Full text of DOI:10.1046/j.1526-0984.2001.82003.x

1
10 E N V I R O N M E N T A L G E O S C I E N C E S
Characterization and Bioremediation of a
Weathered Oil Sludge
W. R. GILES, JR., K. D. KRIEL, and J. R. STEWART
The Department of Biology, The University of Texas at Tyler, Tyler, TX 75799
Of those reported, such as tank bottoms and other refinery
ABSTRACT
•
The feasibility of bioremediating a weathered petroleum sludge
was studied both in the laboratory and in field composting. The
sludge consisted of straight-chained alkanes ranging from C20 to
C , with a halogen content of 4.1% and a specific gravity of
residues, none had undergone long-term storage or exposure
to climatic conditions, two properties that will alter the
chemical composition of the sludge. In addition, there is a
lack of complete characterization of the microbes used in
these studies, and methods of tracking hydrocarbon degra-
dation have ranged from Soxhlet extraction to CO2 evolu-
tion. These inconsistencies make comparisons difficult and
provide little guidance for future studies of other sludges in
the laboratory or in the field.
Sludges that are allowed to remain exposed over extended
periods to biological and abiological processes, such as lim-
ited microbial degradation, volatilization, and auto-oxida-
tion, are known as “weathered sludges” (Floodgate, 1984).
Components of weathered sludges become more and more
recalcitrant because microbial degradation is limited usually
due to oxygen availability and because the hydrocarbon-
degrading organisms degrade the most approachable or bio-
degradable molecules first. In these weathered sludges, the
mixture becomes an enrichment of extremely long-chained
alkanes, and in some cases may contain high concentrations
of aromatic hydrocarbons as well as heavy metals. At high
concentrations, toxicity for microbes can result and may
even pose health risks to persons handling the sludge (Bar-
tha and Bossert, 1984).
Because individual microbes are limited in the types of
hydrocarbons they can metabolize (Britton, 1984), a consor-
tium of hydrocarbon-degrading bacteria is usually em-
ployed in an effort to attain degradation of a wider array of
hydrocarbons, leading to more complete bioremediation. A
process known as cometabolism may also be utilized to in-
crease degradation. In this process, bacteria that partially
metabolize some hydrocarbons (e.g., cleaving a ringed com-
pound) provide products that can be degraded by other bac-
teria. Some non-hydrocarbon-degrading bacteria may then
be used in an effort to eliminate a much broader range of
hydrocarbons (Perry, 1979). When a particular petroleum,
however, contains hydrocarbons that are beyond the degra-
dative abilities of most bacterial consortia, the degradation
process cannot be complete.
3
8
0
.91–0.92 (23–24 API). Twenty indigenous bacteria were iso-
lated from the sludge and identified; all were mesophiles and
grew with or without 3.6% sodium chloride. Nine bacteria used
sludge as the sole carbon and energy source and were used in the
feasibility and composting studies. The sludge-degraders reduced
the total petroleum hydrocarbon (TPH) by 97.4% in 10 weeks
4
in shake cultures with 46 ꢀg/g; at TPH values of 4.1 ꢁ 10 and
5
1
.0 ꢁ 10 ꢀg/g the bacteria reduced the sludge by about 50% in
5
two weeks and at 2.3 ꢁ 10 the sludge was toxic. There was no
evidence that cometabolism occurred, and a surfactant did not
enhance bioremediation. Piles composted with either sawdust or
complete compost (composted agricultural wastes) as bulking
agents were monitored by measuring the most probable numbers
and TPHs through 24 weeks. The initial TPHs ranged from 3.1–
4
3
.6 ꢁ 10 ꢀg/g, and the bioremediation rate and extent were bet-
ter using complete compost (72–75% degradation) than with
sawdust (56–58%). There was no measurable effect on bioreme-
diation by augmenting with the sludge-degraders, and most
5
6
ꢂ1
probable numbers ranged from 10 –10 cells g soil in saw-
7
8
ꢂ1
dust-bulked piles and 10 –10 cells g in complete compost-
bulked piles. After 18 weeks of composting, temperature means
were significantly different in the two bulking materials (32ꢃC
in complete compost and 41ꢃC in sawdust), but both were 41ꢃC
from weeks 19–24, during which time the bioremediation rate
did not change in any pile.
Key Words: weathered sludge, oil sludge, bioremediation, com-
posting.
INTRODUCTION
To date, research on “sludges” has primarily dealt with
those created by spills (Dibble and Bartha, 1979), effluents
•
(
Bossert et al., 1984), tank bottom residues (Cansfield and
Racz, 1978), or other refinery processes not clearly defined
Marshall and Devinny, 1988; Modi et al., 1980; Prado-
Jatar et al., 1993; Shailubhai et al., 1984; Shailubhai, 1986).
(
In the oil fields in East Texas, a process used to remove
and dispose of salt water associated with the production of
crude oil has led to the creation of weathered sludges that
are not conducive to bioremediation due to their anaerobic
©
2001, AAPG/DEG, 1075-9565/01/$15.00/0
Environmental Geosciences, Volume 8, Number 2, 2001 110–122

G I L E S E T A L . : W E A T H E R E D O I L S L U D G E 111
and toxic nature. After the crude oil is separated by heat
sulting material is referred to as filtered sludge (FS) as in
Figure 1.
Methods used to characterize the sludge chemically and
physically are as follows:
from its naturally associated salt water, the residual salt wa-
ter is placed into large concrete- or plastic-lined pits. Minute
droplets of crude oil components, too heavy to be separated
in the heating process, accumulate at the surface of the salt
water. The salt water is then pumped out of the pits and is
disposed of by deep-well injection, leaving a thin layer of
heavy hydrocarbons in the pit covered with approximately
1. general hydrocarbon components: gas chromatography-
mass spectrometry (GC-MS) EPA SW 846 method 8015
(Environmental Protection Agency, 1992)
2
3
4
5
6
. total halogen content: Hach method (Environmental
Protection Agency, 1992)
. polychlorinated biphenyl (PCB) content: EPA SW 846
method 8080 (Environmental Protection Agency, 1992)
. ash content: ASTM-IP method D482-80 (American
Society for Testing and Materials, 1980)
. water content: ASTM method E203 (American Society
for Testing and Materials, 1990b)
. specific gravity and American Petroleum Institute (API)
gravity: ASTM method D1429-74 (American Society
for Testing and Materials, 1976) and ASTM D1298
1
0% salt water. After years of repeated separations, the pits
can become filled with thick layers of sludge. Allowed to
weather without protection from the forces of nature for de-
cades, the sludge becomes primarily composed of the long-
est alkanes and as viscous as axle grease. The weathered
sludge is difficult to handle, has anaerobic microhabitats,
and is potentially toxic to microbes used in bioremediation
(
Bartha and Bossert, 1984; personal observations).
Outdoor storage pits exist in the hundreds in the East
Texas oil fields where the sludge they contain is a financial
liability to the salt water disposal companies that own them.
Most wastes associated with oil and gas exploration, devel-
opment, and production operations, such as pit sludges, are
exempted from federal hazardous waste regulation (Envi-
ronmental Protection Agency, 1988) and, in Texas, the dis-
posal of such wastes is strictly regulated by the Railroad
Commission of Texas (2000). Disposal of the sludge in
these pits is of interest from both an environmental stand-
point as well as an economic one because bioremediation
may be used effectively as a disposal method. Before large-
scale bioremediation can be tried, however, studies are
needed to investigate the feasibility of bioremediating such
wastes.
(American Society for Testing and Materials, 1990a),
respectively
7
8
9
. BTU content: ASTM method D 3286-17 (American
Society for Testing and Materials, 1977)
. pH determination: EPA method SW846 9045 (Envi-
ronmental Protection Agency, 1992)
. ignitability: EPA method SW846 section 2.1.1 (Envi-
ronmental Protection Agency, 1992)
Bacterial Isolation, Identification, and
Culturing Methods
Indigenous bacteria were obtained by spreading sludge on
agar dishes and picking clones from isolation streaks made
on Tryptic Soy agar (TSA) with and without 3.6% sodium
chloride. Multiple slants of each clone were incubated at 25,
35, 40, 45, 50, and 55ꢃC. Conventional microbiological tests
were performed on the isolates, and identification was done
using Bergey’s manuals (Krieg, 1984; Sneath, 1986; Staley,
1989; Holt et al., 1994). Sludge degradation by the isolates
was measured by using Bushnell-Haas basal salts medium
(B-H BSM) with resazurin as an indicator of oxidation as
previously described (Williams et al., 1998). Autoclaved FS
(1%) was used as the sole carbon and energy source, and
each bacterium was scored as sludge-degrader or nonde-
grader after four days at 35ꢃC; controls of B-H BSM, those
with resazurin but no FS, were inoculated and incubated in
the same way to eliminate growth due to organic matter in
the inoculum. Inocula were taken from slant cultures.
For all bioremediation experiments, inocula were from
Tryptic Soy broth (TSB) cultures shaken overnight (16–18
hr, 70 rpm) on a New Brunswick Scientific Model G-52 gy-
roshaker at 35ꢃC, and they consisted of 0.1 mL of each iso-
late of the sludge-degraders. In the cometabolism experi-
ment, inocula included sludge-degraders and nondegraders.
The purpose of this research was to study the feasibility
of bioremediating a weathered petroleum sludge from a salt
water collection pit. Studies included are: (1) chemical and
physical characterization of the sludge; (2) isolation and
identification of the endogenous sludge-degrading microbes;
(
3) determination of the biodegradability of the sludge; (4)
establishment of a sludge concentration nontoxic to bacte-
ria; (5) comparisons of sludge bioremediation rates to estab-
lish the efficacy of cometabolism, exogenous surfactant,
and crude oil-degrading bacteria; and (6) the feasibility of
composting for bioremediation.
MATERIALS AND METHODS
•
Sludge Preparation and Characterization
Sludge that had weathered three decades in an open col-
lection pit was provided by Ricky Clements of East Texas
Salt Water Disposal Co. (Kilgore, Texas). The sludge was
solubilized in Freon (1,1,2-trichloro-1,2,2-trifluoroethane)
and filtered through glass wool to remove any large particu-
lates of decaying plant and animal materials. After remov-
ing the Freon by heating in a water bath (60ꢃC, 1 hr), the re-

1
12 E N V I R O N M E N T A L G E O S C I E N C E S
Those experiments comparing sludge-degrading with crude
oil-degrading bacteria used the consortia in Table 1 from
The University of Texas at Tyler Bacterial Culture Collec-
tion, and the bacteria were maintained and cultured as just
described. Controls for each experiment received no inocu-
lations, and they were prepared and treated in a manner
identical to the experimental flasks.
All laboratory experiments were performed using FS in B-
H BSM at a final volume of 150 mL in 500 mL Erlenmeyer
flasks, shaken on the gyroshaker at 35ꢃC. Experimental treat-
ments and controls were performed in 3 to 15 replicates, and
results of all experiments are reported as means of replicates
with standard errors. ANOVAs were used to identify signifi-
cant changes in total petroleum hydrocarbons.
Feasibility of Bioremediation and Toxicity
of Sludge
A concentration of sludge was used that was low enough
to eliminate the possibility of toxicity and to maximize the
surface area of available carbon source to bacteria, yet high
enough to be efficiently extracted using Freon in a separa-
tory funnel and detected by infrared spectroscopy. To facili-
FIGURE 1. Protocols for sludge preparation, extraction, and quantification.

G I L E S E T A L . : W E A T H E R E D O I L S L U D G E 113
tate such a minute concentration, 15 g of FS were dissolved
in 100 mL of Freon; each experimental and control flask re-
ceived 50 uL of FS, resulting in a TPH of 46 ꢀg/g. Freon
extractions of TPHs were performed at the initiation of the
study to establish a baseline level; this was followed by ex-
traction of experimentals and controls at one-week intervals
for the first four weeks and at the end of 6 and 10 weeks.
There were 15 replicates used to establish a baseline with 13
experimental replicates for weeks 1 and 2, and seven for
each subsequent interval. Controls consisted of three repli-
cates per interval, and their mean TPH level (auto-oxida-
tion) was subtracted from the baseline mean to eliminate the
amount of auto-oxidation; this is referred to as the adjusted
baseline level. Biodegradability was reported at each inter-
val as a percentage of TPH degraded in the experimentals as
compared to the adjusted baseline TPH.
To determine the qualitative changes at the conclusion of
the 10-week period, the seven experimental extractions
were combined, volatized to 25 mL, and analyzed for hy-
drocarbon content by GC-MS. The chromatogram was com-
pared to the chromatogram made on FS before remediation.
To establish the level at which the sludge became toxic to
the bacteria, flasks of B-H BSM containing 5, 10, 25, 50,
and 75% FS (v/v) were chosen empirically; these percent-
4
5
5
5
ages resulted in 4.1 ꢁ 10 , 1.0 ꢁ 10 , 2.3 ꢁ 10 , 4.6 ꢁ 10 ,
5
and 6.9 ꢁ 10 ꢀg/g, respectively. Baseline, control, and ex-
perimental flasks for each interval were in triplicate. Results
are presented after two weeks as the percentage of TPH de-
graded.
Comparative Studies
Cometabolism was investigated after a two-week period
by comparing degradation of sludge as the sole carbon and
energy source by sludge-degraders to that of sludge-degrad-
ers plus nondegraders.
Comparisons were made between experimental cultures
containing 1% exogenous, nonionic surfactant (APG 625CS
Glycoside; Henkel Corporation, Ambler, PA) and those
with no surfactant added over a six-week period, with mea-
surements made at two-week intervals.
A comparison of bioremediation by sludge bacteria to
that of crude oil-degrading bacteria was made after two
weeks, using the consortia in Table 1.
All comparative studies used triplicate cultures and in-
cluded baseline extractions and extractions of experimentals
and controls at the designated intervals; results are presented
as the percentage of TPH degraded. The initial sludge con-
4
centration in these three experiments was 9.0 ꢁ 10 ꢀg/g.
Extraction and Analytical Methods
In the feasibility experiments, FS was extracted from broth
cultures via a separatory funnel (Figure 1). Flasks were ex-

1
14 E N V I R O N M E N T A L G E O S C I E N C E S
TABLE 2. Bacteria isolated and identified from sludge:
Degraders and nondegraders
tracted with Freon (25 mL) divided into two equal portions,
with the extracts being combined before quantification.
In all other laboratory experiments, hydrocarbons (HCs)
were extracted via supercritical fluid extraction using super-
Degraders
Bacillus firmis
Bacillus fastidiosus
Bacillus coagulans
Nondegraders
Bacillus acidocaldarius
Bacillus circulans
Capnocytophaga sp.
critical CO in Freon on a Dionex Series 600 Supercritical
2
Fluid Chromotography/Gas Chromotography. To the aque-
ous sludge, 200 g of hydrocarbon-free sand was added (Fig-
ure 1). After drying the cultures overnight in a drying oven
Brochothrix campestrus
Cellulomonas fimi
Unidentified Gꢅ rod
Enterobacter cloacae
Enterobacter agglomerans
Escherichia blattae
Cellulomonas gelida
Microbacterium imperiale
Curtobacterium faccumfacions
Klebsiella oxytoca
Escherichia adecarboxylata
Aeromonas sp.
(100ꢃC), the sand-sludge mixture was homogenized using a
mortar and pestle and extracted using supercritical CO ex-
2
traction in Method 3560 proposed by Lopez-Avila et al.
Pseudomonas solanacearum
Pseudomonas mendocina
(1992) and used by Williams et al. (1998). Extraction time
was established by extracting samples of soil containing a
known concentration of the sludge used in the laboratory
studies. Extractions were performed in duplicate at 30-min
intervals until results indicated complete extraction by no
increase of TPHs in subsequent extractions. All extracts
were treated with 5 g of silica gel before IR analysis (Per-
FIGURE 2. Mass spectra of FS before and after 10 weeks of bioremediation: (A) before bioremediation and (B) after bioremediation.

G I L E S E T A L . : W E A T H E R E D O I L S L U D G E 115
TABLE 3. Physical and chemical properties of filtered sludge.
kin-Elmer FTIR spectrometer, Model 1640), using a stan-
dard curve according to the EPA Method 418.1 (EPA, 1983).
GC-MS procedures were performed on a Hewlett Packard
Gas Chromatogram, Model HPGC 5890, Series II, inter-
faced with a Hewlett Packard Mass Selective Detector,
Model HP 5971, Series II. The column was a Hewlett Pack-
ard HP-5MS with dimensions of 30 m ꢁ 0.25 um ID. The
flow rate was 1 mL minꢂ1, vacuum-compensated with elec-
tronic pressure control for constant flow rate. The instru-
ment was calibrated using a mixture of straight-chained al-
Specific Gravity
0.91–0.92 (23–24 API)
6.1
pH
BTU
ꢂ1
14,388 BTUIb
Ignitability
Neg.
2.8%
92.1%
5.1%
4.1%
Negligible
Negligible
Water content
Total hydrocarbon content
Ash content
Total halogen content
Heavy metals
PCB
9
18
20
22
24
26
28
30
32
34
36
kanes (C –C , C , C , C , C , C , C , C , C , C ,
3
8
40
42
44
C , C , C , and C ).
FIGURE 3. Feasibility of sludge degradation in aqueous solution through 10 weeks. Bars represent the means of the percent degradation with standard errors of 13 replicates for
weeks 1 and 2 and seven replicates for subsequent weeks at a TPH of 46 ꢀg/g.

1
16 E N V I R O N M E N T A L G E O S C I E N C E S
w/w; American Plant Food Corporation (Galena Park, TX)
were added. Each of the two piles was divided into two
smaller piles for a total of four piles, piles 1 and 2 with saw-
dust and piles 3 and 4 with complete compost. Piles 2 and 4
were inoculated initially and at 3 and 15 weeks with 4.5 L
(0.5 L per isolate) of sludge-degraders, grown in TSB for 24
hr at 35C, while piles 2 and 4 received 4.5 L of water at the
same intervals. After mixing, each of the four piles was cov-
ered with plastic sheeting to maintain a water-holding ca-
pacity (W-HC) of 30% to 90% (Dibble and Bartha, 1979;
Sims et al., 1989; Williams et al. 1998) and to prevent cross-
Composting Study
Two bulking agents were used: sawdust and a composted
material prepared from agricultural waste (complete com-
post). The complete compost was prepared and supplied by
Scott Hammer of Vital Earth Resources, Inc. (Gladewater,
TX). Two piles of 14% sludge were prepared by adding
3
3
2
.74 m of sludge to 16.46 m of each bulking agent and
mixing until homogeneous using a front-end loader. The fi-
nal TPHs of the piles ranged from 3.1–3.6 ꢁ 10 . To each
pile, 37.19 kg of commercial ammonium sulfate (12 N:0 P:0 K,
4
FIGURE 4. Toxicity of sludge at differing TPHs after a two-week period. Bars represent means of percent degradation of three replicates with standard errors.

G I L E S E T A L . : W E A T H E R E D O I L S L U D G E 117
contamination from runoff of rainwater. Internal tempera-
Williams et al. (1998) and changes in the TPHs at three-
week intervals. Four samples were taken from each of the
piles after mixing with a front-end loader at each interval,
and the samples were homogenized for each MPN and TPH
determination. TPHs are presented as the percent of sludge
degraded at each interval and MPNs are expressed as cells
per gram dry weight of sample. False-positive MPNs were
eliminated by using controls identical to the unknowns ex-
cept without the addition of FS. W-HC of the samples was
measured (Klute, 1986) at each interval, and water was
added to each pile as needed to maintain 30–90% W-HC as
tures of the piles were monitored by inserting a temperature
probe (0.3 m long) into each pile and allowing the tempera-
ture gauge to equilibrate before reading. Temperatures were
taken twice weekly (first and fourth days of each week) and
reported as means with standard errors. ANOVAs were
used to determine whether the temperature affected biore-
mediation significantly.
Bioremediation in the composting piles was monitored
for 24 weeks by determining the changes in the most proba-
ble number (MPN) of sludge-degraders as described by
4
FIGURE 5. Comparative degradation of sludge by six bacterial consortia. Bars represent means with standard errors of three replicates at TPH of 9.0 ꢁ 10 ꢀg/g. Consortia 1–5 are
listed in Table 1 and the sludge-degraders are identified in Table 2.

1
18 E N V I R O N M E N T A L G E O S C I E N C E S
recommended by Dibble and Bartha (1979), Sims et al.
1989), and Williams et al. (1998). Initially, the four sam-
ples from each pile were extracted separately for TPHs, but
after three weeks, when the variations in the four TPHs was
minimal as shown by standard deviations of the means, they
were homogenized for all of the remaining extractions.
At the end of the 10-week period, the TPH had an additional
significant reduction (p ꢄ 0.0371 with df ꢄ 1), resulting in
97.4% reduction. The GC-MS performed on the extracted
hydrocarbon after bioremediation is seen in Figure 2B. Only
one peak existed with a retention time of around 28 min;
this corresponds to a straight-chained alkane of C .
(
2
1
In the toxicity experiment, the sludge was degraded by
4
5
RESULTS
40–45% at 4.1 ꢁ 10 to 1 ꢁ 10 ꢀg/g (5–10% sludge), as
•
There were 20 indigenous bacteria isolated from the
shown in Figure 4, but the amount degraded dropped to ap-
proximately 20% at concentrations of 2.3 ꢁ 10 to 4.6 ꢁ
10 ꢀg/g (23–46% sludge). Only 6.7% was degraded at 6.9 ꢁ
10 ꢀg/g (69% sludge).
5
sludge and each grew equally well with or without 3.6% so-
dium chloride. All grew at 25, 35, and 40ꢃC but none grew
at 50 or 55ꢃC. The nine sludge-degraders and 11 nondegrad-
ers are listed in Table 2. As shown in the GC-MS in Figure
5
5
In the cometabolism experiments, 51% of the sludge was
degraded by the two consortia and 52% was degraded by
the sludge-degraders alone, resulting in no significant dif-
ference (p ꢄ 0.4572 with df ꢄ 1).
The difference in bioremediation with surfactant and
without surfactant was small, 49–63% with surfactant and
45–54% without surfactant. The difference was not signifi-
cant (p ꢄ 0.965 with df ꢄ 1).
In the comparative study between the sludge-degrading
bacterial consortium and five crude oil-degrading consortia,
the consortia of oil-degraders reduced the TPHs by 2.5–
34.5% in two weeks, as indicated in Figure 5, while sludge-
degraders reduced the TPH by 52% in the same period.
In the composting experiment, shown in Figure 6, the
2
A, FS had a range of hydrocarbons with approximate re-
tention times of 27 to 40 min, with the most abundant con-
stituent having a retention time of about 32 min. These re-
tention times correspond to straight-chained alkanes ranging
in length from C20 to C38, with the most abundant compo-
nent being C26 (32min). Other chemical and physical prop-
erties of the FS can be seen in Table 3 and include an or-
ganic content of 92.1%, a halogen content of 4.1%, and a
pH of 6.1. The FS has a specific gravity of 0.91–0.92 (23–24
API), and at room temperature it is a semi-solid.
In all bioremediation experiments, auto-oxidation was
less than 1% of the TPH content. As illustrated in Figure 3,
the sludge was degraded by an average of 61% after one
week and 79% after six weeks, which was the next statisti-
cally significant reduction (79%, p ꢄ 0.0128 with df ꢄ 1).
5
7
ꢂ
MPNs ranged from 10 –10 cells g 1 soil throughout the
experiment, except once when the numbers rose to 10 . The
8
FIGURE 6. Changes in MPNs of sludge-degraders in four composting piles through 24 weeks. Piles 1 and 2 were bulked with sawdust and piles 3 and 4 were bulked with complete
compost. Arrows indicate times when bacteria were added (4.5 L) to piles 2 and 4. Samples for MPNs and TPHs were taken at three-week intervals after mixing the piles.

G I L E S E T A L . : W E A T H E R E D O I L S L U D G E 119
FIGURE 7. Degradation of sludge in the four composting piles through 24 weeks. See Figure 6 for legend.
MPNs were consistently higher in the piles inoculated with
sludge-degrading bacteria. The bioremediation of the sludge
throughout the experiment, as seen in Figure 7, was almost
equal in the two piles bulked with complete compost (72%
and 75% at 24 weeks) and essentially equal in the two piles
bulked with sawdust (56–58%). W-HCs were 40–60% in
the four piles through the 24 weeks.
absence of bacteria (p ꢄ 0.480, df ꢄ 1; Table 4). During the
last six weeks of composting, the maximum ambient tem-
peratures were 32ꢃC or above for 38 days, with 18 of those
being 35ꢃC or higher.
DISCUSSION
•
Because the nine sludge-degraders grew at temperatures
The temperature means for the composting piles, shown
in Table 4, were approximately 41ꢃC for piles 1 and 2
below 45ꢃC but not at temperatures of 50ꢃC and higher, they
are mesophiles. Bacteria in this group grow optimally at 20–
45ꢃC with tolerances from 15–45ꢃC (Prescott et al., 1996).
The results of the GC-MS on the FS before remediation
(Figure 2A) confirmed that the sludge fit the characteristics,
both chemically and physically, of a weathered oil sludge.
The sludge was a mixture of only straight-chained alkanes
of C20 to C38 with no evidence of short chains, branched
chains, or cyclic hydrocarbons. This range is similar to a
substance described by Song et al. (1990) as “Bunker C”
and also identified as #6 fuel oil by Bruce and Schmidt
(1994). Hydrocarbons of the type that would give FS a more
(
(
bulked with sawdust) and 34–35ꢃC for piles 3 and 4
bulked with complete compost) after 24 weeks; these rep-
resent significant differences for the four piles (p ꢄ 0.0, df ꢄ
) and between the two bulking agents (p ꢄ 0.0, df ꢄ 1) but
no significance between the piles with or without bacteria (p ꢄ
.518, df ꢄ 1; see Table 4). However, during weeks 19–24,
the mean temperatures for the four piles ranged from 41–
2ꢃC (Table 4), and there were no significant differences
3
0
4
between the following: the four piles (p ꢄ 0.583, df ꢄ 3);
the bulking agents (p ꢄ 0.213, df ꢄ 1); or the presence and
TABLE 4. Temperature means for composting piles and their significance.
Temperature Means for Weeksb
Significance of Temperature Means
Piles and Treatmentsa 1–24 (n ꢄ 41) 1–18 (n ꢄ 30) 19–24 (n ꢄ 11) Analysis of Variance for
n ꢄ 41
n ꢄ 11
1
2
3
4
SD
SDꢅ
CC
41.3 ꢆ 4.3
41.1 ꢆ 4.9
35.2 ꢆ 4.8
34.2 ꢆ 5.7
41.0 ꢆ 4.7
41.1 ꢆ 5.6
33.3 ꢆ 3.8
31.8 ꢆ 4.5
41.9 ꢆ 2.8
41.2 ꢆ 2.4
40.5 ꢆ 2.8
40.6 ꢆ 2.6
Piles 1 vs 2 vs 3 vs 4
Piles 1 ꢅ 2 vs 3 ꢅ 4
Piles 1 ꢅ 3 vs 2 ꢅ 4
p ꢄ 0.0 (df ꢄ 3)
p ꢄ 0.0 (df ꢄ 1)
p ꢄ 0.518 (df ꢄ 1) p ꢄ 0.480 (df ꢄ 1)
p ꢄ 0.583 (df ꢄ 3)
p ꢄ 0.213 (df ꢄ 1)
CCꢅ
a
Treatments: SD, sawdust; CC, complete compost; ꢅ, bacteria added.
Temperatures were recorded twice weekly except for four weeks when only one was recorded.
b

1
20 E N V I R O N M E N T A L G E O S C I E N C E S
fluid character would have been removed during the heating
process used in the crude oil production, and there were no
hydrocarbons with vapor pressures low enough to be vola-
tile under conditions of open storage. The long-chained hy-
drocarbons with low volatility account for its semi-solid
state at room temperature. The less than 1% level of auto-
oxidation in the control flasks is also indicative of weath-
ered oil sludge, as most auto-oxidation would have already
occurred in the weathering process.
Some of the properties of the FS were indicative of its or-
igin and provide some interesting questions about the or-
ganisms isolated directly from the sludge. The 4.1% halo-
gen-containing sludge was covered by a layer of water with
dation (Figure 5; consortia 1, 2, 4, and 5); consortium 3,
however, did not acclimate well to the sludge environment
and degraded only 2.5%.
The MPNs in all four composting piles were always
5
7
ꢂ1
within 10 –10 cells g soil, except once when the num-
8
bers reached 10 (Figure 6). The MPNs were consistently
higher in piles augmented with sludge-degraders, usually in
7
the 10 range. In all instances, where bioremediation oc-
5
6
curred, the MPNs were at least in the 10 –10 range, which
agrees exactly with reports by Williams et al. (1998) for
soils with or without a history of oil pollution and lower
6
8
ꢂ1
than the 10 –10 cells g reported by Bossert and Bartha
(1984) in oil-polluted soils. The W-HCs of the piles were
40–60%, within the recommended range (Dibble and Bar-
tha, 1979; Sims et al., 1989; Williams et al., 1998).
At the TPHs used in the composting study (3–4%), the
MPN method was a valid way to calculate the populations
of hydrocarbon-degrading bacteria because the controls
(those lacking FS) were negative at the dilutions used to de-
termine cell numbers. This was evidence that the MPNs
were due to bacteria utilizing the sludge as sole carbon and
energy source and not some other organic material in the
samples. Williams et al. (1998) reported similar results in
two field studies.
1
0% sodium chloride, a level well above ideal for a sub-
strate for freshwater bacteria. This high halogen content,
undoubtedly resulting from the associated salt-water that
was separated from the crude oil during processing, seem-
ingly had no apparent effect on the bacteria.
It is feasible to bioremediate the sludge since 97% of the
TPH was degraded after 10 weeks when starting with 46 ꢀg/g.
5
The sludge was not toxic at 1 ꢁ 10 ꢀg/g TPH or 10% (Figure
4) where bioremediation occurred. This is twice the 5% opti-
mum concentration reported by Dibble and Bartha (1979).
The result of the GC-MS after 10 weeks of remediation
was qualitative evidence for the removal of hydrocarbons
from the sludge (Figure 2B). The flat baseline of the spec-
trum and the solitary peak with a retention time of 28 min
is indicative of only one remaining hydrocarbon type, cor-
responding to C . It is curious that a peak with the same
Bioremediation of the sludge (Figure 7) was essentially
the same in piles bulked with complete compost (piles 3 and
4) and approximately 16% greater than piles bulked with
sawdust (piles 1 and 2, 56 and 58%, respectively). Augmen-
tation of piles 2 and 4 with sludge-degraders had no measur-
able effect on the extent or rate of bioremediation, contrary
to what one might expect. The lower rate of bioremediation
using sawdust, augmented or not with sludge-degraders
(piles 1 and 2), was a surprise and it suggests that the bulk-
ing agent was more important to the process than microbial
augmentation. The higher rate using complete compost not
augmented with sludge-degraders (pile 3) undoubtedly means
there were hydrocarbon-degraders indigenous to the com-
plete compost.
It is obvious from Figure 7 that the extent and rate of
bioremediation were greater in the piles bulked with com-
plete compost and, from the temperature means in Table 4,
that the complete compost modulated the temperatures better
than sawdust, keeping them within the mesophilic range. At
the 19- to 24-week period, however, the means of all four
piles were in the upper mesophilic range, probably due to the
excessive solar heating when ambient temperatures were as
high as 35ꢃC. There being no significant differences in tem-
peratures between the four piles during weeks 19–24 (Table
4) perhaps explains why the MPNs declined slightly (Figure
6) and why the degradation rate plateaued (Figure 7).
The conclusions that can be drawn from the combination
of the laboratory studies and the field study described are as
follows:
2
1
retention time was present in the sludge before bioremedia-
tion, perhaps indicating that this hydrocarbon was recalci-
trant to mineralization.
Cometabolism has been observed in bioremediation of
petroleum, and it is generally thought to aid in the biodegra-
dation of cyclic and aromatic hydrocarbons primarily (Perry,
1
979). It is not surprising, then, that cometabolism was not
observed in this study.
Surfactants are often used to enhance bioremediation
rates of hydrocarbons by supplementing the natural produc-
tion of surfactants. The lack of enhanced degradation by this
Henkle surfactant does not mean another type might not
yield improved degradation rates. The long-chained alkanes
in this sludge may not be miscible with the nonpolar portion
of the surfactant molecules to the extent that the surfactant
was ineffective.
Those bacteria indigenous to the sludge were far better at
degrading the sludge than the consortia of crude oil-degrad-
ing bacteria (Figure 5). This illustrates the adaptation of in-
dividual bacterial populations to their environment as well
as the acclimation of some populations to environments to
which they are not indigenous. Those consortia that had the
ability to utilize the hydrocarbons in the sludge acclimated
quickly and were able to achieve a modest amount of degra-

G I L E S E T A L . : W E A T H E R E D O I L S L U D G E 121
1
. Weathered oil sludge was primarily long-chained hy-
drocarbons (C –C ) with a specific gravity of 0.91–
American Society for Testing and Materials (1980). Standard test
method for ash from petroleum products. Designation #D 482–
80, vol. 5.01. West Conshohocken, PA.
2
0
38
0
.92 (23–24 API), forming a semi-solid at room tem-
perature.
American Society for Testing and Materials (1990a). Density, rel-
ative density (specific gravity), or API gravity of crude petro-
leum and liquid petroleum products by hydrometer method.
2
3
. At a TPH of 46 ꢀg/g, the indigenous microbes de-
graded 97.4% in 10 weeks in shaken cultures.
4
5
E1
. Bioremediation occurred at 4.1 ꢁ 10 and 1.0 ꢁ 10
g/g in two weeks with toxicity resulting at a TPH of
Designation #D1298–85 (1190) , vol. 05.01 and 10.03 (revised
ꢀ
1994). West Conshohocken, PA.
5
2
.3 ꢁ 10 ꢀg/g and higher.
American Society for Testing and Materials (1990b). Test method
for water using Karl Fischer reagent. Designation #E203–92a,
vol. 15.05 (revised 1994). West Conshohocken, PA.
4
5
6
. Neither cometabolism nor exogenous surfactant had
any measurable effect on biodegradation of the weath-
ered oil sludge.
. None of the five bacterial consortia was as effective at
degrading the sludge as the bacteria indigenous to the
sludge.
. Composting with complete compost and sawdust as
bulking agents were effective means of bioremediating
the oil sludge at TPHs of 3–4 ꢁ 10 , and (a) the rate
and extent were greater in piles bulked with complete
compost than in those with sawdust (73 and 57%, re-
spectively); (b) temperatures in the complete compost-
bulked piles through 18 weeks were in the middle of
the mesophilic range, probably modulated by the com-
plete compost (complete compost, 32ꢃC; sawdust,
Bartha, R., and Bossert, I. (1984). Treatment and disposal of petro-
leum wastes. In R. M. Atlas (Ed.), Petroleum microbiology (pp.
553–577). New York: Macmillan Publishing Co.
Bossert, I., and Bartha, R. (1984). The fate of petroleum in soil
ecosystems. In R. M. Atlas (Ed.), Petroleum microbiology (pp.
435–473). New York: Macmillan Co.
4
Bossert, I., Kachel, W. M., and Bartha, R. (1984). Fate of hydro-
carbons during oily sludge disposal in soil. Applied and Envi-
ronmental Microbiology, 47, 763–767.
Britton, L. N. (1984). Microbial degradation of aliphatic hydrocar-
bons. In D. T. Gibson (Ed.), Microbial degradation of organic
compounds (pp. 89–129). New York: Marcel Dekker, Inc.
Bruce, L. G., and Schmidt, G. W. (1994). Hydrocarbon finger-
printing for application in forensic ecology: Review with case
studies. American Association of Petroleum Geology Bulletin,
78, 1692–1710.
4
1ꢃC); temperatures in the last 6 weeks, however, ap-
proached the thermophilic range in all four piles, dur-
ing which time the bioremediation rate did not change
in any piles; (c) augmentation with the sludge-degrad-
ers kept the MPNs higher in piles bulked with com-
plete compost and sawdust, but there were no corre-
sponding increases in sludge degradation.
Cansfield, P. E., and Racz, G. J. (1978). Degradation of hydrocar-
bon sludges in the soil. Journal of Soil Science, 58, 339–345.
Dibble, J. T., and Bartha, R. (1979). Effect of environmental pa-
rameters on the biodegradation of oil sludge. Applied and Envi-
ronmental Microbiology, 34, 729–739.
ACKNOWLEDGMENTS
Environmental Protection Agency (1983). In Methods for chemical
analysis of water and wastes. United States Environmental Pro-
tection Agency. EPA 600/14–79/020, 1979, revised 1983.
Environmental Protection Agency (1988). Regulatory determina-
tion for oil and gas and geothermal exploration, development
wastes. United States Environmental Protection Agency, Wash-
ington, DC. 53 Fed Reg 25.446, July 6.
•
This research was made possible by the generous finan-
cial support of the East Texas Salt Water Disposal Company,
Kilgore, Texas and the Welch Foundation, Houston, Texas.
The cooperation of Vital Earth Resources, Inc., Gladewater,
Texas was essential for the composting studies. The use of
the Zuckerman Laboratory of Electron Microscopy, Depart-
ment of Biology, The University of Texas at Tyler, Tyler,
Texas in bacterial identification is gratefully acknowledged.
This work is in partial fulfillment of the requirements for a
Master’s of Science (W.R.G.) in the Department of Biology,
The University of Texas at Tyler, Tyler, TX 75799.
Environmental Protection Agency (1992). In Methods for evaluat-
ing solid waste: Laboratory manual—Physical/chemical meth-
ods, SW-846, Revision 2. United States Environmental Protec-
tion Agency, Washington, DC. November.
Floodgate, G. D. (1984). The fate of petroleum in marine ecosys-
tems. In R. M. Atlas (Ed.), Petroleum microbiology (pp. 355–
397). New York: Macmillan Publishing Co.
REFERENCES
•
American Society for Testing and Materials (1976). Test methods
Holt, J. G., Krieg, N. R., Sneath, P. H. A., Staley, J. T., and Will-
iams, S. T., Editors (1994). Bergey’s manual of determinative
bacteriology (9th Edition). Baltimore: Williams and Wilkins.
Klute, A. (1986). Water retention: Laboratory methods. In A.
Klute (Ed.), Methods of soil analysis, part 1: Physical and min-
eralogical methods (2nd Edition), ch. 26. Madison, WI: Ameri-
can Society of Agronomy and Soil Science Society of America.
for specific gravity of water and brine. Designation #D1429–86,
vol. 11.01 (revised 1994). West Conshohocken, PA.
American Society for Testing and Materials (1977). Gross calo-
rific value of coal and coke by the isoperibol bomb calorimeter.
Designation #D3286–91, vol. 05.05 ( revised 1994). West Con-
shohocken, PA.

1
22 E N V I R O N M E N T A L G E O S C I E N C E S
Krieg, N. R., Editor (1984). Bergey’s manual of systematic bacte-
riology (vol. I). Baltimore: Williams and Wilkins.
Railroad Commission of Texas (2000). Texas Railroad Commission,
water protection. 16 Texas Administration Code 3.8(d), West.
Shailubhai, K. (1986). Treatment of petroleum industry oil sludge
in soil. Trends in Biotechnology, 4, 202–206.
Lopez-Avila, V., Benedicto, J., Dodhiwala, N. S., Young, R., and
Beckert, W. F. (1992). Development of off-line SFE-IR method
for petroleum hydrocarbons in soils. Journal of Chromato-
graphic Science, 30, 335–343.
Shailubhai, K., Rao, N. N., and Modi, V. V. (1984). Treatment of
petroleum industry oil sludge by Rhodotorula sp. Applied Mi-
crobiology and Biotechnology, 19, 437–438.
Marshall, T. R., and Devinny, J. S. (1988). The microbial ecosys-
tem in petroleum waste land treatment. Water Science Technol-
ogy, 20, 285–291.
Sims, J. L., Sims, R. C., and Matthew, J. E.. (1989). Bioremediation of
contaminated surface soils. Robert S. Kerr Environmental Re-
search Laboratory, U.S. EPA, ORD. Ada, OK. EPA/600/9–89/073.
Song, H., Wang, X., and Bartha, R. (1990). Bioremediation poten-
tial of terrestrial fuel spills. Applied and Environmental Micro-
biology, 53, 652–656.
Modi, V. V., Rao, N. N., and Shailubhai, K. (1980). Microbial
treatment of petroleum industry sludge and its evaluation. In B.
Patel (Ed.), Management of environment: New Delhi (pp. 172–
1
80). New Delhi, India: Eastern, Wiley & Sons.
Perry, J. J. (1979). Microbial cooxidations involving hydrocar-
bons. Microbiological Reviews, 43, 59–72.
Sneath, P. H. A., Editor (1986). Bergey’s manual of systematic
bacteriology (vol. II). Baltimore: Williams and Wilkins.
Staley, J. T., Editor (1989). Bergey’s manual of systematic bacteri-
ology. (vol. III). Baltimore: Williams and Wilkins.
Prado-Jatar, M., Correa, M., Rodriguez-Grau, J., and Carneiro, M.
(
1993). Oil sludge landfarming biodegradation experiment con-
ducted at a tropical site in eastern Venezuela. Waste Manage-
ment and Research, 11, 97–106.
Williams, D. L., Kriel, K. D., Stewart, G. A., Hulse, R. C., Hol-
somback Jr., J. E., and Stewart, J. R. (1998). Bioremediation of
oil-contaminated soils by stimulating indigenous microbes. En-
vironmental Geosciences, 1, 1–8.
Prescott, L. M., Harley, J. P., and Klein, D. A. (1996). Microbiol-
ogy. Boston: W. C. Brown.
ABOUT THE AUTHORS
•
William Russell Giles, Jr.
James R. Stewart, Ph.D.
James R. Stewart, Ph.D., is a Professor
Emeritus of Biology and Chemistry at The
University of Texas at Tyler (1974–2000)
and he was president of the Texas Branch
of the American Society for Microbiology
from 1999–2001. His research included
bacterial and algal physiology and biore-
mediation of oil spills and other petroleum
products.
William Russell Giles, Jr. received
Bachelor’s and Master’s degrees in 1995
and 1997, respectively. After graduating
from The University of Texas Medical
School at Houston in May 2001, he will
begin a residency in internal medicine and
pediatrics at Herman Hospital, Houston.
Kimberly Diane Kriel
Kimberly Diane Kriel received Bache-
lor’s and Master’s degrees in 1993 and
1999, respectively, and she worked as a re-
search assistant at The University of Texas
Medical Center at Tyler before starting the
home schooling of her two daughters.

Copyright of Environmental Geosciences is the property of Wiley-Blackwell and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for
individual use.