Anal. Chem. 1981, 53, 2107-2109
2107
Deuterated Analogues as Internal Reference Compounds for the Direct Determination of Benzo[ a Ipyrenei and Perylene in Liquid Fuels by Laser-Excited Shpol’skii Spectrometry Yen Yang, Arthur P. D’Sllva, and Velmer A. Fassel* Ames Laboratory and Department of Chemistry, Iowa State Universlw, Ames, Iowa 500 1 1
In Shpol’skii effect spectrometry, poiycycllc aromatlc hydrocarbons and their deuterated analogues exhlblt adequately resolved characteristic quasi-line spectra. Thus, we have utilized deuterated benro[a]pyrene and peryiene as externally added Internal reference compounds to facilitate the direct determination of benro[alpyrene and peryiene in liquid fuels.
Polycyclic aromatic hydrocarbons (PAHs), as a compound class, are known to occur in coal liquefaction products, shale oil, and other liquid fuek. Traditionally, the PAH content of these materials has been assessed by measuring the concentration of a selected surrogate compound. Usually this compound has been benzo[a]pyrene (B[a]P), a well-known potent carcinogen. The unambiguous identification and determination of B[a]P in these highly complex mixtures is a relatively laborious analytical task, which involves considerable fractionation procedures prior to isolation and quantitation of B[a]P by gas-liquid or high-performance liquid chroma tography (I), We have recently reported on the potential of laser-excited Shpol’skii spectrometry (LESS)for the analysis of complex mixtures of alkylated PAHs (2), a task often found difficult by high-resolution chromatographic techniques such as capillary column gas chromatography-mass spectrometry. We have also demonstrated that LESS may be used for the direct determination of several PAHs, including B[a]P, in a coal liquid (SRC-11) and a shale oil sample (3). In these studies, the standard addition approach was employed for the quantitation of trace levels of P.AHs to compensate for any potential luminescence quenching or enhancement resulting from intermolecular effects. Although the standard addition procedure has consistently provided accurate analytical data, the necessity of establishing an analytical curve for each analyte and for each sample renders this approach impractical for routine analyses. In this paper, we report on an alternate but relatively simple procedure to obtain unambiguous quantitative data, by employing completely deuterated PAHs as the internal reference compounds. In principle, luminescence quenching or enhancement effects may be internally compensated by comparing the analyte luminescence to an internal reference compound that responds to these effects to the same degree as the anal@. From a consideration of first principles (see Selection Criteria discussion below), deuterated analogues of the analytes added a t constant concentration levels to each sample should therefore serve as virtually ideal internal reference compounds. In this paper, we demonstrate the efficacy of this approach for the quantitative determination of benzoIa1pyrene and perylene.
APPARATUS AND PROCEDURES Instrumentation. The experimental arrangement has been described previously (2). A brief description of the present instrumental features is given here. An excimer laser with an
approximate 10-ns pulse width was used to excite a tunable dye laser. The excimer laser used a mixture of a rare gas and a halogen as the active medium. The choice of gas mixture determined the output wavelength. In the present study, XeCl lasing at 308 nm was the active medium. This medium was obtained by mixing 60 mbar of xenon, 80 mbar of 5% hydrogen chloride in helium, and 1110 mbar of helium (as the buffer gas). Both the Xe and HCl gases were of ultra high purity (>99.99%) and were obtained from Cryogenic Rare Gas Laboratories, Inc., Newark, NJ. Tho dye laser bandwidth was generally 0.01-0.03 nm. The dye laser beam was focused onto the sample contained in a sample holder which has been described in our earlier communication (2). The luminescence from the sample at 15 K was dispersed with a 0.64-m monochromator. The luminescence was then detected with 13 photomultiplier and the signal was processed by a boxcar averager, which provided the output for a strip-chart recorder. The experimental facilities and operating conditions utilized in the present investigation are summarized in Table I. Reagents and Samples. Benzo[a]pyrene and perylene (99+% purity) and n-octane (99+%) were obtained from Aldrich Chemical, Inc., Milwaukee, WI. Deuterated benzo[a]pyrene, B[a]P-d12,and perylerie, perylene-dlz (>97 atom % D isotope purity), were obtained from Merck and Co., Inc./Isotopes, St,. Louis, MO. The solvent-refined coal liquid (SRC-II), shale oil, and Wilmington petroleum crude oil samples were obtained from thie National Bureau of Standards (NBS), Washington, DC. Thie coal-derived fuel oil blend, no. 1701, and diesel fuel marine produce, no. 4610, were obtained from U.S.EPA/DOE Research Materials Repository, Oak Ridge National Laboratory (ORNL), Oak Ridge, TN. Procedures. The samples were spiked with the previously established reference amount of B[a]P-dlz and perylene-dlz,also in n-octane solutions, and diluted with n-octane by various factors as follows: SRC-II,5 X lo3;shale oil, lo3;Wilmington crude, loe’; coal-derived fuel oil blend, 5 X lo3; and shale-derived diesel fuel, 50. These dilution factors were chosen so that the analyte B[a]P and perylene concentration in the diluted samples would fall within the linear range of the analytical calibration curves. Usually, one trial dilution was required because of the uncertainty in the and@ concentration originally present in the sample. The dilution factor of 50 used for shale-derived diesel fuel was thle minimum factor required to avoid spectral line-broadening observed for sample diluted to a lesser extent (3). The amounts of added inkrnrjJreference B[a]P-dlz and perylene-d12in n-octane solution were controlled volumetrically so that the concentrations of B[a]P-dlz and perylene-dlz in the diluted samples were alwayis 10 and 160 ppb, respectively. At such concentrations no luminescence from any residual undeuterated compound was observed. The diluted sample solutionswere agitated, with a mechanical shaker, at room temperature for at least 10 min and then left unagitated for 2 h, agitated again for 10 min, and then left unagitated for several haws, usually overnight, to allow trace noctane-insoluble asphalbnes to settle to the bottom of the sample solution flask. No attempt was made to isolate the insoluble precipitates from the solution; no asphaltene precipitate was visually observed on dilution of the diesel fuel sample by an:y factor. A clear aliquot of the dilute sample was injected with syringe into the copper sample holder for spectroscopic investigations.
0003-2700/81/0353-2107$01.25/00 1981 Amerlcan Chemical Society
2108
ANALYTICAL CHEMISTRY, VOL. 53, NO. 13, NOVEMBER 1981
Table I. Experimental Facilities and Operating Conditions A. Laser System excimer multigas laser Model EMG101, Lambda Physik (West Germany); operated with XeCl (308 nm) at power level of -2 MW and repetition rate of 30 pulses/s dye laser Model FL 2000, Lambda Physik (West Germany); operated with BBQ dye in cyclohexane and stilbene 3 in ethanol dye laser scan controller Model FL 500, Lambda Physik (West Germany)
B. Spectrometric System monochromator Model HR-640, Instruments SA, Inc. (Metuchen, NJ); a 0.64-m scanning spectrometer, operated at a spectral band-pass of 0.1 nm monochromator Model 980015, Instruments SA, stepping motor Inc. (Metuchen, NJ) controller optical transfer Luminescence focused by a 100 mm focal length X 76.2 mm diameter plano-convex, fused silica lens; positioned at twice the focal length from the entrance slit and sample surface center C. Detector and Signal Processing photomultiplier Type R955 (S-20response), operated at a voltage of 1kV HV power supply Model 110, Pacific Photometric Instrument (Emeryville, CA) boxcar averager Model 162/164/165, Princeton Applied Research (Princeton, NJ); operated with a 100-ns aperture time, 1ps gate time constant and 0.1 S signal processing time constant strip chart recorder Model 5210-5, Houston Instruments (Austin, TX) D. Refrigeration System cryogenic refrigeration Model CSW-202, Air Products and Chemicals, Inc. (Allentown, PA) ~~
~~
~~
SELECTION CRITERIA OF AN INTERNAL REFERENCE COMPOUND I N LESS The normal requirements of an effective internal reference element in analytical atomic spectrometry are well-known. In the analysis of complex organic matrices such as coal liquids and shale oil utilizing LESS, the following criteria for the selection of the internal reference compound and its mode of usage are considered important: (1) Its spectroscopic properties should be as similar as possible to those of the analyte being determined so that the internal reference compounds will provide adequate compensation for the emission variation associated with intermolecular interactions and inner filter and enhancement effects. (2) It should be absent in the sample. (3) It should be in a high state of purity with respect to the analyte being determined. (4)The spectral lines of the internal reference and analyte compound of interest should be free of spectral interferences arising from emissions of other sample constitutents. (5) The spectral lines selected should be free of self-absorption. (6) It is preferable to employ the same excitation wavelength for the reference and analyte, whenever possible.
t
I
I
400
410
I 420
I
430
WAVELENGTH (nm)
Figure 1. Fluorescence spectrum of B[a]P-dlp in n-octane, A,, 379.5 nm, superimposed on the referencespectrum of B[a]P, A, 380.2 nm.
I 4 40
I
I
450
460
I
470 WAVELENGTH l n m )
=
=
I
480
Figure 2. Fluorescence spectrum of peryiene-d12in n-octane, A, = 417.7 nm, superimposed on the reference spectrum of perylene, A,, = 420.3 nm.
Deuterated organic compounds are known to have similar chemical and physical properties as their undeuterated analogues. Because they are available commercially with high purity and are unlikely to be present in “real” samples, these compounds satisfy the stated criteria to an unusually high degree.
RESULTS AND DISCUSSION Isotopic Effects on t h e Shpol’skii-Effect Spectra. Spectral line shifts resulting from isotopically enriched solute and solvent have often been reported for Shpol’skii systems (4-7). Observations on the deuteration effect on the luminescence spectrum of pyrene dissolved in n-pentane and frozen to a solid at 10 K revealed that similar excitation and emission spectra are observed for pyrene and pyrene-dlo and that the spectral line shift between the 0-0 lines was 100 cm-l (4). In the present work, similar spectral shifts were observed for deuterated PAHs. Figure 1shows the observed fluorescence spectrum of B[a]P-d12in n-octane, superimposed on the reference spectrum of B[a]P. It is seen in this figure that B[a]P-d12,as well as B[a]P, exhibit quasi-linear features (fwhm 0.1 nm). The effect of deuterium on the fluorescence of B[a]P resulted in a 1nm (-55 cm-’) blue shift in the 0-0 line, Similar spectral line shifts were also observed for perylene-d12as shown in Figure 2. Such spectrometric properties of deuterated PAH are desirable for an internal reference compound in the determination of PAHs in complex matrices. The proximity of the analytical line pair ensures that both the analyte and internal reference line intensities will vary
-
-
ANALYTICAL CHEMISTRY, VOL. 53, NO. 13, NOVEMBER 1981 2109
Table 11. Concentrations of B[a]P and Perylene in Liquid Fuels (pg/g)
B La3 P
1
standard internal addition reference method method
sample SRC-I1 shale oil Wilmington crude coal-derived fuel oil shale-derived diesel fuel
I3[a]P Concentrations 145 163 13 15 3 88
0.056
Perylene Concentrations SRC-I1 29 Wilmington crude 40 Reference 8. 400
410
430
420
WAVELENGTH ( n r n l
ence.
PERY L E N E
L1I
I J l L ld1* 4 ! L L t ~ b L & J l J L ~
LA 440
450
1 460
I 470
ref data
134/12ga 21a 2;a2.512.7 82.3191.8 0.030/0.045
26’ 38‘
NBS data.
440
Flgure 3. Selectively excited fluorescence spectrum of B[a]P in a shale oil sample with 10 ppb B[a]P-dlpadded as the Internal refer-
I
Reference 1.
-
I 480
WAVELENGTH [nm)
Figure 4. Selectively excited fluorescence spectrum of perylene in a Wilmington crude oil sample wkh 160 ppb perylene-d,, added as the
internal reference.
to a similar extent by the “inner filter and enhancement effects”, because the absorption spectrum of liquid fuel would be expected to be broad and structureless, i.e., the extinction coefficient of a complex mixture remains approximately constant over a small wavelength interval. Analytical Studies. To evaluate the feasibility of the internal reference approach utilizing deuterated PAH analogues, we examined oil samples of widely differing B[aJP contents and complexities. These samples included a petroleum crude, two coal-derived fuels, and two shale-derived crude oils. The analytical calibration curve for the determination of B[a]P in the concentration range of 0.1-100 ppb was linear. For this analytical curve, the 0-0 lines for B[a]P at 402 nm and B[a]P-d12at 403 nm, the most intense lines shown in Figure 1were selected as the analytical line pair in obtaining the data. Typical fluorescence spectra of B[a]P and perylene and their deuterated analogues in the liquid fuel sample spiked with 10 ppb B[a]P-d12and 160 ppb perylene-d12 are shown in Figures 3 and 4. The same excitation wavelengths, 380.2 nm (Figure 3) and 420.3 nm (Figure 4), were utilized for excitation of both analyte and reference com-
pounds as were done for establishing the analytical calibration curves. It is seen in Figures 3 and 4 that both analytical lines are free of spectral interferences. In Table 11, the analytical results on the five samples are compared with values obtained by the standard addition method (3) and the values reported by other investigators who employed independent analytical methodologies ( I , 8). The analytical methodologies employed for obtaining the reference values have generally involved considerable fractionation and separation procedures. For example, in Tomkins’s method ( I ) , which required a minimum of 3 man-days to determhe the B[a]P content of one sample, a B[a]P tracer, [7,10-14C]B[a]P, was added to the samples to evaluate the B[a]P losses throughout the isolation procedures, and the quantitative data were corrected for the B[a]P recovery factor. Thus,Tomkins’s data were probably the most reliable ones available to date. A comparison of the data obtained by the internal reference approach with value!, obtained by other methodologies reveals a level of agreement that is considered excellent in terms of present capabilities. Because the deuterated analogues of the analytes satisfy the operating criteria of an internal reference compound to such a high degree, the direct determination of selected PAHs in complex mixtures in a rapid and simple fashion is portended by the results shown in Table 11. In contrast to other methodologies, only a sample dilution stlep, the addition of the internal reference compound, and cooling to 15 K are requiredl prior to the spectrometric observations. The total elapsed time for an analysis is approximately 1h. LITERATURE CITED Tomklns, B. A.; KiJbOta, H.; Griest, W. H.; Caton, J. E.; Clark, B. R.; Guerln, M. R. Ana!. Chem. 1980, 52, 1332. Yang, Y.; D’Sika, A. P.; Fassel, V. A. Anal. Cbem. 1981, 53, 894. Yang, Y.; D’Sihra, A. P.; Fassel, V. A.; Iles, M. Anal. Cbem. 1980, 52, 1350. Cunningham, K.; Slebrand, W.; Wllllams. D. F. Chem. fbys. Lett.
1973,20,496. Meyer, 8.; Metzgsr, J. L. Specfrochim. Acta, Part A 1972, d8A, 15R2
ieiii,G. L.; Laposa, J. D. J. MOL Spectrosc. 1972, 24, 1249.
Schettino, V. J. Mol. Spectrosc. 1970, 34, 78. Hertz, H. S.: Brown, J. M.; Cheder, S. N.; Guenther, F. R.; Hilpert, L. R.; May, W. E.; Parris, R. M.; Wise, S. A. Anal. Cbem. 1980, 52, 1650.
RE~F,IVED for review June 11,1981. Accepted August 14,19181. This research was supported by the US.Department of E h ergy, Contract No. W-7405-Eng-82, Office of Health and Environmental Research, Physical and Technological Studies, Budget Code GK-01-02-04-3.
