Microscale heat transfer measurements during pool boiling of FC-72: effect of subcooling

https://doi.org/10.1016/j.ijheatmasstransfer.2004.02.008Get rights and content

Abstract

Heat transfer under nucleating bubbles were directly measured using a microheater array with 100 μm resolution with low and high subcooled bulk liquid along with images from below and from the side. The individual bubble departure diameter and energy transfer were larger with low subcooling but the departure frequency increased at high subcooling, resulting in higher overall heat transfer. The bubble growth for both subcoolings was primarily due to energy transfer from the superheated liquid layer-relatively little was due to wall heat transfer during the bubble growth process. Oscillating bubbles and sliding bubbles were also observed in highly subcooled boiling. Transient conduction and/or microconvection was the dominant heat transfer mechanism in the above cases. A description of the experimental setup, the results, and implications for modeling boiling heat transfer are presented. A transient conduction model is developed and compared with the experimental data.

Introduction

The mechanisms by which bubbles transfer energy from a wall are actively being investigated experimentally and numerically. Many mechanisms for bubble heat transfer have been suggested (see Carey [1] for a short review) but the microconvection model of Mikic and Rosenhow [2] and the microlayer model of Cooper and Lloyd [3] are the most widely cited. Recently, a contact line model of bubble heat transfer has also been presented (e.g., Stephan and Hammer [4] and Mitrovic [5]). The large number of experimental studies to date have been supplemented recently by numerical simulations made possible by advances in computer hardware and interface tracking codes. Examples of recent numerical simulations include Welch [6], Son et al. [7], and Yoon, et al. [8]. The proliferation of competing bubble heat transfer models has primarily been due to the inability to make detailed measurements in the vicinity of the bubble.

Yaddanapudi and Kim [9] measured local heat transfer data underneath single bubbles nucleating periodically from a single site for saturated FC-72 at 1 atm (Tsat=56.7 °C) and wall temperature 79.2 °C. They used a heater array with individual heaters 270 μm in size. The bubble departure diameter was about 370 μm, only slightly larger than a single heater. Their results indicated that bubble heat transfer mechanisms were different from the widely accepted view of microlayer evaporation being the dominant heat transfer mechanism in saturated pool boiling. Bubble growth occurred primarily due to energy gained from the superheated liquid layer. Bubble departure resulted in removal of part of the superheated layer, allowing energy to be transferred from the wall through transient conduction and/or microconvection, consistent with the model of Mikic and Rosenhow [2].

Demiray and Kim [10] presented local heat transfer data underneath bubbles nucleating from a single site for single and vertically merging bubbles under conditions similar to Yaddanapuddi and Kim [9], but using an array with heaters 100 μm in size. The surface temperature of the heater array and the bulk fluid temperature during the experiment were 76 and 52 °C, respectively. Bubbles that nucleated at this site alternated between two modes: single bubble mode and multiple bubble mode. In the single bubble mode, discrete bubbles departed from the heater array with a waiting time between the departure of one bubble and nucleation of the following bubble. In the multiple bubble mode, bubble nucleation was observed immediately after the previous bubble departed. The departing bubble pulled the growing bubble off the surface prematurely and the bubbles merged vertically forming small vapor columns. The data indicated that the area influenced by a single bubble departing the surface was approximately half the departure diameter. Microlayer evaporation was observed to contribute a significant, but not dominant, fraction of the wall heat transfer in the single bubble mode. Microlayer evaporation was insignificant in the multiple bubble mode, and heat transfer occurred mainly through transient conduction/microconvection during liquid rewetting as the bubble departed the surface.

The objective of this work is to determine the mechanisms by which heat transfer occurs for bubbles nucleating from a single site, and extends the work of Yaddanapudi and Kim [9] and Demiray and Kim [10] to high subcooling. The array with microheaters 100 μm in size were used to obtain time and space resolved heat transfer measurements. A description of the experimental setup, the results, and implications for modeling boiling heat transfer are presented. A transient conduction model for the wall heat transfer is developed and compared with the data.

Section snippets

Heater array

An array of 96 platinum resistance heater elements deposited on a quartz wafer provided local surface heat flux and temperature measurements. A photograph of the heater array is shown in Fig. 1. Each element in the array was approximately square in shape, nominally 0.01 mm2 in area, and consisted of 2 μm wide Pt lines spaced 2 μm apart. Each heater had a nominal resistance of 8 kΩ with a temperature coefficient of resistance of 0.0019 °C−1. The lines that supply power to the heaters are routed

Data reduction and uncertainty analysis

Because each heater had its own feedback control circuit, we were able to measure the instantaneous heat flux required to maintain each heater at a constant temperature (qraw). Uncertainties in qraw are relatively small since they were computed directly from the measured voltage across the heaters and since the heater resistances do not change much. The maximum uncertainty in the voltage across the heater is 0.04 V. The uncertainty in heater resistance is about 45 Ω. Since the heater

Experimental results

All data were obtained with the wall temperature fixed at Twall=76 °C and P=1 atm (Tsat=57 °C). Two subcooling levels were investigated. The data taken with Tbulk=52 °C will be referred to as low subcooling, while the data taken with Tbulk=41 °C will be referred to as high subcooling.

Advancing contact line heat transfer model description and validation

The experimental data discussed in this paper indicate that transient conduction/microconvection during liquid rewetting of the surface account for most of the wall heat transfer. A transient conduction model for advancing contact line heat transfer is developed next.

Conclusions

A microheater array was used to obtain time and space resolved wall heat transfer data under nucleating bubbles with low and high subcooling. Single bubbles departing the surface gained the majority of their energy from the superheated liquid layer and not from the wall, indicating that microlayer and contact line heat transfer are not significant. Transient conduction/microconvection was the dominant mechanism for bubble heat transfer. Heat transfer measurements and comparison with a simple

Acknowledgements

This work was sponsored by the Office of Biological and Physical Research at NASA under NCC3-783. Mr. John McQuillen was the grant monitor.

References (14)

There are more references available in the full text version of this article.

Cited by (171)

View all citing articles on Scopus
View full text