Alexandria Engineering Journal (2020) 59, 2269–2281

H O S T E D BY
Alexandria University

Alexandria Engineering Journal

www.elsevier.com/locate/aej
www.sciencedirect.com

ORIGINAL ARTICLE

A mathematical study of natural convection flow

through a channel with non-singular kernels: An
application to transport phenomena
Syed Tauseef Saeed a,*, Muhammad Bilal Riaz b,e, Dumitru Baleanu c,f,
Kashif Ali Abro d,e
a
Department of Sciences & Humanities, National University of Computer & Emerging Sciences, 54000 Lahore, Pakistan
b
Department of Mathematics, University of Management and Technology Lahore, Pakistan
c
Department of Mathematics, Cankaya University, Turkey
d
Department of Basic Sciences and Related Studies, Mehran University of Engineering and Technology, Jamshoro, Pakistan
e
Institute for Groundwater Studies (IGS), University of the Free State, South Africa
f
Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan

Received 22 December 2019; revised 7 February 2020; accepted 8 February 2020

Available online 26 February 2020

KEYWORDS Abstract In this manuscript, we have obtained closed form solution using Laplace transform,
Modern fractional operator; inversion algorithm and convolution theorem. The study of mass transfer flow of an incompressible
Inversion algorithm; fluid is carried out near vertical channel. Recently, new classes of differential operators have been
Dufour effect; introduced and recognized to be efficient in capturing processes following the decay law and the
Mass transfer; crossover behaviors. For the study of heat and mass transfer, we applied the newly differential oper-
Convolution; ators say Atangana-Baleanu ðABCÞ and Caputo-Fabrizio ðCFÞ to model such flow. This model for
Convection flow temperature, concentration and velocity gradient is presented in dimensionless form. The obtained
solutions have been plotted for various values physical parameters like a; Df ; Gm ; Gr ; Sc and Pr on
temperature and velocity profile. Our results suggest that for the variation of time the velocity
behavior for CF and ABC are reversible. Finally, an incremental value of prandtl number is
observed for decrease in the velocity field which reflects the control of thickness of momentum
and enlargement of thermal conductivity. Further, dynamical analysis of fluid with memory effect
are efficient for ABC as compared to CF.
Ó 2020 The Authors. Published by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria
University. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/
4.0/).

1. Introduction

* Corresponding author. Heat and mass transfer left a significant impact on non-
E-mail address: [email protected] (S.T. Saeed). Newtonian fluid. The flow of non-Newtonian fluids under
Peer review under responsibility of Faculty of Engineering, Alexandria assumption of heat generation analyzed in different applica-
University. tions of engineering. The engineering processes find numerous
https://doi.org/10.1016/j.aej.2020.02.012
1110-0168 Ó 2020 The Authors. Published by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria University.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
2270 S.T. Saeed et al.

application on heat and mass transfer in a flow fluid across tional time derivative operators. furthermore, Authors [23]
channel such as fire engine, nuclear energy, petroleum reser- investigate the influence of chemical reaction with ramped tem-
voir and transport phenomena. In many systems, the com- perature condition on MHD free convection flow over a verti-
bined effect of temperature and concentration forces often cal plate using non-singular kernel. The solution of MHD
used in modeling of transport phenomena. Furthermore elec- Oldroyd-B fluid with time dependent boundary conditions
tric circuits, heat exchanger, solar panels and thermal protec- using classical model and fractional operators analyzed by
tion system are major application of heat transfer which Riaz et al. [30]. The mass transfer has the impact on rate of
directed impact on fluid system and its geometry [1–8]. heat transfer and mass transfer depends upon concentration
Recently, changes in temperature has significant role in free differences. Imran et al. [24] applied Caputo derivative on dif-
convection flow. Magnetic field effect on infinite plate with ferential type fluid and investigated solutions for temperature,
heat and mass transfer by convection flow discussed by Cham- concentration and velocity. Heat and mass transfer analysis
kha et al. [9]. Further, Chamkha et al. [10] analyzed the lami- for second grade fluid with thermophoresis and thermal radia-
nar and convection flow with heat and concentration field tion analyzed by Das et al. [25]. Ali [35] discussed the novel
inside the rectangular enclosure. In the presence of heat radia- method for a fractional derivative with non-local and non-
tion in MHD convective flow through permeable surface with singular kernel. Furthermore, the solutions of the linear and
ramped wall temperature investigated by Ismail et al. [11]. nonlinear differential equations within the generalized frac-
Influence of heat effect on electrically and chemically fluid tional derivatives is investigated by Ali et al. [36]. Some
under vertical channel investigated by Umavathi et al. [12]. remarkable work regarding heat and mass transfer phenomena
Chamkha et al. [13] presented the micropolar fluid on heated for viscous fluid has been done by researchers [26–29].
vertical preamble surface under effect of chemical and radia- The aim of present article to analyze the closed form solu-
tion parameter. The solution of PDE for heat generation can tion for unsteady free convection under the presence of mass
be obtained by using implicit finite difference schemes. transfer with Dufour effect. CF and ABC modern fractional
Soret and Dufour reaction are two significant circum- operator are used to governs the flow of fluid using partial dif-
stances in the process of heat and mass transfer. Concentration ferential equations ðPDEÞ. LT is used to determine the solu-
and temperature profile generate Dufour and Soret effect tion of given PDE for concentration, temperature and
respectively. Both effects are used to discuss the Fick’s law. velocity profile by satisfying IC0s and BC0s of non-integer
The analytical solution of free convection flow to mass transfer order derivatives. To see the behavior of fluid flow, fractional
in upended channel with consideration of Dufour effect dis- order model is good to explain the dynamics and memory
cussed by Jha et al. [14]. Impact of heat radiation and chemical effect with respect to classical model. Closed form solutions
reaction on viscous incompressible nano-fluid through a investigate in this article are interpreted graphically and com-
preamble surface with Dufour and Soret effect investigated putationally by software Mathcad-15 to study the impact of
by Reddy et al. [15]. Narahari et al. [16] presented the solution different pertinent parameters.
of natural convection flow with ramped wall temperature in
the presence of thermal and mass diffusion between the paral- 2. Fractional formation of natural convection flow through a
lel plates. Furthermore, Author [17] discussed the influence of channel
ramped wall temperature, heat and chemical reaction on fluid
convection flow over infinite plate. He notice that the velocity Free convection and mass transfer fluid flow in a vertical chan-
profile decrease in the case of chemical reaction, inclination of nel are considered with Dufour effect. Let x-axis be the vertical
angle and heat source increase. direction in which initial temperature T1 and C1 are consider
In fluid dynamics, fractional derivatives models have been
as a constant on both plates when t  0. Here, a condition is
analyzed for viscoelastic materials, such as glassy state and
invoked for fluid to be at rest between two bounding walls, this
polymers. Recently, different real life problems have been dis-
is because a Dufour effect with systems of equations shows the
cussed through fractional time derivative operator, namely
influence of temperature and concentration. Meanwhile, at
Caputo-Fabrizio ðCFÞ time fractional derivative and
t ¼ 0þ , concentration and temperature are given as
Atangana-Baleanu ðABCÞ time fractional derivative [18–20]. ðC þC Þt ðT þT Þt
Fractional calculus is an emerging field which is based on dif- C1 þ 0 t0 1 and T1 þ 0 t0 1 . We set the physical quantities
ferent types of kernels. The main significance of kernels is to on spatial y and time t variables between two walls having
provide a better description of the dynamics among complex infinite length. Within such assumptions, governing equations
systems, for instance, collecting the memory at whole and par- under Boussinesq’s approximation with Dufour effect are
tial domain of certain processes. The non-locality of the new derived as:
kernel analyzed the memory structure with alternate scales.
@u @2u
Furthermore, the study of thermal sciences with non-local ¼ t 2 þ gbT  gbT1 þ gbC  gbC1 ; ð2:1Þ
@t @y
exchange plays an important role in non-singular kernel frac-
tional derivative. CF and ABC time fractional operators dis-
2 
2 
@T @ T  @ C
cussed with non-singular kernel [21,32]. Fahd et al. [34] ¼a þ D ; ð2:2Þ
discussed the classification of ordinary differential equations @t @y2 @y2
in the frame of Atangana-Baleanu fractional derivative. Fur-
2 
ther authors [33] investigate the existence theory and numerical @C @ C
¼ Dm ; ð2:3Þ
solutions to smoking model using Caputo-Fabrizio fractional @t @y2
derivative. subject to following employed conditions as defined below
Recently, Riaz et al. [22] investigate the effect of ramped
wall velocity on MHD fractional Oldroyd-B fluid using frac- uðy; 0Þ ¼ 0; Tðy; 0Þ ¼ T1 ; Cðy; 0Þ ¼ C1 ; ð2:4Þ
A mathematical study of natural convection flow through a channel with non-singular kernels 2271

uðy; tÞ ¼ Uo cosðwtÞ; Tðy; tÞ ¼ Tw ; Cðy; tÞ ¼ Cw ; ð2:5Þ 3. Optimal solutions based on modern fractional approaches

uðy; tÞ ! 0; Tðy; tÞ ! T1 ; Cðy; tÞ ! C1 : ð2:6Þ 3.1. Fractional optimality of concentration via CF derivative
In order to avoid lengthiness and exaggerations among gov-
erning equations, we introduce the dimensionless functions Generating Eq. (2.11) for the fractionalized form, we imposed
and parameters in Eqs. (2.1)–(2.6) as illustrated below: Eq. (2.15) on Eq. (2.11), we arrive at
y u t l2 T  T1

y ¼ ; u ¼ ; t ¼ ; t0 ¼ ; h ¼ ; ð2:7Þ @a/ 1 @2C
Tw  T1  ¼ 0; ð3:19Þ
l uo t0 t @ta Sc @y2
C  C1 t t Eq. (3.19) can be manipulated by means of several methods
/¼ ; Sc ¼ ; Pr ¼ : ð2:8Þ
Cw  C1 Dm a but we prefer to employee a powerful technique namely
Laplace transform on Eq. (3.19) with imposed conditions as
The optimal format of set of dimensionless governing equa- discussed in Eqs. (2.12)–(2.14). This mathematical process is
tions have been obtained in terms of non-fractional resulted the suitable expression as
approaches, we have
 2

@u @ 2 u Sc
q
ðy; qÞ ¼ @ /ðy; qÞ :
/ ð3:20Þ
¼ þ Gr h þ Gm /; ð2:9Þ q  aq þ a @y2
@t @y2
The second order partial differential equation say (3.20) can



@h 1 @2h Df @ 2 / have the incomplete solution as investigated in Eq. (3.20)
¼ þ ; ð2:10Þ pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffi
@t Pr @y2 Pr @y2

ðy; qÞ ¼ c1 ey qaqþa
Sc q Sc q
/ þ c2 ey qaqþa : ð3:21Þ


@/ 1 @2C With the help of boundary conditions for concentration, c1
¼ ; ð2:11Þ
@t Sc @y2 and c2 can be traced out in the following equation:
pffiffiffiffiffiffiffiffiffi
with appropriate conditions ey Sc q
qaqþa

ðy; qÞ ¼
/ : ð3:22Þ
uðy; 0Þ ¼ hðy; 0Þ ¼ /ðy; 0Þ ¼ 0; y P 0; ð2:12Þ q

uðy; tÞ ¼ cosðwtÞ; hðy; tÞ ¼ /ðy; tÞ ¼ 1; y ¼ 0; ð2:13Þ Inverting an exponential equation through Laplace transform
subject to the an Appendix (A1), we get final solution of con-
uð1; tÞ ¼ hð1; tÞ ¼ /ð1; tÞ ¼ 0; y ! 1: ð2:14Þ centration equation as:
Z  
2cSc 1 sinðysÞ acts2

/ðy; tÞ ¼ 1  e cSc þs2

ds: ð3:23Þ
2.1. Development of governing equations in terms of CF p 0 sðcSc þ s2 Þ
derivative
Here, c is letting parameter can be taken as c ¼ 1a
1
.

Before developing the non-fractional governing equations in 3.2. Fractional Optimality of Temperature via CF derivative
terms of fractional differential operator, we define the time
fractional differential operators with its Laplace transform.
Generating Eq. (2.10) for the fractionalized form, we imposed
The CF and its Laplace in Caputo sense is defined as [31]
Z t
 Eqs. (2.15) and (2.16) on Eq. (2.10), we arrive at
1 1ðt  1Þ @hðy; sÞ


Dt hðy; tÞ ¼
CF 1
exp  ds; 0 q
1 @2
hðy; qÞ
11 0 11 @s hðy; qÞ ¼
q  aq þ a Pr @y2
< 1 < 1: ð2:15Þ

Df :Sc q
ðy; qÞ:
þ / ð3:24Þ
The LT of CF derivative is Pr q  aq þ a
  sLðhðy; tÞÞ  hðy; 0Þ The solution of second order partial differential equation say
L CF D1t hðy; tÞ ¼ : ð2:16Þ
ð1  1Þs þ 1 (3.20) is analyzed as:
pffiffiffiffiffi
ffi pffiffiffiffiffi
ffi pffiffiffiffiffiffi


cPr q cPr q A1 cSc q
hðy; qÞ ¼ c1 ey qþac þ c2 ey qþac  ey qþac : ð3:25Þ
2.2. Development of governing equations in terms of ABC q
derivative
With the help of boundary conditions for temperature, con-
stants c1 and c2 can be traced out in the following equation:
In this context, The ABC derivative and its Laplace in Caputo pffiffiffiffiffi ffi
pffiffiffiffiffiffi pffiffiffiffiffi ffi

1 cPr q A1 cPr q cSc q
sense is defined as [31] hðy; qÞ ¼ ey qþac þ  ey qþac  ey qþac : ð3:26Þ
Z t
 q q
1 1ðt  1Þ1 @hðy; sÞ
Dt hðy; tÞ ¼
ABC 1
E1  ds: ð2:17Þ Inverting an exponential equation through Laplace transform
11 0 11 @s
subject to the an Appendix (A1), we get final solution of tem-
The LT of ABC derivative is perature equation as
  s1 Lðhðy; tÞÞ  s11 hðy; 0Þ hðy; tÞ ¼ dðy; t; cPr ; caÞ þ A1
L ABC Dat hðy; tÞ ¼ : ð2:18Þ
ð1  1Þs1 þ 1
 ðdðy; t; cPr ; caÞ  dðy; t; cSc ; caÞÞ: ð3:27Þ
2272 S.T. Saeed et al.

3.3. Fractional optimality of velocity field via CF derivative 3.4. Fractional optimality of concentration via ABC derivative

Generating Eq. (2.9) for the fractionalized form, we imposed Generating Eq. (2.11) for the fractionalized form, we imposed
Eqs. (2.15) and (2.16) on Eq. (2.9), we arrive at Eqs. (2.17) and (2.18) on Eq. (2.11), we arrive at

 2

q @ 2 u
ðy; qÞ qa
ðy; qÞ ¼ @ /ðy; qÞ : ð3:32Þ
u
ðy; qÞ ¼ þ Gr 
hðy; qÞ þ Gm  /

ðy; qÞ; Sc a
q  aqa þ a
/
@y2
q  aq þ a @y2
ð3:28Þ The second order partial differential equation say (3.32) can
have the incomplete solution as investigated in Eq. (3.32)
by substitution of
h and /
into (3.28). With minor simplifica- qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 qa  qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 qa 
tion, we get:
y Sc qa aq a þa y Sc qa aq a þa
/ðy; qÞ ¼ c1 e þ c2 e : ð3:33Þ
pffiffiffiffiffi
cq
ffi pffiffiffiffiffi
cq
ffi B ðq þ acÞ pffiffiffiffifficPr q
ffi
u
ðy; qÞ ¼ c1 ey qþac þ c2 ey qþac 
1
2
ey qþac With the help of boundary conditions for concentration field,
q
pffiffiffiffiffi ffi pffiffiffiffiffiffi c1 and c2 can be traced out in the following equation:
B2 ðq þ acÞ y qþac
cPr q B3 ðq þ acÞ y qþac
cSc q
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 e þ e  qa 
q2 q2 1 y
pffiffiffiffiffi ffi
ðy; qÞ ¼ e
/
Sc qa aq a þa
: ð3:34Þ
B4 ðq þ acÞ y qþac
cS c q q
 e : ð3:29Þ
q2
The above solution (3.34) can written in more appropriate
With the help of boundary conditions for velocity field, c1 and form
c2 can be traced out in the following equation: qffiffiffiffiffiffiffiffiffi
c 1 Sc q a
y

 pffiffiffiffiffiffi 1 e
qa þc1 a
q cq
ðy; qÞ ¼
/ : ð3:35Þ
u
ðy; qÞ ¼ ey qþac q1a qa
q2 þ w2

pffiffiffiffiffi ffi
B1 ypqþacffiffiffiffiffi
cq
ffi cPr q Inverting an exponential equation through Laplace transform
þ e  ey qþac subject to the an Appendix (A6)–(A10), we get final solution
q

pffiffiffiffiffi ffi
B1 ac ypffiffiffiffiffi ffi
cq cPr q
for concentration field as:
y
þ 2 e qþac  e qþac
q /ðy; tÞ ¼ hðt; aÞ  ðvðy; t; c1 ; c1 a; 0ÞÞ; ð3:36Þ

pffiffiffiffiffi ffi
B2 ypqþacffiffiffiffiffi
cq
ffi cPr q
where c1 is letting parameter can be taken as c1 ¼ 1
.
þ e  ey qþac 1a
q

pffiffiffiffiffi ffi
B2 ac ypffiffiffiffiffi cq
ffi
y
cPr q 3.5. Fractional optimality of temperature via ABC derivative
þ 2 e qþac  e qþac
q

pffiffiffiffiffi ffi
B3 ypqþacffiffiffiffiffi
cq
ffi cSc q Generating Eq. (2.10) for the fractionalized form, we imposed
 e  ey qþac Eqs. (2.17) and (2.18) on Eq. (2.10), we arrive at
q

pffiffiffiffiffi ffi


B3 ac ypffiffiffiffiffi cq
ffi cSc q qa
1 @2
hðy; qÞ
 2 e qþac  e
y qþac
h ð y; qÞ ¼
q qa  aqa þ a Pr @y2

pffiffiffiffiffi ffi

B4 ypqþacffiffiffiffiffi
cq
ffi cSc q Df :Sc qa
ðy; qÞ:
þ e  ey qþac þ /
q Pr qa  aqa þ a

pffiffiffiffiffi ffi
B4 ac ypffiffiffiffiffi cq
ffi
y cSc q ð3:37Þ
þ 2 e qþac  e qþac
: ð3:30Þ
q The second order partial differential equation say (3.37) can
Inverting an exponential equation through Laplace transform have the incomplete solution as investigated in Eq. (3.37)
qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi
c1 Pr qa c 1 Pr q a c1 Sc qa
subject to the an Appendix (A2)–(A5), we get final solution for y y A y

ðy; qÞ ¼ c1 e
h
q a þac
1
þ c2 e
q a þac
1
 e
1 qa þac1
: ð3:38Þ
velocity equation as: q
1 With the help of boundary conditions for concentration field,
uðy; tÞ ¼ cosðwtÞ   Nðy; t; c; ca; 0Þ
t constants c1 and c2 can be traced out in the following equation:
qffiffiffiffiffiffiffiffiffi
þ B1 ðNðy; t; c; ca; 0Þ  Nðy; t; Pr c; ca; 0ÞÞ c1 Pr qa 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1
y qa þac1 c 1 Pr q a c1 Sc qa
e A y y

ðy; qÞ ¼
h þ
1
 @e
q a þac
1
e
qa þac1
A; ð3:39Þ
þ B1 acð#ðy; t; c; acÞ  #ðy; t; cPr ; caÞÞ q q
þ B2 ðNðy; t; c; ca; 0Þ  Nðy; t; Pr c; ca; 0ÞÞ qffiffiffiffiffiffiffiffiffi 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1
c 1 Pr q a c1 Pr qa c 1 Sc q a
y y
þ B2 acð#ðy; t; c; acÞ  #ðy; t; cPr ; caÞÞ 1 e
qa þac1
A1 Be qa þac1
y
e
qa þac1
C


hðy; qÞ ¼ þ 1a  B  C:
q1a qa q @ qa qa A
 B3 ðNðy; t; c; ca; 0Þ  Nðy; t; Sc c; ca; 0ÞÞ
 B3 acð#ðy; t; c; acÞ  #ðy; t; cSc ; caÞÞ ð3:40Þ
þ B4 ðNðy; t; c; ca; 0Þ  Nðy; t; Sc c; ca; 0ÞÞ Inverting an exponential equation through Laplace transform
B4 acð#ðy; t; c; acÞ  #ðy; t; cSc ; caÞÞ: ð3:31Þ subject to the an Appendix (A6)–(A10), we get final solution
for temperature as:
A mathematical study of natural convection flow through a channel with non-singular kernels 2273
qffiffiffiffiffiffiffiffiffi 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1
hðy; tÞ ¼ hðt; aÞ  ðvðy; t; c1 Pr ; c1 a; 0ÞÞ
 c 1 qa c 1 qa c 1 P r qa
q y B1 y y
þ @e A
qa þac1 qa þac1 qa þac1
u
ðy; qÞ ¼ e e
þA1 hðt; aÞ  ðvðy; t; c1 Pr ; c1 a; 0ÞÞ ð3:41Þ þw
q2 2 q
0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1
A1 hðt; aÞ  ðvðy; t; c1 Sc ; c1 a; 0ÞÞ: c qa c 1 P r qa c 1 qa c 1 P r qa
B1 ac1 @ y qa 1þac1 y qa þac1
A þ @e B 2 y qa þac1
y qa þac1
A
þ a e e e
q :q q
3.6. Fractional optimality of velocity field via ABC derivative 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1
c 1 qa c 1 P r qa c 1 qa c 1 S c qa
B2 ac y y
A  B3 @ e
y y
þ a 1 @e A
qa þac1 qa þac1 qa þac1 qa þac1
e e
q :q q
Generating Eq. (2.9) for the fractionalized form, we imposed 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1
c qa c 1 S c qa c qa c 1 S c qa
Eqs. (2.17) and (2.18) on Eq. (2.9), we arrive at B3 ac1 @ y qa 1þac1 y qa þac1
A B4 @ y qa 1þac1 y qa þac1
A
 a e e þ e e
q :q q
qa @ 2 u
ðy; qÞ 0 qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi 1
u
ðy; qÞ ¼ þ Gr
hðy; qÞ þ Gm /

ðy; qÞ: c qa
B4 ac1 @ y qa 1þac1 y
c 1 S c qa

qa  aq þ a
a @y2 þ a e e
qa þac1
A: ð3:44Þ
q :q
ð3:42Þ
Inverting an exponential equation through Laplace transform
By substituting
h and /
into (3.42), after some minor simplifi- subject to the an Appendix (A6)–(A10), we get final solution
cation, we get: for temperature as:
qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi
c1 qa c1 qa c Pr q a t1a
y qa þac1
y qa þac1 B1 ðqa þ ac1 Þ y q1a þac uðy; tÞ ¼ cosðwtÞ   vðy; t; c1 ; c1 a; 0Þ þ B1 hðt; aÞ
u
ðy; qÞ ¼ c1 e þ c2 e  e 1
CðaÞ
qa :q
qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi  ðvðy; t; c1 ; c1 a; 0Þ  vðy; t; Pr c1 ; c1 a; 0ÞÞ þ B1 ac1 hðt; 0Þ
c Pr q a c Sc q a
B2 ðqa þ ac1 Þ y q1a þac B3 ðqa þ ac1 Þ y qa1 þac  ðvðy; t; c1 ; c1 a; 0Þ  vðy; t; Pr c1 ; c1 a; 0ÞÞ þ B2 hðt; aÞ
 e 1
þ e 1

qa :q qa :q  ðvðy; t; c1 ; c1 a; 0Þ  vðy; t; Pr c1 ; c1 a; 0ÞÞ þ B2 ac1 hðt; 0Þ

qffiffiffiffiffiffiffiffiffi
c Sc qa  ðvðy; t; c1 ; c1 a; 0Þ  vðy; t; Pr c1 ; c1 a; 0ÞÞ þ B3 hðt; aÞ
B4 ðqa þ ac1 Þ y q1a þac
 e 1
: ð3:43Þ  ðvðy; t; c1 ; c1 a; 0Þ  vðy; t; Sc c1 ; c1 a; 0ÞÞ þ B3 ac1 hðt; 0Þ
qa :q
 ðvðy; t; c1 ; c1 a; 0Þ  vðy; t; Sc c1 ; c1 a; 0ÞÞ þ B4 hðt; aÞ
With the help of boundary conditions for concentration field,  ðvðy; t; c1 ; c1 a; 0Þ  vðy; t; Sc c1 ; c1 a; 0ÞÞ þ B4 ac1 hðt; 0Þ
constants c1 and c2 can be traced out:  ðvðy; t; c1 ; c1 a; 0Þ  vðy; t; Sc c1 ; c1 a; 0ÞÞ: ð3:45Þ

Fig. 1 Comparison of velocities with variation of time for ABC and CF.
2274 S.T. Saeed et al.

4. Results and discussion The influence of Gm on velocity field are shown in Figs. 4
and 5. The magnification in buoyancy forces leads to reduce
The investigation of Dufour effect on viscous fluid through a the viscous force which help to enhance the velocity field with
channel has been discussed by using Atangana-Baleanu frac- increase of Gm . For large and small value of time, the behavior
tional time derivative ðABCÞ and Caputo-Fabrizio fractional of velocities are same in Gm . The velocity field of Atangana-
time derivative ðCFÞ with suitable initial and boundary value Baleanu is larger than the velocity for other fractional opera-
problem. The influence of fractional parameter tors. Figs. 6 and 7 shows the behavior of Gr on velocity field.
a; Gm ; Gr ; Df ; Sc and Pr have been analyzed through graphically An increase in Gr , resultant velocity increases. It related to
with different differential operator. i.e. CF and ABC. thermal buoyancy effect, as buoyancy effect rises due to which
The behavior of velocity profile with variation of time for velocity of fluid increases. A comparative study is carried out
ABC and CF discussed in Fig. 1. The velocity profile achieved for fluid velocity profile for all models. Figs. 8 and 9 analyzed
by ABC approach is greater than the velocity obtained with CF the influence of Df on fluid velocity. As increase in Df causes
approach for small time. Whereas for large value of time, the the velocity enhance. Figs. 10 and 11 shows the effect of Sch-
response of velocities behavior are totally reverse. The velocity midt number Sc on velocity profile. The resultant velocity
achieved by CF is larger than the velocity achieved by ABC. reduce with increase in Sc . It is observed that fluid velocity
Another observation is consider for t ¼ 1, both velocities are for ABC fractional model is greater than CF model.
behave alike. The significant role of fractional parameter ðaÞ The resultant fluid velocity reduce with enhance the value
for CF and ABC as shown in Figs. 2 and 3. Clearly, fluid veloc- of Pr . The impact of Pr on fluid velocity analyzed in Figs. 10
ity increases with the increase in a for small and large time. As and 11. The thermal conductivity and thickness of boundary
time increases, the boundary layer thickness become greater layer shrink with increase the value of Pr and viscosity of fluid
and velocity is highest in the vicinity of plate. While making respectively (Figs. 12 and 13). The influence of a; Df ; Pr and Sc
comparison velocity for Atangana-Baleanu model is larger on temperature profile are analyzed in Figs. 14–17 by ABC
because it has nonlocal kernel. The non-integer fractional approach. The temperature reduce with increase in a discussed
order derivative reduce to classical model by a ! 1. in Fig. 14. The increasing behavior of temperature with large

Fig. 2 Profile velocity of a for CF with other parameters are Gm ¼ 5; Gr ¼ 10; Pr ¼ 0:71.

Fig. 3 Profile velocity of a for ABC with other parameters are Gm ¼ 5; Gr ¼ 10; Pr ¼ 0:71.
A mathematical study of natural convection flow through a channel with non-singular kernels 2275

Fig. 4 Profile velocity of Gm for CF with other parameters are Df ¼ 5; Gr ¼ 2; Pr ¼ 2.

Fig. 5 Profile velocity of Gm for ABC with other parameters are Df ¼ 5; Gr ¼ 2; Pr ¼ 2.

Fig. 6 Profile velocity of Gr for CF with other parameters are Df ¼ 2; Gm ¼ 10; Pr ¼ 0:5.
2276 S.T. Saeed et al.

Fig. 7 Profile velocity of Gr for ABC with other parameters are Df ¼ 2; Gm ¼ 10; Pr ¼ 0:5.

Fig. 8 Profile velocity of Df for CF with other parameters are Sc ¼ 0:8; Gm ¼ 8; Pr ¼ 2.

Fig. 9 Profile velocity of Df for ABC with other parameters are Sc ¼ 0:8; Gm ¼ 8; Pr ¼ 2.
A mathematical study of natural convection flow through a channel with non-singular kernels 2277

Fig. 10 Profile velocity of Sc for CF with other parameters are Gr ¼ 5; Gm ¼ 10; Df ¼ 0:8.

Fig. 11 Profile velocity of Sc for ABC with other parameters are Gr ¼ 5; Gm ¼ 10; Df ¼ 0:8.

Fig. 12 Profile velocity of Pr for CF with other parameters are Sc ¼ 0:5; Gm ¼ 5; Df ¼ 0:8.
2278 S.T. Saeed et al.

Fig. 13 Profile velocity of Pr for ABC with other parameters are Sc ¼ 0:5; Gm ¼ 5; Df ¼ 0:8.

Fig. 14 Profile temperature for a with variation of time and other parameters are Gr ¼ 5; Gm ¼ 10; Df ¼ 2; Pr ¼ 0:8.

Fig. 15 Profile temperature for Df with variation of time and other parameters are a ¼ 0:4; Gr ¼ 10; Sc ¼ 1:5; Pr ¼ 3.
A mathematical study of natural convection flow through a channel with non-singular kernels 2279

Fig. 16 Profile temperature for Pr with variation of time and other parameters are a ¼ 0:6; Gm ¼ 5; Sc ¼ 0:4; Pr ¼ 2.

Fig. 17 Profile temperature for Sc with variation of time and other parameters are a ¼ 0:8; Gm ¼ 10; Sc ¼ 2; Pr ¼ 0:8.

value of Df shown in Fig. 15. The effect of Pr on temperature Declaration of Competing Interest
discussed in Fig. 16. As Pr increase, temperature decrease. In
Fig. 17, the value of Sc increase with increase in temperature. The authors declare that they have no known competing finan-
cial interests or personal relationships that could have
5. Conclusion appeared to influence the work reported in this paper.

The comparative study of viscous fluid along vertical plate Appendix A.

with Dufour effect has been analyzed in view of non-local
and non-singular kernel. Caputo-Fabrizio ðCFÞ and

rffiffiffiffiffiffiffiffiffiffiffi
1 aq
Atangana-Baleanu ðABCÞ approach are used to compare the dðy; t; a; bÞ ¼ L1 exp y
q qþb
fluid velocity behavior. The key points have extracted from  
Z
this article. 2a 1 sinðysÞ bts2
¼1 ðA1Þ
2
e aþs ds
p 0 sða þ s Þ
2
 The velocity profile shows opposite behavior for different
values of time while t ¼ 1 the behavior of ABC and CF
rffiffiffiffiffiffiffiffiffiffiffi
1 aq
approach are same. Nðy; q; a; b; cÞ ¼ exp y ðA2Þ
qc qþb
 The magnitude of the velocity increase as increase in frac-
tional parameter. rffiffiffiffiffiffiffiffiffiffiffi Z

ac 2a 1 sinðysÞ s2 bt
 Velocity enhance with increase the value of Gr and Gm . Nðy; t; a; b; cÞ ¼ expðct  y Þ exp  ds
aþc p 0 sða þ s2 Þ a þ s2
 Increase the value of Df and S c , the resultant velocity also Z 1
2

2ac sinðysÞ s bt
increase.  exp  ds ðA3Þ
p 0 sðac þ ða þ cÞs2 Þ a þ s2
 Temperature reduce for increasing values of a and P r .

rffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ABC fractional derivative is more considerable as compared 1 m0 q
to CF. #ðy; q; m0 ; m1 Þ ¼  exp y ðA4Þ
q2 q þ m1
2280 S.T. Saeed et al.
Z 1
[12] J.C. Umavathi, J.P. Kumar, M.A. Sheremet, Heat and mass
#ðy; t; m0 ; m1 Þ ¼ Nðy; s; m0 ; m1 ; 0Þds ðA5Þ
0
transfer in a vertical double passage channel filled with
electrically conducting fluid, Phys. A 465 (2017) 195–216,
sffiffiffiffiffiffiffiffiffiffiffiffiffi! https://doi.org/10.1016/j.physa.2016.07.073.
1 qa þ b


v1 ðy; q; a; b; cÞ ¼ a exp y ðA6Þ [13] A.J. Chamkha, R.A. Mohamed, S.E. Ahmed, Unsteady MHD
q þa qa þ c natural convection from a heated vertical porous plate in a
micropolar fluid with Joule heating, chemical reaction and
Z 1 radiation effects, Meccanica 46 (2011) 399–411.
v1 ðy; t; a; b; cÞ ¼ vðy; t; a; b; cÞgðu; tÞdu ðA7Þ [14] A. O Ajibade, B.K. Jha, Free convection heat and mass transfer
0
flow in a vertical channel with the Dufour effect, Proc. Inst.
pffiffiffiffiffiffiffiffiffiffiffi Z Z Mech. Eng. Part E J. Mech. Eng. 224 (2010) 91–101.
atyy b  c 1 t eat
vðy; t; a; b; cÞ ¼ e  pffiffiffi pffiffi  expðat [15] P.S. Reddy, A.J. Chamkha, Soret and Dufour effects on MHD
2 p 0 0 t convective flow of Al2O3–water and TiO2–water nanofluids
y2 p ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi past a stretching sheet in porous media with heat generation/
 ct   uÞI1 ð2 ðb  cÞutÞdtdu ðA8Þ absorption, Adv. Powder Technol. 27 (2016) 1207–1218, https://
4u
doi.org/10.1016/j.apt.2016.04.005.
1 [16] M. Narahari, Transient free convection flow between long
hðt; 1Þ ¼ ; gðu; tÞ ¼ t1 @ð0; 1; ut1 Þ ðA9Þ vertical parallel plates with ramped wall temperature at one
t1 Cð1  1Þ
boundary in the presence of thermal radiation and constant
mass diffusion, Meccanica 47 (2012) 1961–1976.
tn
@ðk; 1; tÞ ¼ R ðA10Þ [17] M. Narahari, S. Tippa, R. Pendyala, M. Nayan, Ramped
n!Cðk  n1Þ temperature effect on unsteady MHD natural convection flow
past an infinite inclined plate in the presence of radiation, heat
source and chemical reaction, in: Recent Advances in Applied
References and Theoretical Mechanics, Hikari Ltd, Rousse, Bulgaria, 2015,
ISBN 978-1-61804-304-7.
[1] B.K. Jha, A.O. Ajibade, Diffusion-thermo effects on free [18] A. Siyal, K.A. Abro, M.A. Solangi, Thermodynamics of
convective heat and mass transfer flow in a vertical channel magnetohydrodynamic Brinkman fluid in porous medium, J.
with symmetric boundary conditions, J. Heat Transf. 133 (2011), Therm. Anal. Calorim. 136 (2019) 2295–2304.
052502. [19] A. Atangana, On the new fractional derivative and application
[2] L. Florio, A. Harnoy, Combination technique for improving to nonlinear fishers reaction–diffusion equation, Appl. Math.
natural convection cooling in electronics, Int. J. Sci. 46 (2007) Comput 273 (2016) 948–956.
76–92. [20] S. Iyad, T. Abdeljawad, F. Jarad, Monotonicity analysis for
[3] M. Fossa, G. Tanda, Frost formation in vertical channels nabla h-discrete fractional Atangana-Baleanu differences,
under natural convection, Int. J. Multiph. Flow 36 (2010) Chaos, Solit. Fract. 117 (2018) 50–59.
210–220. [21] M. Caputo, M. Fabrizio, A new definition of fractional
[4] J.C. Umavathi, M.A. Sheremet, Mixed convection flow of an derivative without singular kernel, Progr. Fract. Differ. Appl.
electrically conducting fluid in a vertical channel using Robin 1 (2015) 1–13.
boundary conditions with heat source/sink, Eur. J. Mech. B/ [22] T. Abdeljawad, M.B. Riaz, S.T. Saeed, A. Atangana, Role of
Fluids 55 (2016) 132–145. Local and Non-Local Kernels on MHD Flow of Oldroyd-B
[5] J. Umavathi, M. Sheremet, S. Mohiuddin, Combined effect of Fluid under Effect of Ramped Wall Temperature and Velocity
variable viscosity and thermal conductivity on mixed convection (submitted for publication).
flow of a viscous fluid in a vertical channel in the presence of first [23] M.B. Riaz, A. Atangana, T. Saeed, MHD free convection flow
order chemical reaction, Eur. J. Mech. B/Fluids (2016). over a vertical plate with ramped wall temperature and chemical
[6] M.A. Ismael, I. Pop, A.J. Chamkha, Mixed convection in a lid- reaction in view of non-singular kernel. (Book Chapter)
driven square cavity with partial slip, Int. J. Sci. 82 (2014) 47–61. Publisher: Wiley, ISBN: 9781119654162.
[7] M. Narahari, T. Sowmya, P. Rajashekhar, Ramped temperature [24] M.A. Imran, I. Khan, M. Ahmad, N.A. Shah, M. Nazar, Heat
and Dufour effects on transient MHD natural convection flow and mass transport of differential type fluid with non-integer
past an infinite vertical plate in a porous medium, Eur. Phys. J. order time-fractional Caputo derivatives, J. Mol. Liq. 229 (2017)
Plus 130 (2015) 251. 67–75.
[8] D. Kalidas, Magnetohydrodynamics free convection flow of a [25] K. Das, R.P. Sharma, A. Sarkar, Heat and mass transfer of a
radiating and chemically reacting fluid past an impulsively second grade magnetohydrodynamic fluid over a convectively
moving plate with ramped wall temperature, J. Appl. Mech. 79 heated stretching sheet, J. Comput. Des. Eng. 3 (2016) 330–336.
(2012), 061017. [26] G.S. Seth, B. Kumbhakar, S. Sarkar, Soret and Hall effects on
[9] A.J. Chamkha, A.-R.A. Khaled, A.A. Khaled, Hydromagnetic unsteady MHD free convection flow of radiating and chemically
combined heat and mass transfer by natural convection from a reactive fluid past a moving vertical plate with ramped
permeable surface embedded in a fluid–saturated porous temperature in rotating system, Int. J. Eng., Sci. Technol. 7
medium, Int. J. Numer. Methods Heat Fluid Flow 10 (455) (2015) 94–108.
(2000) 477. [27] N.A. Shah, A. Hajizadeh, M. Zeb, S. Ahmad, Y. Mahsud, I.L.
[10] A.J. Chamkha, Double-diffusive convection in a porous Animasaun, Effect of magnetic field on double convection flow
enclosure with cooperating temperature and concentration of viscous fluid over a moving vertical plate with constant
gradients and heat generation or absorption effects, Numer. temperature and general concentration by using new trend of
Heat Transf. Part A Appl. 41 (2002) 65–87. fractional derivative, Open J. Math. Sci. 2 (2018) 253–265.
[11] Z. Ismail, A. Hussanan, I. Khan, S. Shafie, MHD and radiation [28] J.R. Pattnaik, G.C. Dash, S. Singh, Radiation and mass transfer
effects on natural convection flow in a porous medium past an effects on MHD flow through porous medium past an
infinite inclined plate with ramped wall temperature: an exact exponentially accelerated inclined plate with variable
analysis, Int. J. Appl. Math. Stat. 45 (2013) 77–86. temperature, Ain Shams Eng. J. 8 (2017) 67–75.
A mathematical study of natural convection flow through a channel with non-singular kernels 2281

[29] N.A. Shahid, study of heat and mass transfer in a fractional fractional integro-differential equations, Chaos, Solit. Fract. 125
MHD flow over an infinite oscillating plate, SpringerPlus 4 (2019) 194–200.
(2015) 640. [33] S.A. Khan, K. Shah, G. Zaman, F. Jarad, Existence theory and
[30] M.B. Riaz, S.T. Saeed, Comprehensive analysis of integer order, numerical solutions to smoking model under Caputo-Fabrizio
Caputo-Fabrizio and Atangana-Baleanu fractional time fractional derivative, Chaos 29 (2019).
derivative for MHD Oldroyd-B fluid with slip effect and time [34] F. Jarada, T. Abdeljawadb, Z. Hammouchc, On a class of
dependent bounday conditions. Discr. Continuous Dynam. ordinary differential equations in the frame of Atangana-Baleanu
Syst., 2020 (in press). fractional derivative Chaos, Solit. Fract. 117 (2018) 16–20.
[31] M.A. Imran, M. Aleem, M.B. Riaz, R. Ali, I. Khan, A [35] A. Akgül, A novel method for a fractional derivative with non-
comprehensive report on convective flow of fractional (ABC) local and non-singular kernel, Chaos, Solit. Fract., vol. 114, pp.
and (CF) MHD viscous fluid subject to generalized boundary 478–482.
conditions, Chaos, Solit. Fract. 118 (2018) 274–289. [36] E.K. Akgüla, Solutions of the linear and nonlinear differential
[32] C. Ravichandrana, K. Logeswaria, F. Jarad, New results on equations within the generalized fractional derivatives, Chaos:
existence in the framework of Atangana-Baleanu derivative for An Interdiscip. J. Nonlinear Sci., 29 (2), 023108.