Evaluation
of
the
Effectiveness
of
Coatings
in
Reducing
Dislodgeable
Arsenic
from
CCA
Treated
Wood
Quality
Assurance
Project
Plan
Category
II/
Sampling
and
Analysis
Draft

Revision
1.2
U.
S.
Environmental
Protection
Agency
Air
Pollution
Prevention
and
Control
Division
March
2003
Contract
No.
68­
C99­
201
Work
Assignment
No.
4­
38
Project
No.
RN992014.0038
Prepared
for:

U.
S.
Environmental
Protection
Agency
Air
Pollution
Prevention
and
Control
Division
Research
Triangle
Park,
NC
27711
P.
O.
Box
13109
Research
Triangle
Park
North
Carolina
27709
Section
1
Draft
­
Revision
1.2
March
2003
Page
2
of
2
QAPP
DRAFT

Revision
1.2
March
27,
2003
Evaluation
of
the
Effectiveness
of
Coatings
in
Reducing
Dislodgeable
Arsenic
from
CCA
Treated
Wood
Quality
Assurance
Project
Plan
Category
II/
Sampling
and
Analysis
Draft

Revision
1.2
U.
S.
Environmental
Protection
Agency
Air
Pollution
Prevention
and
Control
Division
March
2003
Contract
No.
68­
C99­
201
Work
Assignment
No.
4­
38
Project
No.
RN992014.0038
Prepared
for:

U.
S.
Environmental
Protection
Agency
Air
Pollution
Prevention
and
Control
Division
Research
Triangle
Park,
NC
27711
EPA
Work
Assignment
Manager
__________________________
_______
Mark
Mason
Date
ARCADIS
Geraghty
&
Miller
Work
Assignment
Leader
__________________________
_______
Victor
D

Amato
Date
EPA
QA
Representative
__________________________
_______
Paul
Groff
Date
ARCADIS
Geraghty
&
Miller
QA
Officer
__________________________
_______
Laura
Beach
Nessley
Date
Section
1
Draft
­
Revision
1.2
March
2003
Page
ii
of
9
i
CONTENTS
Section­
Page
1
Project
Description
and
Organization
1­
1
1.1
Overall
Project
Objectives
1­
1
1.2
Background
1­
1
1.3
Experimental
Design,
Scope,
and
Limitations
1­
3
1.3.1
Screening
Study
1­
4
1.3.2
Accelerated
Weathering
Testing
(
Weathering
Chambers)
1­
9
1.3.3
Natural
Weathering
Tests
1­
12
1.3.4
Other
Overall
Study
Limitations
1­
13
1.4
Data
Quality
Objectives
1­
17
1.5
Project
Organization
and
Responsibilities
1­
17
2
Sampling
Approach
2­
1
2.1
Preparation
and
Characterization
of
Wood
Specimens
2­
1
2.1.1
Wood
for
Indoor,
Accelerated,
Artificial
Weathering
Tests
2­
3
2.1.2
Wood
for
Outdoor,
Natural
Weathering
Tests
2­
4
2.2
Selection
of
Coatings
2­
5
2.3
Coating
Application
2­
5
2.3.1
Indoor
Accelerated
Weathering
Tests
2­
6
2.3.2
Outdoor,
Natural
Weathering
Tests
2­
7
2.4
Weathering
Protocols
2­
8
2.4.1
Accelerated
Indoor
Weathering
2­
8
2.4.2
Outdoor
Weathering
2­
12
2.5
Sampling
2­
14
2.5.1
Wipe
Sampling
2­
14
2.5.2
Wood
Sampling
2­
15
2.5.3
Photographs
2­
15
2.5.4
Miscellaneous
Samples
2­
15
3
TESTING
AND
MEASUREMENT
PROTOCOLS
3­
1
3.1
Wipe
Sampling
3­
1
3.2
Sample
Preparation
(
Digestion)
3­
1
3.3
Analysis
by
ICP­
MS
3­
2
3.4
Preparation
and
Analysis
of
Coating
Samples
3­
2
3.5
Preparation
and
Analysis
of
Wood
Samples
3­
2
3.6
Archiving
of
ICP­
MS
Samples
3­
3
3.7
Moisture
Analysis
of
Wood
Specimens
3­
3
ii
4
QA/
QC
Checks
4­
1
5
Data
Reductions
and
Reporting
5­
1
5.1
Data
Reduction
5­
1
5.1.1
Calculation
of
DA
from
Extraction/
Digestion
Fluid
Concentrations
5­
1
5.1.2
Calculation
of
Percent
Reduction
of
DA
5­
1
5.1.3
Assessing
DQI
Goals
5­
2
5.2
Data
Validation
5­
3
5.3
Data
Reporting
5­
3
5.4
Relational
Database
Development
5­
3
5.5
Regular
Reporting
5­
4
6
Assessments
6­
1
Appendix
A:
Test
Plan:
Screening
and
Selection
of
Coatings
for
Reducing
Dislodgeable
Arsenic
from
CCA
Treated
Wood
Appendix
B:
List
of
Key
References
List
of
Figures
Section­
Page
Figure
1.2.1.
Wood
Board/
Specimen
Nomenclature
1­
2
Figure
1.5.1.
Organizational
Chart
for
Sampling/
Analytical
Qualification
Testing
1­
18
Figure
2.1.2.
Example
specimen
identification
scheme
2­
2
Figure
2.1.3.
Schematic
of
Mini­
Deck
Construction
2­
5
List
of
Tables
Section­
Page
Table
1.3.1.
Summary
of
Study
Designs
1­
4
Table
1.3.2.
Selected
Products
for
Screening
Study
1­
7
Table
1.4.1.
Data
Quality
Indicator
Goals
for
Critical
Measurements
1­
17
Table
1.5.1.
Contact
Information
for
Key
Project
Staff
1­
19
Table
2.1.2.
Schedule
of
Mini­
Decks
and
Specimens
for
Outdoor,
Natural
Weathering
Tests
2­
4
Table
2.3.1.
Summary
of
Coating
Application
Measurements
2­
7
Table
2.4.1.
Test
Sequence
for
Chamber
Weathering
Tests
2­
11
Table
2.4.2.
Weather
data
for
outdoor
testing.
2­
13
iii
Table
2.5.1.
Miscellaneous
Samples
to
be
Collected
during
Screening
Study
2­
16
iv
List
of
Acronyms
ACW
Applied
coating
weight
ANOVA
Analysis
of
variance
CCA
Chromated
copper
arsonate
CPSC
Consumer
Product
Safety
Commission
DA
Dislodgeable
arsenic
DCSW
Dry
coated
specimen
weight
DCW
Dry
coating
weight
DQI
Data
quality
indicator
ECW
Ending
coating
weight
FPL
Forest
Products
Lab
IPN
Interpolymer
network
coating
MS/
MSD
Matrix
spikes
and
matrix
spike
duplicates
NCDC
National
Climate
Data
Center
NOAA
National
Oceanic
and
Atmospheric
Administration
PEA
Performance
evaluation
audit
QA/
QC
Quality
Assurance/
Quality
Control
RSD
Relative
standard
deviation
SCW
Starting
coating
weight
USDA
US
Department
of
Agriculture
USW
Uncoated
specimen
weight
WCSW
Wet
coated
specimen
weight
WCW
Wet
coating
weight
WFT
Wet
film
thickness
v
(
Section­
Page)
1
of
20
Draft
­
Revision
1.2
March
2003
1
Project
Description
and
Organization
1.1
Overall
Project
Objectives
The
primary
objective
of
this
project
is
to
evaluate
the
ability
of
selected
coatings
to
reduce
the
amount
of
dislodgeable
arsenic
(
DA)
on
the
surfaces
of
chromated
copper
arsenate
(
CCA)
treated
wood
(
if
resources
permit,
dislodgeable
chromium
may
also
be
measured).
Selected
coatings
will
be
applied
to
new
and
weathered
CCA
treated
southern
yellow
pine.
The
coated
lumber
will
be
subjected
to
accelerated
weathering
in
chambers
and
to
natural
weathering
out
of
doors.
The
ability
of
the
coatings
to
reduce
DA
will
be
evaluated
periodically
by
determining
the
amount
of
arsenic
removed
from
the
surface
of
the
wood
specimens
using
a
wipe
technique
as
the
wood
and
coatings
weather.
For
the
purposes
of
this
study,
DA
is
defined
as
the
amount
of
arsenic
(
ìg/
cm2)
removed
from
the
surface
of
the
test
specimen
by
the
dermal
wipe
procedure
developed
and
demonstrated
by
the
Consumer
Product
Safety
Commission
(
CPSC),
who
are
collaborators
on
this
project
via
an
interagency
agreement
(
CPSC­
I­
03­
1235)
between
EPA
and
CPSC.

The
data
obtained
will
be
used
by
EPA
and
CPSC
staff
in
support
of
efforts
to
inform
the
public
regarding
the
use
and
maintenance
of
existing
CCA­
treated
wood
products,
such
as
decks
and
playground
equipment.
A
supplemental
objective
of
this
study
is
to
evaluate
and
demonstrate
the
use
of
the
test
protocols
and
to
begin
to
understand
their
utility/
realism,
and
to
identify
future
research
needs.
This
second
objective
is
relevant
because
there
are
currently
no
standardized
protocols
for
the
use
of
accelerated
weathering
chambers
to
determine
the
efficacy
of
coatings
to
reduce
DA
from
CCA
treated
wood.
In
that
regard,
this
is
a
pilot
study
that
may
set
the
stage
for
systematic
development
of
standardized
test
methods
that
will
promote
development,
evaluation,
and
demonstration
of
products
that
mitigate
the
potential
for
dermal
contact
with
dislodged
arsenic
from
CCA
treated
wood.

Note
that
very
few
products
are
currently
manufactured
for
the
purpose
of
reducing
DA
from
CCA
treated
wood.
In
this
regard,
EPA
is
evaluating
the
efficacy
of
products
to
perform
a
task
that
is
not
necessarily
related
to
the
manufacturer

s
design
or
intent.
As
such,
the
test
results
should
not
be
construed
to
represent
an
evaluation
of
a
product

s
effectiveness
for
those
purposes
for
which
it
was
designed
and
warranted
by
the
manufacturer.
(
Section­
Page)
2
of
20
Draft
­
Revision
1.2
March
2003
1.2
Background
CCA
is
a
chemical
wood
preservative
injected
under
pressure
to
protect
wood
from
decay
and
insect
damage.
The
manufacturers
of
CCA
treated
wood
have
asked
EPA
to
remove
registration
of
this
product
for
residential
use,
including
playground
equipment,
decks,
and
landscape
timbers,
and
they
intend
total
conversion
to
alternative
treatments
by
December
31,
2003.
However,
there
remains
potential
for
dermal
contact
with
arsenic
and
chromium
residues
on
treated
surfaces,
and
the
potential
is
greatest
for
the
most
susceptible
subpopulation,
infants
and
small
children,
due
to
their
close
contact
with
surfaces
and
hand
to
mouth
activities.
A
recent
field
survey
of
CCA
treated
surfaces
indicated
that
widely
used
deck
sealants
are
often
not
effective
at
preventing
arsenic
contamination
of
surfaces
beyond
six
months.
This
project
will
provide
EPA
with
information
that
can
be
used
to
provide
the
public
with
guidance
on
the
use
of
coatings
to
prevent
contact
with
arsenic
dislodged
from
CCA
treated
wood.

To
provide
consumers
with
effective
guidance,
EPA
must
have
a
basic
understanding
of
the
impact
of
key
variables
on
the
efficacy
of
coating/
sealant
systems.
Key
environmental
variables
include
exposure
to
natural
weathering
phenomena
including
UV
radiation,
condensation,
precipitation,
and
thermal
shock.
Efficacy
of
coatings
may
also
be
impacted
by
level
and
fixation
of
CCA
treatment,
age
and
condition
of
the
wood
at
the
time
of
coating,
and
type
and
dimensions
of
the
treated
wood.
Due
to
the
large
number
of
variables,
and
EPA

s
desire
to
provide
guidance
quickly
for
new
as
well
as
aged,
in­
service
wood,
this
project
will
evaluate
selected
coatings
applied
to
new
and
aged
CCA
treated
wood
using
accelerated
weathering
chambers
located
at
EPA

s
RTP
facilities
and
using
outdoor
natural
weathering
at
a
site
in
North
Carolina.
The
purpose
of
the
chamber
weathering
tests
is
to
accelerate
weathering
processes
so
that
the
impact
of
weathering
on
efficacy
of
coatings
in
reducing
DA
can
be
evaluated
in
a
relatively
short
time
period
(
less
than
one
year).

Approximately
30
sealants
covering
a
wide
range
of
product
types
will
have
been
screened
in
simple
weeklong
tests
on
new
CCA
treated
wood
in
accordance
with
the
test
plan
provided
in
Appendix
A.
Using
the
results
of
this
screening
study,
ten
coatings
will
be
selected
for
use
during
the
indoor,
accelerated,
and
outdoor,
natural
weathering
studies
described
in
this
test
plan.

Before
proceeding,
it
is
essential
to
define
terminology
as
applied
in
this
test
plan.
Wood
nomenclature
used
in
this
test
plan
is
defined
in
Figure
1.2.1.
Note
that
a

board

is
defined
as
the
unit
of
wood
purchased
or
removed
from
an
existing
structure,
while

specimen

refers
to
the
pieces
of
each
board
cut
for
this
project
(
note
that

specimens

are
sometimes
called

coupons

in
accelerated
weathering
testing).
(
Section­
Page)
3
of
20
Draft
­
Revision
1.2
March
2003
0
Figure
1.2.1.
Wood
Board/
Specimen
Nomenclature
1.3
(
Section­
Page)
4
of
20
Draft
­
Revision
1.2
March
2003
Experimental
Design,
Scope,
and
Limitations
Overall
project
objectives
will
be
achieved
through
three
integrated
experimental
efforts:

|
a
screening
study
to
inform
the
process
of
selecting
coatings
to
be
tested
via
weathering
testing,

|
accelerated
weathering
testing
(
using
indoor
test
chambers)
of
selected
coatings,

|
natural
outdoor
weathering
testing
of
the
same
selected
coatings
The
intent
of
the
screening
study
is
to
test
a
wide
variety
of
potentially­
applicable
wood
coating
products
for
their
efficacy
in
reducing
DA
in
a
short­
duration
test
utilizing
conditioned
new
wood.
The
results
of
the
screening
study
will
be
used
to
select
coatings
to
be
further
tested
in
accelerated
weathering
chambers
and
via
natural
outdoor
weathering
tests.

Accelerated
weathering
testing
using
weathering
chambers
is
seen
as
a
potentially
useful
tool
for
evaluating
CCA­
treated
wood
coatings.
Weathering
tests
will
also
be
conducted
using
small
decks
(
mini­
decks)
which
will
be
exposed
to
natural
weathering
conditions
outdoors
at
a
site
in
North
Carolina.
The
results
of
these
two
weathering
tests
will
primarily
be
used
to
evaluate
and
rank
tested
coatings
for
their
efficacy
in
reducing
DA.
Additionally,
a
comparison
of
the
results
of
the
accelerated
and
natural
weathering
studies
will
provide
information
on
the
usefulness
of
weathering
chamber
testing
for
determining
CCA­
treated
wood
coating
efficacy.
In
this
sense,
the
project
can
be
thought
of
as
a
pilot
study:
it
is
hoped
that
the
results
gained
through
its
execution
not
only
support
EPA

s
goals
of
evaluating
and
reducing
risk
of
contact
with
arsenic
dislodged
from
CCA­
treated
wood,
but
also
provide
a
framework
of
methodology
to
inform
the
design
of
future
studies,
in
addition
to
identifying
areas
needing
future
study.
It
is
important
to
note
that
the
study
designs
proposed
for
the
three
testing
components
have
been
developed
with
an
eye
toward
consistency,
so
that
the
results
of
each
can
be
compared
via
similar
statistical
analyses.
A
brief
summary
of
the
study
design
of
the
three
tests
proposed
is
provided
as
Table
1.3.1.

While
the
primary
objective
of
the
testing
described
herein
is
to
evaluate
coatings
for
their
efficacy
in
reducing
DA
when
coated
wood
is
subjected
to
weathering,
available
resources
are
limited
and
dictate
that
the
project
be
focused
in
a
way
that
precludes
the
ability
to
answer
all
of
the
myriad
questions
raised
in
the
development
and
evaluation
of
these
test
plans.
Difficult
choices
have
had
to
be
made
in
a
number
of
important
areas
in
order
to
meet
the
resource
and
time
constraints
posed
by
this
project.
The
objective
of
the
following
discussion
in
this
section
is
to
better
define
the
scope
of
the
proposed
test
plan,
its
limitations,
and
unanswered
questions
that
may
be
applicable
as
a
focus
for
future
research
work.
Where
applicable,
discussion
of
such
limitations
and
decision
rationale
are
additionally
included
in
the
text
of
sections
that
follow
this
introduction.
(
Section­
Page)
5
of
20
Draft
­
Revision
1.2
March
2003
Table
1.3.1.
Summary
of
Study
Designs
Screening
Study
Chamber
Tests
Outdoor
Tests
#
Coatings
~
30
10
10
#
Wood
Sources
1
(
new)
2
(
new,
aged)
2
(
new,
aged)

#
Replicates
3
specimens
3
specimens
3
specimens
#
Uncoated
Controls
>
90
(
to
establish
baseline)
6
(
3
new,
3
aged)

(
not
including
baseline)
6
(
3
new,
3
aged)

(
not
including
baseline)

#
Untreated
Blanks
3
specimens
6
specimens
(
1
per
chamber)
3
specimens
Sampling
Frequency
After
one
week
Monthly
for
selected
coatings
Monthly
Monthly
Total
#
Wipe
Samples
(
not
incl.
baseline)
93
(
one
time)

33/
month
72/
month
69/
month
1.3.1
Screening
Study
Objective
The
objective
of
the
screening
study
is
to
provide
a
limited
evaluation
of
a
variety
of
commerciallyavailable
coatings
for
their
apparent
initial
efficacy
in
reducing
DA
from
the
surfaces
of
new
CCA­
treated
wood
in
a
relatively
simple,
short­
duration
test.

The
results
of
this
screening
evaluation
will
be
used
to
select
coating
products
to
be
more
thoroughly
tested
via
the
larger
pilot
study
to
determine
the
efficacy
of
coatings
to
reduce
DA
from
the
surfaces
of
new
and
aged,
previously
uncoated
CCA
treated
wood
subjected
to
both
accelerated,
indoor,
artificial,
and
outdoor,
natural
weathering
conditions.

Scope
About
30
coatings
will
be
applied
to
46
cm
(
18

)
new
conditioned
CCA­
treated
wood
specimens
in
triplicate.
Wipe
sampling
will
be
conducted
at
7
days
after
coating
application.
Post­
coating
DA
will
be
compared
with
baseline
DA
established
for
each
individual
specimen
to
determine
percent
reductions
in
DA
for
each
specimen.
Individual
baseline
values
of
DA
will
be
determined
for
each
specimen
to
be
coated
and
tested.
The
baseline
DA
of
a
specimen
will
be
determined
by
averaging
the
DAs
from
the
two
adjacent
46­
cm
specimens
on
either
side
of
the
test
specimen.
The
adjacent
specimens
will
be
archived
uncoated
after
sampling
(
and
thus
not
used
in
any
further
testing).
This
step
will
avoid
any
data
analysis
and
coating
efficacy
complications
that
may
arise
from
coating
pre­
rubbed
specimens.
(
Section­
Page)
6
of
20
Draft
­
Revision
1.2
March
2003
Coatings
will
be
ranked
based
upon
initial
DA
reduction
and
10
high­
ranking
coatings
will
be
selected
for
weathering
testing.
The
screening
test
specimens
for
the
selected
coatings
will
be
retained
for
continued
monthly
DA
measurement
during
the
weathering
tests,
as
non­
weathered
controls.
These
samples
will
allow
for
the
monitoring
of
DA
changes
resulting
from
abrasion
(
from
wipe
sampling)
without
the
potentially
confounding
effects
of
weathering.

The
screening
study
is
described
in
a
separate
test
plan
entitled,

Screening
and
Selection
of
Coatings
for
Reducing
Dislodgeable
Arsenic
from
CCA
Treated
Wood

,
which
is
provided
as
Appendix
A.

Data
Product
and
Use
Average
post­
coat
DA
percent
reductions
will
be
calculated
for
each
coating/
wood
type
combination
using
the
individual
DA
results
for
each
of
the
triplicate
specimens
of
each
coating
on
each
of
two
wood
types
(
new
and
aged).
Percent
reduction
in
DA
will
be
calculated
for
each
test
specimen
as
the
percent
difference
between
the
baseline
and
coated
DA
measurements.
The
mean
percent
reduction
of
the
three
measurements
for
each
coating
will
represent
the
initial
DA
reduction.
Coatings
will
be
ranked
based
upon
initial
DA
reduction
and
10
high
ranking
coatings
will
be
selected
for
weathering
testing,
per
the
following
criteria:

|
Overall
ranking
in
terms
of
DA
reduction
|
Subjective
factors,
including:


Ensuring
that
the
final
list
of
selected
coatings
is
representative
of
the
range
of
potentially
effective
products
on
the
market.


Other
factors
which
may
be
identified
as
important
during
screening
tests.

The
coating
ranking
data
will
also
be
compared
with
the
results
of
the
accelerated
and
natural
weathering
tests
to
provide
insight
into
the
applicability
of
the
initial
screening
study
in
predicting
longer­
term
coating
performance.

Limitations
Selection
of
Products
to
be
Screened.
The
selection
of
coatings
to
be
tested
for
efficacy
is
obviously
critical
and
because
of
the
number
and
variety
of
potentially­
applicable
coatings
on
the
market
and
the
budgetary
constraints
of
testing
programs,
this
is
likely
to
be
a
weakness
of
any
such
evaluation
study
to
be
conducted.
The
original
plan
for
this
project
did
not
explicitly
include
a
coatings
screening
phase
but
the
project
team
identified
the
screening
test
to
be
a
needed
addition
despite
its
limitations.
To
put
the
task
into
perspective,
the
goal
of
selecting
coatings
is
to
distill
a
universe
of
hundreds
or
even
thousands
of
potentially
applicable
coatings
to
approximately
10
to
be
fully
tested
via
the
weathering
testing
protocol.
While
well
beyond
the
scope
of
this
project,
a
thorough
review
of
available
coatings
and
their
formulations
(
Section­
Page)
7
of
20
Draft
­
Revision
1.2
March
2003
and
application
techniques
is
needed
to
more
completely
understand
the
characteristics
that
impact
surface
concentrations
of
CCA­
treated
wood
analytes
(
this
could
include
more
focused
involvement
by
the
wood
coating
industry).
For
this
project
however,
the
approach
is
to
gather
basic
formulation
and,
to
a
lesser
extent,
application
information,
on
a
number
of
products
with
reasonable
availability
to
the
project
team
in
North
Carolina
(
project
site).
This
survey
of
available
products
was
primarily
conducted
using
internet
searches
and
visits
to
local
retail
hardware
and
home
improvement
stores.
These
searches
allowed
for
the
development
of
a

master
list

of
specific
products.
This
master
list
of
potential
products
includes
approximately
125
entries,
including
some
that
are
broadly
intended
for
outdoor
wood
use,
as
well
as
some
products
that
are
not
necessarily
intended
for
such
uses,
but
that
were
identified
by
the
project
team
as
promising.

The
list
is
in
spreadsheet
format
and
includes
fields
for
manufacturer,
product
name,
product
type,
cover,
base,
and
main
ingredient.
It
must
be
noted
that
there
are
various
levels
of
classifications
for
coatings
and
that
no
single
categorization
standard
can
be
applied
to
adequately
categorize
each
and
every
product
identified.
Additionally,
many
products
overlap
categories.
Nevertheless,
in
order
to
communicate
effectively
about
the
products
tested,
and
maintain
the
confidentiality
of
product
names,
an
attempt
has
been
made
to
classify
the
products
considered.
As
such,
several
main
descriptors
of
coatings
were
used.
These
include:
base
(
oil
vs.
water),
cover
(
clear,
semi­
transparent,
opaque),
and
product
type,
which
for
this
exercise,
has
been
broken
out
into
the
following:
paints,
primers,
sealants,
stains,
and
other.
The

other

category
embodies
a
vast
variety
of
products,
including,
but
not
limited
to:
varnishes,
epoxies,
lead
encapsulation
products,
rubber
coatings,
fiberglass
coatings,
elastic
vinyl
coatings,
preservatives,
and
other
plastic
coatings.
Additional
classification
descriptors
include:
ingredients
and
surface
(
penetrating
vs.
filmforming

The
master
list
of
about
125
products
includes
roughly
25
paints,
5
primers,
20
wood
sealants,
50
stains,
and
25

other

products.
Out
of
the
paints,
approximately
2/
3
are
water­
based
with
the
balance
oil­
based.
Likewise,
for
the
primers,
2
are
oil­
based
while
3
are
water­
based.
For
the
wood
sealants
and
stains,
most
products
are
oil­
based
with
a
handful
water­
based.
The
cover
for
each
of
these
product
types
is
quite
variable,
and
in
fact,
one

type

of
coating
may
be
available
in
a
range
of
covers
from
clear
to
opaque
(
note
that
existing
research
on
coating
efficacy
suggests
that
opaque
coatings
may
be
more
effective).
Likewise,
the
surface
for
each
of
the
listed
product
types
may
also
be
variable,
depending
on
the
product
(
note
that
existing
research
on
coating
efficacy
suggests
that
film­
forming
coatings
may
be
more
effective,
though
they
may
also
be
more
subject
to
deterioration
via
abrasion).
Paints
and
primers
will
almost
invariably
be
considered
film­
forming
products,
while
sealants,
stains,
and
certainly

other

products
may
be
penetrating
or
film­
forming
depending
on
their
specific
formulation.

From
the
master
list,
35
distinct
products
have
been
selected
based
on
the
following
criteria:

1.
Products
that
are
commonly
used
for
outdoor
wood
treatment,
with
preference
given
to
those
that
either
(
Section­
Page)
8
of
20
Draft
­
Revision
1.2
March
2003
have
been
tested
and/
or
identified
as
promising
by
other
researchers
and
to
those
that
have
performed
well
in
durability
testing
by
Consumer
Reports
magazine.

2.
Products
that
are
typically
not
used
for
outdoor
wood/
deck
treatment,
but
that
have
been
identified
as
having
the
potential
to
meet
efficacy
objectives.

3.
Products
that
are
relatively
straightforward
for
consumers
to
apply
(
i.
e.,
products
that
require
professional
application
have
been
disqualified).
Additionally,
multiple
product
systems
have
generally
not
been
considered,
although
it
is
recognized
that
some
common
products
(
e.
g.,
paints)
may
require
the
application
of
another
product
as
a
primer.
These
situations
will
be
considered
on
a
case­
by­
case
basis.

Table
1.3.2
generically
(
to
preserve
required
product
confidentiality)
lists
and
characterizes
35
products
selected
for
consideration
for
the
screening
study.
Since
it
is
likely
that
several
products
may
be
unavailable
or
require
unusual
or
difficult
coating
methods
not
known
at
this
time,
it
is
anticipated
that
only
30
products
will
be
screening
tested.

Table
1.3.2.
Selected
Products
for
Screening
Study
#
Product
Type
Base
Cover
Comments
1
Paint
Oil
Opaque
Silicone,
alkyd.
Very
durable
marine
finish.
2
Paint
Oil
Opaque
Silicone,
alkyd.
Very
durable
marine
finish.
3
Paint
Water
Opaque
Acrylic
latex.
Designed
for
porches
and
floors.
4
Paint
Water
Opaque
Acrylic
latex.
Exterior.
5
Paint
Water
Opaque
Acrylic
latex.
Exterior.
6
Paint
Oil
Opaque
Alkyd
polyurethane.
Designed
for
porches
and
floors.
7
Paint
Oil
Opaque
Alkyd
polyurethane.
Designed
for
porches
and
floors.
8
Paint
Oil
Opaque
Alkyd
exterior
paint
product
9
Primer
Water
Opaque
Common
latex
primer.
10
Sealant
Oil
Semi
Common
deck
product.
11
Sealant
Oil
Clear
Common
deck
product.
12
Sealant
Oil
Clear
Common
deck
product.
13
Sealant
Oil
Clear
Acrylic/
alkyd/
urethane,
for
decks.
14
Sealant
Water
Clear
Common
deck
product.
15
Sealant
Oil/
Wate
r
Semi
Water,
linseed
oil,
tung
oil.

16
Stain
Oil
Opaque
Gilsonite.
Tar
product.
17
Stain
Opaque
Teflon
acrylic.
18
Stain
Water
Semi
Acrylic
latex.
19
Stain
Water
Semi
Common
deck
product.
Acrylic.
20
Stain
Oil
Semi
Common
oil
alkyd
deck
product.
(
Section­
Page)
9
of
20
Draft
­
Revision
1.2
March
2003
#
Product
Type
Base
Cover
Comments
21
Stain
Oil
Clear
Common
oil
alkyd
deck
product.
22
Stain
Oil
Semi
Contains
UV
absorber.
23
Stain
Oil
Semi
Common
oil
alkyd
resin
deck
product
with
linseed
oil.
24
Stain
Oil
Semi
Oil
alkyd/
acrylic
for
deck,
fence,
and
siding.
25
Stain
Oil
Semi
Oil
alkyd/
acrylic
for
decks.
26
Stain
Oil
Semi
Oil
acrylic.
27
Stain
Oil
Semi
Oil
resin
for
external
wood
products.
28
Stain
Oil
Opaque
Acrylic
color
deck
stain.
29
Stain
Water
Opaque
Common
acrylic
latex
deck
product.
30
Other
Spar
varnish.
31
Other
Interpolymer
network
coating
(
IPN).
32
Other
Urethane.
May
require
professional
application.
33
Other
Elastic
vinyl.
Designed
for
CCA
encapsulation.
34
Other
Polymer.
Designed
for
CCA
encapsulation.
35
Other
Acrylic
mastic.
Designed
for
lead
encapsulation.

Study
Duration.
Another
key
limitation
of
the
screening
study
is
that
it
will
only
test
the
initial
efficacy
of
the
coating.
No
aging
or
weathering
of
the
coating
and/
or
wood
will
be
tested
in
the
screening
study,
primarily
due
to
the
tight
time
schedule
under
which
the
study
is
being
conducted.
It
is
unknown
to
what
degree,
if
any,
initial
efficacy
and
longer­
term
efficacy
relate
or
whether
the
screening
procedure
will
provide
large
enough
differences
between
coatings
to
allow
for
effective
ranking.

Substrate
Selection.
Only
new
CCA­
treated
wood
will
be
tested
in
this
phase
of
study,
even
though
both
new
and
aged
woods
shall
be
tested
in
the
weathering
experiments.

Application
Technique.
It
is
possible
that
the
method
of
applying
coatings
may
contribute
to
measured
DA
levels.
For
example,
applying
coating
using
a
brush
may
cause
physical
displacement
of
dislodged
analytes
and
subsequent
mixing
with
the
applied
coating
and/
or
displacement
of
the
analyte
to
the
finished
coated
surface.
As
such,
a
pre­
qualification
study
to
evaluate
coating
application
techniques
(
e.
g.,
brush
versus
spray)
was
considered
for
inclusion
as
part
of
the
screening
test
plan,
but
later
determined
to
be
outside
of
the
scope
and
resource
allocation
available
for
this
project.
Coatings
will
be
applied
per
manufacturer

s
instructions.
When
a
choice
is
given
of
application
method
is
given
by
the
manufacturer,
brush
application
will
be
used.
(
Section­
Page)
10
of
20
Draft
­
Revision
1.2
March
2003
1.3.2
Accelerated
Weathering
Testing
(
Weathering
Chambers)

Objective
The
primary
objective
of
the
accelerated
weathering
tests
is
to
evaluate
the
effects
of
simulated
weathering
in
a
controlled
environment
on
the
efficacy
of
selected
coating
products
in
reducing
the
DA
from
the
surface
of
new
and
aged
CCA
treated
wood.
A
secondary
objective
will
be
to
evaluate
the
utility
of
such
an
artificial
weathering
test
in
adequately
simulating
outdoor
weathering.

Scope
Ten
(
10)
coatings
which
have
been
selected
via
the
screening
study
previously
described
will
be
applied
to
30­
cm
(
12

)
new
and
aged
CCA­
treated
wood
specimens
in
triplicate.
These
coated
specimens
along
with
uncoated
controls
(
both
untreated
and
CCA­
treated)
will
be
exposed
to
accelerated
weathering
conditions
in
QUV/
Spray
weathering
chambers.
Six
test
chambers
will
be
used
in
a
split­
plot
study
design:
three
chambers
will
contain
new
wood
specimens
while
the
other
three
will
contain
aged
wood
specimens.
Each
specimen
in
each
chamber
will
be
coated
with
one
of
the
10
coatings
selected
per
screening
studies.
Additionally,
one
CCA­
treated,
uncoated
control
sample
and
one
untreated,
uncoated
blank
sample
will
be
included
in
each
chamber.
The
position
of
each
specimen
within
each
chamber
will
be
randomized
at
the
start
of
the
test
and
after
monthly
sampling
events.
DA
(
via
wipe
sampling)
will
be
measured
monthly.
Monthly
DA
results
will
be
compared
with
baseline
DA
for
each
specimen,
determined
by
averaging
DA
results
from
adjacent
specimens,
as
indicated
for
the
screening
study
baseline
determination.

Factors
that
impact
efficacy
of
coatings
which
will
be
evaluated
via
the
artificial
weathering
chamber
experiments
include:

|
UV
degradation
|
Condensation
|
Thermal
shock
|
Erosion
due
to
precipitation
The
proposed
accelerated
weathering
study
is
described
in
more
detail
later
in
this
test
plan.
(
Section­
Page)
11
of
20
Draft
­
Revision
1.2
March
2003
Data
Product
and
Use
The
efficacy
of
each
coating
at
reducing
DA
from
new
and
aged
CCA­
treated
wood
will
be
evaluated
as
a
function
of
time
in
the
weathering
chambers.

Post­
coat
DA
will
be
determined
on
triplicate
specimens
of
each
coating
on
each
of
two
wood
types
(
new
and
aged).
Percent
reduction
in
DA
will
be
determined
monthly,
calculated
as
the
difference
between
baseline
DA
and
that
month

s
DA
measurement.

Weathered
coatings
will
be
ranked
monthly
according
to
their
efficacy
based
on
percentage
reduction
of
DA
from
new
and
aged
CCA
treated
wood.

Although
the
observed
character
of
the
data
must
affect
the
specific
types
of
analysis,
it
is
expected
that
the
following
statistical
methods
will
be
employed:

Analysis
of
data
for
each
given
month.
A
variety
of
analysis
of
variance
(
ANOVA)
studies
will
be
undertaken,
determined
by
the
types
of
model
likely
to
be
valid
from
preliminary
data
examination.
It
is
expected
that
these
will
include
Display
of
the
data
and
calculation
of
summary
statistics
for
each
coating
/
wood
type
combination
for
the
purpose
of
checking
the
assumption
of
constant
variation
among
treatment
combinations,
and
identifying
appropriate
transformations
(
e.
g.
logarithmic)
as
needed.

A
separate
1­
way
ANOVA
of
coatings
for
each
wood
type
with
chambers
entering
as
blocks.
These
analyses
will
provide
additional
information
on
the
validity
of
the
assumption
of
constant
variance
between
wood
types.

A
full
analysis
of
the
3­
replicate
split­
plot
design
to
include
all
coatings,
wood
types
(
and
chambers).
This
will
provide
the
most
complete
and
detailed
conclusions
regarding
e.
g.
coatings
and
their
interactions
with
wood
types,
again
assuming
the
prior
calculations
indicate
model
validity.

Time
histories
of
degradation
A
similar
approach
will
be
taken
to
model
the
changes
in
coating
efficacies
over
the
study
time
period,
involving
linear
and
non­
linear
regressions
as
suggested
by
the
appearance
of
the
data.
The
very
simplest
of
these
­
individual
coating
histories
for
individual
wood
types
(
accounting
for
chamber
differences)
­
seem
likely
to
be
the
most
informative
and
useful.
However,
more
complex
(
multivariate)
analyses
involving
e.
g.
both
wood
types
or
two
or
more
coatings,
will
be
undertaken
as
perceived
desirable
for
applications.

These
same
coatings
will
also
be
similarly
tested
outdoors
on
miniature
decks
in
order
to
allow
for
(
Section­
Page)
12
of
20
Draft
­
Revision
1.2
March
2003
comparisons
to
be
made
between
artificial
and
natural
weathering
and
their
respective
impacts
on
the
efficacy
of
selected
coatings.

Limitations
Accelerated
weathering
chambers.
There
are
currently
no
accepted
standardized
accelerated
weathering
protocols
for
evaluating
the
efficacy
of
coatings
to
reduce
DA
from
CCA­
treated
wood.
Thus,
the
chamber
weathering
experiments
are
exploratory
in
nature.
The
accelerated
weathering
chambers
are
artificial
environments
that
permit
researchers
to
accelerate
certain
aspects
of
weathering.
Prediction
of
service
life
from
chamber
test
data
requires
extensive
method
development
and
validation
with
field
tests.
While
this
is
generally
beyond
the
scope
of
this
project,
some
correlation
with
natural
outdoor
weathering
results
can
be
made,
thus
allowing
some
preliminary
assessment
of
the
applicability
of
chamber
testing
for
evaluating
CCA
wood
coating
efficacy.
Additionally,
the
data
generated
in
this
project
may
be
used
to
inform
future,
more
rigorous
efforts
to
develop
a
standardized
accelerated
weathering
approach
for
evaluating
coating
efficacy.

There
are
many
reasons
that
accelerated
weathering
in
chambers
does
not
completely
simulate
natural
weathering
conditions.
These
include,
but
are
not
limited
to:

|
UVA­
340
nm
lamps
are
specific
to
one
specific
wavelength,
not
the
full
spectrum
of
sunlight
|
DI
water
is
used
for
spraying;
acid
rain
is
not
simulated.

|
Condensation
and
UV
cycles
are
executed
at
much
higher
temperatures
than
typically
encountered
Cut
end
leaching.
Published
literature
suggests
that
cut
ends
tend
to
contain
higher
surface
and
leachate
concentrations
of
total
arsenic
than
the
faces
or
edges
of
the
same
specimen.
For
this
study,
the
ratio
of
cut
end
surface
area
to
face
surface
areas
is
relatively
high
compared
to
typical
applications.
As
such,
there
is
a
potential
for
relatively
high
levels
of
arsenic
to
leach
out
of
the
ends
of
each
specimen.
This
is
of
particular
concern
in
the
accelerated
weathering
chamber
tests
because
of
the
orientation
of
the
specimens
within
the
chambers:
the
specimens
will
be
nearly
vertical,
leaning
slightly
forward.
Such
an
orientation
appears
to
have
the
potential
to
artificially
increase
top
(
front)
face
surface
concentrations
of
CCA
analytes,
as
leachate
from
the
cut
end
at
the
top
of
the
specimen
could
leach
out
the
end
and
then
down
the
front
face
of
the
specimen.

Various
options
have
been
considered
to
deal
with
this
potential
problem.
One
option
that
was
given
serious
consideration
was
to
use
a
consistent
coating
(
Liquid
Nails
has
been
used
by
other
researchers
testing
leaching
of
CCA
wood
analytes)
on
each
cut
end
of
each
specimen.
To
be
consistent
among
coatings,
cut
ends
will
be
coated
using
the
same
coating
applied
to
the
faces
of
each
specimen.
Similar
precautions
(
that
is,
sealing
the
cut
ends
with
the
coating
being
tested)
have
been
employed
for
the
screening
and
outdoor
weathering
studies,
though
leaching
and
redistribution
of
CCA
analytes
is
not
(
Section­
Page)
13
of
20
Draft
­
Revision
1.2
March
2003
expected
to
be
as
great
a
problem,
since
the
screening
and
outdoor
weathering
specimens
will
typically
be
oriented
horizontally
with
their
top
faces
pointed
upwards.

Cross­
Contamination.
CPSC
has
conducted
some
preliminary
cross­
contamination
tests
using
the
QUV/
Spray
weathering
chambers
to
be
used
in
the
artificial
weathering
study.
These
results
suggest
that
cross­
contamination
of
samples
in
the
QUV/
Spray
chambers
is
not
significant.
However,
the
CPSC
crosscontamination
data
is
based
on
relatively
short
duration
testing.
EPA
and
ARCADIS
will
further
test
for
cross­
contamination
during
QUV/
Spray
shakedown
and
startup
tasks.
Furthermore,
sample
dividers
are
being
fabricated
and
tested
to
prevent
splashing
of
spray
(
precipitation)
water
from
one
test
specimen
to
another.
Finally,
blank
(
untreated
wood)
test
specimens
will
be
included
in
each
chamber
to
continually
evaluate
cross­
contamination
effects
throughout
the
study.

1.3.3
Natural
Weathering
Tests
Objective
The
objective
of
the
natural
weathering
tests
is
to
evaluate
the
effects
of
weathering
in
an
actual
outdoor
environment
on
the
efficacy
of
selected
coating
products
in
reducing
the
DA
from
new
and
aged
CCA
treated
wood.

Scope
Ten
(
10)
coatings
which
have
been
selected
via
the
screening
study
previously
described
will
be
applied
to
miniature
decks
constructed
of
new
and
aged
CCA­
treated
wood.
Each
mini­
deck
will
contain
six
decking
specimens,
and
three
of
these
(
i.
e.,
triplicate)
will
be
wipe
sampled
on
a
monthly
basis,
consistent
with
the
data
to
be
obtained
from
the
accelerated
weathering
tests.
The
remaining
three
decking
specimens
will
be
sampled
at
the
end
of
the
test
period
in
order
to
estimate
the
effects
of
re­
rubbing
(
mild
abrasion)
on
coating
efficacy
and
DA
measurements.
These
coated
specimens
along
with
uncoated
controls
(
both
untreated
and
CCA­
treated)
will
be
exposed
to
natural
weathering
conditions
at
a
controlled
site
in
North
Carolina
for
which
good
meteorological
data
is
routinely
collected
and
will
be
used
to
monitor
conditions
during
this
project.
The
study
design
is
similar
to
that
employed
for
the
accelerated
weathering
tests;
that
is,
ten
coatings,
two
wood
types
(
new
and
aged),
and
three
specimens
(
replicates)
per
coating­
wood
type
combination.
Additionally,
two
CCA­
treated,
uncoated
control
mini­
decks
(
one
using
aged
and
the
other
using
new
wood)
and
one
untreated,
uncoated
mini­
deck
will
be
included.
The
position
of
each
mini­
deck
on
the
site
will
be
randomized
at
the
start
of
the
test,
though
their
directional
orientation
will
be
the
same.
DA
will
be
determined
via
wipe
sampling
monthly.
Monthly
DA
results
will
be
compared
with
baseline
DA
determined
on
adjacent
specimens
similar
to
the
approach
described
for
the
screening
and
indoor
weathering
chamber
test
designs,
in
order
to
determine
monthly
percent
reductions
in
DA
for
each
specimen.
(
Section­
Page)
14
of
20
Draft
­
Revision
1.2
March
2003
The
outdoor
weathering
test
offers
a
means
of
evaluating
the
efficacy
of
coatings
on
horizontal
surfaces
of
the
same
width
and
thickness
and
starting
condition
(
e.
g.,
age)
as
the
chamber
studies,
with
additional
stresses
on
the
specimens
resulting
from
their
attachment
to
the
mini­
deck
support
members
and
of
course,
per
exposure
to
natural
weathering
conditions.

The
proposed
natural
outdoors
weathering
study
is
described
in
more
detail
later
in
this
test
plan.

Data
product
and
use
The
efficacy
of
each
coating
for
reducing
DA
on
new
and
aged
CCA­
treated
wood
will
be
evaluated
as
a
function
of
time
exposed
to
natural
weathering
outdoors.

Post­
coat
DA
will
be
determined
by
wipe
sampling
on
triplicate
specimens
of
each
coating
on
each
of
two
wood
types
(
new
and
aged).
Percent
reduction
in
DA
will
be
determined
monthly,
calculated
as
the
difference
between
the
baseline
DA
established
for
each
test
specimen
prior
to
coating
and
that
month

s
DA
measurement.

Weathered
coatings
will
be
ranked
monthly
according
to
their
efficacy
based
on
percentage
reduction
of
DA
from
new
and
aged
CCA
treated
wood.

These
same
coatings
will
also
be
similarly
tested
indoors
in
artificial
weathering
chambers.
The
natural
weathering
study
design,
which
is
consistent
with
the
study
design
for
the
weathering
chamber
tests,
allows
for
similar
statistical
analyses
to
be
applied
to
both
sets
of
data
which
will
aid
in
the
comparison
of
results
between
artificial
and
natural
weathering
and
their
respective
impacts
on
the
efficacy
of
selected
coatings.

Limitations
Stress
factors.
Due
to
the
small
size
of
the
mini­
decks,
the
stress
factors
generated
by
attached
specimens
during
weathering
may
not
be
realistic.

Study
duration.
The
currently
planned
duration
of
this
study
(
approximately
7
months)
is
not
sufficient
to
adequately
compare
artificial
weathering
chamber
results
(
which
may
simulate
several
years
of
certain
weathering
conditions)
to
the
natural
weathering
results.
A
continuation
of
the
outdoor
weathering
study
monitoring
program
beyond
the
currently­
planned
study
duration
could
produce
a
data
set
that
better
allows
such
a
comparison
to
be
made.

1.3.4
Other
Overall
Study
Limitations
Type
and
condition
of
aged
wood
Because
of
the
large
number
of
variables
that
affect
the
weathering
of
existing
CCA­
treated
wood
(
Section­
Page)
15
of
20
Draft
­
Revision
1.2
March
2003
structures,
establishing
a
consistent
and
representative
source
of
aged
wood
for
these
tests
is
relatively
challenging.
It
is
expected
that
different
sources
of
aged
wood
may
have
considerably
different
characteristics
which
are
likely
to
impact
coating
performance.
Because
resources
for
this
project
are
limited,
only
one
source
of
aged
wood
shall
be
used,
taken
from
a
single
existing
outdoor
structure
(
e.
g.,
a
deck).

Predetermined
criteria
were
established
in
order
to
rank
and
accordingly
select
candidate
aged
wood
source
structures.
To
wit,
preferred
source
woods
must
have
been
in
service
for
between
5
and
15
years
with
no
washing
solutions
or
coatings
having
been
applied
within
the
past
5
years.
To
the
extent
possible,
wood
from
the
selected
structure
shall
be
taken
from
areas
of
the
structure
that
have
been
exposed
to
similar
abrasion/
traffic
and
weathering
patterns.
Of
utmost
concern
is
having
a
source
of
consistent,
aged
wood.
The
following
are
important
characteristics
to
be
considered
with
respect
to
the
source
of
aged
wood
used:

|
Location/
site
|
Type
of
use
(
e.
g.,
residential
deck,
etc.)

|
Age
|
Abrasion
pattern
|
Exposure
orientation
(
directional)

|
Exposure
level
(
shading
vs.
direct
exposure,
etc.)

|
Treatment
history
Information
on
these
characteristics
were
gathered
for
multiple
candidate
sources
which
were
then
critically
analyzed
by
EPA
and
ARCADIS
for
conformance
with
specified
criteria
and
completeness
of
specified
information
about
the
source,
in
order
to
select
an
aged
wood
source.
An
excellent
source
of
aged
wood
has
been
selected.
The
source
is
a
stand­
alone
(
i.
e.,
unattached
to
any
other
structure)
deck
with
generally
full
exposure
(
except
for
several
boards

which
will
not
be
used

which
are
located
under
attached
benches),
with
only
moderate
shading
by
adjacent
buildings
during
low
sun
positions.
Given
its
open/
stand­
alone
nature,
abrasion
patterns
appear
very
consistent
and
the
boards
are
visually
similar
to
one
another.
Additional
information
on
this
source
is
being
gathered,
as
it
is
dismantled
under
the
supervision
of
ARCADIS.
The
specific
locations
and
orientations
of
individual
boards
will
be
documented
during
dismantling
of
the
source
structure,
so
that
a
map
of
the
structure
showing
the
location
of
each
specimen
tested
can
be
prepared.
The
source
of
aged
wood
for
the
screening
tests
should
be
sufficient
to
provide
an
adequate
supply
of
wood
for
the
indoor
and
outdoor
weathering
tests.

Re­
rubbing
Effects
and
Baseline
Sampling
(
Section­
Page)
16
of
20
Draft
­
Revision
1.2
March
2003
A
significant
practical
issue
arises
as
a
result
of
the
sampling
process
itself.
Ideally,
initial
surface
wipe
samples
would
be
taken
from
each
specimen
to
be
further
tested.
However,
a
recent
study
by
the
Consumer
Products
Safety
Commission
(
CPSC)
suggests
that,
as
could
be
expected,
the
act
of
sampling
the
surfaces
of
CCA­
treated
wood
in
fact
removes
a
considerable
amount
of
the
DA
from
a
test
specimen.
Furthermore,
wipe
sampling
is
a
form
of

abrasion

which
is
suspected
to
be
a
significant
variable
in
determining
both
uncoated
DA
as
well
as
durability
and
efficacy
of
tested
coatings.
Clearly,
there
is
virtually
no
alternative
to
wipe
sampling
coated
surfaces
to
determine
DA
(
except
perhaps
leachate
sampling
for
which
no
transfer
relationships
have
been
developed
that
relate
mass
leached
from
a
sample
to
amount
transferred
to
a
hand).
While
it
may
be
possible
to
attempt
to
artificially
correct
DA
results
for
the
effects
of
rerubbing
(
i.
e.,
per
the
analysis
of
appropriate
control
samples
and
subsequent
modification
of
measured
DA
on
individual
specimens),
the
decision
has
been
made
to
not
wipe
sample
surfaces
to
be
coated
prior
to
coating,
as
such
an
approach
could
cause
data
analysis
and
coating
efficacy
complications.

Individual
baseline
values
of
DA
will
be
determined
for
each
specimen
to
be
coated
and
tested.
The
baseline
DA
of
a
specimen
will
be
determined
by
averaging
the
DAs
from
the
two
adjacent
specimens
on
either
side
of
the
test
specimen.
The
adjacent
specimens
will
be
archived
uncoated
after
sampling
(
and
thus
not
used
in
any
further
testing).
This
step
will
avoid
any
data
analysis
and
coating
efficacy
complications
that
may
arise
from
coating
the
pre­
rubbed
specimens.

Some
consideration
was
given
to
wipe
sampling
surfaces
to
be
coated
prior
to
coating
and
then
waiting
or
even
exposing
the
specimen
to
weathering
to
induce
more
migration
of
CCA
analytes
to
the
surface
of
the
specimen
prior
to
coating.
While
this
concept
may
be
sound,
there
is
simply
no
data
of
which
we
are
aware
to
support
the
design
of
such
a
method.
That
is,
it
is
not
known
how
much
time
must
elapse
and/
or
under
what
conditions
specimens
must
be
maintained
to
allow
surficial
CCA
analyte
concentrations
to
rebound
to
pre­
wipe
conditions.

Two
other
options
have
been
seriously
considered
to
resolve
this
issue.
The
first
possible
resolution
would
be
to

sacrifice

a
certain
number
of
specimens
to
provide
only
baseline
surficial
concentration
data.
That
is,
use
one
or
more
specimen
per
board
to
establish
average
baseline
surficial
concentrations.
The
other
option
is
to
wipe
sample
the
undersides
of
the
test
specimens
to
establish
the
baseline
DA
of
each
specimen.
This
was
seen
as
a
potentially
good
option
for
new
CCA­
wood
specimens,
but
not
for
aged
specimens,
as
their
top
faces
are
well­
defined
and
of
much
greater
interest
than
their
bottom
faces.
The
top
faces
of
aged
CCA
wood
specimens
would
be
expected
to
have
considerably
different
characteristics
than
their
bottom
faces.
While
the
same
is
not
necessarily
true
of
new
CCA
wood,
CPSC
data
suggests
that
sample
variability
along
the
length
of
a
given
board
is
less
that
the
variability
between
top
and
bottom
faces
of
a
specimen.
As
such,
and
as
previously
indicated,
the
screening
and
weathering
test
plans
will
employ
a
method
whereby
the
DA
of
adjacent
sacrificial
specimens
are
averaged
in
order
to
establish
the
baseline
DA
for
each
individual
test
specimen.

Test
Specimen
Lengths
(
Section­
Page)
17
of
20
Draft
­
Revision
1.2
March
2003
For
this
study,
three
different
test
specimen
lengths
will
be
used
for
the
three
main
testing
components
described.
For
the
screening
study,
46
cm
(
18

)
lengths
shall
be
used,
with
a
43
cm
sampling
length.
These
specimens
have
been
cut
and
are
being
conditioned.
For
the
accelerated
artificial
weathering
chamber
tests,
30
cm
(
12

)
lengths
shall
be
used
(
with
a
25
cm
wipe
length)
because
of
the
dimensional
limitations
of
the
QUV
test
chambers.
For
the
outdoor
natural
weathering
mini­
decks,
a
61
cm
(
24

)
specimen
length
will
be
employed,
with
a
50
cm
sampling
length
to
be
consistent
with
testing
conducted
by
CPSC.
It
is
unclear
what
effect
wipe
length
has
on
measured
DA.
Some
indication
of
its
effect
may
be
derived
per
the
comparison
of
results
of
baseline
DA
measurements
between
specimens
of
different
sampling
lengths.

Coating
Application
Limitations
As
mentioned
previously,
coating
method
effects
will
not
be
rigorously
tested.
Additionally,
the
effect
of
other
coating
application
options
(
for
example,
one
coat
of
a
coating
versus
multiple
coats)
will
not
be
tested
rigorously.

Abrasion
Effects
The
effects
of
abrasion
(
e.
g.,
by
repeated
contact/
walking)
will
not
be
rigorously
tested
in
this
project,
as
it
is
beyond
the
scope
achievable
per
the
available
resources.
Clearly,
abrasion
effects
on
DA
concentrations
as
well
as
on
coating
efficacy
and
durability
is
a
major
issue
that
should
be
addressed
in
the
future.
Additionally,
the
transfer
of
CCA
analytes
via
shoes/
feet,
pets,
and
other
potential
contact
routes
may
be
important
but
cannot
be
addressed
in
this
study.

CCA
Analytes
and
Speciation
The
available
resources
for
this
project
dictate
that
the
focus
be
on
arsenic
and
thus
that
only
DA
be
evaluated.
Potential
contact
with
dislodged
copper
and
particularly
chromium
are
also
important.
Data
for
these
other
analytes
may
be
available
from
the
analytical
lab
employed
and
could
be
procured
at
a
later
date
if
resources
allow.

Additionally,
the
speciation
of
CCA
analytes
could
be
an
important
determinant
of
contact
risks.
Only
total
arsenic
(
and
total
chromium,
if/
as
resources
allow)
will
be
routinely
measured
in
this
study,
due
to
resource
limitations,
as
speciating
arsenic
and
chromium
is
analytically
significantly
more
complex,
and
thus
costly.

Other
Limitations
The
following
issues
will
not
be
rigorously
addressed
by
the
proposed
study:
(
Section­
Page)
18
of
20
Draft
­
Revision
1.2
March
2003
|
Performance
of
coatings
on
wood
of
different
dimensions
that
may
be
encountered
(
note
that
it
may
be
possible
to
test
posts
in
the
outdoor
weathering
study)

|
Directional
exposure
effects
(
though
it
may
be
possible
to
consider
these
to
some
extent
in
the
outdoor
weathering
study)

|
Performance
in
different
climatic
regions
(
NE,
NW,
SW
US)

|
Performance
on
members
oriented
vertically
or
at
angles
(
again,
it
may
be
possible
to
test
posts
in
the
outdoor
weathering
study)

|
Performance
following
various
wood
preparation
techniques
|
Recoat
performance
1.4
Data
Quality
Objectives
The
critical
measurements
for
the
accelerated
and
natural
weathering
tests
are
total
arsenic
(
and,
secondarily,
total
chromium)
concentrations.
Data
quality
indicator
goals
for
arsenic
concentration
in
terms
of
accuracy,
precision,
and
completeness
are
shown
in
Table
1.4.1.

Table
1.4.1.
Data
Quality
Indicator
Goals
for
Critical
Measurements
Analyte
Method
Accuracy
(%
Recovery)
Precision
(%
RSD/
RPD)
Completeness
(%)
Arsenic
(
total)
SW­
846
Method
6020
(
modified)
90­
110
10
90
Chromium
(
total)
SW­
846
Method
6020
(
modified)
90­
110
10
90
(
Section­
Page)
19
of
20
Draft
­
Revision
1.2
March
2003
1.5
Project
Organization
and
Responsibilities
The
EPA
Work
Assignment
Manager
for
this
project
is
Mark
Mason,
who
will
coordinate
involvement
by
other
EPA
staff
and
CPSC
via
an
interagency
agreement
(
CPSC­
I­
03­
1235)
between
EPA
and
CPSC,
as
appropriate.
Key
CPSC
staff
include
Jacque
Ferrante
and
Warren
Porter.
ARCADIS

Work
Assignment
Leader
is
Victor
D

Amato,
PE.
Libby
Nessley,
with
ARCADIS,
serves
EPA
by
providing
Quality
Assurance/
Quality
Control
(
QA/
QC)
management
services,
while
Todd
Thornton
and
Jerry
Revis,
both
with
ARCADIS,
serve
EPA
by
providing
Health
and
Safety
management
services.
Kevin
Bruce,
ARCADIS,
is
the
overall
OLS
Project
Manager.
He
will
support
this
project
by
helping
coordinate
the
wood
preparation,
coating,
sampling,
and
analytical
tasks.
Johannes
Lee,
ARCADIS,
is
the
Assistant
Project
Manager
for
the
OLS
contract,
and,
as
such,
provides
a
variety
of
administrative
support
functions.
Matt
Clayton,
ARCADIS,
will
prepare
and
coat
wood
samples.
Peter
Kariher,
ARCADIS,
will
take
samples,
prepare
samples
via
digestion
protocol,
and
ship
digested
wipe
and
control
samples
to
the
subcontract
analytical
laboratory,
STL­
Savannah
(
Angie
Weimerskirk,
Project
Manager).
Russ
Clayton,
ARCADIS,
leads
an
engineering
and
technician
group
that
will
be
responsible
for
setting
up
and
outfitting
the
indoor
test
chambers,
and
for
building
the
outdoor
miniature
decks.
Krich
Ratanaphruks,
ARCADIS,
will
provide
relational
database
support.
An
organizational
chart
is
provided
as
Figure
1.5.1.
Table
1.5.1
provides
contact
information
for
proposed
staff.
(
Section­
Page)
20
of
20
Draft
­
Revision
1.2
March
2003
0
Figure
1.5.1.
Organizational
Chart
for
Sampling/
Analytical
Qualification
Testing
(
Section­
Page)
21
of
20
Draft
­
Revision
1.2
March
2003
Table
1.5.1.
Contact
Information
for
Key
Project
Staff
Staff
Contact
Organization
Responsibility
Phone
Number
E­
mail
Address
Mark
Mason
US
EPA
WA
Manager
(
919)
541­
4835
Mason.
Mark.@
epa.
gov
Jacque
Ferrante
CPSC
Health
Sciences
(
301)
504­
7259
jferrante@
cpsc.
gov
Warren
Porter
CPSC
Lab
Sciences
(
301)
421­
6421
wporter@
cpsc.
gov
Victor
D

Amato
ARCADIS
WA
Leader
(
919)
544­
4535
vd

amato@
arcadis­
us.
com
Libby
Nessley
ARCADIS
QA
Manager
(
919)
544­
4535
lnessley@
arcadis­
us.
com
Todd
Thornton
ARCADIS
H&
S
Manager
(
919)
544­
4535
tthornton@
arcadis­
us.
com
Jerry
Revis
ARCADIS
H&
S
Manager
(
919)
544­
4535
jrevis@
arcadis­
us.
com
Kevin
Bruce
ARCADIS
PM,
Advisor
(
919)
544­
4535
kbruce@
arcadis­
us.
com
Peter
Kariher
ARCADIS
Lab
Scientist
(
919)
544­
4535
pkariher@
arcadis­
us.
com
Matt
Clayton
ARCADIS
Lab
Scientist
(
919)
544­
4535
mclayton@
arcadis­
us.
com
Russ
Clayton
ARCADIS
Engineering
(
919)
544­
4535
rclayton@
arcadis­
us.
com
Krich
Ratanaphruks
ARCADIS
Database
Technician
(
919)
544­
4535
kratanaphruks@
arcadis­
us.
com
Angie
Weimerskirk
STLSavannah
Analytical
Manager
(
912)
354­
7858
aweimerskirk@
stl­
inc.
com
(
Section­
Page)
1
of
16
Draft
­
Revision
1.2
March
2003
2
Sampling
Approach
Under
this
test
plan,
two
main
series
of
experiments
will
be
conducted:
1)
indoor,
accelerated
weathering
using
artificial
weathering
chambers
and
2)
outdoor,
natural
weathering
using
miniature
decks
placed
outdoors
at
a
site
in
North
Carolina.
Both
series
of
tests
will
include
baseline
measurement
of
DA
(
as
previously
described),
as
well
as
monthly
wipe
sampling
for
measurement
of
DA
after
coating
application
and
as
weathering
progresses.

The
following
subsections
describe
in
detail
the
selection
of
materials
for
testing,
application
of
selected
coatings
to
the
CCA
treated
substrates,
weathering
details,
and
sampling
procedures.
A
project­
specific
health
and
safety
plan
(
HSP)
is
under
development
and
will
be
appended
to
this
QAPP
after
the
HSP
is
finalized
and
approved.

2.1
Preparation
and
Characterization
of
Wood
Specimens
Wood
specimens
will
be
prepared
using
new
and
aged
southern
yellow
pine
that
has
been
originally
CCA
treated
to
0.4
pound
per
cubic
foot,
in
nominal
5/
4

x
6

cross­
sectional
dimensions.
New
southern
yellow
pine
that
has
not
been
CCA
treated
will
be
used
for
blank
control
specimens.
Non­
treated
controls
will
most
likely
be
of
either
1

x
6

or
2

x
6

nominal
dimensions
since
non­
treated
5/
4

x
6

decking
wood
is
not
typically
available,
though
it
may
be
possible
to
easily
obtain
untreated
5/
4

x
6

southern
yellow
pine
locally
from
a
lumber
treating
company.
New
CCA
treated
wood
will
be
purchased
from
a
retail
store
and
will
be
labeled

CCA­
C
Ground
Contact

,
indicating
that
the
product
has
been
treated
with
the

C

formulation
of
CCA
preservative,
which
is
most
commonly
available.
Care
will
be
taken
to
minimize
handling
of
the
primary
(
i.
e.,
6

width)
faces
of
the
board,
with
the
short
edges
of
the
board
preferentially
held
during
transport
and
cutting.
Additionally,
care
will
be
taken
to
minimize
contact
and
especially
rubbing
of
the
primary
faces
of
the
board
with
any
other
surfaces.
Storage
of
the
board/
specimens
shall
be
on
its
cut
ends
(
i.
e.,
short
faces)
or
edges.

One
consistent
source
of
aged
wood
is
required
for
all
of
the
weathering
testing
proposed.
A
full
description
of
this
aged
wood
selection
and
characterization
is
described
in
Section
1.3.4.
(
Section­
Page)
2
of
16
Draft
­
Revision
1.2
March
2003
Both
new
and
aged
boards
will
be
cut
using
a
circular
table
saw
(
or
other
similar
cutting
device)
into
lengths
required
for
use
as
test
specimens
for
the
weathering
tests.
The
indoor,
accelerated,
artificial
weathering
tests
will
require
specimens
of
30
cm
(
12

)
lengths,
while
the
outdoor,
natural
weathering
tests
will
require
specimens
of
61
cm
(
24

)
lengths.
The
saw
will
be
decontaminated
between
cutting
the
different
types
of
wood
utilized
(
new
CCA,
aged
CCA,
untreated)
and
the
untreated
wood
will
be
cut
first
(
after
installation
of
a
new
blade)
to
prevent
cross­
contamination
of
samples.
Decontamination
will
follow
a
similar
protocol
to
that
used
to
clean
the
wipe
sampling
device
between
samples.
15
cm
(
6

)
of
each
end
of
each
board
will
be
removed
and
archived
and
10­
cm
(
4

)
segments
between
each
30­
and
61­
cm
test
specimen
will
be
removed
and
archived,
with
some
of
the
interior
10­
cm
segments
used
to
characterize
the
source
wood
via
moisture
content
and
total
arsenic
(
and
possibly
total
chromium)
analysis.
30­
or
61­
cm
wood
specimens
will
be
visually
inspected
and
those
exhibiting
excessive
amounts
of
deformities,
presence
of
heartwood,
knots,
resin
pockets,
and
other
defects
will
be
disqualified
for
use
in
the
screening
testing
(
it
will
be
allowable
to
avoid
defective
segments
of
wood
during
cutting
and/
or
using
wood
unacceptable
for
the
30­
or
61­
cm
specimens
as
10­
or
15­
cm
end
and
interior
segments
for
archiving,
so
as
to
not
waste
or
require
an
excessive
amount
of
wood).
Each
segment
will
be
identified
with
a
unique
alphanumeric
code
as
follows:

|
New
board
codes
will
be
prefixed
by
the
letter

N

|
Aged
board
codes
will
be
prefixed
by
the
letter

A

for
source
A
and

B

for
source
B
|
Each
new
and
aged
board
will
be
identified
with
a
letter
(
A,
B,
C,

)

|
Each
test
specimen
from
new
and
aged
boards
will
be
identified
with
a
number
(
1,
2,
3,

)

|
Each
end
piece
will
be
identified
with
the
number
of
the
specimen
adjacent
to
it,
followed
by
the
letter

E

|
Each
10­
cm
interior
segment
will
be
identified
with
the
numbers
of
the
specimens
adjacent
to
it
separated
by
a
forward
slash
(

/

)

|
Defective
segments
will
be
suffixed
with
the
letter

D

For
example,
assuming
that
a
366­
cm
(
12

)
length
of
new
CCA
treated
wood
is
used
for
the
30­
cm
indoor
weathering
samples,
samples
from
the
first
board
would
be
identified
from
end­
to­
end,
as
follows:
N­
A­
1E,
N­
A­
1,
N­
A­
1/
2,
N­
A­
2,
N­
A­
2/
3,
N­
A­
3,
N­
A­
3/
4,
N­
A­
4,
N­
A­
4/
5,
N­
A­
5,
N­
A­
5/
6,
N­
A­
6,
N­
A­
6/
7,
N­
A­
7,
N­
A­
7/
8,
N­
A­
8,
N­
A­
8E.
In
this
example,
N­
A­
1,
N­
A­
2,
N­
A­
3,
N­
A­
4,
N­
A­
5,
N­
A­
6,
N­
A­
7,
(
Section­
Page)
3
of
16
Draft
­
Revision
1.2
March
2003
and
N­
A­
8
would
be
the
30­
cm
specimens
to
be
tested,
with
the
remainder
being
end
pieces
or
10­
cm
interior
segments
to
be
archived
for
possible
future
testing.
If,
for
example,
a
segment
of
wood
between
NA
3
and
N­
A­
4
was
deemed
to
be
defective,
it
would
be
labeled,
N­
A­
3/
4­
D.
The
specimen
identification
example
presented
above
is
shown
in
Figure
2.1.2.
Specimen
IDs
will
be
cross­
referenced
with
coating/
application
method
as
described
in
subsequent
subsections.

0
Figure
2.1.2.
Example
specimen
identification
scheme
All
samples
will
be
identified
on
one
cut
end
or
uncut
edge
with
its
identification
code,
as
well
as
with
its

top

side
using
permanent
marker.
Alternatively,
the
identification
code
may
be
correlated
to
a
bar
code
affixed
to
one
cut
end
of
a
specimen.
The
identification
of
the
top
face
will
be
important
for
the
aged
specimens.
For
the
new
specimens,
the
selection
of
the
top
face
of
each
specimen
will
be
the
same
from
specimen
to
specimen
along
a
board.
Documentation
will
be
kept
identifying
all
numbered
specimens
taken
from
each
original
length
of
CCA
treated
wood.

After
cutting
and
identification
marking,
both
sides
of
each
specimen
will
be
rinsed
with
tap
water
using
a
light
setting
of
a
pressure
washer.
Rinsate
water
generated
by
each
source
(
new,
aged,
untreated)
will
be
collected,
preserved
using
nitric
acid,
and
stored
in
TFE
or
PFA
vessels
in
case
needed
for
future
subsampling
and
analysis.
Additionally,
a
sample
of
the
tap
water
used
to
rinse
the
boards
will
be
collected
and
archived
similarly.
The
specimens
will
then
be
stored
in
an
indoor,
temperature­,
and
humiditymonitored
and
controlled
space
prior
to
coating
and
testing.
Temperature
will
be
controlled
to
be
between
15
and
25

C
(~
60­
80

F).
Relative
humidity
will
be
controlled
to
be
between
25
and
75%.
Temperature
and
relative
humidity
will
be
measured
and
recorded
at
least
twice
per
day.
Specimens
will
be
allowed
to
equilibrate
with
storage
conditions
for:
(
Section­
Page)
4
of
16
Draft
­
Revision
1.2
March
2003
|
as
long
as
possible,
but
at
least
7
days,

|
until
moisture
content
is
less
than
20%
(
wet
weight
basis),
or
after
21
days,
whichever
occurs
first
Moisture
content
will
be
measured
on
at
least
one
10­
cm
interior
specimen
per
board,
as
often
as
necessary
to
ensure
that
the
above
criteria
have
been
achieved.
These
10­
cm
interior
specimens
used
for
testing
shall
be
stored
under
the
same
conditions
as
the
30­
and
61­
cm
specimens.
Care
must
be
taken
to
ensure
that
samples
for
measuring
moisture
content
are
taken
from
similar
areas
on
the
10­
cm
specimens
(
using,
for
example,
a
coring
drill
bit),
so
that
moisture
content
over
time
can
be
compared.
Moisture
content
will
be
measured
as
described
in
section
3.0.

All
wood
shall
be
arranged
in
the
artificial
weathering
chambers
with
the
top
face
exposed
(
i.
e.,
facing
the
UV/
Spray
system)
and
in
the
mini­
decks
with
their
top
face
facing
upwards.

2.1.1
Wood
for
Indoor,
Accelerated,
Artificial
Weathering
Tests
For
the
new
CCA­
treated
wood
and
the
source
of
aged
wood,
three
30­
cm
(
12

)
specimens
each
will
be
used
for
each
selected
coating.
The
three
specimens
for
each
coating
shall
be
taken
at
random
locations
from
separate
boards.
A
schedule
of
specimens
to
be
used
is
provided
as
Table
2.1.1.
(
Section­
Page)
5
of
16
Draft
­
Revision
1.2
March
2003
Table
2.1.1.
Schedule
of
Specimens
for
Accelerated,
Artificial
Weathering
Tests
Coating
New
CCA
Aged
CCA

A

Untreated
Wood
Coating
1
3
3
0
Coating
2
3
3
0
Coating
3
3
3
0
Coating
4
3
3
0
Coating
5
3
3
0
Coating
6
3
3
0
Coating
7
3
3
0
Coating
8
3
3
0
Coating
9
3
3
0
Coating
10
3
3
0
Uncoated
3
3
6
(
1/
chamber)

A
table
to
match
specimen
identification
code
with
coating/
sample
ID
will
be
prepared
after
randomized
selection
of
specimens
is
conducted.

2.1.2
Wood
for
Outdoor,
Natural
Weathering
Tests
Sketches
of
the
miniature
decks
to
be
used
for
the
outdoor
weathering
tests
are
shown
in
Figure
2.1.3.
A
schedule
for
the
mini­
decks
and
61­
cm
(
24

)
specimens
to
be
wipe
sampled
is
provided
in
Table
2.1.2.

Table
2.1.2.
Schedule
of
Mini­
Decks
and
Specimens
for
Outdoor,
Natural
Weathering
Tests
New
CCA
Wood
Aged
CCA
Wood
Untreated
Wood
Coating
mini­
decks
61­
cm
specimens
mini­
decks
61­
cm
specimens
mini­
decks
61­
cm
specimens
Coating
1
1
6
1
6
0
0
Coating
2
1
6
1
6
0
0
Coating
3
1
6
1
6
0
0
Coating
4
1
6
1
6
0
0
(
Section­
Page)
6
of
16
Draft
­
Revision
1.2
March
2003
Coating
5
1
6
1
6
0
0
Coating
6
1
6
1
6
0
0
Coating
7
1
6
1
6
0
0
Coating
8
1
6
1
6
0
0
Coating
9
1
6
1
6
0
0
Coating
10
1
6
1
6
0
0
Uncoated
1
6
1
6
1
6
0
Figure
2.1.3.
Schematic
of
Mini­
Deck
Construction
Posts
(
4

x
4

)
to
be
used
in
the
mini­
deck
construction
will
be
CCA­
C
treated.
Posts
will
be
new
CCAtreated
for
all
of
the
mini­
decks.
2

x
4

supports
to
which
decking
is
nailed
will
be
untreated
southern
yellow
pine.
Additionally
these
supports
will
be
slightly
offset
above
the
tops
of
the
posts
to
ensure
that
the
decking
does
not
have
the
opportunity
to
contact
other
CCA­
treated
wood
(
and
untreated
wood
for
the
blank
mini­
deck).
Bracing
will
also
be
untreated
wood.
The
mini­
decks
will
be
free­
standing
(
i.
e.,
posts
will
not
be
set
into
the
ground).
(
Section­
Page)
7
of
16
Draft
­
Revision
1.2
March
2003
A
table
to
match
specimen
identification
code
with
coating/
sample
ID
will
be
prepared
after
randomized
selection
of
specimens
is
conducted.

2.2
Selection
of
Coatings
Ten
coatings
to
be
tested
will
be
selected
per
the
screening
protocol
attached
as
Appendix
A.

2.3
Coating
Application
Unused
aliquots
of
each
coating
tested
shall
be
sampled
in
duplicate,
prepared,
and
analyzed
in
accordance
with
methods
specified
in
Section
3.0.

Staff
performing
coating
and
monthly
wipe
sampling
will
not
be
aware
of
the
specific
coating
being
applied
or
sampled
in
order
to
prevent
real
or
perceived
bias.

2.3.1
Indoor
Accelerated
Weathering
Tests
Coatings
will
be
applied
to
fully
cover
the
top
and
bottom
faces,
uncut
edges,
and
cut
ends
(
to
be
coated
after
all
other
surfaces
have
been;
top
faces
will
be
covered
first)
of
CCA
treated
wood
specimens
in
accordance
with
manufacturers

recommendations
with
appropriate
adaptations
to
the
laboratory
environment,
this
test
plan,
and
the
results
of
screening
tests,
as
described
in
Appendix
A.
In
cases
where
a
choice
of
application
method
(
e.
g.,
spray
or
brush)
is
offered,
brush
application
will
be
employed.

Coating
application
will
take
place
individually
under
a
fume
hood,
on
a
laboratory
bench
top,
or
other
appropriate
location.
Care
shall
be
taken
to
ensure
that
the
top
and
bottom
faces
of
the
specimens
are
not
allowed
to
contact
other
surfaces.

Individual/
dedicated
brushes
and/
or
manual
spray
applicators
will
be
used
to
apply
each
coating
to
each
substrate
(
wood
type).
If
brush
application
is
employed,
brushes
will
be
prepared
for
initial
coating
application
by
cleaning
in
accordance
with
brush
manufacturer

s
recommendations.
Additional
preparations
will
be
conducted
only
in
accordance
with
coating
manufacturer

s
recommendations.
Between
specimens
and
after
a
particular
coating
has
been
applied
to
a
given
group
of
specimens
coated
with
a
(
Section­
Page)
8
of
16
Draft
­
Revision
1.2
March
2003
similar
product,
brushes
will
again
be
cleaned
in
accordance
with
the
brush/
coating
manufacturers

recommendations
and
then
archived
by
storing
in
accordance
with
brush
manufacturer

s
recommendations.
Fresh
cleaning
solution
will
be
used
for
each
coating.
Note
that
brushes
and
the
use
of
cleaning
solution
will
have
been
prequalified
for
use
per
control
samples
taken
during
the
screening
study
(
Appendix
A).

Separate
aliquots
of
coating
liquid
will
be
used
for
each
specimen
to
be
coated
with
a
given
coating,
in
order
to
prevent
cross­
contamination
of
coating
liquid
by
re­
dipping
the
brush
applicator.
Separate
aliquots
of
coating
liquid
will
be
poured
into
disposable
plastic
trays,
which
will
be
discarded
after
application
of
that
coating
to
each
given
specimen.
Coating
remaining
in
the
trays
will
be
composited
so
that
one
sample
is
retained
for
each
coating/
wood
type.
These
samples
will
be
archived
for
possible
future
analyses.

Spray
coating
will
be
used
only
if
explicitly
recommended
over
brushing
by
the
manufacturer.
Spraying,
if
employed,
will
be
conducted
by
pouring
coating
into
a
manual
sprayer.
A
new
sprayer
will
be
dedicated
to
each
coating.
Coating
remaining
in
the
sprayer
will
be
collected
into
one
sample
for
each
coating.
These
samples
will
be
archived
for
possible
future
analyses.
Coatings
having
recommended
application
methods
that
have
not
been
discussed
here
will
be
similarly
managed
on
a
coating­
by­
coating
basis.

Application
procedures
and
any
notable
observations
will
be
documented
for
each
coating.

Coating
application
will
be
assessed
in
two
ways:
gravimetrically,
and
via
measurement
of
wet
film
thickness.
Gravimetric
assessment
will
involve
measuring
the
applied
mass
of
coating
on
each
specimen.
Applied
mass
will
be
measured
gravimetrically
in
two
ways:
1)
calculated
as
the
difference
in
specimen
mass
before
and
after
coating
and
2)
calculated
as
the
difference
between
the
starting
coating
aliquot
and
brush
and
the
post­
coating
aliquot
and
brush
weights.
Neither
method
is
ideal,
as
some
loss
of
coating
mass
due
to
volatilization
is
expected
even
in
the
relatively
short
times
necessary
to
apply
coating.
Additionally,
all
specimens
will
be
weighed
at
the
end
of
the
7­
day
curing
at
the
time
of
final
wipe
sampling
(
for
determination
of
dry
coating
weight)
in
addition
to
pre­
and
post­
application.
Corrections
for
moisture
content
instability
can
be
made
using
the
results
of
moisture
analyses
previously
described,
if
necessary.
Wet
film
thickness
will
be
measured
on
film­
forming
coatings
using
handheld
wet
film
thickness
gauges,
as
manufactured
by
Paul
N.
Gardner
Co.,
Inc.
or
equal.
The
gauges
are
disposable
and
will
be
discarded
after
each
measurement,
after
which
a
new
gauge
will
be
used
on
the
next
coated
specimen.
A
summary
of
the
proposed
parameters
to
be
measured
either
directly
or
by
calculation
is
provided
in
Table
2.3.1.

Table
2.3.1.
Summary
of
Coating
Application
Measurements
(
Section­
Page)
9
of
16
Draft
­
Revision
1.2
March
2003
Description
Acronym
Measurement
or
Calculation
Uncoated
specimen
weight
USW
Weigh
specimen
prior
to
coating
Wet
coated
specimen
weight
WCSW
Weigh
specimen
immediately
post­
coating
Dry
coated
specimen
weight
DCSW
Weigh
specimen
7
days
after
coating
Starting
coating
weight
SCW
Weigh
coating
aliquot
and
brush
prior
to
coating
Ending
coating
weight
ECW
Weigh
coating
aliquot
and
brush
after
coating
Wet
coating
weight
WCW
WCW
=
WCSW

USW
Dry
coating
weight
DCW
DCW
=
DCSW

USW
Applied
coating
weight
ACW
ACW
=
ECW

SCW
Wet
Film
Thickness
WFM
Measured
directly
using
disposable
gauge
(
only
for
film­
formers)

2.3.2
Outdoor,
Natural
Weathering
Tests
Mini­
decks
will
first
be
constructed.
After
construction
and
baseline
characterization
of
DA
(
described
below),
all
exposed
surfaces
of
the
decks
shall
be
coated
in
accordance
with
coating
manufacturers

recommendations,
starting
with
the
top
faces
of
the
decking
specimens
(
six
per
mini­
deck).
Otherwise,
application
method
shall
be
similar
to
that
described
in
Section
2.3.1
for
coating
the
chamber
test
specimens,
with
each
mini­
deck
to
be
coated
using
its
own
dedicated
brush.

Likewise,
measurement
of
coating
application
amount
will
be
similar
to
that
described
for
the
chamber
specimens
in
Section
2.3.1.
For
film­
forming
coatings,
coating
application
will
be
assessed
via
measurement
of
wet
film
thickness,
using
handheld
wet
film
thickness
gauges,
as
manufactured
by
Paul
N.
Gardner
Co.,
Inc.
or
equal.
The
gauges
are
disposable
and
will
be
discarded
after
each
measurement,
after
which
a
new
gauge
will
be
used
for
the
next
measurement.
Wet
film
thickness
(
WFT)
will
be
measured
on
one
randomly­
selected
location
on
each
61­
cm
specimen
of
each
mini­
deck.
WFT
measurements
will
be
recorded
and
compared
with
those
measured
for
the
chamber
test
specimens
to
ensure
consistency
between
methods
of
application.
The
mass
of
coating
applied
will
also
be
measured
gravimetrically,
by
calculating
the
difference
between
the
starting
coating
aliquot
and
brush
and
the
post­
coating
aliquot
and
brush
weights.
Weights
will
be
measured
at
least
twice
for
each
mini­
deck:
after
coating
the
top
decking
surfaces
and
after
coating
the
entire
mini­
deck.
Table
2.3.1
can
be
referenced
for
information
on
the
coating
(
Section­
Page)
10
of
16
Draft
­
Revision
1.2
March
2003
application
evaluation
methods.

2.4
Weathering
Protocols
2.4.1
Accelerated
Indoor
Weathering
Widely
accepted
standards
for
accelerated
weathering
chamber
testing
for
evaluating
the
efficacy
of
CCAtreated
wood
coatings
in
reducing
DA
generally
do
not
exist.
Several
artificial
weathering
protocols
have
been
used
in
related
studies
investigating
the
durability
of
wood
coatings,
including
one
European
standard
which
was
developed
based
on
the
use
of
the
test
chambers
to
be
used
in
this
study.
The
accelerated
weathering
chambers
to
be
used
shall
be
model
QUV/
Spray,
as
manufactured
by
Q­
Panel
Lab
Products,
Cleveland,
OH.

Accelerated
weathering
chambers
may
allow
for
the
accelerated
simulation
of
certain
weathering
factors,
including
UV
exposure,
abrasion
due
to
precipitation,
thermal
shock,
and
condensation.
However,
chamber
weathering
tests
are
artificial
and
may
provide
misleading
information
if
test
parameters
are
not
properly
selected.
In
general,
because
these
results
are
to
be
correlated
with
outdoor
natural
weathering
test
results,
the
goal
of
the
artificial
weathering
protocol
for
this
test
is
to
attempt
to
simulate
natural
weathering
conditions
in
North
Carolina.

Weathering
chambers
shall
be
located
in
E377A­
IEMB

s
Field
Support
Laboratory.
They
shall
be
equipped
with
feedback
control
of
light
intensity,
temperature
sensors,
and
automated
UV
irradiance,
condensation/
humidification,
and
precipitation
systems.

The
following
environmental
factors
that
impact
the
efficacy
of
coatings
will
be
evaluated:

|
UV
light
|
Thermal
shock
|
Precipitation
|
Condensation
Abrasion
is
not
included
in
the
above
list.
Even
though
it
is
an
important
variable
it
is
very
hard
to
quantify
(
Section­
Page)
11
of
16
Draft
­
Revision
1.2
March
2003
and
it
is
not
explicitly
included
in
the
test.
Abrasion
results
from
foot
traffic
and
from
hard
and
prolonged
rain
or
hail
events.
During
the
accelerated
weathering
tests,
some
limited
surface
abrasion
can
be
expected
to
be
caused
by
the
spray
system.

UV
Irradiance
The
QUVs
shall
be
equipped
with
UV­
A
340
nm
lamps
which
can
be
varied
in
intensity.
The
higher
the
irradiance,
the
more
accelerated
the
test.
Because
accelerated
weathering
test
results
will
be
correlated
with
natural
outdoor
weathering
results,
an
irradiance
of
0.68
W/
m2,
which
is
considered
the
maximum
irradiance
of
sunlight,
will
be
used.
In
theory,
this
will
help
facilitate
comparison
between
the
QUV
and
outdoor
test
results.
Additionally,
according
to
the
Operating
Manual
for
the
QUV
system,
it
is
recommended
to
use
either
the
normal
(
0.68
W/
m2)
or
maximum
(
1.38
W/
m2)
set
points
provided
for
this
lamp.
Using
these
precisely
defined
set
points,
as
opposed
to
alternate
settings,
should
allow
for
better
comparison
between
tests
and
potential
replication.

For
the
QUVs,
irradiance
is
set
by
the
user
(

Set
Point

)
and
continuous
feedback
of

Actual
Irradiance

is
generated
by
the
QUV
Solar
Eye
system.
The
Solar
Eye
system
compensates
for
decreasing
irradiance
of
the
bulbs
due
to
natural
aging
by
increasing
the
power
supplied
to
the
lamps.
When
the
lamps
can
no
longer
maintain
the
set
point
irradiance,
the
lamps
must
be
replaced.
The
Solar
Eye
system
shall
be
calibrated
regularly
and
when
indicated
by
the
QUV
system
to
require
calibration.
Actual
irradiance
shall
be
recorded
daily
for
each
Solar
Eye
(
there
are
4
sensors
per
chamber)
daily
Monday­
Friday.

Thermal
Shock
The
UV
lamps
give
off
considerable
heat.
The
QUV
is
equipped
with
an
air
blower
that
operates
continuously
throughout
the
UV
cycle,
as
well
as
a
panel
temperature
sensor
and
a
heating
element
that
heats
the
air
from
the
blower.
Upon
completion
of
the
UV
period,
the
test
samples
can
be
rapidly
cooled
down
by
turning
on
the
water
spray,
thus
creating
thermal
shock
in
the
test
sample.
To
increase
thermal
shock,
the
UV
system
should
be
operated
at
fairly
high
temperatures,
between
50
and
70

C.
The
aforementioned
European
standard
utilizes
a
UV
cycle
temperature
of
60

C.
However,
in
an
effort
to
more
realistically
simulate
natural
conditions,
a
lower
temperature
of
50

C
will
be
used
for
this
project.

The
temperature
of
the
UV
cycle
is
set
by
the
user
and
chamber
temperatures
are
measured/
displayed
(
Section­
Page)
12
of
16
Draft
­
Revision
1.2
March
2003
continuously
on
the
chamber.
The
temperature
sensors
are
calibrated
regularly.
A
monitoring
system
will
be
employed
to
automatically
continuously
record
temperature
in
the
chambers
or
to
manually
record
temperatures
at
least
once
during
each
cycle
on
a
weekly
basis
for
each
chamber.

Precipitation
and
Condensation
According
to
Q­
Panel
Technical
Bulletin
LU­
0821,
spraying
alone
does
not
accurately
reproduce
outdoor
conditions.
Research
has
shown
that,
even
at
inland
sites
in
the
United
States,
outdoor
materials
are
wet
approximately
30
percent
of
the
time.
The
key
reason
for
this
is
dew
formation
as
opposed
to
rain
events.
It
is
believed
that
condensation
represents
dew
formation
more
accurately
than
spraying,
because
condensation
has
a
similar,
deep
wetting
capability
as
opposed
to
spray.
As
such,
condensation
should
be
the
predominant
wetting
mechanism.

In
addition,
during
condensation
the
temperature
can
be
elevated
to
accelerate
the
weathering
process.
According
to
the
Technical
Bulletin,

it
is
not
uncommon
to
find
that
a
10

C
increase
in
condensation
temperature
results
in
a
doubling
of
the
rate
of
deterioration.

A
45

C
condensation
temperature
shall
be
used
for
this
project.
This
is
consistent
with
the
European
protocol
which
utilizes
a
condensation
of
45

C.

Temperature
elevation
is
not
possible
with
spraying,
because
the
QUV
spraying
mechanism
does
not
have
a
heating
mechanism.
However,
chamber
temperature
during
the
spray
cycle
can
be
monitored
and
will
be
recorded
at
least
weekly
during
testing
to
ensure
consistency
during
the
duration
of
testing.

There
are
two
reasons
to
include
spraying
cycles
in
the
test
sequence,
in
addition
to
the
rapid
cooling
effect
that
can
be
achieved
with
spray.
The
first
reason
is
to
induce
abrasion
(
albeit,
a
small
amount).
As
discussed
before,
this
environmental
degradation
mechanism
is
hard
to
quantify.
However,
a
spraying
event,
in
theory,
will
contribute
to
removing
certain
loose,
wood
particles.
The
second
reason
to
include
a
short
spraying
cycle,
would
be
to
wash
off
chemical
components
that
are
likely
to
have
been
leached
out
by
the
condensation,
as
would
be
expected
to
occur
during
rain
events
under
natural
conditions.
These
suggestions
are
congruent
with
findings
by
Arnold,
et
al.
(
1991)
who
state:

the
washing
action
of
water
is
essential
for
removing
solubilized
wood
degradation
products
from
the
wood
surface
and
for
bleaching
the
wood.

Spray
patterns
will
be
manually
measured
and
evaluated
on
a
regular
basis,
in
accordance
with
the
manufacturer

s
recommendations,
using
transparent
spray
plates
provided
by
the
manufacturer.
(
Section­
Page)
13
of
16
Draft
­
Revision
1.2
March
2003
Testing
Sequence
Based
on
the
above,
a
proposed
test
sequence
is
included
in
Table
2.4.1.
The
cycle
takes
about
eight
hours
and
is
to
be
repeated
three
times
per
day.
(
Section­
Page)
14
of
16
Draft
­
Revision
1.2
March
2003
Table
2.4.1.
Test
Sequence
for
Chamber
Weathering
Tests
Mechanism
Time
Temperature
Notes
Spray**
1
min.
Ambient
DI
water.
For
thermal
shock
and
limited
abrasion.
Condensation
4
hours
45

C
Tap
water.
For
deep
wetting.
UV
irradiance
4
hours
50

C
UVA
(
340
nm).
Irradiance
intensity
@
0.68
W/
m2
Total
8
hours
**
note:
QUV
Spray
cycle
flow
rate
is
7
L/
min.
per
chamber.
At
a
one­
minute
cycle
length,
the
monthly
precipitation
amount
is
approximately
equivalent
to
one
year
of
average
rainfall
in
North
Carolina.

This
cycle
is
consistent
with
suggestions
in
the
QUV
Operating
Manual,
which
suggest
between
two
and
eight
hours
of
UV
combined
with
four
hours
of
condensation.
Other
studies
on
sealants
and
coatings
have
used
eight
hours
of
UV
with
four
hours
of
condensation.
A
study
on
wood
specimens
used
five
hours
of
UVA
340
alternated
with
one
hour
of
spray
at
a
chamber
temperature
of
60

C
(
Arnold,
et
al.,
1991).

The
start
of
each
chamber

s
cycle
will
be
staggered
to
ensure
that
sufficient
DI
spray
water
is
available
and
to
allow
the
monthly
sampling
events
to
be
scheduled.
Monthly
sampling
events
will
be
conducted
by
shutting
off
one
unit
at
a
time
at
the
end
of
one
of
its
4­
hour
UV
cycles,
so
as
to
yield
a
relatively
dry
specimen
for
sampling.

Positioning
of
specimens
As
previously
discussed,
a
total
of
six
QUV/
Spray
units
will
be
used.
Three
units
will
be
used
for
new
wood
specimens
and
three
will
be
used
for
aged
wood
specimens.
Each
unit
will
contain
one
specimen
coated
with
each
coating
to
be
tested
along
with
one
treated,
uncoated
control
specimen
and
one
untreated,
uncoated
blank
specimen,
for
a
total
of
12
specimens
per
chamber
(
6
on
each
side
of
the
chamber,
with
3
on
either
side
of
the
UV
intensity
sensor
panel).
Each
of
the
12
specimens
per
chamber
will
be
tested
in
that
particular
chamber
throughout
the
study,
however,
their
positions
within
the
chamber
will
be
randomly
selected
at
the
start
of
the
test
and
on
a
monthly
basis
(
coincident
with
monthly
wipe
sampling
events).

Specimens
will
be
supported
within
the
test
chambers
using
custom­
designed
specimen
dividers
that
(
Section­
Page)
15
of
16
Draft
­
Revision
1.2
March
2003
attempt
to
prevent
the
contamination
of
front
faces
with
cut
end
leachate
and
also
minimize
spray
deflection
between
adjacent
specimens.
Care
shall
be
taken
to
ensure
that
all
available
panel
space
is
occupied,
with
no
more
than
a
1
mm
gap
allowed
on
each
side
of
each
chamber,
per
manufacturer

s
recommendations.
Additionally,
it
will
be
ensured
that
no
loss
of
condensation,
heat,
water,
etc.
is
caused
by
the
use
of
the
wood
specimens
and
custom
sample
supports.

Routine
chamber
monitoring
Checklists
will
be
developed
to
ensure
that
required
maintenance
and
monitoring
items
are
conducted
in
accordance
with
the
chamber
manufacturer

s
recommendations,
this
test
plan,
and
the
objectives
of
this
study.
This
will
include
routine
monitoring
of
UV
intensity
and
spray
distribution
and
calibration
of
UV
sensors
and
temperature
sensors,
all
of
which
have
established
methods
recommended
by
the
chamber
manufacturer.

2.4.2
Outdoor
Weathering
Outdoor
weathering
tests
will
simply
involve
exposing
the
mini­
decks
described
previously
to
natural
outdoor
climatic
conditions
at
a
test
facility
in
Research
Triangle
Park
(
RTP),
North
Carolina.
Mini­
decks
will
be
arranged
randomly
at
the
start
of
testing
and
each
deck
will
be
oriented
have
one
set
of
two
posts
facing
due
south.
Mini­
deck
arrangement/
layout
on
the
site
will
be
documented.

Weather
data
shall
be
collected
for
the
outdoor
weathering
tests.
Data
to
be
collected
are
listed
in
Table
2.4.2.
The
National
Oceanic
and
Atmospheric
Administration
(
NOAA)
in
RTP,
NC
collects
data
on
wind
speed
and
direction,
temperature,
precipitation
amount,
direct
solar
radiation
and
total
solar
radiation
at
the
site
to
be
used
for
mini­
deck
weathering.
Other
parameters
will
be
collected
from
the
National
Climate
Data
Center
(
NCDC),
RDU,
and
should
be
available
in
monthly
summaries,
detailing
specified
conditions
on
a
daily
basis.

At
least
temperature,
irradiance,
and
precipitation
shall
be
recorded
as
parameters
that
affect
weathering
of
wood
outside.
Also,
efforts
will
be
made
to
develop
a
measurement
to
determine
if
dew
formation
is
occurring
overnight.
Research
has
shown
that,
even
at
inland
sites
in
the
United
States,
outdoor
materials
are
wet
approximately
30
percent
of
the
time.
The
key
reason
for
this
is
dew
formation
as
opposed
to
rain
events
(
Technical
Bulletin
LU­
0821,
Q­
Panel).
(
Section­
Page)
16
of
16
Draft
­
Revision
1.2
March
2003
Unfortunately,
dew
point
is
not
measured
or
recorded
by
NOAA
at
this
site.
Data
from
the
Monthly
Climate
Summary
from
the
NCDC,
RDU,
will
be
examined
to
determine
the
possibility
of
dew
formation.

When
fog
is
reported,
dew/
condensation
will
be
present.
The
standard
dew
point
depression
value
(
Temperature

Dew
point
Temperature)
can
be
used
for
determining
the
presence
of
dew.
If
the
temperature
and
dew
point
temperature
are
equal,
dew
is
likely
to
be
present,
because
the
atmosphere
is
saturated,
and
condensation
will
occur.
(
Section­
Page)
17
of
16
Draft
­
Revision
1.2
March
2003
Table
2.4.2.
Weather
data
for
outdoor
testing.

PARAMETER
UNIT
NOA
A
REMARKS
Required
Irradiance
(
UV)
W/
m2
Y
Direct
and
total
radiation
is
available.
Temperature

F
Y
Precipitation,
Duration
hours
Y
Can
be
determined
from
strip
chart,
although
certain
losses
may
occur
due
to
evaporation.
Precipitation,
Amount
inches
Y
Automated
rain
gage.
Dew
Point
(
Measure
of
dew
formation)

F
N
Dew
point
could
be
used
to
calculate
dew
point
depression
(
diff.
with
temp.)
If
DPD
is
small,
there
is
likely
to
be
dew
overnight.

Other
Precipitation,
Type
N/
a
N
Humidity
%
N
Wind
direction
+
speed
Y
Cloud
cover
N
Measure
for
cloud
cover
can
be
deducted
from
total
and
direct
solar
radiation.
pH
rain
N
Likely
to
have
some
effect
on
weathering
Soot,
pollution
N
Likely
to
have
some
effect
on
weathering
The
radiation
instrumentation
has
and
is
calibrated
against
working
standards
that
are
traced
to
world
standards
at
Epply
Laboratories.
This
calibration
is
done
periodically
based
on
the
stability
of
the
instrument.
The
temperature
system
is
checked
against
RDU
(
official
NWS
weather
station)
on
stable
days
and
also
with
a
sling
psychrometer.
The
weighing
rain
gage
is
calibrated
with
weights
and
also
against
a
manual
rain
gage
with
each
precipitation
event.
The
Aerovane
wind
system
records
wind
speed
in
mph
and
only
begins
to
register
at
3
mph.
It
is
also
checked
against
RDU
on
stable
windy
days.
The
operators
of
the
weathering
monitoring
equipment
have
a
great
deal
of
experience
and
their
involvement
and
oversight
is
critical
for
QA/
QC.
(
Section­
Page)
18
of
16
Draft
­
Revision
1.2
March
2003
2.5
Sampling
2.5.1
Wipe
Sampling
Wipe
sampling
shall
be
conducted
and
samples
prepared
and
analyzed
in
accordance
with
the
methods
established
in
section
3.0.

Indoor,
Artificial,
Accelerated
Weathering
Tests
Wipe
samples
shall
be
taken
from
the
top
faces
of
each
specimen
on
a
monthly
basis
by
interrupting
the
QUV/
Spray
cycle,
immediately
following
a
UV
cycle
(
before
the
following
precipitation
cycle
is
implemented),
removing
the
30­
cm
specimens,
and
conducting
wipe
sampling.
Wipe
samples
shall
be
taken
from
the
top
faces
of
each
specimen
only.
The
length
of
wipe
shall
be
25
cm,
as
2.5
cm
adjacent
to
either
cut
end
of
the
specimen
shall
not
be
wiped.
The
back
end
of
the
wipe
backstroke
and
the
front
end
of
the
wipe
frontstroke
shall
be
identified
to
define
the
area
wiped.
The
location
of
the
wipe
stroke
shall
then
be
identified
on
the
specimen
by
marking
its
forward
and
backward
range
on
both
edges
of
the
specimen
using
permanent
marker
on
the
specimen
edge.
This
identified
area
shall
be
the
area
wipe
sampled
during
subsequent
sampling
events.

Individual
baseline
values
of
DA
will
be
determined
for
each
specimen
to
be
coated
and
tested
prior
to
their
initial
placement
inside
the
QUV/
Spray
units.
The
baseline
DA
of
a
specimen
will
be
determined
by
averaging
the
DAs
from
the
two
adjacent
specimens
on
either
side
of
the
test
specimen.
The
adjacent
specimens
will
be
archived
uncoated
after
sampling
(
and
thus
not
used
in
any
further
testing)
in
order
to
avoid
any
data
analysis
and
coating
efficacy
complications
that
may
arise
from
the
pre­
rubbing
of
the
specimen
by
wipe
sampling
for
DA
measurement.

During
sampling
care
must
be
taken
to
ensure
that
the
faces
of
the
specimen
are
not
allowed
to
contact
any
other
surfaces:
handling
and
holding
must
be
via
the
ends
or
edges
of
the
specimens.
A

wipe
sampling
rack

may
need
to
be
constructed
to
hold
the
wood
specimens
in
place
during
wipe
sampling.
After
sampling,
specimens
shall
be
returned
to
the
QUV/
Spray
chambers,
and
assembled
to
randomly
assigned
positions,
as
previously
described.

Outdoor,
Natural
Weathering
Tests
(
Section­
Page)
19
of
16
Draft
­
Revision
1.2
March
2003
Wipe
samples
shall
be
taken
from
the
top
faces
of
three
of
the
six
total
specimens
per
mini­
deck
on
a
monthly
basis
by
wiping
the
specimens
per
the
procedures
described
in
Section
3.0
directly,
on­
site.
The
three
specimens
to
be
sampled
shall
be
consistent
from
month­
to­
month.
The
additional
three
specimens
per
mini­
deck
will
be
sampled
at
the
end
of
the
monitoring
period
(
6­
7
months)
as
a
comparison
with
monthly
sampled
specimens
to
determine
the
effect
of
wipe
sample­
induced
abrasion
on
coating
efficacy.
Wipe
samples
shall
be
taken
from
the
top
faces
of
each
specimen
only.
The
length
of
wipe
shall
be
24
inches.
The
back
end
of
the
wipe
backstroke
and
the
front
end
of
the
wipe
frontstroke
shall
be
identified
to
define
the
area
wiped.
The
location
of
the
wipe
stroke
shall
then
be
identified
on
the
specimen
by
marking
its
forward
and
backward
range
on
both
edges
of
the
specimen
using
permanent
marker.
This
identified
area
shall
be
the
area
wipe
sampled
during
subsequent
sampling
events.
Mini­
decks
shall
be
sampled
only
when
there
has
been
at
least
two
days
since
last

significant
precipitation
event

(
defined
for
these
purposes
as
>
0.1
in.
rain
within
any
24
hour
period).

Individual
baseline
values
of
DA
will
be
determined
for
each
specimen
to
be
coated
and
tested.
The
baseline
DA
of
a
specimen
will
be
determined
by
averaging
the
DAs
from
the
two
adjacent
specimens
on
either
side
of
the
test
specimen.
The
adjacent
specimens
will
be
archived
uncoated
after
sampling
(
and
thus
not
used
in
any
further
testing)
in
order
to
avoid
any
data
analysis
and
coating
efficacy
complications
that
may
arise
from
the
pre­
rubbing
of
the
specimen
by
wipe
sampling
for
DA
measurement.

During
sampling,
again,
it
must
be
ensured
that
the
faces
of
the
specimen
are
not
allowed
to
contact
any
other
surfaces.
(
Section­
Page)
20
of
16
Draft
­
Revision
1.2
March
2003
Unweathered
Control
Specimens
The
screening
test
specimens
for
the
selected
coating
will
be
retained
under
similar
conditions
as
described
for
screening
tested
for
continued
monthly
DA
measurement
during
the
weathering
tests,
as
non­
weathered
controls.
These
samples
will
allow
for
the
monitoring
of
DA
changes
resulting
from
abrasion
(
from
wipe
sampling)
without
the
potentially
confounding
effects
of
weathering.
Monthly
wipe
sampling
will
be
conducted
as
previously
described.

2.5.2
Wood
Sampling
Two,
4­
inch,
interior
wood
specimens
from
each
board
shall
be
sampled,
digested,
and
analyzed
for
total
arsenic
in
accordance
with
the
procedures
described
in
section
3.0.

2.5.3
Photographs
Digital
photographs
of
each
specimen
and
mini­
deck
will
be
made
before
weathering
and
during
monthly
sampling
events.
Visual
observations
per
the
inspection
of
each
specimen
shall
be
documented
in
writing.

2.5.4
Miscellaneous
Samples
Other
miscellaneous
samples
to
be
archived
and/
or
analyzed
are
listed
in
Table
2.5.1.

Table
2.5.1.
Miscellaneous
Samples
to
be
Collected
during
Screening
Study
Sample
Description
#
Samples
to
be
Analyzed
#
Samples
to
be
Archived
Tap
water
used
to
wash
wood
N/
A
1
Rinsate
water
from
wood
washing
N/
A
1
for
each
wood
type
Unaltered
coating
2
for
each
coating
Leftover
coating
to
be
stored
Leftover
brush­
applied
coating
N/
A
1
for
each
coating/
wood
type
Wood
2
per
board
Leftover
wood
to
be
stored
(
Section­
Page)
21
of
16
Draft
­
Revision
1.2
March
2003
Unless
otherwise
stated,
all
samples
indicated
in
Table
2.5.1
to
be
archived
shall
be
held
at
least
until
the
report
of
results
for
the
screening
study
has
been
finalized.
Longer
archiving
times
for
certain
samples
may
be
warranted
upon
further
consideration.
(
Section­
Page)
1
of
3
Draft
­
Revision
1.2
March
2003
3
TESTING
AND
MEASUREMENT
PROTOCOLS
Note
that
this
section
may
be
modified
as
results
of
on­
going
prequalification
testing
of
the
sampling
techniques
and
laboratory
analytical
methods
are
processed.

3.1
Wipe
Sampling
Wipe
sampling
will
be
conducted
in
general
accordance
with
the
method
developed
and
documented
by
CPSC,
using
the
wipe
sampling
device
constructed
by
CPSC.

Wipe
samples
will
be
directly
transferred
to
extraction
or
digestion
vessels
with
no
intermediate
sample
containers
employed.

3.2
Sample
Preparation
(
Digestion)

Wipe
samples
will
be
prepared
for
analysis
using
techniques
similar
to
those
employed
by
other
researchers
including
CPSC
and
Stilwell,
et
al.,
adapted
for
use
with
laboratory
equipment
available
for
this
project.
As
such,
a
microwave­
assisted
extraction
procedure
comparable
to
that
used
in
prior
studies,
and
similar
to
SW­
846
Methods
3051
and
3052,
shall
be
employed.
Steps
involved
in
the
extraction
procedure
are
outlined
following:

1.
All
digestion
vessels
and
volumetric
glassware
will
be
prepared
by
acid
cleaning.
Vessels
will
be
cleaned
by
leaching
with
hot
1:
1
nitric
acid
for
a
minimum
of
two
hours,
then
rinsed
with
deionized
water
and
dried
in
a
clean
environment.

2.
The
wipe
sample
will
be
introduced
to
the
digestion
vessel
and
40

0.1
mL
10%
nitric
acid
added
slowly
to
the
open
vessel
to
allow
for
pre­
extraction.
Once
any
initial
reaction
has
ceased,
the
sample
will
be
capped
and
introduced
into
the
microwave
system.
Using
the
CEM
Corp.
microwave
system,
11
samples
may
be
digested
in
a
single
batch,
with
the
twelfth
vessel
filled
with
only
the
acid
solution
to
be
used
as
a
temperature/
pressure
control.

3.
Using
temperature/
pressure
curves
developed
under
other
research
programs
for
APPCD
as
a
guide,
(
Section­
Page)
2
of
3
Draft
­
Revision
1.2
March
2003
the
samples
will
be
reacted
at
60

5

C
for
a
minimum
of
1
hour.

4.
After
microwave
extraction,
sample
vessels
will
be
allowed
to
cool
for
a
minimum
of
5
min.
prior
to
removing
them
from
the
system.
After
being
allowed
to
cool
to
room
temperature,
each
vessel
will
be
checked
to
ensure
that
they
have
retained
a
seal
(
samples
that
show
signs
of
pressure
relief
will
be
discarded).

5.
Cooled
samples
will
be
carefully
uncapped,
the
contents
transferred
quantitatively
to
a
clean
125
mL
volumetric
flask,
using
a
clean
glass
stir
rod
to
aid
in
squeezing
liquid
from
the
wipe.

6.
Steps
2­
5
will
be
repeated
twice
more,
transferring
the
successive
extraction
liquid
aliquots
to
the
same
volumetric
flask.
After
the
third
iteration,
the
flask
will
be
filled
to
the
mark
with
deionized
water.
Approximately
half
the
final
liquid
volume
will
be
transferred
to
a
PTFE
container,
sealed
and
prepared
for
shipment
to
the
contract
laboratory.
The
remainder
will
be
transferred
to
a
PTFE
container,
sealed
and
archived.

Per
the
specified
analytical
method,
the
hold
time
for
all
metals
other
than
mercury
is
6
months,
and
samples
shall
be
stored
at
4
degrees
C
until
analysis.
Sample
containers
shall
be
of
TFE
or
PFA
in
accordance
with
the
Method
specified
in
Section
3.3.

3.3
Analysis
by
ICP­
MS
Analysis
for
total
arsenic
shall
be
conducted
by
STL
in
Savannah,
GA
using
a
modification
of
SW­
846
Method
6020
(
ICP­
MS).
STL
utilizes
ICP­
MS
for
arsenic
analysis,
modifying
the
technique
to
utilize
hydrogen
plasma,
rather
than
argon
as
classically
performed.
This
modification
eliminates
concerns
over
the
formation
of
Ar40Cl35,
which
can
create
a
positive
bias
when
measuring
As.
STL
is
an
accredited
laboratory,
participating
in
the
CLP
program,
as
well
as
numerous
state
programs.
In
addition
to
obtaining
specific
information
on
laboratory
qualifications,
each
sample
set
submitted
will
include
blind
blanks
and
spiked
samples,
allowing
for
continued
monitoring
of
laboratory
performance.
(
Section­
Page)
3
of
3
Draft
­
Revision
1.2
March
2003
3.4
Preparation
and
Analysis
of
Coating
Samples
Arsenic
and
chromium
in
coatings
material
will
be
determined
in
a
manner
similar
to
that
used
to
analyze
the
wipe
samples
(
acid
digestion/
extraction
followed
by
ICP­
MS).
The
coating
to
be
analyzed
will
be
thoroughly
shaken
to
ensure
homogeneity
and
then
an
aliquot
will
be
transferred
to
a
tared
PTFE
digestion
vessel
and
allowed
to
dry.
Following
loss
of
volatiles
through
drying,
the
residue
will
be
microwave
digested
using
concentrated
nitric
acid
as
described
in
EPA
SW­
846
Method
3052.
Hydrofluoric
acid
may
be
added
if
necessary
to
ensure
complete
digestion
in
accordance
with
the
method.
The
digestate
will
be
quantitatively
transferred
to
a
volumetric
flask
and
diluted
to
a
known
volume
prior
to
submission
to
the
contract
laboratory
for
ICP­
MS
analysis
(
SW­
846
Method
6020).

3.5
Preparation
and
Analysis
of
Wood
Samples
Wood
samples
will
be
analyzed
for
arsenic
and
chromium
content
using
ICP­
MS.
Wood
borings
and/
or
ground
wood
of
known
weight
will
be
digested
using
the
same
protocol
defined
earlier
for
the
wipe
samples
(
SW­
846
Methods
3051
and
3052).
This
procedure
is
consistent
with
American
Wood
Preservers
Association
(
AWPA)
Standard
A7­
93
(
microwave
assisted
nitric
acid
digestion).
Digestates
will
be
analyzed
by
ICP­
MS
in
a
manner
identical
to
that
described
for
the
wipe
samples
(
SW­
846
Method
6020).
This
is
consistent
with
AWPA
Standard
A21­
00.

3.6
Archiving
of
ICP­
MS
Samples
Analysis
of
the
samples
by
ICP­
MS
will
consume
only
a
fraction
of
the
submitted
sample.
Remaining
sample
volume
will
be
archived
by
the
contract
laboratory
until
results
are
confirmed,
then
subsequently
returned
to
ARCADIS
for
archiving
until
the
completion
of
the
project.
Samples
will
be
archived
by
storing
them
in
their
original
sample
containers,
as
shipped
to
the
contract
laboratory,
under
refrigeration.
(
Section­
Page)
4
of
3
Draft
­
Revision
1.2
March
2003
3.7
Moisture
Analysis
of
Wood
Specimens
The
moisture
content
of
wood
samples
will
be
determined
by
ASTM
Method
D4442
(
Primary
Oven
Drying).
A
small
representative
sample
will
be
weighed
prior
to
drying
overnight
at
103
C
in
a
forced
air
oven.
After
24
hours,
the
sample
will
be
cooled
in
a
dessicator,
weighed,
then
returned
to
the
oven.
The
process
will
be
repeated
until
weight
changes
between
weighings
is
within

5%.
Moisture
content
may
also
be
screened
using
a
hand­
held
meter,
but
only
after
the
technique
has
been
qualified
and
calibrated
via
side­
by­
side
testing
with
the
drying
oven
technique,
ASTM
D4442.
(
Section­
Page)
1
of
1
Draft
­
Revision
1.2
March
2003
4
QA/
QC
Checks
One
full
set
of
uncoated
control
specimens
will
be
tested
alongside
coated
samples
for
the
new
and
aged
CCA­
treated
specimens
for
both
weathering
test
protocols.
This
sill
include
two
adjacent
uncoated
control
specimens
to
be
tested
alongside
coated
faces
for
each
specimen
used,
as
previously
described.
Additionally,
untreated
control
specimens
will
be
handled
in
the
same
way
(
with
efforts
to
avoid
crosscontamination
with
CCA
wood
specimens,
of
course)
as
the
treated
samples
in
both
studies
in
order
to
generate
blank
control
wipe
samples
and
assess
the
extent
of
study
design/
method
interferences,
crosscontamination
etc.
Furthermore,
a
variety
of
control
samples
are
proposed,
as
indicated
in
Table
2.5.1.

Each
set
of
digested
wipe
samples
submitted
to
the
subcontract
analytical
laboratory
will
include
an
additional
blind
blank,
one
set
of
three­
concentration
spiked
samples,
and
5%
duplicates
to
assess
laboratory
performance.
So,
for
example,
assuming
that
a
total
of
200
wipe
samples
will
be
taken
for
this
study,
shipped
to
the
subcontract
laboratory
in
a
single
batch,
the
following
additional
samples
will
be
included:

|
one
(
1)
blank
sample
consisting
of
digestion
fluid
only
|
one
(
1)
digestion
fluid
sample
spiked
to
1.0

g/
l
(
0.015

g
in
15
ml
digestion
fluid)
with
As
|
one
(
1)
digestion
fluid
samples
spiked
to
50

g/
l
(
0.75

g
in
15
ml
digestion
fluid)
with
As
|
one
(
1)
digestion
fluid
samples
spiked
to
1000

g/
l
(
15

g
in
15
ml
digestion
fluid)
with
As
|
ten
(
10)
duplicates
(
selected
split
samples
of
digested
wipes
from
actual
samples
generated)

So,
for
example,
assuming
that
330
wipe
samples
are
shipped
to
the
subcontract
laboratory
in
a
single
batch,
the
following
additional
samples
will
be
included:

|
two
(
2)
blank
samples
consisting
of
digestion
fluid
only
|
two
(
2)
digestion
fluid
samples
spiked
to
1.0

g/
l
(
0.015

g
in
15
ml
digestion
fluid)
with
As
and
Cr
|
two
(
2)
digestion
fluid
samples
spiked
to
50

g/
l
(
0.75

g
in
15
ml
digestion
fluid)
with
As
and
Cr
|
two
(
2)
digestion
fluid
samples
spiked
to
1000

g/
l
(
15

g
in
15
ml
digestion
fluid)
with
As
and
Cr
|
seventeen
(
17)
duplicates
(
selected
split
samples
of
digested
wipes
from
actual
samples
generated)

Furthermore,
the
subcontract
analytical
laboratory
will
conduct
analyses
on
project­
specific
matrix
spikes
and
matrix
spike
duplicates
(
MS/
MSD)
for
each
analyte,
in
addition
to
equipment
blanks
run
on
each
batch
of
samples
analyzed
for
this
project.
The
laboratory
will
also
have
chromium
and
copper
analytical
results
(
Section­
Page)
2
of
3
Draft
­
Revision
1.2
March
2003
available
for
all
samples
(
provided
QC
checks
are
acceptable)
if
these
results
are
ever
desired
and
the
project
budget
allows
their
procurement.
(
Section­
Page)
1
of
4
Draft
­
Revision
1.2
March
2003
5
Data
Reductions
and
Reporting
5.1
Data
Reduction
5.1.1
Calculation
of
DA
from
Extraction/
Digestion
Fluid
Concentrations
Raw
data
from
the
subcontract
analytical
laboratory
will
be
reported
in
units
of

g/
l
and
will
represent
the
mass
of
analyte
per
unit
volume
of
extraction/
digestion
solution
sent
to
the
laboratory.
For
standard
wipe
sample
results,
data
will
be
reduced
in
order
to
characterize
the
mass
of
analyte
per
unit
surface
area
wipe
sampled,
in
units
of

g/
cm2,
using
the
following
equation:

0
(
Equation
5.1)

Where:
CDA
=
DA
of
a
sample
(

g/
cm2)

CDF
=
Concentration
of
As
in
extraction/
digestion
fluid
(

g/
l)

V
=
Total
volume
of
extraction/
digestion
fluid
(
ml)

A
=
Area
of
wiped
surface
(
cm2)

5.1.2
Calculation
of
Percent
Reduction
of
DA
Raw
data
from
the
subcontract
analytical
laboratory
will
be
reported
in
units
of

g/
l
and
will
be
converted
to
DA
(
the
mass
of
analyte

arsenic
in
this
case

per
unit
surface
area
wipe
sampled),
in
units
of

g/
cm2,
per
the
calculation
described
in
section
5.1.1.
Percent
reduction
will
be
calculated
for
each
sample
using
the
following
equation:

0
(
Equation
5.2)
(
Section­
Page)
2
of
4
Draft
­
Revision
1.2
March
2003
Where:
RDA
=
Reduction
in
DA
(%)

Cinitial
=
Baseline
DA
(

g/
cm2)

Cfinal
=
Final
DA
(

g/
cm2)

5.1.3
Assessing
DQI
Goals
In
general,
data
quality
indicator
goals
are
based
on
either
(
1)
published
specifications,
(
2)
related
quantities
(
e.
g.,
drift
for
precision),
or
(
3)
engineering
judgment
based
on
previous
experience
with
similar
systems.

Precision
In
order
to
evaluate
the
precision
of
a
measurement,
it
is
necessary
to
make
replicate
measurements
of
a
relatively
unchanging
parameter.
Precision
can
then
be
expressed
as
the
relative
standard
deviation
(
RSD)
of
the
replicated
measurement.
RSD
is
calculated
using
Equation
5.4
and
is
typically
expressed
in
percent.

0
(
Equation
5.4)

Precision
will
be
calculated
using
the
results
of
duplicates
specified
as
control
samples.

Accuracy/
Bias
The
accuracy
of
a
measurement
is
expressed
in
terms
of
percent
bias,
or,
in
some
cases
recommended
by
the
EPA
standard
methods,
in
terms
of
absolute
difference.
Percent
bias
is
defined
as:
(
Section­
Page)
3
of
4
Draft
­
Revision
1.2
March
2003
0
(
Equation
5.5)

Where:
R
=
instrument
response
or
reading
C
=
calibration
standard
or
audit
sample
value
Accuracy
can
take
on
the
units
of
the
measurement,
it
can
be
expressed
as
a
percentage
of
the
average
measurement,
or
it
can
be
expressed
as
a
percentage
of
the
measurement
range.
Accuracy
will
be
calculated
using
the
results
of
matrix
spike
sample
analyses
as
described
for
QA/
QC.

Completeness
The
ratio
of
the
number
of
valid
data
points
taken
to
the
total
number
of
data
points
planned
is
defined
as
data
completeness.
All
measured
data
are
recorded
electronically
or
on
data
sheets
or
project
notebooks.

5.2
Data
Validation
The
subcontract
laboratory
will
be
required
to
submit
calibration
and
QC
data
along
with
each
data
package.
ARCADIS
QA
Officer,
Libby
Nessley
will
validate
at
least
10
percent
of
reported
data
by
reviewing
raw
data
and
data
calculations.
In
addition,
at
least
one
spiked
performance
evaluation
audit
(
PEA)
sample
for
arsenic
will
be
submitted
blind
to
the
laboratory
with
each
sample
set.
Reported
results
for
this
PEA
sample
must
agree
within
10
percent
with
the
known
value.
Failure
to
agree
will
result
in
the
entire
data
set
being
flagged
for
re­
evaluation
up
to
and
including
repeat
analysis.

5.3
Data
Reporting
For
each
series
of
tests,
raw
and
reduced
data
shall
be
reported,
as
applicable.
Coating
efficacy
results
will
be
expressed
in
terms
of
DA
(

g/
cm2)
and
percent
reduction.

All
data
validation
criteria
will
be
reported
along
with
the
associated
data.
In
addition
to
reporting
the
data,
(
Section­
Page)
4
of
4
Draft
­
Revision
1.2
March
2003
results
will
be
discussed
and
recommendations
for
coatings
to
be
tested
in
the
weathering
study
offered
via
narrative
text
in
a
test
report.
This
report
will
be
used
to
support
the
development
of
and
append
the
test
plan/
QAPPs
for
future
phases
of
this
project
including
the

Weathering
Study

(
i.
e.,
the
overall
test
plan/
QAPP).

5.4
Relational
Database
Development
Data
will
be
compiled
into
a
relational
database
that
provides
a
generic
description
of
each
coating
(
formulation
data),
application
date,
method
and
amount,
pretest
treatment
(
e.
g.,
drying
period
before
chamber
testing),
testing
details
(
irradiance,
humidification
and
precipitation
cycles),
temperatures,
sampling
date
and
time,
temperature
at
time
of
sampling,
observations
of
sample
condition,
DA
results,
and
other
applicable
information.

A
prototype
of
database,
user
interfaces,
record
tables,
data
entering,
and
basic
searching
and
filtering
will
be
constructed
and
submit
to
the
WAM
for
review
and
comments.
The
initial
design
of
the
prototype
database
application
will
focus
on
the
ease
of
data
entering
and
minimization
of
data
entry
error
by
the
use
of
predefined
terms
as
much
as
practical.
The
user
interface
will
be
designed
for
the
ease
of
use
for
basic
searching
and
reporting
so
that
the
user
will
not
need
to
be
able
to
program
the
database
in
order
to
retrieve
most
of
the
information.

After
receiving
comments,
modifications
will
be
made
in
consultation
with
the
WAM
and
the
final
specification
of
the
data
fields,
type
of
output,
and
user
interface
will
be
agreed
upon.
Once
the
final
specifications
are
determined,
the
actual
database
and
the
related
querying
and
output
capabilities
(
or
database
application)
implementation
will
begin.
Somewhere
in
the
middle
of
the
development
of
the
database
application,
it
will
be
reviewed
again
the
WAM
for
any
further
comments
and
changes.
This
step
can
be
important
because
experimental
work
will
have
progressed
sufficiently
to
warrant
changes
in
the
database
application

s
specifications.
Depending
on
the
comments,
one
or
more
revisions
of
the
design
may
be
necessary.
The
revisions
will
be
implemented
until
it
is
approved
by
the
WAM.

5.5
Regular
Reporting
ARCADIS
will
provide
the
EPA
WAM
with
weekly
verbal
progress
updates
as
well
as
monthly
written
progress
reports.
Data
reports
will
be
prepared
and
include
all
sampling
and
analysis
data,
calibration
data,
quality
control
data,
and
a
data
quality
evaluation.
(
Section­
Page)
1
of
1
Draft
­
Revision
1.2
March
2003
6
Assessments
Assessments
are
integral
parts
of
a
quality
system.
This
project
is
assigned
a
QA
Category
II
and
will
require
planned
technical
systems
and
performance
evaluation
audits.
The
EPA
QA
Manager
will
coordinate
any
audits
with
the
EPA
WAM.
The
ARCADIS
QAO
will
also
perform
at
least
one
internal
technical
systems
audit
in
the
early
stages
of
this
project.
This
audit
will
be
coordinated
with
the
ARCADIS
WAM.
In
addition,
the
ARCADIS
QAO
will
perform
an
audit
of
data
quality
prior
to
the
release
for
any
formal
reports.
This
audit
will
review
at
least
10%
of
the
data
from
collection
to
reporting.
Calculations
will
be
checked,
laboratory
analytical
reports
will
be
reviewed,
and
hand­
entered
data
will
be
validated.
Appendix
B
Draft
­
Revision
1.2
March
2003
Appendix
A
Test
Plan:
Screening
and
Selection
of
Coatings
for
Reducing
Dislodgeable
Arsenic
from
CCA
Treated
Wood
Appendix
B
Draft
­
Revision
1.2
March
2003
Appendix
B
List
of
Key
References
Appendix
B
Draft
­
Revision
1.2
March
2003
ACC.
2002.
Protocol
for
Hand
and
Wipe
Sampling
of
CCA­
Preserved
Wood.
American
Chemistry
Council.
September
9,
2002.

Arwood
Programme.
2001
a.
Final
Report

Technical
Work
Programme:
A
Reliable
Artificial
Weathering
Test
for
Wood
Coatings.

Arwood
Programme.
2001
b.
Standard
Pr
ENV
927­
6.
Method,
Assessment
and
Evaluation
for
an
Artificial
Weathering
Test
for
Wood
Coatings
Based
on
Fluorescent
Lamp
Apparatus.
January
2001.

ASTM
D
823­
95.
1995.
Standard
Practices
for
Producing
Films
of
Uniform
Thickness
of
Paint,
Varnish
and
Related
Products
on
Test
Panels.

ASTM
D
147­
96.
1996
a.
Standard
Practice
for
Conditioning
and
Handling
of
Nonmetallic
Materials
for
Natural
and
Artificial
Weathering
Tests.

ASTM
E
1792­
96.
1996
b.
Standard
Specification
for
Wipe
Sampling
Materials
for
Lead
in
Surface
Dust.

ASTM
D
358­
998.
1998.
Standard
Specification
for
Wood
to
Be
Used
as
Panels
in
Weathering
Tests
of
Coatings.

ASTM
G
154­
00a.
2000.
Standard
Practice
for
Operating
Fluorescent
Light
Apparatus
for
UV­
Exposure
of
Nonmetallic
Materials.

ASTM
D
169­
01.
2001.
Standard
Guide
for
Application
of
Basic
Statistical
Methods
to
Weathering
Tests.

Cochran
and
Cox

Experimental
Design

,
2
ed.
Wiley,
NY.

Consumer
Reports.
1998
a.
All
Decked
Out/
Ratings.
June
1998.

Consumer
Reports.
2001.
Deck
Treatments

Few
Good
Choices/
Ratings.
June
2001.

Consumer
Reports.
2002
a.
Deck
Treatments

What

s
New/
Ratings.
June
2002.

Consumer
Reports.
2002
b.
House
Stains
that
Hold
up
for
the
Long
Haul/
Ratings.
October
2002.

Fedor,
G.,
P.
Brennen.
May
1990.
Correlation
of
Accelerated
and
Natural
Weathering
of
Sealants.
Adhesives
Age.
Appendix
B
Draft
­
Revision
1.2
March
2003
Fischer,
R.
M.,
W.
D.
Ketola.
1993.
The
Use
of
Nonparametric
Statistics
in
Accelerated
Weathering
Test
Design
and
Development.
SPE
Polymer
Modifiers
and
Additives
Division
Newsletter,
Vol.
19,
No.
2
(
1993),
9­
15.

Hare,
C.
H.
May
1992.
The
Degradation
of
Coatings
by
Ultraviolet
Light
and
Electromagnetic
Radiation.
Journal
of
Protective
Coatings
and
Linings.

Hicks,
Charles,

Fundamental
Concepts
in
the
Design
of
Experiments

,
Saunders
College
Publishing,
Fort
Worth.

Lebow,
S.
August
1996.
Leaching
of
Wood
Preservative
Components
and
Their
Mobility
in
the
Environment,
Summary
of
Pertinent
Literature.
USDA
Forest
Service,
Forest
Products
Laboratory.

Maas,
R.
P.,
S.
T.
Patch,
A.
M.
Stork,
J.
F.
Berkowitz,
G.
A.
Stork.
February
2002.
Release
of
Total
Chromium,
Chromium
VI
and
Total
Arsenic
from
New
and
Aged
Pressure
Treated
Lumber.
Technical
Report
01­
093.
University
of
North
Carolina­
Asheville
Environmental
Quality
Institute.

Mandal,
B.
K.,
K.
T.
Suzuki.
2002.
Arsenic
Round
the
World:
A
Review.
Talanta.
Vol.
58,
201­
235.

Miller,
Denise.
June
4,
2001.
Memorandum
to
Bill
Hinkley.
Evaluation
of
Sealants
for
and
Alternatives
to
the
use
of
CCA­
Treated
Picnic
Tables.

Q­
Labs.
1987.
P.
Brennan.
Improved
UV
Light
Source
Enhances
Correlation
in
Accelerated
Weathering.
Q­
Panel
Lab
Products.
March/
April
1987.

Q­
Labs
­
a.
P.
Brennan,
C.
Fedor.
Sunlight,
UV,
and
Accelerated
Weathering.
Laboratory
Q­
Panel
Lab
Products.
Technical
Bulletin
LU­
0822.

Q­
Labs
­
b.
D.
M.
Grossman.
Know
Your
Enemy:
The
Weather
and
How
to
Reproduce
it
in
the
Laboratory
Q­
Panel
Lab
Products.
Technical
Bulletin
LU­
0821.

Q­
Labs
­
c.
D.
M.
Grossman.
Correlation
Questions
and
Answers,
A
Discussion
of
the
most
Frequently
Asked
Questions
about
Accelerated
Weathering.
Laboratory
Q­
Panel
Lab
Products.
Technical
Bulletin
LU­
0833.

Q­
Labs.
2002.
QUV
Accelerated
Weathering
Tester
Operating
Manual.

Q­
Labs
­
d.
Evaluation
of
Weathering
Effects:
Visual
Inspections
and
Instrumental
Measurements.
Laboratory
Q­
Panel
Lab
Products.
Technical
Bulletin
LU­
9030.

Q­
Labs
­
e.
High
Irradiance
UV/
Condensation
Testers
Allow
Faster
Accelerated
Weathering
Test
Results.
Appendix
B
Draft
­
Revision
1.2
March
2003
Laboratory
Q­
Panel
Lab
Products.
Technical
Bulletin
LU­
8031.

Q­
Labs
­
f.
Controlled
Irradiance
in
Laboratory
Weathering:
Limitations
in
the
State
of
the
Art.
Laboratory
Q­
Panel
Lab
Products.
Technical
Bulletin
LU­
8010.

Q­
Labs
­
g.
A
Choice
of
Lamps
for
the
QUV.
Q­
Lab
Technical
Bulletin
LU­
8160.

Stilwell,
D.
E.
December
1998.
Environmental
Issues
on
the
Use
of
CCA
Treated
Wood.
Fact
Sheet.
Department
of
Analytical
Chemistry,
The
Connecticut
Agricultural
Experiment
Station.

Stilwell,
D.
E.
1999.
Arsenic
in
Pressure
Treated
Wood.
Fact
Sheet.
Department
of
Analytical
Chemistry,
The
Connecticut
Agricultural
Experiment
Station.

Stilwell,
D.
E.
2002
a.
Arsenic
Dislodged
from
CCA
Surfaces

Effects
of
Coatings.
Power
Point
Presentation.

Stilwell,
D.
E.
2002
b.
Uptake
of
Arsenic
by
Plants
Grown
Near
CCA
Preserved
Wood.
2002.
Project
progress
report.

US
CPSC.
2001.
Protocol:
Sampling
Chromated
Copper
Arsenate
(
CCA)

Pressure

Treated
Wood
Playground
Equipment
for
Dislodgeable
Residues
of
Arsenic,
Chromium,
and
Copper.
U.
S.
Consumer
Product
Safety
Commission,
Washington,
D.
C.

US
CPSC.
2002.
Status
Report
on
CCA
Pressure­
Treated
Wood
in
Playground
Equipment.
U.
S.
Consumer
Product
Safety
Commission,
Washington,
D.
C.
February
15,
2002.

US
CPSC.
2003
a.
Mitigation
Cross
Contamination
Study.
Raw
Data.
U.
S.
Consumer
Product
Safety
Commission,
Washington,
D.
C.
January
26,
2003.

US
CPSC.
2003
b.
MEMORANDUM:
Statistical
Analyses
of
CCA­
Treated
Wood
Study
Phases
I
and
II.
U.
S.
Consumer
Product
Safety
Commission,
Washington,
D.
C.
January
10,
2003.

US
CPSC.
2003
c.
MEMORANDUM:
Statistical
Analysis
of
CCA­
Treated
Wood
Study
Phase
III.
U.
S.
Consumer
Product
Safety
Commission,
Washington,
D.
C.
January
13,
2003.

US
CPSC.
2003
d.
Standard
Operating
Procedure
for
the
Collection
and
Analysis
of
Surrogate
Wipes
on
CCA
Treated
Wood.
U.
S.
Consumer
Product
Safety
Commission,
Washington,
D.
C.

USDA.
1995.
Paint,
Stain,
Varnish,
or
Preservative?
It

s
Your
Choice.
USDA
Forest
Service,
Forest
Products
Laboratory.
December
1995.

USDA.
2001.
Coatings
Minimize
Leaching
From
Treated
Wood.
Techline
Durability.
USDA
Forest
Appendix
B
Draft
­
Revision
1.2
March
2003
Service,
Forest
Products
Laboratory.
November
2001.

Veenin,
A.,
T.
Veenin.
2001.
A
Laboratory
Study
on
Effect
of
Coating
Materials
on
Leaching
of
Copper
from
CCA
Treated
Wood.
32nd
Annual
Meeting
of
The
International
Research
Group
on
Wood
Preservation.
May
20­
25,
2001.
Nara,
Japan.

Williams,
R.
S.,
P.
Sotos,
W.
C.
Feist.
1999.
Evaluation
of
Several
Finishes
on
Severely
Weathered
Wood.
Journal
of
Coatings
Technology.
Vol.
71,
No.
895.
August
1999.
