The Impact of pH on the Pore and Structural Characteristics of Acid-Modified Bentonites in Oxalate Solutions

Abstract

The present study aims to create porous materials through the acid activation of bentonites using 0.5 M oxalic acid at different pH values. Two types of bentonites (containing aluminum montmorillonite and ferruginous montmorillonite) were treated with oxalate solutions at pH 1 to 5. During acid activation at the three pH values, Al, Fe, Mg and Si kinetics were monitored; the porosity of the samples was modified; and the specific surface area increased, while the crystal structure did not completely collapse. The optimum conditions occurred at pH 1, where the highest metal leaching was obtained for both samples. For the sample with aluminum smectite, the specific surface increased from 28.1 m²/g to 149 m²/g and the pore volume quadrupled. In the case of samples with ferruginous smectite, the specific surface area rose from 63. 2 m²/g to 372 m²/g and the pore volume increased sixfold. The mechanism of smectite activation was investigated, revealing that at the optimum experimental conditions, which is ferruginous bentonite activation at pH 1, the products have the highest concentration of small 30 to 50 Å pores, which is attributed to the creation of an adequate number of active sites and the formation of aluminum complexes with the oxalate anions. The modified bentonites have elevated porosity; therefore, they could be used as adsorbents in industry.

1. Introduction

Bentonite is a type of clay consisting predominately of montmorillonite, which is a form of smectite mineral. Smectites (including montmorillonite, nontronite, beidelite, saponite and hectorite) are a family of clay minerals with very high cation exchange capacities, high specific surface areas, as well as other unique physical and chemical properties [1,2]. Greece is the largest global producer of bentonite in Europe and the fourth largest in the world [3].

Acid activation is a chemical process traditionally used to treat clays, typically smectites, which involves submersing the clay in an inorganic acid, such as hydrochloric or sulfuric acid solution. Acid activation of smectites has mostly been studied in inorganic acids, resulting in an increase in surface area of the mineral, modification of the material’s porosity and enhancement of its adsorption capacity [4,5]. Acid-activated bentonites have a wide range of applications in various industries. For example, they can be used for the purification of used motor oils [6] and can act as catalysts for a variety of reactions useful for industry [7] or as adsorbents of pollutants like heavy metals and dyes in wastewater treatment [8,9]. Additionally, they are of paramount importance for the vegetable oil industry, which continues to grow year after year [10]. Acid-activated bentonite can efficiently decolorize edible oils through the removal of coloring agents [11,12].

Several studies on the acid activation of smectites with inorganic acids have been reported in the literature. One research work investigated the specific surface area and porosity of acid-activated bentonites using sulfuric acid at several acid concentrations, ranging from 1.5 to 6 M. The surface area increased with acid concentration [13]. Another study focused on the sulfuric acid activation of calcium bentonite with various smectite/acid ratios. The results revealed that acid activation with sulfuric acid 40% lead to the maximum pore volume and specific surface area [14]. A bentonite sample was activated with sulfuric acid 20% at 95 ˚C. The mesoporous character of the sample increased after treatment. Additionally, the specific surface area of activated bentonite was three times larger than that of the raw sample. It was assumed that the product possesses appropriate characteristics for the discoloration of edible oils [15]. The bleaching power of acid-treated bentonites by a dry method was examined, using sulfuric acid with concentrations of 10% to 70%, at 97 ˚C. The maximum surface area was achieved with an acid concentration of 40–45%, while the maximum bleaching power was achieved after activation with 50–60% sulfuric acid. It was observed that the bleaching power depended on the pore size distribution of the material [5]. The structural and textural characteristics of the acid-treated samples indicated that they could be used as effective adsorbents.

An alternative method for the acid activation of smectites could be the use of organic acids, which enhances the dissolution of clay minerals because of the presence of organic ligands [16,17]. It has been reported that the rate of kaolinite dissolution is enhanced in organic acid solutions because of the complexing ability of organic acid’s anions with aluminum. The effect of the presence of oxalic acid on the mechanism and degree of dissolution rate enhancement of kaolinite has been defined [18]. Kaolinite dissolution seems to be catalyzed in the presence of oxalic acid due to the simultaneous effect of the oxalate-promoted and proton-promoted mechanisms [17,19]. Oxalic acid has been proven to enhance the dissolution of clay minerals to the greatest extent among other organic acids. Palygorskite was treated with different organic acids (formic, acetic, tartaric, oxalic and citric acid), among which, oxalic acid presented the best leaching capability [20]. As the study about kaolinite dissolution in citric, oxalic and malic acids revealed, oxalic acid resulted in the dissolution of kaolinite at a higher rate, compared with the other two acids [16].

Some kinetic studies about the oxalic acid treatment of montmorillonites showed that it is more efficient than other organic acids. Oxalic acid improved the adsorption properties of bentonite more efficiently than citric and acetic acid and released more iron from montmorillonite than citric and gluconic acid [21,22]. It has also extracted more silicon from the smectite structure than citric acid. The effect of two mineral acids was also compared to that of the organic acids on montmorillonite, and the results reflected that organic acids affect the minerals through acid attacks combined with a chelating process [23].

Works that deal with kinetics and the mechanism of oxalic acid dissolution of montmorillonite have been carried out in the past in very dilute solutions (0.1 to 1 mmol/L), which do not simulate industrial conditions. The effect of oxalic acid on the dissolution reaction of montmorillonite at 25 ˚C was studied, and its dissolution rates were defined via flow-through dissolution experiments [24]. Furthermore, the adsorption of oxalate on montmorillonite was investigated by conducting batch experiments and using DR-FTIR spectroscopy. It was reported that montmorillonite dissolution was enhanced in the presence of oxalic acid between pH values of 4 to 8, compared to ligand-free solutions (with nitric, acetic and hydrochloric acid used for pH adjustment) because soluble chelates and surface oxalate complexes were formed. Based on the results from this work, the adsorption of oxalate at the edges of montmorillonite platelets was modeled [25].

The bleaching capacity of oxalic acid-activated bentonite has been examined by the authors in the past under elevated pulp density (25 g bentonite/100 mL solution). The material with the maximum bleaching efficiency was obtained after 24 h of activation, in 1 M oxalic acid, at 100 °C. The economic feasibility of the product and its suitability for industrial use was assessed. It was proved that by using 60% recycled oxalic acid, the bleaching properties of the material produced were acceptable. Moreover, the bleaching capacity of the activated Greek bentonites was compared to a commercial bleaching earth, and the results showed that these materials could be used as bleaching earths in industry [12].

In summary, oxalic acid, a low molecular organic acid found ubiquitously in ecosystems [24,26,27], can effectively activate minerals, leading to the production of materials with high adsorption capacities. Oxalic acid activation is also more environmentally friendly in comparison with inorganic acids because its effluents after acid activation can be easily decomposed into CO₂ through both microbiological and photochemical processes [12].

The present work focuses on the oxalic acid activation of two bentonites containing smectites with different occupancies of their octahedral sheets. It aims to investigate the pH at which the specific surface area, pore size and pore distribution are optimized in an acid concentration comparable to the conventional method used and to clarify the modification mechanism.

The dissolution of bentonites is studied at pH 1, 3 and 5, as the dissolution mechanism of the smectite is affected by the oxalate complexes formed with its octahedral atoms and by the active sites on the mineral’s surface, which are pH-dependent. The results of the present study revealed that porous materials with high specific surface areas and diverse porosity are produced. These materials could act as adsorbents and, consequently, be used in various industries, such as in the vegetable oil industry, as bleaching earths.

2. Materials and Methods

The bentonite samples used as raw materials in this study were provided by Imerys Industrial Minerals Greece S.A. They originated from the deposits of Milos Island in Greece (AlBe-G) and from the Gujarat area in India (FeBe-I).

The XRF method (employing a XEPOS apparatus from SPECTRO company) was applied in order to carry out a chemical analysis of the samples. Fourier Transform Infrared (FTIR) spectra of the solid samples were obtained by a Perkin Elmer Spectrum 100. The wavenumber resolution for the spectra was 4.0 cm⁻¹, and a total of 16 scans, in the range 4000–400 cm⁻¹, were collected for each spectrum. The KBr pressed disc technique (0.3 mg of the sample and 200 mg of KBr) was applied.

The mineralogical characterization was carried out with a Bruker D8-Focus X-Ray Diffractometer with nickel-filtered CuKa radiation (l = 1.5406 Å) at 40 kV and 40 mA.

During acid activation, samples of the pulp were collected and filtered. The leachate of the process was analyzed for Al, Si, Fe and Mg content by Atomic Absorption Spectrometry using a Perkin Elmer PinAAcle 900T Atomic Absorption Spectrometer to assess extraction extent.

The specific surface area was determined from nitrogen isotherms (N₂ adsorption–desorption at 77.3 K) through the application of the BET equation. The samples were degassed for 3 h at 100 °C under vacuum before analysis using a NOVA 1200 surface area and pore size analyzer (Quantachrome company). The mean pore diameter was obtained by applying the BJH method.

Batch experiments were carried out in a jacketed glass reactor at 80 °C for 24 h. The solution temperature was kept constant by a silicon-based heat transfer fluid and teflon-coated thermocouple connected to a temperature controller. A glass condenser had been adjusted to the sealed glass reactor to prevent vapor losses. The solution was continuously stirred at 400 rpm by using a mechanical stirrer. A total of 400 mL of the buffer H₂C₂O₄/K₂C₂O₄ solution with a constant total oxalate concentration of 0.5 M and pre-adjusted pH values (pH 1, pH 3 and pH 5) was heated in the glass reactor. A 0.5 M acid concentration was considered appropriate since it closely matches industrial conditions and has also been shown to produce high-quality adsorbents. In a previous study on the effect of acid concentration on the structure and surface of smectites, the metal extraction results revealed that the dissolution of AlBe-G was not significantly affected as the acid concentration increased from 0.5 to 1 M. Moreover, bentonites treated with 0.5 M oxalic acid were used as bleaching earths in soybean oil and reduced the Lovibond red and yellow color to values within the specification range. Considering the above findings along with the cost, it is more reasonable to study acid activation at the various pH values using a 0.5 M concentration of oxalic acid [28].

Bentonite powder was added to the buffer H₂C₂O₄/K₂C₂O₄ solution, creating a suspension with a pulp density of 2% (w/v). Then, 5 mL slurry samples were taken at 10 min, 20 min and 30 min, followed by every hour until 6 h. The conductivity of the buffer solution was measured during the acid activation process and appeared to remain almost constant for 24 h, with a deviation of less than 5% from its initial value. Following acid activation, the pulp was filtered and the solid residue was washed with deionized water and dried at 100 ˚C for 24 h for further analysis.

3. Results

3.1. Material Characterization

3.1.1. Chemical Analysis of Raw Materials

In

Table 1. Chemical analysis of raw materials.

Oxides (wt. %)
	SiO2	Al2O3	Fe2O3	MgO	CaO	Na2O	K2O	TiO2	L.O.I.	Total
AlBe-G	48.28	17.25	3.68	5.28	9.01	0.40	0.30	0.71	15.10	100.00
FeBe-I	46.96	17.66	19.32	2.80	1.75	1.08	0.00	1.39	9.05	100.00

, the chemical composition of the raw materials is presented. It has been deduced previously [28] that AlBe-G contains a Ca-smectite and FeBe-I contains a mixed Na/Ca smectite.

3.1.2. X-Ray Diffraction (XRD)

The full XRD diagram and the 060 maximum of the raw samples (Figure 1) were presented and analyzed in the authors’ previous study [28]. Montmorillonite was the main mineral of the AlBe-G sample. Calcite, dolomite and illite were found also in AlBe-G bentonite. FeBe-I bentonite peaks were assigned to montmorillonite as the predominant mineral, kaolinite, hematite, calcite and calcium titanium oxide.

3.1.3. Infrared Spectroscopy (FT-IR)

The characteristic vibrations of hydroxyl groups, the silicate anion and the octahedral atoms of the untreated samples were presented in the IR spectra in the authors’ previous study [28]. The IR spectra are also presented below (Figure 2), along with a description of the most characteristic bands.

It is observed that the most intensive band for the AlBe-G samples appears at 1040 cm⁻¹, which is attributed to the Si-O stretching vibrations of the tetrahedral layer. Additionally, the band at 3625 cm⁻¹ corresponds to hydroxyl groups bonded with octahedral Al³⁺ [29]. Similarly, a Si-O stretching band for the FeBe-I samples is observed at 1031 cm⁻¹ and 1012 cm⁻¹ [30], and the bands near 3696 and 3623 cm⁻¹ are assigned to the hydroxyl groups connected to the octahedral atoms [29]. Moreover, the bands at 524 cm⁻¹ (for AlBe-G) and 535 cm⁻¹ and (for FeBe-I) are assigned to Si-O-Al (where Al is an octahedral atom) bending vibrations. Additionally, the peak at 468 cm⁻¹ is attributed to Si-O-Si bending vibrations.

The three peaks associated with the hydroxyls bonded to AlAl, AlFe and AlMg are located near 918 (Al₂OH), 881 (AlFeOH) and 845 cm⁻¹ (AlMgOH) in the spectrum of AlBe-G and near 907, 874 and 832 cm⁻¹ in the spectrum of FeBe-I [29].

The peak at 627 cm⁻¹ in the AlBe-G spectrum is assigned to R-O-Si and appears because of perpendicular vibration of the octahedral atoms. It is also characteristic of Al-rich montmorillonites [31]. Additionally, the CO₃ stretches of calcite and dolomite are related to the band from 1382 to 1430 cm⁻¹ [32].

As the previous work revealed, the XRD and IR-spectra indicated that the AlBe-G and FeBe-I samples have different occupancies in their octahedral sheets; AlBe-G contains a dioctahedral aluminum smectite and FeBe-I a dioctahedral ferruginous one.

3.1.4. Pore Analysis

The pore size distribution, the specific surface area and the porosity of raw materials are presented in Figure 3 and

Table 2. Surface area and porosity of raw samples.

	AlBe-G	FeBe-I
Specific surface area (m2/g)	28.11	63.21
Average pore diameter (Å)	65.98	40.20
Total pore volume (cm3/g)	0.046	0.064

. The specific surface area of the AlBe-G sample is 28.11 m²/g and that of FeBe-I is 63.21 m²/g. Furthermore, it was observed that the FeBe-I sample has more pores with smaller diameters compared to the AlBe-G one.

3.2. Experimental Results

3.2.1. Effect of pH on Smectite Dissolution Rate

The metal extraction curves, concerning metals extracted during oxalic acid activation, are presented in Figure 4 for the AlBe-G sample and in Figure 5 for FeBe-I bentonite. In these figures, the vertical axis represents the extraction percentage (%E) for each metal, which is calculated as follows:

$$\mathrm{\%}E={\displaystyle \frac{m_l}{m_s}}\times 100$$

where $m_l$ is the mass of metal extracted in the acidic solution and $m_s$ is the mass of metal in the solid sample before its activation.

The concentration of metals in the solution increased over time. Nevertheless, the Si in the FeBe-I sample at pH 1 seemed to precipitate after 4 h, as indicated by the drop-off to 24 h. It was also observed that dissolution extent and rate are dependent on pH. The major extraction of metals, both in the FeBe-I and AlBe-G samples, occurred at pH 1 (yet %E did not reach 100%). The highest extraction of iron and aluminum in the oxalate solution at pH 1 was in accordance with the results from previous studies on the dissolution of hematite and corundum [33,34]. Subsequently, the extraction of metals was higher at pH 3 than at pH 5. The magnesium in AlBe-G was an exception to this pattern since its release was greater at pH 5 than at pH 3. Additionally, in the same sample, the extraction of iron at pH 3 was almost same as at pH 5.

3.2.2. Effect of pH on Specific Surface Area and Porosity

The modifications in the porosity and specific surface area of the raw and oxalic acid-activated AlBe-G samples are shown in

Table 3. Specific surface area and porosity of raw and treated AlBe-G samples.

	AlBe-G Raw	AlBe-G pH 1	AlBe-G pH 3	AlBe-G pH 5
Specific surface area (m2/g)	28.1	149.0	50.7	92.5
Average pore diameter (Å)	66.0	56.3	82.3	57.2
Total pore volume (cm3/g)	0.05	0.21	0.10	0.13

and Figure 6. The specific surface area increased after acid activation; the material produced after treatment at pH 1 had the highest specific surface area, followed by the samples activated in pH 5 and pH 3. The samples treated at pH 1 and pH 5 had almost the same average pore diameters, while after activation at pH 3, the pore size was higher compared to the other two pH values. The total pore volume increased after treatment and was higher at pH 1 than at pH 3 and 5.

Pore size distribution (PSD) curves for the AlBe-G samples are shown in Figure 5. The number of pores in each sample is associated with the area under the PSD curve of the sample and the two abscissa values defining the curve. The diameters of the pores were between these abscissa values [5]. As observed in Figure 7, the curve of untreated AlBe-G shows that most of the pores in this sample had a diameter between 35 Å and 45 Å. The curves of the samples after acid activation are taller; thus, pores with diameters between these values were more numerous. The acid-activated samples also have wider PSD curves compared to the raw materials, meaning that new pores with diameters smaller than 35 Å and larger than 45 Å were created. The sample treated at pH 1 has more pores with diameters between 15 Å and 45 Å as well as between 45 Å and 80 Å than the other samples.

Table 4. Specific surface area and porosity of raw and treated FeBe-I samples.

	FeBe-I Raw	FeBe-I pH 1	FeBe-I pH 3	FeBe-I pH 5
Specific surface area (m2/g)	63.2	372.3	135.8	94.7
Average pore diameter (Å)	40.2	37.6	48.7	67.0
Total pore volume (cm3/g)	0.06	0.35	0.17	0.16

and Figure 8 concern the pore and surface characteristics of the raw FeBe-I and oxalic acid-activated samples. All activated samples presented higher specific surface areas compared to their raw materials. Furthermore, the sample treated at pH 1 developed the highest specific surface area, followed by the one of bentonite activated at pH 3 and, finally, the one after treatment at pH 5. The average pore diameter of the activated samples increased as pH rose. The total pore volume of the treated samples was higher than that of the raw ones and obtained their highest values when FeBe-I was activated at pH 1.

The pore size distribution curve for FeBe-I shows that the raw material contains many pores with diameter between 35 Å and 45 Å and some pores with diameter between 15 Å and 35 Å. Each acid-activated FeBe-I sample has a higher and wider PSD curve compared to the raw material (Figure 9). The acid-activated samples present more pores between 35 Å and 45 Å compared to the untreated sample. After activation at pH 3 and pH 1, the number of pores with diameters below 35 Å increased compared to the raw material, while the product at pH 1 showed the most pores with this diameter. After activation, pores with diameters between 40 Å and 80 Å also developed, more so at pH 1.

It appears that pH 1 has a greater impact on the iron-bearing minerals. As seen in the PSD diagram of AlBe-G, the pore volume adsorbed by the pores in the 35 to 45 Å range increases from approximately 2 × 10⁻³ (cc/Å/g) for the raw material to about 12 × 10⁻³ (cc/Å/g) after treatment at pH 1. Similarly, the PSC curve of the FeBe-I sample rises from nearly 2 × 10⁻³ (cc/Å/g) to about 24 × 10⁻³ (cc/Å/g), indicating the creation of a more porous sample compared to AlBe-G. This is confirmed by a previous study on the same samples treated at different concentrations of oxalic acid [28], which showed that the iron-bearing sample had twice the pore volume of the aluminum-bearing sample at the same acid concentration. The PSD curve of the iron-bearing sample reached nearly 27 × 10⁻³ (cc/Å/g), while for the AlBe-G sample, it reached about 12 × 10⁻³ (cc/Å/g). The iron-bearing sample also showed enhanced decolorization as far as the Lovibond red and yellow color of oil is concerned.

3.2.3. Structural Modification of the Samples During Acid Activation

In Figure 10 and Figure 11, the FT-IR spectra of the treated AlBe-G and FeBe-I samples in comparison with the untreated ones are depicted respectively. Alterations in the IR spectra indicate modifications in structure according to the literature. Although peaks assigned to free SiO₂ were observed in both samples (at 1100 cm⁻¹ and 1200 cm⁻¹ in the AlBe-G products and at 1100 cm⁻¹, 1200 cm⁻¹ and 800 cm⁻¹ in the FeBe-I products [30]), the tetrahedral sheet has not been transformed into an amorphous silica framework. This is evident from the peak of stretching vibrations of the Si-O bonds in the tetrahedral layer which remained after treatment, and it is clearly seen.

The band near 520 cm⁻¹ is a reliable indicator of smectite dissolution extent. In the treated samples, clear peaks exist in this region, rather than just an inflection, which would reflect significant dissolution of montmorillonite.

It has also been reported that dioctahedral smectites are considered completely dissolved when the vibrations of octahedral cations (OH bending and Al-O-Si) are absent from the IR spectra. In the IR spectra of the present study, the abovementioned peaks remained after acid activation [30,35]. Additionally, the band at 3625 cm⁻¹, corresponding to the hydroxyl groups bonded with octahedral Al³⁺ in AlBe-G, and the bands near 3696 and 3623, attributed to the hydroxyl groups connected to the octahedral atoms in FeBe-I, were preserved after acid treatment.

The three peaks associated with the hydroxyls bonded to AlAl, AlFe and AlMg were reduced after the acid activation, since the octahedral atoms were leached out from the structure.

The abovementioned observations reveal that the smectite atomic structure does not collapse completely after acid activation at the different pH values.

The XRD spectra of acid-activated bentonites are presented in the figures below (Figure 12 and Figure 13). As observed, after 24 h of acid treatment, the (001) peak of montmorillonite decreases in both samples at pH 3 and pH 5 and almost disappears at pH 1. This indicates a loss of coherent layer stacking [35]. However, the peaks of smectite at 2θ 20° remain intact. Additionally, the peaks of calcite and dolomite disappear, suggesting their dissolution after treatment, while the peak of kaolinite in FeBe-I remains.

4. Discussion

From the presented experimental results, we can assume that when AlBe-G and FeBe-I bentonites are treated with oxalic acid at various pH values, new materials with higher specific surfaces are produced. Although a large percentage of metals were leached from the octahedral and tetrahedral sheets of the samples, the IR (Figure 10 and Figure 11) and XRD (Figure 12 and Figure 13) spectra reveal that the smectite structure was not completely destroyed in any of the products. At pH 1, the extraction of metals (Figure 4 and Figure 5) as well as the specific surface area and the total pore volume (Table 3 and Table 4) are maximized, while the average pore diameter has the lowest value. Metals are extracted mainly from the octahedral sheets, which are occupied to a considerable extent by aluminum, as is observed from the chemical analysis and the IR spectra of raw samples (Table 1, Figure 2).

The literature has revealed that the overall dissolution process is the result of two mechanisms that act simultaneously: (i) “proton-promoted mechanism”, where protons replace the exchangeable cations or are adsorbed at the edges of the layers, weakening the bonds of the metal species, and (ii) “ligand-promoted mechanism”, where organic ligands form metal–organic complexes on the surface of the mineral (weakening the metal–oxygen bond) or in the solution, thus increasing the dissolution rate of the mineral [12,17,19,24]. Consequently, organic acids enhance the dissolution of clay minerals and phyllosilicates because of the presence of the organic ligands. Oxalate has been proven to enhance the dissolution of clay minerals to the greatest extent among other organic acids [16,21,22,23].

As per the literature, oxalic acid is a diprotic organic acid, which dissociates as shown in the following equations (Equations (1) and (2)):

$$H_2{C}_2{O}_4\leftrightarrow H^{+}+H{C}_2{O}_4^{-}$$

(1)

$$H{C}_2{O}_4^{-}\leftrightarrow H^{+}+C_2{O}_4^{2-}$$

(2)

Undissociated H₂C₂O₄ and dissociated fractions of $\mathrm{H}{\mathrm{C}}_2{\mathrm{O}}_4^{-}$ and ${\mathrm{C}}_2{\mathrm{O}}_4^{2-}$ coexist in equilibrium and vary as a function of pH. At pH values above 3, the oxalic acid is nearly completely ionized; thus, both ${\mathrm{C}}_2{\mathrm{O}}_4^{2-}$ and $\mathrm{H}{\mathrm{C}}_2{\mathrm{O}}_4^{-}$ exist. At a pH value below 2, the $\mathrm{H}{\mathrm{C}}_2{\mathrm{O}}_4^{-}$ species reach high rates while the ${\mathrm{C}}_2{\mathrm{O}}_4^{2-}$ species are negligible. Between pH 2 and 3, the dominant species is $\mathrm{H}{\mathrm{C}}_2{\mathrm{O}}_4^{-}$. This speciation of oxalate has been depicted in a diagram of a previous study (Figure 14) [36].

It has been documented in studies about oxalic acid’s effect on montmorillonite that, the protonation of the edge aluminum sites as pH decreases creates a positive charge on the surface which permits electrostatic adsorption of oxalate anions on the mineral’s edge. The Al-O bond weakens as well due to this protonation, which leads to a decrease in the electron density of the bond. Consequently, the reaction between oxalate and aluminol sites by ligand exchange and the formation of complexes are empowered. A decrease in pH yields higher protonation of the surface, more OH groups, and more sites for adsorption and H bonding [24,25]. It has also been reported that the adsorption of oxalate on corundum (Al₂O₃) [34] increases as pH decreases. Indicatively, a possible chemical reaction taking place in the oxalic acid–bentonite system could be described by the equation below [36]:

$$>{{\rm M}}^{III}-O{H}^{+}+L^{n-}+H^{+}\leftrightarrow {\left[>{{\rm M}}^{III}-L\right]}^{-\left(n-2\right)}+H_{2}O$$

where M is a trivalent metal (Al³⁺ or Fe³⁺), $L^{n-}$ is the organic ligand with oxidation number n ($C_2{O}_4^{2-}$ or $H{C}_2{O}_4^{-}$) and the symbol “>” means it is a surficial complex.

Taking the abovementioned remarks into consideration, it is assumed that the high metal extraction percentage from montmorillonite and the largest specific surface area both arise at pH 1 because of the high adsorption of oxalate on a smectite surface at decreased pH values. In an acidic environment, as at pH 1, stable complexes of aluminum ions and ligand $\mathrm{H}{\mathrm{C}}_2{\mathrm{O}}_4^{-}$ are formed [37].

The treatment of samples at pH 3 and pH 5 resulted in the production of materials with higher specific surface areas and total pore volumes compared to the raw ones. Among these two pH values, AlBe-G exhibits its highest specific surface area at pH 5, while FeBe-I at pH 3. This differentiation between the two materials exists because of their diverse structure. The ferruginous smectite of FeBe-I bentonite bears much more iron in its octahedral sheets than the smectite of the AlBe-G sample. As per Panias et al. (1996) [36], the dissolution of iron oxides in oxalic acid is optimized at a pH of approximately 3 since; at that pH value, the number of surface-active centers (due to protonation) and of oxalate ions that complex with iron ($\mathrm{H}{\mathrm{C}}_2{\mathrm{O}}_4^{-}$) and accelerate its dissolution rate are in the ideal ratio. Thus, the smectite in the FeBe-I sample with a significant amount of iron in its octahedral sheets exhibits higher metal extraction at pH 3 than at pH 5 (Figure 5) and consequently higher specific surface areas. Furthermore, in the AlBe-G smectite, the octahedral sites are occupied mostly by aluminum and by a significant quantity of magnesium as well, as the pronounced peak at 845 cm⁻¹ in the FTIR spectrum indicates (Figure 2). Magnesium complexes with oxalate in solution from pH 4 to pH 10, as it has been observed [24]. These observations come in accordance with the higher extraction of Mg (Figure 4) and higher specific surface area at pH 5 compared to pH 3 in the AlBe-G sample.

In a previous study, the acid activation of two bentonites containing aluminum and iron in their octahedral sheets was conducted to compare oxalic acid activation with the traditional method using sulphuric acid at an acid concentration of 1 M [38]. The specific surface area of bentonite with more iron in its octahedral sheets appears to increase by approximately 80% when using organic acid for activation. Considering the fact that by using oxalic acid, the products have larger specific surface areas, achieved with a low acid concentration, and also that oxalic acid is more environmentally friendly than mineral acids (its effluents can decompose both photochemically and microbiologically), it is assumed that oxalic acid activation is a promising route to optimize acid activation.

The photodecomposition of oxalic acid is a simple process that could take place in open ponds, either in batch or continuous mode, by exposure to daylight for a period of 4 to 7 days [39].

5. Conclusions

In the present study, the effect of pH on the surface, porosity and dissolution of two different types of smectites (the aluminum one of AlBe-G bentonite, and the ferruginous one of FeBe-I) was investigated to explore the optimum conditions for oxalic acid activation of bentonites and suggest a possible mechanism of dissolution.

It is inferred that during oxalic acid activation at different pH values, the metals are leached out from the smectite structure to some extent, without completely destroying the original structure of the material. As a result, the specific surface area and porosity of the products are enhanced. The largest extraction of metals from the samples is observed after oxalic acid activation at pH 1. At these experimental conditions, the materials produced have the highest pore volume and specific surface area. This can be attributed to the creation of an adequate number of active sites and the formation of aluminum complexes with the oxalate anions. The optimum porosity results are observed for the FeBe-I material at pH 1. In this sample, as pH increases, the specific surface area decreases. For AlBe-G, the specific surface area decreases by increasing pH up to pH 3, but above pH 3, it begins to increase. This behavior is attributed to the diverse chemical content of the octahedral sheets in AlBe-G and FeBe-I.

The experimental results reveal that oxalic acid activation produces materials with diverse porosity and high adsorption capacity, which could be used in industry as adsorbents, like in the vegetable oil industry, as bleaching earths.

Oxalic acid activation can be more effective than traditional inorganic acid activation for bentonites containing iron-rich smectites. A low concentration of 0.5 M is sufficient to achieve high porosity and a large specific surface area. This method is also more eco-friendly, as oxalic acid can be decomposed both microbiologically and photochemically.