Study Of Hazardous Air Pollutant Emissions From Electric Utility Steam Generating Units Interim Final Report, Volume 3 Appendices H-M


          United Slates
          Environmental Protection
          Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-453/R-96-013C
October 1996
          Air
5  EPA  Study of Hazardous Air Pollutant
          Emissions from Electric Utility Steam
          Generating Units — Interim Final Report
          Volume 3.  Appendices H - M

-------
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard,  12(\\ 1 '•••*>»
Chicago,  IL  60604-3590

-------
                                     TABLE OF CONTENTS
  Appendix


  Appendix H


  Appendix I

  Appendix J

  Appendix K

  Appendix L

  Appendix M
                                                                       Page
Summary of Speciation, Environmental Chemistry,  and Fate of Eight HAPs
Emitted from Utility Boiler Stacks	»  .  .   H-l

Summary'of EPRI's Utility Report 	   1-1

Parameter Justifications: Scenario Independent Parameters   	   J-l

Parameter Justifications Scenario-Dependent Parameters  	   K-l

Mercury Partition Coefficient Calibrations 	   L-l

Description of Exposure Models   	   M-l
 L
u
...y
                                              11

-------
 Appendix H—Summary of  Speciation, Environmental  Chemistry,  and
      Fate of Eight HAPs Emitted from Utility Boiler Stacks
Appendix H is followed by Appendix H-l, List of Utility Boiler
Test Reports, and Appendix H-2, Utility Boiler Stack Emisisons of
Furans/Dioxins, PAHs, and Six Trace Metals.

-------
This page is intentionally blank.

-------
H.1  INTRODUCTION

     Under Section 112(n)(1)(A) of the 1990 Clean Air Act
Amendments (CAAA),  Congress mandates that the U.S. Environmental
Protection Agency (EPA) study the hazards to public health
reasonably anticipated to occur as a result of emissions of
hazardous air pollutants  (HAPs) by electric utility steam
generating units (utilities).  EPA was to present the results of
this study in a Report to Congress by November 15, 1995.

     As part of this study, the EPA is conducting direct and
indirect human exposure modeling.  This document was prepared as
part of EPA's indirect exposure modeling effort.  It discusses
the environmental chemistry, speciation, and fate of eight HAPs:
arsenic, cadmium, chromium, lead, mercury, nickel,
dioxins/furans, and polycyclic aromatic hydrocarbons (PAHs).
The eight HAPs were investigated as to their form when emitted
from a utility boiler stack  (Section H.2) and their reactions (if
any) and fate in the atmosphere, water, and soils (Section H.3)
to better understand what HAPs or reaction products of HAPs
humans are most likely to be exposed to via various routes of
exposure (e.g., oral [ingestion], inhalation, dermal [skin])  as a
result of emissions from utilities.  This document summarizes the
available information.

     In preparing this report, test reports, published and
unpublished literature, and other readily available information
such as government documents  (e.g., Agency for Toxic Substances
and Disease Registry [ATSDR] toxicological profiles) and
reference books were reviewed to identify information on the
physical state, species, oxidative state, and anticipated aquatic
and soil reaction products of the eight HAPs.

H.2  UTILITY TEST RESULTS

     As part of a Congressionally mandated study of HAPs from
electric steam generating units  (utilities), the Department of
Energy  (DOE), the Northern States Power Company  (NSP),  and the
Electric Power Research Institute  (EPRI) conducted emissions
tests at over 40 power plants for a variety of HAPs potentially
present.  Nine of the tests were performed for DOE,  eight for
NSP, and the remainder for EPRI.  The EPRI sites were identified
only by a site number; the other sites were identified by name.
Of the tests performed, 33 were on coal-fired boilers,  12 were on
oil-fired boilers,  and 2 were on gas-fired boilers.   The Agency
reviewed test reports from 47 recent tests performed at
43 utility boiler sites.  (Appendix H-l lists the 47 test
reports.)

                               H-l

-------
     Emission control systems tested for the coal-fired boilers
included 13 electrostatic precipitators (ESPs),  3 fabric filters
(FFs),  5 combined ESP flue-gas desulfurization (FGD)  units,
6 combined FF FGD units, and one each of the following: a
combined hot-side, cold-side ESP,  an ESP with ammonia injection,
a combined ESP and pulse-jet FF, and a fluidized-bed combustor
(FBC) boiler with FF.  Emission control systems  for the oil-fired
units were one ESP, one FF,  one pulse-jet FF,  two units that
combined magnesium oxide (MgO) slurry fuel additive and an ESP,
and one unit using MgO fuel additive and a selective catalytic
reduction  (SCR) nitrogen oxides/sulfur oxides (NOX/SOX) removal
process.  Seven of the oil-fired units had no emission control,
and neither of the gas-fired units had emission  controls.  For
some sites, more than one emission control system was tested,
either as a pilot unit in parallel with an existing system or
after replacement of an existing system with a new system.  The
representativeness of the data should allow the  descriptive
statistics presented here to withstand scrutiny.

     All of the tests discussed here were performed in 1990 or
later,  with most of them performed in 1992 and 1993.   The timing
of the testing is important because of test method validity.  For
example, mercury tests performed before the introduction of new
test methods in 1990 may provide suspect results.

     The type of fuel, boiler, and emission controls can affect
HAP emissions from a boiler stack.  The data presented in this
report could be analyzed to determine the qualitative effects of
each of these variables for the trace metals and possibly for the
dioxins/furans and PAHs.

H.2.1  Frequency of HAPs Occurrence
     Appendix H-2 contains air emissions for the 6 trace metals
(including some speciation for mercury, arsenic,  chromium, and
nickel), 16 individual polychlorinated dioxins and furan isomers
and 9 congeners, and 29 PAHs  (classified as polycyclic organic
matter [POM] on the CAAA HAPs list) measured during the testing
at utility boiler stations discussed in Section  2.0.   These
tables are based on 47 tests and 59 HAPs.  Unlike other
industries that measure emissions as metric weight per dry
standard cubic meter  (e.g.,  ng/dscm) , the unit of measure in
Appendix H-2 is lb/1012 British  thermal unit  (Btu) .  This unit  of
measure is commonly used in the utility industry to report
emissions relative to the amount of fuel burned.

     At least one of the metals was measured in all of the
47 tests reported.  In addition, mercury was measured in
speciated forms in 10 emission tests, speciated chromium in

                               H-2

-------
11 tests, speciated nickel in 2 tests,  and speciated arsenic in
2 tests at one site.  Dioxins and furans were measured at 12 test
sites, but many of the reported averages contained one or more
values below the detection limit.  Of the 25 individual dioxin
and furan isomers and congeners, most were found at 10 of the
test sites.  One or more of the 29 PAHs were found at 24 of the
tests.  However, measurements were scattered among sites and
among individual PAHs.   The most frequently measured PAHs were
2-methylnaphthalene, fluorene, fluoranthene, phenanthrene,  and
pyrene.  All 29 PAHs except 1-chloronaphthalene,
dibenz(a,j)acridine, 3-methylcholene,
7,12-dimethyIbenz(a)anthracene, nitranthracene, and perylene were
measured at least once.

H.2.2  Data Quality
     The eight HAPs discussed in this report are present only in
trace quantities in utility boiler stacks.  In many cases,
concentrations are near the detection limit of the analytical
method being used.   Additionally, the sampling process is carried
out in large ducts  (up to about 40 or 50 feet wide) carrying hot,
particulate-laden gas streams that usually have irregular flow
patterns.  The combustion process may be irregular as to fuel
characteristics and will change with load changes-in the
generating system.   For example, coal fired on one day may be
mined from the center of a seam rich in a trace metal, while coal
fired on another day may be mined from a different area of the
seam that is less rich in the trace metal.  When power demand on
a boiler is reduced, less fuel is fired and the combustion zone
changes in dimension.  However the size of the combustion chamber
remains the same.  This situation might change the distribution
of combustion products in the flue gas and the partitioning of
elements between fly ash and bottom ash.   All of these factors
combine to produce final analytical results that should be
accepted with caution.

     Additional uncertainties associated with combining measured
concentrations of an analyte with averages of measured flow rates
in a series of tests made over several days need to be considered
when using these data.   For example, the heating value of a coal
feed may be measured on a Monday, but the concentration of a
compound in the flue gas may not be measured until Tuesday.  A
resulting value of emission rate in lb/1012 Btu will contain
errors caused by the time difference in measurements.

     Because of low or nonexistent concentrations of analytes,
many measurements resulted in nondetects.  Nondetects occurred
when concentration values were so low that the analytical method
could not determine if the analytes were present.  For such

                               H-3

-------
situations, the presence of an analyte in any individual test of
a series (as shown by a concentration above the detection limit)
was taken to mean that the analyte was present.  The question
then arose as to what concentration value to assign to any other
measurement in the series when the measurement was below the
detection limit.  Common choices have been zero, half the
detection limit, or the detection limit.  For this report, tests
showing nondetects were assumed to have an analyte concentration
equal to half the detection limit.

     In Tables H-2-1 through H-2-3 of Appendix H-2, essentially
all values are calculated averages of more than one measurement.
If all measurements forming one average are above the detection
limit, the average is shown in an open box.  If the average
contains one or more nondetect values, the box is shaded.  All
averages contain at least one value above the detection limit,
either at the stack or at some point prior to the stack.

H.2.3  Speciation Data
     For the HAPs examined in this report, all 25 isomers and
congeners of dioxins and furans are reported, as are most of the
individual compounds listed as PAHs.  Species within the trace
metals are available for 20 tests and only for mercury, chromium,
nickel, and arsenic.  Each of the three categories of HAPs is
discussed below.  Numerical emission values are given in Tables
H-2-1 through H-2-3 of Appendix H-2.

     H.2.3.1  Dioxins and Furans.  Table H-l summarizes
Appendix H-2 emissions of dioxins and furans measured at utility
boilers.

     In general, 2,3,7,8-TCDD and 2,3,7,8-TCDF emissions were
approximately 10"6 lb/1012 Btu.  With a few exceptions  at 1C)'5
lb/1012 Btu and  10'4  lb/1012  Btu and one exception at 10~3 lb/1012
Btu, other forms had similar values.

     H.2.3.2  PAHs.  Eight of the PAHs are of particular interest
due to their probable carcinogenicity under EPA's category B2
(weight of evidence from animal testing).  These are
benz(a)anthracene, benzo(b)fluoranthene, benzo(b and
k)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, chrysene,
dibenz(a,h)anthracene, and indeno(l,2,3-c,d)pyrene.  Table H-2
shows median values and ranges for each of the eight as well as
other frequently measured PAHs.
                               H-4

-------
Table H-l.   Emissions  of Dioxins/Furans from Utility Boilers
(lb/1012 Btu)
Specific Dioxin or Furan
2,3,7,8-Tetrach!orodibenzo-p-dioxin
1,2,3,7,8-Pentachlorodibenzo-p-dioxin
1,2, 3,4,7, 8-Hexachlorodibenzo-p-dioxin
1,2,3,6,7,8-Hexachlorodibenzo-p-dioxin
1,2,3,7,8,9-Hexachlorodibenzo-p-dioxin
1,2,3,4,6, 7, 8-Heptachlorodibenzo-p-dioxin
Octachlorodibenzo-p-dioxin
2,3,7,8-Tetrachlorodibenzofuran
1,2,3,7,8-Pentachlorodibenzofuran
2,3,4,7,8-Pentachlorodibenzofuran
1,2,3,4,7,8-Hexachlorodibenzofuran
1,2,3,6,7,8-Hexachlorodibenzofuran
1,2,3,7,8,9-Hexachlorodibenzofuran
2,3,4,6,7,8-Hexachlorodibenzofuran
1,2,3,4,6,7,8-Heptachlorodibenzofuran
1,2,3,4,7,8,9-Heptachlorodibenzofuran
Octachlorodibenzofuran
Tetrachlorodibenzo-p-dioxin
Pentachlorodibenzo-p-dioxin
Hexachlorodibenzo-p-dioxin
Heptachlorodibenzo-p-dioxin
Tetrachlorodibenzofuran
Pentachlorodibenzofuran
Hexachlorodibenzofuran
Heptachlorodibenzofuran
Median
2.30x 10"6
4.38x 10'6
1 .05 x 1 0'5
5.44x 10'6
8.34x 10'6
1.23x 10'6
4.07 x 10*
3.94x 10"6
2.89 x 10'6
6.51 x 10*
8.08 x 10-6
3.99 x 10'6
6.30x 10'6
1.05x 10'5
1.67x 10'5
1.01 x 10'5
1.39 x 10'5
6.61 x 10'6
5.76x 10"6
1.60x 10'5
2.56x 10-5
9.68 x 10"6
1.16x 106
1.50 x 10'5
1.91 x 10'5
Maximum / Minimum
6.51 x 10*/3.50x 10-'
6.51 x 1 0-6 / 6.03 x 10-7
1.52x 10-5 / 1.21 x 10-6
1.83x 1 0-5 / 6.03 x 10-7
1.92x 1 0-5 / 6.03 x 10-7
3.27x 10-4 / 5.99 x 10-7
1.71 x 1 0-3 / 4.79 x 10-6
4.55 x 10-6/6.68x 10'7
2.09 x 10'5 / 6.89 x 10'7
4.83 x 10'5/1.62x 10'6
2.56x 10"/2.79x 10'6
8.77 x 10'5/7.33x 10'7
1.75 x 10* / 6.03 x 10'7
1.88 x 10-*/9.99 x 10'7
1.18x 10'3/ 1.36x 10'6
5.67x lO-4/!^! x 10'6
8.21 x 10'3/7.99x 10'7
5.97x 10'5/ 1.24x 10'6
4.22 x 10'5 76.89 x 10'7
1.80 x 10-* /8.76 x 10'7
6.00 x 10-*/2.34x 10-6
1.49x 10"1/ 1.37x 10"6
2.89 x 10-4 72.89 x 10'6
7.62 x 10-* 7 3.85 x 10"6
2.63 x 10'372.54x 10's
Btu = British thermal unit.
                                 H-5

-------
Table  H-2.   Emissions of  Selected PAHs from Fossil-Fuel-Fired
Utility Boilers
PAHa
Benz(a)anthracene (c)
Pyrene (f)
Benzo(b)fluoranthene (c)
Benzo(b and k)fluoranthene
(0
Benzo(k)fluoranthene (c)
Benzo(a)pyrene (c)
Chrysene (c)
Dibenz(a,h)anthracene (c)
lndeno(1,2,3-c,d)pyrene (c)
2-Methylnaphthalene (f)
Fluorene (f)
Fluoranthene (f)
Phenanthrene (f)
Frequency
of
detection
(from up to
47
individual
tests)
9
14
1
5
1
8
9
4
6
12
13
15
17
Median
(lb/1012Btu)
3.53x 10'3
1.26x 102
808 x 10'3
6.65 x 10'3
3.56x 10'3
1.02x 10'3
3.56 x 10'3
- 2.70 x 10'3
6.40 x 10'3
2.34 x 10'2
1.35 x 10'2
9.89 x 10'3
2.66x 102
Maximum/minimum
(lb/1012Btu)
1.41 x 10° / 1 x 10'3
1.60x 10'1 / 1.21 x 10'3
only one value
4.79 x 10-2/ 1.61 x 10'3
only one value
3.67 x 10'2/2.27x 10"4
5.99 x 10'2/3.30x 10"4
1.32x 10'2/3.30x TO"1
3.59 x TO'2/ 1.65 x 10'3
2.15 x TO'2/ 1.72 x 10'3
1.72 x 10'1 / 1.29 x 10'3
2.81 x 10'3/9.58x 10'2
3.08 x 10'1 79.37 x 10'3
Btu = British thermal unit.
' (c) indicates  suspected carcinogen;  (f) indicates measured frequency greater
than 10.
                                  H-6

-------
     With one exception, all of the median values are  on  the
order of 10"3 lb/1012 Btu.   The  exception,  pyrene,  is
approximately 1.3 x  10~2 lb/1012 Btu.   Ranges  run from 1CT4 lb/1012
Btu to 10'1 lb/1012 Btu.

     Of the PAHs not listed in Table H-2, all, with the exception
of biphenyl,  have median values in the 10"2 lb/1012 Btu or 10"3
lb/1012 Btu range.   Biphenyl has a median  value  of  about  1.8 x 10"
1  lb/1012 Btu.   Ranges generally run from about 10'4  lb/1012 Btu to
10'1 lb/1012 Btu.

     H.2.3.3  Trace Metals.  Four of the metals  (mercury,
chromium, nickel, and arsenic) had speciation data  (see Table
H-3).   All of the metals were measured in nonspeciated form at
nearly all sites.  Most of the measurements were made  in  the  test
site stack or ductwork ahead of the stack.  However, a few
measurements were made at the outlet of a pilot or  full-scale
device operating in parallel with ductwork or other control
devices.  All of the measurements represent values  that,  in the
absence of chemical  reaction, should be found being emitted to
the atmosphere after passing through the  control device  indicated
in Table H-2-3 (Appendix H-2).

     The median value for lead was 3.78 lb/1012  Btu,  with a range
of 1.1 x 10'1 lb/1012 Btu to 176 lb/1012 Btu.   The median value  for
cadmium was 1.24 lb/1012 Btu with a range  of  2.33  x 10"2 lb/1012
Btu to 28.5 lb/1012  Btu.  The four speciated  metals  are discussed
separately.

     H.2.3.3.1  Mercury.  Although total mercury was measured at
most of the sites,  elemental mercury and  ionic mercury were
measured at only 10  sites.  Several of the sites attempted
measuring methyl mercury with the Bloom method.1  However,  the
method does not provide consistently accurate methyl mercury
measurements.  A resolution to this problem  is  to add  reported
methyl mercury values to reported ionic mercury values (for the
same test) and report the sum as ionic mercury.  This  procedure
is followed here.

     Total mercury  had a median value of  3.44 lb/1012 Btu, with a
range of 6.60 x 10'2 lb/1012 Btu to  22.9  lb/1012 Btu.    The sum of
the median values for elemental mercury  (1.44 lb/1012 Btu)  and
ionic mercury  (1.93  lb/1012 Btu) was about the same value as for
total mercury.  This agreement is surprisingly  good,  considering
                                H-7

-------
 [Q
 M
 0)
rH
•rl
 O
(Q
•H
rH
-H
4J
D

•0
 (U
 M
•H
fe
 I
rH
 0)
-rH
 (0
 I
r-l

 (C
 0)
 o
•0
 (U
4J
 u
 (U
 o
 in
 rt
 o
 [0
•H

 I
  i
 K
 •a

*9
1
CO
in


35 -8



,
i





S.
2
o
^



||
28
c "S
1«

|?


b
i
O 0)

c *B
I-
i
n
O
i
*•
B a
£«»



3
f
O
i

g -2
jl
isi
It





i







00
T—


^
z
CO
CO


z





in
CO
«-




CO





'o
•o
c
CO
To
c
u
_c
T
a
a

CO
CO
T-


^
z
CO
CO


z





CO
d





o
CO






"5
_c
•c
a
c
CO
CO
CO


^
z
CN
CO
CO


z





CO
CO




o






'c
1
CM
q
iri


^
Z
*
CO


z





CO
in
*t




CO
CO





'e
1
•5
c.
]
t
0

CM
CO

O
^
d
in
t'


CO

^
^

CM
CO





s
CO




1

'5
1
"a
c
|
t
U Elemental m

CO
CO

CO
o
T*
d
O)
CO


^
in
"m
m
T^

CO
CO





CM
CM




1


'c
CO
- To
c
u
c
>
U Ionic mercur







CM
in

CM

in
CO
O)
CO


z





CO
in




CO
CO






"5
c
I
0

CO
in

O
*^
O
in


O

•*
•-

o>
in





O
CO




A
O5



"5
c
u
_c
£•
c
CO
II Elemental m

V—
*

00
^
0
in
CO


O
CO
CO
CM


^





CO
CM




o>





"5
c
u
>>
|| Ionic mercur







in
CM
CO
d


^
Z
1
d


z





CO
in
CD
d




in






'c
_c
£•
i







-














^_





o



'c
c
t
1
CD
H Elemental m

^
z


z

z


^
z
CS
o
d


z





CM
q
d




-





'c
c
£•
c
.2
c







CO


^
z
CN
»-


Z





05
CM




O





'c
CO
|| Cadmium in coal

m
in
in


^
Z
0)
r-


Z





CM
CO




O
CO






|| Cadmium in coal

:


^
z
o
en
d


z





in
•-




O






|| Cadmium in oil
                                                         H-8

-------
•0

a =5
CD C
TI ™ _

>- o 2
o •- *•
II







1
i
K
CM

Z



in
CM
CM

Z




00
en
en








IV









'6
1
eo
"5
8
.£
o
<

CO
CM

Z



CO
in

Z




00








CO
to









To
8
u
'c
<

01
CO


CO
CM
O
d
00
in
CM
C>

o>
CO


£
6
X
CO
CO
*
en
CO




CD
00
>*•
d








a
*>
CM






"3
8
f
Trivalent arsenic ii


CO


CO
00
CO
o
d
CM
in
00
d

5



b
X
CM
^
CD




CM
d








CM




^_
CD
8
£
o
Pentavalent arseni






o
op

Z



CO
CD
CO

z




in
CM
CO








^









"o
_c
<
IV
en
CO

Z



in
in

Z




in
(O








CO









'6
I
CD
"5
8
c
i
Z
d
CO

Z



tn

Z




oi








CO









1
c
"o
Z
co
CO

Z



CM
rv
CO

Z




00
in








CM









'5
c
I
z

00
in


in
CO
CO
CO

00
tn


X
0
X
CO
CM
CD
00
in




00
CM
co








CM








—
Soluble nickel in o


CO


o>
00
00
O)

CO



b
X
00
r-
CO
co




00
|v
CO








CN








;^
Sulfidic nickel in o


-


-
-

-



-






-








0








ZSi
Metallic nickel in c


0)
CO


t—
CM
OS
in

O)
CO


E
O
X
CM
CM
*
CO




CM
CM








CM








_
'5
c
ID
a
2
X
o
"5
jf
o
Z


O)
00


CD
d
en
tv
CM

00
00



CM
CD

en
00




CO








in








'5
TJ
C
ID
"5
o
o
_c
1
o
u

^


CM
CO
O
d
01

CM


o
X
a>
00
00
,_




CO
CO







e
i


'6
CD
To
o
u
_c
i
D
Hexavalent chrom


o>
00


CM
0>
tv
d
in
CM

O
en



o
q
|V

O)
00




CO
in








CO









hromium in coal
O

^


in
O>
O
d
00
(V
CM

O



CM
IV
rv
d

T—




0
O)








tt



"5
o
o
c
£
3
Hexavalent chrom


o
00


rv
en
CM
d
en
in

IV
rv



00
CO

o
CO




en
in








*—









hromium in oil
O

o
CM


rv
O
d
?

CO
CN



TJ-
q

0
CM




00








r-




~
o
c
£
3
Hexavalent chrom

                                       H-9

-------
•0
o
g
g
«n
K
(D
H

i
o


aximum/minimum (II
BTU)
S

•«
!
€
ID
en



m
1
o
on




i
1
~^



!s
11
• 
e •§
8 Q
1-
?

b
i
e »
^ "S
S w
e
CD
b

£







3

to

X
o
X
CO
rx




0
to
in
CO
IX




§
CO
CD
00
CM


to

X
CD
CO
O
X
05
CM
CM





























"o
CO
"a
o
u







o
u
c







'5
c
1






'c
To
C
O
_c
1
a

to

X
CO
to
X


00
CO


CO
6


f^
CD

CO
CM
05


CM
CO
O

CO
6

'5
1
a
"5
o
u
c
fc
1
I
_s
u



to
CM
tx
04
05
id


CM
CO


*
in


CO
CO

in
in



00
CO
CM
^
CM
CD



'c
^
g
U
8
_c
1
'e



o
10
CO
to
a
CM*
cs






























I
1



o
CO
to
in
CO
10


CD


O>
to
••••
mS^S
CO
CO

CO
CM
Cn
•H

en
in
O

CO
CD




"5
8
c
1
i
I Elemental i



in
O
to
en
id


CO
CO


o
Tj;
in
^•Ml
^^^S
^.
CO

CO

Tt
•H

,_

in
CO
CM
6







~S
8
|
a
c
"c
_e



CD
CO
o
to
CM






























"c
i



-


«-


-
^••1
SBB

*"


^^

•••i





"^






^
.c
g
Elemental i



z


z


z
IHHB
SS3
^
Z


^
Z
••Mi

^
Z

^
z








:=
O
C
1
_e

to

X
CO
CO
oi
b
X
in
00
CM




CO
CM
8
CJ
in
CO
CM




o
CO
to
05
CO


*•
o
X
o
X
q
i—





























'6
1
0
•5
II Cadmium in co






•S
Cadmium in co







U Cadmium in oil






'E
•o
c
CO
I) Arsenic in coal
                                     H-10

-------
 g
•H
 O
 u
to

K

 0
H

•s
IH
o

^
]
1
E
E
*x
ID
S
••^••i
555E5
£

i
in
en

••••
Sf^S

£
1
Q.
£
S
en


£
1
Q.
O











S
H
DO



••••••••
5555551
II
.«

B
N
I-
I
•••••••
5555555
0 «
P
« in
a.

3
f
o
£
ercent of
Spade*
a.
lb/10" BTU







1
j


































|| Arsenic in coal

b
*"*
X
CO
q
CO
b
X
05
CO
CO

O
CO


en
d
•^••B
iiSSS
O
CO




CO
CO
d
••••
^^^—
en
CO

co
CO
cS
d





"5
8
|| Trivalent arsenic in <


^_
b
X
IV
CO
CO
«—



O


£
oi
••••

o




CO
"-
^MH
sssss
CO

q
d




•5
o
u
c
|| Pentavalent arsenic



q
IV
CM





























"6
.0
1

M
b
X
CO
"b
X
in
«•;
CM



























°5
•o
c
CO
•5
8
c
i
.s



CO
o
o
d
5
CO





























"a
o
o
_c
~S
o
Z



O
CO
d
o
CM





























'o
_c
13
o
Z

b
^
X
CO
CM
O
X
^*
CM
^

CO
in


O
in
q
CM


CO
in




5
*

CO
in

CO
(0







'5
c
i
u
"c
£
1



CO
CM
O
X
CO




CO


o
CM
^


CO




CM
O
*~

CO

CO
IV
CO







|| Sulf idle nickel in oil
















-




_


_

-







Metallic nickel in oil



O
o>
o
X
in
CO
CO



en
CO


O
CO
CO
r-


O)
CO




0
^

en
CO

CM







'5
_c
1
'x
o
"5
u
Z

b
T—
X
o
X
CO
CO
«—

05


-.
in


O)




CO
CM


en
CO

CO







U Chromium in coal and oi


r«l
b
X
CO
CM
O
CO



en


CO
*


en




IV
q
>*

-

CO
(0
d
'5
1
CO
"5
o
o
c
E
|| Hexavalent chromiu



d
CO
CO




o
O)


CO
in


o
en




cri


en
CO

CO
in







|| Chromium in coal



CM
O
d
o
CO




o


CO
CD


o




CO
in

-

o
en
d



"a
8
_c
E
H Hexavalent chromiu



o
IV
en
05




o
CO


q
in
"~


0
CO




0
CO


0
CO

en
in
d







|| Chromium in oil



CO
CO
d
CO
CO




o
CM


in
CO
CO


o
CM




CO
CO
CO

0
CM

CO
d




=5
.c
i
Hexavalent chromiu
                                                            H-ll

-------
                 * S
                 in g1
                 » 0
                 * S
                 £ c
                 * 2
                                          8
                                          e
                                       1 o
                                       o~
                                       =o
                 c 1
                 pa
                 •
                                       2 S
                                       tS ~
                                       M_ CO
                   s

                                       ig
                                              S
                   D    ™ os T* >0
                                   O3O)iS
                                   >>«Q.
                                   « « H.  E
in  ^ 01
»  0) 9>
O)  Q) O)
CO  O> <0
£2 S
0)  C V
O  Q) (^

Q.  0> Q.
 . 0-  .
0)  . V
(D  2  05
>»>
                                                       g
a>
>
*
c
                                        -i   „-    rf
                                   •- •- ^ W   0) «  »
                                   OJWBJn   .S S .tt

I
•ri

^
0
o
m
K
O
H

I
uni
her
ish
BTU
en
ed
nonsp
NA = not applica
Open cells indica
                   ^ 1! i §!!'!'! c ^   $-s£   I
                   <5J3  _EEo5O)fe(5«a'   'E.9'E_ou

                   c >» _2 .S .5 ^  *^ .2 .5 *-•  ^ ^ .2 £ .2 -n .2
                                         .
                 g rao-L^STo®  gTBTo « cS'S.Z'B °T5
                     »  ..
                                       *n  w ffl   >%    v*
                                       0}w— +;0*;.. *;
                                                       ID
                                                       a
                   fe gg «* 10 g'
                              ^-
                                                  05
                                                     TJO
                                                       t-
                 a. Si-zccccujooccco « o cc en ir u. cc
                                            H-12

-------
that different test methods were used for measuring total mercury
and for measuring species.  Ranges for elemental and ionic
mercury  (mercury (+2)) were 4.60 x 1CT1 lb/1012  Btu  to  14.7  lb/1012
Btu and 2.72 x 1CT2 lb/1012 Btu  to  5.49 lb/1012 Btu,  respectively.
The split between elemental and ionic mercury was not consistent
from one site to another.  About half the sites had more ionic
mercury, while the rest have more elemental mercury.  The average
proportion of elemental mercury from 10 test sites was 62
percent, while the ionic mercury was 38 percent.  When median
values are used, 43 percent was elemental and 57 percent was
ionic.  For coal only, the average values from 9 samples were 59
percent for elemental mercury and 41 percent for ionic mercury
(median percentages were 40 and 60 for elemental and ionic,
respectively).  For oil, only one sample was found to contain
ionic mercury; elemental mercury was not detected.  Statistics  on
a percentage basis for mercury are given in Table H-4.

      H.2.3.3.2   Chromium.  Total  chromium was measured at most
sites.  The median value was 6.42 lb/1012 Btu, with a range  of
4.11 x 10'1 lb/1012  Btu to 13.8  lb/1012 Btu.   The median value for
the 11 tests showing chromium  (VI) was 8.89 x 10'1 lb/1012  Btu
with a range of 2.37xlO'2  lb/1012 Btu to 6.04 lb/1012 Btu.  The
amount of chromium (VI) for these 11 test sites averaged 11
percent  (median 12 percent) of all speciated chromium, with a
range of 0.4 percent to 34 percent.  Separate tests for chromium
(III) were not made.   When divided between coal- and oil-fired
utilities, chromium (VI) averaged 11 percent  (median 10 percent)
for the four coal-fired sites,  with a range from 0.4 percent to
23 percent.  Seven oil-fired boilers averaged 20 percent chromium
(VI)  (median 23 percent), with a range from 5 percent to 34
percent.  No other sites were available for testing.

      It should be  noted that chromium (VI)  is easily reduced to
chromium  (III) when in reducing conditions.  Stack conditions
lead to rapid reduction of existing chromium  (VI), as experienced
during field testing.2  This factor may  influence (reduce)  the
levels of chromium (VI) measured in the stack and downwind  from
the stack while the chromium (VI)  remains in reducing conditions.
Even in ambient conditions, the half-life of chromium  (VI)  is
about 15 hours.  Table H-5 gives statistics on a percentage basis
for speciated chromium.
                               H-13

-------
 CO
 M
 0)
i-H
•H
 O
CQ
J-)
•H
r-l
•H
•O
 0}
 M
•H
fc
 I
rH
 0)
•H
 W
 O
 W
 C
 o
•H
 en
 w
•H
 U
 M
 (U
 S
  i
 ffi

 a;
 tO
E E
3 S
E E
ii
-ll»
E s s °
°'!l*
ill
0 g 0
if!-!
*" • » W
Ih
o o o
11"
in
1!-.!
SB. 0-
£ VI
|| j
- ^ "«
I|l
p. a
° :i

I? 1 2
ilii





s
0
w


s?
<£>
O)

00
(O
(O
CM
(O

CM
(0


CO
(O

00

O





C
^
D
U
ffi
_ O
a ^
c c
a> a
Is
UJ U
CN
CO

CM
CO
CO
CO
00
CO

00
CO


in
00
CO
O5


o


•o
c
a
"5
o
u
_c
1
s
E
Jo
S
o>

f,.
(O
(O
(D
0)
in

o>
in


O
O)
in

00

a>





c
^.
g
i
"5
S
Is
UJ 0
CN
0>

CO
CO
CO
5


^


o
(0
5
O)


0>






"a
b
£•
i
E
.S





a



*


(D
(B



O




o
c
2-
c
1
s
c
o
H



^
Z
Z
Z


Z


o
o
o
o

Z

«-






•5
.£
£•
a
E
u
'c
_
a.

E
(0
(0

0)

o
•a
 0)
^-*
 u
                                                         0)
                                                        •o
 >•
 3  «

 "1-
 0)  Q-
 E  •
 E  ro
 V  V
iD  E
                                H-14

-------
     H.2.3.3.3  Arsenic.  Total arsenic was measured  at  nearly
all sites, but only one site provided data on  the presence of
arsenic  (III) and arsenic  (V).  Two tests were made at that site.
Total arsenic had a median value of 2.25 lb/1012  Btu  and  a range
of 4 x 1CT2 lb/1012  Btu to  104 lb/1012 Btu.   The  median value for
arsenic  (III) was 4.86 x 10"1  lb/1012 Btu and for arsenic   (V) was
7.72 x 10"1 lb/1012  Btu,  or about half  again as  much arsenic  (V) .
The discrepancy between the median value for total arsenic and
the sum of the median values for arsenic  (III) and arsenic (V)
may be due to using different test methods for total  and for
speciated arsenic.  High and low values for arsenic  (III)  and (V)
were 3 . 03 x 10'1 lb/1012 Btu and  6.69 x 10'1 lb/1012 Btu, and 1.70 x
10'1 lb/1012 Btu  and 1.38 lb/1012  Btu,  respectively.

     H.2.3.3.4  Nickel.  Total nickel was measured at nearly all
sites, but only two sites provided data on speciated  nickel.  The
species measured were soluble nickel  (water-soluble  salts such as
nickel sulfate and nickel chloride), sulfidic  nickel  (such as
Ni3S2, NiS, and Ni3S4) ,  metallic  nickel  (including alloys) , and
oxidic nickel (including NiO, complex oxides,  and silicates).
Total nickel had a median value of 5.45 lb/1012 Btu and a range
of 3.00 x 10'2 lb/1012  Btu  to 2.15  x 103 lb/1012  Btu.   Metallic
nickel was not detected at either site; the distribution of the
remaining three types of species, in lb/1012 Btu  and  in percent
of total nickel species, is shown in Table H-6.

     H.2.3.3.5  Individual Test Site Data  for  Mercury and
Chromium Data.  Individual test site data  for  mercury and
chromium data for each test contributing species information to
this report are given in Tables H-7 and H-8 for  mercury  and
Tables H-9 through H-ll for chromium.

     When the one value of divalent mercury found in  an  oil-fired
boiler is removed, the percentiles for coal-fired boilers change
slightly, as shown in Table H-8.

     Table H-10 shows the chromium data for four coal-fired
boilers.  Table H-ll shows Chromium data for seven oil-fired
boilers.
                               H-15

-------
 0)
rH
•H
 O
OQ
 i
rH
-H
O

T3
 C
 (0
 (0
 o
o

 to
 o
iw

 03
 u
•H
4->
 CQ
•H
J->
 (0
4->
W
-H

 §
 M

g
 in
  i
 (D
 iH
 •s
 E-"
EO E§
O O
g 5 -S 5
•ef 1 8
§ |||
•S *- "S
•el S '5
IS if
a o
"in
| § 11
fe a
a,a ""
CD
= °-S
S c .£
I § 8
£ ~"
C ° "w
.5 e .£
o o a>
S w g-
Is I
Ii°l
"5 —
S> e .£
C 01 O
CO O 0)
DC S; a
1 "
>• e S •=
S la § -
CB M. •g s ,x S <»
1 -§| 1





CA
.£
u
(A
100/66
0)
5)


CO



cn
CO


00
CO

cn
oo

CO



in


To
o
0
.£
E
3
1?
U £
CO
0)
O)


r-



^


CM
r-




t
co



£

"5
o
u
c
§ E

"j" Q (J

100/77
o
05
0
cn


0)
00



cn
CO


O
05

cn
co

CO
CN



CO


CD
o
u
c
E
hromiu
O
O
00
CO
o
T—
O


^



-


o
V—




CO



•*

"5
o
u
c
C c
"5 .2
ll

co
CD
ifl
O)
O
CO
O
CO


O
00



O
CO


rs.
r*

0
CO

cn
CM



£


_.
o
.£
E
hromiu
O
in
CO
o
CM
o
CM


o
CM



o
CM


CO
CM

o
CM

o
CN



r-


'5
c
SE
co .2
> c
co 5
x g
41

                                                   H-16

-------
Table  H-6.   Emission Statistics  for  Nickel Species From Two
Oil-Fired Utility Boilers  (lb/1012 BTU/Percent  of         Species)
Statistic
Average
Maximum
Minimum
Standard deviation
Range
5th percentile
10th percentile
90th percentile
95th percentile
Soluble nickel
628/58
1,240/58
21.3/65
616/NA
1,219/7
31.5/58
63.1/58
1,740/58
2,050/58
Sulfidic nickel
37.8/3
73/3
2.60/8
49.8/NA
70/5
1.89/3
3.78/3
102/3
120/3
Nickel oxides
422/39
835/39
9.00/27
584/NA
826/12
21.1/39
42.2/39
1,170/39
1,380/39
Note:  Data from two sites.  Metallic nickel not detected. Examples of soluble nickel include water-
soluble salts such as nickel sulfate and nickel sulfide. Examples of sulfidic nickel include Ni3S2,
NiS, and Ni3S4.  Examples of nickel oxides include NiO, complex oxides, and silicates. Most of
these compounds have a valence state of 2.
                                      H-17

-------
Table H-J.  Mercury Data for Nine Coal-  and 1 Oil-Fired  Boilers
Test site
1
2
3
4
5
6
7
8
9
10
Fuel
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Oil
Elemental mercury (%)
31
24
85
71
18
9
54
88
96
Not detected
Divalent mercury (%)
69
76
15
29
82
91
46
12
4
up to 1 00
Summary Statistics
Average (%)a
Standard deviation (%)a
Range (%)a
Maximum (%)a
Minimum (%)a
5th percentile (lb/1012 Btula
10th percentile (lb/1012 Btula
90th percentile (lb/1012 Btu)a
95th percentile (lb/1012 Btu)a
62
33
87
96
9
0.170
0.340
9.23
10.9
38
33
87
91
4
0.106
0.212
4.55
5.24
 Results from Site 10 are excluded.
      Table H-8.  Mercury Percentiles for Coal-Fired Boilers
Percentile
5th percentile (lb/1012 Btu)
10th percentile (lb/1012 Btu)
90th oercentile (lb/1012 Btu)
95th percentile (lb/1012 Btu)
Elemental mercury
0.170
0.340
9.23
10.9
Divalent mercury
0.118
0.235
4.73
5.40
                                H-18

-------
Table H-9.
Boilers
Chromium Data for Four Coal- and Seven Oil-fired
Test site
1
2
3
4
5
6
7
8
9
10
11
Fuel
Coal
Coal
Coal
Coal
Oil
Oil
Oil
Oil
Oil
Oil
Oil
Elemental chromium
(%)
99.6
90
91
77
95
91
77
83
66
83
Not detected
Hexavalent chromium
(%)
0.4
10
9
23
5
9
23
17
34
17
up to 1 00
Summary Statistics
Average (%)a
Standard deviation (%)a
Range (%)a
Maximum (%)a
Minimum (%)a
5th percentile (lb/1012 Btu)a
10th percentile (lb/1012Btu)a
90th percentile (lb/1012 Btu)a
95th percentile (lb/1012 Btu)a
85
10
34
99.6
65 .8
0.67
1.34
42.8
51.1
15
10
34
34.2
0.4
0.082
0.163
4.07
4.76
8 Results from Site 11 are excluded
                                 H-19

-------
Table H-10.  Chromium Data for Four Coal-Fired Boilers
Test site
1
2
3
4
Elemental chromium
(%)
99.6
90
91
77
Hexavalent chromium
(%)
0.4
10
9
23
Summary Statistics
Average (%)
Standard deviation (%)
Range (%)
Maximum (%)
Minimum (%)
5th percentile (lb/1012 Btu)
10th percentile (lb/1012 Btu)
90th percentile (lb/1012 Btu)
95th percentile (lb/1012 Btu)
89
9
22
99.6
77.5
0.792
1.58
49.1
58.4
11
9
22
22.5
0.4
0.095
0.190
5.48
6.48
                         H-20

-------
 Table H-ll.   Chromium Data for Seven Oil-Fired  Boilers
Test site
5
6
7
8
9
10
11
Elemental
chromium (%)
95
91
77
83
66
83
Not detected
Hexavalent
chromium (%)
5
9
23
17
34
17
up to 1 00
Summary Statistics
Average (%)"
Standard deviation (%)8
Range (%}a
Maximum (%)a
Minimum (%)"
5th percentile (!b/1012 Btu)a
10th percentile (lb/1012 Btu)a
90th percentile (lb/1012 Btu)a
95th percentile (lb/1012 Btu)a
82
10
29
95
66
0.297
0.594
13.0
15.0
18
10
29
34
5
0.074
0.148
3.33
3.85
8 Results from Site 11 are excluded.
                            H-21

-------
H.3  ENVIRONMENTAL CHEMISTRY AND FATE

     The following sections provide a summary of the
environmental chemistry and fate of HAPs associated with
emissions from utilities.  HAPs of interest include selected
PAHs, dioxins and furans, and trace metals.

H.3.1  Dioxins and Furans
     This subsection discusses the processes that affect the
fate, degradation, and transformation of chlorinated
dibenzodioxins (CDDs) and chlorinated dibenzofuran (CDFs) in the
environment.  In general, the discussion presented here focuses
on CDDs and CDFs as a group and not on individual dioxin and
furan congeners.   Information is also presented on the homologue
groups of dioxins and furans (i.e., hexa-,  hepta-, octa-
chlorodibenzodioxins, etc.) with discussion of congener-specific
fate and transformation processes limited to 2,3,7,8-TCDD.

     H.3.1.1  Transportation and Fate.  CDDs and CDFS are widely
dispersed throughout the environment by atmospheric transport and
deposition.  During transport CDDs and CDFs partition between the
vapor and the particle-bound phase.3   Vapor pressure  greatly
influences the extent to which CDD/CDF congeners are found in the
vapor phase.  Table H-12 lists the vapor pressures and other
chemical properties of homologue groups of CDDs and CDFs.  The
ratio of vapor- to particulate-bound CDDs and CDFs ranges from
6.69 for tetra-CDD (relatively high vapor pressure) to 0.02 for
octa-CDDs  (relatively low vapor pressure)4

     CDDs and CDFs can be physically removed from the atmosphere
through wet deposition, particle dry deposition, and gas phase
dry deposition.5   Wet deposition is very effective at removing
CDDs and CDFs from the atmosphere and can dominate removal
processes in areas of low suspended particulate matter.  However,
in areas with high total suspended particulate matter, dry
deposition  (gravity settling of particles)  can dominate.6

      As seen Table H-12, CDDs  and CDFs have low water
solubilities and vapor pressures and high octanol water partition
coefficients  (Kow)  and  organic carbon partition coefficients
                               H-22

-------
 (Koc).a,9>1°  These  chemical  properties indicate that  CDDs/CDFs are
strongly sorbed  in  soils.11  In subsurface  soils-particularly in
soils with high  organic carbon content—CDDs/CDFs show little or
no vertical migration,  as  they are  resistant to both  leaching and
volatilization.  However,  the mobility of  CDDs/CDFs in soil has
been shown to increase with a carrier solution such as waste oil
or diesel  fuel,  although he research in this area  is  ambiguous.12
Because CDDs/CDFs are resistant to  leaching  and volatilization,
they are mainly  transported to surface waters through soil
erosion.13

      In the aquatic environment, CDDs and  CDFs remain sorbed to
particulate matter,  and dissolved CDDs/CDFs  entering  surface
waters will partition to suspended  solids  or dissolved organic
matter.14   Volatilization  is not expected to  be a significant
loss mechanism for  the higher chlorinated  CDDs/CDFs under normal
conditions.15  The primary removal mechanism  for removing
CDDs/CDFs  from the  water column is  through sedimentation and,
ultimately,  burial  in sediments.  Research has shown  that
concentration of solids settling through the water column was an
order of magnitude  higher  than the  concentration in the suspended
particulate matter.   This  difference in the  concentrations
indicates  that the  solids  tend to scavenge CDDs/CDFs  as they
settle.  Once buried, little recirculation of CDDs/CDFs occurs
within the water column.16

      Aquatic organisms are shown to bioaccumulate  CDDs/CDFs when
exposed to contaminated sediments and to bioconcentrate CDDs/CDFs
dissolved  in water.b'17-18'19'20'21   Although the pathways  for
a  The Kov is defined as the ratio of a chemical's concentration in the octanol phase
   to its concentration in the aqueous phase of a two-phase octanol/water system.  A
   chemical's K^, gives a strong indication of its behavior in the environment.
   Chemicals with low K^ values are considered hydrophilic and tend to have high water
   solubilities and small soil/sediment adsorption coefficients and bioconcentration
   factors.   Conversely,  chemicals with large K^ values are considered hydrophobic and
   have the  opposite tendencies.9  The K00 defines the extent to which  an organic
   chemical  partitions itself between the solid and solution phases of a water-
   saturated soil,  runoff water, or sediment. It is the ratio of the  amount of
   chemical  adsorbed per unit weight of organic carbon in the soil or  sediment to the
   concentration of the chemical in solution at equilibrium.  Koc. is a useful parameter
   because the degree of soil adsorption can affect a chemical's mobility in the
   environment and other fate processes.10

   Bioconcentration refers to the accumulation of a chemical from direct exposure to
   water.  In aquatic organisms, bioconcentration involves uptake via  gills and skin;
   elimination via gills, skin, urine, and feces; and metabolic transformation.17
   Bioconcentration is important because a chemical that bioconcentrates in an
   organism may affect ecological processes at aquatic concentrations  that appear safe
   for the organism according to bioassay criteria for acute or chronic exposure.18  A
   bioconcentration factor (BCF) describes a chemical's tendency to bioconcentrate in
   an organism and is defined as the ratio of chemical concentration in an aquatic
                                                              (continued...)

                                   H-24

-------
accumulation  are not fully understood,  the accumulation may be
primarily  food-chain-based since most of the CDDs/CDFs are
associated with sediments and dissolved organic matter.  Most
likely, uptake  starts with benthic  organisms (e.g., mussels,
chironomids,  etc.)  through ingestion or filtering, and those
organisms  (e.g.,  crabs, suckers, etc.)  consuming benthic
organisms  would then pass the contaminant up the food chain.22

     H.3.1.2  Transformation and Degradation.   CDDs and CDFs are
extremely  stable in the environment under normal conditions.
CDDs/CDFs  do  not hydrolyze appreciably,23 and 2,3,7,8-TCDD  is
recalcitrant  to microbial degradation.24  2,3,7,8-TCDD is also
stable to  oxidation in the ambient  environment.25   The most
significant natural degradation of  CDDs/CDFs is by
photodegradation.26  Photodegradation is slow in water and  on dry
surfaces but  increases in the presence of organic  solvents.27
CDDs and CDFs photodegrade to lower chlorinated CDDs and CDFs.
In general, the rate of photolysis  increases as the degree  of
chlorination  decreases and, within  a congener group, as the
degree of  ortho substitution decreases.  Research has shown that
CDFs degrade  much more rapidly than CDDs.28  In the soil,
photodegradation is limited to only the soil surface and is not
significant below the first few millimeters of soil depth.29

     Atmospheric photodegradation of CDDs/CDFs is not well
characterized but appears to be the most significant mechanism
for degradation of CDDS/CDFs in the vapor phase.  However,
measuring  the rate of atmospheric photodegradation is difficult
because of the  low volatility of CDDs/CDFs.  For airborne
CDDs/CDFs  sorbed to particles, photolysis appears  to proceed very
slowly, and sorption of CDDs onto airborne particulate matter may
even act to stabilize CDDs from photodegradation.  Research has
shown  that the  half-life of gaseous 2,3,7,8-TCDD was several
hours under simulated sunlight.  When the 2,3,7,8-TCDD was  sorbed
onto fly ash  particles, Mill observed only an 8 percent loss of
2,3,7,8-TCDD  after 40 hours of illumination.30
   (...continued)
   organism to that in water.19  Bioaccumulation also describes increases in the
   concentration of chemicals in organisms.  However,  bioaccumulation may be
   distinguished from bioconcentration in that it encompasses accumulation of a
   chemical from all routes of exposure,  including ingestion, and is not limited to
   direct exposure to water or soil.20 Biomagnification is a related term and is the
   systematic increase of a chemical in tissue concentrations at higher levels in the
   food chain.21

                                H-25

-------
H.3.2  Polycyclic Aromatic  Hydrocarbons
     This section discusses the processes  that  affect the fate,
degradation, and transformation of  selected PAHs  in the
environment.  PAHs  selected were  those that are probable
carcinogens and those  that  are present most frequently in utility
emissions.  Probable carcinogens  include benzo(a)anthracene,
benzo(b)fluoranthene,  benzo(k)fluoranthene,  benzo(b and
k)fluoranthene,0 chrysene, benzo(a)pyrene,
dibenzo(a,h)anthracene,  and indeno(l,2,3-c,d)pyrene;  the most
frequently emitted  PAHs  are 2-methylnaphthalene,  fluoranthene,
fluorene, pyrene, and  phenanthrene.

     H.3.2.1   Transportation and Fate.  Transport and fate of
PAHs in the environment  are determined to  a large extent by the
physical and chemical  properties  of the individual PAH.
Table H-13 summarizes  the properties of selected  PAHs as
described in section H.3.2.  Some of the properties correlate
roughly to the molecular weights  of the PAHs, which can be
grouped into the following  categories:

     •    Low-molecular-weight compounds  (142-178 g/mol)--
          2-methylnaphthalene, fluorene, and phenanthrene;

     •    Medium-molecular-weight compounds (202  g/mol)--
          fluoranthene and  pyrene;  and

     •    High-molecular-weight compounds  (228-278 g/mol)--
          benzo(a)anthracene, benzo(b)fluoranthene,
          benzo(k)fluoranthene, chrysene,  benzo(a)pyrene,
          dibenz(a,h)anthracene,  and indeno(1, 2,3-c,d)pyrene.

Much of the discussion of the transport and fate  of PAHs is based
on these categories.31

     PAHs in the atmosphere are present in the  gaseous phase or
sorbed onto particles.  The ratio of particulate  to gaseous PAHs
varies for different PAHs.   For example, the results of a study
of air samples collected in Antwerp, Belgium, indicated that the
ratio of particulate to  gas phase PAHs ranged from 0.03 for
anthracene to 11.5  for benzo(a)fluoranthene and benzo(b)
fluoranthene combined.  Additionally, atmospheric residence time
and transport distance depend on  the size  of particles to which
PAHs are sorbed.  Larger particles  released from  urban sources
deposit close to their source of  origin and can become part of
urban runoff.  Conversely,  smaller  submicron-diameter soot
   Information on the environmental chemistry and fate of benzo (b and k) fluoranthene
   was not readily available from the literature for this study.

                               H-26

-------
 U
o

 in
 CN
 w
ffi
•O
 0)

 U
 0)
rH
 a)
w

4-1
 O

 W
 0)
•H
JJ

 0)

 o

04

rH
 10
 U
•H

 0)

u
 u
-H
 CO
  I
 K

 0)
 rH
 43
 (0
I
C ?
8 5
II

z

**
H)
0


^

o
3
en
E
a
3
O
CO
1
a
£
S z
S.

a
a
\
cp
II
o
S


1

•
to
b
X
in


b
CO
in


r_
o>
in

q
d





6
X
CM


CO
CO
CM
CM




Benz(a)anthracene (c)

•o
O
X
CM
CM
*~


S
m


0
00
in

in
o
O
d



u
b
2
*~
o
f

u
CN
CO
CM
in
CN



u
U Benzo(b)fluoranthene I

S
r*
b
X
CO
^~


£
in


CO
o
CO

CO
o
8
d




u
0
b
X
cn
CO


CM
CO
CM
in
CM



u
|| Benzo(k)fluoranthene 1

3
0
X
CO


b
CO
in


in
CO
"
b
X
CM
CM
1
in




b
X
CO
•*


&
CO
CN
CN




II Chrysene (c)

.
b
X
*


£
co



O
CO

00
CO
o
o
d





n
b
CO
in


u
CN
CO
CM
in
CM




II Benzolalpyrene (c)

,
•
b
^<
CO
^


CM
in
CO


in
rv
CD

CD
O
8
d




0
b
X
00
CM


CO
CO
CO
CM



O
|| Dibenz(a,h)anthracene

•o
0
V—
X
in
0)
CO


b
CM
CO


CO
in
CO

CM
CO
o
d



b
o
o
CM
to
o
0
1

00
CO
CO
CM


"S
c
CD
1
0
CO
CM"
I
CD
I

•
b
X
in
O
in

CO
cn
CO
*
CO

CD
00
CO

in
CM





CO
q
d



u
CM
CM



;T-
2-Methylnaphthalene {

S
0
X
"-


•0
CO
CO
CO


CO

*

b
cn




u
b
in
rv
CO


CM
CO
CD




II Fluorene (f)

.
0
b
X
cn


CO
in
^


CO
^
in

CN
CO
d





0
X
in
*


CO
CM
CM
O
CM




  O

  .  0)
si
•B  E
fc—  (/)
2  w
 O)*;


||


 CD  g^

 9)  0)


 o  cn



 c a.
 «  y
^ _Q)


•g  »

 =  «
 cn •£
 co  ^

 P  °
 E >•-

 cn  cn
 5S
•D  C
 C  O
—  o

fc  5
 ..  CD
 c —
 V  cn

 O  £•
 C  C
•5  «

 co •*•
 O T3

•g  i«^
                                                                           Q. D  O3
                                                                           cn  cn  ^
                                                                           =  «  —   .   .
                                                                           <°  £  < CN  CO
                                                                           cn  -  -  ~  ~
5  t
 CD  "
3

3
•o
CD
eference

eference

lcula
                 CO
5P-  OC QCU
                                                                H-27

-------
particles associated with urban air have residence times of weeks
and are subject to long-range transport.  PAHs can be removed
from the atmosphere by wet and dry deposition, and the relative
importance of each removal process varies according to the PAHs
involved.  For example, dry deposition of benzo(a)pyrene accounts
for three to five times more removal than does wet deposition.34

     In water, PAHs can be removed by volatilization to the
atmosphere and by sorption onto suspended and bottom sediment.
Low-molecular-weight PAHs, which have Henry's law constants
ranging from 10"3 to 10"5 atm-mVmol,  are associated with
significant volatilization from water.  Only limited
volatilization occurs for medium- to high-molecular-weight PAHs,
which have Henry's law constants of less than 10'5.  However, as
exhibited by their Koc values in Table H-13, high-molecular-
weight PAHs strongly sorb onto the organic carbon present in the
suspended and bottom sediments.  Low-and medium-weight PAHs have
only a moderate potential for sorption onto organic carbon.35

     PAHs in soil can be removed by volatilization.
Volatilization of low-molecular-weight PAHs may be substantial.
In one study, volatilization was found to account for 20 percent
of 1-methylnaphthalane and 30 percent of the loss of naphthalene;
both compounds are low molecular weight PAHs.  Volatilization of
higher molecular weight PAHs is not expected to be a significant
removal process from soil.36

     Because of the tendency of high-molecular-weight PAHs to
sorb strongly to soil, they are resistant to leaching.  However,
leaching can occur in the lower-weight PAHs.   In a study
conducted on land cleared by burning, PAHs were found to move
into the soils by partitioning and leaching.   Phenanthrene and
fluoranthene at the site were incorporated into the soil to a
greater extent than the higher-molecular-weight PAHs, such as
indenod, 2, 3-c,d)pyrene.37

     Plants and animals can bioconcentrate PAHs.  Table H-14
provides BCFs of selected species for the PAHs chosen due to
suspected carcinogenicity and frequency of occurrence in emission
tests.  Generally, bioconcentration factors  (BCFs) are larger for
the higher-molecular-weight PAHs.  In water,  bioconcentration and
bioaccumulation of PAHs vary greatly depending on the type of
aquatic organism.  Fish and crustaceans readily bioaccumulate
PAHs from contaminated food, whereas mollusks and polychaete
worms have limited assimilation.  Biomagnification is the
systematic increase in tissue concentrations at higher levels in
                               H-28

-------
     Table H-14.   PAH Bioconcentration Factors  (BCF)  for
     Selected  Speciesa/b
PAHs"
Benz(a)anthracene (c)
Benzo(b)fluoranthene (c)
Benzo(k)fluoranthene (c)
Chrysene (c)
Benzo(a)pyrene (c)
Dibenz(a,h)anthracene (c)
2-Methylnaphthalene (f)
Fluorene (f)
Pyrene (f)
Fluoranthene (f)
Phenanthrene (f)
BCF
10,233
10,000
9.1
10,000
14.1
13,183
14.8
6,095
20,417
2,818
12,882
54,954
8,912,509, 3,235,936
13.8
10,000
2,399
407, 447
502
5.20
2,692
44,668
5.7
1,738
79,433
5.6
324
15,136, 19,055
Species
Daphnia magna
Fathead minnow
polychaet sp
daphnia magna
polychaet sp
daphnia magna
Polychaet sp
Daphnia magna
P. hoyi
Daphnia magna
Daphnia magna
P. hoyi
P. hoyi
Polychaet sp
daphnia magna
algae
None cited
Daphnia magna
Polychaet sp
Daphnia magna
P. hoyi
Polychaet sp
Daphnia magna
P. hoyi
Polychaet sp
Daphnia magna
P. hoyi
No BCF was available for indeno(1,2,3-c,d)pyrene.
All references cited are from Mackay, et al, 1992 (Reference 32).
(c) indicates suspected carcinogen; (f) indicates measured frequency of detection greater than 10.  (See Table H-2)
(Reference 8).
                                         H-29

-------
the food chain, and because many aquatic animals such as mollusks
rapidly eliminate PAHs, biomagnification of PAHs has not been
reported in aquatic animals.38

     Some terrestrial plants can uptake PAHs through the roots or
foliage.  Factors that influence the uptake rate of plants
include concentration of PAH in soil, the solubility and the
molecular weight of PAHs, and the plant species.  Evidence
suggests that plant uptake of PAHs from deposition on foliage
greatly exceeds uptake from roots.  Research has shown that about
30 percent to 70 percent of atmospheric PAHs deposited into a
forest were sorbed onto the leaves and needles of the trees and
then deposited as litterfall.  Furthermore, PAHs can accumulate
in terrestrial animals through the food change or by soil
ingestion.39

     H.3.2.2  Transformation and Degradation.  PAHs in the
atmosphere react with a variety of pollutants such as ozone, NOX,
S02,  and peroxyacetylnitrate.   Reactions of PAHs with ozone or
peroxyacetylnitrate yield diones40  (compounds having two ketone
groups).  Nitro and dinitro PAHs are formed from reactions with
NOX.   S02 reacts with PAHs to form sulfonic acids.41
Additionally, some PAHs have been shown to degrade by oxidation
reactions.42

     Photodegradation of many atmospheric PAHs forms nitrated
PAHs, quinones, phenols, and dihydrodiols.  Some degradation
products are mutagenic.  PAHs absorbed onto particulates with
high organic carbon are more resistant to photodegradation than
the pure compounds.  The dark, high-organic-content particulates
on which the PAHs sorb may absorb more light, thereby making less
light available for photodegradation.43

     Photodegradation is also an important degradation process of
PAHs in water.  The rate and extent of photodegradation vary
widely among PAHs.  For example, one study reported that
anthracene, phenanthrene, and benzo(a)anthracene were susceptible
to photodegradation in water, while chrysene, benzo(a)pyrene,
fluorene,  and pyrene were resistant and are also dependent on
water depth, temperature, and turbidity.  The most common
products of PAH photodegradation in water are peroxides,
quinones,  and diones.44

     PAHs can be chemically oxidized in water by chlorination.
Research has shown that pyrene and benzo(a)anthracene degrade
rapidly upon chlorination.  Indeno(1,2,3-c,d)pyrene has an
intermediate degradation rate.  Of the chemicals considered in
this study, benzo(k)fluoranthene and fluoranthene were the


                               H-30

-------
slowest to degrade.  The byproducts of PAH chlorination are not
fully known but include anthraguinone, a chlorohydration of
fluoranthene, and monochloro derivatives of fluorene and
phenanthrene.45

     PAHs in water can also be oxidized by ozonation.  However,
ozonation is generally slower and less efficient than
chlorination in degrading PAHs.  Some products of PAH ozonation
include benzo(a)anthracene to 7,12-guinone; benzo(a)pyrene to
3,6-, 1,6-, and 4,5-diones; and fluorene to fluorenone.46

     Biodegradation is another removal process of PAHs from
water.  Bacteria degrade PAHs into cis-dihydrodiol via a
dioxetane intermediate, and trans-dihydrodiols are formed through
arene oxide intermediates in fungi.  Arene oxides have been
linked to the carcinogenicity of PAHs.  Additionally, algae'were
found to transform benzo(a)pyrene to oxides, peroxides, and
dihydrodiols.47

     Microbial metabolism is the major degradation process of
PAHs in soil.  Photolysis, hydrolysis, and oxidation are not as
significant.  However, evidence suggests that photolysis and
volatilization may be important for PAHs containing fewer than
four aromatic rings.48

     The rate and extent of biodegradation in soil are affected
by environmental factors, the physical and chemical properties of
the PAHs, and the level of contamination at the site.
Environmental factors include temperature, pH, oxygen content,
soil type, moisture, and the presence of nutrients and other
substrates that can act as cometabolites.  In one study, the type
of PAH and its properties also influenced the biodegradation
rate.   Biodegradation half-lives of PAHs composed of two and
three aromatic rings ranged from 2 to 60 days, while PAHs with
four and five aromatic rings had half-lives greater than
300 days.49   In another study,  the mean half-lives correlated
positively with Kow.  The rate  of biodegradation may also be
altered by the degree of contamination at the site.  Half-lives
may be longer at highly contaminated sites, since other
contaminants may be lethal to microbes.50  Table H-15 gives
half-lives of selected PAHs in soils and other environmental
compartments.

     For some PAHs, the products of microbial degradation in soil
are well known, while for others only limited information is
available.  As in water, bacteria degrade PAHs in soil to
cis-dihydrodiols through dioxetane intermediates, and fungal
                               H-31

Based on estimated
photolysis in water