Cr(III) Adsorption on Green Mesoporous Silica: Effect of Amine Functionalization and pH

Abstract

Contamination of heavy metals, particularly chromium from industrial activities, represents a critical challenge for public health and the environment. The aim of this study is to evaluate the adsorption performance of green mesoporous silica (GMS-24 h), synthesized through a sustainable process from industrial sodium silicate, and its derivative functionalized with amino groups (GMS-24 h–NH₂) for the removal of Cr(III) in aqueous systems. FTIR and CP–MAS NMR characterization confirmed the surface modification and textural property improvement of green mesoporous silica. The adsorption experiments, carried out under varying pH and Cr(III) concentration conditions, were fitted to the Langmuir and Freundlich models. The results showed that GMS-24 h reached a maximum capacity of 303 mg·g⁻¹ at pH 3, while GMS-24 h–NH₂ achieved 370 mg·g⁻¹ at pH 5. The evaluated adsorbents showed up to a 44% increase in efficiency. Preliminary kinetic studies indicated that the pseudo-second-order model accurately describes the process (R² > 0.99), with the rapid stabilization of the system. Diffusion analysis indicated combined mechanisms, with an additional chelation step (N → Cr) in GMS-24 h–NH₂. These findings suggest that GMS–NH₂ has the potential to be a sustainable and economical adsorbent for the remediation of wastewater from the tanning industry in León, Guanajuato, Mexico.

1. Introduction

Environmental remediation is one of the greatest challenges facing humanity today. Water pollution from heavy metals continues to be a priority issue, as these elements, being non-biodegradable, accumulate in the food chain and cause various complications in living beings. Heavy metals are highly toxic and carcinogenic pollutants. The most common heavy metals are lead, arsenic, cadmium, thallium, chromium, and mercury [1,2,3,4].

Chromium is one of the twenty most abundant elements in the Earth’s crust. However, it is highly toxic to plants, animals, microorganisms, and humans due to its carcinogenic and mutagenic properties. Its primary industrial applications include leather tanning; steel, auto part, and dye manufacturing; electroplating; paper production; mining; cement production, [5,6,7,8,9]. The most common valencies of chromium are Cr³⁺ and Cr⁶⁺; the latter is the most harmful due to its high oxidizing power.

In humans, exposure to chromium has been associated with a variety of health concerns, including lung cancer, eczema-like allergies, perforation of the nasal septum, asthma, respiratory diseases, ulcers in the nasal layers, and growth and reproduction disorders. In the case of plants, the effects include defects in roots, stems, and leaves; slowed germination; reduced photosynthesis; nutritional and oxidative imbalances; mutagenesis; and growth impairment [1,2].

In this regard, the harmful effects that chromium can cause in both water and soil are significant. The US Environmental Protection Agency (USEPA) has established a maximum permissible concentration of chromium in drinking water at 0.05 milligrams per liter (mg/L). Thus, the removal of this contaminant from wastewater is a constant challenge in environmental remediation processes due to its toxicity. A variety of methodologies have been proposed for this purpose, including ion exchange, photocatalysis, chemical reduction, electrokinetic remediation, precipitation, bioremediation, adsorption processes, etc. [1,2].

The use of adsorbent materials as a promising solution for the effective removal of chromium from aqueous systems has been proposed. The bioadsorbents employed include grapefruit peel [10], chitosan [11,12,13,14,15], mesoporous silica [16,17,18,19,20], biosilica obtained from rice husks [21,22,23,24,25], activated carbon, etc. [26,27,28,29,30,31,32,33].

The main drawback of its industrial implementation is the cost of the adsorbent. To identify cost-effective silica production methods, the synthesis of silica from sodium silicate has been investigated as a potential alternative. The process entails the acidification of the silicate, a process that results in the precipitation of amorphous silica [34,35,36].

Under acidic conditions, sodium silicate forms silicic acid (Si(OH)₄), which then polymerizes to form silica. The polymerization of Si(OH)₄ occurs in three stages: The process begins with the formation of the monomer, which then undergoes polymerization into particles. Next, the particles grow. Finally, the particles bond to form chains or networks that create a gel-like structure. Stages 2 and 3 are contingent on the pH of the medium: an acidic pH promotes gelation, while an alkaline pH promotes particle aggregation [33,36,37].

This study proposes the use of mesoporous silicas obtained through cost-effective processes, employing industrial-grade sodium silicate as an economical precursor for the formation of silicic acid. Through aggregation processes, this precursor leads to the development of the silica network. This investigation focuses on the effect of the aging time of Si(OH)₄ on the textural properties of the resulting silica, referred to as green mesoporous silica (GMS). This is because it is synthesized via a less polluting route that avoids the use of TEOS and organic solvents. GMS was compared with two typical silicas: one synthesized via simple silicate acidification (precipitated silica) and another obtained from tetraethyl orthosilicate (TEOS).

The adsorption capacity of the silicas against Cr(III) is evaluated, as this ion is frequently used as a leather tanning agent in the local shoe industry and represents a persistent pollutant in León, Guanajuato, Mexico. The maximum adsorption capacity is determined by fitting the experimental data to the theoretical Freundlich and Langmuir models. Furthermore, the impact of pH and Cr(III) concentration on the adsorption process is thoroughly examined. The aim of this study is to assess the impact of surface modification with amino groups on the efficiency of Cr(III) adsorption.

2. Materials and Methods

2.1. Experimental Procedure

2.1.1. Starting Materials

The sodium silicate used was industrial-grade, with 28.7% SiO₂, 8.9% Na₂O, and a density of 1.38 gcm⁻³. The other chemical starting material was reactive-grade from Sigma Aldrich North America, México.

2.1.2. Green Mesoporous Silica (GMS) Synthesis Using Sodium Silicate

GMS was obtained according to the report of Salazar-Hernández et al. [37]. In the reported process, 100 mL of an industrial-grade sodium silicate solution at 25% V was prepared and passed through the Dowex 50WX8 ion exchange resin to yield silicic acid (Si(OH)₄). The acid was eluted in 100 mL of deionized water, and fractions with an approximate pH of 3 were collected. Subsequently, these fractions were then exposed to an aging process for various intervals: 0, 6, and 24 h.

The concentration of Si(OH)₄ was determined with yellow silicon–molybdenum complexes, using UV–visible spectrophotometry with an SQ-2800 UNICO spectrophotometer from México. The measurements were conducted at a wavelength of 400 nm (λ = 400 nm), employing the Deutsche–Einheitsverfahren method [38].

Green mesoporous silica (GMS) was synthesized via a hydrothermal process, in which silicic acid was mixed with a solution of the triblock copolymer P–123 (HO(CH₂CH₂O)₂₀(CH₂CH(CH₃)O)₇₀(CH₂CH₂O)₂₀H (98%, Sigma Aldrich North America, México). The P–123 solution was prepared by dissolving 66 g of the copolymer in 500 mL of H₂O. Si(OH)₄ was added at a volume ratio of 1:2, and the mixture was subsequently heated at 60 °C for 24 h. The resulting solid was recovered using filtration. Subsequently, P–123 was removed via solid–liquid extraction using acidic ethanol under reflux for 24 h. Then, the material was calcined at 600 °C for 2 h.

On the other hand, mesoporous silica obtained by acidification (MS-HCl) was synthesized by adding 10 mL of concentrated hydrochloric acid (37% w; Karal S.A C.V from Leon, Gto., Mexico) to 200 mL of a 12.5% sodium silicate solution under magnetic stirring. The pH was maintained within the range of 1 to 3 until the onset of gelation was observed. Afterward, the resulting gel was washed with deionized water until a neutral pH (approximately 7) was reached; then, the gel was dried at 40 °C for 24 h.

2.1.3. Functionalization of GMS with Amine Groups

The functionalization of silica was carried out using a post-synthesis method. The GMS-24 h material was used for this purpose due to its monodisperse pore distribution and enhanced textural properties.

A mixture containing 83 mmol of GMS-24 h and 0.166 mmol of the alkylalkoxysilane modifier 3-aminopropyl(triethoxy)silane (APTES; 99%, Sigma-Aldrich North America, México) was prepared in 75 mL of ethanol, using NH₄OH as a condensation catalyst. The reaction refluxed for 24 h. The resulting solid was then recovered using filtration, washed with ethanol, and dried at 90 °C for 12 h.

2.1.4. Characterization for GMS and GMS-Amino

GMS and GMS–NH₂ were characterized with Fourier-transform infrared (FTIR) spectroscopy, using an ATR-FTIR Nicolet iS10 (Thermo Scientific, from México), with measurements recorded in the range of 600–4000 cm⁻¹. The average number of scans collected was 32, with a resolution of 4 cm⁻¹. The FTIR analysis confirmed the effective surface modification of the silica. Furthermore, chemical characterization was performed using CP–MAS ¹³C and ²⁹Si NMR (cross-polarization magic-angle-spinning nuclear magnetic resonance of ¹³C and ²⁹Si) spectroscopy, using an AR–Premium COMPACT 600 MHz Varian from México. Talc was used as the reference material, with a spinning rate of 6 kHz, a delay time (D1) of 4 s, a contact time of 3 ms, and a total of 120 scans.

The textural properties of the samples were evaluated by N₂ adsorption–desorption isotherms at −196.15 °C using a Micromeritics ASAP 2010 instrument from México. Prior to measurement, the samples were degassed overnight at 180 °C under a vacuum of 71 mmHg. The specific surface area was calculated using the Brunauer–Emmett–Teller (BET) method, while the mean pore diameter was determined by applying the Barrett–Joymer–Halenda (BJH) method to the desorption branch of the isotherm [39,40].

The zeta potential (ζ-potential) and isoelectric point were determined using a Zetasizer Nano ZS instrument (Malvern Co. Ltd., UK) equipped with a He–Ne laser. The silica samples were dispersed in 1 mM KCl for 24 h prior to measurement. The variation in surface charge detected for MS and MS-modified samples may be associated with adsorption mechanisms and the specific interactions between the silica surface and chromium.

On the other hand, thermogravimetric analysis (TGA/DTA) was performed using a PerkinElmer SII-Diamond Thermogravimetric/Differential Analyzer from México under an oxidizing atmosphere, within a temperature range of 25–1000 °C and a heating rate of 120 °C min⁻¹. Finally, transmission electron microscopy (TEM) micrographs were obtained by suspending the silica samples on gold grids and imaging them with a FEI Talos F200X G2 TEM microscope from México.

2.1.5. Cr(III) Adsorption Studies

The Cr(III) adsorption study was conducted using stock solutions ranging from 247 to 1636 mg·L⁻¹. These stock solutions were prepared by dissolving the Cr₂(SO₄)₃ powder (Karal; 99%) in deionized water. Adsorption experiments were performed in triplicate using a batch method at room temperature (25 °C) under constant stirring at 24 rpm. A total of 0.2 g of GMS or GMS-NH₂ was added to 10 mL of aqueous Cr(III) solution in varying concentrations. The residual chromium in the solution was analyzed using a UV–visible spectrometer (UNICO SQ-2800), with measurements recorded at 10-min intervals over a 60-min period.

The amount of Cr(III) adsorbed at time t (q_t) was calculated using Equation (1), where C₀ and C_t represent the initial and time-dependent concentrations of Cr³⁺ in the solution (mg–L⁻¹), m denotes the mass of the adsorbent used (g), and V represents the volume of the solution (L). Furthermore, the removal efficiency was calculated using Equation (2). All adsorption studies were performed in triplicate.

$$q_t={\displaystyle \frac{\left(C_0-C_t\right)V}{m}}$$

(1)

$$Removal(\mathrm{\%})={\displaystyle \frac{C_0-C_e}{C_0}}X100$$

(2)

3. Results

3.1. Effect of Aging on Textural Properties of GMS

Figure 1a shows the behavior of the surface area as a function of silicic acid concentration in aged solutions. It was observed that, as the aging time increased, the concentration of silicic acid remaining in the solution decreased. This phenomenon is attributable to the agglomeration of particles during the gel formation process, which is in accordance with the silica precipitation mechanisms of silicates [37].

The aging time showed an increase in the surface area up to 24 h, at which point an increase of approximately 12% in SBET was recorded (SBET = 496.5 m² g⁻¹). However, after 48 h, the SBET decreased to 466 m² g⁻¹. Similar trends were observed for pore volume (see Figure 1b). The initial volume of 0.34 cm³/g was measured without aging, and it increased to 0.55 cm³ g⁻¹ after 24 h of aging. Then, it decreased to 0.49 cm³ g⁻¹ in 48 h.

Figure S1 in the Supplementary Materials shows the N₂ adsorption isotherm. According to the IUPAC classification [41], the silicic acid sample without aging (GMS-0 h) shows a mixed isotherm profile, corresponding to type VI. At a relative pressure of approximately 0.72, a small plateau of condensation is observed, indicative of a type IV isotherm (d), which is characteristic of mesoporous materials. The aging time modulates this plateau of condensation (mesoporous fragment): for GMS-24 h, the plateau appears at a relative pressure of 0.7, while for GMS-48 h, it shifts to 0.8. These results suggest that the aggregation process promoted by aging time leads to the formation of mesoporous fragments and additional pore structures.

After 24 and 48 h of aging, the samples display typical type IV isotherms, with capillary condensation occurring at a relative pressure of 0.7 for the 24 h aged sample. At 48 h, a similar isotherm profile is observed; however, a shift in the hysteresis loop is noted, and capillary condensation occurs at a slightly higher relative pressure of 0.8.

Figure 2 shows the modal pore distributions observed; these vary depending on the aging time. The silicic acid sample without aging showed a bimodal pore distribution, with a dominant peak at 3.9 nm (corresponding to the mesoporous fraction) and a broader secondary distribution ranging from 5 to 20 nm. After 24 h of aging, the pore structure shifted to a unimodal distribution centered at 3.9 nm. In contrast, the sample aged 48 h showed a bimodal distribution again, with peaks at 3.9 nm and 7.8 nm.

Table 1. Comparison of textural properties of GMS with other mesoporous silicas.

	[Si(OH)4]; (mg L−1)	ABET [m2 g−1]	Average Pore Volume [cm3 g−1]	Pore Diameter (BJH) [nm]
GMS-0 h	626.1 ± 15	444.5 ± 4.9	0.341 ± 0.02	3.9; broad between 5 and 20
GMS-6 h	265.9 ± 12	491.4 ± 4.6	0.416 ± 0.015	3.9
GMS-24 h	97.7 ± 11	496.5 ± 2.6	0.552 ± 0.015	3.9
GMS-48 h	52.9 ± 8	466 ± 4.2	0.497 ± 0.023	3.9 and 7.8
SiO2-HCl *	–	13.27	0.052	–

* non-porous structure.

presents a comparative analysis of the textural properties of GMS samples with varying aging times, silica obtained from sodium silicate precipitated under acidic conditions (SiO₂–HCl), and mesoporous silica (MS) synthesized from TEOS using Pluronic as a pore-directing agent. The textural properties were obtained by adsorption–desorption N₂ studies using the BET methodology and the BHJ model from pore distribution; the average of the characterization of three samples was reported.

Acidification of sodium silicate results in the formation of non-porous materials due to the agglomeration of silica particles into gel-like structures. In contrast, the aging of silicic acid has been shown to enhance porosity, yielding materials with textural properties comparable to nano-material silica synthesized from TEOS, which typically exhibits A_BET values between 25 and 860 m² g⁻¹ [42].

According to a previous study of these materials reported by other authors [37], the TEM micrographs of GMS samples were aged for 0 and 24 h (Figure 3). The sample without aging (Figure 3a) shows thin, poorly ordered silica layers. In contrast, the sample aged for 24 h (Figure 3b) shows a more ordered stacking of mesoporous silica layers and pore sizes of approximately 4 nm, which closely matches the average diameter determined by the BJH analysis (3.9 nm). These results suggest that aging Si(OH)₄ promotes structural ordering in the mesoporous silica framework. This may be due to the increased size of the oligomers present in the aged stock solution. This could improve the organization of molecules during the formation of the molecular sieve.

3.2. GMS Modification with Amine Groups

The GMS-24 h sample showed a monodisperse pore distribution with enhanced textural properties. Thereafter, the functionalization process involved the addition of amine groups, as shown in Figure 4a. Figure 4b shows the FTIR characterization of GMS. This spectrum confirms the presence of an inorganic silica matrix (SiO₂), with characteristic vibrational bands at 1073 cm⁻¹ (ν Si–O), 947 cm⁻¹ (δ Si–O), and 784 cm⁻¹ (δ Si–OH), corresponding to the SiO₂ lattice (signals marked with a red line). Additionally, adsorbed water is identified by bands at 3483 cm⁻¹ (ν O–H) and 1674 cm⁻¹ (δ O–H) (signals marked with a blue line) [43].

As shown in Figure 4c, the presence of amine groups in GMS–NH₂ is confirmed by bands at 1556 cm⁻¹ (δ N–H) and|cm⁻¹ (ν N–H) (signals marked with a green line). The presence of signals at 1200 cm⁻¹ (Si–C) and 2936–2856 cm⁻¹ (ν C–H) (signals marked with a purple line) suggests alkyl chain incorporation. The silica network is also evident, with bands at 1075.4 cm⁻¹ (ν Si–O–Si), 796 cm⁻¹ (δ Si–O), 945 cm⁻¹ (δ Si–OH), and 698 cm⁻¹ (Si–(CH₂)_n) (signals marked with a red line). Adsorbed water is observed at 3361 cm⁻¹ (O–H) and 1637 cm⁻¹ (δ O–H) (signals marked with a blue line) [43,44].

Figure 5 shows the ¹³C CP–MAS NMR spectra of the GMS-24 h samples that were functionalized with either amine or thiol groups. The spectra confirm the successful crosslinking of the silica surface with the organic modifiers. Three characteristic signals corresponding to methylene groups (–CH₂–) were identified for the GMS-24 h–NH₂ sample. As shown in Figure 5b, the spectrum of ²⁹Si indicates the signal corresponding to the Si–C bond at −67.97 ppm and the siliceous matrix at −101.3 and −110.2 ppm, corresponding to Si types Q² and Q⁴, respectively. This demonstrates the presence of hydrated silica (Q² signal) modified with propyl amino groups.

Figure 6 shows the nitrogen adsorption isotherms and pore size distributions of GMS-24 h samples that were modified with functional groups. A decrease in the volume of adsorbed N₂ was observed, particularly in the sample modified with thiol groups. Changes in pore distribution were also identified; the original pore size of 3.9 nm shifted to 3.2 nm in the sample modified with amino groups. This shift may be attributed to the functional groups acting as an overlay on the pore walls.

Table 2. Textural properties of modified GMS.

	ABET [m2 g−1]	Average Pore Volume [cm3 g−1]	Pore Diameter (BJH) [nm]
GMS-24 h	496.5 ± 2.6	0.552 ± 0.015	3.9
GMS-24 h-NH2	171.1 ± 4.02	0.368 ± 0.012	3.2

summarizes the textural properties of these silica materials. A reduction in BET surface area and pore volume was observed because of surface modification. However, chromium adsorption may be enhanced due to chemical interactions between metal ions and functional groups, as well as pH effects. These effects are discussed in [22].

Figure 7 shows the thermogravimetric analysis (TGA) of green mesoporous silica (GMS) and the modified silica samples. This analysis reveals weight loss as a function of increasing temperature. Between 100 and 200 °C, a loss of adsorbed solvents from the solid surface is observed, and between 300 and 600 °C, the decomposition of organic matter occurs.

Ceramic yield, defined as the percentage of SiO₂ retained in the solid matrix, was 90% for GMS. In contrast, all amine-modified silica samples exhibited a lower ceramic yield of approximately 75%. This reduction is attributed to the presence of organic functional groups grafted onto the silica surface. The degree of surface modification was calculated using Equation (3), resulting in a value of 26 wt%.

$$\%modified={\displaystyle \frac{\int organicweigth loss}{\int total weigthloss}}$$

(3)

The GMS-24h sample showed a large negative surface potential throughout a wide pH range, with an isoelectric point (IEP) around pH 2.0. This behavior is attributed to the presence of deprotonated silanol groups on the silica surface [45]. In contrast, the GMS-24h–NH₂ sample showed a positive surface charge as a function of pH, with an IEP near pH 9.5; as previously the authors have been reported [45]. The effect of this surface charge shift on adsorption capacity was further evaluated.

3.3. Cr(3+) Removal Using GMS-24 h and GMS-24 h-NH2

Cr³⁺ removal studies were conducted using concentrations ranging from 200 to 1600 ppm, which are representative of those found in tannery effluents at various stages of processing [6,9,10,22]. As shown in Figure 8, the pH level had a significant impact on chromium adsorption, with the highest removal rates observed at a pH of 5 for both silica materials. For GMS-24 h, removal efficiencies ranged from 38 to 79% at a pH of 1, 46–78% at a pH of 3, and 43–71% at a pH of 5. In contrast, the amine-modified silica (GMS-24 h–NH₂) showed a removal efficiency of 42–55% at a pH of 1, 33–55% at a pH of 3, and 79–94% at a pH of 5, indicating near-quantitative removal.

These results suggest that electrostatic attraction between the material surface and Cr³⁺ is not the main adsorption mechanism. Instead, coordination between the amine group (–NH₂) and Cr³⁺ appears to play a key role, particularly at pH 5, where the surface charge is less positive, enabling cation approach and interaction.

pH is a key variable that enhances the adsorption capacity of silica materials. This effect can be attributed to the interactions between chromium species in a solution and the surface characteristics of GMS-24 h and GMS-24 h–NH₂. Cr(OH)₃ precipitation occurs at a pH of 6.0. At lower values, the dominant chromium species at the studied pH levels are Cr³⁺ and hydroxy complexes, such as [CrOH]²⁺, [Cr₃(OH)₄]⁵⁺, and [Cr₂(OH)₂]⁴⁺, as shown in the Cr(III) speciation diagram obtained using SHC (Figure 9a).

Furthermore, according to the measurements of the ζ-potential [44], GMS-24 h shows a negative surface charge above pH 2, while GMS-24 h–NH₂ maintains a positive surface charge across all pH values that were evaluated. The corresponding structural representations for GMS-24 h are shown in Figure 9b.

According to the measurements of the ζ-potential for GMS-24 h at pH 1, the silica surface shows a slight positive charge. Therefore, the electrostatic repulsion between Cr³⁺ ions and protonated silanol groups (–OH₂⁺) on the silica surface limits the material’s adsorption capacity. Furthermore, at higher pH values of 3 and 5, the surface charge becomes increasingly negative (as indicated by a decrease in the ζ-potential from −18 mV to −30 mV). Under these conditions, the electrostatic attraction between the negatively charged surface and Cr³⁺ ions, as well as chromium hydroxy complexes, enhances adsorption. This effect becomes more evident as the negative surface charge increases (Figure 9c).

In contrast, GMS-24 h–NH₂ exhibits a positive surface charge at a pH of 1, 3, and 5, with a ζ-potential of approximately +40 mV. In these conditions, the electrostatic repulsion between the positively charged surface and Cr³⁺ ions would, in theory, hinder adsorption. However, the material demonstrates moderate chromium removal (≈50%) at a pH of 1 and 3 and near-quantitative removal at a pH of 5. These results indicate that adsorption in GMS–24 h–NH₂ is governed primarily by coordination interactions between Cr³⁺ ions and amino groups on the surface, rather than by electrostatic attraction.

This observation contradicts the one made for GMS-24 h, where adsorption is predominantly driven by electrostatic interactions (ion exchange) between the negatively charged surface and chromium species (Figure 9d).

Adsorption data were evaluated using the Langmuir and Freundlich isotherm models, which are commonly employed to describe adsorption processes. These models provide key properties, including the maximum adsorption capacity (Q_max) and the adsorption constants (K_L and K_F). Furthermore, they enable the differentiation between monolayer adsorption on homogeneous sites (Langmuir model) and multilayer adsorption on heterogeneous sites (Freundlich model). Additionally, the Gibbs free energy change (ΔG) serves as an indicator of the process’s spontaneity, revealing whether adsorption is favorable and exothermic [44,45,46,47]. The adsorption change, expressed as the normalized standard deviation (Δq), was calculated according to Equation (4). This parameter quantitatively compares the pertinence of each isotherm model for describing the silica adsorption system.

$$∆q(\mathrm{\%})=100X\sqrt{{\displaystyle \frac{\sum {(\frac{q_{exp}-q_{cal}}{q_{exp}})}^2}{N-1}}}$$

(4)

N represents the number of data points, while q_exp and q_cal, in mg·g⁻¹, denote the experimental and model-predicted amounts of Cr³⁺ adsorbed at equilibrium, respectively.

Figure 10 shows the experimental data fitted to adsorption models as a function of pH. A better fit was observed for both silica materials at pH 1 and 3 using the Freundlich model, while an excellent fit was obtained for all adsorption models at pH 5.

Table 3. Cr³⁺ adsorption parameters for GMS-24 h and amine-modified GMS-24 h–NH₂.

Parameter		pH 1		pH 3		pH 5
		GMS-24 h	GMS-24 h-NH2	GMS-24 h	GMS-24 h-NH2	GMS-24 h	GMS-24 h-NH2
Langmuir	Q0 [mg g−1]	72.99	263.16	303.03	48.31	95.24	370.37
	KL [L mg−1]	0.0042	0.00034	0.00039	0.00097	0.0024	0.0045
	R2	0.9632	0.9568	0.8783	0.8675	0.9099	0.9052
	Δq(%)	0.1389	0.1832	0.313	0.1967	0.00012	0.09134
	G [KJ mol−1]	3.59	9.79	9.46	7.23	5.04	3.42
Freundlich	KF[(mg g−1)/(mg L−1)]1/n	4.21	0.042	3.843	0.00197	1.823	3.653
	1/n	0.385	1.14	0.832	1.63	0.527	0.744
	R2	0.9929	0.956	0.9666	0.9907	0.9842	0.9945
	Δq(%)	0.0718	0.3552	0.2199	0.1068	0.1269	0.0561

summarizes the adsorption parameters derived from each model. The Gibbs free energy of the adsorption process was calculated using Equation (5), where ${\mathrm{k}}_C^0$ denotes the equilibrium constant determined as bQ_max according to the Langmuir parameters [37,46]. The resulting negative Gibbs free energy values indicate that the adsorption process is spontaneous in all cases.

Additionally, the separation factor (R_L) was used to assess whether the adsorption process is favorable or unfavorable. R_L is calculated using Equation (6), where C₀ represents the initial concentration (mg–L⁻¹) and K_L represents the Langmuir constant. An R_L value between 0 and 1 indicates favorable adsorption, while an R_L value greater than 1 suggests unfavorable adsorption. Furthermore, an R_L value of 1 corresponds to linear adsorption, and an R_L value of 0 indicates irreversible adsorption [48,49].

$$∆G^0=-RTln{(55.5k}_L)$$

(5)

$$R_L={\displaystyle \frac{1}{1+K_L{C}_0}}$$

(6)

Figure 11 shows the R_L coefficients for chromium adsorption on silica materials. Favorable adsorption was observed for both GMS-24 h and GMS-24 h–NH₂, with R_L values less than 1 in all cases. R_L values for GMS-24 h ranged from 0.49 to 0.13 at pH 1 and from 0.62 to 0.20 at pH 5. As the concentration of Cr³⁺ increased, the R_L values approached zero, indicating a tendency toward irreversible adsorption. However, this behavior was not observed at pH 3, where R_L values ranged from 0.91 to 0.60.

The presence of amine groups on the silica surface modified the R_L values, maintaining favorable adsorption while shifting away from irreversibility. For GMS-24 h–NH₂, the R_L values ranged from 0.922 to 0.12; the lowest R_L value (0.12) was observed at pH 5. According to the Freundlich coefficient 1/n, GMS-24 h shows values below 1, indicating favorable adsorption. In contrast, GMS-24 h–NH₂ showed 1/n > 1 at pH 1 and 3, suggesting heterogeneous and unfavorable adsorption under these conditions. At pH 5, however, 1/n < 1, indicating favorable adsorption.

These findings confirm that the maximum adsorption capacity for both silica materials occurs at pH 5, while moderate adsorption (approximately 55%) is observed at pH 1 and 3.

Figure 12 compares the Q_max values obtained from the Langmuir model for GMS-24 h and GMS-24 h–NH₂ with those of other silica-based materials reported in the literature. Fe₃O₄@nSiO₂@mSiO₂/EDTA, silica modified with magnetite and the complexing agent EDTA [50], showed a Q_max of 30.59 mg·g⁻¹ at pH 3—significantly lower than the values reported in this study for GMS-24 h (303.03 mg·g⁻¹) and GMS-24 h–NH₂ (48.31 mg·g⁻¹) at the same pH.

Biogenic silica (SRH) has been proposed as a low-cost alternative [22]. However, the Q_max values obtained for GMS-24 h (synthesized from sodium silicate) demonstrate superior adsorption capacity. At pH 5, GMS-24 h showed a Q_max approximately six and sixteen times higher than that of SRH. The textural properties of biogenic silica can vary depending on the environmental conditions under which it is obtained, resulting in materials with either high or low porosity [36]. In contrast, the GMS synthesis method yields consistent textural properties, contributing to its reliable adsorption performance.

According to the results obtained, the presence of amine groups on the surface of GMS-24 h significantly improved chromium adsorption. The maximum adsorption capacity (Q_max) of GMS-24 h-NH₂ obtained by the Langmuir model, which indicates the maximum number of adsorbates necessary for monolayer adsorption, increased by factors of 3.6 at pH 1 and 3.88 at pH 5 compared to that of the unmodified material.

The surface modification of SRH with one amine group (SRH–NH₂) and three amine groups (SRH–Triamine) enhanced Cr³⁺ adsorption, reaching 45.45 mg g⁻¹ and 204.08 mg g⁻¹ at pH 4, respectively. In comparison, GMS-24 h–NH₂ demonstrated an adsorption capacity of 48.31 mg g⁻¹ at pH 3 and 370 mg g⁻¹ at pH 5, showing superior adsorption performance [22].

3.4. Kinetic Studies of Cr³⁺ Adsorption in Stock Solutions

The adsorption rate was analyzed using semi-empirical kinetic models: pseudo-first order (PFO), pseudo-second order (PSO), intraparticle diffusion (PID), and external particle diffusion (ED). The quality of fit for each model was evaluated using the normalized deviation parameter, Δq (%), as outlined in [51,52].

As shown in Table S1 (Supplementary Materials), the experimental data fit the PFO kinetic model. The analysis revealed that the PFO model does not adequately describe the adsorption behavior of both GMS-24 h and GMS-24 h–NH₂, with Δq values around 50%. This shows that the PFO model does not provide reliable predictions of the experimental data.

In contrast, the adsorption rate described by the PSO model demonstrated strong alignment with the experimental data, as illustrated in Table S2 (Supplementary Materials). The correlation coefficients (R²) ranged from 0.9945 to 0.9999, and the low Δq values ranged from 0.30% to 20%.

The equilibrium factor R_w was used to explain the kinetic curve behavior under the PSO model and was calculated using Equation (7).

$$R_w={\displaystyle \frac{1}{1+K_2{q}_e{t}_{ref}}}$$

(7)

In Equation (7), t_ref represents a long time in the adsorption process, K₂ represents the kinetic rate constant of the pseudo-second-order (PSO) model, and q_e represents the amount of Cr³⁺ adsorbed at equilibrium, calculated using the PSO model.

The equilibrium factor R_w is used to interpret the kinetic behavior of the system. If 1 > R_w > 0.1, the system gradually approaches equilibrium. If 0.1 > R_w > 0.01, the system approaches equilibrium smoothly. If R_w < 0.01, the system reaches equilibrium abruptly. Finally, if R_w = 1, the system does not approach equilibrium. These classifications are summarized in Table S3 (Supplementary Materials) [37].

Table S3 (Supplementary Materials) shows the kinetic parameters that were fitted to the experimental data using the PSO model. The equilibrium factor, R_w, which is evaluated at various initial Cr(III) concentrations across all studied systems, indicates that the adsorption process smoothly approaches equilibrium. This behavior corresponds to zones I and II in Table S3, suggesting that the systems show favorable adsorption kinetics and well-defined equilibrium profiles (Figure 13).

Figure S2 (Supplementary Materials) shows the chromium adsorption process analyzed using the IPD model. The presence of multiple slopes in the kinetic curves indicates that intraparticle and external diffusion processes, as well as the coordination of Cr(III) ions with functional groups on the silica surface, contribute to the overall adsorption mechanism. Weber’s diffusion model was evaluated across different pH values at an initial Cr(III) concentration of 1636 mg L⁻¹, revealing a similar number of slopes for each studied material. This suggests that the Cr(III) adsorption mechanism on silica is independent of pH. Moreover, four distinct slopes were observed for GMS-24 h and five for the amine-functionalized material (GMS-24 h–NH₂), indicating an additional chelation step between the Cr(III) ions and the amine groups on the surface (Figure 14).

4. Conclusions

• The GMS-24 h material exhibited a high chromium removal capacity at pH 3, with a maximum adsorption of 303 mg·g⁻¹. In contrast, the amine-functionalized GMS-24 h–NH₂ showed its highest adsorption capacity at pH 5, reaching 370 mg·g⁻¹.
• Compared to other green mesoporous silica materials, such as SRH and SRH-Triamine, the two modified GMS samples demonstrated an increase of approximately 44% in adsorption capacity.
• Adsorption kinetics evaluated using the pseudo-second-order (PSO) model showed excellent agreement with the experimental data, with the systems tending rapidly toward equilibrium.
• The Weber diffusion model revealed four kinetic slopes for GMS-24 h and five for GMS-24 h–NH₂, suggesting a similar adsorption mechanism for both materials. The additional slope observed in GMS-24 h–NH₂ is attributed to the chemical coordination step between Cr(III) and surface amine groups (N → Cr).
• No significant changes in the adsorption mechanism were observed with respect to pH. For GMS-24 h, the highest adsorption at pH 3 is associated with ion exchange or electrostatic interaction as the dominant step. At pH 1 and 5, analyte diffusion toward the surface was the predominant mechanism.
• For GMS-24 h–NH₂, the diffusion of Cr(III) to the surface was identified as the dominant step across all studied pH conditions.
• GMS-24 h-NH₂, due to its excellent adsorption capacity for Cr³⁺, can be effectively applied for removing chromium from wastewater originating from the tannery industry in León, Guanajuato, Mexico.