Professional Documents
Culture Documents
Download Liquid Vapor Phase Change Phenomena An Introduction To The Thermophysics Of Vaporization And Condensation Processes In Heat Transfer Equipment Third Edition Van P Carey online ebook texxtbook full chapter pdf
Download Liquid Vapor Phase Change Phenomena An Introduction To The Thermophysics Of Vaporization And Condensation Processes In Heat Transfer Equipment Third Edition Van P Carey online ebook texxtbook full chapter pdf
Download Liquid Vapor Phase Change Phenomena An Introduction To The Thermophysics Of Vaporization And Condensation Processes In Heat Transfer Equipment Third Edition Van P Carey online ebook texxtbook full chapter pdf
https://ebookmeta.com/product/introduction-to-transfer-phenomena-
in-pem-fuel-cells-1st-edition-bilal-abderezzak/
https://ebookmeta.com/product/heat-transfer-4-convection-two-
phase-flows-and-special-problems-1st-edition-michel-ledoux/
https://ebookmeta.com/product/elements-of-heat-and-mass-transfer-
second-edition-gupta/
https://ebookmeta.com/product/an-introduction-to-the-geography-
of-tourism-third-edition-velvet-nelson/
The Physics of Everyday Phenomena _ a Conceptual
Introduction to Physics (2022) TENth Ed 10th Edition
The Physics Of Everyday Phenomena : A Conceptual
Introduction To Physics
https://ebookmeta.com/product/the-physics-of-everyday-phenomena-
_-a-conceptual-introduction-to-physics-2022-tenth-ed-10th-
edition-the-physics-of-everyday-phenomena-a-conceptual-
introduction-to-physics/
https://ebookmeta.com/product/incropera-s-principle-of-heat-and-
mass-transfer-solutions-8th-edition-frank-p-incropera-theodore-l-
bergman-adrienne-s-lavine-david-p-dewitt/
https://ebookmeta.com/product/advances-in-nanofluid-heat-
transfer-1st-edition-hafiz-muhammad-ali/
https://ebookmeta.com/product/computational-fluid-dynamics-and-
heat-transfer-2nd-edition-majumdar/
https://ebookmeta.com/product/heat-transfer-evolution-design-and-
performance-1st-edition-bejan/
Liquid-Vapor Phase-Change
Phenomena
Liquid-Vapor Phase-Change
Phenomena
Van P. Carey
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made
to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all
materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all
material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not
been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any
future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or uti-
lized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopy-
ing, microfilming, and recording, or in any information storage or retrieval system, without written permission from the
publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.
copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-
750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organiza-
tions that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for iden-
tification and explanation without intent to infringe.
Part I T
hermodynamic and Mechanical Aspects of
Interfacial Phenomena and Phase Transitions
vii
viii Contents
xi
xii Preface
Nenad Miljkovoc, and Professor Jungho Kim has also helped guide revisions to update this text. An
expression of gratitude is also due to the many investigators who have contributed to this area over
the past 70 years. It is only through their combined efforts that a clear overview of the physics of
these processes is possible.
Van P. Carey
Berkeley, California
Nomenclature
A surface or cross-sectional area
Af fin area
Ao tube open area
Ap prime surface area
b fin height
Bo boiling number [= q ′′ /Ghlv ]
BoL Bond number [= g(ρl − ρv ) L2 /σ (where the length scale L depends on the circumstances
of interest)]
boiling number (= q n /Ghlv )
cpl liquid specific heat
cpv vapor specific heat
Ca capillary number [= µ lV /σ ]
Co convection number {= [(1 – x)/x]0.8[ρv /ρl ]0.5}
dd bubble departure diameter
dt tube diameter
dh hydraulic diameter based on wetted perimeter
(dP/dz)fr frictional component of two-phase pressure gradient
(dP/dz)l pressure gradient for liquid flow alone through tube
(dP/dz)le pressure gradient for entire flow as liquid through tube
(dP/dz)lo pressure gradient for entire flow as liquid through tube
(dP/dz)v pressure gradient for vapor flow alone through tube
D tube diameter
D1* binary diffusion coefficient for more volatile component
D12 binary diffusion coefficient for species 1 and 2
E mass fraction of liquid phase entrained in the core during annular flow
E ′′ rate of entrainment in mass of droplets per unit time per unit of wall area
f bubble frequency
friction factor
fl friction factor for liquid flowing alone in tube
fv friction factor for vapor flowing alone in tube
F Helmholtz function (= U – TS)
force
Chen correlation parameter
FTD Taitel-Dukler flow regime parameter
Frle Froude number [= G 2 /(ρl2 gD)]
g gravitational acceleration
specific Gibbs function
G Gibbs function (= H – TS)
mass flux through tube or channel
h local heat transfer coefficient
h mean heat transfer coefficient
ĥ specific enthalpy on per unit mass basis
h* mass transfer coefficient
hl heat transfer coefficient for the liquid phase flowing alone in the tube
hle heat transfer coefficient for entire flow as liquid
hlo heat transfer coefficient for entire flow as liquid
hlv latent heat of vaporization per unit mass
xiii
xiv Nomenclature
Hf fin height
jl volume flux of liquid [= G (1 − x ) /ρl ]
jv volume flux of vapor [= Gx /ρv ]
J flux of droplet or bubble embryos through size space; rate of generation of embryos per
unit volume
J* dimensionless droplet flux in size space
Ja Jakob number [= c p ∆T /hlv (where the choices of cp and ∆T depend on the circumstances
of interest)]
k B Boltzmann constant (= 1.3805 × 10 −23 J/K)
k d deposition coefficient in model of entrainment and deposition for annular flow
kl thermal conductivity of liquid
kv thermal conductivity of vapor
KTD Taitel-Dukler flow regime parameter
L tube length
Lb { }
bubble or capillary length scale = [σ /g(ρl − ρv )]1/ 2
Lf fin length
m mass of one molecule
m ′ mass flow rate of condensate in liquid film per unit width of surface
m ′′ mass flux
M mass
mass of one molecule
M molecular weight
NA Avogadro’s number (= 6.02 × 1026 molecules/kg mol)
Nl number of liquid molecules per unit volume
Nn number of embryos of n molecules at equilibrium per unit volume
N n′′ number of embryos of n molecules at equilibrium per unit of interface area
P pressure
Pc critical pressure
Pl ambient liquid pressure
Ppi(T) saturation pressure of pure component i in mixture at temperature T
Pr reduced pressure [= P /Pc ]
Pv ambient vapor pressure
Prl liquid Prandtl number
Prt turbulent Prandtl number (= ε M /ε H )
Prv vapor Prandtl number
q ′′ heat flux
′′
qcrit critical heat flux
′′
qmax maximum (critical) heat flux
′′
qmin minimum heat flux on pool boiling curve
′′
qmkc maximum heat flux limit dictated by kinetic theory for condensation
′′
qmkv maximum heat flux limit dictated by kinetic theory for vaporization
q total heat transfer rate
R ideal gas constant on a per unit mass basis
liquid jet radius
R universal gas constant (= 8.3144 kJ/(kg mol K))
Re Reynolds number
ReF film Reynolds number
Rel Reynolds number for liquid phase flowing alone [= G (1 − x ) D /µ l ]
Rele Reynolds number for entire flow as liquid [= GD /µ l ]
Relo Reynolds number for entire flow as liquid [= GD /µ l ]
ReL film Reynolds number [= 4 m ′ /µ l ]
Nomenclature xv
SUBSCRIPTS
a actual value
b bulk
bp bubble point
c properties evaluated at the critical point
dp dew point
ex exit conditions
f film
fin
i interface
in inlet conditions
l liquid
corresponding to the liquid phase flowing alone
le corresponding to the entire flow as liquid
corresponding to liquid flow in equivalent separate cylinder
lo corresponding to the entire flow as liquid
sat corresponding to saturation conditions
SSL supersaturation limit
v vapor
corresponding to the vapor phase flowing alone
ve corresponding to vapor flow in equivalent separate cylinder
w wall value
∞ far ambient conditions
0 wall value
Author Biography
Van P. Carey, a Professor in the Mechanical Engineering Department, holds the A. Richard Newton
Chair in Engineering at the University of California at Berkeley. Carey is widely recognized for his
research in the areas of micro- and nanoscale thermophysics, interfacial phenomena, and transport
in liquid-vapor phase-change processes. His research interests also include development of new
methods for computational modeling and simulation of energy conversion and transport processes.
Carey’s research has covered a variety of applications areas, including solar thermal power systems,
building and vehicle air conditioning, phase-change thermal energy storage, Rankine cycle power
for manned space missions, heat pipes for aerospace applications, high heat flux cooling of electron-
ics, energy efficiency of information processing systems, microgravity boiling, and nanostructured
surfaces for enhancing droplet evaporation and boiling processes.
Carey is a Fellow of the American Society of Mechanical Engineers (ASME) and the American
Association for the Advancement of Science, and he has also served as the Chair of the Heat Transfer
Division of ASME. Carey has received the James Harry Potter Gold Medal in 2004 for his eminent
achievement in thermodynamics and the Heat Transfer Memorial Award in the Science category (2007)
from the ASME. Carey is also a three-time recipient of the Hewlett Packard Research Innovation Award for
his research on electronics thermal management and energy efficiency (2008, 2009, and 2010), and Carey
received the 2014 Thermophysics Award from the American Institute of Aeronautics and Astronautics.
xix
Introductory Remarks
Liquid-vapor phase-change processes play a vital role in many technological applications. The
virtually isothermal heat transfer associated with boiling and condensation processes makes their
inclusion in power and refrigeration cycles highly advantageous from a thermodynamic efficiency
viewpoint. Liquid-vapor phase-change processes are also encountered in petroleum and chemical
processing, liquefaction of nitrogen and other gases at cryogenic temperatures, and during evapora-
tion or precipitation of water in the earth’s atmosphere.
In addition, the high heat transfer coefficients associated with boiling and condensation have
made the use of these processes increasingly attractive in the thermal control of compact devices
having high heat dissipation rates. Applications of this type include the use of boiling heat transfer
to cool electronic components in mainframe computers and the use of compact evaporators and con-
densers for thermal control of aircraft avionics and spacecraft environments. Liquid-vapor phase-
change processes are also of critical importance to nuclear power plant design, both because they
are important in normal operating circumstances, and because they dominate many of the accident
scenarios that are studied in detail as part of the design evaluation.
The heat transfer and fluid flow processes associated with liquid-vapor phase-change phenomena
are typically among the more complex transport circumstances encountered in engineering applica-
tions. These processes may have all the complexity of single-phase convective transport (nonlinear
effects, transition to turbulence, and three-dimensional or time-varying behavior) plus additional
elements resulting from motion of the interface, nonequilibrium effects, or other complex dynamic
interactions between the phases. Due to the highly complex nature of these processes, development
of methods to predict the associated heat and mass transfer has often proved to be a formidable
task. Nevertheless, the research efforts of numerous scientists over several decades have provided
a fairly clear understanding of many aspects of vaporization and condensation processes in power
and refrigeration systems. On the other hand, some elements of vaporization and condensation
phenomena are not well understood, and research in these areas continues.
There is a vast quantity of published information on liquid-vapor phase-change phenomena in
textbooks, monographs, and journal articles. Because such processes occur in a wide variety of
applications and because the thermodynamic, fluid mechanics, and heat transfer aspects of these
processes appeal to different groups of investigators, technical papers on various aspects of vapor-
ization and condensation processes are found in a number of journals, serial publications, and con-
ference proceedings. English-language publications frequently containing information related to
liquid-vapor phase-change phenomena include the following:
xxi
xxii Introductory Remarks
The above list is intended to provide a starting point for exploration of this area. There are many other
special conference proceedings and publications in applications areas such as electronics cooling,
automotive and aerospace applications that also have content relating to liquid-vapor phase-change
processes.
In addition to the sources listed above, useful information on liquid-vapor phase-change processes
may also be obtained from government agency reports (from, e.g., NASA, the National Bureau of
Standards, or the Nuclear Regulatory Commission) and the reports of research laboratories such as
Sandia National Laboratory, Argonne National Laboratory, and the Electric Power Research Institute.
The presentation of material in this text assumes that the reader is familiar with the basic ele-
ments of classical thermodynamics, fluid dynamics, heat transfer, and interfacial phenomena at
the level of typical coverage in undergraduate mechanical engineering or chemical engineering
programs. Throughout this book, facts from these fields will be recalled as needed for explana-
tion of phenomena of interest. It may be useful to refer to basic texts in these fields, if the reader is
unfamiliar with the concepts discussed. References [0.1–0.18] are useful resources for background
information in these areas.
A word regarding notation in this text is also warranted. The physical diversity of the mecha-
nisms involved in vaporization and condensation processes also makes selection of a consistent
nomenclature a difficult problem. For example, symbols traditionally used for properties in thermo-
dynamic analysis are commonly used to denote other physical quantities in fluid mechanics analysis
or in analysis of heat exchanger performance. To avoid confusion, every effort has been made to
make the definition of variables clear at the location in the text where they are introduced. In addi-
tion, a listing of the nomenclature for the text is provided in the back of the book to provide a quick
means of checking variable definitions.
In macroscopic treatments of systems in which liquid and vapor phases coexist, the boundary
between the bulk phases is usually idealized as a surface at which a discontinuity in properties
occurs. Interfaces between phases are invariably the locations where the net conversion of one phase
into the other occurs in vaporization and condensation processes. For that reason, the thermophysics
of the liquid-vapor interface is of primary importance in many phase-change processes in nature
and in important technological applications. We begin our exploration of liquid-vapor phase-change
processes in Chapter 1 by exploring the nanoscale features of the near-interface region in detail.
Introductory Remarks xxiii
REFERENCES
0.1 Van Wylen, G., and Sonntag, R. E., Introduction to Thermodynamics—Classical and Statistical,
3rd ed., Wiley, New York, NY, 1991.
0.2 Bejan, A., Advanced Engineering Thermodynamics, 4th ed., Wiley, New York, NY, 2016.
0.3 Wark, K., Advanced Thermodynamics for Engineers, McGraw-Hill, New York, NY, 1995.
0.4 Carey, V. P., Statistical Thermodynamics and Microscale Thermophysics, Cambridge University Press,
New York, NY, 1999.
0.5 Cengel, Y. A., and Boles, M. A., Thermodynamics—An Engineering Approach, 8th ed., McGraw-Hill,
New York, NY, 2014.
0.6 Moran, M. J., and Shapiro, H. N., Fundamentals of Engineering Thermodynamics, 7th ed., Wiley,
New York, NY, 2010.
0.7 Carslaw, H. S., and Jaeger, J. C., Conduction of Heat in Solids, 2nd ed., Oxford University Press,
Oxford, 1959.
0.8 Ozisik, M. N., Heat Conduction, Wiley, New York, NY, 1980.
0.9 Arpaci, V. S., Conduction Heat Transfer, Addison-Wesley, Reading, MA, 1966.
0.10 Bird, R. B., Stewart, W. E., and Lightfoot, E. N., Transport Phenomena, 2nd ed., Wiley, New York,
NY, 2001.
0.11 Kays, W. M., Crawford, M. E., and Weigand, B., Convective Heat and Mass Transfer, 4th ed.,
McGraw-Hill, New York, NY, 2004.
0.12 Howell, J. R., Siegel, R., and Menguc, M. P., Thermal Radiation Heat Transfer, 5th ed., CRC Press,
Taylor & Francis, New York, NY, 2011.
0.13 Incropera, F. P., DeWitt, D. P., Bergman, T. L., and Lavine, A., Fundamentals of Heat and Mass
Transfer, 6th ed., Wiley, New York, NY, 2006.
0.14 Mills, A. F., and Coimbra, C. F. M., Basic Heat and Mass Transfer, 3rd ed., Temporal Publishing, LLC,
San Diego, CA, 2015.
0.15 Baehr, H. D., Park, N. J., and Stephan, K., Heat and Mass Transfer, Springer-Verlag, New York, NY,
2006.
0.16 Miller, C. A., and Neogi, P., Interfacial Phenomena—Equilibrium and Dynamic Effects, 2nd ed.,
CRC Press, Boca Raton, FL, 2007.
0.17 Israelachvilli, J. N., Intermolecular and Surface Forces, 3rd ed., Academic Press, Waltham, MA, 2011.
0.18 Sadhal, S. S., Ayyaswamy, P. S., and Chung, J. N., Transport Phenomena with Drops and Bubbles,
Springer-Verlag, New York, NY, 1997.
Part I
Thermodynamic and Mechanical
Aspects of Interfacial Phenomena
and Phase Transitions
1 The Liquid-Vapor
Interfacial Region:
A Nanoscale Perspective
λ0 (1.1)
φ R (r ) = , 9 ≤ k ≤ 15
rk
where λ0 is a constant that varies depending on the type of molecule. Note that by definition, the
potential function is the energy that must be input to bring two molecules from infinite distance
apart to center-to-center spacing r. For a specified interaction potential ϕ, the force between
the molecules at any separation distance r is –dϕ/dr.
3
4 Liquid-Vapor Phase-Change Phenomena
At somewhat larger distances, the forces acting between molecules generally fall into one of the
three categories: (1) electrostatic forces, (2) induction forces, or (3) dispersion forces. Electrostatic
forces between molecules often arise because the molecules have a finite dipole moment (i.e., opposite
sides of the molecule have opposite charges). Common polar molecules (those having non-zero dipole
moments) include water and alcohols. Because the potential function for electrostatic interactions rap-
idly becomes small as r increases, dipole interactions are generally significant only at very short range.
Induction forces arise when a permanently charged particle or dipole induces a dipole in a nearby
neutral molecule. The strength of induction forces depends directly on how easily the initially neu-
tral molecule is polarized. The potential function associated with the induction force interaction
between a dipole molecule and a neutral molecule is inversely proportional to r6.
Dispersion forces are a consequence of transient dipoles that can be induced in molecules or
atoms. Such transient dipoles are mutually induced in adjacent molecules as a result of instanta-
neous asymmetries in the electric field due to moving electrons in each molecule. This type of force
interaction occurs in molecular species of all types and produces an attractive force between the
molecules. The associated potential function for dispersion force interactions is proportional to r–6,
λ DIS α 2p
φ DIS = − (1.2)
r6
where αp is the polarizability of the molecules and λDIS is a constant that varies with the type of mol-
ecule. Short-range attractive dispersion forces play an important role in the thermophysical behavior
of virtually all liquids and vapors near saturation conditions.
For real molecules, multiple force interaction mechanisms may come into play. Hence, the overall
potential function ϕ(r) is generally assumed to reflect the repulsive force behavior at small spacing
and the attractive behavior due to electrostatic, induction and/or dispersion forces at intermediate
distances. Several model variations of ϕ(r) have been proposed which more or less conform to this
general behavior. Perhaps the most well-known of these models is the Lennard-Jones 6–12 potential
which is plotted in Fig. 1.1. Appropriate values of the parameters ε and r0 vary with the type of molecule.
A study of all the implications of the Lennard-Jones potential, and other models like it, is
beyond the scope of this book. The interested reader can find more information on this subject in
references [1.1–1.5]. It is sufficient for our current purposes to note that this variation of the potential
implies that to bring two molecules that are initially very far apart into closer proximity, we must
remove energy. Conversely, if two molecules are close enough to feel attractive forces, but not so
close that repulsive forces come into play, then energy must be supplied to increase the spacing of
the molecules. Clearly, the energy exchanges associated with these processes are consistent with
the input or removal of the latent heat of vaporization during vaporization (moving closely spaced
molecules apart) or condensation (moving widely spaced molecules closer) processes.
The force interactions among molecules largely dictate the thermophysical properties of the sub-
stance. For fluids containing molecules that obey a spherically symmetric potential function ϕ(r),
statistical thermodynamic theory can used to generate relations among thermodynamic properties.
For a system of volume V containing N molecules with a spherically symmetric interaction function,
this type of theoretical model can be used to derive the following relation for the logarithm of the
canonical partition function:
where
∞
∫
av′ = −2π r 2φ(r )dr
D
(1.5)
and bv′ is a mean volume occupied per molecule in the system. In Eq. (1.4), kB is the Boltzmann con-
stant, h is Planck’s constant, M is the molecular mass of the molecule, ξ is the number of translational
and rotational molecular energy storage modes, σs is the symmetry number for the molecule, and θrot,m
is the mean rotational temperature for the molecule if it is a polyatomic species. In Eq. (1.5), D is the
effective diameter of the molecule. When the Lennard-Jones potential function is used, D is usually
taken to be about equal to r0. A simple model to relate bv′ to D is to treat the molecules as hard spheres
of diameter D and make two additional idealizations. The first is to interpret bv′ as the volume per mol-
ecule in the densest possible condition. The second is to take the densest condition to be cubic close
packing of the molecules. Geometry then requires that bv′ is related to the diameter D as
Note that the constant 1.350 in the above relation is the ratio of system volume to volume occupied
by molecules for cubic close packing of spheres. A complete discussion of the derivation of the rela-
tion (1.4) for ln Q for a van der Waals fluid can be found in Carey [1.1].
With the following relations from statistical thermodynamic theory, the equation for ln Q can be
used to generate relations for thermodynamic properties:
∂ln Q
P = k BT (1.7)
∂V T , N
∂(ln Q)
µ = − k BT (1.8)
∂ N V ,T
∂(ln Q)
U = − k BT 2 (1.9)
∂T V , N
6 Liquid-Vapor Phase-Change Phenomena
Substituting Eq. (1.4) into the equations above yields the relations
Nk B T a′ N 2
P= − v 2 (1.10)
V − Nbv′ V
ξN a′ N 2
U= k BT − v (1.12)
2 V
Equation (1.10) is the well-known van der Waals equation of state that can be reorganized to the
form
RT aˆ
P= − v2 (1.13)
vˆ − bˆv vˆ
where v̂ is the molar specific volume, NA is Avogadro’s number, R = NAkB is the universal gas con-
stant, and âv and b̂v are the molar van der Waals constants:
aˆ v = av′ N A2 (1.14a)
The relation for chemical potential can similarly be written in terms of molar specific chemical
potential and molar specific volume:
Equation (1.13) can be reorganized in terms of the mass specific volume v, leading to
RT a
P= − v (1.16)
v − bv v 2
Here M is the molecular mass, R = R / M is the specific gas constant, and av and bv are the mass-
based van der Waals constants:
av = aˆ v / M 2 (1.17a)
bv = bˆv / M (1.17b)
This model analysis thus leads to property relations for a van der Waals fluid. Although it incorpo-
rates many idealizations, the van der Waals model analysis provides a workable framework for pre-
dicting properties of dense gases and liquids that relates macroscopic thermodynamic properties to
The Liquid-Vapor Interfacial Region 7
molecular characteristics. In particular, it relates the van der Waals constants to the physical size of
the molecules and the intermolecular potential function that characterizes force interactions among
molecules. Consistency of the thermodynamic properties with the critical point conditions requires
that the van der Waals constants must be related to the critical pressure Pc and temperature Tc as
27( R / M )2 Tc2
av = (1.18a)
64 Pc
( R / M )Tc
bv = (1.18b)
8 Pc
The model analysis for a fluid of interacting molecules described above incorporates several ide-
alizations. While its property predictions do not exactly match the behavior of real fluids, they
are qualitatively consistent with the behavior of real fluids. Improved property predictions of this
type can be developed by incorporating more accurate non-symmetric potential function models
(see the discussions in references [1.2–1.5]).
Because the predictions of the van der Waals model are expected to be qualitatively consistent with
the behavior of many fluids, and the resulting relations are mathematically simple, we will use it here
to explore the main thermodynamic features of liquid-vapor phase change for a pure fluid system. The
van der Waals equation of state (1.16) can be used to predict the isotherm shapes on a P–v diagram.
Introducing the following reduced properties normalized with their critical point values
Tr = T /Tc (1.19a)
Pr = P /Pc (1.19b)
vr = v /vc (l.19c)
the van der Waals equation of state can be written in the dimensionless form
8Tr 3
Pr = − (1.20)
3vr − 1 vr2
Note that in Eqs. (1.19a)–(1.19c), temperatures are absolute (K) and vc is the critical specific volume.
Using Eq. (1.20), the variation of Pr with vr for fixed reduced temperature Tr defines a dimension-
less van der Waals isotherm on the plot of Pr versus vr . Reduced isotherms of this type are plotted
in Fig. 1.2.
Below the critical temperature (Tr < 1), the van der Waals model predicts that isotherms exhibit
a local minimum and maximum, as indicated in Fig. 1.2. For coexisting liquid and vapor phases,
thermodynamics theory dictates that the necessary conditions for equilibrium are that the tempera-
ture, pressure, and chemical potential in the two phases must be equal.
Tl = Tv , Pl = Pv , µˆ l = µˆ v (1.21)
In Fig. 1.2, this implies that the points corresponding to the reduced specific volume of liquid, vr ,l
and vapor, vr ,v must lie on a line of constant pressure (horizontal line) and be on the same isotherm.
The additional constraint that the chemical potentials be equal is implemented by using Eq. (1.15) to
8 Liquid-Vapor Phase-Change Phenomena
evaluate µˆ l and µˆ v and requiring that Psat = Pl = Pv be chosen so that the computed µˆ l and µˆ v values
are equal. This defines the link between the vapor pressure Psat and the specified temperature as
well as the temperature dependence of the saturation specific volumes.
It should be noted that if a liquid is depressurized isothermally in a quasi-equilibrium process
from initial state A in Fig. 1.2, classical equilibrium thermodynamics predicts that the state point of
the system follows the isotherm (constant Tr line) down until it reaches state point B. Upon reach-
ing point B, classical theory implies that further reduction in pressure can only occur after a phase
change occurs. Heat must be input to convert liquid into vapor as the state point shifts from B to F.
Once all liquid is converted to saturated vapor at point F, further reduction in pressure will cause
the state point to move along the isotherm toward point G. Note that this implies that at specific vol-
umes between vr ,l and vr ,v , the system contains a mixture of two coexisting phases: saturated liquid
at vr ,l and saturated vapor at vr ,v. Classical theory further implies that state points on the equation of
state isotherm between B and F are not observed in an equilibrium system. We will see, however,
that real systems in applications often depart from equilibrium, and it is sometimes possible for the
system to exist in some states on the isotherm between B and F. In fact, these departures from equi-
librium will be shown to play a central role in the onset of phase-change processes.
The relationship between the molecular behavior and macroscopic characteristics for the vapor
phase can, at least qualitatively, be understood from the kinetic theory of gases. Basic elements of the
kinetic theory are developed in Appendix I. Readers unfamiliar with at least the basic aspects of the
kinetic theory of gases are advised to review this appendix before proceeding further in this section.
In Appendix I, the following Maxwell-Boltzmann speed distribution for molecules in a gas is
derived from the kinetic theory of gases:
3/2
m 2
dN c = 4 πN c 2e − mc /2 k B T
dc (1.23)
2πk B T
The Liquid-Vapor Interfacial Region 9
The relation ε = (1/2)mc2 for the kinetic energy of a molecule can be inverted to obtain
Substituting Eqs. (1.24) and (1.25) into Eq. (1.23), the following energy distribution is obtained
In the above relation, dNε is interpreted as the number of molecules having kinetic energies between
ε and ε + dε.
It is often useful to know the fraction of the molecules in the gas that have energies exceeding
a specified value ε*. The number of molecules in the gas with energies above ε*, N>ε* , is given by
∞
N >ε* = dN ε
∫
ε*
(1.27)
The fraction of molecules with energies above ε* is just N >ε* / N . Substituting Eq. (1.26) for the
integrand in Eq. (1.27), dividing both sides by N and evaluating the integral yields
1/2
N >ε* 4 ε*
N
=
πk B T
e −ε*/ k BT + erfc ( ε* k B T ) (1.28)
When the threshold energy ε* is much larger than kBT, the complementary error function term will
be very small and can be neglected. For such conditions N>ε*/N is given by
1/2
N >ε* 4 ε*
= e −ε*/ k BT (ε* >> k B T ) (1.29)
N πk B T
Perhaps the most significant aspect of Eq. (1.29) is that it predicts that the fraction of molecules hav-
ing energies above the threshold value ε* increases rapidly with temperature. This behavior, which
is characteristic of the kinetic energy of molecules in liquids as well as in gases, plays an important
role in determining chemical reaction rates and the equilibrium conditions in two-phase systems.
The equilibrium vapor pressure and its variation with temperature in a system containing satu-
rated liquid and vapor are also consequences of the fact that the energy distribution among the mol-
ecules in the liquid and vapor phases is similar to the Maxwell-Boltzmann distribution. Because the
system exhibits this energy distribution, even at low temperatures some fraction of the molecules
in the liquid will have sufficient energy to escape the cohesive forces of other liquid molecules at
the liquid-vapor interface. If the kinetic energy distribution in the liquid is similar to the Maxwell-
Boltzmann result given by Eq. (1.29), the fraction capable of escaping in this manner will increase
rapidly with temperature.
The rapid increase of the equilibrium vapor pressure suggested by these arguments is character-
istic of most substances. This line of reasoning also suggests that at a given temperature, liquid with
a small cohesive energy will have a higher vapor pressure than one with a large cohesive energy.
Note that if a Lennard-Jones potential is used to model force interactions among molecules in the
fluids, the cohesive energy is expected to vary proportional to the depth of the potential well (ε in
Eq. (1.3)). Since the latent heat of vaporization is also a macroscopic indicator of the cohesive energy
10 Liquid-Vapor Phase-Change Phenomena
of the liquid, it follows that at the same temperature, a liquid with a high latent heat of vaporiza-
tion should have a lower vapor pressure than a liquid with a smaller latent heat. This trend is also
observed for most liquids.
It is clear from this development that the constant bv in the van der Waals equation is associated
with the volume occupied by the molecules themselves and the av/v2 term accounts for attractive forces
between the molecules. These attractive forces, sometimes referred to as van der Waals forces, may in
general be a combination of the dispersion, electrostatic, and induction forces described above.
Because the attractive forces represented by the av /v2 term are the same forces that must be over-
come to separate molecules that are initially closely spaced, the constant av is roughly proportional
to the latent heat of vaporization of the liquid. Note also that the contribution of the av /v2 term varies
strongly with the density (and therefore the spacing) of the molecules. For low-density gases v is
large and the contribution of av /v2 is small. For a liquid, the specific volume is very small and the
contribution of av /v2 will be much larger than for the vapor phase of the same substance.
These arguments can be made more concrete by considering a specific example. It can easily be
shown using density and molecular weight data from Appendix II that for saturated liquid water at
atmospheric pressure (101 kPa) the mean volume occupied per molecule is about 3.1 × 10 –29 m3. If
we envision the molecules as being evenly spaced in a cubic lattice, the mean spacing is approxi-
mately the cube root of the mean volume, or 3.1 × 10 –10 m apart. Since the diameter of a water
molecule is about 2Å (2.0 × 10 –10 m), the center-to-center spacing of the molecules is only about
1.5 molecular diameters in the saturated liquid.
For saturated water vapor at 101 kPa, the same line of reasoning suggests that the volume per
molecule is 5.0 × 10 –26 m3 and the molecular spacing is 3.7 × 10 –9 m or about 19 molecular diam-
eters. If r0 in the Lennard-Jones potential shown in Fig. 1.1 is taken as being about 1 molecular
diameter, then it is clear from these numbers that attractive forces will be very small in the saturated
vapor, but very important in the saturated liquid. This is completely consistent with the magnitude
of these forces implied by the av/v2 term in the van der Waals equation.
Example 1.1
To model nitrogen molecule force interaction with the Lennard-Jones potential (Eq. (1.3)), a value
of 1.31 × 10 –21 J is recommended for ε. Note that this sets the depth of the potential well (Fig. 1.1),
and the value of ε is the energy that must be input for one molecule to escape the attractive pull
of another. In a two-phase system, escape of molecules from a liquid phase into a vapor phase
at the interface is more probable if the translational kinetic energy is larger than ε. For saturated
nitrogen at 77 K, estimate the fraction of the molecules that have translational kinetic energies
larger than ε = 1.31 × 10 –21 J.
Equation (1.28) can be used to predict the fraction of molecules with translational energies
greater than ε. In the equation, we set ε* = ε = 1.31 × 10 –21 J and T = 77 K.
ε* 1.31× 10 −21
= = 1.23
kBT 1.38 × 10 −23 (77)
1/ 2
N>ε * 4ε*
N
=
πkBT
e −ε */ kBT + erfc ( ε* / kBT )
1/ 2
N>ε * 4(1.23)
N
=
π
e −1.23 + erfc ( )
1.23 = 0.366 + 0.117 = 0.483
Thus, the Boltzmann distribution predicts that almost half the molecules have translation energy
values greater than ε = 1.31 × 10 –21 J.
The Liquid-Vapor Interfacial Region 11
For molecules in the center of the large body of liquid, the attractive forces from surrounding mol-
ecules are nominally spherically symmetric and hence they balance out to zero. Near a liquid-vapor
interface, however, things are quite different. Within a few molecular diameters of the interface,
molecules in the liquid must redistribute themselves to accommodate the lack of spherical sym-
metry in the molecular force interactions. Liquid-vapor phase-change processes necessarily involve
the coexistence of a liquid and vapor phase, and the conversion of one phase to the other typically
occurs at the boundary between the bulk phases. This boundary region is usually of central impor-
tance in such processes. To gain a better understanding of the thermophysics of this region, in the
following sections we will examine its characteristics in detail. In the remaining sections of this
chapter we will consider it from a molecular perspective. In Chapter 2, we will examine how the
effects of this region can be treated from a macroscopic point of view.
FIGURE 1.3 Variation of the molecular density across the interfacial region.
In the directions parallel to the interface, however, the decrease in repulsion force between mol-
ecules produced by the increased spacing does not create a force imbalance because of the radial
symmetry of the force interactions. Consequently, there is no impetus to decrease the mean spac-
ing in this direction. The decrease in repulsive forces between immediate neighbors, with little
change in the longer-range attractive forces may thus produce a net tension force among molecules
in the interface region acting equally in all directions parallel to the interface. Although crude, these
arguments clearly imply that the existence of a net tension force on the molecules is a direct con-
sequence of the increased mean molecular spacing and the density gradient in the interface region.
As discussed in the previous section, the intermolecular attractions that give rise to interfacial ten-
sion may result from several different types of molecular forces. These may include forces that are
specific to particular types of molecules, such as the metallic bond or the hydrogen bond, as well as
dispersion forces, which exist in all types of matter and always give an attractive force between adja-
cent atoms or molecules no matter how dissimilar they are chemically. The dispersion forces vary with
the electrical properties of the substances involved and the distance between interacting elements, but
they are independent of temperature. While the magnitude of the attractive force between molecules
may vary, virtually all molecules exhibit long-range attractive force interactions and short-range repul-
sive interactions that can give rise to the increased interfacial free energy described above.
The characteristics of the interfacial region can be explored in a more quantitative way by extend-
ing classical thermodynamic analysis to the interfacial region with the idealization that properties
vary continuously across the region and that local thermodynamic equilibrium applies in a time
averaged sense within small control volumes within the interfacial region. This approach was pio-
neered by van der Waals [1.6] and is usually referred to as the van der Waals theory of capillarity or
the molecular theory of capillarity. It is also sometimes referred to as a mean field theory because
it is based on the idealization that the behavior of each molecule in a localized region is dictated by
the mean field associated with the surrounding molecules.
The Liquid-Vapor Interfacial Region 13
FIGURE 1.4 Variation of mean molar density and volumetric free energy ψ = F/V across the interfacial
region.
The classical van der Waals mean field theory of capillarity is based on the postulate that
the mean properties vary continuously across the transition region between the bulk phases
(see the discussions in Rowlinson [1.6] and Rowlinson and Widom [1.7]). Specifically, the van
der Waals model postulates a continuous variation of mean local density (number density ρn
or molar density ρ̂ = ρn/NA) across the interfacial region, as indicated in Fig. 1.4. The van der
Waals analysis leads to the conclusion that the interfacial tension is equivalent to the excess
interfacial free energy per unit area σ, which is the free energy per unit area above that for
step changes in ρ̂ and the free energy per unit volume at z = 0 (Fig. 1.4). The model postulates
that, in accordance with mass conservation and the second law of thermodynamics, the density
distribution spontaneously adjusts to minimize the total excess free energy (per unit area) in
the interfacial region.
In the van der Waals theory of capillarity, the z = 0 location in the interfacial region is chosen so that
0 ∞
∫
−∞
(ρˆ − ρˆ v ) dz + ∫ (ρˆ − ρˆ ) dz = 0
0
l (1.30)
Equation (1.30) ensures that the mass in the interfacial region with a distributed density profile is the
same as would exist in the region with a discontinuous density step change at z = 0. The free energy
per unit volume is defined as
ψ = F /V (1.31)
Here F is the Helmholtz free energy and V represents the volume of a local system within the inter-
facial region. The relation below stipulates that σ is the free energy per unit area of interface in
excess of that for step changes in ρ̂ and ψ at the interface at z = 0 (Fig. 1.4).
0 ∞
σ=
∫
−∞
ψ − ψ ( ρˆ v ) dz +
∫ ψ − ψ (ρˆ ) dz
0
l (1.32)
The van der Waals model of capillarity is postulated to apply to a system held at constant tempera-
ture with a volume that encompasses the interfacial region over a unit area of the interface. The
14 Liquid-Vapor Phase-Change Phenomena
second law of thermodynamics requires that for such a system (with fixed V and T), equilibrium cor-
responds to a minimum in the Helmholtz free energy. Since volume is fixed, this also corresponds
to a minimum in volumetric free energy ψ = F/V. This equilibrium free energy σlv is the property
commonly referred to as interfacial tension or surface tension.
Determination of the equilibrium interfacial tension using van der Waals theory thus
requires solution of a constrained minimization problem. We must determine the mean density
variation ρˆ ( z ) that minimizes the right side of Eq. (1.32). The interfacial tension or interfacial
free energy σlv is the value of the integral on the right side of Eq. (1.32) for ρˆ ( z ) that satisfies
Eq. (1.30) and minimizes the integral. To execute this scheme requires a means of predicting
ψ ( ρˆ , z ) in the interfacial region that accounts for the effects of the density gradient there. The
classical van der Waals theory of capillarity extends the statistical thermodynamics model
for properties of a van der Waals fluid described in the previous section to evaluate ψ and
other properties in the interfacial region. Once property relations are generated, one of the two
approaches can be used:
1. An empirical relation with an adjustable constant can be postulated for the density profile
and the integrals on the right side of Eq. (1.32) can be numerically evaluated for chosen
values of the adjustable constant to determine the value that minimizes σ.
2. The calculus of variations [1.8] can be applied to derive an integral relation for the mini-
mum σ and the density variation across the interface. Numerical evaluation of the integral
relation predicts the interfacial tension σlv and the density variation across the interfacial
region.
A more detailed discussion of the two methods described above can be found in Appendix III.
Using one of these methods, it is therefore possible to compute predictions of the surface ten-
sion and density profile in the interfacial region for a van der Waals fluid using Eqs. (1.30) and
(1.32), together with the Lennard-Jones model constants and other molecular parameters for
the fluid.
The classic van der Waals mean field theory of capillarity described above has generally been
viewed as an approximate theory. Its predictions are qualitatively similar to real fluid behavior,
but its numerical predictions are inaccurate. The predicted property values and the trends in the
variations of properties with temperature do not agree with those observed for real fluids. Near the
critical point, it is well known that the corresponding states correlation for surface tension that best
matches data for real fluids typically varies proportional to (1 − T/Tc) to an exponential power of
about 1.22 (see the discussion in Poling et al. [1.9]). The van der Waals mean field model predicts
that surface tension varies proportional to (1 − T/Tc)3/2. The van der Waals model also predicts that
the interfacial region thickness varies proportional to (1 − T/Tc) –1/2, whereas recent measurements
(by, for example, Beysens and Robert [1.10]) indicate that the interfacial region thickness varies
about proportional to (1 − T/Tc) –0.62. Despite its shortcomings, the van der Waals theory of capillar-
ity provides useful insight into the connection between molecular properties and interfacial region
thermophysics. In particular, it clearly indicates that interfacial free energy is a consequence of the
attractive force interactions among molecules and the density gradient that exists in the interfacial
region. However, the crudeness of the van der Waals model limits the usefulness of its quantitative
predictions.
In a recent investigation, Carey [1.11] has developed an alternative version of the molecular
theory of capillarity that incorporates a Redlich-Kwong model of fluid properties. The Redlich-
Kwong fluid model was chosen because it is a simple model and its predictions generally agree
better with real fluid saturation property data than predictions of the van der Waals model. The
formulation of Carey [1.11] used the calculus of variations method described in Appendix III
with properties determined using the Redlich-Kwong property model extended to the conditions
in the interfacial region. Using this model analysis, Carey [1.11] obtained integral relations for
The Liquid-Vapor Interfacial Region 15
the interfacial tension and the density variation across the interfacial region that can be cast in
the following dimensionless forms:
ρr ,l
σ lv 1.929 1 Pr ,sat ρr ,v (1 − ρr br )
Pc Li
= 1/4
Tr (1 − Tr )
0.17 ∫
ρr , v
(
1 + br ρr ) 3
− ρr Tr ln
− ρr Tr
ρr (1 − ρr ,v br )
(1.33)
1/2
ar ρr 1 + ρr br br Tr ρr ,v ρr aρ ρr ,v br
– ln – + r r 1 + ρ b dρr
3br Tr1/2 1 + ρr ,v br 1 − ρr ,v br 3br Tr1/2
r ,v r
ρr
z − zv 0.1639 ar P ρr ,v (1 − ρr br )
Li
= 1/4
Tr (1 − Tr )
0.17 ∫ (1 + br ρr ) r ,sat − ρr Tr ln
ρr , v +δρr
3
ρr (1 − ρr ,v br )
(1.34)
−1/2
a ρr 1 + ρr br br Tr ρr ,v ρr a ρr ρr ,v br
−ρr Tr − r 1/2 ln − + r 1/2 dρr
3br Tr 1 + ρr ,v br 1 − ρr ,v br 3br Tr 1 + ρr ,v br
ρr = ρˆ / ρˆ c (1.35)
ar = 3.84732 (1.36)
br = 0.25992 (1.37)
In the above formulation, Li is a characteristic length associated with the size of the interfacial
region. Note that to compute the dimensionless interfacial tension and density profile for a specified
Tr , the values of the dimensionless saturation properties must first be determined. The dimension-
less saturation pressure can be computed using the following theoretically based curve-fit to the
saturation pressure predictions for a Redlich-Kwong fluid [1.11]:
{ (
Pr ,sat = Tr exp −2.9327 Tr−3/2 − 1 )} (1.39)
The above relation predicts values of the saturation pressure that agree closely with full calculations
of Pr,sat for a Redlich-Kwong fluid. If the predicted Pr,sat value from Eq. (1.39) is used, the reduced
saturation density for the liquid and vapor are determined by solving the Redlich-Kwong equation
of state
3Tr a
Pr = − 1/2 r (1.40)
vr − br Tr vr ( vr + br )
for three density values that satisfy the equation for the specified Tr and computed Pr,sat. The
highest and lowest density solutions are the liquid and vapor saturation densities. The saturation
properties can be determined to higher accuracy if Pr,sat and the subsequent density calculations
are iterated to find the values that, when substituted in the chemical potential relations, more
accurately satisfy the equilibrium requirement that µˆ v / RTc = µˆ l / RTc (see reference [1.11] for
details).
16 Liquid-Vapor Phase-Change Phenomena
Carey [1.11] used Redlich-Kwong thermodynamic property relations to determine the saturation
properties, and then determined the dimensionless density profile and interfacial tension by numer-
ically integrating the dimensionless Eqs. (1.33) and (1.34). Asymptotic analysis of the Redlich-
Kwong property relations (1.33) and (1.34) indicates the following power-law dependence of the
surface tension at small 1 − Tr:
σ lv
∼ (1 − Tr )
1.33
Pc ( k B Tc Pc )
1/3
σ lv
= 14.65 (1 − Tr )
1.33
(1.41)
Pc ( k B Tc Pc )
1/3
closely matches the theoretical predictions computed by Carey [1.11] for Tr between 0.6 and 0.98
shown as the solid line in Fig. 1.5. Also shown in Fig. 1.5, are recommended surface tension values
for a variety of fluids from the ASHRAE Fundamentals Handbook [1.12]. It can be seen that the
recommended dimensionless surface tension values exhibit a power-law dependence proportional
to about (1 − Tr)1.25. In contrast, the van der Waals model predicts that surface tension varies pro-
portional to (1 − Tr)1.5. Although the agreement is not perfect, the modified Redlich-Kwong model
agrees fairly well with the recommended values and is a better fit than the classical van der Waals
model.
FIGURE 1.5 Theory prediction of interfacial tension for a Redlich-Kwong fluid computed using Eqs. (1.33)
and (1.34). Also shown are recommended values from the ASHRAE Fundamentals Handbook [1.12].
The Liquid-Vapor Interfacial Region 17
δzi
=
(ρr ,l − ρr ,v )
Li ( dρr dz )z = 0
(1.42)
Li
where the derivative is evaluated at the z = 0 location determined from the integral computation of
the density profile. Differentiating Eq. (1.34) to evaluate the derivative, this equation can be written
in the form
δzi 0.1639 ( ρr ,l − ρr ,v ) ar P ρr ,v (1 − ρr br )
= (1 + br ρr ) r ,sat − ρr Tr ln
Li Tr (1 − Tr ) ρr (1 − ρr ,v br )
1/4 0.17
Li 3
−1/2
(1.43)
ar ρr 1 + ρr br br Tr ρr ,v ρr a ρr ρr ,v br
−ρr T − ln − + r 1/2
3br Tr1/2 1 + ρr ,v br 1 − ρr ,v br 3br Tr 1 + ρr ,v br
z=0
It should be noted that the subscript z = 0 on the right side of Eq. (1.43) implies that the expression
in the square brackets is to be evaluated at the ρr value at z = 0. Full application of this relation thus
requires that the density profile be determined first to establish the ρr value at z = 0. Alternatively, a
good approximate value for δzi/Li can be computed using Eq. (1.43) if the ρr value at z = 0 is taken
to be (ρr,v + ρr,l)/2. This is a good approximation since the density at z = 0 is generally close to the
average for the two phases (Fig. 1.6).
FIGURE 1.6 Reduced density profiles across the interfacial region predicted for a Redlich-Kwong fluid at
various reduced temperatures.
18 Liquid-Vapor Phase-Change Phenomena
The variation of the interfacial region thickness with reduced temperature determined using
Eq. (1.43) with the ρr value at z = 0 determined from the density profile is plotted in Fig. 1.7. The
length scale Li used to normalize z in Figs. 1.6 and 1.7 is on the order of one nanometer for many
common fluids (Table 1.1). Consequently, the thickness of the interfacial region for many fluids is
predicted to be on the order of a few nanometers for temperatures corresponding to 0.6 ≤ Tr ≤ 0.98.
As indicated in Fig. 1.7, the variation of interfacial region thickness predicted by the theory is well
represented by the power-law relation
δzi
= 0.683 (1 − Tr )
−0.67
(1.44)
Li
This relation provides a means of estimating the interfacial region thickness for coexisting vapor
and liquid phases of a pure substance at a specified saturation temperature.
While not exact, molecular theories of capillarity provide a theoretical understanding of the vari-
ations of surface tension and interfacial region thickness with temperature that are observed in real
fluids. Note that the molecular capillarity theories described above indicate that interfacial tension
TABLE 1.1
Values of Li for Various Fluids
Tc(K) Pc (MPa) Li (nm)
N2 126.2 3.400 0.800
CH4 190.6 4.599 0.830
Ar 150.7 4.865 0.753
O2 154.5 5.043 0.751
H2O 647.3 22.129 0.739
NH3 405.6 11.290 0.793
C3H8 (propane) 369.9 4.248 1.063
SF6 318.7 3.760 1.054
R-134a 374.3 4.059 1.084
Hg 1763.2 151.0 0.544
The Liquid-Vapor Interfacial Region 19
and interfacial region thickness are only a function of temperature for a pure fluid. They indicate
that the surface tension vanishes at the critical point and varies about proportional to (1 − Tr)n where
n is about 1.2–1.5. They similarly indicate that the thickness of the interfacial region varies about
proportional to (1 − Tr) –m where m is about 0.5–0.6. The theories predict that the thickness of the
interfacial region grows without bound as T → Tc. These theories further imply that the variation
of these properties conforms to thermodynamic similitude. This suggests that relations among the
properties for different fluids can be reduced to a single relation in terms of properties normalized
with critical point parameters. This insight is useful in developing relations that can be used to pre-
dict surface tension values for different fluids systems.
Molecular theories of capillarity also predict the physical size of the interfacial region. From
Figs. 1.6 and 1.7 and Table 1.1, it can be surmised that the thickness of the interfacial region is on
the order of a few nanometers for a wide variety of fluids under commonly encountered condi-
tions. This suggests that for systems in which characteristic length scales are on the order of a
few nanometers, the properties associated with the interfacial region, including surface tension,
will be altered from those observed in an interface between two extensive bulk phases. This may
be particularly important in interfacial phenomena that occur in micro- and nanoscale system
applications.
Carey [1.11] also noted that the molecular theory of capillarity predicts that the center of the
interfacial region lacks intrinsic stability, suggesting that the interior of the interfacial region
may exhibit high levels of property fluctuations. As discussed in the next section, the interfacial
region property statistics for molecular dynamic simulations appear to be consistent with this
observation.
Example 1.2
For water at atmospheric pressure, use the Redlich-Kwong capillarity theory to estimate the sur-
face tension and the interfacial region thickness. Compare the surface tension value to the tabu-
lated value in Appendix II.
For water, from Table 1.1, Tc = 647.3 K, Pc = 22.1 MPa and Li = 0.739 nm. It follows that
= 14.65 (22.1× 106 )[1.38 × 10 −23 (647.3) / 22.1× 106 ]1/ 3 (1− 0.577)1.33
= 0.076246 N/m
The computed surface tension is within about 29% of the tabulated value of 0.0589 N/m in
Appendix II for water at these conditions. Also, the effective diameter of a water molecule is
estimated to be about 0.28 nm, which implies that the thickness of the interfacial region is about
3 molecular diameters.
Another noteworthy outcome from the molecular theory of capillarity is that it predicts specific
interrelationships among bulk fluid properties and interfacial region properties. Based on the results
of molecular capillarity theory for a Redlich-Kwong fluid, Carey and Wemhoff [1.13] derived the
20 Liquid-Vapor Phase-Change Phenomena
FIGURE 1.8 Variation of interfacial region thickness with reduced temperature predicted using recom-
mended property values from the ASHRAE Fundamentals Handbook [1.12].
following relation relating interfacial region thickness to temperature, interfacial tension and the
changes in molar density and internal energy between the two bulk phases:
ρˆ l − ρˆ v
δzi = 0.071L2i (1 − T / Tc )
−0.34
σ uˆlv (1.45)
lv
This relation can be used to predict interfacial region thickness from the other properties that can be
more directly measured. Figure 1.8 shows a plot of the interfacial region thickness predicted using
Eq. (1.45) for a variety of fluids. The data plotted in this figure were computed using saturation prop-
erties at several temperatures for each fluid. The saturation property data were taken from recom-
mended values in the ASHRAE Fundamentals Handbook [1.12]. The computed δzi values plotted in
Fig. 1.8 indicate that for a variety of common fluids under commonly encountered saturation condi-
tions, the theory described here predicts that the interfacial region thickness is in the range of 1–10
nm. This result, which is consistent with the observed trends in a similar model analysis devised by
Hey and Wood [1.14], is remarkable, given that it seems to apply to molecular species with widely
different molecular structures and interaction potentials.
Example 1.3
Use the property relation (1.45) to predict the interfacial region thickness for saturated nitrogen at
atmospheric pressure using property data from Appendix II.
For nitrogen, from Table 1.1, Tc = 126.2 K, Li = 0.800 nm and
A. WERNER.
CAPE GUARDAFUI
Native Life in East Africa
CHAPTER I
OUTWARD BOUND