- Email: [email protected]
Synthesis of an imprinted polymer for the determination of methylmercury in marine products
Author’s Accepted Manuscript Synthesis of an imprinted polymer for the determination of methylmercury in marine products Roi Rodríguez-Fernández, Elena Peña-Vázquez, Pilar Bermejo-Barrera www.elsevier.com/locate/talanta
PII: DOI: Reference:
S0039-9140(15)30075-8 http://dx.doi.org/10.1016/j.talanta.2015.06.028 TAL15705
To appear in: Talanta Received date: 19 December 2014 Revised date: 9 June 2015 Accepted date: 13 June 2015 Cite this article as: Roi Rodríguez-Fernández, Elena Peña-Vázquez and Pilar Bermejo-Barrera, Synthesis of an imprinted polymer for the determination of methylmercury in marine products, Talanta, http://dx.doi.org/10.1016/j.talanta.2015.06.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Synthesis of an imprinted polymer for the determination of methylmercury in marine products
Roi Rodríguez-Fernández, Elena Peña-Vázquez and Pilar Bermejo-Barreraa,* Department of Analytical Chemistry, Nutrition and Bromatology, Faculty of Chemistry, University of Santiago de Compostela, Avenida de las Ciencias s/n, E-15782, Santiago de Compostela, Spain.
ABSTRACT A molecularly imprinted polymer was synthesized using the precipitation method with methylmercury chloride as the template, phenobarbital as ligand, methacrylic acid (MMA) as monomer, and ethylene glycoldimethacrylate (EDMA) as cross-linking agent. The MIP was characterized using elemental analysis, infrared spectroscopy, energy dispersive X-ray fluorescence and scanning electron microscopy. The operating conditions for solid phase extraction (SPE) were optimized in column mode (pH, loading and elution flow rate using 1M thiourea in 1M HCl). The polymer was used for analyzing the toluene extracts of two reference materials (BCR-463 and TORT-2) with good accuracy. Keywords. Methylmercury, molecularly imprinted polymer (MIP), SPE, HRCSAAS, fish
!"#$ %
1.Introduction The number of imprinted polymers developed for environmental remediation or solid phase extraction (SPE) previous to mercury analysis, mainly in water, has increased sharply in recent years. Most of these polymers were synthesized using Hg(II) as template and complexing ligands to produce selective binding sites. The ligands usually contained a sulfur donating atom such as methacryloyl-(L)-cysteine (MAC) [1], aminothiol monomers [2,3], 1-(2-thiazolylazo)-2-naphthol [4],4-(2-thiazolyazo) resorcinol (TAR) [5] and diphenylthiocarbazone [6,7]. Compounds containing amino groups were also used for developing imprinted polymers for Hg(II) removal from aqueous solutions or preconcentration: 3-isocyanatopropyltriethoxysilane (IPTS) [8] bearing thymine (T) bases, diazoaminobenzene-vinylpyridine copolymers[9], tetrakis(3-hydroxyphenyl)porphyrin[10] or a polyaminated chitosan derivative [11]. Other natural materials have been used as supports for ion imprinting and adsorption of mercury (II), e.g. crop stalks [12] or cellulosic cotton fibers[13]. Monier and AbdelLatif [14] also synthesized ion-imprinted chelating fibers based on poly(ethylene terephthalate) for selective removal of Hg2+. The surface imprinting technique has been used because it provides good accessibility to the analyte and low-mass transfer resistance [15]. Thus, several authors have developed organic-inorganic hybrid materials using thiol-functionalized mesoporous sorbents and methanesulfonic acid [16-18] or cetyltrimethylammonium bromide (CTAB) as a second template to improve the efficiency of the polymer [19,20]. Dakova et al. [21] used silica gel modified with 3-(trimethoxysilyl)propyl methacrylate (TSPM) as supporting material to synthesize a core-shell type imprinted polymer. The outer layer included methacrylic acid (MAA) as monomer and the complexes of Hg(II) with pyrrolidine dithiocarbamate (PDC) or 1-(2-thiazolyazo)-2-naphthol (TAN) as templates.
The sorbent was used for the speciation of mercury in samples of wine, with or without digestion. Najafi et al. [22] coated Fe3O4 magnetic nanoparticles with an ion-imprinting polymer based on N-(pyridine-2-ylmethyl)ethenamine, and these particles were used for the determination of low levels of Hg(II) in fish samples. The imprinting technique was also used for the development of modified carbon electrodes for Hg(II) [23-27], an optical sensor using 9-vinylcarbazole as fluorescent probe [28] and imprinted photonic polymers [29, 30]. Even though the number of the studies appearing in the literature has increased sharply in recent years, only a few studies deal with the development of imprinted polymers and methylmercury as a template [31,32]. Buޠyuޠktiryaki et al.[31] used the dispersion polymerization technique to synthesize imprinted beads that were used as SPE support for the determination of methylmercury and mercury ions in LUTs (a non-defatted lobster hepatopancreas certified reference material) and in several spiked synthetic seawaters. The methylmercury-methacryloyl-(L)-cysteine (MM-MAC) complex was used as monomer, and ethylene dimethacrylate (EDMA) was used as crosslinking agent. Liu et al.[32] used precipitation polimerization to synthesize a methylmercuryimprinted polymer with (4-ethenylphenyl)-4-formate-6-phenyl-2,2'-bipyridine, divinylbenzene (DVB) as crosslinking agent and 2,2ƍ-azobisisobutyronitrile (AIBN) as initiator. They packed the polymer in columns and used it for the determination of methylmercury in aqueous and biological samples (human hair). Recently, an international global treaty to reduce emissions and release of mercury was signed, but the amount of the element in the environment is still increasing [33]. The European Food Safety Authority (EFSA) updated its scientific advice on mercury in food in December 2012. A new Tolerably Weekly Intake (TWI) was established for inorganic mercury (4 µg/kg body weight) and for methylmercury (1.3 µg/kg body
weight, expressed as mercury)[34]. Methylmercury is the predominant form of mercury in seafood (fish and shellfish), and affects the development of the nervous system; this is the reason why unborn children are the most vulnerable group, especially if the mother consumes large amounts of fish. Inorganic mercury is less toxic and can also be found in fish and seafood as well as ready-made meals. The Dietetic Products, Nutrition and Allergy (NDA) Panel published in July 2014 a Scientific Opinion on health benefits of seafood consumption in relation to health risks associated with exposure to methylmercury[35]. Seafood is a source of essential nutrients such as iodine, selenium, calcium, vitamins A and D, and n-3 long-chain polyunsaturated fatty acids (n-3 LCPUFA). The Panel concluded that consumption of about 1-2 servings of seafood per week and up to 3-4 servings per week during pregnancy has been associated with lower risk of coronary heart disease mortality in adults and with better functional neurodevelopment in children, compared to no consumption of seafood. Those amounts agree with the current guidelines in most of the European countries that usually give specific recommendations for toddlers, pregnant and lactating women, taking into account the higher concentration of the contaminants in some species (e.g. swordfish, dogfish, marlin, shark, ray or tuna). The interaction between mercury and phenobarbital is well known because it was used in some classical methods for fast analysis of barbiturates in blood [36], urine [37], plasma and gastric contents [38]. In recent years, amobarbital imprinted microspheres have been synthesized using methacrylic acid (MMA) and EDMA. They have been used for the selective solid-phase extraction of phenobarbital from human urine and medicines [39]. The objective of the present study is the development of a simple procedure to synthesize a methylmercury imprinted polymer using phenobarbital as ligand. The polymer has been characterized using different techniques. The operating
conditions were also studied and the polymer was used for the analysis of methylmercury in two certified reference materials of marine products after an extraction step with toluene.
2. Material and methods 2.1 Instrumentation Mercury was analyzed using a High Resolution Continuum Source Atomic Absorption Spectrometer (HRCSAAS) (Analytik Jena ContrAA 300 model, Jena, Germany), equipped with a flow injection system to perform the vapor generation. The polymer was synthesized using a roll and tilt mixer “Movi-Rod” (Selecta, Barcelona, Spain) placed in a temperature controlled incubation chamber (Boxcult, Selecta). A pH-meter model 720 (ThermoOrion, Waltham, USA) was used to adjust the pHs needed for the experiments. A peristaltic pump (Gilson, Villiers, France) was used in the SPE experiments in column mode. In this case, the polymer was packed into 5 mL SPE cartridges. The elemental analyzer FLASH 1112 from Thermo Finnigan (Waltham, MA, USA) was used to analyze the content of nitrogen, carbon, oxygen, hydrogen and sulfur in the polymer. The EVO LS 15 microscope (Zeiss, Oberkochen, Germany) was used to obtain the micrographs of the MIP and NIP (non imprinted polymer). A labmade spectrometer (Servicios Generales of the University of Santiago de Compostela) with an anode of Mo was used for energy dispersive x-ray dispersion fluorescence measurements. 2.2 Reagents All the solutions were prepared using ultra-pure water of 18ȍcm resistance obtained from a Milli-Q purification device (Millipore Co., Massachusetts, USA).
Methylmercury chloride, MMA, EDMA, phenobarbital and thiourea were supplied by Sigma (Steinhelm, Germany). The stock standard solution (1000 mg L-1) of Hg(II) was from Merck (Darmstadt, Germany). AIBN was purchased from Fluka (Steinhelm, Germany). Tuna fish (BCR-463) certified reference material was obtained from the Community Bureau of Reference (Brussels, Belgium), and TORT-2 Lobster hepatopancreas reference material was from the National Research Council of Canada (NRC, Ottawa, Canada). All the other chemicals (e.g. ammonia for the preparation of the buffer, hydrochloric acid, acetonitrile, toluene …) were purchased from Panreac and Scharlau (Barcelona, Spain). All glass and plastic material was cleaned and kept in 10% (w/w) nitric acid for at least 48 h. The material was then rinsed three times with ultra-pure water before use. 2.3 Synthesis of the Molecularly Imprinted Polymer The precipitation method was used for the preparation of approximately 1.5 g of polymer: 0.075 mmol of MeHgCl and 0.3 mmol of phenobarbital were weighed and mixed with 0.75 mmol of MAA in a clean glass tube. A volume of 12 mL of the porogen (acetonitrile:water 4:1) was added and stirred for 5 min with a vortex. The prepolymerization mixture was kept in the dark overnight. In case of precipitate formation, the solution was either filtered or decanted. After placing the tube in an ice bath, 4.5 mmol of the cross-linker EDMA and 0.25 mmol of AIBN (initiator) were added. The mixture was stirred again for 1 min and purged with argon before closing the tube. Afterwards, the tube was set in the temperature-controlled incubator chamber on the low-profile roller at 60 C, and the polymerization was completed after 24 h. The nonimprinted polymer (NIP) was synthesized following the same procedure but without adding the methylmercury template. The polymers obtained were filtered, washed with
acetonitrile:water 4:1 and dried at room temperature. The resulting polymers are in a white powder format and are easily packed in cartridges to work in column mode instead of batch mode. 2.4 Template removal procedure Portions of MIP (150 mg) were packed in 5 mL syringes between Teflon frits, and the template was completely removed after cleaning with 200 mL of an acidic thiourea solution (1M thiourea in 1M HCl) at a flow rate of 1 mL.min-1. Eluates were analyzed by HRCSAAS to check the complete removal of the template. 2.5 Solid phase extraction procedure The reference materials (BCR-463 Tuna fish and TORT-2 Lobster Hepatopancreas) were analyzed after methylmercury extraction using a modification of the Kwasniak et al. method [40]. Portions of the materials (200 mg) were weighted and transferred to glass centrifuge tubes. Each portion was washed with 5 mL of acetone [41], shaking manually during 15 seconds. Afterwards, a volume of 2.5 mL of 6M HCl was added, and the mixture was shaken during 1 min. Finally, toluene (2.5 mL) was added to perform the extraction of methylmercury. The tubes were sonicated for 30 min at 60°C, and the extracts were centrifuged at 3500 rpm at room temperature for 30 min. The organic fraction was collected and the extraction was repeated using a fresh portion of 2.5 mL of toluene. The sample was sonicated 15 min and centrifuged again. Organic fractions were mixed and stored at 4˚C until analysis. SPE cartridges containing the polymer were conditioned with the NH3/NH4+ buffer solution at pH 8.0. Toluene extracts or buffered aqueous samples were loaded at a flow rate of 0.5 mL min-1. The cartridges were washed with the buffer solution after the loading step, and elution was performed using 10 mL acidic thiourea solution (1M
thiourea in 1M HCl) at the same flow rate (0.5 mL min-1). Thiourea extracts were analyzed by HRCSAAS. The procedure used to treat the samples is shown in Fig. 1.
2.6 Determination of Hg(II) and methylmercury by HRCSAAS The analyses were performed by HRCSAAS after the generation of the vapor using a flow injection system. Samples were transported by a 3% (v/v) HCl solution (carrier), and mixed with the reducing solution (0.2% (w/v) NaBH4 stabilized with 0.05% (w/v) NaOH). The reagents were transferred to the gas-liquid separator through a 500 µL reaction loop, and an 8-way Gilson peristaltic pump (Gilson, Villiers, France) equipped with a 3.18 mm i.d. Tygon tube that was used for extracting the waste from the gasliquid separator. The Hg vapor was separated from the liquid mixture and was swept to the quartz cell using a 25 L h-1 Ar flow. The line used for Hg determination was 253.6492 nm; two hundred pixels were registered, and three analytical pixels (central pixel ± 1) were used to calculate the peak volume selected absorbance (AȜȈ). Detector integration time was 45s (300 spectra recordings), and area mode was used. A reference spectrum of 1M thiourea in 1M HCl was used for background correction, and the instrument selected automatically the pixels used for correction in each measurement (dynamic background correction). The operating parameters for HRCSAAS are shown in Table 1.
3. Results and discussions 3.1 Characterization studies Several techniques were used for the characterization of the molecularly imprinted polymer (MIP) with and without the methylmercury template, and the non-imprinted
polymer (NIP): elemental analysis, energy dispersive X-ray fluorescence, and scanning electron microscopy (SEM). 3.1.1 Microanalysis studies Samples of MIP with methylmercury template, MIP without template and NIP were analyzed to determine the percentage of nitrogen, carbon, hydrogen, sulfur and oxygen. Results are shown in Table 2. Both MIP and MIP without template have a higher percentage of nitrogen due to the trapping of phenobarbital, and an increase in the amount of sulfur was also observed in the MIP without template. This variation was due to the presence of the thiourea used for the extraction of the template. 3.1.2 Scanning electron microscopy (SEM) Digital micrographs of MIP and NIP (Fig.2) were obtained using Scanning Electron Microscopy (SEM). The pictures were taken after applying the procedure for the extraction of the template, and aggregates of particles can be observed. There are no appreciable differences between MIP and NIP in the images. 3.1.3 Energy dispersive X-ray fluorescence The results obtained using this technique indicate a total elimination of the template from the MIP after the treatment with 200 mL of acidic thiourea (Fig.3), and the absence of methylmercury in the polymeric matrix of the NIP. 3.2 Optimization of working conditions In this study, the use of the MIP developed as a solid phase extraction support in column mode was possible. Liu et al.[32] used the column mode for the analysis of methylmercury in human hair samples, but they worked in batch mode to analyze the compound in soil samples. We used 1M thiourea in 1M HCl because an efficient solvent is needed for elution of methylmercury or Hg(II). Thus, Singh and Mishra[5] needed to stir their Hg(II)-TAR imprinted polymer two hours with 1M thiourea in 6M
HCl for extracting Hg(II). A similar procedure was used by Buޠyuޠktiryaki et al. [31] to extract the methylmercury template with 1M thiourea in 8M HCl. 3.2.1 Influence of pH on MIP retention The influence of pH on MIP retention was studied after packing 150 mg portions of the imprinted polymer in 5 mL syringes. A volume of 25 mL of 50 ȝg/L MeHg+ standards buffered at pH 6.0, 7.0, 8.0 and 9.0 were loaded at a 0.5 mL.min-1 flow, after the conditioning of the SPE cartridge at the same pH. Syringes were washed after sample loading and eluted with 10 mL of acidic thiourea solution. Experiments were performed in duplicate and the recoveries were calculated. At pH 6.0 the recovery was 82.2%. Results show that MeHg+ recovery was approximately 100% from pH 7.0 onwards (106.2 ± 9.7% at pH 7.0; 108.9 ± 1.8 % at pH 8.0; 107.3 ± 0.3 % at pH 9.0). We selected pH 8.0 to perform all the following experiments because the results were more reproducible at this pH. The pH used for extraction is higher than that used by Liu et al. [32] with their methylmercury imprinted MIP (5.0) using (4-ethenylphenyl)-4-formate-6-phenyl-2,2'bipyridine as ligand. It is also slightly higher than the pH corresponding to the methylmercury imprinted polymer with the ligand MAC (pH 7.0)[31]. 3.2.2 Comparison of extraction of Hg(II) and methylmercury A volume of 25 mL of a 50 ȝg/L Hg(II) standard buffered at pH 8.0 was loaded at a 0.5 mL.min-1 flow, after the conditioning of the SPE cartridge at the same pH. The experiment was performed in duplicate, and the acidic extracts were analyzed by HRCSAAS. The results of the experiments showed that the retention of Hg(II) at pH 8.0 was 83.0 ± 0.7%. Therefore, the imprinted polymer retains both species, mercury and methylmercury. This is the reason why we extracted the methylmercury from the CRMs
using toluene in the subsequent experiments, and we used the polymer to obtain an aqueous phase that can be easily introduced in the HRCSAAS system. 3.2.3 Influence of sample loading flow rate Portions of 0.2 g of the certified reference materials TORT-2 and BCR-483 were treated following a procedure (section 2.5) based on that proposed by Kwasniak et al.[40] In these first experiments, toluene and hydrochloric acid were added at the same time. Toluene extracts were loaded in duplicate at different flows (0.5, 1.0, 2.5 and 5.0 mL min-1). The recoveries obtained for the analysis of BCR-463 were approximately 50% in all cases. As can be observed in Fig.4, the maximum recovery was obtained at 0.5 mL min-1 for TORT-2; therefore, this was the selected flow for loading the sample. 3.2.4 Influence of elution flow rate Afterwards, the influence of the elution flow rate (0.5, 1.0, 2.5 and 5.0 mL min-1) with acidic thiourea (1M thiourea in 1M HCl) was evaluated. The experiments were performed in duplicate, and the maximum recovery for both certified reference materials (BCR-483 and TORT-2) was obtained with an elution flow rate of 1.0 mL min-1 (Fig.5). In the first experiments (sections 3.2.3 and 3.2.4) toluene and hydrochloric acid were added at the same time and stirred simultaneously with the sample. This procedure seems not to be efficient to perform the extraction of methylmercury from BCR-463, taking into account that the concentration of methylmercury in BCR-463 is approximately 20 times higher than in TORT-2.
3.2.5 Centrifugation time Finally, we studied the influence of centrifugation time to separate the toluene extracts. In these extractions, hydrochloric acid was added to the certified reference materials
previously to the addition of toluene. A volume of 2.5 mL of 6M HCl was added to each sample, shaking during 1 min, and finally toluene (2.5 mL) was used to perform the extraction. Experiments were performed in duplicate using centrifugation times of 10, 20 and 30 min, and results showed that recoveries increased till 100% using centrifugation times of 30 minutes (Fig.6). 3.3 Analytical Performance The software used for the control of the HRCSAAS spectrometer allows the selection of the optimum number of pixels for providing the highest sensitivity and reproducibility. All the measurements during the optimization of the method were performed using 3 analytical pixels (central pixel ± 1), as recommended by the Analytik Jena manufacturer. 3.3.1 Calibration graphs Calibration graphs were prepared with concentrations of 0, 2.5, 5.0, 7.5 and 10.0 µg L-1 of methylmercury in 1.0 M thiourea/1M HCl, and 3% (v/v) HCl (the carrier for the flow injection method of analysis) . The following equation was obtained when using 3% (v/v) HCl as solvent to prepare the standards: A = 0.0547 [Hg(MeHg+)]-0.0118, r = 0.998. In the case of methylmercury in 0.5 M thiourea/1M HCl the equation was: A = 0.0795 [Hg(MeHg+)]-0.0204 , r= 0.997. The slopes of both calibration graphs were observed to be statistically different. We used the calibration in 1.0 M thiourea/1M HCl during all the subsequent experiments. 3.3.2 Limit of detection and quantification The limit of detection (LOD) is defined as the element concentration corresponding to three times the standard deviation of the measurement of a blank (n =11), and the limit of quantification (LOQ) is calculated as ten times the standard deviation of the measurement of a blank. A solution of 3% (v/v) hydrochloric acid was used as a blank. Limits of detection were calculated using the central pixel (CP), 3 pixels (CP ± 1) and 5
pixels (CP ± 2), and the values obtained are shown in Table 3. The best results were obtained using the central pixel (LOD = 6.6 ȝg/Kg, LOQ = 22.0 ȝg/Kg). The values obtained are very small in comparison with the limit value established by European Legislation for mercury in fish (1 mg/Kg) [42]. Moreover, it is assumed that from 60 to 90% of mercury is present in fish as methylmercury [43]. 3.3.3 Reproducibility of the method The relative standard deviation of (RSD(%)) of eleven measurements of a 10 µg L-1 standard was used to estimate the repeatability of the method. Results obtained using 1, 3 or 5 pixels are listed in Table 3. All the values obtained were very similar, ranging from 6.6 to 6.9%. The polymer was also used repeatedly in at least 5-10 cycles of SPE without major changes in its performance. 3.3.4 Sorption capacity of the polymer The sorption capacity of the polymer was studied in toluene and in an aqueous solution (pH 8.0). Solutions of toluene (5 mL) containing amounts of methylmercury (as mercury) ranging from 0.25 to 1.5 µg (50-300 µg L-1) were loaded in the cartridges containing 150 mg of the MIP, and no changes were observed in the loading capacity. The study was also performed after loading buffer solutions (pH 8.0) with concentrations of methylmercury ranging from 0.25 to 5 ȝg (50-1000 µg L-1). A decrease in the recovery was observed from 3 µg (600 µg L-1) of methylmercury onwards. Using this value for the calculation, the capacity of the MIP sorbent would decrease from 20 µg g-1 of polymer onwards. The loading capacity was also calculated for the non-imprinted polymer (NIP). A volume of 5 mL of a 500 µg L-1 solution of methylmercury dissolved in toluene (5 µg), and only 71% of the methylmercury was retained in the NIP, while 100% of the
compound was retained in the MIP. Therefore, there is an imprinting effect added to other type of interactions that affects the retention of the analyte. 3.3.5 Accuracy The tuna fish Certified Reference Material BCR-463 and the lobster hepatopancreas CRM TORT-2 were used to evaluate the accuracy of the SPE method in the determination of methylmercury. A modification of the method (section 2.5) developed by Kwaniak et al.[40] was used for the extraction of methylmercury in toluene. The CRMs were analyzed in triplicate, and results revealed an agreement (t-test, P=0.05) between the experimental concentrations of methylmercury and the certified values. The experimental concentration for BCR-463 was 3.14 ± 0.20 µg/g, and the certified value was 3.04 ± 0.16 µg/g. In the case of TORT-2, the experimental value was 0.155 ± 0.06 µg/g, and the certified value was 0.152 ± 0.013 µg/g of methylmercury (expressed as mercury).
4. Conclusion A molecularly imprinted polymer using methylmercury as a template, phenobarbital as ligand, MMA as monomer and EDMA as cross-linker, was synthesized. The MIP was characterized by elemental analysis, energy dispersive x-ray fluorescence and scanning electron microscopy. The polymer presents versatile operating characteristics in aqueous and organic media (toluene), and works in column mode. The operating conditions were optimized (centrifugation time, loading and elution flow rates), the analytical characteristics were studied, and the material was used to analyze methylmercury in two CRMs of tuna fish and lobster hepatopancreas with good accuracy.
Acknowledgements The authors are grateful for the financial support provided by the Xunta de Galicia (project number: 10PXIB209032PR).
Table 1 Operating parameters for HRCSAAS Vapor generation system Step
Pump 2 (mL min-1) 6
Waste
Time
Reading
Load
Pump 1 (mL min-1) 5
Sample
10
-
Auto zero
0
6
Sample
10
Yes
Reaction
5
6
Carrier
20
Yes
Washing
0
6
Sample
35
Yes
Spectrophotometer Current /A
13
Spectral range /pixels
200
Analytical line for Hg/nm
253.6492
Evaluated pixels
1 or 3 (CP ± 1)
Background correction mode
With reference
Background correction fit
Dynamic
Read time / s
45
Integration mode
Area
Number of spectra
300
Temperature of the quartz cell
150
Table 2 Elemental composition of the polymers Element
MIP (with template)
MIP (without template)
NIP
Carbon
57.19 %
55.26 %
55.74 %
Hydrogen
7.42 %
7.23 %
7.43%
Nitrogen
0.62 %
0.74 %
0.20 %
Sulfur
-
2.08 %
-
Oxygen
22.2 %
20.92 %
22.58 %
Table 3 Analytical performance Number of pixels
LOD (µg L-1)
LOD (µg Kg-1)
RSD(%)
CP
0.13
6.61
6.9
CP ± 1 píxel
0.32
16.0
6.8
CP ± 2 píxeles
0.55
27.7
6.6
Figures Fig.1.Diagram showing the procedure to analyze methymercury Fig.2. Micrographs of a) MIP without methylmercury template b) NIP Fig.3. Energy dispersive X-ray fluorescence a) MIP with methylmercury template b) MIP without template Fig. 4. Optimization of SPE conditions: Influence of loading flow rate Fig. 5. Optimization of SPE conditions: Influence of elution flow rate Fig. 6. Influence of centrifugation time
HIGHLIGHTS - A simple procedure was used to synthesize an MIP with methylmercury and phenobarbital - The polymer was characterized and the conditions for operation were studied - It was tested with aqueous and organic (toluene) solutions - Two CRMs of tuna and lobster hepatopancreas were analyzed using the MIP
[1] M. Andac&, S. Mirel, S. S&enel, R. Say, A. Erso'z, A. Denizli, Int. J. Biol. Macrom. 40 (2007) 159-166. [2] M. Firouzzare, Q. Wang, Talanta 101 (2012) 261-266. [3]M. Firouzzare, Adv. Mater. Res. 718-720 (2013) 645-649. [4] I. Dakova, I. Karadjova, V. Georgieva, G. Georgiev, Talanta 78 (2009) 523-529. [5] D.K. Singh, S. Mishra, Desalination 257 (2010) 177-183. [6] A. Ray, S.N. Gupta, J. Polym. Sci. A: Polym. Chem. 35 (1997) 3729–3734. [7] S. Ashraf, A. Cluley, C. Mercado, A. Mueller, Water Sci. & Technol. 64 (2011) 1325–1332. [8] S. Xu, L. Chen, J. Li, Y. Guan, H. Lu, J. Hazard. Mater. 237-238 (2012) 347-54. [9] Y. Liu, X. Chang, D. Yang, Y. Guo, S. Meng, Anal. Chim. Acta 538 (2005) 85-91. [10] M. Shamsipur, H.R. Rajabi, M.H. Beyzavi, H. Sharghi, Microchim. Acta 180 (2013) 791-799
[11] X. Tang, D. Niu, C. Bi, B. Shen, Ind. Eng. Chem. Res. 52 (2013) 13120–13127. [12] J. Wang, Faming Zhuanli Shenqing (2014) CN 103769059 A 20140507. [13] M. Monier, I.M. Kenawy, M.A. Hashem, Carbohydr. Polym. 106 (2014) 49-59. [14] M. Monier, D.A. Abdel-Latif, Chem. Eng. J. 221 (2013) 452-460. [15] C. Branger, W. Meouche, A. Margaillan, React. Funct. Polym. 73 (2013) 859-875. [16] Z. Fan, Talanta 70 (2006) 1164-1169. [17] Z. Wang, G. Wu, C. He, Microchim. Acta (2009) 151-157. [18] B.S. Batlokwa, L. Chimuka, Z. Tshentu, E. Cukrowska, N. Torto, Water SA 38 (2012) 255-260. [19] G. Wu, Z. Wang, J. Wang, C. He, Anal. Chim. Acta 582(2007) 304-310.. [20] D. Zhao, Y. Wang, G. Lin, H. Zhang, G. Li, S. Wang, Guangpu Shiyanshi 30 (2013) 1555-1561. [21] I. Dakova, T. Yordanova, I. Karadjova, J. Hazard. Mater. 231–232 (2012) 49-56. [22] E. Najafi, F. Aboufazeli, H.R. Lotfi Zadeh Zhad, O. Sadeghi, V. Amani, Food. Chem. 141 (2013) 4040-4045. [23] M.R. Ganjali, T. Alizadeh, F. Azimi, B. Larjani, F. Faridbod, P. Norouzi, Int. J. Electrochem. Sci. 6 (2011) 5200-5208. [24] T. Alizadeh, M.R. Ganjali, M. Zare, Anal. Chim. Acta, 689 (2011) 52-59. [25] X.C. Fu, X. Chen, Z. Guo, C.G. Xie, L.T. Kong, J.H. Liu, X.J. Huang, Anal. Chim. Acta 685 (2011) 21-28.
[26] X.C. Fu, J. Wu, L. Nie, C.G. Xie, J.H. Liu, X.J. Huang, Anal. Chim. Acta 720 (2012) 29-37. [27] H.R. Rajabi, M. Roushani, M. Shmsipur, J. Electroanal. Chem. 693 (2013) 16-22. [28] O. Güney, F.Ç. Cebeci, J. Appl. Polym. Sci. 117 (2010) 2373–2379. [29] X. Hu, Imprinted photonic polymers and methods for their preparation and use. PCT Int. Appl. (2013) WO 2013063772 A1 20130510. [30] X. Hu, W. Hong, B. Zhao, Method for preparing photonic crystal thin film based on ion imprinted technology, Faming Zhuanli Shenqing (2014) CN 103804600 A 20140521. [31] S. Bu'yu'ktiryaki, R. Say, A. Denizli, A. Erso'z, Talanta 71 (2007) 699-705. [32] Y. Liu, Y. Zai, X. Chang, Y. Guo, S. Meng, F. Feng, Anal. Chim. Acta 575 (2006) 159-165. [33] D.P. Krabbenhoft, E.M. Sunderland, Science, 341 (2013) 1457-1458. [34] EFSA Panel on Contaminants in the Food Chain (CONTAM), Scientific Opinion on the risk for public health related to the presence of mercury and methylmercury in food. EFSA Journal 2012; 10 (12):2985 (241 pages). http://www.efsa.europa.eu/en/efsajournal/doc/2985.pdf Last visited: 08/06/15. [35] EFSA Dietetic Products, Nutrition, and Allergies (NDA), Scientific Opinion on health benefits of seafood (fish and shellfish) consumption in relation to health risks associated with exposure to methylmercury, EFSA Journal 2014;12(7):3761 (80 pages). http://www.efsa.europa.eu/en/efsajournal/doc/3761.pdf Last visited: 08/06/15. [36] A. S. Curry, Br. Med. J. 1 (1964) 354-355.
[37] F. Pehr, Clin. Chem. 21 (1975) 1609-1611. [38] E. Z. Helman, Clin. Chem.16 (1970) 797-798. [39] S.G. Hu, S.W. Wang, S.W. He, Analyst 128 (2003) 1485-1489. [40] J. KwaĞniak, L. Falkowska, M. KwaĞniak, Food Chem. 132 (2012) 752-758. [41] AOAC. Metals and other elements. Chapter 9 of Official Methods of Analysis (2005), p. 40 [42] COMMISSION REGULATION (EC) No 629/2008 of 2 July 2008 amending Regulation (EC) No 1881/2006 setting maximum levels for certain contaminants in foodstuffs, Journal of the European Union L 173/6-9. [43] United Nations Environment Programme. Global mercury assessment, December 2002. http://www.unep.org/gc/gc22/Document/UNEP-GC22-INF3.pdf Last visited: 08/06/15.
*Graphical Abstract (for review)
MIP For aqueous and organic samples
Figure 1
! "#! $% &'
%
" .$
' 00 $& /' $& .$
&,-
#! ( #! ) * + (
Figure 2
a
b
Figure 3
b
Intensity
Intensity
a
Energy (keV)
Energy (keV)
Figure 4
BCR-463
120
TORT-2
Recovery (%)
100 80 60 40 20 0 0,5
1,0
2,5
5,0 -1
Loading flow rate (mL min )
Figure 5
TORT-2 BCR-463
120
Recovery (%)
100 80 60 40 20 0 0,5
1,0
2,5
5,0 -1
Elution flow rate (mL min )
Figure 6
BCR-463 TORT-2
120,0
% Recovery (%)
100,0 80,0 60,0 40,0 20,0 0,0 10 min
20 min
30 min
PII: DOI: Reference:
S0039-9140(15)30075-8 http://dx.doi.org/10.1016/j.talanta.2015.06.028 TAL15705
To appear in: Talanta Received date: 19 December 2014 Revised date: 9 June 2015 Accepted date: 13 June 2015 Cite this article as: Roi Rodríguez-Fernández, Elena Peña-Vázquez and Pilar Bermejo-Barrera, Synthesis of an imprinted polymer for the determination of methylmercury in marine products, Talanta, http://dx.doi.org/10.1016/j.talanta.2015.06.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Synthesis of an imprinted polymer for the determination of methylmercury in marine products
Roi Rodríguez-Fernández, Elena Peña-Vázquez and Pilar Bermejo-Barreraa,* Department of Analytical Chemistry, Nutrition and Bromatology, Faculty of Chemistry, University of Santiago de Compostela, Avenida de las Ciencias s/n, E-15782, Santiago de Compostela, Spain.
ABSTRACT A molecularly imprinted polymer was synthesized using the precipitation method with methylmercury chloride as the template, phenobarbital as ligand, methacrylic acid (MMA) as monomer, and ethylene glycoldimethacrylate (EDMA) as cross-linking agent. The MIP was characterized using elemental analysis, infrared spectroscopy, energy dispersive X-ray fluorescence and scanning electron microscopy. The operating conditions for solid phase extraction (SPE) were optimized in column mode (pH, loading and elution flow rate using 1M thiourea in 1M HCl). The polymer was used for analyzing the toluene extracts of two reference materials (BCR-463 and TORT-2) with good accuracy. Keywords. Methylmercury, molecularly imprinted polymer (MIP), SPE, HRCSAAS, fish
!"#$ %
1.Introduction The number of imprinted polymers developed for environmental remediation or solid phase extraction (SPE) previous to mercury analysis, mainly in water, has increased sharply in recent years. Most of these polymers were synthesized using Hg(II) as template and complexing ligands to produce selective binding sites. The ligands usually contained a sulfur donating atom such as methacryloyl-(L)-cysteine (MAC) [1], aminothiol monomers [2,3], 1-(2-thiazolylazo)-2-naphthol [4],4-(2-thiazolyazo) resorcinol (TAR) [5] and diphenylthiocarbazone [6,7]. Compounds containing amino groups were also used for developing imprinted polymers for Hg(II) removal from aqueous solutions or preconcentration: 3-isocyanatopropyltriethoxysilane (IPTS) [8] bearing thymine (T) bases, diazoaminobenzene-vinylpyridine copolymers[9], tetrakis(3-hydroxyphenyl)porphyrin[10] or a polyaminated chitosan derivative [11]. Other natural materials have been used as supports for ion imprinting and adsorption of mercury (II), e.g. crop stalks [12] or cellulosic cotton fibers[13]. Monier and AbdelLatif [14] also synthesized ion-imprinted chelating fibers based on poly(ethylene terephthalate) for selective removal of Hg2+. The surface imprinting technique has been used because it provides good accessibility to the analyte and low-mass transfer resistance [15]. Thus, several authors have developed organic-inorganic hybrid materials using thiol-functionalized mesoporous sorbents and methanesulfonic acid [16-18] or cetyltrimethylammonium bromide (CTAB) as a second template to improve the efficiency of the polymer [19,20]. Dakova et al. [21] used silica gel modified with 3-(trimethoxysilyl)propyl methacrylate (TSPM) as supporting material to synthesize a core-shell type imprinted polymer. The outer layer included methacrylic acid (MAA) as monomer and the complexes of Hg(II) with pyrrolidine dithiocarbamate (PDC) or 1-(2-thiazolyazo)-2-naphthol (TAN) as templates.
The sorbent was used for the speciation of mercury in samples of wine, with or without digestion. Najafi et al. [22] coated Fe3O4 magnetic nanoparticles with an ion-imprinting polymer based on N-(pyridine-2-ylmethyl)ethenamine, and these particles were used for the determination of low levels of Hg(II) in fish samples. The imprinting technique was also used for the development of modified carbon electrodes for Hg(II) [23-27], an optical sensor using 9-vinylcarbazole as fluorescent probe [28] and imprinted photonic polymers [29, 30]. Even though the number of the studies appearing in the literature has increased sharply in recent years, only a few studies deal with the development of imprinted polymers and methylmercury as a template [31,32]. Buޠyuޠktiryaki et al.[31] used the dispersion polymerization technique to synthesize imprinted beads that were used as SPE support for the determination of methylmercury and mercury ions in LUTs (a non-defatted lobster hepatopancreas certified reference material) and in several spiked synthetic seawaters. The methylmercury-methacryloyl-(L)-cysteine (MM-MAC) complex was used as monomer, and ethylene dimethacrylate (EDMA) was used as crosslinking agent. Liu et al.[32] used precipitation polimerization to synthesize a methylmercuryimprinted polymer with (4-ethenylphenyl)-4-formate-6-phenyl-2,2'-bipyridine, divinylbenzene (DVB) as crosslinking agent and 2,2ƍ-azobisisobutyronitrile (AIBN) as initiator. They packed the polymer in columns and used it for the determination of methylmercury in aqueous and biological samples (human hair). Recently, an international global treaty to reduce emissions and release of mercury was signed, but the amount of the element in the environment is still increasing [33]. The European Food Safety Authority (EFSA) updated its scientific advice on mercury in food in December 2012. A new Tolerably Weekly Intake (TWI) was established for inorganic mercury (4 µg/kg body weight) and for methylmercury (1.3 µg/kg body
weight, expressed as mercury)[34]. Methylmercury is the predominant form of mercury in seafood (fish and shellfish), and affects the development of the nervous system; this is the reason why unborn children are the most vulnerable group, especially if the mother consumes large amounts of fish. Inorganic mercury is less toxic and can also be found in fish and seafood as well as ready-made meals. The Dietetic Products, Nutrition and Allergy (NDA) Panel published in July 2014 a Scientific Opinion on health benefits of seafood consumption in relation to health risks associated with exposure to methylmercury[35]. Seafood is a source of essential nutrients such as iodine, selenium, calcium, vitamins A and D, and n-3 long-chain polyunsaturated fatty acids (n-3 LCPUFA). The Panel concluded that consumption of about 1-2 servings of seafood per week and up to 3-4 servings per week during pregnancy has been associated with lower risk of coronary heart disease mortality in adults and with better functional neurodevelopment in children, compared to no consumption of seafood. Those amounts agree with the current guidelines in most of the European countries that usually give specific recommendations for toddlers, pregnant and lactating women, taking into account the higher concentration of the contaminants in some species (e.g. swordfish, dogfish, marlin, shark, ray or tuna). The interaction between mercury and phenobarbital is well known because it was used in some classical methods for fast analysis of barbiturates in blood [36], urine [37], plasma and gastric contents [38]. In recent years, amobarbital imprinted microspheres have been synthesized using methacrylic acid (MMA) and EDMA. They have been used for the selective solid-phase extraction of phenobarbital from human urine and medicines [39]. The objective of the present study is the development of a simple procedure to synthesize a methylmercury imprinted polymer using phenobarbital as ligand. The polymer has been characterized using different techniques. The operating
conditions were also studied and the polymer was used for the analysis of methylmercury in two certified reference materials of marine products after an extraction step with toluene.
2. Material and methods 2.1 Instrumentation Mercury was analyzed using a High Resolution Continuum Source Atomic Absorption Spectrometer (HRCSAAS) (Analytik Jena ContrAA 300 model, Jena, Germany), equipped with a flow injection system to perform the vapor generation. The polymer was synthesized using a roll and tilt mixer “Movi-Rod” (Selecta, Barcelona, Spain) placed in a temperature controlled incubation chamber (Boxcult, Selecta). A pH-meter model 720 (ThermoOrion, Waltham, USA) was used to adjust the pHs needed for the experiments. A peristaltic pump (Gilson, Villiers, France) was used in the SPE experiments in column mode. In this case, the polymer was packed into 5 mL SPE cartridges. The elemental analyzer FLASH 1112 from Thermo Finnigan (Waltham, MA, USA) was used to analyze the content of nitrogen, carbon, oxygen, hydrogen and sulfur in the polymer. The EVO LS 15 microscope (Zeiss, Oberkochen, Germany) was used to obtain the micrographs of the MIP and NIP (non imprinted polymer). A labmade spectrometer (Servicios Generales of the University of Santiago de Compostela) with an anode of Mo was used for energy dispersive x-ray dispersion fluorescence measurements. 2.2 Reagents All the solutions were prepared using ultra-pure water of 18ȍcm resistance obtained from a Milli-Q purification device (Millipore Co., Massachusetts, USA).
Methylmercury chloride, MMA, EDMA, phenobarbital and thiourea were supplied by Sigma (Steinhelm, Germany). The stock standard solution (1000 mg L-1) of Hg(II) was from Merck (Darmstadt, Germany). AIBN was purchased from Fluka (Steinhelm, Germany). Tuna fish (BCR-463) certified reference material was obtained from the Community Bureau of Reference (Brussels, Belgium), and TORT-2 Lobster hepatopancreas reference material was from the National Research Council of Canada (NRC, Ottawa, Canada). All the other chemicals (e.g. ammonia for the preparation of the buffer, hydrochloric acid, acetonitrile, toluene …) were purchased from Panreac and Scharlau (Barcelona, Spain). All glass and plastic material was cleaned and kept in 10% (w/w) nitric acid for at least 48 h. The material was then rinsed three times with ultra-pure water before use. 2.3 Synthesis of the Molecularly Imprinted Polymer The precipitation method was used for the preparation of approximately 1.5 g of polymer: 0.075 mmol of MeHgCl and 0.3 mmol of phenobarbital were weighed and mixed with 0.75 mmol of MAA in a clean glass tube. A volume of 12 mL of the porogen (acetonitrile:water 4:1) was added and stirred for 5 min with a vortex. The prepolymerization mixture was kept in the dark overnight. In case of precipitate formation, the solution was either filtered or decanted. After placing the tube in an ice bath, 4.5 mmol of the cross-linker EDMA and 0.25 mmol of AIBN (initiator) were added. The mixture was stirred again for 1 min and purged with argon before closing the tube. Afterwards, the tube was set in the temperature-controlled incubator chamber on the low-profile roller at 60 C, and the polymerization was completed after 24 h. The nonimprinted polymer (NIP) was synthesized following the same procedure but without adding the methylmercury template. The polymers obtained were filtered, washed with
acetonitrile:water 4:1 and dried at room temperature. The resulting polymers are in a white powder format and are easily packed in cartridges to work in column mode instead of batch mode. 2.4 Template removal procedure Portions of MIP (150 mg) were packed in 5 mL syringes between Teflon frits, and the template was completely removed after cleaning with 200 mL of an acidic thiourea solution (1M thiourea in 1M HCl) at a flow rate of 1 mL.min-1. Eluates were analyzed by HRCSAAS to check the complete removal of the template. 2.5 Solid phase extraction procedure The reference materials (BCR-463 Tuna fish and TORT-2 Lobster Hepatopancreas) were analyzed after methylmercury extraction using a modification of the Kwasniak et al. method [40]. Portions of the materials (200 mg) were weighted and transferred to glass centrifuge tubes. Each portion was washed with 5 mL of acetone [41], shaking manually during 15 seconds. Afterwards, a volume of 2.5 mL of 6M HCl was added, and the mixture was shaken during 1 min. Finally, toluene (2.5 mL) was added to perform the extraction of methylmercury. The tubes were sonicated for 30 min at 60°C, and the extracts were centrifuged at 3500 rpm at room temperature for 30 min. The organic fraction was collected and the extraction was repeated using a fresh portion of 2.5 mL of toluene. The sample was sonicated 15 min and centrifuged again. Organic fractions were mixed and stored at 4˚C until analysis. SPE cartridges containing the polymer were conditioned with the NH3/NH4+ buffer solution at pH 8.0. Toluene extracts or buffered aqueous samples were loaded at a flow rate of 0.5 mL min-1. The cartridges were washed with the buffer solution after the loading step, and elution was performed using 10 mL acidic thiourea solution (1M
thiourea in 1M HCl) at the same flow rate (0.5 mL min-1). Thiourea extracts were analyzed by HRCSAAS. The procedure used to treat the samples is shown in Fig. 1.
2.6 Determination of Hg(II) and methylmercury by HRCSAAS The analyses were performed by HRCSAAS after the generation of the vapor using a flow injection system. Samples were transported by a 3% (v/v) HCl solution (carrier), and mixed with the reducing solution (0.2% (w/v) NaBH4 stabilized with 0.05% (w/v) NaOH). The reagents were transferred to the gas-liquid separator through a 500 µL reaction loop, and an 8-way Gilson peristaltic pump (Gilson, Villiers, France) equipped with a 3.18 mm i.d. Tygon tube that was used for extracting the waste from the gasliquid separator. The Hg vapor was separated from the liquid mixture and was swept to the quartz cell using a 25 L h-1 Ar flow. The line used for Hg determination was 253.6492 nm; two hundred pixels were registered, and three analytical pixels (central pixel ± 1) were used to calculate the peak volume selected absorbance (AȜȈ). Detector integration time was 45s (300 spectra recordings), and area mode was used. A reference spectrum of 1M thiourea in 1M HCl was used for background correction, and the instrument selected automatically the pixels used for correction in each measurement (dynamic background correction). The operating parameters for HRCSAAS are shown in Table 1.
3. Results and discussions 3.1 Characterization studies Several techniques were used for the characterization of the molecularly imprinted polymer (MIP) with and without the methylmercury template, and the non-imprinted
polymer (NIP): elemental analysis, energy dispersive X-ray fluorescence, and scanning electron microscopy (SEM). 3.1.1 Microanalysis studies Samples of MIP with methylmercury template, MIP without template and NIP were analyzed to determine the percentage of nitrogen, carbon, hydrogen, sulfur and oxygen. Results are shown in Table 2. Both MIP and MIP without template have a higher percentage of nitrogen due to the trapping of phenobarbital, and an increase in the amount of sulfur was also observed in the MIP without template. This variation was due to the presence of the thiourea used for the extraction of the template. 3.1.2 Scanning electron microscopy (SEM) Digital micrographs of MIP and NIP (Fig.2) were obtained using Scanning Electron Microscopy (SEM). The pictures were taken after applying the procedure for the extraction of the template, and aggregates of particles can be observed. There are no appreciable differences between MIP and NIP in the images. 3.1.3 Energy dispersive X-ray fluorescence The results obtained using this technique indicate a total elimination of the template from the MIP after the treatment with 200 mL of acidic thiourea (Fig.3), and the absence of methylmercury in the polymeric matrix of the NIP. 3.2 Optimization of working conditions In this study, the use of the MIP developed as a solid phase extraction support in column mode was possible. Liu et al.[32] used the column mode for the analysis of methylmercury in human hair samples, but they worked in batch mode to analyze the compound in soil samples. We used 1M thiourea in 1M HCl because an efficient solvent is needed for elution of methylmercury or Hg(II). Thus, Singh and Mishra[5] needed to stir their Hg(II)-TAR imprinted polymer two hours with 1M thiourea in 6M
HCl for extracting Hg(II). A similar procedure was used by Buޠyuޠktiryaki et al. [31] to extract the methylmercury template with 1M thiourea in 8M HCl. 3.2.1 Influence of pH on MIP retention The influence of pH on MIP retention was studied after packing 150 mg portions of the imprinted polymer in 5 mL syringes. A volume of 25 mL of 50 ȝg/L MeHg+ standards buffered at pH 6.0, 7.0, 8.0 and 9.0 were loaded at a 0.5 mL.min-1 flow, after the conditioning of the SPE cartridge at the same pH. Syringes were washed after sample loading and eluted with 10 mL of acidic thiourea solution. Experiments were performed in duplicate and the recoveries were calculated. At pH 6.0 the recovery was 82.2%. Results show that MeHg+ recovery was approximately 100% from pH 7.0 onwards (106.2 ± 9.7% at pH 7.0; 108.9 ± 1.8 % at pH 8.0; 107.3 ± 0.3 % at pH 9.0). We selected pH 8.0 to perform all the following experiments because the results were more reproducible at this pH. The pH used for extraction is higher than that used by Liu et al. [32] with their methylmercury imprinted MIP (5.0) using (4-ethenylphenyl)-4-formate-6-phenyl-2,2'bipyridine as ligand. It is also slightly higher than the pH corresponding to the methylmercury imprinted polymer with the ligand MAC (pH 7.0)[31]. 3.2.2 Comparison of extraction of Hg(II) and methylmercury A volume of 25 mL of a 50 ȝg/L Hg(II) standard buffered at pH 8.0 was loaded at a 0.5 mL.min-1 flow, after the conditioning of the SPE cartridge at the same pH. The experiment was performed in duplicate, and the acidic extracts were analyzed by HRCSAAS. The results of the experiments showed that the retention of Hg(II) at pH 8.0 was 83.0 ± 0.7%. Therefore, the imprinted polymer retains both species, mercury and methylmercury. This is the reason why we extracted the methylmercury from the CRMs
using toluene in the subsequent experiments, and we used the polymer to obtain an aqueous phase that can be easily introduced in the HRCSAAS system. 3.2.3 Influence of sample loading flow rate Portions of 0.2 g of the certified reference materials TORT-2 and BCR-483 were treated following a procedure (section 2.5) based on that proposed by Kwasniak et al.[40] In these first experiments, toluene and hydrochloric acid were added at the same time. Toluene extracts were loaded in duplicate at different flows (0.5, 1.0, 2.5 and 5.0 mL min-1). The recoveries obtained for the analysis of BCR-463 were approximately 50% in all cases. As can be observed in Fig.4, the maximum recovery was obtained at 0.5 mL min-1 for TORT-2; therefore, this was the selected flow for loading the sample. 3.2.4 Influence of elution flow rate Afterwards, the influence of the elution flow rate (0.5, 1.0, 2.5 and 5.0 mL min-1) with acidic thiourea (1M thiourea in 1M HCl) was evaluated. The experiments were performed in duplicate, and the maximum recovery for both certified reference materials (BCR-483 and TORT-2) was obtained with an elution flow rate of 1.0 mL min-1 (Fig.5). In the first experiments (sections 3.2.3 and 3.2.4) toluene and hydrochloric acid were added at the same time and stirred simultaneously with the sample. This procedure seems not to be efficient to perform the extraction of methylmercury from BCR-463, taking into account that the concentration of methylmercury in BCR-463 is approximately 20 times higher than in TORT-2.
3.2.5 Centrifugation time Finally, we studied the influence of centrifugation time to separate the toluene extracts. In these extractions, hydrochloric acid was added to the certified reference materials
previously to the addition of toluene. A volume of 2.5 mL of 6M HCl was added to each sample, shaking during 1 min, and finally toluene (2.5 mL) was used to perform the extraction. Experiments were performed in duplicate using centrifugation times of 10, 20 and 30 min, and results showed that recoveries increased till 100% using centrifugation times of 30 minutes (Fig.6). 3.3 Analytical Performance The software used for the control of the HRCSAAS spectrometer allows the selection of the optimum number of pixels for providing the highest sensitivity and reproducibility. All the measurements during the optimization of the method were performed using 3 analytical pixels (central pixel ± 1), as recommended by the Analytik Jena manufacturer. 3.3.1 Calibration graphs Calibration graphs were prepared with concentrations of 0, 2.5, 5.0, 7.5 and 10.0 µg L-1 of methylmercury in 1.0 M thiourea/1M HCl, and 3% (v/v) HCl (the carrier for the flow injection method of analysis) . The following equation was obtained when using 3% (v/v) HCl as solvent to prepare the standards: A = 0.0547 [Hg(MeHg+)]-0.0118, r = 0.998. In the case of methylmercury in 0.5 M thiourea/1M HCl the equation was: A = 0.0795 [Hg(MeHg+)]-0.0204 , r= 0.997. The slopes of both calibration graphs were observed to be statistically different. We used the calibration in 1.0 M thiourea/1M HCl during all the subsequent experiments. 3.3.2 Limit of detection and quantification The limit of detection (LOD) is defined as the element concentration corresponding to three times the standard deviation of the measurement of a blank (n =11), and the limit of quantification (LOQ) is calculated as ten times the standard deviation of the measurement of a blank. A solution of 3% (v/v) hydrochloric acid was used as a blank. Limits of detection were calculated using the central pixel (CP), 3 pixels (CP ± 1) and 5
pixels (CP ± 2), and the values obtained are shown in Table 3. The best results were obtained using the central pixel (LOD = 6.6 ȝg/Kg, LOQ = 22.0 ȝg/Kg). The values obtained are very small in comparison with the limit value established by European Legislation for mercury in fish (1 mg/Kg) [42]. Moreover, it is assumed that from 60 to 90% of mercury is present in fish as methylmercury [43]. 3.3.3 Reproducibility of the method The relative standard deviation of (RSD(%)) of eleven measurements of a 10 µg L-1 standard was used to estimate the repeatability of the method. Results obtained using 1, 3 or 5 pixels are listed in Table 3. All the values obtained were very similar, ranging from 6.6 to 6.9%. The polymer was also used repeatedly in at least 5-10 cycles of SPE without major changes in its performance. 3.3.4 Sorption capacity of the polymer The sorption capacity of the polymer was studied in toluene and in an aqueous solution (pH 8.0). Solutions of toluene (5 mL) containing amounts of methylmercury (as mercury) ranging from 0.25 to 1.5 µg (50-300 µg L-1) were loaded in the cartridges containing 150 mg of the MIP, and no changes were observed in the loading capacity. The study was also performed after loading buffer solutions (pH 8.0) with concentrations of methylmercury ranging from 0.25 to 5 ȝg (50-1000 µg L-1). A decrease in the recovery was observed from 3 µg (600 µg L-1) of methylmercury onwards. Using this value for the calculation, the capacity of the MIP sorbent would decrease from 20 µg g-1 of polymer onwards. The loading capacity was also calculated for the non-imprinted polymer (NIP). A volume of 5 mL of a 500 µg L-1 solution of methylmercury dissolved in toluene (5 µg), and only 71% of the methylmercury was retained in the NIP, while 100% of the
compound was retained in the MIP. Therefore, there is an imprinting effect added to other type of interactions that affects the retention of the analyte. 3.3.5 Accuracy The tuna fish Certified Reference Material BCR-463 and the lobster hepatopancreas CRM TORT-2 were used to evaluate the accuracy of the SPE method in the determination of methylmercury. A modification of the method (section 2.5) developed by Kwaniak et al.[40] was used for the extraction of methylmercury in toluene. The CRMs were analyzed in triplicate, and results revealed an agreement (t-test, P=0.05) between the experimental concentrations of methylmercury and the certified values. The experimental concentration for BCR-463 was 3.14 ± 0.20 µg/g, and the certified value was 3.04 ± 0.16 µg/g. In the case of TORT-2, the experimental value was 0.155 ± 0.06 µg/g, and the certified value was 0.152 ± 0.013 µg/g of methylmercury (expressed as mercury).
4. Conclusion A molecularly imprinted polymer using methylmercury as a template, phenobarbital as ligand, MMA as monomer and EDMA as cross-linker, was synthesized. The MIP was characterized by elemental analysis, energy dispersive x-ray fluorescence and scanning electron microscopy. The polymer presents versatile operating characteristics in aqueous and organic media (toluene), and works in column mode. The operating conditions were optimized (centrifugation time, loading and elution flow rates), the analytical characteristics were studied, and the material was used to analyze methylmercury in two CRMs of tuna fish and lobster hepatopancreas with good accuracy.
Acknowledgements The authors are grateful for the financial support provided by the Xunta de Galicia (project number: 10PXIB209032PR).
Table 1 Operating parameters for HRCSAAS Vapor generation system Step
Pump 2 (mL min-1) 6
Waste
Time
Reading
Load
Pump 1 (mL min-1) 5
Sample
10
-
Auto zero
0
6
Sample
10
Yes
Reaction
5
6
Carrier
20
Yes
Washing
0
6
Sample
35
Yes
Spectrophotometer Current /A
13
Spectral range /pixels
200
Analytical line for Hg/nm
253.6492
Evaluated pixels
1 or 3 (CP ± 1)
Background correction mode
With reference
Background correction fit
Dynamic
Read time / s
45
Integration mode
Area
Number of spectra
300
Temperature of the quartz cell
150
Table 2 Elemental composition of the polymers Element
MIP (with template)
MIP (without template)
NIP
Carbon
57.19 %
55.26 %
55.74 %
Hydrogen
7.42 %
7.23 %
7.43%
Nitrogen
0.62 %
0.74 %
0.20 %
Sulfur
-
2.08 %
-
Oxygen
22.2 %
20.92 %
22.58 %
Table 3 Analytical performance Number of pixels
LOD (µg L-1)
LOD (µg Kg-1)
RSD(%)
CP
0.13
6.61
6.9
CP ± 1 píxel
0.32
16.0
6.8
CP ± 2 píxeles
0.55
27.7
6.6
Figures Fig.1.Diagram showing the procedure to analyze methymercury Fig.2. Micrographs of a) MIP without methylmercury template b) NIP Fig.3. Energy dispersive X-ray fluorescence a) MIP with methylmercury template b) MIP without template Fig. 4. Optimization of SPE conditions: Influence of loading flow rate Fig. 5. Optimization of SPE conditions: Influence of elution flow rate Fig. 6. Influence of centrifugation time
HIGHLIGHTS - A simple procedure was used to synthesize an MIP with methylmercury and phenobarbital - The polymer was characterized and the conditions for operation were studied - It was tested with aqueous and organic (toluene) solutions - Two CRMs of tuna and lobster hepatopancreas were analyzed using the MIP
[1] M. Andac&, S. Mirel, S. S&enel, R. Say, A. Erso'z, A. Denizli, Int. J. Biol. Macrom. 40 (2007) 159-166. [2] M. Firouzzare, Q. Wang, Talanta 101 (2012) 261-266. [3]M. Firouzzare, Adv. Mater. Res. 718-720 (2013) 645-649. [4] I. Dakova, I. Karadjova, V. Georgieva, G. Georgiev, Talanta 78 (2009) 523-529. [5] D.K. Singh, S. Mishra, Desalination 257 (2010) 177-183. [6] A. Ray, S.N. Gupta, J. Polym. Sci. A: Polym. Chem. 35 (1997) 3729–3734. [7] S. Ashraf, A. Cluley, C. Mercado, A. Mueller, Water Sci. & Technol. 64 (2011) 1325–1332. [8] S. Xu, L. Chen, J. Li, Y. Guan, H. Lu, J. Hazard. Mater. 237-238 (2012) 347-54. [9] Y. Liu, X. Chang, D. Yang, Y. Guo, S. Meng, Anal. Chim. Acta 538 (2005) 85-91. [10] M. Shamsipur, H.R. Rajabi, M.H. Beyzavi, H. Sharghi, Microchim. Acta 180 (2013) 791-799
[11] X. Tang, D. Niu, C. Bi, B. Shen, Ind. Eng. Chem. Res. 52 (2013) 13120–13127. [12] J. Wang, Faming Zhuanli Shenqing (2014) CN 103769059 A 20140507. [13] M. Monier, I.M. Kenawy, M.A. Hashem, Carbohydr. Polym. 106 (2014) 49-59. [14] M. Monier, D.A. Abdel-Latif, Chem. Eng. J. 221 (2013) 452-460. [15] C. Branger, W. Meouche, A. Margaillan, React. Funct. Polym. 73 (2013) 859-875. [16] Z. Fan, Talanta 70 (2006) 1164-1169. [17] Z. Wang, G. Wu, C. He, Microchim. Acta (2009) 151-157. [18] B.S. Batlokwa, L. Chimuka, Z. Tshentu, E. Cukrowska, N. Torto, Water SA 38 (2012) 255-260. [19] G. Wu, Z. Wang, J. Wang, C. He, Anal. Chim. Acta 582(2007) 304-310.. [20] D. Zhao, Y. Wang, G. Lin, H. Zhang, G. Li, S. Wang, Guangpu Shiyanshi 30 (2013) 1555-1561. [21] I. Dakova, T. Yordanova, I. Karadjova, J. Hazard. Mater. 231–232 (2012) 49-56. [22] E. Najafi, F. Aboufazeli, H.R. Lotfi Zadeh Zhad, O. Sadeghi, V. Amani, Food. Chem. 141 (2013) 4040-4045. [23] M.R. Ganjali, T. Alizadeh, F. Azimi, B. Larjani, F. Faridbod, P. Norouzi, Int. J. Electrochem. Sci. 6 (2011) 5200-5208. [24] T. Alizadeh, M.R. Ganjali, M. Zare, Anal. Chim. Acta, 689 (2011) 52-59. [25] X.C. Fu, X. Chen, Z. Guo, C.G. Xie, L.T. Kong, J.H. Liu, X.J. Huang, Anal. Chim. Acta 685 (2011) 21-28.
[26] X.C. Fu, J. Wu, L. Nie, C.G. Xie, J.H. Liu, X.J. Huang, Anal. Chim. Acta 720 (2012) 29-37. [27] H.R. Rajabi, M. Roushani, M. Shmsipur, J. Electroanal. Chem. 693 (2013) 16-22. [28] O. Güney, F.Ç. Cebeci, J. Appl. Polym. Sci. 117 (2010) 2373–2379. [29] X. Hu, Imprinted photonic polymers and methods for their preparation and use. PCT Int. Appl. (2013) WO 2013063772 A1 20130510. [30] X. Hu, W. Hong, B. Zhao, Method for preparing photonic crystal thin film based on ion imprinted technology, Faming Zhuanli Shenqing (2014) CN 103804600 A 20140521. [31] S. Bu'yu'ktiryaki, R. Say, A. Denizli, A. Erso'z, Talanta 71 (2007) 699-705. [32] Y. Liu, Y. Zai, X. Chang, Y. Guo, S. Meng, F. Feng, Anal. Chim. Acta 575 (2006) 159-165. [33] D.P. Krabbenhoft, E.M. Sunderland, Science, 341 (2013) 1457-1458. [34] EFSA Panel on Contaminants in the Food Chain (CONTAM), Scientific Opinion on the risk for public health related to the presence of mercury and methylmercury in food. EFSA Journal 2012; 10 (12):2985 (241 pages). http://www.efsa.europa.eu/en/efsajournal/doc/2985.pdf Last visited: 08/06/15. [35] EFSA Dietetic Products, Nutrition, and Allergies (NDA), Scientific Opinion on health benefits of seafood (fish and shellfish) consumption in relation to health risks associated with exposure to methylmercury, EFSA Journal 2014;12(7):3761 (80 pages). http://www.efsa.europa.eu/en/efsajournal/doc/3761.pdf Last visited: 08/06/15. [36] A. S. Curry, Br. Med. J. 1 (1964) 354-355.
[37] F. Pehr, Clin. Chem. 21 (1975) 1609-1611. [38] E. Z. Helman, Clin. Chem.16 (1970) 797-798. [39] S.G. Hu, S.W. Wang, S.W. He, Analyst 128 (2003) 1485-1489. [40] J. KwaĞniak, L. Falkowska, M. KwaĞniak, Food Chem. 132 (2012) 752-758. [41] AOAC. Metals and other elements. Chapter 9 of Official Methods of Analysis (2005), p. 40 [42] COMMISSION REGULATION (EC) No 629/2008 of 2 July 2008 amending Regulation (EC) No 1881/2006 setting maximum levels for certain contaminants in foodstuffs, Journal of the European Union L 173/6-9. [43] United Nations Environment Programme. Global mercury assessment, December 2002. http://www.unep.org/gc/gc22/Document/UNEP-GC22-INF3.pdf Last visited: 08/06/15.
*Graphical Abstract (for review)
MIP For aqueous and organic samples
Figure 1
! "#! $% &'
%
" .$
' 00 $& /' $& .$
&,-
#! ( #! ) * + (
Figure 2
a
b
Figure 3
b
Intensity
Intensity
a
Energy (keV)
Energy (keV)
Figure 4
BCR-463
120
TORT-2
Recovery (%)
100 80 60 40 20 0 0,5
1,0
2,5
5,0 -1
Loading flow rate (mL min )
Figure 5
TORT-2 BCR-463
120
Recovery (%)
100 80 60 40 20 0 0,5
1,0
2,5
5,0 -1
Elution flow rate (mL min )
Figure 6
BCR-463 TORT-2
120,0
% Recovery (%)
100,0 80,0 60,0 40,0 20,0 0,0 10 min
20 min
30 min