- Email: [email protected]
Determination of microcystins in environmental water samples with ionic liquid magnetic graphene
Ecotoxicology and Environmental Safety 176 (2019) 20–26
Contents lists available at ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Determination of microcystins in environmental water samples with ionic liquid magnetic graphene
T
Xiaoyan Liua, Shiqian Gaoa,b, Xinyue Lia, Hui Wanga, Xiaowen Jic, Zhanen Zhanga,b,∗ a
School of Environmental Science and Engineering, Suzhou University of Science and Technology, No. 1 Kerui Road, Suzhou, 215009, PR China Jiangsu Key Laboratory of Environmental Science and Engineering, Suzhou University of Science and Technology, No. 1 Kerui Road, Suzhou, 215009, PR China c State Key Laboratory of Pollution Control and Resource Reuse, Center for Hydrosciences Research, School of the Environment, Nanjing University, Nanjing, 210093, PR China b
ARTICLE INFO
ABSTRACT
Keywords: Microcystin Ionic liquid magnetic graphene Magnetic solid phase extraction Ultra high performance liquid chromatography tandem mass spectrometry
Microcystins is a class of monocyclic of heptapeptides with many different isomerides. It has become potential hazardous material in water environment for its toxic, distribution and stability. This project worked on a method for determination of trace microcystin (MC-LR and MC-RR) in environmental waters. The ionic liquid magnetic graphene (IL@MG) was prepared and applied to the concentration and determination of microcystins, based on magnetic solid phase extraction (MSPE), and coupled with ultra high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). The ionic liquid magnetic graphene was prepared by coprecipitatial synthesis and characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM), fourier transform infrared spectroscopy (FT-IR), specific surface area (BET), pore size distribution (BJH) and magnetic hysteresis loop. The experimental parameters of magnetic solid phase extraction, including amount of IL@MG, pH, extraction time and elution solvent were investigated by a univariate method and orthogonal screening. The method showed good linearity in the range of 0.01–10.0 g/L and 0.005–10.0 μg/L for MC-LR and MC-RR, when the pH of water samples was 4.00 and 10.0 mg adsorbents were used to extract targets for 18 min. The lowest detection limit was 0.414 ng/L and 0.216 ng/L for MC-LR and MC-RR respectively. The recoveries of the microcystins were in the range of 83.6–100.9%, and the relative standard deviation was less than 7.59%. The trace amount of MC-LR (0.020 μg/L) and MC-RR (0.003 μg/L and 0.021 μg/L) was detected in actural water samples. Attributed to its simple operator, low detection limit and high sensitivity, this method could be used for the detection of trace microcystins in water samples.
1. Introduction A large number of pollutants containing nitrogen and phosphorus entered the water through surface runoff and sewage pipe network, resulting in eutrophication of water bodies. With the suitable temperature and light intensity, the rapid growth of algae substances lead to water bloom, which causes the release of harmful substances such as microcystins (MCs) (Falconer, 1999). MCs had a certain toxic effect on the spleen and kidney, and it was also a tumor promoter (Kaloudis et al., 2013). In addition, MCs can affect post-embryonic development (Falconer et al., 1999), posing a major threat to water quality and human health. More than 90 kinds of microcystins with variable toxicity had been identified (Sypabekova et al., 2017). Of all MCs, MC-LR was the most common toxic variant which had multiple organ toxicity, genotoxicity and carcinogencity (Neilan et al., 2013). Both the World
Health Organization (WHO) and China's drinking water hygiene regulations stipulated for the minimal permissible concentration of MCs in drinking water (< 1.0 μg/L) (Zhang et al., 2017b). Therefore, it is important to choose the absorbent to establish the proper detection method. At present, the detection methods of microcystins in environmental water are published as the follows. Highly performed liquid chromatography (HPLC) (Pietsch et al., 2001), high performance liquid chromatography tandem mass spectrometry (HPLC-MS/MS) (Liu et al., 2011; Shamsollahi et al., 2014), thin layer chromatography (TLC) (Xu et al., 2013), biological treatment (Meisen et al., 2009), enzyme-linked immunosorbent assays (ELISAs) (Taghdisi et al., 2017), protein phosphatase inhibition assay (PPIA) (Zhang et al., 2017a) and lateral flow immunoassays (LFIAs) (Sedda et al., 2016) has been used to detect MCs. Among these methods, HPLC-MS/MS was a powerful tool that could be
∗ Corresponding author. School of Environmental Science and Engineering, Suzhou University of Science and Technology, No. 1 Kerui Road, Suzhou 215009, PR China. Tel.: +86 13912606902; fax: +86 512-68093060. E-mail address: [email protected] (Z. Zhang).
https://doi.org/10.1016/j.ecoenv.2019.03.063 Received 26 December 2018; Received in revised form 14 March 2019; Accepted 14 March 2019 0147-6513/ © 2019 Elsevier Inc. All rights reserved.
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
confidently used to identify specific variants of MCs, it not only has good reproducibility and low sensitivity, but also can analyze isomers. In addition, it has the advantages of allowing quantitative and qualitative analysis of toxins at the same time with faster speed and more sensitive limits of detection (LOD). The pretreatment method of microcystins is an important process. Solid phase extraction (SPE) (Devasurendra et al., 2018; Li et al., 2017; Wei et al., 2015), accelerated solvent extraction - solid-phase extraction (ASE-SPE) (Hoff and Pizzolato, 2018), online solid-phase extraction (online-SPE) (Vudathala et al., 2017), stir bar sorptive extraction (SBSE) (Ortiz et al., 2017), matrix solid-phase dispersion (MSPD) (Cui et al., 2018) were used for the extraction od MCs before. These pretreatment methods are cumbersome and take a long time to operate. In recent years, a new method - magnetic solid phase extraction (MSPE) (Qian et al., 2017) technology has emerged. This is a new type of pretreatment method that combines solid phase extraction with magnetic and magnetically modified materials. It has convenient separation process, simple operation, and easy separation in complex matrix systems. The adsorbent materials which can be used for magnetic solid phase extraction mainly included magnetic carbon materials (Zhang and Kong, 2011), magnetic molecular imprinted materials (Yan et al., 2013), metal organic framework compounds (Chen et al., 2013), ionic liquids (Zhang et al., 2010b) and magnetic polymer materials (Meng et al., 2011). In this work, ionic liquids (IL) and Fe3O4 were simultaneously introduced to load on graphene. IL refers to a liquid composed entirely by ions. It is usually composed by organic cations and inorganic anions. It has the advantages of non-volatility, high solubility, low vapor pressure, and good thermal stability (Okabe et al., 2017). Simultaneously, graphene has ideal physical and chemical properties, high specific surface area (Zhang et al., 2010a), good stability, high electron mobility (Xiao et al., 2012) and good thermal conductivity (Chang and Wu, 2013). The preparation process of graphene was simple, and the promotion and application of graphene was easy in production (Nardecchia et al., 2013). Modifying graphene by using ionic liquids can reduce the occurrence of agglomeration effects, improve the hydrophilicity of graphene, and solve the shortcomings of uneven distribution in water (Nardecchia et al., 2013). By combining and magnetizing the two materials, a composite nanomaterial with magnetic properties can be formed. In this experiment, we had attempted to explore the possibility of simultaneous extraction and preconcentration of trace MC-LR and MCRR in water samples. Then a new type of magnetic composite material had been prepared by loading ionic liquid and Fe3O4 onto graphene. This synthetic material (IL@MG) was a kind of adsorbent that combined the advantages of high adsorption efficiency, environmental protection and high sensitivity. IL@MG was used as an adsorbent for magnetic solid phase extraction (MSPE), and in combination and in combination with UHPLC-MS/MS to detect microcystins in environmental waters.
heptahydrate of analytically pure were purchased from Sinopharm Chemical Reagent corporation. Ammonia was analytically pure and purchased from Wuxi Prospect Chemical Reagent Company. Anhydrous ethanol was analytically pure and purchased from Wuxi Jingke Chemical Company. The experimental water was ultrapure water made by Milli-Q at 18.2 ΩM (Millipore, USA). Analytical standards of MC-LR and MC-RR were purchased from Standford Analytical Chemicals (USA) and its purity was above 98%. It was prepared by adding pure methanol to a 100.0 μg/mL single standard stock solution and stored at −20 °C. The above single standard stock solution was then diluted to a working standard solution of 1.0 mg/L. 2.2. Experimental methods 2.2.1. Synthesis of IL@MG IL@MG was synthesized by coprecipitatial synthesis method. It was prepared according to the chemical coprecipitation reaction when mixing Fe3+, Fe2+ and graphene under alkaline conditions (Zhu et al., 2014). The accurate weighing 0.0824 g graphene was dispersed in 100 mL ultrapure water, then added 1-butyl-3-methylimidazolium bromide (20.0 g) containing 0.008 mol (2.69 g) 1-butyl-3-methylimidazolium tetrachloride. The mixture was stirred with a glass rod to dissolve it sufficiently, and ultrasonically dispersed for 2 h by KQ300VDE type three-frequency digital ultrasonic cleaner (Kunshan Ultrasonic Instrument Company, China). The above materials were transferred to a 250 mL four-necked flask (Nanjing Pengcheng Chemical Glass Instrument, China), and 0.004 mol (1.11 g) of FeSO4·7H2O was added under nitrogen atmosphere and high-speed stirring process by using DC-12 nitrogen blower (Shanghai Anpu Experimental Technology Company, China) and JJ-1 Precision Forced Electric Mixer (Jintan Analytical Instrument Company, China.). Dropwise slowly added 10 mL of 25% aqueous ammonia during the stirring. It was then stirred at room temperature for 3 h. The precipitate was separated under an external magnetic field. Finally, the materials were alternately washed with anhydrous ethanol and deionized water for three times. It was dried under vacuum circumstance at 60 °C for 12 h by VD115 Vacuum Drying Chamber (Binder, Germany). After removal, it was ground and obtained 1.0301 g of IL@MG powder, and stored after grinding. 2.2.2. Magnetic solid phase extraction Added 100 mL water sample to a 250 mL stoppered conical flask (Nanjing Pengcheng Chemical Glass Instrument, China). Glacial acetic acid was added to adjust the pH to 4.00 by using PHS-3C pH meter (Shanghai Yidian Scientific Instrument Company, China). Accurately weighed 10 mg IL@MG into the above conical flask. The adsorption was carried out for 18 min by using HY-2 variable speed multi-purpose oscillator (Instrument Company, China.). After the magnetic phase separation, the supernatant was decanted and added 1.8 mL of methanol containing 10% ammonia water. After desorbed by using SK-1 Rapid Mixer (Guohua Instrument Company, China) for 1.5 min, using a 0.22 μm organic filter (Haining Dacheng Filter Equipment Company, China) to remove elution solvent in a 10 mL centrifuge tube, repeated the desorption process. Finally, the 3.6 mL of the elution solvent was evaporated by DC-12 nitrogen blower (Shanghai Anpu Experimental Technology Company, China.). After the initial mobile phase was reconstituted to 1 mL, the analysis was analyzed by using UHPLC-MS/MS.
2. Methods and materials 2.1. Instruments and reagents High-quality graphene used in the experiment has a diameter of 0.5–5 μm and a thickness of 0.8–1.2 nm. Its purity was 99%, and the single layer rate is greater than 96%. It was purchased from Xianfeng Nanotechnology corporation (Nanjing, China). The purity of 1-butyl-3methylimidazolium tetrachloride and 1-butyl-3-methylimidazolium bromide (Shanghai Chengjie chemistry, China) are both 99%. FeSO4·7H2O were purchased from Sinopharm (Shanghai, China). Methanol and acetonitrile of chromatographically pure were purchased from Tedia Company of the United States. Formic acid of chromatographically pure was purchased from Aladdin. Methanol, acetone, acetonitrile, ethanol, formic acid, glacial acetic acid and Ferrous sulfate
2.2.3. Chromatographic conditions Analysis experiment was carried out on a TSQ Quantum Access MAX mas spectrometer (Thermo Fisher, USA) equipped with an electrospray ionization (ESI) source and coupled with HPLC System (Dionex, USA). Column: Zorbax Eclipse XDB-C18 reversed phase column (50 mm × 3 mm, 1.8 μm); column temperature: 40 °C; mobile phase: methanol (A) and pure water containing one thousandth of formic acid 21
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
Fig. 1. SEM of IL@MG (a,b), TEM of IL@MG (c,d).
(B). Used gradient elution for separation of MC-LR and MC-RR. The gradient elution procedures were as follows: 0–3min: 5%A-95% A; 3–4.5min, 95% A-100% A; 4.5–5min, 100% A; 5.0–5.1min: 100% A5%A; 5.1–8min: 5% A; flow rate: 0.4 mL/min; injection volume: 10 μL.
material was mesoporous composite. According to Fig. 2 (right), the magnetic hysteresis loops of IL@MG showed almost zero coercivity and remanence, which the maximum saturation magnetization was 34.2 emu/g. The IL@MG loaded with the target analytes could be separated readily from sample medium due to the super-magntism and large saturation magnetization. The data indicated that IL and Fe3O4 were successfully loaded onto the graphene material during the process of synthesis. The ionic liquid possessed by the outer surface of the nano material enabled the experiment to be preferably dissolved in water and better dispersed in the water sample to adsorb the target, thereby making the adsorption effect more ideal.
2.2.4. Mass spectrometry conditions Ion source: electrospray ionization source (ESI+); scan mode: positive ion scan; detection mode: multiple reaction monitoring (MRM); ion transfer tube temperature:350 °C; spray temperature:300 °C; spray voltage:3500 V; sheath gas Pressure:35 psi; Auxiliary gas pressure:15 psi; Collision gas pressure:1.5 mTorr; The parameters of the MRM mode were shown in Table S1 (supporting information).
3.2. Optimization of magnetic solid phase extraction conditions
3. Results and discussion
3.2.1. Optimization of material amount MCs reached extremely high concentrations in aquatic waters, especially at the end of the vegetation season when cyanobacterial blooms collapsed. Both plankton and benthic microorganisms can participate in the adsorption and degradation of high MCs, so the selection of adsorbents is very important for the removal of microcystins. The experiment weighed 2.0–12.0 mg of IL@MG put into the spiked solution, which contains 100 mL 1.0 μg/L MCs. This process was used to investigate the effect of the amount of material on the extraction efficiency of MCs. Fig. 3 showed that the actual detected concentration of MCs increased with the amount of material increasing from 2.0 mg to 10.0 mg, and then the actual detected concentration decreased with the increasing amount of material. It was indicated that 10.0 mg IL@MG had an ideal enrichment ability for MC-LR (actual detected concentration was 0.95 μg/L) and MC-RR (actual detected concentration was 0.95 μg/L). However, when the amount of adsorbent continued to increase, it would hinder the effective elution of the target. It could be explained that although the total adsorption capacity of the adsorbent would increase, the unit adsorption capacity would decrease along with it. In the limited adsorption time, the excess adsorption could not reach adsorption equilibration, which would lead to a decrease in recovery. Therefore, this study selected 10.0 mg IL@MG as the magnetic solid phase extraction adsorbent.
3.1. Surface topography analysis of materials Scanning electron microscopy (SEM), transmission electron microscopy (TEM) and fourier transform infrared spectroscopy (FT-IR) were used to characterize IL@MG. Pictures of SEM and TEM were showed in Fig. 1. The surface wrinkles of the material were obvious, indicating that the material had a large specific surface area. IL and Fe3O4 were mostly attached to the surface of the graphene sheet. Under the applied magnetic field, IL@MG can be effectively separated from the matrix. As seen in Fig. S1 (supporting information), the characteristic peaks of the materials were given by the infrared spectrum. The peak near 3400 cm−1 is the -OH characteristic absorption peak of the adsorbed water molecule. And the absorption peaks in the range of 2895–2929 cm−1 is C-H stretching vibration peaks on the imidazole ring. The absorption peak of 1625 cm−1 is the absorption peak caused by the stretching vibration of C=O in the carboxyl group. The peak of 1389 cm−1 is the stretching vibration absorption peak of COO−. The absorption peak at 1124 cm−1 belongs to the vibration absorption peak of C-Br from ionic liquid and 588 cm−1 belongs to the Fe-O bending vibration absorption peak from Fe3O4. In addition, specific surface area (BET), pore size distribution (BJH) and magnetic hysteresis loop were used to characterize the prepared adsorbent. As shown in Fig. 2(left), the BET surface area and BJH volume of pores were 123.64 m2/g and 0.375 cm3/g, indicating a large surface area and relatively high porosity. Moreover, the pore size was 10.67 nm, which proved that the
3.2.2. Optimization of pH and ionic strength of water samples The pH of the water sample affects the chemical properties of the 22
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
Fig. 2. BET and BJH of IL@MG (left) The magnetic hysteresis loop of IL@MG (right).
analytes and the binding sites on the surface of the adsorbent. In this experiment, the pH of the solution was adjusted to 2.00–12.00 by glacial acetic acid and aqueous ammonia. When the pH was 4.00, the actual detected concentration of MC-RR (0.94 μg/L) and MC-LR (1.0 μg/L) reached the highest level (Fig. 4). Due to the good acid and alkali resistance of IL@MG (Nardecchia et al., 2013), the structure of the adsorbent was kept stable in acidbased solution without decomposition and destruction (Okabe et al., 2017). When the pH was small, the microcapsule molecules with positive, negative and uncharged were simultaneously presented, and the total molecules was electrically neutral. With the increasing of pH (pH > 2.1), the negatively microcystin molecules gradually added. When the pH was 4.0, the negatively microcystin molecules were about 100%. At this time, the electrostatic attraction made IL@MG show strong extraction ability to microcystin molecules, which was in completely agreement with the experimental results. As the pH continued increasing, the number of positive charges on the surface of the adsorbent gradually decreased, and electrostatic repulsion with the negatively charged microcystins molecules resulted in a decrease in adsorption performance. Therefore, in the subsequent measurement of the actual water sample, the pH of the water sample was adjusted to 4.00. The higher ionic strength of the water sample is, the lower solubility of organic matter has in water. In this experiment, the effect of ionic strength on extraction efficiency was investigated by adding different concentrations (0–20%, w/v) of NaCl to the aqueous solutions. The results showed that the ionic strength of the water sample did not affect the extraction efficiency of IL@MG for the two MCs. Therefore, it was not necessary to adjust the ionic strength of the solution (as the data not shown).
Fig. 3. Effect of material dosage on extraction efficiency (Experimental conditions: pH = 4.00; Extraction time: 20 min; Eluent type: methanol (containing 1‰ formic acid); Eluent dosage: 3 mL; Elution time: 3min) (n = 3).
3.2.3. Optimization of extraction time The alternative extraction time was set as 0, 5, 10, 15, 20 and 30 min after the start of experiment respectively. As shown in Fig. 5, when the extraction time was 20 min, both MC-RR and MC-LR reached the adsorption equilibrium, and the actual detected concentration (MCRR was 1.06 μg/L and MC-LR was 0.95 μg/L) were both at the maximum. The adsorption effect was at its best value. When the extraction time was increased from 20 min to 30 min, the actual detected concentration of MCs decreased slightly. This was due to the long extraction time leading to the microcystins adsorbed by the adsorbents dissolved again in the solution (Pavagadhi et al., 2011), resulting in a decrease of actual detected concentration. Therefore, 20.0 min was selected as the extraction time of subsequent experimental.
Fig. 4. Effect of pH on extraction efficiency (Experimental conditions: material dosage：10.0 mg; Extraction time: 20 min; Eluent type: methanol (containing 1‰ formic acid); Eluent dosage: 3 mL; Elution time: 3 min) (n = 3).
23
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
the best extraction efficiency was the methanol containing 10% ammonia (the actual detected concentration of MC-RR was 1.0 μg/L and MC-LR was 0.9 μg/L). So, the eluent was chosen methanol containing 10% ammonia(V/V). When the elution solvent volume is small, the target elutes would be incompletely. If the elution solvent volume is too large, the organic solvent would be wasted and the environment would be polluted. The extraction efficiency of desorption solvent volume was investigated under the volume from 1 to 6 ml. The results were shown in Fig. S2. When the elution solvent volume was 3.0 mL, the actual detected concentration was the largest (MC-RR was 0.92 μg/L and MC-LR was 0.80 μg/L). In addition, on the premise of ensuring higher material extraction efficiency, the experimental operation time was minimized to improve the analysis efficiency. As shown in Fig. S3, when the elution time reached 3 min, the actual detected concentration reached the peak (MC-RR was 0.94 μg/L and MC-LR was 0.96 μg/L) and then decreased. Based on the above experimental results, 3 min was selected as the optimal elution time.
Fig. 5. Effect of extraction time on extraction efficiency (Experimental conditions: material dosage:10.0 mg:pH = 4.00; Eluent type: methanol (containing 1‰ formic acid); Eluent dosage: 3 mL; Elution time: 3min) (n = 3).
3.2.5. Orthogonal experimental study Orthogonal test is the main method to analyze factorial design, and it has a high efficiency, fast and economical experimental design. It uses a standardized table, or an orthogonal table to design the test plan and analyze the test results. A few representative strong test conditions were selected in many test conditions. These representative points had the characteristics of “evenly dispersed, neat and comparable”. Through the data of these several trials, the optimal or better solution can be found. Based on the above single factor conditions, The four analyzed experimental parameters were the amount of IL@MG: A (A1: 9.0 mg, A2: 10.0 mg, A3: 11.0 mg), extraction time: B (B1: 22 min, B2: 20 min, B3: 18 min), eluent dosage: C (C1: 2.4 mL, C2: 3 mL, C3: 3.6 mL) and elution time: D (D1: 2.5 min, D2: 3 min, D3: 3.5 min) as the influencing factors, to proceed a 4-factors-3-horizontal orthogonal test. The peak response value was used as the evaluation index, and the orthogonal test results were shown in Table S2. According to the range R, the influenced factors was B > A > C > D, the optimal combination was A2B3C3D2, and the optimal extraction conditions were as follows: material dosage was 10.0 mg, extraction time was 18 min, desorption solvent volume was 3.6 mL, and elution time was 3 min. 3.3. Repeated use of materials
Fig. 6. Effect of different eluent types on extraction efficiency (Experimental conditions: Material dosage: 10.0 mg; pH = 4.00; Extraction time:20 min; Eluent amount: 3 mL; Elution time: 3 min) (n = 3).
In this experiment, the recovered IL@MG was washed alternately with methanol and ultrapure water for 3 times, and then used for the next extraction separation, so that the extraction was repeated 20 times under optimal conditions. As shown in Table S4, when measured the actual water samples, the lowest recovery rate was 83.6%. So, the recovery rate was preferably not less than 83.6%. From Fig. S4, when the material was reused for 14 times, the recovery rate of MC-LR was 82.8%, which was less than 83.6%. Therefore, we decided to choose no more than 12 times to use repeatedly. Considering the influence on the detection accuracy, the recovered IL@MG material could be reused at least 12 times without reducing the extraction efficiency. Therefore, IL@MG synthesized in this experiment was a rational material with good stability and repeatability.
3.2.4. Optimization of elution conditions The actual detected concentration of the target was directly affected by the selection of the eluted organic solution. IL@MG had excellent acid and alkali resistance and organic solvent properties, and it maintains a stable state in many organic materials. Hence, the range of elution solvent selections were increased to some extent. The surface charge of the target molecule was affected by the acid and alkaline of the eluent due to the negative charge of microcystins. In this experiment, the effects of methanol, acetone, acetonitrile, methanol containing 1‰ formic acid and methanol containing 2%, 4%, 6%, 8%, 10% and 12% ammonia on the extraction efficiency of the target were investigated. As shown in Fig. 6, when pure methanol (the actual detected concentration of MC-RR was 0.78 μg/L and MC-LR was 0.55 μg/L), pure acetone (the actual detected concentration of MC-RR was 0.87 μg/L and MC-LR was 0.88 μg/L) and pure acetonitrile (the actual detected concentration of MC-RR was 0.80 μg/L and MC-LR was 0.85 μg/L) were used, the polar organic solvent showed a more advantageous elution effect. The methanol extraction with 1‰ formic acid was the least efficient (the actual detected concentration of MC-RR was 0.38 μg/L and MC-LR was 0.13 μg/L). After adding a certain proportion of ammonia, the elution effect was significantly improved. The results showed that
3.4. Method evaluation MC-LR and MC-RR standard solutions were respectively prepared at concentrations of 0.005, 0.01, 0.05, 0.1, 1.0, and 10.0 μg/L. Sample preparation was performed under optimal conditions. The linear relationship was established between the ion chromatographic peak area (y) and the mass concentration of the standard solution (x, μg/L) under quantitative chromatographic mass spectrometry conditions. The experimental results were shown in Table S3, in the range of 0.01–1.0 μg/ L, there was a good linear relationship between the detected peak areas 24
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
and the concentration. The correlation coefficient of MC-LR was 0.9997. The linear relationship between the concentration of MC-RR and the detected peak area was also well in the range of 0.05–1.0 μg/L. The correlation coefficient was 0.9995. The limits of detection were (LODs, S/N = 3) from 0.216 to 0.414 ng/L and the limits of quantitation (LOQs, S/N = 10) were ranged 0.721–1.380 ng/L. Chromatograms of 2 microcystins was showed in Fig. S5. In addition, the precisions of the method were also evaluated. According to the established procedure, each water sample was divided into 5 time periods of one day and continuously measured for 5 days, 5 parallel samples were measured for each sample. The relative standard deviation (RSD) of each sample was calculated. The RSDs of daytime were ranged from 3.14% to 7.92%, and the RSDs of days were ranged from 3.98% to 19.06%. It can be seen from the data that the reproducibility and precision of this method were ideal, and the reliability was high.
advantages of easy operation, less operation time, higher sensitivity and environmentally friendly. 4. Conclusions In this experiment, the nanomaterial ionic liquid magnetic graphene was used as adsorbent to detect the MC-LR and MC-RR in water samples by the MSPE-UHPLC-MS/MS system. Due to the merits of high surface area (123.64 m2/g), accessible porosity (10.67 nm) and sensitively magnetic separation property, the as-synthesized IL@MG composite was successfully applied as an effective adsorbent to extract trace MCs. Furthermore, it was also cost-effective because the composite had the performance to reuse at least 12 times. Moreover, the results showed good linear relationships (r = 0.9997, 0.9995) and lower detection limit (0.216–0.414 ng/L), which has obvious advantages compared with similar methods. In summary, the proposed MSPE-UHPLC-MS/MS method exhibited outstanding comprehensive performance was suitable for the detection of trace compounds.
3.5. Determination of actual water samples In summer, there was an outbreak of cyanobacteria, so this experiment selected to collect water samples in August 2018. Real water samples were collected from three water collecting points (Manshan, 2#, Siqian Village) in Tai Lake. Determinate 5 parallel samples respectively for unfiltered water samples, water samples filtered through a 0.45 μm filter membrane (Shanghai Xinya Purification Device Factory, Shanghai, China) by 1000 mL filter bottle, 80 mm Buchner funnel (Guangzhou Code Sharpening Experimental Instrument Company, China) for once, and former mentioned filtered samples with MCs added in making the concentrations to 0.1 μg/L and 1.0 μg/L. All the water samples stored in bottles at 4 °C prior to analysis. As shown in Table S4, the data showed that a small amount of MCs (MC-LR was 0.020 μg/L and MC-RR was 0.021 μg/L) was detected in the actual water samples, and the contaminant contents of unfiltered water samples were greater than the filtered water samples. When the water sample concentrations were 0.1 μg/L and 1.0 μg/L, the recoveries of Manshan were in the range of 83.6–100.4%, and the relative standard deviations did not exceed 6.10%. The recovery rate of 2# points was from 88.3% to 100.9%, and the relative standard deviation was no more than 7.59%. The recovery rate of Siqian Village was from 88.9 to 98.9%, and the relative standard deviation was no more than 6.92%. From the determination of actual water samples, this adsorbent is effective for detection. The synthetic IL@MG can adsorb microcystins in water. In addition, we found that this material can also remove organic compounds such as antibiotics in water. As a result, we can also use this adsorbent to pure environmental water samples.
Acknowledgements This work was jointly supported by the National Natural Science Foundation of China (51778390) and the Major Project of Natural Science Foundation of the Higher Education Institutions of Jiangsu Province (No: 15KJA610003). Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ecoenv.2019.03.063. References Chang, H., Wu, H., 2013. Graphene-based nanocomposites: preparation, functionalization, and energy and environmental applications. Energy Environ. Sci. 6, 3483–3507. Chen, X., et al., 2013. Fe3O4@MOF core-shell magnetic microspheres for magnetic solidphase extraction of polychlorinated biphenyls from environmental water samples. J. Chromatogr. A 1304, 241–245. Cui, Y., et al., 2018. HLB/PDMS-coated stir bar sorptive extraction of microcystins in shellfish followed by high-performance liquid chromatography and mass spectrometry analysis. Food Anal. Methods 11, 1748–1756. Devasurendra, A.M., et al., 2018. Solid-phase extraction, quantification, and selective determination of microcystins in water with a gold-polypyrrole nanocomposite sorbent material. J. Chromatogr. A 1560, 1–9. Falconer, I.R., 1999. An overview of problems caused by toxic blue-green algae (cyanobacteria) in drinking and recreational water. Environ. Toxicol. 14, 5–12. Falconer, I.R., et al., 1999. Hepatic and renal toxicity of the blue-green alga (cyanobacterium) Cylindrospermopsis raciborskii in male Swiss albino mice. Environ. Toxicol. 14, 143–150. Hoff, R.B., Pizzolato, T.M., 2018. Combining extraction and purification steps in sample preparation for environmental matrices: a review of matrix solid phase dispersion (MSPD) and pressurized liquid extraction (PLE) applications. Trac. Trends Anal. Chem. 109, 83–96. Kaloudis, T., et al., 2013. Determination of microcystins and nodularin (cyanobacterial toxins) in water by LC-MS/MS. Monitoring of lake Marathonas, a water reservoir of Athens, Greece. J. Hazard Mater. 263, 105–115. Li, Q., et al., 2017. Analysis of microcystins using high-performance liquid chromatography and magnetic solid-phase extraction with silica-coated magnetite with cetylpyridinium chloride. J. Sep. Sci. 40, 1644–1650. Liu, S., et al., 2011. Preparation and characterization of microcystins-LR by HPLC and HPLC-MS5. In: Cao, Z. (Ed.), Application of Chemical Engineering, Pts 1-3, pp. 120–+. Meisen, I., et al., 2009. Direct coupling of high-performance thin-layer chromatography with UV spectroscopy and IR-MALDI orthogonal TOF MS for the analysis of cyanobacterial toxins. Anal. Chem. 81, 3858–3866. Meng, J., et al., 2011. Preparation of polypyrrole-coated magnetic particles for micro solid-phase extraction of phthalates in water by gas chromatography-mass spectrometry analysis. J. Chromatogr. A 1218, 1585–1591. Munoz, G., et al., 2017. Analysis of individual and total microcystins in surface water by on-line preconcentration and desalting coupled to liquid chromatography tandem mass spectrometry. J. Chromatogr. A 1516, 9–20. Nardecchia, S., et al., 2013. Three dimensional macroporous architectures and aerogels built of carbon nanotubes and/or graphene: synthesis and applications. Chem. Soc. Rev. 42, 794–830. Neilan, B.A., et al., 2013. Environmental conditions that influence toxin biosynthesis in
3.6. Method performance comparison The analytical performance of the IL@MG adsorbent coupled with UHPLC-MS/MS analysis was compared with other methods. As listed in Table S5, several pretreatment methods such as SBSE (Ortiz et al., 2017), SPE (Qian et al., 2017; Zervou et al., 2017), ASE-SPE (Rivasseau et al., 1998), online-SPE (Munoz et al., 2017; Vudathala et al., 2017), MSPD (Cui et al., 2018) and MSPE (Li et al., 2017; Sun et al., 2015) were presented. Obviously, compared with other methods, the LOD (0.414 ng/L) obtained by this method was at the lowest level, which was far below the prescribed limit (1.0 μg/L). The LOD of references (Cui et al., 2018; Qian et al., 2017) were not high. In addition, the extraction time of this method was 18 min, which was less than SBSE (60 min) (Ortiz et al., 2017), SPE (40 min) (Qian et al., 2017) and MSPE (25 min and 30min) (Li et al., 2017; Sun et al., 2015). The results indicated that the extraction efficiency of this method was dominant. In addition, compared to general SPE methods (Munoz et al., 2017; Qian et al., 2017; Vudathala et al., 2017; Zervou et al., 2017), the organic solvent used in this method was less and it was more environmentally friendly. Therefore, this method (MSPE-UPLC-MS/MS) had the 25
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al. cyanobacteria. Environ. Microbiol. 15, 1239–1253. Okabe, T., et al., 2017. Development and performance of a magnetic ionic liquid for use in vacuum-compatible non-contact seals. Precis. Eng. J.Int. Soc. Precis. Eng. Nanotechnol. 47, 97–103. Ortiz, X., et al., 2017. A high throughput targeted and non-targeted method for the analysis of microcystins and anatoxin-A using on-line solid phase extraction coupled to liquid chromatography-quadrupole time-of-flight high resolution mass spectrometry. Anal. Bioanal. Chem. 409, 4959–4969. Pavagadhi, S., et al., 2011. Application of ionic-liquid supported cloud point extraction for the determination of microcystin-leucine-arginine in natural waters. Anal. Chim. Acta 686, 87–92. Pietsch, J., et al., 2001. Simultaneous determination of cyanobacterial hepato- and neurotoxins in water samples by ion-pair supported enrichment and HPLC-ESI-MS-MS. Chromatographia 54, 339–344. Qian, Z.-Y., et al., 2017. Analysis of trace microcystins in vegetables using matrix solidphase dispersion followed by high performance liquid chromatography triple quadrupole mass spectrometry detection. Talanta 173, 101–106. Rivasseau, C., et al., 1998. Determination of some physicochemical parameters of microcystins (cyanobacterial toxins) and trace level analysis in environmental samples using liquid chromatography. J. Chromatogr. A 799, 155–169. Sedda, T., et al., 2016. Determination of microcystin-LR in clams (Tapes decussatus) of two Sardinian coastal ponds (Italy). Mar. Pollut. Bull. 108, 317–320. Shamsollahi, H.R., et al., 2014. Measurement of microcystin -LR in water samples using improved HPLC method. Glob. J. Health Sci. 7, 66–70. Sun, H., et al., 2015. Solid-phase extraction based on mesoporous Fe3O4@mSiO(2) @Cu2+ magnetic nanoparticles coupled with high performance liquid chromatography for the determination of microcystins in water samples. Chin. J. Chromatogr. 33, 449–454. Sypabekova, M., et al., 2017. Selection, characterization, and application of DNA aptamers for detection of mycobacterium tuberculosis secreted protein MPT64. Tuberculosis 104, 70–78.
Taghdisi, S.M., et al., 2017. A novel fluorescent aptasensor for ultrasensitive detection of microcystin-LR based on single-walled carbon nanotubes and dapoxyl. Talanta 166, 187–192. Vudathala, D., et al., 2017. Analysis of microcystin-LR and nodularin using triple quad liquid chromatography-tandem mass spectrometry and histopathology in experimental fish. Toxicon 138, 82–88. Wei, H., et al., 2015. Electrospun polymer nanofibres as solid-phase extraction sorbents for extraction and quantification of microcystins. Environ. Technol. 36, 2796–2802. Xiao, X., et al., 2012. Lithographically defined three-dimensional graphene structures. ACS Nano 6, 3573–3579. Xu, X., et al., 2013. Sensitive determination of total microcystins with GC-MS method by using methylchloroformate as a derivatizing reagent. Anal. Methods 5, 1799–1805. Yan, H., et al., 2013. Synthesis of multi-core-shell magnetic molecularly Imprinted microspheres for rapid recognition of dicofol in tea. J. Agric. Food Chem. 61, 2896–2901. Zervou, S.-K., et al., 2017. New SPE-LC-MS/MS method for simultaneous determination of multi-class cyanobacterial and algal toxins. J. Hazard Mater. 323, 56–66. Zhang, H., et al., 2017a. Simple, high efficiency detection of microcystins and nodularin-R in water by fluorescence polarization immunoassay. Anal. Chim. Acta 992, 119–127. Zhang, K., et al., 2010a. Graphene oxide/ferric hydroxide composites for efficient arsenate removal from drinking water. J. Hazard Mater. 182, 162–168. Zhang, Q., et al., 2010b. Ionic liquid-coated Fe3O4 magnetic nanoparticles as an adsorbent of mixed hemimicelles solid-phase extraction for preconcentration of polycyclic aromatic hydrocarbons in environmental samples. Analyst 135, 2426–2433. Zhang, W., et al., 2017b. Magnetic porous beta-cyclodextrin polymer for magnetic solidphase extraction of microcystins from environmental water samples. J. Chromatogr. A 1503, 1–11. Zhang, Z., Kong, J., 2011. Novel magnetic Fe3O4@C nanoparticles as adsorbents for removal of organic dyes from aqueous solution. J. Hazard Mater. 193, 325–329. Zhu, X., et al., 2014. Preparation of graphene-Fe3O4 nanocomposites using Fe3+ ioncontaining magnetic ionic liquid. Mater. Res. Bull. 59, 223–226.
26
Contents lists available at ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Determination of microcystins in environmental water samples with ionic liquid magnetic graphene
T
Xiaoyan Liua, Shiqian Gaoa,b, Xinyue Lia, Hui Wanga, Xiaowen Jic, Zhanen Zhanga,b,∗ a
School of Environmental Science and Engineering, Suzhou University of Science and Technology, No. 1 Kerui Road, Suzhou, 215009, PR China Jiangsu Key Laboratory of Environmental Science and Engineering, Suzhou University of Science and Technology, No. 1 Kerui Road, Suzhou, 215009, PR China c State Key Laboratory of Pollution Control and Resource Reuse, Center for Hydrosciences Research, School of the Environment, Nanjing University, Nanjing, 210093, PR China b
ARTICLE INFO
ABSTRACT
Keywords: Microcystin Ionic liquid magnetic graphene Magnetic solid phase extraction Ultra high performance liquid chromatography tandem mass spectrometry
Microcystins is a class of monocyclic of heptapeptides with many different isomerides. It has become potential hazardous material in water environment for its toxic, distribution and stability. This project worked on a method for determination of trace microcystin (MC-LR and MC-RR) in environmental waters. The ionic liquid magnetic graphene (IL@MG) was prepared and applied to the concentration and determination of microcystins, based on magnetic solid phase extraction (MSPE), and coupled with ultra high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). The ionic liquid magnetic graphene was prepared by coprecipitatial synthesis and characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM), fourier transform infrared spectroscopy (FT-IR), specific surface area (BET), pore size distribution (BJH) and magnetic hysteresis loop. The experimental parameters of magnetic solid phase extraction, including amount of IL@MG, pH, extraction time and elution solvent were investigated by a univariate method and orthogonal screening. The method showed good linearity in the range of 0.01–10.0 g/L and 0.005–10.0 μg/L for MC-LR and MC-RR, when the pH of water samples was 4.00 and 10.0 mg adsorbents were used to extract targets for 18 min. The lowest detection limit was 0.414 ng/L and 0.216 ng/L for MC-LR and MC-RR respectively. The recoveries of the microcystins were in the range of 83.6–100.9%, and the relative standard deviation was less than 7.59%. The trace amount of MC-LR (0.020 μg/L) and MC-RR (0.003 μg/L and 0.021 μg/L) was detected in actural water samples. Attributed to its simple operator, low detection limit and high sensitivity, this method could be used for the detection of trace microcystins in water samples.
1. Introduction A large number of pollutants containing nitrogen and phosphorus entered the water through surface runoff and sewage pipe network, resulting in eutrophication of water bodies. With the suitable temperature and light intensity, the rapid growth of algae substances lead to water bloom, which causes the release of harmful substances such as microcystins (MCs) (Falconer, 1999). MCs had a certain toxic effect on the spleen and kidney, and it was also a tumor promoter (Kaloudis et al., 2013). In addition, MCs can affect post-embryonic development (Falconer et al., 1999), posing a major threat to water quality and human health. More than 90 kinds of microcystins with variable toxicity had been identified (Sypabekova et al., 2017). Of all MCs, MC-LR was the most common toxic variant which had multiple organ toxicity, genotoxicity and carcinogencity (Neilan et al., 2013). Both the World
Health Organization (WHO) and China's drinking water hygiene regulations stipulated for the minimal permissible concentration of MCs in drinking water (< 1.0 μg/L) (Zhang et al., 2017b). Therefore, it is important to choose the absorbent to establish the proper detection method. At present, the detection methods of microcystins in environmental water are published as the follows. Highly performed liquid chromatography (HPLC) (Pietsch et al., 2001), high performance liquid chromatography tandem mass spectrometry (HPLC-MS/MS) (Liu et al., 2011; Shamsollahi et al., 2014), thin layer chromatography (TLC) (Xu et al., 2013), biological treatment (Meisen et al., 2009), enzyme-linked immunosorbent assays (ELISAs) (Taghdisi et al., 2017), protein phosphatase inhibition assay (PPIA) (Zhang et al., 2017a) and lateral flow immunoassays (LFIAs) (Sedda et al., 2016) has been used to detect MCs. Among these methods, HPLC-MS/MS was a powerful tool that could be
∗ Corresponding author. School of Environmental Science and Engineering, Suzhou University of Science and Technology, No. 1 Kerui Road, Suzhou 215009, PR China. Tel.: +86 13912606902; fax: +86 512-68093060. E-mail address: [email protected] (Z. Zhang).
https://doi.org/10.1016/j.ecoenv.2019.03.063 Received 26 December 2018; Received in revised form 14 March 2019; Accepted 14 March 2019 0147-6513/ © 2019 Elsevier Inc. All rights reserved.
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
confidently used to identify specific variants of MCs, it not only has good reproducibility and low sensitivity, but also can analyze isomers. In addition, it has the advantages of allowing quantitative and qualitative analysis of toxins at the same time with faster speed and more sensitive limits of detection (LOD). The pretreatment method of microcystins is an important process. Solid phase extraction (SPE) (Devasurendra et al., 2018; Li et al., 2017; Wei et al., 2015), accelerated solvent extraction - solid-phase extraction (ASE-SPE) (Hoff and Pizzolato, 2018), online solid-phase extraction (online-SPE) (Vudathala et al., 2017), stir bar sorptive extraction (SBSE) (Ortiz et al., 2017), matrix solid-phase dispersion (MSPD) (Cui et al., 2018) were used for the extraction od MCs before. These pretreatment methods are cumbersome and take a long time to operate. In recent years, a new method - magnetic solid phase extraction (MSPE) (Qian et al., 2017) technology has emerged. This is a new type of pretreatment method that combines solid phase extraction with magnetic and magnetically modified materials. It has convenient separation process, simple operation, and easy separation in complex matrix systems. The adsorbent materials which can be used for magnetic solid phase extraction mainly included magnetic carbon materials (Zhang and Kong, 2011), magnetic molecular imprinted materials (Yan et al., 2013), metal organic framework compounds (Chen et al., 2013), ionic liquids (Zhang et al., 2010b) and magnetic polymer materials (Meng et al., 2011). In this work, ionic liquids (IL) and Fe3O4 were simultaneously introduced to load on graphene. IL refers to a liquid composed entirely by ions. It is usually composed by organic cations and inorganic anions. It has the advantages of non-volatility, high solubility, low vapor pressure, and good thermal stability (Okabe et al., 2017). Simultaneously, graphene has ideal physical and chemical properties, high specific surface area (Zhang et al., 2010a), good stability, high electron mobility (Xiao et al., 2012) and good thermal conductivity (Chang and Wu, 2013). The preparation process of graphene was simple, and the promotion and application of graphene was easy in production (Nardecchia et al., 2013). Modifying graphene by using ionic liquids can reduce the occurrence of agglomeration effects, improve the hydrophilicity of graphene, and solve the shortcomings of uneven distribution in water (Nardecchia et al., 2013). By combining and magnetizing the two materials, a composite nanomaterial with magnetic properties can be formed. In this experiment, we had attempted to explore the possibility of simultaneous extraction and preconcentration of trace MC-LR and MCRR in water samples. Then a new type of magnetic composite material had been prepared by loading ionic liquid and Fe3O4 onto graphene. This synthetic material (IL@MG) was a kind of adsorbent that combined the advantages of high adsorption efficiency, environmental protection and high sensitivity. IL@MG was used as an adsorbent for magnetic solid phase extraction (MSPE), and in combination and in combination with UHPLC-MS/MS to detect microcystins in environmental waters.
heptahydrate of analytically pure were purchased from Sinopharm Chemical Reagent corporation. Ammonia was analytically pure and purchased from Wuxi Prospect Chemical Reagent Company. Anhydrous ethanol was analytically pure and purchased from Wuxi Jingke Chemical Company. The experimental water was ultrapure water made by Milli-Q at 18.2 ΩM (Millipore, USA). Analytical standards of MC-LR and MC-RR were purchased from Standford Analytical Chemicals (USA) and its purity was above 98%. It was prepared by adding pure methanol to a 100.0 μg/mL single standard stock solution and stored at −20 °C. The above single standard stock solution was then diluted to a working standard solution of 1.0 mg/L. 2.2. Experimental methods 2.2.1. Synthesis of IL@MG IL@MG was synthesized by coprecipitatial synthesis method. It was prepared according to the chemical coprecipitation reaction when mixing Fe3+, Fe2+ and graphene under alkaline conditions (Zhu et al., 2014). The accurate weighing 0.0824 g graphene was dispersed in 100 mL ultrapure water, then added 1-butyl-3-methylimidazolium bromide (20.0 g) containing 0.008 mol (2.69 g) 1-butyl-3-methylimidazolium tetrachloride. The mixture was stirred with a glass rod to dissolve it sufficiently, and ultrasonically dispersed for 2 h by KQ300VDE type three-frequency digital ultrasonic cleaner (Kunshan Ultrasonic Instrument Company, China). The above materials were transferred to a 250 mL four-necked flask (Nanjing Pengcheng Chemical Glass Instrument, China), and 0.004 mol (1.11 g) of FeSO4·7H2O was added under nitrogen atmosphere and high-speed stirring process by using DC-12 nitrogen blower (Shanghai Anpu Experimental Technology Company, China) and JJ-1 Precision Forced Electric Mixer (Jintan Analytical Instrument Company, China.). Dropwise slowly added 10 mL of 25% aqueous ammonia during the stirring. It was then stirred at room temperature for 3 h. The precipitate was separated under an external magnetic field. Finally, the materials were alternately washed with anhydrous ethanol and deionized water for three times. It was dried under vacuum circumstance at 60 °C for 12 h by VD115 Vacuum Drying Chamber (Binder, Germany). After removal, it was ground and obtained 1.0301 g of IL@MG powder, and stored after grinding. 2.2.2. Magnetic solid phase extraction Added 100 mL water sample to a 250 mL stoppered conical flask (Nanjing Pengcheng Chemical Glass Instrument, China). Glacial acetic acid was added to adjust the pH to 4.00 by using PHS-3C pH meter (Shanghai Yidian Scientific Instrument Company, China). Accurately weighed 10 mg IL@MG into the above conical flask. The adsorption was carried out for 18 min by using HY-2 variable speed multi-purpose oscillator (Instrument Company, China.). After the magnetic phase separation, the supernatant was decanted and added 1.8 mL of methanol containing 10% ammonia water. After desorbed by using SK-1 Rapid Mixer (Guohua Instrument Company, China) for 1.5 min, using a 0.22 μm organic filter (Haining Dacheng Filter Equipment Company, China) to remove elution solvent in a 10 mL centrifuge tube, repeated the desorption process. Finally, the 3.6 mL of the elution solvent was evaporated by DC-12 nitrogen blower (Shanghai Anpu Experimental Technology Company, China.). After the initial mobile phase was reconstituted to 1 mL, the analysis was analyzed by using UHPLC-MS/MS.
2. Methods and materials 2.1. Instruments and reagents High-quality graphene used in the experiment has a diameter of 0.5–5 μm and a thickness of 0.8–1.2 nm. Its purity was 99%, and the single layer rate is greater than 96%. It was purchased from Xianfeng Nanotechnology corporation (Nanjing, China). The purity of 1-butyl-3methylimidazolium tetrachloride and 1-butyl-3-methylimidazolium bromide (Shanghai Chengjie chemistry, China) are both 99%. FeSO4·7H2O were purchased from Sinopharm (Shanghai, China). Methanol and acetonitrile of chromatographically pure were purchased from Tedia Company of the United States. Formic acid of chromatographically pure was purchased from Aladdin. Methanol, acetone, acetonitrile, ethanol, formic acid, glacial acetic acid and Ferrous sulfate
2.2.3. Chromatographic conditions Analysis experiment was carried out on a TSQ Quantum Access MAX mas spectrometer (Thermo Fisher, USA) equipped with an electrospray ionization (ESI) source and coupled with HPLC System (Dionex, USA). Column: Zorbax Eclipse XDB-C18 reversed phase column (50 mm × 3 mm, 1.8 μm); column temperature: 40 °C; mobile phase: methanol (A) and pure water containing one thousandth of formic acid 21
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
Fig. 1. SEM of IL@MG (a,b), TEM of IL@MG (c,d).
(B). Used gradient elution for separation of MC-LR and MC-RR. The gradient elution procedures were as follows: 0–3min: 5%A-95% A; 3–4.5min, 95% A-100% A; 4.5–5min, 100% A; 5.0–5.1min: 100% A5%A; 5.1–8min: 5% A; flow rate: 0.4 mL/min; injection volume: 10 μL.
material was mesoporous composite. According to Fig. 2 (right), the magnetic hysteresis loops of IL@MG showed almost zero coercivity and remanence, which the maximum saturation magnetization was 34.2 emu/g. The IL@MG loaded with the target analytes could be separated readily from sample medium due to the super-magntism and large saturation magnetization. The data indicated that IL and Fe3O4 were successfully loaded onto the graphene material during the process of synthesis. The ionic liquid possessed by the outer surface of the nano material enabled the experiment to be preferably dissolved in water and better dispersed in the water sample to adsorb the target, thereby making the adsorption effect more ideal.
2.2.4. Mass spectrometry conditions Ion source: electrospray ionization source (ESI+); scan mode: positive ion scan; detection mode: multiple reaction monitoring (MRM); ion transfer tube temperature:350 °C; spray temperature:300 °C; spray voltage:3500 V; sheath gas Pressure:35 psi; Auxiliary gas pressure:15 psi; Collision gas pressure:1.5 mTorr; The parameters of the MRM mode were shown in Table S1 (supporting information).
3.2. Optimization of magnetic solid phase extraction conditions
3. Results and discussion
3.2.1. Optimization of material amount MCs reached extremely high concentrations in aquatic waters, especially at the end of the vegetation season when cyanobacterial blooms collapsed. Both plankton and benthic microorganisms can participate in the adsorption and degradation of high MCs, so the selection of adsorbents is very important for the removal of microcystins. The experiment weighed 2.0–12.0 mg of IL@MG put into the spiked solution, which contains 100 mL 1.0 μg/L MCs. This process was used to investigate the effect of the amount of material on the extraction efficiency of MCs. Fig. 3 showed that the actual detected concentration of MCs increased with the amount of material increasing from 2.0 mg to 10.0 mg, and then the actual detected concentration decreased with the increasing amount of material. It was indicated that 10.0 mg IL@MG had an ideal enrichment ability for MC-LR (actual detected concentration was 0.95 μg/L) and MC-RR (actual detected concentration was 0.95 μg/L). However, when the amount of adsorbent continued to increase, it would hinder the effective elution of the target. It could be explained that although the total adsorption capacity of the adsorbent would increase, the unit adsorption capacity would decrease along with it. In the limited adsorption time, the excess adsorption could not reach adsorption equilibration, which would lead to a decrease in recovery. Therefore, this study selected 10.0 mg IL@MG as the magnetic solid phase extraction adsorbent.
3.1. Surface topography analysis of materials Scanning electron microscopy (SEM), transmission electron microscopy (TEM) and fourier transform infrared spectroscopy (FT-IR) were used to characterize IL@MG. Pictures of SEM and TEM were showed in Fig. 1. The surface wrinkles of the material were obvious, indicating that the material had a large specific surface area. IL and Fe3O4 were mostly attached to the surface of the graphene sheet. Under the applied magnetic field, IL@MG can be effectively separated from the matrix. As seen in Fig. S1 (supporting information), the characteristic peaks of the materials were given by the infrared spectrum. The peak near 3400 cm−1 is the -OH characteristic absorption peak of the adsorbed water molecule. And the absorption peaks in the range of 2895–2929 cm−1 is C-H stretching vibration peaks on the imidazole ring. The absorption peak of 1625 cm−1 is the absorption peak caused by the stretching vibration of C=O in the carboxyl group. The peak of 1389 cm−1 is the stretching vibration absorption peak of COO−. The absorption peak at 1124 cm−1 belongs to the vibration absorption peak of C-Br from ionic liquid and 588 cm−1 belongs to the Fe-O bending vibration absorption peak from Fe3O4. In addition, specific surface area (BET), pore size distribution (BJH) and magnetic hysteresis loop were used to characterize the prepared adsorbent. As shown in Fig. 2(left), the BET surface area and BJH volume of pores were 123.64 m2/g and 0.375 cm3/g, indicating a large surface area and relatively high porosity. Moreover, the pore size was 10.67 nm, which proved that the
3.2.2. Optimization of pH and ionic strength of water samples The pH of the water sample affects the chemical properties of the 22
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
Fig. 2. BET and BJH of IL@MG (left) The magnetic hysteresis loop of IL@MG (right).
analytes and the binding sites on the surface of the adsorbent. In this experiment, the pH of the solution was adjusted to 2.00–12.00 by glacial acetic acid and aqueous ammonia. When the pH was 4.00, the actual detected concentration of MC-RR (0.94 μg/L) and MC-LR (1.0 μg/L) reached the highest level (Fig. 4). Due to the good acid and alkali resistance of IL@MG (Nardecchia et al., 2013), the structure of the adsorbent was kept stable in acidbased solution without decomposition and destruction (Okabe et al., 2017). When the pH was small, the microcapsule molecules with positive, negative and uncharged were simultaneously presented, and the total molecules was electrically neutral. With the increasing of pH (pH > 2.1), the negatively microcystin molecules gradually added. When the pH was 4.0, the negatively microcystin molecules were about 100%. At this time, the electrostatic attraction made IL@MG show strong extraction ability to microcystin molecules, which was in completely agreement with the experimental results. As the pH continued increasing, the number of positive charges on the surface of the adsorbent gradually decreased, and electrostatic repulsion with the negatively charged microcystins molecules resulted in a decrease in adsorption performance. Therefore, in the subsequent measurement of the actual water sample, the pH of the water sample was adjusted to 4.00. The higher ionic strength of the water sample is, the lower solubility of organic matter has in water. In this experiment, the effect of ionic strength on extraction efficiency was investigated by adding different concentrations (0–20%, w/v) of NaCl to the aqueous solutions. The results showed that the ionic strength of the water sample did not affect the extraction efficiency of IL@MG for the two MCs. Therefore, it was not necessary to adjust the ionic strength of the solution (as the data not shown).
Fig. 3. Effect of material dosage on extraction efficiency (Experimental conditions: pH = 4.00; Extraction time: 20 min; Eluent type: methanol (containing 1‰ formic acid); Eluent dosage: 3 mL; Elution time: 3min) (n = 3).
3.2.3. Optimization of extraction time The alternative extraction time was set as 0, 5, 10, 15, 20 and 30 min after the start of experiment respectively. As shown in Fig. 5, when the extraction time was 20 min, both MC-RR and MC-LR reached the adsorption equilibrium, and the actual detected concentration (MCRR was 1.06 μg/L and MC-LR was 0.95 μg/L) were both at the maximum. The adsorption effect was at its best value. When the extraction time was increased from 20 min to 30 min, the actual detected concentration of MCs decreased slightly. This was due to the long extraction time leading to the microcystins adsorbed by the adsorbents dissolved again in the solution (Pavagadhi et al., 2011), resulting in a decrease of actual detected concentration. Therefore, 20.0 min was selected as the extraction time of subsequent experimental.
Fig. 4. Effect of pH on extraction efficiency (Experimental conditions: material dosage：10.0 mg; Extraction time: 20 min; Eluent type: methanol (containing 1‰ formic acid); Eluent dosage: 3 mL; Elution time: 3 min) (n = 3).
23
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
the best extraction efficiency was the methanol containing 10% ammonia (the actual detected concentration of MC-RR was 1.0 μg/L and MC-LR was 0.9 μg/L). So, the eluent was chosen methanol containing 10% ammonia(V/V). When the elution solvent volume is small, the target elutes would be incompletely. If the elution solvent volume is too large, the organic solvent would be wasted and the environment would be polluted. The extraction efficiency of desorption solvent volume was investigated under the volume from 1 to 6 ml. The results were shown in Fig. S2. When the elution solvent volume was 3.0 mL, the actual detected concentration was the largest (MC-RR was 0.92 μg/L and MC-LR was 0.80 μg/L). In addition, on the premise of ensuring higher material extraction efficiency, the experimental operation time was minimized to improve the analysis efficiency. As shown in Fig. S3, when the elution time reached 3 min, the actual detected concentration reached the peak (MC-RR was 0.94 μg/L and MC-LR was 0.96 μg/L) and then decreased. Based on the above experimental results, 3 min was selected as the optimal elution time.
Fig. 5. Effect of extraction time on extraction efficiency (Experimental conditions: material dosage:10.0 mg:pH = 4.00; Eluent type: methanol (containing 1‰ formic acid); Eluent dosage: 3 mL; Elution time: 3min) (n = 3).
3.2.5. Orthogonal experimental study Orthogonal test is the main method to analyze factorial design, and it has a high efficiency, fast and economical experimental design. It uses a standardized table, or an orthogonal table to design the test plan and analyze the test results. A few representative strong test conditions were selected in many test conditions. These representative points had the characteristics of “evenly dispersed, neat and comparable”. Through the data of these several trials, the optimal or better solution can be found. Based on the above single factor conditions, The four analyzed experimental parameters were the amount of IL@MG: A (A1: 9.0 mg, A2: 10.0 mg, A3: 11.0 mg), extraction time: B (B1: 22 min, B2: 20 min, B3: 18 min), eluent dosage: C (C1: 2.4 mL, C2: 3 mL, C3: 3.6 mL) and elution time: D (D1: 2.5 min, D2: 3 min, D3: 3.5 min) as the influencing factors, to proceed a 4-factors-3-horizontal orthogonal test. The peak response value was used as the evaluation index, and the orthogonal test results were shown in Table S2. According to the range R, the influenced factors was B > A > C > D, the optimal combination was A2B3C3D2, and the optimal extraction conditions were as follows: material dosage was 10.0 mg, extraction time was 18 min, desorption solvent volume was 3.6 mL, and elution time was 3 min. 3.3. Repeated use of materials
Fig. 6. Effect of different eluent types on extraction efficiency (Experimental conditions: Material dosage: 10.0 mg; pH = 4.00; Extraction time:20 min; Eluent amount: 3 mL; Elution time: 3 min) (n = 3).
In this experiment, the recovered IL@MG was washed alternately with methanol and ultrapure water for 3 times, and then used for the next extraction separation, so that the extraction was repeated 20 times under optimal conditions. As shown in Table S4, when measured the actual water samples, the lowest recovery rate was 83.6%. So, the recovery rate was preferably not less than 83.6%. From Fig. S4, when the material was reused for 14 times, the recovery rate of MC-LR was 82.8%, which was less than 83.6%. Therefore, we decided to choose no more than 12 times to use repeatedly. Considering the influence on the detection accuracy, the recovered IL@MG material could be reused at least 12 times without reducing the extraction efficiency. Therefore, IL@MG synthesized in this experiment was a rational material with good stability and repeatability.
3.2.4. Optimization of elution conditions The actual detected concentration of the target was directly affected by the selection of the eluted organic solution. IL@MG had excellent acid and alkali resistance and organic solvent properties, and it maintains a stable state in many organic materials. Hence, the range of elution solvent selections were increased to some extent. The surface charge of the target molecule was affected by the acid and alkaline of the eluent due to the negative charge of microcystins. In this experiment, the effects of methanol, acetone, acetonitrile, methanol containing 1‰ formic acid and methanol containing 2%, 4%, 6%, 8%, 10% and 12% ammonia on the extraction efficiency of the target were investigated. As shown in Fig. 6, when pure methanol (the actual detected concentration of MC-RR was 0.78 μg/L and MC-LR was 0.55 μg/L), pure acetone (the actual detected concentration of MC-RR was 0.87 μg/L and MC-LR was 0.88 μg/L) and pure acetonitrile (the actual detected concentration of MC-RR was 0.80 μg/L and MC-LR was 0.85 μg/L) were used, the polar organic solvent showed a more advantageous elution effect. The methanol extraction with 1‰ formic acid was the least efficient (the actual detected concentration of MC-RR was 0.38 μg/L and MC-LR was 0.13 μg/L). After adding a certain proportion of ammonia, the elution effect was significantly improved. The results showed that
3.4. Method evaluation MC-LR and MC-RR standard solutions were respectively prepared at concentrations of 0.005, 0.01, 0.05, 0.1, 1.0, and 10.0 μg/L. Sample preparation was performed under optimal conditions. The linear relationship was established between the ion chromatographic peak area (y) and the mass concentration of the standard solution (x, μg/L) under quantitative chromatographic mass spectrometry conditions. The experimental results were shown in Table S3, in the range of 0.01–1.0 μg/ L, there was a good linear relationship between the detected peak areas 24
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al.
and the concentration. The correlation coefficient of MC-LR was 0.9997. The linear relationship between the concentration of MC-RR and the detected peak area was also well in the range of 0.05–1.0 μg/L. The correlation coefficient was 0.9995. The limits of detection were (LODs, S/N = 3) from 0.216 to 0.414 ng/L and the limits of quantitation (LOQs, S/N = 10) were ranged 0.721–1.380 ng/L. Chromatograms of 2 microcystins was showed in Fig. S5. In addition, the precisions of the method were also evaluated. According to the established procedure, each water sample was divided into 5 time periods of one day and continuously measured for 5 days, 5 parallel samples were measured for each sample. The relative standard deviation (RSD) of each sample was calculated. The RSDs of daytime were ranged from 3.14% to 7.92%, and the RSDs of days were ranged from 3.98% to 19.06%. It can be seen from the data that the reproducibility and precision of this method were ideal, and the reliability was high.
advantages of easy operation, less operation time, higher sensitivity and environmentally friendly. 4. Conclusions In this experiment, the nanomaterial ionic liquid magnetic graphene was used as adsorbent to detect the MC-LR and MC-RR in water samples by the MSPE-UHPLC-MS/MS system. Due to the merits of high surface area (123.64 m2/g), accessible porosity (10.67 nm) and sensitively magnetic separation property, the as-synthesized IL@MG composite was successfully applied as an effective adsorbent to extract trace MCs. Furthermore, it was also cost-effective because the composite had the performance to reuse at least 12 times. Moreover, the results showed good linear relationships (r = 0.9997, 0.9995) and lower detection limit (0.216–0.414 ng/L), which has obvious advantages compared with similar methods. In summary, the proposed MSPE-UHPLC-MS/MS method exhibited outstanding comprehensive performance was suitable for the detection of trace compounds.
3.5. Determination of actual water samples In summer, there was an outbreak of cyanobacteria, so this experiment selected to collect water samples in August 2018. Real water samples were collected from three water collecting points (Manshan, 2#, Siqian Village) in Tai Lake. Determinate 5 parallel samples respectively for unfiltered water samples, water samples filtered through a 0.45 μm filter membrane (Shanghai Xinya Purification Device Factory, Shanghai, China) by 1000 mL filter bottle, 80 mm Buchner funnel (Guangzhou Code Sharpening Experimental Instrument Company, China) for once, and former mentioned filtered samples with MCs added in making the concentrations to 0.1 μg/L and 1.0 μg/L. All the water samples stored in bottles at 4 °C prior to analysis. As shown in Table S4, the data showed that a small amount of MCs (MC-LR was 0.020 μg/L and MC-RR was 0.021 μg/L) was detected in the actual water samples, and the contaminant contents of unfiltered water samples were greater than the filtered water samples. When the water sample concentrations were 0.1 μg/L and 1.0 μg/L, the recoveries of Manshan were in the range of 83.6–100.4%, and the relative standard deviations did not exceed 6.10%. The recovery rate of 2# points was from 88.3% to 100.9%, and the relative standard deviation was no more than 7.59%. The recovery rate of Siqian Village was from 88.9 to 98.9%, and the relative standard deviation was no more than 6.92%. From the determination of actual water samples, this adsorbent is effective for detection. The synthetic IL@MG can adsorb microcystins in water. In addition, we found that this material can also remove organic compounds such as antibiotics in water. As a result, we can also use this adsorbent to pure environmental water samples.
Acknowledgements This work was jointly supported by the National Natural Science Foundation of China (51778390) and the Major Project of Natural Science Foundation of the Higher Education Institutions of Jiangsu Province (No: 15KJA610003). Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ecoenv.2019.03.063. References Chang, H., Wu, H., 2013. Graphene-based nanocomposites: preparation, functionalization, and energy and environmental applications. Energy Environ. Sci. 6, 3483–3507. Chen, X., et al., 2013. Fe3O4@MOF core-shell magnetic microspheres for magnetic solidphase extraction of polychlorinated biphenyls from environmental water samples. J. Chromatogr. A 1304, 241–245. Cui, Y., et al., 2018. HLB/PDMS-coated stir bar sorptive extraction of microcystins in shellfish followed by high-performance liquid chromatography and mass spectrometry analysis. Food Anal. Methods 11, 1748–1756. Devasurendra, A.M., et al., 2018. Solid-phase extraction, quantification, and selective determination of microcystins in water with a gold-polypyrrole nanocomposite sorbent material. J. Chromatogr. A 1560, 1–9. Falconer, I.R., 1999. An overview of problems caused by toxic blue-green algae (cyanobacteria) in drinking and recreational water. Environ. Toxicol. 14, 5–12. Falconer, I.R., et al., 1999. Hepatic and renal toxicity of the blue-green alga (cyanobacterium) Cylindrospermopsis raciborskii in male Swiss albino mice. Environ. Toxicol. 14, 143–150. Hoff, R.B., Pizzolato, T.M., 2018. Combining extraction and purification steps in sample preparation for environmental matrices: a review of matrix solid phase dispersion (MSPD) and pressurized liquid extraction (PLE) applications. Trac. Trends Anal. Chem. 109, 83–96. Kaloudis, T., et al., 2013. Determination of microcystins and nodularin (cyanobacterial toxins) in water by LC-MS/MS. Monitoring of lake Marathonas, a water reservoir of Athens, Greece. J. Hazard Mater. 263, 105–115. Li, Q., et al., 2017. Analysis of microcystins using high-performance liquid chromatography and magnetic solid-phase extraction with silica-coated magnetite with cetylpyridinium chloride. J. Sep. Sci. 40, 1644–1650. Liu, S., et al., 2011. Preparation and characterization of microcystins-LR by HPLC and HPLC-MS5. In: Cao, Z. (Ed.), Application of Chemical Engineering, Pts 1-3, pp. 120–+. Meisen, I., et al., 2009. Direct coupling of high-performance thin-layer chromatography with UV spectroscopy and IR-MALDI orthogonal TOF MS for the analysis of cyanobacterial toxins. Anal. Chem. 81, 3858–3866. Meng, J., et al., 2011. Preparation of polypyrrole-coated magnetic particles for micro solid-phase extraction of phthalates in water by gas chromatography-mass spectrometry analysis. J. Chromatogr. A 1218, 1585–1591. Munoz, G., et al., 2017. Analysis of individual and total microcystins in surface water by on-line preconcentration and desalting coupled to liquid chromatography tandem mass spectrometry. J. Chromatogr. A 1516, 9–20. Nardecchia, S., et al., 2013. Three dimensional macroporous architectures and aerogels built of carbon nanotubes and/or graphene: synthesis and applications. Chem. Soc. Rev. 42, 794–830. Neilan, B.A., et al., 2013. Environmental conditions that influence toxin biosynthesis in
3.6. Method performance comparison The analytical performance of the IL@MG adsorbent coupled with UHPLC-MS/MS analysis was compared with other methods. As listed in Table S5, several pretreatment methods such as SBSE (Ortiz et al., 2017), SPE (Qian et al., 2017; Zervou et al., 2017), ASE-SPE (Rivasseau et al., 1998), online-SPE (Munoz et al., 2017; Vudathala et al., 2017), MSPD (Cui et al., 2018) and MSPE (Li et al., 2017; Sun et al., 2015) were presented. Obviously, compared with other methods, the LOD (0.414 ng/L) obtained by this method was at the lowest level, which was far below the prescribed limit (1.0 μg/L). The LOD of references (Cui et al., 2018; Qian et al., 2017) were not high. In addition, the extraction time of this method was 18 min, which was less than SBSE (60 min) (Ortiz et al., 2017), SPE (40 min) (Qian et al., 2017) and MSPE (25 min and 30min) (Li et al., 2017; Sun et al., 2015). The results indicated that the extraction efficiency of this method was dominant. In addition, compared to general SPE methods (Munoz et al., 2017; Qian et al., 2017; Vudathala et al., 2017; Zervou et al., 2017), the organic solvent used in this method was less and it was more environmentally friendly. Therefore, this method (MSPE-UPLC-MS/MS) had the 25
Ecotoxicology and Environmental Safety 176 (2019) 20–26
X. Liu, et al. cyanobacteria. Environ. Microbiol. 15, 1239–1253. Okabe, T., et al., 2017. Development and performance of a magnetic ionic liquid for use in vacuum-compatible non-contact seals. Precis. Eng. J.Int. Soc. Precis. Eng. Nanotechnol. 47, 97–103. Ortiz, X., et al., 2017. A high throughput targeted and non-targeted method for the analysis of microcystins and anatoxin-A using on-line solid phase extraction coupled to liquid chromatography-quadrupole time-of-flight high resolution mass spectrometry. Anal. Bioanal. Chem. 409, 4959–4969. Pavagadhi, S., et al., 2011. Application of ionic-liquid supported cloud point extraction for the determination of microcystin-leucine-arginine in natural waters. Anal. Chim. Acta 686, 87–92. Pietsch, J., et al., 2001. Simultaneous determination of cyanobacterial hepato- and neurotoxins in water samples by ion-pair supported enrichment and HPLC-ESI-MS-MS. Chromatographia 54, 339–344. Qian, Z.-Y., et al., 2017. Analysis of trace microcystins in vegetables using matrix solidphase dispersion followed by high performance liquid chromatography triple quadrupole mass spectrometry detection. Talanta 173, 101–106. Rivasseau, C., et al., 1998. Determination of some physicochemical parameters of microcystins (cyanobacterial toxins) and trace level analysis in environmental samples using liquid chromatography. J. Chromatogr. A 799, 155–169. Sedda, T., et al., 2016. Determination of microcystin-LR in clams (Tapes decussatus) of two Sardinian coastal ponds (Italy). Mar. Pollut. Bull. 108, 317–320. Shamsollahi, H.R., et al., 2014. Measurement of microcystin -LR in water samples using improved HPLC method. Glob. J. Health Sci. 7, 66–70. Sun, H., et al., 2015. Solid-phase extraction based on mesoporous Fe3O4@mSiO(2) @Cu2+ magnetic nanoparticles coupled with high performance liquid chromatography for the determination of microcystins in water samples. Chin. J. Chromatogr. 33, 449–454. Sypabekova, M., et al., 2017. Selection, characterization, and application of DNA aptamers for detection of mycobacterium tuberculosis secreted protein MPT64. Tuberculosis 104, 70–78.
Taghdisi, S.M., et al., 2017. A novel fluorescent aptasensor for ultrasensitive detection of microcystin-LR based on single-walled carbon nanotubes and dapoxyl. Talanta 166, 187–192. Vudathala, D., et al., 2017. Analysis of microcystin-LR and nodularin using triple quad liquid chromatography-tandem mass spectrometry and histopathology in experimental fish. Toxicon 138, 82–88. Wei, H., et al., 2015. Electrospun polymer nanofibres as solid-phase extraction sorbents for extraction and quantification of microcystins. Environ. Technol. 36, 2796–2802. Xiao, X., et al., 2012. Lithographically defined three-dimensional graphene structures. ACS Nano 6, 3573–3579. Xu, X., et al., 2013. Sensitive determination of total microcystins with GC-MS method by using methylchloroformate as a derivatizing reagent. Anal. Methods 5, 1799–1805. Yan, H., et al., 2013. Synthesis of multi-core-shell magnetic molecularly Imprinted microspheres for rapid recognition of dicofol in tea. J. Agric. Food Chem. 61, 2896–2901. Zervou, S.-K., et al., 2017. New SPE-LC-MS/MS method for simultaneous determination of multi-class cyanobacterial and algal toxins. J. Hazard Mater. 323, 56–66. Zhang, H., et al., 2017a. Simple, high efficiency detection of microcystins and nodularin-R in water by fluorescence polarization immunoassay. Anal. Chim. Acta 992, 119–127. Zhang, K., et al., 2010a. Graphene oxide/ferric hydroxide composites for efficient arsenate removal from drinking water. J. Hazard Mater. 182, 162–168. Zhang, Q., et al., 2010b. Ionic liquid-coated Fe3O4 magnetic nanoparticles as an adsorbent of mixed hemimicelles solid-phase extraction for preconcentration of polycyclic aromatic hydrocarbons in environmental samples. Analyst 135, 2426–2433. Zhang, W., et al., 2017b. Magnetic porous beta-cyclodextrin polymer for magnetic solidphase extraction of microcystins from environmental water samples. J. Chromatogr. A 1503, 1–11. Zhang, Z., Kong, J., 2011. Novel magnetic Fe3O4@C nanoparticles as adsorbents for removal of organic dyes from aqueous solution. J. Hazard Mater. 193, 325–329. Zhu, X., et al., 2014. Preparation of graphene-Fe3O4 nanocomposites using Fe3+ ioncontaining magnetic ionic liquid. Mater. Res. Bull. 59, 223–226.
26