United States
Environmental Protection
Agency
Environmental Research
Laboratory
Gulf Breeze FL 32561
Research and Development
EPA-600/S3-84-071 July 1984
Project Summary
A Survey of the Toxicity and
Chemical Composition of Used
Drilling Muds
Chemical characterization andtoxicity
of oil drilling fluids were investigated
by the Edgerton Research Laboratory
from 1 October 1979 to August 1983
as part of a comprehensive research
program sponsored by the U.S. Envi-
ronmental Protection Agency (EPA) to
determine fate and effects of such
fluids in the marine environment.
Drilling muds used in the research were
supplied by the EPA, the Petroleum
Equipment Suppliers Association (PESA),
and the American Petroleum Institute
(API). The drilling muds were designated
"May 15," "May 29," "Sept. 4,"
"Exxon," "Gilson," "Mobile Bay,"
"Jay Field," and "PESA." Investiga-
tions during the first year centered on
the chemical composition and the acute
toxicity of drilling muds and the effects
of drilling muds on the recruitment of
benthic organisms. In the second year,
studies focused on toxicity testing with
planktonic copepods, chemical charac-
terization of the toxicity test phases,
bioaccumulation studies, and the effects
of muds on larval and adult benthic
organisms. Investigations during the
third and fourth years examined suble-
thal effects of drilling fluids on clam
larvae, trace metal and organic constit-
uents in both drilling fluids and toxicity
test-phases, and the preliminary devel-
opment of a drilling fluid solid phase
toxicity test. Toxic components of the
used drilling muds tested were present
as dissolved components or associated
with very slowly settling particles.
Some used drilling muds contained
lipophilic fractions that were similar to
hydrocarbons found in #2 fuel oil in the
liquid fraction and suspended particu-
lates fraction and contained #2 fuel oil
in whole muds. Muds that contained
those components were more toxic
than those that did not. Juvenile
copepods (Acartia tonsa) were not
more sensitive to toxic drilling mud
solutions than adults of this species. In
general. Cancer irroratus larvae appeared
to exhibit toxicity responses to drilling
muds that were similar to the copepods
tested. Arrested shell development
induced by exposure to drilling muds
appeared to be a sensitive indicator of
stress in bivalve larvae. Total chromium
concentration showed no correlation to
toxicity in the drilling muds that were
tested; however, the highest concentra-
tions of CR(VI), the most biologically
toxic form of chromium, occurred in the
test phases that exhibited the greatest
toxicity to Mercenaria mercenaria
larvae. The muds designated "May 15"
and "Sept. 4" appeared to be relatively
non-toxic to Pseudopleuronectes ameri-
canus and to Menidia menidia. although
the "May 15" mud was toxic to
Neomysis americana and to Acartia
tonsa. A study of the effects of drilling
mud on invertebrate recolonization of
defaunated sediment showed that
recolonization decreased in drilling mud
layered on top of sediment when the
muds were mixed with sediments.
Capitella capitata was much more
numerous in recolonization sediments
that contained drilling mud. Test results
showed that the methods used to
prepare drilling mud test media affect
the apparent toxicity of the muds.
This Project Summary was developed
by EPA's Environmental Research Lab-
-------
oratory. Gulf Breeze, Fl, to announce
key findings of the research project that
is fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
The possible environmental effects of
offshore oil-well drilling operations have
come under increasing scrutiny since the
mid 1970's. The discharge of spent
drilling muds, which are used in large
quantities during drilling operations for
drill-bit lubrication, borehole stabilization,
blow-out prevention, cooling, and removal
of drill cuttings, represent a potential
source of damage to the marine environ-
ment. Drilling muds are composed of
barite, clays (e.g. bentonite, attapulgite),
deflocculants (e.g. lignosulfonate), chela-
tors and organic lubricants in varying
amounts. The concentration of some
drilling mud components is changed
intentionally to match the varied geologi-
cal formations, temperature, and pressure
changes encountered in a drilling opera-
tion. In addition, the composition of a
particular mud is affected by the charac-
teristics of the geological substrate and
the drilling conditions. Because there is
no truly standard drilling mud, the task of
determining the potential toxicological
effects of this heterogeneous assemblage
of mixtures is complex. The most appro-
priate approach for toxicity testing has
been to test the effects of spent drilling
muds on a variety of marine animals
using standard aquatic bioassay tech-
niques.
The toxicity of both the liquid phase and
suspended solid phase of twelve spent
drilling muds was evaluated with a new
bioassay procedure using larvae of the
marine bivalve mollusc, Mercenaria
mercenaria The settled solids phase of
one of these drilling fluids was assessed
for toxicity using both a laboratory and
field recolonization study. These results
are presented below in summary form
under three headings: (1) Effects of used
drilling fluids on the embryonic develop-
ment of the hard clam Mercenaria
mercenaria; (2) Trace metals in drilling
fluid/sea water toxicity-test phases; (3)
Solid phase recolonization studies.
Effects of Used Drilling Fluids
on the Embryonic
Development of the Hard
Clam, Mercenaria mercenaria
Recent investigations have shown that
invertebrate larvae are much more
susceptible than adults to many pollu-
tants This phenomenon has been con-
firmed in our laboratory studies on larval
and adult bivalve molluscs. An under-
standing of the effects of drilling mud
components on bivalve larvae is particu-
larly important in view of the commercial
value of such species asArctica islandica,
Placopecten magellanicus and Spisula
solidissima These species are abundant
in offshore waters where drilling opera-
tions may be initiated. Although the hard
clam, Mercenaria mercenaria, is not
found in these offshore areas, the
convenience of utilizing this species for
laboratory conditioning, rearing, and
toxicity testing makes it a suitable
organism for larval bioassay research. In
addition, studies in our laboratory
showed that the response of M. mercena-
ria larvae to spent drilling muds is similar
to that of larvae of the sea scallop,
Placopecten magellanicus, a bivalve that
is much more difficult to maintain and
culture in the laboratory. A new 48-h
bioassay procedure was developed using
1 -h old, fertilized eggs of M. mercenaria
for assessing the toxicity of both liquid
and suspended solid phases of spent
drilling muds
Materials and Methods
Adult M. mercenaria were conditioned in
the laboratory to allow spawning through-
out the year. Spawning was initiated by
slow temperature increase. Following a
1-h fertilization period, developing eggs
were washed, counted, apportioned to
toxicity test vessels (Pyrex test tubes; 50
ml of test phase; 10-15 embryos mL-1)
and maintained at constant temperature
(27° C) for 48 h. Several concentrations of
both the liquid phase (supernatant from
mud suspension settled 72 h) and the
suspended solid phase (supernatant from
mud suspension settled 1 h. [This test
phase includes the liquid phase compo-
nents.]) of twelve spent drilling muds
were tested. At the completion of the 48-
h bioassay period, animals were fixed
with Lugol's solution and examined
under a dissecting microscope. The
criteria for toxicity were death (i.e. empty
shells) or obvious abnormal shell develop-
ment that would eventually lead to death.
A 48-h EC50 value was generated for
each drilling mud test phase using Probit
analysis.
Results
The suspended solids phase of spent
drilling muds was generally more toxic to
M. mercenaria developing larvae than the
liquid phase fractions alone (Fig. 1, 2).
However, these results were statistically
significant only for the spent muds
designated P1, P2 and P3. The remaining
eight muds did not show an appreciable
difference in toxicity between the two test
phases.
The 48-h EC50 values for the liquid
phase of 8 of the 12 muds ranged from
85-712 ppm (vol/vol mixture of a 72 h-
settled drilling mud suspension and 0.45
/um-filtered natural sea water to yield the
indicated concentration of drilling mud
liquid phase in sea water), whereas the
range for the suspended solids phases of
these muds was 64-382 ppm (vol/vol
mixture of 1 h-settled drilling mud and
0.45 /;m-filtered sea water). The EC50
values for the remaining 4 muds exceeded
2000 ppm, indicating that marked differ-
ences exist in the composition of spent
drilling muds.
Discussion
The results of our tests showed that
fertilized eggs of M. mercenaria are very
sensitive to both the liquid and suspended
solids phases of a diverse assortment of
used drilling muds. This larval quahog
toxicity test should serve as a useful tool
for identifying the potential impact of
drilling muds and other environmental
pollutants on the survival of commercially
important bivalve mollusc species in the
offshore environment. The data generated
m this study should be useful in deter-
mining maximum permissible mud dis-
charge rates in coastal zones that serve
as important seasonal nurseries for
commercially valuable bivalve species.
Developing quahog embryos have been
shown to be sensitive to a variety of used
drilling muds. However, it has been
impossible to correlate toxicity conclu-
sively with specific mud components.
Further toxicological studies are needed
to delineate the effects of turbidity and
particle loading, the adsorption of toxi-
cants to particulates, and the role of
microorganisms in the biodegradation of
various drilling mud components.
Trace Metals in Drilling
Fluid/Sea Water Toxicity Test
Phases
Trace metal analysis of the drilling
mud/sea water toxicity test phases was
conducted in order to identify inorganic
components of muds that may be toxic.
The metals of interest included barium,
cadmium, chromium, copper, manganese,
nickel, lead, and zinc. These metals can
occur in several different forms, or
species, in the marine environment. The
forms that metals assume in solution
affect their bioavailability and, thus, their
toxicity. For this reason, a scheme was
-------
5000-
4000-
3000-
2000-
1000-
*Bars indicate 95% Confidence intervals.
d Liquid Phase
E3 Suspended Solids
Phase
800-
700
600-
500
-^ 400
5
-1 300 •
c
.0
Concentrat
-> N)
8 8
80
70
60
50
40
30
20
F1]
*
=i
E:
E
E
~""
E
E
=
' .-:
'•: ":
=
E
E
=
E
=
',
rh
=
=
E
E
E
E
E
'1
pla
E:
=
—
E
E
^:
E
E
l'
=
E
—
=
=i
E:
E:
'IT
..
••::
SI/- 76 P-8 P-2 P-4 P-5 P-3
Fluid Type
:
E
I
-!
A
irr
E
~
E
E
E
E
SEPT 4
Figure 1. ECso values for M mercenana larvae after a 48-hour exposure to drilling fluid.
developed to determine the various
forms, or species, of the trace elements
listed above, as well as their total concen-
trations in drilling mud/sea water
mixtures.
Three general types of metal species
were targeted in the speciation scheme;
these were: free ionic forms including
inorganic complexes (i.e. chloro-, hydroxy-,
etc.), organically bound metals, and
metals associated with particulates.
These classes of trace elements are
typically considered to be the major types
present.
In addition to the metal species
discussed above, chromium can exist in
two oxidation states in sea water.
Chromium(VI) is present as CrO42~ and is
considered highly toxic, whereas Cr(lll) as
Cr(OH)2+ • 4H2O is much less toxic
Determination of the oxidation state and
species of Cr was also included as part of
the speciation scheme.
Methods
The concentrations of several metals in
sea water mixtures of drilling fluids were
measured simultaneously using dc
plasma emission spectrometry. Total
metal concentrations were measured by
directly aspirating dilute sea water
suspensions of drilling fluids into the
plasma after settling times of 1 h
(suspended solids phase) or 72 h (liquid
phase). Solution phase metals were
determined after centrifugation of the
same mixtures.
Free metal concentrations were deter-
mined by performing equilibrium dialysis
separations of the drill ing mud/sea water
mixtures before spectrometric analysis. A
1000 molecular weight cut-off dialysis
bag containing filtered sea water was
placed in the mixture Metal species
small enough to pass through the pores of
the membrane migrated into the bag and
were measured and assumed to be
primarily free, hydrated metal ion.
Chromium was further speciated by
separating free Cr(lll) using Donnan
dialysis with cation exchange membranes
The difference between total free Cr and
free Cr(lll) gives the Cr(VI) concentration.
The Cr(VI) concentration was also mea-
sured where possible, using differential
pulse polarography
Results
Results from the measurement of
trace metals in drilling fluid-sea water
mixtures showed that the average con-
centrations of the detectable elements
decreased in the order Ba>Cr>Mn>Zn>Cu.
The concentrations of Cd, Ni and Pb were
below the detection limits of the measure-
ment system (0.02, 0.01 and 0.2 mg/L
respectively). All metals exhibited some
particle association in 1 h settled phases
(suspended solids) with Ba being present
principally in the paniculate form.
Chromium and Cu were bound, probably
as lignosulfonate complexes, but Mn and
Zn were primarily in free forms. A
significant portion of the Cr was present
as highly toxic Cr(VI) in two of six muds
analyzed for this form of Cr.
Table 1 gives a breakdown of some Cr
species for some of the muds that were
tested and is an example of the metal
speciation results obtained For each
drilling fluid, the liquid and the suspended
solids test phases were analyzed at
specific concentrations of drilling fluid in
filtered sea water. The liquid phases
showed only a small difference between
solution Cr and total Cr concentrations.
As expected, only a small amount of
presumably colloidal Cr remains sus-
pended after the 72 h settling time.
Conversely, the suspended solids phases
have substantially more Cr associated
-------
5000 -,
4000-
*Bars indicate 95% Confidence Intervals
L_] Liquid Phase
Suspended So/ids
Phase
3000-
2000-
1000 -
800 -
700 -
600-
500-
400-
200-
100-
80-
70-
60-
50-
40-
30-
20-
h
>3(
E
\
E=3
=
=
',
1
E
=
=
100
>
1
^
E
E
E
I
30C
10 >
=
E:
E
— -
1
^
E
=
30
n
00
— I
=
=:
=
=
=
=
h
i
=
E:
=
=
I
p-1
P-6
SEPT 4
Figure 2.
AN31 P-7 MIBLKA51
Fluid Type
ECso values for M. mercenana larvae after a 48-hour exposure to drilling fluids
with particles than is present as dissolved
species.
The free Cr concentrations found for
the liquid phases accounted for approxi-
mately half of the dissolved Cr present for
all but the P8 drilling fluid. This result was
somewhat surprising, based on the rela-
tively high concentration of lignosulfo-
nate and lignite present in these muds
with which Cr could react.
Further speciation of free Cr into free
Cr(lll) and free Cr(VI) for some of the muds
tested is presented in Table 2. An
interesting point is that the free Cr(lll)
values listed (Table 2) are all very close to
0.020 mg/L, while the total free Cr (Table
2, column 1) values vary by almost an
order of magnitude. This strongly suggests
that the Cr(lll) concentrations were
solubility limited.
The difference between free Cr(lll) and
total free Cr gives the free Cr(VI) concen-
tration (Table 2, column 3) The Cr(VI)
concentration found for SV76 was the
highest by far, with P1 being the only
other drilling fluid with a significant
concentration The determination of
Cr(VI) directly by an alternate method,
differential pulse polarography, resulted
in values that were somewhat lower by
approximately a factor of two. These
results indicate that previous assump-
tions asserting the absence of potentially
toxic Cr(VI) in drilling fluids were incorrect.
Chromium (VI) is present in some used
drilling fluids and in dilute sea water
mixtures of these fluids at levels that
could result in toxic effects on marine
organisms
The potential threat of metal toxicity,
bioaccumulation, and food chain biomag-
nification is greater from Cr than from
other elements tested. Although most of
the Cr is probably present as Cr(lll)
complexes of lignosulfonate, its concen-
tration is relatively high The lack of
toxicity usually observed for Cr(lll) has
been attributed to its low solubility
Lignosulfonate forms complexes of Cr(lll)
and increases its solubility, thus increasing
the potential threat of this form of Cr in
drilling muds. In addition, Cr(VI) is often
the form of Cr added to lignosulfonate to
prepare chrome or ferrochrome lignosul-
fonate Chromium(VI) salts are sometimes
added to drilling muds as well, and large
quantities are used during preparation,
making it very likely that some Cr(VI) will
remain unreacted in muds with pH
conditions that are unfavorable for Cr(VI)
reduction. In oxygenated sea water, the
stable form of dissolved Cr is Cr(VI). It has
been demonstrated that Cr(lll) is present,
but slow oxidation occurs in the presence
of 02 and is catalyzed by manganese
oxide. This suggests that any Cr added to
the ocean, regardless of the form, is
potentially harmful.
Solid Phase Recolonization
Studies
The assessment of toxicity in solid
phase pollutants has traditionally been
much more difficult than in water soluble
or suspended solid pollutants. One
approach has been to measure either
survival or physiological and biochemical
effects of solid phase pollutants on
infaunal invertebrates A second approach,
reported here, has involved monitoring
the recolonization of natural, defaunated
sediments containing potentially toxic
solid phase pollutants to determine
toxicity. Laboratory-based and field-
based experiments were used to evaluate
the toxicity of a spent drilling fluid based
on its effects on the recolonization of
natural sediments by benthic organisms
-------
Table 1. Concentration and Spec/at/on of Chromium in Drilling Fluid/Seawater Test Phases
Drilling
Fluid
SV76
P1
P2
P8
Type of
Phase3
Liquid
Suspended
Avg. Liq. (2)
Suspended
Avg. Liq. (21
Suspended
Avg. Liq (2)
Avg Sus. (21
Phase
Cone.
(mULf
3.0
0.15
30
1.0
30
05
30
3.0
Concentration of Chromium (mg/L)
Total
0 59±0.02
0. 135+0.008
0 155+0 004
0 1 10+0.003
0044+0.005
0 083+0.003
027 +002
06 +02
Solution
0.543+0004
0051+0003
0.15 +0.02
0.037+O.OO3
0.041+0.003
0016+0.003
0.264+0.008
0.275+0.003
Free"
0.214+0007
0.067+0.002
0.025+0 002
0031+0.006
^Replicate phases are expressed as an average with number of replicates in parentheses
^Concentrations are mL of whole mud per L of 0.45 um filtered sea water.
"Used 1,000 molecular weight cut-off membranes for one test phase in duplicate (see text).
Table 2. Free Chromium (III) and Chromium (VI) from Donnan Dialysis of Drilling Fluid/
Seawater Liquid Phases3
Drilling
Fluid
SV76
PI
P2
P8
Concentrations of Chromium (mg/L)
Free Crh
0.214+0.007
0.067+0.002
0 025+0 002
0031+0006
Free Cr(lll)
0.025+0.007
0.020+0.004
0014+0.007
0 022±0.009
Cr(VI) by difference*
0.19 ±0.01
0047+0.004
0.011+0.007
0.009+0.009
*AII phases were 3.0 mL of drilling fluid per L of 0.45 um filtered sea water
"From Table 1.
cCr(V// concentrations are the difference between free Cr and Cr(lll) values.
Materials and Methods
The matrices of the laboratory- and
field-based experiments were identical
The tests included three treatments of
sediment: a natural, fine-grained, defaun-
ated reference sediment (control); a dril-
ling mud mixed with the reference sedi-
ment (homogeneous test); and drilling
mud deposited on the surface of the refer-
ence sediment (surface test). Five repli-
cate samples of each treatment were re-
moved after two weeks, four weeks, and
six weeks of recolonization exposure.
Upon collection, samples from both the
laboratory and field studies were sieved
through a 0.25 mm screen and preserved
in 10% formalin. All animals were
subsequently identified to the lowest
possible taxon and counted.
Three parameters were used for
statistical analysis of samples from each
recruitment period of each experiment.
(1) number of individuals; (2) number of
species; and (3) ratio of numbers of
species and individuals. The number of
individuals indicates the overall abund-
ance in a unit area. The number of
species is a measure of variety of species
richness Both of these determine the
third parameter, number of species/num-
ber of individuals (S/N), which is a simple
estimator of diversity, uncorrected for
sample size or evenness of species
distribution. Analysis of variance and the
Student-Newman-Keuls multiple range
test were performed to compare the
treatments for the above parameters
Student's t-test was used for groups of
data in which only two treatments were
being compared. In all statistical tests, a
95% confidence level was used (P <
0.05). Recolonizing populations were
also characterized qualitatively by distri-
bution of individuals within a phylum and
by species predominance. Species were
considered predominant on the following
basis' each predominant species occurred
in at least 60% of the replicates and
contributed at least 4% of the total
animals in a treatment. Less abundant
species were also included until 75% of
the total animals in a sample were
accounted for.
Results
Mean numbers (+ standard deviation)
of animals recovered for all samples
analyzed in both experiments are displayed
in FigureS. Numbers of animals increased
over time in both laboratory- and field-
based experiments. The latter experiments
had a higher number of animals than the
former after both two- and four-week
recruitment periods. In general, the
homogeneous test and control sediments
displayed higher mean and total numbers
of individuals, number of species, and
number of phyla than the surface test
samples.
Discussion
In general, the data show that a used
drilling fluid affected recolonization when
layered on top of defaunated sediment
but not when mixed with it. In both
experiments, deposition of a new layer of
detrital material on top of the drilling fluid
seemed to reduce or reverse the effects;
following six weeks of exposure, the
effects of the drilling fluid were no longer
obvious.
The reduced numbers of individuals
found in surface-test samples could have
been caused by physical or chemical
aspects of the drilling fluid. The evidence
suggests a physical mechanism for the
observed reduction in recolonization in
surface tests. First, if the effect was
chemical, adverse effects would be
expected to occur to a lesser extent in the
homogeneous test samples. This was not
the case. In fact, when homogeneous
samples differed from the other two
treatments, they contained slightly higher
numbers of animals. Second, the effect
ceased when animals were no longer in
direct contact with the layer of drilling
fluid, yet chemical effects would probably
have persisted, since toxicants could
continue to leach upward from the
drilling fluid through a thin detrital layer.
Finally, additional analyses of organic
and trace metal components of the
drilling mud used in the benthic recoloni-
zation toxicity test showed this mud to
have a relatively low toxicity.
-------
200
150
=» -g ;oo
•* -S
50
• Field-based experiment
• Laboratory-based experiment
Control
Homogeneous test
Surface test
Figure 3.
246
Recruitment Period (Wks)
Mean number of individuals collected after two-, four-, and six-week recruitment
periods for control, homogeneous and surface test treatments in laboratory- and
field-based experiments. Data are mean of individuals forn = 5 replicates (except for
lab-based two-week homogeneous test, where n = 4). Vertical bars indicate standard
deviations.
•&U. S. GOVERNMENT PRINTING OFFICE: 1984/759-102/10627
-------
This Project Summary was prepared by staff of the Edgerton Research Laboratory,
New England Aquarium, Boston. MA 021 JO.
T. W. Duke is the EPA Project Officer (see below).
The complete report, entitled "A Survey of the Toxicity and Chemical Composition
of Used Drilling Muds," (Order No. PB 84-207 661; Cost: $13.00, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
Sabine Island
Gulf Breeze, FL 32561
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
00604
------- |