Thermosyphons And Heat Pipes Theory And

Applications Marcia Barbosa Henriques Mantelli

download

https://ebookbell.com/product/thermosyphons-and-heat-pipes-
theory-and-applications-marcia-barbosa-henriques-
mantelli-22147344

Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.

Thermosyphons And Heat Pipes Theory And Applications Marcia Barbosa

Henriques Mantelli

https://ebookbell.com/product/thermosyphons-and-heat-pipes-theory-and-
applications-marcia-barbosa-henriques-mantelli-22147344
Marcia Barbosa Henriques Mantelli

Thermosyphons
and Heat Pipes:
Theory and
Applications
Thermosyphons and Heat Pipes: Theory
and Applications
Marcia Barbosa Henriques Mantelli

Thermosyphons and Heat

Pipes: Theory
and Applications
Marcia Barbosa Henriques Mantelli
Department of Mechanical Engineering
Federal University of Santa Catarina
Florianópolis, Santa Catarina, Brazil

ISBN 978-3-030-62772-0 ISBN 978-3-030-62773-7 (eBook)

https://doi.org/10.1007/978-3-030-62773-7

© Springer Nature Switzerland AG 2021

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
I dedicate this book to my most complex and
unfinished experiment, from which I learn
every day: my family.
To you, Sylvio (my husband), Pedro and Fabio
(my sons), Edi (my mother) and especially to
Fernando (my father, in memoriam), who, as
engineer, inspired me, my brother and sister
to follow the intriguing engineering paths.
To God.
Acknowledgments

Representing the Federal University of Santa Catarina, I thank Prof. Sergio Colle,
who invited me to the adventure of starting a new laboratory. I am really proud to be
a member of the Mechanical Engineering Department of UFSC, Brazil, that hosted
my technical visions, allowing me to build a fruitful laboratory.
Thanks to National Institute for Space Research (INPE) and Brazilian Space
Agency, which lead my first steps to the heat pipe technology development field.
Special thanks to the Brazilian Industry, especially to Petrobras, Brazilian Oil
Company, that trusted our research group, proposing us to be partners of several
development projects, especially in the heat pipe industrial application field. These
many projects along the years allowed us to get the facilities and knowledge we need.
My special thanks to you, my students, who, along the years, helped in the
construction of the knowledge presented in this book. Each one of you is the co-
author of a piece of this work. Special thanks to Luis R. Cisterna and João Victor C.
Batista, who helped in formatting models and building graphics for this book, being
happily available any time of the day, including sleep hours and weekends. My very
special thanks also goes to my student Gabriel S. C. Vieira and Larissa Krambeck,
for proposing and solving some of the Chap. 6 examples. Thanks to Prof. Fernando
H. Milanese and Priscila Gonçalves, who reviewed parts of this book. Also thanks
to the Heat Pipe Laboratory (LABTUCAL) staff, here represented by Dra. Kenia
W. Milanez (manager) and Dr. Juan P. F. Mera (researcher), Leandro da Silva and
Charles Nuernberg (technicians) and Andreza Souza Silva (secretary), who kept the
laboratory running smoothly, so that I could dedicate hours to this book.
To all students that worked in the laboratory along these many years, too many to
mention, who had very important contributions, my deep thanks.
Above all, I thank God for the family, health, intelligence and strength that I am
blessed with.

vii
Contents

1 Introduction to Thermosyphons and Heat Pipes . . . . . . . . . . . . . . . . . . 1

1.1 Working Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Book Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Part I Fundamentals and Fabrication Aspects

2 Physical Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1 Nusselt Model for Wall Condensation . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Liquid–Vapor Phase Change Model . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3 Fluid Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.4 Boiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.5 Closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3 Thermal and Hydraulic Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.1 Thermodynamic Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.2 The Effective Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.3 Liquid Phase Pressure Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.4 Vapor-Phase Pressure Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.5 Vapor Pressure Drop Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.6 Gravity Pressures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.7 Liquid and Vapor Pressure Distribution Curves . . . . . . . . . . . . . . . 71
3.8 Operational Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.9 Geyser Boiling Phenomena in Thermosyphons . . . . . . . . . . . . . . . 95
3.10 Closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
4 Design of Thermosyphons and Heat Pipes . . . . . . . . . . . . . . . . . . . . . . . 111
4.1 Thermosyphons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
4.2 Heat Pipes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
4.3 Heat Pipe Selection of Design Parameters . . . . . . . . . . . . . . . . . . . . 125
4.4 Design Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

ix
x Contents

4.5 Closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5 Fabrication and Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.1 Fabrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.2 Cleaning Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.3 Out-Gassing of Tube Materials and Working Fluids . . . . . . . . . . . 141
5.4 Charging Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
5.5 Sodium Charging—High Temperature Devices . . . . . . . . . . . . . . . 145
5.6 Thermal Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
5.7 Closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
6 Application of Models to Selected Cases . . . . . . . . . . . . . . . . . . . . . . . . . 155
6.1 Ordinary Geometry Thermosyphon . . . . . . . . . . . . . . . . . . . . . . . . . 155
6.2 Small Diameter Thermosyphon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
6.3 Water–Copper Heat Pipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
6.4 Ethanol–Copper Heat Pipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
6.5 Elliptical Cross Section Water-Cooper Heat Pipe . . . . . . . . . . . . . . 186
6.6 Water–Copper Grooved Heat Pipe . . . . . . . . . . . . . . . . . . . . . . . . . . 188
6.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194

Part II Special Devices

7 Classification According to Operational Principles . . . . . . . . . . . . . . . 197
7.1 Loop Thermosyphons—LTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
7.2 Capillary Pumped Loops—CPL . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
7.3 Loop Heat Pipes—LHP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
7.4 Vapor Chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
7.5 Pulsating Heat Pipes (PHP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
7.6 Mini Two-Phase Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
7.7 Closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
8 Classification According to Operational Temperature . . . . . . . . . . . . . 291
8.1 Cryogenic Heat Pipes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
8.2 Intermediate Temperatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
8.3 High Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
8.4 Closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336

Part III Applications

9 Thermosyphon Heat Exchangers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
9.1 Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
9.2 Thermosyphon Heat Exchanger Design Methodology . . . . . . . . . 345
Contents xi

9.3 Thermosyphon Heat Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 346

9.4 External Convection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
9.5 Flow Pressure Drops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350
9.6 Selection of Tube Geometries and Materials . . . . . . . . . . . . . . . . . . 352
9.7 Design Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
9.8 Special Configurations of Thermosyphon Heat Exchangers . . . . . 358
9.9 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
10 Electronics Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
10.1 Electronic Cabinets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
10.2 Electronics Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
10.3 Vapor Chambers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
10.4 Avionics Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
10.5 Closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
11 Other Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
11.1 Cooling Towers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
11.2 House Solar Heating System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
11.3 Ovens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
11.4 Oil Tank Heaters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410
11.5 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416
Nomenclature

Roman Letter Symbols

a, b, c, d Position on the liquid column, plate dimension

  [m]
A Area, disjoining pressure model constant m2
b Width [m]
B Disjoining pressure model constant
BiC Modified Biot number
Bo Bond number  
2
c Sound velocity  m/s 
cp Specific heat J/kg · K
C Dimensionless friction, constant, conductance param-
eter, correction factor
Ca Adjustment parameter for vapor velocity profile
Cf Adjustment parameter for flooding limit
C0 Vapor distribution coefficient for cylindrical tubes
Co Dimensionless confinement number
D Diameter [m]
D+ Non-dimensional diameter
E Free Helmhotlz energy [J ], inernal energy [J]
f Parameter, Fanning (friction) factor, correction factor
F Force [N],
 view  factor, filling ratio
g Gravity m/s2
Gn Geometry non-dimensional parameter
Gr Grashof number
Gr c Corrected Grashof number  
h Convection heat transfer coefficient W/m2 K , groove
height [m]
hlv , hvl Latent
  heat, liquid-vapor and vapor-liquid, respectively
J/kg
H Height [m], hydraulic head [m]

xiii
xiv Nomenclature

I0 , I1 Modified Bessel functions of the first kind, order 0 and

1
J Chilton-Colburn J analogy factor
Ja Jakob number
J0 , J1 Bessel functions of the first kind, order 0 and 1
k Thermal conductivity [W/m · K ]
kb Boltzmann constant
  [J/K]
K Permeability m2
Ka , Kf Kinetic energy constant
Kl Bubble depart parameter
K loss Localized Pressure drop coefficients
Kn Knudsen number
Kp Dimensionless pressure parameter
KT Dimensionless boiling parameter
K0 e K1 Modified Bessel functions of the second kind, order 0
and 1
L Turbulence scale [m]
l Length [m], characteristic length
Lm , Mm , Nm , Om , Pm Qm Superposing solutions for the separation of variables
method
m Mass [kg], adjustable
  constant, fin convection
ṁ Mass flow rate kg/s
 
ṁ" Mass flux kg/m2 s
M Mach number, molar mass [g/mol], fin parameter
n Moles number, adjustable constant, number
N Mesh number [1/m], moles number, number
Nt Merit number for thermosyphons [kg/s5/2 ·K3/4 ]
Nu Nusselt number
 
p Pressure N/m2
P Perimeter [m]
pr Reduced pressure
Pr Prandtl number
Pt Diagonal pitch of tube arrangements [m]
q Heat rate [W]
q’ Heat rate per
 unit length [W/m]
q” Heat flux W/m2
 
q̇ Volume generated heat W/m3
r Distance [m], radius [m]
R Thermal
 resistance [°C/W], universal constant of the gas
J/kg · K
R’ Thermal resistance per unit length [°C/W·m]
Ra Rayleigh number
Re Reynolds number
Rp Surface roughness [μm]
Nomenclature xv

s Pitch [m]
S Crimp factor
t Time [s], thickness [m]
T Temperature [K, ◦C]
u Velocity in the axial direction [m/s]
U Global coefficient of heat transfer [W/m2 ·K]
v Velocity in the radial direction [m/s]
V Vapor mass flux velocity [m/s], volume [m3 ]
V̇ Volume rate [m3 /s]
V+ Non-dimensional vapor drift velocity
w Width [m], eddy velocities [m/s]
We Weber number
x Body force term [N]
x,y Cartesian axis, variable neck radius [m]
X Vapor quality
Xl, Xt Longitudinal and transversal pitch for tube arrangement
[m]
z Axial direction, height of the sintering neck [m]

Greek Letter Symbols

α Void fraction, thermal diffusivity [m2 /s], groove angle

β Momentum correction factor, coefficient of thermal expan-
sion [1/K], thermal coefficient
βe , χe , ςe δe ϕe νe ξe Evaporator geometry dimensions [m]
β1 , β2 Half-angle of the wire-liquid opening and meniscus curva-
ture
 Fin parameter
 v,air Mass diffusion coefficient of vapor in air [m2 /s]
γ Ratio of specific heats, phase change Arckmann factor
δ Film thickness [m]
ε Area porosity, emissivity, effectiveness
ε’ Volumetric porosity
εr Surface roughness
, r0 Molecule potential force parameters
θ Inclination angle, solid-liquid contact angle, temperature
difference
κ Compressibility (1/Pa)
λ Laplace length scale, mean free path
λm , ζm , ηm , υn , τn , m Eigenvalues  
μ Dynamic viscosity (N · s)/m2
η Efficiency  
ν Cinematic viscosity m2 /s
xvi Nomenclature

νe Evaporator pumping region length [m]

ω Mass fraction at the liquid-vapor interface, absolute
humidity
ξ Boiling correlation parameter, variable thickness fin param-
eter  
ρ Density kg/m3
σ Surface tension [N/m], Stefan-Boltzmann constant
σd Collision distance between molecules [ Ȧ]
Mass diffusion
 coefficient

τ Shear stress N/m2
υ Specific volume [m3 /kg]
ϕb Dimensionless bubble release number
ϕc Dimensionless fin condenser parameter
ϕe Evaporator pumping region insulated length [m]
ϕT Modified dimensionless fin condenser parameter
φ Two-phase multiplier, eigenvalue function
(r ) Molecular potential function
χ Pool parameter
Fluid properties parameter, pressure difference
m Percentage of the evaporator heat used in phase-change
 Martinelli parameter
m Mass transfer coeficiente [m/s]

Subscripts and Superscripts

a Adiabatic section, advancing

act Active
air Air
amb Ambient
atm Atmosphere
ax Axial
b Bubble, boiling, base, bulk
c Condenser, condensation, casing, cold
C Characteristic length, characteristic length of condenser
cap Capillary
cc Compensation chamber
cd Condensed
ch Chamber, channel
col Column
cool Cooling
cond Conduction
conv Convection
cont Continuum
Nomenclature xvii

cr Critical
crit Critical
cs Cold source
cw Cooler-wall interface
ct Cooling tower, continuum
d Disjoining, dynamic
D Diameter
db Dry bulb
e Evaporator, entrainment
ef Effective
eq Equivalent
ev Evaporated
ex External
exp Experimental
f Flooding, fin, film
g Gravity, groove
GB Geyser Boiling
h Heater, hydraulic, hot
hb Humid bulb
hem Hemisphere
hs Heat sink
i Interface, ith component, internal, initial
in In
inf Inferior
k Kinetic
l Liquid
L Length
lat Latent
lc Liquid channel
Le Evaporator length
ll Liquid line
lm Logaritmic mean
loss Loss
lv Liquid-vapor
m Mean
ṁL Condensate film
max Maximum
min Minimum
mix Mixture
n Index, nucleation, neck
N Normal direction, natural convection
NCG Non condensable gas
Nu Nusselt
o Overall
op Operational
xviii Nomenclature

out Out
p Pore, pool, pressure, plug, plate
pc Phase change
pm Porous media
r Radial, rescinding
s Surface, solid, sphere, sonic
sat Saturation
st Stagnation point
su start up
sup Superior
t Total, top
tr Transversal, transition
v Vapor, viscous
vl Vapor line
w Wet, wick, wall, wire, water
wf Working fluid
x, y, z Local, direction
X Body force term
β Wick plug
δ Condensate film thickness
σ Surface tension

Numerical

0 (x = 0), (t = 0)
2φ Two-phase flow
∞ Environment, ambient
Chapter 1
Introduction to Thermosyphons
and Heat Pipes

Heat pipes and thermosyphons are highly efficient heat transfer devices that use
two-phase cycles of fluids as the operating principle. Typical heat pipes and/or ther-
mosyphons consist of an evacuated tube casing, within which a controlled amount
of a working fluid is introduced. In heat pipes, capillary forces provided by a
wick, are responsible for the movement of the fluid through the device. In ther-
mosyphons, gravity forces do this job. Although heat pipes and thermosyphons may
have many different geometries and configurations, they are basically composed of
three main regions: evaporator, adiabatic section and condenser. In some applications,
the adiabatic section may not be present.

1.1 Working Principles

Figure 1.1 illustrates the physical principles that drive the thermosyphon operation.
Heat is delivered to the thermosyphon in the evaporator section, reaching the working
fluid, causing liquid–vapor phase change. The generated vapor, due to pressure gradi-
ents inside the thermosyphon, crosses the adiabatic region and reaches the condenser,
where heat is removed. The vapor contained in the condenser region condenses and
the resulting liquid returns to the evaporator by the action of gravity forces. There-
fore, the evaporator must be located in a position inferior to the condenser. Due to
the surface tension effects between the tube material and the liquid, small rivulets
are formed over the tube internal wall.
Heat pipes operates in very similar way, however, the return of the working fluid
from the condenser to the evaporator happens due to the capillary forces resulting
from the presence of a wick structure located inside the device, as illustrated in
Fig. 1.2.
Thermosyphons and heat pipes can also be designed to operate in loops. In this
case, internally smooth tubes connect evaporators to condensers, within which only
vapor or liquid flows. This arrangement avoids the dragging forces between liquid

© Springer Nature Switzerland AG 2021 1

M. B. H. Mantelli et al., Thermosyphons and Heat Pipes: Theory and Applications,
https://doi.org/10.1007/978-3-030-62773-7_1
2 1 Introduction to Thermosyphons and Heat Pipes

Fig. 1.1 Working physical principles of thermosyphons

Fig. 1.2 Working physical principles of heat pipes

1.1 Working Principles 3

and vapor due to countercurrent flows. Schematics of loop thermosyphons and loop
heat pipes are presented in Fig. 1.3.
In a general sense, the casing and the working fluid are considered the main
components of thermosyphons. In heat pipes, besides these two, the wick is another
key element. Usually located in the inner tube wall, the wick structure is respon-
sible for the capillary forces that pump the working fluid from the condenser to the
evaporator.
Casings can be made of metal, ceramic or other material. Besides, several different
liquids can be used as working fluids (liquid nitrogen, water, alcohol, naphthalene,
liquid sodium, etc.). Basically, the selection of these materials depends on the device

Fig. 1.3 Loop thermosyphons (a) and loop heat pipes (b)
4 1 Introduction to Thermosyphons and Heat Pipes

working temperature and its application. The wick, in heat pipes, can be made of
different porous materials such as metal screens, sintered metal powder, longitudinal
grooves, corrugated fiber glass, etc.
Usually, the design of thermosyphons or heat pipes starts with the selection of the
working fluid, which must operate at the required temperature level. After the working
fluid is selected, the casing is chosen. The casing material should be chemically
compatible with the working fluid, avoiding the formation of non-condensable gases
that would block part of the device, decreasing its performance. For heat pipes, the
wick structure must provide the necessary liquid pumping capacity and be compatible
with the working fluid and with the casing material.
Among the several configurations of thermosyphons or heat pipes (loop ther-
mosyphon, loop heat pipe, pulsating heat pipe, vapor chamber, etc.) the engineer must
select the one more suitable for the application: the device must have the expected
thermal performance and be geometrically adaptable. These two-phase technologies
are actually versatile, they can operate at basically any temperature level and they
can assume multiple shapes.
Design procedures and models are presented in details in this book.

1.2 Applications

Thermosyphon and heat pipes can be used in applications where heat needs to be
transferred with high efficiency or where uniform temperatures are required. These
devices are able to manage heat in a myriad of equipment, from very small, such as
electronic components, up to very large ones, such as oil storage tanks in petroleum
refineries. Thermosyphons are more suitable for industrial applications because, as
they do not require porous media to operate, they are easy to construct. Therefore,
the fabrication costs are much reduced when compared to heat pipes. Besides, they
are able to transport at least one order of magnitude more heat than heat pipes.
However, as they need the gravity action to operate, the heat sources (evaporators)
necessarily are positioned at below positions relative to the heat sinks (condensers).
When this condition cannot be fulfilled, such as in microgravity applications or
mobile computers, for instance, heat pipes are the correct technology to be applied.
This book is about theories and applications of thermosyphons and heat pipes. The
major objective of this book is to organize information so that the engineer is able
to select and design the appropriate thermosyphon or heat pipe device for the target
application. However, it was not the intention to duplicate available information.
Only brief reviews are presented for the material that is extensively reported in the
open literature. The interested reader is invited to further explore this material.
1.3 Book Outline 5

1.3 Book Outline

The present book is divided into three major parts.

In PART I, FUNDAMENTALS AND FABRICATION ASPECTS, the
fundamental operation principles of thermosyphon and heat pipes and the physical
phenomena involved are described. Theoretical aspects about condensation in cooled
vertical walls, liquid–vapor and vapor–liquid phase change, liquid–solid interface
tension, wetting phenomena, disjoining pressures, fluid flow in tubes, boiling, etc., are
presented. Thermodynamics aspects are used to construct basic thermal and hydraulic
models for thermosyphons and heat pipes. Liquid and vapor pressure drop distribu-
tion models are presented. Discussion and modelling of the heat transfer operation
limits are shown. Geyser boiling effect, which deeply affects thermosyphon hydraulic
and thermal behaviors, are discussed, as well as mechanisms used to avoid it. Basic
steady-state one-dimensional models, based on the thermal resistance circuits, are
presented and compared with experimental results for both thermosyphons and heat
pipes. Correlations for the coefficients of heat transfer, required for the determination
of the thermal resistances, are presented. Thermosyphons and heat pipes design
methodologies are discussed. Fabrication and experimental test setup design aspects,
are also discussed. To illustrate the modelling and design methodology proposed,
PART I is concluded with the design and testing of several actual thermosyphons.
In PART II, SPECIAL THERMOSYPHON AND HEAT PIPE TECHNOLO-
GIES, several special configuration of thermosyphons and heat pipes are presented
and modelled. First, these devices are classified according to their operational princi-
ples. Then, loop thermosyphons, capillary pumped loops (including loop heat pipes),
pulsating heat pipes and vapor chambers are discussed in details. Mini thermosyphons
and heat pipes technologies and models are also presented, including hybrid heat
pipe/thermosyphons, wire-plate heat pipes and hybrid wire-plate/sintered wick heat
pipes.
In PART III, APPLICATIONS, special attention is given to applications of
thermosyphons and heat pipes. The design and construction aspects of thermosyphon
heat exchangers, largely employed in the industry, are presented. Several special
mini heat pipes and cabinet wall thermosyphon assisted technologies for electronics
cooling, including avionics and smart gadget applications, are discussed. The use of
thermosyphons for saving water in wet cooling towers, for solar heating of houses, for
improving the thermal performance of ovens (vertical and conveyor belt) designed to
bake food (bread, biscuits, etc.) or to dry fruits and medicinal herbs, are also shown.
Thermosyphon equipment designed for the petroleum industry are also presented,
including heaters for oil storage tanks and for natural gas, in gas distribution city
gates.
One of the objectives of this book is to collect the necessary information to design
new devices or to improve already established industrial process and equipment by
the use of thermosyphons and heat pipes. However, the major purpose is to inspire
engineers to use this very promising and effective technology to propose new quality
and creative thermal solutions for many different industrial problems.
 Part I
Fundamentals and Fabrication Aspects
Chapter 2
Physical Principles

The physical principles that drive the operation of thermosyphons and heat pipes are
mostly based on the thermal and mechanical behavior of the several interfaces inside
these devices, especially the interfaces between the working fluid phases (liquid and
vapor) and between the working fluid and the solid surfaces, such as casing, grooves
or wicks. Therefore, most of this chapter is devoted to the understanding of these
physical principles and to the modeling of heat transfer and forces resulting from
these interfaces. In the first part of the chapter, thermosyphon related models are
discussed, followed by heat pipe models.

2.1 Nusselt Model for Wall Condensation

Gravity is the major driving force in thermosyphons, responsible for pushing conden-
sate working fluid from the condenser to the evaporator and for dragging the vapor
from the evaporator to the condenser, by buoyancy effects.
The heat transfer mechanisms and the thermal behavior of the condenser and
evaporator sections of thermosyphons are mostly described based on literature well-
known models, such as condensation over cooled walls and pool boiling. In this
section, these models are briefly presented, highlighting how they can be adapted to
predict the performance of thermosyphons.
Usually, the Nusselt model, briefly treated in this section, is used to predict the
condensate formation over internal walls of the condenser sections of thermosyphons.
In general, the vapor condensation over vertical cooled walls, in the presence of
saturated vapor, can happen in heterogeneous and homogeneous conditions (Collier
and Thome 1994).
Three heterogeneous condensation conditions can be recognized: drop-wise, film-
wise and direct contact. Heterogeneous condensation usually happens close to impu-
rities (such as dust), where the saturated vapor suffers a localized pressure drop. For

© Springer Nature Switzerland AG 2021 9

M. B. H. Mantelli et al., Thermosyphons and Heat Pipes: Theory and Applications,
https://doi.org/10.1007/978-3-030-62773-7_2
10 2 Physical Principles

Fig. 2.1 Schematic of

condensate film for Nusselt
model

thermosyphon equipment design, these models are not usually employed and so they
are not described in this text.
A simple model developed by Nusselt (1916) for homogeneous vapor condensa-
tion over a vertical cooled wall in contact with a pure single-component saturated
vapor, is presented in this Chapter. Stagnant quiescent vapor is considered in contact
with a vertical cooled wall, over which the condensate liquid forms a uniform thin
liquid film, which flows in downward direction, due to gravity action. Its thickness
is zero at the wall top and increases with the descending film. Figure 2.1 shows
the schematic of the condensate film over a cooled vertical surface. The following
hypothesis are assumed:
• The flow is laminar;
• The vapor is saturated and pure;
• The vapor temperature is uniform;
• The gravity is the only external force acting on the film;
• As the vapor is stagnant, no dragging forces over the film external surface are
considered;
• The fluid properties are constant;
• No sensible cooling of the film is considered (negligible when compared to the
phase change latent heat);
• The curvature of the liquid–vapor interface is negligible: no tension forces
are considered and so the temperature of the interface is determined from the
saturation vapor-pressure curve of the fluid.
2.1 Nusselt Model for Wall Condensation 11

• Momentum and energy transfer by advection within the condensate film are
considered negligible.
• The thermal profile of the liquid film is assumed to be fully developed.
Due to these hypotheses, the temperature profile across the film is linear and the
heat transfer is one-dimensional. The heat transfer mechanism is heat conduction
through the film, in the direction: vapor–liquid interface to the wall.
The x-momentum conservation equation, for the element depicted in Fig. 2.1,
neglecting the inertia (advective) terms, can be written as:

∂ 2v 1 dp X
= − , (2.1)
∂y 2 μl dx μl

where v is the velocity in the x direction, μl is the dynamic viscosity, X represents

the body force term and dp/dx is the free stream pressure gradient in the quiescent
vapor region, outside the liquid layer of thickness δ l , given by ρv g. Gravity is the
body force inside the liquid layer, given by ρl g. ρl and ρv are the liquid and vapor
densities, respectively and g is the gravity acceleration. Therefore, the last equation
becomes:

∂ 2v g
= − (ρl − ρv ). (2.2)
∂y 2 μl

which, integrated twice and considering v(0) = 0 and d v/dy|y=δ = 0, results in

the Nusselt velocity profile:
   
g(ρl − ρv )δl2 y 1 y 2
v(y) = − (2.3)
μl δl 2 δl

One should note that the velocity given by this expression depends on both
coordinates y and x, this last indirectly, as δ l depends on x.
The heat for the liquid film condensation q is taken from the wall. The energy balance
equation, for the elementary region of length dx illustrated at Fig. 2.1, is:

dq = hlv d ṁ = qs b dx (2.4)

where hlv is the latent heat of vaporization, ṁ is the mass flux, qs is the heat flux at
the surface and b is the wall width. As the temperature profile within the liquid film is
considered linear, convection heat transfer coefficient can be expressed as h = kl /δl
and, according to the Fourier law, one has:

kl (Tsat − Ts ) kl · T
qs = = (2.5)
δl δl
12 2 Physical Principles

where kl is the thermal conductivity of the liquid film, T is the temperature

difference between the liquid–vapor saturation temperature (T sat ) and the surface
temperature T s .
Therefore, the condensed liquid mass rate, by wall unit width, is given by:

δl (x)
ṁ gρl (ρl − ρv )δl3
= ρl v(y)dy = (2.6)
b 3μl
0

Substituting Eqs. 2.6 and 2.5 in Eq. 2.4, one gets the following ordinary differential
equation:

kl μl (Tsat − Ts )
δl3 d δl = dx (2.7)
gρl (ρl − ρv )hlv

which, integrated from δ(0) = 0 to δ(x), gives:

 1/4
4kl μl (Tsat − Ts )x
δl (x) = (2.8)
gρl (ρl − ρv )hlv

Using the same definition of the liquid film convection heat transfer coefficient
as before (h = kl /δl ), one has:
 1/4
gρl (ρl − ρv )hlv kl3
h(x) = (2.9)
4μl (Tsat − Ts )x

A mean heat transfer coefficient along a characteristic length L along th cooled

wall vertical direction can be determined, through the following expression:

L
1
hL = h dx (2.10)
L
0

Substituting Eq. 2.9 into Eq. 2.10 and performing the integration, the following
expression is obtained, for the mean heat transfer coefficient (Kedzierski et al. 2003):
 1/4
ρl g(ρL − ρv )hlv kl3
hL = 0.943 (2.11)
μl (Tsat − Ts )L

It is interesting to note that hL = 4/3 · h(L). Equation 2.11 is also valid when the
heat flux (and not the temperature difference) is regarded as uniform. In this case,
q is replaced by the mean heat flux q = kl T /L ∫L0 1/δl (x)dx and T is obtained
using the uniform wall temperature (Rose 1988).
2.1 Nusselt Model for Wall Condensation 13

One should note that Eq. 2.9 presents local heat transfer coefficients. Furthermore,
Collier and Thome (1994) suggest that the mean heat transfer coefficient can also be
obtained considering a heat balance over the total length of the cooled surface: the
total heat transferred along the cooled surface of length L is equal to the energy used
for vapor condensation. Therefore, the mean condensation heat transfer coefficient
can be given by:

ṁ(L) hlv
hL = . (2.12)
b L (Tsat − Ts )

Eliminating the temperature difference of this expression by using Eqs. 2.5

and 2.4, one has:

kl ṁ(L) dx
δl (x) = (2.13)
hL L d ṁ

Inserting this expression in Eq. 2.6 and after some algebraic manipulation, the
following differential equation is found:
 1/3
gρl (ρl − ρv )kl3 b 1
dx = hL ṁ1/3 d ṁ (2.14)
3μl ṁ(L) x

which, integrated over the length L, results in the following expression for the
mean heat transfer coefficient:
 1/3
gρl (ρl − ρv )kl3 b
hL = 0.925 (2.15)
μl ṁ(L)

For thermosyphon design, this expression can be more convenient than

Eq. 2.11 , when the wall temperature is not known. Collier and Thome (1994) also
highlights that the heat transfer coefficient can be expressed in terms of the Reynolds
number of the condensate liquid film, which is defined as:

4ṁ(L)
Reδ = . (2.16)
μl b

Substituting the Reynolds number expression on Eq. 2.15 , one gets the following
equation, based on the mean heat transfer coefficient (Eq. 2.12), valid for Reδ ≤ 30:
 1/3
hL μ2l −1/3
= 1.47Reδ (2.17)
kl gρl (ρl − ρv )

As hL = 4/3 · h(L), the last expression, in terms of local condensing heat transfer
coefficient, can be rewritten as:
14 2 Physical Principles

 1/3
h(L) μ2l −1/3
= 1.1Reδ (2.18)
kl gρl (ρl − ρv )

From these last equations, a characteristic length, for the condensation phenomena
over a vertical cooled plate, can be defined as:

μ2l
LC = (2.19)
gρl (ρl − ρv )

As already mentioned, most of the literature models regarding gravity-driven

condensation heat transfer over cooled walls is based on the Nusselt model. However,
the actual condensation heat transfer mechanisms can be quite different from the
model predictions, due to the influence of several factors that affect the phase change
mechanism. Revised Nusselt models are available in the literature and those relevant
for thermosyphon applications, are briefly presented next.
Effect of the Condensate Temperature
According to Collier and Thome (1994), Bromley in 1952 and Rohsenow in 1956
extended the Nusselt theory to consider the subcooling and the non-linear distribution
of the temperature through the film, showing that the latent heat of vaporization, used
in Eq. 2.9 , could be replaced by:

hlv = hlv (1 + 0.68 · Jal ) (2.20)

where Jal is the Jakob number, a relative measurement of the degree of subcooling
of the liquid film, defined as:

cpl (Tsat − Ts )
Jal = (2.21)
hlv

where cpl is the specific heat at constant pressure of the condensate liquid film.
However, Rose (1988) states that it is not evident that this correction improves
significantly the Nusselt model results.
As stated by Collier and Thome (1994), the thermophysical properties of the film
should be evaluated for the effective film temperature, using the following expression:

T = Ts + 0.25(Tsat − Ts ) (2.22)

Minkowycz and Sparrow (1966) suggest the substitution of the number 0.25 to
0.31 in the last equation.
Effect of Vapor Shear Stress
The hypothesis that the interface between laminar falling film is smooth is not always
valid. In counter-current annular film flow, ascending vapor causes shear stress over
the external surface of the descending film, causing ripples and waves. This is the case
2.1 Nusselt Model for Wall Condensation 15

of thermosyphons, where the interface shear stress may cause the liquid to absorb
some vapor, reducing the velocity and increasing the film thickness. In addition, the
convection within the condensation film may affect the heat transfer. Both effects
decrease the heat transfer coefficient and were not considered in Nusselt’s model.
Many literature papers (Chun and Seban 1971; Fujii and Uehara 1972; Rose
1988) deals with vapor shear stress when vapor and liquid film moves in the same
downward directions or in horizontal plates (Koh 1962). As these models cannot be
applied directly to a typical thermosyphon as described in Fig. 1.1 these works will
not be described here (note that some of these models might be useful for describing
the condenser phenomena of some loop thermosyphons).
Koh et al. (1960) studied numerically the two-phase flow problem in laminar
condensation over vertical cooled walls, considering that the condensed liquid flow
induces motions within the vapor, which, in turn, affects the condensate velocities.
In other words, they considered vapor shear forces in their model. They concluded
that the effects of the interfacial shear stress can be especially important for the
condensation of liquids with very small Pr l numbers, such as liquid metals.
Chen, in 1961, proposed the following correlation to correct the Nusselt expression
to account for the vapor drag effects, which compared with numerical results within
1%:
⎡ ⎤1/4
Ja2
1 + 0.68Jal + 0.02 Prll
h = hNu ⎣ ⎦ (2.23)
Ja2
1 + 0.85 Pr
Jal
l
− 0.15 Prll

Pr l /Jal is a dimensionless acceleration effect parameter that takes into account the
effect of the convection within the liquid layer, where the Pr l is the Prandtl number,
given by Prl = cpl μl /kl . This last equation is valid for Jal ≤ 2 and for Jal /Prl ≤ 20,
for liquids with Prl larger than one or less than 0.05. Besides, Rose (1998) sates that,
in most actual applications, the effect of the vapor shear stress in free convection is
very small. Also, if Eq. 2.23 is used, no corrections of the temperature effect in the
latent heat of vaporization (use of Eq. 2.20) is necessary.
Effect of Turbulence
As happens to any boundary layer flow, as the Reynolds number increases, a transition
from laminar to turbulent flow is expected. In general, a regime of laminar flow is
observed over the upper region of the surface, changing to turbulent as the liquid flows
to the rear parts of the cooled plate. For the case of a condensate film flowing down
the wall, even for low Reynolds numbers, the assumption of viscous flow behavior
is questionable, as waves can be easily observed. Experimental data shows that the
heat transfer coefficients can be much greater than those predicted by Eq. 2.17.
Other literature works such as Kutateladze (1981) used similarity analysis to study
the arising waves mechanisms. Carey (1992) suggests, based on the literature, that
waves are expected for flows with Reδ > 33.
16 2 Physical Principles

For long vertical surfaces where both regimes can be observed, Collier and Rose
(1994) proposed the following correlation obtained from experimental data, valid for
Reδ > 2000:
 1/3
hl (x)
(Lc ) = 0.056 · Reδ Pr
1/3 0.2
(2.24)
kl l

The idea is to integrate the local heat transfer coefficient using Eq. 2.9 along the
plate length up to a Reynolds number Ref of 2000 and to use Eq. 2.24 for rest of the
length. Another expression is available in the literature, valid for 1 ≤ Prl ≤ 5 and
Reδ > 1400 (Carey 1992):
 1/3
h νl2 1/3
NuL = = 0.0131 Reδ (2.25)
kl g

where νl is the liquid viscosity.

On the other hand, Kutateladze (1963) suggests the use of the following
correlation, for the laminar wavy film condensation, for 30 < Reδ < 1800:
 1/3
h νl2 Reδ
NuL = = (2.26)
kl g 1.08Re1.22
δ − 5.2

Besides, according to Incropera et al. (2014), Labuntsov recommends the use of

the following expression when the fluid is in turbulent flow:
 1/3
h νl2 Reδ
NuL = = −0.5
 (2.27)
kl g 8750 + 58 Pr l Re0.75
δ − 253

In the present work, special attention is given to counter-current flowing vapor

configuration (reflux condensation condition, such as happens in thermosyphon
applications), which retards the condensate flow and thickens the condensate layer.
Chen et al. (2007) proposed a general expression, applicable to all regimes (laminar-
wave or turbulent), based on correlations for quiescent vapor, i.e., gravity dominated
film condensation for vertical cooled walls, based on the method of Churchill and
Usagi (1974) and on literature correlations, resulting in the following correlation for
the local Nusselt number:
 1/3  1/3 1/2
hl (x) νl2 −1.32 Rel (x)2.4 Pr3.9 Pr1.3 ∗
Nul (x) = = 0.31Rel (x) + l
+ l
τ
kl g 2.37 × 1014 771.6 i
(2.28)

where Rel (x) is the liquid Reynolds number which varies with the length of the
cooled wall and τi is the interface shear stress. For annular film condensation inside
2.1 Nusselt Model for Wall Condensation 17

tubes, the following expression for the vapor phase change dimensionless interfacial
shear stress, is proposed:
τi
τi∗ = (2.29)
ρl (gνl )2/3

The last term of Eq. 2.28 which accounts for the shear stress, may have plus
or minus signs, depending on the direction of the vapor motion: plus if in favor,
or concurrent, minus if against, or countercurrent to the condensate gravity driven
liquid displacement. If the secondary effects are neglected, such as augmentation
of the interfacial shear stress due to phase change, the Fanning friction coefficient
(Seban and Hodgson 1982) can be used to determine interface shear stress. Chen
et al. (2007) proposed the following expression for turbulent flow inside tubes:

τi∗ = −CRel (x)1.8 (2.30)

where the dimensionless group C is defined by:

0.023 μ1.133 μ0.2

C= l v
(2.31)
D2 g 2/3 ρv ρl0.333

and D is the tube diameter.

Expression 2.28 can be averaged along the length of the tube, as the Rel (x) varies
from zero at the top of the cooled wall (or tube) to some value at position x = L,
using the expression:
⎛ Re (L) ⎞−1
 1/3 l
h νl2 d Re (x)
= Rel (L)⎝ ⎠
l
NuL = (2.32)
kl g Nul (x)
o

The following correlation is, therefore, obtained:

 1.8 1/2
−0.44 Rel (L)0.8 Pr 1.3 l Rel (L)
C Pr 1.3
NuL = Rel (L) + l
− (2.33)
1.718 × 105 7.716 × 104

Chen et al. (2007) states that this expression is valid if there is no flooding within
the tube, as in this case, no counter-current could be possible. Therefore, the following
constraints should be respected:
 1/3  2  1/2
g Ck ρv
Rel (x) ≤ D 2 (2.34)
νl Bo1/4 ρl
18 2 Physical Principles

where:
 
√ Bo1/4
Ck = 3.2 tanh (2.35)
2

The Bond number is defined by:

 1/2
(ρl − ρv )g
Bo = D (2.36)
σl

and σ l is the liquid surface tension. If the vapor density is much smaller than
the liquid, the bond number is simplified to Bo = D(ρl g/σl )1/2 . These authors used
literature data and found a good comparison with these correlations.
Other Effects
The presence of non condensable gases in the condensation over vertical cooled walls
has been treated in the literature (Collier and Thome 1994). Minkowycz and Sparrow
(1966) presented a wide ranged analytical investigation of laminar film condensation,
considering the presence of non-condensable gases. Their analytical model included
interfacial resistance, superheating, free convection due to temperature and concen-
tration gradients, mass and thermal diffusion and variable properties of the liquid
and gas–vapor regions. They demonstrated that the noncondensable gas concentra-
tion can affect definitely the heat transfer rate. These models can be actually very
useful for applications where non condensable gas is mixtured with working fluid in
thermosyphons.
Collier and Thome (1994) also present expressions for the Nusselt vapor conden-
sation over inclined surfaces. In this case, Eqs. 2.9 and 2.11 are slightly modified, to
account for the inclination angle θ, resulting in, respectively:
 1/4
g · sin θ · ρl (ρl − ρv )hlv kl3
h(x) = (2.37)
4μl (Tsat − Ts )x
 1/4
ρl g · sin θ · (ρL − ρv )hlv kl3
hL = 0.943 (2.38)
μl (Tsat − Ts )L

2.2 Liquid–Vapor Phase Change Model

Liquid–Vapor Transitions at Molecular Level.

The interaction between two molecules is usually characterized by a potential func-
tion (r), defined as the energy necessary to bring two infinitely distant molecules to
a finite distance r. It depends on the nature of the interaction between molecules and
2.2 Liquid–Vapor Phase Change Model 19

Fig. 2.2 Lennard–Jones

potential curve

can be attractive or repulsive. When very close, two molecules usually exert repul-
sive forces to each other, which increase rapidly as the spacing r decreases. In larger
distances, these forces, which are usually characterized as electrostatic, induction
and dispersion, became attractive.
According to Carey (1992), Lennard–Jones is one of the most known models for
this potential, expressed by:
 
r0 12  r0 6
(r) = 4ε − (2.39)
r r

where the first term at right side represents repulsion and the second the attraction
forces. A schematic plot of this expression is shown in Fig. 2.2. The parameter r0
(distance where the potential is zero) and ε (parameter related to the distance where
the potential is minimum: ε = −rmin , see Fig. 2.2) vary according to the type of
molecules. The variation of the potential implies that energy must be removed to
bring close two molecules that are very far apart from each other. However, if two
molecules are close enough to be attracted (not so close so that repulsive forces takes
place), energy must be supplied to increase the spacing between the molecules. More
information about Lennard–Jones potential can be found, among several books in
the literature, in Tien and Lienhard (1972).
These energy exchanges are associated with the latent heat of vaporization (where
closed molecules are moved apart) or condensation (where distant molecules are
moved closer).
The kinetic theory of gases considers that a gas is composed by a very large
number of particles moving in random directions, which can collide elastically with
one another. Between collisions, it is assumed that the particles move in straight
lines. The pressure can be understood as resulting from the force derived from the
rate of change of momentum of colliding particles with a wall of a recipient where
the gas is confined. The temperature of a gas is directly associated with the random
motion of the particles, so that, at absolute zero temperature, the molecules motion
ceases completely.
20 2 Physical Principles

Individual molecules in the gas may travel in different directions, at different

speeds. The Maxwell–Boltzmann probability distribution is used to describe gaseous
particles (atoms or molecules) speeds of an ideal gas in thermodynamic equilibrium.
The particles are considered to move freely inside a stationary container. Their inter-
action consists of very brief collisions in which they exchange energy and momentum
within each other. A particle speed probability distribution indicates the range of
speeds that a particle selected randomly is likely to be. The distribution depends on
the temperature of the system and on the mass of the particle.
In real gases, several effects, such as van der Waals interactions, vortical flows,
quantum exchange interactions, etc., can make their speed distribution different from
Maxwell–Boltzmann’s. However, the Maxwell speed distribution is an excellent
approximation for rarefied gases at ordinary temperatures, which behavior is similar
to an ideal gas. More information about kinetic gas theory can be found in the
literature (Hirschfelder et al. 1954).
In ideal cases, the Maxwell–Boltzmann speed distribution can be used in the
prediction of the kinetic energy distribution of the molecules, resulting in a distribu-
tion curve as illustrated in Fig. 2.3. In this figure, the hachured area represents the
fraction of molecules that have kinetic energy above the threshold value, represented
by the vertical line. The fact that the energy distribution among molecules of liquid
and vapor are similar to Maxwell–Boltzmann distribution can be used to explain the
relation between the equilibrium pressure and temperature of saturated vapor and
liquid. Due to the kinetic energy distribution, even at low temperatures, a fraction of
the molecules of the liquid has sufficient energy to escape from the cohesive forces of
other liquid molecules. This fraction increases rapidly with temperature. Therefore,
it is expected that a liquid with small cohesive energy have a higher vapor pressure

Fig. 2.3 Molecule energy

distribution
2.2 Liquid–Vapor Phase Change Model 21

than other with large cohesive energy. Actually, the latent heat of vaporization is a
macroscopic indicator of the cohesive energy of the liquid, so that a liquid with high
latent heat of vaporization should have a lower vapor pressure and vice-versa (Carey
1992).
Metastable States
The classical thermodynamics treats the phase change as quasi steady state equilib-
rium conditions in the saturation conditions. However, phase change happens under
non-equilibrium conditions (Carey 1992). Actual processes of vaporization require
superheating temperatures, i.e., temperatures above the saturation temperature levels.
The same observation can be made about condensation, which can be achieved only
after at least part of the vapor is sub-cooled, below the saturation temperature. There-
fore, in a saturation curve plot of a pure substance, there are regions where the liquid
is superheated or vapor is subcooled. These states are known as metastable.
Figure 2.4 shows the P–v Clapeyron diagram for a pure substance. The metastable
regions are highlighted by the gray areas. An isotherm curve, obtained from the Van
der Waals model for real gases, is also plotted. This curve is split in several sections,
representing different thermodynamics conditions: section A-B stands for stable
sub-cooled liquid, section F-G for stable superheated vapor, B-F (horizontal line)
for liquid–vapor equilibrium state, B-C for metastable superheated liquid, E–F for
metastable subcooled vapor and CDE represents a region with no phase stability. The
metastable regions is located between the saturation curve and the spinodal curve.
Phase separation happens for infinitesimal fluctuations in composition and density
of substances within the spinodal curve.

Fig. 2.4 Pure substance

P–V diagram and Van de
Waals curve
22 2 Physical Principles

A state is thermodynamic stable when (∂P/∂v)T < 0. This means that, keeping
the same temperature level, an increase in the pressure necessarily implies in a
decrease of volume (or vice-versa). Observing Fig. 2.4, the CDE line does not fulfill
the stability criteria, as the pressure increases with the volume and therefore no
stability is expected in the region between the saturation and the spinodal curves.
However, curves BC and EF follow the phase stability criteria, but are in a state of
non-thermodynamic equilibrium, which means that, for the same pressure, there is
another state of lower chemical potential. In the metastable regions, any considerable
change in the thermodynamic conditions lead to the phase change, as it represents
lower chemical potential.
Boiling can be considered a process where a discontinuity of the properties of the
substance is observed. When the liquid reaches the saturation condition and there
is an infinitesimal change in the environment, which causes a discontinuity of the
properties of the substance, a “jump” between the liquid metastable to the saturated
vapor states is observed or between the vapor metastable to saturated liquid.
Interfacial Tension
Although, in most two phase (liquid-vapor) interface models, a sharp discontinuity
in the fluid composition and properties (such as density) is assumed, actually there
is a transition that happens within the thin region centered at this interface. The bulk
liquid molecules are closer to each other when compared to those in the liquid–vapor
interface region. As the distance between molecules is larger in the interfacial region
and considering that energy must be supplied to move molecules apart, it is logical
to suppose that the energy per molecule is greater in the interface than in the bulk
region. In other words, there is an additional free energy per unit area in the interface
(Girifalco and Good 1957).
In the bulk liquid region, the repulsion forces act in each molecule in all directions,
so that the resultant force is zero. However, in the regions where the liquid is near
to other medium (vapor, liquid or solid), the mean spacing between molecules is
larger and the repulsive forces are reduced considerably, so that the forces acting
in the molecules are not in equilibrium and the resultant force draws the molecules
toward the liquid bulk region. In the interface, the resultant of the radial forces due
to the interactions among the molecules is null parallel to the interface (Carey 1992).
The resultant force in the direction of the bulk liquid over the liquid–vapor interface
makes the drop of a liquid in contact with its vapor to get the spherical shape.
On the other hand, the interface tension can be interpreted as a force per unit area
acting perpendicular to the density gradient and parallel to the interface, resulting
from the combination of the attractive and repulsive interactions between molecules
of the substance. Due to this spontaneous trend to contract, the liquid surface behaves
like a rubber membrane under tension and the interface tension can be interpreted as
energy stored in the molecules near the interface.
To increase the liquid–vapor interface area, work must be exerted. The surface
free energy E associated with this work divided by the surface area A, is denominated
as surface tension σ, or (Adamson and Gast 1997):
2.2 Liquid–Vapor Phase Change Model 23
 
∂E
σ = (2.40)
∂A T ,p,ni

which dimension is energy by area or force by unity length. The surface tension
depends on the temperature T, pressure p and the number of moles ni of the gas i
component. This equation is valid for interfaces between liquid–gas, liquid–solid and
liquid–liquid, this last one for the interaction between two immiscible liquids. The
surface tension is a fundamental property that characterizes the surface properties of
a fluid. The interface between liquid and gas is a very thin, tridimensional surface
and its properties are different from those of the phases that this surface separates.
Wetting Phenomena–Contact Angle
Usually, in heat transfer equipment that makes use of liquid–vapor phase change
mechanisms, energy is transferred to the liquid by a wall of a container. Hence, the
knowledge of the ability of a liquid to wet a solid surface is of major importance for
their design.
The interaction of liquid and solid surfaces depends mainly on the properties and
characteristics of the solid wall material (including its surface finishing) and of the
liquid. The capacity of the liquid to “wet” the solid surface results from the “affinity”
between liquids and solids, a consequence of the attractive forces among the liquid
and solid molecules. However, the forces between the liquid and the solid molecules
can be repulsive. In this last case, the attraction forces among the liquid molecules
make the liquid drop to assume a spherical shape and, in the limit, the liquid–solid
contacting area reduces to a single point.
For a drop resting in a solid surface, the angle formed by the liquid–vapor interface
and the solid is denominated as “contact angle”. The value of this angle is actually
a quantification of the liquid–solid wettability. The contact angle is usually weakly
dependent of the container shape and of gravity forces. Figure 2.5 illustrates the
geometry of a liquid drop over a solid surface in a vapor environment. The three
phase (solid, liquid and vapor) line is circular and is denominated as the “contact
line”. Forces resulting from the surface tensions: σlv (liquid–vapor), σsl (solid–liquid)
and σsv (solid–vapor) act over the contact line. The following expression results from
horizontal component forces, in equilibrium state:

σsv = σsl + σlv cos θ (2.41)

Fig. 2.5 Drop resting on a solid surface

24 2 Physical Principles

Fig. 2.6 Wetting characteristics of liquid drop over a solid surface

This equation is known as the Neumann Formula or Young Equation (Carey 1992).
Isolating the contact angle θ, one gets:
 
−1 σsv− σsl
θ = cos (2.42)
σlv

If σsv > σsl , the contact angle θ is acute and a liquid wetting condition is perceived.
However, if σsv < σsl , the contact angle is obtuse and a non-wetting condition is
observed. Actually, for a fixed volume liquid drop (considering that no condensation
or vaporization is taking place) the increase of the contact angle θ represents a
decrease of the solid–liquid interface area. On the other side, the decrease of angle
θ results in the liquid spreading, and, as a result, the increasing of the solid–liquid
interface area. In the limit, as θ → 0, the liquid is considered as “full wetting” and
spreads all over the solid surface, forming a thin film. For θ → π the liquid is “non-
wetting” and the liquid-surface contact reduces to a single point. Figure 2.6 shows a
liquid drop in wetting, non-wetting and partial wetting conditions (θ < θ < π ). In
wetting liquids, the adhesive forces between liquid and solid are predominant to the
cohesive forces among the liquid molecules themselves. In non-wetting fluids, the
opposite happens and, if gravity or other field forces are neglected, the liquid drop
shape tends to a sphere.
Due to the surface tension effects, the upper surface of a liquid within a container
is curved, especially close to the wall regions, forming the so-called meniscus. This
curvature can be either concave or convex, depending on the liquid wetting char-
acteristics. Figure 2.7 left side, illustrates the contact angle of a wetting liquid and,
right side, of a non-wetting liquid. In this case, the angle between liquid and surface
is θ < π for the wetting and θ > π for the non-wetting liquid.
Back to Fig. 2.5 that shows a drop resting on a solid surface, the energy dE
(Helmholtz surface free energy) necessary to modify the interface areas, can be
determined considering the three surface tensions: σlv , σsl and σsv acting in the
contact line, by the expression:
2.2 Liquid–Vapor Phase Change Model 25

dE = σlv dAlv + σsl dAsl + σsv dAsv (2.43)

If the liquid–solid area increases by dAlv , the vapor–solid area dAsv decreases by
the same amount. Besides, considering that the drop has a truncated hemispherical
shape, dAlv = dAsl cos θ . So, the last expression can be given as:

∂E
= σlv cos θ + σsl − σsv (2.44)
∂Asl

The meniscus stable form is obtained when the liquid–vapor surface presents the
least free energy, or:

∂E
=0 (2.45)
∂Asl

In the case of a liquid drop, considering no other forces, this shape is a sphere (as
already noted).
Usually, due to the small order of magnitude and the difficulty of measuring the
solid–liquid and liquid–vapor surface tensions, these terms are neglected in the design
of thermosyphons and heat pipes and only the liquid–vapor tension is considered.

Fig. 2.7 Wetting (left) and

partially wetting (right)
liquid in tube
26 2 Physical Principles

Fig. 2.8 Menisci geometry in a two plate channel with angle ϕ

This surface tension property is available for several liquids in many books from the
literature (Faghri 2016; Peterson 1994; Reay and Kew 2006 among others).
In some thermosyphons and heat pipes, the liquid can be within a V shaped
channel, with ϕ angle (see Fig. 2.8). For a fixed volume of liquid, the shape of the
meniscus and of the wall-liquid contact area can be controlled by the ϕ angle. If the
liquid–solid contact angle is θ, the liquid meniscus is concave if φ < 180 − 2 × θ
and convex otherwise. Figure 2.8 illustrates these menisci. Usually, especially for
heat pipe evaporators, large contact areas between the heated walls and the liquid are
desirable.
Capillary Pressure
As described in Chap. 1, the operation principle of thermosyphons and heat pipes
is mainly based on an evaporation and condensation cycles, that allows the transfer
of very large amounts of heat. To guarantee this cycle, liquid must be continuously
supplied to the evaporator. In the case of thermosyphons, the gravity is usually
responsible for this task. However, for heat pipes, a porous wick, usually located
inside the device, do this job.
For the case of heat pipes, the liquid–vapor phase change takes place in the wick.
The evaporation process leads to an increase in the concavity of the meniscus, which,
in turn, leads to an unbalance of the surface tension forces between evaporator and
condenser menisci, resulting in a capillary pressure. Therefore, the capillary pressure
phenomenon is directly related to the meniscus radius of curvature.
Figure 2.9 shows a schematic of a meniscus with curvature radius r x and r y in the
x and y directions, respectively. A small displacement dz in the z direction causes a
deformation of the meniscus area, here represented by a stretch of dx and dy, so that
the original area A = x · y, changes to:

A = (x + dx)(y + dy) − xy (2.46)

Developing the above equation and considering that dx · dy ≈ 0, then:

A = y · dx + x · dy (2.47)
2.2 Liquid–Vapor Phase Change Model 27

Fig. 2.9 Stretched meniscus

geometry with curvature
radius r x and r y

According to Eq. 2.40, the necessary energy to displace the meniscus surface by
dz is equal to the work exerted by the resulting normal force over the meniscus area
(p · xy):

dE = σ (x · dy + y · dx) = p · xy · dz (2.48)

where p is the pressure difference between the liquid and vapor regions, separated
by the meniscus. Considering similarity of triangles, one gets:

x + dx x y + dy y
= and = (2.49)
rx + dz rx ry + dz ry

Therefore, dx and dy can be expressed as, respectively:

x · dz y · dz
dx = and dy = (2.50)
rx ry

Substituting in Eq. 2.48, one gets:

 
xy · dz xy · dz
p · xy · dz = σ + (2.51)
rx ry

p corresponds to the capillary pressure pcap , due to pressure differences between

the liquid and vapor, in both sides of the meniscus and is given by:
28 2 Physical Principles
 
1 1
pcap = p = σ + (2.52)
rx ry

This expression, known as Young–Laplace equation (Carey 1992; Faghri 2016)

is fundamental for the determination of the capillary pressure of a wick structure. It
shows that, for applications where the surface tension does not vary significantly, the
capillary pressure depends mainly of the meniscus curvature.
One of the consequences of the Young–Laplace equation is that the surface tension
observed in a convex liquid–vapor meniscus is larger than that of a concave surface.
This fact can explain the phenomenon associated with the penetration of fluid in
capillary tubes. Two schematics are shown in Fig. 2.10: in the left, a capillary tube
is inserted in a container full of a wetting liquid, while in the right, the liquid is
non-wetting. In the left, liquid penetrates the tube up to a height H, while, in the
right, the liquid inside the tube is forced against the container fluid and a recession
of height -H is observed.
From the left drawing of Fig. 2.10, it is evident that another force acts against
gravity to keep the liquid plug in the high level inside the tube of radius r. The same
force, but in the opposite direction also acts in the right drawing. Considering a
quiescent interface and only the liquid–vapor surface tension, the capillary pressure
over the meniscus area (approximated to π r 2 ) results in a force equal to the sum of

Fig. 2.10 Capillary elevation of liquid in a tube

2.2 Liquid–Vapor Phase Change Model 29

the forces acting over the solid–liquid contact line, or:

pcap · π r 2 = 2π r · σ cos θ (2.53)

Isolating pcap from the last equation one gets:

2σ
pcap = cos θ (2.54)
r
and considering that, for a circular tube, the curvature radius r x and r y are the
same, the radius of the tube and the of meniscus curvature are related by:
r
rx = ry = (2.55)
cos θ
Point b of Fig. 2.10 left side, is located in the basis of the liquid column and is
under the action of the hydraulic pressure of the liquid column of height H, of the
capillary pressure and of the vapor pressure pv (which acts at the meniscus at point
c), i.e.:

pb = ρl gH + pv − pcap (2.56)

However, in point a of this same sketch, the pressure exerted by vapor is composed
of two parcels: column of height H and the vapor pressure at point c, resulting in:

pa = ρv gH + pv (2.57)

As the system is in equilibrium, equating Eqs. 2.56 and 2.57, one has:

2σ
pcap = (ρl − ρv )gH = cos θ (2.58)
r
The maximum capillary pressure happens when cos θ = 1, or θ = 0. In this
case, the liquid wets the tube completely. For the non-wetting liquid, θ > π/2
and the capillary forces pushes the liquid in the container direction, so that a liquid
depression is observed in the tube, as shown in Fig. 2.10 right side. Therefore, for
heat pipe applications, working fluids with high wettability are desirable, aiming the
effectiveness of the capillary action of the wick.
In heat pipes, due to the evaporation process, the working fluid tend to recede
into the wick in the evaporator section and the meniscus radius tends to decrease.
The opposite happens on the condenser region, where the excess of liquid makes the
radius of the meniscus to increase, tending, in the limit, to infinite. Therefore, in the
evaporator, the meniscus pull the liquid from the condenser region though the wick,
while, in the condenser the meniscus pushes the liquid to the evaporator region. The
combination of both effects results in the capillary pumping of the liquid. Figure 2.11
shows a schematic of the meniscus geometry along a metal screen wicked heat pipe,
highlighting the curvature radius of the meniscus in the condenser r c and evaporator
r e . For the limit case where, in the condenser, the meniscus radius is infinite, only the
30 2 Physical Principles

Fig. 2.11 Meniscus radius in a screen wicked heat pipe

evaporator wick is responsible for the capillary pumping, as the capillary pressure
provided by the evaporator region is zero, as stated by Eq. 2.52. This is also the same
pumping condition for the loop heat pipes, where the wicks are mainly concentrated
at the evaporator.
Many different structures can be used as wicks for heat pipes. Schematics
of common wicks are shown in Fig. 2.12 (Reay and Kew 2006). The most simple
one comprises of a capillary tube and is mostly used to transport condensate from the
condenser to the evaporator with low pressure drops. The most usual wicks in elec-
tronic cooling applications are homogeneous, such as superposed metal wire screen
layers and sintered metal powders, in which the pores are interconnected. Due to their

Fig. 2.12 Schematic of usual wicks

2.2 Liquid–Vapor Phase Change Model 31

small pores, these wicks present high capillary pumping capacity. On the other hand,
internal longitudinal grooves, of different cross section geometries, such as rect-
angular and triangular, are not interconnected. They provide small pressure drops
but also small capillary pumping effects. More complex and efficient porous media
designs are also proposed, such as those composed by the combination of grooved
channels and wire screen wicks, for instance.
For modeling purposes, plain smooth meniscus surfaces are usually considered,
as for the contact angle model (Eq. 2.42). Usually, the capillarity of a wick structure is
determined experimentally. The concept of maximum capillary pressure for a small
diameter (capillary) tube is used to describe the pumping capacity of a wick, by
means of an effective meniscus radius (ref ) that characterizes the capillary porous
media. The maximum capillary pressure for circular cross section tube is obtained
when it is filled with perfect wetting liquid, being given by:

2σ
pcap,max = (2.59)
ref

According to Florez et al. (2017), for a capillary structure made of sintered metal
powder, the effective porous radius can be calculated using the concept of hydraulic
diameter of channels formed among packed spheres of uniform radius. Considering
the powder as composed by spheres of same diameter r p , these authors showed that,
among several packing configurations, the cubic is the one that represents better
the actual particle disposal in sintered wicks. Figure 2.13 shows schematics of this
packing configuration and of the cross section of the liquid path through the sintered
wick. To determine the effective porous radius, the concept of hydraulic diameter
is used, where the transversal irregular cross section geometry (see Fig. 2.13 center
drawing), is considered as a circular duct of variable cross section radius. The
√ use of
the hydraulic diameter concept based on the square root of the area (DH = A) and
of the radius of the liquid flow cross section area proposed by Bahrami et al. (2006),
resulted in the duct shown in the right side of Fig. 2.13. Based on this geometry,
the following expression for the radius of the liquid channel,rlc , is proposed (Florez
2017):

Fig. 2.13 Schematic of the cubic metal sphere packing and liquid channel
32 2 Physical Principles
  
(z−rs )2
4rs2 − π rs2 1 − rs2
rlc = (2.60)
2

where z is the length of the channel, varying from zero to rs . These authors considered
that the largest cross section area is observed at the center of the packing z = 0, which
radius is actually the same of the metal sphere that composes the porous medium,
here represented by the effective radius:

ref = rs (2.61)

These authors also stated that this simple model compared with their data within
25%.
For a grooved capillary structure which cross section channel is a rectangle of
width w, the effective meniscus radius can be considered as w (Chi 1976), and the
maximum capillary pressure is Pmax,cap = 2σ/w. Other expressions for capillary
radius of the wick structures shown in Fig. 2.12 are given in Table 2.1.
Many methods are available for measurement of the surface tension (Reay and
Kew 2006). For heat pipe applications, special attention is given to the determination
of the parameter cos θ , which actually determines the capillary force. Data for surface
tension are available in many specialized books. Recently, Faghri (2016) published
a very complete compilation of heat pipes and thermosyphon working fluid data that
includes surface tension parameters.
Disjoining Pressures
The disjoining pressure, usually a phenomena observed in ultra thin films, can be
important in mini or micro heat pipes and thermosyphons or in other devices where
liquid is in intimate contact with sharped grooved metal surfaces, such as wire-plate
mini heat pipes, pulsating heat pipes, miniature rotating heat pipes, among others.

Table 2.1 Effective capillary

Wick structures ref
radius expressions for several
wick strutures Circular cylinder* r r: radius
Rectangular groove* w w: groove width
Triangular groove* w/ cos β w: groove width
β: half angle
Parallel wires* w w: wire spacing
Wire screens* (w + Dw )/2 w: wire spacing
Dw : wire diameter
Sintered metal powder** rs rs rs : sphere radius
Packed spheres 0.41 · rs rs : sphere radius
Source *Chi, 1976, **Florez et al. (2017)
2.2 Liquid–Vapor Phase Change Model 33

Fig. 2.14 Liquid film over an inclined solid surface

Faghri and Zhang (2006) state that, due to disjoining pressure effect, a liquid
with high wetting capability characteristics (polar or non-polar) in contact with a
planar solid wall, forms an extended meniscus, which can be divided into three
regions: equilibrium thin film, micro film and intrinsic region, as shown in Fig. 2.14.
The equilibrium thin film and the microfilm regions actually compose the thin film
region.
As stated by Wayner et al. (1975), if the film is sufficiently thin, the van de Waals
forces between the solid and the liquid are the predominant ones. Consequently, the
intermolecular attractive forces among liquid–liquid and liquid–solid molecules tend
to push back the liquid to the liquid film direction, making the liquid tightly adhered
to the wall and, consequently, no evaporation is observed. In this case, the liquid
vapor interface temperature Tδ is the same of the wall surface, Ts . This adhesion to
the wall forces result in a pressure denominated “disjoining pressure”.
Virtually, if the surface temperature is larger than the saturation temperature,
all the evaporation happens at the microfilm region and the liquid-vapor interface
temperature lies between the vapor (saturation) and the wall temperatures: Tsat ≤
Tδ ≤ Ts . The disjoining and capillary pressures significantly affect the shape of the
meniscus in this region (Faghri 2016).
Moreover, the surface tension is dominant in the intrinsic region, where the effect
of the disjoining pressure can be neglected. Actually, the liquid flow that feeds the
evaporating microfilm is pumped by a meniscus pressure gradient, caused by changes
of the disjoining pressures as the thickness of the liquid layer increases. Therefore,
the larger the meniscus curvature, the larger the meniscus pumping capacity of the
34 2 Physical Principles

intrinsic region (Faghri 2016). As the film becomes thicker, the disjoining pressure
effect becomes negligible. In the meniscus region, the liquid vapor interface tends
to be planar. Actually, the different meniscus curvature among these regions drives
the liquid flow (see Fig. 2.14). Therefore, the pressure difference between vapor and
liquid at a vapor–liquid interface can be given by:

pv − pl = pcap + pd (2.62)

where Pd is the disjoining pressure.

Based on the Lennard Jones interaction potential (see Sect. 2.2.1), the disjoining
pressure, for a pure non-polar wetting liquid film over a horizontal solid substrate,
is given by (Wayner et al. 1975; Faghri and Zhang 2006; Faghri 2016):

pd (δ) = −Aδ −B (2.63)

where A and B are constants that characterize interactions at the molecular and
electrostatic levels (for more information see Potash and Wayner 1972).
Vaporization and Condensation at Liquid–vapor Interfaces
As already observed, the non-equilibrium conditions resulting from the different
curvature radius of the liquid–vapor meniscus, in the condenser and in the evaporator
regions, are the major driving mechanism for the capillary pressure development
within the porous media, which is responsible for the working fluid transportation
in heat pipes.
If the meniscus shape in a wick is concave (as happens for most of the heat
pipe applications), the attraction between molecules is superior to that of a planar
meniscus. This means that the energy necessary to a molecule to leave a concave
surface is larger than that of a planar interface. As a result, the density and the
vapor pressure over a concave surface are greater. The inverse is observed for convex
meniscus.
From a simple hydrostatic pressure balance and based on Eq. 2.58, the saturation
pressure difference between a planar and can be given as:
 
2σ ρsat
pcap = cos θ (2.64)
r ρl

This expression shows that, for very small equivalent porous medium radius (r ef ),
of about 1μm or less, this pressure difference can be quite important (around 20
to 30%) of the vapor pressure. However, for large porous, this difference can be
negligible (Peterson 1994).
Several other effects can influence the liquid–vapor interface such as the temper-
ature. Temperature rises are typically associated with surface tension decrease. In
the limit, when the density of the saturated liquid and vapor are the same, the surface
tension tends to zero. The following expression, to correct the temperature influence
2.2 Liquid–Vapor Phase Change Model 35

on the surface tension (in Newtons per meter) for pure water in contact with its vapor,
is proposed by the International Association for the Properties of Water and Steam
(1994) (see Ghiaasiaan 2008):
    
T 1.25 T
σ = 0.238 1 − 1 − 0.639 1 − (2.65)
Tcr Tcr

with T in Kelvin and where Tcr = 647.15 K is the critical temperature of the water.
This equation shows an almost linear relation of the surface tension with temperature,
as observed for most liquids. Jaspar (1972) proposed linear curve fittings for many
substances, among them several working fluids of heat pipes and thermosyphons.
As stated by Reay and Kew (2006), the following expression, proposed by Fink
and Leibowitz, is valid for liquid metals:
 
T n
σ = σ0 1 − (2.66)
Tcr

where, for sodium for instance,: σ0 = 0.241 N /m, n = 1.126 and Tcr =
2503.7 K.
The presence of one or more other substances (solute) in the working fluid
(solvent) can also have significant influence on the surface tension. Typically, the
solute are surface active and decreases the surface tension, as they concentrate at
the liquid–vapor interface (Carey 1992). Solutes that highly enrich the liquid–vapor
interfaces are called surfactants. A typical case is soap in water.
Another important factor to be considered is the Marangoni effect (Carey 1992),
which is observed mainly in zero or close to zero gravity conditions. As the surface
tension is a function of both temperature and species concentrations, variations of
any of these factors over the liquid–vapor interface can cause motion of the liquid
close to the regions where the surface tension is larger. To understand this effect, one
can consider, for instance, a vapor bubble in a pool, in microgravity conditions (no
natural convection), where one of the pool walls are heated and another one cooled.
A close examination shows that the surface tension over the bubble surface portion
that is in contact with the colder liquid is larger than that in contact with the hotter
liquid. Therefore, the liquid–vapor pressure difference through the coldest interface
area is larger than that of the hottest area. As a result, the bubble moves towards the
hottest wall direction. This effect provokes a liquid force against the density gradient.
In the case of heat pipes operating at zero gravity, a fine vapor layer over a heated
wall may be formed due to this effect, decreasing sensibly the heat transfer rate from
the solid surface.
Hysteresis Effect
The contact angle hysteresis is an important parameter that should be considered in
the design of heat pipes and thermosyphons. This hysteresis is directly related to
the variation of the liquid–solid contact angle, as a result of the direction in which
forces act over the liquid. A typical example is a drop attached to a vertical wall
36 2 Physical Principles

Fig. 2.15 Liquid plug

sustained by capillary forces

under the influence of gravity, as illustrated in Fig. 2.15. The affinity between liquid
and solid (tension at the solid-liquid interface) makes the liquid drop to be attached
to the vertical wall. The action of the gravity force pushes the drop down, decreasing
and increasing the superior and inferior contact angles, respectively.
Taking another example, some liquid drops can stand within straws, even when
subjected to gravity forces, due to the hysteresis effect. If the liquid is able to wet the
straw wall, the hysteresis effect of liquid–solid contact angle due to gravity action
causes the contact angle between the liquid drop and the straw to be different in the
upper (θsup ), and lower (θinf ) liquid–solid interfaces. Consequently, the superior and
inferior liquid–vapor meniscus have different curvature radius, r sup > r inf , resulting
in an upward surface tension force that acts against downward gravity. The meniscus
geometry of a drop within a straw is also shown in Fig. 2.15. Using Eq. 2.59, the
superior meniscus capillary pressure, considering that the liquid wets perfectly the
solid material (cos θ = 0) is 2σ/rsup . For the lower meniscus, the capillary pressure
is 2σ/rinf . The liquid force, resulting from this capillary pressure unbalance, coun-
terbalances the hydrostatic pressure of the liquid plug of height H, resulting in the
following expression, where the vapor density, which is much smaller in comparison
with the liquid, is neglected:
 
1 1
2σ − = ρl gH (2.67)
rsup rinf

In other words, this expression shows that the difference in the radius of curvature
is responsible for the force able to support the weight of the liquid plug.
2.3 Fluid Flow 37

2.3 Fluid Flow

Hagen-Poiseuille Flow
Many models concerning the flow behavior of working fluid in thermosyphons and
heat pipes are based on the Hagen-Poiseuille flow model, briefly described in this
section.
Figure 2.16 shows a schematic of a steady state laminar flow of an uncompressible
fluid in a circular cross section tube with axial velocity u (variable in the r direction),
and a differential element where the forces are sketched, where τr is adjacent layer
shear stress (variable in the r direction). The Hagen-Poiseuille flow hypothesis must
be observed: the radial velocity component v(r) and the gradient of the tube axial
velocity in the axial direction ∂u/∂x are both zero. Considering the flux momentum
null and that only shear and pressure forces act over the differential element, the
following partial differential equation results from a force balance:

d dp
(rτr ) = r (2.68)
dr dx
Applying the Newton law of viscosity τr = −μdu/dr the last equations turns to:
 
μ d du dp
r = (2.69)
r dr dr dx

Fig. 2.16 Flow velocity profile in a Hagen-Poiuseuille flow and force balance in a differential fluid
element
Exploring the Variety of Random
Documents with Different Content
"The school-girl who sends for the autograph of a public man pays
him a graceful compliment, and he should write it for her without a
murmur."
"Just in the same way, the whole public offers a quiet ovation to the
man of reputation when an interviewer presents his card. The
newspaper would never ask for an interview to publish, unless the
masses of its readers desired it. And the interviewer should be met
courteously, and the public man should realize that this sort of thing
is the duty he pays on fame. If he has positively nothing of interest
to say to the interviewer, or is too busily engaged to be interrupted,
he should tell the caller so in a respectful and polite manner. Many a
public man is badly treated by the reporter in print, because he
treated the reporter badly in his house."
"But what have you to say of the interviewer who is well treated,
and then repays the hospitality he has received by an article bristling
with ridicule and untruthful misrepresentations of the personality or
conversation of his entertainer? I have known this to occur."
"I do not believe it occurs very often," Homer answered. "When it
does, there is usually personal malice at the bottom of it, or a
catering to the lowest order of scurrilous journalism. It is a great pity
that the victims in such cases have no dignified redress. A thorough
caning ought to be considered consistent with the situation. But, I
think, as a rule, respectable newspaper men endeavor to do the
right thing by those who have treated them with courtesy in this
matter. The trouble is, journals are not careful enough in the
representatives they send on these commissions. It requires a great
deal of delicate tact to write acceptably of a man's home-life and
personality during his life-time. No thoughtless boy, or sensation-
seeking reporter, should be commissioned with such a task. I
positively know a New York journalist, who possesses a bright mind
and wonderful command of language beside an easy and elegant
deportment, who considers it fair play to gain information through
private letters or confidential conversations with his friends, and
then to use such knowledge for press purposes. He boasts of his skill
in this respect."
"Impossible!" cried Percy, indignantly.
"Quite too possible," Homer replied. "His devotion to journalism, and
his desire to feed the public appetite, has destroyed every particle of
moral principle the fellow ever possessed. Of course, such a man
reflects discredit upon the whole profession. That he is an exception
to the rule, I know, but that he is retained at all upon a respectable
journal, is to be regretted."
"There is still another feature of American journalism to be more
regretted and blushed for, I think," said Percy. "That is, the attitude
of our so called humorists and paragraphers toward public women.
No where else in the world do women occupy so exalted and
honored a position as they occupy in America. No other women in
the world have accomplished so much in various public callings. Yet
no where else are they subjected to such insults as they receive
from the newspapers throughout the United States, from the prima
donna to the President's wife, sister, or daughter."
"Are you not a little extreme in that statement, Mr. Durand?" asked
Homer Orton. "You must recollect that the royal family are discussed
very freely in print, and ladies who have become famous ought to
consider themselves members of the royal family of Genius, and take
newspaper criticisms as a natural consequence."
"It is not newspaper criticisms to which I refer," answered Percy. "Of
course, half the success of an actress, a singer, an author or a
painter depends upon public criticism, and often it happens that the
severer the criticisms the greater the success. But it is the loose
familiarity and the coarse jests of the item-seeker of which I speak.
Only last week I saw a wretched little item, intended to be
humorous, but actually brutal, going the rounds of the press,
concerning the advanced years of a famous opera singer, a woman
who has reflected credit on our nation by her brilliant and stainless
career."
"I saw the item to which you refer," Mr. Elliott said, "and I wondered
if it was consistent with the National boast that Americans are the
kindest and most thoughtful men in the world toward ladies. It
seemed to me an uncalled-for and ungentlemanly incivility toward a
noble lady."
"I often wonder," continued Percy, "if the fellows who perpetrate
those things stop and consider that the public women, whose names
they use so freely, are somebody's sisters, wives, or mothers, and
that, in nine cases out of ten, they lead a public life, or first entered
a public career, to earn a living. If the newspaper men of the country
ever do take this view of the matter, I should think their first impulse
would be to shield and protect and help every self-supporting
woman in the land. At all events, I should think every sensible
journalist would realize that, while it is the province of the
newspaper to furnish able criticism on the voice of the singer, the
book of the author, the speech of the orator, it is not its province to
indulge in poor puns, or insulting comments on the age, the
personal defects, or the domestic life of the singer, author or
speaker. These things should be tabooed by respectable journals,
just as they are tabooed in respectable society. Our journalists
should be as careful, in their references to the private matters of
individuals in print, as they are in conversation in their parlors,
where scandalous or impertinent references to the absent would be
considered 'bad form.' Really, I do not understand how any of us
who read the daily papers dare boast of American chivalry."
"The chivalry of the average man," said Dolores, who approached
the group just at this moment, "consists in protecting a woman
against every man save himself. And now, gentlemen, we are to
have a recitation from Madame Volkenburg. Will you join us and
listen?"
 CHAPTER X.
A DISCOURSE ON SUICIDE.
NE day Mrs. Butler, Dolores, Percy, and several of their
friends went to visit the Latin Quarter—the ancient homes
of the Grisettes—a race rapidly becoming extinct.
"I have always wanted to visit this locality," Dolores said on
her way thither. "It is a phase of Parisian life which has possessed a
curious fascination for me."
"No doubt you have surrounded the Grisettes with a halo of
romance," answered Percy. "If so, it will vanish utterly as you
approach. What sort of beings do you fancy they are?"
"Physically, lovely sirens: mentally frivolous; morally lax, owing to
their education, no doubt. Just the style of woman to fascinate a
romantic student."
Percy laughed. "That is the prevailing idea," he said; "but it is wholly
unlike the reality, as you will see."
What Dolores saw, were groups of contented looking mothers, tidy
housewives, and comfortable young matrons. Women whose lives
were devoted to their homes and families. Universally neat, and
modest in appearance, but in no case strikingly attractive or
beautiful.
"There is every indication here of happy domestic life," Percy said.
"These women make good true consorts, and contented
companions. They exchange their culinary and housekeeping
accomplishments, and their loyalty, for a little affection, protection
and support. On the whole, they lead a very pleasant sort of
existence—while it lasts."
"Their position is far more enviable than that of the average wife,"
Dolores responded, "for if they are unhappy in their relations, they
can at least get away; and I have no doubt they receive more
devotion and loyalty than the majority of married women do. The
position of the latter seems to me the more humiliating of the two."
Percy regarded Dolores with a grave expression.
"You are a strange girl," he said. "Yet extreme as you are in your
ideas, there is much truth in what you say. I have very little respect
for the husbands of my acquaintance. And still I believe God meant
each man to possess one mate, and to be true to her in life and
death. That is my ideal of perfect manhood—though an ideal I never
expect to attain. There was a time when I imagined it possible—but
now I live for the pleasure of the hour, and waste no time in theories
or in moralizing. Life is too short. But of one thing I am sure, Miss
King—positively sure." He paused, and she looked up expecting
some serious remark. "And that is—that you are the most charming
companion in the world."
On their way back to the Avenue Josephine, they saw a beautiful girl
who had just shot herself in the breast, being conveyed to the
hospital. Her lovely features were distorted with pain, and her
agonizing groans, as they lifted her from the street where she had
fallen, were heartrending to hear. Later, as they sat in Dolores'
parlors, they all fell to discussing suicide.
"Terrible as it may seem," Dolores said, "I really cannot think it so
great a crime as many do. We are never consulted in regard to
coming into this world. Life is thrust upon us, and if, as in the case
of that poor girl, perhaps, it becomes an insupportable burden, I
cannot help thinking God will forgive the suffering soul that lays the
burden down. I have always felt the greatest sympathy for suicides.
It is a cowardly act, I own, yet it is a cowardice I can comprehend
and condone. And I think God will surely be as sympathetic as a
mortal."
"You know Dante's description of the Seventh Circle," suggested
Percy, "and the horrors which await the rash soul of a suicide:

"When departs the fierce soul from the body, by itself

Thence torn asunder, to the seventh Gulf
By Minos doomed, into the wood it falls,
No place assigned, but where so ever
Chance hurled it."

"But that was merely the poetic utterance of a visionary mind,"

Dolores answered. "No one in these days believes in a God who
could be guilty of such atrocious punishments for sin or error, as
Dante describes; and then I contend that in many instances suicide
is not a crime, it is merely a cowardly act."
"But laying aside the crime of the act, think what an uncomfortable
position the poor soul may find itself in!" suggested Percy. "To go
where we are not wanted or invited, in this world, is a very
embarrassing situation, you know. And to suddenly thrust yourself,
without an invitation, upon the exclusive society of angels—I must
say I would not have the courage to do it."
"Well, of all things," said Madam Volkenburg, "if any of you ever do
commit suicide, never shoot yourselves, or resort to any disgusting
or painful process. I can tell you of a very swift and painless
method."
"What is it?" they all asked, in chorus, fascinated, as most of us
always are, by a discussion of the horrible.
All but Dolores. She already knew.
"Oh, it is a swift poison," Madam Volkenburg explained. "My
husband, who was a great experimenter in the chemical world, as
you perhaps know, left a package of it among his possessions. It is a
white, brilliant, crystallized substance, and the smallest particle of it,
the moment it mixes with the saliva of the mouth, and is swallowed,
produces instant death, and there is nothing to indicate poison
afterward. It cannot be detected, and it leaves the body quietly
composed as if sudden sleep had overtaken it."
"Why is it not better known?" some one asked.
"Perhaps it is well known; perhaps many of the sudden deaths by
'heart disease,' of which we read so often, occur in this way."
"And it will produce, as Madame Volkenburg says, swift and painless
death, at least upon an animal," added Dolores. "When my little dog
was run over by a carriage wheel, and lay crying in terrible pain, and
I knew he must die, I tested the efficacy of this poison upon him. It
ended his agonies instantly."
"And by the way," spoke Madame laughingly, turning to Dolores, "I
gave you enough of the poison to kill ten dogs, or human beings,
either; and you never returned it to me. Since I have heard your
views upon suicide I think I had better take the dangerous drug out
of your possession."
"If I were anxious to die, I fancy the absence of that drug would not
prevent me from finding the means of self-destruction," Dolores
answered, lightly. And at that moment refreshments were served,
and the conversation turned upon more agreeable subjects.
At the expiration of a month—the swiftest month of his life, it
seemed to Percy,—he was obliged to cut loose from this pleasant
circle, and visit London and Berlin, on the business which had really
brought him abroad.
He felt a curious depression of spirits as he entered the parlors on
the Avenue Josephine the evening preceding his departure—a
depression which he was hardly able to explain to himself. Only this
might be his last interview with his charming friends, and "last
times" are always sad.
Dolores seemed grave, as she welcomed him, and a little later she
said, with a winning frankness, "I never remember to have felt so
lonely at the thought of any other person's departure in my life, Mr.
Durand, as I feel at yours. You have been such an addition to our
circle; you are such a bon comrade; just what a brother would be, I
fancy. How I shall miss you!"
"But I am not your brother, you know," Percy said, and he wanted to
add, "And therefore the association has its dangers." He left it
unsaid, however, thinking she would understand his simple
assertion.
But she did not. She was a woman with a hobby, which precluded
the thought of marriage. And she was a cold woman by nature.
Knowing that Mr. Durand fully understood and respected her views,
she could see no danger in his companionship. She was very lonely
at the thought of his departure. He was her ideal friend, lost as soon
as found.
"It has been a charming month in Bohemia," Percy continued. "I
have thoroughly enjoyed it—it is unlike anything I have ever
experienced before. I have had my fill of conventional society, and I
have drained the cup of reckless pleasures; but this charming
mixture of refinement, esprit and abandon, has been a new element
to me."
"Apropos to your reference to a month in Bohemia," said Dolores, "I
believe Mr. Orton has written a poem on Bohemia, which he has
kindly promised to deliver this evening. Mr. Orton, will you favor us
now?"
"I was not aware that you were a poet, Mr. Orton," Percy remarked,
as the young man arose, and began to affect the bashful-school-girl
air.
"Sir," said the journalist, turning a stern look upon Percy, and
speaking in a sepulchral tone, "I am all that is bad: a newspaper
man, a poet, and—" pointing toward the piano, "the worst remains
to be told; I am a pianist." And then, quickly changing his expression
and voice, he recited in the most admirable manner the following
verses:
 BOHEMIA

Bohemia, o'er thy unatlassed borders

How many cross, with half-reluctant feet,
And unformed fears of dangers and disorders.
To find delights, more wholesome and more sweet
Than ever yet were known to the "elite."

Herein can dwell no pretence and no seeming;

No stilted pride thrives in this atmosphere,
Which stimulates a tendency to dreaming.
The shores of the ideal world, from here,
Seem sometimes to be tangible and near.

We have no use for formal codes of fashion;

No "Etiquette of Courts" we emulate;
We know it needs sincerity and passion
To carry out the plans of God, or fate;
We do not strive to seem inanimate.

We call no time lost that we give to pleasure;

Life's hurrying river speeds to Death's great sea;
We cast out no vain plummet-line to measure
Imagined depths of that unknown To Be,
But grasp the Now, and fill it full of glee.

All creeds have room here, and we all together

Devoutly worship at Art's sacred shrine;
But he who dwells once in thy golden weather,
Bohemia—sweet, lovely land of mine—
Can find no joy outside thy border-line.

"That is just the fear which disturbs my heart, as I am about to cross

the border-line and go back to the common-place world," sighed
Percy, when the applause which succeeded the recitation died away.
"I doubt my ability to enjoy anything, after this delightful
experience."
"Well, now," said Homer Orton, "in response to the encore I ought to
have received, I will give you a few verses appropriate to that
situation, my dear fellow. If you commit them to memory, they may
serve to help you in those dark hours of mental and spiritual pain
which come to every man—the morning after the club supper. They
are called—

PENALTY.

Because of the fullness of what I had,

All that I have seems poor and vain.
If I had not been happy, I were not sad—
Tho' my salt is savorless, why complain?

From the ripe perfection of what was mine,

All that is mine seems worse than naught;
Yet I know, as I sit in the dark, and pine,
No cup could be drained which had not been
fraught.

From the throb and thrill of a day that was,

The day that now is seems dull with gloom;
Yet I bear the dullness and darkness, because
'Tis but the reaction of glow and bloom.

From the royal feast that of old was spread,

I am starved on the diet that now is mine;
Yet, I could not turn hungry from water and bread,
If I had not been sated on fruit and wine."

"Speaking of Bohemia," Dolores said, "with all its charms, I do not

believe I am a Bohemian by nature. I am really fond of ceremonies
and imposing forms. I enjoy the most impressive services in divine
worship. Had I been reared in the Roman Church, I would have
made one of its most devout members. I like conventional life, but I
do not like the people I meet in those circles."
"And yet," Percy answered, "it is generally supposed that in exclusive
circles one finds all that is choice."
"But it is a great mistake," continued Dolores. "It may be true that
whatever is choice is always exclusive; but whatever is exclusive is
not always choice. One finds so little variety in the people one meets
in the so-called best society anywhere. They are all after one
pattern, and society does not tolerate individual tastes and ideas,
you know. So you see I am obliged to select my congenial friends as
I may, and create a Bohemia of my own."
"Which immediately becomes a Paradise," her listener answered
gallantly.
"Don't," ejaculated Dolores with a pained expression, "it sounds so
like—well, so like other men."
"And am I not like other men?" Percy asked, smiling and secretly
pleased. Nothing flatters a man's vanity more than being told he is
not like other men. "I never imagined myself to be a distinct type."
"But you are; or at least you have seemed so to me. And that is why
I have liked you so well."
"Then you do like me?"
Dolores met his gaze without a blush or tremor, frankly, sweetly.
"I do not think I ever met any man before, whom I so thoroughly
liked and respected," she said. "You are my ideal friend."
"Then, perhaps you will consent to correspond with me
occasionally," Percy suggested. "I should have gone away not daring
to ask the favor, believing myself only one of the many on whom you
bestowed your hospitality, but for your kind speech."
As he sat in his room that night, Percy puzzled his brains, trying to
analyze Dolores King's manner and words, and state of mind toward
him.
"She is either the most perfect actress, or the coldest and most
passionless woman on earth," he said, "incapable of any strong
emotion. Or else—or else—she likes me better than she knows. At all
events, it is fortunate for both, that I am going away."
 CHAPTER XI.
A FREAK OF FATE.
ERCY, who had long believed himself to be a perfect
cosmopolitan, quite as much at home in one part of the
globe as in another, was surprised to find that he was
actually homesick after leaving Paris.
With an impatience he could hardly understand, he awaited Dolores'
response to his first letter. When it came, full of bright humor and
sparkling cynicism, pleasant gossip and sincere expressions of regret
at his absence, Percy sat and smoked, and dreamed over it for more
than an hour.
He was trying to analyze his own feelings. When a woman does this,
ten to one she is in love. When a man does it, ten to one he is not.
Percy did not believe himself to be in love.
"At least," he mused, "I could never, even were I a marrying man,
contemplate marriage with Dolores King. She is too cold, too caustic,
too skeptical. In fact, she understands human nature too well. I
should want a wife who would idolize me, who would set me up as a
hero, to worship. I think many a man becomes a hero, through
having some woman over-estimate his worth. Rather than disillusion
her, he acquires the qualities with which her loving imagination has
invested him. Many a man has been saved from yielding to
temptation at the last moment, because he could not shatter the
perfect faith of some trusting heart. Dolores would not surround a
man with any halo. She sees us all as we are—perhaps exaggerating
our defects somewhat. She would suspect a man of evil on the
slightest provocation, and that is the surest way to drive a human
being into wrongdoing.
"But she is a delightful comrade, and so exquisitely beautiful that the
plainest room would seem elegantly furnished if she occupied it.
"She understands the art of entertaining. And time hangs heavy on a
fellow's hands, after he has lost her society. After all, life is too short
to relinquish any pleasure within our grasp, for fear of
consequences." And, rising and tossing aside his cigar, he added
aloud:
"With the Persian poet I can say,

"O threats of Hell, and hopes of Paradise,

One thing at least is certain, this life flies.
One thing is certain, and the rest is lies—
The flower that once has blown forever dies."

A few weeks later, Percy received letters from New York, requesting
him to visit London, there to complete business arrangements with a
large export house, and then to proceed to Copenhagen, where it
would be necessary for him to remain several months in the interest
of the firm.
When the letter arrived, he had just dispatched one to Dolores,
which closed as follows:
"I expect to return to America next month. I go with regret, and yet
no doubt it is for the best. It will cut short our delightful yet
dangerous companionship, but I trust you will permit me to call
upon you and say farewell before I go. In your last, you mentioned
the possibility of leaving Paris soon, but you did not tell me what
your plans were. Wherever you are, I shall, with your permission,
find you, before I sail for America."
What was his astonishment to receive in reply to his letter, the
information that Dolores, accompanied by Mrs. Butler and Madame
Volkenburg were about to start on a journey to the Land of the
Midnight Sun.
"We go direct to Moscow first," wrote Dolores, "stopping there long
enough to drop a tear on the tombs of the Czars; then on to St.
Petersburg; then by steamer down the Gulf of Finland and across
the Baltic to Stockholm; thence by rail to Christiania, where we may
linger some time, as Madame Volkenburg has dear friends there.
From Christiania we go direct to the North Cape. It is our intention
to return via Copenhagen and the Channels, as late in the season as
we can safely make the trip. We do not leave Paris under three
weeks; I hope you will call upon us before your return to America,
as you have promised."
When Percy read this he laughed aloud.
"It is fate," he said. "We are destined to be thrown together. I shall
proceed at once to Copenhagen, and when my charming friends
arrive in Christiania, I shall join them there and make the journey
with them to the North Cape."
It needed this bright prospect to keep Percy's heart cheerful after he
arrived in Copenhagen. Not a person in the city had hung out a sign
of furnished rooms to let; so finally he decided to advertise. After
waiting two days for the advertisement to appear, he rushed off to
the printing office to demand an explanation. The clerk remarked
calmly, that it had been lost, and as the next day was Sunday, he
would be obliged to wait until Monday. On Monday the notice
appeared, badly printed, in a column headed "Servant Girls Wanted."
During that day Percy found a room to his liking, on the
Tordenskjoldsgade, but as he feared an attack of lockjaw if he
attempted to direct any one to his lodgings, he chose apartments on
the Hovedvagtsgade instead. His breakfast, when served, consisted
of a cup of coffee and a cold roll. His dinner, for which he had a
ravenous appetite, was better enjoyed in anticipation than
participation. The soup was devoid of any extract of flesh, fish, or
fowl, but contained quantities of ginger, citron, lemon, and sugar.
This was followed by boiled fish, tasteless and watery, and
cauliflower swimming in sauce composed of milk and black pepper.
There were no side-dishes, and the eagerly-expected dessert
brought only disappointment and bread and cheese.
The next day, Percy was so curious concerning a mysterious plate of
soup which was served, that he made inquiries and learned the
actual ingredients. They consisted of carrots, potatoes, cabbage,
sugar, eels, cinnamon, cherries, plums, and small pieces of pork.
Another soup was made from the first milk of a cow; and what was
known as "beer soup," flavored with various ingredients, was
frequently served.
On inquiry, Percy found that other boarding-houses and hotels
furnished the same mênu, and he could only better his condition by
boarding at the largest hotel at an exorbitant price. Finally he
became reconciled to the fare: esteemed Limburger cheese as a
delicacy, and hailed the advent of every new kind of soup, as he
wrote home to his cousin, "with all the enthusiasm of a scientific
explorer."
His next achievement was learning how to sleep in a Danish bed.
The cot was so narrow, and so rounded in the middle, that if he
forgot himself and fell asleep, the covers were sure to slide off one
side or the other; and any effort to detain them resulted in his own
downfall. Finally, he concluded to lie under the feather bed, instead
of over it; and thus, braced by the wall on one side and two chairs
on the other, and the huge tick settling down over him, he
succeeded in wooing slumber.
After two months devoted to business in Copenhagen, he took
passage one autumn afternoon, in the steamship "Aarhus," for
Christiania, where he was to join Dolores and her party. Passing
through the "Kattegat," a severe wind rendered most of his
companions seasick, and Percy was almost the only one who
escaped the infliction. The next morning, one of the passengers
asked the captain if the storm had been a severe one. For answer he
simply pointed to the smoke-stack, which was encrusted to its very
summit with the salt from the waves which had dashed over it in the
night.
Percy stopped at the beautiful city of Gottenburg for a day, and
made a journey into the Northwest some fifty miles to visit the
famous falls of Trollhatton, which are unsurpassed in all Europe. In a
letter to his cousin that night he wrote as follows:
"On the little cluster of houses, which constitute the village of
Trollhatton, I was surprised to see in bold letters the name of a New
York sewing-machine company. I had seen the sign in France and
Germany, but I hardly expected to find it in this wild, unsettled
portion of Sweden. The same day, in traversing the vast, dreary,
rocky plateau which stretches from Lake Venern to the Skagerak, a
large, freshly-painted sign of 'Fairbank Scales' met my eye. But in
fact, where you find anything good over here in the way of
machinery, you may be sure it is from America.
"In all my travels through Germany I have never seen a reaper, a
mower, or a steel plow. Most of the grain seemed to be cut with a
sickle. In a very few instances I saw men using an awkward sort of
cradle; but they always threw their swath into the standing grain
instead of away from it, and had women follow behind with sickles
to pick it out and lay it in shape, so I did not see that they gained
much.
"It may be true that the American is somewhat given to bragging;
but when he comes to see the clumsy old-fashioned way of doing
things in Europe, and compares it with the methods at home, he
begins to feel that he has a foundation for his boasting. The best
fire-arms, the best cutlery, the best furniture, and the best tools, all
come from America. Even American cheese has found its way all
over Europe, and our various brands of tobacco are as familiar to the
European smoker, as to the Yankee himself."
Two days later found Percy enjoying a delightful interview with his
friends in Christiania; and the next day the happy quartette started
on their journey to the Land of the Midnight Sun.
 CHAPTER XII.
AN EXCITING ICE-BOAT ADVENTURE.
URING six delightful weeks of travel and sightseeing
through the wonderfully picturesque scenery of Sweden
and Norway, Percy was again the comrade and escort of
Dolores.
Day by day a thousand nameless acts of kindness and respectful
unobtrusive attentions, as thoughtful as they were delicate,
endeared him to the heart, at whose portal love, clothed in his most
ancient and most successful disguise of friendship, was effecting an
entrance.
It was late in November when the party returned to Copenhagen.
"My business matters will detain me here a week, possibly ten days,"
Percy said. "You will need that time to thoroughly enjoy the
Thorvaldsen and the Ethnological Museums—which are in their way
the finest in the world. Then I shall be ready to escort you to Paris,
before I report myself at London."
Madame Volkenburg returned to Christiania the day previous to the
intended departure of her friends. But at the expiration of the week,
just as Percy was planning to make an exit from the cold bleak
Island, Dolores sprained her ankle, and was unable to leave her
room during four weeks. Percy found business enough to employ a
few hours of each day in the interest of the London and American
houses, and the remainder of the time he passed agreeably in
entertaining the ladies. He read aloud, told interesting stories of
adventure and travel, and made himself so thoroughly charming that
Dolores forgot her misfortune in view of the happy hours it brought
her.
When she at length declared herself able to proceed upon her
journey another obstacle presented itself. The weather became
unusually cold; and the Sounds, surrounding the island on which
Copenhagen is situated, were packed with jagged blocks of ice, too
thick to be broken by a steamer, but not sufficiently connected to
make it safe for men or teams to venture on them. Our friends were
prisoners, consequently, upon an almost inaccessible island.
"The blockade cannot last forever," Percy said, when he had
informed the ladies of the condition of matters. "That is all the
consolation I can give you at present. It may last a week, or a
month, I am led to understand. In the meantime, we must enjoy
ourselves as best we can. I am very sorry Madame Volkenburg did
not remain with us, to share a little jaunt to Kaskilde—the ancient
Capital of Denmark, which we will make to-morrow."
"What is there to see at Kaskilde?" inquired Dolores.
"A cathedral, of course," Percy answered. "No doubt you are tired of
cathedrals, but this is a famous one: a relic of the ancient grandeur
of the city when it numbered 100,000 souls. Its population is less
than 5,000 now. You will find much to interest you there, as this
building has been the burial-place of nearly all the Danish kings."
Kaskilde was not more than twenty miles from Copenhagen, and
accessible by rail.
Dolores was surprised to find many of the tombs exquisitely carved
with marble and alabaster. One of the most interesting bore the life-
sized figure of Queen Margaret, who died in 1412. The beautifully-
portrayed features, full of expression, were declared to be a correct
likeness of the fair queen.
In the centre of the church upon a large iron slab set into the floor, it
was recorded that "This spot is purchased by Nils Jurgersen, of the
Church, as a resting place for his posterity for all time to come: in
order that his family need not change their burial-place every twenty
years, as other people do." But in spite of this sarcastic reference to
other people, the royal mandate went forth, that no more people not
of royal blood should be buried in the church. And Nils Jurgersen's
descendants are obliged to sleep out of doors, like "other people,"
after all.
High up in the nave of the church stood a huge clock. Before it two
half-sized figures carved in wood. At the end of each hour, the man
struck the time with a hammer upon the face of the clock: while the
quarter-hours were struck by the woman against a small bell.
"This little old couple have been faithful to each other during four
hundred years," said Percy, as he stood beside Dolores watching the
figures. "Is not that a wonderful illustration of constancy?"
"Yes," Dolores answered, laughing. "Such illustrations are readily
found, in wood. But how presumptive of man—to produce such an
example, when the Creator gave him no human precedent!"
"I must tell you about the clock," continued Percy. "Originally, there
were figures of St. George upon a horse, fighting the Dragon. Every
time the clock struck, the Dragon sprang upon the horse, and the
latter gave a wild scream. But there was an old priest who
complained that the noise of this battle disturbed him in his
preaching: so the Knight and the Dragon—wonderful pieces of
mechanism—were destroyed to please one conceited old egotist.
And, furthermore, he commanded that the faithful old couple should
be compelled to keep the Sabbath like other people. The machinery
of the clock was so arranged, in accordance with his wishes, that no
hours have been struck on the Sabbath since that time."
Hanging in a prominent part of the church, Mrs. Butler discovered a
painting which amused her greatly. It represented the devil, well
horned and hoofed, gazing sharply at the pews, in his hand a pencil
and a scroll. On the latter was inscribed: "I make a note of all those
who come late or go around and tattle."
"I wish I were able to purchase this painting and send it over to
America," Mrs. Butler remarked. "We need it there, I am sure."
At the expiration of two weeks, the blockade still continued. The
whole Baltic, as well as the North Sea, was one mass of floating ice,
which the powerful currents and tides in the connecting channels
kept in motion.
If the reader has not visited this portion of Europe, by glancing at
any map he will see that the Northwestern part of Denmark consists
of two islands. The Western is known as Funem, the Eastern as
Zealand.
The "Great Belt," as the channel between them is called, is from
fifteen to twenty miles wide in the narrowest portion, and is so
called to distinguish it from the channel between Funem and the
mainland, known as the "Little Belt."
In ordinary years, these straits remain sufficiently open, so that
steamers can cross regularly; or else they freeze solidly, allowing
sleighs to transfer freight and passengers.
But now, Copenhagen was entirely cut off from all communication
with the outside world.
Percy was told, however, that an effort was being made to carry the
mails across the Sound in a sort of ice-boat.
On investigation, he discovered that these ice-boats were in fact
large, strongly-built fishing-smacks, with iron runners on the bottom.
Each boat carried a crew of eight or ten weather-beaten old
fishermen.
"If you can convey the mails across the Channel in those boats, why
can't you carry passengers?" Percy asked as he stood inspecting the
smacks the day before their intended venture.
The men laughed, and gave him to understand in broken German—
the language he had used—that any man could go who had the
courage to make the attempt.
As he related this to Mrs. Butler and Dolores a little later, he said: "If
I had the least idea when navigation would open and permit you to
make your escape, I would go on the ice-boat to-morrow. Business
cares begin to weigh upon me heavily. But I do not like to leave you
imprisoned here for an indefinite time."
"Why could not we, too, go by the ice-boat?" suggested Dolores.
"Impossible!" cried Percy, aghast.
"By no means. We are experienced travelers, and the adventure
would be exhilarating after our long imprisonment here. If the crew
are opposed, I will go myself and talk them into consenting."
"Although they could not understand a word you speak, I know you
would win their consent to anything," laughed Percy. "But I will see
if the plan is practicable."
An hour later, he returned from a second interview with the ice-boat
crew.
"You can go," he said, "if your courage will sustain you. Reduce your
hand luggage to the smallest possible compass, and be prepared to
start for Korsör this P. M. at five o'clock. We remain there over night.
We take passage in the boat early in the morning. Two other
gentlemen are to accompany us, so we shall not die alone."
In the chilly dawn of the following morning our little party stood
wondering where they were to be stored in those queer-looking
smacks—one loaded with the heavier baggage, the other half-filled
with mail bags. The ladies were soon told to take seats in the rear
boat, among the mail bags; while the men were instructed to run
alongside, and to be prepared to spring into it at a moment's notice.
The crew pulled on a long rope attached to the prow of the boat,
and it gave a lurch forward.
For thirty or forty rods from shore, the ice was solid, and slanted
down toward the water. The boats glided along easily and rapidly.
The ladies laughed gleefully and enjoyed the novel mode of
locomotion.
All the crew, and the three gentlemen passengers, were provided
with huge straw overshoes, the soles fully two inches thick. These
served to protect their feet from the cold, and prevented slipping on
the ice.
"What rare good sport!" cried Dolores, looking like a Russian
princess in her furs, as she smiled up into Percy's face, while he ran
lightly along beside her.
"It is like the coasting days of childhood on a large scale."
Just then there came an ominous cracking sound, and suddenly the
forward boat crashed through the ice, which gave way for rods in
every direction. The rear boat went shooting down an inclined plane
into the water. The ladies shrieked, the crew shouted, the boat
turned over on its side, but was speedily righted.
Percy succeeded in springing into the boat before it reached the
open sea, but the other two passengers clung to its side, their legs
dangling in the icy water.
The forward crew threw out a long rope and a plank, and, getting
out on the ice, pulled the boat along a few lengths. The rear boat
was pushed along in its wake through the broken ice. As they
proceeded farther from the shore, the ice became more uneven.
Where it was strong, the crew propelled the boats by means of the
ropes; but where it was shaky or broken, the oars and boards were
brought into requisition. The old seamen found constant amusement
in the terrified screams of Mrs. Butler, every time they crashed
through the ice, while Dolores seemed to enjoy the excitement with
an almost childish delight.
Upon a sort of sand-bar in one place, which marked the boundary
between the fixed or land-ice, and the loose cakes floating in the
Sound, immense blocks had been crowded into all sorts of fantastic
shapes, forming an irregular rampart thirty or forty feet high.
Beyond this, the crew was able to keep the boat in the water most
of the time, winding in and out among the islands of ice.
Once, they were caught in a narrow strip of water between two ice-
floes.
Then the crew became excited, and hurriedly ran the boat up on the
ice out of harm's way. A few minutes later, the edges of the ice-floes
began to grind together and double up, impelled by the tremendous
currents underneath.
Dolores, who had grown very pale, while they were in this perilous
situation, shivered slightly, as she heard the grinding of the ice-floes,
and suddenly swayed back unconscious.
Percy reached out his arm just in time to receive her inanimate form.
The swoon lasted but a moment: yet during that moment Percy
experienced the delicious pleasure of holding her fair head upon his
shoulder, of clasping her lovely shape against his heart. All his well-
controlled emotions seemed to cry out against their long constraint;
and a sudden desire to seize her in his arms and cover her beautiful
face with kisses might have overruled his reason, his sense of
propriety and his good breeding, had she not opened her eyes and
drawn herself out of his arms.
"How foolish I am," she said. "But I really thought we were to be
crushed between those great ice jaws. I will not be so weak again."
"Please do," whispered Percy. "It was the happiest moment of my
life." His warm audible breath fanned her cheek; his eyes were full of
a fire she had never before seen in them; her blood tingled through
her veins, producing a slight intoxication. Her lids drooped, her
cheek crimsoned, but she did not rebuke him for his speech or his
glance.
A strange, sweet languor filled her heart, and rendered any
commonplace remark impossible. For the first time in her life she
was conscious of a vague pleasure in the close proximity of a human
being.
Out in the middle of the Sound they could see the black smoke of
the steamer "Absolem" awaiting them in a long strip of open sea. By
noon they were within a mile of her: and here the crew stopped
their boats in the middle of an ice-floe, and served a sort of Arctic
dinner.
An hour later, they reached the steamer; the mails and passengers
were transferred and the ice-boats returned to Korsör.
The "Absolem" had been two days and two nights in the ice, and
had narrowly escaped destruction among the sunken rocks, whither
it had been carried by the powerful ice currents.
But now, with an open sea before them, they approached the
western shore. As they neared Nyborg, Percy called the attention of
the ladies to the remarkable thickness of the land ice. And to their
surprise a few moments later, they saw the ship make fast to this
clearly-defined and solid ice pier, and unload its freight and
passengers as readily as if it had been in harbor.
Here they were huddled into Russian sleighs, and driven rapidly to
the station at Nyborg, where they took the train for Hamburg.
Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.

More than just a book-buying platform, we strive to be a bridge

connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.

Join us on a journey of knowledge exploration, passion nurturing, and

personal growth every day!

ebookbell.com