Minami Yoda

Multilayer nano-particle image velocimetry (MnPIV) in microscale Poiseuille flows

Multilayer nano-particle image velocimetry (MnPIV) uses the exponential intensity decay of evanescent-wave illumination with wall-normal distance to obtain near-wall velocity data at different distances from the wall z. Such data can then be used to determine velocity gradients within the first 400 nm next to the wall, and hence wall shear stress. In this paper, the technique is applied to measure near-wall velocities for the incompressible, steady and fully developed Poiseuille flow of four working fluids, namely water and sodium tetraborate solutions at three different molar concentrations (1, 10 and 20 mM), seeded with 100 nm fluorescent colloidal tracers through trapezoidal microchannels with nominal cross-sectional dimensions of 41 µm × 469 µm for pressure gradients up to 0.8 bar m−1. In all cases, the flow Reynolds number based on mean velocity and hydraulic diameter is O(10−1). The tracer images were divided into three layers based on their intensity and the mean velocity of each layer was obtained with a particle tracking method. The distribution of the tracers over z is highly non-uniform, with almost no particles within a particle diameter of the wall, and the distribution varies with the fluid. When corrected for this non-uniform tracer distribution, the MnPIV results are in good agreement with the classic analytical solution for two-dimensional Poiseuille flow based on the measured pressure gradients and channel dimension, and the velocity gradients obtained from the MnPIV results are on average within 5% of the analytical solution.

Gas-liquid two-phase flow in narrow horizontal annuli

Abstract Experimental data associated with the two-phase flow regimes, void fraction and pressure drop in horizontal, narrow, concentric annuli are presented. Two transparent test sections, one with inner and outer diameters of 6.6 and 8.6 mm, and an overall length of 46.0 cm; the other with 33.2 and 35.2 mm diameters and 43.0 cm length, respectively, were used. Near-atmospheric air and water constituted the gas and liquid phases, respectively. The gas and liquid superficial velocities were varied in the 0.02–57 and 0.1–6.1 m s −1 ranges, respectively. The major two-phase flow patterns observed included bubbly, slug/plug, churn, stratified, and annular. Transitional regimes, where the characteristics of two distinct flow regimes could be observed in the test sections, included bubbly-plug, stratified-slug and annular-slug. The obtained flow regime maps were different than flow regime maps typical of large horizontal channels and microchannels with circular cross-sections. They were also different from the flow regimes in rectangular thin channels. The measured average void fractions for the two test sections were compared with predictions of several empirical correlations. Overall, a correlation proposed by Butterworth [Butterworth, D., 1975. A comparison of some void fraction relationships for co-current gas–liquid flow. Int. J. Multiphase Flow 1, 845–850] based on the results of Lockhart and Martinelli (1949) provided the most accurate prediction of the measured void fractions. The measured pressure drops were compared with predictions of several empirical correlations. The correlation of Friedel [Friedel, L., 1979. Improved friction pressure drop correlations for horizontal and vertical two-phase pipe flow. 3R Int. 18, 485–492] was found to provide the best overall agreement with the data.

Measurements of the near-wall hindered diffusion of colloidal particles in the presence of an electric field

Understanding near-wall diffusion of small particles and biomolecules is important in colloid science and many microfluidic devices. Our experimental measurements of the diffusion of 110–460 nm radii suspended particles in the presence of electric fields up to 3.1 kV/m using particle tracking are in agreement with theoretical predictions for diffusion hindered by the presence of a solid surface. The results suggest that the external electric field has little, if any, effect upon the hindered diffusion of colloidal particles, even when the electrophoretic force exceeds the Stokes drag.

Experimental Evaluation of the Thermal Hydraulics of Helium-Cooled Divertors

Abstract Current predictions suggest that the target plate of a divertor, as one of the few solid surfaces directly exposed to the plasma of a magnetic fusion energy reactor, will be subject to steady-state heat fluxes as great as 10 MW/m2. Developing appropriate methods for cooling these divertors with helium is therefore a major technological challenge for plasma-facing components. This paper reviews dynamically similar experimental studies and numerical simulations of the thermal-hydraulic performance of two helium-cooled divertor concepts, the helium-cooled divertor with multiple-jet cooling (HEMJ) and the helium-cooled flat plate divertor, as well as a variant of the HEMJ, the so-called finger-type divertor, performed as part of the ARIES study. The results from these studies are extrapolated to prototypical conditions and used to predict the maximum average heat flux and coolant pumping power requirements for these divertor concepts. These extrapolations can be used to estimate how changes in the operating conditions, such as the helium inlet temperature and the maximum temperature of the divertor pressure boundary, affect thermal performance. Finally, the correlations from these extrapolations are used in the system code developed by the ARIES study.

Experimental Studies of the Thermal Performance of Gas-Cooled Plate-Type Divertors

Abstract The helium-cooled plate-type divertor can reduce the number of divertor modules while accommodating heat fluxes q” up to 10 MW/m2 incident on tungsten-alloy armor. Dynamically similar experimental studies were performed to evaluate the thermal performance of variants of this divertor design at conditions that spanned the prototypical operating Reynolds number Re of 3.3 × 104. In the studies, a jet of air issuing from 0.5 mm and 2 mm wide slots impinged on and cooled a heated planar surface 2 mm away from the slot, then flowed through either a 2 mm wide channel or an array of cylindrical pin fins. The studies indicate that the fins, which increase the cooled surface area by a factor of 3.76, increase the effective heat transfer coefficient (HTC) by as much as 160% at a relatively modest increase in pressure drop of less than 40%. These experimental results were used to determine the thermal performance of the actual plate design with helium cooling under prototypical conditions. Although the benefit of the fins is reduced because the fin efficiency decreases as the HTC increases, the predictions suggest that the fins could increase the maximum q” that can be accommodated by this design to ˜18 MW/m2. Alternatively, for a given heat flux (e.g. 10 MW/m2), adding fins could allow operation of the divertor at lower coolant flow rates, and hence pumping powers.

An Experimental Study of the Helium-Cooled Modular Divertor with Multiple Jets at Nearly Prototypical Conditions

Abstract A new helium (He) loop was used to study the helium-cooled modular divertor with multiple jets (HEMJ) at incident heat fluxes q″ ≤ 6.6 MW/m2 as part of the joint US-Japan effort on plasma-facing components evaluation by tritium plasma, heat, and neutron irradiation experiments (PHENIX). These studies were performed at prototypical pressures of 10 MPa and inlet temperatures ranging from 30 °C to 300 °C. The effect of varying the distance between the inner jets cartridge and the outer shell from 0.44 to 0.9 mm was also investigated. The Nusselt number results for two different tungsten-alloy test sections were in good agreement for q″ = 1.5–6.6 MW/m2. The experiments also suggest that the loss coefficient KL is essentially constant. These and KL results were used to estimate the maximum heat flux that can be accommodated by the divertor under prototypical conditions and the coolant pumping power as a fraction of the incident thermal power β. The agreement over the broad range of experimental parameters studied suggests that these results at near-prototypical conditions can be extrapolated with reasonable confidence to the operating conditions expected for the HEMJ design.

Dynamically Similar Studies of the Thermal Performance of Helium-Cooled Finger-Type Divertors With and Without Fins

Abstract Experimental studies based upon dynamic similarity have been used to evaluate the thermal performance of several modular helium-cooled tungsten divertor designs, including a configuration similar to the helium-cooled modular divertor with multiple jets (HEMJ). Until recently, all of these experiments used air, instead of helium, as the coolant. The average Nusselt number and loss coefficient were determined from cooled surface temperature and pressure drop data. Correlations were developed for the Nusselt number and loss coefficient as a function of the Reynolds number then used to predict the thermal performance of the divertor under prototypical conditions when cooled with high-temperature, high-pressure helium. Recently, experiments were performed using helium and argon to confirm the dynamic similarity assumption. The results indicated that the previous experiments with air, which were performed at the prototypical nondimensional coolant mass flow rate, or Reynolds number, did not account for the differences in the fraction of the incident power conducted through the walls of the divertor versus that convected, i.e., removed, by the coolant. Dimensional analysis and numerical simulations suggest that for a given divertor geometry this fraction can be characterized by the ratio of the thermal conductivities of the divertor material and the coolant. Nusselt number correlations were developed to include the effect of the thermal conductivity ratio. Based on these correlations, the predicted maximum heat flux values that can be accommodated by the HEMJ-like configuration are reduced by [approximately]20% from previous estimates. The results also suggest that the maximum heat flux that can be accommodated by this design can be increased by as much as 19% by adding an array of cylindrical pin fins on the cooled pressure boundary. However, as expected, adding the fins increases the pumping power for the coolant by [approximately]16%. As a fraction of maximum total incident thermal power, however, the pumping power decreases by 2% when the fins are added due to the significant increase in the maximum heat flux.

Verification of Thermal Performance Predictions of Prototypical Multi-Jet Impingement Helium-Cooled Divertor Module

Abstract An experimental investigation of the thermal performance of the Helium-Cooled Multi-Jet (HEMJ) modular divertor design developed by the Karlsruhe Research Center (FZK) was previously performed at Georgia Tech using air at Reynolds numbers (Re) spanning those at which the actual He-cooled divertor is to be operated. More recently, another experimental investigation was performed by the Georgia Tech group for a similar finger-type divertor module using both air and He as coolants. The results of these experiments suggest that, in addition to matching Re, dynamic similarity between the air and He experiments requires that a correction be made to account for the differences in the relative contributions of convection and conduction (through the divertor walls) to the overall heat removal rate by the module. This correction factor depends on the thermal conductivity ratio of the solid to the coolant. Experiments similar to those previously conducted have therefore been performed using air, argon, or He as coolant for test sections constructed of brass or steel thus covering a wide range of thermal conductivity ratio. The resultant correlation between Re, the heat removal rate, and the thermal conductivity ratio from these experiments can be used to predict the thermal performance of HEMJlike divertors at prototypical operating conditions.

Experimental Investigation of Fin Enhancement for Gas-Cooled Divertor Concepts

Abstract The addition of fins to the cooled surface of gas-cooled divertor modules has been proposed as a means to enhance their thermal performance, in the HEMP concept, for example. Such fins enhance heat transfer by significantly increasing the surface area over which convection occurs. However, adding fins also increases pressure losses and manufacturing costs and can adversely affect coolant flow over the cooled surface. More importantly, the high heat transfer coefficients expected with helium (He) cooling may significantly lower the fin efficiency, thereby limiting the extent of heat transfer enhancement to values well below the increase in the area ratio. An experimental investigation was undertaken to quantify the extent of heat transfer enhancement and corresponding pressure loss increase associated with the addition of pin fins to the cooled surface of a modular, helium-cooled, finger-type divertor. Four test cases, including configurations similar to the HEMP and HEMJ concepts, were studied. The results show that the addition of fins to helium jet-cooled finger divertors may not provide enough heat transfer enhancement to justify the associated increases in design complexity and pressure loss. Generalized charts for the thermal performance of helium-cooled divertors have been developed; these allow the designers to estimate the maximum allowable heat flux and corresponding pressure drop for a specified set of operating conditions and maximum operating temperature.

Extending Fluorescence Thermometry to Measuring Wall Surface Temperatures Using Evanescent-Wave Illumination

Cooling microelectronics with heat flux values of hundreds of kW/cm2 over hot spots with typical dimensions well below 1 mm will require new single- and two-phase thermal management technologies with micron-scale addressability. However, experimental studies of thermal transport through micro- and mini-channels report a wide range of Nusselt numbers even in laminar single-phase flows, presumably due in part to variations in channel geometry and surface roughness. These variations make constructing accurate numerical models for what would be otherwise straightforward computational simulations challenging. There is, therefore, a need for experimental techniques that can measure both bulk fluid and wall surface temperatures at micron-scale spatial resolution without disturbing the flow in both heat transfer and microfluidics applications. We report here the evaluation of a nonintrusive technique, fluorescence thermometry (FT), to determine wall surface and bulk fluid temperatures with a spatial resolution of O(10 μm) for water flowing through a heated channel. Fluorescence thermometry is typically used to estimate water temperature fields based on variations in the emission intensity of a fluorophore dissolved in the water. The accuracy of FT can be improved by taking the ratio of the emission signals from two different fluorophores (dual-tracer FT or DFT) to eliminate variations in the signal due to (spatial and temporal) variations in the excitation intensity. In this work, two temperature-sensitive fluorophores, fluorescein and sulforhodamine B, with emission intensities that increase and decrease, respectively, with increasing temperature, are used to further improve the accuracy of the temperature measurements. Water temperature profiles were measured in steady Poiseuille flow at Reynolds numbers of 3.3 and 8.3 through a 1 mm2 heated minichannel. Water temperatures in the bulk flow (i.e., away from the walls) were measured using DFT with an average uncertainty of 0.2 °C at a spatial resolution of 30 μm. Temperatures within the first 0.3 μm next to the wall were measured using evanescent-wave illumination of a single temperature-sensitive fluorophore with an average uncertainty of less than 0.2 °C at a spatial resolution of 10 μm. The results are compared with numerical predictions, which suggest that the water temperatures at an average distance of ∼70 nm from the wall are identical within experimental uncertainty to the wall surface temperature.