Accelerated Solvent Extraction of Semivolatile Organic Compounds

0 downloads 0 Views 138KB Size Report
Organic Compounds from Biomonitoring Samples of Pine Needles and Mosses. Klaus-Dieter Wenzel,*,† Andreas Hubert,‡ Michael Manz,† Ludwig Weissflog,† ...
Anal. Chem. 1998, 70, 4827-4835

Accelerated Solvent Extraction of Semivolatile Organic Compounds from Biomonitoring Samples of Pine Needles and Mosses Klaus-Dieter Wenzel,*,† Andreas Hubert,‡ Michael Manz,† Ludwig Weissflog,† Werner Engewald,‡ and Gerrit Schu 1u 1 rmann†

Department of Chemical Ecotoxicology, UFZ Centre for Environmental Research, Permoserstrasse 15, D-04318 Leipzig, Germany, and Faculty for Chemistry and Mineralogy, Institute for Analytical Chemistry, University of Leipzig, Linnestrasse 3, D-04103 Leipzig, Germany

Accelerated solvent extraction (ASE) was used for the simultaneous extraction of semivolatile organic compounds (SOCs) including chlorobenzenes (1,2,3,4-tetra-, penta-, hexachlorobenzene), HCH isomers (r-, β-, γ-, δ-, E-HCH), DDX (p,p′-DDT, -DDE, -DDD), PCB congeners (28, 52, 101, 138, 153, 180), and PAHs (phenanthrene, anthracene, fluoranthene, pyrene, benzo[a]pyrene) from mosses (Pleurozium schreberi) growing in regional locations (central Germany) and pine needles (Pinus sylvestris L.) from southern Russia (near the Caspian Sea). The results were compared with those obtained by ultrasonic extraction (USE). Using mixed moss samples (thorough cleanup, only minor interference during GC/ MSD analysis) from one location mainly served to optimize two parameters, extraction solvent and temperature. The most favorable extraction conditions proved to be nhexane as the extraction solvent, two temperature stages of 40 and 120 °C, a pressure of 15 MPa, and three static cycles in each case. These conditions were then applied to the extraction of SOCs from the wax and the inner pine needle fraction, which beforehand had undergone extractive separation with dichloromethane. ASE was found to be especially advantageous in the case of higher multiple exposures to pollutants and the resultant complicated matrixes (oleiferous extracts, pollutant/matrix conjugates). Owing to the much better separation of analyte/ matrix, in many cases increases of 1-2 orders of magnitude were obtained in the analytical values of the contaminated sample materials compared to those obtained with USE. Over 130 anthropogenic organic substances which can have biological effects have been detected in European forestlands. Moreover, there is also an abundance of other organic substances accumulating in biomonitoring objects such as mosses and pine needles. These chemicals can significantly contribute to the multiple pollution of a location. The complicated plant matrix with large fractions of structures abundant in lipids, its diversity among † ‡

UFZ Centre for Environmental Research. University of Leipzig.

10.1021/ac9806299 CCC: $15.00 Published on Web 10/17/1998

© 1998 American Chemical Society

different and even the same plant species from one location to another, and the effects of these and numerous other factors1 make the extraction, cleanup, and analysis of lipophilic semivolatile organic compounds (SOCs)2 a highly challenging process. Analytical difficulties in the analysis of pine needles in particular are due to multiple exposures to pollutants, depending on specific locations. Methodological improvements are making it possible to better understand the processes by which anthropogenic pollutants such as SOCs can penetrate and accumulate in plants, as well as the way in which they can also cause damage in the inner plant compartments.3 In view of the known carcinogenic4 and teratogenic5 effects, as well as the endocrinic effects6,7 recently mooted in particular for organochlorines, an improved quantitative determination of persistent organochlorines and polycyclic aromatic hydrocarbons (PAHs) in biological compartments is essential if the hazards of these compounds are to be better assessed. Over the past few years, various extraction methods have been developed to improve automation, shorten extraction times, and reduce the amount of solvent required. One of these techniques is accelerated solvent extraction (ASE), which has been used internationally for about 3 years. Extractions can be carried out at temperatures ranging from room temperature (very gentle conditions) up to 200 °C in order to accelerate extraction, and at pressures in the range of 5-200 atm (0.3-20 MPa) in order to maintain the extraction solvent in a liquid state. The extraction of SOCs from environmental sample material has so far been performed using ultrasonic extraction (USE),8,9 (1) Simonich, S. L.; Hites, R. A. Environ. Sci. Technol. 1995, 29, 2905-2914. (2) Kylin, H.; Nordstrand, E.; Sjo¨din, A.; Jensen, S. Fresenius’ J. Anal. Chem. 1996, 356, 62-69. (3) Schreiber, L.; Scho¨nherr, J. Environ. Sci. Technol. 1992, 26, 153-159. (4) McKee, R. H.; Daughtrey, W. C.; Freeman, J. J.; Federici, J. J.; Phillips, R. D.; Plutnick, R. T. J. Appl. Toxicol. 1989, 9, 270-275. (5) Olsson, M.; Bergman, A.; Jensen, S.; Kihlstro¨m, J. E. In Ecoinforma Press; Hutzinger, O., Fiedler, H., Eds.; Bayreuth, 1990; pp 393-396. (6) Greim, H. Nachr. Chem. Lab. 1998, 46, 63-66. (7) Wolff, M. S.; Toniolo, P. G.; Lee, E. W.; Rivera, M.; Dubin, N. J. Natl. Cancer Inst. 1993, 85, 648. (8) Heemken, O. P.; Theobald, N.; Wenzlawiak, B. W. Anal. Chem. 1997, 69, 2171-2180. (9) Burford, M. D.; Hawthorne, S. B.; Miller, D. J. Anal. Chem. 1993, 65, 14971505.

Analytical Chemistry, Vol. 70, No. 22, November 15, 1998 4827

supercritical fluid extraction (SFE),10-12 Soxtec (Soxhlet extraction),13 steam distillation,14 microwave extraction (MWE),15,16 and more recently also ASE,17-19 with comparable extraction yields being quoted under respectively optimized conditions.8 However, as far as ASE is concerned, investigations have focused on applications only for the soil matrix,20-23 sewage sludge,20,21 sediments,8,23 fly ash,23 and road dust.23 Until now, USE24-27 has been favored for the extraction of SOCs from plant material. The solvent mixture n-hexane/acetone has generally proved to be the most suitable extraction solvent for the ASE applications carried out. Matching or even higher values compared to those obtained with Soxhlet extraction20 have been achieved with ASE for PAHs when using acetone/dichloromethane and toluene. Moreover, a significant relationship has been observed between the extraction temperature, extraction time, and extraction solvent for certain classes of substances analyzed (PAHs, alkanes).8 The aim of this study is to optimize ASE for the extraction of SOCs from plant materials such as mosses and pine needles in order to exploit the advantages of this method (short extraction time, small amounts of solvent) and to achieve results comparable to those obtained with USE. EXPERIMENTAL SECTION Sampling. Mosses. The moss samples of the species Pleurozium schreberi used in this study were collected in October 1996 at biomonitoring sites for mosses in central Germany (LeipzigHalle region) and mixed for temperature- and solvent-dependent ASE investigations. The water content of mosses was determined by drying separate subsamples at 85 °C until the constant final dry weight (dw) had been reached (∼80 wt % water). After sampling, the mosses were kept cool during transport and then frozen at -20 °C until further preparation. Needles. The 2-year-old pine needles of the species Pinus sylvestris L. were collected in July/August 1997 in southern Russia (10) Langenfeld, J. J.; Hawthorne, S. B.; Miller, D. J.; Pawliszyn, J. Anal. Chem. 1993, 65, 338-344. (11) Hawthorne, S. B.; Langenfeld, J. J.; Miller, D. J.; Burford, M. D. Anal. Chem. 1992, 64, 1614-1622. (12) Isco. Application Notes SFE 119, 121; Isco Inc.: Lincoln, NE, 1991. (13) Tremolada, P.; Burnett, V.; Calamari, D.; Jones, K. C. Environ. Sci. Technol. 1996, 30, 3570-3577. (14) Veith, G. D.; Kiwus, L. M. Bull. Environ. Contam. Toxicol. 1977, 17, 631636. (15) Lopez-Avila, V.; Young, R.; Benedicto, J.; Ho, P.; Kim, R.; Beckert, W. F. Anal. Chem. 1995, 67, 2096-2102. (16) Onuska, F. I.; Terry, K. A. J. High Resolut. Chromatogr. 1995, 18, 417421. (17) Dionex. Application Notes ASE 313, 316, 318, 320; Dionex: Sunnyvale, CA, 1994. (18) Ho¨fler, F.; Ezzell, J.; Richter, B. Labor Praxis 1995, 3, 62-67. (19) Ho¨fler, F.; Ezzell, J.; Richter, B. Labor Praxis 1995, 4, 58-62. (20) Popp, P.; Keil, P.; Mo¨der, M.; Paschke, A.; Thuss, U. J. Chromatogr. 1997, A774, 203-211. (21) Enders, B.; Schwedt, G. J. Prakt. Chem. 1997, 339, 250-255. (22) Montero, G. A.; Schnelle, K. B., Jr; Giorgio, T. D. J. Environ. Sci. Health 1997, A32, 481-495. (23) Richter, B. E.; Ezzell, J.; Knowles, D. E.; Ho¨fler, F. Chemosphere 1997, 34, 975-987. (24) Reischl, A.; Reissinger, M.; Hutzinger, O. Chemosphere 1987, 16, 26472652. (25) Umlauf, G.; Hauk, H.; Reissinger, M. Chemosphere 1994, 28, 1689-1699. (26) Wenzel, K.-D.; Mothes, B.; Weissflog, L.; Schu ¨u ¨rmann, G. Fresenius’ Environ. Bull. 1994, 3, 734-739. (27) Wenzel, K.-D.; Weissflog, L.; Paladini, E.; Gantuz, M.; Guerreiro, P.; Puliafito, C.; Schu ¨u ¨ rmann, G. Chemosphere 1997, 34, 2505-2518.

4828 Analytical Chemistry, Vol. 70, No. 22, November 15, 1998

(near the Caspian Sea) and in October 1996 in the Leipzig-Halle region. The ASE investigations were performed on needles from up to three individual trees per site. The water content of the needles was determined by drying separate subsamples at 85 °C until the constant final dw has been reached (∼55-60 wt % water). After sampling, the needles were kept cool during transport and then frozen at -20 °C until further preparation. Standards, Materials, and Solvents. The extracts were quantified using p,p′-DDT, p,p′-DDE, γ-HCH, the internal standard mixture containing a 13C-labeled chlorobenzene cocktail (EM-1725A), PCB 28 (EC-1413), PCB 153 (EC-1406), and the deuterated PAH surrogate cocktail (ES-2044). These standards were supplied by Promochem (Wesel, Germany). All solvents (n-hexane, dichloromethane, toluene, acetone, methanol, diethyl ether) used were of analytical grade (Merck GmbH, Darmstadt, Germany). Forty grams of Florisil 60-100 mesh (Promochem) was conditioned with the double volume of solvents methanol and dichloromethane. After being dried at 100-120 °C in a drying cupboard, it was activated under nitrogen for 4 h at 180 °C. ASE. General Conditions. Extractions were carried out using stainless steel vessels of a Dionex ASE 200 (Dionex GmbH, Idstein, Germany) at temperatures in the range from room temperature up to a maximum of 170 °C and at a pressure of 15 MPa. The extraction of a sample consists of a heating phase of 5 min and three static cycles of 10 min each (total extraction time, 35 min). The dried moss or pine needle samples corresponding to 10 g fresh weight (fw) were used for investigations. The pine needles were divided beforehand by extraction with dichloromethane into the wax (adsorbed on sea sand) and the inner needle fraction (cf. Extraction of Needles below). Both needle fractions and the mosses were separated, placed in vessels, and extracted separately with ∼50 (wax or needle fractions) or ∼30 mL (mosses) of solvent and rinsed with ∼2 mL of the same solvent, the time for rinsing (nitrogen) being 120 s. The stagnant volume was filled up with hydro matrix (bulk Isolute sorbent, Isolute HM-N, International Sorbent Technology GmbH, Bad Homburg, Germany). Extraction of Mosses. Mixed samples from one sampling site were used for the investigations. Before applying ASE, 10 g fw of mosses was dried at room temperature for ∼24 h until a constant final weight had been reached. Double determinations were performed in 22-mL vessels. Samples were accurately weighed, and depending on the water content, ∼2 g dw per 11mL vessel was applied. The extraction temperature for mosses was constant for each double sample. The extractions were performed at room temperature (21 °C), 40, 60, 80, 100, 120, and 170 °C. The extraction solvent was n-hexane. Extraction of Needles. Before applying ASE, 10 g fw of the needles was dried for ∼24 h at room temperature. The outer wax layer was then separated from the inner needle fraction by USE in a glass beaker using 100 mL of dichloromethane for 10 min and filtered. The remaining inner needles on the filter were shredded and placed in a 33-mL vessel containing a mixture of 10 g of Florisil/Al2O3 (anhydrous, γ-aluminia GR, Merck GmbH) 2:1 (preliminary cleaning). The extractions were performed at 40 and 120 °C. The extraction solvent was n-hexane (cf. Table 1). The filtrate containing the wax layer was concentrated to 2 mL, adsorbed on sea sand (purified by acid and calcined GR,

Table 1. Design Used for the Extraction Tests (USE and ASE) of SOCs from Mosses (Pleurozium schreberi) and Pine Needles (Pinus sylvestris L.) method

matrix

solvent

temp (°C)

extraction time (min)

no. of extractions

volume of solvent (mL)

USE

mosses needles needles needles mosses mosses needles

n-hexane n-hexane/acetone (1:9) n-hexane CH2Cl2/acetone (1:1) n-hexane toluene n-hexane/acetone (1:1) CH2Cl2/acetone (1:1) n-hexane

room room room 100 room, 40-170 40 and 120 40 120 40 and 120

10 10 10 35 35 35 35 35 35

3 3 3 1a 1a 1a 1a 1a 1a

300 300 300 60-90 30-60 30-60 60-90 60-90 60-90

ASE

needles a

One sample extraction consisting of a 5-min heating phase and three static cycles of 10 min each.

Merck GmbH), placed in a 33-mL vessel, and extracted under the same conditions as the inner needle fraction. Cleanup of Mosses and Needles. After applying ASE, internal standard solutions were added. The extracted phases were concentrated to 2 mL. This concentrated solution was transferred to column 1 (diameter 1 cm, length 20 cm) containing ∼15 g of deactivated Florisil (4% water) to separate out the hydrophilic plant compounds. It was eluted with 160 mL of n-hexane/dichloromethane 1:1. The first 60 mL of the eluate, containing the principal quantity of the pollutants, was transferred to column 2 (diameter 0.5 cm, length 20 cm) containing ∼3.5 g of activated Florisil to separate out lipophilic plant compounds. Column 2 was eluted with 60 mL of n-hexane/dichloromethane 1:1. All the eluates were combined and concentrated to dryness, the residue being extracted three times with 2 mL of diethyl ether, evaporated to dryness under nitrogen, and dissolved in 200 µL of toluene.26,27 Analysis of Mosses and Needles. GC/MS analysis was carried out using a Hewlett-Packard (HP) 5971 mass spectrometer in SIM mode, coupled with an HP 5890 capillary column gas chromatograph equipped with the HP 7673 autosampler. An HP Ultra 2 capillary column, 25 m × 0.32 mm i.d. × 0.52 µm film thickness, was used. The carrier gas was helium (purity 5.0, Linde GmbH, Ho¨llkriegelskreuth, Germany). One microliter of sample was injected in the splitness mode. The temperature program was as follows: initial temperature 60 °C held for 1 min, followed by a 10 °C min-1 ascent to 260 °C, maintained for 1 min. Calibration standard solutions containing the 13C-labeled organochlorines and the deuterated PAHs in extracted needle matrix were used to determine the detection limit of the analytes. The SOCs in mosses and needle extracts were identified by matching each substance retention time with the retention times of internal 13C-labeled and deuterated standards. Concentrations of analytes were calculated on the basis of recoveries of the internal standards. Recovery in the pine needle matrix for chlorinated organics ranged from 87% for γ-HCH to 96% for PCB 153. PAH recovery was 57-78% for pyrene, phenanthrene, anthracene, and fluoranthene and 25-50% for benzo[a]pyrene.26,27 USE. General Conditions. The samples were shredded with the Ultra Turrax T25 (IKA-Labortechnik, Janke & Kunkel, Staufen, Germany). Extractions (bath sonication; cooling water was circulated through the bath) were carried out using a Transsonic Digital S (Elma, HF 40 kHz, 400 W, Singen, Germany). Extraction time was 10 min for the separation of the wax layer and 3 × 10 min for the extraction of the inner needle fraction or mosses.

Extraction of Mosses. The moss samples of 10 g fw were dried at room temperature for ∼24 h until a constant final weight had been reached. This amount was mixed with 100 mL of n-hexane in a glass beaker, shredded with Ultra-Turrax, and extracted by USE (10 min). The solvent was decanted. The remaining residue was extracted two more times with 100 mL of n-hexane as described above. The three solutions were combined, the internal standard solutions were added, and the extracts were concentrated to 2 mL. Extraction of Needles. The needle samples of 10 g fw were dried at room temperature for approximately 24 h until constant final weight had been reached. The wax layer was separated in a glass beaker with 100 mL of dichloromethane by USE for 10 min and filtered. The filtrate was concentrated to 2 mL. The remaining needles (inner needle fraction) of the filtrate were shredded and extracted by USE (cf. Extraction of Mosses) following two individual extraction steps with first 100 mL of n-hexane and second 100 mL of n-hexane/acetone 1:9 to ensure the complete extraction of PAHs and some organochlorines. Only the first inner needle fraction (n-hexane phase) was washed out with 250 mL of a 10% Na2CO3 (anhydrous, Merck GmbH) solution, and then the organic phase was concentrated to 2 mL. The second extraction phase (n-hexane/acetone 1:9) was concentrated nearly to dryness and taken up with 30 mL of dichloromethane. For the final results, the values of three fractions were used: one wax fraction and two inner needle fractions. Cleanup and Analysis of Mosses and Needles. The cleanup procedure and analysis were performed separately for each extract to improve the purification efficiency for this complicated matrix. Cleanup and analysis were done as described above for ASE. RESULTS AND DISCUSSION Preliminary Investigations. Table 1 shows the extraction parameters for the extraction tests carried out. Work commenced by testing a mixed needle sample with a design based on previous applications by Dionex17 with respect to the soil matrix (n-hexane/ acetone 1:1 as solvent, 100 °C as extraction temperature). Comparing determinations of the inner needle material done using both ASE and USE surprisingly revealed that, under these conditions, only lower levels of pollutants can be detected when using extraction with ASE (cf. Figure 1). Although the highly volatile hexachlorobenzene and low-chlorinated PCBs were not detected with ASE, the ASE conditions were somewhat more favorable for the DDT metabolites DDE and DDD. The lower Analytical Chemistry, Vol. 70, No. 22, November 15, 1998

4829

Figure 1. Comparison between ASE and USE using mixed samples of pine needles (region Leipzig-Halle). A mixture of acetone/n-hexane (1:1 v/v) was used as extraction solvent. Not detectable: DDT, 1,2,3,4-tetraCB, pentaCB, PCB 101, PCB 138, PCB 153, PCB 180, and BaP.

concentrations of SOCs in needles generally measured following ASE were attributed to the polar fraction of acetone in the extraction solvent. In the complicated plant matrix, this results in increased extraction of polar substances contained therein (including chlorophyll and conjugates), which influences cleanup and analysis. One advantage of ASE is that not only the pressure and temperature but also the polarity of the extraction solvent (individually or in mixtures) can be chosen from a wide range and adapted to the respective matrix and substance palette. This is not the case, for example, with SFE for substances with greatly variable physicochemical properties, as extraction with liquid carbon dioxide is prescribed regardless of the properties of the anthropogenic substances to be extracted, and affecting solvent polarity by adding polar modifier volumes is possible only up to 10%. In the case of MWE, which is carried out in pressure vessels which are permeable to microwaves, the solvent mixture cannot be chosen completely freely either, as a component which can be excited by microwaves must be contained. Thus, pure hydrocarbons, such as the n-hexane frequently used for the extraction of SOCs, are unsuitable for the MWE method.21 It became apparent from the initial investigations that the extraction solvent for this matrix had to be varied in order to obtain extraction yields which are at least comparable to those obtained with USE corresponding to the soil extractions. Influence of Temperature on the Extraction Efficiency. The extraction conditions for mosses were optimized for two reasons. First, mosses have an uncomplicated, more homogeneous matrix, which facilitates better cleanup and more sensitive analysis. Judging by to our several years’ experience in using the USE method for needles and mosses, the margin of error upon 4830 Analytical Chemistry, Vol. 70, No. 22, November 15, 1998

multiple determinations in the polluted pine needle matrix is much higher, so that mosses have come to be regarded as a more suitable matrix to obtain fundamental findings concerning the impact of temperature on extraction. Second, as they have no wax layer, mosses can approximately be compared with the inner fraction of pine needles. The extraction solvent used in each case was n-hexane, resulting in fewer polar matrix components (which have a disruptive effect on cleanup and analysis) being extracted at the same time compared to what was observed the extraction mixture n-hexane/acetone 1:1. Although similar results were achieved with toluene, it was not used owing to its higher boiling point and the longer time required to prepare the extracts. Initial investigations at an extraction temperature of 100 °C resulted in significantly greater yields of extracted pollutants, e.g., increases of ∼100% for HCHs, of ∼90% for DDX, and of ∼20% for the PAHs. Under these conditions, ASE and USE produced roughly comparable results for PCBs and chlorobenzene. Subsequently, double determinations were carried out from a larger pool of mixed moss samples from one location at a constant pressure of 15 MPa at each temperature listed (in the range from room temperature to 170 °C). Varying the pressure does not affect the results.28 The following conclusions can be derived from Figure 2 for the five pollutant groups: HCHs, DDX, and Chlorobenzenes. The best extraction conditions prevail at 120 °C. The only exceptions were DDE and δ-HCH, for which the extraction yields at 170 °C were higher by about 10% and 80%, respectively, compared to those at 120 °C. However, an extraction temperature of 170 °C is unadvisable, as (28) Saim, N.; Dean, J. R.; Abdullah, M. P.; Zakaria, Z. Anal. Chem. 1998, 70, 420-424.

Figure 2. Extraction efficiency of different pollutant groups from mosses (Pleurozium schreberi) as a function of the extraction temperature using ASE (solvent is n-hexane).

Analytical Chemistry, Vol. 70, No. 22, November 15, 1998

4831

Table 2. Extraction of SOCs with USE and ASE from the Inner Pine Needle Fraction at Three Trees of One Sitea siteb C ˇ e¨rnyj Jar

Krasnyj Jar

USEc

ASEc

USEc

ASEc

compd

mean

RSD (%)

mean

RSD (%)

mean

RSD (%)

mean

RSD (%)

R-HCH β-HCH γ-HCH δ-HCH -HCH sum/av RSD (%)

0.19 0.04 0.12 nd n 0.35

42.1 25.0 15.4

6.83 6.61 31.5 0.84 6.96 52.7

6.5 8.9 18.6 11.8 25.7 14.3

3.18 0.12 0.80 1.15 0.43 5.68

0.6 25.0 7.5 13.0 9.3 11.1

2.68 2.48 3.55 1.94 0.97 11.6

27.2 25.4 24.5 42.8 18.6 27.7

DDT DDE DDD sum/av RSD (%)

0.14 nd nd 0.14

21.4

0.90 2.01 1.28 4.19

15.4 71.2 12.8 33.1

0.17d 0.63 0.08d 0.88

0.56 0.57 1.56 2.69

3.6 47.4 35.3 28.7

tetraCB hexaCB sum/av RSD (%)

nd nd

nd 0.38 0.38

3.7 3.7

nd 0.22d 0.22

0.26 0.52 0.78

9.8 19.2 14.5

PCB 28 PCB 52 PCB 101 PCB 138 PCB 153 PCB 180 sum/av RSD (%)

nd nd nd nd nd nd

40.0 3.83 11.3 8.00 8.31 2.46 73.6

3.2 4.5 8.4 2.8 7.0 5.9 5.3

0.38d 0.22d 0.19d 0.10d 0.10d nd 0.99

60.7 0.97 2.81 2.01 2.14 1.39 70.0

14.2 21.6 17.1 29.9 22.0 16.5 20.2

PHEN ANT FLUOR PYR BaP sum/av RSD (%)

0.39 0.04 0.38 0.14 nd 1.31

83.5 6.48 34.6 8.68 0.84 134

6.5 8.4 9.2 6.1 36.4 13.3

17.6 1.84 11.6 7.68 0.30d 39.0

52.6 4.11 27.6 12.3 0.27 96.9

20.2 20.9 14.2 26.2 51.9 26.7

27.5

21.4

15.4 50.0 26.3 35.7 31.9

1.6

4.8 1.6 2.8 0.4 2.4

a Concentration is nanograms per gram of dry mass (nd, not detectable). Solvent is n-hexane. b Sites in southern Russia. c Number of the replicated investigations is three. d Only one measuring value because of insufficient extraction efficiency.

very large quantities of matrix components are also extracted. Even after using the cleanup technique, this still led to very impure extracts and, as a result, to considerable disruption of the whole analysis. PCBs. Good conditions prevail at 40 °C. As the extraction temperature rises, the amounts of substance extracted for the analyzed PCBs 101, 138, 153, and 180 slightly decrease. No temperature relationship was observed for the PCBs 28 and 52. PAHs. A sharp rise in the extraction yield takes place above all between 100 and 120 °C for four of the five PAH representatives. For the three PAHs phenanthrene, fluoranthene, and pyrene, the increase is between 17 and 26%; for anthracene a 4-fold increase takes place. Only benzo[a]pyrene is extracted in comparable quantities irrespective of the extraction temperature in the range between room temperature and 120 °C. General Observations. Sufficiently good conditions for the simultaneous extraction of organochlorines and PAHs were achieved when using two successive extraction steps at 40 and 120 °C. Furthermore, the extraction temperatures should not exceed 120 °C, as in many cases plant substances were precipitated in the extraction solution, interfering with cleanup and analysis (due to overlapping peaks). The remarkably high rise in extraction yields of δ-HCH at 170 °C (an exception among 21 SOCs; most other substances could not be detected at this temperature) and PCB 101 at room temperature should be the target of ecotoxicological rather than analytical attention (owing to the possibly stronger bonding force of δ-HCH and the very low 4832 Analytical Chemistry, Vol. 70, No. 22, November 15, 1998

bonding force of PCB 101 to plant tissues such as lignin, membranes, or cell substrates). This behavior could be explained on the basis of the physicochemical properties of the SOCs (e.g., differences in water solubility, lipophility, molecule size, spatial structure), the degree of chlorination, and the affinity to plant tissues. Extraction and Analysis of Separated Wax Layer and Inner Needle Fraction. In the Experimental Section, it can be seen that, in contrast to USE, the needles were not shredded before ASE. For better extraction under the USE conditions, it proved advantageous to enlarge the plant surface to improve the extraction yield. However, this necessitates larger amounts of solvents and longer preparation times (evaporation of larger amounts of solvents). ASE at higher temperatures and pressures in order to accelerate extraction does not require the sample material to be shredded. Moreover, it is certainly difficult to adsorb larger amounts of mushy plant material on sea sand and to place it into vessels. The needle material studied from areas in southern Russia had undergone high multiple exposures to impurities over a period of 2 years. The investigations into this highly polluted needle material were carried out separately on the wax layer and the inner needle fractions. The quantifying pollution fractions in the inner needle are more important than total levels; these fractions are far more likely to be the source of pollutant effects on biological structures and so are of greater relevance for ecotoxicological evaluation. The results are listed in Tables 2-4 using the ASE

Table 3. Comparison of the Extraction Methods USE and ASE for SOCs from the Inner Pine Needle Fractiona siteb Kislovodsk

Elbrus

compd

USE

ASE

USE

R-HCH β-HCH γ-HCH δ-HCH -HCH sum

0.38 0.44 5.02 2.05 0.39 8.28

5.61 0.49 2.91 0.41 0.27 9.69

1.36 0.20 0.56 0.25 0.16 2.53

DDT DDE DDD sum

0.46 0.30 0.76

0.40 0.63 0.11 1.14

tetraCB hexaCB sum

nd 0.26 0.26

0.30 0.39 0.42

PCB 28 PCB 52 PCB 101 PCB 138 PCB 153 PCB 180 sum

0.21 0.18 0.49 nd nd nd 0.88

39.5 0.30 0.27 0.24 0.29 nd 40.6

PHEN ANT FLUOR PYR BaP sum

15.9 4.18 3.32 0.90 0.53 24.8

31.2 2.04 11.2 3.99 0.38 89.4

Godschur ASE

Elista

USE

ASE

USE

ASE

0.96 1.76 2.30 0.76 0.47 6.25

0.29 nd 0.55 0.28 0.13 1.25

4.43 9.76 3.29 1.27 3.55 22.3

0.84 0.39 1.19 0.56 0.25 3.23

17.5 7.61 22.9 1.65 1.11 50.8

0.82 0.45 0.05 1.32

0.19 0.33 1.03 1.55

0.90 0.13 0.03 1.06

1.23 0.61 0.07 1.91

0.17 0.11 0.19 0.47

2.37 2.64 0.98 5.99

nd nd

0.15 0.37 0.52

nd nd

0.02 0.37 0.39

nd nd

nd 0.62

0.10 0.09 0.22 nd 0.19 nd 0.6

49.7 1.05 2.89 2.07 2.14 nd 57.9

nd nd nd nd nd nd

12.9 0.18 0.22 0.24 0.33 1.91 15.8

0.24 0.14 0.27 nd nd nd 0.65

38.5 4.48 12.6 9.28 9.48 3.40 77.7

11.9 2.43 12.2 22.3 0.26 49.7

65.6 2.83 21.2 6.85 0.28 155

2.12 0.71 1.99 6.25 1.15 12.2

19.8 0.55 10.2 1.76 0.31 32.6

7.99 0.72 1.80 0.18 nd 10.7

124 9.69 38.7 9.60 nd 182

a Concentration is nanograms per gram of dry mass (nd, not detectable). Solvent is n-hexane. b Measuring values of one tree at differently polluted sites in southern Russia.

conditions given under the subheading Extraction of Needles. Table 2 lists the findings of triple determinations (three individual trees from the same location) of the inner needle fraction at two different sites. When using ASE, the RSD (relative standard deviation; generally quoted at 20-60% at the same location1) was in an excellent range at both sites considering the plant matrix (location 1, 5-20%; location 2, 15-30%; exceptions, DDE, δ-HCH, and benzo[a]pyrene). The two locations in southern Russia, Cˇ e¨rnyj Jar and Krasnyj Jar, differ with respect to their pollution (the extract phase from location 2 contains a very oily fraction). The difference between the two locations becomes especially apparent upon direct comparison with USE. Whereas at location 2 the extracts obtained using ASE from all three trees could be analyzed, extraction with USE did not produce measurable values for the substance groups HCHs and PAHs in all three trees. However, at some less polluted Russian locations, similar or higher results for individual substances were detected by USE (Kislovodsk, cf. Table 3). The poorer analyte/matrix separation of USE is especially conspicuous at more contaminated sites, e.g., the very oily extracts from Krasnyj Jar. At less contaminated sites, analyte/ matrix separation for USE is partially comparable with that of ASE. One reason for the differences in analyte/matrix separation between USE and ASE could be that bonds between conjugates from substance and plant tissue can be destroyed only under ASE conditions. This was verified by subsequent ASE of the column material Florisil used for cleanup when the quantities of pollutants undetected after USE were found in the expected amounts. Furthermore, the analyzed quantity of pollutants at both locations

was noticeably increased when using ASE (partly by up to 2 orders of magnitude). With respect to the pollutant class of PCBs in particular, very high levels were found in comparison to those previously measured during extensive biomonitoring investigations in the Leipzig-Halle region (central Germany) and Mendoza (Argentina),27 while here none at all was detected by GC/MS following USE. These findings were confirmed by USE/ASE comparisons of other needle samples from other locations in southern Russia (cf. Table 3). Combining our enhanced ASE with chemical ionization using a GC/MSD, which itself improves the detection limit for organochlorines with three or more chlorine atoms in the molecule by up to 3 orders of magnitude, may open up a new possibility for the detection of the tiny amounts of organochlorines and also dioxins in inner needle compartments. The wax layer represents a less complicated matrix for analytical investigations than the inner needle. As was to be expected, this meant that such high increases in extraction yield as those obtained for the inner needle fraction were not observed. Nevertheless, ASE of the wax layer from the same needle material as used for the inner needle fraction (cf. Table 3) also produced remarkable results with substance-dependent increases in the extraction rates of an average of up to 1 order of magnitude compared to USE (cf. Table 4). However, the pollutant levels ascertained in the wax layer are generally below those in the inner needle. This is comparable to results of investigations in Germany and Argentina. When comparing pollutant groups, the largest increases in extraction yields were obtained for PCBs and PAHs. Analytical Chemistry, Vol. 70, No. 22, November 15, 1998

4833

Table 4. Comparison of the Extraction Methods USE and ASE for SOCs from the Wax Fraction of Pine Needlesa siteb Kislovodsk

Elbrus

Godschur

Elista

compd

USE

ASE

USE

ASE

USE

ASE

USE

ASE

R-HCH β-HCH γ-HCH δ-HCH -HCH sum

0.31 0.69 1.02 0.49 0.30 2.81

1.81 1.48 1.96 1.51 1.34 8.10

0.12 0.76 0.14 0.15 0.10 1.27

0.30 0.85 0.44 0.23 0.68 2.50

0.15 0.45 0.54 0.20 0.31 1.65

3.33 0.35 7.29 0.83 0.26 12.1

0.36 0.36 1.21 0.20 0.18 2.31

0.32 0.38 1.47 0.56 0.75 3.48

DDT DDE DDD sum

0.20 0.03 0.16 0.39

2.02 0.50 0.32 2.84

0.19 0.21 0.16 0.56

0.41 0.42 0.12 0.95

0.67 0.26 nd 0.93

22,4 0,45 0,10 23.0

0,22 0,58 0,09 0,89

0.24 0.36 0.13 0.73

tetraCB hexaCB sum

nd nd

nd 0.09 0.09

nd nd

nd nd

nd nd

nd nd

nd nd

nd nd

PCB 28 PCB 52 PCB 101 PCB 138 PCB 153 PCB 180 sum

nd 0.04 nd nd nd nd 0.04

2.97 0.42 0.64 0.81 0.62 nd 5.46

0.11 0.07 0.25 nd 0.08 nd 0.51

1.30 0.26 0.38 0.22 0.04 nd 2.20

nd nd nd nd nd nd

3.74 0.14 nd nd nd nd 3.88

0.21 0.17 0.28 nd nd nd 0.66

0.21 0.15 0.26 0.12 0.21 0.40 1.35

PHEN ANT FLUOR PYR BaP sum

5.70 0.37 2.34 0.88 0.36 9.69

25.0 31.4 11.2 8.60 0.27 81.9

3.44 0.31 3.34 1.42 0.12 9.14

15.6 4.71 6.60 3.69 0.14 32.9

5.16 0.69 2.83 1.09 nd 9.77

9.98 1.25 5.93 2.28 nd 19.4

10.4 0.97 3.37 0.98 nd 15.7

14.2 1.50 6.52 1.83 nd 24.1

a Concentration is nanograms per gram of dry mass (nd, not detectable). Solvent is n-hexane. b Measuring values of one tree at differently polluted sites in southern Russia.

Comparable extractions of vessels filled with sea sand (analogous conditions) to determine the blank value yielded no detectable amounts of pollutant. The higher the multiple contaminations at the location and the more complicated the matrix was, the higher the rates of increase following ASE. These findings show that the key advantage of ASE over USE clearly comprises the greater separation of lipophilic analyte from lipophilic matrix, especially in the case of more difficult plant matrixes and higher multiple contamination. The reason is that, at constantly high pressure, the higher temperatures mean that sufficient energy can be supplied for such separation, this separating effect being especially enhanced when using the nonpolar extraction solvent n-hexane. By contrast, as the degree of multiple contamination at the location concerned increases, bonds between analytes and matrix components cannot be destroyed when using USE. Nevertheless, USE is still a powerful extraction method for SOCs from needles and mosses at less contaminated sites. Placing a mixture of Florisil/Al2O3 into the extraction vessel for the preliminary purification of the needle extracts also had a positive effect. Using the ASE method enables a remarkable improvement in the detection of SOCs in the very complicated pine needle matrix,2 a biomonitoring plant which is globally preferred for the qualitative detection of pollutant levels.26,29,30 These extraction conditions

established ought also to have a fundamental bearing on extractions of SOCs from other plant matrixes. However, alterations to the experimental parameters are conceivable, depending on the complicated plant matrix of the vegetation representative and the plant compartment investigated, as well as the nature of sitespecific multiple exposures.

(29) Calamari, D.; Tremolada, P.; Guardo, A. D.; Vighi, M. Environ. Sci. Technol. 1994, 28, 429-434.

(30) Strachan, W. M. J.; Eriksson, G.; Kylin, H.; Jensen, S. Environ. Toxicol. Chem. 1994, 13, 443-451.

4834 Analytical Chemistry, Vol. 70, No. 22, November 15, 1998

CONCLUSIONS ASE can be used as a very successful technique for the extraction of SOCs from pine needles and mosses. The particular advantage of ASE over USE lies in the clear improvement of the separation of lipophilic pollutants from the plant matrix when using matrix-specific optimized conditions. The choice of a suitable extraction solvent is crucial. ASE, which represents a sharp rise in quality in the extraction of SOCs from pine needles, produces an increase in the analysis results of up to 2 orders of magnitude, especially with highly polluted sample material and a complicated plant matrix (e.g., pine needles). At less contaminated locations, the differences in the results are low between ASE and USE. However, only ASE can aid in assessing more contaminated locations of ecotoxicological and toxicological interest. The findings obtained open up new possibilities for the estimation of regional and global contamination levels in pollution monitoring, their ecotoxicological relevance, and the assessment of the importance of vegetation as a pollutant sink.

ACKNOWLEDGMENT We thank PD Dr. K. Jung for the moss material and Dr. P. Popp for helpful advice. Our thanks also go to Ms. Mothes, Ms. Petre, and Ms. Dipl.-Ing. Heinrich for their excellent technical assistance in performing sample preparation, cleanup, and GC/ MSD analysis of the comparative samples. The needle samples from southern Russia used in this work were provided in the

context of the project INTAS-COPERNICUS, Project ECCA, No. PL 963203, supported by the European Union.

Received for review June 9, 1998. Accepted September 4, 1998. AC9806299

Analytical Chemistry, Vol. 70, No. 22, November 15, 1998

4835