Journal of Chromatography A, 1391 (2015) 917

Contents lists available at ScienceDirect

Journal of Chromatography A

j o ur na l ho me page: www.elsev ier .com/ locate /chroma

Mercur acouple sopre-con pomethyl

Mara Pil na PRaquel D io MDepartment of Analytical Chemistry, Nutrition and Bromatology, Faculty of Chemistry, University of Santiago de Compostela, Avenida das Ciencias, s/n,15782 Santiago de Compostela, Spain

a r t i c l e i n f o

Article history:Received 19 DReceived in reAccepted 24 FAvailable onlin

Keywords:Ionic imprinteMercury speciSeawaterSolid phase exHigh performaInductively cospectrometry

a b s t r a c t

1. Introdu

Althoughpreviously, cury concenthe industrifossil fuels nicant soumercury artions range

CorresponE-mail add

http://dx.doi.o0021-9673/ ecember 2014vised form 23 February 2015ebruary 2015e 3 March 2015

d polymeration

tractionnce chromatographyupled plasma-mass

Trace levels of inorganic mercury, methyl-mercury and ethyl-mercury have been assessed in seawaterby high performance liquid chromatography (HPLC) hyphenated with inductively coupled plasma-massspectrometry (ICP-MS) after solid phase extraction (SPE) pre-concentration with a novel synthesizedionic imprinted polymer. The adsorbent material was prepared by trapping a non-vinylated chelatingligand (phenobarbital) via imprinting of a ternary mixed ligand complex of the non-vinylated chelatingagent, the template (methyl-mercury), and the vinyl ligand (metacrylic acid, MAA). Ethylene dimetacry-late (EDMA) and 2,2-azobisisobutyronitrile (AIBN) were used as cross-linker and initiator reagents,respectively; and the precipitation polymerization technique was used in a porogen of acetonitrile/water(4:1). The best retention properties for methyl-mercury, inorganic mercury and ethyl-mercury speciesfrom seawater were obtained when loading 200 mL of sample adjusted to pH 8.0 and at a ow rate of2.0 mL min1 on a column-packed with 200 mg of the material. Quantitative mercury species recoverieswere obtained using 4 mL of an eluting solution consisting of 0.8% (v/v) 2-mercaptoethanol and 20% (v/v)methanol (pH adjusted to 4.5) pumped at a ow rate of 2.0 mL min1. Mercury species separation wasachieved on a Kinetex C18 column working under isocratic conditions (0.4% (v/v) 2-mercaptoethanol,10% (v/v) methanol, pH 2.5, ow rate 0.7 mL min1). ICP-MS detection was performed by monitoring themercury mass to charge ratio of 202. The limits of quantication of the method were 11, 6.7, and 12 ng L1,for inorganic mercury, methyl-mercury and ethyl-mercury, respectively (pre-concentration factor of 50);whereas, analytical recoveries ranged from 96 to 106%. The developed method was successfully appliedto several seawater samples from unpolluted areas.

2015 Elsevier B.V. All rights reserved.

ction

the use of mercury (in the chloralkali industry and,in the wood pulping industry) is decreasing, high mer-trations are still present in sediments associated withal applications of this metal. In addition, the burning ofand pollution of water by mine tailings can also be sig-rces of mercury [1]. However, low levels of dissolvede expected in the marine environment [2]. Concentra-

from 0.5 to 3.0 ng L1 in open ocean; whereas, values

ding author. Tel.: +34 881814375; fax: +34 981 547141.ress: antonio.moreda@usc.es (A. Moreda-Pineiro).

within the 215 ng L1 range can be found in coastal seawater [1].Nevertheless, local variations from these values can be observed incoastal seawater and inland waters where near to anthropogenicsources, and where mercury associated with suspended materialmay also contribute to the total load [1]. Even at low concentrations,mercury species can drive bio-accumulation and bio-magnicationto levels of this toxic metal in sh, posing human and ecologicalhealth risks [24]. Mercury is thus included in the Annex I (Envi-ronmental Quality Standards for Priority Substances and CertainOther Pollutants) of the Directive 2013/39/UE of the European Par-liament and of the Council (12/08/2013), and a maximum allowableconcentration (MAC) of mercury and its compounds of 70 ng L1 ininland surface waters and other surface waters has been established[5].

rg/10.1016/j.chroma.2015.02.0682015 Elsevier B.V. All rights reserved.y speciation in seawater by liquid chromd plasma-mass spectrometry followingcentration by using an ionic imprinted-mercuryphenobarbital interaction

ar Rodrguez-Reino, Roi Rodrguez-Fernndez, Eleomnguez-Gonzlez, Pilar Bermejo-Barrera, Antontography-inductivelylid phase extractionlymer based on

ena-Vzquez,oreda-Pineiro

10 M.P. Rodrguez-Reino et al. / J. Chromatogr. A 1391 (2015) 917

Despite the better performances (lower limit of detection andbetter separation) obtained when using gas chromatography-inductively coupled plasma-mass spectrometry (GC-ICP-MS), theneed of derivatization of mercury species is drawback. HPLC-ICP-MS methodsample intrneeded whconcentratisamples, suextraction concentratiquantitativminimal/netions for me[7], and SAXexchange reC18 [1118plexation toor the immcles [1418

Current new sorbenbranes, modand mesoppared mainimprinted have been both organposes [19functional interaction,pre-polymechanged to iliary non-vmetal ion aplate and tchelating lioccasions, bmetal ion anlinking) canically immoof the latteing vinyl gso the synformed in thave been umost of thein the polyligandvinyand methyl

The aimtation polymspecies [Hagainst majride). A straligand/viny(MAA) wasposed for demonstratacting withspeciation performancplasma-maaration wathan those 1025 min

(5 min) when comparing to those previously reported applications,and because of the high sensitivity inherent in the use IIP-SPEprocedure, the developed method allows the assessment of mer-cury species in seawater at levels lower than the MAC established

DireL1)

erim

para

lexarom Ps sepn 30ry dem Elchro0 A (alyti8 gu. Othm lemeneanta eratu) eqA). Iny) bA). SPed wl sta

tubiter saore, sized(Millng elny), olybaestrxico

tory de (CHeid

oven

agen

apur purns. Hc (Ba000 igmaolum

Ethyed frSA) bc aciry (I)solvition i, Pb,, although less sensitive allows, however a direct liquidoduction. Pre-concentration procedures are thereforeen using HPLC-ICP-MS because the expected mercuryons are within the ng per liter range, and complexch as seawater, contain high salt levels. Solid phase(SPE) has been demonstrated to be a suitable pre-on technique for extracting mercury species, allowinge recoveries, efcient salt removal, and guarantyinggligible mercury species changes [6]. Recent applica-rcury speciation include anion (quaternary amine type

[8]), and cation (SCX, benzenesulfonic acid-type) [9,10]sins; and also polymeric resins, mainly reverse phase

]. The latter solid supports require mercury species com- enhance analyte interaction with the sorbent [1113],obilization of the complexing reagent on the C18 parti-].trends in SPE are mainly focused on the development ofts such as nanometer-sized materials, egg-shell mem-ied silica beads, molecularly/ion imprinted polymers,orous materials. These new solid supports are pre-ly to improve selectivity [6]. As reported, molecularlypolymers (MIPs) and ionic imprinted polymers (IIPs)proposed for achieving efcient pre-concentration ofic and inorganic analytes, and also for clean-up pur-22]. In contrast to organic molecules, the absence ofgroups in most ions difcults a templatemonomer

and the typical binary (template/vinylated monomer)rization mixtures when synthesizing MIPs must beternary pre-polymerization mixtures in which an aux-inylated reagent (a ligand exhibiting afnity for the

nd adequate functional groups) is mixed with the tem-he monomer. After polymerization, the non-vinylatedgand is therefore trapped via imprinting [22]. On otherifunctional reagents (a ligand exhibiting afnity for thed containing vinyl groups for polymerization after cross

be used, and the complexing ligand is therefore chem-bilized in the polymeric matrix. The main drawbackr approach is the fact that complexing ligands show-roups are scarce and are not commercially available,thesis of these complexing monomers must be per-he laboratory. Although some bifunctional monomerssed for preparing IIPs for mercury recognition [23,24],

synthesized IIPs used non-vinylated ligands trappedmeric matrix after ternary templatenon-vinylatedlated monomer interactions when using Hg(II) [2530]-mercury (Me-Hg) [31] as templates.

of the current work was the application of the precipi-erization approach for synthesizing an IIP for mercury

g(II), Me-Hg, and ethyl-mercury (Et-Hg)] retentionor components in seawater (mainly sodium and chlo-tegy based on using a ternary template/non-vinylatedlated monomer (Me-Hg/phenobarbital/metacrylic acid

adopted. Phenobarbital has not previously been pro-IIP synthesis. However, phenobarbital was recentlyed to exhibit adequate functional groups for inter-

vinylated monomers such as MMA [32]. Mercury(Hg(II), Me-Hg, and Et-Hg) was performed by highe liquid chromatography (HPLC)-inductively coupledss spectrometry (ICP-MS). The chromatographic sep-s attained in 5 min, a shorter chromatographic timereported in other published applications (within therange) [3336]). Because of the short analysis time

in the (70 ng

2. Exp

2.1. Ap

A Fpler) frspecieNex-Iomercupler frophase C18 10ter) anto a Cmenex(100 mPhenolent (Sa temprey, UKNC, USGermaPA, USequippchanne2-stopSeawa(MillipSynthelters scanniGermaeter (mde InfrTecnollaboraelectro2000 (and aning.

2.2. Re

UltrMilli-QsolutioPanreation (1from Ssmall vwater.preparMA, UsulfuriMercuby discalibraMn, Nctive 2013/39/UE for environmental quality standards[5].

ental

tus

LC HPLC (LC pump, column oven, and LC autosam-erkin Elmer (Waltham, MA, USA) was used for mercuryarations. The HPLC system was coupled to a Perkin Elmer0X ICP-MS, instrument which was also used for totalterminations [in such case, a SeaFast SC2 DX autosam-emental Scientic (Omaha, NB, USA) was used]. Reversematographic separations were performed on a Kinetex100 mm length 2.10 mm i.d., 5.0 m particle diame-cal column (Phenomenex, Torrance, CA, USA) connectedard column (4 mm length 3.0 mm i.d.) from Pheno-er tested analytical columns were Kinetex C18 100 Angth 4.60 mm i.d., 2.6 m particle diameter) fromx, and Zorbax Eclipse XDB-C8 (4.6 150 mm) from Agi-Clara, CA, USA). Polymerization was performed by usingre-controlled incubation camera (Stuart Scientic, Sur-uipped with a low-prole roller (Stovall, Greensboro,IPs were packed into 5 mL syringes (Braun, Wertheim,etween replacement Teon frits (Supelco, Bellefonte,E was performed by using a Miniplus 3 peristaltic pumpith a head for delivering low pulse ows of uids in 8ndard ow rate (Gilson, Middleton, WI, USA), and PVCng (3.175 mm i.d.) from SCP Sciences (Quebec, Canada).mples were ltered through Durapore 0.45 m lters

Billerica, MA, USA) by using a vacuum pump (Millipore). IIP material was ltered through Durapore 0.22 mipore). IIP characterization was performed by using aectron microscope EVO LS 15 from Zeiss (Oberkochem,and an energy dispersive X-ray uorescence spectrom-denum-based anode) laboratory-made by RIAIDT (Redeuturas de Apoio Investigacin e ao Desenvolvemento) at the University of Santiago de Compostela. Otherdevices were a Basic20 pH meter with a glasscalomelrison, Barcelona, Spain) for pH measurements, a Reaxoph, Niederbayern, Germany) for vortexing solutions,

model 207 from Selecta (Barcelona, Spain) for IIP dry-

ts

e water of resistivity 18 M cm obtained from aication device (Millipore) was used to prepare all theg(II) stock standard solution (1000 mg L1) was fromrcelona, Spain). Methyl-mercury stock standard solu-mg L1) was prepared from methyl-mercury chloride

(Steinhelm, Germany) by dissolving the reagent in ae of hot methanol, and further dilution with ultrapurel-mercury stock standard solution (1000 mg L1) wasom ethyl-mercury chloride from Alfa Aesar (Ward Hill,y reagent dissolution in a small volume of 5% (v/v)d in methanol before dilution with ultrapure water.

stock standard solution (1000 mg L1) was preparedng mercury (I) chloride from Sigma. Multi-elementstandard 3 (10 mg L1 of Al, Ca, Cd, Cr, Cu, Fe, Mg,

Sr, V, and Zn), and Y (10 mg L1) used as an internal

M.P. Rodrguez-Reino et al. / J. Chromatogr. A 1391 (2015) 917 11

for II

standard wmethanol, (v/v)), hydrchloride anPanreac. Mphenobarbiby Sigma. 2Fluka (Buchtallization (Panreac) a4 C. High pammoniumchased fromwas obtainwater sampsodium chland heptah14 g in 1 L omercury inNational Re

All glassinto 10% (mrinsed three

2.3. Seawat

Surface ies in Galicpolyethylenthrough 0.4diately subjat 20 C.

2.4. Synthe

Solid m(61.0 mg) w(acetonitrila templateof 1:2:2 in and 1:2:4) then stirred(365 L), astudies [38and after sfor 10 min aimmediatelabout its locamera (the60 C for 2 h

eriza, wa

ght aed.k po

reparlumel-meash

mpla

eral ps bee rememenitionn in ain1

) wer of a

syrifor th

arac

nnin NIPsizedericaglomn beeriza].rgy d

thee of ovaFig. 1. Scanning electron microscopy pictures

ere from Perkin Elmer (Shelton, CT, USA). HPLC gradehigh purity nitric acid (96% (v/v)), sulfuric acid (69%ochloric acid (37% (v/v)), and analytical grade sodiumd sodium hydrogencarbonate were purchased frometacrylic acid (MAA), ethylene dimetacrylate (EDMA),tal, thiourea, and 2-mercaptoethanol were supplied,2-Azobisisobutyronitrile (AIBN) was purchased froms, Switzerland). This reagent was puried by crys-

at 20 C after dissolving the reagent in methanolt 5060 C. After purication, this reagent was stored aturity ammonia (25% (v/v)), ethanol, sodium hydroxide,

chloride, and hydroxylamine hydrochloride were pur- Merck (Darmstadt, Germany). HPLC grade acetonitrile

ed from Scharlab (Barcelona, Spain). A synthetic sea-le of low salinity (34.2) was prepared by dissolvingoride and sodium hydrogencarbonate from Panreac,ydrate magnesium sulfate from Merck (32, 0.15, andf ultrapure water, respectively) [37]. ORMS-5 (Elevated

river water) certied reference material was fromsearch Council of Canada (Ottawa, Canada).

and plastic material were rigorously cleaned and kept/m) nitric acid for at least 48 h. The material was then

times with ultrapure water before being used.

er collection

seawater samples were collected from several estuar-ia (northwestern Spain) in pre-cleaned high densitye bottles. After collection, seawater samples ltered5 m lters and subjected to analysis. When not imme-ected to analysis, ltered seawater samples were kept

sis of ionic imprinted polymer particles

ethyl-mercury chloride (18.3 mg) and phenobarbitalere mixed with 51.2 L of MAA and 12 mL of porogene/water 4:1) into 15 mL glass test tubes. This implies

polymlteredoverniobtain

Blanalso pand vo(methysame w

2.5. Te

Sevsyringemust bcomplrecognsolutio1 mL mments200 mLpackedwater

2.6. Ch

ScaIIP andsyntheof sphcles agthat capolym[3840

EnermingabsencHg rem/non-vinylated reagent/vinylated reagent molar ratiothe polymerization mixture. Other molar ratios (1:1:1led to IIPs of poor retention capacity. The mixture was

for 5 min and was kept in the dark overnight. EDMAnd puried AIBN (43.3 mg), according with previous40], were added to the test tubes placed in an ice-bath,

tirring for 1 min, the mixture was purged with argont low temperature (test tubes in an ice-bath), and wasy sealed, and placed in a low-prole roller (33 r.p.mng axis) inside a temperature-controllable incubator

temperature was ramped from room temperature to, and then maintained at 60 C for a further 24 h). Once

material bewell as for Npresent in tthe describ

2.7. IIPs sol

SPE syrpumping 10at a ow raof seawaterNH3/NH4ClP (A) and NIP (B).

tion was nished, the synthesized material was vacuumshed with acetonitrile/water (4:1), and then oven-driedt 40 C. Masses of polymer ranging from 1.5 to 1.7 g were

lymer particles (non-imprinted polymers, NIPs) wereed in the same way as IIPs, using the same massess as for IIP synthesis, but without adding the templatercury chloride). The NIPs were then subjected to the

ing pre-treatment described above.

te removal procedure

ortions (200 mg) of the material were packed into 5 mLtween two Teon frits. The template (methyl-mercury)oved from the polymer particles, leaving free cavities

tary in size, shape, and functionality ready for analyte. This procedure was performed by pumping a thiourean acid medium (1 M thiourea in 1 M HCl) at a ow rate of. Negligible mercury concentrations (ICP-MS measure-e found in the washing/ltrate solutions after passingcidic thiourea. After complete template removal, thenges were extensively washed by pumping ultrapureiourea residues removal.

terization of the synthesized IIP

g electron microscopy (SEM) pictures were taken from (Fig. 1) to study the morphology and the size of the

materials. Both IIP and NIP consisted of agglomeratesl particles (mean diameter lower than 10 m). Parti-eration in NIPs (Fig. 1B) appears to be lower, a fact

attributed to the absence of the template during thetion process as shown when preparing other IIPs/NIPs

ispersive X-ray uorescence (EDXRF) was used for con- presence of mercury in the synthesized IIP (and themercury in NIP), and also to prove an efcient Me-

l from IIPs after leaching with thiourea. Patterns for IIP

fore and after leaching (Fig. 2A and B, respectively), asIP (Fig. 2C) were obtained, showing that mercury is onlyhe unleached IIP particles, and it is totally removed aftered template removal procedure.

id phase extraction

inges containing the polymer were conditioned by mL of a 0.1 M/0.1 M NH3/NH4Cl buffer solution (pH 8.0)te of 2 mL min1. Previous to sample loading, 200 mL

(pH approximately 6.7) was mixed with a 0.1 M/0.1 M buffer solution (pH 8.0) to obtain a pH of 8.0 0.5

12 M.P. Rodrguez-Reino et al. / J. Chromatogr. A 1391 (2015) 917

ached IIP (A), NIP (B), and leached IIP (C).

(2540 mL adjusted toow rate ofof the NH3retained me1.0 mL aliqumercaptoet2.0 mL min

these opera10 mL of ulpassing 10 m8.0 (2 mL mapproximatwith 8 chan

2.8. ICP-MS

Total meCu, Fe, Mn, other toxic Yttrium (25bration wamercury (Hwere separMS. Operata pre-conceditions (2.3Me-Hg andof Hg(II), M

3. Results

3.1. Optimi

Mobile cysteine/msolutions wposed when

s repass msteinical c

mx C18ter), n, thal red

werinete-mer opsitioFig. 2. Energy dispersive X-ray uorescence spectra for the unle

of the buffer solution were needed). Seawater samples a pH of 8.0 were then pumped through the syringe at a

2 mL min1. The syringes were then rinsed with 20 mL/NH4Cl (pH 8.0) buffer solution (2 mL min1), and thercury species were then subsequently eluted with fourots of an eluting solution consisting of 0.8% (v/v) 2-hanol and 20% (v/v) methanol (pH 4.5) at a ow rate of1. A pre-concentration factor of 50 was achieved underting conditions. After elution, the IIPs were treated withtrapure water (2 mL min1) and then conditioned byL of the 0.1 M/0.1 M NH3/NH4Cl buffer solution at pH

in1). Eight samples can simultaneously be treated inely 100 min because of the use of a peristaltic pumpnels.

ndingcury mof l-cyanalyti.d., 5.0Kinetediameadditioa partisignalsboth Kusing 2ered focompo and HPLC-ICP-MS measurements

rcury, as well as other trace elements such as Al, Cd, Cr,Ni, Pb, Sr, and V (assessment of the interferences frommetals on SPE) were determined by ICP-MS (Table 1).

g L1) was used as an internal standard, and cali-s performed within the 0100 g L1 range. Inorganicg(II)), MeHg, and Et-Hg in the extracts after IIP-SPEated by reverse phase (RP) HPLC and detected by ICP-ing conditions are listed in Table 1. Chromatograms ofntrated seawater sample under optimized IIP-SPE con-0 0.13, 0.62 0.035, and 0.45 0.03 g L1 for Hg(II),

Et-Hg, respectively), and a 5.0 g L1 standard solutione-Hg, and Et-Hg is shown in Fig. 3 as an example.

and discussion

zation of operating RP conditions

phases based on acidic aqueous l-cysteine [41], l-ethanol [10,34], and acidic aqueous 2-mercaptoethanolith [35] and without [8] methanol have been pro-

separating mercury species by RP. In accordance with

Table 1Instrumental p

ICPMS

Ar ow rate(L min1)

O2 ow rate(L min1)

KED mode: rate (mL m

Mass-to-charatio

RP-HPLC

Mobile phascompositi

Mobile phasorted by Chang et al. [33], increased background at mer-/z 202 (ICP-MS detection) was observed in the presencee in our preliminary studies when using the selectedolumn (Kinetex C18 100 A, 100 mm length 2.10 mm

particle diameter, Table 1), and also when using the 100 A (100 mm length 4.60 mm i.d., 2.6 m particle

and the Zorbax Eclipse XDB-C8 analytical columns. Ine presence of l-cysteine in the mobile phase resulted inuction of Hg(II) species to Hg(I) (two chromatographice observed when injecting Hg(II) standards when usingx C18 columns). Therefore, RP separations based onrcaptoethaol/methanol mixtures were further consid-timization. A study of the effect of the mobile phasen on the retention times and chromatographic signalsarameters for ICP-MS and HPLC-ICP-MS measurements.

Radiofrequency power (W) 1600Peristaltic pump speed (rpm) 2.5Plasma/auxiliary/nebulizer 18.0/1.2/0.89

0.01

He owin1)

Al, Cr, Mn, Cu, Cd, Hg, and PbMg, Ca, V, Fe, Ni, Zn, and Sr

14

rge Y (internal standard)Mg, Al, Ca, V, Cr, Mn, Fe, Ni, Cu, Zn, Sr,Cd, Hg, and Pb

8924, 27, 43, 51,52, 55, 57, 60,65, 66, 88, 114,202, and 208

Kinetex C18 100 A (100 mmlength 2.10 mm i.d., 5.0 m particlediameter) analytical column connectedto a C8 guard column (4 mmlength 3.0 mm i.d.)Injection volume/L 50

eon

Isocratic: 0.4% (v/v)2-mercaptoethanol, 10% (v/v)methanol, pH 2.5

e ow 0.7 mL min1

M.P. Rodrguez-Reino et al. / J. Chromatogr. A 1391 (2015) 917 13

0

500

1000

1500

2000

2500

3000

3500

4000

4500

0

Inte

nsity

/ cps

5.0 g L-1 standard

Seawater SW1Me-Hg

Fig. 3. Reversstandard, andand 0.45 0.0

(integratedof a mobile10% (v/v) m

3.2. Prelimi

Since mtal residuesbonding delytes by us[23,25,27,2reduction oles for M(v/v) 2-meras an elutobtained wever, signal2 mL aliquomercaptoetconcentratiavoid cross analytes whafter elutioncleaned by acid solutio

3.3. Optimi

3.3.1. Loadirate

Seawatelytes were tpHs 7.0, 7.5without pHlytes assessthe mobile0 and 100 in Figure S1Hg(II) is comately 80%working at to 60% at pHpH of 8.0). Tinefcient econditions w

The loading ow rate (0.5, 1.0, 1.5, and 2.0 mL min1) was alsostudied by using seawater samples spiked with 2.0 g L1 of eachanalytes, and treated with 0.1 M/0.1 M NH4Cl/NH3 buffer solutions

). As shown in Figure S1B (Supplementary electronic infor-), lo

min

ratew rats.

Elutinn

perc0.8eplic100inforusingries using% (vpproas, th

witer, i

obt-spikas siglosedble fr x

s studed ainforchie

than0% w

tativ. Fin

eluectro/v)) l

comd 20and

Elutin of th5.04.54.03.53.02.52.01.51.00.5

Time/ min

Et-Hg

Hg(II)

e phase chromatograms for a 5.0 g L1 Hg(II), Me-Hg, and Et-Hg for a pre-concentrated seawater sample (2.30 0.13, 0.62 0.035,3 g L1 for Hg(II), Me-Hg, and Et-Hg, respectively).

area and shape) of mercury species led to the selection phase consisting of 0.4% (v/v) 2-mercaptoethanol andethanol (pH of 2.5) as listed in Table 1.

nary evaluation of IIP-SPE conditions

ercury species are retained by binding to phenobarbi-, elution must lead to mercury speciesphenobarbitalstruction. First attempts for eluting the retained ana-ing a 0.1 M/0.1 M thiourea/hydrochloric acid solution8,31] led to an incomplete elution, as well as a partialf Hg(II) into Hg(I) and bad chromatographic pro-

e-Hg. The previously optimized mobile phase (0.4%captoethanol, 10% (v/v) methanol, pH 2.5) was trieding solution, and similar chromatograms to thosehen injecting aqueous standards were obtained. How-

intensities were low after eluting with two successivets. Therefore, a solution consisting of 0.4% (v/v) 2-hanol, 10% (v/v) methanol, and pH 2.5 (non-optimizedons) was adopted for further experiments. In order tocontamination and to guarantee that the IIP was free ofen studying experimental loading/elution conditions,

with the proposed solution, the IIP-SPE syringes werepumping 10 mL of 0.1 M/0.1 M thiourea/hydrochloricn.

zation of mercury species IIP-SPE from seawater

ng conditions: effect of sample pH and loading ow

(pH 8.0mation0.5 mLat owing oproces

3.3.2. solutio

Thethe 0.2three r0 and tronic when recovewhen and 0.8up to awhereelutingHowevS2A arein nonThis weries cnegligi

AftepH wawas xtronic were ahigherup to 6quantistudiedanalytetary el20% (v

The0.8 antively, acid).

3.3.3. volumer samples (100 mL) spiked with 2.0 g L1 of each ana-reated with 0.1 M/0.1 M NH4Cl/NH3 buffer solutions of, 8.0, and 9.0. In addition, a fortied seawater sample

adjustment (pH of 6.7) was also tested. After ana-ment by RP-HPLC-ICP-MS (calibration matched with

phase, and covering analyte concentrations betweeng L1) analytical recoveries (three replicates) plotted

A (Supplementary electronic information) showed thatnveniently retained at any pH (recoveries of approxi-); whereas Me-Hg and Et-Hg retention is favored whenhigher pHs (analytical recoveries for Me-Hg were closes of 8.0 and 9.0, and approximately 25% for Et-Hg at ahese low analytical recoveries must be attributed to anlution rather than bad retention (un-optimized elutionere used). Therefore, a pH of 8.0 was nally chosen.

Experima ow ratetive mercurbetween 0 rate (2.0 mL

Regardinrecoveries f4.0 and 5.0In addition,1 mL aliquoical recoverthe highest

3.3.4. EffectExperim

spiked (2.0wer analytes retention was observed when loading at1; whereas, analyte retention increased when loadings within the 1.02.0 mL min1 range. Therefore, a load-e was nally set at 2.0 mL min1 to obtain a faster SPE

g conditions: effect of the composition of the eluting

entage of 2-mercaptoethanol was rst studied within% (v/v) range. Results (mean analytical recoveries ofates after mobile phase matched calibration between

g L1) plotted in Figure S2A (Supplementary elec-mation) showed that analytes release was increased

higher percentages of 2-mercaptoethanol. Analyticalclose to 100% were observed for inorganic mercury

the highest percentages of 2-mercaptoethanol (0.6/v)). In addition, Me-Hg recoveries were also increasedximately 80% (0.6 and 0.8% (v/v) 2-mercaptoethanol);e highest Et-Hg recovery (48%) was obtained whenh solutions containing 0.8% (v/v) 2-mercaptoethanol.t must be mentioned that recoveries plotted in Figureained after subtracting the analyte concentrations founded seawater samples subjected to the same conditions.nicant mainly for inorganic mercury (analytical recov-

to 120% for spiked seawater samples); whereas, it wasor organic mercury species.ing the 2-mercaptoethanol percentage at 0.8% (v/v), theied within the 2.05.0 range (the methanol percentage

t 10% (v/v)). Results in Figure S2B (Supplementary elec-mation) showed that quantitative recoveries for Me-Hgved when using extracting solutions adjusted at pHs

4.0. In addition, Et-Hg recoveries were also increasedhen using higher pHs. Regarding inorganic mercury,

e recoveries were attained within the whole pH rangeally, the effect of the percentage of methanol on thetion was tested. As shown in Figure S2C (Supplemen-nic information), higher methanol percentages (15 anded to higher Et-Hg recoveries.position of the eluting solution was therefore xed at% (v/v) for 2-mercaptoethanol and methanol, respec-a pH of 4.5 (pH adjusted by using 0.1 M hydrochloric

g conditions: effect of the eluting ow rate ande eluting solutionents performed by pumping the eluting solution at

of 0.5, 1.0, 1.5, and 2.0 mL min1 showed quantita-y species recoveries (mobile phase matched calibrationand 100 g L1) for all cases. The highest eluting ow

min1) was therefore chosen for further experiments.g the volume of the eluting solution, quantitativeor the three mercury species were observed when using

mL (Figure S3, Supplementary electronic information). the best elution was achieved after pumping successivets. Since both eluting volumes have led to good analyt-ies, an eluting volume of 4.0 mL was chosen to obtain

pre-concentration factor.

of the sample volume (breakthrough volume)ents were performed by increasing the volume of

g L1 of each analyte) seawater samples from 100 mL

14 M.P. Rodrguez-Reino et al. / J. Chromatogr. A 1391 (2015) 917

0102030405060708090

100110

Ana

lytic

al re

cover

y (%

)

Fig. 4. Effect species by IIP-

to 250 mL. ple volumemercury spare attaineddecreased tTherefore, tTaking intoand the mathe breakth50 can be e

3.3.5. AssesIIP-SPE

Differentivity of thspecies. Sev(low salinitand other ithe optimiztrace elemetion, paramdistributionas shown icies, as wel(the templaHg(II) and Ewere obserciencies ancoefcientsthat the preother mercretained.

Similar eNIP. As shobution ratioions. It can tties for Me-the NIP mathan 50%), iduring the retention. Tand Fe(III) cate that thoffer imprinefciencies

Table 2Extraction efciency (%), distribution ratios (D) and selectivity coefcients (SMe-Hg/M)of for IIP and NIP.

Extraction Distribution Selectivity coefcient

mprin

nic im

A2/AT)A2/A1)Hg/M = unt ount of metal ion enriched by IIP/NIP at equilibrium.l amount of metal ion used in extraction.

distribution ratio for Me-Hg (template)tribution ratio for M (M = Hg(II), Et-Hg, Al(III), Cd(II), Cr(III), Cu(II), Fe(III),i(II), Pb(II), Sr(II), and V(III)).

alytical performances

Calibration. Evaluation of matrix effect external calibrations matched with mobile phase and cov-

analyte concentrations of 0.0, 5.0, 7.5, 15, and 30 g L1

btained in different days. Similarly, ve standard addi-alibrations were also obtained in ve different days by

seawater samples in duplicate with 0.0, 0.10, 0.15,and 0.60 g L1 of each analyte (concentration within30 g L1 after a pre-concentration factor of 50). Theand standard deviation for the slopes of ve standardn graphs were 35 104 2.8 104, 52 104 1.9 104,250200150100Sample volume (mL)

Me-HgHg(II)Et-Hg

of seawater sample volume on the analytical recovery of mercurySPE.

Results in triplicate (Fig. 4) showed that higher sam-s (up to 250 mL) can be used when assessing organicecies (Me-Hg and Et-Hg) since quantitative recoveries

for all cases. However, inorganic mercury recoverieso 87 1% when loading seawater volumes of 250 mL.he breakthrough volume can be established at 200 mL.

account the volume of the eluting solution (4.0 mL),ximum volume of seawater sample without reachingrough volume (200 mL), a pre-concentration factor ofstablished.

sment of the interferences from other toxic metals on

t experiments were performed to evaluate the selec-e synthesized material for interacting with mercuryeral 200 mL aliquots of a synthetic seawater sample

y, 32) [37] spiked Hg(II), Me-Hg, and Et-Hg (1 g L1)ons (2 g L1) as listed in Table 2, were subjected toed IIP-SPE procedure. After ICP-MS measurement fornt assessment and RP-HPLC-ICP-MS for mercury specia-eters such as extraction efciency (analytical recovery),

ratio (D) and selectivity coefcient (SMe-Hg/M), denedn Table 2, were calculated. High extraction efcien-l as high distribution ratios, were obtained for Me-Hgte when synthesizing the IIP material), and also fort-Hg. In addition, low selectivity coefcients (Table 2)ved for mercury species. However, low extraction ef-d distribution ratios (and therefore high selectivity) were obtained for other ions. These results indicate

Ionic iMe-HgHg(II) Et-Hg Al(III) Cd(II) Cr(III)Cu(II) Fe(III)Mn(II)Ni(II) Pb(II) Sr(II) V(III)

Non-ioMe-HgHg(II) Et-Hg Al(III) Cd(II) Cr(III)Cu(II) Fe(III)Mn(II)Ni(II) Pb(II) Sr(II) V(III)

a % = (b D = (c SMe-

A1 = amoA2 = amoAT = totaDMe-Hg =DM = disMn(II), N

3.4. An

3.4.1. Five

ering were otion cspiking0.30, the 0mean additiopared material is selective for Me-Hg (template) andury species; whereas, other transition metals are not

xperiments were performed by using the synthesizedwn in Table 2, lower extraction efciencies and distri-s were obtained for mercury species as well as for otherherefore be concluded that IIP offers imprinting proper-Hg (the template) and also for Hg(II) and Et-Hg; becauseterial shows low afnity (extraction efciencies lowerndicating that those specic cavities formed for Me-HgIIP polymerization are responsible for mercury specieshe relative high extraction efciencies for Cd(II), Cu(II),(Table 2) when using both IIP and NIP materials indi-ese ions are retained by adsorption (the IIP does notting properties for these ions because similar extraction

are observed when using the NIP).

and 42 1Hg, respecthe slopes 48 104 3Me-Hg, andBartletts t(comparisomatched cacally compaseawater wadvantage stechnique i

3.4.2. LimitThe lim

(LOQ) denefciency (%)a ratio (D)b (SMe-Hg/M)c

ted polymer (IIP)97 32 94 16 2.098 49 0.6539 0.27 11980 6.6 4.82.0 0.01 3200

75 3.7 8.673 2.7 121.4 0.012 2667

20 0.21 15257 1.6 203.4 0.033 9701.3 0.012 2667

printed polymer (NIP)47 0.89 43 0.75 1.247 0.89 1.020 0.25 3.679 3.8 0.245.5 0.058 15

65 1.9 0.4865 1.9 0.480.4 0.004 222

22 0.28 3.259 1.7 0.521.6 0.017 520.85 0.009 103

100..DMe-Hg/DM.f metal ion in aqueous solution at equilibrium.04 4.5 104 cps L g1 for Hg(II), Me-Hg, and Et-tively; whereas, mean and standard deviation forof ve matched calibration were 40 104 2.2 104,.6 104, and 39 104 3.1 104 cps L g1 for Hg(II),

Et-Hg, respectively. The application of CochranC andests (comparison of variances), and the ANOVA testn of means) at a 95.0% condence interval led tolibration and standard addition graphs were statisti-rable, and matrix effect due to the high salt content inas found to be unimportant. This fact offers a practicalo that a tedious and time consuming standard additions not necessary.

of detection and limit of quantication of the methodit of detection (LOD) and the limit of quanticationed through the 3/10 criterion as LOD = 3/m and

M.P. Rodrguez-Reino et al. / J. Chromatogr. A 1391 (2015) 917 15

Table 3Comparison of LODs and pre-concentration factor for mercury speciation by SPE.

LOD (ng L1) Pre-concentration factor Ref.

Hg(II) Me-Hg Et-Hg

Off-line phenobarbital IIP HPLC-ICP-MS 3.4 2.0 2.1 50 This workOff-line dithizone-functionalized C18 HPLC-ICP-MS 3.0 3.0 3.0 23.831.2 [16]Off-line sodium diethyldithiocarbamateimmobilized in polyurethane foam

HPLC-ICP-MS 4.6 5.2 100 [18]

Off-line 1,3-bis(2-cyanobenzene)triazenemodied C18

HPLC-UV 1.3 1.0 25 [17]

On-line ammoniumpyrrolidinedithiocarbamate modied C18

HPLC-MS 0.09a 0.37a 0.28a 2500 [12]

Off-line C60 fullerene (analytes are previouslycomplexed with sodiumdiethyldithiocarbamate)

GC-MS 1.5 1.2 1.5 100 [13]

Off-line dithizone-functionalized C18 HPLC-UV 0.54a 0.58a 200 [14]Off-line 2-mercaptoethanol modied C18 HPLC-CV-AFS 800 4300 1400 1000 [15]On-line strong anion-exchange column HPLC-ICP-MS 0.016 0.01 0.009 1025 [8]On-line cation-exchange column HPLC-ICP-MS 0.042 0.016 0.008 1250 [10]On-line C18 0.02

a Absolute L

Table 4Inter-day prec on lev

Concentratio

cover

0.1 0.6

LOQ = 10/mof a blank, culated. By were 3.4, 2tively. SimiLOQ for HgTable 3, LODby other autreatment band GC-MS using UV spvapour atomtion systemthose obtai(Table 3). LOlow enoughare lower t2013/39/eu

3.4.3. RepeaTable 4 s

recovery (seconcentratiperformed

h thater

d (prbtain anted tddedatrixtionydroults certions

Table 5Mercury speci

SW1 SW2 SW3 SW4 SW5 SW6 SW7 SW8

a Concentrab Total mercHPLC-ICP-MS 0.07

OD (expressed as ng).

ision and inter-day analytical recovery (n = 14) at two mercury species concentrati

n level (g L1) Hg(II) Me-Hg

Analytical recovery (%) RSD (%) Analytical re

96 12 12 98 8 99 7 7 100 6

( is the standard deviation of eleven measurementsand m is the mean slope of the calibration) were cal-assuming a pre-concentration factor of 50, LOD values.0, and 2.1 ng L1 for Hg(II), Me-Hg, and Et-Hg, respec-larly, values of 11, 6.7, and 7.0 ng L1 were obtained as(II), Me-Hg, and Et-Hg, respectively. As summarized ins achieved are similar than those previously obtained

thors when using other adsorbents for off-line SPE pre-efore HPLC-ICP-MS detection [16,18], HPLC-UV [17],[13]. LOD achieved are lower than those reported whenectophotometry [14], mass spectrometry [12], and coldic uorescence spectrometry (CV-AFS) [15] as detec-

s for HPLC (Table 3). However, values are higher than

througriver wmethowere oibratioattribu(v/v), aBrCl minteracwith htive resof the conditned when using on-line SPEHPLC-ICP-MS [8,10,11]Qs achieved by applying the current development are

to assess mercury species in seawater since the valueshan the MAC of 70 ng L1 established in the Directive

of the European Parliament and of the Council [5].

tability and accuracy of the methodhows good inter-day precision and inter-day analyticalawater samples spiked with a low and a high analyteson, 0.1 and 0.6 g L1, respectively, and experimentsin two different days). Accuracy was also evaluated

Hg(II) compcan occur. TCRM was thby ICP-MS. determinat(26.2 1.3 p

3.5. Applica

The devples collec

es concentration in some seawater samples.

[Hg(II)] (ng L1) [Me-Hg] (ng L1) [Et-Hg] (n

45.9 2.6 12.4 0.7 9.0 0.612.7 0.4

16 M.P. Rodrguez-Reino et al. / J. Chromatogr. A 1391 (2015) 917

(north-western Spain). Each seawater sample was subjected intriplicate to the proposed IIP-SPE method, and each eluate wasmeasured twice by the proposed RP-HPLC-ICP-MS procedure. Theeluates after IIP-SPE were also analyzed in triplicate for total mer-cury by ICPanalyte, anoccurs as inquantied ition was bealso be seeassessed inconcentratitions in seathere are nmethod. Fin70 ng L1 esParliament

4. Conclus

Mercuryraphy (chroconditions, ate matrix 2-mercaptotrapping inpresence olinker, respcapacities fseparation/samples. Thity for majothe salt matistical comstandard adcedure). Thof 50, and ttoring mercMAC of 70 n

Acknowled

The authInnovacinsupport.

Appendix A

Supplemfound, in th2015.02.06

References

[1] WHO, EnAspects, W

[2] C.H. Lamintercomwater andcruises, L

[3] D. MergleA.H. Sternconcern,

[4] A.M. ScheenvironmAmbio 36

[5] DirectiveAugust 20

[6] A. Moreda-Pineiro, J. Moreda-Pineiro, P. Bermejo-Barrera, Sample pre-treatment methods for organometallic species determination, in: S. Bakirdere(Ed.), Speciation Studies in Soil, Sediments and Environmental Samples, CRC,Press, Boca Raton, 2014, pp. 19201.

[7] P. Houserov, P. Kubn, V. Kubn, Ion exchange preconcentration and sep-tion oectionChengcentructive

analelgad

ifact m of ionl. Chi

Y. Jia, rcury hy hyion ex.L. Ca

mmerm them. AcHous

omatoits py250unozpoun

y follo72.. Snciatioractio, Talaargety in seid ching pr123Yong-ctionaid chrcury ashetaminmodiroche

dos Saanic aline cMartinctive181hen, St statu229

. Rao, ctive. Whitogy: a14) 29iu, Y. ethy

(200u, L. Cmine2+ in wAndacleculacromoakov

thacryercu

akova deterter. 23ordanopart14) 42aja,

impriels of Gnemoseym. Sc-MS. Table 5 lists the obtained concentrations for eachd also for total mercury in seawater. Most mercuryorganic mercury; whereas, Me-Hg and Et-Hg was onlyn one seawater sample. In addition, Me-Hg concentra-tween the LOD and LOQ in two other samples. It cann that total mercury concentrations are similar to theorganic mercury concentrations (and to the mercuryon expressed as the sum of mercury species concentra-water samples SW1, SW5 and SW6), which means thatot other mercury species unresolved by the RP-HPLCally, all concentrations were lower than the MAC oftablished in the Directive 2013/39/eu of the Europeanand of the Council [5].

ions

speciation has been performed by a fast RP chromatog-matographic run of 5 min) working under isocraticand with a mobile phase quite similar to the elu-composition used in the IIP-SPE procedure (formallyethanol and methanol in water). The phenobarbital

the polymeric matrix (IIP adsorbent) prepared in thef MAM and EDMA as vinylated monomer and cross-ectively, has generated a IIP exhibiting recognitionor Me-Hg, Hg(II) and Et-Hg ions, as well as quantitativepre-concentration of mercury species from seawatere synthesized polymeric material has not offered afn-r elements in seawater samples, removing efcientlytrix of seawater. This fact has been veried after sta-parison of matched calibration with mobile phase anddition calibrations (calibration through the IIP-SPE pro-e developed IIP-SPE allowed a pre-concentration factorhe achieved LODs/LOQs resulted low enough for moni-ury species in seawater samples (values lower than theg L1).

gements

ors wish to acknowledge the Ministerio de Ciencia e (project number CTQ2012-38091-C02-02) for nancial

. Supplementary data

entary data associated with this article can bee online version, at http://dx.doi.org/10.1016/j.chroma.8.

vironmental Health Criteria No. 86 Mercury Environmentalorld Health Organisation, Geneva, 1989.

borg, C.R. Hammerschmidt, G.A. Gill, R.P. Mason, S. Gichuki, Anparison of procedures for the determination of total mercury in sea-

recommendations regarding mercury speciation during GEOTRACESimnol. Oceanogr.: Methods 10 (2012) 90100.r, H.A. Anderson, H.M. Chan, K.R. Mahaffey, M. Murray, M. Sakamoto,, Methylmercury exposure and health effects in humans: a worldwideAmbio 36 (2007) 311.uhammer, M.W. Meyer, M.B. Sandheinrich, M.W. Murray, Effects ofental methylmercury on the health of wild birds, mammals and sh,

(2007) 1218. 2013/39/eu of the European Parliament and of the Council of 1213, Off. J. Eur. Union (2013) L 226/1226/17.

aradet

[8] H. conindtion

[9] A. DartuseAna

[10] X.-merapcat

[11] W.RDofroChi

[12] P. chras 242

[13] J. Mcometr66

[14] D.Mspeextphy

[15] J. Mcurliqulow115

[16] Y. funliqume

[17] T. Hconon Mic

[18] J.S.orgon-

[19] A. sele169

[20] L. Cren292

[21] T.Psele

[22] M.Jnol(20

[23] Y. Lof m575

[24] S. XthyHg

[25] M. moMa

[26] I. Dmeof m

[27] I. DandMa

[28] T. Ynan(20

[29] E. Nionlev

[30] O. chePolf mercury species by CE with indirect contactless conductometric, Electrophoresis 27 (2006) 45084515., C. Wu, L. Shen, J. Liu, Z. Xu, Online anion exchange column pre-ation and high performance liquid chromatographic separation withly coupled plasma mass spectrometry detection for mercury specia-ysis, Anal. Chim. Acta 828 (2014) 916.o, A. Prieto, O. Zuloaga, A. de Diego, J.M. Madariaga, Production ofethylmercury during the analysis of certied reference sediments:ic exchange in the sample treatment step to minimise the problem,

m. Acta 582 (2007) 109115.D.-R. Gong, Y. Han, C. Wei, T.-C. Duan, H.-T. Chen, Fast speciation ofin seawater by short-column high-performance liquid chromatog-phenated to inductively coupled plasma spectrometry after on-linechange column preconcentration, Talanta 88 (2012) 724729.irns, M. Ranaldo, R. Hennebelle, C. Turetta, G. Capodaglio, C.F. Ferrari, A.gue, P. Cescon, C. Barbante, Speciation analysis of mercury in seawater

lagoon of Venice by on-line pre-concentration HPLC-ICP-MS, Anal.ta 622 (2008) 6269.erov, D. Matejcek, V. Kubn, High-performance liquidgraphic/ion-trap mass spectrometric speciation of aquatic mercuryrrolidinedithiocarbamate complexes, Anal. Chim. Acta 596 (2007)., M. Gallego, M. Valcrcel, Speciation analysis of mercury and tinds in water and sediments by gas chromatographymass spectrom-wing preconcentration on C60 fullerene, Anal. Chim. Acta 548 (2005)

chez, R. Martn, R. Morante, J. Marn, M.L. Munuera, Preconcentrationn method for mercury compounds in water samples using solid phasen followed by reversed phase high performance liquid chromatogra-nta 52 (2000) 671679.nov, P. Houserov-Pelcov, V. Kubn, Speciation analysis of mer-diments, zoobenthos and river water samples by high-performance

romatography hyphenated to atomic uorescence spectrometry fol-econcentration by solid phase extraction, Anal. Chim. Acta 615 (2008).Guang, M. Chen, P. Jin-Feng, L. Jing-Fu, J. Gui-Bin, Dithizone-lized solid phase extractiondisplacement elution-high performanceromatography-inductively coupled plasma mass spectrometry forspeciation in water samples, Talanta 81 (2010) 17881792.mpur, M.K. Rofouei, A.R. Khorrami, Speciation analysis of mercuryants in water samples by RP-HPLC after solid phase extractioned C18 extraction disks with 1,3-bis(2-cyanobenzene)triazene,m. J. 89 (2008) 131136.ntos, M. de la Guardia, A. Pastor, M.L.P. dos Santos, Determination ofnd inorganic mercury species in water and sediment samples by HPLCoupled with ICP-MS, Talanta 80 (2009) 207211.-Esteban, Molecularly-imprinted polymers as a versatile, highly

tool in sample preparation, Trends Anal. Chem. 45 (2013).. Xu, J. Li, Recent advances in molecular imprinting technology: cur-s, challenges and highlighted applications, Chem. Soc. Rev. 40 (2011)

42.R. Kala, S. Daniel, Metal ion-imprinted polymers novel materials for

recognition of inorganics, Anal. Chim. Acta 578 (2006) 105116.combe, N. Kirsch, I.A. Nicholls, Molecular imprinting science and tech-

survey of the literature for the years 20042011, J. Mol. Recognit. 277401.

Zai, X. Chang, Y. Guo, S. Meng, F. Feng, Highly selective determinationlmercury with methylmercury-imprinted polymers, Anal. Chim. Acta6) 159165.hen, J. Li, Y. Guan, H. Lu, Novel Hg2+-imprinted polymers based onHg2+thymine interaction for highly selective preconcentration ofater samples, J. Hazard. Mater. 237238 (237) (2012) 347350., S. Mirel, S. S enel, R. Say, A. Ersz, A. Denizli, Ion-imprinted beads forr recognition based mercury removal from human serum, Int. J. Biol.l. 40 (2007) 159166.a, I. Karadjova, V. Georgieva, G. Georgiev, Ion-imprinted poly-lic microbeads as new sorbent for preconcentration and speciationry, Talanta 78 (2009) 523529., T. Yordanova, I. Karadjova, Non-chromatographic mercury speciationmination in wine by new coreshell ion-imprinted sorbents, J. Hazard.1232 (231) (2012) 4950.ova, I. Dakova, K. Balashev, I. Karadjova, Polymeric ion-imprintedicles for mercury speciation in surface waters, Microchem. J. 11347.

F. Aboufazeli, H. Reza, L.Z. Zhad, O. Sadeghi, V. Amani, A novel magneticnted nano-polymer for selective separation and determination of lowmercury(II) ions in sh samples, Food Chem. 141 (2013) 40404045.y, F.C . Cebeci, Molecularly imprinted uorescent polymers asnsors for the detection of mercury ions in aqueous media, J. Appl.i. 117 (2010) 23732379.

M.P. Rodrguez-Reino et al. / J. Chromatogr. A 1391 (2015) 917 17

[31] S. Byktiryaki, R. Say, A. Denizli, A. Ersz, Mimicking receptor for methylmer-cury preconcentration based on ion-imprinting, Talanta 71 (2007) 699705.

[32] S.G. Hu, S.W. Wang, S.W. He, An amobarbital molecularly imprinted micro-sphere for selective solid-phase extraction of phenobarbital from human urineand medicines and their determination by high-performance liquid chro-matography, Analyst 128 (2003) 14851489.

[33] L.-F. Chang, S.-J. Jiang, A.C. Sahayam, Speciation analysis of mercury and leadin sh samples using liquid chromatography-inductively coupled plasma massspectrometry, J. Chromatogr. A 1176 (2007) 143148.

[34] A. Castillo, A.F. Roig-Navarro, O.J. Pozo, Method optimization for the determi-nation of four mercury species by micro-liquid chromatography-inductivelycoupled plasma mass spectrometry coupling in environmental water samples,Anal. Chim. Acta 577 (2006) 1825.

[35] X. Jia, Y. Han, X. Liu, T. Duan, H. Chen, Speciation of mercury in water samplesby dispersive liquidliquid microextraction combined with high performanceliquid chromatography-inductively coupled plasma mass spectrometry, Spec-trochim. Acta B 66 (2011) 8892.

[36] C. Ibnez-Palomino, J.F. Lpez-Snchez, . Sahuquillo, Inorganic mercury andmethylmercury determination in polluted waters by HPLC coupled to coldvapour atomic uorescence spectroscopy, Int. J. Environ. Anal. Chem. 92 (2012)909921.

[37] P. Bermejo-Barrera, J. Moreda-Pineiro, A. Moreda-Pineiro, A. Bermejo-Barrera,Usefulness of the chemical modication and the multi-injection technique

approaches in the electrothermal atomic absorption spectrometric determi-nation of silver, arsenic, cadmium, chromium, mercury, nickel and lead insea-water, J. Anal. At. Spectrom. 13 (1998) 777786.

[38] J. Otero-Roman, A. Moreda-Pineiro, P. Bermejo-Barrera, A. Martin-Esteban,Synthesis, characterization and evaluation of ionic-imprinted polymers forsolid-phase extraction of nickel from seawater, Anal. Chim. Acta 630 (2008)19.

[39] J. Otero-Roman, A. Moreda-Pineiro, P. Bermejo-Barrera, A. Martin-Esteban,Inductively coupled plasma-optical emission spectrometry/mass spectrome-try for the determination of Cu, Ni, Pb and Zn in seawater after ionic imprintedpolymer based solid phase extraction, Talanta 79 (2009) 723729.

[40] J. Otero-Roman, A. Moreda-Pineiro, P. Bermejo-Barrera, A. Martin-Esteban,Ionic imprinted polymer for nickel recognition by using the bi-functionalized5-vinyl-8-hydroxyquinoline as a monomer: application as a new solid phaseextraction support, Microchem. J. 93 (2009) 225231.

[41] X. Chen, C. Han, H. Cheng, Y. Wang, J. Liu, Z. Xua, L. Hu, Rapid speciationanalysis of mercury in seawater and marine sh by cation exchange chro-matography hyphenated with inductively coupled plasma mass spectrometry,J. Chromatogr. A 1314 (2013) 8693.

[42] DRAFT, Total and Dissolved Mercury in Water by Bromine MonochlorideDigestion PBM, 2013, Available at http://www.elp.gov.bc.ca/epd/wamr/labsys/pdf/Draft Hg Water 22Feb2013.pdf (accessed 16.05.14).

Mercury speciation in seawater by liquid chromatography-inductively coupled plasma-mass spectrometry following solid phase...1 Introduction2 Experimental2.1 Apparatus2.2 Reagents2.3 Seawater collection2.4 Synthesis of ionic imprinted polymer particles2.5 Template removal procedure2.6 Characterization of the synthesized IIP2.7 IIPs solid phase extraction2.8 ICP-MS and HPLC-ICP-MS measurements

3 Results and discussion3.1 Optimization of operating RP conditions3.2 Preliminary evaluation of IIP-SPE conditions3.3 Optimization of mercury species IIP-SPE from seawater3.3.1 Loading conditions: effect of sample pH and loading flow rate3.3.2 Eluting conditions: effect of the composition of the eluting solution3.3.3 Eluting conditions: effect of the eluting flow rate and volume of the eluting solution3.3.4 Effect of the sample volume (breakthrough volume)3.3.5 Assessment of the interferences from other toxic metals on IIP-SPE

3.4 Analytical performances3.4.1 Calibration. Evaluation of matrix effect3.4.2 Limit of detection and limit of quantification of the method3.4.3 Repeatability and accuracy of the method

3.5 Application to seawater samples

4 ConclusionsAcknowledgementsAppendix A Supplementary dataReferences