- Email: [email protected]
Removal of carotene-like colored compounds by liquid–liquid extraction during polycyclic aromatic hydrocarbons analysis of plant tissue
Available online at www.sciencedirect.com
Chemosphere 70 (2008) 2002–2008 www.elsevier.com/locate/chemosphere
Removal of carotene-like colored compounds by liquid–liquid extraction during polycyclic aromatic hydrocarbons analysis of plant tissue Takayuki Kobayashi, Kengo Morimoto, Kenji Tatsumi
*
National Institute of Advanced Industrial Science and Technology (AIST), 16-1, Onogawa, Tsukuba, Ibaraki 305-8506, Japan Received 6 June 2007; received in revised form 14 September 2007; accepted 14 September 2007 Available online 14 November 2007
Abstract Plants contain a wide variety of chemicals, some of which may have similar chromatographic behavior to polycyclic aromatic hydrocarbons (PAHs). During solid phase extraction (SPE) with Si-gel for instance, the co-elution of carotene-like colored compounds with PAHs has been observed. In this paper, liquid–liquid extraction was applied for the separation and subsequent analysis of PAHs from plant extracts. PAHs containing 2–6 rings, which include naphthalene, phenanthrene, pyrene, benzo[a]pyrene and benzo[ghi]perylene, were used as representative target chemicals. Carotene-like compounds extracted from Komatsuna (Brassica campestris) shoot by acetone followed by Si-gel treatment were incorporated as undesired components in the model matrix. Results showed the feasibility of employing either acetonitrile or 2% (w/v) KOH–methanol as solvents for high PAHs recovery and low extraction of colored fraction. For acetonitrile, 86.9–93.5% of each PAH could be recovered after three extraction cycles (relative standard deviation, RSD < 1.6%) with only about 10% co-extraction of colored fraction. For 2% KOH–methanol, PAHs recoveries ranging from 79.3% to 83.1% after five cycles (RSD < 1.5%) were achieved while the percent extraction of colored fraction was also low at 10%. The relatively higher selectivity of the solvents for PAHs over the colored fraction as well as the solubility of the matrix solution in the solvent may have contributed to these results. On this basis, liquid–liquid extraction is very useful for the pre-treatment of plant extracts for PAHs analysis. 2007 Elsevier Ltd. All rights reserved. Keywords: Clean-up; Phytoremediation; Sample pre-treatment; Solid phase extraction (SPE)
1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are one of the most important classes of toxic pollutants that are ubiquitously found in the environment (Juhasz and Naidu, 2000; Mastral et al., 2003; Chen and White, 2004). Their occurrence is mainly attributed to a variety of anthropogenic activities such as incomplete combustion of fossil-fuels (Junk and Ford, 1980; Wilcke, 2007). Several PAHs, such as fluoranthene and benzo[a]pyrene, which are likely carcinogens, are listed as priority pollutants by United States of Environmental Protection Agency (USEPA, 1985). In *
Corresponding author. Tel.: +81 29 861 8325; fax: +81 29 861 8242. E-mail address: [email protected] (K. Tatsumi).
0045-6535/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2007.09.027
recent years, much attention has been paid to the exposure of plants to PAHs via atmospheric deposition and adsorption from the soil (Simonich and Hites, 1994; Kipopoulou et al., 1999; Howsam et al., 2001; Kazerouni et al., 2001; Berber et al., 2004). On the other hand, there are also numerous studies regarding the application of phytoremediation to treat PAHs contaminated soil using a variety of plant species (Liste and Alexander, 2000; Fismes et al., 2002; Maila and Cloete, 2002; Ke et al., 2003; Gao and Zhu, 2004; Parrish et al., 2006). In all of these studies, accurate and efficient analysis of PAHs content of plant tissues is very important. Plants contain a wide variety of chemicals (e.g. pigments, fatty acids and alcohols) and some of these may have very similar chromatographic behavior to PAHs.
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
The direct injection of untreated plant extracts into the GC/MS may not be appropriate for instrument safety and accuracy of PAHs analysis. Clean-up of plant extracts is therefore very important. Solid phase extraction (SPE) method with silica (Tao et al., 2004; Lin et al., 2007; Yang and Zhu, 2007), florisil (Kazerouni et al., 2001), alumina (Ratola et al., 2006) and C18 (Reilley et al., 1996) has been used for sample purification. However, during SPE with silica gel as stationary phase and dichloromethane as eluent, the co-elution of a yellow-colored fraction, which has a similar visible spectrum pattern with b-carotene, has been quite persistent. In addition, carotene, sterol and alcohol are also co-eluted with PAHs during SPE using florisil (Meudec et al., 2006) to indicate the general difficulty of separating certain compounds. In order to improve sample pre-treatment, a combination of the above-mentioned stationary phases (Tremolada et al., 1996; Wenzel et al., 1998) has been employed during SPE. Size exclusion chromatography (Hubert et al., 2003) has also been considered. However, these methods prolong and thus generally complicate the overall analysis. In the present paper, liquid–liquid extraction was evaluated as a simple alternative to remove colored components from PAHs-containing samples. Specifically, the optimization of extraction conditions (i.e. extraction solvent and number of extraction cycle) for separating 2–6 rings PAHs from contaminating compounds was considered. An extract from the shoot of Komatsuna (Brassica campestris), which is known to contain high amounts of carotene-like chemicals (Hels et al., 2004), was used as model matrix to simulate the yellow-colored fraction commonly encountered during purification of plant extract. 2. Materials and methods 2.1. Reagents and plant material Five types of powdered PAHs (purity >98%) were used in the present study: naphthalene and phenanthrene from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan); pyrene from Nacalai Tesque, Inc. (Kyoto, Japan); benzo[a]pyrene and benzo[ghi]perylene from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). All other reagents were of special grade obtained from Nacalai Tesque or Kanto Chemical Co., Inc. (Tokyo, Japan). The shoot of Komatsuna (B. campestris) was used for the preparation of matrix solution. The vegetable was purchased at a local marketplace, washed with tap water and then stored in the dark at 20 C until use. The 5 PAHs contents of the plant sample were below the detection limit. 2.2. Preparation of PAHs-spiked solution Komatsuna shoots were cut into small pieces and homogenized for 3 min with acetone in an Erlenmeyer flask. The mixture was ultrasonicated for 30 min and filtered through Whatman GF/F glass filter (Whatman Japan
2003
KK., Tokyo). The filtrate was transferred in a separatory funnel and then hexane and distilled water were added. The funnel was shaken and then allowed to stand for a few minutes after which the hexane phase including the remaining stable emulsion was obtained. The hexane extract was washed twice with distilled water, and the excess water was removed over anhydrous Na2SO4. The extract was concentrated in a rotary evaporator at 30 C. To remove chlorophyll and other interfering components, the concentrated solution was deposited onto a silica gel column (Wakogel C300, Wako pure chemical Industries; 12 cm length by 3.2 cm diameter). The column was eluted with hexane (60 ml), and then with dichloromethane. A yellow-colored fraction was eluted from top of the column as soon as the solvent was changed to dichloromethane. The solvent was loaded until the eluent became almost colorless. The hexane and dichloromethane eluents were combined and the resulting solution was evaporated in a rotary evaporator at 30 C to remove the dichloromethane. The concentrated yellow-colored solution was used as matrix solution. It was spiked with the stock solutions of the 3 PAHs (3-ring phenanthrene, 4-ring pyrene and 5-ring benzo[a]pyrene; 500 mg l1 for each PAH in acetone) to come up with concentrations of 0.1 and 1.0 mg l1 for each PAH. The PAHs-spiked solutions were stored in the dark at 20 C until use. 2.3. Optimization for PAHs determination by liquid–liquid extraction Methanol, HCl–methanol, KOH–methanol and acetonitrile were used as solvents for liquid–liquid extraction and the number of extractions was varied from one to five cycles. The concentrations of HCl (v/v) and KOH (w/v) were 1–5%. The volume of extraction solvent was 1.5 ml per one cycle except for acetonitrile, where 3 ml per one cycle was employed. The detailed procedure was as follows: 5 ml of PAHs-spiked solution and the specified volume of extraction solvent were combined in a glass test tube and then subjected to vortex mixing for 30 s. The lower phase was collected and the upper hexane phase was re-extracted with an equal volume of extraction solvent. The operation was repeated up to the specified cycles. In the case of methanol as solvent, 3 ml of hexane was added into the tube before the first extraction. The initial addition of hexane in this case was made to compensate for its high affinity for methanol as compared with the other solvents. This would assure that even after five cycles of extraction, an upper hexane phase would remain. The remaining hexane phase was stored for estimation of percent extraction of colored fraction as described in Section 2.5. All the extracts collected in each cycle were combined in a 100 ml screw glass vial and then 70 ml of 1% (w/v) aqueous NaCl and 10 ml of hexane were added. The vial was shaken and the upper hexane layer was obtained. The hexane extract was washed with distilled water and passed through anhydrous Na2SO4 column. The solution was concentrated to a final
2004
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
volume of approximately 5 ml under a gentle stream of nitrogen inside a Zymark-TurboVap LV evaporator (Caliper Life Sciences, Hopkinton, MA) that was operated with an ice water bath at 4–6 C. After the addition of 100 ll of syringe spike (50 mg l1 anthracene solution in acetone), the solution volume was made up to 10 ml with hexane in a volumetric flask. This solution was used as sample for GC/MS and UV–Vis analysis. All experiments were conducted at ambient temperature (20–25 C) and in triplicates. 2.4. GC/MS analysis and determination of PAHs The determination of PAHs concentration was carried out using a Shimadzu GCMS-QP2010 system (Shimadzu Co., Kyoto, Japan) equipped with AOC-20i autosampler and Quadrex 007-1 capillary column made of 100% dimethylpolysiloxane (0.25 mm ID · 25 m · 0.25 lm film thickness; Quadrex Co., New Haven, CT). Helium (purity >99.9999%) was used as the carrier gas at a constant pressure (100 kPa). Samples (1 ll) were injected in splitless mode. The temperature program for the GC oven was as follows: initial temperature 50 C for 1.5 min, heating to 180 C at 18 C min1 (during analysis with 5 PAHs including naphthalene and benzo[ghi]perylene, the heating rate was reduced to 8 C min1), then to 300 C at 12 C min1 with a final holding time of 10 min. The injector and interface temperatures were both 300 C. Mass spectrometry was performed using the electron ionization (EI) mode and ion source temperature was 200 C. Acquisition was carried out in the selective ion monitoring (SIM) mode, using the following characteristic masses: naphthalene, 128; phenanthrene and anthracene (syringe spike), 178; pyrene, 202; benzo[a]pyrene, 252, and benzo[ghi]perylene, 276. The amount of PAHs in the sample solution was determined, by the absolute calibration method, from the peak area of each PAH and that of anthracene on GC/MS chromatogram. To obtain the calibration curve, hexane solutions with various concentrations of PAHs were prepared (0.1–0.8 mg l1). Percent recoveries of PAHs were calculated from the amounts of PAHs in the sample solution and those in the initially spiked solution. 2.5. UV–Vis analysis and estimation of percent extraction of colored fraction UV–Vis spectra were measured with Jasco V-550 spectrophotometer (Jasco Co., Tokyo, Japan). The remaining hexane phase, which was stored for the estimation of percent extraction of colored fraction as described in Section 2.3 was evaporated to dryness with a rotary evaporator at 40 C and then reconstituted with 5 ml of hexane. Absorption values of the solution at 473 nm and 449 nm were measured. The percent extraction of colored fraction in this case was calculated using the following equation:
Percent extraction ð%Þ ¼
1
Ek sample soln: Ek initially spiked soln:
100
where Ek is the absorbance at k (nm) (i.e. 473 nm and 449 nm: the colored fraction displayed absorption maxima at these wavelengths in hexane) of the sample or initially spiked solution before extraction. No absorption was detected at 473 nm and 449 nm in the hexane solution containing the 5 PAHs. All experiments were performed in triplicates. In the present paper, though quantitatively similar, the terms ‘‘percent recovery’’ and ‘‘percent extraction’’ were used in the case of PAHs and colored fraction, respectively, to distinguish between the target compounds (PAHs) and the impurities being removed (colored fraction) from the sample solution for GC/MS analysis. 2.6. Method validation This experiment was conducted using the best condition that was established from the initial data with the 3 PAHs. To further evaluate its applicability to a wider range of PAHs compounds, 2-ring naphthalene and 6-ring benzo[ghi]perylene were additionally incorporated during this stage. The precision of the method, which is expressed as relative standard deviation (RSD), was evaluated in terms of repeatability and intermediate precision. Repeatability was assessed by performing five replicates on the same day. In the case of intermediate precision, three replicates were analyzed on three different days by the same analyst. The limits of detection (LOD) were calculated using a signal to noise ratio of 3 (S/N = 3). 3. Results 3.1. Effect of methanol solvent system on recovery of PAHs and percent extraction of colored fraction The UV–Vis spectra of the initial solution that was spiked with 3 PAHs and that of the sample solution for GC/MS analysis obtained after extracting with methanol, 2% KOH–methanol and acetonitrile are shown in Fig. 1. All samples were subjected to the same dilutions and number of extraction cycles of five. The spectrum of initial solution showed absorption maxima at 473 nm and 449 nm. The observed wavelengths for maximum absorption as well as the general shape of the spectrum are characteristic pattern of b-carotene (Kochubei, 1968). The same absorption maxima were also detected in the spectra of the sample solutions that were extracted with either 2% KOH–methanol or acetonitrile (Fig. 1). However, their absorption values at 473 nm and 449 nm were lower at around 0.04 for both wavelengths. With these very low absorbances, which are difficult to measure accurately, we therefore defined the percent extraction of colored fraction based on the equation described in Section 2.5. The detailed data for the percent extractions of colored fraction will be discussed in a later section.
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008 0.6 Initially spiked solution Methanol 2%KOH-methanol Acetonitrile
0.5
Absorbance
449
0.4
473
0.3
0.2
0.1
0.0 350
400
450
500
550
600
Wavelength (nm) Fig. 1. UV–Vis spectra of the initially spiked solution (1.0 mg l1 3 PAHs) and sample solution for GC/MS analysis obtained after extracting with methanol, 2% KOH–methanol and acetonitrile (solvent; hexane). All samples have the same dilution ratio and number of extraction cycle is five.
Fig. 2 shows the effect of the number of extraction cycle on the percent recovery of PAHs and the percent extraction of colored fraction from the 1.0 mg l1 PAHs-spiked solution by methanol. The PAHs recovery tended to increase with the number of extractions, so that over 90% of each PAH could be recovered after three cycles (93.5 ± 1.2% for phenanthrene, 99.5 ± 0.3% for pyrene, and 106.8 ± 1.6% for benzo[a]pyrene). For the colored fraction, on the other hand, the percent extractions, which were calculated from the absorbance at either 473 nm or 449 nm, were almost the same and similar results were found in the other solvent systems. Parallel with the PAHs data, the percent extraction of colored fraction also increased with the number of extractions up to the third cycle, after which, the values remained more or less constant (Fig. 2b). The percent extractions of colored fraction after five cycles were 68.4 ± 2.5% and 70.0 ± 2.0% at 473 nm
2005
and 449 nm, respectively. These values were almost consistent with the percentage intensities of methanol solution at 473 nm and 449 nm wavelengths relative to the initially spiked solution (Fig. 1). This indicates that the equation described in Section 2.5 was valid for the calculation of extraction percentage of colored fraction. On the basis of the above results, we further evaluated the extraction of colored fraction with the addition HCl or KOH since plant matrices contain chemicals with various functional groups (e.g. hydroxyl and carboxyl group) whose speciation are pH-dependent. The data for methanol at different acid and base concentrations are shown in Fig. 3. Although recoveries of PAHs decreased with increasing HCl or KOH concentration, over 80% of each PAH could still be recovered at 2% acid or base even after five cycles (Fig. 3a and c). Comparing the PAHs recovery data at the same concentration, no clear differences were found between HCl or KOH addition except that at 5%, relatively higher recoveries were achieved in the former (Fig. 3a and c). On the other hand, the percent extraction of colored fraction with KOH was significantly lower than with HCl at the same concentration. At 2% KOH, the percent extraction at 473 nm and 449 nm were only 9.9 ± 1.7% and 10.7 ± 1.5%, respectively (Fig. 3d), whereas values for 2% HCl were both close to 30% (Fig. 3b). 3.2. Effect of acetonitrile extraction system on recovery of PAHs and percent extraction of colored fraction Fig. 4 shows the effect of the number of extraction cycle on the percent recovery of PAHs and percent extraction of colored fraction from the PAHs-spiked solution (1.0 mg l1) by acetonitrile. The recoveries tended to increase with the number of extractions as in the case of methanol. After three cycles, over 90% for each of the 3 PAHs was recovered (90.0 ± 1.2% for phenanthrene, 92.6 ± 0.7% for pyrene, and 99.6 ± 0.5% for benzo[a]pyrene). Almost similar PAHs recoveries were also obtained
(a) PAHs
(b) Colored fraction
120
100 1 cycle 80
3 cycles 4 cycles
80
5 cycles
60
40
Extraction (%)
Recovery (%)
473 nm 449 nm
2 cycles
100
60
40
20
20
0
0 Phenanthrene
Pyrene
Compound
Benzo[a]pyrene
1
2
3
4
5
Extraction cycles
Fig. 2. (a) Recovery of PAHs and (b) percent extraction of colored fraction from spiked solution (1 mg l1 3 PAHs) by methanol at different number of extraction cycle (n = 3). Percent extraction of colored fraction was estimated from the absorption values at 473 nm and 449 nm.
2006
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
(b) Colored fraction (HCl-methanol)
(a) PAHs (HCl-methanol) 120
100 0%
473 nm
1%
100
60
40
Extraction (%)
Recovery (%)
5%
80
449 nm
80
2%
60
40
20
20
0
0 Phenanthrene
Pyrene
Benzo[a]pyrene
0
Compound
1
2
5
HCl concentration (%, v/v) (d) Colored fraction (KOH-methanol)
(c) PAHs (KOH-methanol) 120
100 0%
473 nm
1%
100
5%
80
Extraction (%)
Recovery (%)
449 nm
80
2%
60
40
60
40
20
20
0 Phenanthrene
Pyrene
0
Benzo[a]pyrene
Compound
0
1
2
5
KOH concentration (%, w/v)
Fig. 3. PAHs recovery and percent extraction of colored fraction from 1 mg l1 PAHs-spiked solution by HCl–methanol (a and b, respectively) and KOH–methanol (c and d, respectively) at different acid or base concentration (n = 3). Percent extraction of colored fraction was estimated from the absorption values at 473 nm and 449 nm. Number of extraction cycle is five.
(b) Colored fraction
(a) PAHs 120
100 1 cycle
473 nm
2 cycles
100
449 nm
80
3 cycles
5 cycles
60
40
Extraction (%)
Recovery (%)
4 cycles 80
60
40
20
20
0
0 Phenanthrene
Pyrene
Compound
Benzo[a]pyrene
1
2
3
4
5
Extraction cycles
Fig. 4. (a) Recovery of PAHs and (b) percent extraction of colored fraction from 1 mg l1 PAHs-spiked solution by acetonitrile at different number of extraction cycle (n = 3). Percent extraction of colored fraction was estimated from the absorption values at 473 nm and 449 nm.
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
2007
Table 1 PAHs recovery, repeatability, intermediate precision and limit of detection (LOD) obtained during extraction of 5 PAHs-spiked solution (1.0 mg l1) with 2% KOH–methanol (five cycles extraction) and acetonitrile (three cycles extraction) PAHs
Naphthalene Phenanthrene Pyrene Benzo[a]pyrene Benzo[ghi]perylene
2% KOH–methanol
Acetonitrile
Recovery (%)
Repeatability (RSD%, n = 5)
Intermediate precision (RSD%)
LOD (lg l1)
Recovery (%)
Repeatability (RSD%, n = 5)
Intermediate precision (RSD%)
LOD (lg l1)
79.3 81.5 81.2 83.1 82.3
1.5 0.5 0.5 1.2 0.9
1.5 1.3 1.4 2.0 1.7
0.27 0.21 0.12 0.34 0.26
86.9 90.1 90.2 93.5 92.5
1.4 0.6 0.9 1.6 1.0
1.6 0.9 2.0 2.3 1.4
0.17 0.21 0.09 0.49 0.19
from 0.1 mg l1 PAHs-spiked solution by three cycles. Interestingly, although the extraction percentage also increased with the number of extractions, the values were quite lower than the methanol data. The percent extraction of colored fraction at 473 nm and 449 nm were only 10.6 ± 1.6% and 8.2 ± 1.5%, respectively, even after three cycles (Fig. 4b), whereas the values were more than 60% for methanol (Fig. 2b). Based on the positive results with pH-adjusted systems in methanol systems, we have also considered the use of HCl and KOH–acetonitrile extraction solvent. Contrary to our expectations, the extraction percentage of colored fraction in the presence of 1% HCl did not decrease much (7.3% and 7.5% at 473 nm and 449 nm, respectively) relative to that of acetonitrile alone even after three cycles. In addition, the PAHs recovery was also slightly lower (87.1 ± 1.0% for phenanthrene, 88.2 ± 0.7% for pyrene, and 90.7 ± 1.4% for benzo[a]pyrene). The same experiment could not be performed with KOH–acetonitrile since KOH crystals do not dissolve in acetonitrile. 3.3. Method validation parameters Table 1 shows the recovery, precision (repeatability and intermediate precision) and LOD obtained during extraction of 5 PAHs-spiked solution (1.0 mg l1) with 2% (w/v) KOH–methanol (five cycles) and acetonitrile (three cycles). The recoveries of 5 PAHs after treatment by 2% KOH–methanol and acetonitrile ranged from 79.3% to 83.1% and 86.9% to 93.5%, respectively. The recoveries of naphthalene in both cases were consistently lower than those of the other PAHs. This might reflect the greater susceptibility of naphthalene to evaporation losses due to its relatively higher vapor pressure (Sims and Overcash, 1983). The repeatability and intermediate precision of both methods showed good precision with RSD values between 0.5% and 2.3%. 4. Discussion In the present paper, we examined the liquid–liquid extraction technique for the removal of colored components from acetone extract of Komatsuna, which contained a mixture of various PAHs. Specifically, the development of a solvent system and appropriate extraction conditions
that maximize PAHs recovery while minimizing colored component co-extraction was mainly considered. Methanol and acetonitrile were employed as primary solvents due to their affinity with hexane. In the case of methanol, the data showed that although over 90% of PAHs was recovered, high extraction of colored fraction was also observed after three cycles (Fig. 2). Hence, to minimize percent extraction of colored fraction, the addition of HCl or KOH to the solvent was considered, taking into account the pH-dependency of colored fraction speciation. Both conditions resulted to significant reductions on the extraction of colored fraction (Fig. 3b and d). Though PAHs recovery also decreased, this occurred at a lesser degree (Fig. 3a and c). These results suggest that the degree of ionization of the matrix components may affect their extraction characteristics. However, it must be mentioned that the dissolution volume of hexane into the methanol phase during extraction also tended to decrease with increasing HCl or KOH concentrations to suggest its possible effect on the observed data as well. Now, comparing the effect of speciation and methanol–hexane dissolution on the colored fraction extraction, the latter may have greatly contributed to the declining values at highly acidic or basic conditions. This is due to the fact that at 2% and 5% acid or base concentrations, the speciation of colored components may already remain unchanged so that any further decrease in its removal may only be attributed to the decreasing dissolution of hexane in methanol. Such dissolution effect is further emphasized in the PAHs data, where decreasing recoveries was also observed at increasing acid or base concentration. In the case of PAHs, however, because its crystals easily dissolve in methanol than in hexane, such high PAHs partitioning in methanol may have contributed to its still high recovery of about 80% even at 2% acid or base concentration. On the basis of these results, the 2% KOH– methanol is an appropriate solvent system for the analysis of PAHs in plant when methanol is being considered as a major solvent. By this process, low colored fraction extraction of about 10% may be achieved (Fig. 3d). Acetonitrile also allowed for high PAHs recovery with low extraction of colored fraction after three cycles (Fig. 4). Though the clear reason for this result is still unknown, the low dissolution volume of hexane in the acetonitrile phase as well as acetonitrile’s low affinity for colored fraction and good selectivity for PAHs may be some
2008
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
of the foremost contributing factors. However, attempts to improve extraction conditions by pH adjustment did not bring about the same positive results as in the case of methanol. Thus, it will not be necessary to add acid or base compounds when acetonitrile is being considered as the major solvent. In conclusion, liquid–liquid extraction was found to be very effective in separating PAHs from carotene-like colored compounds in plant extracts. The use of either acetonitrile or 2% KOH–methanol solvents leads to highly accurate analysis data. For acetonitrile, three extraction cycles were sufficient to maximize PAHs recovery and minimize percentage extraction of colored fraction. For 2% KOH–methanol, five extraction cycles will be necessary. In addition, when this was applied to other plant shoots (i.e. zucchini, green soybean, maize and sweet potato), almost similar percent extraction of colored fraction was also obtained (data not shown), suggesting its possible wide applicability. However with the variability of plant chemical composition, depending on their species, age, environment, etc., extraction conditions must be established on a case by case basis before detailed PAHs analysis are finally made. Nevertheless, with its simple and cheap experimental design, we are confident that this method or its slightly modified version will ultimately find application in routine analysis of PAHs in plant components. Acknowledgements The authors would like to thank Dr. Takashi Otani and Dr. Nobuyasu Seike of the National Institute for AgroEnvironmental Sciences (NIAES) for valuable discussions about the separation of carotene-like compound from plant extract. References Berber, J.L., Thomas, G.O., Kerstiens, G., Jones, K.C., 2004. Current issues and uncertainties in the measurement and modeling of airvegetation exchange and within-plant processing of POPs. Environ. Pollut. 128, 99–138. Chen, G., White, P.A., 2004. The mutagenic hazards of aquatic sediments: a review. Mutat. Res. 567, 151–225. Fismes, J., Perrin-Ganier, C., Empereur-Bissonnet, P., Morel, J.L., 2002. Soil-to-root transfer and translocation of polycyclic aromatic hydrocarbons by vegetables grown on industrial contaminated soils. J. Environ. Qual. 31, 1649–1656. Gao, Y., Zhu, L., 2004. Plant uptake, accumulation and translocation of phenanthrene and pyrene in soils. Chemosphere 55, 1169–1178. Hels, O., Larsen, T., Christensen, L.P., Kidmose, U., Hassan, N., Thilsted, S.H., 2004. Contents of iron, calcium, zinc and b-carotene in commonly consumed vegetables in Bangladesh. J. Food Compos. Anal. 17, 587–595. Howsam, M., Jones, K.C., Ineson, P., 2001. PAHs associated with the leaves of three deciduous tree species II: uptake during a growing season. Chemosphere 44, 155–164. Hubert, A., Popp, P., Wenzel, K.-D., Engewald, W., Schu¨u¨rmann, G., 2003. One-step cleanup for PAH residue analysis in plant matrices using size-exclusion chromatography. Anal. Bioanal. Chem. 376, 53– 60.
Juhasz, A.L., Naidu, R., 2000. Bioremediation of high molecular weight polycyclic aromatic hydrocarbons: a review of the microbial degradation of benzo[a]pyrene. Int. Biodeter. Biodegr. 45, 57–88. Junk, G.A., Ford, C.S., 1980. A review of organic emissions from selected combustion processes. Chemosphere 9, 187–230. Kazerouni, N., Sinha, R., Hsu, C.-H., Greenberg, A., Rothman, N., 2001. Analysis of 200 food items for benzo[a]pyrene and estimation of its intake in an epidemiologic study. Food Chem. Toxicol. 39, 423–436. Ke, L., Wong, T.W.Y., Wong, A.H.Y., Wong, Y.S., Tam, N.F.Y., 2003. Negative effects of humic acid addition on phytoremediation of pyrene-contaminated sediments by mangrove seedlings. Chemosphere 52, 1581–1591. Kipopoulou, A.M., Manoli, E., Samara, C., 1999. Bioconcentration of polycyclic aromatic hydrocarbons in vegetables grown in an industrial area. Environ. Pollut. 106, 369–380. Kochubei, S.M., 1968. Interpretation of the UV absorption spectrum of bcarotene. J. Appl. Spectrosc. 8, 166–168. Lin, H., Tao, S., Zuo, Q., Coveney, R.M., 2007. Uptake of polycyclic aromatic hydrocarbons by maize plants. Environ. Pollut. 148, 614– 619. Liste, H.-H., Alexander, M., 2000. Plant-promoted pyrene degradation in soil. Chemosphere 40, 7–10. Maila, M.P., Cloete, T.E., 2002. Germination of Lepidium sativum as a method to evaluate polycyclic aromatic hydrocarbons (PAHs) removal from contaminated soil. Int. Biodeter. Biodegr. 50, 107–113. Mastral, A.M., Calle´n, M.S., Lo´pez, J.M., Murillo, R., Garcı´a, T., Navarro, M.V., 2003. Critical review on atmospheric PAH. Assessment of reported data in the Mediterranean basin. Fuel Process. Technol. 80, 183–193. Meudec, A., Dussauze, J., Jourdin, M., Deslandes, E., Poupart, N., 2006. Gas chromatographic–mass spectrometric method for polycyclic aromatic hydrocarbon analysis in plant biota. J. Chromatogr. A 1108, 240–247. Parrish, Z.D., White, J.C., Isleyen, M., Gent, M.P.N., Iannucci-Berger, W., Eitzer, B.D., Kelsey, J.W., Mattina, M.I., 2006. Accumulation of weathered polycyclic aromatic hydrocarbons (PAHs) by plant and earthworm species. Chemosphere 64, 609–618. Ratola, N., Lacorte, S., Alves, A., Barcelo´, D., 2006. Analysis of polycyclic aromatic hydrocarbons in pine needles by gas chromatography–mass spectrometry: comparison of different extraction and clean-up procedures. J. Chromatogr. A 1114, 198–204. Reilley, K.A., Banks, M.K., Schwab, A.P., 1996. Organic chemicals in the environment: dissipation of polycyclic aromatic hydrocarbons in the rhizosphere. J. Environ. Qual. 25, 212–219. Simonich, S.L., Hites, R.A., 1994. Importance of vegetation in removing polycyclic aromatic hydrocarbons from the atmosphere. Nature 370, 49–51. Sims, R.C., Overcash, M.R., 1983. Fate of polynuclear aromatic compounds (PNAs) in soil–plant systems. Residue Rev. 88, 1–68. Tao, S., Cui, Y.H., Xu, F.L., Li, B.G., Cao, J., Liu, W.X., Schmitt, G., Wang, X.J., Shen, W.R., Qing, B.P., Sun, R., 2004. Polycyclic aromatic hydrocarbons (PAHs) in agricultural soil and vegetables from Tianjin. Sci. Total Environ. 320, 11–24. Tremolada, P., Burnett, V., Calamari, D., Jones, K.C., 1996. Spatial distribution of PAHs in the U.K. atmosphere using pine needles. Environ. Sci. Technol. 30, 3570–3577. USEPA, 1985. Evaluation and Estimation of Potential Carcinogenic Risks of Polynuclear Aromatic Hydrocarbons. Office of research and development, US Environmental Protection Agency, Washington, DC. Wenzel, K.-D., Hubert, A., Manz, M., Weissflog, L., Engewald, W., Schu¨u¨rmann, G., 1998. Accelerated solvent extraction of semivolatile organic compounds from biomonitoring samples of pine needles and mosses. Anal. Chem. 70, 4827–4835. Wilcke, W., 2007. Global patterns of polycyclic aromatic hydrocarbons (PAHs) in soil. Geoderma 141, 157–166. Yang, Z., Zhu, L., 2007. Performance of the partition-limited model on predicting ryegrass uptake of polycyclic aromatic hydrocarbons. Chemosphere 67, 402–409.
Chemosphere 70 (2008) 2002–2008 www.elsevier.com/locate/chemosphere
Removal of carotene-like colored compounds by liquid–liquid extraction during polycyclic aromatic hydrocarbons analysis of plant tissue Takayuki Kobayashi, Kengo Morimoto, Kenji Tatsumi
*
National Institute of Advanced Industrial Science and Technology (AIST), 16-1, Onogawa, Tsukuba, Ibaraki 305-8506, Japan Received 6 June 2007; received in revised form 14 September 2007; accepted 14 September 2007 Available online 14 November 2007
Abstract Plants contain a wide variety of chemicals, some of which may have similar chromatographic behavior to polycyclic aromatic hydrocarbons (PAHs). During solid phase extraction (SPE) with Si-gel for instance, the co-elution of carotene-like colored compounds with PAHs has been observed. In this paper, liquid–liquid extraction was applied for the separation and subsequent analysis of PAHs from plant extracts. PAHs containing 2–6 rings, which include naphthalene, phenanthrene, pyrene, benzo[a]pyrene and benzo[ghi]perylene, were used as representative target chemicals. Carotene-like compounds extracted from Komatsuna (Brassica campestris) shoot by acetone followed by Si-gel treatment were incorporated as undesired components in the model matrix. Results showed the feasibility of employing either acetonitrile or 2% (w/v) KOH–methanol as solvents for high PAHs recovery and low extraction of colored fraction. For acetonitrile, 86.9–93.5% of each PAH could be recovered after three extraction cycles (relative standard deviation, RSD < 1.6%) with only about 10% co-extraction of colored fraction. For 2% KOH–methanol, PAHs recoveries ranging from 79.3% to 83.1% after five cycles (RSD < 1.5%) were achieved while the percent extraction of colored fraction was also low at 10%. The relatively higher selectivity of the solvents for PAHs over the colored fraction as well as the solubility of the matrix solution in the solvent may have contributed to these results. On this basis, liquid–liquid extraction is very useful for the pre-treatment of plant extracts for PAHs analysis. 2007 Elsevier Ltd. All rights reserved. Keywords: Clean-up; Phytoremediation; Sample pre-treatment; Solid phase extraction (SPE)
1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are one of the most important classes of toxic pollutants that are ubiquitously found in the environment (Juhasz and Naidu, 2000; Mastral et al., 2003; Chen and White, 2004). Their occurrence is mainly attributed to a variety of anthropogenic activities such as incomplete combustion of fossil-fuels (Junk and Ford, 1980; Wilcke, 2007). Several PAHs, such as fluoranthene and benzo[a]pyrene, which are likely carcinogens, are listed as priority pollutants by United States of Environmental Protection Agency (USEPA, 1985). In *
Corresponding author. Tel.: +81 29 861 8325; fax: +81 29 861 8242. E-mail address: [email protected] (K. Tatsumi).
0045-6535/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2007.09.027
recent years, much attention has been paid to the exposure of plants to PAHs via atmospheric deposition and adsorption from the soil (Simonich and Hites, 1994; Kipopoulou et al., 1999; Howsam et al., 2001; Kazerouni et al., 2001; Berber et al., 2004). On the other hand, there are also numerous studies regarding the application of phytoremediation to treat PAHs contaminated soil using a variety of plant species (Liste and Alexander, 2000; Fismes et al., 2002; Maila and Cloete, 2002; Ke et al., 2003; Gao and Zhu, 2004; Parrish et al., 2006). In all of these studies, accurate and efficient analysis of PAHs content of plant tissues is very important. Plants contain a wide variety of chemicals (e.g. pigments, fatty acids and alcohols) and some of these may have very similar chromatographic behavior to PAHs.
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
The direct injection of untreated plant extracts into the GC/MS may not be appropriate for instrument safety and accuracy of PAHs analysis. Clean-up of plant extracts is therefore very important. Solid phase extraction (SPE) method with silica (Tao et al., 2004; Lin et al., 2007; Yang and Zhu, 2007), florisil (Kazerouni et al., 2001), alumina (Ratola et al., 2006) and C18 (Reilley et al., 1996) has been used for sample purification. However, during SPE with silica gel as stationary phase and dichloromethane as eluent, the co-elution of a yellow-colored fraction, which has a similar visible spectrum pattern with b-carotene, has been quite persistent. In addition, carotene, sterol and alcohol are also co-eluted with PAHs during SPE using florisil (Meudec et al., 2006) to indicate the general difficulty of separating certain compounds. In order to improve sample pre-treatment, a combination of the above-mentioned stationary phases (Tremolada et al., 1996; Wenzel et al., 1998) has been employed during SPE. Size exclusion chromatography (Hubert et al., 2003) has also been considered. However, these methods prolong and thus generally complicate the overall analysis. In the present paper, liquid–liquid extraction was evaluated as a simple alternative to remove colored components from PAHs-containing samples. Specifically, the optimization of extraction conditions (i.e. extraction solvent and number of extraction cycle) for separating 2–6 rings PAHs from contaminating compounds was considered. An extract from the shoot of Komatsuna (Brassica campestris), which is known to contain high amounts of carotene-like chemicals (Hels et al., 2004), was used as model matrix to simulate the yellow-colored fraction commonly encountered during purification of plant extract. 2. Materials and methods 2.1. Reagents and plant material Five types of powdered PAHs (purity >98%) were used in the present study: naphthalene and phenanthrene from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan); pyrene from Nacalai Tesque, Inc. (Kyoto, Japan); benzo[a]pyrene and benzo[ghi]perylene from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). All other reagents were of special grade obtained from Nacalai Tesque or Kanto Chemical Co., Inc. (Tokyo, Japan). The shoot of Komatsuna (B. campestris) was used for the preparation of matrix solution. The vegetable was purchased at a local marketplace, washed with tap water and then stored in the dark at 20 C until use. The 5 PAHs contents of the plant sample were below the detection limit. 2.2. Preparation of PAHs-spiked solution Komatsuna shoots were cut into small pieces and homogenized for 3 min with acetone in an Erlenmeyer flask. The mixture was ultrasonicated for 30 min and filtered through Whatman GF/F glass filter (Whatman Japan
2003
KK., Tokyo). The filtrate was transferred in a separatory funnel and then hexane and distilled water were added. The funnel was shaken and then allowed to stand for a few minutes after which the hexane phase including the remaining stable emulsion was obtained. The hexane extract was washed twice with distilled water, and the excess water was removed over anhydrous Na2SO4. The extract was concentrated in a rotary evaporator at 30 C. To remove chlorophyll and other interfering components, the concentrated solution was deposited onto a silica gel column (Wakogel C300, Wako pure chemical Industries; 12 cm length by 3.2 cm diameter). The column was eluted with hexane (60 ml), and then with dichloromethane. A yellow-colored fraction was eluted from top of the column as soon as the solvent was changed to dichloromethane. The solvent was loaded until the eluent became almost colorless. The hexane and dichloromethane eluents were combined and the resulting solution was evaporated in a rotary evaporator at 30 C to remove the dichloromethane. The concentrated yellow-colored solution was used as matrix solution. It was spiked with the stock solutions of the 3 PAHs (3-ring phenanthrene, 4-ring pyrene and 5-ring benzo[a]pyrene; 500 mg l1 for each PAH in acetone) to come up with concentrations of 0.1 and 1.0 mg l1 for each PAH. The PAHs-spiked solutions were stored in the dark at 20 C until use. 2.3. Optimization for PAHs determination by liquid–liquid extraction Methanol, HCl–methanol, KOH–methanol and acetonitrile were used as solvents for liquid–liquid extraction and the number of extractions was varied from one to five cycles. The concentrations of HCl (v/v) and KOH (w/v) were 1–5%. The volume of extraction solvent was 1.5 ml per one cycle except for acetonitrile, where 3 ml per one cycle was employed. The detailed procedure was as follows: 5 ml of PAHs-spiked solution and the specified volume of extraction solvent were combined in a glass test tube and then subjected to vortex mixing for 30 s. The lower phase was collected and the upper hexane phase was re-extracted with an equal volume of extraction solvent. The operation was repeated up to the specified cycles. In the case of methanol as solvent, 3 ml of hexane was added into the tube before the first extraction. The initial addition of hexane in this case was made to compensate for its high affinity for methanol as compared with the other solvents. This would assure that even after five cycles of extraction, an upper hexane phase would remain. The remaining hexane phase was stored for estimation of percent extraction of colored fraction as described in Section 2.5. All the extracts collected in each cycle were combined in a 100 ml screw glass vial and then 70 ml of 1% (w/v) aqueous NaCl and 10 ml of hexane were added. The vial was shaken and the upper hexane layer was obtained. The hexane extract was washed with distilled water and passed through anhydrous Na2SO4 column. The solution was concentrated to a final
2004
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
volume of approximately 5 ml under a gentle stream of nitrogen inside a Zymark-TurboVap LV evaporator (Caliper Life Sciences, Hopkinton, MA) that was operated with an ice water bath at 4–6 C. After the addition of 100 ll of syringe spike (50 mg l1 anthracene solution in acetone), the solution volume was made up to 10 ml with hexane in a volumetric flask. This solution was used as sample for GC/MS and UV–Vis analysis. All experiments were conducted at ambient temperature (20–25 C) and in triplicates. 2.4. GC/MS analysis and determination of PAHs The determination of PAHs concentration was carried out using a Shimadzu GCMS-QP2010 system (Shimadzu Co., Kyoto, Japan) equipped with AOC-20i autosampler and Quadrex 007-1 capillary column made of 100% dimethylpolysiloxane (0.25 mm ID · 25 m · 0.25 lm film thickness; Quadrex Co., New Haven, CT). Helium (purity >99.9999%) was used as the carrier gas at a constant pressure (100 kPa). Samples (1 ll) were injected in splitless mode. The temperature program for the GC oven was as follows: initial temperature 50 C for 1.5 min, heating to 180 C at 18 C min1 (during analysis with 5 PAHs including naphthalene and benzo[ghi]perylene, the heating rate was reduced to 8 C min1), then to 300 C at 12 C min1 with a final holding time of 10 min. The injector and interface temperatures were both 300 C. Mass spectrometry was performed using the electron ionization (EI) mode and ion source temperature was 200 C. Acquisition was carried out in the selective ion monitoring (SIM) mode, using the following characteristic masses: naphthalene, 128; phenanthrene and anthracene (syringe spike), 178; pyrene, 202; benzo[a]pyrene, 252, and benzo[ghi]perylene, 276. The amount of PAHs in the sample solution was determined, by the absolute calibration method, from the peak area of each PAH and that of anthracene on GC/MS chromatogram. To obtain the calibration curve, hexane solutions with various concentrations of PAHs were prepared (0.1–0.8 mg l1). Percent recoveries of PAHs were calculated from the amounts of PAHs in the sample solution and those in the initially spiked solution. 2.5. UV–Vis analysis and estimation of percent extraction of colored fraction UV–Vis spectra were measured with Jasco V-550 spectrophotometer (Jasco Co., Tokyo, Japan). The remaining hexane phase, which was stored for the estimation of percent extraction of colored fraction as described in Section 2.3 was evaporated to dryness with a rotary evaporator at 40 C and then reconstituted with 5 ml of hexane. Absorption values of the solution at 473 nm and 449 nm were measured. The percent extraction of colored fraction in this case was calculated using the following equation:
Percent extraction ð%Þ ¼
1
Ek sample soln: Ek initially spiked soln:
100
where Ek is the absorbance at k (nm) (i.e. 473 nm and 449 nm: the colored fraction displayed absorption maxima at these wavelengths in hexane) of the sample or initially spiked solution before extraction. No absorption was detected at 473 nm and 449 nm in the hexane solution containing the 5 PAHs. All experiments were performed in triplicates. In the present paper, though quantitatively similar, the terms ‘‘percent recovery’’ and ‘‘percent extraction’’ were used in the case of PAHs and colored fraction, respectively, to distinguish between the target compounds (PAHs) and the impurities being removed (colored fraction) from the sample solution for GC/MS analysis. 2.6. Method validation This experiment was conducted using the best condition that was established from the initial data with the 3 PAHs. To further evaluate its applicability to a wider range of PAHs compounds, 2-ring naphthalene and 6-ring benzo[ghi]perylene were additionally incorporated during this stage. The precision of the method, which is expressed as relative standard deviation (RSD), was evaluated in terms of repeatability and intermediate precision. Repeatability was assessed by performing five replicates on the same day. In the case of intermediate precision, three replicates were analyzed on three different days by the same analyst. The limits of detection (LOD) were calculated using a signal to noise ratio of 3 (S/N = 3). 3. Results 3.1. Effect of methanol solvent system on recovery of PAHs and percent extraction of colored fraction The UV–Vis spectra of the initial solution that was spiked with 3 PAHs and that of the sample solution for GC/MS analysis obtained after extracting with methanol, 2% KOH–methanol and acetonitrile are shown in Fig. 1. All samples were subjected to the same dilutions and number of extraction cycles of five. The spectrum of initial solution showed absorption maxima at 473 nm and 449 nm. The observed wavelengths for maximum absorption as well as the general shape of the spectrum are characteristic pattern of b-carotene (Kochubei, 1968). The same absorption maxima were also detected in the spectra of the sample solutions that were extracted with either 2% KOH–methanol or acetonitrile (Fig. 1). However, their absorption values at 473 nm and 449 nm were lower at around 0.04 for both wavelengths. With these very low absorbances, which are difficult to measure accurately, we therefore defined the percent extraction of colored fraction based on the equation described in Section 2.5. The detailed data for the percent extractions of colored fraction will be discussed in a later section.
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008 0.6 Initially spiked solution Methanol 2%KOH-methanol Acetonitrile
0.5
Absorbance
449
0.4
473
0.3
0.2
0.1
0.0 350
400
450
500
550
600
Wavelength (nm) Fig. 1. UV–Vis spectra of the initially spiked solution (1.0 mg l1 3 PAHs) and sample solution for GC/MS analysis obtained after extracting with methanol, 2% KOH–methanol and acetonitrile (solvent; hexane). All samples have the same dilution ratio and number of extraction cycle is five.
Fig. 2 shows the effect of the number of extraction cycle on the percent recovery of PAHs and the percent extraction of colored fraction from the 1.0 mg l1 PAHs-spiked solution by methanol. The PAHs recovery tended to increase with the number of extractions, so that over 90% of each PAH could be recovered after three cycles (93.5 ± 1.2% for phenanthrene, 99.5 ± 0.3% for pyrene, and 106.8 ± 1.6% for benzo[a]pyrene). For the colored fraction, on the other hand, the percent extractions, which were calculated from the absorbance at either 473 nm or 449 nm, were almost the same and similar results were found in the other solvent systems. Parallel with the PAHs data, the percent extraction of colored fraction also increased with the number of extractions up to the third cycle, after which, the values remained more or less constant (Fig. 2b). The percent extractions of colored fraction after five cycles were 68.4 ± 2.5% and 70.0 ± 2.0% at 473 nm
2005
and 449 nm, respectively. These values were almost consistent with the percentage intensities of methanol solution at 473 nm and 449 nm wavelengths relative to the initially spiked solution (Fig. 1). This indicates that the equation described in Section 2.5 was valid for the calculation of extraction percentage of colored fraction. On the basis of the above results, we further evaluated the extraction of colored fraction with the addition HCl or KOH since plant matrices contain chemicals with various functional groups (e.g. hydroxyl and carboxyl group) whose speciation are pH-dependent. The data for methanol at different acid and base concentrations are shown in Fig. 3. Although recoveries of PAHs decreased with increasing HCl or KOH concentration, over 80% of each PAH could still be recovered at 2% acid or base even after five cycles (Fig. 3a and c). Comparing the PAHs recovery data at the same concentration, no clear differences were found between HCl or KOH addition except that at 5%, relatively higher recoveries were achieved in the former (Fig. 3a and c). On the other hand, the percent extraction of colored fraction with KOH was significantly lower than with HCl at the same concentration. At 2% KOH, the percent extraction at 473 nm and 449 nm were only 9.9 ± 1.7% and 10.7 ± 1.5%, respectively (Fig. 3d), whereas values for 2% HCl were both close to 30% (Fig. 3b). 3.2. Effect of acetonitrile extraction system on recovery of PAHs and percent extraction of colored fraction Fig. 4 shows the effect of the number of extraction cycle on the percent recovery of PAHs and percent extraction of colored fraction from the PAHs-spiked solution (1.0 mg l1) by acetonitrile. The recoveries tended to increase with the number of extractions as in the case of methanol. After three cycles, over 90% for each of the 3 PAHs was recovered (90.0 ± 1.2% for phenanthrene, 92.6 ± 0.7% for pyrene, and 99.6 ± 0.5% for benzo[a]pyrene). Almost similar PAHs recoveries were also obtained
(a) PAHs
(b) Colored fraction
120
100 1 cycle 80
3 cycles 4 cycles
80
5 cycles
60
40
Extraction (%)
Recovery (%)
473 nm 449 nm
2 cycles
100
60
40
20
20
0
0 Phenanthrene
Pyrene
Compound
Benzo[a]pyrene
1
2
3
4
5
Extraction cycles
Fig. 2. (a) Recovery of PAHs and (b) percent extraction of colored fraction from spiked solution (1 mg l1 3 PAHs) by methanol at different number of extraction cycle (n = 3). Percent extraction of colored fraction was estimated from the absorption values at 473 nm and 449 nm.
2006
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
(b) Colored fraction (HCl-methanol)
(a) PAHs (HCl-methanol) 120
100 0%
473 nm
1%
100
60
40
Extraction (%)
Recovery (%)
5%
80
449 nm
80
2%
60
40
20
20
0
0 Phenanthrene
Pyrene
Benzo[a]pyrene
0
Compound
1
2
5
HCl concentration (%, v/v) (d) Colored fraction (KOH-methanol)
(c) PAHs (KOH-methanol) 120
100 0%
473 nm
1%
100
5%
80
Extraction (%)
Recovery (%)
449 nm
80
2%
60
40
60
40
20
20
0 Phenanthrene
Pyrene
0
Benzo[a]pyrene
Compound
0
1
2
5
KOH concentration (%, w/v)
Fig. 3. PAHs recovery and percent extraction of colored fraction from 1 mg l1 PAHs-spiked solution by HCl–methanol (a and b, respectively) and KOH–methanol (c and d, respectively) at different acid or base concentration (n = 3). Percent extraction of colored fraction was estimated from the absorption values at 473 nm and 449 nm. Number of extraction cycle is five.
(b) Colored fraction
(a) PAHs 120
100 1 cycle
473 nm
2 cycles
100
449 nm
80
3 cycles
5 cycles
60
40
Extraction (%)
Recovery (%)
4 cycles 80
60
40
20
20
0
0 Phenanthrene
Pyrene
Compound
Benzo[a]pyrene
1
2
3
4
5
Extraction cycles
Fig. 4. (a) Recovery of PAHs and (b) percent extraction of colored fraction from 1 mg l1 PAHs-spiked solution by acetonitrile at different number of extraction cycle (n = 3). Percent extraction of colored fraction was estimated from the absorption values at 473 nm and 449 nm.
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
2007
Table 1 PAHs recovery, repeatability, intermediate precision and limit of detection (LOD) obtained during extraction of 5 PAHs-spiked solution (1.0 mg l1) with 2% KOH–methanol (five cycles extraction) and acetonitrile (three cycles extraction) PAHs
Naphthalene Phenanthrene Pyrene Benzo[a]pyrene Benzo[ghi]perylene
2% KOH–methanol
Acetonitrile
Recovery (%)
Repeatability (RSD%, n = 5)
Intermediate precision (RSD%)
LOD (lg l1)
Recovery (%)
Repeatability (RSD%, n = 5)
Intermediate precision (RSD%)
LOD (lg l1)
79.3 81.5 81.2 83.1 82.3
1.5 0.5 0.5 1.2 0.9
1.5 1.3 1.4 2.0 1.7
0.27 0.21 0.12 0.34 0.26
86.9 90.1 90.2 93.5 92.5
1.4 0.6 0.9 1.6 1.0
1.6 0.9 2.0 2.3 1.4
0.17 0.21 0.09 0.49 0.19
from 0.1 mg l1 PAHs-spiked solution by three cycles. Interestingly, although the extraction percentage also increased with the number of extractions, the values were quite lower than the methanol data. The percent extraction of colored fraction at 473 nm and 449 nm were only 10.6 ± 1.6% and 8.2 ± 1.5%, respectively, even after three cycles (Fig. 4b), whereas the values were more than 60% for methanol (Fig. 2b). Based on the positive results with pH-adjusted systems in methanol systems, we have also considered the use of HCl and KOH–acetonitrile extraction solvent. Contrary to our expectations, the extraction percentage of colored fraction in the presence of 1% HCl did not decrease much (7.3% and 7.5% at 473 nm and 449 nm, respectively) relative to that of acetonitrile alone even after three cycles. In addition, the PAHs recovery was also slightly lower (87.1 ± 1.0% for phenanthrene, 88.2 ± 0.7% for pyrene, and 90.7 ± 1.4% for benzo[a]pyrene). The same experiment could not be performed with KOH–acetonitrile since KOH crystals do not dissolve in acetonitrile. 3.3. Method validation parameters Table 1 shows the recovery, precision (repeatability and intermediate precision) and LOD obtained during extraction of 5 PAHs-spiked solution (1.0 mg l1) with 2% (w/v) KOH–methanol (five cycles) and acetonitrile (three cycles). The recoveries of 5 PAHs after treatment by 2% KOH–methanol and acetonitrile ranged from 79.3% to 83.1% and 86.9% to 93.5%, respectively. The recoveries of naphthalene in both cases were consistently lower than those of the other PAHs. This might reflect the greater susceptibility of naphthalene to evaporation losses due to its relatively higher vapor pressure (Sims and Overcash, 1983). The repeatability and intermediate precision of both methods showed good precision with RSD values between 0.5% and 2.3%. 4. Discussion In the present paper, we examined the liquid–liquid extraction technique for the removal of colored components from acetone extract of Komatsuna, which contained a mixture of various PAHs. Specifically, the development of a solvent system and appropriate extraction conditions
that maximize PAHs recovery while minimizing colored component co-extraction was mainly considered. Methanol and acetonitrile were employed as primary solvents due to their affinity with hexane. In the case of methanol, the data showed that although over 90% of PAHs was recovered, high extraction of colored fraction was also observed after three cycles (Fig. 2). Hence, to minimize percent extraction of colored fraction, the addition of HCl or KOH to the solvent was considered, taking into account the pH-dependency of colored fraction speciation. Both conditions resulted to significant reductions on the extraction of colored fraction (Fig. 3b and d). Though PAHs recovery also decreased, this occurred at a lesser degree (Fig. 3a and c). These results suggest that the degree of ionization of the matrix components may affect their extraction characteristics. However, it must be mentioned that the dissolution volume of hexane into the methanol phase during extraction also tended to decrease with increasing HCl or KOH concentrations to suggest its possible effect on the observed data as well. Now, comparing the effect of speciation and methanol–hexane dissolution on the colored fraction extraction, the latter may have greatly contributed to the declining values at highly acidic or basic conditions. This is due to the fact that at 2% and 5% acid or base concentrations, the speciation of colored components may already remain unchanged so that any further decrease in its removal may only be attributed to the decreasing dissolution of hexane in methanol. Such dissolution effect is further emphasized in the PAHs data, where decreasing recoveries was also observed at increasing acid or base concentration. In the case of PAHs, however, because its crystals easily dissolve in methanol than in hexane, such high PAHs partitioning in methanol may have contributed to its still high recovery of about 80% even at 2% acid or base concentration. On the basis of these results, the 2% KOH– methanol is an appropriate solvent system for the analysis of PAHs in plant when methanol is being considered as a major solvent. By this process, low colored fraction extraction of about 10% may be achieved (Fig. 3d). Acetonitrile also allowed for high PAHs recovery with low extraction of colored fraction after three cycles (Fig. 4). Though the clear reason for this result is still unknown, the low dissolution volume of hexane in the acetonitrile phase as well as acetonitrile’s low affinity for colored fraction and good selectivity for PAHs may be some
2008
T. Kobayashi et al. / Chemosphere 70 (2008) 2002–2008
of the foremost contributing factors. However, attempts to improve extraction conditions by pH adjustment did not bring about the same positive results as in the case of methanol. Thus, it will not be necessary to add acid or base compounds when acetonitrile is being considered as the major solvent. In conclusion, liquid–liquid extraction was found to be very effective in separating PAHs from carotene-like colored compounds in plant extracts. The use of either acetonitrile or 2% KOH–methanol solvents leads to highly accurate analysis data. For acetonitrile, three extraction cycles were sufficient to maximize PAHs recovery and minimize percentage extraction of colored fraction. For 2% KOH–methanol, five extraction cycles will be necessary. In addition, when this was applied to other plant shoots (i.e. zucchini, green soybean, maize and sweet potato), almost similar percent extraction of colored fraction was also obtained (data not shown), suggesting its possible wide applicability. However with the variability of plant chemical composition, depending on their species, age, environment, etc., extraction conditions must be established on a case by case basis before detailed PAHs analysis are finally made. Nevertheless, with its simple and cheap experimental design, we are confident that this method or its slightly modified version will ultimately find application in routine analysis of PAHs in plant components. Acknowledgements The authors would like to thank Dr. Takashi Otani and Dr. Nobuyasu Seike of the National Institute for AgroEnvironmental Sciences (NIAES) for valuable discussions about the separation of carotene-like compound from plant extract. References Berber, J.L., Thomas, G.O., Kerstiens, G., Jones, K.C., 2004. Current issues and uncertainties in the measurement and modeling of airvegetation exchange and within-plant processing of POPs. Environ. Pollut. 128, 99–138. Chen, G., White, P.A., 2004. The mutagenic hazards of aquatic sediments: a review. Mutat. Res. 567, 151–225. Fismes, J., Perrin-Ganier, C., Empereur-Bissonnet, P., Morel, J.L., 2002. Soil-to-root transfer and translocation of polycyclic aromatic hydrocarbons by vegetables grown on industrial contaminated soils. J. Environ. Qual. 31, 1649–1656. Gao, Y., Zhu, L., 2004. Plant uptake, accumulation and translocation of phenanthrene and pyrene in soils. Chemosphere 55, 1169–1178. Hels, O., Larsen, T., Christensen, L.P., Kidmose, U., Hassan, N., Thilsted, S.H., 2004. Contents of iron, calcium, zinc and b-carotene in commonly consumed vegetables in Bangladesh. J. Food Compos. Anal. 17, 587–595. Howsam, M., Jones, K.C., Ineson, P., 2001. PAHs associated with the leaves of three deciduous tree species II: uptake during a growing season. Chemosphere 44, 155–164. Hubert, A., Popp, P., Wenzel, K.-D., Engewald, W., Schu¨u¨rmann, G., 2003. One-step cleanup for PAH residue analysis in plant matrices using size-exclusion chromatography. Anal. Bioanal. Chem. 376, 53– 60.
Juhasz, A.L., Naidu, R., 2000. Bioremediation of high molecular weight polycyclic aromatic hydrocarbons: a review of the microbial degradation of benzo[a]pyrene. Int. Biodeter. Biodegr. 45, 57–88. Junk, G.A., Ford, C.S., 1980. A review of organic emissions from selected combustion processes. Chemosphere 9, 187–230. Kazerouni, N., Sinha, R., Hsu, C.-H., Greenberg, A., Rothman, N., 2001. Analysis of 200 food items for benzo[a]pyrene and estimation of its intake in an epidemiologic study. Food Chem. Toxicol. 39, 423–436. Ke, L., Wong, T.W.Y., Wong, A.H.Y., Wong, Y.S., Tam, N.F.Y., 2003. Negative effects of humic acid addition on phytoremediation of pyrene-contaminated sediments by mangrove seedlings. Chemosphere 52, 1581–1591. Kipopoulou, A.M., Manoli, E., Samara, C., 1999. Bioconcentration of polycyclic aromatic hydrocarbons in vegetables grown in an industrial area. Environ. Pollut. 106, 369–380. Kochubei, S.M., 1968. Interpretation of the UV absorption spectrum of bcarotene. J. Appl. Spectrosc. 8, 166–168. Lin, H., Tao, S., Zuo, Q., Coveney, R.M., 2007. Uptake of polycyclic aromatic hydrocarbons by maize plants. Environ. Pollut. 148, 614– 619. Liste, H.-H., Alexander, M., 2000. Plant-promoted pyrene degradation in soil. Chemosphere 40, 7–10. Maila, M.P., Cloete, T.E., 2002. Germination of Lepidium sativum as a method to evaluate polycyclic aromatic hydrocarbons (PAHs) removal from contaminated soil. Int. Biodeter. Biodegr. 50, 107–113. Mastral, A.M., Calle´n, M.S., Lo´pez, J.M., Murillo, R., Garcı´a, T., Navarro, M.V., 2003. Critical review on atmospheric PAH. Assessment of reported data in the Mediterranean basin. Fuel Process. Technol. 80, 183–193. Meudec, A., Dussauze, J., Jourdin, M., Deslandes, E., Poupart, N., 2006. Gas chromatographic–mass spectrometric method for polycyclic aromatic hydrocarbon analysis in plant biota. J. Chromatogr. A 1108, 240–247. Parrish, Z.D., White, J.C., Isleyen, M., Gent, M.P.N., Iannucci-Berger, W., Eitzer, B.D., Kelsey, J.W., Mattina, M.I., 2006. Accumulation of weathered polycyclic aromatic hydrocarbons (PAHs) by plant and earthworm species. Chemosphere 64, 609–618. Ratola, N., Lacorte, S., Alves, A., Barcelo´, D., 2006. Analysis of polycyclic aromatic hydrocarbons in pine needles by gas chromatography–mass spectrometry: comparison of different extraction and clean-up procedures. J. Chromatogr. A 1114, 198–204. Reilley, K.A., Banks, M.K., Schwab, A.P., 1996. Organic chemicals in the environment: dissipation of polycyclic aromatic hydrocarbons in the rhizosphere. J. Environ. Qual. 25, 212–219. Simonich, S.L., Hites, R.A., 1994. Importance of vegetation in removing polycyclic aromatic hydrocarbons from the atmosphere. Nature 370, 49–51. Sims, R.C., Overcash, M.R., 1983. Fate of polynuclear aromatic compounds (PNAs) in soil–plant systems. Residue Rev. 88, 1–68. Tao, S., Cui, Y.H., Xu, F.L., Li, B.G., Cao, J., Liu, W.X., Schmitt, G., Wang, X.J., Shen, W.R., Qing, B.P., Sun, R., 2004. Polycyclic aromatic hydrocarbons (PAHs) in agricultural soil and vegetables from Tianjin. Sci. Total Environ. 320, 11–24. Tremolada, P., Burnett, V., Calamari, D., Jones, K.C., 1996. Spatial distribution of PAHs in the U.K. atmosphere using pine needles. Environ. Sci. Technol. 30, 3570–3577. USEPA, 1985. Evaluation and Estimation of Potential Carcinogenic Risks of Polynuclear Aromatic Hydrocarbons. Office of research and development, US Environmental Protection Agency, Washington, DC. Wenzel, K.-D., Hubert, A., Manz, M., Weissflog, L., Engewald, W., Schu¨u¨rmann, G., 1998. Accelerated solvent extraction of semivolatile organic compounds from biomonitoring samples of pine needles and mosses. Anal. Chem. 70, 4827–4835. Wilcke, W., 2007. Global patterns of polycyclic aromatic hydrocarbons (PAHs) in soil. Geoderma 141, 157–166. Yang, Z., Zhu, L., 2007. Performance of the partition-limited model on predicting ryegrass uptake of polycyclic aromatic hydrocarbons. Chemosphere 67, 402–409.