3M
Specialty
Materials
3M
Center
St.
Paul,
MN
55
144­
1000
651
733
1110
May
7,
2003
3M
BY
HAND
DELIVERY
Wardner
G.
Penberthy
Director,
Chemical
Control
Division
Office
ofPollution,
Prevention&
Toxics
U.
S.
Environmental
Protection
Agency
Ariel
Rios
Building,
7405M
1200
Pennsylvania
Avenue,
N.
W.
Washington,
D.
C.
20460
Re:
Follow­
up
Questions
Regarding
3M
and
Dyneon
LOl
Environmental
Monitoring
Commitments
Dear
Mr.
Penberthy:

This
letter
responds
to
EPA's
request
in
your
e­
mail
dated
April
30,
2003
for
a
more
detailed
description
ofthe
environmental
monitoring
commitment
contained
in
the
March
13,
2003
3M
Letter
ofIntent
(
LOT).
In
addition,
EPA
has
requested
clarification
ofthe
plans
for
continued
monitoring
ofgroundwater,
surface
water
and
other
environmental
media
outlined
on
page
7,
item
3
of
the
LOT.
This
letter
will
describe:
3M's
environmental
test
methodology,
monitoring
plans
by
plant,
and
summary
calendar
of
monitoring
plans.

As
described
in
section
IT
A
3
in
the
3M
LOT,
3M
has
already
engaged
in
extensive
environmental
monitoring,
including
sampling
in
the
vicinityofits
now
former
PFOA
production
facility
at
Decatur,
AL;
analysis
ofserum
and
liver
samples
from
fish,
birds
and
other
wildlife
for
the
presence
ofPFOA
and
other
constituents;
and
sponsoring
a
Multi­
City
Study
in
which
a
variety
of
environmental
compartments
were
analyzed.
These
data
are
available
through
the
OPPTS
AR­
226
docket.

3M's
environmental
monitoring
commitment
in
the
LOl
is
to
continue
sampling
at
3M
plants
in
Cottage
Grove,
MN
and
Decatur,
AL,
which
were
former
manufacturing
locations
ofPFOA
and
POSF­
based
products.
This
sampling
will
include
ground
water
and
wastewater
monitoring
which
corresponds
with
3M's
voluntary
commitments
and/
or
plans
established
through
permits
with
local
regulators.

The
environmental
monitoring
in
the
vicinity
of
3M's
Decatur,
ALplant
will
include
the
current
operations
ofDyneon,
which,
as
explained
in
the
3M
LOT,
is
a
3M
subsidiary
that
continues
to
use
APFO,
the
ammonium
salt
of
PFOA,
in
the
production
offluoropolymers.
Because
Dyneon's
U.
S.
fluoropolymer
manufacturing
facility
is
located
on
3M's
manufacturing
site,
the
environmental
monitoring
studies
to
satisfy
the
LOT
commitments
will
be
conducted
Wardner
G.
Penberthy
May
7,
2003
Page2
jointly
and
reported
by
3M.
Hence,
the
study
methodology,
sampling
strategy
and
analytical
methods
described
below
are
for
both
3M
and
Dyneon.
1
Data
will
be
submitted
annually
in
final
report
format
by
the
first
ofAugust
of
each
year,
with
the
first
report
in
by
August
1,
2003.
Sampling
will
take
place
monthly,
quarterly,
biannually
and
biennially,
as
described
in
the
text
below.
In
light
of
3M's
phase­
out
of
production
and
significant
reduction
of
use
ofPFOA
in
Decatur,
all
environmental
monitoring
under
the
LOT
is
anticipated
to
terminate
by
December
31,
2006
with
a
final
report
in
by
August
1,
2007.

TEST
METHODOLOGY
AND
VALIDATION:
GROUND
WATER,
SURFACE
WATER,
AND
WASTE
WATEREFFLUENT
Analytical
Test
Method:
3M
ETS­
8­
154.1
"
Determination
of
Perfluorinated
Acids,
Alcohols,
Amides,
and
Sulfonates
In
Water
By
Solid
Phase
Extraction
and
High
Performance
Liquid
Chromatography/
Mass
Spectrometry".
See
Attachment
I.
(
Note
that
this
test
method
supercedes
ETS­
8­
1
15.1
and
155.1,
which
are
referenced
in
historic
test
documents.)

Validation:
This
sampling
and
analysis
method
was
validated
for
the
collection,
extraction,
and
analytical
procedure
for
the
determination
ofPerfluorooctanoate
(
PFOA)
in
groundwater,
surface
water,
and
drinking
water
samples.
Method
accuracy
±
10%.
Method
precision
±
10%.

Method
Summary:
Water
samples
are
collected
from
a
site
ofinterest
and
shipped
cold
to
an
analytical
facility.
PFOA/
APFO
is
extracted
from
4OmL
water
samples
using
Cl
8
solid
phase
extraction
(
SPE)
cartridges.
PFOA/
APFO
is
eluted
from
the
C18
cartridge,
using
methanol.
Separation,
identification,
and
measurement
are
accomplishedby
high­
performance
liquid
chromatography/
tandem
mass
spectrometry
(
HPLC/
MS/
MS)
analysis.
The
concentration
of
each
identified
component
is
measured
by
comparing
the
MS
response
ofthe
quantitation
ion
produced
by
that
compound
to
the
MS
response
ofthe
quantitation
ion
produced
by
the
same
compound
in
an
extracted
calibration
standard
(
external
standard).

TEST
METHODOLOGY
AND
VALIDATION:
FISH
Analytical
Test
Method:
3M
ETS­
8­
23
1.2
"
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices".
SeeAttachmentII.

Validation:
ISO
17025
Accredited
Test
Method,
(
validation
based
on
FDA
guidance
documentation
"
Guidance
for
Industry­
Bioanalytical
Method
Validation,
May,
2001).
This
sampling
and
analysis
method
was
validated
for
the
collection,
extraction,
and
analytical
For
other
clarification
specific
to
Dyneon,
a
letter
from
the
Society
of
Plastics
Institute
(
SPI)
will
address
EPA's
questions
concerning
the
environmental
studies
at
fluoropolymermanuthcturing
sites
and
the
questions
concerning
the
Articles
of
Commerce
Test
Program
in
your
email
to
SPI.
The
current
PFOA
manufacturers
will
address
the
questions
concerning
the
plans
for
envirorimentaimonitoring
at
the
site
of
PFOA
manufacture
in
your
email
to
SPI.
Wardner
G.
Penberthy
May
7,
2003
Page
3
procedure
for
the
determination
ofperfluorooctane
sulfonate
(
PFOS)
in
four
biological
matrices.
The
method
is
a
performance
based
method
requiring
cross­
validation
for
the
biological
matrix
application
of
PFOAIAPFO
in
fish.
Method
accuracy
±
5%.
Method
precision
±
10%.

Method
Summary:
An
amount
of
sample
is
prepared
(
fluids
diluted,
tissues
homogenized,
etc.)
at
a
1/
6
dilution,
or
other
dilution
using
reagent
grade
water.
An
aliquot
ofthe
dilutionlhomogenate
is
spiked
with
the
appropriate
surrogate
or
analyte
mixture.
Acetonitrile
(
ACN)
is
added
as
an
extraction
solvent
and
also
serves
to
precipitate
the
proteins.
The
sample
is
capped,
mixed,
and
put
on
the
centrifuge
to
clarify
the
supernatant.
The
supernatant
is
transferred
to
a
clean
tube,
diluted
with
water,
and
passed
through
a
pre­
conditioned
C18
SPE
cartridge.
Finally,
the
analytes
ofinterest
are
eluted
from
the
SPE
cartridge
and
analyzed
by
high
performance
liquid
chromatography­
electrospray
tandem
mass
spectrometry
(
HPLCES
MS/
MS).

Cottage
Grove,
MN
Groundwater
Monitoring
To
date,
numerous
groundwater
studies
have
been
completed
at
the
site
in
support
ofvarious
regulatory
programs.
These
studies
have
documented
that
a
shallow
unconfined
groundwater
system
is
the
first
aquifer
beneath
the
site.
A
lower
confining
bed
underlies
the
unconfined
aquifer.
Natural
groundwater
flow
in
the
shallow
system
is
generally
from
north
to
south
and
discharges
to
the
Mississippi
River
along
the
southern
boundary
ofthe
site.
These
wells
contain
the
groundwater
in
the
area
ofthe
plant
site
and
would
not
represent
conditions
beyond
the
site
boundaries.

These
studies
focused
on
conventional
substances
and
have
incorporated
fluorochemical
analytes
only
recently.
These
test
results
will
be
included
in
the
upcoming
August,
2003
report
to
EPA
under
the
LOT.

Going
forward,
five
(
5)
existing
on­
site
monitoring
wells
will
be
sampled
and
analyzed
for
PFOA.
The
selection
ofthese
wells
was
based
on
the
site
hydrogeologic
model
and
will
be
representative
ofthe
site
ground
water
conditions,
i.
e.,
both
source
and
non­
source
areas.
These
wells
will
be
sampled
on
a
semiannual
basis
during
the
month
ofMayand
September
commencing
in
2003.

Wastewater
Effluent
Monitoring
The
site
has
a
multi­
phased
wastewater
treatment
plant
that
is
used
to
treat
all
process
wastewaters
generated
at
Cottage
Grove.
Two
ofthe
systems
treat
inorganic
wastewaters
and
the
third
is
an
organic,
biological
treatment
system.
All
ofthe
treated
process
wastewaters
from
these
operations
are
combined
at
a
single
discharge
point~
These
wastewaters
are
then
combined
with
non­
contact
cooling
and
storm
water
and
then
discharged
to
the
Mississippi
River.
Wardner
G.
Penberthy
May
7,
2003
Page
4
Treated
process
wastewater
and
non­
contact
cooling/
storm
water
are
discharged
under
Minnesota
NPDES
Permit
No.
MN000149.
The
designations
in
the
permit
for
the
treated
process
wastewater,
non­
contact
cooling/
storm
water,
and
the
combination
ofthese
two
wastewaters
are
listed
in
the
permit
as
Outfalls
SDOO1,
SDOO2,
and
SDOO3,
respectively.
In
accordance
with
Section
1.12
ofNPDES
Permit,
3M
is
currently
conducting
monthly
sampling
and
analysis
at
Outfall
SDOO1
for
certain
perfluorinated
chemicals,
including
PFOA.

A
24­
hour
composite
sample
is
collected
for
duplicate
analysis
during
each
sampling
event.
In
accordance
with
Section
1.12
par.
5
ofthe
permit,
this
monthly
analysis
at
Outfall
SDOO1
will
be
conducted
through
2004.
The
NPDES
Permit
defines
a
process
to
assess
the
results
and
allow
revision
or
reduction
in
the
frequencyofmonitoring.

Decatur,
AL
Groundwater
Monitoring
To
date,
numerous
groundwater
studies
have
been
completed
at
the
site
in
support
ofvarious
regulatory
programs.
The
natural
groundwater
flow
in
the
shallow
system
is
generally
from
the
southwest
to
the
northeast
and
discharges
to
the
Tennessee
River
along
the
boundary
of
the
site.
These
investigations
have
involved
an
extensive
characterization
ofsite
subsurface
conditions.
Most
recently,
we
conducted
a
dye
trace
study
that
validated
the
earlier
hydrogeological
findings
and
our
site
conceptual
model.

These
studies
have
incorporated
fluorochemical
analytes
only
recently.
Limited
analyses
addressing
PFOA
have
been
conducted
and
results
will
be
included
in
the
upcoming
August,
2003
report
under
the
LOI.

Going
forward
eight
(
8)
existing
on­
site
monitoring
wells
will
be
sampled
and
analyzed
for
PFOA.
The
selection
ofthese
wells
was
based
on
the
site
hydrogeologic
model
and
will
be
representative
of
the
site
ground
water
conditions,
i.
e.,
both
source
and
non­
source
areas.
Groundwater
samples
for
PFOA
analyses
will
be
collected
twice
a
year,
during
the
June
and
December
sampling
events.

Wastewater
Effluent
Monitoring
3M's
Decatur,
Alabama
manufacturing
facilities
obtain
process
water
from
the
City
of
Decatur
Utilities.
In
addition
many
ofthe
manufacturing
operations
utilize
non­
contact
cooling
water,
which
is
obtained
from
the
Tennessee
River.
All
process
wastewaters
are
treated
in
the
site's
wastewater
treatment
facility.
The
system
contains
both
physical­
chemical
and
biological
treatment.
Process
wastewaters
are
mixed
with
non­
contact
cooling
water
prior
to
discharge
to
the
Tennessee
River.
The
discharge
is
permitted
under
Alabama
NPDES
Permit
No.
AL0000205.
The
process
wastewater
discharge
and
combined
process
wastewater/
noncontact
cooling
water
is
designated
as
Outfall
001A
and
001,
respectively.
Wardner
G.
Penberthy
May
7,
2003
Page
5
Sampling
will
be
conducted
quarterly.
Sampling
will
be
conducted
at
Outfall
001
and
will
consist
of24­
hour
composites
with
duplicate
analysis
being
conducted
on
the
collected
sample.

Surface
Waters,
Sediments,
and
Fish
3M
contracted
two
previous
studies
which
characterize
the
concentrations
of
PFOA
and
other
analytes
in
surface
waters,
sediment,
and
fish
in
the
Tennessee
River.
As
part
ofthese
efforts,
representative
samples
ofsurface
water
and
sediment
were
obtained
from
locations
that
are
upstream,
downstream,
and
in
the
vicinity
of
the
plant
wastewater
outfall.
Testing
was
also
conducted
on
a
number
offish
and
clams
from
both
upstream
and
downstream
locations.
The
first
study
was
conducted
in
2000
and
has
been
submitted
to
the
EPA
(
AR­
226).
This
study
report
contains
detailed
descriptions
of
sampling
locations,
sampling
and
analytical
methodologies
as
well
as
data
tabulations
and
results
summaries.
The
second
study
was
completed
recently
and
this
information
will
be
submitted
to
AR­
226
in
the
near
future.

It
is
3M's
intention
to
conduct
additional
sampling
to
augment
the
previous
studies.
These
studies
will
be
conducted
once
every
other
year
with
the
next
set
ofsamples
being
collected
in
2004.
A
project
plan
for
future
work
will
be
prepared
in
the
1st
or
2nd
quarters
of2004.
As
a
part
ofthis
work,
samples
will
be
obtained
upstream,
downstream,
and
adjacent
to
3M's
manufacturing
facilities.

Water
samples
will
be
collected
using
standard
protocols.
Sediments
samples
will
be
collected
with
the
use
ofa
PONAR
dredge
using
the
established
methodologies.
A
boat
mounted
Smith­
Root
type
VT­
A
electro
shocking
device
will
be
utilized
for
fish
collection.
Detailed
descriptions
ofthese
sampling
protocols
and
sampling
locations
will
be
contained
in
the
work
plan.

Monitoring
Schedule
Below
is
a
schedule
ofthe
current
monitoring
plans
under
the
3M
LOT.

Activity
2003
2004
2005
2006
2007
2008
Cottage
Grove
ground
water
biannual
biannual
biannual
biannual
biannual
Cottage
Grove
effluent
monthly
monthly
quarterly
quarterly
quarterly
Decatur
ground
water
biannual
biannual
biannual
biannual
biannual
Decatur
effluent
quarterly
quarterly
quarterly
quarterly
quarterly
Decatur
surface
water,
sediment
&
fish
once
once
Report
submission
by
Aug.
1
by
Aug.
1
by
Aug.
1
by
Aug.
1
by
Aug.
1
by
Aug.
1
Wardner
G.
Penberthy
May
7,2003
Page
6
3M
appreciates
this
opportunity
to
work
with
the
EPA
on
this
LOT
and
on
our
prior,
ongoing
and
future
EHS
measures
for
PFOA.
We
look
forward
to
dialogue
with
the
EPA
regarding
the
content
ofthis
letter.
In
addition,
3M
plans
to
review
the
environmental
monitoring
data
submitted
under
the
LOT
with
EPA
after
two
years
to
determine
whether
any
changes
to
this
overall
monitoring
program
would
be
appropriate
based
on
the
data.
We
anticipate
that
this
review
would
include
discussion
ofmedium,
frequency
ofsampling
and
testing
and
location
ofsampling.

Sincerely,
Please
contact
one
ofus
if
you
have
any
questions.

Michael
A.
Santoro,
Director
Specialty
Materials
EHS&
R
3M
Company
651­
733­
6374
Enclosure
DC\
592269.1
Dr.
George
H.
Millet,
Director
QEHS
Dyneon,
a
3M
Subsidiary
651­
733­
5637
3M
Environmental
Laboratory
Method
Determination
of
Perf!
uorinatedAcids,
Alcohols,
Amides,
and
Sulfonates
In
Water
By
Solid
Phase
Extraction
and
High
Performance
Liquid
Chromatography/
Mass
Spectrometry
Method
Number:
ETS­
8­
154.1
Adoption
Date:
28
Apr
2000
Revision
Date:
5
May,
2003
Effective
Date:
5
May,
2003
William
K.
Reagen
Manager
c2S/~~>`~
3
Date
ETS­
8­
1
54.1
PaQe
I
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
Approved
By:
I
Scope
and
Application
This
method
was
validated
for
the
collection,
extraction,
and
analytical
procedure
for
the
determination
of
Pertluorooctane
sulfonate
(
PFOS),
Perfluorooctane
sulfonylamide
(
FOSA),
and
Perfluorooctanoate
(
PFOA)
in
groundwater,
surface
water,
and
drinking
water
samples.
This
method
may
also
be
applied
to
the
determination
of
other
periluorinated
acids,
alcohols,
amides,
and
sulfonates
in
similar
matrices,
as
long
as
the
defined
QC
elements
are
satisfied
and
with
the
understanding
that
the
method
is
not
validated
for
compounds
outside­
thescope
of
the
original
protocol
This
method
is
based
in
part
on
the
report
"
Method
of
Analysisfor
the
Determination
of
Perfluorooctane
sulfonate
(
PFOS),
Perfluorooctane
sulfonylamide
(
PFOSA),
and
Pertluorooctanoate
(
POAA)
in
Water"
(
see
Section
17),
as
developed
and
validated
by
Exygen
Research
(
formerly
Centre
Analytical
Laboratories,
Inc.).

2
Method
Summary
Water
samples
are
collected
from
a
site
of
interest
and
shipped
cold
to
an
analyticalfacility.
Perfluorinated
acids,
alcohols,
amides,
and
sulfonates
are
extracted
from
4OmL
watersamples
using
C18
solid
phase
extraction
(
SPE)
cartridges.
The
compounds
are
eluted
from
the
C18
cartridge,
using
methanol.
Separation,
identification,
and
measurement
are
accomplished
by
highperformance
liquid
chromatography/
tandem
mass
spectrometry
(
HPLC/
MS/
MS)
analysis.
Highperformance
liquid
chromatography/
mass
spectrometry
(
HPLC/
MS)
maybe
used­
ifthe
defined
QC
elements
are
satisfied.

The
concentration
of
each
identified
component
is
measured
by
comparing
the
MS
response
of
the
quantitation
ion
produced
by
that
compound
to
the
MS
response
of
the
quantitation
ion
produced
by
the
same
compound
in
an
extracted
calibration
standard
(
external
standard).

3
Definitions
3.1
Analytical
Sample
Aportion
of
an
extracted
Laboratory
Sample
prepared
for
analysis.

3.2
Calibration
Standard
A
solution
prepared
from
the
Working
Standard
(
WS)
and
extracted
according
to
this
method.
The
calibration
standard
solutions
are
used
to
calibrate
the
instrument
response
with
respect
to
analyte
concentration.

3.3
Duplicate
Sample
(
DS)

A
DS
is
a
separate
aliquot
of
a
sample,
taken
in
the
analytical
laboratory
that
is
extracted
and
analyzed
separately
with
identical
procedures.
Analysis
of
DSscompared
to
that
of
the
first
aliquot
give
a
measure
ofthe
precision
associated
with
laboratory
procedures,
but
not
with
sample
collection,
preservation,
or
storage
procedures.

3.4
Field
Blank
Control
Sample
(
FB)

ASTM
Type
I
water
placed
in
a
sample
container
in
the
laboratory
and
treated
as
a
sample
in
all
respects,
including
exposure
to
sampling
site
conditions,
storage,
preservation
and
all
analytical
procedures.
The
purpose
of
the
FB
is
to
determine
if
test
substances
or
other
interferences
are
present
in
the
field
environment.

ETS­
8­
154.1
Page
2
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
3.5
Field
Duplicate
(
FD)
A
sample
collected
in
duplicate
at
the
sametime
as
the
sample
and
placed
under
identical
circumstances
and
treated
exactly
the
samethroughout
field
and
laboratory
procedures.
Analysis
of
ED
compared
to
that
of
the
first
sample
gives
a
measure
of
the
precision
associated
with
sample
collection,
preservation
and
storage,
as
well
as
with
laboratory
procedures.

3.6
Field
Matrix
Spike
(
FMS)

A
sample
collected
in
duplicate
to
which
known
quantities
of
the
target
analytes
are
added
in
the
field
at
the
time
of
sample
collection.
Alternatively,
the
known
quantity
of
target
analytes
may
be
added
to
the
sample
bottle
in
the
laboratory
before
the
bottles
are
sent
to
the
field.
A
known,
specific
volume
of
sample
must
be
added
to
sample
container
without
rinsing.
This
may
be
accomplished
by
making
a
"
fill
to
this
level"
line
on
the
outside
ofthe
sample
container.
The
FMS
should
be
spiked
at
approximately
50
 
150%
of
the
expected
analyte
concentration
in
the
sample.
If
the
expected
range
of
analyte
concentrations
is
unknown,
a
low
and
a
high
spike
may
be
prepared
to
increase
the
likelihood
that
a
spike
at
an
appropriate
range
is
made.
The
FMS
is
analyzed
to
ascertain
if
any
matrix
effects,
interferences,
or
stability
issues
may
complicate
the
interpretation
of
the
sample
analysis.

3.7
Field
Spike
Control
Sample
(
FSCS)

An
aliquot
of
ASTM
Type
I
water
to
which
known
quantities
of
the
target
analytes
are
added
in
the
field
at
the
time
of
sample
collection
(
at
an
appropriate
concentration
to
be
determined
by
the
project
lead)
or
in
the
laboratory
prior
to
the
shipment
of
the
collection
bottles,
The
FSCS
is
extracted
and
analyzed
exactly
like
a
sample
to
determine
whether
a
loss
of
analyte
could
be
attributed
to
sample
storage
and/
or
shipment.
A
low
and
high
FSCS
may
be
appropriate
when
expected
sample
concentrations
are
not
known.

3.8
Laboratory
Control
Sample
(
LCS)

An
aliquot
of
ASTM
Type
I
water
to
which
known
quantities
of
the
target
analytes
areadded
in
the
laboratory.
Two
levels
are
included,
one
at
the
LLOQ
(
approx.
25
pg/
mL),
the
other
at
a
concentration
of
approx.
100
 
250
pg/
mL
or
another
concentration
to
be
determined
by
the
project
lead.
The
LCS
is
extracted
and
analyzed
exactly
like
a
laboratory
sample
to
determine
whether
the
methodology
is
in
control,
and
whether
the
laboratory
is
capable
of
making
accurate
measurements
at
the
required
method
detection
limit
and
higher.

3.9
Laboratory
Sample
A
portion
of
a
sample
received
from
the
field
for
testing.

3.10
Limit
of
Detection
(
LOD)

The
LOD
is
the
lowest
concentration
ofan
analyte
that
can
be
measured
and
reported
with
99%
confidence
that
the
analyte
concentration
is
greater
than
zero.
If
required,
the
LOD
may
be
determined
in
several
ways,
including
signal­
to­
noise
ratio
and
statistical
calculations.

3.11
Limit
of
Quantitation
(
LOQ)

The
LOQ
for
a
dataset
is
the
lowest
concentration
(
LLOQ)
or
highest
concentration
(
ULOQ)
that
can
be
reliably
achieved
within
the
specified
limits
of
precision
and
accuracy
during
routine
operating
conditions.

Note:
For
many
analytes,
the
LLOQ
analyte
concentration
is
selected
as
the
lowest
non­
zero
standard
in
the
calibration
curveto
simplify
data
reporting.
Sample
LLOQs
are
matrix­
dependent
3.12
Matrix
Spike
(
MS)

ETS­
8­
1
54.1
Page
3
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
A
matrix
spike
is
an
aliquot
of
a
sample,
to
which
known
quantities
of
target
analytes
are
added
in
the
laboratory.
The
MS
is
extracted
and
analyzed
exactly
like
a
laboratory
sample
to
determine
whether
the
sample
matrix
contributes
bias
to
the
analytical
results.
Thebackground
concentrations
of
the
analytes
in
the
sample
matrixmust
be
determined
in
a
separate
aliquot
and
the
measured
values
in
the
MS
corrected
for
background
concentrations.

3.13
Method
Blank
An
aliquot
of
ASTM
Type
I
water
that
is
treated
exactly
like
a
laboratory
sample
including
exposure
to
all
glassware,
equipment,
solvents,
and
reagents
that
are
usedwith
other
laboratory
samples.
The
method
blank
is
used
to
determine
if
test
substances
or
other
interferences
are
present
in
the
laboratory
environment,
the
reagents,
or
the
apparatus.

3.14
Method
Detection
Limit
(
MDL)
Determination
A
MDL
is
the
statistically
calculated
minimum
amount
of
an
analyte
that
can
be
measured
with
99%
confidence
that
the
reported
value
is
greater
than
zero.
One
of
several
processes
that
may
be
used
to
establish
a
LOD
value
is
found
in
40
CFR
Part
136
Appendix
B.

3.15
Sample
A
sample
is
a
small
portion
collected
from
a
larger
quantity
of
material
intended
to
represent
the
original
source
material.

3.16
Spiking
Stock
Standard
(
SSS)

Asolution
prepared
from
stock
standards
used
to
prepare
the
working
standard.

3.17
Stock
Standard
(
SS)
A
concentrated
solution
of
a
single
analyte
prepared
in
the
laboratory
with
an
assayed
reference
compound.
3.18
Working
Standard
(
WS)

Asolution
of
several
analytes
prepared
in
the
laboratory
from
SSs­
and­
diluted
as
needed
to
prepare
calibration
standards
and
other
required
analyte
solutions.

4
Warnings
and
Cautions
4.1
Health
and
Safety
The
acute
and
chronic
toxicity
of
the
standards
for
this
method
have
not
been
precisely
determined:
however,
each
should
be
treated
as
a
potential
health
hazard.

Unknown
samples
may
contain
high
concentrations
of
volatile
toxic
ct~
nipounds.
Sample
containers
should
be
opened
in
a
hood
and
handled
with
gloves
to
prevent
exposure.

Thelaboratory
is
responsible
for
maintaining
a
safe
work
environment
and
a
current
awareness
of
local
regulations
regarding
the
handling
of
the
chemicals
used
in
this
method.
A
reference
file
of
material
safetydata
sheets
(
MSDS)
should
be
available
to
all
personnel
involved
in
these
analyses.

4.2
Cautions
None
5
Interferences
ETS­
8­
154.1
Page
4
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
M~.
During
extraction
and
analysis,
major
potential
contaminant
sources
are
reagents
and
solid
phase
extraction
devices.

All
materials
used
in
the
analyses
shall
be
demonstrated
to
be
free
from
interferences
under
conditions
of
analysis
by
running
method
blanks.
Parts
and
supplies
that
contain
Teflon
®
should
be
avoided
due
to
the
possibility
of
interference
and/
or
contamination.
These
may
include,
but
are
not
limited
to:
wash
bottles,
Teflon
®
lined
caps,
autovial
caps,
HPLC
parts,
etc.

The
use
of
disposable
micropipettes
or
pipettes
to
aliquot
standard
solutions
is
recommended
to
make
calibration
standards
and
matrix
spikes.

6
Instrumentation,
Supplies,
and
Materials
Note:
Brand
names,
suppliers,
and
part
numbers
are
for
illustrative
purposes
only.
Equivalent
performance
may
be
achieved
using
apparatus
and
materials
other
than
those
specified
here,
but
demonstration
of
equivalent
performance
that
meets
the
requirements
of
this
method
is
the
responsibility
of
the
laboratory
performing
the
analysis.

6.1
Instrumentation
Balance,
analytical
(
display
at
least
0.000lg),
Mettler
HPLC/
MS/
MS
orHPLC/
MS
system,
as
described
in
Section
10.

6.2
Supplies
and
Materials.

Sample
collection
bottles
 
LOPE
(
e.
g.,
NalgeneTM)
narrow­
mouth
bottles
with
screw
cap.
Note:
Do
not
use
Teflon
bottles
or
Teflon
lined
caps.

Coolers
for
sample
shipment.

Ice
for
sample
shipment
Vacuum
pump,
BOchi.

Visiprep
vacuum
manifold,
Supelco.

Sep
PakVac6cc
(
lg)
tCl8
cartridges
(
part#
WAT
036795),
Waters.
5OmLdisposable
polypropylene
centrifuge
tubes,
\/
WR.

I
5mL
disposable
polypropylene
centrifuge
tubes,
VWR.

Disposable
micropipettes
(
50
 
I
OOiJL,
100
 
200pL),
Drummond.

Class
Apipettes
and
volumetric
flasks,
various.

Hypercarb
drop­
in
guard
column
(
4mm)
(
part#
844017
 
400),
Keystone.

Stand­
alone
drop­
in
guard
cartridge
holder,
Keystone.

125mL
LDPE
narrow­
mouth
bottles,
Nalgene.

2mL
clear
HPLC
vial
kit
(
cat#
5181
 
3400),
Agilent/
Hewlett
Packard.

Standard
lab
equipment
(
graduated
cylinders,
disposable
tubes,
etc.),
various.

7
Reagents
and
Standards
Note:
Suppliers
and
catalog
numbers
are
for
illustrative
purposes
only.
Equivalent
performance
may
be
achieved
using
chemicals
obtained
from
other
suppliers.
Do
not
use
a
lesser
grade
of
chemical
than
those
listed.

ETS­
8­
154.1
Page
5
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
7.1
Chemicals
Methanol
(
MeOH),
HPLC
grade,
JT
Baker,
Catalog
No.
JT9093
 
2.

Ammonium
Acetate,
Reagent
grade,
Sigma­
Aldrich,
Catalog
No.
A
 
7330.

ASTM
Type
I
Water,
prepared
in­
house.

Sodium
Thiosulfate,
Reagent
grade,
JT
Baker.

7.2
Standards
Potassium
perfluorooctane
sulfonate
Perfluorooctane
sulfonylamide
Ammonium
perfluorooctanoate
Others
as
required.

7.3
Reagent
Preparation
250mg/
mL
sodium
thiosulfate
solution
 
Dissolve
25g
of
sodium
thiosulfate
in
I
OOmL
reagent
water.

40%
methanol
wash
solution
 
Measure400mL
methanol
and
adjust
volume
to
I
.
OL
with
reagent
water.

100mM
ammonium
acetate
solution
(
Analysis)
 
Weigh
7.71g
of
ammonium
acetate
and
dissolve
in
1
.
OL
of
reagent
water.
Dilute
the
100mM
solution
by
a
factor
of
50
to
makethe
2mM
ammonium
acetate
solution
used
for
mobile
phase
A.

Note:
Alternative
volumes
may
be
prepared
as
long
as
the
ratios
of
the
solvent
to
solute
ratios
are
maintained.

7.4
Spiking
Stock
Standard
(
SSS)
Preparation
Thefollowing
standard
preparation
procedure
serves
as
an
example
and
maybe
changed
to
suit
the
needs
of
a
particular
study.
For
example,
pL
volumes
may
be
spiked
into
volumetric
flasks
when
diluting
stock
solutions
to
appropriate
levels.

lOOpg/
mL
each
PFOS,
PFOSA,
and
POAASSSs
 
Weigh
out
10mg
of
analytical
standard
(
corrected
for
percent
salt
and
purity
 
i.
e.,
10
mg
C8F17S03K
purity
90%
=
8.35mg
C8F17S03)
and
dilute
to
I
OOmL
with
methanol
in
a
I
OOmL
volumetric
flask.
Transfer
to
a
I
25mL
LDPE
bottle
or
other
suitable
container.
Prepare
a
separate
solution
for
each
analyte.
Solutions
may
be
stored
in
a
refrigerator
at4
°
±
2
°
C
for
a
maximum
period
of
6
months
from
the
date
of
preparation.

lpg!
mL
mixed
SSS
 
Add
1.
OmL
each
of
the
lOOpg/
mL
SSSs
(
from
7.4.1)
to
a
lOOmL
volumetric
flask
and
bring
up
to
volume
with
methanol.
O.
lpg/
mL
mixed
SSS
 
Add
10.
OmL
of
the
1.
Opg/
mL­
mixed
solution
(
from
7.4.2)
to
a
lOOmL
volumetric
flask
and
bring
up
to
volume
with
methanol.
O.
OlpgImL
mixed
SSS
 
Add
10.
OmL
of
the
0.
lpg/
mL­
mixed
solution
(
from
7.4.3)
to
a
lOOmL
volumetric
flask
and
bring
up
to
volume
with
methanol.
Storage
Conditions
 
Store
all
SSSs
in
a
refrigerator
at4
°
±
2
°
C
for
a
maximum
period
of
6
months
from
the
date
of
preparation.

7.5
Calibration
Standards
The
following
standard
preparation
procedure
serves
as
an
example
and
may
be
changed
to
suit
the
needs
of
a
particular
study,
provided
the
concentrations
are
calculated
correctly.

ETS­
8­
1
54.1
Page
6
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
1OO~
Jg/
mLeach
PFOS,
PFOSA,
and
POAA
stock
standard
solutions
 
Weigh
out
10mg
of
analytical
standard
(
corrected
for
percent
salt
and
purity)
and
dilute
to
1
OOmL
with
methanol
in
a
IOOmL
volumetric
flask.
Transfer
to
a
125mL
LOPE
bottle
or
other
suitable
container.
Prepare
a
separate
solution
for
each
analyte.
Store
solutions
in
a
refrigeratorat
4
°
±
2
°
C
for
a
maximum
period
of
6
months
from
the
date
of
preparation.

lpgimL
Working
Standard
 
Add
1.
OmL
each
of
the
IOOpg/
mLSS
solutions
(
from
7.5.1)
to
a
lOOmL
volumetric
flask
and
bring
up
to
volume
with
methanol.

O.
lpg/
mL
Working
Standard
 
Add
10.
OmL
of
the
1
.
Opg/
mL
mixed
solution
(
from
7.5.2)
to
a
IOOmL
volumetric
flask
and
bring
up
to
volume
with
methanol.

O.
OlpgImL
Working
Standard
 
Add
l0.
OmL
of
the
0.
lpg/
mL
mixed
solution
(
from
7.5.3)
to
a
I
OOmLvolumetric
flask
and
bring
up
to
volume
with
methanol.

Storage
Conditions
 
Store
all
WSs
in
a
refrigerator
at4
°
±
2
°
C
for
a
maximum
period
of
6
months
from
the
date
of
preparation.
Calibration
Standard
 
Prepare
calibration
solutions
in
ASTM
Type
I
using
the
following
table
as
a
guideline:

Final
Concentration
of
Final
Calibration
Calibration
Standard,
Concentration
Volume
of
Standard
Volume,
mL
,
pglmL,
in
ASTM
Type
I
of
WS,
pg/
mL
WS,
pL
ofASTM
Type
I
Water
Water
0.0
0
40
0
0.010
100
40
25
0.010
200
40
50
0.010
400
40
100
0.10
100
40
250
0.10
200
40
500
0.10
300
40
750
0.10
400
40
1000
1.0
100
40
2500
1.0
400
40
10000
1.0
1000
40
25000
Thestandards
are
processed
through
the
extraction
procedure
(
Section
11),
identical
to
the
laboratory
samples.
The
concentration
of
the
calibration
standard
in
the
final
extract
is
equal
to
8X
the
initial
concentration,
due
to
the
concentration
of
the
standard
during
the
extraction
process.

Storage
Conditions­
Store
all
extracted
calibration
standards
in
l5mL
polypropylene
tubes
at
4
°
±
2
°
C,
for
a
maximum
period
of
two
weeks
from
the
date
ofpreparation
8
Sample
Collection
and
Handling
Note:
Sampling
equipment,
including
automatic
samplers,
must
be
free
of
Teflon
tubing,
gaskets,
and
other
parts
that
may
leach
interfering
analytes
into
the
water
sample.
Automatic
samplers
that
composite
samples
over
time
should
use
refrigerated
polypropylene
samllle
containers
if
possible.
Sample
bottles
should
not
be
rinsed
before
sample
collection.

Labeling:
Each
sample
bottle
must
display
information
regarding
the
collection
of
that
sample,
the
individual
collecting
the
sample,
and
any
matrix
spike
that
has
been
added
to
the
sample.

ETS­
8­
l
54.1
Page
7
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
This
includes
the
volume
and
concentration
of
any
spiking
solution
added
and
the
volume
and
identification
of
any
preservatives
added
in
the
field.

Spiking:
The
spiking
scheme
will
be
clearly
outlined
in
the
sampling
plan,
including
whetherthe
samples
will
be
spiked
in
the
field
or
in
the
laboratory
prior
to
the­
shipment
of
the
bottles­
to­
the
site.
If
spiking
is
to
be
performed
in
the
field,
materials
and
specific
instructions
will
be
included
in
the
sampling
kit.
Be
sure
to
clearly
label
each
bottle
with
spiking
information
if
applicable.

TapWater:
Open
the
tap
and
allow
the
system
to
flush
until
the
watertemperature
(
15
°
±
l0
°
C)
has
stabilized
(
usually
about
two
minutes).
Adjust
the
flow
to
about
500mL/
min
and
collect
samplesfrom
the
flowing
stream.

Ground
Water:
Purge
the
well
of
standing
waterusing
a
pump
~
ra­
baiFer.
Collect
the
sample
directly
from
the
pump
or
from
the
bailer.

Surface
Water:
When
sampling
from
an
open
body
ofwater,
fill
the
sample
container
with
water
from
a
representative
area.

Sample
Dechlorination:
All
samplesshould
be
iced
or
refrigerated
at4
°
±
2
°
C
and
kept
in
the
dark
from
the
time
of
collection
until
extraction.
Residual
chlorine
should
be
eliminated
by
adding
200pL
of
a
250mg/
mL
sodium
thiosulfate
solution
to
each
tap­
water
sample
and
associated
FB
and
FSCS
(
which
may
be
placed
in
each
bottle
before
leaving
for
the
sampling
site
or
done
in
the
field.).

Holding
Time
(
HI):
Results
of
the
time/
storage
study
of
all
target
analytes
showed
that
the
three
compounds
are
stable
for
14
days
in
water
sampleswhen
the
samples
are
dechlorinated
and
stored
as
described
in
the
previous
section
(
see
also
references
in
sectioii
17).
Therefore,
laboratory
samples
must
be
extracted
within
14
days
and
the
extracts
analyzed
within
30
days
of
sample
collection.
Ifthe
HT
exceeds
14
days,
great
care
is
used
when
evaluating
field
spikes
to
avoid
misrepresentation
of
the
sample
concentration.

8.1
Field
Blanks
Process
a
Field
Blank
Control
Sample
(
FB)
along
with
each
sample
set
(
samples
collected
from
the
same
general
sample
site
at
approximately
the
sametime).
At
the
laboratory,
prior
to
sample
collection,
fill
a
sample
container
with
ASTM
Type
I
water,
seal,
and
ship
the
FB
to
the
sampling
site
along
with
the
empty
sample
containers.
Return
the
FB
to
the
laboratory
with
the
filled
sample
bottles.

When
sodium
thiosulfate
is
added
to
samples,
use
the
sameprocedure
to
preserve
the
FB.

8.2
Field
Duplicates
Collect
a
Field
Duplicate
(
FD)
for
every
ten
(
10)
samples
collected
or
per
each
sampling
set,
if
less
than
10
samplesare
collected.

Separate
FDs
must
be
collected
for
each
type
of
water
sample
(
ground,
tap,
etc.)
collected.

Collect
the
FD
immediately
after
the
sample.

Preserve,
store
and
ship
FD
using
the
sameprocedures
as
used
for
the
samples.

8.3
Field
Spike
Control
Sample
(
FSCS)

A
Field
Spike
Control
Sample
(
FSCS)
mustbe
prepared
for
eachsample
shipment
If
multiple
coolers
are
used
to
ship
a
set
of
samples,
each
coolermust
contain
a
FSCS.

At
the
laboratory,
fill
a
sample
container
with
lOOmL
of
ASTM
Type
I
water.
Seal
and
ship
to
the
sampling
site
along
with
the
emptysample
containers
and
FBs.
Samples
may
either
be
spiked
in
the
field
or
in
the
laboratory
prior
to
shipment.
The
method
employed
should
be
consistent
throughout
the
study.
If
the
samples
are
to
be
spiked
in
the
field,
be
sure
to
send
appropriate
supplies
and
instructions
for
the
field
personnel
to
follow.

ETS­
8­
1
54.1
Page
8
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LCIMS.
Seal
and
gently
invert
the
FSCS
to
mix.
Store
and
ship
the
FSCS
using
the
same
procedures
as
used
for
the
samples
Provide
information
on
sample
collection,
preservation,
shipment
and
storage.
List­
applicable
holding
times.
Include
sample
stability
and
extract
storage
requirements.
Reference
the
method
used
for
sample
preparation,
if
applicable.

8.4
Field
Matrix
Spike
(
FMS)

A
Field
Matrix
Spike
(
EMS)
mustbe
prepared
for
each
sampling
location.
One
unspiked
sample
from
the
same
location
mustaccompany
the
EMS
to
determine
endogenous
levels
in
the
sample.
The
samples
should
be
clearly
identifiable
as
being
from
the
same
location.

Samples
may
either
be
spiked
in
the
field
or
in
the
laboratory
prior
to
shipment
The
method
employed
should
be
consistent
throughout
the
study.
Ifthe
samples
are
to
be
spiked
in
the
field,
be
sure
to
send
appropriate
supplies
and
instructions
for
the
field
personne]~
tofollow.

9
Quality
Control
and
Data
Quality
Objectives
Analytical
results
of
the
FB,
FMS,
FD,
and
FSCS
should
be
evaluated
at
the
conclusion
of
the
study
to
help
interpret
the
quality
of
sample
data.
Analytical
results
for
these
control/
duplicate
samples
must
be
reported
with
the
sample
data.

9.1
Solvent
Blanks
Solvent
blanks
are
analyzed
with
each
sample
set
to
determine
contamination
or
carryover.
Aliquots
of
methanol
represent
the
solvent
used
for
the
standard
curve
and­
the­
sample
extraction.
Solvent
blanks
should
have
area
counts
that
are
less
than
50%
ofthe
areacount
of
the
lowest
calibration
standard.

Solvent
blanks
should
be
analyzed
prior
to
and
following
each
calibration
curve,
each
set
of
system
suitability
samples,
and
after
no
more
than
10
unknown
sample
extracts.
If
instrument
carryover
is
a
problem
consecutive
solvent
blanks
may
be
necessary.
In
this
case
the
area
counts
ofthe
solvent
blanks
should
return
to
<
50%
of
the
lowest
calibration
standard
prior
to
the
injection
of
further
standards
or
samples.

9.2
Method
Blanks
A
method
blank
consists
of
an
aliquot
of
ASTM
Type
I
water,
equal
in
volume
to
the
samples,
and
extracted
in
the
same
manner
as
the
samples.
At
least
two
method
blanks
should
be
prepared
and
analyzed
each
day
that
extractions
are
performed
for
a
particular
study
or
project.
When
analyzed
the
area
counts
of
thesesamples
must
be
less
than
50%
of
the
area
count
of
the
lowest
calibration
standard.

9.3
Sample
Replicates
All
samples,
including
field
spikes,
trip
blanks,
etc.,
should
be
extracted
at
least
in
duplicate,
and
in
triplicate
if
difficulties
were
encountered
in
the
sampling
and/
or
holding
conditions
of
the
samples.
The
relative
percent
difference
(
RPD)
of
duplicate
samples
or
relative
standard
deviation
(
RSD)
should
be
less
than
15%
for
the
precision
of
sample
preparation
and
analysis
to
be
considered
in
control.

9.4
Matrix
Spike
Matrix
spikes
are
prepared
for
each
sample
type
and
analyzed
to
determine
the
matrix
effecton
the
recovery
efficiency.
Matrix
spike
recoveries
should
fall
within
±
25%
of
expected
values.
If
the
matrix
spikes
fail,
evaluate
the
lab
control
spikes.
If
the
LCS
are
within
acceptance
criteria
there
may
be
matrix
issues
in
the
samples.
Discuss
these
in
the
final
report.

ETS­
8­
I
54.1
Page
9
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
Matrix
spike
duplicates
are
prepared
periodically
to
measure
the
precision
associated
with
the
analysis.
Analyze
a
matrix
spike
and
matrix
spike
duplicate
(
if
prepared)
in
the
same
run
as
the
original
sample.

Matrix
spike
and
matrix
spike
duplicate
concentrations
should
fall
in
the
mid­
range
of
the
initial
calibration
curveor
should
be
prepared
at
1.5­
5
times
the
endogenous
concentration
of
the
analyte.
Spike
concentrations
should
fall
in
the
low­
range
of
the
initial
calibration
curve
if
extremely
low­
levels
are
expected.
Generally
two
or
more
levels
are
prepared,
one
in
the
low
range
of
the
curve
and
one
in
the
mid­
range.
This
avoids
the
need
to
pie­
screen
unknown
samples
prior
to
preparation.

9.5
Laboratory
Control
Spike
Lab
control
spikes
are
prepared
for
each
study
to
ensure
recovery
of
the
target
analytes.
These
should
be
prepared
at
a
minimum
of
2
levels
and
in
duplicate
or
triplicate.
Recovery
of
these
samples
should
be
within
±
25%
of
expected
values,
and
the
RPD
(
or
RSD)
be

15%.
If
recoveries
fall
outside
these
limits
the
samples
should
be
addressed
in
the
final
report.

10
Calibration
and
Standardization
10.1
Instrument
Setup
Note:
In
this
example,
a
MicroMass
UItimaTM
Liquid
Chromatography
Tandem
Mass
Spectrometer
(
LC/
MS/
MS)
is
used.
Other
brands
of
LC/
MS/
MSs
as
well
as
single
quadrupole
mass
spectrometers
(
LC/
MS)
may
be
used
as
long
as
the
method
criteria
are
met.
Brand
names,
suppliers,
part
numbers,
and
models
are
for
illustrative
purposes
only.
Equivalent
performance
may
be
achieved
using
apparatus
and
materials
other
than
those
specified
here,
but
demonstration
of
equivalent
performance
that
meetsthe
requirements
of
this
method
is
the
responsibility
of
the
laboratory.
The
operatormust
optimizeand
document
the
equipment
and
settings
used.

Establish
the
LC/
MS/
MS
system
and
operating
conditions
equivalent
to
the
following:

Mass
Spec:
Micromass
Ultima
(
Micromass)

Interface:
Electrospray
(
Micromass)

Mode:
Electrospray
Negative,
Multiple
Response
Monitoring
(
MRM)

Harvard
infusion
pump
(
Harvard
Instruments),
for
tuning
Computer:
COMPAQ
Professional
Workstation
AP200
Software:
Windows
NT,
MassLynx
3.3
HPLC:
Hewlett
Packard
(
HP)
Series
1100
HP
Quaternary
Pump
HP
Vacuum
Degasser
HP
Autosampler
HP
Column
Oven
Note:
A
4
x
10mm
Hypercarb
drop­
in
guard
cartridge
(
Keystone,
part
#
844017­
400)
may
be
attached
on­
line
after
the
purge
valve
and
before
the
sample
injector
port
to
trap
any
residue
contaminants
that
may
be
in
the
mobile
phase
and/
or
HPLC
system.

HPLC
Column:
Genesis
C8
(
Jones
Chromatography),
2.1mm
x
50mm,
4i.
tm
ETS­
8­
154.1
Page
10
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
Column
Temperature:
35
°
C
Injection
Volume:
l5pL
Mobile
Phase
(
A):
2mM
AmmoniumAcetate
inASTM
Type
I
water
(
See
7.3.1)

Mobile
Phase
(
B):
Methanol
met
Note:
Other
HPLC
gradients
may
be
used
as
long
as
the
method
criteria
are
It
may
be
necessary
to
adjust
the
HPLC
gradient
in
order
to
optimize
instrument
performance.
Columns
with
different
dimensions
(
e.
g.
2.1mm
x
30mm)
and
columns
from
different
manufacturers
(
Keystone
Betasil
C18
etc.)
may
be
used.

Ions
Monitored:

PFOA
PFOS
FOSA
413
169
5.0
499
99
5.2
498
78
5.8
Other
product
ions
may
be
chosen
at
the
discretion
of
the
analyst~
although
m/
z
99
is
suggested
for
PFOS.
Use
of
the
suggested
primary
ion
is
recommended.
Retention
times
may
vary
slightly,
on
a
day­
to­
day
basis,
depending
on
the
batch
of
mobile
phaseetc.
Drift
in
retention
times
is
acceptable
within
an
analytical
run,
as
long
as
the
drift
continues
through
the
entire
analysis
and
the
standards
are
interspersed
throughout
the
analytical
run.

10.2
Tune
File
Parameters
The
following
values
are
provided
as
an
example.
Actual
values
mayvary
from
instrument
to
instrument.
Also,
thesevalues
may
be
changed
from
time
to
time
in
order
to
optimize
for
greatest
sensitivity.

Analyte
Dwell,
sec
Collision
Energy,
eV
Cone,
V
PFOA
0.2
 
0.4
10
 
25
20
 
30
PFOS
0.2
 
0.4
30
 
60
50
 
80
FOSA
0.2
 
0.4
20
 
50
30
 
60
ETS­
8­
I
54.1
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LCIMS.
Page
11
of
17
Time,
Percent
Mobile
Percent
Mobile
Flow
Rate,
mm
Phase
A
Phase
B
mLlmin
0.0
60
40
0.3
0.4
60
40
0.3
1.0
10
90
0.3
7.0
10
90
0.3
7.5
0
100
0.3
9.0
0
100
0.4
9.5
60
40
0.4
13.5
60
40
0.4
14.0
60
40
0.3
Approximate
Analyte
Primary
Ion
Product
Ion
Retention
Time
(
minutes)
Gas
Flows
Set
Cone
Gas
150L/
hr
Desolvation
700L/
hr
Pressures
Set
Gas
Cell
3.
Oe
 
3mbar
ETS­
8­
I54.1
Page
12
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
Source
Set
Capillary
2.6
 
3.5kV
Hexapole
I
0.5V
Aperture
I
0.2V
Hexapole
2
0.8V
Source
Block
Temp.
100
 
150
°
C
Desolvation
Temo.
250
 
400
°
C
Analyzer
Set
LMResI
12.5
 
15.
OV
HM
Res
I
12.5
 
15.
OV
lEnergy
I
O.
7V
Entrance
 
2V
Exit
IV
LMRes2
11.
OV
HMRes2
11.
OV
lEnergy
2
l.
OV
Multiplier
650V
10.3
Calibration
Curve
Analyze
the
standard
curves
prior
to
each
set
of
samples.
The
validated
method
specifies
that
the
standard
curve
should
be
plotted
using
a
linear
fit,
weighted
1/
xor
unweighted.
However,
the
standard
curve
mayalso
be
plotted
by
quadratic
fit
(
y
=
ax'
+
bx
+
c),
weighted
l/
x
or
unweighted,
using
suitable
software.
The
calibration
curves
may
include
but
should
not
be
forced
through
zero.
The
mathematical
method
used
to
calculate
the
calibration
curveshould
be
applied
consistently
throughout
a
study.
Anychange
should
be
thoroughly
documented
in
the
raw
data.

Ifthe
calibration
curve
does
not
meet
acceptance
criteria
perform
routine
maintenanceor
prepare
a
new
standard
curve
(
if
necessary)
and
reanalyze.

For
purposes
of
accuracy
when
quantitating
low
levels
of
analyte,
it
may
be
necessary
to
use
the
low
end
of
the
calibration
curve
rather
than
the
full
range.
For
example,
when
attempting
to
quantitate
approximately
50
pg/
mL
of
analyte,
generate
a
calibration
curve
consisting
of
the
standards
from
25
pg/
mL
to
1000
pg/
mL
ratherthan
the
full
range
of
the
curve
(
25
pg/
mL
to
25000
pg/
mL).
This
will
reduce
inaccuracy
attributed
to
linear
regression
weighting
of
high
concentration
standards.

High
and/
or
low
points
maybe
excluded
from
the
calibration
curves
to
provide
a
better
fit
over
the
linear
range
appropriate
to
the
data
or
because
they
did
not
meet
tirejire­
determined
acceptance
criteria.
Low­
level
curve
points
should
also
be
excluded
if
their
area
counts
are­
notatleasttwice
that
of
the
method
and/
or
solvent
blanks.
Any
curve
pointmay
be
rejected
due
to
a
bad
injection
or
failing
to
meet
accuracy
requirements
of
±
25%
(
and
±
30%
for
the
LLOQ).
Justification
for
exclusion
of
calibration
curve
points
will
be
noted
in
the
raw
data.
Aminimum
of
6
points
will
be
used
to
construct
the
calibration
curve.

10.4
Continuing
Calibration
Verification
(
CCV)

Continuing
calibration
verifications
(
CCV)
are
analyzed
to
verify
the
accuracy
ofthe
calibration
curve.
Analyze
a
mid­
range
calibration
standard,
one
of
the
same
standards
used
to
construct
the
calibration
curve,
at
a
minimum
after
every
tenth
sample,
not
including
solvent
blanks,
with
a
minimum
of
one
persample
set.
Calibration
verification
injections
must
be
within
±
25%
to
be
considered
acceptable.
The
calibration
curve
and
the
last
passing
CCV
will
then
bracket
acceptable
samples.
Multiple
CCV
levels
may
be
used.

10.5
System
Suitability
A
minimum
of
three
system
suitability
sampleswill
be
injected
at
the
beginning
and
end
of
each
analytical
run.
Typically
these
samples
are
run
prior
to
the
calibration
curve.
The
system
suitability
injections
must
havearea
counts
with
an
RSD
of~
5%
and
a
retention
time
RSD
of

2%
when
evaluated
independently.

11
Procedures
11.1
Extraction
Scheme
Allow
samples
to
equilibrate
to
room
temperature.
Thoroughly
mix
samples
by
gently
inverting
the
sample
bottle.
Measure
4OmLof
sample
into
5OmL
polypropylene
centrifuge
tubes(
Spike
the
Matrix
spikes
as
required*,
replace
lid
and
mix
well).

Note:
*
Samples
may
need
to
be
prescreened
to
determine
an
appropriate
matrix
spike
level
(
typically
50
 
150%
of
sample
concentration).
Alternatively
the
samples
could
be
spiked
at
more
than
one
level,
allowing
for
the
inappropriate
spike
level
to
be
eliminated.

ETS­
8­
1
54.1
Page
13
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
Condition
the
CI8
SPE
cartridges
(
Ig,
6mL)
by
passing
approximately
IOmLmethanol
followed
by
approximately
5OmL
ASTM
Type
I
water(
flow
rate
approximately
2
drop/
sec).
Do
not
let
column
run
dry.

Note:
For
the
following
steps,
maintain
a
­`
I
drop/
sec
flow
rate.
Do
not
allow
the
column
to
run
dry
at
any
time.
Load
the
analytical
sample
onto
the
C18
SPE
cartridge.
Discard
eluate.

Ten
mL
of
the
40%
methanol
in
water
wash
mixture
is
passed
through
the
Cl
8
SPE
cartridge
to
rinse
away
potential
interferences
and
then
discarded.
This
step
must
be
omitted
if
perfluorinated
compoundswith
chain
lengths
less
than
C8
are
targeted
since
these­
will
beJost
during
this
wash
step.

Elute
with
exactly
5mL
of
100%
methanol.
Collect
eluate
into
graduated
1
5mL
polypropylene
centrifuge
tubes.
This
is
the
target
elution
fraction
(
final
volume
approximately
4.5
mL
as
not
all
of
the
solvent
will
leave
the
SPE
column.
This
will
not
affect
the
calculations
in
any
way
since
the
curve
is
also
extracted).

Analyze
a
portion
of
the
target
elution
fraction
eluent
using
negative
electrospray
HPLC/
MS/
MS
or
HPLC/
MS.
Note:
Samples
are
concentrated
by
a
factor
of
eight
during
the
extraction;
Initial
Vol
=
4OmL
­~

Final
Vol.
=
5mL.
Samples
are
stable
at
room
temperature
for
at
least
24
hours.
Analytical
samples
may
be
stored
in
a
refrigerator
at
4
°
±
2
°
C
until
analysis.
Standardization
of
C18
SPE
columns
 
If
poor
recoveries
are
observed,
it
may
be
necessary
to
standardize
the
C18
SPE
columns
in
the
following
manner
before
analyzing
samples.
Use
a
standard
with
an
analyte
concentration
between
1000
and
4000
pg/
mL.
Repeat
the
extraction
scheme
from
the
beginning
up
through
the
eluting
with
 
5mL
100%
methanol.
After
the
eluting
with
 
5mL
100%
methanol
step,
collect
an
additional
post­
elution
fraction
by
eluting
with
an
additional
5mL
of
100%
methanol.
Analyze
both
fractions
by
HPLC/
MS/
MS
or
HPLC/
MS.
If
the
target
fraction
contains
a
minimum
of
85%
of
the
respective
analytes,
it
may
be
considered
acceptable.

If
the
wash
contains
significant
standard
(>
15%),
either
the
wash
volume
or
percentage
of
MeOH
should
be
decreased.
If
the
post­
elution
fraction
contains
significant
standard
(>
15%),
the
target
elution
volume
should
be
increased.

11.2
Sample
Analysis
Set
up
analysis
sample
queue.

Inject
the
samevolume
(
between
5
 
25pL)
of
each
standard,
analytical
sample
and
blank
into
the
instrument.

All
sampleswith
a
concentration>
ULOQ
must
be
diluted
and
reanalyzed.
If
dilution
of
the
final
extract
fails
to
produce
acceptable
results
(
e.
g.
poor
MS
recoveries)
dilute
the
original
sample
and
re­
extract.

12
Data
Analysis
and
Calculations
Calculate
the
analytical
sample
(
extract)
concentration
from
the
standard
curve
using
the
following
equation:

ETS­
8­
154.
I
Page
14
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
Extract
Concentration,
pg/
mL
=
(
Peak
area
 
intercept)

(
slope)

Calculate
the
percent
recovery
ofthe
FSCS
using
the
following
equation:

FSCS
%
rec.
=
(
FSCS
conc.,
pci/
mL)
x
100
(
Conc.
added,
pg/
mL)

Calculate
the
percent
recovery
of
the
MSs
using
the
following
equaticn:

MS
%
rec.
=
(
MS
conc.,
pgfmL
 
Sample
conc.,
pg/
mL)
x
100
(
Conc.
added,
pg/
mL)

13
Method
Performance
Note:
Anymethod
performance
parameters
that
are
not
achieved
must
be
considered
in
the
evaluation
of
the
data.
Nonconformance
to
any
specified
parameters
must
be
described
and
discussed
in
any
reporting
of
the
data.

If
criteria
listed
in
this
method
performance
section
are
not
met,
maintenancemay
be
performed
on
the
system
and
samples
reanalyzed,
or
other
actions
taken
as
determined
by
the
analyst.
Document
all
actions
in
the
raw
data.

If
data
are
to
be
reported
when
performance
criteria
have
not
been­
met,
theiiatamustbe
footnoted
on
tables
and
discussed
in
the
text
of
the
report.

13.1
System
Suitability
A
minimum
of
three
system
suitability
sampleswill
be
injected
at
the
beginning
and
end
of
each
analytical
run.
Typically
thesesamples
are
run
prior
to
the
calibration
curve.
The
system
suitability
injections
musthavearea
counts
with
an
RSD
of

5%
and
a
retention
time
RSD
of

2%
when
evaluated
independently.

13.2
Quantitation
Calibration
Curve:
The
coefficient
of
determination
(
r2)
value
for
the
calibration
curve
must
be
greater
than
or
equalto
0.990.
Each
point
in
the
curve
must
be
within
±
25%
of
the
theoretical
concentration
with
the
exception
of
the
LLOQ,
which
may
be
within
±
30%.

Demonstration
of
Specificity:
Specificity
is
demonstrated
by
chromatographic
retention
time
(
within
3%
of
standard)
and
the
mass
spectral
response
of
unique
ions.

13.3
Sensitivity
Solvent
Blanks
and
Method
Blanks:
Solvent
and
method
blankarea
counts
mustbe<
50%
that
of
the
lowest
standard
used
in
the
calibration
curve.

Limits
of
Quantitation
(
LOQ):
The
lower
LOQ
(
LLOQ)
is
the
lowest
non­
zero
active
standard
in
the
calibration
curve;
the
peakarea
of
the
LLOQ
must
be
at
least
2X
that
of
the
extraction
blank.
By
definition,
the
measured
value
of
the
LLOQ
must
be
within
30%
of
the
theoretical
value.

ETS­
8­
I
54.1
Page
15
of
17
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
13.4
Accuracy
CCV
Performance:
Calibration
verification
injections
must
be
within
±
25%
to
be
considered
acceptable.
Thecalibration
curve
and
the
last
passing
CCV
will
then
bracketacceptable
samples.
MultipleCCV
levels
may
be
used.

MatrixSpikes:
Matrix
spike
percent
recoveries
must
be
within
±
25%
ofthaspiked
concentration.
If
matrix
effects
are
suspected,
evaluate
the
LCS
results
to
determine
if
a
matrix
effects
are
present
and
if
the
method
is
in
control
based
on
compliant
LCS
results.
Discuss
all
results
in
the
analytical
report.

13.5
Precision
Reproducibility:
Reproducibility
of
the
method
is
defined
by
the
results
of
duplicate
or
triplicate
analysis
of
samples.
A
RPD
or
RSD
of

15%
will
be
considered
acceptable.

System
Suitability:
The
system
suitability
injections
musthave
area
counts
with
an
RSD
of

5%
and
a
retention
time
RSD
of

2%
when
evaluated
independently.

14
Pollution
Prevention
and
Waste
Management
15
Sample
extract
waste
and
flammable
solvent
is
discarded
in
high
BTU
containers,
and
glass
pipette
waste
is
discarded
in
broken
glass
containers
located
inthelaboratory.

Records
Each
data
package
generated
for
a
study
must
have
the
followiriginformationirrcluded:
study
or
project
number,
acquisition
method,
integration
method,
sample
name,
extraction
date,
dilution
factor
(
if
applicable),
and
analyst.

Print
the
tune
page,
sample
list,
and
acquisition
method
to
include
in
the
appropriate
study
folder.
Copy
these
pages
and
tape
into
the
instrument
run
log.

Plot
the
calibration
curves
as
described
in
this
method,
then
print
these
graphs
and
store
in
the
study
folder.

Print
data
integration
summary,
integration
method,
and
chromatograms
and
store
in
the
study
folder.

Summarize
data
using
suitable
software
and
store
in
the
study­
folder.

16
Attachments
None.

17
References
"
Method
of
Analysis
for
the
Determination
of
Pertluorooctane
sulfonate
(
PFOS),
Perfluorooctane
sulfonylamide
(
PFOSA),
and
Perfluorooctanoate
(
POM)
in
Water",
E.
Wickremesinhe
and
J.
Flaherty,
Study
Number
023
 
002,
Centre
Analytical
Laboratories,
Inc.,
State
College,
Pennsylvania,
January
2000.

Validation
report
for
the
"
Method
ofAnalysis
for
the
Determination
of
Perfluorooctane
sulfonate
(
PFOS),
Periluorooctane
sulfonylamide
(
PFOSA),
and
Perfluorooctanoate
(
POAA)
in
Water",
E.
Wickremesinhe
and
J.
Flaherty,
Study
Number
023
 
002,
Centre
Analytical
Laboratories,
Inc.,
State
College,
Pennsylvania.

ETS­
8­
1
54.1
Determination
of
Perfluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
Page
16
of
17
18
Affected
Documents
None.

19
Revisions
Revision
Revision
Number
Revision
Number
Date
Updated
to
the
new
format.
Changed
Title.
7
Section
1:
States
the
validation
of3
analytes,
removes
reference
to
EPA
document
that's
no
longer
applicable.
Section
2:
Provided
for
the
extraction
of
more
than
the
3
validated
analytes,
allows
the
use
of
a
LC/
MS
system,
not
only
the
LS/
MS/
MSpreviously
mentioned.
Section
3:
Revised
definitions
for
field
matrix
spike,
field
control
spike,
LLOQ,
methodblank,
and
MDL.
Section
5:
Reworded
the
interferences,
added
recommendation
to
use
disposable
pipettes.
Section
6:
Recategorized
and
pared
down.
Section
7:
Changed
storage
time
to
6
months.
Added
more
calibration
points
to
the
table.
Section
8:
Addedstatement
addressing
labeling
requirements
and
spiking
procedures.
Expanded
section
8.8.
Section
9:
New
Section
Section
10:
Changed
some
of
the
parameters
in
the
tables.
Allowed
for
use
of
different
instrumentation.
Addedinformation
from
section
12
of
previous
version,
extensively
revised.
Section
11
(
section
9
in
previous
version):
Clarification
of
wash
step,
stated
exact
volume
of
eluate
is
5
mL,
revised
standardization
process,
removed
requirement
to
use
LC/
MS/
MS.
Section
12
(
section
13
in
previous
version:
no
changes
Section
13
(
section
14
in
previous
version):
Extensively
rewritten.
Section
14
(
section
15
in
previous
version):
no
changes
Section
15
(
section
16
in
previous
version):
Minorchanges
to
recording
requirements.
Section
16
(
section
17
in
previous
version):
Removed
attachment.
Section
17
(
section
18
in
previous
version):
Removed
reference
to
EPA
document
that
no
longer
applied
to
this
SOP.
Section
18:
New
section.

ETS­
8­
I
54.1
Page
17
of
17
Determination
of
Periluorinated
Compounds
in
Water
Using
SPE
and
LC/
MS.
3M
Environmental
Laboratory
Method
Solid
Phase
Extraction
and
Analysis
of
Perf!
uorinated
Acids,
Alcohols,
Amides,
Sulfonates
and
other
Fluorinated
Compounds
by
High
Performance
Liquid
Chromatography/
Mass
Spectrometry
Method
Number:
ETS­
8­
231.2
Adoption
Date:
13
November
2001
Revision
Date:
Upon
Signing
Effective
Date:
Upon
Signing
William
K.
Reagen
Laboratory
Manager
ETS­
8­
23
1.2
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
C/
A~)/
o~
Date
Page
1
of
23
Approved
By:
I
7~
F
~
nt
r
L
~
1
14~
r14
da~
ftom
~~

Scope
and
Application
This
method
describes
the
extraction
of
target
analytes
from
various
biological
matrices
using
solid
phase
extraction
(
SPE)
followed
by
separation,
identification,
and
quantitation
using
high­
performance
liquid
chromatography
mass
spectrometly
(
HPLC/
MS)
or
tandem
mass
spectrometry
(
HPLCIMS/
MS).
This
method
has
been
validated
for
perfluorooctane
sulfonate
(
PFOS)
in
mouse
and
rat
liver
and
sera
using
HPLC/
MS/
MS
based
on
the
FDA
guidance
document,
"
Guidance
for
Industry
 
Bioanalytical
Method
Validation,
May
2001".
Additionally,
the
performancebased
validations
in
quail
and
mallard
matrices
confirmed
that
the
method
can
be
used
in
matrices
other
than
the
rat
and
mice
matrices
already
validated,
provided
the
concurrent
quality
control
samples
document
the
acceptability
of
each
data
set
as
shown
in
3M
Studies
E0l­
1256
titled
"
Analytical
Phase
Report
for
PFOS:
A
Reproduction
Study
with
the
Mallard"
and
E01­
l245
titled
"
Analytical
Phase
Report
for
PFOS:
A
Reproduction
Study
with
the
Northern
Bobwhite."
3M
Report
E02­
1042
titled
"
Method
and
Validation
Report
for
`
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Biological
Matrices
Method
Number:
ETS­
8­
23
1.1"
is
a
compilation
of
the
results
ofthese
studies.

These
procedures
are
appropriate
for
perfluorinated
acids,
alcohols,
amides,
sulfonates
and
other
fluorinated
compounds
in
any
biological
matrix
provided
that
quality
control
sample
results
meet
the
precision
and
accuracy
requirements
ofthe
study.

2
Method
Summary
This
method
easily
accommodates
a
significant
range
of
initial
sample
weights
and
volumes,
which
is
essential
when
preparing
samples
oflimited
quantities
or
when
large
sample
sizes
are
required
to
provide
low
detection
limits.
When
using
a
1
gram
SPE
cartridge,
sample
amounts
typically
ranging
from
0.02
to
1.0
mLor
g
are
prepared
(
fluids
diluted,
tissues
homogenized,
etc.)
at
a
1/
6,
1/
10,
or
1/
20
dilution
using
reagent
water.
An
aliquot
ofthe
dilution/
homogenate
is
spiked
with
the
appropriate
surrogate
or
analyte
mixture.
Acetonitrile
(
ACN)
is
added
as
an
extraction
solvent
and
also
serves
to
precipitate
the
proteins.
The
sample
is
capped,
mixed,
and
put
on
the
centrifuge
to
clarify
the
supematant.
The
supematant
is
transferred
to
a
clean
tube,
diluted
with
water,
and
passed
through
apre­
conditioned
C18
SPE
cartridge.
Finally,
the
analytes
of
interest
are
eluted
from
the
SPE
cartridge
and
analyzed
by
HPLC/
MS
or
HPLC/
MS/
MS.

Depending
on
sample
availability
and
detection
limits
required,
the
use
of
SPE
cartridges
of
different
sizes
provides
flexibility
for
varying
initial
sample
sizes
and
final
elution
volumes.
Sufficient
quality
control
samples
are
incorporated
into
each
study
to
ensure
that
precision
and
accuracyof
the
study
are
understood.
The
required
quality
control
elementsof
this
performance
based
method
meet
the
cross
validation
requirementsbased
enthe
FDA
guidance
documents.

3
Definitions
3.1
Dilution
A
dilution
expressed
as
1:
5
or
1/
6
is
defined
as:
1
mL
of
sample
+
5
mLs
of
diluent
for
a
total
of
6
mLs
combined,
unless
otherwise
noted.

3.2
SPE
cartridge
A
column
containing
an
open
solvent
reservoir,
retaining
fit,
sorbenthed,
retaining
fit,
andliiertip~
The
sorbent
bed
is
bonded
silica
which
is
designed
to
selectively
retain
or
elute
the
compounds
of
interest
depending
on
the
solvent
conditions.
The
compounds
of
interest
can
then
be
separated
from
difficult
matrices
and
introduced
into
an
appropriate
solvent
for
analysis.
SPE
cartridges
are
available
with
a
wide
range
of
sorbent
beds
in
a
variety
izfsizes,
allowing
the
analyst
to
process
a
wide
array
of
sample
types
andvolumes.

ETS­
8­
23l.
2
Page
2
of
23
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
I
~
~
1
1fllL
~
C
1~
SciJ.
4
~
~
L
L
~~
m
I)
5/
O7i2t~)
u3
3.3
Reagent
grade
water
Waterwith
no
detectable
concentration(
s)
ofthe
target
analyte(
s)
or
less
than
1/
2
the
limit
of
quantitation
(
LOQ).

3.4
Quality
control
sample
Sample
used
to
determine
extraction
precision
andaccuracy
and
to
estimate
the
uncertainty
of
results
generated.

4
Warnings
and
Cautions
4.1
Health
and
Safety
Warnings
Always
wear
appropriate
personal
protective
equipments
such
as
protective
gloves,
eye
protection,
and
appropriate
clothing
when
working
with
biological
matrices,
solvents,
chemicals
and
instrumentation.
For
potential
hazard
information
refer
to
material
safety
data
sheets,
packing
material,
the
3M
Environmental
Laboratory's
Chemical
Hazard
Review,
the
3M
Guide
to
Laboratory
Practices
or
other
information
as
appropriate.

4.2
Cautions
Theanalyst
must
be
familiar
with
the
laboratory
equipmentandpotential
hazards
including,
butnot
limited
to,
theuse
of
biological
materials,
solvents,
high
temperatures,
pressurized
gas
and
solvent
lines,
high
voltage,
and
vacuum
systems.
Referto
the
appropriate
equipmentprocedure
or
operator
manual
for
adilulionalinfbtinatinzraridcaution&

5
Interferences
To
minimize
interferences,
Teflon
should
not
be
used
for
sample
storage
or
anypart
of
instrumentation
that
comes
in
contact
with
the
sample
orextract.

Co­
extracted
matrix
components
may
enhance
or
suppress
the
measured
analyte
signal
in
the
mass
spectrometer.
These
effects
are
minimized
using
this
SPE
technique,
but
the
precision
and
accuracy
of
spike
results
must
be
evaluated
for
possible
effectsof
co­
extracted
matrix
interferences
that
may
be
present.

6
Instrumentation,
Supplies,
and
Materials
The
following
instrumentation,
supplies,
and
materials
are
used
while
performing
this
method.
Equivalent
instrumentation,
supplies,
andmaterials
may
be
used
in
place
ofthose
listed.

6.1
Instrumentation
and
Equipment
Vortexmixer,
VWR,
VortexGenie
2,
or
equivalent
Ultra­
Turrax
T25
tissue
homogenizer
with
various
size
blades,
or
equivalent
Vacuum
Pump
SPE
Extraction
Manifold
Centrifuge,
Mistral
1000,
IEC,
or
equivalent
Shaker,
Eberbach,
VWR,
or
equivalent
Balance
(+/­
0.1000
g)

Micromass,
Quattro
II
or
Ultima
triple
quadrupole
Mass
Spectrometer
equipped
with
an
electrospray
ionization
source,
or
equivalent
HP!
100
or
Agilent
low
pulse
solvent
pumping
system,
solvent
degasser,
column
compartment,
and
autosampler,
or
equivalent
ETS­
8­
23l.
2
Page
3
of
23
Solid
Phase
Extraction
andAnalysis
of
Fluorochemical
Compounds
from
Various
Matrices
I.
~
:~
L~
~.
1
~
i:
rt~
cni..~
~`~
4o
~
~

6.2
Supplies
and
Materials
Eppendorfor
disposable
pipettes,
plastic
or
glass
Dissecting
scalpels
Polypropylene
bottles,
(
Nalgeneor
equivalent),
various
sizes
Volumetric
flasks,
glass,
type
A,
various
sizes
40
mL
glass
vials
(
ICHEM
or
equivalent)

Plastic
sampule
vials,
(
Wheaton
or
equivalent),
6
mL
(
or
other
appropriate
size)

Centrifuge
tubes,
polypropylene,
various
sizes
Labels,
various
sizes
Graduated
pipettes,
glass
Syringes,
capable
ofmeasuring
5
~.
tLto
1000
~
iL
Bottle­
Top
Dispenser(
capable
ofdispensing
5mL
of
solvent)

SPE
extraction
cartridges,
for
example
1
g,
Sep­
Pak6
cc
ti­
functional
C18
(
Waters),
or
other
sizes
and
adsorbents
75
mL
sample
reservoir
(
or
other
appropriate
size)

Crimp
cap
glass
autovials
and
caps,
crimpers
HPLC
analytical
column:
KeystoneBetasil
C
18,
2
X
50
mm,
5
tim,
or
equivalent
7
Reagents
and
Standards
Reagent
gradewater,
ASTM
Type
I
Milli~
QTM,
Kandiyohi
purchased
drinking
water,
or
equivalent
Acetonitrile,
HPLC
grade
or
equivalent
Methanol,
HPLC
grade
or
equivalent
Ammonium
acetate,
reagent
grade
or
equivalent
Various
biological
fluids
or
tissues
from
supplier
or
study
sponsor
for
use
as
control
matrix
7.1
Control
Matrix
Tissue/
Fluid
Preparation
This
method
requires
that
a
control
matrix
be
used
for
the
preparation
of
matrix­
extracted
calibration
standards,
control
spikes,
and
method
blanks.
Therefore,
an
amount
of
control
matrix
should
be
prepared
to
provide
asufficient
amount
throughout
a
study.
Typically
an
aliquot
of
4OmL
is
mixed
with
200
mL
of
reagent
wateror
40g
of
matrix
is
homogenized
with
200
g
reagent
water
andused
as
control
matrix.
Other
amounts
may
be
used
to
maintain
the
same
dilution
ratio
as
the
samples
or
additional
control
matrix
may
be
prepared
during
a
study
asneeded.

Control
matrix
is
prepared
from
one
of
three
sources:
1)
a
study
control
matrix
from
a
study
control
received
with
a
sample
set;
2)
a
commercially
obtained
sample
of
the
same
matrix
as
the
study
matrix;
or
3)
asurrogate
matrix,
also
obtained
commercially,
but
of
a
different
type
than
the
study
matrix.
(
eg.
if
rat
is
used
to
generate
calibration
curves
andQCs
for
amouse
study).

7.2
Reagents
Preparation
Prepare
a
2.0
rnJvI
ammonium
acetate
solution
for
use
as
a
mobile
phase
for
the
liquid
chromatograph:
Weigh
approximately0.300
g
arnmonium
acetate.
Pour
into
a
2000
mL
volumetric
container
containing
reagent
grade
water,
mix
until
all
solids
are
dissolved,
and
bringto
volume
using
reagent
grade
water.
Store
at
roomtemperature.

Note:
When
preparing
different
volumes
or
concentrations
than
those
listed
for
reagents
and
target
analyte
standard
preparation
adjust
accordingly.

ETS­
8­
231.2
Page
4
of
23
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
I)
u~
re,
t
may
4a
u~
adi
a:~
a
4
4a4~
4am
4
7.3
Target
analyte
standard
preparation
Prepare
target
analyte
standard(
s)
for
the
matrix­
extracted
calibration
curve.
Multi­
component
analyte
standards
are
acceptable.
The
following
is
an
example
only
and
may
or
may
not
be
appropriate
for
all
standard
preparations.

Weigh
approximately
100
mg
of
target
analyte
into
a
100
mL
volumetric
flask
and
record
the
actual
weight
in
the
standard
logbook
or
other
appropriate
location.
Bring
to
volume
with
methanol,
or
other
appropriate
solvent,
for
a
stock
standard
of
approximately
1000
ppm
(~
tg/
mL).
The
calculated
concentration
of
this
stock
standard
should
be
corrected
for
purity
and
should
reflect
the
fluorochemical
anion
concentration
that
is
measured.

Dilute
the
stock
solution
with
methanol
for
a
working
standard
solution
of
approximately
50
ppm.
Example
calculation:
l000~
ig/
mLx
5
mL/
100mL
=
50J1g/
mL.
Additional
dilutions
or
serial
dilutions
are
made
as
appropriate.

If
surrogate
standards
or
internal
standards
are
part
of
the
study,
these
stock
solutions
are
prepared
using
the
same
procedures
outlined
above.
Additional
information
regarding
the
appropriate
use
of
surrogates
or
internal
standards
is
found
in
section
9
of
this
method.
Surrogates
and
internal
standards
are
not
required
by
this
method,
but
may
be
used
for
individual
studies.

8
Sample
Handling
It
is
expected
that
samples
are
received
at
the
appropriate
conditions
and
must
be
kept
at
the
appropriate
conditions
until
the
extraction
is
performed,
(
i.
e.
frozen
for
biological
samples).
The
condition
of
samples
upon
receipt
and
storage
conditions
throughout
the
study
should
be
appropriately
documented.

Allow
liquid
samples
and
homogenized
tissue
samplesto
thaw/
warm
to
roomtemperature
prior
to
extraction.

Matrix­
extracted
standards
used
to
calibrate
the
instrument
response
for
each
target
analyte
are
prepared
at
least
once
during
a
study.
With
each
extraction
batch,
matrix
extracted
quality
control
standards
must
be
prepared
to
verify
the
extraction
efficiency.
Alternately,
an
entire
matrix­
extracted
curve
may
be
prepared
with
each
extraction
set.
Extracted
standardsand
samplesare
stored
in
capped
autovials
until
analysia.

Extracted
standardsand
samplesmay
be
stored
at
roomtemperature
or
refrigeralecLatapproximately
4
°
Cuntil
analysis
can
be
performed.
Ifrequired
by
the
study
protocol,
3
post­
extraction
matrix
spikes
(
i.
e.
aknown
amount
of
the
target
analyte
spiked
into
an
extracted
matrixblank)
may
be
stored
with
the
extracted
standardsand
samples
for
evaluation
to
demonstrate
extract
stability.
Alternately
the
precision
and
accuracy
ofmatrix­
extracted
quality
control
samplesmay
be
used
to
document
analyte
stability
during
the
time
frameof
analysis.
Results
will
be
documented
in
the
report.

9
Quality
Control
9.1
Blanks
9.1.1
Solvent
Blank
An
aliquot
ofmethanol
or
other
appropriate
solvent
is
analyzed
periodically
throughout
each
arralyticat
sequence
as
a
solvent
blank
to
serve
as
a
check
for
background
concentrations
of
analyte
or
cross
over
contamination
between
injections.
Solvent
blanks
are
not
extracted.

9.1.2
Method
Blank
An
aliquot
of
1.0
mL
of
reagent
water,
or
other
amount
as
appropriate,
is
used
as
a
method
biank.
Method
blanks
are
extracted
and
analyzed
with
each
extraction
set
following
this
procedure
and
may
or
may
not
be
spiked
with
surrogate
depending
on
study
requirements.
Method
blanks
may
serve
as
a
measure
of
analyte
background
concentrations
not
attributableto
endogenouslevels
in
the
control
matrix.

ETS­
8­
231.2
PageS
of
23
Solid
Phase
Extraction
andAnalysis
of
Fluorochemical
Compounds
from
Various
Matrices
Ducaamnt
may
4a
u~
m4.11
~
arrenL
la
I
4aVSU.
~
aa
4a4
/
~
44~
a
9.1.3
Matrix
Blank
An
aliquot
of
1.0
mL
or
1.0
g
of
matrix
(
diluted
or
homogenized)
is
used
as
a
matrix
blank.
Other
amounts
are
used,
as
appropriate,
in
keeping
with
the
amount
of
sample
available
for
extraction.
Matrix
blanks
are
prepared
from
one
of
three
sources:
1)
a
study
control
matrix
from
a
study
control
received
with
a
sample
set;
2)
a
commercially
obtained
sample
of
thesame
matrix
as
the
study
matrix;
or
3)
a
surrogate
matrix,
also
obtained
commercially,
but
of
a
different
type
than
the
study
matrix.
(
eg.
if
rat
is
used
to
generate
calibration
curves
and
QCs
for
a
mouse
study).
The
matrix
to
use
is
the
same
as
the
matrix
used
for
the
curve.
If
limited
matrix
is
available,
the
number
of
method
and
matrix
blanks
may
be
adjusted,
as
well
as
the
initial
volumes/
weights,
and
will
be
notedin
the
study
protocol
or
in
the
raw
data.

9.1
.3.1
Study
control
matrix
curve
­
if
the
study
control
matrix
is
used
for
the
curve,
preparefour
(
4)
matrix
blanks
using
the
study
control
matrix
(
two
spiked
with
surrogate
and
two
not
spiked).
Ifsurrogates
are
not
part
(
Ifa
study,
duplicates
are
acceptable.

9.1.3.2
Commercially
obtained
matrix
curve
 
if
acommercially
obtained
matrix
is
used
for
the
curve,
prepare
four
(
4)
matrix
blanks
using
the
same
commercially
available
matrix
(
two
spiked
with
surrogate
and
two
not
spiked).
If
surrogates
are
not
partofa
study,
duplicates
are
acceptable.

9.1.3.3
Surrogate
matrix
curve
 
ifa
surrogate
matrix
is
used
for
the
curve,
prepare
four
(
4)
matrix
blanks
using
the
same
surrogate
matrix
and
prepare
four
(
4)
matrix
blanks
using
the
study
control
matrix,
ifavailable.
If
study
control
matrix
is
not
available,
prepare
four
(
4)
matrix
blanks
using
acommercially
available
matrix
of
the
same
type
as
the
study
matrix
(
two
spiked
with
surrogate
and
two
not
spiked).
Ifsurrogates
are
not
partof
a
study,
duplicates
are
acceptable.

9.2
Sample
Replicate
Samples
replicates
are
prepared
accordingto
each
study
protocol,
project
outline,
or
as
determined
by
the
analyst.

9.3
Surrogate
standard
If
surrogate
standard
is
a
component
of
the
study,
all
samples
are
spiked
with
surrogate
standard
prior
to
extraction
to
obtain
an
appropriate
concentration
within
the
target
calibration
curve,
with
the
exception
of
blank
samples
as
described
in
section
9.1.

Typically
surrogate
standard
is
spiked
into
the
1.0
mL
or
1.0
g,
or
other
appropriate
volume,
diluted/
homogenized
sample
removed
for
extraction.

9.4
Initial
Calibration
A
minimum
of
one
set
ofmatrix­
extracted
standards
is
prepared
and
analyzed
to
establish
the
relationship
of
analyte
concentration
to
instrument
response
for
the
target
fluorochemical
analytes.
Details
are
provided
in
Section
10
ofthis
method.

9.5
Quality
Control
(
QC)
Sample
Quality
control
(
QC)
samples
are
used
to
determine
extraction
precision
and
accuracy
and
to
estimate
the
uncertainty
of
results
generated.
Typically
1.0
mL
ofthe
same
matrix
used
to
prepare
the
initial
calibration
curve
is
used
for
each
QC
sample,
and
the
target
analyte
standard
is
spiked
into
the
1.0
mL
or
1.0
g
diluted/
homogenized
sample
removed
for
extraction.
All
spiking
should
be
performed
at
the
same
step
in
the
extraction.
(
i.
e.
Ifthe
curve
is
spiked
into
the
1.0
mL
or
1.0
g
diluted/
homogenized
sample
removed
for
extraction
then
QC
samples,
surrogate
spike,
etc.
should
also
be
spiked
at
that
time.)

A
minimum
of
nine
(
9)
quality
control
samples
(
QC)
will
be
prepared
for
each
matrix
during
the
course
of
a
study
to
allow
the
needed
statistical
calculations
to
be
made.
When
study
samples
are
extracted
in
three
or
fewer
extraction
batches,
aminimum
of
three
QC
samples
must
be
prepared
(
one
at
each
level)
on
each
day
of
sample
extraction,
and
should
total
nine
or
more.
When
3
or
more
extraction
batches
are
prepared,
fewer
than
three
QC
samples
may
be
prepared
with
each
batch
provided
themethod
or
protocol
QCfrequency
requirements
are
met.

ETS­
8­
23l.
2
Page
6
of
23
Solid
Phase
Extraction
and
Analysis
ofFluorochemical
Compounds
from
Various
Matrices
I.
4yam~
e.
m:
aa~
4a
ja~
m.$
uaenL
io~
14
day
lom
(
b'
4a'
244
Thetotal
number
of
QC
sampleswithin
a
study
should
be
at
least
5%
of
thenumber
ofunknownsamples
or
nine
total
QCs,
whichever
is
greater.

The
levels
listed
below
are
target
guidelines
to
be
used
when
spiking
QC
samples
and
may
represent
sample
concentrations
diluted
into
the
range
ofthe
calibration
curve:

Low
level:
3X
to
5X
the
lower
limit
of
quantitation
(
LLOQ),

Mid­
level:
equivalent
to
a
point
near
the
middle
ofthe
calibration
curve,

High
level:
80%
of
theupper
limit
of
quantitation
(
ULOQ)

QC
samples
extracted
with
a
particular
sample
are
initially
analyzed
in
the
same
analytical
run
to
demonstrate
acceptable
quality
control
for
the
batch.
After
the
QC
samples
have
been
successfully
analyzed,
the
QC
samples
do
not
need
to
be
re­
injected
when
a
subset
of
samples
from
a
batch
are
analyzed
since
the
acceptance
criteria
for
the
batch
have
already
been
met.
The
extracted
calibration
curve
and
continuing
calibration
verification
samples
will
document
the
acceptable
performance
of
the
instrumentation.

Each
QC
is
expected
to
show
an
accuracy
85­
115%
of
expectedand
aprecision
of
15%.
A
minimum
of
2/
3
of
all
QC
samples
analyzed
within
an
analytical
run
must
meet
these
criteria.
Ifnot,
the
set
must
either
be
re­
analyzed
or
reextracted

9.6
Continuing
Calibration
Verification
Sample
(
CCV)

One
or
more
initial
calibration
curve
points
are
re­
injected
after
not
more
than
10
sample
injections
to
verify
the
continued
accuracy
ofthe
initial
calibration
curve.

Typically,
two
CCV
sample
levels
are
analyzed
after
every
tenth
sample
injection.
Solvent
blanks
are
not
considered
samples
and
are
not
included
as
such
for
determining
when
QC
samples
will
be
analyzed.
EachCCV
is
expected
to
show
an
accuracy
of
85­
115%
ofexpected.
Ifnot,
the
set
should
eitherbe
re­
analyzed
or
re­
extracted.

9.7
Matrix
Spike
(
MS)
Sample
Matrix
spike
samples
are
not
a
required
component
ofthis
method,
but
maybeiequireclby
protocol
or
project
plan.

If
the
matrices
for
the
curve
are
not
the
same
as
the
sample
matrix
to
be
analyzed,
or
the
protocol/
project
outline
requires
matrix
spike
samples;
the
specific
criteria
regarding
matrix
spikes
will
be
documented
in
the
protocol/
project
outline
and/
or
in
a
note
to
file.

9.8
Control
Charts
Thepercent
recovery
for
each
QC
sample
for
each
analyte
and
matrix
will
be
control
charted
as
per
SOP
ETS­
4­
26,
"
Control
Charts
for
Laboratory
Analyses".

9.9
Sample
Dilution
Any
sample
with
an
analyte
area
greater
than
that
of
the
highest
acceptable
standard
will
need
to
be
diluted
into
the
range
of
the
calibration
curve.
If
samples
are
diluted
into
the
range
of
the
curve
during
analyses
and
enough
sample
remains,
a
post­
run
dilution
validation
may
be
performed
to
verify
sample
values.

To
perform
the
dilution
validation,
one
sample
will
be
separated
into
two
representative
samples
(
i.
e.
two
1.0
mL
aliquots
for
fluid
samples
or
two
1.0
gram
amounts
for
tissue
samples,
or
other
amount
as
determined
by
the
analyst
and
documented
in
a
note
to
file)
then
diluted
using
two
procedures.
The
first
procedure
consists
of
diluting
the
sample
with
additional
blank
matrix
prior
to
extraction
(
fluid
adding
fluid),
while
the
second
procedure
consists
of
diluting
the
extract
with
solvent
post­
extraction
(
methanol
extract
adding
additional
methanol
solvent).

If
the
relative
percent
difference
is
not
within
15%
for
these
two
samples,
additional
testing
may
be
performed
to
determine
which
value
is
a
correct
representation
of
the
sample
concentration
as
determined
by
the
analyst
and
documented
in
anote
to
file.

ETS­
8­
231.2
Page
7
of23
Solid
Phase
Extraction
andAnalysis
of
Fluorochemical
Compounds
from
Various
Matrices
I.).~
mu
a
~
aybe
used.
~
~
4a
d~
:
4a
a
I.

10
Calibration
and
Standardization
10.1
Instrument
Calibration
At
least
one
calibration
curve
will
be
prepared
from
extracted
matrix
standards,
in
the
same
matrix
as
the
samples
if
available,
per
study.
It
will
consist
of
a
minimum
of
nine
concentration
levels.
Additional
calibration
curves
may
be
extractedon
separate
sample
extraction
dates,
as
detennined
by
the
analyst.

Transfer
1.0
mL
of
diluted
control
fluid
or
1.0
g
of
homogenized
control
tissue
to
a
15
mL
centrifuge
tube
using
a
disposable
plastic
pipette.
This
will
be
repeated
while
preparing
aliquots
for
the
calibration
curve.
Be
sure
to
vortex
mix
or
shake
by
hand
the
control
matrix
container
for
approximately
5
seconds
between
aliquots
to
ensure
a
homogenous
sample
is
removed.

Record
each
standard
weight
or
volume
on
the
weight/
volumes
sheet
or
extraction
worksheet,
as
appropriate.

Four
1.0
mL
or
1.0
g
aliquots,
or
other
appropriate
amount
as
described
in
9.1.2,
of
control
matrix
serve
as
matrix
blanks.

A
total
of
9
standards,
four
matrix
blanks,
and
fourmethod
blanks
are
prepared
in
addition
to
the
QC
samples
and
test
samples.
Thenumber
ofstandards
andblanks
may
be
adjusted
as
per
section
9.1,
oras
determinedhy
the
analyst.

Use
Attachment
C,
or
other
appropriate
form,
as
an
aid
in
calculating
the
concentrations
of
the
working
standards.
Referto
section
12
to
calculate
the
actual
concentrationofanalyte
in
each
calibration
standard
and
QC
sample.

Typically
the
target
analyte
standard
is
spiked
into
the
1.0
mL
diluted
fluid
or
1.0
g
homogenized
tissue
sample
removed
for
extraction.
All
spiking
should
be
preformed
at
the
same
step
in
the
extraction.
(
i.
e.
Ifthe
curve
is
spiked
into
the
1.0
mL
or
1.0
g
diluted/
homogenized
sample
removed
for
extraction
then
QC
samples,
surrogate
spike,
etc.
should
also
be
spiked
at
that
time.)

Analyze
the
extracted
matrix
calibration
curve
prior
to
each
set
of
extracts.
The
curve
equation
may
be
determined
by
quadratic
regression
analysis,
weighted
1
/
x,
and
external
standard
using
the
peak
areas
ofthe
target
analyte(
s)
using
MassLynx
or
other
suitable
software.
Other
suitable
regressions
may
be
used
as
appropriate.

Any
level
outside
85%
­
115%
of
nominal
must
be
deactivated,
and
regression
re­
calculated,
except
the
LLOQwhich
must
be
within
20%
ofnominal.
All
levelsmust
show
a
response
greater
than
twice
that
of
the
blank.
Aminimum
of
six
(
6)
levelsmay
be
used
to
plot
the
curve
in
anyone
set,
or
the
set
will
bere­
analyzed.

11
Procedures
11.1
Tissue/
Solid
Sample
Preparation
Obtain
tissue/
solid
samples.

Cut
approximately
1.000
g
oftissue
(+
1­
0.100
g),
or
otherappropriate
amount,
using
adissecting
scalpel.
This
partof
the
procedure
is
best
performed
quickly,
not
allowing
the
tissue
to
thaw.

Weigh
approximately
1.000
g
of
solid
sample
(+
1­
0.100
g).

Weigh
the
sample
directly
into
ataredplastic
sample
vial.

Record
the
weighton
the
weight/
volume
sheet,
extraction
worksheet,
or
other
appropriate
mention.

Return
unused
tissue/
solid
to
the
appropriate
storage
conditions
afterextractionamounts
have
beenremoved.

Weigh
5
mL
(
or
10
mL)
of
reagent
water
into
the
tared
sampule
vial
containing
the
tissue/
solid
sample
then
record
the
weight
on
the
weight/
volume
sheet.
(
For
solid
samples
that
don't
require
homogenization,
skip
the
remaining
steps
in
this
section.)

Homogenize
the
sample.
Put
the
Ultra­
Turrax
grinder
probe
in
the
sample
and
grind
for
approximately
2
minutes,
or
until
the
sample
is
homogeneous.

ETS­
8­
231.2
Page
8
of
23
Solid
Phase
Extraction
andAnalysis
of
Fluorochemical
Compounds
from
Various
Matrices
i)
oaemem
may
~
a
~
aec4
e~
eaa:,
4
1
a
4av~
floe
4aid
44~:
44
Takethe
grinder
apart
and
clean
it
with
methanol
after
each
sample.
Refer
to
ETS­
9­
52
for
more
information.

If
an
amount
other
than
1.000
g
(
not
within
+/­
0.100
g)
is
removed
for
an
initial
weight,
adjust
the
water
volume
accordingly
to
maintain
a
1/
6
dilution.
(
e.
g.
if0.5
g
is
removed
for
extraction,
add
a
total
of2.5
mL
ofwater.),
or
other
ratio
as
documentedin
the
protocol,
the
project
outline,
or
a
note
to
file.

11.2
Fluid/
Liquid
Sample
Preparation
Obtain
fluid/
liquid
sample
and
allow
it
to
thaw/
warm
at
room
temperature
or
in
lukewarmwater.

Label
a
15
niL
polypropylene
centrifuge
tube
with
the
study
number,
sample
ID,
extraction
date
and
analyst
initials.
See
attachedworksheet
(
AttachmentAor
similar
worksheet)
for
documentingthe
remaining
steps.

Vortex
mix
the
sample
for
approximately
15
seconds,
then
transfer
1.0
mL
of
fluid/
liquid,
or
other
available
amount,
to
a
plastic
sampulevial,
or
other
appropriate
plastic
container.

Return
unused
fluid/
liquid
to
the
appropriate
storage
conditions
after
extraction
amounts
have
beenremoved.

Add
5.0
mL
of
reagent
water
to
the
1.0
mL
of
fluid/
liquid
for
a
1/
6
dilution,
or
19
mL
of
reagent
water
to
the
1.0
mL
for
a
1/
20
dilution.

If
a
volume
other
than
1.0
mL
is
removed
for
an
initial
volume,
adjust
the
water
volume
accordingly
to
maintain
the
same
dilutions
as
above.

11.3
Tissue
and
Fluid
Sample
Extraction
After
tissue/
solid
or
fluid/
liquid
samples
have
been
prepared
according
to
sections
11
.1
and
11.2,
vortex
mix
or
shake
by
hand
the
diluted/
homogenized
sample
for
approximately
15
seconds
then
transfer
1.0
mL,
or
other
appropriate
volume,
to
a
clean
15
mLpolypropylene
centrifuge
tube.

Return
unused
diluted/
homogenized
portions
to
the
appropriate
storage
condition
after
extraction
amounts
have
been
removed.

Record
the
volume
ofdiluted
fluid/
liquid
removed
on
the
extraction
worksheet,
(
AttachmentAor
similar
worksheet).

Record
the
weight
of
the
homogenized
tissue/
solid
(
removed
as
a
1.0
mL
volume
or
other
volume)
on
the
weight/
volume
sheet,
the
extraction
worksheet,
or
other
appropriate
location.

Spike
blanks,
samples,
and
standards,
readyfor
extractionwith
surrogate
standard
as
described
hi­
this­
method.

Spike
each
calibration
standard
matrix
with
the
appropriate
amount
of
standard
as
described
in
this
method
for
the
calibration
curve
standardsand
each
QC
sample.

Vortex
mix
the
calibration
curve
samples
and
QC
samples
for
approximately
5
seconds.

To
each
sample
and
standard,
add
5.0
mL
of
acetonitrile,
cap,
and
vortex
mix
or
shake
by
hand
approximately
15
seconds.

Place
all
samples
on
the
shaker
at
an
appropriate
speed
for
20
minutes
to
adequately
mix
(
a
setting
of
approximately
300
rpm
on
the
models
listed
in
section
6.1).

Remove
from
the
shaker
and
centrifuge
at
an
appropriate
speed
for
10
minutes
to
adequately
pellet
the
precipitate
(
a
setting
of
approximately2000
rpm
on
the
models
listed
in
section
6.1).

Add
40.0
mL
of
reagent
grade
water
to
a
clean
50
mL
centrifuge
tube.
Remove
samples
from
the
centrifuge
and
decant
the
supernatant
into
the
water
in
the
50
mL
tube,
taking
care
not
to
introduce
any
of
the
matrix
solids
into
the
solution.
Cap
and
mix
by
inverting
several
times.
In
this
step
the
order
of
addition
may
be
changed
(
i.
e.
the
sample
maybe
put
into
the
centrifuge
tube
and
then
the
water
added).

Attach
the
reservoir
to
the
SPE
cartridgeand
attach
this
reservoir/
cartridgeunitlo
a
vacuum
manifold.

NOTE:
When
running
the
vacuum,
set
the
vacuum
chamberat
approximately
20
kPA
­
to
give
an
approximate
elution
flow
of
5­
7
mL/
min.
Flows
may
vary
through
cartridges
and
the
kPA
may
be
raised
for
slow
tubes
and
drying
after
most
have
beendrawn
down.

Prepare
the
SPE
cartridge
by
washing
twice
with
approximately
5.0
mL
ofmethanol,
followed
by
approximately
two
5.0
niL
aliquots
of
water,
taking
care
not
to
allow
the
column
to
run
to
dryness
after
each
wash.

ETS­
8­
231.2
Page
9
of
23
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
I
aa.
ama
:
nay
4
used
~`
~
aem4.
I'm
~
days
4c
a
(
II5/
O7z2~`
fi
~

After
washing
is
complete,
pour
the
sample
into
the
reservoir/
cartridge
unit
and
allow
all
of
the
liquid
to
pass
through
the
column
to
dryness.

Run
the
vacuum
on
high
for
approximately
5
minutes
to
adequatelydryeach
SPE
cartridge.

Place
a
collection
15
mLpolypropylene
centrifuge
tube
under
each
cartridge
and
elute
with
2.0
mL
ofmethanol.

Spike
extracted
blanks,
samples,
and
standards
with
internal
standard
as
described
in
this
method.

Label
each
glass
autovial,
as
appropriate,
with
the
study
number,
vial
file
archive
number,
identification
number/
gender/
timepoint
(
if
applicable)
or
LIMS
number,
matrix,
final
solvent,
analyte
components
(
if
needed),
extraction
type,
extraction
date,
and
analyst(
s)
performing
the
extraction.

Transfer
each
eluant
to
aglass
autovial
and
cap.

11.4
Extract
Analysis
11.4.1
Software
set­
up
Prior
to
analysis,
a
sample
list
must
be
created
using
the
following
parameters
and
naming
conventions:
instrument
designator
letter,
last
2
digits
of
test
year­
month­
day,
and
a
letter
that
will
increase
through
the
alphabet
with
each
additional
list
for
that
day.
Other
unique
sample
list
names
may
be
used.

Example
Sample
List:
IYYMMDDaor
A020204a
I
=
Initial
ofthe
instrument
name
(
A
=
"
Amelia")

YY=
Test
year
(
02)

MM=
Test
month
(
02)

DD=
Test
day(
04)

a
=
First
sample
list
(
run)
of
the
day
(
the
next
sample
list
will
endwith
`
b',
the
next
c,

and
so
on.)

Filenames
are
created
by
using
the
instrument
designator
letter,
the
last
2
digits
of
the
test
year­
month­
day,
and
a
3­
digit
sequential
file
number
that
starts
with
1
and
increases
by
one
for
each
filename.
Other
unique
filenames
may
be
used.

Example
filename:
IYYMMDD###
or
A020204001
I
=
Initial
of
instrument
name
YY
=
Test
year
MM
=
Test
month
DD
=
Test
day
=
3­
digit
sequential
filenumber
starting
with
1
through999
(
001)

Also,
as
part
of
the
sample
list,
assign
a
mass
spectrometer
acquisition
method
for
acquiring
data,
an
inlet
file
to
designate
HPLC
analytical
parameters,
a
bottle
number,
an
injection
volume,
and
individual
sample
descriptions.

Typically,
analytical
batch
run
sequences
begin
with
a
minimum
of
three
system
suitability
injections,
several
solvent
blanks,
and
acalibration
curve.
Theend
of
the
analytical
sequenceshould
be
markedby
a
closing
CCV.

Typically,
two
CCV
samples
should
be
injected
after
every
tenth
sample,
and
all
sample
extracts
must
be
bracketed
by
the
opening
calibration
curve
and
a
closing
CCV.
Solvent
blanks
should
be
analyzed
periodically
to
monitor
possible
analyte
carryover
and
are
not
considered
sample
extracts
but
may
be
includedas
such.

11.4.2
HPLC
set­
up
Load
the
autosampler
tray
according
to
the
sample
list
prepared
above.

ETS­
8­
231.2
Page
10
of
23
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
a'
as
10.
used.
4
ceo
14
~
ftori
fl~
`
7/~~
fl:'~

Download
the
HPLC
inlet
file
created
in
section
12.4.1.
Record
actual
conditions
in
the
instrument
logbook,
or
other
appropriate
location.
The
following
instrument
conditions
areprovided
as
an
example:

Sample
injection
volume
=
10
IlL
Injections/
sample
=

Cycle
time
=
10.0
minutes
Flow
rate
=
300
pL/
min.

Column:
Keystone
Betasil
C18,
2
x
50mm,
5
urn
particle
size,
or
equivalent
Mobile
phase:
Solvent
A=
2
mM
Animonium
Acetate
in
ASTM
Type
I
water
Solvent
B
=
Methanol
(
HPLC
grade)

Solvent
gradient:

Time
%
B
0.00
10%

1.00
10%

5.50
95%

7.50
95%

8.00
10%

11.4.3
Instrument
set­
up
The
following
instrument
conditions
areprovided
as
an
example:

Drying
gas250
 
400
liters/
hour
ES
nebulizinggas
10
 
15
liters/
hour
Sourceblocktemperature
=
150
°
C
Desolvation
temperature
=
250
°
C
These
settings
may
change
in
order
to
optimize
the
response.
Please
refer
to
the
instrument
tune
pageand
ETS­
9­
24,
"
Operation
and
Maintenance
of
Micromass
®
Triple
Quadrupole
Mass
Spectrometers
Fitted
With
Atmospheric
Pressure
Ionization
Sources,"
for
additional
details.

12
Data
Analysis
and
Calculations
12.1
Calculations
Ifalternative
calculations
are
used,
they
will
be
documented
in
the
raw
data.

Calculate
the
matrix
amount
contained
in
the
initial
dilution
using
the
following
equationn
Initial
Sample
Weight
(
g)
Weight
of
Tissue
Extracted
(
g)
=
.
x
Weight
of
HomogenateAliquot
(
g)
Homogenate
Sample
Weight
(
g)

Initial
SampleVolume
(
mL)
Volumeof
Fluid
Extracted
(
mL)
=
.
x
Volume
of
Fluid
Aliquot
(
rnL)
Diluted
Sample
Volume(
mL)

ETS­
8­
231.2
Page
11
of23
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
I/
oem
em
a
ay
ec
used
if
eurrea.
fe'
14
~
ays
h'om
43/
07/
7403
Calculate
actual
concentrations
ofanalyte
in
calibration
standards
using
the
following
equation:

=
Spike
Concentration
(
ug/
mL)
x
Spiked
Amount
(
mL)
1000
ng
Concentration
(
ng/
mL)
in
Final
SPE
Eluent
Calculate
actual
concentrationof
analyte
in
fluid
(
Ilg/
mL):
(
mL)
Final
SPE
Eluent
1
ug
(
AR
(
ng/
mL
eluent)
x
DF)
x
Final
Eluent
Volume
(
mL)
x
Volume
of
Fluid
Extracted
(
mL)

Calculate
actual
concentration
ofanalyte
in
tissue
(~
g/
g):
=
(
gg/
mL)
1000
ng
(
AR
(
ng/
mL
eluent)
x
DF)
x
Final
Eluent
Volume
(
rnL)
x
1.0
jig
=
(
jig/
g)
Weight
of
tissue
Extracted
(
g)
1000
ng
AR=
Analytical
result
from
MassLynxsummary
DF
=
Dilution
factor
introduced
when
sample
extract
is
diluted
to
fall
within
the
calibration
curverange.

Calculate
spike
percent
recoveries
using
the
following
equation:

Observed
Result
 
Unspike
Sample
Result
%
Recovery
=

Spiking
Level
Calculate
relative
standard
deviation
using
the
following
equation:

Standard
Deviation
Relative
Standard
Deviation
=
x
100
Mean
Calculate
percent
deviationusing
the
following
equation:
x
100
%
Deviation
=
Calculated
Cone.
 
Expected
Cone.

Expected
Cone.
x
100
13
Method
Performance
13.1
System
Suitability
System
suitability
will
be
determined
prior
to
the
start
and
at
the
completion
of
each
analytical
run.
Prior
to
the
calibration
curve
and
after
the
last
sample
of
the
run
three
(
3)
mid­
level
unextracted
calibration
standards
will
be
analyzed.
The
peakarea
precision
and
retention
time
precision
will
be
monitored
at
the
beginning
and
the
end
of
the
run
separately.
The
peak
area
precision
must
be
equal
to
or
less
than
5.0%
RSD,
the
precision
of
the
retention
time
must
be
equal
to
or
less
than
2.5%
RSD.

If
any
item
of
the
system
suitability
fails,
system
maintenance
must
be
completed
prior
to
running
a
second
set
of
system
suitability
samples.
The
system
suitability
must
pass
before
starting
the
calibration.
Ifsystem
suitability
fails
at
the
completion
of
a
run,
the
sample
set
must
be
reanalyzed.

13.2
Uncertainty
Estimation
The
combined
uncertainty
associated
with
this
method,
as
determined
by
the
validation
results,
is
as
follows:
29%
for
tissue
analyses
and
9%
for
fluid
analyses.
Referto
SOP
ETS­
12­
12,
"
Estimation
ofUncertainty
Measurements"
and
ETS­
8­
23
1.2
Solid
Phase
Extraction
and
Analysis
ofFluorochemical
Compounds
from
Various
Matrices
Page
12
of23
1/
or
uncut
may
be
ured,
ii'
current.
ha'
I
a
days
from
(.
15/
07/
2003
the
attached
uncertainty
estimation
report
(
Attachment
E)
for
additional
information
on
calculating
uncertainty
for
methods.

The
uncertainty
estimation
may
change
per
matrix
as
data
are
collected
and
added
to
control
charts,
or
as
control
charts
are
updated
according
to
ETS­
8­
26,
"
Control
Charts
for
Laboratory
Analyses".

13.3
Quantitation
The
coefficient
of
determination
value
for
the
calibration
curve,
plotted
by
regression
using
the
peak
areas
of
the
analyte(
s),
must
be
0.990
or
better.

During
the
validation
it
was
determined
that
four
curve
models
adequately
describe
the
concentration­
response
relationship
for
this
method:
linear­
null
weighted;
linear­
1/
x
weighted;
quadratic­
null
weighted,
and
quadratic­
1/
x
weighted.
However,
it
has
been
determined
that
quadratic­
l/
x
weighted,
external
standard
will
most
often
be
used
for
sample
quantitation.

All
active
calibration
curve
points
must
be
within
15%
of
the
theoretical
value
with
the
exception
of
the
LOQ
point,
which
may
deviate
up
to
20%.

Calibration
standards
with
peak
areas
less
than
two
times
the
curve
matrix
blank
will
be
deactivated
to
disqualify
a
data
range
that
may
be
affected
by
background
levelsof
the
analyte.

Avalid
calibration
curve
must
contain
at
least
6
active
points
above
and
including
the
LOQ.

Ifthe
curve
cannotmeet
these
criteria,
the
sample
set
must
be
reanalyzed
or
reextracted.

13.4
Accuracy
and
Precision
Two
thirds
of
all
quality
control
and
continuing
calibration
verification
samples
are
expected
to
show
an
accuracy
of
85­
115%.

Quality
control
samplesare
expected
to
show
a
precisionof
15%.

14
Pollution
Prevention
and
Waste
Management
Sample
waste
is
disposed
of
in
noninfectious
biohazard
waste
containers.

Flammable
solvent
waste
is
disposed
ofin
high
BTU
containers.

Glass
pipette
waste
is
disposed
of
in
broken
glass
containers
located
in
the
laboratory.

15
Records
Complete
the
extraction
worksheetattached
to
this
method,
or
other
applicable
worksheet,
and
store
withthe
study
raw
data.

Each
page
generated
for
a
study
must
contain
the
following
information
(
if
applicable):
study/
project
or
instrument
number,
acquisition
method,
integration
method,
sample
name,
extraction
date,
dilution
factor
(
if
applicable),
and
analyst.
Other
information
may
be
added
if
applicable
to
the
study.

Print
the
tune
page,
sample
list,
and
acquisition
method
from
MassLynx
to
include
with
the
study
raw
data.
Copy
these
pages
and
tape
into
the
instrument
runlog,
ifapplicable.

Plot
the
calibration
curve
by
the
appropriate
regression.
Print
these
graphs
and
store
with
the
study
raw
data.

Print
data
integration
summary,
integration
method,
and
chromatograms
from
MassLynx,
and
store
with
the
study
raw
data.

Summarize
data
using
suitable
software
(
Excel
7.0
or
LIMS)
and
store
in
the
study
folder.

Back
up
electronic
data
to
appropriate
medium.
Record
in
study
notebook
the
file
name
and
location
of
backup
electronic
data.

Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
~)
omui.
am
clay
be
mmd
if
cu
`
cut.
icr
a
dams
from
1.3/
0
.2003
16
Attachments
AttachmentA:
Extraction
Worksheet
Examples
AttachmentB:
Sample
Weight/
VolumeWorksheet
AttachmentC:
Calibration
Standard
Concentration
Worksheet
AttachmentD:
Dilutions
Summary
Worksheet
Attechrnetn
E:
Uncertainty
Estimation
Report
17
References
3M
report
­
E0l­
1277,
"
Method
Validation
for
ETS­
8­
231"

3M
Method
­
ETS­
4­
26,
"
Control
Charts
for
Laboratory
Analyses"

3M
SOP
­
ETS­
12­
12,
"
Estimation
ofUncertainty
Measurements"

3M
Equipment
Procedure
­
ETS­
9­
24,
"
Operation
and
Maintenance
of
Micromass
®
Triple
Quadrupole
Mass
Spectrometers
Fitted
With
Atmospheric
Pressure
Ionization
Sources"

3MEquipment
Procedure
­
ETS­
9­
52,
"
Operation
andMaintenance
of
a
Tissue
Grinder"

FDA
guidance
document,
"
Guidance
for
Industry
 
Bioanalytical
MethodValidation,
May
2001"

3Mreport
­
E0l­
l256,
"
Analytical
Phase
Report
for
PFOS:
AReproduction
Study
with
the
Mallard"

3Mreport
­
E0l­
1245,
"
Analytical
Phase
Report
for
PFOS:
AReproduction
Study
with
theNorthern
Bobwhite"

3M
report
­
E02­
l042,
"
Method
and
Validation
Report
for
`
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Biological
Matrices
Method
Number:
ETS­
8­
23
1.1"

3MStandard
Operating
Procedure
­
ETS­
12­
1,
"
Analytical
Definitions,
Abbreviations,
Acronyms,
and
Symbols"

18
Affected
Documents
None.

19
Revisions
Revision
Revision
Number
Revision
Description
Minor
formatting
changes.
Added
detailed
information
to
all
sections
concerningthe
02/
18/
02
extraction
procedure,
analytical
procedure,
and
calculations.
Added
attachments
and
references.
2
The
minor
changes
in
this
revision
will
not
affect
themethod
validation
results:
10/
04/
02
Section
1
added
reference
to
validated
method,
added
clarification
to
the
scope
of
the
method,
and
removed
fish
matrix.
Section
9.6,
10.1
added
text
for
consistency
in
spiking
calibration
curves
and
quality
control
samples.
Section
9.8
added
control
chart
section.

Section
12
updated
calculations
and
corrected
serumfinal
concentration
units.
Section
13.2
added
uncertainty
estimation.
Section
13.3
added
text
describing
which
curve
model
to
use
during
sample
analysis.
Section
16
added
additional
extraction
sheets
and
uncertainty
estimation
report.
Section
17
added
additional
references.
Added
clarification
to
text
throughout
method.

ETS­
8­
231.2
Page
14
of23
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
4ocm
an
i
a
cc
be
used
ten
ant.
3m'
14
days
ii
em
fib
a
7/
2
fr3
Attachment
A
 
ExtractionWorksheet
Example
1
Study
Number:
Prep
Date:

Analysts
initials:
Box#:
Method
Revision:
ETS­
8­
231
.2
Matrix:
Sample
Timepoint:

Sample
Volume/
Weight
Number
of
diluted
Amount
and
Amount
and
or
sample
spike
mix
surrogate
spike
Type
of
column
Elution
solvent
Sample
Number
description
removed
used
mix
used
used
and
lot
and
volume
Comments
signature/
Date
Blank
matrix
TN­
A­_________;
Amount
weighed/
aliguoted:
g/
mL
1.
Homogenize
sample
2.
Aliguot
1
mL
of
diluted
matrix
into
15
mL
polypropylene
tube
(
record
volume
or
weight
accordingly)

3.
Spike
samples
accordingly
4.
Add
mL
of
ACN
(
TN­
A­
to
each
diluted
sample
and
shake
or
vortex
mix
5.
Shake
sample
for
20
mm
ata
setting
of
rpm
(
Shaker_~

6.
Centrifuge
sample
for
10
mm
at
a
setting
of
rpm
(
Centrifuge
7.
Add
40
mL
of
water
to
50
mL
polypropelene
centrifuge
tube.

8.
Decant
extract
into
centrifuge
tubes
with
water
9.
Shake
sample
slightly
to
ensure
proper
mixing
10.
Attach
6
mL
C18
SPE
cartridges
and
75
mL
reservoirs
to
vacuum
manifold
11.
Condition
column
with
two
washes
of
 
5
mL
MeOH
(
TN­
A­
­
donot
allow
column
to
goto
dryness
12.
Wash
column
with
two
washes
of
 
5
mL
water
­
do
not
allow
column
to
goto
dryness
13.
Filter
sample
through
conditioned
column,
discarding
filtrate
14.
Allow
column
to
goto
dryness.
After
dripping
stops,
draw
a
high
vacuum
througltcolumn
for
at
leastñminutes
15.
Elute
column
with
solvent
(
TN­
A­
)
into
appropriate
15
mL
centrifuge
tube
16.
Spike
samples
with
uLof
internal
standard
#
conc.

17.
Transfer
sample
into
appropriately
labeled
autovial
and
cap
Note:
In
vacuum
steps
above
setthe
vacuum
chamber
at
approximately
20
kPA
­
this
should
give
approximately
5­
7
mL/
min
elution
flow
Flows
may
vary
through
cartridges
­
kPA
may
be
raised
for
slow
tubes
and
drying
after
most
have
been
drawn
dowriandshut
off.

ETS­
8­
23
1.2
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
Page
15
of23
I/)
uejniem
na:
7
Is
med
fur
n'mc.
Ioi
he
057
s
isan
(
73
ts~
n___________

ast________________

~
tooti_________________

I
g
,,
il~
tiiae9!,===~

`:
tieaasnii
,,,_:,,
y,,~,,__,

 
Attachment
A
 
Extraction
Worksheet
Example
2
/
1
1
~
~

F
UI
`
1
htt
~
jr
d1
Ii~

~
dUll
it!
H
irvn
ii~
th
fl
I
C*
iFti~
WSS
1
nsa
_

Cd
i~
tnW4il,_
y_~_
 
ttiltltaktil
I
man
_____________

ETS­
8­
23
1.2
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
Page
16
of23
I
I
`
ti'

I
L~

~
frTh~
:

;
~
F:_

Cdli
htsWlii_~/._,~.,,,,,

5'

as
ft______

CdlltàbI.
at
I
I______

Cdii
tnt
®
Sa
I
_______
ti________

(
diSk
~

tt'I
®
Lwt_,,_,_,,,,,,,,

 
CdfittMd
I
I
115'
rwra.
rn1remçap
asaaarak3nrnp%:

~`
akacPflwl
s
c.~
acbak5Nws'wa15M
4nkisIrnna*
~
4mt'
5)
tt$'
Dueume~
en
~
ayde
used
i
~
uueres4
d~
14
ciayshme
()
S/
O7/
2~.%)

Attachment
A
 
Extraction
Worksheet
Example
3
4,
1
1
j~.

!
H
h1L4
~.
t
d~
In
~
~
V
ht~!
1
ill
Ifi
d
`
Iii
IUfl
d
Ii
h
h
 
 
­
­

 
 
 
­
­
­
­
­

 
­

I
t1
Cd3ci~
t~
MtIj
__________
tvM~
________
__________

~
_____

?~
a~
ean,
a~,

ta'~~,_~
c__

Cd
Cd~
i
_______
C_______

°
CdoEa1w~
8
1
Cd
W
L~
n1d*~_
LL~_

tIe$

rosa___________

a______________

twfll
~
I
iii
I
~
H
It
ii
P
I
 
~
"..
 
f
~
a~
aa~
a
~
f~
8aa
4,
g&~

oi,~
if
I
f
t
I
p~

a'~\\
~
\
88
I
:`~
t~
 
=
­
t
Ot~
F
n~
Ii~&
s
S~
CIltwS~
pfld~
s.
t
1t~
ssi
~
~~
ve
SOntcwtt*'
tAaa
ETS­
8­
23
1.2
Solid
Phase
Extraction
and
Analysis
ofFluorochemical
Compounds
from
Various
Matrices
Page
17
of23
L)
c:
4
~
1n~
Eu~
tne
b~~.
scn.(
8u$'
reer
~
4
es.
s
~
4on~
~/
O7iI)
24
Attachment
B
­
Sample
Weight/
Volume
Worksheet
Prep
Date(
s):
Analyst(
s):
Sample
Matrix:
Method/
Revision:
Study
Number:
Equipment
Number:

Final
Solvent
&
TN
Number:

Sample
Initial
Water
added
Volume/
Weight
Date
omments
ID
Wt./
Vol.
Wt./
Vol.
Removed
Removed!

gImLIL
g/
mL
mL/
g
Analyst
Form
Completion
Verified
By:

ETS­
8­
23
1.2
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
Page
18
of23
4ocu~
ne1itmay
be
used.
ii
curr'~'
Ibr
1~
days
ii
on
I
Attachment
C:
Calibration
Standard
Concentration
Worksheet
Prep
date(
s):
Analyte(
s):
Sample
matrix:
Method/
revision:

Analyte
mix
std
approx.
0.500
ug/
mL:
Standard
number:
Equipment
number:
Final
solvent
and
TN#:

Blank
Tissue
or
Fluid/
identifier:

Analyte
mix
std
approx.
5.00
ug/
mL:
Analyte
mix
std
approx.
50.0
ug/
mL:

Surrogate
std
approx.
100
ug/
mL:
Actual
concentrations
ofstandards
in
the
analytemix
Analyte
Std
conc
ug/
mL
All
Amount
spiked
mL
All
Final
Volume:
mL
0.500
0.0015
2.00
0.500
0.0030
2.00
0.500
0.0080
2.00
0.500
0.0160
2.00
0.500
0.0320
2.00
5.00
0.0056
2.00
5.00
0.0080
2.00
5.00
0.0160
2.00
5.00
0.0240
2.00
5.00
0.0320
2.00
5.00
0.0400
2.00
50.0
0.005
2.00
50.0
0.006
2.00
Calculated
concentrations
of
standards
in
relation
to
the
final
2.0
mL
solvent
2.0
mL
Final
Volume
Analyte
Final
conc.
ng/
mL
Surrogate
Std
conc
ng/
mL
0.375
100
0.750
Surrogate
2.00
4.00
Final
conc
nglmL
8.00
14.0
0.500
20.0
40.0
60.0
80.0
100
125
150
ETS­
8­
23
1.2
Solid
Phase
Extraction
andAnalysis
of
Fluorochemical
Compounds
from
Various
Matrices
Page
19
of23
2oclInlen~.
may
be
u.
see.
ifs
~
0
m~
t.
14'
14
&
a
a
15001
bc~
ifr/
2Ud3
Attachment
D:
Dilutions
Summary
Worksheet
Study:
Dilution
Date/
Analyst:
Box
Number:
Solvent/
TN
Number:
Extraction
Date/
Analyst:
Matrix/
Timepoint:

SampleNumber
or
Description
Dilutions
1/
1/
1/
1/
1/
1/
1/
Comments
Verified
By:

Notes:
1/
10
dilution
=
I
mL
of
sample
+
9
mL
of
solvent
Form
Completion
Verified
By:

ETS­
8­
23
1.2
Solid
Phase
Extraction
andAnalysis
of
Fluorochemical
Compounds
from
Various
Matrices
Page
20
of
23
14ei,
1
neat
cu~
U
used
1
cur'm
if
4y
`~
fl
c's
fro'n
1)
5/
07/
2143
ETS­
8­
231
.1
Uncertainty
Estimation
Attachment
E:
Uncertainty
Estimation
Report
A
full
validation
for
the
use
of
this
method
for
the
determination
of
PFOS
has
been
performed
on
rat
liver
and
serum;
a
partial
validation
has
been
performed
with
PFOS
on
mouse
liver
and
serum.
A
method
validation
report
has
been
issued
for
these
validations,
as
well
as
a
performance­
based
validation
for
the
analysis
of
PFOS
in
mallard
and
quail
liver
and
serum.
This
report
presents
precision
and
bias
evaluations
through
this
method;
the
results
of
these
evaluations
will
help
in
estimating
the
uncertainty
for
theuse
of
this
method.

1.
Method
summary
Themethod
includes
the
following
procedures:

Weigh
subsample
Weigh
subsample
for
calibration
std
Dilute/
Homogenize
I
Prepare
spike
std
~.

Measure
aliquot
Clean­
up
4~
Final
volume
j
Result
2.
Major
sources
of
uncertainty
The
following
describes
each
ofseveral
majorsources
of
uncertainty
indahoratory
determinations,
the
nature
of
the
uncertainty
associated
with
each,
and
an
estimation
ofthe
extentof
the
error.

Mass
 
Balances
are
used
for
°
measuring
the
weight
of
the
sample
aliquot
for
analysis
(
using
an
analytical
balance),
only
for
solid
matrices
(
e.
g.,
liver.)
°
measuring
the
weight
of
analytical
standard
used
for
the
preparation
of
a
stock
standard
solution(
using
an
analytical
balance.)
As
discussed
in
SOP
ETS­
l2­
012,
the
uncertainty
associated
with
the
standard
weights
can
be
ignored.
Random
uncertainties
associated
with
balance
measurements
were
estimated
using
calibration
datafrom
one
analytical
balance:

ETS­
8­
23
1.2
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
Page
21
of23
Dilute/
Homogenize
Measure
aliquot
LC/
MS
analysis
F
,~
r
Clean­
up
4,
Final
volume
.4,

­
I
LC/
MS
calibration
i'ocumcl
a
may
be
usec.
ifcurrent,
14r
1
~
days
I
`
71407/
2002
Balance
ID
118453734
Logbook
pages
00062­
7
­
00062­
9
Standard
deviation
9.58E­
04
Square
of
standard_
deviation
9.
l7E­
07
n
118
However,
the
uncertainty
associated
by
the
sample
and
neat
standard
weighing
process
is
included
in
the
validation
studies
usingQC
samples.
No
additional
factor
is
needed.

Volume
 
Volumetric
flasks
are
used
for:
°
measuring
the
final
volume
of
spiking
solutions.
°
measuring
the
final
volume
of
calibrating
solutions.
Pipettors
are
used
for:
°
measuring
the
volume
ofthe
sample
aliquot,
only
for
liquid
matrices
(
e.
g.,
serum.)
°
measuring
the
volume
ofwater
added
before
homogenizing.
°
measuring
the
aliquot
taken
for
clean­
up.
°
measuring
the
final
volume
of
extract.
°
measuring
volume
aliquots
of
standard
solutions
in
the
preparation
of
spiking
solutions.
°
measuring
volume
aliquots
of
standard
solutions
in
the
preparation
Dfcalibrating
suhrtiuns.
As
discussed
in
SOP
ETS­
12­
012,
random
uncertainties
associated
with
volume
determinations
were
estimated
using
calibration
data
from
several
pipettors:

Equipment
type
Pipettors
Logbook
01054
Dates
4/
5/
02
­
7/
28/
02
Standard
deviation
1
.33E­
02
Square
of
standard_
deviation
1
.76E­
04
n
69
However,
the
uncertainty
associatedby
the
volume
measurements
of
aliquots
and
extracts
as
well
as
of
standard
solutions
is
included
in
the
validation
studies
using
QC
samples.
No
additional
factor
is
needed.

Calibration
Standards
 
Uncertainties
associated
with
the
neat
standards
are
not
supplied
with
fluorochemical
compounds.
Since
they
are
solid
compounds,
the
random
uncertainties
associated
with
their
purity
is
considered
negligible.
No
additional
factor
is
needed.

Standardization
Process
­
Thestandardization
(
or
calibration)
process
for
most
instrumental
analyses
introduces
uncertainty
because
of
the
inexact
nature
of
the
process.
With
a
correlation
coefficient
of
0.990,
the
minimum
required
by
the
method,
the
standard
error
of
estimatemay
be
10­
20%.
However,
the
uncertainty
associatedby
the
standardization
process
is
included
in
the
validation
studies
usingQC
samples.
No
additional
factor
is
needed.

Extractionand
Analysis
Process
 
With
recoveries
typically
being
acceptable
with
an
accuracy
of
70%
to
130%,
and
with
a
precision
of
15%,
it
is
expected
that
the
greatest
contribution
to
the
uncertainties
will
be
associated
with
the
extraction
and
analysis
process.
The
full
validation
data
generated
with
PFOS
in
rat
liver
and
serum
were
used
to
estimate
this
uncertainty,
using
all
the
%
recoveries
obtained
during
validation,
including
all
sets,
whether
they
met
the
target
recoveries
or
not.
It
is
assumed
to
follow
a
normal
distribution.

ETS­
8­
231.2
Page
22
of
23
Solid
Phase
Extraction
andAnalysis
of
Fluorochemical
Compounds
from
Various
Matrices
!.)
oaument.
cc'
1.
e
usc
4.
il
current.
14r
~
I
my's
fi'nni
05/
07:
214)
3
Method
validation
of
ETS­
8­
23
1
Matrix
Rat
liver
Rat
serum
Standard
deviation
2.06E­
0l
8.75E­
02
Square
of
standard_
deviation
4.24E­
02
7.66E­
03
n
116
52
Homogeneity
 
Not
aparameter
in
the
method
diagram
above,
sample
homogeneity
appears
to
be
a
significant
effect
in
the
analytical
procedure
for
solid
samples
(
e.
g.,
liver).
For
liquid
samples,
it
is
expected
that
this
parameter
will
not
be
significant.
The
homogeneity
of
the
liver
sub­
sample
is
not
taken
into
accountby
the
recoveries
of
the
QC
samples
during
validation,
since
spiking
occurs
after
the
l­
g
subsample
hasbeen
taken.
However,
as
discussed
in
the
EURACHEM/
CITAC
GuideQUAM:
2000.
Pl,
it
is
not
practical
to
measure
homogeneity
directly.
Following
the
same
rationale
presented
in
that
document,
an
uncertainty
factor
of
0.20
is
selected
for
this
parameter.
It
is
assumedto
follow
a
normal
distribution.

3.
An
analysis
of
the
effect
of
contributing
sources
on
combined
uncertainty
A
summary
table
ofthecombined
uncertainties
andexpanded
uncertainties
(
k
=
2)
in
typical
analyses
at
the
3M
Laboratory
are
therefore:

SUMMARY
OF
SOURCES
OF
UNCERTAINTY
rat
liver
rat
serum
Neat
standard
negligible
negligible
Homogeneitystandard
deviation
2.
OOE­
Ol
­

Validation
data
standard
deviation
2.06E­
Ol
8.75E­
02
Homogeneitysquare
of
standard
deviation
4.
OOE­
02
­

Validation
data
square
of
standard
deviation
4.24E­
02
7.66E­
03
Combined
uncertainty
29%
9%
Expanded
uncertainty
57%
18%

With
combined
uncertainty,
u.
a
function
of
the
individual
uncertainties
U,:

And
the
expanded
uncertainty,
U:

with
k
=
2
for
95%
confidence
interval.

As
an
example,
for
rat
liver:
uc=~
tj
Ui
2
U
=
kxu~

u~=
~
J~
2.0OE_
Oi)
2+
(
2X~
6E
 
O1)
2
=
~/(
4.00E
 
02+
4.24E
 
02)
=
0.287
29%

U
=
2
x
0.287
=
0.574
=
57%

ETS­
8­
23
1.2
Solid
Phase
Extraction
and
Analysis
of
Fluorochemical
Compounds
from
Various
Matrices
Page
23
of
23
