-------
Source Loe
Estimates for the end of 1985 indicate that there are approximately
223 municipal waste incinerators in operation in the United States. Of
these 223 units, 109 are conventional and 114 are modular in design.
Conventional and modular unit locations and capacities are given in
Tables 34 and 35, respectively.80
No site specific location information is available for commercial and
industrial waste incinerators. Commercial units are generally located in
urbanized, metropolitan areas with large concentrations of people.
Locations of industrial waste incinerators parallel those of the industries
that use them for waste disposal. The lumber and wood products industries,
the primary metals industry, and the printing industry are the greatest
users of -incinerators for waste disposal. Lumber and wood producers are
primarily in the Southeast and Northwest. Primary metals plants are
predominantly in the Midwest, the Mideast, and the Southwest. The printing,
industry has an essentially nationwide distribution.79
SEWAGE SLUDGE INCINERATION
Process DescrtptH nn
Sewage sludge incineration refers to the oxidation of sludge material
generated by wastewater sewage treatment plants. Polycyclic organic matter
emissions from sewage sludge incineration potentially originate from the
combustion of carbonaceous material in the sludge, from the combustion p'oM
precursors that may exist in the sludge, and from the combustion of
supplemental incinerator fuel (typically natural gas or fuel oil) .
The most prevalent types of incinerators for sludge oxidation are
multiple-hearth and fluidized-bed units. Multiple -hearth incinerators are
relatively simple pieces of equipment, consisting of a steel shell lined
with refractory. The interior of the incinerator is divided by horizontal
121
-------
O
co
g
M
0
ia
H
M
55
•^
W
g
Z
M
S
O
4
,.
>""J
*c
o
f*<
*^*r
lit
J>
0
o
o
w
HHI
H
g
3
<}•
CO
rf
§
(
1
j
i
r
i
«c
4.
5
*
4
i
(
E
4
1
S
J
3
•
jj
"K
n u
(t *
**
n
ij "
I5
S: IM
«
o
~i
u
Facility Loci
^
tl *J
« o
a a
S"
u
„
II
S IH
o
n
C
o
w4
JJ
«
Cl
o
Facility L
^SSSSSSSSS'
^fno.o,^eorteo«o>e
tH rt rt
rtrtieMeMr>cMeM«MN«McM
a,gm 8.2z, • sss
HiiilliiiiiJiii.
•"
•g
ti
s
^
4)
~
O
AJ
«
<-4
•4
V)
o
o.
w
«
«1
1-
II
u
ff)
s
9
•
sl •
• •> e
•! 1 ^
B* I "S
||g
*w 4* O
•e « °
tt W *O
nee
^ J» 09
o "a -
* ** »
« M 3
U O O
S -~. u
n -e js
• «J
5 9ai
on -H o
e u a a
JS S S1 •
v «-4 v *o
a u « ic
•J B « CO
•C I) CK
nge, facility owners
s by consulting curr
on of variables such
t personnel. From I1
453 S
O 4J 0 -1
*> Zi § &
O O u
•H « « -1
*o &i <*4 n
e s u
0 -1 fe u
§ 3 2
*J -H -< B
» "
V •< -O
0 U «
G V K M
2 « ¥ "
•83 S S
0 S g{
8 " S J
J », 0 |
« u «j 2
« -1 I
3 >§*
sl^s
3 « °l
•3 "8 ? t>
£ EJ S
«
4J
-------
H
H
<
£•4
' O3
a
s
H
g
5
w
o
H
2
w
s
o
1
SS
a
§
o
s
0
s
©
i-l
53
0
o
1-3
a
m
S
^gj
C
4
1
a
4
i
J
4
a
0
.p,
JE
, i
*
OS
1
1
••
.C
3
u
=>
GO
U
tQ
W
0)
CO
to
o
U 4J
JI-S.
Is
,g
cations
5
>^
Facilit
"x
03
a.
s
s
CD
S
CO
u 4j
g>^
e
5 S
•> lw
O
CO
e
o
*»-t
jj
o
*J
X
4J
•H
t— <
•*4
U
£
s"c;'"'w'~>-'s->-^'~'2CC~CS
ssstssaass^'sRs^j*-
^^5 a S. E SB gs
= 1 «• _r 5 2 * i 1 « « - = x i
If!i!!IIs!!ili!i
SSSSISlIIIIIssjssIaill I"2 s s o s
~ ~ ~ ~ ^ ^ ^ssscs^^^
sf^ojsfo«^a,sfSSRSSSSSS,f;SSsasSSS5S
«
tNrn^-H^tscn*Hc^csen f
^e>'"Ne*^*? CN' **f t*) •& ~* ^* ^^
.
:
O
H<4 Q
dd£S^ x^ se g _f u
** S ' <» CH" afi 6 Sf 3" S -
• • > jT M e< >4 §. S * o a- Bo' _ "g
»<>^SJ!Je2r
-------
tar
3
C
•»"4
O
V
CO
u
S3
H
CO
g
H
M
55
Ed
S
t"*4
CO
05
O
tjJJ
s
55
M
55
H
U
H
CO
g
»J
M
O
M
S5
2
Q
i
fa
O
CO
S5
M
S
S
*
tn
u
•4
s
r
i
c
t
B
Jj
S
A
c
4-
5
•
&
|
I
V
tt
4J
A
V
EC
DC
n
jj
i
V
<*4
0
1
J
w3
V -1
3
56 "o
o
2
u
o
J
K
.4
-4
U
&
•
A)
O
1
f
co e- o o
eo o ^» CM o en o
~
-
•» wr 10 in ** r«* ^4
r*. *n cj «h N o«
^ <
SB iT *
- i" tT . * §
o •§ 3 £ ^ jj •£
J> N > 4J . U g
i) j: «i »i a o -S
H OJ i> «l S S. -1
il K U -4 -4 F ^4
Q W O 0. to B M
•0
P
c
s
o
o
v>4
ja
Ml
o
c^
s
w
u
91
S
o
"&
•o
n
g
jf
e
M
•
01
ǤP
/* S
0 « 1) ^ .
"3 & o
•O • u
g S Oto
O — . ai
•S !"" 1
S M S g
B n
• -4 « U
JJ JJ • O
S52 &
S*^ ™ w
«T £ * °Q.
§ u «« .a
g g °s
•g«g"
C -t K
Di «H JJ JJ
-J JJ O O
« "S § 4J
« S "* §
a o a o
O ** S JJ
>> -rt O
jj *> S
wi « JJ *4
o jj S x
« -4 O •»
•«- « g
* o u
|«H S«
J « «'•
r <5
g
H
JJ * O 9
'§<«*'•
0 0 g JJ
jj S 3 "*
« S « 3
*J JJ -* J3
151"
B * |1
• X «
Ml JJ «H •
SO •
J3 K **
U > «
£ ^ 8
sll 1
•*n 6 *J
i-g . §
* Lr S ^
n « > -e
:--s
c u a •
u « J |j
"^ " g •"
.5 o S 8 3
fi«^.S4
•
jj
£
124
-------
brick arches into separate compartments or hearths. Alternate hearths are
designed with openings to allow solid material to drop onto the hearth
below. At the center of the unit, a shaft rotates rabble arms that are
located on each hearth. To enable the incinerated material to move inward
and then outward on alternate hearths, teeth on the rabble arms are placed
at an angle. As sludge is fed through the roof of the incinerator, the
rotating rabble arms and rabble teeth push the material across the hearth to
drop holes where it falls to the next hearth. This process continues until
the sterile ash produced by the oxidation steps is discharged from the
bottom of the incinerator. A schematic diagram of a typical multiple-hearth
sewage sludge incinerator is presented in Figure 8.89
The majority of multiple-hearth incinerators have three distinct
operating zones. The first zone includes the top hearths where the
water-laden sludge feed is partially, dried by rising hot combustion gases.
The second operating zone is the incineration/deodorization zone where
temperatures of 760°-980°C (1400-1800°F) are reached and maintained. The
third zone of the multiple-hearth unit is the cooling zone where hot ash
from incineration releases heat to incoming combustion air. •
*
The second technique used to oxidize sewage sludge is fluidized-bed
incineration. Figure 9 represents the basic operations found in a
fluidized-bed unit. In this operation, dewatered sludge is introduced into
the freeboard area of the incinerator just above the fluidized-bed material
(which is.usually sand). Air injected through tuyeres at pressures of from
3 to 5 psig fluidize the bed. Hot combustion gases rising from the bed
evaporate remaining water in the sludge and sludge solids then enter the
fluidized bed. The organic constituents of the sludge are oxidized to
carbon dioxide and water vapor which exit the system as exhaust gases.
During this reaction, the bed is vigorously mixed and the bed temperature is
maintained at 760-927°C (1400-1700°F). Material residence time in the
combustion zone, is 2 to 5 seconds. Remaining inorganic sludge material
either deposits on the bed sand particles and is removed from the bottom of
125
-------
FLUE GASES OUT
I J^-COOLING AIR DISCHARGE
3k-FLOATING DAMPER
DRYING ZONE
COMBUSTION ZONE
COOLING ZONE
ASH DISCHARGE
•SLUDGE INLET
=t
RABBLE ARM AT
EACH HEARTH
•COMBUSTION
AIR RETURN
RABBLE ARM
DRIVE
COOLING AIR FAN
Figure 8. Cross section of a typical multiple-hearth incinerator.
89
126
-------
Sight glass
Exhaust
PrahMt burnsr
Thermocouple
Sludge Inltt
Fluidlzlng
air Inlet
Figure 9. Cross sectioii of a fluidized-bed sewage sludge
incinerator.
127 -
-------
the reactor, or it can be made to exit with the exhaust gases. Air velocity
through the bed is used to control the method of inorganic sludge material
removal. . . •
Controlling the rate of feed of the sludge into the incinerator is the
most critical operating variable for fluidized-bed units. There is an upper
limit on the rate of heat transfer that can be achieved for a given quantity
of sand. If the rate of sludge feed exceeds the burning capacity of the
sand bed, combustion will not be complete. Similarly, either a rapid
increase in the overall furnace load or in the total moisture content of the
sludge will lead to coagulation of the sludge into heavy masses, depress the
bed, and halt combustion. It is also important, for the same reasons, to
ensure that an adequate residence time is available for the sludge to burn
completely.
Electric incinerators and rotary kilns are also used to destroy sewage
sludge, but to a much lesser degree than either multiple-hearth or.
fluidized-bed units. Electric incinerators consist of a horizontally
oriented, insulated furnace. A belt conveyor extends the length of the
furnace. Infrared heating elements are located in the roof above the
conveyor belt. Combustion air is preheated by the flue gases and is
injected into the discharge end of the furnace. To begin the process,
dewatered sludge cake is conveyed into one end of the-incinerator. An
internal roller mechanism levels the sludge into a continuous layer
approximately one inch thick across the width of the belt. The sludge is
dried and then burns as it moves beneath the infrared heating elements. Ash
is discharged into a hopper at the opposite end of the furnace.
The preheated combustion air enters the furnace above the ash hopper
and is further heated by the outgoing ash. The direction of air flow is
countercurrent to the movement of the sludge along the conveyor. Exhaust
gases leave the furnace at the feed end. Electric incinerators consist of a
number of prefabricated modules, which can be linked together to provide
the necessary furnace length.
128
-------
Rotary kilns have been applied in systems only requiring limited sludge
burning' capacity [up to 544 kg (1200 lb)/hr]. The typical kiln is inclined
slightly t,o! the horizontal plane, with the upper end receiving both the
sludge feed and the combustion air. A burner is located at the opposite end
of 'the kiln. The kiln rotates at a speed of about 6 inches per second, with
sludge being combusted as it moves down the kiln. Ash from incineration is
deposited into a hopper located below the burner.
Of the known total number of sewage sludge incinerators in the United
States, approximately 73'percent are multiple-hearth units, 20 percent are
fluidized-bed units, 5 percent are electric incinerators, 1.5 percent are
coincineration with municipal solid waste units and less than 1 percent are
rotary kilns.
Emissions from sewage sludge incinerators are controlled by various
designs of wet scrubbers. Scrubbers have evolved as the preferred method of
control for sludge incinerators because: (1) the treatment plant provides a
relatively inexpensive source of scrubbing water and a system for treating
scrubber effluent and (2) they have been shown to be effective particulate
matter emission control systems.
Prior to 1978, sludge incinerator emissions were controlled
predominantly by impingement tray, yenturi, and venturi/impingement tray
scrubbers. A detailed breakdown of controls prior to 1978 for
multiple-hearth, fluidized-bed, and electric incinerators is provided in
Table 36. After 1978, the dominant emission control system applied to
multiple-hearth arid fluidized-bed sludge incinerators was venturi/impingement
tray scrubbers. Venturi scrubbers were the primary emission control system
for electric incinerators. A detailed breakdown of post-1978 sludge
incinerator controls is given in Table 37. After 1978, a non-scrubbing
system, a baghouse, was also used for emissions control on a sewage sludge
incinerator. As shown in Table 37, since 1978, venturi/impingement tray
scrubbers are the preferred emissions control technology for sewage sludge
incinerators.
129
-------
TABLE 36. DISTRIBUTION OF EMISSION CONTROL TECHNIQUES APPLIED
TO SELECTED SEWAGE SLUDGE INCINERATORS PRIOR TO 1978
89
Venturi/Impingement Tray
Impingement Tray
Venturi
Electric Incinerators
Applications
to Incinerator's
68
23
9
Range of
Control Type
V
Multiple -Hearth Incinerators -
Impingement Tray
Venturi
Venturi/Impingement Tray
Spray Tower
Wet Cyclone
Venturi/Wet Cyclone
Fluidized-Bed Incinerators
Percent of Total
40
- ' 22
20
10
5
3
(in. w.g.)
6 - 9
15 - 32
15 - 35
4 - 9
3. - 4
15
12 - 40
4
17 - 18
venturi
Impingement Tray
Venturi/Wet Cyclone
57
29
14
4 - 9
6 - 9
12
130
-------
TABLE 37. DISTRIBUTION OF EMISSION CONTROL TECHNIQUES APPLIED
TO SEWAGE SLUDGE INCINERATORS AFTER 197889
Control Type
Multiple-Hearth Incinerator..
Venturi/Impingement Tray
Fabric Filter
Impingement Tray
Fluidized-Bed Incinerator?
Venturi/Impingement Tray
Venturi
Electric Incinerators
Venturi
Venturi/Impingement Tray
Applications
to Incinerators
Percent of Total
88
6
6
75
25
75
25
Range of
Pressure Drops
(in. w.g.)
10 - 45
10
42
Not Reported
8-10
. 10
131
-------
Emission Factot-g
Tests of air emissions from sewage sludge incinerators have confirmed
the existence of POM compounds from this source category; however,
quantitative POM emission factors are limited. The only POM emission
factors found in the literature for sewage sludge incinerators are presented
in Table 38. The seven POM compounds listed in Table 38 were specifically
analyzed for during the emission tests.
In tests of another sludge incinerator equipped with a wet scrubber
control device, 20 different POM compounds were-detected. A list of these
20 is provided in Table 39.91 The predominant POM compounds found in sludge
incinerator exhausts were phenanthrene, fluoranthene, fluorene, and pyrene.
The sampling technique used during the tests of the sludge incinerator
measured both particulate and gaseous POM. The greatest proportion of the
POM emissions were consistently found to be in a gaseous state.
Source Locations
According to a 1985 U, S. EPA .assessment, approximately 266 sewage
sludge incinerators are in existence in the United States. The distribution
of these 266 by type is as follows.92
• multiple hearth - 195
• fluidized bed - 54
• electric - 12
• rotary kiln - 2
• municipal solid waste co-combustion
- 3
Of the 266, 155 units are estimated to be in operation. The breakdown of
operating units within each major incinerator type is given below.
• multiple hearth - 119
• fluidized bed - 25
132
-------
TABLE 38.
POM EMISSION FACTORS FOR A SEWAGE SLUDGE
INCINERATOR CONTROLLED BY A WET' SCRUBBER
POM Compound
Acenaphthylene
Pyrene
Fluorene
Carbazole
Fluoranthene
Benzo(a)pyrene
Dibenzofuran
POM
,i i .1
Emission Factor, g/Mg (lb/ton)a>b'c
0.12 - 0.16 (0.00024'- 0.00032)
0.17 - 0.25 (0.00034 - 0.00050)
0.38 - 0.41 (0.00076 - 0.00082)
0.09 - 0.21 (0.00018 -:0.00042)
0.81 (0.0016)
0.002 - 0.007 (0.000004 - 0.000014)
1.1 - 1.3 (0.0022 - 0.0026)
Emission factors are expressed in terms of g (lb) POM emitted per Mg (ton)
of dry sludge fed to the incinerators. The range given is for two fmissJon
test runs.
Emission factors represent the joint final stack emissions of a sewage
sludge incineration facility containing two incinerators, each controlled bv
a wet scrubber. Scrubber exhausts are funneled to a single stack for
atmospheric release.
C
The sampling and analysis procedures used during these tests resulted in
both particulate and gaseous POM being measured.
133
-------
TABLE 39. POM COMPOUNDS IDENTIFIED IN SEWAGE
SLUDGE INCINERATOR EMISSIONS91
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Chrysene
Triphenylene
Benzo(g,h,i)fluoranthene
Benzo(a} anthracene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(e)pyrene
Benzo(a)pyrene
Perylene
0-phenylene pyrene
Dibenzo(a,c)anthracene
Dibenzo(a,h)anthracene
Benzo(b)chrysene
Benzo(g,h,i)perylene
Anthanthrene
134
-------
• electric - 8
• rotary kiln - 1
• municipal solid waste co-combustion - 2
A State-by-State distribution of the number of total and operating
sewage sludge incinerators by type is given in Table 40,92 The States where
sewage sludge incineration is predominantly practiced are New York,
Michigan, Connecticut, Pennsylvania, New Jersey,.Ohio, Virginia, and
California.
A list of wastewater treatment plants where sludge incineration is
thought to be practiced is given in Table 41.93'94 Because the list of
facilities in Table 41 was developed at a different time and by a different
group than that which developed the sludge facility distribution data in •
Table 40, the total number of facilities shown in the two tables differ
slightly.
PETROLEUM CATALYTIC CRACKING - CATALYST REGENERATION
ProcessDescrrDtion
Catalytic cracking processes currently serve a major role in modern
petroleum refineries by substantially increasing the production of gasoline
from a given amount of crude oil. This increased production is accomplished
by cracking heavier feedstocks such as atmospheric or vacuum gas oils to
produce slurry oil, light cycle oil, cracked gasoline, light gases, and
coke. Catalytic cracking takes place in the presence of a catalyst which
can become deactivated through the continual deposition of carbon, in the
form of coke, on active sites. To combat catalyst degradation, catalysts
are regenerated by combusting the coke deposits on the catalyst_- This
combustion of coke or catalyst regeneration process has been found to form
POM emissions, and thus makes catalytic cracker catalyst regenerators
potential POM emission sources.96'97
135
-------
i
«^
m
co
ON
•WDf
Z
M
04
X
H
Q
<«
1
CO
«
CO
w
H
j
U
fa
g
M
H
CO
1
o
u
W
O
•3
CO
b.
o
O
S3
O
B*
H
co
3
H
A
h«*
§
fi
P-
O
w
s
u
•J
e
H
U
*
4.
0
t£
•«
1
***•
U
u
•^
M
.5
•J
e
w
g.
o
_i
v
o
H
00
•H
a
u
0
"e
0
H
M
c
*«
a
&
0
0
o
H
^
5
a
u
u
o
•3
^
^
£
ro
c>
H
•O
'5
b
j:
|
V
fi
I
.f
w
1
V*
9
2
w
.s
u
&
o
19
2
•
o
n
U3
fJ-*tooo-*r».-«— icN
*9>i«)in^HOCsr^c^c«)c«^«9-(^*a-^0c4tv)Csiirt*^«nr^r^
oooo-ds<-}
n*A£*B*^coo;*«-4*ogcC9triCouGt&.A>;kft
<<«wwb.»B«-«»«^jSg^rsxS^a!z
. 136
-------
5
c
^j
2
0
V--
es
tri
co
a*
OW)
z
!-<
U
sw
H
O
U
H
H
tn
>~
CQ
CO
&J
E
H-l
CJ'
£
O
r— •
H
en
£
1
§
S3
CC
Eii
O
z
0
H
ea
5
C-l
en
S
u
i
o
Erf
SK
0,
O
*^
-3
5
H
Cl
J.
c
E-
0
j.
et
V
• n
3 U
; s.
* 0
J
_
a
£
e
c
4.
(fl
j
c
u
1
c
B
-S
i
^
w
(Q
B
l~4
.H
4J
U
u
s
•H
4J
a
b
S.
o
0
u
a
a.
o
_
1
CO
•5
ts
u
01
a.
o
Ml
a
•a
05
"S
'-5
s
s
y
'i
kl
a.
0
a
o
J.
a
u
B
U
«M
O
s
s
w
c
a
QJ
a,
0
1
h*
V
o
'
CM ^^ -* O
0 « « CM - ^ «n
§
•*4 v *e
"5 ' ••< B
S S J5
J£ o eg > 7
IW U S jiui M
>* J J= O CO 0)
| S 1 2 £ S 1
oo
ooo
OOOOOOCM
— ooco
137
-------
TABLE 41. LOCATIONS OF WASTEWATER TREATMENT PLANTS THOUGHT
TO BE USING SEWAGE SLUDGE INCINERATORS93'94
Incinerator Locations
POTW Name
Multiple-Hearth Pnifc«
Anchorage, AK
Palo Alto, CA
San Mateo, CA
South Lake Tahoe, CA .
Truckee, CA
Martinez, CA
Redwood City, CA
San Clemente, CA
Cromwell, CT
Hartford, CT
Killingly, CT
Naugatuck, CT
New Haven, CT
New Haven,- CT
New London, CT
Norwalk, CT
Waterbury, CT
Willimantic, CT
Jacksonville, FL
Pensacola, FL
Atlanta, GA
Atlanta, GA
Cobb County, GA
Savannah, GA
Marietta, GA
Honolulu, HI
Honouliuli, HI
Oahu, HI
Decatur, IL
Rock Falls, IL
Rockford, IL
Point Woronzof STP
Palo Alto WWTF
Not Reported
South Tahoe WWTF
Tahoe-Truckee WWTF
Not Reported
Not Reported
San Clemente WWTF
MDC WPCF
Hartford WPCF
Killingly WPCF
Naugatuck WPCF
East Shore WPCF
Boulevard WPCF
New London WPCF
Norwalk WPCF
Waterbury WPCF
Willimantic WPCF
Buckman Street STP
Main Street Plant
R. M. Clayton
Utoy Creek
Chattahoochee
President Street WPCF
Not Reported
Sand Island WWTF
Honouliuli WWTP
Not Reported
Decatur STP
Rock Falls STP
Rockford S.D. STP
138
-------
TABLE 41. LOCATIONS OF WASTEWATER TREATMENT PLANTS THOUGHT
TO BE USING SEWAGE SLUDGE INCINERATORS93'94 (Continued)
Incinerator Locations
POTW Name
East Chicago, IN
Indianapolis, IN
Indianapolis, IN
Cedar Rapids, IA
Davenport, IA
Johnson County, KS
Shawnee Mission, KS
Kenton County, KY
Algiers, LA
Lake Charles, LA
Lake Charles,'"LA
Annapolis, MD
Baltimore, MD
Riviera Beach, MD
Attleboro, MA
Chicopee, MA
Fall River, MA
Fitchburg, MA
Lawrence, MA
New Bedford, MA
Quincy, MA
Worcester, MA
Ann Arbor, MI
Ann Arbor, MI
Bay City, MI
Bay County, MI
Battle Creek, MI
Detroit, MI
East Lansing, MI
Flint, MI
Grand Rap'ids, MI
Kalamazoo, MI
Lansing, MI
Niles, MI
Owosso, MI
East Chicago STP
Belmont Street Plant
Southport WWTP
Cedar Rapids WPCF
Davenport WWTP
Mission Township STP
(Main Sewer District No.
Turkey Creek MSD #1
1)
Not Reported
New Orleans W.
Plant C
Plant B
Bank STP
Annapolis City STP
Patapsco WWTP
Cox Creek WWTP
Attleboro WWTW
Chicopee WWTP
Fall River STP
Fitchburg East WWTP
Greater Lawrence SD WWTP
New Bedford WWTP
Nut Island WWTP
Upper Blackstone Reg WWTP
Ypsi Community WWTP
Ann Arbor WWTP
Bay City STP
Bay County STP
Not Reported
Detroit STP
East Lansing WWP
Flint WPCF
Grand Rapids
Kalamazoo QQTP
Lansing WWTP
Niles Wastewater Treatment Plant
Owosso WWTP
139
-------
r
TABLE 41. LOCATIONS OF WASTEWATER TREATMENT PLANTS THOUGHT
TO BE USING SEWAGE SLUDGE INCINERATORS93'94 (Continued)
Incinerator Locations
POTW Name
Pontiac, MI
Trenton, MI
Warren, MI
Wayne County, MI
Eagan, MN
St. Paul, MN
St. Eouis, MO
St. Louis, MO
Kansas City, MO
Zephyr Cove, NV
Lebanon, NH
Merrimack, NH
Manchester, NH
Atlantic City, NJ
Bridgewater, NJ
Jersey City, NJ
Parsippany-Troy Hills, NJ
Princeton, NJ
Union Beach, NJ
Wayne Township, NJ
Albany, NY
Amherst, NY
Babylon, NY
Beacon, NY
Buffalo, NY
Dunkirk, NY
Greece, NY
Mamaroneck, NY
Menands, NY
Mechanicville, NY
New Rochelle, NY
New Windsor, NY
Orangeburg, NY
Ossining, NY
Oswego, NY
Pontiac STP
Trenton WWTP
Warren WWTP
Wyandotte STP
Seneca Treatment Plant
Metropolitan TP
Bissell Point STP
Lemay STP
Kansas City Big Blue River STP
Douglas County SID #1 WWTF
Lebanon WWTF
Merrimack WWTP
Manchester WWTP
Atlantic County SA
Somerset-Raritan
West Side STP ' '
Rockaway Val Regn S A TRT
Stony Brook RSA STP #1
Bayshore Regional STP
Mountain View STP
South Albany STP
Not Reported
Southeast SD #3
Beacon WPCP
Birds Island STP
Dunkirk STP
N W Quadrant TP
Mamaroneck San. Sew. District
North Albany STP
Saratoga SD #1
New Rochelle SD STP
New Windsor STP
Orangetown DPW
Ossining SD
East STP
140
-------
TABLE 41. LOCATIONS OF WASTEWATER TREATMENT PLANTS THOUGHT
TO BE USING SEWAGE SLUDGE INCINERATORS93'94 (Continued)
Incinerator Locations
Oswego, NY
Port Chester, NY
Rochester, NY
Rochester, NY
Schenectady, NY
Southampton, NY
Tonawanda, NY
Utica, NY
Wheatfield, NY
Greensboro, NC
Rocky Mount, NC
Akron, OH
Canton, OH
Cincinnati, OH
Cincinnati, OH
Cleveland, OH
Cleveland, OH
Columbus, OH
Columbus, OH
Euclid,.OH '
Youngstown, OH
Tigard, OR
Ambridge, PA
Apollo, PA
Bridgeport, PA
Chester,-PA
Colmar, PA
Erie, PA
Hershey, PA
Johnstown, PA
Norristown, PA
North Wales, PA
Old Forge, PA
Pittsburgh, PA
West Hazelton, PA
Wilkes-Barre, PA
Willow Grove, PA
York, PA
Lemoine Borough, PA
POTW Name
West STP
Port Chester SD STP
Gates Chili Ogden STP
Frank E. Van Lare WWTP
Schenectady STP
Disposal District No. 15
Two Mile.Creek SD Plant 2 •
Oneida County WPCP
Niagara CO SD #1 STP
North Buffalo WTP
Rocky Mount WWTP
Akron WWTP
Canton WWTP
Millcreek WWTP
Little Miami WWTP
Westerly WWTP
Southerly WWTP
Jackson Pike WTP
Columbus-Southerly WWTP
Euclid WWTP
Youngstown WWTP
Durham Regional STP
Ambridge STP
Kiski Valley WPCA
Bridgeport STP
Delcora Chester STP
Hatfield Township STP
Erie City STP
Derry Township WPCP
City of Johnstown
E. Norriton Plymouth TP
Upper Gwynedd Township STP
Lower Lackawanna STP
Alcosan WWTP
Greater Hazelton STP
Wyoming Valley San. Authority
Upper Moreland-Hatboro TP
York WPCC
Cumberland City
141
-------
TABLE 41. LOCATIONS OF WASTEWATER TREATMENT PLANTS THOUGHT
TO BE USING SEWAGE SLUDGE INCINERATORS93'94 (Continued)
Inqinerator Locations
POTW Name
Cranston, RI
Charleston, SC
Columbia, SC
Bristol, TN
Maryville, TN
Nashville, TN
Newport, TN
Alexandria, VA
Arlington, VA
Blacksburg, VA
Fairfax, VA
Newport News, VA
Norfolk, VA
Norfolk, VA
Virginia Beach, VA
Williamsburg, VA
Woodbridge, VA
Bellingham, WA
Vancouver, WA
Clarksburg, WV
Brookfield, WI
DePere, WI
Green Bay, WI
Milwaukee, WI
F_luidized-Bed Units
North Little Rock, AR
San Bernardino, CA
South Bayside, CA
Redwood City, CA
Barstow, CA
Stratford, CT
Cranston WPCF
Plum Island TRT Plant
Metropolitan TRT Plant
Galloway Mill Plant
Maryville Regional .STP
Nashville Central WWTP
Newport WWTP
Alexandria STP
Arlington County WPCP
Lower Straubles STP
Lower Potomac STP
Boat Harbor WPCF
Lamberts Point WPCF
Army Base WPCF
Chesapeake-Elizabeth WPCF
Williamsburg WPCF
Potomac River STP
Bellingham Plant
Vancouver Westside STP
Clarksburg STP
Brookfield STP
DePere WWTP
Green Bay WWTP
South Shore Waste Water TP
Faulkner Lake STP
San Bernardino WWTP #2
South Bayside WWTP
Not Reported
Barstow Regional WWTP
Stratford WPCF
142
-------
TABLE 41. LOCATIONS OF WASTEWATER TREATMENT PLANTS THOUGHT
TO BE USING SEWAGE SLUDGE INCINERATORS93'94 (Continued)
Incinerator Locations
Elkart, IN
Dubuque, IA
Kansas City, KS
Kansas City, KS
New Orleans, LA
Ocean City, MD
Lynn, MA
Port Huron, MI
Duluth, MN '
Independence, MO
Omaha, NE
Somerset-Raritan, NJ
Two Bridges, NJ
Union Beach, NJ
Waldwick, NJ
West Bedford, NJ
1 Hamburg, NY
Port Washington, NY
Poughkeepsie, NY
Shelby, NC
Hazelton, PA
King of Prussia, PA
Tyrone, PA
Clarksville, TN
Edmonds, WA
POTW Name
Elkart WWTF
Dubuque WWTP
Not Reported
KCK WWTP #1 - RAW Point
New Orleans East Bank
Ocean City WWTP
Lynn Regional WPCP
Port Huron STP
WLSSD Regional WWTF
Not Reported
Papillion Creek
Not Reported
Fairfield Sewer Authority
Not Reported
N.W. Bergen
Not Reported
Southtowns STP
Port Washington STP
Arlington SD
Not Reported
Hazelton STP
Trout Run WPC :
Tyrone Borough STP '
Clarksville Main WWTP
Edmonds STP
143
-------
TABLE 41. LOCATIONS OF WASTEWATER TREATMENT PLANTS THOUGHT
TO BE USING SEWAGE SLUDGE INCINERATORS93'94 (Continued)
Incinerator Locations
POTW Name
Electric Units
Wrangell, AK
Decatur, GA
Gainesville, GA
Cynthiana, KY
Bay County, MI
Sylvan Beach, NY
Fayetteville, NC
Greenville, TX
Aberdeen, WA
Rotary Kiln Units
Lake Arrowhead, CA
Los Angeles, CA
Hopewell, VA
Combination Sludge/
Municipal Waste Units
New Canaan, CT
Stamford, CT
Glen Cove, NY
Harrisburg, PA
Other Sludge Units
Rockville, CT
Lyons, KS
Wrangell WWTP
Snapfinger WWTP
Flat Creek WPCP
Not Reported
Not Reported
East Oneida Lake WPCP
Cross Creek Plant
Greenville STP
Aberdeen STP
Lake Arrowhead WWTF
Hyperion WWTP
Hopewell STP
New Canaan WPCF
Stamford WPCF
Glen Cove STP
Harrisburg, STP
Vernon WPCF
Lyons STP
144
-------
TABLE 41. LOCATIONS OF WASTEWATER TREATMENT PLANTS THOUGHT
TO BE USING SEWAGE SLUDGE INCINERATORS93'94 (Continued)
Incinerator Locations
<•
St. Charles,'MO
Independence, MO
Meadow Grove,.NE
Tangier, VA
POTW Name
St. Charles Miss. River STP
Rock Creek WWTP
Meadow Grove WWTP
Tangier STP
NOTE:
"i^••-••-"^»«^—^_^__^__^^_
This listing is subject to change as market conditions change,
facility ownership changes, plants are closed down, etc. The reader
currentlLl? ^/^T °f Particul" faciliti^ by consulting
current listings and/or the plants themselves. The level of POM
ca^clS8 S°m T SlVen faCllity " a function of variables such as
tSn ?;• ShpUt' ^ C0ntr01 measu^, ^d should be determined
through direct contacts with plant personnel. From 1986 .reference
145
-------
Three types of catalytic crackers have been used in the petroleum
industry: fluid-bed catalytic cracking '(FCC) units, Thermofor catalytic
cracking (TCC) units, and Houdriflow catalytic cracking (HCC) units.
Thermofor and Houdriflow crackers are both moving bed designs. Fluid-bed
catalytic crackers greatly dominate over the other two cracking unit types.
As of 1979, FCC units constituted 94 percent of total cracking feed
capacity, TCC units 5^percent, and HCC units 1 percent.95 With the advent
of new and better catalysts, major design and operational changes have been
incorporated in FCC units in recent years. By contrast, no major changes in
moving bed type units have been observed and generally TCC and HCC units are
being phased out.
A process flow diagram of a typical FCC unit is shown in Figure 10.95
In the FCC process, hot regenerated catalyst, mixed with hydrocarbon feed,
is transported into the cracking reactor. The reactor, which is maintained
at about 482°C (900°F) and 15 psig, contains a bed of powdered
silica-alumina type catalyst which is kept in a fluidized state by the flow
of vaporized feed material and steam. ' Cracking of the feed, which
occurs in the riser leading to the reactor and in the fluidized bed, causes
a deposit of coke to form on the catalyst particles. A continuous stream of
spent catalyst is withdrawn from the reactor and steam stripped to remove
hydrocarbons. The catalyst particles are then pneumatically conveyed to. a
catalyst regeneration unit. Hydrocarbon vapors from the cracking process
are fractionated in a distillation column to produce light hydrocarbons,
cracked gasoline, and fuel oil.95
In the catalyst regeneration unit, coke deposits are burned off at
temperatures near 538°C (1000°F) and pressures ranging from 2 to 20 psig.97
This coke combustion process is the source of POM emissions in regeneration
95-97
units. The regenerated catalyst is continuously returned to the
cracking reactor. Heat added to the catalyst during regeneration (coke
combustion) furnishes much of the required heat for the cracking reaction.95
Regenerator flue gases, high in carbon monoxide and unburned hydrocarbons
(including potentially POM compounds) are vented to a carbon monoxide waste
146
-------
CO
03
0)
U
O
00
c
U
CO
i-i
u
>v
r-4
SB
CO
U
A
*O
•H
§
S-.
00
re
O
iH
01
147
-------
heat boiler or directly to the atmosphere.95'97 Waste heat boilers which
'are fired with an auxiliary fuel or contain a catalyst are reported to have
been 99 percent efficient in reducing PAH emissions from a regeneration
unit.
95
In several installations, particulate matter emissions from the
waste heat boiler are controlled by an ESP. Catalytic cracking units
constructed after June 1973 are subject to a new source performance standard
that limits carbon monoxide and particulate matter emissions to such a level
that a waste heat boiler and ESP are generally required for compliance.95
Thermofor and Houdriflow cracking units, illustrated in Figures 11 and
12, operate similarly to FCC units but use beaded or pelleted
95 97
catalysts. ' In both TCC and HCC, the cracking process is initiated by
having regenerated catalyst and vaporized hydrocarbon feed enter the top of
the cracking reactor chamber and travel co-currently downward through the
vessel. As the cracking process proceeds, synthetic crude product is
withdrawn and sent to the synthetic crude distillation tower for.processing
into light fuels, heavy fuels, catalytic gasoline, and wet gas.95 .At the
base of the reactor, the catalyst is purged with steam to remove
hydrocarbons and is then gravity fed into the. catalyst regeneration chamber.
In the regeneration chamber, combustion air is added at a controlled rate to
burn off catalyst coke deposits. As in FCC units, coke burning produces POM
emissions that are released in TCC and HCC catalyst regenerator flue gases.
Regenerated catalyst is collected at the bottom of the chamber and is
conveyed by airlift to a surge hopper above the cracking reactor where it
can be gravity fed back into the cracking process.95
Flue gases from TCC and HCC units are also either vented directly to
the atmosphere or to a carbon monoxide waste heat boiler. Thermofor units
have also been equipped in some installations with direct-fired afterburners
called plume burners. The plume burner is a secondary stage of combustion
built into the catalyst regeneration chambers. This type of burner
successfully increases the clarity of plumes from regeneration flue gases;
however, compared to a carbon monoxide waste heat boiler, the plume burner
is ineffective at reducing POM emissions.
148
-------
VAPOR FEED
STEAM,
PURGE
STEAM
COMBUSTION
AIR '
SURGE
HOPPER
REACTOR
CATALYST
REGENERATOR
i
CATALYST
COOLERS
SYNTHETIC
CRUDE
FLUE GAS (POM)
EMISSIONS
FLUE GAS (POM)
EMISSIONS
LIFT AIR
Figure 11. Diagram of a Thennofar catalytic cracking
process.97
-------
VENT
REACTOR
FEED
SYNTHETIC
CRUDE
PURGE,
STEAM
COMBUSTION
AIR
REACTOR
CATALYST
REGENERATOR
LIFT
DISENGAGER
_r\
STEAM
CATALYST LIFT
PIPE
FLUE GAS (POM)
EMISSIONS
LIFT
ENGAGER
Figure 12. Diagram of a Houdriflow catalytic cracking
process.^7
150
-------
Another means of reducing potential POM emissions from the catalyst
regenerators would be to achieve more complete combustion of carbon monoxide
to carbon dioxide. Recently developed processes such as UOPR hot .
regeneration and Amoco UltracatR may aid in the achievement of lower overall
POM emissions. The relatively higher temperatures for catalyst regeneration
used in the UOP process serve to improve coke combustion efficiency and thus
potentially reduce POM formation and emissions. One drawback to the UOP
process is that due to its higher temperatures, special materials of
construction are required, thus making it more suitable for. new cracking
units as opposed to existing units. The Amoco process, however, is based on
improving the catalytic reactor efficiency and allowing more complete
combustion to occur in the catalys.t regenerator without having to operate at
higher temperatures. Because changes in basic equipment are minimal with
the Amoco process, it is more amenable for retrofitting existing units.95
Emission Factors
Available emission factor data for catalyst regenerators used in FCC,
TCC, and HCC units are presented in Table 42.97 Factors for all three types.
of units generally exhibit a large amount of variability. In uncontrolled
FCC units, pyrene, phenanthrene, and fluoranthene were the predominant
compounds measured. Perylene, anthracene, and coronene were not detected in
uncontrolled emissions from the FCC unit. Benzo(a)pyrene levels were found
to be relatively minor (average of 169 ug/barrel of oil feed versus average
of 133,000 ug/barrel of oil feed for phenanthrene). The positive effect of
carbon monoxide waste heat boilers as control devices for FCC regenerator •
flue gases is well evidenced in Table 42. v
Polycyclic organic matter emissions were in general highest from the
controlled TCC unit (air lift type) and the uncontrolled HCC unit. In the
air lift TCC unit, pyrene, phenanthrene, benzo(g,h,i)perylene, and
benzo(a)pyrene emission levels were the highest of the ten POMs measured.
Similarly, benzo(g,h,i)perylene, benzo(e)pyrene, pyrene, and benzo(a)pyrene
were the most significant compounds measured in uncontrolled HCC unit
151
-------
-------
-------
w=
•
c
X
•
a
1
:
S
J
s
y
s
•'
s,
J
R
o'
• ••• cot »ftciti*4
;•
Itfcrcoct 97
•
V
o*
g
•
3
*
§
S.
I
1
o-
S
•j
8
I
#*
i
8
§
|
e
i'
ft
4
I
\
i
*
E
A
•
0
1
S-
S
I
1
5
i
•*.
t
s
* 2
o
r* tv
S'8
!•
5.8
if
a* •
»•«
i»
ss
£3
'g
i
a
V
^
1
a
4
8-
I
I
\
\
I
*<
1
{
R
5
f*
i
|
?l
P S
8 £?
f S§
• An
• si.
f »?
" f *
A
rt
m *•• i
•»
.'J
• g '
IF
§
« 5-
. w
f
y
l?
c K5
i§
yl
5 £'
v S
o?
§
.»i
r. £ '
M ^
rt
m ^
b 5 '
s-a
•* 0
•^ i
§***
M
" 1
C £,
S **
gi
Is
c ^
sr*
O J"
n
]T
Ht
Ti1
A rf
a 1
* 2.
?Z
?
IT
r?
r»
*
82
,;
»•«
3
3
3
3
3
O
»*
r '
S8
r
s8 a
s^
ire rs
15 ?ff
3 S ?
O r* r*
r»
o
r*
«• *- K»
1.8 =r
B§ "
'I r»
2 ' *
|g 85
§
t!^?
V8 5S
-J
u,
^
•> ' 3
^8*
«' r '
V M 5 V*
£v
s
?f 3
v*.
8§
fg
B« 3
e e
r I
-. s
• Kt
5i
?!
r s
M 1 M
utt
"?S »-
> i 7 o
• •_ y g
II
g
II
"1
J
1
>.*•
M M
S|
,.
«• M
sf
!;
Ii
s»
!?
4* £
3
3
fg
M
X S
>3
* 1
si
WO
si
ft
C.8
*"* X
f
'
f
*•
•
*•
7
5
81
*
B
O
f
i
JT
•
*
j
i
7
|
i
»
1
i
i:
£
3
S
|
•1
•
*
S
9
I
I
s
g
•1
ft
D
w
«
?
1
r
B
S.
V
f
i
r
)
«
i
o
1
I
^
r
i
i
>
>
?
b
p
ik
j
•*
"
H
S
1
(/>
VI
M
O
z
S
1
nn
o
•ti
w
g
1
H
n
o
^
5
M
Z
o
f
M
H
S
W
I
i
M
S*
•Kj
_
-------
emissions. No explanation was offered in Reference 97 of why POM emissions
from the controlled bucket lift TGC unit were so disproportionately less
than that from the air lift TCC unit. Both types of TCC units were equipped
with plume burners.
The data for the HOC unit reinforces the effectiveness of venting
regenerator emissions to carbon monoxide waste heat boilers for POM emission
control. For each of the ten POM compounds measured, the waste heat boiler
reduced uncontrolled HCC regenerator emissions by greater than 99 percent.
Source Locations
Locations of catalytic crackers and their catalyst regenerators are
directly associated with the locations of petroleum refineries. As of
January 1, 1986, there were a total of 189 operating refineries,in the
United States. The States of Texas, California, and Louisiana contain about
41 percent of the total. Not all of the 189 refineries, however, contain
catalytic crackers. A list of the refineries containing catalytic crackers
and that may be potential POM emission sources is given in Table 43.98
SINTERING IN THE IRON AND STEEL INDUSTRY
Process Description
In the iron and steel industry, sintering processes convert materials
such as fine iron ore concentrates, blast furnace flue dust, mill scale,
turnings, coke fines, and limestone fines into an agglomerated product that
is suitable for use as blast furnace feed material. Sintering is necessary
because fine iron ore material, whether in natural or concentrated ores,
must be aggregated to a size and strength to prevent it from being blown out
QQ T on
of the top of a blast furnace. ' A typical sintering operation is
illustrated in Figure 13.101
153
-------
-------
TABLE 43. LOCATIONS OF ACTIVE PETROLEUM REFINERIES WITH
CATALYTIC CRACKERS AS OF JANUARY 198698
Company
Location
Li'on Oil Company
Atlantic Richfield Company
Champlin Petroleum Company
Chevron U.S.A., Inc.
Exxon Company
Fletcher Oil and Refinery Company
Golden West Refining Company
Mobil Oil Corporation
Shell Oil Company
Superior Processing Company
Texaco Refining and Marketing, Inc.
Tosco Corporation
Union Oil of California
Asamera Oil U.S., Inc.
Conoco, Inc.
Texaco Refining and Marketing, Inc.
Chevron U.S.A., Inc.
Clark Oil and Refining Corporation
Marathon Oil Company
Mobil Oil Corporation
Shell Oil Company
Texaco Refining and Marketing, Inc.
Union Oil of California
Amoco Oil Company
Indiana Farm Bureau Cooperative
Association, Inc.
El Dorado, AR
Carson, CA
Wilmington, CA
El Segundo, CA
Richmond, CA
Benicia, CA
Carson, CA
Santa Fe Springs, CA
Torrance, CA
Martinez, CA
Wilmington, CA
Santa Fe Springs, CA
Wilmington, CA
Martinez, CA
Los Angeles, CA
Commerce City, CO
Commerce City, CO
Delaware City, DE
Barber's Point, HA
Blue Island, IL
Hartford, IL
Robinson, IL
Joliet, IL
Wood River, IL
Lawrenceville, IL
Lemont, IL
Whiting, IN
Mt. Vernon, IN
154
-------
TABLE 43. LOCATIONS OF ACTIVE PETROLEUM REFINERIES WITH
CATALYTIC CRACKERS AS OF JANUARY 198698 (Continued) -
Company
Location
Rock Island Refining Corporation
Derby Refining Company
Farmland Industries, Inc.
National Cooperative Refinery Association
Pester Refining Company
Texaco Refining and Marketing, Inc.
Total Petroleum, Inc.
Ashland Petroleum Company
BP Oil, Inc.
Citgo Petroleum Corporation
Conoco, Inc.
Exxon Company
Gulf Products Company
' Hill Petroleum Company
Marathon Oil Company
Murphy Oil U.S.A., Inc.
Placid Refining Company
Shell Oil Company
Tenneco Oil Company
Texaco Refining and Marketing, Inc.
Marathon Oil Company
Total Petroleum, Inc.
Ashland Petroleum Company
Koch Refining Company
Amerada-Hess Corporation
Chevron U.S.A., Inc.
Ergon Refining, Inc.
Cenex
Indianapolis, IN
Wichita, KS
Coffeyville, KS
McPherson, KS
El Dorado, KS
El Dorado, KS
Arkansas City, KS
Catlettsburg, KY
Belle Chasse, LA
Lake Charles, LA
Lake Charles, LA
Baton Rouge, LA
Belle Chasse, LA
Krotz Springs, LA
Garyville, LA
Meraux, LA
Port Allen, LA
Norco, LA
Chalmette, LA
Convent, LA
Detroit, MI
Alma, MI
St. Paul Park, MN
Rosemount, MN
Purvis, MS
Pascagoula, MS
Vicksburg, MS
Laurel, MT
155
-------
TABLE 43. LOCATIONS OF ACTIVE PETROLEUM REFINERIES WITH
CATALYTIC CRACKERS AS OF JANUARY 198698 (Continued)
Company
Conoco, Inc.
Exxon Company
Montana Refining Company
Amerada-Hess Corporation
Coastal Eagle Point Oil Company
Exxon Company
Mobil Oil Corporation
Texaco Refining and Marketing, Inc,
Bloomfie'ld Refining Company
Giant Industries, Inc.
Navajo Refining Company
Amoco Oil Company
Ashland Petroleum Company
Chevron U.S.A., Inc.
Standard Oil Company of Ohio
Sun Cl
Conoco, Inc.
Kerr-McGee Refining Corporation
Sinclair Oil Corporation
Sun Cl
Total Petroleum, Inc.
Atlantic Richfield Company
BP Oil Corporation
Chevron U.S.A., Inc.
Sun Cl
United Refining Company
Mapco Petroleum, Inc.
Amber Refining Company
Location
Billings, MT
Billings, MT
Great Falls, MT
Port Reading, NJ
Westville, NJ
Linden, NJ
Paulsboro, NJ
Westville, NJ
Bloomfield, NM
Gallup, NM
Artesia, NM
Mandan, ND
Canton, OH
Cincinnati,' OH
Lima, OH
Toledo, OH
Toledo, OH
Ponca City, OK
Wynnewood, OK
Tulsa, OK
Tulsa, OK
Ardmore, OK
Philadelphia, PA
Marcus Hook, PA
Philadelphia, PA
Marcus Hook, PA
Warner, PA
Memphis, TN
Fort Worth, TX
-------
TABLE 43. LOCATIONS OF ACTIVE PETROLEUM REFINERIES WITH
CATALYTIC CRACKERS AS OF JANUARY 198698 (Continued)
Company
Location
American Petrofina, Inc.
Amoco Oil Company
Atlantic Richfield Company
Champlin Petroleum Company
Charter International Oil Company
Chevron U.S.A. , Inc.
Coastal States Petroleum Company
Fina Oil and Chemi.cal
Crown Central Petroleum Corporation
Diamond Shamrock Corporation
Exxon Company
Koch Refining Company
LaGloria Oil and Gas Company
Marathon Petroleum Company
Mobil Oil Corporation
Phillips 66 Company
Shell Oil Company
Southwestern Refining Cl
Texaco Refining and Marketing, Inc.
Texas City Refining, Inc.
Union Oil Company of California
Valero Refining Company
Amoco Oil Company
Big West Oil Company
Port Arthur, TX
Texas City, TX
Houston, TX
Corpus Christi, TX
Houston, TX
El Paso, TX
Port Arthur, TX
Corpus Christi, TX
Big Spring, TX
Houston, TX
Sunray, TX
Three Rivers, TX
Baytown, TX
Corpus Christi, TX
Tyler, TX
Texas City, TX
Beaumont, TX
Borger, TX
Sweeny, TX
Deer Park, TX
Odessa, TX
Corpus Christi, TX
Port Neches, TX
E! Paso, TX
Port Arthur, TX
Texas City, TX
Nederland, TX
Corpus Christi, TX
Salt Lake City, UT
•Salt Lake City, UT
157
-------
TABLE 43. LOCATIONS OF ACTIVE PETROLEUM REFINERIES WITH
CATALYTIC CRACKERS AS OF JANUARY 198698 (Continued)
Company
Chevron U.S.A. , Inc.
Phillips 66 Company
RMT Properties, Inc.
Seagull Refining Company
Amoco Oil Company
Mobil Oil Corporation
Shell Oil Company
Texaco Refining and Marketing, Inc,
Murphy Oil U.S.A., Inc.
Amoco Oil Company
• Big West Oil Company
Little American Refining Company
Sinclair Oil Corporation
Wyoming Refining Company
Location
Salt Lake City, UT
Woods Cross, UT
Salt Lake City, UT
Roosevelt, UT
Yorktown, VA
Ferndale, WA
Anacortes, WA
Anacortes, WA
Superior, WI
Casper, WY
Cheyenne, WY
Casper, WY
Sinclair, WY
New Castle, WY
Note:
This listing is subject to change as market conditions change,
facility ownership changes, plants are closed down, etc. The'reader
should verify the existence of particular facilities by consulting
current listings and/or the plants themselves. The level of POM
emissions from any given facility is a function of variables such as
capacity, throughput, and control measures, and should be determined
through direct contacts with plant personnel.
158
-------
ao
•H
U-l
CO
•H
Pn
159
-------
Sintering begins by mixing iron-bearing materials with coke or coal
fines, limestone fines (a flux material), water, and other recycled dusts
(e.g., blast furnace flue dust) to obtain the desired sinter feed
composition. The prepared feed is distributed evenly onto one end of a
continuous traveling grate or strand. After the feed has been deposited on
the strand, the coke on the mixture is ignited by a gas- or oil-fired
furnace. After the coke has been ignited, the traveling strand passes over
windboxes where an induced downdraft maintains combustion in the sinter bed.
This combustion creates sufficient temperatures [1300 to 1500°C (2400 to
2700°F)] to fuse the metal particles into a porous clinker that can be used
as blast furnace feed. '10°
Once the sintering process is completed, the sintered material is
discharged from the sinter strand into a crushing operation. Following
crushing, the broken sinter falls onto.sizing screens where undersized
material is collected and recycled to the start of the sintering process.
The oversized sinter clinker is then sent to. a cooling process. The most
common types of sinter coolers used include circular or straight line moving
. beds, quiescent beds, or shafts. Air or water is used as the cooling medium
in these coolers, with air being prevalent in newer plants and water being
dominant is older plants. The cooled sinter is either sent directly to a
blast furnace, sent to storage, or screened again, prior to blast furnace
usage, to obtain a more precise size specification.99'100
Polycyclic organic matter emissions originate in the sintering process
from the burning of coke and potentially oily materials in the sinter feed.
Potentially, POM emissions may be released from the sinter machine windbox,
from the sinter machine discharge point, and from sinter product processing
operations (i.e., crushing, screening, and cooling). Because of the high
temperatures used in sintering operations, it is probable that sinter plant
POM emissions are in both gaseous and particulate matter forms.99'102
Emissions control at sintering facilities typically involves emissions
collection and conveyance to a standard particulate control device such as a
baghouse, ESP, or wet scrubber. If substantial quantities of POM emissions
160
-------
are in gaseous form, wet scrubbers would likely be the most efficient in
reducing total POM because gaseous compounds would be condensed in the
scrubber.99"102
Emission Factors
Only one emission factor for POM compounds from sintering operations is
available from the literature. Emissions of benzo(a)pyrene have been
determined to range from 600 ug/Mg to 1.1 g/Mg of sinter feed processed.
The precise source of the emissions (windbox, discharge point, etc.) and the
* '
control status of the source are not defined in the literature. Available
data do not indicate whether the range of 600 ug/Mg to 1.1 g/Mg represents
only particulate benzo(a)pyrene or particulate and gaseous
102
benzo(a)pyrene.
Source Locations
Iron and steel sintering facilities are located in conjunction with the
operation of iron and steel blast furnaces. The largest concentration of
sintering processes in tne United States is in the steel producing regions .
of Ohio, Pennsylvania, and Indiana. The American Iron and Steel Institute
indicated that no current central, organized list of sintering facilities
was available. Locations of sinter facilities according to a preliminary
survey in 1977 by the U. S. EPA are identified in Table 44.101
FERROALLOY MANUFACTURING
F.rocess Description
Ferroalloys are crude alloys of iron and one or more other elements
which are used for deoxidizing molten steels and making alloy steels. The
major types of ferroalloys produced are listed in Table 45.103 Ferroalloys
can be produced by five different processes, the primary method of
production being the use of electric arc furnaces (EAF). Emissions of POM
compounds are possible from ferroalloy manufacturing because coke or coal is
161
-------
TABLE 44. LOCATIONS OF IRON AND STEEL INDUSTRY
,101
SINTER PLANTS IN 1977
Company
Republic Steel Corporation
U. S. Steel Corporation
Kaiser Steel Corporation
C. F. & I. Steel Corporation
Interlake Steel Corporation
Calumet Steel Division
Republic Steel Corporation
U. S. Steel Corporation
Granite City Steel Division
Wisconsin Steel Division
Inland Steel Corporation
U. S. Steel Corporation
Youngstown Sheet and Tube
Bethlehem Steel Corporation.
Armco Steel Corporation
Bethlehem Steel Corporation
National Steel Corporation
Great Lakes Steel Company
Bethlehem Steel Corporation
Jones and Laughlin Steel
Republic Steel Corporation
Jones and Laughlin Steel
U. S. Steel Corporation
Republic Steel Corporation
U. S. Steel Corporation
Youngstown Sheet and Tube
Republic Steel Corporation
U. S. Steel Corporation
Plant Location
Gadsden, AL
Bessemer, AL
Fontana, CA
Pueblo, CO
South Chicago, IL
Chicago Heights, IL
Chicago, IL
South Chicago, IL
Granite City, IL
South Chicago, IL
East Chicago, IL
Gary., IN
East Chicago, IN
Burns Harbor, IN
Ashland, KY
Sparrows Point, MD
Detroit, MI
River Rouge, MI
Buffalo, NY
Star Lake, NY
Cleveland, OH
Cleveland,. OH
Lorain, OH
Youngstown, OH
Youngstown, OH
Campbell, OH
Warren, OH
Fairless Hills, PA
162
-------
TABLE 44. LOCATIONS OF IRON*AND STEEL INDUSTRY
SINTER PLANTS IN 197710"1 (Continued)
Company
Plant Location
U. S. Steel Corporation
U. S. Steel Corporation
U. S. Steel Corporation
Jones and Laughlin Steel
U. S. Steel Corporation
Bethlehem Steel Corporation
U. S. Steel Corporation
Alan Wood Steel
Bethlehem Steel Corporation
Wheeling-Pittsburgh Steel
Bethlehem Mines Corporation
Armco Steel Corporation
Lone Star Steel Company
U. S. Steel Corporation
Wheeling-Pittsburgh Steel
Weirton Steel Company
Braddock, PA
Saxonburg, PA
Homestead, PA
Aliquippa, PA
Rankin, PA
Johnstown, PA
McKeesport, PA
Swedeland, PA
Bethlehem, PA
Monessen, PA
Morgantown, PA
Houston, TX
Lone Star, TX
Provo, UT
Follansbee, WV
Weirton, WV
NOTE:
This listing- is subject to change as market conditions change,
facility ownership changes, plants are closed down, etc. The reader
should verify the existence of particular facilities by consulting
current listings and/or the plants themselves. The level of POM
emissions from any given facility is a function of variables such as
capacity, throughput, and control measures, and should be determined
through direct contacts with plant personnel.
163
-------
•TABLE 45. MAJOR TYPES OF FERROALLOYS PRODUCED IN THE UNITED STATES103
Silvery Iron
50 percent Ferrosilicon
65 to 75 percent Ferrosilicon
Silicon Metal
Calcium Silicon
Silicomanganese Zirconium
High-carbon Ferromanganese
Silicomanganese
Ferromanganese Silicon
Charge Chrome and High-carbon Ferrochrome
Ferrochrome Silicon
Calcium Carbide
Low-carbon Ferrochrome
Low-carbon Ferromanganese
Medium-carbon Ferromanganese
Chromium Metal
Manganese Metal
Ferrotitanium
Ferrovanadium
Ferromolybdenum
164
-------
charged to the high temperature smelting furnaces used in the ferroalloy
industry and burned. Because combustion efficiency in the furnace
environment is low, unburned hydrocarbons, including several POM compounds,
are formed and emitted with the furnace exhaust. Ferroalloy production
processes other than EAFs have not been identified as POM emission
99
sources.
The electric arc furnace method of ferroalloy production is depicted in
Figure 14. ' Metal ores and other necessary raw materials such as
quartz or quartzite (slagging materials), alumina (a reducing agent),
limestone, coke or coal, and steel scrap are brought to ferroalloy
facilities by ship, truck, or rail and stored on-site. Depending on its
moisture content and physical configuration, metal ore may need to be dried
and/or sintered prior to being crushed, sized, and mixed with other process
raw materials. Once the proper charge mixture has been prepared, the charge
is weighed and fed to a submerged EAF for smelting.
Three types of EAFs can be used for ferroalloy production. These
three, open, sealed, and semisealed, may be charged continuously or
intermittently. Electric arc furnaces contain three carbon electr6des which
are vertically suspended above the furnace hearth and extend I to 1.5 m (3
to 5 ft) into the charge materials. Three-phase current arcs through the
charge materials from electrode to electrode, and the charge is smelted as
electrical energy is converted to heat. The intense heat around the
electrodes [2204-2760°C (4000-5000°F)] results in carbon reduction of the
metal (e.g., chrome, manganese) and iron oxides in the charge and the
formation of the particular ferroalloy. The molten ferroalloy is
periodically tapped into ladles from tapholes in the lower furnace
wall.103'105
•
The molten ferroalloy is cast into molds and allowed to cool and
solidify. The casts are then removed from the molds, graded, and broken.
The broken ferroalloy is passed through a crusher and screened. The
ferroalloy product is then stored, packaged, and shipped to the consumer.
165
-------
J U
£g
.§
a: B
Ed £C
b. U
.j M
< z
O Id
^g
«
J M
S <3
g<
fc a.
9-
O HH
g^
CxJ H U 01
« z to
Q AM ^^
|
j
b.
Q
U
W 5 O
O X Z
ee w M
< => N
gBi M
U CO
CO
g
O
g
II"
8
-------
Impurities from the smelting process are trapped in a slag which forms
inside the electric arc furnace. The slag is periodically tapped and
treated by a concentration process to recovery metal values. Slag is •
processed in a flotation system, where metal particles sink to the bottom
while slag floats. The recovered metals are recycled to the furnace, and
the remaining slag is removed and disposed of.
Of the three types of EAFs that may be used to produce ferroalloys,
open furnaces are the most common type, and also have the highest potential
for particulate emissions. An open furnace is pictured in Figure 15. 10^ A
hood is usually located 1.8 to 2.4 m (6 to 8 ft) above the furnace crucible
rim. Dust and fumes from the smelting process are drawn into the hood along
with large volumes of ambient air. Advantages of the open furnace include
the ability to stoke it during operation and the flexibility to manufacture
-several types of ferroalloy without altering the furnace design.103"105
The semisealed (or semi -enclosed) furnace is pictured in Figure 16.
A cover seals the top of the furnace except for openings around the
electrodes through which raw material is charged. These furnaces are either
hooded or maintained under negative pressure to collect emissions from
around the electrodes. Because semisealed furnaces cannot be stoked,
crusting and bridging of ferroalloys around the electrodes and charge holes
may prevent uniform descent of the charge into the furnace and blows (jets
of extremely hot gases originating in the high temperature zone near the
electrode tips) may emerge around the electrodes at high velocity.103"105
The third type of EAF, the sealed or closed furnace, is illustrated in
Figure 17. Packing is used to seal the cover around the electrodes and
charging chutes. The furnace is not stoked and a slight positive pressure is
maintained to prevent leakage of the air into the furnace. Care must also
be taken to prevent water leaks which may cause explosive gas release which
could damage the furnace and threaten worker safety. Sealed furnace designs
are specifically used in the manufacture . of narrow families of ferroalloys,
so plants using sealed furnaces have less flexibility to produce different
types of ferroalloys.
167
-------
t
POTENTIAL
FUGITIVE POM
EMISSIONS
flit
ELECTRODES
_ EXTENDING
THROUGH
HOOD
' ' ' ' r ( (
HOOD
•*• DUST;!
.•••». •.
MIX FEED
CHUTE
(TYPICAL)
POM.EMISSIONS
I ' I ' I.' I ' • ' |
INDUCED AIR
Figure 15. Open electric arc furnace.
104
168
-------
POM EMISSIONS
MIX CHUTE
(TYPICAL)
COVER
TAP HOLE
POM EMISSIONS
Figure 16. Semisealed electric arc furnace.
104
169
-------
ELECTRODES .
MIX FEED
(TYPICAL)
ELECTRODE
SEAL
POM EMISSIONS
Figure 17. Sealed electric arc furnace.
104
170
-------
One alternative to the electric arc furnace process which can be used
to produce some low-carbon ferroalloys is a type of exothermic process
involving silicon reduction. A flow diagram of the process for chromium
ferroalloys is shown in Figure 18.104»105 First, chromium ore and lime are
fused together in a furnace to produce a chrome ore/lime melt which is
poured into a reaction ladle (number 1). Then a known quantity of molten
ferrochrome silicon previously produced in another reaction ladle (number 2)
is added to ladle 1. In the ladle, a. rapid, heat-producing reaction results
in the reduction of the chromium from its oxide form and the formation of
low-carbon ferrochrome and a calcium silicate slag. The ferrochrome product
is then cooled, finished, and packaged. Since the slag from ladle 1 still
contains recoverable chromium oxide, it is reacted in ladle 2 with molten
ferrochrome-silicon produced in a submerged arc furnace. The exothermic
reaction in ladle 2 produces the ferrochrome-silicon added to the number 1
ladle during the next production cycle.
A vacuum furnace process can also be used to produce low-carbon
ferrochrome ferroalloys. This type of furnace, pictured in Figure 19, is
charged with high-carbon ferrochrome and heated to a temperature near the
melting point of the alloy. Decarburization occurs as the high-carbon
ferrochrome is oxidized by the silica oxide in the ferrochrome. Carbon
monoxide gas resulting from the reaction is pumped out of the furnace to
maintain a high vacuum and promote decarburization of the
ferrochrome.104'105
The electrolytic process is another alternative to the electric arc
furnace for producing chromium and manganese ferroalloys. Pure chromium and
manganese metal is generally produced this way. Chromite or manganese ore,
high-chromic or high-manganese oxide slags, or ferrochrome or ferromanganese
can be used as raw materials for the process. Preparation of raw materials
can include grinding, calcining, and leaching. In the electrolytic process,
metal ions contained in an electrolytic solution are plated on cathodes by a
low voltage direct current. The pure metal forms a film oh the cathode
about 0.3 cm (1/8 in) thick, which is removed and prepared for
shipment.104'105
171
-------
Chrome
Ore
Lime
Chrome
Ore
i i
Coke
Quart-
Site
Wood
Chips
Chrome Ore/
Lime Melt
Open-Arc Furnace
Ferrochrome
Silicon
LJ.
Ferrochrome—
Silicon
Submerged-Arc
Ferrochrome-
Silicon
34% Cr
Product
Low Carbon
Ferrochrome
70% Cr
Secondary
Throw-away
Slag
Throw-away
Slag
Figure 18. Typical flow chart for the production of low-carbon
ferrochrome by the exothermic silicon reduction
process.105
172
-------
TO INERT
GAS COOLING
TO VACUUM
PUMPING SYSTEM
^REMOVABLE
END CLOSURE
TRACK
HEARTH
CAR
CARBON
RESISTORS'
FURNACE
CHARGE
Figure 19. Vacuum furnace for the production of low-carbon
ferrochrome.104
173
-------
All types of EAFs (open, sealed, semisealed) produce emissions
consisting of a variety of compounds, including POM, in both gaseous and
particulate forms. Baghouses were used to control emissions from 87 percent
of the open-arc ferroalloy .furnaces operating in 1980. Testing of these
control systems indicates total particulate removal efficiency of over
99 percent. Such systems should be very effective at controlling POM
compounds adsorbed onto fine particulate matter emissions. High
pressure-drop venturi scrubbers and electrostatic precipitators have also
been applied to open-arc furnaces producing ferroalloys. Reported total
particulate matter collection efficiencies for scrubbers ranged from 94 to
98 percent. When ESPs were used, the gas was conditioned with ammonia to
enhance particulate resistivity and increase collection efficiency.
Estimated total particulate matter removal efficiencies for the ESPs were
98 percent.99'103'105
In the case.af semisealed furnaces (Figure 16), offgases are drawn from
beneath the furnace cover through ducts leading to a control device.
However, fugitive particulates and fumes escape through the openings around
the electrodes. In some instances, hoods have been placed above the
furnaces to capture these emissions. Wet scrubbers, including both
multistage centrifugal scrubbers and venturi scrubbers, have been used on
semisealed ferroalloy furnaces. Up to 99 percent total particulate matter
removal efficiency has been reported for centrifugal scrubbers. Venturi
units can exhibit even greater efficiencies.
Venturi scrubbers are commonly used to control emissions from sealed
ferroalloy EAFs; however, the use of baghouses at a few installations has
occurred. In general, total uncontrolled emissions vented to a control
device from a sealed furnace are less than from other ferroalloy EAFs
because no air enters sealed furnaces. Resultant gas flows (volumes) to the
control device are only 2 to 5 percent of those from open furnaces.103
174
-------
Emission Factors
Polycyclic organic matter emission .factor data were identified for a
sealed and a semisealed ferroalloy manufacturing EAF. The sealed unit was
tested during the production of silicbmanganese and during the production of
ferromanganese. A high pressure drop wet scrubber was used to control
sealed furnace emissions regardless of the ferroalloy .being produced.
Controlled POM emissions during" silicomanganese production were measured to
be 1.0 g/Mw-h of energy consumed by the furnace. Pre-scrubber POM emissions
from ferromanganese production were 156 g/Mw-h. In the controlled POM
emissions sample, fluorene and anthracene were dominant, constituting,
respectively, 36 and 51 percent of total POM emissions. Seventy percent of
the uncontrolled emissions sample consisted of anthracene (35 percent) and
QQ 1 fi?
fluoranthene (35 percent). '
The semisealed ferroalloy furnace tested produced 50 percent
ferrosilicon and was controlled by a low energy wet scrubber followed by a
flare. Measurements taken after the scrubber but prior to the flare showed
total POM emissions to be 91.0 g/Mw-h. Fluorene constituted 50 percent of
the total POM quantity, followed by pyrene at 19 percent, fluoranthene at
18 percent, and anthracene and phenanthrene at 12 percent. '
A POM species specific list of the compounds and quantities measured
during testing of ferroalloy EAFs is presented in Table 46. "»103
Source Locations
The latest information published by the U. S. Bureau of Mines (BOM) on
the locations of ferroalloy manufacturing facilities in the United States is
given in Table 47. According to these data, as of 1984 there were
46 ferroalloy facilitie's in the United States operated by a total of
30 companies. Ohio and Pennsylvania contain the most ferroalloy facilities
with seven in each State. Ohio, Pennsylvania, Tennessee, and Alabama
together contain almost 57 percent of the total number of facilities
nationwide.
175
-------
176
-------
•o
0)
3
a
e
o
o
o\
"s
o
o
g
i
I
CO ;
W
U
06
o
td
ORS
g
2
O
CO
CO
o
•
o
-* in
• •
O -H
cn
o
o
o 2 2
o o o
i
8 §
-. 0
m
•
e
J
"§
a
o
u
•B
01
t!
a
t. 5
01
u
a
01 OI
a u
I I
1.0
C I
O
2
2
ON
f
g
M
i
a.
0)
h
3
01
§
3
a
B
i
"I
o .2
.2 g
b
3 w
•O B
01
5 8*
g *•
B B
a S
01 01
OI
0
! I
B .
•?
« 3
*u o
B £
01 I
3 S
§ &
i S
1
s
S
S
a ss • 5
MO >, g
•o S -
B « -4 o
a O a
« b 2 "£>
•u 01 b >-7
• IM oi
•o
§
o
I
01
B
b1
X
g
i £.
| a
-
g
B
2
B
g
b
§.
M
B
•rf
u
3
•e
o
O 3
Q
-------
TABLE 47. LOCATIONS OF FERROALLOY PRODUCERS IN
THE UNITED STATES IN 1984
Producer
FERROALLOYS
Affiliated Metals and Minerals,
Inc.
Aluminum Company of America,
Northwest Alloys, Inc.
AMAX, Inc., Climax Molybdenum
Company Division
Ashland Chemical Company
Cabot Corporation, KBI Division,
Penn Rare Metal Division
Dow Corning Corporation
Elkem A/S, Elkem Metals Company
Foote Mineral Company,
Ferroalloys Division
Hanna Mining Company
Hanna Nickel Smelting Company
Silicon Division
International Minerals and
Chemical Corporation, Industry
Group, TAG Alloys Division
A. Johnson and Company, Inc.
Kerr-McGee Chemical Corporation
Macalloy, Inc.
Plant Location
New Castle,. PA
Addy, WA
Langeloth, PA
Columbus, OH
Revere, PA
Springfield, OR
Alloy, WV
Ashtabula, OH
Marietta, OH.
Niagara Falls, NY
Cambridge, OH
Graham, WV
Keokuk, IA
New Johnsonville, TN
Riddle, OR
Wenatchee, WA
Bridgeport, AL
Kimball, TN
Lionville, PA
e
Hamilton (Aberdeen),
MS
Charleston, SC
Type of Furnace
Metallothermic
Electric
Metallothermic
Electric and
metallothermic
Metallothermic
Electric
Electric" and
electrolytic
Electric and
electrolytic
Electric
Electric
Electric
Electric
Electric
Electrolytic
Electric
178
-------
TABLE 47. LOCATIONS OF FERROALLOY PRODUCERS IN
THE UNITED STATES IN 1984 (Continued)
Producer
Plant Location
Type of Furnace
Metallurg, Inc., Shieldalloy
Corporation
Moore McCormack Resources, Inc.,
Globe Metallurgical, Inc.
Ohio" Ferro-Alloys Corporation
Pennzoil Company, Duval
Corporation
Reactive Metals and Alloys
Corporation
Reading Alloys, Inc.-
Reynolds Metals Company
SEDEMA S.A.,.Chemetals
Corporation
SKW Alloys, Inc.
Teledyne, Inc., Teledyne Wah
Chang, Albany Division
Union Carbide Corporation,
Metals Division
Union Oil Company of-California,
Molycorp, Inc.
FERROPHOSPHORUS
Electro-Phos Corporation
FMC Corporation, Industrial
Chemical Division
Monsanto Company, Monsanto
Industrial Chemicals Company
Newfield, NJ
Beverly, OH
Selma, AL
Montgomery, AL
Philo, OH
Powhatan Points, OH
Sahuarita, AZ
West Pittsburg, PA
Robesonia, PA
Sheffield, AL
Kingwood, WV
Calvert City, KY
Niagara Falls, NY
Albany, OR
Marietta, OH
Niagara Falls, NY
Washington, PA
Pierce, FL
Pocatello, ID
Columbia, TN
Soda Springs, ID
Metallothermic
Electric
Electric
Electric
Metallothermic
Electric
. <•
Metallothermic
Electric
Fused-salt
electrolytic
Electric
Metallothermic
Electric
Electric and
metallothermic
Electric
a'
Electric
Electric
Electric
179
-------
TABLE 47. LOCATIONS OF FERROALLOY PRODUCERS IN
THE UNITED STATES IN 1984 (Continued)
Producer
Plant Location
Occidental Petroleum Corporation,
Hooker Chemical Company,
Industrial Chemicals Group
Stauffer Chemical Company,
Industrial Chemical Division
Columbia, TN
Mount Pleasant, TN
Silver Bow, MT
Tarpon Springs, FL
Type of Furnace
Electric
Electric
Electric
Electric
NOTE:
This listing is subject to change as market conditions change
facility ownership changes, plants are closed down, etc. The reader
should verify the existence of particular facilities.by consulting
current listings and/or the plants themselves. The level of POM
emissions from any given facility is a function of variables such as
capacity, throughput, and control measures, and should be determined
through direct contacts with plant personnel.'
180
-------
-------
IRON AND STEEL FOUNDRIES
Process Description
Iron and steel foundries can be defined as those which produce gray,
white, ductile, or malleable iron and steel castings. Cast iron and steels
are both solid solutions of iron, carbon, and various alloying materials.
Although there are many types of each, the iron and steel families can be
distinguished by their carbon content. Cast irons typically contain
2 percent carbon or greater; cast steels usually contain less than 2 percent
carbon.
Iron castings are used in almost all types of equipment, including
motor vehicles, farm machinery, construction machinery, petroleum industry
equipment, electrical motors, and iron and steel industry equipment. Steel
castings are classified on the basis of their composition, and heat
treatment,, which determine their end use. Steel casting classifications
include carbon, low alloy, general purpose structural, heat resistant,
corrosion resistant, and wear resistant. They are used in motor vehicles,
railroad equipment, construction machinery, aircraft, agricultural
equipment, ore refining machinery, and chemical manufacturing equipment.107
The following four basic operations are performed in all iron and steel
foundries.
storage and handling of raw materials
melting of the raw materials
transfer of the hot molten metal into molds
preparation of the molds to hold the molten metal
Other processes present in most, but not all, foundries include:
sand preparation and handling;
mold cooling and shakeout;
181
-------
casting cleaning, heat treating, and finishing;
• ' - coremaking; and
pattern making.
A generic process flow diagram for iron and steel foundries is given in
Figure 20.107
Iron and steel castings are produced in a foundry by injecting or
pouring molten metal into cavities of a mold made of sand, metal, or ceramic
material. Input metal is melted by the use of~a cupola, an electric arc
furnace, or an induction furnace. About 70 percent of all iron castings are
produced using cupolas, with lesser amounts produced in electric arc and
induction furnaces. However, the use of electric arc furnaces in iron
foundries is increasing. Steel foundries rely almost exclusively on
electric arc or induction furnaces for melting purposes. With either type
of foundry, when the poured metal has solidified, the molds are separated
and the castings removed from the mold flasks on a casting shakeout unit.
Abrasive (shotblasting) cleaning, grinding, and heat treating are performed
as is necessary. The castings are then inspected and shipped to another
industry for machining and/or assembly into a final product.107
In a typical foundry operation, charges to the melting unit are sorted
by size and density and cleaned (as required) prior to being put in the
melter. Charges consist of scrap metal, ingot, carbon (coke), and flux.
Prepared charge materials are placed in crane buckets, weighed, and,
transferred into the melting furnace or cupola. The charge in a furnace or
cupola is heated until it reaches a certain temperature and the desired
chemistry of the melt has been attained. Once the desired product is
obtained, the molten metal is either poured out of the furnace into various
sized teeming ladles and then into the molds or it is transferred to holding
furnaces for later use.
The casting or mold pouring operation in iron and steel foundries has
been determined to be a source of POM emissions.108"111 The origin of these
POM emissions is suspected to be the organic binders, including coal powder
182
-------
I
tl*M (
I
t
0)
0)
"O
CO
O
11
S
ui
tn
co
u
•o
fl
03
f
*K
fi
3
co
^
00
£0
•H
_ ..I
i
en
co
0)
u
O r^
^ O
D.rH
(0
U
a* 3
>> o
'
1
IS
«B tk,
•ss
15
if
o
CM
0)
M
183
-------
and coal tar pitch, used to form the sand molds used for molten metal
casting. When the hot molten metal contacts the sand mold pyrolysis occurs
and a plume of smoke is generated which contains a rich mixture of organic
compounds including POMs. In addition to casting, mold preparation and
casting shakeout (removal'from the mold) activities have also been
determined to generate POM emissions. Potential POM emissions from molding
and casting appear to be a function of the type and quantity of organic -
binder used to produce casting" molds.
Emissions of POM from mold preparation, casting, and shakeout
• ; •
operations are fugitive in nature and likely exist in both particulate and
gaseous forms. Fugitive emissions from such sources are generally
controlled with local hooding or building ventilation systems that are
ducted to a control device (predominantly baghouses) or to the
atmosphere.107"111
EmissionFactors
No POM emission factor data for iron and steel foundries could be found
in the literature. However, the existence of POM emissions from molding and
casting operations has been confirmed by air sampling of worker breathing
zones and plant areas associated with these operations. The air sampling
that has been conducted has measured only particulate POM compounds. As
many as 50 POM species have been detected in foundry' air samples.109
Predominant POM compounds that have been detected include benzo(a)pyrene,
benzo(e)pyrene, perylene, phenanthrene, anthracene, fluoranthrene,
benzofluoranthenes, dibenzanthracenes, benzochrysenes, benzo(g,h,i)perylene,
109
and o-phenylenepyrene.
In the quantitative data that are available, POM emissions from using a
coal tar pitch binder appear to be greater than those associated with usine
109
a coal powder binder. The • concentration of benzq(a)pyrene in the
workplace air of two foundries using coal tar pitch binders ranged from less
.than 0.01 to 72 ug/m of air. The average level was 5 ug/m3.109 The level
184
-------
of benzo(a)pyrene as a function of the mass of particulate matter sampled
ranged from less than 0.01 to 6.7 ug/mg, with the average being
1.1 ug/mg.109
In four foundries using a coal powder binder, workplace air had a
benzo(a)pyrene level of less than 0.01 to 0.82 ug/m3 of air.108
level was 0.08 ug/m . The concentration of benzo(a)pyrene in sampled
particulate matter ranged from less than 0.01 to 0.32 ug/mg, with the
average being 0.03 ug/mg.108
The average
Source Locations
The 1980 U. S. EPA background information document for new source
performance standards covering electric arc furnaces in ferrous foundries
indicated that^there were' approximately 4400 iron and steel foundries in the
United States. 7 The States with the greatest percentage of foundries are:
Ohio (10.3 percent)
California (9.7 percent)
Pennsylvania (8.4 percent)
Michigan (7.9 percent)
Illinois (7.2 percent)
New York (6.0 percent)
Wisconsin (4.4 percent)
Indiana (4.4 percent)
As evidenced by these States, foundry locations can be correlated with areas
of heavy industry and manufacturing, and in general, with the iron and steel
production industry (Ohio, Pennsylvania, and Indiana).
Additional information on iron and steel foundries and their locations
may be obtainable from the following trade associations.
American Foundrymen's Society, Des Plaines, Illinois
National Foundry Association, Des Plaines, Illinois
185
-------
Ductile Iron Society, Mountainside, New Jersey
Iron Casting Society, Warrendale-, Pennsylvania
Steel Founders' Society of America, Des Plaines, Illinois
BY-PRODUCT COKE PRODUCTION
Process Description
The by-product coke production source category includes processes used
to treat coal to produce coke and the recovery and treatment of by-product
gases from coking to generate secondary products such as crude tars, light
oil, and ammonia. Coke is one of the basic raw materials used in blast
furnaces to convert iron ore into iron. Approximately 92 percent of the
coke produced in the United States is used for this purpose. Other than
blast furnaces, coke is principally'used in iron foundries, nonferrous
smelters, and the chemical industry.112
In the United States, coke is produced by two methods: the
contemporary by-product recovery or slot oven process and the original
beehive process. Currently, the slot"oven process accounts for
approximately 99 percent of the annual metallurgical coke production in the
United States.
The coking industry is generally classified into two sectors, furnace
and merchant. Furnace plants are owned by or affiliated with iron- and
steel-producing companies that produce coke primarily for consumption in
their own blast furnaces, although they also engage in some intercompany
sales among steel-firms with excesses or deficits in coke capacity.112 In
1984, there were 28 furnace plants, which 'accounted for roughly 92 percent ;
of the total coke production. Independent merchant plants produce coke for
sale on the open market and are typically owned by chemical or coal firms.
The 15 merchant plants in existence in 1984 accounted for about 8 percent of
the total coke produced. These firms sell most of their products to other
firms engaged in blast furnace, foundry, and nonferrous smelting
operations.
186
-------
By-product recovery coking facilities contain three major processing
operations: coal preparation and charging, thermal distillation of coal
(coking), and recovery of coking by-products. A generalized process flow
diagram for by-product recovery operations is shown in detail in
1X2
Figure 21. Although not shown in detail in Figure 21, 'coal preparation
and charging is the initial operation in by-product recovery coking plants.
The coal that is charged to the by-product coke ovens is usually a blend of
two or more low, medium, or high volatile coals that are generally low in
sulfur and ash. Blending is required to control the properties of the
resulting coke, to optimize the quality and quantity of by-products, and to
avoid the expansion exhibited by types of coal that may cause excessive
pressure on the oven walls during the coking process.
Goal is usually received on railroad cars or barges. Conveyor belts
transfer the coal as needed from the barges or from a coal storage pile to
mixing bins where the various types of coal are stored. The coal is
transferred from the mixing bins to a crusher where it is pulverized to a
preselected size between 0.15 and 3.2 mm (0.006 to 0.13 in). The desired
size depends on the response of the coal to coking reactions and the
ultimate coke strength that is required.
112
The pulverized coal is then mixed and blended, and sometimes water and
oil are added to control the bulk density of the mixture. The prepared coal
mixture is transported to coal storage bunkers on the coke oven batteries.
A weighed amount or volume of prepared coal is discharged from the bunker
into a larry car, a vehicle which is driven by electric motors and travels
the length of the battery top on a wide gauge railroad track. The larry car
is positioned over the empty, hot oven, the lids on the charging ports are
removed, and the coal is discharged from the hoppers of the larry car
through discharge chutes. The flow rate from the hoppers to the oven may be
controlled by gravity, a rotary table, or screw feeders. To prevent gases
from escaping during charging, a steam-jet aspirator is used in most plants
to draw gases from the space above the charged coal into the collecting
main.112
187
-------
ffl
(1)
O
U
O
"O
O
O
U-l
2
60
n
(0
CO
-------
Peaks of coal will form directly under the charging ports as the oven
is filled. These peaks are leveled by a steel bar that is cantilevered from
the pusher machine through an opening called the chuck door on the pusher
side of the battery. This leveling process provides a clear vapor space and
exit tunnel for the gases that evolve during coking to flow to the
standpipes and aids in the uniform coking of the coal. After filling, the
chuck door and the topside charging ports are closed. In some plants, the
charging ports are sealed with a wet clay mixture called luting.
112
; The thermal distillation or coking of coal to separate volatile and
nonvolatile components takes place in coke ovens that are grouped in
batteries. A battery consists of 20 to 100 adjacent ovens with common side
walls which contain integral flues/ Coke oven heating systems fall into two
general classes: underjet and gun-flue. In the underjet heating system,
the flue gas is introduced into each flue from piping in the basement of the
battery. The gas flow to each flue can be metered and controlled. The
gun-flue heating.system introduces the gas through a horizontal gas duct
extending the length of each wall slightly below the floorline of each oven.
Short ducts lead upward to a nozzle brick at the bottom of each of the
•119
vertical flues in an oven.
Heat for the coking operation is provided by a regenerative combustion
system located below the ovens. Because the combustion flue gas contains a
significant amount of process heat, two heat regenerators are used for
recovery. These regenerators are located below each oven, one for
combustion air and one for the combustion waste gas. The flow is alternated
between the two at about 30 minute intervals. The slot ovens operate like
chemical retorts in that they are both batch operated, fitted with exhaust
flues (standpipes), and function without the addition of any reagent.
The operation of each oven in the battery is cyclic, but the batteries
usually contain a sufficiently large number of ovens (an average of 57) so
that the yield of by-products is essentially continuous. The individual
ovens are charged and discharged at approximately equal time intervals
189
-------
during the coking cycle. The resultant constant flow of evolved gases from
all the ovens'in a battery helps to maintain a balance of pressure in the
flues, collecting mainland stack. All of the ovens are fired continuously
at a constant rate, irrespective of a particular oven's stage in the coking
cycle. If damage to the refractory occurs in inaccessible locations through
overheating or expansion of coal, repairs may be extremely difficult. A
cooldown takes from 5 to 7 weeks, so a battery shutdown is undertaken only
119 '
as a last alternative.
After, the ovens are filled, coking proceeds for 15 to 18 hours to
produce blast furnace coke and 25 to 30 hours to produce foundry coke. The
coking time is determined by the coal mixture, moisture content of the coal,
rate of underfiring, and the desired properties of the coke. The coking
temperatures generally range from 900 to 1100°C (1652 to 2012°F) and are
kept on the high side" of the range to produce blast furnace coke. Air is
prevented from leaking into the ovens by maintaining a positive back
pressure of about 10 mm (0.4 in) water. The gases and hydrocarbons that are
evolved during thermal distillation are removed through the offtake main and
sent to the by-product plant for recovery.
At the end of the coking cycle, doors at both ends of the oven are
removed and the incandescent coke is pushed out the coke side of the oven by
a ram which is extended from the pusher machine. The coke is pushed through
a coke guide into a special railroad car, called a quench car, which
traverses the coke side of the battery. The quench car carries the coke to
the end of the battery to a quench.tower where it is deluged with water so
that it will not continue to burn after being exposed to air. The quenched
coke is discharged onto an inclined coke wharf to allow excess water to
drain and cool the coke to a reasonable handling temperature.112
Gates along the lower edge of the wharf control the rate of coke
falling on a conveyor belt which carries it to the crushing and screening
system. The coke is then crushed and screened to obtain the optimum size
190
-------
for the particular blast furnace operation in which it is to be used. The
undersize coke generated by the crushing and screening operations is used in
other steel plant processes, stockpiled, or sold.
112
Gases evolved during coking leave the coke oven through the standpipes,
pass into goosenecks, and travel through a damper valve to the gas
collection main which directs them to the by-product plant. These gases
account for 20 to 35 percent by weight of the initial coal charge and are '
composed of water vapor, tar, light oils, heavy hydrocarbons, and other
chemical compounds. '^3
\
The raw coke oven gas exits at estimated temperatures of 760 to 870°C
(1400 to 1598°F) and is shock cooled by spraying recycled flushing liquor in
the gooseneck. This spray cools the gas to 80 to 100°C (176 to 212°F),
precipitates tar, condenses various vapors, and serves as the carrying
medium for the condensed compounds. These products are separated from the
liquor in a decanter and are subsequently processed to yield tar and tar
derivatives.11Z'113
The gas is then passed either to a final tar extractor or an
electrostatic precipitator for additional tar removal. When the gas leaves
the tar extractor, it carries 75 percent of the ammonia and 95 percent of
the light oil originally present when leaving the oven.
The ammonia is recovered either as an aqueous solution by water
. absorption or as ammonium sulfate salt. Ammonium sulfate is crystallized in
a saturator which contains a solution of 5 to 10 percent suifuric acid and
is removed by an air injector or centrifugal pump. The salt is dried in a
centrifuge and packaged.
The gas leaving the saturator at about 60°C (140°F) is taken to final
coolers or condensers, where it is typically cooled with water to
approximately 24°C (75°F), During this cooling, some naphthalene separates
and is carried along with the wastewater and recovered. The remaining gas
191
-------
is passed into a light oil or benzol scrubber, over which is circulated a
heavy petroleum fraction called wash oil or a. coal-tar oil which serves as
the absorbent medium. The oil is sprayed in the top of the packed
absorption tower while the gas flows up through the tower- The wash oil
absorbs about 2 to 3 percent of its weight of light oil, with a removal
efficiency of about 95 percent of .the light oil vapor in the gas. The rich
wash oil is passed to a countercurrent steam stripping column. The steam
and light oil vapors pass upward from the still through a heat exchanger to'
a condenser and water separator. The light oil may be sold as crude or
processed to recover benzene, toluene, xylene, and solvent naphtha.112'113
After tar, ammonia, and light oil removal, the gas undergoes a final
desulfurization process at some coke plants before being used as fuel. The
coke oven gas has a rather high heating value, on the order of 20 MJ/Nm3
(550 Btu/stdft ). Typically, 35 to 40 percent of the gas is returned to
fuel the coke oven combustion system, -and the remainder is used for other
119
plant heating needs. "*'
During by-product recovery plant coking, POM emissions are most likely
to occur from coal charging operations, oven door leaks, topside leaks, coke
pushing operations, coke quenching operations, and battery stacks. The
control of emissions from these sources is generally achieved by using one
or a combination of the following control alternatives.112'114
containment of emissions in the process
.capture techniques (e.g., hoods and enclosures)
add-on control devices
process changes
The applicability of these alternatives to each of the POM-emitting coking
operations is described in Table 48. 114 The effectiveness of the control
measures in Table 48 is likely to vary from plant to plant due to
differences in plant configurations and in the types of POM present in
emissions. Data provided in Reference 115 clearly shows that not all POM
192
-------
TABLE 48. EMISSION CONTROLS USED ON POM EMISSION-
SOURCES IN BY-PRODUCT COKE PLANTS114
POM Emission
Source
Coal Charging
Door Leaks
Type of
Control Needed
Emission Control Techniques
Containment
Containment/Capture
and Control Devices
Topside Leaks
Coke Pushing
Coke Quenching
Containment
Containment/Capture
and Control Devices
Process Changes
Battery Stacks
Containment and/or
Control Devices
Use of staged charging and
aspiration to draw emissions
into the coke battery.
Use and maintenance of doors
designed to close and seal
tightly. Use of collection
hoods on individual doors. Use
of wet scrubber and wet ESP
control devices on collected
emissions.
Application and maintenance of
sealing compounds to leaking
points.
Use of "enclosures over the coke
. side of the battery. Use of
wet scrubber and wet ESP control
devices on collected emissions.
Use of single or multiple
baffles in the quench tower and
use of only clean water for
quenching. Dry quenching is
another option; however, it
would require additional capture
and control devices.
Patching cracks in oven walls as
needed and treat exhaust gases
in wet scrubbers, ESPs, or
baghouses.
17 2
of 17.2-92.4 percent).
WSt ESP t0 contro1 oven door emissions, the
P°M emissions an avera§e of 69 percent (range
the naphthalene component of total POM is
is 95.6 percent (range
193
-------
compounds in an emissions stream are controlled to the same extent (see
footnote a in Table 48) . Examples of the type of POM compounds measured in
coking emissions are given in Tables 49, 50, and si^11^-!!?
In the by-products recovery section of a coking plant, the tar
processing operation, the ammonia processing operation, and the final
cooler/naphthalene handling operation have been identified as potential
113
sources of POM emissions. In tar processing, tar decanting, tar
dewatering and storage, and tar distillation operations are potential POM
emission sources. Emissions are fugitive in nature and are generally *
released directly to the atmosphere.
Excess ammonia liquor treatment has been determined to be a source' of
POM emissions during ammonia processing. Specifically, steam stripping
of the liquor to recover ammonia has been found to generate POM emissions.
In the final cooling/naphthalene handling operations, the cooling tower for
the contact cooler and froth flotation naphthalene separator are potential
POM emission sources. Coke oven gas is cooled by means of a direct contact
spray tower cooler. After contacting the coke oven gas in the final copier,
the water is pumped to a separation device prior to being sent to a cooling
tower. Froth flotation is used to enhance naphthalene separation. Fugitive
POM emissions are potentially released from the froth flotation process.^"13
After separation of the naphthalene, the contact water is sent to an
atmospheric cooling tower prior to being returned to the spray tower cooler.
The cooling tower operation is a potential source of POM emissions because
organic components dissolved in the recirculating water will be air stripped
and- released into the atmosphere.
Emission Factors
Limited emission factor data exist for total POM or individual POM
species emissions from coking and by-product recovery processes. Available
POM data for slot oven coking sources and for by-product recovery sources
194
-------
TABLE 49. SPECIFIC POM COMPOUNDS DETECTED IN
OVEN DOOR LEAK EMISSIONS114"117
POMs in Controlled
Door Leak Emissions
POMs in Uncontrolled
Door Leak Emissions
Naphthalene
Fluoranthene
Pyrene
Benz(c)phenanthrene
Chrysene
Benz (a)anthracene
7,12-Dimethylbenz(a)anthracene
Benzofluoranthenes
Benzo(a)pyrene
Benzo(e)pyrene
Cholanthrene
Indeno(1,2,3-c,d)pyrene
Dibenz (a,h)anthracene
Dibenzacridines
Dibenz(c,g)carbazole
Dibenzpyrenes
3-Methyl cholanthrene
Benzo(a)phenanthrene
Benzo(e)pyrene
Benzofluoranthenes
Benzo(k)fluoranthene
Ghrysene
D ibenzanthracenes
Dibenzpyrene
Dimethylbenz(a)anthracene
Fluoranthene
Indeno(1,2,3-c,d)pyrene
Naphthalene
Pyrene
Benzo(a)pyrene
195
-------
TABLE 50. POM COMPOUNDS DETECTED IN BATTERY
TOPSIDE EMISSIONS114"117
Phenanthrene
Anthracene
Methylphenanthrene/methylanthracene
Fluoranthene
Dihydrobenzo(a,b)fluorene
Pyrene
Benzo(a)fluorene
Benzo(b)fluorene
Benzo(c)phenanthrene
Benz(a)anthracene
Chrysene/triphenylene
Benzo(b,j , k)fluoranthene
Benzo(e)pyrene
Benzo(a)pyrene
Perylene
o-Phenylenepyrene
Benzo(g,h,i)perylene
Anthanthrene
Coronene
Dibenzopyrene
196
-------
TABLE 51. SPECIFIC POM COMPOUNDS DETECTED IN
,114-117
QUENCH TOWER EMISSIONS'
Anthracene
Methyl anthracenes
Fluoranthene
Pyrene
Methyl pyrene and fluoranthene
Benz o ( c )phenanthrene
Chrysene and benz(a)anthracene
Methyl chrysenes
DimethyIbenz(a)anthracene
Benzo(a)pyrene
3-Methyl cholanthrene
7,12-DimethyIbenz(a)anthracene
Dibenz(a,h)anthracene
Dibenzo(a,h)pyrene
Dibenzo(a, i)pyrene
Benz(a)anthracenes
Pyridine
Indeno(1,2,3-c,d)pyrene
Phenanthrene
Phenol
Cresol
Quinoline
197
-------
are given in Tables 52 and 53, respectively.112'114'116'118"120 The lack of
substantive POM emissions data for coking sources prohibits extensive
characterization and comparison; however, as indicated in Table 52, POM
emissions from coking processes are highly variable because coking
conditions can vary widely from plant to plant and within the same plant
from process to process. Coal composition and moisture content vary widely
and these process variables can have a significant bearing on emissions.
Coking times and temperatures can also be varied so as to have marked
impacts on potential POM emissions. The fugitive nature of the majority of
coking process POM emissions complicates emissions control and increases the
potential for widely varying emission estimates. For coke oven sources,
specific source emissions testing should be utilized if possible to
characterize POM emissions.
Although data for POM emissions from coke by-product recovery processes
are also few, it is anticipated that emissions would be as equally variable
as in coking sources. Emissions 'of POM from the by-product plant are in
large part a function of the raw coke gas entering the recovery process. As
the composition of this stream changes with the coking process, so would
potential'POM emissions from by-product recovery.
Source Locations
As of the end of 1984, 43 by-product coke plants with 143 coking
batteries were in existence in the United States. Of these 43, 36 were in
operation. Table 54 lists the existing installations, their
classification as merchant or furnace plants,- and the major uses of their
coke. In terms of production, almost 60 percent of the coke produced in the
United States occurs in Indiana, Pennsylvania,' and Ohio. In 1983, Indiana
-was the leading coke producing State with 23 percent of the national
total.112
198
-------
NG SOURCES
>->
x
B
o
EH
O
CO
ai •
£
ee.
O
E-
C
^f*
f^*m
P
OT
CO
z
g
in
w
•^
H
i
i
; ; ?J
, c
01
01
of
Continent s
Benzofa )pyrene
Emission Factor
|
Total POM
Emission Factor
e
4)
Emission Sourc
tA LTi CO (N
^ *>H »H vN
"P. a.
SB 09
g »
oi jZ 01
» B »
a a
a
-4 01 B "S
O -« _l
b " • — ^
« O O O
B b b b
U B B B
E O O 0
a u u a
. ^ ^1 ^
i j -^J 5
» t :: s
'• - as " § "
*» • OO O *J
-* »4 _ O
g-s *
0
J9 *O Ud
a a" iM1*^ OM
J& 00 X OO
Z 00 b -^ b
"oo "oo *-j1 MJ
B B 00 U r- u
o
^- -g -3 « is
X b VN t-4 o JZ
•^ a --i .». . •§
oojs a e o
« > > —
"5 a a o a
O ^rf 4J Q
§B *J 4J u
O O o eg
• " Z Z S IM
e o o
d
*•* ..4
•0 .-D
IM 01 01
O CO 00
b b
oo a a
xc js js 01
^^. *O U u »ri
00 01 JS
oo ** i-t a
w> b a a —
o a o o —
o J= o o a
2§ I I I
i ~« "»
O 00 2
• *
••4
oe oo oo a
B s a jt
•H ..4 .^ y
| -8-83
S g S w
£ S- § &
V
.S -j? jS £
5 .355
o> c> o S
g-
JS JS JS
*J CO M 4J
•rt b -H .^
o
•o-o-o -o w
• 01 oi a oi a o
a. 'a n."a £"3 a
•H 01 -H 01 -*4 0) 01
=_ co SB SB b
01-0 g" b S" b »
si? §••§ si s
6" S-o S, b
•*<* ft) 01 3
JS g J-g j-g g,
S.3 8^ §5 „.!
BB B 00 E 00 IM .tn
BOI BB aS-fdu
bB bb bb OIQ]
o; o o> a. oi a. S.-S
UU VB t^B CO*H
»M^ .^ •«-) U4 "u
) o -o uj -i IM^B o ^
oi o 5 o « t>
00 U U U 00 01
,x _s ^s eo 3 -S; —
_b"oob"eo2 38
j» a. a. a, m
°. j? ^^ S5 * -g.
o-g e-g "J-g -Tf
u u e u g
f> oi — i iM">a
ja S *o "S ° 5
a a u oo u
•-; -< oo s ~» oi
'2 '2 s "0 M"*
a a »«. o 3 -4 .
> > 00 b o
a a a, o u
•" « —i 01 CS 01
o o . ^ -_
z z e o tn a.
u ts s
a
B
a
B
O
B
» « B 3
1 ! 1 1
J; Jj b oi
o ° •*
a O ci t; a
B
O
CB
a
o
4>
a
» .2
V
§ 2
JS 01
i, and benzof luorani
ra and 23 percent t
rene.
» o >,
B • IM &
01
b 01 -Q
>. 4J B
a. ^a a
g" 1 a
01 JJ 01
JS 8 JS
4j a. t;
e naphthalene, fluor
s, 77 percent were i
e naphthalene, fluor
S = b
OJ O ft)
* 'S *
•g .2 "g
S § S
» CO u
a ax a
§121
(0 -D a
^ V fl* OJ
2 i 1 i
h "? jj to
I X I X
« s 1 2
.1-31
00 *H 4J .^
a u o u
S .2 " .5
> b IM b
< o. o eu
J3 U -0 ' 0
g
01
b .
a" *
N o7 O
a B N
01 oi a
•a o ja
•-S 2S
JS W
B B 00
g A u
01 «J — '
U +* N
a us
b B u
js ai £
B 5 -n
lene, chrysene, .dibenza
ithracene, 7,12-dinethy
ene, dibenzacridines, i
f B U
• s • i!
Eb 01 xS S
OB MS
£ _3 01 B^
Q) Ji &
« "-^ •D <.Tw
*-< O * C N
9 N U 0) B
US) fc- S
•*4 o *n x jo
jj ^ • jj .^
I- W B -0
S, a'-.' S -
2 percent were in
, benzof luoranthene
pyrene, and indeno(
hrysene, benz(c)phe
no(l,2,3-c,d)pyrene
•O 01 *~* u 01
a 01 *2 • B
bo 01 *w
a >, N a
C O. S 0) •
o '-» aj ui 01
(M at £ X B
-»> 0.0)
no* ^
3 S 01 -^
O B B 01 i3
01 01 01 B B
a JS — 01 a
a a js — i
00 -JS u 0
98 percent were in
benzo(a )phenanthrene
thene, pyrene, napht
naphth.alene, fluoran
, benzo(e)pyrene, ch
threne.
• E 01 E
g ^ a _g E js
_O ^3 O 3 b "o
CO B IM B Si °
§ x oT x "e) j?
_ p B, p H u
X E 01 S B 01
w U 01 S
S ~ a — ja f
Q b Q M
oi o •£ o a* xi
- " S " B B
« — *" ™
Of total contrt
Constituents of
dimethylbenz(a]
^Constituents of
benzof 1 uoranthe
dibenzpyrenes,
t.
X
2
• to
QJ 9
V S
b a -
•C CB 01
t; MB
1 T. 2
i § £
JS ~.
ja. u js
u a a
"o 's'o'
» UN
B -HE
pyrene/fluoranthene, bi
ch results are for parl
nz(a,h)anthracene, dibi
luinoline.
01 b '• «
S 3 01 —
•o s o
• 01 oi B
01 U U 01
B o a b
01 b b U
b a. js
a "o
• ,-O *•* E
oi • o a 01
B 01 J3>3 js
01 S UNO.
iracenes, fluoranth
!ne/benz(a )aothrace
ising a modified He
!, 7,12-dinethylben
:ne, phenanthrene, ]
u a B b
B >, -0 01 X
a b 01 b a>
t-l s^^
01 u JSO
b rt m jj «
inthracene, phenanthi
limethylbenz(a)anthn
-green coke. Samplei
>enzo(a)pyrene, 3-nel
tnes, pyridine, indei
IB only.
B S S
•8 « i 4 2 'S
5 g , » - .2
Jf b BUB B
B X a B B U
»•* a. >•< ^
*"* B a as
X a oJZs.Jp
p <— oi o N £
S o b a« s
•S oo oi oi
•ME ** J3 .u
a oi b a a
™ ^ O *J • ^i
0 IM 0 01 3
«J • *4 E u
B a 01 .M
Constituents of
methyl chrysene
Average of test
Constituents of
dibenzo(a,i)pyr
Represents part
fl,
Z E
~ B
= a
n b
S 0
Jc
01 J3
a. o
N
01 B
_
• «
rerage of the four runs
ae, benzo(k)fluorartther
ts the ai
anthracel
E —»
01 B
B •**
01 N
S. ?,
Bt c)
C .£•
U
O> 0)
3 S
_^ Ml
:est runs. This va
fluoranthene, pyr<
01
01 S
a b
b j:
||
a B
01
S-g,
o
CJ
.S E
U >p4
«
(U 00
B
e.
=• I
•M (a
^ O
0! J-
N -J
13 C *
J4
CJ *O
9 V
•a u
B 4_l
9 CJ
§.'
§ ^"
U 4J 01
B E
Z ai ill
Seventy-five PO]
sppcieu consisti
and benzo(a)pyri
199
-------
en
""m
CO
M
r^
CJ
§
fLt
^•4
Cx£
W
^
S
H
OS
§•*
1
f*
g
S
erf
g
1
04
O
6
tS
r*4
S5
O
CO
CO
S
w
,
^3
S
en
1
B
00
•H •
COCO
CO 00
•^ s
B ^*»
• & 00
cu «.
B H
•H O
l— 1 4J
O U
B CO
3*.
ey
B
00
•H •>
COCO
m oo
•!•( J2J
0 **«N»
P3 00
rH ~
B 0
CU 4J
43 O
• r4 pC4
B
00
•H »
COCO
CO OO
S 00
•o tT
PM o
4J
r-4 CJ
CO CO
| i JVj
c^
sion Source
CO
•H
S
S
o
PH
8
vO O
O O
• e
0 0
es
tn, o
o o
0 C3
*
en
U O
-H O
•* 0
cu
00
CO
1-1
o
4J
CO
•e
B
CO
od oo
B B
•i-l t-l -p4
CO Q) t-i
CO AJ CU
cu a w
0 g §
I-l CU CU
CM « Q
I 5 £
CU
3
CO
1— 1
•H
co es
CO *"^
0
4-i
^m
CU
f— 1
CO
•H \n
CO O
^ *
CO O
JJ
O
e3 st
O vO
od
.
CO
B
cu
s
8
43
o
JJ
•o
cu
M-l
CO
8
UH
o
00
33
00
CO
g
cu
expressed in
tors are
ssions.
U •!-)
i2 §
CO
1
,co •
0 0
CU Pit
CO
CO f<
00 CO
•o o
B 4-1
CO
cu "o
4J
HI J_l
J B
3 0)
U S
•t-1 JJ
4J °H
h 44
CO CO
Cu B
44 8
tors represen
Is the major i
u
ssion fai
hthalene
•i-l O.
& co
ca a
o u
-------
g;
3
H
5
§
o
§5
3
o
o
CO
CO
W CM
aj "*
w s
^^ ^1
o
U CO
w u
as H
3 CO
§ «
O W
Ol M
1 SS
LOCATIONS OF BY
PLANTS IN THE U
sf
w
H
CO
0)
o
cj
o
CO
CJ
CO
o
•*"")
CO
s
6
0
•r-l
JJ U
s §
•H i-4
M-l O-i
•H
CO m
CO O
CO
i-(
°
C
O
•H
CO
U
o
B
CO
.—I
PU
0)
SS
I
0
furnace
furnace
CO CO
CO CO
M 14-1
CO CO
cO eg
« «
cu Q;
o y
CO CO
C S
9 3-
fa fa
en
s™^
K
HH
J^j 60
t-i jj
CO CO
1
w
1— 1
i-i cu
« cu
CU JJ
W -a
v B
60 CO
9 i-i
furnace
CO
m
«
•nace
fa
•t
00
eO
y
•H
•
A
0)
CO
^,
cu
^ !
^
gggggggg
SSSS'S'B'SB'
ss s s s ? _ _
<4-l
-------
'
g;
O
i-l
f-t
&
:=>
0 *•*
O 13
B5 0)
B
S I!
8B
O
u
CO v-»
W CN
0 -1
Pi
1
H
(0
1
O
CO
CD
CO
&
i-i
O
s
B
O
••-I
JJ JJ
CO B
O CO
•H i-l
CO U.1
CO O
CO
1— 1
e>
1
B
O
•H
JJ
CO
U
o
B
r-i
CD
i
co
o.
u
^>
•o
§
o
CD CU CD CD CD CU CD CU CU CD
U U CJ U CJ U O WO U
CO CO CQ CO CO CO CO cOcO CO
B S S S B S S SB B
co concocococa coco co
co cc) co cd cd cd cd coca co
CU Q) d) (1) a) 0) 0) CU CU B
u o o o o o o o o id
'cfl eg co eg efl cfl eo coco 43
B BSBBSB BB u
y M M M M M M M *•< *•<
S 333333 S3 CU
^
^E
*J»
^^
<-» ** W
O 1 1 i^ t> ^c
^C] ^Q i-H C ^C;
A PM *H QJ pl| ft
3 J""0 OH S •> CO
g jj-i-iai B*> coco c
JJ «i3 t3 Ti s^'cd S jj S 'a
(j cO>rl-HWhO COB ' E
(2 O [2 (2 c5 ^ PM WE ta
B
O
JJ nj
CO B 0)
o -S jj e
2* rt js °H
O t M CO
° 2 n h
P4 30
cu (4 ja o,
O CJ 4J O
0 H CJ
B cu PL, o h
O CU 1 •*•) CO
09 W ^ ^ **
o «H }j eo
w * i-H O S
co cu S,
>ji -l >» >i M
CO "O *O CO
3 B B 3
•a 3 3 13
^B 0 0 B
02 CD CU OJ
4S 0 O j-
•w cd co jj
0 B B 0
•* 3 3 *
jrjr1" . *" s* &1
13 13 JJ JJ "O 13
a 3 co co s' s
O O i-l i-l O O
CO cd cd S MM
•S-S-S -g -g-g
I^ M U V< MM
^^S S S^
'
^.
P-t
*»
^2
H^ cd
-------
|
1
S
CO
CO
u
0
CM
g
^^
8
CA
6-1
W
C3
Q
o-
PN
I
CQ
CO
LOCATIO
o
In
1
•e
I
jj
o
CN
i-H
fH
•d-
00
ON
i-H
55
M
"CO
H
^H
H
CO
Q
EH
M
^
U
g
PLANTS
CO
0)
cS
MH
O
CO
CU
CO
(4
o
assif icatio
of Plant
i-H
CJ
,
B
•,H
[ i
CO
y
o
jj
B
CO
i-H
. eu
01
S3
' 1
|
b*. &*.
1 1
£ £
Merchant
Merchant
jj <<
o
4J i-H
Ol O
0 W
B
O
JJ
CO
1 *
OM CO
*& (U
O J£
w o
4J
•H
cd
S
CO
TJ
01
is conside
01
CO
3
•0
S
B
0
"B
o
•o*
Ol
JJ
CO
u
•H
*o
•iH
CO
1
1
0
B
M
•
fl
O
•H
JJ
y
O
i-H
JJ
CO
42
4J
JJ
CO
CO
JJ
fl.
CO
r— 1
cu
o
CU
43
e
cu
cu
4J •
CO fl
y o
•O 4J
S CO
•1-4 y
o
CO i-l
parenthese
s at that
jj
B co
•iH -H
CO CU
Ol JJ
43 B
•
jj
B
CO
i-H
CU
CO
•H
f-H
^v
CO
B
B
CU
(X,
£
Cu
•H
cr
•iH
i-H
CU
43
4J
JJ
CO
i-H
CU
01
JJ
CQ
MH
O
B
O
•iH
JJ
y •
3 >>
*o *™^
O i— <
M CO
Cw 9
JJ
Ol fl
3 >
T3 Ol
0)
I-l CO
4J CO
jj
in co
co
ON Ol
»-l i— 1
•o
p§ T>
i-H
B 0
•IH y
§ -
••4 O
CU JJ
B cu
•H 1
CO O
•H U
•o >.
cu co
g s
9 JJ
O B
e co
S-SL
SI
43
JJ
•H
H
Ol
00
Ol
CO
JJ
•H
3
ON
i— 1
i-l
•H
B
announced i
.
i-H
cu
01
4J
co •
CO
•JJ
CO
cu
B
o
^]
MH
o
ompany
y
B
CU
)H
a
cu
JJ
y
M
ndustries,
stries.
Northwest I
Farley Indu
01 CO
CO Ol
>> fl
CO JJ -iH
4J 60 -iH rt
C B i-* B
CO 'fH >r4 CU
i-H JJ y JJ
O.r-4 CO fl)
3 UH "O
•> co
CO C C CU
01 O 0) 43
60 y >
B -5-0
CO >, 60 rH
43 43 3
" >. 0
„ CO B 43
Cu CU CO CO
•pJ. »^
CO -H O C
iri r! *•" w
CU -H UH
B y
§ CO CO CO
UH B 0)
0 M
5*. >H -H 9
4J CO CO CO
•H i-H CO CO
•-H 3 -iH CU
•i-i y g a
y -H cu
« JJ i-H
UH l-i S O
-S.2i3
0) B
60 MH MH O
B 0 0 y
CO
43 CU I-H -O
y y cu B
B > CO
CO CU Ol
O JJ i-H »
O CO JJ
•rH -fH CU 3
JJ X 43 O,
•H CU EH 43
•O 00
fl CU 3
O 43 • O
y JJ co IH •
01 43 i-H
JJ >, > JJ 0)
4•> O
CO CU S JJ CO
0 > S -H Ul
43 y cu
CO •« JJ cfl Cu
co I-H a.
9 to cfl JJ
0) O JJ O C
0043 B CO
B CO CO CO fH
CO i-l to CU
43 l-i d.
« «U 43 43
^ cu y jj
O CO 43 3 -in
JJ CU 4J CO >
jj IH ca co
y cu o eu 4J
01 43 — I-H y
•«-» H -O 43 cfl
43 B CO JJ
3 cfl -H B
CO • IH o
y ca co y
03 JJ 60 >
•H CU fl JJ
•H UH y
60 •> JJ O CU
C fl CO IH
•H S -H B -H
JJ O r-4 O "O
CO 'O >f4
•ft JJ JJ 43
fH T3 B y 60
0> CU B 9
CO co IH 3 O
•H O l-i MH M
43 fH 3 js
H y y co jj
o
ss
203
-------
-------
ASPHALT ROOFING MANUFACTURING
Process Description
Certain processing steps used in the production of asphalt roofing
materials have been shown to be potential sources of POM emissions.
Emissions of POM compounds result primarily from the heating and air blowing
of asphalt mixtures, from asphalt saturation processes, and from evaporation
losses during asphalt material storage and handling.
The asphaltic material used to make roofing grades of asphalt, known
as saturant and coating asphalt, is a product of the fractional distillation
of crude oil. This material is obtained toward the end of the distilling
process and is commonly known as asphalt flux. Asphalt flux is sometimes
blown by the oil refiner or asphalt processor to meet the roofing
manufacturer's specifications. Many roofing manufacturers, however,
purchase the flux and carry out their own blowing.121"123
Handling and storage activities associated with the asphalt flux raw '
material are potential sources of organic emissions, including POM. Asphalt
is normally delivered to the asphalt roofing plant in bulk by pipeline,.
tanker truck, or railcar. Bulk asphalts are delivered in liquid form at
temperatures of 93 to 204°C (200 to 400°F), depending on the type of asphalt
and local practice. ' With bulk liquid asphalt, the most common method
of unloading is to couple a flexible pipe to the tanker and pump the asphalt
directly into the appropriate storage tanks. The tanker cover is partially
open during the transfer. Since this is a closed system, the only potential
sources of emissions are the tanker and the storage tanks. ,The magnitude of
the emissions from the tanker is at least partially dependent on how far the
cover is opened.
Another unloading procedure, of which there are numerous variations, is
to pump the hot asphalt into a large open funnel which is connected to a
surge tank. From the surge tanks, the asphalt is pumped directly into
204
-------
storage tanks. Emission sources under the surge tank configuration are the
tanker, the interface between the tanker and the surge tank, the surge tank,
and the storage tanks. The emissions from these sources are primarily
organic particulate. The quantity of emissions depends on the asphalt
temperature and on the asphalt characteristics.
Asphalt flux is usually stored at 51 to 79°C (124 to 174°F), although
storage temperatures of up to 232°C (450°F) have been noted.
The
temperature is usually maintained with steam coils in the tanks at the lower
temperatures. Oil- or gas-fired preheaters are used to maintain the asphalt
flux at temperatures above 93°C (200°F).121~123
Asphalt is transferred within a roofing plant by closed pipeline.
Barring leaks, the only potential emission sources are the end-points.
These end-points are the storage tanks, the asphalt heaters (if not the
closed tube type), and the air blowing stills.
Saturant and coating asphalts used to manufacture roofing materials are
prepared by blowing air through tanks of hot asphalt flux. Saturant and
coating asphalts are primarily distinguished by the differences in their
softening points. The softening point of saturant asphalts is between 40
and 74 C (104 to 165°F), while coating asphalts soften at about 110°C
(230 F). The configuration of a typical air blowing operation is shown in
Figure 22. This operation consists primarily of a blowing still which is
a tank fitted near its base with a sparger. The purpose of the sparger is-
to increase contact between the blowing air and the asphalt. Air is forced
through holes in the sparger into a tank of hot [204 to 243°C (400 to
470 F)] asphalt flux. The air rises through the asphalt and initiates an
exothermic oxidation reaction. Oxidizing the asphalt has the effect of
raising its softening temperature, reducing penetration, and modifying other
characteristics. Catalysts are sometimes added to the asphalt flux during
air blowing to better facilitate these transformations. The time required
for air blowing of asphalt depends on a number of factors including the
characteristics of the asphalt flux, the characteristics desired for the
205
-------
c
o
•H
0)
a
o
on
•H
§
M
•H
CO
CO
a
CO
co
CO
u
CM
CN
X in
a. «M
co «-
206
-------
finished product, the reaction temperature, the type of still used, the air
injection rate, and the efficiency with which the air entering the still is
dispersed throughout the asphalt. Blowing times may vary in duration from
30 minutes to 12 hours.121
Asphalt blowing is a highly temperature-dependent process, as the rate
of oxidation increases rapidly with-increases in temperature. Asphalt is
preheated to 204 to 243°C (400 to 470°F) before blowing is initiated to
assure that the oxidation process will start an an acceptable rate.
Conversion does take place at lower temperatures but is much slower. Due to
the exothermic nature of the reaction, 'the asphalt temperature rises as
blowing proceeds. This, in turn, further increases the reaction rate.
Asphalt temperature is normally kept at about 260°C (500°F) during blowing
by spraying water onto the surface of the asphalt, although external cooling
may also be used to remove the heat of reaction. The allowable upper limit
to the reaction temperature is dictated by safety considerations, with the
kepi
121
maximum temperature of the asphalt usually kept at least 28°C (5.0°F) below
the flash point of the asphalt being blown.
The design and location of the sparger in the blowing governs how much
of the asphalt surface area is physically contacted by the injected air, and
the vertical height of the still determines the time span of this contact.
Vertical stills, because of their greater head (asphalt height) require less
air flow for the same amount of asphalt-air contact. Both vertical and
horizontal stills (see Figures 23 and 24) are used for asphalt blowing, but
where new design is involved, a vertical type is preferred by the industry
because of the increased asphalt-air contact and consequent reduction in
blowing times. Asphalt losses from vertical stills are also reported to
be less than those from horizontal stills. All recent blowing still
installations have been of the vertical type. Asphalt blowing can be either
a batch process or a continuous operation; however, the majority of
facilities use a batch process.
207
-------
KNOCK OUT BOX I
OR •
WATER VALVE
WATE
STILL
AIR COMPRESSOR
Figure 23. Typical configuration of a vertical asphalt
blowing still.121
air
208.
-------
VALVE 0-0
WATER Q
AIR COMPRESSOR
Figure 24. Typical configuration of a horizontal asphalt air
blowing still.12!
209
-------
Blown asphalt (saturant and coating asphalt) is used to produce asphalt
roofing and siding products according to the process depicted in
Figure 25. A roll of felt is installed on the felt reel and unwound onto
a dry floating looper. The dry floating looper provides'a reservoir of felt
material to match the intermittent operation of the felt roller to the
continuous operation of the line. Felt is unwound from the roll at a faster
rate than is required by the line, with the excess being stored in the dry
looper. The flow of felt to the line is kept constant by raising the top
set of rollers and increasing looper capacity. The opposite action occurs
when a new roll is being put on the felt reel and spliced in, and the felt
supply ceases temporarily. There are no POM emissions generated in this
processing step.
Following the dry looper, the felt enters the saturator where moisture
is driven out and the felt fibers and intervening spaces are filled with
saturant asphalt. (If a fiberglass mat web is used instead of felt, the
saturation step and the subsequent drying-in process are bypassed.) The
saturator also contains a looper arrangement which is almost totally
submerged in a tank of asphalt maintained at a temperature of 232 to 260°C
(450 to 500°F).. 'The absorbed asphalt increases the sheet or web weight by
about 150 percent. At some plants, the felt is sprayed on one side with
asphalt to drive out the moisture prior to dipping. This approach
reportedly results in higher POM emissions than does use of the dip process
alone. The saturator is a significant POM emissions source within the
asphalt roofing process.
The saturated felt then passes through drying-in drums and onto the wet
looper, sometimes called the hot looper. The drying-in drums press surface
saturant into the felt. Depending on the required final product, additional
saturant may also be added at this point. The amount of absorption depends
on the viscosity of the asphalt and the length of time the asphalt remains
fluid. The wet looper increases absorption by providing time for the
saturant asphalt to penetrate the felt. The wet looper operation has been
210
-------
— • .at c
C C i;
jr n
g . •< o I'M
» g 5 =r "3
c o re ts **•
S W Q y ^ T3
1 g
K
cn
£b -£
||_ i
0 £0
! JT
o "32
01 U I.
-. a j)
•± c s u
<5i = « g-3
1 .si
1 S3 <
X 1^ u
5 " s <
C >• m"«
f 1
\
— — u
— X ' "M. ti
n— %, * J
[-*•
[
1
4J
1
K SI
*H «
C C
*
1
m
*-< C ^N
oe
c
"3 U
M o- a
£ = £
E to <
t
01
15
cS
it
4J O
01 0
•
"*•
T
o
b
ta *
CO ^
t
iJ 0
a a.
>. 0 0
i* •* O
O El. -1
1
1
b
•O B
c -o u
— U i -o S
^
»•
3
a
1
i
td
o
01
T T 11
.,.,,_ ^ ^
§ 00 S *J
1 -2" £ n 5 5
"a if. c"c
« __ -i «
TV w
R 1
^M L
•2 e
ee, u
CO
a
o
•H
ca
CO
*H
1
1
j-l
CO
4-1
B
CD
O
P*
CO
0) •
u ca
CO 0)
•H id
•a s
e o
M W
i
01
*« u
£ 3
m u
- S
•§!
f 1
b
£ 3,
OC (U
11
us a.
tn
V Oi
ffB
22
v.
-------
shown to be a significant source of organic particulate emissions within the
asphalt roofing process. Although the wet looper is a potential source of
POM emissions, their severity has not been defined.121'122
If saturated felt is being produced, the sheet bypasses the next two
steps (coating and surfacing) and passes directly to the cool-down section.
For surfaced roofing products, however, the saturated felt is carried to
the coater station where a stabilized asphalt coating is applied to both the
top and bottom surfaces. Stabilized coating contains a mineral stabilizer
and a harder, more viscous coating asphalt which has a higher softening
point than saturant asphalt. The coating asphalt and mineral stabilizer
are mixed in approximately equal proportions. The mineral stabilizer may
consist of finely divided lime, silica, slate dust, dolomite, or other
mineral materials.
The weight of the finished product is controlled by the amount of
coating used. The coater rollers can be moved closer together to reduce the
amount of coating applied to the felt,"or separated to increase it. Many
modern plants are equipped with automatic scales which weigh the sheets in
the process of manufacture and warn the coater operator when the product is
running under or over specifications.
The coater is a significant emissions source within the roofing
production process. It releases asphalt fumes containing organic
particulate, some of which may be POM compounds.121'122
The function of the coater-mixer is to mix coating asphalt and a
mineral stabilizer in approximately equal proportions. The stabilized
asphalt is then piped to the coating pan. The asphalt is piped in at about
232 to 260°C (450 to 500°F). and the mineral stabilizer is delivered by
screw conveyor. There is often a preheater immediately ahead of the
coater-mixer to dry and preheat the material before it is fed into the
coater-mixer. This eliminates moisture problems and also helps to maintain
the temperature above 160°C (320°F) in the coater-mixer. The coater-mixer
212
-------
is usually covered or enclosed, with an exhaust pipe for the air displaced
by (or carried with) the incoming materials. The coater-mixer is viewed as
a potential source of POM emissions, but not a significant one.121'122
The next step in the production of coated roofing products is the
application of mineral surfacing. The surfacing section of the roofing line
usually consists of a multi-compartmented granule hopper, two parting agent
hoppers, and two large press rollers. The hoppers are.fed through flexible
hoses from one or more machine bins above the line. These machine bins
provide temporary storage and are sometimes called surge bins. The granule
hopper drops colored granules from its various compartments onto the top
surface of the moving sheet of coated felt in the sequence necessary to
produce the desired color pattern on the roofing. This step is bypassed for
smooth-surfaced products.121
Parting agents such as talc and sand (or some .combination thereof) are
applied to the top and back surfaces of the coated sheet from parting agent
hoppers. These hoppers are usually of an open-topped, slot-type design,
slightly longer than the coated sheet is wide, with a screw arrangement for
distributing the parting agent uniformly, throughout its length. The first
hopper is positioned between the granule hopper and the first large press
roller, and 0.2 to 0.3 m (8 to 12 in) above the sheet. It drops a generous
amount of parting agent onto the top surface of the coated sheet and
slightly over each edge. Collectors are often placed at the edges of the
sheet to pick up this overspray, which is then recycled to the parting agent
machine bin by open screw conveyor and bucket elevator. The second parting
agent hopper is located between the rollers and dusts the back side of the
coated sheet. Because of the steep angle of the sheet at this point, the
average fall distance from the hopper to the sheet is usually somewhat
greater than on the top side, and more of the material falls off the
sheet.121
In a second technique used to apply backing agent to the back side of a
coated sheet, a hinged trough holds the backing material against the coated
sheet and only material that will adhere to the sheet is picked .up. When
213
-------
the roofing line is not operating, the trough is tipped back so that no
parting agent will escape past its lower lip. . •
Immediately after application of the surfacing material, the sheet
passes through a cool-down section. Here the sheet is cooled rapidly by
passing it around water-cooled rollers in an abbreviated looper arrangement.
Usually, water is also sprayed on the surfaces of the sheet to speed the
cooling process. The cool-down section is not a source of POM emissions.
Following cooling, self-sealing coated sheets usually have an asphalt
seal-down strip applied. The strip is applied by a roller which is
partially submerged in a pan of hot sealant asphalt. The pan is typically
covered to minimize fugitive emissions. No seal-down strip is applied to
standard shingle or roll goods products. Some products are also texturized
at this point by passing the sheet over an embossing roll which forms a
pattern, in the surface of the coated sheet.121
The cooling process for both saturated sheets and coated sheets is
completed in the next processing station known as the finish looper. In the
finish looper, sheets are allowed to cool and dry gradually. Secondly, the
finish looper provides line storage to match the continuous operation of the
line to the intermittent operation of the roll winder. It also allows time
for quick repairs or adjustments to the shingle cutter and stacker during
continuous line operation or, conversely, allows cutting and packaging to
continue when the line is down for repair. Usually, this part of the
process is enclosed to keep the final cooling process from progressing too
rapidly; Sometimes, in cold weather, heated air is also used to retard
cooling. The finish looper is not viewed as a source of POM emissions.121
Following finishing, asphalt sheet destined for use in roll goods is
wound on a mandrel, cut to the proper length, and packaged. When shingles
are being made, the material from the finish looper is fed into the shingle
cutting machine. After the shingles have been cut, they are moved by roller
214
-------
conveyor to manual or automatic packaging equipment. They are then stacked
on pallets and transferred by fork lift to storage areas or waiting
trucks.121
The primary POM emission sources associated with asphalt roofing are
the asphalt air blowing stills (and associated oil knockout boxes) and the
122 124
•felt saturators. ' Additional potential POM emission sources that have
been identified include the wet looper, the coater-mixer, the felt coater,
the seal-down stripper, and air blown asphalt storage tanks. Minor fugitive
emissions are also possible from asphalt flux and blown asphalt handling and
transfer operations .
Process selection and control of process parameters reportedly can be
used to minimize uncontrolled emissions, including POM, from asphalt air
blowing stills ," asphalt saturators, wet loopers, and coaters. Process "
controls include the use of the following:121
dip saturators, rather than spray or spray- dip saturators;
vertical stills, rather than horizontal stills;
asphalts that inherently produce low emissions;
higher flash point asphalts;
reduced temperatures in the asphalt saturant pan;
reduced asphalt storage temperatures; and
lower asphalt blowing temperatures.
Dip saturators have been selected for most new asphalt roofing line
installations in recent years, and this trend is expected to continue.
Recent asphalt blowing still installations have been almost exclusively of
the vertical type because of its higher efficiency and lower emissions.
Vertical stills occupy less space and require no heating during oxidizing
[if the temperature of the incoming flux is above 204°C (400°F)]. Vertical
stills are expected to be used, in new installations equipped with stills and
in most retrofit situations.121
215
-------
Asphalt fluxes with lower flash points and softening points tend to
have higher hydrocarbon emissions because these fluxes generally have been
less severely cracked and contain more low-bailing fractions. Many of these
light ends can be expected to boil off during blowing! Limiting the minimum
softening and flash points of asphalt flux should reduce the amount of
POM-containing fumes generated during blowing since less blowing is required
to produce a-saturant or coating asphalt. Saturant and coating asphalts
with high softening points should reduce POM emissions from felt saturation
and coating operations. However, producing the higher softening asphalt
flux requires more blowing, which increases uncontrolled emissions from the
blowing operation.
Although these process-oriented emission control measures are useful,
emissions capture equipment and add-on emissions control equipment are also
necessary in asphalt roofing material production facilities.- The capture of
potential POM emissions from asphalt blowing stills, asphalt storage tanks,
asphalt tank truck unloading, and" the coater-mixer can and is being achieved
in the industry by the use of enclosure systems around the emission
operations. The enclosures are maintained under negative pressure and the
contained emissions are ducted to the controls specified in Table 55.12i.
Potential emissions from the^saturator, wet looper, and coater are generally
collected by a single enclosure, by a canopy type hood, or by an
enclosure/hood combination. Typically applied controls for POM emission
sources in asphalt roofing plants are summarized in Table 55.
Emission Factors
For the asphalt roofing manufacturing industry, POM emission factor
data?:e*ist for asphalt air blowing stills and for asphalt saturators. The
available information is presented in Table 56.121'124 AS might be
expected, the quantitative results are highly variable. However,
qualitatively the POM compounds identified in the emission streams are very
consistent.
216
-------
TABLE 55. CONTROL DEVICES USED ON POM EMISSION
SOURCES IN ASPHALT ROOFING PLANTS121
Emission Source
Control Device
A. Saturator, wet looper (hot
looper), and coater^
Afterburner
High velocity air filter
Electrostatic precipitator
B. Coater-mixer
High velocity air filter
C. Asphalt blowing still
Afterburner
D. Asphalt storage tanks0
Mist eliminator
These sources usually share a common enclosure, and emissions are ducted to
a common control device.
^ •
Emissions from the coater-mixer are controlled, at some plants, by routing
fumes to the control device used for sources listed in A, above.
Some plants control emissions from storage tanks with the same device used
for processes listed in A and then use a mist eliminator during periods
when the roofing line is not operating (e.g., weekends). Asphalt delivery
can be accomplished via a closed system which vents emissions to the same
control device as that used for the tanks.
217
-------
•3-
CN
a
J-> fa
e 4J
u e
u o
fa u
0)
a« X
ti n ja
H-2C
2«
—i w oo
o _
fa SO
sg fa-
S 2
u
o c ja
H o —
• H s^
•O CO
V a oo
at
•»
U
<~~
cr>
•H
o
O
to
ejt
o
•
o
o
0
•
o
to
•
e
e
u
e
o
o
as
•»
O
CM
•
•4-
T
o
-sr
o
o
fa
*
f
en -N -H
o e e
o o o
X I
o !
OO •
•fa
(•1 O
ON -H
O
O
O O
>0
o
o
o
e
a
u
xco xo
esfa^ esfa"
t-m _ui
ox ox
O to O to
I -« I —
> 01
J2 fa
•e £
oi s
«?• s
— 01
01 J=
> a.
•o o •
~_~ Q
B N B
g B O
3 0) "N
— J to
8 ..S
B £ §
O 0>
— fa 01
u >, a
u a. tu
ai
a.
O
o
in
I
•5
01
fa &. fa
O o
fa ±J fa u
o» Pis
B fa , B
01
u
fa
3
O
CO
B
O
u
0) -rt 0)
§*W g
ta CM
•s
o
z
•§•
fa fa fa
0) >p4 0)
c < c
fa fa
3 >> 3
•o o ,a
fa B fa
01 01 01
B
I
_ffl 0) 01 0)
•P4 01
B O.
u
•a
S S
U AJ -H
01 U
e u
01 4J
2
"S ?
o co
B (0
fa a
S - -
41
eg
fa
a
At
eg
en
o
•e co fa
u U 01 01
eg o oo B
i; ^ eg eg
3 9 fa a
w wo
a o jj o
co en to- u
1°
•H
I
60
B
I S
o
AJ
es
•i. 4 2
g g « e, i
S ^ ij fa eg
S S 5 2 4
» » eg
a. » AJ
X 10 ^
oi a> o
0)0.0)
fa X fa
eg oi o>
a oi *
fa fa —<
O eg CM
AJ «H
u a
eg fa oi
iw o u
AJ B
B u at
O eg fa
01
•"
01 01
g »
-C 01
u a
B ai
eg fa
fa ts
<
At i-N Q)
u «H
oj o>
B .£>
01
u *n
eg
e
01
u
fa
0)
0.
oi
£
"s
oi
u
a
fa 0}
•1
01 S
B 0>
oi Z
B 01
co s
•* e
l«
4
eg
.fi ai
AJ U
a AJ -* B
co a < <
a jo u
218
-------
-------
Uncontrolled total POM emissions from asphalt saturators ranged from
0.001 to 2.1 -g/Mg (2.0 x lO'6 to 0.0042 Ib/ton) . 121 > 124 The values reported
in Reference 124 are much less than those in Reference 121 because few, if
any, vaporous POM emissions were collected by the sampling procedure used in
Reference 124. The sampling procedure used for the Reference 121 samples
apparently was more successful for condensing and collecting vaporous POM.
The same situation exists in the asphalt blowing still -results where the
total POM range is 0.0048 to 15.2 g/Mg (9.6 x 10"6 to
0.030 Ib/ton). 121>122'124'125
emission factors where the total POM factors for afterburner-controlled
saturators range from 0.001 to 0.63 g/Mg (2.0 x 10'6 to O.'OOIS Ib/ton) and
the factors for incineration- controlled blowing stills range from 0 0021 to
0.048 g/Mg (4.2 x lO'6 to 0.001 Ib/ton) . 121 ' 122 • 124 Because of the
differing sampling and analysis procedures used to obtain the factors given
in Table 56, data comparisons between sources is difficult and could lead to
erroneous conclusions.
The POM compounds identified in roofing source emissions were
consistent within a source type (e.g., saturators) and between different
source types. Anthracene/phenanthrene , methyl anthracenes, f luoranthene ,
pyrene, methyl pyrene, chrysene, benz( a) anthracene, methyl chrysenes , benzo
fluoranthenes, benzo (a) pyrene, and benzo (e) pyrene were identified in the
emission measurements of practically every source. In both controlled and
uncontrolled emissions of saturators and blowing stills, methyl anthracenes
predominated. Anthracene/phenanthrene and methyl pyrene/fluoranthene also
repeatedly constituted significant portions of total POM emissions.
Generally, the three POM compound groups constituted between .90 and
95 percent of total POM measured.
Source Locations
As of mid- 1986, there were 94 asphalt roofing manufacturing plants
operating in the United States. A list of all current facilities as
identified by the Asphalt Roofing Manufacturers Association, is provided in
220
-------
Table 57. States containing a relatively significant number of roofing
plants include California; Texas, Ohio, and Alabama. These four States
contain approximately 40 percent of the total number of roofing facilities.
The majority of all plants nationwide are located in urban as opposed to
rural areas.
HOT MIX ASPHALT PRODUCTION
Process Description
In the production of hot mix asphalt (also known as asphalt concrete),
aggregate, which is composed of gravel, sand, and mineral filler, is heated
to eliminate moisture and then mixed with hot asphalt cement. The resulting
hot mixture is pliable and able to be compacted and smoothed. When it cools
and hardens, hot-mix asphalt provides a waterproof and durable pavement for
roads, driveways, parking lots, and runways.
Currently, there are three types of hot mix asphalt plants in use in
the United States: batch-mix, continuous-mix, and drum-mix. Batch-mix and
continuous-mix plants separate the aggregate drying process from the mixing
of aggregate with asphalt cement. Drum-mix plants combine these two
processes. Production capacities for all three types of plants range from
36 to 544 Mg (40 to 600 tons) of hot mix per hour. The production capacity
distribution for the three types of hot mix asphalt plants is presented in
Table 58. Over 80 percent of all hot mix asphalt production plants are
mobile.
127
Raw Materials--
The basic raw material of hot mix asphalt, aggregate, consists of any
hard, inert mineral material mixed with a binding agent to produce hot mix
asphalt. Aggregate typically comprises between 90 and 95 percent by weight
of the asphalt mixture. Since aggregate provides most of the load-bearing
properties of a pavement, the performance of the pavement depends on
selection of the proper aggregate.
221
-------
TABLE 57. ASPHALT "ROOFING MANUFACTURING LOCATIONS
IN THE UNITED STATES IN 1986126
Company
American Roofing Corporation
Bird, Incorporated
Celotex Corporation
Certainteed Corporation
Consolidated Fiberglass Products
Dibiten, U. S. A.
Elk Corporation of America
Evanite/Permaglas, Incorporated
GAF Corporation
The Garland Company
Plant Locations
Chicago, IL
Norwood, MA
Camden, AR
Fremont, CA
Fairfield, AL
Russellville, AL
Goldsboro, NC
Los Angeles, CA
Houston, TX
Lockland, OH
Perth Amboy, NJ
San Antonio, TX
Memphis, TN
Avery, OH
Oxford, NC
Shakopee, -MN
Savannah, GA
Bakersfield, CA
South Gate, CA
Ennis-, TX
Tuscaloosa, AL
Corvallis, OR
Baltimore, MD
Dallas, TX
Erie, PA
Fontana, CA
Millis, MA
Minneapolis, MN
Mobile, AL
Mount Vernon, IN
Savannah, GA
Tampa, FL
Chester, SC
.Cleveland, OH
222
-------
TABLE 57. ASPHALT ROOFING MANUFACTURING LOCATIONS
IN THE UNITED STATES IN 198612 (Continued)
Company
Plant Locations
Georgia-Pacific Corporation
Globe Industries, Incorporated
Iko Industries Limited
Koppers Company
Leatherback Industries
Lunday-Thagard
Manville Corporation
Nord Bitumi U. S., Incorporated
Owens-Corning Fiberglas Corporation
Ardmore, OK
Daingerfield, TX
Franklin, OH
Hampton, GA
Quakertown, PA
Denver, CO
Pryor, OK
Whiting, IN
• .•
Wilmington, DE
Chicago, IL
Franklin, OH
Chicago, IL
Fontana, CA
Houston, TX
Wickliffe, OH
Woodward, AL
. Youngstown, OH
*
Albequerque, KM
Hollister, CA
Auburn, WA
South Gate, CA
Fort Worth, TX
Manville, NJ
Pittsburg, CA
Savannah, GA
Waukegan, IL
Etowan, TN
Waterville, OH
Plattsburgh, NY .
Macon, GA
Atlanta, GA
Brookville, IN
Compton, CA
Denver, CO
223
-------
TABLE 57. ASPHALT ROOFING MANUFACTURING LOCATIONS
IN THE UNITED STATES IN 1986126:(Continued)
Company
Owens-Corning Fiberglas Corporation
(continued)
Siplast, Incorporated
Tamko Asphalt Products, Incorporated
Tremco, Incorporated
U. S. Intec, Incorporated'
W. R. Grace and Company
Plant Locations
Houston, TX
Irving, TX
'Jacksonville, FL
Jessup, MD
Kearny, NJ
Medina, OH
Memphis, TN
Minneapolis, MN
Morehead City, NC
Oklahoma City, OK
Portland, OR
Summit, IL
Aiken, SC
Barrington, NJ
Kansas City, MO
Arkadelphia, AK
Joplin, MO
Frederick, MD
Phillipsburg, KS
Tuscaloosa, AL
•Knoxville, TN
Cleveland, OH
Port Arthur, TX
North Branch, NJ •
Stockton, CA
Cambridge, MA
NOTE:
224
-------
r
TABLE 58. PRODUCTION CAPACITY DISTRIBUTION FOR BATCH, CONTINUOUS,
AND' DRUM-MIX HOT MIX ASPHALT PLANTS127 . •
Type of Plant
Batch-mix Plants
Drum-mix Plants
Continuous-mix Plants
Production
Range, Mg/h (tons/h)
Under 136 (Under 150)
136 - 272 (150 - 300)
273 - 363 (301 - 400)
Over 363 (Over 400)
Under 136 (Under 150)
136 - 272 (150 - 300)
273 -. 363 (301 - 400.)
Over 363 (Over .400)
•
Under 136 (Under 150)
136 - 272 (150 - 300)
273 - 363 (301 - 400)
Over 363 (Over 400)
Percentage of
Plants Within
Production Range
25
63
11
100
15
52
26
100
43
21
19
100
225
-------
Asphalt cement is used as the binding agent for aggregate. It prevents
^sture fro. penetrating the aggregate and it acts as a cushioning agent
Typically, asphalt cement constitutes 4 to 6 percent by weight of a hot »ix
asphalt mixture.
- Asphalt cement is generated as a residue from the distillation of crude
petroleum. It is classified into grades under one of three systems The
most commonly used system classifies asphalt cement based on its viscosity
at 60 C (140 F). The .more viscous the asphalt cement, the higher its
numerical rating. An asphalt cement of grade AC-40 is considered a hard
asphalt [i.e., a viscosity of 4000 grams per centimeter per second (g/cm-s)
(pox...)], while an asphalt cement of grade AC-2.5 is considered a soft
asphalt [i.e., a viscosity of 250 g/cm-s (poises)]. Several western States
use a second grading system that measures viscosity of the asphalt cement
. after a standard simulated aging period. This simulated. aging period
consists of exposure to a temperature of 163°C (325°F) for 5 hours
Viscosity is measured at ^C (140°F) , with grades ranging from AR-1000 for
a soft asphalt cement [1000 g/cm.s (poises) 3 tQ ^.^ ^ & ^ '
cement [16,000 g/cm-s (poises)]. A third grading system is based on the -.
penetration allowed by the asphalt cement/ Grade designation 40 to 50 means
that a needle with a weight attached will penetrate the asphalt cement
between 40 and 50 tenths of a millimeter under standard test conditions.
The hard asphalt cements have penetration ratings of 40 to 50, while the
spft grades have penetration ratings of 200 to 300. 127
The asphalt cement grade selected for different hot mix asphalts
depends on the type of pavement, climate, and type and amount of traffic
expected., Generally, asphalt pavement bearing heavy traffic in warm
that has been removed from existing roadvays. Recycled
hot *ix asphalt is
226
-------
used by a growing number of companies in their hot mix asphalt mixtures. -
The Surface Transportation Assistance Act of 1982 encourages recycling by
;providing a 5 percent increase in Federal funds to State agencies that
recycle asphalt concrete pavement. Rarely does the recycled hot mix asphalt
comprise more than 60 percent by weight of the new asphalt mixture.
Twenty-five- percent recycled hot mix asphalt mixtures are typical in batch
plants while 40 to 50 percent RAP mixtures are typical in drum-mix
plants.127
Rejuvenating agents are sometimes used in recycled hot mix asphalts to
bring the weathered and aged asphalt cement in the recycled mixture up to
the specifications of the new asphalt mixture. Usually, a soft asphalt
cement, a specially prepared high viscosity oil, or a hard asphalt cement
blended with a low viscosity oil are used as rejuvenating agents. The
amount of rejuvenating agent added depends on the properties of the recycled
asphalt and on the specifications for the hot mix asphalt product.
Sulfur has also been used on an experimental basis as a substitute for
a portion of the asphalt cement, in hot mix asphalt mixtures. Tests have
shown that the asphalt.cement/sulfur combination is better able to bind with
aggregate than is asphalt cement alone. Hot mix asphalt pavements
containing the asphalt cement/sulfur combination appear to be stronger and
less susceptible to temperature changes than those containing asphalt cement
alone.
The use of sulfur is not competitive with asphalt cement in asphalt
concrete mixes for several reasons, including environmental questions,
worker objections (odor), and corrosion, all of which result from emissions
of hydrogen sulfide (H,,S) , sulfur dioxide (SO,,) , and elemental sulfur (S).
In addition, sulfur is almost twice as dense as asphalt cement.
Frequently, to make the use of sulfur economically feasible, the cost of
sulfur must be less than half the cost of asphalt cement.127
227
-------
Batch-Mix Plants--
The primary processes of a typical batch-mix hot mix asphalt facility
are illustrated in Figure 26.127 Aggregate of various sizes is stockpiled
at the plant for easy access. The moisture content of the stockpiled
aggregate usually ranges from 3 to 5 percent. The moisture content of
recycled hot mix asphalt typically ranges from 2 to 3 percent. The
different sizes of aggregate are typically transported by front-end loader
to separate cold feed bins and metered onto a feeder conveyor belt through
gates at the bottom of the bins. The aggregate is screened before it is fed
to the dryer to keep oversized material out of the mix.
The screened aggregate is then fed to a rotating dryer, with a burner at
its lower (discharge) end that is fired with fuel oil, natural gas, or
propane. - The dryer removes moisture from the aggregate and heats the
aggregate to the proper mix temperature. Inside the dryer are longitudinal
flights (metal slats) that lift and tumble the aggregate, causing a curtain
of material to be exposed to the heated gas stream. This curtain of
material provides greater heat transfer to the aggregate than would occur if
' the aggregate tumbled along the bottom of the drum towards the discharge
end.^ Aggregate temperature at the discharge end of the dryer is about 149°C
(300 F). The amount of aggregate that a dryer can heat depends on the size
of the drum, the size of the burner, and the moisture content of the
aggregate. As the amount of moisture to be removed from the aggregate
increases, the effective production capacity of the dryer decreases.
Vibrating screens segregate the -'heated aggregate into bins according to
size. A weigh hopper meters the desired amount of the various sizes of
aggregate into a pugmill mixer. The pugmill typically mixes the aggregate
for approximately 15 seconds before hot asphalt cement from a heated tank is
sprayed into the pugmill. The pugmill thoroughly mixes the aggregate and
hot asphalt cement for 25 to 60 seconds. The finished hot mix asphalt is
either directly loaded into trucks or held in insulated and/or heated
228
-------
co
e
o
0)
O
a.
CU
u
CO
iH CO
3 i-l
•H U
U JJ
>- C
CO O
._! 1
0)
u 00'
to c
00 IH
cu >%
!-> V
oo a
oo
*?.
i
•o
I— 1
0
u
cu
00 CO
CO C
o ea
I
•o co
cu cu
iH 4J
•H CO
a. oo
^ cu
u ^
O 00
u QO
OT <
CN
i—I CU
CO U 00
)- cu co
CU -( U
e IH o
•H -H JJ
S fa en
ca cu
^= ^
•u to
CO U
a
CU CO
u E
to
gen
co
cu cu
4J U
i-l O
CO h
co
cu cu
CO
u
e
CO
CO
a
to
CO
o
e
u
CO
e
OJ
cn
cn
co
0)
u
2
a.
CO
u
CN
0)
J-l
00
•H
229
-------
storage silos. Depending on the production specifications, the temperature
of the hot mix asphalt product mix can range from 107 to 177°C (225 to
350 F) at the end of the production process.
When mix containing recycled asphalt is produced, the aggregate is
superheated (compared to totally virgin hot mix production) to about 315°C
(600 F) to ensure sufficient heat transfer to the recycled asphalt when it
is mixed with the: virgin materials. Recycled hot mix asphalt may be added
exther to the pugmill mixer or at the discharge end of the dryer. Rarely is
more than 30 percent recycled asphalt used in batch plants for the
production of hot mix asphalt.
Continuous-Mix Plants--
. Continuous-mix plants are very similar in configuration to batch
Plants. Continuous-mix plants have smaller hot bins (for holding the heated
aggregate) than do batch plants. Little surge capacity is required of these
bxns because the aggregate is continuously metered and transported to the
mixer xnlet by a conveyor belt. Asphalt cement is continuously added to the
aggregate at the inlet of the mixer. The aggregate and asphalt cement are
mxed by the action of rotating paddles while being conveyed through the
mxxer. An adjustable dam at the outlet end of the mixer regulates the
rnxxxng time and also provides some surge capacity. The finished mix is
transported by a conveyor belt to either a storage, silo or surge bin.127
Drum-Mix Plants--
The essential components of a typical virgin hot mix asphalt drum-mix
plant are shown in Fieure 27 ' TV,-,™ j i
i, •«, ,, Drum-mix planes dry the aggregate and m
Wth the asphalt cement in the same drum, eliminating the need for the
extra conveyor belt, hot bins and screens, weigh hopper, and pugmiU.
Although the drum of a dru^-mix plant is much lite the dryer of a batch
Plant, the burner is at the aggregate feed end rather than at the aggregate
discharge end. The veU of aggregate is heated as it flows with the heated
gas stream instead of countercurrent to the gas stream as in a batch plant.
230
-------
r
CO
CO-
OI
y
2
Q.
a.
CO
SJ
O
s
i-l
•o
CB
U
£M
(U
U
231
-------
The burner in a drum-mix plant emits a much bushier flame than does the
burn-er in a batch plant. The bushier -flame is designed to provide earlier
and greater exposure of the virgin aggregate to the heat of the flame. This
design also protects the asphalt cement, which is injected approximately
two-thirds of the way down the length of the drum, away from the direct heat
of the flame. Drum-mix plants typically have more flights in their drums
than do batch dryers to increase veiling of the aggregate and to improve
overall heat transfer. The asphalt cement, which is usually injected from a
Pipe inserted from the discharge end of the rotating drum, coats the
aggregate. The temperature of the mix as it leaves the drum usually ranges
from 107 to 177°C (225 to 350°F). The hot mix asphalt is then transported
by conveyor to a surge bin or to an insulated, and possibly heated, storage
silo for truck load-out.
• Recycled ho.t mix asphalt can also'be used as a raw material in drum-mix "
Plants. Currently in drum-mix plants, recycled hot mix asphalt is
introduced'through a collar midway down the drum and is dried by both the '
superheated aggregate and by the gas stream. The veil of virgin aggregate
created by the flights in the drum keeps the high heat of the burner flame
from reaching the recycled asphalt. Two Vendors have attempted to improve
on this approach by also expanding the drum diameter at the burner end to
allow a shorter, bushier flame and to obtain more efficient heat transfer '
from the burner flame to the virgin aggregate.
One major advantage of drum-mix plants is that they can produce
material containing- higher percentages of recycled hot mix asphalt than
batch plants can produce. With the greater veiling of aggregate, drum-mix
Plants are more efficient than batch plants at transferring heat and
achieving proper mixing of recycled asphalt and virgin materials. The trend
in hot mix asphalt production is towards drum-mix plants 128
232
-------
-------
Indirect Heated Plants--
A potential new commercial production process for hot mix asphalt
involves indirect heating of the aggregate and asphalt cement in a mixer.'
In this process, asphalt cement and preheated aggregate are introduced
through air-locks into a heated, sealed mixing unit. Synthetic heat
transfer fluids are heated to 316 to 343°C (600 to 650°F) by a fuel
efficient diesel- or gas-fired burner. These" synthetic fluids heat the
mixing unit chamber to approximately 149°C (300°F). Steam from the moisture
driven off from the aggregate is piped to the cold feed bins to preheat the
virgin aggregate. This preheating of the aggregate decreases energy costs
for drying. The product hot mix asphalt is transported by an enclosed
conveyor from the mixing unit to a storage silo. Because this process is
sealed, there are no mixer process air emissions. The steam from the mixing
unit condenses as it preheats the cold feed. bins. The only other process
emissions are the gases from the heater unit.
The indirect heated process has been successfully demonstrated with a
pilot-scale plant capable of producing 14 to 18 Mg (15 to 20 tons)/hour of
hot mix asphalt. Stationary and portable indirect heated plants have been
designed with production capacities ranging from 45 to 204 Mg (50 to
225 tons)/hour, and commercial plants of 181 to 272 Mg (200 to
300 tons)/hour production capacity are expected by 1986.127
Polycyclic organic matter emissions in hot mix asphalt plants occur
from the aggregate rotary dryers (due to fuel combustion) and from the hot
mx asphalt mixing vessels (due to heating of the organics-containing
asphalt materials). Most plants employ some form of mechanical collection
typically cyclones, to collect aggregate particle emissions from the rotary
dryers. These cyclones would have a minimal collection efficiency for POM
compounds because the POMs are either in vapor form or would predominantly
exist on fine particles not captured by the cyclones. In many
installations, the recovered aggregate is recycled to the hot mix asphalt
process.
233
-------
Overall, particulate matter emissions from hot mix asphalt mixers are
controlled by wet scrubbers or baghouses.127 Again, their success on POM
emissions is dependent on the form of the POMs (i.e., vapor versus
particulate and fine versus coarse particle). In some installations, the
exhaust stream of the rotary dryer cyclones is vented to the baghouse or
scrubber used for mixer emissions control.129'130 One reference has
indicated that for hot mix asphalt- plants venting dry emissions to the mixer
control device, the POM compounds detected in the mixer control device
emissions were predominantly a function of the rotary dryer and not the
mixer. 50
Because the drum-mix hot mix asphalt process is based on a parallel
flow design (i.e., hot gases and aggregate flow through the dryer in the
same direction); general particulate matter emissions from this process are
less than from a conventional batch process. However, because the asphalt
materials contained in the process are heated to a higher temperature for a
longer time, the drum-mix process potentially may produce greater .levels of
POM emissions. °
In any of the processes used to produce hot mix asphalt, fugitive POM
emissions may occur due to evaporative losses from asphalt handling and
storage. Emissions of this type would be highly variable. No examination
of fugitive POM emissions from hot mix asphalt plants could be found in the
literature.
Emission Facto-rs
Several total POM and speciated POM emission factors were identified in
the literature for hot mix asphalt production facilities. A summary of
these factors for total POM is presented in Table 59.128'132 The limited
data in Table 59 indicate that drum-mix plants potentially have greater POM
emissions than batch plants and plants using recycled hot mix asphalt have
higher POM emissions than those employing only virgin materials.
-------
235
-------
Two references in the literature contain well-documented emission
factors for individual POM species from hot mix asphalt plants." These
factors are presented in Table 60, 61, and 62 to better illustrate what POMs
are predominant in emissions from hot mix asphalt plants.129'132 The
results shown in Tables 60 and 61 for drum-mix plants indicate that for this
type of facility, POM emissions predominantly exist in vapor form.132 The
results in Table 62 indicate that batch plant emissions are predominantly in
particulate form. The emission test methods used to obtain the results
in Tables 60, 61, and 62 were similar with the exception of the resin type
used for organics capture. An EPA Modified Method 5 sampling train was used
for both plants; however, the batch-mix plant test used Tenax resin for
organics collection, while the drum-mix plant test used XAD-2 resin. The
effect, if any, of this difference on the POM emission results cannot be
determined with the information available.. However, as described in
Section 5 (Source Test Procedures), XAD-2 is currently the preferred
organics collection r.esin.
Source Locations ' '
_ There are approximately 2150 companies operating an estimated
4500 hot mix 'asphalt plants in the United States. Approximately 40 percent
of these companies operate only a single plant. Because plants must be
located near the job site, plants are concentrated in areas where the
highway and road network is concentrated.127 Additional information on the.
locations of individual hot.mix asphalt facilities can best be obtained by
contacting the National Asphalt Pavement Association in College Park,
Maryland.
l I
CARBON BLACK MANUFACTURE
Process
The chemical carbon black consists of finely divided carbon produced by
the thermal decomposition of hydrocarbons. Carbon black is a major
industrial chemical used primarily as a reinforcing agent in rubber
compounds, especially tires.133 The manufacture of carbon black is of
236
-------
TABLE 60. INDIVIDUAL POM SPECIES EMISSION FACTORS FOR A DRUM-MIX
HOT MIX ASPHALT PLANT USING VIRGIN FEED MATERIAL132
Emission Factors Under
POM Species
Benz(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(j)fluoranthene
" Benzo(e)pyrene
Benzo(a)pyrene
Indeno(1,2,3 -c,d)pyrene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(k)fluoranthene
Perylene
Benzo(g,h,i)perylene
TOTAL POMa
• •
totals may not equal sum
b.
Knockout Box
Control Device
Only (mg POM/Mg
Asphalt Produced)
rol Situations
Knockout Box
Followed by Venturi
Scrubber (mg POM/Mg
Asphalt Produced)
. 0.13
0.73
0.06
None detected
0.34
0.13
0.16
13.5
2.0
1.3
3.5
0.06
0.06
None detected
22. lb
_
0.02
0.13
0.06
None detected
0.02
None detected
0.01
11.2
0.79
0.31
0.84
0.06
0.02
None detected
13.5°
of individual values due to rounding
237
-------
TABLE 61. INDIVIDUAL POM SPECIES EMISSION FACTORS FOR A DRUM-MIX
HOT MIX ASPHALT PLANT USING RECYCLED FEED MATERIAL132
Emission Factors Under
Different Control S-f f,^f-fnr,*.
POM Species
Benz (a) anthracene
Chrysene
Benzo (b) fluoranthene
Benzo ( j ) fluoranthene
Benzo ( e ) pyrene
Benzo (a) pyrene
Indeno ( 1 , 2 , 3 - c , d) pyrene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo (k) fluoranthene
Perylene
Benzo(g,h,i)perylene
TOTAL POMa
Knockout Box
Control Device
Only (mg POM/Mg
Asphalt Produced)
0.21
1.0
0.01
None detected
0,23
0.06
0.02
25.9
1.9
2.3
4.2
0.1
0.04
None detected
35. 9b
Knockout Box
Followed by Venturi
Scrubber (mg POM/Mg
Asphalt Produced)
0.09
0.29
0.03
None detected
0.06
0.03
0.03
10.0
2.0
1.5
2.9
0.03
0.02
-None detected
16. 9C
j _^—^ ^.^ VJ_ Aitu.jLvj.uuax values due to rounding.
hSf^f^VJ??^ °f thS t0tal P°M 1uantity wa* collected in the back
of the nro^. £™ Me5hod 5 samPling system indicating that the majority
or tne process's POM emissions are in vapor form.
-r
ail of the emissions exiting the scrubber are in vapor form.
238
-------
TABLE 62. INDIVIDUAL POM SPECIES EMISSION FACTORS FOR A BATCH-MIX
HOT MIX ASPHALT PLANT USING VIRGIN FEED MATERIAL129
POM Species
••"^^•^""•"•"""•i^^™.
Dibenzothiophene
Anthracene/phenanthrene
Methylanthracenes/phenanthrenes -
9-Methylanthracene
Fluoranthene
Pyrene
Benzo(c)phenanthrene
Chrysene/benz(a)anthracene
7,12-DimethyIbenz(a)anthracene
3,4-Benzofluoranthene
Benzo(a)pyrene/benzo(e)pyrene/perylene
3-Methylcholanthrene
Dibenz (a,h)anthracene
Indeno(1,2,3-c,d)pyrerie
7H-Dibenzo(c,g)carbazole
Dibenzo(a,h and a,i)pyrene.
TOTAL POM
POM Emission
Factor (mg/Mg
Asphalt Production)a
3.6
4,0
6.9
0.4
0.7
0.8
0.3
0.5
0.3
0.4
0.3
0.3
0.3
0.3
0.3
JL2.
19.7b
239
-------
potential concern for POM emissions because the predominantly used
production process involves the combustion of natural gas and the
high-temperature pyrolysis ofjaromatic liquid hydrocarbons.
Approximately 90 percent of all carbon black produced in the United
States is manufactured by the oil-furnace process, a schematic of which is
given in Figure 28. 133'134 The process streams identified in Figure 28 are
defined in Table 63. Generally, all oil-furnace carbon black plants are
similar in overall structure and operation. The most pronounced differences
in plants are primarily associated with the details of decomposition furnace
design and raw product processing.133 other processes used for carbon black
production are thermal decomposition of natural gas and exothermic
decomposition of acetylene.134
In the oil-fumace process, carbon black is produced by the pyrolysis
of an atomized liquid hydrocarbon feedstock in a refractory-lined steel
furnace. Processing temperatures in the steel furnace range from 1320 to
1540 C (2408 to 2804°F). The heat needed to accomplish the desired
hydrocarbon decomposition reaction is supplied by the combustion of natural
gas.
Feed materials used in the oil-fumace process consist of petroleum
oil, natural gas, and air. Also, small quantities of alkali metal salts may
be added^to the oil feed to control the degree of structure of the carbon
black. The ideal raw material for the production of modern, high
structure carbon blacks is an oil which is highly aromatic, low in sulfur
asphaltenes and high molecular weight resins, and substantially free of
suspended ash, carbon, and water. The reactor for the oil furnace process
consists of a refractory-lined steel furnace which is from 1.5 to 9 m (4.9
to 29.5 ft) in length and 0.15 to 0.76 m (0.49 to 2.5 ft) in internal
diameter. To provide maximum efficiency, the furnace and burner are
designed to separate, insofar as possible, the heat generating reaction from
the carbon forming reaction. Thus, the natural gas feed (stream 2 in
Figure 28) is burned to completion with preheated air (stream 3) to produce
240
-------
C
to
u
CO
o
.0
!-i
CO
0
01
O
to
s
O
M-l
OJ
0)
1
O
co
CO
0)
u
o
03
CM
0)
J-l
M
•H
241
-------
TABLE 63. STREAM CODE FOR THE OIL-FURNACE PROCESS
ILLUSTRATED IN FIGURE 28133
Stream
Identification
1
' 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Oil feed
Natural gas feed
Air to reactor
Quench water
Reactor effluent
Gas to oil preheater
Water to quench tower
Quench tower effluent
Bag filter effluent
Vent gas purge for dryer fuel
Main process vent gas
Vent gas to incinerator -
Incinerator stack gas
Recovered carbon black
Carbon black to micropulverizer
Pneumatic conveyor system
Cyclone vent gas recycle
Cyclone vent gas
Pneumatic system vent gas
Carbon black from bag filter
Carbon black from cyclone
Surge bin vent
Carbon black to pelletizer
Water to pelletizer
Pelletizer effluent
Dryer direct heat source vent
Dryer bag filter vent
Carbon black from dryer bag filter
242
-------
Stream
29
30
31
32
33
34
35
36
37
38
39
TABLE 63. STREAM CODE FOR THE OIL-FURNACE PROCESS
ILLUSTRATED IN FIGURE 28133 (Continued)
^M—M_i^HM.
._
Identification
'™™1—™"^™^~™™™''"—"•"—•—«••
Dryer indirect heat source vent
Hot gases to dryer ."-•
Dried carbon black
Screened carbon black
Carbon black recycle
Storage bin vent gas
Bagging system vent gas
Vacuum cleanup system vent gas
Dryer vent gas
Fugitive emissions
Oil storage tank vent gas
243
-------
a temperature of 1320 to 1540°C (2408 to 2804°F). The reactor is designed
so that this zone of complete combustion attains a swirling motion, and the
.o.il feed (stream 1), preheated to 200 to 370°C (392 to 698°F), is sprayed
itato the center of the zone. Preheating is accomplished by heat exchange
with the reactor effluent and/or by means of a gas-fired heater. The oil is
cracked to carbon and hydrogen with side reactions producing carbon- oxides,
water, methane, acetylene and other hydrocarbon products. The heat transfer
from the hot combustion gases to the atomized oil is enhanced by highly
turbulent flow in the reactor.134-
The reactor converts 35 to 65 percent of the feedstock carbon content
to carbon black, depending on the feed composition and the grade of black
being produced. The yields are lower for the smaller particle size grades
of black. Variables that can be adjusted to produce a given grade of black
include operating temperature, fuel concentration, space velocity in the
reaction zone, and reactor geometry (which influences the degree of
turbulence in the reactor). . A typical set of reactor operating conditions
is given below for high abrasion furnace carbon black.134
_Parameter
Rate of oil feed
Preheat temperature of oil
Rate of air feed
Rate of natural gas feed
Furnace temperature in reaction zone
Rate of carbon black production
Yield of black (based on carbon in oil feed)
JValue
0.76.m /hr
288°C
6653 m3/hr
623 m3/hr
1400°C
390 kg/hr
60 percent
^ The hot combustion gases and suspended carbon black are cooled to about
540 C (1004 F) by a direct water spray in the quench area, which is located
near the reactor outlet. The reactor effluent (stream 5 in Figure 28) is
further cooled by heat exchange in the air and oil preheaters. It is then
sent to a quench tower where direct water sprays finally reduce the stream
temperature to 230°C (446°F).
244
-------
Carbon black is recovered fro* the reactor effluent stream by means of
a bag filter unit. The carbon black laden gases from the quench tower
(stream 8 in Figure 28) enter the bag filter hopper trough below the bag
cell plates at 200 to 230°C (392 to 446°F). The gases flow into the
individual bags of each compartment through the cell plates. The carbon
black collects on the inside of the bags and the filtered gas -flows through
the bags and out the bag filter stacks. During the cleaning cycle of each
compartment, the black is removed from the bag fabric and drops back into
the hopper trough. It passes from the hopper through a hammer mill that
breaks up the lumps and is then transported to the product treatment section
of the plant via a pneumatic conveyor system (stream 16).134
The exhaust gas from the bag filter unit (stream 9 in Figure 28) is
vented .directly to the atmosphere in most carbon black plants
Alternatively, it may be sent to a flare or incinerator to reduce the
c.«ntaminant loading (stream 12) . In addition, 13 to 15 percent of the
effluent (stream 10) may be diverted to produce auxiliary fuel for the raw
product drying operation.
The raw carbon black collected in the bag filter unit must be further
processed to become a marketable product. After passing through the
pulverizer, the black has a bulk density of 24 to 59 kg/m3, and it is too
fluffy and dusty to be transported. It is therefore converted into pellets
or beads with a bulk density of 97 to 171 kg/m3. In this form, it is
dust-free and sufficiently compacted for shipment.
The carbon black is collected from the pneumatic system (stream 16) by
means of a cyclone, a bag filter, or a cyclone and bag filter in
combination. When a cyclone is used, the exhaust gas (stream 17) is -
recycled to the primary bag'filter unit. When a bag filter is used, the
exhaust gas (stream 19) is vented to the atmosphere. The recovered carbon
black „ collected in a covered surge bin. The carbon black is fed from the
surge b« vxa a screw conveyor to the pelletizer, where it is mixed with
one part of water to two parts of black. A binding agent such as molasses
245
-------
sugar, dextrin, or starch, may be added to the pelletizing water. The
pelletizer is a horizontal housing that contains a revolving axial shaft
with pins or spikes mounted on its periphery. Agitation by the pins causes
the mixture of carbon black and water to form nearly spherical particles I 6
to 3.2 mm (0.06 to 0.13 in) in diameter. The pellets are then conveyed to a
dryer for removal of the water.
Rotating horizontal drums operating at 190 to 230°C (374 to 392°F) are
typically used for product drying in carbon black processes. The dryers are
fueled by natural gas, which may be augmented by a portion of the main
process vent gas. From 35 to 70 percent of the combustion gas is charged
directly to the interior of the dryer. After passing through the dryer,
this stream (stream 26) is sent to a bag filter for removal of entrained
carbon black before being vented to the atmosphere. The remaining 30 to
65 percent of the combustion gas (stream 29) acts as an indirect heat source
for the dryejr and is vented directly to the atmosphere.
The dried, pelletized carbon, black (stream 31) is screened and sent to
a covered storage bin via a bucket elevator. Oversize.pellets are removed .'
in the screener and recycled (stream 33) to the pulverizer. From the
product storage bin, the carbon black can be loaded into railroad hopper
cars for bulk shipment or sent to a vacuum bagging system which is
hermetically sealed to prevent emission of carbon black.134
Exhaust gas from the bag filter unit constitutes the main process vent
and the largest source of POM emissions. About two-thirds of the United
States carbon black plants treat the bag filter exhaust stream to control
carbon monoxide and hydrocarbon emissions. Combustion in thermal
incinerators, flares, or carbon monoxide boilers is used for treating the
gases. In the remaining facilities, bag filter exhaust emissions are vented
directly to the atmosphere.135 Emissions from product dryers are
predominantly controlled by high efficiency bag filter units; however, water
scrubbers are also used at a few facilities.134
246
-------
Polycyclic organic matter emissions associated with raw carbon black
production (exclusive of* additional processing steps) appear to be a
function of the efficiency of the product recovery bag filter and, where '
applicable, the destructive or potentially constructive effect of
hydrocarbon and carbon monoxide combustion control devices. Because
decreased efficiency in the product recovery bag filter unit means decreased
carbon black production and lost revenues, it is likely that these bag
filters are maintained by companies at optimum conditions. The use of
combustion control devices in a majority of plants would be expected to
reduce POM emissions by destructing them into constituent compounds and '
elements (water, carbon dioxide,, nitrogen); however, some investigators have
speculated that POM compounds are being formed in the high temperature zone
of the hydrocarbon and carbon monoxide control devices.133 No data were
supplied in Reference 133 to support this POM formation theory.
Emission Factors . .
Several emission factors for POM emissions from carbon black
manufacturing were identified in the literature. All identified emission
factors are applicable to emissions from the main process vent. No
emissions data of any type were available for potential POM sources
associated with raw product processing such as grinding, drying, and
packaging.
The best documented emission factor for POM emissions from carbon black
manufacturing is that developed by Serth and Hughes in Reference 133 Total
uncontrolled POM emissions from the main process vent (product recovery
baghouse) were measured in a series of three tests with the average emission
factor being 1900 mg/Mg (0.0039 Ib/ton) of carbon black produced. Of the
total 1900 mg, 42 percent were acenaphthylene, 26 percent pyrene, and
12 percent methyl- and dimethylanthracenes/phenanthrenes. Known
carcinogenic species constituted about 8 percent by weight of the total POM
quantity
in Table 6.4.
13A breakdown of all POM compounds that were identified is given
. 247
-------
TABLE 64. INDIVIDUAL POM COMPOUNDS MEASURED IN THE TEST
OF AN OIL-FURNACE CARBON BLACK PLANT133
POM Compound
Acenaphthylene
Anthracene/phenanthrene
• Benzo(c)phenanthrene
Benzofluoranthenes
Benzo(g,h,i)fluoranthene
Benzo(g,h,i)perylene/anthanthrene
Benzopyrenes and perylene
Chrysene/benz(a)anthracene
Dibenzanthracenes
Dibenzo(c,g)carbazole
Dibenzopyrenes
Dibenzothiophene
Dimethylanthracenes/phenanthrenes
7,12-Dime thy Ibenz(a)anthracene
Fluoranthene
Indeno(1,2,3-c,d)pyrene
Methylanthracenes/phenanthrenes
Methylcholanthrene
Methylfluoranthene/pyrene
Pyrene
TOTAL
Mean Emission Factor,
mg/Mg (lb/ton)a'b>e'd
800 (0.002)
70 (0.0001)
<2 (trace)
30 (0.00006)
40 (0.00008)
23 (0.00005)
30 (0.00006)
9 (0.00002)
<2 (trace)
<2 (trace)
<2 (trace)
14 (0.00003)
140 (0.0003)
70 (0.0001)
60 (0.0001)
<2 (trace)
100 (0.0002)
<2 (trace)
23 (0.00005)
500 (0.001)
1900 (0.0039-)
^Units are in terms of mg POM/Mg of carbon black produced.
Values given are the average of three test runs.
^l-Sna'ce SSfSS! ££?" *** «~ *"
Mission factors represent particulate and gaseous POM compound constituents,
248
-------
All POM sampling in the Serth and Hughes work was conducted using a EPA
Modified Method 5. A gas cooler and organics trap containing XAD-2 resin
were used to recovery vaporous POM that may be released from the process
vent. All POM samples were separated using liquid chromatography and
anal v^a/^ «««»4«*M. «.._ ^i '_ . . 1 o^
— — — o ^^^*«»^»^» ^
analyzed using gas chromatography-mass spectrometry.
133
A second set of POM emission factor data for carbon black plants was
identified that gave total POM emission factors ranging from 220 mg/Mg
.(0.00045 Ib/ton) of carbon black produced to 490 mg/Mg (0.001 Ib/ton) with
the average being 340 mg/Mg (0.0007 Ib/ton).136 These data were not as well
documented as Reference 133 data in that it was not specified exactly what
emission point the factor applied to and if it was for controlled or
uncontrolled emissions. Of the range of values determined, the 490 mg/Mg
value was deemed by Reference 136 to be the most reliable because it was
measured using a Method 5 train followed by a Tenax adsorbent sampler All
.samples were analyzed by gas chromatography^ass spectrometry.136
In a single test measurement by Battelle of main process vent emissions
from an oil-furnace carbon black plant, an uncontrolled total POM emission
factor of 91 mg/Mg (0.0002 Ib/ton) was determined. No information is
available to define the individual POM species measured or the -techniques
used to sample and analyze for POM. >137
. A summary of available POM emission factors for carbon black production
by the oil- furnace process is given in Table 65.
''• *
Source Loc
In -1985, there were 27 carbon black manufacturing facilities in the
continental United States and one facility in Puerto Rico. Almost
75 percent of all carbon black production occurs in the States of Texas and
Louisiana (41 and 33 percent, respectively). The location of all facilities
^ aimUal Potion capacities in 1985 are provided in
249
-------
TABLE 65. SUMMARY OF POM EMISSION FACTORS FOR
OIL-FURNACE CARBON BLACK PLANTS
Data Sources
References 133 and 134
(Monsanto data)
Total POM Emission
Factor, mg/Mg (lb/ton)a
1200 - 3000 (0.0024 - Q. 0061)
Avg. 1900 (0.0039) >c
Reference 136
(EEA data)
220 - 490 (0.00045 - Q.001)
Avg. 340 (0.0007)
References 134 and 137
(Battelle data)
91 (0.0002)'
^Factors are in terms" of mg POM emissions per Mg of carbon black produced.
^Factors are for uncontrolled emissions from the main process vent.
°2dt12^LSS"?t: °f ?? emiSSi°ns is'*cenaphtylene,"26 percent is pyrene,
and 12 percent is methyl- and dimethylanthracenes/phenanthrenes.
factors are £01- un««n^«n^ emissions. It is'assumed that emissions were
gaseous POM components
A,,**,. v" uncontrolled emissions from the main process vent.
Anthracene, phenanthrene, fluoranthene, pyrene, benzo(a)pyrene benz^
a vi+••>«*•***••«•*** -* — J _•»____ ._ _r-/ * wr^fci^v* \**/ )jy Ldlts f De-tlZ I 3.) "
th«> e costtuentsof al P
Anthracene/phenanthrene constituted 63 percent of total POM pyrene
8r 7 1
250
-------
TABLE 66. LOCATIONS AND ANNUAL CAPACITIES OF
CARBON BLACK PRODUCERS IN 1985138
Company
Ashland Oil, Inc.
Cabot Corporation
Chevron Corporation
Ebonex Corporation
.General Carbon Company
Hoover Color Corporation
J..M. Huber Corporation
Mobay Chemical Corporation
Phillips Petroleum Company
Sid Richardson -Carbon and
Gasoline Company
Union Carbide Corporation
Facility Location
Aransas Pass, TX
Belpre, OH
New Iberia, LA
Franklin, LA
Pampa, TX
Villa Platte, LA
Waverly, WV
Cedar Bayou, TX
Gonroe, TX
El Dorado, AR
Mo j ave, CA
Moundsville, WV
North Bend, LA
Ulysses, KS
Melvindale, MIC
Los Angeles, "CAC
Irvington, NJC
Baytown, TX
Borger, TX
Hawthorne, NJd
Borger, TX
Orange, TX
Addis, LA
Big Spring, TX
Penuelas, PRb
Annual Capacity,
Gg (106 lbs)a
75 (165)
54 (120)
127 (280)
104 (230)
27 (60)
122 (270)
7.3 (160)
9 (20)
52 (115)
49 (108)
24 (54)
73 (162)
110 (242)
38 (84)
5 (10)
<0.5
<0.5
100 (220)
91 (200)
<0.5
125 (275)
61 (135)
61 (120)
50 (110)
4 (8)
251
-------
TABLE 66. LOCATIONS AND ANNUAL CAPACITIES OF
CARBON BLACK PRODUCERS IN 1985138 (Continued)
Company
Witco Chemical. Corporation
Facility Location
Phenix, City, AL
Ponca City, OK
Sunray, TX
TOTAL
Annual Capacity,
Gg (106 lbs)a
25 (55)
57 (125)
45 (100)
1556 (<3431)
Capacities as of January 1, 1985.
fUfii^ d^S n0t ^& the oil-fu'rnace method of carbon black production
acetylene decomposition is used instead" to produce acetylene black.
Carbon black is produced at this facility only for pigment uses.
WOTT** * *PTn •?e>l4**4»^«_^ •»_ • .
num. Jims listing is subject to change as market conditions change
facility ownership changes, plants are closed down, etc. The'reader
should verify the existence of particular facilities by consulting
current listings and/or the plants themselves. The level rfPOM
emissions from any given facility if a function of variables such as
tS^iirocrio^ts^wirpS^^k^should be "•«—
252
-------
SECONDARY LEAD SMELTING
Process Deseripffnrj
One source test of a secondary lead smelting facility has indicated
that secondary lead smelters processing batteries and battery scrap to
recover lead are potential sources of POM emissions.139 . The source of POM
emissions has been theorized to be the polymeric organic casings (plastic
•and rubber) on batteries, which upon combustion in high temperature smelting
furnaces form POM compounds. 9'14°
The secondary lead smelting industry produces lead and lead alloys by
reclaiming lead from scrap. Secondary lead may be refined to produce soft
lead (pure) or alloyed to produce a variety of hard lead alloys. Most of
the lead produced by secondary lead smelters is hard lead used In the
production of lead-acid batteries. Scrap.automobile batteries are the
largest single source of lead-bearing raw material. These batteries contain
approximately 8.2 kg (18 Ibs) of lead per battery consisting of 40 percent
lead alloys and 60 percent lead oxide.
•
Sources of lead-bearing metal used by secondary lead smelters include
scrap batteries from junk dealers, battery plant scrap, and other
miscellaneous scrap. Certain facilities, however, rely exclusively on
non-battery scrap such as wheel balance weights, pipe, solder, drosses, and
lead-sheathed cable. -1
As illustrated in Figure 29, the normal sequence of operations in a
secondary lead smelter are scrap receiving, charge preparation, furnace
smelting, and lead refining and alloying. In the majority of plants, scrap
.batteries are first sawed or broken open to remove the lead alloy plates and
lead oxide paste material. The removal of battery covers is typically
accomplished using an automatic battery feed conveyor system and a slow
speed saw. Hammer mills or other crushing/shredding devices are utilized to
break open battery cases once covers are removed. Float/sink'separation
253
-------
T3
CO
0)
C8
T3
O
U
0)
co
CO
U
•H
CU
4-i
CB
CO
O
CO
^
cu
O.
O
01
U
c
(11
tr
cu
ON
ess
00
254
-------
systems are typically used to separate plastic battery parts, lead
terminals, lead oxide paste, and rubber parts. The majority of lead
smelters recover the crushed plastic materials for recycling. Rubber
casings are usually landfilled. In smelters where battery covers and
casings are removed prior to charging the lead contents into smelting
furnaces,' the potential for POM formation should be greatly reduced Plants
charging whole batteries to smelting furnaces (a minority of -plants) without
any preparation to remove covers and casings would present the greatest
potential for POM emissions from the smelting furnace.
After removing the lead components from the charge batteries, the lead
scrap is combined with other charge materials such as refining drosses flue
dust, furnace slag, coke, limestone, sand, and scrap ion and fed to either a
blast, reverberatory, or rotary smelting furnace. Smelting furnaces are
used to produce crude lead bullion which is refined and/or alloyed into
fznal lead products. Of the 35 existing secondary lead smelters, 17 use a
blast furnace, 8 operate both a blast and reverberatory or rotary furnace 5
operate a reverberatory furnace, and 5 use a rotary furnace.141
A simplified flow diagram of a single secondary lead blast furnace '
system is presented in Figure 30. Blast furnaces'are fueled by coke to
reach smelting operating temperatures of from 430-1320°C (800-2400°F) As
the charge material melts, the iron, silica, and limestone form an
oxidant-retardent flux which floats to the top of the melt. Molten lead
bullion in the bottom of the furnace is tapped almost continuously and cast
into large one ton blocks called buttons or sows. 'Blast furnaces are
operated in both batch and semi-continuous modes. A typical production
range for blast furnaces is 18 to 73 Mg (20 to 80 tons)/day of lead
bullion.
. The emission stream from a blast furnace is typically controlled by
knockout boxes, an afterburner, U-tube coolers, and a baghouse. Knockout
boxes are used to collect large particulate matter, which separates from the
gas flow in the ducts. Afterburners are used to destroy organic emissions
255
-------
"O
to
CO
CO
U
•H
O,
256
-------
U-tube coolers are used to lower the temperature of the gas stream prior to
its ventilation.to' the baghouse to reduce overall particulate matter
emissions. Baghouse dusts are frequently recycled to"the blast furnace.
Smelters using reverberatory furnaces typically have configurations
similar to that shown in Figure 31.141 Reverberatory furnaces are fired
with either gas or oil. Charge materials are heated by radiation from the
burner flames and from the furnace walls. As indicated in Figure 31,
reverberatory furnace charge material typically includes lead scrap,'battery
plates, lead oxides, and recycled flue dusts. Fresh charge material is
added to the furnace as more of the solid material in the furnace becomes
liquid. Material melting in reverberatory furnace takes place at
temperatures of about 1260°C (2300°F) and near atmospheric pressure 141
Molten, semi-soft lead is periodically tapped from the furnace and placed
into molds. Reverberatory furnaces generally produce lead that is purer
than that obtained from, blast furnaces. A typical reverberatory furnace
produces about 45 Mg (50 tons)/day of lead product. Of the total amount of
material input to a reverberatory furnace process, approximately 47 percent
is recovered as lead, 46 percent is slag, and 7 percent leaves the furnace
as particul-ate and metal fume. X
Rotary furnaces in the. secondary lead industry, which are very similar
in operation to reverberatory furnaces, are much less common in the United
States than in industrialized European nations. The rotary furnace is a
batch feed unit that rotates slowly during the heating of the charge
material. One major difference between the rotary furnace and the
reverberatory furnace is that about 70 percent of the sulfur contained in
the rotary furnace charge material is removed in the slag. This is
accomplished by using relatively large amounts of iron (in the form of cast
iron^borings) in the rotary furnace feed. Iron serves the following
distinct purposes.
. - it promotes the reduction of lead sulfate and lead oxide to
metallic lead
257
-------
§
•H
,U
CJ
"O
o
"S
0)
rH
to
-o
§
u
cu
CO
co
>.
CO
0)
u
to
- ^4
cu
•s
OJ
cu
eo
y
•H
O.
ts
H
00
tb
258
-------
it complexes with most of the available sulfur and removes it in
the- slag
Consequently, sulfur dioxide, concentrations in rotary furnace exhausts are
much lower than those at smelters using blast or reverberatory furnaces.141
Rotary furnace charge materials typically consist of lead paste*
batteries, cast iron borings, anthracite, limestone, and soda ash. Natural
gas or fuel oil is used to heat these furnaces to smelting temperatures of
1260 to 1320 C (2300 to 2400°F) . Semi-soft lead product is periodically-
tapped from rotary furnaces and put into molds to await further refining
processes.
Reverberatory and rotary furnaces typically use an emissions control
design consisting of an exhaust gas settling chamber, U-tube coolers and a
baghouse for final emissions control. Dusts collected by the baghouse are
recycled to the furnace as charge material.
As shown in Figure 29, crude lead bullion produced by blast
reverberatory, or rotary furnaces undergoes refining and alloying- processes
to produce final lead products. Refining and alloying processes are
performed in pot furnaces or refining kettles. The process is a batch
operation and may take from a few hours to 2 to 3.. days depending upon the
degree of purity or alloy type required. Refining kettles are gas- or
oal-£lr.d and have' typical capacities of 23 to 136 Mg (25 to 150 tons) of
lead. Refining and alloying activities are conducted at temperatures
ranging from. 320 to 700°C (600 to 1300°F) . tfhen soft lead is desired as a
final product, contaminant elements are removed through careful temperature '
control and by the addition of dross-forming agents. Soft lead refining
typically produces r.fined metal that is greater than 99.97 percent pure
lead.
The production of soft lead typically involves five refining steps In
the first step, the lead is heated to approximately 370 to 510°C (700 to
950 F) which results in the formation of a light dross. This dross consists
259
-------
of lead oxides and other impurities and appears as a dusty black powder.
After removing the light dross, the temperature of the lead is reduced to
300 to 370°C (575 to 700°F) in preparation for the second step, which
involves the addition of sulfur for the removal of copper. After removing
the copper drosses, the kettle temperature is increased to 510 to 650°C (950
to 1200°F) in preparation for the third step. Sodium nitrate is used as a
dressing agent for the purpose of removing tin. The yellow tin drosses are
stored and used as special charge material for the production of high tin
• content blastfurnace metal. The fourth step takes place at 540 to 700°G
(1000 to 1300°F) and involves the addition of sodium nitrate for the removal
of antimony and arsenic drosses. The last step is a final cleaning carried
out by the addition,of sulfur or caustic soda at a lower temperature.141
Hard lead alloys may be classified as either antimonial or
non-antimonial lead alloys, The major alloying agents required in
antimonial lead-are antimony, arsenic, and tin. The production of
antimonial lead alloys generally requires the addition of these elements and
the removal of copper. Alloying agents are generally added at a kettle
temperature of 425 to 480°C (800 to 900°F). The major alloying agent used
in non-antimonial lead alloys is calcium. To produce calcium-lead alloys, '
antimony, arsenic, and tin must first be removed by refining. These
elements, which act as hardening agents in other alloys, are replaced by
copper, "sulfur, and selenium.
Following the final refining step, a sample of the refined metal is
collected, and the alloying specifications are verified by chemical
analysis. When the desired .composition is reached, the molten metal is
pumped from the kettle into the casting machine and cast into lead ingots
which are rectangular bars weighing approximately 25 kg (56 Ibs) each.141
Emission Factors
One set of data have been identified that quantify POM emissions from a
secondary lead smelter processing batteries. Four emission samples have
been obtained from one facility. The data measured were POM concentrations
260
-------
TABLE 67. POM CONCENTRATIONS IN STACK GASES
OF A SECONDARY LEAD SMELTER140 -
POM Compound3
—
Anthracene/phenanthrene
Methyl anthracenes
Fluoranthene
Pyrene
Methyl pyrenes/fluoranthenes
Benzo ( c ) phenanthrene
Chrysene/benz (a) anthracene
Benzo ( a) pyr ene
T~ —
Com
Sample
600
25
160
01
ji
2
10
25
1
™™— — • — . «.
pound Concent:-!-;
1 Sample 2
740
34 '
170
22
2
' 11
23
1
• — —
itions fng/Nr
Sample 3
770
41
330
28
3
17
28
1.
._
B3>b,c.d
Sample 4
940
33
310
30
2
13
25
1
analyzed for.
ontrol device us
wet*
list of just certain coopounds chat »ere
to .easure particulate and
«•»
°°I":rO1
-------
in the stack gases following the final control device. These concentrations
are given in Table 67. ° The predominant POM compounds measured were
anthracene/phenanthrene and fluoranthene. BenZo(a)pyrene was measured but
its levels were only 0.1 percent of the anthracene/phenanthrene levels. The
sampling and analysis procedures used during the tests of the secondary lead
smelter contained mechanisms to capture and measure both particulate and
vaporous POM. The majority of the POM measured was caught in the
sampling train's water impingers.14°
The consistency of the POM data across all samples is atypical for POM
emission sources. Both the results for samples taken on the same day and
the total results for all samples demonstrate a reproducibility that is
unexpected. The relative magnitudes of each POM compound measured are
consistent from sample 1 to sample 4.
Source Ideations
As of May. 1985, the secondary lead smelting industry consisted of
43 facilities operated by 27 companies. A list of the 43 facilities and
their locations is given in Table 68.141
PRIMARY ALUMINUM PRODUCTION
Process Description
•
All primary aluminum in the United States is produced by the
electrolytic reduction of alumina otherwise known as the Hall-Heroult
process. The general procedures for primary aluminum reduction are
illustrated in Figure 32. Aluminum reduction is carried out in shallow
rectangular cells (pots) made of carbon-lined steel with carbon blocks that
are suspended above and extend down into the pot (Figure 33). The pots and
carbon blocks serve as cathodes and anodes, respectively, for the
electrolytical process.142"144
262
-------
TABLE 68. SECONDARY LEAD SMELTERS IN THE
.141
UNITED STATES IN 1985'
State
Alabama
California
Florida
Georgia
Illinois
Indiana
Kansas
Louisiana
Minnesota
Missouri
New Jersey
New York
City
Leeds
Troy
Anaheim
City of Industry
Gardena
San Francisco
Vernon
Tampa
Tampa
Atlanta
Atlanta
Columbus
Fitzgerald
Chicago
Granite City
Beech Grove
East Chicago
Indianapolis
Muncie
Olathe
Baton Rouge
Eagan
Forest City
New Brunswick
Newark
Pedricktown
East Syracuse
Middletown
Company
"winterstate Lead Company, Inc.
Sanders Lead Company
Delco.Remy
Quemetco/RSR Corporation
Alco Pacific
Federated Metals/ASARCO
GNB Batteries/Gould, Inc.
Chloride. Metals
Gulf Coast Lead
National Smelting
Seitzinger/Taracorp
Chloride Metals
Delco Remy
Inland Metals
Taracorp
Refined Metals
U. S. S. Lead
Quemetco/RSR Corporation
Delco Remy
Delco Remy
Schuylkill Metals
Gopher Smelting
Schuylkill Metals
Delco Remy
Federated Metals/ASARCO
National Smelting
Roth Brothers Smelting
Revere Smelting and Refining
Corporation of New Jersey/
RSR Corporation
263
-------
TABLE 68. SECONDARY LEAD SMELTERS IN THE
UNITED STATES IN 1985141 (Continued)
State
Ohio
Oregon
Pennsylvania
Tennessee
Texas
Virginia
Washington
NOTE:
City
Cleveland
St. Helens
Lancaster
Lyon Station
Nesquehoning
Reading
College Grove
Memphis
Rossville
Dallas
Dallas
Frisco
San Antonio
Richmond
•
Seattle
Company
Master Metals
Bergsoe Metal
Lancaster Battery Company
East Penn Manufacturing
Tonolli Corporation
General Battery Corporation
General Smelting and Refining
Refined Metals
Ross Metals, Inc.
Dixie Metals/General Battery
Corporation
Murph Metals/RSR Corporation
GNB Batteries/Gould, Inc.
Standard Industries
Hyman Viener and Sons
Quemetco/RSR Corporation
264
-------
I—
OVERSIZE
BALL
MILL
1
VIBRATING
SCREEN
T"1
BALL
MILL
OVERSIZE
i
(
AIR
CLASSIFIER
f
' PH
BIN
POM
EMISSIONS
POM" '
EMISSIONS
NEEDED FOR PflEBAKED
PROCESS ONLY
Figure 32. General flow diagram for primary aluminum
production.142
265
-------
sa
CO
o
3
•O'
01
C
<
oo
60
•H
fo
266
-------
Cryolite, a double fluoride salt of sodium and aluminum (Na A1F )
serves as an electrolyte and a solvent for alumina. Alumina is added to and
dissolves in the molten cryolite bath. The cells are heated and operated
between 950 and 1000°C (1742 to 1832°F) with heat that results from
. resistance between the electrodes. During the reduction process, the
aluminum is deposited at the cathode where, because of its heavier weight
(2.3 g/cm versus 2.1 g/cm3), it remains as a molten metal layer underneath
the cryolite. The cryolite bath thus also protects the aluminum from the
atmosphere. The byproduct oxygen migrates to and combines with the
consumable carbon anode to form carbon dioxide and carbon monoxide, which
continually evolve from the cell. The basic reaction of the reduction
process is:
A12°3
r-5c — >
1.5C0
Alumina and cryolite are periodically added to the bath -to replenish '
material that is removed or consumed in normal operation. The weight ratio
of sodium fluoride (NaF) to aluminum fluoride .(Air ) in cryolite is l 5
Fluorspar (calcium fluoride) may also be added to lower the bath melting
point. &'
Periodically, the molten aluminum is siphoned or tapped from beneath
the cryolite bath, moved in the molten state to holding furnaces in the
casting area, and fluxed to remove trace impurities. The product aluminum
is later tapped from the holding furnaces and cast into ingots or billets to
await further processing or shipped molten in insulated ladles.142
The process of primary aluminum reaction is essentially one of
materials handling. The true difference in the various process
modifications used by the industry lies in the type of reduction cell used
Three types of reduction cells or pots are used in the United States-
prebake, horizontal stud Soderberg, and vertical stud Soderberg. Prebake
cells constitute the bulk of aluminum production (66 percent), foUowed by
horxzontal Soderberg (21 percent), and vertical Soderberg (13 percent)
267
-------
Both Soderberg cells employ continuously formed consumable carbon anodes
where the anode paste is baked by the energy of the reduction cell itself.
The prebake cell, as indicated by its name, employs a replaceable,
. consumable carbon anode, formed by baking in a separate facility called an
anode bake plant, prior to its use in the cell.
The preparation and operation of the aluminum reduction cells is the
source of potential POM emissions from primary aluminum production. The
magnitude of POM emissions from a typical reduction plant is a function of
the type of reduction cell used.143 Prebaked cell anodes are made by curing
- carbon contained in pitch and coke at relatively high temperatures [~1100°C
(2012 F)]. A flow diagram depicting the production of prebaked cells is
given in Figure 34. The high temperature curing process can potentially
generate POM compounds.
Potentially, POM compounds can be emitted from the prebake cell during
the reduction process when the anodes are lowered into the reduction pot.
However, POM emissions from reduction are expected to be less than that from
prebake cell preparation because the majority of POM emissions have already
been released during the high temperature curing operation.142'144 Data in
Reference 144 support this theory.
Soderberg cell anodes are continuously lowered and baked by conductive
heat from the. molten alumina bath rather than being premolded and baked. A
coke and coal tar pitch paste is packed into a metal shell over the bath.
As the baked anode at the bottom of the shell is consumed, more paste is
added at the top of the shell. As the paste is consumed,. POM emissions are
potentially released. Since the carbon paste is not baked prior to being
placed in the pot, POM emissions from a Soderberg cell (horizontal or
vertical stud) reduction operation would have the potential to be much
greater than those from a prebaked cell reduction operation.
Emissions control at primary aluminum reduction facilities (cell rooms)
is intended primarily for fluoride removal and involves efficient emissions
capture and removal. Emissions capture is generally accomplished by using
268
-------
COAL TAR PITCH
CRUSHER
CALCINED
PETROLEUM u
COKE
TO POTLINE
STACK
Figure 34. Flow diagram for the production of prebake anodes.142
269
-------
precisely designed hooding and ducting systems on reduction cells. The term
hooding includes the use of classical draft hoods and the use of movable
doors, enclosures, and skirts.. Primary emissions removal is achieved
through the use of dry scrubbing systems or wet scrubber/electrostatic
precipitator combination systems. Two types of dry scrubbing systems,
fluidized bed and injected alumina, are found in that industry. Both forms
of dry scrubbers contain baghouse equipment to collect particulate matter
from the chemical absorption scrubbing process. These baghouses would be
effective in removing particulate POM in the emission stream. Standard
design spray tower wet scrubbers and wet electrostatic precipitators used in
the series are also effective primary control systems at aluminum reduction
facilities. The ability of the combination scrubbing/precipitation system
to remove particulate POM should be equal to that of the dry scrubbing
baghouses, and may exceed it because of the combination system's ability to
control some vaporous POM compounds through gas cooling and subsequent
particulate collection. In addition to being a primary control system for
cell room emissions, the wet scrubber/ESP combination is also used as the
primary control system for facilities producing prebake anodes.142
In some primary aluminum installations, secondary control systems'are
also used'to augment the primary systems. The predominant secondary system
is a spray screen scrubber followed by a mist eliminator. The term spray
screen scrubber is applied to wet scrubbers in which the scrubbing liquor is
sprayed into a gas stream and on to screens or. open mesh filters enclosed in
a plenum chamber.
Emission Factors
One set of POM emissions data have been identified for aluminum
smelting plants. Emissions of particulate and vaporous POM have been
measured from an anode paste preparation process, a horizontal Soderberg
reduction cell, and a vertical Soderberg reduction cell at a Swedish
aluminum smelter. The results of these source tests, as reported in the
literature, are given in Table 69. As would be expected", the anode paste
preparation process emissions have greater POM concentrations than either of
270
-------
I" (
Cu ea « f
— s-> 3
as u
•w S >»
H -2 .S1
« «
S 41 41
8 S™
u tS .2
.=
S 8
B S
O 41 41
ft* > =.
I
I
2
o
"8
E
03
O ^
ca s
2 S
41 u
»- o
41 — o B
U O •O —
B U 41 >,
O w u £
u B a. u
O 41
"a
a u e
* g .2
o —•
O.J!
•a — ^ •• — ^ >*
5 = g^ 3^ S
271
-------
Che Soderberg processes. The test data in Table 69 illustrate that the
majority of the POM emissions from the anode past process were particulate
matter and not vapors. The POM emissions of the horizontal Soderberg
process exhibit a similar behavior as evidenced by the Table 69 data.
Conversely, POM emissions from the vertical Soderberg process were
predominantly in vapor form instead of particulate.
145
A list of all the POM compounds identified in emissions from the
aluminum smelter is provided in Table 70.145
Source Locations
« •
As of January 1985, there were 28 primary aluminum reduction plants in
the United States operated by 10 different companies. Washington State has
seven plants, the most of any State in the country. A complete list of all
29 facilities is given in Table 71.146
WOOD CHARCOAL- PRODUCTION
Process Descri-pfrinr.
Charcoal, primarily used for outdoor cooking, is manufactured by the
pyrolysis of carbonaceous raw materials, primarily medium to dense hardwoods
such as beech, birch, maple, hickory, and oak. Softwoods, sawdust,
nutshells, fruit pits, and vegetable wastes are also used in the pyrolysis
process. The high temperature (450 to 510°C) pyrolysis of wood materials is
a potential means of generating POM air emissions.147
Hardwood charcoal .is manufactured by a four-step pyrolysis-process.
Heat is applied to the wood, and as the temperature rises to 100°C (212°F),
water and highly volatile hydrocarbons are distilled off. The wood
temperature remains at approximately 100°C until the moisture content of the
wood has been removed, at which time the volume of distillate production
declines and the wood temperature begins to climb. During the next stage
272
-------
TABLE 70.
POM COMPOUNDS IDENTIFIED IN THE EMISSIONS
OF A PRIMARY ALUMINUM SMELTER145
POM Compounds Measured in Aluminum Smelter Emissions'
1.
2.
3.
4.
5.
' 6.
7.
8.
9.
10.
11.
12.
'13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Naphthalene
2-Methylnaphthalene
1-Methylnaphthalene
Biphenyl
Acenaphthylene
Acenaphthene
Fluorene
2-Methylfluorene
1-Methylfluorene
Phenanthrene
Anthracene
3-Methylphenanthrene
2-Me thyIphenanthrene
2-Methylanthracene
4,5-Methylenephenanthrene
4- and/or 9-Methylphenanthrene
1-Methylphenanthrene
Fluoranthene
Benz(e)acenaphthylene
Pyrene
Ethylmethylenephenanthrene
Benzo(a)fluorene
Benzo(b)fluorene
4-Methylpyrene .
2-Methylpyrene and/or methylfluoranthene
1-Methylpyrene
Benzo(c)phenanthrene
Benz(a)anthracene
273
-------
TABLE 70. POM COMPOUNDS IDENTIFIED IN THE EMISSIONS
,145
OF A PRIMARY ALUMINUM SMELTER143 (Continued)
POM Compounds Measured in Aluminum Smelter Emissions'
29
30.
31.
32.
33.
34.
35.
36.
37.
Chrysene and triphenylene
Benzo (b ) f luoranthene--
Benzo(j+k)fluoranthene
Benzo(e)pyrene
Benzo(a)pyrene
Perylene
Indeno(1,2,3-c,d)pyrene
Dibenz(a,c and/or a,h)anthracenes
Benzo(g,h,i)perylene
™ includ*d anode P*ste Preparation, a horizontal Soderberg
pot. room, and a vertical Soderberg pot room.
274
-------
TABLE 71. LIST OF PRIMARY ALUMINUM PRODUCTION FACILITIES
IN THE UNITED STATES IN 1985146 -
Facility
Alumax, Inc.
Aluminum Company of America
Atlantic Richfield- Company
Consolidated Aluminum Corporation
Kaiser Aluminum and Chemical Corporation
Martin Marietta Corporation
National-Southwire Aluminum Company
Noranda Aluminum, Inc.
Ormet Corporation
Reynolds Metals Company
Location
Mount Holly, SC
Frederick, MD
Ferndale, WA
Alcoa, TN
Badin, NC
Massena, NY
Rockdale, TX
Vancouver, WA
Warrick County, IN
Wenatchee, WA
Columbia Falls, MT
Sebree, KY
New Johnsonville, TN
Chalmette, LA
Mead, WA ' -
Ravenswood, WV
Tacoma, WA
The Dalles, OR
Goldendale, WA
Hawesville, KY
New Madrid, MO
Hannibal, OH
Arkadelphia, AR
Jones Mill, AR
Longview, WA
Massena, NY
Sheffield, AL
Troutdale, OR
NOTE:
275
-------
the wood temperature rises with heat input to approximately 275°C (527°F),
and hydrocarbon distillate yield increases. 'As the third stage begins in'
the vicinity of 275°C, external application of heat is no longer required
since the carbonization reactions become exothermic. During this stage, the
wood temperature rises to 350°C (662°F), and the bulk of hydrocarbon
distillates are produced. At "approximately 350°C, exothermic pyrolysis
ends, and during the final stage, heat is again applied, raising the wood
temperature to 400 to 500°G (752 to 932°F) to remove more of the less
volatile, tarry materials from the product charcoal.
Currently, there are predominantly two types of vessels used to
manufacture wood charcoal, the Missouri-type 'batch kiln and the continuous
Herreshoff furnace. The batch process and kiln account for about 45 percent
of national wood charcoal production. The Missouri-type kiln shown in
Figure 35 is typically constructed of concrete.148 A Missouri-type batch
kiln normally processes about 45 to 50 cords of wood in a 10- to 25-day
cycle. A typical cycle may be structured as follows.
1
5
10
1
- 2 days
8 days'
- 14 days
- 2 days
load wood
pyrolysis
cool
unload charcoal
After the wood is manually loaded in the. kiln, a fire is started
usually at the bottom center of the kiln, by igniting easily combustible
materials placed at this point during the loading. During ignition, a large
amount of air is necessary for the rapid combustion of the starting fuels to
insure the heat level needed for pyrolysis. This air is supplied through
groundline ports in the kiln side walls or through temporary openings under
the kiln door. In some cases, the kiln doors remain open until the burn is
adequately started. Auxiliary ceiling ports in some kilns serve as
temporary stacks and aid ignition by causing greater amounts of air to be
drawn into the kiln through the air ports. They also aid removal of smoke
from the kiln. ** • '
. 276
-------
CO
-a-
ir-i
C
r-t
•H
O
a
M
.0!
CJ
cu
c.
o
en
01
oo
277
-------
Ignition patterns are generally similar for all types of kilns. During
the first 5 to 15 minutes, temperatures in the ignition area will rise
rapidly to about 540°C (1004°F). After much of the fuel has been burned,
the temperatures will quickly drop, often to as low as 150°C (302°F). The
extent of the temperature drop is closely related to conditions of air
supply and to the moisture content of the charge. ' With the establishment of
a suitable ignition zone, however, the temperature gradually increases to
about 280°C (536°F), and the ignition period is considered complete.148
Satisfactory carbonization depends primarily on the maintenance of
proper burning conditions in the pyrolysis zone. Sufficient heat must be
generated first to dry the wood and then to maintain temperatures necessary
for .fttclant carbonization. At the same time, the burning must be limited
so that only sufficient heat is present, to produce good charcoal.
Temperature control is attained by varying the size of'the air port openings
providing air for combustion of wood volatiles.148 '
For the production of good-quality charcoal, kiln temperatures from
about 450 to 510°C (842 to 950°F) are required. Prolonged higher
temperatures will reduce the yield of charcoal without necessarily upgrading
it for recreational use. If, on the other hand, pyrolysis temperatures
remain low, the charcoal may be too smoky for domestic use, and larger than
normal amounts of brands (partially charred wood) will be produced.148
The direction and rate of spread of the pyrolysis zone is associated
with a number of factors, such as location of air ports and stacks, volume
and velocity of the incoming air, wood size and moisture content, piling of
the charge, and design of the kiln. Pyrolysis generally proceeds at a faster
rate at the upper part of the charge, where higher temperatures are
available for longer periods of time. Less rapid pyrolysis takes place near
the kiln floor, where the average temperature usually is lowest. In the
Missouri-type kiln, combustion and carbonization progresses from the top of
the kiln to the floor and from the center to the walls.148
278
-------
Burn progress can be determined by the color of the smoke from the kiln
or by determining the temperature along the vertical distance of the steel
doors. The pyrolysis is completed when fire has reached the floor of the'
kiln as determined by view ports (air intake ports) at the floor level.
This may also be indicated by a marked decrease in the volume of smoke'and a
color change from grayish yellow to bluish white.148
When pyrolysis has been completed, all air ports are sealed for the
start of the cooling cycle. After the ports are sealed, the stacks remain
open until smoking has practically stopped to prevent "the development of -gas
pressure in the kiln. Stacks can usually be sealed from 1 to 2 hours after
the air ports are closed. The kiln is allowed to cool for about 10 to
14 days, before removing the charcoal. Yields of approximately 25 percent
are achieved.
The required pyrolysis time and resultant POM emissions from a
Missouri-type batch kiln vary with kiln capacity, operational practices
wood type, and wood moisture content. Process.reaction gases containing - POM
are -hausted^from the kiln in stacks that run along the side walls of the
vessel. • The charcoal product of a batch kiln processes either sold
directly o-r made into briquettes prior to selling. -
Continuous charcoal production is accomplished in Herreshoff multiple
hearth furnaces. The use of continuous multiple hearth units for charcoal
-production has increased because of the following advantages of the units.
• Lower labor requirements than kiln operations where manual loading
and unloading is needed. Only one man per shift is required for
continuous facilities.
• Consistent yield and quality charcoal with easy control of product
volatile and fixed carbon content.
• Feed of multiple forms of wood waste.
• Off-gases easily collected for further processing.
279
-------
The typical feedstock capacity of continuous wood charcoal furnaces is
2.5 Mg (2.75 tons)/hour.
The operational" principles of the Herreshoff furnace (shown in
Figure 36) are relatively simple. Passing up through the center of the
furnace is a shaft to which are attached two to four rabble arms for each
hearth. As the shaft turns, the hogged wood material resting on the hearth
floors is continually agitated, exposing fresh material to the hot gases
being evolved. A further function of the rabble arms is to move material
through the furnace. On alternate hearths, the teeth are canted to spiral
the material from the shaft toward the outside wall of the furnace or from
the outside wall toward the center shaft. Around the center shaft is an
annular space through which material drops on alternate hearths, while on
the remaining hearths material drops through holes in the outer periphery of
the hearth floor. In this way, material fed at the top of the furnace moves
alternately across the hearths at increasing temperatures until it
discharges from the floor of the bottom hearth. Charcoal exiting from the
furnace is cooled by water sprays and water jacketing on a cooler. These
sprays are controlled automatically by a temperature regulator set for a
given charcoal temperature. As with batch kilns, the charcoal product of
continuous kilns is either sold directly or further processed to briquettes
for sale.148
Initial heat for startup is provided by oil- or gas-fired burners
mounted in the sides of the hearths. When furnace temperature has been
attained, the auxiliary fuel ceases, and combustion air is used to ignite
the evolving wood gases to maintain furnace temperature. Furnace
temperatures range between 480 and 650°C (896 to 1202°F). Exhaust gases
from the charcoal production process are vented to the atmosphere or to
controls through stacks located on top of the furnace,' are used as a heat
source for predrying of feed material and drying of briquettes produced at
an adjacent vessel, or are burned in a waste heat boiler to produce
steam. '
280
-------
POM Emissions
ooling Air Discharge
Floating Damper
,—. /. Feed Material
Drying Zone
Combustion Zone
Cooling Zone
Product
Chareo
Cooling Air Fan'
•Rabble Arm at
Each Hearth
.Combustion
Air Return
Rabble Arm
Drive
Figure 36. Multiple-hearth furnace for charcoal production.148
'281
-------
A 1978 U. S. EPA investigation into wood charcoal production indicated
that many of the batch kilns are relatively old and many, particularly
smaller kilns, are uncontrolled.147'148 in general, the control of
emissions, including POM, from batch wood'charcoal kilns is complicated by
the cyclical nature of the process. Throughout the cycle, both emission
composition and flow rate change. Direct-fired afterburners for the
destruction 'of hydrocarbons have been suggested to be the most feasible
control system; however, these devices would require an auxiliary fuel such
as natural gas. Economic analyses have indicated that for typical batch1
kilns, the operation of afterburners for emissions control would cause firms
to lose money. With the combustion of auxiliary fuel of any type,
a. potential is also created for additional POM emissions. No information is
available on the proportion of batch kilns with afterburner controls or the
effect of afterburner use on POM emissions.147
Continuous wood charcoal furnaces are predominantly controlled by
direct-fired afterburners.147'148 Auxiliary fuel firing is required in
continuous furnace afterburners only during start-up or process upset's
because of the generally higher heating value of continuous furnace exhaust
gases. One facility has been found to be using an incinerator to control
furnace emissions.148
Emission Factors
Polycyclic organic matter emission factor data are available in the
literature only for a Missouri-type batch kiln.147 Five sampling runs were
made and total uncontrolled POM emissions averaged 3.5 g/Mg (0.007 Ib/ton)
of charcoal produced. Reference 147 indicates that the POM samples from
these tests were obtained using a modified Method 5 procedure and sample
analysis was performed by gas chromatography. Benz(c)phenanthrene and
benzo(a)pyrene were identified as constituents of total POM emissions. Four
other POM compounds, dibenz(a,h)anthracene, 3-dimethylcholanthrene,
7,12-dimethylbenz(a)anthracene, and 3,4,5,6-dibenzocarbazole, were
specifically analyzed for but were not detected in any of the samples.147
282
-------
The authors of Reference 147 noted that the results of the batch-kiln
emission tests might be of questionable value due to the difficulty of
sampling the kiln and "the improvisational sampling techniques" used. No
estimate of the accuracy of the test results was provided.
SpurceLocations
Wood charcoal manufacturing facilities are located in 24 States
primarily in Missouri, Arkansas, and in several southeastern States.
list of wood charcoal producers in the United States is provided in
Table 72.148
148
CREOSOTE WOOD TREATMENT
Process Description
Creosote impregnation plants, also called wood treatment plants, have
been, identified as potential air emission sources of POM because creosote
contains significant quantities of POM compounds. Creosote is a product of
the fractional distillation of coal tar, which is a byproduct of bituminous
coal coking. The principal use of creosote is as a wood preservative. " It
is used to treat crossties, switch ties, utility poles, crossarms, marine
and foundation pilings, construction lumber, fence posts, and plywood.144
Treatment is accomplished by either pressure or non-pressure processes
To initiate either process, wood products are debarked and conditioned.
Conditioning, primarily moisture removal, is performed by air seasoning or
kiln drying in the majority of plants. Depending on the particular
preservative to be applied, conditioning may also-be performed by steaming
the wood in the treatment retort, heating the wood in oil under reduced
pressure, or exposing it to hot vapors of organic solvents (vapor drying)
To expedite certain treatment processes, the wood may be pierced by knives
(a process called incising) to provide avenues for penetration of the
preservative solutions.149
283
-------
TABLE 72. CHARCOAL PRODUCERS IN THE UNITED STATES148'3
State
City or County
Producer
Alabama
Arkansas
California
Florida
Georgia
Kansas
I
Kentucky
Illinois
Maryland
Minnesota
Mississippi
Dothan
Tuscumbia
Muscle Shoals
Jasper
Huntsville
Omaha
Green Forest
Yellville
Paris
Scranton
Waldron
Harrison
Paris
Paris
Hot Springs
George
Hatfield
Waldron
Mountain View
Elk Grove
Santa Clara
Milipitas
Ocala
Atlanta
Chetopa
Burns ide
Chicago
White Church
Oakland
Isanti
Bruce
Pachuta
Pachuta
Beaumont
Kingsford Company
Malone Charcoal Company
McKinney Lumber and Plywood
Jasper Charcoal Company
Keeter Charcoal Company
Keeter Charcoal Company
Keeter Charcoal Company
Martin Charcoal Company
Ozark Charcoal Company
Scranton Charcoal Company
Waldron Charcoal Company
Newberry Charcoal Company
Paris Charcoal Company
Arkansas Charcoal Company
Weyerhaeuser Company
George Charcoal Company
Arkansas Charcoal Company
Waldron Charcoal Company
Hinesley and Everett Enterprises
C. B. Hobbs Corporation
C. B. Hobbs Corporation
C. B. Hobbs Corporation
Pioneer Charcoal
Husky Industries
Jayhawk Charcoal Company
Kingsford Company
Great Lakes Carbon Corporation
Kingsford Company
Kingsford Company
Husky Briquetting, Inc.
Blackjack Charcoal Company
Hood Charcoal Company
Masonite Corporation,
Charcoal Division
Ronnies Hickory Chips
In 1978
-------
TABLE 72. CHARCOAL PRODUCERS IN THE UNITED STATES148 (Continued)
State
City or County
Producer
Missouri
Barry
Purdy
Boone
Centralia
Carter
Carter
Ellsinore
Ellsinore
Van Buren
Van Buren .
Cole
Henley
Jefferson City
Steelville
Wesco
Greenfield
Salem
Salem
Salem
Dent
Dent
Salem
Salem
Gasconade
Owensville
Wheatland
• Howell
Mount View
West Plains
Mount View
Peace Valley
Mount View
Mount View
Kansas City
Hocomo
Laclede
Laclede
Vienna
High Gate
Belle
Belle
Belle
Hayden
Iberia
Harris Enterprises
Heaser Charcoal Company
Charles Chrisman Charcoal
L and A Dailing Charcoal Company
Big Springs Industrial
: Carter County Charcoal
Leach Brothers Charcoal
Rozark .Farms
Big Springs Charcoal
Big Springs Charcoal
Stegeman Charcoal Company
Louis Stegeman Charcoal Company
Rich Stegeman Charcoal Company
Hardwood Charcoal Company
Fordell Development Corporation
Pringle Charcoal Company
Carty Charcoal
Floyd Charcoal Company
C and H Charcoal
Langworthy Charcoal Company
Lennox Charcoal Company
Wieberg Charcoal Company
Hobson Charcoal Company
Hickory Charcoal Company
Gene' is Charcoal
J and E Charcoal Company
Missouri Charcoal Company
Craig Charcoal Company
Nubbin Ridge Charcoal Company
Bays Sawmill and Charcoal
Peace Valley Kilns
Old Hickory Charcoal Company
Carr Forest Products
Standard Milling Company
Bakersfield Charcoal Company
Independent S tave Company
Timber Products Company
Wulff Charcoal Company
Kingsford Company
Kingsford Company
W. B. Stockton
H and D Charcoal
Curtis.and Hayes Charcoal
Louis Stegeman Charcoal
285
-------
148
TABLE 72. CHARCOAL PRODUCERS IN THE UNITED STATES148 (Continued)
State
City or County
Producer
Missouri
(continued)
New Jersey
Miller
St. Elizabeth
Neosho
Oregon
Meta
Osage
,0s age
Freeburg
Osage
Qsage
Meta
Osage
Freeburg
Freeburg
Meta
Belle
'Meta
Gainesville
Ozark
St. James
Lake Spring
Vienna
Lesterville
Reynolds
Winona
Shannon
Shannon
Birch Tree
Summersville
Round Springs
Round Springs
Round Springs
Gladden
Branson
Bradleyville
Branson
Raymohdville
Licking
Plato
Seymour
St. Louis
Teterboro
Kalaf Charcoal
Kirkweg Charcoal Company
Neosho Charcoal Products
Greer Springs Company
Barnhart Charcoal
J and M Charcoal Company
Kelly Charcoal Company
Al Luecke Charcoal Company
McDonald Charcoal Company
Ridenhour Charcoal Company
Ripka Charcoal and Lumber
Sugar Creek Charcoal- Company
Wieberg Charcoal Company
Ben Berhorst
Charkol, Inc.
Gene Noblett Charcoal Company
Standard Milling Company
Ozark Forest Charcoal
Wallace <3harcoal Company
Parry Charcoal Company
Lenox Charcoal
Tackett Charcoal Company
Black River Charcoal Company
Copeland Charcoal Company
Dailey Charcoal
George Helmuth Charcoal
Royal Forest Charcoal
Kerr Charcoal
Craig Charcoal
Roaring Springs Corporation
Round Springs Charcoal
Robert Hamilton
Timber Charcoal Company
S and S Charcoal Company
Homer Charcoal Company
Keeter Charcoal Company
Thomason Charcoal Company
Wulff Charcoal Company
H. 0. Charcoal Company
Oak-lite Corporation
Cupples Company, Manufacturers
Degussa, Inc.
286
-------
TABLE 72. CHARCOAL PRODUCERS IN THE UNITED STATES148 (Continued)
State
City or.County
Producer
North Dakota
Ohio
Oklahoma
Oregon
Penrisylvania
South Carolina
Tennessee
Texas
Virginia
West Virginia
Dickinson
Oak Hill
Lucas
West Marion
McArthur
Heavener
Talihina
Clayton
Talihina
Bull Hollow
Springfield
White City
Brookville
Lake City
Jamestown
Red Bank
Cookeville
Tullahoma"
Red Boiling Springs
Spencer
Memphis
Lynchburg
Flatonia
Houston
Jacksonville
Jacksonville
San Antonio
Kenbridge
Belington
Beryl
Maysville
Parsons
Swiss
Bentree
Husky Industries
Victory Charcoal Company
Sun Oil Company
Great Lakes Carbon
Roseville Charcoal
Forest Products Charcoal Company
Forest Products Charcoal Company
Forest Products Charcoal Company
Talihina Charcoal Company
Cherokee Forest Industries
Kingsford Company
Georgia Pacific Corporation
Humphrey Charcoal
T. S. Ragsdale Company, Inc.
Royal Oak Charcoal Company
Cumberland Kingsford
Royal Oak Charcoal Company
Tennessee Dickel Distilling
Cumberland Charcoal Corporation
Royal Oak Charcoal Company
Arkansas Charcoal Company
Jack Daniels Distillery
B and B Charcoal
Pine-0-Pine Company
Campfire Charcoal Company
Char Time Charcoal
National Charcoal Company
Imperial Briquet Corporation
Kingsford
Kingsford
Kingsford
Kingsford
Roseville
Roseville
Charcoal'
Charcoal
Charcoal
Charcoal
Charcoal
Charcoal
287
-------
TABLE 72. -CHARCOAL PRODUCERS IN THE UNITED STATES148 (Continued)
State
City or County
Producer
Wisconsin
Hixton
Husky Industries
NOTE:
This listing is subject to change as market conditions change,
facility ownership changes, plants are closed down, etc. The reader
should verify the existence of particular facilities by consulting
current listings and/or the plants themselves. The level of POM
emissions from any given facility is a function of variables such as
capacity, throughput, and control measures, and should be determined
through direct contacts with plant personnel.
288
-------
Pressurized processes are used to preserve 95 percent of all treated
wood. These processes involve the application of pneumatic or hydrostatic
pressure to expedite the movement of preservative liquid into wood. In the
normal application of preservatives (e.g., creosote), wood is first loaded
on trams and introduced into the pressure vessel. Once in the pressure
vessel, wood can be creosote pressure treated by either the full-cell or the
empty-cell process.
In the full-cell process, an initial vacuum is. applied to the charge
for a period of about 30 minutes. At the end of this period, and while
still maintaining the vacuum, the vessel is filled with creosote. The
vacuum is then released and pressures of 50 to 250 psi are applied to the
system. Pressure is maintained until the required gross absorption of
preservative has been achieved. At the end of the pressure cycle, the
pressure is reduced to atmospheric levels and the preservative liquid in the
vessel is returned to storage. The treated wood will often be subjected to
a final vacuum to remove excess preservative on the surface of the wood.
Once completed, the vacuum is released, the door of the vessel is opened,
and the treated stock is removed. Creosote retentions achieved by the
full-cell process vary from 320 to 480 kg/m3 (20 to 30 lbs/ft3).149
In the empty-cell process, the treatment retort is filled with
preservative while either at" ambient pressure conditions or under an initial
air pressure of 15 to 75 psi. The remainder of the treating process is the
same as that described for the full-cell process. Depending on the
specifications of the customer, wood preservative retentions achieved by the
empty-cell process range from 96 to 208 kg/m3 (6 to 12 lbs/ft3).149
In both the full-cell and empty-cell processes, creosote may be applied
in an undiluted form or.diluted with coal tar or petroleum. Temperatures of
application for creosote and its solutions range from 99 to 110°C (210 to
230°F).
289
-------
Products such as marine pilings are always treated by the full-cell
process. Utility poles, crossties, and fence posts are routinely treated by
the empty-cell process. The amount of preservative retention needed and the
treatment process required are determined by the biological hazard to which
the treated, wood will be subjected in service.
Non-pressurized wood treatment processes are used both commercially and
by individual consumers for home, farm, and garden wood preservation.
Generally, wood treated by non-pressure processes must be seasoned to a
moisture content of 30 percent or less prior to treatment to provide the •
149 •
best results.
Most commercial non-pressure creosote treatments are applied by
cold-soak or thermal processes. In both processes, wood is exposed to the
preservative in an open vessel. The principle behind the cold-soak process
simply entails soaking seasoned wood in the preservative for a fixed period
of time, or until a predetermined gross retention has been achieved. The
thermal process involves exposing wood to hot creosote for 6 to 12 hours
followed by exposure to the preservative at ambient temperature for
2 hours.149
Home and other non-commercial creosote treatments are typically
performed by brush, dip, or spray methods. In these cases, creosote or
creosote-based solutions are manually applied at ambient conditions to wood
and allowed to dry. The amount of retention that is achieved is a function
of wood type, wood moisture content, and wood porosity.149
The creosote wood treatment source category appears to a source of
primarily fugitive POM emissions that are associated with the actual
treatment process and the handling of creosote raw materials and treated
products. Fugitive emissions from treatment occur when the treatment vessel
is opened at the end of the cycle. The duration of such emissions from each
vessel is relatively short because vessels are only opened once or twice
during each working shift.144'150
290
-------
A second source of fugitive POM emissions is during creosote transfer
from an incoming tanker or rail car to plant storage facilities. The method
and frequency of delivery is. a function of plant size and location.
Generally, the larger the facility the more and greater the creosote loads
will be. Increased frequency and quantity means increased potential for
emissions. Transfer of the preservative, whether from rail car or tanker, is
normally accomplished using a closed piping system. In such a system, the
greatest chance for fugitive emissions is at the origin where creosote is
leaving the tanker or rail car and at the end of the transfer where creosote
is entering the storage vessel.144'150
A third potential source of fugitive emissions of POM compounds from
creosote wood treatment plants is evaporative losses from treated wood. If
treated products are stored in a building, emissions of this type would be
largely confined and would not be released to the outside air.144'150
No information concerning currently used or potential control equipment
for POM emissions from creosote treatment processes was identified in .the
literature. •
Emission Factors
No POM emissions or emission factor data were found in the literature
for creosote impregnation plants. The existence of POM emissions in these
facilities has been indicated by area samples of air in and around the
plants and by personal breathing zone air samples that contained POM
compounds. These samples were taken to assess worker exposure to POM
compounds. In one facility, worker breathing zone samples had a
benzo(a)pyrene range of 0.80 to 84 ug/m3.144 An area sample at the
impregnation vessel had a benzo(a)pyrene content of 3.6 ug/m3.144 Both of
these benzo(a)pyrene data points represent collected particulate matter
291
-------
At a second creosote impregnation facility, personal breathing zone air
samples were taken to measure both particulate and gaseous POM compounds.
Breathing zone samples associated with handling creosote treated railroad
ties contained a total POM concentration of 981.2 ug/m3, of which 97 percent
was collected as gaseous POM. These results imply that gaseous POM
emissions from treatment plants may be greater than particulate POM
releases, and that creosote plants with only particulate POM levels may be
greatly underestimating actual POM concentrations in plant air. These
implications should be taken into consideration when attempting to estimate
emissions from a .creosote wood treating process.
Source Locations
Creosote wood treatment plants are located across the country, but they
are predominantly found in the Southeast. Information compiled by the
American Wood-Preservers Association and the American Wood Preservers
Institute indicates that there are roughly 185 creosote treatment plants
nationwide. A list identifying these facilities is given in Table 73.151 .
OIL SHALE RETORTING
Process Description
Oil shale retorting has been identified in the literature as a POM
emissions source category. Retorting produces TOM emissions because it
involves high temperature contact with hydrocarbons and
hydrocarbon-containing rock and because hydrocarbon-containing off-gases
from retorting are typically incinerated.152'153 Oil shale retorting is
performed by two major processes, above-ground or surface retorting and
below-ground or in. situ retorting. In surface retorting, oil shale is mined
and brought to the surface, crushed, and heated either externally or
internally to extract oil from the shale rock. In externally heated
operations, an.external furnace is used to continuously apply'heat to the
shale retort. In an internally heated system, the oil shale furnishes its
own heat because part of its organic matter is burned inside the retort.
292
-------
TABLE 73,
LIST OF CREOSOTE WOOD IMPREGNATION
PLANTS IN THE UNITED STATES151»a
Company
Acme Wood Preserving, Inc.
Alabama Wood Treating Corporation
American Creosote Works, Inc.
American Wood Division of Powe Timber Company
Annadale Plantation
Appalachian Timber Services, Inc.
Arizona Pacific Wood Preserving
Atlantic Wood Industries, Inc.
B and M Wood Products, Inc.
Baldwin Pole and Piling Company
Baxley Creosoting Company, Inc.
J. H. Baxter and Company
Benton Creosoting Works
Birmingham Wood Preserving Company
Broderick Wood Products Company
Brown Wood Preserving Company, Inc.
Burke-Parsons-Bowlby Corporation
Burlington Northern, Inc.
Cahaba Pressure Treated Forest Products
In 1984
Location
Princeton, WV
Mobile, ALa
Pensacola, FL
Jackson, TNa
Louisville, MS
Richton, MS
Georgetown, SC
Sutton, WV
White Plains, KY
Eloy, AZ
Portsmouth, VA
Hainesport, NJ
Port Wentworth, GA
Manor, GA
Bay Minette, AL
Baxley, GA
The Dalles, OR
Eugene, OR
Weed, CA
Laramie, WY
Benton, LA
Birmingham, AL
Denver, CO
Brownville, AL
Louisville, KY
Stanton, KY
Dubois, NV
Goshen, VA
Spencer, WV
Brainerd, MN
Paradise, MT
Somers, MT
Brierfield, AL
293
-------
TABLE 73. LIST OF CREOSOTE WOOD IMPREGNATION
PLANTS IN THE UNITED STATES151 (Continued)
Company
Location
Carolina Creosote Corporation, Inc.
Carolina Wood Preserving Company, Inc.
Cascade Pole Company
Century Forest Industries
Champion International Corporation
Colfax Creosoting Company
Conroe Creosoting Company
F. E. Cooper Lumber Corporation
Crown Zellerback Treated Wood Products
Dant and Russell, Inc.
Duke City Lumber Company
Dura-Wood Treating Company
Easterday Tie and Timber Company
El Dorado Pole and Piling Company, Inc.
Eppinger and Russell Company
Escambia Treating Company
Evr-Wood Treating Company, Inc.
Fernwood Industries
Florida Fence Post Company, Inc.
Fordyce Wood Preservers
Frank Brooks Manufacturing Company
G. C. L. Tie and Treating Corporation
Garland Creosoting Company
Gateway Forest Products, Inc.
General Timber, Inc.
Leland, NC
Scotland Neck, NC
Tacoma, WAa
Lufkin, TX
Cass Lake, MN
Whitewood, SD
Pineville, LA
Conroe, TX
Johnstown, PA
Urania, LA
Gulf port, MS
North Planes, OR
Livingston, TX
Alexandria, LA
Jackson, TN
El Dorado, AR
Chesapeake, VA
Brunswick, GA
Brookhaven, FL
Camilla, GA
Jennings, LA
Fernwood, MS
Ona, FL
Fordyce, AR
Billingham, WA
Sidney, NY
Longview, TX
Mather, PA
Sanford, NC
294
-------
TABLE 73. LIST OF CREOSOTE WOOD IMPREGNATION
PLANTS IN THE UNITED STATES151 (Continued)'
Company
General Wood Preserving Company, Inc.
Glacier Park Company
Glenville Wood Preserving Company, Inc.
Great Lake Timber Company
Hart Creosoting Company
Huxford Pole and.Timber Company, Inc.
Holcomb Creosote Company
Hoosier Treating Company
Indiana Wood Treating Corporation
International Paper Company
Jasper Creosoting Company
Jennison-Wright Corporation
Joslyn Manufacturing and Supply Company
Julian Lumber Company
Kerr-McGee Chemical Corporation
Koppers Company, Inc.
Location
Leland, NG
Somers, MT
Glenville, GA
Ft. Duschene, UT
Jasper, TX
Huxford, AL
Yadkinville, NC
Gosport, IN
.Bloomingdale, IN
De Ridder, LA
Joplin, MO
Longview, WA
Navasota, TXa
Wiggins, MS
Jasper, TX
Granite City, IL
Toledo, OH
Richton, MS
Antlers, OK
Avoca, PA
Indianapolis, IN
Kansas City, MO
Madison, IL
Meridian, MS
Bossier City, LA
Springfield, MO
Texarkana, TX
Columbus, MS
Carbondale, IL
Denver, CO
Florence, SC
Gainesville, FLa
Galesburg, IL
Green Spring, WV
295
-------
TABLE 73. LIST OF CREOSOTE WOOD IMPREGNATION
PLANTS IN THE UNITED STATES151 (Continued)
Company
Location
Koppers Company, Inc. (continued)
The Langley Company
Lufkin Creosoting Company, Inc. •
Madisonville Creosote Works, Inc.
Manor Timber Company, Inc.
Marion Pressure Treating Company
Marshall Wood Preserving Company
McArthur Lumber and Post Company, Inc.
McCormick and Baxter Creosoting Company
t
McCrawie Brothers Wood Preserving Company
L. D. McFarland Company
McFarlahd Cascade
H. P. McGinley, Inc.
Mellott Wood Preserving Company
W. C. Meredith Company, Inc.
T. R. Miller ijikl Company, Inc.
Mississippi Wood Preserving Company
Grenada, MS
Guthrie, KY
Houston, TX
Kansas City, MO
Montgomery, AL
Montgomery, AR
Montgomery, PAa
Nashua, NH
North Little Rock, AR
Oroville, CA
Orrville, OH
Port Newark, NJ
Richmond, VA
Salem, VA
Salisbury, MD
Superior, WIa
Valdosta, GA
Lufkin, TX
Madisonville, LA
Manor, GAa
Marion, LA
Marshall, TX
McArthur, OH
Portland, OR
Stockton, CA
Willacoochee, GA
Eugene, ORa
Olympia, WA
Tacoma, WA
McAlisterville, PA
Needmore, PA
East Point, GA
Brewton, AL
Brookhaven, MS
-------
TABLE 73. LIST OF CREOSOTE WOOD IMPREGNATION
'151 (Continued)
PLANTS IN THE UNITED STATES'
Company
Mixon Brothers Wood Preserving, Inc.
Moultrie Wood Preserving Company
New South Forest Industries
Osser Company
' Oliver Treated Products Company, Inc.
Ouachita-Nevada Treating Company
Pacific Wood Preserving of Bakersfield
Pacific Wood Treating Corporation
Pearl River Wood Preserving Corporation
Perma Treat Corporation
Prentiss Creosote and Forest Products, Inc.
R and K Creosote Company, Inc.
Reddell Creosoted Forest Products, Inc.
San Diego Wood Preserving
Santa Fe Centralized Tie Plant
Seaman Timber Company, Inc.
Sheridan Pressure Treating Lumber
Shollenbanger Wood Treating
W. J. Smith Wood Preserving Company
Southern Pine Wood Preserving Company
Southern Wood Piedmont Company
Stallworth Timber Company, Inc.
Standard Wood Preservers of Shreveport, Inc.
Location
Idabel, OK
Moultrie, GA
Red Hill, SC
Bellingham, WA
Hammond, LA
Reader, AR
Bakersfield, CA
Ridgefield, WA
Picayune, MS
Durham, CT
Prentiss, MSa
Natalbany, LA
Reddell, LA
National City, CA
Somerville, TX
Montevallo, AL
Sheridan, OR
Bernalilco, NM
Denison, TX
Wiggins, MS
Augusta, GA
Baldwin, FL
• Chattanooga, TN
East Point, GA
Gulf, NC
Macon, GA
Spartanburg, SC
Waverly, OH
Wilmington,. NC
i i
Beatrice, At
Shreveport', LAa
297
-------
TABLE 73. LIST OF CREOSOTE WOOD IMPREGNATION
PLANTS IN THE UNITED STATES151 (Continued)
Company
Location
St. Regis Paper.. Company
Superior Tie and Timber
Superior Wood Treating, Inc.
Sweeney Wood Products
J. C. Taylor Lumber Sales
Texarkana Wood Preservative Company
Texas Electric Cooperatives, Inc.
Thompson Industries
Thomasson Lumber Company
Timco, Inc.
Union Lumber Company
Utah-Power and Light
Vermont Correctional Industries
Virginia Wood Preserving
Webster Wood Preserving Company
Western Tar Products Corporation
Western Wood Preserving Company
Wood Preservers, Inc.
Wood Treating, Inc.
Wyckoff Company
Cass Lake, MN
Whitewood, SD
Vivian, LA
Louisville, MS
Lapoint, UT
Sheridan, OR
Texarkana, TX
Jasper, TX
Russellville, AR
Philadelphia, MS
Wiggins, MS
Homerville, GA
Idaho Falls, ID
Windsor, VT
Laurel, VA
Bangor,-WI
Terre Haute, IN
Sumner, WA
Warsaw, VA
Picayune, MS
Bainbridge Island, WA
Seattle, WA
n0n-preasure treati*S techniques in addition to pressure treating
Plants use only non-pressure treating processes.
NOTE: Jhis listing is subject to change as market conditions change,
facility ownership changes, plants are closed down, etc The reader
ldV^if? ** existence of Particular'facilities by consulting
SS an°r thS Plants *««*l™* The level of POM*
facilitv is a function of variables such as
298
-------
Retorting temperatures in an internally heated surface retort range
from 649 to 704°C (1200 to -1300°F) . Due to these relatively high
temperatures, product oil from retorting has a lower naphtha content but
higher aromatic content than oil generated by externally heated retorts.
Retort off -gases, which are diluted by nitrogen from air and carbonate
decomposition in the shale, have a Btu value of about 100 Btu/scf.154
Externally heated surface retorts require a separate heating unit fired
by product gas or residual carbon to extract shale oil. Shale is heated
either by the hot combustion gases or by means of a heat-carrying medium,
typically sand. Retorting temperatures of about 482°C (900°F) are
maintained. Because of the lack of outside air injections, off -gases from
externally heated surface retorts have a higher heating value (950 Btu/scf)
than internally heated surface retorts.154 Diagrams illustrating both types
of surface oil shale retort are given in Figure 37. 154
oil shale retorting generally falls into one of two categories.
True in sigu processing involves the drilling of injection and production
holes from the surface into the oil shale strata to be retorted. These
holes are used to fracture the oil shale and create permeability. Once
permeability is established, hot fluids or fine fronts are passed from
injection wells to production wells where the retorted shale oil is
recovered and pumped to the surface.155
Modified in siai retorting indicates that a portion (15 to 40 percent)
of the oil shale is mined and brought to the surface, while the remainder is
explosively fractured prior to the ignition of the retort.155 The retort is
then ignited with a propane burner and soon after combustion of the oil
shale is self-sustaining. Air and steam are forced into the top of the
retort to control the temperature and rate at which the flame front
progresses downward through the rubblized layers of oil shale. Temperatures
of 871 C (1600 F) are achieved in in situ retort systems.154 As the high
temperature flame front proceeds through the oil.shale rubble, condensed
299
-------
Oil-dial*
mine
Crushing,,
feed
preparation
Fines discard
Retort
Raw snala feed
Separator
Shale preheating
zone
Shale retorting
zone
Combustion zone
Spent-shale
cooling zone
Vapors
I
Spent-shale solids
Internally heated retort
Product gas
Product oil
(to upgrading)
Air
Oil/shato -
mine
Crushing,
» fMd — .
preparation
1
Pines c
(ifreq
iscard
uired)
j"w««*i^n^_ . ,
1 medium M— ^
Raw ~T |
shale feed I |
R«». .y*0™,! .
Heating
*
Separator
X
JL» Product
gas
Product oil
(to upgrading)
Spent shale
b. Externally heated retort
Figure 37. Surface oil shale retorting.154
-------
shale oil, sour water, and low Btu gas flow downward through the retort into
an underground separator room. In this room, shale oil is separated from
the water and both liquids are pumped to the surface.153
Modified and true is Si£u oil shale retorting are illustrated in
Figure 38. "*
Off-gases from surface and la Sifii oil shale retorting, containing POM
compounds, steam, inorganic gases, oil droplets, oil vapors, soot, char,
unbumed oil shale, and spent oil shale, are sent to knockout drums to '
remove most larger particles in the stream. Following the removal of larger
particles, the gas stream is vented to an incinerator to destroy hydrocarbon
compounds. Incinerator exhausts, containing sulfur dioxide and particulate
matter, are sent to wet scrubbers before final venting to the atmosphere.153
Emission Factors
The amount of information .on POM emissions from oil shale retorting is
extremely limited. Two sets of data were located that quantified emissions-
however, these data were expressed as POM concentrations and not emission
factors. In one data set, off-gases from a vertical modified Jfc situ retort
were sampled and analyzed in five test runs. Total POM concentrations in
retort off gas ranged from 49 to 1207 ug/m3, with the average being
550 ug/m . The average offgas flow rate was 100 m3/min. The POM quantities
measured were the total of materials collected by a glass fiber filter and a
Tenax cartridge. The majority of POM compounds measured resided on the
glass filter. Specific POM compounds that were identified included
anthracene, pyrene, benzo( a) anthracene, and chrysene . 152
Three "of the sampling runs were made three to seven days after
initiation of retorting when burn conditions had stabilized. The range of
total POM concentrations for these runs was 49 to 206 ug/m3. The remaining
two sampling runs were made three weeks after retorting began. Total POM
301
-------
'' *" "' Overburden "•'•"!• •'•"••'• '• •' ;-:. '••• '
a- g^-!-^~*=s;5--===i===^S
^' ^ Front movement ~T^Tf~^_T
\ r>—'*1—^—•'•—'-—— ^-~^-^—?^^
t^^T-^^™"^^™^^^^^***^r™t*^^""~*a!Tr**~ *~~*
I
Combunion zone
a. True in situ retorting
To surface
retort
Separation
Combustion zone
Retorting zone
Vapor
condensation
zone
Step 1: Mining Step 2:. Rubblizing
Oil and Gas
Seep 3: Retorting
b. Modified in situ retorting
Figure 38. In situ oil shale retorting.
302
-------
concentrations in off-gases from these runs were 1100 ug/m3 and 1207 ug/m3
The authors of.Reference 152 speculated that the burning pattern in the
retort room may account for the observed increase in total POM over time
As the flame front proceeds down the retort room, increased heating of the
shale further down toward the base of the room is occurring, thereby
allowing greater mobility of POM compounds into the off-gases.
In a second set of data, total PAH compounds contained in the exhaust
gas stream of an incinerator used to control hydrocarbons at a vertical
modified is ^ retort were analyzed. Total PAH concentrations for two
runs were 1142 ug/m3 and 406 ug/m3. Of these total PAH quantities measured
only 2 to 10 percent was identified with specific compounds.« The specific •'
compounds identified are listed below.152
naphthalene
acenaphthylene
acenapthene
fluorene
9,10-dihydroanthracene
• • phenanthrene
anthracene
• '2-methylanthracene
9-methylanthracene
fluoranthene
- . pyrene
o J
benzo(a)anthracene
chrysene
benzo
-------
dtbenzo ( a , h) anthracene
benzo(g,h, i)perylene
In total, over 300 PAH compounds were detected^152
The .sampling train used to collect these samples consisted of a glass
fiber filter followed by an ice bath cold trap followed by two Tenax
cartridges in series. Most of the PAH material collected during the test
resided in the cold trap of the sampling train.
Source
As of January 1986, no commercial scale oil shale retorting was being
performed in the United States. In many cases, retorting projects have
ceased operations because Federal funds supporting the projects have been
withdrawn. Potential retorting that may be performed in the United
States is most likely to occur in either Colorado, Utah, or Wyoming as the
bulk of reclaimable oil shale deposits lie in these States. The primary
area of activity would lie where these three State boundaries meet in
northeastern Utah, northwestern Colorado, and southwestern Wyoming. The map
in Figure 39 illustrates the main zones of oil. shale deposits in these
'States and the United States as a whole.155
ASPHALT PAVING AND COAL TAR PITCH AND ASPHALT ROOFING OPERATIONS
• i
A moderate amount of information exists in the literature that
indicates asphalt paving and roofing and coal tar pitch roofing operations
to be potential POM emission sources.157'159 The principal focus of the
work in the literature has been on the health hazards presented to workers
by these emissions and not on overall environmental implications. For this
reason, the majority of the POM data that exist for these operations is in -
the form of ambient air and worker area personal sample POM concentrations.
Classical emission factors (i.e., mass POM/mass of material used or
304
-------
M
0)
C/3
•o
0)
4J
•pf
c
OJ
c
•H
03
CO
9
a
GJ
-------
produced) are not available. The data that are available demonstrate that
asphalt-paving and roofing and coal tar pitch roofing are fugitive POM
emission sources.
During asphalt paving operations, POM emissions have been measured (as
concentrations) that are associated with pavement preparation, pavement
application, and post-application releases. Polycyclic organic matter
concentrations have been detected both in the breathing zones of paving
workers and in downwind ambient air. The source of the majority of POM
compounds measured in connection with asphalt paving operations is the
asphalt paving material. At two paving operations tested, the paving raw
materials had total POM concentrations of 218 ug/g and 183 ug/g.157'158 in
both of these samples, chrysene and bensoCa)anthracene constituted about
85 percent of the total POM.
Measured concentrations of POM associated with asphalt paving
operations are summarized in Table 74.158 The POM measured was dominated by
fluoranthene, benzo(a)anthracene, pyrene, and chrysene. Benzo
-------
00
in
••H
m •
ss
o
M
^4
^d
CO
£
O
g
s
2J
2
H
s*
S3
04
CO
a
1
3
SB
H4
g
B!t
B
1
*
CO
ss
o
H
ttS
£H
§,
:
§
a
S
^
sf
3
i
Sin
2-1
0
1 S
ca
o» c
«••.-
••-( 4.
en cc
h
60 4.
c c
•*4 01
> c
ca e
"3
ft
•< •
.*" ""
W «
60 4.
e a
*> S
3 e
04 q
.
s
o
4J
ca
0)
cu
o
frfl
(2
i^
•i-l 01
•H
4J CO
a
01 Wi
ia S
3 S!
"*«
S
o-
•l-l
4J
i
o
•H
1
PU
i-l- 01
•< 4J
SS
01 Vi
^ CQ
£) 0)
C9 >9*
J» **
«5
POM Compound
-j «* oo m si- _4 m
«y> •-! r>. o i-i o o
0 es oo ^ o o H en 01* i— i
r*. en cs rs o S o
o o «s o o 2 o
&11
•« S « .0 *< PH
•* o es — i ja o o
0 o . o o es es
oo in ^ o. o o
^ 0 vo CM 0 o o
°I ' 1 1 1 1 1
O ^ ON O% eN qj ^^
O en o o h O
-^ !n S 2 S o o
-« .0 • 0 0 0 <0
01.
2
JS
fi
o
0)
H « -H -H
# Q A 0 r<
HO o G9
«N o ' -H sf cs
O -H O O •
O 0 o O 2
1 1 II i
S S S S £
0 0 2 0 ^
0}
_
c
ca
0)
ticulat
M
8,
x
o
•^
TS
01
h
CO
0)
•o
0)
CO
a
oa
O)
9
0)
u •
o
0.
CO
•H
CO
The sampling and analy
Data not determined.
307
-------
The typical concentrations of POM released from paved asphalt exposed
to sunlight^, as determined by laboratory experiments, are. summarized in
Table 75. The effect of differing light and humidity conditions on POM
and benzo(a)pyrene concentrations from the asphalt are illustrated in
Table 76. Under normal light but zero humidity conditions, POM levels
are only about 20 to 25 percent of what they were when measured under normal
daytime conditions (Table 76). When compared to Table 75 and the results
for experiments 1 and 2, experiments 3 and 4 in Table 76 indicate that POM
releases from freshly paved asphalt are strongly related to light exposure.
An examination of "the benzo(a)pyrene data in Table 76 implies that
benzo^pyrene formation is also strongly related to pavement light
exposure.
All experiments conducted to obtain the data in Tables 75 and 76 were
performed on a simulation of"freshly applied asphalt pavement. No
examination was made of potential releases over time.157
Similar to asphalt paving, several studies have been conducted to
•determine the quantity of POM emissions that asphalt and. coal tar pitch
roofing workers are exposed to. Roofing workers may be exposed to POM
compounds during preparation (heating) of the asphalt or coal tar pitch prior
to application, during application, and during roof tear-off, which must be
performed before a new roof can be applied. New construction roofs would
not have a tear-off step.
Generally, for new asphalt and coal tar-pitch'roofs on existing
structures, the tear-off operation begins by sweeping loose gravel from the
old roof with a power broom. "The existing asphalt or pitch layer is broken
up down to the existing level of insulation using a power cutter. Once the
breaking process is accomplished, the old roof is pried up and scraped from
the surface. After tear-off operations are complete, rigid insulation is
applied with hot asphalt and tar paper is applied over the insulation using
hot asphalt or coal tar pitch. Other materials such as aluminum and rubber
membranes may also be applied -to the insulation using hot asphalt or coal
tar pitch. **
308
-------
TABLE 75. POM CONCENTRATIONS RELEASED FROM FRESHLY
PAVED ASPHALT EXPOSED TO LIGHT157
Experiment
Number •
1.
2.
3.
4..
Test Conditions
Asphalt
Humidity Temperature
72%
76%
85%
78%
42°C (107°F)
43°C
38°C (100°F)
37°C (99°F)
Percent POM
Concentration, that is
3
ng/m Benzo(a)pyrene
2139
1729
2280
2310
14%
14%
21%
0%
309
-------
—
in
p«4
2 §
51
0§
W U
M
>« 0
»J M
S3 SB
0! O
§|
§3
§8
if t«t
z S
£x3 Frii
C3 *«<
so
O
a jj 03
O 4) 01 01 W)
4J ^i ' i_i
O O A
^5 ^E ^S,
^
S & £ •*• M °*
in CN ON •-( en en
en m 1-1 en
en
o en 1-1 en o o
u o w w o u
o o o. o o o
en m -* . m oo oo
g^ ^
O . 0 O O -H ^
r«» oo
•
41 5"
>-4 d
" - _ ** 'w -a o
<§ d s s -a*, <§
. 5 -a- •
•H
,
as
.
•-* «N en ^ m \.p
310
4
*
60
•••4
i— 1
Ot
CO
41
,Ci
4*1
4J
y
9
d
o
u
o
•o
41
CO
00
ca
3
•H
-------
Polycyclic organic matter cbncentrations that have been measured in and
around asphalt and coal tar pitch roofing operations are given in
Table 77. Concentrations associated with coal tar pitch operations are
significantly greater than those from asphalt roofing because coal tar pitch
contains on average 800 times more total POM than asphalt. As would be
expected, concentrations are greatest from the asphalt and pitch preparation
operations because in this step, the materials are being heated to make them
fluid enough for pumping and application. Total POM results for the asphalt'
roofing sites are fairly consistent with the exception of the preparation
source value at site 1, which is about 5 to 7 times greater than preparation
source values for the other asphalt sites. The coal tar pitch data are not
as consistent overall as the asphalt data; however, they do not exhibit any
variability as great as the asphalt preparation source at site 1.
The only emission concentrations information on POM emissions from roof
tear-off operations was contained in Reference 159. In this study POM
concentrations resulting from a coal tar pitch tear-off operation were
significantly greater than those from an asphalt tear-off operation
(i.e., 100 ug/m compared to a trace). Concentrations were also determined
for application operations and these were more consistent. For hot coal tar
pxtch application, source emissions had a total POM concentration of about
30 ug/ny The same concentration at a hot asphalt application site was
20 ug/m . No other information could be identified to confirm the
indication that coal tar pitch tear-off operations are significantly greater
POM emitters than either asphalt tear-off operations or asphalt and coal tar
pitch application operations. 9
TRANSFER AND HANDLING OF ^OAL TAR AND PETROLEUM PITCH
One reference was identified that presented data to indicate that POM '
emissions are released to the atmosphere during handling and transfer
o
operations involving coal tar and petroleum pitch.
UCted ^ ^ Nati°nal InStitUtS ^ OC-
160
this study which
rN Safety and Health
(NIOSH) , an assessment was made of the health hazard presented to workers
311
-------
.
4
a
1
® 5 £ °" ^ ^ °* ^* o -I ao «4 * m-*o« o«
3* 31 • Z S S • ffi* ^^S^^GO^ao*^
0 ® ee e o o e x * «
oSt52oo3-"t5oao
,
SS^"*"*** ^^eo^^oc^aijMiM..
°^**OOO "OOO^woS
ooeeoeHoo«se^p-ooo«
^*
O->5ooo3oC8l°'°aO'n*
_a. o o^ s^S
« «i u a
w a. w
•« u 41 .-> .a
«8 3 aj Ql
•- ee S g 2
-a .1 3. 01 >,
• • <• a — a
eo oa >, § ».-*
S — u k. -0
e *p4 es a) •
e9 *M c X a§ cj
3O SIM
fc* T3 •
• e eu k. .M o « C
" •. » "* 841
g B • >» ej B
CU !A ^ u ™*
2 °3 a a 5 '
1 3 j= £ S ft
-------
from handling coal tar and petroleum pitch. The results of the assessment
showed that coal tar and petroleum pitch handling and transfer operations are
potential sources of fugitive POM1 air emissions.16°
In the coal tar pitch operation that was investigated, coal tar pitch
was being transferred from a river barge to an ocean barge by means of a
crane. In the petroleum pitch operation, pitch was being loaded from
railroad cars to an ocean barge. Railroad cars are positioned over a hole
in the dock so that when a trap door is opened in the bottom of the rail
car, pitch falls out onto a conveyor. The conveyor carries the pitch to a
chute at the edge of the dock, where it drops down and onto a barge.160
Worker exposure to particulate POM emissions during both of these
operations was evaluated by taking personal breathing zone air samples and
handling/transfer area air samples. The POM compounds and concentrations
that were detected are summarized below.160
Compound
Benzo(k)fluoranthene
Benzo(b)fluoranthene
Benzo(a)anthracene
Benzo(e)pyrene
Benzo(a)pyrene
Pyrene
Chrysene
Fluoranthene
Concentration
0.02 - 12.88
0.05 - 34.76
0.11 - 34.76
0.09 - 38.85
0.11 - 38.85
0.46 - 44.99
0.32 - 26.58
0.93 - 3.72
Operation Type
Coal Tar Pitch
Coal Tar Pitch
Coal Tar and
Petroleum Pitch
Coal Tar Pitch
Coal Tar Pitch
Coal Tar and
Petroleum Pitch
Coal Tar Pitch
Petroleum Pitch
While it is not possible to quantify potential POM releases from pitch
handling and transfer with these data, they do indicate that fugitive POM
emissions are occurring.
313
-------
At the river terminal where the Reference 160 tests were conducted, a
coal tar or petroleum pitch transfer operation takes place every 2 to
3 weeks, each lasting about 16 hours. Therefore, as a POM air emissions
source, pitch handling and transfer operations would be intermittent and
variable. They would most likely be located at river and marine terminals
along major water transportation routes that handle industrial commodities
(grain, fertilizer) and basic raw materials (mineral ores, coal).
BURNING COAL REFUSE PILES, OUTCROPS, AND MINES
Process
Because they are sources of highly inefficient combustion, burning coal
refuse piles, outcrops and mines have been identified as potential POM air
emission sources. '2
Coal as it comes from a mine contains various amounts of impurities
such as slate, shale, calcite, gypsum, clay, and pyrite. These waste
impurities are separated from coal prior to its being marketed. This waste
material or coal refuse is commonly piled into banks or stored in
impoundments near coal mines and coal preparation plants. Coarse refuse
(i.e., greater than 595 urn diameter) is deposited into piles by dump trucks,
mine car,s, conveyors, or aerial trams. Indiscriminate dumping and poor
maintenance of refuse pile^are two practices that can result in spontaneous
combustion of refuse piles.
161
Pine coal refuse (i.e., material less than 595 urn diameter) is often
pumped to impoundments or settling ponds as slurry and allowed to settle.
Filters and clarifiers may also be used to aggregate the fine refuse.
Impoundments are usually constructed of existing coal refuse, as it is the
cheapest fill material available. Settled fine refuse is periodically
removed from the impoundment and dumped on the larger coal refuse piles.
Impoundments may also be sources of spontaneous coal combustion.161
314
-------
Spontaneous ignition and combustion of coal refuse piles and
impoundments is mainly an oxidation phenomenon involving coal, associated
pyrite, and impure coal substances. The oxidation of carbonaceous and
pyrite material in the coal refuse is an exothermic reaction. The
temperature of a coal refuse pile or portions of it increases if the amount
of circulating air is sufficient to cause oxidation but insufficient to
allow for dissipation of the resulting heat. The temperature of the refuse
Pile then increases until ignition temperature is reached. Experimental
evidence has indicated that the heat of wetting of coal is greater than the
heat of oxidation of coal; therefore, the presence of moisture in air
accelerates the self-heating process in coal refuse piles. For this reason
the relative humidity of ambient air is a key factor affecting coal refuse '
pile fires.
Coal textural moisture content (i,e., moisture retained in coal pores
and void spaces) is also an important variable in the occurrence of coal
refuse fires. Upon exposure to air, moisture is lost from- the coal pores
thereby leaving a significant area for oxygen adsorption. Increased oxygen
adsorption facilitates greater oxidation and promotes the development of
coal refuse pile fires. X
Oxidation of pyrite impurities in coal refuse piles is another
supplementary factor which enhances the possibility and severity of coal
refuse combustion. Oxidation of pyrite is a highly exothermic reaction that
increases the temperature of surrounding oxygen material and thus increases
the coal's rate of oxidation.161
are:
Other factors contributing to or affecting coal refuse pile combustion
external sources of heat such as steam pipes, sunlight, etc.; and
coal particle size.
315
-------
Fine particles pose competing situations. In one case, a predominance of
fine particles offers greater total surface area to oxidation, thereby
permitting more rapid oxidation. Conversely smaller particles allow for
denser coal packing which can reduce the flow of air through the pile and
decrease the rate of oxidation.161
Coal refuse piles are considered to be burning if they exhibit either
of the following conditions.
or
presence of smoke, fume, flames, thermal' waves above the pile,
fire glow
an internal temperature of 93°C (199°F)
The spontaneous combustion of coal in outcrops and abandoned mines is
also attributable to oxidation phenomena involving coal, moisture, and
pyrite impurities. Other factors affecting combustion in mines and outcrops
include coal rank, coal strata. geology, and the -coal strata temperature
profile. Low- rank coals such as subbituminous or high-volatile bituminous
are more "susceptible to spontaneous combustion than a high-rank coal such as
low-volatile bituminous or anthracite. Low-rank coals contain a greater
amount of moisture and pyrite impurities than high- rank coals, which
enhances their propensity for spontaneous combustion. The presence of
faults in coal seams enhances oxidation by providing channels for greater
volume and more distributed air flow. Coal strata temperature typically
increases with depth. Oxidation rate, therefore, will increase with depth,
making the seam more vulnerable to spontaneous combustion.161
To summarize, spontaneous combustion and resulting emissions from coal
refuse piles, outcrops, and mines are primarily affected by the following
factors .
Coal Refiiaa
oxygen concentration in the pile which is dependent on pile
particle size distribution, type of pile surface, and wind speed
316
-------
type of coal
relative humidity of ambient air
coal moisture content
type of refuse
temperature
Outcros and
oxygen concentrations, which are affected by air leakage through
natural faults and cracks and air leakage through holes caused by
subsidence
type of coal
depth of stratum
relative humidity of ambient air
coal moisture content
temperature
Various techniques exist to control emissions from burning coal refuse
Piles, outcrops, and mines. The majority of these techniques are based on
eliminating the fire's oxygen supply to extinguish it and on preventing the
fxre from spreading. The primary methods that have been applied to refuse
piles are described below. 1
Isolation. - The burning area is isolated from the remainder of the
refuse pile by trenches and is quenched with water or blanketed with
incombustible material.
SliffikeMsg - Some piles are extinguished by leveling the top, then
sealing it /and the sides with fine, incombustible material such as fly
ash, clay, quarry wastes, or acid mine drainage sludge. Heavy seals of
such material are necessary to avoid erosion. The use of clay is
limited as it cracks over hot spots impairing the seal.
Grouting . A slurry of water and finely divided incombustible material
such as pulverized limestone, fly ash, coal silt, or'sand, is forced
into the burning pile so as to provide some cooling action and also to
fill the voids to prevent air from entering the pile.
317
-------
Explosives - Many burning piles have an impenetrable, ceramic-like,
clinker material surface which does not allow the penetration of
slurries and water. In this case, explosive charges are placed deep
into the bank through horizontally drilled holes. The explosion
creates fissures in the fused covering material. Water is then applied
through these crevices and the quenched material is loaded out.
SSSnyins - In this method, water is sprayed over the entire refuse
bank. However, this is only a temporary solution as the pile reignites
and burns, often with renewed vigor, once the water spray is stopped.
Accelerated Combustion and Quenching - The burning refuse material is
lifted by a dragline and dropped through air into a water-filled lagoon
15 to 30 m (49 to 98 ft) below for the purpose of burning off the
combustible material completely during the drop. Another dragline and
bulldozers are used to remove the quenched material from the lagoon
floor and compact it into a tight, dense fill material.
Sanding - Retaining walls are constructed around the perimeter of a
refuse bank after subdividing the surface into a series of' level
.discrete areas and each area is filled with water to flood the fire.
This method has not proven to be successful because flooding with
water may cause explosions due to the formation of water gas and water
penetration into the pile is poor.
g.o.oUnS and TMIm-lon - Water is sprayed on the burning pile from
multiple nozzles and the cooled refuse is mixed, by bulldozer, in a
: one-to-one volume proportion with soil and/or burned oufc refuse from a
nearby area. The mixture is then compacted by heavy equipment.
.. Hydraulic J«t;« - High velocity water connons are used to quench the
Burning refuse material. The quenched material is then relayered and
compacted by a dragline and bulldozer.
With fires in coal outcrops or abandoned mines, the principle of
control is to isolate the burning material and prevent air from reaching it.
The techniques used for this purpose are described below.161
318
-------
_Qut - This method involves digging out the burning and heated
material, then cooling it with water or spreading it on the ground.
This method is effective if the fire is of.recent origin or mild enough
to be accessible.
?ire Barriers - A barrier of incombustible material is used to confine
and isolate the fire from the main body of coal. The isolated area is
also surface sealed to extinguish the fire. The barrier can be an open
trench, between.the fire area and the threatened area, which is
backfilled with incombustible material such as earth, fly ash, or
granulated slag. A plug barrier is used if the overburden is'
excessive, since it is impractical to excavate a trench from outcrop to
outcrop around an abandoned mine fire. -A plug barrier starts at the .
outcrop and terminates when the overburden depth exceeds 20 m (66 ft)
The plug barrier is always used in conjunction with a surface seal A
surface seal on the fire side of the plug has been observed to be
effective in controlling abandoned mine fires if the overburden is in
excess of about 20 m (66 ft).
IlasMng - m this method, the void spaces around an underground fire
are filled with water or an incombustible material such as fly ash
The incombustible material can be applied pneumatically or as a slurry.
Surface SMlfTlg . This technique involves closing the surface openings
surrounding the fire site to prevent ventilation of the fire The
surface seal is established by plowing the surface to a depth of
several meters with an angle dozer to create a blanket of pulverized
earth that effectively seals the surface.
Emission
One emission factor was found in the literature relating to POM
emissions from burning coal refuse piles, outcrops, or mines 161
Particulate POM emissions from a burning coal refuse pile have measured
been measured by using a high-volume filter air sampling device. The total
POM emission factor for burning coal refuse piles developed from these test
319
-------
results is 0.019 mg/hr-m3 of burning coal refuse.— Assuming an average
3 - - - 161
»
coal refuse pile density of 1.5 Mg/m3 (0.05 ton/ft3), an equivalent POM'
emission factor of 0.013 mg/hr-Mg of refuse burned can be calculated. The
POM compounds identified in the collected refuse pile emissions are listed
below.
dibenzothiophene
anthrac ene/phenanthr ene
methylanthracenes/phenanthrenes
9-methylanthracene
fluoranthene
pyrene
benzo ( c ) phenanthr ene
chrysene/benz (a) anthracene
dimethylbenzanthracenes
benzo (k) fluoranthene
benzo (b) fluoranthene
benzo (a)pyrene/benzo (e)pyrene/perylene
3 -methylcholanthrene
dibenzo ( a , h) pyrene
dibenzo(a, i)pyrene
dibenz (a , h) anthracene
dibenz (a , c) anthracene
indeno ( 1 , 2 , 3 - c , d) pyr ene
7H-dibenzo (c , g) carbazole
Source
Burning or potentially-burning coal refuse piles, outcrops, and mines
are linked to coal mining and coal .preparation plant locations. No recent
information on the sources of burning refuse piles, abandoned mines, and
outcrops could be identified from the Bureau of Mines or the Office of
Surface Mining. Data for 1972 indicate that West Virginia, Pennsylvania,
Virginia, and Kentucky accounted for 84 percent of all burning refuse piles
320
-------
and West Virginia and Pennsylvania accounted for 87 percent of the burning
refuse impoundments. Given that these States are still major coal mining
areas, their status as leading sources of burning refuse sites is probably
still valid.161
In January 1976, there were 441 fires in abandoned coal mines and
outcrops. These fires occurred in 18 coal producing States, with Montana,
Wyoming, Colorado, and New Mexico accounting for 66 percent of the fires.
or
A list of States containing burning coal refuse piles, outcrops,
abandoned mines is presented in Table 78.161
PRESCRIBED BURNING AND UNCONTROLLED FOREST FIRES
Process Description
Prescribed burning is defined as the application and confinement of
fire in forest and range management under, specified conditions of weather
fuel moisture, and soil moisture that will accomplish planned benefits such
as fire hazard reduction, control of understory species, seedbed and site
preparation, grazing enhancement, wildlife .habitat improvement, and forest
tree disease control. It differs from uncontrolled forest fires in that it
is used only under controlled conditions and is managed so that beneficial
effects outweigh costs and possible detrimental impacts. Uncontrolled
forest fires refers to fires j:hat are started naturally (lightening),
accidentally, or intentionally in forests that bum and spread in generally
unpredictable patterns. Prescribed burning and uncontrolled forest fires
are both potential sources of POM emissions because combustion in these
environments is Inefficient and incomplete due to the high moisture content
and varying composition of the materials burned.162"164
The firing techniques employed in prescribed burning depend on the kind
of area to be burned and on local burning conditions. The predominant
burning techniques are backing fire, heading fire, ring fire, and
321
-------
TABLE 78. BURNING COAL REFUSE PILES, IMPOUNDMENTS, ABANDONED
MINES, AND OUTCROPS IN THE UNITED STATES BY STATE161
State
Alabama
Alaska
Arizona
Arkansas
Colorado
Illinois
Indiana
Kentucky
Maryland
Montana
New Mexico
.North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
South Dakota
Tennessee
Texas
Utah
Virginia
Washington
West Virginia
Wyoming
TOTAL
&~ia-}>> j_^
Refuse
Active
7
1
...
1
8
0
. 0
28
1
2
' 4
0
7
0
37
...
1
...
30
0
79
0
206
Piles3
Inactive0
1
4
...
...
37
9
6
90
12
16
9
2
...
1
123
...
3
...
76
4
48
19
467
Impoundments
Active Inactive
4 12
1
• « • • • «
• •> — •
-- - . . «
0 2
• • «•
1 94
* • ...
... _„_
... ___
... .„_
0 . 3
• • » • . •
• • •
16 43
• • * ...
0 4
0 2
2 20
0 1
41 42
0 2
65 225
Abandoned
Mines and
Outcrops0
.._<*
7
30
66
6
2
105
39
18
7
1
37
3
1
28
1
2
8
80
441
Indicates not presently burning but could burn
C1975 data.
indicates that data were not reported by a State.
-------
area.ignition. Often, combinations of these techniques are used at a single
burning site. A fifth method of burning known as pile and windrow fires
also exists but is not used frequently.163
Backing fires are ignited on the downwind side of an area and permitted
to spread against the wind. The advance of the active burning zone is slow
and most of the fuel is consumed within this zone. In this way, smoldering
time for the fuel is reduced, and .total combustion efficiency of the fire is
increased. The backing fire produces the least fire intensity of all
techniques, having slow spread rates, a narrow burning zone, and short
flames. It therefore lends itself to use in heavy fuel accumulations and in
removing understory growth and debris where an overstory of crop trees
exists. J
. Heading fires are ignited on the upwind side of an area and spread with
the wind. The active burning zone moves rapidly from fuel element to fuel
element. Under these conditions, many fuel elements are not consumed
completely in the active burning zone. A rather large zone of smoldering
fuel is left behind, producing large quantities of products of incomplete
combustion. This technique, is employed in lighter fuels if the amount of
heat produced will not scorch overstory tree crowns. Heading fires are also
preferred for control of brownspot disease in longleaf pine.163
Ring firing is accomplished by igniting the perimeter of the intended
bum area and allowing the fire to burn towards the center. This technique
results in a rapid,, relatively hot fire and finds particular application in
reducing timber-harvesting residues in clearcut areas.163
Area-ignited fires are set by igniting the intended burn area in many
individual spots and allowing individual fires to burn in all directions as
they come together. This type of fire is frequently employed in clearcut
areas when a rapidly-developed and high-rising convection column is desired
It will have high variability in burning intensity as junction zones of
increased intensity form. J
323
-------
Pile and windrow fires are used in management programs to .effect more
complete consumption of large pieces of material, such as logging residue.
When hand-piling is practiced, the objective is to dispose of only the fine
fuels and the smaller diameter branchwood. The preferred firing technique
calls for igniting the finer fuels around pile perimeters to obtain rapid
heat buildup, permitting the larger fuel elements to become ignited and
consumed. The- extreme fire intensity may adversely affect the soil
immediately beneath piles. In poorly conducted operations, piles and
windrows may contain large amounts of soil when machine piled. This may
result in areas which burn and smolder for days.163
In addition to firing techniques, emissions from prescribed burning and
uncontrolled forest fires are affected primarily by environmental factors
and fuel conditions. The most prominent environmental factors influencing
emissions' are wind speed and direction, rainfall history, and relative
humidity. Secondary environmental factors include'degree of cloud cover,
air temperature, atmospheric stability, and degree of land slope. Wind '
speed, wind direction, and, to a lesser extent, slope of land all determine
how fast a heading fire or a backing fire will spread. Generally, a faster
moving fire front burns less efficiently, producing more smoldering and
greater emissions.
The most important fuel characteristics affecting emissions from
burning are fuel moisture content and fuel loading (i.e!, amount of fuel
per unit area). Fuel arrangement and fuel species composition (i.e., fuel
type, fuel age, and fuel size) are also key variables affecting emissions.
High moisture content reduces combustion efficiency, which in turn produces
greater emissions. Fuel loading level is directly related to emissions, the
more fuel burned, the greater the emissions. Fuel arrangement can affect
burn intensity and completeness by affecting air supply and it may influence
the fire spreading pattern. Fuel composition affects emissions in several
ways. Different fuels (wood, grass, brush, leaves) have varying
compositions, which upon combustion, produce different qualities and
quantities of emissions. Fuels of differing ages contain varying moisture
m
-------
contents (seasoned versus green fuels) and varying organic constituents
which may affect overall burning emissions. Emissions may also be affected
if fuel composition has been modified by organic forest treatment chemicals
such as pesticides, herbicides, etc.163'165
As applied to prescribed burning, the term control technology may be
defined as either alternatives to burning in which prescribed fires are no
longer used, or the use of control techniques in which emissions are
reduced, dispersed, or directed away from population areas. Alternatives to
prescribed burning include mechanical and chemical treatment of forest
areas, improved forest utilization, and no forest treatment of any type
The feasibility of using an alternative approach varies with the needs and
conditions of the particular forest site.
Different types of mechanical treatment can be used to clear away brush
and trees, to prepare land for planting, to break up slash into finer
material, and to dispose of slash or brush by burial in gentle terrain
. These techniques are not effective for control of understory species without
damage to the overstory, for disease control, for reduction of fire hazard
(except burial), or for wildlife habitat improvement. Mechanical choppers
and shredders can be used to clear.away brush and trees for tree planting in
open areas. The debris left behind by these kinds of equipment may have to
be disposed of to avoid a fire hazard. In addition to choppers and
shredders, bulldozers or tractors may be used to clear land for planting and
to bury slash material (logging waste). However, this kind of clearing has
the potential for soil compaction and increased erosion.163
Chemical treatments, such as the application of herbicides, have been
used for seedbed preparation and brush control purposes. The development of
selective herbicides (compounds that are only toxic to certain species) has
made this technique useful in more applications. Such treatments, however
can only be applied to live vegetation and do nothing to reduce fire
hazard. Potentially, the risk of fire may actually increase as the dead
325
-------
vegetation dries. Recent bannings on the use of once major herbicides, such
as 2,4,5-T, have greatly limited the^use of chemical treatment as an
alternative to prescribed burning.^**
163
Improved utilization is an attractive alternative to prescribed burning
that: encompasses both improved harvesting techniques that generate less
slash material and new end uses for material that is normally burned.
Examples of improved harvesting methods that are being used or under
development include:
directional felling to reduce log breakage,
prelogging or postlogging to recover small diameter timber,
better handling techniques that will accept material normally
discarded as slash, and
design of contractual agreements to encourage recovery of small
size material.
The elimination of any form of treatment including prescribed burning
poses no immediate adverse effects to the environment, but it does not
accomplish any of the benefits for which prescribed fires are used
Moreover, no treatment for extended periods of time can lead to increased
risks of losses due to wildfire, insect infestation, and diminished species
diversity and site productivity. The no treatment option cannot be
considered a viable alternative in most situations.
When prescribed burning is used, emissions control and/or emissions
Impact reduction can be effected by utilizing low emission fuel conditions
firing techniques, and meteorological conditions. Fuel conditions can be '
optimized and overall emissions reduced by:
regulating the time between burns to control fuel loading,
burning at lower fuel moisture contents, and
- modifying fuel arrangement to 'facilitate better air flow and more
intense and complete combustion.
326
-------
The ability to reduce emissions by altering the firing technique is
limited because firing techniques are dictated by the type of fuel to be.
burned and the objectives of the prescribed burhing. However, field and
laboratory tests have indicated that different firing techniques do have
varying levels of emissions. For example, backing fires emit less
particulate matter than heading fires.163
Utilizing meteorological conditions to minimize the impact of
prescribed burning emissions involves burning when conditions are best for
directing emission plumes away from receptor areas and for obtaining maximum
atmospheric diffusion of the plume. The key factors to examine in
determining the optimal emissions dispersion.period or situation are mixing
height, atmospheric stability, wind speed, and wind direction. Emissions
dispersal is optimal when the atmosphere is unstable and the mixing height
is high above the earth's surface.-163
A special stability situation occurs when there is an inversion layer
• in which the. air temperature increases with height. An inversion layer acts
as a lid that tends to trap rising emissions near ground level.
Consequently, prescribed burning is often precluded when an inversion layer.
is present. Conversely, inversions lower in elevation than the area where
prescribed burning is being done tend to limit emissions below the inversion
height. In this way, valleys .are often buffered from emissions due to
burning at high mountain elevations.""
163
Currently, the use of meteorological scheduling to specify times for
prescribed burning, for the purpose of reducing emission impacts, is
prevalent.
Emission
No POM emission factor data exist that are based on tests of actual
prescribed burning or uncontrolled forest fires. However, emission factor
data have been developed by the United States Forest Service by simulating
327
-------
forest burning conditions in a laboratory. In these tests, various loadings
of pine needles were burned on a metal table equipped to change slope to
simulate wind effects. All emissions' from burning were channeled through a
large stack where particulate matter was collected by a glass fiber filter
in a modified high-volume sampler. Collected samples were analyzed for POM
compounds by GC/MS. ~
and 81.165
The results of these tests are given in Tables 79, 80,
In Table 79, individual POM species and total POM emission factors
measured for backing and heading fires are presented. These data indicate
that backing fires would generate considerably greater POM emissions than
heading fires. With heading fires, the highest emissions were produced by
the maximum, fuel loading case (2.4 kg/m2). Somewhat unexpectedly, the
highest emissions from backing fires occurred with the minimum fuel loading
case (0.5 kg/m2).165 - . *
As shown in Table 79, the predominant POM compounds found in backing
fire emissions are generally different from those found in heading fire
emissions. In backing fire emissions, chrysene/benz(a)anthracene, pyrene
and methyl pyrene/fluoranthene are predominant. Benzo(a)pyrene constitutes
on average about 2 percent of total POM. In heading fire emissions,
anthracene/phenanthrene, methyl anthracene, methyl pyrene/fluoranthene, and
pyrene are the most.prevalent compounds.165 Benzo(a)pyrene constitutes on
average only about 0.4 percent of heading fire total POM emissions.
The difference in POM emissions during the flaming'and smoldering
phases of a fire are illustrated in Table 80. The smoldering phase would be
expected to produce greater POM emissions because combustion processes
during smoldering are very inefficient. The data in Table 80 confirm this
expectation
Polycyclic organic matter and benzo(a)pyrene emissions, as a function
of total suspended particulate emissions from burning pine needles are
given in Table 81 for backing and heading fires.165
-------
•rm
Sw
SL,
£
^^
Cd
Ori
ct
oa
en
Cd
^J
8
Cd
Cd
SB
Cd
35
S
u
2
I
S3
^Q
as
Ch
en
o
S
Cb
o
.-
«"»
^*
s
i
§
§
H
^
cu
0.
**•
Cd
_3
S
: ;
^
•
0(
C
^
f
<
<
(i
^
d
41
C
H
41
Ch
i
i
1.
b
01
e
e.
a
CQ
t
00
•*
u «• | 1 1 **
£B ^ Hk ^? ^^ ^^ C^ (^ flD ^B ^M ^
S322f>*>eS52222>»0>a>o>
^x«%l'»o"ooSS22233
- B -W^^C^M^^p^
•" -" es -
Q
« « N" ^- *• * ^- •- "1 ^ - 2 S S § 2
~* "* "" —in
04
!- S- ? 1 1 1 5 8 . S 3 8 2 9 g 5 «
s " s a a • a s 3" - " « 5 3 3 3
.
a S
I J § ' .
! S 2 1
S • «l £ 41
tl b o u b 41
Is 1 I ^ 1 I | J
I 2 ' • ^11 = 521 I ' ? |
111 1 i 4 II 1 1 ? M
2= Ug-S =•*•*«• S-T-0. S
1 1 j iii-Hiitiiii*
'
-------
o-e
ca «n
jjjj ^jj
M -H
2 b)
OS OS
OS S
S g
CO
CO Cd
I
vo e> oo
5? P*» rf
O i-=
o ve
o» «n
in « r
co f.
2
en
fcj
CO
|5S3S252S
M p» oo « i*.
t — I-. oo S
g fO *»^ r*.
f+ JJ TJ ^ «" CO CM
3855585
CO C1
VO 00 0<
m m ™
4
"« s
c e
41 *^ J
a. «n -^
SJQ
u
U - vS
C >. O
ai — si* O O
S "S
V 0)
e a
-------
TABLE 81. TOTAL POM AND BENZO (A) PYRENE EMISSION FACTORS FROM
BURNING PINE NEEDLES AS A FUNCTION OF TOTAL
SUSPENDED PARTICIPATE MATTER EMISSIONS165
Fire Type
Backing
Backing
Backing
Heading
Heading
Heading
______ __________
iTn-f **e **J? «»_._
Fuel Loading
kg/m* (lb/ftz)
"
0.5 (0.1)
1.5 (0.3)
2.4 (0.5)
0.5 (0.1)
1.5 (0.3)
2.4 (0.5)
—————— ——^^___.
-————-—————-——_—__
Total POM
Emission Factor3'
13,982
6,254
4,084
873
399
392
1 — — _
Benzo (a)pyrene
Emission Factor3'
274
135
98
3
2
2
•———---- -1.^** -kc.wt»wj._> cs._»@ ^*S '
suspended particulate matter-emitted.
Per g of total
Emission factors represent only particulate matter POM and benz0(a)pyrene.
331
-------
-------
Source L
~
Information provided by the U. S. Forest Service indicates that the
majority of prescribed burning in the United States occurs in the
southern/southeastern part of the country.166 As shown in Table 82, almost
60 percent of national prescribed burning in 1984 was performed in the
southern/southeastern region (Forest Service Region 8). The second most
prevalent source of prescribed burning in 1984 was the Pacific Northwest
which constituted almost 20 percent of the total. California was next in
importance of prescribed burning in 1984 with 10 percent of the national
total.
The locations of uncontrolled forest fires are not as definable as
prescribed burning sites, but the historical record of fires and a knowledge
of the locations of primary forest resources can be used to estimate where
the majority of forest fires are likely to occur. The southern region and
the western part of the country (including California, the Pacific
Northwest, and western mountain States) appear to represent the greatest
potential for POM emissions from forest wildfires.164 Forest Service data
for 1983 indicate that the southern/southeastern region of the United States
constituted 67 percent of the total number of acres burned by wildfires
nationally. The western regions of the country contained 17 percent of the
wildfire burned acreage. The northern region (Idaho, Montana, North Dakota)
of the country contained another 6 percent of acreage destroyed by
wildfires,167 y
AGRICULTURAL BURNING
Process
Agricultural burning involves the purposeful combustion of field crop
row crop, and fruit and nut crop residues to achieve one or a combination of
desired objectives. The typical objectives of agricultural burning are as
follows . '
332
-------
TABLE 82. DISTRIBUTION OF PRESCRIBED BURNING IN
THE UNITED STATES IN 1984a>166
Forest
Service Region
States Included
1
2
3
4
5
6
10
TOTAL
Idaho
Montana
Colorado
Nebraska
North Dakota
South Dakota
Wyoming
Arizona
New Mexico
Nevada
Utah
California
Washington
Oregon
Alabama
Arkansas
Florida
Georgia
Kentucky
Louisiana
Mississippi
North Carolina
Oklahoma
South Carolina
Tennessee
Texas.
Virginia
All others not in
1-8 or 10
Alaska
±984 in this case means fiscal 1984.
There is no Forest Service Region 7.
Total Acres Burned
•'"^•'•"••"^"•^•ii
35,132
19,149
34,860
28,624
91,313
159,006
525,782
10,039
1,123
905,028
333
-------
removal and disposal of agricultural residue at a low cost
preparation of farmlands for cultivation
-' cleaning of vines and leaves from fields to facilitate harvest
operations
disease control
- .direct weed control by incinerating weed plants and weed seeds
indirect weed control by providing clean soil surface for
soil-active herbicides
selective destruction of mites, insects, and rodents
The types of agricultural waste subject to burning include residues such as
rice straw and stubble, barley straw and stubble, wheat residues, orchard
prunings and natural attrition losses, grass straw and stubble, potato and
peanut vines, tobacco stalks, soybean residues, hay residues, sugarcane
leaves and tops, and farmland grass and weeds.
Polycyclic organic matter are created and emitted during agricultural
burning because mixing between the fuel (agricultural residue) and ambient
air is poor and because combustion gases from burning are effectively
quenched by surrounding ambient air. Poor mixing creates pyrolytic (oxygen
deficient) combustion conditions leading to lower temperatures less
efficient combustion, and POM formation and release. Rapid quenching of
combustion gases by the huge volumes of air surrounding agricultural burning
enhances incomplete combustion, thereby permitting the increased release of
unbumed hydrocarbons like POM. Polycyclic, organic matter may be released
from agricultural burning in gaseous form or as a liquid aerosol condensed
on solid particulate matter. 2'168
Potential POM emissions from agricultural burning are related to
factors affecting waste combustion efficiency. Waste combustion efficiency
is influenced by environmental variables, fuel conditions, and the type of
burning or fire management techniques used. Environmental variables
affecting combustion efficiency and restiltant POM emissions include air
temperature, soil moisture, relative and/or absolute humidity, and wind
334
-------
speed and direction. Of the environmental variables, wind speed is the most
important factor affecting potential POM emissions. Conditions of the fuel
(waste) potentially affecting POM emissions during burning are moisture
content, fuel composition, and fuel density. Higher moisture contents and
greater fuel densities generally correlate to increased POM emissions
because either or both of these conditions tend to reduce overall combustion
efficiency. The key fire management techniques influencing POM emissions
from agricultural burning are the type of burning used, backing fire or
heading fire, and the fuel loading level (i.e., the amount of waste burned
in a defined area).
The
ability to reduce POM emissions from agricultural burning
is
to markedly
trend would
be
primarily related to altering combustion conditions to optimize combustion
efficiency. Reducing fuel moisture' content has been shown to
reduce overall emissions from agricultural burning. A similar
be expected for POM emissions because combustion temperatures would
higher with dryer fuel and combustion would be more intense
distributing wastes prior to combustion wou
facilitate more thorough combustion. Waste moisture'content „
predominantly influenced by environmental conditions. Although not directly
controllable, environmental conditions can be optimized in terms of
positive effects on reducing waste moisture content. Lower humidity
air temperatures, and higher wind speeds would tend to reduce
Evenly
aid drying activities and
is
in terms of their
high
levels
waste moisture
m
The use of backfiring (fire progresses in a direction opposite to that
of the wind) instead of headfiring (fire progresses in the same direction as
the wind) techniques for agricultural burning has been shown to
significantly reduce overall particulate emissions. Because backfiring
creates a longer waste residence time in the combustion zone, combustion
should be more complete and potential POM emissions less.168
The most effective means* to control POM emissions from agricultural
burning is to find alternatives to combustion as a means for accomplishing
field sanitation, residue removal; and residue disposal. The most common -
-------
method of disposing of crop residues is to incorporate the material back
into the soil by tilling. This technique only accomplishes residue
disposal. Field sanitation is not.addressed; however, it may not be an
objective in all cases. Mechanical removal of residues is a possible option
to burning; however, it is generally expensive, it does not address field
sanitation, and the problem of waste disposal still exists once all wastes
have been collected. If agricultural burning is only being performed for
field sanitation purposes, chemical applications may be a suitable •
alternative to burning. The effects on the environment of applying
herbicides, fungicides, and pesticides to areas for weed, disease, and pest
control would have to be weighed against the potential air emissions from
burning.168
As with other types of open burning, meteorological scheduling can be
used to lessen the impact to receptor areas, from agricultural burning
emissions.' Utilizing meteorological conditions to minimize the impact of
agricultural burning emissions involves burning when conditions are best for
directing plumes away from receptor areas and for obtaining maximum
atmospheric diffusion of the plume. The key meteorological factors acting to
potentially bring about these conditions are atmospheric mixing height
atmospheric stability, wind speed, and wind direction. Generally, optimal
conditions for burning would be the existence of an unstable atmosphere and a-
high atmospheric mixing height.
Emission Faeto-rg
Few POM emission factors exist for agricultural burning.. The factors
that are available pertain only to benzo(a)pyrene. Burning of whole sugar
cane residue was found to produce a particulate benzo(a)pyrene emission
factor of 0.00027 kg/Mg (0.00053 Ib/ton) of waste burned. Particulate
benzo(a)pyrene emissions from burning sugar cane leaf trash were found to be
0.00021 kg/Mg (0.00042 Ib/ton) of waste burned.168
336
-------
Additional POM emission factors that may be applied to certain types of
agricultural burning are provided in the discussion in Section 4 on
Miscellaneous Open Burning.
Source
types
Agricultural burning is directly correlated with States having a
significant agriculture industry. Major agricultural States comprising the
majority of agricultural burning include California, Louisiana, Florida,
Hawaii, North Carolina, Mississippi, and Kansas . 162
MISCELLANEOUS OPEN BURNING
The miscellaneous, category includes any and all open burning activities
not covered in the discussions on coal refuse banks, prescribed burning,
forest fires, and agricultural burning. The most readily identifiable '
of open burning in the ..miscellaneous category are municipal refuse ope..
burning; open burning of automobile tires, bodies, and components; open
burning of waste railroad ties; and burning of landscaping refuse (grass
clippings, leaves, and branches). The purpose of burning in most of these
cases is volume reduction' to facilitate easier final disposal of the waste
material. In the case of automobile body burning, burning is performed
to expedite the recovery and recycling of usable metal in the automobiles by
removing all organic materials (plastic, vinyl, etc.).170
The procedure of open burning in any of the miscellaneous categories is
relatively simple. The material to be burned (domestic trash, leaves, etc.)
is collected and aggregated in an open space fully exposed to the
atmosphere. The materials are ignited and allowed to bum and smolder until
all combustible material is consumed or the desired degree of volume
reduction is achieved. Combustion efficiency in such operations is typically
poor. Potential POM emissions from such operations are highly variable
because waste moisture content and combustion conditions (air flow, oxygen
levels, waste configuration, degree of exposed surface area) are quite
337
-------
variable from site to site and within the same site. In addition, some
wastes may contain organic constituents that are precursors to POM compounds
or that accelerate POM compound formation.
Generally, there are two means to control POM emissions from
miscellaneous open burning--enclosure of the burning with exhaust
ventilation to standard control devices and prohibition of open burning! In
most areas of the United Sates, open burning of municipal refuse,
automobiles, and grass, leaves, etc., has been greatly restricted, and in
the case of municipal refuse and automobiles, completely prohibited. Open
burning of grass and leaves has been controlled by requiring collection
agencies and the general public to have permits for burning.
The available emission factor data for open burning of municipal refuse
automobiles, and landscaping refuse are presented in Tables 83, 84, and
85. The data in Table 83 represent measured POM emission factors from two
sets of open burning tests. One set of data was measured in the smoke
plumes during outside open burning of municipal refuse, automobile tires,
automobile bodies, and landscaping refuse. A second set of data was
measured in a laboratory research facility designed to simulate and
characterize open burning emissions. The laboratory experiments also burned
automobile components, municipal refuse, and landscaping refuse.
82
^ In Table 84, the results of the laboratory open burning tests are
presented as a function of the amount of waste burned as opposed to being a
function of total particulate matter emissions as in Table 83. The factors
in Table 84 could be determined for the laboratory open burning tests
because all conditions of the tests such as emission rates, flow rates,
waste throughput, etc., could be controlled; whereas, conditions could'not
be duplicated or controlled outdoors.
338
-------
339
-------
tN
Z
ai
09
S
I
O
H
I
Cfe
o
§
i
03
Z
Ed
a.
O
ea
03
•o
I
g
=
41
b «J
JJ
a u
X ai u
ii «-
u a. -.
••4 *~* T3 o
"9
•a
I-a
b 91
i^
« | £
-•a .2»
« B no
.C (• u
*• — •* 0)
IJ5
b b &A m
2 X = «
« fi« *H CD
« i £*
•S3
a 4i
S.S
°a
"a
e
ea
• X • 41
!•* -3S
,^2 . .'"S
! 3 B *
! B 41 Ob
i J3 4^
S «
B e
O 4>
a S
•« a.
I!
X 41
• 0.
41
! II
a B
S £
b b
•i
J= S J(
O. .„ o
O I
03
CO
l-l
X
w
oo
1
•S..2
u
a
u
a
u
CO
5 S 3 u
- * O S
iB b w
i a « jj u
IB"
•* 12 s
« 3 "
- a •
a B a
S 2 *
o s
b
S.3
S.
• *4
U
••«
B
41
Ci
3
W 1M
4) ** 4)
•a w V u
j< — a a
u u u '•
01 — 41
a
u
340
-------
TABLE 85. POM EMISSION FACTORS FOR THE OPEN BURNING OF MUNICIPAL
REFUSE, AUTOMOBILES, AND LANDSCAPING REFUSE82'169
Type of Open Burning
Municipal Refuse
Automobile Tires
Automobile Bodies
Automobile Components
Grass, Leaves, and
Branches
!
Leaf Burning (red oak
leaves)
Leaf Burning (sugar
maple leaves)
Leaf_Burning (sycamore
leaves)
Leaf Burning (composite
of red oak, sugar maple,
and sycamore leaves)
Total POM and Banzofa') pyrene Emission Factto-rg
Total POM Benzo(a)pyrene
mg/kg of mg/ks of
Waste Burned Waste Burned
0.5** - 4.7
240a
110a
190 - 260
2.5a - 9.2
11.2 - 78b
12.0 - 18.7b
13.4 - 22.lb
10.2 - 21.8b
0.088 - 0.340
55a
14a
20 - 29
0.30 - 0.35
b.c
0.097 - 0.41
Not detected - 0.79
0.27 - 0.54
b.c
0.12 - 0.28
b,c
emission factors were calculated from the smoke plume POM emission
factors given in Table 83 (Reference 82) and the emission factors for total
"T 6? emissions &™ ^ AP-42 for open burning. The J. 42
emissjon facers used are as follows: municipal refuse -
kg/Mg; automobile components - 50 kg/Mg; landscaping refuse - 8.5 kg/Mg.
dJT?loped ^ —Pllns at a leaf burning research facility using
h m Jf aBd a TenaX adsorbe*« S«plM were analyzed by GC/MS.
The POM compounds detected include anthracene/phenanthrene , methyl
SSSJS*8' fluoranthene, pyrene, methyl pyrene/f luoranthene , benZO(c)-
phenanthrene, chrysene/benz ( a) anthracene, methyl chrysenes, benzo-
S ™«^W««i-/b«jo<.)pyr«1., perylene, methylcholanthrenes ,
' be^°(S.h.i)perylene, dibenzo( a, h) anthracene,
c(c,g)carbazole, dibenzo(a,i and a,h)pyrenes, and coronene.
CBenzo(a)pyrene emission factors from these tests represent combined
benzo (a)pyrene and benzo (e) pyrene emissions.
-------
Although emission factors for POM from open burning of creosote
railroad ties have not been quantified, this type of open burning has been
shown to produce emissions containing several POM compounds.171 The POM
compounds that have been identified are listed below.171
^
naphthalene
acenaphthylene
fluorene
acenaphthene
phenanthrene
anthracene
fluoranthene
pyrene
chrysene
benz (a)anthracene
benzo(b)fluoranthene
benzo(k)fluoranthene
benzo(a)pyrene
dibenz(a,h)anthracene
o-phenylenepyrene
benzo(g,h,i)perylene
The predominant compounds measured were acenaphthene, acenaphthylene,
benz(a)anthracene, benzo(a)pyrene, and dibenz(a,h)anthracene.
342
-------
-------
REFERENCES FOR SECTION 4
1'
°" "Mi** P?M f113310113 from Stationary Conventional Combustion
Processes, with Emphasis on Polychlorinated Compounds of
^!n?°"P"di°Xin (PCDDs)« Biphenyl (PCBs) , and Dlbenzofuran (DCDFs)
CCEA Issue Paper presented under EPA Contract No. 68-02-3138
Industrial Environmental Research Laboratory, U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina
January 1980.
2' Pa-rAi:Ula£e Polvcyflic Organic Matter. Committee on Biologic Effects
of Atmospheric Pollutants, Division of Medical Sciences, National
Research Council. National Academy of Sciences. Washington D C.
3. Polycyclic Aromatic Hydrocarbons: Evaluation of Sources and Effects
Committee on Pyrerie and Selected Analogues, Board on Toxicology and '
Environmental Health Hazards, Commission on Life Sciences, NaSonal
Research Council. National Academy Press, Washington, D.C'. 1983
4. Mead, R C., G. W. Brooks, and B. K. Post. Summary of Trace Emissions
Oiimramd, R®comm®ndations °f Risk Assessment Methodologies for Coal and
Oil Combustion Sources. Prepared under EPA Contract No. 68-02-3889
P^L^*51?611* 41' Pollu^ant Assessment Branch, U. S. Environmental
Protection Agency, Research Triangle Park, Nor«i Carolina. July 1986.
of Polynuclear Hydrocarbons and
Trt,,™ai f *.u A-'"«"-.;" Generation and Incineration Processes.
Journal of the Air Pollution Control Association. 14(7): 267-278
July 1964. "
Kelly M. E. Sources and Emissions of Polycyclic Organic Matter *EPA
HeenLch°iri^;8^rNc«h LoST^I?1 ^S'S.'K?'
al. Planning Studies for Measurement of Chemical
Ga«*.s of Coal-fired Power Plants. EPRI Report
1983. •" " Research Institute, Palo Alto," California.
9.
PolvuCl <-' ' ^on and Quantification of
Polynuclear Organic Matter (POM) on Particulates from a Coal-fired '
Power Plant. EPRI Report No. EA-1Q92. Electric Power Research
Institute, Palo Alto, California. 1979. research
W> Measurements of POM Emissions from Coal-fired
343
-------
10.
11.
12.
13.
Haile, C. L., et al. Comprehensive Assessment of the Specific
Compounds Present in Combustion Processes. EPA Report
No. 560/5-83-006. Office of Pesticides and Toxic Substances,' U S
Environmental Protection Agency, Washington, D.C. 1983.
Shih, C. C. et al. Emissions Assessment 6f Conventional Stationary
Combustion Systems, Volume III - External Combustion Sources for
Electricity Generation. Prepared under EPA Contract No. 68-02-2197
Industrial Environmental Research Laboratory, U. S. Environmental '
Protection Agency, Research Triangle Park, North Carolina. 1980.
Hangebrauck, R P., et al. Sources of Polynuclear Hydrocarbons in the
Atmosphere Public Health Service Report No. AP-33. U. S. Department
Sfif ™*7Education« and Welfare; Public Health Service. Cincinnati,
unio. iyo/. pp. 5-14.
Suprenant, N. P., et al. Emissions Assessment of Conventional
Stationary Combustion Systems, Volume V - Industrial Combustion
Sources. Prepared under EPA Contract No. 68-02-2197. Industrial
f^™13*?**1 R!S^a5Ch Laborator7. U. S. Environmental Protection
Agency, Research Triangle .Park, North Carolina. 1980.
^
Industrial
T' °-f ?* ?• Field Tests of I«*»trl«l Coal Stoker Fired
Hvdroc Jn°rsanic Trace Element S«d Polynuclear Aromatic
Hydrocarbon Emissions. EPA Report No. 600/7-81-167 U S
Research Tria«Sle Park,' North
16.
17 '
Suprenant, N. p.. et al. Emission Assessment of Conventional
Stationary Combustion Systems, Volume IV - Commercial/Industrial
Combustion Sources. Prepared under EPA Contract No 68-5^^97
PrOteCtion A&™7> ^search Triangle Park, North
from Residential and Small Commercial
Smokeless Operation. EPA Report
A^'
344
-------
20.
21.
22.
23.
24.
25.
26.
27.
lTTi -,- .»«ticulmt. Emissions Characteristics of
Oil-fired Utility Boilers - Phase II. KVB Report No. 1-21610-1107
KVB, Inc., Elmsford, New York. July 1981.
28.
29.
30.
Inventory of Organic Emissions from Fossil
Generation- EPRI Report No. EA-1394.
Electric Power Research Institute, Palo Alto, California. April 1980.
Tennessee Valley Authority. Wood-fired Boiler Test Report - Down-Draft
™
Tennessee Valley Authority.
Wood-fired Boiler Test Report - Stick
and communty
A PolycyoUc Organic Materials Emissions Slsady for Industrial
" 3- NCA Bulletin No. 400. NatJLl
D. Malek, eds. Oregon Graduate Center. Beaverton, Oregon. 1982 .
345
-------
31. Haxtman, M. W. and G. D. Rives. Literature Review and Survey of
Emissions from Residential Wood Combustion and Their Impact EPA
Report No. 660/2-85-047. U. S. Environmental Protection Agency
Research Triangle Park, North Carolina. April 1985.
32. Peters, J. A., et al. An Environmental Assessment of POM Emissions
from Residential Wood-fired Stoves and Fireplaces. In: Polynuclear
Aromatic Hydrocarbons: Chemical Analysis and Biological Fate
Proceedings of the Fifth International Symposium on Polynuclear
^omatic Hydrocarbons, 1980, Columbus, Ohio. M. Cooke and A. Dennis,
eds. Battelle Press, Columbus, Ohio. 1981
Snowderi, W. D., et al. Source Sampling Residential Fireplaces for
Emission Factor Development. EPA Report No. 450/3-76-010 U S
Environmental Protection Agency, Research Triangle Park, North
Carolina. August 1975.
33.
34* £*?• R> C" et al- Investigation of PAH and Polychlorinated Organic
Pollutant Emissions from Wood Combustion Sources Inf Jolynucllar
Aromatic Hydrocarbons: Formation, Mechanism, and MeluremeS
Proceedings of the Seventh International Symposium on Polvnuclear
Aromatic Hydrocarbons, 1982, Columbus, Ohio^M. Cooke^nf A? D^nis,
eds. Battelle Press, Columbus, Ohio. 1983.
35. Knight, C V., et al. Polynuclear Aromatic Hydrocarbons and
Associated Organic Emissions for Catalytic and Noncatalytic Wood
Heaters. In: Polynuclear Aromatic Hydrocarbons: Formation
Mechanism, and Measurement, Proceedings of the Seventh International
MtoST.0!? f01;™*163* Aromatic Hydrocarbons, 1982, Columbus, Ohio.
M. Cooke and A. Dennis, eds. Battelle Press, Columbus, Ohio. 1983
36.
37.
Residential w°°
-------
41. Gibson T. L., A. I. Ricci, and R. L. Williams. Measurement of
S5E- i- -sr-ss^su.
srss-s^ss £*£.- ^SL^sr^ °f
Hydrocarbo-ns, Columbus, Ohio, 1980. M. Cooke and A J Dennis W«=
Battelle Press, Columbus, Ohio. 1981. pp. 707-717! ' '
42. Reference 3, pp. 1-1 to 1-35.
43' M:±:»f;i«n^Kn.!fSL,..!!^=Uar,A--«° Hydrocarbons and the
Press, Columbus, Ohio.
44. Handa, T., T. Yamamura,
Analysis and Derivation
pp.
45. ^rs°^r*. S.,'J. Ingwersen, T. Nielsen, and E. U
Effects of
" of
46.
47. Reference 2, pp. 15-21.
48. Reference 12, pp. 33-37.
49.
1984. U. S,
50< S55^' ?;: "• ?!is2' and E' L« W3mder. Studies on the
51.
52.
347
-------
53.
59.
60.
61.
62.
Hoffman, D. , and E. L. Wynder. Studies on Gasoline Engine Exhaust
Journal of the Air Pollution Control Association. 13? 322-327 1963
Ef«C' ?' C,f^in?Se^iC **•«** Hydrocarbons in Automobile
Effluents. In: Vehicle Emissions --Selected SAE Papers Paper 440C
•presented January 1962 at the SAE Automotive Engineering Confess
Society of Automotive Engineers, New York. pp. 163-174 1964
h,-0;- P'P J1"**™"1 ReP°rt «» Gasoline Composition and Vehicle
Exhaust. Gas Polynuclear Aromatic Content. U. S. Department of Health
Education, and Welfare, Durham, North Carolina. 1970 Health'
56.
57.
58.
Automotive Engineers, Detroit, Michigan. 1970.
of
«* Protection Agency,' Office of Air
Waste Management, Ann Arbor, Michigan. 1974.
Levin, J C. Nilsson, and A. Norstrom. Sampling and
"-!-s'~'
of
Robertson, D. J., R. H. Groth, and T. J. Blasko Organic
Particulate Matter in Turbine Engine Exhaust. Journfl ol
Pollution Control Association. 30(3): 261-266 1980
wo. b8-02-2836. Pollutant Strategies Branch, Office of Air
rianning and Standards, Research Triangle - - -
November 1°'*0 — -~ «*"BJ-«»
*'
*' Black. R. Zveidinger, and
Automotive Engineers, Warrendale, Penn^lvania i981? ^ f
348
-------
64
67.
68.
72.
73.
74.
°f
65.
66.
fr™ ?;; ™di°; S;hue?le- Sampling, Extraction, and Analysis of PAH
from Internal Combustion Engines. In: Handbook of Polycyclic
Hydrocarbons, A. Bjorseth, ed. Marcel Dekker Publishing Smpany.
Grimmer, G. Analysis of Automobile Exhaust Condensates In- Air
Pollution and Cancer in Man. IARC Scientific Publication No' 16
? SfTl- ASTy f°r RSSearch °n Cancer" Ly°«> France.
U. Mohr et al., eds. pp. 24-39. 1977.
Grimmer G., H. Boehnke, and A. Glaser. In: Polycyclic Aromatic
f 2l1-234
*b«..l: Orig. Reihe B
of
Engineers, HarrendaU, Pennsylvania. 1981 - ^
Reference 63. 1981.
^
70.
71' »' 1'*™?°™*^ Protection Agency. Federal Register 45(14)
p. 4161 et seq., January 21, 1980. — •*•>*>*•'**,
349
-------
75.
76.
80.
84.
85.
Bricklemyer B. A and R. S. Spindt. Measurement of Polynuclear
Aromatic Hydrocarbons in Diesel Exhaust Gases. SAE Technical Paper
Presented at the Congress and Exposition. Cobo Hall, Detroit,
Michigan, February 27 - March 3, 1978.
Lee, F., T. J. Prater, and F. Ferris. PAH Emissions from a
Stratified-Charge Vehicle With and Without Oxidation Catalyst"
Sampling and Analysis Evaluation. In: Polynuclear Aromatic '
Hydrocarbons Proceedings of the Third International Symposium on
Chemistry and Biology - Carcinogenesis and Mutagenesis! P. W
a^^*H U T Afc«M^B _. ^3 — *___»* ._ - - _ W *
and P. Leber, eds. Ann Arbor Science Publisher^ Inc.
Michigan. 1979. pp. 83-110.
Jones
77.
78.
Ann Arbor,
?' B< J1"613011' and R- B- Zweidinger. Measurements of
Aromaji° Compounds in the Cylinder of an Operating Diesel
^^ *~~ ^ncyAesearc
-tent of
Pollution Control Association. 30(3): 261-266. 1980."
79. Reference 6, pp. 5-75 to 5-82.
Municipal Waste Combustion Study Data Gathering Phase. Radian '
Corporation. Prepared under EPA Contract No. 68-02-3818 Work
Assignment 31. Pollutant Assessment Branch, U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina. July 1986.
81* £«£!£?* Inf°rmatio« D«<=^ent for Cadmium Emission Sources. Radian
Corporation. Prepared under EPA Contract No. 68-02-3818 Work
Assignment 23. Pollutant Assessment Branch, U. S. Environmental
pp? 106-145 C7> ReS6arch TrianSl* ^rk, North Carolina. May 1985.
82. Reference 12, pp. 14-18.
83.
350
-------
86.
87.
88.
"•
?^yt°?' P< ^ Incineration of Municipal Refuse. Warren Spring
Laboratory, Stevenage, England. 1984. apnng
, C. L., et al. Assessment of Emissions of Specific Compounds
N.. s^s^o^^J^T?^1 SSS^TTi^iSS
Protection Agency, Washington, D.C. June 1984. ^ironmental
Reference 1, pp. 17-21.
Offio of Mr
90' Eifi!'«:./• ,Assess^"t of Organic Emissions, from the
91.
Of
and
Conpounds in
Press, Columbus, Ohio. 1985.
92.
Oho
P^otection Ag«wqr. Seminar Publication:
sUdf ?°mbustion Technology. EPA Report
Sepmber l985. Envir°™tal P-^ction Agency, Cincinnati,
93.
94.
95.
- Eastern
Group to
Reference 81, pp. 71-83.
.7L
Environment Research Laboratory, U. S. Environmental Protection
earCh Trianle Park N< 1980
96. Reference 2, pp. 26-27.
97. Reference 12, pp. 18-25.
98- SS^iafcuns^^ST"7: Oil "*•Gas Journal-
351
-------
99.
100.
103.
104.
105*.
106.
107.
Reference 6, pp. 5-58 to 5-62.
GCA Corporation. Survey of Cadmium Emission Sources. EPA Report
No 450/3-81-013. Office of Air Quality Planning and Standards U. S
Environmental Protection Agency, Research Triangle Park, North
Carolina. September 1981.
?;Jl 2«^ronmental Protection Agency. An Investigation of the Best
Systems of Emission Reduction for Sinter Plants in the Iron and Steel
lT,dus*™ Preliminary Report. 'Office^of Air Quality Planning and
Research
102. Reference 61, pp. 78-79.
U. S. Environmental Protection Agency. A Review of Standards of
S^f??^1106 f™.N^W Stationary Sources - Ferroalloy Production
Facilities. EPA Report No. 450/3-80-041. Office of Air Quality
Planning and Standards, U. S. Environmental Protection Agency
Research Triangle Park, North Carolina. December 1980.
SfcJrf ^irjnmenjal Protection Agency. Background Information for
Standards of Performance: Electric Submerged Arc Furnaces for ' -
w A?n/?n-,f oforr°alloys' V°lume I: Proposed Standards. EPA Report
US £---*"•• 2-fiCe °f Air- Q"*1^ Planning and Standards,
c Prot^ct:ion Agency. Locating and Estimating Air
Sources of chromium. EPA Report No. 450/4-84-007g.
t« Uali7 Plaimin§ and Standards, U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina. July 1984.
!' E; Ferroallovs- m: 1984 Bureau of Mines Minerals
Yearbook. Bureau of Mines, U. S. Department of Interior U S
Government Printing Office, Washington, 'D.C. 1985.
Protecti011 Agency. Electric Arc Furnaces in
InfoCTat*™ ^r Proposed Standards
Office of Air Quality Planning and
'
EPA Report No.
108.
109.
Jou« 5' ?' Polycycli° Aromatic Hydrocarbons in Foundries.
Journal of Toxicology and Environment Health. 6(5-6): 1187-
September/November 1980. .
352
-------
'
f" Identifica*i°* of Polycyclic Aromatic
steel Castins-
111.
112'
McCalla D. R et al. Formation of Bacterial Mutagens from Various
Mould Binder Systems Used in Steel Foundries. In: Polynuclear °
^o±dinSdrrSbT:, J**"1—. -thod.. and MetaboS^
SSE^SV v Eighth International Symposium on Polynuclear
Aromatic Hydrocarbons, Columbus, Ohio, 1983. M. Cooke and
pp. 871-8843' ^^ Battelle Press« Columbus, Ohio. 1985.
Triangle Park, North Carolina. August 1985.
114.
115.
116.
U7-
Reference 6, pp. 5-46 to 5-56.
Barrett, R. E. , et al. Effectiveness of a Wet Electrostatic
* ? f°r ControllinS «« Missions from C^ke O^en Door"
r^^ ?resented at the 71st Annual Meeting of thelir
Paper No 78-1" ASS°Ciation' «™**™> Texas, JuneS25-30 »78.
Reference 3, pp. 2-23 to 2-26.
118. Reference 61, pp. 64-71.
119.
Deteraination of PAH Pollution at Coke Works
120.
figs:
Gas Chromatographic-Mass
, 00.
353
-------
121*
; Protection ASenc7- Asphalt Roofing Manufacturing
Industry - Background Information for Proposed Standards EPA Report
No 450/3-80-021*. Office of Air Quality'pianning andl^anSrds'?
122. Reference 6, pp. 5-62 to 5-72.
123 ' SrD1*' R\TW> AtmosPhe*i<= Emissions from Asphalt "Roof ing Processes
EPA Report No. 650/2-74-101. Control Systems Laboratory U S
Environmental Protection Agency, Research Triangle Park, North'
Carolina. October 1974.
124. Reference 61, pp.. 72-78.
125. Reference 12, pp. 25-26.
126. Letter from the Asphalt Roofing Manufacturers Association to
Cruse, P., Radian Corporation. June 11, 1986. List of asphalt
roofing manufacturing plants. spnaic
127 " ?^?AJ!nV3:r0^en?1JPr0t:eCti0n ASencv- Second Review of New Source
Performance Standards for Asphalt Concrete Plants. EPA Report
~ e °f A±r Qualifcy PlanninS and Standards! U. S.
Agency' Research Triansie
128
Reference 6, pp. 5-63 to 5-68.
'
130.
131.
132.
Q*S*nic Material Emissions from
Paper Present«d at the 70th Annual Meeting of
Toront0' Ontari0' Canada«
Reference 12, pp. 27-28.
Reference 61, pp. 72-76.
Puchs, M. R., et al. Emission Test Report: T. J. Campbell Asohalt
oSirVAf^ °^ah°ma City" Oklah0-- EPA/EMB Report S. 83 -IsP-4
Sff^L? ^r QPaUS7 PlanninS and Standards, U. S. Lvironmental
Protection Agency, Research Triangle Park, North Carolina. May 1984.
5S?'.?' W'^^d T> W> HUgheS' Polvcyelic Organic Matter (POM) and
Trace Element Contents of Carbon Black Vent Gas. Environmental
Science and Technology 14(3): 298-301. March 1980.
' HUgheS' S°urce Assessment: Carbon Black
^^^^
, North Carolina. October
-------
135.
136.
137.
Reference 6, pp. 5-85 to 5-88.
Reference 61, pp. 80-82.
, S. E. andR. E. Barrett. Sampling and Analysis of Source
on Saamles ft-™. a Carbon Black Plant. EPA/EMB Report
of Air Quality Planning and Standards, U. S.
May
142.
143.
144.
138.
139. Reference 3, p. 2-35.
14°'
Measurement
Aromatic
141.
Research Triangle Park, North Carolina. FebruarTl979\
Reference 61, pp. 82-85.
~
146. Reference 138, p. 412.
147. Reference 6, pp. 5-89 to 5-93.
355
-------
148.
153.
154.
155.
156.
157.
Moscow! tz, C. W. Source Assessment: Charcoal Manufactures -
State-of-the-Art. EPA Report No. 600/2- 78- 004z. Industrial
Environmental Research Laboratory, U. S. Environmental Protection
Agency, Research Triangle Park, North Carolina. December 1978.
149.
150.
151.
152.
of taSSSS"? °fnAS?iCUltUre* ^ Biologic and Economic Assessment
of Pentachlorophenol, Inorganic Arsenicals, and Creosote - Volume I-
Wood Preservatives. USDA Technical Bulletin No. 1658-1 U S
Department of Agriculture, Washington, D.C. November 1980.'
Andersson, K. , et al. Sampling and Analysis of Particulate and
.Gaseous Polycyclic Aromatic Hydrocarbons from Coal Tar Sources in the
Working Envxronment. Chemosphere. 12(2): 197-207. 1983.
°0d Preservation Statistics, 1983 and 1984.
^ American Wood
Cas
« and ^tabolism, Proceedings of the
fes-
?Format:ion and Transformation of Particulate Polycyclic
* nnitt:ed fr°m C°al Fired Power Plants ^ 0^ Shale
Retorting. DOE Report No. DE84-012747, DOE/EV/04960-TI U S
Department of Energy, Washington, D.C, April 1984.
Potential' BCh f.^^^^li^tion: The Risks and the
Potential. Chemical Engineering. 88(18): 63-71. September 7, 1981.
Existence of commercial oil shale retorting operation! in the United
356
-------
158.
162
164,
165.
166
167.
Malaiyandi, M. , et al. Measurement of Potentially Hazardous
loo??™ H ^omatic Hydrocarbons from Occupational Exposure During
Roofing and Paving Operations. In: Polynuclear Aromatic
Biol°Sical Chemistry, Proceedings of the
Battelle Press, Columbus, Ohio. 1982. pp. 471-489.
160.
' eds-
Ooraan R and G. M. Liss. Occupational Exposure to Pitch in a
Hon-heated Process - Photoxicity and Pre-malignant Warts? In-
Snes,"
-, »- earc bo
H. S. Environmental Protection Agency, Cincinnati, Ohio. July 1978.
Reference 6, pp. 3-.'3 to 5-9. •
Reference 61, pp. 99-102.
c
357
-------
168" £!™io A-^«- C-TIA n ^- », ._ e* *«-«***•<*•«•«*
.
Burning, State of the Art. EPA.Report
No. 600/2 77
Inustral
. . o. - a Inust
Environmental Research Laboratory, U. S. Environmental Protection
Agency. Research Triangle Park, North Carolina. July 1977
169. Reference 61', pp. 92-97.
170. Reference 6, pp. 5-93 to 5-95.
171. Becker, D,
et
al. Open Burning of Creosote Treated Rail Ties-
eae
Case Study in Health Risk Assessment. p presented
Annual Meeting of the Air Pollution Control Association an
Francisco, California, June 24-29, 1984. Paper No. 84?io2 6
at the 77th
172. Municipal Waste Combustion Study. Emission Data
Base for
358
-------
SECTION 5
SOURCE TEST PROCEDURES
Several sampling and analysis techniques have been employed for the
quantification of POM-. The selection of sampling and analytical techniques
is driven by"the nature of the emissions source, the quantity of POM
present, and the specific POM compounds of interest. With the exception of
real time techniques, quantification of POM involves three steps:
(1) sample collection, (2) sample recovery and preparation, and
(3) quantitative analysis. This section briefly describes general
methodologies associated with each of these steps that have been published
in the literature. No attempt has been made to produce an exhaustive
'listing or a detailed description of the many .methodologies that have been
used. The purpose of this section is to present basic sampling and analysis
principles and examples of how these principles have been applied to various
emission sources. The presentation of these published methods in this
report does not constitute endorsement or recommendation or signify that the
contents necessarily reflect the views and policies of the U. S.
Environmental Protection Agency.
SAMPLE COLLECTION METHODS
The major objective of POM measurement is the quantitative capture and
recovery of both particle-bound and vapor phase constituents, while
simultaneously preserving the integrity of the sample. A second important
factor in sample collection is the ability to capture sufficient quantities'
to allow subsequent chemical analysis. Although collection methods take
different forms, most are similar in principle, utilizing both filtration
and adsorption collection techniques. This section presents an overview of
some of the more prevalent POM sample collection methods as applied to
(1) stationary sources, (2) mobile sources, and (3) fugitive sources and/or
ambient air.
359
-------
Stationary Sources--
Collection of POM material from stationary sources is generally
achieved by using a sampling system that captures both particulate and
condensables. " The most prevalent method is the Modified Method 5
Sampling Train (MMS) which is equipped with a sorbent resin for collection
of condensables. Another method, the Source Assessment Sampling System
(SASS), a high volume variation of MMS, has found application when large
sample sizes are required. Methods which are not specifically designed to
optimize collection of condensables have also been used, and are reported in
the literature. ' A brief description of the MMS and the SASS trains is
provided. General characteristics of each method are compared in Table 86.6
A detailed procedures manual describing each of these methods is available
in a. separate report.
Modified Method 5 (MMS)--The MMS sampling train (shown in Figure 40) is
an adaptation of the EPA Method 5 train commonly used in measuring
particulate emissions.7 The modifications are the addition of a condenser
and a sorbent module between the filter and the impingers. The condenser
cools the gas stream leaving the filter and conditions the streams prior to
entering the sorbent module. The sorbent module contains a polymer resin
designed to adsorb a broad range of volatile organic species. A variety of
resins have been used including Tenax, Chromsorb 102, and XAD-2, with XAD-2 '
being the most widely recommended .for vapor phase organic compounds
including POM. After the sorbent trap, the sample gas is routed through .
impingers, a pump, and a dry gas meter. The MMS train is designed to
operate at flow rates of approximately 0.015 dscmm (0.5 dscfm) over a 4 hour
sampling period. Sample volumes of 3 dscm (100 dscf) are typical.
A major advantage of the MMS train is that the method provides both a
quantitative sample for POM analysis and a determination of particulate
loading (front half filterable particulates) comparable to EPA Method 5. A
disadvantage is that large sampling periods are required to collect enough
sample to support chemical analysis.
360
-------
TABLE 86. COMPARISON OF MODIFIED METHOD 5
,6
TRAIN/SASS CHARACTERISTICS'
Characteristic
Inert materials of construction
Percent isokinecity achievable
Typically used to traverse
Particle-sizing of sample
Sample size over a 4-6 hour
period (dscm)
Sampling flowrate (dscmm)
MM5 Train
Yes
90 - 110
Yes
No
3
0.02 - 0.03
Assuming reasonably uniform, non-stratified flow.
SASS
No
70 - 150a
No
Yes
30
0.09 - 0.14
361
-------
Stack Wail
Filter Holder
Water Jacketed Condenser
Jacketed
Sorfoent Module
Reelrculation Pump
Dry Gas Meter
Air-Tight Pump
Vacuum Line
Figure 40. Schematic of a. Modified Method 5 sampling train.
-------
Source Assessment Sampling Svaten fSASS^.-Th^ SASS train (shown in
Figure 41) is a multi-component sampling system designed for the collection
of particulate, volatile organics and trace metals. Three heated cyclones
and a heated filter allow size fractionation of the particulate sample.
Volatile organic material is collected in a sorbent trap containing XAD-2
resin. Volatile inorganic species are collected in a series of impingers
before the sample gas exits the system through a pump and a dry gas meter.
Large sample volumes are required to ensure adequate recovery of sample
fractions. The: system is designed to operate at a flow rate of 0.113 scmm
(4.6 scfm). Sample volumes of 30 dscm (1000 dscf) are typical.
An advantage of the SASS train is that the sample is collected in a
manner that allows a determination of the amount of POM associated with each
of the particle size fractions. Another advantage is the large quantity of
sample collected, which makes SASS the sampler of choice when a -large
variety of chemical and bioassay analyses are desire*. A disadvantage to
using the SASS train is that the system is not designed to have the ability
to traverse.the stack. Also, the need for constant flow to assure proper
size fractionation renders the SASS train less amenable for compliance
determinations since isokinetic conditions are not achieved. Isokinetic
conditions can be maintained at the sacrifice of particle sizing capability
Another drawback includes potential corrosion of the stainless steel
components of the SASS train by acidic stack gases.
Mobile Sources--
Two general approaches have been used for sampling vehicle exhaust POM
The first and most widely used is a dilution tube sampling arrangement
identical^^the system used for' measuring criteria pollutants from mobile
sources. The second approach involves direct sampling of raw exhaust
gases using condensation techniques.12'13 The following subsections provide
an overview of each of these approaches.
Petition
- Dilution techniques have been widely us^d for
sampling auto exhaust since in theory, dilution helps simulate the
conditions under which exhaust gases condense and react in the atmosphere.
363
-------
t/3
2
U-l
u
D
.n
a
to
fe
-------
Both regulated and nonregulated emissions from mobile sources are sampled
using dilution techniques.10'11'14 Figure 42 shows a diagram of a vehicle
exhaust dilution tube and sampling arrangement.12 Vehicle exhausts are
introduced at an orifice where the gases are cooled and mixed with a supply
of filtered dilution air. The diluted exhaust stream flows at a measured
velocity through the dilution tube and is sampled ispkinetically. Many
investigators have reported particulate-bound POM concentrations based on
filter samples collected from the diluted exhaust stream.15"18 Other
investigators have coupled filtration and adsorption techniques for the
purpose of capturing both gas phase and particulate bound POM.19"21
Table 87 shows a distribution of.particulate bound and gas phase POM
collected from vehicle exhaust.22 As seen in the table, and confirmed by
other investigators, substantial amounts of light three and four ring POMs
(as polycyclic aromatic hydrocarbons) exist in the gas phase.22'23 An
example of^a filtration/adsorption sampling arrangement is shown in
Figure 43. Particle-bound POM is captured by filtration, while a-
polymeric adsorbent trap located downstream of the filter collects gas phase
constituents. Commonly used adsorbent resins include XAD-2, Chromsorb 102,
and Tenax.
The major advantage in using a dilution tube approach is that exhaust
gases are allowed to react and condense onto particle surfaces prior to
sample collection, providing a truer composition of exhaust emissions as
they occur in the atmosphere. Other advantages are that the dilution tube
sampling arrangement is the reference method for sampling regulated
pollutants from vehicle exhaust, and the dilution tube configuration allows
simultaneous monitoring of hydrocarbons, carbon monoxide, carbon dioxide,
and nitrogen oxides. Polycyclic organic matter sampling devices used in'
conjunction with dilution tube arrangements may consist only of filters for
collection of particle bound POM (compounds with five or more rings are
expected to be associated with particulate), or filtration/adsorption
techniques may be used for the collection of both particulate and gas phase
POM. Back-up adsorption techniques are generally recommended for complete
capture of POM because even particle bound POM will readily volatilize from
a filter and pass through in the sample air stream.
365
-------
-------
TABLE 87. DISTRIBUTION'OF POM IN THE PARTICIPATE
AND GAS PHASE FROM VEHICLE EXHAUST.22
"POM
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Chrysene + benz( a) anthracene
Ben2o(e)pyrene
Benzo(a)pyrene .
Benzo(g,h, i)perylene
Picene + dibenzoanthracenes
Anthanthrenes
Dibenzopyrene
Coronene
Filter
40
8
34
36
70
28
9
31
9
1 1
13
3-
4
Microerams /Sample
XAD Trap
16
30
30
40
50
0.1
0.1
0.2
0.2
0.2
0.2
0.2
367
-------
U-l
co
«
00
14-1
-i eo
HI
-------
Condensation Techniques--Polycyclic organic matter emissions from
vehicle exhausts have also been collected vising cryogenic trapping.
Condensation collection systems have been used with both diluted and raw
vehicle exhausts. Figure 44 shows a condensation system for collecting POM
from raw vehicle exhaust.25 After initial filtration, the vehicle exhaust
is condensed using cryogenic vessels cooled by water, dry ice/ethanol, or
liquid nitrogen. Water and other condensable components of the exhaust are
collected in the condensor and recovery flask. Both the condensate and the
collected particulate are analyzed for POM. :
fugitive and Ambler,*: Sampling
Sampling of POM in ambient air, and from specific fugitive emission
sources has been described by several investigators.26'29 Ambient and
fugitive sampling procedures involve the collection of airborne particulate
by means of filtration or impaction. Some methods combine'filtration with
adsorption techniques to ensure collection of vapor phase POM.
High volume samplers are the most commonly used device for collecting
ambient par tide-bound POM. These samples draw ambient air through a
20x25 cm glass fiber filter at rates of 1.1-1.7 cmm (40-60 cfm).
Twenty-four hour sampling periods allow sample volumes of approximately
2039 cm (72,000 cf). An advantage in using high volume samplers is the
large quantity of sample that can be collected in a day. - Disadvantages are
that the high flow rates and long sample periods can contribute to
breakthrough of the more volatile POM as well as potentially contributing to
the formation of artifacts. '
Low volume samplers, which operate at about one tenth the-flow rate of
high volumes, have been used by some investigators to collect particulate
bound POM. Advantages in using low volume samplers are that breakthrough
and artifact formation problems may be less important than with high volume
samplers. The major disadvantage is the small sample size resulting from
use of a low volume sampler. Small sample sizes limit the number and type
of chemical analyses that can be performed.
369
-------
CRYOGENIC VESSELS
(Dry Ice/Ethanol or
Liquid Nitrogen)
RAW
VEHICLE"
EXHAUST
LL
FILTER
HOLDER
WATER COOLED CONDENSER
o
Figure 44. Condensation sampling system for.raw
vehicle exhaust.
-------
The addition of adsorbent materials such as Tenax, Chromsorb 102,
XAD-2, and polyurethane foam downstream of filtration or impaction devices
have been described by some investigators.32'34 These adsorbent materials
collect breakthrough POM which evaporates from particulate during sample
collection. In addition, these materials collect vapor phase POM that are
typically not associated with particulate (e.g., compounds with less than
four rings).
SAMPLE RECOVERY
Quantitative recovery of POM requires the separation of POM from the
remainder of the collected material, as well as efficient removal from
collection media. Solvent extraction techniques which are commonly used for
recovery of POM from filters, adsorbent, and liquid media are briefly
described.
Soxhlet .. .
Soxhlet extraction is generally redognized as the standard method for
preparing a POM-containing solvent extract of solid matrices.34 This
technique is applicable for the extraction of POM from both filter and
sorbent catches. This procedure has been specified as a standard reference
for extraction of POM by the American Society for Testing Materials, the
U. S. Intersociety Committee on Recommended Methods, and the U. S.
Environmental Protection Agency's Procedures Manual for Level -1 '
Environmental Assessment.9'34
Filter samples are folded and placed directly in the extraction chamber
of the soxhiet. Polymeric resins are typically transferred to cellulose or
glass extraction thimbles and then placed in the soxhiet for extraction.
Recommended solvents and extraction periods vary depending on the sample
matrix and the collection media.35'36 Table 88 lists reported soxhiet , ,
extraction recoveries of POM from various sample matrices using a range of' '
extraction periods and solvents.35
371
-------
m
en
OT
04
m
m
8
m VD f.
g
01 e«
•s I
sr ?
r*» vp
i; i if\ C*
0 I • .
*5 *» —
I M
s
O m
es ^
2,
372
-------
f-+
•9
§
B
4J
e
o
u
N*'
vn
w
Cd
U
^H
0!
H
S
Cd
^J
P-
5
CO
a
S-
^
o
g
X
2.
Ex
O :
t
CO >
Cd
M
>
O
Cd
S5
0
&
I
Cd
e_j
S
1
CO
00
00
Cd
Ǥ
H
*•»
J
A
e B
a
*
•t
e
>
e
•>P
t.
u
X M
U B
"3
03
Ml
5 2-
c a
!§
cu eg
J3
"3
i
\
cu
X
bd
IS
X
99
"a.
CO
CO CO CM
< < «
"
2 1 ™T
s
u u
i S !
.2 .2 '
•S -S
4 a
GS 60
. S Q0
"3 0, 'u
1 " -
s- '
SB «
91
•O
I
M
H
S
1
>,
41
••4
n .
u
Q
B
(V
B •
i.
O.
S
§
•3
f
H
I
.-I
I
<
«
•w
I
• —
1"
«
• M
It*
« S ..
fr * «
9i S i
§ 5 5
u e 41 .
• 41 I 41
g W -0 -S
2 B 2 • -a
« - g S o
eg at .»JB
2 2 "^ ai o
E • « 41 »L
* *S 2 S = ^
« ° J5 g « S>
•«» f g £% 1
>, .2 3 .5 g
U *• out!
* * SH « a
> ta U ..j
3 U 4) e*> g g
2 I I a* s.
J O 1* V »* J.V
** e *§ "3 §
_ £ 0 ~4 s <0
w o
^ w -O W ai u-
•
a
o
I
U
•M
i
B
'"*
I
^
a
*
>«
(4
at
>
41
a
373
-------
Typical solvents used for extraction of POM from filters, include
methylene chloride, cyclohexane, or benzene.6'12'21«37 Some investigators
recommend an initial extraction with methylene chloride followed by
subsequent extraction with a more polar solvent such as methanol.38
Solvents- for extraction of polymeric resins are typically chosen based on
the nature of the adsorbent. Methylene chloride followed by methanol is
commonly selected for extracting POM from XAD-2 and Chromsorb 102 resins.
Hydrocarbons, such as pentane followed by methanol, have been recommended
for extracting Tenax.39
Sonication.
Ultrasonic agitation or sonication uses high'intensity ultrasonic
vibration (-20 KHz) to enhance solvent sample contact. Extractions involve
the insertion of a sonication probe into the sample-containing extraction
vessel',' or a sonication bath in which the sample-containing extraction
vessel is set. Filter samples are typically shredded and placed in a glass'
extraction vessel along wi.th solvents. Sonication is typically carried out
for periods ranging from a few minutes to one hour.40 Extracted POM are
then separated from insoluble materials using conventional filtration
techniques. Table 59 lists reported ultrasonic agitation recoveries of POM
from air particulate and coal fly ash using a range of extraction periods
and solvents. Recommended solvents include cyclohexane, benzene,
acetonitrile, tetrahydrofuran, and methylene chloride.42
: *
Solvent Partitioning--
Solvent partitioning, or liquid-liquid extraction is the traditional
procedure for extraction from liquid sample matrices.9'43 The extraction is
typically performed in a separatory funnel by agitation and shaking the
sample-containing liquid with a suitable solvent. Reported solvents include
methylene chloride and cyclohexane.44
374
-------
§
8
2
£
W
U
M
§
of
<-3
9
>"
B3
SB
00
•<
fM
fa1
3
S
a
5
S
>J
a
u
K
§
i— i
<
S
fi
M
2
fa
°
00
M
S
I
H
«
00
_q
Q
H
CD
e
o s
•rt °H
w H
• p.
•c
1
a
.s
.u
u
« J
h 4J
•u e
x
I— 1
o
00
4)
eS J?
JJ a)
e >
4) O
O S
U 4)
,0.
r— 1
0)
IB
I
X
•H
W
^o
2!
4>
^*4
ft .
§
CO
*"! -H es en sf
X X X X x
•5 .2 .5 .g .5 .5 .g
-s s s s a s s
oo m m in m m o
»H
ffi « fa
® ^ « S
"0 n m ^ ^4 o oo
oJ&SS00^-*^
.AtSri~*°°<:r'a*
dO dO 00 {ju _.
S c c j s e
S 5 S •' 2 - §
•-< en g
:
M
3 g aj eg as ^g
SP • « M gg
•
'
.
oa '
41 :
tJ I i ': '
Qj ! i
*™* i t
3 :
U '
•H
4J >'
S. '. ' '•
J
'H i :
•< • i i
»-H
X
c
s
o
en
g
NO
eg
00
3
0
en
1
0
^^
a)
es -4 es es
X X x X
« e c o
see*
0000
en NO NO en
«
CO sO ^M 1^
00
' ?
o -
o
i—4
03
.£
00
«
±*
s—4
M-l
(-1
(S
«
•
4)
C
41
X
jj
Q
4)
•E
u
e
0)
JS
P4
• «
c
4)
Ps
*£?
es
^^
0
a
4)
A
II
PM
ce
a A
§
41
U
a)
*•<
4=
4J
e
03
II
^
JS
4J
03 .£
t
ea
4>
>
O
fflj
a
i
o
d
o
°H
•a
OS
i-i
g
4)
U
C
O
u
^
o
u
e
0
O
^ C
..
«
C
03
3
*O
Jr
CO
4)
1
• M
0)
0)
M
e
4)
II
O A
gj
S
X
1
U
o"
II
O
OA
rS
•H
iJ
'c
0
0)
U
as
U
J
375
-------
-------
IDENTIFICATION AND QUANTIFICATION OF POM ' '
A variety of analytical techniques have been used to quantify the POM
content of complex environmental samples. This section presents a brief
overview of the most commonly used techniques.
High Performance Liquid Chromatography (HPLC)--
The use of liquid chromatography for the determination of specific POM
compounds in complex environmental samples has increased significantly -in
recent years. Detailed reviews are available in the literature that
describe various modes of separation, and applications of liquid .
chromatography (LC) in the measurement of POM.45'52 Although not offering
the high separation efficiency of capillary .Gas Chromatography (GC), HPLC
offers three distinct advantages for POM analysis. First, HPLC offers a
variety of stationary and mobile phases which provide selectivity for the
separation of POM isomers not generally separated by GC. Second, HPLC
coupled with a fluorescence detector provides both sensitivity and
selectivity. Individual POM compounds have characteristic fluorescence
excitation and emission spectra, whereas isomeric POM have very similar if
not identical mass spectra. Finally, HPLC is an extremely useful
fractionation technique for the isolation of POM for subsequent analysis by
other chromatographic or spectroscopic techniques.
, Gas Chromatography (GC)--
Several studies have been performed using gas chromatography for the
separation and determination of POM in environmental samples. Detailed
reviews^re available in the literature that describe various applications
of GC.
The most frequently used detector for GC analysis of POM is the flame
ionization detector (FID). Its general response character makes it ideal
for several classes of compounds, but necessitates an extensive clean-up
376
-------
procedure prior to GC to eliminate possible interfering compounds. The
advantages of using FID include linear response, sensitivity, and day-to-day
quantitative reliability to routine determinations. Typical detection
limits are below 1 ng.
Numerous applications using the combination of Gas Chromatography and
Mass Spectrometry (GO/MS) are also described. EPA Methods 625 and 1625 are
both GC/MS techniques for the determination of POM compounds.55'56
Advantages of GC/MS techniques include a high level of sensitivity for trace
level detection, versatility for the separation of a large number of
compounds, and specificity for absolute identification. The marked
disadvantage is that it is significantly more expensive than other
techniques.
377
-------
REFERENCES FOR SECTION 5
1.
2.
3.
Burlingame, J. D. , et al. Field Test of Industrial Coal Stoker Fired
Boilers for Inorganic Trace Element and Polynuclear Aromatic
Hydrocarbon Emissions. EPA Report No. 600/7-81-167 U S
Environmental Protection Agency, Research Triangle Park,' Research
Triangle Park, North Carolina. October 1981. p. 37.
Sonnichsen T W. Measurement of POM Emissions from Coal-fired Utility
?a£6I?; r ** *ep°f No- J8 -288S" HUctrlc Power Research Institute/
Palo Alto, California. February 1983. p. 3-1.
Reznik' *»"»in«y Characterization
Residential Combustion Equipment. EPA
U- S« Environmental Protection Agency,
Laboratory, Research Triangll Park,
4.
5.
7.
8.
9.
10.
Cottone, L E. Summary Test Report of Test Method Evaluations and
£^68°S f!S? ^TT f°; ^"J5 Woodstoves- ^epared under EPA Contract
No. 68-02-3996. U. S. Environmental Protection Agency, Research
Triangle Park, North Carolina. December 1985. p 2. *esearcn
Jones, P W J. E. Wilkinson, and P. E. Strup". Measurement of
ll%tyr °rpnic Materials and Other Hazardous Organic Compounds in
Stack Gases, State of the Art. EPA Report No. 600/2-77-202 US
' '
Modified
Protection Agency, Research Triangle Park, 'North Carolina^Say 1984.
Reference 4, p. 14.
Reference 6, p. 1.
and W. F. Gutknecht.
378
-------
11.
12.
15.
16
17.
1984 ad
1384 and Later
4°'
Year Heavy Duty Engines.
Regulations for
January 21, 1980.
13.
Lee, F. S., and D. Schuetzle. Sampling, Extraction, and Analysis of
Polycyclic Aromatic Hydrocarbons from Internal Combustion Engines. In-
K^fira.*^ ^-carbons, A. BJorseth, ed. Marcel"
Stenberg, V. R. PAH Emissions from Automobiles. In: ' Handbook of
Polycyclic Aromatic Hydrocarbons - Volume 2. A. Bjorseth and
T. Ramdahl, eds. Marcel Dekker, Inc., New York. 1985. pp. 88-91.
14. Reference 13, p. 94.
Petersen, BAG. C. Chuang, T. L. Hayes, and D. A. Trayser..
Analysis of PAH in Diesel Exhaust Particulate by High Resolution
capillary Column Gas Chromatography/Mass Spectrometry In-
IroSS^ar ^"^Hydrocarbons: Physical and Biological Chemistry,
Proceedings of the Sixth International Symposium on Polynuclear
Aromatic Hydrocarbons, Columbus, Ohio, 1981. M. Cooke, A. J. Dennis
?;-, ; cFisher' eds' BatteHe Press, Columbus, Ohio. 1982.
pp. 641-653.
Choudhury, D R and B. Bush. Polynuclear Aromatic Hydrocarbons in
Diesel Emission Particulates . EPA Report No. 600/9-80-057a
•J: i7!e?«°' R« M' Danne*> and N. A. Clarke, eds. November 1980.
pp. J./D-J.O3.
M' YU> and W' G' Th^ly- Compounds Associated with
!artirlateS" In: P°l^clear Aromatic Hydrocarbons:
Chemical Analysxs and Biological Fate, Proceedings of the Fifth
^ M T0?1^ ™ P°iynUClear Aromatic Hydrocarbons, Columbus,
' eds- Battelle Press' coiumbus'
18. Waraer-Selph, M. A;, and H. E. Dietzmann. Characterization of
Heavy-Duty Motor Vehicle Emissions Under Transient Driving Conditions
S«!S i, JT*7' ReP°rt NO> 600/53-84-104. Atmospheric Sciences'
^oratory, Research Triangle Park, North Carolina.
19. Reference 12, p. 39.
20. Reference 13, p. 96,
21.
sf4' > ' ' Fevris' PAH Emi^ions from a
Stratified-Charge Vehicle With and Without Oxidation Catalyst:
Sampling and Analysis Evaluation. In: Polynuclear Aromatic
Hydrocarbons Proceedings of the Third International Symposium on
CarrtlC Columbus' ^io, 1978T P. W. Jones
379
-------
22. Reference 12, p. 41.
23. Reference 13, p. 98.
24. Reference 12, p. 40.
25. Reference 13, p. 97.
27
28
34.
-, f ' A> Petersen- Review of Sampling and Analysis
Methodology for Polynuclear Aromatic Compounds in Air from Mobile
Sources EPA Report No. 600/4-85-045. U. S. Environmental Protection
Agency, Research Triangle Park, North Carolina. June 1985. pp. 24-36.
J/l-381.
. Inc., Ann Arbor, Michigan. 1979.
i«i i4 Air Monitoring Study of Residential Woodbuming
in Mio Michigan, Presented at the 78th Annual Meeting of the Air
Pollution Control Association, Detroit, Michigan. Jun! 16-21, 1985.
O3~*w.A« L3 pp.
30. Reference 26, p. 28.
31. Lindskog, A Transformation of Polycyclic Aromatic Hydrocarbons Durins
Sampling. Environmental Health Perspectives. 47: 81-84. 1983?
32" TecSieuesJfor't H' J* KraUSS> and M* A' Fox' A Comparison of Two
Compounds in AmbienAir"? JouSal^ S^Alr PollStionaControltiC
Association. 30(2): 166-168. 1980. Control
33.
°u 5* ' an? T/ F' Bidleman- Collection of Airborne Polycyclic
Aromatic Hydrocarbons and Other Organics with a Glass Fiber Filter
Polyurethane Foam System. Atmospheric Environmental. 18: 837-845.
Hvd ; > ?' Cat0n' Extracti°n of Polycyclic Aromatic
Hydrocarbons for Quantitative Analysis. In: Handbook of Polycyclic
Aromatic Hydrocarbons, A. Bjorseth, ed. Marcel Dekker, Inc. /New York.
380
-------
35. Reference 34, pp. 102-103
36. Reference 12, p. 45.
37. Reference 34, pp. 102-104.
38. Reference 5, pp. 41-43.
39. Reference 5, p. 42.
40. Reference 34, p. 107.
41. Reference 34, p. 10a.
42. Reference 34, 'pp.. 107-112.
43. Reference 34, p. 112.
44. Reference 34, p. 114.
45> ?fns' W'» K- °San« and
49 '
- « - . e. ery g-p
Chromatography for PAH Analysis System and Applications
DiCesare. Very High-Speed Liquid
In-
.
Af °ftic «ydroca^ons: Physical and Biological Chemistry
S
Symposium on Polynuclear
o» « , n o
Aromatic Hydrocarbons, Columbus, Ohio, 1981. M. Cooke,
Battelle Press, Gelumbus, Ohio.
237-245
A. J. Dennis
1982.
in the Determination of PAH by High
In: Handbook of Polycyclif
and T- Ramdah1' eds-
46
47. May, W. E. , S. A. Wise. Liquid Chromato graphic Determination of
Polycyclic Aromatic Hydrocarbons in Air Paniculate Extracts
Analytical Chemistry. 56: 225-232. *««-«».
GC/MS
,
Polynuclear Aromatic Hydrocarbons in Fly
Press, Columbus, Ohio. .
pp.
I' L" ^ J° L' DiCesare- ^e Application of High
1'"^?6!1'^11^ ^"^^raphy to the Polycyclic Aromatic
. In: Polynuclear Aromatic Hydrocarbons- Physical
Biological Chemistry, Proceedings of the Sixth International
381
-------
50. James, R. H., R. E. Adams, J. M. Finkel, H. C. Miller, and
L. D. Johnson. Evaluation of Analytical Methods for the Determination
of POHC in Combustion Products. Journal of the Air Pollution Control
Association. 35(9): 959-969. September 1985.
51. Wise, S. A. High-Performance Liquid Chromatography for the
Determination of Polycyclic Aromatic Hydrocarbons. In: Handbook of
Polycyclic Aromatic Hydrocarbons, A. Bjorseth, ed. Marcel Dekker
Inc., New York. 1983: .p. 183. '
52. Federal Register, Volume 49, No. 209. Method 610 Polynuclear Aromatic
Hydrocarbons, pp. 112-120.
53. Bartle, K. S. Recent Advances in the Analysis of Polycyclic Aromatic
Compounds by Gas Chromatography. In: Handbook of Polycyclic Aromatic
Hydrocarbons - Volume 2, A. Bjorseth and T. Ramdahl, eds. Marcel
Dekker, Inc. New York. 1985. pp. 193-237
54. Reference 26, pp. 37-58.
55.
56.
Federal Register, Volume 49, No. 209. October 26, 1984.
Revision B Organic Compounds by Isotopic Dilution'cc/MS.
Federal Register, Volume 49, No. 209. October 26, 1984
Base/Neutrals and Acids, pp. 153-174.
Method 1625
pp. 184-197.
Method 625 -
382
-------
-------
TECHNICAL REPORT DATA
(Flcascnxl Instructions on the reverse before completing
2.
|1. REPORT NO. ___
EPA-450/4-84-607p
4. TITLE AND SUBTITLE
Locating And Estimating Air Emissions From' Sources Of
Polycyclic Organic Matter (POM)
17. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
12. SPONSORING AGSNCY.NAME AND ADDRESS —
Office Of Air Quality Planning And Standards (MD 14)
U. S. Environmental Protection Agency
Research Triangle, NC 27711
3. RECIPIENT'S ACCESSION NO.
5.
September 1987
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPC
1. CONTRACT/GRANT NO.
3. TYPE OF REPORT AND PERIOD COV
4. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
EPA Project Officer: Thomas F. Lahre
MS;
rflACT
Polycyclic organic matter
Polycyclic aromatic hydrocarbons
Polynuclear aromatics
Locating air emissions sources
Toxic substances
Benzo(a)pyrene
EPA
4.77, P*6v,oUSeB1TION,SOBSOUCTE
^IDENTIFIERS/OPEN ENDED TERMS
c. COSATI l-'ieid/Crou
21 NO. OF PAGES"
382
22. PRICE
-------
-------
-------
-------