Rheological Behavior of an Algerian Natural Kaolin: Effect of Dispersant

Abstract

This work presents the study of the rheology ical behavior of Algerian kaolin (DD1) suspensions considering two types of electro-steric dispersants (Hypermer KD1 and Darvan 7) and the evaluation of their effectiveness at neutral pH. The results showed that Darvan 7 exhibits electro-steric behavior at neutral pH, whereas KD1 exhibits purely steric behavior. The addition of a dispersant strongly influenced the rheological behavior of kaolin suspensions. The DD1 suspensions without dispersant exhibited fluidifying plastic behavior (Casson model). The shear stresses decreased significantly with the addition of dispersant, while the significant decrease in viscosity indicated that the dispersant reduced the strength of the particle networks that make up the slurry. The suspensions with 1 wt.% dispersant were consistent with the Bingham model, with a very low yield point. The viscosity of the dispersion reached a minimum when the concentration of the dispersant was 1 wt.%. This value was lower with Darvan 7. The addition of aluminum slag as a source of alumina to KD1 increased its efficiency and lowered the viscosity of the kaolin suspensions.

1. Introduction

Kaolin clay, with the structural formula Al₂O₃·2SiO₂·2H₂O, presents itself in the form of hexagonal lamellar particles formed by the stacking of elementary sheets [1]. A kaolin particle possesses a positive charge on the edge and a negative charge on the surface [2]. The charge difference between the surface and the edges of kaolin particles generates a strong interparticle attraction force. The basal face is composed of tetrahedral siloxane species (–Si–O–Si–) [3,4], while the other face is composed of an octahedral alumina sheet (Al₂O₃). The bond between one sheet and an adjacent sheet is ensured by relatively energetic hydrogen bonds, making the stacks highly cohesive. At the edges of the kaolin crystal, the octahedral alumina and the tetrahedral silica sheets show altered bonds, thus revealing aluminol (Al–OH) and silanol (Si–OH) groups exposed on the surface. Depending on the pH of the suspension, the edge face can be charged due to the protonation and deprotonation of the surface hydroxyl groups [3,5]. At neutral to high pH, kaolinite particles are generally negatively charged in water and thus tend to form stable dispersions.

The properties of suspensions as well as their rheological behaviors are influenced by numerous factors such as suspension concentration, pH, nature, purity and size of the kaolin particles, type of dispersant, suspension temperature, and shear rate. In this context, several studies [3,4,5,6,7] have been conducted to explain the influence of these various factors on the suspension properties and the rheological behavior of pure kaolin.

Rheology is determined by the occurrence of interactions between particles, dictated by repulsive forces (i.e., electrostatic forces, hydration forces, steric forces) and attractive Van der Waals forces [8,9]. The zeta potential (ZP) determines the strength of the electrical double layer (EDL) repulsion forces between particles and identifies the degree of stability of colloidal suspensions [10]. Numerous studies have focused on the possible relationship between zeta potential (ZP) and the rheological properties of various colloidal suspensions [11,12].

It has been demonstrated that higher zeta potential values correspond to a higher potential surface charge of particles [9]. Thus, an increase in ZP leads to a decrease in viscosity or other rheological properties such as shear stress [13]. The point of zero charge (IEP) is the pH value providing a zero-zeta potential value [14]. Understanding the effect of pH is therefore very important for analyzing the electrokinetic effect on surface charge distribution. Apparently, well-dispersed suspensions have relatively low viscosities [11,15].

Generally, when finely divided solids are dispersed in water, the particles tend to form agglomerates. This phenomenon necessitates the addition of mineral or polymeric dispersing agents to prevent the formation of these flocs [16,17]. Polymers can be strongly adsorbed on the particle surfaces. They prevent flocculation at high electrolyte concentrations and ensure the stability of the suspension by allowing homogeneous dispersions, both at high solid content and low viscosity [18]. Stabilization is possible by electrostatic and steric processes, which obstruct particle approach by, respectively, applying the electrostatic repulsive force and creating obstruction by the adsorption of polymeric molecules on the particle surfaces.

In our country (Algeria), several kaolin mines have been exploited (Djebel Debbagh, Tamazert, etc.). Despite the availability of these raw materials, local needs in the ceramic industry are entirely met through imports. With this in mind, we deemed it useful to study the rheological behavior of Kaolin DD1, often neglected in the literature, to explore its potential to meet local raw material needs for the ceramic industry and promote its use in various industrial applications in Algeria. It is pertinent to note that Kolli et. al have previously studied the rheological behavior of Kaolin DD3 using the dispersant Darvan C [19]. This study highlighted the specific characteristics of Kaolin DD3. However, our work focuses on Kaolin DD1, using two types of dispersants (Hypermer KD1 and Darvan 7) to address the gaps in previous studies and to better understand the rheological properties of this specific grade of kaolin. To enhance the efficiency and performance of KD1, we added aluminum slags as a source of alumina to the kaolin suspensions. The addition of this alumina can improve KD1’s properties, such as its chemical reactivity, by providing an additional source of this essential component. This incorporation can optimize the rheological behavior of suspensions and strengthen interactions between particles. Our choice fell on aluminum slag in order to valorize it in ceramic applications due to its classification as toxic waste, and its disposal in landfills is prohibited in many countries [20]. Several studies have valorized this waste to eliminate its toxicity. It has been directly applied as raw material in refractory bricks [21] or in the production of inert filler for construction, road paving, mortar components, polymer composites, adsorbents [22], and cement [23]. Other researchers have valorized aluminum dross in the manufacture of raw materials such as hydrogen gas [22], zeolites X [24], biomaterials, and alumina [25].

2. Materials and Methods

Local natural kaolin, denoted DD1, was used in this study. This kaolin was extracted from Djebel Debbagh near Guelma (north-east of Algeria). The mineralogical composition of the kaolin is given in

Table 1. Mineralogical composition of kaolin DD1 and aluminum slag.

Oxides	SiO2	Al2O3	Fe2O3	CaO	MgO	TiO2	Na2O	K2O
Kaolin wt.%	44.9	37.49	0.12	0.26	0.13	0.1	0.19	0.01
Aluminum slag wt.%	2.5	87	0.15	0.12	0.21	0.2	0.2	0.1

. Its absolute density, measured with a helium pycnometer, is 2.63 g/cm³. The particle size distribution of the powders was characterized with a Malvern Mastersizer Microplus (Figure 1). The mineralogy of the kaolin powder used was studied using X-ray diffraction analysis. The observed spectra (Figure 2) reveal that the kaolin contains, essentially, kaolinite (Index JCPDS 80-0886) and halloysite (JCPDS 29-1489). The kaolin powder observed by SEM shows a structure composed of agglomerates of elongated sheets oriented in all directions (Figure 3) [26]. The aluminum slag used, a waste product from the aluminum industry, was provided by ALGAL company (M’Sila, Algeria). It is composed of 87% alumina (in mass) and is almost entirely converted to α-alumina after thermal treatment at 1400 °C (Figure 4). Its absolute density is 3.41 g/cm³. Aluminum slag is used as a source of alumina to enhance the efficiency and performance of KD1. The mineralogical composition of the kaolin is given in Table 1.

In this study, two dispersants (Hypermer KD1 and Darvan 7) were used to study the rheological behavior of kaolin slurry in order to choose the one most suitable for having a minimum viscosity. According to the technical characteristics given by the suppliers, Hypermer (KD1) is a cationic polymer. It is a polyester/polyamine polymer. It is made of an anchoring group, which is adsorbed on the particle surface, and a polymeric chain with a chemical shape designed to provide ideal steric stabilization of the dispersion.

Darvan 7 (DV7) is an anionic dispersant based on sodium polyacrylate, with the formula [–CH₂–CH(COONa)–]. It is a dispersing agent that has a high-molecular-weight, long-chain polymer that has been used successfully as a general-purpose dispersing agent for ceramics. It is used to lower the viscosity of particulate slurries and to inhibit the settling of particulate solids.

The kaolin slurries consisted of 20 wt.% kaolin and 80 wt.% distilled water and dispersants, which were added to the slurry in amounts of 0.25, 0.5, 0.75, 1, 1.5, and 2% by weight, calculated on kaolin dry mass. These suspensions were continuously mixed with a stirrer at 500 rpm for 5 h to obtain a homogeneous and steady slurry and to study the rheological behavior of kaolin slip without any additions. Slurries without dispersant and other suspensions were prepared in the same conditions. The rheological behavior of the prepared kaolin suspensions was conducted at neutral pH.

Steady shear measurements of the suspensions were carried out using a Haake VT 501 type rheometer, consisting of two coaxial cylinders rotating relative to each other at a constant temperature of 23 °C. This type of rheometer is particularly suitable for rapid and comparative measurements of liquids and slips. Before beginning the measurements, pre-shearing was carried out at a high shear charge of 100 s⁻¹ for 1 min to transmit the equal rheological records to the complete suspension. Due to the rapid sedimentation of kaolin, after several measurement cycles, we adopted the following cycle: the shear rate was raised from 0 s⁻¹ to 100 s⁻¹ and then decreased back to 0 s⁻¹. The time of each step was 1 min. Furthermore, tests for the time dependence of viscosity were performed between the shear rate increasing and decreasing steps. The electrophoretic mobility of kaolin powders dispersed in deionized water with a mass concentration of 2%, as a function of pH, was measured using Nano Partica SZ-100 nanoparticle apparatus (HORIBA Ltd, Kyoto, Japon). To adjust the pH of the diluted solution, we added 1N HCl and 1N NaOH.

The presence of specific atomic groupings and characteristic vibrations of chemical bands was analyzed by Fourier transform infrared (FTIR) spectroscopy using a Perkin-Elmer (PerkinElmer Inc, Waltham, MA, USA) 700-type apparatus operating in the vibration range between 400 cm⁻¹ and 4000 cm⁻¹. The principle involves analyzing translucent pellets obtained by mixing 99% mg of finely ground kaolin with 0.1% mg of potassium bromide (KBr).

3. Results

The FTIR spectra of the kaolin are shown in Figure 5. In the range 4000–3000 cm⁻¹, the spectra are characterized by two bands located at 3697 cm⁻¹ and 3684 cm⁻¹. These bands are due to stretching vibration of the hydroxyls. The presence of stretching vibrations of the Si–O–Si bonds [27,28] is shown by the bands positioned, respectively, at 1983 cm⁻¹ and 1876 cm⁻¹. The band located at 1652 cm⁻¹ is attributed to the vibrations of the deformation of the H–O–H valence bonds of water molecules [29,30]. The bands located at 923 cm⁻¹ and 868 cm⁻¹ are attributed to the vibrations of the Al–OH bonds. The Si–O–Al^VI and Si–O–Al^IV stretching vibrations are, respectively, characterized by bands at 595 cm⁻¹ and 770 cm⁻¹ [24,31]. The deformation of Si–O links was characterized by the presence of a band at 456 cm⁻¹. The band associated with the stretching of Al–O–Al is located at 595 cm⁻¹ [26,31].

The electrophoretic mobility of the natural kaolin particles dispersed in distilled water at different pH values is shown in Figure 6. In the measurements of the zeta potential of kaolin as a function of pH, HCl and NaOH were used as acid and alkaline media, respectively. In this case, we used distilled water because a saline environment reduces the electrostatic repulsion between particles, generating more bonds due to attractive Van der Waals interactions and eventually cation bridges due to cations in the solution, whereas these interactions are weak in distilled water [32]. Zeta potential measurements were carried out between a pH of 2 and 14 after stabilization of the pH values.

As shown in the graph of zeta potential as a function of pH, the isoelectric point (IEP), or point of zero charge of the kaolin particles, was about pH = 5.5. The pH value of the isoelectric point (IEP) that we found in our study is consistent with the results obtained by other researchers. For example, Rubio determined an IEP pH of 6.8 [33], Ersoy found an IEP pH of 2.5 for Ukrainian kaolin [10], and Meltan reported an IEP pH of 7.8 [34]. The value we obtained thus falls within the pH range observed in these studies. It is important to note that the isoelectric point value can vary depending on the pretreatment and prior history of the clay samples [35], which explains the differences between these results. The isoelectric point is designated as the pH where the zetapotential is zero [36]. At pH ≈ 5.5, the positively charged sites are equal to the negatively charged sites on the natural kaolin particles (DD1). The electrophoretic mobility was negative in the pH range 2–13. From pH = 4.5 to pH = 8.5, the magnitude of the negative zetapotential increased rather sharply. Teh et al. [2] show that the degree of deprotonation of the edge positive charge sites increases in this pH range. The magnitude of the negative zetapotential reached a minimum value at pH = 12.5. Indeed, the decrease in the negativity of the zeta potential with increasing pH could be explained by the adsorption of OH⁻ ions on polar groups such as silanol (Si–OH) and aluminol (Al–OH) at the edges of the kaolinite layer, or by the ionization of these groups and the release of H⁺ ions into the water at low pH [37]. However, for a low pH, the concentration of H⁺ ions is increased by adding HCl. The adsorption of H⁺ ions on kaolin particles will compress the diffuse electrical double layer, resulting in lower zetapotential values, whereas the addition of NaOH increases the concentration of OH⁻ ions in the system, resulting in an increase in pH [38]. The adsorption of OH⁻ ions on the kaolinite–water interface results in a large diffuse double layer having a higher zetapotential value [39,40]. Therefore, our obtained results are in agreement with the diffuse electric double layer theory. At the isoelectric point IEP, the surface charge is zero, thereby reducing the electrostatic repulsion between the particles to a minimum, and consequently, the particles in the suspension of kaolin can be easily agglomerated [41].

To characterize and compare the rheological behavior of suspensions with different dispersants at various concentrations, we measured the viscosity at a constant shear rate of 50 s⁻¹. Understanding the interaction between kaolin and the dispersing agent is crucial for optimizing flocculant performance [42]. Figure 7 shows that the viscosity was highly sensitive to variations in dispersant concentration. We observed that suspensions containing insufficient or excessive amounts of dispersant exhibited relatively higher viscosity compared with those with an adequate amount, due to insufficient surface coverage and flocculation by bridging between polymer species, respectively [43].

Thus, the viscosity of the kaolin dispersion decreases as the dispersant concentration increases and reaches a minimum value at a concentration of 1% by weight for both dispersants. Above 1% by weight of dispersant, the viscosity of the kaolin increases, indicating that the adsorption of the dispersing agent on the kaolin particles is saturated. This increase in viscosity can be explained by the fact that at a higher dispersant concentration, the particle network becomes stronger, which slightly increases the viscosity [12,44]. Additionally, the optimal viscosity obtained with the addition of 1% by weight of Darvan 7 (DV7) is half of that obtained with the addition of 1% by weight of Hypermer KD1. Ammonium polymethacrylic acid (Darvan 7) and Hypermer KD1 are both electro-steric dispersants. For the steric effect, the polymer chains of Hypermer KD1 and Darvan 7 adsorbed on the kaolin particles create a physical barrier, preventing the particles from approaching each other sufficiently to agglomerate. All studies conducted on the adsorption of polymers on kaolin conclude that polymer adsorption occurs primarily on the edge surface of kaolin (i.e., on the broken bonds of aluminol (Al–OH) and silanol (Si–OH) groups) via hydrogen bonding [42,45,46].

KD1 is a cationic dispersant based on polyester/polyamine with amine groups that can be protonated in acidic medium to give positively charged ammonium groups (R₄N⁺). These groups are crucial for the electrostatic interaction with the silanolate sites (Si–O⁻) of kaolin. However, in our case, where the studied suspensions have a neutral pH, the electrostatic action mechanism of KD1 could be compromised. Consequently, the main interaction between Hypermer KD1 and kaolin is generally steric. Under these conditions, the effectiveness of KD1 in dispersing negatively charged kaolin particles could be reduced. Darvan 7 is an anionic dispersant based on sodium polyacrylate with the formula [–CH₂–CH (COONa)–]. In solution, the carboxylate group (–COONa) dissociates to release a negatively charged carboxylate ion (–COO⁻) and a sodium cation (Na⁺). The carboxylate ions (–COO⁻) then bind to the surface of the kaolin particles, imparting a negative charge to them. This negative charge brought by the carboxylate ions leads to electrostatic repulsion between the kaolin particles. These electrostatic attractions between the dispersant and kaolin promote the adsorption mechanism [47], thus preventing aggregation and favoring a homogeneous dispersion in the solution.

It is noted that the superior effectiveness of Darvan 7 compared with KD1 in our case is attributed to its electro-steric effect, while KD1 is characterized by a pure steric effect. Moreover, the solubility of Darvan 7 is higher in water compared with that of KD1, giving it the advantage of adsorbing more quickly to the particle surface, forming an adsorbed layer thick enough to prevent particles from approaching too closely. In contrast, KD1 adsorbs better in solvents of medium polarity such as MEK (methyl ethyl ketone) [48].

At neutral pH, the negative charges of kaolin particles can be partially neutralized by the ions present in water, which reduces the electrostatic repulsion between particles. This decrease in electrostatic repulsion allows Van der Waals forces and hydrogen bonds to dominate. These interactions strengthen interparticle bonds and form a three-dimensional flocculated structure where particles are connected randomly but coherently, thus giving the suspension an initial rigidity. This structure requires significant energy to break, resulting in a high yield point. The increase in yield point observed in kaolin suspensions without dispersant (Figure 8) is a direct consequence of this three-dimensional flocculated organization. This behavior, typical of Casson fluids, requires a high shear threshold to break the bonds between particles and initiate flow. Due to this structure, particles strongly resist initial deformation, leading to particularly high viscosity at low shear rates (Figure 9), which reflects the increased resistance and stability of the flocculated structure.

The introduction of a dispersing agent significantly affects the viscosity of suspensions, resulting in a remarkable decrease in viscosity at lower shear rates, and leads to a reduction in shear stress. This is caused by the face-to-face orientation of the kaolin particles in the presence of the dispersing agent [49]. The behavior of kaolin suspensions with a quantity of dispersant varies from 0.5 to 2 wt.% and follows the law of the Bingham model. Shear stresses decrease significantly for a quantity of 1 wt.% of dispersant. An addition of 1 wt.% of DV7 switches the suspension to a perfectly plastic behavior. The optimum dosage of Darvan 7 to attain a good dispersion allows a decrease in the flow point value to 2 Pa. On the other hand, the addition of 1 wt.% by weight of KD1 decreases the flow point value to 4 Pa. The optimal quantity of dispersant to have the lowest viscosity is 1 wt.% Darvan 7 and Hypermer. The value of the viscosity is equal to 0.06 Pa.s for DV7 and twice this value for KD1. However, with an optimal amount of dispersant, the suspensions become ideal. We note that Darvan 7 showed the lowest viscosity in both cases, indicating the best dispersant of the examined two commercial ones.

This difference in yield point and viscosity can be explained by the dispersion mechanisms of the two agents. Darvan 7 provides both electrostatic and steric dispersion effects [50], where the anchoring groups, e.g., the hydroxyl groups of Darvan 7, can be more easily adsorbed onto the surface of the particles, resulting in spatial steric resistance between the particles [51]. These two mechanisms allow for a more significant reduction in yield point and viscosity. In comparison, although KD1 offers effective dispersion, its steric mechanism alone is not as powerful. On the other hand, the rapid interaction of Darvan 7 with kaolin particles leads to a large amount of specific adsorption at the kaolin/water interface [41].

The shear stress of kaolin suspensions without dispersant is high, indicating that the kaolin particles are highly flocculated, which requires a significant force to flow (see Figure 10). The addition of the dispersant reduces interparticle interactions and cohesive forces, which facilitates movement and decreases the shear stress. The shear stress reaches a minimum value with a dispersant concentration of 1 wt.%, at which concentration the dispersants are sufficiently present to maximize particle dispersion. When the dispersant concentration exceeds 1 wt.%, the suspension structure undergoes a substantial change, transitioning from a weakly flocculated state to a more strongly flocculated structure [52]. This structural change is associated with an increase in viscosity across the shear rate range studied. The increase in the yield stress at a dispersant concentration of 2 wt.% indicates that the suspension has flocculated to form an interconnected particulate network, requiring significant energy to disrupt this network [52]. This observation is consistent with the literature on particle dispersions, which indicates that dispersants improve fluidity by reducing flocculation forces, but excessive concentrations can lead to more complex and flow-resistant structures [1].

In order to study the rheological behavior of kaolin suspensions as a function of rest time, we measured the variation in viscosity and shear stress after 24 h. These suspensions were agitated in order to avoid sedimentation of the kaolin. Figure 11 shows that the viscosity increases as the rest time increases. This is certainly due to increased particle–particle interactions. In addition, the shearing stresses increase remarkably with increasing rest time. It is believed that with increasing resting time, water is adsorbed between the kaolin sheets, hence increasing the hydroxyl bonds, which are bonds with more energy [53]. This adsorption leads to an increase in hydrogen bonds formed between the adsorbed water molecules and the hydroxyl groups present on the surfaces of the kaolin sheets. These hydrogen bonds are relatively strong interactions [53] that form a network of hydrogen bridges and increase the interlayer spacing, causing the particles to swell. As a result, the particle network becomes more solid and, consequently, the attraction between the particles increases, thus increasing the viscosity and shear stresses [54,55,56].

To increase the efficiency of KD1, we added 10 wt.% aluminum slag to the kaolin suspensions composed of 20 wt.% kaolin and 80 wt.% distilled water and dispersant (KD1), which was added to the suspension in amounts of 0.5, 1, and 2% by weight, calculated on kaolin dry mass.

Figure 12 shows that the yield point for an amount of 1% by weight of KD1 is 2.6 Pa. This point is lower than that obtained with the use of kaolin alone. Thus, the viscosity decreases with the amount of dispersant added. Additionally, the addition of aluminum slag significantly contributes to the reduction in viscosity, as the partial dissolution of aluminum slag in the kaolin suspension containing KD1 generates Al³⁺ cations [35]. These cations, by adsorbing onto the anionic sites of the kaolinite layers, induce partial neutralization of the surface charges. This phenomenon causes a compression of the electric double layer, reducing the interparticle electrostatic repulsion forces. The consequent reduction in colloidal interactions promotes the particles approach and decreases the flow resistance [57]. This modification of interfacial properties results in a notable decrease in the viscosity of the kaolin suspensions.

In parallel, under the applied shear, the Al₂O₃ particles present in the aluminum slag reorganize into ordered structures in response to the applied stress below a critical shear level, forming compact layers with intercalated liquid films (between the particles/layers) and exhibiting tightly packed particle configurations within the layers. This further reduces colloidal interactions and, consequently, the viscosity of the suspension [52]. Thus, the addition of aluminum slag plays a crucial role not only in neutralizing surface charges but also in reorganizing the particles under shear, leading to a pronounced decrease in the viscosity of the kaolin suspensions.

4. Conclusions

The rheological results showed that the initial behavior of Algerian kaolin suspensions (without dispersant) follows the Casson model but switches to the Bingham model with an optimal amount of dispersant. The suspensions with 1 wt.% of dispersant conform to the Bingham model with a very low yield point. The value of this point with Hypermer KD1 increases to 4 Pa; however, this value decreases to 2 Pa with Darvan 7, and the viscosity of the kaolin DD1 dispersion depends strongly on the dispersing agent concentration.

This comparative study clearly shows that although the optimal amount of dispersant that results in minimum viscosity is the same (1 wt.%) for both dispersants, their effectiveness differs. We note that the minimum viscosity obtained with Darvan 7 is lower than that with the KD1 dispersant. The suspensions exhibit marked yield stresses that decrease with the addition of dispersant, reaching a minimum value at a dispersant concentration of 1% by weight. Beyond this concentration, the yield stress increases.

For Darvan 7, at neutral pH, the steric effect arises from its molecular configuration, which forms a physical barrier on the surface of the kaolin particles, thus preventing their aggregation and promoting their dispersion in the suspension. Additionally, there is an effect resulting from electrostatic interactions between the carboxylate groups (–COO⁻) of Darvan 7 and the negatively charged surface sites of kaolinite, facilitating its adsorption. In contrast, for KD1, at neutral pH, the electrostatic action mechanism is compromised. Consequently, the main interaction between Hypermer KD1 and kaolin is generally steric, which reduces its effectiveness. Additionally, the solubility of Darvan 7 in water is higher compared with that of KD1, which makes the reaction of kaolin particles with Darvan 7 faster than that with KD1, resulting in a larger quantity of Darvan 7 being adsorbed on the kaolin, leading to a significant reduction in rheological parameters. We note that Darvan 7 is the best dispersant for stabilizing Algerian kaolin slurries. The incorporation of aluminum slag as a source of alumina in the kaolin suspension with KD1 leads to a significant decrease in viscosity and shear stress. For a concentration of 1% KD1, the yield point is reduced to 2.6 Pa, which reduces colloidal interactions and decreases flow resistance.