Continuous Production of Mesoporous Silica with Controlled Structure via Aerosol-Assisted Evaporation-Induced Self-Assembly

Abstract

This work demonstrates a viable and attractive continuous synthesis route for spherical SBA-15 mesoporous silica particles using the aerosol-assisted evaporation-induced self-assembly (AA-EISA) technique. This approach integrates aerosol processing with a self-assembly mechanism driven by solvent evaporation, offering distinct advantages over conventional sol–gel and hydrothermal methods. The influence of ethanol content, pH, P123 template amount, and reactor temperature (T _r) on the formation of SBA-15 spheres was systematically investigated using gas physisorption, thermogravimetric analysis (TGA), small-angle X-ray scattering (SAXS), and electron microscopy. In the absence of ethanol, the desired spherical morphology was not obtained; instead, the products consisted of mixed, fiber-like elongated features and bundled aggregates, highlighting ethanol’s pivotal role in directing spherical particle formation. An ethanol/tetraethyl orthosilicate (TEOS) molar ratio of 1 was sufficient to maintain a spherical morphology during continuous runs up to 12 h. For production extending beyond 12 h, a ratio of 3 was required to ensure precursor stability and to balance surface area with pore volume. Adjusting the pH to 3 enabled pore expansion to ∼20 nm and a pore volume of 1 cm³/g, with a moderate surface area of 194 m²/g. Lower ethanol and acid usage suggest a more sustainable process. A P123/TEOS molar ratio of 0.015 was sufficient to yield well-ordered mesostructures with minimal reactor fouling. Reducing T _r from 400 to 200 °C decreased energy consumption without sacrificing quality, however, a further reduction to 120 °C compromised the morphology. Key parameters affecting continuous AA-EISA production are given, showing potential for a robust, energy-efficient, and environmentally conscious route to produce mesoporous silica particles.

1. Introduction

Interest in ordered mesoporous silica relates to their distinctive properties, such as a well-defined pore structure with a high surface area complemented by a narrow pore size distribution. Being able to easily tailor these materials increases their potential by making them more attractive for a broader range of applications, such as drug delivery, sensors, adsorption, catalysis, and separations. Over the past two decades, numerous ordered mesoporous materials were synthesized, such as the MCM, SBA, FDU, TCPS, TUD, and HMS families of materials. − Among these, SBA-15 is relatively straightforward to synthesize using the triblock copolymer Pluronic, P123 (PEO₂₀-PPO₇₀-PEO₂₀) as a template. SBA-15 has high thermal and mechanical stability, a two-dimensional hexagonally ordered (p6mm) pore alignment, high surface area, large pore volume, and well-defined mesopores with a tunable size. ,, This makes it attractive for diverse applications, including catalysis, where it serves as an excellent support material, or for controlled drug release. Additionally, SBA-15 can be used to adsorb pollutants and in chromatographic separation and purification.

The synthesis of nanostructured porous silica can be significantly influenced by the amount of ethanol (EtOH), template, pH of the solution, and reaction temperature. Several studies report the effect of these parameters on the structural properties of mesoporous silica when synthesized using the conventional batch sol–gel method. ,,− Pluronic P123 is often favored as a template for biomedical applications due to its biocompatibility, ease of removal, and reduced environmental impact. Also, average pore diameters for mesoporous silica created with P123 of 7–10 nm are easily obtained, making it attractive to enhance the accessibility and diffusivity of large molecules. , Pluronic P123 surfactant self-assembles into micelles in an acidic environment and serves as the mesopore structure-directing agent, where the Critical Micelle Concentration (CMC) is 0.030 and 0.005 weight-to-volume percent (w/v %) in water at 25 and 30 °C, respectively. The acid, mainly HCl, helps to maintain the mixture’s acidity around pH = 2, near the point of zero charge (PZC) of the evolving silica surface. EtOH is generally added to act as a cosolvent and cosurfactant. However, additional EtOH will be released from the tetraethyl orthosilicate (TEOS) hydrolysis, which can change the CMC and micellization process. A total of 1–10 wt % EtOH produces spherical micelles, 20–30 wt % produces elongated micelles, and more than 30 wt % does not form micelles. Moreover, it was demonstrated that the surface area for the synthesized mesostructured silica decreased from 653 to 278 m²/g as the pH of the solution was increased from 1.2 to 7.

The conventional batch process to produce SBA-15 consists of two steps. The first step is the aging of the mixture after adding all the SBA-15 precursors, and the second step is hydrothermal crystallization. The aging step usually controls the external morphology, such as a hexagonal prism, short rod, or long rod. , The hydrothermal crystallization step usually controls the pore size of mesoporous SBA-15 nanoparticles, where increasing the temperature of the silica/surfactant nanocomposite increases the pore size. ,,, However, both steps consume a significant amount of time, from several hours to days, to produce SBA-15 with the desired properties, such as a high degree of order of the pores. Other disadvantages of the batch process include limited production capacity, which can constrain output and hinder scalability. Additionally, bottlenecks in the process often lead to increased product prices and inefficiencies. The necessity for costly human intervention introduces the risk for errors, resulting in more product defects and inconsistent quality. These factors collectively reduce the overall efficiency and cost-effectiveness of the batch production method. In this regard, the continuous and scalable aerosol-assisted evaporation-induced self-assembly (AA-EISA) approach is advantageous compared to traditional batch synthesis, as it reduces the synthesis time significantly, producing spherical mesoporous silica nanoparticles in a few seconds. Thus, the AA-EISA approach is a versatile and effective method for fabricating structured materials by merging self-assembly driven by solvent evaporation into the aerosol process. This technique increases the quantities of processed materials and allows frequent sampling for product quality checks.

Despite this, and perhaps surprisingly so, over the past three decades since nanostructured mesoporous silica was introduced, there has been a dearth of literature on synthesizing such materials using the AA-EISA approach. − The AA-EISA method, a novel approach to micro and nanomaterials synthesis, was pioneered by C. Jeffrey Brinker at Sandia National Laboratories and the University of New Mexico. Their publications, since the early 2000s, have shed light on forming mesostructured materials using aerosol-assisted processes. − ,, They synthesized mesostructured spherical nanoparticles and laid the foundation for using block copolymers in AA-EISA processes using 5% P123 with TEOS, HCl, and EtOH to produce a mixture with the molar ratio TEOS:EtOH:water:HCl:P123 of 1:22:5:0.004:0.0096.

In 2008, Pitchumani et al. reported small-scale, continuous production of spherical nanostructured porous silica via AA-EISA using a mixture of TEOS:EtOH:water:HCl:P123 with the molar ratio of 1:30:40:0.0525:(0.008–0.010). However, the increase of P123 did not show any significant impact on the material properties. Then, they adopted a statistical approach to determine the optimal conditions for nanostructured silica synthesis using the factorial design methodology to determine the effect of furnace temperature, template, water, and EtOH on the product’s textural properties. It was noted that the BET surface area was impacted proportionally by P123 content and inversely by water content. Nonetheless, the strong interaction between P123 and EtOH makes them interdependent, so they need to be considered together. The reactor temperature (T _r) and EtOH content also proportionally impact the total pore volume. However, the pH was not investigated for the AA-EISA method, where the precursor concentration in the droplet changes along the reactor, affecting the residence time and the localized pH. All previous AA-EISA studies used the exact pH of 1.2 in their synthesis.

Around 2009, the horizon for AA-EISA expanded to create new amorphous aluminosilicate materials with hierarchical porosity along with embedded nanocrystals of zeolite into the amorphous mesostructured aluminosilicate matrix. ,, Since then, AA-EISA attracted more attention to synthesize functional nanostructured inorganic and hybrid porous materials for various applications. , An example includes WO₃–SiO₂–Al₂O₃ as metathesis catalyst, leading to a higher propene yield, and TiO₂–SiO₂ epoxidation catalysts, leading to a high epoxidation activity.

Those AA-EISA studies did not discuss their recipes’ scalability for industrial production. It is worth noting that a significant amount of EtOH in the mixture hampers industrial applicability and scalability due to flammability and explosion risks. Similarly, using an excessive amount of the Pluronic template (P123) will release more CO₂ during the calcination step, where the template is burned to create the mesoporosity. The higher reactor temperature also consumes more energy and releases more CO₂ during production. On the other hand, increasing pH means less use of HCl, less hazardous waste generation, lower capital costs for manufacturing equipment, and longer life cycles.

Therefore, this work aims to reduce the AA-EISA reactor temperature to synthesize spherical mesoporous silica with a high surface area by using the lowest amount of EtOH, P123, and HCl to make this process more viable and attractive for industrial production and commercialization.

2. Experimental Section

2.1. Materials

Tetraethyl orthosilicate (TEOS, reagent grade 99%) used as the silica source, ethanol (EtOH, absolute), hydrochloric acid (HCl, ACS reagent 37%), and triblock copolymer block-poly­(ethylene glycol)-block-poly­(propylene glycol)-block-poly­(ethylene glycol) EO₂₀PO₇₀EO₂₀ (Pluronic P123) were obtained from Sigma-Aldrich Company Ltd. and used as received without any further purification. Deionized Water (DI) was obtained from a water purifier (RiOs-3, Millipore).

2.2. Experimental Setup

Scheme a shows the schematic of the continuous aerosol-assisted evaporation-induced self-assembly (AA-EISA) apparatus. Polypropylene bottles were used as feed tanks and fixed on a hot plate. The tank was connected to a peristaltic pump (Minipuls 3, Gilson) to transfer the solution to a Collison nebulizer (BGI Inc.). The atomized droplets were carried with nitrogen gas toward the aerosol reactor, which is a three-zone tubular furnace (VCTF7, Vecstar). The resulting powder was collected inside a customized stainless steel vacuum filtration system (MV050, Cytiva) using standard 47 mm filter paper with 2.5 μm particle retention (1005–047, Whatman). A vacuum filtration flask filled with DI water was used as a scrubber to absorb gas impurities, such as HCl, before release into the ventilation.

1. Schematic Diagram Illustrating the (a) Continuous Aerosol-Assisted Evaporation-Induced Self-Assembly (AA-EISA) Setup, (b) EISA Mechanism, Crossing Critical Micelle Concentration (CMC) Levels due to Evaporation: CMC 1 (where the First Micelles Appear) and CMC 2 (Formation of Ordered Mesophases).

2.3. Synthesis of Mesoporous Silica (SBA-15) Precursor via AA-EISA

Molar ratios of the gel, TEOS:EtOH:H₂O:P123, are 1:X:170:Y, where X = 0, 1, 3, 5, and 10 mol EtOH/mol TEOS, and Y = 0.010, 0.016, and 0.020 mol P123/mol TEOS. The base case is X = 3 and Y = 0.010. For this, 4.74 g of P123 was dissolved in 252.70 g of deionized water and 11.40 g of EtOH, then stirred at 500 rpm for 1 h at room temperature until a clear solution was obtained. Then, the pH of the solution was adjusted to 1.2 with HCl (37 wt %), and the resulting solution was stirred at 500 rpm for 10 min 17.19 g TEOS was added to the solution and continuously stirred at 500 rpm for 1h to form a clear solution with pH ∼ 1.2. T _r was set to 400 °C before atomizing the mixture. Then, the mixture was transferred to the feed tank and stirred at 500 rpm at room temperature. A peristaltic pump transferred the mixture at 0.50 rpm to the Collison nebulizer, which was operated at 101.3 kPa pressure and a nitrogen gas flow rate of 5000 mL/min during the experiment. This is where the solvents begin to evaporate, initiating the EISA mechanism, as depicted in Scheme b. After completing the run, the dried powder sample was collected and calcined at 550 °C for 6 h at a heating rate of 1 °C/min to remove the organic templates.

2.4. Characterization Techniques

The gas physisorption measurements were conducted on an Autosorb IQ2 instrument (Anton Paar, Quantachrome Instruments) using both argon (Ar at 87 K) and nitrogen (N₂ at 77 K). Before the measurements, sample degassing was carried out under vacuum at 50 °C for 1h, 100 °C for 5h, 300 °C for 10h, and finally at 350 °C for 3h. Both Ar (87 K) and N₂ (77 K) isotherms were measured over the whole pressure range (approximate relative pressure (P/P ₀) ranged from 10^–7 up to 1) for the same sample. The analysis was carried out with Quantachrome software ASiQWin and following the International Standardization Organization (ISO) recommendations in their ISO 9277:2022 for BET calculations for the porous solid surface area and ISO 15901–2 for pore size distribution and porosity of solid materials. The specific (nominal) surface area was estimated using the Brunauer–Emmett–Teller (BET) theory, where P/P ₀ ranged from 0.05 to 0.30. The pore size distribution was estimated based on the adsorption branch data using the nonlocal density functional theory (NLDFT) method and the cylindrical pore model kernel, which assumes a cylindrical pore geometry.

The thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were coconducted on a TGA/DSC 3+ instrument (Mettler Toledo). The small-angle X-ray scattering (SAXS) spectra were measured on a Ganesha 300XL instrument (Xenocs, SAXSLAB) using 1.5 mm borosilicate glass capillaries. Scanning electron microscopy (SEM) images were acquired with a Zeiss XB1540 Crossbeam (Carl Zeiss) and Zeiss NVision 40 instrument with laser ablation. The samples were sputter-coated with gold for 10 s to reduce charging using an Edwards Sputter coater (Edwards). Transmission electron microscopy (TEM) images were collected with a JEOL 2100 at 200 kV TEM with a Gatan camera. Samples were dispersed in methanol and then drop-casted onto holey carbon-coated copper grids (E.M. Resolutions). The acquired data were analyzed, and figures were plotted using OriginPro 2021b software (OriginLab).

3. Results and Discussion

3.1. Aerosol-Assisted Evaporation-Induced Self-Assembly Mechanism

The synthesis mechanism for EISA is depicted in Scheme b. The solvent starts evaporating after the droplets enter the aerosol reactor. During the evaporation process, the concentration of solutes (TEOS and surfactants (P123)) significantly increases within the droplets. ,, As the solvent evaporates, the concentration of surfactants gradually reaches CMC 1, triggering the formation of P123 surfactants into micelles. Further evaporation passes CMC 2, leading to the formation of ordered mesophases consisting of these micelles. Simultaneously, inorganic species form around these structures, as part of a cooperative process between the micelles and an evolving inorganic phase. The hydrolysis and condensation of the metal alkoxides proceed within the confined environment of the droplet, ultimately forming an inorganic network templated by the surfactant structure. Complete evaporation of the solvent solidifies the structure, creating a mesostructured, composite particle. A porous material emerges after calcination, which removes the organic phase.

3.2. Textural Properties by Gas Physisorption Analysis

Gas physisorption analysis of all the mesoporous silica samples prepared using the AA-EISA technique was conducted as per the IUPAC and the ISO , recommendations with N₂ (77K) for meso/macroporous materials. Previous studies indicated an overestimation of approximately 20–25% in BET surface area measured with N₂, compared to Ar. Because nitrogen interacts more strongly with silica surfaces due to its quadrupole moment, it is adsorbed more readily in the micropores than argon, leading to an overestimation of the BET surface area. Therefore, the samples were tested with both Ar (87 K) and N₂ (77 K), and it was demonstrated in Table S1 that the N₂ BET surface area overestimation, indeed, ranged from 23 to 36%. Because the results derived using Ar gas are more reliable than those using N₂, they are used to further analyze the textural properties, which are shown in Table .

1. Textural Properties of Samples Measured Using Ar Physisorption.

m-SiO2-n (n =)	T r (°C)	EtOH/TEOS molar ratio (X)	P123/TEOS molar ratio (Y)	pH	S BET $\left(\frac{{\mathrm{m}}^2}{\mathrm{g}}\right)$	D p (nm)	V total ( $\frac{{\mathrm{c}\mathrm{m}}^3}{\mathrm{g}}$ )
1	400	0	0.015	1.2	611	6	0.95
2	400	1	0.015	1.2	273	10.5	0.93
3	400	3	0.015	1.2	300	10.5	0.95
4	400	5	0.015	1.2	297	11	0.97
5	400	10	0.015	1.2	290	11	0.99
6	400	3	0.010	1.2	253	9	0.75
7	400	3	0.020	1.2	415	16	1.03
8	400	10	0.015	3.0	194	21	1.06
9	300	3	0.010	1.2	231	9	0.63
10	200	3	0.010	1.2	211	8.5	0.53
11	120	3	0.010	1.2	190	6	0.30

^aSurface Area-MultiPoint BET (S _BET).

^bNLDFT Modal Pore Diameter (D _p).

^cTotal pore volume (V _total) at P/P₀ = 0.995. The error margins for the physisorption data are approximately 5%.

Table indicates that the mesoporous silica produced through AA-EISA without EtOH (denoted m-SiO₂-1) has the highest surface area of 611 m²/g. Compared to m-SiO₂-1, adding EtOH at a molar ratio of EtOH/TEOS, X = 1, leads to m-SiO₂-2 with a surface area that has more than halved and a pore diameter increased by 70%, while only slightly impacting the total pore volume. Further increasing the EtOH content tended to marginally enhance the surface area and total pore volume while maintaining the pore diameter nearly constant, as depicted in Figure a. Figure b shows the effect of EtOH on the Ar adsorption/desorption isotherms for nanostructured mesoporous silica synthesized via AA-EISA, which aligned with type IV isotherms in the IUPAC classification. The mesopore sorption behavior of m-SiO₂-1 exhibits the typical H1 hysteresis loop observed in SBA-15, with delayed capillary condensation occurring at a relative pressure, P/P₀, ranging between 0.5 and 0.7. The H1 loop usually indicates a narrow range of uniform mesopores and controlled ordering of the pores.

1.

(a) Trends of textural properties as a function of EtOH/TEOS molar ratio (X). (b) Ar adsorption/desorption isotherms for varied X. (c) Trends of textural properties as a function of P123/TEOS molar ratio (Y). (d) Ar adsorption/desorption isotherms for varied Y. (e) Trends in textural properties as a function of reactor temperature (T _r). (f) Ar adsorption/desorption isotherms for varied T _r. The error bar represents 5% systematic error from the physisorption equipment.

In contrast, the remaining samples (m-SiO₂-2, m-SiO₂-3, m-SiO₂-4, and m-SiO₂-5) display different but uniform mesopore sorption behavior resembling a combination of an H1 hysteresis loop and an H2-b loop, where the capillary condensation takes place at a relative pressure, P/P₀, ranging between 0.45 and 0.995. This suggests the presence of more complex pore structures forming networks or pore blocking with a substantial size distribution of neck widths. Figure S2a shows the effect of EtOH on the pore size distributions of the synthesized samples with the NLDFT method using the adsorption branch of the isotherm. The m-SiO₂-1 exhibits a significant and sharp peak in its pore size distribution, with pore widths predominantly around 6 nm and a smaller fraction of micropores around 0.4 nm. The other samples show broader pore size distributions, featuring a significant peak around 10.5–11 nm, along with some pores with 12 nm pore width. It appears that the addition of EtOH correlates with an increase in pore size and a wider distribution.

The effect of the P123 template on the synthesis of mesoporous silica was investigated using molar ratios of P123/TEOS, Y = 0.010, 0.015, and 0.020. The positive, nonlinear relationship between P123 and the textural properties is shown in Figure c. It was observed that increasing the template P123’s content while keeping the other parameters constant for m-SiO₂-6 appears to have a notable impact on the surface area, pore diameter, and total pore volume. For example, increasing the P123 content by 50% for m-SiO₂-3 resulted in a 19% increase in surface area compared to m-SiO₂-6, whereas doubling the template quantity for m-SiO₂-7 led to a substantial 64% increase. Similarly, the pore diameter exhibited a 15% and 77% increase with a 50% and 100% addition of P123, respectively. However, the effect on total pore volume was comparatively less pronounced, rising from 27% to 37% for identical additions. Figure d shows the influence of the effect of P123 on the Ar adsorption/desorption isotherms for nanostructured mesoporous silica synthesized via AA-EISA, which aligned with a type IV isotherm as classified by IUPAC. Figure S2b illustrates the effect of P123 on the pore size distributions of the synthesized samples with the NLDFT method using the adsorption branch of the isotherm. The pore size distributions for m-SiO₂-3 and m-SiO₂-6 show a similar pattern but with a 15% peak shift from 9 to 10.5 nm. However, m-SiO₂-7 shows a broader distribution without any sharp peaks. Increasing the content of P123 to Y = 0.020 can disrupt the dominant peak and create broader pore size distributions.

This adjustment can significantly impact material properties and performance in various applications. The resulting mix of different pore sizes can increase the total surface area, which is advantageous for catalysis and adsorption. However, it is important to note that this adjustment could also introduce some variability in catalytic performance, as different pores may have varying levels of accessibility and activity.

T _r was varied from 120 to 400 °C to understand the effect of the synthesis temperature on the mesoporous silica’s texture. It was observed that increasing the reaction temperature while keeping the other parameters constant appears to significantly increase the surface area along with the pore diameter and the total pore volume. This is particularly evident at lower temperatures, as illustrated in Figure e. There seems to be a proportional relationship between the temperature and the surface area that follows the following linear trend with a coefficient of determination, R ²∼ 0.997

$$S_{\mathrm{B}\mathrm{E}\mathrm{T}}=0.2218T_{\mathrm{r}}+164.7$$ 1

Like the surface area, the total pore volume and diameter tend to increase when the temperature rises, but the pore diameter stabilizes and becomes constant above 300 °C. This trend is most likely because higher temperatures provide more kinetic energy to the reactant molecules, accelerating the condensation and polymerization of silica around the P123 templating agent. , This increased reaction rate can help to form a more ordered and extensive network of mesopores, which improves the structural integrity, strength, and stability. This can be supported by the type IV isotherm shown in Figure f, which suggests more complex pore structures and networks or pore blocking with a substantially broadened size distribution of neck widths.

The mesopore sorption behavior for m-SiO₂-11 shows the conventional H1 hysteresis loop for SBA-15, with delayed capillary condensation at P/P₀ between 0.47 and 0.8. The H1 loop usually reveals a narrow range of uniform mesopores and controlled ordered pores. However, the rest of the samples shows different but uniform mesopore sorption behavior closer to a combination of an H1 hysteresis loop along with an H2-b loop, where the capillary condensation takes place at P/P₀ between 0.4 and 0.45 to 0.995.

The pore size distribution for m-SiO₂-10 is broader, with split peaks at 4.5 and 9 nm. Both m-SiO₂-6 and m-SiO₂-9 show similarly wide pore size distributions with 9 and 10.5 nm split peaks. On the other hand, reducing the temperature to 120 °C narrows the pore size distributions for m-SiO₂-11 significantly, to a peak around 6 nm pore diameter.

The pH of the solution was increased to understand its effect on the synthesized mesoporous silica’s texture. It was observed that increasing the pH from 1.2 for m-SiO₂-5 to 3.0 for m-SiO₂-8 reduced the surface area by 33% from 290 to 194 m²/g, but the total pore volume increased by only 7% from 0.99 to 1.06 cm³/g, which aligns with the trend in reported values. For a pH value of 7, no sample was collected in the filter (no notable silica formation) due to the short residence time in the AA-EISA. This contradicts a previously reported successful batch synthesis of mesostructured silica at a pH of 7, which can be explained by the insignificant rate for the TEOS hydrolysis reaction for pH values closer to 7 and the short residence time in the AA-EISA. On the other hand, this increased the pore diameter by 89% from 11 to 21 nm, as shown in Figure S2d. The type IV isotherms in Figure S2e illustrate the impact of pH on the Ar adsorption/desorption isotherms, where the behavior of m-SiO_2–5 (prepared at pH ∼ 1.2) resembles a combination of H1 and H2-b hysteresis loops, but m-SiO₂-8 (pH = 3.0) demonstrates behavior closer to a combination of H1 and H2-a.

3.3. Thermogravimetric Analysis and Differential Scanning Calorimetry

For mesoporous silica, TGA provides crucial information about its thermal stability, decomposition behavior, and the presence of adsorbed species or residual template materials. DSC is often used to study phase transitions of substances confined within the pores. When integrated, TGA and DSC becomes valuable to analyze the incorporated organic components. Figure shows the TGA graphs of different samples prepared using AA-EISA by varying the reaction conditions and reactant ratios, and Table shows the TGA wt % losses per temperature range along with the DSC peaks. Usually, the initial mass loss in TGA up to 200 °C indicates the evaporation or desorption of physically adsorbed water weakly bound to the surface and within the pores of mesoporous silica as the dominant contribution in this first phase. During a second phase, the mass lost due to the template decomposition occurs between 200 and 400 °C. Above 400 °C, the third phase is considered an excellent indicator of thermal stability and may also include minor contributions from residual carbon burnout and progressive condensation or dehydroxylation of surface silanol groups, leading to siloxane bond formation and water release. The DSC traces display two upward exothermic peaks that mirror these regimes: first at 130–160 °C that accompanies the onset of mass loss, and a second one near 600 °C attributed to burnout of residual carbon and the beginning of silica densification, together with possible silanol condensation or dehydroxylation at elevated temperature.

2.

Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) profiles for AA-EISA–derived SBA-15. (a) Effect of EtOH/TEOS molar ratio, X, on mass loss; (b) effect of template loading, P123/TEOS molar ratio, Y; (c) effect of solution pH; (d–f) corresponding DSC heat-flow (HF) traces (exotherm up) for the X, Y, and pH series, respectively.

2. TGA wt % Losses per Temperature Range (°C) and DSC Peaks.

m-SiO2-n (n = )	Wt % losses per temperature range (°C)			DSC peaks
	25–200	200–400	400–1000	1st	2nd
1	56	22	1	160	600
2	53	7	1	139	600
3	53	6	1	138	600
4	55	5	2	140	600
5	39	8	2	146	600
6	41	7	1	140	600
7	49	7	1	130	600
8	43	8	1	135	600

In the EtOH/TEOS samples shown in Figure a,d, the ethanol-free sample m-SiO₂-1 exhibits the largest template step of 22 wt % between 200 and 400 °C and the lowest final residue of 21 wt %. Introducing ethanol reduces the 200–400 °C loss to 5–8 wt % for m-SiO₂-2 through m-SiO₂-5 and lowers the low-temperature loss from 56 wt % to 39 wt % at the highest ethanol level, increasing the residue to 51 wt % and indicating more efficient detemplation and higher inorganic yield. In the samples with varied P123/TEOS shown in Figure b,e, varying the template loading from Y = 0.010 to 0.020 keeps the 200–400 °C loss essentially constant at about 7 wt %, which implies near-saturation of incorporated template. The high-temperature loss remains about 1 wt %, confirming comparable stability and indicating only minor high-temperature contributions from residual carbon burnout and silanol condensation or dehydroxylation. In the samples with varied pH shown in Figure c,f, raising pH from about 1.2 to about 3.0 increases the 25–200 °C loss from 39 wt % to 43 wt %, while the template step remains at 8 wt % and the loss above 400 °C stays at 1–2 wt %, giving residues of 51 and 48 wt %, respectively. The consistently small mass change above 400 °C and the subdued high-temperature exotherm across all panels confirm the robust thermal stability of the AA-EISA SBA-15 frameworks, while also showing that residual carbon burnout and silanol condensation or dehydroxylation are limited at elevated temperature.

3.4. Small-Angle X-ray Scattering Analysis

SAXS is a robust technique for characterizing the nanostructure of materials with characteristic length scales ranging from 1 to 100 nm, such as mesoporous silica. Also, it can provide information about mesoporous materials’ periodic pore arrangement, unit cell size, and pore wall thickness when combined with gas adsorption measurements. The SAXS patterns in Figure show clear Bragg peaks for some of the samples, corresponding to the primary scattering due to the ordered arrangement of pores. The measured momentum transfer (q value in Å ^–1) for those peaks is related to the distance between planes of pores, d-spacing (d in nm), using eq

$$d(\mathrm{n}\mathrm{m})=\frac{2\pi }{10q({\mathrm{\AA }}^{-1})}$$ 2

3.

Small-angle X-ray scattering (SAXS) profiles for SBA-15 prepared by AA-EISA: (a) effect of ethanol/TEOS molar ratio, X; (b) effect of template loading, P123/TEOS molar ratio, Y; (c) effect of solution pH; and (d) effect of reactor temperature (T _r).

The unit cell parameter (a ₀) describes the distance between the centers of adjacent mesopore channels. It was calculated from the d-spacing and assuming a hexagonal structure using eqs -()

$$\mathrm{T}\mathrm{h}\mathrm{e}\mathrm{f}\mathrm{i}\mathrm{r}\mathrm{s}\mathrm{t}\mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k}\mathrm{c}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{n}\mathrm{d}\mathrm{s}\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{e}(100),a_{100}=\frac{2d_{100}}{\sqrt{3}}$$ 3

$$\mathrm{T}\mathrm{h}\mathrm{e}\mathrm{s}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{d}\mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k}\mathrm{c}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{n}\mathrm{d}\mathrm{s}\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{e}(110),a_{110}=2d_{110}$$ 4

$$\mathrm{T}\mathrm{h}\mathrm{e}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{r}\mathrm{d}\mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k}\mathrm{c}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{n}\mathrm{d}\mathrm{s}\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{e}(200),a_{200}=\frac{4d_{200}}{\sqrt{3}}$$ 5

The pore wall thickness (t) is the difference between the unit cell parameter (a ₀) and the pore diameter D _p (nm) obtained using Ar adsorption. The textural properties of all samples were calculated and summarized in Table .

3. Textural Properties of Samples Measured Using SAXS.

m-SiO2-n (n = )	q (Å–1)	Plane	d (nm)	a Plane (nm)	D p (nm)	t (nm)
1	0.068	100	9	11	6	4
	0.119	110	5
	0.137	200	5
2	0.053	100	12	14	10.5	3
3	0.054	100	12	14	10.5	3
4	0.052	100	12	14	11	3
5	0.051	100	12	14	11	3
6	0.053	100	12	14	9	5
9	0.052	100	12	14	9	5
10	0.057	100	11	13	8.5	4
11	0.069	100	9	11	6	5

For samples showing only one clear SAXS reflection, a ₀ was calculated from the (100) reflection. For m-SiO₂-1, which showed three reflections, the reported a ₀ value represents the average calculated from the (100), (110), and (200) reflections. The corresponding wall thickness was then calculated using this representative a ₀ value.

The SAXS patterns for m-SiO₂-7 and m-SiO₂-8 did not display any clear Bragg peaks, suggesting disorganized and irregular pore arrangement or a broader pore size distribution without a dominant pore size. Those results align well with the previously estimated pore size distributions, illustrated in Figure S2b,d for m-SiO₂-7 and m-SiO₂-8. Additionally, some relatively insignificant peaks are marked by the red boxes in Figure next to the dominant peak. Those features appear to be broad without a distinct maximum in intensity, so they do not qualify as well-defined peaks but could indicate weak or short-range ordering or incomplete mesophase formation. Additionally, the 2D SAXS patterns for nanostructured mesoporous silica samples are illustrated in Figure S3.

The SAXS patterns of the samples used to study the EtOH effect are illustrated in Figure a. The m-SiO₂-1 pattern exhibits three q peaks at approximately 0.07, 0.12, and 0.13 Å^–1, corresponding to the d-spacing of approximately 9, 5, and 5 nm, respectively.

Although m-SiO₂-1 exhibits the highest estimated wall thickness, this does not contradict its high S _BET. The wall thickness was estimated indirectly from t = a ₀ - D _p, whereas S _BET reflects the total accessible surface area measured by Ar physisorption. The smaller pore diameter of m-SiO₂-1 increases the surface-area-to-volume ratio of the mesopores, which can lead to a higher measured surface area despite the larger calculated wall thickness. Therefore, S _BET and t are related structural parameters but are not expected to vary inversely in a simple one-to-one manner.

The patterns for m-SiO₂-2, m-SiO₂-3, m-SiO₂-4, and m-SiO₂-5 each display a single prominent q peak at approximately 0.05 Å^–1, corresponding to a d-spacing of 12 nm.

SAXS patterns for samples used to study the impact of P123 are displayed in Figure b. The patterns for m-SiO₂-3 and m-SiO₂-6 exhibit distinct q peaks around 0.05 Å^–1, corresponding to a d-spacing of 12 nm. The SAXS patterns for the samples used in the study of the impact of pH are shown in Figure c. In particular, the m-SiO₂-5 pattern shows a distinctive q peak around 0.05 Å^–1, corresponding to a real-space distance of 12 nm, which is missing from the m-SiO₂-8 pattern. Figure d displays the SAXS patterns for the samples used to study the effect of reaction temperature. These patterns reveal a distinct q peak for m-SiO₂-6, (the same for m-SiO₂-9), m-SiO₂-10, and m-SiO₂-11 at 0.053, 0.057, and 0.069, which correspond to a d-spacing of 12, 11, and 9 nm, respectively.

3.5. Scanning Electron Microscopy Analysis

SEM was used to investigate the external morphology of the synthesized crystallite surface and the additional porosity in the textural properties, as well as particle dimensions. Figure displays SEM images depicting the impact of EtOH on the external morphology of the produced nanostructured mesoporous silica at different magnifications.

4.

SEM images for mesoporous silica via AA-EISA for different values of X (EtOH/TEOS): (a) m-SiO₂-1, (b) m-SiO₂-2, (c) m-SiO₂-3, (d) m-SiO₂-4, and (e) m-SiO₂-5; Y (P123/TEOS): (f) m-SiO₂-6, (c) m-SiO₂-3, and (h) m-SiO₂-7; solution pH: (i) m-SiO₂-5, (j) m-SiO₂-8; and reactor temperature: (f) m-SiO₂-6, (g) m-SiO₂-9, (k) m-SiO₂-10 and (l) m-SiO₂-11.

Figures a and S4b show mixed fiber-like elongated features and a bundle morphology but not the targeted spherical morphology for m-SiO₂-1 (without EtOH addition). This supports the previous suggestions of conventional SBA-15 formation due to P123 micellar growth in the precursor solution, and this sample has the highest surface area too. Also, this suggests that the EtOH produced from the TEOS hydrolysis (approximately 5 wt %) is insufficient to form spherical micelles and prevent their growth and assembly into a hexagonal structure. Those results align with the conventional SBA-15 synthesis but contradict some reported results for AA-EISA syntheses without EtOH, possibly due to precursor stability during long runs.

The SEM images in Figure b for m-SiO₂-2 indicate that adding one mole of EtOH per mole of TEOS (X = 1) successfully produced a spherical morphology, with particle diameters ranging from 0.7 to 5 μm. However, the images also revealed the presence of conventional SBA-15 fiber morphology mixed in due to micelles growing in the precursor solution to form those fiber-like elongated features, as highlighted by the red circles. However, moving to X = 3 for m-SiO₂-3 eliminates the conventional SBA-15 fiber morphology and produces spherical particles with particle diameters ranging from 0.7 to 5 μm, as shown in Figure c. Nevertheless, some broken large shells appear in the mixture with 0.37 μm wall thickness. Moreover, at X = 5 for m-SiO₂-4 can increase the particle size range to 0.7–16 μm and produce more broken shells with approximately 1 μm thickness, as shown in Figure d. On the other hand, Figure e shows that X = 10 produces a particle size range of 0.9–7 μm for m-SiO₂-5 with a significant amount of broken shells, some with approximately 0.9 μm thickness and a lot of debris in the mix. Interestingly, Figure b−d for m-SiO₂-2, m-SiO₂-3, and m-SiO₂-4, respectively, show fingerprint-like surface striations, a unique pattern shaped like a fingerprint, but it disappears once we add more EtOH to X = 10, as illustrated in Figure e for m-SiO₂-5.

The template (P123) effect on the external morphology of the produced nanostructured mesopore silica is shown in Figure . Increasing the P123 content while keeping the other parameters constant significantly increases the shell thickness, as illustrated in the SEM images, creating particles with rougher surfaces. For example, the broken large shells in Figure f, for m-SiO₂-6 with Y = 0.01, look thin and fragile, but increasing it to 0.015 for m-SiO₂-3 increases the broken shells’ wall thickness to approximately 0.37 μm, as shown in Figure c. This aligns perfectly with the Ar sorption data, where the surface area, pore diameter, and total pore volume increase as the P123 template content increases. Intriguingly, increasing the template content to Y = 0.02 for m-SiO₂-7 produces particles with rougher surfaces without noticing any broken shells, as shown in Figure h, but with much debris in the mix, like m-SiO₂-5, although with the same unique pattern. This provides a better understanding of the template P123’s effect on the morphology toward a material with a more suitable design for intended applications in catalysis, adsorption, etc. For example, one way to synthesize micromacroporous zeolites uses mesoporous silica particles to act as a hard template during crystallization, , and the silica dissolution rate plays a critical role in creating the required macroporosity. So, mesoporous silica particles with thin walls like m-SiO₂-6 are a better choice as compared to m-SiO₂-7.

The effect of pH on the nanostructured mesoporous silica’s external morphology is also shown in Figure e,j. Increasing the pH to 3.0 seems to produce smaller particles ranging from 4 to 0.7 μm diameter and with a fragile structure and a significant amount of smaller debris, as illustrated in Figure i for m-SiO₂-8. This is compared to a pH of 1.2 for m-SiO₂-5, which produces a particle size range of 7–0.9 μm, as shown in Figure j.

Similarly, Figure g,k,l show the effect of the aerosol reactor temperature, T _r, on the external morphology of the produced nanostructured mesopore silica. Thin and fragile, broken large shells are expected to be seen due to low P123 molar ratio, Y = 0.01, as seen in Figure f for m-SiO₂-6 at 400 °C. Lowering the temperature to 300 °C produces a similar morphology with some incomplete spheres. Operating at T _r = 200 °C produces more even particles and shells that are slightly thicker, allowing some to maintain their integrity, as demonstrated in Figure k. From the SEM image in Figure l for m-SiO₂-11, it seems that 120 °C was insufficient to produce silica with a spherical morphology but this created agglomerated and fused silica spheres. To avoid this, T _r must be above 200 °C, and the selection of T _r will depend on the surface area, pore diameter, and total pore volume for the intended application.

Figure S4 shows more SEM images for nanostructured mesoporous silica m-SiO₂-1, m-SiO₂-2, m-SiO₂-3, m-SiO₂-4, m-SiO₂-5, m-SiO₂-6, m-SiO₂-7, and m-SiO₂-8.

The thick, fractured shells lead to a question about the nature of the mesoporous silica spheres: Are they solid or do they have a hollow core–shell structure? Further investigation was conducted using SEM with focused ion beam (FIB) milling using gallium ions to cut one of the m-SiO₂-3 particles, as shown in Figure .

5.

SEM image for spherical SBA-15 mesoporous silica (m-SiO₂-3) with FIB milling to expose the solid inner structure, and “Pa 1” shows the channel diameter of 10 nm.

The m-SiO₂-3 particle, after FIB milling, exposes its internal structure. The exposed cross-section shows a compact/dense morphology rather than an empty hollow core, indicating that the spherical particles are predominantly dense mesoporous silica with clear pore channels. One channel was measured to be approximately 10 nm wide, similar to the pore diameter measured by Ar adsorption at 11 nm. These observations suggest the formation of two types of spheres: the first is solid spheres, which are seen in the previous SEM images for relatively small particles, and the second is a hollow core–shell structure, which is indicated by the thick, fractured shells for relatively large particles. Also, for the large droplets, it seems that EtOH will evaporate faster than the droplets shrink, forcing the TEOS to diffuse to the droplet’s outer layer and start forming and thickening the shell. The final particle size will depend on each droplet’s initial size and precursor concentration within the droplet, which depends heavily on the nebulizer’s characteristics. In this case, the Collison nebulizer was used because of its easy, reliable, and scalable operation, but the variable output significantly impacted product formation, quality, and particle size distribution. Therefore, for future work, it is recommended to use constant-output nebulizers or remove the large droplets with aerosol traps specially designed for specific sizes.

3.6. Transmission Electron Microscopy Analysis

TEM provides insights into the internal microstructural morphology of individual particles, such as the organization of the mesopores, and the pore sizes. Figure displays TEM images for the nanostructured mesoporous silica samples prepared in this study. It should be noted that TEM provides local information from selected particles and specific orientations, whereas SAXS provides a bulk-averaged measure of mesostructural ordering. Therefore, TEM and SAXS are interpreted together, but a direct one-to-one correlation is not always expected, especially for samples with disordered regions, broad pore-size distributions, or variable pore orientations.

6.

TEM images of AA-EISA-derived mesoporous silica samples: (a) m-SiO₂-1, (b) m-SiO₂-2, (c) m-SiO₂-3, (d) m-SiO₂-4, (e) m-SiO₂-5, (f) m-SiO₂-6, (g) m-SiO₂-7, (h) m-SiO₂-8, (i) m-SiO₂-9, (j) m-SiO₂-10, and (k) m-SiO₂-11. The samples correspond to variations in EtOH/TEOS molar ratio, P123/TEOS molar ratio, solution pH, and reactor temperature, as summarized in Table .

Figure a presents a high-resolution TEM image and a zoomed-in part for m-SiO₂-1, which reveals a relatively consistent mesopore size throughout the particle and hexagonal arrangement of the pores. This is consistent with the well-defined peaks observed via SAXS analysis, as illustrated in Figure . Furthermore, it confirms a uniform pore size distribution, with an average diameter of around 6 nm. EtOH addition produces a structure with concentric layers, and an irregular and nonuniform distance between the layers, as shown in Figure b–d, and e for m-SiO₂-2, 3, 4, and 5. These images show onion layer formation and a folded structure induced by evaporation during AA-EISA. The rose-like structure in the layers seems to become more compact and folded as the EtOH content increases. For these samples, the TEM images mainly reflect local concentric or folded mesopore domains, while the SAXS patterns represent the average ordering across the whole sample.

By maintaining X = 3 for EtOH and reducing the P123 content to Y = 0.010 for m-SiO₂-6 (Figure f), the structure transforms into short, arc-shaped stripe domains with frequent discontinuities and misalignments. This microstructure indicates a smaller micelle population and weaker packing correlations, which is consistent with the lower pore diameter and pore volume obtained from gas sorption. On the other hand, increasing the P123 content to Y = 0.020 for m-SiO₂-7 (Figure g) maintained the same concentric, rose-like structure as m-SiO₂-3, shown in Figure c, but with more tightly folded layers. The pH plays a critical role in the TEOS hydrolysis rate during the silica layer formation, so increasing it to 3 produces a loose concentric, rose-like structure with a more complex pore network for m-SiO₂-8 as shown in Figure h, and consistent with the broad pore size distribution and large average diameter of 21 nm noted earlier.

Reducing T _r from 400 °C for m-SiO₂-6 to 300 and 200 °C did not change the morphology much for m-SiO₂-9 and m-SiO₂-10, as shown in Figure i,j. However, reducing it to 120 °C destroys the complex pore network, and some of the fused agglomerated silica forms a hexagonal pattern, as illustrated in Figure k for m-SiO₂-11.

3.7. Multi-Technique Analysis

Table indicates that the mesoporous silica produced through AA-EISA without EtOH (denoted m-SiO₂-1) has the highest surface area of 611 m²/g. These findings are consistent with those of traditional SBA-15 synthesized using the sol–gel method, which typically has a surface area of around 600–700 m²/g. However, this value is significantly higher than a previously reported value of 298 m²/g for the surface area of SBA-15 synthesized with AA-EISA without EtOH. The difference in surface areas is likely linked to the stability of the precursor solution during synthesis and over time, which suggests that this is a critical factor in determining the resulting surface area. The visual observations during the synthesis support this theory, as the precursor solution changes from a clear to a milky solution 3 h after the start of the atomization, similar to the milky color for the traditional SBA-15 precursor solution during synthesis.

Figure S1 supports this statement, showing the stability of the solution after 9 h of ethanol addition in a batch test. Also, SAXS, as well as SEM and TEM imaging, further confirmed the formation of SBA-15 with a high surface area. X = 1 was sufficient to produce a spherical morphology during continuous operation up to 12 h, but it reduced the surface area by more than half and increased the pore diameter by ∼70%, while only slightly affecting the total pore volume. For production extending beyond 12 h, a ratio of 3 was required to ensure precursor stability and to balance surface area with pore volume. In the absence of ethanol, a spherical morphology could not be sustained, indicating that ethanol plays a critical role in stabilizing micelle formation during AA-EISA. Further increasing the EtOH content tended to marginally enhance the surface area and total pore volume while maintaining the pore diameter nearly constant.

Also, increasing the amount of P123 increases the number and size of these micelles, leading to larger and more pores. As T _r was varied, the behavior of the P123 templating agent at higher temperatures impacted the micelles to grow larger and more uniformly, creating larger and more regular pores, as shown in Figure S2c. ,, The pore size distributions for m-SiO₂-10 show broader pore size distributions with split peaks at 4.5 and 9 nm. Both m-SiO₂-6 and m-SiO₂-9 show similar wide pore size distributions with 9 and 10.5 nm split peaks, which indicates that 300 °C is the limit for pore expansion due to the expansion capacity of the micelles. On the other hand, lowering the temperature to 120 °C reduces the pore expansion and hinders the micellar growth, which narrows the pore size distributions for m-SiO₂-11 significantly to 6 nm pore width. Also, it was observed that the amount of silica produced at pH = 3.0 (m-SiO₂-8) was almost half that produced at pH = 1.2 (m-SiO₂-5). On the other hand, this increased the pore diameter by 89% from 11 to 21 nm.

Consistent with these trends, SAXS shows a strong (100) reflection of the p6 mm lattice with shifts in q₁₀₀ tracking lattice dilation; peak broadening at higher EtOH or higher P123/TEOS aligns with the broader pore-size distributions from gas sorption, and SEM and TEM confirm preservation (or loss) of spherical morphology depending on EtOH level and temperature.

Integrated thermal analysis (TGA/DSC) further corroborates the structure–property map. Three regimes are observed by TGA: 25–200 °C for desorption of physisorbed water/solvent, 200–400 °C for Pluronic P123 decomposition, and >400 °C for assessment of framework stability. In the EtOH series (m-SiO₂-1 to m-SiO₂-5), the template-related mass loss between 200 and 400 °C drops from 22 wt % without EtOH (m-SiO₂-1) to 5–8 wt % with EtOH present (m-SiO₂-2 to 5), while the final residue increases from 21 wt % to as high as 51 wt %, indicating less template entrapment and more efficient removal of template when ethanol is used. The DSC traces (exotherm up) show two reproducible features that align with these steps: a small exotherm at 130–160 °C accompanying volatile removal and a broad exotherm near 600 °C attributed to burnout of residual carbon and onset of silica densification.

Varying Y (P123/TEOS) from 0.010 to 0.020 yields a nearly constant 200–400 °C loss of about 7 wt % (m-SiO₂-6, −3, −7), consistent with near-saturation of incorporated template; increasing pH from ∼ 1.2 (m-SiO₂-5) to ∼ 3.0 (m-SiO₂-8) slightly increases the 25–200 °C loss (from 39 to 43 wt %) at similar template loss (∼8 wt %) and maintains ≤2 wt % loss above 400 °C, confirming comparable high-temperature stability across pH. The first-phase loss is largest, and its onset is at slightly higher temperature in the EtOH-free sample, consistent with narrower pores that retain water more strongly. EtOH-stabilized spherical mesopores release volatiles at lower temperature.

Collectively, the concordance of adsorption, SAXS, SEM and TEM with TGA/DSC analysis demonstrates that moderate EtOH, optimized P123/TEOS, and T _r ≈ 200–300 °C maximize mesostructural order and yield while preserving robust thermal stability.

3.8. Industrial Assessment for Scaleup and Commercialization

This work reduced the AA-EISA reactor temperatures to synthesize spherical mesoporous silica with a high surface area by using decreased amounts of EtOH, P123, and HCl, which offers opportunities to make this synthesis route more viable and attractive for industrial production and commercialization.

3.8.1. Process Safety Considerations in Ethanol-Based AA-EISA Production

Ethanol is a flammable solvent with a low flash point (approximately 13 °C) and a broad flammability range (explosive limits between 3.1 and 27.7 vol % in air), presenting serious fire or explosion risks during aerosolization and thermal processing. Usually, nitrogen gas is used as an inert carrier in lab-scale synthesis and production to mitigate the fire or explosion risks. However, this is not a cost-effective solution for industrial production. Reducing the ethanol content in the AA-EISA production process offers considerable safety and sustainability benefits, especially for large-scale operations. This can be achieved by reducing the recipe’s EtOH content to maintain the reactor’s vapor concentration below flammable limits, mitigating ignition hazards. This adjustment simplifies safety compliance, diminishes the need for extensive inert gas blanketing, and minimizes the risk of vapor accumulation in closed systems.

Moreover, operating with reduced ethanol enhances energy efficiency, lowers solvent handling and recovery costs, and aligns with green manufacturing principles. Therefore, optimizing ethanol content is essential for facilitating safer, more cost-effective, and environmentally responsible large-scale production of mesoporous silica. This study shows that the EtOH/TEOS molar ratio, X, can be reduced to 1 while maintaining a spherical morphology, whereas X = 3 is optimal for ensuring precursor feed stability and balancing surface area and pore volume. The amount of ethanol released during the TEOS hydrolysis is calculated as follows (one mole of TEOS produces 4 mol of EtOH)

$$\mathrm{E}\mathrm{t}\mathrm{O}\mathrm{H}\mathrm{r}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{s}\mathrm{e}\mathrm{d}=\frac{17.19\mathrm{g}}{208.33\mathrm{g}/\mathrm{m}\mathrm{o}\mathrm{l}\mathrm{e}}{\mathrm{T}\mathrm{E}\mathrm{O}\mathrm{S}}^{*}{\frac{4\mathrm{m}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{s}\mathrm{E}\mathrm{t}\mathrm{O}\mathrm{H}}{1\mathrm{m}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{T}\mathrm{E}\mathrm{O}\mathrm{S}}}^{*}\frac{46.07\mathrm{g}}{\mathrm{m}\mathrm{o}\mathrm{l}\mathrm{e}}\mathrm{E}\mathrm{t}\mathrm{O}\mathrm{H}=15.21\mathrm{g}\mathrm{E}\mathrm{t}\mathrm{O}\mathrm{H}$$ 6

Adding this to the added ethanol for a molar ratio of 3 (11.40 g) resulted in 26.61 g of ethanol in the mixture, which is approximately 33.73 mL of EtOH using 0.79 g/mL as the density. The total ethanol concentration in the mixture is 11.62 vol %. Assuming complete evaporation at 25 °C and 1 atm at an atomization rate of 0.12 mL min⁻¹, results in a liquid ethanol flowrate of 0.01394 mL min⁻¹. Multiplying this by the density of ethanol at 25 °C (0.789 g mL⁻¹) and dividing by its molar mass (46.07 g mol⁻¹) leads to a molar flow rate of 2.39 × 10⁻⁴ mol min⁻¹. Applying the ideal-gas law by multiplying it by the ideal gas molar volume of ∼ 24.47 L mol⁻¹ to convert it to gas volume, results in an ethanol vapor flow rate of ∼5.86 mL min⁻¹. With the Collison nebulizer delivering N² at 5000 mL min⁻¹, the total gas stream is ∼5006 mL min⁻¹, giving an ethanol concentration of:

$$\mathrm{E}\mathrm{t}\mathrm{O}\mathrm{H}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{a}\mathrm{s}=\frac{\mathrm{E}\mathrm{t}\mathrm{O}\mathrm{H}\mathrm{Vapor}\mathrm{V}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{F}\mathrm{l}\mathrm{o}\mathrm{w}\mathrm{R}\mathrm{a}\mathrm{t}\mathrm{e}}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{G}\mathrm{a}\mathrm{s}\mathrm{V}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{F}\mathrm{l}\mathrm{o}\mathrm{w}\mathrm{R}\mathrm{a}\mathrm{t}\mathrm{e}}=\frac{5.86\mathrm{m}\mathrm{L}/\min }{5006\mathrm{m}\mathrm{L}/\min }=0.12\mathrm{v}\mathrm{o}\mathrm{l}\%$$ 7

This is far below the lower flammability limit (LFL) of ethanol–air mixtures (∼3.3 vol %), confirming the operation is well within safe limits for explosion or ignition hazards under nitrogen or even air.

3.8.2. Operational and Safety Considerations for P123 in AA-EISA Production

The concentration of the P123 template is crucial for the efficiency and sustainability of commercial-scale AA-EISA synthesis. Higher levels of P123 increase the viscosity of the precursor solution, which can lead to blockages of lines and the filter, material build-up inside the reactor, and potential fouling or burnout. These issues elevate maintenance costs and downtime, raising operational expenditures (OPEX). From the capital expenditure perspective (CAPEX), frequent cleaning and reactor maintenance may require more robust equipment or process modifications, thus increasing initial investment costs.

Environmentally speaking, excessive use of P123 releases more CO₂ during the calcination step, where the template is burned to create the mesoporosity, and can generate larger volumes of unreacted surfactant-containing waste streams that need treatment to prevent environmental release and ecological harm due to potential bioaccumulation. This can complicate waste management and increase associated costs.

Therefore, reducing the concentration of P123 to an optimal level (balancing template efficiency with operational stability) minimizes environmental impact, lowers CAPEX and OPEX, and ensures safer, more sustainable production of mesoporous silica materials. The results showed that a P123/TEOS molar ratio, Y = 0.015 was sufficient to achieve high-quality mesostructures with reduced reactor fouling and filter clogging. Reducing the molar ratio to Y = 0.01 was more effective in improving operation, reducing filter blockage, and preventing material accumulation and burnout inside the reactor tube, though this will lower the surface area, average pore diameter, and total pore volume.

3.8.3. Increasing pH by Reducing Acid Addition: Environmental & CAPEX Impact

Elevating the synthesis pH by decreasing the use of HCl acid in the AA-EISA process significantly diminishes the environmental impact and chemical footprint associated with it. Acidic effluents necessitate neutralization prior to disposal, and extensive acid usage contributes to corrosive waste streams and poses greater environmental hazards. Reducing acid addition makes the process greener and safer, aligning with sustainable manufacturing practices.

Operating at a higher pH leads to several benefits, including: (1) Reduced corrosion rates of reactor tubing and system components, which lessens the need for costly corrosion-resistant materials like Inconel or Teflon linings. (2) Decreased maintenance frequency and downtime, resulting in improved equipment longevity and lower replacement costs. (3) Simplified neutralization and post-treatment infrastructure (smaller reactors or reduced caustic dosing) minimize CAPEX for effluent treatment. The corrosion rate can increase significantly due to the synergistic effect between ethanol and chloride in HCl.

Consequently, increasing the pH from 1.2 to approximately 3.0 enhances environmental compliance and reduces CAPEX, although this comes at the expense of lower product yield and surface area. Reducing the acid content and adjusting the pH to 3.0 increases the average pore diameter to 20.1 nm and total pore volume to 1 cm³/g, but with a moderate decline in surface area to 194 m²/g and reduced yield. These findings present new opportunities for tailoring mesoporous structures for specific applications, such as enzyme immobilization, while lowering acid consumption and environmental impacts.

3.8.4. Reactor Temperature Considerations for Commercial-Scale AA-EISA Synthesis of Mesoporous Silica

In the context of scaling up the AA-EISA method for synthesizing mesoporous silica, T _r is a critical parameter that directly affects product quality, operational safety, environmental performance, and economic viability. Empirical evidence shows that T _r ∼ 200 °C strikes an optimal balance. This temperature ensures sufficient solvent evaporation to facilitate micelle formation and structural ordering while maintaining the final material’s spherical morphology.

On the other hand, reducing T _r to 120 °C, although it may lower energy consumption, leads to incomplete self-assembly, structural deformation, and diminished textural properties. From environmental and economic perspectives, operating at moderate temperatures decreases fuel consumption, minimizes CO₂ emissions, and limits the thermal degradation of reagents and reactor materials. This approach contributes to cost savings and enhances sustainability. Safety considerations are critical at commercial scales, where solvent volatility poses significant risks. Keeping reactor operating temperatures low helps mitigate these hazards, preventing excessive pressure build-up and reducing the likelihood of ignition or template degradation. Therefore, T _r around 200 °C is recommended for large-scale AA-EISA operations as it offers a favorable compromise between mesostructure fidelity, process safety, environmental responsibility, and economic feasibility. Lowering T_r from 400 to 200 °C saves energy with some impact on the textural properties and morphology of the produced silica.

3.8.5. Production Scalability for Spherical SBA-15 via AA-EISA

The current setup was designed for the continuous, lab-scale production of spherical nanostructured porous silica via AA-EISA. The molar composition used for this synthesis was designed to reduce the ethanol content and ensure stable feed for smooth, continuous production. The scalability of the method was demonstrated by sustained operation over multiple days, with production yields varying based on synthesis conditions and run duration. For example, a typical run of the m-SiO₂-10 formulation yielded approximately 57% recovery during the first 15 h, while extending the production for another 26 h increased the yield to 62%. This resulted in a cumulative 60% average recovery from the filter paper. The recovery yields were calculated based on the actual collected calcined sample over the expected amount of produced silica, assuming 100% conversion of TEOS. This substantial enhancement in throughput, combined with the preservation of morphological and textural uniformity, underscores AA-EISA’s promise as a robust and potentially industrially viable strategy for continuously manufacturing mesostructured materials, although scale-up studies will have to be carried out to validate these findings.

4. Conclusion

Spherical mesoporous silica nanoparticles were effectively synthesized using the AA-EISA method, employing SBA-15 precursors, focusing on investigating the influence of EtOH, P123, and pH on the structural properties. The effect of each parameter was determined using Ar and N₂ gas physisorption, TGA, SAXS, and multiple microscopic techniques such as SEM and TEM.

These characterization studies suggest that the EtOH addition plays a pivotal role in stabilizing the precursor solution and preventing conventional SBA-15 formation by impeding P123 micelle growth. The optimal EtOH addition was determined to be a molar ratio of 3 (normalized to TEOS) over a continuous synthesis period of 1–3 days via AA-EISA. This enhances industrial scalability by minimizing EtOH usage, thereby reducing fire or explosion hazards. Increasing the P123 template content while maintaining other parameters can significantly enhance surface area, pore diameter, total pore volume, and sphere wall thickness. It is advisible to utilize less than a 0.020 molar ratio of P123 (normalized to TEOS), particularly for applications requiring rapid silica dissolution rates. Demonstration of successful synthesis at increased pH from 1.2 to 3.0 via AA-EISA presents a significant cost-saving opportunity in terms of capital, operation, and maintenance. This adjustment augments pore diameter and total pore volume but diminishes surface area. Increasing the reaction temperature while keeping other parameters constant substantially boosts surface area, pore diameter, and total pore volume. Optimal operating conditions typically involve temperatures above 120 °C, preferably around 200 °C, depending on specific application requirements.

Therefore, the AA-EISA approach represents a versatile and efficient method for producing structured materials. This technique outperforms traditional methods by merging aerosol processing with solvent evaporation-driven self-assembly. It enhances production capacity and allows for frequent quality checks through regular sampling, improving consistency and reliability while lowering production costs. Additionally, AA-EISA enables precise control of material properties, thereby enhancing performance and functionality in the final products. Its capacity to generate uniform, high-quality materials on a larger scale signifies a significant advancement over traditional fabrication methods. The AA-EISA precursor preparation time was reduced to 2 h at room temperature compared to the conventional 24–48 h. Moreover, the production yield varies depending on the conditions and the run length. For example, in m-SiO₂-10, the cumulative yields for recovery were 60%, but the yield improved by approximately 7%, which is a significant improvement for industrial applications. Future work should focus on increasing the run time for more than 2 days and investigating the intensification of the atomization process.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/cbe.6c00038.

• Additional experimental details, synthesis, SAXS analysis, Ar physisorption isotherms and pore-size distributions, SEM images, and supplementary tables (PDF)

The authors declare no competing financial interest.