Ionic-Liquid-Micelle-Functionalized Mesoporous Fe3O4 Microspheres

Aug 14, 2014 - The accuracy of the proposed method was investigated by recovery in herb and granules of Radix et Rhizoma Rhei, yielding values between...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/JAFC

Ionic-Liquid-Micelle-Functionalized Mesoporous Fe3O4 Microspheres for Ultraperformance Liquid Chromatography Determination of Anthraquinones in Dietary Supplements Shuai-Shuai Hu,†,§ Wan Cao,†,§ Han-Bin Dai,†,§ Jian-Hua Da,†,§ Li-Hong Ye,‡ Jun Cao,*,† and Xing-Ying Li† †

College of Material Chemistry and Chemical Engineering, Hangzhou Normal University, Hangzhou 310036, China Integrated Chinese and Western Medicine Hospital of Zhejiang Province, Hangzhou 310003, China



ABSTRACT: A magnetic solid-phase extraction method using ionic liquid (IL)-micelle-functionalized mesoporous Fe3O4 microspheres (MFMs) was proposed for the preconcentration of anthraquinones in dietary supplements. The analytes were then determined by ultraperformance liquid chromatography combined with an ultraviolet detector. The extraction parameters, such as the choice of ILs, the concentrations of ILs and MFMs, the pH of diluent, and the concentration of acetic acid in the eluent, were presented. Under the optimized conditions, the limits of detection and limits of quantitation were 0.4−2.8 ng mL−1 and 1.4−9.4 ng mL−1, respectively. The accuracy of the proposed method was investigated by recovery in herb and granules of Radix et Rhizoma Rhei, yielding values between 89.25% and 96.48%. The use of the proposed method in the sample pretreatment of complex dietary supplements is feasible due to the high surface area and excellent adsorption capacity of MFMs after modification with IL. KEYWORDS: anthraquinones, ionic liquids, magnetic solid-phase extraction, mesoporous Fe3O4 microspheres, Radix et Rhizoma Rhei, ultraperformance liquid chromatography



INTRODUCTION Mesoporous materials contain mesopores, which have diameters between 2 and 50 nm. In recent years, mesoporous materials have attracted considerable attention because of their desirable properties, such as large pore volume, high surface area, controllable wall composition, tunable mesoporous channels with a well-defined pore-size distribution, and modifiable surface properties.1 In this field, chemically synthesized nanoparticles with mesoporous structures and magnetic characteristics have been widely applied in catalysis,2 environmental remediation,3 extraction,4 and biotechnology.5,6 However, several unavoidable problems are associated with magnetic mesoporous materials (MMMs), such as their inherent instability over long periods because of their tendency to aggregate.7 For this reason, it is necessary to coat the MMMs with a layer of novel materials, such as silica and ionic liquids (ILs), which could overcome many of these limitations and may have great potential for the extraction of complex plant samples. ILs, also known as organic salts, have melting points at or below 100 °C. ILs usually consist of an organic cation (e.g., imidazolium, pyrrolidinium, or pyridinium) and an inorganic or organic anion (e.g., tetrafluoroborate, bromide).8 Since their first discovery in 1914, ILs have been widely investigated due to their unique and desirable characteristics, such as negligible vapor pressure, tunable viscosity and miscibility with water, high thermal stability, and nonflammability.9 Recently, ILs have found potential applications in sample extraction,10,11 spectroscopy,12 density measurements,13 matrix-assisted laser desorption/ionization mass spectrometry,14 and chromatography.15 In the past several decades, IL-functionalized MMMs have been a © XXXX American Chemical Society

topic of growing interest because ILs may provide a promising and effective approach to modifying the physical properties of these MMMs and incorporating certain functionalities due to the interaction with the cationic part of the ILs.16 Experts predict that this material has great application prospects in such areas as sample separation and extraction. Magnetic solid-phase extraction (MSPE), which is based on the use of sorbents with paramagnetic features, has been successfully applied in various fields, including biological,17,18 plant ingredient,19 food,20 and environmental water analyses.21 Compared with the conventional solid-phase extraction (SPE) method, MSPE features a simpler overall procedure and presents a similar extraction efficiency; however, it reduces the amount of sorbent required. The success of MSPE has been based on two basic aspects. On one hand, the sorbent is uniformly dispersed in the sample solution, rather than being packed into a SPE cartridge, which increases the contact surface between the sample and the sorbent and thereby the extraction efficiency. On the other hand, the sorbent is easily recovered after the extraction by applying an external magnetic field, without requiring filtration or centrifugation. Furthermore, nanosized magnetic materials, such as graphene and graphenebased Fe3O4 oxide nanocomposite,22,23 have attracted much interest in MSPE due to their large surface areas and unique physical and chemical properties. However, to the best of our knowledge, the application of IL-functionalized mesoporous Received: May 16, 2014 Revised: August 13, 2014 Accepted: August 14, 2014

A

dx.doi.org/10.1021/jf502323f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

A (0.1% formic acid, v/v) and B (methanol) at a flow rate of 0.5 mL min−1 as eluents. The gradient eluent program was as follows: 0−1 min, 25%−30% B; 1−3 min, 30%−45% B; 3−5 min, 45%−60% B; 5−7 min, 60%−80% B; 7−8 min, 80%− 100% B; 8−8.5 min, 100%−100% B. An Acquity UPLC BEH C18 analytical column (1.7 μm, 2.1 mm × 50 mm) was supplied by Waters Co. During the separation, the injection volume was 1 μL for standards and sample solutions. UV detection was performed at a wavelength of 254 nm. Preparations of Standard Solution and Sample Solutions. Standard solutions of emodin (500 μg mL−1), rhein (250 μg mL−1), aloe emodin (500 μg mL−1), physcion (250 μg mL−1), and chrysophanol (500 μg mL−1) were prepared in methanol and then diluted to the appropriate concentration using pure water prior to MSPE. Rhei Radix et Rhizoma sample was prepared according to the approach described in the Chinese Pharmacopoeia 2010 without any modifications. First, a 0.15 g sample of powdered rhubarb herb was extracted with methanol (25 mL) by refluxing for 60 min. Next, the solution was cooled to room temperature, and the weight lost was compensated with methanol in the extraction process. After shaking and filtering, 5 mL of filtrate was transferred to a small flask and evaporated to dryness. The residue was redissolved in 10 mL of 8% HCl and then sonicated at 100 W (40 kHz) for 2 min. Next, 10 mL of chloroform was added in acidified solution. The solution was refluxed for 60 min and allowed to cool. The mixture was then transferred to a separating funnel and extracted three times with 10 mL of chloroform, and the combined solutions were dried to a residue under reduced pressure. The residue was dissolved with methanol and transferred to a 10 mL volumetric flask, and methanol was added to the mark. The rhubarb granules were prepared using the same method. Preparation of MFMs. Monodispersed Fe3O4 microspheres of average size were synthesized according to previous reports.25,26 For a typical synthesis of Fe3O4 microspheres, FeCl3 (0.406 g) and NaAc (1.026 g) were dissolved in 49.02 mL of ethylene glycol and 0.98 mL of water. After vigorous stirring of the mixture for 8 h, the homogeneous solution was transferred to a Teflon-lined stainless-steel autoclave (100 mL in volume), sealed, and heated to 200 °C. After a 12 h reaction period, the autoclave was cooled to room temperature. The obtained Fe3O4 microspheres were washed with ethanol and water and then dried in a vacuum for 12 h. Finally, the resulting Fe3O4 MMNPs were stored at 25 °C before being used for the following MSPE process. Material Characterization. Transmission electron microscope (TEM) (Zeiss Supra 55, Oberkochen, Germany) was used to obtain the micrographs of MFMs (100 kV), which were used to characterize their particle size and morphology. MSPE Procedure. The extraction procedure, elaborately depicted in Figure 2, was as follows. First, 20 μL of the standard/sample solutions was homogeneously mixed with 6.6 mg mL−1 MFMs and 0.32 mg mL−1 [C12mim]Cl dispersions in a 1.5 mL Eppendorf tube. Subsequently, 1000 μL of purified water was slowly added to the tube, and the mixture was vortexed for 10 s at fast speed before being allowed to stand for 5 min. A magnet was deposited outside the bottom of the centrifuge tube for 1 s to separate the Fe3O4-retrieval anthraquinone sorbents from the sample solution. Next, the supernatant was decanted, and each of the target analytes was eluted from the sorbents using 200 μL of methanol (containing 1% acetic acid). The collected eluents (eluted thrice, yielding

Fe3O4 microspheres (MFMs) sorbents for SPE has not been reported. Rhei Radix et Rhizoma, the root or rhizome of perennial herbaceous plant Rheum palmatum L., Rheum tanguticum Maxim. ex Balf., or Rheum officinale Baill., is not only consumed as a health food supplement in China or other countries, but it also represents one of the most popular medicinal herbs used in clinical practice for the treatment of blood stagnation syndrome as well as a purgative agent. Recent publications have demonstrated that anthraquinone derivatives were thought to be the major active components responsible for the pharmacological activities and therapeutic efficacy.24 In this study, the use of IL-coated MFMs sorbents combined with ultraperformance liquid chromatography (UPLC) was developed to preconcentrate and determine the anthraquinones in dietary supplements, namely, aloe emodin, rhein, emodin, chrysophanol, and physcion. Furthermore, various parameters, including IL type, concentration of ILs and MFMs, diluent pH, and concentration of acetic acid in eluent were optimized in detail. The experimental results indicated that the established method was reliable and feasible for the analysis of real rhubarb samples.



MATERIALS AND METHODS Reagents and Standards. 1-Dodecyl-3-methylimidazolium bromide ([C12mim]Br), 1-dodecyl-3-methylimidazolium chloride ([C 12 mim]Cl), 1-butyl-3-methylimidazolium tetrafluoroborate (bminBF4), and 1-butyl-3methylimidazolium polystyrenesulfonate (bminPSSA) were purchased from Shanghai Cheng Jie Chemical Co., Ltd. (Shanghai, China). Methanol (HPLC grade) was supplied by Sigma-Aldrich Shanghai Trading Co., Ltd. (Shanghai, China). Purified water was obtained from Hangzhou Wahaha Group Co., Ltd. (Hangzhou, China). All other chemicals and reagents were of analytical grade and were used without further purification. The investigated standards, including emodin, rhein, aloe emodin, and physcion, were purchased from Shanghai Winherb Medical Technology Co., Ltd. (Shanghai, China). Chrysophanol was obtained from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). The purity of each standard was higher than 98%, and their structures are shown in Figure 1. The herb and granules of Rhei Radix et Rhizoma were collected from the local drugstore (Hangzhou, China). UPLC Analysis. The separation and quantitation of anthraquinones was accomplished using UPLC with ultraviolet (UV) detection (Waters Co., Milford, MA, U.S.A.) with solvent

Figure 1. Chemical structures of the five compounds investigated. B

dx.doi.org/10.1021/jf502323f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Figure 2. Schematic of the magnetic SPE process using IL-functionalized MFMs.

600 μL) were transferred to a new Eppendorf tube and then evaporated to dryness in an 80 °C water bath. The residues were redissolved with 50 μL of methanol, sonicated at 100 W (40 kHz) for 2 min, and then centrifuged at 13 000 rpm for 5 min. Finally, 1 μL of the standard or sample solution of interest was injected into UPLC for identification and quantification of the anthraquinones.



RESULTS AND DISCUSSION Characterization of MFMs. The morphology and size of MFMs were characterized by TEM with 200 nm (A), 100 nm

Figure 4. IL selection. Extraction conditions: sample volume, 1230 μL; spiked working solution, 20 μL; IL concentration, 0.32 mg mL−1; MFMs concentration, 6.6 mg mL−1; pH, 11. Desorption conditions: 200 μL of methanol (containing 1.0% acetic acid), thrice. Types of ILs: (A) [C12mim]Br, (B) [C12mim]Cl, (C) bminBF4, (D) bminPSSA. Peak identification: 1, aloe emodin; 2, rhein; 3, emodin; 4, chrysophanol; 5, physcion.

Figure 3. TEM image of MFMs with 200 nm (A), 100 nm (B), and 50 nm (C) in scale bars.

Selection of ILs. The principal aims of this study were to investigate the unique characteristics of an adsorbent comprising MFMs and ILs in dispersion and to select a suitable type of IL for the preparation of dispersants. To this end, four types of ILs ([C12mim]Br, [C12mim]Cl, bminBF4, and bminPSSA) containing different inorganic anions were selected in this work. For all IL species investigated, a good dispersion was achieved by adding MFMs to the IL solutions. The dispersions, which were optically inhomogeneous at the microcosmic level, exhibited more intense UV absorption with increasing IL alkyl chain length. As shown in Figure 4, a significantly higher extraction efficiency was obtained by mixing with long-chain ILs ([C12mim]Br, [C12mim]Cl) in the magnetic functionalization process than when using shortaliphatic-chain ILs (bminBF4 and bminPSSA). This may due to the following. (a) The long-alkyl-chain ILs possessed surface-

(B), and 50 nm (C) in scale bars. As indicated from the TEM image (Figure 3), the sample consisted of a large quantity of tiny primary Fe3O4 nanocrystals with an average size of 100 nm. In addition, the uniform shape and hierarchical structure of mesoporous Fe3O4 were observed from Figure 3. When the nanocrystals were used as building blocks, they were assembled into a mesoporous structure. It was obvious that the TEM image confirmed the loosely packed character with lots of void space between microspheres. To achieve the best extraction efficiency of MSPE for the target analytes, several variables, including IL type, concentration of ILs and MFMs, diluent pH, and concentration of acetic acid in the eluent, were investigated in depth in the optimization process following a univariate method. C

dx.doi.org/10.1021/jf502323f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Figure 7. Influence of pH on the extraction of five compounds. Extraction conditions: sample volume, 1230 μL; spiked working solution, 20 μL; [C12mim]Cl concentration, 0.32 mg mL−1; MFMs concentration, 6.6 mg mL−1; pH, 5−13. Peak identification: 1, aloe emodin; 2, rhein; 3, emodin; 4, chrysophanol; 5, physcion. The other conditions were the same as in Figure 4.

Figure 5. Effect of IL concentration on the extraction of five compounds. Extraction conditions: sample volume, 1230 μL; spiked working solution, 20 μL; MFMs concentration, 6.6 mg mL−1; pH, 11; [C12mim]Cl concentration, 0−0.48 mg mL−1. Peak identification: 1, aloe emodin; 2, rhein; 3, emodin; 4, chrysophanol; 5, physcion. The other conditions were the same as in Figure 4.

Figure 6. Influence of MFMs concentration on the extraction of five compounds. Extraction conditions: sample volume, 1230 μL; spiked working solution, 20 μL; [C12mim]Cl concentration, 0.32 mg mL−1; pH, 11; MFMs concentration, 3.3−16.5 mg mL−1. Peak identification: 1, aloe emodin; 2, rhein; 3, emodin; 4, chrysophanol; 5, physcion. The other conditions were the same as in Figure 4.

Figure 8. Effect of acetic acid concentration on the extraction of five compounds. Extraction conditions: sample volume, 1230 μL; spiked working solution, 20 μL; [C12mim]Cl concentration, 0.32 mg mL−1; MFMs concentration, 6.6 mg mL−1; pH, 11. Desorption conditions: 200 μL of methanol (containing 0%−5% acetic acid), thrice. Peak identification: 1, aloe emodin; 2, rhein; 3, emodin; 4, chrysophanol; 5, physcion.

active characteristics similar to conventional cationic surfactants and may be amphiphilic when dissolved in water.27 Moreover, the IL with longer alkyl chains at the cation featured a higher distribution ratio in the functionalization process. (b) Imidazolium-based ILs with longer alkyl side chains strengthened the directionality of hydrogen bonds and other weak forces (e.g., van der Waals),28 which provided a greater interaction with MFMs and target analytes. In addition, the application of [C12mim]Cl-functionalized magnetic dispersion achieved the optimum UV absorption due to the stronger conjugated and inductive effect induced by the chloride anions. It also can be clearly seen that the addition of four ILs in MSPE ultimately yielded yellow liquids. Among these ILs, the deepest color was observed using [C12mim]Cl dispersion. The results indicated that [C12mim]Cl employed in functional magnetic materials indeed had a significant effect on the extraction of analytes in the types of ILs investigated. Therefore, [C12mim] Cl was selected as the optimal IL and used in further study.

Effect of IL Concentration. The IL concentration generally has a strong effect on the extraction efficiency and the site of action of target compounds. During the optimization of the method, the effect of IL concentration was evaluated at concentrations ranging from 0 to 0.48 mg mL−1. The results indicated that when the concentration of [C12mim]Cl in the dispersions increased from 0 to 0.32 mg mL−1, the detected anthraquinone peak areas greatly improved by factors of 20 to 60 (Figure 5). It was also apparent that the maximum adsorption of target analytes was obtained with 0.32 mg mL−1 of [C12mim]Cl in the dispersion. However, further increasing the concentration of ILs to 0.48 mg mL−1 resulted in serious deterioration of the extraction efficiency, although the dispersion of the sample was good for high IL concentrations, indicating that the partition of the analytes between the sorbents and aqueous phase was reduced. Additionally, the yellow color of the solutions was observed in the final step of the extraction process, especially when 0.32 mg mL −1 D

dx.doi.org/10.1021/jf502323f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Table 1. Linear Regression Data and Precision of Investigated Compounds precision (RSD%) linear range analytes aloe emodin rhein emodin chrysophanol physcion

−1

calibration curves y y y y y

= = = = =

intraday (n = 6)

278813x− 2683.5 480700x − 213.04 596132x − 1846.2 151816x − 2730.5 119604x − 2919.1

interday (n = 3)

(μg mL )

r

2

peak area

retention time

peak area

retention time

0.0122−1.2200 0.0115−1.1500 0.0084−0.8400 0.0245−2.4500 0.0219−2.1900

0.9995 0.9994 0.9998 0.9997 0.9998

0.10 0.25 0.20 0.23 0.11

0.30 0.27 0.24 0.22 0.18

1.11 1.12 1.03 1.19 0.95

0.66 0.60 0.48 0.43 0.34

Table 2. Repeatability, Limits of Detection (LODs), Limits of Quantitation (LOQs), and Recovery of Investigated Compounds repeatability (RSD%) n = 3 analytes

peak area

retention time

aloe emodin rhein emodin chrysophanol physcion

3.08 2.21 1.82 3.55 4.32

0.44 0.46 0.28 0.27 0.21

LOD −1

LOQ −1

recovery

(ng mL )

(ng mL )

%

1.1 0.6 0.4 1.5 2.8

3.8 2.1 1.4 4.9 9.4

89.25 92.66 95.32 90.22 96.48

[C12mim]Cl was added into the MFMs dispersion. The overall results indicated that the extraction efficiency was significantly improved by various interactions between the ions themselves,1,29,30 such as van der Waals forces, electrostatic interactions, and hydrogen bonding. As a result, 0.32 mg mL−1 [C12mim]Cl was chosen in this experiment. Influence of MFMs Concentration. The concentration of MFMs is also a vital parameter for magnetic SPE, as it is directly related to the extraction efficiency of the tested analytes. In this study, to investigate the adsorption capacity of MFMs and to choose the optimal content, 0.32 mg mL−1 ILs and 20 μL of rhubarb-herb-extracting solution were added into diluents containing MFMs ranging in concentration from 3.3 to 16.5 mg mL−1. As illustrated in Figure 6, the adsorption capacity sharply increased with the MFMs concentration up to 6.6 mg mL−1 but then decreased significantly when 9.9 mg mL−1 MFMs was used in the dispersion, decreasing the efficiency of the magnetic SPE. Furthermore, the UV absorption of the five compounds remained substantially unchanged as the MFMs concentration increased from 9.9 to 16.5 mg mL−1. These observations demonstrated that the capacity of MFMs adsorbing compounds was saturated, and it was difficult to elute more analytes from the sorbent in the presence of high concentrations of MFMs in dispersions. According to the above results, 6.6 mg mL−1 was selected as the optimal MMNP concentration and employed in the following studies. Effect of pH. The pH of the sample-extracting solution played an important role in the extraction process, as it determined the type of compounds present and the state of the ILs. Additionally, it was a critical parameter for the MSPE method optimization. In the experiment, the pH of the sample solution was adjusted using 1 M phosphoric acid or 1 M sodium hydroxide to values ranging from 5 to 13. The effect of pH on the extraction efficiency of the analytes is shown in Figure 7. It was apparent that with the increment of pH values from 5 to 11, the magnitude of the extracted peak areas of the target compounds was progressively increased, especially in the

Figure 9. Chromatograms obtained from the analysis of (A) standard solutions, (B) Radix et Rhizoma Rhei sample (sample solution, 20 μL; extraction volume, 1230 μL) using the unenriched method, (C) Radix et Rhizoma Rhei extracted by IL coated MFMs SPE under optimum conditions, (D) Radix et Rhizoma Rhei granule sample (sample solution, 20 μL; sample volume, 1230 μL) using the unenriched method, (E) Radix et Rhizoma Rhei granules extracted by IL-coated MFMs SPE under optimum conditions. Peak identification: 1, aloe emodin; 2, rhein; 3, emodin; 4, chrysophanol; 5, physcion.

E

dx.doi.org/10.1021/jf502323f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Table 3. Comparison of Proposed Extraction Procedure with Other Published Methods for the Extraction of Anthraquinones sample volume

LODs

recovery

analysis time

sorbenta

extraction methodb

instrumental techniquec

(mL)

(ng/mL)

(%)

(min)

C18 CTAB-SCMNPs CTAB-MNPs IL-MFMFMs

SPE magnetic-SPE magnetic-SPE magnetic-SPE

HPLC-UV HPLC-FLD HPLC-DAD UPLC-UV

4 0.1−1 0.02 0.02

230−4610 0.2−10 2.6−6.6 0.4−2.8

81.2−106 92.8−109.9 96.7−106.6 89.25−96.48

45 20 50 8

reference 32 33 34 this work

a

CTAB-SCMNPs, cetyltrimethylammonium bromide−silica-coated magnetic nanoparticles; CTAB-MNPs, cetyltrimethylammonium bromidemagnetic nanoparticles; IL-MFMFMs, ionic-liquid-micelle-functionalized mesoporous Fe3O4 microspheres. bSPE, solid-phase extraction; magneticSPE, magnetic solid-phase extraction. cHPLC-UV, high-performance liquid chromatography-ultraviolet; HPLC-FLD, high-performance liquid chromatography-fluorescence detection; HPLC-DAD, high-performance liquid chromatography-diode array detection; UPLC-UV, ultraperformance liquid chromatography-ultraviolet.

case of the absorption of rhein (peak 2) and chrysophanol (peak 4). This result is ascribed to the presence of the hydroxyl groups (OHs) in the molecular structures of anthraquinones. The lone-pair electrons exposed in OHs could interact with the π electrons in the imidazole structure of IL-coated MFMs, which increased the extraction efficiency of the target analytes. However, the peak areas of the analytes greatly decreased when the pH value increased to 13, indicating that stronger alkalinity of the solution resulted in more difficult desorption of target analytes from functionalized magnetic materials. This finding may imply that the enrichment effect was strongly pH dependent. As a result, a pH of 11 was found to be optimal. Concentration of Acetic Acid in the Eluent. After the IL-coated MFMs and extracted compounds were separated from the aqueous phase, an elution procedure with an organic solvent was required to remove the target compounds from the sorbents. On the basis of our experience and the nature of the desorption solvent, methanol was chosen as an eluent. In addition, the eluent efficiency of methanol was investigated by adding from 0% to 5% acetic acid to adjust the eluent pH. Initially, poor extraction efficiencies were obtained, and the aloe emodin and physcion signals (peaks 1 and 5) were extremely weak (Figure 8) in pure methanol. The extraction efficiency then improved as the acetic acid concentration increased from 1% to 2%. However, when larger volumes of acetic acid (from 3% to 5%) were applied, the peak areas of the five compounds decreased sharply. This may be due to the decreased hydrogen bonding, ionic interactions, and solubilities between the analytes and eluent.31,34 Thus, based on the above observations, 2% acetic acid in methanol was chosen as the optimal eluent. Validation of the Proposed Method. Once optimized, the proposed process was analytically characterized in terms of linearity, precision, repeatability, limits of detection (LODs), limits of quantification (LOQs), and recovery. The calibration curves were established using five target compounds at concentrations ranging from 0.0084 to 2.4537 μg mL−1. The results summarized in Tables 1 and 2 indicate good linearity over the concentration range investigated, and the correlation coefficients of determination (r2) were all greater than 0.9994. To confirm the repeatability, five different working solutions prepared from rhubarb samples using MSPE were analyzed, and the RSDs of peak areas and retention times were found to be lower than 4.32% and 0.46%, respectively. The intra- and interday precision of the proposed method were also calculated as the RSD among concentrations in the middle of the linear range. As listed in Table 1, the RSDs of the peak areas and retention times based on six replicated injections for interday precision were in the range of 0.95%−1.19% and 0.18%−0.30%, respectively. For the intraday precision test, the peak areas and

retention times were less than 0.25% and 0.66%, respectively. The LOD values calculated based on a signal-to-noise ratio (S/ N) of 3 ranged from 0.0004 to 0.0028 μg mL−1. Using an S/N ratio of 10, the LOQ values of these analytes were 0.0014− 0.0094 μg mL−1. Good recoveries were obtained by IL-coated MFMs SPE, ranging from 89.25% to 96.48%. The results indicated that the proposed method could provide good qualitative results for the analysis of the target analytes. Sample Analysis. The optimized IL-based MFMs SPE coupled with UPLC method was applied to the extraction of anthraquinones in dietary supplements collected from a local drugstore. Figure 9 presents representative chromatograms of the standards (Figure 9A) and extracted samples (Figure 9B− E) under the determined optimal conditions. Without the sample pretreatment step (Figure 9B,D), five anthraquinones were not obviously detected by UPLC (below the LOQ of the method) in rhubarb herb and rhubarb granules. On the other hand, target compounds were notably determined using magnetic SPE (Figure 9C,E) without any interference. It was observed that the obtained extraction efficiency of the samples using magnetic SPE was remarkably greater, and the enrichment factors for all analytes were in the range from 23 (for physcion) to 60 (for emodin). The rhubarb herb and rhubarb granules contained 0.0146 mg g−1 and 0.0071 mg g−1 aloe emodin, 0.0089 mg g−1 and 0.0047 mg g−1 rhein, 0.0056 mg g−1 and 0.0024 mg g−1 emodin, 0.1411 mg g−1 and 0.0692 mg g−1 chrysophanol, and 0.0334 mg g−1 and 0.0227 mg g−1 physcion, respectively. It was obvious that the contents of anthraquinones in the rhubarb granules were much lower than those in the rhubarb herb, which may due to the complexity of the production process. The experimental results clearly indicated that the proposed method was a useful tool for the analysis of multiple ingredients in real plant samples. Comparison with Other Methods. Table 3 presents a comparison of the proposed method with other counterparts based on the use of SPE. The comparison is performed on the basis of some operational (sample volume, sorbent) and analytical (LODs, recovery, and analysis time) properties. In the majority of the cases, the extraction method is performed using the principles of magnetic SPE. Our proposal provides low LODs (0.4−2.8 ng/mL) for target analytes by using the lowest amount of sample volume, which is comparable with the HPLC-FLD method. Moreover, the recoveries are comparable with the other counterparts, requiring a shorter analysis time. In summary, IL-MFMFMs can be considered as a promising alternative for the extraction of phytochemical compounds. In conclusion, a novel IL-coated MFMs SPE method coupled with UPLC was developed for the preconcentration of five anthraquinones in dietary supplements. The contents of target F

dx.doi.org/10.1021/jf502323f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

(9) Ho, T. D.; Zhang, C.; Hantao, L. W.; Anderson, J. L. Ionic liquids in analytical chemistry: fundamentals, advances, and perspectives. Anal. Chem. 2014, 86, 262−285. (10) Alothman, Z. A.; Yilmaz, E.; Habila, M.; Shabaka, A.; Soylak, M. Ligandless temperature-controlled ionic liquid-phase microextraction of lead(II) ion prior to its determination by FAAS. Microchim Acta 2013, 180, 669−674. (11) Liang, R.; Bao, Z.; Su, B.; Xing, H.; Yang, Q.; Yang, Y.; Ren, Q. Feasibility of Ionic Liquids as Extractants for Selective Separation of Vitamin D3 and Tachysterol3 by Solvent Extraction. J. Agric. Food Chem. 2013, 61, 3479−3487. (12) Yilmaz, E.; Soylak, M. Ionic liquid-linked dual magnetic microextraction of lead(II) from environmental samples prior to its micro-sampling flame atomic absorption spectrometric determination. Talanta 2013, 116, 882−886. (13) Bwambok, D. K.; Thuo, M. M.; Atkinson, M. B. J.; Mirica, K. A.; Shapiro, N. D.; Whitesides, G. M. Paramagnetic ionic liquids for measurements of density using magnetic levitation. Anal. Chem. 2013, 85, 8442−8447. (14) Laremore, T. N.; Zhang, F. M.; Linhardt, R. J. Ionic liquid matrix for direct UV-MALDI-TOF-MS analysis of dermatan sulfate and chondroitin sulfate oligosaccharides. Anal. Chem. 2007, 79, 1604− 1610. (15) Cao, J.; Li, P.; Yi, L. Ionic liquids coated multi-walled carbon nanotubes as a novel pseudostationary phase in electrokinetic chromatography. J. Chromatogr. A 2011, 1218, 9428−9434. (16) Shao, Y. F.; Yan, B.; Li, Q. P. Magnetic Mesoporous Silica Nanosphere Supported Europium(III) Tetrakis(β-diketonate) Complexes with Ionic Liquid Compounds as Linkers. Eur. J. Inorg. Chem. 2013, 3, 381−387. (17) Li, X. S.; Wu, J. H.; Zhao, Y.; Zhang, W. P.; Gao, Q.; Guo, L.; Yuan, B. F.; Feng, Y. Q. Preparation of magnetic polymer material with phosphate group and its application to the enrichment of phosphopeptides. J. Chromatogr. A 2011, 1218, 3845−3853. (18) He, H.; Yuan, D. H.; Gao, Z. Q.; Xiao, D. L.; He, H.; Dai, H.; Peng, J.; Li, N. Mixed hemimicelles solid-phase extraction based on ionic liquid-coated Fe3O4/SiO2 nanoparticles for the determination of flavonoids in bio-matrix samples coupled with high performance liquid chromatography. J. Chromatogr. A 2014, 1324, 78−85. (19) Qing, L. S.; Xue, Y.; Liu, Y. M.; Liang, J.; Xie, J.; Liao, X. Rapid magnetic solid-phase extraction for the selective determination of isoflavones in soymilk using baicalin-functionalized magnetic nanoparticles. J. Agric. Food Chem. 2013, 61, 8072−8078. (20) Wang, L.; Zang, X. H.; Chang, Q. Y.; Zhang, G. J.; Wang, C.; Wang, Z. Determination of triazole fungicides in vegetable samples by magnetic solid-phase extraction with graphene-coated magnetic nanocomposite as adsorbent followed by gas chromatography−mass spectrometry detection. Food Anal. Methods 2014, 7, 318−325. (21) Li, N.; Qi, L.; Shen, Y.; Li, Y. P.; Chen, Y. Amphiphilic block copolymer modified magnetic nanoparticles for microwave-assisted extraction of polycyclic aromatic hydrocarbons in environmental water. J. Chromatogr. A 2013, 1316, 1−7. (22) Farré, M.; Sanchís, J.; Barceló, D. Analysis and assessment of the occurrence, the fate and the behavior of nanomaterials in the environment. TrAC, Trends Anal. Chem. 2011, 30, 517−527. (23) Han, Q.; Wang, Z.; Xia, J.; Chen, S.; Zhang, X.; Ding, M. Facile and tunable fabrication of Fe3O4/graphene oxide nanocomposites and their application in the magnetic solid-phase extraction of polycyclic aromatic hydrocarbons from environmental water samples. Talanta 2012, 101, 388−395. (24) Wang, G. Y.; Chen, F. F.; Shi, Y. P. Ultra-performance LCphotodiode array-eλ-ESI-MS/MS screening method for the detection of radical-scavenging natural antioxidants fromRadix et Rhizoma Rhei. J. Sep. Sci. 2011, 34, 268−277. (25) Xuan, S. H.; Wang, F.; Lai, J. M. Y.; Sham, K. W. Y.; Wang, Y. X. J.; Lee, S. F.; Yu, J. C.; Cheng, C. H. K.; Leung, K. C. F. Synthesis of biocompatible, mesoporous Fe3O4 nano/microspheres with large surface area for magnetic resonance imaging and therapeutic applications. ACS Appl. Mater. Interfaces 2011, 3, 237−244.

components were successfully determined with satisfactory precision, recovery, and repeatability. Compared with traditional extraction modes, this proposed method has the advantages of simplicity, rapidity, ease of operation, and environmental-friendliness. Therefore, IL-functionalized MFMs SPE could have high analytical potential for the extraction of other analytes from complex plant samples.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Fax: +86 571 2886 7909. Tel.: +86 571 2886 7909. Author Contributions §

S.-S.H., W.C., H.-B.D., and J.-H.D. contributed equally to this work. Funding

This study was supported by General Program of National Natural Science Foundation of China (No. 81274065), Research on Public Welfare Technology Application Projects of Zhejiang Province (No. 2014C37069), Changjiang Scholars and Innovative Research Team in Chinese University (IRT 1231), Young and Middle-Aged Academic Leaders of Hangzhou (2013−45), Scientific Research Foundation of Hangzhou Normal University (2011QDL33), and the NewShoot Talents Program of Zhejiang province (2013R421044, 2014R421019). Notes

The authors declare no competing financial interest.



REFERENCES

(1) Reyes-Gallardo, E. M.; Lasarte-Aragonés, G.; Lucena, R.; Cárdenas, S.; Valcárcel, M. Hybridization of commercial polymeric microparticles and magnetic nanoparticles for the dispersive microsolid phase extraction of nitroaromatic hydrocarbons from water. J. Chromatogr. A 2013, 1271, 50−55. (2) Wang, Y. G.; Ren, J. W.; Liu, X. H.; Wang, Y. Q.; Guo, Y.; Guo, Y. L.; Lu, G. Z. Facile synthesis of ordered magnetic mesoporous gammaFe2O3/SiO2 nanocomposites with diverse mesostructures. J. Colloid Interface Sci. 2008, 326, 158−165. (3) Mashhadizadeh, M. H.; Karami, Z. Solid phase extraction of trace amounts of Ag, Cd, Cu, and Zn in environmental samples using magnetic nanoparticles coated by 3-(trimethoxysilyl)-1-propantiol and modified with 2-amino-5-mercapto-1,3,4-thiadiazole and their determination by ICP-OES. J. Hazard. Mater. 2011, 190, 1023−1029. (4) Lasarte-Aragonés, G.; Lucena, R.; Cárdenas, S.; Valcárcel, M. Effervescence assisted dispersive liquid-liquid microextraction with extractant removal by magnetic nanoparticles. Anal. Chim. Acta 2014, 807, 61−66. (5) Mashhadizadeh, M. H.; Amoli-Diva, M.; Pourghazi, K. Magnetic nanoparticles solid phase extraction for determination of ochratoxin A in cereals using high-performance liquid chromatography with fluorescence detection. J. Chromatogr. A 2013, 1320, 17−26. (6) Liu, X. D.; Yu, Y. J.; Li, Y.; Ning, S. L.; Liu, T. T.; Li, F. J.; Duan, G. L. Restricted access magnetic core-mesoporous shell microspheres with C8-modified interior pore-walls for the determination of diazepam in rat plasma by LC-MS. Talanta 2013, 106, 321−327. (7) Liu, H.; Lin, C.; Ma, Z.; Yu, H. B.; Zhou, S. H. Gold nanoparticles on mesoporous SiO2-coated magnetic Fe3O4 spheres: a magnetically separatable catalyst with good thermal stability. Molecules 2013, 18, 14258−14267. (8) Joshi, M. D.; Steyer, D. J.; Anderson, J. L. Evaluating the complexation behavior and regeneration of boron selective glucaminium-based ionic liquids when used as extraction solvents. Anal. Chim. Acta 2012, 740, 66−73. G

dx.doi.org/10.1021/jf502323f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

(26) Zhang, D. P.; Lu, C. H.; Ni, Y. R.; Xu, Z. Z.; Zhang, W. B. Effect of water on size-controllable synthesis of mesoporous Fe3O4 microspheres and their applications in waste water treatment. CrystEngComm 2013, 15, 4755−4764. (27) Jungnickel, C.; Łuczak, J.; Ranke, J.; Fernández, J. F.; Müller, A.; Thöming, J. Micelle formation of imidazolium ionic liquids in aqueous solution. Colloids Surf., A 2008, 316, 278−284. (28) Greaves, T. L.; Drummond, C. J. Solvent nanostructure, the solvophobic effect and amphiphile self-assembly in ionic liquids. Chem. Soc. Rev. 2013, 42, 1096−1120. (29) Galán-Cano, F.; Alcudia-León, M. d. C.; Lucena, R.; Cárdenas, S.; Valcárcel, M. Ionic liquid coated magnetic nanoparticles for the gas chromatography/mass spectrometric determination of polycyclic aromatic hydrocarbons in waters. J. Chromatogr. A 2013, 1300, 134−140. (30) Hu, S. S.; Yi, L.; Li, X. Y.; Cao, J.; Ye, L. H.; Cao, W.; Da, J. H.; Dai, H. B.; Liu, X. J. Ionic liquid-based one-step micellar extraction of multiclass polar compounds from hawthorn fruits by ultrahighperformance liquid chromatography coupled with quadrupole timeof-flight tandem mass spectrometry. J. Agric. Food Chem. 2014, 62, 5275−5280. (31) Bouri, M.; Gurau, M.; Salghi, R.; Cretescu, I.; Zougagh, M.; Rios, Á . Ionic liquids supported on magnetic nanoparticles as a sorbent preconcentration material for sulfonylurea herbicides prior to their determination by capillary liquid chromatography. Anal. Bioanal. Chem. 2012, 404, 1529−1538. (32) Fernand, V. E.; Dinh, D. T.; Washington, S. J.; Fakayode, S. O.; Losso, J. N.; van Ravenswaay, R. O.; Warner, I. M. Determination of pharmacologically active compounds in root extracts of Cassia alata L. by use of high performance liquid chromatography. Talanta 2008, 74, 896−902. (33) Zhu, L.; Pan, D.; Ding, L.; Tang, F.; Zhang, Q. L.; Liu, Q.; Yao, S. Z. Mixed hemimicelles SPE based on CTAB-coated Fe3O4/ SiO2NPs for the determination of herbal bioactive constituents from biological samples. Talanta 2010, 80, 1873−1880. (34) Zhang, H. F.; Shi, Y. P. Pretreatment properties of CTAB coated Fe3O4 nanoparticle mixed hemimicelle sorbents for the analysis of herbal medicine samples. Curr. Anal. Chem. 2012, 8, 150−158.

H

dx.doi.org/10.1021/jf502323f | J. Agric. Food Chem. XXXX, XXX, XXX−XXX