COF-SiO2@Fe3O4 Core-Shell Composite for Magnetic Solid-Phase Extraction of Pyrethroid Pesticides in Vegetables

1. Introduction

Pyrethroids pesticides (PYRs) were a class of synthetic insecticides in the 1970s and possessed low cost, wide insecticidal, low toxicity, low residue and environmental friendliness. They played an important role in pest control and were widely used in the production of vegetables, fruits and other agricultural products. [1]. However, growing evidence indicates that PYRs may be endocrine disruptors, which can impair the endocrine function of animals and have estrogenic effects on the environment [2]. Toxic substances can kill embryos prior to and after implantation, or malformations of various organs [3]. Long-term exposure to pyrethroids and their metabolites may lead to endocrine-disrupting effects and sublethal toxicity. With the increasing awareness of food safety, some organizations have established the maximum residue limits (MRLs) of PYRs in fruits and vegetables, such as 0.01–0.5 mg·kg^-1 in the European Union [4] and 0.01–10 mg·kg^-1 in China (GB 2763-2021). Therefore, it is of great significance to establish a simple, rapid and accurate detection method for PYRs.

The matrix of vegetable samples contains large amounts of pigments, cellulose and minerals, which could dramatically impede the detection of trace PYRs in food. Therefore, efficient enrichment and purification of multiple PYRs in vegetable samples is crucial before instrumental analysis. Recently, the sample preparation method for pesticide residue analysis in vegetable mainly involves liquid-liquid extraction (LLE) [5], QuEChERS methods [6,7] and solid-phase extraction (SPE) [8]. SPE has obvious advantages in enrichment of analytes, purification of matrix, and low organic solvent consumption [9]. Compared to LLE and QuEChERS, SPE is more suitable for sample pretreatment of trace components in complex matrix samples. However, the SPE procedure is usually expensive, time consuming and tedious.

Magnetic solid-phase extraction (MSPE) has the advantages of easy separation, convenient operation and time-saving qualities[10,11]. In the MSPE process, the magnetic sorbents are directly dispersed in the sample solution for rapid and efficient extraction of analytes, and then quickly separated by an external magnetic field. MSPE effectively compensates for the shortcomings of SPE. At present, MSPE is widely used in the field of environmental governance [12,13], biotechnology [14,15], medicine [16] and food [17,18]. MSPE technology mainly relies on magnetic Fe₃O₄ nanoparticles and their surface adsorbent, which significantly affects the selectivity and efficiency of a MSPE approach. Recently, novel porous materials such as molecularly imprinted polymers (MIPs) [19,20], porous carbon [21,22], microporous organic networks (MONs) [23,24] and metal organic frameworks (MOFs) [25,26] have been employed as adsorbents for separation and enrichment. However, they still have drawbacks such as low adsorption capacity, poor selectivity and weak temperature tolerance, which limit their applications to some extent. Therefore, it is urgently needed to explore the adsorbents with excellent extraction performanc to solve these problems.

Covalent organic framework (COF) is a new type of porous polymer material, which can be constructed with organic building units by covalent bonds of elements (C, O, N, H, etc.) [27]. The structure and surface properties are mainly dependent on covalently linked topological schemes and organic monomers. It has excellent characteristics such as adjustable pore size and high chemical stability, making it have superior application potential for a variety of applications, such as catalysis [28,29], sensing [30], optoelectronic devices [31] and separation [32,33]. In this paper, the target analyte PYRs contains benzene rings and heteroatoms. Benzidine and 3,3,5,5-tetraaldehyde biphenyl was selected as ligands to synthesize the COF material with π- electron system and enriched hydroxyl groups in the structure. And the strong π-π stacking and hydrophobic interactions of the COF and PYRs leads to better adsorption properties. However, the inconvenient separation of COFs from solution due to COFs low density is a challenge and limits their application in the field of separation and enrichment. Fortunately, the combination of COFs and Fe₃O₄ provides a pathway to solve the above problem and reduce large mass loss of COFs.

During this study, a new core-shell material, COF-SiO₂@Fe₃O₄, was successfully synthesized under solvothermal conditions. Using it as an extractant, combined with GC-MS, a sample pretreatment method for MSPE of trace amounts of PYRs in vegetables was established. The synthesized COF-SiO₂@Fe₃O₄ has the advantages of good thermal stability, fast separation ability, and good selectivity for PYRs. At the same time, COF-SiO₂@Fe₃O₄ can be collected quickly through magnets, which is environmentally friendly. The developed MSPE-GC-MS method utilizing COF-SiO₂@Fe₃O₄ composites was applied for the enrichment and determination of PYRs in vegetables. A schematic diagram of the synthesis of COF-SiO₂@Fe₃O₄ and determination of PYRs by MSPE is shown in Scheme 1.

2. Results and Discussion

2.1. Characterization of COF-SiO₂@Fe₃O₄

The surface morphology of the prepared materials was characterized by SEM and TEM. The SEM images of Fe₃O₄, COF-SiO₂@Fe₃O₄ and the TEM images of COF-SiO₂@Fe₃O₄ were shown in Figure 1. It can be seen that the Fe₃O₄ materials presents a spherical structure with rough surface (Figure 1a). After the modification of a COF shell, a relatively smooth surface was observed in the SEM image of COF-SiO₂@Fe₃O₄ (Figure 1b). Through TEM combined with mapping analysis, COF-SiO₂@Fe₃O₄ contains five elements of Fe, C, O, Si and N, which proves the successful synthesis of COF-SiO₂@Fe₃O₄. Among them, the contents of Fe, C and O elements were mostly 25%, 37% and 35%, respectively. The C element was mainly derived from the synthesis of covalent organic frameworks, and the N element was mainly derived from the ligand benzidine. The core-shell structure of COF-SiO₂@Fe₃O₄ can be clearly seen by mapping.

The XRD patterns of Fe₃O₄, SiO₂@Fe₃O₄, NH₂-SiO₂@Fe₃O₄ and COF-SiO₂@Fe₃O₄ materials were shown in Figure 2. In Figure 2a, characteristic diffraction peaks could be observed at 2θ=30.1° (200), 35. 5° (311), 43.5° (400), 53.7° (422), 57.2° (511) and 62.5° (440), which were all attributed to the magnetic center body, indicating that the synthesized material has a good crystal structure [34]. In the XRD pattern of SiO₂@Fe₃O₄, there were no other diffraction peaks emerged, indicating that the coated SiO₂ shell was amorphous. For COF-SiO₂@Fe₃O₄, the newly appeared peaks located at 2θ=16.3° was speculated to be related to the encapsulation of COF. Compared with Fe₃O₄, the characteristic peaks of SiO₂@Fe₃O₄, NH₂-SiO₂@Fe₃O₄ and COF-SiO₂@Fe₃O₄ were not significantly different, indicating that the gradual reaction of Fe₃O₄ to COF-SiO₂@Fe₃O₄ does not cause changes in the crystal phase.

The infrared spectra of Fe₃O₄, SiO₂@Fe₃O₄, NH₂-SiO₂@Fe₃O₄ and COF-SiO₂@Fe₃O₄ (4000–500 cm^-1) were shown in Figure 2b. The characteristic peak at 3440 cm^-1 was the stretching vibration peak of -OH group. The typical band at 577 cm^-1 was assigned to the Fe-O-Fe vibration, which was the evidence for the existence of Fe₃O₄. The characteristic band at 1640 cm^-1 indicated the presence of carboxyl groups (curve a) [35]. The peak at 1080 cm^-1 was the tensile vibration of the Si-O group (curve b), indicating that SiO₂ had been successfully loaded onto the surface of the particles. After the amination modification of Fe₃O₄@SiO₂, a new peak appears at 1560 cm^-1 (curve c), which was -NH₂ on the surface of SiO₂ nanoparticles. The peak was disappeared after the formation of COF-SiO₂@Fe₃O₄, but new peaks appeared at 1250 cm^-1 and 1480 cm^-1, corresponding to C=N and aromatic C=C groups, respectively (curve d), indicating that COF was successfully attached to the surface of magnetic nanoparticles. The above results showed that COF-SiO₂@Fe₃O₄ was successfully synthesized by covalent bonding between monomers.

The saturation magnetization values of the Fe₃O₄, SiO₂@Fe₃O₄, NH₂-SiO₂@Fe₃O₄ and the COF-SiO₂@Fe₃O₄ nanocomposites were measured to be 71.85, 68.67, 67.97 and 59.59 emu⋅g^-1, respectively. The hysteresis curves of all magnetic nanoparticles were S-type, which indicated their superparamagnetic characteristics (Figure 2c). Among them, the magnetization of COF-SiO₂@Fe₃O₄ was the lowest, which was 12.26 emu⋅g^-1 lower than that of Fe₃O₄. This was due to the decrease of magnetism caused by the COF wrapped on Fe₃O₄. Such high saturation magnetism of the COF-SiO₂@Fe₃O₄ nanocomposites was sufficient to achieve the demand for magnetic separation. As displayed in the inset of Figure 2C, the COF-SiO₂@Fe₃O₄ nanocomposites homogeneously dispersed in aqueous solution could be rapidly gathered in 0.5 min together with the assistance of an external magnet, and thus the solution became clear and transparent immediately.

The mass ratios of different components and the thermal stability of NH₂-SiO₂@Fe₃O₄ and COF-SiO₂@Fe₃O₄ nanocomposites were examined by TGA. As presented in Figure 2d. The temperature detection range was between 30°C and 800°C, and the heating rate was 10°C⋅min^-1. The NH₂-SiO₂@Fe₃O₄ and COF-SiO₂@Fe₃O₄ showed 8.19 wt% and 17.05 wt% loss in the temperature range of 30℃–800℃, respectively. The NH₂-SiO₂@Fe₃O₄ showed 1.26 wt% loss below 140 °C, which was attributed to the weight loss of the volatilization of water adsorbed on the COF-SiO₂@Fe₃O₄ structure [36]. The mass loss between 140°C and 640°C is 6.26 wt%, which is the loss of amino groups. The COF-SiO₂@Fe₃O₄ showed 0.92 wt% loss below 200°C, which was attributed to the weight loss of the absorbed water. The mass loss between 200 °C and 800 °C was 14.83 wt%, mainly due to the loss of surface COF shell.

2.2. Optimization of MSPE Parameters

Effect of Adsorbent Amount

The adsorbent amount plays an important role in the MSPE process and was optimized with the COF-SiO₂@Fe₃O₄ adsorbents ranging from 5 to 25 mg. As shown in Figure 3a, the recoveries presented a significantly enhanced when the employed adsorbent amount was changed from 5 to 10 mg, and then the increase of adsorbent amount did not enhance the extraction efficiency of the five PYRs. Therefore, in order to improve the recoveries and save the amount of adsorbent, 10 mg of COF-SiO₂@Fe₃O₄ was optimal and used as the adsorbent in subsequent experiments.

Effect of Extraction Time

The extraction time usually has some effects on the extraction efficiency of MSPE. Insufficient extraction time will lead to inadequate adsorption, and while too long extraction time was not necessary and might bring about some losses. The effect of extraction time was explored under vigorous oscillation for 5, 10, 15, 20 min, and the results were shown in Figure 3b. When the extraction time was 10 min, the recoveries were the highest, indicating that the adsorption between the adsorbent and the target component reached dynamic equilibrium. The recoveries decreased slightly when the extraction time was prolonged. This was because the COF-SiO₂@Fe₃O₄ adsorption site was completely occupied, and the saturated adsorption capacity was obtained at 10 min. After that, a small amount of analyte broke away from the adsorbent surface and re-entered the sample solution because the adsorption was not firm. Therefore, 10 min was sufficient for the extraction process and chosen as the preferred extraction time.

Effect of Elution Solvent

The selection of suitable elution solvent is very important for the efficient desorption of target analytes from the extracted adsorbents. Here, acetonitrile, methanol and acetone were respectively used as the elution solvent to evaluate the elution performances. As shown in Figure 3c, acetone was employed as the elution solvent significantly improve the desorption efficiency of the target compounds. The recoveries of allethrin were low, when methanol was used as eluent solvent. The recoveries of five PYRs were generally low when acetonitrile was used as eluent solvent. Therefore, acetone was selected as the best eluent solvent in the subsequent experiments.

Effect of Elution Time

The effect of elution time was explored under ultrasound of 1, 3, 5, 7 min, and the results were shown in Figure 3d. When increasing from 1 min to 3 min, the recoveries increased by 80–109%. When the elution time increased from 3 to 5 min, the recoveries of PYRs remained basically unchanged, because the adsorption and desorption equilibrium was reached at 3 min. When the ultrasonic time increased from 5 to 7 min, the recoveries of PYRs increased slightly. This may be due to ultrasonic heat release, which makes the organic solvent volatilize, resulting in an increase in PYRs concentration in acetone. Therefore, 3 min was selected as the best elution time.

2.3. Adsorption Mechanism of PYRs

The adsorption behavior of adsorbents to target analytes is affected by their structures. According to the structures of COF-SiO₂@Fe₃O₄ and PYRs, it can be inferred that there are two main factors affecting the adsorption process. On the one hand, the surface functional groups (–NH₂, –OH, C=N) of COF-SiO₂@Fe₃O₄ and PYRs have strong hydrogen bond interactions [37]. On the other hand, from the structural formula of PYRs, PYRs possess abundant benzene rings and have strong π-π stacking interactions with the synthesized COF-SiO₂@Fe₃O₄ with π-π conjugation [38].

2.4. Method Validation

The quantitative analysis of the five PYRs were further evaluated by COF-SiO₂@Fe₃O₄ based MSPE coupled with GC-MS. With the optimized conditions, method validations were also studied here, including linearity, limits of detection (LODs, S/N=3), limits of quantification (LOQs, S/N = 10), enrichment factors (EFs) and reproducibility. The results were summarized in Table 1. The good linearity of the developed method was obtained with correlation coefficients (r) higher than 0.9990 in the range of 5–100 μg·kg^-1 (1.00–100 μg·kg^-1 for bifenthrin and 2.5–100 μg·kg^-1 for fenpropathrin, respectively). The LODs for the five PYRs were calculated to be 0.3–1.5 μg·kg^-1. Their corresponding LOQs were found to be 0.9–4.5 μg·kg^-1. The EFs of PYRs, defined as the ratio of the concentration of the analytes in the extract to that in the original sample, were ranged from 4.4–12.4. The inter-day RSDs were obtained by extracting standard solution five times within a day, and the intra-day RSDs were determined by extracting standard solution that had been independently prepared for contiguous six days. The inter- and intra-day RSDs were in the range of 1.9–6.2% and 2.3–7.0%, respectively, indicating the acceptable reproducibility. In addition, the reproducibility of the COF-SiO₂@Fe₃O₄ nanocomposites was assessed by the batch-to-batch RSDs. The result showed that the batch-to-batch RSDs were less than 4.2%, implying the good synthetic reproducibility of the COF-SiO₂@Fe₃O₄.

2.5. Real Sample Analysis

In order to evaluate the reliability and feasibility of the developed method in practical application, three vegetables cucumber, cabbage and lettuce were collected for the extraction and determination of PYRs. Each sample was subjected to five repeated analyses. The results showed that any of the PYRs were not detected in the vegetables. In order to further verify the method developed in the actual sample, the recoveries were analyzed by analyzing vegetable samples mixed with different concentrations of PYRs. Therefore, the three vegetables were spiked with PYRs standards at three concentrations of 5, 10, 20 μg⋅kg^-1 for low, medium and high level, and the extraction procedure was conducted, with RSDs (n=3) and average recoveries given in Table 2. The recoveries of five PYRs were 80.2–116.5% with RSDs of 2.3–6.7% for cucumber, 81.7–114.7% with RSDs of 2.1–6.8% for cabbage and 81.6–116.7% with RSDs of 2.4–7.0% for lettuce. The results indicated that the proposed MPSE-GC-MS method could be used for the enrichment and determination of PYRs in complex samples.

2.6. Comparison of COF-SiO₂@Fe₃O₄, NH₂-SiO₂@Fe₃O₄ and Fe₃O₄ as MSPE Adsorbents

Fe₃O₄ (10 mg), NH₂-SiO₂@Fe₃O₄ (10 mg), COF-SiO₂@Fe₃O₄ (10 mg) were used as adsorbents for MSPE of blank Chinese cabbage extract (50 μg·kg^-1) with standard addition of PYRs, and GC-MS detection was performed. The results were shown in Figure 4. The enrichment effects of NH₂-SiO₂@Fe₃O₄ and COF-SiO₂@Fe₃O₄ on PYRs were better than that of Fe₃O₄. Using COF-SiO₂@Fe₃O₄ as the adsorbent of MSPE, the impurity peaks of the chromatographic curve are less, especially the impurities that interfere with the target allethrin are eliminated, and the accuracy of the method is improved. From the analysis of structural differences, it can be seen that COF-SiO₂@Fe₃O₄ is a COF material with benzene ring and imine group wrapped outside the magnetic core, which can produce a strong π-π stacking effect with PYRs, thus having good selectivity for PYRs.

3. Experimental

3.1. Materials and Chemicals

Benzidine (95%) and 3,3,5,5-tetraaldehyde biphenyl (97%) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Allethrin, tetramethrin, bifenthrin, fenpropathrin, cyhalothrin were obtained from Beijing Putian Tongchuang Biological Technology Co., Ltd. (Beijing,China). FeCl₃.6H₂O and dimethyl sulfoxide (DMSO) were purchased from Tianjin Kemiou Chemical Reagent Co., Ltd. (Tianjin, China). Anhydrous sodium acetate, ethylene glycol, anhydrous ethanol, tetrahydrofuran (THF), toluene, Tetraethoxysilane (TEOS), 3-aminopropyltriethoxysilane (APTES) and dimethyl silicone oil were purchased from Tianjin Damao Chemical Reagent Factory (Tianjin, China). Chromatographic grade acetonitrile was obtained from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Glacial acetic acid was purchased from Tianjin Fuchen Chemical Reagent Factory (Tianjin, China). Ammonia water was purchased from Yantai Far East Fine Chemical Co., Ltd. (Yantai, China).

3.2. Equipment

GC/MS-QP 2010 Ultra (Shimadzu Corporation, Japan), SC-3612 centrifuge (Anhui Zhongke Zhongjia Scientific Instrument Co., LTD, China), SN-QX-20D Ultrasonic cleaning machine (Shanghai Shangdun Instrument Equipment Co., LTD, China), 2K-82B vacuum drying oven (Shanghai Instrument Experimental Factory, China), DHG-9023A blast drying oven (Shanghai Yiheng Scientific Instrument Co., LTD, China), SHB-Ⅲ circulating water multi-purpose vacuum pump (Zhengzhou Great Wall Technology & Trade Co., LTD., China), RE-52AA rotary evaporator (Shanghai Yarong biochemical instrument factory, China), YTGT-12 dry nitrogen blowing instrument (Shanghai Night extension Technology Co., LTD, China), Milli-Q Ultrapure water apparatus (Millipore Corporation, USA).

Scanning electron microscopy (SEM) images were obtained with a scanning electron microscope (GeminiSEM 300, ZEISS, Germany). Transmission electron microscopy (TEM) images were recorded by a Transmission Electron Microscope (F200x, FEI Talos, USA). X-ray diffraction (XRD) data were achieved by X-ray Difraction (3kw, Rigaku, Japan). Fourier-transform infrared spectroscopy (FT-IR) was taken on Nicolet iS20 spectrometer (Thermo Fisher, USA). The magnetization curves were measured by a vibrating sample magnetometer (VSM) (7404 LakeShore, USA). Thermogravimetric analysis (TGA) was performed using a thermogravimetric analyzer (TGA/DSC 3+, Mettler Toledo, Switzerland).

3.3. Preparation of Standard Solution

The stock standard solutions of five PYRs were individually prepared at the concentration of 5 μg·mL^-1 in acetone, and stored at 4°C before use. The working standard solutions were obtained freshly before use by diluting the stock solution to the desired concentration (0.02, 0.04, 0.06, 0.08, 0.1, 0.2, 0.4 μg·mL^-1).

3.4. Preparation of Magnetic Materials

Synthesis of Fe₃O₄ magnetic nanoparticles

Monodisperse Fe₃O₄ magnetic nanoparticles (MNPs) were synthesized by solvothermal method [39]. Briefly, FeCl₃·6H₂O (1.352 g) and anhydrous sodium acetate (3.6 g) were dissolved in ethylene glycol (40 mL). The obtained homogeneous yellow solution was transferred to autoclave, and then heated to 200 °C for 8 h. After reaction, the product was collected by magnet and washed with ultrapure water and ethanol for several times and then dried at 60 °C in a vacuum drying oven.

Synthesis of SiO₂@Fe₃O₄ Magnetic Nanospheres

Fe₃O₄ (1 g) was dispersed in a mixture of ethanol (120 mL), ultrapure water (30 mL) and concentrated ammonia (2.4 mL). After ultrasonic dispersion for 5 min, TEOS (0.5 mL) was added and stirred for 12 h at room temperature. By hydrolysis of TEOS in an alkaline environment, the surface of Fe₃O₄ was coated with a silica layer containing -OH. The obtained brown precipitates were collected by magnetic separation and washed with a ultrapure water and ethanol for three times. Finally, the resultant SiO₂@Fe₃O₄ nanocomposites were dried in vacuum at 60 °C.

Synthesis of NH₂-SiO₂@Fe₃O₄ Magnetic Nanospheres

SiO₂@Fe₃O₄ (1.00 g) was dispersed in toluene (100 mL). Then APTES (Silane coupling agent) (10 mL) was added. APTES will bond to the surface of Fe₃O₄ nanoparticles to form the ammoniated NH₂-SiO₂@Fe₃O₄. The reaction mixture refluxed at 110 °C for 8 h, and magnetically separated after natural cooling and washed with toluene and ethanol for four times. Finally, the resultant NH₂-SiO₂@Fe₃O₄ nanocomposites were dried at 60 °C for further use.

Synthesis of COF-SiO₂@Fe₃O₄ Magnetic Nanoparticles

NH₂-SiO₂@Fe₃O₄ (0.2 g), benzidine (0.185 g) and 3,3,5,5-tetraaldehyde biphenyl (0.133 g) were added into 80 mL DMSO and ultrasonically dispersed for 10 min to form a stable dispersion. Then, add 2.5 mL glacial acetic acid slowly. The mixture was transferred to autoclave, and then heated to 200 °C for 3 days. After reaction, the product was collected by magnet and washed with THF and methanol for several times and then dried at 60 °C in vacuum.

3.5. MSPE pretreatment

The prepared COF-SiO₂@Fe₃O₄ nanocomposites were used to extract PYRs from aqueous sample solution. Firstly, 20 g crushed cabbage sample was put into a centrifuge tube, 20 mL acetonitrile was added, shaken for 20 min, centrifuged at 4500 r·min^-1 for 15 min, and the supernatant was extracted. The supernatant was evaporated to about 3 mL, blown to nearly dry with nitrogen, and redissolved with ultrapure water to 100 mL. Then 10 mg COF-SiO₂@Fe₃O₄ nanocomposites were dispersed in 20 mL sample solution under vigorous oscillation for 10 min to adsorb the target analytes until adsorption equilibrium. Then, the COF-SiO₂@Fe₃O₄ nanocomposites with adsorbed PYRs were collected by using an external magnet and eluted with 1 mL acetone under 3 min of ultrasound. The desorption solution was collected, filtered through a 0.22 μm filter and injected for GC-MS analysis.

3.6. GC-MS conditions

The GC was fitted with an Rxi-5si1MS column (30 m × 0.25 mm × 0.25 µm) (Shimadzu, Japan). Helium (99.999% purity) was utilized as the carrier gas. The inlet temperature was 250℃ and the sample was injected at 1 μL without splitting (1.08 mL·min^-1). The initial oven temperature is controlled at 150℃ (hold for 0.5 min), then with a rate of 25℃·min^-1 to 180℃, and finally with a rate of 10℃·min^-1 to 250℃ (hold for 8 min). The ion source temperature, and Interface temperature were 230℃, and 250℃, respectively. The solvent delay was set to 2 min (bypassing the solvent peak). Separation of the PYRs was sufficient to set up the full scan in the range 40–500 m/z. The selected ion monitoring (SIM) mode was used. Sample analysis was performed with the electron ionization source set at 70 eV. The mass spectral parameters of 5 PYRs (50 μg·L^-1) were shown in Table 3.

3.7. Recovery

Recovery rate (ER) and enrichment factor (EF) were used to evaluate the extraction and enrichment ability of COF-SiO₂@Fe₃O₄ magnetic nanoparticles on PYRs. The calculation formula are as follows:

ER=(C_M×V_M)/(C₀×V_aq)

(1)

EF=C_M/C₀

(2)

where C₀ is the concentration of PYRs added to the sample solution before magnetic solid phase extraction, C₀=50 μg·L^-1. C_M is the concentration of PYRs in acetone elution after magnetic solid phase extraction. V_aq represents the volume of sample solution before magnetic solid phase extraction, V_aq=20 mL. V_M is the volume of acetone eluent after magnetic solid phase extraction.

Experiments were repeated three times to validate the repeatability of the results. The results presented were the average of the three experiments conducted. Origin software was used as statistical treatment to reach the conclusion described in the work.

4. Conclusions

In this research, core-shell structured magnetic nanocomposites COF-SiO₂@Fe₃O₄ were synthesized and utilized as adsorbents for the preconcentration of five PYRs via MSPE. The interaction between the benzene in COF-SiO₂@Fe₃O₄ and the benzene of analytes facilitated the effective extraction of PYRs from sample solutions. The combination of COF-SiO₂@Fe₃O₄ based MSPE with GC-MS led to the development of a fast, simple, highly efficient, and sensitive method for the determination of trace PYRs, demonstrating a high enrichment factor, wide linear range, low detection limits, and good reproducibility. Furthermore, the successful application in the selective enrichment and determination of trace PYRs in vegetables suggests that the COF-SiO₂@Fe₃O₄ nanocomposites hold great potential as a novel adsorbent in sample pretreatment.