Determination of nine bisphenols in sewage and sludge using dummy molecularly imprinted solid-phase extraction coupled with liquid chromatography tandem mass spectrometry

Journal of Chromatography A, 1552 (2018) 10–16

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Determination of nine bisphenols in sewage and sludge using dummy molecularly imprinted solid-phase extraction coupled with liquid chromatography tandem mass spectrometry Xiaoli Sun a,∗ , Junyu Peng b , Muhua Wang a , Jincheng Wang b , Chunlan Tang c , Luoxing Yang a , Hua Lei a , Fang Li b , Xueli Wang d , Jiping Chen b,∗ a

Department of Chemistry, Lishui University, Lishui 32300, China CAS Key Laboratory of Separation Sciences for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China c Medical School of Ningbo University, Ningbo 315211, China d School of Environmental Science and Engineering, Chang’an University, Xi’an 710054, China b

a r t i c l e

i n f o

Article history: Received 11 October 2017 Received in revised form 29 March 2018 Accepted 1 April 2018 Available online 3 April 2018 Keywords: Bisphenols Dummy molecularly imprinted polymer Matrix effect SPE HPLC–MS/MS

a b s t r a c t This paper describes the determination of bisphenol A (BPA), bisphenol S (BPS), bisphenol F (BPF), bisphenol E (BPE), bisphenol B (BPB), bisphenol AF (BPAF), bisphenol AP (BPAP), bisphenol Z (BPZ) and tetrabromobisphenol A (TBBPA) in sewage and sludge samples. A highly class-selective dummy molecularly imprinted polymer was used for solid phase extraction (SPE) and clean-up of the samples. Bisphenols was quantified by high-performance liquid chromatography–tandem mass spectrometry (HPLC–MS/MS). The developed method had acceptable recoveries (43.6–101%), precision (RSDs: 1.5–15%) and matrix effects (−6.7 to 28%). The method limits of quantitation (MLOQs) for nine bisphenols in sewage and sludge samples were 0.0007–16.3 ng L−1 and 0.0004–8.28 ng g−1 dry weight (dw), respectively. The method was applied to a survey of a municipal wastewater treatment plant (WWTP) in Dalian. All of the tested bisphenols, except BPB and BPZ, were presented in the analyzed samples. BPA, BPS, and BPF with the concentrations 412, 109 and 66.4 ng L−1 in the WWTP influent, respectively, were the predominant bisphenols. The results demonstrated that BPS and BPF have become the most frequently used substitutes of BPA. © 2018 Elsevier B.V. All rights reserved.

1. Introduction Bisphenols (BPs) are a class of chemicals with similar structure widely used in the plastics manufacturing industry, mainly including bisphenol A (BPA), bisphenol S (BPS), bisphenol F (BPF), bisphenol E (BPE), bisphenol B (BPB), bisphenol AF (BPAF), bisphenol (BPAP) and bisphenol Z (BPZ). The release of BPs into the food and environmental has caused wide concern due to its potential health risks [1–5]. The adverse effects of BPA on reproductive [6,7], immune [8,9] and central nervous [10,11] systems have been well documented. Comparative estrogenic activities have also been reported for BPS, BPF, BPE, BPB and BPAF [12–14]. Recent studies shown that BPF and BPS (the most important substitutes of BPA

∗ Corresponding authors. E-mail addresses: [email protected], [email protected] (X. Sun), [email protected] (J. Chen). https://doi.org/10.1016/j.chroma.2018.04.004 0021-9673/© 2018 Elsevier B.V. All rights reserved.

[15,16]) are not necessarily safer and there is a need to remove all of the bisphenols from consumer merchandise [17–20]. The occurrence of BPA in environmental matrices, human samples and foodstuffs has been abundantly reported [21–24]. BPA was found in 92.6% of the 2517 participants in the United States (U.S.) [25]. Limited studies have shown that other bisphenols were also detected in matrices such as river water [26], sediment [15], wastewater [27], indoor dust [28], milk [29] and soft drink [30]. BPS, BPF and BPAF has been found not only in river water and sediment [26,31] but also in indoor dust [28]. BPB was detected in canned foods in European countries [32,33]. High concentration of BPS was identified in currency bills and paper products [34]. BPF was reported to be the predominant substitute of BPA in foodstuffs [35]. Due to the ultra-low concentration of bisphenols in environmental samples, high sensitive detection instruments and efficient sample pretreatment techniques are both indispensable. At present, solid-phase extraction (SPE) coupled with

X. Sun et al. / J. Chromatogr. A 1552 (2018) 10–16

LC–MS/MS detection is the most frequently used method. The (U) HPLC–ESI–MS/MS with MRM mode provided a highly effective method for the quantitative determination of bisphenols [26,31,36]. Sample preparation based on hydrophilic–hydrophobic balance (HLB) [27], mixed-mode anionic exchange (MAX) and mixed-mode cationic exchange (MCX) [31,37] sorbents were previously reported as effective clean-up methods before LC–MS/MS analysis of BPs. However, significant signal suppressions for BPs detection were observed when HLB or HLB + MAX were used for sample preparation of sewage and sludge samples [26]. Such signal suppressions were caused by co-elution of matrix components which have influences on the ionization efficiency of BPs. Therefore, highly selective sample preparation method able to remove or minimize the co-elution components was needed. Molecularly imprinted polymers (MIPs) are tailor-made materials with high affinity and selectivity for template molecules [38,39]. Molecularly imprinted solid-phase extraction (MISPE) is the most widely used area of MIPs. The high selectivity of the MIPs made them suitable for enriching of ultra-trace analytes in complex matrices. Until now, MISPE used in sample pretreatment of food, biological and environmental samples were extensively reported [40,41]. However, inherent drawbacks of MIPs such as template bleeding and low class-selectivity, limited theirs application in real sample analysis. In our previous work, highly class-selective MIPs for bisphenol analogues were prepared by using 1,1,1-Tris(4hydroxyphenyl)ethane (THPE) and phenolphthalein (PP) as dummy templates [42,43]. Sample preparation methods based on THPE-DMISPE were demonstrated to have great potential in complex sample pretreatment including sediment and human urine. Since dummy templates were used, the THPE-DMISPE method was free from template bleeding problem, and can be used in the routine analysis of trace BPs. The previous works, however, focused on the methodology, and BPs cannot be detected in real samples due to the low sensitivity of HPLC detection. In this work, the THPE-DMISPE procedure was first used for sample pretreatment of real sewage and sludge samples in the HPLC–MS/MS analysis. Matrix effects in the detection of BPs were carefully studied and compared with the commercial SPE sorbents. The linearity, accuracy, precision, and sensitivity of the developed DMISPE-HPLC–MS/MS method were evaluated. Finally, the optimized method was applied to the quantitation of nine BPs in sewage and sludge samples collected from a WWTP in Dalian, China. 2. Experimental 2.1. Chemicals and reagents Bisphenol A (BPA), bisphenol S (BPS), bisphenol F (BPF), bisphenol E (BPE), bisphenol B (BPB), bisphenol AP (BPAP), bisphenol AF (BPAF), bisphenol Z (BPZ), 1,1,1-Tris(4-hydroxyphenyl)ethane (THPE), trifluoroacetic acid (TFA) and ethylene dimethacrylate (EGDMA) were supplied by J&K Chemical Ltd. Chemical Reagent Co. (Beijing, China). Methacrylic acid (MAA), 4-Vinylpyridine (4VP) and 2,2 -azobisisobutyronitrile (AIBN) were purchased from Acros (NJ, USA). 13 C12 -labeled BPA and TBBPA were obtained from Cambridge Isotope Laboratories (Andover, MA). The methanol, ace® tonitrile and water used were LiChrosolv hypergrade for LC–MS (Merck KGaA, Germany) and formic acid was purchased from Fisher Scientific (NJ, USA). 2.2. Preparation of SPE column Dummy molecularly imprinted polymer was synthesized by the method described previously [43]. Briefly, THPE was used

11

as the dummy template, with 4-vinylpridine and acetonitrile as functional monomer and polymerization solvent, respectively. SPE cartridges with a 3 mL volume were packed with 200 mg of the THPE-DMIP sorbents and used for sample pretreatment. 2.3. Sample collection Four sewage samples (Sewage 1–4) and one sludge sample (Sludge 1) were collected from a WWTP in Dalian, China. The sampling locations in the WWTP are shown in Fig. S1. To inhibit the microbial activity, Formaldehyde (1%, v/v) was added to each sample immediately. Samples were then sealed in glass jars and transported to the laboratory at room temperature within a short time (5–10 min) for further processing. Water samples were filtered with a 0.45 ␮m glass membrane and stored at −20 ◦ C. Sludge sample was freeze-dried, homogenized, and held at −20 ◦ C until analysis. 2.4. Sample pretreatment 2.4.1. Sewage samples After adjusting to pH 9.0 using sodium hydroxide solution, sewage samples (100 mL of Sewage 1, 300 mL of Sewage 2/3/4) were spiked with 20 ng of 13 C12 -BPA and 13 C12 -TBBPA internal standards and percolated through the THPE-DMIP cartridges (preconditioned with 3 mL acetonitrile and 3 mL water) at a flow rate of 3 mL min−1 . The cartridges were then vacuum-dried for 30 min and selectively washed with 3 mL acetonitrile to remove interferences. Bisphenols were finally eluted using 12.0 mL of methanol. The eluate was evaporated under a stream of high purity nitrogen gas and reconstituted in 1.0 mL with methanol–water (50:50, v/v) for HPLC–MS/MS analysis. 2.4.2. Sludge sample Freeze-dried sludge sample (0.2 g) was spiked with 20 ng of 13 C -BPA and 13 C -TBBPA internal standards and allowed to 12 12 stand at room temperature (∼15 ◦ C) for 24 h (in dark). The sludge sample was then extracted with 5 mL methanol–water (pH = 12.0) mixture (5:3, v/v) by ultrasound for 5 min and shaking for 30 min. After centrifugation at 4500g for 5 min, the supernatant was collected and transferred into a glass tube. The extraction process was repeated twice. The extracts were combined and evaporated to ∼4 mL under a stream of N2 . After diluted to 10 mL with water, the extract was adjusted to pH 9.0 and loaded onto the THPE-DMIP column. The column was then rinsed with 3 mL water and vacuumdried for 30 min. After drying, the column was further washed with 3 mL acetonitrile and eluted with 12 mL methanol. The eluate was dried under a stream of nitrogen and reconstituted in 1.0 mL with methanol–water (50:50, v/v) for HPLC–MS/MS analysis. 2.5. Instrument and analytical conditions Sample analysis was performed using a TSQ Quantum Access MAX mass spectrometer coupled with an Accela HPLC System (Thermo Fisher System, San Jose, CA, USA). HPLC separation was conducted on a Hypersil GOLD C18 column (150 × 2.1 mm; 3 ␮m). The mobile phase consisted of methanol (A) and water (B). The gradient program with a flow rate of 200 ␮L min−1 was as follows: 35% A to 100% A (25 min), held for 5 min. The injection volume was 10 ␮L. The mass spectrometric analysis was performed in negative ion mode (ESI) with multiple-reaction monitoring (MRM). The capillary voltage was maintained at −2.5 kV. Source and de-solvation chamber temperature were both at 300 ◦ C. Nitrogen gas was used as the cone and de-solvation gas at 5 psi and 20 psi, respectively. The MRM transitions, collision energy and tube lens value used for

12

X. Sun et al. / J. Chromatogr. A 1552 (2018) 10–16

Table 1 Selected MRM transitions and optimized potentials of the target compounds. Compound

Molecular weight

Precursor (m/z) [M−H]−

(MW) BPS BPF BPE BPA BPA-13C12 BPB BPAP BPAF BPZ TBBPA TBBPA-13C12

250.27 200.23 214.26 228.29 243.29 242.32 290.36 336.23 268.35 543.87 558.87

248.9 199.1 213.0 227.2 239.2 241.1 289.0 335.1 267.0 542.7 554.9

Quantitation

Confirmation

Tube lens (V)

Product (m/z)

CE (eV)

Product (m/z)

C (eV)

108.1 93.3 198.0 212.0 224.2 212.0 274.0 264.9 222.9 445.7 296.7

29 22 20 21 21 20 22 24 34 31 34

92.2 105.1 196.9 133.2 139.2 225.9 195 197.0 173.1 290.6 430.8

40 22 30 30 30 20 35 38 27 34 31

each of the bisphenols in quantitative determination are summarized in Table 1. The schematic molecular structures of BPs analyzed are shown in Fig. S2.

2.6. Quality control The procedural blank was assessed with every batch of samples, and trace of BPS, BPA, TBBPA and BPAF (approximately 0.01, 0.01, 0.21 and 0.02 ng mL−1 , respectively) were found. A midpoint calibration standard (20 ng for each bisphenols) and methanol were injected after every 10 samples as a check for the drift in instrumental sensitivity and carry-over of target analytes between samples, respectively. Ultra-low level of BPAF was found (approximately 0.0002 ng mL−1 ) when acetonitrile- isopropanol (1:1, v/v) was used as the needle washing solvent.

2.7. Method validation The reliability of the method was carefully studied in terms of linearity, precision, recovery and sensitivity. To estimate linearity, seven-point internal standard calibration curves (BPF, BPE: 1.0, 2.0, 5.0, 10.0, 20.0, 50.0, 100 ng mL−1 ; BPA, BPB, BPZ: 0.5, 2.0, 5.0, 10.0, 20.0, 50.0, 100 ng mL−1 ; BPAP, TBBPA: 0.2, 0.5, 2.0, 5.0, 20.0, 50.0, 100 ng mL−1 ; BPS, BPAF: 0.005, 0.05, 0.5, 5.0, 20.0, 50.0, 100 ng mL−1 ) were obtained using 13 C12 -BPA and 13 C12 -TBBPA as internal standards. Each calibration curve was analyzed individually by using weighed (1/x) linear regression. For determining the intra-day precision and accuracy, five replicate analyses (3 samples each time) of sewage and sludge samples spiked with bisphenols (5 ng and 20 ng) were performed on the same day. The background concentrations of nine bisphenols in the original samples were determined at the same time and subtracted from the total concentrations to calculate the absolute recovery. The precision of the method was determined by calculating the relative standard deviation (RSD %) at each spiking level. The inter-day trueness and precision were calculated by performing five replicate analyses (3 samples each time) on five consecutive working days. The measurement uncertainty at a confidence level of 95% (a coverage factor k = 2) was evaluated systematically according to ISO/TS 21748:2010. In correspondence to the strategy proposed by Song et al. [36], matrix effects (MEs) were evaluated using the formula ME (%) = (1 − Am /A0 ) × 100%, where A0 represents the peak area of a pure standard solution and Am represents the peak area of a sample extract spiked at the same level of the standard (n = 5). The MLOQ defined as the analyte concentration with a signal-to-noise ratio (S/N) of 10.

88 80 85 88 88 80 90 88 95 100 100

3. Results and discussion 3.1. Optimization of sample preparation ESI–MS has been frequently used in the quantitative analysis of trace environmental pollutants due to its high specificity and sensitivity. However, Ion suppression or enhancement resulted from the presence of matrix compounds has often been reported. Such matrix effects were caused by the presence of matrix compounds, which have big influence on the ionization efficiency of the target analytes in the ESI. Thus, high efficient purification of complex environmental samples is necessary to enhance the accuracy and precision of ESI–MS analysis.

3.1.1. Washing and elution conditions In previous work, SPE methods for cleanup of BPs in sediment and river water samples were developed in our lab using HPLC detection. Both high cleanup efficiency and recovery were achieved by using acetonitrile and methanol–TFA (98:2, v/v) as washing and elution solvents. However, due to the complex of the sewage and sludge sample matrices, it is necessary for further validation and optimization of the washing and elution conditions. After adjusting to pH 9.0 using 0.2 mol L−1 sodium hydroxide, sewage samples (100 mL STP influent) were spiked with 20 ng of mixed internal standards and passed through THPE-DMIP columns (pre-conditioned with 3 mL acetonitrile and 3 mL water) at a flow rate of 3 mL min−1 . Columns were cleaned with different volumes of acetonitrile (3.0 mL, 4.0 mL and 5.0 mL) after vacuum-dried for 30 min. The results indicate that, high recoveries of BPs (>85.6%) can be achieved both when 3 mL and 4 mL of acetonitrile were used, 5 mL washing of acetonitrile will cause low recovery of TBBPA with a value of 64.8% (Fig. S3). Finally, 3 mL acetonitrile was selected as the washing solvent, to avoid possible loss of BPs caused by nonspecific adsorption of interfering substances. The elution condition was further optimized by comparing the matrix effects of the samples eluted by 6 mL methanol–TFA (98:2, v/v) and 12 mL methanol, which can both guarantee the BPs recovery (>87.8%). The results of matrix effects are shown in Fig. 1, lower matrix suppression (3.8–28.4%) were found when 12 mL methanol was used. This may due to the high elution ability of methanol–TFA (98:2, v/v) for the impurities which can form strong acid-base interaction with 4-VP monomer, causing more matrix effect in the HPLC–MS/MS analysis of BPs. So, 12 mL methanol was used as the elution solvent in this work. Using the above optimized washing and elution condition, matrix effects of STP influent (100 mL of sewage 1), STP effluent (300 mL of sewage 4) and activated sludge (0.2 g, d.w.) were all tested and values of 3.8–28%, −6.7 to 8.4% and 5.2–15% were found respectively (Table 2), much lower than other reported work [26].

X. Sun et al. / J. Chromatogr. A 1552 (2018) 10–16

13

Table 2 Matrix effect of STP effluent, STP influent and activated sludge after THPE-DMISPE process. Sample

STP influent STP effluent Activated sludge

Matrix effect (%) BPS

BPF

BPE

BPA

BPB

BPAP

BPAF

BPZ

TBBPA

6.3 ± 0.2 8.4 ± 0.1 8.2 ± 0.2

3.8 ± 0.5 −5.3 ± 0.7 5.2 ± 0.8

8.3 ± 0.6 8.3 ± 0.7 15 ± 1.5

18 ± 1.7 −6.7 ± 0.9 13 ± 1.9

7.5 ± 0.5 4.7 ± 0.7 15 ± 1.4

13 ± 1.4 2.6 ± 0.2 8.6 ± 1.0

13 ± 0.6 3.8 ± 0.2 11 ± 0.4

28 ± 3.6 −0.1 ± 0.1 14 ± 0.4

20 ± 1.4 0.5 ± 0.1 14 ± 1.3

Fig. 2. Recoveries of bisphenols from sludge sample using different extraction solvents. Fig. 1. Influence of elution condition on matrix effects of bisphenols detection (mean ± SD, n = 3).

3.1.2. Sludge extraction The extraction conditions including solvent, time, extraction mode and extraction cycles were optimized. Good recovery and reproducibility was found when sludge samples were extracted by ultrasound for 5 min and shaking for 30 min (3 times repeats). Meanwhile, different extraction solvents including methanol, acetone, methanol-acetone mixture and methanol-water mixture were tested. The results indicate that sludge samples extracted with methanol, acetone and methanol-acetone mixture have good recoveries for all BPs (>73.7%), but unfortunately, causing serious problems in the subsequent SPE process. As described in Section 2.4.2, the combined extracts were concentrated to ∼4 mL and diluted to 10 mL with ultrapure water before loading on to the THPE-DMIP cartridges in order to provide a reverse phase solid-phase extraction mode. A large number of insoluble substances were generated simultaneously when water was added, and such insoluble substances were difficult to remove by filtration or centrifugation. Further study indicated that this problem can be solved using methanol-water mixture as extraction solvent, proportions including methanol: water (9:1, v/v), methanol: water (4:1, v/v), methanol: water (7:3, v/v), methanol: water (5:3, v/v) and methanol: water (3:2, v/v), were tested. Methanol: water (5:3, v/v) was finally selected to thoroughly avoid this problem and obtain high recoveries as far as possible. Additionally, methanol: water (5:3, v/v) (pH = 12) was used in order to provide an alkaline environment which conducive to the dissociation of BPs (pKa = 8.13–10.3) [42]. After dissociation, the molecular polarities of BPs increased and thus improve the extraction efficiency of the BPs (51.3–96.7%), especially for TBBPA. The detailed information of recoveries using different extraction solvents was shown in Fig. 2. Due to the difficulty in prepare samples with insoluble substances (do not separate under 12,000 rpm/more than 10 filters was need for just one sample), only methanol-acetone (1:1, v/v) and methanol: water (4:1, v/v) were compared with methanol: water (5:3, v/v) and methanol: water (5:3, v/v) (pH = 12).

Fig. 3. Comparison of the matrix effect by using different SPE process (mean ± SD, n = 3).

3.2. Comparison of DMISPE with commercial SPE sorbents In order to prove the real superiority of the developed DMISPE method for the selective enrichment of BPs, the results obtained with THPE-DMIP sorbents were compared with those obtained with commercial SPE sorbents. The commercially available SPE cartridges tested were a hydrophilic-hydrophobic balance column (HLB) and a hydrophilic-hydrophobic balance column coupled with a strong anion exchanger column (HLB + MAX). The testing of the commercial SPE cartridges has been assessed under their own optimum conditions reported by the literature [26]. STP influent (100 mL) spiked with 20 ng of mixed internal standards was used in this experiment. As shown in Fig. 3, sample cleaned by HLB cartridge has much higher matrix suppression (23.6–60.2%) than HLB + MAX (12.8–47.3%) and THPE-DMIP (3.8–28.4%). The low matrix suppression of THPE-DMIP was contributed to its ultra-high selectivity toward BPs, which allow removing most of the matrix interferences during the ACN washing

14

X. Sun et al. / J. Chromatogr. A 1552 (2018) 10–16

Table 3 Recoveries, intra-day precision, and MLOQ of bisphenols in environmental samples (n = 5). Analyte

BPS BPF BPE BPA BPB BPAP BPAF BPZ TBBPA

Spiking level (ng)

5 20 5 20 5 20 5 20 5 20 5 20 5 20 5 20 5 20

effluent

influent

Sludge

Rec. (%) a

Intra-pre. (%) b

MLOQ c

Rec. (%) a

Intra-pre. (%) b

MLOQ c

Rec. (%) a

Intra-pre. (%)3) b

MLOQ c

101 98.6 86.7 88.4 90.2 93.9 88.6 92.5 92.3 95.0 86.4 90.7 93.5 92.5 82.2 85.1 94.9 96.7

3.0 4.2 10 9.1 8.9 6.1 5.8 3.9 7.7 8.3 6.9 7.1 4.0 3.7 10 5.5 3.7 3.4

0.002

83.8 85.7 92.4 89.6 90.3 87.4 94.1 96.2 87.8 86.9 94.5 94.0 85.9 87.0 85.2 99.7 97.0 98.6

4.6 1.2 8.0 15 5.7 4.6 3.2 5.6 7.8 12 4.4 5.8 2.1 5.1 11 6.9 2.9 3.7

0.007

91.3 90.9 86.7 94.4 90.5 96.7 88.4 92.3 86.3 94.5 93.6 91.9 91.0 89.3 83.6 90.6 43.6 51.3

4.5 1.5 10 13 8.1 2.8 9.3 7.0 13 7.5 3.7 5.3 4.2 2.8 9.2 4.3 3.5 4.8

0.003

4.97 1.14 0.98 2.20 0.56 0.0002 0.52 0.05

16.3 3.42 3.84 6.79 1.80 0.0007 2.55 0.20

8.28 1.85 1.80 3.69 0.64 0.0004 0.93 0.09

Recoveries; b intra-day precision, calculated as the relative standard deviation (n = 5); c the method limits of quantitation, the units of MLOQ are ng L−1 for water and ng g−1 (d.w.) for sludge. a

Table 4 Concentrations of bisphenols in sewage and sludge samples collected from a WWTP in Dalian, China. Samples/number

Sewage 1 Sewage 2 Sewage 3 Sewage 4 sludge

BPs in sewage (ng L−1 ) or sludge (ng g−1 d.w.) BPS

BPF

BPE

BPA

BPB

BPAP

BPAF

BPZ

TBBPA

109 79.0 55.9 11.9 3.98

66.4 45.2 23.7
9.39 5.66
412 360 212 30.0 63.6


1.16 0.83
1.03 0.78 1.45 0.48 0.26

ND ND ND ND ND

3.78 3.55 0.68 0.60 0.93

ND: not detected; Sewage 1: influent; Sewage 2: S3D effluent; Sewage 3: DN effluent; Sewage 4: effluent (Fig. 2).

step efficiently. As a result, low matrix effect was found during the HPLC–MS/MS analysis. Moreover, compared to HLB + MAX process, THPE-DMISPE reduced not only the time but also the cost. 3.3. Method performance To evaluate the applicability of the proposed THPE-DMISPE method for real sample analysis, the recovery and precision were determined using WWTP influent, effluent and sludge samples spiked at two concentration levels (5 and 20 ng). The results are listed in Table 3. The linearity of the established method was estimated with correlation coefficient greater than 0.99 for all BPs. The HPLC–MS/MS chromatograms of a spiked STP influent sample with a spiking level of 20 ng are shown in Fig. 4. The recoveries of BPs ranged from 82.2% to 101% for the sewage samples and from 43.6% to 96.7% for the sludge sample, better than recently reported work [27]. The low recovery of TBBPA in the sludge sample was due to the low extraction efficiency caused by its strong hydrophobicity (log Kow value for TBBPA was 9.7), which enhance its binding to the solid matrix. The precision of the method, expressed as the relative standard deviation (RSD%), was quantified by performing five spiked samples during the same day (interday precision) and on five different days (intra-day precision). The intra-day precision was less than 15%, and the inter-day precision was less than 13%. The measurement uncertainty was calculated and the results are shown in Table S1. The relative uncertainty of BPs at different spiking level did not exceed 15% at a confidence level of 95%. The instrument limits of quantitation were estimated using a signal-to-noise ratio of 10. The method limits of quantitation

(MLOQs) in the reference matrix were then calculated based on the instrument detection limits, recoveries, enrichment factors and matrix effect. The MLOQs ranged from 0.0002 to 16.3 ng L−1 in sewage samples and from 0.0004 to 8.28 ng g−1 (d. w.) in sludge samples. The above results indicate that THPE-DMIP served as an efficient SPE sorbent for class-selective extraction of bisphenols from complex matrices with outstanding practicality and superiority.

3.4. Method application The developed procedure was applied to the analysis of 9 bisphenols in sewage and sludge samples collected from a local WWTP. Concentrations of BPs measured are summarized in Table 4. All of the tested BPs, except BPB and BPZ, were present in the analyzed samples. In these samples, BPA, BPS and BPF were the predominant bisphenols. In particular, BPA occurred at high levels, with 412 ng L−1 and 63.6 ng g−1 (d.w.) in the influent and sludge sample, respectively. According to the concentrations of bisphenols detected in the influent and effluent samples, the total removal rates of BPS, BPA, BPAF and TBBPA were calculated, and values of 89.2%, 92.7%, 53.4% and 84.1% were achieved, respectively. Such values were close to the previous reported literature [44]. In the case of BPA, the removal rates of the SEDIPAC S3D, BIOFOR DN and BIOFOR CN units were 12.6%, 35.8% and 44.4%, respectively. Although the removal efficiency of the bisphenols is high, there is still a release of bisphenols to the environment which might cause long term effects to the ecosystem. Further research is needed to improve the removal effi-

X. Sun et al. / J. Chromatogr. A 1552 (2018) 10–16

15

Fig. 4. HPLC–MS/MS chromatograms of a spiked STP influent sample (20 ng for each bisphenol).

ciency of BPA and its structural analogues, especially for BPF and BPS. 4. Conclusion An efficient method based on THPE-DMISPE for selective determination of nine BPs in sewage and sludge samples coupled with HPLC–MS/MS detection was developed. The proposed method shows satisfactory precision, good recoveries and adequate limits of detection for BPs in complex samples. Superiority of the clean-up efficiency was also proved by comparing the matrix effect with commercial SPE sorbents. This method was then successfully applied to the determination of BPs in sewage and sludge samples collected from a WWTP in Dalian, China. High concentration of BPA, BPS and BPF were found in the influent sample, indicate that BPS and BPF has become the most frequently used substitutes of BPA. Acknowledgements This research was supported by the National Natural Science Foundation of China (Grant Nos. 21607067, 41502326, 81603266) and the CAS Key Laboratory of Separation Sciences for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.chroma.2018.04. 004.

References [1] F.S. vom Saal, C. Hughes, An extensive new literature concerning low-dose effects of bisphenol A shows the need for a new risk assessment, Environ. Health Perspect. 113 (2005) 926–933. [2] L.N. Vandenberg, T. Colborn, T.B. Hayes, J.J. Heindel, D.R. Jacobs Jr., D.H. Lee, T. Shioda, A.M. Soto, F.S. vom Saal, W.V. Welshons, R.T. Zoeller, J.P. Myers, Hormones and endocrine-disrupting chemicals: low-dose effects and nonmonotonic dose responses, Endocr. Rev. 33 (2012) 378–455. [3] L.N. Vandenberg, I. Chahoud, J.J. Heindel, V. Padmanabhan, F.J.R. Paumgartten, G. Schoenfelder, Urinary, circulating, and tissue biomonitoring studies indicate widespread exposure to bisphenol A, Environ. Health Perspect. 118 (2010) 1055–1070. [4] Y.Q. Huang, C.K. Wong, J.S. Zheng, H. Bouwman, R. Barra, B. Wahlstrom, L. Neretin, M.H. Wong, Bisphenol A (BPA) in China: a review of sources, environmental levels, and potential human health impacts, Environ. Int. 42 (2012) 91–99. [5] D. Chen, K. Kannan, H. Tan, Z. Zheng, Y.L. Feng, Y. Wu, M. Widelka, Bisphenol analogues other than BPA: environmental occurrence, human exposure, and toxicity—a review, Environ. Sci. Technol. 50 (2016) 5438–5453. [6] F.S. Vom Saal, P.S. Cooke, D.L. Buchanan, P. Palanza, K.A. Thayer, S.C. Nagel, S. Parmigiani, W.V. Welshons, A physiologically based approach to the study of bisphenol a and other estrogenic chemicals on the size of reproductive organs daily sperm production, and behavior, Toxicol. Ind. Health 14 (1998) 239–260. [7] M.V. Maffini, B.S. Rubin, C. Sonnenschein, A.M. Soto, Endocrine disruptors and reproductive health: the case of bisphenol-A, Mol. Cell. Endocrinol. 254–255 (2006) 179–186. [8] C. Sawai, K. Anderson, D. Walser-Kuntz, Effect of bisphenol A on murine immune function: modulation of interferon-␥, IgG2a, and disease symptoms in NZB × NZW F1 mice, Environ. Health Perspect. 111 (2003) 1883–1887. [9] E.M. Clayton, M. Todd, J.B. Dowd, A.E. Aiello, The impact of bisphenol A and triclosan on immune parameters in the U.S. population, NHANES 2003–2006, Environ. Health Perspect. 119 (2011) 390–396. [10] J. Zeng, H. Kuang, C. Hu, X. Shi, M. Yan, L. Xu, L. Wang, C. Xu, G. Xu, Effect of bisphenol A on rat metabolic profiling studied by using capillary electrophoresis time-of-flight mass spectrometry, Environ. Sci. Technol. 47 (2013) 7457–7465. [11] S. Matsuda, D. Matsuzawa, D. Ishii, H. Tomizawa, J. Sajiki, E. Shimizu, Perinatal exposure to bisphenol A enhances contextual fear memory and affects the

16

[12] [13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23] [24]

[25]

[26]

[27]

[28]

X. Sun et al. / J. Chromatogr. A 1552 (2018) 10–16 serotoninergic system in juvenile female mice, Horm. Behav. 63 (2013) 709–716. M. Chen, M. Ike, M. Fujita, Acute toxicity, mutagenicity, and estrogenicity of bisphenol-A and other bisphenols, Environ. Toxicol. 17 (2002) 80–86. Y. Li, K.A. Burns, Y. Arao, C.J. Luh, K.S. Korach, Differential estrogenic actions of endocrine-disrupting chemicals bisphenol A, bisphenol AF, and zearalenone through estrogen receptor alpha and beta in vitro, Environ. Health Perspect. 120 (2012) 1029–1035. A. Macczak, M. Cyrkler, B. Bukowska, J. Michalowicz, Eryptosis-inducing activity of bisphenol A and its analogs in human red blood cells (in vitro study), J. Hazard. Mater. 307 (2016) 328–335. C. Liao, F. Liu, H.B. Moon, N. Yamashita, S. Yun, K. Kannan, Bisphenol analogues in sediments from industrialized areas in the United States, Japan, and Korea: spatial and temporal distributions, Environ. Sci. Technol. 46 (2012) 11558–11565. C. Liao, F. Liu, H. Alomirah, V.D. Loi, M.A. Mohd, H.B. Moon, H. Nakata, K. Kannan, Bisphenol S in urine from the United States and seven Asian countries: occurrence and human exposures, Environ. Sci. Technol. 46 (2012) 6860–6866. L. Peyre, P. Rouimi, G. de Sousa, C. Helies-Toussaint, B. Carre, S. Barcellini, M.C. Chagnon, R. Rahmani, Comparative study of bisphenol A and its analogue bisphenol S on human hepatic cells: a focus on their potential involvement in nonalcoholic fatty liver disease, Food Chem. Toxicol. 70 (2014) 9–18. C. Helies-Toussaint, L. Peyre, C. Costanzo, M.C. Chagnon, R. Rahmani, Is bisphenol S a safe substitute for bisphenol A in terms of metabolic function? An in vitro study, Toxicol. Appl. Pharmacol. 280 (2014) 224–235. C.D. Kinch, K. Ibhazehiebo, J.H. Jeong, H.R. Habibi, D.M. Kurrasch, Low-dose exposure to bisphenol A and replacement bisphenol S induces precocious hypothalamic neurogenesis in embryonic zebrafish, Proc. Natl. Acad. Sci. U. S. A. 112 (2015) 1475–1480. S. Eladak, T. Grisin, D. Moison, M.J. Guerquin, T. N’Tumba-Byn, S. Pozzi-Gaudin, A. Benachi, G. Livera, V. Rouiller-Fabre, R. Habert, A new chapter in the bisphenol A story: bisphenol S and bisphenol F are not safe alternatives to this compound, Fertil. Steril. 103 (2015) 11–21. Y. Huang, C. Wong, J. Zheng, H. Bouwman, R. Barra, B. Wahlström, L. Neretin, M. Wong, Bisphenol A (BPA) in China: a review of sources, environmental levels, and potential human health impacts, Environ. Int. 42 (2012) 91–99. C.A. Staples, P.B. Dorn, G.M. Klecka, S.T. O’Blook, L.R. Harris, A review of the environmental fate, effects, and exposures of bisphenol A, Chemosphere 36 (1998) 2149–2173. L.N. Vandenberg, R. Hauser, M. Marcus, N. Olea, W.V. Welshons, Human exposure to bisphenol A (BPA), Reprod. Toxicol. 24 (2007) 139–177. S.L. Stacy, M. Eliot, A.M. Calafat, A. Chen, B.P. Lanphear, R. Hauser, G.D. Papandonatos, S. Sathyanarayana, X. Ye, K. Yolton, J.M. Braun, Patterns, variability, and predictors of urinary bisphenol A concentrations during childhood, Environ. Sci. Technol. 50 (2016) 5981–5990. A.M. Calafat, X. Ye, L.Y. Wong, J.A. Reidy, L.L. Needham, Exposure of the U.S. population to bisphenol A and 4-tertiary-octylphenol: 2003–2004, Environ. Health Perspect. 116 (2008) 39–44. Y. Yang, L. Lu, J. Zhang, Y. Yang, Y. Wu, B. Shao, Simultaneous determination of seven bisphenols in environmental water and solid samples by liquid chromatography-electrospray tandem mass spectrometry, J. Chromatogr. A 1328 (2014) 26–34. Q. Sun, Y. Wang, Y. Li, M. Ashfaq, L. Dai, X. Xie, C.P. Yu, Fate and mass balance of bisphenol analogues in wastewater treatment plants in Xiamen City, China, Environ. Pollut. 225 (2017) 542–549. C. Liao, F. Liu, Y. Guo, H.B. Moon, H. Nakata, Q. Wu, K. Kannan, Occurrence of eight bisphenol analogues in indoor dust from the United States and several

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39] [40]

[41]

[42]

[43]

[44]

Asian countries: implications for human exposure, Environ. Sci. Technol. 46 (2012) 9138–9145. L. Grumetto, O. Gennari, D. Montesano, R. Ferracane, A. Ritieni, S. Albrizio, F. Barbato, Determination of five bisphenols in commercial milk samples by liquid chromatography coupled to fluorescence detection, J. Food Prot. 76 (2013) 1590–1596. H. Gallart-Ayala, E. Moyano, M.T. Galceran, Analysis of bisphenols in soft drinks by on-line solid phase extraction fast liquid chromatography-tandem mass spectrometry, Anal. Chim. Acta 683 (2011) 227–233. C. Liao, F. Liu, H.B. Moon, N. Yamashita, S. Yun, K. Kannan, Bisphenol analogues in sediments from industrialized areas in the United States, Japan, and Korea: spatial and temporal distributions, Environ. Sci. Technol. 46 (2012) 11558–11565. L. Grumetto, D. Montesano, S. Seccia, S. Albrizio, F. Barbato, Determination of bisphenol A and bisphenol B residues in canned peeled tomatoes by reversed-phase liquid chromatography, J. Agric. Food Chem. 56 (2008) 10633–10637. S.C. Cunha, C. Almeida, E. Mendes, J.O. Fernandes, Simultaneous determination of bisphenol A and bisphenol B in beverages and powdered infant formula by dispersive liquid-liquid micro-extraction and heart-cutting multidimensional gas chromatography-mass spectrometry, Food Addit. Contam. Part A Chem. Anal. Control Expo Risk Assess. 28 (2011) 513–526. C. Liao, F. Liu, K. Kannan, Bisphenol s, a new bisphenol analogue, in paper products and currency bills and its association with bisphenol a residues, Environ. Sci. Technol. 46 (2012) 6515–6522. C. Liao, K. Kannan, Concentrations and profiles of bisphenol A and other bisphenol analogues in foodstuffs from the United States and their implications for human exposure, J. Agric. Food Chem. 61 (2013) 4655–4662. S. Song, M. Song, L. Zeng, T. Wang, R. Liu, T. Ruan, G. Jiang, Occurrence and profiles of bisphenol analogues in municipal sewage sludge in China, Environ. Pollut. 186 (2014) 14–19. S. Lee, C. Liao, G.J. Song, K. Ra, K. Kannan, H.B. Moon, Emission of bisphenol analogues including bisphenol A and bisphenol F from wastewater treatment plants in Korea, Chemosphere 119 (2015) 1000–1006. F.G. Tamayo, E. Turiel, A. Martin-Esteban, Molecularly imprinted polymers for solid-phase extraction and solid-phase microextraction: recent developments and future trends, J. Chromatogr. A 1152 (2007) 32–40. L. Chen, X. Wang, W. Lu, X. Wu, J. Li, Molecular imprinting: perspectives and applications, Chem. Soc. Rev. 45 (2016) 2137–2211. A. Beltran, F. Borrull, R.M. Marcé, P.A.G. Cormack, Molecularly-imprinted polymers: useful sorbents for selective extractions, TrAC Trends Anal. Chem. 29 (2010) 1363–1375. A. Martín-Esteban, Molecularly-imprinted polymers as a versatile, highly selective tool in sample preparation, TrAC Trends Anal. Chem. 45 (2013) 169–181. X. Sun, J. Wang, Y. Li, J. Jin, B. Zhang, S.M. Shah, X. Wang, J. Chen, Highly selective dummy molecularly imprinted polymer as a solid-phase extraction sorbent for five bisphenols in tap and river water, J. Chromatogr. A 1343 (2014) 33–41. X. Sun, J. Wang, Y. Li, J. Jin, J. Yang, F. Li, S.M. Shah, J. Chen, Highly class-selective solid-phase extraction of bisphenols in milk, sediment and human urine samples using well-designed dummy molecularly imprinted polymers, J. Chromatogr. A 1360 (2014) 9–16. A.S. Stasinakis, G. Gatidou, D. Mamais, N.S. Thomaidis, T.D. Lekkas, Occurrence and fate of endocrine disrupters in Greek sewage treatment plants, Water Res. 42 (2008) 1796–1804.