A selective method for the determination of benzo[a]pyrene in soil using porous graphitic carbon liquid chromatography columns

A selective method for the determination of benzo[a]pyrene in soil using porous graphitic carbon liquid chromatography columns

Talanta 59 (2003) 845 /848 www.elsevier.com/locate/talanta Short communication A selective method for the determination of benzo[a]pyrene in soil u...

98KB Sizes 10 Downloads 109 Views

Talanta 59 (2003) 845 /848 www.elsevier.com/locate/talanta

Short communication

A selective method for the determination of benzo[a]pyrene in soil using porous graphitic carbon liquid chromatography columns Ameera Al-Haddad Department of Chemistry, University of Bahrain, P.O. Box 32038, Isa Town, Bahrain Received 17 September 2002; accepted 20 November 2002

Abstract A back-flushing procedure using porous graphitic carbon (PGC) HPLC columns has been used successfully for the cleanup of soil samples for the determination of benzo[a]pyrene in ppb levels by an ODS-fluorescence HPLC column. The procedure was tested on nine random soil samples taken from an industrial area of the Kingdom of Bahrain. The mean percent recovery from the PGC column was 96% and the average coefficient of variation for the whole method was 5.2%. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Benzo[a]pyrene; Porous graphitic carbon; Soil analysis; Polycyclic aromatic hydrocarbons

1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are fused-ring compounds formed in both natural and manmade processes such as the incomplete combustion of various kinds of materials. They are released to the atmosphere and then contaminate the air-suspended particulates. Both the fallout of these particulates and PAHs in the vapor phase pollute the environmental soil and sand. Some PAHs have mutagenic and carcinogenic activities

E-mail address: [email protected] (A. Al-Haddad).

and benzo[a]pyrene is of special interest since many researchers have demonstrated that it is probably the most potential animal carcinogen [1]. The established carcinogenic activity of benzo[a]pyrene and their extensive environmental occurrence have led to a widespread interest in the development of analytical methods for their isolation and measurement in various matrices. These methods reported in the literature include mass spectrometry, fluorescence spectrometry, gas chromatography, thin-layer chromatography, liquid chromatography and combination of these techniques. In most cases, the first analytical step is solvent extraction of the organic compounds

0039-9140/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0039-9140(02)00619-7

846

A. Al-Haddad / Talanta 59 (2003) 845 /848

from the sample. This is followed by liquid column chromatography cleanup and separation. Silica gel columns were mostly used for this purpose [2]. In this paper, the chromatography of benzo[a]pyrene on porous graphitic carbon (PGC) columns is studied and a back-flushing procedure was used for the cleanup of soil samples for the determination of benzo[a]pyrene. PGC is a novel chromatographic material [3] consisting of porous carbon spheres. It has a surface area of about 150 m2 g1, a particle porosity of 70%. PGC is the only carbon that can be used as a packing material for HPLC, because of its strength and ability to withstand the high-pressure gradients used in HPLC [4]. PGC has been used for the fractionation of chlorinated aromatic compounds including PCBs, PCDFs, PCDDs and pesticides and for the cleanup of soil samples for the determination of PCDDs and PCDFs [5,6].

2. Experimental 2.1. LC equipment LKB model 2150 LC solvent delivery system, PE series 3000 fluorescence detector and a Rheodyne Model 7125 syringe loading sample injector with a 100 ml sample loop. 2.2. PGC/LC column Fifty /4.7 mm2 i.d., 7 mm particle size (Hypercarb, Shandon Scientific Ltd, UK). Eluent was methanol, dichloromethane and cyclohexane at 5 ml min 1.

2.4. Chemicals Benzo[a]pyrene was purchased from BDH (Poole, UK). All solvents used were of HPLC grade (Fluka plc). 2.5. Analysis method Soil samples were collected from an industrial area of the Kingdom of Bahrain. They were air dried and sieved through a 2 mm mesh to remove large particles. Each soil sample (100 g) was subjected to ultrasonic extraction using cyclohexane. The extracted material was reduced to 2 ml using a rotary evaporator. The extract was then applied to a column packed from the top with anhydrous sodium sulphate, sulphuric acid (1:1 w/ w), silica, anhydrous sodium sulphate. With cyclohexane as the elution solvent, the second 10 ml portion is collected as the benzo[a]pyrene fraction. This fraction was then concentrated to about 50 ml and applied to the PGC column. The PGC column was flushed with 40 ml of methanol in the forward direction, then back flushed with 10 ml dichloromethane (as an intermediate solvent) then with 40 ml of cyclohexane to recover the benzo[a]pyrene. The later fraction was concentrated and quantified by the ODS column using acetonitrite: water (75:25 v/v) as the mobile phase. Calibration curves for benzo[a]pyrenes were constructed for this purpose using fluorescence as the detection device. Peak identification was established by comparing the retention times of peaks in the sample chromatogram with those in the standard chromatograms. Further confirmation of benzo[a]pyrene identity was obtained by GC-FID.

3. Results and discussion 2.3. ODS/LC column One hundred and fifty /4.7 mm2 i.d. reversedphase octadecylsilane column. The ODS column was connected to the fluorescence detector which was operated with an excitation wavelength of 254 nm (slit width 10 nm) and an emission wavelength of 420 nm (slit width 10 nm). Eluent was acetonitrite /water (75:25 v/v), at 2 ml min1.

The PGC column produced a suitable fraction to allow the rapid determination of benzo[a]pyrene in soil samples covering a wide range of complexities. The detection limit of this technique, determined as 3 times the detector noise, was 10 ppb (10 ng g1). The average recovery for 50 ng ml1 benzo[a]pyrene standard was 96% on the PGC Column and about 87% for the whole method. The

A. Al-Haddad / Talanta 59 (2003) 845 /848

analysis time, including re-equilibrium times was under 2 h. Fig. 1a shows an ODS chromatogram of a benzo[a]pyrene standard solution. The fluorescence was used because its sensitivity and the selectivity are considerably superior to that of flame ionization detectors generally used for the

847

GC analysis of benzo[a]pyrene. Furthermore, samples analysed by GC require laborious sample preparation and cleanup procedures. Fig. 1 shows the GC-ODS chromatography obtained for a typical soil extract before (Fig. 1b) and after (Fig. 1c) cleanup by PGC. Soil sample extracts were cleaned up by PGC using the back-flushing procedure described earlier. In both cases, the same pretreatment method was applied to the extracts, since the relatively low capacity of PGC necessitated the removal of the bulk of coextracted organics, which would otherwise overload the PGC column. Fig. 1c demonstrates the effectiveness of PGC in eliminating a considerable amount of the fluorescence material in the soil sample extracts making the measurement of benzo[a]pyrene by the ODS-fluorescence HPLC system relatively easy and reliable. Table 1 presents the quantitative results for benzo[a]pyrene in nine soil samples collected from the different parts of the country. The mean concentration value was approximately 37.9 ppb. It is believed that the main source of benzo[a]pyrene pollution is the various petroleum/ oil industries distributed around the country. The results of different aliquots of each sample showed good reproducibility and the precision obtained for the whole method was typically 5.2% which is relatively better than precision values reported elsewhere for trace level benzo[a]pyrene determination in different sample matrices. The retention volume of benzo[a]pyrene Table 1 Concentration of benzo[a]pyrene in soils Sample no.

Fig. 1. ODS-fluorescence chromatograms of (a) standard benzo[a] pyrene; (b) soil extract before PGC cleanup and (c) soil extract after PGC cleanup. 150/4.7 mm2 i.d. column; EX. Wavelength: 254 nm; EM. Wavelength: 420 nm; Eluent: acetonitrile: water (75:25 v/v) at 2 ml min1.

1 2 3 4 5 6 7 8 9

Concentration ng g 1

CV (%)

30.2 16.5 32.3 16.0 18.5 55.9 92.0 16.5 63.6

3.9 3.8 6.6 4.7 6.5 5.9 3.6 6.6 5.2

CV, coefficient of variation for six replicates.

848

A. Al-Haddad / Talanta 59 (2003) 845 /848

on the PGC column was reproducible even after analysing many soil samples. A relatively rapid and reliable cleanup method has been used and applied in the preparation of soil samples for the qualitative and quantitative data of parts per billion levels of benzo[a]pyrene. The method can probably be applied to other matrices. This cleanup method is now used in our laboratory to carry out a survey of the background level of benzo[a]pyrene in the soil of the Kingdom of Bahrain.

References [1] M.S. Zedeck, J. Environ. Pathol. Toxicol. 3 (1980) 537. [2] S. Tsuge, K. Nishimura, M. Suzuki, H. Hayasha, Anal. Sci. 4 (1988) 115. [3] J.H. Knox, B. Kaur, J. Chromatogr. 352 (1986) 3 /25. [4] J.H. Knox, K.K. Unger, H. Muller, J. Liq. Chromatogr. 6 (S-1) (1983) 1. [5] C.S. Creaser, A. Al-Haddad, Anal. Chem. 61 (11) (1989) 1300. [6] A. Al-Haddad, J. AOAC Int. 77 (2) (1994) 437.