Ionic imprinted polymer solid-phase extraction for inorganic arsenic selective pre-concentration in fishery products before high-performance liquid chromatography – inductively coupled plasma-mass spectrometry speciation

Elsevier Editorial System(tm) for Journal of Chromatography A Manuscript Draft Manuscript Number: JCA-19-1449R2 Title: Ionic imprinted polymer - solid phase extraction for inorganic arsenic selective pre-concentration in fishery products before high performance liquid chromatography - inductively coupled plasma-mass spectrometry speciation Article Type: Full length article Keywords: Inorganic arsenic speciation, fish, ionic imprinted polymer, HPLC-ICP-MS Corresponding Author: Professor Antonio Moreda-Pineiro, Ph.D Corresponding Author's Institution: University of Santiago de Compostela First Author: Antonio Moreda-Pineiro, Ph.D Order of Authors: Antonio Moreda-Pineiro, Ph.D; B.K. Kolita Kamal Jinadasa; Elena Peña-Vázquez; Pilar Bermejo-Barrera Abstract: Low levels of inorganic arsenic [As(III) and As(V)] in fishery products have been selectively isolated from fish extracts (1 g of wet fish samples pre-treated with 10 mL of 1:1 methanol/water under sonication at 25 °C for 30 min) by ionic imprinted polymer based solid phase extraction procedure (on-column mode). The selective adsorbent was synthesized using sodium (meta) arsenite as a template, 1-vinyl imidazole as a functional monomer, divinylbenzene as a cross-linker, and 2,2'- azobisisobutyronitrile as an initiator. Optimized pre-concentration conditions imply fish extract (10 mL) pH adjustment at 8.5 before loading (flow rate of 0.25 mL min-1), and elution with ultrapure water (2 mL) at 0.25 mL min-1. A pre-concentration factor of 10 was finally obtained after evaporation to dryness (N2 stream) and redissolution in 0.2 mL of ultrapure water before HPLC-ICP-MS. Synthesized material was found to pre-concentrate inorganic arsenic species; whereas organic arsenic compounds, mainly arsenobetaine (the major organoarsenic compound in fish/seafood products), were not found to interact with the adsorbent. The developed selective method gave limits of quantification of 1.05 and 1.31 µg kg-1 for As (III) and As (V), respectively, and good precision [relative standard deviations lower than 12% in fish extracts spiked at several As (III) and As (V) levels]. The proposed method was finally applied to the selective determination of As (III) and As (V) species in several fishery products. Highlights: > Selective determination of low concentrations of As(III) and As(V) in the presence of large concentrations of organoarsenic compounds > Speciation of low concentrations of inorganic arsenic in fishery products > Simple and low cost synthesis of ionic imprinted polymer for solid phase extraction *Highlights (for review) 1 Ionic imprinted polymer solid-phase extraction for inorganic arsenic 1 selective pre-concentration in fishery products before high-performance 2 liquid chromatography – inductively coupled plasma-mass spectrometry 3 speciation 4 Kamal K. Jinadasa, Elena Peña-Vázquez, Pilar Bermejo-Barrera, Antonio Moreda-Piñeiro* 5 Trace Element, Spectroscopy and Speciation Group (GETEE), Strategic Grouping in Materials 6 (AEMAT), Department of Analytical Chemistry, Nutrition and Bromatology. Faculty of 7 Chemistry. Universidade de Santiago de Compostela. Avenida das Ciencias, s/n. 15782, 8 Santiago de Compostela. Spain 9 *[email protected] 10 Abstract 11 Low levels of inorganic arsenic [As(III) and As(V)] in fishery products have been selectively 12 isolated from fish extracts (1.0 g of wet fish samples pre-treated with 10 mL of 1:1 13 methanol/water under sonication at 25 °C for 30 min) by ionic imprinted polymer (IIPs) based 14 solid phase extraction procedure (on-column mode). The selective adsorbent was synthesized 15 using sodium (meta) arsenite as a template, 1-vinyl imidazole as a functional monomer, 16 divinylbenzene as a cross-linker, and 2,2’-azobisisobutyronitrile as an initiator. Optimized pre-17 concentration conditions imply fish extract (10 mL) pH adjustment at 8.5 before loading (flow 18 rate of 0.25 mL min-1), and elution with ultrapure water (2 mL) at 0.50 mL min-1. A pre-19 concentration factor of 50 was finally obtained after evaporation to dryness (N2 stream) and re-20 dissolution in 0.2 mL of ultrapure water before HPLC-ICP-MS. Synthesized material was 21 found to pre-concentrate inorganic arsenic species; whereas organic arsenic compounds, 22 mainly arsenobetaine (the major organoarsenic compound in fish/seafood products), were not 23 found to interact with the adsorbent. The developed selective method gave limits of 24 quantification of 1.05 and 1.31 µg kg-1 for As (III) and As (V), respectively, and good 25 *Manuscript Click here to view linked References 2 precision [relative standard deviations lower than 12% in fish extracts spiked at several As (III) 26 and As (V) levels]. The proposed method was finally applied to the selective determination of 27 As (III) and As (V) species in several fishery products. 28 29 Keywords 30 Inorganic arsenic speciation, fish, ionic imprinted polymer, HPLC-ICP-MS 31 32 1. Introduction 33 Fish and shellfish tend to bio-accumulate pollutants from seawater and have been identified as 34 the major As exposure pathway to humans [1-3]. Several organoarsenic species are generated 35 when inorganic arsenic is metabolized by marine biota and, in addition to un-metabolized 36 inorganic As (III & V), organoarsenic species such as monomethylarsonic acid (MA), 37 dimethylarsinic acid (DMA), arsenobetaine (AB), arsenocholine (AC), trimethylarsine oxide 38 (TMAO) and tetramethylammonium ion (TETRA) are found in marine organisms [4]. 39 Moreover, arsenosugars, as well as arsenolipids, have also been reported as organoarsenic 40 species in seafood products such as seaweed [5]. Among all these As species, inorganic As 41 [arsenite, As (III); and arsenate, As (V)] are the most toxic As forms (10 to 60 times more toxic 42 than other As species) and have been classified as Group I carcinogens [6]. MA and DMA are 43 considered to be moderately toxic and cancer promoters; whereas, AB and AC are innocuous 44 [6]. The maximum lethal dose (MLD-50) for As (III), As (V), MA, DMA, AC, and AB have 45 been established at 14, 20, 700-1800, 700-2600, 6500, and >10,000 mg kg-1 body weight, 46 respectively [2,7]. 47 Arsenic speciation is therefore necessary since toxicity depends on the As form, while AB (the 48 non-toxic species) is the most abundant As species in fish and shellfish [8]. The European 49 Commission (Regulation 2015/1006) has set maximum levels only for inorganic arsenic in rice 50 3 and rice-based products [9]. New sample preparation techniques to allow inorganic As pre-51 concentration, to decrease interferences and to exclude co-elution are therefore needed [10]. In 52 addition, high-performance liquid chromatography (HPLC) is the separation/determination 53 technique of choice for As speciation because of its operational simplicity [11]. The assessment 54 of low levels of inorganic As species in the presence of high concentrations of organoarsenic 55 forms, mainly AB, is troublesome. Small dilution factors and/or the application of a pre-56 concentration stage are therefore required to assess inorganic As forms. However, these 57 methodologies increase the analytical signals from organic As species, which can overlap the 58 inorganic As responses, as usually occurs when performing As speciation with 59 chromatographic methodologies based on anion-exchange conditions. Several sample pre-60 treatments have been developed to assess inorganic As. Recent proposals are based on using 61 liquid-based microextraction techniques [12,13], but most of the developments use solid-phase 62 extraction (SPE) and micro-solid phase extraction (µ-SPE) processes as sample pre-treatments 63 [14-17]. SPE (and also more recently µ-SPE) is one of the most common and popular 64 techniques for target pre-concentration due to its flexibility, speed, simplicity, high degree of 65 safety, and low-cost. In addition, SPE is a well-known environmentally friendly methodology 66 that is easy to automate [17,18]. Advances on SPE imply the use/development of new 67 adsorbents which allow effective target pre-concentration with a small amount of adsorbent 68 and guarantee selectivity. In addition to commercial silica-based strong anion exchange 69 cartridges [14], new nano-based adsorbents have been also proposed [15-17]. 70 The molecular imprinting technique (MIT) has offered an appealing way to prepare highly 71 selective materials. Adsorbents for SPE based on MIT are currently being developed for 72 several organic and inorganic targets, including large biomolecules. Materials prepared using 73 MIT are usually referred to as an ionic imprinted polymer (IIP) when the target (the template 74 used for generating the recognition cavities) is an ion. Difficulties can occur when using ions as 75 templates because these simple structures lack functional groups that can interact with the 76 4 monomer (vinylated reagent). A basic strategy consists of trapping a chelating ligand (a non-77 vinylated reagent) via imprinting of binary/ternary mixed ligand complexes of ions with a non-78 vinylated chelating agent and a vinyl ligand (monomer) [19]. After polymerization, the non-79 vinylated ligand is not chemically bound to the polymer chains, but instead is trapped inside 80 the polymeric matrix and allows the polymer-ion interactions. The selectivity of the prepared 81 IIPs can be enhanced when using bifunctional reagents (a reagent exhibiting chelating 82 properties through the ion, and also containing vinyl groups for polymerization) [19]. 83 The trapping technique using methacrylic acid as a vinylated reagent (monomer) and 84 hydroquinone as a non-vinylated reagent (chelating agent) has been used by Alizadeh and 85 Rashedi for preparing (bulk polymerization), an IIP-based membrane electrode sensitive and 86 selective to As (III) species [20]. Bifunctional reagents such as 1-vinyl imidazole and 4-vinyl 87 pyridine have been proposed by Tsoi et al. for preparing a selective adsorbent for As (V) by 88 bulk polymerization [21]. These two developments, however, are focused on assessing only 89 one inorganic As form, and selectivity through the other inorganic As species (and also 90 organoarsenic species) was not performed/reported. In addition, as is the case with other 91 speciation procedures for inorganic As by SPE/µ-SPE and liquid-based microextraction, 92 oxidation or a reduction stage was required to assess total inorganic As by atomic 93 spectrometry-based techniques [12-17], and potentiometry [20]. The aim of the current 94 proposal has been the development of a highly sensitive and selective method for assessing 95 inorganic As species [As(III & V)] in samples containing high concentrations of organoarsenic 96 species such as fishery products. A SPE method based on a new IIP as an adsorbent combined 97 with HPLC hyphenated with inductively coupled plasma mass spectrometry (ICP-MS) has 98 been developed. The IIP adsorbent was prepared by using the precipitation polymerization 99 method, As (III) as a template, and 1-vinyl imidazole as a bifunctional monomer. IIP-SPE 100 optimization was performed to allow a simultaneous pre-concentration of As (III) and As (V) 101 species for a further determination by HPLC-ICP-MS. 102 5 2. Material and methods 103 2.1. Instrumentation 104 Arsenic speciation was performed with a Flexar LC HPLC (LC pump, column oven, and LC 105 autosampler) from Perkin Elmer (Waltham, MA, USA) hyphenated to a Perkin Elmer Nex-Ion 106 300X ICP–MS. A PRP×100 column (10 µm, 100 × 4.1 mm) connected to a PRP×100 guard 107 column (10 µm, 25 × 2.3 mm) from Hamilton (Reno, NV, USA) was used for inorganic As 108 separation. The same ICP-MS instrument connected to a SeaFast SC2 DX autosampler 109 (Elemental Scientific, Omaha, NB, USA) was used for total As determinations. Synthesized 110 IIP and NIP were characterized by Fourier transform infrared spectrometry with ATR 111 correction using a Spectrum-Two FT-IR from Perkin Elmer, and by scanning electron 112 microscopy (SEM) with a ZEISS EVO LS 15 (Carl Zeiss, Oberkochen, Germany). A low-113 profile roller (Stovall, Greensboro, NC, USA), placed inside a Boxcult temperature-controlled 114 chamber (Stuart Scientific, Surrey, UK), was used for IIP synthesis. Samples were digested 115 (total As determination) using an Ethos Plus microwave accelerated system (Milestone, 116 Sorisole, Italy) with 100-mL closed Teflon vessels. SPE was performed by using an 8-channel 117 Miniplus 3 peristaltic pump (Gilson, Middleton, WI, USA), and PVC 2-stop tubing (1.52 mm 118 i.d.) from SCP Sciences (Baie-D'Urfe, Quebec, Canada). Other equipment included an 119 USC60TH ultrasonic cleaner bath (45 kHz, 120 W) from VWR (Leuven, Belgium), an 120 ultracentrifuge (Sigma 2K15, Osterode, Germany), a pH meter with glass combined electrode 121 (Crison Basic 20, Barcelona, Spain), a domestic Taurus blade grinder (Taurus, Barcelona, 122 Spain), an oven model 207 from Selecta (Barcelona, Spain), and a Classic ML analytical 123 balance (Metler Toledo, Columbus, OH, USA). 124 2.2. Reagents 125 All aqueous solutions were prepared with ultrapure water with a resistivity of 18.2 MΩ cm 126 obtained from a Milli Q-A10 system from Millipore Co. (Billerica, MA, USA). Sodium (meta) 127 6 arsenite (NaAsO2), used as a template for IIP synthesis, was obtained from Sigma-Aldrich (St. 128 Louis, MO, USA). Arsenite and arsenate stock standard solutions, 1000 mg L−1, were from 129 Panreac (Barcelona, Spain). Standard solutions of MA, DMA, AB and AC (1000 mg L−1) were 130 prepared by dissolving the appropriate amounts of MA (CH3AsO(ONa)2·6H2O) purchased 131 from Carlo Erba (Milan, Italy); DMA (C2H6AsNaO2·3H2O), purchased from Merck 132 (Darmstadt, Germany); and AB (AsC5H11O2) and AC (AsC5H14O), both from Tri Chemical 133 Laboratory Inc. (Yamanashi, Japan). Bifunctional monomer 1-vinyl imidazole (≥99.0% purity) 134 and initiator 2,2’-azobisisobutyronitrile (AIBN) were obtained from Fluka (Buchs, 135 Switzerland); whereas, divinylbenzene (DVB) was from Sigma-Aldrich. 1-vinyl imidazole was 136 used directly without a purification stage; whereas, AIBN and DVB were subjected to a 137 purification process before use. AIBN was dissolved in methanol (50–60 °C) and was then re-138 crystallized at –20 °C (AIBN). Regarding DVB, the reagent was passed through a mini-column 139 containing a 0.50 g of neutral alumina (Merck) for removal of impurities. Nitric acid (Hyper 140 pure, 69%), and 33% of hydrogen peroxide were supplied by Panreac (Barcelona, Spain). 141 NexIon Setup Solution (10 μg L-1 of Be, Ce, Fe, In, Li, Mg, Pb, and U in 1% HNO3) was 142 purchased from Perkin Elmer (Shelton, CT, USA). Ammonium chloride, ammonium 143 hydroxide and methanol (Chromasolv) were from Merck; whereas, ammonium hydrogen 144 carbonate and acetic acid (glacial) were from Panreac and ammonium dihydrogen phosphate 145 was from BDH (Poole, United Kingdom). Multi-element standard solutions (cross-reactivity 146 study) were prepared by combining single Ca, Co and Mg stock standard solutions (1000 mg 147 L−1) from Merck, single Cr, Hg, K, P, Pb and Zn stock standard solutions (1000 mg L−1) from 148 Scharlab (Barcelona, Spain), and single Cd, Cu, Fe and Na stock standard solutions (1000 mg 149 L−1) from Perkin Elmer. Internal standard solutions (Ge, Sc, and Rh) were prepared from 150 single-element standards (1000 mg L-1) purchased from Perkin Elmer. DORM–2 (dogfish 151 muscle) and DOLT-3 (dogfish liver) were from National Research Council of Canada (Ottawa, 152 Canada), and BCR 278 (mussel tissue) was from Community Bureau of Reference – 153 7 Commission of the European Communities (Brussels, Belgium). Other consumables were 154 Durapore 0.22 µm membrane filters (Millipore), replacement Teflon frits (Supelco, Bellefonte, 155 PA, USA), and disposable syringes (sterile, 5 mL) from Dispomed (Gelnhausen, Germany). 156 2.3. Sample pre-treatments 157 The following fish samples were used: yellowfin tuna (Thunnus albacares) and swordfish 158 (Xiphias gladius) obtained from Tropic Frozen Foods (Pvt) Ltd (Sri Lanka), bluefin tuna 159 (Thunnus thynnus) and blue shark (Prionace glauca) also as frozen products, and fresh cod 160 (Gadus morhua) bought in local supermarkets in Santiago de Compostela. Approximately 0.2 161 kg of flesh was blended, homogenized and stored in pre-cleaned polyethylene bottles with 162 hermetic seals at -20 °C before use. 163 Fish samples, as well as CRMs were subjected to microwave-assisted acid digestion in 164 triplicate for further total As determination by ICP-MS (electronic supplementary information, 165 ESI). Approximately 1.0 g of homogenized fresh fish samples (0.2 g of dried CRMs) were 166 placed in the microwave reactors and pre-digested with 3.0 mL of 69% (v/v) HNO3 plus 4.0 167 mL of ultrapure water and 1.0 mL of H2O2 for 15 min at room temperature. Pre-digested 168 samples were then subjected to microwave-assisted acid digestion (1st step: from room 169 temperature to 180 °C for 15 min; 2nd step: 180 °C for 10 min). Acid digests were made up to 170 25 mL with ultrapure water and stored in polyethylene bottles at room temperature before 171 measurements. Two reagent blanks were prepared in each sample pre-treatment set. 172 For As species isolation, homogenized fish samples (1.0 g for wet samples and 0.20 g for dried 173 CRMs) were weighted in 50 mL centrifuge tubes and subjected to an ultrasound-water bath 174 extraction process using 10 mL of 1:1 methanol/ultrapure water as an extractant and 30 min 175 sonication at 45 kHz (120 W) and room temperature. This extracting solution was selected 176 from the available literature [10, 22-25]. The suspensions were then centrifuged at 3000 rpm 177 for 10 min, and the supernatants were kept in polyethylene bottles and stored at – 20 °C. 178 2.4. Synthesis of As-ion imprinted polymer 179 14 studied using aqueous standards at 4.0 µg L-1 each. They were subjected to the optimized 335 IIP/NIP-SPE procedure in triplicate. After HPLC-ICP-MS (inorganic arsenic and 336 organoarsenic species) and ICP-MS (metallic ions) determination (measurement conditions in 337 ESI), parameters such as extraction efficiency, distribution ratio (D) and selectivity coefficient 338 (SAs(III)/X), defined as shown in Table 2, were calculated. As expected, the polymeric material 339 (IIP) discriminates As(III) and As(V) species (extraction efficiencies of 92% and 90%, 340 respectively) from organoarsenic forms (extraction efficiencies within the 0-17% range) and 341 also other metallic ions (extraction efficiencies lower than 10%). In addition, extraction 342 efficiencies were lower than 50% for As(III) and As(V) when using NIP (Table 2), 343 highlighting the imprinted recognition capacities of the synthesized IIP. According to Table 2, 344 high distribution ratios and low selectivity coefficients were observed for As (III) and As (V); 345 whereas, other ions and organoarsenic species showed low distribution ratios and high 346 selectivity coefficients. These findings demonstrate that IIP offers selective recognition 347 properties for inorganic As (As(III) and As(V)). 348 3.4. Analytical performances 349 3.4.1. Calibration 350 In addition to aqueous As(III) and As(V) calibrations (concentrations within the 0 – 20 µg L-1 351 range), several standard addition graphs were performed by spiking fish extracts with 352 increasing As(III) and As(V) concentration up to 0.4 µg L-1 before IIP-SPE (concentrations 353 within the 0 - 20 µg L-1 after IIP-SPE). Since mean slopes for aqueous calibration and standard 354 addition (Table 3) are different, a calibration based on standard additions throughout the IIP-355 SPE on-column procedure is needed for achieving accurate results. 356 3.4.2. Limit of detection and limit of quantification 357 The pre-concentration factor (PF) of the method, defined as the ratio of sample extract to eluate 358 [29,30], was found to be 50 for both As(III) and As(V) species. In addition, experiments in 359 triplicate were performed to establish the enhancement factor (EF), which is defined as the 360 15 ratio of the slope of calibration curves with and without pre-concentration [29,30]. EFs of 59 361 and 58 were obtained for As(II) and As(V). Similar EF and PF values indicate quantitative 362 recovery for the pre-concentration procedure in absence of matrix effect. As previously 363 commented, the existence of the matrix effect is exceeded/compensated by using the standard 364 addition technique. 365 The limit of detection (LOD) and the limit of quantification (LOQ) were calculated according 366 to the Eurachem guideline as three (LOD) and ten (LOQ) times the standard deviation of an 367 analytical blank [31]. Analytical responses were then expressed as concentrations dividing by 368 the mean slope of the standard addition graph. Table 3 lists the LOD and LOQ values after 369 taking into account the pre-concentration factor of 50 and referring to sample mass. These 370 values are quite lower than the maximum levels for inorganic As in rice and rice-based 371 products (0.3 mg kg-1 for total inorganic As) established by the EU [9]. 372 3.4.3. Precision and accuracy 373 Precision and analytical recovery have been established through intraday and inter-day assays 374 which implied several fish extracts spiked with As(III) and As(V) at concentration levels of 375 0.04, 0.20, and 0.40 µg L-1 (concentrations of 2.0, 10, and 20 µg L-1 after pre-concentration). 376 Intraday assays were performed by preparing three standard additions in three different days, 377 replicating seven times one of the tested concentration levels each day (0.2, 2.0, or 5.0 µg L-1); 378 whereas, the other concentration levels were replicated twice. Inter-day assays were assessed 379 by preparing six standard additions in seven different days but replicating each concentration 380 level twice. Both intraday and inter-day precision, as well as intraday and inter-day analytical 381 recoveries (Table 4), were found to be good. 382 Inter-day and intraday assays were performed with the same set of 6 IIP-packed syringes, 383 which implies 25 adsorption-desorption cycles each one. As shown in Figure 5, the analytical 384 relative recovery for As(III) and As(V) solutions at 0.2 µg L-1 (10 µg L-1 after pre-385 16 concentration) was found to be within the 80-100% range. These relative recovery values were 386 calculated as reported in Equation 1: 387 Equation 1 388 where Cmeasured is the total concentration of the target analyte measured after the addition of a 389 certain concentration of analyte (Cadded) to a fish extract sample. All these findings imply that 390 the same IIP syringe can be re-used at least 25 times without losing recognition/sorption 391 properties. 392 Three CRMs showing certified total As contents were analysed for total As (microwave-393 assisted acid digestion and ICP-MS) and also for As(III) and As(V) (UAE, IIP-SPE and HPLC-394 ICP-MS). Table 5 lists the found total As concentrations in the three CRMs, which are in good 395 agreement with the certified total As concentrations after applying a t-test at a 95% confidence 396 range. Although As(III) and As(V) concentrations are not certified in the analysed CRMs, 397 found As(III) and As(V) concentrations after applying the proposed method (data listed in 398 Table S2, ESI section) were quite similar to some reported data for inorganic As species in 399 these CRMs [3]. Differences between the amounts of reported As(III)/As(V) concentrations in 400 these CRMs suggest the need for improving As speciation methods for inorganic (As(III) and 401 As(V)) assessment, and also the preparation of new CRMs with certified/indicative 402 As(III)/As(V) concentrations. 403 3.5. Application 404 The developed and validated method was applied to fish samples commercially available in 405 markets in Sri Lanka and Spain. Results in Table 6 shows that the percentage of inorganic As 406 was found to be within the 8-20 % range, although inorganic As in one swordfish sample 407 (coded as Swordfish 1) was 61%, mainly due to the high As(V) concentration found in this 408 sample (0.28±0.01mg kg-1). 409 410 17 Conclusions 411 The synthesized IIP has been shown to offer high selectivity for inorganic As (As(III) and 412 As(V)). The development of an IIP-SPE procedure has allowed the quantification of very low 413 concentrations (lower than 0.02 mg kg-1 for As (V)) of these two toxic As species in 414 commercially available fish. The high selectivity of the material offers the possibility of 415 inorganic As speciation by HPLC methods because organoarsenic compounds (mainly AB) are 416 not retained/pre-concentrated, which implies the avoidance of co-elution and allows for the 417 accurate determination of low concentrations of As(III) and As(V). The prepared material 418 could also be useful for pre-concentrating inorganic As in other complex samples exhibiting 419 low inorganic As levels, or for samples with moderately low concentrations of inorganic As 420 that are analysed by less sensitive hyphenated techniques. 421 422 Acknowledgment 423 This work was supported by the Dirección Xeral de I+D – Xunta de Galicia Grupos de 424 Referencia Competitiva (project number 6RC2014/2016 and ED431C2018/19), and 425 Development of a Strategic Grouping in Materials - AEMAT (grant ED431E2018/08). The 426 authors also wish to thank Dr. M. José Pazos-Guldrís (Unidade de Microscopía) at Rede de 427 Infraestruturas de Apoio á Investigación e ao Desenvolvemento Tecnolóxico – Universidade de 428 Santiago de Compostela) SEM technical support. 429 430 Figure captions 431 Figure 1. HPLC-ICP-MS chromatograms for pre-concentrated As(III) and As(V) standards 432 within the 0-20 µg L-1 range (A), and for a pre-concentrated extract from a cod fish sample (B). 433 434 18 Figure 2. FT-IR spectra (A) of synthesized IIP, synthesized IIP after template removal, and 435 NIP, and SEM images of synthesized IIP (B), synthesized IIP after template removal (C), and 436 NIP (D). 437 438 Figure 3. Effect of the extract pH (A), loading flow rate (B), elution flow rate (C), and elution 439 volume (D) on the As(III) analytical recovery and on the extracted As(V) concentration. 440 441 Figure 4. Effect of the fish extract volume (breakthrough volume) on the As(III) and As(V) 442 analytical recovery. 443 444 Figure 5. Analytical recovery for As(III) and As(V) standards at 0.2 µg L-1 using the same IIP 445 syringe over two weeks (25 sorption/desorption cycles). 446 447 19 Figure 1. 448 449 450 0 500 1000 1500 2000 2500 0 1 2 3 4 5 6 7 8 75As cps Time (min) (A) 75As cps Time (min) (B) 0 200 400 600 800 1000 1200 0 1 2 3 4 5 6 7 8 1.0 µg L-1 2.0 µg L-1 5.0 µg L-1 10 µg L-1 20 µg L-1 As(III) As(V) As(III) As(V) 20 Figure 2. 451 452 453 454 455 456 (B) (C) (D) 2924 1602 1510 1525-1575 1456 880-725 88 90 92 94 96 98 100 102 400900140019002400290034003900 Wave number (cm-1) IIP before template removal IIP after template removal NIP Transmitance (%) (A) 21 Figure 3. 457 458 459 460 0 2 4 6 0 10 20 30 40 50 60 70 7.0 7.5 8.0 8.5 9.0 [As(V)] (µg L-1) As (III) analytical recovery (%) pH (A) As(III) As(V) As(III) As(V) As (III) analytical recovery (%) [As(V)] (µg L-1) (D) 0 1 2 3 4 5 6 0 10 20 30 40 50 60 1 2 3 4 5 Elution volume (mL) As(III) As(V) [As(V)] (µg L-1) As (III) analytical recovery (%) (B) 0 1 2 3 4 0 10 20 30 40 50 60 70 0.25 0.5 0.75 1.0 Loading flow rate (mL min-1) As(III) As(V) As (III) analytical recovery (%) [As(V)] (µg L-1) (C) 0 1 2 3 4 5 0 10 20 30 40 50 60 70 80 90 0.25 0.5 0.75 1.0 Elution flow rate (mL min-1) 22 Figure 4. 461 462 463 0 20 40 60 80 100 0 20 40 60 80 10 25 37.5 50 100 Breakthrough volume (mL) As(III) As(V) As (III) analytical recovery (%) As (V) analytical recovery (%) 23 Figure 5. 464 465 466 467 0 20 40 60 80 100 120 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Number of sorption/desorption cycles As (III) As (V) Analytical recovery (%) 30 Table 6. Total As and As(III) and As(V) concentrations (mean ± SD, n=3) in several fish 488 samples 489 Total As (mg kg-1)a As (III) (mg kg-1)b As (V) (mg kg-1)b Inorganic As (mg kg-1)b,c Inorganic As percentage (%) Swordfish-1 0.60±0.02 86.3±2.8d 0.28±0.01 0.37±0.01 61±3 Swordfish-2 1.8±0.1 0.33±0.05 13.9±2.9d 0.35±0.05 19±5 Yellowfin tuna 2.0±0.1 0.27±0.03 24.5±3.4d 0.29±0.03 15±3 Tuna sp. 0.31±0.01 39.3±1.3d 7.3±0.3d 0.047±0.001 15±1 Blue shark 4.9±0.3 0.98±0.17 9.7±5.9d 0.99±0.17 20±13 Cod 0.34±0.02 15.0±4.7d 12.1±5.6d 0.027±0.007 8±4 (a) Microwave assisted acid digestion and ICP-MS measurement; (b) Ionic imprinted polymer based solid phase extraction and HPLC-ICP-MS measurement; (c) Sum of As(III) and As(V) concentration; (d) Concentration expressed in µg kg-1 490 491 31 References 492 [1] Z. Šlejkovec, Z. Bajc, D.Z. Doganoc, Arsenic speciation patterns in freshwater fish, Talanta, 62 (2004) 931-936. [2] S. McSheehy, J. Szpunar, R. Morabito, P. Quevauviller, The speciation of arsenic in biological tissues and the certification of reference materials for quality control, TrAC Trends Anal. Chem. 22 (2003) 191-209. [3] T. Llorente-Mirandes, R. Rubio, J. F. López-Sánchez, Inorganic arsenic determination in food: A review of analytical proposals and quality assessment over the last six years, Appl. Spectrosc. 71 (2017) 25–69. [4] A. Leufroy, L. Noël, V. Dufailly, D. Beauchemin, T. Guérin, Determination of seven arsenic species in seafood by ion exchange chromatography coupled to inductively coupled plasma-mass spectrometry following microwave assisted extraction: Method validation and occurrence data, Talanta, 83 (2011) 770-779. [5] European Food Safety Authority, Scientific opinion on arsenic in food, EFSA J. 7 (10):1351 (2009) 1-199. [6] WHO, Technical Report Series 959, Evaluation of Certain Food Additives and Contaminants, 72nd Report of the Joint FAO/WHO Expert Committee on Food Additives. Geneva, Switzerland, 2010. [7] M. Contreras-Acuña, T. García-Barrera, M.A. García-Sevillano, J.L. Gómez-Ariza, Arsenic metabolites in human serum and urine after seafood (Anemonia sulcata) consumption and bioaccessibility assessment using liquid chromatography coupled to inorganic and organic mass spectrometry, Microchem. J. 112 (2014) 56-64. [8] J. Moreda–Piñeiro, E. Alonso-Rodríguez, V. Romarís-Hortas, A. Moreda-Piñeiro, P. López-Mahía, S. Muniategui-Lorenzo, D. Prada-Rodríguez, P. Bermejo-Barrera, Assessment 32 of the bioavailability of toxic and non-toxic arsenic species in seafood samples, Food Chem., 130 (2012) 552-560. [9] COMMISSION REGULATION (EU) 2015/1006 of 25 June 2015 amending Regulation (EC) No 1881/2006 as regards maximum levels of inorganic arsenic in foodstuffs. Official Journal of the European Union L161 (2015) 14-16. [10] Á.H. Pétursdóttir, H. Gunnlaugsdóttir, Selective and fast screening method for inorganic arsenic in seaweed using hydride generation inductively coupled plasma mass spectrometry (HG-ICPMS), Microchem. J. 144 (2019) 45-50. [11] H. Rekhi, S. Rani, N. Sharma, A, K, Malik, A review on recent applications of highperformance liquid chromatography in metal determination and speciation analysis, Crit. Rev. Anal. Chem. 47 (2017) 524-537. [12] J.M.R Castor, L. Portugal, L. Ferrer, L. Hinojosa-Reyes, J.L. Guzmán-Mar, A. Hernández-Ramírez, V. Cerdà, An evaluation of the bioaccessibility of arsenic in corn and rice samples based on cloud point extraction and hydride generation coupled to atomic fluorescence spectrometry, Food Chem. 204 (2016) 475-482. [13] H. Shirkhanloo, A. Rouhollahi, H.Z. Mousavi, Ultra-trace arsenic determination in urine and whole blood samples by flow injection-hydride generation atomic absorption spectrometry after preconcentration and speciation based on dispersive liquid-liquid microextraction, Bull. Korean Chem. Soc. 32 (2011) 3923-3927. [14] G. Chen, T. Chen, SPE speciation of inorganic arsenic in rice followed by hydridegeneration atomic fluorescence spectrometric quantification, Talanta 119 (2014) 202-206. [15] C. Magoda, P.N. Nomngongo, N. Mabuba, Magnetic iron–cobalt/silica nanocomposite as adsorbent in micro solid-phase extraction for preconcentration of arsenic in environmental samples. Microchem. J. 128 (2016) 242-247. 33 [16] S.A. Jabasingh, T. Ravi, A. Yimam, Magnetic hetero-structures as prospective sorbents to aid arsenic elimination from life water streams. Water Sci. 32 (2018) 151-170. [17] N. Tavakkoli, S. Habibollahi, S.A. Tehrani, Separation and preconcentration of Arsenic (III) ions from aqueous media by adsorption on MWCNTs, Arab. J. Chem. 10 (2017) S3682- S3686. [18] E. Najafi, F. Aboufazeli, H.R. Lotfi Zadeh Zhad, O. Sadeghi, V. Amani, A novel magnetic ion imprinted nano-polymer for selective separation and determination of low levels of mercury(II) ions in fish samples, Food Chem. 141 (2013) 4040-4045. [19] T. Prasada Rao, R. Kala, S. Daniel, Metal ion-imprinted polymers—Novel materials for selective recognition of inorganics, Anal. Chim. Acta 578 (2006) 105–116. [20] T. Alizadeh, M. Rashedi, Synthesis of nano-sized arsenic-imprinted polymer and its use as As3+ selective ionophore in a potentiometric membrane electrode: Part 1. Anal. Chim. Acta, 843 (2014) 7-17. [21] Y.K. Tsoi, Y.M. Ho, K.S.Y. Leung, Selective recognition of arsenic by tailoring ionimprinted polymer for ICP-MS quantification, Talanta, 89 (2012) 162-168. [22] J.J. Sloth, E.H. Larsen, K. Julshamn, Determination of organoarsenic species in marine samples using gradient elution cation exchange HPLC-ICP-MS, J Anal. At. Spectrom. 18 (2003) 452-459. [23 A. Moreda-Pi eiro, E. Pe aque , P. ermeloerbello, P. ermejoarrera, . Moreda-Pi eiro, E. Alonsoodr gue , S. Muniategui-Loren o, P. L pe -Mah a, . Pradaodr gue , Matrix solid-phase dispersion as a sample pretreatment for the speciation of arsenic in seafood products, Anal. Chem. 80 (2008) 9272-9278. [24] B. Batista, L. Nacano, S. De Souza, F. Barbosa Jr, Rapid sample preparation procedure for As speciation in food samples by LC-ICP-MS, Food Addit. Contam. Part A 29 (2012) 780- 788. 34 [25] C.M.M. Santos, M.A.G. Nunes, I.S. Barbosa, G.L. Santos, M.C. Peso-Aguiar, M.G.A. Korn, E.M.M. Flores, V.L. Dressler, Evaluation of microwave and ultrasound extraction procedures for arsenic speciation in bivalve mollusks by liquid chromatography–inductively coupled plasma-mass spectrometry, Spectrochim. Acta Part B: At. Spectr. 86 (2013) 108-114 [26] A.A. Ammann, Arsenic speciation analysis by ion chromatography - A critical review of principles and applications, Am. J. Anal. Chem. 2 (2011) 27-45. [27] Y. Zhai, Y. Liu, X. Chang, X. Ruan, J. Liu, Metal ion-small molecule complex imprinted polymer membranes: Preparation and separation characteristics, React. Funct. Polym. 68 (2008) 284-291. [28] J. Otero Romaní, A. Moreda Piñeiro, P. Bermejo Barrera, A. Martin Esteban, Synthesis, characterization and evaluation of ionic-imprinted polymers for solid-phase extraction of nickel from seawater, Anal. Chim. Acta 630 (2008) 1-9. [29] R. Saxena, K. Madaan, S. Bansal, M. Saxena, N. Sharma, A Review on nanomaterials as solid phase extractants for determination of lead in environmental samples, J. Appl. Chem. 11 (2018) 27-38. [30] R. M. Toudeshki, S. Dadfarnia, A.M.H. Shabani, Surface molecularly imprinted polymer on magnetic multi-walled carbon nanotubes for selective recognition and preconcentration of metformin in biological fluids prior to its sensitive chemiluminescence determination: Central composite design optimization, Anal. Chim. Acta 1089 (2019) 78-89. [31] EURACHEM guide, The fitness for purpose of analytical methods, LGC, Teddington, UK, 1998. Electronic Supplementary Material (online publication only) Click here to download Electronic Supplementary Material (online publication only): Electronic Supplementary Information.docx Authors Contributions Section Kamal K. Jinadasa: Formal analysis, Investigation, Validation, Visualization, Writing- Original draft preparation Elena Peña-Vázquez: Data curation, Supervision, Validation, Writing- Reviewing and Editing Antonio Moreda-Piñeiro: Conceptualization, Methodology, Software, Validation, Writing- Reviewing and Editing, Supervision Pilar Bermejo-Barrera: Resources, Project administration, Funding acquisition, Writing- Reviewing and Editing *Credit Author Statement Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: *Conflict of Interest Statement