Probing acid/base chemistry and adsorption mechanisms of hydrolysable Al(III) species with a clay system in aqueous solution

Abstract

The acid/base chemistry of hydrolysable Al(III) species with a clay (bentonite and kaolin) system was investigated at 35 °C in the expanded solution pH range from 1 to 9. The adsorption capacity (q_e) and mechanism of Al(III) species on clays were examined by means of UV-Vis spectra, optical microscope, zeta potential testing, SEM, XPS and XRD analysis. The results demonstrate that the q_e of Al(III) increases with increasing solution pH, and the maximum adsorptions of bentonite and kaolin are 13.64 mg g⁻¹ and 1.06 mg g⁻¹, respectively. Each Al(III) species offers a different contribution to the total adsorption capacity. The cation Al³⁺ is the dominant species below pH 4 and the anion Al(OH)₄⁻ is the main species above pH 6. The solution pH values change in the single clay or hybrid clay/Al(III) solution because of the acid–base dissolution and competitive adsorption of H⁺/OH⁻ with Al(III) species. The particle sizes (d₈₀) of the clays after adsorption at different pH were first associated with thermodynamics. The zeta potential variation of clays was first connected with the total charge numbers of Al(III) species in solution. Zeta potential, XPS and XRD studies indicate that charge neutralization and ion exchange dominate the adsorption process at lower pH and surface complexation and precipitation at higher pH.

---

1. Introduction

The mineral processing industry faces increasing challenges to seek sustainable strategies for their beneficiation of wastewater and hazardous tailings. To meet the requirements of environmental and sustainable development, water recycling and reuse technologies are becoming an essential part of mineral processing activities. However, the increasing use of recycled water often has negative influences on the recovery and grade of minerals in ore beneficiation, particularly in mineral flotation.^1–6 These have been traced to the accumulation of dissolved ions/compounds, clays and chemicals, which can alter the solution chemistry of a flotation system.

The wastewater from bauxite flotation after repeated cycles was analyzed, and the results show that there are some major metal cations, metal complexes and a larger number of clays. The Al³⁺ content and the suspended clay content are sometimes up to 100 mg L⁻¹ and 500 mg L⁻¹, respectively, because of fine grinding of low-grade bauxite and complete disintegration of minerals. Like most hydrolysable metal ions, Al³⁺ forms specific stoichiometric compounds once the pH of the solution exceeds a compound-specific threshold.^7,8 For instance, Al in acidic solution mainly exists in the form of Al³⁺, and in alkaline solution Al easily hydrolyses to form hydroxy-Al ions. Generally, the mineral flotation is conducted in neutral or alkali conditions, and the produced wastewater is alkaline. Therefore, Al³⁺ easily hydrolyses in flotation pulp to form hydroxy-Al ions. These chemical species may inhibit the interaction between reagents and minerals through the formation of surface coatings.⁹

Interactions between hydroxy-Al and clay minerals, such as montmorillonite, kaolinite, vermiculite and oxisol, in suspension have been investigated in some studies. Walker et al.¹⁰ studied the adsorption kinetic of aluminum by aluminosilicate clay minerals in the pH range from 3.0 to 4.1. The mechanism was a simple electrostatic cation exchange between adsorbed Al and the clay surface. Vermiculite was governed by both internal ion diffusion and electrostatic attraction.¹⁰ The factors influencing Al hydrolysis have also been investigated.¹¹ Al hydrolysis increased with increasing initial Al concentration and solution pH, reaching a maximum value at pH 4.5. The solution pH change was observed after Al was adsorbed onto clays.¹² The clay minerals had much higher adsorption affinity for hydroxy-Al than Al³⁺, and the selective adsorption of hydroxy-Al promoted Al hydrolysis.¹³ However, the former studies were mainly performed at lower pH values (<7.5) and focused on the influencing factors, kinetics and solution pH change, without considering the difference in the clay structures and properties and their change after adsorption. Therefore, a study of the hydrolysis of Al in suspensions at higher pH and the effect and change of clay structures and properties would be of practical significance in mineral processing.

In this work, the adsorption of Al(III) species onto two clay minerals (bentonite and kaolin) in aqueous solution was investigated over an extended pH range from 2 to 9. The main objectives are as follows: (1) investigating the Al(III) dissolution of clays and solution chemistry changes in acid/base solution, (2) inspecting the total Al(III) adsorption capacity and the theoretical Al(III) species adsorption capacity at different pHs and examining the solution chemistry change, and (3) acquiring a further understanding of adsorption mechanisms through inspecting the variation of morphology, surface charge and microstructures of clay minerals after adsorption. This study would offer a perspective to understand the acid/base chemistry and adsorption mechanisms of hydrolysable Al(III) species and clays in the system of bauxite flotation using recycled water.

2. Experimental

2.1. Materials

The bentonite is Na-bentonite obtained from a mineral processing plant in India. The kaolin (analytical grade) was purchased from Tianjin Kermel Reagent Technologies Co., Ltd, China. Both the clay samples were further dried at 105 °C for 24 h. The specific surface area of samples was measured using a surface area tester (NOVA ST-2000, America). The specific surface areas of bentonite and kaolin were 17.77 m² g⁻¹ and 8.92 m² g⁻¹, respectively, and the pore radii D_v(r) are 18.97 nm and 19.28 nm, respectively, calculated from N₂ adsorption and desorption isotherms (Fig. S1†). The chemical composition of bentonite is shown in Table 1.

Table 1 Chemical composition of bentonite (wt%)^a

SiO2	Al2O3	CaO	MgO	K2O	Na2O	MnO	LOI
a LOI: the heat loss in N2 atmosphere.
59.46	20.63	2.62	2.74	0.13	3.31	0.12	10.99

---

Unless specifically noted, all reagents used in the experiment were of analytical grade and were used without further purification. A 1000 mg L⁻¹ stock solution of Al(III) was prepared by dissolving 13.90 g of Al(NO₃)₃·9H₂O in ultrapure water and then diluting it to 1000 mL. Working solutions of Al(III) were prepared from the stock solution by dilution with ultrapure water.

Other main reagents including 8-hydroxy-7-iodo-5-quinolinesulfonic acid (Ferron, C₉H₆INO₄S, Aladdin) and sodium acetate trihydrate (CH₃COONa, Aladdin) were used for Al(III) determination. Sodium hydroxide (NaOH) was used to adjust the pH and was purchased from Sinochem, China. Ultrapure water (18.2 MΩ cm) was generated with a Barnstead NANO Pure Diamond Water Purification System.

2.2. Characterization

Al(III) ion concentrations were tested by ferron photometric method in accordance with HG/T 3525-2011. The photometric measurements were performed at 370 nm with a model TU-1901 double-beam ultraviolet and visible spectrophotometer (UV-Vis) with ultrapure water as the reference solution.

The zeta potentials of the clays before and after adsorption of Al(III) were measured using a JS94H2 model microscopic electrophoresis apparatus (Powereach.com, China) and each test was repeated four times. The pH values of the samples were checked by acidometer PHSJ-4A. The particle sizes of the clay samples were observed and analyzed by using an optical microscope and software (ZEISS Axio Scope, A1, Germany), and d₈₀ was defined as the apparent particle size relevant to 80% of cumulative distribution. Scanning electron micrographs (SEM) were obtained using a scanning electron microscope (JOEL JSM-7500F, Japan), and the samples were sputtered with platinum before observation. The Al/Si ratio of the clay surfaces was characterized by energy dispersive X-ray spectroscopy (EDX). X-ray photoelectron spectroscopy (XPS) analyses were performed using a Kratos Axis Supra (Kratos Analytical, United Kingdom) fitted with an Al Kα monochromator source and a multidetection analyzer. The electron energy analyzer was operated with a pass energy of 20 eV enabling high resolution spectra to be obtained. A step size of 0.1 eV was employed and each peak was scanned five times. The binding energies were referenced to the C 1s line at 284.8 eV from adventitious carbon. X-ray diffractions (XRD) were obtained using a D8 Advance (Bruker) X-ray diffractometer with Cu Kα radiation (λ = 1.5418 Å). The samples were scanned at a scan rate of 0.02 s⁻¹ in a 2θ range of 4–80.

2.3. Adsorption experiments

The experiments were conducted using a batch equilibration approach at a constant temperature (35 °C). The 250 mL Erlenmeyer flasks containing the clays and metal solution were shaken for 1 hour in a thermostatic rotary shaker, operating at 150 rpm, and then centrifuged (5 min, 8000 rpm) immediately. The concentration of Al³⁺ in the supernatant was determined by UV-Vis spectrometry. The concentration of clay was 8 g L⁻¹ for bentonite and 100 g L⁻¹ for kaolin, and the initial concentration of Al(III) ions was set as 100 mg L⁻¹. The initial pH of solution was varied from 1 to 9 adjusted by 0.1 M HNO₃ or 0.1 M NaOH solutions. In addition, a blank experiment was carried out under the same conditions without adding Al(III) ions to investigate the Al(III) dissolution of clays and to eliminate its effect on the adsorption capacity of Al(III). After adsorption, the final pH of the solution was measured.

The adsorption percent (%) aluminum ions onto clays and the adsorption capacity q_e (mg g⁻¹) after equilibrium were calculated using the following equations:

where C₀ is the initial concentration of aluminum ions (mg L⁻¹), C_e is the final or equilibrium concentration of aluminum ions (mg L⁻¹), V is the volume of the aluminum solution (mL), and m is the weight of clay (g).

3. Results and discussion

3.1. Acid/base chemistry of clays

Bentonite and kaolin are aluminosilicate minerals with smectitic structures. Their major components are montmorillonite and kaolinite, respectively. They hold a permanent negative structural charge, ≡X⁻ on the basal surfaces, and a smaller variable charge that can be either positive or negative, ≡SOH on the mineral edges, which can release or acquire H⁺ depending on the solution pH, as displayed in eqn (3) and (4).^14,15

≡SOH⁰ + H⁺ ↔ ≡SOH₂⁺ (3)

≡SOH⁰ ↔ ≡SO⁻ + H⁺ (4)

It is well known that some ions can be dissolved from bentonite and kaolin in acidic or basic conditions because of the reactions shown in the eqn (5)–(7). In order to correct the adsorption capacities of Al(III) species onto clay minerals, the leaching amounts of Al³⁺, q_Al, in a blank adsorption experiment were measured, and the change of solution properties was examined, as displayed in Fig. 1.

Al₂O₃ + 6HNO₃ = 2Al(NO₃)₃ + 3H₂O (5)

Al₂O₃ + 2NaOH = 2NaAlO₂ + H₂O (6)

SiO₂ + 2NaOH = Na₂SiO₃ + H₂O (7)

Fig. 1 (a) Al(III) dissolution amounts and (b) pH change at different initial pHs.

At pH = 1, the q_Al reaches the maximum of 1.57 mg g⁻¹ and 1.05 mg g⁻¹ for bentonite and kaolin, respectively. The q_Al for kaolin is low, approximately 0.06 mg g⁻¹ in the pH range of 2–9. Additionally, the dissolution reactions of bentonite and kaolin caused a change in solution pH (Fig. 1(b)). The final solution pH turns less acidic when the initial pH is below 7, owing to the consumption of H⁺ by reacting with the clays (eqn (5)). On the other hand, the final pH turns less basic when the initial pH is above 7, owing to the OH⁻ ions reacting with the clays (eqn (6) and (7)). The change of solution pH could also be ascribed to the adsorption of H⁺ or OH⁻ onto edge sites and basal surface sites of clays.¹⁶

3.2. Acid/base chemistry and adsorption of Al³⁺ species

3.2.1. Theoretical calculation of Al³⁺ hydrolysis. In the case of trivalent Al³⁺ ions, it is known that there is a sequence of hydrolysis equilibria with increasing pH, which can be written as follows:⁸

Al³⁺ → Al(OH)²⁺ → Al(OH)₂⁺ → Al(OH)₃ → Al(OH)₄⁻ (8)

Each step in the hydrolysis process has an appropriate equilibrium constant. Further, hydrolysis causes the release of a hydrogen ion into solution. High pH values promote dissociation and vice versa. Furthermore, as each proton is released, the decreasing positive charge makes further dissociation more difficult. The equilibrium parameters of Al³⁺ ions (Table S1†) in the hydrolysis process were calculated by a chemical reaction and equilibrium software package, HSC Chemistry 6.0, which includes an extensive thermodynamic database. The mole fraction (f_i) of hydrolyzed Al(III) species relative to total soluble metal concentration at 35 °C is depicted in Fig. 2. Herein, only monomeric hydrolysis products are shown. It is apparent that the cationic Al³⁺ ions are the absolutely dominant dissolved species below pH 4 and the anion Al(OH)₄⁻ are the principle species when the solution pH is above 6. At pH 4–6, the species include Al³⁺, Al(OH)²⁺, Al(OH)₂⁺ and Al(OH)₄⁻.

Fig. 2 Mole fraction of hydrolyzed Al(III) species relative to total soluble metal concentration at 35 °C.

3.2.2. Adsorption of Al³⁺ species on clays. Solution pH is an important factor in the adsorption process, which affects the surface charge of clays and the degree of ionization of Al(III).^17,18 Fig. 3 shows the effect of initial pH on adsorption capacity and adsorption percentage of Al(III) onto clays, which was corrected for the dissolution of Al(III) from bentonite and kaolin. At lower pH, the absorption of Al(III) increases rapidly with increasing pH. However, the absorption has little change in the pH range of 5–9. Similar adsorption trends were also found in other reports.^19–21 This could be attributed to the formation of a Al³⁺ hydrate layer coating on the surface, which decreases the adsorption of Al(III) onto clays. Therefore, the adsorption at higher pH values also involves the formation of metal complexes (Al(OH)²⁺, Al(OH)₂⁺, Al(OH)₄⁻, etc.) and hydroxide precipitation (Al(OH)₃). At all pH conditions, bentonite has a higher adsorption capacity than that of kaolin, which is likely due to the bentonite used in this study having a larger specific surface area than kaolin.

Fig. 3 (a) Total adsorption capacity and (b) adsorption percentage of Al(III) at different pHs.

In order to investigate the contribution of each Al(III) species to the total adsorption capacity at different pHs, the adsorption capacity of each Al(III) species, q_ei, was calculated using eqn (9), in which f_i is the mole fraction of each Al(III) species. It is necessary to note that the q_ei is theoretical adsorption capacity since the proportions of each Al(III) species in solution might change due to the preferential adsorption of certain species.

q_ei = q_e × f_i (9)

As shown in Fig. 4, the adsorbed species is mainly Al³⁺ over the pH range 1–4 (eqn (10) and (11)), while over the pH range 6–9 the adsorbed species is principally the anion species Al(OH)₄⁻. In the pH range 4–6, the adsorption includes Al³⁺, Al(OH)²⁺, Al(OH)₂⁺, Al(OH)₃ and Al(OH)₄⁻.

3≡X⁻ + Al³⁺ ↔ ≡X₃⁻·Al³⁺ (10)

3≡X⁻·Na⁺ + Al³⁺ ↔ ≡X₃⁻·Al³⁺ + 3Na⁺ (11)

Fig. 4 Adsorption capacity of each Al(III) species on (a) bentonite and (b) kaolin at different pHs.

The effects of Al(III) adsorption on solution pH are shown in Fig. 5. Similar to the blank experiment, the solution pH after adsorption turns less acidic at initial pH < 7 and turns less basic at initial pH > 7. This could be attributed to the consumption of H⁺ or OH⁻ in the dissolution reaction of clays and the adsorption reaction of H⁺ or OH⁻ onto clay surfaces. Significantly, the pH variation in the Al(III) adsorption experiment is smaller than that in the blank experiment. These results could be attributed to the competitive adsorption of Al³⁺ with H⁺ at low pH and Al(OH)₄⁻ with OH⁻ at high pH, which hinders the adsorption of H⁺ or OH⁻ onto clay surfaces. However, for the Al(III)–kaolin system, the solution pH after adsorption of Al(III) is still lower than that in the blank experiment at initial pH > 7. This should be ascribed to the higher adsorption percentage of kaolin, which promotes the hydrolysis of Al(III), that is, increases the consumption of OH⁻.

Fig. 5 pH change after Al(III) adsorption on (a) bentonite and (b) kaolin.

3.3. Effect of particle size and morphology of the clays on adsorption of Al³⁺ species

3.3.1. Particle size. The effect of clay particle size on the adsorption of metal ions has been investigated in former studies.^22,23 In order to further explore the adsorption behavior, the effect of the change in clay particle size after adsorbing Al³⁺ species was examined in detail. Fig. 6 shows the microphotographs of clay particles after adsorption. Apparently, the particle size of bentonite decreases while that of kaolin increases with increasing pH. In order to display more clearly the particle size variation of clays after adsorption, the d₈₀ of the clays was obtained from the cumulative distribution (Fig. S2†) and the results are listed in Table 2.

Fig. 6 Microphotograph of clay particles of (a–d) bentonite and (e–h) kaolin after adsorption at varied pH.

Table 2 Particle size of clay samples at different pHs (μm)^a

Samples	0	1	2	3	4	5	6	7	8	9
a Note: the clays before adsorption are denoted by “0”, and the clays after adsorption at pH 1–9 are respectively labeled as “1–9”.
d80-Bentonite	2.34	8.33	7.83	7.83	8.35	6.81	7.83	7.54	4.80	4.91
d80-Kaolin	5.60	5.94	5.78	7.24	8.00	8.10	13.60	10.24	11.05	8.22

---

Clay particles can be stabilized or aggregated by surface charge/electrostatic forces and van der Waals forces based on the DLVO theory.²⁴ Theoretically, the particle sizes of bentonite and kaolin would become larger with an increase in adsorbed Al(III) as the surface negative charges are partly neutralized and the electrostatic forces between clay particles turn weaker. However, the actual situation is relatively complex. To explore the reason for the particle size change, we correlated the d₈₀ of clays with the adsorption capacity q_e. As illustrated in Fig. 7, the d₈₀ values of bentonite and kaolin exhibit different changing trends with increasing q_e. These results could be explained by the thermodynamics of adsorption processes.

Fig. 7 d₈₀ of (a) bentonite and (b) kaolin after adsorption as a function of q_e.

The thermodynamic parameters for the Al(III) adsorption process, enthalpy (ΔH, kJ mol⁻¹) and entropy (ΔS, kJ K⁻¹ mol⁻¹), were calculated from the temperature-dependent adsorption isotherms. The isotherm adsorption experiment was performed with initial Al³⁺ concentrations of 10–100 mg L⁻¹ at three different temperatures, 25, 35 and 45 °C (see ESI†). The endothermic adsorption process of bentonite (ΔH = 32.86 kJ mol⁻¹) arouses an increase in the entropy (ΔS = 0.10 kJ K⁻¹ mol⁻¹) and further results in increased repulsive forces between particles. Therefore, the d₈₀ values of bentonite decrease with increasing q_e. Relatively, the exothermic adsorption process of kaolin (ΔH = −42.93 kJ mol⁻¹) causes a decrease in the entropy (ΔS = −0.15 kJ K⁻¹ mol⁻¹) and finally leads to a decrease of repulsive forces between particles. Therefore, the d₈₀ values of kaolin increase with increasing q_e.

3.3.2. Surface morphology. The SEM images of bentonite and kaolin before and after adsorption at pH = 4 were also investigated and presented in Fig. 8. It is apparently seen from Fig. 8(a) and (c) that the schistose structure of bentonite is tightly and evenly stacked. The channels and spaces among these lamellae make the metal adsorption feasible on the inner and outer surfaces. Comparatively, kaolin has smaller and broken lamellae, which results in the lower Al(III) adsorption capacity. After Al(III) adsorption, the surface morphology of the two clays changed greatly, with more loose lamellae, curly edges and partially exfoliated slices, as observed in Fig. 8(b) and (d). These phenomena probably occur because the Al(III) adsorption arouses a change of layer interaction.

Fig. 8 SEM images of (a) bentonite, (b) bentonite–Al(III), (c) kaolin and (d) kaolin–Al(III).

Fig. 9 shows the content of Al and Si on the surfaces of bentonite and kaolin before and after adsorbing Al(III). As shown in Fig. 9(a) and (b), the Al/Si ratio of bentonite increases from 0.50 to 0.56. The Al/Si ratio of kaolin increases from 1.23 to 1.30, as shown in Fig. 9(c) and (d). These results declare the successful adsorption of Al(III) onto the clay surfaces.

Fig. 9 EDX spectra of (a) bentonite, (b) bentonite–Al(III), (c) kaolin and (d) kaolin–Al(III).

3.4. Further exploring adsorption mechanisms

3.4.1. Charge neutralization study. Charge neutralization occurs when the charged surface adsorbs the oppositely charged ions or species.⁸ Since zeta potential reflects the surface charge of a solid particle and its interactions with the surrounding medium,^25,26 it is significant to study the effect of pH on the zeta potential of clays before and after Al(III) adsorption. As seen from Fig. 10, the zeta potentials of bentonite and kaolin before adsorption (ζ₀) are strongly pH dependent and become more negative with increasing pH. Both clays have no iso-electric point (IEP). Bentonite exhibits a higher negative potential than kaolin over the whole pH range, which is one of the reasons that bentonite has a higher Al(III) adsorption capacity than kaolin. Combined with the poor Al(III) adsorption at lower pH in Fig. 3, the less negative potential of clay results in a decrease in the electrostatic attraction between Al(III) and the clay surface. Furthermore, this process involves competition between H⁺ ions and Al(III) for the surface sites on clay minerals at lower pH.²⁷

Fig. 10 Zeta potentials of (a) bentonite and (b) kaolin before and after adsorption at different pHs.

The zeta potentials (ζ_i) of the two clays after Al(III) adsorption are also depicted in Fig. 11. It can be seen that the zeta potential increases after adsorption, and it exhibits charge reversals at pH 1–3 for bentonite and at pH 2 for kaolin. The charge reversal occurs owing to the specific adsorption of the positive Al(III) species Al³⁺ onto the clay surfaces.²⁶ The zeta potential variation of the two clays after adsorption, Δζ, was calculated using eqn (12), and the results can be found in Fig. 11.

Δζ = ζ₀ − ζ_i (12)

where ζ₀ and ζ_i are the zeta potential of clay before and after adsorption, respectively.

Fig. 11 Zeta potential variation of bentonite and kaolin.

It is noteworthy that the variations of zeta potential (Δζ) for bentonite and kaolin decrease with pH increase, though the positive Al(III) adsorption capacity increased with solution pH. In fact, the variation of zeta potential is related to the total charge number of Al(III) species (q_e, e) calculated by the following equation:

where z_i is the charge of each hydrolysed Al(III) species, including Al³⁺, Al(OH)²⁺, Al(OH)₂⁺ and Al(OH)₄⁻. The calculated q_e at varied pH can be found in Fig. 12.

Fig. 12 Total charge number of Al(III) species at different pHs.

The total charge numbers of Al(III) species in solution decreased with increasing solution pH, which may be the reason for the decline of Δζ with solution pH. The q_e of Al(III) species in solution are positive in the pH range of 2–5, and thus the electrostatic attraction governs the adsorption process in this pH range. The total Al(III) species are negatively charged in the pH range of 6–9, hence the electrostatic interactions between clay and Al(III) species are excluded; apparently, the complexation of hydrolysed Al(III) species onto clay surfaces dominates the system. It is noteworthy that bentonite has a small zeta potential variation at pH 2. The main reason probably lies in cation exchange in the interlayer not causing the large increase in surface charge at this pH value. It could be concluded that electrostatic attraction and surface complexation are the mechanisms of adsorption of Al(III) onto clays in the pH range 1–9.

3.4.2. Structure changing of clays. To further investigate the adsorption mechanism of Al(III) species onto clay, the crystal structures of two clays after adsorption at varied pH were examined by XRD. The change of basal spacing can reflect the adsorption reaction occurring in the interlayers of the clay.^6,28 Based on the XRD patterns of the two clays before and after Al(III) adsorption (Fig. S3†), the basal spacings, d(001), of each sample are listed in Table 3.

Table 3 Basal spacing of two clay samples at different pHs (Å)^a

Samples	0	3	4	6	8	9
a Note: the clays before adsorption are denoted by “0”, and the clays after adsorption at pH 1–9 are respectively labeled as “1–9”.
d(001)-Bentonite	12.34	15.46	15.40	15.52	16.06	15.56
d(001)-Kaolin	6.99	7.07	7.14	7.10	7.07	7.09

---

As displayed in Table 3, the basal spacing presents a different increase for two clays after adsorption. For bentonite, the basal spacing gradually increases from the initial value of 12.34 Å to 15.56 Å after adsorption. These could be attributed to the fact that hydrolysed Al³⁺ species has a bigger ion diameter (4.75 Å) than interstratified Na⁺ (0.95 Å),²⁹ which were adsorbed into the interlayer of bentonite or ion-exchanged with the interstratified Na⁺ ions. Comparatively, kaolin has narrower basal layers than bentonite. It is difficult for Al³⁺ species to enter the kaolin interlayers. Therefore, the basal spacing of kaolin after Al(III) adsorption only has minor increase, from 6.99 Å to 7.14 Å.

3.4.3. XPS investigation. In order to obtain molecular level information for Al(III) adsorption onto clays at different solution pHs, the chemical binding of Al(III) on the clay surface was characterised by XPS. The high resolution XPS Al 2p spectra are shown in Fig. 13. It can be seen that the aluminum peak is shifted slightly as the pH increases. There are three main separation peaks at pH 4 and two separation peaks at pH 6, while there are multiple separation peaks at pH 5. The XPS binding energies obtained from the Al 2p core levels are listed in Table 4.

Fig. 13 Al 2p XPS spectra of (a, c and e) bentonite and (b, d and f) kaolin after Al(III) adsorption at different pHs.

Table 4 The binding energy of Al 2p in clay samples at different pHs

pH	Binding energy
	Bentonite–Al(III)	Kaolin–Al(III)
4	74.9, 74.8, 74.3	75.3, 74.9, 74.4
5	74.6, 74.5, 74.4, 74.1, 73.7	75.4, 74.8, 74.7, 74.5, 74.1
6	74.1, 73.4	74.6, 74.0

---

The XPS binding energies cannot be interpreted to identify hydrolysable Al(III) species since no reference about hydrolysable Al(III) species has been studied. Herein, the confirmation of Al(III) species adsorbed onto the clay surfaces is inferred according to the distribution of Al(III) species at different pHs (Fig. 2). In Fig. 13(a), (c) and (e), the three peaks at 74.9, 74.8 and 74.3 eV for the pH 4 bentonite sample should be attributed to Al³⁺, Al(OH)²⁺ and Al(OH)₂⁺ species, respectively, while the five peaks at 74.6, 74.5, 74.4, 74.1 and 73.7 eV for the pH 5 bentonite sample should be assigned to Al³⁺, Al(OH)²⁺, Al(OH)₂⁺, Al(OH)₃ and Al(OH)₄⁻ species, respectively. The two peaks at 74.1 and 73.4 eV for pH 6 bentonite sample should be ascribed to Al(OH)₃ and Al(OH)₄⁻ species, respectively. In order to further confirm the Al(III) species adsorbed on the clay samples, the high resolution XPS spectra and binding energies of O 1s are shown in Fig. 14 and Table 5, respectively.

Fig. 14 O 1s XPS spectra of (a, c and e) bentonite and (b, d and f) kaolin after Al(III) adsorption at different pHs.

Table 5 The binding energy of O 1s in clay samples at different pHs

pH	Binding energy
	Bentonite–Al(III)	Kaolin–Al(III)
4	533.7, 531.9	532.4, 531.8
5	532.6, 532.3, 531.6, 530.8	532.9, 532.2, 531.6, 530.5
6	533.8, 531.7	532.4, 531.8

---

In Fig. 14(a), (c) and (e), the two peaks for the pH 4 bentonite sample could be attributed to Al(OH)²⁺ and Al(OH)₂⁺ species, respectively, while the four peaks for the pH 5 bentonite sample should be assigned to Al(OH)²⁺, Al(OH)₂⁺, Al(OH)₃ and Al(OH)₄⁻ species, respectively. The two peaks for the pH 6 bentonite sample should be ascribed to Al(OH)₃ and Al(OH)₄⁻ species, respectively. The XPS spectra of kaolin samples at different pHs have similar results (Fig. 13(b), (d) and (f) and Fig. 14(b), (d), (f)). Compared to the spectra at pH 4 and pH 6, however, the quantity of the species of kaolin at pH 5 is different. The main peaks for the kaolin sample at pH 5 are attributed to Al(OH)²⁺, Al(OH)₂⁺, Al(OH)₃ and Al(OH)₄⁻ species. The peaks for kaolin samples at pH 4 and pH 6 are also mainly assigned to Al(OH)₃ and Al(OH)₄⁻ species, respectively. The differences in the binding energies for the same species on the bentonite and kaolin samples could be due to the differences in the binding sites on the clay surfaces. In conclusion, the XPS results are consistent with the above discussion.

4. Conclusions

This study of the acid/base chemistry and adsorption mechanisms of hydrolysable Al(III) species and a clay system is an important part in understanding the solution chemistry of the bauxite flotation system using recycled water. The adsorption of Al(III) species onto clay minerals in aqueous medium is strongly pH dependent and increases with solution pH. At low pH values (pH < 4), the adsorbed species are mainly Al³⁺ ions, while at higher pH values (pH > 6) principally the negative anion species Al(OH)₄⁻ are absorbed. The maximum adsorption capacities of bentonite and kaolin are 13.64 mg g⁻¹ and 1.06 mg g⁻¹, respectively, indicating that bentonite has a stronger affinity with Al(III) species than with kaolin. The Al(III) adsorption can arouse changes of the solution pH and particle sizes of clays. The solution pH turns less acidic when the initial pH is under 7, and the pH turns less basic when the initial pH is above 7. The particle size of bentonite decreases while that of kaolin decreases with increasing Al(III) adsorption capacity. In the absence of Al(III) ions, both bentonite and kaolin samples yield negative zeta potentials in the pH range of 2–9. After contacting with Al(III) solutions, the zeta potentials of the two clays increase, relating to Al(III) adsorption on the clay surfaces. The interaction mechanism of Al(III) with clays involves electrostatic attraction, ion-exchange, surface complexation and precipitation.

Acknowledgements

The authors acknowledge financial support provided by the National Natural Science Foundation (No. 51674225), the National Science Fund for Young Scholars of China (No. 51404213 and 51404214), the Development Fund for Outstanding Young Teachers of Zhengzhou University (No. 1421324065) and the University Key Research Project of Henan Province (17A450001).

References

1. S. Farrokhpay and M. Zanin, Powder Technol., 2012, 23(4), 493–497.
2. N. J. Shackleton, V. Malysiak, D. De Vaux and N. Plint, Miner. Eng., 2012, 38, 53–64.
3. W. Liu, C. J. Moran and S. Vink, Miner. Eng., 2013, 53, 91–100.
4. Y. Huang, G. Han, J. Liu and W. Wang, J. Hazard. Mater., 2016, 301, 46–55.
5. Y. Huang, W. Chai, G. Han, W. Wang, S. Yang and J. Liu, J. Hazard. Mater., 2016, 307, 318–327.
6. W. Chai, Y. Huang, S. Su, G. Han, J. Liu and Y. Cao, Physicochem. Probl. Miner. Process., 2017, 53(1), 264–278.
7. C. T. Driscoll and W. D. Schecher, Environ. Geochem. Health, 1990, 12(1–2), 28–49.
8. J. Gregory, Particles in water: properties and processes, CRC Press, New York, 2006.
9. L. M. Shengo, S. Gaydardzhiev and N. M. Kalenga, Miner. Eng., 2014, 65, 145–148.
10. W. J. Walker, C. S. Cronan and H. H. Patterson, Geochim. Cosmochim. Acta, 1988, 52(1), 55–62.
11. R. Xu, S. Xiao, J. Li, D. Tiwari and G. Ji, Pedosphere, 2007, 17(1), 90–96.
12. S. C. Hodges and L. W. Zelazny, Soil Sci. Soc. Am. J., 1983, 47(2), 206–212.
13. M. J. Philip and W. Z. Lucian, The environmental chemistry of aluminum, CRC Press, New York, 1995.
14. X. Gu, L. J. Evans and S. J. Barabash, Geochim. Cosmochim. Acta, 2010, 74(20), 5718–5728.
15. X. Gu and L. J. Evans, Geochim. Cosmochim. Acta, 2008, 72(2), 267–276.
16. E. Tertre, S. Castet, G. Berger, M. Loubet and E. Giffaut, Geochim. Cosmochim. Acta, 2006, 70(18), 4579–4599.
17. P. Srivastava, B. Singh and M. Angove, J. Colloid Interface Sci., 2005, 290(1), 28–38.
18. X. Gu and L. J. Evans, J. Colloid Interface Sci., 2007, 307(2), 317–325.
19. D. S. Eldridge, R. J. Crawford and I. H. Harding, J. Colloid Interface Sci., 2015, 454, 20–26.
20. M. J. C. Calagui, D. B. Senoro, C. Kan, J. W. L. Salvacion, C. M. Futalan and M. Wan, J. Hazard. Mater., 2014, 277, 120–126.
21. Z. Melichova and L. Hromada, Pol. J. Environ. Stud., 2013, 22(2), 457–464.
22. T. Kogure, K. Morimoto, K. Tamura, H. Sato and A. Yamagishi, Chem. Lett., 2012, 41(4), 380–382.
23. R. Donat, A. Akdogan, E. Erdem and H. Cetisli, J. Colloid Interface Sci., 2005, 286(1), 43–52.
24. B. W. Ninham, Adv. Colloid Interface Sci., 1999, 83(1–3), 1–17.
25. D. Niriella and R. P. Carnahan, J. Dispersion Sci. Technol., 2006, 27(1), 123–131.
26. A. Chakir, J. Bessiere, K. E. Kacemi and B. Marouf, J. Hazard. Mater., 2002, 95(1–2), 29–46.
27. X. Zhang, S. Lin, X. Lu and Z. Chen, Chem. Eng. J., 2010, 163(3), 243–248.
28. T. Kozaki, Y. Sato, M. Nakajima, H. Kato, S. Sato and H. Ohashi, J. Nucl. Mater., 1999, 270(1–2), 265–272.
29. E. R. Nightingale, J. Phys. Chem., 1959, 63(9), 1381–1387.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra22923a

‡ These authors as co-first author contributed equally to this work.

---