.30.? J
i. :4...

1.0K. 3:1".
.. .
h 3. la! a

 

. : 3.3
A If 3

 

1...: .3.
n. {a v

 

51.33.-.

 

guy: ma. . . ‘ . . w t:
a; .3. , Wm?

 

 

LIBRARY
Michigan State
University

This is to certify that the
thesis entitled

POOL BOILING HEAT TRANSFER
FROM POROUS-COATED SURFACES IN FC-72:
THE EFFECTS OF SUBCOOLING AND
NON-BOILING IMMERSION TIME

presented by

KUIYAN XU

has been accepted towards fulfillment
of the requirements for the

MS. degree in Department of Mechanical
Mnmfim

 

 

Major Professor’s Singatdre
4/21 52$; 25m”

Date

MSU Is an Afllrmadve Acflon‘Equnl Opportunity Won

.- ‘._.-._.--.--‘.~ .—

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 c:/CIFIC/DateDue.indd-p.15‘

 

POOL BOILING HEAT TRANSFER FROM POROUS-COATED SURFACES
IN FC-72: THE EFFECTS OF SUBCOOLING AND
NON-BOILING IMMERSION TIME
By

Kuiyan Xu

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department Of Mechanical Engineering

2005

ABSTRACT
POOL BOILING HEAT TRANSFER FROM POROUS-COATED SURFACES
IN FC-72: THE EFFECTS OF SUBCOOLING AND
NON-BOILING IMMERSION TIME
By

Kuiyan Xu

The present research is an experimental study of pool boiling behavior of surfaces
coated with thin porous layers. The fluid employed is FC-72, a highly-wetting dielectric
perfluorocarbon with zero ozone-depletion potential (ODP). This creates the potential for
electronic cooling application. Different surfaces, including the super-smooth surface
(SSS), the High Flux TM surface (HF S), and the new electrochemical deposition surface
(EDS) were tested, and the test results were compared. Both subcooled and saturated
fluid pools were studied. The boiling hysteresis phenomenon was studied for these
surfaces under different boiling conditions, which include the fluid bulk temperature and
the non-boiling immersion time. Results of the study showed that the porous-coated
surface dramatically enhanced the nucleate boiling heat transfer performance. The
boiling hysteresis phenomenon is more prominent on porous-coated surfaces than on

smooth surfaces, and subcooling can deteriorate this phenomenon.

ACKNOWLEDGEMENTS

I would like to express my gratitude to my advisor, Dr. John R. Lloyd, for
providing me the opportunity to perform this research and to pursue my graduate study
under his direction. His patience, support and guidance were invaluable to the
completion of this work.

I would like to thank the US Civilian Research and Development Foundation for
sponsoring this work (grant number RPl-2337-ST-02). I truly appreciate 3M United
States for providing the FC-72 sample to complete this experiment.

In addition, I wish to acknowledge the efforts of my parents for their education
and support in my life. They taught me how to face difficulties and never give up.

Most of all I wish to express my particular gratitude to my wife Rong and
daughter Heting for their understanding, support and sacrifice all these years. I could not

have gone so far without you.

iii

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. vi
LIST OF FIGURES ................................................................................ vii
LIST OF APPENDICES ............................................................................ ix
NOMENCLATURE ................................................................................. x
CHAPTER 1
INTRODUCTION .................................................................................... l
1.1 Pool Boiling .................................................................................. 2
1.2 Enhanced Surfaces ........................................................................... 4
1.2.1 Machined Surfaces ..................................................................... 4
1.2.2 Porous-coated Surfaces ............................................................... 8
1.3 Parametric Effects ........................................................................... 11
1.3.1 Pressure ............................................................................... 11
1.3.2 Subcooling ........................................................................... 12
1.3.3 Electric Field ......................................................................... 13
1.4 Nucleate Boiling Mechanisms ............................................................ 15
1.4.1 Boiling on Plain Surfaces ........................................................... 15
1.4.2 Boiling on Porous Coated Surfaces ............................................... 18
1.5 Boiling Hysteresis .......................................................................... 20
1.5.1 Boiling Hysteresis Characteristics ................................................ 21
1.5.2 Boiling Hysteresis Mechanisms ................................................... 22
T08 Hysteresis ...................................................................... 24
TD Hysteresis ........................................................................ 25
CHAPTER 2
EXPERIMENTAL APPARATUS AND PROCEDURE ...................................... 28
2.1 Experimental Apparatus .................................................................. 28
2.1.1 Test Facility ........................................................................ 28
2.1.2 Test Surfaces ........................................................................ 31
Super Smooth Surface (SSS) ................................................... 31
Electrochemical Deposition Surface (EDS) .................................... 33
High Flux TM Surface (HFS) ...................................................... 33
Surface Microscopic Characteristics ............................................. 33
2.2 Experimental Procedure .................................................................. 35
CHAPTER 3
RESULTS AND DISCUSSION .................................................................. 37
3.1 Pool Boiling Enhancement of Porous-coated Surfaces .............................. 37

iv

3.1.1 Saturated Pool Boiling ............................................................ 37

3.1.2 Subcooled Pool Boiling ............................................................ 40

3.2 Effects of Nonboiling Immersion Time ................................................ 46

3.3 Boiling Hysteresis Phenomenon ......................................................... 49

3.4 Boiling Curves with Water as Working Fluid .......................................... 50
CHAPTER 4

SOCIAL AND ETHICAL CONSIDERATIONS ............................................... 53
CHAPTER 5

CONCLUSIONS .................................................................................... 54
CHAPTER 6

RECOMMENDATIONS ........................................................................... 56

APPENDICES ....................................................................................... 57

BIBLIOGRAPHY ................................................................................... 74

Table

3.1

A]

8.1

3.2

8.3

8.4

8.5

8.6

8.7

B8

8.9

01

LIST OF TABLES

Title Page
Effect of non-boiling immersion time on incipient superheat and
temperature overshoot ............................................................ 47
Change of surface temperature after each voltage increase ..................... 58

Experimental data of the SSS in 5K subcooled FC-72 with 72 hours
of non-boiling immersion time ...................................................... 61

Experimental data of the SSS in saturated FC-72 with 72 hours
of non-boiling immersion time ...................................................... 62

Experimental data of the EDS in 5K subcooled F C-72 with 72 hours
of non-boiling immersion time ...................................................... 63

Experimental data of the EDS in saturated FC-72 with 72 hours
of non-boiling immersion time ...................................................... 64

Experimental data of the HFS in 5K subcooled FC-72 with 72 hours
of non-boiling immersion time ...................................................... 65

Experimental data of the HF S in saturated FC-72 with 72 hours
of non-boiling immersion time ...................................................... 66

Experimental data of the HF S in 5K subcooled FC-72 with 48 hours
of non-boiling immersion time ...................................................... 67

Experimental data of the HFS in 5K subcooled FC-72 with 24 hours
of non-boiling immersion time ...................................................... 68

Experimental data of the HFS in 5K subcooled F C-72 with 0 hour
of non-boiling immersion time ...................................................... 69

Heat flux uncertainty for different surfaces ........................................ 73

vi

Figure
1.1

1.2

1.3

1.4
1.5

1.6

1.7

1.8
1.9
2.1
2.2
2.3
2.4
2.5
2.6

3.1

3.2

LIST OF FIGURES

Title Page
Boiling curve and boiling regimes for a plain horizontal tube .................. 3

Boiling of methanol on a horizontal tube. (a) Nucleate boiling. (b)
Transition boiling. (c) Film boiling .............................................. 5

Theoretical models of (a) Micro-porous. (b) Porous non-conducting

coated surfaces. (c) Porous conducting surfaces ............................... 9
Magnified cross-section of a plain, smooth metallic surface ................... 15
Cavity and bubble geometry ........................................................ 16

Nucleate boiling mechanisms. (a) Bubble agitation, (b) Vapor-liquid
exchange, (c) Evaporation .......................................................... 17

Conceptual model of boiling in a porous matrix of sintered metallic

particles ............................................................................. 19
Characteristics curves of nucleate pool boiling hysteresis ...................... 22
Activation of neighboring cavities .................................................. 26
Schematic of experimental setup ................................................... 29
Heat transfer test section with front Teflon sleeve removed .................... 3O
Electrochemical deposition experimental set-up ................................. 32
SEM image of the SSS ............................................................... 34
SEM image of the HFS ............................................................... 34
ESEM image of the EDS ............................................................ 35

Boiling curves of different surfaces in saturated F C-72 with 72 hours of
non-boiling immersion time 38

Boiling curves of different surfaces in 5 K subcooled FC-72 with 72
hours of non-boiling immersion time ............................................ 41

vii

3.3

3.4

3.5

3.6

3.7

3.8

3.9

A.1

Boiling curves of SSS in 5K subcooled and saturated FC-72 with 72

hours of non-boiling immersion time

Boiling curves of EDS in 5K subcooled and saturated F C-72 with 72

hours of non-boiling immersion time

Boiling curves of HF S in 5K subcooled and saturated FC-72 with 72

hours of non-boiling immersion time

Patch boiling of HF S in 5 K subcooled F C-72 ...............................

Boiling curves of HF S in 5K subcooled F C-72 with different non-

boiling immersion time ..........................................................
Hysteresis area ....................................................................
Boiling curves of different surfaces in 10 K subcooled water ..............

Change of surface temperature after each voltage increase .................

viii

.43

.44

.45

....47

....48

....50

....51

....6O

Appendix

LIST OF APPENDICES

Title Page
Steady State Verification .................................................. 58
Experimental Data .......................................................... 61
Uncertainty Analysis ....................................................... 7O

ix

NOMENCLATURE

a length (m)

b width (m)

E power (W)

hfg latent heat of vaporization (kJ/kg)
L length (m)

M mass (kg)

P pressure (Pa)

q heat flux (W/mz)

q” heat flux (W/mz)

R universal gas constant

R resistance

r radius (m)

T temperature (°C)

AT wall superheat (K)

At temperature difference (°C)
U voltage (V)

Greek Symbols

[3 contact angle

7 contact angle

6 conic angle

6 wall superheat
p density (kg/m3)
0 surface tension (N/m)
(p conic angle
Subscripts

b bulk liquid

cav cavity

g residual gas

in inside

L local

out outside

3 saturation

sat saturation

sub subcooled

v vapor

w boiling surface

xi

CHAPTER 1

INTRODUCTION

Industrial demands for more efficient boilers and evaporators have spurred the
development of methods to increase boiling efficiency. Enhancement of nucleate
boiling heat transfer has received ever-growing interest in a variety of industries, such as
refrigeration, air-conditioning, and chemical process etc. Many different boiling
surfaces with enhancement characteristics have been created in the past. These surfaces
take a number of forms from simple low integral fins with various fin profiles to more
complicated re-entrant cavity type surfaces, such as structured and porous coated
surfaces.

In this work the fluid employed is F C-72, a highly-wetting dielectric
perfluorocarbon with zero ozone-depletion potential (ODP). This creates the potential
for electronic cooling application. Different surfaces, including the super-smooth
surface (SSS), the High Flux TM surface (HFS), and the new electrochemical deposition
surface (EDS) were tested, and the test results were compared. A pool boiling facility
was designed to perform the experiment. Both subcooled and saturated fluid pools were
studied. The boiling hysteresis phenomenon was studied for these surfaces under
different boiling conditions, which include the fluid bulk temperature and the non-boiling

immersion time of the surface.

1.1 POOL BOILING

Pool boiling refers to a situation in which the bulk liquid is quiescent and its
motion near the surface is due to free convection and to mixing induced by bubble
growth and detachment. Pool boiling heat transfer can be described with reference to
the pool boiling curve, which shows the relationship between the heat flux leaving the
heated wall and temperature difference between the heated surface and the surrounding
bulk fluid, also known as the wall superheat.

Figure 1.1 shows the boiling curve and boiling regimes for boiling on the outside
of a typical plain, smooth tube in water. The curve is divided into four distinct heat
transfer regimes: single-phase natural convection, nucleate pool boiling, transition boiling,
and film boiling.

As the wall superheat increases from an initial value of zero, the regime is in
single-phase natural convection regime until the first vapor bubbles form on the heated
wall (Point A in Figure 1.1), which is called the incipience of boiling.

As the heat flux increases to a higher value, point B, the nucleate pool boiling
regime has been attained. In this regime the surface is densely populated with vapor
bubbles, and most of the heat change is through direct transfer from the surface to the
liquid which is in motion at the surface, not through the vapor bubbles rising from the
surface.

With the increase of the superheat, a maximum in the heat flux is reached (Point

C in Figure 1.1). This point is the peak nucleate heat flux and usually termed as the

2

critical heat flux.

 

 
    
  
 
     
 
 
 
 

 

 

......., - ......., r - fir...
Peak Nucleate Heat Flux
1000': 1
. Fully Transition :
« Developed Boiling -
500: Nucleate ‘1
. Boiling ,
f." -l
E
3 1001
x . . i
V 1 Discrete .
g 50* Nucleate "
C : Boiling '
‘6 . a
a:
I
10'} A . Nucleation ‘3
j / Superheat Film 3
5- ” Belling ..
' ll Minimum
Ill Heat Flux
[I Natural Convection
/
1 I U V I YTIIT I I I ITI I II' I I j I
ll 10 100 1000

 

Wall Superheat (K)

Figure 1.1 Boiling curve and boiling regimes for a plain horizontal tube [42].

After the curve reaches point D, the minimum heat flux point, the regime enters
the film boiling, in which the surface is totally covered by a vapor blanket. In this
regime, heat is transferred from the surface to the liquid via conduction and radiation of
the vapor.

Vapor formation and bubble dynamics in the regime of nucleate boiling, transition

boiling, and film boiling are shown in Figure 1.2. These photographs are from boiling

of methanol on a horizontal tube.

1.2 ENHANCED SURFACES

Enhanced surfaces exhibit significantly higher heat transfer than smooth plain
surfaces. Thome [42] proposed two important parameters to consider for enhancing
boiling heat transfer: boiling nucleation and nucleation site density. Boiling nucleation
is the wall superheat required for boiling incipience, and nucleation site density is the
number of active boiling sites per unit area of the heated surface. A surface is
considered enhanced when either the surface requires a lower wall superheat for boiling
incipience, or it can generate a higher number of active nucleation Sites than a smooth

plain surface.

1.2.1 Machined Surfaces

Machined surfaces are those treated with mechanical methods and contain
grooves and cavities which can act as active nucleation site of boiling. One of the
earliest surfaces for enhanced boiling performance was a plain surface roughened with

emery paper or other abrasives. Corty and Foust [16] used different grades of emery

 

Figure 1.2 Boiling of methanol on a horizontal tube. (a) Nucleate boiling. (b)
Transition boiling. (c) Film boiling. Photographs courtesy of Professor J. W.
Westwater, University of Illinois at Urbana—Champaign [21].

paper to produce nickel surfaces with scratches ranging from 0.254 to 25.4 urn across by
0.05 to 0.635 um deep. They demonstrated that increasing roughness alone has
limitations on enhancement of the boiling process. Berenson [4, 5] used a combination
of mechanical roughening and oxidation to treat the surface. He found that the slope of
the boiling curve is higher for the roughened surfaces than for the mirror-finished surface,
and that variation of surface roughness can hardly affect the critical heat flux.

The first patent for an extended surface on the purpose of heat transfer
enhancement was Obtained by Still [40]. He applied this in the forced convection to air
on the outside of a tube. In 1960, Webber [45] successfully applied the finned tube
reboiler in a refinery light end unit. These studies demonstrated the viability of finned
surfaces in industrial applications. Myers and Katz [31] demonstrated that the fins can
not only increase the surface area but also change the boiling process to enhance
performance.

Chien and Webb [10] proposed two key geometric characteristics of the surfaces
are subsurface tunnels and surface pores or fin gaps. They believe that the important
dimensional parameters include the following: tunnel pitch, tunnel height, tunnel width,
tunnel base radius, tunnel shape, pore diameter, and pore pitch. Chien and Webb [10, 11,
and 13] performed a series studies using R-l 1, R-22, R-123, and R-134a as the working
fluids for an extensive variety of parameter combinations. They found that: (1) the
boiling heat transfer rate decreases as the tunnel height is reduced; (2) sharp tunnel

comers provide a greater enhancement; and (3) the boiling coefficient is strongly

6

influenced by pore size at a given heat flux.

Ramaswamy et a1 [35] investigated an enhanced surface consisting of a stacked
network of interconnecting channels which makes it highly porous. They studied the
effect of varying the pore size, pitch and height on the boiling performance with
fluorocarbon FC-72 as the working fluid. They found that a larger pore and a smaller
pitch resulted in higher heat dissipation at all heat fluxes. The heat dissipation increases
with an increase in the pore size (for the same pore pitch). In the range of pore sizes
tested (90—320 um), the largest pore size resulted in maximum heat dissipation. It was
found that the effect of pore pitch on heat transfer performance was more significant than
the pore size.

J iang et a1 [22] investigated the microfilm evaporation and two-phase flow inside
the microstructure of a machined porous surface and developed a new microscale heat
transfer model for boiling heat transfer on porous surfaces. They used R11, R22, R134a,
and water as the working fluids and determined the optimum microstructure for the given
conditions. Their results demonstrated that the heat transfer performance of porous
surfaces is augmented by enhancing the evaporation of the thin liquid films inside the
microchannels. They believe that the surface tension and evaporation of the working
fluids are the two most important parameters. For boiling in fluids with large surface
tensions, it is better to choose larger pore diameter and smaller space between two pores.
They also found that sharp channel comer angles, rough surfaces and crossed channels

can bring larger heat transfer area inside the channels.

7

1.2.2 Porous-coated Surfaces

Porous-coated surfaces are widely used in heat exchangers, heat pipes, and
electronic cooling. In these surfaces, an irregular matrix of potential nucleation sites is
formed by means of poor welding, sintering or brazing of particles, electrolytic
deposition, flame spraying, bonding of particles by plating, galvanizing, plasma spraying
of a polymer, or metallic coating of a foam substrate [8].

Milton got his patent for a porous sintered metallic coating in 1968 [30]. After
further improvement, he developed the commercial surface-High Flux TM. This surface
contains many irregular cavities similar to those of coral [23]. The particles are sintered
onto the surface and form randomly shaped cavity openings that are approximately 0.04
to 0.13 mm in diameter. The thickness of the porous coating is approximately 0.645
mm. The porosity of the matrix is 45%. Since the High Flux TM surface can provide
superior boiling heat transfer performance, it is often used as a benchmark for evaluating
boiling performance of other porous coated surfaces [1].

Thome [42] proposed that the key parameters for porous-coated surfaces include
particle size, layer thickness, particle material, and porosity. But it is difficult to
determine an Optimal combination of these parameters to maximize boiling performance.

With FC-72 as the working fluid, Chang and You [9] investigated particle size
effects on the boiling performances of micro-porous enhanced surfaces using five

different sizes of diamond particles. They compared the coating thicknesses with the

8

superheated layer thickness and classified the coatings into two groups: ‘micro-porous’

and ‘porous’ coatings (Figure 1.3). The micro-porous surfaces correspond to the

superheated lquid layer]

 

 

 

 

(b)

 

Figure 1.3 Theoretical models of (a) Micro-porous. (b) Porous non-conducting coated

surfaces. (c) Porous conducting surfaces [9].

coatings with thickness 5 100 um and the porous coatings have a thickness > 100 um.

The superheated layer thickness, 899, is calculated using one-dimensional transient heat
conduction. The micro-porous coatings show different characteristics of boiling
performance compared to porous coatings in incipient superheat, nucleate boiling and
CHF. A significant decrease of incipience superheat and an increase of CHF was
observed for the micro-porous surface regime [9].

Kim et a1 [24] investigated the mechanism of nucleate boiling heat transfer
enhancement from plain and microporous coated 390 um diameter platinum wires. The
materials they used for the microporous coated wire is the dielectric DOM (8-12 pm
synthetic Diamond particles/Omegabond 101 epoxy binder/Methyl-ethyl-keytone carrier).
Their results showed that the microporous coatings augment nucleate boiling
performance through increased latent heat transfer in the low heat flux region and
through increased convection heat transfer in the high heat flux region. The results
showed that the CHF for the microporous coated surface is significantly enhanced over
the plain surface. They believed that this is due to decreased latent heat transfer
(decreased vapor generation rate) and/or increased hydrodynamic stability from increased
vapor inertia, both of which are a direct result of increased nucleation site density.

Cieslinski [14] performed series experiments of heat transfer with distilled water
as the working fluid. He used stainless steel tubes of different diameters and flat
horizontal surfaces ranging from a smooth lapped finish through sandblasted surfaces, to
porous coated surfaces. Various methods of variation of surfaces, such as electrolytic

treatment, plasma spraying, gas-flame spraying and modified gas-flame spraying were

10

employed to form metal coatings. The results showed that the main parameters of the
porous coating that influence boiling heat transfer are thickness and porosity. It is not
affected by the methods of fabrication, providing the contact between the porous matrix

and the substrate is good.

1.3 PARAMETRIC EFFECTS
Pool boiling heat transfer performance can be affected by many parameters, such
as system pressure, subcooling, and electric field imposed on the working fluid. Some of

these parameters can change the boiling performance dramatically.

1.3.1 Pressure

Thome [42] reported that the boiling curves of plain surfaces should move to the
left with increasing pressure. Rainey et a1 [33] investigated the effect of pressure,
subcooling, and dissolved gas on pool boiling heat transfer from 1 cm2 plain and
microporous flat copper blocks immersed in FC-72. They found that the incipient
superheat increases with decreasing pressure for both microporous and plain surfaces.
The difference in the nucleate boiling performance between the plain and microporous
surfaces appears to decrease with increased pressure. The CHF values also show a
significant increasing trend with increased system pressure.

Nakayama et al [32] investigated the influence of system pressure on saturated

pool nucleate boiling heat transfer with R-ll as the working fluid. They found that at

11

smaller superheats, the higher the system pressure, the higher heat fluxes are sustained at
reduced wall superheat.

Zhang et a1 [46] performed experiments on pool boiling heat transfer from a
circumference-interrupted T-finned (CIT) tube and a Thermoexcel-E type tube at system
pressures ranging fiom l to 6 bar with ethyl alcohol and R-113 as the working fluids.
They demonstrated that with the increasing system pressure, the boiling heat transfer
coefficient increases at first, but the improvement is smaller for higher pressures. There
exists an optimal system pressure beyond which the boiling performance tends to

diminish.

1.3.2 Subcooling

The term surface subcooling here means the temperature difference between
saturation temperature of the working fluid and the surface. According to Thome [42],
the boiling curve moves to the right as the subcooling increases. Even small levels of
subcooling can dramatically affect the heat transfer process and substantially reduce the
boiling heat transfer coefficient.

Bajorek [3] investigated the effect of subcooling on boiling on a low-finned
capper tube. He obtained data for water, ethanol, and several mixtures at 101 kPa for
subcooling ranging from 0 — 40 K. He found that the largest drop in the heat transfer
coefficient occurred in the first 10 K of subcooling, where boiling performance was

reduced by 40% to 50%.

12

Rainey et a1 [34] found the most prominent effect of subcooling on boiling was
that regardless the subcooling level, the fully developed nucleate boiling curves collapse
onto a single line. They believe that as subcooling increases, the bubble departure
diameters and frequencies decrease which reduces the amount of heat transferred through
latent heat and microconvection. The increased subcooling also decreases the
superheated liquid layer thickness which increases natural convection and Marongoni
convection heat transfer. These combined effects result in relative insensitivity of the
nucleate boiling curve to liquid subcooling.

Demiray et al [ 17] measured the heat transfer under nucleating bubbles using a
microheater array with different subcooling. They found that the individual bubble
departure diameter and energy transfer were larger with low subcooling. But as the
subcooling increases, the bubble departure frequency increases which results in higher

overall heat transfer.

1.3.3 Electric Field

Leontiev et a1 [26] investigated the effect of an electric field on the boiling of
liquid nitrogen. They believed that an external electric field can cause an additional
surface force directed into the gas phase and essentially influences the nucleation and
bubble dynamics with a dielectric liquid as working fluid. Both uniform and
non-uniform electric fields were applied on horizontal smooth and corrugated surfaces.

The results showed that uniform electric field with an intensity of up to 107 V/m had no

13

essential influence on the initial part of the boiling curve of the smooth surface. The
non-uniform electric field, created by a metallic pin attached to the test surface, reduced
the incipience superheat of both the smooth and corrugated surfaces as compared to the
results with the uniform electric field.

Vorob’ev et a1 [44] found that the presence of an external electric field always
stimulates vapor bubble formation in superheated liquids. The uniform and
non-uniform electric fields initiate both condensation of superheated vapor and boiling of
superheated liquid. Provided that the field intensity exceeds some threshold value, the
vapor condensation would be intensified with the presence of an external electric field,
regardless it is uniform or not. But for the non-uniform field, this threshold value is
smaller.

Snyder et a1. [39] performed an experiment to produce a dielectrophoretic (DEP)
force over the length of a horizontal platinum wire heater. They concluded that the
overall boiling heat transfer coefficient in the presence of an electric field can be modeled
as the summation of a heat transfer coefficient due to bubble dynamics and a heat transfer
coefficient due to electroconvection. In terms of the relationship between the bubble
dynamics and the heat transfer, they believed that the effect of a variable DEP force is
similar to the effect of a variable buoyancy force. The heat transfer in the presence of
an electric field will be enhanced if the effective gravity acts to hold the vapor bubbles
near the heated surface, while at the same time permitting access of the liquid to the

surface in order to prevent dryout.

l4

1.4 NUCLEATE BOILING MECHANISMS

1.4.1 Boiling on Plain Surfaces

Thome [42] states that a surface appearing to be smooth is actually covered with
small cavities of various sizes, shapes, and depths (Figure 1.4). Clark et a1 [15] have
demonstrated that boiling nucleation from a smooth surface occurs at microscopic

imperfections in the surface.

Machined Surface

 

Nuclei

Figure 1.4 Magnified cross-section of a plain, smooth metallic surface [42]

These cavities are on the order of 01-10 pm in diameter and act as nucleation
sites due to their ability to trap vapor [37]. The trapped vapor grows into a bubble and
detaches as the surface is heated. Once the bubble departs from the surface, a liquid
film again covers the cavity where some vapor remains trapped. This remaining trapped

vapor will act as an embryo for the next bubble.

15

Bankoff [4] and Griffith and Wallis [18] developed the criteria for a stable vapor
trapping conical cavity. It is based on the contact angle, 7, and the cavity cone angle, 0,
as shown in Figure 1.5 as

0 <y<90° (1.1)

When 0 and y satisfy this condition, the incipient superheat is calculated from the

Laplace and the Clausius-Clapeyron equations, to get

ZaT

SO!

At =
pvhfgrc

(1.2)
where At is the temperature difference between the trapped vapor temperature and the
distant liquid temperature, a is the liquid surface tension, Tsat is the distant fluid
saturation temperature, hfg is the latent heat of vaporization, pv is the density of the vapor,
and rcav is the radius of the nucleation site. For such low contact angle liquids as FC-72,

Equation 1.1 will be valid only for very steep walled conical cavities. If this criterion is

met, Equation 1.2 predicts a minimum incipient superheat of only 1.8K for FC-72 [28].

LlOUlD l {C

 

 

Rx:

VAPOR SOLID

Figure 1.5 Cavity and bubble geometry [29].

16

Hsu and Graham [19] summarized the heat transfer mechanisms responsible for
heat transfer in nucleate pool boiling on plain, smooth surfaces as follows (see Figure
1.6 ) :

Liquid motion

\ f

I I
77””
-¢- ’\ X x .4”
I

UNI/IU/NNHH/U/

Heated surface

\022

(b) Ill/ll] l/f/lll/U/l/N/T/lllf

 

(a)

 

 

 

 

 

 

 

Superheated
liquid
I Evaporation
I A
(c) I/I/T/I/H‘ III/nm/m/mr
Microlayer

Figure 1.6 Nucleate boiling mechanisms. (a) Bubble agitation, (b) Vapor-liquid
exchange, (c) Evaporation [l9].

1. Bubble agitation (Figure 1.6 a): Improved liquid-phase convection results
from the motion imparted to the liquid from the growth and departure of vapor bubbles.

The natural convection process has been transformed into a “forced” convection process.

17

Heat is transferred from the surface by the superheated liquid.

2. Vapor-liquid exchange (Figure 1.6 b): Convection to the liquid is enhanced
by the quenching of the heated wall by fresh liquid rushing in after the leaving of a vapor
bubble. The thermal boundary layer covering the bubble and the adjacent wall is
continuously removed.

3. Evaporation (Figure 1.6 c): Vapor bubbles grow by virtue of the heat
conducted into the liquid from the heated wall and then to the bubble surface, where
phase change happens. Vaporization occurs from (1) a liquid microlayer trapped
between the heated wall and the bottom of a rapidly growing bubble and (2) the original
thermal boundary layer covering the top of the bubble.

The actual boiling process is a combination of these three heat transfer

mechanisms.

1.4.2 Boiling on Porous Coated Surfaces

Bergles and Chyu [8] postulate the mechanisms of boiling from porous coatings.
Consider nucleation in the porous matrix shown in Figure 1.7. If the matrix material is
poorly wetted, vapor or vapor plus noncondensable gas will be retained in the interstitial
space when the temperature is reduced below the saturation temperature. Altemately, if
the liquid has more wetting ability, re-entrant and double re-entrant cavities are required
to permit stable dropwise formation so that the cavities are not filled with subcooled

liquid. In any case, the matrix increases the probability that nucleation sites are

18

available which will remain active for repeated cycles of heating and cooling.

Griffith and Wallis [18] demonstrated that the geometry of the micro-cavity
containing trapped vapor is important to the bubble nucleation process. They found the
re-entrant type of micro-cavities were stable, easily activated boiling sites. Benjamin and
Westwater [5] reported the construction of an artificial re-entrant cavity with a mouth
diameter of 102um, which was described to be stable, remaining active down to low wall

superheats for boiling mixture of water and ethylene glycol.

    

  
 
  

    

: \ --._ ~' ‘—

' ‘
. - first. mater.
"////

l\\“-.--..1’

E¢§§cponous MATRIX

,, RE-ENTRANT CAVlTlES

4%???5U35TRATE SERVING AS INITIAL

gggg; LIQUID NUCLEATION srrts
VAPOR

Figure 1.7 Conceptual model of boiling in a porous matrix of sintered
metallic particles [8].

Since then, porous metallic coatings have received considerable attention.
Formed by bonding metal particles with diameters ranging from 44 to 1000m to a base
surface (coating thickness ranges from 250 to 2000um), porous metallic coatings have
demonstrated to greatly enhance nucleate boiling heat transfer performance.

Andrianov et a1 [2] believed that porous coatings essentially transform heat
transfer laws at bulk boiling. They proposed that there were two possible modes of
boiling corresponding to the bubble mechanism of the vapor departure from the outer
surface of the coating, whereas there was only one mode on the smooth surface. In
bubble mode I, the vapor generating zones inside the porous coating are not connected
with each other through the vapor-filled pores (heat transfer law q~0n corresponds to nyél ,
0 is the wall superheat). In bubble mode 11, vapor filled pores are connected over the
whole system, and the evaporation zone is separated from the heated surfaces by a thin
layer of vapor film which is stabilized in the body of coating (heat transfer law q~0n
corresponds to n=1). A nonequilibrium phase transition from one bubble boiling mode
to the other is accompanied by an abrupt change of 0 or a discontinuity in derivatives.

Thome [43] believed that the primary enhancement mechanisms for re-entrant
type enhanced surfaces were: enhanced nucleation from the larger embryonic bubbles,
increased thin film evaporation due to the large internal area of the porous structure, and

two-phase convection within the porous structure.

1.5 BOILING HYSTERESIS

20

1.5.1 Boiling Hysteresis Characteristics

Research on the boiling hysteresis phenomenon can be traced back to the 19503,
when Corty and Foust [16] found that the immediate past history of the boiling surface
had a pronounced effect on the superheat required for boiling incipience, which was far
beyond that required in normal boiling conditions. This is defined as temperature
overshoot (TOS), which means the surface temperature is highly superheated preceding
the transition from the natural convection state to the nucleate boiling state [3 8].

Boiling hysteresis can be defined as the discrepancy between the boiling curves
obtained by increasing and then decreasing the applied heat flux. For the boiling
hysteresis to exist in the boiling developing process, Shi et a1 [38] termed it as
Temperature Deviation (TD) hysteresis. The two kinds of boiling hysteresis phenomena,
TOS hysteresis and TD hysteresis, are generalized in Figure 1.8.

Porous-coated surfaces have proven to be an efficient method to increase the
performance of boiling heat transfer. But experiments by Bergles [8] and Marta [28]
have demonstrated that boiling hysteresis phenomenon is more prominent on
porous-coated surfaces than on smooth surfaces.

It is obvious that the boiling hysteresis may jeopardize cooling process of
high-power density components. This will restrict the application of pool boiling

cooling in the industrial fields.

21

 

 

 

AT

W

Figure 1.8 Characteristics curves of nucleate pool boiling hysteresis:
F DNPB-fully—developed nucleate pool boiling; PNPB-partial nucleate pool boiling;
DNC-departure from natural convection; NC-natural convection [38].

1.5.2 Boiling Hysteresis Mechanisms

The nucleation of vapor bubbles from embryonic vapor/gas pockets in
microcavities on the heated surface has been proven to account for boiling incipience and
much of the ebullient heat transfer from metallic surfaces to the boiling fluids [36]. If
the surfaces are immersed in the liquid with high wetting characteristics, a large

percentage of the cavities would be filled by the liquid. The cavities which trapped

22

some residual gases would function as the first nucleation sites.

Shi et al [38] analyzed the boiling hysteresis mechanisms. Consider an
embryonic bubble with a radius of re growing out from a conical cavity. Inside the
bubble there may include some noncondensable gases. To achieve mechanically stable
of the bubble, the internal pressure of the bubble must exceed the external or the local
pressure by an amount proportional to the surface tension dividing by the radius of
curvature at the interface. Assuming the bubble to be hemispherical and considering the
presence of the initial residual gas, the mechanical equilibrium for this embryonic bubble
is:

20'

Pv+Pg-PI =7“ (1.3)

where PV and Pg are the vapor partial pressure and initial partial pressure of the residual

gas in the bubble respectively. PL is the external or local ambient pressure.
From the thermodynamic nucleation theory, the liquid superheat needed to sustain

the growth of this embryonic bubble should be

 

T
ATwi=Twi—Ts= s 20_Pg (14)
pvhfg crc '

where c is a constant related to the thermal layer thickness surrounding the bubble.

Assuming the residual gas is an ideal gas, so
3m RT
1) z __8_L

C

g

where mg is the initial mass of residual gas in the cavity. Equation 1.1.2 can then be

23

expressed as

T3 20' 3ng];
‘—‘3— (1.6)

 

AT 2
pvhfg crc 4727'

T08 Hysteresis

TOS hysteresis occurs at the incipience of the boiling process. The first bubbles
generate from the cavities in which a small amount of residual gas preexisted. The
liquid superheat needed for the first bubble generation can be predicted by equation 1.6.
After the first bubbles leave the surface, more residual gas (vapor) will be trapped in

these cavities. The liquid superheat needed for further bubble grth is

3m RT
ATW2 = Ts 2" _ gr 3 v
pvhfg are 4711;,

 

(1.7)

where mgc is an equivalent mass of the new residual gas-vapor mixture in the cavities.
ATW2 is the normal surface superheat under the given heat flux. Because mgc> mg,

then ATW2<ATw1. This is why hysteresis will occur at the boiling incipience. The

temperature overshoot can be expressed by the following equation:

 

3QER
ATTOS = ATW, — ATM = 4mfp.hfg (mg, — mg) (1.8)

From Equation 1.8 it can be seen that the larger the mass of residual gas (vapor)

trapped in the cavities, the smaller the temperature overshoot will have at the boiling

24

incipience. When Inge: mg, there will be no occurrence of temperature overshoot at all.

TD Hysteresis

Shi et a1 [38] interpreted the occurrence of the TD hysteresis with the vapor
propagation after the incipience of boiling. As shown in Figure 1.9, the growth of the
first bubbles may activate the neighboring cavities which retained less or no residual gas
at a lower wall superheat. With the bubble interface spreading along the surface, the
vapor front would enter the neighboring cavity providing the liquid contact angle B>
(1t-(p). Then the neighboring cavity would be filled with gas (vapor) and be activated
(Figure 1.9 a). In another case, if [3< (rt-(p), which means either the conical angle of the
cavity (p is small, or the contact angle of the fluid [3 is small (Figure 1.9 b), the cavity
might initially have a small amount gas in it and not be activated at the initial wall
superheat. If the first growing bubble covers this cavity, the liquid inside the cavity
would be separated from the bulk liquid and becomes locally highly superheated by the
heated wall. So the small embryonic bubble would grow because of liquid evaporation

and be activated in the next bubble generation.

25

 

 

 

/’ \\ "'-
///""'\ \\_________
__ ________// \ ~\
1/ B _ \t‘”
/

 

(b)

Figure 1.9 Activation of neighboring cavities [3 8].

After the fully developed nucleate boiling is reached, if the heat flux is then
decreased, the cavities which were not activated in increasing heat flux are now active.
From equation 1.7 it can be found they need lower wall superheat. So the boiling is

augmented compared with the increasing process and causes the deviation of boiling

curves, which is termed as the TD hysteresis.

26

Bergles and Chyu [8] explained the large scale boiling curve hysteresis for High
Flux TM surfaces with the flooding of the porous matrix with liquid, which resulted only
relatively small sites are available for nucleation. They believed that for a specific
surface, the contact angle of the working fluid would strongly influence the extent of
temperature overshoot.

Tehver et a1 [41] investigated the relation between the effectiveness of heat
transfer and structural parameters of a plasma-sprayed coating with F-113 as the working
fluid. They tested different coatings such as aluminum, bronze, copper and corundurn.
The great variety hysteresis phenomena they observed suggest there are two main factors:
( 1) the boiling incipience conditions determined by the availability of active center
embryos and (2) the structure of a porous material characterized by non-uniform pore
cross sections for liquid and vapor flows. The joint influence of these two factors,
which depend on the wetting contact angle and the experimental conditions, results in a
great variety of boiling curves for porous surfaces.

This work will investigate the influence of subcooling and non-boiling immersion
time of test surfaces to the nucleate pool boiling performance. A new porous coating is
studied in this work using highly wetting fluid FC-72. Subcooled pool boiling data in
F C-72 and water are compared and discussed to demonstrate the effect of contact angle

between the fluid and the surface for the two fluids.

27

CHAPTER 2

EXPERIMENTAL APPARATUS AND PROCEDURE

2.1 EXPERIMENTAL APPARATUS

2.1.1 Test Facility

The experimental setup~ for this study is shown in Figure 2.1. The test fluid,
F072, is a highly-wetting dielectric perfluorocarbon produced by the Industrial
Chemical Products Division of 3M. The test liquid was contained within a 3500 ml
Pyrex beaker that was submerged in a 40cm><20cm><27cm vessel. This vessel served as
a constant temperature water bath. A Fisher 90 Refrigerated Bath was used to heat the
water to 56°C for the saturated tests, or 51°C for the 5 K subcooled test. An insulation
layer was applied to prevent heat losses of the water bath. The tests were conducted at
atmospheric pressure. An external, water-cooled condenser made by copper tube was

used during the test to prevent loss of the testing liquid.

A copper-constantan (Type T) thermocouple was placed within the test vessel to
measure the bulk test liquid temperature. The test section was mounted horizontally and

immersed in the test liquid.

The test section used in this research consisted of the test surface, the
25mm><45mm><45mm copper block with two embedded T-type thermocouples and a
cartridge heater. Since copper has very high thermal conductivity, it is an ideal heat

transfer material.

28

Thermocouple Condenser

 

 

 

 

read-out l
Power —— Bulk
supply temperature

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

———1Refrigerated batt

Figure 2.1 Schematic of experimental setup.

 

The copper block was covered by Teflon sleeves on all sides except the top one.
Silicon sealant was applied between the copper block and the Teflon sleeves to prevent
liquids from entering the inside of the test section. Figure 2.2 shows the structure of the

test section with the front Teflon sleeve removed.

Test surfaces are soldered on the top surface of the copper block. When

29

soldering the test surface, first cut the solder into 1 cm pieces. Second, turn on the
power to the cartridge heater until the thermal couple reading reaches 140 °C. Then
apply the solder to the upper surface of the copper block. After the solder is melted, put
the test surface onto the copper block carefully. To make the test surface and the copper
block upper surface contact tight, apply a weight on the test surface until the thermal

couple reading drops to room temperature.

Teflon Sleeve

Test Surface

Thermocouple if, . .. _ Cartridge Heater
Leads * ‘ ’ *

 

Figure 2.2 Heat transfer test section with front Teflon sleeve removed.

To measure the surface temperature of the test specimen, two copper-constantan

thermocouples (30 gauge, 0.075cm in diameter) were inserted and soldered into two

30

holes drilled in the copper block (1 -mm in diameter, S-mm in depth).

The thermocouples were connected to an OMEGA HH23 Microprocessor
Thermometer. The test surfaces were heated by the 0.635cm diameter by 3.81cm long
125-Watt Chromalox model CIR-1015 cartridge heater. The heater was first coated with
Chromalox Boron Nitride Lubricoat and then inserted into a pre-drilled hole in the copper
block. A Powerstat Variable Autotransforrner was used to control the input of the power

into the heater and the value of the voltage was displayed in a Wavetek 15XL multimeter.

2.1.2 Test Surfaces

Super Smooth Surface (SSS)

The copper super smooth surface was prepared by Ngai [33] following the
preparation procedure developed by Aitcheson [1]. At first, distilled water was used to
wash off the dirt on the copper surface. Then acetone was used to dissolve the possible
oil and grease. Afier that, the surface was wiped clean with a sofi cloth before the
polishing process. Sandpaper of different grits and diamond polishing compounds of
different grit sizes were used successively to process the surface. The copper surface
was first sanded 50 times in one direction using 320-grit sandpaper, then cleaned and
sanded 50 times in the direction perpendicular to the previous sanding direction.
Sandpapers of 400-grit, 600-grit, 1200-grit and 4000-grit, diamond-polishing compounds
of 6 micron, 3 micron, 1 micron and 0.25 micron were used progressively to remove the

scratches from the previous grit and get a smooth mirror-like surface.

31

Electrochemical Deposition Surface (EDS)

The electrochemical deposition surface (EDS) was created by Nicole Aitcheson [1]
using the procedure developed by Lloyd [27]. Figure 2.3 shows the experimental set-up
used for electrochemical deposition. The electrolyte solution was an aqueous solution of
1.5 molar reagent grade sulfuric acid (H2804) and 0.05 molar copper sulfate (CuSO4).
The copper ions of the CuSO4 served as the transferred species of ions and the H2804 was

the supporting electrolyte.

  

Cathode

 
 
  
  

    

.41" Reference
Electrode

2a ‘

Figure 2.3 Electrochemical deposition experimental set-up [1].

 
  

The anode and cathode were made from thick copper sheets with dimensions of
54.3 cmX30.5 cmX0.32 cm and 6.0 cm><7.5 cmx0.32 cm respectively. The anode surface
area was 34 times larger than the surface area of the cathode to insure that the electrode

reaction would be controlled by the cathode. A 14 gauge copper wire was inserted into a

32

polyethylene test tube with a capillary hole drilled in the bottom to serve as the reference

electrode. The cathode was the test surface for electrochemical deposition.

High Flux T” Surface (HFS)

The High Flux TM surface (HFS) was acquired from the commercially available
High Flux TM tube by UOP. First, one side of the tube was cut along the axial direction,
and then it was flattened carefully with a plastic hammer. After that, it was cut into

pieces of 4.5cmX4.5cm to be tested.

Surface Microscopic Characteristics

The test surfaces were studied with a high-resolution Scanning Electron
Microscope (SEM) and an Environmental Scanning Electron Microscope (ESEM).
Figure 2.4 shows the SEM image of the SSS. By taking 50 measurements on different
locations of the surface, the average width of the scratches was calculated and found to be
approximately 200nm [33].

The SEM image of HFS is shown in Figure 2.5. The lengths of the surface
structure vary in a wide range, with a maximum length of 48 um approximately [33].

Figure 2.6 shows the ESEM image of EDS. The particle-like surface formations

are the visible top surfaces of the columnar protrusions. Aitcheson [1] determined the

33

 

Figure 2.5 SEM image of the HFS [33].

34

 

Figure 2.6 ESEM image of the EDS [1].

average column-face sizes using the reference scale provided on the ESEM image.
Column-face sizes were measured at 20 different locations and were then taken an

average. The average column-face sizes of the EDS ranged from 0.7 pm to 1.5um.

2.2 EXPERIMENTAL PROCEDURE

Before the test, the surfaces were immersed in the working fluid for 0, 24, 48, and

35

72 hours to study the effects of the non-boiling immersion time.

In each test run, the working fluid was first heated to the saturated temperature

under atmospheric pressure, i.e. 56°C, for the saturated pool boiling experiment using the

Fisher 90 Refrigerated Bath. For the 5 K subcooled boiling case, the working fluid was
heated to 51°C. Then the cold water was opened to run through the condenser, which

will change the FC-72 vapor back to liquid state. Once the test liquid reached its
saturation temperature, it was left at this stage for 2 hours for degassing purposes.
During this interval, insulation layers were applied outside the water bath vessel to

prevent the heat losses.

After the degassing process, data acquisition was initiated. The heat-flux was
controlled by varying the input voltages using the Variable Autotransformer. Starting
from 20V, the voltage was first increased to 120V in a step by step process with an
increment of 10V, and then decreased from 120V to 20V with a decrement of 10V.
When the power to the cartridge heater was turned on, about 12 minutes were needed to
reach steady state conditions. After that, a steady-state condition was usually reached 8
minutes after each voltage increase. This was verified by the experimental data (see
Appendix A). The surface temperature and bulk fluid temperature were measured when
steady-state was reached. After all the data were collected, the cartridge heater was
powered off, but the water for the condenser was kept running until the F C-72 bulk
temperature reached room temperature to prevent loss of the FC-72. Then the test

section was removed and the FC-72 was carefully poured back into the storage container.

36

CHAPTER 3

RESULTS AND DISCUSSION

The data were analyzed and boiling curves of test surfaces under different
conditions were plotted. In the plot, the x-axis was the wall surface superheat in K and
the y-axis was the heat flux in W/mzK. Heat flux was calculated using the power input

of the cartridge heater divided by the upper surface area of the c0pper block.

3.1 POOL BOILING ENHANCEMENT OF POROUS COATED SURFACES

3.1.1 Saturated Pool Boiling

Figure 3.1 shows the saturated pool boiling curves of the three surfaces. In the
natural convection section of the curves, the SSS has higher heat flux than the HFS at any
given surface superheat. This may be due to the fact that there might have been some
gases trapped in the micro-cavities of the HFS and this porous coated layer reduced the

heat transfer in the natural convection region.

After the start of nucleate boiling, the SSS has the highest surface superheat and
the lowest boiling heat transfer coefficient at a constant heat flux among the three
surfaces, which can be seen in Figure 3.1. The EDS demonstrated lower heat flux
values than those of the HFS for a specified surface superheat. This means that the HF S
has a better nucleate boiling heat transfer performance than the EDS. The saturated
boiling experimental data show that at the same heat flux value (6.6 k/mz), the EDS

experiences an enhancement in the boiling heat transfer coefficient by 75%, while the

37

10'

Heat flux q" (W/m 2)

Boiling curves of different surfaces in saturated
FC-7 2

o.1—~~~

Surface superheat (K)

 

r- SSS Incr A. SSS Deer r EDS Incr
. —e- EDS Deer --I— HFS Incr ——e— HFS Decr

Figure 3.1 Boiling curves of different surfaces in saturated FC-72 with 72 hours

of non-boiling immersion time.

38

HF S by 276% compared to the SSS. From Figure 3.1, it can be seen that compared with
the SSS (18.2 K), the boiling incipience superheat of the EDS (5.9 K) is reduced by 12.3

K (68%), while that of the HFS (8.5 K) reduced by 9.7 K (53%). This means that

porous-coated surfaces can reduce the boiling incipience superheat.

For both the subcooled and saturated boiling conditions, the nucleate boiling of
the SSS occurs at a much higher surface superheat than that of the HFS or the EDS.
This is due to the fact that the SSS has lower site density than the porous coated surfaces,
and thus requires a higher surface temperature to dissipate the same amount of heat as the

porous coated surfaces.

According to Kim and Rainey [24], the increased heat transfer coefficients of the
microporous surfaces are not because of the increased area generated by the microporous
coating. This enhancement is a direct result of its significantly higher active nucleation
site density. Rainey and You [34] showed that an increase in surface area shifts the
boiling curve vertically, not horizontally, and has a significant effect only on the single

phase natural convection heat transfer.

Many researchers believe that the nucleate boiling enhancement from
porous-coated surfaces is the result of increased active nucleation site density. Kurihara
and Myers [25] proposed that the increased numbers of nucleation sites may enhance the
heat transfer by providing more convection heat transfer from increased bubble agitation

and/or increased latent heat transfer.

The difference of boiling heat transfer enhancement between the EDS and the
HFS can be interpreted as follows. As can be seen in Figure 2.5, the HFS is

characterized with re-entrant cavities, which are absent on the EDS (see Figure 2.6).

39

Griffith and Wallis [18] demonstrated that the geometry of the microcavity containing
trapped vapor was directly related to the bubble nucleation process. They found that the
re—entrant type cavities were stable, easily activated boiling sites. In this work, since
FC-72 is a highly—wetting fluid, the porous matrix of the EDS and the HFS was easily
flooded with FC-72. But the re-entrant cavities in the HFS trapped more
non-condensable gases than in the EDS which served as active nucleation Sites during

boiling incipience.

3.1.2 Subcooled Pool Boiling

When the bulk liquid temperature is less than the saturation temperature,
subcooled boiling can occur on the heated surface if the surface temperature is
sufficiently higher than the saturation temperature. Figure 3.2 shows the pool boiling
curves of the three different surfaces with 5K subcooling of the working fluid. The SSS
has the largest surface superheat for nucleate boiling (22.6K), which is larger than those
of the EDS (6.9K) and HFS (9.5K). In other words, compared to the SSS, the EDS
reduced the nucleate boiling superheat by 15.7 K (69%), while the HFS reduced it by
13.1 K (58%). This demonstrates that porous coated surfaces reduce the superheat of

boiling incipience.

The 5 K subcooled boiling experimental data Show that at the same heat flux
value (6.6 k/mz), the EDS increases the boiling heat transfer coefficient by 80%, while

the HF S increases it by 239% compared to the SSS.

From Figure 3.3, 3.4, and 3.5 it can be found that the incipience superheats in

40

Boiling curves of different surfaces in 5 K
subcooled FC-7 2

1o ,
,- ,° 9
P p p” A .
d 1’ ,’ M
0 'i‘ #4

El
0
O
.I
D
V

Heat flufx q" (W/m2)

!
,/ *
as" ,6 «n l
l 52/
5O l/
0.1
Surface superheat (K)

 

 

r
|

ifs—ssmcr A» $3 Decr 4 EDS Incr .
i o EDS Decr .1 HFSlncr a HFS Deer!

 

 

Figure 3.2 Boiling curves of different surfaces in 5 K subcooled FC-72 with 72 hours of
non-boiling immersion time.

41

subcooled boiling were higher than those of the saturated cases. This was anticipated
because in subcooled boiling as the bubbles departed from the heated surfaces, subcooled
liquid filled the spaces left by the departing bubbles. It would take some time for the

filling subcooled liquid to reach saturation conditions.

From Figure 3.3, 3.4, and 3.5 it can be seen that subcooling makes the boiling
curves move to the right, whether the surfaces are super smooth surface or porous coated

surfaces. Because the subcooled heat transfer coefficient is defined as [42]

sub = 5:? (3.1)

it decreases as the bulk liquid temperature Tb decreases.

Hui and Thome [20] studied the effect of liquid subcooling on the nucleate
boiling curve of benzene boiling on a vertical brass disk. Their results demonstrate that
even small levels of subcooling can substantially affect the heat transfer process and

significantly reduce the boiling heat transfer coefficient.

From the boiling curves of the SSS (Figure 3.3) and the EDS (Figure 3.4) it is
observed that the fully developed nucleate boiling curves of the subcooled cases
collapsed on the curves of the saturated cases. This observation is consistent with the
results of Rainey and You [34]. They tested the effect of subcooling on the pool boiling
of microporous surfaces in FC-72 with a liquid subcooling range of O (saturation) to 50 K.
Due to the restriction of the heater power, not enough data were collected to plot the fully

developed boiling curves of the HF S for the subcooled case (Figure 3.5) in this study.

42

1o 7
. “ti
/ ..
L 99 i
NA 1 j/ ”I
1 Cd! /
.V M / I
.U‘ 1 ****************** ~71 //-- . ‘
5% 1 g‘.’ y 1001
a /
Q) /
:r:
e f/ ‘
l
} l
'
0.1 , r
Surface superheat (K)

_ ; SAT Incr I: SAT Decr'4-.;susina -----a—— SUB Deer

Figure 3.3 Boiling curves of SSS in 5K subcooled and saturated FC-72 with 72 hours
of non-boiling immersion time.

43

Boiling curves of EDS

10
l 1/
44 .
l '0' o
a 4
.V 4:1,"; .
’o- 1 . ........ _- ,_,/ - » .. _. . , . -- .
B 1. 10 100-
E ; 15'" p‘ 9 ,
A ° [AI 9
1 6° ‘ o
0AM -____._._-l__ ._ __ _ ------- ‘.

Surface superheat (K)

 

 

l+ SAT Incr 44.»--. SAT Dari ~ 3337.167" 1) SUB—BEEF
\ . _ _____._--.__.__’c .___ . ..__’,.____.... 7.— . ____- L_.___. 1

Figure 3.4 Boiling curves of EDS in 5K subcooled and saturated FC-72 with 72 hours
of non-boiling immersion time.

44

Boiling curves of HF S
10T_-__--_--.-_-_--- _,------ ----- . -----
f 5'
4 ,r
,4' .4" .;
p O R
4/ a 4
(a Q a
K I
: f“ A c: I j
1 e1- ............ ------------------------- .
l x .. . ~
1 :3 1 10 100
* '74 4‘ a - fir:
l *5 .
‘ 0)
A D 2‘ .
A D A I
0.1 ......................... _ _______ ‘
Surface superheat (K)
T SATiherT A SE02...“ '7 'SUBTich o TSUB 08¢?
L- A _h A A A

Figure 3.5 Boiling curves of HF S in 5K subcooled and saturated FC-72 with 72

hours of non-boiling immersion time.

45

The boiling curves in Figure 3.2 Show that the subcooling does influence the
natural convection part and the initial nucleate boiling part of the boiling curve. It was
found that the subcooled pool boiling on porous coated surfaces is characterized by a
phenomenon of patch boiling, which can be seen from the boiling curves of increased
heat flux. Figure 3.6 shows the patch boiling of the HFS in 5 K subcooled FC-72.
This phenomenon can be explained by flooding of the porous matrix with liquid so that
only relatively small sites are available to act as active nucleation sites [8]. The
distribution of the active nucleate sites on the surface was random. After vapors were
generated from these active nucleate sites, they were quenched by the subcooled liquid in

the vicinity before it can spread over the surface and activated other sites.

3.2 EFFECTS OF NONBOILING IMMERSION TIME

Figure 3.7 shows the boiling curves of HFS in 5K subcooled FC-72 with a
non-boiling immersion time of 0 hour, 24 hours, 48 hours, and 72 hours. The
experimental data of 0 hour and 24 hours showed that the non-boiling immersion can

affect the temperature overshoot of incipient boiling, as is seen in Table 3.1.

As the non-boiling immersion time increasing, more and more microcavities in
the porous matrix, which served as the active nucleate sites, were flooded with the low
contact angle working liquid. When this happens, a very large wall superheat may be
needed to activate the nucleate sites. After the starting of nucleate boiling, the heat
transfer coefficient was dramatically increased due to the existence of phase change. So

a smaller wall superheat was needed to transfer the same amount of heat.

46

 

Figure 3.6 Patch boiling of HF S in 5 K subcooled F C-72 with 72 hours of non-boiling
immersion time.

Table 3.1 Effect of non-boiling immersion time on incipient superheat

and temperature overshoot.

 

 

 

 

 

. . . Surface superheat Surface superheat
Time ofnon-borlrng . . . . . . Temperature
immersion (Hours) before Incrprent after 1nc1prent overshoot (K)
boiling (K) boiling (K)
0 8.4 8.1 0.3
24 8.9 7.4 1.5
48 9.6 7.8 1.8
72 9.5 7.4 2.1

 

 

 

 

 

 

47

Boiling curves of HFS with different immerssion

time

10 -------
K ¢.l\ \A

ill/I I

Heat flux (1" (w/m 2)
c. -\&T ‘4‘
1:1

/
F 4
a; «tr-
0.1 ------- .
1 10 100 ’

Surface superheat (K)

e 72 hrs Incr L»- 72ThrsDecr A 46%. $ch A 48 hrs Dear
a l— 24 hrslncr 7 ct 24 hrsDecr f: 0 hrs Incr“ -0 — 0 hrs Decr

Figure 3.7 Boiling curves of HF S in 5K subcooled F C-72 with different non-boiling
immersion time.

48

But after immersion times of 24 hours and greater, the temperature overshoot
shows no dependence on the immersion time. This is in consistent with the conclusions
of Miller et al [29]. They believe that once the noncondensable were removed from the
cavities, the surface reached a steady condition in which the nucleation sites continue to

retain vapor embryos.

3.3 BOILING HY STERESIS PHENOMENON

The data presented in Figure 3.1 indicate that HFS is characterized by a large
temperature overshoot which leads to dramatic boiling curve hysteresis. The free
convection curve was followed typically until about AT=8.5K when there was an
explosive formation of vapor, which reduce AT to 2.1K. The HF S temperature
overshoot was 6.4 K, which was larger than that of the EDS (2.5 K). The SSS showed

no temperature overshoot in the saturated pool boiling case.

Due to the irregularities of various hysteresis curves, the use of temperature
difference is not enough to completely describe the hysteresis characteristics. Shi et a1
[38] introduced a parameter called hysteresis area, which is defined as the shaded area

enclosed by the normal boiling curve and the hysteresis curve, as shown in Figure 3.8.

Comparing the hysteresis areas in Figure 3.3, 3.4, and 3.5, it is obvious that the
HF S has the largest area, while the SSS has the smallest one. In other words, the boiling
hysteresis phenomenon is the most prominent on the HFS among the three surfaces. For
the same surface, the hysteresis area of subcooled boiling is larger than that in the

saturated boiling case. So it can be concluded that subcooling can deteriorate the

49

boiling hysteresis phenomenon.

 

 

 

 

Figure 3.8 Hysteresis area [38].

3.4 BOILING CURVES WITH WATER AS WORKING FLUID

Ngai [33] and Aitcheson [1] studied the nucleate boiling heat transfer
performance of these three surfaces with 10 K subcooled water as working fluid. They
used a cartridge heater with much higher power than the one used in this study. Figure

3.9 shows the boiling curves plotted using their data. From this figure it can be seen

50

Heat flux q" (W/m 2)

Boiling curves of different surfaces
in 10 K subcooled water

100000 - A V --,f- --._ ---AL --

1 10 100

Surface superheat (K)

 

 

[ T SSS —-—----~-EDS— L HFS“

Figure 3.9 Boiling curves of different surfaces in 10 K subcooled water.

51

that in the natural convection region, for a given superheat, the SSS has the smallest
convection heat transfer coefficient, the EDS has the medium one, while the HF S has the

largest one.

In Figure 3.9, the EDS shows no prominent heat transfer enhancement compared
to the SSS after incipience of boiling, whereas that of the HF S is obvious. There is no
temperature overshoot in the EDS curve, but there some small ones in the HF S curve.
This is probably due to different contact angles of water and F C-72. Since water has
large contact angle and low-wetting ability than FC-72, the liquid penetration into the
microcavities in the water test should be less than in the FC-72 test. The sizes of
microcavities in the HF S are lager than those in the EDS, so more potential active
nucleation Sites in the HFS are flooded with water than in the EDS. This interprets the

temperature overshoots in the HFS curve.

52

CHAPTER 4

SOCIAL AND ETHICAL CONSIDERATIONS

In this study, the fluid employed is FC-72, a highly-wetting dielectric
perfluorocarbon (PFC) with zero ozone-depletion potential (ODP). So using F C-72 as
working fluid will not destroy the ozone layer. This creates the potential for electronic
cooling application. F C-72 is low-solubility substance and has insignificant toxicity to
aquatic organisms. PFCS have high global warming potentials (GWP) so precautions
should be taken to prevent direct release of this substance to the environment. If FC-72
is exposed to extreme conditions of heat from misuse or equipment failure, toxic

decomposition products that include hydrogen fluoride and perfluoroisobutylene can

occur [47].

The porous-coated surfaces can be widely applied in boilers, heat exchangers,
heat pipes, and electronic cooling. Since these surfaces can dramatically enhance the
boiling heat transfer performance, this can decrease the energy consumption and bring

considerable social benefit.

53

CHAPTER 5

CONCLUSIONS

Three different surfaces, including the super-smooth surface (SSS), the High Flux
TM surface (HFS), and the electrochemical deposition surface (EDS) were tested in the
experimental study of pool boiling behavior in subcooled and saturated FC-72. The
effects of subcooling and non-boiling immersion time were investigated. Following are

the primary conclusions from this study:

1. Microporous coatings enhance nucleate boiling heat transfer due to their
significantly higher nucleation site density. Compared with the smooth surface, the
porous-coated surface can reduce the superheat of boiling incipience and increase the

boiling heat transfer coefficient.

2. Subcooling does not appear to influence the fully developed part of the boiling
curve comparing with the saturated boiling curve, but it does influence the natural
convection part and the initial nucleate boiling part of the boiling curve. The subcooled
pool boiling on porous coated surfaces is characterized by a phenomenon of patch boiling.

The subcooling makes the boiling curves move to the right.

3. The boiling hysteresis phenomenon is more prominent on porous-coated
surface than on the smooth surface, but subcooling is shown to deteriorate the boiling

hysteresis.

4. Non-boiling immersion was seen to increase the temperature overshoot of

incipient boiling. But after immersion times of 24 hours and greater, the temperature

54

overshoot shows no dependence on the immersion time.

5. The HFS enhanced the nucleate boiling heat transfer performance more
effectively than the EDS, but the EDS has smaller boiling incipient superheat and
Temperature Overshoot and Temperature Deviation hysteresis, which provides the EDS

potential for industrial application.

6. The contact angle of the working fluid was a significant influence on the

boiling hysteresis and the temperature overshoots.

55

CHAPTER 6

RECOMMENDATIONS

Suggestions for future work in the area of nucleate pool boiling from

porous-coated surfaces in FC-72 are as follows:

1. Rebuild the test section with a larger power cartridge heater and investigate the

CHF of porous-coated surfaces and compare it with that of the smooth surface.

2. Study the bubble dynamics of the nucleate pool boiling from porous-coated
surfaces with FC-72 as the working fluid, including the bubble departure diameter,

frequency etc to investigate the boiling enhancement mechanism more thoroughly.

3. Investigate the possible solutions to diminish boiling hysteresis phenomenon
on porous-coated surfaces in order to secure these surfaces to work more effectively in

the industrial application.

56

APPENDICES

57

APPENDIX A

STEADY STATE VERIFICATION

When increasing the heat flux by increasing the voltage input of the heater, a
steady state is needed before the acquisition of data. An experiment was conducted to
find the time interval necessary to reach the steady state.

The power of the cartridge heater was increased by increments of 10 Volts.
Temperature data of the surface were taken very minute. The data showed in the table
below and in Figure A.1 verified that when the power of cartridge heater was turned on,
about 12 minutes were needed to reach the steady state condition. After that, a

steady-state condition was usually reached 8 minutes after each voltage increase.

Table A.1 Change of surface temperature after each voltage increase

 

Voltage Time Temp. for TC #1
(V) min) (°C)
20.4 50.7
51.5
51.9
52.3
52.7

53

53.2
53.5
53.5
53.6
53.7
53.7
53.8
53.8
53.8
54.1
54.4
54.7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

25.4

 

 

 

 

 

 

asaaeaesaomwmmewwe

 

58

Table A.1 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

19 54.8
20 55
21 55.2
22 55.1
23 55.2
24 55.2
25 55.1
30.5 26 55.6
27 56
28 56.3
29 56.5
30 56.6
31 56.7
32 56.8
33 56.9
34 56.8
35 56.9
36 57
37 57
40.5 38 58
39 58.8
40 59.4
41 59.6
42 59.8
43 6O
44 6O
45 60
46 60.1
47 60.2
48 60.1
49 60.2
50 60.1
51 60.1
50.5 52 61.2
53 60
54 59.3
55 59.2
56 59.2
57 59.1
58 59.1
59 59.2
60 59.1
61 59.1
62 59.2
60.7 63 60
64 60.2

 

59

 

Temperature (°C)

Table A.1 (cont’d)

 

 

 

 

 

 

 

 

 

65 60.3
66 60.5
67 60.4
68 60.5
69 60.5
70 60.5
71 60.6

 

 

 

 

 

 

 

 

 

 

 

 

 

 

62
*
9'" i
60 WM - e“. |
x l
58 :e l
56 ‘4
J!"
.e
e. -
52 “f 1
O !
e
50
l
48 I I I T I I I I I I I I I T ‘
0 51015202530354045505560657075l

 

'.’ 20.4v i- 2.4-w A 30.5v x 4o.5v x 50.5% Sow

Figure A.1 Change of surface temperature after each voltage increase.

60

APPENDIX B

EXPERIMENTAL DATA

Table B.1 Experimental data of the SSS in 5K subcooled FC-72
with 72 hours of non-boiling immersion time.

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (W/m2) (°C) CC) (°C) co) (K)
20.6 0.192459799 55.6 55.7 55.7 51.7 4
30.8 0.430236271 58.4 58.4 58.4 51.4 7
40.4 0.740232788 61.2 61.3 61.3 51.2 10.1
50.6 1.161198914 65.4 65.4 65.4 51.1 14.3
60.7 1.671025086 69.5 69.6 69.6 51 18.6
70.1 2.228648736 73.8 73.8 73.8 51.2 22.6
80.8 2.960931151 76.4 76.4 76.4 53.3 23.1
80.8 2.960931151 75.2 75.3 75.3 53.3 22
90.5 3.714520383 77.2 77.2 77.2 54.5 22.7
100.5 4.580761821 78.6 78.7 78.7 55.4 23.3
110.6 5.547736706 79.8 79.8 79.8 56 23.8
120.7 6.60724069 80.8 80.9 80.9 56.6 24.3
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (W/mzl (°C) (°C) (°C) r0) (K)
120.7 6.60724069 80.8 80.9 80.9 56.6 24.3
110.7 5.557773315 79.5 79.5 79.5 56.4 23.1
100.5 4.580761821 78.3 78.3 78.3 56.3 22
90.4 3.706316034 76.9 77 77 56.2 20.8
80.5 2.938984856 74.9 74.9 74.9 56 18.9
70.3 2.241383846 72.2 72.2 72.2 55.8 16.4
60.9 1.682054924 69.1 69.1 69.1 54.8 14.3
50.8 1.170396494 64.7 64.8 64.8 52.9 11.9
40.6 0.747579966 60.6 60.6 60.6 51 9.6
30.4 0.419133867 57.7 57.6 57.6 49.8 7.8
20.8 0.196215024 54.2 54.2 54.2 48.6 5.6

 

 

 

 

 

 

 

61

 

Table B.2 Experimental data of the SSS in saturated FC-72

with 72 hours of non-boiling immersion time.

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) IWIm’) (°C) re) (°C) (°C) (K)
20.5 0.190595793 59.4 59.5 59.5 56.9 2.6
30.3 0416380943 61.3 61.3 61.3 56.6 4.7
40.5 0.743901842 64.2 64.2 64.2 56.2 8
50.4 1.152037618 67.8 67.9 67.9 56.1 11.8
60.7 1.671025086 71.5 71.5 71.5 56.1 15.4
70.9 2.279806869 74.6 74.7 74.7 56.5 18.2
80.3 2.924399345 75.9 75.9 75.9 57 18.9
80.3 2.924399345 75.2 75.2 75.2 57 18.2
90.5 3.714520383 76.4 76.5 76.5 57 19.5
100.6 4.5898823 78 78 78 57.2 20.8
110.9 5.577873745 79.2 79.3 79.3 57.3 22
120.2 6.552613056 80.7 80.7 80.7 57.3 23.4
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (WIm’) (°CL (°C) (°C) £0) (K)

120.2 6.552613056 80.7 80.7 80.7 57.3 23.4
110.8 5.567818995 79.4 79.4 79.4 57.3 22.1
101 4.626454923 78.1 78.1 78.1 57.3 20.8
90.8 3.739187856 76.8 76.9 76.9 57.2 19.7
80.6 2.946291217 75.4 75.4 75.4 57.1 18.3
70.8 2.273380355 73.1 73.1 73.1 57 16.1
60.3 1649074255 70 70 70 56.9 13.1
50.8 1.170396494 66.5 66.6 66.6 56.2 10.4
30.5 0.421895862 59.5 59.5 59.5 50.1 9.4
40.6 0747579966 62.5 62.6 62.6 54.9 7.7
20.5 0.190595793 56.7 56.7 56.7 49.2 7.5

 

 

 

 

 

 

 

62

 

Table B.3 Experimental data of the EDS in 5K subcooled F C-72
with 72 hours of non-boiling immersion time.

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (Wimi) (°C) (°C) (°C) (°C) (K)
20.3 0186894992 56.8 56.8 56.8 52.8 4
30.5 0421895862 58.5 58.5 58.5 52.5 6
40.6 0.747579966 59.1 59.2 59.2 52.3 6.9
40.6 0.747579966 58.1 58.1 58.1 52.3 5.8
50.8 1.170396494 59.2 59.2 59.2 52.2 7
60.5 1 .660031529 60.3 60.3 60.3 52.1 8.2
70.5 2.254155238 61.2 61.3 61.3 53.3 8
81 2.975607367 62.7 62.7 62.7 54.7 8
90.1 3681757409 64.2 64.2 64.2 55.4 8.8
100.9 4617298161 65.9 66 66 56 10
11 1 5.587937565 68 68 68 56.2 11.8
120.3 6.563520441 69.8 69.8 69.8 56.4 13.4
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T 8qu Superheat
(V) (Wm?) (°C) (°C) (°C) (°C) (K)
120.3 6.563520441 69.8 69.8 69.8 56.4 13.4
110.8 5.567818995 68 68 68 56.2 11.8
100.8 460815047 66.1 66.1 66.1 56 10.1
90.8 3.739187856 64.2 64.3 64.3 55.6 8.7
80.4 2.931687565 62.6 62.6 62.6 55.1 7.5
70.2 2.235011755 61.1 61.1 61.1 54.8 6.3
60.8 167653547 60.1 60.1 60.1 54.7 5.4
50.5 1.156613731 58.6 58.7 58.7 54.5 4.2
40.8 0754963427 57.8 57.8 57.8 54.3 3.5
31 0435841896 56.8 56.8 56.8 54.1 2.7
20.5 0190595793 55.1 55.1 55.1 53.1 2

 

 

 

 

 

 

 

63

 

Table 8.4 Experimental data of the EDS in saturated FC-72

with 72 hours of non-boiling immersion time.

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (WIm’) (°C) (°C) (°C) (°C) (K)
20.7 0194332876 58.1 58.1 58.1 55.6 2.5
30.7 0427447064 59.8 59.8 59.8 55.3 4.5
40.4 0740232788 61.1 61.2 61.2 55.3 5.9
40.4 0.740232788 58.7 58.7 58.7 55.3 3.4
50.4 1 . 152037618 59.4 59.4 59.4 55.2 4.2
60.7 1671025086 60.3 60.4 60.4 55.3 5.1
70.6 2.26055454 61.5 61.5 61.5 55.6 5.9
80.4 2.931687565 62.8 62.8 62.8 55.8 7
90.5 3.714520383 64.4 64.5 64.5 56.2 8.3
100.4 4.571650412 66.2 66.2 66.2 56.5 9.7
110.7 5.557773315 68 68 68 56.5 11.5
120.3 6.563520441 69.9 70 70 56.6 13.4
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
m sz) (°C) (°C) (°C) (°C) (K)

120.3 6.563520441 69.9 70 70 56.6 13.4
110.7 5.557773315 68.1 68.1 68.1 56.5 11.6
100.6 4.5898823 66.2 66.2 66.2 56.2 10
90.1 3.681757409 64.4 64.5 64.5 55.9 8.6
81.5 3.012456643 63 63 63 55.6 7.4
70.5 2.254155238 61.5 61.5 61.5 55.4 6.1
60.9 1682054924 60.2 60.2 60.2 55.2 5
50.6 1.161198914 58.9 59 59 55.1 3.9
40 0725647277 58 58 58 54.9 3.1
30.3 0416380943 56.9 57 57 54.6 2.4
20.8 0196215024 56.3 56.3 56.3 54.4 1.9

 

 

 

 

 

 

 

64

 

Table 8.5 Experimental data of the HFS in 5K subcooled F C-72
with 72 hours of non-boiling immersion time.

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (sz) 4°C) 4°C) (°C) (°C) (K)
20.8 0.196215024 54.7 54.7 54.7 51 3.7
30.3 0.416380943 57.9 58 58 50.7 7.3
40.5 0.743901842 60.1 60.1 60.1 50.6 9.5
40.5 0.743901842 58 58 58 50.6 7.4
50.6 1.161198914 58.5 58.5 58.5 50.5 8
60.7 1.671025086 59.4 59.4 59.4 50.5 8.9
70.3 2.241383846 59.9 60 60 50.5 9.5
80.8 2.960931151 60.8 60.8 60.8 50.9 9.9
90.6 3.722733804 61.6 61.6 61.6 53.5 8.1
100.9 4.617298161 62.3 62.3 62.3 55.5 6.8
110.1 5.49768972 62.9 63 63 55.7 7.3
120.2 6.552613056 63.9 63.9 63.9 56.8 7.1
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
M (WIm’) (°C) (°C) (°C) (°C) (K)
120.2 6.552613056 63.9 63.9 63.9 56.8 7.1
110.9 5.577873745 63.2 63.2 63.2 56.6 6.6
100.9 4.617298161 62.3 62.3 62.3 56.4 5.9
90.1 3.681757409 61.5 61.6 61.6 56.2 5.4
80.6 2.946291217 60.8 60.8 60.8 55.8 5
70.8 2.273380355 60.1 60.1 60.1 55.6 4.5
60.6 1.665523772 59.4 59.5 59.5 55.5 4
50.4 1.152037618 58.6 58.6 58.6 55.2 3.4
40.4 0.740232788 58 58 58 55.1 2.9
30.8 0430236271 57.1 57.1 57.1 54.6 2.5
20.7 0.194332876 54.6 54.6 54.6 52.4 2.2

 

 

 

 

 

 

 

65

 

Table 3.6 Experimental data of the HF S in saturated F 072

with 72 hours of non-boiling immersion time.

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (sz) (°C) (°C) (°C) (°C) (K)
20.8 0.196215024 59.9 59.9 59.9 56.7 3.2
30.4 0.419133867 62.4 62.4 62.4 56.3 6.1
40.7 0.751267162 64.6 64.7 64.7 56.2 8.5
40.7 0.751267162 58.4 58.4 58.4 56.2 2.2
50.7 1.165793169 58.7 58.7 58.7 56.2 2.5
60.6 1.665523772 59.1 59.1 59.1 56.3 2.8
70.6 2.26055454 59.6 59.7 59.7 56.4 3.3
80.6 2.946291217 60.2 60.2 60.2 56.6 3.6
90.5 3714520383 60.8 60.8 60.8 56.7 4.1
100.8 4.60815047 61.4 61.5 61.5 56.8 4.7
111.4 5.628283554 62.2 62.2 62.2 56.8 5.4
121 6.640126117 63.1 63.1 63.1 56.8 6.3
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (Wimz) (°C) (°C) (°C) (°C) (K)
121 6.640126117 63.1 63.1 63.1 56.8 6.3
110.9 5.577873745 62.3 62.3 62.3 56.8 5.5
100.7 4.59901185 61.5 61.5 61.5 56.8 4.7
90 3.673589342 60.7 60.7 60.7 56.7 4
80.4 2.931687565 60.1 60.1 60.1 56.5 3.6
70.6 2.26055454 59.4 59.5 59.5 56.3 3.2
60.4 1654548357 58.9 58.9 58.9 56.1 2.8
50.7 1.165793169 58.1 58.1 58.1 55.6 2.5
40.2 0.732921891 57.2 57.3 57.3 55.2 2.1
30.8 0.430236271 56.6 56.6 56.6 54.8 1.8
20.8 0.196215024 54.4 54.5 54.5 52.9 1.6

 

 

 

 

 

 

 

66

 

Table 3.7 Experimental data of the HFS in 5K subcooled FC-72
with 48 hours of non-boiling immersion time.

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) sz) (°C) (°C) (°C) (°C) (K)
20.7 0.194332876 54.8 54.8 54.8 50.7 4.1
30.4 0.419133867 58.2 58.3 58.3 50.4 7.9
40.9 0.758668764 59.8 59.8 59.8 50.2 9.6
40.9 0.758668764 58 58 58 50.2 7.8
50.4 1.152037618 58.6 58.7 58.7 49.9 8.8
60.5 1.660031529 59.7 59.7 59.7 49.7 10
70.5 2.254155238 59.9 59.9 59.9 49.5 10.4
80.2 2.917120196 60.5 60.6 60.6 49.4 11.2
90.3 3.698120755 61.3 61.3 61.3 50.4 10.9
100.6 4.5898823 61.9 62 62 53.6 8.4
110.5 5.537709168 62.6 62.6 62.6 55 7.6
120.8 6.618193429 63.6 63.6 63.6 56 7.6
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T 8qu Superheat
(V) (Win12) (°C) (°C) (°C) (°C) (K)
120.8 6.618193429 63.6 63.6 63.6 56 7.6
110.9 5.577873745 62.6 62.6 62.6 55.9 6.7
100.2 4.553454807 61.8 61.9 61.9 55.9 6
90.8 3.739187856 61.1 61.1 61.1 55.8 5.3
80.3 2.924399345 60.3 60.3 60.3 55.5 4.8
70.9 2.279806869 59.5 59.6 59.6 55.3 4.3
60.5 1.660031529 58.7 58.7 58.7 54.8 3.9
50.9 1.175008889 58.1 58.1 58.1 54.6 3.5
40.9 0.758668764 57.3 57.3 57.3 54.2 3.1
30.6 0.424666928 56.2 56.3 56.3 53.7 2.6
20.3 0.186894992 53.5 53.5 53.5 51.3 2.2

 

 

 

 

 

 

 

67

 

with 24 hours of non-boiling immersion time.

Table B.8 Experimental data of the HF S in 5K subcooled FC-72

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) (sz) (°C) (°C) (°C) (°C) (K)
20.3 0.186894992 54.9 54.9 54.9 51 3.9
30.9 0.433034548 58.2 58.2 58.2 50.6 7.6
40.7 0.751267162 59.5 59.5 59.5 50.6 8.9
40.7 0.751267162 57.9 58 58 50.6 7.4
50.8 1.170396494 58.8 58.8 58.8 50.4 8.4
60.5 1 .660031529 59.6 59.6 59.6 50.3 9.3
70.8 2.273380355 60.5 60.5 60.5 50.3 10.2
80.8 2.960931151 61.3 61.4 61.4 50.9 10.5
91 3.75567819 61.1 61.1 61.1 53 8.1
100.7 4.59901185 61.8 61.8 61.8 54.6 7.2
110.9 5.577873745 62.5 62.6 62.6 55.4 7.2
120.8 6.618193429 63.5 63.5 63.5 56 7.5
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
()0 MM (°C) (°C) (°C) (5) (K)

120.8 6.618193429 63.5 63.5 63.5 56 7.5
110.9 5.577873745 62.7 62.7 62.7 55.8 6.9
100.8 4.60815047 61.8 61.8 61.8 55.6 6.2
90.6 3.722733804 60.9 61 61 55.4 5.6
80.3 2.924399345 60.3 60.3 60.3 55.2 5.1
70.4 2.247765006 59.6 59.6 59.6 55 4.6
60.6 1.665523772 58.8 58.9 58.9 54.8 4.1
50.6 1.161198914 58.1 58.1 58.1 54.5 3.6
40.8 0.754963427 57.3 57.3 57.3 54.2 3.1
30.6 0.424666928 56.2 56.3 56.3 53.7 2.6
20.3 0.186894992 53.5 53.5 53.5 51.2 2.3

 

 

 

 

 

 

 

68

 

Table B.9 Experimental data of the HF S in 5K subcooled FC-72
with 0 hour of non-boiling immersion time.

 

Increased heat flux

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) sz) (°C) (°C) (°C) (°C) (K)
20.9 0.198106242 55.2 55.2 55.2 51.1 4.1
30.2 0.413637089 58.1 58.1 58.1 50.7 7.4
40.2 0732921891 58.9 58.9 58.9 50.5 8.4
40.2 0732921891 58.5 58.6 58.6 50.5 8.1
50.7 1.165793169 59.6 59.6 59.6 50.3 9.3
50.7 1.165793169 56.9 59 59 50.3 8.7
60.7 1.671025086 60.1 60.1 60.1 50 10.1
60.7 1 .671025086 59.6 59.6 59.6 50 9.6
70.6 2.26055454 60.4 60.5 60.5 49.9 10.6
80.7 2.953606648 60.6 60.6 60.6 51.2 9.4
80.7 2.953606648 60.1 60.1 60.1 51.2 8.9
90.6 3.722733804 60.7 60.8 60.8 53.7 7.1
100.9 4.617298161 61.7 61.7 61.7 54 7.7
110.3 5.517681303 62.4 62.4 62.4 55.1 7.3
120.4 6.574436898 63.1 63.2 63.2 55.6 7.6
Decreased heat flux
Voltage Heat Flux TC#1 TC#2 T Avg T Bulk Superheat
(V) sz) (°C) (°C) (°C) (°C) (K)
120.4 6.574436898 63.1 63.2 63.2 55.6 7.6
110.4 5.5276907 62.6 62.6 62.6 55.6 7
100.7 4.59901185 61.7 61.7 61.7 55.4 6.3
90.7 3730956294 60.8 60.8 60.8 55.2 5.6
80.5 2.938984856 60.1 60.1 60.1 55 5.1
70.2 2.235011755 59.3 59.4 59.4 54.8 4.6
60.3 1.649074255 58.7 58.7 58.7 54.7 4
50.2 1 .142912603 58 58 58 54.4 3.6
40.7 0.751267162 57.2 57.2 57.2 54.1 3.1
30.5 0.421895862 56.3 56.3 56.3 53.6 2.7
20.3 0.186894992 53.7 53.8 53.8 51.4 2.4

 

69

 

APPENDIX C

UNCERTAINTY ANALYSIS

C.1 Temperature Measurement Uncertainty
The last digit of reading from the OMEGA HH23 Microprocessor Thermometer
was 0.1 °C, thus the uncertainty in the temperature measurement was 0.05 °C.
The uncertainties in measurements of the length and width of the upper surface, a
and b, are
da=db=i0.5mm=i5x10‘4m (CI)
The uncertainties in measurements of the voltage input U and the resistance of the
cartridge heater R are
dU = i0.05V (C.2)

dR = i005!) (C3)

C2 Heat Flux Uncertainty
Heat flux was calculated using the power input of the cartridge heater divided by

the upper surface area of the copper block.
E
"= — C4
9 A ( )

where A is the area of the copper block upper surface.

The electric power generated by the cartridge heater

E = ]? (C5)

The uncertainty of E becomes

70

dE= l-—UdU: l—dR

R +R
—E[— 2dU +dR]
U R
So it gives
£=_°dv.d_°° (8.7)
E U R
Similarly, the uncertainty of A is
44:41.22 (8..)
A a b

In the calculation of the heat flux, it was assumed that all heat generated by the
cartridge heater transferred through the upper surface of the copper block, where the test
surface was soldered. Although Teflon has a good insulation effect, there was heat loss
from the sides and bottom of the test section. This is a kind of systematic error.

Assume the inside temperature of Teflon sleeves Tin equals to the copper block
temperature Tin and the outside temperature of Teflon sleeves Tom equals to the
temperature of bulk liquid Tbulk. The heat losses by conduction through Teflon sleeves

are

T T.

61);... = ‘k ”—1";— (C9)

where k is the thermal conductivity of Teflon in W/m K and L is the thickness of Teflon
sleeves in m. Since k is a function of temperature, here it is approximated with the

value under the average temperature of Tin and Tom. Then the systematic error is

71

qloss Atom!

E , where Am. is the surface area of the copper block which is covered by the

Teflon sleeves.
Hence, the heat flux uncertainty is

dq " dE M q lbss Atom!
.. = + + ((3.10)
q E A E

 

 

 

Table C] shows the uncertainties of heat flux for the surfaces under different
experimental conditions. It can be found that the heat flux uncertainty ranges from

3.26% for the high heat flux value to 37.55% for the low heat flux value.

72

Table C.l Heat flux uncertainty for different surfaces.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . Non-boiling
Surface Type 35:25:33,? Heat Flux Immersion V023)” Uncertainty
Time (Hour)
Increased 72 20.6 0.21 19694
Increased 72 120.7 0.0579482
I
SSS 5 K Subcoo ed Decreased 72 120.7 0.0579482
Decreased 72 20.8 0.2789861
Increased 72 20.3 0.218067
Increased 72 120.3 0.0426307
EDS 5 K bcool
8" 9d Decreased 72 120.3 0.0426307
Decreased 72 20.5 0.1210789
Increased 72 20.8 0.1945056
Increased 72 120.2 0.0338765
HFS 5 K bcool d
S” e Decreased 72 120.2 0.0338765
Decreased 72 20. 7 0.1281979
Increased 72 20. 5 0.1 502643
72 1 .2 0.0569913
sss Saturated '"C'eased 2°
Decreased 72 120.2 0.0569913
Decreased 72 20.5 0.3755165
Increased 72 20.7 0.142856
EDS Saturated Increased 72 120. 3 0.0426427
Decreased 72 120.3 0.0426427
Decreased 72 20.8 0.1141195
Increased 72 20.8 0.1741235
I 72 1 0.0326491
HFS Saturated ncreased 12
Decreased 72 121 0.0326491
Decreased 72 20.8 0.1002312
Increased 48 20.7 0.214255
HFS 5 K Su I ed Increased 48 120.8 0.0344422
Decreased 48 120.8 0.0344422
Decreased 48 20. 3 0.1319857
Increased 24 20.3 0.2124147
| ased 24 120.8 0. 0 7
HFS 5 K Subcooled me 0343 3
Decreased 24 120.8 0.0343343
Decreased 24 20.3 0.1366736
Increased 0 20.9 0.2108578
d 120.4 .
HF S 5 K Subcool ed Increase 0 0 0345028
Decreased 0 120.4 0.0345028
Decreased 0 20.3 0. 1414794

 

 

 

 

 

 

73

 

BIBLIOGRAPHY

74

BIBLIOGRAPHY

. Aitcheson, N. L. 2003. New surfaces for the control of boiling heaflraasfer: site
formation at incipience. MS. Thesis, Michigan State University.

. Andrianov, A. B., Borzenko, V. I., and Malyshenko, S. P. 1999. Elementary processes
at boiling on the surfaces with porous coatings. Russian Journal of Engineering

Thermoghysics, 9(1-2), pp. 5 1-67.

. Bajorek, S. M. 1988. An experimental and theoretical investigation of mtflicomponent
pool boiling on smooth and finned surfaces. Ph. D. Dissertation, Michigan State
University.

. Bankoff, S. G. 1959. The prediction of surface temperature of incipient boiling.
Chemical Engineering Progress Syr_nmsium Series, 55, No. 29, pp.87-94.

. Benjamin, J. E. and Westwater, J. W. .1961. Bubble growth in nucleate boiling in a
binary mixture. International Development of Heat Transfer, pp. 212-218.

. Berenson, P. J. 1960. Transition boiling heat transfer from ficrizontal surfag. Sc. D.
Dissertation, Massachusetts Institute of Technology.

. Berenson, P. J. 1962. Experiments on pool boiling heat transfer. International Journal
of Heat and Mass Transfer, 5, pp. 985-999.

. Bergles, A. E. and Chyu, M. C. 1982. Characteristics of nucleate pool boiling from
porous metallic coatings, ASME Journal; of Heat Transfer, 104, pp. 279-285.

. Chang, J. Y. and You, S. M. 1997. Boiling heat transfer phenomena from micro-porous
and porous surfaces in saturated FC-72. International J oumal of Heat and Mass
Transfer 40 (18), pp. 4437-4447.

 

10. Chien, L. and Webb, R. L. 1998. A parametric study of nucleate boiling on structured
surfaces. Part I: Effect of tunnel dimensions. ASME Journfi of Heat Transfer. 120, pp.
1042-1048.

11. Chien, L. and Webb, R. L. 1998. A parametric study of nucleate boiling on structured
surfaces. Part II: Effect of pore diameter and pore pitch. ASME Journal of Heat
Transfer 120, pp. 1049-1054.

 

75

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Chien, L. and Webb, R. L. 1998. Measurement of bubble dynamics on an enhanced
boiling surface. Experimental Thermal and Fluid Science 16, pp. 177-186.

 

Chien, L. and Webb, R. L. 2001. Effect of geometry and fluid property parameters on
performance of tunnel and pore enhanced boiling surfaces. Experimental Thermal and
Fluid Science, 8, pp. 329-339.

Cieslinski, J. T. 2002. Nucleate pool boiling on porous metallic coatings.
Experimental Thermal and Fluid Science 25, pp. 557—564.

 

Clark, H. B., Strengewater, P. S. and Westwater, J. W. 1959. Active sites for nucleate
boiling. Chemicaflgineering Progess Symposium Series, 55 (29), pp. 103-110.

Corty, C. and Foust, A. S. 1955. Surface variables in nucleate boiling. Chemical
Engineering Progress Symposium Series, 51, No.16, 1-12.

Demiray, F. and Kim, J. 2004. Microscale heat transfer measurements during pool
boiling of FC-72: effect of subcooling. International JouLnal of Heat and Mass
Transfer 47, pp. 3257-3268.

 

Griffith, P. and Wallis, J. D. 1960. The role of surface conditions in nucleate boiling.
Chemical Engineering Progress Symposium Series 56, No.49, pp. 49-63.

 

Hsu, Y. Y. and Graham, R. W. 1976. Transport processes in boiling and two-phase
system. Washington, DC: Hemisphere.

Hui, T. O. and Thome, J. R. 1985. A study of binary mixture boiling: Boiling site
density and subcooled heat transfer. International Journal of Heat aid Mass Transfer,
28, pp.919-928.

Incropera, F. P. and DeWitt, D. P. 1981. Fundamentals of heap transfer. John Wiley &
Sons

Jiang, Y.Y., Wang, W.C., Wang, D. and Wang, B.X. 2001. Boiling heat transfer on
machined porous surfaces with structural optimization. International Journal of Heat
and Mass Transfer 44. 443-456.

 

Kedzierski, M. A. 1995. Calorimetn'c and visual measurements of R123 pool boiling
on four enhanced surfaces. NISTIR 5732. US. Department of Energy, Washington,
DC.

76

24. Kim, J. H., Rainey, K. N., You, S. M. and Pak, J. Y. 2002. Mechanism of nucleate
boiling heat transfer enhancement from microporous surfaces in saturated F 072.
ASME Journal of Heat Transfer 124, pp. 500-506.

 

25. Kurihara, H. M. and Meyers, J. E. 1960. The effects of superheat and surface
roughness on boiling coefficients. AIChE J. 6(1), pp. 83-91.

 

26. Leontiev, A. 1., Lloyd, J. R., Malyshenko, S. P., Borzenko, V. I., Dunikov, D. O.
Eronin, A. A. and Nazarova, O. V. 2003. New effects in interfacial heat and mass
transfer of boiling and evaporation in micro-scale porous materials. 2003 ASME
IMECE Proceedings, Washington DC, November 2003.

27. Lloyd, J. R. 1971. Laminar, t_ransition, and turbulent nat1_1ral convection adiacent to
vertical and uvaard-facing inclined surfagea. Ph. D. Thesis, University of Minnesota.

28. Marto, P. J. and Lapere, Lt. V. J. 1982. Pool boiling heat transfer from enhanced
surfaces to dielectric fluids, ASME Journal of Heat Transfer 104, pp. 292-299.

 

29. Miller, J. W., Gebhart, B. and Wright, N. T. 1990. Effects of boiling history on a
microconfigured surface in a dielectric liquid. International Communications of Heat
and Mass Transfer 17, pp. 389-398.

 

30. Milton, R. M. 1968. Heat exchange system. US. Patent 3,3 84,154. May 2.

31. Myers, J. E. and Katza, D. L. 1953. Boiling coefficients outside horizontal tubes.
Chemical Engineering Progress Symposiw Series. 49, No.5, pp. 107-114.

32. Nakayama, W., Daikoku, T. and Nakajima, T. 1982. Effects of pore diameters and
system pressure on saturated pool nucleate boiling heat transfer from porous surfaces.
ASME Journal of Heat Transfer 104, pp. 286—291.

 

33. Ngai, 8., Lloyd, J. R., Leontiev, A. I. and Malyshenko S. P. 2004. Creation of nano-
scaled surface structure for the enhancement of boiling heat transfer. 2004 ASME
IMECE Proceeding; Anaheim, Califomia, November 2004.

34. Rainey, K. N., You, S. M. and Lee, S. 2003. Effect of pressure, subcooling, and
dissolved gas on pool boiling heat transfer from microporous surfaces in FC-72.
ASME Journal of Heat Transfer 125, pp. 75-83.

 

77

35. Ramaswamy, C., Joshi, Y., Nakayama, W. and Johnson, W. B. 2003. Effects of
varying geometrical parameters on boiling from microfabricated enhanced structures.
ASME Journal of Heat Transfer 125, pp. 103-109.

 

36. Ronsenow, W. M. 1982. Pool boiling. Handboon Multiplaase Systems, McGraw-
Hill-Hemisphere, New York.

37. Shakir, S. 1987. Boiling Incipience and Heat Transfer on Smooth and Enhfled
Surfaces. Ph. D. Dissertation, Michigan State University.

38. Shi, M.H., Ma, J. and Wang, B.X. 1993. Analysis on hysteresis in nucleate pool
boiling heat transfer. International Journal of Heat and Mass Transfer, 36 (18), pp.
4461-4466.

39. Snyder, T. J., Chung, J. N. and Schneider, J. B. 1998. Competing effects of
dielectrophoresis and buoyancy on nucleate boiling and an analogy with variable
gravity boiling results. ASME Journal of Heat Transfer, 120, pp. 371-379.

40. Still, W. J. 1929. Heat transmitting tube. U. S. Patent 1,716,743, June 11.

41. Tehver, J ., Sui, H. and Temkina, V. 1992. Heat transfer and hysteresis phenomena in
boiling on porous plasma-sprayed surface. Experimental Thermal and Fluid Science.
5. pp.7l4-727.

42. Thome, J. R. 1990. Enhanced Boiling Heat Transfer. Hemisphere Publishing Corp.

43. Thome, J. R. 1992. Mechanisms of enhanced nucleate pool boiling. Proceedings of
the Engineering Founcgtion Conference on P001 and External Flow Boiling, ASME,

New York, pp. 337-343.

44. Vorob'ev, V.S. and Malyshenko, S.P. 2002. Nucleus formation in polarized dielectric
media. Experimental Thermal and Fluid Science 26, pp. 833-839.

 

45. Webber, W. O. 1960. Under fouling conditions-Pinned tubes can save money.
Chemical Engineering, March, pp. 149-152.

46. Zhang, H. J ., Li, L. J. and Tong, Q. M.1999. Boiling heat transfer from mechanically
fabricated porous tubes at atmospheric and super atmospheric pressures. Heat
Transfer-Asian Research. 28 (8), pp. 640-648.

78

47. 3M Material Safety Dy Sheet FC-72 Flporinert Brand Electronic Liquid.
09/05/2002.

79

‘1'!» l l
I. " ‘l‘l

lllfqlf'l

   

1113111111111)131:1