1
TELECON
Call
To:
Gail
Froiman,
US
EPA
Office
of
Environmental
Information
202­
566­
0697
Call
From:
Jill
Lucy
Project:
US
EPA
Office
of
Water,
Engineering
and
Analysis
Division
304(
m)
Detailed
Studies
Charge
Number:
0172.01.005.063
Date:
November
13,
2003
Subject:
Polycyclic
Aromatic
Compounds
(
PACs)
Wastewater
Discharges
Reported
to
the
Toxic
Release
Inventory
(
TRI)

I
contacted
Ms.
Froiman
to
investigate
whether
the
Office
of
Environmental
Information
had
any
information
concerning
the
reliability
of
PACs
discharges
reported
to
TRI
(
based
on
discussions
with
industry,
etc.)
or
any
suggestions
of
approaches
to
using
the
TRI
data
for
EAD's
analysis.

BACKGROUND
To
prepare
the
2004/
2005
effluent
guidelines
plan,
EAD
is
evaluating
industrial
point
source
wastewater
discharges
of
toxic
or
non­
conventional
pollutants.
The
purpose
of
the
evaluation
is
to
establish
a
schedule
for
the
annual
review
and
revision
of
all
effluent
guidelines
(
i.
e.,
a
prioritized
list
of
existing
effluent
guidelines).
EAD's
evaluation
is
also
being
used
to
identify
categories
of
point
sources
discharging
toxic
or
non­
conventional
pollutants
for
which
EPA
has
not
published
effluent
guidelines
(
i.
e.,
new
categories).

EAD
analyzed
readily
available
information
to
rank
industries
based
on
potential
hazard.
The
"
readily
available"
information
includes
the
TRI
data
and
Permit
Compliance
System
(
PCS)
data
for
2000.

To
estimate
the
potential
hazard
of
the
discharges,
EAD
multiplied
the
mass
discharged
by
their
toxic
weighting
factors
(
TWFs)
­
these
have
been
used
over
the
years
for
other
effluent
guidelines
calculations.
EAD
assigns
chemical
toxicity
based
on
both
aquatic
life
effects
and
human
health
effects
and
additively
combines
them
into
one
pollutant­
specific
TWF.
EAD
then
uses
the
calculated
"
toxic­
weighted
pound
equivalents"
in
its
cost
effectiveness
analysis
for
developing
new
effluent
guidelines.

EAD
has
TWFs
for
eight
of
the
compounds
in
the
PACs
category;
however,
sites
report
to
TRI
2
Telecon
­
November
13,
2003
only
the
sum
of
the
discharges
for
the
PACs
category
(
not
specific
compounds).
For
the
screening
level
analysis,
EAD
estimated
the
potential
hazard
for
PACs
discharges.

The
PACs
Category
reportable
to
TRI
(
as
required
by
EPCRA
Section
313),
includes
21
compounds.
In
addition,
benzo(
g,
h,
i)
perylene
is
also
a
PACs,
but
reported
separately
for
TRI
purposes.
From
the
August
2001
EPCRA
Section
313
Guidance
for
Reporting
Toxic
Chemicals:
Polycyclic
Aromatic
Compounds
Category
document:

Most
of
the
21
listed
PACs
are
products
of
incomplete
combustion;
see
Table
1­
1.
Two
exceptions,
7,12­
dimethylbenz(
a)
anthracene
and
3­
methylcholanthrene,
are
produced
in
small
quantities
as
research
chemicals
and
are
not
products
of
incomplete
combustion
(
1).
Twelve
of
the
21
are
reported
to
be
found
in
fossil
fuels
(
1).
EPCRA
Section
313
PAC
category
chemicals
are
also
found
in
coal
tar
and
coal
distillates.

NOTES
ON
CALL
Ms.
Froiman
noted
that
there
has
been
some
discussion
with
industry
on
the
potential
for
over
reporting
of
PACs
discharges.
Sites
may
use
estimated
production
rates
(
i.
e.,
manufacture
of
PACs
as
products
of
incomplete
combustion)
as
the
release
rates.
PACs
can
be
created
or
destroyed
during
combustion.

For
TRI
reporting,
sites
may
use
monitoring
data
or
engineering
calculations/
estimates
to
determine
the
amount
of
PACs
discharged
to
the
wastewater.
More
reporting
data/
estimate
factors
are
available
for
PACs
air
emissions
than
for
PACs
wastewater
discharges
(
e.
g.,
EPA
has
developed
emission
factors
for
PACs).
Ms.
Froiman
noted
that
we
could
use
various
methods
to
assist
in
developing
estimates
of
PACs
discharges
(
or
to
determine
the
validity
of
TRI
data):

1.
To
estimate
wastewater
discharges,
we
could
assume
the
PACs
concentration
in
wastewater
releases
equals
those
in
air
emissions.
EPA
has
air
guidance
data
on
seven
PACs
(
e.
g.,
Locating
&
Estimating
Air
Emissions
from
Sources
of
Polycyclic
Organic
Matter,
EPA­
454/
R­
98­
014):
benz(
a)
anthracene,
benzo(
a)
phenanthrene
(
chrysene),
benzo(
b)
fluoranthene,
benzo(
k)
fluoranthene,
benzo(
a)
pyrene,
dibenz(
a,
h)
anthracene,
and
indeno(
1,2,3­
cd)
pyrene.

2.
PACs
are
not
very
soluble
in
water,
and
this
solubility
limit
could
provide
a
concentration
limit
for
the
amount
of
PACs
discharged
in
the
wastewater.
[
However,
this
would
neglect
PACs
that
may
be
released
as
in
an
oil
layer,
oil/
water
emulsion,
or
associated
with
suspended
solids].

3.
Compare
the
developed
emission
factors
for
the
seven
PACs
compounds
(
listed
above)
with
the
concentration
in
fuels.
If
the
values
are
similar,
use
the
concentration
in
the
fuels
to
estimate
wastewater
discharges.

Ms.
Froiman
suggested
we
could
evaluate
PACs
estimates
as
reported
to
TRI
by
using
reporting
data
for
benzo(
g,
h,
i)
perylene
­
since
this
compound
is
reported
separately
to
TRI,
we
may
be
able
3
Telecon
­
November
13,
2003
to
determine
how
sites
calculated
the
value
and
assume
the
PACs
category
releases
were
reported
in
the
same
manner.

Some
other
benchmarking
data
sources
include
the
Center
for
Disease
Control's
human
exposure
of
chemicals.
There
is
a
January
2003
report
(
publication
number
NCEH­
02­
0716)
that
provides
the
metabolite
of
individual
PACs
(
the
data
does
show
a
fair
number
of
nondetects).

Other
possible
wastewater
data
sources
we
discussed
included
determining
if
other
Office
of
Water
projects
contained
data
on
PACs.
Some
possible
data
sources
include
mobile
diesel
data,
impact
assessment
of
petrochemicals
on
water,
or
drinking
water
standards.

Some
American
Petroleum
Institute
(
API)
documents
may
also
be
helpful
(
e.
g.,
those
referenced
in
the
TRI
guidance
document
for
PACs).

Ms.
Froiman
also
suggested
talking
with
individuals
who
have
completed
TRI
reports
for
industry
­
they
may
have
more
data
than
needed
to
be
reported
(
i.
e.,
specific
PACs
compound
data
that
was
summed
to
report
to
TRI
for
the
PACs
Category).

Ms.
Froiman
and
I
also
discussed
the
Risk
Screening
Environmental
Indicators
(
RSEI)
program.
The
program
does
have
some
general
assumptions
that
may
not
translate
accurately
to
the
discharges
reported
to
TRI.
The
model
assumes
unspecified
compounds
are
a
higher
toxicity
(
e.
g.,
all
chromium
compounds
are
assigned
the
toxicity
for
chrome­
6
 
this
toxicity
is
higher
than
other
chromium
compounds).
