Aminated Spherical SiO₂ Synthesized from Fly Ash and Its Application for Pb²⁺ and Cu²⁺ Sorption

Abstract

Fly ash was utilized as raw material for the preparation of spherical SiO₂ (SS), which was subsequently ammonified using APTES (H₂NCH₂CH₂CH₂Si(OC₂H₅)₃) to obtain aminated spherical SiO₂ (SSN). The physicochemical properties of SS and SSN were systematically characterized. Notably, SS exhibited a remarkable specific surface area and pore volume, enabling it to accommodate abundant nitrogen-containing groups. These functional groups served as crucial active sorption sites, significantly enhancing the sorption capacity of SiO₂ for Pb²⁺ and Cu²⁺ ions. Thus, the removal efficiency was above 99.9% when using dosages of 4 and 6 g/L SSN in solutions containing 200 mg/L of Pb²⁺ and Cu²⁺, respectively. Additionally, SSN showed a higher theoretical maximum sorption capacity for Pb²⁺ and Cu²⁺ ions, as determined by the Langmuir isotherm model, with values of 185.2 mg/g and 86.2 mg/g, respectively. These results surpass those reported in previous studies on adsorbents derived from fly ash. The chemical reactions that occurred between the aqueous cations and nitrogen-containing groups were identified as the pivotal factors governing the sorption of Pb²⁺ and Cu²⁺. This study presents a practical approach to fly ash utilization, along with the effective removal of Pb²⁺ and Cu²⁺ from water.

1. Introduction

Heavy metal ions, particularly Pb²⁺ and Cu²⁺, are significant water contaminants due to their widespread distribution, non-biodegradability, and toxicity, which can result in varying degrees of harm to human health. [1]. Various strategies have been developed for the removal of these toxic metal ions from water, including chemical precipitation, ion exchange, membrane filtration coagulation/flocculation, electrochemical treatment, and sorption [2]. The sorption method has gained increasing attention due to its straightforward operation and exceptional removal efficacy [3]. However, the exorbitant cost of conventional adsorbents poses a significant challenge that must be addressed [4,5]. Therefore, the investigation of low-cost adsorbents synthesized from industrial waste has garnered significant interest [6,7,8].

Fly ash (FA), a prominent industrial solid waste generated in metallurgy, thermal power, the chemical industry, and other sectors, not only occupies significant space in the natural environment, but also poses severe respiratory health risks due to its propensity for dusting. Therefore, finding a rational solution for the disposal of this industrial solid waste is an urgent issue that demands attention [9,10]. One approach to address the disposal of FA is to enhance its added value through scientific research for its application elsewhere [11]. A proven method for the treatment of fly ash involves utilizing it as an economically viable raw material for synthesizing silicon-based materials due to its abundant Si component. Among silicon-based materials, spherical SiO₂ exhibits promising characteristics as a porous material, due to its exceptional structural stability and high specific surface area, which signifies its immense potential in pollutant sorption [12,13,14]. However, despite possessing a competent pore structure, pure SiO₂ consistently exhibits inadequate sorption capacity for heavy metal ions [15]. McManamon et al. [16] synthetized spherical SiO₂ with a considerably high specific surface area (871 m²/g). However, the sorption capacity of Pb²⁺ was limited to only 18 mg/g. This poor sorption ability is attributed to the insufficient presence of surface active complexation sites, resulting in predominantly electrostatic or hydrogen bond interactions for bonding most heavy metal species [6]. Therefore, it is imperative to employ an effective modification method for enhancing the sorption capacity of SiO₂. Based on previous studies, the amino (-NH₂) group has demonstrated a significant complexation ability towards heavy metal species, such as Pb²⁺, Cu²⁺, HCrO⁻, and Cr₂O₇²⁻ [17,18,19]. Meanwhile, a previous study has demonstrated the viability of ammoniation modification for SiO₂, indicating its effectiveness [12]. Hence, in this work, we propose a novel approach wherein fly ash is utilized as a silicon source and the extraction of silicon is achieved through high-temperature activation with Na₂CO₃, followed by acid leaching and alkaline leaching. Subsequently, P123 is employed as a template agent for the synthesis of spherical SiO₂. And then, the SiO₂ surface is modified using 3-aminopropyl-tri-ethoxy-silane (APTES) to introduce nitrogen-containing groups. Furthermore, the physicochemical properties, and the Pb²⁺ and Cu²⁺ sorption ability of spherical SiO₂, and the ammoniation of SiO₂, are analyzed.

The present study employs the one-step reflux condensation method to modify industrial waste-derived SiO₂ in order to obtain aminated SiO₂, which exhibits high efficiency in adsorbing Pb²⁺ and Cu²⁺ ions from aqueous solutions.

2. Experiments

2.1. Materials

Fly ash (FA) was collected from the thermal power plants operated by Inner Mongolia Datang International Togtoh Power Generation, Co., Ltd. (Hohhot, China).

The chemicals, namely sodium carbonate (Na₂CO₃), sodium hydroxide (NaOH), nitric acid (HNO₃), and hydrochloric acid (HCl), were procured from China National Medicines Co., Ltd., Shanghai, China. The 3-aminopropyl-tri-ethoxy-silane (APTES, Mw = 221.37) was obtained from Shanghai Macklin Biochemical Co., Ltd., Shanghai, China. Copper sulphate (CuSO₄) and lead nitrate (Pb(NO₃)₂) were purchased from Tianjin Fengchuan Chemical Reagent Technology Co., Ltd., Tianjin, China. All the chemical reagents used in this study were of analytical reagent grade and required no further purification.

2.2. Preparation of the Samples

Fly ash (FA) weighing 100 g was mixed with 100 g of Na₂CO₃ powder and, subsequently, subjected to calcination in a muffle furnace at a heating rate of 25 °C/min, reaching a peak temperature of 900 °C, and held for a duration of 2 h. The resulting product was designated as FAC.

Then, 100 g FAC was soaked in 1000 mL of 50% HNO₃ for 12 h, and the filtering residue (named SP) was collected and utilized as a silicon precursor after vacuum filtration.

The SP (10 g) was mixed with 100 mL of NaOH solution (15% mass fraction), followed by heating in an oil bath at 100 °C for 1 h. Subsequently, the mixture was separated using high-speed centrifugation and the resulting supernatant was collected. Then, the supernatant (7.4 g) was mixed with deionized water (22 g) and added, dropwise, to a solution composed of P123 (polyethylene oxide–polypropylene oxide–polyethylene oxide, 2 g) and 2 mol/L HCl (60 g). The resulting mixture was then transferred to a high-pressure autoclave and heated at 100 °C for 24 h. After natural cooling, the product was washed, filtered, dried, and ground. The obtained SiO₂ material was designated as SS. The aminated silica material was synthesized as follows: a mixture of 0.5 g of SS, 50 mL of absolute alcohol, and 1.25 mL of APTES (H₂NCH₂CH₂CH₂Si(OC₂H₅)₃) was subjected to ultrasonic oscillation for 50 min. Subsequently, the mixture was stirred and refluxed at 50 °C for 6 h. Afterwards, the suspension underwent five washes with absolute alcohol, followed by centrifugation in order to achieve separation. The resulting powder was collected and designated as SSN.

2.3. Characterization

The raw material components (FA) were analyzed using X-ray fluorescence (XRF-1800, Shimadzu, Kyoto, Japan), with an instrument working voltage of 40 kV and a working current of 95 mA, at a flow rate of 8 mL min⁻¹. The X-ray diffraction (XRD) patterns were measured using an X-ray diffractometer (BRUKER D8, Bruker, Madison, WI, USA), with Cu Kα radiation (λ = 1.5406 Å), operating at 40 kV and 30 mA. The scanning speed was 8 °/min, and the scanning range was 0°–80°. The acceleration voltage of the scanning electron microscopy (SEM, Nova NanoSEM 450, FEI, Hillsboro, OR, USA) was 50 V–30 kV and the magnification was 20–1000 k. An automatic adsorption instrument (ASAP-2460, Micromeritics, Norcross, GA, USA) was used to determine the N₂ adsorption–desorption isotherms. The specific surface area was calculated using the Barrett–Emmett–Teller (BET) method, and the pore volume and pore size distribution were calculated using the Barrett–Joyner–Halenda (BJH) method. Fourier transform infrared (FT-IR) spectroscopy was conducted via an infrared absorption spectrometer (Nicolet 560 IR, Thermo Nicolet, Waltham, MA, USA). The scanning range of the spectrum was 4000–400 cm⁻¹. The elemental composition of the fresh and used adsorbents was analyzed using X-ray photoelectron spectroscopy (XPS, ESCALAB250Xi, Thermo Scientific, Waltham, MA, USA). Thermogravimetric analysis was performed under a nitrogen atmosphere with Setaram equipment (TGA-DTA, STA449F3 Jupiter, Netzsch, Bavaria, Germany); the samples were heated from 40 to 810 °C at 10 °C/min and the sample mass introduced was 25 mg. The concentrations of various elements in the solutions were analyzed by an inductively coupled plasma spectrometer (ICP-OES, Prodigy, Teledyne Leeman Labs, Hudson, NH, USA).

The point of zero charge (pHpzc) of SS and SSN was determined through the modified pH drift method. According to which 0.05 g of SS (or SSN) was put into a beaker and mixed with 50 mL of different NaNO₃ solutions, of which the concentration was 0.01 mol/L, and the initial pH was adjusted at 1, 2, 3, 4, and 5, respectively. The mixtures were stirred at room temperature for 24 h. The pHpzc accords to the pH value, where the curve crosses the line of pH_initial = pH_final.

2.4. Sorption Experiments

The Pb²⁺ and Cu²⁺ solutions with different concentrations were prepared by dissolving agent Pb(NO₃)₂ and CuSO₄·5H₂O in deionized water, in which the pH value was adjusted by 0.1 mol/L NaOH and 0.1 mol/L H₂SO₄ (or HNO₃).

The sorption experiments were conducted on sample SS and SSN. Briefly, 50 mL of Pb²⁺ and Cu²⁺ solution (200 mg/L) were mixed with 0.05 g of the adsorbent and put into conical flasks and stirred with a magnetic stirrer at room temperature for 4 h.

To investigate the effect of the contact time, sorbent dosage, initial pH, and solution concentration on the sorption behavior, further sorption experiments were carried out on SSN as a representative. The experimental conditions were as following: sorbent amount of 0.025, 0.05, 0.10, 0.20, and 0.40; initial pH of 1, 2, 3, 4, and 5; contact time of 5, 10, 30, 60, 90, 120, and 240 min; Pb²⁺ (or Cu²⁺) concentration of 20, 50, 80, 120, 150, 180, and 200 mg/g, respectively.

All of the sorption experiments were conducted in triplicate, and the average values were reported. The corresponding error bars for the values were determined.

The suspensions were separated by centrifugation, and then filtered after the sorption experiments. The concentrations of Pb²⁺ or Cu²⁺ in the solutions before and after sorption were determined by ICP-OES. The Pb²⁺ or Cu²⁺ sorption ability was evaluated using the following Equations (1) and (2):

$$q=\frac{(c_o-c_t)\times v}{m}$$

(1)

$$\eta =\frac{c_o-c_t}{c_o}\times 100\%$$

(2)

where q (mg/g) means the sorption capacity, v (L) means the solution volume, m (g) means the mass of the adsorbent, η (%) means the pollutant removal efficiency, C₀ (mg/L) means the initial concentration of the pollutant (Pb²⁺ or Cu²⁺), C_t (mg/L) means the concentration at the time t (min).

The sorption kinetic models were plotted according to pseudo-first-order model and pseudo-second-order model [20]. And the isotherm parameters were plotted according to the Langmuir and Freundlich models [21,22]. The details are provided by the following Equations (3)–(6):

$$\mathrm{Pseudo-first-order} \mathrm{model} (\mathrm{PF}):In(q_e-q_t)=In{q}_e-k_1\times t$$

(3)

$$\mathrm{Pseudo-second-order} \mathrm{model} (\mathrm{PS}):\frac{t}{q_t}=\frac{1}{k_2\times {q_e}^2}+\frac{t}{q_e}$$

(4)

where t (min) means the sorption time, Q_e (mg/g) and q_t (mg/g) mean the equilibrium sorption capacity and the sorption capacity at the sorption time, k₁ and k₂ (min⁻¹) are constant coefficients of the pseudo-first-order and the pseudo-second-order kinetic equation. With t utilized as the horizontal coordinate and ln (q_e − q_t) (or t/q_t) used as the vertical coordinate, the linear fits are plotted to the resulting data, in which the corresponding kinetic parameters are analyzed as follows:

$$\mathrm{Langmuir} \mathrm{isotherm}:\frac{c_e}{q_e}=\frac{1}{q_m\times k_l}+\frac{c_e}{q_m}$$

(5)

$$\mathrm{Freundlich} \mathrm{isotherm}:\log q_e=\log k_f+\frac{1}{n}\log c_e$$

(6)

where q_e (mg/g) is the equilibrium sorption capacity and c_e (mg/L) is the equilibrium concentration after the sorption process, q_m (mg/g) is the maximum sorption capacity, k_l and k_f are the Langmuir constant and Freundlich constant, respectively, and 1/n is the intensity of the sorption.

3. Results and Discussion

3.1. Physicochemical Properties

The chemical composition of FA and SP were determined by X-ray fluorescence (XRF), and the results are displayed in

Table 1. Average chemical composition of original fly ash (FA) and silicon precursor (SP).

Sample	Component (%)
	SiO2	Al2O3	Fe2O3	CaO	TiO2
FA	37.40	46.22	3.05	5.17	3.51
SP	79.65	4.46	0.12	0.43	2.27

; FA mainly consists of Si and Al components, with SiO₂ and Al₂O₃ content of 37.40% and 46.22%, respectively. Comparatively, the content of Al₂O₃ and other metallic oxides in SP were significantly decreased, of which the content of Al₂O₃ was reduced to 4.46%, and the contents of Fe₂O₃ and CaO were reduced to below 1%. Meanwhile, the content of SiO₂ steeply increased to 79.65%. These findings indicate that most of the impurities in FA were effectively removed through Na₂CO₃ mixed high-temperature activation combined with HNO₃ leaching, making SP a promising silicon precursor.

To further elucidate the evolving trends in the phase composition during the preparation process, X-ray diffraction analysis was conducted on samples FA, FAC, SP, SS, and SSN. The corresponding results are presented in Figure 1. The presence of a significant amount of mullite (Al₆Si₂O₁₃), along with Al₂O₃ and SiO₂, was observed in FA (Figure 1a). FAC, on the other hand, consisted of NaAlSiO₄, Na₂SiO₃, and Al₂O₃ (Figure 1b). The observed changes can be attributed to the successful transformation of Al₆Si₂O₁₃ into Al₂O₃ and NaAlSiO₄, as well as the conversion of SiO₂ into Na₂SiO₃ during the high-temperature activation process, with a mixture of Na₂CO₃ (the related reaction equations are described in Equations (7)–(9)) [23]. In the SP pattern, shown in Figure 1c, a prominent broad peak around 25° was observed, which can be attributed to the presence of amorphous SiO₂. Additionally, minor interfering peaks were exclusively detected, as shown in Figure 1c, indicating the existence of residual metallic oxides and inorganic salts. The results indicated that the removal of metallic oxide and inorganic salt from FAC, as well as the sufficient transformation of Si species into SiO₂, were achieved through a series of processes including HNO₃ leaching, filtering, and drying (the related reaction equations are described in Equations (7)–(13)) [24]. Additionally, following P123 treatment, the phase curve exhibited enhanced smoothness and the interfering peaks in regard to SS were eliminated (as shown in Figure 1c), indicating further purification of the SiO₂ component. Moreover, SSN (in Figure 1c) exhibited similar phases to SS, with slight shifting and widening observed in the characteristic peak of SiO₂, which could be attributed to NH₂ group grafting [9]. Meanwhile, based on the small angle XRD pattern of SS (in Figure 1d), a prominent characteristic peak near 1° was observed, indicating the presence of an ordered mesoporous structure. However, following ammoniation modification, this peak exhibited a noticeable reduction in intensity and shifted towards larger angles. This phenomenon may be attributed to the obstruction of the pore structure caused by residual APTES macromolecules.

Al₆Si₂O₁₃ (s) + 3Na₂CO₃ (s) + 4SiO₂ (s) → 6NaAlSiO₄ (s)+ 3CO₂ (g)

(7)

SiO₂ (s) + Na₂CO₃ (s) → Na₂SiO₃ (s)+ CO₂ (g)

(8)

2Na₂CO₃ (s) + Al₆Si₂O₁₃ (s) → 2Na₂SiO₃ (s) + 3Al₂O₃ (s) + 2CO₂ (g)

(9)

NaAlSiO₄ + 4HNO₃ → Al(NO₃)₃ + NaNO₃ + H₂SiO₃ + H₂O

(10)

Na₂SiO₃ + 2HNO₃ → 2NaNO₃ + H₂SiO₃

(11)

Al₂O₃ + 6HNO₃ → 6Al(NO₃)₃ + 3H₂O

(12)

H₂SiO₃ = SiO₂ → H₂O

(13)

The SEM images of SS and SSN are exhibited in Figure 2. According to the SEM images in Figure 2a,b, SS exhibited the successful synthesis of a spherical structure with a uniform particle size, utilizing P123 as the template agent. In contrast, the ammoniation modification of SSN maintained its spherical structure, but resulted in a larger and more uneven particle size, along with an increased agglomeration tendency (in Figure 2c,d).

To further investigate the effectiveness of ammoniation modification and explore the physicochemical properties of SS and SSN, comprehensive analysis, including FITR, XPS, and TGA, was conducted on these samples. The corresponding results are presented in Figure 3.

Figure 3a displays the FTIR spectra of SS and SSN, wherein the wide and intense bands around 3450 cm⁻¹ can be attributed to the stretching vibrations of the -OH bonds [25], the peaks observed at 1640 cm⁻¹ can be attributed to the absorption of free water molecules [26], the characteristic peaks at 1098 and 800 cm⁻¹ are indicative of the Si–O–Si bonds, while the bond stretching vibration of Si–OH is attributed to the peak observed at 963 cm⁻¹ [27]. The characteristic peak in the SSN spectra near 2950 cm⁻¹ implied the existence of -CH₂- groups apart from the residual APTES macromolecules [28,29]. In addition, the sorption peak near 1539 cm⁻¹ and 1350 cm⁻¹ belonged to the -NH₂ and -CH groups on the surface, as well as the peak near 694 cm⁻¹, which was attributed to the -NH- group that had substituted the oxygen atom in the skeleton of SSN [30]. The presence of nitrogen-containing functional groups on SS was confirmed through the successful grafting after the APTES treatment. Additionally, the ammoniation modification resulted in a significant decrease in the adsorption peak of Si–OH, indicating the effective coverage of the SiO₂ surface by the nitrogen-containing groups.

XPS analysis was utilized to determine the chemical environment of the elemental composition of Si and N on SS and SSN, and the spectra are exhibited in Figure 3b–d. The survey spectra presented in Figure 3b reveals the presence of a subtle N 1s characteristic peak within the SSN. In Figure 3c, the characteristic peaks of 401.94 eV and 399.74 eV represent the -NH₃⁺ and -NH₂ groups on the surface of the SSN [31]. In comparison, no apparent characteristic peak of the N element could be observed in SS, which further indicated that the nitrogen-containing functional groups found in the spectra of SSN were introduced by the APTES modification. The Si 2p spectra are shown in Figure 3d, in which the characteristic peak of SiO₂ (101.93 eV) has been reduced and has shifted slightly to lower the energy level after the ammoniation modification process, which could be attributed to the cover of the -NH₃⁺ and -NH₂ groups [6].

The curves of the TG and DTG analysis are shown in Figure 3, which provide information of on the composition and thermal stability. Evidently, SS exhibited three stages of weightlessness (Figure 3e). In the first stage, desorption of the absorbed water molecules and surface hydroxyl radicals occurred within the temperature range of 40 to 120 °C, resulting in a weight loss of 0.504%. This weight loss corresponded to the peak observed around 83 °C on the derivative thermogravimetric (DTG) curves. And then, in the second stage (120 to 594 °C), a weight loss of 0.241% was ascribed to the polycondensation of amorphous SiO₂. In the last stage (594 to 810 °C), a weight loss of 1.048% was observed. Meanwhile, the weight loss trend of SSN was divided into three stages (Figure 3f). The first stage, in the range of 40 to 130 °C, which corresponded to the temperature on the DTG curves of 96 °C, was ascribed to the desorption of the absorbed water molecules and surface hydroxyl radicals. This stage caused a weight loss of 2.961%. The second stage, created by the decomposition of NH₂ groups and the polycondensation of amorphous SiO₂ at 558 °C (in the range of 130 to 600 °C), caused 6.872% of weight loss, which further provided NH₂ groups on the surface of SSN. In the third stage (600 to 810 °C), the NH₂ groups have been decomposed completely, but the polycondensation of amorphous SiO₂ was still in progress and caused a weight loss of 0.939%. In summary, sample SS and SSN exhibited a huge difference in regard to their thermal stability, of which there was a weight loss of 1.793% and 10.772% in the range of 40 to 810 °C, respectively. The wide variations in weight loss may be attributed to the decomposition of nitrogen-containing groups during the heating process.

To evaluate the effect of the APTES treatment, the grafting percentage and the conversion ratio of the monomer to polymer were determined through Equations (14) and (15), which is relevant to the results from the curves in the TGA analysis [28].

$$G=\frac{m_s-m_N}{m_N}\times 100\%$$

(14)

$$Conversion=\frac{G\times W_M}{W_s}\times 100\%$$

(15)

where G (%) is the grafting percentage, m_s (%) and m_N (%) are the ash content of SS and SSN after the TGA analysis, Conversion (%) means the conversion ratio of the monomer to polymer, W_M (g) and W_s (g) are the mass of the APTES and SS used in the synthesis process.

By doing the calculations, the grafting percentage is 10.04% and the conversion ratio of the monomer to polymer is 23.80%, indicating that the nitrogen-containing functional groups have been successfully grafted onto the SSN.

The specific surface area and pore structure parameters of SS and SSN were determined, and the results are given in

Table 2. The specific surface area (SSA), total pore volume (PV), average pore diameter (PD), and zero-point potential (pH_z) of SS and SSN.

Sample	SSA (m2/g)	PV (cm3/g)	PD (nm)	pHpzc
SS	677.5	0.023	3.42	2.62
SSN	351.4	0.012	3.07	3.64

. SS shows a high specific surface area (677.5 m²/g) and pore volume (0.023 cm³/g). The competent pore structure parameter of SS might be due to its spherical structure, which indicates its efficiency in carrying nitrogen-containing groups [16]. In addition, the specific surface area was observed to decrease to 351.2 m²/g, accompanied by a reduction in the total pore volume to 0.012 m²/g and a decrease in the average pore diameter to 3.07 nm, following the ammoniation modification. The observed decrease can be attributed to the pore blockage caused by residual APTES macromolecules. The N₂ adsorption–desorption isotherms of SS and SSN are displayed in Figure 4a,b, in which both of these corresponded to type IV isotherms, indicating the mesoporous structure of SS and SSN [9]. In addition, according to the pore size distribution curves (in Figure 4a,b), sample SS and SSN showed a uniform pore distribution, which primarily ranged within approximately 3 nm. The point of zero charge (pHpzc) of SS and SSN were determined and are exhibited in Table 2. SS showed a low pHpzc (2.62), and the reported similarity of this value to that of other pure SiO₂ in previous studies can be attributed to the presence of Si–OH groups, which readily undergo deprotonation for the capture of OH- ions (as in Equation (14)) [9]. In contrast, SSN shows a noticeably higher pHpzc. The observed increase can be attributed to the introduction of -NH₂ groups, which exhibit a greater propensity for protonation in solution (as in Equation (15)) [32]. In brief, SS and SSN had a relatively low pHpzc value and, thus, could be negatively charged in water, and could easily attract heavy metal cations, in theory [32].

Si-OH + OH⁻ → Si-O⁻ + H₂O

(16)

-NH₂ + H⁺ → -NH₃⁺

(17)

3.2. Sorption Ability and Behavior

3.2.1. Batch Sorption Experiment

In order to investigate the Pb²⁺ and Cu²⁺ removal efficiency by SS and SSN, batch sorption experiments were conducted. The results showed that the sorption capacity of SS for Pb²⁺ was 2.4 mg/g, while that of SSN was significantly higher at 106.0 mg/g (Figure 4). Similarly, the sorption capacity of SS for Cu²⁺ was found to be 2.0 mg/g, whereas SSN exhibited a higher sorption capacity of 62.2 mg/g. Obviously, the ammoniation modification had significantly improved the Pb²⁺ and Cu²⁺ sorption ability of SiO₂ (SS), of which the Pb²⁺ and Cu²⁺ sorption capacity had increased 44.2 times and 31.1 times, respectively. After conducting the aforementioned analysis, SS had a low pHpzc and tended to undergo deprotonation for attracting cations. However, it lacked active sites for fixing Pb²⁺ and Cu²⁺, thereby resulting in its limited sorption capacity [16]. Although the specific surface area and pore volume of SSN had drastically decreased compared to SS, the ammoniation modification had added the -NH₂ groups to the surface and provided many of the active sorption sites for fixing Pb²⁺ and Cu²⁺ [7]. Meanwhile, it can be inferred that the Si–OH and -Si- groups exhibit limited sorption ability towards Pb²⁺ and Cu²⁺, while the nitrogen-containing groups account for the remarkable sorption capacity of SSN towards Pb²⁺ and Cu²⁺. By doing the calculations, the Pb²⁺ and Cu²⁺ sorption capacity of SSN (106.0 mg/g and 62.2 mg/g) could be converted into 0.51 mmol/g and 0.97 mmol/g, respectively. Obviously, more Cu²⁺ cations were captured than Pb²⁺ cations. This may be owing to the ionic radius of Cu²⁺ (0.73 Å), which is that much smaller than the value of Pb²⁺ (1.19 Å), for which the Cu²⁺ cations could be transferred and diffused in the pore structure of the adsorbent more easily.

3.2.2. Further Sorption Experiments

Effect of Different Factors

In order to investigate the effect of the sorbent dosage on the Pb²⁺ and Cu²⁺ removal efficiency, the dosage of SSN of 0.05, 0.1, 0.2, 0.3, and 0.4 g was adopted to carry out the sorption experiments, with 50 mL of Pb²⁺ and Cu²⁺ solution (200 mg/L), for 240 min. And as shown in Figure 5 the removal efficiency increased gradually due to the increase in the sorbent dosage, and the value of Pb²⁺ and Cu²⁺ went from 57.03% and 22.53% to above 99.9% when the sorbent dosage at 0.05 g rose to 0.2 g and 0.3 g, respectively (in Figure 6b), which might be due to the possibility that the increase in the sorbent dosage would provide more active sites [33]. The results indicated that the aminated SiO₂ synthesized from FA exhibited a remarkable ability to effectively remove Pb²⁺ and Cu²⁺ in high concentration solutions.

To investigate the impact of the contact time, sorbent dosage, solution pH, and initial concentration on the sorption behavior, a series of relevant sorption experiments were conducted using SSN. The mathematical relationship between the contact time and the Pb²⁺ and Cu²⁺ sorption ability of SSN is presented in Figure 6a. The relevant experimental conditions were as following: 0.05 g of SSN was put into 50 mL of Pb²⁺ and Cu²⁺ solutions in 200 mg/L, and the value of the sorption capacity for different contact times (for 1 to 240 min) was detected. Accordingly, the sorption capacity of Pb²⁺ exhibited rapid growth within the first 80 min, followed by a slower growth trend between 80 and 180 min. Finally, at 210 min, the maximum Pb²⁺ sorption capacity (106.0 mg/g) was achieved. Meanwhile, the Cu²⁺ sorption capacity showed a high increasing trend in the first 30 min, and then it became gentle until it reached the maximum (55.6 mg/g) at 210 min. The significant initial increase in the Pb²⁺ and Cu²⁺ sorption capacity can be attributed to the abundance of active sites and heavy metal cations present in the solution. And, subsequently, the active sites and heavy metal cations in the solution reduced continuously due to the sorption process, which significantly decreased the opportunities for contact between the sorption sites and the adsorbate.

The pH of the solution, which influences the surface charge of the sorbent and the speciation of heavy metals, plays a crucial role in the sorption process. [34]. To investigate this influence, sorption experiments were conducted using 50 mL of the Pb²⁺ and Cu²⁺ solutions, with an initial concentration of 200 mg/L, adjusted to a pH range of 1 to 5. A dosage of 0.2 g of SSN was employed for each experiment, with a duration of 240 min. The results of the experiments are displayed in Figure 6c; according to which SSN showed almost no sorption ability for Pb²⁺ and Cu²⁺ when the pH was 1. The subsequent increase in pH to 4 resulted in a sharp rise in the Pb2+ removal efficiency, exceeding 99.9%, which approached the equilibrium value for removal efficiency. Meanwhile, the Cu²⁺ removal efficiency exhibited a gradual increase to 5% within the pH range of 1 to 2. Subsequently, it experienced a sharp rise to 90.86 as the pH increased to 4 and, thereafter, demonstrated a smooth growth trend in regard to the removal efficiency with further increases in the pH. A higher pH value is beneficial for Pb²⁺ and Cu²⁺ sorption, as SSN exhibits a greater affinity for OH⁻ ions, facilitating deprotonation and cation attraction. [35].

Figure 6d illustrates the effect of the initial Pb²⁺ and Cu²⁺ concentrations on the sorption capacity, in which 0.05 g of SSN was used as an adsorbent to treat 50 mL of the Pb²⁺ and Cu²⁺ solution at pH 4 and 5, respectively, in different concentrations for 240 min. Clearly, the Pb²⁺ sorption capacity increase from 12.1 mg/g to 93.1 mg/g occurred due to the initial concentration increase from 20 mg/L to 200 mg/L; the Cu²⁺ sorption capacity increase from 10.6 mg/g to 55.8 mg/g occurred after the initial concentration increase from 20 mg/L to 200 mg/L. The sorption capacity significantly increased with the increase in the initial concentration, and no clear equilibrium value for the sorption capacity was observed in the corresponding function. This suggests that an even higher sorption capacity could be achieved at higher concentrations of Pb²⁺ and Cu²⁺ cations.

Isotherm and Kinetics of Sorption

To further study the kinetic behavior and mechanisms in Pb²⁺ and Cu²⁺ sorption, the pseudo-first-order (PF-order) and pseudo-second-order (PS-order) models were used to fit the sorption experiments, and the results are exhibited in Figure 7a,b and

Table 3. Fitted parameters of isotherm sorption models.

	Adsorbate	qe(exp) (mg/g)	Pseudo-First-Order Model			Pseudo-Second-Order Model
			k1 (min−1)	qe (mg/g)	R2	k2 (g/mg·min−1)	qe (mg/g)	R2
SSN	Pb2+	106.032	0.0088	63.8924	0.9018	0.0106	94.3396	0.9901
SSN	Cu2+	62.212	0.0303	30.7349	0.8478	0.0177	56.4972	0.9983

. When the regression coefficient (R²) of the formula PS-order is larger than this value for the PF-order, it is implied that the PS-order model fits to the sorption behavior better. The aforementioned examples demonstrate that the chemical reaction plays a pivotal role in constraining the rate of the steps involved in these sorption processes [36].

Langmuir and Freundlich isotherms were utilized to fit the linear regression sorption isotherm data according to the sorption experiments, and the results are shown in Figure 7c,d. And, as illustrated in

Table 4. Fitted parameters of the isotherm sorption models.

Adsorbent	Adsorbate	Isotherm	Parameters	R2
SSN	Pb2+	Langmuir	qm = 185.1852 mg/g, b = 0.0008 L/mg	0.9018
SSN	Pb2+	Freundlich	n = 1.3277, kf = 2.7063	0.9901
SSN	Cu2+	Langmuir	qm = 86.2069 mg/g, b = 0.0009 L/mg	0.8397
SSN	Cu2+	Freundlich	n = 1.6432, kf = 2.2991	0.9901

, the correlation coefficient (R²) of the Freundlich isotherm model is larger than the value of the Langmuir isotherm model. Apparently, the Freundlich isotherm exhibits a superior fit to the experimental data, which indicates that the adsorption of Pb²⁺ and Cu²⁺ cations onto the surface of SSN occurs through multilayer sorption [21]. In addition, the theoretical maximum Pb²⁺ and Cu²⁺ sorption capacity by SSN was 185.2 mg/g and 86.2 mg/g. In addition, the Pb²⁺ and Cu²⁺ sorption capacity of different adsorbents synthesized from fly ash, as reported in previous articles and this study, are presented and compared in

Table 5. Comparison of Pb²⁺ and Cu²⁺ sorption capacity of different fly ash-derived adsorbents.

	Adsorbate	Dosage (g/L)	pH	qe/(mg/g)	Ref
Activated coal fly ash	Pb2+	4	3.0	49.8	[37]
Silica synthesized from fly ash	Pb2+	1	5.0	90.0	[6]
NaOH- activated-coal fly ash	Pb2+	2	5.0	126.6	[38]
Aminated spherical SiO2 synthesized from fly ash	Pb2+	1	Natural pH	185.2	This work
Fly ash/magnetite material	Cu2+	10	5.0	17.4	[39]
Fly ash/steel slag-based geopolymer	Cu2+	2	Natural pH	38.4	[40]
Fly ash-derived calcium silicate hydrate	Cu2+	1	Natural pH	57.8	[13]
Aminated spherical SiO2 synthesized from fly ash	Cu2+	1	Natural pH	86.2	This work

. These findings demonstrate the potential applicability of aminated SiO₂ synthesized from fly ash for the effective sorption of Pb²⁺ and Cu²⁺.

Recycling Performance of SSN

The recycling performance of SSN was assessed through cyclic sorption–desorption experiments, using Pb²⁺ and Cu²⁺ solutions, with a concentration of 200 mg/L. The dosage of SSN for Pb²⁺ sorption was fixed at 4 g/L, while for Cu²⁺ sorption it was set at 6 g/L. After each sorption cycle, the Pb²⁺ (or Cu²⁺) adsorbed by SSN was separated with heavy metal cations using the following process: filtration, immersion, and stirring of 0.1 mol/L HNO₃ for 1 h, deionized water filtration, and final drying at 60 ℃ for 8 h. As shown in Figure 8, the sorption efficiency decreases after an increase in the cycle number. After four iterations of the sorption–desorption process, the sorption efficiency of Pb²⁺ and Cu²⁺ decreased to 49.0% and 33.3%, which might be attributed to the loss of active adsorption sites in the sorption–desorption process, as well as the pore structure collapse caused by the acidic solution that would weaken the sorption ability [18,41].

3.3. Mechanisms for Sorption

The FTIR spectra of SSN, SSN-Pb, and SSN-Cu (the Pb²⁺ and Cu²⁺ adsorbed SSN) are presented in Figure 9. Accordingly, the adsorption strength of the characteristic peaks assigned to the -NH₂ groups (at 1539 cm⁻¹) were weakened, which might be due to the complex reaction between the nitrogen atoms and the Pb²⁺ (or Cu²⁺) cations [18]. In addition, the weakening of the characteristic peaks for -NH- (694 cm⁻¹), C–N (1350 cm⁻¹), and -CH₂- was due to the surface-covering Pb (or Cu) atoms in the adsorbent [9,29]. Obviously, the nitrogen-containing functional groups were the key active sites for Pb²⁺ and Cu²⁺ sorption.

The XPS analysis was run for SSN, SSN-Pb, and SSN-Cu to determine the surface chemical environment of the elemental composition for N, Pb, and Cu, and the results are shown in Figure 10. The characteristic peaks of Pb 4f and Cu 2p are visibly observed for SSN-Pb and SSN-Cu, respectively (Figure 10a–c), indicating that the Pb and Cu cations have attached to the surface of SSN. In addition, the characteristic peaks of the -NH₃⁺ (401.94 eV) and -NH₂ (399.74 eV) groups in the N 1s pattern (Figure 10d) had shifted slightly in terms of higher energy, which could be attributed to the complex reaction between these nitrogen-containing groups and the heavy metal cations [17]. The weakening of the adsorption peak for the -NH- group (694 eV) was attributed to the surface-covering Pb (or Cu) atoms in the adsorbent. These findings illustrate that -NH₃⁺ and -NH₂ groups play a key role in Pb²⁺ and Cu²⁺ sorption.

Therefore, the proposed Pb²⁺ and Cu²⁺ sorption mechanisms by SSN were as follows: In the sorption process, the surface Si–OH groups were deionized first and became negatively charged (as in Equation (7)), which attracted the heavy metal cations (Pb²⁺ and Cu²⁺) to the surface of the adsorbent. Meanwhile, the -NH₂ groups were protonated to form -NH₃⁺ (as in Equation (8)). Subsequently, the -NH₃⁺ groups were involved in the ion exchange with the Pb²⁺ and Cu²⁺ cations in order to fix them on the SSN (as in Equations (20) and (21)) [32,42]. Although the -NH₃⁺ groups showed electrostatic repulsion to Pb²⁺ and Cu²⁺, the electrostatic attraction from Si–O− could secure the Pb²⁺ and Cu²⁺ cations to the surface of the SNN, which made these chemical reactions possible. In sum, despite that the Si–O⁻ groups could attract heavy metal cations, they lacked the ability to fix them, thus they had a poor sorption ability in regard to pure SiO₂. In contrast, the ammoniation modification process had provided abundant nitrogen-containing active sites, so that SSN had the complete ability to attract and fix heavy metal cations and, thus, had efficient Pb²⁺ and Cu²⁺ removal ability.

Si-OH + OH- → Si-O⁻ + H₂O

(18)

-NH₂ + H⁺ →-NH₃⁺

(19)

-NH₃⁺ + Pb²⁺ → -NH₂Pb⁺ + H⁺

(20)

-NH₃+ + Cu²⁺ → -NH₂Cu⁺ + H⁺

(21)

4. Conclusions

Spherical SiO₂ (SS), with a high specific surface area (677.5 m²/g) and pore volume (0.23 cm^3/g), was synthesized from fly ash, and it was ammonified by APTES (H₂NCH₂CH₂CH₂Si(OC₂H₅)₃) to obtain the aminated spherical SiO₂ (SSN) in this work. According to the characterization analysis, nitrogen-containing groups (-NH₃⁺, -NH₂, and -NH-) were successfully introduced into the spherical SiO₂ without damaging its spherical structure, which significantly improved the Pb²⁺ and Cu²⁺ sorption ability. As a result of the Langmuir model, the theoretical maximum Pb²⁺ and Cu²⁺ sorption capacity of SSN was 185.2 mg/g and 86.2 mg/g. Moreover, the removal efficiency for Pb²⁺ and Cu²⁺ stayed at 49.0% and 33.3% after the cyclic sorption–desorption process was repeated four times. The efficient Pb²⁺ and Cu²⁺ sorption ability was ascribed to the nitrogen-containing groups. This work provides a strategy for utilizing solid waste as raw material to prepare aminated SiO₂ for the removal of Pb²⁺ and Cu²⁺ from water.