874 IEEE JOURNAL OF OCEANIC ENGINEERING, VOL. 40, NO.

4, OCTOBER 2015

Rheological Behavior of Artificial Flocculated

Clay/Guar-Gum Suspensions
Priyanthi M. Amarasinghe and Andrei Abelev

Abstract—In this paper, artificial flocs, similar to the ones that aqueous suspension at the sediment-water interface, which is re-
are found in a recently deposited cohesive seabed, were prepared ferred to as “fluid mud” in some cases [1]–[3], or deposited on
using Ca-montmorillonite and kaolinite (two of the most abun- the seafloor, representing the most recent addition to the consol-
dant clay minerals), guar gum (a polysaccharide closely resem-
bling natural marine organic matter), and artificial sea water, in idating sediment layer. Understanding the viscous behavior of
order to investigate the effect of the composition on their rheolog- this type of material is important in many applications of ocean
ical response. Shear rate effects on viscosity of these suspensions engineering. The applications range from remote acoustic sed-
were also systematically addressed. These effects typically have a iment identification and classification to the prediction of the
strong influence on the overall rheological response of materials amount of impact and subsequent burial during placement, re-
incorporating any polymeric substance. It was found that the ad-
dition of guar gum, even as little as 5% relative to the clay con- trieval, or movement of objects on, near, or in these cohesive
tent, increases the viscosity of the floc suspensions significantly. The seafloors. The erosion and hydrodynamic transport parameters
higher the guar-gum and clay content, the greater is the viscosity of these flocs also largely depend on their viscous properties and
of the floc suspensions, within the guar-gum loading range investi- settling velocity. The settling velocity of flocs is primarily deter-
gated in this study. The floc suspensions with higher solid content mined by the size, solid concentration, and the density distribu-
were found to exhibit a slight shear thinning behavior, apparently
due to the breakdown of the floc structure under high shear stress. tion of the flocs [4]. Floc breakup and aggregation during turbu-
This behavior is more pronounced in montmorillonite floc suspen- lent flow depend on the equilibrium floc size, the residence time
sions than in the kaolinite suspensions due to the differences in of the floc in the turbulent water column, and the characteristics
the clay mineral surface charge properties and surface-to-mass ra- of the turbulent flow [5]. Floc strength and overall suspension
tios. We explore different rheological models in their applicability viscosity properties are major factors in attaining equilibrium
to describing the observed viscous response and conclude that a
cross-type formulation may be most appropriate. conditions during turbulent flows.
During the formation of flocs in saline marine environments,
Index Terms—Montmorillonite, kaolinite, guar gum, viscosity,
a variety of organic matter is aggregated together with clay par-
flocculation, rheology.
ticles. Decomposition of terrestrial plants and weathering of
clay rich soils and bedrocks deliver substantial amounts of or-
I. INTRODUCTION ganic and inorganic material to large river systems [6]. This or-
ganic matter and clay particles, transported by the rivers, are dis-

R ECENT marine sediments are highly heterogeneous and charged into the lagoons and estuaries. Significant flocculation
complex materials, in which mineralogy, organic matter and aggregation of these materials may start as these materials
type and composition, salinity, and depositional characteristics mix with seawater. Changes in the physico–chemical proper-
have a strong influence on the mechanical properties such as ties of these constituents continue to take place as a result of
viscosity. Understanding the viscous behavior of newly formed progressively increasing salinity of the sea water in the estuary
cohesive marine sediments, composed of flocculated matter, is compared to the fresh water in the river. Often, these suspended
important in many applications such as penetration of heavy ob- sediments settle within 5 km from the coastline due to floccula-
jects in the cohesive seafloors, hydrodynamic transport of surfi- tion and aggregation of the particles [7], although longer trans-
cial seafloor sediments, drag forces on seafloor surface and sub- ports offshore could be expected under stronger flow conditions.
surface vessels, and acoustic properties, among others. In addition to the terrestrial organic matter, marine organic
Viscosity of clay suspensions is expected to be influenced matter is also incorporated into flocs, especially in environments
by the presence of organic matter either in aggregated or un- with rapid primary productions. Whereas the terrestrial organic
aggregated forms. Clay and organic materials are present as an matter, such as humic acid, tends to retard flocculation, the ma-
rine organic matter is enriched in polysaccharides and thus tends
to encourage flocculation [8], [9].
Manuscript received January 30, 2014; revised June 21, 2014; accepted
November 12, 2014. Date of publication January 12, 2015; date of current To predict the viscous behavior of flocculated marine sed-
version October 09, 2015. This work was supported by the Office of Naval iment suspensions, a comprehensive understanding of the
Research/Naval Research Laboratory (ONR/NRL) and National Research
relative effects of the different types of organic matter and
Council (NRC) Research Associateship under Grant N00173-08-2-C-005.
Associate Editor: M. Richardson. various clay minerals is required. A systematic study, inves-
P. M. Amarasinghe was with the National Research Council, Naval Research tigating these effects by varying one parameter (e.g., clay
Laboratory, Washington, DC 20375 USA.
mineral, clay concentration, or organic matter type and concen-
A. Abelev is with the Naval Research Laboratory, Washington, DC 20375
USA (e-mail: [email protected]). tration, salinity, etc.) at a time is therefore required. The current
Digital Object Identifier 10.1109/JOE.2014.2375632 work, as part of a broader inquiry into the rheological and the

0364-9059 © 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
AMARASINGHE AND ABELEV: RHEOLOGICAL BEHAVIOR OF ARTIFICIAL FLOCCULATED CLAY/GUAR-GUM SUSPENSIONS 875

Fig. 1. Atomic structures of kaolinite (left) and Ca-montmorillonite (right).

micromechanical response of individual flocs and floccular suspensions into larger aggregate particles with a greater set-
suspensions, investigates how the charge neutral guar gum, a tling velocity. However, these flocculated particles tend to be
polysaccharide that closely resembles natural marine organic very porous and may break apart when sheared by vigorous
matter, and two of the most abundant in nature clay minerals mixing.
affect the rheological properties of the floccular suspensions. In this work, we focus our attention on studying the coagu-
A preliminary study conducted earlier focused on the mor- lation and flocculation of clay and guar gum as a function of
phological analysis of the flocs which were prepared using Xan- the relative composition and clay mineral type in ionic solution.
than gum as the organic material [10]. In general, the floccu- Two methods were tested for floc preparation, imposing variable
lation behavior of clay sediments depends on the type of the degrees of shearing in the suspensions during mixing. The flocs
clay mineral and organic matter, as well as the ionic strength were imaged and characterized optically, using a microscope. A
(salinity) of the water via the mechanisms of the colloidal in- series of rheological experiments were then conducted with the
teraction forces. For instance, xanthan gum is more effective goal of establishing the dependence of viscosity of these suspen-
than chitin in promoting flocculation of montmorillonite [11]. sions on floc composition, floc formation processes, and testing
Meanwhile, addition of humic acid to clay mixtures inhibits conditions.
the aggregation of Na-montmorillonite particles [8]. In general,
the amount of organic matter present in sediments suspensions II. EXPERIMENTAL
greatly influences the mechanical properties, such as viscosity A. Materials
and yield stress [39].
Ca-montmorillonite (STx-1) and kaolinite (KGa-1b), repre-
In this study, we systematically investigate different methods
senting two of the most abundant clay minerals found in nature,
of controlled preparation of artificial marine clay/guar-gum floc
were obtained for this study from the Clay Mineral Society
suspensions and conduct a series of rheological tests on the re-
(Chantilly, VA, USA). Ca-montmorillonite is a 2:1 phyllosili-
sulting material. We also attempt to describe the influence of
cate. The atomic structure of montmorillonite clay mineral is
floc compositions such as clay concentration, guar-gum content
composed of an octahedral sheet, which is sandwiched between
(relative to clay content and absolute), and the ionic strength of
two tetrahedral sheets, as shown in Fig. 1. These structures
the water on the viscous behavior and the shear flow charac-
are called tetrahedral–octahedral–tetrahedral (TOT) layers. To
teristics of floc suspensions, as a function of the applied shear
form clay particles, several of these TOT layers were stack one
rate. This type of comparison will help to reveal the roles of dif-
above another with an interlayer space gap between them. For
ferent constituents, such as organic materials and clays, on the
dry clays the interlayer distance is about 1 nm [16]. Normally,
overall rheologic behavior of suspended cohesive oceanic sed-
clay layers are negatively charged due to isomorphous substi-
iments. A concise version of the current study can be found in
tution. The negative charges of the layers are balanced by the
[12]. The term “organic matter” will also be used to refer to the
cations that occur in the interlayer. The cations that occupy
guar-gum component of our simulated marine sediment suspen-
the interlayer in Ca-montmorillonite are calcium. The cation
sions throughout the remainder of the paper.
exchange capacity and the surface area of Ca-montmorillonite
are 91.5 meq/100 g and 753.5 m /g, respectively [16]. The
A. Coagulation and Flocculation atomic structure of kaolinite clay mineral is composed of an
octahedral sheet and a tetrahedral sheet. The layers are called
Double-layer repulsive, van der Waals attractive, and inter- tetrahedral–octahedral (TO) layers. Hydrogen bonding between
layer forces are the most prominent physico–chemical forces layers results in formation of kaolinite clay particles. The cation
acting on clay particles [13]. Electrolyte solutions decrease the exchange capacity and the external surface area of kaolinite
double-layer repulsive forces acting on any two particles [14]. A are 2–10 meq/100 g and 10–30 m /g, respectively [16]. Miner-
decrease in the double-layer repulsive force results in a decrease alogical and chemical analysis of STx-1 and KGa-1b clays can
in the distance between particles. This process is called coagula- be found in several baseline studies [17]–[19].
tion. An addition of organic material may bridge the OH groups Guar gum, obtained from MP Biomedicals (Santa Ana, CA,
of clay particles, which leads to a flocculated structure [15]. Co- USA), was used as the organic material that represents a typical
agulation and flocculation are the mechanisms that make clay microbial extracellular polysaccharide, abundant in the marine
876 IEEE JOURNAL OF OCEANIC ENGINEERING, VOL. 40, NO. 4, OCTOBER 2015

environment. Guar gum is a high molecular weight polysac-

charide. A schematic of a guar-gum repeating unit is shown in
Fig. 2. Guar gum does not contain any functional groups that
contribute to any charge. Therefore, it is considered as a charge
neutral polymer [20]. Guar gum also is water soluble. Guar-gum
solutions contain some ionic groups as a result of inadvertent
hydrolysis. These ionic groups assist flocculation once mixed
with clay. It has been found that the hydrogen bond between the
broken edges of the layered clays and the polymers is the main
binding mechanism of these types of polymers to layered clays
[21]. However, the affinity of guar gum to the clay surfaces is
different from the affinity of guar-gum derivatives: nonionic hy-
droxypropyl and cationic hydroxypropyl guar gum [22]. Floccu-
Fig. 2. Molecular structure of guar-gum repeating unit.
lation efficiency is greater in cationic hydroxypropyl guar gum
compared to linear guar gum. High-affinity polymer molecules
may increase the electrostatic interaction between the clay and torque values that are high enough to be measured (with suf-
the polymer molecules, resulting in an increased viscosity of the ficient resolution) even in the low viscosity suspensions. The
flocculated aggregates [23]. Additional details on flocculation inner and outer radii of the inner cylinder are 19.7 and 20.5 mm,
of clay with guar gum may be found elsewhere [40]. Artificial respectively. The inner and outer radii of the cup are 19.2 and
sea salt was used to produce solutions with varying salinities, 21.0 mm, respectively. The gap between two surfaces is 0.5 mm.
typical of the estuarine and marine environments. Although, some high concentration floc samples (created using
Ca-montmorillonite) tend to contain elongated flocs longer than
B. Testing Equipment 500 m, since they are also thin and narrow, it is believed that
A Brookfield R/S plus Rheometer with the double gap at- any error due to the geometry of the attachment is insignificant.
tachment DG3 DIN was used to test the viscosity of the floc To confirm this, an additional test was performed using the at-
suspensions. The soft solid tester (SST) configuration and the tachment CC25 that has 1.06-mm gap between the walls on the
instrument data processing and analysis software (Rheo3000, sample with the highest concentration (5 g/L of clay, 0.25 g/L
Middleboro, MA, USA) has been a powerful feature for mea- of guar gum, and 35 PSU of salinity, also named as 5-5-35 as
suring the shear stress in other studies of soft clay materials explained in Section II-C) as the case where any effect of a large
[24]. The resistance of the flocs to shearing, due to floc size, floc size would be more pronounced. The outer and inner radii
volume fraction, and interfloc bridging, is expected to contribute of the CC25 attachment are 13.56 and 12.5 mm, respectively.
to the overall viscosity in the floc suspensions. In the Brookfield Due to the low torque resolution in these very low viscosity sus-
R/S Rheometer, the torque required to rotate the spindle in the pensions, the lowest shear rate that yields reliable viscosity data
sample is measured. The tests could be conducted under either in CC25 measuring system is 1250 s , which is beyond the
constant shear stress or constant shear rate conditions. The latter shear rate range that is of interest. The viscosity of the suspen-
was chosen due to possible significant effect of the shearing sion that was measured using the CC25 attachment at 1250-s
rates on polymeric suspensions [38]. In cylindrical spindles, the shear rate was 1.894 mPa.s. This value is in line with the vis-
shear rate and the shear stress are given by cosity values of the same suspension that was measured using
DG3DN attachment in the range 250–1000-s shear rate. The
(1) measuring ranges of the shear rate and the shear stress for the
DG3DN measuring system are 0.43–4 340 s and 0.83–83 Pa,
and respectively. The nominal viscosity range of the instrument is
from 0.19 mPa s to 190 Pa s. The filling volume of the mea-
suring cup is 16 mL.
(2) The floc suspensions were examined under a Nikon Inverted
Microscope, the Eclipse Ti-S version, to perform the morpho-
respectively. In these equations, , and are the logical analysis of the flocs. Four different objectives with 4X,
angular velocity of the spindle (radians per second), the radius 10X, 20X, and 40X magnification lenses were used. The mi-
of container (centimeters), the radius of the spindle (centime- croscope is equipped with a camera and the Nikon image pro-
ters), the radius at which the shear rate is being calculated (cen- cessing and analysis software. For most floc imaging, the dif-
timeter), the torque (dyne centimeters), and the effective length ferential interference contrast (DIC) microscopic illumination
of the spindle (centimeters), respectively. Thus, during the tests, method was employed, providing the best imaging results of all
the shear rates are set to the desired value while the resulting the techniques available. The microscope slides that hold the
torque is measured. Details of the viscosity calculation method floc suspensions for imaging were prepared by creating a 5-mm
can be found in Brookfield R/S Rheometer operating instruc- diameter hole on Plexiglas and mounting 0.25-mm-thick glass
tions manual, No. M08-29. In the double gap measuring sys- slide underneath, thus allowing the use of all the objectives, in-
tems, an inner (hollow) cylinder is mounted in the center of the cluding the highest magnification one (40X) with minimal in-
cup. This geometry provides an increased shear area to achieve terferences.
AMARASINGHE AND ABELEV: RHEOLOGICAL BEHAVIOR OF ARTIFICIAL FLOCCULATED CLAY/GUAR-GUM SUSPENSIONS 877

Fig. 3. Comparison of the MMT flocs created using (a) 25-PSU and (b) 35-PSU salinities.

C. Preparation of Floc Suspensions for Rheological Analyses of excess (unassociated) guar gum in the suspension. The goal
was to study the effects of flocculation on the mechanical
In many natural river sediments, it is believed that clay min- properties of the resulting suspensions as opposed to the uni-
eral fragments may come in contact with most organics before formly distributed but unaggregated mass of clay and organic
the river reaches the estuarine plume zone, where the mixture components. In this preliminary work, clay, guar gum, and
encounters increased water salinity [25]. Therefore, to mimic salt water were mixed using a magnetic stirrer as well as a
this natural process, the clay (particle sizes smaller than 2 m) Brookfield R/S Rheometer with the double gap attachment.
and guar gum were mixed first, and then the salt water was intro- In this latter case, the exact shearing conditions are known and
duced to the mixture during the preparation of floc specimens. In can be controlled, which is not the case with magnetic stirrers,
this study, the flocs that were created using montmorillonite are which create irregular and nonuniform turbulent flows in the
referred to as MMT flocs and the flocs that were created using container. To find out the most suitable mixing rate—those
kaolinite are referred to as KOL flocs. that create more stable floccular structures—mixtures were
A complete set of preliminary tests was performed on suspen- sheared in the rheometer with several shearing rates ranging
sions prepared using the montmorillonite clay mineral. How- from 1 to 800 s , and the resulting flocs were examined under
ever, due to that study's preliminary nature and limited scope, the microscope. Fig. 4 shows the comparison of the size and
as well as lower surface activity in KOL suspensions versus morphology of the resulting MMT flocs that were prepared
the MMT ones, only the latter were tested. These preliminary using the magnetic stirrer and the rheometer methods. This
studies also included an investigation of the effect of salinity on preliminary work also indicated that shearing the mixtures in
the morphology of the MMT-floccular suspension, since salinity the rheometer with a rate of 50 s , which corresponds to a
may influence the yield stress of sensitive clays [26], [27]. In this linear velocity of 2.1 cm/s (0.07 ft/s), tends to produce larger
preliminary work, three floc suspensions were prepared with and denser flocs. The magnetic stirrer technique did not result
the salinity set to 5, 25, and 35 practical salinity units (PSU) in larger flocs and did not appear to aggregate as much of
and keeping the guar gum and the montmorillonite clay con- the guar gum into the floc structures, leaving a larger amount
tents constant at 17% and 83% by weight, respectively. Size of material unabsorbed. Fig. 4 also shows that very low and
and morphologies of the resulting flocs were then examined very high mixing rates in the rheometer did not yield distinct
under the microscope. Fig. 3 shows the comparison of the MMT (large) and well-formed floc structures either, thus, the choice
flocs created using 25- and 35-PSU salinities. It was found that of a moderate mixing velocity of 50 s .
the suspensions prepared with 35-PSU salinity (typical for sea Once the optimal mixing rates were established, the prelim-
water) resulted in larger and denser floccular structures. There- inary work was extended to determine the optimum shearing
fore, the salinity was kept constant at 35 PSU for the remainder (or mixing) time. Thus, the floc suspensions were sheared at
of this study, as we were studying other floc suspensions' com- rate of 50 s for different durations. Results showed (Fig. 5)
positional effects on the rheological properties. that shearing the flocs for 5–10 min tends to produce larger and
A literature review indicates [28] that the mixing rates, denser flocs. Literature survey [31], [32] as well as our micro-
used in the preparation of the suspensions, affect the resulting scopic observations of flocs revealed that longer mixing times,
floc size. Typical river flow velocities range from 3 to 180 in excess of these limits, do not change the size and morphology
cm/s (0.1–6 ft/s) [29]. Normally, the sedimentation of river of the flocs significantly These findings show that the exact and
transported material takes place when the river velocities are uniform shear fields within the suspension can be imposed and
between 0.48 and 30 cm/s (0.016 and 1 ft/s) [30]. A prelim- that the magnitudes of the velocities imposed comply with those
inary study was conducted to find the most effective mixing typically found in riverine environments. It was thus decided to
technique that creates larger flocs and leaves the least amount make all the floc specimens by mixing the ingredients at a rate
878 IEEE JOURNAL OF OCEANIC ENGINEERING, VOL. 40, NO. 4, OCTOBER 2015

Fig. 4. Comparison of the MMT flocs created using (a) the magnetic stirring technique and the rheometer mixing technique at constant shear rates of (b)
1 s , (c) 50 s , and (d) 800 s .

of 50 s for 25 min in the double gap attachment of the Brook- and 0.469 ft/s), respectively. These velocities are within the
field rheometer. range of the typical river flow velocities [29] but may be higher
The current study pursues several goals in attempt to system- than some low-energy littoral environments where some cohe-
atically investigate the rheological behavior of floccular sus- sive, fine-grained sediment deposition may occur. The interest
pensions, including: 1) the effects of two different clay min- in these higher linear velocities (or corresponding shear rates)
erals (Ca-montmorillonite and kaolinite); 2) total clay concen- is also in addressing problems typical of object–seafloor inter-
tration, used in floc formation; 3) total guar-gum concentration action, including free-fall penetration and burial, object drag
(defined as percentage of the clay mass); and 4) a wide range of through the suspended mud layer, or resistance to seafloor traffi-
the strain rates imposed on these floccular suspensions during cability by bottom (or near-bottom) machines, mechanisms, and
shear. To accomplish these goals, three sets of floccular suspen- unmanned vehicles or crawlers. Since these velocities are higher
sions were prepared from each clay mineral type using three than the flocs' settling velocities [4], any error due to settling of
different clay concentrations: 1, 3, and 5 g/L. To understand flocs on the bottom of the measuring system is thus minimized.
the effect of guar-gum content on the viscosity of the floc sus- Rheometer experiments were performed as soon as the sus-
pensions, varying amounts of guar gum were added to each of pensions were placed in the attachment and the duration of the
the three sets of clay concentrations mentioned above. Since the testing period was kept at 5 min or below. Test results show that
percent of organic matter in typical marine sediments may vary the viscometer measurements become stabilized (reach a steady
anywhere between 0% and 15% [33] of the total mass of solids state condition) within close to 1 min. No significant fluctua-
in suspension, we selected several guar-gum-to-clay ratios (by tions in the measurements occur during the rest of the testing
weight), including: 0%, 1%, 3%, 4%, and 5%. This range of time. This observation is also believed to indicate insignificant
solids concentrations of 1–5 g/L is rather typical in suspensions floc settling to the bottom of the attachment. To avoid any turbu-
such as fluid mud [34]. The salinity of all the floc suspensions lent flow or vortices occurring between the two walls, the max-
was kept at a constant value of 35 PSU, representing a typical imum shear rate implemented was set as 1000 s based on both
ocean environment. Reynolds and Taylor criteria, as discussed further in Section III.
All the tests were conducted using the constant shear rate Before the experiments, the authors noted considerable
mode of the instrument. The shear rates used in this study were roughness of the cylinder walls and decided that no additional
250, 500, 750, and 1000 s , which correspond to linear veloci- roughening was needed to mitigate any possible wall-slip
ties of 11.73, 23.47, 35.20, and 46.94 cm/s (0.117, 0.234, 0.352, effects. In general [35], this wall slip is most pronounced in
AMARASINGHE AND ABELEV: RHEOLOGICAL BEHAVIOR OF ARTIFICIAL FLOCCULATED CLAY/GUAR-GUM SUSPENSIONS 879

Fig. 5. Comparison of the MMT flocs created with different shear times: (a) 3 min, (b) 5 min, (c) 10 min, and (d) 25 min.

highly viscous fluids. In our case, the floc concentrations were TABLE I
very low as were the overall viscosity values of the suspensions CONSTITUENTS OF THE FLOCCULAR SPECIMENS FOR KAOLINITE
AND MONTMORILLONITE SUSPENSIONS
tested. The combination of these two considerations led us
to believe that any errors due to a possible wall slip were
insignificant.
Nomenclature of the floc specimens was set as follows. The
first number indicates the concentration of clay (MMT or KOL)
in grams per liter in the specimen. The second number indicates
the weight percentage of guar gum relative to the weight of the
clay. The third number represents the salinity of the specimen
in practical salinity units. Thus, for example, a specimen la-
beled “1-5-35” will have clay concentration of 1 g/L, guar-gum
concentration of 0.05 g/L (or 5%, relative to clay content), and
salinity of 35 PSU. Table I lists the specimens prepared for this
study as well as their individual constituents for both, MMT-
and KOL-based floccular suspensions.

D. Tylor and Reynolds Number Calculation

The choice of a measuring technique has a profound effect
on the rheological behavior of fluids. Depending on a number
of factors, there is a shear rate beyond which the laminar flow
of the fluid turns into an unstable flow. Taylor vortices and tur-
bulent flow are the most concerning instabilities in rheological
analysis of low viscosity fluids [36], [37], when using a con- when the bob is rotating too fast against the stationary cup re-
centric-cylinder configuration with the internal cylinder rotating sulting in centrifugal and inertial forces within the fluid mass.
and the external cylinder stationary. These instabilities occur Therefore, it is important not to exceed the upper boundary
880 IEEE JOURNAL OF OCEANIC ENGINEERING, VOL. 40, NO. 4, OCTOBER 2015

TABLE II
TAYLOR AND REYNOLDS NUMBERS FOR WATER AND 1-0-35 KOL-FLOC SUSPENSION

of the shear rates, allowing for a stable laminar flow in such

rheometer experiments. The Taylor number is given by

(3)

where , and are the angular velocity in radians per Fig. 6. Variation of viscosity of water in the shear rate range of 250–2400 s .
second or sec , the radius of the bob in millimeters, the den-
sity of the fluid in kilogram per cubed meter (kg/m ), and the
viscosity of the fluid in Pascal-second (Pa s), respectively. is
the ratio between the cup diameter and the bob diameter.
The critical Taylor number is 41.2 [36], [38]. The Reynolds
number is given by

Fig. 7. Variation of shear stress with time for the 5-5-35 floc specimens sheared
(4) at constant shear rates of (a) 250 s and (b) 1000 s .

The critical Reynolds number is considered as 1000 [38].

66.8, and 63.4, respectively. The Reynolds values for water at
the same shear rates are 407.9, 565.1, 809.8, and 1603.4.
III. RESULTS AND DISCUSSION All the viscosity values measured using the Brookfield R/S
Rheometer instrument were calibrated using the Cannon vis-
A. Taylor and Reynolds Number Verification cosity standard N2 (Decane) solution. This standard fluid dis-
Higher shearing rates are associated with higher Taylor and plays the Newtonian flow behavior, as expected [41]. The vis-
Reynolds flow regimes, due to instability in the flow. The low cosity values, therefore, do not depend on the shear rate applied
viscosity fluids reach the critical Taylor and Reynolds numbers as long as the laminar flow condition is maintained. The ranges
relatively quickly with increasing shear rates for the current of Taylor and Reynolds numbers calculated for Decane for the
double gap testing configuration. Water exhibits the lowest vis- shear rate range 250–1000 s are 5–15 and 167–291, respec-
cosity measured in this study. The 1-0-35 KOL-floc specimen tively. Both are well within the laminar flow regime.
exhibits the lowest viscosity among all the floc specimens tested
in this study. Table II shows the Taylor and Reynolds numbers B. Viscosity of MMT-Floc Suspensions
calculated for water and the 1-0-35 KOL-floc suspension for all It was observed that during shear, suspensions come to a
the shear rates used. steady state after some time. Two examples of shear stress
Out of all the samples tested, only 1-0-35 KOL-floc suspen- versus time relationship for the 5-5-35 MMT-floc suspension
sion and water at 1000-s shear rate show slightly elevated are shown in Fig. 7. As can be seen in the figure, the average
Taylor numbers of 45.0 and 42.5, respectively, compared to the shear stress reaches a near-constant value very quickly. While
critical value of 41.2. Taylor numbers in all the other tests were the data shown in Fig. 7(a) appear to reach this steady-state
smaller than the critical value of 41.2. All the samples, including value after approximately 60 s, the shear stress in Fig. 7(b) in-
water, exhibit Reynolds numbers smaller than 1000. However, dicates small continuous reduction in value up to about 200 s of
it was observed that the shear rates higher than 1000 s tend to runtime. This reduction, however, is very small and is similar
elevate the viscosity values significantly due to the occurrence to the overall variability in the stress values at any one time.
of vortices and turbulence flows. The viscosity values of water For this reason, we believe that the viscosity values of the floc
measured using the shear rates between 250 and 2400 s are suspensions may be calculated by averaging the shear stress
shown as an example in Fig. 6. A sudden increase in the vis- values after the steady state is attained. While this steady-state
cosity is observed after 1000 s . The Taylor numbers for water condition may be considered to begin after about 100 s for
at 1400-, 1800-, 2200-, and 2400-s shear rates are 50.3, 56.8, all tests conducted, it was decided to average the shear stress
AMARASINGHE AND ABELEV: RHEOLOGICAL BEHAVIOR OF ARTIFICIAL FLOCCULATED CLAY/GUAR-GUM SUSPENSIONS 881

Fig. 8. Relative viscosity values of the MMT-floc suspensions. (a)–(d) Floc suspensions containing 0%, 1%, 4%, and 5 % guar gum relative to the weight of the
clay.

Fig. 9. Comparison of the size and morphology of the MMT flocs with (a) 1-5-35 and (b) 5-5-35 compositions.

values for 200–300-s time interval, for added measurement Fig. 9. The flocs are larger and appear denser in the 5-5-35
reliability. specimen compared to the 1-5-35 specimen. Denser flocs may
The relative viscosity values of the floc suspensions were dissipate larger shearing energies, resulting in elevated viscous
calculated by taking the ratio between the viscosity of floc properties compared to the smaller and more loosely associated
specimens and the viscosity of water at a given shear rate. flocs. A slight shear thinning behavior is observed with the floc
Fig. 8 shows the relative viscosity values of the MMT-floc suspensions with higher clay and guar-gum concentrations. As
suspensions. Fig. 8(a)–(d) represents the floc suspensions can be seen in Fig. 8(c) and (d), floc suspensions with higher
containing 0%, 1%, 4%, and 5% guar gum relative to clay MMT concentrations (e.g., 5-5-35, 5-4-35, and 3-5-35) exhibit
content, respectively. Although little flocculation is visible higher viscosity values when sheared at lower shear rates. The
in the microscopic images (not included), it can be observed viscosity then starts to decrease with increasing shear rate. An
in Fig. 8(a) that the contribution of the clay fraction to the image of the floc specimen 5-5-35 after applying 1000-s shear
viscosity is insignificant when there is no guar gum present in rate is shown in Fig. 10. In comparison to the image that was
the suspensions. The highest viscosity is observed when the taken before applying shear [Fig. 9(b)], the flocs appear broken
guar-gum content is at 5% of the clay content, by weight. It is into smaller subparts. As the shear rate increases from 250 to
evident that increasing the amount of the organic component 1000 s , larger flocs break up into smaller flocs, resulting in
to 5% results in the higher overall viscosity of the floc suspen- lower viscosity values of the suspension.
sions. This observation may be attributed to the greater amount
of material present, naturally increasing the overall viscosity C. Viscosity of KOL-Floc Suspensions
of the suspensions. Comparison of the size and morphology Fig. 12 shows the variation of the relative viscosity of the
of the flocs for the 1-5-35 and 5-5-35 suspensions is given in KOL-floc suspensions. Fig. 12(a)–(d) represents the floc sus-
882 IEEE JOURNAL OF OCEANIC ENGINEERING, VOL. 40, NO. 4, OCTOBER 2015

Fig. 11. Shear rate versus shear stress for (a) MMT-floc suspensions and (b)
KOL-floc suspensions prepared with 5-g/L clay concentration. The linear re-
gression for the water is shown as a dotted line.

colloids into a 3-D network. Increasing the amount of guar gum

Fig. 10. Change in the size and morphology of the 5-5-35 floc specimen after may strengthen the guar-gum/clay network up to an optimum
applying 1000-s shear rate.
dosage. During shear, breaking and reforming of these bridges
is a continuous process even after the steady state is attained.
pensions containing 0%, 1%, 4%, and 5% guar-gum content, Change in the rheological behavior of different guar-gum/clay
relative to clay. The KOL-floc suspensions show somewhat sim- suspensions may then be attributed to the different magnitudes
ilar viscous behavior to MMT-floc suspensions. Similarly to the of binding energies between the clay particles and the guar gum.
MMT-floc suspensions, the contribution of amount of clay to the The change is more prominent in MMT-floc suspensions com-
overall viscosity of the suspension is insignificant when there pared to KOL-floc suspensions due to the higher clay surface
is no organic matter present. The highest viscosity is observed charge densities, resulting in higher initial clay–polimer particle
when the guar-gum content is 5% of the clay weight, indicating bonding.
that this organic component plays a major role in the viscosity
of the KOL-floc suspensions as well. The higher clay concentra- D. Viscosity Models
tions result in elevated viscosity values in the suspensions due to The examination of the rheological response of the suspen-
the greater amount of material present in the suspensions. How- sions in Fig. 11 may suggest that a Bingham plastic model may
ever, in spite of the same amount of materials present, the vis- be appropriate to describe the overall response within the strain
cosity is considerably lower in KOL-floc suspensions compared rate range explored during our tests, and if we were to ignore the
to the MMT-floc suspensions. This behavior may be attributed slight shear thinning effects present in only some of the suspen-
to the low amount of flocculated materials that can be seen sion, as noted earlier and shown in Figs. 8 and 12. The Bingham
in KOL-floc suspensions compared to the MMT-floc suspen- function is shown in
sions, which subsequently may be contributing to the fewer floc
breakage events during shear, where the smaller flocs may be (5)
more stable under equivalent flow conditions. Another notice-
able difference between the viscosities of MMT- and KOL-floc where is the yield stress under essentially static shear con-
suspensions is the less prominent shear thinning behavior in ditions. While this model may be appropriate in describing the
KOL-floc suspensions. This difference may also be attributed overall response in only a portion of the parameter space ex-
to the initially lower amount of flocculation in KOL-floc sus- plored (higher range of shear rates and lower ranges of clay and
pensions compared to the MMT-floc suspensions. Unlike MMT guar-gum concentrations), in general, we expect that the actual
flocs, the breaking of flocs into smaller flocs, which seems to yield stress is near zero. This is suggested by the fact that the
be the apparent cause for shear thinning nature, is less promi- suspensions tested herein are made with low overall concentra-
nent in KOL-floc specimens. The comparison of the size and tion of clay and guar-gum material (per volume of water) and no
morphology of MMT flocs, before and after shear, is shown in settlement or consolidation time was allowed before testing, as
Fig. 13. to result in any significant transformation of the suspension to a
Most floc suspensions tested exhibited Newtonian behavior solid soil layer, where such static yield stress may become an ap-
within the range of the shear rate explored. The relationship be- propriate descriptor. Linear regressions yielded essentially zero
tween the shear stress and the shear rate for the specimens cre- apparent yield stress (within the accuracy of the experiments)
ated using 5-g/L concentration is shown in Fig. 11. The linear re- in all but two cases: 212 and 136 mPa in 5-5-35 and 5-4-35
gression that fits the results of water and its equation are shown MMT-floccular suspension samples, respectively.
in the figure. A schematic model of the actual response of the suspensions
When small amount of guar gum is added to the colloidal clay may be suggested, as shown in Fig. 14. The yield stress is best
suspension, guar gum, in its ionized form, will adsorb to the described as an apparent yield stress , indicating its limited
charged clay colloids in such a way that it will bridge the clay applicability.
AMARASINGHE AND ABELEV: RHEOLOGICAL BEHAVIOR OF ARTIFICIAL FLOCCULATED CLAY/GUAR-GUM SUSPENSIONS 883

Fig. 12. Relative viscosity values of the KOL-floc suspensions. (a)–(d) Ffloc suspensions containing 0%, 1%, 4%, and 5 % guar gum relative to the clay
weight, respectively.

Fig. 13. Comparison of the size and morphology of the 5-4-35 KOL flocs in suspension (a) before and (b) after shear at 1000 s .

relative to clay. While this exercise averages over a wide range

of shear rates, thus ignoring some of the important dependen-
cies, as discussed above, it may still be useful in exploring
these guar-gum content effects in general. The results for
MMT suspensions are shown in Fig. 15(a), and the results for
KOL suspensions are shown in Fig. 15(b). The fitted curves,
shown in the MMT chart are given only as an approximation
of the observable trends, separately for the 5-g/L concentration
and jointly for the essentially indistinguishable 1- and 3-g/L
concentrations. Thus, 5-g/L clay concentration suspensions
show a smooth and more substantial increase in the apparent
viscosity from 0% to 5% guar-gum-to-clay ratio, than the 1-
and 3-g/L clay concentrations. We may also note a possible
change in trends at around 3% guar-gum-to-montmorillonite
Fig. 14. Example of a viscous pseudoplastic material with a pseudoyield
stress . ratio, where the rise is more dramatic at higher content values.
On the other hand, kaolinite suspensions [Fig. 15(a)] show
In addition to the apparent yield stress, an apparent Bingham no definitive trend as a function of guar-gum-to-clay content
viscosity could also be examined (regression over entire range ratio, but only increased apparent Bingham viscosity values at
of shear rates tested) as a function of the guar-gum content, higher overall clay content, as noted before.
884 IEEE JOURNAL OF OCEANIC ENGINEERING, VOL. 40, NO. 4, OCTOBER 2015

Fig. 15. Apparent Bingham viscosity as a function of relative guar-gum content and total clay content for two clay minerals: (a) MMT and (b) KOL.

Fig. 16. Apparent Bingham viscosity as a function of total guar-gum content and total clay content for two clay minerals: (a) MMT and (b) KOL.

It may be illustrative to show that the overall increase in seen in MMT versus KOL suspensions are due to the higher
guar-gum content is not the sole contributing factor to the ob- surface charge densities of the clay minerals and possibly to the
served changes of the apparent viscosity. For this purpose, we higher surface area-to-clay volume characteristics that result in
could plot the apparent Bingham viscosity values not versus the the stronger guar-gum-to-clay mineral bonds developing during
guar-gum-to-clay ratio as before, but versus the total guar-gum flocculation, in turn resulting in the higher overall resistance to
content, present in the suspensions. Fig. 16 shows such an ex- shearing.
ample. There are noticeable changes in trends that occur at some To summarize, we can state that the suspensions tested do
critical guar-gum content value—demonstrating, perhaps, a bi- not appear to exhibit any true yield stress—the kind that can be
linear nature of response. Additionally, we notice that these measured at very low, essentially static, shear rates (perhaps on
changes in trends no longer occur at the same value of the ab- the order of 0.01–1.0 s [38]), which were beyond the reso-
scissa. This would suggest that not only the overall availability lution capacity of our instrument for these very low resistance
of the guar gum in the suspensions, but also the combination suspensions. In this case, the more appropriate model of de-
of the clay mineral presence and the availability of guar gum scribing the rheological response of these suspensions could be
in the suspension to produce the floccular structures that are, in the cross model for shear thinning fluids or a somewhat sim-
fact, responsible for the overall rheological effects described, plified Sisko equation [38]. These formulations include a dedi-
are important. As noted before, the more pronounced effects cated shear thinning constant, among other material constants,
AMARASINGHE AND ABELEV: RHEOLOGICAL BEHAVIOR OF ARTIFICIAL FLOCCULATED CLAY/GUAR-GUM SUSPENSIONS 885

and generally describe viscosity as a decay function from the incorporated into floc structures (as suggested by microscope
initial (static or low rate viscosity) to the final (very high observations) is incorporated at much lower binding potential
strain rate) viscosity that is much easier detached (bonds/flocs fractured) during the
suspension shearing process.
(6)
ACKNOWLEDGMENT
where and are material constants (cross constant and cross The authors would like to thank Dr. P. Valent, associate ed-
exponent). Again, since the actual determination of the very low itor Dr. M. Richardson, and three anonymous reviewers for pro-
rate viscosity was not possible due to device limits and the very viding constructive comments on this manuscript.
low shear resistance of the highly dilute suspensions under con-
sideration, we did not attempt an actual model fit. Such an ex- REFERENCES
ercise would be speculative in this case. [1] G. P. Allen, J. C. Salomon, P. Bassoullet, Y. Du Penhoat, and C. de
Grandpré, “Effects of tides on mixing and suspended sediment trans-
port in macrotidal estuaries,” Sedimentary Geol., vol. 26, pp. 1–3, 1980.
IV. SUMMARY AND CONCLUSION [2] R. Kirby and W. R. Parker, “Distribution and behavior of fine sediment
in the Severn Estuary and inner Bristol Channel, UK,” Can. J. Fisheries
Artificial floc suspensions were prepared using Ca-montmo- Aquatic Sci., vol. 40, pp. 83–95, 1983.
rillonite, kaolinite, guar gum, and sea salt to investigate the ef- [3] J. T. Wells, C. E. Adams, Jr., Y. A. Park, and E. W. Frankenberg, “Mor-
phology, sedimentology and tidal channel processes on a high-tide-
fect of the constituents and the shear rate on the rheological range mudflat, West Coast of South Korea,” Mar. Geol., vol. 95, pp.
behavior of the flocs. Fifteen different floc suspensions from 111–130, 1990.
each clay type were prepared by mixing three different clay [4] M. J. Fennessy, K. R. Dyer, and D. A. Huntley, “INSSEV: An instru-
ment to measure the size and settling velocity of flocs in situ,” Mar.
concentrations with five different guar-gum concentrations and Geol., vol. 117, pp. 107–117, 1994.
35-PSU salinity water. The flocs were examined under an op- [5] J. C. Winterwerp, “A simple model for turbulence induced flocculation
tical microscope. The produced Ca-montmorillonite flocs were, of cohesive sediment,” J. Hydraulic Res., vol. 36, pp. 309–326, 1998.
[6] M. Dagg, R. Benner, S. Lohrenz, and D. Lawrence, “Transformation
in general, larger in size than the Kaolinite flocs. of dissolved and particulate materials on continental shelves influenced
Viscosity of these floc suspensions was measured by im- by large rivers: Plume processes,” Continental Shelf Res., vol. 24, pp.
posing shear forces with four different (and constant) shear rates 833–858, 2004.
[7] M. J. Devlin and J. Brodie, “Terrestrial discharge into the great barrier
from 250 to 1000 s . Microscopic imaging was carried out to reef lagoon: Nutrient behavior in coastal waters,” Mar. Pollution Bull.,
investigate the morphology of the flocs. It was found that both vol. 51, pp. 9–22, 2005.
MMT- and KOL-floc suspensions exhibit somewhat similar [8] Y. Furukawa, J. L. Watkins, J. Kim, K. J. Curry, and K. R. H. Bennett,
“Aggregation of montmorillonite and organic matter in aqueous media
viscous behavior with increasing organic matter content. The containing artificial seawater,” Geochem. Trans., vol. 10, p. 2, 2009.
addition of organic materials changes the viscous behavior of [9] T. Ayukai and E. Wolanski, “Importance of biologically mediated re-
clay suspensions significantly. The higher the organic content, moval of fine sediments from the fly river plume, Papua New Guinea,”
Estuarine Coastal Shelf Sci., vol. 44, pp. 629–639, 1997.
the higher is the viscosity of the floc suspensions. The floc [10] A. Abelev, “Examination and manipulation of clay aggregates—Initial
suspensions that have higher clay and organic contents exhibit inquiry,” Naval Res. Lab. (NRL), Washington, DC, USA, NRL Tech.
slight shear thinning behavior, especially above approximately Rep. NRL/FR/7420-11-10, 2011, 33 pp.
[11] Y. Furukawa and J. L. Watkins, “Effect of organic matter on the floccu-
3%–4% organic matter content for the highest clay concentra- lation of colloidal montmorillonite: A modeling approach,” J. Coast.
tion tested. This behavior can be attributed to the breakdown of Res., vol. 28, pp. 726–737, 2012.
floccular particles under high shear force. The shear thinning [12] P. M. Amarasinghe and A. Abelev, “Rheological behavior of floc-
culated suspensions of artificial marine cohesive sediment,” in Proc.
behavior is less pronounced in KOL-floc suspensions. The MTS/IEEE OCEANS Conf., San Diego, CA, USA, Sep. 23–27, 2013.
reason could include the lack of larger or denser flocs, present [13] A. Anandarajah and P. M. Amarasinghe, “Discrete element study of
in the KOL-floc suspensions as compared to the MMT-floc the swelling behavior of NA-montmorillonite,” Geotechnique, vol. 19,
pp. 173–183, 2012.
suspensions. Shear thinning behavior was not observed at lower [14] K. Ma and A. C. Pierre, “Colloidal behaviour of montmorillonite in the
organic matter and clay concentrations. In fact, a very mild presence of Fe ions,” Colloids Surfaces, vol. 155, pp. 359–372, 1999.
shear-thickening effect may be seen at low clay concentrations. [15] P. Mpofu, J. Addai-Mensah, and J. Ralston, “Investigation of the effect
of polymer structure type on flocculation, rheology and dewatering be-
At very lower clay concentrations, all the suspensions behave haviour of kaolinite dispersions,” Int. J. Miner. Process., vol. 71, pp.
in essentially Newtonian manner. 247–268, 2003.
Evaluation of rheological models suggested that the nature of [16] H. Van Olphen, An Introduction to Clay Colloid Chemistry. New
York, NY, USA: Wiley Interscience, 1977, p. 70 and 255.
viscous response of the floccular suspensions tested lies not only [17] S. J. Chipera and D. L. Bish, “Baseline studies of the clay minerals
in the overall amount of guar gum used in sample preparation, society source clays: Powder x-ray diffraction analyses,” Clays Clay
but also in the combination of guar-gum and clay mineral con- Minerals, vol. 49, pp. 398–409, 2001.
[18] A. R. Mermut and A. F. Cano, “Baseline studies of the clay minerals so-
tent, expressed as constant guar-gum-to-clay ratio, by weight. ciety source clays: Chemical analyses of major elements,” Clays Clay
This fact stresses the importance of flocculation mechanisms on Minerals, vol. 49, pp. 381–386, 2001.
the rheological response and suggests that certain changes in be- [19] J. E. Kogel and S. A. Lewis, “Baseline studies of the clay minerals
society source clays: Chemical analysis by inductively coupled
havior may occur after some critical value of guar-gum-to-clay plasma-mass spectroscopy (ICP-MS),” Clays Clay Minerals, vol. 49,
is reached, especially evident in MMT suspension. The actual pp. 387–392, 2001.
reason behind this change in the trends may be assumed to lie [20] R. P. Gambrell, G. Zhang, R. Nugent, and L. Land, “Physical, chem-
ical, and biological evaluation of the application of biodegradable poly-
in the nature of guar-gum-to-clay binding, after some satura- mers to restoring coastal wetlands,” Coastal Restoration and Enhance-
tion value. We speculate that the additional guar gum, even if ment Through Science and Technology (CREST), 2010, 63 pp.
886 IEEE JOURNAL OF OCEANIC ENGINEERING, VOL. 40, NO. 4, OCTOBER 2015

[21] J. Wang and P. Somachandran, “Study of galactomannose interaction [37] S. I. Pia, “On turbulent flow between rotating cylinders,” Ph.D. disser-
with solids using AFM, IR and allied techniques,” J. Colloid Interface tation, Eng. Appl. Sci. Dept., California Inst. Technol., Pasadena, CA,
Sci., vol. 309, pp. 373–383, 2007. USA, 1939, 90 pp.
[22] B. Zhang, H. Su, X. Gu, X. Huang, and H. Wang, “Effect of structure [38] T. G. Metzger, The Rheology Handbook. Hanover, Germany: Vin-
and charge of polysaccharide flocculants on their flocculation perfor- centz Network GmbH & Co KG, 2006, p. 176.
mance for bentonite suspensions,” Colloids Surfaces A, vol. 436, pp. [39] R. W. Faas and S. I. Wartel, “Rheological properties of sediment
443–449, 2013. suspensions from Eckernforde and Kieler Forde Bays, Western Baltic
[23] A. Alemdar, N. Öztekin, N. Güungüor, Ö. I. Ece, and F. B. Erim, “Ef- Sea,” Int. J. Sediment Res., vol. 21, no. 1, p. 24, 2006.
fects of polyethyleneimine adsorption on the rheological properties [40] Y. Furukawa, A. H. Reed, and G. Zhang, “Effect of organic matter
of purified bentonite suspensions,” Colloids Surfaces A, vol. 252, pp. on estuarine flocculation: A laboratory study using montmorillonite,
95–98, 2005. humic acid, xanthan gum, guar gum and natural estuarine flocs.,”
[24] A. Abelev and P. Valent, “Strain-rate dependence of strength of the Geochem. Trans., vol. 15, no. 1, p. 1, 2014.
Gulf of Mexico soft sediments,” IEEE J. Ocean. Eng., vol. 28, no. 1, [41] Cannon Instruments, Inc., Private Communication 2013.
pp. 25–31, Jan. 2012.
[25] M. Preda and M. E. Cox, “Chemical and mineralogical composition of
marine sediments, and relation to their source and transport, Gulf of
Carpentaria, North Australia,” Mar. Syst., vol. 53, pp. 169–186, 2005.
[26] J. Locat and D. Demers, “Viscosity, yield stress, remolded strength, and
liquidity index relationships for sensitive clays,” Can. Geotech. J., vol.
25, pp. 709–806, 1988.
[27] S. W. Jeong, J. Locat, and S. Leroueil, “The effect of salinity and shear Priyanthi M. Amarasinghe received the Ph.D. de-
history on the rheological characteristics of illite-rich and Na-montmo- gree in geotechnical engineering from North Dakota
rillonite-rich clays,” Clays Clay Minerals, vol. 60, pp. 108–120, 2012. State University, Fargo, ND, USA, in 2009. Her doc-
[28] P. Ofori, A. V. Nguyen, B. Firth, C. McNally, and O. Ozdemir, “Shear- toral research involved effect of clay-fluid molecu-
induced floc structure changes for enhanced dewatering of coal prepa- larinteractions on mechanical properties of expansive
ration plant tailing,” Chem. Eng. J., vol. 172, pp. 914–923, 2011. clay minerals.
[29] K. Schulze, M. Hunger, and D. Doll, “Simulating river flow velocity As a Postdoctoral Fellow and a Research Scien-
on global scale,” Adv. Geosci., vol. 5, pp. 133–136, 2005. tist, she was also involved in molecular modeling of
[30] H.-P. Kozerski, “Determination of areal sedimentation rates in rivers clay minerals using molecular dynamics and discrete
by using plate sediment trap measurements and flow velocity—Settling element method. Recently, as a Research Scientist at
flux relationship,” Water Res., vol. 36, pp. 2983–2990, 2002. Brimrose Corporation, Baltimore, MD, USA, she has
[31] Y. Yuan and R. R. Farnood, “Strength and breakage of activated sludge been working collaboratively with the U.S. Naval Research and Army Research
flocs,” Powder Technol., vol. 199, pp. 111–119, 2010. Laboratories on characterization of semiconductor materials for IR imaging
[32] J. Labille, F. Thomas, M. Milas, and C. Vanhaverbeke, “Flocculation technology.
of colloidal clay by bacterial polysaccharides: Effect of macromolecule
charge and structure,” J. Colloid Interface Sci., vol. 284, pp. 149–156,
2005.
[33] L. Burone, P. Muniz, A. M. S. Pires-Vanin, and M. Rodrigues, “Spatial Andrei Abelev received the M.Sc. and Ph.D. degrees
distribution of organic matter in the surface sediments of Ubatuba Bay in civil engineering with expertise in soil mechanics
(Southeastern—Brazil),” Anais da Academia Brasileira de Ciências, from the Johns Hopkins University, Baltimore, MD,
vol. 75, pp. 77–90, 2003. USA, in 2001.
[34] P. A. Traykovski, P. L. Wiberg, and W. R. Geyer, “Observations and He has been a Civil Engineer and a Research
modeling of wave-supported sediment gravity flows on the Po prodelta Scientist at the Naval Research Laboratory (NRL),
and comparison to prior observations from the Eel shelf,” Continental Washington, DC, USA, since 2005. He conducts
Shelf Res., vol. 27, pp. 375–399, 2007. experimental, theoretical, and numerical studies
[35] B. J. Medhi, A. A. Kumar, and A. Singh, “Apparent wall slip velocity of behavior and structure of marine and terrestrial
measurements in free surface flow of concentrated suspensions,” Int. sediments under various loading conditions, remote
J. Multiphase Flow, vol. 37, pp. 609–610, 2011. sensing characterization of terrain properties, and
[36] G. I. Taylor, “Stability of a viscous liquid contained between two ro- air-borne and in situ exploration of sea ice structure and properties. Prior to
tating cylinders,” Philosoph. Trans. Roy. Soc. Lond. A, vol. 223, pp. NRL, he was a Research Scientist at the University of Southern Mississippi,
289–343, 1923. Stennis Space Center, MS, USA.