Trace Determination of Petroleum Pollutants in Water Samples by Dispersive Liquid-Liquid Microextraction Method

Kobra Sadat Hasheminasab1

Ali Reza Fakhari 1

Majid Baghdadi 2

1Faculty of Sciences, Department ofChemistry, Shahid BeheshtiUniversity, Tehran, Iran

2Faculty of Environment, University ofTehran, Tehran, Iran

Research Article

Trace Determination of Petroleum Pollutants inWater Samples by Dispersive Liquid–LiquidMicroextraction Method

Since there is not any report about simultaneous microextraction of 42 petroleumpollutants (n-alkanes and polycyclic aromatic hydrocarbons) in water, in the presentstudy dispersive liquid–liquid microextraction (DLLME) using organic solvents lighterthan water was applied for simultaneous determination of these compounds in watersamples by GC-flame ionization detection (GC-FID). In order to achieve the bestextraction efficiency, optimization of variables affecting DLLMEmethodwas carried out.In this method, an appropriate mixture of extraction and disperser solvents is injectedinto an aqueous sample, which led to formation of a cloudy solution. The analytesbecome enriched in the extraction solvent, which is dispersed throughout the bulkaqueous sample. After centrifugation, quantification of the analytes in the organic phasecan be performed by GC-FID. The main advantage of this technique is its quite largesurface area between the fine droplets of the extraction solvent and water sample, andthe accordingly fast extraction kinetics that results in the rapid achieving of a state ofequilibrium and the high enrichment factors (EFs). Under the optimum conditions, EFsin the range of 280–446 were obtained. The proposed method provided low limits ofdetection (0.1–0.98mg L�1) and acceptable extraction repeatabilities. It was linear overtwo orders of magnitude. Finally, petroleum pollutants have been determined indifferent type of water samples by this method.

Keywords: n-Alkanes; Analytical techniques, Polycyclic aromatic hydrocarbons (PAHs), Samplepreparation

Received: January 26, 2013; revised: August 5, 2013; accepted: August 21, 2013

DOI: 10.1002/clen.201300044

1 IntroductionIn recent years, there has been considerable interest in thedetermination of contaminants of petroleum hydrocarbons inresidential indoor air, soil, and water samples [1]. Considerableamounts of petroleum compounds are discharged into the environ-ment through industrial effluents [2]. They are considered as one ofthe most frequent soil and groundwater pollutants due to theirleakages during storage and racking by transportation lines orimprudent handling, and by accidents [3–5]. It has been estimatedthat around 20� 106 t of petroleum hydrocarbons pollute marinewaters and estuaries annually [6]. The nature of petroleum

hydrocarbon contamination is highly variable. The more commonmost functional categories of compounds found in petroleum

products are n-alkanes, branched alkanes, cycloalkanes, and aromaticcompounds (e.g., www.newenglandtesting.com) [7, 8].Normal alkanes (n-alkanes) from petroleum sources are an

important feed stock for the petrochemical industries. The longchain alkanes can be converted to lubricant and fuel additives,plasticizers, industrial surfactants, flotation agents, and solvents [9].n-Alkanes have a lower solubility and toxicity than aromatichydrocarbons with the same number of carbons. They are degradedby microorganisms to toxic polycyclic aromatic hydrocarbons(PAHs) [10]. In mammals, alkanes can solubilize fats or mucousmembranes and can penetrate myelin sheath nerve fibers [11].PAHs are hazardous organic chemicals consisting of two or

more benzenoid groups [12]. PAHs are composed primarily during theincomplete combustion of fossil fuels and coal, cracking process inpetrochemical industry, and degradation of lubricating oils anddyes [13]. These compounds are considered as priority pollutants dueto their carcinogenicity, high toxicity, and mutagenicity [14, 15].Considering the importance of petroleum hydrocarbons as

environmental pollutants, monitoring of these compounds in watersamples is very important for both human health and environment.Many standardmethods were presented by Environmental ProtectionAgency, American Society for Testing and Materials, and AmericanPetroleum Institute [13, 16–24].Many analytical techniques have been applied for the determina-

tion of petroleum contaminants. Between them, GC-flame ionization

Correspondence: Professor A. R. Fakhari, Faculty of Sciences, Departmentof Chemistry, Shahid Beheshti University, G. C., P.O. Box 19396-4716, Evin,Tehran, IranE-mail: [email protected]

Abbreviations: DLLME, dispersive liquid–liquid microextraction; EF,enrichment factor; GC-FID, GC-flame ionization detection; LLE, liquid–liquid extraction; LOD, limit of detection; PAH, polycyclic aromatichydrocarbon; R, recovery; RR, relative recovery; RSD, relative standarddeviation

1

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9

detection (GC-FID) and HPLC are widely used techniques [25, 26].GC–MS is used for identification and to provide structuralinformation for the individual petroleum compounds present inenvironmental samples [3, 27].Sample preparation is a critical step in the analytical procedure for

the analysis of petroleum hydrocarbons in environmental samples.Liquid–liquid extraction (LLE) has been the most commonly usedsample preparation method for aqueous samples. But this method istime consuming and requires large amounts of organic solvents thatare toxic and expensive [28]. Recently, microextraction methodshave gained interest in analytical chemistry because of the highenrichment factors (EFs), fast sample preparation, and low solventconsumption in, they have gained interest in the analyticalchemistry. Furthermore, they are compatible with many analyticalinstruments [29–31].In 2006, Assadi and co-workers developed a novel liquid-phase

microextraction technique, named dispersive liquid–liquid micro-extraction (DLLME) [32, 33]. This method is a miniaturized kind ofLLE that uses microliter volumes of extraction solvent. DLLME usesof a binary mixture of two organic solvents (extractant anddispersant) for the extraction and concentration of organic andinorganic species from aqueous samples. DLLME presents extremelyfast extraction kinetics, low cost, high EFs, and low consumption oforganic solvents [34]. It has been used for the extraction of a widevariety of organic compounds, i.e., PAHs [18, 35–40] phenols [41],herbicides [42], pesticides [43], polychlorinated biphenyls [44],carbendazim, and thiabendazole [45], heavy metals [46].Since there is no report about simultaneous extraction of 42

petroleum pollutants (n-alkanes and PAHs) in water samples usingmicroextraction methods, the present work aims mainly at theoptimization of DLLME method coupled to GC–FID for the simulta-neous analysis of petroleum pollutants in water samples. Theproposedmethod was successfully applied to different water samples.

2 Materials and methods

2.1 Chemicals and standard solutions

n-Hexane, n-heptane, and cyclohexane as extraction solvents wereSupraSolv grade and purchased from Merck (Darmstadt, Germany).Other chemicals such as methanol, acetonitrile, acetone, biphenyl(internal standard), and sodium chloride were of analytical grade andpurchased from Merck. PAHs and n-alkanes standard solutions werepurchased from Dr. Ehrenstorfer (Reference Materials, Augsburg,Germany). Working solutions containing both PAHs and n-alkaneswere prepared by proper dilution.

2.2 GC analysis

Separation and detection of analytes were performed using anAgilent 7890N Gas Chromatograph (Centerville Road, Wilmington,USA) equipped with an FID detector and a split/splitless injector. AnHP-5 (5% diphenyl and 95% dimethyl polysiloxane) fused silicacolumn (30m length, 0.32mm id, and 0.25mm film thickness) fromSupelco (Bellefonte, PA, USA) was employed. In order to dataacquisition and processing, ChemStation software was used. The GCoven temperature program was as follows: the oven temperature washeld at 60°C for 1min, then increased to 295°C at a rate of3.2°Cmin�1 and held for 6min at this temperature. The temperatureof the detector (FID) was 330°C and it was fed with 35mLmin�1 of

hydrogen, 350mLmin�1 of air, and 35mLmin�1 of nitrogen asauxiliary gas. The mode of injection was splitless at 320°C. Nitrogen(with 99.999% purity) was used as the carrier gas with a constantcolumn flow of 1mLmin�1.

2.3 Sample collection and conditioning

Water samples were collected from the Jajrood River (Tehran, Iran)and from Caspian Sea (Anzali, Iran). Water samples were collected inthe amber glass bottles, which were pre-washed with chromic acid,stored at 4°C.Water samples should be extractedwithin seven days ofsampling [47]. In order to assess contamination introduced duringsample preparation activities, a blank sample was prepared andit was extracted and analyzed exactly like the samples.

2.4 DLLME

An aliquot (8mL) of water sample was poured into a 10mL glass testtube, which was specially designed for collection of low densitysolvents [48]. Organic solution containing 15mL extraction solventand 1mL disperser solvent was rapidly injected into the watersample using 1mL syringe. Cloudy solution was rapidly formed assoon as the fine droplets of the immiscible extraction solventdispersed in the water sample and the analytes were extracted intothe formed fine droplets. After 1min, the formed emulsion wascentrifuged at 5000 rpm for 4min to separate the phases. A fewmicroliters of distilled water were added to the vial through theglass tube deliberately fixed on the side of the vial and the floatedorganic phase containing extracts ran up into the capillary tubeon the top of the vial [49]. Then this organic phase was collectedusing a 10mL microsyringe. 2mL of the collected organic phase wasthen directly injected to GC-FID.

2.5 Calculation of enrichment factor, extractionrecovery, and relative recovery

The EF was defined as the ratio of the final analyte concentrationin the organic phase (Corg, final) and the initial concentration ofanalyte in the sample solution (Cs, initial):

EF ¼ Corg;finalCs;initial

ð1Þ

Extraction recovery (R%) was calculated according to thefollowing equation for each analyte:

R% ¼ Vo

Vs

� �Corg;finalCs;initial

� �� 100 ¼ EF

Vo

Vs

� �� 100 ð2Þ

where Vo is the volume of organic phase, Vs is the volume of sample.Relative recovery (RR%) was acquired from the following equation:

RR% ¼ Cfound � CrealCadded

� 100 ð3Þ

where Cfound, Creal, and Cadded are the concentration of analyte afteraddition of known amount of standard into the real sample, theconcentration of analyte in real sample, and the concentration ofknown amount of standard which was spiked into the real sample,respectively.

2 K. S. Hasheminasab et al.

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9

3 ResultsIn order to obtain optimum extraction conditions, relative peak area(the ratio of the peak area for each analyte to peak area of internalstandard) was used to evaluate the extraction efficiency underdifferent conditions. The optimizationwas carried out on an aqueoussolution containing 50mg L�1 of each petroleum pollutants.

3.1 Optimization of DLLME operation parameters

3.1.1 Selection of the extraction solvent

Selection of the extraction solvent was done on the basis of its goodgas chromatographic behavior, extraction capability, and formationof an appropriate cloudy state in the presence of a disperser wheninjected into an aqueous solution [50].Traditional extraction solvents in DLLME are mostly chlorinated

ones (such as chloroform, dichloromethane, etc.), which are denserthan water. These solvents are high toxic and not suitable for theextraction of non-polar compounds [51]. Taking these considerations,we used less toxic solvents (n-hexane, n-heptane, and cyclohexane),which provide better extraction efficiency for non-polar analytes.

These solvents are lighter than water and must be collected from the

surface of the water sample. The process was done by taking 20mLof each solvent. As it is shown in Fig. 1, n-hexane and n-heptanehave relative preference over cyclohexane. n-Hexane due to its higherextraction efficiency than n-heptane was selected as the extractionsolvent and was used in further studies.

3.1.2 Selection of the disperser solvent

Themiscibility of the disperser solvent in the organic phase (extractionsolvent) and the aqueous phase (sample solution) is the mainpoint for the selection of the disperser solvent [52]. The dispersionabilities of acetone, acetonitrile, and methanol were investigatedin this section. Therefore, extraction was carried out using 1mL ofeach disperser solvent containing 20mL of n-hexane. Accordingto the extraction results (Fig. 2) acetone is selected as dispersersolvent.

3.1.3 Effect of the volume of extraction solvent

In DLLME, the volume of extraction solvent is an importantparameter because of its impact on the EF. In order to evaluatethe effect of volume of extraction solvent on the extraction efficiency,a series of experiments were performed using 1mL acetone anddifferent volumes of n-hexane (15, 20, 25, and 30mL). On the basis of

Figure 1. The effect of the extraction solvent on theextraction efficiency of (A) n-alkanes and (B) PAHcompounds obtained from DLLME. Extractionconditions: volume of the water sample, 8mL;volume of the disperser solvent (acetone), 1mL,analyte concentration 50mg L�1, room temperature.

Trace Determination of Petroleum Pollutants 3

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9

the results shown in Fig. 3, 15mL of n-hexane was selected forsubsequent experiments.

3.1.4 Effect of the volume of disperser solvent

The effect of the volume of disperser solvent on the extractionefficiency was investigated. Different volumes of acetone (0.25, 0.50,1.0, 1.5, and 2.0mL) containing an aliquot of 20mL n-hexane wereemployed. According to the results shown in Fig. 4, at low volumes ofacetone (<1mL), the cloudy state is not formedwell, so the extractionprocess is disturbed. Higher extraction efficiency was obtained using1mL acetone.

3.1.5 Effect of the extraction and centrifugation time

In DLLME method, extraction time is defined as an interval timebetween injection of mixture of disperser and extraction solvents,and centrifugation [49]. The effect of the extraction time wasinvestigated in the range of 0–25min. According to the results, theequilibrium time has no significant effect on the extractionefficiencies of analytes. Because the contact surface of the extractingsolvent/aqueous sample was extremely large, the state of equilibriumwas achieved instantly. Therefore, in experiments the centrifugationwas performed immediately after the emulsification. In order toachieve the highest extraction efficiency in the shortest time,extraction was performed for only 1min.

In DLLME method, the most time-consuming step is thecentrifugation of sample solution. Therefore, it is important toestablish the right centrifugation time to achieve phase separationand the highest analysis frequency. The effect of centrifugation timewas investigated in the range of 2–8min at 5000 rpm. Higherextraction efficiency was obtained using a 4min centrifugation time.

3.1.6 Effect of salt addition

In order to optimize the best amount of salt, different amounts ofNaCl (0–20%w/v) were added to the aqueous sample. Because of therapidmass transfer of these analytes in DLLME process, the amount ofNaCl had not any significant effect on the extraction efficiency.Therefore, NaCl was not added.

3.2 Method evaluation

To investigate the practical applicability of the DLLMEmethod underoptimized extraction conditions, the figures of merit analyticalcharacteristics were evaluated. The results are shown in Tab. 1. Thecalibration plots were constructed by extracting seven aqueousstandards containing all the analytes at increasing concentrations.In order to improve repeatability, biphenyl (40mg L�1) was added tothe standard solutions as an internal standard and the ratio ofthe peak area for each analyte to peak area of internal standardwas considered for drawing calibration curves. With this preconcen-tration technique, all the analytes exhibited good linearity over therange studied with good correlation coefficients (r2> 0.986). Errors

for the slope and intercept were calculated. In addition, the EF of eachanalyte was calculated based on the ratio of final concentration ofanalyte in the organic phase to initial concentration of analyte inthe water sample. The EFs were found in the range of 280–446.The intraday and interday relative standard deviations (RSDs) foreach compound were determined at 10mg L�1 concentration for fivereplicate experiments. The limit of detection (LOD) was estimated asthree times the signal-to-noise ratio. These LOD data showed thatthe sensitivity of this method was good enough to ensure reliablemeasurements.

3.3 Comparison of DLLME with other reportedmethods

Since there is not any report about simultaneous microextractionof petroleum pollutants (n-alkanes and PAHs) in water samples,

Figure 2. The effect of different disperser solvents on the extractionefficiency of petroleum pollutants. Extraction conditions: volume ofthe water sample 8mL, concentration level at 50mg L�1, volume of theextraction solvent (n-hexane) 20mL, room temperature.

Figure 3. The effect of the volume of n-hexane on the extraction efficiencyof petroleum pollutants obtained from DLLME. Extraction conditions: watersample volume, 8mL; disperser solvent (acetone) volume, 1mL, analytesconcentration 50mg L�1, room temperature.

Figure 4. The effect of volume of acetone on extraction efficiency ofpetroleum pollutants obtained from DLLME. Extraction conditions: watersample volume 8mL, concentration level at 50mg L�1, extraction solvent(n-hexane) volume 20mL, room temperature.

4 K. S. Hasheminasab et al.

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9

the proposed method was compared with the other methodswhich were applied for the extraction of n-alkanes [8, 53, 54]and PAHs [55–58] separately (Tab. 2). The extraction time ofDLLME method is shorter than the other microextraction methods.In headspace solvent microextraction and hollow fiber liquidphase microextraction methods, the extraction time is longerthan DLLME according to the fact that less contact surfacearea increases the extraction time. The solid phase extraction,solid-phase microextraction, and headspace solid-phase micro-extraction methods are more expensive and time consuming.The main advantage of the DLLME method is its quite largesurface area between the fine droplets of the extraction solventand the water sample, and accordingly its fast extraction

kinetics results in the rapid achieving of a state of equilibriumand higher EFs.

3.4 Real sample analysis

The DLLME-GC-FID technique was successfully applied to thedetermination of petroleum pollutants in several water samples.Analytical results and the relative recoveries for the spiked realwater samples are shown in Tab. 3. Spike concentration foreach compound is lower than solubility of each compound inwater. Because of the rapid mass transfer with the DLLME,

the salts present in real water samples have no effect on the

Table 1. Some analytical performance date of DLLME method for petroleum pollutants

Compound LDRa) Smb) Sb

c) r2 EFd)LODe)

(mg L�1)

RSDf) % (n¼ 5)

Intraday Interday

C10 1–200 0.07 0.25 0.999 300 0.10 5.3 5.8C11 1–200 0.2 0.17 0.999 387 0.23 4.1 4.3C12 1–200 0.19 0.11 0.998 314 0.26 4.2 4.9C13 1–200 0.08 0.19 0.998 323 0.73 5.9 5.4C14 1–200 0.12 0.20 0.998 290 0.82 7.9 8.4C15 1–200 0.13 0.09 0.987 387 0.71 3.5 3.9C16 1–200 0.18 0.17 0.988 335 0.53 4.7 4.9C17 1–200 0.12 0.25 0.998 373 0.33 4.6 5.7C18 1–200 0.15 0.13 0.987 290 0.37 8.1 8.6C19 1–200 0.06 0.15 0.998 328 0.42 8.2 8.7C20 1–200 0.08 0.16 0.988 299 0.36 6.9 6.7C21 1–200 0.13 0.21 0.999 350 0.37 7.5 7.9C22 1–200 0.08 0.25 0.997 380 0.45 4.9 6.6C23 1–200 0.15 0.16 0.999 446 0.44 5.1 6.8C24 1–200 0.11 0.18 0.996 399 0.39 5.5 6.2C25 1–200 0.09 0.19 0.988 320 0.50 7.9 7.7C26 1–200 0.14 0.15 0.987 412 0.53 4.8 5.1C27 1–200 0.12 0.19 0.987 280 0.31 8.7 7.9C28 1–200 0.08 0.21 0.999 385 0.37 5.5 5.9C29 1–200 0.05 0.16 0.995 298 0.26 6.2 6.6C30 1–200 0.06 0.21 0.995 391 0.34 4.8 5.8C31 1–200 0.12 0.17 0.997 421 0.34 5.4 5.7C32 1–200 0.09 0.19 0.988 346 0.28 5.3 5.7C33 1–200 0.21 0.25 0.987 387 0.5 4.7 4.9C34 1–200 0.11 0.23 0.988 403 0.45 8.1 8.6C35 1–200 0.09 0.21 0.997 332 0.54 4.2 4.9Naphthalene 1–200 0.13 0.16 0.988 368 0.34 7.7 8.2Acenaphthylene 1–200 0.2 0.18 0.988 321 0.39 4.7 5.0Acenaphthene 1–200 0.13 0.23 0.986 398 0.34 6.4 6.5Fluorene 1–200 0.13 0.22 0.999 361 0.45 3.8 5.4Phenanthrene 1–200 0.08 0.18 0.988 400 0.36 6.5 7.2Anthracene 1–200 0.05 0.15 0.999 403 0.44 6.5 6.9Fluoranthene 1–200 0.16 0.25 0.987 443 0.56 6.4 7.3Pyrene 1–200 0.16 0.15 0.999 373 0.38 7.5 8.0Benzo[a]anthracene 1–200 0.11 0.29 0.999 396 0.42 4.9 5.7Chrysene 2–200 0.15 0.12 0.986 360 0.67 6.3 6.6Benzo[k]fluoranthene 2–200 0.16 0.25 0.988 387 0.72 5.9 6.5Benzo[b]fluoranthene 4–200 0.14 0.22 0.994 399 0.98 6.1 5.9Benzo[a]pyrene 2–200 0.09 0.15 0.986 356 0.16 6.4 7.1Indeno[1,2,3-cd]pyrene 4–200 0.10 0.20 0.988 385 0.18 6.5 7.8Dibenzo[a,h]anthracene 2–200 0.12 0.17 0.987 369 0.96 7.3 7.9Benzo[g,h,i]perylene 2–200 0.19 0.18 0.986 337 0.53 7.9 8.3

a) Linear dynamic range (mg L�1).b) Standard deviations of the slope.c) Standard deviations of the intercept.d) Enrichment factor.e) LOD (signal-to-noise¼ 3).f) Intraday and interday repeatability (RSDs%) for spiked concentration of 10mg L�1 for petroleum hydrocarbons.

Trace Determination of Petroleum Pollutants 5

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9

Table 2. Comparison of the proposed method with the other methods

Compound MethodLOD

(mg L�1)LDR

(mg L�1) RSDExtraction time

(min) Refs.

n-Alkanes HSME 0.1–4 0.5–400 to 5–200 �7.2 8 [8]SPME 0.1–0.3 0.5–30 �9.54 20 [53]HSPME 50–150 150–3000 to 450–4500 �8.6 20 [54]

PAHs SPME 0.001–0.029 0.01–10 45 [55]HSPME 0.03–0.3 0.1–50 �10.2 30 [56]SPE 0.026–0.82 0.2–100 to 1–100 �9.7 60 [57]LPME 0.35–0.60 1.2–12 �6.0 20 [58]

n-Alkanes and PAHs Proposed method 0.1–0.96 1–200 to 4–200 �8.8 <1 –

LDR, linear dynamic range; HSME, headspace solvent microextraction; SPME, solid phase microextraction; HSPME, headspace solid phasemicroextraction; SPE, solid phase extraction; LPME, liquid phase microextraction.

Table 3. Spiked recoveries of three real water samples with a spiked concentration of 20mg L�1 for petroleum hydrocarbons

Compound

Drinking water Jajrood river water Caspian sea water

Cinitial (mg L�1) RRb) (%) Cinitial (mg L

�1) RR (%) Cinitial (mg L�1) RR (%)

C10 NDa) 97� 2 ND 99� 3 ND 103� 5C11 ND 91� 7 ND 95� 4 ND 90� 2C12 ND 89� 6 ND 97� 5 ND 80� 6C13 ND 95� 3 ND 87� 8 ND 89� 5C14 ND 94� 5 ND 95� 4 ND 90� 3C15 ND 97� 4 ND 89� 7 26 87� 4C16 ND 102� 6 ND 90� 5 ND 98� 7C17 ND 80� 3 ND 93� 3 ND 102� 4C18 ND 89� 4 ND 89� 5 ND 89� 6C19 ND 100� 4 ND 98� 5 ND 83� 4C20 ND 73� 5 ND 94� 6 9.1 89� 6C21 ND 93� 5 ND 91� 8 ND 92� 2C22 ND 91� 4 ND 96� 5 ND 97� 5C23 ND 97� 5 ND 98� 7 8.3 99� 6C24 ND 87� 8 ND 81� 6 16 98� 3C25 ND 89� 4 ND 87� 5 23 83� 5C26 ND 93� 2 ND 89� 8 26 80� 6C27 ND 102� 5 ND 93� 3 28.0 101� 8C28 ND 98� 6 ND 80� 4 21.6 97� 2C29 ND 90� 4 ND 93� 4 24.0 91� 4C30 ND 80� 4 ND 89� 5 27.0 89� 4C31 ND 93� 4 ND 93� 6 19 87� 6C32 ND 99� 8 ND 89� 2 13.4 83� 3C33 ND 102� 4 ND 88� 3 ND 90� 7C34 ND 89� 3 ND 99� 5 ND 95� 5C35 ND 99� 3 ND 98� 2 ND 99� 2Naphthalene ND 103� 4 ND 96� 7 ND 96� 5Acenaphthylene ND 88� 4 ND 89� 5 ND 99� 3Acenaphthene ND 90� 3 ND 88� 3 ND 98� 4Fluorene ND 98� 8 ND 89� 6 ND 94� 8Phenanthrene ND 99� 5 12.1 83� 8 ND 89� 7Anthracene ND 102� 5 ND 85� 3 ND 88� 4Fluoranthene ND 97� 4 13.3 95� 7 ND 94� 4Pyrene ND 99� 5 8.2 84� 6 ND 87� 4Benzo[a]anthracene ND 99� 4 ND 85� 5 ND 89� 6Chrysene ND 88� 8 ND 90� 3 ND 90� 4Benzo[k]fluoranthene ND 87� 5 ND 86� 4 11.0 83� 5Benzo[b]fluoranthene ND 88� 5 ND 80� 7 ND 87� 8Benzo[a]pyrene ND 80� 4 ND 83� 6 11.2 80� 8Indeno[1,2,3-cd]pyrene ND 83� 4 ND 92� 5 12.5 87� 7Dibenzo[a,h]anthracene ND 93� 5 ND 87� 7 ND 84� 9Benzo[g,h,i]perylene ND 90� 6 ND 82� 6 ND 90� 6

a) Not detected.b) Relative recovery (%, n¼ 3).

6 K. S. Hasheminasab et al.

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9

extraction yield. Figure 5 depicted typical chromatograms ofreal water samples and the spiked sea water obtained aftermicroextraction.

4 Concluding remarksThe DLLME method using organic solvents lighter than water wasused for the simultaneous determination of 42 petroleum pollutantsin water samples by GC-FID. The main advantage of this techniqueis its quite large surface area between the fine droplets of theextraction solvent and water sample, and the accordingly fastextraction kinetics that results in the rapid achieving of a state ofequilibrium and the high EFs. This method was successfully appliedto the determination of petroleum pollutants in different watersamples.

Acknowledgments

The authors gratefully appreciated financial support from theResearch Affairs of Shahid Beheshti University.

The authors have declared no conflict of interest.

References[1] M. Farajzadeh, A. A. Matin, A New PVC-Activated Charcoal Fiber

Coated on Silver Wire; Application in Determination of n-Alkanesin the Headspace of Soil Samples by SPME-GC, Anal. Sci. 2002, 18,77–81.

[2] A. Pavlova, R. Ivanova, Determination of Petroleum Hydrocarbonsand Polycyclic Aromatic Hydrocarbons in Sludge from WastewaterTreatment Basins, J. Environ. Monit. 2003, 5, 319–323.

Figure 5. Chromatograms of (A) drinking water, (B) Jajrood river water, (C) Caspian sea water, and (D) spiked Caspian sea water at the concentrationlevel of 20mg L�1 of petroleum pollutants by DLLME-GC-FID.

Trace Determination of Petroleum Pollutants 7

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9

[3] E. S. Brodskii, I. M. Lukashenko, G. A. Kalinkevich, S. A. Savchuk,Identification of Petroleum Products in Environmental SamplesUsing Gas Chromatography and Gas Chromatography–MS, J. Anal.Chem. 2002, 57, 486–490.

[4] M. Martienssen, O. Reichel, M. Schirmer, Einsatz oberflächenaktiverSubstanzen zur Verbesserung der biologischen Abbaubarkeit vonMineralölkohlenwasserstoffen, Chem. Ing. Tech. 2003, 75, 1749–1755.

[5] M. Voyevoda, W. Geyer, S. Mothes, An Improved Gas Chromatograph-ic Method for the Determination of C6–C12 Petroleum Hydro-carbons in Soil Water, Clean – Soil Air Water 2008, 36, 164–170.

[6] S. Patin, Environmental Impact of the Offshore Oil and Gas Industry,Ecomonitor Publishing, New York 1999.

[7] M. Bishop, Petroleum Hydrocarbons and Petroleum HydrocarbonsDeterminations, New England Testing Laboratory, Providence, RI 1997.

[8] M. K. Zanjani, Y. Yamin, S. Shariati, Analysis of n-Alkanes in WaterSamples by Means of Headspace Solvent Microextraction and GasChromatography, J. Hazard. Mater. 2006, 136, 714–720.

[9] S. C. Vishnoi, S. D. Bhagat, V. B. Kapoor, S. K. Chopra, R. Krishna,Simple Gas Chromatographic Determination of the Distributionof Normal Alkanes in the Kerosene Fraction of Petroleum, Analyst1987, 112, 49–52.

[10] J. K. Volkman, A. T. Revill (Eds.), Oil Pollution and MicrobialRegulation, in Environmental and Ecological Chemistry, UNESCO-EOLSS,Paris 2006.

[11] National Park Service, Environmental Contaminants EncyclopediaAlkanes Entry (Ed: R. J. Irwin), National Park Service and ColoradoState University, USA 1997.

[12] R. M. Marce, F. Borrull, Solid-Phase Extraction of Polycyclic AromaticCompounds, J. Chromatogr. A 2000, 885, 273–290.

[13] Y. Liu, Y. Hashi, J. M. Lin, Continuous-Flow Microextraction and GasChromatographic–Mass Spectrometric Determination of PolycyclicAromatic Hydrocarbon Compounds in Water, Anal. Chim. Acta 2007,585, 294–299.

[14] T. Luan, S. Fang, Y. Zhong, L. Lin, S. M. N. Chan, C. Lan, N. F. Y. Tam,Determination of Hydroxyl-Metabolites of Polycyclic AromaticHydrocarbons by Fully Automated Solid-Phase MicroextractionDerivatization and Gas Chromatography–Mass Spectrometry, J.Chromatogr. A 2007, 1173, 37–43.

[15] L. Guo, H. Kee Lee, Development of Multiwalled Carbon NanotubesBased Micro-Solid-Phase Extraction for the Determination of TraceLevels of Sixteen Polycyclic Aromatic Hydrocarbons in Environmen-tal Water Samples, J. Chromatogr. A 2011, 1218, 9321–9327.

[16] US EPA, Methods for Chemical Analysis of Water and Wastes, US EPA,Washington, DC 1986.

[17] S. Ozcan, A. Tor, M. E. Aydin, Determination of Polycyclic AromaticHydrocarbons in Waters by Ultrasound-Assisted Emulsification-Microextraction and Gas Chromatography–Mass Spectrometry, Anal.Chim. Acta 2010, 665, 193–199.

[18] L. Guo, H. K. Lee, Low-Density Solvent-Based Solvent DemulsificationDispersive Liquid–Liquid Microextraction for the Fast Determina-tion of Trace Levels of Sixteen Priority Polycyclic AromaticHydrocarbons in Environmental Water Samples, J. Chromatogr. A2011, 1218, 5040–5046.

[19] V. Ghasemzadeh-Mohammadi, A. Mohammadi, M. Hashemi,R. Khaksar, P. Haratian, Microwave-Assisted Extraction andDispersive Liquid–Liquid Microextraction Followed by Gas Chroma-tography–Mass Spectrometry for Isolation and Determination ofPolycyclic Aromatic Hydrocarbons in Smoked Fish, J. Chromatogr. A2012, 1237, 30–36.

[20] Annual Book of Standards,Water (II), Vol. 11.02, American Society forTesting and Materials, Conshohocken, PA 1995.

[21] H. Xu, Z. Ding, L. Lv, D. Song, Y.-Q. Feng, A Novel DispersiveLiquid–Liquid Microextraction Based on Solidification of FloatingOrganic Droplet Method for Determination of Polycyclic Aromat-ic Hydrocarbons in Aqueous Samples, Anal. Chim. Acta 2009, 636,28–33.

[22] D. Stainken, M. Miller, Symposium on Waste Testing and QualityAssurance Proceeding, EPA, Washington, DC 1988.

[23] C. Fan, S. Krishnamurthy, C. T. Chen, T. A. O’Shay, K. B. Hoddino,Analysis of Soil Contaminated with Petroleum Constituents, ASTM STP 1221,American Society for Testing andMaterials, Conshohocken, PA 1994.

[24] J. R. Marquart, G. B. Dellow, E. R. Freitas, Determination of NormalParaffins in Petroleum Heavy Distillates by Urea Adduction and GasChromatography, Anal. Chem. 1968, 40, 1633–1637.

[25] M. A. Farajzadeh, A. A. Matin, Semi-Micro Solvent Extraction as aRapid and Efficient Preconcentration Technique for the Determina-tion of n-Alkanes in Oil-Contaminated Water and Soil Samples byCapillary Gas Chromatography, Chromatographia 2002, 55, 225–229.

[26] International Organization for Standardization, ISO 17993, WaterQuality – Determination of 15 Polycyclic Aromatic Hydrocarbons (PAH) byHPLC with Fluorescence Detection after Liquid–Liquid Extraction in Water,Beuth Verlag, Berlin 2002.

[27] US Environmental Protection Agency, Compendium Method TO-13A:Determination of Polycyclic Aromatic Hydrocarbons (PAHs) in Ambient AirUsing Gas Chromatography/Mass Spectrometry (GC–MS), Compendium ofMethods for the Determination of Toxic Organic Compounds in Ambient Air,2nd Ed., US EPA, Washington, DC 1999.

[28] P. Liang, J. Xu, L. Guo, F. Song, Dynamic Liquid-Phase Micro-extraction with HPLC for the Determination of Phoxim in WaterSamples, J. Sep. Sci. 2006, 29, 366–370.

[29] A. R. Fontana, M. M. de Toro, J. C. Altamirano, One-StepDerivatization and Preconcentration Microextraction Techniquefor Determination of Bisphenol A in Beverage Samples by GasChromatography–Mass Spectrometry, J. Agric. Food Chem. 2011, 59,3559–3565.

[30] A. Sarafraz-Yazdi, A. Amiri, Liquid-Phase Microextraction, TrendsAnal. Chem. 2010, 29, 1–14.

[31] A. R. Fontana, J. C. Altamirano, Sensitive Determination of 2,4,6-Trichloroanisole in Water Samples by Ultrasound Assisted Emulsifi-cation Microextraction prior to Gas Chromatography–Tandem MassSpectrometry Analysis, Talanta 2010, 81, 1536–1541.

[32] M. Rezaee, Y. Assadi, M. R. Milani Hosseini, E. Aghaee, F. Ahmadi,S. Berijani, Determination of Organic Compounds in Water UsingDispersive Liquid–Liquid Microextraction, J. Chromatogr. A 2006,1116, 1–9.

[33] S. Berijani, Y. Assadi, M. Anbia, M. R. Milani Hosseini, E. Aghaee,Dispersive Liquid–Liquid Microextraction Combined with GasChormatography-Flame Photometric Detection Very Simple, Rapidand Sensitive Method for the Determination of OrganophosphorusPesticides in Water, J. Chromatogr. A 2006, 1123, 1–9.

[34] T. Rodríguez-Cabo, I. Rodríguez, M. Ramil, R. Cela, Dispersive Liquid–Liquid Microextraction Using Non-Chlorinated, Lighter than WaterSolvents for Gas Chromatography–Mass Spectrometry Determina-tion of Fungicides in Wine, J. Chromatogr. A 2011, 1218, 6603–6611.

[35] A. N. Panagiotou, V. A. Sakkas, T. A. Albanis, Application ofChemometric Assisted Dispersive Liquid–Liquid Microextraction tothe Determination of Personal Care Products in Natural Waters,Anal. Chim. Acta 2009, 649, 135–140.

[36] A. Mousa, C. Basheer, A. R. Al-Arfaj, Determination of PhthalateEsters in Bottled Water Using Dispersive Liquid–Liquid Micro-extraction Coupled with GC–MS, J. Sep. Sci. 2013, 36, 2003–2009.

[37] C. Stalikas, Y. Fiamegos, V. Sakkas, T. Albanis, Developments onChemometric Approaches to Optimize and Evaluate Microextrac-tion, J. Chromatogr. A 2009, 1216, 175–189.

[38] M. T. Pena, M. C. Casais, M. C. Mejuto, R. Cela, Development of anIonic Liquid Based Dispersive Liquid–Liquid MicroextractionMethod for the Analysis of Polycyclic Aromatic Hydrocarbons inWater Samples, J. Chromatogr. A 2009, 1216, 6356–6364.

[39] X. Wang, L. Lin, T. Luan, L. Yang, N. F. Y. Tam, Determination ofHydroxylated Metabolites of Polycyclic Aromatic Hydrocarbons inSediment Samples by Combining Subcritical Water Extraction andDispersive Liquid–Liquid Microextraction with Derivatization, Anal.Chim. Acta 2012, 753, 57–63.

[40] V. Vickackaite, E. Pusvaskiene, Dispersion-Solidification Liquid–Liquid Microextraction for Volatile Aromatic Hydrocarbons Deter-mination: Comparison with Liquid Phase Microextraction Based onthe Solidification of a Floating Drop, J. Sep. Sci. 2009, 32, 3512–3520.

8 K. S. Hasheminasab et al.

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9

[41] J. López-Darias, M. Germán-Hernández, V. Pina, A. M. Afonso,Dispersive Liquid–Liquid Microextraction versus Single-Drop Micro-extraction for the Determination of Several Endocrine-DisruptingPhenols from Seawaters, Talanta 2010, 80, 1611–1618.

[42] Q. H. Wu, C. Wang, Z. M. Liu, C. X. Wu, X. Zeng, J. L. Wen, Z. Wang,Dispersive Solid-Phase Extraction Followed by Dispersive Liquid–Liquid Microextraction for the Determination of Some SulfonylureaHerbicides in Soil by High-Performance Liquid Chromatography, J.Chromatogr. A 2009, 1216, 5504–5510.

[43] E. Zhao, W. Zhao, L. Han, S. Jiang, Z. Zhou, Application of DispersiveLiquid–Liquid Microextraction for the Analysis of Organophospho-rus Pesticides in Watermelon and Cucumber, J. Chromatogr. A 2007,1175, 137–140.

[44] J. Hu, L. Fu, X. Zhao, X. Liu, H. Wang, X. Wang, L. Dai, DispersiveLiquid–Liquid Microextraction Combined with Gas Chromato-graphy-Electron Capture Detection for the Determination ofPolychlorinated Biphenyls in Soils, Anal. Chim. Acta 2009, 640, 100–105.

[45] Q. H. Wu, Y. P. Li, C. Wang, Z. M. Liu, X. H. Zang, X. Zhou, Z. Wang,Dispersive Liquid–Liquid Microextraction Combined with HighPerformance Liquid Chromatography–Fluorescence Detection forthe Determination of Carbendazim and Thiabendazole in Environ-mental Samples, Anal. Chim. Acta 2009, 638, 139–145.

[46] P. Liang, E. Zhao, F. Li, Dispersive Liquid–Liquid MicroextractionPreconcentration of Palladium inWater Samples and Determinationby Graphite Furnace Atomic Absorption Spectrometry, Talanta 2009,77, 1854–1857.

[47] Environmental Research Institute, Analysis of Extractable TotalPetroleum Hydrocarbons (ETPH) Using Methylene Chloride Gas Chromato-graph/Flame Ionization Detection, Environmental Research Institute,University of Connecticut, Storrs, CT 1999.

[48] A. Saleh, Y. Yamini, M. Faraji, M. Rezaee, M. Ghambarian, Ultrasound-Assisted Emulsification Microextraction Method Based on ApplyingLow Density Organic Solvents Followed by Gas ChromatographyAnalysis for the Determination of Polycyclic Aromatic Hydrocarbonsin Water Samples, J. Chromatogr. A 2009, 1216, 6673–6679.

[49] N. Mashkouri Najafi, H. Tavakoli, Y. Abdollahzadeh, R. Alizadeh,Comparison of Ultrasound-Assisted Emulsification and Dispersive

Liquid–Liquid Microextraction Methods for the Speciation ofInorganic Selenium in Environmental Water Samples Using LowDensity Extraction Solvents, Anal. Chim. Acta 2012, 714, 82–88.

[50] M. A. Farajzadeh, D. Djozan, R. Fazeli Bakhtiyari, Use of a CapillaryTube for Collecting an Extraction Solvent Lighter than Waterafter Dispersive Liquid–Liquid Microextraction and Its Applicationin the Determination of Parabens in Different Samples by GasChromatography-Flame Ionization Detection, Talanta 2010, 81,1360–1367.

[51] A. Zgoła-Grze�skowiak, T. Grze�skowiak, Dispersive Liquid–LiquidMicroextraction, Trends Anal. Chem. 2011, 30, 1382–1399.

[52] N. Fatahi, Y. Assadi, M. R. Milani Hosseini, E. Zeini Jahromi,Determination of Chlorophenols in Water Samples Using Simulta-neous Dispersive Liquid–Liquid Microextraction and DerivatizationFollowed by Gas Chromatography-Electron-Capture Detection, J.Chromatogr. A 2007, 1157, 23–29.

[53] M. Farajzadeh, M. Hatami, Analysis of n-Alkanes at Sub Microgramper Liter Level after Direct Solid Phase Microextraction fromAqueous Samples, Anal. Sci. 2002, 18, 1221–1225.

[54] M. Farajzadeh, M. Hatami, A New Selective SPME Fiber for Somen-Alkanes and Its Use for Headspace Sampling of Aqueous Samples,J. Sep. Sci. 2003, 26, 802–808.

[55] A. J. King, J. W. Readman, J. L. Zhou, Determination of PolycyclicAromatic Hydrocarbons in Water by Solid Phase Micro-Extraction-Gas Chromatograhpy–Mass Spectrometry, Anal. Chim. Acta 2004, 523,259–267.

[56] Dj. Djozan, Y. Assadi, Monitoring of Polycyclic Aromatic Hydro-carbons in Water Using Headspace Solid-Phase Microextractionand Capillary Gas Chromatography, Microchem. J. 1999, 63, 276–284.

[57] Y. Huang, Q. Zhou, G. Xie, Development of Micro-Solid PhaseExtraction with Titanate Nanotube Array Modified by Cetyltri-methylammonium Bromide for Sensitive Determination of Poly-cyclic Aromatic Hydrocarbons from Environmental Water Samples,J. Hazard. Mater. 2011, 193, 82–89.

[58] L. Hou, H. K. Lee, Application of Static and Dynamic Liquid-PhaseMicroextraction in the Determination of Polycyclic AromaticHydrocarbons, J. Chromatogr. A 2002, 976, 377–385.

Trace Determination of Petroleum Pollutants 9

© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com Clean – Soil, Air, Water 2013, 41 (9999), 1–9