Proc. of the Third Intl. Conf.

on Advances in Applied Science and Environmental Engineering - ASEE 2015

Copyright Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-055-2 doi: 10.15224/ 978-1-63248-055-2-12

THEORETICAL APPROACH TO STUDY

DISTURBANCES DUE TO MECHANICAL
SOURCE IN A GENERALIZED
THERMOELASTIC DIFFUSIVE HALF
SPACE
K.C. Deshmukh, J. J. Tripathi and G.D. Kedar,
Abstract Disturbances caused by impulsive mechanical
source in a homogeneous, isotropic half-space are studied
within the context of generalized thermoelastic diffusion with
one relaxation time. A two dimensional half space with a
permeating substance in contact with the bounding plane is
considered in axisymmetric distribution. The chemical
potential is assumed to be a known function of time. Integral
transform technique is used to find the analytic solution in the
transform domain by using direct approach. Inversion of
transforms is done employing a numerical scheme.
Mathematical model is prepared for Copper material and
numerical results for temperature, stress, displacement,
chemical potential and concentration are obtained and
illustrated graphically.

theory of thermoelastic diffusion. The theory of Nowacki

uses Ficks law. Sherief et al. [7] introduced the theory of
thermoelastic diffusion in the framework of Lord-Shulman
theory by introducing thermal relaxation time parameter and
diffusion relaxation parameters governing the field
equations. Many researchers [8-13] studied various types of
problems in thermoelastic diffusion. Tripathi et al. [14, 15]
studied problems on generalized thermoelasticity in a semiinfinite solid circular cylinder with one relaxation time and
discussed a problem of generalized thermoelastic diffusion
in a thick circular plate with axisymmetric heat supply.
Elhagary [16] solved a two dimensional generalized
thermoelastic diffusion problem for a half-space subjected to
harmonically varying heating.

Keywords: impulsive, half space, diffusion, generalized,

thermoelastic, relaxation time.

The objective of this work is to study the

effects of impulsive mechanical source on thermoelastic
diffusion interactions in a half space under axisymmetric
distribution within the context of Lord-Shulman theory of
generalized thermoelastic diffusion (TEDLS). The Classical
coupled thermoelastic diffusion theory (TEDCT) is
recovered as a special case. Analytic solutions for
temperature, concentration, chemical potential, displacement
and stresses are obtained in the Laplace transform domain
using direct approach. Numerical inversion of Laplace
transforms are performed using Gaver-Stehfast Algorithm
[17-19] and all integrals were evaluated using Rombergs
integration technique [20] with variable step size. A
mathematical model is prepared for Copper material and
results are discussed along with the graphical representation.

Introduction

Generalized thermoelasticity theories are successful in

removing the paradox of infinite speed of propagation of
thermal signals inherent in the classical coupled
thermoelasticity introduced by Biot [1]. Lord and Shulman
[2] developed a theory modifying the Fourier law of heat
conduction by introducing the heat flux rate and a relaxation
time for the special case of an isotropic body. The heat
equation associated with this theory is of wave type.
Diffusion in thermoelastic solids is a transport
phenomena governed by Ficks law which states that the
passive movement of molecules or particles is along the
concentration gradient. Thermoelastic diffusion involves the
coupling of the fields of temperature, mass diffusion and
strain. It has a wide range of applications in geophysics and
industries. In particular, diffusion is used to form the base
and emitter in bipolar transistors, form integrated resistors,
form the source/drain regions in Metal oxide
semiconductors (MOS) transistors and dope poly-silicon
gates in MOS transistors.
Study of phenomenon of diffusion is used to improve the
conditions of oil extractions and is of great deal of interest
for oil extraction companies. Nowacki [3-6] developed the

I.

Formulation of the Problem

We shall consider a homogeneous isotropic

thermoelastic solid occupying the region z 0 .The z-axis
is taken perpendicular to the bounding plane pointing
inwards. The problem is considered within the context of the
theory of generalized thermoelastic diffusion with one
relaxation time. We shall assume that the initial state of the
medium is quiescent at a temperature T0 . An impulsive
mechanical source is assumed to act at the origin of the
cylindrical co-ordinate system (r , , z ) having isothermal
boundary and the chemical potential is a known function of
time.

K.C. Deshmukh* , G.D. Kedar

Department of Mathematics, R.T.M. Nagpur University, Nagpur
India.
J.J.Tripathi
Dr. Ambedkar College, Deekshabhoomi, Nagpur-440010, Maharashtra.

The problem is thus two-dimensional with all considered

functions depending on the spatial variables r and z as
well as on the time variable t .

India.

57

Proc. of the Third Intl. Conf. on Advances in Applied Science and Environmental Engineering - ASEE 2015
Copyright Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-055-2 doi: 10.15224/ 978-1-63248-055-2-12

where (.) denotes the Dirac delta function , Q 0 is the

magnitude of the force, f (r ) is a known function and
H (t ) is a Heaviside unit step function.

The displacement vector, thus, has the form u (u,0, w) .

For the two dimensional problem, the components of
strain tensors can be written in the form,
err

u
u
w
1 u w
, e , ezz
, erz

r
r
z
2 z r

and

e is the cubical dilatation given by,

u u w 1
ru w

r r z r r
z

(1)

Applying the Laplace and Hankel transform to a

function f (r , z, t ) defined by

f (r , z, s) L f (r , z, t ) e st f (r , z, t )dt

2
2
where 2 1
2
r
r r z 2

r2

u ( )

2 w ( )

C
2u
1
2
2
r
r
r
t

C
2w
1
2
2
z
z
z
t

On taking Laplace and Hankel transform of both sides of

equations (2)-(5), after using equation (8) (supressing the
primes for convenience) , we get,

(2)

D
D

(3)

2
(4)
C E 1 0 div u a 0 C
k 2 0
t 2
t

2
D 2 2 (div u ) Da 2 0 2 C Db 2C 0 (5)
t
t

2 e e 1 0 2 C

(6a)

rr 2 err e 1 0 2C

(6b)

zz 2 ezz e 1 0 2 C

(6c)

rz erz

(6d)

r z 0

(6e)

P 2 e bC a ( 0 )

f * ( , z, s) H f (r , z, s) f (r , z, s) r J 0 ( r )dr

The governing equations are of the form,

2u

Analytic Solution

II.

2 * s 0 s 2 * e * 1C

2 e * 1 D 2 2

3 D 2 2 C * 0

a1

P (r, 0, t ) H (t ) f (r )

, 0r

0r

(15)

2
s 2 s 0 1 1 s 3 s 2 s 1 1

( 3 1) 2 ss 1

a3

s 4 2
s 0 1s 1
( 3 1)

k1

k2

k3

,e*,C * 0

where k1 , k 2 and k 3 are

characteristic equation given by,

The roots

rz (r, 0, t ) 0

a2

(r, 0, t ) 0

(10)

s 0 1 1 1 2 3 1 1
s

( 3 1) 2 s 1 3 s

k 6 a1 k 4 a2 k 2 a3 0

, 0r

(15)

a1 , a 2 , a3 are giving by

The boundary conditions of the problem in dimensionless

form at z 0 are taken as

Q0 (r ) (t )
2r

(14)

Equation (15) can also be written as,

where CE / k is the dimensionless characteristic

length, c1 2 , is the speed of propagation of
isothermal elastic waves.

zz (r , 0, t )

2 s 0 s 2 C *

a1 D 4 a 2 D 2 a3 * , e * , C * 0

where the coefficients

r c1 r , z c1 z , u c1 u , w c1 w , t c12 t,
1 c12 , P P / 2 , ij ij /( 2 ) , 0 c12 0
(8)
1 0 /( 2 ) , C 2C / 2

(9)

(13)

Eliminating the transformed e , C and from (13), (14)

and (15), we obtain the following six order differential
equation

To facilitate the solution, the following dimensionless

variables are introduced

0r

2 s 2 e * D 2 2 * D 2 2 C *

(7)

1
2 p1 sin( p2 ) a1
3

(12)

k2

1
a1 p1
3

k3

1
a1 p1
3

58

roots

of

the
(17)

k1 , k 2 and k3 are given by,

k1

(11)

the

(16)

3 cos( p 2 ) sin( p 2 )

3 cos( p2 ) sin( p 2 )

Proc. of the Third Intl. Conf. on Advances in Applied Science and Environmental Engineering - ASEE 2015
Copyright Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-055-2 doi: 10.15224/ 978-1-63248-055-2-12

where
p1 a1

2a 3 9a1 a 2 27a 3
sin
3a 2 , p 2
and 1

3
2 p3

Applying Laplace transform to eqns. (6a)-(7) and making

use of the solutions given in eqns. (22)-(26), we obtain the
stress components and the chemical potential in the Laplace
transform domain,

The solutions of equation (16) are of the form,

* Ai ( , s) e qi z

(18)

2 r 0

i 1
3

e * Ai ( , s) e qi z

(19)

rr

i 1

C * Ai( , s) e qi z

(20)

i 1

where Ai ; Ai and Ai , i 1, 2, 3 are parameters depending

on and s . Substituting from (18), (19), (20) into (14) and
(15), the parameters Bi ( , s) and Ci ( , s) , i 1, 2, 3 can be

zz

rz

k s s 1 k ,
s s k 1 s
k s s 1 s k s s s

s s k 1 s
4

where f i

2
2

1
2

Ai ( , s) e qi z J 0 ( r ) d
0

i 1

P (r , z, s)

2 q 2

B( , s) e q z

q
2 J r d
1
3 q (1 q )

qi z
i
i i2

e
2
i 1 (q i q )

2
2

A e
0

i 1

qi z

J 0 r d

(23)

A ( , s) 0

(24)

i 1

w (r , z, s)

C ( , s) e qz

J 0 r 3 i q i
qi z d
2
e
i 1 (q q 2 )

(32)

3
2 2
q2
B( , s) 2 i i 2 Q0
2

i 1 ( qi q )

Applying Laplace transform to eqns. (2)(3) and making use

of eqns. (22)(24), the solutions for the displacement
components in the Laplace transform domain as

(31)

Applying Laplace transform and Hankel transform on both

sides of boundary conditions (9)-(12) and using equations
(22)-(31), we get,
3

B( , s) e qz

2
u (r , z, s) J1 r 3
d

0
2 i 2 e qi z
i 1 (qi q )

(30)

i fi 3 di 1 , i 2 2 fi / 2 di 1Ai

(22)

C Ai( , s) e qi z J 0 ( r ) d

0 i 1

(29)

e Ai( , s) e qi z J 0 ( r ) d

0 i 1

q C ( , s) e q z

2
J 0 r d G
i q i

qi z

2
i 1 (q q 2 )

G J 0 r i e qi z d
i

Applying the inversion of Hankel transform to equations

(18), (19) and (20), we get,

1
2

di

(21)

(27)

qz

1
B( , s) e

r
3

2 3
1

2 r
d G
i

qi z

J 0 r i 1 (qi q 2 )

(28)

expressed in terms of Ai ( , s) as

Ai( , s) f i Ai ( , s) , Ai( , s) d i Ai ( , s)

B( , s)e q z

2
J1 r
d G

2 i 2 e qi z
i 1 (qi q )

2 q2

(25)

q (1 qi )
B( , s) 2 i i2
0
2
i 1 ( q i q )

(2 / 2 ) i Ai f * ( )
i 1

(33)

(34)

(35)

Equations (32)-(35) is a system of linear equations

with A1 , A2 , A3 and B as unknown parameters. Solving the
above system of linear equations, the complete solution of
the problem is obtained in the Laplace transform domain.

(26)

where the parameters B( , s) and C ( , s) depend on and

s only.

III.

2
Also, q 2 2 2 s 2 , C ( , s) B( , s) ,
q

Inversion of Double
transforms

The formula for the inverse of the Laplace transform as

obtained by Gaver [17] and Stehfast [18, 19] is given below.

i 1 2 f i 2 1 di Ai

59

Proc. of the Third Intl. Conf. on Advances in Applied Science and Environmental Engineering - ASEE 2015
Copyright Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-055-2 doi: 10.15224/ 978-1-63248-055-2-12

thereafter. As the disturbance travels through the medium, it

encounters sudden changes, resulting in a non uniform
pattern of the curves which shows the effect of coupling of
the fields of temperature, diffusion and strain.

By this method the inverse f (t ) of the Laplace transform

f (s) is approximated by,
K

f (t ) (ln 2 / t ) D( j , K ) F j ln 2 / t

(36)

In figure 2, the concentration C shows an oscillatory

behavior throughout the medium. The values of
concentration fall sharply till r 3 and then gradually
decrease to zero with the increase in radial distance. The
magnitudes of values of concentration for TEDLS theory are
more than TEDCT theory throughout the medium.
Particularly, if we observe the region 3 r 6 , the nonuniformity in the graphs is clearly visible. This can be
attributed to the effect of coupling between the fields of
temperature, diffusion and strain.

j 1

With
D( j, K ) (1) j M

min( j , M )

nm

n M (2n)!
( M n) ! n ! (n 1)! ( j n)! (2n j )!

(37)

where K is an even integer, whose value depends on the

word length of the computer used. M K / 2 and m is the
integer part of the ( j 1) / 2 . The optimal value of K was
chosen as described in Gaver-Stehfast algorithm, for the fast
convergence of results with the desired accuracy. This
method is easy to implement and very accurate for functions
of the type e t . The Romberg numerical integration
technique [20] with variable step size was used to evaluate
the integrals involved. All the programs were made in
mathematical software Matlab.
IV.

Figure 3 exhibits the variation of zz along the radial

direction. One can observe that the variation in values of
zz for TEDLS and TEDCT theory is seen throughout the
medium. The axial stress values are tensile in the medium. A
sharp fall in zz values is seen till r 3 and then is
gradually decreases. It is also observed that the values of
axial stress are more for TEDLS theory than TEDCT theory
up to r 4 and then TEDCT theory predicts higher axial
stress as compared to TEDLS theory. It is also observed that
the axial stresses for TEDLS theory become compressive
after r 8 .

Numerical results and

discussion

The chemical potential is taken as

f 2 (r ) P0 H (r1 r )
where P0 is constant.
On taking Hankel transform, we get,
*
f 2 ( ) (r1 P0 / ) J 1 ( ) .
Mathematical model is prepared with Copper material for
purposes of numerical computations. The material constants
of the problem are thus given in S.I. units [9]:

V.

Conclusion

In this work, the effect of an impulsive mechanical source

on a two dimensional thermoelastic half-space in contact
with a permeating substance was investigated. The method
used
in this study provides quite a successful approach in
T0 293 K , 8954 kg m 3 , 0 0.02 , 0.2 , k 386 JK 1 m 1dealing
s 1 ,
with thermoelastic diffusion problems without any
assumed restriction on the field variables. Coupling of the
t 1.78 10 5 K 1 , C 1.78 10 5 K 1 ,
diffusion field, temperature and strain plays an important
3.86 1010 Nm 2 , 7.76 1010 Nm 2 , c1 4.158 10 3 ms 1 ,
role in the deformation of an elastic body. As the
a 1.2 10 4 m 2 / s 2 k , b 0.9 10 6 m 5 / kg s 2 ,
disturbance travels through the medium, it encounters
D 0.88 10 8 kg s / m 3 , c E 383.1 J .Kg 1 K 1 ,
sudden changes, resulting in a non uniform pattern of the
curves. It was observed that the chemical potential of the
Using these values it was found that, 8886.73 s.m 2
diffusive material attains a steady state. The results of this
1
2
0.0168 Nm.J , 4, 1 5.43 , 2 0.533 , 3 36.24 .
problem are very useful in the two dimensional problems in
It should be noted that a unit of non-dimensional time
axisymmetric half-space which have various geophysical
12
corresponds to 6.5 10 s , while a unit of non-dimensional
and industrial applications.
8
length corresponds to 2.7 10 m . The computations were
carried out for non-dimensional time t 0.05 .

References

Figures 1-3 exhibit the variations of , C and zz with

distance r. The variations of the various components with
distance r are shown a) Solid line for TEDCT theory b)
Dotted line for TEDLS theory. The numerical simulations
are done at the bounding plane i.e. z 0 .

[1]
[2]
[3]

Fig. 1 exhibits the variation of as a function of radius. It

is observed that follows a non-uniform pattern as distance
r increases. TEDLS and TEDCT theories show large
variations throughout the medium. It is seen that for TEDLS
theory, temperature has a positive value at r 0 and then
follows an oscillatory pattern whereas for TEDCT theory,
the values of temperature at r 0 is negative and then it
gradually increases and follows an oscillatory pattern

[4]
[5]
[6]
[7]

60

M. Biot, Thermoelasticity and irreversible thermodynamics, J.

Appl. Phys., vol. 27, pp. 249-253, 1956.
H. Lord, Y. Shulman, A generalized dynamical theory of thermoelasticity, J. Mech. Phys. Solids, vol. 15, pp. 299-309, 1967.
W. Nowacki, Dynamical problems of thermodiffusion in solids I,
Bull. Acad. Pol. Sci. Ser. Sci. Technol., vol. 22, pp.5564, 1974.
W. Nowacki, Dynamical problems of thermodiffusion in solids II,
Bull. Acad. Pol. Sci. Ser. Sci. Technol., vol. 22, pp. 129135, 1974.
W. Nowacki, Dynamical problems of thermodiffusion in solids III,
Bull. Acad. Pol. Sci. Ser. Sci. Technol., vol. 22, pp. 257266, 1974.
W. Nowacki, Dynamical problems of thermo diffusion in elastic
solids, Proc. Vib. Probl., vol. 15, pp. 105128, 1974.
H. H. Sherief, F. A. Hamza, H. A. Saleh, The theory of generalized
thermoelastic diffusion, Int. J. Eng. Sci., vol. 42, pp. 591608, 2004.

Proc. of the Third Intl. Conf. on Advances in Applied Science and Environmental Engineering - ASEE 2015
Copyright Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-055-2 doi: 10.15224/ 978-1-63248-055-2-12
[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]
[18]
[19]
[20]

H. H. Sherief, H. A. Saleh, A half space problem in the theory of

generalized thermoelastic diffusion, Int. J. Solid Struct., vol. 42, pp.
44844493, 2005.
N. M. El-Maghraby, A two-dimensional generalized thermoelasticity
problem for a half-space under the action of a body force, Journal of
Thermal Stresses 31(2008) 557568.
A. M. El-Sayed, A two-dimensional generalized thermoelastic
diffusion problem for a half-space, Math. and Mech. of Solids (2014)
DOI: 1081286514549877.
N. Sharma, P. Ram, R. Kumar, Plane strain deformation in
generalized thermoelastic diffusion, Int. Journal of Thermophy., vol.
29, pp. 1503-1522, 2008.
R. Kumar, T. Kansal, Propagation of lamb waves in transversely
isotropic thermoelastic diffusive plate, Int. J Solids Struct., vol. 45,
pp. 58905913, 2008.
J. N. Sharma,
N. K. Sharma, K. K. Sharma, Diffusion in
Generalized Thermoelastic Solid in an Infinite Body with Cylindrical
Cavity, Pro. Nat. Acad. of Sci., vol. 83(4), pp. 353-364, 2013.
J. J. Tripathi, G. D. Kedar, K. C. Deshmukh, Dynamic Problem of
Generalized Thermoelasticity for a Semi-infinite Cylinder with Heat
Sources, Journal of Thermoelasticity, vol. 2 , pp. 01-08, 2014.
J. J. Tripathi, G. D. Kedar, K. C. Deshmukh, Generalized
thermoelastic diffusion problem in a thick circular plate with
axisymmetric heat supply, Acta Mechanica, 2015 doi:
10.1007/s00707-015-1305-7
M. Elhagary, A two-dimensional generalized thermoelastic diffusion
problem for a half-space subjected to harmonically varying heating,
Acta Mechanica, vol. 224, pp. 47114722, 2013.
D. P. Gaver, Observing Stochastic processes and approximate
transform inversion, Operations Res., vol. 14, pp. 444-459, 1966.
H. Stehfast, Algorithm 368, Numerical inversion of Laplace
transforms, Comm. Assn. Comp. Mach., vol. 13, pp. 47-49, 1970.
H. Stehfast, Remark on algorithm 368, Numerical inversion of
Laplace transforms, Comm. Assn. Comp., vol. 3, pp. 624, 1970.
W. H. Press, B. P. Flannery, , S. A. Teukolsky, W. A. Vetterling,
Numerical Recipes, Cambridge University Press, Cambridge, the art
of scientific computing, 1986.

Fig.2. Concentration

Fig.3. Axial stress

C distribution along the radial direction

zz distribution along the radial direction

About Author (s):

Dr. K. C. Deshmukh is working as a

professor and head Department of
mathematics, R.T.M.Nagpur University,
Nagpur. Having teaching experience of
30 years. Published more than 100
research papers in reputed international
journals. Guided 18 students for Ph.D
degree. Three major research projects
completed and more than 18 years of
research experience

Fig.1.Temperature

Dr. G.D. Kedar is working as an

associate professor in the Department of
mathematics, R.T.M.Nagpur University,
Nagpur. Having teaching experience of
23 years. Published 17 research papers
in reputed international journals.
Guided 1 student for Ph.D degree.

distribution along the radial direction

Mr. J.J. Tripathi is working as a

research scholar in the Department of
mathematics,
R.T.M.
Nagpur
University, Nagpur. Published 2
research papers in reputed international
journals.

61