The physics of rewetting in water reactor emergency core cooling

Surface rewetting is essential for the re-establishment of normal and safe temperature levels following dryout in rod clusters or boiler tubes, or following postulated loss-of-coolant accidents in water reactors. Rewetting experiments have been performed with tubes and rods with a wide range of materials and experimental conditions (surface temperatures 300–800°C, constant water flows 0.1–30 g s⁻¹). The physical processes involved in the rewetting of high temperature surfaces have been shown to be identical for both falling water films and bottom flooding. The variation of rewetting velocity with mass flow has been determined, and shown to be independent of hydraulic diameter over the range 0.2–6 mm of practical interest. Data have also been obtained on the mass ‘carryover’ fraction. Theoretical solutions for the rewetting velocities have been obtained by analysis of thermal conduction in the surface. At low mass flows, effectively one-dimensional (axial) conduction cools the surface ahead of the rewetting front, and gives agreement with experiment. At higher mass flows the rewetting velocity is substantially independent of surface thickness and conductivity. The present data and the available world data for rewetting are shown to be in agreement with the theory.     

An experimental investigation on the quenching of a hot vertical heater by water injection at high flow rate

An experimental investigation on rewetting has been carried out by injecting water from the top of a hot vertical heater. Tests have been performed with varied range of experimental conditions (200-500 °C surface temperatures, constant water flow rates 5.77-30.98 g s⁻¹). Effect of several coolant injection systems on the hydrodynamics of rewetting has been studied. It is observed that for a particular range of flow rate and initial wall temperature (21.58 g s⁻¹, 300 °C) a circumferentially symmetric wet front is observed for the region closer to the coolant injection point even while using sub-cooled water. Rewetting velocity has been calculated from the temperature transients measured during the experiment and was found to vary within 1.0-20.0 cm s⁻¹. Two different rewetting models ( [Sahu et al., 2006] and [Sahu et al., 2008a]) have been used to compare the present experimental data and the comparison is found to be fairly good in both the cases. It has been observed that the flow rate varies linearly with effective Biot number (M) and varies inversely with magnitude of precursory cooling (N) in the present investigation.     

Neutronic and thermal hydraulic design of the graphite moderated helium-cooled high flux reactor

A new design concept for a high flux reactor was investigated, where a graphite moderated helium-cooled reactor with pebble fuel elements containing (²³⁵U, ²³⁸U)O₂ TRISO coated particles was taken as its design base. The reactor consists of an annular pebble bed core, a central heavy water region, and inner, outer, top, and bottom graphite reflectors. The maximum thermal neutron flux in its central heavy water region as high as 10¹⁵ cm⁻² s⁻¹ with thermal neutron spectral purity of more than two orders of magnitude and a large usable volume of more than 1,000 liters can be achieved by (1) diluting the fissile material in the core and (2) optimizing the core-reflector configuration. The in-core thermal-hydraulic analysis was done to obtain adequate information about the coolant flow pattern and pressure distribution, core temperature profile, as well as other safety aspects of the design. To protect the reactor during off-normal or accident events, a large margin of temperature difference between the operating fuel temperature and the fuel limit temperature is confirmed by lowering the coolant inlet and core rise temperatures.     

A review of spray-cooling and bottom-flooding work for lwr cores

Rewetting or the re-establishment of water films on the hot, fuel rod surfaces is of fundamental importance following a loss-of-coolant accident in a water cooled reactor. Two techniques are used: rewetting by a falling film of water (top sparay) and rewetting by an upflow of water (bottom flooding). Considerable theoretical and experimental work has been done to investigate the effects of different operating parameters on these techniques, and that work is summarized here.     

Response to G. Yadigaroglu''s letter

The rewetting or quench temperature is the temperature of a hot solid surface at which a liquid can reestablish contact with the dry surface. An estimation of this temperature is essential in predicting the rate at which the coolant quenches the core of a light-water reactor (LWR) after a loss-of-coolant accident. The present study reviews and evaluates previous work in this area and presents a model relating experiments to theory for the different possible types of reflood in LWRs. It is postulated that, with the exception of those cases of top reflood by a film in a single-rod geometry and bottom reflood with a very low mass flow rate, the quench temperature corresponds to either the minimum film boiling temperature or the Leidenfrost temperature. In cases where there are such exceptions, the quench temperature corresponds to the critical heat flux temperature. New correlations for the rewetting or quench temperature are presented.     

An analysis of the wet-side heat-transfer coefficient during rewetting of a hot dry patch

A physical model for the rewetting of a hot dry patch is proposed. The mechanisms governing the rate of resorption of the dry area are the conduction of heat from the dry to the wet area and the removal of this heat by a large local heat-transfer coefficient. The coefficient is assumed to be directly proportional to the cube of the difference between the surface and saturation temperature (h = RT^ß). The wall temperature at the interface between the wet and dry side is assumed equal to the sputtering value, and the cubic relationship between the heat-transfer coefficient and surface temperature difference is assumed valid up to this temperature. Experimental data from Harwell on rewetting rates for a stainless steel tube in a steam environment at various pressures are analysed to determine R. The energy equations are solved numerically by a quasi-linearization technique combined with the finite-difference method.     

Countercurrent flow limitation in inclined channels with bends

The countercurrent flow limitation phenomenon in an inclined round channel connected to bends at both ends is experimentally studied. The channel inner diameter is 1.9 cm, the bend radius divided by channel diameter is 10, and the channel length divided by channel diameter is varied in the 105–315 range. Countercurrent flow rates are measured with liquid and gas superficial velocities in the 0.015–0.21 m s⁻¹ and 0.1–3.1 m s⁻¹ ranges respectively. Air and water at room temperature and atmospheric pressure are used in the experiments. The liquid injection system is a constant-head plenum, and the channel angle of inclination with respect to the vertical line is varied in the 0°–60° range. The measured liquid and gas flow rates for all the angles of inclination are correlated using Wallis' flooding correlation, with a unique value for each of the two constants in the correlation.     

Mixed convection heat transfer to carbon dioxide flowing upward and downward in a vertical tube and an annular channel

This paper addresses three main subjects in supercritical heat transfer: (1) difference in thermal characteristics between upward and downward flows; (2) effect of simulating flow channel shape; (3) evaluation of the existing supercritical heat transfer correlations. To achieve the objectives, a series of experiments was carried out with CO₂ flowing upward and downward in a circular tube with an inner diameter of 4.57 mm and an annular channel created between a tube with an inner diameter of 10 mm and a heater rod with an outer diameter of 8 mm. The working fluid, CO₂, has been regarded as an appropriate modeling fluid for water, primarily because of their similarity in property variations against reduced temperatures. The mass flux ranged from 400 to 1200 kg/m² s. The heat flux was varied between 30 and 140 kW/m² so that the pseudo-critical point was located in the middle of the heated section at a given mass flux. The measurements were made at a pressure of 8.12 MPa, which corresponds to 110% of the critical pressure of CO₂. The difference between the upward and downward flows was observed clearly. The heat transfer deterioration was observed in the downward flow through an annular subchannel over the region beyond the critical point. Several well-known correlations were evaluated against the experimental data, and new correlations were suggested for both a tube and an annular channel.     

Design and experiment of hot gas duct for the HTR-10

A horizontal coaxial double-tube hot gas duct is a key component connecting the reactor pressure vessel and the steam generator pressure vessel for the 10 MW High Temperature Gas-cooled Reactor—Test Module. Hot helium gas from the core outlet flows into the steam generator through the liner tube, while helium gas after being cooled returns to the core through a passage formed between the inner tube and the duct pressure vessel. Thermal insulation material is packed into the space between the liner tube and the inner tube to resist heat transfer from the hot helium to the cold helium. The thermal compensation structure is designed in order to avoid large thermal stress because of different thermal expansions of the duct parts under various conditions. According to the design principal of the hot gas duct, the detailed structure design and strength evaluation for it has been done. A full-scale duct test section was then made according to the design parameters, and its thermal performance experiment was carried out in a helium test loop. With helium gas at pressure of about 3.0 MPa and a temperature over 900 °C, the continuous operation time for the duct test section lasted 98 h. At a helium gas temperature over 700 °C, the cumulative operation time for the duct test section reached 350 h. The duct test section also experienced 20 pressure cycles in the pressure range of 0.1–3.4 MPa, 18 temperature cycles in the temperature range of 100–950 °C. Thermal test results show an effective thermal conductivity of the hot gas duct thermal insulation is 0.47 W m⁻¹ °C⁻¹ under normal operation condition. In addition, a hot gas duct depressurization test was carried out; the test result showed that the pressure variation occurred on the liner tube was not more than 0.2 MPa for an assumed maximum gas release rate.     

Hydrodynamically-controlled rewetting

Transient cooling of a hot tube by a falling liquid film is analysed. Vapor production from the liquid film and the sputtered droplets can produce a countercurrent vapor velocity which exceeds the flooding limit, and rewetting becomes hydrodynamically-controlled rather than heat conduction-controlled. A criterion shows that conduction-controlled rewetting prevails at higher coolant flow rates and flooding conditions at lower flow rates. A solution is obtained for the liquid coolant vaporized during its fall from the sputtering film front. The required thermal radiation properties are also presented. Detailed calculation based on this analysis shows good agreement with experimental results.     

Effect of thermal radiation on rewetting of top spray emergency coolant

A general physical model for top spray rewetting during an emrgency core cooling (ECC) transient is proposed which takes into account thermal radiation in the dry region. The model is employed to study the effect of thermal radiation on rewetting a single rod and a 3 × 3 rod bundle up to 2100°F. The results show that rewetting in a bundle is slower than for an isolated rod, due to reduced thermal radiation heat transfer in the dry region. Also, there is a definite correlation between the decreased radiation heat flux Δq_R and the corresponding decrease in rewetting velocity Δu. Values of Δu are not significant unless Δq_R is larger than 6000 Btu/hr ft², where Δq_R cannot exceed a value of 6000 Btu/hr ft² below a temperature of 1100°F, even in the most adverse conditions. Hence, it is concluded that radiant heat transfer does not significantly affect rewetting velocities up to an initial rod temperature of 1100°F. Beyond this temperature, the rewetting velocities change by more than 1.5% and hence radiation must be included in the model for top spray rewetting.     

The quenching of irradiated fuel pins

This paper presents the results of falling film rewetting experiments on two types of irradiated fuel pin and on a complementary range of tubes and heaters. Preliminary measurements of rewetting rate in bottom flooding are presented and a brief comparison is made between falling film and bottom flooding rewetting, including the effects of surface finish. It is demonstrated that fuel pin rewetting behaviour is broadly consistent with that for tubes and heaters and that there are no large systematic effects of irradiation. Surface finish is shown to have an important influence on the falling film rewetting rate and could have influenced the fuel pin behaviour. The preliminary findings for bottom flooding are that there are smaller differences between various heaters than for falling film, and surface finish does not significantly change rewetting behaviour. It is shown that subcooling local to the quench front is an important parameter in rewetting, and can be used as a basis for correlating data. This is in agreement with recent ideas.     

Experimental comparisons for bottom reflooding in tight LWHCR and standard LWR geometries

The NEPTUN test facility is currently being used for reflooding experiments in tight hexagonal geometry, representative of light water high conversion reactors (LWHCRs). Results of parametric studies, based on over sixty forced-feed bottom reflooding experiments carried out with the NEPTUN-III (p / d = 1.13) test bundle, show that flooding rate is the most important, single thermal-hydraulics parameter.Direct comparisons with earlier NEPTUN experiments in standard LWR geometry indicate — on the basis of pressure difference considerations — that much smaller flooding rates may be expected to occur in tight LWHCR cores. The corresponding NEPTUN-III experiments show long-lasting rod surface temperature excursions with relatively high maximum temperatures being reached, and some of the more detailed experimental data collected is used to explain this behaviour. In spite of the above, rewetting of the tight-LWHCR geometry bundle was found to occur in all experiments with reasonably LWHCR-representative values for the various thermal-hydraulics parameters.     

Non-stationary outflow and rarefaction waves in flashing liquid

The outflow of high pressure liquid (in particular, water) to the atmosphere from a closed tube (of length a few metres and diameter more than a few centimetres) because of sudden destruction of one bottom is theoretically investigated. Evaporation takes places on the nucleus bubbles. The number of nuclei depends on the quality of the liquid or its purification. The process involves flashing evaporation of the liquid.There are two rarefaction waves at the initial stage. The velocity of the first wave (elastic forerunner) is sound speed in the one phase liquid and equals about 1000 m s⁻¹. After the elastic forerunner the liquid becomes superheated because the pressure drops and evaporation begins.The velocity of the second rarefaction wave is about 1–10 ms s⁻¹. There is intensive bubbly evaporation on and after the second wave. Intensity of the outflow is determined by the intensity of evaporation on the interface of the bubbles and by intensity of fragmentation of the bubbles because of their relative slip velocity in the liquid (0.1–1 m s⁻¹). The fragmentation of the bubbles significantly intensifies the evaporation because of augmentation of the bubbly interface.The degree of non-equilibrium or superheating behind the forerunner in water grows with the increasing initial temperature T₀. For T₀<530−540 K this superheating is negligible and the process may be described by an equilibrium scheme. For T₀ above 0.95T_cr≈605 K homogeneous nucleation is possible.After forerunner reflection from the closed bottom, intense evaporation is initiated near the bottom. Then the equalization of the pressure along the tube occurs (quasi-static homobaric stage).There is good correlation with experimental data.     

Critical heat flux of water in vertical round tubes at low pressure and low flow conditions

An experimental study on critical heat flux (CHF) has been performed for water flow in vertical round tubes under low pressure and low flow (LPLF) conditions to provide a systematic data base and to investigate parametric trends. Totally 513 experimental data have been obtained with Inconel-625 tube test sections in the following conditions: diameter of 6, 8, 10 and 12 mm; heated length of 0.31.77 m; pressure of 106951 kPa; mass flux of 20277 kg m⁻² s⁻¹; and inlet subcooling of 50654 kJ kg⁻¹, thermodynamic equilibrium critical quality of 0.3231.251 and CHF of 1081598 kW m⁻². Flow regime analysis based on Mishima & Ishii’s flow regime map indicates that most of the CHF occurred due to liquid film dryout in annular-mist and annular flow regimes. Parametric trends are examined from two different points of view: fixed inlet conditions and fixed exit conditions. The parametric trends are generally consistent with previous understandings except for the complex effects of system pressure and tube diameter. Finally, several prediction models are assessed with the measured data; the typical mechanistic liquid film dryout model and empirical correlations of (Shah, M.M., 1987. Heat Fluid Flow 8 (4), 326–335; Baek, W.P., Kim, H.G., Chang, S.H., 1997. KAIST critical heat flux correlation for water flow in vertical round tubes, NUTHOS-5, Paper No. AA5) show good predictions. The measured CHF data are listed in Appendix B for future reference.     

Reflooding experiments with LWR-type fuel rod simulators in the QUENCH facility

The QUENCH experiments at the Karlsruhe Research Center are carried out to investigate the hydrogen generated during reflooding of an uncovered Light Water Reactor (LWR) core. The QUENCH test bundle is made up of 21 fuel rod simulators approximately 2.5 m long. The Zircaloy-4 rod cladding is identical to that used in PWRs (Pressurized Water Reactors) with respect to material and dimensions. Pellets are made of zirconia to simulate UO₂. After a transient phase with a heating rate of 0.5–1 K s⁻¹ water of approx. 395 K is admitted from the bottom when the test bundle has reached its pre-determined temperature. Except for the flooding (quenching) phase, the QUENCH test phases are conducted in an argon/steam atmosphere at 3 g s⁻¹ each. The results of the first two experiments, QUENCH-01 (with pre-oxidation of 300 μm oxide layer thickness at the cladding outside surface) and QUENCH-02 (reference test without pre-oxidation), are compared in the paper. The pre-oxidized LWR fuel rod simulators of QUENCH-01 were quenched from a maximum temperature of 1870 K. In the second bundle experiment, QUENCH-02, quenching started at 2500 K. Pre-oxidation apparently prevented a temperature escalation in the QUENCH-01 test bundle, while the QUENCH-02 test bundle experienced a temperature excursion which started in the transient phase and lasted throughout the flooding phase. The different behavior of the two experiments is also reflected in hydrogen generation. While the bulk of H₂ was produced during pre-oxidation of test QUENCH-01 (30 g), the largest amount, i.e. 170 g, of hydrogen was generated during the reflooding phase of test QUENCH-02, at a maximum production rate of 2.5 g s⁻¹ as compared to 0.08 g s⁻¹ in test QUENCH-01. Similarities between the two experiments exist in the thermo-hydraulics during the quench phase, e.g. in the cooling behavior, the quench temperatures, and quench velocities.     

The effect of thermal diffusion from fuel pellets on rewetting of overheated water reactor pins

This paper presents the results of a finite difference solution of a conduction equation for the rewetting of a hot tube containing a filler material. The results show that the effect of a filler is always to reduce the rewetting velocity compared with an empty tube and reasonable agreement is obtained with previous experimental work. The effects of a gas gap on the rewetting of a UO₂-filled Zircaloy tube are discussed. A simple physical model is also presented which shows that the dominant parameter in determining the effect of a filler is (kpc)^1/2. It is suggested that previous theories for rewetting rate derived for empty tubes can be modified to include the effects of a filler by the use of a conduction correction term.     

Influence of a flow obstacle on the occurrence of burnout in boiling two-phase upward flow within a vertical annular channel

When a flow obstruction such as a cylindrical spacer is set in a boiling two-phase flow within an annular channel, the inner tube of which is used as a heater, the temperature on the surface of the heating tube is severely affected by its existence. In some cases, the cylindrical spacer has a cooling effect, and in the other cases it causes the dryout of the cooling water film on the heating surface resulting in the burnout of the heating tube.In the present paper, we have focused our attention on the influence of a flow obstacle on the occurrence of burnout of the heating tube in boiling two-phase flow.The results are summarized as follows:

(1)When the heat flux approaches the burnout condition, the wall temperature on the heating tube fluctuates with a large amplitude. And once the wall temperature exceeds the Leidenfrost temperature, the burnout occurs without exception.

(2)The trigger of dryout of the water film which causes the burnout is not the nucleate boiling but the evaporation of the base film between disturbance waves.

(3)The burnout never occurs at the downstream side of the spacer. This is because the dryout area downstream of the spacer is rewetted easily by the disturbance waves.

     

Vapour void fraction in subcooled flow boiling

Experimental data on steam void fraction and axial temperature distribution in an annular boiling channel for low mass-flux forced and natural circulation flow of water with inlet subcooling have been obtained. The ranges of variables covered are: mass flux = 1.4 × 10⁴−1.0 × 10⁵ kg/hr m²; heat flux = 4.5 × 10³−7.5 × 10⁴ kcal/hr m²; and inlet subcooling = 10–70°C. The present and literature data match well with the theoretical void predictions using a four-step method similar to that suggested by Zuber and co-workers.     

An analytical solution to a two-dimensional model of the rewetting of a hot dry rod

The model considers a hot dry rod of infinite length cooled by a film of liquid moving along its surface. The heat transfer coefficient is assumed to be constant on the wet side and zero on the dry side of the rewetting front, and the liquid film is assumed to move at constant speed. We derive an analytical formula relating the temperature difference in the rod, the temperature at the rewetting front, the wet side heat transfer coefficient, and the rewetting speed. The formula is thought to apply to the rewetting of a fuel rod during emergency cooling by flooding.