ME729: Modelling of Mechanical Properties of
Materials
Sumit Basu, Associate Professor, Department of
Mechanical Engineering, IIT Kanpur, Kanpur 208016, UP.
Email: [email protected], Ph: +91 (0) 512 259 7506,
Course Objectives
Learn to model the constitutive behaviourof materials
i.e devise laws describing how a particular material
deforms under application of loads which can be body
forces (eg. gravity), surface tractions (contact with other
bodies), thermal (heating and consequent expansion) or
electrical loads or chemical action (eg. material in a
corrosive environment) on the material.
What is a constitutive equation?
Deformation Rate Temperature
Other internal variablesthat characterise physical
quantities like the microstructure of the material,
history of prior deformation etc
}
Method I: Based on Phenomenology
Test under
uniaxial
conditions
Guess all parameters
that can have an
influence on the stress-
strain response. Vary
them one by one and
get curves.
Curve fit:
Construct a
function F( ) for
multi-
dimensional
situations, that,
when applied to
the uniaxial
case, explains all
the experimental
results.
Phenomenological models contd
Biaxial stretching tests on polyethylene
terephthalate (PET) conducted at ESA Lab,
IIT Kanpur.
From www.grantadesign.com
Extensive testing has led to material
property handbooks used routinely
by practising engineers.
Such information is used routinely for
design of components
From: http://www-materials.eng.cam.ac.uk/
Bottom up routes to F( )
Macroscopic structure deformed
under load
Deformations at the
macro-level are smoothly
varying
zoom
At a macro-point, the material is
homogeneous
A micro-
structure
underlies every
macro-point.
Properties are
determined by
the
microstructure
F(..) at this point
obtained through
micromechanics at the
level of the
microstructure
Suggested Reading:
1.Mechanics of Solids and Materials, Asaro and Lubarda, Cambridge Univ
Press, 2006.
2.Crystals, Defects and Microstructures, Philips, Cambridge Univ Press,
2001
3.Foundations and Applications of Mechanics, vol 1, Jog, Narosa, 2005
4.Introduction to the Mechanics of a continuous Medium, Malvern, Prentice-
Hall
5.Nonlinear Solid Mecahnics, A continuum approach for engineering,
Holzapfel, Wiley, 2000.
6.The Physics of Polymers, Strobl, Springer, 1996
7.Fluid Mechanics of Viscoelasticity, Huilgol and Phan-Thien, Elsevier,
1997
8.Mechanics of the Cell, Boal, Cambridge Univ Press, 2002.
9.Research papers
Evaluation procedure:
1.2 quizzes, after first and second mid semester
periods respectively. Open book, open notes, use of
laptops and internet allowed. (30%)
2.About 6 assignments. You do not need to submit
them.
3.1 one hour mid semester exam, open book, open
notes, use of laptops and internet allowed. (30%)
4.3 hour end semester exam, open book, open notes,
use of computers and internet allowed. (40%)
A test generally determines properties under restricted
conditions
A UTM determines uniaxial
properties
Strain gauges
determine surface
strains
Real life is seldom uniaxial!
Loading and geometry renders a real life
problem non-uniaxial!
Maity et al. (2007a,b), IEEE
TDEI
Any observable,
macroscopic property P is
given by
Average over a
statistically
representative periodic
volume
On top of that, materials are
inhomogeneous!
Macroscopic properties are homogenised forms of
microscopic properties.
Homogenisation scheme applied to a typical
heterogeneous single crystal superalloy: (a) RVE and
(b) predicted contours of accumulated inelastic strain
under [010] uniaxial loading
The micro picture is always inhomogeneous, while the macro
may be homogeneous.
From Busso, 2004
Micro may be big!
Bandopadhyay et al Constr. Bldg
Mater, 2006
Maitra et al, ASCE J Civil Engng
Mater, submitted
aggregates
Air voids
Asphalt binder
Main crack
Intense localisation of
deformation, porous
material.
Micro may be local
Basu and Narasimhan,
JMPS, 1999,2000
Interfaces: properties are difficult to test
Substrate
Coating
From Hohlfelder
(1998)
.
Pressure (MPa)
Volume(micro-l)
Experimental data from Hohlfelder et al., 1998
Blister tests are used to assess
integrity of thin films on hard
substrates, eg in electronic
packaging
Funded by ISRO
Simulating interfaces
Kulmi and Basu, Model Simul Mater Sc
Engng, 2006
Sudarkodi and Basu, JMPS, submitted.
Funded by DST
500 mm
crack
Shear bands
craze
Bulk
Bulk
Fibrils
~750nm ~50nm
crazes
Micro may be local and small!
Basu and van der Giessen, IJP,
2002
Basu et al., Polymer, 2005
Experiments:
simple loading,
macroscopic data.
Model:
Generalised but
should explain
experimental
observations under
restricted
conditions.
Material
inhomogeneities
interfaces
Damage
Size effects
1. Concept of strain and deformation (Jog) (2 lectures)
i.Lagrangian and Eulerian descriptions of motion, deformation gradient tensor
ii.Transformation of length elements
iii.Transformation of area and volume elements
iv.Velocity, acceleration and rates of deformation
v.Examples of simple motions
2. Conservation laws and stress measures (Jog) (5 lectures)
i.Transport theorems, balance of mass, momentum and angular momentum
ii.Cauchy theorem and problems
iii.Momentum balance in reference configuration
iv.Variational theorems
v.Thermodynamics of deformation (first and second laws) and problems
vi.Example of modeling of growing mass (Asaro+papers)
vii.Assignment
Topics to be covered
1. General theories of constitutive behaviour of materials (4 lectures, Asaro)
i.Introduction to constitutive laws
ii.Applications of the laws of thermodynamics, concept of thermodynamic potentials.
iii.Example of linear thermoelasticity
iv.Irreversible thermodynamics and internal variables
v.Assignment
vi.Concept of frame indifference and objectivity (cover Jaumann and other rates)
vii.Assignment
2. Specific constitutive theories and applications-1, elasticity (12 lectures) (Asaro)
i.Non-linear and linear elasticity (2)
ii.Example of non linear elasticity: rubber elasticity (1)(Strobl)
iii.Biological membranes + surface elasticity
iv.Simple problems: hole in an infinite plate (1)
v.Green’s function solutions: indentation problems (1)
vi.Inclusion problem: Eshelby formalism (1) (Asaro+Mura)
vii.Effective properties of composites (3)
viii.Fields due to dislocations (2)
ix.Misc problems involving dislocations (near interfaces, dislocation arrays etc.) (2)
Assignment
1. Specific constitutive theories and applications-II, plasticity (11 lectures) Asaro+papers
i.Phenomenological theory of plasticity, pressure dependent yield, kinematic hardening. (4)
ii.Discussion of porous plasticity (1)
iii.Assignment
iv.Micromechanics of crystallographic slip (1)
v.Crystal plasticity, example of shear localisation (2)
vi.Strain gradient plasticity (1)
vii.Discrete dislocation plasticity (1)
viii.Open issues in metal plasticity esp. size effects (1)
ix.Computer assignment
2. Phenomenological viscoelasticity (1)
3. Microscopic viscoelasticity: Rouse, dumbbell and reptation models (1) (Huilgol)
4. Assignments
5. Length scale bridging issues and open questions (5 lectures) (Phillips+papers)
i.Problems with Multiple time and length scales
ii.Historic examples of multiscale modeling
iii.Multiscale models in fracture, cohesive models
iv.Dislocation dynamics, Molecular dynamics
v.Molecular dynamics of metals, quasicontinuum approaches
Molecular dynamics of amorphous materials
A quick introduction to Continuum Mechanics
Kinematics of deformation
P
p
Motion:
Notes
Example
Example
An example
Animation of the
motion described in
the previous slide . The
arrows denote
velocities at the
corners of the
rectangle in the
deformed
configuration.
Notes
Material Derivative
Velocity profile of the problem in Slide 31 at
times t=0.5 (blue) and 0.8 (green)
Example on the application of the
material derivative
Deformation Gradient tensor
The deformation gradient tensor maps an
infinitesimal line element tangent to a
curve on the reference configuration to an
infinitesimal line element to a curve on the
deformed/current configuration
P
p
Notes
Deformation Gradient: in terms of the
displacement gradient
P
p
Concept of strain
Physical significance of the Green Lagrange strain tensor
Example
Mapping surfaces and volumes
Proof
P
p
P
p
Example:
Polar decomposition
Definition
Proof
Uembodies only stretch. Polar decomposition splits the
motion into a stretch and a rotation
Spectral decomposition of C
Example
Example: finding Uin 2-d
Exercise
Rates of deformation
Example: Lin terms of U
Some more rates
Balance Laws and Stress Measures
Reynolds Transport theorem
}}}
See explanation
See explanation
Mass conservation
Example
Example
Balance of momenta
Notes
Stress in the reference configuration:
First Piola Kirchoff stress
Equation of motion in terms of the 1
st
PK stress
Second Piola Kirchoff stress tensor
Cooking up stresses!
Balance of angular momentum
Principal stresses and directions
Example: find the maximum shear stress directions at a
point
Example: The meaning of the first Piola Kirchoff stress
Reference configuration
Current configuration
Note that the surface with a cross in the reference configuration
goes to the surface with a cross in the deformed configuration.
This means that the motion is a combination of a rotation about
the X
3axis followed by a stretch in the X
1direction and a
contraction in the X
2direction.
Balance Laws: The two laws of thermodynamics
Balance of energy: first
law of thermodynamics
P
p
Rate of thermal work
Second law of thermodynamics
Clausius Duhem inequality in the reference
configuration
Entropy due to heat
conductionRate of stress work
Example:Moving discontinuity surfaces:
Application of balance laws
Internal variables
Internal variables quantify the missing dissipation!!
Types of thermomechanical processes
Free energy
Example: Second law and constitutive
equations
Example: Thermodynamic restrictions on elastic
solids
Objectivity and Frame Indifference
Objectivity and Frame Indifference
Proof
Proof
Example: Is velocity objective?
Example: What about acceleration?
Coriolis acceleration
Centrifugal
acceleration
Some tensors
Proof
Proof
Rate of Cauchy stress
Objective and material response
The requirement that constitutive equations be ‘Material
Frame Indifferent’ poses restrictions on the nature of these
equations!
Example: Jaumann rate
Putting it all together: Case Study I:
Mechanics of a growing mass.
A very important problem in biomechanics. Growth and remodelling of tissues takes
place during normal developmental growth, healing processes and pathological
conditions
Objectives
•Formulate the basic balance equations for the problem
•Assume a constitutive framework consistent with the
Clausius Duhem inequality and the principle of
objectivity.
•Solve a simple problem
Momentum principles
Energy principle: 1
st
law of thermodynamics for a
continuum with growing mass
2
nd
law for a continuum with growing mass
Building a constitutive framework
Example: circumferential growth of a blood vessel
Reversible materials: hyperelasticity
Isotropic Green elasticity
Usual assumptions…
This is the usual linear elastic case….
Alternate forms of expressing the constitutive equation for
Green elastic materials
Incompressible hyperelasticity
Further forms of hyperelastic
constitutive laws
Examples: Mooney Rivlin rubber
Inflating a spherical balloon
Snap through
Snap back
For more read:Needleman (1977), Inflation of spherical balloons, Int J Solids
Structures, v13, p409.
How do we get the free energy?
Case study 1: a linearly elastic crystal
Linking micro to macro: introduction to statistical
mechanics
subsystem
reservoir
A short review of statistical mechanics
This equation links the micro to the macro world!
Polymer chain as a random walk
Compare this with the Mooney Rivlin material
Linear elastic isotropic materials
Navier equation
Potential Methods for problems in Linear Elasticity:
Papkovich Neuber potentials
See, http://www.engin.brown.edu/courses/en224/for details
Green’s function
Some Fourier transforms
Useful Fourier transforms
Concept of averaging over space
X
Y
x
y
V
Macrostress
Microstress
Representative
Volume Element
(RVE)
B
H
Example:
Phase averages
RVE
Basic idea of homogenisation
techniques
Properties of the localisation tensor
To find the stiffness Cor the compliance D, knowledge of the localisation
tensors are sufficient.
The simplest assumption: Voigt
model
Strain everywhere is same as
the average strain
Maybe valid for low contrast
cases
Rule of mixtures
Reuss model
Mean field approaches: dilute inclusions
Low volume fraction
Each particle sees an infinite
matrix around it.
A non-mathematical overview of the dilute
particle theory
Composite subjected to macrostrain
Assumption:if volume fraction f<0.25
All points in the matrix also has strain
since the matrix does not feelthe particles
Inside the particles, according to Eshelby’s
theory it can be shown that strains are
constant.
The strain in the particles can be found by
using the Eshelby tensor.
Thus the localisation tensor in the matrix is the identity tensor and that in the
particles can be found (constant). The total localisation tensor can thus be
derived.
Concept of eigenstrains
Unconstrained
expansion
Eigenstrains
Eigenstrains are non-stress strains.
Stress is caused only by the elastic
strain.
The Eshelby procedure
I
II
III
IV
VVI
Disturbance strain due to the
particle.
Main result of Eshelby’s (1957) analysis
Eshelby tensor
Ref: Micromechanics of defects in solids, T. Mura, 1982
Estimates of average moduli from Eshelby’s concept of the
equivalent inclusion.
A digression: how do we invert fourth order
tensors?
Alternate method: E bases
Beyond dilute distributions: estimating matrix
strains
Dilute distribution assumes that matrix strains ~
At higher volume fractions, matrix strains are non-uniform
A particle is ‘embedded’ in a matrix
with strains
The localisation tensor is same as
in the case when the matrix had
strains
Self consistent schemes
When it is not clear which is the matrix and which are particulates i.e
phases are not distinct.
Assume:the particles are embedded in a medium with the overall
elastic modulus
This approximately accounts for the inclusion-inclusion interaction
Since both sides are functions of , iterative solution of the above
equation is necessary
How good are the estimates: bounds
Ref: Hill, J Mechanics Physics Solids (1963)