- Email: [email protected]
Photo-oxidation of biodegraded crude oil and toxicity of the photo-oxidized products
Chemosphere 44 (2001) 1145±1151
www.elsevier.com/locate/chemosphere
Photo-oxidation of biodegraded crude oil and toxicity of the photo-oxidized products H. Maki, T. Sasaki, S. Harayama
*
Marine Biotechnology Institute, Kamaishi Laboratories, 3-75-1 Heita, Kamaishi, Iwate 026-0001, Japan Received 15 March 2000; accepted 27 June 2000
Abstract We investigated the physicochemical changes resulting from irradiation by sunlight of biodegraded crude oil. An Arabian light crude oil sample was ®rst subjected to microbial degradation. n-Alkanes and aromatic compounds such as naphthalenes, ¯uorenes, dibenzothiophenes and phenanthrenes possessing short, alkyl side chain(s) were almost completely degraded, while the contents of the saturated and aromatic fractions were reduced by 70% and 40%, respectively. This biodegraded oil was then suspended in seawater and exposed to sunlight irradiation for several weeks. The most remarkable change caused by the irradiation was a substantial decline in the aromatic fraction with a concomitant increase in the resin and asphaltene fractions. A 13 C-nuclear magnetic resonance (NMR) spectroscopic analysis showed that the aromaticity of the biodegraded oil was signi®cantly lower in the irradiated sample. A ®eld desorption±mass spectrometric (FD±MS) analysis showed that sunlight irradiation reduced the average molecular weight of the oil components and formed oxygenated compounds. Consistent with this observation is that the oxygen content in the oil increased as the irradiation was prolonged. The bioavailability of the biodegraded oil was increased by the photo-oxidation: the growth of seawater microbes was minimal when the non-irradiated biodegraded oil was used as the source of carbon and energy; however, growth was signi®cant when irradiated biodegraded oil was used. The concentration of dissolved organic carbon (DOC) increased linearly during the sunlight irradiation of the biodegraded oil, and this increase was matched by an increase in ultraviolet-absorptive materials in the seawater. The photochemically formed, water-soluble fraction (WSF) showed acute toxicity against the halophilic crustacean, Artemia. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Crude oil; Biodegradation; Photo-oxidation; Bioavailability; Toxicity
1. Introduction A vast amount of mineral oil is being released into the ocean by both accidental oil spills and as a result of human activity on land, this chronic annual discharge of oil into the ocean being estimated as approximately 320
*
Corresponding author. Present address: National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki 305-0053, Japan. Tel.: +81-193-26-6544; fax: +81-193-26-6584. E-mail address: [email protected] (S. Harayama).
million tons (Etkin et al., 1998). The discharged oil is removed by various natural factors, i.e., volatilization, dispersion, precipitation, biodegradation and photooxidation (Wolfe et al., 1994). Microbial degradation has been applied to the removal of spilled oils in the marine environment ever since the serious tanker accident to the Exxon-Valdez occurred in Alaska in 1989 (Pritchard and Costa, 1991; Bragg et al., 1994; Swannell et al., 1996). However, the natural biodegradability of crude oil is limited. In laboratory experiments, the speed of the biodegradation of crude oil has been shown to diminish sharply after the petroleum components have been degraded by 30% (Maki et al., 1997). Besides the
0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 2 9 2 - 7
1146
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
biological action, sunlight irradiation is another eect which alters the physicochemical properties of crude oil in the natural environment. It is possible that crude oil components, which are recalcitrant to biodegradation are degraded by sunlight irradiation. Although many reports have so far been published characterizing the photo-chemistry of crude oil, they have focused on the physicochemical changes to intact crude oil, rather than to biodegraded crude oil (Larson et al., 1977, 1979; Tjessem and Aaberg, 1983; Payne and Phillips, 1985; éstgaard et al., 1987; Jacquot et al., 1996; Garrett et al., 1998; Nicodem et al., 1998). Moreover, these studies mainly employed gas chromatography (GC) and/or combined GC and mass spectrometry (GC±MS) for component analyses, despite the fact that GC methods can only detect the volatile components of crude oil (Poston et al., 1988; Jacquot et al., 1996; Nicodem et al., 1998). The GC-detectable compounds in crude oil are generally susceptible to biodegradation (Ishihara et al., 1995; Maki et al., 1997). In other words, the photochemical alteration of crude oil components that are resistant to biodegradation cannot be characterized by the GC techniques. Thin layer chromatography coupled with ¯ame ionization detector (TLC±FID) has been employed to evaluate the biodegradability of total petroleum including the non-volatile fraction in crude oil (Goto et al., 1994; Ishihara et al., 1995; Maki et al., 1997; Sugiura et al., 1997). In this present study, we applied TLC±FID, 13 C-nuclear magnetic resonance spectroscopy (NMR) and ®eld desorption±mass spectrometry (FD±MS) to survey the photo-chemical alteration of components in an already biodegraded crude oil sample. The water-soluble fraction (WSF) of crude oil is known to be signi®cantly increased by photo-oxidation (Payne and Phillips, 1985; Sydnes et al., 1985; éstgaard et al., 1987; Poston et al., 1988; Ali et al., 1995), and considerable attention has been paid to the ecological eect of photo-chemically generated WSF because it is known to exert signi®cant toxicity against aquatic organisms (Schneider and Gominger, 1976; Sydnes et al., 1985; éstgaard et al., 1987; Poston et al., 1988). However, there has been little work done on WSF derived from biodegraded crude oil. In this study, we characterize WSF, which has been formed by the sunlight irradiation of a biodegraded crude oil sample. 2. Materials and methods 2.1. Preparation of the biodegraded and photo-oxidized crude oil samples The oil used was Arabian light crude that had been heated at 230°C to evaporate the volatile fractions. The oil thus treated is hereafter called ``weathered crude oil''.
When analyzed by silica-gel column chromatography according to the method of the Japan Petroleum Institute (JPI, 1983), it was found to contain 61% of saturated hydrocarbons (the n-hexane eluate), 33% of aromatic hydrocarbons (the n-hexane:benzene 1:1 eluate), 4% of resins (the methanol eluate), and 2% of asphaltenes (the chloroform eluate). Chloroform (2 ml) containing 0.5 g ml 1 of the weathered crude oil was poured into each of several 2-l Sakaguchi ¯asks (a conical ¯ask used for microbial cultivation), and the chloroform was evaporated by heating the ¯asks overnight at 50°C. Fresh seawater (500 ml) was then added to each ¯ask (the ®nal oil concentration was 2 g l 1 ). 3 g of a slow-release solid, granular nitrogen fertilizer (Super IB; Mitsubishi Chemicals, Tokyo, Japan) and 0.5 g of a phosphorous fertilizer (Linstar 30, Mitsubishi chemicals) were supplemented to the seawater medium in each ¯ask. After adding these fertilizers, the ¯asks were agitated by a reciprocating shaker at 120 rpm and 25°C for 3 weeks to grow the indigenous microbes in seawater on the weathered crude oil. After the cultivation, the biodegraded oil in the medium was extracted with dichloromethane (DCM), the excess DCM being removed from the extract by using a rotary evaporator. The extracted oil was weighed and then dissolved in chloroform. Aliquots of the chloroform solution, each containing 30 mg of biodegraded oil, were poured into 100-ml Erlenmeyer ¯asks. After the chloroform had been evaporated, 30 ml of seawater was added to each ¯ask. The ®nal concentration of the biodegraded oil was 1 g l 1 . These ¯asks were capped with aluminum foil, moved to the roof of a building and agitated at 80 rpm by a rotary shaker. The ¯asks were exposed to sunlight for 3 weeks, the sunlight intensity in the daytime being in the range of 564±2 800 mol m 2 s 1 . No light with a wavelength shorter than 300 nm passed through the ¯ask glass. Non-irradiated controls were provided by the Erlenmeyer ¯asks containing seawater and the biodegraded oil, which were wrapped in aluminum foil and were incubated for 3 weeks. One ¯ask was used each week to extract the oil as already described. To sample the WSF, 18-ml aliquots of seawater were taken from each ¯ask, and the WSF samples were stored at 4°C until needed for the chemical analyses and the Artemia bioassay. 2.2. Analyses The extracted oil samples were dried, dissolved in a determined volume of chloroform, and then subjected to TLC±FID (Iatroscan MK-5; Iatron, Tokyo, Japan) (Goto et al., 1994). The chloroform solution (1 ll) was applied to one end of a silica-gel rod (Chromarods SIII, Iatron) and developed in three steps, ®rst with n-hexane over 10 cm, second with n-hexane:toluene (20:80 in volume) over 5 cm, and ®nally with methanol:DCM (5:95 in volume) over 2 cm. The amounts of hydrocar-
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
bons thus separated were determined by FID. This method separated the weathered crude oil into four peaks. We have called these four peaks as the saturated, aromatic, resin and asphaltene fractions in our previous works (Goto et al., 1994; Ishihara et al., 1995; Maki et al., 1997; Sugiura et al., 1997). Although we use the same nomenclature in this report, we have recently found that the saturated fraction contains not only saturates, but also alkylaromatics with long alkyl side chains (Dutta et al., manuscript in preparation). The oil samples were also subjected to an analysis by combined GC±MS (GC/MS-QP5000, Shimadzu, Kyoto, Japan) to quantify the series of n-alkanes (C13±33 ), C1±4 alkylnaphthalenes, C0±4 -alkyl¯uorenes, C0±4 -alkyldibenzothiophenes, C0±7 -alkylphenanthrenes (C0 indicates a non-substituted compound), and the biomarker, 17a(H), 21b(H)-hopane. The GC±MS analysis was conducted by the method of Wang et al. (1994). All values obtained by the instrumental analyses were normalized to that of 17a(H), 21b(H)-hopane (Prince et al., 1994). FD±MS was performed with a JMS-700 MStation equipped with a silicon emitter (JEOL, Tokyo, Japan) under the following conditions: emitter current, 0±40 mA; rate, 4 mA min 1 ; cathode voltage, 8.0 kV; scanning range, 50±3 000 m/z. 13 C NMR spectra were recorded by a Unity 300 spectrometer (Varian, Palo Alto, CA, USA) at 300 MHz, the intensity of each chemical shift being determined relative to internal tetramethylsilane in CDCl3 (Wako Pure Chemicals, Tokyo, Japan). The aromaticity is de®ned as the ratio of aromatic carbon to total carbon. The amount of aromatic carbon was obtained from the integrated intensity of these peaks between 100 and 160 rpm in the 13 C-NMR spectrum, while that of aliphatic carbon was found from the integrated intensity of the peaks between 4 and 70 ppm (Sugiura et al., 1997). The aromaticity index (fa) was calculated according to the equation fa (amount of aromatic carbons)/(amount of aromatic and aliphatic carbons). An elemental analysis to determine the oxygen content was conducted with a CHN instrument (MT-5; Yanako Analytical Instruments, Kyoto, Japan) equipped with a platinum column. Combustion was for 5 min at 750°C, and the bridge current for the detector was set at 100 mA. D L -alanine was used as the standard. The WSF adsorption spectrum was obtained by a Shimadzu UV2200 Ultraviolet-visible spectrophotometer. Dissolved organic carbon (DOC) was determined by a total organic carbon meter (TOC-5000, Shimadzu) equipped with an auto-sampler. 2.3. Artemia assay Artemia larvae were obtained from the eggs that are commercially available as a feed for tropical ®sh in a domestic aquarium. The Artemia eggs were incubated overnight in fresh seawater at 25°C. The hatched Art-
1147
emia larvae were recovered by a glass pipette and transferred into another vessel ®lled with fresh seawater. The recovered Artemia larvae were incubated overnight at 25°C and then used for an acute toxicity assay of WSF that had been formed by the photo-oxidation of the biodegraded oil. The Artemia assay was performed on 6-well plastic plates. A 5-ml aliquot of the WSF sample was poured into each well, and then 5 or 6 Artemia larvae were placed in each well. The number of dead Artemia larvae in each well was counted after incubating at 25°C for 24 h. 3. Results and discussion 3.1. Preparation of the biodegraded crude oil The weathered Arabian light crude oil was biodegraded by indigenous marine microbes at 25°C for 3 weeks. A 90±100% loss of n-alkanes (C13±33 ), C1±4 -alkylnaphthalenes, C0±4 -alkyl¯uorenes, C0±4 -alkyldibenzothiophenes and C0±7 -alkylphenanthrenes was identi®ed by the GC±MS analysis, while 40% and 70% reductions of the aromatic and saturated fractions, respectively, were detected by TLC±FID. The extent of biodegradation was very similar to that observed previously (Maki et al., 1997; Sugiura et al., 1997). The results of the 13 C-NMR analysis show that the total aromaticity of the biodegraded crude oil was 0.193, while that of the weathered (non-biodegraded) crude oil was 0.08. Thus, the aromatic components were enriched in the biodegraded oil as a result of the preferential biodegradation of the aliphatic components in the weathered crude oil. 3.2. Photo-oxidation of the biodegraded oil The biodegraded oil was suspended in seawater and exposed to sunlight for 3 weeks. The oil recovery was around 100% from the samples after 0 and 7 days, while it was 61% and 68% from the samples after 14 and 28 days, respectively. The oil recovery from the dark controls, in contrast, was always around 100% even after incubating for 28 days. During this period, no signi®cant biodegradation of the oil was expected to occur for two reasons. First, no nutrient was supplemented to the seawater during the period of sunlight irradiation. Our previous experiments showed no signi®cant growth of microorganisms and hence, no signi®cant biodegradation of weathered crude oil in natural seawater (Ishihara et al., 1995; Maki et al., 1997). Second, most of the biodegradable components in the weathered crude oil had been decomposed within 3 weeks, and further growth in a batch culture did not signi®cantly increase the degree of biodegradation (our unpublished observations). Thus, under prolonged sunlight irradiation,
1148
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
part of the oil seemed to have been transformed into a form which could not be extracted by DCM. 3.3. Characterization of the biodegraded oil after photooxidation When the DCM-extracted oil was analyzed by TLC± FID, the most remarkable change observed after irradiation by sunlight was a substantial decline in the aromatic fraction, and a signi®cant increase in the resin and asphaltene fractions (Fig. 1). The results of the 13 CNMR analysis also demonstrate the reduction in aromatic carbons by the sunlight irradiation: the chemical shift between 100 and 170 ppm in the 13 C-NMR spectrum detected the aromatic carbons, the intensity of such signals getting smaller as the photo-oxidation proceeded (Figs. 2(a) and (b)). The calculated aromaticity on the basis of the 13 C-NMR spectrum was plotted against the irradiation time (Fig. 2(c)). The decline in aromaticity determined by 13 C-NMR, however, was not matched by the abundance of the aromatic fraction determined by TLC±FID. The decrease in the aromatic fraction measured by TLC±FID exceeded 90% (Fig. 1(b)) in the ®rst week, in contrast to only a 37% decrease in aromaticity in the ®rst week as determined by 13 CNMR. This dierence could be explained by assuming that most of the aromatic compounds were converted to a polar form (i.e., moved to the resin or asphaltene fraction) either by cleavage of the aromatic rings, or by the introduction of polar groups such as hydroxyl and carbonyl, but that more than half of the aromatic rings survived after sunlight irradiation for 3 weeks.
A smaller but signi®cant decrease in the amount of the saturated fraction was observed. Saturates do not absorb light above 220 nm, and they are therefore relatively inert to photo-oxidation, although they could suer sensitized photo-oxidation (Rontani, 1997). The observed decrease in the saturated fraction, however, was not due to sensitized photo-oxidation of the saturated compounds, but instead due to the photo-oxidation of alkylaromatics with long alkyl side chains that were fractionated in the saturated fraction (Dutta et al., manuscript in preparation). Thus, the aromatic compounds present in the saturated and aromatic fractions were photo-chemically transformed into more polar compounds that migrated to the resin and/or asphaltene fractions. These aromatic compounds would be transformed into ones containing oxygen such as phenols, alcohols, ketones, aldehydes and carboxylic acids (Larson et al., 1977, 1979; Payne and Phillips, 1985; Poston et al., 1988). As shown in Fig. 3, the oxygen content was increased by up to 4.3-fold in the sample that had been photo-oxidized for 3 weeks. To evaluate the molecular weight distribution of the biodegraded oil and its photo-oxidized samples, we applied an FD±MS analysis. As shown in Fig. 4, the molecular weight distribution was shifted to a lower range after the photo-oxidation: the mean molecular weight (Mw ) of the biodegraded oil without sunlight irradiation was 943, whilst that of the sample sunlight-irradiated for 21 days was 575. In the spectrogram of the photo-oxidized sample, a distinct series of peaks became evident around m/z 300±700, in which the m/z dierence between the neighboring peaks was 16, corresponding to the
Fig. 1. Light-induced compositional change of the biodegraded oil determined by TLC±FID. The areas of the four fractions, saturated, aromatic, resin and asphaltene, on TLC±FID chromatograms (a) were determined, and the proportional area of each fraction is presented (b).
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
1149
Fig. 3. Increase in the oxygen content of oil exposed to sunlight irradiation.
Fig. 2. 13 C-NMR spectra of biodegraded oil without sunlight irradiation (a), and with sunlight irradiation for 21 days (b). The aromaticity of the samples before and 14 and 21 days after the sunlight irradiation is also shown (c).
mass of the oxygen atom. This series of peaks might represent multiplied oxygenated derivatives of the parent compounds. 3.4. Bioavailability of the oil subjected to both biodegradation and photo-oxidation The bioavailability of the photo-oxidized products generated from the biodegraded oil was investigated. The biodegraded oil samples with and without being
Fig. 4. Field desorption±mass spectra of biodegraded oil not exposed (a), and exposed to sunlight irradiation for 21 days (b).
exposed to sunlight irradiation were extracted by DCM, and each extract was dried in an Erlenmeyer ¯ask into which a seawater-based medium supplemented with the nitrogen and phosphorus fertilizers was then added to a concentration of 1 g of oil per litre. Each ¯ask was incubated at 25°C. The heterotrophic bacteria indigenous to seawater grew on the biodegraded oil to the concentration of 1:0 107 cfu ml 1 in 10 days, while they grew to the concentration of 6:6 107 cfu ml 1 on the
1150
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
biodegraded and photo)oxidized oil sample. This observation implies that the photo-oxidation of the biodegraded oil could have produced some compounds, which were readily available to microbes in the seawater. 3.5. Formation of the WSF by photo-oxidation of the biodegraded crude oil and its toxicity The seawater was recovered from the sunlight-irradiated samples, and WSF in the seawater was characterized. Both the irradiated and dark-control ¯asks were prepared in triplicate. As shown in Fig. 5, ultravioletabsorbing material was formed in the irradiated-oil seawater as photo-oxidation proceeded, whereas almost no absorption was apparent in the non-irradiated dark controls. The concentration of DOC in the irradiated-oil seawater linearly increased with irradiation time, in contrast to no signi®cant increase in DOC in the dark controls (Fig. 5). The ®nal DOC concentration of WSF was 134 mg l 1 . Since the carbon content in the original crude oil sample was ca. 85%, approximately 16% of the biodegraded crude oil had been converted to WSF by the photo-oxidation process. WSF that had been formed by photo-oxidation of the biodegraded oil was also evaluated for its acute toxicity against the halophilic crustacean, Artemia. As shown in Fig. 6, the toxicity of WSF against Artemia increased as the photo-oxidation proceeded, all the Artemia larvae dying when seawater from the samples that had been irradiated for 25 days was used. The dark control showed hardly any signi®cant increase in toxicity against Artemia. These observations indicate that crude oil remains capable of releasing some toxic materials into seawater by photo-oxidation, even after it
Fig. 6. Acute toxicity against Artemia of WSF formed by the photo-oxidation of biodegraded oil. Irradiated (s), dark control (e).
has been extensively biodegraded. Although the biodegradability of WSF formed by the photo-oxidation of crude oil remains to be elucidated, it is expected to be resistant to microbial degradation because of its toxicity not only against crustaceans but also against microbes (Larson et al., 1977, 1979). There is an increasing interest in the natural attenuation of spilled oil as the cost associated with active cleanup eorts are rather large. The selection between cost-eective natural attenuation and man-made remedial engineering may be determined by various factors, but a full characterization of the site contaminated by oil, including an evaluation of the speed of natural attenuation, is required before the implementation of any remedial techniques. The results obtained in this study will provide some insights into evaluating the physicochemical changes and ecotoxicological eects of spilled oil after being subjected to intrinsic and/or man-stimulated bioremediation. Acknowledgements The authors are indebted to Mr. Etsuro Sasaki and Ms. Kaori Sasaki for their technical assistance, and to Mr. Kazutetsu Nojima of JEOL in Tokyo for the FD± MS analysis. This work was supported by the New Energy and Industrial Technology Development Organization (NEDO).
Fig. 5. Time-course plots for the increase in absorbance at 218 nm (A218 ) and the DOC concentration of WSF in seawater after exposing suspended biodegraded oil to sunlight irradiation. A218 in irradiated seawater (d), DOC in irradiated seawater (h), A218 in dark control seawater (´), DOC in dark control seawater (s).
References Ali, L.N., Fauzi, R.F., Mantoura, C., Rowland, S.J., 1995. The dissolution and photo-degradation of Kuwaiti crude oil in seawater, part 1: Quantitative dissolution and analysis of the seawater-soluble fraction. Mar. Environ. Res. 40, 1±17.
H. Maki et al. / Chemosphere 44 (2001) 1145±1151 Bragg, J.R., Prince, R.C., Harner, E.J., Atlas, R.M., 1994. Eectiveness of bioremediation for the Exxon Valdez oil spill. Nature 368, 413±418. Etkin, D.S., Wells, P., Nauke, M., Campbell, J., Grey, C., Koefoed, J., Meyer, T., Reddy, S., 1998. In: Proceedings of the 21st Arctic and Marine Oil Spill Program Technical Seminar. Environment Canada, Edmonton, Alberta, pp. 903±910. Garrett, R.M., Pickering, I.J., Haith, C.E, Prince, R.C, 1998. Photo-oxidation of crude oils. Environ. Sci. Technol. 32, 3719±3723. Goto, M., Kato, M., Asaumi, M., Shirai, K., Venkateswaran, K., 1994. TLC±FID method for evaluation of the crude-oildegrading capability of marine microorganisms. J. Mar. Biotechnol. 2, 45±50. Ishihara, M., Sugiura, K., Asaumi, M., Goto, M., Sasaki, E., Harayama, S., 1995. Oil degradation in microcosms and mesocosms. In: Microbial Process Bioremediation, vol. 3. Battelle Press, pp. 101±116. Jacquot, F., Guiliano, M., Doumenq, P., Munoz, D., Mille, G., 1996. In vitro photo-oxidation of crude oil maltenic fractions: Evolution of fossil biomarker and polycyclic aromatic hydrocarbons. Chemosphere 33, 671±681. Japan Petroleum Institute (JPI) standard, 1983. Analytical method for asphalt by column chromatography (JPU-5S22-83). JPI, Tokyo, Japan (in Japanese). Larson, R.A., Bott, T.L., Hunt, L.L., Rogenmuser, K., 1979. Photo-oxidation products of fuel oil and their antimicrobial activity. Environ. Sci. Technol. 13, 965±969. Larson, R.A., Hunt, L.L., Blankenship, D.W., 1977. Formation of toxic products from a #2 fuel oil by photo-oxidation. Environ. Sci. Technol. 11, 492±496. Maki, H., Ishihara, M., Harayama, S., 1997. Application of wastewater sludge to microbial degradation of crude oil. In: In Situ and On-site Bioremediation, vol. 4. Battelle Press, pp. 371±375. Nicodem, D.E., Guedes, C.L.B., Correa, R.J., 1998. Photochemisry of petroleum I. Systematic study of a Brazilian intermediate crude oil. Mar. Chem. 63, 93±104. éstgaard, K., Aaberg, A., Klungsùyr, J., Jensen, A., 1987. Comparative studies of phytotoxicity and chemical composition of aqueous oil solution aected by evaporation, illumination and extraction. Water Res. 21, 155±164.
1151
Payne, J.R., Phillips, C.R., 1985. Photo-chemistry of petroleum in water. Environ. Sci. Technol. 19, 569±579. Poston, T.M., Bean, R.M., Kalkwarf, D.R., Thomas, B.L., Clark, M.L., Killand, B.W., 1988. Photo-oxidation products of smoke generator fuel (SGF) no. 2 fog oil and toxicity to Hyallela azteca. Environ. Sci. Technol. 7, 753±762. Prince, R.C., Eimendorf, D.L., Lute, J.R., Hsu, C.S., Haith, C.E., Senius, J.D., Dechert, G.J., Douglas, G.S., Butler, E.L., 1994. 17a, 21b(H)-hopane as a conserved internal marker for estimating the biodegradation of crude oil. Environ. Sci. Technol. 28, 142±145. Pritchard, P.H., Costa, C.F., 1991. EPA Alaska oil spill bioremediation project. Environ. Sci. Technol. 25, 372± 379. Rontani, J.F., 1997. Mechanistic aspects of saturated hydrocarbon photo-oxidation induced by hydrogen atom abstraction. Trends Photochem. Photobiol. 4, 125±136. Schneider, A., Gominger, D., 1976. A preliminary study of the toxic eects of irradiated vs non-irradiated water soluble fractions of #2 fuel oil. Bull. Environ. Contam. Toxicol. 10, 595±603. Sugiura, K., Ishihara, M., Shimauchi, T., Harayama, S., 1997. Physicochemical properties and biodegradability of crude oil. Environ. Sci. Technol. 31, 45±51. Swannell, R.P.J., Lee, K., McDonagh, M., 1996. Field evaluations of marine oil spill bioremediation. Microbiol. Rev. 60, 342±365. Sydnes, L.K., Burkow, I.C., Stene, A., Lùnnning, S., 1985. On the formation of short-lived, toxic, water-soluble compounds, during illumination of crude oil on seawater. Mar. Environ. Res. 16, 115±126. Tjessem, K., Aaberg, A., 1983. Photo-chemical transformation and degradation of petroleum residues in the marine environment. Chemosphere 12, 1373±1394. Wang, Z., Fingas, M., Ken, L., 1994. Fractionation of a light crude oil and identi®cation and quantitation of aliphatic, aromatic, and biomarker compounds by GC±FID and GC± MS, part II. J. Chromatogr. Sci. 32, 367±382. Wolfe, D.A., Hameedi, M.J., Galt, J.A., Watabayashi, G., Short, J., O'Claire, C., Rice, S., Michel, J., Payne, J.R., Braddock, J., Hanna, S., Sale, D., 1994. The fate of oil spilled from the Exxon Valdez. Environ. Sci. Technol. 28, 560A±568A.
www.elsevier.com/locate/chemosphere
Photo-oxidation of biodegraded crude oil and toxicity of the photo-oxidized products H. Maki, T. Sasaki, S. Harayama
*
Marine Biotechnology Institute, Kamaishi Laboratories, 3-75-1 Heita, Kamaishi, Iwate 026-0001, Japan Received 15 March 2000; accepted 27 June 2000
Abstract We investigated the physicochemical changes resulting from irradiation by sunlight of biodegraded crude oil. An Arabian light crude oil sample was ®rst subjected to microbial degradation. n-Alkanes and aromatic compounds such as naphthalenes, ¯uorenes, dibenzothiophenes and phenanthrenes possessing short, alkyl side chain(s) were almost completely degraded, while the contents of the saturated and aromatic fractions were reduced by 70% and 40%, respectively. This biodegraded oil was then suspended in seawater and exposed to sunlight irradiation for several weeks. The most remarkable change caused by the irradiation was a substantial decline in the aromatic fraction with a concomitant increase in the resin and asphaltene fractions. A 13 C-nuclear magnetic resonance (NMR) spectroscopic analysis showed that the aromaticity of the biodegraded oil was signi®cantly lower in the irradiated sample. A ®eld desorption±mass spectrometric (FD±MS) analysis showed that sunlight irradiation reduced the average molecular weight of the oil components and formed oxygenated compounds. Consistent with this observation is that the oxygen content in the oil increased as the irradiation was prolonged. The bioavailability of the biodegraded oil was increased by the photo-oxidation: the growth of seawater microbes was minimal when the non-irradiated biodegraded oil was used as the source of carbon and energy; however, growth was signi®cant when irradiated biodegraded oil was used. The concentration of dissolved organic carbon (DOC) increased linearly during the sunlight irradiation of the biodegraded oil, and this increase was matched by an increase in ultraviolet-absorptive materials in the seawater. The photochemically formed, water-soluble fraction (WSF) showed acute toxicity against the halophilic crustacean, Artemia. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Crude oil; Biodegradation; Photo-oxidation; Bioavailability; Toxicity
1. Introduction A vast amount of mineral oil is being released into the ocean by both accidental oil spills and as a result of human activity on land, this chronic annual discharge of oil into the ocean being estimated as approximately 320
*
Corresponding author. Present address: National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki 305-0053, Japan. Tel.: +81-193-26-6544; fax: +81-193-26-6584. E-mail address: [email protected] (S. Harayama).
million tons (Etkin et al., 1998). The discharged oil is removed by various natural factors, i.e., volatilization, dispersion, precipitation, biodegradation and photooxidation (Wolfe et al., 1994). Microbial degradation has been applied to the removal of spilled oils in the marine environment ever since the serious tanker accident to the Exxon-Valdez occurred in Alaska in 1989 (Pritchard and Costa, 1991; Bragg et al., 1994; Swannell et al., 1996). However, the natural biodegradability of crude oil is limited. In laboratory experiments, the speed of the biodegradation of crude oil has been shown to diminish sharply after the petroleum components have been degraded by 30% (Maki et al., 1997). Besides the
0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 2 9 2 - 7
1146
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
biological action, sunlight irradiation is another eect which alters the physicochemical properties of crude oil in the natural environment. It is possible that crude oil components, which are recalcitrant to biodegradation are degraded by sunlight irradiation. Although many reports have so far been published characterizing the photo-chemistry of crude oil, they have focused on the physicochemical changes to intact crude oil, rather than to biodegraded crude oil (Larson et al., 1977, 1979; Tjessem and Aaberg, 1983; Payne and Phillips, 1985; éstgaard et al., 1987; Jacquot et al., 1996; Garrett et al., 1998; Nicodem et al., 1998). Moreover, these studies mainly employed gas chromatography (GC) and/or combined GC and mass spectrometry (GC±MS) for component analyses, despite the fact that GC methods can only detect the volatile components of crude oil (Poston et al., 1988; Jacquot et al., 1996; Nicodem et al., 1998). The GC-detectable compounds in crude oil are generally susceptible to biodegradation (Ishihara et al., 1995; Maki et al., 1997). In other words, the photochemical alteration of crude oil components that are resistant to biodegradation cannot be characterized by the GC techniques. Thin layer chromatography coupled with ¯ame ionization detector (TLC±FID) has been employed to evaluate the biodegradability of total petroleum including the non-volatile fraction in crude oil (Goto et al., 1994; Ishihara et al., 1995; Maki et al., 1997; Sugiura et al., 1997). In this present study, we applied TLC±FID, 13 C-nuclear magnetic resonance spectroscopy (NMR) and ®eld desorption±mass spectrometry (FD±MS) to survey the photo-chemical alteration of components in an already biodegraded crude oil sample. The water-soluble fraction (WSF) of crude oil is known to be signi®cantly increased by photo-oxidation (Payne and Phillips, 1985; Sydnes et al., 1985; éstgaard et al., 1987; Poston et al., 1988; Ali et al., 1995), and considerable attention has been paid to the ecological eect of photo-chemically generated WSF because it is known to exert signi®cant toxicity against aquatic organisms (Schneider and Gominger, 1976; Sydnes et al., 1985; éstgaard et al., 1987; Poston et al., 1988). However, there has been little work done on WSF derived from biodegraded crude oil. In this study, we characterize WSF, which has been formed by the sunlight irradiation of a biodegraded crude oil sample. 2. Materials and methods 2.1. Preparation of the biodegraded and photo-oxidized crude oil samples The oil used was Arabian light crude that had been heated at 230°C to evaporate the volatile fractions. The oil thus treated is hereafter called ``weathered crude oil''.
When analyzed by silica-gel column chromatography according to the method of the Japan Petroleum Institute (JPI, 1983), it was found to contain 61% of saturated hydrocarbons (the n-hexane eluate), 33% of aromatic hydrocarbons (the n-hexane:benzene 1:1 eluate), 4% of resins (the methanol eluate), and 2% of asphaltenes (the chloroform eluate). Chloroform (2 ml) containing 0.5 g ml 1 of the weathered crude oil was poured into each of several 2-l Sakaguchi ¯asks (a conical ¯ask used for microbial cultivation), and the chloroform was evaporated by heating the ¯asks overnight at 50°C. Fresh seawater (500 ml) was then added to each ¯ask (the ®nal oil concentration was 2 g l 1 ). 3 g of a slow-release solid, granular nitrogen fertilizer (Super IB; Mitsubishi Chemicals, Tokyo, Japan) and 0.5 g of a phosphorous fertilizer (Linstar 30, Mitsubishi chemicals) were supplemented to the seawater medium in each ¯ask. After adding these fertilizers, the ¯asks were agitated by a reciprocating shaker at 120 rpm and 25°C for 3 weeks to grow the indigenous microbes in seawater on the weathered crude oil. After the cultivation, the biodegraded oil in the medium was extracted with dichloromethane (DCM), the excess DCM being removed from the extract by using a rotary evaporator. The extracted oil was weighed and then dissolved in chloroform. Aliquots of the chloroform solution, each containing 30 mg of biodegraded oil, were poured into 100-ml Erlenmeyer ¯asks. After the chloroform had been evaporated, 30 ml of seawater was added to each ¯ask. The ®nal concentration of the biodegraded oil was 1 g l 1 . These ¯asks were capped with aluminum foil, moved to the roof of a building and agitated at 80 rpm by a rotary shaker. The ¯asks were exposed to sunlight for 3 weeks, the sunlight intensity in the daytime being in the range of 564±2 800 mol m 2 s 1 . No light with a wavelength shorter than 300 nm passed through the ¯ask glass. Non-irradiated controls were provided by the Erlenmeyer ¯asks containing seawater and the biodegraded oil, which were wrapped in aluminum foil and were incubated for 3 weeks. One ¯ask was used each week to extract the oil as already described. To sample the WSF, 18-ml aliquots of seawater were taken from each ¯ask, and the WSF samples were stored at 4°C until needed for the chemical analyses and the Artemia bioassay. 2.2. Analyses The extracted oil samples were dried, dissolved in a determined volume of chloroform, and then subjected to TLC±FID (Iatroscan MK-5; Iatron, Tokyo, Japan) (Goto et al., 1994). The chloroform solution (1 ll) was applied to one end of a silica-gel rod (Chromarods SIII, Iatron) and developed in three steps, ®rst with n-hexane over 10 cm, second with n-hexane:toluene (20:80 in volume) over 5 cm, and ®nally with methanol:DCM (5:95 in volume) over 2 cm. The amounts of hydrocar-
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
bons thus separated were determined by FID. This method separated the weathered crude oil into four peaks. We have called these four peaks as the saturated, aromatic, resin and asphaltene fractions in our previous works (Goto et al., 1994; Ishihara et al., 1995; Maki et al., 1997; Sugiura et al., 1997). Although we use the same nomenclature in this report, we have recently found that the saturated fraction contains not only saturates, but also alkylaromatics with long alkyl side chains (Dutta et al., manuscript in preparation). The oil samples were also subjected to an analysis by combined GC±MS (GC/MS-QP5000, Shimadzu, Kyoto, Japan) to quantify the series of n-alkanes (C13±33 ), C1±4 alkylnaphthalenes, C0±4 -alkyl¯uorenes, C0±4 -alkyldibenzothiophenes, C0±7 -alkylphenanthrenes (C0 indicates a non-substituted compound), and the biomarker, 17a(H), 21b(H)-hopane. The GC±MS analysis was conducted by the method of Wang et al. (1994). All values obtained by the instrumental analyses were normalized to that of 17a(H), 21b(H)-hopane (Prince et al., 1994). FD±MS was performed with a JMS-700 MStation equipped with a silicon emitter (JEOL, Tokyo, Japan) under the following conditions: emitter current, 0±40 mA; rate, 4 mA min 1 ; cathode voltage, 8.0 kV; scanning range, 50±3 000 m/z. 13 C NMR spectra were recorded by a Unity 300 spectrometer (Varian, Palo Alto, CA, USA) at 300 MHz, the intensity of each chemical shift being determined relative to internal tetramethylsilane in CDCl3 (Wako Pure Chemicals, Tokyo, Japan). The aromaticity is de®ned as the ratio of aromatic carbon to total carbon. The amount of aromatic carbon was obtained from the integrated intensity of these peaks between 100 and 160 rpm in the 13 C-NMR spectrum, while that of aliphatic carbon was found from the integrated intensity of the peaks between 4 and 70 ppm (Sugiura et al., 1997). The aromaticity index (fa) was calculated according to the equation fa (amount of aromatic carbons)/(amount of aromatic and aliphatic carbons). An elemental analysis to determine the oxygen content was conducted with a CHN instrument (MT-5; Yanako Analytical Instruments, Kyoto, Japan) equipped with a platinum column. Combustion was for 5 min at 750°C, and the bridge current for the detector was set at 100 mA. D L -alanine was used as the standard. The WSF adsorption spectrum was obtained by a Shimadzu UV2200 Ultraviolet-visible spectrophotometer. Dissolved organic carbon (DOC) was determined by a total organic carbon meter (TOC-5000, Shimadzu) equipped with an auto-sampler. 2.3. Artemia assay Artemia larvae were obtained from the eggs that are commercially available as a feed for tropical ®sh in a domestic aquarium. The Artemia eggs were incubated overnight in fresh seawater at 25°C. The hatched Art-
1147
emia larvae were recovered by a glass pipette and transferred into another vessel ®lled with fresh seawater. The recovered Artemia larvae were incubated overnight at 25°C and then used for an acute toxicity assay of WSF that had been formed by the photo-oxidation of the biodegraded oil. The Artemia assay was performed on 6-well plastic plates. A 5-ml aliquot of the WSF sample was poured into each well, and then 5 or 6 Artemia larvae were placed in each well. The number of dead Artemia larvae in each well was counted after incubating at 25°C for 24 h. 3. Results and discussion 3.1. Preparation of the biodegraded crude oil The weathered Arabian light crude oil was biodegraded by indigenous marine microbes at 25°C for 3 weeks. A 90±100% loss of n-alkanes (C13±33 ), C1±4 -alkylnaphthalenes, C0±4 -alkyl¯uorenes, C0±4 -alkyldibenzothiophenes and C0±7 -alkylphenanthrenes was identi®ed by the GC±MS analysis, while 40% and 70% reductions of the aromatic and saturated fractions, respectively, were detected by TLC±FID. The extent of biodegradation was very similar to that observed previously (Maki et al., 1997; Sugiura et al., 1997). The results of the 13 C-NMR analysis show that the total aromaticity of the biodegraded crude oil was 0.193, while that of the weathered (non-biodegraded) crude oil was 0.08. Thus, the aromatic components were enriched in the biodegraded oil as a result of the preferential biodegradation of the aliphatic components in the weathered crude oil. 3.2. Photo-oxidation of the biodegraded oil The biodegraded oil was suspended in seawater and exposed to sunlight for 3 weeks. The oil recovery was around 100% from the samples after 0 and 7 days, while it was 61% and 68% from the samples after 14 and 28 days, respectively. The oil recovery from the dark controls, in contrast, was always around 100% even after incubating for 28 days. During this period, no signi®cant biodegradation of the oil was expected to occur for two reasons. First, no nutrient was supplemented to the seawater during the period of sunlight irradiation. Our previous experiments showed no signi®cant growth of microorganisms and hence, no signi®cant biodegradation of weathered crude oil in natural seawater (Ishihara et al., 1995; Maki et al., 1997). Second, most of the biodegradable components in the weathered crude oil had been decomposed within 3 weeks, and further growth in a batch culture did not signi®cantly increase the degree of biodegradation (our unpublished observations). Thus, under prolonged sunlight irradiation,
1148
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
part of the oil seemed to have been transformed into a form which could not be extracted by DCM. 3.3. Characterization of the biodegraded oil after photooxidation When the DCM-extracted oil was analyzed by TLC± FID, the most remarkable change observed after irradiation by sunlight was a substantial decline in the aromatic fraction, and a signi®cant increase in the resin and asphaltene fractions (Fig. 1). The results of the 13 CNMR analysis also demonstrate the reduction in aromatic carbons by the sunlight irradiation: the chemical shift between 100 and 170 ppm in the 13 C-NMR spectrum detected the aromatic carbons, the intensity of such signals getting smaller as the photo-oxidation proceeded (Figs. 2(a) and (b)). The calculated aromaticity on the basis of the 13 C-NMR spectrum was plotted against the irradiation time (Fig. 2(c)). The decline in aromaticity determined by 13 C-NMR, however, was not matched by the abundance of the aromatic fraction determined by TLC±FID. The decrease in the aromatic fraction measured by TLC±FID exceeded 90% (Fig. 1(b)) in the ®rst week, in contrast to only a 37% decrease in aromaticity in the ®rst week as determined by 13 CNMR. This dierence could be explained by assuming that most of the aromatic compounds were converted to a polar form (i.e., moved to the resin or asphaltene fraction) either by cleavage of the aromatic rings, or by the introduction of polar groups such as hydroxyl and carbonyl, but that more than half of the aromatic rings survived after sunlight irradiation for 3 weeks.
A smaller but signi®cant decrease in the amount of the saturated fraction was observed. Saturates do not absorb light above 220 nm, and they are therefore relatively inert to photo-oxidation, although they could suer sensitized photo-oxidation (Rontani, 1997). The observed decrease in the saturated fraction, however, was not due to sensitized photo-oxidation of the saturated compounds, but instead due to the photo-oxidation of alkylaromatics with long alkyl side chains that were fractionated in the saturated fraction (Dutta et al., manuscript in preparation). Thus, the aromatic compounds present in the saturated and aromatic fractions were photo-chemically transformed into more polar compounds that migrated to the resin and/or asphaltene fractions. These aromatic compounds would be transformed into ones containing oxygen such as phenols, alcohols, ketones, aldehydes and carboxylic acids (Larson et al., 1977, 1979; Payne and Phillips, 1985; Poston et al., 1988). As shown in Fig. 3, the oxygen content was increased by up to 4.3-fold in the sample that had been photo-oxidized for 3 weeks. To evaluate the molecular weight distribution of the biodegraded oil and its photo-oxidized samples, we applied an FD±MS analysis. As shown in Fig. 4, the molecular weight distribution was shifted to a lower range after the photo-oxidation: the mean molecular weight (Mw ) of the biodegraded oil without sunlight irradiation was 943, whilst that of the sample sunlight-irradiated for 21 days was 575. In the spectrogram of the photo-oxidized sample, a distinct series of peaks became evident around m/z 300±700, in which the m/z dierence between the neighboring peaks was 16, corresponding to the
Fig. 1. Light-induced compositional change of the biodegraded oil determined by TLC±FID. The areas of the four fractions, saturated, aromatic, resin and asphaltene, on TLC±FID chromatograms (a) were determined, and the proportional area of each fraction is presented (b).
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
1149
Fig. 3. Increase in the oxygen content of oil exposed to sunlight irradiation.
Fig. 2. 13 C-NMR spectra of biodegraded oil without sunlight irradiation (a), and with sunlight irradiation for 21 days (b). The aromaticity of the samples before and 14 and 21 days after the sunlight irradiation is also shown (c).
mass of the oxygen atom. This series of peaks might represent multiplied oxygenated derivatives of the parent compounds. 3.4. Bioavailability of the oil subjected to both biodegradation and photo-oxidation The bioavailability of the photo-oxidized products generated from the biodegraded oil was investigated. The biodegraded oil samples with and without being
Fig. 4. Field desorption±mass spectra of biodegraded oil not exposed (a), and exposed to sunlight irradiation for 21 days (b).
exposed to sunlight irradiation were extracted by DCM, and each extract was dried in an Erlenmeyer ¯ask into which a seawater-based medium supplemented with the nitrogen and phosphorus fertilizers was then added to a concentration of 1 g of oil per litre. Each ¯ask was incubated at 25°C. The heterotrophic bacteria indigenous to seawater grew on the biodegraded oil to the concentration of 1:0 107 cfu ml 1 in 10 days, while they grew to the concentration of 6:6 107 cfu ml 1 on the
1150
H. Maki et al. / Chemosphere 44 (2001) 1145±1151
biodegraded and photo)oxidized oil sample. This observation implies that the photo-oxidation of the biodegraded oil could have produced some compounds, which were readily available to microbes in the seawater. 3.5. Formation of the WSF by photo-oxidation of the biodegraded crude oil and its toxicity The seawater was recovered from the sunlight-irradiated samples, and WSF in the seawater was characterized. Both the irradiated and dark-control ¯asks were prepared in triplicate. As shown in Fig. 5, ultravioletabsorbing material was formed in the irradiated-oil seawater as photo-oxidation proceeded, whereas almost no absorption was apparent in the non-irradiated dark controls. The concentration of DOC in the irradiated-oil seawater linearly increased with irradiation time, in contrast to no signi®cant increase in DOC in the dark controls (Fig. 5). The ®nal DOC concentration of WSF was 134 mg l 1 . Since the carbon content in the original crude oil sample was ca. 85%, approximately 16% of the biodegraded crude oil had been converted to WSF by the photo-oxidation process. WSF that had been formed by photo-oxidation of the biodegraded oil was also evaluated for its acute toxicity against the halophilic crustacean, Artemia. As shown in Fig. 6, the toxicity of WSF against Artemia increased as the photo-oxidation proceeded, all the Artemia larvae dying when seawater from the samples that had been irradiated for 25 days was used. The dark control showed hardly any signi®cant increase in toxicity against Artemia. These observations indicate that crude oil remains capable of releasing some toxic materials into seawater by photo-oxidation, even after it
Fig. 6. Acute toxicity against Artemia of WSF formed by the photo-oxidation of biodegraded oil. Irradiated (s), dark control (e).
has been extensively biodegraded. Although the biodegradability of WSF formed by the photo-oxidation of crude oil remains to be elucidated, it is expected to be resistant to microbial degradation because of its toxicity not only against crustaceans but also against microbes (Larson et al., 1977, 1979). There is an increasing interest in the natural attenuation of spilled oil as the cost associated with active cleanup eorts are rather large. The selection between cost-eective natural attenuation and man-made remedial engineering may be determined by various factors, but a full characterization of the site contaminated by oil, including an evaluation of the speed of natural attenuation, is required before the implementation of any remedial techniques. The results obtained in this study will provide some insights into evaluating the physicochemical changes and ecotoxicological eects of spilled oil after being subjected to intrinsic and/or man-stimulated bioremediation. Acknowledgements The authors are indebted to Mr. Etsuro Sasaki and Ms. Kaori Sasaki for their technical assistance, and to Mr. Kazutetsu Nojima of JEOL in Tokyo for the FD± MS analysis. This work was supported by the New Energy and Industrial Technology Development Organization (NEDO).
Fig. 5. Time-course plots for the increase in absorbance at 218 nm (A218 ) and the DOC concentration of WSF in seawater after exposing suspended biodegraded oil to sunlight irradiation. A218 in irradiated seawater (d), DOC in irradiated seawater (h), A218 in dark control seawater (´), DOC in dark control seawater (s).
References Ali, L.N., Fauzi, R.F., Mantoura, C., Rowland, S.J., 1995. The dissolution and photo-degradation of Kuwaiti crude oil in seawater, part 1: Quantitative dissolution and analysis of the seawater-soluble fraction. Mar. Environ. Res. 40, 1±17.
H. Maki et al. / Chemosphere 44 (2001) 1145±1151 Bragg, J.R., Prince, R.C., Harner, E.J., Atlas, R.M., 1994. Eectiveness of bioremediation for the Exxon Valdez oil spill. Nature 368, 413±418. Etkin, D.S., Wells, P., Nauke, M., Campbell, J., Grey, C., Koefoed, J., Meyer, T., Reddy, S., 1998. In: Proceedings of the 21st Arctic and Marine Oil Spill Program Technical Seminar. Environment Canada, Edmonton, Alberta, pp. 903±910. Garrett, R.M., Pickering, I.J., Haith, C.E, Prince, R.C, 1998. Photo-oxidation of crude oils. Environ. Sci. Technol. 32, 3719±3723. Goto, M., Kato, M., Asaumi, M., Shirai, K., Venkateswaran, K., 1994. TLC±FID method for evaluation of the crude-oildegrading capability of marine microorganisms. J. Mar. Biotechnol. 2, 45±50. Ishihara, M., Sugiura, K., Asaumi, M., Goto, M., Sasaki, E., Harayama, S., 1995. Oil degradation in microcosms and mesocosms. In: Microbial Process Bioremediation, vol. 3. Battelle Press, pp. 101±116. Jacquot, F., Guiliano, M., Doumenq, P., Munoz, D., Mille, G., 1996. In vitro photo-oxidation of crude oil maltenic fractions: Evolution of fossil biomarker and polycyclic aromatic hydrocarbons. Chemosphere 33, 671±681. Japan Petroleum Institute (JPI) standard, 1983. Analytical method for asphalt by column chromatography (JPU-5S22-83). JPI, Tokyo, Japan (in Japanese). Larson, R.A., Bott, T.L., Hunt, L.L., Rogenmuser, K., 1979. Photo-oxidation products of fuel oil and their antimicrobial activity. Environ. Sci. Technol. 13, 965±969. Larson, R.A., Hunt, L.L., Blankenship, D.W., 1977. Formation of toxic products from a #2 fuel oil by photo-oxidation. Environ. Sci. Technol. 11, 492±496. Maki, H., Ishihara, M., Harayama, S., 1997. Application of wastewater sludge to microbial degradation of crude oil. In: In Situ and On-site Bioremediation, vol. 4. Battelle Press, pp. 371±375. Nicodem, D.E., Guedes, C.L.B., Correa, R.J., 1998. Photochemisry of petroleum I. Systematic study of a Brazilian intermediate crude oil. Mar. Chem. 63, 93±104. éstgaard, K., Aaberg, A., Klungsùyr, J., Jensen, A., 1987. Comparative studies of phytotoxicity and chemical composition of aqueous oil solution aected by evaporation, illumination and extraction. Water Res. 21, 155±164.
1151
Payne, J.R., Phillips, C.R., 1985. Photo-chemistry of petroleum in water. Environ. Sci. Technol. 19, 569±579. Poston, T.M., Bean, R.M., Kalkwarf, D.R., Thomas, B.L., Clark, M.L., Killand, B.W., 1988. Photo-oxidation products of smoke generator fuel (SGF) no. 2 fog oil and toxicity to Hyallela azteca. Environ. Sci. Technol. 7, 753±762. Prince, R.C., Eimendorf, D.L., Lute, J.R., Hsu, C.S., Haith, C.E., Senius, J.D., Dechert, G.J., Douglas, G.S., Butler, E.L., 1994. 17a, 21b(H)-hopane as a conserved internal marker for estimating the biodegradation of crude oil. Environ. Sci. Technol. 28, 142±145. Pritchard, P.H., Costa, C.F., 1991. EPA Alaska oil spill bioremediation project. Environ. Sci. Technol. 25, 372± 379. Rontani, J.F., 1997. Mechanistic aspects of saturated hydrocarbon photo-oxidation induced by hydrogen atom abstraction. Trends Photochem. Photobiol. 4, 125±136. Schneider, A., Gominger, D., 1976. A preliminary study of the toxic eects of irradiated vs non-irradiated water soluble fractions of #2 fuel oil. Bull. Environ. Contam. Toxicol. 10, 595±603. Sugiura, K., Ishihara, M., Shimauchi, T., Harayama, S., 1997. Physicochemical properties and biodegradability of crude oil. Environ. Sci. Technol. 31, 45±51. Swannell, R.P.J., Lee, K., McDonagh, M., 1996. Field evaluations of marine oil spill bioremediation. Microbiol. Rev. 60, 342±365. Sydnes, L.K., Burkow, I.C., Stene, A., Lùnnning, S., 1985. On the formation of short-lived, toxic, water-soluble compounds, during illumination of crude oil on seawater. Mar. Environ. Res. 16, 115±126. Tjessem, K., Aaberg, A., 1983. Photo-chemical transformation and degradation of petroleum residues in the marine environment. Chemosphere 12, 1373±1394. Wang, Z., Fingas, M., Ken, L., 1994. Fractionation of a light crude oil and identi®cation and quantitation of aliphatic, aromatic, and biomarker compounds by GC±FID and GC± MS, part II. J. Chromatogr. Sci. 32, 367±382. Wolfe, D.A., Hameedi, M.J., Galt, J.A., Watabayashi, G., Short, J., O'Claire, C., Rice, S., Michel, J., Payne, J.R., Braddock, J., Hanna, S., Sale, D., 1994. The fate of oil spilled from the Exxon Valdez. Environ. Sci. Technol. 28, 560A±568A.