Colloids and Surfaces A: Physicochem. Eng.

Aspects 468 (2015) 103–113

Contents lists available at ScienceDirect

Colloids and Surfaces A: Physicochemical and

Engineering Aspects
journal homepage: www.elsevier.com/locate/colsurfa

Physicochemical behaviors of kaolin slurries with and without

cations—Contributions of alumina and silica sheets
Lavanya Avadiar a , Yee-Kwong Leong a,∗ , Andy Fourie b
a
School of Mechanical and Chemical Engineering (M050), The University of Western Australia (UWA), 35 Stirling Highway, Crawley, WA 6009, Australia
b
School of Civil and Resource Engineering (M051), The University of Western Australia (UWA), 35 Stirling Highway, Crawley, WA 6009, Australia

h i g h l i g h t s g r a p h i c a l a b s t r a c t

• Zeta potential-pH and yield stress-pH

behaviors observed.
• Silica and kaolin behaviors were sim-
ilar to each other with and without
cations.
• However, alumina’s behaviors were
less similar to (silica’s) and kaolin’s.
• Silica sheets compared to alumina
sheets control kaolin behaviors more.
• Silica and kaolin particle interactions
with hydrolyzed products could be
similar.

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

Article history: Numerous studies on the surface charge-pH and yield stress-pH behaviors of kaolin slurries with respect
Received 9 October 2014 to the pH-dependency of alumina (Al) sheets, silica (Si) sheets and edges in kaolin particles have been
Received in revised form 4 December 2014 carried out. There is, however, still a paucity of studies that explain if the alumina sheets or the silica
Accepted 9 December 2014
sheets in kaolin particles contribute in greater extents to the surface charges of kaolin particles that will
Available online 19 December 2014
consequently affect the kaolin particle interactions and yield stresses. In this study, the contributions
of the alumina and the silica sheets to the zeta potential-pH and yield stress-pH behaviors of Riedel
Keywords:
kaolin particles, which display no isoelectric points (IEPs), are analyzed by comparing these behaviors of
Cationic hydrolysed products
Alumina slurries
Riedel kaolin slurries to that of alumina and silica slurries with and without Mg2+ and Ca2+ cations. With
Silica slurries and without cations, the behaviors of Riedel kaolin slurries were more similar to the behaviors of silica
Kaolin slurries compared to alumina slurries. The principal reason for this similarity is due to the high concentrations of
Zeta potential-pH negative charges on both silica and Riedel kaolin particles at all pH compared to the lower concentrations
Yield stress-pH of negative charges on alumina particles. This explains that the permanently negatively charged silica
sheets contribute in greater extents to the surface charge-pH and yield stress-pH behaviors of Riedel
kaolin slurries compared to the alumina sheets.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction

Abbreviations: CR, charge reversals; IEPs, isoelectric points; BET, Brunauer,
Emmett, Teller; SEM, scanning electron microscopy; PSDs, particle size distribu-
tions; wt%, weight percentages; ESA, electrokinetic sonic amplitude; dwb%, dry
weight percentages; HRTEM, high-resolution transmission electron microscopy.
∗ Corresponding author. Tel.: +61 8 6488 3602; fax: +61 8 6488 1024.
E-mail addresses: avadil01@student.uwa.edu.au (L. Avadiar),
yeekwong.leong@uwa.edu.au (Y.-K. Leong), andy.fourie@uwa.edu.au (A. Fourie).
Kaolin particles comprise of one octahedral alumina (Al) sheet
and one tetrahedral silica (Si) sheet that fit together to form the
basic repeating unit layer of the particle (Fig. 1) [1]. In the alumina
and silica sheets, isomorphic substitution of Mg2+ for Al3+ and of
Al3+ for Si4+ respectively occurs. This substitution results in silica
sheets to be pH-independent and permanently negatively charged.

http://dx.doi.org/10.1016/j.colsurfa.2014.12.019
0927-7757/© 2015 Elsevier B.V. All rights reserved.
104 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113
Fig. 1. A schematic diagram representing the kaolin crystal layer structure and the
primary building blocks of the kaolin particle.
Adapted from Palomino and Santamarina [1].
On the other hand, despite the substitution, the presence of surface
hydroxyls ( OH) on the alumina sheets causes these sheets to dis-
play patch-wise charge heterogeneity that is dependent on the pH
and ionic strength of surrounding medium. This results in alumina
sheets to display an isoelectric point (IEP) between pH 6.0–8.0 [2].
At the edges of the kaolin particles, bonds on these alumina and sil-
ica sheets are broken to comprise of aluminol (Al OH) and silanol
(Si OH) groups respectively. These groups (i.e. broken bonds) cause
the edges on kaolin particles to display charge heterogeneity and
carry an IEP between pH 5.0 and 7.0 [3].
Numerous studies, including studies by Lagaly [4], Flegmann
et al. [5] and Rand and Melton [6] have analyzed kaolin particle
interactions and consequent kaolin rheological behaviors in rela-
tion to these surface charge-pH properties of alumina sheets, silica
sheets and especially edges in kaolin particles. Teh et al. [7], via
sequential zeta potential-pH and yield stress-pH measurements
on the same kaolin suspensions, have shown that most (low CaO
content) kaolin slurries obey the yield stress-DLVO force model
(i.e. the yield stress decreases linearly with increasing square of
zeta potential) to indicate that high kaolin yield stresses occurs at
low kaolin particle zeta potential magnitudes. These studies have
indicated that at low pH when alumina sheets and edges are pos-
itively charged (i.e. when average zeta potential magnitudes are
low), particle interactions and thus high yield stresses occur within
most kaolin slurries where at high pH when alumina sheets, sil-
ica sheets and edges are all negatively charged (i.e. when average
zeta potential magnitudes are high), weak interactions or complete
de-flocculation and thus low or negligible yield stresses are respec-
tively recorded within most kaolin slurries. This explains the yield
stress-zeta potential-pH behavior of pristine kaolin slurries with-
out cations.
The zeta potential-pH and yield stress-pH behaviors of kaolin
slurries with cations have also been analyzed in literature [8,9]. For
example, Mpofu et al. [8] stated that the first hydrolysis product
of Ca(II) ions forms at pH > 10.5 to increase dramatically in con-
centrations at pH 10.5–14 that caused the reductions in (negative)
zeta potential magnitudes and induced a charge reversal (CR) from
negative to positive charges at pH > 10 within kaolin slurries. They
attributed the CR to the specific adsorption of Ca(OH)+ and possi-
bly the surface nucleation of Ca(OH)2 [10]. Atesok et al. [11] also
observed that Ca adsorption concentrations on Na-kaolinite parti-
cles increased with pH to display a sharp increase at pH > 10 where
they attributed the dramatic reduction in negative zeta potentials at
pH 11.3 to the adsorption of CaOH+ . Divalent Mg2+ and Ca2+ cations
hydrolyze to form hydroxy complexes (Mg(OH)+ and Ca(OH)+ ) and
hydroxide precipitates (Mg(OH)2 and Ca(OH)2 ) at high pH that
adsorb to possibly alter and reverse kaolin particle surface charges
from negative to positive at the specific (high) pH regions of prod-
uct adsorptions [12]. In addition, these adsorptions could enhance
or cause variations in kaolin particle interaction forces at the spe-
cific pH regions of hydrolyzed product adsorption to alter the kaolin
slurry flow, i.e. rheological-pH behavior [5,8,13].
Particles could also (naturally) carry cations in sufficiently high
concentrations to display varied physicochemical behaviors. An
example is the Unimin kaolin particle, which is understood to
carry sufficiently high Ca2+ concentrations. The presence of Ca(II)
hydrolyzed products is observed to cause Unimin kaolin slurries to
display high yield stresses at high pH while most kaolin slurries,
such as Riedel, Sigma and Fluka, with negligibly high cationic con-
centrations commonly display high yield stresses only at low pH
[7,9].
While knowledge on the yield stress-zeta potential-pH rela-
tion of (most) pristine kaolin slurries and literature on the effects
of cations on the zeta potential and yield stress of these kaolin
slurries are abundant, the knowledge on if it is the alumina
or the silica sheets that contribute in greater extents to these
zeta potential (i.e. surface charge) and yield stress (i.e. particle
interactions and rheology) behaviors of kaolin slurries with and
without cations is still scarce. Due to the varied surface charge-
pH densities of alumina sheets vs. silica sheets in kaolin particles,
it is possible that cationic hydrolyzed products could adsorb in
greater extents on one of these interacting sites that would control
the behaviors of kaolin particles. While the dominant interacting
particle site that controls the behaviors of homogeneous parti-
cles such as silica and alumina are clearly evident, identifying
the dominant interacting site of kaolin particles, which comprise
of heterogeneous surface charge densities, is not as straight-
forward.
As a result, the aim of this study was to analyze the contrib-
utions of the alumina sheets and the silica sheets in the Riedel
kaolin particles to the zeta potential-pH and yield stress-pH behav-
iors of these Riedel kaolin particle slurries. (This study did not
include analysis on the edges of the Riedel kaolin particles.) Sim-
ilar to most kaolin particles, Riedel kaolin particles (which are of
low CaO content) display no IEP (or IEP at very low pH) [9] and
obey the yield stress-DLVO force model [14] to display high yield
stresses at low zeta potential magnitudes and vice versa. In order
to analyze the contributions of the alumina and silica sheets to
Riedel kaolin behaviors, the zeta potential-pH and the yield stress-
pH behaviors of Riedel kaolin slurries were compared to the zeta
potential-pH and the yield stress-pH behaviors of alumina and sil-
ica slurries with and without Mg2+ and Ca2+ cations. The yield
stress-pH data comparisons is understood to confirm or alter possi-
ble inferences made from the zeta potential-pH data comparisons,
which could involve discrepancies with regard to the kaolin par-
ticle zeta potentials measured (explained in Section 2.2). Alumina
and silica slurries were chosen for comparison to the Riedel kaolin
slurries as the constituent elements in kaolin particles are alu-
mina and silica (Fig. 1). Similarities in the behaviors of the alumina
and/or the silica slurries to the Riedel kaolin slurries indicated
that the alumina and/or the silica sheet(s) contributed in greater
extents to the Riedel kaolin surface charges and particle interac-
tions.
 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113 105

Fig. 2. SEM images of (a) alumina (Alusion), (b) silica (SFP-30 M) and (c) Riedel kaolin particles. Magnifications and scales at: (a) 5000× and 10 ␮m, (b) 25,000× and 1 ␮m
and (c) 20,000× and 1 ␮m.

2. Materials and methods platelet shapes due to the covalent bonding interactions between
the particles.
2.1. Materials and their characterization The particle size distributions (PSDs) of the alumina and silica
powders were measured on the Malvern Mastersizer Microplus in
The platelet ␣-Al2 O3 (alumina) powder used is a commercial water at pump speeds of 1400 rpm under ultrasonic power and
sample with a trade name Alusion that is sourced from Advanced with polyphosphate dispersant added. The PSD of Riedel kaolin
Nanotechnology (ANT) Limited, now known as Antaria, in Perth, powder was measured similarly. The alumina particles displayed
Australia. The spherical SiO2 (silica) powder used is a commercial a narrower size distribution compared to the silica and Riedel
sample with a trade name SFP-30M that is sourced from Denka, kaolin particles (Fig. 3). While the d50 of the alumina particles
Japan. The kaolin powder used is the Riedel kaolin powder used in was 10.17 ␮m, the d50 of the Riedel kaolin particles was lower at
[9]. For complete characterization of the Riedel kaolin powder, refer 4.00 ␮m and the d50 of the silica particles was lowest at 2.83 ␮m.
to [9]. Both alumina and silica powders used comprise of >99.5% of This indicates that the alumina powder comprises of larger parti-
␣-Al2 O3 and SiO2 concentrations. The insignificant impurity con- cles with similar sizes compared to the smaller kaolin and smallest
centrations indicate the insignificant effect of impurities on the silica particles with slightly divergent particle sizes.
physicochemical behaviors of alumina and silica slurries. Kaolin particles comprise of non-uniform, heterogeneous
The Brunauer, Emmett, Teller (BET) surface areas of the alu- charge densities with respect to pH and ionic strength. These het-
mina and silica powders is 1.8 [15,16] and 6.2 m2 /g respectively. erogeneous charge densities are attributed to the isomorphous
The lower BET surface area of the alumina powders could indi- substitutions in alumina and silica sheets and the protonation or
cate the larger thicknesses and the larger particle sizes of the
alumina than the silica particles. However, comparisons between
these BET surface areas could comprise of discrepancies if BET sam-
ple preparation conditions such as the particle drying temperatures
employed were varied [9].
The scanning electron microscopy (SEM) images of the alumina
and silica powders were obtained using a Zeiss 55 field emission
SEM and sample preparation and imaging methods similar to when
imaging the Riedel kaolin powder (Supplementary data of [9]).
According to the magnifications and scales of the respective SEM
images, the alumina followed by Riedel kaolin and then silica par-
ticles are in order of decreasing sizes (Fig. 2). The alumina and silica
particles were also observed to vary morphologically to the Riedel
kaolin particles. These variations included the lower concentra-
tions of micro-islands on the platelet alumina particles than on the
platelet Riedel kaolin particles and the spherical shapes of the silica
particles. Silica particles generally exist to be in spherical and not
Fig. 3. PSDs of alumina (Alusion), silica (SFP-30 M) and Riedel kaolin powders.
106 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113
de-protonation of Al OH and Si OH groups in alumina sheets
and edges of kaolin particles. Alumina and silica particles, how-
ever, comprise of homogeneous surface charges with respect to
pH and ionic strength where isomorphic substitutive defects are
absent in these pure particle samples. These differences in charges,
shapes (morphology) (Fig. 2) and sizes (Fig. 3) between the alumina,
silica and Riedel kaolin particles analyzed could affect the zeta
potential-pH and yield stress-pH behaviors of these slurries. Even
so, significant comparisons can be made qualitatively between the
behavioral trends displayed by these slurries.
(Zeta potential-pH and yield stress-pH plots of Unimin kaolin
slurries obtained from [9] was included in the zeta potential-pH and
yield stress-pH plots of Riedel kaolin slurries with Mg2+ and Ca2+
cations of this study to verify the results of [9]. Unimin kaolin parti-
cles comprise of similar particle morphology, microstructures and
sizes to Riedel kaolin particles but are understood to carry higher
concentrations of Ca2+ cations to Riedel kaolin particles [9]).
2.2. Methods of analysis
All slurries were prepared by adding appropriate amounts of
respective powders, 2.0–6.0 M HNO3 solutions and then 0.1 weight
percentages (wt%) MgCl2 or CaCl2 stock solutions (when required)
to deionized water and then sonicated using the Branson digital
Sonifier with a 2.5 cm probe for 20–40 s. MgCl2 and CaCl2 were
obtained in anhydrous forms from Optigen Scientific and Ajax
Finechem Pty Ltd respectively and were prepared into stock solu-
tions using distilled water. HNO3 addition ensured that the initial
pH of all slurries were ∼3.0. Subsequent 0.7–8.0 M KOH additions
allowed the pH increments of these slurries in a stepwise man-
ner from pH 3.0 to 12.0 and 13.0 for zeta potential and yield stress
measurements respectively while zeta potentials and yield stresses
of slurries were correspondingly measured. Beyond these pH val-
ues, zeta potentials and yield stresses of slurries recorded were
observed to result in inconsistent trends which altered erratically
with pH due to the effects of high slurry ionic strengths. While KOH
additions caused increments in slurry ionic strengths of less than
an order of magnitude and while an order of magnitude change in
ionic strength is commonly employed in the study of effect of ionic
strength on the zeta potentials and yield stresses of slurries, these
increments are considered insufficient to affect the zeta potentials
and yield stresses of slurries recorded.
Slurry solids concentrations were varied with respect to mea-
surements undertaken so as to obtain accurate zeta potential and
yield stress values that explain the particle properties and slurry
behaviors analyzed. Solids concentrations required for zeta poten-
tial measurements are low and thus were fixed at 5 wt% for all
slurries. Solids concentrations required for yield stress measure-
ments are high where the solids concentration of alumina, silica
and kaolin slurries without MgCl2 or CaCl2 were 55, 70 and 40 wt%
respectively and with MgCl2 or CaCl2 were 55, 55 and 40 wt%
respectively. These differences in the yield stress slurry solids con-
centrations ensure that substantial and measurable yield stress
values were obtained. While these variations in solids concentra-
tions will affect the magnitudes of yield stresses of slurries, they will
not affect the yield stress-pH trends of these slurries. In this study,
since the effect of Mg(II) and Ca(II) hydrolyzed products formed
with respect to pH are analyzed, the yield stress-pH trends and not
the yield stress magnitudes are of interest. Variations in yield stress
slurry solids concentrations are thus considered acceptable in this
study.
Zeta potential was measured on a ZetaProbe manufactured by
Colloidal Dynamics, USA. To each zeta potential-pH data obtained
of slurries with cations, the ZetaProbe was programmed to carry out
a background electrolyte correction so as to eliminate the effects
of cationic concentrations on the electrokinetic sonic amplitudes
(ESA) measured. (Corrected) zeta potentials were then plotted with
respect to pH.
The ZetaProbe functions on an ESA technique which is described
in O’Brien et al. [17]. This technique was developed to measure zeta
potentials of spherical particles with low surface conductance. This
includes the spherical silica particles but not the platelet alumina
and kaolin particles analyzed in this study. This is because while the
alumina particles comprise of low surface conductance, they are
of thin hexagonal platelet shapes. Loewenberg and O’Brien [18],
however, explained that a disc-shaped particle behaves approxi-
mately like a sphere of smaller radius where these alumina particles
can be approximated to be oblate spheroids with the same aspect
ratio as the hexagonal plates and can be analyzed for their dynamic
mobility and zeta potential using the standard theory for spherical
particles that is employed by the ZetaProbe [19].
Kaolin particles, on the contrary, exist as plates and carry
anomalously high surface conductance at very low electrolyte
concentrations to comprise of non-uniform surface charges (i.e.
positive edges and negative faces) especially at low pH. Loewen-
berg and O’Brien’s [18] particle property approximations that were
applicable to alumina particles are not applicable to kaolin par-
ticles as uncertainties exist concerning the dynamic mobility and
zeta potentials of kaolin particles calculated at low pH and low
ionic strength (i.e. at high surface conductance) using the standard
theory. As particle velocities and consequently ESA measured will
be affected by these varied and separate charges (i.e. zeta poten-
tials) on both the edges and the faces of kaolin particles, the zeta
potentials of kaolin particles calculated by the ZetaProbe could be a
representation of the average zeta potential value of the kaolin par-
ticles instead [20]. As a result, to maintain accuracy, only general
differences in the zeta potential-pH trends or magnitudes between
kaolin slurries with varying cationic concentrations or in compar-
ison to alumina or silica slurries and no individual magnitudes of
kaolin particles were mentioned in this study.
However, these average kaolin particle zeta potentials will not
indicate the local and individual surface charges of the alumina
faces, silica faces and edges on the kaolin particles to indicate the
local interactions between these interacting kaolin particle sites.
Zhao et al. [21] analyzed the surface charge potentials of mus-
covite clays to explain the different positive and negative charges
on the basal planes and edges of these particles that will conse-
quently affect the local particle interactions and thus rheology. As
a result, in this study, zeta potential-pH and yield stress-pH data of
kaolin slurries with cations were analyzed independent from one
another. As explained in the Introduction, since the yield stress-
zeta potential-pH behaviors of pristine kaolin slurries without any
cations, such as of Riedel kaolin slurries, have been well character-
ized with respect to the surface charges of the interacting kaolin
particle sites, the zeta potential-pH and yield stress-pH data of
these pristine Riedel slurries, on the other hand, were analyzed in
relation to one another.
Yield stress,  y (Pa) was calculated from the maximum torque, S
(%) values that were measured on the Brookfield vane viscometers.
S represents the amount of torque that results in the irreversible
deformation, i.e. the rupture and collapse of most or all slurry net-
work structural bonds that consequently causes the slurry to yield.
These S values were measured on the Brookfield LVDV-II, RVDV-
II and HBDV-II vane viscometers each of varied spring viscometer
constants, VC (or torque at 100% scale reading) of 0.0673, 0.7187 and
5.7496 mNm respectively. The medium and small four-blade vanes
of vane constants, K 3.757 × 10−6 and 4.538 × 10−7 m3 respectively
were used. The vane rotational speeds employed in these yield
stress measurements were kept low at 0.3–0.8 rpm. At these speeds,
the yield stress measured is negligibly affected by the vane rota-
tional speeds [22]. From S values obtained at each respective pH, the
yield stress,  y was calculated with respect to pH using the equation
 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113
Fig. 4. (a) Zeta potential-pH and (b) yield stress-pH behaviors of alumina, silica and
Riedel kaolin slurries.
 y = [(S ÷ 100) × Vc] ÷ K. Yield stresses obtained were then plotted
with respect to pH.
After each pH change, zeta potential was measured immedi-
ately and yield stress was measured ∼5 min after. As the dissolution
kinetics of alumina at room temperature and at high pH is slow, the
precipitation of tricalcium aluminate, Ca3 AlO6 , precipitates from
the presence of Ca(OH)2 precipitates at high pH is not anticipated
within the alumina slurries in this study.
3. Results and discussion
3.1. Zeta potential-pH and yield stress-pH behaviors of alumina
and silica slurries vs. kaolin slurries – without cations
It was observed that the zeta potentials of Riedel kaolin particles
were less positive than that of alumina particles and less negative
than that of silica particles at all pH analyzed (Fig. 4a). For exam-
ple, from pH ∼3.0 to 12.0, the zeta potentials of alumina, silica
and Riedel kaolin particles ranged from 38.2 to −31.6 mV, −21.3
to −91.3 mV and −15.6 to −48.8 mV respectively. Riedel kaolin’s
surface charges in relation to alumina’s and silica’s surface charges
are attributed to the simultaneous contributions of positive charges
from the alumina sheets and negative charges from the silica sheets
107
(where the edges would also contribute to kaolin’s surface charges)
and where these charges are averaged to provide the kaolin particle
zeta potentials.
Despite the general variations in zeta potential magnitudes
between these particles, the surface charges of Riedel kaolin par-
ticles were more similar to the surface charges of the silica than
the surface charges of the alumina particles. Silica, Riedel kaolin
and most other kaolin particles such as Sigma and Fluka are (net)
negatively charged even at a low pH of ∼3.0 [7,9] unlike alumina
particles. The low IEPs of silica and kaolin particles, which are
between pH 1.0–2.0 [23–25] and at pH 2.35 [26] respectively, are
indicative of the high negative surface charges carried by both these
particles at most pH. (In this study, IEPs were not displayed by
the silica and Riedel kaolin particles as zeta potential measure-
ments were carried out only from pH ∼3.0.) The IEP of the platelet
␣-alumina particles was instead observed at a higher pH of 7.4
in this study, which is close to the IEPs obtained for these par-
ticles by Husin et al. [15] (at pH 7.0–8.0) and Khoo et al. [16]
(at pH 8.0). The higher IEP is indicative of the lower negative
charge concentrations of the alumina particles compared to the
silica and kaolin particles at most pH. This similarity in surface
charges between silica and kaolin particles indicates the greater
control of the permanently negatively charged silica sheets than
the pH-dependent alumina sheets on the surface charges of kaolin
particles.
The yield stress-pH trends of alumina, silica and Riedel kaolin
slurries were observed to co-relate to the surface charges of the
respective particles (Fig. 4b). This resulted in Riedel kaolin slurries
exhibiting similar yield stress-pH trends to the silica slurries but
different yield stress-pH trends to the alumina slurries. The sim-
ilarity in yield stress-pH trends between silica and Riedel kaolin
slurries also includes most other kaolin slurries such as Sigma and
Fluka reported previously by [7,9]. Alumina, silica and Riedel kaolin
slurries each displayed the highest yield stress at the lowest par-
ticle zeta potential magnitude. As a result, silica and Riedel kaolin
slurries displayed high yield stresses at pH ∼3.0 at (net) negative
zeta potentials of −21.3 and −15.6 mV respectively while alumina
slurries displayed high yield stress at pH 6.2, which is close to the
pHIEP at 7.4 of net zero surface charge. (Conductivity differences
between slurries prepared for zeta potential (0.4 mS/cm) and yield
stress (1.4 mS/cm) measurements is inferred to cause the slight
mismatch between alumina slurry pHIEP and pH of maximum yield
stress.) From these respective pH values, the yield stresses of alu-
mina, silica and Riedel kaolin slurries decreased due to increasing
zeta potential magnitudes and repulsive forces between interacting
particles.
Despite the similarity between the yield stress-pH trends of
silica and kaolin slurries, the yield stresses of silica slurries are
affected by the properties and interactions between silica parti-
cles while the yield stresses of kaolin slurries are affected by the
properties and interactions between silica sheets, alumina sheets
and edges. While the high silica slurry yield stress is attributed to
van der Waals forces, the high kaolin slurry yield stress is con-
tributed by unlike charge attractions between edges and faces in
addition to van der Waals forces [4]. The absence of a yield stress
at high pH for both silica and Riedel kaolin slurries is attributed to
the high inter-repulsive forces as reflected by the high (net) neg-
ative particle zeta potentials (Fig. 4a). A previous study by Lagaly
[4] showed evidence of weak edge-edge interactions being present
at high pH within kaolin slurries. If this interaction is present to
induce weakly inter-connected network structures, kaolin slurries
will display low yield stresses at high pH. However, while the edges
and alumina sheets on kaolin particles are understood to affect the
kaolin particle surface charges and interactions, the silica sheets
are observed to control these kaolin particle properties to greater
extents.
108 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113

3.2. Zeta potential-pH and yield stress-pH behaviors of alumina

and silica slurries vs. kaolin slurries – with Mg2+ and Ca2+ cations

Abrupt and significant zeta potential-pH and yield stress-pH

trend variations were observed within alumina, silica and Riedel
kaolin slurries from pH > ∼6 to 12/13 with MgCl2 and CaCl2 added
(Figs. 5–8). (Refer to Supplementary data for the insignificant trend
variations of these slurries at pH < 6.) While zeta potential and
yield stress trend variations of these slurries were insignificant at
pH < 6 and conductivity variations of these slurries with the respec-
tive cation concentrations added were negligible between pH < 6
to pH > 6, the effect of conductivity increments from the addition
of these cations is inferred to make insignificant contributions on
these slurry trend variations observed from pH > 6. These trend
variations are instead attributed to the formation and adsorption of
Mg(II) and Ca(II) hydrolyzed products, such as MgOH+ and CaOH+ ,
onto alumina, silica and Riedel kaolin particle surfaces from pH > 6.
The adsorptions of such hydrolyzed products onto particle surfaces,
even of low dielectric constants, are thermodynamically favored as
these products carry low ionic charges that concomitantly reduces
the ion-solvent interaction in the adsorption process unlike the
high interaction energies involved with hydrated products [27].
These Mg(II) and Ca(II) hydrolyzed product formations and
adsorptions, however, occurred at an earlier pH of ∼6 compared
to a pH of ∼10, which was specified in literature [9]. The ear-
lier pH of cation hydrolysis and adsorption onto particles could
occur via surface induced hydrolysis of un-hydrolyzed metal ions
[28,29]. Surface induced hydrolysis involves the release of one pro-
ton from the primary hydration sheath of the adsorbing cation
that significantly reduces the energy barrier to facilitate cationic
adsorption onto particle surfaces [29]. In this case, from pH > 6,
magnesium and calcium would exist as Mg(OH)+ and Ca(OH)+ com-
plexes respectively to interact with the oxides of alumina, silica and
kaolin particles in the following manner: (oxide particle surface)
O− · · ·Mg(OH)+ /Ca(OH)+ [28,29].
As pH increased to ∼10, the hydrolysis of Mg2+ and Ca2+ cations
begin [9]. At low cationic concentrations, as adopted in this study
(<0.2 M), Mg(OH)+ /Ca(OH)+ complexes and Mg(OH)2 /Ca(OH)2
(solid) precipitates form from pH 10 [30]. While Mg(OH)+ and
Ca(OH)+ complexes adsorb to reduce particle negativity, Mg(OH)2
and Ca(OH)2 precipitates adsorb and nucleate to induce positive
potentials to the surface [11], which can then lead to CR points and
‘curved’ or abrupt changes in zeta potential-pH trends.
While Mg(OH)+ adsorptions and Mg(OH)2 precipitations dis-
played such ‘curved’ zeta potential-pH trends with CR within
alumina, silica and Riedel kaolin slurries (Fig. 5), Ca(OH)+ adsorp-
tions and Ca(OH)2 precipitations resulted in moderately straight
zeta potential-pH trends (Fig. 6). The turbidity-pH results of Mg2+
and Ca2+ solutions from Avadiar et al. [9] explain this variation
in trends. With similar Mg2+ and Ca2+ concentrations added, the
turbidity-pH results display increasing Mg2+ and Ca2+ solution
turbidity from pH 10 that is higher for the Mg2+ than the Ca2+ solu-
tions. This indicates that Mg(OH)2 and Ca(OH)2 precipitates form
from pH 10 but with Mg(OH)2 forming in higher concentrations
to Ca(OH)2 despite the similar Mg2+ and Ca2+ cationic concentra-
tions added. Baes and Mesmer [30] supported this to explain that
Mg2+ hydrolysis is obvious at pH ∼9–10 just before the precipi-
tation of Mg(OH)2 while Ca2+ cations carry lower tendencies to
hydrolyze. With increasing pH to 12, Ca(OH)2 continues to form
in increasing concentrations as inferred from the increasing tur-
bidity while Mg(OH)2 concentrations reduce slightly as inferred
Fig. 5. Zeta potential-pH behaviors of (a) alumina, (b) silica and (c) Riedel kaolin
from the reducing turbidity [9]. This indicates that Mg(OH)2 could slurries with varying MgCl2 concentrations. Plot of Unimin kaolin slurries [9]
be nearly completely formed by pH 12 while Ca(OH)2 completely included in (c) for comparison.
forms at a higher pH. Clark and Cooke [12] stated that Mg(OH)2
and Ca(OH)2 completely form at pH 10 and 13 respectively. Parks
[31] reported the IEP of (solid) Mg(OH)2 to be at pH ∼12 while
 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113 109

Fig. 6. Zeta potential-pH behaviors of (a) alumina, (b) silica and (c) Riedel kaolin Fig. 7. Yield stress-pH behaviors of (a) alumina, (b) silica and (c) Riedel kaolin slur-
slurries with varying CaCl2 concentrations. Plot of Unimin kaolin slurries [9] ries with varying MgCl2 concentrations. Plot of Unimin kaolin slurries [9] included
included in (c) for comparison. in (c) for comparison.
110 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113
Fig. 8. Yield stress-pH behaviors of (a) alumina, (b) silica and (c) Riedel kaolin slur-
ries with varying CaCl2 concentrations. Plot of Unimin kaolin slurries [9] included
in (c) for comparison.
Atesok et al. [11] and Zhang [32] reported the IEP of (solid) Ca(OH)2
to be higher at pH > 12 and 12.9 respectively. The higher Ca(OH)2
IEP emphasizes the later pH of Ca(OH)2 to Mg(OH)2 completion of
formation. As a result, Ca(OH)2 concentrations formed are inferred
to be insufficient to result in positive potentials and CR within these
zeta potential slurries observed in this study at pH ≤ 12. Indepen-
dent to these zeta potential-pH trends, the yield stress-pH trends
of the corresponding slurries also indicated the later pH of Ca(OH)2
to Mg(OH)2 formation from the shifts in the maxima yield stress
points of slurries with CaCl2 compared to MgCl2 to a higher pH
(Figs. 7 and 8).
(Refer to Supplementary data for reasons why zeta potentials
might not relate precisely to yield stresses of these slurries.)
Comparing alumina, silica and Riedel kaolin slurries, silica and
Riedel kaolin slurries displayed CR with sufficiently high MgCl2
concentrations of 1.25 and 2.5 dry weight percentages (dwb%)
respectively added while alumina slurries displayed CR with lower
MgCl2 concentrations of just 0.05 dwb% added (Fig. 5). This is
because the more negatively charged silica and kaolin particles
require higher concentrations of positive charges from Mg(OH)+
and positive potentials from Mg(OH)2 to neutralize and reverse
their surface charges compared to the less negatively charged alu-
mina particles. The BET surface area of platelet alumina particles is
also lower than that of silica and Riedel kaolin particles, inhibiting
the exposure of its (low concentrations of) negatively charged sites.
This led to alumina particles carrying higher concentra-
tions of positive charges than silica and Riedel kaolin particles
from pH > 6 with similar cationic concentrations of 0.5 dwb%
MgCl2 /CaCl2 added (Figs. 5 and 6). The increasing zeta potential
gaps between alumina slurries with cations and pristine slur-
ries without cations explained the greater Mg(OH)+ and Ca(OH)+
concentrations adsorbed on alumina particles with increasing pH
from >6. Higher Mg(OH)+ and Ca(OH)+ surface coverage are also
expected at higher MgCl2 and CaCl2 concentrations added to
alumina slurries. These high Mg(OH)+ /Ca(OH)+ -induced positive
charge concentrations is inferred to induce electrosteric repul-
sive forces where these Mg(OH)+ /Ca(OH)+ -adsorbed layers would
act as electrosteric layers that cause increments in inter-particle
separation distances that are larger than the separation distances
between alumina particles at the IEP of the pristine slurries. This
consequently led to the yield stress reductions of alumina slurries
from pH 6–10 (Figs. 7a and 8a). These yield stress reductions were
consequently higher with higher MgCl2 and CaCl2 concentrations
added into alumina slurries where even with high concentra-
tions of 5 dwb% CaCl2 added, negligible attractive interactions were
induced between alumina particles by the Ca(OH)+ layers adsorbed.
Such yield stress reductions were, however, not observed within
alumina slurries with 0.05 dwb% CaCl2 (Fig. 8a). From pH ∼6–10,
this slurry displayed higher yield stresses than the yield stresses
of pristine alumina slurries. This explains that the low Ca(OH)+
concentrations adsorbed on alumina particles at 0.05 dwb% CaCl2
addition were insufficient to induce strong electrosteric repulsive
forces but which induced an attractive force, most likely charged
patch attractions. This is because greater concentrations of nega-
tively charged alumina particle surfaces (i.e. at high pH: after its
IEP at pH 6.2 – Fig. 4a) would remain bare with the low cationic
concentrations adsorbed to interact attractively with the positively
charged Ca(OH)+ adsorbed-alumina particle patches [13]. From the
approximate similarity in yield stress magnitudes, the strength
of this charged patch interaction at pH > 6 is inferred to be simi-
lar to the strength induced by van der Waals forces of attraction
at the yield stress peak (or IEP) of the pristine alumina slurries.
Alumina slurries with 0.05 dwb% MgCl2 (Fig. 7a), however, did
not display such yield stress increments but displayed approxi-
mately similar yield stress values as the pristine alumina slurries
from pH 6–10. If the steric effect due to MgOH+ adsorption is
 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113
significant, a considerable yield stress reduction should be
observed. The approximately similar yield stresses indicate that
charged patch attraction could also be present at similar strengths
to the weak electrosteric repulsive forces within these slurries with
0.05 dwb% MgCl2 . The strength of this charged patch attraction is,
however, not as strong as that induced by CaOH+ adsorption.
From pH > 10, alumina slurries with MgCl2 and CaCl2 displayed
yield stress increments (Figs. 7a and 8a). At the vicinity of high
pH of 10, a water molecule covalently links to the metal ion core
of the Mg(OH)+ /Ca(OH)+ complexes that causes these complexes
to lose a proton and form Mg(OH)2 /Ca(OH)2 precipitates on these
particles. The Mg(OH)2 and Ca(OH)2 precipitated layer on alumina
particles is inferred to induce an attractive force that is stronger
than the steric forces that is usually induced by such precipitated
layers on particles. While such precipitated layers increase the
interacting distances between particles, van der Waals of attrac-
tion will grow to be weak. While hydrolysis products formed are
usually amorphous to comprise of lower Hamaker constants than
their crystalline oxide counterparts [13], the hamaker constants of
Mg(OH)2 and Ca(OH)2 are expected to be small and thus could not
contribute significantly to the effective hamaker constant of alu-
mina particles to explain the yield stress increments via van der
Waals forces. This attractive force should be a precipitate-induced
attractive interaction. The strength, i.e. the yield stress induced
by this attractive interaction was higher within alumina slurries
with Ca(OH)2 precipitates at pH > 12 (i.e. when these precipitates
form in significant concentrations) than within alumina slurries
with Mg(OH)2 precipitates from pH 10 (i.e. when these precipitates
form) (Figs. 7a and 8a). The strength of this attractive interaction
was also higher with Ca(OH)2 and lower with Mg(OH)2 than the
strength of van der Waals forces of attraction at the IEP of pristine
alumina slurries.
The alumina slurry with a lower CaCl2 dosage of 0.2 dwb% dis-
played higher yield stresses than the alumina slurry with a higher
CaCl2 dosage of 0.5 dwb% from pH 10 (Fig. 8a). This indicates that
concentrations of Ca(OH)2 vs. Ca(OH)+ are lower on alumina par-
ticle surfaces from pH 10. The lower Ca(OH)2 concentration is due
to the slow hydrolysis transformation rate of Ca(OH)+ and the
high pH of Ca(OH)2 complete formation [9,12]. While adsorbed
Ca(OH)+ complexes continue to act as electrosteric layers that
reduce yield stresses, Ca(OH)2 precipitations induce an attractive
force that increases slurry yield stresses from pH 10. The signifi-
cantly lower yield stress increment exhibited by the alumina slurry
with 0.05 dwb% CaCl2 from pH 10, on the other hand, could be due to
the arrangement of Ca(OH)2 precipitates on alumina particles. The
transformation of Ca(OH)+ positive patches (that were involved in
charged patch interactions) to Ca(OH)2 precipitates could result in
these precipitates adsorbed to be concentrated on a few particles
and poorly distributed across the slurry microstructures. This could
inhibit the formation of a continuous and inter-connected slurry
network structure that is of high strength. The low Ca(OH)2 con-
centrations formed with only 0.05 dwb% CaCl2 could also reduce
the degrees of precipitate-induced attractive interactions within
these slurries.
On the other hand, from pH 6–10, Mg(OH)+ and Ca(OH)+
adsorptions on silica and Riedel kaolin particles reduced
the negative charge concentrations on these particles
(Figs. 5b and c, and 6b and c). Similar to on alumina particles,
greater Mg(OH)+ and Ca(OH)+ concentrations adsorbed on silica
and Riedel kaolin particles with increasing pH and with increasing
MgCl2 /CaCl2 dosages added. This caused the greater reductions in
silica and Riedel kaolin particles’ surface charges. These reductions
consequently reduced the inter-particle-repulsive forces within
silica slurries. While not relating the kaolin particle zeta potential
magnitude reductions to the particle’s reducing repulsive forces,
the very similar zeta potential-pH trends between silica and Riedel
111
kaolin can be used to explain that inter-particle-repulsive forces
reduced within Riedel kaolin slurries with MgCl2 /CaCl2 added.
Particles could thus approach one another and exist at inter-
particle separation distances closer than the distances between
these particles within the pristine slurries to experience attractive
inter-particle forces. The silica and Riedel kaolin slurry yield
stress increments from pH 6 indicate that inter-repulsive forces
reduced that led to attractive interactions within these slurries
(Figs. 7b and c, and 8b and c). This attractive force should be an
ion-facilitated attractive interaction. (Refer to Supplementary data
for reasons why silica vs. kaolin slurry yield stresses is higher.)
From pH ∼10, silica and Riedel kaolin slurries with MgCl2 dis-
played yield stress reductions (Fig. 7b and c) while these slurries
with CaCl2 displayed yield stress increments (Fig. 8b and c). This
caused silica slurries with 0.2, 0.5 and 1.25 dwb% MgCl2 to display
yield stress peaks at pH ∼9.8, 9.2 and 8.5 respectively while sil-
ica slurries with 0.2 and 0.5 dwb% CaCl2 to display peaks at pH
∼12 and 11.35 respectively and silica slurries with 1.25 dwb% CaCl2
to display no peaks with yield stresses increasing to pH 13. This
also caused kaolin slurries with 0.5 and 2.5 dwb% MgCl2 to display
peaks at pH ∼9.45 and 9.35 respectively while kaolin slurries with
0.5 dwb% CaCl2 to display a peak at pH ∼12.22 and kaolin slur-
ries with 2.5 dwb% CaCl2 to display no peaks with yield stresses
increasing to pH 13.
This explains that the strength of Mg(OH)+ -(ion)-facilitated
attractive interactions between pH ∼6–10 is greater than the
strength of Mg(OH)2 -precipitate-induced attractive interactions
from pH > ∼10 within these silica and Riedel kaolin slurries ana-
lyzed. Steric forces induced by Mg(OH)2 layers on silica and Riedel
kaolin particles could be higher to override possible precipitate-
induced attractive forces and to contribute to the slurry yield stress
reductions from pH > ∼10. The strength, i.e. the yield stress induced
by these ion-facilitated interactions is higher than the strength of
van der Waals forces of attraction and hydrogen bonding within
pristine silica slurries at pH 3 and similar or slightly higher, depend-
ing on the cation dosage, to the strength of electrostatic edge-face
interactions within pristine kaolin slurries at pH 3.
Kaolin slurries with 2.5 dwb% MgCl2 concentrations, however,
displayed approximately constant yield stresses from pH > ∼10
(Fig. 7c). This indicates that Mg(OH)2 concentrations adsorbed
could be sufficiently high at this higher MgCl2 dosage to overcome
the steric forces and to induce precipitate-induced attractive inter-
actions at similar strengths to the steric forces. The yield stress
induced by this precipitate-induced attractive interaction is seen
to be only slightly lower to that induced by the ion-facilitated
attractive interaction. This explains that with higher MgCl2 con-
centrations added and thus Mg(OH)2 precipitates formed, the yield
stresses of silica and kaolin slurries from pH ∼10 would be higher.
The silica and Riedel kaolin slurry yield stress increments with
CaCl2 from pH > 10 is attributed to Ca(OH)+ and gradually increasing
Ca(OH)2 concentrations that induce ion-facilitated and precipitate-
induced attractive interactions respectively (Fig. 8b and c). The
strengths of these interactions are significantly higher within silica
slurries and similar or slightly higher within kaolin slurries than the
strengths of the pristine silica and kaolin slurries at pH 3 respec-
tively. The subsequent yield stress reductions from pH > ∼12 within
silica slurries with 0.2 and 0.5 dwb% CaCl2 and kaolin slurries with
0.5 dwb% CaCl2 is attributed to the low Ca(OH)2 concentrations pro-
duced, which acted as steric Ca(OH)2 layers that overcame possible
precipitate-induced attractive interactions. With sufficiently high
CaCl2 concentrations of 1.25 and 2.5 dwb% added into silica and
kaolin slurries respectively, sufficient Ca(OH)2 concentrations were
inferred to be produced to overcome these steric forces and induce
a precipitate-induced attractive interaction and slurry yield stress
increments from pH > ∼12. This cation dosage effect is similar to
with MgCl2 added where the effects (i.e. yield stress peaks and
112 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113
Fig. 9. A schematic representation of other possible kaolin crystal layer structures.
Adapted from Ma and Eggleton [34].
subsequent yield stress reductions) were observed at an earlier
pH with MgCl2 (Fig. 7b and c) than with CaCl2 due to the higher
tendency of Mg2+ hydrolysis.
Ca(OH)+ adsorptions and Ca(OH)2 precipitations induced higher
slurry network strengths than Mg(OH)+ adsorptions and Mg(OH)2
precipitations on alumina, silica and Riedel kaolin particles
(Figs. 7 and 8). This is despite the similar MgCl2 and CaCl2 con-
centrations added into the slurries. This could be attributed to
the larger Ca(OH)+ than Mg(OH)+ complex ion size and the larger
Ca(OH)2 than Mg(OH)2 precipitate size. This results in Ca(OH)+ ,
which comprises of lower hydration enthalpies than Mg(OH)+ ,
to adsorb easily and at higher degrees onto particle surfaces (i.e.
stern layers) and Ca(OH)2 , which comprises of lower precipitation
enthalpies than Mg(OH)2 , to form and adsorb in higher degrees
onto particle surfaces. The spherical cap areas at the closest point
of interaction for Ca(OH)+ and Ca(OH)2 are also understood to be
larger than Mg(OH)+ and Mg(OH)2 respectively. This reduces the
inter-particle-separation distances with the former. These vari-
ations between Ca2+ to Mg2+ hydrolyzed products consequently
increase the probabilities of attractive interactions with the former.
The ion-facilitated attractive interaction within silica and Riedel
kaolin slurries could be ion-facilitated van der Waals forces of
attraction. These attractive forces are propagated primarily by
the reduction in inter-particle separation distances, which occur
upon the adsorption of Mg(OH)+ /Ca(OH)+ complexes on silica and
Riedel kaolin particles and upon the consequent reduction in inter-
particle-repulsive forces. These attractions link Mg(OH)+ /Ca(OH)+
complexes that are adsorbed onto particles to the oxides of other
particles. The precipitate-induced attractive interaction could be
precipitate bridging interaction. Mg(OH)+ /Ca(OH)+ complexes that
are adsorbed onto particles transform to form Mg(OH)2 /Ca(OH)2
precipitates at the pH of hydrolyzed product phase change (i.e. at
pH ∼10). These adsorbed precipitates then adsorb onto another
available interacting particle site to propagate greater particle
interactions via precipitate bridging interactions. A schematic rep-
resentation of the precipitate bridging interaction can be found in
[33]. (Refer to Supplementary data for more information on these
ion-facilitated and precipitate-induced attractive interactions.)
Zeta potential-pH and yield stress-pH behaviors of Riedel kaolin
slurries with and without cations are closer to the behaviors of sil-
ica than alumina slurries. This indicates that the silica sheets are
affecting the behaviors of Riedel kaolin in greater extents than the
alumina sheets. Fig. 10 in Gupta and Miller [2] shows the estimated
surface potential-pH behaviors of the alumina face and the silica
face in kaolin particles obtained via surface force measurements.
These behaviors were observed to be similar to the respective zeta
potential-pH behaviors of the alumina and silica slurries (Fig. 4)
analyzed in this study. This similarity was especially apparent at
pH < 8 when the IEP of the alumina face (at pH ∼6.3 [2]) was at
approximately similar pH range to that of the alumina slurries (at
pH 7.4 (Fig. 4)) and when the silica face did not display an IEP (i.e. at
the lowest pH of 3, the surface potential was ∼−30 mV) [2] similar
to the silica slurries (Fig. 4). According to these surface potential-
pH measurements (i.e. bearing in mind that properties between the
Riedel kaolin particles analyzed in this study and the kaolin parti-
cles analyzed in [2] could be different and thus only focusing on the
general similarities in the surface potential behaviors), it could be
unlikely that the silica face in Riedel kaolin would carry sufficient
negative charge densities to override the positive charge densities
on the alumina face and the edge of the kaolin particle and to con-
sequently result in the net negative surface charge of the kaolin
particle at low pH.
The net negativity of Riedel kaolin particles might then indi-
cate that there could exists low concentrations of pyrophyllite-like
or smectite-like layers as surface layers of the Riedel kaolin crys-
tal structures [34] as shown in Fig. 9 different from Fig. 1. Ma and
Eggleton [34] examined natural kaolin samples via high-resolution
transmission electron microscopy (HRTEM) to indicate that besides
the expected 1:1 Si:Al kaolin crystal structure, kaolin particles could
comprise of a mixture of 1:1 Si:Al layers and 2:1 Si:Al surface (or
cover) layers. They indicated that these surface layers could occur
on one or both sides of the kaolin crystal structure to result in
tetrahedral silica sheets terminating at both sides of the kaolin
crystal. They explained that one or several 2:1 pyrophyllite-like sur-
face layers could occur on one side of the kaolin crystal structure
of well-ordered kaolinites while one or several 2:1 smectite-like
 L. Avadiar et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 103–113
layers could occur on one or both sides of the kaolin crystal struc-
ture of poorly-ordered kaolinites.
4. Conclusions
The zeta potential-pH and yield stress-pH behaviors of alumina,
silica and Riedel kaolin slurries with and without Mg2+ and Ca2+
cations show that the silica sheets contribute in greater extents
than the alumina sheets to control the surface charges and rheo-
logical behaviors of Riedel kaolin slurries with and without cations.
In addition, the similarity between the zeta potential-pH and yield
stress-pH trends of Unimin kaolin slurries (base titration) to the
trends of Riedel kaolin slurries with CaCl2 (base titration), which
displayed generally straight trend-lines (Figs. 6c and 8c), and not
MgCl2 (base titration), which displayed ‘curved’ trend-lines at high
pH (Figs. 5c and 7c), emphasizes the results of Avadiar et al. [9],
which explained that Unimin’s varied yield stress-pH behavior is
attributed to the sufficiently high Ca(II) concentrations on Unimin
kaolin particles.
Acknowledgments
We wish to acknowledge The Australian Research Council (ARC)
for partially funding this project via DP1096528. We also wish to
thank the reviewers for making this a better paper.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.colsurfa.
2014.12.019.
References
[1] A.M. Palomino, J.C. Santamarina, Fabric map for kaolinite: effects of pH and
ionic concentration on behavior, Clays Clay Miner. 53 (2005) 209–222.
[2] V. Gupta, J.D. Miller, Surface force measurements at the basal planes of ordered
kaolinite particles, J. Colloid Interface Sci. 344 (2010) 362–371.
[3] E. Tombacz, M. Szekeres, Surface charge heterogeneity of kaolinite in aque-
ous suspension in comparison with montmorillonite, Appl. Clay Sci. 34 (2006)
105–124.
[4] G. Lagaly, Principles of flow of kaolin and bentonite dispersions, Appl. Clay Sci.
4 (1989) 105–123.
[5] A.W. Flegmann, J.W. Goodwin, R.H. Ottewill, Rheological studies on kaolinite
suspensions, Proc. Br. Ceram. Soc. 13 (1969) 31–45.
[6] B. Rand, I.E. Melton, Particle interactions in aqueous kaolinite suspensions.
I. Effect of pH and electrolyte upon the mode of particle interaction in
homoionic sodium kaolinite suspensions, J. Colloid Interface Sci. 60 (1977)
308–320.
[7] E.J. Teh, Y.K. Leong, Y. Liu, A.B. Fourie, M. Fahey, Differences in the rheology and
surface chemistry of kaolin clay slurries: the source of the variations, Chem.
Eng. Sci. 64 (2009) 3817–3825.
[8] P. Mpofu, J. Addai-Mensah, J. Ralston, Influence of hydrolyzable metal
ions on the interfacial chemistry, particle interactions, and dewatering
behavior of kaolinite dispersions, J. Colloid Interface Sci. 261 (2003)
349–359.
113
[9] L. Avadiar, Y.K. Leong, A. Fourie, T. Nugraha, P.L. Clode, Source of Unimin kaolin
rheological variation – Ca2+ concentration, Colloids Surf. A 459 (2014) 90–99.
[10] R.O. James, T.W. Healy, Adsorption of hydrolyzable metal ions at the
oxide–water interface. II. Charge reversal of SiO2 and TiO2 colloids by adsorbed
Co(II), La(III), and Th(IV) as model systems, J. Colloid Interface Sci. 40 (1972)
53–64.
[11] G. Atesok, P. Somasundaran, L.J. Morgan, Adsorption properties of Ca2+ on Na-
kaolinite and its effect on flocculation using polyacrylamides, Colloids Surf. 32
(1988) 127–138.
[12] S.W. Clark, S.R.B. Cooke, Adsorption of calcium, magnesium, and sodium ion by
quartz, Trans. AIME 241 (1968) 334–341.
[13] Y.K. Leong, Yield stress and zeta potential of nanoparticulate silica dispersions
under the influence of adsorbed hydrolysis products of metal ions-Cu(II), Al(III)
and Th(IV), J. Colloid Interface Sci. 292 (2005) 557–566.
[14] P.-I. Au, S.-Y. Siow, L. Avadiar, E.-M. Lee, Y.-K. Leong, Muscovite mica and koalin
slurries: yield stress–volume fraction and deflocculation point zeta potential
comparison, Powder Technol. 262 (2014) 124–130.
[15] H. Husin, Y.K. Leong, J. Liu, H.J. Choi, W.L. Zhang, Surface force arising from
adsorbed graphene oxide in alumina suspensions with different shape and size,
AIChE J. 59 (2013) 3633–3641.
[16] K.S. Khoo, E.J. Teh, Y.K. Leong, B.C. Ong, Hydrogen bonding and interparti-
cle forces in platelet alpha-Al2 O3 dispersions: yield stress and zeta potential,
Langmuir 25 (2009) 3418.
[17] R.W. O’Brien, D.W. Cannon, W.N. Rowlands, Electroacoustic determination of
particle size and zeta potential, J. Colloid Interface Sci. 173 (1995) 406–418.
[18] M. Loewenberg, R.W. O’Brien, The dynamic mobility of nonspherical particles,
J. Colloid Interface Sci. 150 (1992) 158–168.
[19] W.N. Rowlands, R.W. O’Brien, The dynamic mobility and dielectric response of
kaolinite particles, J. Colloid Interface Sci. 175 (1995) 190–200.
[20] P.B. Laxton, J.C. Berg, Relating clay yield stress to colloidal parameters, J. Colloid
Interface Sci. 296 (2006) 749–755.
[21] H. Zhao, S. Bhattacharjee, R. Chow, D. Wallace, J.H. Masliyah, Z. Xu, Probing
surface charge potentials of clay basal planes and edges by direct force mea-
surements, Langmuir 24 (2008) 12899–12910.
[22] N.Q. Dzuy, D.V. Boger, Yield stress measurement for concentrated suspensions,
J. Rheol. 27 (1983) 321–349.
[23] P. Wilhelm, D. Stephan, On-line tracking of the coating of nanoscaled silica with
titania nanoparticles via zeta-potential measurements, J. Colloid Interface Sci.
293 (2006) 88–92.
[24] D.C. Agrawal, R. Raj, C. Cohen, In-situ measurement of silica-gel coating on
particles of alumina, J. Am. Ceram. Soc. 73 (1990) 2163–2164.
[25] R.K. Iler, The Chemistry of Silica, A Wiley-Interscience Publication, New York,
1979.
[26] M. Alkan, O. Demirbas, M. Dogan, Electrokinetic properties of kaolinite in
mono- and multivalent electrolyte solutions, Microporous Mesoporous Mater.
83 (2005) 51–59.
[27] R.O. James, T.W. Healy, Adsorption of hydrolyzable metal ions at the
oxide–water interface. III. A thermodynamic model of adsorption, J. Colloid
Interface Sci. 40 (1972) 65–81.
[28] J.A. Davis, J.O. Leckie, Surface ionization and complexation at the oxide/water
interface II. Surface properties of amorphous iron oxyhydroxide and adsorption
of metal ions, J. Colloid Interface Sci. 67 (1978) 90–107.
[29] H.M. Jang, D.W. Fuerstenau, The specific adsorption of alkaline-earth cations at
the rutile/water interface, Colloids Surf. 21 (1986) 235–257.
[30] C.F. Baes, R.E. Mesmer, The Hydrolysis of Cations, A Wiley-Interscience Publi-
cation, New York, London, Sydney, Toronto, 1976.
[31] G.A. Parks, The isoelectric points of solid oxides, solid hydroxides, and aqueous
hydroxo complex systems, Chem. Rev. 65 (1965) 177–198.
[32] S. Zhang, A new nano-sized calcium hydroxide photocatalytic material for the
photodegradation of organic dyes, RSC Adv. 4 (2014) 15835–15840.
[33] Z. Shumin, Y.K. Leong, L. Avadiar, Yield stress- and zeta potential-pH behavior
of washed alumina suspension in the presence of CaCl2 /MgCl2 – hydrolysis
product formation pH and precipitate bridging, Int. J. Miner. Process. (2014)
(submitted for publication).
[34] C. Ma, R.A. Eggleton, Surface layer types of kaolinite: a high-resolution trans-
mission electron microscope study, Clays Clay Miner. 47 (1999) 181–191.