, the samples did not crack or collapse during the long-term heating process, and always maintained a satisfactory crushing load, revealing that this core-shell breeding ceramic can be used for a long time under severe operating conditions.

Production of tritium using a high-temperature gas-cooled reactor (HTGR) has been studied for a prior engineering test with tritium handling and for the startup operation of a demonstration fusion reactor. For this purpose, the hydrogen absorption speed of Zr in a Li-loading rod for the reactor operation is experimentally measured, and an analysis model is presented to evaluate the tritium outflow from the Li rod in a high-temperature engineering test reactor (HTTR). On the basis of the presented model, the structure of the Li-loading rod for the demonstration test using the HTTR is proposed.

In order to efficiently grasp the tritium behavior in advanced core–shell breeding materials, this study adopts a route of injecting tritium out-of-pile to deal with the problems of fewer platforms, long periods, and high costs for traditional neutron irradiation-tritium release experiments. Here, tritium adsorption and desorption experiments were carried out with Li4TiO4–Li2TiO3 core–shell breeding materials before and after long-term heating up to 30 days. The purpose is to study whether the structural changes caused by long-term heating will affect the adsorption and desorption behavior of tritium on the sample surface. The results show that the lack of chemically adsorbed water caused by long-term heating will significantly weaken the tritium adsorption capacity of the sample, but will not affect the desorption behavior of Ar, 1% H2 + Ar and water vapor on tritium, so all samples show a very low tritium retention.

In this study, hydrogen permeation through Flinabe (equimolar LiF + NaF + BeF2 molten salt) including 0.1 wt% Ti particles is experimentally and analytically investigated. Ti particles mixed with Flinabe can suppress H2 permeation for a certain time when the upstream side is supplied with H2. The diffusion time lag after activation process is over 20 times larger than that before activation. H2 diffusion behavior in the Flinabe + Ti system is analyzed using an unsteady-state diffusion equation. The effective diffusivity in the Flinabe + Ti system is around 1/27 of that in the Flinabe system. It is confirmed that the activation of Ti particles is important for the overall permeation process in the Ti-Flinabe system.

Understanding of Li mass transfer behavior in a tritium breeding blanket is an important issue from viewpoints of establishment of tritium cycle and tritium safety. In this work, weight reduction property of Li2TiO3 with excess Li pebbles, which were fabricated by National Institutes for Quantum and Radiological Science and Technology, was investigated and the amount of Li mass loss and the rate of Li mass loss in water vapor atmosphere at elevated temperatures were evaluated. The Li mass loss proceeding at 900 °C with a relatively high rate was limited. The Li mass loss of Pebble210 (Li/Ti = 2.10) was 1.2 wt% and that of Pebbe211 (Li/Ti = 2.11) was 1.4 wt%, eventually. The rate of the Li mass loss increased with increasing temperature and it seemed to increase proportionally to a square root of water vapor pressure. An empirical formula for the Li mass loss was proposed as a function of temperature, heating time and water vapor pressure in a purge gas.

A high-temperature, gas-cooled reactor (HTGR) is proposed as a tritium production device that has the potential to produce a large amount of tritium using the 6Li(n,α)T reaction without major changes to the original reactor core design. In an HTGR design, generally, boron is loaded into the core as a burnable poison to suppress excess reactivity. In this study, lithium is loaded into the HTGR core aiming to produce thermal energy and tritium simultaneously and is loaded instead of boron as a burnable poison. The nuclear characteristics and fuel temperature were analyzed to confirm the nuclear and thermal feasibility of a lithium-loaded HTGR. It was shown that the analysis results satisfied the design requirements and hence the nuclear and thermal feasibility was confirmed for a lithium-loaded HTGR that produces thermal energy and tritium.

Large quantities of tritium are required to start up fusion reactors and conducts engineering tests using tritium for a fusion blanket system. However, tritium is very rare and kg orders of tritium must be produced artificially. Tritium production, by 6Li(n,α)T reaction using a high temperature gas-cooled reactor (HTGR) has been proposed. This method considers the loading of Li rods into burnable poison holes in the HTGR. In this paper, the Li rod suited for use in the High Temperature engineering Test Reactor (HTTR) was designed, and tritium production and leakage from Li-rod capsules were evaluated by adjusting the thicknesses of LiAlO2, alumina, and Zr layers. An irradiation test scenario to be conducted in the HTTR for demonstration of the Li rod’s tritium production and containment performance was presented.

Extraction of hydrogen from a Li-Pb flow to inert Ar one using a random packed tower is experimentally investigated for development of a more efficient tritium recovery system of DEMO fusion reactors. A small-scale tower of 57.2 mm in diameter and 300 mm in height packed with SS304 Raschig rings is supplied with two counter-current flows, which one is liquid Li-Pb absorbing hydrogen of 5.06 kPa in equilibrium pressure from the top of the tower and another is Ar gas from the bottom. Overall mass-transfer rates are experimentally determined as a function of Li-Pb flow rates and Ar gas ones at constant temperature of 673 K. Overall mass-transfer capacity coefficients are determined by fitting a transfer equation of hydrogen in a random packed tower to experimental variations of its concentration in gas under given conditions of the space velocity of liquid or gas, and the values of the mass-transfer capacity coefficient increase with Li-Pb flow rates almost linearly. The hydrogen recovery ratios between the top and bottom of the tower were almost independent of the Li-Pb flow rate, and its values were between 20% and 30%.

Separation of tritium from tritiated water is analyzed based on the theoretical plate model comparatively for the three cases among a water distillation tower, the Girdler-Spevack (G-S) bithermal exchange process and the combined electrolysis chemical exchange (CECE) process. The McCabe-Thiele diagrams to design large-scale detritiation systems are drawn for each, and tritium concentration profiles in each system are compared. It is clarified how detritiation behavior in the distillation tower is enhanced by an increase in the separation factor. The number of theoretical stages and internal flow rates to achieve detritiation of the bottom-to-top concentration ratio of xB/xD = 100 in cases of the G-S and CECE processes, are estimated based on the equilibrium stage separation factor reported in the past. A water distillation tower of 108 mm in diameter and 1000 mm in height packed with structured packing coated with Zeolite 13X are experimentally tested for detritiation of a large amount of wastewater to be exhausted from nuclear reactors. Separation performance between HTO and H2O under reduced pressure is experimentally verified in the water distillation tower. Enhancement of the xB/xD ratio is experimentally proved under the total reflux condition as a function of evaporation rate. Activation on surfaces of adsorbent coated on the structured packings enhances the stage separation factor. The enhancement ratio is affected by liquid-gas dispersion in the column and flow instability in the packed tower.

To launch a fusion reactor stably, it is necessary to prepare sufficient amount of tritium by an external tritium source. Tritium production using nuclear reactions by neutron and Li in a high temperature gas-cooled reactor (HTGR) is an attractive method. An important issue is tritium confinement in high temperature conditions of HTGR. Covering Li compound by Al2O3 is a promising method because hydrogen permeability in Al2O3 is quite low. Furthermore, it is expected that inserting Zr between Li compound and Al2O3 suppresses tritium permeation because Zr has a large hydrogen absorption capacity. For example, a cylindrical multilayer structure consisted of Zr – Li compound – Zr – Al2O3 is a candidate structure to confine generated tritium. To evaluate hydrogen permeation rate through cylindrical Zr material, in this work, hydrogen permeation experiments were carried out for two samples of one-side sealed Zr pipe. Hydrogen permeation rate in one sample was proportional to as the square root of hydrogen pressure but that in another sample, which had retained a certain amount of hydrogen before the experiment, did not indicate an obvious pressure dependence regardless of almost the same experimental procedure except hydrogen permeation direction. Since observed hydrogen permeation rate in Zr was faster than that in Al2O3, under the multi confinement structure by Zr – Li compound – Zr – Al2O3, the generated tritium in Li compound diffuses promptly in whole Zr inside Al2O3 layer and it is expected tritium is stored stably in the structure. Keywords: Hydrogen, Permeation, Zirconium, Pipe

For the decommissioning operation of tritium handling facility, surveillance of contamination with tritium on floor, wall and roof was performed. Significant amount of tritium contamination was detected in approximately 86% of total surface measurement points and maximally 70 mm of scraping was required to eliminate tritium contamination. In some samples for measurement of tritium depth profile, the highest tritium concentration was observed at the inside but not the surface. This measurement result of depth profile is also discussed with our previous calculation results considering diffusion, adsorption, desorption of tritium in the concrete and isotope exchange reaction between tritiated water vapor in the concrete bulk and atmospheric water vapor.

Tritium is required for research and development activities for the deuterium–tritium (DT) fusion reactor and fueling the DEMOnstration Power Station (DEMO). However, tritium is a very rare nuclide and must be produced artificially. Tritium production by loading Li compounds (Li rods) into burnable poison holes of a high-temperature gas-cooled reactor (HTGR) has been proposed (H. Matsuura, et al., Nucl. Eng. Des. 243 (2012) 95–101.). Al2O3 and Zr are used to prevent tritium leaks. Nuclear reaction heat caused by the nuclear reaction (e.g., 6Li(n,α)T reaction) can cause a spatial temperature profile in the Li rods and may change its tritium containment performance, because Al2O3 and Zr performance strongly depend on these temperatures. The effect of nuclear reaction heat by the 6Li(n,α)T reaction on the tritium containment performance of the Li rods was evaluated by simulation. The temperatures of the Li rods for the high-temperature engineering test reactor (HTTR) and gas turbine high-temperature reactor 300 (GTHTR300) increased by 36 K and 46 K, and the leaked tritium decreased by 32% and 37% via nuclear reaction heat, respectively.

Tritium is essential for generating D-T reactions in a fusion reactor, which is expected to be used as a next power generation technology. If tritium is released into the environment due to unexpected accidents, it may transfer to human body through plants eventually. Therefore, it is needed to understand the behavior of tritium in plants from the viewpoint of radiation protection. In this study, an airtight plant cultivation system was constructed and Arabidopsis thaliana was cultivated in the humus supplied with tritiated water. Then, the amount of tritium accumulated in the collected samples was investigated by water immersion, drying, isotope exchange and combustion. It was suggested that the mass transfer rate of exchangeable organically bound tritium to tissue free water by the isotope exchange reaction was faster than the mass transfer rate of tissue free water tritium to the surrounding water. The percentage of non-exchangeable organically bound tritium was in the range from 2.5 to 9.0 % to the total amount of tritium accumulated.

Tungsten (W) is a candidate material of plasma-facing walls for fusion reactors. When a W redeposition layer is formed on the plasma-facing wall by sputtering, the behavior of tritium in the in-vessel components is possible to be affected. Therefore, it is necessary to understand the tritium permeation behavior through the W deposition layer as well as W bulk. In the previous study, the hydrogen permeating through the W deposition layer growing on a nickel substrate was measured experimentally. In this work, the TMAP simulation code was used to analyze the time variation in the measured hydrogen permeation flux and mass transfer parameters were obtained. The results suggested that the permeation flux initially increased with increasing temperature but decreased when the thickness of the W deposition layer was greater than the hydrogen injection range. The obtained recombination coefficient was several orders of magnitude larger than that reported for W bulk. The obtained values were used to predict hydrogen isotope permeation through the plasma-facing wall in the DEMO reactor. The results suggested that the W deposition formed on the plasma-facing wall of the DEMO reactor significantly reduces tritium permeation rate.

FLiNaBe is a promising liquid blanket material for a fusion reactor. Since hydrogen solubility for FLiNaBe is low, one concern is the permeation of bred tritium. In order to increase effective solubility of hydrogen, the addition of Ti powder was proposed. In this study, the influence of Ti addition on the behavior of tritium generated inside FLiNaBe by neutron irradiation was investigated. The samples of FLiNaBe and FLiNaBe mixed with Ti of 5 wt% and 0.5 wt% were prepared and the solid-state samples were irradiated by thermal neutrons at Kyoto University Research Reactor. The neutron-irradiated samples were heated to 700 °C in an Ar gas flow. The released tritium was collected by a water bubbler and the temporal variation of TF and HT release was observed. Initially, the most of tritium was released as HT from both FLiNaBe with and without Ti but eventually, the most of released tritium was HT after long time heating from FLiNaBe with Ti. This result suggests that the addition of Ti can decrease the release of TF very much. The amount of tritium released from FLiNaBe with Ti was lower than that from FLiNaBe without Ti. This result suggests that the addition of Ti can suppress the release rate of tritium from FLiNaBe.

Solid tritium breeding materials are expected to be used in high-temperature conditions for a long time in a fusion DEMO reactor. Thus, it is important to understand how bred tritium releases from the long-term heated material under the environment as close to a DEMO condition as possible to establish a tritium fuel cycle and keep it safe. In this work, the tritium release behavior from the irradiated Li2TiO3 pebbles that was preheated for 720 h at most was observed by heating at 1000 °C under 1000 Pa H2/Ar gas flow. The release peaks of HTO and HT were observed around 300 °C due to the desorption of the chemisorbed water. Also, the broad HTO peak was observed in a higher temperature region despite purging H2/Ar gas. This result suggests that this tritium was released without exchanging with H2 but with combining with oxygen in the pebbles. Moreover, the released tritium amount decreased as the pre-heating time. Finally, the amount of tritium that could not be released by the heating experiment was quantified by dissolving samples with an acid solution. Besides, the total tritium release ratio was discussed.

The production of tritium (T) using high-temperature gas-cooled reactors (HTGRs) has been studied for a prior engineering research with T handling and initial T possession in demonstration fusion reactors. Stable containment of T in Li-loading rods during HTGR operation is a critical issue. This study investigates this for an irradiation test to examine T-containment performance in Li-loading rods and develops an analytical model of evaluating the amount of T outflow to a He coolant. The hydrogen absorption characteristics, including the deterioration of the hydrogen absorption speed after Zr has sufficiently absorbed the hydrogen, is experimentally measured assuming an HTGR setting. We present an analytical model of evaluating the T outflow from a Li rod and, on the basis of this model, estimate the total T outflow, assuming the presence of a gas-turbine high-temperature reactor of 300 MWe with a nominal capacity and a high-temperature engineering test reactor. It is demonstrated that, by loading a sufficient amount of Zr into the Li rod, the T outflow can be suppressed to less than a small percent of the total T produced during 360 days of reactor operation.

The estimation of tritium permeation rate through the plasma facing wall into coolant is required to discuss tritium balance in a D-T fusion plant, to design tritium recovery system and to perform safety assessments. In this work, tritium permeation rates in the blanket first wall and the divertor were estimated by numerical analysis for simplified multi-layer structures with considering the temperature distribution in recent JA DEMO condition. The permeation rate in the blanket first wall, which was a double layer consisting of tungsten and F82H, was estimated to be 0.69 g/day. The permeation rate in the divertor, which was a triple layer consisting of tungsten, copper and copper alloy or F82H, was estimated to be 0.013 g/day. When the permeation rate in tritium breeding region in the blanket can be reduced by three orders of magnitude due to a permeation barrier, total tritium permeation rate in the blanket and the divertor was estimated to be 0.71 g/day.

Understanding of Li evaporation properties is important because Li mass transfer decreases tritium breeding ratio and influences tritium behavior possibly. In QST, the pebbles of Li2TiO3 with excess Li were fabricated by the optimized sintering condition recently and it was confirmed that the pebbles have a good tritium release property. In this work, the weight reduction of the pebbles at elevated temperatures was investigated considering the releases of water vapor and CO2 from the pebbles. Furthermore the mechanism of Li mass loss from Li2TiO3 with excess Li pebbles was discussed by comparing the difference between the pebbles formed by the optimal fabrication method, Pebble210 and ones formed by the previous fabrication method, Pebble211. The release behaviors of water vapor from Pebble210 and Pebble211 were almost same except a small peak at 350 °C. A large amount of CO2 was released from Pebble211 at over 700 °C which is close to the melting point of Li2CO3. On the other hand, the release of CO2 from Pebble210 was not observed. The weight reduction excluding the contribution of water vapor for Pebble210 was smaller than that for Pebble211. The difference of weight reduction between Pebble210 and Pebble211 are caused by the presence of Li2CO3.

It is important to understand tritium desorption behavior from plasma-facing materials of a fusion reactor in order to discuss effective tritium recovery method from in-vessel components. However, basic behavior of hydrogen isotopes in W deposition layer is not understood completely. In this study, characterized tungsten deposition layer formed by hydrogen plasma sputtering was exposed to gaseous tritium at 300 °C or 500 °C and tritium desorption behavior by vacuum heating was investigated by the imaging plate technique. For comparison, bare tungsten substrates were exposed to gaseous tritium in the same condition. Initial tritium activity in the deposition layer was much higher than that in the bare substrate. Tritium desorption behavior from tungsten deposition layer was different by the temperature of the layer during tritium exposure process. By heating at 500 °C for 1 h, 97.5% of tritium was desorbed from the layer exposed to tritium at 300 °C. On the other hand, by heating at 500 °C for 2 h, only 44.6% of tritium was desorbed from the layer exposed to tritium at 500 °C. To recover most tritium from W deposition layer and W substrate, heating at above 700 °C is required.

Performance of tritium production for fusion reactors, using a high-temperature gas-cooled reactor (HTGR) is examined. From the viewpoints of tritium recovery and environmental safety, tritium outflow from Li rods should be suppressed to the same level as the liquid radioactive waste from the pressurized water reactors (PWRs) in Japan. Methods for suppressing tritium leakage from Li rods are studied. The tritium outflow is reevaluated accurately on the basis of non-equilibrium simulations and the influence of coolant temperature on tritium leakage is clarified. The approach using Zr in the Li rod to reduce the tritium pressure and the resulting suppression of tritium leakage are also investigated.The results of the non-equilibrium simulation show that the tritium outflow is approximately 40% lower than the outflow reported in a previous study. Although the electric power generation efficiency is reduced, lowering the coolant temperature to 600 K results in a reduction of the tritium outflow to ~1/...

The Joint Special Design Team for Fusion DEMO was organized in 2015 to enhance Japan’s DEMO design activity and coordinate relevant research and development (R&D) toward DEMO. This paper presents t...

H2 permeation behavior in a new molten salt of Flibe mixed with 0.5 wt% Ti powder is experimentally and analytically investigated. Ti powder included in Flibe can suppress H2 permeation for...

The deposited W layers formed on the W plate by hydrogen plasma sputtering were exposed to deuterium plasma in EAST together with bare W plate. In TDS measurement, the deuterium release was clearly observed from the deposited W layer in addition to the release of hydrogen which was incorporated during the sputtering-deposition processes. On the other hand, the release of hydrogen isotope was not detected from the bare W plate. This suggests that the formation of deposited W layers increases tritium inventory in the plasma confinement vessel. Although the thermocouple contacting to the backside of the W plate did not indicate a remarkable temperature rise, deuterium release peaks from the W layer were close to that from the W layer irradiated by 2keV D2+ at 573K. It was found by glow discharge optical emission spectrometry analysis that retained deuterium in the W layer has a peak at the depth of 50nm and gradually decreases toward the W substrate. From X-ray photoelectron spectroscopy analysis, it was evaluated that W oxide existed just at the surface and W atoms in the bulk of deposited W layer were not oxidized. These data suggest that hydrogen isotopes are not retained in W oxide but grain boundaries. Keywords: Deuterium retention, Deposited tungsten, EAST

In order to make proof of the recovery of hydrogen isotopes from a liquid lithium (Li) blanket, we experimented the recovery of deuterium (D) dissolved in Li by means of yttrium (Y) metal at 300°C....

Recovery of hydrogen dissolved in Li-Pb eutectic alloy by mean of a bubbling tower is experimentally investigated. Mass-transfer coefficients to predict tritium recovery rate are experimentally determined when Ar and Ar+H2 gas bubbles are injected into Li-Pb through an I-shaped nozzle under the conditions of temperature 573–773 K and H2 partial pressure of 1 Pa–0.1 MPa. The results are fitted by an analytical equation based on diffusion and solution in Li-Pb. So that, the rate-determining step is hydrogen diffusion through a boundary layer formed in Li-Pb-gas interface and absorption and desorption are found to be almost reversible.

With the aim of developing a method for the recovery of tritium from tritium-bearing hydrocarbons, it was shown experimentally that methane can be decomposed directly into hydrogen and carbon in RF plasmas via reactions initiated by electrons. Measurements performed with CH4 and CH3T in a helium RF plasma indicate that the degree of decomposition of CH3T is substantially smaller than that of CH4. This is considered to be caused by a very low concentration of CH3T. It was found that a majority of tritium dissociated from CH3T is retained in the plasma reactor. However, a certain amount of retained tritium could be removed by a discharge-cleaning of oxygen.

The approximate estimation of tritium permeation rate under the acceptable assumption from a safety point of view is surely useful to progress the design activities for a fusion DEMO reactor. Tritium permeation rates in the blanket and the divertor were estimated by the simplified evaluation model under the recent DEMO conditions in the water-cooled blanket with solid breeder as a first step. Plasma driven permeation rates in tungsten wall were calculated by applying Doyle & Brice model and gas driven permeation rates in F82H were calculated for hydrogen-tritium two-component system. In the representative recent DEMO condition, the following tritium permeation\ rates were obtained, 1.8 g/day in the blanket first wall, 2.3 g/day in the blanket tritium breeding region and 1.6 g/day in the divertor. Total tritium permeation rate into the cooling water was estimated to be 5.7 g/day.

A high temperature gas cooled reactor is proposed as an external tritium source for stably starting up of fusion reactors. A Li rod coated with zirconium and alumina have been proposed to confine produced tritium. Although the tritium leakage from the Li rod has been estimated by the simulation, it is necessary to evaluate the tritium confinement performance experimentally. In this study, the test assembly simulating the Li rod composed of Inner Zr tube - Outer Zr tube - Al2O3 tube was built up, and tritium confinement experiments were conducted. The test assembly successfully confined tritium, which was contained in the Ar gas mainly as HTO and slightly as HT, for 12 h and 87 h at a high temperature of 700 °C. It was suggested that an oxide layer formed on the Zr tube contributed to suppress tritium permeation in HTO.

Transient- and steady-state hydrogen permeation from Li-Pb forced convection flow in a permeable tube to outside Ar purge gas was investigated between 400–600 °C. The values of the steady-state permeation rate increased with the increase of the Li-Pb flow rate. It was found that the overall permeation rates were limited by diffusion in a Li-Pb boundary layer developed from flow inlet. The effect of the boundary layer was correlated in terms of the mass-transfer coefficient. The values of the mass-transfer coefficient at 600 °C were compared with those of 400 °C and 500 °C obtained beforehand. Judged from these data of mass-transfer coefficients, it can be predicted that the effect of boundary layer varies with the increase of Li-Pb flow rate at different temperature conditions.

It is important to control the concentration of hydrogen isotopes as impurities included in Li by Y hot trap for the International Fusion Materials Irradiation Facility (IFMIF). Impurities limit goals in IFMIF are determined as

Understanding of hydrogen accumulation behavior in tungsten deposits is important from viewpoints of tritium economy and tritium safety in fusion reactors. Some reports indicate that a large amount of hydrogen isotope is incorporated into tungsten deposits growing by hydrogen plasma sputtering. However, the mechanism of hydrogen incorporation is not clarified yet. In this work, tungsten deposits were formed at different circumstances in the sputtering device and the amount of incorporated hydrogen was measured. The implantation of hydrogen reflected from the sputtering target was mainly contributed to the hydrogen incorporation. The contribution of the implantation of reactive hydrogen from plasma was smaller than that of the reflected hydrogen in the experimental condition. A detectable amount of tungsten deposits was formed even in the shadow region from the sputtering target. This suggests that a certain amount of tungsten atoms, which lost its initial energy by collision with molecular hydrogen, diffuses in the gas phase and adheres in the shadow region.

Development of tritium transport model in natural soil is an important issue from a viewpoint of safety of fusion reactors. The spill of a large amount of tritiated water to the environment is a concern accident because huge tritiated water is handled in a fusion plant. In this work, a simple tritium transport model was proposed based on the tritium transport model in porous materials. The overall mass transfer coefficient representing isotope exchange reaction between tritiated water and structural water in soil particles was obtained by numerically analyzing the result of the percolation experiment of tritiated water into the soil packed bed. Saturated hydraulic conductivity in the natural soil packed bed was obtained to be 0.033 mm/s. By using this value, the overall mass transfer capacity coefficients representing the isotope exchange reaction between tritiated water percolating through the packed bed and overall structural water on soil particles was determined to be 6.0 × 10−4 1/s. This value is much smaller than the mass transfer capacity coefficient between tritiated water vapor and water on concrete material and metals.

Fluoride molten salt Flibe (2LiF + BeF2) is a promising candidate for the liquid blanket of a nuclear fusion reactor, because of its large advantages of tritium breeding ratio and heat-transfer fluid. Since its melting point is higher than other liquid candidates, another new fluoride molten salt Flinabe (LiF + NaF + BeF2) is recently focused on because of its lower melting point while holding proper breeding properties. In this experiment, hydrogen permeation behavior through the three molten salts of Flibe (2LiF + BeF2), Fnabe (NaF + BeF2) and Flinabe are investigated in order to clarify the effects of their compositions on hydrogen transfer properties. After making up any of the three molten salts and purifying it using HF, hydrogen permeability, diffusivity and solubility of the molten salts are determined experimentally by using a system composed of tertiary cylindrical tubes. Close agreement is obtained between experimental data and analytical solutions. H2 permeability, diffusivity and solubility are correlated as a function of temperature and are compared among the three molten salts.

Understanding of hydrogen isotope behaviors in plasma-facing wall is important from viewpoints of fuel control and tritium safety. Tungsten (W) is a candidate material of plasma-facing components. In this study, hydrogen diffusion coefficient and hydrogen solubility constant in the W deposition layer formed by hydrogen plasma sputtering were obtained by gas-driven permeation experiments. The hydrogen diffusion coefficient in the W deposition layer was smaller than that in bulk W, and the hydrogen solubility constant was larger than that in bulk W. Then, hydrogen permeation behavior through the W deposition layer growing on nickel plate by hydrogen plasma sputtering was observed in the temperature range from 77 to 145 °C. Initially, the hydrogen permeation flux steeply increased but then gradually decreased with increasing the thickness of the deposition layer. From the analysis of the permeation behavior by TMAP code, the recombination coefficient on the surface of the deposition layer was evaluated. The obtained recombination coefficient was smaller than that on bulk W.

Tritium breeding ceramic materials are placed at high temperatures for a long period in a fusion DEMO reactor. Therefore, the understanding of Li mass loss phenomena and its influence on tritium behavior are important. In this study, the pebbles of Li2TiO3 with excess Li were heated at 900 °C for 30 days in a 1000 Pa H2/Ar flow and tritium sorption and recovery experiments were carried out. Li mass loss by the heating was evaluated to be 0.7 wt%. The value of Li mass loss was almost same as that for 3 days heating at the same condition. Tritium sorption capacity for the heated pebbles at 600 °C and 900 °C were almost same as that for the pebbles as received. Tritium sorbed in the pebbles could not be recovered effectively by the 1000 Pa H2/Ar purge at room temperature and 300 °C but it could be recovered at 600 °C and 900 °C. The influence of the long-time heating on the behavior of tritium sorbed in the pebbles was not large.

Solid tritium breeding materials are used in a high temperature condition for a long time in a fusion reactor blanket. It is expected that a certain amount of Li will be evaporated and the pebble structure will be changed. Understanding of Li mass transfer behavior in the blanket is an important issue from viewpoints of establishment of tritium cycle and tritium safety. In this work the pebbles of Li2TiO3 with excess Li were heated at 900 °C in 1000 Pa H2/Ar flow for a long time (72, 240, 720, 1200 h). The amount of Li mass loss under the hydrogen atmosphere was 0.665 wt% which was less than that under the water vapor atmosphere observed in the previous study. No significant Li mass loss was observed after 240 h. By heating for 240 h, the grain diameter increased from 1.8 to 3.1 μm and the specific surface area decreased from 0.72 to 0.20 m2/g. After 240 h, no significant grain growth and no significant decrease in the specific surface area were observed. From these results, it seems that there is a relation between Li mass loss and grain growth.

The conceptual design of Japan’s fusion demonstration plant (JA DEMO) is now being developed. In this paper, an overall plant system concept related to tritium handling in the water-cooling system is developed to give a concrete shape to the present JA DEMO concept as an electric power plant. The basic condition of tritium permeation from the in-vessel components to the primary cooling system is evaluated to be 5.7 g-T/day. The tritium concentration of the primary coolant is assumed to be 1 TBq/kg similar to the heavy water reactor condition. The capacity of the water detritiation system (WDS) is assessed, and the bypass feed water from the primary cooling loop is evaluated to be 94 kg/h under the tritium extraction efficiency of 0.96. Based on those specific parameters, the existing WDS in the heavy water reactor is found to be applicable to that of JA DEMO. Configuration of the primary heat transfer system (PHTS) is also discussed. Based on the heavy water reactor experience, tritium permeation through a steam generator (SG) to the secondary cooling system in PHTS is evaluated at 11.77TBq/year/loop (318 Ci/year/loop), which is found to be less than the restricted amount of tritium disposal for a pressurized water reactor in Japan. The key effect of the heavy water reactor experience is reduction of tritium permeation by oxide layer formed on SG pipes. Finally, a confinement concept of tritium release from PHTS is discussed under the condition of ex-vessel loss of coolant accident (LOCA). A pressure suppression system is installed to prevent the upper tokamak hall from pressurizing at the ex-vessel LOCA, and the tritium leakage from the upper tokamak hall is consequently restrained. The resultant early public dose at the plant site boundary can be reduced to 1.8 mSv, which is negligibly smaller than 100 mSv of the no-evacuation limit recommended by IAEA.