Applied Clay Science 87 (2014) 87–96

Contents lists available at ScienceDirect

Applied Clay Science

journal homepage: www.elsevier.com/locate/clay

Research paper

Impact of clay mineral particle morphology on the rheological properties

of dispersions: A combined X-ray scattering, transmission electronic
microscopy and flow rheology study
Christian Blachier a,⁎, Alain Jacquet a, Martin Mosquet a, Laurent Michot b, Christophe Baravian c
a
Lafarge Centre de Recherche, 95 rue du Montmurier, 38291 Saint Quentin Fallavier, France
b
Laboratoire Environnement et Minéralurgie, Université de Lorraine, CNRS, UMR 7569, BP40, 54501 Vandoeuvre lès Nancy Cedex, France
c
Laboratoire d'Energétique et de Mécanique Théorique et Appliquée, Université de Lorraine, CNRS, UMR 7563, BP160, 54501 Vandoeuvre lès Nancy Cedex, France

a r t i c l e i n f o a b s t r a c t

Article history: The rheological features of aqueous dispersions of three clay minerals, illite, kaolinite and montmorillonite are
Received 10 September 2012 analyzed using the notion of effective volume fraction that links the morphological properties of the suspended
Received in revised form 4 November 2013 solids to their hydrodynamical behavior. The morphological information obtained from an appropriate treatment
Accepted 6 November 2013
of the flow curves is confronted to the results obtained by small and wide angle X-ray scattering and transmission
Available online 13 December 2013
electron microscopy. The excellent agreement obtained by independent measurements proves that rheology can
Keywords:
be used as a relevant tool to obtain information about the morphology of clay mineral particles. The fruitfulness of
Clay this approach is further evidenced by analyzing the influence of the presence of various clay minerals on the rhe-
Rheology ology of a concentrated calcareous sand. It was proved that using effective volume fractions, the behavior of var-
Morphology ious sand–clay mineral mixtures can be easily rationalized.
Effective volume © 2013 Elsevier B.V. All rights reserved.
SAXS
TEM

1. Introduction
The rheological properties of natural clay minerals have been exten-
sively studied for the past century due to their importance in numerous
environmental and industrial applications. For instance, the role of clay
mineral in seismic response (Barnes et al., 2002; Matsuda et al., 2004;
Mochizuki et al., 2005; Saffer et al., 2009; Zhu et al., 2011), subaqueous
landslides (Biscontin et al., 2004; Elverhoi et al., 2010; Marr et al., 2002)
and continental landslides (Geertsema and Torrance, 2005; Hungr et al.,
2001; Khaldoun et al., 2009; Wan and Kwong, 2002) is well recognized,
though not fully understood. In terms of industrial applications, the rhe-
ological properties of clay minerals play a crucial role in numerous
fields. Clay mineral based muds have been used for a long time in oil-
drilling operations (Darley and Gray, 1991; Maitland, 2000) where
their mechanical properties stabilize the well. In the cement industry,
clay minerals present in the sands used for cement manufacturing can
have a detrimental effect on the rheological properties of the disper-
sions that leads to water additions that impair the final mechanical
properties (Kroyer et al., 2003).
For all these reasons, numerous studies investigate the rheology of
clay mineral dispersions (Baravian et al., 2003; Brandenburg and
Lagaly, 1988; Callaghan and Ottewill, 1974; Durán et al., 2000; Heller
⁎ Corresponding author. Tel.: +33 474 828 191.
E-mail address: christian.blachier@lafarge.com (C. Blachier).
and Keren, 2001; Keren, 1988; Lagaly, 1989; Lubetkin et al., 1984; M'
Ewen and Mould, 1950; Mourchid et al., 1995; Neumann and Sansom,
1971; Ramos-Tejada et al., 2001; Ramsay, 1986; Rand et al., 1980; Ten
Brinke et al., 2007; Vali and Bachmann, 1988; Willenbacher, 1996).
However, such studies either assess the role of the physico-chemical
conditions (pH, temperature, ionic strength…) on the rheological fea-
tures of the dispersions, or try to find a link between mineralogy and
mechanical properties. For this reason, the present understanding is
rather confused and very few works have tried to provide a general
framework to rationalize the wide spectrum of observed rheological be-
haviors. Studies have recently concentrated their efforts on well-
defined size-selected swelling clay minerals and have shown that parti-
cle anisotropy was a crucial parameter involved in the control of flow
properties (Michot et al., 2004, 2009; Paineau et al., 2011a,b; Philippe
et al., 2011). In the present paper, three different clay minerals, kaolin-
ite, illite and montmorillonite exchanged with calcium, i.e. in conditions
close to the most frequently encountered natural ones, have been stud-
ied. Some experiments on the sodium-exchanged forms are also pre-
sented as this parameter is known to have a strong influence on
dispersion rheology. On the basis of an effective approach, these exper-
iments show how morphological parameters can be directly obtained
from an appropriate treatment of flow curves. The morphological indi-
cators thus obtained are then confronted with the determination of av-
erage aspect ratios by X-ray scattering and transmission electron
microscopy measurements. Finally it was shown that the approach

0169-1317/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.clay.2013.11.004
88 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96
Fig. 1. Schematic representation of a clay mineral particle (aspect ratio = d/e).
developed in the present manuscript can be used to understand the in-
fluence of clay minerals on the rheological properties of granular
suspensions.
2. Materials and methods
Three clay mineral samples were used in this study: montmorillon-
ite Bleu de Sardaigne (Mt), Illite du Puy (Ilt) and Kaolinite BS3 (Kaol).
The raw samples were characterized by X-ray diffraction and infrared
spectroscopy. Such analyses revealed the presence of quartz, feldspar,
Fig. 2. Viscosity as a function of shear stress for different volume fractions of (a) Na+-Mt, (b) Ca2+-Mt, (c) Ca2+-Ilt, (d) Ca2+-Kaol. Continuous black lines correspond to the best fit of Eq. (6).
dolomite and calcite as the major impurities. Prior to use, the samples
were then purified and homoionized under their sodium and calcium
forms (Blachier et al., 2009). A 40 g/L clay mineral dispersion was first
decarbonated during 2 h at 90 °C with 1 M sodium acetate solution at
pH 5 by addition of acetic acid. After centrifugation, homoionization to
Na+ and Ca2+ clay minerals was carried out by exchanging the solid
three times in 1 M NaCl or CaCl2 solution. The dispersion was then dia-
lyzed several times in deionized water until the water was chloride free
(AgNO3 and conductivity test). Finally, the fraction b2 μm was collected
after decantation of the impurities in Imhoff cones. After this purifica-
tion stage, XRD and FT-IR patterns revealed only trace amounts of feld-
spar and dolomite, showing the efficiency of the process.
Highly concentrated clay mineral dispersions were prepared by os-
motic stress to ensure dispersion homogeneity. Dialysis tubes (Visking)
with a molecular mass cut off of 14,000 Da (Paineau et al., 2009) were
placed in a 75 g/L PEG 20,000 (Roth) solution and experiments were
stopped after two weeks. Concentrated clay mineral pastes were recov-
ered and their mass concentrations were determined by mass loss upon
drying one night at 105 °C. Clay mineral dispersions were then prepared
by dispersing various amounts of these clay mineral pastes in MilliQ
Water.
Rheological measurements on clay mineral dispersions were per-
formed using an AR2000 TA Instruments apparatus. The cone and
plate geometry used was 6 cm in diameter with a 2° angle and a
59 μm truncature. All experiments were carried out at 25 °C. A pre-
 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96 89

shearing stage of 2 min at 1000 s−1 was performed before flow exper- polydispersity is present in the system or when particles are non-
iments (up–down ramp 0.1–1000 s−1 logarithmic step, 10 points per spherical. Deviation from a spherical particle shape results in signifi-
decade for a total time around 20 min). cant viscosity increase for a given volume fraction (Barnes et al.,
The morphology of clay mineral particles was characterized by 1989). To take this feature into account, it has been proposed that
transmission electron microscopy (TEM) and small and wide angle X- one should consider the hydrodynamic volume of the particle rather
ray scattering (SAXS–WAXS). Transmission electron micrographs than its real solid volume. In such an approach, an effective hard
were obtained on a Philips CM20 microscope (20 to 200 KeV) on sphere volume fraction (ϕeff) is defined. This accounts for the fluid
samples prepared by depositing diluted clay mineral dispersions volume trapped by the particles (Baravian et al., 2003; Lubansky
(100 mg/L) onto a copper mesh grid. et al., 2005; Mendoza and Santamaria-Holek, 2009, 2010; Petrie,
Small and wide angle X-ray scattering (SAXS–WAXS) experiments 1999; Santamaria-Holek and Mendoza, 2010).
were carried out on dispersions or pastes placed in 1 mm thick capil-

laries at beamline A2 (Hasylab at DESY, Hamburg, Germany) using a ϕeff −2
ηr ¼ 1− : ð2Þ
wavelength of 0.150 nm and sample to detector distances of 3 m and ϕm
30 cm, respectively.
Additional experiments were carried out using a calcareous granular In the case of disk-like particles smaller than a few microns, due to
material formed by mixing three size fractions. The final material, that Brownian motion, nanometric disk-shape objects are free to rotate
represents a classical model of granular materials used in various appli- and trap the suspending fluid around particles that rotates with them.
cations, was then composed of 29 m% of grains with diameters between According to an effective hard sphere model, the volume fraction can
0.5 and 1 mm, 58 m% mass of grains with diameters between 0.1 and therefore be replaced by the hydrodynamic volume of the disks. An ef-
0.5 mm and 13 m% of grains with diameters b 50 μm. Rheological mea- fective volume fraction can then be considered accounting for the
surements were performed using a vane geometry adapted to an AR whole volume of the equivalent sphere of radius R generated by the
1000 rheometer (TA instruments). The vane used is formed with 6
blades, 17 mm high and 11 mm wide and is placed 5 mm above the
bottom of a stainless steel cylinder, 30 mm high. As shown recently
(Ancey and Coussot, 1999; Baravian et al., 2002), such a system is well
adapted to the measurements of pastes as its design prevents most par-
asitic phenomena such as wall slip or segregation. In order to limit sed-
imentation phenomena that perturb measurement, the granular
material was dispersed in a 25% aqueous solution of Emkarox® HV45.
Such a medium is perfectly Newtonian, has a density close to that of
water (1.02 kg/m3) and a viscosity of 0.087 Pa.s, i.e. around two orders
of magnitude higher than that of water. A pre-shearing stage of 2 min at
a shear rate of 50 s−1 was first performed. Flow curves were then mea-
sured under controlled shear rate with up–down ramps between 0.1
and 50 s-1 with logarithmic steps (10 points per decade, 15 s per
point).

3. Results and discussion

3.1. Rheology

The flow behavior of aqueous dispersions is mainly governed by the

solid volume fraction (ϕ). When solid particles are added to a fluid, the
viscosity η of the resulting dispersions increases. Many models have
been proposed to describe this rheological feature by providing expres-
sions relating the solid volume fraction to the relative viscosity
ηr = η/ηf, where ηf is the suspending fluid viscosity. At very low par-
ticle concentration (ϕ b 1%), an analytical solution to the problem can
be obtained and was developed by Einstein (2005) as ηr = 1 + 2.5 ϕ.
This expression was extended to higher volume fractions (ϕ b 5%), by
Batchelor (1977) as ηr = 1 + 2.5 η + (2.5 η)2. Due to the complexity
of multiple hydrodynamic interactions, no analytical expression can
be obtained for higher volume fractions, and effective approaches
must then be used for describing the range of volume fraction in
which viscosity rapidly increases. Taking into account crowding ef-
fects, approaches describing the whole volume fraction range have
then been developed by Krieger and Dougherty (1959) or Quemada
(1977). The expression developed by Quemada and based on mini-
mum energy dissipation principle, writes:


η ϕ −2
ηr ¼ ¼ 1− ð1Þ
ηf ϕm

where (ϕm) is the maximum packing volume fraction parameter that,

for monodisperse spheres, evolves from 0.63 at 0 shear to 0.72 under Fig. 3. Evolution with volume fraction ϕ of (a) infinite shear viscosity and (b) yield stress
infinite shear (Andrew et al., 1991). Such values increase when for Na+-Mt, Ca2+-Mt, Ca2+-Ilt, Ca2+-Kaol.
90 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96

free rotation of the disk-like particle. The value of R depends on the

thickness (e) and diameter (d = 2r) of the particles (Fig. 1). The effec-
tive volume fraction can be calculated as:

V sphere 4 R3
ϕeff ¼ ϕ¼ 2 ϕ ð3Þ
V disk 3r e

with
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
d2 þ e2
R¼ : ð4Þ
2

It can be noted that in the specific case of strongly anisotropic parti-

cles such as Na+-swelling clay mineral (dN N e), the average radius R of
the equivalent sphere reaches the value of r, the radius of the disk-like
particle. The expression of ϕeff is therefore simplified as followed:

4r
ϕeff ≈ ϕ: ð5Þ
3e

Fig. 1 illustrates this concept in the case of a model clay mineral

particle.
Flow curves were adjusted using a viscosity-shear model already
used for clay minerals in order to obtain the rheological parameters
(Baravian et al., 2003). This model inspired from Quemada hard sphere
equation (Quemada, 1977; Quemada and Berli, 2000) links the viscosity
with shear stress (σ) and the infinite-shear viscosity (η∞). The model
equation is:


ð1 þ PeÞ 2
ηðσ Þ ¼ η∞ ð6Þ
χ þ Pe

In this equation, the Peclet number (Pe) links the applied shear
stress to a critical shear stress (σc): when σ ≈ σc, hydrodynamic effects
are comparable to interaction energy and/or thermal Brownian effects.

σ
Pe ¼ : ð7Þ
σc

The χ parameter is an adimensional term whose value indicates the

liquid or solid trend of the sample. This parameter is equal to 1 for New-
tonian fluids, 0 b χ b 1 for shear thinning fluids, −1 b χ ≤ 0 for solid-
Fig. 4. Evolution with effective volume fraction (see text for details) of (a) relative infinite
shear viscosity and (b) yield stress for Na+-Mt, Ca2+-Mt, Ca2+-Ilt, Ca2+-Kaol.

Fig. 5. SAXS and WAXS measurement: Variation of the scattering intensity as a function of
q for Ca2+-Ilt and Ca2+-Kaol dispersions (50 g/L). The star corresponds to a parasitic dif- Fig. 6. SAXS and WAXS measurement: Variation of the scattering intensity as a function of
fraction peak due to the Kapton polymer used as a window. q for Na+-Mt and Ca2+-Mt dispersions (50 g/L).
 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96
Fig. 7. Variation of the average interparticular distance with inverse volume fraction for
Na+-Mt dispersions.
like dispersions at rest. For granular suspensions that can be described
as viscoplastic fluids, χ is always negative and the following equation
can be used to determine the yield stress (σy):
σ y ¼ −χσ c
where σy is the yield stress for solid-like dispersions. This viscosity
model can be fitted by adjustment of the three parameters η∞, χ and
σc using a least square method.
Fig. 2 displays the flow curves obtained for dispersions of Na+-Mt,
Ca -Mt, Ca2+-Ilt and Ca2+-Kaol for increasing solid volume fraction.
2+
All the presented curves were obtained on the decreasing ramp and it
must be pointed out that up and down ramps yield viscosity curves
that are almost superimposed, which shows that thixotropic effects
are negligible and that Fig. 2 presents the steady state flow curves of
the systems investigated in the present paper. For all samples, very di-
lute dispersions exhibit quasi-Newtonian behavior, as the viscosity of
the dispersions is nearly independent of shear stress. Upon increasing
solid concentration, the dispersions display a shear thinning behavior
and a yield stress appears for volume fractions that depend on the
Fig. 8. (a) TEM micrograph and (b) size distribution histograms of Ca2+-Mt bleu de Sardaigne.
91
nature of the clay minerals, and for Mt, on the nature of the compensat-
ing cation. Indeed, the flow curves corresponding to Na+-Mt exhibit a
shear thinning behavior for volume fractions approximately one order
of magnitude lower compared to the other samples. Fig. 2 also displays
the best fits obtained using Eq. (6). It appears that satisfactory fits can be
obtained in all cases. For very high volume fractions, a second inflexion
can appear on the experimental curves near the yield stress that can be
assigned to inhomogeneities in the shear rate distribution inside the
cone and plate geometry. In that case, only the high shear section of
the curve is fitted. In any case, for each of the studied case, it is possible
to determine the yield stress (σy) and the infinite-shear viscosity (η∞)
from such a fitting procedure.
Fig. 3 displays the evolution of η∞/ηwater (Fig. 3a) and σy (Fig. 3b) as a
function of solid volume fraction for the four samples. For all samples,
the sol–gel transition (SGT) can be defined as the volume fraction for
which a yield stress appears. According to such a definition, Na+-Mt
dispersions exhibit a SGT at much lower volume fraction than all
other samples, a classical feature of swelling clay minerals. The SGT
for Ca2+-Mt and Ca2 +-Kaol is rather close around 0.14, whereas the
value is higher for Ca2+-Ilt (ϕsol/gel ≈ 0.2). The evolution of the relative
infinite shear viscosity (Fig. 3a) reveals various interesting features. All
samples display similar shape with a strong increase in viscosity with
volume fraction followed by a pseudo-plateau at higher volume frac-
tions, which is particularly striking in the case of Na+-Mt. Such a
pseudo-plateau reveals that particles orientate along the velocity
streamlines (Michot et al., 2009; Paineau et al., 2011a,b). The viscosity
value at the pseudo-plateau is approximately ten times that of water,
ð8Þ which shows the exceptional shear thinning behavior of clay mineral
dispersions. The curves are shifted towards higher volume fraction in
the order Na+-Mt N Ca2 +-Mt ≈ Ca2 +-Kaol N Ca2 +-Ilt, i.e. the same
order as that observed on the yield stress curves.
The evolution of the infinite shear viscosity as a function of effec-
tive volume fraction (Eq. (2)) with ϕm = 0.72 is shown in Fig. 4a. In
that graph, ϕeff, is obtained for all samples by multiplying the actual
volume fraction by a coefficient α that is adjusted for each sample.
Using such a procedure, all infinite viscosity curves can be adjusted
to Eq. (2) (thick black line in Fig. 4a) for volume fractions before
those corresponding to the appearance of a pseudo-plateau. The
values of α obtained are 16, 3.1, 2.7 and 1.8 for Na+-Mt, Ca2 +-Mt,
Ca2 +-Kaol and Ca2 +-Ilt, respectively. Similar values of α are obtain-
ed using other models describing the evolution of the infinite shear
viscosity as a function of the volume fraction (Mendoza and
Santamaria-Holek, 2009, 2010).
92 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96

Fig. 9. (a) TEM micrograph and (b) size distribution histograms of Ca2+-Ilt du Puy.

According to its definition (Eq. (3)), the values of α are directly relat-
ed to the morphology of the particles. Na+-Mt particles then appear
much more anisometric than all other samples with a much higher
value of α. In that specific case, the fit of the data (Fig. 4a) is slightly
less convincing. This is certainly related to the fact that, as already
shown in the case of Na+-nontronite (Michot et al., 2009), for very an-
isotropic particles the infinite-shear viscosity versus volume fraction
curve can be best fitted using an ellipsoidal model rather than a spher-
ical one. Furthermore, for very anisometric particles the value of ϕm
could be as high as 1 (Michot et al., 2009). The treatment of the curve
of Na+-Mt with ϕm = 1 (not shown) yields an α value of 35. Fig. 4b dis-
plays the evolution with ϕeff of the yield stress obtained from applying
Eqs. (6)–(8). Compared to the results displayed in Fig. 3b, using the no-
tion of effective volume fraction allows rationalizing the behavior of
various clay mineral samples. Simple rheological experiments treated
with an appropriate model then appear as relevant tools to characterize
the morphology of clay minerals, a crucial parameter in numerous ap-
plications. Still, before using such a method any further, it is necessary
to collate the morphological parameters thus obtained with those ob-
tained by more direct techniques.
3.2. SAXS–WAXS measurement
To assess the structure of clay mineral particles in dispersions,
synchrotron-based small and wide angle X-ray scattering experiments
(SAXS–WAXS) were performed. Using such a technique, the structure
of aqueous clay mineral dispersions can be investigated in situ, without
any drying step. Fig. 5 displays, for both Kaol and Ilt, the evolution with
the wave vector q (q = 4πsinθ/λ, where θ is the scattering angle and λ
the wavelength) of the intensity scattered by dispersions at 50 g/L. In
the SAXS region, both curves display monotonous variation of the
scattered intensity with slopes of −3 and −3.2 in log-log scale, for Ilt

Fig. 10. (a) TEM micrograph and (b) size distribution histograms of Ca2+-Kaol AGS BS3.
 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96 93

and Kaol, respectively. In the WAXS region, typical peaks at 10.1 Å and similar to that observed for Ca2+-Mt with even scarcer large particles
7.2 Å are observed for Ilt and Kaol respectively. The (002) reflection of and an average diameter around 40 nm. Very similar results (not
Ilt around 5 Å is also observed. In order to properly take into account shown) are obtained for Na+-Ilt, in agreement with the non-swelling
the form factor of the particle, the intensity of the SAXS patterns was character of the sample.
multiplied by qx where x is the slope of the scattering signal (I = f(q) Micrographs corresponding to Ca2 +-Kaol (Fig. 10) and Na+-Kaol
in logarithmic scale) in the small angle region. (not shown) reveal well-crystallized particles with mainly hexagonal
Using Scherrer's equation, it is then possible in the WAXS region to shapes. In terms of size distribution, the histogram is rather broad,
determine the average number of ordered layers in the elementary par- which shows that even after purification Kaol sample remains polydis-
ticles (Kübler and Jaboyedoff, 2000). Indeed, Scherrer's equation can be perse. The average diameter of Kaol particles can be determined around
written as: 150 nm.
High-resolution TEM experiments were also carried out to obtain
K:λ edge views of some particles, which allow determining directly an aver-
e¼ ð9Þ
ω: cosθ age aspect ratio for the particles. Typical micrographs are presented in
Fig. 11a,b and c, for Ca2 +-Mt, Ca2 +-Ilt and Ca2 +-Kaol, respectively.
where e is the particle thickness, λ the wavelength of the radiation Kaol and Ilt exhibit stacks of layers with nearly constant thickness,
(0.15 nm), θ the angle of the considered Bragg reflection, ω the width whereas the values are more dispersed for Ca2+-Mt (Fig. 11a).
at the half of the intensity of the peak (2θ scale), and K a constant
close to 0.9. Using such an equation, values of 31 nm and 49.7 nm are
obtained for Ilt and Kaol, respectively. This corresponds in the case of
Ilt to around 30 TOT layers and in the case of Kaol to around 70 TO
layers.
Fig. 6 displays the SAXS–WAXS curves obtained for Ca2 +- and
+
Na -Mt. The two curves display significantly different features. In
the case of Ca2 +-Mt, a slope of around − 2.5 is observed in the SAXS
region and a clear peak corresponding to a distance of 19.3 Å is ob-
served. Such a peak corresponding to an interlayer with an average
of three layers of water molecules shows that in the case of Ca2 +-
Mt, due to ion correlations (e.g., Pellenq et al., 1997), the layers are as-
sociated in packs of a few layers (Hight et al., 1962).
Application of Scherrer's formula to this peak yields a coherent scat-
tering domain around 15 nm, i.e. particles of 8 layers. In the case of
Na+-Mt, the curve in the SAXS region exhibits a slope of − 2 in log-
log scale, whereas no correlation peak is obtained in the WAXS region.
The combination of these two features reveals that dispersions of
Na+-Mt are formed with dispersed clay mineral layers. Still, a diffuse
peak can be observed in the low angle region that reveals correlation
distances between layers. This peak can be used to obtain information
on the thickness of the objects in the dispersion. Indeed, upon variation
of the clay mineral concentration in the dispersions, the position of this
peak changes. It was shown recently (Michot et al., 2006, 2008, 2009;
Paineau et al., 2009, 2011a,b) that in the case of swelling clay mineral
samples at low ionic strength, and in the gel region the interparticular
distance D (with D = 2π/q* where q* is the wave vector where the
peak is observed) evolves as ϕ−1 where ϕ is the volume fraction. Such
lamellar swelling allows deducing the thickness of the individual scat-
tering objects that is the slope of the curve D = f(1/ϕ). Fig. 7 displays
in the case of the sample investigated in the present study, the evolution
of D vs. ϕ. According to such data, the average thickness obtained is
8.57 Å. Such a value is near the theoretical value of the interlayer dis-
tance (d001 = 9.6 Å) (Bergaya and Lagaly, 2013) showing the layers
of Na+-Mt are almost completely dispersed.

3.3. TEM characterization

Whereas SAXS–WAXS provides information about the thickness of

the particles, the lateral dimensions of the samples can be measured
by TEM. Fig. 8 displays a typical micrograph obtained on Ca2+-Mt. The
shape of the particles is rather irregular, but in first approximation,
they can be considered as disks. The diameter can be calculated as the
longest chord inside a particle. The size distribution histogram deduced
from the analysis of 3 micrographs (around 100 particles) reveals that
most particles have sizes ranging between 10 and 100 nm. Still, scarce
larger particles (up to the micrometer) can also be observed and the
resulting average diameter is then 50 nm.
Ca2+-Ilt particles (Fig. 9) appear with well-defined edges and main- Fig. 11. Lateral view of (a) Ca2+-Mt, (b) Ca2+-Ilt, (c) Ca2+-Kaol by High Resolution TEM
ly display a polygonal shape. The size distribution histogram is rather microscopy.
94 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96

Table 1 50% as, for lower values, the up and down ramps are not superimposed
Morphological parameters of the three clay minerals obtained from analysis of TEM micro- any longer and the shape of the curves clearly indicates the occurrence
graphs and SAXS–WAXS measurements.
of sedimentation effects (Fig. 12, bottom curve).
Na+-Mt Ca2+-Mt Ca2+-Ilt Ca2+-Kaol
α
Rheologie 16 3.1 1.8 2.7
SAXS–WAXS Thickness(nm) 0.86 15 31 50
TEM Thickness(nm) – 16 30 47
Diameter(nm) 50 50 40 150
α
Morphology 38 2.5 1.8 2.4

Table 1 summarizes the main morphological data obtained by SAXS–

WAXS and by TEM. It appears that in terms of thickness, both methods
lead to very similar results for all samples.
Table 1 also shows a comparison between the values of α derived
from the rheological analysis and the values than can be directly obtain-
ed from morphological analyses using Eq. (3). Except for Na+-Mt, the
agreement between values derived using totally independent methods
is rather remarkable. As mentioned earlier, the discrepancy observed for
Na+-Mt is likely related to the fact that for such very anisometric sys-
tems, an ellipsoidal model would be more adapted (Michot et al.,
2009). Furthermore, as already reported, the use of a ϕm value of 1 for
this system yields an α value of 35, i.e. very close to that deduced
from SAXS and TEM measurements. In any case, the results presented
here clearly show that the rheological behavior of numerous clay miner-
al systems can be understood on the basis of their anisotropic features
and that appropriate treatment of rheological curves can yield valuable
morphological information.

3.4. Sand suspensions

To further illustrate this point, we analyze the rheological behavior

of a complex system formed with a polydisperse sand system in
which small proportions of various clay mineral samples are added.
Fig. 12 displays the flow curves obtained for pure sand. As mentioned
in the materials and methods section, to avoid sedimentation effects,
the experiments were carried out in a higher viscosity fluid, i.e. a 25%
aqueous solution of Emkarox® HV45. In such conditions, valid experi-
ments can be obtained for high volume fractions as the curves corre-
sponding to either increasing or decreasing shear stress are nearly
superimposed (Fig. 12). This holds for volume fractions higher than

Fig. 12. Viscosity as a function of shear stress for different volume fractions of calcareous
sand. Empty symbols correspond to increasing shear stress and filled ones to decreasing Fig. 13. Evolution of the yield stress with volume fraction for a calcareous sand filled with
shear stress. Thick black lines correspond to the best fit of Eq. (6). various proportions of different clay minerals. (a) 2% (b) 3% and (c) 5% clay mineral added.
 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96
Fig. 13 displays the evolution with ϕ of the yield stress of the same
granular material upon addition of various proportions of different
clay mineral materials, i.e. Ilt du Puy, Sardaigne Mt, Arizona Mt and Wy-
oming Mt. These two latter materials were chosen after size selection
(Michot et al., 2004; Paineau et al., 2011a), as they present average par-
ticle dimensions of 200 and 400 nm, respectively, i.e. significantly
higher than both Ilt and Sardaigne Mt (see above). As revealed by
Fig. 13, anisotropy has a crucial influence on the yield stress, the more
anisotropic particles leading to the development of a yield stress at
lower total volume fraction, whatever the amount of clay mineral
added in the sand. Furthermore, the addition of increasing clay mineral
amounts shifts the yield stress curves towards significantly lower vol-
ume fractions.
In order to try to rationalize all these results in terms of particle an-
isotropy, it is first necessary to analyze the rheological behavior of the
various clay minerals in Emkarox® HV45 as, due to possible interactions
between clay mineral materials and Emkarox, the morphological prop-
erties of the particles may be different from those observed in water.
Fig. 14a displays the evolution with volume fraction of the yield stress
in Emkarox for the 4 clay mineral samples added to the sand. It exhibits
the same order as that deduced from Fig. 13.
Fig. 14. Evolution with (a) volume fraction and (b) effective volume fraction of the yield
stress of various clay minerals in Emkarox.
95
Using the same approach as that carried out for samples dispersed
in water (Fig. 4), it is possible to use the notion of effective volume
fraction by using for each clay mineral material a coefficient α that is
related to the morphology of the objects in dispersion. Using values of
6, 8.5, 11.5 and 18.5 for Ilt du Puy, Sardaigne, Arizona and Wyoming Mt
samples, respectively, leads to the curves displayed in Fig. 14b. It clearly
appears that using such an approach allows obtaining a master curve
for the evolution of yield stress with volume fraction in Emkarox. The
values thus obtained should then in principle allow understanding
the effects of various clay minerals on the rheological behavior of
sand–clay mineral mixtures. Indeed, for each sand–clay mineral mix-
ture an effective volume fraction ϕeff can be defined as ϕeff = β *ϕ
where β = 1*m%sand + α*dsand/dclay*m%clay where α is the same as
that determined from Fig. 14.
As shown in Fig. 15, using such a procedure clearly rationalizes the
behavior of all the investigated mixtures as a single master curve relat-
ing yield stress to volume fraction can then be obtained.
4. Conclusions
The present study clearly shows that the rheological behavior of
aqueous clay mineral dispersions can be rationalized on the basis of
their morphological features. Furthermore, an adequate treatment of
flow curves based on an effective approach allows determining a pa-
rameter that is directly related to the aspect ratio of the particles. The
collation of this parameter with a combination of SAXS–WAXS and
TEM data confirms the relevance of the proposed treatment. It then ap-
pears that simple rheological experiments carried out with a systematic
variation in volume fraction can provide direct morphological informa-
tion on clay minerals. As shown by simple experiments, the approach
developed in the present paper for pure aqueous dispersions can be ex-
tended to understand the role of clay minerals on the rheological fea-
tures of granular materials. This can be of prime importance in
numerous industrial and environmental situations where clay minerals
are a minor component of complex mixtures (soils, pastes, cement
mixes…). Finally, it must be emphasized that the approach developed
in the present paper can in principle be able to analyze dispersions
with either attractive or repulsive interactions. In the case of repulsive
interactions such as those described in the present study, the morphol-
ogy of individual particles should be considered. In the case of attractive
Fig. 15. Evolution of the yield stress with effective volume fraction (see text for details) for
a calcareous sand filled with various proportions of different clay minerals.
96 C. Blachier et al. / Applied Clay Science 87 (2014) 87–96
interactions, aggregate structure and their hydrodynamical properties
have to be taken into account. It would certainly be worthwhile in
that context to analyze the rheological behavior of clay minerals at dif-
ferent pH and ionic strengths as such conditions are also relevant to nu-
merous fields such as sedimentation in estuaries, water treatment or
cement manufacturing.
References
Ancey, C., Coussot, P., 1999. Transition from frictional to viscous regime for granular sus-
pensions. C.R. Acad. Sci. Paris Sér. II 327, 215–222.
Andrew, D., Jones, R., Leary, B., Boger, D.V., 1991. The rheology of a colloidal suspension of
hard spheres. J. Colloid Interface Sci. 147, 479–495.
Baravian, C., Lalante, A., Parker, A., 2002. Vane geometry with a large finite gap. Appl.
Rheol. 45, 1421–1439.
Baravian, C., Vantelon, D., Thomas, F., 2003. Rheological determination of interaction po-
tential energy for aqueous clay suspensions. Langmuir 19, 8109–8114.
Barnes, H.A., Hutton, J.F., Walters, K., 1989. An introduction to rheology. Rheology series,
3. Elsevier0-444-87140-3.
Barnes, P.M., Nicol, A., Harrison, T., 2002. Late Coenozoic evolution and earthquake poten-
tial of an active listric thrust complex above the Hikurangi subduction zone, New
Zealand. Geol. Soc. Am. Bull. 114, 1379–1405.
Batchelor, G.K., 1977. The effect of Brownian motion on the bulk stress in a suspension of
spherical particles. J. Fluid Mech. 83, 97–117.
Bergaya, F., Lagaly, G., 2013. Handbook of Clay Science, 2nd ed. Elsevier.
Biscontin, G., Pestana, J.M., Nadim, F., 2004. Seismic triggering of submarine slides in soft
cohesive soil deposits. Mar. Geol. 203, 341–354.
Blachier, C., Michot, L.J., Bihannic, I., Barrès, O., Jacquet, A., Mosquet, M., 2009. Adsorption
of polyamine on clay minerals. J. Colloid Interface Sci. 336, 599–606.
Brandenburg, U., Lagaly, G., 1988. Rheological properties of sodium montmorillonite dis-
persions. Appl. Clay Sci. 3, 263–279.
Callaghan, I.C., Ottewill, R.H., 1974. Interparticle forces in montmorillonite gels. Faraday
Discuss. 57, 110–118.
Darley, H.C.H., Gray, G.R., 1991. Composition and Properties of Drilling and Completion
Fluids, 5th edition. Gulf Publishing Company, Houston.
Durán, J.D.G., Ramos-Tejada, M.M., Arroyo, F.J., González-Caballero, F., 2000. Rheological
and electrokinetic properties of sodium montmorillonite suspensions. J. Colloid Inter-
face Sci. 2000 (229), 107–117.
Einstein, A., 2005. Zur Theorie der Brownschen Bewegung. Ann. Phys. (Leipzig) 14, 248–258.
Elverhoi, A., Breien, E., De Blasio, F., Harbitz, C., Pagliardi, M., 2010. Submarine landslides
and the importance of the initial sediment composition for run-out length and final
deposit. Ocean Dyn. 60, 1027–1046.
Geertsema, M., Torrance, J.K., 2005. Quick clay from the Mink Creek landslide near terrace,
British Columbia: geotechnical properties, mineralogy, and geochemistry. Can.
Geotech. J. 42, 907–918.
Heller, H., Keren, R., 2001. Rheology of Na-rich montmorillonite suspension as affected by
electrolyte concentration and shear rate. Clay Clay Miner. 49, 286–291.
Hight Jr., R., Higdon, W.T., Darley, H.C.H., Schmidt, P.W., 1962. Small Angle X-ray Scatter-
ing from montmorillonite suspension. J. Chem. Phys. 37, 502–510.
Hungr, O., Evans, S.G., Bovis, M.J., Hutchinson, J.N., 2001. A review of the classification of
landslides of the flow type. Environ. Eng. Geosci. 7, 221–238.
Keren, R., 1988. Rheology of aqueous suspension of sodium/calcium montmorillonite. Soil
Sci. Soc. Am. J. 52, 924–928.
Khaldoun, A., Moller, P., Fall, A., Wegdam, G., De Leeuw, B., Meheust, Y., Fossum, J.O., Bonn,
D., 2009. Quick clays and landslides of clayey soils. Phys. Rev. Lett. 103.
Krieger, I.M., Dougherty, T.J., 1959. A mechanism for non-Newtonian flow in suspensions
of rigid spheres. Trans. Soc. Rheol. 3, 137–152.
Kroyer, H., Lindgren, H., Jakobsen, H.J., Skibsted, J., 2003. Hydration of Portland cement in
the presence of clay minerals studied by 29Si and 27Al MAS NMR spectroscopy. Adv.
Cem. Res. 15, 103–112.
Kübler, B., Jaboyedoff, M., 2000. Illite crystallinity. C.R. Acad. Sci. Paris Earth Planet. Sci.
331, 75–89.
Lagaly, G., 1989. Principles of flow of kaolin and bentonite dispersions. Appl. Clay Sci. 4,
105–123.
Lubansky, A.S., Boger, D.V., Cooper-White, J.J., 2005. Batchelor's theory extended to elon-
gated cylindrical or ellipsoidal particles. J. Non-Newtonian Fluid Mech. 130, 57–61.
Lubetkin, S.D., Middleton, S.R., Ottewill, R.H., 1984. Some properties of clay–water disper-
sions. Philos. Trans. R. Soc. Lond. A 311, 353–368.
M' Ewen, M.B., Mould, D.L., 1950. Gelation of montmorillonite. Nature 166, 437–438.
Maitland, G.C., 2000. The colloid and interface science of oil and gas production. Curr.
Opin. Colloid Interface Sci. 5, 301–311.
Marr, J.G., Elverhoi, A., Harbitz, C., Imran, J., Harff, P., 2002. Numerical simulation of mud-
rich subaqueous debris flows on the glacially active margins of the Svalbard-Barents
sea. Mar. Geol. 188, 351–364.
Matsuda, T., Omura, K., Ikeda, R., Arai, T., Kobayashi, K., Shimada, K., Tanaka, H., Tomita, T.,
Hirano, S., 2004. Fracture-zone conditions on a recently active fault: insights from
mineralogical and geochemical analyses of the hirabayashi NIED drill core on the
Nojima fault Southwest Japan which ruptured in the 1995 Kobe earthquake.
Tectonophysics 378, 143–163.
Mendoza, C.I., Santamaria-Holek, I., 2009. The rheology of hard sphere suspensions at ar-
bitrary volume fractions: an improved differential viscosity model. J. Chem. Phys. 130,
044904.
Mendoza, C.I., Santamaria-Holek, I., 2010. Rheology of concentrated emulsions of spheri-
cal droplets. Appl. Rheol. 20, 23493.
Michot, L.J., Bihannic, I., Porsch, K., Maddi, S., Baravian, C., Mougel, J., Levitz, P., 2004. Phase
diagrams of Wyoming Na-Montmorillonite clay: influence of particle anisotropy.
Langmuir 20, 10829–10837.
Michot, L.J., Bihannic, I., Maddi, S., Funari, S.S., Baravian, C., Levitz, P., Davidson, P., 2006.
Liquid-crystalline aqueous clay suspensions. Proc. Natl. Acad. Sci. U. S. A. 103,
16101–16104.
Michot, L.J., Bihannic, I., Maddi, S., Baravian, C., Levitz, P., Davidson, P., 2008. Sol/gel and
isotropic/nematic transitions in aqueous suspensions of natural nontronite clay. In-
fluence of particle anisotropy. 1. Features of the I/N transition. Langmuir 24,
3127–3139.
Michot, L.J., Baravian, C., Bihannic, I., Maddi, S., Moyne, C., Duval, J.F.L., Levitz, P., Davidson,
P., 2009. Sol/gel and isotropic/nematic transitions in aqueous suspensions of natural
nontronite clay. Influence of particle anisotropy. 2. Gel structure and mechanical
properties. Langmuir 25, 127–139.
Mochizuki, K., Nakamura, M., Kasahara, J., Hino, R., Nishino, M., Kuwano, A., Nakamura, Y.,
Yamada, T., Shinohara, M., Sato, T., Moghaddam, P.P., Kanazawa, T., 2005. Intense PP
reflection beneath the aseismic forearc slope of the Japan Trench subduction zone
and its aseismic slip subduction. J. Geophys. Res. Solid Earth 110, 1–16.
Mourchid, A., Delville, A., Lambard, J., Lécolier, E., Levitz, P., 1995. Phase diagram of colloi-
dal dispersions of anisotropic charged particles: equilibrium properties, structure and
rheology of laponite suspensions. Langmuir 111, 1942–1950.
Neumann, B.S., Sansom, K.G., 1971. The rheological properties of dispersions of laponite, a
synthetic hectorite-like clay, in electrolyte solutions. Clay Miner. 9, 231–243.
Paineau, E., Antonova, K., Baravian, C., Bihannic, I., Davidson, P., Dozov, I., Imperor-Clerc,
M., Levitz, P., Madsen, A., Meneau, F., Michot, L.J., 2009. Liquid-crystalline nematic
phase in aqueous suspensions of a disk-shaped natural beidellite clay. J. Phys.
Chem. B 113, 15858–15869.
Paineau, E., Michot, L.J., Bihannic, I., Baravian, C., 2011a. Aqueous suspensions of natural
swelling clay minerals. 2. Rheological characterization. Langmuir 27, 7806–7819.
Paineau, E., Bihannic, I., Baravian, C., Philippe, A.M., Davidson, P., Levitz, P., Funari, S.S.,
Rochas, C., Michot, L.J., 2011b. Aqueous suspensions of natural swelling clay minerals.
1. Structure and electrostatic interactions. Langmuir 27, 5562–5573.
Pellenq, R.J.M., Caillol, J.M., Delville, A., 1997. Electrostatic attraction between two
charged surfaces: a (N, V, T) Monte Carlo simulation. J. Phys. Chem. B 101,
8584–8594.
Petrie, C.J.S., 1999. The rheology of fibre suspension. J. Non-Newtonian Fluid Mech. 87,
369–402.
Philippe, A.M., Baravian, C., Imperor-Clerc, M., De Silva, J., Paineau, E., Bihannic, I.,
Davidson, P., Meneau, F., Levitz, P., Michot, L.J., 2011. Rheo-SAXS investigation of
shear-thinning behaviour of very anisometric repulsive disc-like clay suspensions.
J. Phys. Condens. Matter 23, 194112.
Quemada, D., 1977. Rheology of concentrated disperse systems and minimum energy dis-
sipation principle. I. Viscosity-concentration relationship. Rheol. Acta 16, 82–94.
Quemada, D., Berli, C.L.A., 2000. Rheological modeling of microgel suspensions involving
solid–liquid transition. Langmuir 16, 7968–7974.
Ramos-Tejada, M.M., Arroyo, F.J., Perea, R., Durán, J.D.G., 2001. Scaling behavior of the
rheological properties of montmorillonite suspensions: correlation between in-
terparticle interaction and degree of flocculation. J. Colloid Interface Sci. 235,
251–259.
Ramsay, J.D.F., 1986. Colloidal properties of synthetic hectorite clay dispersions: I. Rheol-
ogy. J. Colloid Interface Sci. 109, 441–447.
Rand, B., Pekenc, E., Goodwin, J.W., Smith, R.W., 1980. Investigation into the existence of
edge-face coagulated structures in Na-montmorillonite suspensions. J. Chem. Soc.
Faraday Trans. 76, 225–235.
Saffer, D.M., Frye, K.M., Marone, C., Mair, K., 2009. Laboratory results indicating complex
and potentially unstable frictional behavior of smectite clay. Geophys. Res. Lett. 28,
2297–2300.
Santamaria-Holek, I., Mendoza, C.I., 2010. The rheology of concentrated suspensions of
arbitrarily-shaped particles. J. Colloid Interface Sci. 346, 118–126.
Ten Brinke, A.J.W., Bailey, L., Lekkerkerker, H.N.W., Maitland, G.C., 2007. Rheology modifi-
cation in mixed shape colloidal dispersions. Part I: pure components. Soft Matter 3,
1145–1162.
Vali, H., Bachmann, L., 1988. Ultrastructure and flow behavior of colloidal smectite disper-
sions. J. Colloid Interface Sci. 126, 278–291.
Wan, Y.S., Kwong, J., 2002. Shear strength of soils containing amorphous clay-size mate-
rials in a slow-moving landslide. Eng. Geol. 65, 293–303.
Willenbacher, N., 1996. Unusual thixotropic properties of aqueous dispersions of laponite
RD. J. Colloid Interface Sci. 182, 501–510.
Zhu, J.B., Perino, A., Zhao, G.F., Barla, G., Ki, J.C., Ma, G., Zhao, J., 2011. Seismic response of a
single and a set of filled joints of viscoelastic deformational behaviour. Geophys. J. Int.
186, 1315–1330.