Facile Preparation and Characterization of Silica Nanoparticles from South Africa Fly Ash Using a Sol–Gel Hydrothermal Method

Abstract

Silica nanoparticles (SNPs) consist of several applications which include lightweight aggregates, energy storage, and drug delivery. Nevertheless, the silica reagents used in SNP synthesis are both costly and hazardous. As a result, it is critical to look for other sources of silica. For this research, a simple sol–gel hydrothermal approach is used to make SNPs from South African fly ash (SAFA). SAFA is classified as fly ash class F according to X-ray fluorescence (XRF) analysis. The wide-angle X-ray diffraction (XRD) pattern reveals the structural composition of SAFA and the amorphous phase of extracted SNPs, while Fourier transform infrared (FTIR) examination reveals the presence of silanol and siloxane groups. Basic SNPs were generally spherical with diameters of about 60 nm, according to scanning electron microscopy (SEM) and transition electron microscope (TEM) studies. The presence of SiO₂ is confirmed by energy-dispersive X-ray spectroscopy (EDX) spectrum analysis. Particle size assessment indicates particle sizes ranging from 48 nm to 87 nm in diameter, with a mean diameter of 67 nm. The application of SNPs in wastewater treatment demonstrated that they can be used to remove Cd²⁺ from an aqueous solution. This research offers new ideas for using South African fly ash in SNP manufacturing.

1. Introduction

A byproduct of burning coal in a thermal power station is fly ash. Eskom, South Africa’s national power company, generates 34 million tonnes of coal fly ash annually from the use of about 120 million tonnes of coal, of which 95% is not recycled and the remaining 5% is recycled in the cement and building materials industries [1]. Fly ash is made up of mostly spherical, solid/hollow, and amorphous pieces of tiny, powdery particles [2], and it has a low bulk density and a great amount of distinct surface area [3]. Fly ash can range in hue from a deep red to orange, blue, white, or even yellow depending on the amount of unburned carbon and iron present. Coal ash has a specific gravity that is usually about 2.0, but to a great extent, ranges from 1.6 to 3.1.

A combination of several factors, inclusive of the shape of the particle, gradation, as well as chemical composition, are accountable for this variation [4]. Based on its Si, Al, and Fe oxide material, fly ash has been classified into two groups per ASTM C618 [5]. Fly ash is rated as class C or F. Fly Ash grade C is found to consist of high calcium content, while grade F, is rich in silica material [5]. However, it is getting more difficult to identify the character of fly ash because of its heterogeneous structure and physical characteristics. Catalysts alongside catalyst aids, pigments, adsorbents, mechanical substrates, electrical and thermal insulators, thin film substrates, and exhaust gas filters are examples of silica applications [6,7,8,9,10,11].

As a result of its large aluminum (Al) and silicon (Si) composition, fly ash could be a useful supplementary resource of refined alumina and silica, assuming that sufficient commercially viable processes can be established [12]. With the use of chemical vapor condensation (CVC) [13], reverse microemulsion (RME) [14], precipitation [15], and the sol–gel method [16,17,18], silica has successfully been isolated from a range of agricultural bio-resources, including sugarcane [19,20,21,22], rice husk ash [23,24,25,26,27], maize cob ash [28,29,30,31,32], bagasse ash [33], clay [34], wheat husk ash, and coffee husk ash, [35,36,37]. Owing to its large concentration of SiO₂ and Al₂O₃, various research has explored coal fly ash from the synthesis of silica. Silica from coal ash can also be employed for adsorption or the retention of certain heavy metals from wastewater [38,39,40,41], either explicitly or as fillers in a polymer matrix. It was discovered that the mineralogical and chemical makeup of the generated material was just as vital as the Si quantity in fly ash together with its retrieval [41,42].

When tolerable amounts of cadmium, Cd (II), and other metals are surpassed in the aquatic environment, they have the potential to harm human physiology along with other biological systems [43]. Heavy metals cannot be ruined or deteriorated in any way. The toxicity of heavy metals could result from polluted drinking water (e.g., lead pipes), elevated concentrations of the surrounding air close to emission sources, or consumption via the food chain [44]. Due to the obvious rising usage of heavy metals in industry, metallic compounds are more readily available in natural water sources [45,46]. Metal contaminants have been removed from water using a variety of processes, including adsorption, precipitation, membrane filtration, and ion exchange [47]. Adsorption, on the other hand, is a cost-effective and efficient method of eliminating heavy metals, organic contaminants, and dyes from contaminated waterways.

The preparation of silica nanoparticles from South African fly ash has been restricted and limited to a solitary study by Mathibela et al. (2020). They synthesized SNPs from a sol–gel process with the utilization of sulphuric acid as the catalyst, polyethylene glycol (PEG) as the surfactant, and reported 99.3 wt. %, SiO₂ purity, and 26% yield [12]. From previous research, 37.3% [38] and about 62% SiO₂ [39] have been reported from Chinese fly ash. According to the author’s expertise, there has been no exposition of the sol–gel hydrothermal template-free synthesis method for silica preparation from SAFA. In this study, a facile extraction of silica from South African fly ash (SAFA) was carried out using a sol–gel, hydrothermal template-free method. Extracted silica was characterized by X-ray diffraction (XRD), X-ray fluorescence (XRF), scanning electron microscope (SEM)/energy-dispersive X-ray spectroscopy (EDX), Fourier transform infrared spectroscopy (FTIR), and transition electron microscope (TEM). In addition, prepared SNPs were exploited for the elimination of Cd (II), from wastewater.

2. Experiment

2.1. Material and Methods

Coal fly ash was gathered from Lethabo power station in South Africa. Unburned carbon and floating beads were removed from the samples using froth flotation [48].

2.2. Silica Extraction

The extraction process of silica from South African fly ash (SAFA) was carried out via a sol–gel, hydrothermal template-free method. SAFA samples containing 50 g (Figure 1a) were introduced to 500 mL of 2 M NaOH in a beaker and heated at 100 °C on a hotplate for 2 h. The solution was stirred incessantly to liquefy the presence of silica in the SAFA, thereby obtaining a solution of sodium silicate. This solution then underwent filtration with the use of filter paper, with deionized water used in washing the residue. The filtrate was allowed to cool and thereafter, underwent titration with a 2 M HCl up to a pH range of 7.5–8.0. The sample solution was poured into a Teflon autoclave and later put into a muffle furnace for 24 h at 373 K for the hydrothermal process to yield silica aqua gels (Figure 1b). The silica aqua gels were refluxed with 1 M HCl for 2 h and dissolved in 1 M NaOH with constant stirring for 1 h. After, the solution was titrated with 1 M HCl to pH 7.5. The resulting white silica xerogel (Figure 1c) was dried at 80 °C and subjected to successive washing cycles. This way, minerals, and impurities are easily separated from the silica. This is because it is more effective to wash dried silica (xerogel) in distilled water (dH₂O) than to wash silica gel (aquagel) before drying, to extract the silica minerals [31]. The extracted SNPs were dried at 110 °C for 6 h (Figure 1d). To measure the production of SNPs concerning the SAFA composition and the purity level of the final SNPs, the product masses were reported by the XRF analysis in

Table 1. XRF analysis of SAFA and SNP.

Composition	Concentration Wt. (%)
	SAFA	SNP
Al2O3	31.66	0.15
BaO	0.19	0.00
CaO	5.33	0.02
Cr2O3	0.06	0.00
Fe2O3	4.10	0.05
K2O	0.84	0.03
MgO	1.54	0.00
Na2O	0.20	0.01
P2O5	0.66	0.00
SiO2	52.03	98.76
SO3	0.35	0.00
TiO2	1.72	0.00
LOI	1.33	0.98
Total	100.00	100.00

.

2.3. Characterizations

Elemental composition was determined by the “MagiX PRO” wavelength dispersive XRF spectrometer from PANalytical. XRD analysis of crystal structure was analyzed using Philips/PANalytical X’Pert Powder Diffractometer at 40 kV acceleration voltage as well as a 40 mA current from 2θ 15° to 55°. Surface morphology was analyzed using a TESCAN VEGA3 SEM. TEM was performed using a JEM-2100 JEOL electron microscope attached with Oxford X-Max for EDX analysis. The surface area of the samples was studied by Brunauer-Emmett-Teller (BET), using a micrometric ASAP 2460. Inductively Coupled Plasma Optical Emission spectroscopy systems (ICP-OES) analysis was utilized for sorption analysis, and IRAffinity-IS was utilized for FTIR analysis. Hitachi STA TG/DSC was used for TGA/DTA analysis, while UV-1800 spectrophotometer was used for ultraviolet-visible diffuse reflection spectroscopy (UV–vis DRS) analysis.

2.4. Application of Extracted SNPs in the Removal of Cd²⁺ from Aqueous Solution

The influence of the quantity of sorbent (m/V, 0.50 g/L, 0.74 g/L, and 1.00 g/L) and removal efficiency on the initial concentration of Cd²⁺ in an aqueous solution was investigated. Cd²⁺ solutions were made by deliquescing various quantities of cadmium nitrate in a 1 L dH₂O. Across all the sorption studies, 50 mL of Cd²⁺ solution was used, and the solution’s pH was changed by 0.1 M NaOH or 1 M HCl. For Inductively Coupled Plasma Optical Emission spectroscopy systems (ICP-OES) analysis, the solution underwent filtration via a filter membrane of 0.45 µm after sorption. The precise Cd²⁺ concentration in the initial solution alongside the filtrate after the sorption process was measured. The operating solution’s pH was modified to pH 5.0 using 1.0 M HCl or 0.1 M NaOH based on a previous study [38,40]. Each sorption was performed in triplicate at 30 °C, and 240 min of reaction time. The sorption capacity (q_e) and the efficiency of removal were estimated from Equations (1) and (2).

$$q_e=\frac{\left(C_o-C_e\right)V}{m}$$

(1)

$$\mathrm{Removal} \mathrm{efficiency} (\%)=\frac{\left(C_o-C_e\right)}{C_o}$$

(2)

where C_o and C_e are metal concentrations at initial and final concentrations, respectively; sorbent mass = m, and solution volume is V.

3. Results and Discussions

3.1. XRF Analysis

The constituent composition of SAFA as determined by XRF analysis is shown in Table 1. It contains mainly Al₂O₃ (31.66%), SiO₂ (52.03%), Fe₂O₃ (4.10%), and CaO (5.33%). After extraction, SNP yield from SAFA was 28.6%, and silica content was 98.56%; this was slightly higher than 24% and similar to 99.3% previously reported using the template method [7]. According to the American Society for Testing and Materials (ASTM C618), this SAFA can be classified as class F owing to its large composition of alumina and silica [5].

3.2. XRD Analysis

Analysis from the XRD of SAFA and SNP illustrated in Figure 2 reveals that SAFA samples contain crystalline phases of Hydrosodalite [Na₈(AlSiO₄)₆(OH)₂·nH₂O], Anhydrite [CaSO₄], Quartz (SiO₂), and Mullite (Al₆Si₂O₁₃), with a substantial silica amorphous phase at 2-theta 19–25°. This had similarities with prior reports from disparate coal fly ash [4,49]. A broad diffraction peak was observed for SNP at 2-theta = 19–25°. It was however observed that the broad peak correlated to the amorphous fraction has been shifted to lower 2 theta angles in sample SNP compared to sample SAFA. The broad diffraction peak confirms the predominantly amorphous nature of extracted SNPs [12,16,31]. This was consistent with the ICDD database PDF # 01-089-8935.

3.3. FTIR Analysis

FTIR spectra are shown in Figure 3. The O-Si-O vibrational modes characteristic peak was noticed at 472 cm⁻¹ [50]. The Si–O stretching vibration was detected at 581 cm⁻¹, symmetrical stretching vibration network of Si-O-Si was assigned to 807 cm⁻¹ [17,18]. Band 1071 cm⁻¹ was because of the intermittent stretch vibration due to Si-O-Si that is typical of SNPs [12,31], this band was observed to have higher intensity in the extracted SNPs. The O-H bond symmetric bending was observed at the 1419 cm⁻¹ band [23], and the 1652 cm⁻¹ broadband was a result of the O-H bond bending vibration which emanated from the Si-OH silanol group. While the 3452 cm⁻¹ and 3471.13 cm⁻¹ bands were because of the O-H bond stretching vibration of the Si-OH silanol group which is a result of the silica surface adsorbed H₂O molecules [23,32,51]. SAFA includes more absorbed water molecules than SNPs, implying that the samples are mostly made up of absorbed water. However, after silica extraction, the amount of absorbed water decreases.

3.4. SEM/TEM Analysis

The morphology of SAFA particles is shown in Figure 4a. From the SEM image, SAFA occurs as a set of particles with unusual shapes, as well as spheroidal and ellipsoidal bodies of different measurements. These discoveries are in good alignment with those stated by Kutchko et al. [2] and Zhu et al. [52], respectively. These earlier experiments have shown that SAFA particle diameter is inconsistent, with small-sized particles adhering to the surfaces of certain spheroidal bodies, which were as well discovered within this research. SEM image of SNP, as shown in Figure 4b. revealed that primary SNPs formed agglomeration. SEM investigation through the Scanning Probe Image Processor revealed particle size variation between 64 nm and 97 nm diameter, with a 65 nm mean size diameter.

TEM of SNPs depicted in Figure 5a, revealed that primary SNPs were roughly spherical with sizes of about ≥60 nm and aggregated to form an agglomeration of micron-size. Strong Si, and O intensity illustrated in the EDX Spectrum of Figure 5b, certified silica (SiO₂) to be the sample’s prevailing element. This is around 98% of the overall component. The presence of C and Cu is from the carbon-coated copper grids used in the TEM analysis. There are however negligible impurities that are usually associated with the process of sol–gel extraction. Strong peak intensities of Si and O demonstrated in Figure 4 (EDX Spectrum), indicated silica (SiO₂) as the principal material of the sample. This represents approximately 98.56% of the total material.

3.5. Bandgap

The prepared SNP optical bandgap (Eg) was evaluated through the measurement of powdered spectral absorption via UV–vis absorbance spectroscopy. The bandgap was subsequently extracted from the resulting tauc plot (Figure 6). The (hv) quantity is displayed along the abscissa in a standard tauc plot while being depicted along the ordinate. The relationship between the photon energy (h) and absorption coefficient (α) demonstrated in Equation (3), can be applied in the calculation of the bandgap.

$${\left(\alpha hv\right)}^2=c \left(hv-Eg\right)$$

(3)

From the equation, α symbolizes the optical coefficient of absorption, c designates the direct transition constant, with h represented as Planck’s constant, and ν denoting the incident photon frequency. The linear intercept of the curve of (αhυ)² vs energy (hυ), was used to evaluate the optical bandgap of the material [53]. The band gap of the prepared SNPs was reported as 4.79 eV. This was lower than the previously reported value by Maj, S. (1988) [54] and similar to the 4.25 eV reported by Khedkar et al. (2019) [55].

3.5.1. Thermal Analysis (TGA-DTA)

From the TGA-DTA illustrated in Figure 7, a weight loss of two-step could be recognized. This weight loss which measured 143 °C (step 1) was attributed to the physically adsorbed dehydration of H₂O. The chemical-bound water from the preparation technique was allotted to the minimal weight loss observed within 143 °C and 535 °C (step 2) [16]. Beyond 535 °C, there was no more weight loss observed, indicating that the retrieved SNPs had reached thermal stabilization. Sharp endothermic peaks within the DTA investigation of a comparable temperature range supported this conclusion [56]. However, beyond 765 °C (stage 3), there was a shape weight loss seen, and the amorphous silica transforms into crystalline silica.

3.5.2. BET Surface Area (S_BET) Analysis

Nitrogen adsorption/desorption isotherms of extracted SNPs prepared from SAFA are displayed in Figure 8. Following the classification system used by the International Union of Pure and Applied Chemistry (IUPAC), the isotherm appeared to be that of TYPE IV with a well-defined peak at high P/P0, which is typical of mesoporous materials [16,57,58]. A large surface area of 18,651 m²/g was also reported in this study. The surface area of prepared nano silica and their sources are shown in

Table 2. Surface area of previously prepared nano silica.

Sources	Surface Area (m2 g−1)	References
Palm kernel shell ash	438	[59]
Rice Husk	327	[60]
Silica-rich rock	422	[61]
Coal Fly Ash	1865	[62]
Oil Shale ash	697	[63]

.

The prepared SNPs’ pore size distribution was determined using desorption isotherms by the Barett–Joyner–Halenda (BJH) approach as shown in Figure 9. At high corresponding pressures of 0.7 and 1.0, respectively, the isotherms exhibited a unique hysteresis loop type H3, which displayed the existence of 35.71% mesoporous SNPs (2–50 nm) and 64.29% microporous SNPs (above 50 nm), exhibiting a bimodal distribution. The average pore diameter was determined as 46.3 nm.

3.5.3. Cd²⁺ Removal from Aqueous Solution Utilizing Prepared SNPs

The sorption behavior may be influenced by the original concentrations of metal ions and the dosage of SNPs. The original Cd²⁺ concentration affected SNPs sorption positively. Figure 10 explained that the reduced dosages of 0.5 g/L and 0.75 g/L resulted in the system achieving faster sorption saturation with an increasing original Cd²⁺ concentration. This thus suggests reduced sorption capacities. Similar results have been previously reported [38,40]. As shown in Figure 11, the purge efficacy of Cd²⁺ swiftly decreases with increasing initial concentrations of 0.05 g/L and 0.75 g/L dosages accordingly. However, a 1.00 g/L dosage suggests a higher sorption capacity of SNPs at a higher dosage.

4. Conclusions

SNPs were extracted using a sol–gel hydrothermal template-free technique from South Africa Fly Ash (SAFA). The occurrence of silica in SAFA and the amorphous nature of derived silica have been discovered by XRF studies. Extracted silica yield was 28.6% and purity was 98.56% with a negligible mineral contaminant. The crystalline phase of SAFA and the amorphous nature of extracted SNP have been investigated by XRD. The existence of SiO₂ in the samples was confirmed by EDX results, with the FTIR data displaying the primary chemical group present in the samples. SEM and TEM revealed that primary SNPs were roughly spherical with sizes of about ≥60 nm. This study will add to the awareness of a template-free, low-cost extraction of SNPs from SAFA, as well as being a great performance and economical sorbent for disposing of wastewater by the materials which emanate from waste. The efficient use of SAFA can have significant economic and environmental benefits as a useful raw material for advanced high-tech applications.