Mechanics of Composite Materials

Mechanics of Composite
Materials
Course contents
• Introduction • Long fiber composites
– Definitions – Fiber orientation
– Why use composite materials – Sandwich structures
– Benefits and drawbacks over metals and non-reinforced – Layup properties
plastics – Types of plies/mats weaving
– Different types of fiber reinforcements
– Levels of analysis of composites • Analysis of composite plies
– Composites in nature – Analysis homogenization laws
– Typical microstructures – Case of Iso-Strain vs Iso-Stress
– 3D woven composites
• Manufacturing of composites – Treatment as anisotropic material
– Manufacture of thermoplastic resin composites – Elastic behavior and Voigts notation
– Manufacture of thermosets plastic composites – Constants of elasticity
– Reverse engineering an existing polymer matrix – Transformation of reference
composite part – Invariants of compliance and stiffness
• Types of composite materials • Analysis of layups
– Classification based on fiber length – Behavior of thin plates
– Classification based on matrix type – Asymmetrical plates
– Non-polymer matrix composites – Introduction to failure of plies
• Mechanical Behavior of Composites • Failure theory of Composite plates
– Effect of fiber percentage – Maximum Stress based criteria for first ply failure
– Effect of short or long fiber reinforcements – Energy based criteria
– Effect of type of reinforcement – Delamination and propagation of cracks
– Effect of matrix type
Introduction
• What is a composite material?
– A composite material (also called a composition material
or shortened to composite, which is the common name) is
a material made from two or more constituent materials
with significantly different physical or chemical properties
that, when combined, produce a material with
characteristics different from the individual constituents.
– Difference between a composite and an alloy?

Introduction
• Material Definitions
– Homogenous
– Heterogeneous
– Isotropic
– Anisotropic
– Transversely isotropic
– Orthotropic
Introduction
• Why to use composites?
– Lighter weight and higher strength to weight ratio
– Directionally preferential properties
– Controlled coefficient of expansion
– Dielectric properties
– Different phase and class of materials can be joined to get interesting
properties.
Longitudinal carbon fibers/epoxy Glass fiber web/epoxy Glass fiber rings

Introduction
• Comparison to metals.
– Lighter
– Non conductive (in case of polymer matrix)
– Easier to manufacture
– Can be used as they are manufactured.
– Directional properties.
Introduction
• Comparison to non reinforced polymers.
– Superior strength
– Superior damping and noise absorption
– Higher strength to weight ratio hence lighter
components
Introduction
Matrice
Continuous
Fibers
Disontinous fibers
Particle or whiskers
filler
Unidirectional discontinous fibers
Unidirectional
Particulate
composite Woven Tissue
Composite
Quasi-isotropic
Random Discontinous fibers
Multidirectional Composite
Introduction Matrix
Fibre
• Levels of analysis of Micromechanical
composite Ply
components
Macromechanical
Layup
Structural analysis
Structure
Introduction
• Composites in nature
– The cells in natural materials are aligned in more or less one
direction.
– Naturally materials are either ductile or brittle and such composite

structures give them aggregate modified properties
– Wood is a composite – it is made from long cellulose fibres (a

polymer) held together by a much weaker substance called lignin.
Cellulose is also found in cotton, but without the lignin to bind it
together it is much weaker. The two weak substances – lignin and
cellulose – together form a much stronger one.
– Bones
• Calcium Phosphate - CaPO3 amorphous – Fragile
• Firbrous Collagen – Matrix – Hard but less stiff
Introduction
• Composites in nature
– It is important to understand wood as a
cellular and composite structure.
However, at different length scales, from
the cellulose molecule to the
macrostructure of lumber as we have
become accustomed to look at it visually.
The figure illustrates the hierarchy of
length scales of the atomic structure of
cellulose to the structure of a tree trunk.
The basic functional block of wood is the
glucose polymer known as cellulose,
which occurs as (most of the time) in
crystalline fiber. The other critical
component of wood is lignin, which is a
complex and amorphous material
containing phenyl groups. Lignin is
critical for structural properties and also
for the (relative) insensitivity of wood to
the environment.
Introduction
• Typical microstructures
– Natural is normally
cellular in nature.
– There are three types of
principal structural
composite materials
• MMC: Metallic Matric
Composite
• CMC: Ceramic Matrix
Composites
• PMC: Polymer Matrix
Composites
Introduction
• Typical Properties
– Properties of the
constituents of
composites
Manufacturing of composites
• Manufacturing of polymer processes
– Open Mould
• Composite materials (resin and fibers) are placed in an
open mold, where they cure or harden while exposed to
the air. Tooling cost for open molds is often inexpensive,
making it possible to use this technique for prototype and
short production runs.
– Closed Mould
• Composite materials are processed and cured inside a
vacuum bag or a two-sided mold, closed to the
atmosphere. Closed molding may be considered for two
cases: first, if a two-sided finish is needed; and second, if
high production volumes are required.
• Open Mould
– Hand layup
• Hand lay-up is the most common and least expensive open-
molding method because it requires the least amount of
equipment. Fiber reinforcements are placed by hand in a mold and
resin is applied with a brush or roller. This process is used to make
both large and small items, including boats, storage tanks, tubs
and showers.
• Open Mould
– Sprayup
• Spray-up is similar to hand lay-up but uses special equipment—
most notably a chopper gun—to cut reinforcement material into
short fibers, add them to resin and deposit the mixture (called
chop) on to a molding surface. Spray-up is more automated than
hand lay-up and is typically used to produce large quantities.
• Open Mould
– Filament Winding
• Filament winding is an automated process that applies resin-
saturated, continuous strands of fiber reinforcements over a
rotating cylindrical mold. It’s used for creating hollow products like
rocket motor casings, pipes, stacks, and chemical storage tanks.
Filament winding is less labor-intense than other open-molding
processes.
• Closed Mould
– Vacuum Bag Molding
• This manufacturing process is designed to improve the mechanical properties of
laminate (two or more layers of fiber reinforcement bonded with a resin). A
vacuum is created to force out trapped air and excess resin, compacting the
laminate. High-fiber concentration provides better adhesion (between layers of
sandwich construction). In addition, vacuum bag molding helps eliminate excess
resin that builds up when structures are made using (open-molding) hand lay-up
techniques.
• Closed Mould
– Vacuum Infusion Processing
• Vacuum infusion processing (VIP) is a technique that uses vacuum pressure to drive
resin into a laminate. Vacuum infusion is typically used to manufacture very large
structures. Vacuum infusion produces strong, lightweight laminates and offers
substantial emissions reductions (compared to open-molding processing and wet
lay-up vacuum bagging). This process uses the same low-cost tooling as open
molding and requires minimal equipment.
• Closed Mould
– Resin Transfer Molding
• Resin transfer molding (RTM), sometimes called liquid molding, is a closed-molding
method in which reinforcement material is loaded into a closed mold, the mold is
clamped, and resin is pumped in (through injection ports) under pressure. This
process produces complex parts with smooth finishes on all exposed surfaces. The
process can be simple or highly automated–and cycle times are speedy. By laying
up reinforcement material dry inside the mold, any combination of materials and
orientation can be used, including 3-D reinforcements.
• Closed Mould
– Compression Molding
• Compression molding is a manufacturing process in which composite materials are
sandwiched between two matching molds under intense pressure and heat (from
250° to 400° F) until the part cures. This technique is used to rapidly cure large
quantities of complex fiberglass-reinforced polymer parts. Compression molding
features fast molding cycles and high part uniformity. The process can be
automated. In addition, labor costs are low and it provides design flexibility and
nice surface finishes.
• Closed Mould
– Pultrusion
• Pultrusion is used to form composites into long, consistent shapes like rods or bars.
Continuous strands of reinforcement are pulled through a resin bath to saturate
them, then pulled through heated steel molds that sculpt the composites into
continuous lengths. The process operates continuously so it can be readily
automated. Labor costs are low and finished products are very strong. Pultrusion is
used to make products such as beams, channels, pipes, tubing, fishing rods and golf
club shafts.
• Closed Mould
– Reinforced Reaction Injection Molding
• Reinforced reaction injection molding (RRIM) is widely used to make external and
internal automotive parts. In this process, two (or more) resins are heated
separately and combined with milled glass fibers. The mixture is injected into a
mold under high pressure and compressed. The resin cures quickly. RRIM
composites feature many processing advantages, including very fast cycle time, low
labor, low mold-clamping pressure and low scrap rate. The RRIM process requires
special resins and reinforcements.
• Closed Mould
– Centrifugal Casting
• In centrifugal casting, reinforcements and resin are deposited against the inside
surface of a rotating mold. Centrifugal force holds them in place until the material
cures or hardens. Centrifugal casting is used to produce hollow parts (like pipes
with two smooth surfaces). It’s especially well-suited for producing structures with
large diameters, such as pipes for oil and chemical industry installations and
chemical storage tanks. Centrifugal casting is increasingly being used to produce
telephone, street light and other poles.
• Closed Mould
– Continuous Lamination
• Continuous lamination is used to make flat or corrugated sheets and panels for
products used in truck and RV sidewalls, road signs, skylights, building panels and
electrical insulating materials. It’s a highly automated process in which fibers and
resin are combined, sandwiched between two plastic carrier films—and guided
through a conveyor process. Forming rollers shape the sheets, and the resin is
cured (in an oven or heating zone) to form the composite panel. Panels are
automatically trimmed to the desired width and length.
• Reverse Engineering a Polymer Matrix
Composite part
– Determination of laminate sequence
• All polymer matrices are brunt off at elevated temperatures such as 600oC leaving
behind the fiber reinforcements, which can be physically inspected to determine
their ply or layup sequence.
• Fiber to matrix mass percentage can also be determined by this process
𝐵 𝑀𝑓 𝜌𝑚
𝑀𝑓 = × 100% 𝑉𝑓 = × 100%
𝐴 𝜌𝑓 − 𝑀𝑓 (𝜌𝑓 − 𝜌𝑚 )
𝑀𝑚 = 100 − 𝑀𝑓 % 𝑉𝑚 = 100 − 𝑉𝑓 %
Composite part
– Determination of Polymer type
• The most widely used technique is that of Differential Scanning Calorimetery (DSC)
Differential scanning calorimetry or DSC is a

thermoanalytical technique in which the
difference in the amount of heat required to
increase the temperature of a sample and
reference is measured as a function of
temperature
• Endothermic step,
glass transition
• Exothermic peak
due to cold
crystalization
• Endothermic peak
due to melting
• Reverse Engineering a Polymer
Matrix Composite part
– Glass transition Temperature
• The glass–liquid transition or glass transition for
short is the reversible transition in amorphous
materials (or in amorphous regions within
semicrystalline materials) from a hard and
relatively brittle "glassy" state into a viscous or
rubbery state as the temperature is increased.
An amorphous solid that exhibits a glass
transition is called a glass. The reverse
transition, achieved by supercooling a viscous
liquid into the glass state, is called vitrification.
The glass-transition temperature Tg of a
material characterizes the range of
temperatures over which this glass transition
occurs. It is always lower than the melting
temperature, Tm, of the crystalline state of the
material, if one exists.
Composite part
– Heat capacity
• Heat capacity or thermal capacity is a measurable physical quantity equal
to the ratio of the heat added to (or removed from) an object to the
resulting temperature change
Composite part
– Crystallization of polymers is a process associated
with partial alignment of their molecular chains.
These chains fold together and form ordered
regions called lamellae, which compose larger
spheroidal structures named spherulites.
Composite part
– Typical DSC Plots for Thermosets
Long fiber reinforced Composites
• Long fiber composites
– Fiber orientation
– Sandwich structures
– Layup properties
– Types of plies/mats weaving
• Long fiber composites
– Also known as continuous fiber composites,
where the fiber reinforcement is continuous
between the applied load and the fixation or
between two edges of the component
• Fiber Orientation
– The fiber orientation
can be measured with
reference to the
principal loading
direction or the global
coordinate system in
which the specimen is
aligned
• Sandwich Structures
– The composite structures can be stiffened by using
a sandwich configuration, without significantly
increasing the weight of the said structure.
• Sandwich Structures
– Bending stiffness relates directly to the
moment of inertia
– K = M/EI
• Layup Properties
– Symmetrical Layups also known
as Mirror Symmetry
– Asymmetrical Layups introduce
interesting coupling properties
• Bending torsion Coupling for
example
– Asymmetrical Layups introduce
interesting coupling properties
• Bending torsion Coupling for
example
– General Behavior
• Dependence on Vf
• Filament
– A single strand of the material is
known as a filament, usually is of the
same diameter
• Fiber
– Is a combination of a number of
filaments wound together to make up
a braid or a rope type structure
• Tex
– Is the specific weight per meter length
usually expressed in grams, this can be
both for a filament or a fiber, e.g 200
tex carbon filament is 0.007mm
• Types of Weaving and Mats
• Mats are specified by their

– Weave Type
– Tex number of the fiber e.g. 300, 700 etc.
– Specific weight in gm/m2
Mechanical Behavior
• Effects of fiber percentage
– Rule of Mixtures
– In general 𝑃𝑐 = 𝑉1 𝑃1 + 𝑉2 𝑃2 + 𝑉3 𝑃3 + ⋯ + 𝑉𝑛 𝑃𝑛
– Therefore for bi-material or bi-phase composites

– 𝐸𝑙 = 𝑉𝑓 𝐸𝑓 + (1 − 𝑉𝑓 )𝐸𝑚
Mechanical Behavior
– Iso – Strain
• All the constituents are
subjected to the same strain
– Iso – Stress
• All the constituents
experience the same stress
– Each hypothesis brings a

very different result
according to its treatment
Mechanical Behavior
– Iso – Strain
• All the constituents are subjected to the same
strain
• Assuming 𝜀𝑐 = 𝜀
• From the force transmitted through the section we
can see that 𝜎𝐶 = 𝑉𝑎 𝜎𝑎 + 𝑉𝑏 𝜎𝑏 = 𝑉𝑎 𝐸𝑎 𝜀𝑐 + 𝑉𝑏 𝐸𝑏 𝜀𝑐
• Therefore the modulus becomes
𝜎𝐶
• 𝐸𝑐 = = 𝑉𝑎 𝐸𝑎 + 𝑉𝑏 𝐸𝑏
𝜀𝑐
Mechanical Behavior
– Iso – Stress
• All the constituents are subjected to the
same stress
𝜎𝑐 𝜎
• Assuming 𝜀𝐴 = , 𝜀𝐵 = 𝑐
𝐸𝐴 𝐸𝐵
𝑉 𝜎 𝑉 𝜎
• 𝜀𝐶 = 𝑉𝑎 𝜀𝑎 + 𝑉𝑏 𝜀𝑏 = 𝑎 𝑐 + 𝑏 𝑐
𝐸𝑎 𝐸𝑏
• Therefore the modulus becomes
1 𝜀𝑐 𝑉𝐴 𝑉𝐵
• = = +
𝐸𝑐 𝜎𝑐 𝐸𝐴 𝐸𝐵
Mechanical Behavior
• Voigt and Reuss Notations
– Iso – Strain is known as the
Voigt Notation or the Rule
of Mixtures
– Iso – Stress is known as the
Reuss Notation or the
Inverse Rule of Mixtures
– Hill has proposed that an
appropriate notation be
developed where the
loading is not according to
these extreme cases.
Mechanical Behavior
• Strength of Composites
– As for the simplest case of rule of mixtures the strength of
composites can also be defined as
– 𝑆𝑐 = 𝑉𝑓 𝑆𝑓 + 𝑉𝑚 𝑆𝑚
– However the brittleness of the fibers limit the application of
stress to the matrix.
– 𝑆𝑐 = 𝑉𝑚 𝑆𝑚
– To accurately model this behaviour the failure of the ductlie
matirx at the strain of the brittle fiber can be taken into
consideration.
– 𝑆𝑐 = 𝑉𝑓 𝑆𝑓 + 𝑉𝑚 σ′𝑚
– Where 𝜎′𝑚 = 𝐸𝑚 𝜀′𝑓
– With 𝜀′𝑓 being the strain at failure of the fibers
Mechanical Behavior
• Strength of Composites
– Therefore a limiting minimum value of the fibers
volume ratio is required for the strength of the
composite to exceed that of the matrix alone
𝑆𝑐 𝑆𝑐 = 𝑉𝑓 𝑆𝑓 + 𝑉𝑚 σ′𝑚
𝑆𝑐 = 𝑉𝑚 𝑆𝑚
0 Vf 1
Mechanical Behavior
• Some common Nomenclature
– UD Ply; is a ply of fabric where the fibers are in
only one direction
– Woven Ply; is a ply of fabric with the fibers in a
mutually orthogonal direction
– Mat; a ply of fabric with random fibers
Mechanical Behavior
– In common terminology “m” will signify
the matrix, “f” the fiber and “c” the
composite
– The directions in case of UD and Woven
would be specified by “L” as in the
direction of the fiber and “T” as
tangential to it.
E fL – Is the young’s modulus of the
composite in the direction of the fiber
sometime replaced by EL
E cT – Is the young’s modulus of the composite
in the direction transverse to that of the
fibers sometimes replaced by ET
E cT
E cL
Mechanical Behavior
– Quality of composite based on volume of
voids in the material
– Vair > 5% average quality composite

– Vair < 1% hi quality composite material
Analysis of Composite Plies
• Longitudinal and transverse Modulus of
Elasticity
E cL  E fl Vf  E m Vm
  f    Vm 
1 V

E cT  E f    E m 
Elasticity
Vf    Vm
1    

   f G f   mG m   c G c Gc  Gf   G m 
2
Elasticity lt/2
hm/2
1 matrix
1
hf
fiber
1
hm/2
matrix
l l
– This case is that of Iso-Strain
l cT    c  cL h f  h m     f  fL h f   m  mL h m
 c   f V f   m Vm
2
Elasticity lt/2
hm/2
1 matrix
1
hf
fiber
1
hm/2
matrix
l l
– This case is that of Iso-Strain
l cT    c  cL h f  h m     f  fL h f   m  mL h m
 c   f V f   m Vm
• For the improvement of the calculated
properties in the transverse direction we
use the correction factor of Halpin Tsai.
E cL  E fL Vf  E m Vm  c   f V f   m Vm
– For Transverse direction we have

M 1   Vf M f M m  - 1
  (1)
Mm 1 - Vf M f M m   
– Where ξ is 2 for EcT and 1 for GcLT the module
is determined empirically.
Determine for a fiber glass / epoxy with 70% volume fraction of fibers:
a) The density of the composite and the mass fractions of the fibers and the matrix
b) The modules EL and ET at determine the ratio of the load supported by the fibers
in the logitudinal direction by the fibers with that of the composite.
c) The coefficeints of Poisson vLT and vTL
d) The Modulus of Shearing GLT.
e) Also compare the values of ET and GLT as calculated using the Halpin Tsai
formulation
Les propriétés du verre et de l’époxy:

ρf=2.5 g cm-3, vf=0.2, vm=0.3, ρepoxy=1.35 g cm-3, Ef=70 GPa, Em=3.6 GPa
E cL  E fL Vf  E m Vm  c   f V f   m Vm
M 1   Vf M f Mm -1
  (1)
Mm 1 - Vf M f Mm  
• For the estimation of woven plies we can
superpose two UD plies on top of the
other
Tissé Unidir 1 Unidir 2
Thickness e e k e (1-k) k  n1 n1  n 2 

Reinforcemnt
Vf Vf k Vf (1-k)
percentage
• For the estimation of a 3D composite
component
• General Hypotheses
– Elastic Domain with reversibility of
phenomena
– Small deformations
– Progressive application of loads
– Linear behaviors
– Existence of an elastic potential
1 ∂W ( ε )
W (ε ) = C ijkl ε ij ε kl σ ij = = Cijkl ε kl
2 ∂ε ij
∂ σ ij ∂ σ kl ∂ 2 W ()
C ijkl = = C ijkl 
∂ ε kl ∂ ε ij ∂ ij ∂ kl
• Stiffness Tensor
– In tensor form the Hooke’s law can be given as
 ij  cijkl  kl with i , j  1,2,3 1  U i U j  1
 ij    U i, j  U j,i 
 
2  x j x i  2
 ij  sijkl  kl with i , j  1,2,3
– In general any strain can produce stress
 ij  cijkl  kl with i , j  1,2,3
– The tensor is of the form Cijkl
– i, j, k, l = 1, 2, 3
– 3 x 3 x 3 x 3 = 81 elements
– For example the contribution of ε31 to σ11 is given by C1131ε31
 ij  c ij11 11  c ij12 12  c ij13 13  c ij21  21  c ij22  22  c ij23  23
 c ij31  31  c ij32  32  c ij33  33
ε 11 × c 1111
ε 22 × c 1122
ε 33 × c 1133
ε 12 × c 1112
ε 21 × c 1121 σ 11
ε 13 × c 1113
ε 31 × c 1131
ε 23 × c 1123
ε 32 × c 1132
– Symmetries
• Symmetry of stress : ij = ji c ijkl  c jikl
• Symmetry of Strain :  kl =  lk c ijkl  c ijlk
 11   s 1111 s 1112 s 1113 2s 1114 2s 1115 2s 1116   11 

  s s 2222 s 2223 2s 2224 2s 2225 2s 2226   22 
 22   2212  

  33 
  s 3313 s 3323 s 3333 2s 3334 2s 3335 2s 3336    33 

    
  23   2s 4414 2s 4424 2s 4434 4s 4444 4s 4445 4s 4446    23 
  31   2s 5515 2s 5525 2s 5535 4s 5545 4s 5555 4s 5556    31 
    

  12 
  2s 6616 2s 6626 2s 6636 4s 6646 4s 6656 4s 6666  
 12 
– The stress and strain tensors are of the
second order
    ij  where i , j  1,2,3
    ij 
– And written as a vector we have
  11,  22 ,  33 ,  23 , 13 , 12  
 
   ,  ,
11 22 33 , 2  23 , 2  13 , 2  
12 
– And the Mechanical Behavior is written as

  C 
  S 
– The Voigt Notation can be used to reduce the
Tensor using Symmetries
C  
  11  C 11 C 12 C 13 C 14 C 15 C 16    11 
   C C 22 C 23 C 24 C 25 C 26   22 
 22   12  
  C 13
  33  C 23 C 33 C 34 C 35 C 36    33 

   
  23  C 14 C 24 C 34 C 44 C 45 C 46    23 
  31  C 15 C 25 C 35 C 45 C 55 C 56    31 
    

 12  C 16 C 26 C 36 C 46 C 56 C 66  
  12 

• Compliance Tensor
– The Voigt Notation can be used to reduce the Tensor using
Symmetries
S  
– With the strain as

 12  s1211 11  s1222 22  s1233 33  2 s1223 23  2 s1213 13  2 s1212 12
– Which becomes
212  S 6111  S 62  22  S 63  33  S 64  23  S 65 13  S 66 12
• Hooke’s Law
– The Rigidity Matrix can be divided in 4 quadrants as
 I II 
 
III IV 
– The elements II and III define the coupling behavior

• The tension and compression induces shearing
• The expansion creates shearing behavior
Such coupling is absent in isotropic materials

• Hooke’s Law
– The Rigidity Matrix can be
divided in 4 quadrants as
 I II 
 
III IV 
– The elements II and III define

the coupling behavior
• The tension and compression
induces shearing
• The expansion creates shearing
behavior
Such coupling is absent in isotropic

materials
• Hooke’s Law 12,1
– The Rigidity Matrix can be  11  S 16   
E1
divided in 4 quadrants as
12,2
 I II   22  S 26  
E2

 
III IV  12,3
 33  S 36   
E3
– The elements II and III define
the coupling behavior  12,23
2 23  S 46   
• The tension and compression G 23
induces shearing  12,13
• The expansion creates shearing 2 13  S 56   
behavior G 13

2 12  S 66  
Such coupling is absent in isotropic G 12
materials
• Transformation of matrices
– The anisotropic properties are defined in
the local coordinates as (1, 2, 3) they
need to be translated to (x, y, z) for
inclusion of real world loads and
boundary conditions
• Transformation of matrices x ' y' z'
– The transformation matrix is designated T11 T12 T13 
by [T] = (Tij) T   T21 T22 T23 

T31 T33 
– The Matrix is to be designated between T32
XYZ and X’Y’Z’
– The transformation can be calculated by

rotation about a single axis at a time
– The inverse of the matrix is its transpose
T 1  T t T 1   T ji [T]
– The stress and strains translate from one T11 T12 T13 
reference to the other by the following T   T21 T22 T23 

relationships T31 T32 T33 
 ij'  Tki Tlj  kl   ij  Tik T jl  kl'

 ij'  Tki Tlj  kl   ij  Tik T jl  kl'
– And the rigidities can be transformed as
'
cijkl  c pqrs T pi Tqj Trk Tsl  cijkl  c 'pqrs Tip T jq Tkr Tls
[T]
– In general then T11 T12 T13 
T   T21 T23 
'  M  
T22

T31 T32 T33 
 
 '  M   
– Where
D1  2A   D1  A  
M      ; M     
 B  D 2   2B  D 2 
[T]
– And
D1  2A   D1  A  

M      ; M    
 B  D 2   2B  D 2 
– Where
– Also
D1 t 2B t   D1 t B t 

M 
1
 M 
t
  M 1
 M t
 
 A t
D 2 t   2A t
D 2 t 
C '  M C M t ou C   M t C 'M 
S '  M S M t ou S   M t S 'M 

z z’
– Simplification leads to y’
cos   sin  0
P    sin  cos  0
 y
 0 0 1  
x x’
z z’
– Simplification leads to y’
 c2 s2 0 0 0 2 sc 
 2 
 s c2 0 0 0  2 sc  y

 0 0 1 0 0 0  x x’
M     
 0 0 0 c s 0 
 0 0 0 s c 0  c
2
s2 0 0 0  sc 
  2 
  sc sc 0 0 0 cos 2   s c2 0 0 0 sc 
 0 0 1 0 0 0 
M     
 0 0 0 c s 0 
 0 0 0 s c 0 
 
 2 sc  2 sc 0 0 0 cos 2 
– Orthotropic Materials
– Quasi-Isotropic Transverse
– Example of Quasi-Isotropic Transverse
• Carbon Mat
C 11  C 12
C 66  2C 11  C 12  2C 66  sin 2  cos 2   C 66 C 66 
2
– To translate into Material properties
C 11  C 22
C  C  0
 44 55
C 11  C 12  C 33  0

C 11  C 12  2 C 2
13
 C 33
E1
12  1 
E3
1  12
13  31 
2
• Compliance • Elasticity
• Hooke’s law in terms of invariants
– Most composite structures an be treated as shells where
their thicknesses are very less.
  11  12 0
 
     12  22 0  or     11 ,  22 , 0, 0, 0,  12 
 0 0 
 0
 11   S 11 S 12 0   11 
    
  22    S 12 S 22 0   22 
 2   0 S 66   
 12   0  12 
 11   Q 11 Q 12 0   11 
    
 22    Q 12 Q 22 0    22 
   0 Q 66   
 12   0 212 
• Transformation of Stresses and Strains
 l   c 2 s2 2 sc    x 
   2  
 t    s c2  2 sc   y    l t  T  x y
    sc sc c 2  s 2   x y 
 lt  
  x   c2 s2  sc    l 
   2    
y    s c2 sc    t      T ' l t
   2 sc   xy
 xy   2 sc c 2  s 2   l t 
 1   yx  xy 
   xy x x Ex
E Ey G xy 
 x   x x    
     xy 1  xy    G xy Ex  xy G xy
 y 
   y   xy y y Ey
   E x Ey G xy      
 xy    y 1   xy  G xy Ey  xy G xy
x
 
 E x Ey G xy 
• Transformation of Stresses and Strains
1  c4 s4  1  
   c 2 s 2  - 2 tl  
E x (  )  El Et  Glt Et  
t y
1  s4 c4  1   l
   c 2 s 2  - 2 tl  
E y (  )  El Et  Glt Et  
  1
G xy (  )  1 4 c 2 s 2  
1 
 2 tl
 c2 s2
 
  
2

  E l Et Et  Glt 
x
 yx
Ey
  
 tl
Et
c 4
  1
 s 4 - c 2 s 2  
1 1 
- 
 El Et Glt 
 yx  s2 c2 
G xy
   2cs -  c 2 - s 2    E
tl
-
1
 

 El Et  t 2Glt 
 yx  c2 s2 
G xy
   2cs -  c 2 - s 2    E
tl
-
1
 

 El Et  t 2Glt 
• Transformation of Stresses and tStrains y l
 El  ltE tl 
 0 
  l  1   tl  lt 1   tl  lt
  l 
    tlE l Et  
 t   0  t 
  1   tl  lt 1   tl  lt    
 lt   0 0 G lt   lt 
  x
 
 x  c 2 s2  2 sc   l   l 
   2    
 y   s c2 2 sc   t   T   t 
   sc  sc c 2  s 2   l t    x   El  tl E l 0   x 
 xy     lt     
 
  y   T   lt E t Et 0  T '   y 
 l   c2 s2 sc  x  x     0 0 G lt   
 2   xy    xy 
     
 t    sc    y   T '    y 
2
 s c
    2 sc 2 sc c2  s2     
 lt     xy   xy 
x  E 11 E 12 E 13   x 
     
y   E 21 E 22 E 23   y 
  E 31 E 32 E 33   
 xy    xy 
• Transformation of Stresses and Strains y
t l
E 11 = c 4 E l + s 4 E t + 2 c 2 s 2 ( νtl E l + 2 Glt )
E 22 = s 4 E l + c 4 E t + 2 c 2 s 2 ( νtl E l + 2 Glt ) 
x
2 2 2 2 2
E 33 = c s ( E l + E t - 2 νtl E l ) + ( c - s ) Glt
E 12 = c 2 s 2 ( E l + E t - 4 Glt ) + ( c 4 + s 4 )νtl E l
E 13 = -cs {( c 2 E l - s 2 E t - ( c 2 - s 2 )( νtl E l + 2 Glt )}
E 23 = -cs {( s 2 E l - c 2 E t + ( c 2 - s 2 )( νtl E l + 2Glt )}
El Et
El  Et 
1   tl  lt 1   tl  lt
• Transformation E 1,2 Engineering Constants in Local Direction
Flow
Q 1,2 S 1,2 Mathematical Constants
 Fiber Orientation
Qx,y S x,y Transformed Mathematical

Constants
Repère (1,2)  (L,T) E x,y Transformed Engineering Constants

Analysis of Layups
• Behavior of thin plates
• Layup is a combination of plies laid one
on top of the other in one plane
z z
y h
x
Ny ek
Nx Txy
Analysis of Layups
– The Strains of the membrane are considered as plane bi-
dimensional
– The non zero Strains are written as u 0 v 0
v 0  0 xy  
u 0 y x
 0x   0y 
x y z
k 1
n
k 1

Nx    x k  e k    E 11 0 x  E 12  0 y  E 13  0 xy  e k
k k k

y
N x  A 11  0 x  A 12  0 y  A 13  0 xy
x
Ny
Nx Txy
A 11  
n
k 1
 k
E 11  ek  A 12  
n
k 1
 k
E 12  ek  A 13 
n

 E 13  e k
k 1
k

Analysis of Layups
– Similarly for Ny and Txy
– The non zero Strains are written
as
Ny  A 21  0 x  A 22  0 y  A 23  0 xy
n
 k
A 2j   E 2j  ek
k 1
 z
T xy  A 31  0 x  A 32  0 y  A 33  0 xy
n
 k
A 3j   E 3j  e k
k 1

y
– So the Force - strain field is given x

by Ny
Nx Txy
 Nx   A 11 A 12 A 13    0x 
      Nx   A 11 A 12 0    0x 
 Ny    A 21 A 22 A 23    0y      

 T xy   A 31 A 32 A 33    0 xy   Ny    A 21 A 22 0    0y 
    
 T xy   0 0 A 33    0 xy 
   
n
k 1
 k
A ij   E ij  e k  A ji   0x 
Nx
 0y 
Ny
 0 xy 
T xy
h h h
Analysis of Layups
– Average stresses in terms of
strains are thus
Nx Ny T xy
 0x   0y   0 xy  z
h h h
  0x   A 11 A 12 A 13    0x 
  A  
A 23 
1
  0y 
 h  21 A 22   
 0 y
  0 xy   A 31 A 32 A 33    0 xy  y
   
x
1 n
 k ek 
h
 A ij    E ij  h  Ny
k 1   Nx Txy
– Where the proportion of each ply
in the layup is given by e k  pk
h
1 0 90 45  45
0 / 90 / 45 / 45  A ij  E ij  p 0  E ij  p 90  E ij  p 45  E ij  p  45
h
Analysis of Layups
– Thin plates under bending
z
moments
h h
2 2
Mx    x zdz My    y zdz
h h y
2 2
h
2 x
M xy  h  xy zdz My
Mx

2
Mxy Myx
Analysis of Layups
– In terms of displacements or
deflections
 2w0
 w 0  x   0 x  z
u  u 0  z x  x 2
 
 w 0 2w0
v  v 0  z  y   0 y  z
 y   y 2
w  w 0  2w0
   xy   0 xy  z  2
  x y
  x   0 x   Kx 
     
  y    0 y   z K y   x , y , z    m x , y   z  K x , y 
     
 
 xy   0 xy  K
 xy 
Analysis of Layups
– The bending stiffness can then be
developed as
 zk  k   k   k   
2  w0   E12  z 0 y  z 2  w 0   E13  z 0 xy  z 2 2  w 0
n 2 2 2
   dz 
Mx    
  E11  z 0 x  z
x 2   y 2   xy  
k 1 
 k 1
z         
zk  z k 1
 E1j zdz    E1j zdz

z k 1 zk
n  3

 k z k  z k 1  w 0
3 2
 3 3 2

k z k  z k 1  w 0 
k z k  z k 1
3 3

2w 0   
Mx  
  E 11  E 12  E 13 2
 

k 1    
3 2 3 2 3 x y
 x y 
2w0 2w 0 2w 0

n
z 3
 z 3k 1 
with 
k
M x   D11  D12  D13 2 D1j  E11 k
x 2 y 2 xy k 1
3
Analysis of Layups
– The bending stiffness can then be
developed as
 M x  D11 D12 D13   2w   M x  D11 D12 D13  Kx 

           
0
     x  2
     
 M y  D 21 D 22 D 23    2
w0   M y  D 21 D 22 D 23  Ky 
   

y 2      
           
M xy  D 31 D 32 D 33   2  2
w 0  M xy  D 31 D 32 D 33  K xy 
     xy       
     
n k z 3
k  z k3 1 
D ij  E ij
3
k 1
Analysis of Layups
– The aggregate bending is thus
k k k
 x  E 11 x  E 12  y  E 13  xy
k k k  k 2w0 k  w0
2
k 2w0 
x  E 11 0 x  E 12  0 y  E 13  0 xy  z  E 11  E 12  E 13 2 
  x 2
 y 2
x y 
 
 x   x ( membrane )   x ( bending )
x x
membrane bending
Analysis of Layups
– A coupling of bending and
tension also exists for
asymmetric layups
zk
k z 2
k  z k2 1 
 E ij zdz  E ij
2
z k 1
n k z 2
k  z k21 
B ij  E ij
2
k 1
2w 0 2w 0 2w 0

Mx  B11 0 x  B12  0 y  B13  0 xy  D11  D12  D13 2
x 2
y 2 xy
M x  B11 0 x  B12  0 y  B13  0 xy  D11K x  D12K y  D13 K xy

Analysis of Layups
– A coupling of bending and tension also exists for asymmetric
layups
zk
k z 2
k  z k2 1 
 E ij zdz  E ij
2
z k 1
n k z 2
k  z k21 
B ij  E ij
2
k 1
– So for applied moments the behavior would become

2w 0 2w 0 2w 0
Mx  B11 0 x  B12  0 y  B13  0 xy  D11  D12  D13 2
x 2
y 2 xy
M x  B11 0 x  B12  0 y  B13  0 xy  D11K x  D12K y  D13 K xy
– Similarly for Normal Forces

N x  A 11  0 x  A 12  0 y  A 13  0 xy  B11K x  B12K y  B13 K xy
Analysis of Layups
– The complete stiffness matrix would thus be written as
 Nx       0x 
       
N
 y   A B 
 0y 
  
 N xy          0 xy 
    
 Mx       K
 x
 My    B
 
D
  Ky 
       
M xy         K xy 
A ij 
n
E  k
 ek  n k z 2
k  z k21  n k z 3
k  z k3 1 
ij B ij  E ij
2
D ij  E ij
3
k 1 k 1 k 1
Analysis of Layups
– Explanation of each term Coupling Bending-Tension
Coupling Tension-Shearing
 N x   A11 A12 A13    x0   B11 B12 B13   k x 

      
 N y    A12 A22 A23    y0    B12 B22 B23   k y 
  
N  A B   0  N   A
 xy   13 A23  0 
A33   xy   B13 B23 B33   k xy 
   
M  B D   K   M x   B11 B12 B13    x0   D11 D12 D13   k x 
      
 M y    B12 B22 B23    y0    D12 D22 D23   k y 
  
M   B  0 
B33   xy   D13 D33   k xy 
 xy   13 B23 D23
Coupling Bending-Tension Coupling Bending-Torsion
 Nx   E 11 E 11 E 16    0x  E 11 E 11 E 16  kx  
n  zz zz

     0     
 Ny     E 12 E 22 E 26   y  dz   E 12 E 22 E 26  ky  zdz 
N  k 1  z k 1 
E 66   0  z k 1 
E 66  k  
 xy   E 16 E 26
k  xy  E 16 E 26
k  xy  
 Mx   E 11 E 11 E 16    0x  E 11 E 11 E 16  kx  
n  zz zz
      0      2  
 My     E 12 E 22 E 26   y  zdz   E 12 E 22 E 26  ky  z dz 
M  k 1  z k 1 
E 66   0  z k 1 
E 66  k  
 xy  
 E 16 E 26
k  xy  E 16 E 26
k  xy  

Failure in Composite Materials
– Modes of failure
• Failure in Plies
• Fiber Breakage
• Failure in Matrix
• Delamination
• Micro Buckling
• Fiber Pull out
– Theories of failure based on maximum stress and first ply failure
• Tsai Wu Criteria
• Puck Criteria
• Hill Criteria
2 2 2
 L   T   T'   1 1 1 
          2  2  2  L  T
 X   Y   Z  X Y Z 
 1 1 1   1 1 1 
  2  2  2  L  T '    2  2  2  T  T '
X Y Z   X Y Z 
2 2 2
     
  LT    LT '    TT '   1
 S LT   S LT '   S TT ' 
2
  LT 
2 2
 L   T   1 1 1 
     2  2  2  L T 
     1
 X   Y  X Y Z   S LT 
• Von Misses Stress is given by
• In composites the failure criterion would be

• After determining the constants of the polynomial the remaining equation

comes out to be
– Theories of failure based on stresses decomposed into the
matrix and fiber strengths
• Hashin damage Criteria
These are interacting failure criteria where more than one

stress components have been used to evaluate the different failure
modes. These criteria were originally developed for unidirectional
polymeric composites, and hence, applications to other type of
laminates and non-polymeric composites have significant
approximations. Usually Hashin criteria are implemented within two
dimensional classical lamination approach for point stress calculations
with ply discounting as the material degradation model. Failure indices
for Hashin criteria are related to fibre and matrix failures and involve
four failure modes. The criteria are extended to three dimensional
problems where the maximum stress criteria are used for transverse
normal stress component.
Tensile fibre failure for σ11 ≥ 0
2
  11   122   132   1 failure
   
 1 no failure
2
 XT  S12
Compressive fibre failure for σ11 < 0

2
  11    1 failure
  
 XC    1 no failure
Tensile matrix failure for σ22 + σ33 > 0
 22   33   122   2
2
 232   22 33   1 failure
  13

  1 no failure
2 2
Y T S 23 S122
Compressive matrix failure for σ22 + σ33 < 0
  Y  2           2  2     2
  2
  1 failure
   1     
C 22 33 22 33 23 22 33 12 13
 1 no failure
2 2 2
 2 S 23    YC  4 S 23 S 23 S12
Interlaminar tensile failure for σ33 > 0
2
  33    1 failure
  
 ZT    1 no failure
Interlaminar compression failure for σ33 < 0
2
  33    1 failure
  
 ZC    1 no failure
• Strain Energy Release Rate
» Application for the failure of materials
» Griffith Energy Based Approach
» When the crack grows a certain energy is released, using
the Ignis approach Grffith was able to calculate this
release rate
» This is calculated by considering β = π as calculated by

Ignis in the stress triangle.
» The surface energy S associated with crack of length a is
then given by
» Where ϒ is the surface energy in Joules/m2

» The total energy associated to the cracking is the –ive
energy of the crack surfaces being unloaded and the +ive
energy delivered to create the cracks.
» Setting the derivative to zero we can find the critical crack

length.
» For plasticly deforming materials a general strain energy

release rate coefficient is defined
» Compliance Calibration
» Half Cantilever Beam
– Theories of failure based Strain Energy Release
ENF
DCB
– Theories of failure based Strain Energy Release 


Mechanics of Composite Materials

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mechanics of Composite Materials

Uploaded by

Copyright:

Available Formats

Mechanics of Composite

– Difference between a composite and an alloy?

Longitudinal carbon fibers/epoxy Glass fiber web/epoxy Glass fiber rings

• Levels of analysis of Micromechanical

– Naturally materials are either ductile or brittle and such composite

– Wood is a composite – it is made from long cellulose fibres (a

Differential scanning calorimetry or DSC is a

• Mats are specified by their

– Therefore for bi-material or bi-phase composites

– Each hypothesis brings a

– Vair > 5% average quality composite

– For Transverse direction we have

Les propriétés du verre et de l’époxy:

Tissé Unidir 1 Unidir 2

Thickness e e k e (1-k) k  n1 n1  n 2 

• Symmetry of Strain :  kl =  lk c ijkl  c ijlk

 11   s 1111 s 1112 s 1113 2s 1114 2s 1115 2s 1116   11 

– And the Mechanical Behavior is written as

– With the strain as

– The elements II and III define the coupling behavior

Such coupling is absent in isotropic materials

– The elements II and III define

Such coupling is absent in isotropic

XYZ and X’Y’Z’

– The transformation can be calculated by

– The inverse of the matrix is its transpose

 ij'  Tki Tlj  kl   ij  Tik T jl  kl'

– And the rigidities can be transformed as

D1  2A   D1  A  

D1 t 2B t   D1 t B t 

C '  M C M t ou C   M t C 'M 

S '  M S M t ou S   M t S 'M 

E 13 = -cs {( c 2 E l - s 2 E t - ( c 2 - s 2 )( νtl E l + 2 Glt )}

E 23 = -cs {( s 2 E l - c 2 E t + ( c 2 - s 2 )( νtl E l + 2Glt )}

 Fiber Orientation

Qx,y S x,y Transformed Mathematical

Repère (1,2)  (L,T) E x,y Transformed Engineering Constants

– So the Force - strain field is given x

 E1j zdz    E1j zdz

2w0 2w 0 2w 0

 M x  D11 D12 D13   2w   M x  D11 D12 D13  Kx 

2w 0 2w 0 2w 0

M x  B11 0 x  B12  0 y  B13  0 xy  D11K x  D12K y  D13 K xy

– So for applied moments the behavior would become

M x  B11 0 x  B12  0 y  B13  0 xy  D11K x  D12K y  D13 K xy

– Similarly for Normal Forces

 N x   A11 A12 A13    x0   B11 B12 B13   k x 

• Von Misses Stress is given by

• In composites the failure criterion would be

• After determining the constants of the polynomial the remaining equation

These are interacting failure criteria where more than one

Tensile fibre failure for σ11 ≥ 0

Compressive fibre failure for σ11 < 0

Tensile matrix failure for σ22 + σ33 > 0

Compressive matrix failure for σ22 + σ33 < 0

Interlaminar tensile failure for σ33 > 0

» This is calculated by considering β = π as calculated by

» Where ϒ is the surface energy in Joules/m2

» Setting the derivative to zero we can find the critical crack

» For plasticly deforming materials a general strain energy

You might also like