Study of bubble growth in water pool boiling through synchronized, infrared thermometry and high-speed video

https://doi.org/10.1016/j.ijheatmasstransfer.2010.05.041Get rights and content

Abstract

High-speed video and infrared thermometry were used to obtain time- and space-resolved information on bubble nucleation and heat transfer in pool boiling of water. The bubble departure diameter and frequency, growth and wait times, and nucleation site density were directly measured for a thin, electrically-heated, indium–tin-oxide surface, laid on a sapphire substrate. These data are very valuable for validation of two-phase flow and heat transfer models, including computational fluid dynamics with interface tracking methods. Here, detailed experimental bubble-growth data from individual nucleation sites were used to evaluate simple, commonly-used, but poorly-validated, bubble-growth and nucleate-boiling heat-transfer models. The agreement between the data and the models was found to be reasonably good. Also, the heat flux partitioning model, to which our data on nucleation site density, bubble departure diameter and frequency were directly fed, suggests that transient conduction following bubble departure is the dominant contribution to nucleate-boiling heat transfer.

Introduction

Nucleate boiling is an effective mode of heat transfer; and one of the most studied physical phenomena in science and engineering. At low heat flux, where isolated bubble growth occurs, the growth cycle can be qualitatively described as follows [1]. Once the liquid layer above the heater surface reaches the required superheat, ΔTs, to activate a given nucleation site, a bubble begins to form and pushes the surrounding liquid outward, except for a thin liquid microlayer that remains in contact with the wall underneath the bubble. Evaporation occurs at the bubble surface and through the microlayer, thus fueling further bubble growth. When the size of the bubble is sufficiently large, buoyancy causes the bubble to detach from the surface; new fresh liquid floods the surface, and the cycle restarts.

For decades, nucleate-boiling heat transfer has been predominantly an empirical science, accompanied by relatively simple models based on hypotheses that are not always fully justified. For example, the widely popular Rohsenow’s correlation for nucleate boiling is based on the assumption that single-phase convection and nucleate boiling are analogous physical processes, and can be both correlated in terms of the Reynolds and Prandtl number of the liquid phase; for nucleate boiling the characteristic velocity and length are assumed to be the downward liquid velocity and the most unstable Taylor wavelength, respectively; then, an empirical constant, Csf, is determined to fit the experimental data for any fluid/surface combination [2].

As researchers are now finally moving away from the rough empiricism of the past, and start to develop more mechanistic models of nucleate-boiling heat transfer, the need for high-quality high-resolution data on the bubble nucleation and growth cycle is increasing. Specifically, nucleation site density, bubble departure diameter and frequency data are a necessary input for the source terms in interfacial area transport models [3] and CFD ‘multi-fluid’ models [4], [5], [6], as well as semi-empirical models for boiling heat transfer, such as the heat flux partitioning model of Kurul and Podowski [7], Kolev’s bubble interaction model [8] or the more recent hybrid numerical-empirical model of Sanna et al. [9]. Furthermore, time-resolved temperature distribution data for the boiling surface and direct visualization of the bubble cycle are needed for validation of ‘first principle’ models of bubble nucleation and growth, based on interface tracking methods, in which the geometry of the vapor/liquid interface is not assumed, but rather calculated from a marker function advected according to the Navier–Stokes equations [10], [11], [12].

However, gathering the detailed data needed for validation of advanced simulation models is not straightforward. The traditional approaches based on thermocouples and high-speed visualization of the boiling process suffer from several shortcomings; for example, the thermocouples can only measure temperature at discrete locations on the boiling surface, thus no information on the temperature distribution about a nucleation site can be obtained. Further, thermocouples (including micro-thermocouples) have relatively long response time, thus are unsuitable for studying the bubble nucleation and growth phenomena, which have time scales of the order of milliseconds. The usefulness of high-speed video is typically limited by poor optical access to the nucleation site and interference from adjacent bubbles. Second-generation two-phase flow diagnostics, such as multi-sensor conductivity and optical probes [13], [14] and wire-mesh probes [15], can measure bubble diameter and velocity near the boiling surface. However, these approaches are intrusive, and also produce data only at discrete locations within the boiling fluid. It was not until the early 2000s that new possibilities for generating time-resolved multi-dimensional data on the bubble nucleation and growth cycle have opened up with the introduction of infrared-based visualization of thermal patterns on the boiling surface by Theofanous et al. [16]. Another recent high-resolution approach is based on arrays of micro-heaters, individually controlled to achieve a constant temperature boundary condition, as described by Demiray and Kim [17].

In this paper we present an approach based on synchronized infrared thermometry and high-speed video through a transparent heater that enables simultaneous measurement of the nucleation site density, bubble-growth rate (including bubble departure diameter), bubble departure frequency (including wait time), and time-resolved 2D temperature distribution on the boiling surface.

The experimental facility used in this research is described in Section 2, the data reduction methodology in Section 3, and the results for bubble nucleation and growth and nucleate-boiling heat transfer in Sections 4.1 Bubble nucleation and growth, 4.2 Nucleate-boiling heat transfer, respectively.

Section snippets

Pool boiling facility

The experiments were conducted in the facility shown in Fig. 1. A thin film made of Indium–Tin-Oxide (ITO) was electrically heated. ITO is composed of 90% In2O3 and 10% SnO2 by weight. Boiling occurred on the upward facing side of this film which had an exposed area of 30 × 10 mm2, and was 0.7 μm thick. The ITO was vacuum deposited onto a 0.4-mm thick sapphire substrate. This heater was connected to a DC power supply to control the heat flux at the surface. The cell accommodating the test fluid was

Data reduction and uncertainty

The raw data obtained for each heat flux are in the form of hundreds of frames, each representing a two-dimensional infrared intensity distribution on the heater surface (see Fig. 3). The conversion from IR intensity to temperature is done via a calibration curve, obtained using vendor-supplied blackbody simulators; with an accuracy of about 2%, or 2 °C. The nucleation sites appear as short-lived dark (cold) spots on the IR image. The edge of the nucleation sites is sharp because there is very

Experimental results

Boiling curves for the various experimental runs discussed in this paper are shown in Fig. 6. The critical heat flux (CHF) consistently occurred between 900 kW/m2 and 1000 kW/m2. In our experiments CHF was always accompanied by a large sudden excursion of the heater surface temperature resulting in physical burnout of the heater. The boiling curves are generated by taking the average temperature for a ∼5 × 5 mm2 area in the center of the heater across a set of IR images for each heat flux shown.

Conclusions

Synchronized high-speed video and infrared thermometry were used to obtain time- and space-resolved information on bubble nucleation and boiling heat transfer. This approach provides a detailed and systematic method for investigating the fundamentals of nucleate boiling. Data on bubble departure diameter and frequency, growth and wait times, and nucleation site density can be effortlessly measured for all nucleation sites on the heater surface. The main findings of the study are as follows:

  • -

Acknowledgements

C.G.’s doctoral project and purchase of the IR camera were supported by the King Abdulaziz City of Science and Technology (KACST, Saudi Arabia). The authors would like to thank Dr. Jim Bales and the Edgerton Center at MIT for providing access to the HSV and for their generous support and advice. Special thanks to Dr. Hyungdae Kim (MIT) and Prof. Gretar Tryggvason (Worcester Polytechnic Institute) for reviewing the article.

References (30)

  • D. Bestion, D. Lucas, M. Boucker, H. Anglart, I. Tiselj, Y. Bartosiewicz, Some lessons learned from the use of...
  • W.K. In, T.-H. Chun, CFD analysis of a nuclear fuel bundle test for void distribution benchmark, N13-P1259, in:...
  • N. Kurul, M.Z. Podowski, Multidimensional effects in forced convection subcooled boiling, in: Proceedings of the 9th...
  • N. Kolev

    How accurately can we predict nucleate boiling?

  • A. Sanna, C. Hutter, D.B.R. Kenning, T.G. Karayiannis, K. Sefiane, R.A. Nelson, Nucleate pool boiling investigation on...
  • Cited by (0)

    View full text