Gpc

ZCB

a

ARRAA

KGSCUE

1

rwdtfbsscoihs

h0

Journal of Chromatography A, 1355 (2014) 219–227

Contents lists available at ScienceDirect

Journal of Chromatography A

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

raphene based solid phase extraction combined with ultra higherformance liquid chromatography–tandem mass spectrometry forarbamate pesticides analysis in environmental water samples

hihong Shi, Junda Hu, Qi Li, Shulan Zhang, Yuhuan Liang, Hongyi Zhang ∗

ollege of Chemistry and Environmental Science, Hebei University, Key Laboratory of Analytical Science and Technology of Hebei Province,aoding 071002, China

r t i c l e i n f o

rticle history:eceived 22 November 2013eceived in revised form 13 April 2014ccepted 29 May 2014vailable online 12 June 2014

eywords:raphenePEarbamate pesticidesPLC–MS/MS

a b s t r a c t

In this paper, graphene, a new sorbent material, was synthesized and used for solid-phase extraction (SPE)of the six carbamate pesticides (pirimicarb, baygon, carbaryl, isoprocarb, baycarb and diethofencarb) inenvironmental water samples. The target analytes can be extracted on the graphene-packed SPE cartridge,and then eluted with acetone. The eluate was collected and dried by high purity nitrogen gas at roomtemperature. 1 mL of 20% (v/v) acetonitrile aqueous solution was used to redissolve the residue. The finalsample solution was analyzed by ultra performance liquid chromatography-tandem quadrupole massspectrometry (UPLC–MS/MS) system. Under optimum conditions, good linearity was obtained for thecarbamates with correlation coefficient in the range of 0.9992–0.9998. The limits of detection (S/N = 3) forthe six carbamate pesticides were in the range of 0.5–6.9 ng L−1. Relative standard deviations (RSD) for fivereplicate determinations were below 5.54%. RSD values for cartridge-to-cartridge precision (n = 7) were in

nvironmental water the range of 1.27–8.13%. After proper regeneration, the graphene-packed SPE cartridge could be re-usedover 100 times for standard solution without significant loss of performance. The enrichment factorsfor the target analytes were in the range of 34.2–51.7. The established method has been successfullyapplied to the determination of carbamate pesticide residues in environmental water samples such asriver water, well water and lake water.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

As less-toxic alternatives to organophosphorus and organochlo-ine classes, carbamates, composed of the ester of carbamic acidith various substituents, are widely used in agricultural pro-uction as pesticides [1,2]. Although carbamates can disintegrateo some extent, they have been found frequently remaining inruit, vegetables and crops as a result of excessive use [3]. Car-amate pesticides may also enter into the environmental waterystems through various paths, including spraying, soil seepage,torage and the discharge of waste water, leading to possibleontamination of the environmental water [4,5]. As inhibitorsf acetylcholinesterase, carbamate pesticides could affect nerve

mpulse transmission, inducing dramatic toxicological effects inuman beings. Moreover, carbamates and their metabolites areuspected to be carcinogens and mutagens [6]. So carbamate

∗ Corresponding author. Tel.: +86 312 5079357.E-mail address: hyzhang@hbu.edu.cn (H. Zhang).

ttp://dx.doi.org/10.1016/j.chroma.2014.05.085021-9673/© 2014 Elsevier B.V. All rights reserved.

pesticides are included on the priority list issued by the UnitedStates Environmental Protection Agency (EPA) [7].

Therefore, the monitoring of the carbamate residue levels in var-ious environmental water systems is of special concern to humanhealth and environmental safety. The European Union Directiveon drinking water quality (98/83/EC) established a maximumallowed concentration of 0.1 �g L−1 for each individual pesticideand 0.5 �g L−1 for total pesticides [8]. In this sense, reliable, sen-sitive and rapid analytical methods are urgently needed for thedetermination of carbamate pesticides at trace levels.

Different techniques have been employed for the determina-tion of carbamate pesticides in water and the most commonlypreferred methods are liquid chromatography (LC) [9–11] and gaschromatography (GC) [12] coupled to a large number of detec-tors. As the thermolability of carbamate pesticides may lead todifficulty in direct GC analysis, carbamate pesticides are usually

derivatized on-line [13] or off-line [14] for GC analysis. In this case,some authors do not recommend the use of GC for the analysisof carbamate pesticides and consider LC to be the most conve-nient technique [15]. As the carbamate pesticides are usually found

2 ogr. A

acVeemtattpoe[as

aahcGccs[mpdatt[cgai

tsTcst�tpgaqtcgspbs

2

2

ic

20 Z. Shi et al. / J. Chromat

t trace concentration levels, it is necessary to carry out pre-oncentration and/or cleanup steps prior to instrumental analysis.arious sample preparation techniques have been employed for thextraction of carbamates from water sample, such as liquid–liquidxtraction (LLE) [16], solid phase extraction (SPE) [17], solid-phaseicroextraction (SPME) [18], hollow fiber liquid-phase microex-

raction (HF-LPME) [10], single-drop microextraction (SDME) [19]nd dispersive liquid–liquid microextraction (DLLME) [20]. SPE as aechnique well known for its large enrichment capacity is presentlyhe most extended method for the preconcentration of carbamateesticide residues from water samples. In SPE procedure, the choicef appropriate adsorbent is a critical factor to obtain good recov-ry and high enrichment factor [21,22]. Octadecyl-bonded silica23], polymeric sorbent [24], and mixed-mode sorbent [25] havelready been used for the SPE of carbamate pesticides from wateramples.

In recent years, various carbon-based materials have beendopted as SPE adsorbents because of their specific propertiesnd high stability. Graphene, a new class of carbon nanomaterial,as sparked much interest because of its remarkable mechani-al, thermal and electronic properties since it was discovered byeim in 2004 [26,27]. With a two dimensional honeycomb latticeomposed of carbon atoms, graphene possesses an ultra-high spe-ific surface area (theoretical value 2630 m2 g−1) [28], and bothides of the planar sheets are available for molecule adsorption29,30]. Besides, as graphene is an electron-rich, hydrophobic nano-

aterial with large specific area and �–� electrostatic stackingroperty [31,32], it has been served as an extraordinarily won-erful adsorbent or extraction material [33]. Nowadays, graphenend graphene-based materials have been used as adsorbents forhe extraction and preconcentration of chlorophenols [34], glu-athione [35], sulfonamide antibiotics [36], phthalate acid esters37], organophosphate pesticides [38], heavy metals [39], poly-yclic aromatic hydrocarbons [40], macrolides [41] and malachitereen [42]. Amine modified graphene has been used to remove fattycids and other interfering substances for the analysis of pesticidesn oil crops [43].

In this paper, the performance of graphene-packed SPE car-ridge for the extraction of carbamate pesticides from wateramples prior to UPLC–MS/MS analysis was first demonstrated.he six targeted carbamate pesticides are pirimicarb, baygon,arbaryl, isoprocarb, baycarb and diethofencarb, their chemicaltructures are shown in Fig. 1. These compounds have benzyl, naph-hyl or pyrimidyl in their structures, so they can exhibit strong-stacking interaction with the large delocalized �-electron sys-

em of graphene, to be selectively adsorbed on graphene. Thearameters influencing the extraction efficiency were investi-ated, including the type and volume of eluent solvent, the pHnd volume of sample solution. For the efficient separation anduantification of the six carbamates, the UPLC and MS/MS condi-ions were optimized. The established method was validated andartridge-to-cartridge precision was evaluated. The performance ofraphene-packed SPE cartridge was compared with conventionalorbents such as C18 and graphitized carbon black. Finally, theroposed method was applied to the determination of the six car-amate pesticide residues in river water, lake water and well wateramples.

. Experimental

.1. Chemicals and materials

Pirimicarb (99.2%), diethofencarb (99.5%), baygon (99.5%),soprocarb (99.2%), carbaryl (99.5%) and baycarb (99.5%) were pur-hased from Dikma technologies Co., Ltd. (Beijing, China). The

1355 (2014) 219–227

standard stock solutions of carbamates were prepared in darkbrown flask with methanol as solvent and stored in the dark at−18 ◦C. The standard working solution was freshly prepared bydiluting the stock solution with water. Graphite powder (99%)and hydrazine hydrate (50%) were purchased from J&K technol-ogy Co., Ltd. (Beijing, China). KMnO4, P2O5, K2S2O8, H2O2 (30%)and concentrated H2SO4 (95–98%) were of analytical grade andwere purchased from Huaxin Chemicals Co., Ltd. (Baoding, China).Acetontrile, formic acid and methanol were of HPLC grade andwere purchased from MREDA technologies Co., Ltd. (Beijing, China).Experimental water was doubly distilled de-ioned water.

The empty SPE cartridges (3 mL) and SPE frits were purchasedfrom Dikma technologies Co., Ltd. (Beijing, China). AGT CleanertODS C18 cartridges were purchased from Agela Techonologies INC.(Delaware, USA). VARIAN Bond Elut PRS cartridges were purchasedfrom Varian Co. (USA). Envi-carb graphitized carbon black car-tridges were purchased from Supelco Co. (USA).

2.2. Instrumentation

Chromatographic separation was performed on an ACQUITYTM

Ultra Performance Liquid Chromatography system (Waters, Mil-ford, MA, USA), consisting of a binary solvent delivery system andan autosampler. MS/MS detection was performed on a Xevo® TQtandem quadrupole mass spectrometer (Waters, USA) equippedwith an electrospray ionization (ESI) source. Data were acquiredand processed with MassLynx V4.1 software.

JEM-100SX Transmission Electron Microscope (TEM) (Jeol Ltd,Japan), JEM-7500F Scanning Electron Microscope (SEM) (Jeol Ltd,Japan) and TU-1901 UV–vis spectrometer (Persee, China) were usedto characterize the lab-produced graphene. The SPE experimentswere performed on an HSE series solid-phase extraction devicewith a vacuum pump (Tianjin HengAo technology development Co.,Ltd., Tianjin, China). MTN-2800D pressure blowing concentratorwas purchased from Auto Science Co., Ltd. (Tianjin, China).

2.3. Chromatographic conditions

The chromatographic separation was performed on an ACQUITYUPLC® BEH C18 column (2.1 × 100 mm i.d., 1.7 �m, Waters, madein Ireland) preceded by a BEH C18 VanGuardTM pre-column(2.1 × 5 mm i.d., 1.7 �m, Waters, made in Ireland). The mobile phaseconsisted of (A) 0.1% formic acid solution and (B) acetonitrile. Theeluting conditions were as follows: 0–4 min, linear gradient from30% to 40% B; 4–6 min, linear gradient from 40% to 45% B; 6–6.5 min,linear gradient from 45% to 90% B; 6.5–6.6 min, the compositionof B dropped from 90% to 30%; 6.6–8.0 min, the composition of Bwas kept at 30%. The flow rate was 0.4 mL min−1. The strong washvolume was 200 �L (90% acetonitrile, 0.1‰ formic acid) and theweak wash volume was 600 �L (10% acetonitrile, 0.1‰ formic acid).The column temperature and autosampler temperature were main-tained at 40 ◦C and 15 ◦C, respectively. The injection volume was10 �L.

2.4. Mass spectrometric conditions

Mass spectrometry was performed on a Waters Xevo® TQ tan-dem quadrupole mass spectrometer equipped with electrosprayionization (ESI) source. The conditions of ESI source were as fol-lows: Source temperature 150 ◦C; Desolvation gas temperature,550 ◦C; Desolvation gas (N2) flow rate, 850 L h−1; Cone gas (N2) flow

rate, 50 L h−1; Capillary voltage, 4.00 kV; Collision gas (Ar) flow rate,0.15 mL min−1. All the six compounds were analyzed in positive ESImode and multiple-reaction monitoring (MRM) mode was selectedfor quantification.

Z. Shi et al. / J. Chromatogr. A 1355 (2014) 219–227 221

O

O

NH

O

Baygon

O NH

O

Carbaryl

N

N

N

CH3

CH3

O

CH3

NCH3

CH3

O

CH3

Pirimicarb

Diethofencarb

OCONHCH3

CH(CH3)2

Isoprocarb

OHN

O

Baycarb

NH

O

O

O

O

of the

2

m5adw0tTa1scatfaaW11wbcw

gaodba

2

utm

Fig. 1. Chemical structures

.5. Synthesis and characterization of graphene

Graphite oxide was synthesized by a modified Hummersethod [44,45]. Concentrated H2SO4 (12 mL) was added into a

0 mL beaker and heated in an 80 ◦C water bath. Then 2.5 g K2S2O8nd 2.5 g P2O5 were added into the beaker. Next, 3 g Graphite pow-er was accurately weighed and added into the beaker. The mixtureas stirred and kept at 80 ◦C for 4.5 h. Then, it was diluted with

.5 L of water and left overnight. After that, the mixture was fil-ered through a 0.22 �m membrane and washed with 1 L of water.he product was dried at 60 ◦C. This pre-oxidized graphite wasdded into 120 mL concentrated H2SO4 in an ice bath. After that,5 g KMnO4 was slowly (0.5 g min−1) added to the mixture undertirring. The temperature must be kept below 20 ◦C during this pro-ess. Then the ice bath was removed and the mixture was stirredt 35 ◦C for 2 h. After that, 250 mL of water was added to the mix-ure in an ice bath to keep the temperature below 50 ◦C and stirredor another 2 h. Then the mixture was diluted with 0.7 L of watergain. After the addition of water, 20 mL of H2O2 (30%, v/v) wasdded, causing the color turning into yellow along with bubbling.hen no bubble was produced, the mixture was centrifuged for

0 min at 5000 rpm, and washed with 1 L of HCl (1:10, v/v) and L of water until the pH was 7.0. The product was dark yellowhen it was washed with hydrochloric acid and it turned into dark

rown when it was washed with water. In the mean time, the vis-osity increased with the increase of the number of washing. Afterashing, the product was dried at 50 ◦C.

Hydrazine was used to reduce graphene oxide to synthesizeraphene. Graphite oxide (0.5 g) was dispersed in 500 mL of water,nd ultrasonicated for 1 h to exfoliate graphite oxide to graphenexide. Then 12 mL of hydrazine hydrate (50%) was added to theispersion. The mixture was refluxed and stirred for 24 h in an oilath at 95 ◦C. The final product was filtered, washed with water,nd dried at 50 ◦C.

.6. Solid-phase extraction procedures

Graphene (30 mg) was placed in a 3 mL empty SPE cartridgesing an upper frit and a lower frit to avoid adsorbent loss. Prioro extraction, the SPE cartridge was preconditioned with 3 mL

ethanol, 3 mL acetone, 3 mL acetonitrile and 9 mL doubly distilled

six carbamate pesticides.

water. The sample solution (50 mL) was passed through the car-tridge at a flow rate of 1 mL min−1. Then 5 mL acetone was usedto elute the analytes retained on the cartridge. The eluent wascollected and dried at room temperature under the nitrogen pro-tection. 1 mL of 20% (v/v) acetonitrile aqueous solution was usedto redissolve the residue. The final solution was filtered through a0.22 �m membrane, and 10 �L of the solution was injected into theUPLC–MS/MS system for analysis.

3. Results and discussion

3.1. Characterization of graphene

Characterization of the lab-produced graphene was carried outby using TEM, SEM and UV–vis scanning. From the TEM image(Fig. 2A), we can see clearly the transparent laminate structure withintrinsic wrinkles, which is a characteristic feature of the single-layer graphene sheet. The SEM image (Fig. 2B) indicates that thegraphene agglomerates consist of randomly aggregated and crum-pled nanosheets. In addition to this, UV–vis spectra were used toconfirm the graphene oxide and graphene. As shown in UV–visspectra (Fig. 2C), graphene oxide dispersion has an absorption peakat 230 nm, and graphene dispersion has a wide absorption peak at260 nm. This phenomenon is in accordance with the description inthe literature [34].

3.2. Optimization of UPLC–MS/MS conditions

In order to realize good separation of the six carbamates,the composition of mobile phase and elution program wereoptimized. Experimental results showed that complete separa-tion of the six carbamates could be achieved within 5.5 minby using the chromatographic conditions described in Section2.3. For the recondition of the column, the total run time was8 min.

The mass spectrometric detection was performed by using pos-itive ion electrospray MS/MS in MRM mode. The qualitative ion

pairs were selected based on their full scan mass spectra obtainedby directly infusing standard solutions into the mass spectrometerESI source. Two transitions between the parent ion and the mostabundant daughter ions were monitored for the identification of

222 Z. Shi et al. / J. Chromatogr. A 1355 (2014) 219–227

e. (C)

ewpp

3

te

TO

Fig. 2. (A) TEM image of graphene. (B) SEM image of graphen

ach compound, and the ion pair with relatively higher intensityas selected for quantification. For each analyte, the optimizedarameters including cone voltage, collision voltage, qualitative ionairs and quantitative ion pair are listed in Table 1.

.3. Optimization of SPE procedures

Several factors affecting SPE performance were investigated inhis paper. These parameters included the type and volume of elu-nt solvent, the pH of sample solution, as well as the volume of the

able 1ptimized multiple reaction monitoring (MRM) parameters for the detection of the carba

Compound Cone voltage (V) Collision voltage (V)

Isoprocarb22 14

22 8

Carbaryl20 25

20 12

Baycarb20 15

20 8

Baygon15 15

15 8

Pirimicarb28 20

28 16

Diethofencarb16 30

16 10

UV–vis spectra of graphene and graphene oxide dispersion.

sample. Extraction recovery (R) was used to evaluate the extractionefficiency and R is expressed as follows:

R = CV

CoVo× 100%

where C is the analyte concentration (ng mL−1) in the reconstitutedsolvent, Co is the initial concentration of analyte in water sample. V

and Vo are the volumes of the reconstituted solvent and water sam-ple, respectively. Three replicates were performed for these studies.Statistical analysis of the data (significance level) was carried outby using statistical software IBM SPSS statistics 19.

mate pesticides.

Qualitative ion pair (m/z) Quantitative ion pair (m/z)

194.20 → 95.00194.20 → 95.00194.20 → 137.10

202.05 → 127.05202.05 → 145.00202.05 → 145.00

208.09 → 95.00208.09 → 95.00208.09 → 152.00

210.15 → 111.00210.15 → 111.00210.15 → 168.10

239.17 → 72.00239.17 → 72.00239.17 → 182.10

268.18 → 124.03268.18 → 226.15268.18 → 226.15

Z. Shi et al. / J. Chromatogr. A 1355 (2014) 219–227 223

Fig. 3. Effect of the type of eluent on the extraction recoveries of six carbamatepbo

3

towdacOStmtsbwisof

3

et1aitdFwifoiado

3

e

Fig. 4. Effect of volume of eluent on the extraction recoveries of six carbamate

the extraction recoveries of baygon, carbaryl, isoprocarb and bay-carb decreased sharply (p < 0.001 for baygon, carbaryl and baycarb;p < 0.01 for isoprocarb), and the extraction recoveries for pirimicarb

esticides. Other conditions: sample, 1 mL of sample solution containing 25 ng ofaygon, diethofencarb and isoprocarb, 50 ng of carbaryl, 30 ng of baycarb, and 5 ngf pirimicarb; volume of eluent, 5 mL.

.3.1. Effect of the type of eluent solventSelection of the type of eluent solvent is of vital importance for

he extraction efficiency of the analytes. In this work, four kindsf eluent solvents (acetone, acetonitrile, methanol and ethanol)ere tested. 1 mL of sample solution containing 25 ng of baygon,iethofencarb and isoprocarb, 50 ng of carbaryl, 30 ng of baycarb,nd 5 ng of pirimicarb was loaded onto the graphene-packed SPEartridge, and eluted with 5 mL of the potential eluent solvent.ther experimental conditions were carried out as described inection 2.6 and the results are shown in Fig. 3. Among the poten-ial eluent solvents, methanol showed relatively low recovery for

ost of the carbamates, while acetone exhibited the highest extrac-ion recoveries for carbaryl, baycarb and diethofencarb, showingignificant difference from the other three solvents (p < 0.01). Foraygon and isoprocarb, acetone demonstrated similar recoveryith acetonitrile and ethanol (p > 0.05). For the elution of pirim-

carb, acetone behaved similar with acetonitrile (p > 0.05). In thistudy, acetone was proved to be more effective compared withther tested solvents, and it was chosen as eluent solvent in theollowing experiments.

.3.2. Effect of the volume of eluentThe volume of eluent is an important parameter for the efficient

lution of the analytes. Therefore, the effect of the volume of ace-one on the extraction efficiency of the analytes was investigated.

mL of sample solution containing 25 ng of baygon, diethofencarbnd isoprocarb, 50 ng of carbaryl, 30 ng of baycarb, and 5 ng of pir-micarb was loaded onto the graphene-packed SPE cartridge, andhen eluted with 1.0–7.0 mL of acetone. Other experimental con-itions were carried out as described in Section 2.6. As shown inig. 4, the extraction recoveries for all the six carbamates increasedith the increase of the volume of acetone, and reached the max-

mum value with 5 mL of acetone as eluent solvent. Then with theurther increase of the volume of acetone, different behaviors werebserved for the carbamates: the extraction recoveries for pirim-carb, carbaryl and diethofencarb kept almost constant (p > 0.05),nd the recoveries of baygon, baycarb and isoprocarb showed aecreasing trend (p < 0.01). Based on an overall consideration, 5 mLf acetone was employed in the following experiments.

.3.3. Effect of pH of the sampleThe pH of the sample solution is another important factor influ-

ncing the extraction efficiency for the reason that pH not only

pesticides. Other conditions: sample, 1 mL of sample solution containing 25 ng ofbaygon, diethofencarb and isoprocarb, 50 ng of carbaryl, 30 ng of baycarb, and 5 ngof pirimicarb; eluent solvent, acetone.

affects the existing state of targeted analytes but also influencesthe charge species and density on the sorbent surface [46]. Inthis work, the effect of pH on the extraction efficiency was evalu-ated at pH 3.00, 5.00, 6.80, 8.20, 10.00 and 12.00, and phosphatebuffer solutions were used to adjust the pH. It should be men-tioned that pH 8.20 was the original pH of the solution which wasprepared by diluting the mixed standard solution with water. Asshown in Fig. 5, when the pH was increased from 3 to 6.80, theextraction recoveries increased significantly for pirimicarb, baycarb(p < 0.001), diethofencarb (p < 0.01) and carbaryl (p < 0.05), whilefor baygon and isoprocarb, no significant increase was observed(p > 0.05). From pH 6.8 to 8.2, the extraction recoveries for all the sixcarbamates didn’t change much (p > 0.05). Then from pH 8.2 to 10.0,

Fig. 5. Effect of sample pH on the extraction recoveries of six carbamate pesti-cides. Other conditions: sample, 1 mL of sample solution containing 25 ng of baygon,diethofencarb and isoprocarb, 50 ng of carbaryl, 30 ng of baycarb, and 5 ng of pirim-icarb; eluent, acetone; eluent volume, 5 mL.

224 Z. Shi et al. / J. Chromatogr. A 1355 (2014) 219–227

Table 2Linearity, limit of detection and limit of quantification for the six carbamate pesticides.

Compound Linear range (�g L−1) Regression equation R LOD (ng L−1) LOQ (ng L−1)

Pirimicarb 0.005–5 Y = 22,794.25 + 604,919.40X 0.9996 0.5 1.5Baygon 0.025–100 Y = 56,511.95 + 69,356.45X 0.9998 5.6 18.6Carbaryl 0.010–200 Y = 66,523.57 + 30,251.73X 0.9992 1.0 4.9Isoprocarb 0.025–100 Y = −56,730.70 + 74,346.90X 0.9992 3.0 12.7Baycarb 0.030–60 Y = −21,980.45 + 62,965.53X 0.9996 6.9 23.3Diethofencarb 0.025–50 Y = 38,362.25 + 125,428.78X 0.9995 5.0 16.9

Table 3Results for precision test.

Compound Repeatability (n = 5) Cartridge-to-cartridgeprecision RSD% (n = 7)

C1 (ng mL−1) RSD% C2 (ng mL−1) RSD% C3 (ng mL−1) RSD%

Pirimicarb 0.005 3.09 2 0.97 5 2.91 1.88Baygon 0.025 4.19 10 0.97 25 2.56 1.27Carbaryl 0.010 5.54 20 4.22 50 3.05 8.13

ataarwwtp

3

TacpoIc(

Fc5v

of pirimicarb were loaded onto the cartridges and eluted underthe optimized experimental conditions. Three replicates were car-

Isoprocarb 0.025 3.33 10

Baycarb 0.030 1.57 12

Diethofencarb 0.025 0.89 10

nd diethofencarb kept almost constant (p > 0.05). It was reportedhat carbamate pesticides are likely to decompose in strong acidicnd strong alkaline conditions, especially in strong alkaline solutiont pH > 10 [6]. In our experiment, the original pH values of the envi-onmental water samples are as follows: river water, pH = 7.01; lakeater, pH = 6.99; well water, pH = 6.64. They are all in the pH rangeith relatively constant and high extraction efficiency. Therefore,

o facilitate the extraction process, no adjustment of sample pH waserformed in the following experiments.

.3.4. Effect of sample volumeIn this part, the influence of sample volume was investigated.

he graphene-packed SPE cartridge was loaded with 1–100 mLqueous standard solutions containing 25 ng of baygon, diethofen-arb and isoprocarb, 50 ng of carbaryl, 30 ng of baycarb, and 5 ng ofirimicarb in all cases. Other experimental conditions were carried

ut as described in Section 2.6. The results are illustrated in Fig. 6.t could be seen that extraction recoveries did not change signifi-antly for pirimicarb, baygon, carbaryl, baycarb and diethofencarbp > 0.05) when the sample volume increased from 25 mL to 50 mL.

ig. 6. Effect of sample volume on the recoveries of six carbamate pesticides. Otheronditions: amount of carbamates, 25 ng of baygon, diethofencarb and isoprocarb,0 ng of carbaryl, 30 ng of baycarb, and 5 ng of pirimicarb; eluent, acetone; eluentolume, 5 mL.

4.01 25 1.07 1.623.13 30 4.45 1.461.03 25 2.42 3.11

But when sample volume increased up to 100 mL, the extractionrecoveries significantly reduced for all of the carbamates (p < 0.001).Therefore, 50 mL was employed as the loading volume.

3.4. Comparison with other sorbent materials

To evaluate the validity of graphene adsorbent, the extrac-tion efficiency of graphene was compared with several commonlyused commercialy available sorbent materials including VARIANBond Elut PRS, AGT CleanertTM ODS C18 and ENVI-Carb graphitizedcarbon black (GCB). The same amounts (30 mg) of different adsor-bents were accurately weighed and packed in 3 mL SPE cartridges.50 mL of sample solutions containing 25 ng of baygon, diethofen-carb and isoprocarb, 50 ng of carbaryl, 30 ng of baycarb, and 5 ng

ried out for each sorbent material. As shown in Fig. 7, graphenehas highest recoveries among those studied adsorbents except for

Fig. 7. Comparison of performance of graphene with several other adsorbents (PRS,C18 and graphitized carbon black (GCB)) for the SPE of six carbamate pesticides.Other conditions: sample, 50 mL of aqueous solution containing 25 ng of baygon,diethofencarb and isoprocarb, 50 ng of carbaryl, 30 ng of baycarb, and 5 ng of pirim-icarb; eluent, acetone; eluent volume, 5 mL.

Z. Shi et al. / J. Chromatogr. A 1355 (2014) 219–227 225

Table 4Recoveries of carbamate pesticides from river water, lake water and well water.

Compound Added (ng mL−1) River water recovery (%) Lake water recovery (%) Well water recovery (%)

Pirimicarb0.005 85.0 111.0 91.61 103.1 95.1 104.75 83.6 81.1 97.8

Baygon0.025 96.1 85.3 96.15 97.4 93.7 101.225 86.9 85.4 87.9

Carbaryl0.01 110.4 106.4 97.410 100.6 81.8 106.850 92.4 81.2 110.3

Isoprocarb0.025 103.2 101.5 97.15 106.0 107.9 104.925 94.9 87.7 85.0

Baycarb0.03 97.6 84.3 108.96 110.3 110.7 105.530 100.7 93.6 97.30.025 101.9 100.9 103.05 100.8 90.7 100.3

csim

tett3

Ft

Diethofencarb25 88.2

arbaryl. This is probably because that carbaryl has a naphthyl in itstructure and the �–� interaction between carbaryl and graphenes stronger compared with other sorbents. The results proved the

erits of graphene as SPE adsorbent.Furthermore, compared with commercially available SPE car-

ridges, the reusability of graphene-packed SPE cartridge is

xcellent. After proper regeneration, the graphene-packed SPE car-ridge could be reused. The regeneration procedures are as follows:he cartridge was eluted with 9 mL of acetone, 3 mL of methanol,

mL of acetonitrile and 9 mL of water in series. From the results of

ig. 8. The TIC chromatograms of the well water sample (top one) and the spiked well waR = 2.78 min; Isoprocarb, tR = 3.54 min; Baycarb, tR = 4.88 min; Diethofencarb, tR = 5.15 min

84.1 102.9

our experiment, the graphene-packed SPE cartridge could be usedat least 100 times for the adsorption of standard solution withoutsignificant loss of performance.

3.5. Method validation

The proposed method was validated in the following aspects:linearity, limit of detection (LOD), limit of quantification (LOQ),precision and recovery.

ter sample (bottom one). Pirimicarb, tR = 1.00 min; Baygon, tR = 2.39 min; Carbaryl,.

2 ogr. A

3

csibot(tm

3

alls5m

cfdcb3

3

mlstTtrat

3e

arwwdspaLteTwt

4

Utga

[

[

[

[

[

26 Z. Shi et al. / J. Chromat

.5.1. Linearity, LOD and LOQTo establish the calibration curves, 50 mL aqueous solutions

ontaining seven concentration levels of carbamate pesticidetandards were extracted and analyzed under the optimized exper-mental conditions. The results are summarized in Table 2. It coulde seen from Table 2 that excellent linear relationships werebtained for all the six carbamate pesticides with the correla-ion coefficients (R) ranging between 0.9992 and 0.9998. LODsS/N = 3) and LOQs (S/N = 10) varied from 0.5 to 6.9 ng L−1 and 1.5o 23.3 ng L−1, respectively. These data indicated that the proposed

ethod has excellent sensitivity.

.5.2. PrecisionPrecision was evaluated by carrying out repeatability test

t three different concentration levels. For each concentrationevel, five replicate samples were processed with the estab-ished method. The results are presented in Table 3. The relativetandard deviations (RSD) for repeatability test were below.54%, which demonstrates good precision of the establishedethod.To evaluate the cartridge-to-cartridge precision, seven repli-

ate graphene-packed SPE cartridges were prepared, and usedor the sample extraction according to the experimental proce-ures described in Section 2.6. The RSD values for the seven testedartridges are as follows: pirimicarb 1.88%, baygon 1.27%, car-aryl 8.13%, isoprocarb 1.62%, baycarb 1.46% and diethofencarb.11%.

.5.3. RecoveryRelative recovery experiment was carried out by spiking carba-

ate pesticide standards at low, medium and high concentrationevels into river water sample, lake water sample and well waterample containing known amount of carbamate pesticides. Thenhe spiked samples were processed by the established method.hree replicates were carried out for each concentration level, andhe results of relative recovery are summarized in Table 4. Theecoveries for the carbamate pesticides in river water, lake waternd well water (40 m-deep) samples were in the range from 81.1%o 111.0%.

.6. Application of the established method to the analysis ofnvironmental water samples

Under the optimized conditions, the established method waspplied to the determination of the six carbamate pesticideesidues in various environmental water samples. Water samplesere collected from different regions in China, including riverater (Tangshan, China), lake water (Taihu, China), forty-meter-eep well water (Qinhuangdao, China). The experimental resultshowed that baygon and carbaryl were not found in the water sam-les. Isoprocarb was not detected in river water and well water,nd it was found in lake water, but the concentration was less thanOQ. Baycarb, diethofencarb and pirimicarb were detected in allhe water samples, but the concentration levels were less than LOQxcept that 0.0062 �g L−1 of pirimicarb was detected in well water.he TIC chromatograms of the well water sample and the spikedell water sample are shown in Fig. 8. The enrichment factors for

he carbamate pesticides were in the range of 34.2–51.7.

. Conclusion

In this paper, graphene-based SPE was combined with

PLC–MS/MS to determine the carbamate pesticides for the first

ime. Possessing large surface area and strong adsorption ability,raphene proved to be a good adsorbent for the SPE enrichmentnd purification of the six carbamate pesticides. At the same time,

[

[

1355 (2014) 219–227

graphene-packed SPE cartridge has fine reusability (a graphene-packed SPE cartridge could be used at least 100 times based onthe experience in our lab). The cartridge-to-cartridge precisionwas good. Satisfactory linearity, repeatability and recovery wereachieved with graphene-packed SPE coupled to UPLC–MS/MS forthe analysis of six carbamate pesticides. LOD and LOQ are lowerthan other existing methods, demonstrating that the proposedmethod is highly sensitive. The results indicated that the proposedmethod could be used efficiently for the determination of tracecarbamates in various water samples.

Acknowledgments

Financial support from the National Natural Science Foundationof China (20875020, 20575016) and the Natural Science Foundationof Hebei Province China (B2013201234) are gratefully acknowl-edged. The authors also thank the Human Resources and SocialSecurity Department of Hebei Province for the financial supportfrom the Scientific and Technological Foundation for Selected Over-seas Chinese Scholars (2011). Special thanks are given to the projectsponsored by the Scientific Research Foundation for the ReturnedOverseas Chinese Scholars, State Education Ministry of China.

References

[1] S. Moinfar, M.R.M. Hosseini, Development of dispersive liquid-liquid microex-traction method for the analysis of organophosphorus pesticides in tea, J.Hazard. Mater. 169 (2009) 907–911.

[2] C. Przybylski, V. Bonnet, Combination of H-1 nuclear magnetic resonance spec-troscopy and mass spectrometry as tools for investigation of the thermolyticand solvolytic effects Case of carbamates analysis, J. Chromatogr. A 1216 (2009)4787–4797.

[3] X.H. Wang, J. Cheng, H.B. Zhou, X.H. Wang, M. Cheng, Development of a simplecombining apparatus to perform a magnetic stirring-assisted dispersive liquid-liquid microextraction and its application for the analysis of carbamate andorganophosphorus pesticides in tea drinks, Anal. Chim. Acta 787 (2013) 71–77.

[4] E. Ballesteros, M.J. Parrado, Continuous solid-phase extraction and gas chro-matographic determination of organophosphorus pesticides in natural anddrinking waters, J. Chromatogr. A 1029 (2004) 267–273.

[5] J. Cheng, Y.T. Xia, Y.W. Zhou, F. Guo, G. Chen, Application of an ultrasound-assisted surfactant-enhanced emulsification microextraction method for theanalysis of diethofencarb and pyrimethanil fungicides in water and fruit juicesamples, Anal. Chim. Acta 701 (2011) 86–91.

[6] A. Santalad, L. Zhou, F.J. Shang, D. Fitzpatrick, R. Burakham, S. Srijaranai, J.D.Glennon, J.H.T. Luong, Micellar electrokinetic chromatography with ampero-metric detection and off-line solid-phase extraction for analysis of carbamateinsecticides, J. Chromatogr. A 1217 (2010) 5288–5297.

[7] US Environmental Protection Agency, National Survey of Pesticides in DrinkingWater Wells, Phase II Report, EPA 570/9-91-020, National Technical Informa-tion Service, Springfield, VA, 1992.

[8] EU Council, EU Council Directive on the Quality of Water Intended for HumanConsumption, 98/83/CE, 1998.

[9] C.Y. Hao, B. Nguyen, X.M. Zhao, E. Chen, P. Yang, Determination of residual car-bamate, organophosphate, and phenyl urea pesticides in drinking and surfacewater by high-performance liquid chromatography/tandem mass spectrome-try, J. AOAC Int. 93 (2010) 400–410.

10] G.Y. Zhao, C. Wang, Q.H. Wu, Z. Wang, Determination of carbamate pesticides inwater and fruit samples using carbon nanotube reinforced hollow fiber liquid-phase microextraction followed by high performance liquid chromatography,Anal. Methods 3 (2011) 1410–1417.

11] N. Makihata, T. Kawamoto, K. Teranishi, Simultaneous analysis of carbamatepesticides in tap and raw water by LC/ESI/MS, Anal. Sci. 19 (2003) 543–549.

12] H. Chen, R.W. Chen, S.Q. Li, Low-density extraction solvent-based sol-vent terminated dispersive liquid-liquid microextraction combined with gaschromatography-tandem mass spectrometry for the determination of carba-mate pesticides in water samples, J. Chromatogr. A 1217 (2010) 1244–1248.

13] L. Guo, H.K. Lee, Low-density solvent based ultrasound-assisted emulsifi-cation microextraction and on-column derivatization combined with gaschromatography-mass spectrometry for the determination of carbamate pes-ticides in environmental water samples, J. Chromatogr. A 1235 (2012) 1–9.

14] E.Y. Yang, H.S. Shin, Trace level determinations of carbamate pesticides in sur-face water by gas chromatography-mass spectrometry after derivatization with9-xanthydrol, J. Chromatogr. A 1305 (2013) 328–332.

15] J.M. Soriano, B. Jiménez, G. Font, J.C. Moltó, Analysis of carbamate pesticidesand their metabolites in water by solid phase extraction and liquid chromatog-raphy: a review, Crit. Rev. Anal. Chem. 31 (2001) 19–52.

16] S.M. Goulart, R.D. Alves, A.A. Neves, J. Humberto de Queiroz, T. Condé de Assis,M.E.L.R. de Queiroz, Optimization and validation of liquid-liquid extraction

ogr. A

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

Z. Shi et al. / J. Chromat

with low temperature partitioning for determination of carbamates in water,Anal. Chim. Acta 671 (2010) 41–47.

17] M.P. Garcıı́a de Llasera, M. Bernal-González, Presence of carbamate pesticides inenvironmental waters from the northwest of Mexico: determination by liquidchromatography, Water Res. 35 (2001) 1933–1940.

18] G.Y. Zhao, S.J. Song, C. Wang, Q.H. Wu, Z. Wang, Solid-phase microextrac-tion with a novel graphene-coated fiber coupled with high-performance liquidchromatography for the determination of some carbamates in water samples,Anal. Methods 3 (2011) 2929–2935.

19] X.H. Wang, J. Cheng, X.F. Wang, M. Wu, M. Cheng, Development of an improvedsingle-drop microextraction method and its application for the analysis of car-bamate and organophosphorus pesticides in water samples, Analyst 137 (2012)5339–5345.

20] R. Sousa, V. Homem, J.L. Moreira, L.M. Madeira, A. Alves, Optimisation andapplication of dispersive liquid-liquid microextraction for simultaneous deter-mination of carbamates and organophosphorus pesticides in waters, Anal.Methods 5 (2013) 2736–2745.

21] C.F. Poole, New trends in solid-phase extraction, TrAC, Trends Anal. Chem. 22(2003) 362–373.

22] Z.P. Zang, Z. Hu, Z.H. Li, Q. He, X.J. Chang, Synthesis, characterization andapplication of ethylenediamine-modified multiwalled carbon nanotubes forselective solid-phase extraction and preconcentration of metal ions, J. Hazard.Mater. 172 (2009) 958–963.

23] L.L. El Atrache, R.B. Sghaier, B.B. Kefi, V. Haldys, M. Dachraoui, J. Tortajada,Factorial design optimization of experimental variables in preconcentration ofcarbamates pesticides in water samples using solid phase extraction and liquidchromatography-electrospray-mass spectrometry determination, Talanta 117(2013) 392–398.

24] S. Lissalde, N. Mazzella, V. Fauvelle, F. Delmas, P. Mazellier, B. Legube, Liquidchromatography coupled with tandem mass spectrometry method for thirty-three pesticides in natural water and comparison of performance betweenclassical solid phase extraction and passive sampling approaches, J. Chro-matogr. A 1218 (2011) 1492–1502.

25] N. Luque, S. Rubio, Extraction and stability of pesticide multiresidues from nat-ural water on a mixed-mode admicellar sorbent, J. Chromatogr. A 1248 (2012)74–83.

26] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V.Grigorieva, A.A. Firsov, Electric field effect in atomically thin carbon films, Sci-ence 306 (2004) 666–669.

27] K.P. Loh, Q.L. Bao, P.K. Ang, J.X. Yang, The chemistry of grapheme, J. Mater. Chem.20 (2010) 2277–2289.

28] M.D. Stoller, S. Park, Y. Zhu, J. An, R.S. Ruo, Graphene-based ultracapacitors,Nano Lett. 8 (2008) 3498–3502.

29] Q. Liu, J.B. Shi, G.B. Jiang, Application of graphene in analytical sample prepa-ration, TrAC, Trends Anal. Chem. 37 (2012) 1–11.

30] N.S. Ye, P.Z. Shi, Q. Wang, J. Li, Graphene as solid-phase extraction adsorbent forCZE determination of sulfonamide residues in meat samples, Chromatographia76 (2013) 553–557.

31] M.J. McAllister, J.L. Li, D.H. Adamson, H.C. Schniepp, A.A. Abdala, J. Liu, M.Herrera-Alonso, D.L. Milius, R. Car, R.K. Prud’homme, I.A. Aksay, Single sheet

[

1355 (2014) 219–227 227

functionalized graphene by oxidation and thermal expansion of graphite,Chem. Mater. 19 (2007) 4396–4404.

32] Q. Su, S.P. Pang, V. Alijani, C. Li, X.L. Feng, K. Müllen, Composites of graphenewith large aromatic molecules, Adv. Mater. 21 (2009) 3191–3195.

33] R. Sitko, B. Zawisza, E. Malicka, Graphene as a new sorbent in analytical chem-istry, TrAC, Trends Anal. Chem. 51 (2013) 33–43.

34] Q. Liu, J.B. Shi, L.X. Zeng, T. Wang, Y.Q. Cai, G.B. Jiang, Evaluation of graphene asan advantageous adsorbent for solid-phase extraction with chlorophenols asmodel analytes, J. Chromatogr. A 1218 (2011) 197–204.

35] K.J. Huang, Q.S. Jing, C.Y. Wei, Y.Y. Wu, Spectrofluorimetric determination ofglutathione in human plasma by solid-phase extraction using graphene asadsorbent, Spectrochim. Acta A: Mol. Biomol. Spectrosc. 79 (2011) 1860–1865.

36] Y.B. Luo, Z.G. Shi, Q. Gao, Y.Q. Feng, Magnetic retrieval of graphene: Extractionof sulfonamide antibiotics from environmental water samples, J. Chromatogr.A 1218 (2011) 1353–1358.

37] X.L. Wu, H.J. Hong, X.T. Liu, W.B. Guan, L.X. Meng, Y. Ye, Y.Q. Ma, Graphene-dispersive solid-phase extraction of phthalate acid esters from environmentalwater, Sci. Total Environ. 444 (2013) 224–230.

38] S. Wu, X.Q. Lan, L.J. Cui, L.H. Zhang, S.Y. Tao, H.N. Wang, M. Han, Z.G. Liu, C.G.Meng, Application of graphene for preconcentration and highly sensitive strip-ping voltammetric analysis of organophosphate pesticide, Anal. Chim. Acta 699(2011) 170–176.

39] Y.K. Wang, S.T. Gao, X.H. Zang, J.C. Li, J.J. Ma, Graphene-based solid-phaseextraction combined with flame atomic absorption spectrometry for a sensitivedetermination of trace amounts of lead in environmental water and vegetablesamples, Anal. Chim. Acta 716 (2012) 112–118.

40] Y.B. Luo, J.S. Cheng, Q. Ma, Y.Q. Feng, J.H. Li, Graphene-polymer composite:extraction of polycyclic aromatic hydrocarbons from water samples by stir rodsorptive extraction, Anal. Methods 3 (2011) 92–98.

41] J.B. Wu, Y.S. Qian, C.L. Zhang, T.L. Zheng, L.Y. Chen, Y.B. Lu, H.H. Wang,Application of graphene-based solid-phase extraction coupled with ultrahigh-performance liquid chromatography-tandem mass spectrometry fordetermination of macrolides in fish tissues, Food Anal. Methods 6 (2013)1448–1457.

42] L.Y. Chen, Y.B. Lu, S.Y. Li, X.J. Lin, Z.M. Xu, Z.Y. Dai, Application of graphene-based solid-phase extraction for ultra-fast determination of malachite greenand its metabolite in fish tissues, Food Chem. 141 (2013) 1383–1389.

43] W.B. Guan, Z.N. Li, H.Y. Zhang, H.J. Hong, N. Rebeyev, Y. Ye, Y.Q. Ma, Aminemodified graphene as reversed-dispersive solid phase extraction materialscombined with liquid chromatography-tandem mass spectrometry for pesti-cide multi-residue analysis in oil crops, J. Chromatogr. A 1286 (2013) 1–8.

44] W.S. Hummers, R.E. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc.80 (1958) 1339.

45] Q. Liu, J.B. Shi, L.X. Zeng, T. Wang, Y.Y. Cai, G.B. Jiang, Evaluation of graphene asan advantageous adsorbent for solid-phase extraction with chlorophenols as

model analytes, J. Chromatogr. A 1218 (2011) 197–204.

46] G.Y. Zhao, S.J. Song, C. Wang, Q.H. Wu, Z. Wang, Determination of triazineherbicides in environmental water samples by high-performance liquid chro-matography using graphene-coated magnetic nanoparticles as adsorbent, Anal.Chim. Acta 708 (2011) 155–159.