3M
Environmental
Laboratory
Report
No.
W1878
Study
Title
Hydrolysis
Reactions
of
Pertluorooctane
Sulfonate
(PFOS)

Data
Requirement:
Based
on
OPPTS:
835.21
10
Author
Thomas
L.
Hafield,
Ph.
D.

Study
Completion
Date
March
27,2001
Performing
Laboratory
3M
Environmental
Laboratory
Building
2­
3E­
09,935
Bush
Avenue
St.
Paul,
MN
55106
Project
Identification
3M
Laboratory
Report
No:
W1878
Total
Number
of
Pages
71
Page
1
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
This
page
has
been
reserved
for
specific
country
requirements.

Page
2
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Statement
of
Non­
Compliance
StudyTitle:
HydrolysisReactionsofPerfluorooctaneSulfonate(
PFOS).
StudyIdentificationNumber:
W1878
This
study
does
not
comply
with
the
requirements
of
the
US
EPA
Good
Laboratory
Practices
Standards
at
40
CFR
Part
792
(TSCA).
However,
many
GLP
standards
were
used
in
the
development
of
the
analytical
method
(Appendix
A),
and
the
quality
assurance
procedures
followed
in
this
study
were
based
on
the
practices
described
in
the
GLP
documentation.

Spo'nsor
Representative
Date
Page
3
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Quality
Assurance
Statement
StudyTitle:
HydrolysisReactions
of
PemuorooctaneSulfonate(
PFOS).
StudyIdentificationNumber:
W1878
The
following
table
provides
details
of
the
audits
performed
by
the
3M
Environmental
Laboratory
Quality
Assurance
Unit
(QAU).

Date
Reported
to
Management
Study
Director
Inspection
Dates
Phase
I
I
8/
11/
00,
8/
16­
17/
00
I
Dataand
Draft
Report
I
8/
18/
00
I
8/
18/
00
I
I
9/
7,8,
11/
00
I
Dataand
Draft
Report
I
9/
11/
00
I
9/
11/
00
1
3/
13­
14/
0
1
I
DraftReport
I
3/
14/
01
1
3/
14/
01
­1
Page
4
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No
.
W1878
Table
of
Contents
Statement
of
Non­
Compliance
............................................................................................
3
Quality
Assurance
Statement
..............................................................................................
4
List
of
Tables
........................................................................................................................
6
List
of
Figures
.......................................................................................................................
6
Study
Personnel
and
Contributors
.......................................................................................
6
Location
of
Archives
.............................................................................................................
7
Summary
..............................................................................................................................
8
Introduction
..........................................................................................................................
9
Summary
of
Kinetics
Model
...............................................................................................
10
Materials
and
Methods
.......................................................................................................
11
Chemical
Characterizations
..........................................................................................
11
Sample
Preparation
......................................................................................................
11
Sample
Analysis
............................................................................................................
12
Deviations
......................................................................................................................
12
Results
and
Discussion
.....................................................................................................
14
Data
Quality
Objectives
(DQO's)
..................................................................................
14
Anomalous
Analytical
Results
.......................................................................................
14
Statistical
Methods
and
Calculations
............................................................................
15
Data
Summary
and
Discussion
.....................................................................................
15
Conclusions
........................................................................................................................
19
References
.........................................................................................................................
20
Signatures
..........................................................................................................................
21
Appendix
A:
Analytical
Method
..........................................................................................
22
Appendix
B:
Kinetics
Model
...............................................................................................
38
Appendix
C:
Selected
Analytical
and
Kinetics
Results
.....................................................
48
Appendix
D:
Selected
Chromatograms
.............................................................................
59
Page
5
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
List
of
Tables
Table
1.
Summary
of
Results
Based
on
PFOS
Concentrations
........................................
8
Table
2.
Summary
of
Results
Based
on
the
Mean
and
Precision
of
PFOS
Measurements
......................................................................................................
8
Table
3.
Characterizations
of
Test
and
Reference
Substances
......................................
1
1
Table
4.
Observed
(50'
C)
Slopes
of
PFOS
Concentrations
in
Aqueous
Buffered
Solution
for
Various
pH
Levels.
..........................................................................
15
Table
5.
Calculated
Slope
of
PFOS
Concentrations
in
Aqueous
Buffered
Solutions
Using
Data
Pooled
Over
All
pH
Levels
...............................................................
16
Table
6.
Degradation
Rate
and
Half
Life
of
PFOS
in
Aqueous
Buffered
Solutions
Based
on
the
Concentration
Mean
and
Standard
Deviation
.........................................
18
List
of
Figures
Figure
1.
Structure
of
the
Potassium
Salt
of
PFOS
...........................................................
9
Figure
3.
Pooled
PFOS
Data
and
Slope
Regression.
......................................................
17
Figure
2.
Observed
PFOS
Degradation
for
Various
pH
levels.
.......................................
16
Study
Personnel
and
Contributors
Study
Director
Thomas
L.
Hatfield,
Ph.
D.
3M
Environmental
Laboratory
Building
2­
3E­
09
935
Bush
Avenue
St.
Paul,
MN
55106
(651)
778­
7863
Sponsor
3M
Corporation
Professional
Services
Contributing
Personnel
Kuruppu
Dharmasiri,
Ph.
D.
Kevin
Macklin
Mark
T.
McCann
Anthony
E.
Scales
Angela
Schuler
Joseph
J.
S.
Tokos,
Ph.
D.
(Pace
Analytical
Services,
Inc.,
1700
Elm
St.,
Minneapolis,
MN
55144)

Page
6
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Location
of
Archives
The
3M
Environmental
Laboratory
will
retain
the
original
data
documents
and
digital
copies
of
the
original
data
related
to
this
work
for
at
least
10
years
following
the
effective
date
of
any
related
final
ruling.
Information
may
obtained
through
written
inquiry
addressed
as
follows:

3M
Environmental
Laboratory
Building
2­
3E­
09
935
Bush
Avenue
St.
Paul,
MN
55106
Page
7
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Summary
We
report
here
the
results
of
our
study
of
the
hydrolysis
of
perfluorooctane
sulfonate
(hereafter,
PFOS).
Our
methods
are
described
below
and
in
Appendix
A
to
this
work;
our
results
are
based
on
the
observed
concentrations
of
PFOS
in
buffered
aqueous
solutions
as
a
function
of
time.
3"
s
Environmental
Laboratory
staff
developed
the
study
procedures;
they
are
based
on
EPAs
OPPTS
Guideline
Document
835.21
I
O
'
but
do
not
fulfill
all
the
requirements
of
the
guideline.
The
chosen
analytical
technique
was
high
petformance
liquid
chromatography
with
mass
spectrometry
detection
(HPLUMS).
Table
1
summarizes
the
results
of
the
study.

During
this
study,
we
prepared
and
examined
samples
at
six
different
pH
levels
between
1.5
and
11
.O
over
a
period
of
49
days,
and
our
results
indicate
no
dependence
of
the
degradation
rate
of
PFOS
on
the
sample
pH
level.
We
have
excluded
the
"Day
0
samples
because
they
were
improperly
stored
before
analysis;
see
the
"Deviations"
section
below.
Our
observations
of
the
PFOS
concentrations
for
the
42
day
incubation
period
(IDay
7"
through
"Day
49),
pooled
over
the
six
observed
pH
levels,
are
presented
in
Table
1.

I
Table
I.
Summary
of
ResultsBasedon
PFOS
Concentrations.
I
Calculated
slope
upper
limit
(20)
Calculated
slope
Calculated
slope
lower
limit
(20)
(day')

6.66
x
IO­
'
5.30
x
1
0­
4
2.98
X
10"
(day­
')
(day­
')

These
results
indicate
no
degradation
of
PFOS;
the
derived
positive
slope
does
not
provide
a
half­
life
estimate.

The
mean
value
and
precision
of
PFOS
concentrations
do
provide
an
estimate
of
the
PFOS
half­
life,
presented
in
Table
2.

I
Table
2.
Summary
of
Results
Based
on
the
Mean
and
Precision
of
PFOS
Measurements
Half
Life
Page
8
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Introduction
Three
primary
chemical
routes
of
environmental
degradation
are
hydrolysis,
photolysis,
and
biodegradation.
Studies
of
these
routes
provide
information
on
the
environmental
persistence
of
both
the
"parent"
compounds
and
their
reaction
products,
and
are
ideally
carried
out
over
the
range
of
chemical
conditions
pertinent
to
both
environmental
and
metabolic
processes.
The
hydrolysis
of
the
potassium
salt
of
PFOS
(or,
more
generally,
its
degradation
in
the
presence
of
H20)
is
addressed
in
this
report.
The
structure
of
the
PFOS
salt
is
illustrated
in
Figure
1.

Fiaure
I.
Structure
of
the
Potassium
Salt
of
PFOS
F
F
F
F
F
F
F
F
O
..

Page
9
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Summary
of
Kinetics
Model
A
full
mathematical
description
of
the
kinetics
model
employed
in
this
study
is
presented
in
Appendix
B.
The
data
in
similar
studies
typically
allow
two
independent
estimates
of
the
hydrolytic
half­
life,
but
only
one
estimate
is
available
from
our
PFOS
data.

A
first
estimate
is
usually
based
on
the
observed
degradation
of
the
compound
in
dilute,
appropriately
buffered
aqueous
solutions.
Equation
BIO
describes
the
estimated
half­
lifeintermsoftheestimatedtotalhydrolysisrate
cp
:

=
­
In(
2)

kP
Eq.
1
We
attempted
to
determine
the
quantity
Gp
from
the
experimental
data
as
described
in
Appendix
B.
The
data
corresponding
to
"Day
7"
were
used
to
determine
the
relative
concentration
ratios
(see
the
"Deviations"
section
below;
see
also
Equations
B8
and
B9).
The
derived
rates,
both
at
various
pH
levels
and
when
pooled
over
all
pH
levels,
were
poorly
determined,
and
indicate
only
that
PFOS
does
not
degrade
in
aqueous
solution.

A
half­
life
second
estimate
(see
Equation
B37)
is
available
from
the
mean
p
and
standard
deviation
0
of
the
observed
PFOS
concentrations,
assuming
that
they
were
essentially
constant
over
the
experimental
portion
of
the
study.
This
estimate
is
Eq.
2
where
A
t
represents
the
sample
incubation
period.

All
the
samples
used
in
this
study
were
maintained
a
reaction
temperature
of
50"
(&
3")
C.
The
quoted
results,
valid
for
the
reaction
temperature
25"
C,
were
calculated
from
our
experimental
results
according
to
methods
described
in
Appendix
B
(Eq.
B38
and
B39).

Page
10
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Materials
and
Methods
Details
of
the
characteristics
of
the
test
materials,
sample
preparation
techniques,
and
analytical
methods
are
presented
in
Appendix
A
(ETS­
8­
13.0,
"Hydrolysis
of
Perfluorooctane
sulfonate
(PFOS)
and
Analysis
by
High
Performance
Liquid
Chromatography
with
Mass
Spectrometry
Detection").
A
summary
of
these
items
is
provided
below,
as
well
as
a
description
the
known
deviations
from
the
procedures
of
Appendix
A.
3M
prepared
and
analyzed
the
samples
included
in
this
study
between
May
24
and
July
12,1999.

Chemical
Chamctemizations
Table
3
describes
the
sources
and
properties
of
the
materials
used
in
this
work.
These
materials
were
used
to
prepare
both
the
samples
and
the
quantitative
standards
used
to
quantify
them.
For
this
reason,
and
because
Equation
1
involves
only
ratios
of
the
parent
and
product
concentrations,
the
resulting
rate
and
half­
life
estimates
are
largely
independent
of
the
material
purity
levels.

1
Table
3.
Characterizations
of
Test
and
Reference
Substances
PFOS
(Potassium
Salt)
THPFOS
Source
ICN
Biomedicals
3M
ICPIPCP
Division
Chemical
Lot
Number
1
Batch
#
171
I
Batch
#
53406
I
Physical
Description
I
LightcoloredpowderBrownwaxysolid
Molecular
Weight
(gm
mole­`)
538
428
Sample
Preparation
We
prepared
four
1
.O­
mL
aqueous
buffer
samples
(a
sample,
a
duplicate,
a
triplicate,
and
a
matrix
spike
at
each
of
six
pH
levels
(1.5,3,
5,7,
9
and
11)
for
analysis
at
eight
time
intervals
(0,
7,
14,21,28,
35,42
and
49
days).
Buffered
solutions
containing
501.5
ng/
mL
of
the
potassium
salt
of
PFOS
and
368
ng/
mL
of
THPFOS
(3,
3,4,4,
5,
5,
6,6,
7,
7,
8,8,
8­
tridecafluorooctane
sulfonic
acid),
the
latter
serving
as
a
surrogate
for
the
compound
PFOS,
formed
the
basis
of
all
these
samples.
The
chosen
buffer
solutions
are
described
fully
in
Appendix
A.

All
the
samples
were
prepared
simultaneously,
and
all
but
the
"Day
0
samples
were
placed
in
an
orbital
incubator/
shaker
maintained
at
50"
(+
3")
C.
After
the
appropriate
incubation
times,
subsets
of
the
sample
vials
were
removed
from
the
incubator;
they
were
then
spiked
(as
required)
with
the
PFOS
solution,
diluted
1O:
l
with
methanol
containing
the
internal
standard
THPFOS,
and
refrigerated
at
­4"
C.
Similar
treatment
of
the
"Day
0
samples
is
prescribed
by
the
analytical
method
(see
Appendix
A);
they
are
Page
11
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
to
be
agitated
for
a
minimum
of
three
minutes,
diluted
and
spiked
as
described
immediately
above,
and
then
refrigerated
rather
than
incubated.
In
this
study,
the
"Day
0
samples
were
not
refrigerated;
see
the
"deviations"
section
below.
Except
during
the
relatively
short
periods
of
time
required
to
prepare
them,
all
the
samples
were
shielded
from
light.

Eight
calibration
standards
containing
PFOS
(301
to
802
ng/
ml)
served
as
the
quantitative
basis
of
the
study.
All
these
standards
were
prepared
at
the
appropriate
pH
levels
using
the
buffer
solutions
described
in
Appendix
A.

Sample
Analysis
The
equipment
we
used
for
the
HPLC/
MS
analysis
was
a
Hewlett
Packard
model
1
100
equipped
with
a
Dionex
lonPac@
NG­
1
HPLC
column
(aqueous
ammonium
acetate/
methanol
solvent
gradient)
and
an
ALS
Model
G1322A
degassing
module.
An
ALS
Model
G1315A
column
heater
maintained
the
column
temperature
at
40°
C,
a
quaternary
pump
supplied
a
column
flow
rate
of
0.3
mumin,
and
an
ALS
Model
G1313A
auto­
sampler
provided
5
pL
sample
injections.
The
detector
was
a
Hewlett
Packard
MSD
mass
spectrometer,
operated
in
negative­
mode
electrospray
ionization
mode;
anions
of
PFOS
and
THPFOS
were
detected
at
the
mass­
to­
charge
(m/
z)
ratios
499,
427,
respectively.
We
processed
the
resulting
data
using
the
computer
program
HP
ChernStation
for
LC
(Rev.
A.
06.0).
Further
analytical
details,
including
the
gradient
elution
program,
instrument
and
detector
parameters,
and
performance
specifications,
are
presented
in
Appendix
A.

Deviations
After
incubation
and
prior
to
analysis,
all
samples
were
stored
at
temperatures
(­
4"
C
or
­20"
C)
rather
than
at
the
prescribed
temperature
of
4"
C
prescribed
in
the
Method.

The
pH
meter
calibration
was
performed
with
only
two
standards
(at
pH
=
7.0
and
10.0)
rather
than
at
the
three
Method­
prescribed
pH
levels
(4.0,
7.0,
and
10.0).

Only
three
ions
were
monitored
in
the
analysis,
instead
of
the
seven
specified
in
the
Method.

Two
samples
(PFOS­
157
and
PFOS­
1
13)
were
quenched
with
18
mL
of
MeOH
rather
than
the
prescribed
9
mL,
and
the
concentration
results
were
corrected
to
reflect
this
fact.

The
buffer
solutions
prepared
for
use
as
"Day
0"
samples
were
stored
at
room
temperature
for
seven
days
prior
to
spiking
with
PFOS
test
analyk
and
processing.
This
led
to
evaporation
of
the
solutions
and
artificially
raised
the
"Day
0"
concentrations.
Two
additional
sets
of
samples
were
prepared
and
analyzed
as
origmlly
intended
for
generation
of
the
"Day
0"
and
"Day
7"
data.
The
results
of
these
additional
analyses
indicated
no
measurable
PFOS
degradation
in
the
first
seven
days.
On
the
basis
of
these
additional
results,
we
have
excluded
the
"Day
0"
data
h
m
the
data
set
used
to
estimate
the
PFOS
degradation
rates
and
half­
life;
the
"Day
7"
data
are
used
in
the
calculation
of
the
concentration
ratios
of
Equation
B8.
For
these
"Day
0"
and
"Day
7"
re­
runs,
the
ammonium
acetate
solution
was
5
mM
rather
than
the
2
mM
specified
in
the
method,
and
Page
12
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
the
sample
injection
was
10
pL
rather
than
the
specified
10
pL;
the
continuing
calibration
verification
sample
also
exceeded
the
allowed
15%
for
these
re­
runs
samples.

Page
13
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Results
and
Discussion
Data
Quality
Objedies
(DQO's)
We
briefly
describe
the
data
quality
objectives
applied
in
this
study
below.
Appendix
A
describes
them
in
greater
detail.
With
the
exceptions
of
the
anomalous
results
noted
below,
all
the
DQO's
were
met.
Appendix
C
presents
the
results
for
each
sample
set,
organized
by
pH
level.
Calibrations.
The
minimum
acceptable
coefficient
of
determination
(P)
for
linear
fits
to
calibration
data
is
0.990.
The
acceptance
criterion
for
individual
calibration
points
is
that
their
values
fall
within
k
25%
of
the
linear
fit
value;
data
outside
this
range
are
excluded
and
the
linear
fit
is
recalculated.
No
more
than
two
points
may
be
rejected
from
a
calibration
data
set.
Data
for
the
high
or
low
calibration
standards
may
be
rejected,
though
this
results
in
a
smaller
effective
calibration
range.
The
average
results
of
calibrations
performed
before
and
after
the
analytical
procedures
are
used
to
calculate
the
analyte
concentrations.
Continuing
Calibration
Verification
(CCV).
Selected
calibration
samples
are
examined
at
the
beginning
of,
during,
and
at
the
end
of
each
analytical
procedure.
The
results
may
not
deviate
by
more
than
k
15%
of
the
known
values.
Matrix
Spikes.
The
acceptable
percent
spike
recovery
range
is
70%
to
130%;
recoveries
outside
this
range
%
place
the
analysis
out
of
control,
and
require
intervention
by
the
Team
Leader
or
designee.
Analyte
specificity
is
demonstrated
by
acceptable
analyte
spike
recoveries.
Identically
Prepared
Samples.
Triplicate
sample
results
with
relative
standard
deviations
(RSDs)
greater
than
25%
place
the
analysis
out
of
control,
and
require
intervention
by
the
Team
Leader
or
designee.
Solvent
Blanks.
Concentration
results
for
solvent
blanks
exceed
neither
5%
of
the
highest
calibration
standard
nor
25%
of
the
lowest
calibration
level.
System
Suitability.
Suitability
was
demonstrated
by
either
an
abbreviated
mass­
to­
charge
(m/
z)
check­
tune
or
performance
of
a
full
auto­
tune
routine.

Anomalous
Analytical
Results
With
the
following
exceptions,
our
analytical
results
met
or
exceeded
the
data
quality
objectives
of
Appendix
A.

0
Spike
Recoveries.
The
triplicate
samples
PFOS­
157
through
159
(pH
=
7.0)
failed
to
meet
this
DQO
and
were
rejected.

Page
14
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Statistical
Methods
and
Calculations
Using
functions
provided
in
Microsoft@
Excel@
software,
we
calculated
means,
standard
deviations,
and
first­
order
rate
constants
(see
Appendix
B,
Equation
B8)
for
various
subsets
of
the
acquired
data.
Our
linear
regressions
included
the
determination
of
constant
terms,
that
is,
the
regression
fits
were
not
forced
to
pass
through
the
origin.

As
described
in
Appendix
B
(Equations
B38
and
B39),
rates
measured
at
50°
C
were
extrapolated
to
25°
C
by
dividing
by
a
factor
of
10;
this
approximation
is
valid
for
reactions,
such
as
these,
with
Arrhenius
heats
of
activation
near
18
Kcal/
mole.*

Data
Summary
and
Discussion
The
LOQ
is
defined
as
the
concentration
of
the
lowest
(accepted)
standard
in
the
calibration
set
for
which
the
known
concentration
exceeds
400%
of
the
indicated
solvent
blank
level
(see
Appendix
A).
During
this
study,
the
LOQ
for
PFOS
was
308
ng/
mL.
Results
for
the
surrogate
compound
THPFOS
were
very
consistent
throughout
the
study.
The
percent
relative
standard
deviations
of
the
measured
values,
calculated
for
each
pH
level,
ranged
from
1.2
%
to
3.4%.

Table
4
presents
the
results
of
the
slope
determinations
(see
Equation
B8)
at
six
pH
levels
and
50°
C.

I
~~
~~~__

Table
4.
Observed
(50"
C)
Slopes
of
PFOS
Concentrations
in
Aqueous
Buffered
Solution
for
Various
pH
Levels.

I+
3.0
5.0
7.0
9.0
11
Observed
Slope
(day­")

0.00025
0.00032
0.0001
1
0.0001
5
0.00057
0.00030
Percent
(20)
Rate
Uncertainty
(day­")

111
73
618
253
89
161
Figure
3
illustrates
these
values,
which
are
all
positive,
and
the
fact
that
they
are
generally
only
poorly
determined;
their
percent
relative
20
(95%
confidence)
uncertainties
range
from
73%
to
618%.
The
data
do
not
indicate
degradation
of
PFOS
at
any
of
the
six
pH
levels.

In
the
absence
of
a
clear
trend
relating
the
degradation
rate
to
sample
pH,
it
is
appropriate
to
"pool"
the
data
over
pH
level
and
determine
the
concentration
slope
using
the
entire
data
set.
Figure
3
illustrates
the
results
of
this
pooled
analysis
according
to
Equation
1,
and
Table
5
summarizes
the
results
of
the
analysis.
It
also
indicates
no
degradation
of
PFOS,
and
the
positive
slope
does
not
provide
a
half­
life
estimate.

Page
15
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Table
5.
Calculated
Slope
of
PFOS
Concentrations
in
Aqueous
Buffered
Solutions
Using
Data
Pooled
Over
All
pH
Levels.

Calculated
slope
Calculated
slope
Calculated
slope
lower
limit
(20)
upper
limit
(20)
(day­
')
2.98
x
104
(day'')
(day­
')
6.66
x
IO*
5.30
x
Figure
2.
Observed
PFOS
Degradation
for
Various
pH
levels.

0.04
'
0.03
­

0.02
­

n
0.01
­

\

B
E
0.00
­
­

­0.01
­

­0.02
­

­0.03
!
I
I
I
I
_.­­_
pH
1.5
pH
3.
C
,
pH
5.0
­­­
,pH
7.
C
­.
I­

pH
9.
C
­1
­1
pH
11
0
10
20
30
40
50
Time
(days)

Page
16
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Figure
3.
Pooled
PFOS
Data
and
Slope
Regression.

0.05
0
Solid
Line:
Dashed
Lines:
y
=
2.983E­
04~
­
5.616E­
03
20
limits(
slopeandintercept)
=
1.481E­
01
0
10
20
30
40
50
60
time
(days)

The
mean
and
standard
deviation
of
the
PFOS
data
do
provide
a
useful
estimate
of
its
half­
life.
Details
of
the
related
calculations
are
presented
in
below
(see
Appendix
B,
Equations
B36
and
837).
The
maximum
degradation
rate
is
given
in
Equation
3:

and
the
minimum
half­
life
is
given
in
Equation
4
Eq.
3
Eq.
4
We
note
that
in
both
Equations
3
and
4,
the
mean
PFOS
concentration
(pp)
and
standard
deviation
(
op
)
can
be
either
molar
or
mass
quantities.
Table
6
presents
the
results
of
the
calculation.

Page
17
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
I
Table
6.
Degradation
Rate
and
Half
Life
of
PFOS
in
Based
on
the
Concentration
Mean
and
Standard
Deviation
1
Page
18
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Conclusions
We
have
performed
a
study
of
the
aqueous
hydrolysis
of
perfluorooctane
sulfonate
(PFOS).
Six
different
pH
levels
were
included
in
the
study,
which
were
carried
out
at
50°
C
and
extrapolated
to
25°
C.
Our
results
based
on
direct
observation
of
the
PFOS
concentration
indicate
no
hydrolytic
degradation
at
any
individual
pH,
nor
do
the
data
pooled
over
all
six
pH
levels.
From
the
mean
value
and
precision
of
PFOS
concentrations,
we
estimate
the
hydrolytic
half­
life
of
PFOS
to
be
greater
than
41
years.

Page
19
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
References
~~

1.
"Fate,
Transport
and
Transformation
Test
Guidelines:
835.21
I
O
:
Hydrolysis
as
a
Function
of
pH,"
U.
S.
EPA
Office
of
Prevention,
Pesticides
and
Toxic
Substances,
publication
number
712­
C­
98­
057,
January
1998.
2.
Experimental
Physical
Chemistry",
F.
Daniels,
et
al.,
McGraw
Hill
Book
CO.
(New
York),
p.
131,
1962.

Page
20
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Signatures
Page
21
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Appendix
A:
Analytical
Method
Method
ETS­
8­
13.0,
"Hydrolysis
of
Perfluorooctane
Sulfonate
(PFOS)
and
Analysis
by
High
Performance
Liquid
Chromatography
with
Mass
Spectrometry
Detection."

This
Appendix
presents
the
analytical
method
employed
in
this
study.

Page
22
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
3M
ENVIRONMENTAL
LABORATORY
METHOD
HYDROLYSIS
OF
~RFLUOROOCTANE
SULFONATE
(PFOS)
AND
ANALYSIS
BY
MASS
SPECTROMETRY
DETECTION
HIGH
PEFWORMANCE
LIQUID
cHR0MATOGRAPHYWIT.
H
MethodNumber:
ETS­
8­
13.0
Adoption
Date:
a
q
los
Effective
Date:

Approved
by:

Laboratory
Page
23
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
1.0
SCOPE
AND
APPLICATION
1.1
This
method
defines
the
steps
for
analysis
of
perfluorooctane
sulfonate
(PFOS)
hydrolysis
products
by
high
performance
liquid
chromatography
(HPLC)
with
mass
spectrometry
(MS)
detection
and
quantitation.
It
is
based
on
EPA
OPPTS:
835.21
10
(Reference
18.1).
PFOS
anion
is
detected
and
quantified
by
this
method
using
the
anion
of
1
,
1,2,2­
tetrahydro(
trideca)
fluorooctane
sulfonic
acid
(THPFOS)
as
the
internal
standard.
Representative
chemical
structures
are
shown
in
Attachment
A.

tetrahydro(
trideca)
fluorooctane
sulfonic
acid
(THPFOS).
1.2
Compatibleanalytes.
Perfluorooctanesulfonate
(PFOS)
and
1,
1,2,2
1.3
Acceptable
matrices
for
analysis.
Aqueous
solutions
at
various
buffered
pHs.
1.4
This
method
is
defined
as
performance­
based.
Target
analyte
or
surrogate
matrix
spike
recoveries
(100
f
30%)
are
used
for
each
sample
matrix
to
evaluate
method
performance.
(Refer
to
Section
10
for
the
quality
control
parameters
to
be
analyzed
by
this
method.
Refer
to
Section
14
for
the
quality
assurance
evaluation
criteria
for
this
method,)

2.0
SUMMARY
OF
METHOD
2.1
Aliquots
of
PFOS
stock
solution
are
added
to
vials
that
contain
b
e
e
r
s
at
pH
1.5,3.0,
5.0,
7.0,9.0,
and
11.0.
The
vials
are
then
placed
in
an
orbital
incubatorhhaker
set
at
50
4
3
"C.
Sets
of
vials
are
removed
at
designated
intervals
and
the
date
and
time
recorded.
.
The
aqueous
sample
from
the
hydrolysis
of
PFOS
is
diluted
1
to
10
with
methanol
(MeOH)
containing
internal
standard.
PFOS,
incubated
in
one
of
the
several
buffers
is
separated
on
a
reverse
phase
Dionex
IonPac@
NG­
1
HPLC
column
using
an
ammonium
acetateMeOH
solvent
gradient,
with
detection
by
electrospray
ionization
mass
spectrometry
in
the
negative
mode.

3.0
DEFINITIONS
3.1
Method
blank.
An
analyte­
free
matrix
to
which
all
reagents
are
added
in
the
same
3.2
3.3
3.4
volumes
or
proportions
as
used
in
the
sampling
processing.
The
method
blank
is
carried
through
the
complete
sample
preparation
and
analytical
procedure.
Because
the
reagent
in
a
hydrolysis
experiment
is
the
aqueous
buffer,
which
is
the
solvent,
the
classical
definition
of
a
method
blank
is
limited
in
applicability.
The
method
blank
is
used
to
document
contamination
resulting
from
the
entire
preparation
and
analytical
process.
Solvent
blank.
A
sample
of
analyte­
free
medium
(for
example,
10:
90
buffered
water/
MeOH)
that
is
not
taken
through
the
sample
preparation
process.
This
blank
is
used
to
evaluate
instrument
contamination.
Sample
triplicates.
Three
samples
taken
from
and
representative
of
the
same
sample
source
and
separately
carried
through
all
steps
of
the
extraction
and
analytical
procedures
in
an
identical
manner.
Matrix
spike
(MS).
Prepared
by
adding
a
known
mass
of
target
analyte
to
specified
amount
of
a
sample
matrix.
This
assumes
that
an
independent
estimate
of
target
analyte
concentration
is
available.
Matrix
spikes
are
used
to
determine
the
effect
of
the
matrix
on
the
method's
recovery
efficiency.

ETS­
8­
1
3
.O
PFOS
Hydrolysis
and
Analysis
by
HPLCiMS
Page
2
of
15
Page
24
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
3.5
3.6
3.7
3.8
3.9
3.10
4.0
Internal
standard
(IS).
A
known
amount
of
a
compound
or
element
similar
in
analytical
behavior
to
the
compound
or
eIement
of
interest,
added
to
al1
samples
and
standards,
and
carried
through
the
entire
measurement
process.
IS
responses
provide
a
reference
for
evaluating
and
controlling
the
precision
and
bias
of
the
applied
analytical
process.
Calibration
standard.
A
dilution
of
various
amounts
of
a
stock,
intermediate
or
purchased
standard
to
achieve
standard
solutions
in
a
concentration
range
of
interest.
Hydrolytic
half­
lives
resulting
fiom
these
analyses
are
calculated
based
on
analytical
ratios
and
not
absolute
numbers.
Therefore,
results
do
not
depend
on
the
purity
of
the
standards
used.
Continuing
calibration
verification
(CCV).
Standards
analyzed
during
an
analytical
run
to
verify
the
continued
accuracy
of
the
calibration
curve.
This
solution
may
or
may
not
be
prepared
fiom
a
different
source
or
lot
number
than
the
calibration
curve
standards.
Dilution.
A
step
in
the
hydrolysis
study
procedure
in
which
a
solvent
is
added
to
the
test
analytehuffer
solution
to
prepare
it
for
instrumental
analysis.
This
step
occurs
after
the
vials
are
removed
from
incubation
and
before
the
samples
are
analyzed.
If
the
solvent
used
is
miscible
with
the
test
analyte/
bufTer
solution,
the
diluting
solvent
is
merely
added
and
mixed.
If
the
diluting
solvent
is
non­
miscible,
a
liquid­
liquid
extraction
is
performed.
IS(
s)
may
be
incorporated
into
the
diluting
solvent,
if
desired.
Limit
of
quantitation
(LOQ).
The
lowest
concentration
that
can
be
reliably
measured
within
specified
limits
of
accuracy
(see
Sections
14.1
and
14.2)
and
precision
(see
Section
14.3)
during
routine
laboratory
operating
conditions.
The
LOQ
is
generally
5
to
10
times
the
minimum
concentration
with
a
99%
confidence
limit
that
the
concentration
is
greater
than
zero.
However,
it
may
be
nominally
chosen
within
these
guidelines
to
simplify
data
reporting.
For
many
analytes,
the
LOQ
is
selected
as
the
lowest
non­
zero
standard
in
the
calibration
curve
that
is
greater
than
4
times
the
level
of
the
solvent
blanks.
Sample
LOQ
are
highly
matrix­
dependent.
Residuals.
The
percentage
difference
between
that
actual
known
concentration
in
a
standard,
versus
the
result
obtained
fiom
the
back
calculation
using
the
calibration
curve's
linear
regression
formula.
For
the
purposes
of
the
study,
the
acceptance
criteria
for
the
residuals
are
k
25%.

Residual
=
known
conc.
­
calculated
conc.
x
100
known
conc.

WARNINGS
AND
CAUTIONS
4.1Healthandsafetywarnings
4.1.1
Wear
the
proper
lab
attire
for
all
parts
of
this
procedure.
Wear
gloves
and
proper
eyewear
at
all
times.
4.1.2
Handle
all
solvents
in
a
hood
for
all
parts
of
the
described
sample
preparation
procedure.
Whenever
possible
and
practical,
dilute
samples
with
solvent
in
a
hood.

ETS­
8­
13
.O
PFOS
HydroIyis
and
Analysis
by
HPL#
MS
Page
3
of
15
Page
25
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
4.1.3
For
potential
hazards
of
each
chemical
used,
refer
to
material
safety
data
sheets,
packing
materials,
and
3M
Environmental
Laboratories
Chemical
Hazard
Review.

4.2
Cautions
4.2.1
Rinse
all
glassware
in
which
standards
are
prepared
with
1:
l
acetoneMeOH,
then
dry
to
reduce
the
possibility
of
contamination.
4.2.2
Ensure
that
the
HPLC
mobile
phases
are
prepared
fresh
prior
to
beginning
a
run
sequence,
and
that
there
is
sufficient
quantity
to
complete
the
run.
Do
not
allow
the
pump
to
run
dry.

the
computer
to
save
all
run
data.
4.2.3
Ensure
that
before
starting
the
run
sequence
there
is
ample
hard
disk
space
on
4.2.4
Ensure
that
there
is
enough
nitrogen
in
the
supply
tank
to
complete
sequence
rUnS.

5.0
INTERFERENCE
5.1
Contaminants
in
solvents,
reagents,
glassware,
and
other
sample
processing
or
analysis
hardware
may
cause
interference.
Use
the
routine
analysis
of
laboratory
method
blanks
to
demonstrate
that
there
is
no
such
interference.

interference
at
low
detection
levels.
The
routine
analysis
of
solvent
blanks
must
be
used
to
demonstrate
that
there
is
no
such
interference.
5.2
,
Contamination
from
columns,
HPLC
tubing,
and
detector
components
may
cause
6.0
EQUIPMENT
6.1
Analyticalbalancesensitive
to
0.1
mg
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
6.10
6.1
1
Shaker,
incubator
capable
of
maintaining
temperature
at
50
k
3
OC
Hewlett­
Packard
(HP)
1
100
HPLC
System,
or
equivalent
6.3.1
Pump,
quaternary,
Model
G13
1
1
A,
or
equivalent
6.3.2
Solvent
degasser,
Model
G1322A
or
equivalent
6.3.3
Autosampler,
ALSModelG1313A,
variableinjectionvolumecapable
6.3.4
Columnheater,
ModelG1316A
Dionex
IonPac@
NG­
1
column,
35
mm
x
4.0
mm,
10
pm
packing
,
or
equivalent
Mass
spectrometer.
Hewlett­
Packard
LC/
MSD,
or
equivalent,
capable
of
operating
in
the
seIected­
ion­
monitoring
mode
Clock,
digital.
Only
one
clock
should
be
used,
to
insure
unambiguous
documentation
of
the
correct
performance
of
procedures.
Centrifuge
capable
of
maintaining
3000
rpm
for
5
min
pH
meter.
Corning
Model
308
pWTemperature
Meter
with
3­
in­
1
gel
filled
combination
electrode
(pWreference/
temperature),
or
equivalent
Refigerator,
capable
of
maintaining
4
­t
2
OC
Data
system.
A
PC
computer
capable
of
controlling
the
HPLC
system
as
well
as
recording
and
processing
signals
from
the
detector
Data
analysis
software:
Hewiett­
Packard
­
ChemStation',
Version
A.
6.03
or
more
recent.

ETS­
8­
13.0
Page
4
of
15
PFOS
Hydrolysis
and
Analysis
by
HPLUMS
Page
26
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
7.0
SUPPLIES
AND
MATERIAIS
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
Vials,
40­
mL
VOA
(I­
CHEM
or
equivalent)
Crimp­
cap
autovials,
1.5­
mL
Labels
Graduated
pipets,
glass,
disposable,
l­
mL
to
10­
mL
Pasteur
pipets,
glass,
disposable
Hamilton
Gastight@
syringes
(precision
f
1
%
of
total
volume),
1
0­
pL
to
1
000­
pL
Volumetric
flasks,
various
sizes
Beakers,
glass,
various
sizes
Automatic
pipettor,
capable
of
dispensing
10­
5000
pL
8.0
REAGENTS
AND
STANDARDS
8.1
Methanol(
MeOH):
HPLC/
SPEC/
GCgradefromEMScience,
or
equivalent
8.2
8.3
8.4
8.5
8.6
8.7
­

18.0
Mi2
water.
Water
with
lower
resistance
must
not
be
used.
Ammonium
acetate
(approximately
2
mM).
This
is
chromatographic
solvent
A
(see
Section
12.3.1).
Example:
An
acceptable
buffer
solution
is
made
by
weighing
out
0.15
grams
of
ammonium
acetate
(reagent
grade)
into
a
weigh
boat
and
then
quantitatively
transferring
to
a
1
­L
volumetric
flask.
Add
10
mL
of
MeOH
as
a
preservative,
dilute
to
the
mark
with
18.0
MSZ
water
and
mix
thoroughly.
Calibration
and
Standard
Stock
Solutions
All
weights
should
be
recorded
to
the
nearest
0.0001
g
8.4.1
PFOS
prepared
in
MeOH.
(Example:
A
PFOS
stock
solution
is
prepared
at
a
concentration
of
approximately
10,000
pg/
mL
by
weighing
0.1000
g
of
PFOS
in
a
10
mL
volumetric
flask
and
bringing
to
the
mark
with
MeOH.
This
solution
is
diluted
in
MeOH
to
make
additional
standards
as
needed.)
8.4.2
THPFOS
prepared
in
MeOH.
(Example:
A
THPFOS
stock
solution
is
prepared
at
a
concentration
of
approximately
10,000
pg/
mL
by
weighing
0.1000
g
of
THPFOS
in
a
10
mL.
volumetric
flask
and
bringing
to
the
mark
with
MeOH.
This
solution
is
diluted
in
MeOH
to
make
additional
standards
as
needed.)
Internal
Standard
(THPFOS)/
Diluting
Solvent
8.5.1
THPFOSprepared
in
MeOH.
(
Example:
a
THPFOS
internal
standard
solution
is
prepared
to
a
concentration
of
approximately
400
ng/
mL
by
diluting
100
pL
of
a
10,000
pg/
mL
THPFOS
stock
standard
(Section
8.4.2)
into
2.5
L
of
MeOH.
This
solution
will
be
used
for
the
final
MeOH
dilution
of
samples
prior
to
analysis.
Buffers
for
calibration
of
pH
Meter.
Purchased
pH
calibration
standards
of
pH
4.0,
7.0,
and
10.0
(Mallinckrodt
or
equivalent).
Buffer
solutions
for
hydrolysis
study.
Prepare
buffer
solution
of
pH
5.0
using
guidelines
fiom
Fate,
Transport
and
Transformation
Test
Guidelines
(Reference
1
8.1).
Prepare
buffer
solutions
of
pH
=
1.5,3.0,7.0,9.0,
and
11.0
using
guidelines
f?
om
CRC
Handbook
of
Chemistry
and
Physics
(Reference
18.2).
Prepare
the
buffer
solutions
in
l­
L
ETS­
8­
13.0
PFOS
Hydrolysis
and
Analysis
by
HPLC/
MS
Page
5
of
15
Page
27
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
quantities.
Calibrate
a
portable
pWtemperature
meter
using
purchased
pH
calibration
standards
of
pH
4.0,7.0,
and
10.0,
and
measure
the
pH
of
all
buffer
solutions.
Prepare
buffer
solutions
of
pH
1.5,3.0,5.0,7.0,9.0,
and
11.0
at
ambient
room
temperature.
The
concentrations
are
given
below.
Record
final
pH
measurements
of
all
buffers.
Store
buffers
in
sealed
glass
containers.
8.7.1
8.7.2
8.7.3
8.7.4
8.7.5
8.7.6
pH
1.5
a)
207
mL
of
0.1
N
HCl
(reagent
grade)

b)
125
mL
of
0.2
M
KC1
(reagent
grade)

c)
Adjust
pH
to
1.5
with
additional
1
N
HC1
d)
Bring
to
a
final
volume
of
1
L
with
18.0
Ma
water
pH
3.0
a)
10.2
g
potassium
biphthalate
(reagent
grade)
dissolved
in
approx.
600
mL
b)
pH
adjusted
with
1
N
HCl
to
3.0
c)
Bring
to
a
final
volume
of
1
L
with
18.0
MQ
water
pH
5.0
a)
Approximately
3.8777
g
ammonium
acetate
(reagent
grade)
added
to
250
b)
Add
250
mL
0.052
M
acetic
acid
(reagent
grade)

c)
Add
18.0
MQ
water
to
about
900
mL
d)
Add
to
glacial
acetic
acid
(about
0.5
mL)
to
adjust
to
pH
5.0
e)
Bring
to
a
final
volume
of
1
L
with
18.0
Mi2
water
pH
7.0
a)
500
mL
0.1
M
KHzPO,
buffer
(reagent
grade)

b)
291
mL
0.1
N
NaOH
(reagent
grade)

c)
Adjust
pH
to
7.0
with
either
1
N
HCl
or
1
N
NaOH
d)
Bring
to
a
final
volume
of
1
L
with
18.0
MQ
water.

pH
9.0
a)
46
mL
of
0.1
N
HCl
b)
125
mL
of
0.1
M
borax
(NqBO,*
H,
O)
(reagent
grade)

c)
Adjust
pH
to
9.0
with
either
1
N
HCI
or
1
N
NaOH
d)
Bring
to
a
final
volume
of
1
L
with
18.0
M
a
water.

p
H
11.0
a)
500
mL
0.05
M
NaIICO2
(reagent
grade)
b)
227
mL,
0.1
N
NaOH
1
8.0
MQ
water
mL
18.0
MQ
water
ETS­
8­
13.0
Page
6
Of
15
Page
28
of
71
PFOS
Hydrobsis
and
Analysis
by
HPLUMS
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
c)
Add
water
to
950
mL
d)
Adjust
pH
to
1
1.0
with
1
N
NaOH
e)
Bring
to
a
final
volume
of
1
L
with
18.0
MR
water
8.8
Test
AnalyteandSpikesolutions:
8.8.1
PFOS
test
analyte
solution.
[Example:
An
analyte
solution
of
PFOS
is
prepared
by
adding
0.5
mL
of
10,000
pg/
mL
PFOS
standard
(Section
8.4.1)
to
st
1
0­
mL
volumetric
flask
and
diluting
to
the
mark
with
MeOH.
A
10­
pL
aliquot
of
this
solution
is
then
added
to
1
mL
buffer
(Section
12.1.5),
resuIting
in
a
final
concentration
of
approximately
500
n
g
/d
after
MeOH
dilution
(Section
12.1.11)].
8.8.2
Spiking
solution.
pxample:
A
spiking
solution
is
preparedby
adding
0.1
mL
of
an
approximately
10,000
ppm
PFOS
standard
(Section
8.4.1)
to
a
10­
mL
volumetric
flask
and
diluting
to
the
mark
with
MeOH.
Final
concentration
after
addition
of
the
spike
(Section
12.1.12)
and
MeOH
dilution
(Section
12.1.1
1)
is
approximately
100
n
g
/d
of
PFOS].

9.0
SAMPLE
HANDLING
9.1
Record
times
of
initial
preparation
and
dilution
on
the
fluorochemical
degradation
9.2
Once
the
9.0
mL
of
diluting
solvent
(Section
8.5.1)
has
been
added
to
the
hydrolysis
(hydrolysis)
analysis
sample
preparation
sheet
(Attachment
B).

mixtures,
the
samples
should
be
analyzed
within
24
hours.
Alternatively,
the
samples
should
be
stored
at
4
f
2
"C
before
dilution
with
MeOH
and
until
analysis
can
be
performed.

10.0
QUALITY
CONTROL
10.1
Internal
Standards.
IS(
s)
are
added
in
a
constant
concentration
to
all
standards,

10.2
Sample
triplicates.
Prepare
and
analyze
all
samples
in
triplicate
to
provide
a
measure
of
samples,
and
matrix
spikes.

the
precision
of
analysis.
The
analyst
shall
accept
RSD
values
(25%.
RSD
values
of
25%
or
greater
should
be
noted.
Appropriate
steps
must
be
taken
to
correct
the
problem
before
analysis
is
allowed
to
proceed
(e.
g.
sample
re­
runs,
additional
blanks,
etc.).
Consult
with
the
Team
Leader
or
designee
for
direction
and
final
acceptance
or
rejection
of
the
analytical
run.

study.
Concentrations
of
the
spike
should
be
approximately
equal
to
a
mid­
range
calibration
standard.
The
matrix
spike
sample
should
be
analyzed
immediately
following
the
sample
triplicates
to
which
it
corresponds.
The
analyst
shall
accept
percent
spike
recoveries
of
100
rt
30%.
Spike
recoveries
outside
of
this
range
should
be
noted.
Appropriate
steps
must
be
taken
to
correct
the
problem
before
analysis
is
allowed
to
proceed
(e.
g.
reruns,
additional
blanks
or
spikes,
etc.).
Before
the
analysis
is
allowed
to
proceed,
consult
with
the
Team
Leader
or
designee
for
direction
and
final
acceptance
or
rejection
of
the
analytical
run.
10.3
Matrix
spikes.
Prepare
a
post­
hydrolysis
matrix
spike
for
each
of
the
pHs
used
in
the
ETS­
8­
13.0
PFOS
Hydrolysis
and
Analysis
by
HPLUMS
Page
7
of
15
Page
29
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
10.4
Solvent
Blank.
Solvent
blanks
should
be
run
before
and
after
every
calibration
curve,
CCV,
and
after
no
more
than
20
injections.
Acceptable
values
for
the
blank
are
values
below
the
limit
of
quantitation
(LOQ)
of
the
instrument
(Section
3.9).
If
analyte
carry­
.over
is
a
problem,
use
back­
to
back
solvent
blanks.

11.0
CALIBRATION
AND
STANDARDIZATION
11.1
Standard
preparation.
Prepare
at
least
six
calibration
standards
of
PFOS
in
MeOH.
Standards
from
approximately
300
n
g
/d
to
800
n
g
/d
PFOS
are
recommended.
This
solution
should
also
contain
appropriate
concentrations
of
the
internal
standard.
11.2
Calibration
standards.
Analyze
the
set
of
calibration
standards
at
the
beginning
of
the
m,
after
every
20
injections,
and
at
end
of
the
run.
Use
the
data
reduction
software
program
for
linear
regression
calculations
to
relate
the
analyte
peak
area
ratio
versus
the
amount
ratio
from
the
internal
standard.
ExternaI
standard
calibration
may
be
used
if
data
review
shows
a
problem
with
the
internal
standard
analysis.
Consult
with
the
Team
Leader
or
designee
for
direction
prior
to
performing
the
external
calibration
methodology.

12.0
PROCEDURES
12.1Sampleandspikepreparation
12.1.1
12.1.2
12.1.3
12.1.4
12.1s
12.1.6
12.1.7
12.1.8
12.1.9
Before
spiking
with
any
of
the
stock
standards,
transfer
approximately
1
mL
of
the
solution
to
an
autovial
and
cap
it.
Use
this
smaller
volume
for
spiking
to
minimize
the
effects
of
evaporation
from
stock
solutions.
Determine
the
number
of
time
points
that
will
be
analyzed.
Each
time
point
will
have
at
least
four
vials
for
each
pH,
multiplied
by
the
number
of
pHs
analyzed.
One
vial
at
each
level
will
be
labeled
as
"sample",
"duplicate",
triplicate",
or
"spike".
Obtain
the
appropriate
number
of
40­
mL
VOA
vials
with
caps
and
cardboard
boxes.
Prepare
appropriate
sample
preparation
worksheets
and
create
labels
and
affix
to
the
vials.
The
labels
should
include
the
sample
number,
pH,
time
point
and
initials
of
the
analyst.
Record
the
pH
of
each
buffer
solution,
Remove
the
cap
of
the
VOA
vial
and
add
1
mL
of
the
appropriate
buffer
solution
to
all
of
the
pre­
labeled
vials.
Always
replace
the
cap
immediately
after
any
addition
to
minimize
evaporation.
To
all
of
the
vials,
add
10
pL
of
the
PFOS
test
analyte
solution
(Section
8.8.1)
with
a
1
0­
pL
Gastight@
syringe.
Record
the
time
of
addition
for
each
vial.
For
"Time
Zero"
samples
only,
proceed
to
section
12.1.11.
For
all
other
samples,
continue
on
to
section
12.1.7.
Make
sure
that
the
cap
has
been
firmly
tightened
and
place
the
samples
back
in
the
cardboard
case.
Place
the
case
into
a
pre­
warmed
incubatodshaker
for
the
appropriate
time.
Record
the
time,
temperature,
and
rate
of
shaking.
The
temperature
is
determined
by
the
conditions
of
the
experiment.
Continue
to
manually
monitor
the
incubator
temperature
daily
during
the
entire
incubation.
Record
the
temperature
on
the
sample
preparation
sheet
(Attachment
B).
Remove
the
case
from
the
incubator
at
the
designated
preset
time.

ETS­
8­
13
.O
PFOS
Hy&
olysis
and
Analysis
by
HPLC/
MS
Page
8
of
15
Page
30
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
12.2
12.3
12.4
12.1.10
Remove
the
vials
from
the
case
and
place
in
racks.
Allow
the
vials
to
cool
for
approximately
15
minutes
to
room
temperature.
Alternatively,
refrigerate
the
vials
at
4
k
2
"C
if
solutions
are
to
be
quenched
at
a
later
date.
12.1.11
Using
a
10­
mL
graduated
pipet,
add
9
mL
MeOH
containing
approximately
400
ng/
mL
THPFOS
internal
standard
(Section
8.4.2)
to
each
vial.
12.1.12
Using
a
25­
pL
Gastight
syringe,
add
10
pL
of
PFOS
spiking
solution
(Section
8.8.2)
to
the
"spike"
vials.
Invert
each
vial
several
times
to
mix
the
contents.
12.1.13
Aliquot
approximately
1
mL
of
each
sample
to
the
appropriately
labeled
autovial.
Cap
the
vials
and
mark
the
bottom
of
the
meniscus.
12.1.14
Place
the
vials
in
the
HPLC
autosampler.
Instrument
set
up
12.2.1
Check
that
the
appropriate
HPLC
column
is
in
the
instrument
for
each
analysis.
12.2.2
Check
that
the
correct
eluent
solutions
are
in
bottles
to
be
used
and
that
enough
12.2.3
Place
the
samples
in
the
autosampler
tray
and
construct
a
sequence
table
with
is
available
to
complete
the
sequence
run.

appropriate
calibration
standards,
calibration
check
standards
and
solvent
blanks.

information:
sample
or
standard
ID,
method
name,
one
injection
per
sample.

03
1499s).
Save
data
to
a
subdirectory
labeled
with
analysis
date
(e.
g.
031499).

"STANDBY"
on
HP
systems).
12.2.4
Verify
that
all
samples
and
standards
are
positioned
correctly.
Enter
sequence
12.2.5
Save
sequence
as
analysis
date
(e.
g.
on
March
14,1999
save
sequence
as
12.2.6
Set
post­
sequence
cornmand
macro
to
shut
down
system
(Example:

HPLC
set
up:
12.3.1
Analysis
of
PFOS
Samples
from
pHs
1.5,3.0,5.0,7.0,9.0
and
11.0.
Install
the
column:
Dionex
IonPac@
NG­
l,
4.0
x
35
mm,
10
pm
or
equivalent.
Solvent
A:
2mM
Ammonium
Acetate
in
1%
MeOH
(Section
8.3)

Solvent
B:
MeOH
Solvent
Gradient:

Time,
minutes
0.3
mIJmin
95
5
8.0
0.3
mL,/
min
95
5
4.0
0.3
mL/
min
40
60
1
.o
0.3
mLJmin
40
60
0.0
Flow
Rate
%B
%A
Post
time:
5.0
minutes;
Column
Temperature:
40"
C
Recommended
Mass
Spectrometer
set
up;
*
12.4.1
Analysis
of
PFOS
at
pHs
1.5,3.0,5.0,7.0,9.0
and
11.0,

ETS­
8­
13.0
PFOS
Hydrolysis
and
Analysis
by
HPLCYMS
Page
9
of
15
Page
31
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
­
Drying
Gas
Temp
3500
V
Capillary
Voltage
300
"C
`Example
conditions
are
applicable
to
Hewlett
Packard
HPl
100
equipment
only.

Time
1
SIM
1
Identity
I
Gain
I
Fragmentor
I
Dwell
I
.
..
.
5.6
210
70
I
.o
PFOS
aualitative
ion
500
6.2
210
70
1
.o
PFOS
quantitation
ion
(M­
H)
499
6.2
210
70
1
.O
PFHS
13C,
(M­
H)
qualitative
ion
400
5.9
210
70
1
.O
THPFOS
(M­
H),
quantitation
ion
427
I
I
*
"Quant"
ion
for
THPFOS;
**"
Qua&
'
ion
for
PFOS
..

1­
2­
3­

12.5
12.6
13.0
pduorohexanesulfonaide
perfluorohexanesulfonate
perfluorooctanesulfonamide
Autosampler
set
up*

Autosampler:

5.0
pL
Injection
volume:
None
Autosampler
Program:
ALS
Model
G13
13A
*Example
conditions
are
applicable
to
Hewlett
Packard
HP1100
equipment
only.

Sample
analysis
12.6.1
12.6.2
12.6.3
12.6.4
12.6.5
Enter
the
standard,
sample
and
QC
information
into
the
sequence
table.
Analyze
calibration
standards,
up
to
20
injections,
and
the
calibration
standards
again.
If
more
than
20
injections
are
to
be
run,
analyze
a
continuing
calibration
standard
(CCV)
after
every
20
and
run
the
calibration
standards
again
at
the
end
of
the
sequence.
Run
solvent
blanks
after
the
highest
calibration
standard,
before
and
after
the
CCV,,
and
after
the
set
of
samples
to
check
for
any
malyte
carryover.

exceed
five
characters
if
the
sequence
contains
more
than
99
lines.
Place
the
standards,
samples,
and
QC
(matrix
spikes
and
sample
blanks)
into
the
autosampler
tray
according
to
the
order
they
are
listed
in
the
sequence.
Save
the
sequence
table
with
a
name
corresponding
to
today`
s
date.
(e.
g.
if
today
is
December
1,1998,
save
the
sequence
as
120198.)
Start
the
sequence.
Identify
the
electronic
acquisition
files
with
an
appropriate
prefix.
Do
not
DATA
ANALYSIS
AND
CALCULATIONS
13.1
Peak
Evaluation:
Peaks
must
be
symmetric
in
shape
and
identified
by
extracting
compound
specific
ions.
Peaks
considered
for
calibration
must
have
peak
heights
greater
than
5
(five)
times
the
baseline
noise
for
that
region
of
the
chromatogram.
Peak
integration
is
from
baseline
to
baseline
through
a
peak
using
automatic
or
manual
integration.
Compounds
with
isomers
present
as
a
shoulder
or
as
a
discrete
second
peak
ETS­
8­
13.0
PFOS
Hydrolysis
anddnalysis
by
HPLC'MS
Page
10
of
15
Page
32
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
13.2
13.3
13.4
13.5
14.0
should
be
integrated
with
the
parent
compound
unless
otherwise
noted.
Quantitation
data
are
calculated
using
THPFOS
as
the
internal
standard.
However,
external
standard
calibration
may
be
acceptable.
Consult
with
the
Team
Leader
or
designee
for
direction
prior
to
performing
the
external
calibration
methodology.
Calculation
of
k
Calculate
the
PFOS
concentrations
in
each
of
the
pH
matrices
using
the
curves
obtained
from
the
calibrations
and
the
internal
standard.
Assuming
first­
order
kinetics
a
rate
constant
(k)
can
be
determined
by
plotting:

concentrations
determined
at
some
elapsed
time
t
and
at
t
=
0,
respectively.
The
slope
of
the
resulting
line
is
k.
The
iL
value
for
this
plot
should
be
>
0.800.
For
rZ
values
less
than
0.800,
consult
the
Team
Leader
or
designee
for
direction.
Calculate
the
PFOS
concentrations
in
each
of
the
pH
matrices
versus
t
using
the
curves
obtained
from
the
calibrations
and
the
internal
(or
external)
standard.
Matrix
spikes.
Calculate
the
percent
recovery
for
each
of
the
matrix
spikes.
Using
the
observed
matrix
spike
recoveries,
calculate
the
average
spike
recovery.
Calculate
the
matrix
spike
percent
recoveries
using
the
following
equation:

YO
Recovery
=
(observed
spiked
sample
result
­
observed
sample
result)
x
actual
amount
spiked
Sample
triplicates.
Calculate
the
relative
standard
deviation
(RSD)
for
the
triplicate
samples:

RSD
=
StandardDeviationx
100%
Mean
Conc.

This
parameter
is
also
known
as
the
coefficient
of
variation,
a
measure
of
the
confidence
of
the
mean.

METHOD
PERFORMANCE
14.1
Coefficient
of
Determination
(3.
An
acceptable
coefficient
of
determination
(13,
for
14.2
linear
curves
is
0.990
or
greater.
The
curves­
should
be
examined
closely
for
linea&
and
intercept,
particularly
for
accuracy
of
quantitation
at
the
low
and
high
ends
of
the
curve.
The
acceptance
criteria
for
curve­
fitting
residuals
(Section
3.10)
is
&
25%.
Alternative
methods
of
curve
fitting
(e.
g.
quadratic)
may
be
necessary
in
some
cases.
An
acceptable
correlation
coefficient
(r)
for
qkdratic
curves
is
0.990
or
greater.
Record
in
the
raw
data
the
reasons
for
using
quadratic
equations.
Matrix
spikes.
Performance
based
warning
and
control
limits
are
recommended
for
measurements
of
percent
spike
recovery
from
matrix
spike
samples.
The
analyst
shall
accept
percent
spike
recovery
values
of
100
&
30%.
Spike
recoveries
outside
of
this
range
place
the
analysis
out
of
control.
Appropriate
steps
must
be
taken
to
correct
the
problem
before
analysis
is
allowed
to
proceed.
Consult
the
Team
Leader
or
designee
for
direction.

ETS­
8­
13.0
PFOS
Hydrolysis
and
Analysis
by
HPLC/
MS
Page
11
of
15
Page
33
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
14.3
14.4
14.5
14.6
15.0
Sample
Triplicates.
Performance
based
warning
and
control
limits
are
recommended
for
measurements
of
relative
percent
difference
of
replicate
samples.
The
analyst
shall
accept
RSD
values
<
25
%.
RSD
values
of
25%
or
greater
place
the
analysis
out
of
control.
Appropriate
steps
must
be
taken
to
correct
the
problem
before
analysis
is
allowed
to
proceed.
Before
analysis
is
allowed
to
proceed,
consult
the
Team
Leader
or
designee.
Continuing
calibration
verification.
If
the
percent
difference
for
the
amount
of
measured
analyte
is
greater
than
15%
from
the
true
value,
relative
to
the
initial
standard
curve,
stop
the
run.
Only
those
samples
analyzed
before
the
last
acceptable
calibration
check
standard
will
be
used.
Reanalyze
remaining
samples
with
a
new
calibration
curve.
Limit
of
Quantitation.
The
limit
of
quantitation
is
equal
to
the
lowest
standard
in
the
calibration
curve
that
is
greater
than
4
times
the
level
of
the
solvent
blanks.
Solvent
blanks.
Solvent
blanks
should
show
no
more
than
a
5%
carry
over
from
a
high
standard
or
calibration
check
standard.
If
so,
two
solvent
blanks
may
be
necessary
to
rule
out
instrumental
contamination.
If
peaks
with
greater
than
25%
the
peak
area
of
a
low
standard
value
are
observed
in
sequential
solvent
blanks,
the
run
should
be
stopped.
This
indicates
instrument
contamination.
The
instrument
shall
be
maintained
by
thoroughly
cleaning
the
electrospray
source,
and
replacing
or
cleaning
columns,
tubing,
etc.

POLLUTION
PREVENTION
AND
WASTE
MANAGEMENT
15.1
Dispose
of
sample
waste
by
placing
in
high
or
low
BTU
containers
as
appropriate.
Use
15.2
Collect
HPLC
solvent
waste
in
the
satellite
accumulation
can.
Empty
into
the
flammable
15.3
Use
smaller
bore
columns
when
possible
to
minimize
waste
generation.
broken
glass
containers
to
dispose
of
glass
pipettes.

storage
drum
in
the
hazardous
waste
collection
area
on
the
2nd
floor.

16.0
16.1
16.2
163
16.4
16.5
16.6
16.7
16.8
REcoRlDs
Print
out
hard
copies
of
all
graphics
and
data
analysis
summaries
for
archiving
Sign
and
date
all
graphics
and
label
with
instrument
ID.
Fill
out
the
hydrolysis
sample
preparation
worksheet
completely,
making
sure
to
include
all
initials
and
dates.
Print
chromatograms
and
internal
standard
reports
for
all
analyses.
Print
calibration
tables
and
curve
information
and
store
in
the
raw
data
file.
Store
hydrolysis
sample
preparation
worksheets
in
the
raw
data
file.
Enter
all
standard
preparation
information
in
the
standards
preparation
logbook.
Make
a
photocopy
of
the
logbook
page
and
include
the
copy
in
the
raw
data
file.
Archive
electronic
data
to
appropriate
media
when
necessary.

17.0
ATTACHMENTS
17.1
Attachment
A.
RepresentativeChemicalStructures
17.2
Attachment
B.
HydrolysisSampleLogsheet
ETS­
8­
13.0
PFOS
Hydroiysis
and
Analysis
by
HPLUMS
Page
12
of
15
Page
34
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
18.0
BIBLIOGRAPHY
18.1
Fate,
Transport
and
Transformation
Test
Guidelines
Office
of
Prevention,
Pesticides
and
Toxic
Substances
(OPPTS)
835.21
10
Hydrolysis
as
a
Function
of
pH,
EPA
712­
C­
98­
057,
January
1998.

Operational
Definitions
of
pH,"
Robert
C.
Weast,
Ph.
D.,
1988,
p.
D­
87.
18.2
CRC
Handbook
of
Chemistry
and
Physics,
1st
Student
Edition,
``
Buffer
Solutions
19.0
AFFECTED
DOCUMENTS
None.

20.0
REVISIONS
Revision
Number
ETS­
8­
13
.O
PFOS
Hydrolysis
and
Analysis
by
HPLUMS
Page
13
of
15
Page
35
of
71
BACK
TO
MAIN
m
c
z
0
0
\\
;o
BACK
TO
MAIN
."
..
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Appendix
B:
Kinetics
Model
This
Appendix
includes
a
mathematical
description
of
the
kinetics
model
employed
in
the
study.
Appendix
Page
38
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Kinetics
Model
BI.
Reaction
Components
and
Rates
The
arguments
below
are
based
on
the
following
idealized
set
of
reactions
representing
the
hydrolysis
of
a
parent
compound
P
and
its
hydrolysis
products
A,,
which
number
N.
The
actual
hydrolysis
reactions
that
occur
under
neutral,
acidic,
and
basic
conditions
are
subsumed
in
these
equations,
and
are
assumed
to
proceed
with
pseudo­
first
order
rates
k,,
(for
the
parent)
and
k,,
(for
the
parent's
hydrolysis
products).

I­
'
+
H,
O
w
n,
A,+
Y,
(m=
ltoN)
krm
A,
+
H20
w
Y,,
k
h
(m=
1
toN)
(B2)

where
the
general
symbols
Y,,
and
Ym2
represent
all
the
other
hydrolysis
products.

B2.
Parent
Compound
Concentrations
Equation
B1
indicates
that
the
pseudo­
first
order
differential
change
in
the
parent
concentration
P
is
given
by
which
is
equivalent
to
the
separable
differential
equation
dp=­[
z
P
n,
k,,]
dt
Equation
B4
may
be
directly
integrated
.to
obtain
the
general
solution
Ink]=
­x
n,
k,,
t
+C
(
my,
]

With
the
initial
condition
P(
t
=
0)
=
Po,
the
specific
solution
to
Equation
B4
is
P
=Po
exp
[­&
n,
k,,
t]=
P,
e­
kpt
using
the
additional
definition
of
the
total
parent
hydrolysis
rate
Page
39
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
N
k,
=
n,
k,,
.

Equation
B6
can
be
re­
written
in
a
form
that
allows
a
least­
squares
estimate
of
the
total
parent
hydrolysis
rate:

k,
t
=­
In
(i)

Using
the
initial
(t
=
0)
measured
value
of
the
parent
concentration
Po
and
later
values
P
measured
at
later
times
t
,
one
can
calculate
and
plot
the
(linear)
quantity
[­
In
(P/
Po)]
versus
time
and
obtain
a
least
­squares
estimate
of
the
slope
of
the
line.

The
resulting
slope
is
the
least­
squares
estimate
c,
of
the
total
parent
hydrolysis
rate.

Equation
B6
indicates
that
over
a
period
of
time
TI':
(the
parent
hydrolysis
half­
life)
the
parent
concentration
P
is
reduced
through
hydrolysis
by
a
factor
of
two,
where
A
least
squares
estimate
?
vt
of
the
parent
hydrolysis
half­
life
is
therefore
available
from
B3.
ProductCompoundConcentrations
The
pseudo­
first
order
differential
changes
in
the
product
concentrations
4,
(using
Equations
B2
and
B6)
are
dA,
=
(
n,
kp,
P
­
k,,
A,)
dt
=
(
nmkPmPO
e­
kp
­
k,,
A,)
dt
(B11)

and
the
(first
order,
non­
separable)
differential
equation
governing
the
product
concentrations
is
%+
k,
A,
=
n,
k,,
Po
e­
kpt
.
dt
The
"standard
form"
of
Equation
B12
is
Page
40
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Ab
+
S
(t)
A,
=
Q(
t)

where
the
"function"
S
(t)
is
actually
a
constant:

and
Q(
t)
=
n,
k,,
Po
e­
kpt
.

The
general
solution
A,
to
Equation
B12
is
contained
in
where
and
There
are
two
cases
of
Equation
818
to
consider.
In
the
circumstance
that
k,
=
k,
,
which
occurs
only
when
the
hydrolysis
rate
of
the
mth
product
is
identical
to
the
total
parent
hydrolysis
rate,
the
general
solution
to
Equation
B18
is
(for
k,
=
k,)

A,
ekpt
=
n,
k,,
Po
t
+
C
and,
using
the
initial
condition
A,(
t
=
0
)
=
A,
,
the
specific
solution
to
Equation18
is
(for
k,
=
k,
)

A,
=
(nmkPmPO
t
+
A,)
e­
kpt
We
note
that
when
k,
=
k,
=
0
(that
is,
when
both
the
parent
and
potential
product
are
hydrolytically
stable),
Equation
B7
requires
(also)
that
k,,
=
0
,
so
Equation
B20
becomes
Page
41
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Am
=
A,

indicating,
as
required,
that
the
productconcentration
does
not
change
with
time.

The
circumstance
k,
=
kp
is
highly
improbable,
and
is
neglected
in
the
remainder
of
this
discussion.
However,
the
reader
should
bear
in
mind
that
the
expressions
derived
below
do
not
hold
when
the
parent
hydrolysis
rate
k,
and
the
product
hydrolysis
ratek,
approach
each
other.

In
the
more
probable
case,
for
which
k,,
#
k,
(i.
e.
that
the
hydrolysis
rate
of
the
mth
product
is
different
from
the
total
parent
hydrolysis
rate),
the
general
solution
to
Equation
B18
is
and
the
specific
solution
to
Equation
B18
with
the
initial
condition
Am(
t
=
0)
=
Am,
is
nrnkPmPO
e­
k,
t
­
nmkPmPO
e­
kp
t
kF'
­
kArn
1
kP
­
kAm
Of
greatest
interest
here
is
the
case
in
which
the
product
compounds
are
known
to
be
hydrolytically
stable,
that
is,
when
k,
=
0
for
all
m.
In
this
case,
Equation
B23
becomes
(for
hydrolytically
stable
products)

A,
=AmO
+
nmkPmPO
(1­
e
­k
p
t
)
.
kP
B4.
Relationships
Between
the
Parent
and
Compound
Concentrations
Equations
B7
and
B24
can
be
combined
to
obtain
(for
hydrolytically
stable
products)

Page
42
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
so
that
(for
hydrolytically
stable
products)

or
(for
hydrolytically
stable
products)

If
the
changes
in
the
product
concentrations
are
all
small
compared
to
the
original
parent
concentration,
that
is,
if
we
may
use
the
expression
(valid
for
­1
5
X
I
1
)

1
1
1
4
ln(
l+
X)=
X­
­x2
+­
x3­­
x4+
.....
3
2
L
J
and
Equation
823
becomes
(for
hydrolytically
stable
products
and
Page
43
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
or
(for
hydrolytically
stable
products
and
Z
A
,
­Amo
m
k,
t
E
m=
l
B5.
Parent
Half­
Life
Estimates
Based
on
Limits
of
Quantification
of
the
Products
In
every
experimental
determination
of
k,
,
there
is
some
set
of
values
A
r
Q
(the
"limits
of
quantitation")
below
which
the
product
concentrations
A,
cannot
be
reliably
measured.
If
during
an
experiment
carried
out
over
the
period
of
timeA
t
all
the
product
concentrations
A,
remain
below
their
limits
of
quantitation,
then
the
maximum
possible
value
of
the
rate
k,
is
obtained
by
assuming
(for
all
the
products)
that
1)
A,,
=
0
and
2)
at
time
t
=
A
t
,
the
product
concentrations
have
increased
to
the
values
A,
=
AZQ.
With
these
assumptions,
the
experimental
data
indicate
that
the
reaction
rate
k,
is
less
than
some
maximum
value
(kp)­
as
follows:

(for
hydrolytically
stable
products
at
concentrations
below
the
limits
of
quantitation)

Under
the
same
circumstances
and
assumptions,
the
experimental
data
indicate
that
the
parent
half­
life
TI':
(see
Equation
B9)
is
greater
than
the
value
(TVi)
.
as
follows:
mm
(for
hydrolytically
stable
products
at
concentrations
below
the
limits
of
quantitation)

TV2
1
=­­
P
h­
42)
­
A
t
Po
In(
2)
(kP
1­

The
reader
should
note
that
Equations
B32
and
B33
are
valid
only
when
both
1)
the
products
are
hydrolytically
stable
and
2)
the
concentrations
of
all
the
potential
products
are
measured.
Otherwise,
the
quantity
(k,),,
in
Equation
B32
may
not
actually
represent
the
maximum
possible
value
of
the
rate
constant
k,
,
and
the
related
result
in
Equation
B33
for
(TVi)
.
is
also
questionable.
mm
Page
44
of
79
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
B6.
Parent
Half­
Life
Estimates
Based
on
Limits
of
Quantification
and
Experimental
Precision
of
Product
Concentrations
In
certain
experiments,
some
hydrolysis
products
are
present
at
quantifiable
but
essentially
constant
concentrations
over
the
time
(A
t
)
of
the
experiment.
In
this
case,
it
is
the
experimental
precision
of
the
measured
product
concentrations,
rather
than
the
limits
of
quantitation,
which
contribute
to
the
estimate
of
the
maximum
value
of
the
parent
hydrolysis
rate
k,
.
If
the
set
of
concentrations
measured
for
the
mth
product
have
the
mean
value
p,
and
standard
deviation
CY,,
the
data
do
not
exclude
the
possibility
that
the
product
concentration
increased
from
the
initial
value
G,,,
­p,
to
the
value
0,
+
p,
at
time
t
=
A
t
.
Taking
this
possibility
to
be
the
actual
case
for
the
measured
products,
the
maximum
value
of
the
quantity
(A,
­A,,
)
is
20,.
This
reasoning
suggests
that
the
following
estimate
of
the
maximum
parent
hydrolysis
rate
is
appropriate:

(for
hydrolytically
stable
products
at
either
1)
constant
measured
concentrations
with
standard
deviation
Om
or
2)
concentrations
below
the
limits
of
quantitation)

r
1
kp
5
k
P
),
=
­1
I
A
r
Q
+
c20,
j.

'0
*
Below
LOQ
Cons
tan
t
Under
these
circumstances
and
assumptions,
the
experimental
data
indicate
that
the
parent
half­
life
T1'i
is
greater
than
the
value
(T
v
i
)
.
as
follows:
nun
(for
hydrolytically
stable
products
at
either
1)
constant
measured
concentrations
with
standard
deviation
Om
or
2)
concentrations
below
the
limits
of
quantitation)

r
7
­1
The
reader
should
note
that
Equations
B34
and
B35
are
valid
only
when
both
1)
the
products
are
hydrolytically
stable
and
2)
the
concentrations
of
all
the
potential
products
are
measured.

B6.
Parent
Half­
Life
Estimates
Based
on
the
Experimental
Precision
of
Parent
Concentrations
In
certain
experiments,
the
hydrolytic
parent
remains
at
an
essentially
constant
concentration
over
the
time
(A
t
)
of
the
experiment.
In
this
case,
it
is
the
experimental
precision
of
the
measured
parent
concentrations
that
determines
the
maximum
value
of
the
parent
hydrolysis
rate
k,
.
If
the
set
of
concentrations
measured
for
the
parent
have
the
mean
value
pLp
and
standard
deviation
op,
the
data
do
not
exclude
the
possibility
Page
45
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
where
R
=
1.99
x
Kcal
mole­
'
K­
'
is
the
ideal
gas
constant.
Using
the
valueB2
AH,
=I8
Kcallmole,
the
rate
ratio
kl/
k2
at
the
corresponding
temperatures
T,
=298
K
and
T,
=323
K
is
that
the
product
concentration
increased
from
the
initial
value
pP
­
crP
to
the
value
pp
+
op
at
time
t
=
A
t
.
This
reasoning
suggests
that
the
following
estimate
of
the
maximum
parent
hydrolysis
rate
is
appropriate:

(for
essentially
constant
parent
concentrations
with
mean
value
pp
and
standard
deviation
Gp
)

Under
these
circumstances
and
assumptions,
the
experimental
data
indicate
that
the
parent
half­
life
T1'i
is
greater
than
the
value
(T
v
i
)
.
as
follows:
nun
(for
essentially
constant
parent
concentrations
with
mean
value
pp
and
standard
deviation
Gp
)

B8.
Temperature
Dependence
of
the
Reaction
Rate
and
Half­
Life
In
order
to
increase
the
speed
of
the
reactions
of
interest,
we
conducted
this
experimental
study
using
samples
maintained
at
the
temperature
50°
C
=
323
K.
Of
greater
interest
are
the
corresponding
results
for
the
environmentally
important
temperature
25°
C
=
298
K.

When
the
Arrhenius
activation
energy
for
a
reaction
is
AH,,
Equation
838
B1
provides
the
following
relationship
between
the
hydrolysis
rates
(kl
and
k2)
for
that
reaction
at
two
different
absolute
temperatures
(
Tl
and
T2):

Page
46
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Equation
B39
indicates
that
the
hydrolysis
reactions
of
interest
proceed
approximately
ten
times
more
slowly
at
25°
C
than
at
the
chosen
experimental
temperature
of
50°
C.
Accordingly,
the
rate
reactions
reported
here
for
the
temperature
25°
C
are
ten
times
lower
than
those
measured
at
50°
C,
and
the
hydrolysis
half­
life
estimates
reported
here
for
25°
C
samples
are
ten
times
longer
than
those
calculated
from
the
50°
C
experimental
data.

References
to
Appendix
B:

''
I.
N
Levine,
"Physical
Chemistry,"
McGraw­
Hill
(New
York),
pp.
498­
501
(1978).

'*
F.
Daniels,
et
ai.,
"Experimental
Physical
Chemistry",
McGraw
Hill
(New
York),
p.
131
(1962).

Page
47
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Appendix
C:
Selected
Analytical
and
Kinetics
Results
This
Appendix
includes
selected
sample
data
and
their
related
kinetics
results.
For
brevity,
some
solvent
blank
and
continuing
calibration
verification
results
that
did
meet
the
data
quality
objectives
have
been
excluded
from
the
following
tables.
However,
all
data
failing
to
meet
the
data
quality
objectives
are
included
below.

Page
48
of
71
BACK
TO
MAIN
­r
BACK
TO
MAIN
z
0
?
E
2
a
1­
BACK
TO
MAIN
u
A
t
*
vl
0
BACK
TO
MAIN
s
E
s
s
B
m
5
8
g
s
N
6
r
s
E
­
s
$0
s
BACK
TO
MAIN
ae
r
$5
0
z
NNNNNNNNNNNNNNNN
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
NNNNNNNP
d
d
d
d
d
d
d
u
BACK
TO
MAIN
z
0
BACK
TO
MAIN
z
0
BACK
TO
MAIN
co
b
2
4
0
BACK
TO
MAIN
03
b
03
5
0
z
BACK
TO
MAIN
,,

3M
Environmental
Laboratory
Report
No.
W1878
Figure
3.
Pooled
PFOS
Data
and
Slope
Regression
0.05
n
p"

+
0.00
E4
­0.05
0
0
­/.o
*­­­­­

O
F
I
S
!

6
­­­­­
W
0
8
3
­9.
I
I
­
­
­
0
e­­­­­

8
0
0
­
V
0
__________._.­.­.­­
­­.­­­­­..­.­­­­.­­
­0
0
Solid
Line:

Dashed
Lines:
y
=
2.983E­
04~
­
5.616503
20
limits
(slope
and
intercept)
R2
=
1.481E­
01
I
I
I
1
1
10
20
30
40
time
(days)
50
60
SUMMARY
OUTPUT
Multiple
R
Regression
Sfatisfics
0.384852267
R
Square
0.148111267
Adjusted
R
Square
0,126267966
Standard
Error
0.010265446
Observations
41
ANOVA
.
..
.
df
SS
MS
F
Regression
Signitiiance
F
10.0007145380.0007145386.7806266280.0129706
Residual
Total
39
0.0041097960.000105379
40
0.004824334
Coefficients
Standard
Enor
Stat
P­
value
Intercept
Lower95%
­0.0056162880.003550759­
1.5817147620.121791782­
0.0127983690.001566793­
0.0127983690.0015658
Upper
95%
L
m
r
95.0%
pper95.0%

X
Valiable
I
0.0002982790.0001145482.6039636380.0129706
6.65838Ea5
0.0005299736.65838E­
050.00053
6.66E­
055.30E­
04
Page
58
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Appendix
D:
Selected
Chromatograms
A
representative
set
of
chromatograms
from
the
present
study
is
included
in
this
Appendix.

Page
59
of
71
BACK
TO
MAIN
Calib.
Data
Modified
:

Calcuiate
Based
on
Rel.
Reference
Window
:
Abs.
Reference
Window
:
Rel.
Non­
ref.
Window
:
Abs
.
Non­
ref
.
'
Window
:
MultTplFer
..
Dilution
Sample
Amount
Uncalibrated
Peaks
.'
Partial
Calibration
:
Correct
All
Ret.
Tiines:
*%.

9/
24/
1999
10:
08:
51
AM
Y
5
;
00.0
%
I
0.000
min
1.0000'
1.0000
I
0.00000
I
I
not
reported
Yes,
identified
pe.
aks
are
recalibrated
No,
only
for
identified
peaks
Cunie
Type
Origin
Weight
'
Linear
Included
*
.E
q
~d
.

Recalibration
Settings:
Average
Response
Average
allcalibrations
Average
Retention
Time:
FloatingAverage
New
75%

Calibration
Report
Options
:
Printout
of
recalibrations
within
a
sequence:
Calibration
Table
after
Recalibration
Normal
Report
after
Recalibration
If
the
sequence
is
done
with
bracketing:
Results
of
first
cycle
(ending
previous
bracket)

Signal
1:
.MSD1
427,
EIC=
426.7:
427.7
Signal
2,:
MSDl
499,
EIC=
498.7:
499.7
Signal
3:
MSDl
399,
EIC=
398.7:
399.7
RetTime
Lvl
Amount
,
,
lPPbl
I
­­I
1
11
2'.
22
3
33
4
44
5
55
6
66
7
77
8
Area
I
300.90000
300.90000
401.20000
401.20000
451.35000
451.35000
501.50000
501.50000
6.86155e4
541.62000
6.91006e4
541.62000
6.74844e4
601.80000
7.89434e4
601.80000
7.90989e4
.702.
IOOOO
8.70628e4
702.10000
9.13195e4
802.40000
1.00442e5
­­­­­­­­_­
3.78048e4
3.80593e4
4.74013e4
5.14873@
4
5.60681e4
5.86619e4
6.94725e4
Amt/
Area
Re€
Grp
Name
1
­­­­­­­­­­1
­­­1
­­1
­~,­­­­­­­­­­­­­
7.95930e­
3
.,
PFWS
7.90609e­
3
8.46390e­
3
7.79221e­
3
a.
05003e­
3
7.69409e­
3
7.21869e­
3
7.30884e­
3
7.83813e­
3
.
8.02586e­
3
7.62318e­
3
7.60820e­
3
8.06430e­
3
7.68839e­
3
7.98866e­
3
Page
60
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Method
C:\
HPCHEM\
l\
METHODS\
O922­
5O.
M
RetTime
Lvl
Amount
Area
Amt
/Area
[minl
Sig
[PPbl
­­­­­­­1
­­1
­­1
­­­­­­­­­­~­­­­­­­­­­­­­­­­­­­
88
802.40000
1.00128e5
8.01373e­
3
5.931
1
1
367.92000
4.73070e5
7.77729e­
4
2
367.92000
4.60726eS
­7.98565e­
4
3
367.92000
4.39572e5
8.36996e­
4
~
­~

4
367.92000.
4.90981e5
7.49357e­
4
5
367.92000
4.67449e5
7.87080e­
4
6
367.92000
4.57579e5
8.04058e­
4
7
.367.92000
4.3.5372e5
8.45071e­
4
8
367.92000
4.52504e5
8.13075e­
4
11
367.92000
4.61022e5
7.98052e­
4
22
367.92000
4.61213e5
7.97722e­
4
44
367.92000
4.55268e5
8.08140e­
4
55
367.92000
4.57342e5
8.04475e­
4
6.6­
367,92000
4.52098e5
,8.
'13806e­
4
77
367.92000
4.56825eS
8.05384e­
4
.88
367.92000'4.52021e5
8..
13945e­
4
21
300.90000
9.24706e5
3.25401e­
4
2
401.20000'1.21379e6
3.30534e­
4
22
401.20000
1.22485e6
3.27551e­
4
3..
451.35000
1.32239e6
3.41314e­
4
33
451.35000
1.34445e6
3.35713e­
4
4
501.50000
1.56794e6
3.19846e­
4
44
501.50000
1.51715e6
3.30554e­
4
5
541.62000
1.636.0le6
3.31062e­
4
55
541.62000
1.63668e6
3.30926e­
4
6
601.80000
1.76299e6,
3.41353e­
4
66
601.80000
1.79159e6
3.35902e­
4
7
702.10000
2.03577e6
3.44881e­
4
77
102.10000
2.07740e6
3.37970e­
4
8
802.40000.2.31841e6
3.46099e­
4
8
8
802.40000
2.34731e6
3.41839e­
4
.­
.
.
..­

33367.920004.53927e58.10526e­
4
6.2222
1
300.900009.20859e53.26760e­
4
PFOS.
.
I
=­­­­
­­­­
P=
s==
PIOIE=
P=
10=
I==
I=
3rP=
rPI91=
P=
P=
P=~==~===
x~==================

Peak
Sum
Table
'

'OI=
P=~=
Pe====
rpt=
ll­­­­­===~========~=~=~==~~~===~~­­­­~­­­­­­
­­­­
­­­­­­
0=
13=
r
Page
61
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Method
C:\
HPCHEM\
1\
METHODS\
O922_
50.
M
THPFOS
at
exp.
RT:
5.931
MSDl
427,
EIC=
426.7:
427.7
Correlation:
­0.99395
Residual
Std.
Dev.:
12673.93865
Formula:
y
=
mx
+
b
'

'..
m:
1244.66095
.

b:
­3.54902e­
10
x:
Amount
[ppbl
y:
Area
PFOS
at
exp.
RT:
6.222
MSDl
499,
EIC=
498.7:
499.7
Correlation:
0.99860
Residual
Std.
Dev.:
31589.73759
Formula:
y
=
mx
+
b
m:
2878.88299
b:
51022.56267
x:
Amount
Cppbl
y:
Area.

Page
62
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Batch
Run
#
1
of
60
Data
File
D:\
DATA\
O92399\
PFOSOOOl.
D
Sample
Name:
MeOH
Blank
Sorted
Signal
Calib.
Data
Modified
:
9/
24/
1999
10:
08:
51
AM
Multiplier
,1.0000
.
Dilution
1.0000
Signal
1:
MSDl
427,
EIC=
426
.`
7:
427.7
Totals
:
0.00000
Totals
:
0.00000
Page
63
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
I
Batch
Run
#
1
of
60
Data
File
D:\
DATA\
O92399\
PFOSOOOl.
D
Sample
Name:
MeOH
Blank
Totals
:
0.00000
1
Warnings
or
Errors
:

warning
:
Calibrated
compound(
s1
not
found'

Sorted
By
Signal
Calib.
Data
Modified
:
'
9/
24/
1999
10:
08:
51
AM
.
.
­­
Multiplier
1.0000
Dilution
*'
i;
oooo
Signal
1:
MSDl
427,
EIC=
426
.7:
427.7
Totals
:
0.00000
Name
Totals
:
'
0.00000
Signal
3:
MSDI
399,
EIC=
398,7:
399.7
Peak
RetTime
Type
WidthArea
Area
Name
#.
[minl
[minl
%.
­­­­
I­­;­­­­~­­­­­­
J­­­­­­­
J­­­­­­­­­­
l­­­­­­­­
l­­­­­­­­­­­­­­­­­­­­­
1
5.642
0.0000
0.00000
0.0000
PFHS
Totals
:
'
0.00000
Page
64
of
71
BACK
TO
MAIN
Batch
Run
#
1
qf
60
Data
F
i
l
e
D:\
DATA\
O92399\
PFOSOOOl.
D
1
Warnings
or
Errors
:
Sample
Name:
MeOH
Blank
­+

Page
65
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Batch
Run
.#
6
of
60
I
Data
F
i
l
e
D:\
DATA\
O92399\
PFOSOOO6.
D
Sample
Name:
L4
Std,
501.50
Sorted
By
Signal
Calib.
DataModified
:
9/
24/
1999
10:
08:
51
AM
Multiplier
1.0000
Dilution
1.0000
Signal
1:
MSDl
427,
EIC=
426.7:
427.7
Page
66
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Batch
Run
#
6
of
60
Data
File
D:\
DATA\
O92399\
PFOSOOO6.
D
Sample
Name:
L4
Std,
501.50
Page
67
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Batch
Run
#
18
of
60
Data
F
i
l
e
D:\
DATA\
092399\
PPOSOOl€
i.
D
Sample
Name:
PFOS­
035
Sorted
By
Calib.
Data
Modifi,
ed
:
Multiplier
..
.
Dilution
Signal
I:
MSDl
427,
61C0426
Page
68
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Batch
Run
#
18
of
60
Data
File
D:\
DATA\
O92399\
PFOSOO18.
D
Sample
Name:
PFOS­
035
Page
69
of
71
BACK
TO
MAIN
3M
Environmental
Laboratory
Report
No.
W1878
Batch
Run
#
19
of
60
'

Data
File
D:\
DATA\
O92399\
PFOSOOl9.
D
Sample
Name:
PFOS­
036
'===
13=
IPI==
1====
t=
'DPP­­­
­­­=
s­­­­­­­­,=­­­­­
­­­­­­­­­
­­­­­=­­­­­=­­­=­­­
_­­­­
__­
­­­==
r==
n=====
xs
'InjectionDate
:
9/
23/
1999
2
:0
8
:5
8
PM
Seq.
Line
:
19
Sample
1
:
PFOS­
036
Vial
:
18
Acq.
Operator:
:
MTM
'tnj
:
.I
Acq.
Method
i
:
C:\
HPd\
l\
METHODS\
PFO5­
NG2.
M
Last
changed
j
:
9/
23/
1999
9:
SO:
OO
AM
by
MTM
Analysis
Methbd
:
C:\
HPCHEM\
l\
METHOD5\
0922_
50.
M
Last
changed
:
9/
24/
1999
10:
12:
55
AM
by
MTM
SIM
Analysis
'jES­)
for
THPFOS,
PFOS,
and
PFHS
using
4mmx35mm
Dionex
IonPac
NG1
column,
SlN
12879.

.MTM
*,.
''
InjVolume
:
5
,d
(Results
are
from
a
previously
saved
Batch)

_C­­­­­­

Signal
2
:
MSDl
499,
EIC=
498.7:
499.7
RetTime
Type
Area
Amt/
AreaAmount
G
r
p
,Name
[minl
[PPbl
­­­­­­­I
­­­­­­I
­­­­­­­­­­l
­­­­­­­­­­l
­­­­­­­­­­l
­­~­­­­­­­­­­­­­­­­­­
'

6.231
BB
1.89862e63.38022e­
4641.77689
PFOS
Totals
:
641.77689
Page
70
of
71
BACK
TO
MAIN
Batch
Run
#
19
of
60
Data
F
i
l
e
D:\
DATA\
o9,2399\
PFOSOOl9.
D
S
i
g
n
a
l
3
:
MSDl
399,
EIC=
398.7:
399.7
RetTime
Type
Area
Amt/
Area
h
i
n
l
Totals.
:
3M
Environmental
Laboratory
Report
No.
W1878
Sample
Name:
PFOS­
036
c
680.89256
Page
71
of
71
BACK
TO
MAIN
