Professional Documents
Culture Documents
Full download The classical Stefan problem: basic concepts, modelling and analysis with quasi-analytical solutions and methods Gupta S.C. file pdf all chapter on 2024
Full download The classical Stefan problem: basic concepts, modelling and analysis with quasi-analytical solutions and methods Gupta S.C. file pdf all chapter on 2024
Full download The classical Stefan problem: basic concepts, modelling and analysis with quasi-analytical solutions and methods Gupta S.C. file pdf all chapter on 2024
https://ebookmass.com/product/solved-problems-in-classical-
electromagnetism-analytical-and-numerical-solutions-with-
comments-j-pierrus/
https://ebookmass.com/product/applied-statistics-theory-and-
problem-solutions-with-r-dieter-rasch-rostock/
https://ebookmass.com/product/dynamic-system-modelling-and-
analysis-with-matlab-and-python-1st-edition-jongrae-kim/
https://ebookmass.com/product/classical-and-analytical-mechanics-
theory-applied-examples-and-practice-1st-edition-poznyak/
Basic Methods of Policy Analysis and Planning Pearson
eText 3rd Edition, (Ebook PDF)
https://ebookmass.com/product/basic-methods-of-policy-analysis-
and-planning-pearson-etext-3rd-edition-ebook-pdf/
https://ebookmass.com/product/roman-error-classical-reception-
and-the-problem-of-romes-flaws-basil-dufallo/
https://ebookmass.com/product/solutions-manual-to-accompany-an-
introduction-to-numerical-methods-and-analysis-james-f-epperson/
https://ebookmass.com/product/calibration-in-analytical-science-
methods-and-procedures-pawel-koscielniak/
https://ebookmass.com/product/understandable-statistics-concepts-
and-methods-12th-edition/
The Classical Stefan Problem
Basic Concepts, Modelling and Analysis with
Quasi-Analytical Solutions and Methods
The Classical Stefan
Problem
Basic Concepts, Modelling and Analysis with
Quasi-Analytical Solutions and Methods
New Edition
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or
mechanical, including photocopying, recording, or any information storage and retrieval system, without
permission in writing from the publisher. Details on how to seek permission, further information about
the Publisher’s permissions policies and our arrangements with organizations such as the Copyright
Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/
permissions.
This book and the individual contributions contained in it are protected under copyright by the Publisher
(other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our understanding, changes in research methods, professional practices, or medical treatment
may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and
using any information, methods, compounds, or experiments described herein. In using such information
or methods they should be mindful of their own safety and the safety of others, including parties for
whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any
liability for any injury and/or damage to persons or property as a matter of products liability, negligence
or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in
the material herein.
ISBN: 978-0-444-63581-5
Some symbols and abbreviations which are common throughout the book are given below. They have
also been explained in the text, occasionally, but not every time they occur. In addition to these symbols,
several other symbols have been used in the text, and they have been explained at the places they occur.
The parameters occurring in the equations could be dimensionless quantities or may have dimensions.
At those places where the equations are in the dimensionless form, the method of dimensionalization has
been mentioned or referred to reference indicated in the text.
For the notations used for the function spaces, the reader is referred to Appendices A–D. The same
have been explained in the text, occasionally.
Standard notations have been used for the numbering of equations, figures, definitions and proposi-
tions. For example, Eq. (7.2.7) refers to the seventh equation in the second section of Chapter 7.
xv
xvi List of Symbols
Subscripts
Greek Symbols
Abbreviations
The first edition of The Classical Stefan Problem: Basic Concepts, Modelling and Analysis
with Quasi-Analytical Solutions and Methods was published by Elsevier in 2003 as a volume
in the North-Holland Applied Mathematics and Mechanics book series. The main objective
was to discuss comprehensively, insofar as possible, the theoretical aspects of classical
formulations and analysis of some of the topics of the study of the Stefan problem.
The Stefan problem, which has some characteristic features, forms only a small part of
a bigger class of problems known as Free Boundary Problems. Even in 2003, the existing
literature on Stefan problems was so vast that it seemed feasible to discuss only classical
formulations of Stefan problems related to topics such as supercooling, variational inequality,
hyperbolic Stefan problems, inverse problems, existence and uniqueness and other aspects of
analysis. While discussing variational inequalities, inverse problems, analysis aspects, etc.,
the discussion of weak solutions was unavoidable, and they were discussed as needed and
not comprehensively. There are weak solutions, which are as good as classical solutions. To
bridge the gap between other basic sciences and mathematics, and to deepen the understanding
of the book’s contents, some definitions, theory and results from thermodynamics, metallurgy,
physics, applied mathematics, etc., were included as separate chapters.
A 5-year effort by the sole author produced the earlier edition. The positive reviews and
readers’ indirect encouragement provided the author inspiration and courage to undertake
this new project. This new edition features an extensive Chapter 12, which deals with
quasianalytical solutions and methods of classical Stefan and Stefan-like problems. Because
the class of Stefan-like problems is very large, only those problems whose formulations
are similar to those of Stefan problems and whose physics and formulations can be easily
explained have been included. Rather than publish the contents of Chapter 12 as a separate
book, with the aim of bridging theoretical and solution aspects of Stefan problems, Chapter 12
has been introduced along with earlier chapters.
A solution method is a procedure, and it is not confined only to Stefan problems. It
can be applied to any mathematical physics problem. However, for illustrative purposes, the
discussion is focused only on formulations of classical Stefan problems and some Stefan-like
problems. It is easier to explain the method with clarity in a concise way than the solution, as
describing the solutions requires too much space.
Chapter 12 is divided into 10 sections and each section into several subsections. First,
Section 12.1 begins with an overview of the aims, objectives, and contents of the chapter.
Some preliminaries, such as Green’s functions in various geometries, similarity variable, and
similarity solution, are discussed.
A lengthy Section 12.2 is devoted to exact analytical solutions pertaining to various ge-
ometries, including ellipsoidal and paraboloidal. Sections are assigned to various geometries,
different types of heat equations (such as with parameters depending on temperature and
xix
xx Preface to the New Edition
space variables), Stefan problems with kinetic conditions, equations with fractional derivatives,
multiple-phase problems and dilute binary alloy problems.
Section 12.3 is about series solutions of various types, including short-time solutions.
Section 12.4 deals with the analytical-numerical solutions of Stefan problems. Here the
term analytical-numerical is used for those solutions in which a complete analytical solution
cannot be obtained, and after some analytical derivation, numerical solutions are attempted
with the help of some suitable numerical schemes. The analytical derivation part should be
dominant and should have some variety if possible. Terms, such as semianalytical solution
and quasianalytical solution, are also used, but we prefer analytical-numerical solution. The
Adomian decomposition method, variational iteration method, integral equation approach and
regularization of Dirac-delta function are also discussed.
Section 12.5 is about analytical-numerical solutions of inverse Stefan problems. In
addition to the methods discussed in Section 12.4, the homotopy analysis method and some
regularization methods are also discussed.
The analytical-numerical solutions of hyperbolic Stefan problem are discussed in Sec-
tion 12.6, with the background information provided in Chapter 8. A rigorous background of
deriving Green’s function in the planar case is also briefly discussed.
Section 12.7 is about the use of complex variable methods in solving solidification/melting
problems and Hele-Shaw problems. The singularity development in suction problems, types
of singularity and its possible removal are described.
Approximate solutions and methods are discussed in Section 12.8. A major portion
of this section is devoted to the heat-balance integral method and its refinements and
variations, such as RIM, ARIM and hybrid methods. Weighted residual methods, such as the
Galerkin method and the orthogonal collocation method, are discussed briefly with only a few
illustrative examples. This section also discusses the first variation, variational principles and
the derivation of Euler’s equation for a given functional using calculus of variations. Finally,
the section describes the method of constructing a functional for a given problem whose first
variation or Euler’s equation will be the required differential equation.
A considerable amount of literature exists on perturbation solutions because of their
easiness in application. Therefore the emphasis in Section 12.9 is on applications of the
homotopy analysis method and the homotopy perturbation method. Regular perturbation and
singular perturbation methods applied to solutions of Stefan problems in various geometries
as well as a variety of formulations are discussed. Applications of the methods of strained
coordinates and matched asymptotic expansions are also illustrated.
Section 12.10 offers brief reviews of some supplementary references. Chapter 12 con-
cludes with an extensive bibliography of about 455 references.
The presentation of material in all 12 chapters is characterized by discussions based on the
thorough study of full-length research papers. The discussion includes my own comments
on many published works in Chapter 12. Reporting purely numerical solutions was never
the objective of the chapter, as that would require several separate volumes. However, for
analytical-numerical solutions, the highlights of numerical solutions and results are given very
briefly along with names of the software used if given in the referenced paper.
Invariably, the author thinks conceptually in terms of obtaining the solution first and then
devising a method to achieve it. This is why solution comes first in the book title, followed by
method.
Preface
This volume emphasizes studies related to classical Stefan problems. The term ‘Stefan
problem’ is generally used for heat transfer problems with phase-changes such as from the
liquid to the solid. Stefan problems have some characteristics that are typical of them, but
certain problems arising in fields such as mathematical physics and engineering also exhibit
characteristics similar to them. The term ‘classical’ distinguishes the formulation of these
problems from their weak formulation, in which the solution need not possess classical
derivatives. Under suitable assumptions, a weak solution could be as good as a classical
solution. In hyperbolic Stefan problems, the characteristic features of Stefan problems are
present but unlike in Stefan problems, discontinuous solutions are allowed because of the
hyperbolic nature of the heat equation. The numerical solutions of inverse Stefan problems,
and the analysis of direct Stefan problems are so integrated that it is difficult to discuss one
without referring to the other. So no strict line of demarcation can be identified between a
classical Stefan problem and other similar problems. On the other hand, including every related
problem in the domain of classical Stefan problem would require several volumes for their
description. A suitable compromise has to be made.
The basic concepts, modelling, and analysis of the classical Stefan problems have been
extensively investigated and there seems to be a need to report the results at one place. This
book attempts to answer that need. Within the framework of the classical Stefan problem
with the emphasis on the basic concepts, modelling and analysis, I have tried to include
some weak solutions and analytical and numerical solutions also. The main considerations
behind this are the continuity and the clarity of exposition. For example, the description of
some phase-field models in Chapter 4 arose out of this need for a smooth transition between
topics. In the mathematical formulation of Stefan problems, the curvature effects and the
kinetic condition are incorporated with the help of the modified Gibbs–Thomson relation.
On the basis of some thermodynamical and metallurgical considerations, the modified Gibbs–
Thomson relation can be derived, as has been done in the text, but the rigorous mathematical
justification comes from the fact that this relation can be obtained by taking appropriate limits
of phase-field models. Because of the unacceptability of some phase-field models due to their
so-called thermodynamical inconsistency, some consistent models have also been described.
This completes the discussion of phase-field models in the present context.
Making this volume self-contained would require reporting and deriving several results
from tensor analysis, differential geometry, nonequilibrium thermodynamics, physics and
functional analysis. I have chosen to enrich the text with appropriate references so as not to
enlarge the scope of the book. The proofs of propositions and theorems are often lengthy and
different from one another. Presenting them in a condensed way may not be of much help to
the reader. Therefore only the main features of proofs and a few results have been presented to
suggest the essential flavour of the theme of investigation. However at each place, appropriate
references have been cited so that inquisitive readers can follow them on their own.
xxi
xxii Preface
Each chapter begins with basic concepts, objectives and the directions in which the subject
matter has grown. This is followed by reviews—in some cases quite detailed—of published
works. In a work of this type, the author has to make a suitable compromise between length
restrictions and understandability. I have followed my best judgement in this regard. I hope the
readers will appreciate my efforts.
S.C. Gupta
Bangalore
Acknowledgements
I gratefully acknowledge the financial support from the Department of Science and Tech-
nology, Ministry of Science and Technology, Government of India, without which it would
not have been possible for me to undertake this book-writing project. The administration
of the Indian Institute of Science, Bangalore, and the Department of Mathematics, provided
infrastructural support for which I am extremely thankful. During my visit to the University of
Oulu, Finland, in 1997, I collected lot of material relevant for this book. I sincerely thank my
host, Professor Erkki Laitinen and Department of Mathematics, Oulu, for their kind hospitality.
The idea of writing this book nucleated during my visit to University of Rosario, Argentina, in
1991, and fructified 11 years later. I thank Professor D.A. Tarzia, Department of Mathematics,
Universidad Austral, Rosario, for sending me the bibliography prepared by him, and some
reprints which I could not get from other sources.
I am grateful to Professor Adimurthi, Department of Mathematics, TIFR, at IISc,
Bangalore, and Professor A.K. Lahiri, Department of Metallurgy, Indian Institute of Science,
Bangalore, for reading and responding to some portions of this book. My grateful thanks
to Dr. Thomas Chacko, Professor of English and Communication Skills, Department of
Management Studies, for style editing this volume with a great deal of attention. There are
many old friends and well-wishers who sent reprints and preprints that proved very useful for
me. Without naming them individually, I thank all of them for their selfless help.
I thank the Publishing Editor and staff of Elsevier and the Series Editor Professor
J.D. Achenbach of the book series ‘Applied Mathematics and Mechanics’ for their cooperation.
The excellent typing in Latex and drawing of figures in computer graphics were done by
Susheel Graphics, Bangalore. Without their cooperation the work would not have taken the
final shape.
I dedicate this New Edition to my wife whose constant inspiration and support helped
me in completing this long-term project. In the initial stages the help rendered by Dr Bryan
Davis, Solutions Production Director, Science and Technology Series, Elsevier and at the later
stage the constant cooperation given by the Editor, Dr A. Koch and the Project in charge,
Ms Amy Clark is gratefully acknowledged. The offer from the Chairman Department of
Mathematics, Indian Institute of Science, Bangalore, to use some of available facilities of
the Department and some contingency grant from the Centre of Continuing Education, Indian
Institute of Science, Bangalore, is an unforgettable experience for a retired Professor. Without
naming any individual, I thank the authors who sent me reprints. The production of the
manuscript was well handled by S&T book production centre at Chennai, Tamil Nadu, India.
xxiii
Chapter 1
The term Stefan problem can be best understood with the help of an example for which the
reader is referred to Section 1.3. Our concern in the present section is to understand the notion
of a free boundary which is a typical feature of the Stefan and Stefan-like problems. Therefore
this chapter begins with some examples of Stefan and Stefan-like problems which demonstrate
the existence of an unknown boundary, commonly known in the literature as a ‘free boundary’
or a moving boundary. In the context of solidification/melting problems, with which Stefan
problems are commonly associated, the free boundary is also called a phase-change boundary
or a melting front or a freezing front. Some authors use the term free boundary when the
unknown boundary is static and moving boundary when it is time dependent. In this volume
we do not make any distinction between ‘free’ and ‘moving’ boundaries. The term boundary
is used for a surface also. In many of the examples considered in this section, the identification
of the free boundary and the mathematical formulation of the problem are rather easy but in
some of them even the identification of the free boundary is difficult. The problems given here
are from various fields of mathematics, physics and engineering and demonstrate the existence
of free boundaries. Our interest in examples given in this section is more on demonstrating the
existence of a free boundary and its typical characteristics than justifying the formulation.
Problem 1.1.1 (Steady-State Heat Conduction With a Free Boundary). Find the steady-
state temperature T(x, y) satisfying the equation
∂ 2T ∂ 2T
+ = 0, (1.1.1)
∂x2 ∂y2
If f (x, y) is known throughout the plane, then one boundary conditions on R1 will be
∇ (T − f )|R1 = 0. (1.1.4)
The problem is to determine the temperature T(x, y) in D, and the unknown boundary R1 .
Two boundary conditions are to be prescribed on R1 . One, because it is a boundary and one
more boundary condition is required to determine an unknown boundary. Note that R1 can be
determined only by solving the system (1.1.1)–(1.1.4). Such an unknown boundary is called a
free boundary or a moving boundary. In ‘boundary value problems’ of mathematical physics,
the boundary of the region under consideration is completely known. Conditions (1.1.3)
and (1.1.4) are called free boundary conditions and (1.1.2) is a fixed boundary condition.
Although Eq. (1.1.1) is linear, free boundary problems are nonlinear problems because of
the nonlinearity of the boundary conditions at the free boundary.
Problem 1.1.2 (Steady-State Free Surface Flow With Surface Tension). Consider a two-
dimensional steady, incompressible, irrotational flow in a long channel. x and y axes are taken
along the length and depth of the channel, respectively, with the bottom of the channel taken as
y = 0 and the upper surface of water, as a free surface or a free boundary denoted by y = η(x),
where η(x) is unknown. If u(x, y) and v(x, y) are the velocity components in x and y directions,
respectively, then
∂u ∂v
+ =0 (equation of continuity), (1.1.5)
∂x ∂y
∂v ∂u
− =0 (irrotationality condition). (1.1.6)
∂x ∂y
where H is the given total water head of water, ρ is the density, g the acceleration due to gravity
and q is the velocity vector. If the bottom of the channel is a rigid boundary, then
v = 0, on y = 0. (1.1.8)
and
q · n = 0. (1.1.10)
Here, p̄ is the atmospheric pressure (known), σ is the surface tension (known) and Rc is
the radius of curvature of the free surface, taken as positive when the centre of curvature
is situated above the free surface and n is the unit outward normal to the free surface. Rc
can be expressed in terms of derivatives of η(x). Free boundary has been taken as static in
The Stefan Problem and Its Classical Formulation Chapter | 1 3
this problem. Eq. (1.1.10) implies that the normal component of the fluid velocity at the free
boundary is zero.
Problem 1.1.3 (Free Surface Flow With Time Dependent Free Surface). Let the free
surface in Problem 1.1.2 be time dependent and represented by y = η(x, t) with η|t=0 being
given. Surface tension effects will be neglected. If the velocity field q is expressed as
q = ∇φ, (1.1.11)
then from Eqs (1.1.5), (1.1.6) it is easy to conclude that φ(x, t) satisfies the equation
∂ 2φ ∂ 2φ
+ = 0, b(x) < y < η(x, t). (1.1.12)
∂x2 ∂y2
Here, y = b(x) is the equation of the bottom of the channel. The momentum equation can be
written as (cf. [3])
∂ q 1
− grad(p/ρ),
+ grad |q|2 − q ∧ curl q = F (1.1.13)
∂t 2
represents body forces. On substituting q from Eq. (1.1.11) in Eq. (1.1.13) and integrating
F
with respect to x, we obtain
∂φ 1
+ |∇φ|2 + gy + p/ρ = ψ(t), (1.1.14)
∂t 2
provided the density is taken as constant and the gravitational field is the only force field. The
arbitrary function ψ(t) can be absorbed in φ(t) and Eq. (1.1.14) becomes
∂φ 1
+ |∇φ|2 + gy + p/ρ = 0. (1.1.15)
∂t 2
∂φ
= 0 on y = b(x), (1.1.16)
∂n
where n denotes the unit outward normal to y = b(x). On the unknown free boundary
y = η(x, t), the two conditions are given by
∂φ 1
+ |∇φ|2 + gη(x, t) + p/ρ = 0, (1.1.17)
∂t 2
and
∂φ
= V · n, (1.1.18)
∂n
where V is the velocity of the free boundary and n is the unit outward drawn normal on it.
Eq. (1.1.18) can be expressed in terms of quantities already defined. Let,
Then
∂F ∂F ∂F
DF = dx + dy + dt = 0, (1.1.20)
∂x ∂y ∂t
or
∂F ∂η dx dy
− =− + = V · ∇F = (V · n)|∇F|, n = ∇F/|∇F|. (1.1.21)
∂t ∂x dt dt
or
∂η ∂φ ∂η ∂φ
=− + . (1.1.23)
∂t ∂x ∂x ∂y
Eq. (1.1.12) is to be solved using the boundary conditions (1.1.17) and (1.1.23), the fixed
boundary condition (1.1.16) and the prescribed η(x, t) at t = 0. In this problem the velocity
of the fluid is not time dependent but the free boundary is time dependent. Such problems
are called quasi-steady state free boundary problems or degenerate free boundary problems.
We shall see later that the term degenerate free boundary problem is used for other types of
problems also.
εf (x, t) can be considered as the disturbance at the flat surface y = y0 . Substituting Eq. (1.1.24)
in Eqs (1.1.12), (1.1.16), (1.1.17), (1.1.23), and in the changed equations retaining only linear
terms in ε and dropping higher order terms of ε, we obtain
∂u
= 0 on y = b(x), (1.1.26)
∂n
∂u
+ gf = 0 on y = y0 , (1.1.27)
∂t
∂u ∂f
= on y = y0 . (1.1.28)
∂y ∂t
In Eq. (1.1.27), we have taken p = −ρgy0 which comes from the contribution of the zeroth-
order terms in Eq. (1.1.17).
The Stefan Problem and Its Classical Formulation Chapter | 1 5
∂ 2u ∂u
+g = 0 on y = y0 . (1.1.29)
∂t2 ∂y
Since y = y0 is not a free boundary, only one boundary condition is to be prescribed on it. For
further information about Problems 1.1.2 and 1.1.3, see [3–5].
Problem 1.1.4 (A Problem of Reproductive Toxic Mass Diffusion). Let u(x, t) be the
concentration of a toxic mass which is diffusing in a region , where
= {x ∈ R : 0 ≤ x ≤ 1}. (1.1.30)
If the concentration exceeds a certain value uv in a portion of , then it is called a toxic region.
Let the reproduction rate of toxic mass in the toxic region be P and in the nontoxic region αP,
0 < α < 1. The toxic and nontoxic regions are separated by a surface S, where
and
u(x, t) and the free boundary x = φ(t) are to be obtained by solving the following system of
equations. For u > uv
For u < uv
The term d0 + d1 u accounts for the mass loss due to the bottom leakage, and other similar
factors. d0 , d1 , α and P are positive constants. Mass diffusion coefficient has been taken to be
unity which is possible by suitably defining the time and/or length scales.
and
u(φ(t), t) = uv . (1.1.36)
It may be noted that the velocity of the free boundary is not explicitly occurring in Eq. (1.1.35)
which was so in Problems 1.1.2 and 1.1.3. The boundary conditions at the free boundary
in which the velocity of the free boundary is not occurring explicitly are known as implicit
free boundary conditions. Eqs (1.1.33)–(1.1.36) are to be supplemented with a suitable initial
condition at t = 0 and with boundary conditions at the fixed boundaries x = 0 and x = 1.
Some results on the existence of solution of the above problem and the regularity of the free
boundary can be found in [6].
6 The Classical Stefan Problem
Problem 1.1.5 (Gas Flow Through Porous Media). The equation of state for an isentropic
(constant entropy) flow of an ideal gas in a homogeneous porous media is given by [7]
ρ(x, y) = ρ0 pα ≥ 0, (1.1.37)
where ρ(x, y) is the density and p(x, y) is the pressure. ρ0 ∈ R+ and α ∈ (0, 1] are constant.
The conservation of mass gives
∂ρ
= −γ
div(ρ V) , (1.1.38)
∂t
where γ is the porosity of the medium. According to Darcy’s law [8], the velocity V of the gas
flow in a porous medium is given by
β ∈ R+ is the permeability of the medium and η ∈ R+ is the viscosity of the gas. V and p can
be eliminated from Eqs (1.1.38), (1.1.39) and we obtain
∂ρ β
= 1/α
∇ 2 (ρ m ), ρ ≥ 0, (1.1.40)
∂t ηγρ0 (1 + α)
where m = 1 + 1/α. The diffusion is called ‘fast’ if m > 1, and ‘slow’ if m < 1.
By suitably choosing the time and/or length scales, the following equation can be obtained
from Eq. (1.1.40):
∂ρ
= ∇ 2 (ρ m ), ρ ≥ 0. (1.1.41)
∂t
If α ∈ (0, 1], then m ∈ [2, ∞). Eq. (1.1.41) is the porous media equation and it arises also in
other contexts such as population dynamics and plasma physics [7]. In order to calculate the
mass flux of the gas, the r.h.s. of Eq. (1.1.41) can be written as
mρ m−1 grad ρ is the mass flux and mρ m−1 can be taken as diffusivity. Diffusivity vanishes
as the density ρ tends to zero. Therefore Eq. (1.1.41) is a nonlinear degenerate equation in
the neighbourhood of any point where ρ = 0 but is nondegenerate and uniformly parabolic
(see [9] and Eq. (7.3.26)) in the neighbourhood of any point at which ρ is away from zero.
Such problems are called degenerate parabolic-elliptic problems. An important consequence
of nonlinear degeneracy is that there is a finite speed of propagation of a disturbance from rest
which is in contrast to the parabolic heat equation in which the speed of heat propagation is
infinite. The finite speed of propagation may give rise to waiting time solutions. Eq. (1.1.41)
is to be supplemented with an initial condition if the region is infinite and with both initial and
boundary conditions if the region considered is finite. The existence of a free boundary in such
problems can be illustrated with the help of the following example. Let
∂ρ
= ∇ 2 (ρ m ), −∞ < x < ∞, t > 0, (1.1.43)
∂t
The Stefan Problem and Its Classical Formulation Chapter | 1 7
> 0 for x ∈ RI = (a1 , a2 ), −∞ < a1 < a2 < ∞
ρ(x, 0) = (1.1.44)
= 0 for x ∈ R\RI .
This problem, generally, does not have a classical solution. The classical solution of a
problem can be roughly stated to be a solution in which the dependent variable possesses
continuous derivatives of the order required in the problem formulation. The mathematical
definition of a classical solution will be discussed later but at present it would suffice to state
that the solution ρ(x, t) of Eqs (1.1.43), (1.1.44) may not possess the required continuous
derivatives. For t > 0, gas will be diffusing to the right of x = a2 and to the left of x = a1 and
thus giving rise to two moving boundaries x = Si (t), i = 1, 2. Let S1 be moving towards +∞
and S2 moving towards −∞. Using a weak formulation of the above problem in Eqs (1.1.43),
(1.1.44), several interesting results on the behaviour of Si (t), i = 1, 2 have been obtained
in [10, 11]. The following proposition indicates that in some cases, the interface Si (t) starts
moving only after an elapse of time t∗ > 0.
Proposition 1.1.1. There exist numbers ti∗ ∈ [0, +∞) for i = 1, 2 such that Si (t) is strictly
monotone for t ∈ (ti∗ , +∞) and
If ti∗ > 0, then Si (t), i = 1, 2 remain stationary for ti∗ units of time [11].
In this case ti∗ is called a waiting time. It has been proved in [11] that the interface is
Hölder continuous under certain conditions and if the interface is in motion, then one expects
it to move with the velocity V of the gas, and
dSi m
= − lim Wx (x, t), i = 1, 2; W= ρ m−1 . (1.1.46)
dt x→Si (t)± m−1
The limits in Eq. (1.1.46) are taken as x approaches the boundary of the region from within the
region in which ρ(x, t) > 0. Eq. (1.1.46) can also be obtained from the mass balance condition
at the interface which states that the jump in the density at the interface multiplied by the
velocity of the interface is equal to the jump in the mass flux across the interface.
In a problem symmetrical with respect to x, it is sufficient to consider the region
0 ≤ x < ∞ with a single interface x = S1 (t) and a1 = a2 > 0. S1 (t) should satisfy
Eq. (1.1.46) for i = 1, and another condition to determine the unknown S1 (t) may be
prescribed as
S1 (t) a1
ρ(x, t)dx = ρ(x, 0)dx. (1.1.47)
0 0
Conditions of the type Eq. (1.1.47) are called nonlocal boundary conditions at the free
boundary. x = 0 is now a fixed boundary and the boundary condition on it is given by
∂ρ
= 0. (1.1.48)
∂x x=0
Note that we have two conditions prescribed at the free boundary, viz., Eqs (1.1.46), (1.1.47).
Another random document with
no related content on Scribd:
XXIV
MRS. POLAR BEAR'S ADVENTURE
The long, dark winter was on the wane. Months of cold starlight and
terrific winds, with numberless storms of heavy snow, had gone by.
Little by little the streak of light on the horizon, the thin shadows
which it cast over the snowfields, and the gentler quality of the air
increased; so that every one who lived in this far Arctic region stirred
in his winter sleep and there was preparation for a short and very
busy summer.
Some of the animals had been abroad, indeed, throughout the whole
dark night of the polar winter; such of them for instance as the lovely
white fox and the great polar bear. For it was not their custom to
crawl away, as many did, into the deep snow-banks, there to sleep it
out; for they knew that even this season of blackness and appalling
cold had plenty of food for them, and they were always insatiably
hungry.
But Mr. Bear's wife was of a different turn of mind, and although she
knew that her husband would not provide for her quite as she would
like to be fed, she was willing to go deep into the snow and dig out
for herself a warm bed away beneath the surface. There she had
stayed, never so much as venturing to the opening after the real
night had set in.
And there her cubs were born. Two of them there were. The good
Mrs. Bear was so delighted with their beauty that she was impatient
for the warm days to come when she could take them out and show
them to her relatives and friends.
"Perhaps, too, their father will be back by the time summer comes,"
she thought.
And then she was suddenly glad that he was not around just now; for
he was very quick-tempered, and if the babies annoyed him at all, he
would be pretty sure to cuff them. And one blow of Mr. Bear's paw
would finish the career of any baby bear in the world.
So the two little creatures, clad in the whitest of fur from head to foot,
their claws as black as ebony, and their wide eyes as yellow as
amber, lay snuggled against the great warm body of their mother for
all the weeks of the departing winter.
Suddenly, as they rolled over and looked upward through the snow
cavern, they saw for the first time what seemed to them a great big
eye staring down at them.
"That's only the hole in the roof," Mrs. Bear explained. "And pretty
soon you will see that it is all blue and beautiful above that window—
and then we will go out and away."
What that meant they did not know; for life so far as they had known
it consisted of meals and sleep and endless playtime on the icy floor
of their cavern. But they were to know more about it very soon. A
white wing flashed by one morning, and a land voice called down the
depths of their cave.
It was Mr. Burgomaster, the good-natured gull. He had come
purposely to call on Mrs. Bear, for he had two stirring pieces of
information to give her.
He perched by the edge of her skylight, and wasted no words in
relating the news.
"There's a whale being driven ashore; and the mists have hidden the
birds."
He was gone before Mrs. Bear could so much as thank him for
coming; and she was, indeed, deeply obliged. No one but good Mr.
Burgomaster would ever have taken such pains.
What he said sounded strange enough, but it meant everything to
Mrs. Bear. When a whale was disabled in the far depths of the sea,
or had been caught in the currents and gales in such a way that he
must surely drift to shore, he was as good as dead and devoured.
For in shallow water he would be helpless and once his enormous
bulk was stranded on the rocks or the jagged capes of ice he could
only give himself up to his enemies.
Mrs. Bear, however, would have been very cautious about venturing
to the scene of the banquet, if the great flocks of birds, which were
sure to be on hand, were not hidden from view as they hovered
above it. Clouds of excited gulls that came nearer and nearer to the
shore were a signal of what was about to happen. And the bears, the
foxes, and the wolves were not the only ones who knew it. Men, with
their ferocious packs of dogs, their long lassos of walrus hide, and
their terrible spears, knew well enough what the noisy birds were
announcing.
But all would be well if the fogs hung low, and the gathering flocks of
sea-birds were thereby hidden.
Mrs. Bear explained the situation to her cubs.
"Of course, your mother would not have built her nursery here," she
ended, "if she thought those terrible creatures with the wolfish dogs
and the ropes were within miles and miles of the spot. But you can
never tell when they may turn up. They come with their dogs over
endless tracks of snow and ice to find us, and they travel fast. You
must lie as quietly as you can while I am gone. Amuse yourselves in
only the quietest way. Don't call out at all; and go to sleep again, like
good children."
With that Mrs. Bear rose to her hind feet and reached upward along
the snow walls of her house. Then, balancing herself on a ridge of
the ice which was for all the world like a side shelf, she made a
ponderous leap through the opening into broad daylight. For at last it
was the real day, and a glorious glimmer of sunlight behind the fogs
showed that summer was coming.
It was good to breathe the free air, and Mrs. Bear shook herself
violently to straighten out the creases of her heavy coat. She would
have liked to roar, loud and long, but she was trained by experience
never to speak in a fog.
"You can't tell who's hearing you," her own mother used to say.
So she only trundled her mighty bulk downward across the ice and
snow, to its very edge, where it suddenly broke off and formed an
embankment. Below this there was a narrow beach, or what
appeared to be one—a strip of confused and tumbled blocks of ice
and jagged rocks.
There was a sudden whizzing of wings above her head, and the
wailing cries of a hundred little gulls and the many crowds of birds
that were hurrying to eat of whale fat. Mrs. Bear broke off in their
direction; and soon the sound of snarling voices, the yelps of the
quarrelsome foxes, and the vicious bark of the wolves met her ears.
Yes, she was none too early, for evidently the assemblage of
animals, all as famished as herself, were fighting over the repast.
They were not so polite to Mrs. Bear as they might have been, for
they begrudged her any share of the whale's body. But she paid little
attention to any one, and went to work lustily on her first meal of the
season.
After the first mouthfuls, however, she felt wonderfully good-
humored; for such is the effect of a meal, and it is pleasant to stop
and talk a bit when you know there is more to follow.
"I must thank you, Mr. Burgomaster," was her first remark. "You were
kind to call me in time. This is a good beginning to the summer."
The white-winged gull, largest of all the birds that were present, and
by far the best mannered, only begged Mrs. Bear to remember that
they had been friends for many years.
"And I propose to name my children," Mrs. Bear announced, as this
delicious dinner began to increase her fine spirits, "I propose to
name the babies after you and your wife: Odin and Olga. That's what
they shall be."
Mr. Burgomaster was at a loss how to express his gratitude for this
compliment. But he needed to say little, for such a generous tribute
is not repaid in words.
Something he said later on, however, in which he quoted Dr.
Penguin, brought forth her assent on the subject of eating too much,
for she added, "True, true, it is not wise to overeat at your first meal
of the year. A relative of mine did that once, and was unable to climb
over the path to his door."
So, taking as goodly an amount of provender with her as she could
carry away, Mrs. Bear went home to feed her babies. They were far
more interested in this new and appetizing breakfast than in the
names which she gave them, you may be sure; and from then until
the whale was used up and only his bones were left to dry in the
winds, Mrs. Bear was continually carrying meals to her cave.
By this time the winter was gone, and the roof of the snowhouse fell
in. The melting drifts drenched every ledge and cranny of their home,
and it was time to be wandering.
"You must do exactly as I tell you," Mrs. Bear kept saying, "and you
must never stray from me a minute. For we are going to start on our
journey, and there will be a great many dangers to guard against."
When little Odin and Olga trotted along beside their mother, with the
whole world before them, and a keen appetite with them, they were
as alert and excited as any two bears in the world could be.
The great rolling, blue water, the ice that floated on its surface and
shone like white ivory in the sun, the patches of green grass on the
sides of the hills, and the rocks black with snow water, made a
dazzling scene.
Their long day began with a wonderful feat on the part of Mother
Bear. After they had swum to a low, wide ice floe, which was a little
way from shore, and Odin and Olga were just learning to use the
hairy pads of their feet in climbing the sides of the small iceberg,
Mrs. Bear gave a sudden plunge into the water, and disappeared
from view. She swam far out, her nose barely coming to the surface,
and the rest of her body entirely concealed. Then, rising to the
surface, she brought back with her a huge fish which she had
stunned with a blow of her mighty paw.
"It's all in the way you slide into the water," she said; and then, as
they ate greedily of this morsel, she told them of diving for sea-lions
and of capturing them by coming up from under the prey.
"You will swim under water great distances, as soon as you learn to
hunt," she said, "and you will learn to make no noise about it."
This was the truth, as not only the seals and the sea-lions, but plenty
of the great fish, could bear witness.
But, as events of the day were to turn, little Odin and Olga were near
to never growing up at all; for the very danger which their mother
most dreaded was speedily approaching. While they were playing
first on the ice cakes and then on the shore, and Mrs. Bear had
about made up her mind that they would stay that night at a point not
far distant, where she saw many sea-birds fluttering, and where, she
reasoned, the fishing and seal hunting might be good, the hunters
with their trained dogs were fast approaching the very spot.
For your Eskimos have their own way of reading the signs; and as
many birds had been flocking in this direction, the men had steadily
pursued the trail. Day after day they had traveled, and they felt sure
that they were coming upon at least a herd of seals or of walruses.
And they hoped, of course, to bag a great white bear.
But Odin's mother had assured herself that there was no danger, or
it would have been revealed during the time that the whale had
attracted such crowds of her brother animals. She did not perceive
that her enemies knew exactly how prone the well-fed bear is to
linger near the spot of her recent feedings.
"That is just the place to spend the night, out there," she said; "for on
those points that reach out into the sea, you can escape by land or
by water, as you have to. Remember that, too, children."
Little Olga stopped to rub her head at this. She was trying to
remember so many things! Mrs. Bear told her it was nothing, and
that learning things was the whole of life anyway.
When Mrs. Bear and her twins reached the icy point, there were the
friendly Penguins to meet them and to exclaim over the children.
They were having a fine visit when suddenly a dull roar far below
them on the shore made every one sit up and listen.
It came again and lasted longer. It was a new sound to the children,
but Mrs. Bear recognized it.
"That's an iceberg breaking up," she said at last. "Not a pleasing
sound, but one you'll soon get used to."
Night came and they curled up, all three, in a snug corner under the
ice shelves of the point. The wind was high and the sea was noisy,
but they were too well tucked away to care.
And they little dreamed of what was going on around them.
For scarcely had the sun gone down, when the Eskimos with their
teams of wolfish dogs were on the spot. Little by little they had crept
to the end of the point, and one by one they stationed themselves at
intervals to wait, like so many sentinels, for the morning.
Mrs. Bear would never reach the water alive; and escape back to the
mainland was impossible. There were enough dogs and men on
hand to cover the avenues of escape.
Before little Odin and Olga were awake sufficiently to see anything at
all, Mrs. Bear had faced her first ambushed enemies. From where
the cubs cowered in their corner, they saw their mother rear on her
hind legs and then drop with a terrible force, hitting the dogs right
and left as she landed among them. There were thunderous noises,
and her own mighty roars were almost drowned by the snarling of
the dogs and the shouting of the men, who were fast closing in. She
was bleeding already and several of the dogs were lying dead
around her.
Mrs. Bear stood truly at bay. One man, more courageous than the
rest, came running up with his pointed spear, ready to take aim. A
terrific noise arrested him—a noise in which all else was nothing.
The land seemed to reel and topple; the great ice shelves came
crashing down.
Men and dogs ran for their lives; and to save themselves they
plunged bodily into the sea. For the whole point of ice had broken
from the mainland and, like a ship that is rocking and righting itself, it
was sending up mighty waves and eddies on every side.
The motions were less gigantic after a while, and the new iceberg
had found itself. Already it was moving forward, and the wind was
driving it foot by foot into the outgoing tide.
Mrs. Bear knew precisely what to do. Twice in her life she had
traveled on ice floes, though never on so large a one as this.
"Here we are, and here we stay," she said. "By and by we'll come to
islands, or so close to shore that we can swim back to land. It will be
a long time before we are carried out beyond this gulf, and we're
sure to escape before then."
She was a little too cheerful, perhaps, for some of her own kin had
gone that way so far into the great southerly current that they were
never seen again. But Mrs. Bear was one of those happy beings who
always look for the best, not the worst; and she was too joyous over
this sudden deliverance to heed any new perplexity.
Long weeks afterward, when Mother Bear's wounds were healed,
and Odin and Olga had indeed learned how to live by diving and
hunting under water, they came to a narrow bay where the land was
green on both sides. The distance from their iceberg was but little;
and they plunged in, while Mr. Burgomaster circled over them
excitedly. He was a wonderful mariner, Mr. Burgomaster, and, being
such a good friend, he had flown back and forth over land and sea,
following them on their icy ship.
"You'll know where you are, Mrs. Bear, when this fog lifts," he said.
"You will find that you have come to a beautiful shore where there
are berries and all kinds of refreshing things that bears like. It was a
good day that the iceberg started you on your trip."
"All things, Mr. Burgomaster," said wonderful Mother Bear, as she
crawled out of the water and shook her shaggy fur, "all things
happen for the best!"
THE END
*** END OF THE PROJECT GUTENBERG EBOOK FOREST
FRIENDS ***
Updated editions will replace the previous one—the old editions will
be renamed.
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside the
United States, check the laws of your country in addition to the terms
of this agreement before downloading, copying, displaying,
performing, distributing or creating derivative works based on this
work or any other Project Gutenberg™ work. The Foundation makes
no representations concerning the copyright status of any work in
any country other than the United States.
• You pay a royalty fee of 20% of the gross profits you derive from
the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”
• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.
1.F.
1.F.4. Except for the limited right of replacement or refund set forth in
paragraph 1.F.3, this work is provided to you ‘AS-IS’, WITH NO
OTHER WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PURPOSE.
Please check the Project Gutenberg web pages for current donation
methods and addresses. Donations are accepted in a number of
other ways including checks, online payments and credit card
donations. To donate, please visit: www.gutenberg.org/donate.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.