DAf

ZD

a

ARRAA

KDCPEBHc

1

ngassi

otts

f

(

h0

Journal of Chromatography A, 1394 (2015) 46–53

Contents lists available at ScienceDirect

Journal of Chromatography A

jo ur nal ho me pag e: www.elsev ier .com/ locate /chroma

issolution of biological samples in deep eutectic solvents:n approach for extraction of polycyclic aromatic hydrocarbons

ollowed by liquid chromatography-fluorescence detection

ahra Helalat–Nezhad, Kamal Ghanemi ∗, Mehdi Fallah–Mehrjardiepartment of Marine Chemistry, Faculty of Marine Science, Khorramshahr University of Marine Science and Technology, P.O. BOX 669, Khorramshahr, Iran

r t i c l e i n f o

rticle history:eceived 6 December 2014eceived in revised form 19 March 2015ccepted 20 March 2015vailable online 27 March 2015

eywords:eep eutectic solventholine chloride-oxalic acidolycyclic aromatic hydrocarbons (PAHs)xtractioniological samplesigh performance liquidhromatography-fluorescence detection

a b s t r a c t

A novel sample preparation method based on the complete dissolution of marine biological samplesin choline chloride-oxalic acid (ChCl-Ox) deep eutectic solvent was developed for fast and efficientextraction of eight polycyclic aromatic hydrocarbons (PAHs) using minimum volumes of cyclohexane.The extracted PAHs were purified and then measured by high-performance liquid chromatography-fluorescence detection (HPLC-FL). The effect of key parameters on extraction recoveries and precisionwas investigated. At optimized conditions, the studied samples were dissolved under atmospheric pres-sure in ChCl-Ox (1:2) at 55 ◦C for 30 min, which is considerably lower than the temperature used in theclassical and current methods. After dissolution, it took approximately 20 min to quantitatively extractthe PAHs from ChCl-Ox using 5 mL cyclohexane. Depending on the analyte, the developed method waslinear over the calibration range 1.0–250, 2.0–250, and 5.0–250 ng g−1, with r2 > 0.996. The detection lim-its of the method were between 0.50 and 3.08 ng g−1. The intra-day and inter-day precisions (based onthe relative standard deviation, n = 5) of the spiked PAHs at a concentration level of 50 ng g−1 were better

than 12.6% and 13.3%, respectively. Individual PAH recoveries from spiked marine fish and macroalgaesamples were in the range of 71.6% to 109.6%. For comparison, the spiked samples were also subjected tothe Soxhlet extraction method. The simplicity of the procedure, high extraction efficiency, short analysistime, and use of safe and inexpensive components suggest the proposed method has a high potential forutilization in routine trace PAH analysis in biological samples.

. Introduction

The increasing social demand for analytical methods and theeed for fast, accurate, precise, selective, and sensitive methodolo-ies oblige us to consider the use of reagents that are innocuous, ort least less toxic than those currently used. We should also con-ider a drastic reduction in the amount of samples, reagents, andolvents employed for analysis, as well as to minimize, decontam-nate, and neutralize the waste generated [1].

In nearly all analytical methods used for determining tracerganic pollutants, sample preparation, especially involving extrac-

ion, occupies a strategic place [2]. This step generally extracts thearget analytes from their matrices into solution and renders themuitable for analysis [1,2].

∗ Corresponding author. Tel.: +98 6153533322/9161310959;ax: +98 6153533322.

E-mail addresses: Kamal.ghanemi@kmsu.ac.ir, kamalghanemi@gmail.comK. Ghanemi).

ttp://dx.doi.org/10.1016/j.chroma.2015.03.053021-9673/© 2015 Elsevier B.V. All rights reserved.

© 2015 Elsevier B.V. All rights reserved.

Polycyclic aromatic hydrocarbons (PAHs) are a large group oforganic compounds that are included in the European Union (EU)and the US Environmental Protection Agency (US EPA) prioritypollutant list because of their mutagenic and carcinogenic prop-erties [3–6]. Most people are exposed to PAHs predominantly fromdietary sources, especially seafood [7–9]. In the marine environ-ment, PAHs are available biologically to marine species via thefood chain, as waterborne compounds, and from contaminatedsediments. As lipophilic compounds, they can easily cross lipidmembranes and have the potential to accumulate in aquatic orga-nisms, and may interfere in normal DNA functioning [3,10].

The mode of extraction for PAHs is highly dependent on thematrix. For solid-based matrices such as food samples, sediments,soil, and marine organisms, classical extraction methods suchas Soxhlet extraction with nonpolar solvents are used [11]. Thismethod utilizes large amounts of solvent as well as sample and

is considerably time consuming [12]. Another separation proto-col of PAHs from lipophilic products is based on saponification inbasic alcoholic solution, a process that requires 2–4 h. Moreover,the presence of alcohol in the hydrolytic mixture interferes with

hrom

toasnflate

tfatlsjc

lGhacd

anarplIal[

I(tevtisD((Traihw[

o[bDromyir

Z. Helalat–Nezhad et al. / J. C

he alkylated PAH derivatives [10]. Consequently, current meth-ds tend to minimize the consumption of solvents, sample amount,nd extraction time with the aid of additional energy and/or pres-ure to the mixture of sample and solvent [13]. These methods,amely, microwave-assisted extraction (MAE) [14], supercriticaluid extraction (SFE) [15], pressurized liquid extraction (PLE) [16],nd ultrasound-assisted extraction (USE) [17], differ in the way thathe energy is supplied to the system and the kind of extracting fluidmployed.

Recently, some methods have used microwaves in combina-ion with water instead of organic solvents for extracting PAHsrom solid matrices [18]. Although this technique can consider-bly increase the environmental-friendliness of the method, somearget analytes, e.g., 5-ring PAHs, are fairly insoluble in water,eading to a very small amount extracted into the final organicolvent. Therefore, these analytes produce low signals that mayust be detectable with highly sensitive instruments such as gashromatography-mass spectrometry (GC-MS).

An alternate approach for the extraction of analytes from bio-ogical samples is the use of ionic liquids (ILs). For example,ermán-Hernández and co-workers utilized aggregates of 1-exadecyl-3-butyl imidazolium bromide in a focused-microwave-ssisted extraction method followed by high-performance liquidhromatography with ultraviolet and fluorescence detection toetermine the 15 + 1 EU priority PAHs in toasted cereals [19].

The widespread use of ILs in both academia and industry isttributed to their unique combination of properties that includeegligible vapor pressure, high thermal stability, low/no volatility,nd ease of handling [20]. But typically, one cannot simply evapo-ate ILs to recover analytes, as analytical separations are generallyerformed with organic solvents. Many reports have also high-

ighted the hazardous nature and poor biodegradability of mostLs [21]. In addition, processes for the synthesis of ILs are notlways environmentally friendly because they generally requirearge amount of salts and solvents for complete anion exchange22].

To overcome the limitations of high price and toxicity ofLs, a new generation of green solvents—deep eutectic solventsDESs)—have emerged [23]. A DES is generally composed of two orhree non-toxic components that are capable of associating withach other through hydrogen bonds [22]. DESs typically have aery high depression in freezing point and are liquids at tempera-ures ranging from 21 ◦C to 70 ◦C [23]. Choline chloride (ChCl), annexpensive, biodegradable, and non-toxic quaternary ammoniumalt, is widely used as one of the components in the formation ofESs [24]. When combined with non-toxic hydrogen bond donors

HBDs), such as carboxylic acids (e.g., oxalic acid), urea, or polyolse.g., glycerol), ChCl is capable of rapidly forming a liquid [22–24].hese mixtures can be prepared with high purity and do noteact with water, which allows for easy storage. In addition, theyre biodegradable, biocompatible, non-toxic, non-flammable, andnexpensive [22]. Charge delocalization that is achieved throughydrogen bonding between the halide anion (e.g., Cl− in ChCl)ith an HBD moiety is responsible for formation of the liquid

23].Liquid ChCl mixtures have been used in applications in vari-

us fields, including drug solubilization [25], biodiesel purification26], and electrodeposition of metals [23]. Moreover, they haveeen proven to dissolve macromolecules. For example, in 2013,ai et al. [27] utilized different types of DESs composed of natu-

al constituents, e.g., proline-malic acid (PMA), for the extractionf phenolic metabolites of various polarities from safflower. Most

ajor phenolic compounds were recovered from the DES with

ields between 75% and 97%. They also found that the H-bondingnteractions between DES molecules and phenolic compounds areesponsible for their high extractability.

atogr. A 1394 (2015) 46–53 47

Similarly, Bi and co-workers used alcohol-based DESs that areprepared by mixing ChCl with different alcohol-based HBDs toextract flavonoids (myricetin and amentoflavone) from plants [28].

We recently reported the application of some DESs for com-plete dissolution of marine biological samples, which facilitatedthe quantitative extraction of some studied metals with smallvolumes of dilute nitric acid for determination by flame atomicabsorption spectrometry (FAAS) [29]. Therefore, based on theseexperiments, we developed a new, green, and efficient method forthe extraction of PAHs from marine biological samples, e.g., fishand macroalgae samples, after their complete dissolution in thestudied DESs. The extracted PAHs were then analyzed with high-performance liquid chromatography coupled with a fluorescencedetector. Generally, the chemical characteristics of PAHs are simi-lar within a given ring number [30]; therefore, PAHs that containthree, four, and five aromatic rings were selected, because morethan 80% of USEPA PAHs belong to these ring classes. As a model,8 PAHs including phenanthrene (Ph), anthracene (An), fluoran-thene (Flt), pyrene (Pyr), benz[a]anthracene (BaA), chrysene (Chry),benzo[e]pyrene (BeP), and benzo[a]pyrene (BaP) were studied. Theoptimum analytical conditions for quantitative recoveries of PAHswere investigated. To our knowledge, this is the first report of usingDESs for the extraction of organic pollutants from solid-based bio-logical matrices.

2. Experimental

2.1. Reagents and solutions

All reagents used in the experiment were of analytical reagentgrade and used without further purification. Choline chloride(C5H14NClO, 99.0%) was purchased from Sigma (St. Louis, MO,USA). High-purity oxalic acid (Ox) was supplied by Merck (Darm-stadt, Germany). The standard mixture of PAHs (Ph, An, Flt, Pyr,BaA, Chry, BeP, and BaP) containing 500 �g mL−1 of each com-ponent in toluene was purchased from Supelco (Bellefonte, PA,USA). Stock solutions containing 10 �g mL−1 of PAHs were pre-pared by dilution of the standard mixture in acetonitrile and storedat 4 ◦C in darkness. Fresh calibration solutions were prepared dailyfrom stock solutions. HPLC-grade acetonitrile, dichloromethane,n-hexane, toluene, cyclohexane, and water were purchased fromMerck. Double distilled deionized water was used throughout. Allglassware was washed three times with n-hexane and methanolbefore use.

2.2. Instrumentation

PAHs were quantitatively determined with the Knauer HPLCinstrument (Berlin, Germany) with a Wellchrom series solventorganizer (k-1500), binary pump (k-1001), vacuum degasser, elec-trical injection system, and RF-10 AXL fluorescence detector. Thecolumn temperature was maintained at 25 ◦C in an oven (S-4000)with a thermostat. PAHs were separated on a Nucleosil® LC-PAHcolumn (250 mm × 4.6 mm i.d., particle size 5 �m, pore size 100 A)from Macherey-Nagel (Düren, Germany). A personal computerequipped with Ezchrom 3.1.7 software (Knauer) for the LC systemwas used to acquire and store the chromatographic data. A rotaryevaporator (Heidolph Laborota 4011-digital, Germany) with bathtemperature ≤30 ◦C was used for the fast solvent evaporation ofthe extracts.

2.3. Sample collection and pretreatment

In this study, two types of marine fish samples i.e., Kafshak(Platichthys flesus) and Shoorideh (Otolithes ruber) were boughtfresh from a local fish market in Khorramshahr, Iran. These fish

4 hrom

aocftpsirwpwsw

2

afifcc(Tpe

2

potawidtuosm

pvpbwwtt

2

DimODuisuc

8 Z. Helalat–Nezhad et al. / J. C

re usually caught from the Musa estuary (north-western regionf the Persian Gulf, Iran). They have a high demand and are widelyonsumed by the people in this region. To study extraction per-ormance and optimize the extraction conditions as indicated inhe individual experiments, an uncontaminated Hamour fish sam-le (Epinephelus coioides) was prepared at the marine fish researchtation (Bandar Emam Khomeini, Iran). All samples were kept coldn an icebox during transportation to the laboratory. At the labo-atory, the fish samples were washed thoroughly with deionizedater and the muscle tissues were separated and cut into smallieces. A marine macroalgae sample (Enteromorpha intestinalis)as collected from Bushehr (north of the Persian Gulf, Iran). All

amples were freeze-dried for about 24 h, and the dried samplesere ground to fine powders and sieved through a 125 mesh.

.4. Preparation of spiked samples

To evaluate the extraction performance during the optimizationnd validation of the method, the spiking of the uncontaminatedsh sample (previously checked to be PAH-free) was performed as

ollows: 1 g of the freeze-dried and powdered sample was mixedompletely with 250 �L of acetonitrile containing a known con-entration of each PAH to get a desired concentration of analytes50 ng g−1). The samples were then stored in the dark to dry.hen, all samples including non-spiked and spiked samples werereserved in pre-cleaned polyethylene bottles at 4 ◦C until thextraction was conducted.

.5. Preparation of eutectic mixtures

There is no limit to the number or type of DESs that can be pre-ared from available chemicals, because there are a large numberf salts and hydrogen bond donors that can be used to preparehese solvent mixtures [31]. Amides, carboxylic acids, and alcoholsre 3 major groups of HBDs that are widely used, in combinationith ChCl, for producing DESs [23,24,31]. Our preliminary exper-

ments indicated that ChCl-Ox DES systems have more ability toissolve marine biological samples and extract the analytes fromheir matrices than other common eutectic solvents such as ChCl-rea and ChCl-glycerol. These findings are in coherence with thoseutlined in other studies. For example, in 2009, Morrison et al. [25]howed that 80% of the studied drugs are about 3 to 22,000 timesore soluble in carboxylic-based DESs than in other types.Therefore, in this work, the eutectic mixtures of ChCl-Ox were

repared at molar ratios of 1:1, 1:2, and 1:3, and utilized as sol-ents for the extraction of the PAHs from the biological samples. Torepare these mixtures, ChCl was mixed with Ox in a 50-mL round-ottomed flask, which was then placed in a small water-bath, whichas heated using a hot-plate stirrer. The temperature of the DESas continuously monitored with a thermometer (±0.1 ◦C), and

he eutectic mixture was formed by stirring the two componentsogether at 45 ◦C until a homogeneous, colorless liquid was formed.

.6. Proposed eutectic-based extraction method

Before beginning the procedure, the temperature of the studiedES was increased to 50–70 ◦C and maintained at each stud-

ed temperature. Exactly 0.10 g of the biological sample (fish oracroalgae) was added to a known volume (1.5–3.5 mL) of ChCl-x and stirred at 120 rpm for a specified time interval (20–70 min).uring this stage, the entire sample was dissolved, and a relativelyniform solution was obtained. Then, known volumes of extract-

ng solvent were added and stirred for 20–60 min. The supernatantolution was completely separated, and passed through a col-mn of 0.3 g silica gel; the column was then washed with 5 mL ofyclohexane; the collected solvents were evaporated in the rotary

atogr. A 1394 (2015) 46–53

evaporator at 35 ◦C until almost dry (about 3 min). The samplewas then diluted with 1.0 mL of acetonitrile and passed througha 0.2-�m PTFE filter (Macherey-Nagel) to prevent any suspendedparticles from entering into the HPLC column. Finally, the solutionwas injected into the HPLC-FL system for analysis. The extractionswere repeated in triplicate (n = 3). With each series of extractions,a similar blank procedure was also conducted without any samplematerial. After optimization, this method was applied to deter-mine the prevalence of the PAHs in different fish and macroalgaesamples.

2.7. Liquid chromatographic condition

All sample injections were held constant at 20 �L through theuse of a fixed-volume injection loop. The flow rate of the mobilephase (A) acetonitrile and (B) water was set at 1.0 mL min−1. ThePAHs were separated using the following gradient program: 65%A was maintained for 2 min, a linear gradient was used from 65%A at 2 min to 100% A at 14 min, and 100% A was maintained for10 min. Thereafter, the mobile phase was returned to the initialcondition (35% A) and held for 5 min to prepare the column for thenext injection. The column temperature was maintained at 25 ◦C.The chromatographic data were acquired at the following exci-tation/emission wavelengths: 250/375 nm (Ph, An), 237/440 nm(Flt), 270/376 nm (Pyr), 330/410 nm (BaA, Chry, and BeP), and384/410 nm (BaP). It was possible to simultaneously monitor 5ex/em pairs, hence only one injection was required for the anal-ysis. The quantitative analysis of individual PAHs in the sampleswas carried out by comparing the peak areas of the individual PAHswith the peak areas of the PAHs in the standard mixture.

2.8. Soxhlet extraction method

According the procedure described by S. Mora et al. [32], twograms of the spiked sample was extracted using Soxhlet for 8 hwith 250 mL of methanol. The extracts were then saponified byadding 20 mL of 0.7 M KOH and 30 mL of water and refluxing for2 h. The resulting mixture was transferred into a separating fun-nel and extracted 3 times with n-hexane (once with 90 mL, twicewith 50 mL). Then the extracts were combined, filtered throughglass wool and dried with anhydrous sodium sulfate. The extractswere concentrated by rotary evaporation down to 5 mL. Finally,the extract was cleaned up and fractionated by passing it througha silica/alumina column. Elution was performed using 20 mL ofn-hexane to yield the first fraction (containing the aliphatic hydro-carbons), then 30 mL of n-hexane: dichloromethane (90:10) andfollowed by 20 mL of n-hexane: dichloromethane (50:50). Thesetwo eluents containing the aromatic hydrocarbons (PAH) werecombined, rotary evaporated to near dryness and finally dilutedin 1 mL acetonitrile for HPLC-FL analysis.

3. Results and discussion

To achieve maximum efficiency, the effect of experimental vari-ables such as the type and volume of DES, dissolution time andtemperature, amount of sample, type and volume of extracting sol-vent, and extraction time were investigated and optimized. Theseparameters control the dissolution of biological samples in DESsand the extraction of PAHs from their matrices.

3.1. Optimizing the dissolution and extraction parameters

3.1.1. Effect of the type and composition of DESDESs are a unique class of multi-component solvent systems

with varying physico-chemical properties [22,23]. To facilitate theextraction of PAHs from their matrices to organic solvent, it was

Z. Helalat–Nezhad et al. / J. Chromatogr. A 1394 (2015) 46–53 49

Fig. 1. Effect of the composition of ChCl:Ox on the recovery of PAHs. Conditions:volume of ChCl:Ox, 2.5 mL; dissolution temperature, 55 ◦C; dissolution time, 30 min;amount of sample, 0.10 g; volume of cyclohexane, 5 mL; extraction time, 20 min.

Fdv

npctmrDSowm9

3

o(sp

irPtCcoicOdt

ig. 2. Effect of the volume of ChCl:Ox (1:2) on the recovery of PAHs. Conditions:issolution temperature, 55 ◦C; dissolution time, 30 min; amount of sample, 0.10 g;olume of cyclohexane, 5 mL; extraction time, 20 min.

ecessary to completely dissolve the biological sample in an appro-riate DES. Changing the type and/or the composition of the DESan significantly impact the sample dissolution and efficient extrac-ion of analytes from their matrices. In this experiment, ChCl was

ixed separately with Ox, at 1:1, 1:2, and 1:3 (ChCl: Ox) molaratios. Then, 0.10 g of the spiked sample was added to 2.5 mL of eachES under investigation and the procedure conducted as stated inection 2.6. As can be seen in Fig. 1, the best extraction recoveriesver the range 73.4–102.0% were obtained when the spiked sampleas completely dissolved in ChCl: Ox (1:2). At 1:1 and 1:3 ChCl: Oxolar ratios, the extraction recoveries of all PAHs did not exceed

3.3% and 79.4%, respectively.

.1.2. Effect of the volume of ChCl-OxThe volume of ChCl-Ox has a significant effect on the dissolution

f the biological samples and its constituents. Different volumes1.5–3.5 mL) of ChCl-Ox (1:2) were added to 0.10 g of the spikedample and the samples were processed according to the generalrocedure.

As can be seen in Fig. 2, when the volume of ChCl-Ox (1:2)ncreased from 1.5 to 2.5 mL, the extraction efficiencies increasedapidly, and with 2.5 mL of ChCl-Ox (1:2), about 78.5–101.1% ofAHs were extracted into the organic solvent; this trend was rela-ively constant up to 3.0 mL of ChCl-Ox (1:2). However, when thehCl-Ox (1:2) volume was increased to 3.5 mL, the extraction effi-iency of all studied PAHs decreased slightly, compared to 2.5 mLf ChCl-Ox (1:2). The volume of organic solvent used for extract-ng PAHs from 1.5 to 3.0 mL of ChCl-Ox was sufficient, but this

ould not extract completely all PAHs from larger volumes of ChCl-x, because of incomplete mixing of the organic solvent with theissolved biological sample and the studied DES. This indicatedhat 2.5 mL was the optimum volume of ChCl-Ox (1:2), which was

Fig. 3. Effect of the dissolution temperature on the recovery of PAHs. Conditions:volume of ChCl:Ox (1:2), 2.5 mL dissolution time, 30 min; amount of sample, 0.10 g;volume of cyclohexane, 5 mL; extraction time, 20 min.

adopted in subsequent tests. Lower volumes of ChCl-Ox (1:2) wereinsufficient for complete dissolution of the biological samples andled to lower extraction efficiency and precision.

3.1.3. Effect of dissolution temperatureTemperature can influence several physico-chemical properties

of a eutectic-based solvent such as viscosity, surface tension, diffu-sivity, conductivity, and heat capacity [33]. Increasing temperaturegenerally reduces the viscosity and the surface tension of a DESand increases the diffusivity of the solvent, which can facilitatethe penetration of DES in the matrices of the sample, leading tomore breaking down of the intermolecular bonds, and thereforeenhancing their solubility [28]. In this work, the effect of increas-ing temperature on the solubility and extraction of PAHs from theirmatrices was studied over the range 50–70 ◦C. As can be seen inFig. 3, the best extraction recoveries of 76.2–113.5% for all stud-ied PAHs were obtained when the temperature was held between55–60 ◦C. At temperatures greater than 65 ◦C, the extraction recov-eries of both the 3- and 4-ring PAHs were rapidly reduced, butthe extraction recoveries of the 5-ring PAHs reached 64.8–74.9%.This effect may be related to the more evaporation of the 3- and4-ring PAHs at higher temperatures, because of their inherentlyhigher vapor pressure, compared to the 5-ring PAHs. At temper-atures lower than 50 ◦C, the dissolution of the sample in ChCl-Ox(1:2) was incomplete and led to relatively low extraction recover-ies, for some PAHs. Therefore, to obtain quantitative and preciseresults, the temperature of the ChCl-Ox was set at 55 ◦C. Comparedto microwave-assisted and accelerated extraction methods, thismethod can proceed successfully at relatively lower temperaturesand at atmospheric pressure [34–36].

3.1.4. Effect of extraction conditionsDepending on the extraction technique and type of sample

matrix, various organic solvents have been used for the extractionof PAHs [37,38]. The nature of the solvent is of prime importanceand should meet three criteria: efficiency, selectivity, and compati-bility. Therefore, the following considerations should be taken intoaccount when selecting an appropriate solvent for efficient extrac-tion of PAHs. First of all, PAHs should have high solubility in thesolvent. Secondly, the solvent should be immiscible or have low sol-ubility in DESs to minimize the loss of solvent in the eutectic-basedsolution. Thirdly, the solvent should have low vapor pressure toprevent evaporative loss during extraction. Based on these criteria,

and from our own experience and literature reports, four organicsolvents, dichloromethane, toluene, cyclohexane, and n-hexanewere studied. Fig. 4 shows that the best recoveries (72%–110%) ofall PAHs were achieved using cyclohexane and dichloromethane

50 Z. Helalat–Nezhad et al. / J. Chromatogr. A 1394 (2015) 46–53

Fig. 4. Effect of the type of extracting solvent on the recovery of PAHs. Conditions:volume of ChCl:Ox (1:2), 2.5 mL; dissolution temperature, 55 ◦C; dissolution time,30 min; amount of sample, 0.10 g; extraction time, 20 min.

Fig. 5. Effect of the extraction time on the recovery of PAHs. Conditions: volumeoa

frmducih

sw5wotaeo2

tactanss

pttt

Fig. 6. Effect of the volume of cyclohexane on the recovery of PAHs. Conditions:volume of ChCl:Ox (1:2), 2.5 mL; dissolution temperature, 55 ◦C; dissolution time,30 min; amount of sample, 0.10 g; extraction time, 20 min.

Fig. 7. Effect of the dissolution time on the recovery of PAHs. Conditions: volume

f ChCl:Ox (1:2), 2.5 mL; dissolution temperature, 55 ◦C; dissolution time, 30 min;mount of sample, 0.10 g; volume of cyclohexane, 5 mL; extraction time, 20 min.

or extraction. Solvents such as n-hexane and toluene providedecoveries in the range of 41%–96%. However, cyclohexane wasore selective than dichloromethane for extraction of PAHs, since

ichloromethane extracted larger amounts of lipids that were sol-ble and/or turn to solid at room temperature and then partlyo-eluted with PAHs during the cleanup stage. These results aren accordance with previous reports [5,34–37]. Therefore, cyclo-exane was selected as a proper extracting solvent.

Next, the time required for extracting the analytes from the dis-olved sample (in ChCl-Ox) to the organic solvent (cyclohexane)as studied over the range of 5–30 min. As can be seen in Fig. 5, after

min of solvent extraction, the recoveries of the 3- and 4-ring PAHsere in the range of 26.3 to 47.4% and the extraction efficiencies

f the 5-ring PAHs were around 82.0%. By increasing the extrac-ion time, the extraction recoveries of all PAHs were increased,nd reached maximum levels (80.7–115.8%) at 20 min. At higherxtraction times, a small decrease in the extraction of all PAHs wasbserved, and therefore the optimum extraction time was set at0 min.

The volume of the organic solvent is the next important issuehat was carefully studied in the range of 2.5 to 20 mL. The resultsre depicted in Fig. 6. As shown, using just 2.5 mL of cyclohexanean extract PAHs over the range of 54.4 to 77.5%. By increasinghe volume of the extracting solvent to 5 mL, the recoveries werelso increased and ranged between 85.2 to 108.7%. This trend didot change considerably up to 10 mL. Therefore, to reduce the con-umption of the organic solvent, the extraction volume of 5 mL waselected as the optimum value.

As reported, the consumption of organic solvents in the pro-osed method is very low, guaranteeing safety of operators and

he environment. In addition, because of the quick dissolution ofhe studied biological samples in ChCl-Ox, the quantitative extrac-ion of PAHs can be done within a reasonable amount of time and

of ChCl:Ox (1:2), 2.5 mL; dissolution temperature, 55 ◦C; amount of sample, 0.10 g;volume of cyclohexane, 5 mL; extraction time, 20 min.

without using any apparatus like ultrasonic and microwave gen-erators.

3.1.5. Effect of the dissolution timeDepending on the temperature and type of the DES and analyte,

varying time intervals between 25 to 60 min have been reported forthe dissolution of biological samples and extraction of their ingredi-ents [27,28,38]. The dissolution time of the sample in ChCl-Ox (1:2)is an important issue that can influence the extraction recoveries ofPAHs. Therefore, the effect of the dissolution time on the extractionwas studied at five time intervals between 20–60 min. According tothe results presented in Fig. 7, the extraction recoveries of all PAHsincreased as the dissolution time of the sample in ChCl-Ox (1:2) wasincreased from 20 to 40 min. The best results (recoveries and preci-sions) for all PAHs was obtained when the sample was completelydissolved in ChCl-Ox (1:2) within 30 min, therefore this time wasselected as the optimum dissolution time. However, above 40 min,about 20 to 65% decrease for the extraction of the 3- and 4-ringPAHs was observed, compared to the maximum recoveries. In con-trast, only up to 35% decrease in the recoveries of the 5-ring PAHswas reported, and the results were still repeatable. This trend maybe related to slight evaporation of studied PAHs due to heating forlonger times after the complete dissolution of the sample in theeutectic solvent.

The time, on average, required for dissolution in ChCl-Ox andquantitative extraction of PAHs is comparable to that (about30 min) in the MAE and PLE methods, without applying high

pressures or sophisticated technologies to accelerate the extractionprocess [18,34,36]. This represents a high potential of the eutectic-based green solvents for use in different extraction technologies.

Z. Helalat–Nezhad et al. / J. Chromatogr. A 1394 (2015) 46–53 51

Table 1Analytical performance of the proposed method at optimized conditions.

Analytes Linear range (ng g−1) r2 LOD (ng g−1) LOQ (ng g−1) Intra-day precision (RSD%, n = 5) Inter-day precision a (RSD%, n = 5)

Ph 5.0–250 0.9995 3.08 3.50 12.2 11.1An 5.0–250 0.9994 2.91 4.00 9.9 10.2Flu 5.0–250 0.9981 0.77 2.05 9.8 13.3Pyr 2.0–250 0.9989 0.62 1.50 12.6 8.3BaA 1.0–250 0.9997 0.50 0.85 7.5 6.6Chry 1.0–250 0.9997 0.65 0.90 5.4 7.5

6.1 5.63.9 4.0

3

2psIeept

3

cseadsda

fs5ttrtttwo

3

tab2ia

nq1

Pcor

BeP 2.0–250 0.9988 0.79 1.07

BaP 1.0–250 0.9960 0.53 0.90

a Based on 5 days.

.1.6. Effect of the amount of sampleThe effect of the amount of sample that can be dissolved in

.5 mL ChCl-Ox was studied at 0.10, 0.15, and 0.20 g, using thereviously determined optimized conditions. Before analysis, theamples were spiked with a constant amount (10 ng) of each PAH.t was found that with increasing the amount of the sample, thextraction recovery of all PAHs gradually decreased, and the highestxtraction recoveries can be obtained with 0.10 g of the sam-le. Therefore, 0.10 g of the fish or macroalgae sample was usedhroughout.

.1.7. Cleanup stageExtracts from foods, biological, and environmental matrices

ontain not only the hydrophobic analytes of interest but alsoeveral other hydrophobic compounds, frequently lipids. These co-xtracted components not only interfere with the analytes butlso frequently affect or damage the chromatographic column,ecreasing its useful life and deteriorating the chromatographicystem performance [5]. Among the several cleaning procedureseveloped for PAHs extracts, those based on solid-phase extractionre the most convenient in practice.

In this experiment, 0.30 g of florisil and/or silica gel was usedor cleanup of the extracted PAHs. After passing all the extractedolvent (5 mL) through the sorbent, the column was washed with

mL of cyclohexane. The collected solvents were evaporated inhe rotary evaporator, and the steps were followed as described inhe general procedure. However, both the sorbents gave the sameesults for the 3- and 4-ring PAHs, but the extraction recoveries ofhe 5-ring PAHs obtained by the silica gel were up to 7.0% higherhan those using florisil. Therefore, the silica gel was selected ashe appropriate sorbent for cleaning up the extracts. The same trendas observed by Pena et al. [5] during cleaning up the PAHs extracts

btained from fish samples, after applying MAE.

.2. Analytical performance

Optimized experimental conditions were employed to validatehe analytical performance of the proposed method. The resultsre summarized in Table 1. The linearity of the calibration plotsased on the analyte peak areas was obtained between 1.0–250,.0–250, and 5.0–250 ng g−1, depending on the analytes, with sat-

sfactory coefficients of determination (r2) ranging between 0.9960nd 0.9997.

Limits of detection (LODs) of the method, based on a signal-to-oise ratio of 3, were in the range of 0.50–3.08 ng g−1. The limit ofuantification (LOQ), which is defined as a signal-to-noise ratio of0:1, was obtained in the range of 0.85–4.00 ng g−1.

The intra-day and inter-day precisions (5 days) of the spiked

AHs at concentration levels of 50 ng g−1 were evaluated by cal-ulating the relative standard deviations (RSD) of 5 replicate runsf the proposed procedure. The intra-day precisions were in theange of 3.9–2.6%, showing good repeatability of the method. The

Fig. 8. Comparison of the Soxhlet method and the proposed method.

inter-day precisions based on replications at 5 different days werein the range of 4.0–13.3%.

3.3. Comparison

The Soxhlet extraction technique has been commonly used as abenchmark technique in the extraction of PAHs from environmen-tal samples [36]. Therefore, the extraction recovery of the proposedmethod was compared with the data obtained by Soxhlet extrac-tion. To do this, the biological sample spiked with 50 ng g−1 of eachPAH was subjected separately to the Soxhlet procedure and theproposed method. The results are summarized in Fig. 8, based on 3replications. It can be clearly seen that the performance of the cur-rent method is considerably higher than that of the Soxhlet method.The observations (Fig. 8) show that the eutectic solvent is a keyfactor in the efficient dissolution of the total sample, which thenfacilitated the fast extraction of the analytes from their matrices tothe organic solvent. All these facts indicate that in addition to sav-ing time and energy as well as considerably low consumption oforganic solvent, higher extraction recoveries can also be obtainedby the proposed method, compared to Soxhlet extraction.

The extraction results obtained are also comparable with somerecent extraction methods like matrix solid-phase dispersion[39], ultrasonication extraction coupled with magnetic solid-phaseclean-up [6], homogeneous liquid–liquid extraction [40], andacid-induced cloud point extraction [41]. However, this ChCl-Oxextraction method has demonstrated a particular advantage in theoverall volume of organic solvent, which was much less than thoseof the above-mentioned procedures. Additionally, because of thecomplete dissolution of the sample in ChCl-Ox, the studied PAHscan be extracted quickly, without exposure to high temperatures

and pressures like that applied in the routine extraction methodssuch as MAE and PLE.

52 Z. Helalat–Nezhad et al. / J. Chromatogr. A 1394 (2015) 46–53

Table 2Concentration of individual PAHs found in fish and macroalgae samples by the proposed method.

Platichthys flesus Otolithes ruber Enteromorpha intestinalis

PAHs Spiked Detected Recovery (RSD) a Spiked Detected Recoveries (RSD) Spiked Detected Recoveries (RSD)

(ng g−1) (ng g−1) (%) (%) (ng g−1) (ng g−1) (%) (%) (ng g−1) (ng g−1) (%) (%)

Ph – 200.8 – – 151.4 – – 88.3 –50 242.7 83.8 (2.8) 50 201.9 101.0 (2.4) 50 125.4 74.2 (2.1)An – 5.8 – – <LODb – – <LOD –50 44.6 77.6 (10.0) 50 37.6 75.2 (8.8) 50 36.3 72.6 (5.7)Flu – <LOD – – <LOD – – <LOD –50 52.8 105.6 (11.7) 50 45.3 90.6 (6.5) 50 43.6 87.2 (3.6)Pyr – 5.2 – – 2.6 – – 3.5 –50 56.6 102.8 (8.3) 50 47.9 90.6 (7.1) 50 42.2 77.4 (5.5)BaA – 1.0 – – <LOD – – <LOD –50 45.8 89.6 (7.1) 50 43.5 87.0 (9.3) 50 36.3 72.6 (8.7)Chry – 1.3 – – 1.5 – – 1.7 –

50 47.0 91.4 (10.5) 50 45.6 88.2 (4.7) 50 38.6 73.8 (10.4)BeP – <LOD – – 2.9 – – 4.2 –50 46.0 92.0 (5.2) 50 47.8 89.8 (2.9) 50 59.0 109.6 (3.0)BaP – 2.5 – – 6.6 – – 3.8 –50 42.4 79.8 (9.4) 50 49.4 85.6 (4.2) 50

a n = 5.b <LOD = lower than limit of detection.

Ffi

3

bfiPimtteoaschoaFsrhsAds

[4] S.S. Cai, J. Stevens, J.A. Syage, Ultra high performance liquid

ig. 9. HPLC–FL chromatograms of the analysis of the non-spiked (A) and spiked (B)sh samples by.

.4. Analysis of macroalgae and fish samples

The practical utility of the proposed method was evaluatedy the determination of PAHs in marine biological samples (i.e.,sh muscle and macroalgae), collected from different sites of theersian Gulf. The muscles of two kinds of marine fish samples,ncluding Kafshak (P. flesus) and Shoorideh (O. ruber) as well as the

arine macroalgae (E. intentinalis) were selected for study usinghis method. After pre-treatment, each sample was subjected tohe proposed procedure, under optimal conditions. Additionally,ach sample was spiked with 50 ng g−1 of PAH, and the recoveriesf extraction as well as the precisions (RSDs, n = 5) were calculatednd summarized in Table 2. The extraction recoveries of the spikedamples ranged from 70.0% to 109.6%, with RSDs of below 11.7%. Asan be seen, the concentration of Ph in all samples is considerablyigher than other PAHs. Similarly, S. de Mora et al. [32] pointedut the considerably high concentrations of Ph in water, sediment,nd fish samples collected from the north of the Persian Gulf. Inig. 9, the HPLC-FL chromatograms of the extracted PAHs from thepiked (at 50 ng g−1) and non-spiked samples of a fish muscle of O.uber are presented. Since the concentration of Ph was relativelyigh, the other studied PAHs were not clearly observed in the non-piked chromatogram, but each of PAHs was detected by HLPC-FL.

ll these facts demonstrate that this new ChCl-Ox extraction proce-ure has a great potential for application to genuine environmentalamples.

39.6 71.6 (4.9)

4. Conclusion

Deep eutectic solvents are a new class of green solvents thathave recently gained high utility in various fields such as sepa-ration, dissolution, and extraction. In the present work, a novelextraction technique based on the complete dissolution of marinebiological samples in ChCl-Ox deep eutectic solvent was developedfor convenient and quantitative extraction of PAHs. The extractedPAHs were then determined with HPLC-FL. This method providessome operational advantages such as simplicity of experimentalprocedure, use of low-cost eutectic solvent, relatively high-speedsample preparation, consumption of low amount of sample, andlow organic solvent usage. Thus, this method demonstrates poten-tial for routine trace PAH analysis in biological samples. Comparedto routine extraction methods, such as MAE and PLE, the process ofdissolution and extraction of PAHs can be conducted at relativelylower temperatures at atmospheric pressure. Under the most favor-able conditions, the developed method exhibited particularly goodlimits of detection, linearity, and satisfactory repeatability. Overall,this study demonstrated that the proposed method is an efficientapproach for the sample preparation of marine biological samplessuch as fish muscle and macralgae. This new approach opens aninteresting and innovative way in the field of sample preparationmethods for effectively handling complex samples.

Acknowledgments

The financial and technical support provided by the ResearchCouncil of Khorramshahr University of Marine Science and Tech-nology through grant (2013) to conduct this study is gratefullyacknowledged.

References

[1] M. Guardia, S. Garrigues, in: M. Guardia, S. Garrigues (Eds.), Challenges in Ana-lytical Chemistry, RSC Publishing, Cambridge, UK, 2011, pp. 1–12.

[2] J. Pawliszyn, Sample Preparation: Quo Vadis? Anal. Chem. 75 (2003)2543–2558.

[3] M.J. Ramalhosa, P. Paíga, S. Morais, C. Delerue–Matos, M.B.P.P. Oliveira, Analysisof polycyclic aromatic hydrocarbons in fish: evaluation of a quick, easy, cheap,effective, rugged, and safe extraction method, J. Sep. Sci. 32 (2009) 3529–3538.

chromatography–atmospheric pressure photoionization-mass spectrom-etry for high-sensitivity analysis of US Environmental Protection Agencysixteen priority pollutant polynuclear aromatic hydrocarbons in oysters, J.Chromatogr. A 1227 (2012) 138–144.

hrom

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

Z. Helalat–Nezhad et al. / J. C

[5] T. Pena, L. Pensado, C. Casais, C. Mejuto, R. Phan–Tan–Luu, R. Cela, Optimiza-tion of a microwave-assisted extraction method for the analysis of polycyclicaromatic hydrocarbons from fish samples, J. Chromatogr. A 1121 (2006) 163–169.

[6] F. Yang, Y. Long, R. Shen, C. Chen, D. Pan, Q. Zhang, Q. Cai, S. Yao, Ultrasonicationextraction coupled with magnetic solid-phase clean-up for the determinationof polycyclic aromatic hydrocarbons in soils by high-performance liquid chro-matography, J. Sep. Sci. 34 (2011) 716–723.

[7] K. Kalachova, J. Pulkrabova, L. Drabova, T. Cajka, V. Kocourek, J. Hajslova, Sim-plified and rapid determination of polychlorinated biphenyls, polybrominateddiphenyl ethers, and polycyclic aromatic hydrocarbons in fish and shrimpsintegrated into a single method, Anal. Chim. Acta 707 (2011) 84–91.

[8] J. Sánchez–Avila, R. Tauler, S. Lacorte, Organic micropollutants in coastal watersfrom NW Mediterranean Sea: Sources distribution and potential risk, Environ.Int. 46 (2012) 50–62.

[9] A. Mohammadi, V. Ghasemzadeh–Mohammadi, P. Haratian, R. Khaksar, M.Chaichi, Determination of polycyclic aromatic hydrocarbons in smoked fishsamples by a new microextraction technique and method optimisation usingresponse surface methodology, Food Chem. 141 (2013) 2459–2465.

10] L. Liguori, K. Heggstad, H.T. Hove, K. Julshamn, An automated extractionapproach for isolation of 24 polyaromatic hydrocarbons (PAHs) from variousmarine matrixes, Anal. Chim. Acta 573–574 (2006) 181–188.

11] C. Anyakora, in: L.M.L. Nollet, L.S.P. DeGelder (Eds.), Handbook of Water Anal-ysis, 3rd ed., CRC Press, Taylor & Francis Group, 2014, pp. 807–818.

12] M.D. Luque de Castro, F. Priego–Capote, Soxhlet extraction: Past and presentpanacea, J. Chromatogr. A 1217 (2010) 2383–2389.

13] T. Reemtsma, J.B. Quintana, in: T. Reemtsma, M. Jekel (Eds.), Organic Pollutantsin Water Cycle, WILEY–VCH Verlag GmbH & Co. KGaA, Weinheim, 2006, pp.1–6.

14] D. Castro, K. Slezakova, M.T. Oliva–Teles, C. Delerue–Matos, M.C. Alvim–Ferraz,S. Morais, M.C. Pereira, Analysis of polycyclic aromatic hydrocarbons in atmo-spheric particulate samples by microwave assisted extraction and liquidchromatography, J. Sep. Sci. 32 (2009) 501–510.

15] M.N. Baig, G.A. Leeke, P.J. Hammond, R.C.D. Santos, Modelling the extraction ofsoil contaminants with supercritical carbon dioxide, Environ. Pollut. 159 (2011)1802–1809.

16] N. Barco–Bonilla, J.L.M. Vidal, A.G. Frenich, R. Romero–González, Comparisonof ultrasonic and pressurized liquid extraction for the analysis of polycyclicaromatic compounds in soil samples by gas chromatography coupled to tandemmass spectrometry, Talanta 78 (2009) 156–164.

17] C. Sánchez–Brunete, E. Miguel, J.L. Tadeo, Rapid method for the determinationof polycyclic aromatic hydrocarbons in agricultural soils by sonication-assistedextraction in small columns, J. Sep. Sci. 29 (2006) 2166–2172.

18] L. Guo, H.K. Lee, Microwave assisted extraction combined with solventbar microextraction for one-step solvent-minimized extraction, cleanup andpreconcentration of polycyclic aromatic hydrocarbons in soil samples, J. Chro-matogr. A 1286 (2013) 9–15.

19] M. Germán–Hernández, V. Pino, J.L. Anderson, A.M. Afonso, A novel in situpreconcentration method with ionic liquid-based surfactants resulting inenhanced sensitivity for the extraction of polycyclic aromatic hydrocarbonsfrom toasted cereals, J. Chromatogr. A 1227 (2012) 29–37.

20] B. Tang, W. Bi, M. Tian, K.H. Row, Application of ionic liquid for extraction andseparation of bioactive compounds from plants, J. Chromatogr. B 904 (2012)1–21.

21] M.C. Bubalo, K. Radosevic, I.R. Redovnikovic, J. Halambek, V.G. Srcek, A briefoverview of the potential environmental hazards of ionic liquids, Ecotox. Envi-ron. Safe. 99 (2014) 1–12.

22] Q. Zhang, K.D.O. Vigier, S. Royer, F. Jérôme, Deep eutectic solvents: syntheses,

properties and applications, Chem. Soc. Rev. 41 (2012) 7108–7146.

23] T. Beyersdorff, T.J.S. Schubert, U. Welz-Biermann, W. Pitner, A.P. Abbott,K.J. McKenzie, K.S. Ryder, in: F. Enders, A.P. Abbott, D.R. MacFarlane (Eds.),Electrodeposition from ionic liquids, WILEY–VCH Verlag GmbH & Co. KGaA,Weinheim, 2008, pp. 15–42.

[

atogr. A 1394 (2015) 46–53 53

24] A.P. Abbott, G. Capper, D.L. Davies, K.J. McKenzie, S.U. Obi, Solubility of metaloxides in deep eutectic solvents based on choline chloride, J. Chem. Eng. Data51 (2006) 1280–1282.

25] H.G. Morrison, C.C. Sun, S. Neervannan, Characterization of thermal behavior ofdeep eutectic solvents and their potential as drug solubilization vehicles, Int. J.Pharm. 378 (2009) 136–139.

26] A. Hayyan, M.A. Hashim, F.S. Mjalli, M. Hayyan, I.M. AlNashef, A novelphosphonium-based deep eutectic catalyst for biodiesel production fromindustrial low grade crude palm oil, Chem. Eng. Sci. 92 (2013) 81–88.

27] Y. Dai, G. Witkamp, R. Verpoorte, Y.H. Choi, Natural deep eutectic solvents as anew extraction media for phenolic metabolites in Carthamus tinctorius L, Anal.Chem. 85 (2013) 6272–6278.

28] W. Bi, M. Tian, K.H. Row, Evaluation of alcohol-based deep eutectic solvent inextraction and determination of flavonoids with response surface methodologyoptimization, J. Chromatogr. A 1285 (2013) 22–30.

29] E. Habibi, K. Ghanemi, M. Fallah–Mehrjardi, A. Dadolahi–Sohrab, A novel diges-tion method based on a choline chloride–oxalic acid deep eutectic solvent fordetermining Cu, Fe, and Zn in fish samples, Anal. Chim. Acta 762 (2013) 61–67.

30] A. Battaglia, D.J. Morgan, D.J. Linz, T.D. Hayes, in: P.T. Kostecki, E.J. Calabrese, M.Bonazountas (Eds.), Hydrocarbon Contaminated Soils, vol. 2, Lewis Publishers,Chelsea, 1992, pp. 257–278.

31] Y. Dai, J.V. Spronsen, G. Witkamp, R. Verpoorte, Y.H. Choi, Natural deep eutecticsolvents as new potential media for green technology, Anal. Chim. Acta 766(2013) 61–68.

32] S. Mora, I. Tolosa, S.W. Fowler, J. Villeneuve, R. Cassi, C. Cattini, Distribution ofpetroleum hydrocarbons and organochlorinated contaminants in marine biotaand coastal sediments from the ROPME Sea Area during 2005, Mar. Pollut. Bull.60 (2010) 2323–2349.

33] K. Shahbaz, F.S. Mjalli, M.A. Hashim, I.M. AlNashef, Prediction of deep eutec-tic solvents densities at different temperatures, Thermochim. Acta 515 (2011)67–72.

34] K. Oukebdane, F. Portet–Koltalo, N. Machour, F. Dionnet, P.L. Desbène, Com-parison of hot Soxhlet and accelerated solvent extractions with microwaveand supercritical fluid extractions for the determination of polycyclic aromatichydrocarbons and nitrated derivatives strongly adsorbed on soot collectedinside a diesel particulate filter, Talanta 82 (2010) 227–236.

35] N. Itoh, A. Fushimi, T. Yarita, Y. Aoyagi, M. Numata, Accurate quantification ofpolycyclic aromatic hydrocarbons in dust samples using microwave-assistedsolvent extraction combined with isotope-dilution mass spectrometry, Anal.Chim. Acta 699 (2011) 49–56.

36] W. Wang, B. Meng, X. Lu, Y. Liu, S. Tao, Extraction of polycyclic aromatichydrocarbons and organochlorine pesticides from soils: A comparison betweenSoxhlet extraction, microwave-assisted extraction and accelerated solventextraction techniques, Anal. Chim. Acta 602 (2007) 211–222.

37] H. Zhang, B. Tang, K.H. Row, A green deep eutectic solvent-based ultrasound-assisted method to extract astaxanthin from shrimp byproducts, Anal. Lett. 47(2014) 742–749.

38] S. Mulyono, H.F. Hizaddin, I.M. Alnashef, M.A. Hashim, A.H. Fakeeha, M.K.Hadj–Kali, Separation of BTEX aromatics from n-octane using a (tetra-butylammonium bromide + sulfolane) deep eutectic solvent–experiments andCOSMO-RS prediction, RSC Adv. 4 (2014) 17597–17606.

39] V.F. González, E. Concha–Grana, S. Muniategui–Lorenzo, P. López–Mahía,E. Fernández–Fernández, D. Prada–Rodríguez, A matrix solid-phasedispersion–gel permeation chromatography–programmed temperaturevaporisation–GC-MS procedure for the analysis of polycyclic aromatichydrocarbons in mussel samples, J. Sep. Sci. 33 (2010) 3741–3750.

40] X. Zhao, X. Liu, Z. Zhao, C. Huang, M. Zhang, H. Wang, X. Wang, Homo-geneous liquid–liquid extraction combined with high performance liquid

chromatography– fluorescence detection for determination of polycyclic aro-matic hydrocarbons in vegetables, J. Sep. Sci. 32 (2009) 2051–2057.

41] F. Merino, S. Rubio, D. Pérez–Bendito, Acid-induced cloud point extractionand preconcentration of polycyclic aromatic hydrocarbons from environmentalsolid samples, J. Chromatogr. A 962 (2002) 1–8.