Journal of Chromatography A, 1406 (2015) 40–47
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Magnetic solid phase extraction and gas chromatography–mass spectrometrical analysis of sixteen polycyclic aromatic hydrocarbons Ying Cai a , Zhihong Yan b , Manh NguyenVan a,c , Lijia Wang a , Qingyun Cai a,∗ a b c
State Key Laboratory of Chemo/Biosensing & Chemometrics, College of Chemistry & Chemical Engineering, Hunan University, Changsha 410082, China Key Laboratory of Modern Preparation of TCM, Ministry of Education, Jiangxi University of Traditional Chinese Medicine, Nanchang 330004, China Faculty of Chemical Technolology, Hanoi University of Industry, Hanoi, Viet Nam
a r t i c l e
i n f o
Article history: Received 1 April 2015 Received in revised form 8 June 2015 Accepted 10 June 2015 Available online 20 June 2015 Keywords: Fluorenyl functionalized MNPs Polycyclic aromatic hydrocarbons Water sample – interaction
a b s t r a c t Fluorenyl functionalized superparamagnetic core/shell magnetic nanoparticles (MNPs, Fe3 O4 @SiO2 @Flu) were prepared and characterized by transmission electron microscope, X-ray diffraction and infrared spectroscopy. The MNPs having an average diameter of 200 nm were then used as solid-phase extraction sorbent for the determination of 16 priority pollutants polycyclic aromatic hydrocarbons (PAHs) in water samples designated by United States Environmental Protection Agency (U.S. EPA). The main influencing parameters, including sorbent amount, desorption solvent, sample volume and extraction time were optimized. Analyses were performed on gas chromatography–mass spectrometry (GC–MS) using selected ion monitoring (SIM) mode. Method validation proved the feasibility of the developed sorbents for the quantitation of the investigated analytes at trace levels. Limit of detection ranging from 0.5 to 4.0 ng/L were obtained. The repeatability was investigated by evaluating the intra- and inter-day precisions with relative standard deviations (RSDs) lower than 13.1%. Finally, the proposed method was successfully applied for the determination of PAHs in water samples with the recoveries in the range of 96.0–106.7%. © 2015 Elsevier B.V. All rights reserved.
1. Introduction It is well known that water pollution has become an important problem related to public health. Among the pollutants, polycyclic aromatic hydrocarbons (PAHs) have raised global concerns because of their widespread distribution and carcinogenic toxicity [1]. The U.S. EPA has formulated regulations for the monitoring and control of PAHs and has developed methods for their measurement in air, water, food, and other matrixes [2]. The EPA has designated 16 PAHs as priority pollutants [3], and eight of them are considered to be possible carcinogens, namely: benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, dibenzo[a,h]anthracene, indenol[1,2,3-cd]pyrene and benzo[g,h,i]perylene [4]. PAHs in different matrices were characterized and quantified mainly by HPLC or GC-based methods, such as Method 610 which deals with the measurement of these PAHs in wastewater by either HPLC/UV and FLD or gas chromatography with flame ionization detection (GC/FID) [5]. Method 8100 analyzes PAHs by GC/FID [6].
∗ Corresponding author at: State Key Laboratory of Chemo/Biosensing & Chemometrics, College of Chemistry & Chemical Engineering, Hunan University, Changsha 410082, China. Tel.: +86 73188821848. E-mail addresses: [email protected], [email protected] (Q. Cai). http://dx.doi.org/10.1016/j.chroma.2015.06.024 0021-9673/© 2015 Elsevier B.V. All rights reserved.
Methods 525, 625, and 8270 which analyze several PAHs and many other environmental pollutants in drinking water, wastewater, and solid wastes, respectively, by GC/MS [7–9]. Method 8275 which is a thermal extraction coupled with capillary GC/MS procedure for the quantitative determination of targeted PCBs and the 16 EPA priority pollutants PAHs in soils, sludges, and solid wastes [10]. However, the trace content in surface water makes the determination of PAHs a challenging task. Effective extraction and preconcentration techniques prior to instrumental analysis are often required. The widely used pre-treatment method at present for PAHs from aqueous samples are mainly sorbent based techniques, such as solid phase extraction (SPE) [11–14], microsolid-phase extraction (-SPE) [15], and magnetic solid-phase extraction (MSPE) [16–29]. MSPE is a procedure for the preconcentration of target analytes from large volumes by using magnetic or magnetizable sorbents [30], which can be easily separated from sample solutions by an external magnetic field without the need of additional centrifugation or filtration. Moreover, since the sorbent can be dispersed in sample solution, the contact surface between analytes and sorbents is very large, which results in a rapid mass transfer and rapid extraction equilibrium. For the above reasons, MSPE has attracted increasing interests and found wide applications in the field of sample preparations. Several MNPs were synthesized for the extraction of U.S. EPA 16 PAHs, they are C18 functionalized Fe3 O4 nanoparticles
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
(magnetic C18 ) [31], diphenyl functionalization of Fe3 O4 MNPs (Fe3 O4 diphenyl) [32], carbon–ferromagnetic nanocomposite (carbon–Fe3 O4 ) [33] and n-octadecylphosphonic acid modified mesoporous MNPs (OPA/MMNPs) [34]. Generally, – conjugative effect and hydrophobic interaction were contributed to adsorb PAHs. To the best of our knowledge, there is no report about preparation of condensed cyclic structural MNPs, and their application as SPE sorbents for adsorption of PAHs. Also, there is lack of report about the adsorption capacity of the MNPs. In the present study, a new kind of MNPs termed as fluorenyl functionalized magnetic silica nanoparticles (Fe3 O4 @SiO2 @Flu) was developed to extract PAHs from water samples. Fe3 O4 @SiO2 @Flu MNPs were synthesized by coating/modification magnetite Fe3 O4 nanoparticles with silica, subsequently 3-isocyanatopropyltriethoxysilane and finally 9-fluorenylmethyl chloroformate. To our knowledge, this study is the first report with respect to the introduction of fluorenyl to functionalize MNPs for the extraction of PAHs. Due to the condensed cyclic structure, rich -electrons exist on the surface of fluorenyl, which leads to the adsoption of PAHs through the – conjugative effect. Finally, adsorption capacity of Fe3 O4 @SiO2 @Flu was studied, and the sorbents was used for enrichment of PAHs from environmental water samples. 2. Experimental 2.1. Reagents and materials All reagents were procured from major suppliers. Tetraethoxysilane (TEOS, 99%), 3-aminopropyltriethoxysilane (APTES, 99%) and 9-fluorenylmethyl chloroformate were obtained from Adamas Reagent Ltd. (Switzerland). HPLC grade methanol, n-hexane and acetonitrile (99%) were bought from Merck (Darmstadt, Germany). Ferric chloride hexahydrate (FeCl3 ·6H2 O, 99%), ferrous sulfate heptahydrate (FeSO4 ·7H2 O, 99%), ammonia (26%), hydrazine hydrate (99%), isopropanol (99%), triethylamine (99%), toluene (99%) were purchased from Sinopharm Chemical Reagent Co. Ltd. (China). Ultrapure water was used through the experiment which was prepared using a Milli-Q system water purification system (Millipore Inc., USA). 610/8100 PAH standard solution was bought from ANPEL Laboratory Technologies (Shanghai) Inc. The PAH standard solution contains naphthalene (Nap), acenaphthylene (Aceny), acenaphthene (Acen), fluorene (Flu), phenanthrene (Phen), anthracene (Ant), fluoranthene (Fluo), pyrene (Pyr), chrysene (Chr), benzo[a]anthracene (B[a]A), benzo[b]fluoranthene (B[b]F), benzo[k]fluoranthene (B[k]F), benzo[a]pyrene (B[a]P), indeno[1,2,3-cd]pyrene (I[1,2,3-cd]P), dibenzo[a,h]anthracene (D[a,h]A) and benzo[g,h,i]perylene (B[ghi]P), each at 2 mg/mL in benzenen/methylene dichloride (1/1, v/v). The PAH stock solution was prepared in methanol at a concentration of each at 2 g/mL, and kept at 4 ◦ C in darkness. PAH working solutions were prepared by proper dilution of the stock solution. Tap water, river water and waste water were used for real sample investigation. Tap water was taken from our lab tap after flowing for 10 min. River water sample was collected from the Ganjiang River in Nanchang (Jiangxi, China) and collected at 10 cm depth below the water surface. Waste water sample was collected from water treatment plants effluents (Jiangxi, China). Water samples were analyzed within 24 h.
41
(FTIR) were recorded on Vertex 70 (Bruker Optics, Germany). After being dried at 80 ◦ C in a vacuum oven for 12 h, samples were mixed with KBr to fabricate a KBr pellet for FTIR analysis. PAHs were extracted from water samples with the assistance of an ultrasonic instrument KQ-600KDE. The magnetic property was analyzed using a vibrating sample magnetometer (VSM, Model 7410, Lake Shore Cryotronics, Inc., Westerville, OH, USA). Analyses were conducted using an Agilent 7890A GC (Agilent Technologies, USA) system coupled to an Agilent 5975C mass spectrometer (Agilent Technologies, USA). Helium (purity 99.999%) was employed as the carrier gas at a flow rate of 1.8 mL/min. Samples (1 L) were injected in splitless mode. The injector temperature was set at 280 ◦ C and the interface temperature maintained at 280 ◦ C. The chromatographic separation of the PAHs was performed on a 30 m × 0.25 mm × 0.25 m HP5 capillary column (Agilent). The GC oven was initially held at 80 ◦ C for 1 min and increased to 250 ◦ C at 25 ◦ C/min and held for 6 min, then increased to 300 ◦C at 10 ◦ C/min and held for 2 min. Preliminarily, full scan electron ionization data were acquired to determine appropriate masses under the following conditions: ionization energy: 70 eV, mass range: 35–350 amu, scan time: 3 scan/s. PAH standards and samples were finally analyzed in selective ion monitoring (SIM) mode for quantitative determination of the analytes. The masses monitored were as follows: 3–5 min, m/z 128, 127, 102 for naphthalene; 5–6 min, m/z 152, 154 for acenaphthylene and acenaphthene; 6–7 min, m/z 166, 165 for fluorene; 7–8 min, m/z 178 for phenanthrene and anthracene; 8–10 min, m/z 202, 203, 200 for fluoranthene and pyrene; 10–13 min, m/z 228, 226, for chrysene and benzo[a]anthracene; 13–18 min, m/z 252 for benzo[b]fluoranthene, benzo[k]fluoranthene and benzo[a]pyrene; 18–22 min, m/z 276, 278 for indeno[1,2,3cd]pyrene, dibenzo[a,h]anthracene and benzo[g,h,i]perylene. The solvent delay time was 3 min (to bypass the solvent peak). For all the investigated analytes the corresponding ion ratios were used for confirmation purposes. Signal acquisition and data processing were performed using the Agilent MSD Chemstation (G1701EA, Agilent Technologies). 2.3. Preparation of Fe3 O4 @SiO2 @Flu magnetic nanoparticles 2.3.1. Preparation of Fe3 O4 MNPs The Fe3 O4 MNPs were prepared by chemical co-precipitation. Briefly, FeCl3 ·6H2 O (4.0 g) was dissolved in deionized water (30 mL) in a three-necked round bottom flask, followed by addition of hydrazine hydrate (2 mL) and FeSO4 ·7H2 O (10.90 g) to prepare a stock solution. Afterwards, ammonia solution (35 mL 26.5% w/w) was added into the stock solution under vigorous stirring, followed by dropwise addition of ammonia solution (26.5% w/w) until the solution pH reached 9. The solution was stirred at room temperature for 30 min, aged at 80 ◦ C for 60 min, and then cooled to room temperature. The product was magnetically collected, and washed with water, finally vacuum-dried at 60 ◦ C for 12 h.
2.2. Instrumentations and analytical conditions
2.3.2. Encapsulation of the MNPs with silica (Fe3 O4 @SiO2 ) Nanoparticles (1 g) were dispersed in a mixture of 2-propanol (100 mL) and ultrapure water (8 mL), sonicated for 15 min, then added with ammonia solution (10 mL, 26.5% w/w) and TEOS (8 mL) sequentially. After stirring the mixture for 12 h at 45 ◦ C, the MNPs were collected by a magnet, washed with water and ethanol respectively, and vacuum-dried at 60 ◦ C for 12 h. Through this procedure, the ferro-MNPs were encapsulated with a layer of mesoporous silica.
The size and morphological characterization of the particles were observed by transmission electron microscopy (TEM, JEM2100F, JEOL Co., Tokyo, Japan). Fourier transform infrared spectra
2.3.3. Preparation of fluorenyl coated MNPs (Fe3 O4 @SiO2 @Flu) The preparation scheme of Fe3 O4 @SiO2 @ Flu is depicted in Fig. 1. Firstly, the above Fe3 O4 @SiO2 nanoparticles (3 g) were
42
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
Fig. 1. Preparation scheme of Fe3 O4 @SiO2 @Flu.
suspended in toluene (60 mL), then triethylamine (1 mL) and APTES (6 mL) were added under argon atmosphere. The suspension was mechanically stirred and refluxed at 110 ◦ C for 24 h. The reaction was stopped and the APTES MNPs were cooled to room temperature, washed with toluene, acetone and ethanol respectively, finally dried under vacuum at 60 ◦ C. Next, the APTES MNPs (3 g) were mixed with toluene (40 mL) and triethylamine (1 mL) under argon atmosphere, after stirring for 30 min in ice-bath, 9-fluorenylmethyl chloroformate (3.8 g) was added. The mixture was refluxed for 4 h at room temperature and then 12 h at 110 ◦ C. The reaction was stopped and the as-prepared Fe3 O4 @SiO2 @Flu MNPs were cooled to room temperature, washed with ethanol and water/ethanol (1/1 v/v). The resulting product was dried under vacuum at 60 ◦ C for 12 h.
gurantee a complete separation of the sorbents from solution even analyte was loaded. FTIR spectra were acquired for Fe3 O4 @SiO2 and Fe3 O4 @SiO2 @Flu MNPs between 4000 and 400 cm−1 (Fig. S1). As seen in the figure, the appearance of characteristic peaks
2.4. Extraction procedure 40 mg Fe3 O4 @SiO2 @Flu magnetic sorbents was added into 100 mL PAHs-spiked water sample. After sonicating the mixture for 30 s to accelerate the dispersion of the sorbents in solution, the MNPs were isolated rapidly from the solution by applying an external magnetic field via a supermagnet (Nd–Fe–B). An aliquot of 1.0 mL n-hexane was introduced to desorb the captured PAHs from the nanoparticles by sonication for 15 s. Finally, the extract was separated from the suspension with a supermagnet and filtered through a 0.22 m polytetrafluoroethylene membrane syringe filter, 1 L of the filtrate was injected into GC–MS system for analysis. 3. Results and discussion 3.1. Characterization of Fe3 O4 @SiO2 @Flu MNPs The prepared MNPs were characterized with TEM, VSM and FTIR. TEM was used to obtain structure information of the Fe3 O4 , Fe3 O4 @SiO2 and Fe3 O4 @SiO2 @Flu MNPs. As shown in Fig. 2, the synthesized Fe3 O4 MNPs are quasi-spherical in shape at an average size of 20 nm in diameter. The TEM image of Fe3 O4 @SiO2 MNPs shows that a uniform gray silica shell is coated on the Fe3 O4 MNPs with an average diameter of 100 nm (Fig. 2b). The silica shell provides abundant silanol groups for further chemical modification. The Fe3 O4 @SiO2 @Flu MNPs (Fig. 2c) are larger than Fe3 O4 @SiO2 MNPs with an average diameter of 200 nm, confirming the immobilization of flu shell. Sufficient magnetic properties are important for MSPE. The magnetic properties of the prepared nanoparticles were investigated with a VSM. Fig. 3 shows the magnetization curves of Fe3 O4 , Fe3 O4 @SiO2 and Fe3 O4 @SiO2 @Flu at 300 K, giving the magnetic saturation values of 71.6, 30.7 and 27.48 emu/g, respectively. The coating results in decreases in the magnetic saturation values. The magnetic saturation value of Fe3 O4 @SiO2 @Flu sorbents can
Fig. 2. TEM images Fe3 O4 (a), Fe3 O4 @SiO2 (b) and Fe3 O4 @SiO2 @Flu (c).
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
43
3.2. Optimization of extraction conditions
Fig. 3. Magnetization curves of Fe3 O4 , Fe3 O4 @SiO2 and Fe3 O4 @SiO2 @Flu.
at 1089 and 475 cm−1 are attributed to the Si O Si stretching vibration and Si O symmetric stretching vibration, respectively. After modification with fluorenyl, there displays a prominent peak at 768 cm−1 , which is characteristic of H on fluorene ring. In addition, a band at 1768 cm−1 corresponding to the bending vibration of skeleton of aromatic ring appears. These results confirmed the success surface modification of the MNPs.
The effects of several parameters on extraction efficiency were studied, such as the sorbent amount, type of desorption solvent, water sample volume and extraction time. The influence of all these parameters was evaluated in terms of peak areas. The optimization experiments were conducted using water sample spiked with 200 ng/L of each PAH. Each experiment was performed in triplicate. The effect of amount of Fe3 O4 @SiO2 @Flu MNPs was investigated with the amount varying from 10 to 60 mg. Fig. 4a demonstrates that the peak areas of almost all the 16 PAHs increase continuously with the increase of the sorbent amount from 10 to 40 mg. Further increasing the sorbent amount over 40 mg results in no obvious increase of the peak area. These results indicated that 40 mg of sorbents are sufficient to extract PAHs. Excess sorbent may retain analytes and result in a decrease in peak area. So, 40 mg of sorbent was used for the following experiments. An appropriate type of elution solvent is fundamental for extraction efficiency by Fe3 O4 @SiO2 @Flu MNPs. Four types of solvents were selected as desorption solvent, including acetone, dichloromethane, n-hexane and acetonitrile. As shown in Fig. 4b, dichloromethane and n-hexane yields higher analytical signals than acetone and acetonitrile. For most PAHs, n-hexane was the best desorption solvent. The effect of sample volume on extraction efficiency of PAHs was investigated with the volume changing from 100 to 2000 mL. As shown in Fig. 4c, the signals of all the PAHs declined significantly with increasing the sample volume from 100 to 2000 mL.
Fig. 4. Effect of (a) sorbent amount, (b) type of desorption solvent, (c) sample volume, (d) extraction time on the peak areas of PAHs. Amount of sorbent, 40 mg; sample volume: 100 mL; desorption solvent, n-hexane (1.0 mL); extraction time, 5 min.
44
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
Fig. 5. Adsorption isotherms for PAHs on Fe3 O4 @SiO2 @Flu MNPs.
This behavior occurs possibly because that when increasing the sample volume, especially for 500–2000 mL, the magnetic sorbents would be dispersed more widely, and some sorbents will loss during the recovery process. So, 100 mL was selected as sample volume. Fig. 4d shows the effect of extraction time on the extraction efficiency. One can see that there is no obvious variation of the peak area when increasing the extraction time from 2 to 20 min, and when extraction time was increased to 30 min, the analytical signals decreased a little. With this clarification, 5 min was sufficient to achieve satisfactory extraction efficiency for most PAHs. Therefore, extraction time of 5 min was applied in the following MSPE procedure. Based on the above discussion, the optimal conditions of MSPE with Fe3 O4 @SiO2 @Flu sorbent for PAHs are: 40 mg Fe3 O4 @SiO2 @Flu MNPs, 100 mL solution volume, 5 min extraction time, and n-hexane as the desorption solvent. 3.3. Stability of Fe3 O4 @SiO2 @Flu The stability of the Fe3 O4 @SiO2 @Flu sorbent was evaluated based on the reproducibility of extraction efficiency using different batches of magnetic sorbent and the reusability of the sorbent. The reproducibility of the extraction was investigated using three different batches of Fe3 O4 @SiO2 @Flu MNPs with water sample spiked with 200 ng/L of each PAH. The peak area and RSD values of 16 PAHs extracted with the three different batches of magnetic sorbents are listed in Table S1. The RSD values are lower than 8.7%, indicating a good reproducibility. The regeneration of the magnetic sorbent was investigated with a random batch of Fe3 O4 @SiO2 @Flu as sorbent. The used Fe3 O4 @SiO2 @Flu (40 mg) was regenerated by rinsing it with 1 mL of n-hexane twice to make sure that no residual PAHs were remained. The regenerated sorbent was applied in MSPE of water samples which were spiked with 200 ng/L of each PAHs. As shown in Fig. S2, after 10 times of recycling, there is no significant change in the peak areas for each analyte extracted by the Fe3 O4 @SiO2 @Flu MNPs, indicating that the Fe3 O4 @SiO2 @Flu sorbents are stable and durable during MSPE procedure with good reusability. 3.4. Adsorption capacity In adsorption isotherm studies, solutions with different initial concentrations (0.05–5.0 g/mL) were added. Fe3 O4 @SiO2 @Flu MNPs (40 mg) and PAHs solution (100 mL) were mixed in a flask, then MNPs were separated by applying an external magnetic field
after adsorption equilibrium. The equilibrium concentration of PAHs was determined by GC–MS system. The condition for the determination was the same as Section 2.2. As shown in Fig. 5, the calculated maximal adsorption capacities of Nap, Aceny, Acen, Flu, Phen, Ant, Fluo, Pyr, Chr, B[a]A, B[b]F, B[k]F, B[a]P, I[1,2,3-cd]P, D[a,h]A and B[ghi]P on Fe3 O4 @SiO2 @Flu were 325, 411, 252, 258, 230, 268, 432, 436, 335, 393, 297, 349, 215, 227, 338 and 211 ng/mg, respectively. 3.5. Analytical figures of merit Under the optimal conditions mentioned above, a series of experiments with regard to the linearity (linear ranges and coefficient of determination), limit of detection (LOD), limit of quantification (LOQ), precision and enrichment factor (EF) were performed to validate the proposed method. In order to investigate the possible matrix effect on determination, the linearity of the proposed method was estimated by analyzing Ganjiang river water samples spiked with various concentrations of PAHs over a range of 10 and 1000 ng/L (10, 50, 100, 250, 500, 1000 ng/L). The calibration curves were obtained by plotting the mean peak area versus sample concentration. The results are listed in Table 1. All the analytes show good linearity with coefficient of determination ranging from 0.9900 to 0.9998. The LOD and LOQ were calculated based on the standard deviation of the response and the slope of calibration curve [35] from the following equations: LOD = 3.3Sb/m (1) LOQ = 10Sb/m (2) where Sb is the standard deviation of the y-intercept of regression line and m is the slope of the calibration curve. The LOD values were from 0.5 (pyr) to 4.0 ng/L (I[1,2,3-cd]P), and the LOQ values were from 1.7 to 13.3 ng/L. Accuracy was calculated in terms of recovery rate (R%) on real water samples as follows: R% = (c 1 /c 2 ) × 100% where c1 is the measured concentration and c2 is the concentration calculated from the quantity spiked into the sample. Relative extraction recoveries ranged from 96.6 to 105.7%. The EF of the method was calculated using the following equation [36,37]: EF = (V s /V e ) × R% where Vs is the sample volume, Ve is the elution volume, and R% is percent recovery. The EF were determined to be 96.6, 103.4,
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
45
Table 1 MSPE–GC/MS analytical figures of merit for the16 PAHs. Compound
Linearity range (ng/L)
R2
Calib. curvea (n = 3)
LOD (ng/L)
LOQ (ng/L)
Recoveries (%)
RSD (%) (n = 3)
Enrichment factor
Nap Aceny Acen Flu Phen Ant Fluo Pyr B[a]A Chr B[b]F B[k]F B[a]P I[1,2,3-cd]P DB[a,h]A B[ghi]P
10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000
0.9992 0.9950 0.9962 0.9941 0.9897 0.9900 0.9904 0.9818 0.9989 0.9983 0.9997 0.9998 0.9998 0.9990 0.9988 0.9989
y = 1017.6x + 8517.1 y = 149.91x + 17825 y = 195.72x + 19164 y = 226.57x + 24799 y = 412.26x + 57672 y = 393.05x + 15494 y = 789.74x + 77017 y = 789.37x + 216948 y = 710.28x + 11930 y = 695.18x + 18077 y = 481.45x − 2913.3 y = 435.14x + 328.57 y = 590.41x − 2441.6 y = 287.43x − 4630 y = 339.03x − 5536.1 y = 507.83x + 1575.4
0.8 1.9 1.6 1.3 1.4 1.1 0.6 0.5 1.4 1.7 2.5 2.9 4.0 3.4 2.9 3.1
2.7 6.3 5.3 4.3 4.7 3.7 2.0 1.7 4.7 5.7 8.3 9.7 13.3 11.3 9.7 10.3
96.6 103.4 101.5 100.8 105.3 103.8 105.7 101.0 99.0 98.6 100.5 100.9 103.6 99.9 103.8 99.6
3.5 2.7 2.1 1.8 1.9 2.0 1.5 2.2 1.9 1.4 2.7 2.7 4.9 5.0 6.2 6.9
96.6 103.4 101.5 100.8 105.3 103.8 105.7 101.0 99.0 98.6 100.5 100.9 103.6 99.9 103.8 99.6
a
y is peak area and x is compound concentration (ng/L).
101.5,100.8, 105.3, 103.8, 105.7, 101.0, 99.0, 98.6, 100.9, 100.5, 103.6, 99.9, 103.8 and 99.6 for Nap, Aceny, Acen, Flu, Phen, Ant, Fluo, Pyr, B[a]A, Chr, B[b]F, B[k]F, B[a]P, I[1,2,3-cd]P, DB[a,h]A and B[ghi]P, respectively. Repeatability or intra-day precision was evaluated by spiking Ganjiang River samples with appropriate amounts of PAHs at three different levels (50, 100, 400 ng/L). Six replicates were performed and analyzed on the same day. In order to determine reproducibility or inter-day precision, samples that were spiked with the same amount of PAHs, were analyzed in six consecutive days at three replicate levels. The intra and inter-day precisions were expressed as the percentage relative standard deviation (RSD%) (Table 2). The intra-day precision was from 0.8 to 7.5% and the inter-day precision for the PAHs were below 13.1%, illustrating the good repeatability achieved by the suggested procedure. These results imply that the proposed method can be applied to the analysis of real samples containing PAHs at trace level.
sample, benzo[a]pyrene and dibenzo[a,h]anthracene were detected except naphthalene, acenaphthylene and anthracene. The relative recoveries of PAHs at two concentration levels are in the range of 96.0–106.7%, with RSDs within 6.1%. These results imply that the established method can be applied to the analysis of PAHs at trace level in real samples. 3.7. Comparison with other sorbents The proposed magnetic sorbent is compared with other solidbased sorbents [11–15,31–34] in Table 3. The comparison is performed on the basis of some operational (sorbent amount, extraction time and sample volume) and analytical properties (GC–MS elution time, LOD, recovery and precision). Firstly, for comparison with traditional SPE, the amount of magnetic Fe3 O4 @SiO2 @Flu sorbent was much less than SPE sorbents such as C18 [11] and carbon nanotubes [13], and the MSPE procedure was much quicker than SPE [12–15]. Secondly, for comparison with other magnetic sorbents, the Fe3 O4 @SiO2 @Flu shows similar recoveries to the most of other sorbents, and the LODs were comparable to others. It needs less amount of sorbent than magnetic C18 [31] and OPA/MMNPs [34], but more amount of sorbent than Fe3 O4 diphenyl [32] and carbon-Fe3 O4 Fe3 O4 /C [33]. Finally, our GC–MS elution time was the shortest. In addition, the abundant electrons of naphthyl provide potent – stacking interactions with PAHs, contributing to the selectivity to PAHs. The adsorption equilibrium in the process of extraction can be quickly achieved due to the good
3.6. Analysis of real water samples To further evaluate the method applicability, Ganjiang river water, tap water and waste water samples were analyzed. Unspiked water samples and water samples spiked with 16 PAHs at two concentration levels (50 and 100 ng/L) were analyzed by the proposed method (n = 3). The results are listed in Table S2. Naphthalene, acenaphthylene and anthracene were found in both tap water and Ganjiang river water samples, and in waste water Table 2 The precision of MSPE method. Compound
Nap Aceny Acen Flu Phen Ant Fluo Pyr B[a]A Chr B[b]F B[k]F B[a]P I[1,2,3-cd]P DB[a,h]A B[ghi]P
Intra-day precision (RSD%, n = 6)
Inter-day precision (RSD%, n = 6)
50 ng/L
100 ng/L
400 ng/L
50 ng/L
100 ng/L
400 ng/L
4.3 1.2 3.2 1.3 1.6 1.6 2.3 1.8 2.1 2.4 6.1 1.9 4.5 5.1 7.4 7.5
2.0 1.9 2.8 1.9 1.0 2.8 1.6 1.2 3.4 1.9 4.6 6.2 4.9 3.9 7.2 4.7
5.5 2.1 1.6 0.8 2.1 1.6 2.1 2.0 2.8 2.9 3.1 2.9 2.7 3.6 2.1 2.9
2.5 5.2 2.5 3.7 3.6 6.1 5.4 5.6 10.2 10.4 6.9 11.6 12.7 12.5 11.6 11.2
2.2 5.9 2.2 4.5 5.1 8.6 6.9 7.2 4.6 11.6 6.0 11.3 11.7 10.8 12.2 13.1
2.5 3.9 2.5 2.7 3.0 3.8 4.8 3.6 5.6 5.8 9.4 10.1 10.6 12.8 11.7 10.9
hydrophobicity of Fe3 O4 @SiO2 @Flu. Considering these results, the proposed sorbent is a sensitive, efficient, convenient and reliable material for the pre-concentration of trace PAHs.
[31] [32] [33] [34] This work
[11] [12] [13] [15] [14]
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
Ref.
46
n.r. n.r. 35–133 n.r. 96.6–105.7 35–99 88–97 85–94 61.9–119.1 96.6–105.7 2.0–10
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Magnetic solid phase extraction and gas chromatography–mass spectrometrical analysis of sixteen polycyclic aromatic hydrocarbons Ying Cai a , Zhihong Yan b , Manh NguyenVan a,c , Lijia Wang a , Qingyun Cai a,∗ a b c
State Key Laboratory of Chemo/Biosensing & Chemometrics, College of Chemistry & Chemical Engineering, Hunan University, Changsha 410082, China Key Laboratory of Modern Preparation of TCM, Ministry of Education, Jiangxi University of Traditional Chinese Medicine, Nanchang 330004, China Faculty of Chemical Technolology, Hanoi University of Industry, Hanoi, Viet Nam
a r t i c l e
i n f o
Article history: Received 1 April 2015 Received in revised form 8 June 2015 Accepted 10 June 2015 Available online 20 June 2015 Keywords: Fluorenyl functionalized MNPs Polycyclic aromatic hydrocarbons Water sample – interaction
a b s t r a c t Fluorenyl functionalized superparamagnetic core/shell magnetic nanoparticles (MNPs, Fe3 O4 @SiO2 @Flu) were prepared and characterized by transmission electron microscope, X-ray diffraction and infrared spectroscopy. The MNPs having an average diameter of 200 nm were then used as solid-phase extraction sorbent for the determination of 16 priority pollutants polycyclic aromatic hydrocarbons (PAHs) in water samples designated by United States Environmental Protection Agency (U.S. EPA). The main influencing parameters, including sorbent amount, desorption solvent, sample volume and extraction time were optimized. Analyses were performed on gas chromatography–mass spectrometry (GC–MS) using selected ion monitoring (SIM) mode. Method validation proved the feasibility of the developed sorbents for the quantitation of the investigated analytes at trace levels. Limit of detection ranging from 0.5 to 4.0 ng/L were obtained. The repeatability was investigated by evaluating the intra- and inter-day precisions with relative standard deviations (RSDs) lower than 13.1%. Finally, the proposed method was successfully applied for the determination of PAHs in water samples with the recoveries in the range of 96.0–106.7%. © 2015 Elsevier B.V. All rights reserved.
1. Introduction It is well known that water pollution has become an important problem related to public health. Among the pollutants, polycyclic aromatic hydrocarbons (PAHs) have raised global concerns because of their widespread distribution and carcinogenic toxicity [1]. The U.S. EPA has formulated regulations for the monitoring and control of PAHs and has developed methods for their measurement in air, water, food, and other matrixes [2]. The EPA has designated 16 PAHs as priority pollutants [3], and eight of them are considered to be possible carcinogens, namely: benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, dibenzo[a,h]anthracene, indenol[1,2,3-cd]pyrene and benzo[g,h,i]perylene [4]. PAHs in different matrices were characterized and quantified mainly by HPLC or GC-based methods, such as Method 610 which deals with the measurement of these PAHs in wastewater by either HPLC/UV and FLD or gas chromatography with flame ionization detection (GC/FID) [5]. Method 8100 analyzes PAHs by GC/FID [6].
∗ Corresponding author at: State Key Laboratory of Chemo/Biosensing & Chemometrics, College of Chemistry & Chemical Engineering, Hunan University, Changsha 410082, China. Tel.: +86 73188821848. E-mail addresses: [email protected], [email protected] (Q. Cai). http://dx.doi.org/10.1016/j.chroma.2015.06.024 0021-9673/© 2015 Elsevier B.V. All rights reserved.
Methods 525, 625, and 8270 which analyze several PAHs and many other environmental pollutants in drinking water, wastewater, and solid wastes, respectively, by GC/MS [7–9]. Method 8275 which is a thermal extraction coupled with capillary GC/MS procedure for the quantitative determination of targeted PCBs and the 16 EPA priority pollutants PAHs in soils, sludges, and solid wastes [10]. However, the trace content in surface water makes the determination of PAHs a challenging task. Effective extraction and preconcentration techniques prior to instrumental analysis are often required. The widely used pre-treatment method at present for PAHs from aqueous samples are mainly sorbent based techniques, such as solid phase extraction (SPE) [11–14], microsolid-phase extraction (-SPE) [15], and magnetic solid-phase extraction (MSPE) [16–29]. MSPE is a procedure for the preconcentration of target analytes from large volumes by using magnetic or magnetizable sorbents [30], which can be easily separated from sample solutions by an external magnetic field without the need of additional centrifugation or filtration. Moreover, since the sorbent can be dispersed in sample solution, the contact surface between analytes and sorbents is very large, which results in a rapid mass transfer and rapid extraction equilibrium. For the above reasons, MSPE has attracted increasing interests and found wide applications in the field of sample preparations. Several MNPs were synthesized for the extraction of U.S. EPA 16 PAHs, they are C18 functionalized Fe3 O4 nanoparticles
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
(magnetic C18 ) [31], diphenyl functionalization of Fe3 O4 MNPs (Fe3 O4 diphenyl) [32], carbon–ferromagnetic nanocomposite (carbon–Fe3 O4 ) [33] and n-octadecylphosphonic acid modified mesoporous MNPs (OPA/MMNPs) [34]. Generally, – conjugative effect and hydrophobic interaction were contributed to adsorb PAHs. To the best of our knowledge, there is no report about preparation of condensed cyclic structural MNPs, and their application as SPE sorbents for adsorption of PAHs. Also, there is lack of report about the adsorption capacity of the MNPs. In the present study, a new kind of MNPs termed as fluorenyl functionalized magnetic silica nanoparticles (Fe3 O4 @SiO2 @Flu) was developed to extract PAHs from water samples. Fe3 O4 @SiO2 @Flu MNPs were synthesized by coating/modification magnetite Fe3 O4 nanoparticles with silica, subsequently 3-isocyanatopropyltriethoxysilane and finally 9-fluorenylmethyl chloroformate. To our knowledge, this study is the first report with respect to the introduction of fluorenyl to functionalize MNPs for the extraction of PAHs. Due to the condensed cyclic structure, rich -electrons exist on the surface of fluorenyl, which leads to the adsoption of PAHs through the – conjugative effect. Finally, adsorption capacity of Fe3 O4 @SiO2 @Flu was studied, and the sorbents was used for enrichment of PAHs from environmental water samples. 2. Experimental 2.1. Reagents and materials All reagents were procured from major suppliers. Tetraethoxysilane (TEOS, 99%), 3-aminopropyltriethoxysilane (APTES, 99%) and 9-fluorenylmethyl chloroformate were obtained from Adamas Reagent Ltd. (Switzerland). HPLC grade methanol, n-hexane and acetonitrile (99%) were bought from Merck (Darmstadt, Germany). Ferric chloride hexahydrate (FeCl3 ·6H2 O, 99%), ferrous sulfate heptahydrate (FeSO4 ·7H2 O, 99%), ammonia (26%), hydrazine hydrate (99%), isopropanol (99%), triethylamine (99%), toluene (99%) were purchased from Sinopharm Chemical Reagent Co. Ltd. (China). Ultrapure water was used through the experiment which was prepared using a Milli-Q system water purification system (Millipore Inc., USA). 610/8100 PAH standard solution was bought from ANPEL Laboratory Technologies (Shanghai) Inc. The PAH standard solution contains naphthalene (Nap), acenaphthylene (Aceny), acenaphthene (Acen), fluorene (Flu), phenanthrene (Phen), anthracene (Ant), fluoranthene (Fluo), pyrene (Pyr), chrysene (Chr), benzo[a]anthracene (B[a]A), benzo[b]fluoranthene (B[b]F), benzo[k]fluoranthene (B[k]F), benzo[a]pyrene (B[a]P), indeno[1,2,3-cd]pyrene (I[1,2,3-cd]P), dibenzo[a,h]anthracene (D[a,h]A) and benzo[g,h,i]perylene (B[ghi]P), each at 2 mg/mL in benzenen/methylene dichloride (1/1, v/v). The PAH stock solution was prepared in methanol at a concentration of each at 2 g/mL, and kept at 4 ◦ C in darkness. PAH working solutions were prepared by proper dilution of the stock solution. Tap water, river water and waste water were used for real sample investigation. Tap water was taken from our lab tap after flowing for 10 min. River water sample was collected from the Ganjiang River in Nanchang (Jiangxi, China) and collected at 10 cm depth below the water surface. Waste water sample was collected from water treatment plants effluents (Jiangxi, China). Water samples were analyzed within 24 h.
41
(FTIR) were recorded on Vertex 70 (Bruker Optics, Germany). After being dried at 80 ◦ C in a vacuum oven for 12 h, samples were mixed with KBr to fabricate a KBr pellet for FTIR analysis. PAHs were extracted from water samples with the assistance of an ultrasonic instrument KQ-600KDE. The magnetic property was analyzed using a vibrating sample magnetometer (VSM, Model 7410, Lake Shore Cryotronics, Inc., Westerville, OH, USA). Analyses were conducted using an Agilent 7890A GC (Agilent Technologies, USA) system coupled to an Agilent 5975C mass spectrometer (Agilent Technologies, USA). Helium (purity 99.999%) was employed as the carrier gas at a flow rate of 1.8 mL/min. Samples (1 L) were injected in splitless mode. The injector temperature was set at 280 ◦ C and the interface temperature maintained at 280 ◦ C. The chromatographic separation of the PAHs was performed on a 30 m × 0.25 mm × 0.25 m HP5 capillary column (Agilent). The GC oven was initially held at 80 ◦ C for 1 min and increased to 250 ◦ C at 25 ◦ C/min and held for 6 min, then increased to 300 ◦C at 10 ◦ C/min and held for 2 min. Preliminarily, full scan electron ionization data were acquired to determine appropriate masses under the following conditions: ionization energy: 70 eV, mass range: 35–350 amu, scan time: 3 scan/s. PAH standards and samples were finally analyzed in selective ion monitoring (SIM) mode for quantitative determination of the analytes. The masses monitored were as follows: 3–5 min, m/z 128, 127, 102 for naphthalene; 5–6 min, m/z 152, 154 for acenaphthylene and acenaphthene; 6–7 min, m/z 166, 165 for fluorene; 7–8 min, m/z 178 for phenanthrene and anthracene; 8–10 min, m/z 202, 203, 200 for fluoranthene and pyrene; 10–13 min, m/z 228, 226, for chrysene and benzo[a]anthracene; 13–18 min, m/z 252 for benzo[b]fluoranthene, benzo[k]fluoranthene and benzo[a]pyrene; 18–22 min, m/z 276, 278 for indeno[1,2,3cd]pyrene, dibenzo[a,h]anthracene and benzo[g,h,i]perylene. The solvent delay time was 3 min (to bypass the solvent peak). For all the investigated analytes the corresponding ion ratios were used for confirmation purposes. Signal acquisition and data processing were performed using the Agilent MSD Chemstation (G1701EA, Agilent Technologies). 2.3. Preparation of Fe3 O4 @SiO2 @Flu magnetic nanoparticles 2.3.1. Preparation of Fe3 O4 MNPs The Fe3 O4 MNPs were prepared by chemical co-precipitation. Briefly, FeCl3 ·6H2 O (4.0 g) was dissolved in deionized water (30 mL) in a three-necked round bottom flask, followed by addition of hydrazine hydrate (2 mL) and FeSO4 ·7H2 O (10.90 g) to prepare a stock solution. Afterwards, ammonia solution (35 mL 26.5% w/w) was added into the stock solution under vigorous stirring, followed by dropwise addition of ammonia solution (26.5% w/w) until the solution pH reached 9. The solution was stirred at room temperature for 30 min, aged at 80 ◦ C for 60 min, and then cooled to room temperature. The product was magnetically collected, and washed with water, finally vacuum-dried at 60 ◦ C for 12 h.
2.2. Instrumentations and analytical conditions
2.3.2. Encapsulation of the MNPs with silica (Fe3 O4 @SiO2 ) Nanoparticles (1 g) were dispersed in a mixture of 2-propanol (100 mL) and ultrapure water (8 mL), sonicated for 15 min, then added with ammonia solution (10 mL, 26.5% w/w) and TEOS (8 mL) sequentially. After stirring the mixture for 12 h at 45 ◦ C, the MNPs were collected by a magnet, washed with water and ethanol respectively, and vacuum-dried at 60 ◦ C for 12 h. Through this procedure, the ferro-MNPs were encapsulated with a layer of mesoporous silica.
The size and morphological characterization of the particles were observed by transmission electron microscopy (TEM, JEM2100F, JEOL Co., Tokyo, Japan). Fourier transform infrared spectra
2.3.3. Preparation of fluorenyl coated MNPs (Fe3 O4 @SiO2 @Flu) The preparation scheme of Fe3 O4 @SiO2 @ Flu is depicted in Fig. 1. Firstly, the above Fe3 O4 @SiO2 nanoparticles (3 g) were
42
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
Fig. 1. Preparation scheme of Fe3 O4 @SiO2 @Flu.
suspended in toluene (60 mL), then triethylamine (1 mL) and APTES (6 mL) were added under argon atmosphere. The suspension was mechanically stirred and refluxed at 110 ◦ C for 24 h. The reaction was stopped and the APTES MNPs were cooled to room temperature, washed with toluene, acetone and ethanol respectively, finally dried under vacuum at 60 ◦ C. Next, the APTES MNPs (3 g) were mixed with toluene (40 mL) and triethylamine (1 mL) under argon atmosphere, after stirring for 30 min in ice-bath, 9-fluorenylmethyl chloroformate (3.8 g) was added. The mixture was refluxed for 4 h at room temperature and then 12 h at 110 ◦ C. The reaction was stopped and the as-prepared Fe3 O4 @SiO2 @Flu MNPs were cooled to room temperature, washed with ethanol and water/ethanol (1/1 v/v). The resulting product was dried under vacuum at 60 ◦ C for 12 h.
gurantee a complete separation of the sorbents from solution even analyte was loaded. FTIR spectra were acquired for Fe3 O4 @SiO2 and Fe3 O4 @SiO2 @Flu MNPs between 4000 and 400 cm−1 (Fig. S1). As seen in the figure, the appearance of characteristic peaks
2.4. Extraction procedure 40 mg Fe3 O4 @SiO2 @Flu magnetic sorbents was added into 100 mL PAHs-spiked water sample. After sonicating the mixture for 30 s to accelerate the dispersion of the sorbents in solution, the MNPs were isolated rapidly from the solution by applying an external magnetic field via a supermagnet (Nd–Fe–B). An aliquot of 1.0 mL n-hexane was introduced to desorb the captured PAHs from the nanoparticles by sonication for 15 s. Finally, the extract was separated from the suspension with a supermagnet and filtered through a 0.22 m polytetrafluoroethylene membrane syringe filter, 1 L of the filtrate was injected into GC–MS system for analysis. 3. Results and discussion 3.1. Characterization of Fe3 O4 @SiO2 @Flu MNPs The prepared MNPs were characterized with TEM, VSM and FTIR. TEM was used to obtain structure information of the Fe3 O4 , Fe3 O4 @SiO2 and Fe3 O4 @SiO2 @Flu MNPs. As shown in Fig. 2, the synthesized Fe3 O4 MNPs are quasi-spherical in shape at an average size of 20 nm in diameter. The TEM image of Fe3 O4 @SiO2 MNPs shows that a uniform gray silica shell is coated on the Fe3 O4 MNPs with an average diameter of 100 nm (Fig. 2b). The silica shell provides abundant silanol groups for further chemical modification. The Fe3 O4 @SiO2 @Flu MNPs (Fig. 2c) are larger than Fe3 O4 @SiO2 MNPs with an average diameter of 200 nm, confirming the immobilization of flu shell. Sufficient magnetic properties are important for MSPE. The magnetic properties of the prepared nanoparticles were investigated with a VSM. Fig. 3 shows the magnetization curves of Fe3 O4 , Fe3 O4 @SiO2 and Fe3 O4 @SiO2 @Flu at 300 K, giving the magnetic saturation values of 71.6, 30.7 and 27.48 emu/g, respectively. The coating results in decreases in the magnetic saturation values. The magnetic saturation value of Fe3 O4 @SiO2 @Flu sorbents can
Fig. 2. TEM images Fe3 O4 (a), Fe3 O4 @SiO2 (b) and Fe3 O4 @SiO2 @Flu (c).
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
43
3.2. Optimization of extraction conditions
Fig. 3. Magnetization curves of Fe3 O4 , Fe3 O4 @SiO2 and Fe3 O4 @SiO2 @Flu.
at 1089 and 475 cm−1 are attributed to the Si O Si stretching vibration and Si O symmetric stretching vibration, respectively. After modification with fluorenyl, there displays a prominent peak at 768 cm−1 , which is characteristic of H on fluorene ring. In addition, a band at 1768 cm−1 corresponding to the bending vibration of skeleton of aromatic ring appears. These results confirmed the success surface modification of the MNPs.
The effects of several parameters on extraction efficiency were studied, such as the sorbent amount, type of desorption solvent, water sample volume and extraction time. The influence of all these parameters was evaluated in terms of peak areas. The optimization experiments were conducted using water sample spiked with 200 ng/L of each PAH. Each experiment was performed in triplicate. The effect of amount of Fe3 O4 @SiO2 @Flu MNPs was investigated with the amount varying from 10 to 60 mg. Fig. 4a demonstrates that the peak areas of almost all the 16 PAHs increase continuously with the increase of the sorbent amount from 10 to 40 mg. Further increasing the sorbent amount over 40 mg results in no obvious increase of the peak area. These results indicated that 40 mg of sorbents are sufficient to extract PAHs. Excess sorbent may retain analytes and result in a decrease in peak area. So, 40 mg of sorbent was used for the following experiments. An appropriate type of elution solvent is fundamental for extraction efficiency by Fe3 O4 @SiO2 @Flu MNPs. Four types of solvents were selected as desorption solvent, including acetone, dichloromethane, n-hexane and acetonitrile. As shown in Fig. 4b, dichloromethane and n-hexane yields higher analytical signals than acetone and acetonitrile. For most PAHs, n-hexane was the best desorption solvent. The effect of sample volume on extraction efficiency of PAHs was investigated with the volume changing from 100 to 2000 mL. As shown in Fig. 4c, the signals of all the PAHs declined significantly with increasing the sample volume from 100 to 2000 mL.
Fig. 4. Effect of (a) sorbent amount, (b) type of desorption solvent, (c) sample volume, (d) extraction time on the peak areas of PAHs. Amount of sorbent, 40 mg; sample volume: 100 mL; desorption solvent, n-hexane (1.0 mL); extraction time, 5 min.
44
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
Fig. 5. Adsorption isotherms for PAHs on Fe3 O4 @SiO2 @Flu MNPs.
This behavior occurs possibly because that when increasing the sample volume, especially for 500–2000 mL, the magnetic sorbents would be dispersed more widely, and some sorbents will loss during the recovery process. So, 100 mL was selected as sample volume. Fig. 4d shows the effect of extraction time on the extraction efficiency. One can see that there is no obvious variation of the peak area when increasing the extraction time from 2 to 20 min, and when extraction time was increased to 30 min, the analytical signals decreased a little. With this clarification, 5 min was sufficient to achieve satisfactory extraction efficiency for most PAHs. Therefore, extraction time of 5 min was applied in the following MSPE procedure. Based on the above discussion, the optimal conditions of MSPE with Fe3 O4 @SiO2 @Flu sorbent for PAHs are: 40 mg Fe3 O4 @SiO2 @Flu MNPs, 100 mL solution volume, 5 min extraction time, and n-hexane as the desorption solvent. 3.3. Stability of Fe3 O4 @SiO2 @Flu The stability of the Fe3 O4 @SiO2 @Flu sorbent was evaluated based on the reproducibility of extraction efficiency using different batches of magnetic sorbent and the reusability of the sorbent. The reproducibility of the extraction was investigated using three different batches of Fe3 O4 @SiO2 @Flu MNPs with water sample spiked with 200 ng/L of each PAH. The peak area and RSD values of 16 PAHs extracted with the three different batches of magnetic sorbents are listed in Table S1. The RSD values are lower than 8.7%, indicating a good reproducibility. The regeneration of the magnetic sorbent was investigated with a random batch of Fe3 O4 @SiO2 @Flu as sorbent. The used Fe3 O4 @SiO2 @Flu (40 mg) was regenerated by rinsing it with 1 mL of n-hexane twice to make sure that no residual PAHs were remained. The regenerated sorbent was applied in MSPE of water samples which were spiked with 200 ng/L of each PAHs. As shown in Fig. S2, after 10 times of recycling, there is no significant change in the peak areas for each analyte extracted by the Fe3 O4 @SiO2 @Flu MNPs, indicating that the Fe3 O4 @SiO2 @Flu sorbents are stable and durable during MSPE procedure with good reusability. 3.4. Adsorption capacity In adsorption isotherm studies, solutions with different initial concentrations (0.05–5.0 g/mL) were added. Fe3 O4 @SiO2 @Flu MNPs (40 mg) and PAHs solution (100 mL) were mixed in a flask, then MNPs were separated by applying an external magnetic field
after adsorption equilibrium. The equilibrium concentration of PAHs was determined by GC–MS system. The condition for the determination was the same as Section 2.2. As shown in Fig. 5, the calculated maximal adsorption capacities of Nap, Aceny, Acen, Flu, Phen, Ant, Fluo, Pyr, Chr, B[a]A, B[b]F, B[k]F, B[a]P, I[1,2,3-cd]P, D[a,h]A and B[ghi]P on Fe3 O4 @SiO2 @Flu were 325, 411, 252, 258, 230, 268, 432, 436, 335, 393, 297, 349, 215, 227, 338 and 211 ng/mg, respectively. 3.5. Analytical figures of merit Under the optimal conditions mentioned above, a series of experiments with regard to the linearity (linear ranges and coefficient of determination), limit of detection (LOD), limit of quantification (LOQ), precision and enrichment factor (EF) were performed to validate the proposed method. In order to investigate the possible matrix effect on determination, the linearity of the proposed method was estimated by analyzing Ganjiang river water samples spiked with various concentrations of PAHs over a range of 10 and 1000 ng/L (10, 50, 100, 250, 500, 1000 ng/L). The calibration curves were obtained by plotting the mean peak area versus sample concentration. The results are listed in Table 1. All the analytes show good linearity with coefficient of determination ranging from 0.9900 to 0.9998. The LOD and LOQ were calculated based on the standard deviation of the response and the slope of calibration curve [35] from the following equations: LOD = 3.3Sb/m (1) LOQ = 10Sb/m (2) where Sb is the standard deviation of the y-intercept of regression line and m is the slope of the calibration curve. The LOD values were from 0.5 (pyr) to 4.0 ng/L (I[1,2,3-cd]P), and the LOQ values were from 1.7 to 13.3 ng/L. Accuracy was calculated in terms of recovery rate (R%) on real water samples as follows: R% = (c 1 /c 2 ) × 100% where c1 is the measured concentration and c2 is the concentration calculated from the quantity spiked into the sample. Relative extraction recoveries ranged from 96.6 to 105.7%. The EF of the method was calculated using the following equation [36,37]: EF = (V s /V e ) × R% where Vs is the sample volume, Ve is the elution volume, and R% is percent recovery. The EF were determined to be 96.6, 103.4,
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
45
Table 1 MSPE–GC/MS analytical figures of merit for the16 PAHs. Compound
Linearity range (ng/L)
R2
Calib. curvea (n = 3)
LOD (ng/L)
LOQ (ng/L)
Recoveries (%)
RSD (%) (n = 3)
Enrichment factor
Nap Aceny Acen Flu Phen Ant Fluo Pyr B[a]A Chr B[b]F B[k]F B[a]P I[1,2,3-cd]P DB[a,h]A B[ghi]P
10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000 10–1000
0.9992 0.9950 0.9962 0.9941 0.9897 0.9900 0.9904 0.9818 0.9989 0.9983 0.9997 0.9998 0.9998 0.9990 0.9988 0.9989
y = 1017.6x + 8517.1 y = 149.91x + 17825 y = 195.72x + 19164 y = 226.57x + 24799 y = 412.26x + 57672 y = 393.05x + 15494 y = 789.74x + 77017 y = 789.37x + 216948 y = 710.28x + 11930 y = 695.18x + 18077 y = 481.45x − 2913.3 y = 435.14x + 328.57 y = 590.41x − 2441.6 y = 287.43x − 4630 y = 339.03x − 5536.1 y = 507.83x + 1575.4
0.8 1.9 1.6 1.3 1.4 1.1 0.6 0.5 1.4 1.7 2.5 2.9 4.0 3.4 2.9 3.1
2.7 6.3 5.3 4.3 4.7 3.7 2.0 1.7 4.7 5.7 8.3 9.7 13.3 11.3 9.7 10.3
96.6 103.4 101.5 100.8 105.3 103.8 105.7 101.0 99.0 98.6 100.5 100.9 103.6 99.9 103.8 99.6
3.5 2.7 2.1 1.8 1.9 2.0 1.5 2.2 1.9 1.4 2.7 2.7 4.9 5.0 6.2 6.9
96.6 103.4 101.5 100.8 105.3 103.8 105.7 101.0 99.0 98.6 100.5 100.9 103.6 99.9 103.8 99.6
a
y is peak area and x is compound concentration (ng/L).
101.5,100.8, 105.3, 103.8, 105.7, 101.0, 99.0, 98.6, 100.9, 100.5, 103.6, 99.9, 103.8 and 99.6 for Nap, Aceny, Acen, Flu, Phen, Ant, Fluo, Pyr, B[a]A, Chr, B[b]F, B[k]F, B[a]P, I[1,2,3-cd]P, DB[a,h]A and B[ghi]P, respectively. Repeatability or intra-day precision was evaluated by spiking Ganjiang River samples with appropriate amounts of PAHs at three different levels (50, 100, 400 ng/L). Six replicates were performed and analyzed on the same day. In order to determine reproducibility or inter-day precision, samples that were spiked with the same amount of PAHs, were analyzed in six consecutive days at three replicate levels. The intra and inter-day precisions were expressed as the percentage relative standard deviation (RSD%) (Table 2). The intra-day precision was from 0.8 to 7.5% and the inter-day precision for the PAHs were below 13.1%, illustrating the good repeatability achieved by the suggested procedure. These results imply that the proposed method can be applied to the analysis of real samples containing PAHs at trace level.
sample, benzo[a]pyrene and dibenzo[a,h]anthracene were detected except naphthalene, acenaphthylene and anthracene. The relative recoveries of PAHs at two concentration levels are in the range of 96.0–106.7%, with RSDs within 6.1%. These results imply that the established method can be applied to the analysis of PAHs at trace level in real samples. 3.7. Comparison with other sorbents The proposed magnetic sorbent is compared with other solidbased sorbents [11–15,31–34] in Table 3. The comparison is performed on the basis of some operational (sorbent amount, extraction time and sample volume) and analytical properties (GC–MS elution time, LOD, recovery and precision). Firstly, for comparison with traditional SPE, the amount of magnetic Fe3 O4 @SiO2 @Flu sorbent was much less than SPE sorbents such as C18 [11] and carbon nanotubes [13], and the MSPE procedure was much quicker than SPE [12–15]. Secondly, for comparison with other magnetic sorbents, the Fe3 O4 @SiO2 @Flu shows similar recoveries to the most of other sorbents, and the LODs were comparable to others. It needs less amount of sorbent than magnetic C18 [31] and OPA/MMNPs [34], but more amount of sorbent than Fe3 O4 diphenyl [32] and carbon-Fe3 O4 Fe3 O4 /C [33]. Finally, our GC–MS elution time was the shortest. In addition, the abundant electrons of naphthyl provide potent – stacking interactions with PAHs, contributing to the selectivity to PAHs. The adsorption equilibrium in the process of extraction can be quickly achieved due to the good
3.6. Analysis of real water samples To further evaluate the method applicability, Ganjiang river water, tap water and waste water samples were analyzed. Unspiked water samples and water samples spiked with 16 PAHs at two concentration levels (50 and 100 ng/L) were analyzed by the proposed method (n = 3). The results are listed in Table S2. Naphthalene, acenaphthylene and anthracene were found in both tap water and Ganjiang river water samples, and in waste water Table 2 The precision of MSPE method. Compound
Nap Aceny Acen Flu Phen Ant Fluo Pyr B[a]A Chr B[b]F B[k]F B[a]P I[1,2,3-cd]P DB[a,h]A B[ghi]P
Intra-day precision (RSD%, n = 6)
Inter-day precision (RSD%, n = 6)
50 ng/L
100 ng/L
400 ng/L
50 ng/L
100 ng/L
400 ng/L
4.3 1.2 3.2 1.3 1.6 1.6 2.3 1.8 2.1 2.4 6.1 1.9 4.5 5.1 7.4 7.5
2.0 1.9 2.8 1.9 1.0 2.8 1.6 1.2 3.4 1.9 4.6 6.2 4.9 3.9 7.2 4.7
5.5 2.1 1.6 0.8 2.1 1.6 2.1 2.0 2.8 2.9 3.1 2.9 2.7 3.6 2.1 2.9
2.5 5.2 2.5 3.7 3.6 6.1 5.4 5.6 10.2 10.4 6.9 11.6 12.7 12.5 11.6 11.2
2.2 5.9 2.2 4.5 5.1 8.6 6.9 7.2 4.6 11.6 6.0 11.3 11.7 10.8 12.2 13.1
2.5 3.9 2.5 2.7 3.0 3.8 4.8 3.6 5.6 5.8 9.4 10.1 10.6 12.8 11.7 10.9
hydrophobicity of Fe3 O4 @SiO2 @Flu. Considering these results, the proposed sorbent is a sensitive, efficient, convenient and reliable material for the pre-concentration of trace PAHs.
[31] [32] [33] [34] This work
[11] [12] [13] [15] [14]
Y. Cai et al. / J. Chromatogr. A 1406 (2015) 40–47
Ref.
46
n.r. n.r. 35–133 n.r. 96.6–105.7 35–99 88–97 85–94 61.9–119.1 96.6–105.7 2.0–10
