Page
1
of
30
July
29,
2004
Impact
of
Heat
Aging
on
Sediment
Toxicity
of
Ester/
Olefin
Based
Drilling
Muds
to
Leptocheirus
plumulosus
M.
L.
Cano
1,
S.
Rabke
2,
J.
Candler
2,
P.
B.
Dorn
3,
J.
T.
Louallen
3,
and
P.
Scott
4
1
Shell
Chemical
LP,
2
M­
I
SWACO,
3
Shell
Global
Solutions,
4
Marathon
Oil
Corporation
Page
2
of
30
July
29,
2004
ACKNOWLEDGEMENTS
This
research
was
funded
through
the
Synthetic
Based
Mud
(
SBM)
Research
Group
(
a
consortium
of
oil
and
gas
companies,
drilling
mud
suppliers,
and
drilling
mud
chemical
manufacturers).
We
greatly
appreciate
the
efforts
of
the
members
of
the
Sediment
Toxicity
Work
Group,
which
oversaw
and
reviewed
this
study.

Sediment
Toxicity
Work
Group
Members:

Andrew
Glickman,
ChevronTexaco
­
CoChair
Cheryl
Hood,
Baker
Hughes
Inteq
 
CoChair
Rene
Bernier,
ChevronTexaco
John
Candler,
M­
I
SWACO
Manuel
Cano,
Shell
Chemical
LP
James
Clark,
ExxonMobil
Philip
Dorn,
Shell
Global
Solutions
Jan
Farmer,
BP
Michael
Fefer,
PetroCanada
Carole
Fleming,
ChevronTexaco
Stefano
Garavini,
Sasol
North
America
Jeffry
Gee,
Chevron
Phillips
John
Hall,
Baroid/
Halliburton
Larry
Henry,
ChevronTexaco
Gail
Korenaga,
ChevronTexaco
Burney
Lee,
BP
Jeff
Louallen,
Shell
Global
Solutions
Peter
Loggenburg,
Alochem
Bob
MacGregor,
Baroid
Allen
Neilsen,
Sasol
North
America
Khai
Nguyen,
M­
I
SWACO
James
O'Reilly,
ExxonMobil
Mike
Parker,
ExxonMobil
Stephen
Rabke,
M­
I
SWACO
James
Ray,
Shell
Global
Solutions
Larry
Reitsema,
Marathon
Jesse
Roberts,
BHP
Billiton
James
Robinson,
BP
Paul
Scott,
Marathon
Maryam
Shekarchian,
Shrieve
Tom
Shumate,
Halliburton
Don
Van
Slyke,
Unocal
Jennifer
Walden,
Marathon
Diana
Wong,
Shell
Global
Solutions
Thomas
Purcell
 
API
Staff
Page
3
of
30
July
29,
2004
SUMMARY
Olefins,
esters
and
more
recently
olefin/
ester
blends
are
synthetic
based
drilling
fluids
(
SBF)
that
have
been
used
to
meet
regulatory
requirements.
This
study
provides
information
on
the
toxicity
performance
of
ester/
olefin
blends
under
high
temperature
well
conditions.
An
inter­
industry
group
designed
the
study
to
examine
the
effects
of
temperature
and
time
on
the
sediment
toxicity
of
ester/
olefin
muds.
Commercial
testing
laboratories
familiar
with
the
fluids
and
testing
protocols
conducted
the
tests.
The
first
phase
of
the
study
indicated
that
the
temperature
(
275
 
300
°
F)
and
time
(
16
 
160
hr)
scales
of
the
tests
bracketed
the
region
where
ester/
olefin
muds
transitioned
from
passing
to
failing
the
sediment
toxicity
limitation.
The
second
phase
of
the
study
used
a
statistical
experimental
design
to
identify
statistically
significant
factors
impacting
sediment
toxicity.
Four
factors
were
examined:
(
1)
time,
(
2)
temperature,
(
3)
alkalinity
via
the
absence
or
presence
of
green
cement,
and
(
4)
ester
content.
The
base
stock
blends
included
a
100
%
internal
olefin
(
IO1618),
a
blend
of
90%
IO1618
and
10%
traditional
ester
and
a
blend
of
70%
IO1618
and
30%
traditional
ester.
The
base
fluids
were
formulated
into
drilling
muds
and
tested
in
a
temperature
range
from
275
to
350
°
F
and
time
exposure
range
to
temperature
from
16
to
160
hours.

A
statistical
model
was
developed
to
predict
sediment
toxicity
ratios
(
STR)
as
a
function
of
time,
temperature,
ester
content,
and
alkalinity
via
the
presence
or
absence
of
green
cement.
The
model
was
developed
using
the
data
from
the
Phase
2
toxicity
tests.
Data
from
the
Phase
1
toxicity
tests
were
used
to
independently
test
and
validate
the
model.

The
sediment
toxicity
tests,
analytical
measurements
conducted
on
the
muds
used
for
Phases
1
and
2
of
this
study
along
with
the
statistical
STR
model
supported
the
hypothesis
that
toxicity
was
negatively
affected
by
ester/
olefin
blends
and
downhole
conditions
(
time,
temperature)
and
formulation
(
alkalinity
via
the
presence/
absence
of
green
cement).
The
data
suggest
that
under
some
conditions,
for
the
types
of
muds
used
in
this
study,
sediment
toxicity
will
increase
when
the
mud
is
exposed
to
elevated
temperatures
for
extended
periods
of
time.
Ester/
olefin
muds
with
>
10%
ester
are
likely
to
break
down
and
produce
breakdown
products
that
will
result
in
sediment
toxicity
and
failure
of
the
STR
<
1.0
criteria
required
for
cuttings
discharge.

2­
Ethyl
hexanol
(
2­
EH)
as
an
indicator
of
ester
hydrolysis
was
correlated
with
increased
STR
and
can
be
used
as
a
surrogate
to
screen
for
potential
sediment
toxicity.
On
average,
when
2­
EH
concentrations
exceed
2.0
wt%,
it
is
likely
that
muds
will
fail
sediment
toxicity
compliance
tests.
At
2­
EH
concentration
of
>
4
wt%,
it
is
95%
certain
that
the
muds
will
fail.

The
data
from
this
study
support
a
temperature
threshold
of
~
300
°
F
for
which
operators
should
be
cautious
with
the
use
of
drilling
fluids
containing
the
type
of
esters
(
traditional
esters)
used
in
this
study.
Above
300
°
F
these
materials
may
break
down
and
result
in
increased
sediment
toxicity
for
the
drilling
muds.
Other
types
of
esters
were
not
examined
in
this
study.

INTRODUCTION
Synthetic
based
drilling
fluids
(
SBF)
and
other
non­
aqueous
fluids
(
NAF)
have
been
accepted
for
use
in
the
US
Gulf
of
Mexico
(
USEPA,
2001a).
As
part
of
the
technology­
based
limits
that
were
promulgated
by
USEPA,
any
synthetic
based
mud
(
SBM)
must
meet
environmental
requirements
for
base
fluids
as
well
as
the
discharged
SBM
cuttings
offshore.
Base
fluids
must
meet
stock
limitations
for
anaerobic
degradation,
PAH
measurement
and
sediment
toxicity
by
performing
as
good
as
or
better
than
a
corresponding
C1618
internal
olefin
or
ester
reference
fluid.

During
the
data
collection
efforts
for
the
effluent
limitation
guidelines
(
USEPA,
2001b),
industry
provided
to
USEPA
information
indicating
that
the
technical
temperature
limitations
for
ester
Page
4
of
30
July
29,
2004
based
fluids
was
300
°
F
and
that
the
presence
of
cement
contamination
further
decreased
the
functional
temperature
range
of
ester
based
fluids
(
USEPA,
Docket
W­
98­
26).

Reports
from
the
literature
suggest
that
esters
are
thermally
stable
to
only
moderately
high
temperatures,
in
the
region
of
300
°
F,
and
that
these
temperatures
may
be
reduced
considerably
when
the
esters
are
exposed
to
high
pH.

Young
(
Attachment
Ester
52)
found
in
laboratory
experiments
that
sharp
changes
in
viscosity
of
esters
were
observed
at
a
temperature
of
about
310
°
F.
He
concluded
that
the
esters
were
unstable
above
that
temperature.
Patel
et
al.
(
Attachment
Ester
53)
in
laboratory
experiments
of
muds
made
with
a
group
of
esters
of
different
lengths
of
acid
and
alcohol
chains
found
that
the
temperatures
of
decomposition
ranged
from
225
to
above
275
°
F.
These
experiments
establish
a
thermal
limit
for
the
use
of
esters
at
no
higher
than
about
300
°
F.

Both
Patel
et
al.
(
Attachment
Ester
53)
and
Friedheim
(
Attachment
Ester
54)
have
studied
the
hydrolysis
of
esters
and
have
concluded
that
this
mechanism
may
be
a
major
factor
in
the
apparent
thermal
decomposition
of
these
compounds.
Field
evidence
is
provided
by
Shell's
experience
at
its
Spirit
well
the
Gulf
of
Mexico
in
1995,
when
the
conventional
ester
encountered
green
cement.
The
downhole
temperature
was
255
°
F,
and
the
flowline
temperature
(
the
temperature
of
the
mud
as
it
exited
the
well
to
the
shakers)
was
about
200
°
F;
the
mud
quickly
gelled,
becoming
unusable,
upon
contact
with
the
cement
at
that
elevated
temperature.
Thus,
under
the
usual
conditions
in
which
the
mud
is
run,
with
a
small
amount
of
excess
lime
in
the
water
phase,
the
mud
is
stable
to
a
temperature
in
excess
of
250
°
F,
possibly
as
high
as
300
°
F;
however,
when
the
mud
encounters
the
much
higher
pH
of
green
cement,
the
hydrolysis
temperature
(
the
stability
limit)
is
reduced
considerably,
in
the
field
case
to
about
200
°
F.
No
other
field
cases
of
hydrolysis
are
known
at
this
time.
Thus,
this
type
of
failure,
while
it
is
known,
has
not
proven
to
be
a
general
problem
in
the
field
use
of
muds
made
from
conventional
esters.

No
data
are
available
on
the
new
low
viscosity
esters,
and
so
no
evaluations
of
temperature
stability
are
possible.
Experience
with
other
esters
suggests,
however,
that
the
maximum
operational
temperature
will
likely
be
no
higher
than
about
300
°
F
and
that
this
temperature
will
likely
be
considerably
lowered
at
high
pH,
if
the
mud
is
run
with
too
much
excess
lime
in
the
water
phase,
for
example,
or
if
the
mud
encounters
green
cement.

USEPA
used
this
and
other
information
to
develop
a
technology
limitation
for
synthetic
based
muds.
After
the
development
of
the
guidelines
and
incorporation
of
the
limitations
into
General
Discharge
permits
the
USEPA
collected
additional
information
on
ester/
olefin
blends
and
indicated
that
the
temperature
stability
of
ester/
olefin
blends
may
be
higher
than
for
drilling
fluids
that
were
based
solely
on
ester
fluids.

I
n
documentation
from
2003
(
USEPA,
2003)
the
USEPA
recognized
limitations
of
using
ester
or
ester/
olefin
blends
in
high
temperature
wells
where
SBM
may
break
down.
However,
the
2003
information
indicated
the
upper
temperature
range
for
ester/
olefin
blends
was
in
the
range
of
350
 
375
°
F
(
USEPA,
2003).

However,
there
may
be
some
very
rare
occasions
where
ester/
internal
olefin
blends
cannot
be
used
due
to
technical
drilling
limitations.
In
particular,
we
are
still
concerned
that
the
molecular
structure
of
esters
presents
more
potential
drilling
problems
than
pure
C16­
C18
Internal
olefins
in
very
high
temperature
Page
5
of
30
July
29,
2004
environments
(
e.
g,.
greater
than
350
°
F).
In
particular,
some
exploration
of
deep
gas
reserves
in
shallow
waters
may
involve
drilling
in
very
high
temperatures
environments
(
e.
g,.
greater
than
350
°
F)
at
great
depths
(
e.
g.,
18,000
to
20,000
feet
vertical).
In
these
very
high
temperatures
environments
(
e.
g,.
greater
than
350
°
F)
esters
can
also
begin
thermal
degradation
(
i.
e.,
break­
down
to
their
parent
acids
and
alcohols)
which
will
transform
the
ester­
SBFs
into
a
thick
mass
(
Docket
No.
W­
98­
26,
Record
No.
IV.
A.
a.
3).
The
50/
50
C16­
C18
internal
olefins/
ester
blend
reference
above
is
rated
up
to
350
°
F.
Increasing
the
ratio
of
C16­
C18
internal
olefins
in
C16­
C18
internal
olefins/
ester
blends
will
increase
the
upper
temperature
range
available
for
the
blend
(
e.
g.,
a
80/
20
C16­
C18
internal
olefins/
ester
blend
may
able
to
perform
at
375
°
F).
It
is
important
to
note
that
these
very
high
temperature
environments
are
very
rarely
seen
in
deep
water
wells
were
down­
hole
temperatures
are
typically
below
300
°
F.
Therefore,
in
order
to
promote
SBFs
as
a
pollution
prevention
technology
for
the
exploration
of
shallow
water
deep
gas
reserves,
we
would
recommend
a
case­
by­
case
enforcement
discretion
for
any
operators
that
can
demonstrate
the
need
to
use
pure
C16­
C18
internal
olefins
(
see
40
CFR
435.11(
rr))
in
order
to
compensate
for
very
high
temperature
environments
(
e.
g,.
greater
than
350
°
F).
You
may
wish
to
require
operators
to
obtain
your
written
permission
on
a
case­
by­
case
basis
before
any
discharges
of
pure
C16­
C18
internal
olefins
associated
with
drill
cuttings.

The
general
mechanism
for
chemical
degradation
of
esters
is
known
to
be
via
hydrolysis,
and
typically
an
increase
in
2­
ethyl
hexanol
(
2­
EH)
is
observed
when
traditional
esters
hydrolyze.
While
the
general
mechanism
is
understood,
the
effects
of
temperature,
time
and
drilling
fluid
formulation
factors
have
not
been
documented
in
a
formal
inter­
industry
study.
Thus,
there
was
a
need
to
develop
the
data
to
document
the
effects
of
downhole
and
formulation
conditions
on
sediment
toxicity
compliance
results.

A
study
commissioned
by
the
inter­
industry
SBM
Toxicity
Workgroup
and
SBM
Technology
Assessment
Workgroup
set
objectives
to
understand
in
laboratory
controlled
conditions
the
effect
upon
toxicity
of
factors
involved
in
high
temperature
environments
upon
SBM
with
respect
to
formulation,
temperature,
alkalinity
via
the
presence/
absence
of
"
green
cement",
and
time
under
temperature.
The
testable
hypothesis
for
this
study
was
that
toxicity
was
negatively
affected
by
ester/
olefin
blends
and
downhole
conditions
(
time,
temperature)
and
formulation
(
presence/
absence
of
green
cement).

Unlike
laboratory
conditions,
field
conditions
cannot
be
prescribed
to
control
time
and
temperature
exposure
conditions.
Under
normal
circulating
conditions,
drilling
fluids
stay
at
bottom
hole
temperatures
for
only
a
few
minutes.
However,
when
bits
are
replaced,
or
casing
strings
are
run,
the
drilling
fluid
may
be
exposed
to
bottom
hole
temperatures
for
many
hours.
In
some
cases,
hurricane
evacuations,
well
control
problems
and
other
non­
scheduled
events
can
expose
the
drilling
fluid
to
bottom
hole
temperatures
for
many
days.
In
addition,
other
factors
such
as
removal
of
contaminants
with
solids
control
equipment
and
volatilization
add
to
the
dynamics
of
wellsite
conditions.
While
the
exposure
conditions
may
not
match
exact
well­
site
conditions
and
dynamics,
the
statistical
relationships
between
downhole
conditions
and
sediment
toxicity
identified
in
the
study
are
expected
to
be
predictive
of
relationships
under
field
conditions.

Of
primary
concern
to
offshore
operators
is
compliance
with
NPDES
permit
limitations.
A
synthetic
drilling
fluid
that
meets
the
technical
needs
of
the
wellbore
but
fails
to
comply
with
NPDES
limits
for
cuttings
discharge
is
of
no
practical
use
to
an
operator.
Operators
need
to
be
confident
that
a
drilling
fluid
rated
to
a
specific
temperature
limitation
will
perform
at
bottom
hole
temperatures
with
routine
exposure
and
contamination
conditions.
Page
6
of
30
July
29,
2004
EXPERIMENTAL
Mud
preparation
Drilling
muds
were
prepared
in
2
phases.
All
muds
were
prepared
using
a
C1618
internal
olefin.
The
ester
used
for
the
ester/
olefin
blend
muds
was
a
traditional
ester
manufactured
from
a
vegetable
oil
(
palm
kernel
oil)
and
2­
ethyl
hexanol
(
2­
EH).
Each
mud
sample
was
mixed
using
a
dispersator
in
a
plastic
container
over
a
period
of
2.5
hours.
The
products
were
added
in
the
order
as
listed
in
mud
formulation
tables
(
Tables
1
and
2)
at
around
3000­
4000
rpm.
The
fluids
were
allowed
to
mix
for
an
additional
30
minutes
after
adding
all
the
products.
The
muds
were
prepared
by
a
service
company's
laboratory
and
a
third
party
contract
laboratory
added
the
cement
contamination.
The
contract
laboratory
also
hot
rolled
the
muds
for
the
prescribed
time
(
16,
40,
88,
or
160
hr)
and
temperature
(
275,
300,
325,
or
350
°
F).
The
same
contract
laboratory
performed
mud
analyses
to
verify
mud
properties.

I
n
Phase
1,
all
muds
had
mud
property
analyses
and
analytical
analyses
(
see
below)
performed.
Only
muds
A
and
D
had
sediment
toxicity
tests
performed.
In
Phase
2
all
muds
has
analytical
analyses
(
see
below)
and
sediment
toxicity
tests
performed.

Two
different
sources
of
alkalinity
were
used.
In
Phase
1,
alkalinity
was
included
in
the
muds
in
the
form
of
lime.
In
Phase
2,
alkalinity
was
included
in
the
muds
in
the
form
of
green
cement.

Table
1:
Phase
1
drilling
fluid
preparation.

Mud
Product
A
B
C
D
E
G
H
C1618
IO,
bbl
0.567
0.567
0.567
0.63
0.63
0.441
0.708
Ester,
bbl
0.063
0.063
0.063
­
­
0.189
0.079
Organophilic
clay,
ppb
6.0
6.0
6.0
6.0
6.0
6.0
­

Lime,
ppb
­
3.0
­
3.0
3.0
3.0
­

1
°
emulsifier,
ppb
7.0
7.0
7.0
7.0
7.0
7.0
­

2
°
emulsifier,
ppb
2.0
2.0
2.0
2.0
2.0
2.0
­

Water,
bbl
0.15
0.15
0.15
0.15
0.15
0.15
0.19
CaCl2
(
95%),
ppb
19.2
19.3
19.2
19.2
19.3
19.3
24.2
Barite,
ppb
194
196
194
194
196
196
­

Fluid
loss
control,
ppb
1.0
1.0
1.0
1.0
1.0
1.0
­

OCMA
clay,
ppb
25.0
25.0
25.0
25.0
25.0
25.0
­

Green
cement
1,
ppb
­
­
15.0
­
­
­
­

1
F
or
the
cement
contaminated
muds,
19.3
g
of
Class
H
cement
was
added
to
10.3
g
of
water.
This
mixture
was
then
added
to
620
g
(
450
ml)
of
each
mud
and
stirred
for
10
minutes.
Page
7
of
30
July
29,
2004
Table
2:
Phase
2
drilling
fluid
preparation.

Mud
Products
B
I
E
J
G
K
C1618
IO,
bbl
0.567
0.567
0.63
0.63
0.441
0.441
Ester,
bbl
0.063
0.063
­
­
0.189
0.189
Organophilic
clay,
ppb
6.0
6.0
6.0
6.0
6.0
6.0
1
°
emulsifier,
ppb
7.0
7.0
7.0
7.0
7.0
7.0
2
°
emulsifier,
ppb
2.0
2.0
2.0
2.0
2.0
2.0
25%
CaCl2,
ppb
59.5
59.5
59.5
59.5
59.5
59.5
Barite,
g
198
198
198
198
198
198
Fluid
loss
control,
ppb
1.0
1.0
1.0
1.0
1.0
1.0
OCMA
clay,
ppb
25.0
25.0
25.0
25.0
25.0
25.0
Green
cement1,
ppb
15.0
15.0
15.0
1
F
or
the
cement
contaminated
muds,
19.3
g
of
Class
H
cement
was
added
to
10.3
g
of
water.
This
mixture
was
then
added
to
620
g
(
450
ml)
of
each
mud
and
stirred
for
10
minutes.

Sediment
toxicity
tests
The
96­
h
sediment
toxicity
tests
followed
the
method
outlined
in
the
EPA
Region
6
General
Permit
(
USEPA,
2001a).
A
laboratory
that
routinely
conducts
industry
compliance
tests
carried
out
the
sediment
toxicity
tests.
For
each
sediment
toxicity
test,
six
treatments
were
prepared
with
formulated
sediment,
five
with
the
test
material
and
a
negative
control.
The
sediment
toxicity
tests
were
conducted
in
2
phases:
(
1)
Phase
1
for
selected
temperature
and
time
conditions
for
muds
A
and
D,
and
(
2)
Phase
2
for
selected
temperature
and
time
conditions
for
muds
B,
I,
E,
J,
G,
and
K.
Phase
2
sediment
toxicity
tests
were
carried
out
with
a
statistical
design
(
see
below).
Phase
1
tests
were
performed
on
April
8­
12,
2004
and
Phase
2
tests
wee
performed
on
June
17­
22,
2004.
Table
3
shows
the
specific
time
and
temperature
conditions
that
were
tested
for
the
various
muds.
The
formulated
sediment
used
for
these
tests
was
prepared
at
the
test
laboratory.
The
dry
components
of
the
formulated
sediment
were
mixed
with
20
ppt
synthetic
seawater
the
day
before
the
treatments
were
prepared.
Test
concentrations
(
3,
10,
34,
115,
and
383
ml/
kg)
were
prepared
as
ml
of
drilling
fluid
per
kg
of
dry
formulated
sediment.
Each
treatment
was
mixed
for
ten
minutes
with
a
hand­
held
mixer.
With
each
batch
of
toxicity
test
there
was
a
reference
mud
(
11.5
ppg,
based
on
a
C1618
internal
olefin
fluid)
toxicity
test
that
was
also
conducted.
This
reference
test
is
an
industry
standard
and
was
used
to
determine
the
sediment
toxicity
compliance
ratio
(
STR;
reference
LC50/
test
material
LC50
 
1.0).

One
hundred
Leptocheirus
plumulosus
(
five
replicates
of
20)
were
each
exposed
to
formulated
sediment
spiked
with
the
test
material
at
five
different
ml/
kg
concentrations
and
negative
control.
Each
replicate
contained
approximately
150
ml
formulated
sediment
and
600
ml
20
ppt
overlying
water.
All
treatments
were
prepared
and
dispensed
18
to
24
hours
prior
to
initiating
the
test.
Treatments
were
kept
in
a
dedicated
environmental
chamber
with
14
hours
light
and
10
hours
dark
at
20
±
1
°
C
and
the
test
was
aerated.
At
24­
hour
intervals
temperature,
dissolved
oxygen,
pH,
and
salinity
were
measured
in
each
treatment.
After
96
hours,
all
five
replicates
of
each
treatment
were
terminated
and
the
number
of
surviving
organisms
recorded.
This
survival
information
was
used
to
determine
the
concentration
that
caused
50%
mortality
(
LC50)
and
the
95%
confidence
intervals.
Quality
assurance
and
quality
control
parameters
were
within
acceptable
ranges
(
USEPA,
2001).
Page
8
of
30
July
29,
2004
Analytical
measurements
2­
Ethyl
hexanol
(
2­
EH)
and
ester
in
the
heat­
aged
muds
were
measured
using
gas
chromatography/
mass
spectrometry
(
GC/
MS).
Olefin
and
paraffin
content
were
determined
by
difference.
A
sub­
sample
of
each
test
material
was
placed
in
a
40­
ml
vial
and
centrifuged
for
approximately
60
minutes
or
until
adequate
separation.
Then,
0.1
ml
of
supernatant
liquid
was
taken
from
the
top
by
pipette.
This
sample
was
decanted
into
a
2­
ml
vial
and
diluted
10
times
with
0.9
ml
of
methylene
chloride.
The
combined
sample
was
homogenized
by
gentle
shaking
and
inversion.

From
the
diluted
sample
portion,
a
1­
µ
l
sample
was
obtained
using
a
10­
µ
l
autosampler
syringe.
The
sample
was
then
directly
injected
into
an
Agilent
model
6890
gas
chromatograph
with
an
Agilent
model
5973N
mass
selective
detector.
The
column
used
was
an
HP
19091S­
433
(
HP­
5MS
5%
Phenyl
Methyl
Siloxane,
30
m
length,
250
µ
m
diameter,
0.25
µ
m
film
thickness).
A
temperature
program
was
used
(
initial
temperature
50
°
C,
ramp
at
10
°
C/
min
up
to
325
°
C).
Total
run
time
per
sample
was
29.5
min.
The
injector
temperature
was
260
°
C
and
the
detector
temperature
was
280
°
C.

Each
component
of
the
base
fluid
blend
(
C1618
internal
olefin,
ester)
and
2­
EH
was
analyzed
to
determine
its
chromatographic
retention
time.
The
2­
ethyl
hexanol
was
determined
to
be
present
in
the
3.8
to
4.2
minute
interval,
ester
was
determined
to
be
present
in
the
18
to
20
minute
range,
the
olefin
was
determined
to
be
in
the
10
to
13.8
minute
range,
and
another
insignificant
quantity
of
material
was
present
in
the
2.5
to
8
minute
range
(
this
material
was
reported
as
paraffin).
The
ester
material
was
observed
as
five
peaks
on
the
chromatograms.
These
peaks
corresponded
to
the
C8,
C10,
C12,
C14,
and
C16
fatty
acid­
based
esters.
Two
(
C8
and
C10
fatty
acid­
based
esters)
of
these
five
peaks
were
small
and
co­
eluted
with
the
olefin
material.
As
a
result,
the
three
largest
ester
peaks
(
C12,
C14,
and
C16
fatty
acid­
based
esters),
which
accounted
for
~
86%
of
the
ester
material
were
used
to
quantify
the
amount
of
ester
present
in
each
sample.

Results
were
calculated
based
on
the
area
under
the
peaks
in
relation
to
the
total
area
corrected
for
baseline.
All
compounds
were
assumed
to
have
equal
response
factors.
For
the
heat­
aged
samples,
quantitative
analysis
was
performed
only
on
the
ester
and
2­
ethyl
hexanol
components.
The
paraffin
and/
or
olefin
content
were
determined
by
subtracting
the
ester
and
2­
EH
values
from
100%.
Percent
ester
was
determined
as
follows:

100
13755
.
1
area
corrected
of
sum
peaks
ester
3
of
sum
ester
%
×
×
=

Detection
limits
for
2­
EH
and
ester
were
0.01
wt%.
Trace
paraffin
concentrations
were
<
2
wt%.
Spike
recovery
experiments
for
2­
EH
(
1
wt%)
added
to
Mud
D
(
IO
mud
with
lime)
showed
recoveries
of
72
%.
The
reported
results
were
not
corrected
for
recovery.

I
n
Phase
1,
measurements
were
made
for
muds
A,
B,
C,
D,
E,
G,
and
H.
In
Phase
2,
measurements
were
made
for
muds
E,
B,
G,
J,
K,
and
I.

Statistical
experimental
design
for
Phase
2
Data
developed
in
Phase
1
of
the
study
was
used
to
plan
Phase
2,
which
consisted
of
a
series
of
sediment
toxicity
tests
for
selected
time
and
temperature
heat
aging
conditions
for
6
different
muds.
Due
to
the
large
number
of
possible
combinations
of
parameters,
a
statistical
experimental
design
was
employed.
For
this
study,
there
were
four
factors
that
were
studied
to
determine
their
effects
on
the
sediment
toxicity
ratio
(
STR):
ester
content
of
ester/
olefin
muds
(
expressed
as
%
ester),
absence
or
presence
of
green
cement
(
alkalinity
effects;
yes/
no),
exposure
temperature
(
°
F),
and
exposure
time
(
hr).
For
each
factor,
the
design
was
selected
to
Page
9
of
30
July
29,
2004
determine
the
primary
or
main
effect
independently
of
all
the
other
factors.
Additionally,
the
design
was
set
up
to
allow
for
estimation
of
two­
way
interaction
effects
(
e.
g.
when
the
impact
of
one
factor
depends
on
the
setting
of
another
factor)
as
well
as
potential
non­
linear
effects
for
all
factors
except
green
cement
content.

Experimental
logistical
constraints
necessitated
performing
the
experiments
in
two
"
blocks".
The
assignment
of
each
experiment
to
Block
1
or
Block
2
was
done
in
a
manner
that
did
not
compromise
the
ability
to
estimate
the
effect
of
the
factors
under
study.

I
n
order
to
estimate
all
these
effects,
it
was
necessary
to
study
percent
ester
(
0,
10,
30),
temperature
(
275,
300,
350
°
F)
and
time
(
16,
40,
160
hr)
at
three
levels
and
green
cement
(
0,
15
ppb)
at
two
levels.
Performing
all
possible
combinations
would
have
required
54
experiments
in
total.
To
minimize
the
number
of
experiments,
a
statistical
experimental
design
was
used
to
allow
for
the
estimation
of
all
the
desired
effects
in
less
than
half
the
number
of
runs.
The
trade
off
in
doing
fewer
runs
was
the
inability
to
estimate
higher
ordered
interactions
(
three­
way
and
above).
However,
experience
has
shown
that
such
effects
are
rarely
significant
either
statistically
or
practically.

When
doing
fewer
than
the
full
number
of
experiments,
the
primary
goals
in
selecting
the
subset
were
allocation
and
balance.
Allocation
means
that
the
total
number
of
runs
is
relatively
evenly
spread
among
the
levels
being
studied
for
each
factor
(
e.
g.,
similar
number
of
runs
with
high
temperature
as
low
temperature).
Balance
means
that
for
a
given
level
of
say
temperature,
there
are
a
similar
if
not
equal
number
of
runs
at
high
time
and
low
time,
high
percent
ester
and
low,
etc.

Table
3
below
demonstrates
the
final
design
chosen.
This
table
also
visually
demonstrates
the
allocation
and
balance
referred
above.
Page
10
of
30
July
29,
2004
Table
3:
Sediment
toxicity
testing
experimental
design
matrix.
In
Phase
1,
two
muds
(
A
and
D)
were
tested
on
April
8­
12,
2004.
These
Phase
1
tests
are
designated
with
a

.
In
Phase
2,
which
employed
a
statistical
design,
six
muds
were
tested
and
are
designated
as
Muds
E,
B,
G,
J,
K,
and
I.
Two
separate
groups
(
blocks)
of
experiments
were
conducted
(
Block
#
1,
June
17­
21,
2004;
Block
#
2,
June
18­
22,
2004).
An
IO1618
reference
mud
test
was
conducted
with
each
block.
Each
time
and
temperature
condition
tested
is
denoted
by

for
Block
#
1
and

for
Block
#
2
(
2

or
2

indicates
a
duplicate
experiment
was
conducted).
Alkalinity
sources
are
either
lime
(
L)
or
green
cement
(
GC).

Sediment
Toxicity
Test
Matrix
IO
90/
10
IO/
ester
70/
30
IO/
ester
Alkalinity
Source
Brine
Time
(
hr)
T
(
F)
16
40
160
16
40
88
160
16
40
160
E
B
G
None
no
275




None
no
300



None
no
325
None
no
350




J
I
K
GC
no
275



GC
no
300



GC
no
325
GC
no
350

2

2

,
2



D
A
L
no
275


L
no
300


L
no
325
L
no
350




C
L,
GC
no
275
L,
GC
no
300
L,
GC
no
325
L,
GC
no
350
H
None
yes
275
None
yes
300
None
yes
325
None
yes
350
Page
11
of
30
July
29,
2004
RESULTS
AND
DISCUSSION
Phase
1
 
screening
experiments
During
Phase
1
of
the
study,
seven
muds
were
prepared
and
subjected
to
elevated
temperatures
for
different
periods
of
time
(
0
 
160
hr).
Prior
work
in
industry
laboratories
suggested
that
the
sediment
toxicity
of
ester­
containing
muds
could
increase
under
certain
conditions
of
time
and
temperature.
In
addition,
the
sediment
toxicity
has
been
observed
to
increase
concomitantly
with
the
level
of
2­
ethyl
hexanol
(
2­
EH)
in
the
mud.
2­
Ethyl
hexanol
can
be
produced
via
the
hydrolysis
of
ester
products
that
are
based
on
2­
EH.
Thus,
for
the
muds
prepared
for
Phase
1,
the
2­
EH
concentration
was
measured
at
each
time
and
temperature
condition.
In
addition,
a
few
selected
muds
were
used
to
carry
out
sediment
toxicity
tests
with
L.
plumulosus
(
see
Table
3).

Analytical
measurements
of
Phase
1
muds
Each
of
the
prepared
muds
was
analyzed
for
2­
EH,
ester
content,
olefin
content,
and
paraffin
content.
The
data
for
Phase
1
are
shown
in
the
Appendix.
In
addition,
the
2­
EH
values
are
shown
as
a
function
of
temperature
and
time
in
Table
4.

I
n
general,
longer
times
and
higher
temperatures
resulted
in
the
formation
of
higher
levels
of
2­
EH.
The
removal
of
all
mud
constituents
except
brine
(
Mud
H)
reduced
but
did
not
eliminate
the
amount
of
2­
EH
formed.
For
the
longest
time
and
temperature
conditions
studied,
the
2­
EH
concentration
was
2
 
2.6
wt%.
For
these
muds,
a
corresponding
decrease
in
ester
content
was
observed
(
see
Appendix).
No
2­
EH
was
detected
in
muds
that
did
not
include
ester
as
part
of
the
formulation.

Sediment
toxicity
of
Phase
1
muds
Eight
of
the
prepared
mud
samples
were
tested
for
sediment
toxicity
to
L.
plumulosus.
Two
mud
types
were
assessed:
(
1)
an
internal
olefin
(
IO)
mud
prepared
with
lime
(
Mud
D)
and
(
2)
a
90/
10
olefin/
ester
blend
mud
prepared
with
lime
(
Mud
A).
The
sediment
toxicity
results
are
shown
in
Table
5.

Each
result
is
reported
as
an
LC50
(
ml/
kg)
along
with
its
corresponding
95%
confidence
interval.
A
corresponding
C1618
IO
reference
mud
(
11.5
ppg)
was
also
tested
for
sediment
toxicity.
This
reference
mud
had
an
LC50
of
79
ml/
kg.
The
reference
LC50
was
used
to
calculate
a
sediment
toxicity
ratio
(
STR,
LC50
reference
mud/
LC50
test
mud)
according
to
USEPA
procedures
(
USEPA,
2001).
The
calculated
STRs
are
shown
in
Table
6.
In
order
for
a
mud
to
be
in
compliance,
the
STR
must
be
 
1.0.

The
ester
containing
mud
was
more
toxic
(
lower
LC50,
higher
STR)
than
the
IO
mud.
One
of
the
ester
mud
samples
(
time
of
40
hr
and
temperature
of
350
°
F)
had
an
STR
>
1.0
(
1.6)
which
would
have
resulted
in
a
failure
for
a
field
compliance
test.

These
Phase
1
results
support
the
hypothesis
that
toxicity
was
negatively
affected
by
the
presence
of
ester
in
the
ester/
olefin
blends
and
downhole
conditions
(
time,
temperature).
These
results
were
used
as
a
basis
to
design
the
experiments
for
Phase
2.
Page
12
of
30
July
29,
2004
Table
4:
Phase
1
concentrations
of
2­
ethyl
hexanol
(
2­
EH)
in
drilling
muds
as
a
function
of
time,
temperature,
and
presence
or
absence
of
an
alkalinity
source
(
L
=
lime,
GC
=
green
cement).
Areas
highlighted
indicate
concentrations
of
2­
EH
>
0.5
wt%.

2­
ethyl
hexanol
(
wt
%)

IO
90/
10
IO/
ester
70/
30
IO/
ester
Alkalinity
Source
Brine
Time
(
hr)

T
(
F)
16
40
160
16
40
88
160
16
40
160
E
B
G
None
no
275
0.05
0.09
0.3
0.14
0.4
1.37
None
no
300
ND
0.05
0.41
0.1
0.35
1.86
None
no
325
0.06
0.16
0.89
0.38
0.92
2.55
None
no
350
0.15
0.42
0.9
0.78
2.18
2.52
J
I
K
GC
no
275
GC
no
300
GC
no
325
GC
no
350
D
A
L
no
275
ND
ND
ND
0.05
0.1
0.85
0.55
L
no
300
ND
ND
ND
0.1
0.15
0.54
0.6
L
no
325
ND
ND
ND
0.2
0.45
1.19
1.74
L
no
350
ND
ND
ND
0.3
0.7
1.89
2.06
C
L,
GC
no
275
0.08
0.2
0.39
1.29
L,
GC
no
300
0.06
0.3
0.93
1.87
L,
GC
no
325
0.27
0.84
2.36
L,
GC
no
350
0.62
1.59
2.61
H
None
yes
275
ND
ND
0.06
None
yes
300
ND
0.02
0.10
None
yes
325
ND
ND
0.44
None
yes
350
ND
0.06
0.77
Page
13
of
30
July
29,
2004
Table
5:
Phase
1
sediment
toxicity
(
LC50)
to
L.
plumulosus.
The
reference
mud
(
11.5
ppg)
had
an
LC50
of
79
ml/
kg
(
68
 
94).
Areas
highlighted
indicate
2­
EH
concentrations
>
0.5
wt%.
Alkalinity
sources
were
either
L
=
lime
or
GC
=
green
cement.

LC50
(
ml/
kg)

IO
90/
10
IO/
ester
70/
30
IO/
ester
Alkalinity
Source
Time
(
hr)

T
(
F)
16
40
160
16
40
88
160
16
40
160
E
B
G
None
275
None
300
None
325
None
350
J
I
K
GC
275
GC
300
GC
325
GC
350
D
A
L
275
>
240
>
240
L
300
151
(
121
 
189)
109
(
9
 
133)

L
325
L
350
>
240
183
(
135
 
248)
226
(
206
 
247)
50
(
43
 
58)

C
L,
GC
275
L,
GC
300
L,
GC
325
L,
GC
350
Page
14
of
30
July
29,
2004
Table
6:
Phase
1
sediment
toxicity
ratios
(
STR)
based
on
data
in
Table
5.
Areas
highlighted
indicate
2­
EH
concentrations
>
0.5
wt%.

Alkalinity
sources
were
either
L
=
lime
or
GC
=
green
cement.
Sediment
Toxicity
Ratio
(
STR)

IO
90/
10
IO/
ester
70/
30
IO/
ester
Alkalinity
Source
Time
(
hr)

T
(
F)
16
40
160
16
40
88
160
16
40
160
E
B
G
None
275
None
300
None
325
None
350
J
I
K
GC
275
GC
300
GC
325
GC
350
D
A
L
275
<
0.3
<
0.3
L
300
0.5
0.7
L
325
L
350
<
0.3
0.4
0.3
1.6
C
L,
GC
275
L,
GC
300
L,
GC
325
L,
GC
350
Page
15
of
30
July
29,
2004
Phase
2
 
detailed
experiments
During
Phase
2
of
the
study,
six
new
muds
were
prepared
and
subjected
to
elevated
temperatures
for
different
periods
of
time
(
0
 
160
hr).
Five
of
these
muds
were
equivalent
to
muds
used
in
Phase
1with
green
cement
used
as
an
alkalinity
source
in
place
of
lime.
The
additional
mud
was
a
70/
30
internal
olefin/
ester
blend
contaminated
with
green
cement.
For
the
muds
prepared
for
Phase
2,
the
2­
EH
concentration
was
measured
at
each
time
and
temperature
condition.
In
addition,
all
of
the
muds
were
used
to
carry
out
sediment
toxicity
tests
with
L.
plumulosus.

Analytical
measurements
of
Phase
2
muds
Each
of
the
prepared
muds
was
analyzed
for
2­
EH,
ester
content,
olefin
content,
and
paraffin
content.
The
data
for
Phase
2
are
shown
in
the
Appendix.
In
addition,
the
2­
EH
values
are
shown
as
a
function
of
temperature
and
time
in
Table
7.

I
n
general,
longer
times
and
higher
temperatures
resulted
in
the
formation
of
higher
levels
of
2­
EH.
For
the
longest
time
and
temperature
conditions
studied,
the
2­
EH
concentration
was
5
 
10
wt%.
For
these
muds,
a
corresponding
decrease
in
ester
content
was
observed
(
see
Appendix).
No
2­
EH
was
detected
in
muds
that
did
not
include
ester
as
part
of
the
formulation.

Sediment
toxicity
of
Phase
2
muds
A
selection
of
the
prepared
muds
(
22
muds,
plus
3
replicates
were
tested
for
sediment
toxicity
to
L.
plumulosus.
The
sediment
toxicity
tests
were
carried
out
in
two
blocks
(
see
Experimental)
due
to
logistical
constraints
for
the
experiments.
The
two
blocks
of
experiments
were
performed
on
consecutive
days.
The
sediment
toxicity
results
are
shown
in
Table
8.

Each
result
is
reported
as
an
LC50
(
ml/
kg)
along
with
its
corresponding
95%
confidence
interval.
A
corresponding
C1618
IO
reference
mud
(
11.5
ppg)
was
also
tested
for
sediment
toxicity
with
each
block.
This
reference
mud
had
an
LC50
of
100
and
66
ml/
kg,
respectively
for
Blocks
#
1
and
#
2.
The
reference
LC50
was
used
to
calculate
a
sediment
toxicity
ratio
(
STR,
LC50
reference
mud/
LC50
test
mud)
according
to
USEPA
procedures
(
USEPA,
2001).
The
calculated
STRs
are
shown
in
Table
9.

The
ester
containing
muds
were
more
toxic
(
lower
LC50,
higher
STR)
than
the
IO
muds.
In
general,
higher
times
and
temperatures
resulted
in
higher
STRs.
Five
of
the
ester/
olefin
mud
samples
had
a
STRs
>
1.0
(
1.4
 
2.9)
which
would
have
resulted
in
failures
for
field
compliance
tests.

These
Phase
2
results
support
the
hypothesis
that
toxicity
was
negatively
affected
by
esterester
olefin
blends,
downhole
conditions
(
time,
temperature)
and
formulation
(
alkalinity
via
the
presence/
absence
of
green
cement).
Page
16
of
30
July
29,
2004
Table
7:
Phase
2
concentrations
of
2­
ethyl
hexanol
(
2­
EH)
in
drilling
muds
as
a
function
of
time,
temperature,
and
presence
or
absence
of
an
alkalinity
source.
Areas
highlighted
indicate
2­
EH
concentrations
>
0.5
wt%.
Alkalinity
sources
were
either
L
=
lime
or
GC
=
green
cement.
2­
ethyl
hexanol
(%)

IO
90/
10
IO/
ester
70/
30
IO/
ester
Alkalinity
Source
Time
(
hr)

T
(
F)
16
40
160
16
40
88
160
16
40
160
E
B
G
None
275
ND
0.15
0.15
1.22
None
300
ND
0.22
1.06
None
350
ND
1.27
1.5
0.58
J
I
K
GC
275
ND
0.64
0.4
GC
300
ND
1.53
2.22
GC
350
ND
1.31
2.42,
2.40,
2.46,
2.72
4.99
9.75
D
A
L
275
L
300
L
350
C
L,
GC
275
L,
GC
300
L,
GC
350
Page
17
of
30
July
29,
2004
Table
8:
Phase
2
­
Sediment
toxicity
of
heat­
aged
muds
to
L.
plumulosus.
Data
reported
as
LC50
in
ml/
kg
(
95%
confidence
interval).

Areas
highlighted
indicate
2­
EH
concentrations
>
0.5
wt%.
Alkalinity
sources
were
either
L
=
lime
or
GC
=
green
cement.

LC50
(
ml/
kg)

IO
90/
10
IO/
ester
70/
30
IO/
ester
Alkalinity
Source
Time
(
hr)

T
(
F)
16
40
160
16
40
88
160
16
40
160
E
B
G
None
275
189
(
152
 
235)
82
(
68
 
99)
>
383
161
(
145
 
178)

None
300
>
383
121
(
98
 
155)
145
(
116
 
179)

None
350
256
(
209
 
315)
68
(
56
 
81)
84
(
72
 
98)
>
383
J
I
K
GC
275
122
(
102
 
147)
108
(
94
 
125)
171
(
130
 
224)

GC
300
153
(
125
 
187)
74
(
63
 
86)
39
(
34
 
44)

GC
350
>
383
73
(
62
 
83)

129
(
109
 
153)
116
(
98
 
138)

74
(
62
 
88)

63
(
53
 
75)

46
(
39
 
55)
31
(
27
 
36)
34
(
29
 
40)

D
A
L
275
L
300
L
350
F
C
L,
GC
275
L,
GC
300
L,
GC
350
Page
18
of
30
July
29,
2004
Table
9:
Phase
1
­
Sediment
toxicity
ratios
(
STR)
of
heat­
aged
muds.
Comparisons
were
made
to
a
11.5
ppg
reference
mud
prepared
with
a
1618
internal
olefin
base
fluid.
The
reference
mud
toxicities
were
100
ml/
kg
and
66
ml/
kg
for
the
sediment
toxicity
experiments
conducted
on
6/
17­
6/
21/
04
and
6/
18­
6/
22/
04,
respectively.
Areas
highlighted
indicate
2­
EH
concentrations
>
0.5
wt%.
Alkalinity
sources
were
either
L
=
lime
or
GC
=
green
cement.
Sediment
Toxicity
Ratio
(
STR)

IO
90/
10
IO/
ester
70/
30
IO/
ester
Alkalinity
Source
Time
(
hr)

T
(
F)
16
40
160
16
40
88
160
16
40
160
E
B
G
None
275
0.5
0.8
<
0.3
0.6
None
300
<
0.2
0.5
0.5
None
350
0.4
1.0
0.8
<
0.3
J
I
K
GC
275
0.8
0.6
0.6
GC
300
0.4
0.9
1.7
GC
350
<
0.3
0.5
0.9
0.9
1.4
1.0
1.4
2.1
2.9
D
A
L
275
L
300
L
350
C
L,
GC
275
L,
GC
300
L,
GC
350
Page
19
of
30
July
29,
2004
Sediment
toxicity
ratio
(
STR)
statistical
model
The
experimental
results
of
Phase
2
were
used
to
develop
a
statistical
model
to
analyze
the
toxicity
and
2­
EH
data
sets.
A
series
of
statistical
models
were
constructed.
With
the
available
data
it
was
possible
to
build
models
for
both
test
LC50
and
the
resulting
STR.
Models
for
both
could
be
used
to
generate
similar
predicted
STRs
for
a
given
combination
of
factors/
inputs.
However,
since
STR
is
the
most
relevant
parameter
from
the
perspective
of
passing
or
failing
a
compliance
test,
one
specific
STR
model
was
selected
for
data
analysis
and
is
summarized
below.
For
the
specific
model
selected,
the
STR
data
were
not
log
transformed.
This
model
was
developed
via
standard
analysis
of
variance
(
ANOVA)
and
regression
methods
(
Iman
and
Conover,
1983).
Table
10
provides
a
summary
of
the
significant
effects
for
the
final
STR
model.

Table
10:
Parameter
estimates
for
STR
model.

Term
Estimate
Std
Error
t
Ratio
Prob
>
|
t|

Intercept
­
0.904246
0.062217
15.85
<.
0001
Ester
wt%,
E
­
0.014261
0.078952
4.40
0.0003
Green
cement,
G
­
0.032288
0.062221
6.75
<.
0001
Time
(
hr),
t
­
0.003024
0.072279
4.56
0.0002
Temperature
(
°
F),
T
0.0053333
0.065185
3.07
0.0066
Ester
wt%*
Green
Cement,
E
x
G
0.0024638
0.074656
3.71
0.0016
Ester
wt%*
Time,
E
x
t
0.000215
0.086763
2.68
0.0154
Green
Cement*
Time,
G
x
t
0.0005831
0.072197
4.36
0.0004
Note
that
all
p­
values
(
Prob>|
t|)
are
less
than
the
0.05
threshold,
corresponding
to
the
95%
confidence
level.
Equation
1
gives
the
resulting
prediction
formula
for
STR:

Gt
Et
EG
T
t
G
E
STR
000583
.
0
0002
.
0
000246
.
0
005
.
0
003
.
0
0323
.
0
014
.
0
904
.
0
+
+
+
+
 
 
 
 
=
(
1)

where
E
=
percent
ester
in
the
base
fluid
used
for
the
drilling
mud
G
=
alkalinity
present
in
green
cement
(
ppb)
used
in
the
drilling
mud
t
=
time
(
hr)
mud
was
heated
at
temperature
T
T
=
temperature
(
°
F)
at
which
the
mud
was
held
for
time
t
This
model
accounted
for
approximately
80%
of
the
variability
in
STR.
Figures
1
and
2
represent
the
model
in
a
more
visual
way.
Figure
1
is
a
plot
of
the
predicted
versus
the
actual
STR
for
each
experiment.
Page
20
of
30
July
29,
2004
Figure
1:
Comparison
of
model
predicted
versus
actual
for
STR
model.
Dashed
lines
represent
expected
variability
about
the
fitted
model.

0.0
0.5
1.0
1.5
2.0
2.5
3.0
S
T
R
A
c
t
u
a
l
.0
.5
1.0
1.5
2.0
2.5
3.0
3.5
STR
Predicted
P<.
0001
RSq=
0.86
RMSE=
0.2755
Figure
2:
Effect
profiles
for
STR
model.
Vertical
dashed
red
lines
correspond
to
the
current
setting
for
each
variable,
yielding
the
current
predicted
STR
that
is
noted
with
the
horizontal
dashed
line.

S
T
R
2.941
0.1551
0.952235
Ester
%(
0,30)
0
3
0
15
GC(
0,15)
0
1
5
9
Time(
16,160)
1
6
1
6
0
61
Temp(
275,350)
2
7
5
3
5
0
318
The
overall
effect
of
each
factor
is
generally
to
increase
the
STR
by
decreasing
the
LC50
of
the
test
mud.
Page
21
of
30
July
29,
2004
There
were
three
significant
interaction
effects:
(
1)
Ester
x
Lime,
(
2)
Ester
x
time,
and
(
3)
Lime
x
time.
The
significance
of
these
interaction
effects
indicates
that
these
parameters
work
together
in
addition
to
their
individual
effect
to
determine
the
STR.
For
example,
if
the
Ester
x
Lime
interaction
is
considered,
the
effect
of
percent
ester
on
STR
depends
on
whether
lime
is
present
as
well.

There
were
four
experiments
in
which
the
test
LC50
was
greater
than
the
maximum
concentration
studied
(>
383
ml/
kg).
Several
approaches
for
handling
these
"
detection
limited"
experiments
via
both
substitution
and
exclusion
were
considered.
The
resulting
"
best
model"
in
each
case
gave
effects
and
predictions
that
did
not
differ
significantly
either
from
a
practical
or
statistical
perspective.
As
a
result,
a
substitution
method
was
used
for
the
model
summarized
here.
This
method
used
the
relationship
between
survival
percent
and
LC50
to
estimate
the
LC50s
that
were
>
383
ml/
kg.

Application
of
the
sediment
toxicity
ratio
(
STR)
statistical
model
The
section
above
gave
a
brief
summary
of
the
final
model
in
terms
of
quality
and
estimated
effects.
However,
the
model
can
be
used
to
address
research
questions
of
interest.
The
experimental
results
shown
above
support
the
hypothesis
that
toxicity
was
negatively
affected
by
ester/
olefin
blends,
downhole
conditions
(
time,
temperature),
and
formulation
(
presence/
absence
of
green
cement).
However,
due
to
large
variability
in
the
sediment
toxicity
results,
it
is
difficult
to
quantify
these
effects.
As
a
result,
the
model
was
used
to
develop
quantitative
boundaries
for
some
of
these
negative
effects.

An
example
of
the
use
of
the
model
is
shown
below
in
Figure
3.

The
95%
prediction
limits
are
quite
broad
(
±
0.6
STR
units)
due
to
the
large
observed
variability
in
the
sediment
toxicity
results
and
corresponding
ratios.
Nevertheless,
the
model
suggests
that
on
average
at
temperatures
>
285
 
290
°
F,
the
STR
for
this
scenario
(
10%
ester,
presence
of
15
ppb
green
cement,
and
time
=
80
hr)
will
be
>
1.0.

Figure
4
shows
a
contour
plot
of
STR
=
1
as
a
function
of
temperature,
%
ester,
time,
and
absence
or
presence
of
15
ppb
green
cement.
Each
line
represents
a
constant
STR
of
1.0
at
a
specific
point
in
time.
Areas
above
each
line
at
a
fixed
time
indicate
combinations
of
temperature
and
%
ester
where
on
average
the
STR
will
be
>
1.0.
Areas
below
each
line
at
a
fixed
time
indicate
combinations
of
temperature
and
%
ester
where
on
average
the
STR
will
be
<
1.0.
Up
to
160
hr,
for
range
of
muds
tested
(
ester
content
up
to
30%),
in
the
absence
of
green
cement,
the
average
STR
does
not
go
above
1
for
temperatures
>
325
 
350
°
F.
However,
in
the
presence
of
green
cement
(
15
ppb),
the
average
STR
begins
to
go
above
1
in
the
ester
range
of
10
 
30%
at
temperatures
>
300
°
F.

The
statistical
model
(
Equation
1)
was
used
to
estimate
temperature
thresholds
that
would
result
in
statistically
significant
effects
on
the
STR.
These
results
are
shown
in
Table
11.
At
temperatures
above
295
°
F
and
less
than
275
°
F,
the
model
predicts
that
the
STR
will
be
>
1.0
with
95%
confidence,
when
the
ester
content
and
times
are
10%
and
160
hr
and
30%
and
160
hr,
respectively
(
in
the
presence
of
green
cement).
Thus,
based
on
the
predictive
statistical
model,
at
a
temperature
of
300
°
F,
ester­
containing
muds
with
>
10%
ester
in
the
presence
of
15
ppb
green
cement
will
likely
fail
the
mud
sediment
toxicity
test
as
they
will
have
STR
>
1.0
with
a
certainty
of
95%
confidence.
Page
22
of
30
July
29,
2004
Figure
3:
Predicted
STR
(
with
95%
prediction
limits)
as
a
function
of
temperature
for
the
scenario
10%
ester
content,
presence
of
15
ppb
green
cement,
and
time
of
80
hr.

0.0
0.5
1.0
1.5
2.0
260
270
280
290
300
310
320
330
340
350
360
Temperature
(
F)
Sediment
Toxicity
Ratio
Model
Prediction
95%
Prediction
Limit
Page
23
of
30
July
29,
2004
Figure
4:
Contour
plot
of
STR
=
1
as
a
function
of
temperature,
%
ester,
time,
and
absence
or
presence
of
15
ppb
green
cement.
Each
line
represents
a
constant
STR
of
1.0
at
a
specific
point
in
time.
Areas
above
each
line
at
a
fixed
time
indicate
combinations
of
temperature
and
%
ester
where
on
average
the
STR
will
be
>
1.0.
Areas
below
each
line
at
a
fixed
time
indicate
combinations
of
temperature
and
%
ester
where
on
average
the
STR
will
be
<
1.0.

t
=
40
h
t
=
80
h
t
=
120
h
t
=
160
h
t
=
40
h
t
=
80
h
t
=
120
h
t
=
160
h
­
100
0
100
200
300
400
500
0
5
10
15
20
25
30
35
%
Ester
Temperature
(
F)
for
STR
=
1
­
Green
Cement
+
Green
Cement
STR
>
1
STR
<
1
The
statistical
model
(
Equation
1)
was
also
used
to
estimate
ester
content
thresholds
that
would
result
in
statistically
significant
effects
on
the
STR.
These
results
are
shown
in
Table
12.
At
ester
contents
above
>
10%,
the
model
predicts
that
the
STR
will
be
>
1.0
with
95%
confidence,
when
the
time
approaches
160
hr
(
in
the
presence
of
green
cement).
Thus,
based
on
the
predictive
statistical
model,
ester­
containing
muds
with
>
10%
ester
in
the
presence
of
green
cement
will
likely
fail
the
mud
sediment
toxicity
test
as
they
will
have
STR
>
1.0
with
a
certainty
of
95%
confidence.

Finally,
the
STR
statistical
model
developed
with
the
data
from
Phase
2
was
validated
with
the
data
from
Phase
1.
Thus,
the
independent
data
from
Phase
1
were
used
to
"
test"
the
STR
model.
This
analysis
assumed
that
the
alkalinity
from
the
lime
and
the
alkalinity
of
the
green
cement
added
to
the
muds
had
equivalent
effects.
In
the
comparison,
one
unit
of
lime
(
3
ppb)
was
assumed
to
equal
one
unit
of
green
cement
(
15
ppb).
These
results
are
shown
in
Table
13.
For
6
of
the
7
sediment
toxicity
tests
in
Phase
1,
the
STR
model
was
able
to
predict
the
measured
STR
within
the
95%
prediction
limits.
Page
24
of
30
July
29,
2004
Table
11:
Estimated
temperature
thresholds
(
°
F)
for
having
95%
confidence
that
a
specific
mud
(
time,
ester
content,
alkalinity
via
presence
or
absence
of
green
cement)
will
pass
(
STR
<
1)
or
fail
(
STR
>
1).

Threshold
Temp
(
F)

STR
<
1
STR
>
1
Ester
(%)
Green
Cement
(
ppb)
Time
(
hr)
95%
Confidence
Pass
95%
Confidence
Fail
0
0
16
<
275
>
350
0
0
160
<
312
>
350
5
0
16
<
275
>
350
5
0
160
<
300
>
350
10
0
16
<
275
>
350
10
0
160
<
285
>
350
30
0
16
<
303
>
350
30
0
160
<
275
>
350
0
15
16
<
310
>
350
0
15
160
<
275
>
350
5
15
16
<
285
>
350
5
15
160
<
275
>
350
10
15
16
<
275
>
350
10
15
160
<
275
>
295
30
15
16
<
275
>
350
30
15
160
<
275
<
275
Page
25
of
30
July
29,
2004
Table
12:
Estimated
ester
percentage
thresholds
(%)
for
having
95%
confidence
that
a
specific
mud
(
time,
alkalinity
via
presence
or
absence
of
green
cement)
will
pass
(
STR
<
1)
or
fail
(
STR
>
1)
at
a
temperature
of
300
°
F.

Ester
Threshold
(%)

STR
<
1
STR
>
1
Temperature
(
°
F)
Green
Cement
(
ppb)
Time
(
hr)
95%
Confidence
Pass
95%
Confidence
Fail
300
0
16
<
30
­­

300
0
160
<
5
­­

300
15
16
<
3
>>
30
300
15
160
­­
>
10
Table
13:
Comparison
of
predicted
and
measured
STRs
from
the
preliminary
(
Phase
1)
data
set.

Ester
(%)
Lime
(
ppb)
Time
(
hr)
Temperature
(
°
F)
Lower
95%
Pred
Limit
Predicted
STR
Upper
95%
Pred
Limit
Measured
STR
0
3
1
16
275
­
0.506
0.170
0.845
<
0.3
0
3
16
350
­
0.079
0.570
1.218
<
0.3
0
3
40
350
0.073
0.707
1.341
0.4
0
3
160
300
0.397
1.127
1.857
0.5
10
3
16
275
­
0.207
0.431
1.069
<
0.3
10
3
160
300
1.023
1.698
2.373
0.7
10
3
40
350
0.413
1.020
1.627
1.6
1
For
the
model
analysis,
it
was
assumed
that
3
ppb
lime
was
equivalent
to
15
ppb
green
cement.

2­
Ethyl
hexanol
correlates
with
increased
STR
Another
hypothesis
for
the
study
was
that
the
presence
of
2­
EH
as
an
indicator
of
ester
hydrolysis
in
a
drilling
mud
would
lead
to
increased
sediment
toxicity
(
higher
STR).
This
hypothesis
was
tested
in
both
Phases
1
and
2.
The
raw
data
for
the
2­
EH
measurements
are
shown
in
the
Appendix.
Figure
5
is
a
plot
of
STR
as
a
function
of
2­
EH
concentration.
In
general,
the
STR
increases
as
the
concentration
of
2­
EH
increases.
The
data
from
Phase
2
were
used
to
develop
a
linear
correlation
for
STR
as
a
function
of
2­
EH
(
R2
=
0.85).
This
correlation
is
shown
as
the
predicted
line
in
Figure
5
along
with
the
95%
prediction
limits
of
the
correlation.
The
data
from
Phase
1
are
consistent
with
the
Phase
2
results.
On
average,
at
2­
EH
concentrations
>
2
wt%,
the
correlation
predicts
that
STR
will
be
>
1.0.
With
95%
confidence,
the
correlation
predicts
that
STR
will
be
>
1.0
when
the
2­
EH
concentration
exceeds
4
wt%.
These
results
are
consistent
with
the
STR
statistical
model
as
they
also
indicate
that
at
ester
levels
>
10%,
temperatures
>
~
300
Page
26
of
30
July
29,
2004
°
F,
and
longer
times
(
40
 
160
hr),
it
is
likely
that
STR
will
be
>
1.0
and
the
muds
will
fail
the
sediment
toxicity
test.

Figure
5:
STR
as
a
function
of
2­
EH.

0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0
2
4
6
8
10
2­
Ethyl
Hexanol
(%)
STR
Prediction
95%
Prediction
Limit
95%
Confidence
Limit
Phase
2
Phase
1
CONCLUSIONS
I
n
Phase
1
of
this
study
eight
bioassay
tests
were
conducted
on
laboratory
prepared
samples
tested
at
temperature
ranges
from
275
to
350
°
F.
Four
samples
used
an
internal
olefin
(
IO1618)
base
fluid
while
four
samples
used
ester/
olefin
blends.
There
were
no
IO1618
drilling
fluid
sample
failures
and
one
ester/
IO
sample
failure
in
Phase
1.

I
n
Phase
2
of
the
study
26
bioassay
tests
were
conducted
on
laboratory
prepared
samples
at
temperature
ranges
from
275
to
350
°
F.
Six
samples
used
IO
1618
base
fluid
while
20
samples
used
ester/
olefin
blends.
There
were
no
IO
1618
drilling
fluid
sample
failures
and
there
were
5
ester/
IO
drilling
fluid
sample
failures.

The
toxicity
limitation
failures
of
ester/
olefin
blends
at
or
below
350
°
F
temperature
exposures
indicate
the
potential
for
compliance
test
failures
under
field
conditions.

I
n
order
to
better
understand
the
relationships
of
time,
temperature
and
other
factors
on
the
toxicity
performance,
a
statistical
design
and
analysis
of
the
test
results
was
carried
out
to
assess
the
probability
of
passing
or
failing
the
toxicity
test.
The
sediment
toxicity
tests
and
analytical
measurements
conducted
on
the
muds
used
for
Phases
1
and
2
of
this
study
support
the
hypothesis
that
toxicity
was
negatively
affected
by
ester/
olefin
blends
and
some
high
temperature
Page
27
of
30
July
29,
2004
well
downhole
conditions.
The
data
suggest
that
under
some
conditions,
for
the
types
of
muds
used
in
this
study,
sediment
toxicity
will
increase
when
the
mud
is
exposed
to
elevated
temperatures
for
extended
periods
of
time.
Ester/
olefin
muds
with
>
10%
ester
are
likely
to
break
down
and
produce
breakdown
products
that
will
result
in
sediment
toxicity
and
failure
of
the
STR
<
1.0
criteria
required
for
compliance.

2­
Ethyl
hexanol
was
correlated
with
increased
STR
and
can
be
used
as
a
surrogate
to
screen
for
potential
sediment
toxicity
for
ester/
olefin
muds
that
contain
traditional
esters.
On
average,
when
2­
EH
concentrations
exceed
2.0
wt%,
it
is
likely
that
muds
will
fail
sediment
toxicity
compliance
tests.
At
2­
EH
concentration
of
>
4
wt%,
it
is
95%
certain
that
the
muds
will
fail.

The
data
from
this
study
support
a
temperature
threshold
of
~
300
°
F
for
which
operators
should
be
cautious
with
the
use
of
drilling
fluids
containing
the
types
of
esters
(
traditional
esters)
tested
in
this
study.
Above
300
°
F
these
materials
may
break
down
and
result
in
increased
sediment
toxicity
for
the
drilling
muds.
Other
types
of
esters
were
not
examined
in
this
study.

Consequently,
while
some
individual
ester/
olefin
drilling
muds
that
have
been
exposed
to
certain
time
and
temperature
conditions
can
be
identified
as
passing
the
toxicity
limitations,
the
statistical
analysis
of
sediment
toxicities
of
heat­
aged
ester/
olefin
drilling
muds
shows
that
under
a
broader
evaluation
of
temperature,
time
and
routine
wellbore
contamination
that
the
functional
temperature
limitation
for
ester/
olefin
blends
is
lower
than
expected
by
USEPA
in
development
of
the
effluent
limitation
guidelines.

REFERENCES
Iman,
R.
L.
and
W.
J.
Conover.
1983.
A
Modern
Approach
to
Statistics.
John
Wiley
&
Sons,
New
York.

USEPA:
Docket
Number
W­
98­
26,
VI.
A.
a.
13,
Index
to
the
Rulemaking
Record
Effluent
Limitations
Guidelines
and
Standards
for
Synthetic­
Based
Drilling
Fluids
and
other
Non­
Aqueous
Drilling
Fluids
in
the
Oil
and
Gas
Extraction
Point
Source
Category
(
40
CFR
435).

USEPA.
2001a.
"
Final
NPDES
General
Permit
for
New
and
Existing
Sources
and
New
Dischargers
in
the
Offshore
Subcategory
of
the
Oil
and
Gas
Extraction
Category
for
the
Western
Portion
of
the
Outer
Continental
Shelf
of
the
Gulf
of
Mexico
(
GMG290000),"
(
66
FR
65209)
Federal
Register
v.
66
(
Dec
18,
2001)
65209.

USEPA
(
United
States
Environmental
Protection
Agency).
2001b.
Effluent
Limitation
Guidelines
and
New
Source
Performance
Standards
for
the
Oil
and
Gas
Point
Source
Category.
40
CFR
Parts
9
and
35.
Federal
Register
U.
S.
Environmental
Protection
Agency,
Washington,
DC.

USEPA.
2003.
Memorandum,
Linda
Y.
Boornazian
and
Mary
T.
Smith,
Clarification
of
Technology­
based
Sediment
Toxicity
and
Biodegradation
Limitations
and
Standards
for
Controlling
Synthetic­
based
Drilling
Fluid
Discharges,
October
10,
2003.
Page
28
of
30
July
29,
2004
APPENDIX
Table
A­
1:
Phase
1
measurements
for
2­
EH,
ester,
olefin,
and
paraffin.

Mud
Temperature
(
°
F)
Time
(
hr)
Lime
Brine
Green
Cement
2­
ethyl
hexanol
(
wt
%)
Ester
(
wt
%)
Olefin
(
wt
%)
Paraffin
(
wt
%)
A
Ambient
0
yes
no
no
ND
11.9
88.1
TRACE
A
275
16
yes
no
no
0.05
11.9
88.05
TRACE
A
300
16
yes
no
no
0.10
11.5
88.4
trace
A
275
40
yes
no
no
0.10
10.9
89
trace
A
300
40
yes
no
no
0.15
11.2
88.65
trace
A
325
16
yes
no
no
0.20
11.0
88.8
trace
A
350
16
yes
no
no
0.30
10.6
89.1
TRACE
A
325
40
yes
no
no
0.45
9.9
89.65
trace
A
300
88
yes
no
no
0.54
8.85
90.61
trace
A
275
160
yes
no
no
0.55
8.45
91
trace
A
300
160
yes
no
no
0.60
9.3
90.1
trace
A
350
40
yes
no
no
0.70
8.8
90.5
TRACE
A
275
88
yes
no
no
0.85
9.0
90.15
trace
A
325
88
yes
no
no
1.19
7.5
91.31
trace
A
325
160
yes
no
no
1.74
5.69
92.57
trace
A
350
88
yes
no
no
1.89
4.44
93.67
trace
A
350
160
yes
no
no
2.06
3.46
94.54
trace
B
Ambient
0
no
no
no
ND
10.3
89.7
trace
B
300
16
no
no
no
ND
10.6
89.4
trace
B
275
16
no
no
no
0.05
10.5
89.45
trace
B
300
40
no
no
no
0.05
10.3
89.65
trace
B
325
16
no
no
no
0.06
10.4
89.54
trace
B
275
40
no
no
no
0.09
10.0
89.91
trace
B
350
16
no
no
no
0.15
9.36
90.49
trace
B
325
40
no
no
no
0.16
8.8
91.04
trace
B
275
160
no
no
no
0.30
8.84
90.86
trace
B
300
160
no
no
no
0.41
8.41
91.18
trace
B
350
40
no
no
no
0.42
8.53
91.05
trace
B
325
160
no
no
no
0.89
6.71
92.4
trace
B
350
160
no
no
no
0.90
5.68
93.42
trace
C
Ambient
0
no
no
yes
ND
10.6
89.4
trace
C
300
16
no
no
yes
0.06
10.3
89.64
trace
C
275
16
no
no
yes
0.08
10.0
89.92
trace
C
275
40
no
no
yes
0.20
9.7
90.1
trace
C
325
16
no
no
yes
0.27
8.9
90.83
trace
C
300
40
no
no
yes
0.30
9.0
90.7
trace
C
275
88
no
no
yes
0.39
8.48
91.13
trace
C
350
16
no
no
yes
0.62
8.01
91.37
trace
C
325
40
no
no
yes
0.84
7.3
91.86
trace
C
300
88
no
no
yes
0.93
7.3
91.77
trace
C
275
160
no
no
yes
1.29
5.43
93.28
trace
C
350
40
no
no
yes
1.59
5.45
92.96
trace
C
300
160
no
no
yes
1.87
4.41
93.72
trace
C
325
160
no
no
yes
2.36
3.13
94.51
trace
C
350
160
no
no
yes
2.61
1.31
96.08
trace
Page
29
of
30
July
29,
2004
Table
A­
1
continued:
Phase
1
measurements
for
2­
EH,
ester,
olefin,
and
paraffin.

Mud
Temperature
(
°
F)
Time
(
hr)
Lime
Brine
Green
Cement
2­
ethyl
hexanol
(
wt
%)
Ester
(
wt
%)
Olefin
(
wt
%)
Paraffin
(
wt
%)
D
Ambient
0
no
no
no
ND
ND
100
TRACE
D
275
16
no
no
no
ND
ND
100
TRACE
D
350
16
no
no
no
ND
ND
100
TRACE
D
300
16
no
no
no
ND
ND
100
trace
D
325
16
no
no
no
ND
ND
100
trace
D
350
40
no
no
no
ND
ND
100
TRACE
D
275
40
no
no
no
ND
ND
100
trace
D
300
40
no
no
no
ND
ND
100
trace
D
325
40
no
no
no
ND
ND
100
trace
D
300
160
no
no
no
ND
ND
100
trace
D
350
160
no
no
no
ND
ND
100
trace
D
275
160
no
no
no
ND
ND
100
trace
D
325
160
no
no
no
ND
ND
100
trace
G
Ambient
0
no
no
no
0.08
31.9
68.02
trace
G
300
16
no
no
no
0.10
33.4
66.5
trace
G
275
16
no
no
no
0.14
33.4
66.46
trace
G
300
40
no
no
no
0.35
32.0
67.65
trace
G
325
16
no
no
no
0.38
32.7
66.92
trace
G
275
40
no
no
no
0.40
31.8
67.8
trace
G
350
16
no
no
no
0.78
29.19
70.03
trace
G
325
40
no
no
no
0.92
28.7
70.38
trace
G
275
160
no
no
no
1.37
28.99
69.64
trace
G
300
160
no
no
no
1.86
26.17
71.97
trace
G
350
40
no
no
no
2.18
24.95
72.87
trace
G
350
160
no
no
no
2.52
24.22
73.26
trace
G
325
160
no
no
no
2.55
23.60
73.85
trace
H
Ambient
0
no
yes
no
ND
11.8
88.2
ND
H
275
16
no
yes
no
ND
11.9
88.1
ND
H
300
16
no
yes
no
ND
11.8
88.2
ND
H
350
16
no
yes
no
ND
11.8
88.2
ND
H
325
16
no
yes
no
ND
11.4
88.6
trace
H
275
40
no
yes
no
ND
11.9
88.1
ND
H
325
40
no
yes
no
ND
10.5
89.5
trace
H
300
40
no
yes
no
0.02
12
87.98
ND
H
350
40
no
yes
no
0.06
11.9
88.04
ND
H
275
160
no
yes
no
0.06
10.18
89.76
trace
H
300
160
no
yes
no
0.10
11.17
88.73
trace
H
325
160
no
yes
no
0.44
9.64
89.92
trace
H
350
160
no
yes
no
0.77
7.94
91.29
trace
ND
=
non­
detect.
The
detection
limit
was
0.01
wt
%.
Trace
=
<
2
wt
%.
Page
30
of
30
July
29,
2004
Table
A­
2:
Phase
2
measurements
for
2­
EH,
ester,
olefin,
and
paraffin.

Mud
Temperature
(
°
F)
Time
(
hr)
Green
Cement
2­
Ethyl
Hexanol
(
wt
%)
Ester
(
wt
%)
Olefin
(
wt
%)
Paraffin
(
wt
%)
B
275
40
no
0.15
9.38
90.47
trace
B
300
40
no
0.22
9.07
90.71
trace
B
350
40
no
1.27
7.42
91.31
trace
B
350
160
no
1.50
5.49
93.01
trace
B
350
40
no
2.40
5.85
91.75
trace
B
350
40
no
2.46
5.61
91.93
trace
B
350
40
no
2.72
5.61
91.67
trace
B
275
40
yes
0.64
9.08
90.28
trace
B
350
16
yes
1.31
7.84
90.85
trace
B
300
40
yes
1.53
7.49
90.98
trace
B
350
40
yes
2.42
5.87
91.71
trace
B
350
160
yes
4.99
4.19
91.82
trace
E
275
16
no
0
ND
100
trace
E
300
40
no
0
ND
100
trace
E
350
160
no
0
ND
100
trace
E
275
160
yes
0
ND
100
trace
E
300
40
yes
0
ND
100
trace
E
350
16
yes
0
ND
100
trace
G
275
16
no
0.15
18.54
81.31
trace
G
350
16
no
0.58
17.59
81.83
trace
G
300
16
no
1.06
16.66
82.28
trace
G
275
160
no
1.22
17.50
81.28
trace
G
275
16
yes
0.40
17.93
81.67
trace
G
300
40
yes
2.22
15.80
81.98
trace
G
350
160
yes
9.75
4.19
86.06
trace
ND
=
non­
detect.
The
detection
limit
was
0.01
wt
%.
Trace
=
<
2
wt
%.
