Received 7 March 2006; received in revised form 3 May 2006; accepted 12 May 2006Available online 19 June 2006
bstract
Analysis of polycyclic aromatic hydrocarbons (PAHs) standards in model systems was carried out by solid-phase microextraction (SPME) coupledo a direct extraction device (DED) and subsequent gas chromatography/mass spectrometry (GC/MS). PAHs standard was added to gelatine systemst different concentrations. Extraction process was carried out by SPME-DED at 25 ◦C for 60 min. Polydimethylsiloxane 100 �m (PDMS 100 �m),ivinylbenzene/polydimethylsiloxane 65 �m (DVB/PDMS 65 �m) and polyacrilate 85 �m (PA 85 �m) SPME fibres were tested. SPME-DEDatisfactorily extracted PAHs with a molecular weight (MW) lower than 206 from the gelatine system. All fibres showed a good reproducibilityresidual standard deviation (RSD) between 5.24% and 18.25%), linearity (regression coefficients between 0.8959 and 0.9983) and limit of detection
LOD) (between 0.008 and 0.138 ng mL−1). Presence of PAHs in different smoked meat products was also tested by SPME-DED. Different low
W PAHs were satisfactorily detected from all the foodstuffs studied. SPME-DED appears as a rapid, non-destructive technique for primarycreening of low MW PAHs in solid matrixes.
2006 Elsevier B.V. All rights reserved.
eis[a
ettliclt
eywords: SPME; PAHs; Direct extraction; Model systems; Smoked foods
. Introduction
The relation between the presence of polycyclic aromaticydrocarbons (PAHs) in foodstuffs and the development ofutagenic and carcinogenic processes is well known [1]. The
ccurrence of PAHs in foods may result from their sorptionrom a contaminated environment (unprocessed foods) or fromechnological processes (processed foods) [2]. PAHs have beenound in charcoal-broiled meat, smoked/grilled foods, fats andils, plant materials, seafood, liquid smokes and beverages3].
Most methodologies for analysis of PAHs in food systems areaborious, time consuming and need organic solvents. More-ver, the sampling step may deteriorate the quality of severalolid foods. Solid-phase microextraction (SPME) is an alterna-
ive technique that can overcome some of these disadvantages.PME integrates sampling, extraction, concentration and sam-le introduction in a simple process, and uses no solvent during
xtraction [4]. Previous studies have demonstrated the feasibil-ty of SPME for analysing PAHs in different types of samples,uch as environmental [5–7], biological [8,9] and foodstuffs10]. Nevertheless, a portion of sample is also needed for SPMEnalysis.
In previous papers we have described the use of a directxtraction device (DED) [11,12], which enables the introduc-ion of the SPME fibre in the core of solid matrixes, allowinghe analysis of volatile compounds from solid foods with no orittle deterioration of the product. Such device (Fig. 1) has annternal chamber in which the SPME fibre is exposed. This smallhamber has several holes that allow volatile compounds to equi-ibrate between the solid matrix and the internal headspace ofhe DED.
Extraction of volatile compounds from dry cured hams andate in one step, without sampling and thus, with only minor orone physical damage of the foodstuffs by SPME couple to DEDas been previously reported [11,12]. Moreover, Ventanas and
uiz [13] effectively extracted volatile nitrosamines by SPME-ED from solid model systems.The aim of this work was to study the feasibility of using
PME-DED for extraction of PAHs from solid matrixes
752 D. Martin, J. Ruiz / Talanta 71 (2007) 751–757
F rgemt e proc
mae
2
2
M1cbbai(
(bw
attps
2
4
(topaaaiacrod
2
Doocawu
st
ig. 1. Scheme of the use of SPME-DED in model systems of gelatine. An enlahe matrix to the headspace of the DED. The equilibriums that are implied in th
imicking solid foodstuffs and real smoked meat productsnd to compare different types of SPME stationary phases forxtraction of these compounds from the model systems.
. Experimental
.1. Reagents and materials
PAHs standard was supplied by Sigma–Aldrich (QTM PAHsix, Sigma–Aldrich, St. Louis, USA). This solution contained
Methanol (HPLC grade) was obtained from Scharlau ChemieBarcelona, Spain). Gelatine (300 bloom, A type) was suppliedy Sigma–Aldrich (St. Louis, USA). SPME stationary phasesere acquired from Supelco.Smoked meats products were obtained in a small village from
n area of Extremadura (Spain) in which home processing ofhis type of products is common. These products were: smokedenderloin, smoked jowl, smoked dry cured sausage and smokedotato sausage. All of them had been directly exposed to themoke for 30 days.
.2. Preparation of PAHs standards
PAHs stock solution at 8 �g mL−1 was prepared by diluting�L of the PAHs standard (2000 �g mL−1) in methanol.
8c
a
ent of a section of the gelatine shows the diffusion process of some PAHs fromess are also showed.
PAHs standard working gelatines at different concentrations0.1, 0.5, 1, 5 and 10 ng mL−1) were prepared by addition ofhe appropriate volume of the PAHs stock solution to 32 mLf gelatine solutions contained in polypropylene tubes witholyethylene caps. Gelatine solutions (20%) were prepared byddition of the appropriate amount of gelatine to distilled waternd microwave heating. The PAHs standard stock solution wasdded when the gelatine was at 60 ◦C. Working gelatines weremmediately sealed, vortexed for 30 s and stored at 2–4 ◦C untilnalysis. Gelatines of each concentration were prepared in dupli-ate, except for the 10 ng mL−1 ones, used for calculating theeproducibility, which were prepared in quintuplicate. Analysisf PAHs in gelatines by SPME-DED was carried out within 3ays after preparation.
.3. SPME-DED extraction of PAHs
PAHs were extracted from gelatines by directly inserting theED into the core of the gelatine and subsequent introductionf the SPME fibre into the DED following the procedure previ-usly described [12]. In order to control the temperature, tubesontaining the gelatines were kept in a thermostatised water batht 25 ◦C. Extraction time was 60 min. Fig. 1 shows the DED, theay in which it is coupled to SPME, and how SPME-DED issed to extract PAHs from gelatines.
Different SPME stationary phases were tested: polydimethyl-iloxane 100 �m (PDMS 100 �m), divinylbenzene/polydime-hylsiloxane 65 �m (DVB/PDMS 65 �m) and polyacrilate
5 �m (PA 85 �m). Parameters studied for each fibre were: effi-acy, reproducibility, linearity and limit of detection (LOD).
PAHs from smoked meat products were extracted followingprocedure similar to that used for gelatines, directly inserting
D. Martin, J. Ruiz / Talanta
Table 1PAHs included in the studied standard, showing the selected ion for detectionin MS (SIM mode) and obtained retention times
he DED into the product and subsequently inserting a PDMS100 �m) SPME fibre. Extraction conditions were 60 min atoom temperature (approximately 25 ◦C).
.4. Gas chromatography–mass spectrometry conditions
After extraction, analytes were desorbed onto the gas chro-atography column by inserting the fibre in the injector thatas set at the appropriate temperature (250 ◦C for PDMS andVB/PDMS or 300 ◦C for PA).
Analyses were performed using an Agilent 6890 series gashromatograph (Agilent, Avondale, USA) coupled to a masselective detector (Agilent 5973, Agilent, Avondale, USA).nalytes were separated using a 5% phenyl–methyl silicone
HP-5) bonded phase fused silica capillary column (Hewlett-ackard, 50 m × 0.32 mm i.d., film thickness 1.05 �m), operat-
ng at 60.2 kPa of column head pressure, resulting in a flow of.2 mL min−1 at 40 ◦C. The SPME fibres were desorbed during0 min. The injection port was in splitless mode. The tempera-ure program was 130 ◦C during 0.5 min, raised to 290 ◦C at aate of 5 ◦C min−1 and maintained at this temperature for 42 min.
The transfer line to the mass spectrometer (MS) was main-ained at 290 ◦C. The mass spectra were obtained by electronicmpact at 70 eV, a multiplier voltage of 1824 V and collectingata at a rate of 2.83 scan s−1. Detection of compounds was car-ied out in SIM mode. Selected ions chosen for each compoundnd retention times are shown in Table 1. In order to establish theetention time for each analyte, stock solutions were previouslynjected under the same chromatographic conditions as thoseor detection of PAHs in SIM mode but using the mass selectiveetector in scan mode over the m/z range of 30–550.
. Results and discussion
.1. PAHs from model systems of gelatine
Given that it would be very difficult to achieve an homo-eneous concentration of PAHs in those real food samples for
pi[s
71 (2007) 751–757 753
hich SPME-DED was initially developed (mainly whole pieceeat and fish products), we have tested the feasibility of usingPME-DED for analysing PAHs in solid samples in model sys-
ems made with gelatine. This allows a correct dilution of PAHshile the gelatine is still liquid, while showing similar charac-
eristics to muscle foods when it is solid, since it is quite elasticnd the matrix is constituted by proteins.
Fig. 2 shows the chromatogram of the stock solution of PAHsirectly injected in the gas chromatography–mass spectrome-ry (GC–MS) and the chromatograms obtained using differentPME stationary phases for SPME-DED extraction of PAHst 10 ng mL−1 from solid gelatines. Chromatographic areasbtained after extraction of PAHs at 10 ng mL−1 from solidelatines using SPME-DED are shown in Fig. 3.
Selection of the studied fibres was based on available scien-ific literature and Supelco recommendations. PDMS is a liquidnd non-polar polymer [14] that is commonly used in the extrac-ion of PAHs due to the non-polar nature of these compounds.A is a solid-phase for polar or slightly polar compounds [14]nd it has also tested for analysis of PAHs [15,18]. DVB/PDMSs a porous phase where DVB microspheres are immobilized byDMS [14]. However, found scientific literature dealing with
he application of this phase for PAHs is limited. In our case,DMS, DVB/PDMS and PA phases showed similar efficacy in
he extraction of PAHs from the standard working gelatines usingPME-DED (Fig. 3).
PDMS phases with lower thickness (7 and 30 �m) have beenlso used for PAHs extraction, since they improve extraction ofemivolatile compounds [5,6]. However, we previously tested a�m PDMS SPME fibre for extraction of PAHs from gelatiney SPME-DED and poorer results than using the 100 �m onend the other two phases (PA, PDMS/DVB) were observed.
The three evaluated phases only extracted 9 of the 16 PAHsresent in the standard, more specifically those firstly eluted andith lower molecular weight (MW) (NA, ACL, 2-BrNA, AC,L, PHE, AN, FA and PY). Thus, the lack of extraction of highW PAHs seems to be a drawback of the SPME-DED method.igh MW PAHs are those mainly recognized in carcinogenic
nd mutagenic processes [16]. However, the non carcinogenic-ty of most low MW PAHs has not been proved and they are stillnder research. On the other hand, Hansen et al. [17] pointedut that NA and PY might function as markers for carcinogenicAHs. Thus, this technique would be a rapid and interestingethod for a primary screening in the determination of the pres-
nce of low MW PAHs, both in solid foodstuffs or other kindf solid matrixes in which the presence of this specific toxicompounds would be interesting.
PAHs show a wide range of volatility, so the release of thoseith lower volatility to the headspace inside the DED during
xtraction is limited. These could explain the extraction of onlyPAHs but not the whole 16 PAHs in the standard and the
ecrease in chromatographic area of the 9 PAHs extracted withecreasing volatility of the analytes using the three evaluated
hases (Fig. 3). This fact was also observed by other authorsn the extraction of PAHs by headspace-SPME. Doong et al.18] satisfactorily extracted only 8 of 16 PAHs present in wateramples with 100 �m PDMS phase and PY was the heaviest
754 D. Martin, J. Ruiz / Talanta 71 (2007) 751–757
Fig. 2. Chromatograms obtained for the PAHs stock solution (8 �g mL−1) and the PAHs extracted by SPME-DED from gelatine model systems (10 ng mL−1) bydifferent SPME fibres.
D. Martin, J. Ruiz / Talanta 71 (2007) 751–757 755
Fig. 3. Effect of different SPME fibre coating on the extraction efficiencies forPAHs (10 �g l−1) from solid gelatines by SPME-DED.
Table 2Reproducibility (R.S.D.%) for studied PAHs extracted from solid gelatines bySPME
which caused a peak with a signal-to-noise ratio of 3 are shown
TR
H
NA2AFPAFP
A 13.05 17.30 5.68Y 9.18 18.25 9.33
AH found. Similarly, Eriksson et al. [5], studying presence ofAHs in soils by SPME, were not able to extract analytes witheavier MW than PY.
Reproducibility of the method expressed as relative stan-ard deviation (R.S.D.) for each studied PAH with consideredPME coatings is shown in Table 2. Values ranged from 5.24
o 17.72% for PDMS, 8.16 to 18.25% for DVB/PDMS and 5.68o 9.33% for PA. DVB/PDMS showed the worse reproducibilityor most compounds while PDMS and PA generated the bestesults for most PAHs (ACL, AC, FL, PHE, AN and PY). It haseen hypothesised that mixed SPME coatings show worse repro-
ucibility due to condensation of compounds into the poroushase [19]. R.S.D. values for most compounds were below 18%or the three kinds of SPME coatings. These results are simi-
iff
able 3egression equations and regression coefficients (R2) for studied PAHs extracted from
AP PDMS 100 �m PDMS/DVB
Equation R2 Equation
A y = 24643x − 7468.1 0.9681 y = 20922x −CL y = 18737x − 3114.3 0.9890 y = 16414x −-BrNA y = 4277.8x − 1284.3 0.9814 y = 3654.8x −C y = 11766x − 1003.1 0.9909 y = 9827.8x −L y = 11161x − 1404.4 0.9911 y = 9657.6x −HE y = 6987.7x + 952.99 0.9886 y = 6294.8x −N y = 6089.7x + 775.91 0.9983 y = 5869.7x −A y = 1847.9x + 911.95 0.9929 y = 1454.5x +Y y = 1438.8x + 941.88 0.9672 y = 1049.8x +
A 0.044 0.043 0.047Y 0.102 0.075 0.138
ar to those observed for other PAHs analysis techniques such aiquid–liquid partition [20], Soxhlet, microwave assisted extrac-ion or supercritical fluid extraction [21]. R.S.D. levels foundor PAHs using SPME by some authors were similar or slightlyower [7,18,22]. Non-equilibrium conditions used in our caseould partially explain such higher values. In addition, in com-lex matrixes, analytes can be trapped in the cells or interactith non-charged regions of the proteins, hindering the diffusion
nd, therefore, the achievement of the final equilibrium [23–25].btained reproducibility for PAHs by SPME-DED was better
han that observed in the case of extraction of nitrosamines fromodel systems by SPME-DED [13].The linearity of the SPME-DED method for analysing PAHs
rom solid gelatines was evaluated by plotting the calibrationurves of obtained chromatographic areas versus the concentra-ion of PAHs in gelatines ranging from 0.1 to 10 ng mL−1 ofach individual PAH. Most PAHs showed acceptable regres-ion coefficients above 0.9 for all studied SPME coatingsTable 3), PDMS showing the best linearity for most PAHs. Theorst regression coefficients were found for DVB/PDMS. Some
uthors have pointed out that those coatings which lead to worseeproducibility corresponded with those with a worse linearity26]. In our case, DVB/PDMS was also the SPME coating withworse linearity.
LOD, defined as the concentration of analytes in the gelatines
n Table 4. These values ranged from 0.008 to 0.102 ng mL−1
or PDMS, from 0.016 to 0.075 ng mL−1 for DVB/PDMS androm 0.012 to 0.138 ng mL−1 for PA. For all these three SPME
solid gelatines by SPME
65 �m PA 85 �m
R2 Equation R2
14399 0.9081 y = 19222x − 2486.9 0.93818769.8 0.9416 y = 16211x − 4823.4 0.97051454.7 0.9874 y = 4294.9x − 1916.2 0.95944041.8 0.9262 y = 9546.9x − 1883.5 0.96204480.4 0.9556 y = 10147x − 1481.7 0.96671804.1 0.9439 y = 6814.4x + 1796.9 0.9686827.87 0.9433 y = 6148.2x + 82.104 0.9726660.31 0.9555 y = 1574.3x + 1357.4 0.9540992.55 0.9314 y = 1094.7x + 471.91 0.8959
7 alanta
pi
lpabhpdpmttrhhSTsc
aGbFsf
M(aPmh
3
o
wa(t
dduPqpFsdestaltcbttev[t
McbcSut
56 D. Martin, J. Ruiz / T
hases, there was a general trend toward losing sensitivity withncreasing MW of PAHs and subsequent decrease in volatility.
Obtained LOD for PAHs by SPME-DED were similar or evenower than those reported for several PAHs using other analyticalrocedures. Dugay et al. [27] found LOD in lettuce between 4.36nd 39.04 ng g−1 when using saponification and SPE followedy GC–MS. Using SPME, Poon et al. [8] found LOD of PAHs inuman serum between 2.1 and 6.6 ng mL−1 with 100 �m PDMShase. The lower LOD found in our study is most likely partiallyue to the simplicity of the solid matrix compared to real sam-les, in which other components such as salt, fat and so on,ight interfere with extraction. Moreover, the small volume of
he chamber inside the DED could also contribute to enhancehe sensitivity of the method. In headspace-SPME, the lower theatio between the volumes of the headspace and the sample, theigher the sensitivity [14]. In our experiment, the volume of theeadspace is that of the chamber inside the DED in which thePME fibre is exposed, which in our prototype was 0.09 cm3.he sample volume is that of the gelatine (32 mL), giving a verymall ratio between both volumes of 2.8 × 10−3, which surelyontributes to the good sensitivity of the method.
Reported LOD values for PAHs in the scientific literaturere often above those obtained for SPME-DED. For example,omaa et al. [2] found levels of 6.1 ng g−1 for PHE in chickenarbecued wings, 18 ng g−1 for ACL in bacon or 60 ng g−1 forL in pork sausages. Thus, SPME-DED appears as a sufficientlyensitive technique for detection of low MW PAHs at the levelsrequently found in foodstuffs.
In conclusion, SPME-DED satisfactorily extracts only lowW PAHs (MW below 206) from solid matrixes. PDMS
100 �m), DVB/PDMS (65 �m) or PA (85 �m) showed goodnd similar efficacy, reproducibility, linearity and LOD, butDMS stands out from the others because its better results forost studied PAHs and other desirable characteristics like its
igher resistance or lower conditioning time.
.2. PAHs from smoked meat products
In order to evaluate the potential application of SPME-DEDn the extraction of PAHs from real foodstuffs contaminated
picd
Fig. 4. PAHs found in smoked meats products by S
71 (2007) 751–757
ith these toxic compounds, different smoked products werenalysed by SPME-DED with the previous selected SPME phasePDMS 100 �m). PAHs found for each smoked product withheir respective chromatographic areas are shown in Fig. 4.
As for gelatine gels, PAHs with MW above PY were notetected. Main reasons explaining this fact are the same as thoseiscussed previously for gelatines. Curiously, despite all prod-cts were smoked together at the same conditions, a differentAH profile was obtained for each one, both qualitatively anduantitatively. NA, ACL, FL, PHE and AN were detected in allroducts, but AC was only extracted from smoked tenderloin andA and PY were not found in smoked dry cured sausage. NAhowed the highest chromatographic area in the case of smokedry cured sausage and smoked tenderloin, whereas the high-st value corresponded to FA for smoked jowl and to PHE formoked potato sausage. These different results might be relatedo the different nature, structure or composition of the matrixesssayed. Moreover, their different content in lipids or salt couldead to a different release of PAHs into the headspace. In addi-ion, the presence of natural casings in some of them (smoked dryured sausage and smoked potato sausage) could have limited,ut not totally prevented, diffusion of PAHs from the surfaceo inside the product [28]. Furthermore, it should be consideredhat during ripening and smoking of these products, a great vari-ty of characteristic aroma compounds are generated [28]. Sucholatile compounds might interfere in the extraction process29], since the concentration of volatile compounds derived fromhe ripening and smoked processes is higher than PAHs levels.
In conclusion, SPME-DED satisfactorily extracts only lowW PAHs from smoked foodstuffs. Thus, it seems that the appli-
ation of SPME-DED on solid matrixes and foodstuffs is limitedy the volatility of the targeted compounds. However, this cir-umstance does not invalidate at all the potential application ofPME-DED on PAHs analysis, since SPME-DED would be aseful screening method for those studies in which the aim ishe evaluation of the presence of low MW PAHs, without sam-
ling and any laboratorial manipulation of the solid matrix. Thiss especially interesting in solid foodstuffs, in which the depre-iation of the product due to sampling could be an importantrawback.
PME-DED with their chromatographic areas.
alanta
A
Cpcdut
R
[[[[[
[
[
[
[[[[
[[[[[
D. Martin, J. Ruiz / T
cknowledgements
This research was supported by a Consejerıa de Sanidad yonsumo (Junta de Extremadura) project (ref. 03/84) “Puesta aunto de un metodo para deteccion de nitrosaminas e hidro-arburos policıclicos aromaticos en alimentos de manera noestructiva mediante microextraccion en fase solida acoplada an dispositivo de extraccion directa (SPME-DED)”. Diana Mar-in thanks the Ministry of Education for funding her research.
eferences
[1] National Toxicology Program, Department of Health and Human Ser-vices, Report on Carcinogens, available: http://ntp.niehs.nih.gov/ntp/roc/eleventh/profiles/s150pah.pdf.
[2] E. Gomaa, J. Gray, S. Rabie, C. Lopez-Bote, A. Booren, Food Addit. Con-tam. 5 (1993) 503.
[3] E. Menichini, B. Bocca, in: B. Caballero, L.C. Trugo, P.M. Finglas (Eds.),Encyclopedia of Food Sciences and Nutrition, Academic Press, Amster-dam, 2003, p. 4616.
[4] C.L. Arthur, J. Pawliszy, Anal. Chem. 62 (1990) 2145.[5] M. Eriksson, J. Faldt, G. Dalhammar, A.K. Borg-Karlson, Chemosphere
44 (2001) 1641.[6] J.J. Langenfeld, S.B. Hawthorne, D.J. Miller, Anal. Chem. 68 (1996) 144.[7] M. Llompart, K. Li, M. Fingas, J. Chromatogr. A 824 (1998) 53.[8] K. Poon, P.K.S. Lam, M.H.W. Lam, Anal. Chim. Acta 396 (1999)
303.
[
[[
71 (2007) 751–757 757
[9] S. Waidyanatha, Y. Zheng, S.M. Rappaport, Chem. Biol. Interact. 145(2003) 165.
10] M.D. Guillen, M.C. Errecalde, J. Sci. Food Agric. 82 (2002) 945.11] A.I. Andres, R. Cava, J. Ruiz, J. Chromatogr. A 963 (2002) 83.12] J. Ruiz, J. Ventanas, R. Cava, J. Agric. Food Chem. 49 (2001) 5115.13] S. Ventanas, J. Ruiz, Talanta, in press.14] J. Pawliszyn, Solid Phase Microextraction. Theory and Practice, Wiley-
VCH, New York, 1997.15] G. Gmeiner, C. Krassnig, E. Schmid, H. Tausch, J. Chromatogr. B 705
(1998) 132.16] Agency for Toxic Substances and Disease Registry, available: http://
Health 63 (1991) 247.18] R. Doong, S. Chang, Y. Sun, J. Chromatogr. A 879 (2000) 177.19] T. Gorecki, X. Yu, J. Pawliszyn, Analyst 124 (1999) 643.20] S. Moret, L.S. Conte, J. Chromatogr. A 882 (2000) 245.21] N. Saim, J.R. Dean, M.P. Abdullah, Z. Zakaria, J. Chromatogr. A 791 (1997)
361.22] M. Llompart, K. Li, M. Fingas, Talanta 48 (1999) 451.23] Z. Baines, E. Morris, Food Hydrocol. 1 (1987) 197.24] A. Boland, K. Buhr, P. Giannouli, S. van Ruth, Food Chem. 86 (2004) 401.25] J. Guinard, C. Marty, J. Food Sci. 60 (1995) 727.26] P. Popp, A. Pasckhe, Chromatographia 46 (1997) 419.
27] A. Dugay, C. Herrenknecht, M. Czok, F. Guyon, N. Pages, J. Chromatogr.
A 958 (2002) 1.28] J.A. Maga, Smoke in Food Processing, CRC Press, Boca Raton, FL, 1988.29] D.D. Roberts, P. Pollien, C. Milo, J. Agric. Food Chem. 48 (2000)
