ANALYST, JANUARY 1987, VOL. 112 31 Analytical Reference Materials Part Vl.* Development and Certification of a Sediment Reference Material for Selected Polynuclear Aromatic Hydrocarbonst Hing-Biu Lee, Geeta Dookhran and Alfred S. Y. Chau Quality Assurance and Methods Section, Analytical Methods Division, National Water Research Institute, Environment Canada, Burlington, Ontario L7R 4A6, Canada A naturally contaminated environmental sediment reference material (EC-1) was developed and analysed for selected polynuclear aromatic hydrocarbons (PAH). Freeze-dried and homogenised subsamples of EC-1 were Soxhlet extracted and the extracts were cleaned-up on activated silica gel and alumina columns. The levels of PAH in this material were determined by using the following three independent analytical methods: (1) GC - FID, (2) GC - MS and (3) reversed-phase HPLC with fluorescence and UV detectors.Up to a total of 72 replicate determinations were performed and the results obtained by each method were in good agreement with each other. lnterlaboratory PAH results for EC-1 obtained in a round-robin study also confirmed the in-house data. The results for ten PAH of which the between-method difference was less than k 10% were pooled t o generate the certified values. Keywords: Pol ynuclear aromatic hydrocarbon determination; certified reference material; quality assurance; sediment samples Polynuclear aromatic hydrocarbons (PAH) are ubiquitous environmental pollutants as they are naturally occurring and are also formed by the pyrolysis of carbonaceous materials at high temperatures.The routine determination and monitoring of PAH in environmental samples are essential because of their persistence and carcinogenic and mutagenic properties. Although PAH levels in open lake or marine surface waters are low, they are readily adsorbed and accumulated by sediments and particulate matter and pg g-1-ng 8-1 levels of PAH have been reported in many sediment samples.2-5 Methods for the determination of PAH in environmental samples have been reviewed previously.6 The most popular detection techniques involve either gas chromatography (GC) or high-performance liquid chromatography (HPLC) . The determination of PAH by HPLC with a fluorescence detector has been used by many workers7-9 as many PAH are highly sensitive to this detector.The judicious selection of excitation and emission wavelengths also makes the detector highly specific, which minimises the interferences from unresolved co-eluting PAH resulting from highly complicated mix- tures.7-10 In many instances the fluorescence detector is complemented by a variable-wavelength UV detector for the detection of those PAH with a low sensitivity to the former. 11-13 GC determinations of PAH are usually carried out with high-resolution capillary columns.2-5J4J5 The quality of these chromatograms is much higher than those obtained by packed columns and therefore capillary columns are considered essential for complicated environmental samples. Although PAH are generally detected by a flame ionisation detector (FID), electron-capture detectorsl6J7 can also be used under special circumstances for these compounds.The determina- tion of PAH by a sensitive and selective mass spectrometer interfaced to a high resolution capillary column is currently the most powerful approach.lgJ9 In this instance, mass spectra of a sample provide positive identification of known PAH or structural information for unknown PAH. By using the selected ion monitoring (SIM) technique, sub-nanogram amounts of PAH can easily be detected. * For Part V of this series, see reference 26. t This material is currently not for sale and not available for general distribution. Two interlaboratory studies on the determination of PAH in sediment samples have been reported.20.21 The results of these studies and that organised by our section22 indicated that widespread results were obtained from different laboratories.For many parameters, interlaboratory relative standard devia- tions (after the rejection of outliers) ranged from 30 to 60%, suggesting that there was a need to improve the accuracy of data obtained from PAH determination in environmental samples. Although several certified reference materials (CRM) for PAH determination have been prepared by the US National Bureau of Standards,23 only one of them, a sample of urban particulate matter, was in an environmental matrix. In order to fulfil quality assurance and method research require- ments, the development of sediment CRMs for PAH is therefore necessary. In this paper, we describe the development and certification of the first in a series of sediment reference materials for method evaluation and other in-house and interlaboratory quality assurance applications.Certified or reference values for the following 14 PAH are reported: phenanthrene (phen), anthracene (anth), fluoranthene (F), pyrene (Py), benzo- [alanthracene (B[a]A) , chrysene (chry) , benzo[b]fluoran- thene (B[b]F), benzoCk1fluoranthene (B[k]F), benzo[e]py- rene (B[e]P), benzo[a]pyrene (B[a]P), perylene (pery), inde- no[l23cd]pyrene (I[cd]P), dibenz[ah]anthracene (D[ah]A) and benzo[ghi]perylene (B[ghi]P). All of the above, except B[e]P and perylene, are listed as US EPA priority pollu- tan ts .24 Experimental Preparation of Sediment Reference Material Approximately 450 kg of wet sediment were collected from a landfill site in Hamilton Bay, Ontario, Canada. The sample, coded EC-1, was freeze-dried, crushed, sieved, blended and subsampled. Further details of this procedure have been published elsewhere.25.26 Extraction of PAH in Sediments A 10.00 g aliquot of EC-1 was extracted in a Soxhlet apparatus with 350 ml of 59 + 41 V/V acetone - hexane for 8 h at a rate of 8 cycles h-1.For the comparison of recoveries, Soxhlet32 ANALYST, JANUARY 1987, VOL. 112 extraction with other solvent systems and extraction using a soni~ator25~26 were also evaluated. The results of this compari- son are given under Results and Discussion. Clean-up of Sediment Extracts The combined organic extract was shaken with 400 ml of 2% KHC03 in a 1-1 separating funnel for 1 min with frequent venting.After the layers had separated, the aqueous layer was drained into a 500-ml separating funnel and then discarded. The organic layers in the two separating funnels were combined and passed through 100 g of anhydrous Na2S04 in a column. The funnels were washed with 2 x 10 ml of hexane and the washings were again applied to the column. After the last trace of solvent was removed from the Na2S04 column by vacuum suction, the dry extract was evaporated down to ca. 5 ml using a rotary evaporator with a 35 "C water-bath. A 400 x 10 mm i.d. glass clean-up column with either a coarse porosity fritted disc or a glass-wool plug was filled with a freshly prepared slurry of 10.0 g of silica gel (Davison grade 923, 100-200 mesh activated at 130 "C for 18 h before use) in hexane with 1 cm of anhydrous Na2S04 at the top.The concentrated sediment extract in hexane was quantitatively transferred on to the column and drained just into the Na2S04 layer. The sample flask was rinsed with 2 ml of hexane and the rinsing again applied to the column. This process was repeated twice. The column was then eluted with 50 ml of hexane and the eluate was discarded. This fraction contains chloroben- zenes, PCBs and several chlorinated insecticides if they are also present in the sample. The column was further eluted with 60 ml of 40 + 60 V/V dichloromethane - hexane. This fraction was collected in a 250-ml round-bottomed flask and was evaporated down to ca. 5 ml using a rotary evaporator as described above. After the addition of 20 rnl of hexane and 3 ml of isooctane, the evaporation was repeated until the volume was ca.3 ml. A second clean-up column was prepared by filling a 230 X 5 mm i.d. disposable Pasteur pipette having a glass-wool plug at the bottom with 5 cm of activated neutral alumina (Woelm, Brockmann activity 1 , 100-200 mesh) and 5 mm of anhydrous sodium sulphate at the top. This column was eluted with 5 ml of hexane and the eluate was discarded. The concentrated extract, after silica gel column clean-up, was applied to the column, rinsing through with 3 x 1 ml of hexane. The column was further eluted with hexane until a total of 10.0 ml of hexane was collected. This fraction contained aliphatic hydrocarbons and other non-polar co-extractives that had not been removed by the silica gel column.The PAH were removed from the alumina column by elution with toluene until a volume of 10.0 ml was collected. Gas Chromatography with Flame Ionisation Detection (GC - FID) A Hewlett-Packard 5880A gas chromatography equipped with a Grob-type split - splitless capillary injection port, a flame ionisation detector, a 7671A autosampler and Level IV terminals was used. A 30 m X 0.25 mm i.d. DB-5 fused-silica capillary column of 0.1 pm film thickness (J and W Scientific) operating under the following conditions was used for PAH analysis. Temperatures: injection port and detector, 275 "C; oven initial temperature 70 "C, hold 1.5 min at 70 "C, programming rate 1, 30 "C min-1 (from 70 to 160 "C), rate 2, 2 "C min-1 (from 160 " to 260 "C), hold 10 min at 260 "C. Flow-rates: hydrogen, 30 ml min-1; air, 240 ml min-I; detector make-up gas (helium), 25 ml min-1.Carrier gas, helium; column head pressure, 15 lb in -2. Splitless valve on for 90 s. A 2-p1 aliquot of the final extract was injected in the splitless mode without dilution. Gas Chromatography with Mass Spectrometry (GC - MS) The system consisted of a Hewlett-Packard 5880A gas chromatograph, as described above, a 5970B mass-selective detector (MSD) a 9816s computer and a 9133XV disc drive. The DB-5 capillary column was directly interfaced with the electron-impact ion source (70 eV) for maximum sensitivity. The GC operating conditions were identical to those used in the FID determination, except that the detector gases were not needed and the column head pressure was 4 lb in-2.A 2-p1 aliquot of a 20-fold diluted sample extract was analysed. The data were acquired by the following two modes: (a), linear scanning from rnlz 50 to 300 in order to obtain abundance data of major fragments for compound identification purposes; and (b), selected ion monitoring (SIM) for quantitative analysis. In the latter instance, the following molecular ions characteris- tic of PAH were monitored: 1, rnlz 178 for phen and anth; 2, rnlz 202 for F and Py; 3, rnlz 228 for B[a]A and chry; 4, mlz 252 for B[b]F, B[k]F, B[e]P, B[a]P and pery; 5, mlz 276 for I[cd]P and B[ghi]P; and 6, rnlz 278 for D[ah]A. The dwell time for each ion was 100 ms. Three labelled internal standards, i.e., phen-dlo, chry-d12 and B[ghi]P-Wl2 were used for the calibration of response factors.High-performance Liquid Chromatography (HPLC) A system including a Waters Model 510 pump, a Rheodyne 7125 loop injector and 20- j.d loop, a 4.6 mm i.d. x 25 cm long Zorbax ODS column (DuPont, 5-6 ym particle size) and a Schoffel Model FS 970 fluorescence detector were used. Mobile phase (isocratic), 85 + 15 acetonitrile - water; flow-rate, 1.0 ml min-1. The detector wavelength was set at 280 nm (excitation) and 389 nm (emission). A 20-pl aliquot of a 100-fold diluted sample extract was injected. Standards and Standard Solutions Most PAH standards are available from Aldrich Chemical or Eastman Kodak. Certified reference materials of PAH are also available from the Commission of European Communi- ties, BCR, Brussels. Individual stock solutions were prepared by dissolving 50.0 mg of each PAH in a 100-ml low actinic calibrated flask with toluene; some PAH required gentle heating or sonification to dissolve.Appropriate amounts of the 14 PAH stock solutions in proportions similar to those found in EC-1 were pipetted into a low actinic calibrated flask and diluted to volume with toluene. This solution, which contained PAH at yg ml-1 levels, was used in the GC analyses as an external standard. A standard for the HPLC analysis of EC-1 was prepared by diluting the above solution with the HPLC mobile phase. Results and Discussion The sediment reference material EC-1 was originally prepared and analysed for PCBs.25 It is a fine (200-325 mesh), silty clay sediment naturally contaminated with many toxic organics and metals. Although homogeneity tests were not performed for PAH before subsampling, subsequent determinations on various lots of sediment subsamples did not reveal any inhomogeneity for PAH (Table 1).Therefore, EC-1 is considered sufficiently homogeneous for use as a PAH reference material. Soxhlet extraction has been used by many workers for the determination of organics, including PAH, in sediments. Although most of the extraction in this work was carried out with 41 + 59 hexane - acetone, comparative extraction was also investigated using solvent systems such as 1 + 1 benzene - methanol, cyclohexane and hexane. The results for theANALYST, JANUARY 1987, VOL. 112 33 determination of PAH in EC-1 under various extraction conditions are given in Table 2. No difference in PAH recoveries from EC-1 subsamples was observed between the four solvent systems, although the non-polar solvents (cyclohexane and hexane) gave extracts much lighter in colour.Further experiments by Soxhlet extraction were all carried out with the 41 + 59 hexane - acetone because this solvent was easy to evaporate and was also used in our multi-residue extraction procedure. A longer extraction time (24 vs. 8 h) did not produce higher results (Table 2). Ultrasonic extractions25 of EC-1 with 1 + 1 hexane - acetone were also carried out, and again recoveries of PAH by this technique were identical to those obtained by the Soxhlet method (Table 2). As different solvent systems and different extraction methods gave the same results, it was therefore concluded that the extraction recoveries of PAH from EC-1 were quantitative in these experiments.Table 1. Homogeneity test of various lots of EC-1 subsamples for PAH. Concentrations in pg g-1 Bottling sequence . . Start Middle End Bottleno. . . 144 434 1014 2029 3769 4784 F . . . . 19.6 21.9 20.5 20.9 21.2 20.6 B[u]A. . . . 8.9 7.5 8.3 7.8 8.0 8.5 B[a]P . . . . 5.2 4.6 4.5 4.9 4.7 4.4 I[cd]P. . . . 5.0 5.0 4.3 5.6 4.5 4.7 A rotary evaporator was used to evaporate organic extracts containing PAH. At a water-bath temperature of 40 "C, this technique was found to be satisfactory for the 14 PAH determined in this work. Quantitative recoveries of the hydrocarbons were obtained unless the solution was evapor- ated to dryness. However, if the determination of the more volatile PAH, such as naphthalene, is required, a Kuderna - Danish evaporator equipped with a three-stage Snyder column should be used to minimise evaporative losses of the volatiles.Silica gel, neutral alumina and Florisil have been commonly used for the clean-up of sediment extracts containing PAH.2,11>27,28 Fully activated silica gel and neutral alumina both gave 295% recoveries of all PAH when microgram amounts of the hydrocarbons were spiked directly on to the columns. Activated Florisil also worked well for most PAH, however, as reported earlier, the satisfactory recovery of B[a]P could not be obtained on this column.11 Gel permeation chromatography with Sephadex LH-20 is also a popular approach to the clean-up of sediment extracts for PAH determination,2.20>28?29 especially for the separation of ali- phatic and aromatic hydrocarbons.In this work, clean-up of sediment extracts was carried out using a 20-g Sephadex LH-20 column with a 1 + 1 benzene - methanol elution system according to the method of Giger and Schaffner.2 However, it was found that in our work the Sephadex column did not further improve the clean-up of the EC-1 extracts after they were subjected to the silica gel and alumina columns. Table 2. Concentrations of PAH (pg g-1) in EC-1 obtained under various extraction conditions. Average of three analyses Solventsystem* . . . . A B C D A E Replicates . . . . . . 3 3 3 3 3 3 Extraction method . . Soxhlet Soxhlet Soxhlet Soxhlet Soxhlet Ultrasonic Extractiontime . . . . 8 h 8 h 8 h 8 h 24h 3x3min Phen . . . . .. . . 15.9 Anth . . . . . . . . 1.3 F . . . . . . . . 22.2 Py . . . . . . . . 17.2 B[a]A . . . . . . . . 8.5 Ghry . . . . . . . . 8.4 B[b]F . . . . . . . . 9.0 B[k]F . . . . . . . . 4.2 B[e]P . . . . . . . , 5.3 B[a]P . . . . . . . . 5.0 Pery . . . . . . . . 0.8 D[ah]A . . . . . . 1.4 B[ghi]P . . . . . . 4.3 I[cd]P . . . . . . . . 5.2 16.0 1.7 21.5 15.1 8.3 8.3 7.9 4.4 4.9 4.8 0.7 5.2 0.8 2.6 14.7 1.2 23.3 15.8 8.2 8.2 8.3 5.2 5.1 5.1 1 .o 5.3 1.5 4.8 15.9 1.3 22.7 16.8 8.0 7.5 7.8 4.0 4.6 4.9 0.8 2.9 1.3 2.9 15.2 1.5 22.4 16.2 8.2 8.7 7.8 4.6 5 .0 5.1 0.8 5.3 1.5 4.4 14.2 1 .o 21.4 16.8 10.2 7.5 8.2 3.6 4.4 4.9 0.8 5.4 1.2 4.4 * A = 41 + 59 hexane - acetone; B = 1 + 1 benzene - methanol; C = cyclohexane; D = hexane; and E = 1 + 1 hexane - acetone. Ti meimin Fig.1. gel and alumina column clean-up Total ion current chromatogram of EC-1 extract after silica loo 80 r 10 4 20 30 40 Time/m in34 ANALYST, JANUARY 1987, VOL. 112 Table 3. Mean concentrations of selected PAH (pg g-1, dry mass) in reference material EC-1. Uncertainty is one standard deviation Table 5. Certified concentration of selected PAH in reference material EC-1. Uncertainty is one standard deviation Phen . . Anth F . . Py . . BbIA Chry . . B[bIF B[kIF BkIP BbIP I[CdlP Pery . . D[ah]A B[ghi]P GC-FID GC-MSD HPLC* No. of analyses 30 12 30 . . . . . . 15.8f 1.2 16.2+ 1.5 14.9+0.4$ . . . . . . 1.3k0.3 1 . 2 f 0 . 3 0.8+0.1$ . . . . . . 22.5k2.0 23.6f 1.9 23.8k2.1 . . . . . . 16.8k 1.9 17.8-t 1.5 16.2f2.1 . . . . . . 8.5fO.9 8.7k1.0 8 . 8 f 0 . 6 . . . . .. 9.2f0.9-t 9.6f l.lt 7.920.84 . . . . . . 7.6k 1.2 8.5f0.9 8.0k0.5 . . . . . . 4.5f0.6 4.4k0.6 4.4k0.3 . . . . . . 5 . 2 f 0 . 6 5.7k0.8 5.4f0.3$ . . . . . . 5.4kO.7 5.8k0.7 5.0f0.6 . . . . . . 0.8fO.2 1.7k0.2 1.7k0.1$ . . . . . . 5.8fO.5 6.1k1.0 5.5k0.4 . . . . . . 1.3 20.2 1.5 k0.2 1.1 20.2 . . . . . . 4.6kO.7 5.4f1.0 4.9k0.4 * Results obtained by fluorescence detection except phen, anth, t Results include triphenylene. $ Results obtained by UV detection and no. of replicates was five. chry, B[e]P and pery. ~~~~ ~ Table 4. Interlaboratory results on selected PAH (pg g-l) in EC-1 No. of Parameter results* Range Median Mean f s.d. Phen . . Anth F . . Py . . B [ 4 4 BPIF WIF B[eIP BbIP I[CdlP Chry . . Pery . . D[ah]A B[ghi]P . . . . 11 9.9-24.35 . . . . 9 0.35-13.18 .. . . 15 14.87-45.3 . . , . 13 9.58-26.0 . . . . 11 4.6-15.6 . . . . 10 6.7-44 .O . . . . 11 3.68-15.2 . . . . 11 2.8-16.61 . . . . 8 3.12-7.76 . . . . 15 2.61-30.0 . . . . 7 1.05-2.19 . . . . 11 3.12-7.6 . . . . 10 1.44-11.0 . . . . 13 0.45-20.31 16.80 1 S O 21.81 18.50 7.60 8.80 6.75 3.63 5.36 4.50 1.16 4.90 2.35 4.73 16.57 f 4.59 3.92 f 4.70 23.45 f 7.39 18.42 f 5.21 8.41 f 3.04 13.70 f 11.86 8.08 k 3.64 5.58 f 4.17 5.55 f 1.50 6.58 f 6.77 1.49 f 0.48 5.10 k 1.36 3.62 f 2.35 7.32 f 6.48 * Some laboratories did not provide all the results obtained. Therefore, Sephadex and Florisil column clean-ups were not employed. At the early stage of PAH determination, a 12-m OV-1 capillary column was used. This column successfully resolved all 14 PAH of interest, including the following isomeric pairs: phen and anth, B[a]A and chry, and B[b]F and B[k]F. However, even better resolution of the PAH could be obtained by using a 30-m, thin-film DB-5 column.In the latter instance, base-line resolution was observed for all 14 PAH except B[b]F and B[k]F. This efficient column provided adequate resolution of the PAH and co-extractives in this complex EC-1 sample (Fig. 1). Initially, the PAH in EC-1 were identified by their retention times as they were chromatographed on the OV-1 and DB-5 columns. Their identities were positively confirmed by operat- ing the MSD in the scanning mode and comparing the mass spectra with authentic standards. For all 14 PAH, the match quality were better than 9800 (best match = 10000). As chrysene and triphenylene were not resolved on our GC columns and their mass spectra were very similar, we were not able to tell which one of the two, or whether a mixture of both PAH was present in EC-1.Nevertheless, this peak was quantitated against a chrysene standard in subsequent deter- minations. By monitoring only the characteristic molecular ions of the PAH, the MSD was extremely selective, as indicated by the reconstructed multi-ion current profile of an EC-1 extract (Fig. 2) versus the total ion current chromato- Concentration/ Parameter CLg g-l Phen . . . . . . 15.8 k 1.2 F . . . . . . . . 23.2 f 2.0 Py . . . . . . 16.7f2.0 B[a]A . . . . . . 8.7 f 0.8 B[b]F . . . . . . 7.9 k 0.9 B[k]F . . . . . . 4.4 f 0.5 B[e]P . . . . . . 5.3 f 0.6 B[a]P . . . . . . 5.3 k 0.7 I[cd]P .. . . . . 5.7 f 0.6 B[ghi]P.. . . . . 4.9 f 0.7 gram (Fig. 1). As the molecular ion is the most abundant ion for each PAH, SIM quantitation of these PAH was also highly sensitive. Under the conditions used the MSD was approxi- mately 100 times more sensitive than the FID in these PAH determinations. Because of availability of equipment, only nine of the 14 PAH in EC-1 were determined by HPLC. The samples were chromatographed on a reversed-phase C18 column and the nine PAH were detected by a filter fluorimeter (Table 3). Under isocratic conditions, several PAH eluted closely with each other and were not separated by modification of mobile phase composition. For example, B[a]A and chry, and B[b]F, pery and B[e]P were the two groups of co-eluting PAH. However, by operating the fluorescence detector at he, 280 and he, 389 nm, B[a]A was selectively detected in the presence of chry as the latter had very little fluorescence sensitivity at such wavelengths .7JO For similar reasons, the determination of B[b]F was not significantly interfered with by the presence of B[e]P and perylene in the same sample. Extracts of EC-1 were also determined by an independent laboratory for the 14 PAH using the gradient elution HPLC technique detailed in the US EPA Method 610.24 Under these conditions, all 14 PAH were resolved and they were quanti- tated by the fluorescence and UV detectors connected in series.The GC - FID, GC - MSD and HPLC results of the PAH in EC-1 are summarised in Table 3. Recently, we organised an interlaboratory study on the determination of PAH in sediment samples, including EC-1.22 The interlaboratory results submitted by 14 Canadian partici- pants were generated by a variety of extraction, clean-up and detection methods.Interlaboratory round-robin results alone are generally considered insufficient to certify environmental reference materials as the precision and accuracy of these results are obtained in an uncontrolled manner. Nevertheless, based on what we have learned in the 100 or more interlabora- tory studies organised by us, the interlaboratory mean or median results are usually good estimates of the true values in unknown samples. As shown in Table 4, the interlaboratory results further confirmed our in-house results as the two were in good agreement with each other.Levels for ten PAH (phen, F, Py, B[a]A, B[b]F, B[k]F, B[e]P, B[a]P, I[cd]P and B[ghi]P) in EC-1 were certified and their values are listed in Table 5. In these instances, the agreement between the in-house results using three indepen- dent detection techniques (Table 3 ) were all better than k 10%. The certified values were obtained by calculating the weighted averages of the pooled in-house results26 and the uncertainty was one standard deviation. Values for chrysene could not be ascertained because both GC methods were unable to resolve chrysene and triphenylene. The other three PAH are relatively minor components in EC-1 and the results listed in Table 3 are for information only. Currently, this sediment reference material is kept at -20 "C in the dark. No degradation of PAH in the sample hasANALYST, JANUARY 1987, VOL.112 been detected since the work was initiated approximately two years ago. In conclusion, a naturally contaminated lake sediment certified reference material was developed and certified for ten PAH at the yg 8-1 level. The certified values were derived by repetitive in-house analysis using three different methodol- ogies, i.e., GC - FID, GC - MS and HPLC techniques. These values were further confirmed by an interlaboratory study. This material is a valuable tool in the development and evaluation of analytical methods for PAH, and for the generation of accuracy statements in in-house and interlabora- tory quality assurance activities in such analysis. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Dipple, A., in Searle, C .E., Editor, “Chemical Carcinogens,” American Chemical Society, Washington, DC, 1976, ACS Monograph 173, pp. 245-314. Griger, W., and Schaffner, C . , Anal. Chem., 1978, 50, 243. Laflamme, R. E., and Hites, R. A., Geochim. Cosmochim. Acta, 1978, 42, 289. Bjarseth, A., Knutzen, J., and Skei, J., Sci. Total Environ., 1979, 13, 71. Bieri, R. H., Cueman, M. K., Sith, C. L., and Su, C . W., Int. J. Environ. Anal. Chem., 1978, 5 , 293. Lee, M. L., Novotny, M., and Bartle, K. D., “Analytical Chemistry of Polycyclic Aromatic Compounds ,” Academic Press, New York, 1981. Das, B. S., and Thomas, G. H., Anal. Chem., 1978, 50, 967. Ogan, K., Katz, E., and Slavin, W., Anal. Chem., 1979, 51, 1315. Konash, P. L., Wise, S. A., and May, W. E . , J. Liq.Chromatogr., 1981,4, 1339. Das, B. S., and Thomas, G. H., “Proceedings of the 9th Materials Research Symposium, April 10-13, 1978, Gaithers- burg, MD ,” National Bureau of Standards Special Publication 519, April 1979. Dunn, B. P., and Armour, R. J., Anal. Chem., 1980,52,2027. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 35 Joe, F. L., Jr., Salemme, J., andFazio, T., J. Assoc. OF. Anal. Chem., 1982, 65, 1395. Black, J. J., Dymerski, P. P., and Zapisek, W. F., Bull. Environ. Contam. Toxicol., 1979, 22, 278. Lee, M. L., and Wright, B. W., J. Chromatogr. Sci., 1980,18, 345. Lee, M. L., Vassilaros, D. L., and Later, D. W., Znt. J. Environ. Anal. Chem., 1982, 11, 251. BjGrseth, A., and Eklund, G., J. High Resol. Chromatogr. Chromatogr. Commun., 1979, 2 , 22. Miller, D. A., Skogerbock, K., and Grimsrud, E. P., Anal. Chem., 1981, 53, 464. Konig, J., Balfanz, E., Funcke, W., and Romanowski, T., Anal. Chem., 1983, 55, 599. Howard, A. G., and Mills, G. A., Int. J . Environ. Anal. Chem., 1983, 14, 43. MacLeod, W. D., Jr., Prohaska, P. G., Gennero, D. D., and Brown, D. W., Anal. Chem., 1982, 54,386. Hilpert, L. R., May, W. E . , Wise, S. A., Chesler, S . N., and Hertz, H. S., Anal. Chem., 1978, 50, 458. Lee, H. B., and Chau, A. S. Y., 1985, unpublished results. May, W. E., Chesler, S. N., Hertz, H. S . , and Wise, S. A., Int. J. Environ. Anal. Chem., 1982, 12, 259. US Environmental Protection Agency, “Guidelines Establish- ing Test Procedures for the Analysis of Pollutants,” Federal Register, Volume 49, No. 209, Washington, DC, Oct. 26,1984, Chau, A. S . Y., and Lee, H. B., J. Assoc. 08. Anal. Chem., 1980,63,947. Lee, H. B., Hong-You, R. L., and Chau, A. S. Y., Analyst, 1986, 111, 81. Tan, Y. L., J. Chromatogr., 1979, 176, 319. Giger, W., and Blumer, M., Anal. Chem., 1974,46, 1663. Ramos, L. S., and Prohaska, P. G., J. Chromatogr., 1981,211, 284. pp. 112-120. NOTE-Reference 25 is to Part V of this series. Paper A61208 Received June 27th, 1986 Accepted August 8th, 1986