Exploring Adsorption Performance of Functionalized Mesoporous Silicas with a Different Pore Structure as Strong Cation-Exchange Sorbents for Solid-Phase Extraction of Atropine and Scopolamine

Abstract

In this work, mesoporous silicas with two types of mesoporous structures were synthesized and functionalized with sulfonic acid groups: MCM-41-SO₃H (honeycomb-like hexagonal structure) and MSU-2-SO₃H (three-dimensional porous structure with wormhole pores). The synthesized materials were characterized by X-ray diffraction, transmission electron microscopy, scanning electron microscopy, nitrogen adsorption–desorption, Fourier-transform infrared spectroscopy, ²⁹Si solid-state nuclear magnetic resonance spectroscopy, and elemental analysis. The obtained functionalized materials were evaluated as sorbents for strong cation-exchange solid-phase extraction (SPE) to determine their efficiency in the adsorption and desorption of tropane alkaloids (atropine and scopolamine). The loading solvents, loading volume, analyte concentration, and elution volume were studied, using 50 mg of both materials. Analyses were carried out by ultra-high performance liquid chromatography coupled to triple quadrupole tandem mass spectrometry. The synthesized MCM-41-SO₃H material presented the highest recovery efficiency and has proven to be a promising sorbent for strong cation-exchange SPE of atropine and scopolamine in aqueous media. The high degree of functionalization of MCM-41-SO₃H and the high accessibility of the sulfonic groups for the target analytes, due to the regularity and uniformity of their pores, maximize the contact between the alkaloids and the sorbent, favoring efficient adsorption.

1. Introduction

In recent years, the application of mesoporous silicas in sorbent-based extraction techniques for sample preparation has gained research interest [1]. This is due to the unique properties and remarkable advantages of these materials such as (i) large surface area and high pore volume, with uniform mesopores of 2–10 nm; (ii) well-defined pore network arrangement (e.g., 2D hexagonal, 3D cubic, wormhole, etc.); (iii) tunability of the morphology and size of the particles; and (iv) good thermal and chemical stability. Furthermore, mesoporous silicas can be easily functionalized with organic ligands that allow for tuning the chemical properties of the surface and expanding the range of their application for the extraction and purification of various analytes by solid-phase extraction (SPE) [2]. The choice of the type of sorbent packed in cartridges for SPE is crucial, as it regulates aspects such as capacity and selectivity, depending on the interaction of the sorbent with the functional groups of the analyte (e.g., hydrophilic, hydrophobic, ion-exchange, or mixed-mode) [3]. In this sense, the selection of the appropriate mesoporous silica sorbent is important to optimize the extraction performance of the SPE process since it determines the specific surface area, the accessibility of the active sites, and the interaction with the target molecules, allowing better recovery and purification of the analytes in a complex matrix [4].

Although SBA-15 (Santa Barbara Amorphous type material-15) has been one of the most studied mesoporous silicas as sorbent for SPE [5,6,7], other mesoporous silicas also show great potential. For example, MCM-41 (Mobil Composition of Matter) and MSU-X (Michigan State University material) are widely recognized for their adsorptive properties. MCM-41, with its small-pore hexagonal structure, exhibits a remarkable order that facilitates the adsorption of a variety of analytes, including metals [8,9,10], pesticides [11,12], and pharmaceuticals [13,14,15] in different samples. On the other hand, MSU-X lacks a regular packing order but exhibits a three-dimensional porous structure with wormhole-like pores and uniform diameters, which gives it versatility in the adsorption of substances in complex matrices, including applications in the analysis of pesticides [16,17] or metals [18], among others. Nevertheless, there is still a requirement to investigate the potential of new sorbents for SPE with diverse core structures and functionalities, depending on the specific use for which they are intended, to promote the retention of different analytes and achieve high enrichment and good recovery. These factors are mandatory to reduce detection limits to a level compatible with the detection technique [19]. Furthermore, thanks to these new sorbents, miniaturized sample preparation procedures can be developed, using smaller quantities of sorbents, solvents, and reagents, to comply with the precepts of Green Analytical Chemistry (GAC) [20,21]. This is of particular importance in food safety, where sample preparation is a critical and indispensable step to determine the presence of contaminants at low concentrations in complex food matrices [2,22].

Nowadays, there is great interest in the development of analytical methodologies that facilitate the monitoring of natural toxins in food products, to comply with current regulations and thus contribute to food safety [23]. For example, the recently published Commission Regulation (EU) 2023/915 establishes the need to control tropane alkaloids (TAs), setting limits for atropine and scopolamine in processed cereal-based infant foods, unprocessed sorghum, millet, maize, buckwheat as well as for herbal infusions, ranging from 1 to 50 µg/kg as the sum of atropine and scopolamine [24]. TAs are secondary metabolites naturally produced by different plants, mainly of the Solanaceae family. These alkaloids are known for their toxicity, which is due to their ability to bind to muscarinic acetylcholine receptors in the central and autonomic nervous system, causing symptoms such as dry mucous membranes, ataxia, and hallucinations, and in high doses can even be lethal [25]. Food contamination with TAs has become an issue of concern due to the increase in cases of food poisoning. The main contamination pathways are often related to the occurrence of TAs-producing plants in fields for crop production that can lead to contamination of the harvested foods, being seeds of Datura stramonium as the common source of contamination. Given the wide range of products affected, it is crucial to implement stringent control and regulatory measures [26]. However, the detection of these alkaloids in food presents significant challenges, since they are often found in very low concentrations. In addition, the identification and quantification of these substances can be complicated by interferences arising from the complexity of food samples. In this context, proper sample preparation becomes an essential step for the accurate determination of alkaloids in food matrices [27].

To date, different commercial sorbents have been widely used for SPE of TAs in food analysis. In general, mixed-mode polymeric sorbents with anionic functional groups have been used to extract TAs in acidic solutions, as is the case of the commercial Oasis MCX^® and Strata-XC^® cartridges containing a chemically modified polymeric sorbent with polar and strong cation-exchange (SCX) groups [28,29,30]. Nevertheless, the development of new materials for their application as sorbents in SPE has become essential to obtain materials with greater adsorption capacity, which allows the development of miniaturized protocols that minimize the generation of waste in this sample preparation stage. In this context, mesoporous silicas have gained increasing interest as a viable alternative to commercial sorbents [27]. For example, González-Gómez et al. have evaluated sulfonic acid-functionalized SBA-15 mesoporous silicas (150 mg) as SPE sorbents for the analysis of TAs in gluten-free grains and flours and demonstrated excellent extraction performance compared to an analogous commercial material (MFE-PAK^® SCX) [30]. In addition, other mesoporous silicas such as HMS (hexagonal mesoporous silica) have been studied for atropine and scopolamine determination in thyme, basil, and coriander samples, using 75 mg of material in SPE showing recoveries close to 100% [31]. However, it is interesting to evaluate other mesoporous silicas not explored to date, presenting different pore structures and other textural characteristics in the evolution to green miniaturized sorbent-based sample preparation protocols.

In this sense, this work has aimed to evaluate the adsorption of atropine and scopolamine in two mesoporous materials with different pore structures. Through this evaluation, we seek to understand the potential of these materials as sorbents in SPE and how the specific characteristics of the pore structure affect the adsorption capacity. Additionally, we aim to develop more efficient sorbents that improve the accuracy and sensitivity in detecting these compounds in various matrices, thus contributing to significant advances in the control of natural toxins in foods.

2. Materials and Methods

2.1. Chemicals, Reagents, and Standard Solutions

The atropine reference standard (≥99%) and scopolamine hydrobromide (≥98%) were acquired from Sigma-Aldrich (St. Louis, MO, USA). To prepare the individual standard solutions (1000 mg/L), the precise amount of solid substance was weighed and dissolved in methanol (MeOH) of HPLC−MS quality purchased from Scharlab (Barcelona, Spain). The solution was then stored in the dark at −20 °C. Working solutions at the desired concentrations were prepared in MeOH and stored in the dark at −20 °C. Ultra-pure water (resistance 18.2 MΩ.cm) was purified using the Millipore Milli-Q-System (Billerica, MA, USA), ethanol (EtOH) absolute HPLC grade, and 32% (v/v) ammonia solution was acquired from Scharlab (Barcelona, Spain). Formic acid (FA) and acetonitrile (ACN) LC/MS grade were purchased from Fisher Chemical (Loughborough, UK). For the SPE process, empty 3 mL cartridges, polyethylene frits with a pore size of 0.20 µm, and nylon membrane filters with a pore size of 0.45 µm provided by Scharlab (Barcelona, Spain) were used.

Tetraethylorthosilicate (TEOS) 98% (MW = 208.33, d = 0.934 g/mL), Tergitol^® NP-9 (MW = 616.82), sodium silicate SiO₂·NaOH (MW = 242.23, d = 1.390 kg/m³), and cetyltrimethyammonium bromide (CTAB) (MW = 364.46) were purchased from Sigma-Aldrich (Schnelldorf, Germany), while sodium fluoride 99% (MW = 41.99) was purchased from Fluka (Buchs, Switzerland). Sulfuric acid 96% (MW = 98.08, d = 1.840 kg/m³) was acquired from Panreac (Barcelona, Spain). Hydrochloric acid 37% (MW = 36.46, d = 1.2 g/mL) and hydrogen peroxide 30% (MW = 34.01, d = 1.110 g/cm³) were obtained from Scharlab (Barcelona, Spain). (3-Mercaptopropyl) triethoxysilane 94% (MPTES) (MW = 196.34, d = 1.057 g/mL) was purchased from Alfa Aesar (Karlsruhe, Germany).

2.2. Synthesis and Functionalization of the Mesoporous Silicas

The synthesis of the MSU-2 was carried out following the previous work of Pérez-Quintanilla et al. [32]. A milky suspension was prepared by adding TEOS to a 0.08 M solution of Tergitol^® NP-9 in Milli-Q water. This suspension was then aged without stirring for 20 h, and then a 0.24 M sodium fluoride solution was added dropwise with stirring, aiming to obtain a NaF/TEOS molar ratio of 0.025/1. The resulting solution was placed in a shaking bath at 55 °C for 48 h. Finally, the product obtained was filtered, washed with Milli-Q water, dried, and calcined in air at 600 °C for 12 h.

The synthesis of the MCM-41 was carried out following the previous work of Pérez-Quintanilla et al. [33]. First, sodium silicate was prepared, with a composition of 14% NaOH, 27% SiO_2, and 59% water. Then, 4.20 g of 98% H₂SO₄ and 140 mL of water were added dropwise to 65.45 g of the sodium silicate solution. The solution was kept stirred at room temperature for 30 min. A total of 58.66 g of CTAB was dissolved in 176 mL of water and added to the mixture. After this, the resulting gel was mixed with 70 mL of water, keeping the process stirred for a period of 30 min at room temperature. The formed gel was transferred to a Teflon-coated autoclave and heated at 121 °C for 144 h. Subsequently, the resulting solid was separated by vacuum filtration and subjected to a drying process at 100 °C for 8 h. Finally, the surfactant was removed by a calcination process in air at 530 °C for 6 h.

MSU-2 and MCM-41 were subjected to further functionalization. For this, 2.5 g of bare mesoporous silica was suspended in 250 mL of 0.1 M HCl, and 1.0 g of MPTES was added. The mixture was stirred for 7 h at room temperature and transferred to a reactor for 24 h at 100 °C. The solid obtained was filtered, washed with Milli-Q water, and allowed to dry at 50 °C overnight. The recovered materials (MSU-2-SH or MCM-41-SH) were mixed with 300 mL of 2M HCl and then 11.6 g H₂O₂ (30%) was added to oxidize the thiols to sulfonic acid. After 5 min of stirring at room temperature, the mixture was heated in a reactor for 6 h at 100 °C. The resulting materials designated as MSU-2-SO₃H and MCM-41-SO₃H, were filtered and washed with Milli-Q water.

2.3. Characterization of Mesoporous Silicas

The synthesized materials were characterized by X-ray diffraction (XRD) to evaluate the structure of the materials. Low-angle powder XRD patterns of the silicas were performed to determine if the materials showed the typical long mesoscopic ordered structure. XRD patterns were obtained on a Philips Diffractometer model PW3040/00 X’Pert MPD/MRD (Philips, Amsterdam, The Netherlands) at 45 kV and 40 mA, using Cu Kα radiation (λ = 1.5418 Å). The samples were treated in power and placed in a sample holder. The sample and detector were rotated and the XRD patterns were collected from 0 to 10°. Transmission electron microscopy (TEM) images were performed on JEOL JEM 1010 (Tokyo, Japan) with an accelerating voltage of 80 kV. The samples were dispersed in acetone and deposited on carbon-supported grids. Scanning electron microscopy (SEM) images were scanned by a Nova Nano SEM230 (FEG-SEM) (Denton, TX, USA) with an energy-dispersive spectrometry system (EDS). The metallizer used was Leica ACE600 (Wetzlar, Germany), using a gold target and depositing a layer of less than 5 nm. Nitrogen gas adsorption–desorption isotherms were performed with a Micromeritics ASAP 2020 (Norcross, GA, USA) analyzer. The isotherms were measured at −196 °C with an interval of relative pressures (P/P₀) from 1 × 10⁻⁴ to 0.99. Previously, samples were degassed at 90 °C under vacuum for 10 h. The Brunauer–Emmett–Teller (BET) method was utilized to obtain the specific surface area (SBET), and the Barrett–Joyner–Halenda (BJH) model was used to calculate the pore volume and pore size distribution by the desorption branches of isotherms. The total pore volume (Vt) was estimated from the desorbed amount at a relative pressure P/P₀ of 0.97. Fourier-transform infrared (FT-IR) spectra were recorded with a Spotlight 200i, Perkin Elmer (Waltham, MA, USA) spectrometer in the region 4000–400 cm⁻¹ using KBr pellets with 16 scans and a resolution of 4 cm⁻¹. ²⁹Si solid-state nuclear magnetic resonance (²⁹Si-PDA-MAS-NMR) spectra were recorded on a Bruker Avance III-WB Spectrometer 400 MHz, operating at 79.4 MHz to ²⁹Si nucleus, Half-pi pulse: 6 μs, number of accumulations: 1000, rotation speed: 12,000 Hz, time between pulses: 60 s. Finally, elemental analysis (% S) was performed using a microanalyzer Flash 2000 Thermo Fisher Scientific Inc. (Hampton, NY, USA).

2.4. Evaluation of Mesoporous Silica as Sorbents in Solid-Phase Extraction

The functionalized mesoporous silicas prepared and characterized were tested as sorbents in SPE for TAs extraction before ultra-high performance liquid chromatography coupled to triple quadrupole tandem mass spectrometry (UHPLC-TQ-MS/MS) analysis. The SPE procedure was carried out by packing polypropylene cartridges with 50 mg of the prepared sorbents. Both ends were sealed with polyethylene frits. In addition, a nylon filter membrane (0.45 µm) was incorporated at the bottom of the bed to prevent material loss during the process. First, the cartridges were conditioned with 3 mL of Milli-Q water, followed by equilibration with 3 mL of the same solvent used for the loading step. They were then loaded with the standard solution at 0.8 mL/min. The cartridges were then vacuum dried using a Supelco Visiprep SPE 12-port vacuum manifold (Sigma Aldrich, St. Louis, MO, USA) connected to a vacuum pump at 7.6 psi for 3 min. Elution of the analytes retained on the cartridge was performed with 5% ammonia solution in MeOH, followed by another vacuum drying process for 3 min. Finally, the eluate was evaporated to dryness using a vacuum/nitrogen manifold system (Schlenk line) at ambient temperature and reconstituted in 0.3 mL of MeOH for subsequent injection into the UHPLC-TQ-MS/MS. Several loading solvents, including EtOH, MeOH, H₂O, and a mixture of MeOH:H₂O (50:50, v/v), all acidified with 1% HCl (pH 1) and analyte concentrations, in the range of 0.1 to 30 µg/L, were studied. In addition, sample loading volumes from 5 to 20 mL and analyte elution volumes from 3 to 6 mL, were studied (Figure 1). Studies were performed in triplicate, and the results were expressed as percentage recovery of atropine and scopolamine. The recovery was calculated by comparing the results obtained by UHPLC-MS/MS for a loading solvent spiked with the target analytes under the study conditions prior to the SPE with the result obtained by a loading solvent spiked with the analytes after the SPE (blank control).

2.5. UHPLC-TQ-MS/MS Analysis

Liquid chromatographic separations were carried out using an Elute UHPLC system (Bruker Daltonics Inc., Billerica, MA, USA), consisting of an Elute UHPLC HPG 1300 pump with two pairs of individually controlled linear drive pump heads coupled in series, an Elute autosampler maintained at 10 °C, and an Elute CSV column oven preheater equipped with an Intensity Solo 2 C18 column (100 mm × 2.1 mm, 2.0 μm, Bruker Daltonics Inc., Billerica, MA, USA) maintained at room temperature. Optimum separation conditions were achieved using a solvent mixture as a mobile phase consisting of 0.1% FA in water (solvent A) and 0.1% FA in ACN (solvent B) in a gradient elution mode as follows: 0.00 min 95% solvent A; 0.00–3.00 min 70% solvent A; 3.00–4.00 min 95% solvent A; 4.00–5.00 min 95% solvent A at a flow rate of 0.450 mL/min, and an injection volume of 2 µL (partial injection), resulting in an analysis time of 5 min, including equilibration time.

Detection was carried out using a Bruker EvoQ LC-TQ Elite triple quadrupole mass spectrometer (Bruker Daltonics Inc., Billerica, MA, USA) equipped with a heated electrospray ionization interface (HESI). The instrument was operated in positive ionization mode. The mass spectrum is registered with multiple reaction monitoring (MRM) of three transitions per compound (Q1 resolution 2.5; Q3 resolution 2.5; scan time 93.80 ms). Optimization of MRM collision-induced dissociation (CID) energies was performed by direct infusion of individual standard solutions (1 μg/mL) at a flow rate of 5 μL/min. HESI source parameters were optimized with multiple runs of UHPLC-MS/MS using gradient and flow rate-optimized conditions. The optimum HESI source parameters were: N2 as drying and nebulizer gas, argon as collision gas, collision pressure at 2.0 mTorr, spray voltage positive at 1535 V, spray amp positive 100.0 µA, cone pressure 20.0 psi, probe pressure 25.0 PSI, nebulizer pressure 60 psi, cone temperature 280 °C, and probe temperature 225 °C. The MS parameters for each analyte are listed in

Table 1. Retention time and optimal parameters to mass spectrometry detection by MRM mode for the analysis of atropine and scopolamine using the UHPLC-TQ-MS/MS method developed in this work.

Compound	Rt (min)	Precursor Ion (Q1, m/z)	Product Ions (Q3, m/z)	Collision Energy (eV)	Scan Time (ms)
Atropine	2.950	290.5	123.90 *	20.5	31.26
			93.60	29.0
			91.10	34.0
Scopolamine	2.465	303.5	137.90 *	12.0	23.45
			155.80	9.5
			120.60	16.0

* Productions used for quantification. Chromatographic conditions with the optimized gradient elution: 0.00 min 95% solvent A; 0.00–3.00 min 70% solvent A; 3.00–4.00 min 95% solvent A; 4.00–5.00 min 95% solvent A. Mobile phase: 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (solvent B). Flow rate: 0.450 mL/min. Column at room temperature and total run time analysis of 5 min.

. Therefore, the most intense production of each analyte in its MS² spectrum was used for quantification, while other less intense product ions (at least one of them was mandatory) were controlled for confirmation purposes (Table 1, Figure 2). The software tqControl 2.2.0.1 (Bruker Daltonics, Billerica MA, USA) was used for data processing. Linear ranges were established, from 0.007 to 100 µg/L for atropine and from 0.14 to 100 µg/L for scopolamine. LODs and LOQs values ranged from 0.002 to 0.007 µg/L for atropine and from 0.04 to 0.14 µg/L for scopolamine.

2.6. Statistical Analysis

Statistical analysis of the optimization process was performed with IBM SPSS Statistics version 29.0.2.0 (20) software (SPSS Inc., Chicago, IL, USA), using Student’s t-test for comparison of two means or one-way analysis of variance (ANOVA) and Duncan’s post hoc multiple range test (significant differences at p ≤ 0.05) for comparison of more than two means, and different letters were used to denote them. All experiments were performed in triplicate, and the results are presented as the means of the values ± standard deviation (SD) obtained in this investigation.

3. Results

Two mesoporous silicas with different structural characteristics such as MCM-41 and MSU-2 were evaluated for adsorption studies due to their well-defined properties and their ability to accommodate a variety of analytes. While MCM-41 presents an ordered, cylindrical pore structure, MSU-2 features a highly interconnected three-dimensional (3D) wormhole-like framework. These structural differences allow us to evaluate how the configuration of the material influences the adsorption process of the target analytes.

3.1. Characterization of Mesoporous Silica

3.1.1. X-Ray Diffraction

Figure 3 shows the XRD patterns for MCM-41 and MSU-2. The XRD pattern of MCM-41 shows a very sharp diffraction peak at 2.17° (100), accompanied by two additional peaks of lower intensity (110 and 200) at 3.76° and 4.36°, corresponding to d-spacing values of 40.63, 23.46, and 20.26 Å, respectively. A unit cell parameter, a₀, of 46.92 Å was obtained. The X-ray diffractogram of MSU-2 (Figure 3) shows a single, much broader peak, indicating that this material exhibits a lower mesoscopic order compared to MCM-41, with a d-spacing value of 60.25 Å, and a unit cell parameter, a₀, of 69.57 Å.

$$d={\displaystyle \frac{k\lambda }{\beta cos\theta }}$$

where k is a constant (k = 0.9), λ is the wavelength of X-rays (1.5418 Å), β is the full width at half-maxima of the diffraction peak line (in radians), and θ is the diffraction angle yielding values of 380 and 219 Å for MCM-41 and MSU-2, respectively. The obtained values of the crystal sizes of these materials are related to the sharpness and position of the diffractogram peaks. Consequently, the larger the crystals of a particular component, the sharper the peaks on the XRD pattern for each crystal plane. Additionally, the narrower X-ray diffraction peak of MCM-41 compared to the diffraction peak of MSU-2 is related not only to the crystal size but also to a more ordered pore distribution in MCM-41 than in MSU-2, as evidenced by the TEM images indicated in Section 3.1.2 [34,35].

3.1.2. Transmission Electron Microscopy Measurements

The type and pore distribution of the materials were evaluated by TEM, as illustrated in Figure 4. TEM micrographs of the MCM-41 sample exhibit a well-defined and ordered porous structure, showing a parallel arrangement of cylindrical pore channels in the [100] direction (Figure 4a) and a hexagonal honeycomb-like arrangement along the [001] axis (Figure 4a inset) [34]. On the other hand, the less ordered structure of the MSU-2 silica, with a wormhole-like pattern, is illustrated by the TEM micrographs in Figure 4b. These micrographs show that the silica particle contains a large number of channels that are regular in diameter, although they lack long-range packing order. Furthermore, TEM confirms that the entire material is composed of this porous structure, with no non-porous components present, and reveals the spherical shape of the particles [35].

3.1.3. Scanning Electron Microscopy Measurements

SEM micrographs of the mesoporous silica MCM-41 (Figure 5a) reveal agglomerates of curved prismatic rods (with particle sizes of approximately 2 µm in length and 0.5 µm in width), which form large particles of disordered shapes due to their interlocking [36]. The SEM images of MSU-2 show spherical particles with a very homogeneous size distribution around 2 µm with a sponge-like appearance (Figure 5b) [35].

3.1.4. Nitrogen Adsorption–Desorption Isotherms

The adsorption isotherms for both materials are type IV according to the IUPAC, in the isotherms, an abrupt increase in adsorbed volume occurs at a relative pressure (P/P₀) of approximately 0.4, indicative of nitrogen capillary condensation within the mesopore structure. The isotherms also show that the adsorption volume is higher in the case of MCM-41 due to its larger pore volume. The hysteresis loop presented by MCM-41 and MSU-2 is of type H1 according to the IUPAC classification, which is associated with porous materials consisting of regular-shaped aggregates or compacts and a narrow pore size distribution [35,36]. This is consistent with what has been observed in the SEM and TEM images for these materials. The average BJH pore diameter for MCM-41 is 26.6 Å, and the wall thickness is 26.32 Å. On the other hand, for MSU-2, pore diameter of 33.6 Å and wall thickness of 18.58 Å was found. In the case of the functionalized materials, it can be seen that the attachment of organic ligands to the pore walls has increased the effective thickness of the walls. This does not significantly alter the pore diameter but has reduced the internal volume of the pores and surface area as can be seen in the data on their textural properties. These data on wall thickness, volume, and pore diameter demonstrate why MCM-41 has a higher number of pores per gram of material. All these physical parameters obtained from nitrogen isotherms, such as the BET surface area, average BJH pore diameter, pore volume, and wall thickness of the materials are presented in

Table 2. Textural properties and functionalization degree of the prepared functionalized mesoporous silicas.

Sample	BET Specific Surface Area a (m2/g)	Pore Diameter BJH b (Å)	Pore Volume c (cm3/g)	Wall Thickness d (Å)	L0 e (mmol SO3H/g)
MCM-41	1101	26.6	1.03	26.32	-
MCM-41-SO3H	870	25.3	0.69	-	1.12
MSU-2	852	33.6	0.78	18.58	-
MSU-2-SO3H	648	25.6	0.52	-	1.08

^a BET: Specific surface area calculated by the Brunauer–Emmett–Teller method. ^b BJH: Pore size distribution obtained using the Barret–Joyner–Halenda model on the desorption branch. ^c Pore volume: Total pore volume measured at relative P/P⁰ = 0.97. ^d Wall thickness for MCM-41 = a₀—pore size, where $a_0={\displaystyle \frac{2}{\sqrt{3}}}· d_{100}$ and Wall thickness for MSU-2 = d₁₀₀—BJH pore diameter. ^e L₀: Degree of functionalization estimated through the sulfur percentage determined by elemental analysis.

, where it is observed that the MCM-41 has a larger surface area (1101 m²/g) than the MSU-2 (852 m²/g).

Figure 6 shows the pore size distribution using the BJH method in the desorption branch of the functionalized and unfunctionalized materials. It can be observed that MCM-41 presents a narrower pore distribution than MSU-2 (Figure 6a), indicative of a higher uniform mesoporosity in the structure. After functionalization (Figure 6b), a decrease in SBET, BJH pore diameter, and pore volume took place (see Table 2) in both materials. The adsorbed volume in the adsorption isotherms of the two materials decreased, and the hysteresis loops were reduced (Figure 6c,d). These changes can be attributed to the presence of sulfonic groups on the surface and pores of the materials, which partially block nitrogen molecule adsorption. Additionally, it is observed that the inflection point of the isotherm in the adsorption branch is shifted slightly toward lower relative pressures because of the earlier filling of the monolayer due to the reduction in the pore volume of the material [37].

3.1.5. FT-IR Spectra

Figure 7 shows the FT-IR patterns for the functionalized and non-functionalized materials covering a range from 4000 to 400 cm⁻¹. The functionalized materials show the characteristic bands of the organic ligands attached to the material. The peaks observed in the spectra at 1620 cm⁻¹ and 3400 cm⁻¹ have been attributed to OH vibration and stretching vibrations of physisorbed water. Also, a weak peak is observed at 3640 cm⁻¹ associated with the surface hydroxyl groups involved in forming hydrogen bonds with other neighboring silanol groups. The intensity of these signals is reduced in the functionalized materials because of the decrease in the number of free silanol groups. The presence of sulfonic groups in functionalized mesoporous silica is confirmed by the identification of bands characteristic of the ligands, like deformation vibration bands of -CH₂ detected at 1447 cm⁻¹ due to the alkyl chains of the ligand. The peaks located around 1410 cm⁻¹ correspond to S=O stretching vibrations present in undissociated acidic sulfonic groups, while the peak at 1344 cm⁻¹ is attributed to stretching vibrations of -SO₃⁻ species. The spectra also can be identified as characteristic bands for aliphatic stretching vibrations (C-H) of pendant alkyl chains located around 3000–2800 cm⁻¹. In addition, bands corresponding to the Si-O vibration at 790 cm⁻¹, the symmetric Si-OH vibration at 950 cm^−1, and a large band between 1000 and 1240 cm⁻¹ attributed to the asymmetric Si-O-Si vibrations are distinguished [38]. The greater intensity of these bands in the functionalized MCM-41 is a consequence of the higher degree of functionalization that this material presents compared to MSU-2.

3.1.6. NMR Spectra

Figure 8 shows the ²⁹Si-PDA-MAS-NMR spectra of the functionalized and non-functionalized materials.

In these spectra, it can be observed that the two materials present three signals due to the structural silica sites Q⁴ ((SiO)₄Si) at −110 ppm (MCM-41 and MCM-41-SO₃H), at −109 ppm (MSU-2 and MSU-2-SO₃H), and the silanol sites Q³ ((SiO)₃SiOH), and Q² sites ((SiO)₂Si(OH)₂), (−102 ppm and −91 ppm for MCM-41 and MCM-41-SO₃H), (−101 ppm and −90 ppm for MSU-2 and MSU-2-SO₃H). In the two materials, the signal from the Q⁴ sites is the most intense, which indicates a high degree of condensation. In the case of MCM-41-SO₃H, it is observed that the Q³ and Q² sites are more intense than in MSU-2-SO₃H, which indicates a greater presence of surface silanol groups [39]. Additionally, after the functionalization, two signals can be observed at a lower chemical shift (δ) due to the Si atoms bonded to the carbons of the sulfonic ligands (-R), T sites, corresponding to the T³ ((SiO)₃Si-R) (−67 and −66 ppm for MCM-41-SO₃H and MSU-2-SO₃H, respectively), and T² ((SiO)₂Si(OH)-R) (−57 ppm in both materials) sites. The high intensity of T sites in the functionalized materials indicates a high degree of functionalization as is confirmed by the elemental analysis.

3.1.7. Elemental Analysis

Finally, the degree of functionalization (L₀, mmol of the sulfonic groups attached to mesoporous silica) of MCM-41-SO₃H and MSU-2-SO₃H were estimated based on sulfur content calculated through elemental analysis and found to be 1.12 mmol/g and 1.08 mmol/g, respectively. The higher degree of functionalization of MCM-41 is attributed to the larger surface area and higher number of free silanol groups, along with a more ordered structure and accessible pore distribution, as evidenced by gas physisorption and XRD. Compared to MSU-2, MCM-41 has a more ordered pore distribution that facilitates better ligand diffusion, leading to higher functionalization and improved analyte diffusion toward the active sites in MCM-41 functionalized with sulfonic groups.

3.2. Evaluation of Mesoporous Silicas as Sorbents in SPE

In the purification step by SPE, the selection of a suitable sorbent is crucial, as it directly influences the experimental results. The affinity of the sorbent toward the target analytes is a determining factor to be considered. The target TAs have basic characteristics (pKa values of 9.43 and 7.75 for atropine and scopolamine, respectively), with a ternary amine protonated at low pH [40,41]. Since the pH of media determines whether the analytes are in their ionic or molecular form, this variable plays a crucial role in strong cation-exchange SPE. In the functionalized silicas, their negatively charged sulfonic acid group has a strong cation-exchange capacity, so this sorbent interacts with the positively charged TAs via electrostatic (ionic) interactions that can be achieved by adding to the loading solvent HCl. However, other types of interactions can also occur, such as hydrophobic interactions between non-polar alkyl chains and the non-polar parts of the analytes, or hydrogen bonds between the free silanol groups of the silica or the sulfonic groups and suitable functional groups present in the analytes, such as hydroxyl or carbonyl groups. These multiple interactions enhance the efficiency of the sorbent material in capturing the target TAs. Figure 9 shows the types of interactions described that can occur between functionalized mesoporous silicas and target TAs.

Therefore, to evaluate the recovery efficiencies (%) of the functionalized mesoporous silicas, cartridges were packed with 50 mg of MCM-41-SO3H or MSU-2-SO3H and the following protocol was followed: (1) conditioning the cartridge with 3 mL of H2O; (2) equilibration with 3 mL of the acidified loading solvent; (3) loading with 10 mL of a 0.1 µg/L acidified solution (pH 1) of the target analytes in EtOH, MeOH, H2O or MeOH:H2O (50:50, v/v); and (4) elution with 3 mL of 5% ammonia solution in MeOH. Figure 10 shows the atropine and scopolamine recoveries obtained for the two different sorbent materials and four loading solvents. As can be seen, in general, the best recoveries were obtained with acidified H₂O as loading solvent, with recoveries >85% for both analytes and both materials, followed by EtOH, (recoveries >70%), while with MeOH and the MeOH:H₂O (50:50, v/v) the recoveries were generally below 50%. The affinity of analytes such as scopolamine and atropine for mesoporous materials (MCM-41 and MSU-2) functionalized with sulfonic groups in different media can be explained by several reasons related to the properties of the solvent and the interaction of the analytes with the functional groups of the material (sulfonic groups). Water is a highly polar solvent that can form an extensive network of hydrogen bonds. This ability to form hydrogen bonds facilitates the interaction between the sulfonic groups of the mesoporous material and the analytes, thus increasing the affinity in an aqueous medium at pH 1 [42]. In contrast, although MeOH and EtOH are also polar solvents, their ability to form hydrogen bonds is lower and their structure is less extensive. This can reduce the interaction with the sulfonic groups of the mesoporous material, resulting in a lower affinity compared to water. Additionally, at pH 1, both in water and in alcohols, analytes such as scopolamine and atropine will be predominantly in their protonated form. However, the ionization of these analytes may be more favored in an aqueous medium due to the higher dielectric constant of water (79.99), significantly higher than that of MeOH (33.30) and EtOH (24.55), which allows for better stabilization of charged species [43,44].

This greater stabilization in water can result in a higher affinity of the analytes for the sulfonic groups of the mesoporous material. On the other hand, although both EtOH and MeOH are polar solvents, EtOH has a larger molecular structure and can form stronger and more extensive hydrogen bonds and Van der Waals interactions than MeOH [45]. This can facilitate better solvation of the sulfonic groups and the analytes, improving the affinity of the analytes for the material in EtOH compared to MeOH. On the other hand, the differences in performance for these two materials can be explained by considering different factors like the surface characteristics of the sorbents and the number of functional groups present in them. On one hand, MCM-41-SO₃H presents a higher degree of functionalization (1.12 mmol ligand/g) than MSU-2-SO₃H (1.08 mmol ligand/g), which allows for greater interaction with the analytes. However, this higher degree of functionalization is not particularly significant, suggesting that its better performance as sorbent for these analytes is due to its textural properties. These include a larger pore volume (0.69 cm³/g vs. 0.52 cm³/g) after functionalization, which implies less pore blockage and greater accessibility to bulky analytes like atropine and scopolamine. Additionally, MCM-41-SO₃H has a more ordered pore structure with greater long-range order, as observed in XRD data, and a more ordered pore distribution, as seen in TEM images. This allows for better and faster diffusion of the analytes throughout the entire volume of the particle, enabling them to reach the active sites of the material where analyte–ligand interactions occur more effectively.

To elute the maximum amount of the target alkaloids, it is important to use a desorption solvent that is powerful enough to break the analyte–sorbent ionic interactions while also minimizing the volume of solvent required. Based on our previous works [29] a 5% methanolic ammonia solution was used for this task and the effect of elution solvent volume was evaluated (Figure 11). As can be seen, the increase in the elution volume from 3 to 6 mL of the 5% methanolic ammonia solution did not improve the recovery percentage for atropine. On the other hand, in the case of scopolamine, the increase in the elution volume resulted in a significant improvement with both sorbents when the loading solvent was EtOH (reaching over 95% and 80% for MCM-41-SO₃H and MSU-2-SO₃H, respectively), and there were no significant differences when the loading solvent was H₂O. These results highlight that, in general, the poorer recoveries observed when using MSU-2-SO₃H as a sorbent compared with MCM-41-SO₃H are due to lower retention of the analytes in this material and not to a lack of elution, especially when the loading was performed in EtOH.

Next, the breakthrough volume was determined. The breakthrough volume in SPE refers to the volume of solvent that can pass through the sorbent before the target analytes start to elute. So, when the breakthrough volume is reached, the analyte begins to appear in the eluate, indicating that the sorbent can no longer retain it effectively. To investigate this, increasing volumes (5, 10, and 20 mL) of 0.1 µg/L atropine and scopolamine solutions in acidified (pH 1) EtOH and H₂O were passed through the SPE cartridges containing 50 mg of MCM-41-SO₃H or MSU-2-SO₃H as sorbents. As seen in Figure 12, when the loading solvent was H₂O, no significant differences were observed in the recoveries of both analytes with both materials. However, when the loading solvent was EtOH, there was a significant reduction in retention capacity when increasing the loading volume from 10 to 20 mL. This effect was more pronounced when the sorbent was MSU-2-SO₃H (recoveries around 25%) compared to MCM-41-SO₃H (recoveries around 60%). These results can be explained by polar interactions, especially hydrogen-bond interactions of EtOH tend to reduce retention by the sorbents and result in smaller breakthrough volumes. Therefore, if the solvent most suitable for the prior solid–liquid extraction of analytes is EtOH, the low breakthrough volume observed with this loading solvent should be considered to achieve adequate recoveries, which is especially necessary for very diluted extracts.

Finally, additional experiments were carried out to evaluate the adsorption capacity of these materials, maintaining the optimal loading volume of 10 mL. These studies were carried out in a concentration range of 0.1–30 µg/L and 50 mg of MCM-41-SO₃H and MSU-2-SO₃H as a sorbent (Figure 13). As can be observed, the adsorption capacity of MCM-41-SO₃H was superior to that of MSU-2-SO₃H in both solvents. For MSU-2-SO₃H, the increase in analyte concentration resulted in a significant reduction in retention for both analytes when the solvent was EtOH, with recoveries as low as 10% for a concentration of 10 µg/L. Similarly, for MSU-2-SO₃H, the increase in analyte concentration from 0.1 to 10 µg/L led to a significant reduction in recovery, with recoveries of approximately 80% for 10 µg/L. However, when the loading was performed in H₂O, the increase in concentration up to 30 µg/L resulted in a significant reduction in recovery when the sorbent was MSU-2-SO₃H, which was more evident for scopolamine (recoveries around 45%). In contrast, for MSU-2-SO₃H, as shown in Figure 13, this same concentration allowed for recoveries close to 100% for both analytes. These results demonstrated the effectiveness of MCM-41-SO₃H as a sorbent material for SPE of tropane alkaloids in aqueous samples or extracts.

4. Conclusions

In this study, two types of mesoporous silicas with sulfonic acid groups were synthesized and functionalized: MCM-41-SO₃H, with hexagonal honeycomb structure, and MSU-2-SO₃H, with a three-dimensional interconnected pore structure. These silicas were characterized and their performance as strong cation-exchange sorbents was evaluated in the solid phase extraction of atropine and scopolamine, using ultra-high performance liquid chromatography coupled to triple quadrupole tandem mass spectrometry. The results indicated that the MCM-41-SO₃H material presented the highest recovery efficiency, around 100% for both analytes when acidified water was used as the loading solvent. This was attributed mainly to the accessibility of the sulfonic groups, favored by the regularity and uniformity of their pores. These findings highlight MCM-41-SO₃H as a promising sorbent for solid-phase extraction of tropane alkaloids in aqueous solutions, with potential applications in the control of natural toxins in food. Furthermore, this work underlines the importance of the pore structure of the mesoporous silicas in the optimization of extraction processes, opening new perspectives for the development of more efficient and sustainable miniaturized sample preparation protocols, in line with the principles of Green Analytical Chemistry.