CHAPTER 1

INTRODUCTION

1.1 Background of the Study

The study of fluid mechanics has long been a cornerstone of applied mathematics and

engineering science, underpinning the analysis and design of a vast range of industrial,

biological, and environmental systems. Historically, the Newtonian fluid model — which

assumes a linear relationship between shear stress and the rate of strain — has served as the

dominant framework for theoretical and computational investigations. While this model is

adequate for simple fluids such as water and air, it is fundamentally inadequate for the large

class of fluids encountered in modern industrial practice, including polymer melts,

lubricating greases, biological fluids, foodstuffs, paints, and drilling muds. These materials

exhibit complex rheological behaviour, including shear-thinning, shear-thickening,

viscoelasticity, yield stress, and memory effects, none of which can be captured by the

Newtonian constitutive equation.

1
 Non-Newtonian fluid mechanics has consequently emerged as an active and

expansive field, with its theoretical foundation rooted in rational continuum mechanics. The

constitutive modelling of non-Newtonian fluids presents considerable mathematical

challenges, as it necessitates the introduction of additional material parameters and higher-

order derivative terms that substantially increase the complexity of the governing equations.

Among the broad hierarchy of non-Newtonian models — which include power-law,

Bingham plastic, Carreau-Yasuda, Oldroyd-B, Maxwell, and higher-order fluids — the

second-grade fluid occupies a distinguished position as the simplest model capable of

accounting for normal stress differences while retaining a mathematically tractable structure.

The constitutive equation for a second-grade fluid introduces two material parameters

beyond viscosity: the coefficients of elasticity α₁ and α₂, whose signs are constrained by

thermodynamic considerations. Specifically, the Clausius-Duhem inequality and the

requirement of stability of the rest state demand that μ ≥ 0, α₁ ≥ 0, and α₁ + α₂ = 0, as

established by Dunn and Fosdick (1974). When α₁ > 0, the fluid exhibits energy storage

(elastic behaviour); when α₁ < 0, it dissipates elastic energy in an anomalous sense that can

give rise to physically unexpected flow phenomena.

The stretching-sheet problem constitutes one of the most fundamental and extensively

studied configurations in boundary-layer fluid mechanics. The problem was first formulated

and solved exactly by Crane (1970), who demonstrated that the steady, laminar, two-

dimensional flow of a Newtonian fluid over a flat elastic sheet stretching linearly in its own

plane admits a closed-form exponential solution for the velocity field. The elegance of this

exact solution and the direct relevance of the configuration to manufacturing processes —

2
particularly the continuous casting and drawing of plastic films, fibres, and metal sheets —

have motivated a large and still-growing body of research extending the Crane problem to

increasingly realistic physical settings. These extensions include non-Newtonian rheology,

magnetic fields, porous media, non-uniform heating, thermal radiation, chemical reactions,

and nanofluid effects.

The present study is situated at the intersection of three important physical

phenomena: non-Newtonian fluid behaviour (modelled by the second-grade fluid), porous

medium effects (modelled by the Darcy drag term), and non-uniform thermal effects (viscous

dissipation and internal heat generation/absorption). The simultaneous consideration of these

effects is physically realistic for polymer processing, lubrication engineering, and biological

flow through tissue, and it generates a mathematically challenging coupled system of

nonlinear ordinary differential equations (ODEs) that are not amenable to standard analytical

treatment. The accurate and efficient solution of this system is the central methodological

concern of this thesis.

1.2 Problem Background

1.2.1 Non-Newtonian Fluids in Industrial Applications

The prevalence of non-Newtonian fluids across modern industry is both widespread

and consequential. In polymer processing — the manufacture of synthetic fibres, plastic

3
films, rubber sheets, and composite materials — the working fluid is invariably a viscoelastic

polymer melt or solution whose flow properties are strongly rate-dependent and history-

dependent. The quality of the final product, including its microstructure, mechanical strength,

and optical properties, is directly determined by the velocity and temperature fields within

the processing zone. In lubrication engineering, greases and synthetic oils are non-Newtonian

fluids whose shear-thinning behaviour under high-pressure, high-shear conditions is

deliberately engineered to protect machine surfaces. In biomedical engineering, blood,

synovial fluid, and mucus are non-Newtonian biological fluids whose anomalous flow

properties are central to the function of the circulatory, skeletal, and respiratory systems. In

food processing, concentrated suspensions, emulsions, and gels exhibit yield-stress and

viscoelastic behaviour that determines mixing efficiency, pumping power requirements, and

product texture.

The second-grade fluid model is particularly appropriate for dilute polymer solutions,

biological fluids under moderate shear rates, and certain lubricating oils. Its constitutive

equation introduces the first and second Rivlin-Ericksen tensors, capturing the leading-order

elastic effects that manifest as non-zero first and second normal stress differences — a

characteristic departure from Newtonian behaviour that is observable in phenomena such as

the Weissenberg rod-climbing effect, extrudate swell, and die swell in polymer extrusion.

While more sophisticated models (e.g., Oldroyd-B, FENE-P) may be required for strongly

elastic or shear-thinning fluids, the second-grade model provides a balance of physical

realism and mathematical tractability that makes it an important canonical model in

theoretical studies.

4
1.2.2 The Stretching Sheet Configuration

The stretching sheet problem, first solved by Crane (1970) for a Newtonian fluid, has

become the paradigmatic model for continuous surface processes in manufacturing. In these

processes, a flat surface — which may be a polymer sheet, a metallic strip, a rubber film, or a

fibre — is extruded continuously from a die or slit and drawn along its length. The stretching

velocity of the surface imparts momentum to the adjacent fluid, generating a boundary layer

whose velocity and temperature distributions determine the rate of cooling or heating and,

ultimately, the mechanical properties of the product. The mathematical idealisation assumes

that the sheet stretches linearly with velocity u_w = cx, where c is the stretching rate and x is

the distance from the slot, and that the flow occupies the half-space above the sheet.

The beauty of Crane's solution lies in the fact that, for a Newtonian fluid with a

linearly stretching sheet, the nonlinear partial differential equations of boundary-layer theory

reduce to a single nonlinear ODE that admits an exact closed-form solution. This property

has been shown to carry over to the second-grade fluid: Liu (2005) demonstrated that the

fourth-order momentum ODE governing second-grade flow over a stretching sheet in a

porous medium and a magnetic field also admits an exact exponential solution, provided that

the constitutive parameters satisfy a specific algebraic constraint. The existence of this exact

solution for the velocity field is exploited systematically in the present work as an

unimpeachable mathematical benchmark against which approximate methods are validated.

5
1.2.3 Role of Porous Media and Thermal Effects

The embedding of the stretching sheet in a porous medium introduces an additional

Darcy drag force that opposes fluid motion and modifies the velocity boundary layer. Porous

media are relevant in a variety of engineering contexts, including packed-bed reactors,

groundwater flow, geothermal systems, filtration, and the manufacture of fibrous materials.

The Darcy resistance term adds a linear damping effect to the momentum equation, whose

magnitude is governed by the permeability of the medium k'. In the dimensionless

formulation, this effect is characterised by the porosity parameter λ₂ = ν/(ck'), which

represents the ratio of viscous resistance to the stretching-induced inertia.

The thermal aspects of the problem are governed by the energy equation, which in the

present formulation includes three physically important effects beyond standard convective-

diffusive heat transfer. First, viscous dissipation — characterised by the Eckert number Ec —

converts kinetic energy of the fluid into thermal energy through the action of viscous

stresses, acting as a distributed volumetric heat source within the boundary layer. This effect

is significant in high-speed polymer processing and lubrication flows. Second, a non-uniform

internal heat source or sink, modelled following Abel et al. (2010) with space-dependent

coefficient A* and temperature-dependent coefficient B*, accounts for the chemical,

radiative, or metabolic heat generation that may occur within the fluid. Third, wall suction —

characterised by the suction parameter R — draws fluid toward the sheet, thinning both the

velocity and thermal boundary layers and enhancing heat transfer. The interplay of these

6
thermal mechanisms generates a rich and physically complex temperature distribution that

motivates the development of accurate solution methods.

1.3 Problem Statement

The governing equations of steady boundary-layer flow and heat transfer for a

second-grade fluid over a permeable stretching sheet embedded in a porous medium, with

viscous dissipation and non-uniform internal heat generation, constitute a coupled system of

nonlinear ordinary differential equations: a fourth-order momentum ODE and a second-order

energy ODE. While the momentum ODE admits an exact closed-form solution for the

velocity field — an exponential decay f'(η) = e^{−sη}, where s satisfies a cubic algebraic

equation — no such closed-form solution exists for the energy equation, owing to its

nonlinear coupling through the Prandtl number, the Eckert number, and the heat source

parameters.

The conventional approach to solving such problems has been the Differential

Transform Method (DTM), introduced by Zhou (1986). The DTM constructs a global power-

series approximation through simple algebraic recurrences and has been applied extensively

to boundary-layer problems, including the second-grade fluid configuration studied by Abel

et al. (2010). However, the DTM suffers from a fundamental and well-documented

convergence limitation: as a global polynomial approximation, its radius of convergence is

often insufficient to span the physical domain [0, ∞). For boundary-layer problems with far-

7
field conditions imposed at large η, the DTM polynomial invariably diverges beyond a

moderate value of η — typically η ≈ 2–3 for the present problem — rendering it unreliable

for large-domain predictions. This divergence cannot be fully remedied by the DTM-Padé

technique, whose accuracy depends on a non-trivial choice of approximant order and is not

guaranteed for stiff or strongly nonlinear systems.

The critical gap identified in the existing literature is therefore twofold. First, there is

a need for an accurate, efficient, and systematically improvable method that overcomes the

convergence limitation of the classical DTM for the second-grade fluid boundary-layer

problem over the full domain. Second, the existing literature contains a mathematical

discrepancy in the derivation of the cubic equation for the velocity parameter s, particularly

for the case λ₁ = 0.5, whose resolution requires a rigorous rederivation from first principles.

The present work addresses both gaps through the systematic application of the Multistep

Differential Transform Method (MsDTM) and a corrected derivation of the exact solution.

1.4 Research Objectives

The primary objective of this study is to apply the Multistep Differential Transform

Method (MsDTM) to obtain accurate semi-analytical solutions for the coupled momentum

and energy equations governing second-grade fluid flow and heat transfer over a permeable

stretching sheet embedded in a porous medium, and to conduct a systematic parametric

8
investigation of the resulting velocity and temperature fields. The specific objectives are as

follows:

(1) To derive the exact closed-form solution for the velocity field of the second-grade

momentum ODE by direct substitution, and to obtain the corrected cubic algebraic

equation for the decay parameter s, resolving the discrepancy present in the existing

literature for the case λ₁ = 0.5.

(2) To develop fully explicit, step-by-step recurrence relations for the MsDTM applied to

both the fourth-order momentum ODE and the second-order energy ODE, including

all convolution terms arising from the product nonlinearities.

(3) To implement the MsDTM shooting algorithm in Maple 18 and to validate the

numerical results against the exact velocity solution and a fourth-order Runge-Kutta

(RK4) temperature benchmark, demonstrating accuracy better than 10 ⁻¹² throughout

the domain [0, 8].

(4) To conduct a systematic convergence study of the MsDTM with respect to the

number of terms per subinterval (N) and the number of subintervals (M), establishing

the exponential convergence rate and comparing the performance against the classical

DTM.

9
 (5) To investigate the effects of the governing dimensionless parameters — the second-

grade parameter λ₁, the porosity parameter λ₂, the suction parameter R, the Prandtl

number Pr, the Eckert number Ec, and the heat source parameters A* and B* — on

the velocity and temperature profiles, the skin-friction coefficient, and the local

Nusselt number.

1.5 Scope of the Study

The scope of this study is circumscribed by the following physical and mathematical

assumptions. The flow is assumed to be steady, laminar, two-dimensional, and

incompressible. The fluid is modelled as a second-grade fluid with constant viscosity and

constant elastic coefficients; temperature-dependent properties are not considered. The

stretching sheet is assumed to be flat and to stretch linearly, with velocity u_w = cx;

nonlinear or exponential stretching laws are beyond the scope of the present work. The

porous medium is homogeneous and isotropic, characterised by a constant permeability k';

anisotropic or heterogeneous porous media are excluded. Magnetic field effects

(magnetohydrodynamics) are not included in the present formulation, though the

methodology is directly extendable to the MHD case. Thermal radiation and chemical

reaction are also excluded, as their inclusion would require additional modelling assumptions

beyond the scope of the present study.

10
 The computational domain is truncated at η_max = 8, beyond which the velocity and

temperature solutions are verified to be below 10 ⁻⁵ for all parameter combinations

considered. The MsDTM implementation uses M = 80 equal subintervals of width h = 0.1

and N = 15 terms per subinterval as the standard configuration, which is shown to provide

essentially machine-precision accuracy for all cases. The complete Maple 18 implementation

is available from the corresponding author upon reasonable request.

1.6 Significance of the Study

1.6.1 Mathematical Significance

From a mathematical perspective, this study makes three principal contributions.

First, it provides the first rigorous rederivation of the exact velocity solution for the second-

grade fluid over a stretching sheet in a porous medium, using direct substitution of the ansatz

f'(η) = e^{−sη} into the full fourth-order ODE without approximation, and thereby corrects

an algebraic inconsistency that has propagated through several published works. Second, it

provides fully explicit MsDTM recurrence relations for a fourth-order nonlinear ODE with

product nonlinearities — a level of detail absent from most prior publications that cite the

DTM but relegate the derivational steps to a brief summary — making the methodology fully

reproducible and directly extensible to related problems. Third, it presents a comprehensive

and systematic convergence analysis, including both N-convergence and M-convergence (h-

convergence) studies with pointwise error tables, establishing the exponential convergence

11
rate O(h^{2N}) predicted by local Taylor series theory and confirming saturation at floating-

point precision for N = 15, M = 80.

1.6.2 Physical and Engineering Significance

From a physical perspective, the parametric investigations conducted in this study

provide quantitative insight into the interplay between fluid elasticity, porous medium

resistance, suction, and thermal effects in a configuration directly relevant to industrial

manufacturing. The finding that the second-grade parameter λ₁ thickens the velocity

boundary layer and reduces wall shear stress has direct implications for the design of

polymer extrusion and film-drawing processes, where the wall shear stress determines the

tension in the drawn material and therefore its mechanical integrity. The finding that

increasing the porosity parameter λ₂ or the suction parameter R simultaneously thins both

boundary layers and enhances the Nusselt number provides design guidance for industrial

cooling systems that employ porous substrates or suction mechanisms to augment heat

transfer from continuously moving surfaces.

The analysis of viscous dissipation, through the Eckert number Ec, reveals the

possibility of a temperature overshoot above the wall temperature for large Ec, indicating that

frictional heating within the boundary layer can dominate over conductive cooling in high-

speed processing. This finding is directly relevant to the design of high-speed polymer

drawing and lubrication systems, where excessive frictional heating can degrade material

properties or cause thermal damage. The investigation of the heat source parameters A* and

12
B* provides insight into the thermal management of processes involving exothermic or

endothermic chemical reactions within the fluid, with direct application to reactive polymer

processing and biological flow through metabolically active tissue.

1.6.3 Methodological Significance

From a methodological perspective, the MsDTM framework developed in this thesis

provides a broadly applicable and computationally efficient alternative to standard numerical

methods (such as finite difference, finite element, or shooting with RK4) for nonlinear

boundary-layer problems. Unlike fully numerical methods, the MsDTM produces a semi-

analytical representation of the solution as a piecewise polynomial, which can be evaluated at

any point in the domain without additional computation and which provides explicit

expressions for the solution's derivatives. This is advantageous in engineering design

contexts where parametric sensitivity — i.e., the dependence of flow quantities on system

parameters — must be assessed rapidly. The complete Maple 18 implementation provided

with this thesis serves as a template that can be directly adapted to a broad class of related

boundary-layer problems, including MHD flow, three-dimensional stretching, and nanofluid

configurations.

1.7 Brief Overview of the Literature

13
 Research on non-Newtonian fluid flow over stretching sheets has expanded

dramatically over the past five decades, evolving from Crane's seminal exact solution for

Newtonian fluids to complex multi-physics models incorporating magnetic fields, porous

media, radiation, chemical reactions, and nanofluids. The following brief overview identifies

the principal lines of research that converge in the present work; a comprehensive critical

review is provided in Chapter 2.

The exact solution framework was extended from Newtonian to second-grade fluids

by several authors. Liu (2005) obtained an exact solution for the velocity field of a second-

grade fluid over a stretching sheet in a porous medium and a transverse magnetic field,

establishing the existence of an exponential solution and the cubic equation relating the

decay constant to the governing parameters. Abel et al. (2010) built upon Liu's velocity

solution to investigate the thermal problem with non-uniform heat source/sink and viscous

dissipation, employing the classical DTM and the DTM-Padé technique. This work is the

most direct predecessor of the present study and has been widely cited; however, as

demonstrated in Chapter 5, the DTM solutions reported by Abel et al. are inaccurate beyond

η ≈ 2, and the exact velocity value reported for λ₁ = 0.5 contains an algebraic error.

The Multistep Differential Transform Method was introduced by Odibat et al. (2010)

as a systematic remedy for the convergence failure of the classical DTM and was applied to

second-grade fluid flow over a stretching/shrinking sheet by Rashidi et al. (2011). The

present work extends the Rashidi et al. framework to include the thermal problem with non-

uniform heat source/sink and viscous dissipation, provides the corrected exact solution, and

14
offers a substantially more detailed convergence analysis. Table 1.1 below summarises the

key studies that are most directly related to the present work.

Table 1.1 Key Related Studies in Second-Grade Fluid Flow over Stretching Sheets

Author(s) & Year Configuration Method Key Contribution

Crane (1970) Newtonian fluid, Exact First exact closed-form solution

Liu (2005) Second-grade fluid, Exact Exact velocity solution; cubic for
porous medium, MHD s

Abel et al. (2010) Second-grade fluid, DTM, DTM-Padé Extended Liu's model to thermal
porous medium, heat problem
source, viscous dissipation

Odibat et al. Nonlinear ODEs (general) MsDTM Introduced domain decomposition

Rashidi et al. Second-grade fluid, MsDTM First MsDTM application to

Das et al. (2016) Second-grade MHD, Numerical MHD and convective BC

Sidahmed & Salah Second-grade, radiation, Analytical/ Radiation and reaction effects
(2022) chemical reaction Numerical

Present Study Second-grade fluid, MsDTM + Exact Corrected exact solution; full
porous medium, suction, convergence analysis; complete
heat source, viscous recurrences
dissipation

1.8 Thesis Organisation

15
 This thesis is organised into six chapters, each addressing a distinct component of the

study. The organisation is as follows:

Chapter 1 (Introduction): presents the background, motivation, problem statement,

objectives, scope, and significance of the study. A brief overview of the literature is provided

to contextualise the present work within the broader research landscape.

Chapter 2 (Literature Review): provides a comprehensive and critical review of the

existing literature on second-grade fluid flow and heat transfer, stretching sheet problems,

and semi-analytical solution methods. The chapter identifies the specific research gaps that

motivate the present work.

Chapter 3 (Mathematical Formulation): presents the physical configuration,

governing partial differential equations, similarity transformations, reduced ODE system, and

exact velocity solution. The derivation of the corrected cubic equation for the decay

parameter s is given in full detail, with an explicit correction of the literature discrepancy.

Chapter 4 (Methodology): develops the DTM and MsDTM frameworks in detail,

including full derivations of all recurrence relations for both the momentum and energy

ODEs. The shooting algorithm, convergence criteria, and Maple 18 implementation are

described. The RK4 benchmark scheme for temperature is also detailed.

16
 Chapter 5 (Results and Discussion): presents all numerical results, including

convergence analyses, pointwise error tables, velocity and temperature profiles, and

parametric investigations of all governing parameters. Literature comparisons are provided,

and the corrected exact solution is validated. The physical implications of each parametric

trend are discussed.

Chapter 6 (Conclusions and Future Work): summarises the principal findings of

the study, restates the specific contributions, acknowledges the limitations, and proposes

directions for future research.

Three appendices are included: Appendix A provides the complete Maple 18 code for

the MsDTM implementation; Appendix B contains the full derivations of the similarity-

transformed ODEs and the corrected cubic equation; and Appendix C tabulates extended

parametric results for the skin-friction coefficient and the local Nusselt number.

1.9 Chapter Summary

This chapter has established the motivation, context, and objectives of the present

study. The industrial prevalence of non-Newtonian fluids, the physical relevance of the

stretching-sheet configuration, and the importance of accurate thermal analysis have been

argued in detail. The fundamental problem — the convergence failure of the classical DTM

for large-domain boundary-layer problems — has been identified, and the MsDTM has been

17
proposed as the systematic remedy. The five specific objectives of the study have been

stated, and the scope and significance of the work have been delineated from mathematical,

physical, and methodological perspectives. The thesis organisation has been described to

guide the reader through the subsequent chapters. Chapter 2 now provides a comprehensive

review of the existing literature to position the present work precisely within the established

body of knowledge.

18