Viscoelastic Fluid

Related terms:

Slurry, Shear Thinning, Viscous Fluid, Mucus, Deformation, Hydrodynamics, Vis-

cosity

View all Topics

Learn more about Viscoelastic Fluid

Introduction to Wave Propagation

Vimal Saxena, ... Ludmila Adam, in Handbook of Borehole Acoustics and Rock
Physics for Reservoir Characterization, 2018

> Read full chapter

Specialty Polymers & Polymer Process-

ing
Joseph Zimmerman, in Comprehensive Polymer Science and Supplements, 1989

8.3.3 Draw Resonance

As we have mentioned above, isothermal spinning can lead to severe instabilities
and large fluctuations in fiber diameter when some critical attenuation factor is
reached and when there is a discontinuity introduced in the form of rapid quench
or solidification of some sort. For Newtonian fluids, theoretical analysis104–106 has
indicated that the critical attenuation factor is about 20. For power law fluids,
theoretical analysis107,108 has predicted that the critical value of AF will decrease as
n decreases. Examples are AF = 2.8 for n = 0.3 and 4.8 for n = 0.5. Cooling of the
threadline tends to stabilize it against draw resonance110 since much higher values of
AF are achieved in commercial spinning. On the other hand, some studies indicate
that cooling can increase the severity of draw resonance in viscoelastic fluids (see
ref. 111 for a discussion). It is fortunate that, in the air gap spinning process used
for spinning of p-aramids, relatively low stretch factors are needed to achieve high
orientation so that draw resonance can be avoided.

In one experimental study of draw resonance,110 the critical draw down ratio was
about 13 for polypropylene and the severity of the diameter oscillations increased
more rapidly with increasing draw down when the air gap was 12 cm compared
to 6 cm. In another study112 the critical draw down ratio was about ten. There are
some data113,114 indicating that the power law exponent n for elongational flow of
PP is about 0.4–0.5. One of these studies113 shows an effect of spinneret capillary
dimensions on the critical stretch ratio for draw resonance with longer L/D ratios
giving higher critical ratios. In another study,115 the critical stretch ratio was also
affected by capillary flow variables and reached values as low as five, the theoretical
value for n = 0.5.109 Finally, in the spinning of H2SO4 solutions of PPD-T,116 the
critical jet stretch for draw resonance was found to increase from about 12 to 21
as the size of the air gap was reduced from 2 cm to 0.35 cm. Effects of capillary
dimensions were also observed. If there was a bulge formed below the spinneret
under the spinning conditions used, the critical attenuation factor would have been
larger than the numbers quoted for critical draw ratio.

In certain cases, it is possible to encounter severe draw resonance even when there
is no external rapid quench. Thus it has been found that high molecular weight
liquid crystalline polyarylates, which spin well at low molecular weights, manifest
draw resonance at quite low spin stretch factors and this represents a formidable
barrier to realizing the tensile strength potential of these polymers.77 It was found
that proper selection of polymer composition and spinning below the melting point
(but above the freezing point) of the polymer eliminates draw resonance and permits
attainment of higher strength fibers.

> Read full chapter

Free and mixed convection bound-

ary-layer flow of non-Newtonian fluids
Ioan Pop, Derek B. Ingham, in Convective Heat Transfer, 2001

10.1 Introduction
It is well known that most fluids which are encountered in chemical and allied
processing applications do not adhere to the classical Newtonian viscosity pos-
tulate and are accordingly known as non-Newtonian fluids. One particular class
of materials which are of considerable practical importance is that in which the
viscosity depends on the shear stress or on the flow rate. Most slurries, suspensions
and dispersions, polymer solutions, melts and solutions of naturally occurring
high-molecular-weight, synthetic polymers, pharmaceutical formulations, cosmet-
ics and toiletries, paints, biological fluids, synthetic lubricants and foodstuffs, exhibit
complex rheological behaviour which is not experienced when handling ordinary
low-molecular-weight Newtonian fluids such as air, water, silicon oils, etc. Due to the
importance of the applications of non-Newtonian fluids for the design of equipment
and in industrial processing, considerable efforts have been directed towards the
analysis and understanding of such fluids. Non-Newtonian fluid behaviour has
been the subject of recent books by Astarita and Marrucci (1974), Schowalter (1978),
Crochet et al. (1984), Tanner (1985), Bird et al. (1987) and Siginer et al. (1999). Further,
a fairly large body of fundamental research on non-Newtonian fluid flow can also
be found in a number of excellent review articles, e.g. Cho and Hartnett (1982),
Shenoy and Mashelkar (1982), Crochet and Walters (1983), Shenoy (1986), Irvine, Jr.
and Karni (1987), Andersson and Irgens (1990) and Ghosh and Upadhyay (1994).

Real fluids and their mathematical models are classified into the following three
types, see Andersson and Irgens (1990):

(i) Time-independent fluids for which the properties are independent of time;

(ii) Time-dependent fluids for which the properties change with time as the fluid
is deformed;
(iii) Viscoelastic fluids which exhibit both viscous and elastic behaviour, e.g. elastic
recovery after deformation, and stress relaxation.

The time-independent fluids may be subdivided into four types, depending on the
general nature of the viscosity function, as follows:

(10.1)

where is the shear rate and is the shear stress. The four types are as follows:

(i) Viscoplastic fluids (e.g. Bingham-fluids);

(ii) Pseudo-plastic or shear thinning fluids;

(iii) Dilatant or shear thickening fluids;

(iv) Newtonian fluids.

In dealing with the complexities of non-Newtonian fluid flows, methods that allow
the description, interpretation and correlation of fluids properties are required. A
number of mathematical models and techniques have been proposed to describe
the rheological behaviour of such fluids, see for example Rosen (1979) and Bird et
al. (1987). Some empirical models have been found to correlate the viscosity data
adequately for various types of material through the use of a limited number of
meaningful parameters. Despite the trend to develop constitutive theories through
the application of continuum theories, simple models have been developed which
describe the non-Newtonian behaviour of fluids which have useful applications in
industry, see for example Cramer and Marchello (1968). Ideally, a simple model
should give an accurate fit with all the available data and have a minimum number
of independent constants which can be easily evaluated and they have some physical
basis.

Such a model which has been most widely used for non-Newtonian fluids, and is fre-
quently encountered in chemical engineering processes, is the empirical Ostwaldde
Waele model, or the so-called power-law model, defined as follows:

(10.2)

where and n are material parameters, is called the consistency coefficient and n is
the power-law index with n being non-dimensional and the dimension of depends
on the value of n. The quantity is not the viscosity in a classical sense unless n is
unity. The parameter n is an important index to subdivide fluids into pseudo-plastic
fluids n < 1 (most macromolecular fluids are of this kind with 0.2 < n < 0.6, see
Bird et al., 1987) and dilatant fluids n > 1. Clearly, the power-law model described by
Equation (10.2) reduces to the Newtonian model when n = 1 and then the consistency
coefficient is the dynamic viscosity. Details of this class of fluids can be found in
the review articles by Metzner (1965) and Andersson and Irgens (1990).

Other important non-Newtonian fluids are those fluids which contain certain ad-
ditives and some naturally occurring fluids such as animal blood. Physically these
fluids may form suitable non-Newtonian fluid models which can be used to analyse
the behaviour of exotic lubricants, colloidal suspensions, liquid crystals, etc. A math-
ematical model for the description of such fluids, which exhibit certain microscopic
effects arising from the local structure and microrotations of the fluid elements
is that of a microfluid, first introduced by Eringen (1966). As this model is not
easily amenable to theoretical treatment, a subclass, known as micropolar fluids,
was further proposed by Eringen (1972) and such fluids exhibit the micro-rotational
inertia. They can support couple stresses and body couples only, and may represents
fluids consisting of bar-like or sphere-like elements. The theory of micropolar fluids
has generated a considerable amount of interest and many flow problems have been
studied, see for example Ariman et al. (1973), Pop et al. (1998c) and Rees and Pop
(1998) for detailed references.

However, more theoretical and experimental work is still required in the area of
convective flow of non-Newtonian fluids for both power-law and micropolar fluids.
> Read full chapter

Polymer Properties
Jean-Francois Joanny, S. Jean Candau, in Comprehensive Polymer Science and
Supplements, 1989

7.1 Introduction
Many of the industrial applications of polymers are related to their unusual hy-
drodynamic and rheological behavior; a common way of increasing the viscosity
of a liquid (e.g. an oil) is to introduce a polymeric additive. The relaxation times
associated with the internal motions of polymers in solution (or in the melt) are much
longer than the relaxation times for simple molecules, and can be as long as several
seconds. This, in turn, leads to very high viscosities for polymer solutions and very
slow motions of the macromolecules. The response of polymer solutions to strong
hydrodynamic forces is even more unusual, and characteristic of non-Newtonian
viscoelastic fluids. This non-Newtonian behavior, related to the internal elasticity of
polymer chains, is the basis of such properties as drag reduction or shear thinning
of polymer solutions.

In this chapter we describe the hydrodynamic properties of polymer solutions close

to thermodynamic equilibrium, but ignore all the fascinating non-linear effects. We
will also limit ourselves to neutral linear flexible macromolecules, although very
similar theories have been applied to more complicated systems such as polyelec-
trolytes, branched polymers and rod-like polymers.

The classical approach to hydrodynamic properties of polymer solutions is described

in detail in the books of Ferry,1 Bird et al.,2 Flory3 and Yamakawa.4 The non-draining
Zimm description of dilute solutions is quite consistent with experimental results
obtained by viscometry, concentration gradient diffusion (CGD) measurements
and sedimentation. This classical approach, however, is not equally consistent for
semi-dilute solutions.

In recent years, our understanding of hydrodynamic properties has been consid-

erably improved by the development of new experimental techniques, the most
important of these being quasi-elastic light scattering5–6 (QELS), which is described
in detail in Volume 1, Chapter 8 of this work. In a QELS experiment one measures the
time correlation function of the intensity of light scattered by the solution for a given
transfer wavevector q = [(4πn)/ ] sin /2, where n is the refractive index of the medium,
is the wavelength of the scattered light and is the scattering angle. This intensity
correlation function is directly proportional to the time correlation function of the
concentration; at low wavevectors it allows investigation of the diffusion constant of
the solution; at higher wavevectors it probes the internal motions of polymer chains.
QELS has now become one of the easiest ways of characterizing a polymer solution
(or any colloidal system) by its diffusion constant (or hydrodynamic radius).

Some more refined techniques have been developed to probe length or time scales
which are not accessible by QELS. Amongst these are neutron spin echo (NSE)
scattering,7 which has allowed hydrodynamic measurements at the shorter length
scales inaccessible by light-scattering, also pulsed field gradient nuclear magnetic
resonance,8 forced Rayleigh light-scattering9 and photobleaching fluorescence re-
covery,10 which have allowed a detailed measurement of self-diffusive motions of
polymer molecules, chemically labelled with fluorescent dyes, in more concentrated
solutions and in the melt.

On the theoretical side, the introduction of scaling laws,11 by analogy with critical
phenomena, has provided a more complete description of dilute and semi-dilute
polymer solutions. Most physical properties of polymer solutions can, in certain
asymptotic limits, be expressed as power laws of a few physical parameters, such
as the degree of polymerization N, the concentration c or the shift from the Flory
compensation temperature (T −  )/ . The exponents of these power laws are uni-
versal, in the sense that they do not depend explicitly on the chemical nature of the
polymer and the solvent but on the fact that the polymer-solvent pair belongs to
a certain universality class: linear polymers in a good solvent, randomly branched
polymers in a solvent. Very few exact results are known for these exponents: a
very useful, and usually very good, approximation is obtained by using a mean field
Flory approximation. A more systematic approach uses the renormalization group
theory, which provides expansions of these exponents. The critical exponents are
not the onlyuniversal quantities in a polymer solution; some combinations of the
amplitudes of the power laws are also universal numbers which can be determined
both experimentally and theoretically. More generally, physical quantities can be
expressed as scaling functions of some reduced variables, these scaling functions
becoming power laws in certain asymptotic limits.

The scaling law approach, and renormalization group calculations, have been exten-
sively developed for the static properties of polymer solutions. Many exponents and
universal ratios are known very accurately, but the determination of full scaling func-
tions is much harder and one often needs some reexponentiations (non-universal)
to get results which are comparable with experiment.

The development of scaling laws for dynamic properties has been pioneered by de
Gennes11 and seems to provide a good description of hydrodynamic properties of
dilute and semi-dilute solutions. Systematic dynamic renormalization calculations13
have been performed for dilute solutions. Such calculations, however, do not seem to
be as systematic for semi-dilute solutions, where further approximations are needed.
Nevertheless, these approximations appear to yield quantitatively good results.

We shall use the dynamical scaling theory to describe the hydrodynamic properties
of polymer solutions, focusing mainly on the expected universal behavior. We use a
Flory approximation for the power law behavior, which turns out to be a much easier
approach and allows a simple understanding of the important physical features often
masked by a heavier formalism. For comparison with experiment we shall sometimes
quote more detailed results obtained by renormalization group calculations. We
will also discuss briefly the deviations from universal behavior related to crossover
effects.

Scaling theory suggests using reduced variables (which are combinations of mea-
surable physical quantities) for the presentation of experimental results. This allows
elimination of the non-universal behavior due to the specific chemical nature of
the polymer or solvent, which is not usually well known, and leads to universal
scaling functions which contain more information than the determination of a single
exponent. We shall use such a presentation as often as possible in this chapter.

In Section 7.2 we describe briefly the static scaling laws for polymers both in
good and solvents. Section 7.3 is devoted to the discussion of the hydrodynamic
properties of dilute solutions, which are often used to characterize polymers. The
hydrodynamic properties of semi-dilute solutions are divided into two groups;
collective properties and single chain properties, which are described in Sections 7.4
and 7.5 respectively.

> Read full chapter

Specialty Polymers & Polymer Process-

ing
A.I. Isayev, in Comprehensive Polymer Science and Supplements, 1989

11.4 Rheology of Rubber Compounds

Rubber compounds are elastomers usually highly filled with carbon black. Similar
to two-phase particulate-filled thermoplastic melts,31–38 they are non-Newtonian
viscoelastic media with inherent yield-stress.39–44 Rubber compounds are highly
thixotropic in nature. The yield stress and non-Newtonian and thixotropic behavior
are dependent upon the concentration of the carbon blacks, their activity, particle
size and their distribution and shape.32,33,41 Due to the presence of the yield stress,
rubber compounds show solid-like behavior at stresses below the yield stress and
viscoelastic-fluid behavior above it. In particular, Figure 2 shows a dependence of
viscosity on shear stress for styrene–butadiene rubber (SBR) filled with carbon black.
The rheological behavior observed in this figure is typical of rubber compounds.
At low shear-stresses, the gum SBR exhibits a constant shear-viscosity, while car-
bon-black-filled SBR shows an unbounded value of the shear viscosity, indicative of
the presence of a yield value. At high shear-stresses, the compounds and the gum
rubber show a decrease of viscosity, the behavior typical of a non-Newtonian fluid.

Figure 2. Viscosity–shear-stress behavior of SBR 1 500 filled with carbon black N326
at various concentrations, , at 100 °C: ○,  = 0.00; Δ,  = 0.10; ,  = 0.20; ,  = 0.30
(from ref. 79)

There have been some efforts towards developing a constitutive equation to describe
the observed rheological behavior in two-phase systems. However, these attempts
have so far led to a qualitative description of the real behavior of rubber compounds.
In particular, development of a fluid model with yield stress has a long history.45–48
By combining the Von Mises yielding criterion with the Navier–Stockes Newton-
ian-fluid equation of motion, a three-dimensional Bingham fluid model has been
developed.49–51 This theory has been improved by introducing viscosity as a function
of the invariants of the rate of deformation tensor,52 in order to accommodate
non-Newtonian behavior after yielding. Further, the time-dependent yield value
has been introduced53 to explain thixotropic behavior. An attempt54 has also been
made to extent Oldroyd’s plastic non-Newtonian viscous constitutive equation52 to
describe a plastic material exhibiting viscoelastic behavior beyond the yield value. In
recent years, there have been further theoretical and experimental studies of the rhe-
ological behavior of particle-filled melts exhibiting yield behavior, viscoelasticity and
thixotropy.55–60 Hutton’s yield-viscoelastic constitutive equation54 has been extended
by including explicit expressions for the time-dependent yield-stress and non-linear
memory function. The White model55–57 has been tested on several particulate-filled
polymers, but in a narrow range of strain rates.

In some rubber compounds and pure elastomers a further complication arises due
to the presence of the slip effect during flow along the solid boundary, such as
occurs in processing equipment.43,62–65 Typically, the slip velocity for a particular
rubber compound depends on temperature, shear stress, shear rate and pressure.66
In particular, the presence of hydrostatic pressure reduces slip velocity, while an
increase in shear stress and temperature leads to an increase in slip velocity. As
an example, Figure 3 shows data obtained on an EPDM-based rubber compound
(EPDM is ethylene–propylene diene monomer). Slip velocity can reach values as
high as 10 cm s −1 at high shear-stress. The occurrence of slip effects in processing
equipment, e.g. in injection molding, may lead to several undesirable phenomena
such as non-homogeneity of the compound, poor mixing or unstable feeding. At the
same time, the slip effect can be beneficial in reducing the pressure needed to fill the
mold or push a compound through the dies. In the capillary rheometer, the presence
of the slip effect leads to a dependence of flow curve on the capillary diameter.
Namely, flow curves obtained in capillaries of a smaller diameter are shifted toward
lower shear-stresses or higher shear-rates in comparison with those obtained in
capillaries with a large diameter. Although the problem of slip has a long history,62
it is still not well understood, especially for the case of injection molding of rubber
compounds.
Figure 3. Slip velocity vs. shear stress for an EPDM-based rubber compound (from
ref. 66)

As a processibility indicator, many rubber molders rely on the Mooney viscosity.2

However, the Mooney viscosity is not indicative of the complicated rheological
behavior of rubber compounds, since it determines viscosity at one particular
shear rate. In injection molding the flow curve measured over a wide range of
shear rates is required.67 Shear rates which rubber compounds experience in the
molding system can be as low as a few reciprocal seconds and as high as tens of
thousands. Obviously, viscosity determined at one shear rate, such as the Mooney
viscosity, cannot characterize the viscous behavior of a rubber compound.5,9,67 In
this regard, the use of capillary rheometers can be quite beneficial for the evaluation
of the processibility of rubber compounds.31,68 In particular, the Instron capillary
rheometer has been employed for these purposes.4,67,69 The recent development of
the Monsanto processibility tester, which has been specifically designed for rubber
compounds, has provided an important tool for measuring the viscous properties
of rubber compounds under injection molding conditions.9,70–73 In addition, the
development of the capillary or slit rheometer on the base of injection molding
screws is another alternative.74,75 In this case, a capillary or a slit die with pressure
transducers installed along its length can be attached to the injection screw. The
injection unit can be used to push rubber compounds through the die at various
velocities thus giving a range of high shear-rates realized during molding. The
experiments can be conducted under isothermal and non-isothermal conditions.
Thus, such a rheometer would allow measurement of the flow curves at various
temperatures. In this case, even a change of flow curve during the early stage
of vulcanization can be detected. The latter is highly important for cavity-filling
simulation.

A simple rheological model which has been employed in order to simulate the
injection molding process is a power-law model which is given by the following
expression for the apparent viscosity

(1)

where A(T) is called the consistency and n is the power-law index. The value of A is
usually based on the Arrhenius-type temperature dependence given by

(2)

where B and Tb are material constants.

For many rubber compounds, the power-law model is known to be a good approx-
imation in the high shear-rate region only. Thus, a more general form of the rheo-
logical equation is needed. Several equations are employed for injection molding of
rubber compounds to fulfill this requirement. In particular, the Rubbers and Plastics
Research Association (RAPRA)76 uses a so-called Klein viscosity equation,19 which is
a second-order function of temperature and shear rate, namely

(3)

Here, Aij are material constants. The drawback of this model is that it predicts a
maximum viscosity at some low shear-rate and a minimum viscosity at some finite
value of temperature. In addition, this viscosity function shows crossover points
in flow curves corresponding to different temperatures. Thus, in order to correctly
apply this viscosity model, a careful consideration should be given to the range of
temperatures and shear rates over which it is applied.

A more general rheological equation, which does not exhibit the above-mentioned
deficiencies, is the so-called modified Cross model given by the following expres-
sion77,78

(4)

where

(5)

is the zero shear-rate viscosity. This equation can be applied to a wide range of shear
rates and includes four material parameters, namely A, Tb, * and n. In particular,
it has been widely used in the simulation of injection molding of thermoplastics77
and, most recently, of rubber compounds.73 However, even the latter equation is
not applicable to those rubber compounds which exhibit yield stress leading to an
unbounded value of viscosity at low shear-rates. In order to correct this deficiency,
the Cross equation has recently been modified by incorporating yield stress.79 This
led to

(6)

where Y is the yield stress, and A and B are material constants. Evidently, these three
constants should be temperature dependent, although this was not specified in ref.
79 which dealt with isothermal flow.

In addition to viscosity, extrudate swell is an important characteristic of the behavior

of a rubber compound. According to Oda et al.,80 the extrudate swell is a criterion
for determining the allowable gate-size to avoid undesirable jetting during cavity
filling.
ScienceDirect is Elsevier’s leading information solution for researchers.
Copyright © 2018 Elsevier B.V. or its licensors or contributors. ScienceDirect ® is a registered trademark of Elsevier B.V. Terms and conditions apply.

> Read full chapter