Salt strengthened the bond formed and quickened the bonding process of 3–5 wt. % hectorite gels during the structural rejuvenation process. This even occurred at 0.002M KCl. Microstructure showed exfoliated, flexible platelet bonding in (+)edge–(−)face configurations. The display of prominent aging time-dependent behavior is due to the structural rejuvenation process being controlled by the electric double layer (EDL) repulsive force. Salt increased the lower energy paths to bonding in the (+)edge–(−)face configurations and weakened the EDL force to form stronger bonds. The Leong model time constant data supported the faster bonding process. In shear, the gels with a weakened EDL repulsive force caused by 0.01 and 0.1M KCl treatment were unable to display EDL force-control time-dependent behavior in the stepdown shear stress response. This situation was remedied by increasing the negative charge density of platelets with adsorbed P2O74−. The amount of P2O74− needed was higher at 0.1M KCl.

Our understanding of the effects of surface chemistry, such as salt and pH, on charge and shape anisotropic 2:1 (smectite) clay gel behavior remained incomplete despite decades of studies.1–9 Low salt addition effects were rarely studied on yield stress gels. The analysis of these surface chemistry effects rarely focused on the charge and shape anisotropic nature of the clays and the colloidal forces influencing the different shapes of the platelet–platelet bonding configurations. Strong ionic strength effects on yield stress of natural hectorite gels at relatively low salt concentration were reported recently, but the cause and mechanism involved remained unclear.10 These gels at ∼4 wt. % solids displayed several-fold increase in the aging yield stress upon the addition of only 0.01M KCl. The time-dependent aging behavior was not extinguished by the salt addition as high as 0.1M KCl. This study aims to uncover the cause and mechanisms involved. An additional stepdown shear rate test will be deployed to evaluate the effect of colloidal forces, particularly the electric double layer (EDL) repulsive forces, on the stepdown shear stress response of salt treated gels. A pronounced time-dependent behavior in the stepdown shear stress response is a strong indication of a strong EDL repulsive force in operation.11 This behavior is a consequence of the slowing down of the platelet–platelet bonding process. The effects of even much lower salt concentration on aging yield stress and stepdown shear rate behavior were also evaluated. Other factors such as pH and nature of salt alkali metal ions and anions were also covered.

The EDL repulsive force was confirmed to be responsible for the time-dependent behavior of clay gels in the structural rejuvenation mode just recently.11 In 2018, based on the information of a very open gel microstructure with the faces of the interacting platelets in the network being far apart and the display time-dependent behavior in the yield stress during aging, we postulated that the strong face–face EDL repulsive force was responsible.9,12,13 Later, we found that these clay gels, Laponite (a synthetic hectorite), hectorite, and NaMnt, also display time-dependent behavior in stepdown shear stress response.10 The response of the stepdown shear stress showed an immediate increase with time, sharply initially and then more gradually with time. This behavior is due to the EDL repulsive force in the face–face configuration being strong to disrupt the flow-aligned platelets and getting them to bond and form a stronger structure even in the presence of an imposed shear. At times, it is necessary to increase the negative charge density of the platelets, such as with an adsorbed anionic additive, to produce the same time-dependent response. More types of clay gels, such as kaolin suspension, were found to display structural rejuvenation-type time-dependent behavior by this anionic adsorption method of strengthening the EDL repulsive force.11 For kaolin, both the silica and alumina faces must be negatively charged for the display of the EDL repulsive force-controlled time-dependent behavior.

A strong EDL repulsive force in the face–face configuration will slow the bonding process by heightening the energy barrier to bonding in many approach configurations. As this EDL repulsive force gets stronger, even more approach configurations will acquire this insurmountable energy barrier. The charge and shape anisotropic nature of the hectorite particles will permit bonding to occur even when the EDL face–face repulsion is strong. Low energy paths to bonding are available.10 For example, the (+)edge–(−) face approach configuration at ∼90° angle is one such path.14–20 When the EDL face–face repulsive force is very strong, only those edge–face approach angles close to 90° will lead to bonding. This angle will broaden as this EDL repulsive force gets weaker. The (+)edge–(−)face bond forming the clay gel network structure is a well-accepted fact.14–17 Computer simulations relied on this (+)edge–(−)face attractive interaction to form the clay gel network structure.14,15 Artifact-free clay gel microstructures captured by cyro-SEM showed such bonding forming the network structure.12,13

At the silica face, the diffuse layer that counters the negative charge is net positive. At the edge, the diffuse layer will be net negative.10 The path of the 90° (+)edge–(−)face approach should produce the lowest barrier energy to bonding. The initial overlapping of these two oppositely charge diffuse layers should lead to attraction, bringing the face and edge in contact. The acute edge–face approach angle can also lead to bonding when the EDL repulsive force is weakened sufficiently. The bonding process will quicken further when more low energy paths to bonding are created as the EDL repulsive force weakens further. Overlapping face–face attraction at the edge can also occur and was captured theoretically for Laponite disk interactions.15,16 This attractive interaction formed sheets, which were also captured theoretically17 and experimentally.12 A strong charge anisotropy condition, i.e., high positive edge and high negative face charge density, is required for this to occur. Laponite has long been regarded as a model system for intensive investigations because of the belief that its nanodiscotic particles are monodisperse with interesting anisotropic charge properties. Most of these studies focused on answering fundamental questions on the phase state21–27 and the properties and behavior of the dispersion.15,28–31 A number of model representations of the surface charge properties and potential15,16,30,32 were developed, such as the point quadripole28 and potential of mean force.30 Some were used to model the interaction between disks in various orientation configurations.15,16,28,30

When the EDL repulsive force is weak, the stepdown shear stress will display a mild time-dependent behavior. At times, the time-dependent behavior was barely discernible. The bonding process becomes very fast as most approach configurations lead to bonding. The stepdown shear stress remained high unlike that of the gel in the flocculated phase state. In this state, the van der Waals attractive force dominates the platelet–platelet interactions and forms stable compact aggregates quickly. The stepdown shear stress is very low in value and displays a time-independent response.9,10 The imposed shear rate of 10 s−1 is unable to break the bond of the platelet aggregates depleting the concentration of particle or aggregate interactions. The breakdown of these aggregates at a high shear rate of 1000 s−1 caused the shear stress to decrease by an unusually large amount with time.9,10 The decrease is very sharp initially and then more gradually. A hectorite gel with 0.5M KCl was reported to be in the flocculated phase state.10 For the Laponite dispersions, the onset of the flocculated phase state occurred at a much lower salt concentration, 0.07–0.02M 1:1 salt.21–27 For NaMnt or bentonite gel, the onset salt concentration was much higher 0.2–0.3M NaCl or KCl.9,21 The stepdown shear rate behavior of hectorite gels at a salt concentration less than 0.5M is unknown. This study will determine the effect of salt concentration on the surface forces in play from the response of the stepdown shear stress.

The strength of the EDL repulsive force can be manipulated. It can be increased by increasing the negative charge density or potential of the platelet and the thickness of the electric double layer. Adsorption of an anionic additive, such as pyrophosphate P2O74−, and increasing the gel pH are two such methods.11 Reducing the soluble salt content in the clay gel is one way of increasing the EDL thickness. This study aims to uncover the mechanism of the salt effect on the speed of the structural rejuvenation process in relation to the EDL repulsive force effect on the strength and configuration of the (+)edge–(−)face bond and the speed of bonding.

In aging, the presheared gel is allowed to recover undisturbed with rest time. Hence, the EDL repulsive force required for time-dependent behavior need not be as strong. The initial phase of the structural rejuvenation process involved bond reformation.9 At long times, the dominant process involved the breaking of weak bonds and moving and orienting the platelets to form a stronger bond and structure. It is thus possible for a gel to display time-dependent behavior in aging but not in shear.

The combined aging and stepdown shear rate method was able to provide definitive information on the controlling surface force responsible for the behavior displayed. Other time-dependent methods, such as hysteresis loops of ramp-up and ramp-down flow curve33,34 and storage modulus,34 may require the aid of other methods to identify the controlling forces.

The natural hectorite SHCa-1, a swelling clay, was sourced from the Clay Minerals Society (CMS), USA. This hectorite was obtained from a mine in Red Mountain, which is in the same region as the Newberry Springs mine near Hector, CA. The physical and chemical data of this source clay are available on CMS’s website. Analytical grade (AR) Na4P2O7 · 10H2O chemical was sourced from Sigma Aldrich. It contained 39% by weight of P2O74−. The KCl used was also from Sigma Aldrich.

Hectorite is a 2:1 smectite clay mineral with two tetrahedral silica sheets sandwiching an octahedral magnesia sheet. The magnesia sheet usually contained a significant amount of Li(I) substitution imparting a high negative layer charge to the octahedral sheet. The unit structure of the hectorite SHCa-1 has the following chemical formula (Mg.56 Na.42 K.05)[Mg4.60 Li1.39 MntrTi.01][Si7.75 Al.17 Fe(III).05]O20(OH)4. This hectorite has a layer charge of −1.57e per formula unit. Most of the layer charges are in the octahedral sheet, −1.35e. The tetrahedral layer charge is only −0.22e. The negative pH-dependent charge of the silanol group at the face is not known. There are two possible sources for the edge positive charge: (i) the pH-dependent –Mg(OH) group, which can remain positive at pH > 13,35 and (ii) permanent positive charge if the Mg2+ at the edge is replaced by Al3+ or Fe3+, which are present in this hectorite. The cation exchange capacity (CEC) was reported to be 43.9 meq/100 g and a surface area of 63.19 m2/g. A higher CEC value of 62 meq/100 g for the same hectorite, which was determined with no acid or alkali addition, was reported.36 The dissolution of structural Li and Mg was reported to increase with decreasing pH. Both these Li and Mg ions concentrations were found to be constant from pH 8. The solubility normalized to the edge surface area from pH 8 was 10−6.3 mol m−2 for Li(I) and 10−7 mol m−2 for Mg (II). The concentration of soluble Si was much higher 10−4.3 mol m−2. The dissolution reaction was claimed to be located at the edge of the platelets.

According to Elementis, the company selling natural hectorite rheological modifier made from hectorite mined in Newberry Springs, CA, the hectorite particle has a thin rectangular or lath shape morphology with a dimension of 50 × 250 nm2 and a thickness of 1 nm determined by AFM. In contrast, NaMnt or bentonite from Wyoming has a square shape morphology with the same thickness of 1 nm but a much longer width and a length of 300 nm. Szabó et al.37 displayed the AFM image of the source hectorite clay showing elongated rectangular particles with a measured thickness of ∼2.1 nm. In contrast, AFM images of NaMnt or bentonite showed ellipsoidal shape particles with a thickness of 1.2 nm.38 There were, however, two populations: one 320–400 nm long and 250 nm wide and the other 200–250 nm long and 120 nm wide. The smaller platelet size explains the much larger aging yield stress displayed by the hectorite gels compared to NaMnt at the same solid concentration.10 The hectorite gel will have more platelets interacting attractively.

As-received hectorite SHCa-1 contained as much as 50 wt. % impurities. These impurities were removed by sedimentation of very dilute suspensions containing ∼0.5 wt. % solids prepared in de-ionized (DI) water and the layered platelets were exfoliated by sonication with a high intensity probe. The supernatant containing the exfoliated platelets was collected and dried in an oven at 105 °C. The use of high chemical potential water, i.e., salt-depleted or DI water, also helped the exfoliation process by hydrating interlayer surfaces, forming EDL, and separating the layer.

The purified hectorite gels were prepared by sonicating for ∼2 min an appropriate mass of dried platelets and DI water with a Mixonic sonic probe. The mixture of water and dried platelets was first allowed to sit for a few days to allow the wetting process to occur prior to sonication. The platelets in the low ionic strength environment exfoliated well after the sonication. The gels were again allowed to sit for a few days before any testing. The hectorite gels have a natural pH ∼ 10. Salts at the appropriate concentration were added to the exfoliated gels, stirred, and then rested. KCl, NaCl, and LiCl salts were evaluated. Preparing the gels in concentrated salt solutions must be avoided as salt hindered the exfoliation process. The formation of ionic bridges and strong binding of the interlayer surfaces were suggested to occur. The resultant gels will have a much lower yield stress.39,40 HCl and HNO3 acids were also evaluated.

During aging, the yield stress of the presheared gels was measured as a function of rest time. The gels were allowed to rest undisturbed until the yield stress measurement was performed at a set of predetermined times. The yield stress was measured directly with vane Brookfield viscometers with different spring constants. Each yield stress was measured in a location in the gel not disturbed by a previous measurement. Prior to commencing the aging test, the gels (∼100 g) were first presheared by stirring in circular and side-to-side directions for 2 min. This will breakdown the structure to a steady state or an equilibrium state. In the stepdown shear rate experiment, an Anton Paar MCR72 rheometer was used. A 1° cone-and-plate geometry was chosen. With this geometry, the gel will experience the same shear rate everywhere in the gap. In this test, the gel was sheared for 5 min at 1000 s−1 and then step down abruptly to 10 s−1 and sheared for another 10 min or 600 s. Prior to the test, the gel was stirred with the aim of getting it to attain the steady state faster at the high shear rate. This manual stirring, however, can also bring about a gel structure with a stress state lower than the steady state stress at 1000 s−1. When this happens, the shear stress at 1000 s−1 will rise until the steady state is reached.

The stepdown shear rate test was also performed on hectorite gels treated with both KCl salt and Na4P2O7 · 10H2O. The salt treated hectorite gels of known mass and solid concentration were added with known amounts of Na4P2O7 · 10H2O solids weighed accurately with a four-digit balance. The concentrations of the added additive were converted to dwb% or g of P2O74− per 100 g of hectorite solids.

The zeta potential of the clay suspensions was measured using a ZetaProbe manufactured by Colloidal Dynamics Inc., USA. This instrument measures the electrokinetic sonic amplitude (ESA), which then converts it to dynamic mobility and subsequently to zeta potential. The electroacoustic theory employed by this instrument is based on rigid spherical particles. Dissolved ions also contribute to the ESA, and this contribution must be corrected at high salt or low particulate concentrations. The corrections were made with ESA measured on solutions with the same concentration of KCl as in the gels.

The microstructures were captured on ice sublimed and metal coated gel samples with cryo-SEM. These samples were first subjected to high-pressure cryo-freezing or HF (∼2000 bars) and then cool rapidly at a superfast rate of ∼25 000 °C/s. Amorphous ice formed in the samples in this process was sublimed and coated with Pt.

The effect of pH on the zeta potential of 0.5 and 1 wt. % hectorite gels in Fig. 1 showed a negative zeta potential at all pH. The dependence of the zeta potential on pH is relatively weak. The zeta potential between the two solids loading deviated at pH below 6. Their zeta potential varied ∼−58 mV at pH 10.2 to ∼−50 mV at pH 6. The 1 wt. % gel became perceptibly more viscous at pH below 6. At pH 4, the zeta potential was −40 mV for the 0.5 wt. % gel compared to −50 mV for the 1 wt. %. During the potentiometric titration, the 0.5 wt. % dispersion displayed an electrical conductivity of 0.24–0.53 mS/cm, which corresponded to an equivalent KCl concentration of 0.0016–0.0037M. For the 1 wt. % dispersion, the equivalent KCl concentration varied from 0.0028 to 0.005M. The titration was from pH 10.2 to 3.5.

FIG. 1.

The effect of pH on the zeta potential of 0.5 and 1 wt. % SHCa-1 hectorite dispersions.

FIG. 1.

The effect of pH on the zeta potential of 0.5 and 1 wt. % SHCa-1 hectorite dispersions.

Close modal

At high conductivity, the soluble ions can contribute significantly to the ESA signal used in the zeta potential calculation. The correction error can be minimized by using a high solids hectorite gels in the potentiometric titration. The effect of KCl concentration on the zeta potential of 0.9 and 4 wt. % hectorite in Fig. 2 showed a gradual decrease in the magnitude of the zeta potential. The zeta potential is −75 mV at 0.0025M KCl decreased to −25 mV at 0.5M KCl for the 0.9 wt. % hectorite. For the 4 wt. % hectorite gel, the zeta potential remained essentially unchanged with a value of ∼−60 mV for the KCl concentration ranging from 0.008 to 0.08M.

FIG. 2.

The effect of KCl concentration on the zeta potential of 0.9 and 4 wt. % hectorite dispersions. Below 0.002M KCl, the uncorrected and corrected zeta potential for the background electrolyte were almost the same for the 4 wt. % solid suspensions. The zeta potential of 0.9 wt. % solid suspension required background correction from 0.01M KCl.

FIG. 2.

The effect of KCl concentration on the zeta potential of 0.9 and 4 wt. % hectorite dispersions. Below 0.002M KCl, the uncorrected and corrected zeta potential for the background electrolyte were almost the same for the 4 wt. % solid suspensions. The zeta potential of 0.9 wt. % solid suspension required background correction from 0.01M KCl.

Close modal

The effects of KCl concentration on the aging behavior of 3.61 and 4.7 wt. % hectorite gel (more than 3 years old) in Figs. 3(a) and 3(b) showed a marked increase in the yield stress by the addition of a small amount of salt, 0.002M KCl. For the 3.61 wt. % gel, the increase was as much as 1.63 times at the start aging (t = 1 min) and 1.52 times at the end of aging (t = 10 000 min) compared to the untreated gel at the same aging time. The increase in the gel electrical conductivity was quite small. This gel with an inherent conductivity of 1.04 mS/cm (0.007M KCl equivalent) increased to 1.28 mS/cm (0.009M KCl equivalent), representing a 23% increase. The decrease in the thickness of the electric double layer or EDL according to Eq. (1) should be even smaller. High ionic strength normally induced stronger particle–particle attraction via van der Waals force, which does not normally occur at this level of salt concentration for isotropic, homogeneously charged colloidal suspension. A different gel strengthening process must be present. Upon increasing the KCl concentration to 0.01M, the yield stress increased further, by 3.38 times at the start of aging and 2.94 times at the end. This increase continued but to a smaller extent, 4.36 times at the start to 3.9 times at the end for 0.1M KCl. At the end of the aging test, the gels with 0.01 and 0.1M KCl were stirred and rested for 1 min before performing a yield stress measurement. The same value of 288.6 Pa for the commencing aging yield stress at t = 1 min was measured for the gel with 0.1M KCl. That gel with 0.01M KCl displayed a yield stress 4.9% higher.

FIG. 3.

Effect of KCl concentration on the aging behavior of (a) 3.6 and (b) 4.7 wt. % hectorite gels. For each gel concentration, four sets of aging data were obtained, i.e., for KCl concentration of 0 (untreated), 0.002, 0.01, and 0.1M. The Leong model given by Eq. (4) was used to fit the aging data. The solid lines were the model fits and the corresponding model equations were also given in the figures. The model time constant 1/Kr decreased with salt concentration. These gels were more than 3 years old. They should be at the state of physicochemical equilibrium.

FIG. 3.

Effect of KCl concentration on the aging behavior of (a) 3.6 and (b) 4.7 wt. % hectorite gels. For each gel concentration, four sets of aging data were obtained, i.e., for KCl concentration of 0 (untreated), 0.002, 0.01, and 0.1M. The Leong model given by Eq. (4) was used to fit the aging data. The solid lines were the model fits and the corresponding model equations were also given in the figures. The model time constant 1/Kr decreased with salt concentration. These gels were more than 3 years old. They should be at the state of physicochemical equilibrium.

Close modal

For the 4.7 wt. % gel (3 years old), the increase in the yield stress at the start of aging by the addition of 0.002M KCl was 1.3 times. See Fig. 3(b). The commencing aging yield stress increase (t = 2 min) was 2 and 2.6 times for 0.01 and 0.1M KCl concentration, respectively. This revealed that a more concentrated gel produced a smaller increase.

All gels with 0–0.1M of added KCl displayed time-dependent aging behavior, which is characterized by the yield stress increasing with time. The temporal increase in the aging yield stress or the structural rejuvenation process is spontaneous. This requires a strong EDL repulsive interacting force between platelets in the face–face configuration being present.9,10,12,13 An open structure showing individual hectorite platelets interacting in the (+) edge–(−)face configuration has been reported previously for a gel with and without 0.05M KCl.9,10,12 A similar structure is expected for the gel with 0.1M KCl added. This suggests that the nature of platelet–platelet interaction in the gel has not been affected at this range of KCl concentration. This is not the case with the gel with 0.5M KCl.10 

The EDL thickness or Debye length κ−1 is given by.41 
(1)
where ɛo is the permittivity in free space, ɛ is the relative permittivity, k is the Boltzmann constant, T is the temperature, ρ∞,i is the number density of the ith ion in the bulk, e is the electronic charge, and zi is the valency of the ith ion. For a 1:1 electrolyte, the EDL thickness is κ1=0.304c nm at T = 298 K or 25 °C and c is the salt concentration in M. The thickness returned a value of 3.04 nm for c of 0.01M. The inherent EDL thickness of the 5% hectorite gel is 3.63 nm, which was reduced to 3.24 nm after the addition of 0.002M KCl.

The unexpected strengthening of the gel structure by 0.002M KCl is not usually seen in isotropic and homogeneously charged colloidal suspensions, such as zirconia and alumina.42,43 A much higher salt concentration is needed with isotropic suspensions to increase the strength of interparticle attraction. Here, the van der Waals attractive force dominates the particle–particle interaction. In hectorite, the strengthening of the gel by 0.002M KCl addition is due to stronger bonds being formed in the (+)edge–(−)face. Bonds of different strength are formed, but a larger proportion of them would have a higher strength. The bond strength is dependent on a few factors such as the angle of edge–face attraction, area of interaction, edge morphology, and the extent of like-charge EDL overlapping near the bond. The EDL repulsive interaction was weakened by the decrease in its EDL thickness of 11%, from 3.63 to 3.24 nm. According to Eq. (2), the extent of the weakening in the face–face interaction configuration is only 3.3%, assuming that the surface potential being unaffected by the small amount of salt added. The zeta potential was not affected by the 0.002M KCl addition. Note that not all the assumptions in Eq. (2) are met such as unlike edge charges, length of interacting sheets, and irregular shapes of the platelets. This extent of weakening could facilitate bond formation by increasing the number of lower energy paths to bonding. The range of the (+)edge–(−)face approach angle resulting in bonding would broaden as a result. The bond strength enhancement and the increase in the low energy paths to bonding were sufficient to bring a ∼1.6-fold increase in the aging yield stress.

As the KCl concentration increased, the EDL gets thinner, and the EDL repulsive force weakened further. Higher fraction of the (+)edge–(−)face bond formed would be even stronger. More low energy paths will also be created. The 0.01M KCl gel with a conductivity of 2.14 mS/cm equivalent to 0.016M KCl produced a ∼3.4 times increase in the yield stress. The EDL thickness being 2.4 nm represented a 34% decrease. Assuming no change in the surface potential, the extent of EDL repulsive interaction energy or force lowering at a given separation distance in the face–face configuration is 28% according to Eq. (2). The no change in surface potential assumption may be valid as the change in zeta potential is small. See Fig. 2. The 0.1M KCl gel with a conductivity of 11.53 mS/cm (equivalent to 0.09M KCl) displayed an increase in the commencing aging yield stress by ∼4.4 times, showing a slower rate of increase. The EDL is now 1.01 nm thick. The extent of EDL repulsive force lowering was 51%.

The EDL interaction energy per unit area WEDL between two similarly charged parallel plates is41,
(2a)
where ψo is the surface potential (mV) and H is the face–face separation distance. The repulsive interaction energy increases with the surface potential and EDL thickness κ−1. This equation does not take into account of the flexible platelets with oppositely charged edge. For plates interacting in KCl solution,
(2b)
The corresponding van der Waals interaction energy per unit area is36 
(3)
where A131 is the Hamaker constant of the clay minerals in water. The negative sign denotes the attractive interaction energy. At low to moderate salt concentrations, the van der Waals attractive interaction energy is weak due to uneven edge surface and low interaction area between the edge and face even at a low separation distance. The more consequential attractive interaction is the oppositely charge attraction between edge–face. At high salt concentration, i.e., 0.5M KCl, however, the van der Waals force dominates the platelet–platelet interactions in all configurations.10 
An aging model based on yield stress was used to fit the aging data. The Leong model based on a second order kinetics of particle bond formation in producing the network structure was used. The theory or assumed physics and the derivation of the Leong model were provided in detail by de Kretser and Boger.44 The model is given by
(4)
where τy0 is the yield stress at zero aging time of the presheared gel, τy is the yield stress of the gel at infinite time, and 1/Kr is the model time constant, which measures the time needed to attain 0.68τy. This time constant reflects the speed of the structural rejuvenation process.

This model described the week-long aging behavior of the 3.6 and 4.7 wt. % hectorite gels at all KCl concentrations very well. See the result in Fig. 3. The time constant 1/Kr decreased with KCl concentration for both gels. The values were 714.3, 500, 312.5, and 200 min for the 3.61 wt. % gel with 0.0, 0.002, 0.01, and 0.1M KCl. For the 4.7 wt. % gel, the values were 714.3, 555.6, 500, and 454.5 min for 0.0, 0.002, 0.01, and 0.1M KCl. The time constant thus decreased with decreasing strength of the EDL repulsive force. This is a strong substantiation of a faster bonding process at the higher salt concentration. Higher concentration gels displayed a slower bonding process. The initial phase of the structural rejuvenation process involved bond formation. At longer time, breaking of weak platelet bonds and moving and orienting platelets by the EDL repulsive force to form stronger bonds become more important. This phase of structural rejuvenation becomes dominant several weeks later.9,10

The EDL repulsive force played an important role in determining the time-dependent response of the stepdown shear stress in the shear rate test.11 In this test, the 3.61% hectorite gels with 0.01 and 0.1M KCl were used. See Figs. 4(a) and 4(b). The 0.01M KCl gel showed a stepdown shear stress displaying almost time-independent behavior while that with 0.1M KCl showed time-independent behavior after a short period of decreasing stepdown shear stress. They are the same gels used in the aging study. High consistency gels tend to display much milder time-dependent behavior in the stepdown shear stress response. The difference between the initial stepdown shear stress and the higher steady or constant shear stress is small. A high commencing stepdown shear stress reflects a bonding state being close to saturation. This saturation bonding density occurring at the steady or equilibrium shear stress state should increase with salt concentration. It should decrease with increasing negative charge density of the hectorite platelets caused by a stronger EDL repulsive force. The stronger repulsive force reduced the number of low energy paths to bonding, slowing down the bonding process and accentuating the time-dependent response of the stepdown shear stress.11 The stepdown shear stress of the 0.01M KCl gel treated with 0.15 dwb% P2O74− (dwb% = g additive/100 g solids) showed a pronounced time-dependent behavior. This stress increased sharply initially and then more gradually until it reached the steady state. At the end of shearing, i.e., 600 s later, this stress was 1.66 times larger than that at the onset. A similar time-dependent behavior was displayed at 0.24 dwb% P2O74−. Although the stepdown shear stress was smaller, the stress at the end of the stepdown was 2.26 times larger than that at the onset. This type of time-dependent behavior characterized the EDL repulsive force-control structural rejuvenation process in shear.10 

FIG. 4.

The effect of P2O74− on the stepdown shear stress behavior of 3.61 wt. % hectorite gel with (a) 0.01M, (b) 0.1M, and (c) 0.002M KCl. The panels on the right, (b), (d), and (f), are magnifications of the corresponding panels on the left.

FIG. 4.

The effect of P2O74− on the stepdown shear stress behavior of 3.61 wt. % hectorite gel with (a) 0.01M, (b) 0.1M, and (c) 0.002M KCl. The panels on the right, (b), (d), and (f), are magnifications of the corresponding panels on the left.

Close modal

The gel with 0.1M KCl required a higher pyrophosphate concentration to produce a similar but much weaker time-dependent response. The use of 0.15 dwb% P2O74− did not strengthen the EDL repulsive force sufficiently. No display of the EDL repulsive force-control time-dependent behavior in the stepdown shear stress was seen. The initial stepdown shear stress was, in fact, larger by 16% than that at the end of stepdown. The EDL repulsive force controlled-time dependent behavior was observed at 0.37 dwb% P2O74−. The stepdown shear stress being relatively high displayed a value at the end of stepdown of only 1.22 times larger than that at the onset. At 0.74 dwb% P2O74−, it was slightly larger, 1.33 times. The highly compressed EDL, from 3.63 to 1 nm, required a much larger negative charge density via a higher P2O74− adsorption to generate an EDL repulsive force with sufficient strength to produce the time-dependent behavior. The stepdown shear stress remained much more elevated at a given P2O74− concentration compared to the gel with 0.01M KCl. This reflected a higher bonding density. In the absence of an imposed shear, the EDL repulsive force in this gel remained strong enough to produce time-dependent behavior in the aging even without the addition of P2O74−. It is quite remarkable that significant P2O74− adsorption and strong EDL repulsion can still occur at this level of ionic strength—a remarkable property of charge and shape anisotropic colloidal materials. The effect of P2O74− on the stepdown shear rate behavior was also evaluated for the gel with 0.002M KCl, and the result is shown in Fig. 4(c). The gel showed time-dependent behavior in the stepdown shear stress response even without P2O74− being used. The addition of P2O74− increased the stepdown shear stress at the end more than the initial value. Without additive, the stepdown shear stress at the end for the two tests were 1.24 and 1.42 times. At 0.12 and 0.28 dwb% P2O74−, it is 2.49 and 2.21 times, respectively.

The reason for the significant increase in the aging yield stress by the small amount of added salt is now clear. The added salt facilitated the platelet–platelet bonding process and produced stronger bond. It increased the number of approach configurations for bonding markedly, and this was aided by the marginally weakended EDL repulsive force and by the absence of an imposed shear. The platelet–platelet bonding process initiated first by the overlap of the positively charged EDL of the face and the negatively charged EDL of the edge can now occur at a more acute angle of approach.11 Note that the zero-degree approach angle, i.e., parallel plate configuration, is unlikely to result in bonding except when it occurs near the edge, i.e., overlapping face–face bonding at the platelet edge. A strong charge anisotropy condition is required.16,17 The feature of the EDL at the boundary of the negative and positive EDLs near the edge is unclear or not well-defined and may play an important role in facilitating bonding when compressed by a small amount of added salt.

At 0.01M KCl, the number of approach configurations favorable to bonding increased further, resulting in a much larger aging yield stress increase by an average of 3.1 times for the 3.61 wt. % gel. This increases further but at a slower rate to 4.2-times when the salt added was increased to 0.1M KCl. The compression of the EDL became significant, leading generally to stronger bonds being formed. This is especially true when the overlap of unlike charge EDLs in the location of the edge–face bond did not involve like charge EDLs overlap further away especially for the acute angle edge–face bonds. When the EDL is very thick, like charge EDLs overlap involving the silica faces, near the acute angle, edge–face bonds can be significant.

The schematic of platelet–platelet interactions in Fig. 5 is used to reinforce our explanations of the salt effects on the time-dependent behavior of hectorite gels shown in Figs. 3 and 4 to substantiate our premise. The need of strong face–face EDL repulsion for time-dependent behavior in the structural rejuvenation mode is illustrated. This face–face mode of interaction is commonly encountered by the platelets in the gel, and the strong EDL repulsive force they experienced will move or bounce them around until they are in the right edge–face configuration to bond. This edge–face approach has a low energy barrier and so bonding occurs easily.14–19 The schematic also shows the effects of salt concentration on the nature of (+)edge–(−)face bonding. Low salt concentration or thick EDL restricts bonding in the edge–face configuration to those at or close to 90°. The acute angle edge–face bond is unstable due to significant overlap of like-charge EDLs at the bond. See the image in Fig. 5. This will slow down the bonding process and cause the gel to display time-dependent behavior. At high salt concentrations, the thin EDL layer permits stable edge–face bonds to form at very acute angles. The overlapping of like-charge EDLs at the bond encountered in the case of thick EDL is avoided. This increases the low energy paths to bonding, thereby speeding up the bonding process. Time-dependent behavior may not be discernible especially when the bonding speed is very fast particularly in the stepdown shear rate mode.

FIG. 5.

Schematic of platelet–platelet interactions with views taken across a vertical plane through the platelets. (a) Strong EDL repulsion caused by like-charge EDLs overlap in the face–face configuration. This condition needs to be met for all hectorite gels showing time-dependent behavior in the structural rejuvenation mode. (b) Condition of low salt gel with a thick double layer, stable bonding can only form when the (+)edge–(−)face approach is ∼90°. Unstable acute angle (+)edge–(−)face bond is formed especially when the extent of like-charge EDL overlap is significant. The bonding process in the gel is slow. (c) Condition of high salt gel with a thin double layer, a stable edge–face bond is formed even when the angle is very acute. The energy barrier to bonding, which was insurmountable when the EDL is thick, is now weak enough for bonding to occur. The bonding process in the gel is fast.

FIG. 5.

Schematic of platelet–platelet interactions with views taken across a vertical plane through the platelets. (a) Strong EDL repulsion caused by like-charge EDLs overlap in the face–face configuration. This condition needs to be met for all hectorite gels showing time-dependent behavior in the structural rejuvenation mode. (b) Condition of low salt gel with a thick double layer, stable bonding can only form when the (+)edge–(−)face approach is ∼90°. Unstable acute angle (+)edge–(−)face bond is formed especially when the extent of like-charge EDL overlap is significant. The bonding process in the gel is slow. (c) Condition of high salt gel with a thin double layer, a stable edge–face bond is formed even when the angle is very acute. The energy barrier to bonding, which was insurmountable when the EDL is thick, is now weak enough for bonding to occur. The bonding process in the gel is fast.

Close modal

If the gel has not reached the physicochemical equilibrium state, the added salt would produce a much greater effect. The salt produced a very large increase in the aging yield stress with fresh gels because the yield stress of the untreated fresh gel was very low. See Fig. 6(a) for the effect of KCl on 5 wt. % fresh gels, ∼1-week old. The salt concentration in the range of 0.01–0.1M KCl produced a commencing aging yield stress of ∼400 Pa. This yield stress is very close to that obtained by the 3 year-old gel. The commencing yield stress of the untreated fresh gel was only ∼10 Pa. This yield stress was found to increase with the age of the gel. It was ∼50 Pa for a 4-month-old gel9 as seen in Fig. 6(b) and ∼270 Pa for the 3-year-old gel. This suggests that the salt-treated gels reached its physicochemical equilibrium state much faster. The effect of 0.002M KCl on the yield stress is also greater for fresher gel, 4 months old compared to that obtained with the 3-year-old gels. See Fig. 6(b), which showed, on average, 2.4 times increase in the aging yield stress by 0.002M KCl. This is significantly more than the 1.3 times obtained with the 3-year-old 4.7 wt. % gel.

FIG. 6.

(a) Effect of KCl concentration on the aging behavior of fresh 5.0 wt. % hectorite gels. These gels were 1 week old at the start of the aging test. (b) The effect of 0.002M KCl on 4.6 wt. % hectorite gels that were 4 months old. The Leong model time constant is 4255 and 2062 min for this gel without and with 0.002M KCl. A few drops of a 10 wt. % KCl solution were added to the gel to the level 0.002M KCl. The electrical conductivity increased from 1.115 to 1.432 mS cm−1. (c) Effect of alkali metal ions on the aging behavior of 5 wt. % hectorite gels. (0.05M LiCl: pH 8.96-8.4, zeta potential of −51 mV; 0.05M NaCl: pH 9.1 and 4.54 wt. % masterbatch sample: pH 9.9, zeta potential of −47.3 mV). The aging test was performed on 1–2-day-old gels. (d) The effect of HNO3, HCl, and pH on the aging behavior fresh 5 wt. % hectorite gels. These gels were 4 to 3 weeks old. The natural pH of fresh gel ranged from 9.4 to 9.8. The HCl treated gels with a pH of 8.45 and 6.43 have a solid concentration of 4.45 and 4.34 wt. %.

FIG. 6.

(a) Effect of KCl concentration on the aging behavior of fresh 5.0 wt. % hectorite gels. These gels were 1 week old at the start of the aging test. (b) The effect of 0.002M KCl on 4.6 wt. % hectorite gels that were 4 months old. The Leong model time constant is 4255 and 2062 min for this gel without and with 0.002M KCl. A few drops of a 10 wt. % KCl solution were added to the gel to the level 0.002M KCl. The electrical conductivity increased from 1.115 to 1.432 mS cm−1. (c) Effect of alkali metal ions on the aging behavior of 5 wt. % hectorite gels. (0.05M LiCl: pH 8.96-8.4, zeta potential of −51 mV; 0.05M NaCl: pH 9.1 and 4.54 wt. % masterbatch sample: pH 9.9, zeta potential of −47.3 mV). The aging test was performed on 1–2-day-old gels. (d) The effect of HNO3, HCl, and pH on the aging behavior fresh 5 wt. % hectorite gels. These gels were 4 to 3 weeks old. The natural pH of fresh gel ranged from 9.4 to 9.8. The HCl treated gels with a pH of 8.45 and 6.43 have a solid concentration of 4.45 and 4.34 wt. %.

Close modal

Like KCl, the addition of 0.05M LiCl and NaCl also caused a large increase in the aging yield stress of a 5 wt. % hectorite gel. See Fig. 6(c). These gels were only a day-old when the test was conducted. There is a small but appreciable difference in the yield stress enhancement in the following order, K > Na > Li, which appeared to be related to the size of the hydrated alkali metal ions. The radii of the hydrated ions are 340, 276, and 232 pm for Li+, Na+, and K+, respectively. The smaller hydrated K+ ions can bind more strongly to the platelet surface as it can get closer to the opposite charged sites displacing some of the fixed ions forming the Stern layer. This should reduce the surface potential of the hectorite particles, enabling stronger platelet–platelet attraction. There were computer simulations that captured the swelling inhibition effect of K+, revealing the possible mechanism45,46 for the stronger platelet–platelet attraction. However, the zeta potential of the gel with 0.05M LiCl is the same as the masterbatch gel containing no added salt, ∼−50 mV. This display of the time-dependent behavior implied that the EDL repulsive force remained quite strong. The (Li + Na)/Si ratio was reported to determine the gel strength of Laponite.47 K was not included in this study. Similarly, the yield stress of pure spherical silica suspension was found to be higher in K salt than that in Na or Li salts.48 An ion–ion correlation force was invoked to explain the larger increase in the yield stress by K+.

The effects of HCl and HNO3 on the aging behavior of 5 wt. % SHCa-1 gels that were a few weeks old in Fig. 6(d) showed a large commencing yield stress of ∼300 Pa. The gel concentration was slightly lower, and this explained the slightly lower magnitude compared to that obtained with gels treated with 0.05M salt. The aging behavior of the treated gel was not affected by the pH or nature of the anions. The high yield stress was caused by a relatively small decrease in the pH from ∼9.8 to 8.5 despite no significant change in the magnitude of the negative zeta potential. The magnitude of the yield stress was the same for the HCl-treated gels at pH 7.57 or 8.45. The gel treated with HNO3 to a pH of 7.9 produced the same aging yield stress. The pH of the gel did increase with time. The gel treated with HCl to a pH of 6.43 increased to pH 7.57 overnight. This gel acquired a conductivity of 2.7 mS/cm or 0.02M KCl equivalent. Another gel treated with HCl to a pH of 8.11 also increased to pH 8.45 overnight has a lower conductivity of 1.7 mS/cm, displaying almost identical aging behavior. The 4.6 wt. % gel treated with HNO3 to a pH of 7.9 and a conductivity of 2.53 mS/cm also displayed similar aging behavior. The display of aging time-dependent behavior revealed that the (+)edge–(−)face attraction remained the main bonding type in all these treated gels. The EDL repulsive force was weakened by the pH reduction but remained strong enough to control the structural development process. It breaks weak platelet bonds, moving and orienting these platelets to form a stronger bond. The lowering of the pH facilitated bonding, increasing the bond density and hence the yield stress. The number of low energy paths for bonding increased significantly, covering a wider range of edge–face angle and a broader range of edge–face orientations.

Earlier, it was advocated that the stepdown shear rate test alone be used to determine the flocculated phase state.9,10 In view of this study, the use of the time-independent aging yield stress test result to complement the stepdown test result would provide a more accurate determination of the flocculated state. In the flocculated phase state, both the stepdown shear stress and aging yield stress do not increase immediately with time. This additional test is required because of the display of very weak time-dependent behavior in the stepdown shear stress by the gels treated with moderate salt concentration.

The microstructure of a 4-month-old, 5 wt. % hectorite gel in Figs. 7(a) and 7(b) showed flexible exfoliated platelets forming the open network structure. The platelets were small (<1 μm) and very thin and flexible with curled surfaces and edges. Some of the edges were quite jagged, forming highly curl-up strings radiating outward. Many of the platelets were also bent with curled edges exposing a large part of their faces when they were seated at an angle from vertical. No two edge–face bonding configurations are the same as seen in the magnified images of microstructure sections highlighted in Figs. 7(a) and 7(b). Some of these varied edge–face configurations were also quite difficult to describe. The range of edge–face configuration and orientation was increased by the flexibility and jagged nature of the platelets. The face area at the edge may bond to the face of another platelet well away from its edge in the face–face configuration. This can be seen in platelets with an L-shaped edge resting on the face of another platelet.

FIG. 7.

The microstructure of 4.6 wt. % hectorite gels without added KCl prepared 4 months earlier. Flexible curled platelet bonding between face–edge configuration can be seen clearly in (c) and (d). The face at the edge was bent by 90°.

FIG. 7.

The microstructure of 4.6 wt. % hectorite gels without added KCl prepared 4 months earlier. Flexible curled platelet bonding between face–edge configuration can be seen clearly in (c) and (d). The face at the edge was bent by 90°.

Close modal

The microstructure of the gel treated with 0.002M KCl in Fig. 8(a) and at higher magnification of selected sections in Fig. 8(b) also displayed a very open network structure. The flexible platelets with curled edges appeared to adopt a more vertical orientation. The angle of fracture of the high pressured cryo-frozen gel sample may present a view that appeared to show the marked difference in the platelet orientation in the network. The edge–face bonds forming the network structure were also varied.

FIG. 8.

(a) The microstructure of 4-month-old hectorite gel with 0.002M KCl at 4.6 wt. % solids and (b) the magnified image of three sections highlighted by three blue circles in (a). The KCl addition increased the conductivity from 1.12 to 1.43 mS/cm. This microstructure was captured 3 weeks after salt addition. (c) The microstructure of 4 wt. % hectorite gel with 0.05M KCl.

FIG. 8.

(a) The microstructure of 4-month-old hectorite gel with 0.002M KCl at 4.6 wt. % solids and (b) the magnified image of three sections highlighted by three blue circles in (a). The KCl addition increased the conductivity from 1.12 to 1.43 mS/cm. This microstructure was captured 3 weeks after salt addition. (c) The microstructure of 4 wt. % hectorite gel with 0.05M KCl.

Close modal

The microstructure of the 4 wt. % hectorite gel with 0.05M KCl possessed features very similar to that displayed by the gels with 0 and 0.002M KCl added. The edge–face bond formed the open network structure. A high bonding density in the gel is expected to increase with salt concentration. It is, however, difficult to quantify in these 2D microstructure images and complicated by the angle of fracture of the sample.

Hectorite gels appeared to take quite a while to attain the physicochemical equilibrium state. This was reflected by its changing stepdown shear stress response spanning months. See Fig. 9. The 4.54 wt. % hectorite gel was prepared by sonication on 19th April 2023. The response curve of the stepdown shear stress moved increasingly to the higher shear stress region with time since preparation. This upward movement appeared to stop on 14th November 2023, i.e., 7 months later. A much later test on 24th July 2024 showed a slight increase in the stepdown shear stress. The shear stress at the onset of stepdown moved increasingly closer to the steady stepdown shear stress, reflecting a progressively milder time dependent response as the gel aged. The degree of time-dependent behavior can also be measured by the relative increase in the stepdown shear stress at the end of shearing. The test conducted 26 days later [test on (T):15 May 2023] showed that the stepdown shear stress at the end of shearing, 600 s later, was 2.11 times larger than that at the start. This decreased to 1.48 times when tested 298 days since preparation (T:2 February 2024). The shear stress at 1000 s−1 at the end of shearing at 300 s also increased with the age of the gel. The solid concentration of the gel was kept the same by dilution with water, especially for those that were several months old. Some of the characterization was performed before and after dilution of the gel to its original concentration. Earlier, it was reported that a hectorite gel with an age of 2.7 years displayed yield stress 2.6 times that at the start.11 Both its solids concentration and electrical conductivity had also increased with time. It also displayed a lower pH. The concentration was found to have increased from 6.3 to 6.72 wt. % solids and the conductivity from 1.5 to 1.76 mS/cm. The pH was reduced from 9.4 to 8.6.

FIG. 9.

The variation of stepdown shear rate behavior of 5 wt. % hectorite gel with time since preparation.

FIG. 9.

The variation of stepdown shear rate behavior of 5 wt. % hectorite gel with time since preparation.

Close modal

As the gel aged, the steady state shear stress at the high and low shear rates were high. This means that the bond concentration is high. The rate of bond breaking and reforming must also be high and equal at the steady state. The speed of bonding was hastened by the progressive weakening of the EDL repulsive force as the gel aged. The bonds formed would therefore, in general, be stronger too. The initial stepdown shear stress being closer to the steady value as the gel gets older is an indication of a faster bonding speed at the point of stepdown so much so the bond density or concentration is close to that at the steady state and, hence, the milder time-dependent behavior.

The result appeared to show that the gel has reached the physicochemical equilibrium state at 7 months. The slight increase in the equilibrium shear stress after 15 months highlighted the difficulty in pinpointing the exact time of this event being attained.

A similar experiment was performed by Shahin and Joshi49 on a 2.8 wt. % Laponite gel. The variation of the storage modulus with oscillatory shearing time was determined as a function of time since its preparation or it was kept idle. Prior to the modulus characterization, the gel was first presheared in the oscillatory shear mode to a steady state at a shear stress of 70 Pa and at a fixed frequency of 0.1 Hz. The storage modulus measured at 10 Pa and 0.1 Hz was found to increase with time since its preparation, a result qualitatively similar to the stepdown shear stress result obtained here shown in Fig. 9. The increase continued even after 52 days since its preparation. The effect of NaCl concentration in the range of 0.1–7 mM was also evaluated. The storage modulus also increased with salt concentration.

The addition of 0.002M of KCl caused a marked increase in the aging yield stress of 3.61 and 4.7 wt. % hectorite gel that was more than 3 years old despite that the inherent salt content was more than three times that of the added salt. The added salt strengthened the bond formed and increased the number of low energy paths for bond formation in the (+)edge–(−)face configurations. This occurred despite the marginal weakening of the EDL repulsive force and a 11% reduction in the EDL thickness. The usual stress response to the low amount of added salt is due to its effect on the (+)edge–(−)face bond found mainly in charge and shape anisotropic clay gels.

The aging yield stress increased further with KCl addition up to 0.1M. More low energy paths are created, hastening the bonding process further. Even stronger bonds are also produced. Bonding occurred over an even wider range of angle covering a greater variety of edge–face interactions, such as curled edge–face interaction.

The gels at all salt concentrations displayed time-dependent aging behavior, a strong indication of EDL repulsive force controlling the structural rejuvenation process at rest. This spontaneous process involved bond forming in the initial phase of aging with EDL repulsive directing the edge–face bonding. Later, the breaking of weak platelet bonds, moving and orienting them to form stronger bond and structure by EDL repulsive force, becomes the more dominant process. Leong models fitted the aging data very well at all salt concentrations. The model time constant decreased with increasing salt concentration. This is a strong substantiation of a faster bonding process occurring.

The 3.61 wt. % gels with 0.01 and 0.1M KCl did not show EDL repulsive force-control time-dependent behavior in the stepdown shear stress response. Increasing the negative charge density of the platelets by pyrophosphate adsorption caused the stepdown shear stress of these gels to display this time-dependent behavior prominently. In shear, a higher negative charge density of the platelets is needed to produce this time-dependent behavior.

A relatively small decrease in pH also caused a large increase in the yield stress in aging. A reduced negative charge density of the platelets was responsible.

The gel structure remained open and porous after treatment with 1:1 salt up to 0.05M concentration. Li(I) salt produced a smaller increase in the aging yield stress compared to K(I) and Na(I).

Natural hectorite gels take months to reach physicochemical equilibrium, and this was accompanied by a decrease in pH and an increase in electrical conductivity, resulting in an increasing yield stress or shear stress. The gel with a higher shear stress displayed a milder form of time-dependent behavior in the stepdown shear stress response.

We wish to thank Dr. Pengfei Liu for the gel preparation and capturing some of microstructure images presented. We acknowledge the use of Microscopy Australia facilities within the CMCA@UWA, which are funded by the University, plus State and Commonwealth Governments. We would like to thank the referees for making this a better paper.

The authors have no conflicts to disclose.

Yee-Kwong Leong: Conceptualization (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Writing – original draft (equal); Writing – review & editing (equal). Peta Clode: Methodology (equal); Visualization (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
H.
Van Olphen
, “
Rheological phenomena of clay sols in connection with the charge distribution on the micelles
,”
Discuss. Faraday Soc.
11
,
82
84
(
1951
).
2.
S. W.
Jeong
,
J.
Locat
, and
S.
Leroueil
, “
The effects of salinity and shear history on the rheological characteristics of illite-rich and Na-montmorillonite-rich clays
,”
Clays Clay Miner.
60
,
108
120
(
2012
).
3.
D.
Penner
and
G.
Lagaly
, “
Influence of anions on the rheological properties of clay mineral dispersions
,”
Appl. Clay Sci.
19
,
131
142
(
2001
).
4.
B.
Abu-Jdayil
, “
Rheology of sodium and calcium bentonite–water dispersions: Effect of electrolytes and aging time
,”
Int. J. Miner. Process.
98
,
208
213
(
2011
).
5.
M. Y.
Du
,
P. F.
Liu
,
J. E.
Wong
,
P. L.
Clode
,
J.
Liu
, and
Y. K.
Leong
, “
Colloidal forces, microstructure and thixotropy of sodium montmorillonite (SWy-2) gels: Roles of electrostatic and van der Waals forces
,”
Appl. Clay Sci.
195
,
105710
(
2020
).
6.
J.
Ren
,
Y.
Deshun
, and
R.
Zhai
, “
Rheological behavior of bentonite-water suspension at various temperatures: Effect of solution salinity
,”
Eng. Geol.
295
,
106435
(
2021
).
7.
H.
Kimura
,
M.
Sakurai
,
T.
Sugiyama
,
A.
Tsuchida
,
T.
Okubo
, and
T.
Masuko
, “
Dispersion state and rheology of hectorite particles in water over a broad range of salt and particle concentrations
,”
Rheol. Acta
50
,
159
168
(
2011
).
8.
M.
Benna
,
N.
Kbir-Ariguib
,
A.
Magnin
, and
F.
Bergaya
, “
Effect of pH on rheological properties of purified sodium bentonite suspensions
,”
J. Colloid Interface Sci.
218
,
442
455
(
1999
).
9.
Y. K.
Leong
,
P.
Liu
,
P.
Clode
, and
J.
Liu
, “
Ageing behaviour spanning months of NaMt, hectorite and Laponite gels: Surface forces and microstructure – A comprehensive analysis
,”
Colloids Surf., A
630
,
127543
(
2021
).
10.
Y. K.
Leong
and
P. L.
Clode
, “
Time-dependent clay gels: Stepdown shear rate behavior, microstructure, ageing, and phase state ambiguity
,”
Phys. Fluids
35
,
123329
(
2023
).
11.
Y. K.
Leong
, “
Direct evidence of electric double layer (EDL) repulsive force being responsible for the time-dependent behavior of clay gels in the structural rejuvenation mode
,”
J. Phys. Chem. B
128
,
3784
3793
(
2024
).
12.
M.
Du
,
J.
Liu
,
P. L.
Clode
, and
Y. K.
Leong
, “
Surface chemistry, rheology and microstructure of purified natural and synthetic hectorite suspensions
,”
Phys. Chem. Chem. Phys.
20
,
19221
19233
(
2018
).
13.
Y. K.
Leong
,
M.
Du
,
P. I.
Au
,
P.
Clode
, and
J.
Liu
, “
Microstructure of sodium montmorillonite gels with long aging time scale
,”
Langmuir
34
,
9673
9682
(
2018
).
14.
M.
Dijkstra
,
J. P.
Hansen
, and
P. A.
Madden
, “
Statistical model for the structure and gelation of smectite clay suspensions
,”
Phys. Rev. E
55
,
3044
3053
(
1997
).
15.
G.
Odriozola
,
M.
Romero-Bastida
, and
F.
de J Guevara-Rodríguez
, “
Brownian dynamics simulations of Laponite colloid suspensions
,”
Phys. Rev. E
70
,
021405
(
2004
).
16.
B.
Jönsson
,
C.
Labbez
, and
B.
Cabane
, “
Interaction of nanometric clay platelets
,”
Langmuir
24
,
11406
11413
(
2008
).
17.
M.
Delhorme
,
B.
Jönsson
, and
C.
Labbez
, “
Monte Carlo simulations of a clay inspired model suspension: The role of rim charge
,”
Soft Matter
8
,
9691
9704
(
2012
).
18.
A.
Shahin
and
Y. M.
Joshi
, “
Physicochemical effects in aging aqueous Laponite suspensions
,”
Langmuir
28
,
15674
15686
(
2012
).
19.
B.
Ruzicka
,
E.
Zaccarelli
,
L.
Zulian
,
R.
Angelini
,
M.
Sztucki
,
A.
Moussaïd
,
T.
Narayanan
, and
F.
Sciortino
, “
Observation of empty liquids and equilibrium gels in a colloidal clay
,”
Nat. Mater.
10
,
56
60
(
2011
).
20.
K.
Suman
and
Y. M.
Joshi
, “
Microstructure and soft glassy dynamics of an aqueous Laponite dispersion
,”
Langmuir
34
,
13079
13103
(
2018
).
21.
J.-C. P.
Gabriel
,
C.
Sanchez
, and
P.
Davidson
, “
Observation of nematic liquid-crystal textures in aqueous gels of smectite clays
,”
J. Phys. Chem.
100
,
11139
11143
(
1996
).
22.
A.
Mourchid
,
A.
Delville
,
J.
Lambard
,
E.
Lecolier
, and
P.
Levitz
, “
Phase diagram of colloidal dispersions of anisotropic charged particles: Equilibrium properties, structure, and rheology of Laponite suspensions
,”
Langmuir
11
,
1942
1950
(
1995
).
23.
A.
Mourchid
,
E.
Lecolier
,
H.
Van Damme
, and
P.
Levitz
, “
On viscoelastic, birefringent, and swelling properties of Laponite clay suspensions: Revisited phase diagram
,”
Langmuir
14
,
4718
4723
(
1998
).
24.
H.
Tanaka
,
J.
Meunier
, and
D.
Bonn
, “
Nonergodic states of charged colloidal suspensions: Repulsive and attractive glasses and gels
,”
Phys. Rev. E
69
,
031404
(
2004
).
25.
P.
Mongondry
,
J. F.
Tassin
, and
T.
Nicolai
, “
Revised state diagram of Laponite dispersions
,”
J. Colloid Interface Sci.
283
,
397
405
(
2005
).
26.
S.
Jabbari-Farouji
,
H.
Tanaka
,
G. H.
Wegdam
, and
D.
Bonn
, “
Multiple nonergodic disordered states in Laponite suspensions: A phase diagram
,”
Phys. Rev. E
78
,
061405
(
2008
).
27.
B.
Ruzicka
and
E.
Zaccarelli
, “
A fresh look at the Laponite phase diagram
,”
Soft Matter
7
,
1268
1286
(
2011
).
28.
M.
Dijkstra
,
J. P.
Hansen
, and
P. A.
Madden
, “
Gelation of a clay colloid suspension
,”
Phys. Rev. Lett.
75
,
2236
2239
(
1995
).
29.
F.
Capuani
,
I.
Pagonabarraga
, and
D.
Frenkel
, “
Lattice-Boltzmann simulation of the sedimentation of charged disks
,”
J. Chem. Phys.
124
,
124903
(
2006
).
30.
D.
Ebrahimi
,
A. J.
Whittle
, and
R. J.-M.
Pellenq
, “
Mesoscale properties of clay aggregates from potential of mean force representation of interactions between nanoplatelets
,”
J. Chem. Phys.
140
,
154309
(
2014
).
31.
F. A.
de Melo Marques
,
R.
Angelini
,
G.
Ruocco
, and
B.
Ruzicka
, “
Isotopic effect on the gel and glass formation of a charged colloidal clay: Laponite
,”
J. Phys. Chem. B
121
,
4576
4582
(
2017
).
32.
M.
Bier
,
L.
Harnau
, and
S.
Dietrich
, “
Surface properties of fluids of charged platelike colloids
,”
J. Chem. Phys.
125
,
184704
(
2006
).
33.
A. J. W.
ten Brinke
,
L.
Bailey
,
H. N. W.
Lekkerkerker
, and
G. C.
Maitland
, “
Rheology modification in mixed shape colloidal dispersions. Part II: Mixtures
,”
Soft Matter
4
,
337
348
(
2008
).
34.
L.
Bailey
,
H. N. W.
Lekkerkerker
, and
G. C.
Maitland
, “
Smectite clay – inorganic nanoparticle mixed suspensions: Phase behaviour and rheology
,”
Soft Matter
11
,
222
236
(
2015
).
35.
X.
Liu
,
X.
Lu
,
M.
Sprik
,
J.
Cheng
,
E. J.
Meijer
, and
R.
Wang
, “
Acidity of edge surface sites of montmorillonite and kaolinite
,”
Geochim. Cosmochim. Acta
117
,
180
190
(
2013
).
36.
L.
Delavernhe
,
M.
Pilavtepe
, and
K.
Emmerich
, “
Cation exchange capacity of natural and synthetic hectorite
,”
Appl. Clay Sci.
151
,
175
180
(
2018
).
37.
T.
Szabó
,
J.
Wang
,
A.
Volodin
,
C.
van Haesendonck
,
I.
Dekany
, and
R. A.
Schoonheydt
, “
AFM study of smectites in hybrid Langmuir-Blodgett films: Saponite, Wyoming bentonite, hectorite, and Laponite
,”
Clays Clay Miner.
57
,
706
714
(
2009
).
38.
A.
Cadene
,
S.
Durand-Vidal
,
P.
Turq
, and
J.
Brendle
, “
Study of individual Na-montmorillonite particles size, morphology, and apparent charge
,”
J. Colloid Interface Sci.
285
,
719
730
(
2005
).
39.
W.-Z.
Chang
and
Y. K.
Leong
, “
Ageing and collapse of bentonite gels—Effects of Li, Na, K and Cs ions
,”
Rheol. Acta
53
,
109
122
(
2014
).
40.
P. I.
Au
and
Y. K.
Leong
, “
Rheological and zeta potential behaviour of kaolin and bentonite composite slurries
,”
Colloids Surf., A
436
,
530
541
(
2013
).
41.
J. N.
Israelachvili
,
Intermolecular and Surface Forces
(
Academic Press
,
1992
).
42.
Y. K.
Leong
,
N.
Katiforis
,
D. B. O’C.
Harding
,
T. W.
Healy
, and
D. V.
Boger
, “
Role of rheology in colloidal processing of ZrO2
,”
J. Mater. Process. Manuf. Sci.
1
,
445
453
(
1993
).
43.
S. B.
Johnson
,
G. V.
Franks
,
P. J.
Scales
,
D. V.
Boger
, and
T. W.
Healy
, “
Surface chemistry–rheology relationships in concentrated mineral suspensions
,”
Int. J. Miner. Process.
58
,
267
304
(
2000
).
44.
R. G.
de Kretser
and
D. V.
Boger
, “
A structural model for the time-dependent recovery of mineral suspensions
,”
Rheol. Acta
40
,
582
590
(
2001
).
45.
G.
Odriozola
and
J. F.
Aguilar
, “
Stability of K-montmorillonite hydrates: Hybrid MC simulations
,”
J. Chem. Theory Comput.
1
,
1211
1220
(
2005
).
46.
X.
Li
,
Q.
Li
,
S.
Yang
, and
G.
Yang
, “
Swelling of clay minerals: Dual characteristics of K+ ions and exploration of critical influencing factors
,”
Phys. Chem. Chem. Phys.
21
,
1963
1971
(
2019
).
47.
K.
Torii
and
T.
Iwasaki
, “
Synthesis of hectorite
,”
Clay Sci.
6
,
1
16
(
1987
).
48.
G. V.
Franks
, “
Zeta potentials and yield stresses of silica suspensions in concentrated monovalent electrolytes: Isoelectric point shift and additional attraction
,”
J. Colloid Interface Sci.
249
,
44
51
(
2002
).
49.
A.
Shahin
and
Y. M.
Joshi
, “
Irreversible aging dynamics and generic phase behavior of aqueous suspensions of laponite
,”
Langmuir
26
(
6
),
4219
4225
(
2010
).