Effect of Oxycations in Clay Mineral on Adsorption—Vanadyl Exchange Bentonites and Their Ability for Amiloride Removal

Abstract

The presence of drugs in aquatic bodies is a prevailing issue, and their removal by adsorption is an effective treatment. Among the adsorbents, those based clay minerals have been proposed. Bentonite is a clay mineral that is widely studied as an adsorbent due to its unique physicochemical properties, such as cation exchange capacity (CEC), intercalation, and adsorption. The properties of bentonites can be improved through chemical modifications, such as the incorporation of organic and/or inorganic compounds. These modifications allow for the efficient removal of different contaminants, including pharmaceutical compounds. In this work, raw sodium bentonite (Na⁺-Bent) and vanadyl bentonites were prepared using 100 (BentV1), 300 (BentV3), and 500% (BentV5) of the cationic exchange capacity of the Na⁺-Bent and further used for amiloride removal from aqueous solution. Analysis of X-ray fluorescence and Na⁺ in solution after interaction indicated that the principal mechanism of interaction between bentonite and ions was the ion exchange between sodium of the matrix and vanadyl in solution. Infrared spectroscopy suggested the contribution of coordination of the interlayer water with the vanadyl ions and hydrogen bonding between vanadyl and structural OH. X-ray diffraction analysis indicated that vanadyl ions were incorporated onto Na⁺-Bent. Amiloride adsorption was better at pH 5.8, using a solid dosage of 75 mg of Na⁺-Bent, 25 mg of BentV1 and BentV5, and 50 mg of BentV3. The adsorption occurred briefly until 20 min, and maximum removal values were 457.08, 374.64, 102.56, and 25.63 mg·g⁻¹ for Na⁺-Bent, BentV1, BentV3, and BentV5, respectively. At lower drug concentrations (48.78 and 91.24 mg·g⁻¹ for Na⁺-Bent and BentV3), the best performance was obtained for the BentV3 sample.

1. Introduction

The occurrence of drugs on the aquatic medium in concentrations of μg·L⁻¹ or ng·L⁻¹ can result in damage to animal and plant lives [1,2] and long-term water contamination due to low biodegradability [3,4,5,6,7]. Therefore, new technologies for drug removal that are frequently detected in aquatic ecosystems are a prevailing issue [7,8,9,10,11,12]. Among the treatments proposed for environmental remediation, adsorption is a promising alternative considering its relatively simple operation. The development of new and efficient adsorbents, including those based on clay minerals, has been described in recent years [13,14]. Clay minerals are natural or synthetic phyllosilicates, and their properties are adequate for adsorption, such as cation exchange capacity [15,16], expansion, high surface specific area [17], and presence of acidic and basic sites on the surface. These properties are important for their interaction with organic pollutants, especially drugs [18]. Among clay minerals, bentonite (Bent) has been used in environmental remediation processes [19,20,21]. The term bentonite is a rock with at least 50% smectite, particularly montmorillonite [22]. Montmorillonite (Mt) is a 2:1 phyllosilicate with layers in nanometers (≈1 nm) and a net negative charge in the lamella in the 0.2–0.6 range. The negative charge occurs through the exchange of trivalent cations (Al³⁺) for divalent (Mg²⁺) in the octahedral layer [23,24], and its neutralization occurs by exchangeable hydrated cations that are present in the interlamellar region, especially Na⁺ and Ca²⁺. The surface properties of bentonites can also be changed through chemical modification processes [25], including ion exchange with organic and inorganic cations [26], organic and inorganic anion bonds at the surface edge, intercalation of organic compounds [27], acid activation [28], and pillarization [29]. All modifications enable new properties and new adsorption sites to bentonites and can expand their applications in environmental remediation [30,31,32]. Ion exchange reactions in bentonites by inorganic [33] or organocations [34] have been widely explored for drug removal [25,35]. It is known that the nature of the interlayer cation plays a vital role in drug interactions [36,37,38,39]. However, although the use of sodium form of Mt (Na-Mt) is reported for drug removal [40,41], the low sedimentation and high swelling capacity of sodium bentonites can limit their application in wastewater treatment operations, especially in column adsorption processes [40,42]. Therefore, the nature of the interlayer cation and type of the drug are important aspects for the performance of Mt. Additionally, the type of interaction between organic species and Mt in adsorption is mainly influenced by the drug speciation, which is pH-dependent [36,43,44,45]. Amiloride (N-Amidino-3,5-diamino-6-chloropyrazinecarboxamide hydrochloride hydrate, Figure 1) is an oral diuretic drug that promotes potassium retention by the organism and the reduction of sodium and chlorine by excretion and hypertension. It is often used as a booster for other diuretics and to combat congestive heart failure since it has a weaker action than thiazide diuretics [46]. Studies have shown that amiloride does not undergo metabolic transformation and approximately half of the oral dose can be excreted through urine. In addition, the remainder of the unmetabolized drug can be recovered in the feces [47,48]. This can cause a major environmental problem, considering that the wastewater from sewage treatment plants is the most common source of these contaminants, which are released by the population through excreta or final disposal, and collected in sewage networks that do not have treatment plants designed to remove them [49,50]. Moreover, amiloride is photosensitive and undergoes a rather complex photodegradation process involving many different photoproducts [51]. Amiloride mineralization has been proposed as an alternative for drug removal at concentrations of 10 mg·L⁻¹. Its photodegradation was evaluated using catalysts based on ZnO:g-C₃N₄ heterostructures with different amounts of g-C₃N₄ (15%, 50%, and 85%) [52]. Depending on the water pH during photodegradation, simultaneous protonation can occur, resulting in more chemical species that can be photodegraded and, consequently, producing more photoproducts [53]. Therefore, amiloride removal by adsorption on inorganic matrices containing photoactive species may result in the possibility of photodegradation of the drug later, as reported in the literature for other drugs, including ciprofloxacin [54], tetracycline [55], amoxicillin, diclofenac sodium, and paracetamol [30]. Clay minerals have also been described for drug photodegradation [56]. However, amiloride adsorption by clay minerals has been poorly reported in the literature. For example, a modified chitosan magnesium phyllosilicate was used for amiloride removal at pH 5.8, resulting in a maximum adsorption capacity of 55.74 mg·g⁻¹ at 1500 mg·L⁻¹ after 24 h [57].

In this context, the present study aimed to evaluate the use of vanadyl-exchanged bentonites at different concentrations as adsorbents for amiloride hydrochloride at pH 5.8, which is the natural pH of the species in an aqueous solution. The influence of the amount of vanadyl ions (VO²⁺) incorporated into the clay mineral was evaluated and compared to sodium bentonite. Vanadyl derivatives proved to be more active and selective in oxidation processes [58]. Vanadyl ion was adsorbed on hydrated hectorite and the study indicated that hydrolysis of VOH was promoted at low levels of adsorption [59]. The hydrolyzed product was adsorbed on the clay surfaces, with a ligand environment that was partially aqueous and partially hydroxide in nature [59]. Chiral bis(oxazoline) ligands are used to promote the enantioselective vanadium-catalyzed oxidation of sulfides with alkyl hydroperoxides. Several bis(oxazoline)-VO complexes have been prepared and supported by cation exchange in Laponite [60]. Vanadium oxides in clay minerals have been also reported as potential catalysts for oxidation reactions [61,62]. In a recent study, vanadium species were incorporated into a free aluminum layered silicate, followed by heat treatment to form V₂O₅. Vanadium oxide was anchored on the silicate surface, and the optical and magnetic properties of the hybrid materials suggested its potential application as a catalyst [63]. Therefore, the preparation of vanadyl exchange bentonites is described for the first time, and their performance for amiloride removal was evaluated as a function of the vanadyl content in the silicate matrix.

2. Materials and Methods

2.1. Chemicals and Materials

A sodium bentonite sample was obtained from the Bentonise Bentonita Company, Campina Grande, Brazil. The CEC was 88 cmol(+)/kg, as measured by the ammonium-exchange determination [64,65]. The chemical composition of the bentonite indicated the following constituents: SiO₂ (52.98%), Al₂O₃ (18.35%), Fe₂O₃ (3.96%), Na₂O (2.56%), MgO (2.47%), TiO₂ (0.18%), K₂O (0.22%), and CaO (0.01%) [61]. Vanadyl sulfate hydrate (VOSO₄ H₂O, 97% purity) was supplied by Alfa Aesar (Haverhill, MA, USA), and amiloride hydrochloride hydrate (C₆H₈ClN₇O·HCl·xH₂O, purity 98%, CAS 2016-88-8) from Sigma Aldrich (St. Louis, MO, USA) was used in the adsorption tests. All the chemicals were used without prior treatment. Distilled water was used as the solvent in all the procedures.

2.2. Quartz Removal

Initially, quartz removal of raw bentonite was performed using the centrifugation/decantation method [65]. A sample of approximately 100 g was washed with 500 mL of distilled water, centrifuged, dried at 50 °C for 24 h, and sieved through a 200 mesh. Therefore, sodium bentonite (Na⁺-Bent) was used in the adsorbent preparation and adsorption tests.

2.3. Preparation of the Vanadyl Exchanged Bentonites

Na⁺-Bent (10.0 g) was reacted with 100 mL vanadyl sulfate solutions at concentrations of 100%, 300%, and 500% of the CEC of the clay. At this condition, salt amounts of 1.76, 5.29, and 8.81 g in vanadyl sulfate/Na⁺-Bent dispersions were maintained at 200 rpm and 30 °C for 144 h in a thermostabilized bath. The vanadyl sulfate solution in the reaction vessel was repositioned every 48 h to control the volume of the solvent. The solids were separated by centrifugation at 7500 rpm for 5 min and washed with distilled water. Finally, the prepared bentonites were dried at 60 °C for 72 h and named as BentV1, BentV3, and BentV5 for the solids prepared with vanadyl sulfate at 100%, 300%, and 500% of clay CEC, respectively.

2.4. Adsorption of Amirolide

Both Na⁺-Bent and exchanged bentonites were used in the adsorption tests. The influence of the adsorbent dosage was determined for 25, 50, 75, and 100 mg of the solid dispersed in 50 mL of 50 mg·L⁻¹ amiloride solution at pH 5.8. At pH 5.8, the cationic form of amiloride is predominant in the solution, and its biological activation occurs at the same value [66]. The dispersions were maintained in orbital agitation at 30 °C in a TE-420 model Tecnal incubator for 24 h. After each determination, the adsorbent samples were recovered by centrifugation, and the solids were dried at 50 °C for further characterization. Amirolide concentration in the solution was monitored before and after adsorption at 286 nm by UV-VIS spectrometry using a Shimadzu spectrometer UV-2550 (Shimadzu, Kyoto, Japan) model. The amount of adsorbed amiloride was calculated using Equation (1):

$$q=\frac{\left(C_i-C_e\right)·V_c}{m}$$

(1)

where C_i and C_e are the initial and equilibrium amiloride concentrations, respectively; V_c is the volume of the drug solution; and m is the mass of the solid used in each test. The influence of the time in the adsorption was performed at 10–120 min intervals at pH 5.8, and the optimal adsorbent dosage and the same procedure were adopted. The equilibrium isotherms were evaluated for 10–100 mg·L⁻¹ and 10–500 mg·L⁻¹ initial amiloride concentrations at the optimal time obtained in the kinetic tests at pH 5.8.

2.5. Characterizations

XRD diffractometry was performed at room temperature in a Shimadzu XD3A model, with a scan rate of 0.03 s⁻¹, using CuKα (λ) as the source radiation (λ = 0.15406 nm), between 2θ of 2.5 to 80° and a fixed power source (30 kV and 30 mA)). The average crystallite size was calculated based on the Scherrer equation [67], defined as:

$$D=\frac{K\lambda }{\beta cos\theta }$$

(2)

where D is the crystallite size; K is a constant whose value depends on the shape of the particle (which is equal to 0.89 for spherical particles); λ is the wavelength of the electromagnetic radiation (λCu = 1.5406 Å); θ is half the angle of diffraction; and β is the full width at half maximum (FWHM) of the peak crystallite size on the 002 plane, whose contributions due to instrumental enlargement and strain were discounted.

Fourier transform infrared spectroscopy (FTIR) was performed using a Bomem MB-series FT spectrometer, Quebec, Canada. The samples were prepared using KBr pellets at a concentration of 2%. The FTIR spectra (4000–400 cm⁻¹) were recorded with a resolution of 4 cm⁻¹ and 32 accumulations.

X-ray fluorescence (XRF) was performed using a Shimadzu model EDX-7000 (Shimadzu, Kyoto, Japan) under a vacuum, with a 10 mm collimator, scanning from sodium to uranium.

Inductively coupled plasma optical emission spectrometry (ICP OES) was performed using a Spectro, Arcos model (SPECTRO, Kleve, Germany) and used to quantify sodium in the solution.

The zeta potential (ζ) was monitored using a Zetasizer Nano Zs (Malvern Panalytical, Malvern, UK) for isoelectric titration through pH titration. Solutions of 0.100 mol·L⁻¹ NaOH or 0.500 mol·L⁻¹ HNO³ were used to adjust the pH.

Scanning electron microscopy (SEM) images were obtained using a TESCAN VEGA3 SEM (TESCAN, Brno, The Czech Republic), corresponding to a tungsten thermionic emission system, suitable for high and low vacuum operations and equipped with an energy dispersive X-ray spectrometer (EDS), manufactured by Oxford Instruments (Abingdon, UK).

2.6. Kinetic and Equilibrium Models

Experimental adsorption data used in the kinetic and equilibrium models are described in Supplementary Materials Figure S1. For kinetics, experimental data were adjusted to fit the three kinetic models (as shown in Equations (1)–(3) in Supplementary Material SI1): Pseudo-first-order [68], pseudo-second-order [69], and simplified Elovich equation [70], assuming αβt >> 1.

The equilibrium isotherms were analyzed using the Langmuir [71], Freundlich [72], and Temkin [73] models (following Equations (4)–(6) in Supplementary Material SI1).

Standard deviation (SD) was used to verify which equation models were best suited to describe the experimental data [74]. Equation (3) was used for SD determination in the following form:

$$SD=\sqrt{\frac{1}{n_p-p}{\displaystyle \sum }_i^n{\left(q_{i,exp}-q_{i,model}\right)}^2}$$

(3)

where q_i,exp and q_i,model are the experimental adsorbed amiloride amount and theoretical amount obtained by the kinetic or equilibrium models; n_p is the number of performed experiments; and p is the number of parameters of the fitted model.

3. Results

3.1. Characterizations of the Sodium and Vanadyl Exchanged Bentonites before and after Amiloride Adsorption

3.1.1. X-ray Fluorescence (XRF)

The presence of vanadyl after the exchange process was monitored using X-ray fluorescence (

Table 1. Chemical composition by XRF for bentonites before and after interaction with amiloride.

	Samples
	BentV1		BentV3		BentV5
% Oxide	Before	After	Before	After	Before	After
SiO2	54.07	57.90	52.64	53.69	51.26	51.26
Al2O3	16.57	17.70	16.20	16.58	15.91	15.90
Fe2O3	15.25	11.83	14.22	13.23	14.38	14.38
V2O5	5.13	4.35	10.44	9.11	11.92	11.92
SO3	2.81	2.84	1.64	2.24	2.47	2.47
CaO	2.27	1.67	1.58	1.58	0.76	0.76
TiO2	-	-	-	-	-	-
K2O	0.82	0.79	0.77	0.71	0.73	0.73
MgO	0.79	0.65	0.61	0.61	0.46	0.46
BaO	-	-	1.82	2.13	1.93	1.93
MnO	-	-	-	-	0.06	0.46
Na * (mg·L−1)	351.4	-	334.1	-	334.4	-

* Values quantified by ICP OES for residual sodium in the solution.

). The analysis revealed that part of the vanadyl was incorporated into the clay samples with vanadyl concentrations of 75.6, 153.8, and 175.6 cmol(+) kg⁻¹, corresponding to 86, 175, and 199% of CEC for BentV1, BentV3, and BentV5, respectively. The use of vanadyl in high proportions also resulted in better incorporation of the ions in the clay sample. However, a significant amount of vanadyl remained in the solution for the three conditions, as indicated by X-ray fluorescence analysis. After interaction with amiloride, the amounts of vanadyl in solids with maximum amounts of adsorbed drug were 64.0, 134.2, and 145.0 cmol(+) kg⁻¹, suggesting that vanadyl cations can be exchanged with the cationic drug during the adsorption process.

Sobrenadant analysis of the solution after the interaction showed the presence of sodium (Table 1). Sodium amounts were 76.7, 70.8, and 70.8 cmol(+) kg⁻¹, corresponding to 87% of CEC for BentV1 and 80% of CEC for both BentV3 and BentV5. Results indicated that ion exchanged between vanadyl and sodium was a predominant process of interaction considering that the amount of Na⁺ in all cases is close to CEC of the raw bentonite sample (88 cmol(+) kg⁻¹). Moreover, it suggested that vanadyl can be incorporated on external sites of the bentonite or (vanadyl/sulphate) ionic pair intercalated as observed for the interaction between ammonium salts and bentonite at concentrations higher than the CEC of the clay sample [58].

3.1.2. X-ray Diffractometry

The XRD patterns of the samples are presented in Figure 2, with the phases indexed by the ICDD 00-060-0318 and 01-070-8055 files. Sodium bentonite XRD indicated that Mt is the principal constituent. The characteristic Mt reflection was observed at 2θ = 7.24°, representing the basal spacing d₍₀₀₁₎ of 1.22 nm, typical of sodium Mt [75]. Other reflections were also observed at 2θ = 19.8, 28.5, 34.9, and 61.8°, which were indexed to the Mt phase in concordance with the 00-060-0318ICDD file. The reflection at 61.9° was associated with the (060) plane of dioctahedral phyllosilicate [23]. The impurity of quartz in the raw sample was associated with reflection at 26.5°, in agreement with the 01-070-8055 ICDD file. The structure of the bentonite changed after modification. The interlayer distances (calculated using Bragg’s law) and the thickness of the bentonite (calculated using the Debye-Scherer equation) were different post-modification. The data for the samples are presented in

Table 2. Parameters calculated according to the XRD results.

Sample	2θ	d (nm)	∆d * (nm)	D (nm)
Na+-Bent	7.21	1.22	0.26	8.86
BentAmil	6.20	1.42	0.46	5.10
BentV1	6.58	1.34	0.38	4.86
BentV1Amil	6.76	1.30	0.34	7.15
BentV3	6.43	1.37	0.41	5.86
BentV3Amil	6.83	1.29	0.33	5.62
BentV5	6.13	1.44	0.48	5.21
BentV5Amil	6.52	1.35	0.39	4.94

* Mt thickness 0.96 nm [21]. The values of 2θ, d, and D refer to the 001 plan.

. After reaction with vanadyl, the (001) reflection changed for lower 2θ values, indicating an increase in the basal spacing, which are 1.34, 1.37, and 1 nm for BentV1, BentV3, and BentV5, respectively. The thickness of the bentonite (0.96 nm) indicated that values of the free spacing to accommodate the amiloride molecules were 0.38, 0.41, and 0.04 nm for BentV1, BentV3, and BentV5, respectively. Therefore, the XRD data suggested that samples with higher amounts of vanadyl can present more difficulty in accommodating amiloride molecules. Values of D indicated a decrease in the crystallite size in the 001 direction after vanadyl exchange and amiloride adsorption without a regular relationship between the concentration of vanadyl and amiloride, except for BentV1Amil, where the D value was higher than BentV1.

In the sample that interacted with amiloride, it was observed that d₍₀₀₁₎ increased from 1.22 to 1.42 nm for Na⁺-Bent, which is an indication of the amiloride intercalation in the interlayer spacing of the Mt. For the vanadyl exchanged bentonites, d₍₀₀₁₎ reduced for all samples. The d₍₀₀₁₎ values changed from 1.34 to 1.30 nm, 1.37 to 1.29 nm, and 1.44 to 1.35 nm in BentV1, BentV3, and BentV5, respectively. Broader reflections were observed in all samples after the interaction with amiloride, especially Na⁺-Bent, BentV3, and BentV5, suggesting a reduction in the crystallinity of the solids after the drug interaction [76]. The XRD results suggested that amiloride was adsorbed on the surface of the vanadyl exchanged bentonites, while amiloride intercalation was predominant for Na⁺-Bent. It is inferred that the entrance of vanadyl in the interlayer of the precursor sample decreased the free exchange sites, and the drug was preferentially adsorbed on the surface of the vanadyl bentonites.

3.1.3. Zeta Potential

According to the zeta potential (ζ) measurements of the solids before and after cation exchange (Figure 3), the surface charge was always negative in aqueous media, as expected for particles with negative structural charges. The solid particles were submitted to a process involving only cationic exchange, not compromising the bentonite structure, resulting in the substitution of Al³⁺ in the octahedral layer by Mg²⁺ or Fe²⁺ and Si⁴⁺ in the tetrahedral layer by Fe³⁺ or Al³⁺.

3.1.4. Spectroscopy

FTIR spectroscopy was useful for the monitoring of alterations in the OH absorptions and presence of the amiloride in the solids (Figure 4).

The FTIR spectra of the sodium and vanadyl bentonites showed characteristic bands of the inorganic matrix. Vibrations related to structural OH stretching (Al–OH and Mg–OH) shifted from 3626 to 3634 cm⁻¹ as vanadyl ions were incorporated in the clay sample, which suggested that structural OH groups are also involved in the interaction [77]. The bands at 3437 cm⁻¹ related to OH stretching in water and Si-OH were unchanged [77], and OH deformation in water occurred at 1635 cm⁻¹ without alteration in their position [77]. Bands at 1117 and 1043 cm⁻¹ are assigned to Si–O–Si and Si–O–Al antisymmetric and symmetric stretching, respectively. These latter bands changed to higher wavenumbers after vanadyl was exchanged and reached 1126 and 1060 cm⁻¹. The shift for higher wavenumbers indicates stronger bonding, which may be associated with the coordination of the interlayer water with the vanadyl ions, considering that vanadyl species form stable complexes with water molecules in five coordination [78]. The formation of interlayer complexes can result in weak interactions between water and the Si–O–Si and Si–O–Al surfaces. The bands at 911 and 835/842 cm⁻¹ were assigned to OH bending in Al₂OH and AlMgOH of the octahedral layer of clay minerals, thus reflecting the partial substitution of Al³⁺ by Mg²⁺ [76]. At 462 cm⁻¹, the bands were attributed to Si–O and Al–O–Si bending [79]. Vanadyl ions were assigned to the bands between 843 and 622 cm⁻¹, which are attributed to the V-O antisymmetric and symmetrical stretching [80], while the band at 622 cm⁻¹ probably overlapped with the clay mineral vibrations. FTIR spectra of the amiloride-adsorbed samples (Figure 5) showed the presence of the drug on solids. All of the absorption of the inorganic part was present. The broad and intense band at 3420 cm⁻¹ is attributed to the N-H symmetrical stretching of the primary amine substituted in the pyrazine ring [81]. Two other broad bands were observed at 3339 and 3177 cm⁻¹, which were assigned to N-H asymmetric and symmetric stretching of guanidine, respectively [81,82]. Other characteristic bands were observed at 1676 cm⁻¹, corresponding to the C=O stretching of the disubstituted amide; 1642 cm⁻¹, attributed to the NH₂ deformation of the guanidine; 1543 cm⁻¹, associated with stretching of the tetrasubstituted pyrazine ring; 1383 cm⁻¹, attributed to the C-N stretching of the carboxamide; 1065 and 1250 cm⁻¹, attributed to the ortho-substituted chlorine and the C-N group in the pyrazine ring, respectively [81,82]. The bands between 841/845 and 622 cm⁻¹, assigned to V-O antisymmetric and symmetric stretching [80], were unchanged after amiloride.

3.1.5. SEM and EDX Analysis

The morphologies of the samples before and after amiloride adsorption were examined by SEM, as shown in Figure 6i and Supplementary Materials Figure S1i, respectively. Raw Bent presented a typical flake morphology, which remained unchanged after vanadyl exchange, independent of the specific composition. For the amiloride-loaded samples, the morphology was the same, indicating that solids were stable under the conditions used (Figure S1i).

EDX spectra and elemental mapping of the samples are presented in Figure 6ii and Figure S1ii, respectively. For samples before adsorption, EDX spectra showed the presence of the main elements in raw Bent (Si, Al, Mg, O) and vanadyl bentonites (Si, Al, Mg, O, V). Furthermore, EDX vanadium mapping (Figure S2) indicated that the element was uniformly distributed in the samples before and after amiloride adsorption.

3.2. Mechanism of Amiloride Clay Interaction

Based on characterization, the interaction between the drug and the clay mineral surface is mostly electrostatic, resulting in a cation exchange process with hydrated vanadium cations [61]. The drug can also interact by hydrogen bonding with the nitrogen and oxygen in its interlayer water molecules and the terminal OH of the phyllosilicate structure [76]. However, considering the stability formed by the vanadyl ion coordinated by water molecules, [78] the coordination by amiloride in the vanadyl ions is unlikely, as indicated by the infrared spectra. A preferable coordination by the water molecules in relation to the drug is also possible considering the pKa values of the species involved. In fact, vanadyl ions have pKa 5.38 [78] and are stronger acids than water (pKa = 15.9) and amiloride (pKa = 8.4) [51]. Therefore, water has a higher pKa among the species and, consequently, has a stronger conjugate base, favoring coordination with the vanadium oxidation in an acid-base reaction between the vanadyl ion and ligands. Therefore, the drug molecules were probably inserted in the interlayer region parallel to the phyllosilicate sheets, considering the size of the molecule and d₍₀₀₁₎ value (Figure 7). Although bentonite is composed of expandable phyllosilicate, its basal spacing does not increase after amiloride adsorption on the modified samples. Similar results were also reported for the adsorption of thiabendazole [37] and amitriptyline [45] by bentonites.

3.3. Adsorption

3.3.1. Adsorbent Dosage

The influence of the adsorbent dosage on amiloride adsorption by bentonites was evaluated for masses between 25 to 100 mg (see Figure S3). Thus, the adsorbent concentrations in the suspension were 0.5, 1.0, 1.5, and 2 mg·L⁻¹. The results showed that the maximum removal of 89.8% was obtained with 75 mg of Na⁺-Bent, while the performance of the exchanged vanadyl bentonites improved by 96.3, 94.7, and 86.5% at 50 mg adsorbent dosage for BentV1, BentV3, and BentV5, respectively.

3.3.2. Influence of Time

The effect of time on amiloride adsorption (Figure S4) was monitored for 120 min. Fast adsorption of the drug was obtained at 20 min for Na⁺-Bent, BentV1, and BentV5 and 60 min for BentV3. Maximum amounts of the adsorbed drug were 80.7, 36.1, 91.5, and 20.6 mg·g⁻¹ for Bent, BentV1, BentV3, and BentV5, respectively. The results indicated that the presence of vanadyl on bentonite drastically affected its performance, possibly due to the blocking of the adsorption sites. The experimental adsorption data were fitted to kinetic models (Figure S4), and the results were (

Table 3. Kinetic parameters obtained from the pseudo-first-order, pseudo-second-order, and Elovich equations for nonlinear fitting of amiloride adsorption on vanadyl-bentonites (experimental conditions: 30 °C, pH 5.8, and 50 mg·L⁻¹ amiloride solution).

Pseudo-First-Order Model
Adsorbent	qteor (mg·g−1)	qexp (mg·g−1)	k1 (min−1)	R2	SD (mg·g−1)
Na+-Bent	80.10 ± 0.26	80.69 ± 0.04	0.34 ± 0.02	0.9993	0.69
BentV1	35.32 ± 0.31	36.12 ± 0.02	0.36 ± 0.03	0.9940	0.86
BentV3	86.23 ± 1.99	91.45 ± 0.04	0.28 ± 0.05	0.9624	5.33
BentV5	20.64 ± 0.09	20.64 ± 0.01	1.006 ± 0.42	0.9984	0.26
Pseudo-Second-Order Model
Adsorbent	qteor (mg·g−1)	qexp (mg·g−1)	k2 (g·mg−1·min−1)	R2	SD (mg·g−1)
Na+-Bent	80.98 ± 0.22	80.69 ± 0.04	0.026 ± 0.003	0.9998	0.39
BentV1	36.28 ± 0.23	36.12 ± 0.02	0.027 ± 0.003	0.9980	0.4
BentV3	90.86 ± 1.56	91.45 ± 0.04	0.006 ± 0.001	0.9988	3.05
BentV5	20.59 ± 0.13	20.64 ± 0.01	0.84 ± 0.04	0.9978	0.31

) better adjusted for the pseudo-second-order model for all of the investigated systems, as shown by the R² and SD values. The results obtained from characterization of the amiloride/vanadyl-bentonites hybrids indicated that chemisorption contributed to the interaction mechanism.

3.3.3. Equilibrium Isotherms

The influence of concentration on amiloride adsorption was determined to be between 10 and 500 mg·L⁻¹ at pH 5.8 and 30 °C for 150 min. The isotherms for vanadyl bentonites showed an increase in q_e, achieving constant concentrations (Ci) of 100 mg·L⁻¹ for Na⁺-Bent and BentV1, 50 mg·L⁻¹ for BentV3, and 30 mg·L⁻¹ for BentV5 (Figure 6). Maximum amounts (q_max) of adsorbed drug by Na⁺-Bent, BentV1, BentV3, and BentV5 were 457.08, 374.64, 102.56, and 25.63 mg·g⁻¹, respectively (Figure 8). There are no studies on amiloride adsorption in species exchanged with vanadyl in the literature. However, amiloride adsorption for release purposes has been found for talc phyllosilicate in its natural form and talc phyllosilicate hybridized with chitosan [57]. In that study, the modified material presented better amiloride adsorption performance, which had maximum adsorption of 55.74 mg·g⁻¹ in the hybrid and 49.53 mg·g⁻¹ in the raw-talc, both in contact with an adsorbate solution with a concentration of 1500 mg·L⁻¹ [57]. The obtained adsorption capacity was also higher than the one observed for chitosan (4.71 mg·g⁻¹) [57].

The adsorption parameters were evaluated according to the Langmuir, Freundlich, and Temkin adsorption models (

Table 4. Adsorption parameters of amiloride on vanadyl-bentonites at 30 °C and pH 5.8 according to the Langmuir, Freundlich, and Temkin models.

Langmuir
Adsorbent	qteor (mg·g−1)	qmax (mg·g−1)	k (10−1 L·mg−1)	R2
Na+-Bent	457.08 ± 7.56	434.53 ± 2.17	0.081 ± 0.01	0.9948
BentV1	374.64 ± 26.88	343.46 ± 1.71	0.115 ± 0.03	0.8972
BentV3	102.56 ± 5.90	91.61 ± 0.04	0.637 ± 0.12	0.9287
BentV5	25.63 ± 0.76	21.2 ± 0.01	0.316 ± 0.04	0.98767
Freundlich
Adsorbent	n	kf (mg·g−1) (mgL·mg−1)−1/n	R2	SD (mg·g−1)
Na+-Bent	3.39 ± 0.43	91.14 ± 16.57	0.9282	47.76
BentV1	3.24 ± 0.58	74.72 ± 18.36	0.7727	64.92
BentV3	4.70 ± 1.06	47.96 ± 5.81	0.7396	16.21
BentV5	2.87 ± 0.43	8.21 ± 1.09	0.9347	1.90
Temkin
Adsorbent	bT (102 J·mol−1)	AT (L·mg−1)	R2	SD (mg·g−1)
Na+-Bent	0.01 ± 0.001	2.23 ± 0.60	0.9581	36.47
BentV1	0.01 ± 0.002	1.24 ± 0.45	0.8398	54.51
BentV3	0.06 ± 0.01	11.07 ± 7.80	0.7872	463.20
BentV5	0.18 ± 0.01	2.84 ± 0.71	0.9679	1.33

). From the R² values, the experimental data were better fitted by the Langmuir model for all of the studied solids, indicating that chemisorption may be the predominant mechanism involved in amiloride adsorption. The Langmuir model is based on homogeneous adsorption sites.

As observed in the kinetic study, the adsorption was lower by the vanadyl-exchanged bentonites. However, Na⁺-Bent presented a better performance in comparison to the other solids. Equilibrium occurred at higher drug concentration, while at lower amiloride concentration (20–70 mg·L⁻¹), the performance of the adsorbent followed the order BentV3 > Na⁺-Bent > BentV1 > BentV5. The adsorbed capacities were 15.75 and 37.14 mg·g⁻¹ at 20 mg·L⁻¹, 26 and 56.6 mg·g⁻¹ at 30 mg·L⁻¹, and 48.8 and 91.2 mg·g⁻¹ at 50 mg·L⁻¹ for Bent and BentV3, respectively. The maximum performance occurred for BentV3 at 50 mg·L⁻¹, while Na⁺-Bent increased its performance at an initial drug concentration of 70 mg·L⁻¹. Therefore, Na⁺-Bent was less efficient at lower amiloride concentrations, which are typical drug occurrence conditions in aquatic bodies [1,19,83].

4. Conclusions

The incorporation of vanadyl in sodium bentonite was performed at different proportions of the CEC. However, partial incorporation of the initial vanadyl was observed.

The difference in the proportion of VO²⁺ ions resulted in different behaviors in the amiloride adsorption, and the material prepared at 100% CEC showed better performance, with maximum adsorption of 343.46 mg·g⁻¹ at 500 mg·L⁻¹ initial amiloride concentration in an equilibrium time of 20 min.

At lower amiloride concentrations, BentV3 showed better performance than all of the solids, including Na⁺-Bent at 20–70 mg·L⁻¹. The adsorption of Na⁺-Bent was higher than the exchanged bentonites at concentrations above 100 mg·L⁻¹.