Instrument techniques for rheometry

Ying Y. Hou and Hamida O. Kassim

Citation: Review of Scientific Instruments 76, 101101 (2005); doi: 10.1063/1.2085048

View online: http://dx.doi.org/10.1063/1.2085048
View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/76/10?ver=pdfcov
Published by the AIP Publishing

Articles you may be interested in

Continuous lubricated squeezing flow: A novel technique for equibiaxial elongational viscosity measurements on
polymer melts
J. Rheol. 54, 1083 (2010); 10.1122/1.3474599

Industrial application of ultrasound based in-line rheometry: From stationary to pulsating pipe flow of chocolate
suspension in precrystallization process
Rev. Sci. Instrum. 75, 3164 (2004); 10.1063/1.1790585

Industrial application of ultrasound based in-line rheometry: Visualization of steady shear pipe flow of chocolate
suspension in pre-crystallization process
Rev. Sci. Instrum. 74, 5255 (2003); 10.1063/1.1626024

Obtaining the shear rate profile of steady laminar tube flow of Newtonian and non-Newtonian fluids from nuclear
magnetic resonance imaging and laser Doppler velocimetry data
J. Rheol. 46, 351 (2002); 10.1122/1.1446881

Direct Numerical Evaluation of Shear Rates in Concentric Cylinder Viscometry Using LeastSquares Cubic
Splines
J. Rheol. 33, 745 (1989); 10.1122/1.550037

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 REVIEW OF SCIENTIFIC INSTRUMENTS 76, 101101 共2005兲

REVIEW ARTICLE

Instrument techniques for rheometry

Ying Y. Houa兲 and Hamida O. Kassim
Department of Applied Science, Faculty of Engineering Science and the Built Environment,
London South Bank University, 103 Borough Road, London SE1 0AA, United Kingdom
共Received 21 December 2004; accepted 14 August 2005; published online 25 October 2005兲
This article presents a review of some latest advances in rheology measuring techniques.
Consideration is given to the modification and approaches in conventional measuring techniques
and also to the development of specialty instruments. A number of sensing technologies such as
nuclear-magnetic-resonance imaging and ultrasonic pulse Doppler mapping have recently been
adopted to produce viscoelastic measurements for both Newtonian and non-Newtonian materials.
The working principles of these technologies and their applications are described. Other recent
developments in modifications of conventional rheometers for performance enhancement and for
complex material characterizations have been thoroughly discussed. Some instrument designs and
their special applications, such as interfacial rheometers, extensional rheometers, and high-pressure
rheometers, have also been evaluated in detail. © 2005 American Institute of Physics.
关DOI: 10.1063/1.2085048兴
I. INTRODUCTION Over the past 20 years, there has been some significant
progress in the innovation of conventional viscoelastic mea-
Rheology measurements are important in both industrial suring techniques. A variety of viscometers and rheometers
production and scientific research. The most commonly used with rapid response times, broad ranges of measuring geom-
measuring techniques for viscosity and rheological proper- etries, and accurate velocity and temperature control have
ties can generally be grouped into three broad categories: been developed.5–7 This development has not only enabled
substantial improvements in the performance
共1兲 Capillary technique, in which viscosity is determined
of conventional instrumentation, but also allowed the appli-
from the measurements of both flow rate and pressure
cation of the viscoelastic measuring techniques to be
drop.
successfully extended into a number of previously uncharted
共2兲 Rotational technique, in which viscosity is calculated
areas, such as electro- and magnetorheological fluid
from torque and rotor speed measurements.
measurements,8,9 interfacial shears determination,10,11 soft
共3兲 Falling/rolling ball technique, in which the viscosity is
solids viscoelastic analysis,12 and online multiphase slurry
deduced from the measurements of the time taken by a measurements.13
ball of known density rolling or falling through a certain This review aims to summarize some of the latest devel-
distance in the test fluid. opments in the rheometry instrumentation techniques. It
Although rheological measurements have been a well- commences with a comprehensive review of technical inno-
established practice for centuries, there is still a range of vations in the modification of conventional equipment for
performance enhancement and complex material character-
limitations associated with traditional instruments. Apart
izations. The review continues with a focus on the develop-
from tedious manual operation, the lack of ability to investi-
ment of specialty instruments for interfacial, soft solid, ma-
gate flow conditions at the fluid boundaries, and the local
terial property investigations, and precision measurements.
flow- or temperature-induced inhomogeneities inside the flu-
Both the operating principles of the techniques and their ap-
ids, have been widely considered as some of the major re-
plications will be discussed in detail.
strictions in using traditional measuring techniques, espe-
cially when trying to characterize the rheological properties
of complex materials.1,2 Moreover, with the advent of the
advanced materials such as electro- and magnetorheological
fluids and soft compressible solids, there are increasing II. MODIFICATION APPROACHES TO CONVENTIONAL
MEASURING TECHNIQUES
needs for prolonged shearing studies and dynamic rheologi-
cal investigations. These are well beyond the capability of A. Velocimetry
conventional viscoelastic measuring techniques.3,4 For an isothermal and fully developed laminar flow in a
tube, the viscosity, by definition, is expressed as the depen-
a兲
Author to whom correspondence should be addressed; electronic mail: dence of the shear stress ␶ upon the shear rate ␥ at a particu-
[email protected] lar radial position r,

0034-6748/2005/76共10兲/101101/19/$22.50 76, 101101-1 © 2005 American Institute of Physics

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-2 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

␶共r兲
␩共r兲 = . 共1兲
␥共r兲
In the above equation, ␥共r兲 and ␶共r兲 are defined, respectively,
as

du共r兲
␥共r兲 = , 共2兲
dr

⌬p
␶共r兲 = r, 共3兲
2L
FIG. 1. A mean velocity profile generated from NMR imaging for car-
in which u共r兲 is the local flow velocity, and ⌬p is the pres- boxymethyl cellulose solution 共see Ref. 26兲.
sure drop across the whole length of the tube L.
Thus, for example, for a conventional capillary tube vis- G • u = Gu共r兲cos ␪ , 共5兲
cometer, the determination of the viscosity of a fluid in ques-
tion often requires the measurements of both flow rate and in which ␥ is the gyromagnetic ratio, u共r兲 is the magnitude of
pressure drop. For non-Newtonian flow measurements, this the fluid velocity at any point, and ␪ is the angle between the
can be painfully tedious, as a series of diameters of capillar- velocity and PFG vectors.
ies would have to be used to obtain the set of ⌬p and flow To generate a simple form of velocity map, only two
rate data required, in order to complete a full range of images will be needed, each with a different magnitude or
viscosity-shear rate curves.14 duration of the PFG signal. A recent investigation shows that
To overcome the problem, one obvious approach is to for a cylindrical Couette flow, the one-dimensional 共1D兲 pro-
incorporate some of the advanced monitoring techniques ca- jection measurement which exploits the circular symmetry of
pable of mapping the flow velocity profile into the traditional the velocity field can produce accurate and most speedy ve-
rheometer. This not only enables the shear rate distribution in locity results.20 NMR has been used to investigate a variety
the fluid to be determined accurately in just one simple op- of flows and rheological properties from simple fluids to
eration, but also allows the viscosity information 共␩兲 to be complex suspension,21,22 from bulk viscosity to viscoelastic
extracted analytically, provided that a best-fit approximation characterization,23,24 and from steady capillary flow to rotat-
expression u共r兲 to the measured velocity profile data can be ing concentric cylinder geometry.25
Arola et al.26 carried out a NMR flow measurement on a
deduced. Several investigations on using modern imaging
non-Newtonian fluid. A pulsed gradient spin-echo 共PGSE兲
techniques such as nuclear magnetic resonance and ultra-
pulse sequence was applied to a 1% aqueous carboxymethyl
sound Doppler effect to acquire flow velocity profiles have
cellulose 共CMC兲 solution flowing through a 26.2 mm bore
recently been reported.15,16
polyvinyl chloride 共PVC兲 tubing for the velocimetry mea-
1. Nuclear-magnetic-resonance „NMR… imaging surement. Figure 1 illustrates one of the velocity profiles
The nuclear-magnetic-resonance imaging rheometer was obtained with a corresponding pressure drop of 11.1 kPa/ m.
first patented by Maneval et al.17 This technique is used to It was found that the velocity data were best fitted by the
obtain macroscopic fluid rheological information via NMR following fourth-order polynomial expression:26
imaging measurements and pressure gradient inside a pipe u共r兲 = 657.7 − 1.156r2 − 0.015r4 . 共6兲
flow. NMR flow measurement is based on the principle of
The velocity was then subsequently differentiated at each
encoding displacement in either the magnitude 共bolus track-
radial position to yield shear rate. Figure 2 compares the
ing or time of flight兲 or the phase 共in the plane transverse to
viscosity-shear rate data calculated from the NMR velocity
the polarizing static magnetic field兲 of the local nuclear mag-
netization at any point in the sample. Different NMR tech-
niques which can be exploited to observe fluid flows have
been documented.18,19 The phase-based technique is gener-
ally considered to be more convenient in allowing quantita-
tive interpretation of fluid velocities. In this technique, a pair
of bipolar 共or unipolar if separated by a 180° radio-frequency
pulse兲 of pulsed-field-gradient 共PFG兲 signals of equal mag-
nitude G, duration ␦, and separation ⌬ is usually incorpo-
rated into a NMR sequence, such that the phase of the NMR
signal becomes dependent upon the displacement of the
fluid. For a constant velocity u, the phase ⌽␯ is described by
the following equation:

⌽v = ␥␦⌬Gu, 共4兲 FIG. 2. Results deduced from the velocity profile shown in Fig. 1, along
with measurements from the Haake Rotovisco RV20 viscometer using the
where same fluid 共see Ref. 26兲.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-3 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

profile with those obtained from a Haake Rotovisco RV20

viscometer on the same fluid. It can be seen that there is an
excellent agreement between NMR measurements and RV20
data for shear rate exceeding 10 s−1. The discrepancies have
been mainly attributed to the NMR velocity measurement
resolution and shear rate calculation method.26
Gibbs et al.27 also used NMR flow imaging to measure
the velocity fields of aqueous xantham gum and guar gum
solutions, respectively. The experiments were carried out on
a steady flow of the solutions in a 1.2-cm-i.d. cylindrical
polymethylmethacrylate pipe, using a NMR method capable
of producing maps of velocity with a spatial resolution of the
order of 200 ␮m and a velocity resolution of approximately
2% of the mean velocity measured. The attention of the in-
vestigation was focused on the wall slip effect studies and
observation of bulk rheological behaviors in complex poly-
mer solutions. Figure 3 depicts the flow imaging results for
1% and 0.2% xantham gum and 1% guar gum solutions,
respectively. Clearly, large deviations from the parabolic pro-
file can be observed for all the solutions investigated. The
higher the polymer concentration, the more significant the
deviation will be. However, none of the results shows evi-
dence of apparent wall slip, as there is no significant devia-
tion from zero velocity at the tube wall. Figure 4 illustrates
the plots of dependence of viscosity 共r · ␥−1兲 upon shear rate
共␥兲, as defined from Eqs. 共1兲–共3兲 for all the polymer solu-
tions. The data were calculated from the differentiation of the
velocity profiles in Fig. 3. The results indicate a near power-
law behavior 共␶ = k␥n兲 for the 1% xantham gum solution, as
the data are well presented by
log共␩兲 = c − 0.86 log共␥兲 − 4.6 ⫻ 10−2 log2共␥兲, 共7兲
corresponding to a power-law index n of 0.14 at a shear rate
of 1 s−1. This has been found consistent with other research-
ers’ observations.28
For the guar solution, it is shown that the data are best
fitted by a relationship proposed by Cross,29
␩ ⬀ 关1 + 共a␥兲b兴−1 , 共8兲
where a = 0.21 and b = 0.81, respectively.
In this work, the author claimed that the NMR velocim-
etry results have been found to have a general agreement
with those obtained by the cone-and-plate viscometry. The
FIG. 3. NMR velocity profiles for aqueous xantham gum solutions. The
latter measured a power-law index of 0.11 for shear rates points represent the average of 50 measurements in the radial vicinity, and
between 0.01 and 100 s−1 for the 1% xantham gum, and a the error bars represent the standard deviation of these 50 points 共see Ref.
and b of 1.2 and 0.65, respectively, for shear rates between 27兲. 共a兲 1% xantham gum with Vmean = 0.12 cm/ s, 共b兲 0.2% xantham gum
0.01 and 100 s−1 for the 1% guar gum. The only significant with Vmean = 0.075 cm/ s, and 共c兲 1% guar gum with Vmean = 4.08 cm/ s.
difference is observed for the guar gum in the low shear rates
range 共below 2 s−1兲, where the cone-and-plate data are show-
ing higher viscosity measurements. spans, the velocities of the fluid along the line of the mea-
suring points can be deduced. Repeating this procedure nu-
merous times would allow the entire velocity profile of a
2. Ultrasound pulse Doppler „UPD… mapping flow fluid to be mapped out. The position of each measuring
The ultrasound velocimetry measurement utilizes the point can be determined from the time elapsed between the
well-known Doppler effect, which states that a sound wave pulse emitted and the echoes received, provided that the
scattered by a moving particle would produce a measurable speed of sound in the traveling medium is known. Figure 5
frequency shift in the wave that is directly proportional to the illustrates a schematic diagram of the measuring principle of
velocity of the particle. Thus, by emitting an ultrasound a typical UPD viscometer. After transmitting an ultrasound
pulse of a known frequency into a moving fluid and evalu- pulse of a known frequency f, the sensor is switched to the
ating the frequencies of the reflected echoes at different time receiving mode. The received Doppler echoes are mixed in a
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-4 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 6. A measured velocity profile of glycerine water solution using UPD

mapping 共see Ref. 31兲.

f Dc s
us = u cos ␣ = , 共9兲
2f
where cs denotes the speed of sound in the measuring me-
dium, and ␣ is the incline angle between the sensor and the
direction of the flow.
An extended technique, gradient-UPD 共GUPD兲 method,
has also been documented.30,31 This technique is the combi-
nation of the UPD method and the gradient-laser-Doppler
anemometer. The former emphasizes real time measurement,
whereas the latter is focused on gradient measurement.
The UPD technique has been applied to both Newtonian
共glycerine in water solution兲 and non-Newtonian fluids 共in-
cluding polyacrylamide solution, hydroxpropyl guar gum,
mayonnaise, etc.兲 for velocity profile and subsequent viscos-
ity measurements.30,31 All the investigations indicate that the
UPD method can yield comparable results to those obtained
by commercial rheometers. Figures 6 and 7 show the veloc-
ity profile and the viscosity data from the UPD measure-
ments on a glycerine water solution. The fluid was pumped
FIG. 4. Plot of r共␥兲−1 vs ␥ 共see Ref. 27兲. 共a兲 1% aqueous xantham gum. The
through a 3-m long, 13.2 mm bore Plexiglas pipe. The ultra-
line corresponds to Eq. 共7兲. 共b兲 0.2% aqueous xantham gum. The line cor-
responds to log共␩兲 = c⬘ − 0.651 log共␥兲 − 1.19⫻ 10−2 log2共␥兲. 共c兲 1.0% aqueous sound sensor assembly was mounted approximately halfway
guar gum. The dashed line corresponds to 关1 + 共1.2␥兲0.65兴−1 and the dotted in the pipe and placed in a bath filled with the fluid to avoid
line corresponds to 关1 + 共1.2␥兲0.81兴−1. air from absorbing the sound waves. A pulse of 4 MHz was

demodulator, and filtered to eliminate noises. The Doppler

frequencies of the echoes 共f D兲 can then be revealed by ap-
plying the fast Fourier transform 共FFT兲 algorithm. The flow
velocity u at each measuring point can be calculated by using
the following equation:

FIG. 5. A schematic diagram illustrating the measurement principle of the FIG. 7. The viscosity data measured using the GUPD technique and a com-
UPD technique 共see Ref. 31兲. mercial rheometer, respectively 共see Ref. 31兲.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-5 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

shows a typical velocity profile of mayonnaise measured us-

ing the UPD velocimetry. It exhibits clear features of wall
slip effect and shear-thinning behavior.30
Brunn et al.33 have recently performed rheological stud-
ies of a body lotion by using an UPD viscometer. They found
that although the method had been employed successfully
two disturbing points had been realized.
共1兲 To approximate experimental data by a continuous curve
is rather time consuming.
共2兲 Experimental data are always noisy to some extent.

Wunderlich and Brunn34 specifically investigated the

wall information correction for UPD measurements and also
gave out the comparison between theory and experiment.
FIG. 8. A possible velocity profile from a complex liquid in a tubular flow
共see Ref. 30兲.
They concluded that a large deviation from the expected pro-
files is found near the tube wall, and UPD has demonstrated
many advantages, provided that the measurements are taken
far enough from the tube wall.
transmitted over a duration of 1 ␮s before it was switched
over to the receiving mode. In order to obtain a sound sta-
tistics for the measurements, the pulse was repeated 128 B. Spectroscopy
times for each profile. The authors claim that the viscosities Apart from the velocimetry imaging, both ultrasound
calculated from the UPD velocity profile exhibit a good and NMR have also been adapted to produce spectroscopy
agreement to those obtained on a commercial viscometer viscosity measurements. The ultrasound spectroscopy viscos-
over the entire measurement range31 共see Fig. 7兲. ity 共USV兲 measurement exploits the idea that the reflection
Wunderlich and Brunn32 tried different theoretical ex- of a shear wave at the interface between a solid and liquid
pressions to fit a UPD measured velocity profile on a provides information about the viscosity.30 Unlike the UPD
4000 ppm polyacrylamide solution. They demonstrated that viscosity measurement in which longitudinal wave reflec-
a seemingly good fit of the velocity profile does not neces- tions are measured, in the ultrasound spectroscopy viscosity
sarily guarantee equivalently a good fit of the viscosity data. measurement a shear transducer is employed and brought in
In industrial practice, a simple expression of u共r兲 with mod- contact with the sample being analyzed. By measuring the
erate approximation may well be in favor of a complex ex- magnitude and phase of the reflected shear waves as a func-
pression of u共r兲 with rather accurate approximations. tion of frequency, the reflection coefficient, defined as the
Brunn et al.30 pointed out that the UPD viscometer can ratio of reflected ultrasound to the incident signal, can then
be much more versatile. Figure 8 illustrates what the velocity be calculated. From this the dynamic shear rheology of the
profile of a complex fluid may look like. Obviously both the sample can be deduced. The biggest drawback of the tech-
wall slippage and yield point can be determined from the nique is that it only measures the rheological properties
velocity profile, provided that the wall shear stress is known. within one shear wavelength depth of the sample surface,
Moreover, it would also be possible to detect the phase rather than the properties of the whole sample.30 The probed
distance would increase as the shear wave frequency de-
boundaries from the irregularities in the velocity profile,
creases.
should separation of the complex fluids occur. Figure 9
The ultrasound spectroscopy technique for viscosity
measurement is still in its infancy. Much still has to be in-
vestigated to understand the fundamentals of the process and
the potential capability of the technique. Greenwood et al.35
pointed out that for light liquids such as sugar water solu-
tions, the reflection coefficient is a function of the product of
the density and viscosity of the liquid. For highly viscous
liquids such as silicone oils, the reflection coefficient is de-
pendent on the shear modulus of the liquid. Kulmyrzaev and
McClements36 investigated the dynamic shear rheology of
different concentrations of aqueous honey solutions using the
ultrasound spectroscopy technique and a conventional Cou-
ette viscometer, respectively. Figure 10 compares the results.
As expected, both techniques show that the viscosity of
honey increases with concentration. Kulmyrzaev and
McClements36 claimed a high correlation between the two
FIG. 9. The velocity profiles of a mayonnaise sample determined by a UPD results, indeed they claimed a value of r2 ⬎ 0.97. They also
viscometer sensor at different flow rates 共a, b, and c兲 共see Ref. 30兲. claimed that the results from the ultrasonic device are less
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-6 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 10. Concentration dependence of the viscosity of diluted honey sam-

ples measured using a Couette viscometer 共0.1 s−1兲 and an ultrasonic spec-
trometry 共10 MHz兲 at 25 ° C 共see Ref. 36兲.

reliable at a honey concentration below 70 wt %, therefore,

the equipment is restricted to measuring highly viscous
samples 共above 0.3 Pa s兲. Based on the visual observation
of Fig. 10 alone and with no further data presented in this
paper,36 the present authors could not agree on such a close
value of correlation between the two results. However, the
figure clearly shows that 共1兲 the ultrasonic equipment gave
constantly a lower value of viscosity than the conventional
Couette equipment and 共2兲 there is approximately a constant
difference between the two results.
The most likely explanation is that when a signal of high
frequency is used, the natural relaxation property of the
sample is disturbed and the honey exhibits shear thinning.
Consequently, it will give a lower value than from conven-
tional Couette equipment. FIG. 11. Schematic model for the breakage of gel structure in aqueous hy-
Recently reported is the use of NMR spectroscopy aluronan solution 共see Ref. 38兲. 共a兲 Solution at rest, shear rate of 0 s−1 for
intermolecular hydrogen bonds between chains. 共b兲 Shear rate of 7.4 s−1 for
coupled with a conventional rotational viscometer for the disruption of hydrogen bonds between larger aggregates. 共c兲 Shear rate of
characterizations of sol-gel transition of ultrahigh molecular 194 s−1 for breakage of many intermolecular hydrogen bonds.
weight polyethylene 共UHMWPE兲 decalin solutions and the
breakage of gel structure in aqueous hyaluronan 共HA兲 solu-
tions, respectively.37,38 Figure 11 shows a schematic model been incorporated into a conventional capillary tube viscom-
for the breakage of HA gel structure under different shear eter. Figure 12 illustrates a schematic diagram of the so-
rates. The model has been supported by both the conven- called mass-detecting capillary viscometer 共MDCV兲. The de-
tional cone-and-plate rheometry measurements and 13C vice mainly consists of a pipette pump, a filling tube, a
NMR spectroscopy studies. The former indicated that as the capillary tube, and a precision balance equipped with a com-
shear rate increased from 7.4 s−1 to above 100 s−1, the vis- puterized data acquisition system. The diameter and length
cosity of the HA solution changed from erratic perturbations of the capillary tube are carefully chosen to ensure the fric-
to a universal-shear thinning behavior. The latter, on the tion loss in the capillary tube, which is the dominant loss in
other hand, showed the stretching and disaggregating of the the system. Before a measurement starts, the pipette pump
␤-sheet tertiary structure in the HA solution at shear rates of sucks up the test fluid from the receptacle to a preset position
7.4 and 194 s−1, respectively. in the falling tube. The fluid is then allowed to flow back into
the receptacle freely by opening up the pipette pump to the
C. Other techniques atmosphere. The fluid mass variation over time in the recep-
Shin et al.39 recently developed an innovative approach tacle is recorded by the electronic data logging system at a
to the capillary viscometry. A mass-detecting assembly has resolution of 0.01 g and a frequency of 25 measurements/ s.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-7 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 13. The mass variation of the collected fluid as a function of time 共see
Ref. 40兲.

rates. It also combined the flow rate and pressure drop mea-
surements into a single measurement of liquid-mass variation
FIG. 12. A schematic diagram of the mass-detecting capillary viscometer with time. Figure 13 shows the mass variation as a function
共see Ref. 39兲. of time for soymilk.40 Figure 14 compares the viscosity mea-
surements from a MDCV and a commercial rotating viscom-
Assuming that the flow inside the capillary tube is isothermal eter 共Physica, UDS-200兲. It can be seen that there is an
and fully developed laminar, and there is no slip at the wall, excellent agreement between these two measurements.
the shear rate at the capillary wall can be calculated using the Moreover, it is worth noting that the MDCV can measure the
classical Weissenberg-Rabinowitsch equation,39 viscosity at a much lower shear rate than the rotating vis-

冋 册
cometer.
1 d ln Q A similar idea has also been explored by other
␥共t兲 = ␥共a兲 3 + , 共10兲
4 d ln ␶ workers.41,42 Instead of using a precision balance to measure
the fluid mass variations with time, the authors employed a
where Q is the fluid volumetric flow rate, ␥共a兲 is the appar-
charge-coupled device 共CCD兲 sensor, together with a com-
ent or Newtonian shear rate at the wall and is defined as
puterized data acquisition system to record the variation of
32 dm共t兲 the test fluid level rising in a tube of known diameter. Figure
␥共a兲 = , 共11兲
␳␲d3c d共t兲 15 illustrates the configuration of the so-called scanning cap-
illary tube viscometer. The viscosity of the fluid being mea-
in which ␳ is the fluid density, dc is the diameter of the sured can subsequently be expressed as a function of varying
capillary tube, and m共t兲 is the fluid mass in the receptacle fluid levels in the rising tube with time. Figure 16 shows the
measured at time t. Therefore, dm共t兲 / dt is actually the fluid viscosity measurements for mineral oil at 25 ° C using a dual
mass flow rate inside the capillary tube, and can be related to scanning capillary tube viscometer and a commercial rotat-
the volumetric flow rate Q by the following expression:
dm共t兲
Q=␳ . 共12兲
dt
The shear stress at the wall ␶ is given by
gdc
␶= 关m⬁ − m0 − m共t兲兴, 共13兲
␲Lcd2f
where d f is the diameter of the falling tube, Lc is the length
of the capillary tube, m0 and m⬁ are the masses of the fluid
measured at the beginning and the end of the experiment,
respectively, in the receptacle. Thus, the viscosity can be
determined as a function of m共t兲,

␩=
␳gd4c 关m⬁ − m0 − m共t兲兴
2
8Lcd f dm共t兲/dt
3+ 冉
d ln Q
d ln ␶
冊 −1
. 共14兲

Clearly, the incorporation of the mass-detecting technique

has enabled the conventional capillary viscometer to measure FIG. 14. The flow curves of soymilk at room temperature obtained with a
the fluid viscosity continuously over a broad range of shear MDCV and rotating viscometer 共see Ref. 40兲.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-8 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

effect of surface viscosity on Marangoni instability. Since

then the important role of surface viscosity in interfacial pro-
cesses has attracted wider attention. The investigation of the
interfacial behavior and properties of surfactants, protein,
and their mixtures is getting more and more popular. A de-
tailed discussion and comprehensive review on inter-
facial rheology and corresponding techniques have been
published.46–48 Several techniques have been proposed for
the measurement of interfacial rheological behavior. Four
types of devices for the measurement of interfacial rheologi-
cal behavior have been used, namely, biconical bob,49
knife-edge,50 thin biconical bob,51 and deep-channel interfa-
cial rheometers.52,53 Generally, these rheometers can be di-
vided into two classes: indirect shear rheometer and direct
FIG. 15. A schematic diagram of the scanning dual-capillary-tube viscom- shear rheometer.54
eter 共see Ref. 42兲. The first category includes deep-channel and knife-in-
wall rheometers.55,56 This kind of interfacial rheometer has
ing viscometer, respectively. It can be seen that for the scan- high sensitivity and is suitable for measuring viscoelastic
ning capillary tube viscometer, the measurements come close properties of liquid-liquid interfaces. However, they cannot
to a documented standard viscosity of 9.9 cP for the test oil be used when the upper liquid layer is opaque, such as when
at 25 ° C within the whole range of shear rates. The rotating crude oil is in contact with an aqueous layer. In addition, the
viscometer measurements, however, give increasing errors at interfacial viscosity cannot be calculated until the velocity
the lower end of the shear rate range. profile of the flow is known. The velocity profile is usually
Table I presents the features and merits of the modifica- measured by following the movement of small particles
tions to conventional techniques discussed in this section. placed at the interface. It is cumbersome to perform such
measurements due to the difficulty of positioning such a tiny
III. DEVELOPMENT OF SPECIALTY INSTRUMENT particle accurately.57
The second category consists of rotating disk, biconical
A. Interfacial rheometer
bob 共usually for oil/water and air/water interfaces兲, knife-
It has been long since Hagen 共1845兲 introduced the find- edge, or ring geometries 共air/water interfaces兲. This kind of
ing that the viscoelastic behaviors of the interfacial region interfacial rheometer can directly measure the torque or ro-
differ from those of the adjoining phases.43 In the earlier tational speed of a disk located at the interface.58–60
stage of the study, several researchers attempted using simple Figures 17 and 18 show a sectional view of a deep-
devices, such as a floating needle rotating in a magnetic field channel interfacial rheometer. The annular space between the
to measure the viscosity of a gas-liquid surface. It was soon inner and outer cylinders on the rotating dish constructs a
found out that some corresponding problems were encoun- wide channel. The dish is designed to either rotate at a con-
tered with surface-tension gradients when this method was stant rate or oscillate sinusoidally, whereas the cylinders are
actually used. After that, Boussinesq44 demonstrated a sys- fixed. The dish is driven by a motor that adjusts the ampli-
tematic theoretical analogy on surface viscosity. This theory tude and frequency of the oscillations. The fluid sample in
defines the rheological properties of the interfacial phase as the channel tends to rotate due to the rotation of the dish and
the interfacial dilatational viscosity and the interfacial shear is sheared by the presence of the channel walls. This aug-
viscosity. The former measures the resistance to deformation, ments the effects of the interface and therefore enables the
whereas the later illustrates the resistance of a surface to instrument to determine surface viscosity.59 Usually the
flow. Subsequently, Sternling and Scriven45 confirmed the maximum possible frequency of this instrument is 1 Hz. The
system temperature can be controlled to within ±0.1 ° C. As
mentioned earlier, to calculate the surface shear viscosity the
velocity profile in the channel has to be solved. This can
usually be determined by the velocity at the centerline of the
interface 共vc兲, the channel width 共y 0兲, the height of the liquid
in the channel 共x0兲, and the velocity of the basement of the
channel 共vb兲; with a series of assumptions, the velocity at the
centerline of the interface can be written as61

vc 4 1
= , 共15兲
V̄b ␲ cosh ␲D 1 + ␲␧0/y 0␩0
FIG. 16. The viscosity measurement for mineral oil at 25 ° C with a scan-
ning dual-capillary-tube viscometer 共SDCV兲 and a rotating viscometer 共RV兲
共see Ref. 42兲. under an assumption that a velocity v*c is defined as
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-9 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

TABLE I. Summary of the different modifications approaches to conventional techniques.

Techniques → Velocimetry Spectroscopy

Features ↓ NMRI UPD USV NMR

Shear rate range A wide range of A relative wide Limited A wide range
shear rate 共such range of shear rate of shear rate
as 0.1– 1000 s−1兲 共such as 5 – 80 s−1兲 共such as 0.1– 1000 s−1兲

Velocity resolution 2% of the mean Do not know Do not know Do not know
velocity
measured

Specification for Normal range of Usually less than Usually less Can be up to
temperature control water bath 100° C than 100° C; 200° C
extreme case
can be up to 500 ° C

Applicability to Suitable for Mainly single Mainly single Suitable for

filled/multiphase multiphase phase phase multiphase
systems

Claim for optical Suitable for both Mainly suit for Suitable for Mainly suit for
transparency opaque as well as transparent both opaque as transparent
transparent systems well as systems
systems transparent
systems

Batch or continuous use Batch Continuous Continuous Batch

Suitability of Suitable Suitable Suitable Suitable

velocimetry to measure
close to wall

Other indispensable Not applicable Speed of sound Speed of sound Not applicable
information—speed of sound needed needed

Advantage/disadvantage Advantages: Advantages: Advantages: Advantages:

over existing method 共1兲 Noninvasive. 共1兲 Nonintrusive. 共1兲 One 共1兲 Noninvasive.
共2兲 A single 共2兲 Materials measurement 共2兲 A single
measurement. ranges from can get a result. measurement
共3兲 Suitable gases to soft 共2兲 Nonintrusive. will suffice.
for solid. 共3兲 Wall 共3兲 Can be used
experimental 共3兲 Complex effect can be for
and analytical modulus and detected. experimental
purposes as complex bulk 共4兲 Can and analytical
well as for online modulus can be apply for an purposes as
process obtained. opaque well as for
monitoring and 共4兲 Wall sample. process
control. slippage and Disadvantages: monitoring and
共4兲 By yield point can be 共1兲 Only control.
varying the determined. measures one 共4兲 By varying
pressure 共5兲 Phase shear wavelength the pressure
gradient, the boundaries can depth of gradient, the
range of shear be detected. the sample range of shear
rate can be 共6兲 Easy to surface. rates can be
extended. operate. 共2兲 Only limited extended.
共5兲 Gives Disadvantages: to low viscous
three-dimensional 共3D兲 velocity 共1兲 Time samples 共⬍0.3
information. consuming. Pa s兲.
共2兲 Focusing signal 共3兲 High
noisy to some frequency may
extent. influence the
共3兲 Results not easily natural
reproducible. relaxation
property of the
sample.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-10 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 17. The deep-channel surface viscometer 共see Ref. 61兲.

v*c
= 4/␲ cosh共␲D兲, 共16兲
V̄b
then the surface viscosity ␧0 can be written as FIG. 19. Deep-channel surface viscometer data for 2% aqueous carboxy-
methyl cellulose 共CMC兲-air interface 共see Ref. 61兲.
␧ 0/ ␩ 0 = 共v*c − vc兲y 0/␲vc . 共17兲
Since Eqs. 共15兲–共17兲 are based on the assumption that there
have an inverse relationship with D, which follow the ex-
is a flat interface in analyzing the flow in the deep-channel
pected trend from the theory described in Eqs. 共15兲 and 共16兲.
surface viscometer, a correct interfacial velocity v**c for a It can also be seen from Figs. 19–21 that all experimental
curved interface, which caused by the wet wall of the chan-
nel, from the interface having zero surface viscosity is de- data fall on a straight line parallel to that for v**
c / V̄b. This is
scribed as consistent with the theory for a Newtonian interface.60 More-
over, the discrepancy in the two lines is an indication of the
c /V̄b = 0.904vc /V̄b = 3.616/␲ cosh共␲D兲,
v** 共18兲
*
difference in viscosity between the two fluids. The discrep-
ancy for liquid-air interface 共Figs. 19 and 20兲 is less than that
where v*c
represents the velocity that would be measured for
of liquid-liquid interface 共Fig. 21兲.
an interface having zero surface viscosity. D is the ratio of
To closely examine the interfacial viscosity for non-
the height of the interface in the channel to the width of the
Newtonian material, Lopez et al.63 and Lopez and Hirsa64
channel 共as defined in Figs. 19–21兲. ␩0 is the fluid viscosity
have recently performed some investigations by using an op-
at zero shear rate.
tical annular channel rheometer. This device has a similar
To validate the theory above, Pintar et al.61 and Pintar62
structure as a deep-channel rheometer. It consists of two sta-
carried out some experimental investigations on the surface
viscosity of aqueous solutions of carboxymethyl cellulose
共CMC兲 and carbopol with the deep-channel surface viscom-
eter. Figures 19–21 show the experimental surface viscosity
data for three different interfaces. Indicated on each figure is
c / V̄b against D, computed using Eq. 共18兲, and
the plot of v**
experimental data on the shape of each interface. The data

FIG. 20. Deep-channel surface viscometer data for the 0.02% carbopol-air
FIG. 18. Basic model of the deep-channel surface viscometer 共see Ref. 61兲. interface 共see Ref. 61兲.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-11 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 23. Azimuthal velocity profiles at the interface driven with Re= 250 for
c0 mg/ m2 as indicated; the broken lines are measurements using digital
particle imaging velocimetry 共DPIV兲 and the solid lines are computed 共see
Ref. 63兲.

共DPIV兲. The velocimetry images seed particles 共8.1 ␮m

polystyrene兲 from below by placing a thin laser light sheet.
FIG. 21. Deep-channel surface viscometer data for the 2% carboxymethyl Figure 23 shows the measurement results of surface azi-
cellulose 共CMC兲-benzene interface 共see Ref. 61兲.
muthal velocity, v⬘, obtained by the optical channel DPIV,
together with computation that is based on predicted New-
tionary cylinders of different diameters 共ro and ri兲 in a dish. tonian behavior.62 The results are illustrated by the velocity
As show in Fig. 22, the floor of the dish is rotated at a against the nondimensional gap coordinate x = 共r − ri兲 / 共ro
constant angular velocity. The main differences between this − ri兲. As we can see from Fig. 23, with lower concentration,
optical annular channel and the stainless-steel deep-channel co = 0.4 and 0.8 mg/ m2, the measurements show a good
viscometer 共conventional兲 can be summarized as follows. agreement with the computations, confirming the Newtonian
behavior throughout the interface. However, the discrepancy
共1兲 The optical channel is not restricted to small annular
is getting bigger and bigger with the increase of concentra-
gap-to-depth ratios.
tion, implying a non-Newtonian behavior. The investigator
共2兲 Flow inside the optical channel is steady and axisym-
claimed that the observed shear-thinning nature of the veloc-
metric and the interface remains flat due to the fact that
ity profile is associated with a phase transition at c0
the range of angular velocity used is from the Stokes
⬇ 0.9 mg/ m2 at low Reynolds numbers.
flow limit to large inertia.
Figures 24 and 25 show the schematic diagram of the
共3兲 Inside the optical channel, the interface is completely
covered and the radial surface velocity is zero. As a
result, the surface viscosities do not contribute to the
radial stress balance, and surface dilatational viscosity
can be ignorable.
共4兲 The optical channel allows an examination of non-
Newtonian interfacial hydrodynamics with nonlinear
flow.
共5兲 The optical channel that is used to measure surface shear
viscosity is of much greater dimensions.

In optical channel system, the flow on the air/water interface

is measured by using digital particle image velocimetry

FIG. 22. Schematic of transparent annular channel geometry 共see Ref. 63兲. FIG. 24. The biconical bob interfacial viscometer 共see Ref. 48兲.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-12 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 25. Schematic diagram of the experimental setup for a biconical bob
interfacial viscometer 共see Ref. 48兲.

biconical bob rheometer. This rheometer overcomes the op-

erational disadvantages of the deep-channel rheometer.57 It
can be utilized for the opaque oil/aqueous system. As shown
in Fig. 24, a biconical bob, suspended by a torsion wire, is FIG. 26. Interfacial viscoelastic parameters for the interface of 0.05% PVA
positioned at the interface contained in a cylindrical dish. in water, in contact with oil phase 共see Ref. 57兲.
The dish is driven to oscillate sinusoidally with small ampli-
tude. By measuring the amplitude ratio and phase lag of the
bob with respect to the dish, the interfacial viscosity can natural frequency of the pendulum system, ␮共1兲* is shear vis-
therefore be determined accordingly.57 The theory is based cosity of phase 1 共dimensionless兲, and I* is moment of inertia
on the Boussinesq model for the surface stress tensor.48 of the bob pendulum.
Lately, the biconical bob interfacial rheometer has been
T共␴兲* = 关r* + 共␬* − ␧*兲div共␴兲V共␴兲*兴P + 2␧*D共␴兲* , 共19兲 commercially available with a much higher sensitivity.54 The
where ␬ is the surface dilatational viscosity, ␧ is the surface
* * working principle of the instrument can be explicitly sum-
shear viscosity, marized as follows: Different materials can be identified by
different Boussinesq numbers, B0, defined as the ratio of the
1 interface drag to the subphase drag,
D共␴兲* = 关P • ⵜ共␴兲V共␴兲* + 关ⵜ共␴兲V共␴兲*兴T • P兴, 共20兲
2
␩s
where D共␴兲* is the surface rate of deformation tensor, V共␴兲* is B0 = , 共25兲
␩bulk • R
the surface velocity vector, ⵜ共␴兲 is the surface gradient op-
eration, div共␴兲 is the corresponding surface divergence opera- for highly viscous interfacial film case, B0 → ⬁, and the fol-
lowing relationship will meet:54
tion, and P is the projection tensor that transforms every
vector on the dividing surface into its tangential component. 8
r* is the thermodynamic surface tension. Having made a M = R32␩bulk⍀ + 4␲R22␩s⍀, 共26兲
3
number of assumptions 共see Appendix兲, a series of math-
ematical deductions was made, and eventually, the following where M is the disk torque, ⍀ is the angular velocity, and
equations for amplitude ratio and phase angle ␾ were ␩bulk and ␩s are bulk and interface viscosities, respectively.
established:48 Consequently, by measuring the disk torque and the angular
velocity, the interfacial viscoelastic parameters can be calcu-
S = 2␻N␧R21/兵共2␻N␧R21兲2 + 关K共1 − ␻2兲共1 − R21兲兴2其1/2 , 共21兲 lated.
So far, a number of researchers have performed experi-
tan ␾ = K共1 − ␻2兲共1 − R21兲/共2␻N␧R21兲, 共22兲 mental investigations on different materials with the biconi-
cal bob interfacial viscometer.54,57,65,66 For example, using
K = K*/共2␲␻*0␮共1兲*R*3兲, 共23兲 both biconical and deep-channel rheometers, Nagarajan et
al.57 conducted a set of experiments for macromolecular sur-
␻*0 = 共K*/I*兲1/2 , 共24兲
factant containing a system of 0.05 wt % polyvinyl alcohol
where S is amplitude ratio of the induced oscillation on the 共PVA兲 in water contacted with the oil mixture. These results
bob to the imposed oscillation on the dish, ␻ is frequency of shown in Fig. 26 illustrated the viscous component ␩⬘ and
oscillation, N␧ is a dimensionless number denoting the ratio imaginary part ␩⬙ of interfacial viscosity as a function of
of shear viscous forces at the interface to those in phase 1, R1 frequency. It is clearly seen that there is a dependence of
is the radius of the bob, ␾ is the phase angle between the complex viscosity on the frequency. However, there is also
induced oscillation on the bob and the imposed oscillation on discrepancy in the values measured by the two techniques.
the dish, K* is the surface dilatational viscosity, ␻*0 is the The author claimed that the error is due to the difference in
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-13 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 28. Steady-state interfacial shear viscosity of sorbitan tristearate films

共air/water interface兲 at different surfactant concentrations 共see Ref. 54兲.

this instrument, in terms of sensitivity, has been much im-

proved in comparison with the device used in the earlier
study by Lee et al.65 Nevertheless, the investigator54 did
FIG. 27. Surface loss modulus as a function of frequency with surfactant
concentration as a parameter 共gelatin-surfactant solutions at 35 ° C兲 共see make a clear statement that at lower surfactant concentra-
Ref. 63兲. tions the measurements were getting much more difficult due
to the extremely small torques involved. Each measurement
value has to be averaged from more than 300 000 raw data
sensitivities between the two instruments, and the deep- points.54
channel rheometer usually generates results with a higher
sensitivity.57
Lee et al.65 also carried out a series of rheological stud- B. Extensional rheometer
ies for the surfaces between aqueous gelatin-surfactant solu- The measurement of extensional rheological parameters
tions and air using a biconical bob rheometer. Figure 27 has been pursued for the last three decades since extensional
shows the plot of surface loss modulus as a function of fre- flow is recognized to be one of the major types of deforma-
quency at different surfactant concentrations. The results in- tion in most industrial polymer processing sequences.67,68 A
dicate that the minor concentration changes cannot be easily good review on the earlier developed extensional rheometers
recognized by the instrument although there is an obvious was well presented by Dealy 共1978兲.69 In this review, some
trend that the loss modulus increases with an increase of devices for measuring rheological properties of molten poly-
frequency when the frequency is less than 0.6 rad/ s. It is mers were meticulously classified and compared in terms of
concluded that the mechanical properties in the interfacial their strengths and weakness. However, for the last two de-
region may be altered more significantly than the viscosity cades, experimental and instrumental techniques have been
changes caused by concentration alteration.65 This again con- considerably improved.70,71 This section gives a brief de-
firmed that the instrument sensitivity of a biconical bob in- scription on the developed extensional rheometers.
terfacial rheometer is not high enough for this case; espe- Extensional flow plays an important role in many indus-
cially by the time the authors performed the investigations. try applications,72,73 such as polymer process, coating pro-
A more interesting and more recent investigation has cess, inkjet application, and even in the baking process. The
been reported using Physica’s modular compact rheometer ratio of shear to extension will differ in each specific appli-
共MCR兲 interfacial accessory, a biconical bob interfacial cation. It is therefore important to understand the behavior of
rheometer.54 This MCR is equipped with an air-bearing and materials in both shear and extensional flow fields. As a re-
electronically commutated motor system. The sensitivity of sult, the need to develop extensional rheometry has become
the biconical bob interfacial rheometer is therefore expected apparent. However, the difficulty encountered with an exten-
to be improved. The MCR has been used for investigation on sional rheometer is that steady-state values cannot be easily
different protein films. Figure 28 presents a set of results obtained in contrast to shear rheometry. Furthermore, steady-
measured with spread films of sorbitan tristearate at the air/ state values are not the most relevant ones because elonga-
water interface at two surfactant concentration levels. It is tional motions are generally transient.74 The most sensible
evident from Fig. 28 that the interface features shear thinning value is the step-input extensional viscosity. In this situation,
and with the decrease of surfactant concentration from 31.6 the fluid is initially undeformed or at rest and then is sud-
⫻ 1018 to 1.6⫻ 1018 mol/ m2 interfacial viscosity signifi- denly subjected to a constant rate of extension. During this
cantly drops. This in turn confirms that the performance of process, the extensional viscosity is measured as a function
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-14 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 29. General layout of the modified rotational rheometer 共MRR兲: 共1兲 top FIG. 30. Transient uniaxial extensional viscosity results for a low-density
fixture coupled to torque transducer, 共2兲 bottom fixture driving plate, 共3兲 polyethylene 共IUPAC-X兲 at 150 ° C shown on each line 共see Ref. 76兲. Open
rollers, 共4兲 clamp, 共5兲 high-temperature bath, 共6兲 spring, 共7兲 main arm, and symbols, MRR; solid lines, RME; and dashed lines, fiber windup.
共8兲 sample 共see Ref. 76兲.

⍀Rr
of strain or time.75 It is recognized to be extremely difficult ␧˙ = . 共28兲
L
to develop an extensional rheometer that produces this ideal
motion. So far, this technique has been reportedly used for testing
Generally, the extensional rheometers that have been de- various materials, such as polyethylenes,91 polystyrene,92
veloped so far can be divided into three groups: uniaxial isobutylene-isoprene copolymer,93 and polyisobutylene.76
extensometers, biaxial extensometers, and other stretching Figure 30 presents the transient extensional viscosity results
rheometers. In the first group there are modified rotational for a low-density polyethylene76 performed on the MRR, at
rheometers,76 converging channel extensional rheometers,77 150 ° C. These results are also compared with those obtained
extensional rheometers for high viscosity measurements,78 using the Rheometrics RME and the fiber-windup
falling-plate extensional rheometers,71 and extensional rhe- technique,94 respectively. The close agreement between the
ometers for polymer melts.79,80 The second group includes MRR and other techniques is apparent in Fig. 30, consider-
opposed jet extensional rheometers,81 equibiaxial and planar ing the difficulty of replicating these measurements. The
rheometers,82 and parallel-plate rheometers.83 In the final authors76 claim that these results reflect good performance of
group there are triaxial rheometers,84 film-blowing the MRR. Moreover, the MRR appears not only able to yield
apparatus,85 and filament-stretching rheometers.86,87 One ex- correct values for creep curve, but also capable of monitoring
ample from each group is hereby discussed in detail. the inflexion point in the transient viscosity.

2. Biaxial extensometers
1. Uniaxial extensometers An opposed jet device, the Rheometrics RFX, is a type
So far, a few attempts have been made in modifying of extensional rheometer. This system is considered as one of
normal rotational rheometers for extensional viscosity the most popular methods for measurement of extensional
measurements.88–90 The most notable achievement of a flow properties.81 This equipment utilizes the opposed jets to
modified rotational rheometer 共MRR兲 is to make use of the create extensional flow 共Fig. 31兲. Two identical cylindrical
driving and measuring capabilities of a commercial rota- nozzles are immersed in a reservoir of sample fluid, as
tional rheometer to measure extensional viscosity. Figure 29 shown in the Fig. 31. The nozzles are separated by a certain
illustrates the general layout of a modified rotational rheom- distance. When the fluid is drawn from the gap at a certain
eter. This modification is based on the Weissenberg rheogo- flow rate and approximately uniaxial extensional flow field is
niometer. A motor-controlling angular speed is linked in the created. There is a point at the geometric center of the flow
bottom plate. This rotation is transmitted to the rollers via a field, where no net fluid motion exits. From this point, fluid
belt system. Tensile force is measured and temperature con- accelerates and reaches a maximum velocity at the nozzle
trol is through sample immersion in a silicone oil bath. Fi-
nally, the transient uniaxial extensional viscosity, defined as
the ratio of tensile stress to strain rate under the assumption
of homogeneous extension, can be calculated by76

L ˙
␩+e = M共t兲e␧t , 共27兲
␲⍀Rr2R20

where ⍀ and Rr are the angular speed and radius of the

rollers, respectively, L is the sample length, M共t兲 is the mea-
sured torque, R0 is the initial sample radius, and ␧˙ is the FIG. 31. Schematic of Rheometrics RFX extensional rheometer 共see Ref.
theoretical strain rate, 81兲.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-15 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 32. The apparent extensional viscosity for a Newtonian fluid at three FIG. 33. The effect on the extensional flow curves of 5.8% solutions of
different gaps to diameter ratios. At each ratio, the data are comprised of polyvinylpyrrolidone 共PVP兲 on the addition of sodium dodeclsulphate
measurements made using three different jet sizes 共see Ref. 81兲. 共SDS兲 and glycerol 共see Ref. 81兲.

entrance. This acceleration creates a uniaxial extension in the The dominance of fluid inertia at high velocity is well
flow direction, which is accompanied by compression in the pronounced for low viscosity fluids. The consequence is that
orthogonal plane. A biaxial extension is also created when a non-Newtonian behavior can be wrongly predicted for a
the fluid is expelled from the nozzles towards the geometric Newtonian fluid with low viscosity but high fluid inertia.
center.81 Each nozzle is attached to a separated arm that stays These results indicate the need for a better understanding of
vertically in the reservoir. One arm is stationary and attached the inertia behavior. Such understanding will enhance the
to a transducer, the other is moveable to adjust the gap be- ability to differentiate inertial effects from true non-
tween the nozzles. When measurement starts, the fluid field Newtonian behavior. As a result, a series of further tests on
disturbs the balance of the stationary arm and tends to make polyvinylpyrrolidon 共PVP兲 solution containing sodium dode-
this arm vibrate. The transducer applies a counter torque, T, cylsulphate 共SDS兲 was carried out at a range of jet sizes to
which restores the balance of the arm, keeping it stationary. avoid the inertial effects. Figure 33 shows the extensional
This torque is converted into a force, F. The average exten- viscosity flow curves for the addition of solvent 共glycerol兲 to
sional stress, ␴␧, can therefore be calculated as 5.8% PVP solution and 5.8% PVP with 120 mM SDS solu-
tion, respectively. It is apparent that this opposed jet measur-
4T 4F ing system is capable of identifying the changes in the solu-
␴␧ = 2 = , 共29兲
␲D l ␲D2 tion characteristics. For example, the viscosity increases
significantly with the addition of solvents and SDS.81
where T is a torque applied to the transducer, D is the diam-
eter of the nozzle, l is the length of the stationary arm, and F
3. Other major stretching rheometers
is the measured force. In the meantime, the extension rate
With the development of industrial materials, such as
imposed on the fluid can also be determined as
soft solids, e.g., foam and fiber, there is an urgent need to
2Q know their viscoelastic properties before they are actually
␧˙ = , 共30兲 used. For example, foam is particularly useful for drilling in
␲GD2
the petroleum industry due to its low density coupled with its
where G is the gap between the nozzles of diameter D, and Q good carrying capacity. By using foam the pressure of the
is the volumetric flow rate. The extensional viscosity is then drilling fluid can be maintained below the formation
given by the extensional stress over the extension rate. pressure.95 As a result, the rate of penetration can be consid-
␴␧ FG erably enhanced, and the formation damage, differential
␩␧ = = . 共31兲 sticking risks, and lost circulation can be largely protected.
␧˙ 2Q The rheological characterization of foam is rather
There are two components to the measured force. One is complex95–97 due to the fact that foam is formed by dispers-
the force due to extensional stress, and the other is the force ing gas into an aqueous solution stabilized by surfactants.
due to fluid inertia. At high fluid velocities the contribution Experimental study of flow for foams has been long prac-
from fluid inertia becomes dominant and therefore is signifi- ticed. Herzhaft95 reported the use of a parallel-plate rheom-
cant. This is reflected as an apparent increase in viscosity eter for rheological measurements on a stable foam. It has
with an increasing extension rate. To test the significance of been recognized that slip effect and extension flow make the
inertial effect, Eastman et al.81 used this measuring system precise measurement difficult. One improvement in this as-
for Newtonian liquids. It is expected that for Newtonian be- pect is the use of a triaxial rheometer, an example of other
havior the viscosity will be unchanged with the extensional major stretching rheometers.
rate. Any deviation in viscosity was attributed to fluid inertia. The triaxial rheometer is a totally novel instrument. The
Figure 32 presents the results obtained from three different design is to try to achieve a uniform strain field and to reduce
jet sizes and at three different G / D ratios. It was noticed that the error associated with the free boundaries.84 This rheom-
the apparent viscosity increased considerably as the G / D eter is a translating parallel-plate type. It measures the re-
ratio reduced. For a fixed G / D ratio, the variation in viscos- sponse to applied uniform strains. The deformation of
ity value also illustrated that flow pattern varies with the sample that is fixed between two parallel plates is adjusted
changing of the gap between nozzles. by generating translational motion on the plates. A local
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-16 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 34. 共a兲 Section of the transducer head 共see Ref. 84兲. 共b兲 Schematic of
sample deformation: 共above兲 the resulting stress distribution in compression
and 共below兲 the resulting stress distribution in shear. The arrows indicate the
relative motion of the plates 共see Ref. 84兲. FIG. 35. Compressive stress vs strain for chloroprene foam with 共a兲 ho / rb
= 0.10 and 共b兲 h0 / rb = 0.41 共see Ref. 84兲.

stress transducer is therefore used to measure the correspond-

ing three-dimensional 共xz, yz, and zz兲 components of stress. ing dissolved gases are vital to industrial processes. There
The main key component of the instrument is the triaxial are also needs to investigate the effect of pressure on the
stress transducer. viscosity of carbonaceous material, especially to monitor the
Figures 34共a兲 and 34共b兲 show the transducer head and viscosity changes during pitch material process, which cer-
the schematic of sample deformation. As shown in these tainly help to understand the rate and degree of penetration
figures, this instrument eliminates edge effects by only during impregnation. Over the last 40 years, a number of
measuring the stress on an interior region of the plate. experimental devices that facilitate the rheological property
The two transducer plates, “a” and “b,” are attached to the measurement of different fluids under high pressure have
base plate a to absorb the free edge effect at transducer b and been developed.98–102 However, most of these devices are
thus measure the true material stress response at transducer pressure-driven devices such as capillary rheometer103 or ex-
plate a. trusion slit dies104 and falling-body devices such as falling
Alkhagen and Toll84 conducted experiments to validate needle,105 falling cylinder,106 or rolling sphere devices.107 It
the instrument. The test material was a closed-cell foamed has been recognized that these two kinds of devices inherited
chloroprene rubber. Two types of tests were actually per- some limitations. For example, for pressure-driven devices,
formed: compression and flow curves for both transducer the nature of a pressure-driven flow limits their utility, espe-
plates a and b. The results are shown in Figs. 35共a兲 and cially the large pressure drop limits the concentration of fluid
35共b兲, for the compression test, under a fixed strain rate of that can be measured. For falling-body devices, they can be
−␧˙ zz = 0.01 s−1. With the increase of sample thickness, the operated at constant pressure, which allows the direct mea-
response from load cells at a and b is getting bigger, imply- surement of pressure effects on rheology. However, the fall-
ing that the free edge has a significant effect on the stress on
plate b.
To confirm the performance of the instrument, a parallel
test was also carried out in a conventional parallel disk con-
figuration in a Rheometrics RDAll. As shown in Fig. 36, a
good agreement was achieved between the two instruments.
It is also well summarized that the instrument is capable of
measuring three strain components in different directions and
minimizing the relative motion between plates. Therefore, it
is favorably suitable for compressible solid materials.

C. High-pressure rheometer
Monitoring and studying the pressure effect on material
FIG. 36. Comparison between a triaxial rheometer 共squares兲 and a conven-
viscosity are becoming increasingly important. This is be- tional parallel disk rheometer 共circles兲. The resulting initial shear modulus,
cause the rheological parameters of polymer melts contain- G, is indicated 共see Ref. 84兲.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-17 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

FIG. 38. Viscosity of a petroleum oil sample with Physica’s MCR under
different pressure levels 共see Ref. 110兲.

M
shear stress: ␶ = 共32兲
2␲Ri2l

␲ R2c
shear rate: ␥˙ = n, 共33兲
15 共Rc − R2i 兲
2

where M is torque, n is speed, Ri is the outer radius of the

spindle, Rc is the inner radius of the cup, and l is effective
length of the spindle defined as the length of the lateral sur-
face, which actually contributes to the shear stress.
Figure 38 shows a viscosity measurement result for a
petroleum oil product obtained from a commercial high-
pressure Couette-type rheometer 共Physica’s MCR兲, which
FIG. 37. 共a兲 Schematic of the high-temperature and high-pressure 共HTHP兲
has similar working principle to the one described above.110
viscometer 共see Ref. 109兲. 共b兲 Schematic of the cup-spindle assembly 共see This test was carried out at a constant shear rate of 30 s−1. As
Ref. 109兲. it can be seen from this figure, the onset of fluid gelatinizing
changed from 22.3 ° C under normal pressure 共1.0 bar兲 to
24.8 ° C under high pressure 共3.0 bars兲, indicating that the
ing body density and fluid viscosity determine the shear rate, pressure effect is significant. Meanwhile, one can notice that
which usually results in difficulty in measurements of flow there is a slight discrepancy in the absolute viscosity value
curve parameters for non-Newtonian materials. for the two measurements of different pressure levels. It is
High-pressure rotational devices such as high-pressure believed that the error mainly resulted from the magnetically
Couette and parallel-plate geometries can offer the measure- coupled drive shaft, which has a mechanical bearing,
ments both at constant pressure and over a variety of differ- whereas the test under normal pressure is performed by using
ent shear rates.108 Nevertheless, these devices are quite dif- an air-bearing and electronically controlled motor, which
ficult to design and manufacture because both material demonstrates a much higher sensitivity.111
viscoelastic and device rotational information need to be To overcome the drawback associated with falling-body
and rotational devices, a magnetically levitated sphere rhe-
transferred under pressure.
ometer is designed.112 This instrument combines the prin-
Figures 37共a兲 and 37共b兲 show a high-temperature and
ciple of falling ball with that of a rotational device. In this
high-pressure 共HTHP兲 rheometer designed to measure pitch
rheometer, a magnetic sphere submerged in a test fluid
viscosity at elevated pressure.109 This is basically a Couette-
sample within a high-pressure housing that is made by an
type rheometer. The main features of this design include a optically transparent sapphire tube. By driving the cylindri-
special double-delta metal-to-metal seal. This seal allows cal tube, a shear flow around the levitated sphere can be
the pressurization of the vessel for high-pressure 共Pmax generated. The main working principle is that the fluid sam-
= 7.6 MPa兲 viscosity measurements. A motor mounted on an ple’s viscosity is directly related to the difference in mag-
almost friction-free bearing drives the spindle. A magnedrive netic force needed to keep the sphere at rest and under fluid
is used to connect the motor shaft and the spindle shaft due motion.112 As shown in Fig. 39, the device consists of several
to the fact that a magnetic drive system is able to ensure parts, a motor to drive the cylindrical tube, an electromagnet
leak-proof rotation of the spindle at both high temperature to generate magnetic force, a water bath to adjust tempera-
and high pressure. The fluid viscosity can be measured by ture, and an optical system to monitor the movement of mag-
shearing the fluid between a stationary cup and a rotating netic sphere. This instrument allows operation at constant
spindle. The corresponding shear stress and shear rate can be pressure conditions and eliminates the needs of headspace
calculated as follows: normally used with the high-pressure rotational devices. The
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-18 Y. Y. Hou and H. O. Kassim Rev. Sci. Instrum. 76, 101101 共2005兲

Although some of the techniques discussed in this paper

to date are still at an early development stage, and the appli-
cations are mainly limited to laboratory scale investigations,
most of them have a high potential of being adopted to in-
dustrial operations. We expect a future in which there will be
a variety of instruments having advanced monitoring tech-
niques, powerful flow calculation abilities, precise tempera-
ture control devices, and capable of making fast and accurate
measurements even in a harsh, aggressive plant environment.

ACKNOWLEDGMENTS
FIG. 39. Diagram of a magnetically levitated sphere rheometer 共see Ref.
112兲. The authors wish to acknowledge the financial sponsor-
ship from the Knowledge Transfer Partnership Scheme be-
tween London South Bank University and Anton-Paar Ltd.
only shortcoming of this device is that it needs to be cali- The authors also wish to acknowledge the kind permission of
brated against a fluid of known viscosity because of the non- publishers and authors of the figures utilized in this paper.
homogeneous nature of the flow. By simply investigating the
force balance about the magnetically levitated sphere, the
viscous force imposed on the sphere can be determined. It is APPENDIX
also known that the viscoelastic force is proportional to the The assumptions for Eqs. 共25兲–共28兲 are as follows.
shear stress on the surface of the sphere. Therefore, the vis-
cosity of a Newtonian fluid can be derived from 共a兲 The liquids are incompressible and Newtonian,
共b兲 The zero-thickness disk bob is positioned vertically
K共I − I0兲 such that the phase interface is a plane.
␩= , 共34兲
v 共c兲 The amplitude of the oscillation is so small that
where K is a calibration constant, which is specific to the ␪0 艋 1,
rheometer, the sphere geometry, and the magnetic properties
of the coil and sphere, obtained after a calibration. I and I0 共j兲 d共j兲*R*2␻*0␪0
NRe ⬅ 艋 1,
are the currents supplied to the magnetic coil when the fluid ␮共j兲*
is moving and at rest, respectively. v is the maximum veloc-
where ␪0 is the amplitude of the angular displacement of the
ity at the tube wall.
dish, ␻*0 is a characteristic frequency of oscillation, which is
To evaluate the performance of this instrument, some
defined as the natural frequency of the system, R* is the inner
experimental comparisons between this device and a com-
radius of the dish, and d共j兲* and ␮共j兲* are the density and
mercially available rheometer equipped with a Couette ge-
viscosity of phase j, respectively.
ometry are also implemented in this work.112 It is concluded
共d兲 There is no mass transfer across the liquid-liquid phase
that these experimental results demonstrated a maximum er-
interface.
ror of 5% between the device and the Couette. For most
共e兲 Any surfactant present in the phase interface is uni-
cases, measurements for polymer sample to accuracy of ap-
formly distributed over the entire interface.
proximately 2% could be achievable.
共f兲 The interfacial stress-deformation behavior can be rep-
resented by the linear Boussinesq surface liquid model.

IV. FORWARD LOOKS 1

G. Mason, B. A. Buffham, M. J. Heslop, and B. Zhang, Chem. Eng. Sci.
The incorporation of modern sensing technologies and 53, 2665 共1998兲.
2
various computerized monitoring devices has undoubtedly S. Kim, Y. I. Cho, W. N. Hogenauer, and K. R. Kensey, J. Non-Newtonian
Fluid Mech. 103, 205 共2002兲.
resulted in some significant improvements in both the perfor- 3
H. M. Laun, C. Kormann, and N. Willenbacher, Rheol. Acta 35, 417
mance and application range of the conventional viscometers 共1996兲.
4
and rheometers. Complex rheological properties, such as D. R. Gamota and F. E. Filisko, J. Rheol. 35, 399 共1991兲.
5
shear thinning, yield stress, wall slippage, and phase separa- Th. G. Mezger, The Rheology-Handbook 共Vincentz, Hannover, 2002兲, Vol.
1, Chap. 10, pp. 201–213.
tion can now be investigated from a single measurement, 6
H. O. Kassim and E. G. Killick, IChemE Symp. Ser. Rugby 140, 259
regardless of whether the sample is clear or opaque. The 共1997兲.
7
accuracy and reliability of the measurement is improved. SVM 3000 Stabinger Viscometer Instruction Handbook 共Anton Paar
Some designs have allowed one to enter the previously GmbH, Graz, 2002兲; www.anton-paar.com.
8
J. W. Kim, M. S. Cho, H. J. Choi, S. G. Kim, and M. S. John, Polym. Test.
uncharted rheological measurement area. Special rheological 20, 913 共2001兲.
9
properties, such as interfacial viscosity and extensional vis- S. Odenbach, Th. Rylewicz, and M. Heyen, J. Magn. Magn. Mater. 201,
cosity, both for solid and liquid, can now be measured with a 155 共1999兲.
10
D. Langevin, Adv. Colloid Interface Sci. 88, 209 共2000兲.
higher accuracy. The reliability and repeatability of these 11
B. S. Murray, A. Ventura, and C. Lallemant, Colloids Surf., A 143, 211
measurements have been improved significantly, and the cor- 共1998兲.
12
responding theoretic models have also been improved. D. A. Reinelt, J. Colloid Interface Sci. 132, 491 共1989兲.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22
 101101-19 Techniques and specialty instrument Rev. Sci. Instrum. 76, 101101 共2005兲

13
F. N. Shi and T. J. Napier-Munn, Int. J. Min. Process. 47, 153 共1996兲. 62
A. J. Pintar, Ph.D. thesis, Illinois Institute of Technology, 1968.
14
H. A. Barnes, Appl. Rheol. 9, 102 共1999兲. 63
J. M. Lopez, R. Miraghaie, and A. H. Hirsa, J. Colloid Interface Sci. 248,
15
B. Ouriev, Appl. Rheol. 10, 148 共2000兲. 103 共2002兲.
16 64
M. K. Cheung, R. L. Powell, and M. J. McCarthy, AIChE J. 43, 2596 J. M. Lopez and A. Hirsa, J. Colloid Interface Sci. 229, 575 共2000兲.
共1997兲. 65
H. O. Lee, T. S. Jiang, and K. S. Avramidis, J. Colloid Interface Sci. 146,
17
J. E. Maneval, K. L. McCarthy, M. J. McCarthy and R. L. Powell, U. S. 90 共1991兲.
Patent No. 5,532,593 共2 July 1996兲. 66
R. Borbás, B. S. Murray, and É. Kiss, Colloids Surf., A 213, 93 共2003兲.
18
A. Caprihan and E. Fukushima, Phys. Rep. 198, 195 共1990兲. 67
F. N. Cogswell, Trans. Soc. Rheol. 16, 383 共1972兲.
19
J. M. Pope and S. Yao, Concepts Magn. Reson. 5, 281 共1993兲. 68
J. M. Dealy, Polym. Eng. Sci. 11, 433 共1971兲.
20 69
A. D. Hamlon, S. J. Gibbs, L. D. Hall, D. E. Haycock, W. J. Frith, and S. J. M. Dealy, J. Non-Newtonian Fluid Mech. 4, 9 共1978兲.
Ablett, Magn. Reson. Imaging 16, 953 共1998兲. 70
D. F. James and K. Walters, Techniques in Rheological Measurements
21
L. Frydman, J. S. Harwood, D. N. Garnier, and G. C. Chingas, J. Magn. 共Chapman and Hall, London, 1993兲.
Reson., Ser. A 101, 240 共1993兲. 71
R. W. G. Shipman, M. M. Denn, and R. Kennigns, J. Non-Newtonian
22
K. Kose, Phys. Rev. Lett. 72, 1467 共1994兲. Fluid Mech. 40, 281 共1991兲.
23
S. A. Altobelli, R. C. Givler, and E. Fukushima, J. Rheol. 35, 721 共1991兲. 72
S. Muke, I. Ivanov, N. Kao, and S. N. Bhattacharya, J. Non-Newtonian
24
A. L. Graham, S. A. Altobelli, E. Fukushima, L. A. Mondy, and T. S. Fluid Mech. 101, 77 共2001兲.
Stephens, J. Rheol. 31, 191 共1991兲. 73
J. Meissner and J. Hostettler, Rheol. Acta 33, 1 共1994兲.
25
C. J. Rofe, R. K. Lambert, and P. T. Callaghan, J. Rheol. 38, 875 共1994兲. 74
C. S. Petrie, Rheol. Acta 34, 12 共1995兲.
26 75
D. F. Arola, G. A. Barrall, R. L. Powell, K. L. McCarthy, and M. J. D. F. James and G. M. Chandler, J. Non-Newtonian Fluid Mech. 35, 421
McCarthy, Chem. Eng. Sci. 52, 2049 共1997兲. 共1990兲.
27 76
S. J. Gibbs, D. Xing, T. A. Carpenter, and L. D. Hall, J. Rheol. 38, 1757 J. M. Maia, J. A. Covas, J. M. Nóbrega, T. F. Dias, and F. E. Alves, J.
共1994兲. Non-Newtonian Fluid Mech. 80, 183 共1999兲.
28
J. O. Carnali, J. Appl. Polym. Sci. 43, 929 共1991兲. 77
K. Sadeghy and D. F. James, J. Non-Newtonian Fluid Mech. 90, 127
29
M. M. Cross, J. Colloid Sci. 20, 417 共1965兲. 共2000兲.
30
P. O. Brunn, J. Vorwerk, and R. Steger, Appl. Rheol. 3, 20 共1993兲. 78
R. Muller and D. Froelich, Polymer 26, 1477 共1985兲.
31
M. Müller, P. O. Brunn, and T. Wunderlich, Appl. Rheol. 7, 204 共1997兲. 79
H. Münstedt, J. Rheol. 23, 421 共1979兲.
32
Th. Wunderlich and P. O. Brunn, Flow Meas. Instrum. 10, 201 共1999兲. 80
A. Bach, H. K. Rasmussen, and O. Hassager, J. Rheol. 47, 429 共2003兲.
33 81
P. O. Brunn, Th. Wunderlich, and M. Müller, Flow Meas. Instrum. 15, J. R. Eastman, J. W. Goodwin, and A. M. Howe, Colloids Surf., A 161,
139 共2004兲. 329 共2000兲.
34
Th. Wunderlich and P. O. Brunn, Flow Meas. Instrum. 11, 63 共2000兲. 82
P. Hachmann and J. Meissner, J. Rheol. 47, 989 共2003兲.
35 83
M. S. Greenwood, J. R. Skorpik, and J. A. Bamberger, AIChE Sixth Con- J. M. Dealy and S. S. Soong, J. Rheol. 28, 355 共1984兲.
ference on Food Engineering, Dallas, TX, 1999 共unpublished兲, pp. 691– 84
M. Alkhagen and S. Toll, J. Rheol. 46, 31 共2002兲.
85
696. C. D. Han, Rheology in Polymer Processing 共Academic, New York, 1976兲.
36
A. Kulmyrzaev and D. J. McClements, J. Food. Eng. 45, 219 共2000兲. 86
V. Tirtaatmadja and T. Sridhar, J. Rheol. 37, 1081 共1993兲.
37
H. T. Chiu and J. H. Wang, Polymer 40, 6859 共1999兲. 87
G. H. McKinley and T. Sridhar, Annu. Rev. Fluid Mech. 34, 375 共2002兲.
38 88
E. Fischer, P. T. Callaghan, F. Heatley, and J. E. Scott, J. Mol. Struct. C. W. Macosko and J. M. Lornston, SPE Tech. Pap. 19, 461 共1973兲.
602–603, 303 共2002兲. 89
R. W. Connely, L. J. Garfield, and G. H. Pearson, J. Rheol. 23, 651
39
S. Shin, S. W. Lee, and D. Y. Keum, Rev. Sci. Instrum. 72, 3127 共2001兲. 共1979兲.
40
S. Shin and D. Y. Keum, J. Food. Eng. 58, 5 共2003兲. 90
M. Rides, C. R. G. Allen, and S. Chakravorty, NPL Report CMMT 共A兲 44
41
Y. I. Cho, W. T. Kim, and K. R. Kensey, Rev. Sci. Instrum. 70, 2421 共National Physical Laboratory兲, Teddington, Middlesex, UK 共1996兲.
共1999兲. 91
H. M. Laun and H. Münstedt, Rheol. Acta 15, 517 共1976兲.
42
S. Kim and Y. I. Cho, Rev. Sci. Instrum. 71, 3188 共2000兲. 92
A. E. Everage and R. L. Ballman, J. Appl. Polym. Sci. 20, 1137 共1976兲.
43
G. H. L. Hagen, Abh, Koeniglichen Akad. Wiss., 41 共1845兲. 93
J. F. Stevenson, AIChE J. 18, 540 共1972兲.
44
M. J. Boussinesq, C. R. Acad. Sci. 156, 983 共1913兲. 94
M. Padmanabhan, L. J. Kasehagen, and C. W. Macosko, J. Rheol. 40, 473
45
C. V. Sternling and L. E. Scriven, AIChE J. 5, 514 共1959兲. 共1996兲.
46 95
M. K. Vora, M.S. thesis, Illinois Institute of Technology, 1970. B. Herzhaft, J. Colloid Interface Sci. 247, 412 共2002兲.
47 96
D. A. Edwards, H. Brenner, and D. T. Wasan, Interfacial Transport Pro- R. K. Prud’homme and S. A. Khan, Foams, Theory, Measurements and
cesses and Rheology 共Butterworth-Heinemann, Boston, MA, 1991兲. Applications, Surfactant Science Series Vol. 57 共Dekker, New York, 1996兲.
48 97
Y. C. Ray, H. O. Lee, T. L. Jiang, and T. S. Jiang, J. Colloid Interface Sci. B. Herzhaft, Oil Gas Sci. Technol. 54, 587 共1999兲.
119, 81 共1987兲. 98
L. J. Gerhardt, C. W. Manke, and E. Gulari, J. Polym. Sci., Part B: Polym.
49
D. T. Wasan, L. Gupta, and M. K. Vora, AIChE J. 17, 1287 共1971兲. Phys. 35, 523 共1997兲.
50 99
T. S. Jiang, J. C. Chen, and J. C. Slattery, J. Colloid Interface Sci. 96, 7 C. Kwag, C. W. Manke, and E. Gulari, J. Polym. Sci., Part B: Polym.
共1983兲. Phys. 37, 2771 共1999兲.
51
S. G. Oh and J. C. Slattery, J. Colloid Interface Sci. 67, 516 共1978兲. 100
J. R. Royer, Y. J. Gay, J. M. DeSimone, and S. A. Khan, J. Polym. Sci.,
52
R. J. Mannheimer and R. S. Shechter, J. Colloid Interface Sci. 32, 195 Part B: Polym. Phys. 38, 3168 共2000兲.
共1970兲. 101
M. Lee, C. B. Park, and C. Tzogannakis, Polym. Eng. Sci. 39, 99 共1999兲.
53
R. Nagarajan and D. T. Wasan, Rev. Sci. Instrum. 65, 2675 共1994兲. 102
Y. C. Bae and E. Gulari, J. Appl. Polym. Sci. 63, 459 共1997兲.
54 103
P. Erni, P. Fischer, and E. J. Windhab, Third International Symposium on M. Mackley, R. Marshall, and J. Smenlders, J. Rheol. 39, 1293 共1995兲.
Food Rheology and Structure, 9-13 February 共ETH Zurích, Zurich, Swit- 104
C. D. Han and C. Y. Ma, J. Appl. Polym. Sci. 28, 851 共1983兲.
105
zerland 2003兲, pp. 411–415. J. A. Roetling, J. J. Gebhardt, and F. G. Rouse, Carbon 25, 233 共1987兲.
55 106
F. C. Goodrich and A. K. Chatterjee, J. Colloid Interface Sci. 34, 36 E. Kiran and Z. Gokmenoglu, J. Appl. Polym. Sci. 58, 2307 共1995兲.
共1970兲. 107
M. D. Hersey and H. Shore, Mech. Eng. 共Am. Soc. Mech. Eng.兲 50, 221
56
J. V. Addison and R. S. Shechter, AIChE J. 25, 32 共1979兲. 共1928兲.
57 108
R. Nagarajan, S. I. Chung, and D. T. Wasan, J. Colloid Interface Sci. 204, F. Koran and J. M. Dealy, J. Rheol. 43, 1279 共1999兲.
53 共1998兲. 109
P. M. Khamdare, J. W. Zondlo, P. B. Stansberry, and A. H. Stiller, Carbon
58
R. J. Addison and R. A. Burton, J. Colloid Interface Sci. 32, 73 共1970兲. 38, 881 共2000兲.
59 110
N. Lifshutz, M. G. Hegde, and J. C. Slattery, J. Colloid Interface Sci. 37, Y. Y. Hou, Project Progress Report No. 4 共Anton-Paar Ltd, Hertford,
73 共1971兲. Hertfordshire, UK, 2004兲.
60 111
P. B. Briley, A. R. Deemer, and J. C. Slattery, J. Colloid Interface Sci. 56, Physica MCR- Modular Compact Rheometer Series 共Anton-Paar GmbH,
1 共1976兲. Graz 2004兲; www.anton-paar.com
61 112
A. J. Pintar, A. B. Isreal, and D. T. Wasan, J. Colloid Interface Sci. 37, 52 J. R. Royer, Y. J. Gay, M. Adam, J. M. DeSimone, and S. A. Kham,
共1971兲. Polymer 43, 2375 共2002兲.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
132.206.7.165 On: Thu, 11 Dec 2014 21:24:22