photooxidation of cude oil

8/9/2019 Photooxidation of Cude Oil

1/12

The Roles of Photooxidationand Biodegradation in Long-term

Weathering of Crude andHeavy Fuel OilsROGER C. PRINCE*, ROBERT M. GARRETT, RICHARD E. BARE, MATTHEW J. GROSSMAN,

TODD TOWNSEND, JOSEPH M. SUFLITA, KENNETH LEE§, EDWARD H. OWENS,

GARY A. SERGY, JOAN F. BRADDOCK§§, JON E. LINDSTROM & RICHARD R. LESSARD

ExxonMobil Research and Engineering Co., Annandale, NJ 08801, USA

Institute for Energy and the Environment and the Department of Botany and Microbiology, University of Oklahoma,

Norman, OK 73019, USA

§Department of Fisheries and Oceans, Dartmouth, Nova Scotia, Canada B2Y 4T3

Polaris Applied Sciences, Inc., Bainbridge Island, WA 98110, USA

Environment Canada, #200, 4999––98th Ave. Edmonton, AB, Canada T6B 2X3§§Institute of Arctic Biology, University of Alaska Fairbanks, Fairbanks, AK 99775, USA

Shannon & Wilson, Inc., Fairbanks, AK 99709, USA

ExxonMobil Research and Engineering Co., Fairfax, VA 22037, USA

Although spilled oil is subject to a range of natural processes, only combustion, photooxidation and bio-degradation destroy hydrocarbons and remove them from the biosphere. We present laboratory data thatdemonstrate the molecular preferences of these processes, and then examine some oil residues collectedfrom previously documented releases to confirm the important roles that these processes play in removingspilled oil from both marine and terrestrial environments. 2003 Elsevier Science Ltd. All rights reserved.

Introduction

Crude and heavy fuel oils that escape into the en-

vironment, either from natural seeps or from acci-

dental spills, become subject to a variety of physical,

chemical and biological phenomena. Small molecules

evaporate (Fingas, 1999; Sharma et al., 2002), and are

either degraded photochemically (Poisson et al., 2000;

Hurley et al., 2001), or are washed from the atmo-

sphere in rain and then biodegraded (Arzayus et al.,

2001). Under particularly aggressive aeration in water,

as happened in the spill from the OSSA II pipeline into

the flood-stage Rııo Desaguadero on the Bolivian Al-

tiplano in January 2000, this evaporation can extend

into molecules with >30 carbon atoms (Douglas et al.,2002; Prince et al., 2002), but evaporation is more

usually limited to molecules with less than about 15

carbon atoms (Payne et al., 1991; Fingas, 1999). Ter-

restrial spills may soak into the ground, as happened

in the Nipisi, Rainbow and Old Peace River pipeline

spills in the Lesser Slave Lake area of Northern Al-

berta spill (Blenkinsopp et al., 1996). Some molecules,

particularly aromatic hydrocarbons and small polar

molecules such as naphthenic acids, dissolve if suffi-

cient water is present (Lafargue & Le Thiez, 1996;

Burns et al., 2000), and again these are eventually

biodegraded (Herman et al., 1994). Spills at sea or on

lakes and rivers often disperse into the water column,

Spill Science & Technology Bulletin, Vol. 8, No. 2, pp. 145–156, 2003

2003 Elsevier Science Ltd. All rights reserved

Printed in Great Britain

1353-2561/03 $ - see front matter

doi:10.1016/S1353-2561(03)00017-3

145

*Corresponding author.

E-mail address: [email protected] (R.C. Prince).

http://mail%20to:%[email protected]/http://mail%20to:%[email protected]/


2/12

dramatically increasing the surface area available for

microbial colonization and biodegradation. This oc-

curred during the Braer spill (Thomas, 1993; Thomas& Lunel, 1993), and accelerating the process by the

application of chemical dispersants is an important

option for minimizing the environmental impact of

spills that do not naturally rapidly disperse (Fiocco &

Lewis, 1999; Lessard & DeMarco, 2000; Page et al.,

2000). Dispersants were successfully applied on a large

scale on the spill from the Sea Empress (Lunel et al.,

1997). Other physical processes that occur in water

include the formation of water in oil emulsions,

known as mousses (Fingas et al., 2001), and the in-

teraction of the oil with fine particles of sediment

(Owens, 1999). The oil–fine particle aggregates dis-perse in the water column, often as neutrally buoyant

particles, and the oil is more available for biodegra-

dation (Weise et al., 1999). Concentrations of oil in

local sediments where oil–fine particle interactions

have occurred are typically very low (Boehm et al.,

1987; Sergy et al., 1999). Some spills spontaneously

ignite, as happened during the Haven spill (Martinelli

et al., 1995), and deliberate ignition of spills is an ac-

cepted response option in some situations, such as that

of the New Carissa (Gallagher et al., 2001). The ulti-

mate fate of spilled hydrocarbons that are not col-

lected, burnt or photooxidized is biodegradation, and

stimulating this biodegradation by adding fertilizers

was successful on shorelines oiled following the spill

from the Exxon Valdez (Prince & Bragg, 1997).

Untreated terrestrial spills are not usually subjected

to any dilution, and while biodegradation eventually

removes the majority of the hydrocarbons, it appar-

ently leaves the majority of the resins and polar frac-

tions of the oil. Bioremediation to stimulate the

removal of the hydrocarbons can be an effective

treatment (McMillen et al., 1995; Prince et al., 1997;

Braddock et al., 1997; Radwan et al., 2000).

In contrast, when considering the long-term

weathering of oil spills in the marine environment it isimportant to bear in mind that the vast majority of

most spilled oil is physically dispersed so that it is

impossible to find, and hence to study. As examples,

most of the 2.5 million gallons of Bunker C spilled in

Chadabucto Bay, Nova Scotia, Canada, in February

1970 from the Arrow has ‘‘disappeared’’ (Owens et al.,

1993). So has that from the Baffin Island Oil Spill

experiment conducted on the northern tip of Baffin

Island, Nunavut, Canada in August 1981 (Owens

et al., 1994). In both cases, only remnants are left on

the shorelines. Nevertheless, analysis of these rem-

nants allows us to ascertain how the oils have altered

since the spill, and thus gain some insights into the

likely fate of the oil that has left the beaches.

In this paper we will review what is known about

the photochemistry and biodegradation of crude oils,

principally from work in the laboratory, and then use

this information to implicate these processes in the

transformation of oil spilled in marine and terrestrialenvironments.

Methods

The analyses of this paper rely principally on the

results of gas chromatography coupled with mass

spectrometry. This is a powerful technique that, in

selected ion monitoring mode allows the analysis of a

range of individual hydrocarbons, and in total ion

mode allows an estimation of the total hydrocarbon.

We focus here on normal and iso-alkanes, polycyclicaromatic hydrocarbons and their alkylated forms, and

hopanes and sterane biomarkers (see Douglas et al.,

1992). GC/MS is unable to analyze the majority of the

heteroatom-containing molecules in crude oils and

refined products, often called resins, asphaltenes and/

or polar compounds. These can be analyzed by thin

layer chromatography (Barman, 1996), and we report

some data from laboratory experiments using this

technique. This technique is not able, however, to

distinguish between polar compounds present in crude

oils and non-petrogenic polar compounds in envi-

ronmental samples, so we do not include any data

using this technique on samples collected from spill

sites.

The foundation of our approach is to follow

changes in the chemical composition of the oil, de-

termined with gas chromatography and mass spect-

rometry, using a conserved internal marker in the oil

as a reference compound. Providing we have a sample

of the initial oil, whether in a laboratory experiment or

in examining samples from a historical spill, we can

then determine how much of an individual analyte has

been lost from the experimental or field sample.

Although most hydrocarbons in crude oil are bio-

degradable, some, such as the biomarkers (Peters &Moldowan, 1993) that are molecular fossils of that

biomass that gave rise to the crude oil, are remarkably

resistant to biodegradation (Prince et al., 1994). This is

true for both aerobic (Prince et al., 1994) and anaer-

obic (Caldwell et al., 1998) biodegradation, and they

are also resistant to photooxidation (Garrett et al.,

1998). They can thus serve as conserved internal

markers within the oil, and the loss of other com-

pounds can be assessed with reference to them by

simple proportion.

17a(H)21b(H)hopane is abundant enough in most

crude and heavy fuel oils to be a particularly useful

conserved internal marker. In laboratory experiments

we have shown that it is not generated during oil

biodegradation (Prince et al., 1994), and we have not

seen evidence for its significant biodegradation in

R.C. PRINCE et al.

146 Spill Science & Technology Bulletin 8(2)


3/12

laboratory or field studies (although we have seen

some evidence for its eventual biodegradation in soil

(Prince et al., 1997)). We can calculate the percentdepletion of other analytes within the oil using the

equation:

%Loss ¼ ½ðð A0= H 0Þ ð As= H sÞÞ=ð A0= H 0Þ 100

Where: As and H s are the concentrations of the target

analyte and hopane in the oil sample, respectively, and

A0 and H 0 are the concentrations in the initially spilled

oil.

Recently Wang et al. (2001) have reported the ap-

parent biodegradation of hopanes in oil remaining

from the 1974 Metula spill, and Bost et al. (2001) have

reported degradation of hopanes and norhopanes bya microbial consortium enriched from a creosote-

contaminated site. Although we have seen no evidence

that biodegradation of 17a(H)21b(H)hopane has oc-

curred in any of the samples discussed here, if it had

occurred our estimates of the extent of biodegradation

of other compounds would be underestimates.

Photooxidation of crude oil

Figure 1 presents total ion gas chromatograms of

an artificially weathered Alaskan North Slope crude

oil (treated to have lost 30% of its initial weight byevaporation) exposed to the atmosphere in a shallow

dish in the dark or exposed to a laboratory UV lamp

(Garrett et al., 1998). There is clearly almost no effect

of the illumination on the total ion GC/MS chroma-

tograms (Fig. 1, left), but the illumination has a sig-

nificant effect on the composition of the oil determined

with thin layer chromatography. The saturates are

unaffected, but the majority of the aromatic hydro-

carbons have been converted to resins or polar mole-

cules (Fig. 1, right). When the aromatic hydrocarbons

are measured with selected ion monitoring GC/MS

(Douglas et al., 1992), a clear pattern emerges; as

shown in Fig. 2, the four-ring chrysene is substantially

more affected than the three-ring phenanthrene and

dibenzothiophene, and in each family the extent of

loss increases with increasing alkylation. Although

smaller aromatics such as naphthalene and benzene

derivatives were not present in the oil used in Figs. 1 &

2, we may surmise that these compounds would be lesssusceptible to photooxidation than phenanthrene and

dibenzothiophene, and recent work bears this out,

even with substantial alkylation (Dutta & Harayama,

2001). Fortunately, although resistant to photooxida-

tion, such molecules are readily biodegraded (Prince,

2002). These patterns of photooxidative loss, with

larger polycyclic aromatic hydrocarbons lost before

smaller ones, and more alkylated compounds lost be-

fore their less alkylated congeners, is quite different

from that seen in biodegradation (Elmendorf et al.,

1994), as we shall see below.

Fig. 1 Photooxidation of an artificially weathered crude oil. On the left are total ion mass chromatograms of an artificially weathered AlaskanNorth Slope crude oil, before and after exposure to a laboratory UV source. On the right is the composition of the oil determined by thin layerchromatography.

Fig. 2 Photooxidation of an artificially weathered crude oil. Thefigure shows the relative losses of alkylated polycyclic aromatic

hydrocarbons caused by exposure to a laboratory UV source. TheC0, C1, C2, C3 nomenclature indicates the number of alkyl carbonson the parent molecule, regardless of position. For example, C2includes dimethyl and ethyl forms.

PHOTOOXIDATION AND BIODEGRADATION OF CRUDE AND HEAVY FUEL OILS

Spill Science & Technology Bulletin 8(2) 147


4/12

Combustion of crude oil

As noted above, burning is sometimes considered asan option for dealing with spilled oil (Buist et al.,

1999; Yoshioka et al., 1999), including in marshes (Lin

et al., 2002). Burning can be an effective way of re-

moving large amounts of oil, but there is always some

residue left when the oil is unable to maintain a high

enough temperature for complete combustion, usually

because the slick becomes so thin that heat loss to the

substrate becomes a dominant process. The residual

oil is typically slightly enriched in pyrogenic hydro-

carbons such as fluoranthene and pyrene, although the

total amount of these compounds in the environment

is reduced by a successful burn (Garrett et al., 2000).None of the samples discussed here has any history,

nor shows any evidence, that combustion was involved

in the weathering processes.

Aerobic biodegradation of crude oil

The aerobic biodegradation of hydrocarbons has

been intensively studied in the last century, and hun-

dreds of cultures of hydrocarbon-degrading aerobic

microorganisms have been studied (Prince, 2002).

These organisms rely upon oxygen as both the initial

oxidant of the hydrocarbon, and the terminal electron

acceptor of respiratory electron flow. Almost all hy-

drocarbons are known to be biodegraded, although

individual strains of organisms typically degrade only

a limited range of substrates. A typical aerobic bio-

degradation pattern for a hydrocarbon fuel is shown

in Fig. 3––on the left, total ion GC/MS chromato-

grams of an intermediate fuel oil that has undergone

aerobic biodegradation by an inoculum collected from

the New Jersey coast, on the right the loss of indi-

vidual compounds identified by selected ion moni-

toring GC/MS. Within one week the n-alkanes,

exemplified here by heptadecane, were essentiallycompletely removed, as were approximately 50% of the

phenanthrene and dibenzothiophene. In contrast, only

some 10% of the branched alkane pristane (2,6,10,14-

tetramethylpentadecane) had been degraded after one

week, but there had been substantial degradation by

three weeks, at which time chrysene biodegradation

had begun. More than 60% of the chrysene was lost at

12 weeks. Figure 4 shows the pattern of degradation

of the alkyl polycyclic aromatics in this experiment; it

is clear that within each family, the unsubstituted

parent compound is degraded most readily, and that

increasing alkylation slowed biodegradation (Elmen-dorf et al., 1994). In summary, under aerobic condi-

tions the n-alkanes are the most readily degraded

hydrocarbons, and the biodegradation of polycyclic

aromatic hydrocarbons decreases with increasing size

and alkylation. These patterns are essentially the op-

posite of those seen for photooxidation.

Perhaps surprisingly, biodegradation in the field

does not usually show very much isomer specificity; all

the isomers of, for example, the methyl dibenzothi-

ophenes or phenanthrenes are lost at more or less the

same rate, although this is not necessarily apparent at

first inspection. For example, the left panel of Fig. 5

shows the methyldibenzothiophenes and methylphe-

nanthrenes of the Arrow cargo oil, and a sample from

Black Duck Cove, Nova Scotia, Canada, a site still

contaminated with oil from the 1970 spill (see below).

While at first glance it seems that there has been

a dramatic preference for the loss of some isomers

(e.g. 4-methyldibenzothiophene) and not others (e.g.

1-methyldibenzothiophene), in fact all the isomers

have been substantially removed when compared to

the residual hopane in the oil (Fig. 5 right panel). We

Fig. 3 Aerobic biodegradation of a heavy fuel oil (IF30). On the leftare total ion mass chromatograms of the oil at various times into theexperiment, and on the right are the relative losses of representativesaturate and aromatic compounds in the oil.

Fig. 4 Aerobic biodegradation of a heavy fuel oil (IF30). The figureshows the relative losses of alkylated polycyclic aromatic hydro-carbons in the samples from Fig. 3, with similar nomenclature toFig. 2.

R.C. PRINCE et al.



5/12

have seen no consistent pattern of preferential degra-

dation of individual isomers in the samples we have

collected from the sites discussed here.

As we will discuss below, all the samples we have

collected from the field show changes that we can

readily interpret as the combination of evaporation,

aerobic biodegradation and photochemistry. Never-theless, Fayad and Overton (1995) have reported a

quite different pattern of biodegradation in a labora-

tory experiment with mousse collected during the Gulf

oil spill. In the absence of added nutrients (just indi-

genous nutrients from the Gulf seawater), and at 5

and 20 g oil per liter of seawater, they reported sub-

stantial degradation of aromatic but not aliphatic

hydrocarbons in 144 h. When nutrients were added,

this preference was reversed! There are some puzzling

aspects to this work, including the observation that

much less biodegradation was seen with 10 g oil per

liter of seawater. Unfortunately no data on potential

conserved biomarkers is reported, so it is possible that

the oils in the different tests, albeit from a single

sample of mousse, may have been from a hetero-

geneous mixture of oils. Otherwise it is very hard to

reconcile the data from the different experiments. And

in the absence of data on a conserved internal marker,

their observation of the apparent preferential biode-

gradation of 4-methyldibenzothiophene (their Fig. 3)

may well be akin to that seen in Fig. 5 here, and in fact

reflect very extensive biodegradation of all isomers.

Anaerobic biodegradation of crude oil

The anaerobic degradation of hydrocarbons has

only been clearly demonstrated in the last decade or so

(Heider et al., 1999; Spormann & Widdel, 2000), and

very few defined cultures are in laboratory captivity

(Prince, 2002). In the absence of oxygen, sulfate, which

is reduced to sulfide, and carbon dioxide, which is

reduced to methane, are the most likely oxidants in

most terrestrial and aquatic environments. Although

sulfate reduction and methanogenesis have been well

studied, their involvement in hydrocarbon biode-gradation has not been fully documented, and both

the substrate range and preference of anaerobic hy-

drocarbon-degrading communities are largely un-

known.

We have shown that linear alkanes in crude oil are

readily degraded under sulfate-reducing conditions

by microorganisms from marine sediments (Caldwell

et al., 1998). More recently, we have found that mi-

croorganisms from a terrestrial subsurface environ-

ment catalyze a similar range of crude oil n-alkane

biodegradation under both sulfate-reducing and

methanogenic conditions (Fig. 6). Compared to aero-

bic biodegradation, extensive degradation of branched

alkanes and aromatic hydrocarbons seems to lag far

behind that of the n-alkanes, and under optimal con-

ditions, the anaerobic process, in general, is likely

to be slower than the aerobic process. Nevertheless,

it is apparent that spilled crude oil is subject to bio-

degradation in both aerobic and anaerobic environ-

ments.

The Arrow spill

The wreck of the Arrow in February 1970 released

2.5 million gallons of Bunker C fuel oil into Chedab-

ucto Bay, Nova Scotia, Canada (45 N, 61 W). Only

48 km of an estimated 305 km of oiled shoreline were

cleaned after the spill, and there are still traces of

Fig. 5 Aerobic biodegradation of methyldibenzothiophene and methylphenanthrene isomers. On the left are selected ion chromatograms(m= z ¼ 198 and 192) of cargo from the Arrow and oil from the beach of Black Duck Cove, Nova Scotia, Canada. On the right is the percentageloss of the individual isomers in the field sample.




6/12

residual oil in parts of the affected area that can be

attributed to the spill (Wang et al., 1994; Prince et al.,

1998). Figure 7 shows chromatograms of the original

cargo and of two samples collected in October 1997

from Black Duck Cove, one of the areas where small

amounts of oil can still be found. We note that the

residual surface oil is not very noticeable to the un-informed eye, since the oil is associated with asphalt

pavements in a beautiful day-use park.

One sample, a subsurface sheen, was collected from

the surface of water filling a shallow pit dug in the

intertidal zone of the sheltered beach, while the other

is a sample of exposed asphalt from above the high

tide mark. Both have lost substantial amounts of

their initial hydrocarbon, some 40% and 60% respec-

tively. Note that the subsurface sheen sample still has

molecules with less than 20 carbons, although these

are not resolvable alkanes, while the surface sample

has lost most of these. We attribute this differenceto more extensive evaporation of the surface sam-

ple, coupled with extensive biodegradation in both

samples, and photooxidative loss of alkylated phe-

nanthrenes and chrysenes in the exposed sample

(Fig. 8).

Fig. 7 Samples from Black Duck Cove, Nova Scotia, Canada. On the left are total ion mass chromatograms of the cargo oil and two samplesfrom the cove. On the right are the relative losses of representative saturate and aromatic compounds in the oil.

Fig. 6 Anaerobic biodegradation of an artificially weathered crude oil. On the left are total ion mass chromatograms of an artificially weatheredAlaskan North Slope crude oil, before and after biodegradation under sulfate-reducing and methanogenic conditions. Inocula came from FortLupton, Colorado, and the cultures were incubated for approximately a year. On the right are the relative losses of representative saturated andaromatic compounds in the oil.

R.C. PRINCE et al.



7/12

The Baffin Island Oil Spill (BIOS) Project

The Baffin Island Oil Spill (BIOS) Project (Sergy &

Blackall, 1987) released approximately 15 m3 of lightly

weathered (8% weight loss) Lago Medio crude oil onto

the water adjacent to a shoreline on Cape Hatt,

northern Baffin Island, Nunavut, Canada (72310 N,

79500 W) in August 1981. About 45% of the oil

stranded on the previously pristine adjacent beach,

and this subsequently weathered naturally without any

cleanup efforts. By 1989 there had been an approxi-

mately 80% decrease in the total oiled area (Owens

et al., 1994), and in September 2001 we estimated that

coverage had decreased to less than 5% of the initial

area (Owens et al., 2002; Prince et al., 2002).

Wang et al. (1995) examined oil samples collected

in 1993 from this site, and Fig. 9 shows chromato-

grams of the spilled oil and of three samples collectedin September 2001 (Prince et al., 2002). The hydro-

carbon content of these three samples are very dif-

ferent, with the subsurface sample being almost

unchanged in the 20 years since it was spilled with the

exception of the loss of parent and methyl phenanth-

renes and dibenzothiophenes, which we tentatively

attribute to evaporation. In contrast, the oil on some

surface granules was extensively degraded, having lost

approximately 90% of its total hydrocarbons, but it

has still only lost about 20% of its methylchrysenes.

We attribute most of these losses to biodegradation. In

contrast, the oil extracted from a surface pavementhad lost only approximately 50% of its total hydro-

carbon and only 50% of its pristane, but 40% of its

methylchrysenes. We attribute this to rather less ex-

tensive biodegradation coupled with rather more ex-

tensive photooxidation, and the relative losses of the

alkylated forms of chrysene bear this out (Fig. 10).

The Poker-Caribou Creeks Research Watershed experi-

ment

The Poker-Caribou Creeks Research Watershed oil

spill experiment was conducted in 1976 during the

construction of the Trans Alaska Pipeline in order to

examine the potential effects of an oil leak from the

pipeline (Collins et al., 1994). Two 7570 l spills (one in

February and one in July) of hot (57 C) Prudhoe Bay

crude oil were conducted in an open black spruce

(Picea mariana) forest. Samples were collected from

the winter spill site and from an adjacent reference site

in June 2001. The most likely mechanisms for oil loss

at this site are evaporation, biodegradation and

Fig. 9 Samples from the BIOS site, Nunavut, Canada. On the left are total ion mass chromatograms of the initially spilled oil and three samplesfrom the beach. On the right are the relative losses of representative saturate and aromatic compounds in the oil.

Fig. 8 Samples from Black Duck Cove, Nova Scotia, Canada. Thefigure shows the relative losses of alkylated polycyclic aromatichydrocarbons in the samples from Fig. 6.




8/12

photooxidation. Loss by water-washing does not seem

an important fate, since there is no evidence for sig-

nificant water movements at the site and the spill area

did not expand significantly over the first 13 years

(Collins et al., 1994). Figure 11 (left panel) shows gas

chromatograms of the initially spilled oil and of three

samples collected from the site approximately 25 years

later. Perhaps surprisingly, the oil in the mineral soil

horizon (8–18 cm below the surface) is essentially

unchanged despite its 25 years in the environment; less

than 30% of the heptadecane and less than 5% of the

pristane has been lost, and the only significant loss

seems to be of the aromatics, including phenanthreneand dibenzothiophene and their alkyl derivatives (Fig.

11, right panel). Interestingly, this sample still contains

molecules such as dodecane that are thought to be

relatively readily lost by evaporation. In contrast, the

sample from the oiled organic soil horizon (0–8 cm

from the surface) is substantially weathered, having

lost 40% of its total hydrocarbon, and almost 90% of

its heptadecane and methylphenanthrene and methyl-

dibenzothiophene. We attribute most of this loss to

biodegradation. A sample from an oiled surface twig is

even more weathered. Figure 12 shows that neither

chrysene nor its alkyl substituted forms has been lostfrom the soil samples, consistent with quite limited

biodegradation and no significant photooxidation in

these heavily oiled samples (4.5% and 34% oil by

weight, respectively) that were overlain by 5 cm of

moss. Nevertheless, the oil from a surface twig has lost

substantial amounts of these compounds in a manner

consistent with photooxidation rather than biodegra-

dation. Rather similar results, with some samples re-

maining essentially unaltered after 25 years in the

environment, have been obtained by Wang et al.

(1998) in samples from the Nipisi spill near the Lesser

Slave Lakes in northern Alberta. Whether plants can

recolonize the Poker-Caribou site as natural weath-

ering proceeds remains to be seen. Already some

mosses and lichens are beginning to creep across the

surface from unoiled areas adjacent to the site, and it

is possible that the oiled layers may eventually be

Fig. 11 Samples from the Poker-Caribou Flats oil spill site. On the left are total ion mass chromatograms of the initially spilled oil and threesamples from the site. On the right are the relative losses of representative saturate and aromatic compounds in the oil.

Fig. 10 Samples from the BIOS site, Nunavut, Canada. The figureshows the relative losses of alkylated polycyclic aromatic hydro-carbons in the samples from Fig. 8.

R.C. PRINCE et al.



9/12

buried by these plants, and then provide a substratum

for succession of other species.

The Erika spill

The Erika, carrying about 30,000 tonnes of heavy

fuel oil as cargo, broke up in a severe storm off the

coast of Brittany on 11 December 1999 (Oudot, 2000).

The left hand panel of Fig. 13 presents a gas chro-

matogram of the cargo oil, and of a sample collected

on 29 March, 2000, at Le Croisic, less than four

(winter) months after the spill. The right hand panel

shows that almost 20% of the hydrocarbon had been

lost, even in this short time. The loss of naphthalenes

may well be due to evaporation, but the loss of hepta-

decane, the phenanthrenes and the dibenzothiophenes

is most likely due to biodegradation. As we saw in

Fig. 3, the fact that no chrysenes had been lost sug-

gests that biodegradation is just beginning, but also

suggests that biodegradation will eventually removemuch of the hydrocarbon content of the spilled oil.

The fate of the resins and asphaltenes, which make up

the majority of the spilled cargo, is less clear, as we will

discuss below.

Discussion

Crude oils and refined products are typically com-

posed of many thousands of individual compounds

(Tissot & Welte, 1984), and we have focussed on only

a few of them in this paper. Nevertheless the com-pounds we have discussed include the most abundant

and most biodegradable, the alkanes, and represen-

tative polycyclic hydrocarbons on the USEPA priority

pollutant list (Keith & Telliard, 1979), which are

usually of the most environmental concern. The ma-

jority of the other hydrocarbons are also biodegrad-

able (see McMillen et al., 1995; Prince, 2002), and it is

reasonable to expect that their biodegradation occurs

concomitantly with the alkanes and polycyclic aro-

matic hydrocarbons. Less is known about the bio-

degradation of many of the polar oil compounds,

including those commonly called resins and asphalt-

enes. Current knowledge suggests that these species

are not very biodegradable, nor are they subject to

significant photooxidative destruction, so they are

likely to remain in the environment for a long time.

Fortunately their inertness seems to be mirrored by

their environmental impact, and indeed in the absence

of hydrocarbons they are difficult to distinguish from

modern soil and sediment components such as humic

and fulvic acids (Burdon, 2001; Rice, 2001). They

completely lack the ‘‘oiliness’’ and ‘‘stickiness’’ asso-

ciated with crude and heavy fuel oils.

Fig. 13 A sample from the Erika spill. On the left are total ion mass chromatograms of the initially spilled oil and a sample from a beach. On theright are the relative losses of representative saturate and aromatic compounds in the oil.

Fig. 12 Samples from the Poker-Caribou Flats oil spill site. Thefigure shows the relative losses of alkylated polycyclic aromatichydrocarbons in the samples from Fig. 10.




10/12

The samples described here show that both biode-

gradation and photooxidation play important roles in

the long-term weathering of crude and heavy fuel oilsin the environment. The most degraded sample de-

scribed from the Arrow Bunker C spill (Fig. 7) has lost

60% of its initial hydrocarbon, essentially all of its

parent, C1, C2 and C3 phenanthrenes, and more than

half of its C1 and C2 chrysenes, the latter by photo-

oxidation. The most degraded sample from the BIOS

crude oil experiment (Fig. 9) had lost 85% of its total

hydrocarbon, and the most degraded sample from the

Poker-Caribou Flats crude oil experiment had lost

70% of its hydrocarbon (Fig. 11). Biodegradation can

be quite rapid, as seen in the sample from the Erika,

which has lost significant amounts of phenanthrene in just a few winter months.

Nevertheless, at both the BIOS and Poker-Caribou

Creeks Research Watershed sites there are also still

some samples that are essentially unchanged from the

date of the spill. What causes this heterogeneity? At

BIOS the least degraded samples have probably never

been inundated by the tide, and so their biodegrada-

tion may be limited by the availability of water. At

Poker-Caribou Creeks Research Watershed the least

degraded samples are from very heavily oiled soil that

has been undisturbed since the spill, and biodegrada-

tion may be limited by inhospitably oily conditions.

But this is not the only heterogeneity that exists at

these sites. At BIOS, for example, the shoreline site

was oiled in 1981. By 1989 only some 20% was still

oiled (Owens et al., 1994), and this had decreased to

only some 5% by 2001. The remaining oil was patchily

distributed, and it is not obvious why some areas re-

mained oiled while adjacent areas were apparently

completely clean. It seems likely that the major loss of

oil was caused by physical factors, but why is the effect

so heterogeneous? The entire beach likely freezes in

the winter, and the areas that have lost oil do not seem

more exposed, or to be more subject to run-off, than

the areas with residual oil. Understanding both thecauses of the physical weathering processes, including

oil–fines interactions (Owens, 1999), and their hetero-

geneity, and integrating this with our understanding

of photooxidation and biodegradation, may allow us

to construct useful models to predict the long-term

weathering of spilled oil, but this remains a challeng-

ing goal.

Acknowledgements— We are grateful to Dr. Jean Oudot for theErika samples.

References

Arzayus, K.M., Dickhut, R.M., Canuel, E.A., 2001. Fate of atmospherically deposited polycyclic aromatic hydrocarbons(PAHs) in Chesapeake Bay. Environmental Science & Techno-logy 35, 2178–2183.

Barman, B.N., 1996. Hydrocarbon-type analysis of base oils andother heavy distillates by thin-layer chromatography with flame-

ionization detection and by the clay–gel method. Journal of Chromatographic Science 34, 219–225.Blenkinsopp, S., Sergy, G., Lambert, P., Wang, Z., Zoltai, S.C.,

Sitanen, M., 1996. Long-term recovery of peat bogs oiled bypipeline spills in northern Alberta. In: Proceedings of theNineteenth Arctic and Marine Oil Spill Program TechnicalSeminar. Environment Canada, Calgary, Alberta, pp. 1335– 1354.

Boehm, P.D., Steinhauer, M.S., Green, D.R., Fowler, B., Hum-phrey, B., Fiest, D.L., Cretney, W.J., 1987. Comparative fate of chemically dispersed and beached crude oil in subtidal sedimentsof the Arctic nearshore. Arctic 40 (1), 133–148.

Bost, F.D., Frontera-Suau, R., McDonald, T.J., Peters, K.E.,Morris, P.J., 2001. Aerobic biodegradation of hopanes andnorhopanes in Venezuelan crude oils. Organic Geochemistry 32,105–114.

Braddock, J.F., Ruth, M.L., Catterall, P.H., Walworth, J.L.,

McCarthy, K.A., 1997. Enhancement and inhibition of micro-bial activity in hydrocarbon-contaminated arctic soils: implica-tions for nutrient-amended bioremediation. EnvironmentalScience & Technology 31, 2078–2084.

Buist, I., McCourt, J., Potter, S., Ross, S., Trudel, K., 1999. In situburning. Pure & Applied Chemistry 71, 43–66.

Burdon, J., 2001. Are the traditional concepts of the structures of humic substances realistic? Soil Science 166, 752–769.

Burns, K.A., Codi, S., Duke, N.C., 2000. Gladstone, Australia fieldstudies: weathering and degradation of hydrocarbons in oiledmangrove and salt marsh sediments with and without theapplication of an experimental bioremediation protocol. MarinePollution Bulletin 41, 392–402.

Caldwell, M.E., Garrett, R.M., Prince, R.C., Suflita, J.M., 1998.Anaerobic biodegradation of long-chain n-alkanes under sulfate-reducing conditions. Environmental Science & Technology 32,

2191–2195.Collins, C.M., Racine, C.H., Walsh, M.E., 1994. The physical,chemical and biological effects of crude oil spills after 15 years ona black spruce forest, interior Alaska. Arctic 47, 164–175.

Douglas, G.S., McCarthy, K.J., Dahlen, D.T., Seavey, J.A.,Steinhauer, W.G., Prince, R.C., Elmendorf, D.L., 1992. Theuse of hydrocarbon analyses for environmental assessment andremediation. Journal of Soil Contamination 1, 197–216.

Douglas, G.S., Owens, E.H., Hardenstine, J., Prince, R.C., 2002.The OSSAII pipeline spill: the character and weathering of the spilled oil. Spill Science and Technology Bulletin 7, 135– 148.

Dutta, T.K., Harayama, S., 2001. Analysis of long side chainalkylaromatics in crude oil for evaluation of their fate in theenvironment. Environmental Science & Technology 35, 102– 107.

Elmendorf, D.L., Haith, C.E., Douglas, G.S., Prince, R.C., 1994.

Relative rates of biodegradation of substituted polycyclic aro-matic hydrocarbons. In: Hinchee, R.E., Leeson, A., Semprini,L., Ong, S.K. (Eds.), Bioremediation of Chlorinated andPolycyclic Aromatic Hydrocarbon Compounds. Lewis Publish-ers, Boca Raton, FL, pp. 188–202.

Fayad, N.M., Overton, E., 1995. A unique biodegradation patternof the oil spilled during the 1991 Gulf War. Marine PollutionBulletin 30, 239–246.

Fingas, M.F., 1999. The evaporation of oil spills: development andimplementation of new prediction methodology. In: Proceedingsof the 1999 International Oil Spill Conference. AmericanPetroleum Institute, Washington, DC, pp. 185–194.

Fingas, M.F., Fieldhouse, B., Lane, J., Mullin, J.V., 2001. Whatcauses the formation of water-in-oil emulsions. In: Proceedingsof the 2001 International Oil Spill Conference. AmericanPetroleum Institute, Washington, DC, pp. 109–114.

Fiocco, R.J., Lewis, A., 1999. Oil spill dispersants. Pure and AppliedChemistry 71, 27–42.

Gallagher, J.J., Hile, H.B., Miller, J.A., 2001. The old New Carissa;a study in patience. In: Proceedings of the 2001 Oil SpillConference. American Petroleum Institute, Washington, DC,pp. 85–90.

R.C. PRINCE et al.



11/12

Garrett, R.M., Pickering, I.J., Haith, C.E., Prince, R.C., 1998.Photooxidation of crude oils. Environmental Science & Tech-

nology 32, 3719–3723.Garrett, R.M., Guenette, C.C., Haith, C.E., Prince, R.C., 2000.Pyrogenic polycyclic aromatic hydrocarbons in oil burn residues.Environmental Science & Technology 34, 1934–1937.

Heider, J., Spormann, A.M., Beller, H.R., Widdel, F., 1999.Anaerobic bacterial metabolism of hydrocarbons. FEMS Mi-crobiology Review 22, 459–473.

Herman, D.C., Fedorak, P.M., MacKinnon, M.D., Costerton, J.W.,1994. Biodegradation of naphthenic acids by microbial popula-tions indigenous to oil sands tailings. Canadian Journal of Microbiology 40, 467–477.

Hurley, M.D., Sokolov, O., Wallington, T.J., Takekawa, H.,Karasawa, M., Klotz, B., Barnes, I., Becker, K.H., 2001.Organic aerosol formation during the atmospheric degradationof toluene. Environmental Science & Technology 35, 1358– 1366.

Keith, L.H., Telliard, W.A., 1979. Priority Pollutants I.––a perspec-

tive view. Environmental Science & Technology 13, 416–423.Lafargue, E., Le Thiez, P., 1996. Effect of water washing on light

ends compositional heterogeneity. Organic Geochemistry 24,1141–1150.

Lessard, R.R., DeMarco, G., 2000. The significance of oil spilldispersants. Spill Science & Technology Bulletin 6, 59–68.

Lin, Q.X., Mendelssohn, I.A., Carney, K., Bryner, N.P., Walton,W.D., 2002. Salt marsh recovery and oil spill remediation afterin-situ burning: Effects of water depth and burn duration.Environmental Science & Technology 36, 576–581.

Lunel, T., Rusin, J., Halliwell, C., Davies, L., 1997. The netenvironmental benefit of a successful dispersant operation at theSea Empress incident. In: Proceedings of the 1997 InternationalOil Spill Conference. American Petroleum Institute, Washing-ton, DC, pp. 185–194.

Martinelli, M., Luise, A., Tromellini, E., Sauer, T.C., Neff, J.M.,

Douglas, G.S., 1995. The M/C Haven oil spill: Environmentalassessment of exposure pathways and resource injury. In:Proceedings of the 1995 Oil Spill Conference. American Petro-leum Institute, Washington, DC, pp. 679–685.

McMillen, S.J., Requejo, A.G., Young, G.N., Davis, P.S., Cook,P.D., Kerr, J.M., Gray, N.R., 1995. Bioremediation potential of crude oil spilled on soil. In: Hinchee, R.E., Vogel, C.M.,Brockman, F.J. (Eds.), Microbial Processes for Bioremediation.Battelle Press, Columbus, OH, pp. 91–99.

Oudot, J., 2000. Biodegradability of the Erika fuel oil. ComptesRendus de LAcademie Des Sciences Serie III––Sciences de LaVie––Life Sciences 323, 945–950.

Owens, E.H., 1999. The interaction of fine particles with strandedoil. Pure and Applied Chemistry 71, 83–93.

Owens, E.H., Sergy, G.A., McGuire, B.E., Humphrey, B., 1993. The1970 Arrow Spill: what remains on the shoreline 22 years later.In: Proceedings of the Sixteenth Arctic and Marine Oil Spill

Program Technical Seminar. Environment Canada, Calgary,Alberta, pp. 1149–1167.

Owens, E.H., Humphrey, B., Sergy, G.A., 1994. Natural cleaning of oiled coarse sediment shorelines in Arctic and Atlantic Canada.Spill Science & Technology Bulletin 1, 37–52.

Owens, E.H., Sergy, G.A., Prince, R.C., 2002. The fate of strandedoil at the BIOS site twenty years after the experiment. In:Proceedings of the Twentyfifth Arctic and Marine Oil SpillProgram Technical Seminar. Environment Canada, Calgary,Alberta, pp. 1–11.

Page, C.A., Bonner, J.S., Sumner, P.L., McDonald, T.J., Autenr-ieth, R.L., Fuller, C.B., 2000. Behavior of a chemically-dispersedoil and a whole oil on a near-shore environment. Water Research34, 2507–2516.

Payne, J.R., Clayton, J.R., McNabb, G.D., Kirstein, B.E., 1991.Exxon Valdez oil weathering and behavior, model predictionsand field observations. In: Proceedings of the 1991 InternationalOil Spill Conference. American Petroleum Institute, Washing-ton, DC, pp. 641–654.

Peters, K.E., Moldowan, J.M., 1993. The Biomarker Guide;Interpreting molecular fossils in petroleum and ancient sedi-ments. Prentice-Hall, Englewood Cliffs, NJ.

Poisson, N., Kanakidou, M., Crutzen, P.J., 2000. Impact of non-methane hydrocarbons on tropospheric chemistry and the

oxidizing power of the global troposphere: 3-dimensional mod-elling results. Journal of Atmospheric Chemistry 36, 157–230.Prince, R.C., 2002. Petroleum and other hydrocarbons, biodegra-

dation of. In: Bitton, G. (Ed.), Encyclopedia of EnvironmentalMicrobiology. John Wiley, New York, pp. 2402–2416.

Prince, R.C., Bragg, J.R., 1997. Shoreline bioremediation followingthe Exxon Valdez oil spill in Alaska. Bioremediation Journal 1,97–104.

Prince, R.C., Elmendorf, 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(H),21b(H)-hopane as a conserved internal marker forestimating the biodegradation of crude oil. EnvironmentalScience & Technology 28, 142–145.

Prince, R.C., Drake, E.N., Madden, P.C., Douglas, G.S., 1997.Biodegradation of polycyclic aromatic hydrocarbons in ahistorically contaminated site. In: Alleman, B.C., Leeson, A.(Eds.), In situ and on-site bioremediation 2. Battelle Press,

Columbus, OH, pp. 205–210.Prince, R.C., Garrett, R.M., Haith, C.E., Vandermeulen, J.H.,

Cobanli, S., Mossman, D.C., Lee, K., 1998. The role of biodegradation in the weathering of oil from the 1970 Arrowspill. In: Proceedings of the Twentyfirst Arctic and Marine OilSpill Program Technical Seminar. Environment Canada, Cal-gary, Alberta, pp. 717–727.

Prince, R.C., Owens, E.H., Sergy, G.A., 2002. Weathering of anArctic oil spill over 20 years: the BIOS experiment revisited.Marine Pollution Bulletin 44, 1236–1242.

Prince, R.C., Stibrany, R.T., Hardenstine, J., Douglas, G.S., Owens,E.H., 2002. Aqueous Vapor Extraction; a previously unrecog-nized weathering process for oil spills in vigorously aeratedwater. Environmental Science & Technology 36, 2822–2825.

Radwan, S.S., Al Mailem, D., El Nemr, I., Salamah, S., 2000.Enhanced remediation of hydrocarbon contaminated desert soil

fertilized with organic carbons. International Biodeterioration &Biodegradation 46, 129–132.Rice, J.A., 2001. Humin. Soil Science 166, 848–857.Sergy, G.A., Blackall, P.J., 1987. Design and conclusion of the

Baffin Island Oil Spill project. Arctic 40 (Suppl. 1), 1–9.Sergy, G., Guenette, C.C., Owens, E., Prince, R.C., Lee, K., 1999.

Treatment of oiled sediment shorelines by sediment relocation.In: Proceedings of the 1999 International Oil Spill Conference.American Petroleum Institute, Washington, DC, pp. 549–554.

Sharma, V.K., Hicks, S.D., Rivera, W., Vazquez, F.G., 2002.Characterization and degradation of petroleum hydrocarbonsfollowing an oil spill into a coastal environment of south Texas,USA. Water Air and Soil Pollution 134, 111–127.

Spormann, A.M., Widdel, F., 2000. Metabolism of alkylbenzenes,alkanes, and other hydrocarbons in anaerobic bacteria. Bio-degradation 11, 85–105.

Thomas, D., 1993. The Braer incident at first hand. Petroleum

Review (UK) 47, 263–265.Thomas, R., Lunel, T., 1993. The Braer Incident; dispersion in

action. In: Proceedings of the Sixteenth Arctic Marine OilspillProgram Technical Seminar, Edmonton, Alberta, pp. 843–859.

Tissot, B.P., Welte, D.H., 1984. Petroleum Formation and Occur-rence. Springer-Verlag, Berlin.

Wang, Z., Fingas, M., Sergy, G., 1994. Study of 22-year-old Arrowoil samples using biomarker compounds by GC/MS. Environ-mental Science & Technology 28 (9), 1733–1746.

Wang, Z., Fingas, M.F., Sergy, G., 1995. Chemical characterizationof crude oil residues from an Arctic Beach by GC/MS and GC/FID. Environmental Science & Technology 29, 2622–2631.

Wang, Z., Fingas, M., Blenkinsopp, S., Sergy, G., Landriault, M.,Sigouin, L., Lambert, P., 1998. Study of the 25-year-old NipisiOil Spill: persistence of oil residues and comparisons betweensurface and subsurface sediments. Environmental Science &Technology 32, 2222–2232.

Wang, Z., Fingas, M.F., Sigouin, L., Owens, E.H., 2001. Fate andpersistence of long-term spilled Metula oil in the marine saltmarsh: degradation of biomarkers. In: Proceedings of the 2001International Oil Spill Conference. American Petroleum Insti-tute, Washington, DC, pp. 115–125.




12/12

Weise, A.M., Nalewajko, C., Lee, K., 1999. Oil-mineral fineinteractions facilitate oil biodegradation in seawater. Environ-

mental Technology 20, 811–824.

Yoshioka, G., Wong, E., Grossman, B., Drake, W., Urban, B.,Hudon, T., 1999. Past in situ burning possibilities. Spill Science

& Technology Bulletin 5, 349–351.

R.C. PRINCE et al.


photooxidation of cude oil

Documents