A design concept and characteristics for a high flux subcritical reactor (HFSR) are presented. By replacing the control rods with photoneutron generators, we could improve safety and perform radioactive waste burning in subcritical reactors that have primary system size, power density and cost comparable to a pressurized water reactor (PWR). The initial design is based on a small PWR, dry fuel cycle, intense resonance neutron source and power monitoring system with in-core gamma-ray detectors, now under development in the US, Korea, Russia and Ukraine. An important aspect of the HFSR is the reactor's modular design. To increase neutron source intensity the module is divided into two zones : a booster with fission electric cells (FEC) and a blanket with spent/depleted fuel. Neutrons are mostly generated in the booster surrounded the targets. A neutron gate (absorber and moderator) imposed between two subcritical zones permits fast neutrons from the booster flow to the blanket. Neutrons moving in the reverse direction are moderated and absorbed. However, these in-core applications require somewhat smaller high-brightness neutron generators that are presently available. An inexpensive method of obtaining large neutron fluxes is target-distributed electron accelerators (TDA), in which a FEC electrical field compensates for lost beam energy in the thin photon production targets. The FEC is essentially a high-voltage power source that directly converts the kinetic energy of the fission fragments into electrical potential of about 2 MV. The charge deposited by the electron beam in the target could be used to suppress the flow of secondary electrons across the gap between the FEC electrodes. Although the low and medium energy accelerator accelerators are mainly seen as support equipment for particle physicists, recent advances in space technology have made it possible to use these accelerators as neutron sources for energy generation systems. The advanced design features, based essentially on the studies in dispersive wave and space physics (wave model of observed relativistic phenomena, ultrasonic concept the plant seismic resistance increase, microwave x-ray source/separator), were recently presented at the 5th Sakharov Conference and 7th Wigner Symposium. Significant savings in cost can be realized by exploring this study in areas as diverse as radio astronomy navigation, space propulsion, monochromatic computed tomography and X-ray lithography.

View full abstract

Small sized fast reactors have a potential to use a power source applicable to diversified social needs. In the feasibility studies on commercialized fast breeder reactor cycle system of Japan nuclear cycle development institute (JNC), we will make a concept of small sized fast reactor with various requirements of economical competitiveness, reactor safety, very long lived core, etc. The design requirements for a small lead-bismuth cooled fast reactor which is one of promising concepts are defined as follows; -electric power : 50 MWe, -fuel type : mixed nitride fuel, -reactor type : tank type (elimination of secondary cooling system), -coolant circulation system : natural circulation, -fuel exchanging system : handling of each fuel subassembly, and -core life : 30 years. In this design, the reactor type was tank type and the steam generator is placed inside the reactor vessel, because there is no high reactivity between lead-bismuth and steam. The heat mass balance and the steam pressure for the turbine are chosen to minimize the total mass of the reactor vessel and the steam generator. The reactor coolant outlet and inlet temperature are 505 and 335 degree-C and the reactor vessel height becomes rather high (18.5m) to keep natural circulation force. The steam temperature (400 degree-C) and pressure (6 MPa) are designed to make the steam generator compact and the reactor vessel diameter is achieved to be 3.85m. Passive safety is enhanced to avoid a core disruptive accident even at anticipated transient without scram. The constitution of a decay heat removal system is chosen which is a system of one PRACS and one reactor vessel air cooling system (RVACS) driven by natural circulation. Protection from corrosion of metallic alloys in liquid lead-bismuth is ensured with using 9Cr steel and control of oxide composition in coolant. But further elemental studies are needed. The construction cost of the plant is estimated at 800,000 yen/k We which is much higher than the target value, that is, 350,000 yen/kWe (5.5 yen/kWh), because of a scale demerit coming from a low power and a usage of heavy reactor components for lead-bismuth coolant. But it would have some possibility to reduce the construction cost with a improvement of core efficiency, an increasing of a reactor power, a common use of some BOP components among reactors and a learning effect of construction.

View full abstract

Corrosion tests were carried out on austenitic AISI 316L and 1.4970 steel and on martensitic MANET 10Cr steel in flowing (up to 1.3 m/s) Pb/Bi. The concentration of oxygen in the liquid alloy was controlled at (10)^<-6> wt%. Specimens consisted of tube and rod sections in original state and after alloying of Al into the surface. Martensitic steels develop thick protective magnetite and spinel layers that spall off and grow again during long term exposure (7200 h). At 600℃ the oxide scale changes to thin protective spinel layers with partial dissolution attack at some spots. Austenitic steels exhibit thin spinel layers at 420℃ and thick spinel and magnetite layers at 550℃ which are protective. The spinel layer is also replaced after it brakes off. Severe attack of the liquid alloy occurs on austenite at 600℃ already in the 2000 h exposure period. Steels with 8 - 15 wt% Al alloyed into the surface suffer no corrosion attack at all experimental temperatures and exposure times.

View full abstract

The alkaline extraction method is being investigated as a method to remove polonium from a lead-bismuth cooled reactor. Tellurium was used as a surrogate for these preliminary experiments to determine if migration from the lead-bismuth to the molten alkaline (NaOH) could be measured. Experiments provided direct evidence that molten NaOH can effectively remove tellurium from lead or lead-bismuth, as a reduction of three to four orders of magnitude in the tellurium concentration in the metals was observed during the experiments. The experiments also showed a higher than expected concentration of NaOH in the lead or lead-bismuth. Graphite, alumina, nickel and nickel-chromium were found unsatisfactory as crucible materials, while zirconium was found to be compatible at the conditions of interest.

View full abstract

After the first phase of the 10MW high temperature gas-cooled test module reactor (HTR-10) with steam-turbine cycle, the preliminary design of the power conversion unit (PCU) with direct gas-turbine cycle for the next phase of HTR-10 project was finished by the Institute of Nuclear Energy Technology (INET) of Tsinghua University and OKBM (Russia). The main characteristics of the preliminary design, including the PCU structure, the thermal circuit and equipment parameters are described detailed in this paper. Implementation of research program for gas-turbine cycle on the HTR-10 reactor will make it possible to gain necessary experience in gas-turbine cycle design for high-temperature gas-cooled reactor and creation of equipment of large commercial nuclear power plant in the future.

View full abstract

The High Temperature Gas-cooled Reactor (HTGR) is particularly attractive due to its capability of producing high temperature helium gas as well as its inherent safety characteristics. Hence, perspective of HTGR as possible future nuclear energy source was discussed in the review of "Long-term Program for Research, Development and Utilization of Nuclear Energy" by the Atomic Energy Commission of Japan, and the High Temperature Engineering Test Reactor (HTTR), which is the first HTGR in Japan, was successfully constructed at the Oarai research establishment of the Japan Atomic Energy Research Institute. The HTTR attained the first criticality on November 10,1998 and achieved the full power of 30MW and the reactor outler coolant temperature of about 850℃ on December 7,2001. The purpose of the HTTR project is to establish and upgrade HTGR technologies. It is widely recognized to the nuclear community that the timely and successful operation and tests of the HTTR are major milestones in development of the HTGR and high temperature nuclear process heat application. A process heat application system will be coupled to the HTTR, where hydrogen will be produced directly from the nuclear energy.

View full abstract

This paper presens the numerical simulations of droplet breakup under an impulsive acceleration by the Moving Particle Semi-implicit method and the effect of the density ratio on a critical Weber number. The simulation results of droplet breakup are consistent with experimental observations. It is shown that the critical Weber number is inversely proportional to density ration. Furthermore, the Weber number on the onset of breakup is investigated, which is approximately 0.4 for the above density ratios.

View full abstract

An experimental study was carried out to investigate the steam-water two-phase flow of coolant in the Neutron Reflector (N/R) during the reflooding of large break Loss OF Coolant Accident (LOCA). The experimental apparatus simulated the geometry of one hole of a typical N/R cooling-hole (including the corresponding thermal capacity), the lower and the upper plenum. The steam and water were used as the practical working fluids. The water flow rate at the inlet was kept constant at a fixed level through the injection pump. The boundary conditions including the LOCA phenomena, which were estimated by an analytical code, were supplied through the measurement of the system pressure and the entrance temperature of the water. The system pressure, temperature, and flow rate of steam and water were measured simultaneously to investigate the thermal-hydraulic behaviors at the end of the cooling-hole. Further, the behavior of liquid lumps and droplets flow during the simulated reflooding was observed at the outlet of the test section. The closed inspection revealed that the flow condition out of the test section were changing during the simulated reflooding and was classified into three stages. At the initial stage, the liquid lumps and the droplets flowed continuously like a fountain. At the medium stage, misty flow appeared, i. e. dispersed small droplets and/or liquid lumps flowed in steam flow. At the final stage, the liquid lumps flowed periodically with a little amount of steam flow. In this study, the effects of the reflooding velocity, the initial metal temperature, the reflooding water temperature and the upper plenum pressure were clarified. For the lower reflooding velocity, the ratio of steam flow rate increased. For the higher initial metal temperature, the integration of steam and water flow rate increased. The reflooding water temperature and the pressure of the upper plenum at the experimental conditions had little effects on the thermal hydraulic behavior of the coolant at the outlet of the test section.

View full abstract

Velocity measurements without any device have indicated a bucket shaped flow pattern confirming the need to use a flow distribution device to attain the required distribution. It is observed that the presence of a FDD has improved the bucket shaped profile, by reducing the velocities in the vicinity of the downcomer. The fine tuned FDD has further improved the velocity profile.

View full abstract

The numerical modules for simulating the motion of the Lagrangian particles and their interactions with the surrounding Eulerian fluids are currently being developed at Korea Atomic Energy Research Institute (KAERI). The basic concept employed in the modules is similar to that of TEXAS-V, which is the Lagrangian-Eulerian one-dimensional code. Unlike TEXAS-V, the motion of the particles can be simulated in one, two or three dimensions. The modules are not the stand-alone computer codes. Rather, they are designed to be incorporated into the existing multi-phase multi-dimensional computer codes. This will introduce the ability to model the discrete medium fields into the existing computer codes with the minimal effort. Currently, the models for simulating the movements of the particle groups, the heat transfer process and the fragmentation of the particles due to the interaction with the surrounding fluid have been completed. The next stage of the development is to implement the modules to the multiphase fluid code K-FIX. The finished code will be tested, compared and verified with the experiment data and/or the results from other computer codes.

View full abstract

For the development of nuclear reactors and the assessment of their safety features, the development of computer code with the high quantity database from the measurement as well as the understanding of the multiphase flow physics are necessary. In this study, the characteristics of bubbly flow in a vertical tube are investigated using Wire Mesh Tomography (WMT). Local void fraction is detected from the dependency of electrical conductivity on the local void fraction. The developed Wire Mesh Sensor is a circular type with two parallel measuring planes to have the capability of gas velocity and bubble size evaluation. The experiment is conducted in a 50mm ID tube at the fully developed condition (93D). The mean bubble size is treated as a constant parameter independent from the superficial gas and liquid velocity by using the bubble generator with a water sub flow. The capability of WMT for studying local characteristic of bubbly flow is presented in Fig 1 and verified by comparing with the literature in Fig 2. In addition, the initial condition effect is still observed at long pipe length. The effects of superficial gas and liquid velocity and the additional bubble intensity on the void fraction distribution are presented. These effects are supposed to change the lateral lift force in both magnitudes and directions which induce the bubble migrated toward to or depart from the wall.

View full abstract

Feasibility study on the future light water reactor has been carried out. Small- and medium-sized natural circulation boiling water reactor (BWR) has been one of the concepts. One of the indispensable topics is thermohydraulic instability during startup. Occurrence of the instability during the startup negatively affects safety, reliability and operability, as it would cause complexity in raising and maneuvering reactor power. The purpose of the study is to experimentally investigate driving mechanism of major instabilities simulated in a natural circulation experimental loop, under a predetermined range of system operating pressure and inlet subcoolings. Pressure range of 0.1 up to 0.7 MPa, input heat flux range of 0 up to 577kW/m^2,and inlet subcoolings of 5,10 and 15K respectively, were applied in the experiments. The objective of this paper is to offer new experimental data as the basis to propose rational startup procedure, in which various thermohydraulic instabilities can be prevented. The study clarifies that four (4) kinds of thermohydraulic instability might occur even up to a higher pressure of 0.7 MPa. The clarified sequence instabilities are as follows : (1) geysering induced by condensation accompanied by flashing, (2) oscillation induced by hydrostatic head fluctuation, (3) density wave oscillations, and (4) flashing accompanying those instabilities. The experiments confirmed that the geysering region gets narrower and suppressed with the increased system pressure. With chimneys, natural circulation can be achieved reliably and more easily. However, the flashing in the chimney cannot be avoided at low system pressure. Stable two-phase natural circulation can be established if the system pressure is increased beyond 0.7 MPa, after the high frequency density eave oscillation thoroughly suppressed. The experiments were analyzed based on frequency domain of each instability phenomenon.

View full abstract

To provide a dynamic and mechanistic constitutive model for the interfacial area concentration in the two-fluid model, a two-group interfacial area transport equation is developed for confined two-phase flows in this series of papers. This paper presents the modeling of bubble interaction mechanisms in the two-group interfacial area transport equation for confined gas-liquid two-phase flow. The transport equation is applicable to bubbly, cap-turbulent, and churn-turbulent flow regimes. In the two-group interfacial area transport equation, bubbles are categorized into two groups : spherical/distorted bubbles as Group 1 and cap/slug/churn-turbulent bubbles as Group 2. Thus, two sets of equations are used to describe the generation and destruction rates bubble number density, void fraction, and interfacial area concentration for the two groups of bubbles due to bubble expansion and compression, coalescence and disintegration, and phase change. Five major bubble interaction mechanisms are identified for the gas-liquid two-phase flow of interest, such as, bubble coalescence as a result of random collision driven by turbulent eddies, the entrainment of the bubbles following in the wake of preceding bubbles, bubble breakup due to turbulent impact and surface instability for large bubbles, and small bubble shearing-off at the rim of large bubbles. These interaction mechanisms are analytically modeled as the source/sink terms for the transport equations based on certain assumptions for the confined flow. The models include both intra-group (within a certain group) and inter-group (between two groups) bubble interactions. A complete set of models close the two-group interfacial area transport equation, which has embedded the mechanisms of the flow regime transition. The comparisons of the prediction by the one-dimensional two-group interfacial area transport equation with experimental data are presented in the second paper of this series.

View full abstract

The bubble interaction mechanisms have been analytically modeled in the first paper of this series to provide mechanistic constitutive relations for the two-group interfacial area transport equation, which was proposed to dynamically solve the interfacial area concentration in the two-fluid model. This paper presents the evaluation approach and results of the two-group interfacial area transport equation based on available experimental data obtained in the confined flows at Purdue University, namely, 11 data sets in or near bubbly flow and 13 sets in cap-turbulent and churn-turbulent flows. The two-group interfacial area transport equation is evaluated in steady state, one-dimensional form. Also, since the experiments were performed under adiabatic, air-water two-phase flow conditions, the phase change effect is omitted in the evaluation. To account for the inter-group bubble transport, the void fraction transport equation for Group-2 bubbles is also used to predict the void fraction for Group-2 bubbles. Agreement between the data and the model predictions is promisingly good and the average relative difference for the total interfacial area concentration between the 24 data sets and predictions in within 7%. The model evaluation demonstrates the capability of the two-group interfacial area transport equation focused on the current confined flow to predict the interfacial area concentration over a wide range of flow regimes.

View full abstract

The critical heat fluxes (CHFs) of subcooled water flow boiling for the test tube inner diameters (d-3 and 6mm) and the heated lengths (L=67,120 and 150mm) are systematically measured for the flow velocities (u=4.0 to 13.3m/c), the inlet subcoolings (ΔT_<sub, in>=48 to 148 K), the outlet subcoolings (ΔT_<sub, out>=10.5 to 95.1 K), the inlet pressure (P_<in>=753 to 995 kPa) and the outlet pressure (P_<out>=720 to 887 kPa). The SUS304 tubes of L=67,120 and 150mm for d=3mm and L=150mm for d=6mm are used. The values of L/d are 22,40 and 50 for d=3mm, and 25 for d=6mm, respectively. The CHFs, q_<cr, sub>, for a fixed ΔT_<sub, out> become gradually lower with an increase in the L/d in the whole experimental range. The CHF correlation against outlet subcooling, which has been previously derived for L/d lower than 16,was modified to new one containing the L/d effect based on these experimental data. Furthermore, the relation between q_<cr, sub> and L/d for a fixed ΔT_<sub, in> was checked. The values of q_<cr, sub> for a fixed ΔT_<sub, in> became exponentially lower with the increase in L/d. CHF correlation against inlet subcooling has been given based on the experimental data for L/d ranging from 4.08 to 50. The correlations against outlet and inlet subcoolings can describe not only the CHFs obtained in this work for the inner diameter of 3 and 6mm at the outlet pressure of around 800 kPa but also the authors' published CHFs data (1611 points) for the wide ranges of P_<in>=159 kPa to 1 MPa, d=3 to 12mm, L=33 to 150mm and u=4.0 to 13.3 m/s within 15% difference for 30 K&le;ΔT_<sub, out>&le;140K and 40 K&le;ΔT_<sub, in>&le;151 K.

View full abstract

The subcooled flow boiling critical heat fluxes (CHFs) and the heat transfer coefficients (HTCs) data for L=49,99 and 149mm with 9-mm inner diameter were applied to thermal analysis on the Mono-block type divertor of LHD. Incident CHFs for the divertor with the cooling tube diameter, d, of 10mm and the carbon armor outer diameter, D, of 26 and 33mm were numerically analyzed based on the measured CHFs and HTCs at the inlet pressure of around 800 kPa. The numerical solutions were also compared with those for the Flat-plate type divertor, which were numerically analyzed for the divertor with the cooling tube diameter d=10mm and the divertor width, w, ranging from 16 to 30mm. It is confirmed that the ratio of the one-side heating CHF data, q_<cr, inc>, to the uniform heating CHF data, q_<cr, sub>, can be represented as the simple equation based on the numerical solutions. The values of the q_<cr, inc> for L=50,100 and 150mm were estimated with various D/d and w/d at higher pressures.

View full abstract

Series of slow transient and rapid blowdown calculations were performed to evaluate potential code problems for supercritical applications. The slow transients investigated transitions between sub-critical and supercritical pressures and temperatures. The rapid blowdown calculations simulated the geometry of the Edwards pipe experiment, but with supercritical initial conditions. Both series of calculations investigated a wide range of supercritical thermodynamic states. Overall, 20 of the 27 of the slow transient cases failed with the original code, and the probability of failure increased dramatically near the critical point. Eight of the 85 rapid blowdown calculations encountered water property failures and did not run to completion. The failures generally occurred near the critical point. Various code updates were implemented to correct the code execution failures. Changes were made to the reset water property derivatives at the critical point, the number of pressure and temperature points used to generate the steam tables near the critical point, the interpolations used for specific volume and isothermal compressibility, extrapolations for metastable states near the critical point, the interfacial heat transfer coefficient for vapor near the critical point, and to the transport properties. The modified code was able to successfully execute all of the slow transients and rapid blowdowns. In addition, the modified code was able to successfully execute a loss-of-coolant accident in a light water reactor pressurized to supercritical conditions.

View full abstract

Framatome-ANP developed S-RELAP5,a RELAP5 based thermal hydraulic code and PANBOX, which is a 3D code for calculation of neutronic and thermal hydraulic core kinetics. For core thermal hydraulics the COBRA 3-CP code is introduced as integral part into PANBOX. The coupling of all three codes has provided the powerful code system R/P/C. The capability of the code system has been tested by application to several suitable transients. The measured values of the data recording system of the plant were kindly put at our disposal. For the purpose of this presentation a particularly interesting transient with "loss of load" combined with a temporary coast down of one main coolant pump is discussed. The cause of this transient was the accidental disconnection of a 400 kV power switch at an off site transformation station. As a consequence the change over of the turbine control from power to speed control was managed by resetting the power nominal value to station service power by dropping 5 (uneven) control rod pairs, which resulted in an azimuthal asymmetric neutron distribution over the core. The turbine valves were spontaneously closed in order to limit the overspeed at 110%. Due to the simultaneous voltage break down (83%) in one of the station power net bars and the consequent long duration switch towards the reserve net the main consumers on this bar suffered coast down (i.e. one (of four) main coolant pumps and one (of two) main feedwater pumps). After change to the reserve net the consumers were reactivated again according to the design procedure. The reconnection of the main coolant pump caused a cold water plug to enter the core and a distortion of the massflow therein resulting in a significant power peak. The key phenomena of the given transient, which have to be reproduced by the calculation are detection of "Loss of Load" by comparison of the reactor and generator power and the synchronous and asymmetric drop of 5 control rod pairs with the result of temperature decrease and following reactivity increase (check of the I&C model). coast down and reactivation of the main coolant pump in one coolant circuit and the formation of an accentuated low temperature profile up to the core region. The physical progress of the transient is dependent of a double asymmetry : the pattern of an a odd number (5) of fallen control rod pairs in the core. the temperature distribution, which causes interaction with the asymmetrically distributed reactivity sinks. Therefore a 3-D calculation is indispensable for a stringent simulation. The presented calculation was carried out as a part of the validation procedure for the R/P/C code system. Without any essential corrections of input values or code adjustments the represented results could be achieved. Even in details, the calculation and measurement results meet surprisingly well.

View full abstract

The vibration of nuclear steam piping is usually associated with pressure fluctuations emanating from flow disturbances such as steam generator nozzles, bends, or other pipe fittings. Flow separation at pipe tees and within steam chest manifolds or headers generate pressure fluctuations that propagate both upstream to steam generators as well as downstream to the steam turbine. Steady-state acoustic oscillations at various frequencies occur within the piping, possibly exciting structural vibrations. This paper focuses on the assessment of the origin of the disturbances using signal analyses of two dynamic pressure recordings from pressure transducers located along straight runs in the steam piping. The technique involves performing the cross spectrum to two dynamic pressure signals in piping between (1) the steam generator and steam chest header, and (2) between the header and steam turbine outlet. If, at a specified frequency, no causality occurs between the two signals then the cross spectra magnitude will be negligible. Of interest here is the value of the phase between the two signals for frequencies for which the magnitude of the cross spectrum is not negligible. It is shown in the paper that the direction of the dominant waves at all frequencies can be related to the phase angle from the cross spectrum. Cross-spectral analyses has been employed to determine the direction of the dominant acoustic waves in the piping for various frequencies for which there are signals. To prove the technique, synthetic spectra are generated comprised of harmonic waves moving both upstream and downstream between two locations. The method shows that the direction of the larger magnitude wave -- whether propagating upstream or downstream -- is directly related to the slope of the unwrapped phase angle versus frequency correlation. Indeed, the slope of this line can be related to the acoustic velocity of the wave. The method is then applied to dynamic pressure recordings obtained in a nuclear steam system. Plots of cross-spectra phase versus frequency taken in straight runs of steam piping yield correlations that are nearly linear, and, moreover, the slope of the line is closely related to the acoustic velocity at the corresponding steam pressure.

View full abstract

This paper presents the results of TEXAS-V computer code simulations of FARO L-14,L-28,and L-33. The old break up model and new break up model are exercised for the comparative simulations. As these experimental data set cover a wide range of ambient pressures, sub-cooling of the water pool, and the melt jet diameters, the results of the simulation will be beneficial in assessing the TEXAS-V code capability in predicting the steam explosion phenomena in prototypic reactor case. As it turned out that current model had some deficiencies, the modules for the fragmentation, equation of state, and interfacial area for cach flow regime in TEXAS--V are improved for the simulation of FARO L28 and FARO L-33.

View full abstract