Laser

© All Rights Reserved

6 tayangan

Laser

© All Rights Reserved

- Solidwork Flow Simulation Tutorial
- Motivation Letter for PhD application
- Conduction
- TUP038
- Streamlight TLR-2 HL G – Lighting Up the Night
- Investigation of Thermal Insulation on Ice Coolers
- Conduction
- con pasco
- Hmt Kkk Unit 1, 2
- A a A
- Ground Loop Report
- Laser Properties
- Ingolf V. Hertel et al- Surface and bulk ultra-short pulsed laser processing of transparent materials
- Wei Yu et al- Direct acceleration of solid-density plasma bunch by ultraintense laser
- Full Range CST Berger
- Highlight4000l Ds Final
- Heat Chap04 081
- 1 (6)
- Thermal Stresses Induced by a Point Heat Source in a Circular Plate by Quasi-static Approach
- Conduction and Heat Exchangers

Anda di halaman 1dari 185

Bekir S. Yilbas

Ahmad Y. Al-Dweik

Nasser Al-Aqeeli

Hussain M. Al-Qahtani

Laser Pulse

Heating of Surfaces

and Thermal Stress

Analysis

Materials Forming, Machining and Tribology

Series Editor

J. Paulo Davim

http://www.springer.com/series/11181

Bekir S. Yilbas Ahmad Y. Al-Dweik

of Surfaces and Thermal

Stress Analysis

123

Bekir S. Yilbas Ahmad Y. Al-Dweik

Nasser Al-Aqeeli Department of Mathematics and Statistics

Hussain M. Al-Qahtani King Fahd University of Petroleum

Mechanical Engineering Department and Minerals

King Fahd University of Petroleum Dhahran

and Minerals Saudi Arabia

Dhahran

Saudi Arabia

ISBN 978-3-319-00085-5 ISBN 978-3-319-00086-2 (eBook)

DOI 10.1007/978-3-319-00086-2

Springer Cham Heidelberg New York Dordrecht London

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of

the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,

recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or

information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar

methodology now known or hereafter developed. Exempted from this legal reservation are brief

excerpts in connection with reviews or scholarly analysis or material supplied specifically for the

purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the

work. Duplication of this publication or parts thereof is permitted only under the provisions of

the Copyright Law of the Publishers location, in its current version, and permission for use must

always be obtained from Springer. Permissions for use may be obtained through RightsLink at the

Copyright Clearance Center. Violations are liable to prosecution under the respective Copyright Law.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this

publication does not imply, even in the absence of a specific statement, that such names are exempt

from the relevant protective laws and regulations and therefore free for general use.

While the advice and information in this book are believed to be true and accurate at the date of

publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for

any errors or omissions that may be made. The publisher makes no warranty, express or implied, with

respect to the material contained herein.

Preface

High power lasers can be used as the tools to treat and machine the materials in

industry with precision of operation, high processing speed, and low cost.

Although laser processing of metallic materials has several advantages over the

conventional methods, the development of high temperature gradients in the

treated region gives rise to the formation of high stress levels in the region treated.

In some instances, this limits the practical applications of the material processed

by a high power laser beam. Although considerable research studies were carried

out to examine the thermal stress developed during laser processing, the models

developed have short comings to reflect the actual physical problems. This is

mainly because of the complicated nature of the problem, which involves with

multi-physics phenomena. Online experimentation of the thermally induced stress

field is extremely difficult and costly because of the limitations in sensing systems,

which operate at tremendously elevated temperatures during laser processing.

Therefore, the model studies still satisfactorily provide physical insight into the

thermally induced processes enabling to understand the relations between the

process parameters and thermal response of the treated material.

Although the physical processes related to the laser heating and thermal stress

development are complicated, analytical solution to the problem is possible

through incorporating some useful assumptions in the analysis. Since the tem-

perature gradient along the absorption depth is much higher than that of the

direction normal to the absorption depth, heating situation may reduce to one-

dimensional problem. In addition, the assumption of heating of a semi-infinite

body can be justified after comparing the depth of absorption with the thickness of

the substrate material, which is significantly larger than the absorption depth.

Analytical solution to the laser pulse heating problem yields the functional relation

between the dependent and independent parameters, despite the unavoidable

number of assumptions, which may be reduced in the numerical solutions. Nev-

ertheless, validation through the experimentation of the findings is necessary to

fulfill the required accuracy of the solutions.

In this book, thermal stress development during laser pulse heating of metallic

surfaces is formulated for various laser pulse parameters and heating conditions

incorporating the temporal variation of laser pulse intensity, the convection

cooling of the surface resembling the assisting gas, and the consideration of

v

vi Preface

heating situations are classified presenting the closed form solutions, accordingly.

Analytical solution is also presented for two-dimensional heating situation for non-

equilibrium energy transfer in the irradiated region. However, some cases pertinent

to laser heating and thermal stress development are not presented in this book due

to space limitations and, therefore, these cases are left for the future treatments.

Acknowledgments

We would like to acknowledge the role of King Fahd University of Petroleum and

Minerals in extending a strong support from the beginning to end facilitating every

means during the preparation of this book. The author wishes to thank the col-

leagues who contributed to the work presented in the book through previous

cooperation of the author. In particular, thanks to Dr. Muammer Kalyon, Dr. S. Z.

Shuja, Dr. Saad Bin Mansoor, and all our graduate students.

vii

Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.2 Step Input Laser Pulse Heating . . . . . . . . . . . . . . . . . . . . . . . . 6

2.2.1 Stress Free Boundary at the Surface . . . . . . . . . . . . . . . 6

2.2.2 Stress Continuity Boundary at the Surface . . . . . . . . . . . 19

2.3 Time Exponentially Varying Laser Pulse Heating . . . . . . . . . . . 26

2.3.1 Stress Free Boundary at the Surface . . . . . . . . . . . . . . . 26

2.3.2 Stress Free Boundary and Convection at the Surface. . . . 36

2.3.3 Stress Boundary at the Surface . . . . . . . . . . . . . . . . . . . 54

2.4 Entropy Analysis Due to Thermal Stress Field . . . . . . . . . . . . . 64

2.5 Findings and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

2.5.1 Step Input Pulse Heating . . . . . . . . . . . . . . . . . . . . . . . 67

2.5.2 Time Exponentially Varying Laser Pulse Heating . . . . . . 71

2.5.3 Entropy Analysis Due to Thermal Stress Field . . . . . . . . 79

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . ........ . . . 85

3.2 Surface Heat Source Consideration . . . . . . . . . . ........ . . . 86

3.2.1 Step Input Pulse Heating . . . . . . . . . . . . ........ . . . 87

3.2.2 Exponential Pulse Heating . . . . . . . . . . . ........ . . . 92

3.3 Volumetric Source Consideration . . . . . . . . . . . ........ . . . 95

3.3.1 Step Input Pulse Heating . . . . . . . . . . . . ........ . . . 96

3.3.2 Exponential Pulse Heating . . . . . . . . . . . ........ . . . 103

3.4 Entropy Analysis . . . . . . . . . . . . . . . . . . . . . . . ........ . . . 108

3.5 Findings and Discussion . . . . . . . . . . . . . . . . . . ........ . . . 110

3.5.1 Surface Heat Source Consideration . . . . . ........ . . . 110

3.5.2 Volumetric Heat Source Consideration . . ........ . . . 114

3.5.3 Entropy Generation Rate . . . . . . . . . . . . ........ . . . 117

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........ . . . 119

ix

x Contents

for Stress Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 121

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 121

4.2 Formulation of Energy Transport in Metallic Substrates

at Microscopic Level . . . . . . . . . . . . . . . . . . . . . . . . . ...... 122

4.3 Thermal Stress Field: Consideration of Surface

and Volumetric Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

4.3.1 Surface Heat Source Consideration . . . . . . . . . . . . . . . . 134

4.3.2 Volumetric Heat Source Consideration . . . . . . . . . . . . . 140

4.4 Thermal Stress Field: Two-Dimensional Consideration . . . . . . . 147

4.5 Findings and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

4.5.1 Surface Heat Source Consideration . . . . . . . . . . . . . . . . 154

4.5.2 Volumetric Heat Source Consideration . . . . . . . . . . . . . 158

4.5.3 Two-Dimensional Analysis. . . . . . . . . . . . . . . . . . . . . . 160

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164

5.1 Equilibrium Heating. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

5.1.1 Step Input Laser Pulse Intensity . . . . . . . . . . . . . . . . . . 168

5.1.2 Exponential Laser Pulse Intensity . . . . . . . . . . . . . . . . . 168

5.2 Cattaneo Heating Model and Thermal Stresses . . . . . . . . . . . . . 170

5.2.1 Exponential Laser Pulse Intensity . . . . . . . . . . . . . . . . . 171

5.2.2 Step Input Laser Pulse Intensity . . . . . . . . . . . . . . . . . . 171

5.3 Non-Equilibrium Heating . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

5.3.1 Surface Heat Source Consideration . . . . . . . . . . . . . . . . 173

5.3.2 Volumetric Heat Source Consideration . . . . . . . . . . . . . 174

5.3.3 Two-Dimensional Heating . . . . . . . . . . . . . . . . . . . . . . 174

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

Chapter 1

Introduction

methods such as precision of operation, local treatment, and low cost. Laser at high

intensity when interacts with the solid surface, the absorption takes place. This in

turn causes internal energy gain of the substrate material and heat release from the

irradiated region. Since the process, in general, is fast, temperature gradients remain

high in the irradiated region. This results in high thermal strain and thermally

induced stresses in this region. Moreover, in laser treatment process, the end product

is important from the application point view. The high stress levels formed in the

irradiated region may cause failure of the surface through stress induced cracking.

Consequently, a care needs to be taken during the laser treatment process. This

chapter provides the information about the importance and limitations of the laser

treatment process in terms of the thermal stresses formed in the irradiated region.

Lasers are considered to be one of the effective tools for laser treatment of metallic

surfaces. High power laser, when focused at the surface, generates excessive heat,

which enables the surface to reach the melting temperature of the substrate

material. In some cases, laser heating is associated with the phase change, such as

melting and evaporation at the surface. In this case the heating process becomes

complicated because of the size of the heated, which is in the order of fraction of

millimeter. Therefore, controlling the phase process becomes necessary to avoid

surface asperities, such as cavities, which are formed during surface evaporation.

One of the methods to secure the control melting at the surface is to model the

physical processes involved during the laser interaction with surface. Although

model studies involve with assumptions simplifying the mathematical arrange-

ments, experimenting the process is difficult, timely, and expensive because of the

high temperature involvement and small size of heated region during the short

time period. The time limitation in measurements is attributed to the laser pulse

length, which has to be shorter to be effective for the treatment of the surfaces.

Laser heating process can be classified into two categories. The first category

involves with the solid state heating without the phase change at the irradiated

surface. The heating is governed by the conduction process, which can be

and Thermal Stress Analysis, Materials Forming, Machining and Tribology,

DOI: 10.1007/978-3-319-00086-2_1, Springer International Publishing Switzerland 2014

2 1 Introduction

governed by the Fourier heating law. This consideration has shortcomings when

the heating duration becomes comparable with the thermalization time of the

substrate material, which is on the order of fraction of nanoseconds for most of the

metals. The Fourier heating low fails to predict the correct temperature rise in this

region due to the consideration of infinite heat wave speed resulting in equilibrium

heating in the irradiated region. Since the absorption depth of the incident radiation

is small for metallic substrates, the heat wave propagation at a finite speed occurs

in the irradiated region. This is particularly true for laser short-pulse heating of the

metallic surfaces. In this case, the short heating duration of solid surfaces, by a

laser beam, initiates non-equilibrium energy transport in the irradiated region. In

addition, electrons gain excess energy from the irradiated field and increase their

temperature. This, in turn, results in thermal separation of electron and lattice sub-

systems in the irradiated region. Since electron excess energy transfer to lattice site

takes place through the collisional process, lattice site temperature increases

gradually as the heating period progresses. The rates of electron and lattice site

temperature increase change for different absorption depths and laser pulse shapes.

The energy transfer in the irradiated region governs by non-equilibrium transport,

which can be modeled through electron kinetic theory approach. The resulting

lattice temperature equation becomes hyperbolic due to wave nature of the heating

process during the short interaction time. Since the depth of absorption of the

irradiated field is considerably small and heating duration is extremely short, the

closed form solution of the resulting hyperbolic equation provides useful insight

into the heating process. On the other hand, the phase change takes place at high

laser power intensities. The irradiated material undergoes first melting and sub-

sequent evaporation during the heating period. Although phase change involves

with the energy transport by conduction and convection, the convection effect may

be neglected due to the small depth of the melt layer. Therefore, introducing the

convective boundary condition at the surface resembling the surface melting and

evaporation satisfies the physical phenomena taking place during the high intensity

laser irradiation at the surface. Moreover, this assumption simplifies the heating

problem such that the analytical solution to the resulting energy equation becomes

feasible.

In laser heating, thermal strain is developed because of the presence of the

temperature gradient in the irradiated solid substrate. Depending on the thermal

expansion coefficient of the substrate material and the temperature gradient,

thermal strain results in excessive thermal stress levels in the substrate material. In

laser non-conduction heating situation, melting and evaporation of the substrate

material occur. Since the materials expands freely in the vapor and molten phases,

the stress level reduces to zero in these phases. Moreover, below the melt zone

conduction heating takes place in the solid substrate during the laser surface

treatment process. The thermal stresses can only developed in the solid phase of

the substrate material and the energy transfer taking place in the solid substrate is

limited with the conduction heating process. Consequently, the solid phase of the

substrate material is subjected to the stress levels during the surface treatment

1 Introduction 3

development across the vapor front and the melt surface. The magnitude of recoil

pressure depends on the laser power intensity across the irradiated spot and

material properties. Since the laser pulse, in general, decays exponentially with

time, the pressure is expected to decay with a similar trend of the power intensity.

The recoil pressure can be considered as a stress source, which acts externally at

the surface of the substrate material. In some cases, such as low rate of evapo-

ration, the recoil pressure induced forces can be ignored at the surface. This allows

introducing stress free condition at the surface. However, the presence of thin films

at the surface causes stress continuity at the interface between the substrate

material and the coating. In this case, the stress gradient may be considered ay zero

rather than the stress free surface.

Laser short-pulse heating results in non-equilibrium energy transfer in the

irradiated region and the speed of the thermal cannot be ignored as indicated

earlier. However, lattice-site temperature rise becomes gradual during the laser

short-pulse heating because of the collisional energy transport between the excited

electrons and the lattice phonons. The thermal stress is developed in the lattice

sub-system only because of the work done against the thermal expansion.

Application of the classical treatment of thermal stress analysis is useful to

understand the mechanical response of the laser irradiated region to the short-pulse

thermal loading. In addition, the energy deposited in a laser irradiation pulse is

stored in the substrate material in terms of the internal energy gain and the

mechanical work done during the thermal expansion. If the heating process takes

place in a short time period, thermal energy transport becomes non-equilibrium

while influencing the mechanical response of the material in the heated region.

Although the thermomechanical coupling between the temperature and stress

fields is small, it cannot be neglected for the non-equilibrium heating situation.

Consequently, uncoupled solutions of thermal and mechanical fields lead to the

incorrect results associated with the actual physical process. On the other hand,

during the long laser pulse heating (pulses longer than the thermalization time),

thermomechanical coupling may not be important in the energy equation so that

the uncoupled solution of the problem becomes visible to explore both the tem-

perature and the stress fields. The analytical solution for the coupled non-equi-

librium energy transport or for uncoupled equilibrium energy transport provides

the functional relations between the laser parameters and the workpiece properties

in time and space for thermal and stress fields.

Chapter 2

Equilibrium Laser Pulse Heating

and Thermal Stress Analysis

Abstract When the heating duration becomes greater than the thermalization time

of the substrate material, equilibrium heating takes place in the laser irradiated

region. In this case, the classical Fourier heating law governs the energy transport.

Although the heating process is complicated, some useful assumptions enable to

obtain the closed form solution for temperature and stress fields. Since the ana-

lytical solution provides the functional relation between the dependent variable

and the independent parameters, it provides better physical insight into the heating

problem than that of the numerical analysis. In this chapter, equilibrium heating of

solid surfaces heated by a laser beam is considered. The closed form solution for

the resulting temperature and stress fields are presented for various heating situ-

ations. The study also covers the phase change taking place at the irradiated region

during the laser treatment process.

2.1 Introduction

Laser pulse heating of metallic surfaces can be modeled using the Fourier heating

law for the pulse lengths greater than the thermalization time of the substrate

material. Laser pulses can be considered as a step input pulse type or a time

exponentially decaying pulses. In addition, the boundary conditions are important

and the solution changes for different boundary conditions resembling the actual

physical configurations. Therefore, the analytical solution for the laser heating

problem depends on the pulse type and the boundary conditions incorporated in

relation to the physical situations. During the laser heating process, thermal strain is

developed because of the temperature gradient generated in the solid substrate.

Depending on the thermal expansion coefficient of the substrate material and the

temperature gradient, thermal strain results in excessive thermal stress levels in the

substrate material. Therefore, thermal stress developed because of step input laser

and time exponentially varying pulses for various boundary conditions are con-

sidered, and the closed form solutions for temperature and stress fields are presented

in line with the previous studies [18].

and Thermal Stress Analysis, Materials Forming, Machining and Tribology,

DOI: 10.1007/978-3-319-00086-2_2, Springer International Publishing Switzerland 2014

6 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

The laser spot diameter at the workpiece surface is small and heat transfer in the

radial direction is considerably smaller than that corresponding to the axial direc-

tion. Therefore, one-dimensional modeling of the laser heating process predicts

reasonably accurate temperature distribution at the surface and inside the substrate

material. In addition, the absorption depth of laser irradiated radiation is consider-

ably small in the surface as compared to the thickness of the substrate material;

consequently, the substrate material can be assumed as a semi-infinite body.

The boundary conditions influence the thermal stress distribution in the irradiated

region. In, general two-boundary conditions can be considered associated with the

practical laser heating applications. The first boundary condition involves with

stress free surface where the surface free to expand during the heating cycle. The

second boundary condition is stress continuity at the surface because of the

presence of the film at the surface. The closed form solutions for laser heating and

thermal stress development are given below the appropriate sub headings in line

with the previous studies [13].

The laser heating pulse can be considered as a step input pulse. In the actual laser

pulse, there exists the pulse rise and fall times and the actual laser pulse deviates

slightly from the step input pulse due to small pulse rise and fall times. However, it

is convenient to use a step input pulse in the analysis for the mathematical sim-

plicity. In order to construct a single step intensity pulse, it is necessary to subtract

two unit step functions with a time shift between them. The first step function will

start at t 0 and the other will start at t Dt. The step intensity pulse is therefore:

SPt 1t 1t Dt 2:1

where:

( )

1; jt [ 0

1t 2:2

0; t\0

and

( )

1; t [ Dt

1t Dt 2:3

0; t\Dt

a single step intensity pulse is given here; however, the analyses related to two

2.2 Step Input Laser Pulse Heating 7

successive pulses are similar to the single step intensity pulse and, therefore, it is

avoided herein. The Fourier heat transfer equation for a laser heating pulse can be

written as:

o2 T I 1 d 1 oT

2

C1 SPtedx 2:4

ox k a ot

where:

I1 1 rf Io 2:5

Initially, the substrate material is assumed to at a constant uniform temperature;

therefore, the initial condition is:

At t = 0 ) Tx; 0 T0 2:6

Since the heating duration is short and no assist gas cooling is assumed;

therefore, the radiation and convective losses from the surface are negligible.

Consequently, the corresponding boundary condition is:

oT

At x 0 ) 0 2:7

ox x0

Since the depth of the irradiated region is limited with the depth of absorption,

the size of the absorption depth is considerably smaller than the depth of the

workpiece; therefore a semi-infinite workpiece is considered. This assumption

leads to a boundary condition of constant temperature at an infinitely depth below

the surface, i.e. the influence of the laser heating pulse is negligible at a depth

infinitely long from the surface. Therefore,

At x 1 ) T1; t 0 2:8

The Laplace transform of Eq. 2.4 with respect to t, results:

o2 T I1 d 1

2

C1 SPsedx sTx; s Tx; 0 2:9

ox k a

where:

1 eDts

SPs 2:10

s s

Introducing the initial condition and rearranging Eq. 2.9 yields:

o2 T

I1 d C1 SPsedx T0

k2 T 2:11

ox2 k a

where k2 = s/a and s is the transform variable. Equation 2.11 has the solution:

I1 d C1 SPs dx T0

Tx; s Aekx Bekx e 2:12

k d2 s=a s

8 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

where A and B are constants. Introducing the boundary conditions determines the

constants A and B, i.e.:

I1 d2 C1 SPs

B 0; and A :

kkd2 s=a

Substitution of A and B in Eq. 2.12 yields:

2 p kx

s I1 d apC 1 SPse I1 dC1 SPsedx T0

Tx; 2 2:13

k s d k2 k d2 k2 s

The inverse Laplace transform of Eq. 2.13 gives the temperature distribution

inside the substrate material in non-dimensional form. This is possible by defining

dimensionless quantities, which are:

Tkd

x xd : t ad2 t : T 2:14

I1

The dimensionless temperature distribution becomes:

8 2

39

> x

p x >

>

> e erfc t p

7>>

> 6

>

>

>

x t

e 6 t 2 t

7>>

>

>

>

> e e U1 6 7 >

>

< 2 6

7 =

4 x p x 5

T T0 C1 e erfc

t p

>

> 2 t >

>

>

> >

>

>

> p

>

>

>

> 2 t x2 x >

>

>

: p e 4t x erfc p

>

;

p 2 t

8 x t Dt 9

>

> e e U1 Dt >

> 2:15

>

> 2

3>>

>

> p >

>

>

> x

x >

>

>

> 6

e erfc t Dt p 7 >

>

>

> e

t Dt 6 2 t Dt 7 >

>

>

< 6 7>=

2 4 6

7

C1 p x 5

>

> e erfcx

t Dt p >

>

>

> 2 t Dt >

>

>

> >

>

>

> " p

# >

>

>

> Dt >

>

>

> 2 t x2

4t Dt x >

>

>

: p e x erfc p >

;

p

2 t Dt

Zz

2 2

erfcz 1 erf z and erfcz p et dt

p

0

2.2 Step Input Laser Pulse Heating 9

To solve for the stress distribution inside the substrate material, equation

governing the momentum in a one-dimensional solid for a linear elastic case can

be considered [9], i.e.:

o2 rx 1 o2 rx o2 T

2

2 2 c2 2 2:16

ox c1 ot ot

where c1 is the wave speed in the solid

s

E 1t

c1 and c2 qaT

q 1t

where t is Poissons ratio, q is the density of the solid and aT is the thermal

expansion coefficient of the solid.

Initial and boundary conditions for the temperature field in Eq. 2.16 are similar to

those given for Eq. 2.4. Initially substrate material is considered as free from the

stress. In addition, as time extends to infinity, the stress field vanishes in the substrate

material. Therefore, the initial and final conditions for the stress field become:

At t 0 ) rx 0 2:17

and

At t 1 ) rx 0 2:18

The consideration of no mechanical force at the surface prior to laser irradiation

pulse leads to stress free boundary conditions at the surface. In addition, the

effective depth of laser irradiation is considerably smaller than the workpiece

thickness. Therefore, the assumption of semi-infinite body holds in the stress

analysis. In this case, the thermal strain disappears at a depth infinitely long from

the surface. Consequently, the corresponding boundary conditions yield:

At x 0 ) rx 0 2:19

and

At x 1 ) rx 0 2:20

Taking the Laplace transform of Eq. 2.16 with respect to time yields:

o2 r

x 1 :

2

2 s2 rx x; s srx x; 0 rx x; 0

ox c 2:21

h1 : i

s sTx; 0 Tx; 0

c2 s2 Tx;

x x; s and Tx;

perature, respectively in the x and s domains.

By substituting the initial conditions, Eq. 2.21 reduces to:

o2 r

x s2 s c2 sT0

x x; s c2 s2 Tx;

r 2:22

ox2 c1 2

10 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

variation, Eq. 2.13, and substituting into Eq. 2.22 and solving for the stress field,

yields:

" #

2 p

o2 r

x s2 2 I1 d aC1 SPsekx I1 dC1 SPsedx T0

x c2 s

r p c2 sT0

ox2 c1 2 k s d2 k2 k d2 k2 s

2:23

The complementary and the particular solutions for Eq. 2.23 are:

sx sx

r

x c D1 ec1 D2 e c1 2:24

While the particular solution has two parts, the first part is:

ps

x p1 D3 e ax

r 2:25

x p2 D4 edx

r 2:26

p

c2 s2 I1 d2 C1 SPs a

D3 p 2 2:27

k s d k2 k2 s2 =c21

and

c2 s2 I1 dC1 SPs

D4 2 2:28

k d k2 s2 =c21 d2

So, the general solution for the stress field becomes:

sx sx

x g D1 ec1 D2 e c1 D3 egx D4 edx

r 2:29

Then,

sx

ps

x g D2 e c1 D3 e ax D4 edx

r 2:30

x 0, the

Consider the boundary condition at the surface, where at x 0 ) r

constant in Eq. 2.30 becomes:

D2 D3 D4 2:31

2.2 Step Input Laser Pulse Heating 11

Then:

ps sx sx

x x; s D3 e

r ax D3 e c1 D4 edx D4 e c1 2:32

Finding the solution for rx in the x and t domain, we should take the inverse

Laplace transform for each term in Eq. 2.32. To accomplish this, the following

definitions are introduced:

ps sx

Term1 D3 e ax Term2 D3 esx

c1

2:33

Term3 D4 edx Term4 D4 e c1

Consequently, the solution for the stress distribution becomes the summation of

the inverse Laplace transforms of the above terms.

Therefore, the inverse of Laplace transform of Terms (Term1, Term2, Term3,

Term4) are:

p " ps ps #

I 1 d2 C 1 c 2 a e ax e ax eDts

Term1 = p 2 p 2

k s d s=a 1=a s=c21 s d s=a 1=a s=c21

2:34

After partial fraction of the above equation:

Term1

2 ps p 4 psx p ps 3

ae ax sa e a c21 a se ax

p 6 psd2 a2 d2 c2 sa c2 d2 a2 d2 c2 ad2 s 7

I 1 d2 C 1 c 2 a 6 6 1 1 1

7

7

6

k 6 p

as x Dts p 4 psx Dts 2

p psx Dts 77

4 ae e sa e a e c1 a se a e 5

p 2 2 2 2

2 2 2 2

2

sd 2

a d c1 sa c1 d a d c1 ad s

2:35

p

let G1 I1 d Ck1 c2 a, then:

2

where

ps p 4 psx

ae s

1 eDts

a sa e a 1 eDts

Term11 p 2 ; Term21 2 and

sd a2 d c1 2 sa c1 2

p ps 2:37

c1 2 a se ax 1 eDts

Term31 2

s2 a2 d c1 2 ad2 s

The inverse of Laplace transform of Term1 is -1[Term1], i.e.:

12 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

"

!#

1 a 1 x2 1 x2

G1 Term11 G1 2 p exp p exp

d pt 4at pt Dt 4at Dt

2:39

1 G1 Term21

2 2

r

3

2 x c1 c21 tc1 x t x

6 p

exp p e a a erfc c1 p 7

6 pt 4 a a 2 at 7

6 r !

7

6 c1 c21 tc1 x t x 2 x2 7

6 p e a a erfc c1 p p exp 7

6 a 2 at 4at Dt 7

6 a pt Dt 7

a3 6 ! 7

G1 2 6 r 7

2 a2 d c21 66 c1 c21 tDtc1 x t Dt x 7

7

6 p e a a erfc c1 p 7

6 a a 2 at Dt 7

6 r ! 7

6 2 7

4 c1 c1 tDtc1 x t Dt x 5

p e a a erfc c1 p

a a 2 at Dt

2:40

1 G1 Term31

2 2

3

2 x pad2 tdx p x

6 p exp a erfc d at p 7

6 pt 4at 2 at

6

7

7

6 p 2 p x 2 x 2

7

6 adead tdx erfc d at p p exp 7

6 2 at pt Dt 4at Dt 7

c21 a 6 ! 7

G1 2 2 6 7

2

2d c1 a d 62 6 p ad2 tDtdx p x 7

ade erfc d at Dt p

7

6 7

6 2 at Dt 7

6 ! 7

6 p 7

4 p ad2 tDtdx x 5

ade erfc d at Dt p

2 at Dt

2:41

However, the Term 2 is:

p " sx sx

#

c2 d2 I1 C1 a e c1 e c1 eDts

Term2 p 2 p 2

k s d s=a 1=a s=c21 s d s=a 1=a s=c21

2:42

Let:

p

c2 d2 I1 C1 a

G2 2:43

k

After the partial fraction, Term2 can be written as:

Term2 G2 Term12 Term22 Term32 2:44

2.2 Step Input Laser Pulse Heating 13

where

sx p s x

a e c1 c21 sae c1

Term12 2 p : Term22 2

d s d c1 2 a2 d2 ad2 s

and

p 4 cs x

sa e 1

: Term32 2 2 2:45

d a d c1 2 as c1 2

2

or

"

1 a 1 x

G2 Term12 G2 2 p 1 t

d pt x=c1 c1

# 2:47

1 x

p 1 t Dt

pt Dt x=c1 c1

and

1 G2 Term22

2 r

!

3

1 p ad2 t p x x

6 p

a de erf a d t 1 t 7

c2 a 6 pt x=c1 c1 c1 7

6 7

G2 2 12 2 6 r

!

7

2

d a d c1 46 p ad2 tDt p 7

1 x x 5

p ade erf ad t Dt 1 t Dt

pt Dt x=c1 c1 c1

2:48

and

1 G2 Term32

2 r

!

3

1 c1 c21 t c1 x x

6 p p e a erf p t 1 t 7

a3 6 pt x=c1 a a c1 c1 7

6 7

G2 2 6 r

!

7

a2 d c1 6

2

4 1 c1 c21 tDt c1 x x 5

7

p p e a erf p t Dt 1 t Dt

pt Dt x=c1 a a c1 c1

2:49

where erf (y) is the error function of the variable y.

However, Term3 is:

" #

c2 I1 dC1 sedx sedx eDts

Term3 2:50

k s2 =c21 d2 d2 s=a s2 =c21 d2 d2 s=a

14 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

Let:

c2 I1 dC1

G3 2:51

k

The Term3 can be written as:

Term3 G3 Term13 G3 Term23 G3 Term33 2:52

After the partial fraction, the Term 3 becomes:

Term13 : Term23

2ds c1 dc1 ad 2ds c1 dc1 ad

and

c1 2 a2 edx

Term13 2:53

c1 2 a2 d2 s ad2

The inverse of Laplace transform of Term3 comes out to be:

where

1 G3 Term13 G3 e ec1 dt 2:55

2dc1 ad

and

c21 aedx h c1 dt i

1 G3 Term23 G3 e ec1 dtDt 2:56

2dc1 ad

and

i

1 G3 Term33 G3 e e 2:57

c21 a2 d2

However, Term4 is:

" #

sx sx

c2 I1 dC1 se c1 se c1 eDts

Term4

k s2 =c21 d2 d2 s=a s2 =c21 d2 d2 s=a

2:58

Let:

c2 I1 dC1

G4 2:59

k

2.2 Step Input Laser Pulse Heating 15

Term4 G4 Term14 G4 Term24 G4 Term34 2:60

After partial fraction of Term4, it yields:

sx sx

c1 2 ae c1 c1 2 ae c1

Term14 : Term24

2ds c1 dc1 ad 2ds c1 dc1 ad

and

sx

c1 2 a2 e c1

Term34 2:61

c1 2 a2 d2 s ad2

The inversion of Term 4 is:

where

1 G4 Term14

"

#

c21 a c1 d tDtcx x c1 d tcx x

G4 e 1

1 t Dt e 1

1 t

2dc1 ad c1 c1

2:63

and

1 G4 Term24

"

#

c21 a c1 d tcx x c1 d tDtcx x

G4 e 1

1 t e 1

1 t Dt

2dc1 ad c1 c1

2:64

and

1 G4 Term34

"

#

c21 a2 ad2 tcx x ad2 tDtcx x

G4 2 e 1

1 t e 1

1 t Dt

c 1 a2 d2 c1 c1

2:65

where 1 t cx1 is a unit step function.

The closed form solution of stress distribution can be written as:

where -1 represents the inverse sign of Laplace transform.

16 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

dimensionless quantities are defined, i.e.:

c1 krx

c1

and rx 2:67

ad I1 da2 c2 C1

U[1] is the unit step function, which is U1 t cx , and U1 Dt is the

1

unit step function for the time shift, i.e. U1 Dt 1 t Dt cx

1

Therefore, for the dimensionless stress distribution, the following results are

obtained:

rx 1 rx 11 rx 21 rx 31 2:68

where

2

1 x 1 x2

rx 11 p exp p exp 2:69

pt 4t pt Dt 4t Dt

and

2 2

3

2 x c2 t c1 x

p x

6 p exp c1 e erfc c1 t p

1

pt 4t 2 t 7

6

7

6 p

7

6 2 x 7

6 c1 ec1 t c1 x erfc c1 t p 7

6 2 t 7

6

7

6 2 x 2 7

6 7

1 6 p exp 7

rx 21 2 6 pt Dt 4t Dt 7

2 1 c 1 6 !7

6 p 7

6 x 7

6 c1 ec2

1 t Dt c1 x erfc c

t Dt p 7

6 1

2 t Dt 7

6 7

6 ! 7

6 p

7

4 c2 t Dt c1 x

x 5

c1 e 1 erfc c1 t Dt p

2 t Dt

2:70

and

rx 31

2 2

3

2 x t x

p x p x

p exp e erfc t p et x erfc t p

6 pt 4t 2 t 2 t 7

6 7

6

!7

6 2 x 2 p x 7

c 2 6 7

1 2 6 p exp

et Dt x erfc t Dt p 7

2 1c1 6 6 p t Dt 4t Dt 2 t Dt 7

7

6 ! 7

6 p x 7

4

t Dt x

5

e erfc t Dt p

2 t Dt

2:71

2.2 Step Input Laser Pulse Heating 17

However, rx 2 is:

rx 2 rx 12 rx 22 rx 32 2:72

where

1 1

rx 12 q

U1 q

U1 Dt 2:73

p t x =c1

p t Dt x =c1

and

20 s!

1 3

1 x

B

6 @q et erf

C 7

6 t A U1 7

6 p t x =c c1 7

2

c1 6 1 7

rx 22 6 0 1 7

1c2 6 s! 7

1 6 7

6 B 1

Dt

x C 7

4 @q

e t

erf t Dt A U 1Dt 5

p t Dt x =c1 c 1

2:74

and

2 0 1 3

s!

1 x

6 B c2 t C 7

6 @q

c1 e 1 erf c1 t A U1 7

6

p t x =c1 c1 7

1 6 7

rx 32 6 0 1 7

1 c2 6 s

! 7

1 6 7

6 B 1 c2 x C 7

4 @q

c e 1 t Dt erf c

t Dt

A U 1 Dt 5

1 1

c1

p t Dt x =c1

2:75

The term rx 3 is:

rx 3 rx 13 rx 23 rx 33 2:76

where

x h i

c2

1 e

rx 13

ec1 t Dt ec1 t 2:77

2 c1 1

and

rx 23 1 e 1 ec1 t Dt 2:78

2 c1 1

and

ex h t i

c2

1

rx 33 2 e et Dt 2:79

c1 1

18 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

The term rx 4 is:

rx 4 rx 14 rx 24 rx 34 2:80

where

" #

c2 c

t

Dt x

c

t x

rx 14 1 e U1 Dt e

1 c 1 c

1 1 U1 2:81

2 c1 1

and

" #

c2 c

t x

c

t

Dt x

rx 24 1 e

1 c 1 c

1 U1 e 1 U1 Dt 2:82

2 c1 1

and

" #

c2

1

t xc

t Dt xc

rx 34 2 e 1 U1 e 1 U1 Dt 2:83

c1 1

rx rx 1 rx 2 rx 3 rx 4 2:84

The solution of Eqs. 2.15 and 2.84 provides data for temperature and stress

fields. The mathematical arguments for the stress equation due to two successive

pulse heating situation are not given inhere due to the length arguments. However,

the closed form solution for the stress field (Eq. 2.84) can be used for heating

situation due to two successive pulses, provided that the following conditions are

satisfied:

rx rx first pulse rx second pulse 2:85

rx first pulse can be obtained after setting Dt t1 dt1 h1 in Eq. 2.84, where dt*1

is the pulse length of the first pulse and h1 is the cooling period after the first pulse.

It should be noted that the duration of the cooling period of the first pulse is the

same of the first pulses length (h1 = dt*1). This requires that 0 t t1 dt1 h1

(covering the heating and the cooling periods;

the second pulse initiates after the

cooling period of the first pulse. rx f secondpulse can be obtained when setting

Dt t2 dt2 h2 in Eq. 2.84, where dt*2 is the pulse length of the second pulse

and h2 is the cooling period after the second pulse. This requires that

t1 dt1 h1 t t2 dt2 h2 . In the present simulations the pulse lengths and

cooling periods of the first and second pulses of the two successive pulses are set

the same (dt*1 = dt*2 and h1 = h2).

2.2 Step Input Laser Pulse Heating 19

The zero stress gradient at the surface represents the case occurring during laser

heating of the coated surfaces. In this case, the surface of the substrate material is

coated by a high absorbent material, such as paint, to improve the absorptivity at

the surface. However, the absorbent material has different mechanical properties,

but the stress continuity across the substrate material and thin layer of the coat

occurs. This results in zero stress gradient at the surface of the substrate material.

The heat transfer equation for a laser step input heating pulse can be written

similar to Eq. 2.5, which is:

o2 T I1 d dx 1 oT

e 2:86

ox2 k a ot

where, I1 1 rf Io (similar to Eq. 2.5), x is the distance, t is the time, k is the

thermal conductivity, a is the thermal diffusivity, d is the absorption coefficient rf

is the reflection coefficient, and Io is the peak power intensity. Since the solution of

Eq. 2.86 is similar to Eq. 2.4, the boundary conditions and the solution of Eq. 2.86

is given lightly.

The initial and boundary conditions are:

At time t 0 ! Tx; 0 0 2:87

and

oT

At the surface x 0 ! 0

ox x0 2:88

and at x 1 ! Tt; 1 0

The solution of Eq. 2.86 can be obtained possibly through Laplace transfor-

mation method, i.e., with respect to t, the Laplace transformation of Eq. 2.86

yields:

o2 T I1 d dx 1

2

e sT Tx; 0 2:89

ox ks a

where T Tx; s. Using the initial condition, Tx; 0 0; Eq. 2.87 yields:

o2 T I1 d dx

q2 T e 2:90

ox2 ks

where q2 as . Equation 2.89 has a solution:

I1 ad

T Aeqx Beqx edx 2:91

ks s ad2

where A and B are the constants and they are calculated through the boundary

conditions. Substituting boundary condition, ooxT 0 at the surface (x 0), it gives:

20 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

I1 ad2

AB 2:92

qks s ad2

B yields:

I1 d2

B 2:93

qks q2 d2

Hence, Eq. 2.91 becomes:

I1 ad I 1 d2

T 2

edx 2

eqx 2:94

ks s ad 2

qks q d

Rearrangement of Eq. 2.94 yields:

I1 ad dx I1 d eqx eqx

T e 2:95

ks s ad2 2kqs q d q d

I1 p x 1

Tx; t 2 atierfc p edx

k 2 at 2d

1 ad2 tdx p x

e erfc d at p 2:96

2d 2 at

1 ad2 tdx p x

e erfc d at p

2d 2 at

where erf is the error function, erfc is the complementary error function, and ierfc

is the integral of complementary error function, which are:

Z x

2 2

erf v p ev dv

p 0

erfcv 1 erf v 2:97

1 v2

ierfcv p e verfcv

p

Introducing dimensionless quantities as:

kd

s ad2 t : x0 xd : T 0 T 2:98

I1

Substituting the dimensionless quantities in Eq. 2.96, it yields:

2.2 Step Input Laser Pulse Heating 21

0

p x 1 0

T 0 x0 ; s 2 sierfc p ex

2 s 2

1 sx0 p x0

e erfc s p 2:99

2 2 s

1 sx0 p x0

e erfc s p

2 2 s

Equation 2.99 is used to compute the dimensionless temperature profiles inside

the substrate material.

The equation governing the momentum in one-dimensional solid for a linear

elastic case can be used to formulate the thermal stress field in one-dimensional

semi-infinite solid. Therefore, Eq. 2.16 is used to formulate the momentum

equation.

The new initial conditions for Eq. 2.16 are:

and at t 1 !; for the stress equation, rx 0

2:100

At t 0 !; for the heat transfer equation,T 0

and at t 1 !; for the heattransfer equation, T 0

In the case of zero stress gradient at the surface, the relevant boundary con-

ditions are:

orx orx

At x 0 !; for thestress equation, 0 and 0

ox ot

oT

and at x 0 !; for the heat transferequation, T 0: 0 2:101

ot

Atx 1 !; for the stress equation, rx 0

and at x 1 !; for the heat transferequation, T 0

The solution of thermal stress equation (Eq. 2.16) is possible using the Laplace

transformation method, i.e. Laplace transformation of Eq. 2.16 with respect to

t yields:

o2 rx 1 2 :

2 s rx x; s srx x; 0 rx x; 0

ox2 c1 2:102

h : i

c2 s2 T x; s sTx; 0 T x; 0

where rx x; s and T x; s are the Laplace transform of thermal stress and tem-

perature, respectively. Introducing the initial condition, Eq. 2.102 yields:

o2 rx s2

2 rx x; s c2 s2 T x; s 2:103

ox2 c1

22 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

ps

T c1 e ax c2 edx 2:104

where:

p

I1 d d a

c1 p and

k s s d2 as

2:105

I1 d 1

c2

k s d2 as

" pp ps #

o2 r x s 2 I1 d d a se ax sedx

2 rx x; s c2 2 s 2 s 2:106

ox2 c1 k d a d a

" pp #

I1 d d a s

M1 s c2 2 s

k d a

" # 2:107

I1 d s

M2 s c2 2 s

k d a

o2 rx s2 ps

2

2 rx x; s M1 se ax M2 sedx 2:108

ox c1

Equation 2.108 has homogenous rx h and particular rx p solutions, i.e.:

rx rx h rx p 2:109

rx h D1 ec1 x D2 ec1 x

s s

2:110

and the particular solution has two parts. The first part is:

ps

rx p1 G1 e ax 2:111

M1 s

G1 s s2

2:112

a c21

rx p1 G2 edx 2:113

2.2 Step Input Laser Pulse Heating 23

M2 s

G2 2:114

d2 cs2

2

rx rx h rx p1 rx p2

or

s

x cs x M1 s pasx M2 s dx

rx D1 ec1 D2 e 1 s s2

e 2 s2 e 2:115

a c2

1

d c2

1

where D1 and D2 are constants and they will be calculated through boundary

conditions. The coefficient D1 must be zero, since c1 [ 0 then rx can be finite.

Equation 2.115 becomes:

ps

rx D2 ec1 x f1 se ax f2 sedx

s

2:116

where

2 3

pp

I1 d 4 d a s

f1 s c2 5 2:117

k sd2 s 1 s2

a a c1

or

" #

2 p

I 1 d 1

f1 s c2 c21 a a p 2 2:118

k s s ad2 s ca1

and

" #

I1 d s

f2 s c2 c21 a 2:119

k s ad2 s2 c21 d2

I1 d2 p

C10 c2 c21 a a 2:120

k

and

I1 d

C20 c2 c21 a 2:121

k

D2 can be found through taking the derivative of Eq. 2.116 and setting to zero

and knowing that D1 is zero, as explained before, it yields:

r

s s

D2 f1 s df2 s 0 2:122

c1 a

24 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

Therefore, D2 becomes:

c1 c1

D2 p f1 s df2 s 2:123

sa s

Substituting D2 and setting D1 as zero into Eq. 2.108, it yields:

2 3

I1 da s

x 1 1

r c2 c31 e c1 4 c2 2

5

x k 2

s s ad s a 1 s ad s c 1 d s c 1 d

2 p 3 2:124

2 p as x cs

I1 d e e 1

c2 c21 a a4p c2 5

k s s ad2 s 1 s ad2 s c1 ds c1 d

a

sx

e c1

Term1 c2

s s ad2 s a1

sx

e c1

Term2 2

s ad s c1 ds c1 d

p psx 2:125

ade a

Term3 p c2

s s ad2 s a1

sx

e c1

Term4 2

s ad s c1 ds c1 d

Using partial fraction expansion, the following relation can be obtained for Term1:

sx sx sx

1 e c1 c1 d e c1 c d e c1

Term1 4 2

2

4 1 2:126

c1 d s a2 d c21 d s ad c1 2 2 c2

2 c d s 1

a 1 a

2 3

1 c1 d ad2

tcx

6 7

6 c1 d a2 d4 c2 d2 e

1

7

x 6 1 7

1

Term1 1 t 6 7 2:127

c1 6 6 c2 7

c1 d 1 t x

7

4 c4 e a c1 5

2 2

a2 c1 d

1

"

x c1 d

ead tc1

2

1 x

Term2 1 t 2 2

c1 ad c1 d ad c1 d

# 2:128

1 c1 dtcx1 1 c1 dtcx1

e e

2 c1 d ad2 2 c1 d ad2

2.2 Step Input Laser Pulse Heating 25

Using partial fraction expansion, the following relation can be obtained for

Term3:

p " #

d a pasx 1 1

Term3 p e c2 2

2 2 2:129

s a ad

1

s ca1 ad2 ca1 s ad2

or

p " ps ps #

da a e ax e a x

Term3 2 p 2 p 2:130

ad c21 s s ca1 s s ad2

8 2

39

> p x >

>

> e erfc d at p 7 >

dx

>

>

> ad2 t 6 2 at >

>

>

> e 6 7 >

>

>

> p 6

7 >

>

>

> 2d a 4 p x 5 >

>

>

> e dx

erfc d at p >

>

p >< >

=

da a 2 at

1

Term3 2 2 r

3 2:131

ad c21 >>

> xca1 t x >

>

>

>

> c2 6 e erfc c p 7 >

a 2 at 7 >

1

>

> eat 6

1

>

>

>

> p 6 r 7>>

>

> 2c1 a 4 6

7 >

>

>

> xc1 t x 5 >

>

>

: e a erfc c1 p >

;

a 2 at

The Laplace inversion of Term4 yields:

ad2 2 1

1 Term3 4 2 2

ead t 2

ec1 dt ec1 dt edx 2:132

2

a d c1 d 2 c1 d ad

Using the dimensionless quantities and summing the terms Term1, Term2,

Term3, and Term4 gives the Laplace inversion of dimensionless stress, i.e.:

(" c0 s x0 c0 s x00 0 2

#

ec1 sc1 x

0 0

I 1 d2

0

0 2 e 1 c1 e 1 c1

rx c2 c1 0 0 0 1 u

k 2 c01 1 2 c1 1 c1 c1 1

es 0 p x0 0 p x0

ex erfc s p ex erfc s p

2 1 c01 2 2 s 2 s

0 2

ec1 s c01 x0 0

p x0 c01 x0 0

p x0

e erfc c1 s p

e erfc c1 s p

2c01 1 c01 2 2 s 2 s

" #)

es 1 0

x0 c s c 0

s

e 2 0 e 1 e 1

1 c01 2 c1 1

2:133

26 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

0

As indicated earlier, when s\ cx0 (or t\cx1 ), the step function is zero (u 0) and

1

0 0

it has a value of one (u 1) when s [ cx0 . Therefore, during the time interval s\ cx0

1 1

the step function in Eq. 2.133 is set to zero (u 0) or else it is set to 1.

Equation 2.133 is used to compute the dimensionless stress distribution for zero

stress gradient at the surface.

Since the laser power intensity varies with time, temporal variation of the pulse

intensity is incorporated in the stress analysis. In addition, the laser pulse

intensity distribution can be resembled through employing a time exponentially

varying pulse intensity profile. The boundary conditions are important formu-

lating the stress states in the irradiated material. Consequently, the closed form

solutions for the thermal stress generated inside the substrate material are pre-

sented due to different boundary conditions for the time exponentially decaying

laser pulse under the following sub-headings in line with the previous studies

[35].

The Fourier heat transfer equation due to time exponentially decaying laser

heating pulse can be written as:

o2 T I1 d bt dx 1 oT

e e 2:134

ox2 k a ot

where I1 = (1-rf) Io.

It can be assumed that the convection losses from the surface are neglected

during the heating pulse. This implies the insulated boundary at the substrate

surface. In addition, since the substrate material is assumed to be semi-infinite

body, temperature at a depth approaching to infinity becomes the same as the

initial temperature of the substrate material. Therefore, the corresponding

boundary conditions are:

The insulated boundary condition at the surface yields:

At the free surface:

oT

x 0 ) 0 2:135

ox x0

2.3 Time Exponentially Varying Laser Pulse Heating 27

infinitely below the surface, i.e.:

At x 1 ) T1; t 0 2:136

Due to the simplicity the initial temperature of the substrate material is set

to zero and temperature is assumed to be uniform inside the solid substrate at

time equal to zero (before heating initiates). Therefore, the initial condition

becomes:

At t 0 ) Tx; 0 0 2:137

The closed form solution of Eq. 2.134 can be obtained through a Laplace

Transformation method after employing the appropriate boundary and initial

conditions. In this case, the Laplace Transformation of Eq. 2.134 with respect to t,

results:

o2 T I1 d edx 1

sTx; s Tx; 0 2:138

ox2 k s b a

Introducing the initial condition and rearranging Eq. 2.138 yields:

o2 T 2 I1 d edx

h T 2:139

ox2 k s b

where h2 = s/a and s is the transform variable. Equation 2.139 has the solution:

I1 dedx

Tx; s Aehx Behx 2:140

k s b d2 h2

where A and B are constants. Introducing the boundary conditions will allow

determining the constants A and B, i.e.:

I1 d2

A 0; and: B 2:141

khs b d2 h2

After substituting the values of A and B in Eq. 2.140, it yields:

" #

s I1 d d exphx expdx

Tx; 2:142

ks b h h2 d2 h2 d2

The inverse Laplace Transform of Eq. 2.142 gives the temperature distri-

bution inside the substrate material in space (x) and time (t) domain as follows

[35]:

28 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

Tx;t

8 2 r!

3 9

>

> b x p >

>

>

> 6 exp ix Erfc p i bt 7 >

>

>

> r 6 a 2 at 7 >

>

>

> a 6 7 >

>

>

> id expbt 66 7 >

>

>

> r !

7 >

>

>

> b 6 p

7 >

>

>

< 4 b x 5 >

=

I1 d a exp ix Erfc p i bt

a 2 at

2k bad >2

> >

>

>

> 2

3>>

>

>

>

> x p p x >

>

>

> 6 expErfc p d at expdxErfc d at p 7 >

>

>

> 2 6 2 at 2 at 7 > >

>

> expad t4 5 >

>

>

> >

>

: ;

2expbt dx

2:143

where Erfc is the complementary error function. Equation 2.143 is the closed form

solution for temperature distribution. The temperature distribution in non-dimen-

sional form is possible by defining dimensionless quantities and substituting in

Eq. 2.143. The dimensionless quantities are:

Tkd b

x xd: t ad2 t: T :b 2 2:144

I1 ad

The dimensionless temperature distribution becomes:

T x ;t

8 2 r! 3 9

>

x p

>

>

> b >

>

>

> 6 exp ix p i b 7 >

>

>

> r 6 t

2 t 7 >

>

>

> t 6 7 >

>

>

> i exp b

6 7 >

>

>

> b 6 r!

7 >

>

>> 6 b

x p 7 >

>

1 t < 4 exp ix

Erfc p i b 5 =

t

2 t

2 b t >

> >

>

>

> 2

3>>

>

>

>

> x p p x >

>

> exp x Erfc p t expx Erfc t p >

>

> 6 7 >

>

> exp t 6

2 t 2 t 7>

>

>

>

> 4 5 >

>

>

:

>

;

2expb x

2:145

To obtain the closed form solution for the stress distribution inside the semi-

infinite one-dimensional substrate material, equation governing the momentum in

a one-dimensional solid for a linear elastic case can be considered. In this case, the

momentum equation becomes the same as Eq. 2.16.

Since the substrate surface is free to expand during the heating process, a free

stress condition is adopted at the surface of the substrate material. Moreover, as the

depth below the free surface increases to infinity, the thermal stress approaches the

initial stress conditions, which is the stress free condition, i.e. it is assumed that

the substrate material is initially free from any stresses. Therefore, the corre-

sponding boundary conditions are:

At x 0 ) rx 0 2:146

2.3 Time Exponentially Varying Laser Pulse Heating 29

and

At x 1 ) rx 0 2:147

For the temperature, the boundary conditions is:

oT

At x 0 ) 0 2:148

ox

and

At x 1 ) T 0 2:149

Since the substrate material is assumed to be initially stress free and the thermal

stress becomes the same as the initial stress condition as time approaches infinity,

the initial and final conditions for the stress field yield:

At t 0 ) rx 0 2:150

and

At t 1 ) rx 0 2:151

The initial condition for temperature is:

At t 0 ) Tx; 0 0 2:152

Since the laser pulse decays exponentially with time, the final and the initial

temperature of the substrate material becomes the same as the time approaches

infinity, i.e.:

At t 1 ) Tx; 1 0 2:153

Taking the Laplace Transformation of Eq. 2.16 with respect to time yields:

o2 r

x 1 2 :

2

2 s r x x; s srx x; 0 rx x; 0

ox c1 2:154

h : i

s sTx; 0 T x; 0

c2 s2 Tx;

x x; s and Tx;

perature respectively in the x and s domains.

By substituting the initial conditions, Eq. 2.154 reduces to:

o2 r

x s2 s

r x x; s c2 s2 Tx; 2:155

ox2 c21

Considering the temperature distribution for time exponentially varying pulse

and substituting into Eq. 2.155, and solving for the stress field, yields:

o2 r

x s2 2 I1 d d exphx expdx

r

x x; s c 2 s 2:156

ox2 c21 ks b hh2 d2 h2 d2

30 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

o2 r

x s2 I1 d2 c2 s2 I1 dc2 s2

2rx x; s 2

ehx edx 2:157

ox 2 c1 ks bh h2 d ks b h2 d2

Now let W1 and W2 are defined as:

I1 d2 c2 s2

W1 2:158

ks bh h2 d2

and

I1 dc2 s2

W2 2:159

ks b h2 d2

Then Eq. 2.157 becomes:

o2 r

x s2

r x x; s W1 ehx W2 edx 2:160

ox2 c21

The homogeneous and the particular solutions for Eq. 2.160 are:

sx sx

r

x h C1 ec1 C2 e c1 2:161

while the particular solution has two parts, the first part is:

rx p1 Q1 ehx

2:162

W1

Q1 2 2:163

h2 cs2

1

rx p2 Q2 edx

2:164

W2

Q2 22 2:165

d cs2

1

sx sx

x g C1 ec1 C2 e c1 Q1 ehx Q2 edx

r 2:166

2.3 Time Exponentially Varying Laser Pulse Heating 31

Then,

sx W1 hx W2

rx g C2 e c1

2 e 2 s2 edx 2:167

h2 cs2 d c2

1 1

sx

rx g C2 e c1 g1 sehx g2 sedx

2:168

where:

I1 d2 c2 s2

g1 s 2:169

ks bh h2 d2 h2 cs2

2

and

I1 dc2 s2

g2 s 2:170

ks b h2 d2 d2 cs2

2

Now substituting for h s=a and simplifying the expressions for g1(s) and

g2(s) yields:

p

I1 d2 c2 saac21

g1 s c2

2:171

ks bh s ad2 s a1

and

g2 s 2:172

ks b s ad2 s2 c21 d2

Rearranging the two expressions as:

" p #

s

g1 s C3 2:173

s bh s ad2 s c21 =a

and

" #

s2

g2 s C4 2:174

s b s ad2 s c1 ds c1 d

where:

p

I1 d2 c2 aac21

C3 2:175

k

32 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

I1 c2 c21 da

C4 2:176

k

Using partial fraction, the following relations can be obtained:

2 p p 3

s a2 s

6 ad2 b ad2 c2 =a s ad2

6 1 c21 ab c21 a2 d2 s c21 =a 77

g1 s C3 6 p 7

4 s 5

2

2

b ad b c1 =a s b

2:177

and

2 3

ad2 c1

6 2 7

6 ad b a2 d2 c21 s ad2 2c1 d bc1 ads c1 d 7

6

g2 s C4 6 7

c1 b2 7

4 5

2

2 2

2c1 d bc1 ads c1 d 2

b ad b c1 d s b

2:178

Consider the boundary condition at the surface in the stress field, where at

x0)r x 0, the constant in Eq. 2.167 becomes:

C2 g1 s g2 s 2:179

Then:

ps sx sx

x x; s g1 se

r ax g1 se c1 g2 sedx g2 se c1 2:180

Finding the solution for rx in the x and t domain, we should take the inverse

Laplace Transform for each term in 2.180. To perform this, the following desig-

nations are introduced:

ps sx

Term1 g1 se ax Term2 g1 se c1 2:181

sx

Term3 g2 sedx Term4 g2 se c1

Consequently, the solution for stress distribution becomes the summation of the

inverse Laplace Transforms of the above terms. Therefore, the Laplace inversion

of Terms (Term1, Term2, Term3, Term4) are:

2.3 Time Exponentially Varying Laser Pulse Heating 33

1 Term1

8 2 0

13 9

>

> dx

p x >

>

>

> 6 B e Erfc d at 2 at C7 > p >

>

> 1 6 2 p B C 7 >

>

>

> x2

6p e 4at d a e B d2

at C7 >>

>

> 6 B

C 7 >

>

>

>

2 2 2

2 ad b ad c1 =a 4 pt @ dx p x A5 >>

>

> p >

>

>

> e Erfc d at >

>

>

> 2 at >

>

>

> >

>

>

> 2 0 r

1 3 >

>

>

> c1 x t x >

>

>

> e Erfc c1

a p >

>

>

< 6 B a 2 at C 7 >

=

a 2 6 2 x 2 c 1

c2 t B C 7

C3 2 2 6p e 4at p e a B 1

C7

> 2

2 c1 ab c1 a d 4 pt 2 6 a B r

C 7 >

>

> @ c1 x t x A5 >

>

>

> e a c1 p >

>

>

> a at >

>

>

> 2 >

>

>

> 2 0 p

1 3 >

>

>

> p

x >

>

>

> xi a b >

>

>

> 6 B e Erfc i bt p C 7 >

>

>

> 1 6 2 p B 2 at C 7 >

>

>

> 6 p

x 2

bt B C 7 >

>

>

> 2 6 e 4at i b e B p

C 7 >

>

>

> 2 bad bc 2 =a 4 pt @ p

x A 5 >

>

>

> 1

e xi b

a Erfc i bt p >

>

: ;

2 at

2:182

and

8 q 9

> p 2 >

>

> p ad e

1 2 ad tx=c 1

Erf 2

ad t x=c1 >

>

>

> ptx=c1 >

>

>

> >

>

>

> 2 2 >

>

>

> ad c 2 =aad b >

>

>

> 0 q

1

q

1 >

>

>

> c2 >

>

>

< p

1 c 2 1

a1 e a tx=c1 Erf c 2

1

t x=c1 >

=

1

Term2 C3 a2 B B p tx=c a C

1

2 C

>

> @ A>>

>

> c1 ab c21 a2 d2 >

>

>

> >

>

>

> >

>

>

> p p

>

>

>

> p

1 i b e btx=c1

Erf i b t x=c >

>

>

> ptx=c1

1 >

>

>

> >

>

: 2 2 ;

b c1 =a ad b

x

1 t

c1

2:183

and

8 2

9

>

>

> a2 d2 ead t c1 ec1 dt >

>

>

< ad2 ba2 d2 c2 2c1 d bc1 ad

> >

=

1 dx 1

Term3 C4 e

>

> c1 ec1 dt b2 ebt >

>

>

> >

>

: 2c d bc ad 2

2 2d ;

2

1 1 b ad b c 1

2:184

34 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

2 2

3

a2 d2 ead tx=c1 c1 ec1 dtx=c1

6 2 7

6 ad b a2 d2 c21 2c1 d bc1 ad 7

1 Term4 C4 6

6

7

7

4 c1 ec1 dtx=c1 b2 ebtx=c1 5

2

2 2 d2

2c1 d bc1 ad b ad b c

1

x

1 t

c1

2:185

where 1t cx1 is a unit step function and Erf (y) is the error function of the

variable y. It should be noted that the value of a unit step function takes one or

zero, i.e.:

x x x x

t ) 1 t 0 and t [ ) 1 t 1 2:186

c1 c1 c1 c1

The closed form solution of stress distribution can be written as:

-1

where represents the inverse sign of Laplace Transformation.

Presenting the stress distribution in dimensionless form, the additional

dimensionless quantities are defined, i.e.:

c1

c1 2:188

ad

and

krx

rx 2 2:189

I1 da2 c

2 c1

U [1] is the unit step function, which is U1 t xc . The values of unit

1

step function are:

x x x x

t ) U1 t

0 and t

[ ) U1 t

1 2:190

c1 c1 c1 c1

2.3 Time Exponentially Varying Laser Pulse Heating 35

Therefore, for the dimensionless stress distribution, the followings are resulted:

8 2 0

13 9

> x

p x >

>

> p

x 2 e t p >

>

>

> 1 6 t e 4t t B 2 t C7 >

>

>

> 6 B C 7 >

>

>

> 6 p B

C 7 >

>

>

> t

b 1 c 2 4 p 2 @ p x A 5 >

>

>

> 1 e x

erfc t p >

>

>

> >

>

>

> 2 t >

>

>

> 2 0

13 >

>

>

> p x >

>

>

> p

x 2 e c1 x

erfc c

t p >

>

< 6 t e 4t c1 t B

2 1

2 t C7 =

1 6 c 1 t e B C7

rx 1 2 2 6 p B

C7

>

> c1 1 c1 t b 4 p 2 @ p x A5 >

>

>

>

ec1 x erfc c1 t p >

>

>

> >

>

>

> 2 t >

>

>

> 2 0 p p

1 3 >

>

>

> b x >

>

>

> 2 p e ix t erfc i b p >

>

>

> t

3=2 6 e

x

i b

e b B

2 t C 7 >

>

>

> 6 4t

B C 7 >

>

>

> 6 p B p

C7 >

>

>

> b

t b c t 4 2 p 2 @ b p x A 5 >

>

>

: 1 e ix

t

erfc i b p >

;

2 t

2:191

8 2 3 9

>

> q >

>

>

> 1 6 t t

x

7 >

>

>

> 4pqt e =c 1 t x =c1 5 >

>

>

>

b 1 c 2

x =c

>

>

>

> t 1 p t >

>

>

>

1 >

>

>

> 2 3 >

>

>

> q

>

>

< 1 1 =

6 2

c1 t c1 x

2 7

rx 2 4pqc1 e Erf c1 t c1 x 5 U1

> c 1

> 2 2

c1 b =t p t x =c1 >

>

>

> 1 >

>

>

> 2 3>>

>

>

>

>

> p q

>>

>

>

> t 3=2 6 t p

b 1x

=c t

7>>

>

>

>

> 4pqi b e 1 erf i b 1x =c t

5 >

>

>

: t b b c 2 t

1 >

;

1 p t x =c1

2:192

8

9

>

> t et x t c1 ec1 t x >

>

> >

>

>

>

> 2 b 2 c1 t b c1 1 >

>

>

> 1 c 1 t >

>

>

> >

>

>

< c

t

x >

=

t c1 e 1

rx 3

2:193

>

> 2 c1 t b c1 1 >

>

>

> >

>

>

> 2 b x >

>

>

> b t e >

>

>

> >

>

>

: b t b 2 c1 t 2 >

;

8 9

t c1 ec1 t x

>

> t et x =c1 t c1 ec1 t x >

>

>

> 2 >

>

>

> 2 c1 t b c1 1 > >

< 1 c1 t b 2 c1 t b c1 1 =

rx 4 U1

>

>

b 2 t eb 1x =c1 t

>

>

>

> >

>

>

>

:

2 >

>

;

b t b 2 c1 t

2:194

36 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

rx rx 1 rx 2 rx 3 rx 4 2:195

Equations 2.145 and 2.195 are used to compute the dimensionless temperature

and stress distributions inside the substrate material.

In practical applications, an inert assisting gas is used to protect the surface from

high temperature exothermic reactions. This intern results in surface cooling due to

convection effect of the assisting gas. Therefore, the convective boundary condi-

tion needs to be incorporated for laser gas assisted heating situations.

The Fourier heat transfer equation for a laser pulse decaying exponentially with

time can be written similar to Eq. 2.134. Therefore, the boundary conditions and

the closed form solution of the heat equation is given lightly.

The substrate material is considered as a semi-infinite body and heated by a

laser beam on the surface. The convective boundary condition is assumed be on

the substrate surface. In addition, as the depth is considered to extend to infinity

and the temperature to go down to zero. Heating occurs in the surface region

during the laser pulse. Therefore, the corresponding boundary conditions are:

h i h

At x 0 ) T0; t T0 2:196

ox x0 k

and

At x 1 ) T1; t 0 2:197

Initially substrate material is assumed to be at uniform temperature. Therefore,

the initial condition is:

At t 0 ) Tx; 0 0 2:198

The Laplace transformation of Eq. 2.16 with respect to t, results in:

o2 T I1 d edx 1

sTx; s Tx; 0 2:199

ox2 k s b a

Introducing the initial condition and rearranging the Eq. 2.199, it yields:

o2 T 2 I1 d edx

g T 2:200

ox2 k s b

where g2 = s/a and s is the Laplace transform variable. Equation 2.203 has the

solution:

2.3 Time Exponentially Varying Laser Pulse Heating 37

Tx; 2:201

ks b d2 g2

where A and B are constants. Introducing the boundary conditions will enable

calculation of the constants A1 and A2, i.e.:

A1 = 0, and:

I1 dh dk hT0

A2 2 2:202

2

ks b d g h kg s h kg

Tx; 2 2 2:203

ks b d g2 h kg sh kg k d g2 s b

which gives the solution for the temperature distribution in the Laplace domain.

The inverse Laplace Transform of Eq. 2.206 provides the temperature distri-

bution within the substrate material in space x and time t. The mathematical

arrangements of the Laplace inversion of Eq. 2.203 are given in the previous study

[35]. Therefore, the equation after the Laplace inversion is given below as:

8 p p 9

ead tdx Erfc 2px ead tdx Erfc 2px

2 2

>

> d at d at >

>

>

> at at >

>

>

> p p p p

>

>

>

> 2 h a 2 h a >

>

>

> 2 b ad k d a 2 b ad k d a >

>

>

> >

>

>

> p

p

p

p

>

>

>

< >

3=2

I1 da h kd bt

e e b=axi

Erfc 2p x

bt i e bt

e b=axi

Erfc p bt i =

x

Tx; t at 2 at

k 2 > p p p p >

>

> 2 b ad2 h k a bi 2 b ad2 h k a bi >

>

>

> >

>

>

> >

>

>

> p >

>

>

> p

hx=k h2 =k2 at p

x h dx ad 2

t bt >

>

>

> h a e e Erfc k at ke e e >

>

>

> h2 a h2 a

2 at

p

>

>

: k 2 2 ;

k2 b k2 ad ab ad h kd

" !

#

x x h p

T0 Erfc p ehx=k eh =k Erfc p

2 2

at

2 2 at k

2:204

where Erfc is the complementary error function. Equation 2.204 is the closed form

solution for temperature distribution. The temperature distribution can be

expressed in a non-dimensional form by introducing dimensionless quantities and

substituting in Eq. 2.204. The dimensionless quantities are:

Tkd h

x xd: t ad2 t: T : h : b bt

I1 dk

38 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

82 p p p p 39

> x t

b t i >

> 6 e e Erfc 2pt t

x x

> x ex et Erfc 2p t eb e b x i Erfc 2p >

>

> t t 7>>

>

> 6 1

1

p 7>>

>

>

> 6 2 b 1 h 2 b 1 h

2b 1 h b i 7 >

>

>

> 6 7 >

>

>

>66 7 >

>

>

> p

p 7 >

>

>

> 6 b t b x i x h x h 2 t x

p

7

7 >

>

> 6

>6 e e Erfc 2 t b t i h e e

p Erfc 2 t h t

p

e x

e t

e b t >

>

>

> 7 >

>

>4 p 5>>

>

>

> b i b

2 2

1 b 1h 1 >

>

>

> 2 b 1 h 2 h h >

>

< =

T x ;t h 1

>

> >

>

> 2

> p p p 3 > >

>

> p >

>

>

> e x t

e Erfc x

p t e x t

e Erfc x

p t e c

e c x i

Erfc x

p c t i >

>

> 6

> 2 t 2 t 2 t 7 >

>

>

> 6 p

7 >

>

> 6

> 2 c 1h 1 2 c 1h 1 2 c 1h c i

7>>

>

> 6 7>>

>

> 7>>

> 6

> p

6 ec t e c x i Erfc p p p >

>

> 7>

2

x t i h x h t x

>

> 6 c h e e Erfc p h t e x

t

e e

c t 7>>

>

> 4 2 t

p

2 t

5>>

>

> >

>

: 2c 1h c i 2 h 2 c h 2 1 c 1h 1 ;

2:205

In order to solve for the stress distribution within the substrate it is possible to

consider the equation governing the momentum in a one-dimensional solid for the

linear elastic case. Therefore, Eq. 2.16 is used to formulate thermal stress distri-

bution in the substrate material.

In order to solve the momentum equation (Eq. 2.16) it is necessary to establish

the initial conditions for stress and temperature fields. In this case, the substrate

material is assumed to be free from stresses initially (at time = 0) and as the time

extends to infinity, the stress free state must apply in the substrate. The same initial

condition for the temperature is applied as in Eq. 2.198 provided that as time

approaches infinity, the temperature in the substrate material reduces to zero. This

is due to the fact that the laser pulse decays exponentially with time; therefore, as

time approaches infinity, the laser pulse intensity becomes zero. Therefore, the

initial and boundary conditions for the stress field are:

At t = 0 ) rx 0 2:206

and

At t = 1 ) rx 0 2:207

and

At x 0 ) rx 0 2:208

and

At x = 1 ) rx 0 2:209

Taking the Laplace Transformation of Eq. 2.16 with respect to time yields:

x 1 h 2

o2 r :

i h

2

: i

s r

x x; s srx x; 0 r x; 0 c 2 s Tx; s sTx; 0 T x; 0

ox2 c21 x

2:210

2.3 Time Exponentially Varying Laser Pulse Heating 39

x x; s and Tx;

temperature respectively in the x and s domains.

By substituting the initial conditions, Eq. 2.210 reduces to:

o2 r

x s2 s

r x x; s c2 s2 Tx; 2:211

ox2 c21

Considering the temperature distribution in a Laplace domain for an expo-

nentially decaying pulse with time, equation (Eq. 2.200), and substituting it into

Eq. 2.211, and solving for the stress field, yields:

"

o2 r

x s2 2 I1 dh dkegx

2rx x; s c2 s

ox 2 c1 ks b d2 g2 h kg

# 2:212

hT0 egx I1 dedx

sh kg k d2 g2 s b

I1 dh dkc2 s2 hT0 c2 s

M1 2 2:213

2

ks b d g h kg h kg

and

I1 dc2 s2

M2 2 2:214

k d g2 s b

Then, Eq. 2.212 becomes:

o2 r

x s2

r x x; s M1 egx M2 edx 2:215

ox2 c21

The complementary and the particular solutions of Eq. 2.215 are:

sx sx

r

x h A3 ec1 A4 e c1 2:216

While the particular solution has two parts, the first part is:

ps

x p1 G1 e ax

r 2:217

M1

G1 2 2:218

g2 cs2

1

40 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

x p2 G2 edx

r 2:219

M2

G2 22 2:220

d cs2

1

sx sx

x g A3 ec1 A4 e c1 G1 egx G2 edx

r 2:221

Then, Eq. 2.21 reduces to:

sx

ps

rx g A4 e c1 G1 e ax G2 edx

2:222

Consider the boundary condition of the stress field at the surface, where at

x 0 ) orx 0, the constant in Eq. 2.222 becomes:

r

c1 s

A4 G1 s dG2 s

s a

Therefore, Eq. 2.222 becomes:

ps c1 sx

x x; s G1 se

r ax

G1 s p e c1

sa

2:223

c1 d sx

G2 sedx G2 s e c1

s

Finding the solution for rx in the x and t domain, one should take the inverse

Laplace Transform for each term in Eq. 2.223. To do this, the following terms are

introduced:

ps sx

Term1 G1 se ax Term2 G1 s pc 1

sa

e c1

sx 2:224

Term3 G2 sedx Term4 G2 s c1sd e c1

Consequently, the solution for the stress distribution is the summation of the

inverse Laplace Transforms of the above terms. Therefore, the Laplace inversion

of Terms (Term1, Term2, Term3, Term4) can be stated as follows:

Term1 is composed of the terms:

Term1 Term11 Term 21 2:225

2.3 Time Exponentially Varying Laser Pulse Heating 41

where

2 ps 3

ax

I1 dh dkc2 4 se

Term11

k 2 p 5

d s=a s b h k s=a 1=a s=c2 1

2:226

and

2 p 3

as x

e

Term21 hT0 c2 4 p 5 2:227

h k s=a 1=a s=c21

I1 dh dkC2

Let C10 2:228

k

and

C20 hT0 c2 2:229

Therefore, the Laplace transformation of Term1 can be written as:

where

1 Term411 1 Term511 2:231

1 1 1

Term611 Term711 Term811

The Laplace transformations of the terms are:

2 p p

3

h a h22athxk h at x

p

6 k e Erfc p

2 at 7

k

c21 h2 k3 a2 aC10 6 k 7

1 Term111

2 6 p p

7

2 2 2 2 2 2 2

2 a h c1 k h a k b h k d 2 4 h a h 2athxk

2 h at x 5

ek Erfc p

k k 2 at

2:232

and

2

3

p btpbx p x

p 6 b e a Erfc bt p

c2 ha3 bC10 6 2 at 7 7

1 Term211 2

1 2 6 p p p

7

2 b ad h2 a k2 b c1 ab 4 b x 5

be axbt Erfc bt p

2 at

2:233

42 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

p

c31 ha4 aC10

1 Term311

2 a2 h2 c21 k2 c21 ab d2 a2 c21

2 r

3

c1 c21 tc1 x t x

6 p e a a Erfc c1 p 7 2:234

6 a a 2 at 7

6

6 7

r

7

4 c1 1 1c2 t c x

t x 5

p e a a Erfc c1 p

a a 2 at

and

p

1 c21 ha2 adC10

Term411

2 b ad2 d2 a2 c21 h2 k2 d2

2

3

p ad2 tdx p x

6 a de Erfc d at p 2:235

6 2 at 7 7

6

7

4 pad2 tdx p x 5

a Erfc d at p

2 at

and

p

1 c21 h2 k3 a2 aC10

Term511

2 a2 h2 c21 k2 h2 a k2 b h2 k2 d2

2 2

p p

3

2 x h a h22athxk h at x

6 p

exp e Erfc p 7

k

6 pt 4at k k 2 at 7

6 p

p

7

4 h a h2 at hx h at x 5

2 k p

e k Erfc

k k 2 at

2:236

and

p

1 c21 ka2 abC10

Term611

2 c21 ab h2 a k2 b b ad2

2 2

q

pb

3

2 x p

x

6 p exp btx a Erfc bt p 7

6 pt 4at 2 at 7

66 p pb

7

7

4 btx a

p x 5

be Erfc bt p

2 at

2:237

2.3 Time Exponentially Varying Laser Pulse Heating 43

and

p

c21 ka3 aC10

1 Term711

2 a2 h2 c21 k2 c21 ab d2 a2 c21

2 2

r

3

2 x c1 c21 tc1 x t x

6 p exp p e a a Erfc c1 p 7

6 pt 4at a a 2 at 7

66 7

r

7

4 c1 1 1c2 t c x

t x 5

p e a a Erfc c1 p

a a 2 at

2:238

p

c21 ka2 ad2 C10

1 Term811

2 b ad2 d2 a2 c21 k2 d2 h2

2 2

3

2 x p ad2 tdx p x

6 p

exp

4at

ade Erfc d at p 7

6 pt 2 at 7

6

7

4 p 2 p x 5

adead tdx Erfc d at p

2 at

2:239

The Laplace inversion of Term21 can be written as:

2:240

1 Term321 1 Term421

Therefore, the Laplace inversions of composing terms are:

2 p p

3

h a h22athxk h at x

p e Erfc p 7

c2 hka aC20 6

k

6 k k 2 at 7

1 Term121 1 2 2 6 p

p

7

2h a h c21 k2 4 h a h2 athx h at x 5

ek 2 k Erfc p

k k 2 at

2:241

and

2 r

3

c1 c21 tc1 x t x

p p e a a Erfc c1 p

1 c1 ha2 aC20 6 6 a a 2 at 7 7

Term221 6 r

7

2 a2 h2 c21 k2 4 c1 c21 tc1 x t x 5

p e a a Erfc c1 p

a a 2 at

2:242

44 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

2 2

p p

3

2 x h a h22 hxk h x

p

6 p

exp e Erfc p

2 at 7

k

c2 ka aC20 6 pt 4 k k 7

1 Term321 12 2 6 p

p

7

2 a h c21 k2 4 h a h22athx h at x 5

e k k Erfc p

k k 2 at

2:243

and

2 2

r

3

2 x c1 c21 tc1 x t x

p

6 p

exp p

e a a Erfc c p

a 2 at 7

1

c2 ka aC20 6 pt 4 a 7

1 Term421 12 2 6 r

7

2 a h c21 k2 4 c1 c21 tc1 x t x 5

p e a a Erfc c1 p

a a 2 at

2:244

The Laplace transform of Term2 can be written as:

Therefore, the Laplace inversions of composing terms are:

1 Term412 1 Term512 2:246

1 1 1

Term612 Term712 Term812

Knowing that:

I1 dh dkc2 c1

C30 p 2:247

ak

and

hT0 c2 c1

C40 p 2:248

a

Hence:

1 Term112

k2 c21 k2 h2 a2 h2 a k2 b h2 k2 d2

2 1 3

p 2:249

6 pt x=c1 7 x

6 7

6 p 71 t

4 h a h2 at hpa p

5 c1

e k Erf

2

t x=c1

k k

2.3 Time Exponentially Varying Laser Pulse Heating 45

and

1 Term212 2

b ad h2 a k2 b c21 ab

2 3

1

p

2:250

6 pt x=c1 7 x

46 7 1 t

p bt p 5 c1

be Erf bt x=c1

and

c2 ha5 C30

1 Term312 1 2

a2 h2 c21 k2 c1 ab d2 a2 c21

2 1 3

p 2:251

6 pt x=c1 7 x

6 7

6

7 1 t c

4 c1 c21 c1 p 5 1

t

p e Erf p t x=c1

a

a a

and

1 Term412

b ad d2 a2 c21 h2 k2 d2

2

2 3

1

p

2:252

6 pt x=c1 7 x

46 7 1 t

p ad2 t p p 5 c1

ade Erf ad t x=c1

and

c21 k5 a2 C30

1 Term512

a2 h2 c21 k2 h2 a k2 b h2 k2 d2

2:253

h2 a h22atx=c1 x

Dt x=c1 2 e k 1 t

k c1

and

p h i

1 c21 ka2 aC30 btx=c1

Term612 2 2

D t x=c 1 be

2 2

c1 ab

h a k b b ad

x

1 t

c1

2:254

46 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

p

c21 ka4 aC30

1 Term712

a2 h2 c21 k2 c21 ab d2 a2 c21

2:255

c21 c21 tx=c1 x

Dt x=c1 e a 1 t

a c1

and

p

1 c21 ka aC30

Term812

b ad2 d2 a2 c21 k2 d2 h2

h i

2:256

2 x

Dt x=c1 ad2 ead tx=c1 1 t

c1

The Laplace transformation of Term22 can be written as:

2:257

1 Term322 1 Term422

1 aC40 x

Term122 p 1 t 2:258

h pt x=c1 c1

and

c2 k2 aC40

1 Term222 21 2

h a h c21 k2

" p p r

#

1 h a h22at h a x x

p e Erf

k t 1 t

pt x=c1 k k c1 c1

2:259

and

ha3 C40

1 Term322 2 2

a h c21 k2

" r

#

1 c1 c21 t c1 x x

p p e a Erf p t 1 t

pt x=c a a c1 c1

2:260

2.3 Time Exponentially Varying Laser Pulse Heating 47

and

p

k aC40 x

1 Term422 D t x=c 1 1 t 2:261

h2 c1

and

p

c2 k3 aC40 h 2 a h2 a x

1 Term522 12 Dt x=c1 2 e k2 tx=c1 1 t

h2 c 1 k 2 h2 a2 k c1

2:262

and

p

Term622 2 2 Dt x=c1 e a 1 t

c 1 k h2 a2 a c1

2:263

and

" #

I1 dc2 s2 edx

Term3 2 2:264

k d s=a s b d2 s2 =c21

and

8 9

>

> c21 c1 dt dx c31 aec1 dtdx c21 ab2 edx >

>

>

>

2 d2 b ad2 >

>

I dc < 2b c1 dc1 ad 2c1 d bc1 ad b 2

c 1

=

1 1 2

Term3

k > > 2 3

c a d e2 ad 2

tdx >

>

>

> >

>

: 2 21 2 2

;

a d c1 b ad

2:265

and

" #

sx

I 1 d2 c 2 c 1 se c1

Term4 2 2:266

k d s=a s b d2 s2 =c21

and

8 9

>

> c21 aec1 dtx=c1 c21 aec1 dtx=c1 >

>

>

> >

I d2

c c < 2db c1 dc1 ad 2dc1 d bc1 ad > = x

1 1 2 1

Term4 1 t

k >

> c21 abebtx=c1

2

c2 a2 ead tx=c1 >

> c1

>

> 2 1 >

>

: 2 2 2

2 2 2

2 ;

c1 d b b ad c1 a d b ad

2:267

48 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

where 1t cx1 is a unit step function, Erf (y) is the error function of the variable y

and Dt x=c1 is the Dirac delta function. The unit step function has the values

of 0 for t cx1 and 1 for t [ cx1 :

The closed form solution of the stress distribution can be written as:

where -1 represents the inverse sign of the Laplace transformation.

The additional dimensionless quantities are defined to present the stress dis-

tribution in the dimensionless form, i.e.:

c1 krx x

c1 : rx and U1 t

2:269

ad I1 da2 c2 c1

where U (1) is the dimensionless unit step function. Therefore, for the dimen-

sionless stress distribution, the followings are resulted:

rx 1 rx 11 rx 21 2:270

where

rx 11 rx 111 rx 211 rx 311 rx 411 rx 511 rx 611 rx 711 rx 811

2:271

In the dimensionless form:

2

3

h2 t h x

p x

h e Erfc h t p

h 1h2 c2 6 2 t 7

1 6 7

rx 111 6

7

2h2 1 h2 c2 2

1 h b =t

4 h2 t h x p x 5

h e Erfc h t p

2 t

2:272

and

2 p p

3

b bt x x

6 e Erfc b p

h 1h c2

1 b

6 2 t 7 7

rx 211

6 p

7

2b =t 1h2 b =t c2

1 b =t 4

b p x 5

eb t x Erfc b p

2 t

2:273

2.3 Time Exponentially Varying Laser Pulse Heating 49

and

2

3

2 c2 p

x

1 t c1 x Erfc

c1

t p 7

h 1h c3 6 c1 e 2 t

1 6 7

rx 311 2 2

2

2

6

7

2 h c1 1 c1 c1 b =t 4 2 p

x 5

c2

1 e

c1 t c1 x

Erfc c1 t p

2 t

2:274

and

2

3

t x

p x

Erfc t p 7

h 1h c2 6e

1 6 2 t 7

rx 411 6

7

2h2 1 1 c2

1 1 b

=t 4

t x p

x 5

e Erfc

t p

2 t

2:275

and

h 1h2 c2

rx 511 1

2h2 1 h2 c2 2

1 h b =t

2 2

3

2 x h2 t h x

p x

p

exp h e Erfc h t p

6 pt 4t 2 t 7

6 7

6

7

4 2 p

x 5

h t h x

h e Erfc h t p

2 t

2:276

and

h 1c2

1 b

rx 611

2t b =t 1h2 b =t c2

1 b =t

2 2

3

2 x

p

6 pt

exp 7

6 4t 7

6 r p p

7 2:277

6

b b b x x 7

6 7

6 e t Erfc b p

7

6 t 2 t 7

6 7

6 r p p

7

4 b b b x x 5

e t Erfc b p

t 2 t

50 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

h 1c4

rx 711 2 1 2

2 h c1 1 c2

2

1 c1 b =t

2 2

3

2 x c2 t c1 x

p x

6 p

exp c1 e Erfc c1 t p 7

1

6 pt 4t 2 t 7

6

7

4 2 p

x 5

c1 ec1 t c1 x Erfc c1 t p

2 t

2:278

and

h 1c2

rx 811 1

2 1 h2 1 c2

1 1 b =t

2 2

3

2 x t x

p x

6 p

exp

4t

e Erfc t p 7 2:279

6 pt 2 t 7

6

7

4 p

x 5

et x Erfc t p

2 t

The dimensionless form of (r*x)21 is:

rx 21 rx 121 rx 221 rx 321 rx 421 2:280

where

2

3

h2 t h x

p x

Erfc h t p 7

T0 h c2 6h e

1 6 2 t 7

rx 121 2 6

7 2:281

2 h c2 1

4 2

h t h x

p

x 5

h e

Erfc h t p

2 t

and

2

3

c2 p x

c e 1 t c1 x Erfc c t p

T0 h2 c1 6 6

1 1

2 t 7

7

rx 221 2 6

7 2:282

2 h c2 4 c2 t c x p x 5

1

c1 e 1 1 Erfc c t p

1

2 t

2.3 Time Exponentially Varying Laser Pulse Heating 51

and

T0 h c2

1

rx 321 2

2 h c2 1

2 2

3

2 x h2 t h x

p x

6 p

exp

4t

h e Erfc h t p 7

6 pt 2 t 7

6

7

4 2 p

x 5

h eh t h x Erfc h t p

2 t

2:283

and

2 2

3

2 x c2 p x

p

exp c e 1 t c1 x Erfc c t p

T0 c2 6 4t 1 1 7

1 h 6 pt 2 t 7

rx 421 2 6

7

2 h c2 4 2 p x 5

1

c1 ec1 t c1 x Erfc c1 t p

2 t

2:284

The dimensionless form of (r*x)2 is:

rx 2 rx 12 rx 22 2:285

where

rx 12 rx 112 rx 212 rx 312 rx 412

2:286

rx 512 rx 612 rx 712 rx 812

and

h 1h c3

rx 112 1

h2 1 h2 c2 2

1 h b =t

2 3

s! 2:287

1

x 7

6 h2 t

4q

h e Erf h t c 5 U1

p t x =c 1

1

52 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

h 1c3

1 h

rx 212

b 1h b =t c2

=t 2

1 b =t

2 3

r s!

6 1 b b x 7

4q

eb Erf b 5 U1

p t x =c t c1 t

1

2:288

and

h 1c3 1h

rx 312 2

h c2

1 1 c2

1 c2

1 b =t

2 3

s! 2:289

1

x 7

6 c2 t

4q

c1 e 1 Erf c1 t c 5 U1

p t x =c 1

1

and

h 1c3

1 h

rx 412

1 h2 1 c2

1 1 b =t

2 3

s! 2:290

6 1 x 7

4q

et Erf t 5 U1

p t x =c c1

1

and

h 1c3

rx 512 1

h2 1 h2 c2 2

1 h b =t

2:291

2 h2 x

2 h t c1

D t x =c1 h e U1

and

h 1c3

1

rx 612

b =t 1h2 b =t c2

1 b =t

2:292

b b bcx

D t x =c1 e 1 U1

t

2.3 Time Exponentially Varying Laser Pulse Heating 53

and

h 1c3

rx 712 1 2

h2

c2

1 c2

1 1 c1 b =t 2:293

h i

c2

D t x =c1 c2 1 e

1 t c1 x U1

and

h 1c31

t xc

rx 812 D t x =c1 e 1 U1

1 h2 1 c2

1 1 b =t

2:294

The dimensionless for of (r*x)22 is:

rx 22 rx 122 rx 222 rx 322 rx 422 2:295

where

T0 c1

rx 122 q

U1 2:296

p t x =c1

and

2 3

s!

3

T c 6 1 h2 t

x 7

rx 222 2 0 1 2 4q

h e Erf h t c 5 U1

h c1 p t x =c 1

1

2:297

and

2 3

s!

T c h2 1

x 7

0 1 2 6

2

rx 322 4q

c1 ec1 t Erf c1 t 5 U1

h2 c1 p t x =c1 c1

2:298

and

T c

rx 422 0 1 D t x =c1 U1 2:299

h

and

T0 c3

1

2 h2 x

2 h t c1

rx 522 D t x =c 1 h e U1 2:300

h h2 c21

54 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

T c h h c2

i

rx 622 20 1 2 D t x =c1 c2

1 e

1 t c1 x U1 2:301

h c1

and

c3 t ec1 t x c3 t ec1 t x

rx 3 1 1

2 b c1 t c1 1 2 c1 t b c1 1

2:302

c2 t b2 eb x c2 t x

1 t e

1 2

b t b c2 1 t

2 b t 1 c2

1

and

2 3

c3

1 t ec1 t x c3

1 t ec1 t x

6 2 c1 t b c1 1 7

6 2 b c1 t c1 1 7

rx 4 6

6 b b x

t x

7 U1

7 2:303

2 c t c

4 c3

1 t b e 1 c3

1 t e 1 5

2 2

2

b t c 1 t b 2

b t c1 1

rx rx 1 rx 2 rx 3 rx 4 2:304

The dimensionless temperature (Eq. 2.205) and stress distributions (Eq. 2.304)

are computed during the heating pulse.

The Fourier heat transfer equation due to time exponentially decaying laser pulse

can be written similar to Eq. 2.134. Therefore, the boundary conditions and the

solution of the heat equation is given lightly.

In the analysis, no heat convection is considered from the free surface of the

substrate material. The depth well below the surface (x % ?), temperature

remains the same. Therefore, the corresponding boundary conditions are:

At the surface:

oT

0

At x 0 ) 2:305

ox
x0

and

At depth infinity:

At x 1 ) T1; t 0 2:306

2.3 Time Exponentially Varying Laser Pulse Heating 55

initial condition is:

Initially:

At t = 0 ) Tx; 0 0 2:307

The Laplace Transformation of Eq. 2.16 with respect to t, results:

o2 T I1 d edx 1

sTx; s Tx; 0 2:308

ox2 k s b a

Introducing the initial condition and rearranging Eq. 2.308 yields:

o2 T 2 I1 d edx

h T 2:309

ox2 k s b

where h2 = s/a and s is the transform variable. Equation 2.309 has the solution:

I1 dedx

Tx; s Aehx Behx 2:310

k s b d2 h2

where A and B are constants. Introducing the boundary conditions will allow

determining the constants A and B., i.e.:

" #

s I1 d d exphx expdx

Tx; 2:311

k s b h h 2 d 2 h2 d2

The inverse Laplace Transform of Eq. 2.311 gives the temperature distribution

inside the substrate material in space x and time t domain as follows:

I1 d a

Tx; t

2k b ad2

8 2 r! 3 9

>

> b x p

>

>

>

> 6 exp ix Erfc p i bt 7 >

>

>

> r 6 a 2 at 7 >

>

>

> a 6 7 >

>

>

> id expbt 6 7 >

>

>

> 6 r ! 7

7 >

>

>

> b 6 p

>

>

>

< 4 exp ix b x 5 >

=

Erfc p i bt

a 2 at

>

> >

>

>

> 2

3>>

>

>

>

> x p p x >

>

> 6 expdx Erfc p d at exp Erfc d at p 7 > >

>

>

> 2 6 2 at 2 at 7 >

>

>

> exp ad t 4 5 >

>

>

> >

>

: ;

2 expbt dx

2:312

where Erfc is the complementary error function. Equation 2.312 is the closed form

solution for temperature distribution. The temperature distribution in non-dimen-

sional form is possible by defining dimensionless quantities and substituting in

Eq. 2.312. The dimensionless quantities are:

56 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

Tkd b

x xd: t ad2 t: T :b 2 2:313

I1 ad

The dimensionless temperature distribution becomes:

1 t

T x ; t

2 b t

8 2 r! 3 9

> p

>

>

> b x >

>

>

> 6 exp ix Erfc p i b 7 >

>

>

> r 6 t

2 t 7 >

>

>

> t 6 7 >

>

>

> i expb

6 7 >

>

>

> b 6 r !

7 >

>

>

> 6 b

x p 7 >

>

< 4 exp ix p

i b 5 =

t 2 t

>

> >

>

>

> 2

3>>

>

>

>

> x p

p x

>

>

>

> 6 expx Erfc p

t expx Erfc t p

7 >

>

>

> expt 6 2 t 2 t 7>

>

>

> 4 5 >

>

>

> >

>

: ;

2 expb x

2:314

To solve for the stress distribution inside the substrate material, equation

governing the momentum in a one-dimensional solid for a linear elastic case can

be considered and Eq. 2.16 can be incorporated in the analysis.

To solve the stress equation (Eq. 2.16), two boundary conditions for the stress

fields should be defined. The first boundary condition is time exponentially

decaying stress at the surface due to recoil pressure developed. It should be noted

that the recoil pressure generated at surface decays almost exponentially with time

and it acts as a stress at the surface as soon as the recoil pressure is generated.

Moreover, as the distance below the surface increases further (extends almost to

infinity), the temperature gradient diminishes; in which case, thermal stress

approaches zero in this region. Therefore, the corresponding boundary conditions

are:

Stress at the surface due to recoil pressure:

At x 0 ) rx kx0 ro ebt eat 2:315

Stress free as depth approaches infinity:

At x 1 ) rx kx1 0 2:316

Initially, the substrate material is assumed to be free from thermal stress and as

the time approaches infinity, heating diminishes, the substrate material also

becomes free from thermal stress. In this case, initial and final conditions for the

stress field are:

Initially stress free substrate material:

At time t 0 ) rx kt0 0 2:317

2.3 Time Exponentially Varying Laser Pulse Heating 57

At time t 1 ) rx kt1 0 2:318

In the case of temperature term in Eq. 2.16, initially substrate material is

assumed at a zero uniform temperature as similar to initial condition for Eq. 2.311.

As time approaches infinity, temperature becomes zero, since heating diminishes.

It should be noted that the laser pulse intensity decays exponentially with time.

Therefore, initial and final conditions for temperature term in Eq. 2.16 are:

Initially:

At t 0 ) Tx; 0 0 2:319

and

Finally:

At t 1 ) Tx; 1 0 2:320

Taking the Laplace Transformation of Eq. 2.16 with respect to time yields:

o2 r

x 1 2 : h : i

2

s r

x x; s srx x; 0 r x x; 0 c 2 s Tx; s sTx; 0 T x; 0

ox2 c21

2:321

where r s are the Laplace Transforms of thermal stress and tem-

x x; s and Tx;

perature respectively in the x and s domains.

By substituting the initial conditions, Eq. 2.16 reduces to:

o2 r

x s2 s

r x x; s c2 s2 Tx; 2:322

ox2 c21

Considering the temperature distribution for time exponentially varying pulse,

Eq. 2.318, and substituting into Eq. 2.322, and solving for the stress field, yields:

o2 r

x s2 2 I1 d d exphx expdx

2rx x; s c2 s 2:323

ox2 c1 ks b hh2 d2 h2 d2

o2 r

x s2 I1 d2 c2 s2

2rx x; s ehx

ox 2 c1 ks bh h2 d2

2:324

I1 dc2 s2

edx

ks b h2 d2

58 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

I1 d2 c2 s2

N1 2:325

ks bh h2 d2

and

I1 dc2 s2

N2 2:326

ks b h2 d2

Equation 2.16 becomes:

o2 r

x s2

r x x; s N1 ehx N2 edx 2:327

ox2 c21

The homogeneous and the particular solutions for 2.328 are:

sx sx

rx h C1 ec1 C2 e c1 2:328

while the particular solution has two parts, the first part is:

rx p1 Q1 ehx

2:329

N1

Q1 2 2:330

h2 cs2

1

rx p2 Q2 edx

2:331

N2

Q2 2 2 2:332

d cs2

1

sx sx

rx g C1 ec1 C2 e c1 Q1 ehx Q2 edx

2:333

Then,

sx N1 hx N2

rx g C2 e c1

2 e 2 s2 edx 2:334

h2 cs2 d c2

1 1

2.3 Time Exponentially Varying Laser Pulse Heating 59

sx

rx g C2 e c1 g1 sehx g2 sedx

2:335

where:

I1 d2 c2 s2

g1 s 2:336

ks bh h2 d2 h2 cs2

2

and

I1 dc2 s2

g2 s 2:337

ks b h2 d2 d2 cs2

2

Now substituting for h s=a and simplifying the expressions for g1(s) and

g2(s) yields:

p

I1 d2 c2 saac21

g1 s c2

2:338

ks bh s ad2 s a1

and

g2 s 2:339

ks b s ad2 s2 c21 d2

Rearranging the two expressions as:

" p #

s

g1 s C3 2:340

s bh s ad2 s c21 =a

and

" #

s2

g2 s C4 2:341

s b s ad2 s c1 ds c1 d

where:

p

I1 d2 c2 aac21

C3 2:342

k

and

I1 c2 c21 da

C4 2:343

k

60 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

2 p p 3

s a2 s

6 ad2 bad2 c2 =as ad2 c2 abc2 a2 d2 s c2 =a 7

6 1 1 1 1 7

g1 s C3 6

6 p 7

7

4 s 5

2

2

b ad b c1 =a s b

2:344

and

2 3

ad2 c1

6 2 7

6 ad b a2 d2 c21 s ad2 2c1 d bc1 ads c1 d 7

g2 s C4 6

6

7

7

4 c1 b2 5

2c1 d bc1 ads c1 d b ad2 b2 c21 d2 s b

2:345

Consider the boundary condition at the surface, where:

Laplace transformation of the boundary condition gives:

1 1

s ro

r 2:347

sb sa

Then, the constant in Eq. 2.334 becomes:

1 1

C2 ro g1 s g2 s 2:348

sb sa

Then:

ps sx

x x; s g1 se

r ax g1 se c1

2:349

sx 1 1 sx

g2 sedx g2 se c1 ro e c1

sb sa

Finding the solution for rx in the x and t domain, we should take the inverse

Laplace Transform for each term in Eq. 2.349. To perform this, the following

designations are introduced:

ps sx

Term1 g1 se ax Term2 g1 se c1

sx

Term3 g2 sedx Term4 g2 se c1 2:350

1 1 sx

Term5 ro e c1

sb sa

2.3 Time Exponentially Varying Laser Pulse Heating 61

Consequently, the solution for stress distribution becomes the summation of the

inverse Laplace Transforms of the above terms.

The details of Laplace Transformation of the terms are given in Appendix 1.

Therefore, the Laplace inversions of Terms (Term1, Term2, Term3, Term4,

Term5) are:

8 2 0

13 9

>

> dx

p x >

>

>

> 6 B e Erfc d at p

2 at C 7>>

>

> 1 6 2 x2 p d2 at B C7 >>

>

>

2 2 2 6

> pe 4at d ae B C7 >>

>

>

> 6 B

C 7 >

>

2 ad b ad c1 =a 4 pt

>

> @ p x A 5 >

>

>

> dx

e Erfc d at p >

>

>

> 2 at >

>

>

> >

>

>

> 2 0 r

1 3 >

>

>

> >

>

>

> c1 x t x >

>

>

> 6 B e a Erfc c

1 p C 7 >

>

< a2 6 2 x 2 c c2 t B a 2 at C 7 =

1 6 1 1

B C 7

Term1C3 2 2 p

e 4at p e a

r

> 2 c ab c a 2 d2 6 pt a B

C7 >

>

> 1 1 4 @ c1 x t x A5 >>

>

> e a Erfc c1 p >

>

>

> a 2 at >

>

>

> >

>

>

>

> 2 0 p

13 >>

>

>

> b p

x >

>

>

>

> 6 B e xi a

Erfc i bt p C7 > >

>

>

> 1 6 2 x p B 2 at C7 >>

> C7 >

2

> 6pe 4at i be B bt >

>

>

> 2 2 =a 6 pt B pb

C7 >>

>

>

> 2 bad bc 1 4 @ p

x A 5 >

>

>

: e xi a

Erfc i bt p >

;

2 at

2:351

and

8 q 9

> p 2 >

>

> p

1 ad e 2 ad tx=c1

Erf 2

ad t x=c1 >

>

>

> p tx=c >

>

>

>

1

>

>

>

> 2 2 >

>

>

> ad c2 =a ad b >

>

>

>

1 >

>

>

> 0 q q

1 >

>

>

> c2

>

>

=

c12 c12

< p a e a

1 1 tx=c1

Erf t x=c 1 x

1

Term2 C3 a2 B B p tx=c a C 1 t

>

1

2 C

>

> @ c1 ab c21 a2 d2 A>>

>

c1

>

> >

>

>

> >

>

>

> >

>

>

> p btx=c1 p >

>

>

> p

1 i be Erf i bt x=c >

>

>

> p

1 >

>

>

>

tx=c 1

>

>

: 2 ;

b c1 =a

2:352

8 2

9

>

a2 d2 ead t

> c1 ec1 dt >

>

>

> >

>

2 < 2

ad ba d c1 2 2 2c d bc ad =

1 dx 1 1

Term3 C4 e

>

> c1 ec1 dt b2 ebt >

>

>

> >

>

: 2c d bc ad 2 2 2 2 ;

1 1 b ad b c1 d

2:353

62 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

2 2 3

a2 d2 ead tx=c1 c1 ec1 dtx=c1

6

6 ad2 b a2 d2 c21 2c1 d bc1 ad 7 7

1 Term4 C4 6 71 t x

6 7 c1

4 c1 ec1 dtx=c1 b2 ebtx=c1 5

2

2 2

2c1 d bc1 ad b ad b c21 d

2:354

and

!

1 b tcx a tcx x

Term5 ro e 1

e 1

1 t 2:355

c1

where 1 t cx1 is a unit step function and Erf (y) is the error function of the

variable y.

The closed form solution of stress distribution can be written as:

1 Term5 2:356

-1

where represents the inverse sign of Laplace Transformation.

Presenting the stress distribution in dimensionless form, the additional

dimensionless quantities are defined, i.e.:

c1

c1 2:357

ad

and

krx

rx 2 2:358

I1 da2 c

2 c1

and

kro

ro 2 2:359

I1 da2 c 2 c1

U [1] is the unit step function, which is U1 t cx

1

Therefore, for the dimensionless stress distribution, the followings are resulted:

2.3 Time Exponentially Varying Laser Pulse Heating 63

8 2 0

13 9

> x

p x >

>

> p

6 x B e erfc t p

C7 >

>

>

> 6

2

B 2 t C7 >

>

>

> 1 6 t e 4t t B C 7 >

>

>

> 6 p B

C 7 >

>

>

> 2

4 p 2 @ p

A 5 >

>

>

> t b 1 c1 x

x >

>

>

> e erfc t p >

>

>

> 2 t >

>

>

> >

>

>

> 2 0

13 >

>

>

> p x >

>

>

> e c

x

erfc c

t p >

>

> 6p x 2

2 B

1

C7 >

< 2 t C7 =

1

1 6 t e 4t c c1 t B

t e

rx 1 2 6 p 1 B C7

6 B

C7

>

>

> c1 1

2

c1 t b 4 p 2 @ p x A5 >

>

>

>

> ec1 x erfc c1 t p >

>

>

> 2 t >

>

>

> >

>

>

> 2 0 p

1 3 >

>

>

> b p x >

>

>

> e ix

t erfc i b p >

>

>

> 6 x 2 p B C 7 >

>

>

> t

3=2 6 e 4t i b e b B 2 t

C 7 >

>

>

> 6

6 p B C 7 >

>

>

> B p

C 7 >

>

>

>

b t b c1 t 4

2 p 2 @ p x A 5 >

>

>

: e ix b

t erfc i

b p >

;

2 t

2:360

and

8 2 3 9

>

> q >

>

>

> 1 6 t 7 >

>

>

>

> 2 4p

qt et x =c1 Erf t x =c1 5 >

>

>

>

t b 1c1 p t x =c >

>

> >

>

>

>

1 >

>

>

> 2 3 >

>

>

> >

>

>

< q

>

=

1 6 1 c 2

t c

x

2 7

rx 2 2 2 4pqc1 e 1 1 c1 t c1 x 5 U1

>

>

c1 1

c1 b =t

p t x =c1 >

>

>

> >

>

>

> >

>

>

> 2 3>>

>

>

> p q

>>

>

>

>

3=2

t 6 t p

b

1x

=c t

7> >

>

>

>

> 4pqi b e 1 erf i b 1x

=c t

5 >

>

>

: 2

1 >

;

t b b c t 1 p t x =c 1

2:361

and

8 9

> t et x t c1 ec1 t x t c1 ec1 t x

>

>

> 2 >

>

>

> b 2 c t b

c 1 2 c t b c 1 > >

>

< 1 c t 1 1 1 1 >

=

1

rx 3

>

> b 2 t eb

x >

>

>

> >

>

>

>

:

2 >

>

;

b t b 2 c1 t

2:362

64 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

and

8 9

t et x =c1 t c1 ec1 t x

>

>

t c1 ec1 t x

>

>

> >

>

>

> 2

>

>

>

>

< 1 c t b 2 c 1 t b c 1 1 2 c 1 t b c 1 1 >

=

1

rx 4 U1

> >

b 2 t eb 1x =c1 t

>

> >

>

> >

>

>

>

>

>

>

: b t b 2 c t 2 ;

1

2:363

and

!

b t x

c a t x

c x

rx 5 ro e 1 e 1 1 t 2:364

c1

rx rx 1 rx 2 rx 3 rx 4 rx 5 2:365

Equations 2.314 and 2.365 are used to compute the dimensionless temperature

and stress distributions inside the substrate material.

thermal system. It also aids to quantify the loss work in the system because of the

heat transfer. Although the formulation of the entropy generation rate for laser

pulse heating is the same for different pulse types, the amount of entropy gener-

ation rate becomes different for different pulses. Consequently, the general for-

mulation of entropy generation rate is given below in line with the previous studies

[68].

Entropy generation in the solid substrate can be written as (2.10):

DS q

r: 2:366

Dt T

or

DS 1 q

r:q 2 rT 2:367

Dt T T

Consider the relation:

h hT; e 2:368

2.4 Entropy Analysis Due to Thermal Stress Field 65

or

oh oh

dh dT de 2:369

ot e oe T

oh

However, the term can be written as [11]:

oe T

oh oCk

q Ck T T T o . . . 2:370

oe T oT To

where Ck is the heat capacity and its value at reference temperature is assumed to

be zero. If T To is the almost the same order of To , then:

oCk

Ck T To 2:371

oT To

It is known that:

oh

Cp 2:372

oT e

Therefore:

oCk

qdh qCp dT To de 2:373

oT To

oCk 3E

aT 2:374

oT To 1 2m

Therefore, Eq. 2.369 can be rearranged as:

dh dT 3E de

q qCp aT To 2:375

dt dt 1 2m dt

It was also shown that [11]:

dh

r:q q 2:376

dt

Therefore Eq. 2.375 becomes:

dT 3E de

r:q qCp aT T o 2:377

dt 1 2m dt

66 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

oT

Inserting Eq. 2.377 into Eq. 2.367 and knowing that q k , it yields:

ox

DS 1 k dT 3E To de

rT 2 qCp aT 2:378

Dt T T dt 1 2m T dt

However, the strain can be written in terms of displacement (U), i.e.: e oU

ox .

Therefore Eq. 2.378 becomes:

DS 1 k dT 3E To d oU

rT 2 qCp aT 2:379

Dt T T dt 1 2m T dt ox

In the non-dimensional form:

DS 1 1 2 dT

3EaT To d oU

rT 2:380

Dt T T dt qCp 1 2m T dt ox

where

S

U Ud and S 2:381

qCp ad2

The first term on the right hand side of Eq. 2.380 represents the entropy gen-

eration due to temperature field while the second term is entropy generation due to

stress field.

Equation 2.381 is used to compute the entropy generation due to temperature

and stress fields.

The lost work can be evaluated using a Gouy-Stodola theorem [10]. In this case,

the lost work can be written as:

Z

Wlost To dS

2:382

where Wlost is the dimensionless lost work and To is the reference temperature.

Equation 2.383 is used to compute the lost work.

In order to formulate the displacement, the relation between the stress field and

the displacement needs to be established. However, the thermal strain along the

x-axis can be written as [11]:

1 m1 2m 1m

ex rx aT DT 2:383

1 mE 1m

where m, E, and aT are Poissons ratio, elastic module, and thermal expansion

coefficient of a solid material, respectively. The equation of motion in a solid

substrate in the one-dimensional form can be written as:

orx oV

q 2:384

ox ot

2.4 Entropy Analysis Due to Thermal Stress Field 67

where V is the velocity, which can be written in terms of displacement (V) as:

oU

V 2:385

ot

The equation of motion becomes:

orx o2 U

q 2 2:386

ox ot

Therefore, Eq. 2.386 gives the relation between the thermal stress and the

displacement, which is needed for the entropy calculations.

The findings of the thermal stress developed in the laser irradiated region is

presented below under the appropriate sub-headings in line with the previous

studies [18].

A closed form solution for the stress field is obtained using the Laplace transform

method for a single and two successive step input pulses. However, the analytical

solution obtained is limited with the free stress condition at the surface, which is

the most common situation in laser machining applications. The stress field in the

solid phase of the substrate material is modeled although the liquid and gas phases

are developed during the laser high intensity irradiation. This is because of the fact

that the thermal strain generated in these phases is insignificant due to the free

molecular activity in these phases. The elastic stress field is considered because of

minimizing the complications associated with the non-linear equation of motion

when the stress level exceeds the elastic limit of the substrate material. The non-

dimensional closed form solution for the stress field can be applicable to any

material, provided that the laser pulse shape is in the step input form. In line with

the previous study [13], the two pulse modes are considered in the analysis,

which are a single step input pulse and two successive step input pulses as shown

in Figs. (2.1) and (2.2) while Table 2.1 gives the laser properties used in the

simulations.

Figures (2.3) and (2.4) show non-dimensional temperature profile inside the

substrate material for three different heating periods for single and two successive

step input pulses. The differences in temperature profiles appear in the surface

68 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

pulse used in the analysis and

1.0E+00

PULSE INTENSITY

in the simulations

8.0E-01

6.0E-01

4.0E-01

2.0E-01

Heating Period Cooling Period

0.0E+00

0.0E+00 5.0E-05 1.0E-04 1.5E-04 2.0E-04

TIME (t*)

region where, the magnitude of temperature is high and the temperature gradient is

small. The small temperature gradient in the surface region results from the energy

gain in the substrate material from the irradiated field. It should be noted that the

amount of the absorbed laser power varies exponentially with the depth inside the

substrate material (Beer Lamberts law), i.e. it is maximum at the surface and

decays exponentially with increasing depth. Consequently, the energy gain from

the irradiated field dominates the diffusional energy transport from the surface

vicinity to the solid bulk. It should be noted that the diffusional energy transport

enhances as the temperature gradient increases. Moreover, as the depth below the

surface increases, the temperature gradient becomes large due to a sharp change in

temperature in this region and smaller amount of absorbed energy from the irra-

diated field with increasing depth. Therefore, diffusional energy transport domi-

nates internal energy gain of the substrate material through absorption in this

region. Since the time domain selected is close to each other, the temperature

variation along the depth is almost similar, particularly at some depth below the

surface. It should be noted that the selection of a close time domain is due to

examination of stress wave propagation in the substrate material, i.e. a long time

domain obscures the appearance of the stress wave within the surface region, since

it has a considerably high speed. In the case of the heating situation for two

successive pulses, some variation in temperature profiles in the surface region

occurs at different heating durations. As the heating duration increases

(t* = 0.003), the influence of the second pulse on the temperature rise is more

pronounced in the surface region. This, in turn, elevates the temperature rise at the

surface.

Figure (2.5) shows non-dimensional stress inside the substrate material for three

heating periods while Fig. (2.6) shows non-dimensional stress distribution in the

region close to the surface. Thermal stress is compressive in the region of the

surface and becomes tensile with increasing depth from the surface. This situation

can be seen from Fig. (2.6). As the depth increases beyond the absorption depth

(x* = 1), it decays sharply. However, the location at which the thermal stress

reaches its first peak varies with the heating period. This is because of the stress

wave propagation inside the substrate material. The depth at which the wave

2.5 Findings and Discussions 69

1.2E+00

t* = t 1* t* = t1* + dt 1* t* = t 2* t* = t 2* + dt 2* t* = t 3*

1.0E+00

PULSE INTENSITY

8.0E-01

6.0E-01 C1 C2

4.0E-01

0.0E+00

0.0E+00 5.0E-05 1.0E-04 1.5E-04 2.0E-04

TIME (t*)

Fig. 2.2 Two successive dimensionless laser step input pulses used in the analysis and in the

simulations (dt1* = dt2*)

properties used in the

simulations Dimensionless pulse length 0.00006 0.00006

(sad2 , s is the pulse length)

Dimensionless intensity 1 1

Io

( , Io laser peak

kd

power intensity)

Dimensionless cooling period 0.00006 0.00006

between pulses (h1 and h2)

temperature profiles inside t* = 0.001

the substrate material t* = 0.0015

4.5E-05

TEMPERATURE (T*)

dimensionless heating

periods for one step input

3.0E-05

pulse

1.5E-05

0.0E+00

0 0.4 0.8 1.2

DISTANCE (x*)

reaches its full amplitude depends on the wave speed and the duration of the wave

that travels. Consequently, longer the heating period results in the deeper the

location at which wave reaches its maximum amplitude.

70 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

temperature profiles inside

the substrate material

obtained at three different 9.0E-05

TEMPERATURE (T*)

dimensionless heating

periods for two successive

pulses 6.0E-05

t* = 0.001

3.0E-05

t* = 0.002

t* = 0.003

0.0E+00

0 0.3 0.6 0.9 1.2

DISTANCE (x*)

thermal stress inside the

substrate material for

different dimensionless 0.5

STRESS (*x)

pulse

0

t* = 0.001

-0.5 t* = 0.0015

t* = 0.002

-1

0 0.4 0.8 1.2

DISTANCE (x*)

thermal stress in the surface

vicinity of the substrate

material for different 0.5

STRESS (*x)

dimensionless heating

periods for single pulse

0

t* = 0.001

-0.5

t* = 0.0015

t* = 0.002

-1

0 0.05 0.1

DISTANCE (x*)

2.5 Findings and Discussions 71

Figure (2.7) shows dimensionless stress levels inside the substrate material at

different heating periods. Stress field inside substrate material is compressive. This

is due to the stress boundary condition introduced at the sur-

face oT =ox jx 0 0 . The stress level is high in the surface vicinity and as the

distance increases it reduces sharply. Moreover, the stress level reduces to zero at

some distance below the surface. The depth, where the zero stress occurs,

increases with progressing time. This indicates that the stress wave propagates into

the substrate material with a wave speed c1.

Figure (2.8) shows the temporal variation of dimensionless stress field at dif-

ferent depths inside the substrate material. The stress level is considerably high at

the surface and as the depth increases, the stress level reduces. Moreover, the

magnitude of stress level inside the substrate material rises after certain period

from the pulse beginning. This indicates that the stress wave propagates into the

substrate material. However, the intensity and shape of the stress wave is modified

by the stress boundary condition at the surface, i.e. stress wave is compressive and

its magnitude increases with progressing time.

locations inside the substrate material. Temperature rises rapidly in the early

heating period and the rate of temperature rise reduces as the heating progresses.

This occurs because of the energy gain and energy transport mechanisms inside the

stress distribution inside the

substrate material for

DIMENSIONLESS STRESS

-20

different heating periods

-40

t = 0.01

t = 0.04

-60 t = 0.08

t = 0.1

-80

-100

0 1 2 3 4 5

DISTANCE (x*)

72 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

of dimensionless stress at

different locations inside the

-20

substrate material

DIMENSIONLESS STRESS

-40

x=0

-60 x=1

x=2

-80 x=3

-100

0 0.03 0.06 0.09 0.12

TIME (t*)

DIMENSIONLESS TEMPERATURE

of dimensionless temperature

at different locations inside

0.0012

the substrate material

0.0008 x* = 0

x* = 0.5

x* = 1

0.0004

x* = 1.5

0

0.0E+00 6.0E-03 1.2E-02 1.8E-02 2.4E-02 3.0E-02

TIME (t*)

substrate material. In this case the internal energy gain by the substrate material

from the irradiated field dominates over the conduction losses in the early heating

period. As the heating period progresses internal energy gain and temperature

gradient in the surface region increase. Consequently, a stage is reached when the

conduction losses from the surface region to the solid bulk becomes important due

to high temperature gradient developed in this region with progressing time. Since

the pulse intensity decreases with time as shown in Fig. (2.10), temperature also

decreases with further progressing heating time. Therefore, temperature curve

decays after reaching its maximum value. The rise of temperature at some depth

below the surface is not as high as that corresponding to the surface. This occurs

because of the energy absorbed by the substrate material at some depth below the

surface, i.e. energy absorbed decreases exponentially with increasing depth

(Lamberts law). As the heating progresses further (t* [ 6 9 10-3) temperature in

the surface region (x* \ 1) reduces while at some depth below the surface

(x* C 1) increases gradually with heating period. This occurs because of the dif-

fusional heating of the substrate material at some depth below the surface due to

high temperature gradient in this region. In the region limited by the absorption

2.5 Findings and Discussions 73

intensity distribution

0.8

0.6

0.4

0.2

0

0 0.01 0.02 0.03 0.04

TIME (t*)

depth (x* \ 1) energy gain by the substrate material reduces considerably with

further progressing time because of time exponentially decaying pulse intensity.

However, high temperature gradient enhances the energy loss from this region to

solid bulk, i.e. conduction losses dominates over the internal energy gain of the

substrate material. This, in turn, results in reducing temperature in this region with

further progressing time.

Figure (2.11) shows temporal variation of dimensionless stress at different

locations inside substrate material. The occurrence of peak stress at different

periods indicates the propagation of the wave inside the substrate material, i.e.

location x* = 0.1, the peak stress occurs at about t* = 0.3 9 10-2 while at

x* = 1.5 it occurs at about t* = 2.4 9 10-2. The magnitude of maximum stress is

high at x* = 0.5. This is because of the development of the maximum temperature

gradient at this location at t 1:25 102 . The stress level decreases with

increasing time and as the time progresses further it decays gradually.

The closed form solution for the stress distribution due to an exponentially

decaying pulse with time and the convective boundary condition at the surface

have been derived. Steel was employed to simulate the temperature and stress

fields. The laser pulse parameter (b) and substrate material properties are given

Table 2.2. Figure (2.12) shows the laser pulse used in the simulations.

Figure (2.13) shows the dimensionless temperature distribution within the

substrate material for various dimensionless heating periods. The influence of the

heat transfer coefficient on the temperature distribution becomes significant when

the dimensionless heat transfer coefficient reaches h 2:02 102

108 W=m2 K . In this case, the temperature and its gradient in the surface region

are reduced. This can also be seen from Fig. (2.14), in which the temperature

74 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

x* = 0.1

variation of dimensionless

DIMENSIONLESS STRESS

x* = 0.5

stress distribution at different

4 x* = 1

locations inside the substrate

x* = 1.5

material

2

-2

0.0E+00 1.0E-02 2.0E-02 3.0E-02 4.0E-02

TIME (t*)

Table 2.2 Material properties and laser pulse parameter used in the simulations

q d a aT m E Cp k b

(kg/m3) (1/m) (m2/s) (1/K) (Pa) (J/kgK) (W/mK) (1/s)

7930 6.16 9 106 3.7 9 10-5 1.6 9 10-5 0.29 2.10 9 1011 510 150 1.53 9 1011

gradient is shown. The temperature gradient in the surface region is reduced to its

minimum. At the point of minimum temperature gradient, the internal energy gain

by the substrate from the irradiated area is balanced by the diffusional energy

transport from the substrate to the solid bulk. In this case, the depth beyond the

point of minimum temperature gradient diffusional energy transport dominates

over the internal energy gain of the substrate material due to absorption of irra-

diated field. The point of minimum temperature gradient changes with the heat

transfer coefficient, which is more pronounced for the heating period of 0.021.

Moreover, the sharp decay in the temperature gradient in the surface region

x 0:1 is because of: (i) the absorption process, i.e. the absorbed energy

decreases exponentially with increasing depth (Beer Lamberts law), and (ii) the

internal energy gain in the surface region is high and diffusional energy transport

due to the temperature gradient from this region to the solid bulk is low, i.e. the

increase in temperature due to diffusional energy transport in the neighboring

region is low; therefore, the temperature profile is governed by the internal energy

gain in this region.

Figure (2.15) shows the dimensionless stress distribution within the substrate

material for different dimensionless heat transfer coefficients and times. The

thermal stress is zero at the surface as a result of the surface boundary condition

used in the analysis and it increases sharply close to the surface. The thermal stress

is tensile in this region due to expansion of the surface. As the depth increases

x [ 0:06, the stress becomes compressive, as a result of the thermal strain

developed in this region, i.e., at this depth and beyond the material contracts

resulting in a compressive thermal stress field. The influence of the heat transfer

coefficient on the stress development is considerable, as illustrated by

2.5 Findings and Discussions 75

in the simulations

NORMALIZED INTENSITY

0.75

0.5

0.25

0

0.0E+00 1.0E-02 2.0E-02 3.0E-02 4.0E-02

TIME

temperature distributions Time = 0.015

TEMPERATURE

h* = 2.02E-4

1.4E-02 h* = 2.02E-3

h* = 2.02E-2

h* = 2.02E-1

1.3E-02

1.2E-02

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

2.0E-02

Time = 0.018

h* = 2.02E-4

TEMPERATURE

1.8E-02 h* = 2.02E-3

h* = 2.02E-2

1.6E-02 h* = 2.02E-1

1.4E-02

1.2E-02

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

2.2E-02

TEMPERATURE

2.0E-02

1.7E-02 h* = 2.02E-4

h* = 2.02E-3 Time = 0.021

1.5E-02 h* = 2.02E-2

h* = 2.02E-1

1.2E-02

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

76 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

temperature gradient within h* = 2.02E-4

the substrate material -4.0E-03 h* = 2.02E-3

Time = 0.015

h* = 2.02E-2

dT/dx

-8.0E-03 h* = 2.02E-1

-1.2E-02

-1.6E-02

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

0.0E+00

h* = 2.02E-4

-4.0E-03 h* = 2.02E-3

Time = 0.018

h* = 2.02E-2

dT/dx

-8.0E-03 h* = 2.02E-1

-1.2E-02

-1.6E-02

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

0.0E+00

h* = 2.02E-4

-4.0E-03

h* = 2.02E-3

Time = 0.021

dT/dx

h* = 2.02E-2

-8.0E-03 h* = 2.02E-1

-1.2E-02

-1.6E-02

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

h 0:0202 109 W/m2 K . In this case, the stress developed is compressive and

with a high magnitude in the vicinity of the surface and decays sharply as the depth

increases. However, the compressive stress wave is developed at some point below

the surface. The magnitude of the stress wave is lower at this point as time

progresses. In addition, the magnitude of the thermal stress levels, corresponding

to a heat transfer coefficient other than h 0:0202, increases with an increase in

time, provided that this increase is less than 10 %.

Figure (2.16) shows the change in thermal stress for different dimensionless

heat transfer coefficients with time. Since the substrate material is considered

initially to be stress free, i.e. the magnitude of the stress level is zero at time t = 0.

The stress level is tensile in the in the vicinity of the surface x 0:001. This is

because of the free expansion of the surface, in which case the strain becomes

2.5 Findings and Discussions 77

h = 2.02E-4

stress distribution within the

1.0E-01 h = 2.02E-3

substrate material Time = 0.015

h = 2.02E-2

STRESS

0.0E+00 h = 2.02E-1

-1.0E-01

-2.0E-01

-3.0E-01

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

2.0E-01

h* = 2.02E-4

1.0E-01 h* = 2.02E-3

h* = 2.02E-2 Time = 0.018

STRESS

0.0E+00 h* = 2.02E-1

-1.0E-01

-2.0E-01

-3.0E-01

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

2.0E-01

h* = 2.02E-4

1.0E-01 h* = 2.02E-3

Time = 0.021

h* = 2.02E-2

STRESS

0.0E+00

h* = 2.02E-1

-1.0E-01

-2.0E-01

-3.0E-01

0.0 0.1 0.2 0.3 0.4 0.5

DISTANCE

positive. However, the thermal stress becomes compressive as the depth increases

beyond x 0:01. This occurs as a result of the compression of the substrate

material. The effect of the heat transfer coefficient on the stress distribution

behavior with time is significant for a heat transfer coefficient of h 0:0202. In

this case, the temperature gradient in the surface region differs significantly from

those corresponding to the other heat transfer coefficients. This, in turn, generates a

compressive stress wave propagating into the material. The magnitude of the stress

wave is reduced as it propagates into the substrate material. Moreover, the stress

wave behavior disappears at a depth of 0.1, i.e. the stress behavior beyond this

depth changes significantly. This shows that the stress wave does not attain a depth

of x 0:1 during the time interval considered in the present study.

78 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

dimensionless stress 0.0E+00

distribution with time

STRESS

-2.0E-01 h* = 2.02E-4

h* = 2.02E-3

-4.0E-01 x = 0.001

h* = 2.02E-2

-6.0E-01 h* = 2.02E-1

-8.0E-01

0.0E+00 5.0E-03 1.0E-02 1.5E-02 2.0E-02 2.5E-02

TIME

2.0E-01

0.0E+00

STRESS

-2.0E-01 h* = 2.02E-4

h* = 2.02E-3

-4.0E-01

x = 0.01 h* = 2.02E-2

-6.0E-01

h* = 2.02E-1

-8.0E-01

0.0E+00 5.0E-03 1.0E-02 1.5E-02 2.0E-02 2.5E-02

TIME

2.0E-01

0.0E+00

STRESS

-2.0E-01 h* = 2.02E-4

-4.0E-01 h* = 2.02E-3

x = 0.03 h* = 2.02E-2

-6.0E-01

h* = 2.02E-1

-8.0E-01

0.0E+00 5.0E-03 1.0E-02 1.5E-02 2.0E-02 2.5E-02

TIME

2.0E-01

0.0E+00

STRESS

-2.0E-01 h* = 2.02E-4

h* = 2.02E-3

-4.0E-01

h* = 2.02E-2 x = 0.1

-6.0E-01

h* = 2.02E-1

-8.0E-01

0.0E+00 5.0E-03 1.0E-02 1.5E-02 2.0E-02 2.5E-02

TIME

material at different heating periods. The decay of temperature in the surface

region is slower as compared to some depth below the surface. This occurs

because of the insulated boundary condition at the surface as well as the internal

energy gain from the irradiated field in this region. In this case, the amount of

energy absorbed from the irradiated field in the surface region is higher than that

2.5 Findings and Discussions 79

temperature distributions

inside the substrate material t* = 0.01

TEMPERATURE (T*)

for different time 1.2E-03 t* = 0.02

t* = 0.03

8.0E-04

4.0E-04

0.0E+00

0 1 2 3 4

DISTANCE (x*)

corresponding to some depth below the surface due to the Lamberts law of

absorption, i.e., energy absorbed by the substrate reduces exponentially with

distance away from the surface. Consequently, energy absorption enhances the

internal energy gain in the surface region, which dominates over the conduction

energy transfer from this region to the solid bulk. However, as the depth below the

surface increases further, amount of energy absorbed reduces and conduction

losses dominates the internal energy gain of the substrate material. Hence, the

temperature gradient changes considerably in this region.

Figure (2.18) shows stress levels inside the substrate material at different

periods. It is evident that the stress wave propagates into the substrate material

with a wave speed. The magnitude of peak stress reduces as the depth below the

surface increases towards the solid bulk. This indicates that material dumps

gradually the magnitude of stress level as the wave propagates towards the solid

bulk. Moreover, the time occurrence of peak stress inside the substrate material

differs for the stress free condition and stress condition at the substrate surface.

This is because of the temporal variation of the stress distribution at the surface

and the temporal variation of laser pulse intensity distribution, which differ, i.e.

pulse intensity decays with a progressing time while stress distribution at the

substrate increases to reach its peak, then decreases with increasing time as shown

in Fig. (2.18).

different heating periods. Entropy generation is low in the surface region and it

increases steadily with increasing depth from the surface. The increase in entropy

80 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

stress distribution inside the t* = 0.01

substrate material at different 10

t* = 0.02

times for stress free and stress

t* = 0.03

conditions at the surface 8

STRESS

6

No, Stress Sur.

4

-2

0 0.6 1.2 1.8 2.4 3

DISTANCE (x*)

entropy generation profiles Time = 0.4

due to temperature field Time = 0.8

inside the substrate material 3 Time = 1.2

at different heating periods

ENTROPY (S*)

Time = 1.6

0

0 0.5 1 1.5 2 2.5

DISTANCE (x*)

in magnitude in the region next to the surface vicinity (Fig. (2.20)). However,

temperature is less in this region next to the surface vicinity. Consequently,

increase in entropy due to term rT 2 is enhanced by the term (T12 ), i.e. T12 rT 2

increases with depth. Moreover, in the region further away from the surface

rT 2 does not attain high values as much as in the surface region, but term (T12 )

increases considerably due to low temperature in this region. Similarly, the term

(T1 dTdt ) increases with increasing depth, since (T12 ) attains higher values as compared

to term (T1 dTdt ). Therefore, entropy generation due to temperature field increases

with depth. Moreover, entropy generation in the early heating period is high. In

this case, rapid rise of temperature in the surface region results in high rate of

entropy generation in the early heating period.

2.5 Findings and Discussions 81

temperature gradient inside

the substrate material at

-0.06

different dimensionless

heating periods

Time = 0.4

-0.12

dT*/dx*

Time = 0.8

Time = 1.2

-0.18 Time = 1.6

-0.24

-0.3

0 1 2 3 4 5

DISTANCE (x*)

different heating periods. Entropy generation in the surface region is high and as

the distance increases from the surface it reduces, particularly sharply for the long

heating period (t 1:6). Attainment of high entropy generation in the surface

region is because of high magnitude of surface displacement in this region.

Entropy contribution of displacement is presented by the term dtd oU ox in entropy

equation. High entropy generation in the surface region is high values of the term

d oU

dt ox in this region. When comparing the magnitude of entropy generation due

to stress field with its counterpart corresponding to temperature field, it is evident

that entropy generation due to temperature field is considerably high, i.e. the ratio

of entropy is in the order of 10-5. This is because of the small displacement of the

surface during the heating process.

entropy generation profiles Time = 0.4

Time = 1.2

substrate material at different

Time = 1.6

ENTROPY (S*)

0.0E+00

-5.0E-06

0 0.5 1 1.5 2 2.5

DISTANCE (x*)

82 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

Figure (2.22) shows the dimensionless volumetric entropy generation inside the

substrate material due to temperature field. Entropy generation is high in the

surface region and reduces sharply as the depth increases from the surface. The

sharp decay of the entropy generation in the region next to the surface vicinity is

because of the temperature gradient, which varies significantly in this region.

Moreover, entropy generation

due to temperature field is influenced considerably

by the energy diffusion Tk rT 2 as well as by the energy storage oT ot .

Consequently, entropy profiles do not exactly follow the square of temperature

gradient, i.e. the influence of energy storage on entropy generation is none zero. It

was shown that at some depth below the surface energy balance attains among the

energy absorbed from the irradiated field, internal energy gain of the substrate

material and the diffusional energy transport from the surface region to the solid

bulk. Consequently, entropy generation reduces to minimum in the region close to

the location of the equilibrium point, i.e. an equilibrium state is reached. As the

distance increases from the point of the equilibrium state, entropy generation

increases sharply. In this case, the diffusional energy transport dominates the

internal energy gain due to the energy absorbed from the irradiated field. Conse-

quently, the decay of temperature gradient changes in this region. As the depth

increases further towards to solid bulk, entropy generation attains a steady value,

i.e. a steady state is reached. As the heating period progresses, the magnitude of

entropy generation reduces; in which case, the laser power intensity reduces with

progressing time. Moreover, the location of minimum entropy generation moves

away from the surface as heating period progresses.

Figure (2.23) shows the dimensionless volumetric entropy generation inside the

substrate material due to stress field at different heating periods. Although the

entropy generation does not follow the stress distribution (Fig. 2.24), the location

of peak entropy generation coincides with the location of the peak stress inside the

substrate material. The negative sign of the entropy generation rate at some depth

below the surface is because of the negative magnitude of stress field (compressive

DIMENSIONLESS ENTROPY

material due to temperature 4.0 t = 0.01

GENERATION

periods 3.0 t = 0.03

2.0

1.0

0.0

0.0 0.5 1.0 1.5 2.0

DIMENSIONLESS DISTANCE

2.5 Findings and Discussions 83

DIMENSIONLESS ENTROPY

entropy generation inside the Dimensionless Time

material due to stress field at t = 0.01

0.3 t = 0.02

different heating periods

GENERATION

t = 0.03

0.0

-0.3

-0.6

0.0 0.5 1.0 1.5 2.0

DIMENSIONLESS DISTANCE

Dimensionless Time

thermal stress inside the 4.0

DIMENSIONLESS STRESS

t = 0.01

material at different heating t = 0.02

periods t = 0.03

2.0

0.0

-2.0

0.0 0.5 1.0 1.5 2.0

DIMENSIONLESS DISTANCE

stress field), which is due to the propagation of the stress wave and, in all cases, the

entropy generation is positive. It should be noted that the negative sign in the

entropy generation curve is the indication of direction of the stress field and, in all

cases, the entropy generation is positive. The entropy generation inside the sub-

strate material appears like a cyclic with time. This is because of the stress

behavior, which occurs in the wave form. It should be noted that since the sub-

strate material is assumed to be stress free at initial state and elastic in nature, once

the thermal field diminishes it returns the initial state. Consequently, the final net

balance of entropy generation should be zero. This situation can be seen from Fig.

(2.23), i.e. the final net balance of entropy generation is zero.

Figure (2.25) shows the dimensionless total entropy generation inside the

substrate material, due to temperature and stress fields, at different heating periods.

Entropy profile in the surface region follows the entropy profile corresponding to

the temperature field. This indicates that entropy generation in this region is

dominated by the temperature field and the entropy generation due to stress field is

negligibly small. As the depth increases further towards the solid bulk, entropy

generation due to stress fields become important. In this case, the magnitude of

84 2 Equilibrium Laser Pulse Heating and Thermal Stress Analysis

total entropy generation 5.0

Dimensionless Time

inside the material at different

4.0 t = 0.01

heating periods t = 0.02

GENERATION

3.0 t = 0.03

2.0

1.0

0.0

0.0 0.5 1.0 1.5 2.0

DIMENSIONLESS DISTANCE

entropy generation due to temperature field becomes less than the peak values of

the entropy generation due to stress field. This, in turn, results in cyclic appearance

on the entropy curve in the region at some depth below the surface.

References

1. M. Kalyon, B.S. Yilbas, Analytical solution for thermal stresses during laser pulse heating

process. Proc. Inst. Mech. Eng. Part C, J. Mech. Eng. Sci. 215, 14291445 (2001)

2. B.S. Yilbas, M. Kalyon, Repetitive laser pulse heating with a convective boundary condition

at the surface. J Phys D Appl Phys 34, 222231 (2001)

3. B.S. Yilbas, N. Ageeli, Thermal stress development due to laser step input pulse intensity

heating. J Therm Stresses 29(8), 721751 (2006)

4. B.S. Yilbas, N. Ageeli, M. Kalyon, Laser induced thermal stresses in solids: exponentially

time decaying pulse case. Lasers Eng 14(1), 81101 (2004)

5. B.S. Yilbas, N. Al-Ageeli, Formulation of laser induced thermal stresses: stress boundary at

the surface. Proc. Inst. Mech. Eng. Part C: J. Mech. Eng. Sci. 217, 423434 (2003)

6. B.S. Yilbas, N. Ageeli, Thermal stresses due to exponentially decaying laser pulse and a

convection boundary at the surface. Lasers Eng 16, 235265 (2006)

7. B.S. Yilbas, M. Kalyon, Analytical approach for entropy generation during a laser pulse

heating process. AIChE J 52, 19411950 (2006)

8. B.S. Yilbas, Entropy analysis due to temperature and stress fields in the solid irradiated by a

time exponentially varying laser pulse. Heat Transfer Eng J 26(8), 8089 (2005)

9. A.D. Kovalenko, Thermoelaticity: basic theory and applications. (Wolters-Noordhoff

Publishing, Groningen, 1969), p. 190

10. A. Bejan, Entropy generation minimization (CRC press, New York, 1995)

11. D.Y. Tzou, Macro-to-microscale heat transfer. (Taylor and Francis, Washington, 1997),

pp. 231234

Chapter 3

Analytical Solution of Cattaneo

and Thermal Stress Equations

energy transport in the irradiated region. In this case, thermal separation of elec-

tron and lattice sub-systems takes place. The thermal communication of these sub-

systems occurs through the collisional process and the electrons transfer some of

their excess energy during this process. Although electron temperature attains

significantly high values due to the energy gain from the irradiated field through

absorption, lattice site temperature remains low. Since the heated region is limited

within a small volume, temperature gradients remain high across the irradiated

region despite the attainment of low temperature field. Consequently, high tem-

perature gradients cause the development of high thermal stress field in the small

region. This limits the practical applications of the laser treatment process at

microscopic scales. In this chapter, heat transfer at micro-scale is formulated and

temperature field is presented analytically. The closed for solutions for the

temperature and stress fields are obtained for various heating situations.

3.1 Introduction

Short-pulse laser heating of solid surfaces causes the heat wave generation in the

irradiated region and the energy transport to the solid bulk takes place through the

heat wave at a finite speed. In this case, heat is conducted in the solid due to the near

neighborhood excitation via changing of momentum and energy on a microscopic

scale in a wave form. The average communication time between the neighborhoods

is associated with the phonon relaxation time. The thermal communication takes

place in a dissipative nature resulting in the thermal resistance in the solid medium.

If the heating duration is longer than the relaxation time, the speed of the heat wave

propagation approaches to infinity and the heat wave equation reduces to the

classical Fourier heat equation. However, the Fourier heating model fails to predict

the temperature propagation speed and suffers anomalies when the heating duration

and Thermal Stress Analysis, Materials Forming, Machining and Tribology,

DOI: 10.1007/978-3-319-00086-2_3, Springer International Publishing Switzerland 2014

86 3 Analytical Solution of Cattaneo and Thermal Stress Equations

is comparable or less than the phonon relaxation time. Therefore, the heat equation

incorporating the wave nature of the energy transport in solids becomes unavoid-

able to formulate correct temperature filed in the heated region. Moreover, the

heating of sub-micron sized solid devices during the short heating period results in

non-equilibrium heating situation in the solid; in which case, the governing equa-

tion takes the hyperbolic form. The heating situation can be modeled using the

Cattaneos equation, since Cattaneos heating model is governed by the hyperbolic

temperature equations. Hence, Cattaneos heating model becomes appropriate to

account for the temperature propagation speed while eliminating the anomalies

arise from the Fourier heating model.

Thermal energy deposited into the substrate material causes internal energy

gain of the substrate material and thermal expansion because of the presence of

temperature gradients in the heated region. The work done during thermal

expansion of the substrate material can be associated with the energy loss in terms

of the internal energy gain. This is because of the fact that not all the deposited

energy is transformed into the internal energy gain of the substrate material, but

some fraction of the deposited energy is dissipated through the mechanical work

done during the thermal expansion. In reality, the temperature field developed in

the solid is responsible for the thermal expansion of the substrate. This, in turn,

generates the thermal stress field within the heated region. Since the heat equation

associated with the energy deposition, diffusion, and internal energy gain is in a

hyperbolic form for the short heating durations, investigation into the solution of

the hyperbolic heat and the thermal stress equations becomes necessary to asses

the thermal and mechanical responses of the substrate material. Although the

numerical solution is possible for such heating situation, the analytical solution of

Cattaneos heating equation is fruitful to generalize the temperature field in terms

of the heat source parameters and the material properties. The application of the

Laplace transformation method provides the exact solution to the Cattaneos

heating equation.

In the current chapter, the model study in relation to the short-pulse laser heating

of surfaces is presented and the analytical solutions for the heat equation, in the

form of Cattaneos equation, and thermal stress equation are provided for surface

and volumetric heat sources in line with the previous studies [15]. The influence of

laser pulse shape, namely step input and the exponential pulses, on temperature and

stress fields is also considered in the analytical treatment of the problem.

In order to simplify the laser heating process, energy deposition from the irradiated

source can be considered as a surface heat source. The incident irradiated energy is

absorbed within the surface skin of the substrate material; however, consideration

of small absorption depth of the substrate material for the incident radiation can

simplify the necessity for the volumetric heat source consideration to the surface

3.2 Surface Heat Source Consideration 87

complications associated with the solution of the temperature and stress fields.

Two pulses, including the step input and exponential pulses, are considered in the

below analysis.

The model presented considers the Cattaneo heating in line with the previous study

[1]. The Cattaneo equation can be written as:

oq

q KrT 3:1

ot

where q krT, e is the relaxation time, K is the conductivity tensor, q is the

heat flux vector, and T is the temperature. The Cattaneo equation yields to the

hyperbolic heat conduction equation for the temperature field, which is propaga-

q

tive with a speed c qCkp e, where q is the density and Cp is the specific heat

capacity.

o2 T oT o2 T

K 2 qCp s 2 3:2

ox ot ot

The heat equation can be nondimensionalized through introducing the fol-

lowing dimensionless parameters:

t x T j

t x p T p 3:3

2e 4ej=qCp P 4ej=qCp

where the P is the pulse intensity.

The resulting dimensionless equation is thus:

o2 T oT o2 T

2 2 3:4

ox2 ot ot

with dimensionless initial and boundary conditions:

T0 at t0 3:5

q f t at x0 3:6

The closed form solution of Eq. 3.4 can be possible by adapting the Laplace

transformation method. Taking the Laplace transformation of the governing

Eq. 3.4, initial and boundary conditions (Eqs. 3.5 and 3.6) results in:

b

o2 T b 2s T

s2 T b 0 3:7

ox2

88 3 Analytical Solution of Cattaneo and Thermal Stress Equations

b

oT

b

q b Fs

sT at x0 3:8

ox

The last equation can be rewritten as:

oT 1

Fss 1 at x 0; 3:9

ox s

The step input pulse intensity can be written as:

f t Put s ut 3:10

Since the model is linear, one can use the superposition principle to compute

the temperature distribution due to the two terms of the pulse function separately.

It is convenient to get the solution due to the second term of the f(t) first and then

use the mathematical formula:

where L1 is the inverse Laplace Transform. Solution of the governing equation in

the s-domain (Laplace domain) can be represented as:

p p

2 2

Tx; s C1 ex s 2s C2 ex s 2s 3:12

For the solution to be bounded C2 must vanish and therefore,

p

2

Tx; s C1 ex s 2s 3:13

By applying the transformed boundary condition, the constant C1 can be

determined and the solution in the transformed domain can be finally written as:

p

2

1 2s ex s 2s

Tx; s p 3:14

ss 2

The first term of the solution can be inversely transformed as,

!

1 p hpi

1 x ss2

L p e et Io t2 x2 ut x 3:15

ss 2

where I0 is the modified Bessel function of the first kind and u is the unit step

function.

The second term can be inversed using the rule:

Z t

1

L1 Fs f k dk 3:16

s 0

Therefore, the final solution in the time domain can be expressed as:

3.2 Surface Heat Source Consideration 89

hpi Z t hpi

Tx; t ut x et Io t2 x2 2 ek Io k2 x2 dk 3:17

0

To solve for the stress distribution inside the substrate material, equation

governing the momentum in a one-dimensional solid for a linear elastic case can

be considered [6], i.e.:

o2 r 1 o2 r o2 T

2

2 2 c2 2 3:18

ox c1 ot ot

The preceding equation is rewritten using the following dimensionless quan-

tities along with the previously described ones (Eq. 3.3):

r

r eqCp 2c2 Pj

r c1 c1 c2 p 3:19

E j Cp Eq2 ej=Cp q

2

Therefore,

o2 r 1 o2 r o2 T

2

2 2 c2 2 3:20

ox c1 ot ot

By applying the Laplace transformation, the dimensionless stress field equation

in the s-domain is given by:

s2 r

^x p

^00 x

r M 2 e ss2x

3:21

c21

where M2 cp2 ss2

s2 2s

and c1 and c2 are the dimensionless wave speed and the

dimensionless thermal modulus, respectively. Therefore, the solution of stress

equation in the Laplace domain becomes:

p

2 ss2x

sx

sx c M 2 e

^ B1 ec1 B2 e c1 1 2

r 3:22

s c1 s 2 s

For boundness of the solution B1 0, and the condition of free stress (i.e.

r 0) on the surface is imposed as this stage to calculate the other coefficient B2

as:

c2 M2

B2 2 1 2 3:23

s c1 s 2c1 s

^x; s r

r ^h x; s r

^p x; s 3:24

where

sx

c21 c2 s 2e c1

^h p

r

s2 2s c21 s 2c21 s

90 3 Analytical Solution of Cattaneo and Thermal Stress Equations

p

c21 c2 s 2e ss2x

^p p 2

r 3:25

s2 2s c1 s 2 s

The stress component r

^h can be written as a multiplication of two subcom-

ponents as:

1 2

^h x; s r

r ^h x; s^

rh x; s 3:26

where

1 c2 c2 s 2 sx

r

^h 21 ec1

c1 s 2c21 s

and

2 1

r

^h p 3:27

2

s 2s

The inverse Laplace transform of the two subcomponents is carried out as the

following:

1 1

rh x; t L1 r^h

2c1 xc1 t

!

2 1 x 2 c2 1

x

c1 c2 2 d t 2 e 1 U t 3:28

c1 1 c1 c2 1 c1

1

and

2 2

rh x; t L1 r^h et J0 it 3:29

^h are:

1 1 1 1 c21 c2 s 2 csx

rh x; t L ^h L

r 2 e 1

c1 s 2 s

!

2 2

2c c 2 sx c c 2 sx

L1 2 1 e c1 2 1 e c1

c1 1 c21 s 2c21 s c1 1

! 2

1 2c21 c2 csx 1 c1 c2 csx

L e 1 L e 1

c21 1 c21 s 2c21 s c21 1

2c1 xc1 t

!

c21 c2 x 2c21 c2 c2 1

x

2 d t 2 e 1 U t

c1 1 c1 c21 1 c1

2c1 xc1 t

!

2 1 x 2 c2 1

x

c1 c2 2 d t 2 e 1 U t

c1 1 c1 c2 1 c1

1

3.2 Surface Heat Source Consideration 91

and

!

1

2 1 2 1

rh x; t L r

^h L p et J0 it 3:30

ss 2

1 2

Having obtained rh x; t and rh x; t one can get rh x; t by employing the

convolution theorem:

Z t

1 2

rh x; t ^h s^

r rh t s ds 3:31

0

^p can be written can be written

as a multiplication of two subcomponents:

r ^1

^p x; s r r2

p x; s^ p x; s 3:32

where

c21 c2 s 2

^1

rp

c21 s 2 s

and

p

e ss2x

^2

r p p 3:33

s2 2s

The inverse Laplace transform is carried out for both subcomponents:

0 2c2 t 1

21

c 1

B dt 2e 1 C

r1

p x; t L

1

^1

r p c21 c2 @ 2 2 A 3:34

c1 1 c2 1 1

p

r2

p x; t L 1

r

^ 2

p e t

I 0 t2 x2 Ut x 3:35

Having obtained rp1 x; t and rp2 x; t one can get rp x; t by employing the

convolution theorem:

Z t

rp x; t ^1

r p s^r2

p t s ds 3:36

0

92 3 Analytical Solution of Cattaneo and Thermal Stress Equations

2

c1 c2 s 2

r1

p x; t L1

r

^ 1

p L1

c21 s 2 s

!

1c21 c2 2c21 c2

L 2

c21 1 c1 1 c21 s 2c21 s

2 !

1 c1 c2 1 2c21 c2

L L

c21 1 c21 1 c21 s 2c21 s

2 2c2 t

!

c1 c2 2c21 c2 21

c 1

2 dt 2 e 1

c1 1 c21 1

and

p !

e ss2x hpi

r2

p x; t

1

L ^2

r p L 1

p et Ut xI0 t2 x2 3:37

ss 2

All components of the stresses have been transformed to (x, t) domain, and we

are ready to get the stresses as,

rx; t rh x; t rp x; t 3:38

Equations 3.17 and 3.38 can be used to compute the temperature and thermal

stress distributions inside the substrate material.

Laser heating pulse temporal behavior can be formulated through the time varying

exponential function. In this case, the formulation of the heat equation and the

relevant boundary conditions become identical to those presented in Sect. 2.2.1.

The analysis presented below is line with the previous study [2]. However, the

surface heat source can be formulated through the exponential pulse as shown in

Fig. 3.1 and it takes the form:

Since the model is linear, one can use the superposition principle to compute

the temperature distribution due to the two terms of the pulse function separately

and add them to get the final solution. In the sequel, one term namely, eb1 t is

considered as a pulse source and then the total solution is simply:

T Tjeb1 t Tjeb2 t 3:40

3.2 Surface Heat Source Consideration 93

of the exponential pulse

The solution of the governing heat equation (Eq. 3.9) in the s-domain can be

represented as:

p p

b x; s C1 epa

x px

es2 s es2 s

T C2 e a 3:41

For the solution to be bounded C2 must vanish.

By applying the transformed boundary condition, the constant C1 can be

determined and the solution in the transformed domain can be finally written as:

p p

px ss1=e

ea

b x; s

T q e a=e 3:42

Ks b ss 1=e

as:

b x; s T^1 x; sT^2 x; s

T 3:43

The inverse of the two terms can individually be written as:

p !

1 ^

1 1 px ss1=e

L T1 x; s L p e a=e

ss 1=e

s! " #

t

2e 1 x2 x

e Io 2

t U t p 3:44

2e a=e a=e

and

p p

ae ae bt

L T^2 x; s L1

1

e 3:45

Ks b K

where Io is the modified Bessel function of the first kind, U is the unit step

function.

94 3 Analytical Solution of Cattaneo and Thermal Stress Equations

Z t

Tx; t T^1 sT^2 t sds 3:46

0

rate among the constitutive variables [7]. However, most of the materials, the

strain rate is of the same order of the temperature rate; in which case, the thermo-

mechanical coupling becomes small for the small displacements and it can be

neglected [7, 8]. This simplifies the physical problem to yield the closed form

solution for the thermal stress field in the solid substrate. Consequently, after

assuming the homogenous isotropic structure, the temperature distribution of the

semi-infinite media in the s-domain can be represented in a dimensionless form:

p

s2 2sx

^ s s 2e

Tx; p 3:47

s b s2 2s

The corresponding dimensionless stress field equation in the s-domain is given

by:

s2 b

r x c2 s2 s 2 p

ss2x

b

r x p e 3:48

c21 s b s2 2s

c1 and c2 are the dimensionless wave speed and the dimensionless thermal

modulus, respectively.

p

2 2 ss2x

sx

sx c 2 s s 2 c 1 e

b

r B1 ec1 B2 e c1 p 3:49

s b s2 2s sc21 s 2 s

For boundness of the solution B2 0, and the condition of free stress (i.e.

b

r 0) on the surface is imposed as this stage to calculate the other coefficient B2

as:

c21 c2 s2 s 2

B2 p 3:50

sc21 s 2c21 s s b s2 2s

Therefore,

p

sx

c21 c2 ss 2e c1 c21 c2 ss 2e ss2s

^x; s p 2

r p 2

s2 2s c1 s 2c21 s s b s2 2s c1 s 2 s s b

3:51

where r^x; s r

^h s; x r

^p x; s. The stress component r ^h can be written as a

1 2

multiplication of two subcomponents as r ^h x; s r

^h s; x^

rh x; s. The inverse

Laplace transform of the two subcomponents is carried out as the following:

3.2 Surface Heat Source Consideration 95

1 1

rh L1 r

^h

!

2c1 xc1 t

c21 c2 xb

tb 4c2 c2 1 x c2 c2 x

2

bb 2ec1 2 1 2 e 1 U t 21 d t

c1 b 2 b c1 1 c1 c1 1 c1

3:52

2 2

rh x; t L1 r

^h et J0 it 3:53

Z 1 2

t

rh x; t 0

^h s^

r rh t sds 3:54

^p can be written can be written

as a multiplication of two subcomponents:

r ^1

^p x; s r r2

p x; s^ p x; s 3:55

r1 1 1

p x; t L ^

rp

2c2

!

1 2b b2 bt 4c21 1

c2 1

c21 c2 d t e e 1

c21 1 c21 b 2c21 b c21 12 c21 b 2c21 b

3:56

p

r2

p x; t L1 2

^

r p e t

I 0 t 2 x2 U t x 3:57

One can get r

Z t

rp x; t ^1

r r2

p s^ p t sds 3:58

0

All components of the stresses have been transformed to (x, t) domain, and we

are ready to get the stresses as,

rx; t rh x; t rp x; t 3:59

The temperature and stress distributions inside the substrate material can be

computed from Eqs. (3.46) and (3.59), respectively.

The incident irradiated energy is absorbed in the skin of the substrate material

surface according with the Beer Lamberts law. In the case, a volumetric heat

source should be incorporated in the Cattaneo equation to account for the

96 3 Analytical Solution of Cattaneo and Thermal Stress Equations

and thermal stress are presented in line with the previous studies [3, 4].

In line with the previous study [4], the Cattaneo heat equation incorporating the

volumetric heat source can be written as:

o2 T oT o2 T

a F 0 f tgx e 3:60

ox2 ot ot2

k

where Fo f tgx is the source term and a is the thermal diffusivity a qCe .

Fo

Moreover, F0 Fo qCe is the normalized laser peak intensity Fo is the laser

peak intensity, Ce is the specific heat, f(t) is temporal function representing the

laser pulse intensity variation, g(x) is the absorption term, and e is the relaxation

time. The initial and boundary conditions are:

oT

T0 and 0 at t0 3:61

ot

oT

0 at x0 3:62

ox

The step input pulse intensity is considered as shown in Fig. 3.2, which can be

written as:

f t U x s U x 3:63

of laser step input pulse

3.3 Volumetric Source Consideration 97

where U x is the unit step function and s is the width of the step function. The

heating due to absorption of the laser beam is assumed to be exponential along the

depth of the substrate to resemble the actual heating process:

gx edx 3:64

Since the model is linear, one can use the superposition principle to compute

the temperature distribution due to the two terms of the temporal part of the pulse

(i.e. f t) separately. It is convenient to get the solution due to the second term of

the first and then use the mathematical formula:

Taking the Laplace transformation of the governing Eqs. 3.60 and the boundary

conditions (Eq. 3.62) results in:

o2 T^ F0

a sse 1T^ edx 3:66

ox2 s

and

oT^

0 at x0 3:67

ox

The solution of the governing equation in the s-domain can be represented as:

p p

^ x ps aes

2

p

x s2 es F0 exd

Tx; s C1 e C2 e a 3:68

ses2 s ad2

For the solution to be bounded, C2 must vanish and therefore,

p

^ s C1 e pa

x s2 es F0 exd

Tx; 3:69

ses2 s ad2

By applying the transformed boundary condition, the constant C1 can be

determined as:

p

F0 d a

C1 p

s ses 1es2 s ad2

Therefore, the temperature distribution in the s-domain can be obtained as:

p

xd

p x ps2es

F 0 e F 0 a de a

T^ 2

p 3:70

ses s ad s ses 1es s ad2

2 2

98 3 Analytical Solution of Cattaneo and Thermal Stress Equations

T^p F0 exd ^

hs 3:72

and

p

^

T^h F0 ad /s 3:73

where

^ 1

hs 3:74

s s e s ad2

2

and

^

/s ^ ^

hsws 3:75

where

p

x ps aes

2

^ e

ws p 3:76

sse 1

To invert the particular part T^p into the time domain, partial fraction and

standard inverse Laplace tables are employed to invert hs and for sake of brevity,

i.e.:

h tpc p p

p

1 p t c1 p t ct

ht p 2 e e c 12 ce 2e c 1 e 2e 3:77

2 c ad

Therefore, the particular solution of the temperature in the space-time domain

is:

p

F0 edx h tpc p

p

p t c1 p t ct

Tp x; t p 2 e e c 12 ce 2e c 1 e 2e 3:78

2 c ad

Similarly, the Laplace inversion of the function wt can be written as:

" s# " #

1 x 2 x

2et

wt e Io t2 U t p 3:79

2e a=e a=e

where I0 is the modified Bessel function of the first kind and U is the unit step

function

Having obtained the inversion of ^ ^

hs and ws, one can use the convolution

^

technique along with Eqs. 3.773.79 to calculate the inverse Laplace of /s as:

Zt

/t hswt sds 3:80

0

3.3 Volumetric Source Consideration 99

Th x; t F0 edx /t 3:81

By combining the homogeneous and the particular solutions of the overall

temperature distribution is obtained as:

T x; t Tp x; t Th x; t 3:82

The stress distribution inside the substrate material due to step volumetric pulse

is obtained by solving the equation that governs the momentum in an elastic linear

solid, which is given in a dimensionless form in the s-domain (Laplace domain) as:

s2 r

^x p

^00 x

r 2

M1 edx M2 e ss2x 3:83

c1

where

F0 c 2 s 2

M1 and

s b s2 2s d2

3:84

F0 c2 s2 d

M2 p

s b s2 2s d2 s2 2s

and C1 and C2 are the dimensionless wave speed and the dimensionless thermal

modulus, respectively.

Solution of Eq. 3.83 can be written as:

sx c21 M1 dx c2 M p

csx

^ B1 ec1 B2 e

r 1 2

e 2 1 2 e ss2x 3:85

c21 d s2 s c1 s 2 s

For the boundness of the solution B1 0, and the condition of free stress (i.e.

^ 0) on the surface is imposed as this stage to calculate the other coefficient B2

r

as:

!

M 1 M 2

B2 c21 2 2 3:86

c1 d s2 s c21 s 2 s

r ^1 x; s r

^x; s r ^2 x; s 3:87

where

sx

c21 c2 F0 s2 e c1 edx

^1

r 3:88

s2 2s d2 s2 c21 d2

100 3 Analytical Solution of Cattaneo and Thermal Stress Equations

sx p

dc21 c2 F0 s e c1 e ss2x

^2 2

r p 3:89

c1 s 2 s ss 2 d2 ss 2

^1 and r

The inversions of r ^2 will be carried out next.

For convenience, the first part is divided into two parts,

^1 x; s r

r ^1a x; s r

^1b x; s 3:90

where

c21 c2 F0 s2 edx

^1a

r 3:91

s2 2s d2 s2 c21 d2

and

sx

c21 c2 F0 s2 e c1

^1b

r 3:92

s 2s d2 s2 c21 d2

2

Using standard partial fraction techniques and Laplace tables, the inverse

Laplace transform of the a-part can be evaluated as:

p p

c2 c21 F0 c21 d2 1 1 d2 1 1 p

t d2 11 xd

r1a x; t

p 2 e

2 d2 1 c21 1 d2 4c21

p p

c2 c21 F0 c21 d2 1 1 d2 1 1 p

t d2 11 xd

p 2 e

2 d2 1 c21 1 d2 4c21

c2 c21 F0 c2 c2 F0

ec1 tdxd 2 1 ec1 tdxd

2dc1 2 c1 d d 2d c1 1 d 2c1

3:93

csx

It is clear that r ^1a e where A exd and therefore, the Laplace s-

^1b A r 1

Here again, U is the unit step function. Thus, the b-part can be inverted as:

xd sx sx

r1b x; t e U t r1a x; t 3:95

c1 c1

where r1a has already been calculated in Eq. 3.93.

Consequently, the first part of the total stress is explicitly determined as:

r1 x; t r1a x; t r1b x; t 3:96

3.3 Volumetric Source Consideration 101

Analogous to the first part of stress, this part of stress can be decomposed into:

^2 x; s r

r ^2a x; s r

^2b x; s 3:97

where

sx

dc2 c2 F0 e c1

^2a x; s 2

r 1 p 3:98

c1 s 2 s ss 2 d2 ss 2

and

p

dc21 c2 F0 e ss2x

^2b x; s 2

r p 3:99

c1 s 2 s ss 2 d2 ss 2

The convolution theorem is employed to get the inverse Laplace transform of

both parts. First, the a-part can be written as a multiplication of two functions:

^ a x; s N

^2a x; s R

r ^ a x; s 3:100

where

^ a x; s c21 c2 F0 d

R 3:101

c21 s 2 s ss 2 d2

sx

e c1

b a x; s p

N 3:102

ss 2

Therefore, the inversion of r^2a is obtained as:

Z t

1

r2a x; t L r^2a x; s Ra x; s Na x; t sds 3:103

0

where

2 2c2 t

c1 1 c2 c21 F0 d c2 1

1

1 ^

Ra x; t L Ra x; s e 1

2

c21 1 d2 4c21

p p

c2 c21 F0 d c21 d2 1 1 d2 1 1 p

t d2 11

p 2 e

2 d2 1 c21 1 d2 4c21

p p

c2 c21 F0 d c21 d2 1 1 d2 1 1 p

t d2 11

p 2

e

2 d2 1 c21 1 d2 4c21

3:104

102 3 Analytical Solution of Cattaneo and Thermal Stress Equations

and

x x cx t

Na x; t L 1 ^

Na x; s U t I0 t e1 3:105

c1 c1

Similarly, the b-part can be expressed as a product of two functions as:

^ b x; s N

^2b x; s R

r ^ b x; s 3:106

where

^ b x; s dc21 c2 F0

R 3:107

c21 s 2 s ss 2 d2

and

p

^ e ss2x

Nb x; s p 3:108

ss 2

As before, the inversion of r ^2b is obtained using the convolution theorem:

Z t

r2b x; t L1 r

^2b x; s Rb x; s Nb x; t sds 3:109

0

where

2 2c2 t

c1 1 c2 c21 F0 d c2 1

1

1 ^

Rb x; t L Rb x; s e 1

2

c2 1 d2 4c21

1p p

c2 c21 F0 d c21 d2 1 1 d2 1 1 p

t d2 11

p 2 e

2 d2 1 c21 1 d2 4c21

p p

c2 c21 F0 d c21 d2 1 1 d2 1 1 p

t d2 11

p 2

e

2 d2 1 c21 1 d2 4c21

3:110

and

hpi

Nb x; t L1 N^ b x; s et Ut xI0 t2 x2 3:111

Having obtained r2a x; t and r2b x; s using Eqs. 3.103 and Eq. 3.109

respectively, the second part of stress is completely determined in the time

domain:

r2 x; t r2a x; t r2b x; t 3:112

3.3 Volumetric Source Consideration 103

rx; t r1 x; t r2 x; t 3:113

where r1 and r2 have already been computed as in Eqs. 3.96 and 3.112,

respectively.

The temperature and thermal stresses fields can be obtained from Eqs. 3.82 and

3.113.

The temporal variation of the volumetric source term in the heat conduction

equation can be represented by an exponential function, which can be constructed

through subtraction of two exponential functions. The analysis presented below in

in line with the previous study [4]. However, Catteneo equation pertinent to the

conduction heating including the volumetric heat source is similar to Eq. 3.60, i.e.:

o2 T oT o2 T

a F 0 f tgx e 3:114

ox2 ot ot2

where F0 F0 =qCe with initial and boundary conditions:

oT

T0 and 0 at t0 3:115

ot

and

oT

0 at x0 3:116

ox

The laser pulses are not exactly in a step pulses, but having decaying intensity

tail. Similarly, the rise of the pulse intensity is not exactly step, but with increasing

rate. This can be described as an exponential pulse. In addition, introducing the

function f t eb1 t eb2 t provides a convenience in mathematical arrange-

ments for obtaining the closed form solution. The heating pulse takes the expo-

nential form:

and

gx edx 3:118

Since the model is linear, one can use the superposition principle to compute

the temperature distribution due to the first term of f t, i.e. eb1 t only, and then the

solution due to the second term is achieved by replacing b1 with b2 . In the sequel a

generic b will be used in the formulation to refer to either b1 or b2 .

104 3 Analytical Solution of Cattaneo and Thermal Stress Equations

Taking the Laplace transformation of the governing Eq. 3.114 and the boundary

conditions (Eqs. 3.115 and 3.116) results in:

o2 T^ F0 dx

a 2

sse 1T^ e 3:119

ox sb

and

oT^

q K

^ es^

q0 at x0 3:120

ox

The last equation can be rewritten as:

oT^

at x = 0: 3:121

ox

The solution of the governing equation in the s-domain can be represented as:

p p

^ x ps aes

2

p

x s2 es I0exd

Tx; s C1 e C2 e a 3:122

s b es2 s ad2

For the solution to be bounded, C2 must vanish and therefore:

p

^

px ss1=e I0exd

Tx; s C1 e a=e

3:123

s b es2 s ad2

By applying the transformed boundary condition, the constant C1 can be

determined as:

p

F0 d a

C1 p

s b ses 1 es2 s ad2

Therefore, the temperature distribution in the s-domain can be obtained as:

p

xd

p x ps2es

b I0e I0 a de a

T 2

p 3:124

2

s b s e s ad s b sse 1 s e s ad2

2

where the homogeneous and particular solutions of the temperature are:

T^p I0 exd ^

hs 3:126

and

p

^

T^h I0 ad /s 3:127

3.3 Volumetric Source Consideration 105

where

^ 1

hs 3:128

s b s e s ad2

2

and

^

/s ^ ^

hsws 3:129

where

p

x ps aes

2

^ e

ws p 3:130

sse 1

To invert the particular part T^p into the time domain, partial fraction and

standard inverse Laplace tables are employed to invert hs and for sake of brevity,

tedious intermediate steps are glossed over here.

1 h tpc p

ht p 2 e e c 1 2be

2 c ad eb2 b

p

p 3:131

p t2be c1 p t ct

2 ce 2e c 1 2be e 2e

is:

Fo edx h tpc p

Tx; t p 2 e e c 1 2be

2 c ad eb2 b

p

p 3:132

p t2be c1 p t ct

2 ce 2e c 1 2be e 2e

" s# " #

1 x 2 x

2et

ut e Io t2 U t p 3:133

2e a=e a=e

where Io is the modified Bessel function of the first kind and U is the unit step

function.. Having obtained the inversion of ^ ^ s, one can use the con-

hs and w

volution technique to calculate the inverse Laplace of /s as:

Zt

;t hsut sds 3:134

0

106 3 Analytical Solution of Cattaneo and Thermal Stress Equations

Th x; t Fo edx ;t 3:135

By combining the homogeneous and the particular solutions of the overall

temperature distribution is obtained as:

Tx; t Tp x; t Th x; t 3:136

The stress distribution inside the substrate material due to exponentially

decaying volumetric pulse is obtained by solving the equation that governs the

momentum in an elastic linear solid, which is given in a dimensionless form in the

s-domain as:

s2 r

^x p

^00

r 2

M1 edx M2 e ss2x 3:137

c1

where

F0 c 2 s 2

M1 and

s b s2 2s d2

3:138

F0 c2 s2 d

M2 p

s b s2 2s d2 s2 2s

and c1 and c2 are the dimensionless wave speed and the dimensionless thermal

modulus, respectively.

Solution of Eq. 3.137 can be written as:

sx c21 M1 c21 M2 p

csx dx ss2x

^ B1 ec1 B2 e

r 1 2

e 2

e 3:139

c21 d s2 sc1 s 2 s

For the boundness of the solution B1 0, and the condition of free stress (i.e.

^ 0) on the surface is imposed as this stage to calculate the other coefficient B2

r

as:

" #

M 1 M 2

B2 c21 2 2 3:140

c1 d s2 sc21 s 2 s

^x; s r

r ^1 x; s r

^2 x; s 3:141

where

sx

c21 c2 F0 s2 e c1 edx

^1

r 3:142

s b s2 2s d2 s2 c21 d2

3.3 Volumetric Source Consideration 107

and

sx p

dc21 c2 F0 s e c1 e ss2x

^2

r p 3:143

s b c21 s 2 s ss 2 d2 ss 2

The inversions of r1 and r2 will be carried out next.

Utilizing t-shifting rule of Laplace transform, the partial fraction technique and

the standard Laplace tables, the inverse of stress component r ^1 can be written as:

x x

r1 x; t kx; t U t k x; t 3:144

c1 c1

where

8 9

>

> 2b2 etb 2et m sinhtv v d2 c21 2b d2 coshtv >>

>

> >

>

1 2 < b2 2b d2 b2 c2 d2 v d2 l2 4c21 d2 bg =

xd 1

k c1 c2 F0 e

2 >

> c1 ec1 td c1 ec1 td >

>

>

> >

>

: ;

c1 d bdl 2c1 c1 d b2c1 dl

2

et v coshtv 2c21 b d l m sinhtv

h21 c21 c2 F0 d

v d2 l2 4c21 d2 bg

2c2 t

1

!

2c21 le l betb

2 2

c1 g b d l2 4c21 c21 g b bg d2

x

t x x

w21 ec1 U t I0 t

c1 c1

h22 h21

hpi

w22 et Ut xI0 t2 x2

3:145

p

v d2 1

gb2

3:146

l c21 1

m d2 c21 b 1 b 1 2c21 b

The second part of the stress is now written as a summation of two sub-

functions as:

^2 x; s r

r ^21 x; s r

^22 x; s 3:147

where

sx

c21 c2 F0 d s e c1

^21 p 2

r 3:148

ss 2 c1 s 2 s s b ss 2 d2

108 3 Analytical Solution of Cattaneo and Thermal Stress Equations

and

p

c21 c2 F0 d s e ss2x

r

^22 p 2 3:149

ss 2 c1 s 2 s s b ss 2 d2

To invert r

^21 , the convolution theory is employed here as well:

Z t

r21 x; t h21 sw21 t sds 3:150

0

Similarly, the inversion of r ^21 is carried out using the convolution theory:

Z t

r22 x; t h22 sw22 t sds 3:151

0

where h22 and w22 are given in Eq. 3.145 Therefore, the inverse Laplace transform

of r2 is simply written as:

r2 x; t r21 x; t r22 x; t 3:152

Hence, the two components of the transformed stress have been inverted and as

a final step, one has to sum both terms to get the total stress distribution in space-

time domain:

rx; t r1 x; t r2 x; t 3:153

electron and lattice sub-systems. The collisional process occurring between elec-

trons and lattice site governs the energy transfer in the irradiated region. Since

electrons gain energy from the irradiated field through the absorption process, their

excess energy increases rapidly and electrons attain higher temperature than that of

the lattice phonons. As the irradiated pulse ends, this process continues until the

temperature equilibrium between electrons and lattice phonons are achieved. Since

the energy transfer takes place in a short time, a hyperbolic heating model may

represent the heat transfer taking place in the irradiated material. Heat transfer due

to conduction and the work done during the formation of thermal stresses are

irreversible and cause entropy generation in the heated region. The entropy gen-

eration rate can be considered one of the measures of the energy dissipated due to

the thermodynamic irreversibility during the laser heating process. However,

3.4 Entropy Analysis 109

for the thermodynamic irreversibility taking place in the temperature and stress

fields. In addition, it facilitates the parametric analysis for improved understanding

of thermodynamics of the laser irradiation process. The analytical formulation of

the entropy generation rate is given in line with the previous work [5].

The analytical solutions presented for the temperature and stress fields can be

used to predict the entropy generation rate during heating process. Yilbas [9]

formulated the entropy generation due to temperature and stress fields. Following

his approach, the entropy rate can be written as:

DS 1 dT k 2 3E T0 de

qCp rT aT 3:154

Dt T dt T 1 2m T dt

The strain de dt can be calculated from the stress field. For instance, for the

exponential laser pulses, the following equations can be incorporated in Eqs. 3.155

and 3.158 to compute the entropy generation rate. The temperature distribution can

be evaluated from:

Z

Tx; t t

0

T^1 sT^2 t sds 3:155

where

p !

1

1 1 px ss1=e

L T^1 x; s L p e a=e

ss 1=e

s! " #

t 1 x 2 x

e2e Io t2 U t p 3:156

2e a=e a=e

and

c

L1 T^2 x; s L1 cetb 3:157

sb

here I0 is the modified Bessel function of the first kind and U is the unit step

function.

The stress distribution can be obtained from:

rx; t rh x; t rp x; t 3:158

where

Z t

1 2

rh x; t ^h s^

r rh t sds 3:159

0

110 3 Analytical Solution of Cattaneo and Thermal Stress Equations

and

1 1

rh x; t L1 r

^h

!

2c1 xc1 t

c21 c2 xb

tb 4c2 c2 1

x

2 bb 2ec1 2 1 2 e 1 U t

c1 b 2 b c1 1 c1

c21 c2 x

2 d t

c1 1 c1

3:160

and

2 2

rh x; t L1 r

^h et J0 it 3:161

and

2c2

!

1 2b b2 bt 4c21 21t

r1 1 1

rp c21 c2

p x; t L ^ 2

dt 2 2

e 2 2

e c1 1

c1 1 c1 b 2c1 b 2 2

c1 1 c1 b 2c1 b

3:162

and

p

r2 1 2

rp et I0

p x; t L ^ t 2 x2 U t x 3:163

The similar arguments can be used for the step input laser pulses for the entropy

calculations.

The findings, form the solution of Cattaneos equation, are presented for two

different heat source considerations and two different laser pulse types. In addition,

thermodynamic irreversibility and entropy analysis due to temperature and stress

fields are given in details. The results obtained from the simulations of the ana-

lytical solutions are presented under the appropriate sub-headings in line with the

previous studies [15].

The surface heat source consideration is divided into two categories including the

step input laser pulse and time exponentially decaying laser pule.

3.5 Findings and Discussion 111

The results obtained from the analytical solutions of Cattaneo and the thermal

stress equations are presented below in line with the previous study [1].

Figure 3.3 shows temporal variation of dimensionless stress distribution at

different locations below the surface. Thermal stress developed appears to be in

wave form such that it decays first sharply and later the decay becomes gradual in

the amplitude. The stress wave propagates into the substrate material with a

constant speed and with almost constant peak amplitude, provided that the

amplitude decays incrementally due to the damping effect of the substrate material.

During the heating cycle, the wave generated is tensile with positive amplitude.

However, in the cooling cycle, it becomes compressive with negative amplitude.

The generation of the tensile wave in the heating cycle is attributed to the thermal

expansion of the substrate material in the surface region. This generates a positive

thermal displacement at the surface. Consequently, the stress wave generated in

the surface vicinity becomes tensile. Once the thermal stress wave is generated, it

propagates into the substrate material with the speed of sound, c, [19]. Conse-

quently, at depth below the surface and at the time it reaches any location, the

wave appears as repeating at this particular location. In the case of the cooling

cycle (t C 400), the sudden cooling of the surface results in contraction in the

surface region and generates a compression wave propagating into the substrate

material. Since the tensile wave is generated in an earlier stage than the com-

pressive wave, both waves do not meet at any location in the substrate material.

Consequently, canceling of the amplitudes of the tensile and the compression

waves in the substrate is less likely to occur. The compression wave behaves

similar to the tensile wave; in which case, the wave propagates at constant speed

and the wave amplitude decays slowly as it propagates into the substrate material

due to the damping effect of the substrate material.

Figure 3.4 shows dimensionless thermal stresses developed inside the substrate

material for different times in the heating period. The thermal stress developed

during the heating cycle appears as the stress wave propagating at a constant wave

stresses versus time at

different locations.

(x1 = 100; x2 = 200;

x3 = 300; x4 = 400)

112 3 Analytical Solution of Cattaneo and Thermal Stress Equations

stresses distribution inside the

substrate material for

different times. (t1 = 70;

t2 = 124; t3 = 179;

t4 = 233)

speed. The wave front is sharp and a tail is formed at the back of the wave. The tail

of the wave is due to the temperature gradient developed inside the substrate

material. In this case, the thermal strain developed inside the substrate material

during the heating cycle is responsible for the formation of the wave tail, which

decays gradually towards the surface. As the wave further propagates into the

substrate material the tail of the wave extends into the substrate material. More-

over, the amplitude of the tail of the wave reduces in the substrate material due to

the change in the temperature gradient along the depth below the surface.

Exponential Pulse Heating:

The findings of the closed form solutions of Cattaneo and thermal stress equations

are presented in line with the previous study [2].

Figure 3.5 shows temporal variation of dimensionless temperature at different

locations inside the substrate material. The peak temperature attains the maximum

at the surface and as the locations move at some distance below the surface, the

peak temperature reduces. Moreover, the location of the peak temperature below

of dimensionless temperature

at different locations inside

the substrate material

3.5 Findings and Discussion 113

the surface changes in time due to the finite speed of temperature in the substrate

material. In general, the material response to the heating pulse at the surface is

slow, since the maximum peak pulse intensity occurs at dimensionless time t = 10

while the peak temperature occurs at around dimensionless time t & 25 at the

surface. This is associated with the pulse profile and the energy transfer taking

place in the surface region during the heating cycle of the short pulse. In this case,

the energy deposited to the substrate material during the early heating duration is

not sufficient to rise temperature at the same rate as the rate of pulse rise. As the

heating period progresses, the rate of temperature rise becomes high, particularly

at the surface. The internal energy gain in the surface vicinity from the heating

pulse is responsible for the high rate of temperature rise at the surface. However,

as the heating period progresses further, temperature decays from its peak value.

The rate of temperature decay is slower than that corresponding to the heating

pulse. This is particularly true at some depth below the surface. In this case, the

internal energy gain in the surface vicinity results in the attainment of high tem-

perature and diffusional energy transfer from the surface region to the solid bulk

takes place in a finite speed, which is less than the decay rate of the pulse intensity.

As the pulse intensity decays beyond 5 % of its peak intensity, temperature decay

becomes gradual. Consequently, the temporal gradient of temperature becomes

small; the Cattaneo equation tends to reduce the Fourier equation. Therefore,

diffusional energy transport with infinite speed governs the energy transfer in the

substrate material. This is more pronounced in the region below the surface.

Figure 3.6 shows temporal behavior of dimensionless thermal stress at different

locations inside the substrate material. The compressive stress waves are formed

first in the surface vicinity and they moved into the substrate material as the

heating period progresses. The compression due to initial heating results in the

formation of the compressive waves. However, the expansion of the initially

compressed surface results in tensile wave formation as the time progresses. This

appears as positive wave amplitude in the tail of the stress wave. This situation is

true for all the locations below the surface. The magnitude of the compressive

wave is larger than the amplitude of the tensile part of the wave. This indicates the

of dimensionless thermal

stress at different locations

inside the substrate material

114 3 Analytical Solution of Cattaneo and Thermal Stress Equations

initially formed wave propagates in compression form into the substrate material.

Moreover, the rate of rise of the amplitude of the compressive wave is higher than

that of the decay rate of the wave. This behavior is similar to temperature response

of the heated substrate material. Consequently, rapid heating and gradual cooling

of the substrate material in the surface region generates a compressive wave in a

similar fashion, provided that the rates of temperature rise and decay are not the

same as the rates of rise and decay of the thermal stresses.

The step input pulse intensity resembling the laser pulse is incorporated in the

analysis and the absorption of the laser beam is modeled after considering the vol-

umetric heat source. The findings are presented in line with the previous study [3].

Figure 3.7 shows the temporal variation of temperature distribution at different

depths inside the substrate material. Temperature rises sharply at the surface

reaching its maximum prior to the step input pulse ends. However, the rise of

temperature is slightly less towards the end of the pulse as compared to its

counterpart corresponding to the mid-pulse-length. The high rate of rise at the mid-

pulse-length is associated with the internal energy gain of the substrate material

from the irradiated field. In addition, the rate of heat diffusion from surface region

towards the solid bulk is less in the early heating period. This enhances the rise of

temperature at the surface. Once the laser pulse ends, temperature first decays

sharply at surface and as the heating period progresses, temperature decay

becomes gradual. The wave nature of heat transfer becomes important during the

rapid decay of temperature at the initiation of the cooling period. This is because

2

of ootT2 , which becomes significantly large during this period. However, once

of dimensionless temperature

at different depths below the

surface

3.5 Findings and Discussion 115

2

temperature decay becomes gradual the term ootT2 becomes small and the wave

nature of the heating replaces with the diffusional heating in the late cooling

period. The temporal behavior of temperature changes as the depth below the

surface increases. In this case, the rise and decay rates of temperature reduce,

which are more pronounced at dimensionless depths x3 and x4. Temporal behavior

of temperature at different depths reveals that the energy transfer from the irradiant

surface as well as the internal energy gain due to absorption do not result in the

rapid rise of temperature at some depth below the surface, which opposes to its

counterpart observed at the surface. This is attributed to the internal energy gain

and energy loss from this region towards the solid bulk through the wave motion

and diffusion. Since the temporal gradient of temperature oT ot becomes small at

some depth below the surface, the second time derivative of temperature also

becomes small. This, in turn, results in diffusional heat transfer from the surface

region to the solid bulk of the material with progressing time. Consequently, the

wave nature of heat transfer replaces with the diffusional heat transfer at some

depth below the surface during both heating and cooling periods.

Figure 3.8 shows temporal variation of thermal stress at different depths below

the surface. Thermal stress is compressive during the heating cycle irrespective of

the locations inside the substrate material. The stress waves compressive and they

have similar rise and fall times. The stress wave generated in the early heating

period has smaller amplitude as compared to those generated at different periods.

This is associated with the temperature gradient, which is low in the early heating

period because of the temperature rise. It should be noted that the temperature

gradient becomes high once the heating period progresses as observed from the

temperature distribution inside the substrate material (Fig. 3.3). Consequently the

amplitude of the stress wave becomes high as the heating cycle progresses towards

the pulse ending. Moreover, once the heating pulse ceases, stress wave becomes

tensile. Consequently the expansion of the free surface after the compression is

responsible for the formation of the tensile stress waves.

of dimensionless thermal

stress at different depths

below the surface

116 3 Analytical Solution of Cattaneo and Thermal Stress Equations

The closed form solutions of Cattaneo and thermal stress equations are presented

for time exponentially varying short pulse. The pulse is treated as a volumetric

source to incorporate the absorption of the incident laser irradiation. The findings

are presented below in line with the previous study [4].

Figure 3.9 shows temporal variation of dimensionless temperature at different

locations inside the substrate material. The peak temperature attains the maximum

at the surface and as the locations move at some distance below the surface, the

peak temperature reduces. Moreover, the location of the peak temperature below

the surface changes in time due to the finite speed of temperature in the substrate

material. In general, the material response to the heating pulse at the surface is

slow, since the maximum peak pulse intensity occurs at dimensionless time

t = 2.5 while the peak temperature occurs at around dimensionless time t 7:5 at

the surface. This is associated with the pulse profile and the energy transfer taking

place in the surface region during the heating cycle of the short pulse. In this case,

the energy deposited to the substrate material during the early heating duration is

not sufficient to rise temperature at the same rate as the rate of pulse rise. In

addition, the temperature gradient developed in the surface vicinity enhances the

heat diffusion from the surface to the solid bulk as the heating cycle progresses.

This contributes to temperature rise at the surface such that the delay in the peak

temperature is observed at the surface. Moreover, the rate of rise of temperature in

the early heating period is slow, which is more pronounced at locations some depth

below the surface. However, as the heating period progresses, the rate of tem-

perature rise becomes high, particularly at the surface. The internal energy gain in

the surface vicinity from the heating pulse is responsible for the high rate of

temperature rise at the surface. However, as the heating period progresses further,

temperature decays from its peak value. The rate of temperature decay is slower

than that corresponding to the heating pulse. This is particularly true at some depth

below the surface. In this case, the internal energy gain in the surface vicinity

results in the attainment of high temperature and diffusional energy transfer from

temperature distribution

versus time at different

depths,

x1 0; x2 1; x3 3,

and x4 5

3.5 Findings and Discussion 117

the surface region to the solid bulk takes place in a finite speed, which is less than

the decay rate of the pulse intensity. As the pulse intensity decays beyond 5 % of

its peak intensity, temperature decay becomes gradual. Consequently, the temporal

gradient of temperature becomes small; the Cattaneo equation tends to reduce the

Fourier equation. Therefore, diffusional energy transport with infinite speed gov-

erns the energy transfer in the substrate material. This is more pronounced in the

region below the surface.

Figure 3.10 shows temporal behavior of dimensionless thermal stress at dif-

ferent locations inside the substrate material. The compressive stress waves are

formed first in the surface vicinity and they moved into the substrate material as

the heating period progresses. The compression due to initial heating results in the

formation of the compressive waves. However, the expansion of the initially

compressed surface results in tensile wave formation as the time progresses. This

appears as positive wave amplitude in the tail of the stress wave. This situation is

true for all the locations below the surface. The magnitude of the compressive

wave is larger than the amplitude of the tensile part of the wave. This indicates the

initially formed wave propagates in compression form into the substrate material.

Moreover, the rate of rise of the amplitude of the compressive wave is higher than

that of the decay rate of the wave. This behavior is similar to temperature response

of the heated substrate material. Consequently, rapid heating and gradual cooling

of the substrate material in the surface region generates a compressive wave in a

similar fashion, provided that the rates of temperature rise and decay are not the

same as the rates of rise and decay of the thermal stresses.

Entropy generation rate due to temperature and stress fields are evaluated for time

exponentially varying laser short-pulse. The findings are presented below in line

with the previous findings.

thermal stresses distribution

with time at different depths,

x1 0; x2 1; x3 3,

and x4 5

118 3 Analytical Solution of Cattaneo and Thermal Stress Equations

entropy generation due to

heat transfer and stress

developed with depth at

different times

Figure 3.11 shows dimensionless entropy generation rate due to heat transfer

and thermal stress developed inside the substrate material for different heating

periods. Entropy generation rate is low in the surface region as compared to at

some depth below the surface. This is attributed to 1/T term in the entropy

equation, which becomes high when temperature reduces. Although temperature

gradient is high at some depth below the surface, its contribution to entropy

generation rate is lower than that of the term 1/T in the entropy equation. This

situation can also be observed when examining the entropy curves at different

heating periods. In this case, entropy generation rate reduces with progressing time

in the surface region despite the fact that the temperature gradient in the region

next to the surface vicinity increases with progressing time.

Figure 3.12 shows entropy generation rate due to thermal stress developed

inside the substrate material. Entropy generation rate in the surface region is

negligibly small and increases slightly as the depth below the surface increases.

This behavior is associated with the propagation of the thermal stress waves; in

entropy generation due to

stress developed inside the

material for different times

3.5 Findings and Discussion 119

which case, the maximum magnitude of stress wave increases with progressing

time. When comparing the entropy generation rate due to heat transfer and thermal

stress developed, it is evident that entropy generation rate due to heat transfer is

significantly higher than that of thermal stress developed. This is because of the

energy dissipated by heat transfer, which is significantly higher than the work done

during the thermal expansion of the substrate material.

References

1. H. Al-Qahtani, B.S. Yilbas, The closed form solutions for Cattaneo and stress equations due to

step input pulse heating. Physica B 405(18), 38693874 (2010)

2. H. Al-Qahtani, B.S. Yilbas, Temperature and stress fields for short pulse heating of solids.

J. Thermophys. Heat Transf. 25(1), 173176 (2011)

3. H. Al-Qahtani, B.S. Yilbas, Closed form solution of Cattaneo equation including volumetric

source in relation to laser short-pulse heating. Can. J.Phys. 89(7), 761767 (2011)

4. H. Al-Qahtani, A closed form solution of temperature and stress fields for laser short-pulse

heating of a solid: exponentially decaying volumetric source. Lasers Eng. 22(12), 109124

(2011)

5. H. Al-Qahtani, B.S. Yilbas, Entropy generation rate during laser short-pulse heating:

contribution of heat transfer and thermal stress. Lasers Eng. (2012) (in print)

6. A.D. Kovalenko, Thermoelaticity: Basic Theory and Applications (Wolters-Noordhoff

Publishing, Groningen, 1969), p. 190

7. B.A. Boley, J.H. Weiner, Theory of Thermal Stresses (Robert E. Krieger Publishing Company,

Malabar, 1985)

8. T. Darabseha, M. Najia, M.A. Al-Nimr, Transient thermal stresses in an orthotropic cylinder

under the hyperbolic heat conduction model. Heat Transf. Eng. 29, 632642 (2008)

9. H. Al-Qahtani, B.S. Yilbas, Entropy generation rate during laser pulse heating: effect of laser

pulse parameters on entropy generation rate. Opt. Lasers Eng. 46, 2733 (2008)

Chapter 4

Analytical Treatment of Hyperbolic

Equations for Stress Analysis

Abstract Laser ultra-short pulse heating of metallic surfaces causes the hyperbolic

behavior of energy transport in the heated region. The consideration of the parabolic

nature of the non-equilibrium heating situation fails to formulate the correct heating

process. Although heating duration is ultra-short, material response to the heating

pulse is not limited to only heat transfer and the mechanical response of the heated

surface also becomes important. Consequently, mechanical response of the surface

under ultra-short thermal loading becomes critical in terms of the generation of the

high stress levels. In this chapter, hyperbolic behavior of heat transfer is introduced

in the laser heated region. The closed for solutions for the temperature and stress

fields are obtained for various heating situations. Two-dimensional effect of heating

on temperature rise is also considered for nano-scale applications.

4.1 Introduction

When the heating duration becomes less than the thermalization time of the substrate

material and the size of the irradiated region becomes comparable to the lattice

phonon mean free path, the equation governing the laser heating becomes hyperbolic

in nature. The analytical solution of hyperbolic heat equation can be achieved using

the Lie symmetry, or Laplace transformation technique, or perturbation method.

However, the formulation of the physical problem requires deep understanding of

electron and phonon behavior in the metallic substrates and microscopic modeling

of the process. When the short-pulse laser irradiation interacts with the metallic

surface, thermal separation of the electron and the lattice sub-systems takes place in

the irradiated region during the short heating period. The thermal communication

between electron and lattice sub-systems results in non-equilibrium energy transport

in the heated region. The collisional process taking place between excited electrons

and the lattice sub-system governs the energy transfer from the electron sub-system

to the lattice sub-system. This process continues until the thermal equilibrium is

established between the sub-systems. However, for the heating duration, which is

and Thermal Stress Analysis, Materials Forming, Machining and Tribology,

DOI: 10.1007/978-3-319-00086-2_4, Springer International Publishing Switzerland 2014

122 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

through the collisional process while dominating over the diffusional energy transfer

in the solid. In this case, the parabolic heating model fails to describe the physical

insight into heat transfer in the substrate material. Consequently, electron kinetic

theory approach incorporating the electron-lattice site collisions between the lattice

and electrons sub-systems at microscopic level becomes essential to account for the

formulation of the energy transport in the solids. Moreover, the closed form solution

for the governing equation of the physical problem becomes fruitful, since it pro-

vides the functional relation between the independent variables, such as time and

space, and the dependent variable, such as temperature. In order to obtain the ana-

lytical solution to the heat equation, some useful assumptions can be made such as;

electronphonon collision is elastic and the energy transport process based on the

collisions process can be reduced to one-dimensional form. In addition, if the

heating process is limited with the solid heating while excluding the phase change

process, elastic collisions between electrons and lattice site are justifiable. The size

of the absorption depth is considerably smaller than the irradiated spot size, which

enables to simply the problem into one-dimensional space. In this chapter, analytical

solution for the laser short-pulse heating of the metallic substrate is presented in line

with the previous studies [14]. The study covers the one-dimensional and two-

dimensional hyperbolic equations and the thermal stress development during the

short time period.

Substrates at Microscopic Level

The formulation of the heat equation from the electron kinetic approach at micro-

scopic level is given below in the line with the previous study [1]. During the laser

heating of the metallic substrates, electrons within the absorption depth of the

substrate material gain energy from the irradiated field through the absorption

process. This, in turn, increases the electron energy and results in transferring of their

excess energy to lattice site through scattering, which depends upon the duration of

the interaction. In the case of short-pulses (slightly higher than the electronphonon

interaction time), electrons undergo few collisions with lattice site, since the elec-

tronphonon collision time is in the order of 0.02 ps. Electrons in the surface region

continuously gain energy from the irradiated field, which in turn results in energy

differences between the electrons and the lattice site in this region. The specific heat

capacity of electron is much smaller than its counterpart corresponding to the lattice

site; consequently, electron temperature increases rapidly while lattice site tem-

perature increase is gradual during the short heating duration. The temperature

differential between electron and lattice sub-systems results in non-equilibrium

energy transport in the substrate material. However, energy distribution of the

excited electrons may not be uniform in the surface region and also varies with time.

This causes temperature differential occurring in the electron sub-system.

4.2 Formulation of Energy Transport in Metallic Substrates at Microscopic Level 123

in the surface region (x = 0

is the free surface)

In order to formulate the electron kinetic theory approach for the laser short-pulse

heating process, the investigation into the electron motion in the surface region of

the metallic substrates is necessary. The electron motion in the surface region of the

metals due to the irradiated field can be shown schematically in Fig. 4.1.

Electrons travel from surface to solid bulk as well as bulk to solid. Electrons

reaching the surface can escape once their energy exceeds the surface potential

barrier. Consequently, the number of electrons with X fraction reaching the surface

can manage to escape. In order to account for the reflected electrons from the surface,

a mirror image at the surface is considered [5]. Consider the location A in Fig. 4.1, the

net flow of Nsx ( N6 ) number of electrons towards the surface occurs and X fraction

of these electrons may escape from the surface. The situation, which occurs at

location A in Fig. 4.1 is an exact mirror image of that corresponding to point B, i.e.

(1 w) Nsx number electrons still flow towards the surface from location B, i.e.

(1 w) Nsx electrons flow to the right and N6 to the left at location B (Fig. 4.1). In the

case of location A approaches to location B, the number of electrons Nsx , which flow

from s to x changes discontinuously at s 0 and at s x. These changes can be

summarized as follows:

In the negative direction of x

N

1\s\x: Nxs

6

N

x\s\0: Nxs 1 w 4:1

6

N

0\s\1: Nxs

6

and

In the positive direction of x

N

1\s\0: Nsx

6

N

0\s\x: Nsx 1 w 4:2

6

N

x\s\1: Nsx

6

124 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

N

Nsx Nxs 2 w 4:3

6

where Nxs is the number of electrons, which flow from x to s. It should be

noted that changes in Nsx follows a distribution, which can be described by a

rectangle function of unit height and width x centered on the positive s 2x.

Consequently, the rectangle function can be written as:

Y s x Y 2s x

2

4:4

x 2x

where

Y 2s x 1

0 for j2s xj [ 4:5

2x 2

and

Y 2s x

1 for j2s xj\ x 4:6

2x

Therefore, the electron distribution can be described as:

Y 2s x

N N

Nsx 1w with Nsx Nxs 1 w 4:7

6 2x 6

It should be noted that electron energy, which is characterized by temperature

Te s; t, is augmented from the initial Te s; t by an amount equal to that absorbed

in travelling from s to x. The total amount of energy, which is absorbed in an

element dn, area A in time dt is:

Io A f 0 n dt dn 4:8

where Io and f 0 n are the laser peak power intensity and the intensity distribution

function in the absorption depth of the solid material, since all the beam energy is

absorbed in the x-axis. The electron density can vary along the x-axis, in partic-

ular, the number of electrons travelling from ds to dx may not be the same as that

from dx to ds. Therefore, the portion of energy which is absorbed by electrons

which travel from ds to dx in dt is:

Nsx

Io A dt dn f 0 n 4:9

Nsx Nxs

where Nsx and Nxs are the number of electrons which travel from s to x and from x

to s, respectively. The total number of electrons which travel from ds to dx in this

time is:

Nsx AV dt 4:10

where V is electron mean velocity. Hence, the average energy absorbed by one

electron in dn in time dt is:

4.2 Formulation of Energy Transport in Metallic Substrates at Microscopic Level 125

f 0 n

Io dn 4:11

Nsx Nxs V

and the total amount of energy absorbed by this electron from dx to ds is:

Z s

f 0 n

Io dn 4:12

x Nsx Nxs V

This expression gives the extra energy gain by the electrons in travelling from

ds to dx.

Electrons receive energy from the irradiated field and make collisions among

themselves as well as lattice site ions and they transfer some fraction of their

excess energy through the collisional process, i.e. electrons after the first collision

scatter and make further collisions with less energetic electrons and lattice site

ions. The energy exchange between energetic electrons, due to absorption of the

irradiated field, and other species can be formulated after considering the collision

probability of energetic electrons.

Consider the probability of electrons travelling a distance x without making a

collision is [6]:

x

exp 4:13

k

where x\2k and k is the mean free path of the electrons. Consider Fig. 4.1), the

probability of electrons, which make collision in B can be written as:

x

1 exp 4:14

k

or

dx

1 1 4:15

k

or

dx

4:16

k

provided that x\2k. The probability of electrons which last collided in B now

colliding in A is:

ds jx sj dx

exp 4:17

k k k

However, the number of electrons (Nsx ) leaving the location A in Fig. 4.1, area A in

time dt is Nsx AV dt where Nsx is the number density of electrons which transfer

energy from dx to ds, and V is the mean electron velocity. The number these

electrons which have just collided in location A is:

126 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

ds dx jx sj

Nsx AV dt exp 4:18

k k k

where dsk dxk expjxs j

k is the probability of electrons just collided in location A. If the

temperature of lattice site in dx is Tl x; t and the temperature of the electrons

when they arrive at dx is Te s; t (allowing for absorption on the way), then the

energy transfer to the lattice site in dx from collisions with electrons in which the

electrons give up a fraction f of their excess energy is:

ds dx j x sj

Nsx AV dt exp f Ee El 4:19

k k k

where Ee and El are the energy of electron and lattice ion, respectively. The

analysis related to f is given below:

The fraction of electron excess energy transfer during the time comparable or

slightly greater than the electronphonon collision time (sp ) can be written in

terms of the energy balance across the section dx in the substrate material, i.e.:

Electron energy)in (Electron energy)out Te in Te out

f or f ;

Electron energy)in (Phonon Energy) Te in Tl

where (Te ) in is the temperature of an electron entering the section, (Te ) out tem-

perature of the an electron leaving the section, and Tl is the phonon temperature. f

takes the values 0 f 1.

Integrating the contributions from all such infinitely small strips as to the

energy in location B (Fig. 4.1) gives:

Z 1

ds dx j x sj

Nsx AVdt exp f Ee El ds 4:20

1 k k k

In this case, energy transfer during Dt (Dt sp , where sp is the electronphonon

collision time) due to absorption of irradiated field and the collisional process can

be written after incorporating electron distribution function [7]:

Z 1 Y 2s x

DEtrans Vf kB j x sj

2

Nsx 1 w exp Te s; tds

AdxDt 1 k 2x k

Z 1 Y 2s x

Vf kB j x sj

2

Nsx 1 w exp Tl x; tds

1 k 2x k

Z 1 Z s

Io f Nsx jx sj

2 N N

exp f 0 ndnds

1 k sx xs k x

4:21

where

Q

Nsx 1 w 2sx

x

4:22

Nsx Nxs 2w

4.2 Formulation of Energy Transport in Metallic Substrates at Microscopic Level 127

where f is the fraction of electron excess energy, which transfers to lattice site

during a single electron lattice site collision. The first term on the left hand side of

Eq. 4.21 is energy gain by the substrate material through the collisional process,

first and second terms on the right hand side represent electron and lattice energies,

and third term on the right hand site is the energy gain of the electrons due to the

irradiated field.

The final temperature of the electrons in dx after the collisional process can be

readily found from the conservation of energy, i.e.:

Total electron energy after collision = Total electron energy in during dt-

Change of lattice site energy

Total electron energy after collision is:

Z 1 Y 2s x

VkB j x sj

2

N sx 1 w exp Te s; t f Tl x; tds

1 k 2x k

4:23

Total electron energy carried into dx during dt is:

Z 1 Y 2s x

VkB 1 f j x sj

N sx 1 w exp Te s; tds

1 k2 2x k

Z 1 Q Z s 4:24

Io 1 f 1 w 2sx

2x j x sj

exp f 0 ndnds

1 k2 2w k x

Z 1 Y 2s x

VkB j x sj

2

N sx 1 w exp Te s; t f Tl x; tds

1 k 2x k

Z 1 Y 2s x

VkB 1 f jx sj

Nsx 1 w exp Te s; tds

1 k2 2x k

Z 1 Q 2sx Z s

Io 1 f 1 w 2x jx sj

2

exp f 0 ndnds

1 k 2 w k x

4:25

Equations 4.21 and 4.25 can be re-written after considering the electron distri-

butions for the lattice element dx apart and for electrons passing an area A, i.e.:

128 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

Z 1 Z 1

DEtrans fk jx sj fk jx sj

3

exp Te s; tds 3

exp Tl x; tds

AdxDt 1 k k 1 k k

Z j xj Z j xj

fk jx s j fk jx sj

w 3 exp Tl x; tds w 3 exp Te s; tds

0 k k 0 k k

Z 1 Z s

Io f 1 jx sj

2 2 w

exp f 0 ndnds

1 k k x

Z 1 Z s

Io f w jx sj

2

exp f 0 ndnds

1 k 2 w k x

4:26

and

Z 1

k j x sj

2

exp Te s; t f Tl x; tds

1 k k

Z j xj

wk j x sj

exp Te s; t f Tl x; tds

0 k2 k

Z 1 Z j xj

k1 f j x sj k1 f j x sj

exp Te s; tds exp Te s; tds

1 k2 k 0 k2 k

Z 1 Z s

Io 1 f j x sj

2

exp f 0 ndnds

1 k 2 w k x

Z j xj Z s

Io 1 f j x sj

exp f 0 ndnds

0 k2 2 w k x

4:27

where k is the thermal conductivity, which makes use of the simple kinetic theory

result for the electron thermal conductivity [8]:

NVkB k

k : 4:28

3

The energy content of the small lattice site element dx apart can be written as:

DE A dx Cl Tl x; t 4:29

where Cl q Cp. The energy gain of the small lattice element during the small

time interval dt is DE DE

dt . The expansion of dt yields:

" #

DE 1 0 dt2 00

Et dt E t E t Et 4:30

dt dt 2!

or

DE dt 00

E0 t E t . . . 4:31

dt 2!

4.2 Formulation of Energy Transport in Metallic Substrates at Microscopic Level 129

DE o dt o2

A dx Cl Tl x; t Cl Tl x; t 4:32

dt ot 2! ot2

When the time increment approaches to thermal relaxation time (dt ! ss ),

Eq. 4.32 reduces to:

oe DE o o

Cl Tl x; t ss Tl x; t 4:33

ot dt A dx ot ot

where e is the volumetric energy content of lattice site. The energy gain of the

small lattice element through collisional energy transport can also be written as:

DE 1 DEtrans o DEtrans

ss 4:34

A dx Dt A dx Dt ot Dt

Substituting Eq. 4.26 into Eq. 4.34 yields the change of lattice site energy

which is:

0Z 1 Z 1 1

fk jx sj fk jx sj

3 exp Te s; tds 3 exp Tl x; tds

B 1 k k 1 k k C

B C

B Z Z C

B j xj

jx sj

j xj

jx sj C

B

fk

w 3 exp Tl x; tds

fk

w 3 exp Te s; tds C

DE B k k k k C

B 0 0 C

B Z 1 Z s C

AdxDt B jx sj

C

B

Io f 1

exp 0

f ndnds C

B 2 2 w k C

B 1 k x C

B Z 1 s Z C

@ Io f w jx sj

A

0

2 2 w

exp f ndnds

1 k k x

0Z 1 Z 1 1

fk jx sj fk jx sj

exp T e s; tds exp T l x; tds

B 1 k3 k 1 k

3 k C

B C

B Z Z C

B j xj

jx sj

jxj

jx sj C

B w

fk

exp T x; tds w

fk

exp T s; tds C

oB C

3 l 3 e

B 0 k k 0 k k C

ss B Z 1 Z s C

B j j

C

B I o f 1

exp

x s

f 0

ndnds C

B 2 2 w k C

B 1 k x C

B Z s Z C

@ 1

Io f w jx sj

A

0

2

exp f ndnds

1 k 2 w k x

4:35

Equations 4.27 and 4.35 are the energy transport equations of interest for laser

short pulse heating process. However, for small rise of electron temperature during

the low intensity laser heating pulse prevents electron escape from the surface.

Consequently, the term w in Eqs. 4.27 and 4.35 becomes zero.

Equations 4.27 and 4.35 can be transformed into differential equations. The

method of solution to be used in the following analysis is the transformation of the

simultaneous differentialintegral Eqs. 4.27 and 4.35 using the Fourier integral

transformation, with respect to x [7]. This is due to the fact that the resultant ordinary

130 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

differential equations may then be handled much more conveniently. Consider first

reduction of the set of equations to the differential equation of heat conduction.

The Fourier transformation of a function f x is defined by:

Z 1

Ff x expixxf xdx Fx 4:36

1

Z 1

1

f x Fxexpixxdx 4:37

2p 1

Z 1

f ngx sds 4:38

1

f x:gx 4:39

and the transform of function expjkxj is:

2k

4:40

1 x 2 k2

Therefore, the Fourier transform of the function:

Z 1

k j x sj

IX 3

exp Tl x; tds 4:41

1 k k

will be a constant factor (the value of integral) multiplying the transform of the

function Tl x; t, i.e.:

Z 1

kf jx sj

FIX 3 Tl F exp ds 4:42

k 1 k

or

Z 1

kf jx sj

FIX Tl F exp H jsjds 4:43

k3 1 k

where H jsj 1 for 1\s\1.Therefore:

Z 1

kf jx sj

FIX 3 Tl F exp F fH jsjdsg

k 1 k

4:44

kf 2k

3 Tl dx

k x k2 1

2

4.2 Formulation of Energy Transport in Metallic Substrates at Microscopic Level 131

where dx is the Dirac delta function. Since this function only has a value of 1 at

x 0, then the transform is:

kf

Tl 4:45a

k2

Using these results, the Eqs. 4.27 and 4.35 can be Fourier transformed, the

result of which is:

oe k f 2k kf Io df 2k 2d

Te 2 Tl

ot k3 x2 k2 1 k 2k x2 k2 1 d2 x2

4:45b

o kf 2k kf Io df 2k 2d

sp Te 2 Tl

ot k3 x2 k2 1 k 2k x2 k2 1 d2 x2

and

k k1 f 2k Io d1 f 2k 2d

Te f Tl Te

k2 k3 x 2 k2 1 2k x 2 k2 1 d2 x 2

4:46

If the transform function Te is obtained from Eqs. 4.45a, 4.45b using 4.46, the

result is:

oe o 2d

f x2 k2 x2 k f Tl sp x2 k f Tl Io df 2

ot ot d x2

o 2d

sp Io df 2 4:47

ot d x2

Insertion of oeot in terms of Tl and multiplication of Eq. 4.47 which is in the

transform domain, by ix2 corresponds to second order differential in the real

plane. Hence the inversion of the above equation gives:

o k2 o2 oTl o2 Tl o o2 Tl

1 ss C l k s p

ot f ox2 ot dx2 ot dx2

o

Io df expdj xj sp Io df expdj xj

ot

4:48

Equation 4.48 is similar to that obtained from the quasi-ballistic

o2 Tapproach

[9]. It

should be noted that the time derivative of diffusion sp ot dx2 o l

and source

(oto Io df expdj xj) terms are included in Eq. (4.48) as similar to the that obtained

from the quasi-ballistic approach [9].

Parabolic Heating Model:

First, let us consider the two-equation model presented previously [10]. The two-

equation model representing the energy exchange mechanism during phonon

132 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

be written as [10]:

oTe s; t

Ce r : krTe s; t GTe s; t Tl x; t S

ot 4:49

oTl x; t

Cl GTe s; t Tl x; t

ot

Te s; t and Tl x; t are the electron and lattice site temperatures, S is the laser

source term (Io d expdx), and Ce and Cl are the electron and lattice heat

capacities, respectively. G is the electronphonon coupling factor, given by [11]:

2

p2 me N V

G 4:50

6sp Te s; t

where me , N, V, and sp are electron mass, electron number density, electron drift

velocity, and the electronphonon collision time, respectively.

Now, let us consider the equation

2 derived from the electron kinetic theory

approach (Eq. 4.48). If the terms sp oto odxT2l and sp oto Io df expdj xj are neglected in

Eq. (4.48) for all t values, Eq. 4.48 becomes:

o k2 o2 oTl o2 Tl

1 ss C l k Io df expdj xj 4:51

ot f ox2 ot dx2

which is the same as previously formulated kinetic theory model [12], i.e. the time

derivative of diffusion and source terms are omitted in the previously formulated

kinetic theory approach. Eq. 4.51 can be re-written as:

oTl o2 Tl k2 o2 oTl o2 T l

Cl k 2 f 2 q Cp Cl ss 2 Io d expdj xj 4:52

ot dx ox ot ot

Equation 4.52 is a third order partial differential equation, which can be decom-

posed into second and first order two differential equations, i.e., when Eq. 4.52 is

decomposed into two equations, the resulting probable differential equations are:

oTe o2 Te

A B 2 CTe Tl Io d expdj xj

ot dx 4:53

oTl

D CTe Tl

ot

where A, B, C, and D are the coefficients. To find the values of A, B, C, and D, the

following procedure is adopted, i.e.:

o2 Tl oTe oTl

D 2 C 4:54

ot ot ot

4.2 Formulation of Energy Transport in Metallic Substrates at Microscopic Level 133

or:

oTe D o2 Tl oTl

4:55

ot C ot2 ot

Similarly:

2

o2 oTl o Te o2 Tl

D C 4:56

ox2 ot ox2 ox2

or:

o2 Te D o2 oTl o2 Tl

4:57

ox2 C ox2 ot ox2

Substitution of Eqs. 4.57 and 4.55 into Eq. 4.53, and inserting CTe Tl DoTotl

into Eq. 4.53, it yields:

oTl BD o2 oTl o2 Tl AD o2 Tl

D A B Io d expdj xj 4:58

ot C ox2 ot ox2 C ot2

After equating Eqs. 4.52 and 4.58, the coefficients A, B, C, and D can be calcu-

lated, i.e.:

f kss

A

k2

Bk

fk f k ss 4:59

C 2 1

k q Cp k2

f k ss

D q Cp 2

k

Equation 4.53 is identical to Eq. 4.49 given in the two-equation model. Conse-

quently, setting the coefficients of equations Eqs. 4.53 and 4.49, it yields:

f k ss

Ce

k2

fk f k ss

2

1 G 4:60

k q Cp k2

f kss

q Cp 2 Cl

k

where ts Ce =G [10] and Ce cTe (where c is constant; for example, c 96:6 m3JK 2

for copper). Moreover, the electron phonon coupling factor is temperature depen-

dent, which can be written as [11]:

134 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

2

p2 m N V

G 4:61

6 sp Te Te

Equation 4.61 can be used for lattice site temperature less than the Debye tem-

perature. Since the electronphonon collision time sp T1e , the electron phonon

coupling factor becomes constant as temperature increases [13, 14], i.e. it becomes

independent of temperature.

The electron mean free path (k) can be written as [15]:

k Vs 4:62

where s is the electron relaxation time, which is [16]:

3m k

s

p2 N kB2 Te

where N and m are electron number density and effective mass of free electrons,

respectively. kB is the Boltzmanns constant.

and Volumetric Sources

The surface and volumetric source term is associated with the laser output pulse

intensity. In the case of surface heat source consideration, it is assumed that the

laser pulse energy is deposited at the surface of the irradiated material. However,

the volumetric heat source considers the absorption of the incident laser beam

within the skin of the substrate surface. Since the absorption takes place according

to the Beer Lamberts law, exponential decay of the laser intensity along the

absorption depth defines the volumetric heat source in the heat equation.

Equation 4.48 represents the temperature field in the lattice sub-system. The

analysis is presented in line with the previous study [2] and the one-dimensional

form of Eq. 4.48 can be written as:

2

o k2 o2 oTL o2 TL o o TL

1 ss CL k 2 sp k 2

ot f ox2 ot ox ot ox 4:63

f dI t expdjx j

where ss is the electronphonon characteristic time (ss CE =G), G is the elec-

tronphonon coupling factor, k is the mean free path of electrons, f is the fraction

4.3 Thermal Stress Field: Consideration of Surface and Volumetric Sources 135

of excess energy change, CL and CE are the lattice and electron heat capacities,

respectively, k is the thermal conductivity, sp is the electron mean free time

between electronphonon coupling, I t I expd t where I is the laser

peak power intensity, expd t is the temporal distribution function of laser

pulse, d is the absorption coefficient. x is the distance along the x-axis and t is the

time variable. TL and TE are the lattice site and electron temperatures, respectively.

Introducing the following equalities and dimensionless variables:

fk CE

k2 ; ss

G G

4:64

TE TL t k d2 CE

hE ; hL ; x x d; t ; a ;

T T CE =G G CL

to Eq. 4.63 yields finally:

o3 hL o2 hL o2 hL ohL

a 1 l 2

a 2 b expx expc t 0 4:65

ox ot ox2 ot ot

sp G f I d l c

where l CL ; b To G ; c d GCE :

This model provides the improved energy transport equation including ballistic

effects with volumetric source in dimensionless form. Once the lattice site tem-

perature is determined, the electron temperature can be found from:

ohL

hE hL 4:66

ot

Equations 4.65 and 4.66 describe temperature distribution in the lattice and

electron sub-systems.

Moreover, the coefficients of Eq. 4.66 are considered to be independent of

temperature. Hence:

o 3 hL o2 hL o2 hL ohL

a 1 l a 2 b expx expct 0;

ox2 ot ox2 ot ot

ohL

hE hL 4:67

ot

In the absence of the volumetric source, one can get that b 0. So the system

(Eq. 4.67) reduces to the following system:

o3 hL o2 hL o2 hL ohL

a1 l a 2 0;

ox2 ot ox2 ot ot

ohL

hE hL 4:68

ot

Since the absorption depth of the substrate material is significantly smaller than the

thickness of the substrate material, the assumption of a semi-infinite solid heated

136 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

by a time decaying source from the surface can be justifiable. The boundary

conditions for the problem can be written as follows:

ohE ohE

0; t I ek1 t ; 1; t 0

ox ox 4:69

hE x; 1 h ; hL x; 1 h

through the relationI k dI T , and the dimensionless power k1 is related to the

dimensional one d through the relationk1 d GCE c.

Recently, Ylbas et al. [17] constructed the closed form solutions for temper-

ature distribution with surface heat sources. The surface heat source corresponds to

the short duration contact of the gold film with a heat source at the surface. The

solution of the boundary value problem can be given, when:

k1 [ 1 & k1 \ or k1 \1 & k1 [ 4:70

1l 1l

as follow:

I

hL x; t h Expk1 t xx

xk1 k1

4:71

I

hE x; t h Expk1 t xx

xk1

q

where x a k1 k1k 1

1 1l

:

n be noted that the solution

o x; t in Eq. 4.71 is only

valid for k1 [ 1 & k1 \ 1l or k1 \1 & k1 [ 1l . In which case, e is of

2

order 10 for metals and k1 \1for short pulse heating situation.

The equation governing the momentum in one-dimensional solid for a linear

elastic case can be written as [18]:

o2 rx 1 o2 rx o2 T L

2

2 2 c2 2 4:72

ox c1 ot ot

where c1 is the wave speed in the solid.

q

c1 Eq and c2 1m1m q aTL where mis Poissons ratio, qis the density of the

solid and aTL is the thermal expansion coefficient of the solid.

Our goal in this paper will be solving Eq. 4.72 to find the exact solution for

thermal stresses corresponding to the closed form solutions for temperature dis-

tribution hL with surface heat sources. So by using the dimensionless variables in

2

Eq. 4.65, the dimensionless stress isrx c2 dd2 T0 rx . Equation 4.72 can be

4.3 Thermal Stress Field: Consideration of Surface and Volumetric Sources 137

Therefore Eq. 4.72 becomes:

o2 rx 2 o rx

2

A B2 Expc t x x 4:73

ox2 ot2

q

I

where A c1dc d and B c xc :

rx x; 0 0; rx x; 1 0 4:74

Now, applying the Fourier sine transform for Eq. 4.73 with respect to t, by

requiring that, for a physically meaningful system, rx x; t and its first derivatives

in t vanish as t ! 1. To obtain the analytical solution for Eq. 4.73, Fourier sine

and Fourier cosine transformations are used. In this case, Eq. 4.73 yields:

r r

o2 2 2 2 2 2 2 Expc x x

2

U x; p A p rx x; 0 A p U x; p B p 4:75

ox p p c 2 p2

where p is the Fourier sine variable and U x; p is the Fourier sine of rx x; t with

q R

1

respect to t, which is define by Ux; p p2 0 rx x; t sinp tdt, and the

q R

1

inverse Fourier sine of rx x; t is given by rx x; t p2 0 Ux; p sinp tdp.

To apply the Fourier sine transform, with respect to t, in Eq. 4.75, one can use

the following formulas [19]:

2 r

o 2 2

Fs ux; t; t; p p Fs fux; t; t; pg p ux; 0:

ot2 p

r 4:76

at 2 p

Fs fe ; t; pg ; a [ 0:

p a2 p2

By applying the initial condition for Eq. 4.75, one can get

r

o2 2 2 2 2 Expc x x

U x; p A p U x; p B p 4:77

ox2 p c 2 p2

For the boundary conditions, two cases are considered.

Case 1 (stress-free boundary condition):

The thermal stress at the surface is set to zero. The relevant boundary conditions

are:

rx 0; t 0; rx 1; t 0 4:78

Now, applying the Fourier sine transform for Eq. 4.77 with respect to x, by

requiring that, for a physically meaningful system, rx x; t and its first derivatives

in x vanish as x ! 1, as well as using the formulas [19]:

138 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

r

o2 2 2 o

Fc 2 ux; t; x; s s Fc fux; t; x; sg u0; t:

p ox

2 r

o 2

Fs 2

ux; t; x; s s2 Fs fux; t; x; sg s u0; t:

ox p

r 4:79

ax 2 a

Fc fe ; x; sg ; a [ 0:

p a2 s 2

r

2 s

Fs feax ; x; sg ; a [ 0:

p a2 s 2

The resulting equation is:

r

2 2 1

sU 0; p s2 V s; p A2 p2 V s; p B2 p s 2

p p c p c2 x2 s2

2

4:80

where s is Fourier sine variable and V s; p is the Fourier sine of U x; p with

q R

1

respect to x; which is define by V s; p p2 0 U x; p sins xdx, and the

q R

1

inverse Fourier sine of rx x; t is given by U x; p p2 0 V s; p sins xds:

q R

2 1

p 0 rx x; t sinp tdt gives:

U0; p 0 4:81

So, Eq. 4.80 reduces to:

2 2 ps

V s; p B 4:82

p c2 p2 s2 c2 x2 s2 A2 p2

Applying the inverse Fourier sine transform for Eq. 4.82 with respect to x, and

using tables of Fourier sine transform [19, 20], one can obtain:

4.3 Thermal Stress Field: Consideration of Surface and Volumetric Sources 139

r Z 1

2

U x; p V s; p sins xds

p 0

r Z 1

2 B2 p 2 s

sins xds

p c p p 0 s c x s2 A2 p2

2 2 2 2 2

r Z 1

2 B2 p 2 s s

sins xds

p c2 p2 c2 x2 A2 p2 p 0 s2 A2 p2 s2 c2 x2

2 B2 p s s

F s ; s; x Fs ; s; x

p c2 p2 c2 x2 A2 p2 s2 A2 p2 s2 c2 x2

r

2 B2 pcosA p x exp( c x x

p 2 2

A2 p2 c2 p2 c x2A

Therefore:

r

2 B2 p cosA p x exp( c x x

U x; p 4:83

p 2 2

A2 p2 c2 p2 c x2 A

Applying the inverse Fourier sine transform for Eq. 4.83 with respect to t, one can

have:

8 cx

c t

>

> e coshc

> A x e A t coshc x x

>

> ; tAx

>

>

cx

< ec x x e c t e A t

2

rx x; t c2 xB2 A2 cx

>

> e c A x sinhc t e c x x sinh t

>

> A

>

> ; t \ Ax

> cx

: ec x x e c t e A t

4:84

Case 2 (stress gradient-free boundary condition):

The thermal stress gradient at the surface is set to zero. The relevant boundary

conditions are:

orx

0; t 0; rx 1; t 0 4:85

ox

Now, applying the Fourier cosine transform for Eq. 4.77 with respect to x, by

requiring that, for a physically meaningful system, rx x; t and its first derivatives

in x vanish as x ! 1 and the re-arrangement results in:

r

2 o 2 p

U 0; p s2 W s; p A2 p2 W s; p B2 c x 2

p ox p c p2 c2 x2 s2

4:86

140 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

q R

1

respect to x; which is define by W s; p p2 0 U x; p coss xdx, and the

q R

1

inverse Fourier cosine of rx x; t is given by U x; p p2 0 W s; p coss xds:

q R

2 1

p 0 rx x; t sinp tdt gives:

o

U 0; p 0 4:87

ox

So, Eq. 4.86 reduces to:

2 2 p

W s; p B cx 2 4:88

p c p2 s2 c2 x2 s2 A2 p2

Applying the inverse Fourier cosine transform for Eq. 4.88 with respect to x,

one can have:

r

2 B2 A exp( c x x p c x sinA x p

U x; p 4:89

p A3 2 2

p2 c2 p2 c x2 A

Applying the inverse Fourier sine transform for Eq. 4.24 with respect to t, one can

obtain by using Fourier cosine transform [19, 20]. Therefore,

8 x cx

>

> ec t sinh c A x e A t sinhc x x

>

> A

>

> ;tAx

2

> cx

< exp( c x x ec t e A t

B

rx x; t 2 2 x c A x c x

c x A2 >

> e sinhc t e c x x

sinh t

>

> A A

>

> ; t\A x

>

: exp( c x x ec t ecAx t

4:90

Equations 4.84 and 4.90 can be used to compute the stress distribution in accor-

dance with the physical properties.

Equation 6.67 is applicable for the volumetric heat source heating. Since the

irradiated spot size is small, one can assume a semi-infinite substrate material

being subjected to laser heated. The analysis presented below is in line with the

4.3 Thermal Stress Field: Consideration of Surface and Volumetric Sources 141

previous study [3]. The boundary conditions for the problem can be written as

follows:

ohE ohE

0; t 0; 1; t 0

ox ox 4:91

hE x; 1 h ; hL x; 1 h

Recently, Ylbas et al. [17] constructed the closed form solutions for temperature

distribution with volumetric heat sources. The solution of the boundary value

problem

n can be given, owhen:n o

c [ 1 & c \ 1l or c\1 & c [ 1l as follow:

hL x; t h H Expct x x Expx ct

xc

4:92

c

hE x; t h H Expct x x cH Expx ct

xc

q

1c b

where x acc1l 2 and H c2 cal11a :

It should

n be noted that theo solution

n for hL x; t and

o hE x; t in Eq. 4.92 is only

valid for c [ 1 and c\ 1l or c\1 and c [ 1l . In which case, e is of order

102 for metals and c\1for short pulse heating situation.

Now, the equation governing the momentum in one dimensional solid for a

linear elastic case can be written as [18]:

o2 rx 1 o2 rx o2 TL

2

2 2 c2 2 4:93

ox c1 ot ot

q

E

where c1 is the wave speed in the solid, which is c1 q, c2 1m

1m q aTL , m is

Poissons ratio, qis the density of the solid, and aTL is the thermal expansion

coefficient of the solid.

Our aim is solving Eq. 4.93 to find the exact solution for thermal stresses corre-

sponding to the closed form solutions for temperature distribution hL with volumetric

heat sources. So by using the dimensionless variables Eq. 4.64, the dimensionless

2

stress is rx c2 dd2 T0 rx . Equation 4.93 can be re-written after incorporation

Eq. 4.92, in which temperature is presented. Therefore Eq. 4.93 becomes:

o2 rx 2

2 o rx 2 1

A B Expct c x x Expc t x 4:94

ox2 ot2 cx

p

where A c1dc d and B H :

The initial conditions for Eq. 4.94 are as follows:

rx x; 0 0; rx x; 1 0 4:95

142 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

Now, applying the Fourier sine transform for Eq. (4.94) with respect to t, by

requiring that, for a physically meaningful system, rx x; t and its first derivatives

in t vanish as t ! 1, To apply the Fourier sine transform, with respect to t, in

Eq. 4.96. One can use the following formulas [19, 20]:

2 r

o 2 2

Fs 2 ux; t; t; p p Fs fux; t; t; pg p ux; 0:

p

r 4:96

at 2 p

Fs fe ; t; pg ; a [ 0:

p a2 p2

Therefore:

r

o2 2 2

U x; p A p rx x; 0 A2 p2 U x; p

ox2 r p

2 B2 Expc x x c x Expx

p 4:97

p cx c 2 p2

where p is the Fourier sine variable and U x; p is the Fourier sine of rx x; t with

q R

1

respect to t, which is define by Ux; p p2 0 rx x; t sinp tdt, and the inverse

q R

1

Fourier sine of rx x; t is given by rx x; t p2 0 Ux; p sinp tdp:

By applying the initial conditions for Eq. 4.97, one can get:

r

o2 2 2 2 B2 Expc x x c x Expx

2

U x; p A p U x; p p 4:98

ox p cx c 2 p2

For boundary conditions, two cases are considered. These are given below:

Case 1 (stress-free boundary condition):

The thermal stress at the surface is set to zero. The relevant boundary conditions

are:

rx 0; t 0; rx 1; t 0 4:99

Now, applying the Fourier sine transform for Eq. 4.98 with respect to x, by

requiring that, for a physically meaningful system, rx x; t and its first derivatives

in x vanish as x ! 1, Therefore:

r

2

s U 0; p s2 V s; p A2 p2 V s; p

p

4:100

2 B2 p s s

cx 2

p c x p2 c 2 s 2 c 2 x 2 s 1

4.3 Thermal Stress Field: Consideration of Surface and Volumetric Sources 143

q R

1

respect to x; which is define by V s; p p2 0 U x; p sins xdx, and the

q R

1

inverse Fourier sine of rx x; t is given by U x; p p2 0 V s; p sins xds:

q

Applying the boundary conditions by using the formula Ux; p p2

R1

0 rx x; t sinp tdt, it gives:

U0; p 0 4:101

So, Eq. 4.100 reduces to:

2 B2 p s s

V s; p c x 4:102

p c x p2 c2 s2 A2 p2 s2 c2 x2 s2 1

Applying the inverse Fourier sine transform [19, 20] for Eq. 4.107 with respect

to x, one can have:

r Z 1

2

U x; p V s; p sins xds

p 0

r Z 1

2B2 p 2 1 s s

c x sins xds

p c x p2 c2 p 0 s2 A2 p2 s2 c2 x2 s2 1

r Z 1

2B2 p 2 s s

sins xds

p c x p2 c2 c2 x2 A2 p2 p 0 s2 A2 p2 s2 c2 x2

r Z 1

2B2 p 2 s s

sins xds

p p2 c2 1 A2 p2 p 0 s2 A2 p2 s2 1

2B2 p s s

F s ; s; x F s ; s; x

p c x p2 c2 c2 x2 A2 p2 s2 A2 p2 s2 c2 x 2

2

2B p s s

F s ; s; x F s ; s; x

p p2 c2 1 A2 p2 s2 A2 p2 s2 1

r r

2

2 B pcosA p x exp( c x x 2 2 pcosA p x exp( x

B

p c x A2 p2 c2 p2 c2 x2 2 p A2 p2 c2 p2 A12

A

4:103

or

r

2 B2 p cosA p x exp( c x x

U x; p

p c x A2 p2 c2 p2 c2 x2 2

A

r

2 2 p cosA p x exp( x

B

4:104

p A2 p2 c2 p2 A12

Applying the inverse Fourier sine transform [20, 21] for Eq. 4.104 with respect to

t, it yields:

144 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

r r Z 1 0 1

2 B2 2 @ 1 p cos A p x exp( c x x p cos A p x exp( x

rx x; t

A sinp tdp

p A2 p 0 cx p2 c2 p2 cAx2

2 2

p2 c2 p2 A12

r 0 1 r 0 1

2 B2 p cos A x p 2 B2

p

Fs @ ; p; tA e c x x

Fs @ ; p; tA

p A2 c x p2 c2 p2 cAx2

2 2

p A2 c x p2 c2 p2 cAx2

2 2

r ! r !

2 B2 p cosA x p 2 B2 x p

Fs

; p; t e Fs

; p; t

p A2 p2 c2 p2 A12 p A2 p2 c2 p2 A12

4:105

where

0 1 0 8 91

< =

p 1 p p

Fs @ ; p; tA 2 2 @Fs ; p; t Fs ; p; t A

c 2 x2

p2 c2 p2 A2 c x 2 2 2

p c : p2 c 2 x2

;

A2 c A2

r r !

1 p c t cx

p 2 ec t e A t

cx

2 2 e e A t A

c x

2 c

2 2 2 c2 x2 A2

A

4:106

and

! ( )!

p 1 p p

Fs

; p; t

1 Fs ; p; t Fs

2 1 ; p; t

p2 c2 p2 A12 A2 c

2 p2 c 2 p A2

r r !

1 p ct 1

p 2 ect eAt

1

t

1 e e A A

A2 c

2 2 2 1 A 2 c2

4:107

and

0 1 0 8 91

<p cosA x p =

p cosA x p 1 p cosA x p

Fs @ A

; p; t 2 2 @

Fs ; p; t Fs ; p; t A

2 2 2

p c p A2c 2 x2 c x 2 p 2 c2 : 2 c2 x 2 ;

A2 c p A2

8 p p cx

> p c t p At

> 2e cosh c A x 2e

>

<

coshc x x; t [ A x

1 pp e2 c t e2cAxt

2 2 ; tAx

c2 >

c x 2 2

>

A2 : ppec A x sinhc t pp ec x x sinh

c x t; t \ A x

>

2 2 A

4:108

4.3 Thermal Stress Field: Consideration of Surface and Volumetric Sources 145

and

! ( )!

p cosA x p 1 p cosA x p p cosA x p

Fs

; p; t

1 Fs ; p; t Fs

2 ; p; t

p2 c2 p2 A12 A2 c

2 p2 c 2 p A12

8 p p 1

p ct

>

> 2e coshcAx

> p2eAt coshx; t [ A x

1 < pp e2ct e 2 A1 t

1

2 > 2 2 ; t Ax

c >

: ppecAx sinhct ppex sinh 1 t; t\A x

>

A 2

2 2 A

4:109

Therefore, the solution becomes:

8 B2

c t cx

c x

>

> c3 x x2 A2 e coshc A x e A t coshc x x ec x x ec t e A t

>

>

>

< 1A B2 1

ec t cosh c A x eA t cosh x ex e c t eAt

1

; t Ax

2 c2

rx x; t B2

c A x

c x

>

>

> c3 x x2 A2 e sinhc t ec x x sinh cAx t ec x x ec t e A t

>

>

: 1A B2 1

ec A x sinhc t ex sinh A1 t ex ec t eA t ; t\Ax

2 c2

4:110

Case 2 (stress continuity boundary condition at the surface):

The thermal stress gradient at the surface is set to zero. The relevant boundary

conditions are:

orx

0; t 0; rx 1; t 0 4:111

ox

Now, applying the Fourier cosine transform for Eq. 4.98 with respect to x, by

requiring that, for a physically meaningful system, rx x; t and its first derivatives

in x vanish as x ! 1, as well as using the formulas in appendix, results in:

r

2 o

U 0; p s2 W s; p A2 p2 W s; p

p ox

2 2 p 1 1

B 4:112

p p2 c 2 s 2 c 2 x 2 s 2 1

where s is Fourier cosine variable and W s; p is the Fourier cosine of U x; p with

q R

1

respect to x; which is define by W s; p p2 0 U x; p coss xdx, and the

q R

1

inverse Fourier cosine of rx x; t is given by U x; p p2 0 W s; p coss xds:

q R

2 1

p 0 rx x; t sinp tdt gives:

146 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

o

U 0; p 0 4:113

ox

So, Eq. 4.112 will reduce to:

2 2 p 1 1

W s; p B 4:114

p p2 c 2 s 2 A2 p2 s 2 c 2 x 2 s 2 1

Applying the inverse Fouriercos transform [20, 21] for Eq. 4.114 with respect

to x, it yields:

r Z 1

2

U x; p W s; p coss xds

p 0

r Z 1

2 2 p 2 1 1 1

B 2 coss xds

p p c2 p 0 s2 A2 p2 s2 c2 x2 s2 1

r Z 1

2 2 p 2 1 1

B 2 coss xds

p p c2 c2 x2 A2 p2 p 0 s2 A2 p2 s2 c2 x2

r Z 1

2 p 2 1 1

B2 2 coss xds

p p c2 1 A2 p2 p 0 s2 A2 p2 s2 1

2 p 1 1

B2 2 F c ; s; x F c ; s; x

p p c2 c2 x2 A2 p2 s 2 A2 p 2 s2 c2 x2

2 p 1 1

B2 2 F c ; s; x F c ; s; x

p p c2 1 A2 p2 s2 A2 p2 s2 1

r

2 B2 A exp( c x x p c x sinA x p

p c x A3 2 2

p2 c2 p2 c Ax2

r

2 B2 A exp( x p sinA x p

p A3 p2 c2 p2 A12

Therefore,

r

2 B2 A exp( c x x p c x sinA x p

U x; p

p c x A3 2 2

p2 c2 p2 c Ax2

r

2 B2 A exp( x p sinA x p

4:115

p A3 p2 c2 p2 A12

Applying the inverse Fourier sine transform for Eq. 4.115 with respect to t, it

yields:

4.3 Thermal Stress Field: Consideration of Surface and Volumetric Sources 147

r r Z 1 0 1

2 B2 2 @ 1 A exp( c x x p c x sin A x p A exp( x p sin A x p A sinp tdp

rx x; t

p A3 p 0 cx 2 2

p2 c2 p2 c Ax2 p2 c2 p2 A12

r 0 1 r 0 1

2 B2 @ p A 2 B Fs @

2

sinA x p A

exp( c x x Fs ; p; t ; p; t

p A2 c x p2 c2 p2 cAx2

2 2

p A3 p2 c2 p2 cAx2

2 2

r ! r !

2 B2 p 2 B2 sinA x p

2

exp( x Fs

1

; p; t 3

Fs

1

; p; t

pA 2 2 2

p c p A2 pA 2 2 2

p c p A2

4:116

or

8 x

> B2 cx cx

>

>

> 3 2 2

ec t sinh c A x e A t sinh c x x exp( c x x ec t e A t

>

> c x x A A

>

> ;tAx

>

>

> B2 1 c t

>

> e

1 1

sinh c A x eA t sinh x exp( x e c t e A t

< 2 2

1 A c A c

rx x; t x c x

>

> B2 cx

>

> ec A x sinhc t ec x x sinh t exp( c x x ec t e A t

>

> c3 xx2 A2 A A

>

>

> ; t\A x

>

>

> B2 1 c A x 1

>

: e sinhc t e x sinh t exp( x ec t e A t

1

1 A 2 c2 A c A

4:117

Equations 4.11 and 4.117 can be used to compute the stress fields in the substrate

material subjected to the laser heating pulse.

laser short-pulse heating can be written in line with the previous study [4];

therefore, the heat equation in two-dimensional cylindrical coordinate system

yields:

o k2 1 o o2 o2 oTL o

1 ss 2 2 C L 1 sp

ot f r or or oy ot ot

2 2 4:118

1o o o

k TL f d Tr I t expdjy j

r or or 2 oy2

where ss is the electronphonon characteristic time (ss CE =G), G is the elec-

tronphonon coupling factor, k is the mean free path of the electrons, f is the

fraction of excess energy change, CL and CE are the lattice and electron heat

capacities, respectively, k is the thermal conductivity, sp is the electron mean free

time between electronphonon coupling, I t I expd t where I is the

laser peak power intensity, expd t is the temporal distribution function of

laser pulse, d is the absorption coefficient, Tr 1 rf where rf is surface

reflectivity, r is the distance along r-axis, y is the distance along y -axis, t is the

148 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

time variable and TL is the lattice site temperatures. Introducing the following

equalities and dimensionless variables:

fk CE

k2 ; ss

G G

4:119

TL T t k d2 CE

hL ; x r d; y y d; t ; a ;

T CE =G G CL

to Eq. 4.118 yields finally:

3 2

o hL o3 hL 1 o2 hL o hL o2 hL 1 ohL o2 hL ohL

a 1 l 2 a

ox2 ot oy2 ot x oxot ox2 oy2 x ox ot2 ot

b Tr expy expc t

4:120

s G

where l Cp L ; b f I dTl

oG

c

; c d GCE :

This model is the improved energy transport equation including ballistic effects

with the presence of the source term in dimensionless form.

By using Maple program, one can find that Eq. 4.120 admits the following Lie

point symmetries:

1o o o 1o o

X1 hL : X2 : XF Fx; y; t 4:121

c ot ohL oy c ot ohL

where Fx; y; t is a solution for the following homogenous equation:

3 2

o hL o3 hL 1 o2 hL o hL o2 hL 1 ohL o2 hL ohL

a 1 l 2

2

a 2

2

2 0

ox ot oy ot x oxot ox oy x ox ot ot

4:122

And since the invariants of X1 are:

x; y; ec t hL 4:123

Then the corresponding similarity solution can be given as:

This similarity solution reduces the boundary conditions, so the boundary value

problems in both of the following two cases can be reduced.

Case 1: Volumetric source:

Now, consider a semi-infinite cylindrical nano-sized wire subjected to a laser

short-pulse heating (Fig. 4.2).

The boundary conditions for the problem can be written as follows:

4.4 Thermal Stress Field: Two-Dimensional Consideration 149

of the irradiated element and

the coordinate system

x=L

Symmetry Axis

oTL

0; y ; t 0 L L; y

; t T

or

oTL

r; 0; t 0 TL r; 1; t T 4:125

oy

oTL

r; y ; 1 0 TL r; y ; 1 T

ot

Using the dimensionless variables Eq. 4.109 yields the dimensionless boundary

conditions:

ohL

0; y; t 0 hL L d; y; t 0

ox

ohL

x; 0; t 0 hL x; 1; t 0 4:126

oy

ohL

x; y; 1 0 hL x; y; 1 0

ot

The similarity solution hL ux; y ec t transforms Eq. 4.120 to the second

order partial differential equation (PDE):

o2 u o2 u 1 ou

A2 ux; y B2 Tr ey 4:127

ox2 oy2 x ox

p p p

c 1c b

where A r ; B r and r a l 1c and transforms the

dimensionless boundary conditions to the following boundary conditions:

150 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

ou

0; y 0; uL d; y 0

ox 4:128

ou

x; 0 0; ux; 1 0

oy

Now, applying the Fourier cosine transform for Eq. 4.127 with respect to y, by

requiring that, for a physically meaningful system, ux; y and its first derivatives

in y vanish as y ! 1, which results in:

r r

o2

2 2

2 ou 1o 2 B2 Tr

U x; s s A U x; s x; 0 U x; s

ox2 p oy x ox p s 2 1

4:129

where s is Fourier cosine variable and U x; s is the Fourier cosine of ux; y with

q R

1

respect to y; which is define by U x; s p2 0 ux; y coss ydy, and the inverse

q R

1

Fourier cosine of U x; s is given by ux; y p2 0 U x; s coss yds.

Applying the boundary conditions on y; Eq. 4.128, it gives:

r

o2 1o

2 2

2 B2 Tr

2

U x; s U x; s s A U x; s 4:130

ox x ox p s 2 1

p

The transformation z s2 A2 xtransforms Eq. 4.130 to the inhomogeneous

modified Bessel equation:

r

o2 1o 2 B2 Tr

U x; s U z; s U z; s 4:131

oz2 z oz p s2 1s2 A2

It should be noted that Eq. 4.131 is the inhomogeneous

n modified oBessel

equation for real value of A: i.e. when c [ 1 & c \ 1l or

n o

c\1 & c [ 1l

. In which case, e is of order 102 for metals and c\1:

So, the general solution of Eq. 4.131 can be given as:

p

U x; s F1 sBesselI 0; s2 A2 x

p r

2 B2 Tr

4:132

F2 sBesselK 0; s2 A2 x

p s2 1s2 A2

where BesselI and BesselK are the modified Bessel functions of the first and

second kinds, respectively.

Applying the Fourier cosine to the boundary conditions on x, Eq. 4.128, it

gives:

oU

0; s 0 U L d; s 0 4:133

ox

4.4 Thermal Stress Field: Two-Dimensional Consideration 151

r

p !

2 B2 Tr BesselI 0; s2 A2 x

U x; s 1

p 4:134

p s 2 1 s 2 A2 BesselI 0; L d s2 A2

Then the final solution is the inverse Fourier cosine transform to U x; s, i.e.:

r Z

c t 2 c t 1

hL x; y; t ux; y e e U x; s coss yds 4:135

p 0

It should be noted that U x; s doesnt have singularity and although the range of

integration is from 0 to 1, the improper integral is convergent and can be evaluated

numerically at given x; y by using adaptive GaussKronrod quadrature [21]. This

method is included in MATLAB, where the function quadgk(fun,a,b,0 RelTol0 ,1e-

3,0 AbsTol0 ,1e-3) attempts to approximate the integral of a scalar-valued function,

fun, from a to b using high-order global adaptive quadrature and take both of

relative error tolerance and absolute error tolerance as 10-3. Moreover, the function

of quadgk attempts to satisfy that error bound \= max(AbsTol,RelTol*|Q|) so

that the results have an excellent accuracy.

Also, since the value of the modified Bessel function of the first kind exceeds

the range of MATLAB environment for large s. One can avoid any undefined

numerical results by scaling the modified Bessel function of the first kind using the

function besseli(nu,z,1),this function is included in MATLAB, where the function

besseli(nu,z,1),scales the modified Bessel function of the first kind, besseli(nu,z),

by exp(-abs(real(z))).

Case 2: Surface source:

In the absence of the volumetric source, we get that b 0. Hence, the Eq. 4.122

reduces to the following equation:

3 2

o hL o3 hL 1 o2 hL o hL o2 hL 1 ohL o2 hL ohL

a 1 l 2

2 a 2

2 2 0

ox ot oy ot x oxot ox x ox ot ot

4:136

Now, assume a semi-infinite substrate material heated with a time decaying source

from the surface (Fig. 4.2). The boundary conditions for the problem can be

written as follows:

oTL

0; y ; t 0 TL L; y ; t T

or

oTL I Tr

r; 0; t expd t TL r; 1; t T 4:137

oy k

oTL

r; y ; 1 0 TL r; y ; 1 T

ot

152 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

Using the dimensionless variables Eq. 4.119 yields the dimensionless boundary

conditions

ohL

ox 0; y; t 0 hL L d; y; t 0

ohL

oy x; 0; t I Tr expc t hL x; 1; t 0 4:138

ohL

ot x; y; 1 0 hL x; y; 1 0

where the dimensionless source amplitude I is related to the dimensional one I

through the relation:

I I

4:139

k d T

The similarity solution hL ux; y ec t transforms Eq. 4.136 to the second order

partial differential equation (PDE):

o2 u o2 u 1 ou

A2 ux; y 0 4:140

ox2 oy2 x ox

p p

c 1c

where A r and r a l 1c and transforms the dimension-

less boundary conditions to the following boundary conditions:

ou

0; y 0 uL d; y 0

ox

4:141

ou

x; 0 I Tr ux; 1 0

oy

Now, applying the Fourier cosine transform for Eq. 4.140 with respect to y, by

requiring that, for a physically meaningful system, ux; y and its first derivatives

in y vanish as y ! 1, as well as using the formulas in appendix, results in:

r

o2

2 2

2 ou 1o

U x; s s A U x; s x; 0 U x; s 0 4:142

ox2 p oy x ox

where s is Fourier cosine variable and U x; s is the Fourier cosine of ux; y with

q R

1

respect to y; which is define by U x; s p2 0 ux; y coss ydy, and the inverse

q R

1

Fourier cosine of U x; s is given by ux; y p2 0 U x; s coss yds:

Applying the boundary conditions on y; Eq. 4.138, gives:

r

o2 1o

2 2

2

2

U x; s U x; s s A U x; s I Tr 4:143

ox x ox p

p

Similarly, the transformation z s2 A2 x transforms Eq. 4.143 to the inho-

mogeneous modified Bessel equation:

4.4 Thermal Stress Field: Two-Dimensional Consideration 153

r

o2 1o 2 I Tr

U x; s U z; s U z; s 4:144

oz2 z oz p s 2 A2

So, the general solution of Eq. 4.144 can be given as:

p p

U x; s F1 sBesselI 0; s2 A2 x F2 sBesselK 0; s2 A2 x

r

2 I Tr

p s2 A2

4:145

where BesselI and BesselK are the modified Bessel functions of the first and

second kinds, respectively.

Applying the Fourier cosine to the boundary conditions on x; Eq. 4.138, gives:

oU

0; s 0 U L d; s 0 4:146

ox

Therefore, the solution of Eq. 4.144 can be given finally written as:

r

p !

2 I Tr BesselI 0; s2 A2 x

U x; s 1

p 4:147

p s2 A 2 BesselI 0; L d s2 A2

Then, the final solution is the inverse Fourier cosine transform to U x; s, i.e.:

r Z

c t 2 c t 1

hL x; y; t ux; y e e U x; s coss yds 4:148

p 0

Eq. 4.148, is convergent and it can be evaluated numerically at given x; y.

Therefore, Eqs. 4.117 and 4.148 are used to compute the stress field in the

substrate material subjected to the laser heating pulse.

The findings of the stress equations are presented under the appropriate sub-

heading and in line with the previous studies [24]. Two different stress boundary

conditions are incorporated to account for the stress continuity at the surface and

the stress free boundary conditions. In addition, the surface and volumetric heat

sources are considered in the analysis. The surface heat source assumes the

deposition of laser energy at the surface of the irradiated material while the vol-

umetric heat source considers the absorption of the incident irradiated energy in

the absorption depth of the material according to the Beer Lamberts law.

154 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

Two different boundary conditions are incorporated in the thermal stress analysis,

which are stress free boundary at the surface and stress continuity at the surface.

The stress free boundary condition represents the free expansions of the surface

while stress continuity corresponds to the presence of optically transparent film at

the surface. The discussions will be presented in line with the previous study [2].

Table 4.1 gives the parameters used in the simulations.

Figure 4.3 shows dimensionless lattice site temperature distribution inside the

substrate material for different dimensionless heating times. Lattice temperature

increases along the x-axis due to energy gain from the excited electrons through

the collisional process. Since electrons undergo several collisions with increasing

distance, lattice site temperature increases along the x-axis accordingly. As the

heating period increases further, lattice site temperature increase becomes almost

steady inside the substrate material. Temperature gradient is high in the early

heating period due to high rate of energy gain of lattice site from electrons through

the collisional process in the early heating period. Temperature gradient reduces

gradually along the x-axis as the heating time progresses.

Case I: Stress free boundary

Figure 4.4 shows temporal variation of dimensionless thermal stress at different

dimensionless locations inside the substrate material. Temporal behavior of ther-

mal stress is in the wave form and it propagates at a constant speed into the

substrate material. The maximum amplitude of the thermal stress increases slightly

with progressing time, which is attributed to the temperature gradient developed in

the substrate material, which modifies the stress field in the substrate material. The

rise of stress amplitude is higher than its decay in the stress wave. This indicates

that stress wave dies gradually with progressing time while it propagates into the

substrate material. In addition, the thermal stress generated is compressive in the

surface region, i.e. stress value is negative. As the heating period progresses, it

becomes tensile, i.e. stress value becomes positive. The compressive stress initi-

ation at some depth below the surface x [ 0 reveals that material undergoes

contractions due to sudden thermal expansion of the surface. As the heating period

progresses, temperature field inside the substrate increases resulting thermal

expansion of the substrate material where it is initially compressed by the thermal

expansion of the surface. Consequently, compressive stress replaces with tensile

stress in the surface region with progressing time. The magnitude of compressive

stress reduces with progressing time, which is attributed to the high temperature

gradient (dT/dx) dtjdx developed in the solid bulk with progressing time.

Figure 4.5 shows dimensionless stress distribution inside the substrate material

for different dimensionless heating periods. Thermal stress is tensile in the surface

region and it becomes compressive as the distance from the surface increases

towards the solid bulk. The tensile behavior of thermal stress is attributed to

thermal expansion of the surface, since stress free boundary condition is in con-

sidered at the surface. As the heating progresses, the region of tensile stress

4.5 Findings and Discussions 155

the solution of temperature

filed of gold CL 2:8

106 J m3 K1

CE 2:1

104 J m3 K1

G 2:6

1016 W m3 K1

k 315 W m1 K1

d 109 m1

d 0:03

1013 s1

sp 0:024

1012 s

T 300 K

h 1

I 1013 W/m2

E 207

109 Pa

q 7930 kg/m3

aTL 16

106 K1

m 0.3

lattice site temperature

distribution inside the

substrate material for

different dimensionless

heating periods

extends further inside the substrate material. This is because of the extension of

heated region inside the substrate material with progressing time. Therefore,

thermally expanded region extends further inside the substrate material resulting in

tensile stress in this region. The maximum magnitude of dimensionless tensile

stress is on the order of 30, which is about 20 MPa in a dimensional form. The

maximum tensile stress is less than the yielding limit of the substrate material and

crack does not form at the surface vicinity. On the other hand, some distance

below the surface, material cannot expands freely which results in the formation of

compressive stress in this region as observed from Fig. 4.5. The magnitude of

tensile stress increases with progressing time due to the increasing temperature

gradient below the surface. This also suppresses the compressive stress magnitude

inside the substrate material due to extension of the heated region below the

surface.

156 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

of dimensionless stress at

different dimensionless

locations inside the substrate

material for free stress at the

surface case (Case 1)

stress distribution inside the

substrate material for

different dimensionless

heating periods and for free

stress at the surface case

(Case 1)

Figure 4.6 shows temporal variation of dimensionless thermal stress at different

dimensionless locations inside the substrate material. The magnitude of thermal

stress wave increases sharply below the surface and it decays gradually with

progressing time. The time corresponding to the maximum stress becomes longer

as the depth below the surface increases. This is because of the propagation of the

stress wave; in which case, time taken for the wave to reach at some depth below

the surface becomes long. The consideration of stress continuity at the surface

modifies the temporal stress distribution inside the substrate material resulting in

decreasing peak stress in the substrate material. The stress continuity condition at

the surface results in compressive stress waves inside the workpiece. In this case,

the surface of the workpiece is not free to expand during the heating process

because of the presence of stress boundary at the surface.

4.5 Findings and Discussions 157

of dimensionless stress at

different dimensionless

locations inside the substrate

material for stress continuity

at the surface case (Case 2)

Figure 4.7 shows dimensionless stress distribution inside the substrate material

for different dimensionless times. Thermal stress remains high in the region close

to the surface and as the depth below the surface increases; it forms a peak first

and, then, decays gradually with increasing distance towards the solid bulk. The

attainment of high thermal stress levels in the surface region is associated with

the presence of stress boundary at the surface; in which case, a region close to the

surface cannot expand freely during the heating process. When comparing

Figs. 4.4 and 4.7, it can be observed that the behavior of the stress wave due to

stress free surface condition and stress continuity at the surface is different. In this

case, tensile stress region totally replaces with the compressive stress with high

amplitude for the stress continuity at the surface case. In addition, the rise and fall

of the stress waves also differ significantly. The maximum magnitude of dimen-

sionless compressive stress is on the order of 80, which is about 50 MPa in a

stress distribution inside the

substrate material for

different dimensionless

heating periods and for stress

continuity at the surface case

(Case 2)

158 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

dimensional form. Consequently, the maximum compressive stress is less than the

yielding limit of the substrate material and crack does not initiate at the surface

vicinity.

The discussions for the findings are presented in line with the previous studies [3].

The properties used in the simulations are similar to those given in Table 4.1.

Figure 4.8 shows dimensionless lattice site temperature distribution along the

substrate material for different heating periods. Lattice site temperature increases

with increasing depth below the surface, which is more pronounced with

increasing heating periods. This is associated with electron excess energy transfer

to the lattice site through the collisional process with progressing time. In this case,

the number of electron lattice site collisions increases with progressing time

resulting in energy gain of the lattice site. Since the electron excess energy transfer

to lattice is considerably small during a single collision, successive collisions with

progressing time enhances lattice site temperature increase inside the substrate

material. In addition, the number of collisions increases with increasing depth;

which in turn elevates lattice site temperature rise with increasing depth below the

surface. Lattice temperature increase is sharp in the surface region in the early

heating period and it becomes gradual as the depth below the surface increases.

This is associated with electron energy transfer through the collisional process,

which increases with increasing depth below the surface [1]. As the heating period

increases, lattice temperature along the depth below the surface becomes gradual.

Figure 4.9 shows dimensionless thermal stress developed inside the substrate

material for different heating periods for stress free surface boundary condition.

Thermal stress is tensile in the surface region for all heating periods and it

becomes compressive at same depth below the surface. The tensile stress behavior

is attributed to the free expansion of the surface during the heating period.

lattice temperature 1.000000

distribution for different

heating durations

0.999999

t=10

L

0.999998 t=20

t=30

0.999997

0.999996

0 50 100 150 200 250 300

x

4.5 Findings and Discussions 159

stress distribution inside the

workpiece for stress free t=10

2.0E-03

boundary condition at the t=20

surface t=30

0.0E+00

-2.0E-03

-4.0E-03

-6.0E-03

0 50 100 150 200 250 300

x

the surface; in this case, material in this region is not free to expand and free

expansion of the surface results in contraction in this region. However, as the

heating period progresses, temperature gradient in the surface region reduces

causing less thermally induced strain in the surface region. This lowers the

maximum tensile stress magnitude in the surface region. The wave nature of

heating results in wave nature of thermal stress fields in the irradiated region. As

the thermal stress reduces at the surface region, compressive stress magnitude

increases below the surface. Moreover, the depth of the tensile stress region

increases below the surface as the heating period progresses. In the case of stress

continuity at the surface (Fig. 4.10), stress wave is modified in the surface region

such that peak stress reduces with progressing time. Moreover, stress field inside

the substrate material becomes compressive for all heating periods; which is

because of the constraint introduced at the surface. The peak stress moves further

into the substrate material with progressing time.

variation of dimensionless

stress distribution at different

2.0E-03

locations inside the

workpiece for stress free

boundary condition at the 0.0E+00

surface

-2.0E-03

x=100

x=200

-4.0E-03 x=300

-6.0E-03

0 30 60 90 120 150

t

160 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

Figure 4.11 shows temporal behavior of the thermal stress at different dimen-

sionless depths below the surface. Thermal stress is compressive at dimensionless

depths x 100 in the early heating period. However, as the heating period pro-

gresses it becomes tensile in these regions. This is attributed to wave nature of

thermal stress, which propagates into the substrate material with progressing time.

The attainment of the high compressive stress is balanced with low magnitude

tensile stress with progressing time at a certain location inside the substrate

material. When comparing the time occurrence of the peak stress, it is evident that

thermal stress wave propagates at a constant speed inside the substrate material.

This is because of the assumption of constant acoustic wave speed in the stress

analysis. In the case of stress continuity at the surface Fig. 4.12, thermal stress

remains compressive inside the substrate material and peak stress decays with

increasing depth below the surface. In this case, temporal distribution of stress

becomes more extended over the time as compared to that corresponding to close

to the surface.

absorption of the laser irradiated energy. The findings are discussed in line with the

previous study [4]. Tables 4.2 and 4.3 gives the properties used in the simulations.

Figure 4.13 shows dimensionless temperature distribution along the x-axis for

different dimensionless time. It should be noted that the y-axis location is y = 0,

which corresponds to the free surface of the irradiated element. Temperature

decays along the x-axis towards the outer radius of the wire due to the consid-

eration of low temperature boundary at the outer edge of the wire. However,

temperature decay in the central region of the wire x 0:025 is gradual and it

becomes sharp as the distance increases towards the wire edge. In this case, energy

variation of dimensionless

stress distribution at different

locations inside the -2.0E-03

workpiece for stress

continuity boundary

condition at the surface

-4.0E-03

x=100

-6.0E-03 x=200

x=300

-8.0E-03

0 30 60 90 120 150

t

4.5 Findings and Discussions 161

stress distribution inside the

workpiece for stress t=10

-2.0E-03

condition at the surface t=30

-4.0E-03

-6.0E-03

-8.0E-03

0 50 100 150 200 250 300

x

absorbed in the surface region of the workpiece enhances the internal energy gain

at the substrate material in the irradiated region. Consequently, radial heat transfer

along the x-axis is much less than that of along the y-axis due to the absorption,

which takes place along the y-axis. Moreover, the presence of low temperature

boundary at the wire edge acts like a heat sink while lowering temperature sharply

towards the nano-sized wire edge. This is more pronounced for silver, then, fol-

lows chromium, and copper. Since temperature rise for chromium and copper

along the x-axis is much lower than that of silver, temperature decay at central and

towards of the wire edge appears to be gradual. As the heating period progresses,

temperature reduces because of the intensity decay with progressing time (expo-

nential heating pulse). When comparing temperature distribution due to silver,

chromium and copper, it can be seen that the maximum temperature occurs for

silver, then follows chromium, and copper.

d

107 CL

106 k G

1016 CE

104 ss

1012

1=m J =m3 K W =m K W =m3 K J =m3 K s

Silver 7.1 1.5 35 14.4 1.1664 0.081

Chromium 6.7 3.3 94 42.4 5.8088 0.137

Copper 6.7 3.43 386 26 4.0612 0.1562

the simulations

d 5

1012 s1

sp 6

1012 s

T 300 K

I 1:5

1015 W=m2

f 1

rf 0:99

L 25

109 m

162 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

temperature variation along

the x-axis for three materials

and three heating periods

Figure 4.14 shows dimensionless lattice temperature inside the nano-wire for

different dimensionless time. The x-axis location is at the center of the wire.

Temperature decay at the surface vicinity of the wire is gradual and it becomes

sharp in the region next to the wurface vicinity. Since the volumetric heat source is

considered in the analysis, energy absorbed by the substrate material within the

absorption depth is responsible for the gradual decay of temperature in the surface

vicinity. In this case, energy absorbed from the irradiated field increases internal

energy gain of the material while increasing temperature in the irradiated region.

Since, the absorption takes place according to the Lamberts Beer law, absorbed

4.5 Findings and Discussions 163

temperature variation along

the y-axis for three materials

and three heating periods

power reduces with increasing depth below the surface. This, in turn, gives rise to

sharp decay of temperature in the region next to the surface vicinity. Since the

laser pulse intensity reduces with time (exponential pulse), temperature inside the

nano-wire reduces due to low energy being absorbed in the nano-wire with pro-

gressing time. As the depth below the surface increases, temperature decay

becomes gradual. It should be noted that increasing temperature gradient enhances

the diffusional energy transport from the surface region to the solid bulk. However,

this may not occur substantially during the short time period. The maximum

temperature is higher for silver, then, follows chromium, and copper.

164 4 Analytical Treatment of Hyperbolic Equations for Stress Analysis

variation of dimensionless

surface temperature variation

for three materials. The x, y-

axes locations are at x = 0

and y = 0

distribution at the nano-wire center for three different substrate materials. The rate

of temperature decay follows almost the rate of laser pulse intensity decay, par-

ticularly for silver. This is associated with the short-heating period. Electrons gain

energy from the irradiated field and transfer their excess energy to the lattice site

through the collisional process. Consequently, depending on the rate of excess

energy gain by electrons from the irradiated field and number of collisions during

the short-period of heating, lattice site temperature increases accordingly.

Although, the number of collisions between electrons and lattice site increases

with progressing time, energy absorbed by electrons from the irradiated field

reduces-because of time decay of the laser pulse, which is in the exponential decay

form. This, in turn, does not affect notably the rate of lattice site temperature decay

with progressing time. Temperature attains high values for silver, then, follows

chromium, and copper.

References

1. B.S. Yilbas, Improved formulation of electron kinetic theory approach for laser short-pulse

heating. Int. J. Heat Mass Transf. 49(1314), 22272238 (2006)

2. B.S. Yilbas, A.Y. Al-Dweik, Non-equilibrium heating and thermal stress development.

J. Thermophys. Heat Transf. (2012) (in print)

3. B.S. Yilbas, A.Y. Al-Dweik, Closed form solutions for thermal stress field due to non-

equilibrium heating during laser short-pulse irradiation. Physica B 407(12), 21692175

(2012)

4. B.S. Yilbas, A.Y. Al-Dweik, Laser short pulse heating of metal nano-wires. Physica B (2012)

(in print)

5. B.S. Yilbas, S.Z. Shuja, Laser short-pulse heating of surfaces. J. Phys. D Appl. Phys. 32,

19471954 (1999)

6. C.L. Tien, J.H. Lienhard, Statistical Thermodynamics (Hemisphere, Washington DC, 1979)

References 165

7. B.S. Yilbas, A.F.M. Arif, Material response to thermal loading due to short pulse laser

heating. Int. J. Heat Mass Transf. 44, 37873798 (2001)

8. H.E. Elsayed-Ali, M.A. Norris, M.A. Pessot, G.A. Mourou, Time-resolved observation of

electron-phonon relaxation in copper. Phys. Rev. Lett. 58, 12121215 (1987)

9. G. Chen, Ballistic-diffusive heat-conduction equations. Phys. Rev. Lett. 86, 22972300

(2001)

10. G.L. Eesley, Generation of nonequilibrium electron and lattice temperatures in copper by

picosecond laser pulses. Phys. Rev. B 33, 21442151 (1986)

11. L.B. Loeb, The Kinetic Theory of Gases (Dover, New York, 1961)

12. M.I. Kaganov, I.M. Lifshiftz, L.V. Tanatarov, Relaxation between electrons and crystalline

lattice. Soviet Phys. JETP 4, 173178 (1957)

13. S.D. Brorson, A. Kazeroonian, J.S. Moodera, D.W. Face, T.K. Cheng, E.P. Ippen, M.S.

Dresselhaus, G. Dresselhaus G., Femtosecond room-temperature measurement of the

electron-phonon coupling constant k in metallic superconductors. Phys. Rev. Lett. 64,

21722175 (1990)

14. M. Honner, J. Kunes, On the wave diffusion and parallel nonequilibrium heat conduction.

ASME J. Heat Transf. 121, 702707 (1999)

15. T.Q. Qiu, L. Tien, Femtosecond laser heating of multi-layer metals-I analysis. Int. J. Heat

Mass Transf. 37, 27892797 (1994)

16. S.D. Brorson, J.G. Fujimoto, E.P. Ippen, Femtosecond electron heat-transport dynamics in

thin gold film. Phys. Rev. Lett. 59, 19621965 (1987)

17. B.S. Yilbas, A.Y. Al-Dweik, Exact solution for temperature field due to non-equilibrium

heating of solid substrate. Physica B 406(23), 45234528 (2011)

18. A. Kovalenko, Thermoelasticity (Basic Theory and Applications) (Wolters-Noordhoff,

Publishing, Groningen, 1969)

19. L. Debnath, D. Bhatta, Integral Transforms and Their Applications, 2nd edn. (Chapman &

Hall/CRC, New York, 2007)

20. H. Bateman, Tables of Integral Transforms, vol. 1 (McGraw-Hill, New York, 1954)

21. L.F. Shampine, Vectorized adaptive quadrature in MATLAB. J. Comput. Appl. Math. 211,

131140 (2008)

Chapter 5

Concluding Remarks

laser and workpiece materials. The main laser parameters include the pulse length

and the laser power intensity while the important material properties are the

thermal conductivity, absorption depth, elastic modules, thermal expansion coef-

ficient, and Poissons ratio. Depending upon the duration of the laser pulse, the

heating process can be classified into two categories, which are equilibrium and

non-equilibrium heating situations. In the case of equilibrium heating, duration is

longer than the thermalization time of the substrate material and Fourier heating

law governs the heating process. However, laser pulse durations comparable and

shorter than the thermalization time of the substrate material, wave behavior of

stress field takes place and the hyperbolic nature of the governing equations are

used to account for the wave behavior. Consequently, care must be taken to

formulate the heating and thermal stress problems in line with the physical aspects

and scale of the problem. The conclusions derived from the body of this book are

presented according to the following sub-headings and in line with the previous

studies [118].

The maximum laser pulse duration is limited with the thermalization time of the

substrate material for the equilibrium heating situation. In this case, volumetric

and surface heat sources can be incorporated in the analysis to resemble the laser

heating process. Since the laser pulse profile has significant effect on the thermal

stress field, step input and exponential lasers pulses should be considered to

identify this effect. In addition, insulated and convection boundary conditions at

the surface together with stress free surface and stress continuity at the surface are

important concerns influencing the thermal stress behavior in the irradiated region.

Therefore, in line with the previous studies [18], the followings are concluded.

and Thermal Stress Analysis, Materials Forming, Machining and Tribology,

DOI: 10.1007/978-3-319-00086-2_5, Springer International Publishing Switzerland 2014

168 5 Concluding Remarks

In the analysis, stress free surface and zero stress gradient at the surface conditions

are considered. Temperature and stress fields are computed for two cases sepa-

rately. In general, it is observed that stress level increases considerably at some

depth below the surface. This is more pronounced as the heating period progresses.

The behavior of thermal stress inside the substrate material differs considerably as

the boundary condition at the surface changes, i.e. stress free surface and zero

stress gradient at the surface conditions result in different stress behavior inside the

substrate material. In the case of stress free surface condition, the stress wave

propagates towards the solid bulk of the substrate material, since at the surface

stress is kept at zero. The stress level increases as it propagates inside the substrate

material and the stress level enhances as the heating period progresses. In the case

of zero stress gradient at the surface condition, stress is compressive in the surface

vicinity of the substrate material and it becomes tensile as the distance from the

surface increases. The stress level reaches maximum at some depth below the

surface. The magnitude of stress reduces sharply to almost zero as the depth inside

the substrate material increases further. The point of sharp decay of stress level

varies with heating time while the location of maximum stress inside the substrate

material remains almost constant. Temporal behavior of the thermal stress inside

the substrate material differs considerably as the boundary condition at the surface

changes. In the case of stress free surface, stress peaks occur inside the substrate

material as the stress wave propagates. The stress level attains low values in the

early heating period for the zero stress gradient at the surface. However, as the

heating period progresses, stress level increases rapidly, provided that the mag-

nitude of stress level exceeds its counterpart corresponding to stress free condition.

Consequently, the boundary condition at the surface modifies the stress wave

propagation and its level inside the substrate material. Temporal variation of

maximum stress inside the substrate material differs for two boundary conditions

employed in the present study. In the case of zero stress gradient at the surface

condition, maximum stress attains low values in the early heating period and it

increases rapidly as the heating period progresses. The maximum stress follows

almost an exponential increase with time. However, maximum stress behavior is

almost parabolic with time for stress free surface condition. In this case, maximum

stress increases rapidly in the early heating period and the rate of rise of maximum

stress becomes less as the heating period progresses.

Volumetric heat source is considered to resemble the absorption of the laser pulse

during the heating process. The conclusions for various boundary conditions are

given below.

5.1 Equilibrium Heating 169

It is found that the stress wave propagates into the substrate material, and the stress

level behind the wave front is tensile and after the wave front it is compressive.

The magnitude of maximum stress level varies inside the substrate material.

Temperature rises rapidly in the early heating period and the rate of temperature

rise reduces as the heating period progresses. In this case, internal energy gain

dominates over the conduction losses due to high temperature gradient. Thermal

stress wave propagates into the substrate material at a speed of c1. The magnitude

of stress wave reduces at different locations inside the substrate material as the

heating period progresses. This is because of the magnitude of temperature gra-

dient corresponding to different heating periods, since pulse intensity varies

exponentially with time.

during the surface ablation, time exponential stress distribution is considered at the

free surface of the solid substrate. The stress free boundary condition results in low

magnitude of stress waves and the stress boundary at the surface modifies the

shape of the stress wave developed inside the substrate material. In this case, the

stress curve becomes similar to that introduced at the surface. The magnitude of

stress wave reduces as the depth below the surface increases towards the solid

bulk. The time corresponding to the stress peak inside the substrate material differs

for stress free boundary and stress boundary at the surface cases. This is because of

the temporal distributions of the laser pulse intensity and stress at the surface,

which differ considerably.

The zero stress gradient at the surface is considered to incorporate the stress

developed during the laser heating of surfaces initially coated or interfaced with

other substrate surface. Temperature rise due to internal energy gain is important

in the early heating period and as the heating period progresses, conduction energy

transfer from the surface vicinity results in gradual temperature rise inside the

substrate material. This is due to the temporal behavior of the pulse power, which

reduces exponentially with progressing heating period. Stress level in the surface

region attains considerably high values. The stress wave is compressive in nature

and propagates with a wave speed c1. Moreover, the rise of the stress wave is

modified by the stress boundary condition at the surface.

170 5 Concluding Remarks

cooling effect of the assisting gas during the heating process. The influence of the

heat transfer coefficient on temperature profiles is significant as the dimensionless

heat transfer coefficient at the surface increases to 0.0202. The temperature gra-

dient is reduced to its minimum at some point below the surface. At a depth

beyond the point of minimum temperature gradient the diffusional energy transport

dominates over the gain in internal energy of the substrate from the irradiated area.

The point of minimum temperature gradient changes for high heat transfer coef-

ficient of h 0:0202. Moreover, the heat transfer coefficient influences the

temperature gradient in the surface region, which is significant for h 0:0202.

The thermal stress developed in the vicinity of the surface is tensile and as the

depth increases it becomes compressive. This is because the thermal strain

developed in the vicinity of the surface, which is positive, and at some point below

the surface, it becomes negative due to the compressive effect of the substrate. The

thermal stress wave is generated within the substrate material for a heat transfer

coefficient of 0.0202. The magnitude of stress wave is reduced as the depth

increases from the surface towards the bulk solid.

Entropy generation due to temperature field decays sharply in the surface region,

which is due to the behavior of temperature gradient and energy storage in this

region. The location of minimum entropy generation due to temperature field

inside the substrate material is in the same region of the temperature equilibrium as

described earlier. The location of minimum entropy generation moves away from

the surface as the heating period progresses. Entropy generation due to stress field

shows a cyclic behavior provided that the maximum entropy occurs at the location

where the stress is the maximum inside the substrate material. The cyclic behavior

of the entropy is because of the propagation of the stress waves. Entropy gener-

ation due to temperature field dominates over its counterpart corresponding to the

stress field in the surface region. As the depth from the surface increases towards

the solid bulk, entropy generation due to stress field becomes important; in which

case, entropy peaks appear on the entropy curve. In the case of stress field, the

negative sign of the entropy generation is because of the compression component

of the stress waves and, in all cases, the entropy generation is positive.

The conclusions derived from the solution of Cattaneo heat equation and the

thermal stress are given below in line with the previous studies [913].

5.2 Cattaneo Heating Model and Thermal Stresses 171

It is observed that temperature rise in the early heating period is gradual. As the

heating period progresses, temperature rises rapidly reaching its maximum, which

is more pronounced at the surface. The time occurrence of the peak temperature

changes as the distance below the surface increases. Temperature decay rate

becomes small when the pulse intensity reduces 5 % of its peak value. In this case,

temporal gradient of temperature becomes small and the wave nature of the

heating replaces with the diffusional heating. Temperature decay is sharp inside

the substrate material during the heating cycle. This results in large temperature

gradients and stress field below the surface. Temporal behavior of thermal stress

reveals that the compressive stress waves are formed due to initial contraction of

the surface during the early heating period. The compression wave reaches its peak

value rapidly and decays gradually similar to the pulse intensity distribution. The

stress wave generated propagates at a constant speed inside the substrate material.

The thermal expansion of the surface during the late heating period results in the

tensile wave formation in the surface region. This appears as a tensile tail in the

compressive wave generated earlier. The magnitude of the tensile wave is sig-

nificantly lower than that of the compressive wave.

Temperature decay is sharp inside the substrate material during the heating cycle.

This results in large temperature gradients and stress field below the surface.

Temporal behavior of thermal stress reveals that the compressive stress waves are

formed due to initial contraction of the surface during the early heating period. The

compression wave reaches its peak value rapidly and decays gradually similar to the

pulse intensity distribution. The stress wave generated propagates at a constant

speed inside the substrate material. The thermal expansion of the surface during the

late heating period results in the tensile wave formation in the surface region. This

appears as a tensile tail in the compressive wave generated earlier. The magnitude

of the tensile wave is significantly lower than that of the compressive wave.

The high rate of increase in temperature is attributed to the internal energy gain of

the substrate material from the heat source and small amount of energy transfer

by diffusion from the surface vicinity to the solid bulk in the early heating period.

172 5 Concluding Remarks

In this case, qT/qt, and q2T/qt2 attain high values and varies sharply with progressing

time. Consequently, the heat diffusion is not governed by the classical Fourier law in

the early heating period. This situation is also observed onset of the initiation of the

cooling period for which temperature decays rapidly with progressing time. How-

ever, as the heating progresses, the rate of temperature rise becomes gradual and the

term q2T/qt2 becomes small in Cattaneo equation. The heat diffusion is governed

mainly by the Fourier law. This is also true during the long cooling periods. Thermal

stress developed in the surface vicinity is in the form of a stress wave which

propagates into the substrate material with a constant speed. Moreover, the stress

wave is tensile in the heating cycle due to the thermal expansion of the surface while

it is compressive in the cooling cycle because of the thermal contraction of the

surface region during the cooling period. The thermal wave generated has a tail with

decaying amplitude and it extends over the heated region.

Temperature and stress distribution are demonstrated for a step input laser short-

pulse heating of metallic surfaces. It is found that the temperature rise towards the

laser mid-pulse-length is rapid, which results in large temporal gradient of tem-

o2 T

perature (oT

ot and ot2 ). In this case, wave nature of heat transfer dominates and

governs the heat transfer in the substrate material. This is true during the heating

and in the early cooling periods, except at some depth below the surface; in which

case, temporal variation of temperature is not significantly high and resulting in

diffusional heat conduction. This situation is also observed during the late cooling

periods, i.e. wave nature of the heat transfer is replaced by the diffusional con-

duction governed by Fourier law. Thermal stress generated during the heating

period is compressive and the stress wave amplitude is influenced by the volu-

metric heat source in the early heating period. However, stress wave becomes

tensile in the cooling period. This is attributed to the expansion of the irradiated

surface during the cooling period.

Entropy generation rate due to heat transfer and thermal stress field is computed

for the time exponentially varying laser short-pulse. The volumetric heat source

resembling the laser irradiated energy is incorporated in the analysis. Stress free

boundary condition is considered to account for the free expansion of the surface

during the laser heating pulse in the analytical solution. It is observed that tem-

perature gradient remains low in the surface vicinity because of the internal energy

gain of the substrate material during the heating pulse. The maximum value of

thermal stress increases with time, which is attributed to the formation of high

temperature gradient with progressing time. Thermal stress is tensile in the surface

5.2 Cattaneo Heating Model and Thermal Stresses 173

region and becomes compressive with increasing depth below the surface. This is

associated with free expansion of the surface. As the heating progresses, the depth

of tensile stress extends inside the substrate material. Entropy generation rate,

during the heating pulse attains low values in the surface region and increases with

increasing depth below the surface, which is true for all heating periods. The

attainment of low entropy generation rate is associated with the term 1/T in the

entropy equation, which is low in the surface region. Entropy generation rate due

to heat transfer is significantly higher than that of due to thermal stress developed

for all heating periods.

Non-equilibrium energy transfer takes place during the laser-short pulse heating

process. Thermal separation results in thermally communicating two sub-systems

co-existed in the irradiated region. Although the temperature gradient is low in the

lattice sub-system, high thermal stress field is developed due to short-heating

duration. The findings due to non-equilibrium heating situation are given below in

line with the previous findings [1418] under the appropriate sub-headings.

Two boundary conditions are introduced for thermal stress analysis. These include

stress free boundary and stress continuity at the surface conditions. Stress free

boundary condition assumes free expansions of the surface during the heating

process while stress continuity boundary condition incorporates the presence of

optically transparent coating at the surface. It is found that lattice temperature

increases along the x-axis because of electron excess energy transfer through the

collisional process. The temperature gradient remains high in the surface region

during the initial heating period, which in turn results in high magnitude of thermal

stress wave generation in the surface region. The thermal stress wave behavior

differs significantly for different stress boundary conditions at the surface. Stress

free boundary condition causes free expansion of the surface resulting in tensile

thermal stress in the surface region. As the distance from the surface increases,

tensile behavior replaces with compressive behavior. The maximum magnitude of

compressive stress increases with increasing depth below the surface. In the case

of the stress continuity boundary condition, the amplitude of thermal stress

remains high in the surface region and thermal stress becomes compressive inside

substrate material for all heating periods. Stress wave decays gradually with

progressing distance along the x-axis towards the solid bulk.

174 5 Concluding Remarks

Two boundary conditions for thermal stress were introduced, namely stress free

surface and stress continuity at the surface. It is found that lattice site temperature

rise is sharp in the early heating period and it becomes gradual as the heating

period progresses. This is attributed to the collisional energy transfer from elec-

trons to lattice site during the heating period. Thermal stress developed in the

surface region is tensile while it becomes compressive at same depth below the

surface for stress free boundary condition. This is attributed to free expansion of

the surface. Thermal stress wave propagates into the substrate material at a con-

stant speed and thermal stress behavior in the surface region is modified due to

stress continuity condition at the surface; in which case, the peak stress reduces

with progressing time. In the case of stress continuity at the surface, thermal stress

becomes compressive inside the substrate material for all heating periods.

solution of temperature field in the irradiated region is formulated after considering

two-dimensional axisymmetric heating situation. It is found that temperature decay

inside the substrate material is gradual in the surface vicinity, which is associated

with the internal energy gain from the irradiated field in this region. Temperature

decay is sharp in the region next to the surface vicinity. Although high temperature

gradient enhances heat diffusion from surface region to the solid bulk, this is not

substantiated due to the short heating duration. Temperature decays sharply towards

the edge of the nano-wire because of the low temperature boundary condition at the

wire edges. Temporal variation of temperature at the irradiated spot center follows

almost the laser pulse intensity variation with time. This is attributed to energy gain

by electrons from the irradiated field; in which case, electrons transfer their excess

energy to lattice site through the collisional process.

References

1. M. Kalyon, B.S. Yilbas, Analytical solution for thermal stresses during laser pulse heating

process. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 215, 14291445 (2001)

2. B.S. Yilbas, M. Kalyon, Repetitive laser pulse heating with a convective boundary condition

at the surface. J. Phys. D Appl. Phys. 34, 222231 (2001)

3. B.S. Yilbas, N. Ageeli, Thermal stress development due to laser step input pulse intensity

heating. J. Therm. Stresses 29(8), 721751 (2006)

4. B.S. Yilbas, N. Ageeli, M. Kalyon, Laser induced thermal stresses in solids: Exponentially

time decaying pulse case. Lasers Eng. 14(1), 81101 (2004)

References 175

5. B.S. Yilbas, N. Al-Ageeli, Formulation of laser induced thermal stresses: Stress boundary at

the surface. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 217, 423434 (2003)

6. B.S. Yilbas, N. Ageeli, Thermal stresses due to exponentially decaying laser pulse and a

convection boundary at the surface. Lasers Eng. 16, 235265 (2006)

7. B.S. Yilbas, M. Kalyon, Analytical approach for entropy generation during a laser pulse

heating process. AIChE J. 52, 19411950 (2006)

8. B.S. Yilbas, Entropy analysis due to temperature and stress fields in the solid irradiated by a

time exponentially varying laser pulse. Heat Transf. Eng. J. 26(8), 8089 (2005)

9. H. Al-Qahtani, B.S. Yilbas, The closed form solutions for Cattaneo and stress equations due

to step input pulse heating. Phys. B 405(18), 38693874 (2010)

10. H. Al-Qahtani, B.S. Yilbas, Temperature and stress fields for short pulse heating of solids.

J. Thermophys. Heat Transf. 25(1), 173176 (2011)

11. H. Al-Qahtani, B.S. Yilbas, Closed form solution of Cattaneo equation including volumetric

source in relation to laser short-pulse heating. Can. J. Phys. 89(7), 761767 (2011)

12. H. Al-Qahtani, A closed form solution of temperature and stress fields for laser short-pulse

heating of a solid: Exponentially decaying volumetric source. Lasers Eng. 22(12), 109124

(2011)

13. H. Al-Qahtani, B.S. Yilbas, Entropy generation rate during laser short-pulse heating:

Contribution of heat transfer and thermal stress. Lasers Eng., in print (2012)

14. B.S. Yilbas, Improved formulation of electron kinetic theory approach for laser short-pulse

heating. Int. J. Heat Mass Transf. 49(1314), 22272238 (2006)

15. B.S. Yilbas, A.Y. Al-Dweik, Non-equilibrium heating and thermal stress development.

J. Thermophys. Heat Transf., in print (2012)

16. B.S. Yilbas, A.Y. Al-Dweik, Closed form solutions for thermal stress field due to non-

equilibrium heating during laser short-pulse irradiation. Phys. B 407(12), 21692175 (2012)

17. B.S. Yilbas, A.Y. Al-Dweik, Laser short pulse heating of metal nano-wires. Phys. B, in print

(2012)

18. B.S. Yilbas, S.Z. Shuja, Laser short-pulse heating of surfaces. J. Phys. D Appl. Phys. 32,

19471954 (1999)

Index

A E

Adaptive Gauss-Kronrod quadrature, 151 Elastic module, 66

Analytical solution, Cattaneo equation. See Electron-phonon collision, 122, 126, 132, 134

Cattaneos equation Energy transport, microscopic level, 122

Analytical treatment, hyperbolic equations. conservation of energy, 127, 128

See Energy transport, microscopic level electron distribution, 123, 124

electron excess energy transfer, 126

electron kinetic theory approach, 122

B electron mean free path, 134

Beer Lamberts law, 68, 74, 95, 134, 153 electron movement in surface

Boltzmanns constant, 134 region, 123f

electron phonon coupling

factor, 133, 134

C electrons, collision, 125

Cattaneo heat equation, 82, 96 electrons, without collision, 125

See also Cattaneos equation, 000 energy balance, 126

Cattaneo heating model and thermal Fourier integral transformation, 129

stresses, 81, 170 Fourier inversion, 130

exponential laser pulse intensity, 171 Fourier transformation, 130

step input laser pulse intensity, 171, 172 kinetic theory approach, 132

Cattaneos equation, 81, 82 rectangle function, 124

surface heat source consideration, 8695, simple kinetic theory, 128

110114 total electron energy, 127

volumetric source consideration, 95110, Entropy analysis, thermal stress field, 6467,

114119 108110

dimensionless entropy generation, 80f, 81f,

82f, 83f, 84f

D dimensionless stress distribution, 80f

Debye temperature, 134 dimensionless temperature gradient, 81f

Dimensionless form, 4850, 50, 51, 5153, 53, dimensionless thermal stress, 83f

54 elastic module, 66

Dimensionless stress field equation, 89, 94 entropy generation, 64, 65

Dimensionless temperature distribution, 8, 28, Gouy-Stodola theorem, 66

38, 56, 73, 75f, 78, 79f, 116f, 160, 162f non-dimensional form, 66

Dirac delta function, 48, 131 Poissons ratio, 66

and Thermal Stress Analysis, Materials Forming, Machining and Tribology,

DOI: 10.1007/978-3-319-00086-2, Springer International Publishing Switzerland 2014

178 Index

step input pulse, findings, 7981 Lamberts law, 72, 79, 162

strain, in terms of displacement, 66 Laplace transformation, 7, 9, 19, 21, 27, 29,

thermal expansion coefficient, 66 36, 38, 39, 55, 60, 104

time exponentially decaying pulse, find- See also Inverse Laplace transform

ings, 8284 boundary conditions, 87, 97

Equilibrium heating, 167 respect to time, 9, 38, 39, 57

exponential laser pulse intensity, 168170 temperature, 9, 21, 57

step input laser pulse intensity, 168 of Term1, 41

Exponential laser pulse intensity, 168, 171 of Term2, 44

convection boundary condition at surface, of Term22, 46

170 thermal stress, 9, 21, 38, 39, 57

entropy analysis, 170 t-shifting rule of, 107

exponential stress distribution at surface, Laser heating process, 13, 86, 108

169 conduction heating, 2

stress continuity at surface, 169 Fourier heating law, 2

stress free surface consideration, 169 one-dimensional modeling, 6

surface heat source considerations, 171 short-pulse heating, 3

volumetric heat source considerations, 171 solid state heating, 1

Lasers, 1

F

Fourier heat transfer equation, 7, 26, 36, 54 M

Fourier heating law, 1, 5, 167, 170, 172 Metallic substrates at microscopic level, for-

Fourier heating model, 85, 117 mulation of energy transport. See

Fourier cosine transform, 137, 139, 140, 145, Energy transport, microscopic level

150, 152 Modified Bessel functions, 88, 93, 97, 109,

Fourier sine transform, 137, 138, 140 150, 151

Fourier transformation, 130

N

G Non-equilibrium energy transport, 2, 3, 121

Gouy-Stodola theorem, 66 Non-equilibrium heating, 3, 85, 173

surface heat source consideration, 173

two-dimensional heating, 174

H volumetric heat source consideration, 174

Hyperbolic equations. See Energy transport,

microscopic level

P

Parabolic heating model, 131134

I Partial differential equation (PDE)

Inverse Fourier cosine transform, 140, 146, second order, 149, 152

151, 153 third order, 132

Inverse Fourier sine transform Partial fraction, 11, 12, 32, 36, 60, 98, 100,

Inverse Laplace transform, 11, 20, 55, 56 105, 107

of Term1, 11, 12, 24, 33, 4044, 60, 61 Term1, 24

of Term2, 12, 13, 24, 33, 4447, 60, 61 Term2, 12

of Term3, 13, 14, 25, 33, 47, 60, 61 Term3, 14, 25

of Term4, 15, 33, 47, 48, 60, 62 Term4, 15

of Term5, 60, 62 Poissons ratio, 9, 66, 136, 141, 167

K Q

Kinetic theory, 2, 121, 122, 128, 132 Quasi-ballistic approach, 131

Index 179

Short-pulse laser heating, 81 time exponentially varying laser pulse

Step input laser pulse heating, 6 heating, 3654

stress continuity boundary at surface, closed form solution of stress distribution,

1926 48

stress free boundary at surface, 618 dimensionless form, 4850, 50, 51, 5153,

Step input laser pulse intensity, 168 53, 54

entropy analysis, 172, 173 dimensionless stress distribution, 77f, 78f

surface heat source consideration, 171, 172 dimensionless temperature distribution, 38,

volumetric heat source consideration, 172 75f

Step input pulse intensity, 88, 96, 114 dimensionless temperature gradients, 76f

Stress boundary at surface, time exponentially findings, 7378

varying laser pulse heating, 5464 Fourier heat transfer equation, 36

boundary conditions, 54 general solution for stress field, 40

closed form solution of stress inverse Laplace transform, 37, 40, 41

distribution, 62 Laplace inversions of composing terms, 43,

complementary error function, 55, 56 44, 4446, 4648

dimensionless stress distribution, 6264 Laplace transformation, 36, 37

dimensionless temperature distributions, Laplace transformations of terms, 4143

79f Laplace transformation to time, 38, 39

findings, 78, 79 laser pulse in simulations, 75f

general solution for stress field, 58, 59 solving momentum equation, 38

initial conditions, 55 solving stress distribution, 38

inverse Laplace transform, 55 temperature distribution in Laplace

Laplace inversions of terms, 61, 62 domain, 39

Laplace transformation of boundary Stress free boundary at surface, step input laser

condition, 59, 60 pulse heating, 618

Laplace transformation to time, 56, 57 complementary error function, 8

partial fraction, 60 dimensionless laser step input pulses, 69f

solving stress distribution, 56 dimensionless temperature profiles, 69f,

solving stress equation, 56 70f

stress distribution in dimensionless form, dimensionless thermal stress, 70f

62 findings, 6770

stress due to recoil pressure, 56 inverse Laplace transform, 8, 9

Stress continuity boundary at surface, step inverse of Laplace transform of terms,

input laser pulse heating, 1926 1015

dimensionless stress distribution, 71f laser heating pulse, 6

dimensionless stress, temporal variation, laser pulse properties in simulations, 69t

72f pulses in analysis and simulations, 68f

findings, 71 stress distribution, dimensionless form,

heat transfer equation, 19 1618

initial and boundary conditions, 19, 20 stress equation, closed form, 15

inverse Laplace transformation, 20, 21 Stress free boundary at surface, time expo-

inversion of Laplace nentially varying laser pulse heating,

transformation, 2426 2736

Laplace transformation, 19, 20 boundary conditions for temperature, 29

new boundary conditions, 22, 23 closed form solution of stress distribution,

new initial conditions, 21, 22 34

zero stress gradient, 21 closed form solution, Laplace transforma-

Stress distribution, 11, 32, 38 tion method, 27

closed form, 15, 16, 34, 48 complementary error function, 28

dimensionless form, 1618, 26, 28, 34, 48, dimensionless temperature, temporal vari-

54, 62, 71f, 74f, 77f, 78f, 80f, 111, 154, ation, 72f, 74f

156f, 157, 159f, 160f, 161f findings, 7173

180 Index

Stress free boundary at surface, time expo- Thermal stress. See Cattaneo heating model

nentially varying laser pulse heating (cont.) and thermal stresses; Energy transport,

initial conditions for temperature, 29, 30 microscopic level; Entropy analysis,

inverse Laplace transform, 27, 28, 3234 thermal stress field

normalized power intensity distribution, Thermoelasticity, 94

73f Time exponentially varying laser pulse heat-

partial fraction, 32 ing, 26

stress distribution in dimensionless form, stress boundary at surface, 5464

3436 stress free boundary and convection at

stress field, general solution for, 30, 31 surface, 3654

Stress free substrate material, 56, 57 stress free boundary at surface, 2736

Surface heat source consideration, Cattaneos

equation, 86

convolution theorem, 95 V

dimensionless stress distribution, 112f Volumetric source consideration, Cattaneos

dimensionless stress field equation, 94 equation, 95, 96

dimensionless stress versus time, 111f convolution theorem, 102

exponential pulse heating, 9295, 112114 dimensionless entropy generation, 118f

exponential pulse, temporal variation, 93f dimensionless temperature distribution

findings, 110114 versus time, 116f

inverse Laplace transform, 95 dimensionless temperature, temporal vari-

step input pulse heating, 8792, 111 ation, 114f

temperature distribution, 93 dimensionless thermal stress, temporal

variation, 115f

dimensionless thermal stresses distribution

T with time, 117f

Thermal expansion coefficient, 5, 65, 136, 167 entropy generation rate, 117119

Thermal strain, 2, 5, 9, 66, 67, 74, 112, 170 exponential pulse heating, 103108, 116,

Thermal stress field 117

entropy analysis, 6467 Laplace inversion, 97

stress continuity boundary condition at laser step input pulse, temporal variation,

surface, 145147, 156158 96f

stress gradient-free boundary condition, modified Bessel function, 97

139, 140 space-time domain, 97

stress-free boundary condition, 137139, step input pulse heating, 96103, 114, 115

142145, 154156 temperature distribution, 104

surface heat source considerations,

134140, 154158

surface source, 151153 Y

two-dimensional considerations, 147149, Youngs modulus, 65

160164

volumetric heat source considerations,

140147, 158160 Z

volumetric source, 148151 Zero stress gradient, 19, 21, 168, 169

- Solidwork Flow Simulation TutorialDiunggah olehMilad Ah
- Motivation Letter for PhD applicationDiunggah olehSakhawat Hussain
- ConductionDiunggah olehsandyengineer13
- TUP038Diunggah olehParticle Beam Physics Lab
- Streamlight TLR-2 HL G – Lighting Up the NightDiunggah olehTwobirds Flying Publications
- ConductionDiunggah olehkevinjorgeramos
- con pascoDiunggah olehMiguel Angel
- A a ADiunggah olehiamfikri
- Investigation of Thermal Insulation on Ice CoolersDiunggah olehInternational Organization of Scientific Research (IOSR)
- Hmt Kkk Unit 1, 2Diunggah olehSiddharth Garkhail
- Ground Loop ReportDiunggah olehParwez Khan
- Laser PropertiesDiunggah olehtanushrisharma
- Ingolf V. Hertel et al- Surface and bulk ultra-short pulsed laser processing of transparent materialsDiunggah olehPocxa
- Wei Yu et al- Direct acceleration of solid-density plasma bunch by ultraintense laserDiunggah olehItama23
- Full Range CST BergerDiunggah olehManas Mondal
- Highlight4000l Ds FinalDiunggah olehKarthick Nagarajan
- Heat Chap04 081Diunggah olehoddomancan
- 1 (6)Diunggah olehZana Aziz
- Thermal Stresses Induced by a Point Heat Source in a Circular Plate by Quasi-static ApproachDiunggah olehSteven Hall
- Conduction and Heat ExchangersDiunggah olehSrini
- High Power LaserDiunggah olehShahin
- Transport 2 Ch 1Diunggah olehAndrew Horowitz
- Advance Transport Phenomena – Assignment 2Diunggah olehSandeep Challa
- D541e1111_ILEE_cFL-02_Mk2Diunggah olehMichaelBozickovic
- 978-1-61470-887-2_ch1Diunggah olehthefernandoman
- 1-s2.0-S0143974X16302498-mainDiunggah olehRaul Andres
- T-UTC-0941Diunggah olehFaustoVinicioAcuñaCoello
- Hitec RamizDiunggah olehRamiz Kurbegovic
- 2 HT Lecture - Conduction.pdfDiunggah olehAngelo Ignacio
- Yb CaF2 Laser CrystalDiunggah olehLaser Crylink

- ART-Additive Manufacturing and Its Societal Impact-A Literature ReviewDiunggah olehJuan Colombo
- Solid State Chemistry_LSmartDiunggah olehJuan Colombo
- Lean managmente beyond manufacturing.pdfDiunggah olehFidel Roman
- Subdivision Surfaces JPetersDiunggah olehJuan Colombo
- Carbon Nanotubes as Nanodelivery Systems_MChoonDiunggah olehJuan Colombo
- Tribo Fatigue LSosnovskiyDiunggah olehJuan Colombo
- Theory of ElasticityDiunggah olehlin
- Introduction to Quality and Reliability Engineering_R-JiangDiunggah olehJuan Colombo
- System DynamicsDiunggah olehJuan Colombo
- Friction and Wear_GStraffenili.pdfDiunggah olehJuan Colombo
- Ch1.pdfDiunggah olehJuan Colombo

- E-153Diunggah olehDači Kalender
- forcesvirtuallabDiunggah olehapi-249694223
- A3_Unsteady_Heat_Advection+Convection_Physics based ApproachDiunggah olehAravindSomu
- Starting High Inertia LoadsDiunggah olehviswajd
- 08r05310403 Antennas and Wave PropagationDiunggah olehandhracolleges
- 1.4 MaterialsDiunggah olehSharifahHafizahAl-juf
- EMF 2m 6m 10m-FlowerDiunggah olehFLOWER
- Alexander Trofimov Report 2012Diunggah olehvangelis90
- thin airfoil.pptDiunggah olehmgskumar
- Question Paper 3 (Topic 4 - Heat P3)Diunggah olehNur Syidah
- Lee Binary Mixtures Bragg WilliamsDiunggah olehKonkman
- PRO II 8.0 Component and Thermophysical Properties - Reference Manual.pdfDiunggah olehLuis Carlos Ayllón Escobar
- Civil IV II PartDiunggah olehAnonymous 8EYiR2EGQ
- MTech_Thermal_Engineering.pdfDiunggah olehAnonymous mXicTi8hB
- F.E. Walker- A Comparison of the Classical and a Modern Theory of DetonationDiunggah olehGhoree23456
- ejsr_21_4_06AlexanderDiunggah olehioancozma
- AuroraDiunggah olehGigi Gheorghe
- HeatEnginesVol 1 Chapter 2 RSDiunggah olehLissa Hannah
- 109317003-Deflection-of-Curved-Bars.docDiunggah oleh伟伦
- HC Verma Concepts of Physics PDF Exercise Solutions of Chapter 1 n2u_education PhysicsDiunggah olehAnonymous SZIuqImz
- ObjectivesDiunggah olehHia Jing
- ColluMaurizio-HSMV2008Diunggah olehMaurizio
- The First Reactor.Chicago pile 1.CP-1. 40th Anniversary.1982 DOEDiunggah olehjleydenco
- LOSS REDUCTION IN RADIAL DISTRIBUTION SYSTEM BY OPTIMAL PLACEMENT OF CAPACITOR USING DIFFERENTIAL EVOLUTION METHODDiunggah oleheditor3854
- prs_w02d2_qonlyDiunggah olehakirank1
- final exam a review 2014Diunggah olehapi-273978634
- Work-Energy.pdfDiunggah olehHasan Nawaz
- phy1.pdfDiunggah olehraokvkp
- Computational ChemistryDiunggah olehshadow_anchor
- RP_166Diunggah olehmacarronm62016

## Lebih dari sekadar dokumen.

Temukan segala yang ditawarkan Scribd, termasuk buku dan buku audio dari penerbit-penerbit terkemuka.

Batalkan kapan saja.