Rheology 19 PDF

Rheology of polymers and
polymer composites
Composite Materials Engineering - A.Pegoretti - 2019 1

Content
1) Viscosity
2) Rheology of non-Newtonian fluids
3) Temperature and pressure dependency of viscosity
4) Rheological measurements (rheometry)
5) Rheology of particles-reinforced composites

1) Viscosity

Viscosity
Viscosity is a measure of the resistance of a fluid which is being deformed by either shear
stress or tensile stress.
As shown in the figure below, suppose that a liquid fills the space between two parallel plates of
area A each and gap h, the lower plate is fixed, and force F is needed to move the upper plate in
parallel at velocity V.
Whenever Vh/ν < 1500 (ν = η/ρ: kinematic viscosity), laminar flow is maintained, and a linear
velocity distribution, as shown in the figure, is obtained. Such a parallel flow of uniform velocity
gradient is called the Couette flow.

Viscosity
In this case, the force per unit area necessary for moving the plate, i.e.
the shearing stress (τ), is proportional to V and inversely proportional to h.
Using a proportional constant η, it can be expressed as follows:
F V
τ = =η
A h
Viscosity
The velocity distribution in the case where the fluid is flowing is as shown below:
Such a flow where the velocity u in the x direction changes in the y direction is called shear flow.
For such a flow, the shear stress τ on the section dy, distance y from the solid wall, is given by the
following equation: V dv
τ =η →τ =η
h dy
This relation was found by Newton through experiment, and is called Newton's law of viscosity.

Viscosity
The proportional constant η is called the viscosity, the coefficient of viscosity or the dynamic
viscosity. The units of viscosity are Pa s (pascal second) in SI, and g/(cm s) = Poise in CGS:
1 P = 0.1 Pa s
The value ν obtained by dividing viscosity η by density ρ is called the kinematic viscosity or the
coefficient of kinematic viscosity: 2
η $m ' In the CGS system of units
ν= & s ) 1 cm2/s is called 1 St (stokes)
ρ % (
2) Rheology of non-Newtonian fluids

Non-Newtonian fluids
Fluids are generally classified according to the relationship between the shearing stress and the velocity
gradient.
Rheological curves

For non-Newtonian fluids, an apparent viscosity ηa can be defined as the slope of the rheological curve:
dτ
ηa =
dγ

The law most frequently adopted to fit rheological curves of non-Newtonian fluids is the
power law equation:
τ = Aγ
n (Ostwald – de Waele relationship)
0.7
0.6
0.5
τ (MPa)
0.4
0.3
0.2
0.1
0.0
0 200 400 600 800 1000
dγ/dt (s-1)
n = 1 Newtonian fluid
n < 1 pseudoplastic fluid
ηa = Bγ m
n > 1 dilatant fluid m=n-1
Non-Newtonian fluids 0.8
0.6
log ηa (kPa s)
0.4
m
0.2
0.0
-0.2
-0.4
0 0.5 1 1.5 2 2.5 3
log dγ/dt (s -1)

The flux index m can be obtained from a linear regression of the data plotted in a
double logarithmic scale
log ηa = log A + m log γ

Several empirical equations have been proposed to fit rheological curves of non-Newtonian
fluids, such as:
Parabolic equation: τ = A + Bγ + Cγ 2
Logarithmic equation: τ = A + B log γ
B
Hyperbolic equation: τ = A+
γ
Steiger-Ory equation: τ = Aγ + Cγ 3
Casson equation: τ = A + B γ
Coefficients (A, B and C) are determined by fitting experimental data.

Till now (apparent) viscosity has been considered as a time-independent quantity.

Some non-Newtonian pseudoplastic fluids show a time-dependent change in viscosity.
In other words, rheological curves may display some hysteresis and, consequently,
(apparent) viscosity may depend on time.
Thixotropy: the longer the fluid undergoes shear stress, the lower its viscosity.
Rheopexy: the longer the fluid undergoes shearing force, the higher its viscosity
(less common).

γ = constant
Both behaviors are usually reversible under quiescent conditions.

3) Temperature and pressure
dependency of viscosity

Temperature dependency of viscosity
In the case of gases, increased

temperature makes the molecular
movement more vigorous and
increases molecular mixing so that
the viscosity increases.
In the case of a liquid, as its

temperature increases molecules
separate from each other, decreasing
the attraction between them, and so
the viscosity decreases.
The relation between the

temperature and the viscosity is thus
reversed for gases and for liquids.

For Newtonian liquids the viscosity manifests an Arrhenius type temperature dependency:
E
η(T ) = η(T0 )e RT
where T0 is a reference temperature and E is an activation energy.
T (°C)
0 20 40 60 80 100
2.0
1.5
ηa (mPa s)
1.0
0.5
0.0
280 300 320 340 360
T (K)
For non-Newtonian liquids an approach based on the specific free-volume (Vf) concept is
usually adopted:
V f = V − V0 Doolitle (1951)
where V is the specific volume experimentally measured and V0 is the specific volume
effectively occupied by the material. Doolitle considered V0 as the value of V at 0 K.
Doolitle verified that the viscosity of low-molecular weight hydrocarbons depends on the
fractional free volume ( f = Vf /V ) as follows:
b
f
η=ae

In 1950 Fox and Flory postulated a linear dependence of fractional free volume on the
temperature above Tg:
f (T ) = f (Tg ) + α f T − Tg ( )
In 1956 Williams-Landel and Ferry (WLF) proposed an expression which accounts for the
temperature dependence of the viscosity of non-Newtonian liquids:
log
η(T ) −C1 T − Tg
=
( )
η(Tg ) C2 + T − Tg ( )
b f (Tg )
where!!!!!C1 = and C2 =
! 2.303 f (Tg ) αf
The validity of this equation generally extends in the range Tg < T < Tg +100 °C

At the glass transition temperature, most polymers present similar αf and f(Tg) values:
αf = 4.84 x 10-4 K-1 ( range 2.9 ÷ 7 x 10-4 K-1 )
f(Tg) = 0.025 ( range 0.015 ÷ 0.036 )
b~1
14
12
This means that constants 10
logηa (Pa s)
C1 = 17.4 and C2 = 51.6 can be
considered as universal and that 8
most liquids have a similar viscosity 6

(about 1013 Pa s) at the glass
transition temperature. 4
0
0 20 40 60 80 100
T-Tg (K)
Pressure dependency of viscosity
An increased pressure causes a decrease in the free volume. Therefore, an increase in Tg is
observed when pressure is increased.
dTg
Tg = Tg0 + β ( P − P0 ) β=
dP
β = 0.020 ± 0.006 K/atm

Pressure dependency of viscosity
Therefore, a pressure-viscosity
relationship can be established by
inserting a pressure dependent glass
transition temperature into WLF
equation:
η (T, P) −C1 ⎡⎣T − Tg0 − β ( P − P0 )⎤⎦

log =
η (T, P0 ) C2 + ⎡T − Tg0 − β ( P − P0 )⎤
⎣ ⎦

4) Rheological measurements
(rheometry)

Rheometry
In measurements of viscosities, several flow geometries can be used, the most important
being capillary, concentric cylinders (Searle or Couette), cone-plate and plate-plate
geometries.

Rheometry
The type of geometry used depends on the shear rate at which one needs to measure the viscosity.

Capillary rheometry
This procedure, based on the relationship between pressure drop and flow rate, was
independently developed by Hagen and Poiseuille. It is one of the most widely used
techniques for studying the flow behavior of fluids.

Capillary rheometry
Let us assume that a Newtonian liquid flows through a capillary tube of radius R and length ΔL.
Once steady-state conditions are reached, the stresses acting on a section can be represented as:
A simple balance of forces gives:

r dP r ΔP
2 2
(P + dP) π r = P π r + 2 π r τ (r) dz τ (r) = =
2 dz 2 ΔL

Capillary rheometry
The shear stress increases as r increases, reaching a maximum value at the wall of the capillary tube.
The value of the shear stress at the wall is given by:
R ΔP
τw =
2 ΔL
Now, for a Newtonian fluid we can write:
dV (r) ΔP r
τ (r) = η γ = η where V(r) is the fluid axial velocity dV (r) = dr
dr η ΔL 2
Therefore:
ΔP
V (r) =
4 η ΔL
[ R2 − r 2 ]
This expression indicates that the fluid presents a distribution of velocities whose profile
corresponds to a paraboloid of revolution; the velocity of the fluid reaches its maximum
value [Vmax] at the center of the capillary and 0 at the capillary wall.

Capillary rheometry Vmax =
ΔP
R2
4 η ΔL
Speed profile for a Newtonian fluid
" r2 %
V (r) = Vmax $1− 2 ' r = 0 → γ = 0
# R &
r
The shear rate can be now γ = −2 Vmax
computed as: R2
Vmax
r = R → γ w = −2
dV (r) R
γ =
dr

Capillary rheometry
The shear rate can be correlated to the volumetric mass flow Q through the capillary.
In fact, the flow of the fluid through an infinitesimal cylinder defined by r and r + dr is: dQ = 2 π r V(r) dr.
r+dr
Then, the volumetric flow rate can be written as:

r
R R R
$ r2 '
Q = ∫ dQ = 2π ∫ rV (r)dr = 2π Vmax ∫ r &1− 2 ) dr
0 0 0 %
R (
R2
Therefore: Q = π Vmax
2
r 4Qr
The shear rate is given by: γ = −2 Vmax γ = −
R2 π R4
4Q
The maximum shear rate value is reached at the wall (r=R): γ w =
π R3

Capillary rheometry
For a non-Newtonian fluid the velocity profile is not parabolic. In fact, assuming a power-law
rheological behavior, we can write:
n 1/n
# dV (r) & " ΔP %
τ (r) = A γ n = A % dV (r) = $ r1/n dr
$ dr (' # 2A ΔL '&
Therefore:
n " ΔP % (n+1)/n (n+1)/n
V (r) =
1+ n $# 2 A ΔL '&
[R − r ]
In other terms we can write:
( " r % (n+1)/n +
V (r) = Vmax *1− $ ' -
# R&
) ,
Where n is the flux index

Capillary rheometry
Following the same procedure as before, it is possible to show that the shear rate at the wall for a
non-Newtonian liquid is related to the volumetric mass flow through the following equation:
3n + 1 3n + 1 # 4 Q & 3n + 1
γ w,NN = 3
Q= % 3(
= γ w
nπ R 4n $π R ' 4n
3.5
3.0
2.5
2.0
The term (3n+1)/4n
is called 1.5
Rabinowitsch correction
1.0
0.5
0 0.5 1 1.5 2
n
Capillary rheometry
Another important phenomenon to consider is
the pressure losses at the capillary entrance.
Bagley correction
R ΔP − ΔPE
τw =
2 ΔL

Capillary rheometry
Reynolds number
The Reynolds number is a dimensionless parameter defined as
ρV D Where: ρ is the fluid density in kg/m3

Re = V fluid velocity in m/s
η D pipe diameter in m
η fluid viscosity in Pa s
Physically, the Reynolds number represents the ratio of inertial forces to viscous forces.
The Reynolds number helps us to predict the transition between laminar and turbulent flows.
Laminar flow is highly organized flow along streamlines. As velocity increases, flow can become
disorganized and chaotic with a random 3-D motion superimposed on the average flow velocity. This is
known as turbulent flow.
Laminar flow occurs in flow environments where Re < 2000.
Turbulent flow is present in circumstances under which Re > 4000.
The range of 2000 < Re < 4000 is known as the transition range.

Capillary rheometry
In a regime of turbulent flow the phenomenon of melt fracture may occur.

Rotational rheometry: coaxial cylinders
There are two types of coaxial cylinders rheometers: Searle and Couette models.
Searle model Couette model M

τ=
2 π Rb2 h
2 ω Rb2 Rc2
γ = 2 2 2
r Rc - Rb
Where:
ω = angular velocity
r = radial coordinate
Rb = radius of inner cylinder (bob)
Rb = radius of outer cylinder (cup)
M = torque
h = height of the cylinders

The shear rate is zero at the stationary wall and maxima at the rotating wall:
Searle model Rc
r
Rb
Rc2
Searle r = Rb → γ w = 2 ω 2 2
Rc - Rb
2 ω Rb2 Rc2
γ = 2 2 2
r Rc - Rb
Rb2
Couette r = Rc → γ w = 2 ω 2 2
Rc - Rb

dV ω dr ω ω
γ = = = 
dh dr tan α tan α α
If α small
(usually 1° < α < 5°)
ω 3 M
γ  τ=
α 2 π r3
It is important to observe that the shear rate is constant !

Rheology of particles-reinforced composites
To study the effect solid particles on viscosity of fluid (polymer melts) it is necessary to
deal with the rheological behaviour of suspensions.
If η is the viscosity of the suspending fluid (matrix), ηc the viscosity of the suspension, and
Φ the volume fraction of the solid phase, the following parameters can be defined:
!$
1) Relative viscosity !" =
!
ηc − η
2) Specific viscosity ηsp = = ηr −1
η
3) Inherent viscosity
ln ηr
ηi =
Φ
ηsp
4) Intrinsic viscosity [η ] = lim
Φ→0 Φ

A conversion from the weight (Wf) to the volume fraction (Vf) of fibers can be performed as:
1
Vf =
ρf ⎛ 1 ⎞
1 + ⎜ − 1⎟
ρm ⎝ W f ⎠
For a WPC the following densities can be estimated for polymer matrix (rm)
Polymer Density (g/cm3)

Polyethylene, LDPE 0.89 – 0.92 On the other hand, the
Polyethylene, HDPE 0.94 – 0.97 density of wood filler (rf) can
Polypropylene, PP 0.90 – 0.94 change depending on the

Polyvinylchloride , PVC 1.40 – 1.45 wood type and water content
Polystyrene , PS 0.96 – 1.04 from 0.3 to 1.2 g/cm3.
Epoxies 1.15 – 1.25
Phenol formaldehyde resins (PF) 1.16 – 1.21
Assuming a dilute dispersion of uniform, rigid, non-interacting spheres, Einstein* in 1906
derived an equation expressing the increase in viscosity of the dispersion as:
4.0
ηc = η (1+ 2.5Φ )
3.5
3.0
η/η0
2.5
2.0
* the first paper (1906) reported the front factor 1.5

of the volume fraction to be 1. Einstein corrected
this in 1911.
1.0
0 0.2 0.4 0.6 0.8 1

The Einstein equation can be adopted only for very low filler concentrations.

The Einstein equation has been modified to exted its predition to higher volume fractions.
In particular, the viscosity relation is generally expressed as a polynomial series in Φ:
ηc = η (1 + 2.5Φ + kΦ 2 + ...)
Which can be rearranged as:
ηc − η
= 2.5 + kΦ + ...
ηΦ

ηc − η
= 2.5 + kΦ + ...
ηΦ
The departure from the Einstein equation is due to hydrodynamic interactions between the
particles and to other interparticle forces.
The prediction capabilities of Einstein equations is therefore not enough to be used for
particle reinforced composites for which filler volume fraction as high as 30% can be easily
reached.
Thomas and Muthukumar* considered full hydrodynamic interactions between three
hard spheres and derived, using the multiple scattering theory:
!" = ! 1 + 2.5) + 4.83) - + 6.43) /
However, each successive term extends the equation's applicability to only slightly higher F.
* C.U. Thomas and M. Muthukumar. Three-body hydrodynamic e®ects on viscosity of suspensions of spheres. J. Chem. Phys., 94:5180, 1991.
Another model to be mentioned for the effective viscosity of Newtonian suspensions of

monosized spheres is the one derived by Hsueh and Becher**:
!" = ! 1 + 2.5) + 6.25) - + 10) / + 13.75) 2 + 17.5) 3 + 21.25) 4 + ⋯
** C.H. Hsueh and P.F. Becher. Effective viscosity of suspensions of spheres. J. Am. Ceram. Soc., 88:1046, 2005.

The most widely used equation for concentrated suspensions was formulated by
Dougherty and Krieger* (or equivalent formulas, e.g. Quemada, Maron-Pierce or
Mooney):
().+ ∅-
&
!" = ! 1 −
&'
Where Fm is an adjustable parameter related to the volume fraction at which spheres

become close packed. Typical values for Fm range from 0.6 to 0.7 for monodisperse
spheres. The -2.5 used in the exponent is used in order to re-obtain the Einstein's
solution when F → 0.
J.M. Krieger and A. Dougherty. A mechanism for non-newtonian flow iin suspensions of rigid spheres. Tran. Soc. Rheol., 3:137, 1959.

In the following figures the predictions of existing solutions for the normalized effective
viscosity of suspensions of monodisperse hard spheres are compared.

For studying the rheological behaviour of viscoelastic suspensions the dynamic storage (G’)
and loss (G”) shear moduli are often determined at a constant temperature T and at
various frequencies w.
The materials used for the study contain 30%, 50%, and 70% of white fir fibers L=170 µm, D= 16 µm) .

' '
$′ $”
A complex viscosity can be defined as: !∗ = +
& &

Shear viscosity of HDPE filled with various weight percentages of 40 mesh maple with and
without maleated polyethylene (MAPE)

Rheology 19 PDF

Uploaded by

Copyright:

Available Formats

Rheology 19 PDF

Uploaded by

Document Information

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

Rheology 19 PDF

Uploaded by

Copyright:

Available Formats

Rheology of polymers and

Composite Materials Engineering - A.Pegoretti - 2019 1

2) Rheology of non-Newtonian fluids

3) Temperature and pressure dependency of viscosity

4) Rheological measurements (rheometry)

5) Rheology of particles-reinforced composites

Composite Materials Engineering - A.Pegoretti - 2019 2

Composite Materials Engineering - A.Pegoretti - 2019 3

Composite Materials Engineering - A.Pegoretti - 2019 4

Composite Materials Engineering - A.Pegoretti - 2019 6

Composite Materials Engineering - A.Pegoretti - 2019 8

Composite Materials Engineering - A.Pegoretti - 2019 9

Composite Materials Engineering - A.Pegoretti - 2019 10

log dγ/dt (s -1)

Composite Materials Engineering - A.Pegoretti - 2019 12

Composite Materials Engineering - A.Pegoretti - 2019 13

Parabolic equation: τ = A + Bγ + Cγ 2

Logarithmic equation: τ = A + B log γ

Steiger-Ory equation: τ = Aγ + Cγ 3

Coefficients (A, B and C) are determined by fitting experimental data.

Composite Materials Engineering - A.Pegoretti - 2019 14

Till now (apparent) viscosity has been considered as a time-independent quantity.

Composite Materials Engineering - A.Pegoretti - 2019 15

Both behaviors are usually reversible under quiescent conditions.

Composite Materials Engineering - A.Pegoretti - 2019 16

Composite Materials Engineering - A.Pegoretti - 2019 17

In the case of gases, increased

In the case of a liquid, as its

The relation between the

Composite Materials Engineering - A.Pegoretti - 2019 18

where T0 is a reference temperature and E is an activation energy.

Composite Materials Engineering - A.Pegoretti - 2019 20

Composite Materials Engineering - A.Pegoretti - 2019 21

αf = 4.84 x 10-4 K-1 ( range 2.9 ÷ 7 x 10-4 K-1 )

f(Tg) = 0.025 ( range 0.015 ÷ 0.036 )

This means that constants 10

most liquids have a similar viscosity 6

β = 0.020 ± 0.006 K/atm

Composite Materials Engineering - A.Pegoretti - 2019 23

η (T, P) −C1 ⎡⎣T − Tg0 − β ( P − P0 )⎤⎦

Composite Materials Engineering - A.Pegoretti - 2019 24

Composite Materials Engineering - A.Pegoretti - 2019 25

Composite Materials Engineering - A.Pegoretti - 2019 26

Composite Materials Engineering - A.Pegoretti - 2019 27

Composite Materials Engineering - A.Pegoretti - 2019 28

A simple balance of forces gives:

Composite Materials Engineering - A.Pegoretti - 2019 29

Now, for a Newtonian fluid we can write:

Composite Materials Engineering - A.Pegoretti - 2019 30

Composite Materials Engineering - A.Pegoretti - 2019 31

Then, the volumetric flow rate can be written as:

Composite Materials Engineering - A.Pegoretti - 2019 32

In other terms we can write:

Where n is the flux index

Composite Materials Engineering - A.Pegoretti - 2019 33

Composite Materials Engineering - A.Pegoretti - 2019 35

ρV D Where: ρ is the fluid density in kg/m3

Laminar flow occurs in flow environments where Re < 2000.

Turbulent flow is present in circumstances under which Re > 4000.