Abstract
In this study, magnetic mesoporous silica-Fe3O4-graphene oxide nanoparticles (Fe3O4@GO@mSiO2) were synthesized and used as sorbents for magnetic solid-phase extraction (MSPE) of trace amounts of quercetin in natural samples (spinach, green pepper, dill, and red onion). The sorbent produced was characterized by Fourier transform infrared (FTIR), scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDAX), X-ray diffraction (XRD), vibrating sample magnetometry (VSM), and X-ray photoelectron spectroscopy (XPS). The effects of various experimental factors on the percent recovery of quercetin, including extraction time, desorption time, sample solution pH, and adsorbent amount were investigated. The Fe3O4@GO@mSiO2 strategy showed excellent stability and sensitivity for the determination of quercetin, with a suitable linear range of 20–800 µg L−1 and a detection limit of 5.2 µg L−1. The data indicate that Fe3O4@GO@mSiO2 has a specific surface area and suitable adsorption capacity for the determination of quercetin.
1. Introduction
Quercetin is a flavonol found in many vegetables and fruits. In general, quercetin occurs in the form of aglycones and glycosides, mainly in leaves. Absorption of this compound occurs in the large and small intestine, where it undergoes glucuronidation, sulfidation, and methylation to improve its hydrophilicity. After metabolism, which takes place mainly in the intestine, it is distributed throughout the organism and excreted through feces, urine, and exhalation of carbon dioxide. Despite its cytotoxic effects in vitro, in vivo studies on animal models confirm its safety. Due to its anti-inflammatory properties and inhibition of radicals, this compound may protect against cancer, cardiovascular disease, chronic inflammation, oxidative stress, and neurological diseases [1]. However, to discover the potential of this flavanol, it is necessary to measure its bioavailability, which is significantly limited due to its hydrophobicity. In this context, nanoscience is one of the emerging research areas that has numerous applications in the fields of industry, science, ecology, engineering, chemistry, biomedicine, and pharmaceuticals [2, 3]. There are many traditional methods for extraction and adsorption of pharmaceutical compounds and active ingredients from plants. However, due to their complexity, high operating costs, and low efficiency, these methods have some limitations [4]. Recently, the use of nanotechnology as a promising tool has overcome these problems and limitations [5]. Nanomaterials have high adsorption capacity due to their large surface area and activity sites, so they have attracted much attention [6].
One of the most significant developments in nanotechnology is graphene and its derivatives, the so-called magnetic graphene oxide- (MGO-) based materials [7]. Graphene is a single-layer, two-dimensional (2D) nanomaterial in which the carbon atoms are honeycombed together by covalent bonds to form a flat sheet structure [3]. Unlike graphene, the exact structure of graphene oxide has not yet been determined, but several models have been proposed [8]. GO consists of various oxygen-containing functional groups, including hydroxyl, carboxyl, and epoxy groups [9], which are used as a suitable Candida adsorbent in the extraction and separation of organic and inorganic materials [10]. GO nanosheets adsorb insoluble molecules by noncovalent bonding via π-π interactions, but interfacial interaction with the matrix and insolubility are among their disadvantages [11]. It should be noted that the dispersion of GO nanosheets after adsorption is very high and their separation from solution is a very time-consuming and complicated process. To solve this problem, magnetic graphene nanocomposites are usually used [12]. The use of Fe3O4 nanoparticles on graphene results in a composite with magnetic properties that has numerous applications in biomedicine, magnetic fluids, drug delivery, nanomedicine, bioengineering, biotechnology, biosensors, cell imaging, energy storage and antifungal activity, catalysis, and material extraction [2, 11]. MGO has very favorable chemical and physical properties, including easy functionalization, high chemical and physical stability, oxidation specific surface area, size coordination, and surface binding active site [2].
Nowadays, magnetic Fe3O4 nanoparticles are widely used in the preparation of nanocomposite adsorbents, mainly due to their easy synthesis and numerous applications [6]. Meanwhile, pure Fe3O4 nanoparticles are easily oxidized in air and do not show high adsorption capacity in complex environments [13]. The vertical coating of mesoporous silica on graphene nanosheets can improve the surface properties of graphene [11].
In recent years, mesoporous silica nanoparticles (MSNs) have been developed and are increasingly used due to their properties such as mesoporous nature, easy functionalization, biocompatibility, nontoxicity, adjustable porosity, and stability [14]. The Stober method is the most widely used and suitable method to prepare micrometre-sized silica particles with spherical and monodisperse morphology, and this method pioneered the development of a chemical reaction for MSN [15]. These methods are known for acidic, basic, and neutral conditions, and by simply changing the reaction parameters, this synthesis approach leads to the formation of MSN with different shapes and sizes [16].
In the present work, an attempt was made to prepare an adsorbent of Fe3O4@GO@mSiO2 with graphene oxide properties as a nanosheet and porous mesoporous silica with magnetic properties by the Hammer method for the extraction of quercetin traces and subsequent spectrophotometry. These Fe3O4@GO@mSiO2 nanocomposites have large pores and voids that can significantly increase the surface area. The superparamagnetic properties of Fe3O4 contribute to the rapid separation of the adsorbent from the matrix solution. Based on these considerations, the preconcentration and determination of quercetin in the various real samples can be easily achieved.
2. Experimental Setup
2.1. Materials
Quercetin powder was supplied by Sigma-Aldrich Chemical Company (Missouri, United States). A stock solution of quercetin (100 μg mL−1) in methanol was prepared and stored at 4°C before use. Methanol (MeOH), sulfuric acid, hydrochloric acid, sodium hydroxide, ethanol (EtOH), potassium permanganate (KMnO4), hydrogen peroxide (H2O2 30%), ferric chloride hexahydrate (FeCl3.6H2O), ferric chloride hexahydrate (FeCl2.4H2O), cetyltrimethylammonium bromide (CTAB), ammonium nitrate, natural graphite powder, and tetraethyl orthosilicate (TEOS) were of analytical grade and purchased from Merck (Darmstadt, Germany). All reagents and solvents were used in this work as received without further purification.
2.2. Instrumentation
The ultraviolet-visible (UV-Vis) spectrophotometer, model SPECTROD 250 Analytic Jena from the USA, was used to record the absorption spectra. Fourier transform infrared (FTIR) analyzes were recorded using a Bruker Vector 22 spectrometer (Germany). SEM images and energy dispersive X-ray spectroscopy (EDAX) were performed using a field emission-scanning electron microscopy (FE-SEM) from Mighty-8 Instrument, TSCAN Company, Prague. The synthesized materials were also characterized by X-ray diffraction (XRD) and vibrational sample magnetometry (VSM).
2.3. Preparation of Graphene Oxide (GO)
GO was synthesized by the modified Hummer method. Briefly, 1 g of graphite and 23 mL of concentrated H2SO4 were added to a 250 mL conical flask with stirring in an ice bath. Then, 3 g of KMnO4 was added to the mixture. After a while, the ice bath was removed, and the resulting dark green solution was stirred at 35°C for one hour. Then, 46 mL of water was slowly added to the reaction mixture and placed in boiling water. The reaction mixture was kept at this temperature for another 30 minutes. Then, 10 mL of 30% H2O2 and 140 mL of distilled water were added to stop the reaction. The obtained precipitate was centrifuged and washed several times with 5% HCl and ethanol and dried in vacuo at 60°C.
2.4. Preparation of Fe3O4@GO Nanoparticles
The method for the synthesis of Fe3O4@GO nanoparticles was described by Wang et al. In summary, 0.1 g GO was kept in 150 ml deionized water containing 1.6 g FeCl3.6H2O and 0.6 g FeCl2.4H2O for 30 min at room temperature with stirring. Then, it was added dropwise to a 250 mL double-neck flask containing 25 mL of 30% ammonia solution with vigorous stirring. The solution was kept under nitrogen atmosphere at 25°C for 30 minutes and then heated to 80°C for one hour. The product was magnetically separated and washed with deionized water and ethanol, respectively. The Fe3O4@GO nanoparticles were dried in vacuo at 50°C.
2.5. Preparation of Fe3O4@GO@mSiO2
The synthesis of Fe3O4@GO@mSiO2 nanoparticles was described by Yin. Briefly, 50 mg Fe3O4@GO nanoparticles were ultrasonically dispersed in 50 mL deionized water containing 500 mg CTAB for 30 minutes. Then, 400 mL of deionized water and 50 mL of NaOH (0.01 M) were added and sonicated for 10 minutes. The mixture was heated at 60°C for 30 minutes. Then, 2.5 mL of TEOS/ethanol (v/v: 1/4) solution was added. The mixture was stirred at 60°C for 12 hours. The final product was magnetically separated and washed with deionized water. To remove CTAB, these nanoparticles were then mixed with 15 ml of methanol containing 120 mg of ammonium nitrate and stirred at 60°C for 12 hours. Finally, after 12 hours, they were separated with a magnet and washed several times with ethanol. The Fe3O4@GO@mSiO2 nanoparticles were dried in vacuo at 70°C.
2.6. Extraction Procedure
Adsorption was performed using 5 mL of quercetin sample solution at a concentration of 500 ng mL−1 (pH 8) and 2 mg Fe3O4@GO@mSiO2 as sorbent in a 10 mL glass vial for 45 minutes with magnetic stirring of 500 rpm at room temperature. Then, the solution was filtered with a magnet and washed several times with distilled water. Desorption was carried out with 1 mL of methanol to 2 mg Fe3O4@GO@mSiO2 for 30 min with a magnetic stirrer at 500 rpm. Finally, the solution was separated from the nanoparticles with a magnet and adsorption was measured with a spectrophotometer at 372 nm (Scheme 1).
2.7. Real Sample Preparation
The MSPE method was tested on real samples by analyzing samples of spinach, dill, green bell pepper, and red onion. These food samples were prepared for the solid-phase magnetic extraction method. All food samples were first washed with tap water and then with distilled water. Then, the samples were cut into cubes and exactly 5 g were weighed into 50 mL Falcon tubes. Then, 30 mL of methanol was added to each sample and stirred on a shaker for 24 hours. The samples were then centrifuged for 15 min, and the solid phase was separated from the liquids using an aqueous filter paper. The solvents were evaporated using a rotary apparatus, and the obtained extract was transferred to a 25 mL Falcon tube and diluted with distilled water after adjusting the pH [17].
3. Results and Discussion
The properties and characteristics of magnetic oxide-graphene nanoparticles coated with porous silica nanoparticles have been investigated using various techniques.
3.1. Characterization of Fe3O4@GO@mSiO2 Nanoparticles
The chemical structure of Fe3O4@GO@mSiO2 was confirmed by FTIR spectroscopy, as shown in Figure 1. The FTIR spectra of Fe3O4@GO@mSiO2 showed the stretching vibrations of the O-H and C-OH groups on the graphene oxide sheets at 3444 cm−1 and 1238 cm−1, respectively. Also, the peak observed at 1731 cm−1 is related to the stretching vibration of the C=O bond, confirming the presence of carboxyl and carbonyl groups in the graphene oxide structure. The aliphatic C-H peak in the region of 2925 cm−1 is also visible in this spectrum. These peaks indicate the presence of epoxide, carboxyl, and hydroxyl groups in the structure of graphene oxide. Comparison of the spectra of graphene oxide and magnetic graphene shows a strong decrease in the peak of oxygen-containing functional groups in magnetic graphene compared to graphene oxide. The absorption peak at 576 cm−1 is related to the stretching vibration of Fe-O during the Fe3O4 preparation step. This peak confirms that Fe3O4 nanoparticles were successfully combined with graphene nanoparticles. Fe3O4@GO@mSiO2 showed three new peaks at 1047 cm−1, 761 cm−1, and 568 cm−1 compared to Fe3O4@GO, which relate to symmetric and asymmetric stretching vibrations and bending vibrations of Si-O-Si. The presence of these three peaks indicates the binding of mSiO2 to the surface of Fe3O4.
The surface morphology of GO, Fe3O4@GO@mSiO2 was studied by scanning electron microscopy (SEM). As shown in Figure 2(a), graphene oxide exhibits laminated sheets with many wrinkles, and after coating with Fe3O4 and mesoporous silica nanoparticles, the surfaces were completely roughened (Figure 2(b)).
(a)
(b)
(c)
(d)
The EDAX spectrum of Fe3O4@GO@mSiO2 was recorded to determine the elemental composition. Figure 2(c) shows the energy dispersive X-ray spectroscopy (EDAX) spectrum of the sorbent, and the presence of the elements C, O, Fe, and Si can be found to be 33.8%, 8.26, 2.20%, and 81.21% (w/w), respectively. These results confirm the successful synthesis of Fe3O4@GO@mSiO2 as sorbent.
XRD analysis was used to evaluate the crystallinity of the prepared samples. In Fe3O4@GO, the diffraction peaks at 2θ angles are 52/30, 84/35, 46/43, 99/53, 58/57, 02/63, and 60/74 degrees, corresponding to crystal plates (220), (311), (400), (422), (511), (440), and (533), respectively. For Fe3O4@GO@mSiO2, the diffraction peaks at angles 2θ are about 52/18, 32/30, 71/35, 49/43, 93/53, 38/57, 94/62, and 54/74 degrees, and the crystal plates are (111), (220), (311), (400), (422), (511), (440), and (533). These peaks are consistent with the results reported in other research articles. No diffraction peaks of graphene oxide are seen in the diffraction patterns of Figure 2(d), and only specific peaks related to the crystal of magnetic iron oxide nanoparticles are visible. This may be due to the tiny amount of graphene oxide in the nanocomposite, or it may be that it peels and exfoliates between the graphene oxide layers during the synthesis process of crystal growth, and the graphene oxide diffraction peak disappears.
The magnetic behavior of the Fe3O4@GO@mSiO2 nanosorbent was determined using a vibrating sample magnetometer. Figure 3 shows the magnetic hysteresis loops of the samples, which are S-shaped and pass almost through the origin, indicating that the samples have near zero hysteresis magnetism. These results indicate the superparamagnetic behavior of the synthesized samples. The VSM measurements for Fe3O4 nanoparticles are (65/65 emu/g), Fe3O4@GO (emu 34/56), and Fe3O4@GO@mSiO2 (13/34 emu/g). Addition of graphene oxide and mesoporous silica to the nanoparticles reduces the saturation magnetization. The decreased saturation magnetization in nanocomposites may be related to the presence of nonmagnetic layers GO and mSiO2, the lower charge of Fe3O4 on the surface of mSiO2 and GO, the particle size of Fe3O4, and the electron exchange between Fe3O4 and GO and mSiO2.
The chemical status of the element in Fe3O4@GO and Fe3O4@GO@mSiO2 was further investigated by XPS (Figure 4). The positions of the C 1s, O 1s, and Fe 2p peaks can be seen in the spectrum of Fe3O4@GO. Compared with Fe3O4@GO, two additional peaks of Si 2s and Si 2p have also appeared in the full spectrum of Fe3O4@GO@mSiO2, which is a reason for the presence of mesoporous silica in the synthesized mesoporous silica magnetic graphene nanocomposite.
3.2. Optimization of MSPE Conditions
In order to optimize the magnetic solid-phase extraction method, various parameters such as extraction time, desorption time, adsorbent amount, and pH of the sample solution were investigated.
3.2.1. Effect of Extraction Time and Desorption Time
Optimization of extraction and desorption time is an essential and fundamental parameter affecting the sensitivity of extraction procedures. The extraction and desorption times were studied in a range from 5 to 40 min. The results in Figures 5(a) and 5(b) show that as the time increases, the efficiency of extraction and desorption of the analyte increases. As can be seen, the maximum extraction efficiency was reached after 40 min, and the maximum desorption time was 45 min. Finally, 40 min and 45 min were used as the optimum extraction and desorption times for further experiments.
(a)
(b)
(c)
(d)
3.2.2. Effect of Sorbent Amount
The effect of sorbent amount on the preconcentration of quercetin was investigated in the range of 2.0–8.0 mg. As shown in Figure 5(c), the absorbance increased with increasing amount of sorbent, while the difference between the values of 6 and 8 mg of adsorbent was not very large. Therefore, based on the results, 8 mg of adsorbent can be selected as the optimum amount.
3.2.3. Effect of Extraction Solvent Type
To evaluate the effects of the type of extraction solvent, methanol, ethanol, acetone, acetonitrile, and chloroform were used as solvents. The results showed that the highest amount of adsorption signals was associated with the use of methanol as solvent. The desired solvent must be able to dissolve the quercetin molecule and overcome the bond between quercetin and adsorbent. In addition, the extraction solvent must have the same polarity as the quercetin in order for the quercetin to be properly desorbed in the solvent. Due to the presence of hydroxyl groups, quercetin is considered a polar molecule, and considering the pKa value of quercetin (pKa: 6.9) at pH 8.0, the composition of quercetin is almost molecularly neutral. Based on the data obtained, methanol (aprotic solvent) was considered as the optimal extraction solvent for the desorption of quercetin and was used in the next experiments.
3.2.4. Effect of Extraction Solvent Volume
The effect of the volume of the extraction solvent was also evaluated. The results showed that increasing the volume of extraction solvent causes a decrease in the amount of adsorbed signal, as the concentration of quercetin decreases with the increase in the volume of extraction solvent. These results showed that the volume of 1 ml of methanol is sufficient for the desorption of quercetin and that the highest concentration of the extracted sample can be obtained with this volume. It should be noted that although a higher concentration factor would be expected with smaller volumes of solvent, this was not used due to a significant error in the separation of quercetin from the sample solution. Finally, the volume of 1 mL of solvent was considered as the optimal extraction solvent volume and was used in the next experiments.
3.2.5. Effect of Sample Solution pH
The effect of sample pH plays a significant role in SPE procedures. Therefore, the pH of aqueous sample solutions in the range of 2–11 was investigated. From the results (Figure 5(d)), the adsorption of quercetin gradually increased with increasing pH from 2 to 8 and then decreased at a higher pH. The possible adsorption mechanism between the sorbent and quercetin composition depends on the degree of protonation and deprotonation of quercetin and its interaction with the sorbent surface. Low adsorption in acidic environment may be due to the presence of abundant proton ions and protonation of quercetin hydroxyl groups with pKa = 6.9 and silanol groups on the sorbent surface and subsequent electrostatic repulsion between these groups. With a gradual increase in pH up to 8, the positive charge on the surface of the analyte and sorbent decreases, and hydrogen bonding may occur between them, ultimately leading to the highest amount of adsorption. In contrast, at a pH above 8, the adsorption decreases again, which may be due to the electrostatic repulsion in acidic environments between the negative charges of the sorbent surface and the deprotonated quercetin molecules. The highest adsorption was found at pH 8, where the strongest interaction between the sorbent and the quercetin composition occurs. Therefore, pH 8 was chosen as the optimum pH for further work.
3.2.6. Extraction Capacity
The extraction capacity of the proposed Fe3O4@GO@mSiO2 nanosorbent was calculated using equation (1) and a series of aqueous quercetin solutions ranging from 1 to 0.02 μg mL−1. As shown in Figure 6, the amount of extracted quercetin increases with increasing initial quercetin concentration up to 0.8 μg mL−1 and then remains constant. The extraction capacity of Fe3O4@GO@mSiO2 nanoparticles was calculated to be 1.89 mg g−1.
In equation (1), Q (mg g−1) is the adsorption capacity of Fe3O4@GO@mSiO2, C (μg L−1) is the extracted concentration of quercetin obtained by the calibration line equation, V (mL) is the volume of quercetin solution, and m (mg) is the weight of the Fe3O4@GO@mSiO2.
3.3. Isotherm Study
The adsorption mechanism of quercetin in Fe3O4@GO@mSiO2 was evaluated by using two isotherm models, Langmuir (equation (1) and Freundlich (equation (2)).where Ce (mg L−1) is the amount of quercetin in solution and qe (mg g−1) and qmax (mg g−1) are the equilibrium and maximum adsorption capacities of the adsorbent, respectively. kL (L mg−1) is the Langmuir constant, which represents the affinity of the binding site and the adsorption energy. Langmuir parameters such as KL and qmax are calculated from the intercept and slope of the linear plot of Ce/qe vs. Ce. KF (L mg−1) and n are Freundlich constants representing adsorption capacity and adsorption intensity, respectively, and are calculated from the linear plot of log qe versus log Ce.
The Langmuir isotherm assumes that the binding sites on the surface of sorbent are homogeneous, and that adsorption occurs in a monolayer pattern with no interaction between the adsorption sites. In contrast, the Freundlich isotherm assumes that the adsorption process is based on a heterogeneous surface and that adsorption occurs in multiple layers. Table 1 shows the parameters for the two models. On the bases of fitted curves and data, it can be seen that the R2 of the linear Langmuir (0.9959), is higher than that of Freundlich (0.9703). Thus, the Langmuir model best describes the adsorption process of quercetin and confirms that the adsorptive sites are homogeneous on Fe3O4@ GO@mSiO2.
3.4. Selectivity of Fe3O4@GO@mSiO2 Nanosorbent
The influence of foreign species on the determination of quercetin was studied under optimized conditions. For this purpose, sample solutions containing 500 ng mL−1 quercetin and different concentrations of interfering species were prepared. From the experimental results presented in Table 2, it can be seen that when the extraction amount of the analyte is changed by more than 5%, most of the cations can be retained in the pores of the adsorbent and reduce the absorption of quercetin. Organic compounds with relatively similar structures, such as gallic acid and caffeine, can also be adsorbed. It can be concluded that the adsorbent used in this project has relatively good performance for quercetin but is not very selective. To increase the selectivity, the silanol groups in the pores of the adsorbent need to be changed.
3.5. Method Validation
Under the optimized conditions, the performance of the developed method for the determination of quercetin was investigated according to the measurement procedure. The calibration curve was linear in the range of 20–800 μg L−1. The limit (LOD) was 5.2 μg L−1 and was calculated using the equation LOD = 3Sb/m, where Sb and m are the standard deviations of three blank repeats and the slope of the calibration curve, respectively. Additional performance data for this method, including recovery and preconcentration factor, are shown in Table 3. In addition, a comparison of the proposed method with the other reported preconcentration methods for the extraction of quercetin from real samples was made in Table 4. The results show that the proposed method for the determination of quercetin provides superior or comparable adsorption capacity for quercetin compared to the literature reports.
3.6. Real Sample Analysis
The method was used to determine quercetin in samples of spinach, green bell pepper, red onion, and dill weed by the standard addition method. Different concentrations of quercetin were added and analyzed by the proposed method. As shown in Table 5, recoveries ranging from 94.00 to 107.00% and RSD values ranging from 2.43 to 4.49 were obtained by the proposed method. These results indicate that the proposed method is suitable for the separation and preconcentration of quercetin from real samples.
4. Conclusions
In this study, an attempt was made to prepare an adsorbent of Fe3O4@GO@mSiO2 with graphene oxide properties as a nanosheet and porous mesoporous silica with magnetic properties by the Hammer method to extract quercetin and analyze it by a simple spectroscopic method. The prepared sorbent was characterized by FTIR, FESEM, EDX, XRD, VSM, and XPS techniques. The pH of the aqueous solution as a parameter affecting the interaction between quercetin and the synthesized adsorbent showed that a strongly acidic or alkaline environment was not suitable due to the reduction of electrostatic interaction and hydrogen bonding between the analyte and the sorbent. The results showed that the synthesized magnetic nanoparticle sorbent can adsorb and extract small amounts of quercetin from aqueous solutions under optimal conditions, nonselectively. It was successfully used to measure quercetin in real plant samples (spinach, green pepper, dill, and red onion).
Data Availability
Our research data during this study are included in the article.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
The study was supported by the University of Kurdistan-9713005203.